Gene Therapy & Molecular Biology Volume 3 Isuue A

! !

Gene Therapy & Molecular Biology FROM BASIC MECHANISMS TO CLINICAL APPLICATIONS

!

! ! ! "#$%&'!(! )**%'!+! +,-,./!0111! ! ! ! !

Gene Therapy and Molecular Biology Vol 3

Table of contents Gene Ther & Mol Biol Vol 3, August 1999

Pages

Type of Article

Article title

Authors (corresponding author is in boldface)

1-14

Review Article

Routes of vector application for brain tumor gene therapy

Nikolai G. Rainov, Xandra O. Breakefield, Christof M. Kramm

15-23

Research Article

Efficient in vivo expression of a reporter gene in rat brain after injection of recombinant replicationdeficient Semliki Forest virus

Kenneth Lundstrom, J. Grayson Richards, J. Richard Pink, and Francois Jenck

25-33

Research Article

Establishment of an assay to determine adenovirus-induced endosome rupture required for receptor-mediated gene delivery

Daniela Schober, Nora Bayer, Robert F. Murphy, Ernst Wagner and Renate Fuchs

35-44

Review Article

Gene regulation in Herpesvirus saimiri and its implications for the development of a novel gene therapy vector

Adrian Whitehouse and Alex J. Stevenson

45-56

Review Article

Regulation of papillomavirus transcription and replication; insights for the design of extrachromosomal vectors

Alison A. McBride

57-65

Review Article

Gene transfer with adeno-associated virus 2 vectors: the growth factor receptor connection

Cathryn Mah, Keyun Qing, Jonathan Hansen, Benjawan Khuntirat, Mervin C. Yoder, and Arun Srivastava

67-74

Research Article

Hepatocyte-specific gene expression by a recombinant adeno-associated virus vector carrying the apolipoprotein E enhancer and !1antitrypsin promoter

Torayuki Okuyama, Motomichi Kosuga, Satori Takahashi, Kyoko Sasaki, and Masao Yamada

75-78

Minireview

Human cytomegalovirus (HCMV) nuclease: implications for new strategies in gene therapy

Elke Bogner

Gene Therapy and Molecular Biology Vol 3

79-89

Review Article

Application of recombinant Herpes Simplex Virus-1 (HSV-1) for the treatment of malignancies outside the central nervous system

George Coukos, Stephen C. Rubin, and Katherine L Molnar-Kimber

91-101

Review Article

Transcriptional repression in cancer gene therapy: targeting HER-2/neu overexpression as an example

Mien-Chie Hung and Shao-Chun Wang

103-112

Review Article

Gene therapy targeting p53

John Nemunaitis

113-121

Research Article

Targeted therapy of CEA-producing cells by combination of E. coli cd/HSV1-tk fusion gene and radiation

Dao-song Xu, Xin-yao Wu, Yun-fei Xia, Ling-hua Wu, Chao-quan Luo, Yin-hao Yang, Lu-qi Zhong, and Bin Huang

123-131

Research Article

Efficacy of antiherpetic drugs in combined gene/chemotherapy of cancer is not affected by a specific nuclear or cytoplasmic compartmentation of herpes thymidine kinases

Bart Degrève, Erik De Clercq, Anna Karlsson, and Jan Balzarini

133-148

Review Article

Glioblastoma multiforme: molecular biology and new perspectives for therapy

Giorgio Palù, Luisa Barzon, and Roberta Bonaguro

149-155

Review Article

Gene-based vaccine strategies against cancer

Daniel Lee, Ken Wang, Liesl K. Nottingham, Jim Oh, David B. Weiner, and Jong J. Kim

157-165

Review Article

Rational vaccine design through the use of molecular adjuvants

Jong J. Kim, Liesl K. Nottingham, Jim Oh, Daniel Lee, Ken Wang, Mera Choi, Tzvete Dentchev, Darren Wilson, Devin M. Cunning, Ara A. Chalian, Jean Boyer, Jeong I. Sin, and David B. Weiner

167-177

Review Article

In vivo production of therapeutic antibodies by engineered cells for immunotherapy of cancer and viral diseases

Mireia Pelegrin, Danièle Noël, Mariana Marin, Estanislao Bachrach, Robert M. Saller, Brian Salmons, and Marc Piechaczyk

179-187

Research Article

Use of DNA priming and vaccinia virus boosting to trigger an efficient immune response to HIV-1 gp120

Dolores Rodríguez, Juan Ramón Rodríguez, Mercedes Llorente, Pilar Lucas, Mariano Esteban, Carlos Martínez-A. and Gustavo del Real

189-196

Review Article

Gene therapy approaches to the treatment of hemoglobinopathies

Linda Gorman and Ryszard Kole

Gene Therapy and Molecular Biology Vol 3

197-206

Research Article

Intramuscular injection of plasmid DNA encoding intracellular or secreted glutamic acid decarboxylase causes decreased insulitis in the nonobese diabetic mouse

Jingxue Liu, Maria Filippova, Omar Fagoaga, Sandra Nehlsen-Cannarella, and Alan Escher

207-221

Review Article

Muscle-based tissue engineering for the musculoskeletal system

DS Musgrave and Johnny Huard

223-232

Review Article

Helper-dependent adenoviral vectors as gene delivery vehicles

Manal A. Morsy, Diane M. Harvey, and C. Thomas Caskey

233-241

Review Article

Gene transfer into muscle for the treatment of muscular dystrophy and haemophilia

Geoffrey Goldspink, Maria Skarli and Paul Fields

243-248

Review Article

Gene therapy for arthritis

Sherry Thornton and Raphael Hirsch

249-256

Review Article

Antisense gene therapy in the longterm control of hypertension

Craig H. Gelband, Michael J. Katovich, Mohan K. Raizada

257-269

Research Article

Construction and deployment of triple Ling Ren, Shani L. Schalles, Weihua Pan, Corinne E. Isom, Sarah E. Loy, Jiaribozymes targeted to multicatalytic Hai Lee, Catharine M. Benedict, Mary T. proteinase subunits C3 and C9 Pickering, James S. Norris, and Gary A. Clawson

271-280

Review Article

Development of hammerhead ribozymes for HIV-1 gene therapy: principles and progress

A. Ramezani and Sadhna Joshi

281-291

Review Article

Use of antisense oligonucleotides to study homeobox gene function

Olubunmi Afonja, Takashi Shimamoto, John E. Smith, Jr., Long Cui, and Kenichi Takeshita

293-300

Research Article

Potential application of dominant negative retinoic acid receptor genes for ex vivo expansion of hematopoietic stem cells

Yoji Ogasawara, Yutaka Hanazono, Hiroshi Kodaira, Masashi Urabe, Hiroyuki Mano, Akira Kakizuka, Akihiro Kume, Keiya Ozawa

301-310

Research Article

Optimized expression of serotonin receptors in mammalian cells using inducible expression systems

Peter Vanhoenacker, Walter Gommeren, Walter H.M.L. Luyten, JosĂŠe E. Leysen and Guy Haegeman

Gene Therapy and Molecular Biology Vol 3

311-325

Research Article

Identification of a negative regulatory mechanism for the repair of U5 long terminal repeat DNA by the human immunodeficiency virus type 1 integrase DNA polymerase

Brian E. Udashkin, Andrea Acel, Avi Shtvi, Benjamin Alt, Henry Triller, Mark A. Wainberg and Emmanuel A. Faust

327-345

Review Article

Brian A. Lenzmeier and Jennifer K. Nyborg

347-354

Review Article

Molecular mechanisms of viral transcription and cellular deregulation associated with the HTLV-1 Tax protein What does acetylcholinesterase do in hematopoietic cells?

355-371

Review Article

The ETS-domain transcription factors: lessons from the TCF subfamily

Shen-Hsi Yang, Paula R. Yates, Yi Mo and Andrew D. Sharrocks

373-378

Roxanne Y.Y. Chan and Bernard J. Jasmin

D.A. Spandidos, G. Sourvinos, S. Transcriptional activation of the ras oncogenes and implications of BRCA1 Miyakis in the cell cycle regulation through p53 checkpoint

379-385

Review Article

Nuclear receptor coactivators as potential therapeutical targets: the HATs on the mouse trap

Arndt Benecke and Hinrich Gronemeyer

387-395

Review Article

High mobility group protein HMGIC: a molecular target in solid tumor formation

Erik Jansen, Marleen M.R. Petit, Eric F.P.M. Schoenmakers, Torik A.Y. Ayoubi, and Wim J.M. Van de Ven

397-412

Review Article

Replication of simple DNA repeats

Maria M. Krasilnikova, George M. Samadashwily and Sergei M. Mirkin

413-422

Research Article

Separation of the DNA replication and transactivation activities of EBNA1, the origin binding protein of Epstein-Barr virus

Derek F.J. Ceccarelli and Lori Frappier

423-435

Research Article

The activation of the lysozyme locus in development is a cooperative process

Matthias C. Huber and Constanze Bonifer

437-445

Review Article

Mechanisms involved in regulation of the estrogen-responsive pS2 gene

Ann M. Nardulli, Jongsook Kim, Jennifer R. Wood, and Lorene E. Romine

Gene Therapy and Molecular Biology Vol 3

447-453

Review Article

Biological function of the USF family of transcription factors

Michèle Sawadogo, Xu Luo, Mario Sirito, Tao Lu, Preeti M. Ismail, Yibing Qyang, and Marilyn N. Szentirmay

455-464

Review Article

The role of chromatin in the establishment of enhancer function during early mouse development

Luca Rastelli and Sadhan Majumder

465-474

Review Article

Molecular mechanisms that regulate hyaluronan synthesis

Paraskevi Heldin

Gene Therapy and Molecular Biology Vol 3, page 1 Gene Ther Mol Biol Vol 3, 1-14. August 1999.

Routes of vector application for brain tumor gene therapy Review Article

Nikolai G. Rainov1, Xandra O. Breakefield2, Christof M. Kramm3 1

Department of Neurosurgery, Faculty of Medicine, Martin-Luther-University, Halle, Germany, 2 Molecular Neurogenetics Unit, Department of Neurology, Massachusetts General Hospital, and Neurosciences Program, Harvard Medical School, Boston, MA, USA University Children’s Hospital, Heinrich-Heine-University Medical Center, Duesseldorf, Germany.

__________________________________________________________________________________ Corresponding author: Nikolai G. Rainov, M.D., Martin-Luther-University Halle, Dept. Neurosurgery, Magdeburger Str. 16, D-06097 Halle, Germany. Tel: +49 345 5571399; Fax: +49 345 5571412; E-mail: nikolai.rainov@medizin.uni-halle.de Key words: adenovirus, brain neoplasms, gene therapy, gene transfer, herpes simplex virus, intra-carotid delivery, liposomes, plasmids, retrovirus. Received: 25 September 1998; accepted: 5 October 1998

Summary The development of highly efficient virus and non-virus vector systems for gene transfer to and gene therapy of brain tumors has advanced to the stage of clinical trials, but has still not successfully addressed some major limiting factors, such as the inability of a single delivery modality or therapeutic transgene to target a maximum number of tumor cells in diffuse or multifocal tumors, such as human glioblastoma, and to confer eradicating cytotoxicity to the whole neoplastic mass. Moreover, the choice of vectors and the route of their administration dramatically affect both the efficiency of tumor transduction and its spatial distribution, as well as the extent of transgene expression within a brain tumor and outside it, in the surrounding tumor cell-infiltrated tissue. Three main routes of vector delivery to experimental brain tumors are reviewed in this paper: stereotactic or direct intratumoral inoculation; intrathecal and intraventricular injection; and intravascular infusion with or without modification of the blood-brain-tumor-barrier. The pros and cons of all these modes of application are discussed in respect to the specific and unique features of tumors in the central nervous system. We conclude that, at the present time, there is no ideal vector or unconditionally efficient application mode, and so the successful approaches to brain tumor gene therapy need to combine different application routes with different vectors and therapeutic genes designed to address the individual features of different tumor types. The intravascular vector delivery route, although at an early stage of development, seems to be the most pervasive and demonstrates the greatest therapeutic potential in animal experiments, but for human use it should be combined either with direct intratumoral vector injections or with CSF vector delivery.

transduced cells surrounding transgene-expressing cells, still the transgene-bearing vector must be delivered to a substantial number of tumor cells (1-10%) throughout the tumor (Moolten, 1996; for review see Kramm et al., 1995, and Spear et al., 1998). The choice of vectors and the route of their administration have been demonstrated to affect both tumor transduction efficiency and spatial distribution, as well as the extent and stability of transgene expression within a tumor, in invasive tumor cells, and in the surrounding normal brain (Zlokovic and Apuzzo, 1997).

I. Introduction The advancement of gene therapy for brain tumors through the stage of animal models into clinical trials has not succeeded in eliminating major limiting factors, such as the inability of a single vector or delivery mode to target a pool of tumor cells large enough to confer cytotoxicity to the whole tumor (Ram et al., 1997). Even with the bystander effect elicited by many therapeutic genes, which is responsible for the killing of non-

1

Gene Therapy and Molecular Biology Vol 3, page 2 Three main modes of vector delivery to experimental brain tumors have been extensively studied and compared in animal models and, in some cases, in clinical trials: stereotactic intratumoral inoculation of virus suspension or vector-producing cells (VPC) (Badie et al., 1994; Boviatsis et al., 1994a and 1994b; Bramson et al., 1997; Culver et al., 1992; Eck et al., 1996; Izquierdo et al., 1995; Kramm et al., 1997; Mineta et al., 1994: Oldfield et al., 1993; Rainov et al., 1996; Ram et al., 1993 and 1997); intrathecal and intraventricular injection of virus or VPC (Bajocchi et al., 1993; Kramm et al., 1995, 1996, 1997; Oldfield et al., 1995; Oshiro et al., 1995; Ram et al., 1994; Rosenfeld, 1997; Vincent et al., 1996a); and more recently, intravascular application of virus vectors (Barnett et al., 1998; Chauvet et al., 1998; Doran et al., 1995; Kroll and Neuwelt, 1998; Muldoon et al., 1997; Neuwelt et al., 1991; Nilaver et al., 1995; Rainov et al., 1995 and 1998). The study of the modes of application and the factors which limit vector distribution and propagation in a brain tumor is of great importance to the improvement of present gene therapy strategies and the development of more efficient approaches. Therefore, the present paper will review the routes and methods for delivery of gene therapy vectors to malignant brain tumors, and will focus on strategies which may have the potential of improving the efficiency of gene transfer to brain tumors in vivo.

1991; Mineta et al., 1995); adenovirus (AV) (Badie et al., 1994 and 1998; Boviatsis et al., 1994a; Chen et al., 1994; Eck et al., 1996; Izquierdo et al., 1996; Le Gal La Salle et al., 1993; Maron et al., 1996; Perez-Cruet et al., 1994, Puumalainen et al., 1998); adeno-associated virus (AAV) (Mizuno et al., 1998; Okada et al., 1996); and liposomeDNA complexes (lipoDNA) (Gennuso et al., 1993; Yagi et al., 1994; Zerrouqi et al., 1996; Zhu et al., 1996); or for implantation of retrovirus (RV) producing cells (VPC) (Rainov et al., 1996; Ram et al., 1993; Culver et al., 1992; Takamiya et al., 1993; Tamiya et al., 1995). Since the life cycle of replication-conditional HSV and AV vectors is lytic or damaging to the host cell, there have been no available cell-based vector producer systems for in vivo use of these viruses (Kramm et al., 1995).

A. Direct intratumoral injection of vectors Stereotactic surgery methods combined with 3D computer reconstruction and imaging databases provide powerful options for tumor gene therapy (Kelly, 1997). Volumetric stereotactic procedures can be modified for individually planned delivery of viruses or VPC to single or multiple tumor foci. Despite the potential for high spatial accuracy, direct intra- or peritumoral injections have several disadvantages, such as limited vector distribution to a few millimeters surrounding the injection site (Boviatsis et al., 1994a; Rainov et al., 1996; Lal et al., 1994), and the need of multiple injections of either virus or VPC suspension even with large volumes of inoculum of up to 0.5 ml per injection site (Muldoon et al., 1997; Ram et al., 1997) (Fig. 1A and B). Since the number of stereotactic injection sites is limited for practical reasons by length of surgery and increasing risk of hemorrhage with every new intracerebral puncture track, this mode of application can only provide vector delivery to small intracerebral foci or limited tumor areas (Spear et al., 1998).

II. Intratumoral delivery of vectors The earliest and most straightforward approach to delivery of gene therapy vectors to brain tumors is the stereotactic intratumoral injection (Short et al., 1990; Boviatsis et al., 1994a) or the direct injection after open surgery for brain tumor removal (Ram et al., 1997). It offers the advantages of low systemic toxicity, reduced vector loss, and high local vector concentrations, and can be employed either for application of concentrated vector suspension, as in the case of herpes-simplex-virus type 1 (HSV) (Andreansky et al., 1993; Boviatsis et al., 1994b; Kaplitt et al., 1994; Kramm et al., 1997; Martuza et al.,

Fig. 1: Microscopic appearance of 9L tumors in syngeneic Fischer rats. A. Photomicrograph of tumor tissue in an animal from the TK/GCV group, 7 days after intratumoral grafting of retrovirus-packaging cells (CRIP-MFG-TK) in a 1:5 ratio of producer cells to tumor cells. This section was stained immunohistochemically for HSV-TK. Note the high number of HSV-TK-positive cells (darkbrown, arrows) in the tumor (magnification 200 x, 20 µm frozen section, counterstained with hematoxylin). B. Photomicrograph of tumor tissue 12 days after grafting of CRIP-MFG-TK cells and 5 days after start of GCV application. Necroses (arrowheads) are visible inside the tumor area. HSV-TK-positive cells (arrows) are still detectable, but numbers are much lower (magnification 200 x, 20 µm frozen section, counterstained with hematoxylin). C. X-gal staining 10 days after implantation of 9L cells and 6 days after intrathecal injection of replication-conditional HSV vector bearing the lacZ gene demonstrates widespread distribution of vector in leptomeningial 9L tumor cells (blue, arrows) directly contacting the CSF. In the tumor parenchyma (T), only a few cells display X-gal staining, B = normal brain (75x magnification, 20 µm frozen section, counterstained with hematoxylin and eosin, H&E). D. Frontal tumor (T) in an animal 7 days after intracerebral and intrathecal implantation of 9L cells and 2 days after intrathecal inoculation with HSV vector used in C. Extensive X-gal staining (blue, arrows) is seen in tumor areas which have broken into the lateral ventricle, while other parts of the tumor (T) with no CSF contact show essentially no staining. Normal brain (B) is not affected by intrathecal HSV application (100x magnification, 20 µm frozen section, counterstained with H&E). E. Photomicrograph of intracerebral tumor in the BK/HSV group 24

2

Gene Therapy and Molecular Biology Vol 3, page 3 hours after ipsilateral intra-carotid virus injection in the presence of bradykinin (BK). This section was double-stained for ß-gal (blue cells, arrows) and HSV-TK (dark-brown, arrowheads). Note the higher number of stained tumor cells at the tumor/brain border (T = tumor, B = brain) and the absence of transgene protein staining in normal brain (magnification 200x, 20 µm frozen section, counterstained with H&E). F. Photomicrograph of 9L gliosarcoma in the BK/HSV group 48 hours after virus injection in the presence of BK. This section was double-stained for ß-gal (arrows) and HSV-TK (arrowheads). Note the higher intensity of tumor staining, probably due to secondary spread of replication-conditional HSV vector and infection of neighboring tumor cells. Staining is limited to the tumor and does not extend into surrounding normal brain (T = tumor, B = brain) (magnification 200x, 20 µm frozen section, counterstained with H&E).

3

Gene Therapy and Molecular Biology Vol 3, page 4 area. Infusion time did not affect distribution, and the volume infused was closely related to the size of the distribution area (Kroll et al., 1996). The same group (Muldoon et al., 1995) infused AV or HSV into normal rat brain for 2 hours at a rate of 0.2 Âľl/min and found widespread infection in tissue volumes of 40 mm3 (replication-defective AV) and up to 200 mm3 (replication-conditional HSV). When applied to rat brain tumors, this technique was able to mediate delivery of virus particles to tumors with an approximate volume of 100 mm3, and also beyond the tumor borders into the surrounding brain tissue (Nilaver et al., 1995). Virus vectors, however, do not travel in the extracellular space of the brain solely by diffusion, since they bind to receptors and are taken up by cells, and because they are very large (e.g. HSV diameter = 150 nm). Tumor or brain cells near the injection or infusion site may take up many more virus particles than cells distant to it, which reduces the particle numbers of the suspension that diffuses further.

Although several reports have previously demonstrated that this mode of vector delivery may be efficient in rodents (Badie et al., 1994; Boviatsis et al., 1994c; Bramson et al., 1997; Culver et al., 1992; Izquierdo et al., 1995; Mineta et al., 1994; Rainov et al., 1996, Ram et al., 1994; Tamiya et al., 1995), it does not reach the same degree of efficiency in humans (Raffel et al., 1994; Ram et al., 1997). Part of the problem seems to be that human glioblastomas (GBM) are much larger, more randomly shaped, and more diffusely infiltrating than the rodent glioma models (Izquierdo et al., 1997; Kramm et al., 1995; Zlokovic and Apuzzo, 1997). Further, they have a lower fraction of dividing tumor cells which limits on site propagation of replication-conditional HSV and integration of RV (Ram et al., 1997, Harsh et al., in preparation, Puumalainen et al., 1998), and, since most vectors derive from common pathogens, the immune system may block infection of tumor cells. Herrlinger et al. (1998) have investigated the role of the immune system in HSV-mediated gene transfer and found that rats preimmunized to HSV had dramatic decrease in transduction efficiency to brain tumors.

The above data demonstrate that convection-enhanced vector delivery is sufficient for targeting a relatively small and circumscribed rodent brain tumor implant. If this technique should be applied to human GBM, much larger tumor volumes have to be targeted and, since there is no selectivity in the delivery mode itself, normal brain tissue may be overloaded with vectors leaking out of the tumor mass.

Direct intratumoral injections into the walls of the tumor resection cavity, although they can be performed under direct visual control and with multiple vector depots very close to each other, have the same basic limitations as stereotactic procedures. Moreover, the depth of injection is limited to 10-15 mm from the resection border, which seems to be insufficient to reach tumor cells migrating away from the main tumor mass. Thus, both stereotactic and "free-hand" injection techniques are inefficient in cases of multiple tumor foci and diffusely infiltrating tumors.

III. Intrathecal and intraventricular vector application The intrathecal gene therapy approach is attractive because access to the cerebrospinal fluid (CSF) is minimally invasive and distribution of virus vectors and VPC may be facilitated by CSF circulation, thus overcoming distribution barriers in solid tumors. Intrathecal delivery seems to be best suited for treatment of leptomeningial tumor manifestations. These are found in adults in secondary intracerebral tumors of carcinomas and lymphomas. The most frequent primary brain tumor in children, the medulloblastoma, often spreads from its primary location in the cerebellum and the fourth ventricle via the CSF pathways along the entire spinal cord down to the cauda equina. Moreover, leukemic leptomeningiomatosis is a frequent site of relapse of acute lymphoblastic leukemia, the most frequent pediatric cancer.

B. Bulk convection-enhanced flow methods An alternative method for efficient and widespread delivery of macromolecules and particles to tumors is convection-enhanced infusion, which is used to supplement simple diffusion and to improve vector distribution by bulk flow inside and outside the tumor (Bobo et al., 1994; Lieberman et al., 1985; Muldoon et al., 1997). Stereotactic injection and subsequent infusion by maintaining a positive pressure gradient is able to improve the distribution of large molecules in animal models (Lieberman et al., 1995). The volume of distribution seems to increase linearly with the infusion volume, if relative small molecules are used (Bobo et al., 1994). Kroll et al. (1996) used convection for delivery of MION, superparamagnetic iron oxide nanoparticles with a size comparable to that of viruses (Shen et al., 1993), to normal rat brain and found out that the concentration of the agent is of primary importance for the size of the distribution

Retroviral vectors, as well as AV and HSV vectors, have been investigated for their use for gene therapy of leptomeningial tumors after intrathecal administration. Ram et al. (1994) implanted retrovirus producer cells into the leptomeningial space of rats, which have been intrathecally challenged with syngeneic tumor cells some 4

Gene Therapy and Molecular Biology Vol 3, page 5 days prior. Prolonged survival was achieved by subsequent GCV treatment. Gene transfer was demonstrated in tumor foci growing in the cistern magna, the injection site of the VPC. Toxicity and gene transfer into normal cells was also evaluated in rats and nonhuman primates without tumors after single and repeated intrathecal application of retrovirus producer cells (Oshiro et al., 1995). Only choroid plexus cells, and no other normal CNS structures, showed transgene expression. Magnetic resonance imaging of brains of non-human primates revealed no pathological changes. In total, no significant toxicity was observed either in rats or in nonhuman primates, even after repeated intrathecal application of retrovirus producer cells with or without subsequent GCV treatment. Interestingly, measurable titers of retroviral particles were detected in lumbar, as well as in cisternal, CSF samples indicating an effective circulation of vector particles within the CSF. According to a former study, CSF does not inactivate retroviruses or lyse vector producer cells, as it occurs when these cells are incubated with serum of the same species (Russell et al. 1995). However, significant CSF retroviral titers could only be detected in vivo over 24 h (Oshiro et al., 1995). The retroviral studies in rats and non-human-primates by Ram et al. (1994) and Oshiro et al. (1995) were performed as preclinical studies for a clinical trial aiming to treat leptomeningial carcinomatosis by intrathecal application of retrovirus producer cells liberating retroviral particles bearing the HSV-tk gene for sensitization of transduced tumor cells towards subsequent GCV treatment (Oldfield et al., 1995). Despite the encouraging data in animals, this clinical trial was closed prematurely after toxic side effects occurred in the first patient (Anderson et al., 1995).

gene transfer into ependymal and leptomeningial cells, as well as into cerebral blood vessels. No marked toxicity was observed in these studies. This was also true for two studies which investigated gene transfer into rodent leptomeningial tumor masses by replication-deficient AV vectors administered intrathecally. Viola et al. (1995) demonstrated AV-mediated gene transfer into the main tumor mass at the intrathecal injection site. Some limited gene transfer was also noted in tumor manifestations in the cauda equina and along the nerve roots emerging from the spinal cord. Vincent et al. (1996a) observed gene transfer into tumor cells along the entire neural axis after intrathecal administration of AV vectors. These authors achieved a significantly longer survival of treated animals by combining intrathecal delivery of AV vectors carrying the HSV-tk transgene with subsequent GCV treatment, but had no long-term survivors. In contrast to this, when replication-conditional herpes vectors with the HSV-tk gene were injected intrathecally in a similar model of rodent leptomeningial neoplasia as above, long-term survival was achieved in approximately 90% of the animals treated with GCV (Kramm et al., 1996b). One reason for these diverging results may be that replication-conditional HSV vectors replicate in dividing tumor cells, and not in non-dividing normal cells, thereby producing and releasing new vector particles on site which move freely through the CSF. After intrathecal application of replication-conditional herpes vectors, Kramm et al. (1996a) also showed extensive gene transfer into leptomeningial tumors along the entire spinal axis (Fig. 1C ), as well as into parenchymal brain tumors (Fig. 1D). Additionally, ependymal and endothelial cells, as well as neurons projecting to the ventricles, showed marked transgene expression during the first two days after injection of herpes vectors. Five and more days after vector application, normal cells no longer showed transgene expression. However, there was a high degree of toxicity to animals, probably due to inflammatory reaction to the virus, which was apparently absent after intrathecal application of retroviral and adenoviral vectors in rodents (Ram et al., 1994; Viola et al., 1995), but has been noted in humans (Ram et al., 1997; Eck, 1997, personal communication). The viral genesis of symptoms in the HSV study in rats is strongly suggested by the fact that GCV treatment, which blocks virus replication, significantly improved and curtailed this toxicity (Kramm et al., 1996b). The ambivalent potential of intrathecal delivery is emphasized by studies with a new generation of replication-conditional HSV vectors (Kramm et al., 1997, Mineta et al., 1995). The prototype of this new generation herpes vector was designed to be safer than the preceding vectors (ribonucleotide reductase-deleted) by deletion of viral neurovirulence genes (gamma 34.5). Application of this new vector intrathecally in

An interesting alternative to retroviral gene transfer was demonstrated by Vrionis et al. (1996a and b) who showed that therapeutic efficiency can also be achieved in a rat model of leptomeningial neoplasia by co-mixture of native and HSV-tk transduced cells in the same tumor with subsequent GCV treatment. This therapeutic approach relies mainly on the bystander effect describing that close proximity of non-transduced tumor cells with TK-positive tumor cells sensitizes non-transduced cells to GCV treatment. Vrionis et al. (1996a) demonstrated that an HSV-TK/GCV system which exploits the bystander effect at a relative low effector-to-target cell ratio (1:1) is more effective for treatment of leptomeningial neoplasia than the gene therapy approach with intrathecal application of retrovirus producer cells. Intrathecal application of adenoviral vectors has also been used for gene therapy. Bajocchi et al. (1993) showed gene transfer into ependymal cells following direct injection into the ventricles. Ooboshi et al. (1995 and 1997) injected replication-deficient AV (1x109 pfu) intrathecally and demonstrated Ă&#x;-galactosidase (Ă&#x;-gal) 5

Gene Therapy and Molecular Biology Vol 3, page 6 combination with GCV in the same rodent model of leptomeningial neoplasia as described above, was associated with no apparent toxicity or mortality, but also with no significant prolongation of survival of treated animals (Kramm et al., 1997).

terms of selective entry for CNS neoplasms (Rainov et al., 1995). In addition to the BTB, some other factors limit intravascular vector delivery to brain tumors. In order to infect a maximum number of tumor cells, virus vectors must be delivered in sufficiently high titers and should not be inactivated by serum factors (Muldoon et al., 1997). High interstitial fluid pressure within tumors also acts to decrease entry of macromolecules and particles (Jain, 1987 and 1994). Larger tumors generally have a higher interstitial pressure than smaller tumor foci (Leunig et al., 1992), which theoretically limits efficiency of vector delivery to human brain tumors (Boucher et al., 1996). Human glioblastomas also have a variable degree of vascularization, and their microvasculature and hemodynamics vary considerably (Warnke et al., 1987). These obstacles for brain tumor vector delivery call for alternative strategies to circumvent them in order to make delivery of vectors more efficient by additional penetration-enhancing or barrier-modulating techniques.

Rosenfeld et al. (1995) used an adeno-associated virus (AAV) vector to transduce medulloblastoma cells in a nude rat model of leptomeningial disease. After intrathecal application, tumor cells transduced with the marker gene Ă&#x;-galactosidase were detected in tumors, as well as in ependymal and subependymal cells, but not in normal brain parenchyma. No evidence of virus toxicity was noted during the course of the experiment. In conclusion, leptomeningial neoplasia, which represents a main problem in the management of primary and secondary brain tumors, especially in children, is a good target for future gene therapy approaches for intrathecal delivery of therapeutic genes.

IV. Intravascular vector application Intravascular methods of vector application make use of a natural and ubiquitously distributed network of arteries, veins and capillaries, which is present in every normal tissue and is even denser in malignant tumors. Intravascular applications, intra-arterial injection of virus vectors in particular, appear to have the greatest potential to date for delivering a vector to the largest proportion of tumor cells and surrounding tissues without afflicting mechanical injury to normal brain tissue or having other toxic consequences (Spear et al., 1998; Muldoon et al., 1997). Intra-arterial vector application with or without disruption of the blood-brain-barrier (BBB) or the bloodtumor-barrier (BTB) seems to offer a solution to the difficulties of vector distribution by employing the extensive tumor neovasculature for transgene delivery to all vascularized tumor foci (Muldoon et al., 1997). In contrast to the normal BBB, which consists of endothelial cells bound together with tight junctions and wrapped by astrocytic processes, and which limits the entry of substances into the interstitial and intracellular space of the brain, the brain tumor neovasculature has a somewhat more permeable barrier (Cox et al., 1976; Inamura and Black, 1994; Long, 1979; Yamada et al., 1982). Although the BTB may be more or less leaky, it still limits delivery of high molecular weight substances to tumor tissue and to immediately adjacent, partially tumor-infiltrated areas of the brain (Groothuis et al., 1991). The varying permeability found throughout the BTB in the majority of malignant brain tumors (BergstrĂśm et al., 1983; Burger et al., 1988) may restrict vector penetration, especially those with a larger size (HSV = 150 nm in diameter, AV = 70100 nm). On the other hand, the existence of a tight BBB throughout the normal brain provides an advantage in

A. Intravascular vector delivery without BTB modulation The average size of a HSV particle is about 150 nm and that of an AV particle is 70-100 nm, and because they are so large, their penetration through normal brain capillaries or tumor neocapillaries is poor (Rainov et al., 1995). There are only a few studies being done with virus or non-virus vectors injected intravascularly without modulation of the BTB or the BBB. Chauvet et al. (1998) injected AV vectors into the middle carotid artery (MCA) of a dog with a benign intracranial meningioma and were able to achieve a high percentage of transduced tumor cells without any concomitant toxic effects to the CNS. Meningioma, however, unlike astrocytoma or glioblastoma, have excessive fenestration and leakiness of tumor capillaries, which probably facilitate virus vector entry (McDermott and Wilson, 1996). Our studies (Rainov et al., 1995) and those of other investigators (Neuwelt et al., 1991; Nilaver et al., 1995) have demonstrated increased transduction rates of tumor cells by HSV or AV after osmotic or pharmacologic barrier disruption, as compared to delivery of vectors across the intact BBB and BTB. Delivery studies of HSV particles and MION across the unmodified BBB and the BTB have shown that there is a small percentage (2-5%) of 9L gliosarcoma cells which can be targeted with HSV, presumably through the somewhat leaky BTB in 9L tumors, or by secondary spread from infected endothelial cells in neocapillaries producing replication-conditional vectors (Rainov et al., 1995). While HSV particles tend to infect cells in the periphery of the tumor and at the tumor/brain border, MION are delivered throughout the tumor and accumulate to some extent in the tumor center

6

Gene Therapy and Molecular Biology Vol 3, page 7

Fig. 2: Visual comparison of ß-gal expression in large and small 9L tumor foci in syngeneic Fischer rats injected with replicationdeficient AV and liposome-DNA complexes (lipoDNA) bearing the lacZ gene with and without blood-tumor-barrier disruption by bradykinin (BK). A. Photomicrograph of a small intracerebral tumor focus (< 0.5 mm) in the AV group 48 hours after ipsilateral intracarotid virus injection in the absence of BK. Note the relatively high number of X-gal stained cells (blue, arrows) in the tumor periphery and, to a certain extent, in the tumor center (T), B = tumor-infiltrated surrounding brain (magnification 300x, 20 µm frozen section, counterstained with hematoxylin). B. Photomicrograph of an intracerebral tumor focus in the AV group 48 hours after intra-carotid BK infusion and AV vector injection. An increased number of stained cells is distributed somewhat more evenly throughout the tumor (T), B = normal brain (magnification 200x, 20 µm frozen section, counterstained with hematoxylin). C. Photomicrograph of a tumor in the lipoDNA group 48 hours after vector injection in the absence of BK. Note the high number of X-gal stained cells (blue, arrowheads) throughout the tumor (T). Endothelial cells in capillaries (V) near the tumor/brain border are also stained positively (arrows), B = normal brain (magnification 200x, 20 µm frozen section, counterstained with NeutralRed). D. Photomicrograph of a tumor in the lipoDNA group 48 hours after intra-carotid BK infusion and vector injection. The number of X-gal stained cells (blue, arrows) throughout the tumor (T) is somewhat higher than in the absence of BK, V = tumor vessel, B = normal brain (magnification 200x, 20 µm frozen section, counterstained with NeutralRed).

products in this tumor model is particularly high in small tumor foci (<0.5 mm) away from the main tumor mass. In these foci, almost half of all tumor cells are transduced. A few endothelial cells in normal brain capillaries are also transduced by AV-mediated gene transfer. Intracarotid delivery of non-viral vectors, such as liposome-plasmid

(Rainov et al., 1995). Intra-carotid delivery of AV vector to 9L rat gliosarcoma without BTB disruption results in transgene expression in 3-10% of tumor cells, predominantly located at the tumor-brain border, as well as randomly distributed throughout the tumor (Fig. 2A) (Rainov et al., in press). Virus-mediated expression of marker gene 7

Gene Therapy and Molecular Biology Vol 3, page 8 DNA-complexes (lipoDNA), without BTB disruption renders more than 30% of the tumor cells positive for the marker gene (Rainov et al., in press). The pattern of distribution is homogenous throughout the tumor, with a slightly higher transduction rate in the tumor periphery (Fig. 2C). Although lipoDNA-mediated gene transfer without barrier modification has increased efficacy as compared to HSV- and AV-mediated gene transfer, it is less tumor-specific, since a considerable number of endothelial and glial cells also express the respective transgene. LipoDNA complexes represent alternative vehicles for gene transfer and avoid some of the unwanted features of virus vectors (Hug and Sleight, 1991). The route of in vivo administration may affect dramatically the uptake of liposomes by normal and tumor cells. Intravenously injected liposomes are taken up mainly by the reticulo-endothelial system (RES), particularly in the liver and spleen (Hug and Sleight, 1991). Intra-arterial application of lipoDNA to tumors has not been investigated extensively, and little is known about transduction efficiency in brain tumors in vivo (Gennuso et al., 1993).

high efficiency of osmotic barrier disruption. HSV and AV vector delivery to brain and intracerebral tumors was increased up to four fold by hypertonic mannitol (Neuwelt et al., 1991b; Nilaver et al., 1995). When virus was administered intra-arterially without barrier modification, virtually no infection was detected of either tissue type. MION can penetrate efficiently through the disrupted BBB in rats, and have the advantage of being imageable by MRI. After intra-arterial mannitol infusion, glial cells were predominantly infected by AV, while HSV and MION targeted neurons more efficiently (Muldoon et al., 1998). The degree of barrier opening correlated with the transduction efficiency of glial and neuronal cells (Doran et al., 1995). Osmotic BBB disruption in combination with intra-arterial administration of viral vectors may offer a method of global delivery to treat disseminated brain tumors (Nilaver et al., 1995), although its specificity is far from optimal.

C. Intravascular vector delivery with pharmacological BBB and BTB disruption Vasoactive agents for modification of the BBB and BTB have been identified through studies of peritumoral brain edema and effects on systemic capillaries (Black, 1992; Chan et al., 1983; Cloughesy and Black, 1995). The BTB can develop transient increases in permeability with the intra-arterial delivery of vasoactive agents, while the normal BBB resists the effects of these compounds because of additional biochemical and physical barriers (Inamura and Black, 1994). Vasoactive compounds, including leukotrienes (Black and Chio, 1992; Chio et al., 1992), bradykinin (BK) and its analog RMP-7 (Barnett et al., 1998; Black et al., 1997; Doctrow et al., 1994; Elliott et al., 1996a and 1996b; Inamura et al., 1994a; Matsukado et al., 1996; Nakano et al., 1996; Rainov et al., 1995; Rainov et al., 1998), histamine (Inamura et al., 1994b; Nomura et al., 1994), and calcium antagonists (Matsukado et al., 1994) appear to selectively increase permeability in abnormal brain tumor capillaries.

B. Intravascular vector delivery with osmotic BBB and BTB disruption BBB and BTB can be manipulated to increase permeability for gene therapy vectors, such as viruses or non-viral particles. Several studies have focused on transient osmotic disruption of the BBB and the BTB, and this technique has been well characterized in animal models and in humans as an enhancer of chemotherapeutic drugs and vector delivery to brain tumors (Doran et al., 1995; Neuwelt et al., 1987; Neuwelt et al., 1991b; Nilaver et al., 1995; Z端nkeler et al., 1996). The mechanism of osmotic disruption of the barrier includes shrinkage of endothelial cells with subsequent opening of the capillary tight junctions, which is achieved by application of hypertonic solutions of sugars or salts into the arterial system (Rapoport and Robinson, 1986). Infusion of mannitol is most commonly used because of its relatively low toxicity and the applicability to humans (Muldoon et al., 1998). Mannitol offers the possibility of global delivery of drugs and virus vectors throughout the vasculature, which can reach even infiltrating tumor foci distant to the main mass (Neuwelt and Hill, 1987; Neuwelt et al., 1991a and b). With mannitol disruption of the BBB and BTB, however, delivery and uptake of therapeutic agents is less specific to the tumor and tends to spread the toxic agents throughout the whole affected hemisphere, which may increase toxicity to normal brain tissue (Z端nkeler et al., 1996).

BK, a nonapeptide hormone with peripheral vasodilatation effect, permeabilizes the vascular endothelium in brain capillaries at low concentrations (10 g/kg/min), when delivered intra-arterially in rodents, and its barriermodifying effects are specific to brain tumor neocapillaries (Inamura and Black, 1994). BK exerts its effects by interaction with specific B2 receptors (Hess et al., 1992) on endothelial cells, which mediate contraction of the endothelial cell cytoskeleton with subsequent temporary opening of the tight junctions (Doctrow et al., 1994; Inamura et al., 1994a; Sanovich et al., 1995) and may also increase the rate of pinocytosis/transcytosis in endothelial cells (Raymond et al., 1986). It has also been

Studies of delivery of virus and non-virus vector particles across the BBB and BTB have demonstrated the

8

Gene Therapy and Molecular Biology Vol 3, page 9 demonstrated that BK and RMP-7 increase intracellular free calcium levels (Doctrow et al., 1994) and stimulate a nitric oxide-mediated pathway in tumor vasculature and/or in tumor cells itself (Nakano et al., 1996).

transduce brain tumor cells, and that BTB modification by BK further increases the number of transgene-expressing tumor cells without apparent adverse effects (Rainov et al., in press). With AV and lipoDNA, it remains to be determined whether the increase in transduction following BK infusion will result in long-term survival in experimental brain tumor models.

BTB disruption by low-dose BK can facilitate selective uptake of HSV vectors administered through the carotid artery to single or multiple tumor foci in the rodent brain, with essentially no infection of normal neurons and glia (Rainov et al., 1995). Transgene expression after intra-arterial BK infusion and HSV vector bolus injection is particularly intense in the periphery of the tumor, a zone with distinct biological and biomechanical properties such as high mitotic rate, angiogenesis, parenchymal invasion, and low interstitial pressure (Boucher et al., 1996). Up to 25% of tumor cells in this region express transgene proteins after BK/HSV administration, as compared to less than 0.1% of cells in normal brain tissue (Fig. 1E and F). In contrast, MION uptake is increased by BK predominantly in the tumor center and has less effect at the infiltrating edge (Rainov et al., 1995). Furthermore, this study demonstrated HSV infection of multiple bilateral tumor nodules by unilateral BK infusion and HSV injection, which suggests that BK may have generalized effects beyond the site of infusion in rat brain with extensive collateralization.

To replace BK with a new, longer and more selectively acting derivative suitable for human studies, the synthetic nonapeptide RMP-7, H-Arg 1-Pro2-HydroxyPro3-Gly4-Thi5Ser6-Pro7-Tyr(Me)-!-(CH2NH)8-Arg9-OH, a BK analog with three amino acid substitutions, was designed by Alkermes, Inc. (Cambridge, MA). It is more resistant to angiotensin-converting enzyme (ACE) due to the replacement of alanine with 2-thienyl-alanine (Thi 5) and to neutral endopeptidase and carboxypeptidase I, due to replacement of phenylalanine with the Tyr(Me)- !(CH2NH)8 group (Elliot et al, 1996a). Replacement of proline with hydroxyproline (HydroxyPro3) removes the undesirable action on the B1 receptor. RMP-7 has a 5-10 times longer half-life in the blood circulation than BK, is 100-fold more potent in mice, and acts more selectively on endothelial cells by binding to the B2 receptor only, without undesirable blood pressure drops (Elliot et al., 1996a and b). RMP-7 has been FDA-approved for human studies.

The increased rate of tumor infection by HSV after BK infusion has been exploited for eradication of intracerebral 9L tumors in syngeneic rats. In this model, virus vector concentration appears to influence survival rates in a dosedependent fashion when GCV is given systemically starting three days after BK/HSV application and continued for 14 days (Rainov et al., 1998). A concentration of 1x1010 pfu HSV was able to eradicate tumors in 80% of the treated animals, while 1x109 pfu eliminated tumors in 40% of the rats, and 1x108 pfu was sufficient for prolonged survival, but not for permanent tumor cures. No apparent complications of intra-arterial HSV injection were encountered in this study (Rainov et al., 1998).

RMP-7 was tested extensively as an adjunctive therapeutic agent for primary and recurrent malignant gliomas. In humans, intra-carotid or intravenous infusion of low-dose RMP-7 (0.1 g/kg/min for 15 min) was able to increase the delivery to brain tumors of intravenously injected low and high molecular weight tracers, such as aminoisobutyric acid or dextrane, and of cytotoxic agents, such as methotrexate and carboplatin (Elliott et al., 1996b; Matsukado et al., 1996; Muldoon et al., 1998). There are data to suggest that RMP-7 is at least equivalent to BK in terms of enhancement of virus and non-viral particles to brain tumors (Barnett et al., 1998). This study compared BK and RMP-7 and, among other findings, demonstrated no significant difference in the enhancement of HSV delivery across the BTB, which confirms the potential of RMP-7 for application for gene therapy of human brain tumors (Barnett et al., 1998).

In another study, intra-carotid delivery of AV and lipoDNA to 9L rat gliosarcoma with and without BKmediated BTB modification was compared (Rainov et al., in press). For AV-mediated gene transfer, BK infusion increased the amount of transgene-expressing tumor cells from 5 to 19 % (Fig. 2B) and enhanced expression in the center of larger tumor foci. BK infusion prior to lipoDNA injection was able to increase the number of transduced tumor cells from 30% to more than 50%, and to produce a more homogeneous pattern of transgene distribution in the tumor (Fig. 2D). The relatively low tumor specificity of lipoDNA transfer remains unchanged by BK application, with extensive delivery to normal tissue as well. These findings indicate that intra-carotid application of virus and non-virus vectors can preferentially and effectively

In conclusion, the issues of delivery of gene therapy vectors to tumors in the brain seem still to be underappreciated in the literature. Unfortunately, the development of new and more specific virus and non-virus vectors does not address the difficulties of accessing the maximum number of tumor cells in diffuse or invasive and multifocal tumors, such as human GBM. Since there are no ideal vectors or unconditionally efficient application modes or delivery routes, in the future, gene therapeutic approaches for brain tumors will be a combination of different application routes, vectors and transgene 9

Gene Therapy and Molecular Biology Vol 3, page 10 Black KL, Cloughesy T, Huang SC, Govin YP, Zhou Y, Grous J, Nelson G, Fraahari K, Hoh CK, and Phelps M (1997) Intracarotid infusion of RMP-7, a bradykinin analog and transport of gallium-68 ethylenediamine tetra-acetic acid into human gliomas. J Neurosurg 86, 603-609.

combinations, designed to account for the individual features of different tumor types. The intravascular route of vector delivery should gain a much higher popularity and will be aided either by direct intratumoral injections, as in the case of adult multifocal GBM, or by widespread CSF vector delivery as in the case of pediatric brain tumors.

Bobo RH, Laske DW, Akbasak A, Morrison PF, Dedrick RL, and Oldfield EH (1994) Convection-enhanced delivery of macromolecules in the brain. Proc Natl Acad Sci USA 91, 2076-2080.

Acknowledgements

Boucher Y, Leunig M, and Jain RK (1996) Tumor angiogenesis and interstitial hypertension. Cancer Res 56, 4264-4266.

These studies were supported in part by grant 015VE1997 from the State Ministry of Culture and Education of Saxony-Anhalt, Germany, to NGR, and NCI grant CA69246 to XOB

Boviatsis EJ, Chase M, Wei M, Tamiya T, Hurford RK Jr, Kowall NW, Tepper RI, Breakefield XO, and Chiocca EA (1994a) Gene transfer into experimental brain tumors mediated by adenovirus, herpes-simplex-virus (HSV), and retrovirus vectors. Hum Gene Ther 5, 183-191.

References

Boviatsis EJ, Park JS, Sena-Esteves M, Kramm CM, Chase M, Efird JT, Wei MX, Boviatsis EJ, Scharf JM, Chase M, Harrington K, Kowall NW, Breakefield XO, and Chiocca EA (1994b) Antitumor activity and reporter gene transfer into rat brain neoplasms inoculated with herpes simplex virus vectors defective in thymidine kinase or ribonucleotide reductase. Gene Ther 1, 323-331, 1994

Anderson WF, Miller AD, Bordignon C (eds) (1995) ORDA reports. Recombinant DNA advisory committee (RAC). Data management report - December 1994. Hum Gene Ther 6, 535-548. Andreansky SS, He B, Gillespie GY, Soroceanu L, Markert J, Chou J, Roizman B, and Whitley RJ (1996) The application of genetically engineered herpes simplex viruses to the treatment of experimental brain tumors. Proc Natl Acad Sci USA 93, 11313-11318.

Boviatsis EJ, Park JS, Sena-Esteves M, Kramm CM, Chase M, Efird JT, Wei MX, Breakefield XO, and Chiocca EA (1994c) Long-term survival of rats harboring brain neoplasms treated with ganciclovir and a herpes simplex virus vector that retains an intact thymidine kinase gene. Cancer Res 54, 5745-5751.

Badie B, Hunt K, Economou JS, and Black KL (1994) Stereotactic delivery of a recombinant adenovirus into a C6 glioma cell line in a rat brain tumor model. Neurosurgery 35, 910-915.

Bramson JL, Hitt M, Gauldie J, and Graham FL (1997) Preexisting immunity to adenovirus does not prevent tumor regression following intratumoral administration of a vector expressing IL-12, but inhibits virus dissemination. Gene Ther 4, 1069-1076.

Badie B, Kramar MH, Lau R, Boothman DA, Economou JS, and Black KL (1998) Adenovirus-mediated p53 gene delivery potentiates the radiation-induced growth inhibition of experimental brain tumors. J Neuro-oncol 37, 217-222.

Burger PC, Heinz ER, Shibata T, and Kleihues P (1988) Topographic anatomy and CT correlations in the untreated glioblastoma multiforme. J Neurosurg 68, 698-704.

Bajocchi G, Feldman SH, Crystal RG, and Mastrangeli A (1993) Direct in vivo gene transfer to ependymal cells in the central nervous system using recombinant adenovirus vectors. Nat Genet 3, 229-234.

Chan PH, Fishman RA, Caronna J, Schmidley JW, Prioleau G, and Lee J (1983) Induction of brain edema following intracerebral injections of arachidonic acid. Ann Neurol 13, 625-632.

Barnett FH, Rainov NG, Ikeda K, Shuback DE, Elliott P, Kramm CM, Chase M, Qureshi N, Harsh G IV, Chiocca EA, and Breakefield XO (1998) Selective delivery of herpes virus vectors to experimental brain tumors using RMP-7. Cancer Gene Ther, in press.

Chauvet AE, Kesava PP, Goh CS, and Badie B (1998) Selective intraarterial gene delivery into a canine meningioma. J Neurosurg 88, 870-873.

Bergstrรถm M, Collins VP, Ehrin E, Ericson K, Eriksson L, Greitz T, Halldin C, Von Holst H, Langstrรถm B, Lilj A, Lundquist H, and Nagren K (1983) Discrepancies in brain tumor extent as shown by computed tomography and positron emission tomography using [68Ga] EDTA, [11C]glucose, and [11C]methionine. J Comput Assist Tomogr 7, 1062-1066.

Chen SH, Shine HD, Goodman JC, Grossman RG, and Woo SL (1994) Gene therapy for brain tumors: regression of experimental gliomas by adenovirus-mediated gene transfer in vivo. Proc Natl Acad Sci USA 91, 3054-3057. Chio CC, Baba T, and Black KL (1992) Selective blood-tumor barrier disruption by leukotrienes. J Neurosurg 77, 407-410.

Black KL (1992) Biochemical opening of the blood-brain barrier. Adv Drug Rev 15, 37-52.

Cloughesy TF, and Black KL (1995) Pharmacological bloodbrain barrier modification for selective drug delivery. J Neuro-oncol 26, 125-132.

Black KL, and Chio CC (1992) Increased opening of bloodtumor barrier by leukotriene C4 is dependent on size of molecules. Neurol Res 14, 402-404.

Cox DJ, Pilkington GJ, and Lantos PL (1976) The fine structure of blood vessels in ethylnitrosurea-induced tumours of the rat

10

Gene Therapy and Molecular Biology Vol 3, page 11 nervous system: with special reference to the breakdown of the blood-brain-barrier. Br J Exp Pathol 57, 419-430.

Huard J, Lochmuller H, Acsadi G, Jani A, Massie B, and Karpati G (1995) The route of administration is a major determinant of the transduction efficiency of rat tissues by adenoviral recombinants. Gene Ther 2, 107-115.

Culver KW, Ram Z, Wallbridge S, Ishii H, Oldfield EH, and Blaese RM (1992) In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors. Science 256, 1550-1552.

Hug P, and Sleight RG (1991) Liposomes for the transformation of eukaryotic cells. Biochim Biophys Acta 1097, 1-17.

Doctrow SR, Abelleira SM, Curry LA, Heller-Harrison R, Kozarich JW, Malfroy B, McCarroll LA, Morgan KG, Morrow AR, Musso GF, Smart JL, Straub JA, Turnbull B, and Gloff CA (1994) The bradykinin analog RMP-7 increases intracellular free calcium levels in rat brain microvascular endothelial cells. J Pharmacol Exp Ther 271, 229-237.

Inamura T, and Black KL (1994) Bradykinin selectively opens blood-tumor barrier in experimental brain tumors. J Cereb Blood Flow Metab 14, 862-870.

Doran SE, Dan Ren X, Betz AL, Pagel MA, Neuwelt EA, Roessler BJ, and Davidson BL (1995) Gene expression from recombinant viral vectors in the CNS following blood-brain barrier disruption. Neurosurgery 36, 965-970.

Inamura T, Nomura T, Ikezaki K, Fukui M, Pollinger G, and Black K (1994b) Intracarotid histamine infusion increases blood tumor permeability in RG2 glioma. Neurol Res 16, 125-128.

Eck SL, Alavi JB, Alavi A, Davis A, Hackney D, Judy K, Mollman J, Phillips PC, Wheeldon EB, and Wilson JM (1996) Treatment of advanced CNS malignancies with the recombinant adenovirus H5.010RSVTK: a phase I trial. Hum Gene Ther 7, 1465-1482.

Izquierdo M, Cortes M, de Felipe P, Martin V, Diez-Guerra J, Talavera A, and Perez-Higueras A (1995) Long-term rat survival after malignant brain tumor regression by retroviral gene therapy. Gene Ther 2, 66-69.

Inamura T, Nomura T, Bartus RT, and Black KL (1994a) Intracarotid infusion of RMP-7, a bradykinin analog: A method for selective drug delivery to brain tumors. J Neurosurg 81, 752-758.

Izquierdo M, Cortes ML, Martin V, de Felipe P, Izquierdo JM, Perez-Higueras A, Paz JF, Isla A, and Blazquez MG (1997) Gene therapy in brain tumours: implications of the size of glioblastoma on ist curability. Acta Neurochir Suppl 68, 111-117.

Elliott PJ, Hayward NJ, Dean RL, Blunt DG, and Bartus RT (1996) Intravenous RMP-7 selectively increases uptake of carboplatin into rat brain tumors. Cancer Res 56, 39984005. Elliott PJ, Hayward NJ, Huff MR, Nagle TL, Black KL, and Bartus RT (1996) Unlocking the blood-brain barrier: A role for RMP-7 in brain tumor therapy. Exp Neurol 141, 214224.

Izquierdo M, Martin V, de Felipe P, Izquierdo JM, PerezHigueras A, Cortes ML, Paz JF, Isla A, and Blazquez MG (1996) Human malignant brain tumor response to herpes simplex thymidine kinase (HSVtk)/ganciclovir gene therapy. Gene Ther 3, 491-495.

Gennuso R, Spigelman MK, Chinol M, Zappulla RA, Nieves J, Vallabhajosula S, Alberto-Paciucci P, Goldsmith SJ, and Holland JF (1993) Effect of blood-brain barrier and bloodtumor barrier modification on central nervous system liposomal uptake. Cancer Invest 11, 118-128.

Jain RK (1987) Transport of molecules in the tumor interstitium: a review. Cancer Res 47, 3039-3051. Jain RK (1994) Barriers to drug delivery in solid tumors. Sci Am 271, 58-65.

Groothuis DR, Vriesendorp FJ, Kupfer B, Warnke PC, Lapin GD, Kuruvilla A, Vick NA, Mikhael MA, and Patlak CS (1991) Quantitative measurements of capillary transport in human brain tumors by computed tomography. Ann Neurol 30, 581-588.

Kaplitt MG, Tjuvajev JG, Leib DA, Berk J, Pettigrew KD, Posner JB, Pfaff DW, Rabkin SD, and Blasberg RG (1994) Mutant herpes simplex virus induced regression of tumors growing in immunocompetent rats. J Neuro-oncol 19, 137147.

Harsh GR, Hochberg F, Louis DN, Silver JS, Qureshi N, Bankiewicz K, Breakefield XO, and Chiocca EA Implantation of TK-retrovirus producer cells into human recurrent malignant gliomas: assessment of gene transfer eeficiency and producer cell permanence. Manuscript in preparation.

Kelly PJ (1997) Stereotactic procedures neurosurgery. Exp Neurol 144, 157-159.

for

molecular

Kramm CM, Chase M, Herrlinger U, Jacobs A, Pechan PA, Rainov NG, Sena-Esteves M, Aghi M, Barnett FH, Chiocca EA, and Breakefield XO (1997) Therapeutic efficiency and safety of a second-generation replication-conditional HSV-1 vector for brain tumor gene therapy. Hum Gene Ther 8, 2057-2068.

Herrlinger U, Kramm CM, Aboody-Guterman KS, Silver JS, Ikeda K, Johnston KM, Pechan PA, Barth RF, Finkelstein D, Chiocca EA, Louis DN, and Breakefield XO (1998) Preexisting herpes simplex virus 1 (HSV-1) immunity decreases, but does not abolish, gene transfer to experimental brain tumors by a HSV-1 vector. Gene Ther 5, 809-819.

Kramm CM, Rainov NG, Sena-Esteves M, Barnett FH, Chase M, Herrlinger U, Pechan PA, Chiocca EA, and Breakefield XO (1996b) Long-term survival in a rodent model of disseminated brain tumors by combined intrathecal delivery of herpes vectors and ganciclovir treatment. Hum Gene Ther 7, 1989-1994.

Hess JF, Borkowski JA, Young GS, Strader CD, and Ransom RW (1992) Cloning and pharmacological characterization of a human bradykinin (BK-2) receptor. Biochem Biophys Res Comm 184, 260-268.

Kramm CM, Rainov NG, Sena-Esteves M, Chase M, Pechan PA,

11

Gene Therapy and Molecular Biology Vol 3, page 12 Chiocca EA, and Breakefield XO (1996a) Herpes vectormediated delivery of marker genes to disseminated central nervous system tumors. Hum Gene Ther 7, 291-300.

Saunders Comp, Philadelphia, pp 2782-2825. Mineta T, Rabkin SD, and Martuza RL (1994) Treatment of malignant gliomas using ganciclovir-hypersensitive, ribonucleotide reductase-deficient herpes simplex viral mutant. Cancer Res 54, 3963-3966.

Kramm CM, Sena-Esteves M, Barnett FH, Rainov NG, Schuback DE, Yu JS, Pechan PA, Paulus W, Chiocca EA, and Breakefield XO (1995) Gene therapy for brain tumors. Brain Pathol 5, 345-381.

Mineta T, Rabkin SD, Yazaki T, Hunter WD, and Martuza RL (1995) Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. Nat Med 1, 938-943.

Kroll RA, and Neuwelt EA (1998) Outwitting the blood-brain barrier for therapeutic purposes: osmotic opening and other means. Neurosurgery 42, 1083-1099.

Mizuno M, Yoshida J, Colosi P, and Kurtzman G (1998) Adenoassociated virus vector containing the herpes simplex virus thymidine kinase gene causes complete regression of intracerebrally implanted human gliomas in mice, in conjunction with ganciclovir administration. Jpn J Cancer Res 89, 76-80.

Kroll RA, Pagel MA, Muldoon LL, Roman-Goldstein S, and Neuwelt EA (1996) Increasing volume of distribution to the brain with interstitial infusion: dose, rather than convection, might be the most important factor. Neurosurgery 38, 746752.

Moolten FL (1986) Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: paradigm for a prospective cancer control strategy. Cancer Res 46, 52765281.

Lal B, Indurti RR, Couraud PO, Goldstein GW, and Laterra J (1994) Endothelial cell implantation and survival within experimental gliomas. Proc Natl Acad Sci USA 91, 96959699. Le Gal La Salle G, Robert JJ, Berrard S, Ridoux V, StratfordPerricaudet LD, Perricaudet M, and Mallet J (1993) An adenovirus vector for gene transfer into neurons and glia in the brain. Science 259, 988-990.

Muldoon LL, Kroll RA, Pagel MA, Roman-Goldstein S, and Neuwelt EA (1997) Delivery of therapeutic genes to brain and intracerebral tumors. In Chiocca EA, and Breakefield XO (eds): Gene Therapy for Neurological Disorders and Brain Tumors. Boston: Humana Press, pp 128-139.

Leunig M, Yuan F, Menger MD, Boucher Y, Goetz AE, Messmer K, and Jain RK (1992) Angiogenesis, microvascular architecture, microhemodynamics, and interstitial fluid pressure during early growth of human adenocarcinoma LS174T in SCID mice. Cancer Res, 65536560.

Muldoon LL, Nilaver G, Kroll RA, Pagel MA, Breakefield XO, Chiocca EA, Davidson BL, Weissleder R, and Neuwelt EA (1995) Comparison of intracerebral inoculation and osmotoic blood-brain barrier disruption for delivery of adenovirus, herpesvirus and iron oxide nanoparticles to normal rat brain. Am J Pathol 147, 1840-1851.

Lieberman DM, Laske DW, Morrison PF, Bankiewicz KS, and Oldfield EH (1985) Convection-enhanced distribution of large molecules in gray matter during interstitial drug infusion. J Neurosurg 82, 1021-1029.

Nakano S, Matsukado K, and Black KL (1996) Increased brain tumor microvessel permeability after intracarotid bradykinin infusion is mediated by nitric oxide. Cancer Res 56, 40274031.

Long DM (1979) Capillary ultrastructure in human metastatic brain tumors. J Neurosurg 51, 53-58.

Neuwelt EA, and Hill SA (1987) Chemotherapy administered in conjunction with osmotic blood-brain barrier modification in patients with brain metastases. J Neuro-oncol 4, 195-207.

Maron A, Gustin T, Le Roux A, Mottet I, Dedieu JF, Brion JP, Demeure R, Perricaudet M, and Octave JN (1996) Gene therapy of rat C6 glioma using adenovirus-mediated transfer of the herpes simplex virus thymidine kinase gene: long-term follow-up by magnetic resonance imaging. Gene Ther 3, 315-322.

Neuwelt EA, Goldman D, Dahlborg SA, Crossen J, Ramsey F, Goldstein SM, Braziel R, and Dana B (1991a) Primary CNS lymphoma treated with osmotic blood-brain barrier disruption: prolonged survival and preservation of cognitive functions. J Clin Oncol 9, 1580-1590.

Martuza RL, Malick A, Markert JM, Ruffner KL, and Coen DM (1991) Experimental therapy of malignant glioma by means of a genetically engineered virus mutant. Science 252, 854856.

Neuwelt EA, Pagel MA, and Dix RD (1991b) Delivery of ultraviolet inactivated 35S-herpesvirus across an osmotically modified blood-brain barrier. J Neurosurg 74, 475-479. Nilaver G, Muldoon LL, Kroll RA, Pagel MA, Breakefield XO, Davidson BL, and Neuwelt EA (1995) Delivery of herpesvirus and adenovirus to nude rat intracerebral tumors after osmotic blood-brain barrier disruption. Proc Natl Acad Sci USA 92, 9829-9833.

Matsukado K, Inamura T, Nakano S, Fukui M, Bartus RT, and Black K (1996) Enhanced tumor uptake of carboplatin and survival in glioma-bearing rats by intracarotid infusion of bradykinin analog, RMP-7. Neurosurgery 39, 125-134. Matsukado K, Nomura T, Ikezaki K, and Fukui M (1994) Selective increase in blood-tumor barrier permeability by calcium antagonists in transplanted rat brain tumors. Acta Neurochir Suppl 60, 403-405.

Nomura T, Ikezaki K, Matsukado K, and Fukui M (1994) Effect of histamine on the blood-tumor barrier in transplanted rat brain tumors. Acta Neurochir Suppl 60, 400-402. Okada H, Miyamura K, Itoh T, Hagiwara M, Wakabayashi T, Mizuno M, Colosi P, Kurtzman G, Yoshida J (1996) Gene therapy against an experimental glioma using adeno-

McDermott MW, and Wilson CB (1996) Meningiomas. In: Youmans JR (ed) Neurological surgery. 4th ed., WB

12

Gene Therapy and Molecular Biology Vol 3, page 13 associated virus vectors. Gene Ther 3, 957-964.

arterially to experimental brain neoplasms. Hum Gene Ther 6, 1543-1552.

Oldfield EH, Ram Z, Chiang Y, and Blaese RM (1995) Intrathecal gene therapy for the treatment of leptomeningial carcinomatosis. GTI 0108. A phase I/II study. Hum Gene Ther 6, 55-85.

Ram Z, Culver KW, Oshiro EM, Viola JJ, DeVroom HL, Otto E, Long Z, Chiang Y, McGarrity GJ, Muul LM, Katz D, Blaese RM, and Oldfield EH (1997) Therapy of malignant brain tumors by intratumoral implantation of retroviral vectorproducing cells. Nat Med 3, 1354-1361.

Oldfield EH, Ram Z, Culver KW, Blaese RM, DeVroom HL, and Anderson WF (1993) Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Hum Gene Ther 4, 39-69.

Ram Z, Culver KW, Walbridge S, Blaese RM, and Oldfield EH (1993) In situ retroviral-mediated gene transfer for the treatment of brain tumors in rats. Cancer Res 53, 83-88.

Ooboshi H, Rios CD, and Heistad DD (1997) Novel methods for adenovirus-mediated gene transfer to blood vessels in vivo. Mol Cell Biochem 172, 37-46.

Ram Z, Walbridge S, Shawker T, Culver KW, Blaese RM, and Oldfield EH (1994) The effect of thymidine kinase transduction and ganciclovir therapy on tumor vasculature and growth of 9L gliomas in rats. J Neurosurg 81, 256-260.

Ooboshi H, Welsh MJ, Rios CD, Davidson BL, and Heistad DD (1995) Adenovirus-mediated gene transfer in vivo to cerebral blood vessels and perivascular tissue. Circ Res 77, 7-13.

Ram Z, Wallbridge S, Oshiro EM, Viola JJ, Chiang Y, Meuller SN, Blaese RM, and Oldfield EH (1994) Intrathecal gene therapy for malignant leptomeningial neoplasia. Cancer Res 54, 2141-2145.

Oshiro EM, Viola JJ, Oldfield EH, Walbridge S, Bacher J, Frank JA, Blaese RM, and Ram Z (1995) Toxicity studies and distribution dynamics of retroviral vectors following intrathecal administration of retroviral vector-producer cells. Cancer Gene Ther 2, 87-95.

Rapoport SI, and Robinson PJ (1986) Tight-junctional modification as the basis of osmotic opening of the bloodbrain barrier. Ann NY Acad Sci 481, 250-267.

Perez-Cruet MJ, Trask TW, Chen SH, Goodman JC, Woo SL, Grossman RG, and Shine HD (1994) Adenovirus-mediated gene therapy of experimental gliomas. J Neurosci Res 39, 506-511.

Raymond JJ, Robertson DM, and Dinsdale HB (1986) Pharmacological modification of bradykinin induced breakdown of the blood-brain barrier. Can J Neurol Sci 13, 214-220.

Puumalainen A, Vapalahti M, Agrawal RS, Kossila M, Laukkanen J, Lehtolainen P, Viita H, Paljärvi L, Vanninen R, and Yla-Herttuala S (1998) Ă&#x;-galactosidase gene transfer to human malignant glioma in vivo using replicationdeficient retroviruses and adenoviruses. Hum Gene Ther 9, 1769-1774.

Rosenfeld MR, Bergman I, Schramm L, Griffin JA, Kaplitt MG, and Meneses PI (1997) Adeno-associated viral vector gene transfer into leptomeningial xenografts. J Neuro-oncol 34, 139-144. Russel DW, Berger MS, and Miller AD (1995) The effects of human serum and cerebrospinal fluid on retroviral vectors and packaging cell lines. Hum Gene Ther 6, 635-641.

Raffel C, Culver K, Kohn D, Nelson M, Siegel S, Gillis F, Link CJ, Villablanca JG, and Anderson WF (1994) Gene therapy for the treatment of recurrent pediatric malignant astrocytomas with in vivo transduction with the herpes simplex thymidine kinase gene/ganciclovir system. Hum Gene Ther 5, 863-890.

Sanovich E, Bartus RT, Friden PM, Dean RL, Le HQ, and Brightman MW (1995) Pathway across blood-brain barrier opened by the bradykinin agonist, RMP-7. Brain Res 705, 125-135. Shen T, Weissleder R, Papisov M, Bogdanov A, and Brady TJ (1993) Monocrystalline iron oxide nanocompounds (MION): physicochemical properties. Magn Res Med 29, 599-604.

Rainov NG, Dobberstein KU, Heidecke V, Dorant U, Chase M, Kramm CM, Chiocca EA, and Breakefield XO (1998) Longterm survival in a rodent brain tumor model by bradykininenhanced intra-arterial delivery of a therapeutic herpes simplex virus vector. Cancer Gene Ther 5, 158-162.

Short MP, Choi BC, Lee JK, Malick A, Breakefield XO, and Martuza RL (1990) Gene delivery to glioma cells in rat brain by grafting of a retrovirus packaging cell line. J Neurosci Res 27, 427-439, 1990

Rainov NG, Ikeda K, Qureshi N, Grover S, Herrlinger U, Pechan P, Chiocca EA, Breakefield XO, and Barnett FH (1998) Intra-arterial delivery of adenovirus vectors and liposomeDNA complexes to experimental brain neoplasms. Hum Gene Ther, submitted.

Spear MA, Herrlinger U, Rainov NG, Pechan P, Weissleder R, and Breakefield XO (1998) Targeting gene therapy vectors for CNS malignancies. J Neurovirol 4, 133-147. Takamiya Y, Short MP, Moolten FL, Fleet C, Mineta T, Breakefield XO, and Martuza RL (1993) An experimental model of retrovirus gene therapy for malignant brain tumors. J Neurosurg 79, 104-110.

Rainov NG, Kramm CM, Aboody-Guterman K, Chase M, Ueki K, Louis DN, Harsh GR 4th, Chiocca A, and Breakefield XO (1996) Retrovirus-mediated gene therapy of experimental brain neoplasms using the herpes simplex virus-thymidine kinase/ganciclovir paradigm. Cancer Gene Ther 3, 99-106.

Tamiya T, Wei MX, Chase M, Ono Y, Lee F, Breakefield XO, and Chiocca EA (1995) Transgene inheritance and retroviral infection contribute to the efficiency of gene expression in solid tumors inoculated with retroviral vector producer cells. Gene Ther 2, 531-538.

Rainov NG, Zimmer C, Chase M, Kramm C, Chiocca EA, Weissleder R, and Breakefield XO (1995) Selective uptake of viral and monocrystalline particles delivered intra-

13

Gene Therapy and Molecular Biology Vol 3, page 14 Vincent AJ, Esandi MD, van Someren G, Noteboom JL, Avezaat CJ, Vecht C, Smitt PA, van Bekkum DW, Valerio D, Hoogerbrugge PM, and Bout A (1996a) Treatment of leptomeningial metastases in a rat model using a recombinant adenovirus containing the HSV-tk gene. J Neurosurg 85, 648-654.

Dedrick RL, Herscovitch P, and Oldfield EH (1996) Quantification and pharmacokinetics of blood-brain barrier disruption in humans. J Neurosurg 85, 1056-1065.

Vincent AJ, Vogels R, Someren GV, Esandi MC, Noteboom JL, Avezaat CJ, Vecht C, Bekkum DW, Valerio D, Bout A, and Hoogerbrugge PM (1996b) Herpes simplex virus thymidine kinase gene therapy for rat malignant brain tumors. Hum Gene Ther 7, 197-205. Viola JJ, Ram Z, Wallbridge S, Oshiro EM, Trapnell B, TaoCheng JH, and Oldfield EH (1995) Adenovirally mediated gene transfer into experimental solid brain tumors and leptomeningial cancer cells. J Neurosurg 82, 70-76. Vrionis FD, Wu JK, Qi P, cano WG, Cherington V (1996a) Preservation of the bystander cytocidal effect of irradiated herpes simplex virus thymidine kinase (HSV-tk) modified tumor cells. J Neurooncol 30:225-236 Vrionis FD, Wu JK, Qi P, Cano WG, Cherington V (1996b) Tumor cells expressing the herpes simplex virus-thymidine kinase gene in the treatment of Walker 256 meningeal neoplasia in rats. J Neurosurg 84:250-257 Warnke PC, Friedman HS, Bigner DD, and Groothuis DR (1987) Simultaneous measurements of blood flow and blood-totissue transport in xenotrasnplanted medulloblastomas. Cancer Res 47, 1687-1690. Wei MX, Tamiya T, Chase M, Boviatsis EJ, Chang TK, Kowall NW, Hochberg FH, Waxman DJ, Breakefield XO, and Chiocca EA (1994) Experimental tumor therapy in mice using the cyclophosphamide-activating cytochrome P450 2B1 gene. Hum Gene Ther 5, 969-978. Yagi K, Hayashi Y, Ishida N, Ohbayashi M, Ohishi N, Mizuno M, and Yoshida J (1994) Interferon-beta endogenously produced by intratumoral injection of cationic liposomeencapsulated gene: cytocidal effect on glioma transplanted into nude mouse brain. Biochem Mol Biol Int 32, 167-171. Yamada K, Ushio Y, Hayakawa T, Kato A, Yamada N, and Mogami H (1982) Quantitative autoradiographic measurements of blood-brain barrier permeability in the rat glioma model. J Neurosurg 57, 394-398. Zerrouqi A, Rixe O, Ghoumari AM, Yarovoi SV, Mouawad R, Khayat D, and Soubrane C (1996) Liposomal delivery of the herpes simplex virus thymidine kinase gene in glioma: improvement of cell sensitization to ganciclovir. Cancer Gene Ther 3, 385-392. Zhu J, Zhang L, Hanisch UK, Felgner PL, and Reszka R (1996) A continuous intracerebral gene delivery system for in vivo liposome-mediated gene therapy. Gene Ther 3, 472-476. Zlokovic BV, and Apuzzo ML (1997) Cellular and molecular neurosurgery: pathways from concept to reality - part II: vector systems and delivery methodologies for gene therapy of the central nervous system. Neurosurgery 40, 805-813. Z端nkeler B, Carson RE, Olson J, Blasberg RG, DeVroom H, Lutz RJ, Saris SC, Wright DC, Kammerer W, Patronas NJ,

14

Gene Therapy and Molecular Biology Vol 3, page 15 Gene Ther Mol Biol Vol 3, 15-23. August 1999.

Efficient in vivo expression of a reporter gene in rat brain after injection of recombinant replicationdeficient Semliki Forest virus Research Article

Kenneth Lundstrom, J. Grayson Richards, J. Richard Pink, and Francois Jenck F. Hoffmann-La Roche, Research Laboratories, CH-4070 Basel, Switzerland __________________________________________________________________________________________________ Correspondence: Kenneth Lundstrom, Ph.D. Tel: 41-61-687 8653, Fax: 41-61-688 4575, E-mail: kenneth.lundstrom@roche.com Keywords: Semliki Forest virus; in vivo expression; rat brain; !-galactosidase Received: 13 October 1998; accepted: 20 October 1998

Summary Recombinant replication-deficient Semliki Forest virus (SFV) expressing bacterial -galactosidase was injected into the amygdala and striatum of male Wistar rats. Reporter gene expression was detected up to 28 days post-injection. The maximal expression levels were obtained 1-2 days postinjection. I n s i t u hybridization studies demonstrated high expression of LacZ mRNA until day 2, but no signal was detected 4 days post-injection. N o significant change i n body weight and temperature, exploratory locomotor behavior and forced motor performances were observed after SFV-LacZ injections. The neuronal gene transfer with SFV vectors did not trigger any major cell toxicity.

al., 1990). The drawback with retroviruses is that only relatively low virus titers can be achieved and only dividing cells are infected. Adenovirus vectors are capable of infecting non-dividing as well as dividing cells and their transduction frequency is generally high (Haffe et al., 1992). The duration of expression is, however, limited due to cellular and humoral immune responses induced by the virus infection (Yang et al., 1994). Adeno-associated virus (AAV) are replication-deficient parvoviruses. They are nonpathogenic and nonimmunogenic, but can replicate in cell culture only in the presence of adenovirus or helper virus (Clark et al., 1995). AAV have only a limited packaging capacity of foreign DNA (<4.5 kb), but can integrate into the host genome. Herpes simplex virus (HSV) offers very good infectivity and allows large inserts of foreign DNA to be introduced (~30 kb). Virus infection can be maintained indefinitely in a latent state, but HSV infections generally show severe cytoxicity to cells. This effect has been reduced by deletion of some viral genes,

I. Introduction A multitude of different methods and vehicles have been developed to increase the efficiency of delivery of recombinant genes in vivo for gene function and gene therapy applications. The non-viral delivery vehicles include naked DNA and a variety of liposome-DNA complexes consisting of cationic lipids (Filion and Phillips, 1997). Naked DNA is highly sensitive to degradation with a half-life of only 5 min when injected intravenously (Lew et al., 1995), whereas the lipid structures can offer an increased protection. However, the low delivery efficiency is a considerable drawback using these vectors (Boulikas 1996). Viral vectors have offered the possibility to achieve higher transfection frequencies. Retroviral vectors are capable of very high transduction rates and even retrovirus producer cell lines can be used for gene delivery (Markowitz et al., 1990). The retrovirus delivered transgene can be stably integrated into the host genome to provide long-term gene expression (Miller et 15

Lundstrom et al: Semliki Forest virus in expression of LacZ gene in rat brain like ICP27 and ICP34.5 from the HSV genome (Howard et al., 1998). Lentiviruses offer good infectivity and longterm expression, and are therefore potential candidates as vectos for gene therapy (Verma and Somia, 1997). Despite the variety of gene delivery methods available, there are still needs for improvements and modifications of existing vectors as well as development of new vector technology. Recently, Sindbis virus, a member of the Alphavirus family, was used for successful high level delivery and expression of !-galactosidase in mouse nucleus caudata/putamen and nucleus accumbens septi (AltmanHamandzic et al., 1997). The goal of our study was to examine the ability of Semliki Forest virus (SFV) vectors (Liljestrรถm and Garoff, 1991), a closely related Alphavirus, to infect neuronal cells in vivo by direct delivery to a desired location in rat brain. SFV has an extremely broad host range which allows efficient infection of many cell types, including post-mitotic cells. In vivo packaging results in high titer (up to 1010 infective particles/ml) replication-deficient recombinant SFV particles. Recombinant SFV-LacZ virus particles were injected into the amygdala and striatum of rat brain. These regions play important roles in controlling motor functions and in regulating emotional states, respectively. We have investigated the neuropathological consequences of the SFV inoculation at different time points. Macroscopical analyses were carried out using neurological and behavioral parameters and the microscopical studies were performed on fixed brain tissue.

II. Results and Discussion A. Injection of SFV-LacZ into rat brain High-titer recombinant SFV-LacZ virus was generated as described in the Experimental procedures and as schematically illustrated in F i g . 1 . The infectivity of the SFV-LacZ virus was tested by infection of BHK cells in 6well plate cultures followed by X-gal staining. 100% infectivity was achieved with a multiplicity of infection (MOI) of 4. To enable the infection of a reasonable large population of cells, 1 x 105 SFV-LacZ particles were injected into the amygdala and striatum of male Wistar rats (F i g . 2 ), respectively, as described in the Experimental procedures.

B. Behavioral studies To study the effect of reporter gene expression based on virus vector delivery, rats receiving SFV-LacZ and control animals injected with sterile culture medium were subjected to behavioral studies. No significant change in body weight and temperature, exploratory behavior and

16

F i g . 1 . S e m l i k i Forest virus v e c t o r s for i n v i v o g e n e d e l i v e r y . In vitro transcribed RNA from pSFV3-LacZ and pSFV-Helper2 were cotransfected into BHK cells to in vivo package recombinant SFV-LacZ particles. These were injected into the amygdala and striatum of rat brain.

forced motor performances was observed between the two groups at any time post-injection. Physical and behavioral parameters recorded 1, 7 and 14 days post-injection are described in Table 1. Body weights did not differ significantly and a similar gradual increase was recorded for both SFV-LacZ injected rats and control animals, which is indicative of a good general condition of the animals (normal food and water intake, healthy metabolism). No infection-induced hyperthermia was detected as the body

Gene Therapy and Molecular Biology Vol 3, page 17

Table 1.

Fig. 2. Stereotaxic localization of injection sites. Injection cannulae were lowered into the right striatum (top) and left amygdala (bottom) for local delivery of SFV-LacZ virus or vehicle. Stereotaxic coordinates are from G. Paxinos and C. Watson (The Rat Brain in stereotaxic coordinates, Academic Press, 1997).

17

Lundstrom et al: Semliki Forest virus in expression of LacZ gene in rat brain be delivered to the caudate putamen and central amygdala for local expression in the infected cells, mainly neurons, and does not appear to spread into neighbouring regions, except via the ventricular system into ependymal cells of the lateral third and fourth ventricles. Modifications of the injection procedure (i.e. decrease in injection speed, volume or virus concentration) might further reduce or eliminate this spread. Whereas LacZ transcripts were only detected in the first 48 h after injection, !-galactosidase could still be found after 4 weeks. This is most probably due to the high stability of this particular enzyme.

temperature remained normal. No difference between the groups in spontaneous exploration of a novel environment by the animals was recorded in measurement of total distance and vertical activity which is indicative of a state of normal emotional reactivity of the rats. As the animals were still recovering from surgery (1 day post-injection) scores were lower for both SFV-LacZ injected and control animals on day 1 when compared to 7 and 14 days postinjection. No impairment was seen at any post-injection date on forced motor performance; virus-injected animals even performed better (i.e. remained longer on the rotating rod) than control rats on some occasions (1 and 14 days post-injection). No statistically significant differences in muscular strength were detected at any time following injection. Equally, no differences were observed at 2, 4, 21 and 28 days post-injection in groups of smaller size (n = 2, controls; n = 3, virus injections; data not shown). This neurobehavioral evaluation suggests that central injection of recombinant replication-deficient SFV particles has no major consequences on the general health and on regular sensorimotor functions of male Wistar rats.

III. Conclusions Our results clearly demonstrate the feasibility of using SFV vectors for efficient infection of neuronal cells in different regions of the rat brain. We could obtain local expression of !-galactosidase, mostly due to the replication-deficient nature of the recombinant SFV particles. The infection rate at the injection site was extremely high and the duration of the recombinant protein expression at least 28 days. This is comparable to the duration of bacterial !-galactosidase expression obtained with other viral vectors, like adenovirus (Neve 1993) and herpes virus (Fotaki et al., 1997). A further suggestion of the exceptionally high stability of the recombinant !galactosidase came from our in situ hybridization experiments, where we demonstrated that no LacZ mRNA could be detected after 48 h post-injection. Similar observations have been demonstrated in vitro in BHK cells infected with SFV-LacZ virus by RT-PCR techniques, where LacZ mRNA disappears approximately 65 h postinfection (Lundstrom, unpublished data). The kinetics of other recombinant proteins might be different and could result in faster degradation of the gene product. However, the transient nature of the protein expression is evident from our results. Although this will exclude the use of SFV vectors, at least in their present form, for long-term expression, the lack of neuronal cell damage caused by the SFV infection should allow efficient transient gene expression in short term studies. Fast generation of sitespecific knock-in and knock-out gene expression studies should be possible. Our behavioral studies also demonstrated that the SFV injections did not trigger any widespread inflammatory response or extensive cell destruction, although a local inflammatory response was evident at 14 and 28 days post-injection. There were no change in the animals’ exploratory locomotor behavior or forced motor performance, further indications of intact neuronal cells, compared to control animals.

C. -galactosidase expression Brain sections from rats injected with SFV-LacZ particles and medium, respectively, were stained with X-gal at the different time points (1, 2, 4 , 7 , 14 and 28 days post-injection) and in situ hybridization with a LacZ gene specific probe was carried out at 1 h, 24 h, 48 h and 4 days post-injection. Twenty-four hours after injection into striatum and amygdala, both LacZ mRNA and !galactosidase were detected at the site of injection as well as in ventricular ependymal cells throughout the brain (F i g . 3 ). The !-galactosidase expression was restricted to the infected cells and their processes and was not observed in other brain regions. LacZ mRNA was restricted to perikarya and no hybridization signal was found in cell processes. Whereas both transcript and recombinant protein were detected at 48 h post-injection (mainly at the injection sites), no mRNA was present at later time points. !galactosidase, on the other hand, was detected, albeit in ever decreasing amounts, 4, 7, 14 (F i g . 4 e ) and even 28 days (F i g . 4 f ) post-injection. In order to determine the presence or not of a toxic effect of the viral infection, adjacent sections were also stained with Toluidine Blue. Using the vehicle-injected animals as controls, virus-induced inflammation (in the form of local glioses) at the injection site could be observed 1-4 weeks after administration. Experiments are in progress with marker protein staining for astroglioses (GFAP) and microglioses (OX-42). Over all, our findings suggested that a reporter gene can

18

Gene Therapy and Molecular Biology Vol 3, page 19

Fig. 3. Regional distribution and cellular localization in rat brain of LacZ mRNA and -galactosidase 2 4 h p o s t - i n j e c t i o n . !-galactosidase (blue precipitate) is detected not only at the striatal injection site but also in the ependyma throughout the ventricular system (arrowhead and arrow, respectively in a). Note the expression not only in the neuronal cell bodies (b ) and ependymal cells (c ), but also in presumptive neuronal processes (b ). LacZ mRNA is also detected at the site of injection and in the ependyma (arrowhead and arrow, respectively, in d). The cellular sites of synthesis of LacZ in presumptive neurons and ependymal cells (of the fourth ventricle) are illustrated in e and f , respectively.

19

Lundstrom et al: Semliki Forest virus in expression of LacZ gene in rat brain

Fig. 4. Regional distribution and cellular localization in rat brain of LacZ mRNA and -galactosidase 2 , 1 4 a n d 2 8 d a y s p o s t - i n j e c t i o n , r e s p e c t i v e l y . !-galactosidase (blue precipitate) is detected at the striatal and amygdala injection sites (white and black arrowheads, respectively). Note the expression not only in the neuronal cell bodies, but also in their processes (b ). LacZ mRNA is also detected at the striatal and amygdala injection sites (white and black arrowheads, respectively, in c ). The cellular sites of synthesis of LacZ in presumptive neurons are illustrated in d. a-d illustrate 2 days postinjection, e and f the regional distribution of !-galactosidase at 14 and 28 days post-injection, respectively.

20

Gene Therapy and Molecular Biology Vol 3, page 21

IV. Experimental procedures C. Behavioral studies

A. Cell cultures and recombinant SFV production

1 . S e n s o r i m o t o r f u n c t i o n was evaluated using a rotarod paradigm in which animals were required to walk on a rotating bar. The bar was 10 cm wide, 5 cm in diameter, 40 cm above the bench and rotated twice per minute. Trained animals were able to follow the slow regular movement of the bar for several minutes. Mild sedation or motor impairment translates into incoordination on the rotating rod and the animals fall off the bar. Time spent on the rotating rod is measured in seconds and maximal cut-off time is 60 s (non-impaired animal).

BHK-21 cells were grown in a mixture of F12-MEM/Iscove (1:1) in 10% FCS (Gibco-BRL) for in vivo packaging of recombinant SFV particles (Lundstrom et al., 1994). Briefly, in vitro transcripts from pSFV3-LacZ (SFV replicase genes + LacZ gene) and pSFV-Helper 2 (SFV structural genes) (Berglund et al., 1993) were co-electroporated into BHK-21 cells (F i g . 1 ). In vivo packaged recombinant SFV particles were collected 24 hours later by harvesting the medium from the cell cultures. The SFV particles were activated with "chymotrypsin and the titer of the virus stocks determined by infection of defined numbers of BHK-21 cells with different dilutions of recombinant SFV-LacZ followed by X-gal staining. The titers were generally in the range of 1 x 10 9 infectious particles / ml. The virus stocks were filter sterilized through 22 µm filter (Millipore) and no further purification or concentration was necessary. The virus stocks were diluted to 1 x 108 infectious particles / ml prior to use.

2 . Muscular c a p a c i t y was evaluated using a grip strength procedure consisting of a quantitative assessment of forelimb grip strength. A triangular bar, 2 mm in diameter, 5 cm wide was connected to a digital strain gauge. This device was used to measure graded changes in the forelimb grip strength of rats. Animals held by the tail grasped the bar and were then gently pulled away from the bar with a smooth steady pull until they released the triangle. The strain gauge remained fixed at its maximum deflection, which was the force required to break the animal’s grip. Three readings were taken for each animal and the maximum of 3 permissible readings was recorded as the grip strength score (in g).

B. Injections of recombinant SFV into rat brain

3 . E x p l o r a t i o n o f a n o v e l e n v i r o n m e n t in a test of free exploratory activity was measured in activity monitors (40 x 40 x 30 cm, Omnitech Electronics) placed in a soundproof room with central light. Locomotion was monitored via a grid of invisible infrared light beams. Horizontal and vertical activity were used in this study to describe the dynamic picture of rats. A vertical sensor monitoring rearing and jumping activity was attached 8 cm above the cage floor. An analyzer constantly collected the beam status information from the activity monitor and activity detected by the horizontal sensors was expressed as total distance run during the 30 min test.

Male Wistar rats (lbm RoRo, SPF, Biological Research Labs Ltd, Switzerland) were housed individually under controlled laboratory conditions (temperature 20 ± 2 o C, relative humidity 50-60%) with ad libitum access to food and water and were maintained on a normal 12 h light-12 h dark cycle (6 am-6 pm). Rats weighed 250-300 g at the time of surgery. They were stereotaxically microinjected under general anesthesia with Ketamine/Xylazine (200/10 mg/kg ip) in physiological saline under thermoregulatory control and oxygen supplementation. Craniotomy was performed using a fine dental drill for injection at one site located over the right striatum (0.2 mm anterior and 2.6 lateral to bregma; Paxinos and Watson, 1997) and at another site located over the left amygdala (2.6 mm posterior and 4.0 mm lateral to bregma). Stainless steel injectors attached to a stereotaxic holder were then lowered 5.0 mm ventral to the skull surface in the striatum and 8.0 mm in the amygdala (F i g . 2 ). These were connected via polyethylene tubing containing viral or control solutions to a 10 µl Hamilton syringe on a microinfusion pump (Harvard PHD 2000). Solutions were infused in a volume of 1 µl over 2 min (0.5 µl/min). The injection needle was left in place for 2 additional min before being slowly withdrawn over 1 min. The wound was then sutured and animals kept warm for 3-4 h after surgery. Post-operative buprenorphine (0.05 mg/kg) analgesic treatment (sc) was given for the next day.

The experimental procedures used in this study received approval by the local ethics committee and were performed in accordance with international and Swiss federal regulations and guidelines on animal experimentation.

D. Histological analyses On each test day (1, 2, 4, 7 , 14, 21 and 28 days postinjection) two to three animals in each group (virus and vehicle) were sacrificed for histological analysis. Directly following CO2 inhalation euthanasia, animals were transcardially perfused (clamped dorsal aorta) with 20 ml of 4% paraformaldehyde (PFA) for fixed brain extraction. Brains were stored for 4 h in the same PFA solution, then cryoprotected in 30% sucrose at 4 o C overnight and stored at -80 o C until sectioned. Free-floating sections were cut on a freezing microtome at 40 µm then reacted for !-galactosidase, as well as LacZ mRNA. Some sections were stained for 0.5-1.0 min in 0.5% Toluidine Blue (Fluka 89640) in 0.2 M acetate buffer pH 4.5.

A group of 21 rats were stereotaxically injected with the SFV-LacZ virus (105 particles/ µl) and 12 control rats received sterile vehicle (culture medium). Animals were carefully evaluated on day 1, 2, 4, 7, 14, 21 and 28 for consequences of viral injections on general health (global appearance, measures of body weight and rectal temperature), sensorimotor coordination, muscular capacity and exploration of a novel environment.

E. Enzyme histochemistry of -galactosidase 21

Lundstrom et al: Semliki Forest virus in expression of LacZ gene in rat brain Oxidation solution was prepared as follows. 80 µl Nonidet P-40 and 0.04 g sodium deoxycholate were added to the oxidation stock solution (40 ml 10x PBS, 360 ml 2x distilled H2 O, 0.65 g potassium ferricyanide (K3 Fe(CN)6 ), 0.84 g potassium ferrocyanide (3 H 2 O) (K 4 Fe(CN)6 3 H2 O), 0.16 g MgCl 2 6 H2 O) and stirred thoroughly and filtered (45 µm). The resulting bright yellow solution was stored at room temperature under light-tight conditions. !-galactosidase was visualized by adding 10 mg of X-gal substrate (Boehringer Mannheim 1680293) to 0.25 ml DMSO. Once dissolved, 10 ml oxidation solution was added with careful mixing to avoid the formation of air bubbles. Tissue sections were rinsed ( 2 x 15 min) in PBS, then reacted with the X-gal solution overnight at 31-33 o C in a dark box. The reacted sections (an insoluble blue indoyl precipitate reaction for !-galactosidase) were then rinsed again (2 x 15 min) in PBS, post-fixed in icecold 4% PFA for 15 min, rinsed in PBS, mounted on precleaned glass slides, counterstained with 1% Neutral Red (Sigma), dehydrated and coverslipped with DePeX.

examined with brightfield optics using a Zeiss Axiophot.

G. Imaging The regional and cellular distribution of X-gal and LacZ mRNA were recorded as digital images using a ProgRes high resolution color camera and Adobe Photoshop software.

Acknowledgements We are grateful to Mr. Andreas Kunz for his help with SFV-LacZ virus production and Ms. Martine Maco and Ms. Martine Kapps for their excellent technical assistance in the neurosurgical injections and behavioral evaluation of rats. Ms. Fabienne Goepfert is acknowledged for X-gal staining of brain sections, Ms. Zaiga Bleuel for help with in situ hybridizations and Mr. Jürg Messer for his assistance in imaging of brain sections.

References

F . In situ hybridization For selected time points we also investigated the regional and cellular expression of LacZ transcripts using a 60-mer oligonucleotide probe (nucleotides 3001-3060) selective for the LacZ gene (Casadaban et al., 1983). The hybridization procedure has been previously described (Saura et al., 1996). Briefly, 12 µm cryostat sections of fresh-frozen rat brains (1 h, 24 h, 48 h and 4 days post-injection) and 30 µm freezingmicrotome sections of perfusion-fixed rat brains (1, 2, 4, 7, 14 and 28 days post-injection) were used. The cryostat sections were mounted on Superfrost Plus# slides then fixed in 4% PFA in PBS, pH 7.4 for 20 min followed by three 5 min washes in PBS. The oligonucleotide was ordered from Genosys Biotechnologies and labeled at the 3’ end with terminal deoxynucleotidyl-transferase (BRL) and [35 S] dATP (New England Nuclear). The labeled probe was separated from unincorporated nucleotides with a Biogel P30 spin column (twice 4 min at 1600 x g, Sorvall SW24). Sections were hybridized with 50 µl of a solution with the following composition: 4 x SSC, 20% dextran sulfate, 0.25 µg/ml herring sperm DNA (denatured) , 50% deionized formamide (BRL), 0.1 M dithiotreitol (DTT) (Fluka), 0.5 x Denhard’s solution and the 35 S-labeled probe (3 x 10 5 cpm). Sections were covered with strips of Fujifilm# and incubated in moist chambers at 43o C overnight. Following removal of the strips, the sections were washed twice in a solution containing 1 x SSC and 10 mM DTT for 15 min at 55o C, then in 0.5 x SSC with 10 mM DTT once for 15 min at room temperature. After a dip in 2 x distilled H 2 O, sections were dehydrated in ethanol, exposed (for up to 4 weeks) to sheet film (Hyperfilm#, !Max, Amersham) or dipped in Ilford K5 nuclear emulsion to reveal the regional and cellular localization of the mRNA, respectively. The film or emulsion was developed in Kodak PL12 or Kodak D19, respectively, then transferred to Kodak Rapid Fix. Nissl- or Neutral Red-counterstained sections were

22

Altman-Hamandzic S, Groceclose C, Ma J-X, Hamandzic D, Vrindavanam NS, Middaugh LD, Parratto NP and Sallee FR (1 9 9 7 ) Expression of !-galactosidase in mouse brain: utilization of a novel nonreplicative Sindbis virus vector as a neuronal gene delivery system. Gene Ther 4, 815822. Berglund P, Sjöberg M, Garoff H, Atkins GJ, Sheahan BJ and Liljeström P (1 9 9 3 ) Semliki Forest virus expression system: Production of conditionally infectious recombinant particles. B i o / T e c h n o l o g y 11, 916-920. Boulikas T ( 1 9 9 6 ) Liposome DNA delivery and uptake by cells. O n c o l R e p 3, 989-995. Casadaban MJ, Martinez-Arias A, Shapira SK and Chou J ( 1 9 8 3 ) !-galactosidase gene fusions for analyzing gene expression in Escherichia coli and yeast. Meth E n z y m o l 100, 293-310. Clark KR, Voulgaropoulus F, Fraley DM and Johnson PR ( 1 9 9 5 ) Cell lines for the production of recombinant adeno-associated virus. Hum Gene Ther 6, 1329-1341. Filion MC and Phillips NC ( 1 9 9 7 ) Toxicity and immunomodulatory activity of liposomal vectors formulated with cationic lipids toward immune effector cells. B i o c h e m B i o p h y s A c t a 1329, 345-356. Fotaki ME, Pink JR and Mous J ( 1 9 9 7 ) Tetracyclineresponsive gene expression in mouse brain after amplicon-mediated gene transfer. Gene Ther 4, 901908. Haffe HA, Danel C, Longenecker G, Metzger M, Setoguchi Y et al., ( 1 9 9 2 ) Adenovirus-mediated in vivo gene transfer and expression in normal rat liver. Nature Genet 1, 372-378. Howard MK, Kershaw T, Gibb B, Storey N, MacLean AR, Zeng B-Y, Tel BC, Jenner P, Brown SM, Woolf CJ, Anderson PN, Coffin RS and Latchman DS ( 1 9 9 8 ) High efficiency

Gene Therapy and Molecular Biology Vol 3, page 23 gene transfer to the central nervous system of rodents and primates using herpes virus vectors lacking functional ICP27 and ICP34.5. Gene Ther 5, 1137-1147. Lew D, Parker SE, Latimer T, Abai AM, Kuwahara-Rundell A, Doh SG, Yang Z-Y, Laface D, Gromkowski SH, Nabel GJ, Manthorpe M and Norman J ( 1 9 9 5 ) Cancer gene therapy using plasmid DNA: pharmacokinetic study of DNA following injection in mice. Hum Gene Ther 6, 553564. Liljestrom P and Garoff H (1 9 9 1 ) A new generation of animal cell expression vectors based on the Semliki Forest virus replicon. B i o / T e c h n o l o g y 9, 1356-1361. Lundstrom K, Mills A, Buell G, Allet E, Adami N and Liljestrรถm P (1 9 9 4 ) High-level expression of the human neurokinin-1 receptor in mammalian cell lines using the Semliki Forest virus expression system. Eur J Biochem 224, 917-921. Markowitz S, Hesdorffer C, Ward M, Goff S and Bank A ( 1 9 9 0 ) Retroviral gene transfer using safe and efficient packaging cell lines. Ann NY Acad Sci 612, 407-414. Miller DG, Adam MA and Miller AD ( 1 9 9 0 ) Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. M o l C e l l B i o l 10, 4239-4242. Neve RL ( 1 9 9 3 ) Adenovirus vectors enter the brain. Trends Neurosci 16, 251-253. Paxinos G and Watson C ( 1 9 9 7 ) in The Rat Brain in stereotaxic coordinates, Academic Press. Saura J, Bleuel Z, Ulrich J, Mendelowitsch A, Chen K., Shih JC, Malherbe P, Da Prada M and Richards JG ( 1 9 9 6 ) Molecular neuroanatomy of human monoamine oxidases A and B revealed by quantitative enzyme radioautography and in situ hybridization histochemistry. N e u r o s c i e n c e 70, 755-774. Verma IM and Somia N ( 1 9 9 7 ) Gene therapy - promises, problems and prospects. Nature 389, 239-242. Yang Y, Nunes FA, Berencsi K, Furth EE, Gonczol E and Wilson JM ( 1 9 9 4 ) Cellular immunity to viral antigens limitsE1-deleted adenoviruses for gene therapy in cystic fibrosis. Nature Genet 7, 362-369.

23

Gene Therapy and Molecular Biology Vol 3, page 25 Gene Ther Mol Biol Vol 3, 25-33. August 1999.

Establishment of an assay to determine adenovirusinduced endosome rupture required for receptormediated gene delivery Research Article

Daniela Schober, Nora Bayer, Robert F. Murphy1, Ernst Wagner2 and Renate Fuchs Department of General and Experimental Pathology, University of Vienna, A-1090 Vienna, Austria 1 Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, USA 2 Boehringer Ingelheim R&D Vienna, A-1121 Vienna, Austria __________________________________________________________________________________________________ Correspondence: Renate Fuchs, Department of General and Experimental Pathology, University of Vienna, W채hringer G체rtel 18-20, A-1090 Vienna, Austria. Phone: +43-1-40-400-5127; Fax: +43-1-40-400-5130; E-mail: renate.fuchs@akh-wien.ac.at Key words: adenovirus entry, endosome rupture, flow cytometry Received: 9 October 1998; accepted: 19 October 1998

Summary Successful human gene therapy requires methods to transfer recombinant genes to cells efficiently. One p o s s i b i l i t y i s t o use adenoviral-based vectors. The entry route o f adenovirus involves endocytic uptake, penetration of modified viral particles into the cytoplasm by endosome rupture, transport to the nuclear pore complex, disassembly of modified particles and import of the DNA into the nucleus. Since endosome rupture is a rate-limiting step in foreign gene expression, we developed a two-step assay to quantitative virus-mediated membrane rupture. Following endosome labeling of HeLa cells with a pH-sensitive (FITC-dextran) and pH-insensitive (Cy5-dextran) fluidphase marker in the absence or presence of replication-defective adenovirus type 5 (Ad5), first, the pH o f labeled compartments was determined by flow-cytometry o f c e l l suspensions. When compared to control cells, the pH of labeled compartments was elevated by co-internalization of Ad5 indicating endosome l y s i s and penetration o f the marker into the pH-neutral cytoplasm. Second, single-organelle flow analysis (SOFA) of cell homogenates of the same cells was applied t o q u a n t i t a t e t h e a m o u n t o f l a b e l e d a s w e l l a s u n l a b e l e d v e s i c l e s i n t h e p r e s e n c e o f Ad5. Our results demonstrate that adenovirus internalized for 10 min into HeLa cells destroys about 30% of endosomal compartments. This assay can be applied t o rapidly screen various gene delivery systems for their ability to disrupt endosomal membranes and to enter the cytoplasm.

proteins. In contrast to other DNA viruses, adenovirus infects non-dividing cells and replicates in the nucleus but rarely integrates into the host genome. Expression of foreign genes can be achieved by direct DNA insertion into the viral genome or conjugation to the virus. Moreover, replication deficient adenoviruses have been generated that can be propagated in cell lines expressing the complementing viral proteins (Graham et al., 1977; Jones and Shenk, 1979). Recombinant adenoviruses have also been used for gene transfer (Fujita et al., 1995).

I. Introduction Adenovirus (Ad) is widely used as a vehicle to deliver foreign DNA into the cell of interest. This is due to the broad expression of adenoviral receptors on human cells and the particular mechanism of uncoating of this virus. 47 serotypes of adenoviruses have been characterized and classified into 6 subgroups (A - F). Adenovirus is a nonenveloped virus with an icosahedral capsid shell and protruding fibers anchored to the penton base proteins. The linear double stranded DNA is attached to four core 25

Schober et al: Adenovirus-induced endosome rupture So far, the internalization pathway and mechanism of uncoating of adenoviruses of subgroup C (e.g. Ad2 and Ad5) are fairly well characterized, although not fully understood at the molecular level. After binding to primary cellular receptors (MHC-class I complex and coxsackievirus-adenovirus receptor) via the distal portion of its fiber protein (Bai et al., 1994; Bergelson et al., 1997; Hong et al., 1997), subsequent internalization of the virus requires binding of five RGD motifs in the penton base to ! v"-integrins (Wickham et al., 1993; Nemerow et al., 1994). Following endocytosis via clathrin coated pits and vesicles (Wang et al., 1998), adenovirus undergoes a series of modifications of its capsid proteins in the low pH environment of endosomes that ultimately results in rupture of the endosomal membrane (Greber et al., 1993). The low pH-dependent, virus-induced endosome lysis also requires the presence of ! v"-integrins (Wickham et al., 1994). During endocytosis and penetration into the cytoplasm, capsid proteins are degraded or dissociated and finally the DNA core is freed from the hexon (Greber et al., 1993; Greber et al., 1996). Binding of the DNA core to the nuclear pore complex results in its disassembly and import of the viral genome into the nucleus (Greber and Kasamtsu, 1996; Greber et al., 1997). Thus, the capacity of adenovirus to rupture the endosome and the selective targeting of viral DNA to the nucleus provides a powerful vehicle for gene transfer (Curiel et al., 1991; Wagner et al., 1992; Wagner, 1998).

resolve distinct fluorescent endocytic vesicles in cell homogenates, a method termed single-organelle flow analysis (SOFA) (Murphy, 1985, 1990; Murphy et al., 1989; Wilson and Murphy, 1989). In the present study we co-internalized replicationdefective adenovirus type 5 (Ad5) together with pHsensitive (FITC) and pH-insensitive (Cy5) derivatives of the fluid-phase marker dextran into HeLa cells. The pH of labeled compartments was determined by flow cytometry and, in addition, the integrity of total vesicles and fluorescent endosomes was evaluated by SOFA. Using these techniques, we here demonstrate that Ad5 elevates the pH of labeled compartments, suggesting endosome lysis and access of the dextran into the pH neutral cytoplasm. This conclusion was verified by SOFA, in that a reduction in the number of labeled as well as unlabeled vesicles in the presence of Ad5 was observed. The results confirm the utility of these flow cytometric methods for monitoring adenovirus-induced endosome lysis.

II. Results A. Flow cytometry of HeLa cell endosomes Fluid-phase markers are non-specifically internalized into cells and can therefore be used to label all endocytic vesicles, depending on the internalization conditions applied (time, temperature). Furthermore, fluid-phase markers do not bind to cellular membranes and are released into the cytoplasm when endosomes are lysed (Yoshimura, 1985; Defer et al., 1990). Under control conditions, internalized markers will be exposed to the low pH environment of intact endosomes whereas in the presence of membrane disrupting agents they will be released into the pH-neutral cytoplasm (see F i g . 1 ).

Adenovirus has long been known to increase the permeability of the plasma or endosomal membrane at low pH (Seth et al., 1985; Seth, 1994). In this process, cointernalized macromolecules can gain access to the cytoplasm. Using an in vitro assay (a so-called endosome leakage assay), we have recently shown that adenovirus leads to release of small and large molecules from isolated endosomes when incubated in low pH buffer (Prchla et al., 1995). Although this in vitro assay allows determination of conditions required for endosomal content release, it is laborious and time consuming. Since efficient gene transfer primarily depends on endosome rupture, we sought to establish a rapid assay to determine the endosomolytic activity of adenovirus in vivo. Macromolecules and viruses taken up by endocytosis are exposed to the low pH environment of the endosomes due to the activity of the vacuolar proton ATPase (Mellman et al., 1986; Mukherjee et al., 1997). Therefore, we took advantage of the pH-dependence of FITC-derivatives to selectively label endosomes and measured endosomal pH using flow cytometry and the dual fluorescence (dual fluorochrome) ratio method (Murphy et al., 1984; Cain and Murphy, 1986). Flow cytometry can also be applied to

F i g u r e 1 . Receptor mediated adenovirus entry and virusinduced endosome rupture.

26

Gene Therapy and Molecular Biology Vol 3, page 27 F i g u r e 2 . Experimental set-up for FACS and SOFA analysis of HeLa cells infected with adenovirus type 5 (Ad5) in the presence of FITC- and Cy5-dextran.

We therefore internalized a pH-sensitive (FITC) and insensitive (Cy5) derivative of the fluid-phase marker dextran (MW 70 kD) in the presence or absence of Ad5 (1000 particles / cell) into HeLa cells for 10 min at 34째C, followed by a 10 min chase in marker-free medium (for experimental set-up see F i g . 2). Under this condition primarily late endocytic compartments will be labeled with the marker (Schober et al., 1998). Flow cytometry of cell suspensions was then used to determine the amount of internalized marker (reflected by Cy5 fluorescence) and endosomal pH (reflected by the ratio of FITC and Cy5 fluorescence). The total amount of marker internalized was found to be stimulated by adenovirus by 40% when compared to controls. For pH measurements, a standard curve was generated for each sample by measuring the FITC/Cy5 ratio as a function of external pH. As shown in F i g . 3, increasing the pH of the external medium (in the presence of permeant ions) from 4.5 to 7.5 results in a linear increase in the FITC/Cy5 ratio. Based on this calibration curve, an average endosomal pH of 6.3 +/- 0.1 was obtained when marker was internalized in the absence of Ad5 (F i g . 4 ). When Ad5 was co-internalized with the dextran, the pH of labeled compartments was increased to 7.3 +/- 0.1, suggesting virus-induced endosome rupture and release of internalized marker into the pH neutral cytoplasm.

Figure 3 . Normalized pH calibration curves of FITC/Cy5-dextran labeled endosomes obtained by flow cytometry of cell suspensions and post-nuclear supernatants (PNS), respectively. HeLa cell endosomes were labeled as described in F i g . 2 . Cell suspensions or PNS were incubated with pH buffers containing permeant anions and azide to deplete cellular ATP and nigericin to equilibrate internal with external pH.

27

Schober et al: Adenovirus-induced endosome rupture

B. Single-organelle flow analysis (SOFA) of HeLa cells infected with adenovirus

non-fluorescent objects can be detected in this fashion. In order to reproducibly measure endocytic vesicles, a method for choosing a consistent threshold is needed. As described previously (Wilson and Murphy, 1989), we chose a threshold value of SS just above the maximum value observed when sheath fluid without sample was analyzed.

To verify whether the increase of the "endosomal" pH of labeled compartments in the presence of adenovirus is due to loss of internalized fluid-phase marker from acidic organelles such as endosomes, we applied single-organelle flow analysis (SOFA). When a cell homogenate or postnuclear supernatant (PNS) is subjected to SOFA, the following parameters can be analyzed at the same time: forward scatter (FS) and side scatter (SS), both of which are related to size and optical density, and FITC- and Cy5fluorescence (Wilson and Murphy, 1989). Using these parameters the following information can be obtained: (i ) size distribution and number of vesicles of a certain size; (i i ) the degree of co-localization of two distinct fluorescent markers; and (i i i ) the internal pH of individual vesicles. Our goal was to determine (for cells treated and untreated with adenovirus) the number of total endocytic vesicles (unlabeled and fluorescent), the number of fluorescent endosomes, and the average pH of intact endosomes.

As an illustration of the SOFA method, all endocytic compartments (endosomes, lysosomes, recycling vesicles) of HeLa cells were labeled by continuous internalization of FITC-and Cy5-dextran for 2 h at 34째C. Thereafter, cells were rapidly cooled, washed and homogenized with a ballbearing homogenizer ensuring minimal destruction of vesicles during homogenization (Balch and Rothman, 1985). Nuclei and unbroken cells were removed by centrifugation and the resulting PNS was subjected to SOFA (see Materials and Methods). For control purposes, a PNS was prepared from unlabeled cells and also analyzed by SOFA. To differentiate large vesicles (such as late endosomes) from small vesicles, an analysis window was created with a lower FS value just above the maximum observed for sheath fluid alone (as above for SS). As depicted in F i g . 5 and Table 1, about 29% of the total events (objects) detected in both unlabeled (F i g . 5 A ) and labeled (F i g . 5E) samples were in this large vesicle window, while about 61% of all events fell in a corresponding window for small vesicles. In histograms displaying the fluorescence parameters, a region defining events positive for FITC and/or Cy5 fluorescence was created to exclude essentially all events from unlabeled cell homogenates. When either large or small vesicles from F i g . 5 A were depicted in dual fluorescence histograms, a minute number (0.2%) of vesicles from unlabeled cells were detected in the fluorescence-positive region (F i g . 5 B and F). This confirmed that the region was appropriately defined. (F i g . 5 C ). In the PNS of FITC/Cy5-dextran labeled HeLa cells, 64% of all vesicles counted (small and large) were fluorescent (F i g . 5 G , T a b l e 1). However, whereas the majority (83%) of the large vesicles contained FITC/Cy5-dextran (F i g . 5F), only 47% of the small vesicles were fluorescent (F i g . 5 H , Table 1). Thus, nearly all large vesicles can be defined as endocytic vesicles due to labeling with internalized fluorescent fluid-phase marker. Consequently, the number of vesicles in this population was used as one indication of endosome disruption by adenovirus.

When measuring the properties of subcellular organelles by SOFA, a criterion must be chosen to define which objects in an organelle suspension are to be analyzed. This is normally accomplished by using a threshold on a light scattering parameter (FS or SS), since both fluorescent and

Having defined the flow-cytometer settings to analyze endosomes, FITC- and Cy5-dextran was internalized into HeLa cells in the absence or presence of Ad5 (10 min pulse, 10 min chase) as in F i g . 2 . Cell homogenates were prepared, centrifuged at low speed and the resulting supernatant (PNS) was subjected to SOFA. As shown in F i g . 6 A , the total number of large vesicles was reduced by co-internalization of Ad5 to 69% of controls. Further support for endosome lysis due to adenovirus entry is

F i g u r e 4 . Influence of adenovirus on the pH of labeled compartments of HeLa cells. HeLa cell endosomes were labeled with FITC/Cy5-dextran without or with Ad5 as described in F i g . 2 . Cell suspensions were analyzed by flow cytometry and the internal pH was calculated using the pH calibration curve shown in F i g . 3 . Values depicted are the mean +/- SD from 3 independent experiments.

28

Gene Therapy and Molecular Biology Vol 3, page 29

F i g u r e 5 . Selection of gates to define small, large and fluorescent vesicles for SOFA. All endocytic compartments in HeLa cells were labeled by continuous internalization with 2 mg / ml FITC-dextran and 0.1 mg / ml Cy5-dextran for 2 h at 37째C. A postnuclear supernatant was prepared as described in Materials and Methods and subjected to SOFA (E-H ). Results for a post-nuclear supernatant from unlabeled HeLa cells are also shown (A-D).

29

Schober et al: Adenovirus-induced endosome rupture unlabeled sample

labeled sample

100.000

100.000

events counted

vesicles (% of total)

nonfluorescent (% of total) (% of total) fluorescent

fluorescent (% of total) (% of gated)

large vesicles

27.8

0.2

5.0

24.0

83.0

small vesicles

62.7

0.2

32.0

29.0

47.0

T a b l e 1 . Quantitation of large, small and fluorescent vesicles in post nuclear supernatants analyzed by SOFA. The data presented in F i g . 5 are summarized and expressed as % of total events counted (F i g . 5 B , D , F , H ). In addition, when endosomes had been labeled with FITC/Cy5-dextran (F i g . 5 E , F , H ) fluorescent vesicles were also normalized to the amount observed in the respective FS gate for large and small vesicles. vesicles was determined based on FS and SS histograms (as in F i g . 5 A ). (B ) The number of large fluorescent endosomes in scatter-gated histograms of FITC and Cy5 fluorescence is shown (as in F i g . 5 F ). Data are expressed as percent of the corresponding values obtained in the absence of virus.

F i g u r e 7 . Influence of adenovirus on the internal pH of residual intact endosomes. The internal pH of large fluorescent endosomes for the samples in F i g . 6 B was calculated using the pH calibration curve shown in F i g . 3 .

F i g u r e 6 . Adenovirus internalization reduces the number of large vesicles (A) and fluorescent endosomes (B ). FITC/Cy5-dextran was internalized into HeLa cells in the absence or presence of Ad5 as described in F i g . 2 . A PNS was prepared and analyzed by SOFA. (A) The number of total large

30

Gene Therapy and Molecular Biology Vol 3, page 31 provided when bona fide endosomes, i.e. fluorescent vesicles, are analyzed. In Ad5-infected cells, the number of FITC/Cy5-positive endosomes was decreased to 73% of controls (F i g . 6 B ). This demonstrates that quantitation of large vesicles by SOFA can be used to reflect endosome rupture.

lysed) has not been quantitated. Our results show that relatively small amounts of Ad5 (1000 particles / cell) when co-internalized for 10 min with the fluid-phase marker dextran and chased for an additional 10 min (to label late endosomes but not lysosomes (Prchla et al., 1994)) stimulated dextran uptake by 40%. This is in good agreement with published data (Defer et al., 1990). Under the same conditions, the virus destroyed about 30% of all endocytic compartments. Since the number of large vesicles in FS/SS histograms primarily reflect endocytic compartments (F i g . 5 and Table 1), the endosomolytic activity of a given agent can be rapidly analyzed without prior endosome labeling with fluorescent tracers. Thus, this system offers the advantage to rapidly screen DNA-delivery systems such as low pH activated liposomes for their ability to lyse endosomes and to enter the cytosol.

Finally, the influence of adenovirus on the average internal pH of residual intact endosomes was determined. Endosomes maintain their low intravesicular pH after cell homogenization at 4째C for up to 20 h in the absence of highly permeant ions (Fuchs et al., 1989; Wilson and Murphy, 1989). A pH calibration curve similar to that obtained for whole cells was created for endosomes in the PNS analyzed by SOFA (F i g . 3 ). Using this calibration curve, an average internal pH of 6.0 +/- 0.1 was found for large vesicles (endosomes) from control cells (F i g . 7), while co-internalization of Ad5 slightly decreased the pH of residual intact endosomes (5.5 +/- 0.2). This confirms that the increase in pH of labeled compartments observed in cell suspensions by flow cytometry is indeed due to release of internalized marker into the cytoplasm, rather than to alteration of endosomal pH per se.

Using adenovirus as gene delivery system one has to bear in mind that destruction of endosomes will also affect subsequent endocytic uptake of nutrients, hormones, and growth factors as well as signaling events from endosomes. So far, it is unknown how rapidly the endosomal system is regenerated after adenovirus internalization. Presumably, this may depend on whether for the particular cell of interest, transport to lysosomes occurs by endosome maturation or by carrier vesicles (Murphy, 1993; Gruenberg and Maxfield, 1995). We intend to apply SOFA to investigate the recovery of the endocytic system after adenovirus infection.

III. Discussion We here demonstrate that flow cytometry is a rapid and sensitive technique that can be used to analyze the endosome-disrupting potential of adenovirus. The two-step analysis involves 1/ determination of endosomal pH of labeled compartments by flow cytometry of cell suspensions, and 2/ SOFA of cell homogenates of the same cells. The first analysis indicates the potential endosome leakage induced by the virus that it then verified and quantified by SOFA. Our results show that short cointernalization (10 min) of Ad5 and fluid-phase marker results in rupture of about 30 % of endocytic vesicles.

Acknowledgments This work was supported by Austrian Science Foundation grants P-10618-MED and P-12967-GEN to R.F.

IV. Materials and Methods

Adenoviruses of subgroup B and C have been shown to increase the rate of fluid-phase uptake and in addition to permeabilize the plasma or endosomal membrane for small and large molecules (Yoshimura, 1985; Otero and Carrasco, 1987). In particular, adenoviruses type 2 and 5 (subgroup C) are known to enter the cytoplasm by endosome lysis. Comparison of the data presented in this investigation with previous studies are difficult, because in former studies large quantities of adenovirus (2000 - 5000 particles / cell) were used and the internalization conditions applied resulted in labeling of early and late endosomes as well as of lysosomes (Defer et al., 1990). Furthermore, permeabilization of the plasma membrane could not be differentiated from endosome rupture. Adenovirus-mediated enhancement of cytoplasmic delivery has mainly been analyzed using toxins or toxin-conjugates that inhibit protein synthesis (Seth, 1994). So far, adenovirus-induced endosome rupture in vivo (i.e. the number of endosomes

A. Chemicals All chemicals were obtained from Sigma unless specified. Fluorescein isothiocyanate-conjugated (FITC)-dextran (FD 70) was extensively dialyzed against Tris-buffered saline pH 7.4 (TBS) and finally against phosphate buffered saline (PBS) before use. Cy5.18-OSu (Cy5) was obtained from Amersham (UK) and coupled to dextran (M r 70 kD) as described (Rybak and Murphy, 1998).

B. Cell culture and virus propagation HeLa cells (Wisconsin strain, kindly provided by R. Rueckert, University of Wisconsin) were grown in monolayers in MEM-Eagle (GIBCO) containing heat-inactivated 10% fetal calf serum; in suspension culture Joklik's MEM (GIBCO) supplemented with 7% horse serum was used. Adenovirus serotype 5 mutant dl 312 (Ad5), a replication incompetent strain deleted in the E1a region was propagated in 293 cells

31

Schober et al: Adenovirus-induced endosome rupture (Graham et al., 1977; Jones and Shenk, 1979).

at pH 7.5. Finally, the average pH of labeled compartments was determined using the pH calibration curve.

C. Endosome labeling for flow cytometry F. Flow cytometry

HeLa suspension cells (2x10 7 ) were preincubated in 2 ml DMEM containing 10% FCS for 30 min at 37°C. For labeling of all endocytic compartments, HeLa cells were incubated in 2 ml fresh medium with serum containing 2 mg / ml FITC-dextran and 0.1 mg / ml Cy5-dextran for 2 h at 37°C. To determine the influence of adenovirus on endosomal pH and endosome integrity, endosomes were labeled by incubation of HeLa cells in 2 ml DMEM containing 10% FCS, 6 mg / ml FITC-dextran, and 1 mg / ml Cy5-dextran without or with Ad5 (MOI of 1000) for 10 min at 34°C followed by a 10 min chase in DMEM in the absence or presence of Ad5. Internalization was halted by addition of ice-cold PBS (pH 7.4), pelleting the cells and washing the pellet twice with 30 ml ice cold PBS. The final cell pellet was resuspended in 2 ml PBS and divided into 7 aliquots. One aliquot (further diluted with PBS to 500 µl) was analyzed immediately by flow cytometry, the remaining aliquots were used for generation of the pH calibration curve (see below).

A dual laser FACS Calibur (Becton Dickinson Immunocytometry Systems) equipped with argon-ion and reddiode lasers was used. FITC-fluorescence (488 nm excitation) was measured using a 530 nm band pass filter (30 nm band width) and Cy5-fluorescence (635 nm excitation) was measured using a 661 nm band pass filter (16 nm band width). Forward light scatter and 90° (side)-scatter, along with both fluorescence values, were collected in list mode using 256channel resolution. For flow cytometry of cell suspensions, data for 10.000 cells were collected, while 100.000 events per sample were collected for SOFA. The following threshold parameters were defined for SOFA (see also Results and F i g . 5 ): 1/ Forward scatter (FS) and side scatter (SS): As described (Wilson and Murphy, 1989), a threshold value of SS just above the maximum value observed when sheath fluid without sample was analyzed was chosen. To differentiate large vesicles (such as late endosomes) from small vesicles, an analysis window was created with a lower FS value just above the maximum observed for sheath fluid alone (as above for SS). 2/ FITC and Cy5 fluorescence: In dual fluorescence histograms, threshold parameters were set after analyzing the PNS of unlabeled cells. Thus, a region was defined for FITC and/or Cy5 fluorescence positive events.

D. Preparation of post-nuclear supernatant (PNS) for SOFA All manipulations were carried out at 4°C. Following endosome labeling, the cells were washed twice with 50 ml PBS and pelleted. The cell pellet was resuspended in 4 vol. PBS and homogenized with a ball-bearing homogenizer (Balch and Rothman, 1985). The resulting homogenate was centrifuged for 15 min at 4300 g (Rotixa/RP, Hettich) to obtain the postnuclear supernatant (PNS). The PNS was diluted 1:5 with PBS (pH 7.4) and immediately subjected to SOFA and generation of the pH standard curve, respectively (Murphy et al., 1989; Wilson and Murphy, 1989; Murphy, 1990).

References Bai, M., Campisi, L. and Freimuth, P. (1 9 9 4 ). Vitronectin receptor antibodies inhibit infection of HeLa and A549 cells by adenovirus type 12 but not by adenovirus type 2. J . V i r o l . 68, 5925-32. Balch, W.E. and Rothman, J.E. (1 9 8 5 ). Characterization of protein transport between successive compartments of the Golgi apparatus: asymmetric properties of donor and acceptor activities in a cell-free system. Arch. B i o c h e m . B i o p h y s . 240, 413-425.

D. Generation of pH standard curves and calculation of internal pH for flow cytometry and SOFA 50 µl aliquots of cells or PNS were resuspended in 250 µl of buffers of various pH. Buffers (pH 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5) were obtained by mixing 50 mM HEPES with 50 mM MES (both containing 50 mM NaCl, 30 mM ammonium acetate, 40 mM sodium azide and 1 µM nigericin) accordingly. The samples were left on ice for 5 min for ATP-depletion and for equilibration of intravesicular pH.

Bergelson, J.M., Cunningham, J.A., Droguett, G., Kurt Jones, E.A., Krithivas, A., Hong, J.S., Horwitz, M.S., Crowell, R.L. and Finberg, R.W. (1 9 9 7 ). Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. S c i e n c e . 275, 1320-3. Cain, C.C. and Murphy, R.F. (1 9 8 6 ). Growth inhibition of 3T3 fibroblasts by lysosomotropic amines: correlation with effects on intravesicular pH but not vacuolation. J . C e l l P h y s i o l . 129, 65-70.

E. Calculation of the pH of labeled compartments

Curiel, D., Agarwal, S., Wagner, E. and Cotten, M. (1 9 9 1 ). Adenovirus enhancement of transferrin-polylysinemediated gene delivery. P r o c . N a t l . A c a d . S c i . U S A 88, 8850-8854.

The mean fluorescence value for each fluorochrome of experimental samples (8 parallels) and samples of the pH standard curve (duplicates) was calculated and the corresponding mean autofluorescence of unlabeled cells was subtracted from each. The ratio of the resulting average FITC and Cy5 values was calculated for each condition and normalized to the value obtained for that sample after clamping

Defer, C., Belin, M., Caillet-Boudin, M. and Boulanger, P. (1 9 9 0 ). Human adenovirus-host cell interactions: comparative study with members of subgroups B and C. J . V i r o l . 64, 3661-3673.

32

Gene Therapy and Molecular Biology Vol 3, page 33 Fuchs, R., Male, P. and Mellman, I. (1 9 8 9 ). Acidification and ion permeabilities of highly purified rat liver endosomes. J . B i o l . C h e m . 264, 2212-2220.

Bowser, R. (1 9 8 9 ). Determination of the biochemical characteristics of endocytic compartments by flow cytometric and fluorometric analysis of cells and organelles. F l o w C y t o m e t r y : A d v a n c e d Research a n d C l i n i c a l A p p l i c a t i o n s , 221-254 (Yen, A. ed.) CRC Press, Inc., Florida.

Fujita, A., Sakagami, K., Kanegae, Y., Saito, I. and Kobayashi, I. (1 9 9 5 ). Gene targeting with a replicationdefective adenovirus vector. J . V i r o l . 69, 6180-6190.

Nemerow, G., Cheresh, D. and Wickham, T. (1 9 9 4 ). Adenovirus entry into host cells: a role for ! v integrins. T r e n d s C e l l B i o l . 4, 52-55.

Graham, F., Smiley, J., Russell, W. and Nairn, R. (1 9 7 7 ). Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J . G e n . V i r o l . 36, 5974.

Otero, M. and Carrasco, L. (1 9 8 7 ). Proteins are cointernalized with virion particles during early infection. V i r o l o g y 160, 75-80.

Greber, U. and Kasamtsu, H. (1 9 9 6 ). Nuclear targeting of SV40 and adenovirus. T r e n d s C e l l B i o l . 6, 189-195.

Prchla, E., Kuechler, E., Blaas, D. and Fuchs, R. (1 9 9 4 ). Uncoating of human rhinovirus serotype 2 from late endosomes. J . V i r o l . 68, 3713-3723.

Greber, U., Suomalainen, M., Stidwill, R., Boucke, K., Ebersold, M. and Helenius, A. (1 9 9 7 ). The role of the nuclear pore complex in adenovirus DNA entry. EMBO-J. 16, 5998-6007. Greber, U., Webster, P., Weber, J. and Helenius, A. (1 9 9 6 ). The role of the adenovirus protease on virus entry into cells. EMBO-J. 15, 1766-1777.

Prchla, E., Plank, C., Wagner, E., Blaas, D. and Fuchs, R. (1 9 9 5 ). Virus-mediated release of endosomal content in vitro: different behavior of adenovirus and rhinovirus serotype 2. J . C e l l B i o l . 131, 111-123.

Greber, U., Willetts, M., Webster, P. and Helenius, A. (1 9 9 3 ). Stepwise dismantling of adenovirus 2 during entry into cells. C e l l 75, 477-486.

Rybak, S.L. and Murphy, R.F. (1 9 9 8 ). Primary cell cultures from murine kidney and heart differ in endosomal pH. J . C e l l P h y s i o l . 176, 216-22.

Gruenberg, J. and Maxfield, F.R. (1 9 9 5 ). Membrane transport in the endocytic pathway. C u r r . O p i n . C e l l B i o l . 7 , 552-563.

Schober, D., Kronenberger, P., Prchla, E., Blaas, D. and Fuchs, R. (1 9 9 8 ). Major and minor receptor group human rhinoviruses penetrate from endosomes by different mechanisms. J . V i r o l . 72, 1354-1364.

Hong, S.S., Karayan, L., Tournier, J., Curiel, D.T. and Boulanger, P.A. (1 9 9 7 ). Adenovirus type 5 fiber knob binds to MHC class I alpha 2 domain at the surface of human epithelial and B lymphoblastoid cells. EMBO-J. 16, 2294-2306.

Seth, P. (1 9 9 4 ). Mechanism of adenovirus-mediated endosome lysis: role of the intact adenovirus capsid structure. B i o c h e m . B i o p h y s . R e s . C o m m u n . 205, 1318-1324.

Jones, N. and Shenk, T. (1 9 7 9 ). An adenovirus type 5 early gene function regulates expression of other early viral genes. P r o c . N a t l . A c a d . S c i . U S A 76, 3665-3669.

Seth, P., Pastan, I. and Willingham, M. (1 9 8 5 ). Adenovirusdependent increase in cell membrane permeability. J . B i o l . C h e m . 260, 9598-9602.

Mellman, I., Fuchs, R. and Helenius, A. (1 9 8 6 ). Acidification of the endocytic and exocytic pathways. A n n . R e v . B i o c h e m . 55, 663-700.

Wagner, E. (1 9 9 8 ). Effects of membrane-active agents in gene delivery. J . C o n t r o l . R e l . 53, 155-158.

Mukherjee, S., Ghosh, R. and Maxfield, F. Endocytosis. P h y s i o l . R e v . 77, 753-803.

Wagner, E., Zatloukal, K., Cotten, M., Kirlappos, H., Mechtler, K., Curiel, D. and Birnstiel, M. (1 9 9 2 ). Coupling of adenovirus to transferrin-polylysine/DNA complexes greatly enhances receptor-mediated gene delivery and expression of transfected genes. P r o c . N a t l . Acad. Sci. USA 89, 6099-6103.

(1 9 9 7 ).

Murphy, R. F., Powers, S. and Cantor, C. R. (1 9 8 4 ). Endosome pH measured in single cells by dual fluorescence flow cytometry: rapid acidification of insulin to pH 6. J . C e l l B i o l . 98, 1757-1762.

Wang, K., S, H., KapoorMunshi, A. and Nemerow, G. (1 9 9 8 ). Adenovirus internalization and infection require dynamin. J . V i r o l . 72, 3455-3458.

Murphy, R.F. (1 9 8 5 ). Analysis and isolation of endocytic vesicles by flow cytometry and sorting: demonstration of three kinetically distinct compartments involved in fluidphase endocytosis. P r o c . N a t l . A c a d . S c i . U S A 82, 8523-6.

Wickham, T., Filardo, E., Cheresh, D. and Nemerow, G. (1 9 9 4 ). Integrin alpha v beta 5 selectively promotes adenovirus mediated cell membrane permeabilization. J . C e l l B i o l . 127, 257-264.

Murphy, R.F. (1 9 9 0 ). Ligand binding, endocytosis, and processing. F l o w C y t o m e t r y a n d S o r t i n g , 355-366 (Melamed, R. M., Lindmo, T., Mendelssohn, M. L. eds.) Wiley-Liss, Inc., New York.

Wickham, T., Mathias, P., Cheresh, D. and Nemerow, G. (1 9 9 3 ). Integrins alpha v beta 3 and alpha v beta 5 promote adenovirus internalization but not virus attachment. C e l l 73, 309-319.

Murphy, R.F. (1 9 9 3 ). Models of endosome and lysosome traffic. A d v . C e l l M o l . B i o l . M e m b r . 1, 1-7.

Wilson, R.B. and Murphy, R.F. (1 9 8 9 ). Flow-cytometric analysis of endocytic compartments. M e t h o d s C e l l

Murphy, R.F., Roederer, M., Sipe, D.M., Cain, C.C. and

33

Schober et al: Adenovirus-induced endosome rupture B i o l . 31, 293-317

Struct. Funct. 10, 391-404.

Yoshimura, A. (1 9 8 5 ). Adenovirus-induced leakage of coendocytosed macromolecules into the cytosol. C e l l

34

Gene Therapy and Molecular Biology Vol 3, page 35 Gene Ther Mol Biol Vol 3, 35-44. August 1999.

Gene regulation in Herpesvirus saimiri and its implications for the development of a novel gene therapy vector Review Article

Adrian Whitehouse* and Alex J. Stevenson Molecular Medicine Unit, University of Leeds, St James's University Hospital, Leeds, LS9 7TF United Kingdom __________________________________________________________________________________ *Correspondence: A. Whitehouse Tel: +44-113-206 5865; Fax: +44-113-244 4475; E-mail: A.Whitehouse@leeds.ac.uk Received: 7 July, 1998; accepted: 23 July 1998

Summary We have investigated the potential of HVS as a human gene therapy vector and found that it is capable of infecting an extremely broad spectrum of human cell lines and primary cultures with efficiencies that are at least as good as (and in many cases better than) currently available vector s y s t e m s . L i k e o t h e r s w e found that the virus was capable o f stably transferring a functional heterologous gene by virtue o f episomal maintenance. Although transduced clones can be established in all cases, we have also been able to demonstrate low levels of virus production from t h e s e c e l l s . T h i s f i n d i n g n e c e s s i t a t e s t h e d e v e l o p m e n t o f disabled mutants for potential future clinical applications. Fundamental research carried out in this laboratory has identified the interactions between the two known transcriptional regulatory genes encoded by HVS. Overall, these results suggest that ORF50 and ORF57 are ideal essential candidate genes to delete in order to produce a replicationdisabled HVS. This will provide the basis for a novel gene therapy vector which is theoretically capable of addressing the problems faced by current vector systems.

of the immediate early genes. These viruses could only replicate in ‘helper’ cell lines which provided the IE gene product in trans (DeLuca et al., 1985; DeLuca and Schaffer, 1987). The vectors were engineered to carry heterologous genes in the deleted portions of their genomes, but were often toxic to the cells which they infected (Glorioso et al., 1985; Johnson et al., 1992; Sabel et al., 1995).

I. Herpesviruses as gene therapy vectors Herpesviruses are classified as large DNA viruses having genomes of between 100 and 250 kb. They are divided into alpha, beta and gamma sub-groups on the basis of their biological and genetic properties (Roizman et al., 1981). As a family their advantages as gene therapy vectors relate to an ability to package large DNA insertions and establish lifelong latent infections in which the genomic material exists as a stable episome. Nearly all of the research in this field has focused on the use of Herpes simplex virus (HSV) vectors for gene transfer to the nervous system (Coffin and Latchman, 1996). HSV encodes several proteins which modulate viral and cellular gene expression via a temporal cascade of immediate-early (IE), early and late genes (Honess and Roizman, 1974; DeLuca and Schaffer, 1985). These first generation HSV based vectors were disabled by the deletion of one or more

A second broad category of HSV based vectors are amplicons. These are plasmids containing an HSV lytic replication origin and terminal packaging signals. They can be amplified and packaged into infectious HSV-1 particles in the presence of helper-virus (Spaete and Frenkel, 1982; Kwong and Frenkel, 1984; Geller and Breakefield, 1988; Geller and Freese, 1990). As such they constitute a cloning vehicle which can efficiently carry genetic information between prokaryotic and eukaryotic cells. Amplicons retain many of the characteristics of

35

Whitehouse and Stevenson: Herpesvirus saimiri as a gene therapy vector standard HSV vectors but viral stocks tend to have lower titres, making them less useful for gene therapy applications.

homologous sequences are found in approximately equivalent locations and in the same relative orientation. However, conserved gene blocks are separated by unique genes with respect to each virus (Albrect and Fleckenstein, 1990; Nicholas et al., 1992; Russo et al., 1996; Virgin et al., 1997).

A problem shared by all HSV based vector systems is the fact that the genome has no ‘latent origin’ of DNA replication, meaning that a state of episomal maintenance cannot be established in dividing cells. However, Epstein Barr virus (EBV), a member of the gamma-herpesviruses, is capable of establishing a latent state in dividing cells where the viral episome replicates co-ordinately with cell division and is inherited by all progeny cells. Such a vector derived from EBV may be suitable for stem cell gene therapy. However, EBV is associated with a number of human malignancies and lymphoproliferative disorders, necessitating extensive modification of the virus genome to eliminate those genes involved in transformation.

III. Potential of HVS as a gene therapy vector All herpesvirus vector systems which have previously been assessed were based on human herpesviruses and are inevitably likely to be ineffective in the majority of individuals due to the inherent immune response induced by the wild type virus. A herpesvirus of non-human origin, capable of infecting human cells without a cytopathic effect therefore represents an attractive candidate as a gene therapy vector, as there should be no immediate inate immune response in the recipient. Earlier publications have demonstrated that a selectable HVS has the ability to persist in a variety of human cell lines for long periods of time apparently without the production of infectious progeny (Grassmann and Fleckenstein 1989; Simmer et al., 1991).

One example of an alternative strategy, which has been employed with some success in the laboratory, is the construction of HSV amplicons containing EBV sequences that maintain the plasmid as an episome in the infected cell nucleus (Wang and Vos, 1996; Wang et al., 1997). Such developments are bound to lead to improvements in Herpes Simplex virus based vectors and the eventual creation of ‘niche’ gene therapy applications. However, all of the current options suffer from inherent problems and limitations which are far from trivial. This review highlights a potential alternative, Herpesvirus saimiri (HVS), and illustrates the importance of basic research in the quest for better vector systems.

In order to assess the potential of HVS as a possible gene therapy vector, we have generated a HVS recombinant virus (based on a non-transforming strain, A-11) which expresses EGFP (Cormack et al., 1996), and the neomycin resistance gene under the control of distinct promoters. These heterologous genes have been cloned into the repeat regions of the HVS genome, theoretically preventing alteration of the wild type virus phenotype (Figure 1 ). Analysis using this virus, which can be grown to a high titre, has demonstrated infection of a wide variety of human cell lines at approaching 100% efficiency including, A549 (lung carcinoma), HT-29 (colonic adenocarcinoma), MIA-PACA (pancreatic carcinoma), K562 (chronic myelogenous leukaemia), Jurkat (T-cell lymphoma), Molt-4 (T-cell leukaemia) and Raji (Burkitt's lymphoma) cells (Figure 2) (Stevenson et al., 1999). In contrast to previously published work we have detected low levels of virus replication in all of these cell lines at early stages post infection, even in the absence of apparent cytopathic effects. However, the virus DNA is clearly able to establish a latent episomal state within the cell which segregates to the progeny upon division. Figure 3 shows the development of a clone of A549 cells resulting from a single infected cell. The period of this experiment was four weeks, but the clone is still growing (and remains bright green) six months later. This result and similar data generated in other cell lines are extremely encouraging and we believe this system offers enormous potential for the delivery of therapeutic genes to cancerous cells, as well as to bone marrow and stem cells.

II. Herpesvirus saimiri Herpesvirus saimiri is a lymphotrophic rhadinovirus (2 herpesvirus) of squirrel monkeys (Saimiri sciureus), which persistently infects its natural host without causing any obvious disease. However, HVS infection of other species of New World primates results in fulminant polyclonal T-cell lymphomas and lymphoproliferative diseases (Fleckenstein and Desrosiers, 1982). Certain strains of HVS are also capable of transforming human T lymphocytes to continuous growth in vitro (Beisinger et al., 1992). The genome of HVS (strain A11) consists of a unique internal low G+C content DNA segment (L-DNA) of approximately 110 kbp which is flanked by a variable number of 1444 bp high G+C content tandem repetitions (H-DNA) (Albrecht et al., 1992). Analysis indicates it shares significant homology with other herpesviruses: EBV, bovine herpesvirus 4, Kaposi’s sarcoma-associated herpesvirus (KSHV or human herpesvirus 8) and murine gammaherpesvirus 68 (MHV68) (Albrect and Fleckenstein, 1990; Bublot et al., 1992; Gompels et al., 1988a;b; Neipel et al., 1997; Russo et al., 1996; Virgin et al., 1997). The genomes of EBV, KSHV, MHV68 and HVS have been shown to be generally colinear, in that 36

Gene Therapy and Molecular Biology Vol 3, page 37

F i g u r e 1 . Construction of GFP/Neo virus. A recombinant virus was generated by transfection of OMK cells with the recombination vector followed by super-infection with wild type virus. Recombinants were initially selected for by the addition of Geneticin to the culture medium followed by two rounds of plaque purification.

F i g u r e 2 . Examples of human cancer cell lines infected with GFP/Neo virus. (a) SW480 (colonic carcinoma) (b) HT-29 (colonic carcinoma) (c) Miapaca (pancreatic carcinoma).

F i g u r e 3 . Segregation of GFP amongst dividing human lung carcinoma cells (A549). A549 cells were infected with GFP/Neo virus and selected in the presence of Geneticin. The figure shows the development of an individual clone over the period of four weeks.

37

Whitehouse and Stevenson: Herpesvirus saimiri as a gene therapy vector

F i g u r e 4 . Diagrammatic representation of the ORF 50 transcripts. ORF 50 produces two transcripts, the first is spliced containing a single intron and is detected at early times during the productive cycle, whereas the second is expressed later and is produced from a promoter within the second exon.

al., 1992), a sequence-specific transactivator (Gruffat et al., 1990). The HVS R gene or ORF 50 produces two transcripts. The first is spliced containing a single intron and is detected at early times during the productive cycle, whereas the second is expressed later and is produced from a promoter within the second exon (Figure 4). The spliced transcript is 5-fold more potent in activating the delayed-early ORF 6 promoter. The function of the nonspliced transcript is unclear (Nicholas et al., 1992; Whitehouse et al., 1997a). Further analysis of ORF 50 indicates it responds to particular DNA sequences specifically contained within the promoters of the genes it transactivates. Deletion and gel retardation analysis have identified a consensus ORF50-recognition sequence, CCN 9GG, required for transactivation by both ORF 50 transcripts (Whitehouse et al., 1997b). The response elements have significant homology to the EBV.R response element consensus sequences, GNCCN9GGNG. It

To develop this virus further as a gene therapy vector and to minimise the risk of pathogenicity, disabled HVS vectors are required. In order to generate a replicationdisabled vector, genes essential for the replication of the virus must be deleted. Ideal candidate genes to disable viruses are those expressed early in the viral replication cycle and which are involved in the regulation of viral gene expression. The following section discusses these genes and the role they play in HVS replication.

IV. Gene regulation in HVS A. The ORF 50 gene products We have recently identified the two major transcriptional regulating genes encoded by HVS. The first transcriptional activator is homologous to the EBV BRLF1 gene product, R (Nicholas et al., 1991; Albrecht et

38

Gene Therapy and Molecular Biology Vol 3, page 39 has been shown by guanine methylation studies that the CCN 9GG motif is essential for EBV.R binding, suggesting that the R binds to adjacent major grooves of the DNA (Gruffat et al., 1990; 1992; Gruffat & Sergeant, 1994). The ORF 50 response elements map to within 32 bp which contain a CCN 9GG motif. However, the flanking sequences are significantly different to the EBV.R response elements, suggesting that the ORF 50 gene products have different sequences required for recognition and fixation of the proteins to their target. At present we are unable to determine using gel retardation analysis if the ORF 50 gene products bind directly to the response elements, or whether the retarded complex identified is due to the recruitment of host cell proteins by ORF 50.

target gene promoter sequences and appear to be mediated at the post-transcriptional level. At present we are unable to determine the actual effect of ORF 57 i.e. whether it affects the processing, transport or translational efficiency of mRNA. The more widely studied ORF 57 homologue, ICP27, appears to act post-transcriptionally by affecting mRNA processing suggesting ICP27 regulates usage of polyadenylation sites as a means of controlling gene expression (McGregor et al., 1996; McLauchlan et al., 1992). It has also been demonstrated that a bacterially expressed ICP27 fusion protein specifically binds to the 3’ ends of RNA leading to accumulation and increased half life of the mRNAs (Brown et al., 1995). The RNA binding motif (residues 138 and 152), is similar to an RGG box motif and this is believed to be an RNA binding determinant (Mears and Rice, 1996). Furthermore, it has recently been shown that ICP27 shuttles between the nucleus and cytoplasm. Shuttling occurs only at late stages during infection and is dependent on the coexpression of HSV late mRNAs, suggesting that ICP27 facilitates the export of late mRNAs (Soliman et al., 1997; Phelan et al., 1997). However, not all ICP27 homologues, including ORF 57, contain an homologous RGG box motif. Nevertheless, ORF 57 does encode an arginine-rich amino terminus, which may contain alternative RNA binding determinants. Deletion and mutational analysis of the N-terminal region of ORF 57 may help to clarify its role, if any, in RNA binding.

We believe that ORF 50 probably binds to the response sequences because of its homology with EBV.R protein which has been purified and shown to specifically recognise its response elements (Gruffat & Sergeant, 1994). However, further analysis of the ORF 50 response element by mutagenesis is required as is the production of purified ORF 50 gene products to investigate these hypotheses. The EBV.R protein has been shown to transactivate three promoters. We believe that ORF 50s gene products also transactivate multiple promoters. We have searched the HVS genome for additional ORF 50 response elements using the motif, CCN 9GG and have identified 69 putative response elements. Further characterisation of these putative elements to localise in a promoter region of a defined ORF expressed delayed-early or late in the virus replication cycle, has identified 10 possible promoter regions which may contain ORF 50 responsive elements (Whitehouse et al., 1997b). We are currently examining these genes for possible transactivation by either of the ORF 50 gene products and investigating whether late genes are transactivated by the later ORF 50b transcript. Alternatively, the ORF 50b gene product, which has been shown to transactivate ORF 6 to a lesser extent, may compete with ORF 50a for binding to the response elements, thus acting as a negative regulator of transcription.

In addition to ORF 57’s transactivating capabilities we have demonstrated that it can downregulate gene expression, specifically on intron-containing genes (Whitehouse et al., 1998a;b). In addition, the more widely studied homologue, ICP27, has been shown to be involved in the switch from early to late gene expression (McGregor et al., 1996; McLauchlan et al., 1989; 1992; Rice et al., 1993; Sandri-Goldin et al., 1995) and in the downregulation of viral IE and early genes. It is also required for the expression of late genes (McLauchlan et al., 1989; Rice et al., 1993; Sandri-Goldin et al., 1995; Sandri-Goldin and Mendoza, 1992). Furthermore, ICP27 contributes to the shut off of host cell protein synthesis and contributes to a decrease in cellular mRNA levels during infection, as deletion mutant infections result in increased levels of cellular protein synthesis and mRNA than do wild type infections (Hardwicke and Sandri-Goldin, 1994; Hardy and Sandri-Goldin, 1994; Hibbard and SandriGoldin, 1995; Schroder et al., 1989).

B. The ORF 57 gene product The second transactivator encoded by ORF 57 is homologous to genes identified in all classes of herpesviruses. These include the EBV transactivator encoded by BMLFI, ICP27 of HSV, BICP27 in bovine herpesvirus 1, ORF 4 encoded by varicella-zoster virus, UL69 in human cytomegalovirus, and ICP27 in equine herpesvirus 1 (Davidson and Scott, 1986; Kenney et al., 1989; Nicholas et al., 1988; Perera et al., 1994; Singh et al., 1996; Winkler et al., 1994; Zhao et al., 1995). The ORF 57 gene product has transregulatory functions which, unlike ORF 50’s gene products, are independent of the

HVS contains 76 major open reading frames, of which only 4 contain introns. This suggests that this virus makes limited use of the host cellular splicing machinery. Preliminary experiments have shown that during HVS infection, antigens associated with the small nuclear ribonucleoproteins (snRNPs), which are subunits of splicing complexes (reviewed in Kramer, 1995), are 39

Whitehouse and Stevenson: Herpesvirus saimiri as a gene therapy vector redistributed in the nucleus and become concentrated in specific intranuclear structures (Cooper et al., 1999). This redistribution has also been observed during herpes simplex virus infection (Phelan et al., 1993; Sandri-Goldin et al., 1995). Sequence analysis has shown that ORF 57 is more highly conserved with respect to other members of the ICP27 family at the 3’-terminal region of the gene. We believe that the ORF 57 gene product contains a functional domain within the C-terminus which is required for the repressor function of this protein. It has been demonstrated that the C-terminal domain of ICP27 must remain intact for its inhibitory effect (McMahon and Schaffer, 1990; Sandri-Goldin et al., 1995). This region contains a cysteine-histidine rich region which resembles a single “zinc finger-like” motif or “zinc knuckle” which is conserved in all ICP27 homologs including ORF 57 (histidine residue 383 and cysteine residues 387 and 392 in ORF 57). Similar motifs occur in a number of splicing factors (Sandri-Goldin and Hibbard, 1996). Further studies are been undertaken to determine if this domain is essential for the repressor activity of ORF 57 and to determine which cellular genes interact with ORF 57.

downregulated at a similar time during the replication cycle. This series of events regulating gene expression in HVS differs from other herpesviruses. IE genes in all herpesviruses are defined as those which can be transcribed efficiently in the absence of de novo protein synthesis. Therefore, they mostly encode transcriptional regulators which are required for viral gene expression. However, despite their obvious role in virus replication the major IE genes are not conserved amongst herpesviruses. For example, during HSV replication five IE genes; ICP0, ICP4, ICP22, ICP27, ICP47 are expressed in the absence of viral protein synthesis. The fact that only one of these genes is conserved in HVS (ORF 57 is homologous to ICP27), may be unsurprising as HSV and HVS belong to different subfamilies of the herpesvirus genera. However, EBV, a member of the same subfamily as HVS, also differs from HVS in the IE genes it encodes. Upon reactivation, two major IE genes are expressed which are the key transactivating genes in EBV. The first, the IE BZLF1 gene product, Z, is sufficient to trigger reactivation, when overexpressed in latently infected cells (Buisson et al., 1989; Furnari et al., 1994; Rooney et al., 1989). Z is able to transactivate several promoters containing Z responsive elements, as well as to regulate its own promoter (Furnari et al., 1994; Liebermann et al., 1989; Packman et al., 1990; Rooney et al., 1989). The second IE protein, the BRLF1 gene product, R, is also a sequence specific transactivator. HVS does not encode a Z homologue. However, ORF 50 is homologous to the EBV R protein. Overall, this suggests that the two genes encoded by HVS which are homologous to genes found in other herpesviruses play a critical role in the HVS replication cycle.

C. A novel feedback mechanism which regulates HVS gene expression More recently, we have demonstrated that these two major transcriptional control genes interact to regulate HVS gene expression via a novel feedback mechanism summarised in F i g u r e 5 . (Whitehouse et al., 1998b). The ORF 57 gene is produced at low levels early in the replication cycle until transactivated by the early ORF 50a gene product. Sequences within the ORF 57 promoter contain an ORF 50 response element which are essential for transactivation by the ORF 50a gene product and which result in an increase in RNA levels of the ORF 57 transcript. In addition, ORF 50a transactivates other genes which contain ORF 50 response elements within their promoters, for example the major DNA binding protein (Nicholas et al., 1992; Whitehouse et al., 1997a;b). Once transactivated by ORF 50a the ORF 57 gene product has several functions. As discussed previously, it has been shown to transactivate a range of HVS genes through posttranscriptional modification. Second, it downregulates ORF 50a, due to the presence of the intron within its coding region (Whitehouse et al., 1998a). Therefore we believe a feedback mechanism is in operation involving ORF 50a and ORF 57, which regulates gene expression in HVS, whereby a gene is downregulated by the product of the gene is has previously transactivated. Third, we believe the intron containing ORF 57 gene is responsible for its own downregulation by the same mechanism as that with which it represses ORF 50a, as both genes are

Acknowledgements This work was supported in part from grants from Yorkshire Cancer Research, The Candlelighters Trust, West Riding Medical Trust, Medical Research Council and the Wellcome Trust.

References Albrecht, J. C., and Fleckenstein, B. (1 9 9 0 ). Structural organization of the conserved gene block of herpesvirus saimiri coding for DNA polymerase, glycoprotein B, and major DNA binding protein. V i r o l o g y 174, 533-542. Albrecht, J. C., Nicholas, J., Biller, D., Cameron, K. R., Biesinger, B., Newman, C., Wittman, S., Craxton, M. A., Coleman, H., Fleckenstein, B. and Honess, R. W. (1 9 9 2 ). Primary structure of the herpesvirus saimiri

40

Gene Therapy and Molecular Biology Vol 3, page 41

F i g u r e 5 . Schematic representation of the role and interactions of the ORF 57 and ORF 50 genes which regulate gene expression in the HVS replication cycle. (Abbreviations, DE - Delayed Early; RE - Response Elements). genome. J . V i r o l . 66, 5047-5058.

Coffin, R. S. and Latchman, D. S. (1 9 9 6 ). Herpes simplex virus-based vectors. In: Latchman, D. S. (ed). G e n e t i c m a n i p u l a t i o n o f t h e n e r v o u s s y s t e m . Academic Press: London 99-111.

Beisinger, B., Muller-Fleckenstein, I., Simmer, B., Lang G., Wittmann, S., Platzer, E., Desrosiers, R. C. and Fleckenstein, B. (1 9 9 2 ). Stable growth transformation of human T lymphocytes by herpesvirus saimiri. P r o c . N a t l . A c a d . S c i . U S A 89, 3116-3119.

Cooper M, Goodwin DJ, Hall KT, Stevenson AJ, Meredith DM, Markham AF, and Whitehouse A (1 9 9 9 ) The gene product encoded by ORF57 of Herpesvirus Saimiri regulates the redistribution of the splicing factor, SC-35. J . G e n . V i r o l . In Press.

Brown, C. R., Nakamura, M. S., Mosca, J. D., Hayward, G. S., Straus, S. E. and Perera, L. P. (1 9 9 5 ). Herpes simplex virus trans-regulatory protein ICP27 stabilises and binds to 3’ ends of labile mRNA. J . V i r o l . 69, 7187-7195.

Cormack, B.P., Valdivia, R. H. and Falkow, S. (1 9 9 6 ). FACS-optimised mutants of the green fluorescent protein (GFP). Gene 173, 33-38.

Bublot, M., Manet, E., Lequarre, A. S., Albrecht, J. C., Nicholas, J., Fleckenstein, B., Pastoret, P. P. and Thiry, E. (1 9 9 2 ). Genetic relationships between bovine herpesvirus 4 and the gamma-herpesviruses Epstein-Barr and herpesvirus saimiri. V i r o l o g y 190, 654-665.

Davison, A. J. and Scott, J. E. (1 9 8 6 ). The complete sequence of varicella-zoster virus. J . G e n . V i r o l . 67, 1759-1816.

Buisson, M., Manet, E., Trescol-Biemont, M.-C., Gruffat, H., Durand, B. and Sergeant, A. (1 9 8 9 ). The Epstein-Barr virus (EBV) early protein EB2 is a post transcriptional activator expressed under the control of EBV transcription factors EB1 and R. J . V i r o l . 63, 5276-5284.

DeLuca, N. A. and Schaffer, P. A. (1 9 8 5 ). Activation of immediate-early, early and late promoters by temperature sensitive and wild-type forms of herpes simplex virus type 1 protein ICP4. M o l . C e l l . B i o l . 5, 1997-2009. DeLuca, N. A. and Schaffer, P. A. (1 9 8 7 ). Activities of herpes simplex virus type 1 (HSV-1) ICP4 genes specifying nonsense peptides. N u c l e i c Acids R e s 15, 44914511.

Burd, C. G. and Dreyfuss, G. (1 9 9 4 ). Conserved structures and diversity of functions of RNA-binding proteins. S c i e n c e 265, 615-621.

41

Whitehouse and Stevenson: Herpesvirus saimiri as a gene therapy vector DeLuca, N. A., McCarthy, A. M. and Schaffer, P. A. (1 9 8 5 ). Isolation and characterisation of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate-early regulatory protein ICP4. J . V i r o l . 56, 558-570.

to the decrease in cellular mRNA levels during infection. J . V i r o l . 68, 4797-4810. Hardy, W. R. and Sandri-Goldin R. M. (1 9 9 4 ). Herpes simplex virus inhibits host cell splicing, and the regulatory protein ICP27 is required for this effect. J . V i r o l . 68, 7790-7799.

Fleckenstein, B. and Desrosiers, R. C. (1 9 8 2 ). Herpesvirus saimiri and herpesvirus ateles, p. 253-332. In B. Roizman (ed.), T h e h e r p e s v i r u s e s , vol. 1. Plenum Press, New York.

Hibbard, M. K. and Sandri-Goldin, R. M. (1 9 9 5 ). Argininerich regions succeeding the nuclear localisation region of herpes simplex virus type 1 regulatory protein ICP27 are required for efficient nuclear localisation and late gene expression. J . V i r o l . 69, 4656-4667.

Furnari, F. B., Zacny, V., Quinlivan, E. B., Kenney, S. and Pagano, J. S. (1 9 9 4 ). RAZ, an Epstein-Barr virus transdominant repressor that modulates the viral reactivation mechanism. J . V i r o l . 68, 1827-1836.

Honess, R. W. and Roizman, B. (1 9 7 4 ). Regulation of herpesvirus macromolecular synthesis I. Cascade regulation of the synthesis of three groups of viral proteins. J . V i r o l . 14, 8-19.

Geller, A.I. and Breakefield, X.O. (1 9 8 8 ). A defective HSV-1 vector expresses Escherichia coli b-galactosidase in cultured peripheral neurons. S c i e n c e 241, 1667-1669.

Johnson, P. A., Miyanohara, A., Levine, F., Cahill, T. and Freidmann, T. (1 9 9 2 ). Cytotoxicity of a replicationdefective mutant of herpes simplex virus type 1. J . V i r o l . 66, 2952-2965.

Geller, A.I. and Freese, A. (1 9 9 0 ). Infection of cultured central nervous system neurons with a defective herpes simplex virus 1 vector resulting in stable expression of Escherichia coli b-galactosidase. P r o c . N a t l . Acad. S c i . U S A 87, 1149-1153.

Kenney, S., Kamine, J., Holley-Guthrie, E. A., Mar, E.-C., Lin, J. C., Markovitz, D. and Pagano, J. (1 9 8 9 ). The Epstein-Barr virus immediate-early gene product, BMLF1, acts in trans by a posttranscriptional mechanism which is reporter gene dependent. J . V i r o l . 63, 3870-3877.

Glorioso, J. C., DeLuca, N. A. and Fink, D. J. (1 9 9 5 ). Development and application of herpes simplex virus vectors for human gene therapy. A n n u R e v M i c r o b i o l 49, 675-710.

Kramer, W. (1 9 9 5 ). The biochemistry of pre-mRNA splicing, p.35-55. In A. I. Lamond (ed.), Pre-mRNA processing. Springer-Verlag, R.G. Landes Company, Austin, Tex.

Gompels, U. A., Craxton, M. A. and Honess, R. W. (1 9 8 8 a ). Conservation of gene organisation in the lymphotrophic herpesvirus saimiri. J . V i r o l . 62, 757-767.

Kwong, A.D. and Frenkel, N. (1 9 8 4 ). Herpes simplex virus amplicon: effect of size on replication of constructed defective genomes containing eukaryotic DNA sequences. J . V i r o l . 51, 595-603.

Gompels, U. A., Craxton, M. A. and Honess, R. W. (1 9 8 8 b ). Conservation of glycoprotein H (gH) in herpesvirus: Nucleotide sequence of the gH gene from herpesvirus saimiri. J . G e n . V i r o l . 69, 2819-2829.

Lieberman, P. M., Hardwick, J. M. and Hayward, S. D. (1 9 8 9 ). Responsiveness of the Epstein-Barr virus NotI repeat promoter to the Z transactivator is mediated in a cell-specific manner by two independent signal regions. J . V i r o l . 63, 3040-3050.

Grassmann, R. and Fleckenstein, B. (1 9 8 9 ). Selectable recombinant herpesvirus saimiri is capable of persisting in a human cell line. J . V i r o l . 63, 1818-1821. Gruffat, H. and Sergeant, A. (1 9 9 4 ). Characterization of the DNA-binding site repertoire for the Epstein-Barr virus transcription factor R. N u c l e i c A c i d s R e s 22, 11721178.

McGregor, F., Phelan, A., Dunlop, J. and Clements, J. B. (1 9 9 6 ). Regulation of herpes simplex virus poly(A) usage and the action of the immediate-early protein IE61 in the early-late switch. J . V i r o l . 70, 1931-1940.

Gruffat, H., Duran, N., Buisson, M., Wild., F., Buckland, R. and Sergeant, A. (1 9 9 2 ). Characterization of an Rbinding site mediating the R-induced activation of the Epstein-Barr virus BMLF1 promoter. J . V i r o l . 66, 4652.

McLauchlan, J., Phelan, A., Loney, C., Sandri-Goldin, R. M. and Clements, J. B. (1 9 9 2 ). Herpes simplex virus IE63 acts at the posttranscriptional level to stimulate viral mRNA 3’ processing. J . V i r o l . 66, 6939-6945.

Gruffat, H., Manet, E., Rigolet, A. and Sergeant, A. (1 9 9 0 ). The enhancer factor R of Epstein-Barr virus (EBV) is a sequence specific DNA binding protein. N u c l e i c A c i d s R e s 18, 6835-6843.

McLauchlan, J., Simpson, S. and Clements, J. B. (1 9 8 9 ). Herpes simplex virus induces a processing factor that stimulates poly(A) site usage. C e l l 59, 1093-1105. McMahon, L. and Schaffer, P. A. (1 9 9 0 ). The repressing and enhancing functions of the herpes simplex virus regulatory protein ICP27 map to C-terminal regions. J . V i r o l . 64, 3471-3485.

Hardwick, J. M., Tse, L., Applegren, N., Nicholas, J. and Veliuona, M. A. (1 9 9 2 ). The Epstein-Barr virus R transactivator (Rta) contains a complex, potent activation domain with properties different from those of VP16. J . V i r o l . 66, 5500-5508.

Mears, W. E. and Rice, S. A. (1 9 9 6 ). The RGG box of the herpes simplex virus ICP27 protein mediates RNA binding activity and determines in vivo methylation. J . V i r o l . 70, 7445-7453.

Hardwicke, M. A. and Sandri-Goldin, R. M. (1 9 9 4 ). The herpes simplex virus regulatory protein ICP27 contributes

42

Gene Therapy and Molecular Biology Vol 3, page 43 Neipel, F., Albrecht, J. C. and Fleckenstein, B. (1 9 9 7 ). Cellhomologous genes in Kaposi’s sarcoma associated rhadinovirus human herpesvirus 8: Determinants of its pathogenicity? J . V i r o l . 71, 4187-4192.

S., Chang, Y. and Moore, P. S. (1 9 9 6 ). Nucleotide sequences of the Kaposi sarcoma-associated herpesvirus (HHV8). P r o c . N a t l . Acad. S c i . USA 93, 1486214867.

Nicholas, J., Cameron, K. R., Coleman, H., Newman, C. and Honess, R. W. (1 9 9 2 b ). Analysis of nucleotide sequence of the rightmost 43 kbp of herpesvirus saimiri (HVS) LDNA: General conservation of genetic organization between HVS and Epstein-Barr virus. V i r o l o g y 188, 296-310.

Sabel, B.A., Vick A. and Holt, V. (1 9 9 5 ). Neurotoxic reactions of CNS following gene transfer with defective herpes simplex virus (HSV-1) vector. Neuroreport 6, 2447-2449. Sandri-Goldin, R. M. and Hibbard, M. K. (1 9 9 6 ). The herpes simplex virus type 1 regulatory protein ICP27 coimmunoprecipitates with anti-sm antiserum, and the C terminus appears to be required for this interaction. J . V i r o l . 70, 108-118.

Nicholas, J., Coles, L. S., Newman, C. and. Honess, R. W. (1 9 9 1 ). Regulation of the herpesvirus saimiri (HVS) delayed-early 110-kilodalton promoter by HVS immediate-early gene products and a homolog of the Epstein-Barr virus R transactivator. J . V i r o l . 65, 24572466.

Sandri-Goldin, R. M. and Mendoza, G. E. (1 9 9 2 ). A herpesvirus regulatory protein appears to act posttranscriptionally by affecting mRNA processing. G enes D e v . 6, 848-863.

Nicholas, J., Gompels, U. A., Craxton, M. A. and Honess, R. W. (1 9 8 8 ). Conservation of sequence and function between the product of the 52-kilodalton immediate-early gene of the herpesvirus saimiri and the BMLF1-encoded transcriptional effector (EB2) of the Epstein-Barr virus. J . V i r o l . 62, 3250-3257.

Sandri-Goldin, R. M., Hibbard, M. K. and Hardwicke, M. A. (1 9 9 5 ). The C-terminal repressor region of herpes simplex virus type 1 ICP27 is required for the redistribution of small nuclear ribonucleoprotein particles and splicing factor SC35; however, these alterations are not sufficient to inhibit host cell splicing. J . V i r o l . 69, 6063-6076.

Packham, G., Economou, A., Rooney, C.M., Rowe, D.T. and Farrell, P.J. (1 9 9 0 ). Structure and function of the Epstein-Barr virus BZLF1 protein. J . V i r o l . 64, 21102116.

Schoder, H. C., Falke, D., Weise, K., Bachman, M., CarmoFonseca, M., Zaubitzer, T. and Muller, W. E. G. (1 9 8 9 ). Change of processing and nucleoplasmic transport of mRNA in HSV-1 infected cells. Virus Res. 13, 61-78.

Perera, L. P., Kaushal, S., Kinchington, P. R., Mosca, J. D., Hayward, G. S. and Straus, S. E. (1 9 9 4 ). Varicella-zoster virus open reading frame 4 encodes a transcriptional activator that is functionally distinct from that of herpes simplex virus homolog ICP27. J . V i r o l . 68, 24682477.

Simmer, B., Alt, M., Buckreus, I., Fleckenstein, B., Platzer, E. and Grassmann, R. (1 9 9 1 ). Persistence of selectable herpesvirus saimiri in various human haematopoietic and epithelial cell lines. J . G e n . V i r o l . 72, 1953-1958.

Phelan, A., Carmo-Fonseca, M., McLauchlan, J., Lamond, A. I. and Clements J. B. (1 9 9 3 ). A herpes simplex virus type 1 immediate early gene product IE63, regulates small nuclear ribonucleoprotein distribution. P r o c . N a t l . Acad. Sci. USA 90, 9056-9060.

Singh, M., Fraefel, C., Bello, L. J., Lawrence, W. C. and Schwyzer, M. (1 9 9 6 ). Identification and characterisation of BICP27, an early protein of bovine herpesvirus 1 which may stimulate mRNA 3’ processing. J . G e n . V i r o l . 77, 615-625.

Phelan, A., Dunlop, J., and Clements, J. B. (1 9 9 6 ). Herpes simplex virus type 1 protein IE63 affects nuclear export of virus intron-containing transcripts. J . V i r o l . 70, 52555265.

Spaete, R.R. and Frenkel, N. (1 9 8 2 ). The herpes simplex virus amplicon: a new eukaryotic defective-virus-cloningamplifying vector. C e l l 30, 295-304.

Rice, S. A., Lam, V. and Knipe, D. M. (1 9 9 3 ). The acidic amino-terminal region of the herpes simplex virus type 1 alpha protein ICP27 is required for an essential lytic function. J . V i r o l . 6 7 , 1778-1787.

Stevenson, A.J., Cooper, M., Griffiths, J.C., Gibson, P.G., Whitehouse, A., Jones, E.F., Kinsey, S.E., Markham, A.F. and Meredith, D.M. (1 9 9 9 ) Assessment of herpesvirus saimiri as a potential gene therapy vector. J Med Virol, In Press.

Roizman, B., Carmichael, L. E., Deinhardt, F., de-The, G., Nahmias, A. J., Plowright, W., Rapp, F., Sheldrick, P., Takahashi, M. and Wolf, K. (1 9 8 1 ). Herpesviridae. Definition, provisional nomenclature and taxonomy. I n t e r v i r o l o g y 16, 201-217.

Virgin, H. W., Latreille, P., Wamsley, P., Hallsworth, K., Weck, K. E., Dal Canto, A. J. and Speck, S. H. (1 9 9 7 ). Complete sequence and genomic analysis of murine gammaherpesvirus 68. J . V i r o l . 71, 5894-5904. Wang, S. and Vos, J. (1 9 9 6 ). A hybrid herpesvirus infection vector based on Epstein-Barr virus and herpes simplex virus type 1 for gene transfer into human cells in vitro and in vivo. J . V i r o l . 70, 8422-8430.

Rooney, C. M., Rowe, D. T., Ragot, T. and Farrell, P. J. (1 9 8 9 ). The spliced BZLF1 gene of Epstein-Barr virus (EBV) transactivates an early EBV promoter and induces the virus production cycle. J . V i r o l . 63, 3109-3116.

Wang, S., Di, S., Young, W.-B., Jacobson, C. and Link, C. J. Jr (1 9 9 7 ). A novel herpesvirus amplicon system for in

Russo, J. J., Bohenzhy, R. A., Chein, M. C., Chen, J., Yan, M., Maddalena, D., Parry, J. P., Peruzzi, D., Edelman, I.

43

Whitehouse and Stevenson: Herpesvirus saimiri as a gene therapy vector vivo gene delivery. Gene Ther 4, 1132-1141. Whitehouse, A., Carr, I. M., Griffiths, J. C. and Meredith, D. M. (1 9 9 7 a ). The herpesvirus saimiri ORF 50 gene, encoding a major transcriptional activator homologous to the Epstein-Barr Virus R protein, is transcribed from two distinct promoters of different temporal phases. J . V i r o l . 71, 2550-2554. Whitehouse, A., Cooper, M. and Meredith, D. M. (1 9 9 8 a ). The IE gene product encoded by ORF 57 of herpesvirus saimiri modulates gene expression at a posttranscriptional level. J . V i r o l . 72, 856-861. Whitehouse, A., Cooper, M., Hall, K. and Meredith, D. M. (1 9 9 8 b ). The open reading frame (ORF) 50a gene product regulates ORF 57 gene expression in herpesvirus saimiri. J . V i r o l . 72, 1967-1973. Whitehouse, A., Stevenson, A. J., Cooper, M. and Meredith, D. M. (1 9 9 7 b ). Identification of a cis-acting element within the herpesvirus saimiri ORF6 promoter that is responsive to the HVS.R transactivator. J . G e n . V i r o l . 78, 1411-1415. Winkler, M., Rice, S. A. and Stamminger, T. (1 9 9 4 ). UL69 of human cytomegalovirus, an open reading frame with homology to ICP27 of herpes simplex virus, encodes a transcriptional activator. J . V i r o l . 68, 3943-3954. Zhao, Y., Holden, V. R., Smith, R. H. and O’Callaghan, D. J. (1 9 9 5 ). Regulatory function of the equine herpesvirus 1 ICP27 gene product. J . V i r o l . 69, 2786-2793.

44

Gene Therapy and Molecular Biology Vol 3, page 45 Gene Ther Mol Biol Vol 3, 45-56. August 1999.

Regulation of papillomavirus transcription and replication; insights for the design of extrachromosomal vectors Review Article

Alison A. McBride Laboratory of Viral Diseases, National Institutes of Allergy and Infectious Disease, National Institutes of Health, Bethesda, Maryland 20892-0445, Building 4, room 137, 4 CENTER DR MSC 0445, Bethesda, MD 20892-0445 __________________________________________________________________________________ Correspondence: Alison A. McBride, Ph.D., Tel: 301-496-1370; Fax: 301-480-1497; E-mail: alison_mcbride@nih.gov Received: 16 October 1998; accepted: 20 October 1998

Summary The papillomaviruses infect and replicate in the stratified layers of skin and mucosa and give rise t o b en i g n l e si ons c a l l e d wa r t s o r pa pi l l o m as . Th e v i r us i nf e c t s ba sa l ep it he li al ce ll s an d wi th in these persistently infected c e l l s the viral genome i s maintained at l o w l e v e l s as extrachromosomally replicating viral DNA. The genomes of papillomaviruses can also be stably maintained as high copy number extrachromosomal elements in cell lines and within these cells the viral genomes replicate in synchrony with cellular DNA. The E1 and E2 viral proteins regulate viral transcription, initiation of replication and long term episomal maintenance of viral genomes. This review will describe the functions of the E1 and E2 proteins and discuss how these functions can be exploited in the design of extrachromosomal replicating vectors for gene therapy.

episomal maintenance of viral genomes within replicating cells (Piirsoo et al., 1996). Papillomavirus genomes and the E2-TA protein interact with mitotic chromosomes in dividing cells and this association is likely to be important for genome segregation (Skiadopoulos and McBride, 1998; Lehman and Botchan, 1998).

I. Introduction Certain DNA viruses, such as papillomavirus or Epstein-Barr virus, are able to maintain their genomes as stable extrachromosomal elements in the nuclei of infected cells. The papillomaviruses are small DNA viruses that infect basal epithelial cells and replicate in terminally differentiating keratinocytes. These viruses have been isolated from a wide range of vertebrates and they exhibit both host species and tissue specificity. Viral DNA replication has been studied mostly in bovine papillomavirus type 1 (BPV-1) and the human papillomaviruses (HPV), HPV-1, -11, -16 and -18 and -31. The viral E1 and E2 proteins are important for initiation of viral DNA replication and for regulation of viral transcription. The E1 protein is the primary viral replication initiator protein (Ustav and Stenlund, 1991a; Mohr et al., 1990; Yang et al., 1991) and E1 also functions as a transcriptional repressor (Sandler et al., 1993; Le Moal et al., 1994); the viral E2 protein(s) are transcriptional regulatory proteins that regulate the expression of the other viral gene products and, in addition, play an important role in DNA replication. The E2 transactivator protein is also required for long-term

II. The papillomavirus life cycle and function of the viral proteins Papillomaviruses infect and replicate in stratified epithelium and give rise to benign lesions called warts or papillomas. Papillomaviruses infect the lower basal layer of cells of a stratified epithelium (Figure 1). The !6"4 integrin protein, expressed exclusively in this cell layer, acts as a receptor for the virus (Evander et al., 1997). Damage to the superficial layers of the epithelium is probably necessary to allow access of virus to the basal layer. Within basal cells the viral genome is amplified to a low copy number and maintained as an extrachromosomally replicating circle of double stranded DNA (Figure 2). DNA replication in these cells probably requires the viral E1 and E2 replication proteins. The viral E5 protein is also expressed in basal cells. E5 45

McBride: papillomavirus in design of extrachromosomal vectors

F i g u r e 1 . Diagram of differentiating cells in a stratified epithelium and expression of viral functions in a papilloma. Papillomaviruses infect basal skin cells; within these cells the viral genome is replicated extrachromosomally and early gene products are expressed. Viral DNA amplification and late gene expression only occur in differentiating cells.

Figure 2. Circular genomic map of bovine papillomavirus type 1 (BPV-1). The early ORFs (E1-E8) and late ORFs (L1 and L2 are indicated). The LCR (long control region) contains regulatory elements for transcription and DNA replication such as the origin and minichromosome maintenance element (MME). E2 DNA binding sites are represented by red circles and promoters by arrows.

46

Gene Therapy and Molecular Biology Vol 3, page 47 (Gilbert and Cohen, 1987; Ravnan et al., 1992). Long term, stable maintenance of papillomavirus-derived plasmids requires expression of the E1 and E2 proteins, the replication origin and a region from the LCR, that has been designated a minichromosome maintenance element (MME) (Piirsoo et al., 1996). This element contains multiple high affinity E2 binding sites. Recent studies have shown that both the BPV1 E2 transactivator protein and BPV genomes are associated with cellular chromosomes at mitosis (Skiadopoulos and McBride, 1998; Lehman and Botchan, 1998). This could be the mechanism by which approximately equal numbers of viral genomes are segregated to daughter cells at cell division to ensure that all basal cells of a papilloma contain viral DNA . The third stage of viral replication is vegetative DNA replication, which is required to generate progeny virus. Vegetative DNA replication only occurs as the basal cells of a papilloma migrate upwards and differentiate in the stratum spinosum layer. Increased expression of the E2 proteins also occurs in the stratum spinosum and may be important for amplification of viral DNA (Burnett et al., 1990). The E2 protein is important for initiation of viral DNA replication but it has also been shown that HPV-31 E2 can arrest cells in S phase (Frattini et al., 1997). Clearly this could be important for vegetative replication by allowing sustained synthesis of viral DNA. There appears to be a switch from bidirectional theta replication in the maintenance stage of replication to a rolling circle mode in the vegetative stage (Flores and Lambert, 1997). Little else is known about vegetative viral DNA replication because of the requirement for terminally differentiating keratinocytes and difficulties in reproducing these conditions in a culture system. However, great advances are being made by replicating papillomaviruses in organotypic raft cultures and in xenografts of mice and these systems are proving to be very useful in studying the entire viral life cycle (reviewed in Meyers and Laimins, 1994).

stimulates the activity of growth factor receptors expressed by the cell and induces cellular proliferation (reviewed in Howley, 1995). Enhanced proliferation of basal cells may be important to increase the population of infected cells and to provide a suitable environment for establishment of a productive viral lesion. As basal cells differentiate and migrate up to the stratum spinosum, expression of the E2 proteins is greatly increased and vegetative DNA replication begins (Burnett et al., 1990; Howley, 1995). The cells in this layer do not normally divide nor express cellular proteins necessary for DNA replication. Therefore, the viral E7 protein is required to induce the differentiated keratinocytes to enter S-phase and synthesize cellular replication proteins by binding to and inactivating the cellular retinoblastoma protein, Rb (reviewed in Jones and Munger, 1996). However, the conflicting signals of cell cycle progression and differentiation induce the p53 protein, which in turn signals cells to undergo apoptosis or growth arrest. The viral E6 protein can inactivate this function of p53 by targeting it for degradation by the ubiquitin-proteasome pathway (reviewed in Kubbutat and Vousden, 1996). The viral E4 protein is also abundant in the more differentiated layers of a papilloma. It has been hypothesized that E4 may function as a nuclear structural protein, an RNA splicing and transport factor, or in release of viral particles from the papilloma (reviewed in Howley, 1995). In the upper differentiated layers of the papilloma, the viral capsid proteins L1 and L2 are synthesized and virions are assembled (reviewed in Howley, 1995).

III. Different modes of DNA replication in the papillomavirus life cycle Three modes of DNA replication take place in the papillomavirus life cycle: initial DNA amplification, maintenance replication and vegetative replication. After initial uptake of the virus, the virion particle is uncoated and the genome transported to the nucleus of the basal cell where it is presumed to be amplified to a low copy number (Zhou et al., 1995). Presumably, a low level of the E1 and E2 proteins must be expressed early after infection since there is no evidence that they are in the viral particle. Most experimental studies have examined transient DNA replication in cultured cells, a system that is probably most analogous to this initial amplification stage and which requires the E1 and E2 proteins and the viral replication origin (Ustav and Stenlund, 1991a; Ustav and Stenlund, 1991a). Stable episomal maintenance is the second stage of papillomavirus DNA replication. In a papilloma, the infected basal cells proliferate and maintain low levels of extrachromosomal viral DNA. The genomes of papillomaviruses can also be stably maintained as high copy number extrachromosomal elements in cell lines (Law et al., 1981) and within these lines the viral genomes replicate in synchrony with cellular DNA. The viral genome copy number remains constant overall but the genomes are replicated by a random choice mechanism

IV.Transcriptional regulation by the viral E1 and E2 proteins Papillomavirus transcription is regulated primarily by the viral E2 gene products. These proteins regulate transcription by binding to specific DNA sites located in the viral genomes (see Figure 2). In bovine papillomavirus type 1 several gene products are expressed from the E2 ORF and they have been shown to function as transcriptional activators and repressors (Figure 3). cDNA species that could potentially encode truncated human papillomavirus E2 repressor proteins have been cloned but, as yet, no such proteins have been identified. Some HPVs may have evolved a mechanism to both activate and repress viral transcription with the full-length E2 protein (see Figure 5, reviewed in McBride and Myers, 1997; Fuchs and Pfister, 1994).

47

McBride: papillomavirus in design of extrachromosomal vectors

F i g u r e 3 . A structural and functional map of the BPV-1 E2 proteins.

F i g u r e 4 . A structural and functional map of the BPV-1 E1 proteins.

The full-length E2 protein from all papillomaviruses consists of a 200 amino acid N-terminal transactivation domain linked to a 100 amino acid C-terminal DNA binding and dimerization domain by a flexible hinge region of variable length and sequence (reviewed in McBride and Myers, 1997; McBride et al., 1989; Dostatni et al., 1988). The E2-TA protein activates transcription by binding to specific DNA binding sites that are located within enhancer elements in the viral genome (reviewed in McBride and Myers, 1997). There are seventeen different E2 binding sites in the BPV-1 genome that vary in affinity for the E2 protein over two orders of magnitude (Li et al., 1989) (Figure 2). The well-studied genital-associated HPV genomes contain only four E2 sites in the LCR (Figure 5).

The C-terminal domain of E2 binds specifically to DNA as a dimer. The X-ray crystal structure of the Cterminal 85 amino acids of E2 bound to DNA was the first example of an anti-parallel "-barrel DNA binding structure (Hegde et al., 1992). The DNA binding domain forms an eight-stranded anti-parallel "-barrel made up of four strands from each subunit. A pair of !-helices symmetrically positioned on the outside of the barrel contain the amino acids residues that are required for specific DNA interaction. The DNA binding domain of the Epstein Barr virus EBNA1 protein has a very similar structure to the E2 DNA binding domain despite no sequence similarity (Bochkarev et al., 1995).

48

Gene Therapy and Molecular Biology Vol 3, page 49 function both by direct competition with E2-TA for binding to the E2 DNA binding sites and by the formation of inactive heterodimers with the full-length E2-TA protein (Lim et al., 1998; Barsoum et al., 1992) (Figure 5). In several HPVs associated with the anogenital tract, the full length E2 protein appears to repress the promoter located upstream from the E6 gene (reviewed in McBride and Myers, 1996, 1997; Fuchs and Pfister, 1994). This probably occurs when the E2 dimer binds to E2 DNA binding sites that overlap binding sites for the cellular SP1 and TFIID transcription factors. Recent studies have indicated that these proximal E2 binding sites have lower affinity for the E2 protein than the E2 binding sites are located further upstream from the promoter start site. This has led to a model in which low levels of E2 bind to the higher affinity upstream E2 sites and activate transcription, but at high levels of E2 protein the lower affinity proximal E2 sites are occupied leading to transcriptional repression (Figure 5). The situation is probably even more complex in a papilloma as the levels and activities of the E2 proteins and cellular transcription factors are likely modulated by cell cycle and epithelial differentiation.

The 200 amino acid E2 transactivation domain, unlike many other transactivation domains, appears to have a very constrained structure that is easily disrupted by deletion or certain non-conservative point mutations (reviewed in McBride and Myers, 1997). The transactivation domain is also critical for the replication function of the E2 protein and for interaction with the E1 protein (reviewed in McBride and Myers, 1997). The exact mechanism of transactivation has not been elucidated but probably involves interaction with components of the basic transcriptional machinery. BPV1 E2 has been shown to interact with SP1, TBP, TFIIB and a novel cellular protein, AMF-1 (Li et al., 1991; Steger et al., 1995; Rank and Lambert, 1995; Breiding et al., 1997). In BPV-1, the E2 ORF encodes three different polypeptides; the E2-TA transactivator protein is encoded by the entire ORF and two smaller polypeptides, E2-TR and E8/E2, are encoded by the 3' half of the ORF. The shorter polypetides function as transcriptional repressors by antagonizing the function of E2-TA (Hubbert et al., 1988; Lambert et al., 1987; Spalholz et al., 1985; Lambert et al., 1989; Choe et al., 1989). The repressors contain only the DNA binding/dimerization domain and

F i g u r e 5 . Mechanisms of transcriptional regulation by the papillomavirus E2 proteins. A. BPV1 expresses a transcriptional transactivator with a transactivation domain and DNA binding/dimerization domain. Two shorter repressor proteins contain only the DNA binding/dimerization domain and repress E2 transactivation by forming heterodimers with the transactivator and by competing for the binding to the E2 sites in the viral genome. B . In many HPVs the full-length E2 protein can activate transcription by interacting with higher affinity E2 binding sites upstream from the transcriptional start site. At higher levels of E2, the lower affinity sites close to the promoter become occupied. This displaces essential cellular factors, SP1 and TFIID and results in repression of basal promoter activity.

49

McBride: papillomavirus in design of extrachromosomal vectors binding of an E1/E2 complex to the replication origin, which is located just upstream from P89.

V. Initiation of viral DNA replication by the E1 and E2 proteins In addition to the cell’s replication machinery, papillomavirus DNA replication requires the full-length E2 transactivator protein, the viral E1 protein and the replication origin (Ustav and Stenlund, 1991a; Ustav et al., 1991b; Ustav and Stenlund, 1991a). The minimal origin of replication consists of an E1 binding site, an E2 binding site and an AT rich region that may facilitate origin unwinding. (Ustav et al., 1991b). The E1 protein has several replication-associated activities such as originspecific binding (Wilson and Ludes-Meyers, 1991) and helicase activities (Yang et al., 1993) and forms a complex with the E2 transactivator (Mohr et al., 1990; Blitz and Laimins, 1991; Seo et al., 1993; Spalholz et al., 1993; Sedman and Stenlund, 1995) (Figure 4). The E1 and E2 sites have relatively low affinity for their respective proteins but together they cooperatively bind to the origin with high affinity (Figure 6). After the initial binding of an E1/E2 complex to the origin, the E1 protein oligomerizes to form a trimer or hexamer that encircles the DNA and E2 dissociates from the origin (Sedman and Stenlund, 1996, 1998). The E1 helicase function of the hexamer then unwinds the DNA at the origin to allow DNA synthesis to begin (Sedman and Stenlund, 1998) (Figure 6). The E2-TA transactivator plays an auxiliary role in replication by enhancing and regulating the functions of the E1 protein. In addition to cooperatively binding to the origin with the E1 protein, E2 alleviates repression of replication by nucleosomes (Li and Botchan, 1994) and interacts with cellular replication proteins such as RPA (Li and Botchan, 1993).

VI. Viral genome plasmid maintenance and genome segregation

F i g u r e 6 . Model of initiation of viral DNA replication. The E1 and E2 proteins initiate DNA replication by cooperatively binding to specific sites in the viral origin of replication. It has been proposed that an E1-alone complex then assembles in a ring-like hexamer structure around the DNA and the helicase activity of E1 unwinds the origin to allow access of the cellular replication machinery.

Rodent cells transformed by BPV-1 maintain approximately 50 to 200 copies of the viral genome indefinitely as extrachromosomal nuclear plasmids (Law et al., 1981). Cell lines derived from cervical carcinomas can also maintain human papillomavirus genomes as extrachromosomal elements (Bedell et al., 1991). Plasmids containing the minimal viral replication origin replicate transiently in cells expressing the E1 and E2 proteins but the replicated DNA is lost with time. Long-term stable maintenance of origin-containing plasmids also requires regions from the LCR that contain multiple high affinity E2 DNA binding sites in addition to the replication origin (Piirsoo et al., 1996). This region has been designated the minichromosome maintenance element (MME) and can be substituted by ten tandem copies of E2 DNA binding sites, suggesting that the E2 protein and the E2 DNA binding sites are important for genome segregation (F i g u r e 7 ). This finding is supported by the observation that the E2TA protein and BPV-1 genomes are associated with

The viral E1 replication protein can also function as a transcriptional repressor. Inactivation of E1 increases the immortalizing or growth transforming potential of HPV16 and BPV-1, respectively (Schiller et al., 1989; Lambert and Howley, 1988; Romanczuk and Howley, 1992) and this correlates well with the frequent disruption of E1 and/or E2 expression found in HPV-associated carcinomas. The E1 protein of BPV-1 can repress E2-mediated transactivation of the viral P89 promoter, which expresses the E6 and E7 gene products (Sandler et al., 1993; Le Moal et al., 1994). This is probably a consequence of 50

Gene Therapy and Molecular Biology Vol 3, page 51

F i g u r e 7 . Requirements for long-term episomal maintenance of papillomavirus genomes.

scaffold or chromosomal periphery. The chromosomal periphery is a region around the condensed chromatids that contains many proteins, some of which form a network of fibrils and granules (Hernandez-Verdun and Gautier, 1994). Several components of the nuclear matrix are found in the perichromosomal region as well as a number of “passenger� proteins from the nucleus and nucleoli. The E2-TA protein (but not the E2-TR or E8/E2 proteins) has been shown to be associated with the nuclear matrix (Hubbert et al., 1988) and it will be interesting to determine whether the same interactions are important for the association with mitotic chromosomes. Nuclear matrix attachment sites have also been identified in the BPV-1 genome (Adom and Richard-Foy, 1991; Adom et al., 1992; Tan et al., 1998) and it is possible that these sites are also important for interaction of the genomes with mitotic chromatin instead of, or in addition to, E2 DNA binding sites. Although the overall viral copy number in a population of BPV-1 transformed cells remains relatively constant, several studies have shown that individual cells contain a wide range of copy numbers (Roberts and Weintraub, 1988; Ravnan et al., 1992; Ravnan and Cohen, 1997). Stewart et al. (1994) also demonstrated that there is significant randomization in replication and/or partitioning. This suggests that segregation does not occur by a very precise mechanism and is consistent with the model that the E2 proteins and viral genomes randomly associate with mitotic chromatin as passenger molecules. This model would also predict that the viral copy number depends on the levels of the E2-TA protein. A similar phenomenon has been observed for EpsteinBarr virus (EBV). EBV infects and immortalizes Blymphocytes and the viral genome is maintained indefinitely as an extrachromosomal element. The EBNA-1

condensed mitotic chromosomes in dividing cells (Skiadopoulos and McBride, 1998; Lehman and Botchan, 1998) (F i g u r e 8 ) and supports a model in which viral genomes are attached to mitotic chromatin indirectly via the E2 protein and E2 DNA binding sites (F i g u r e 9 ). This interaction would ensure that approximately equal numbers of viral genomes are segregated to daughter cells. Viral genomes that replicate as extrachromosomal plasmids may also require a mechanism to ensure that they are not lost from the nucleus during cell division. Association with cellular chromosomes would ensure that viral genomes are enclosed in the nuclear membrane during telophase. The genomes may also interact with some cellular component that ensures that they are in a transcriptionally active region of the nucleus as the cells move into the G1 stage of the cell cycle. The BPV-1 E2-TA protein interacts with mitotic chromatin in the absence of viral genomes. Conversely, the E2-TR and E8/E2 proteins are dispersed throughout the cell during mitosis and are excluded from mitotic chromatin. This indicates that the DNA binding domain of the E2 protein is not sufficient for the interaction with mitotic chromosomes and suggests that the interaction is not mediated by binding to cellular DNA sequences. The finding that a DNA-binding defective E2-TA protein retains the ability to interact with mitotic chromatin also supports this. Furthermore, deletions within the Nterminal domain abrogate the ability of E2 to interact with mitotic chromosomes. These findings indicate that the Nterminal transactivation domain of E2-TA is necessary for the interaction (Skiadopoulos and McBride, 1998). As yet, it is not known what component of mitotic chromatin is important for interaction of the E2 protein with mitotic chromatin. One possibility is that E2 interacts with some constituent of the chromosomal 51

Gene Therapy and Molecular Biology Vol 3, page 52

F i g u r e 8 . Papillomavirus genomes and the E2-TA transactivator protein are associated with cellular chromosomes in mitotic cells. E2 proteins were detected in COS7 cells expressing the E2-TA protein by indirect immunofluorescence using an E2-specific antibody. Panels A and B show COS-7 cells as a control. Panels C and D show COS-7 cells expressing E2-TA. In panels A and C, cellular DNA was detected by propidium iodide staining. In panels B and D, FITC-labeled E2 protein is detected in the same field of cells. BPV DNA was detected by fluorescent in situ hybridization in C127 cells (E and F) and 137 cells (that contain BPV-1) (G and H). In panels E and G cellular DNA was detected by the propidium iodide signal. In panels F and H the same field of cells are shown with the FITC-labeled BPV DNA signal. Mitotic cells are indicated by white arrows.

52

Gene Therapy and Molecular Biology Vol 3, page 53 This maintenance requires both E2-TA and the multiple E2 binding sites in the MME element of papillomavirus (Piirsoo et al., 1996) or EBNA-1 and the multiple EBNA1 binding sites in the oriP element of Epstein-Barr virus (Yates et al., 1985). In both cases the viral proteins and genomes are associated with condensed cellular chromosomes during mitosis (Skiadopoulos and McBride, 1998; Lehman and Botchan, 1998; Grogan et al., 1983, Harris et al., 1985). The EBNA-1 protein promotes prolonged nuclear retention of plasmids containing EBNA1 DNA binding sites even in the absence of replication (Krysan et al., 1989; Middleton and Sugden, 1994) and it has been proposed that this is due to the association with mitotic chromosomes.

VIII. Papillomavirus-derived gene therapy vectors Gene therapy vectors that replicate and are retained extrachromosomally have several advantages over those that integrate in a random fashion into the host genome. These vectors will persist in proliferating cells and should not generate mutations by insertion into the cellular chromosomes. Such vectors can be maintained at a high copy number and are not susceptible to positional effects, such as inactivation, that are dependent on the integration site. EBV-based vectors that contain the EBNA-1 gene and the oriP replication origin have been developed and can express foreign gene products in primate and human cells (reviewed in Calos, 1996). Another class of EBV vectors have been developed that only contain the EBNA-1 gene and repeats of the EBNA-1 binding site required for nuclear retention. In these vectors the oriP origin has been replaced with a cellular replication origin and the resulting vectors are able to replicate in a wider range of mammalian cells (Krysan et al., 1989). The EBNA-1 protein also has the advantage that it is not recognized by the cell-mediated immune system (Levitskaya et al., 1995) as it is resistant to the proteasome-mediated degradation that is required for antigen presentation (Levitskaya et al., 1997). However, there is a report that the EBNA-1 protein can cause lymphomas in transgenic mice expressing this protein in B-cells (Wilson et al., 1996). Sarver et al. (1981) first described the use of papillomaviruses as vectors in 1981. In general, these vectors comprised the 69% transforming region of the virus (the genome minus the late region) and the foreign gene to be expressed. A newer vector only contains the LCR and the E1 and E2 genes (Ohe et al., 1995). However, these vectors have a limited host range and, in some cases, insertion of a foreign transcriptionally active foreign gene causes the plasmid to integrate (Waldenstrom et al., 1992). This is probably because the small papillomavirus genomes are very compact and contain multiple overlapping genes and regulatory signals that can be inadvertently disrupted. The presence of an active heterologous enhancer and promoter could interfere with viral replication by transcriptional interference. Using the detailed knowledge of the mechanisms of papillomavirus

Figure 9 . This diagram shows a model in which papillomavirus genomes are linked via the E2-TA protein to condensed mitotic chromosomes.

protein of EBV is a transcriptional transactivator and a replication protein and it is the only viral protein required for replication and maintenance of plasmids containing the oriP origin of replication (which contains a number of repeated EBNA DNA binding sites) (Yates et al., 1985). The EBNA 1 protein and EBV genomes have also been shown to be randomly associated with mitotic chromatin (Grogan et al., 1983; Harris et al., 1985) and it has been suggested that these properties might be important for the genome segregation and nuclear retention function of EBNA-1. The EBNA-1 protein also promotes prolonged nuclear retention of plasmids containing EBNA-1 DNA binding sites but no origin of replication (Krysan et al., 1989; Middleton and Sugden, 1994) and it has been proposed that this is due to the interaction of plasmids with mitotic chromosomes.

VII. Similarities between the papillomavirus E2 and Epstein-Barr virus EBNA-1 protein The Epstein-Barr virus EBNA-1 protein and the papillomavirus E2-TA protein have common roles in the life cycles of their respective viruses (Grossman and Laimins, 1996). Both proteins are transcriptional transactivators that activate transcription by binding to specific binding sites within the viral genomes. Notably, both proteins have dimeric DNA binding. domains with almost identical anti-parallel "-barrel structures, despite no amino acid homology (Bochkarev et al., 1995). Both viruses replicate and maintain their genomes as extrachromosomal elements in persistently infected cells. 53

McBride: papillomavirus in design of extrachromosomal vectors replication and genome maintenance, it should be possible to generate a new class of papillomavirus vectors. These vectors could express the E1 and/or E2 genes from different promoters suitable for a specific cell type and either viral or cellular replication origins could be incorporated, as has been described for EBV-based vectors (Krysan et al., 1989). The addition of repeated E2 binding sites may be sufficient to maintain the vector as an episome when either a viral or cellular replication origin is used.

Dostatni, N., Thierry, F., and Yaniv, M. (1 9 8 8 ). A dimer of BPV-1 E2 containing a protease resistant core interacts with its DNA target. EMBO J. 7, 3807-3816. Evander, M., Frazer, I.H., Payne, E., Qi, Y.M., Hengst, K., and McMillan, N.A. (1 9 9 7 ). Identification of the a6 integrin as a candidate receptor for papillomaviruses. J . V i r o l . 71, 2449-2456. Flores, E.R. and Lambert, P.F. (1 9 9 7 ). Evidence for a switch in the mode of human papillomavirus type 16 DNA replication during the viral life cycle. J . V i r o l . 71, 7167-7179. Frattini, M.G., Hurst, S.D., Lim, H.B., Swaminathan, S., and Laimins, L.A. (1 9 9 7 ). Abrogation of a mitotic checkpoint by E2 proteins from oncogenic human papillomaviruses correlates with increased turnover of the p53 tumor suppressor protein. EMBO J. 16, 318-331.

References Adom, J.N., Gouilleux, F., and Richard-Foy, H. (1 9 9 2 ). Interaction with the nuclear matrix of a chimeric construct containing a replication origin and a transcription unit. B i o c h i m . B i o p h y s . A c t a 1171, 187-197.

Fuchs, P.G. and Pfister, H. (1 9 9 4 ). Transcription of papillomavirus genomes. I n t e r v i r o l o g y 37, 159-167.

Adom, J.N. and Richard-Foy, H. (1 9 9 1 ). A region immediately adjacent to the origin of replication of bovine papilloma virus type 1 interacts in vitro with the nuclear matrix. Biochem. Biophys. Res. Commun. 176, 479-485.

Gilbert, D.M. and Cohen, S.N. (1 9 8 7 ). Bovine papilloma virus plasmids replicate randomly in mouse fibroblasts throughout S phase of the cell cycle. C e l l 50, 59-68. Grogan, E.A., Summers, W.P., Dowling, S., Shedd, D., Gradoville, L., and Miller, G. (1 9 8 3 ). Two Epstein-Barr viral nuclear neoantigens distinguished by gene transfer, serology, and chromosome binding. P r o c . N a t l . A c a d . S c i . U S A 80, 7650-7653.

Barsoum, J., Prakash, S.S., Han, P., and Androphy, E.J. (1 9 9 2 ). Mechanism of action of the papillomavirus E2 repressor: repression in the absence of DNA binding. J . V i r o l . 66, 3941-3945. Bedell, M.A., Hudson, J.B., Golub, T.R., Turyk, M.E., Hosken, M., Wilbanks, G.D., and Laimins, L.A. (1 9 9 1 ). Amplification of human papillomavirus genomes in vitro is dependent on epithelial differentiation. J . V i r o l . 65, 2254-2260.

Grossman, S.R. and Laimins, L.A. (1 9 9 6 ). EBNA1 and E2: a new paradigm for origin-binding proteins? Trends. M i c r o b i o l . 4, 87-89. Harris, A., Young, B.D., and Griffin, B.E. (1 9 8 5 ). Random association of Epstein-Barr virus genomes with host cell metaphase chromosomes in Burkitt's lymphoma-derived cell lines. J . V i r o l . 56, 328-332.

Blitz, I.L. and Laimins, L.A. (1 9 9 1 ). The 68-kilodalton E1 protein of bovine papillomavirus is a DNA binding phosphoprotein which associates with the E2 transcriptional activator in vitro. J . V i r o l . 65, 649656.

Hegde, R.S., Grossman, S.R., Laimins, L.A., and Sigler, P.B. (1 9 9 2 ). Crystal structure at 1.7 A of the bovine papillomavirus-1 E2 DNA- binding domain bound to its DNA target. Nature 359, 505-512.

Bochkarev, A., Barwell, J.A., Pfuetzner, R.A., Furey, W.J., Edwards, A.M., and Frappier, L. (1 9 9 5 ). Crystal structure of the DNA-binding domain of the Epstein-Barr virus origin-binding protein EBNA 1. C e l l 83, 39-46.

Hernandez-Verdun, D. and Gautier, T. (1 9 9 4 ). The chromosome periphery during mitosis. B i o e s s a y s 16, 179-185.

Breiding, D., Sverdrup, F., Grossel, M.J., Moscufo, N., Boonchai, W., and Androphy, E.J. (1 9 9 7 ). Isolation of a BPV1 E2 transactivation domain binding factor required for both transcriptional activation and DNA replication. V i r o l o g y 221, 34-43.

Howley, P.M.(1 9 9 5 ). Papillomavirinae: The viruses and their replication. In V i r o l o g y (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) Lippincott-Raven, Philadelphia and New York. 2045-2076. Hubbert, N.L., Schiller, J.T., Lowy, D.R., and Androphy, E.J. (1 9 8 8 ). Bovine papilloma virus-transformed cells contain multiple E2 proteins. P r o c . N a t l . A c a d . S c i . USA 85, 5864-5868.

Burnett, S., Strom, A.C., Jareborg, N., Alderborn, A., Dillner, J., Moreno-Lopez, J, Pettersson, U., and Kiessling, U. (1 9 9 0 ). Induction of bovine papillomavirus E2 gene expression and early region transcription by cell growth arrest: correlation with viral DNA amplification and evidence for differential promoter induction. J . V i r o l . 64, 5529-5541.

Jones, D.L. and Munger, K. (1 9 9 6 ). Interactions of the human papillomavirus E7 protein with cell cycle regulators. S e m i n . C a n c e r B i o l . 7, 327-337.

Calos, M.P. (1 9 9 6 ). The potential of extrachromosomal replicating vectors for gene therapy. Trends. G e n e t . 12, 463-466.

Krysan, P.J., Haase, S.B., and Calos, M.P. (1 9 8 9 ). Isolation of human sequences that replicate autonomously in human cells. M o l . C e l l B i o l . 9, 1026-1033.

Choe, J., Vaillancourt, P., Stenlund, A., and Botchan, M. (1 9 8 9 ). Bovine papillomavirus type 1 encodes two forms of a transcriptional repressor: Structural and functional analysis of new viral cDNAs. J . V i r o l . 63, 1743-1755.

Kubbutat, M.H.G. and Vousden, K.H. (1 9 9 6 ). Role of E6 and E7 oncoproteins in HPV-induced anogenital malignancies. S e m i n . V i r o l . 7, 295-304. Lambert, P.F. and Howley, P.M. (1 9 8 8 ). Bovine papillomavirus type 1 E1 replication-defective mutants

54

Gene Therapy and Molecular Biology Vol 3, page 55 are altered in their transcriptional regulation. J . V i r o l . 62, 4009-4015.

amino-terminal domain. P r o c . N a t l . A c a d . S c i . U S A 86, 510-514.

Lambert, P.F., Hubbert, N.L., Howley, P.M., and Schiller, J.T. (1 9 8 9 ). Genetic assignment of multiple E2 gene products in bovine papillomavirus-transformed cells. J . V i r o l . 63, 3151-3154.

McBride, A.A. and Myers, G.(1 9 9 6 ). The E2 proteins .in Human P a p i l l o m a v i r u s e s (Myers, G. , Baker, C., Wheeler, C., Halpern, A., McBride, A., and Doorbar, J., Eds.) .Los Alamos National Laboratory, Los Alamos.

Lambert, P.F., Spalholz, B.A., and Howley, P.M. (1 9 8 7 ). A transcriptional repressor encoded by BPV-1 shares a common carboxy-terminal domain with the E2 transactivator. C e l l 50, 69-78.

McBride, A.A. and Myers, G.(1 9 9 7 ). The E2 proteins: an update .in H u m a n P a p i l l o m a v i r u s e s 1 9 9 7 (Myers, G., Baker, C., Munger, K., Sverdrup, F., McBride, A. , and Bernard, H.-U., Eds.) .Los Alamos National Laboratory, Los Alamos. http://hpvweb.lanl.gov/

Law, M.F., Lowy, D.R., Dvoretzky, I., and Howley, P.M. (1 9 8 1 ). Mouse cells transformed by bovine papillomavirus contain only extrachromosomal viral DNA sequences. P r o c . N a t l . Acad. S c i . USA 78, 2727-2731.

Meyers, C. and Laimins, L.A. (1 9 9 4 ). In vitro systems for the study and propagation of human papillomaviruses. C u r r . T o p . M i c r o b i o l . I m m u n o l . 186, 199-215. Middleton, T. and Sugden, B. (1 9 9 4 ). Retention of plasmid DNA in mammalian cells is enhanced by binding of the Epstein-Barr virus replication protein EBNA1. J . V i r o l . 68, 4067-4071.

Le Moal, M.A., Yaniv, M., and Thierry, F. (1 9 9 4 ). The bovine papillomavirus type 1 (BPV1) replication protein E1 modulates transcriptional activation by interacting with BPV1 E2. J . V i r o l . 68, 1085-1093.

Mohr, I.J., Clark, R., Sun, S., Androphy, E.J., MacPherson, P., and Botchan, M.R. (1 9 9 0 ). Targeting the E1 replication protein to the papillomavirus origin of replication by complex formation with the E2 transactivator. S c i e n c e 250, 1694-1699.

Lehman, C.W. and Botchan, M.R. (1 9 9 8 ). Segregation of viral plasmids depends on tethering to chromosomes and is regulated by phosphorylation. P r o c . N a t l . Acad. S c i . U S A 95, 4338-4343. Levitskaya, J., Coram, M., Levitsky, V., Imreh, S., Steigerwald-Mullen, P.M., Klein, G., Kurilla, M.G., and Masucci, M.G. (1 9 9 5 ). Inhibition of antigen processing by the internal repeat region of the Epstein-Barr virus nuclear antigen-1. Nature 375, 685-688.

Ohe, Y., Zhao, D., Saijo, N., and Podack, E.R. (1 9 9 5 ). Construction of a novel bovine papillomavirus vector without detectable transforming activity suitable for gene transfer. Hum. Gene Ther. 6, 325-333. Piirsoo, M., Ustav, E., Mandel, T., Stenlund, A., and Ustav, M. (1 9 9 6 ). Cis and trans requirements for stable episomal maintenance of the BPV-1 replicator. EMBO J. 15, 1-11.

Levitskaya, J., Sharipo, A., Leonchiks, A., Ciechanover, A., and Masucci, M.G. (1 9 9 7 ). Inhibition of ubiquitin/proteasome-dependent protein degradation by the Gly-Ala repeat domain of the Epstein-Barr virus nuclear antigen 1. P r o c . N a t l . A c a d . S c i . U S A 94, 12616-12621.

Rank, N.M. and Lambert, P.F. (1 9 9 5 ). Bovine papillomavirus type 1 E2 transcriptional regulators directly bind two cellular transcription factors, TFIID and TFIIB. J . V i r o l . 69, 6323-6334.

Li, R. and Botchan, M.R. (1 9 9 3 ). The acidic transcriptional activation domains of VP16 and p53 bind the cellular replication protein A and stimulate in vitro BPV-1 DNA replication. C e l l 73, 1207-1221.

Ravnan, J.-B. and Cohen, S.N. (1 9 9 7 ). Transformed mouse cell lines that consist predominantly of cells maintaining bovine papillomavirus at high copy number. V i r o l o g y 213, 526-534.

Li, R. and Botchan, M.R. (1 9 9 4 ). Acidic transcription factors alleviate nucleosome-mediated repression of DNA replication of bovine papillomavirus type 1. P r o c . N a t l . A c a d . S c i . U S A 91, 7051-7055.

Ravnan, J.-B., Gilbert, G.M., Ten Hagen, K.G., and Cohen, S.N. (1 9 9 2 ). Random-choice replication of extrachromosomal bovine papillomavirus (BPV) molecules in heterogeneous clonally-derived BPV-infected cell lines. J . V i r o l . 66, 6946-6952.

Li, R., Knight, J., Bream, G., Stenlund, A., and Botchan, M. (1 9 8 9 ). Specific recognition nucleotides and their DNA context determine the affinity of E2 protein for 17 binding sites in the BPV- 1 genome. Genes D e v . 3, 510-526.

Roberts, J.M. and Weintraub, H. (1 9 8 8 ). Cis-acting negative control of DNA replication in eukaryotic cells. C e l l 52, 397-404.

Li, R., Knight, J.D., Jackson, S.P., Tjian, R., and Botchan, M.R. (1 9 9 1 ). Direct interaction between Sp1 and the BPV enhancer E2 protein mediates synergistic activation of transcription. C e l l 65, 493-505.

Romanczuk, H. and Howley, P.M. (1 9 9 2 ). Disruption of either the E1 or the E2 regulatory gene of human papillomavirus type 16 increases viral immortalization capacity. P r o c . N a t l . Acad. S c i . USA 89, 31593163.

Lim, D.A., Gossen, M., Lehman, C.W., and Botchan, M.R. (1 9 9 8 ). Competition for DNA binding sites between the short and long forms of E2 dimers underlies repression in bovine papillomavirus type 1 DNA replication control. J . V i r o l . 72, 1931-1940.

Sandler, A.B., Vande Pol, S.B., and Spalholz, B.A. (1 9 9 3 ). Repression of bovine papillomavirus type 1 transcription by the E1 replication protein. J . V i r o l . 67, 5079-5087. Sarver, N., Gruss, P., Law, M.F., Khoury, G., and Howley, P.M. (1 9 8 1 ). Bovine papilloma virus deoxyribonucleic acid: a novel eucaryotic cloning vector. M o l . C e l l . B i o l . 1, 486-496.

McBride, A.A., Byrne, J.C., and Howley, P.M. (1 9 8 9 ). E2 polypeptides encoded by bovine papillomavirus type 1 form dimers through the common carboxyl-terminal domain: Transactivation is mediated by the conserved

55

McBride: papillomavirus in design of extrachromosomal vectors Schiller, J.T., Kleiner, E., Androphy, E.J., Lowy, D.R., and Pfister, H. (1 9 8 9 ). Identification of bovine papillomavirus E1 mutants with increased transforming and transcriptional activity. J . V i r o l . 63, 1775-1782.

Wilson, J.B., Bell, J.L., and Levine, A.J. (1 9 9 6 ). Expression of Epstein-Barr virus nuclear antigen-1 induces B cell neoplasia in transgenic mice. E M B O J . 15, 3117-3126.

Sedman, J. and Stenlund, A. (1 9 9 5 ). Co-operative interaction between the initiator E1 and the transcriptional activator E2 is required for replicator specific DNA replication of bovine papillomavirus in vivo and in vitro. E M B O J . 14, 6218-6228.

Wilson, V.G. and Ludes-Meyers, J. (1 9 9 1 ). A bovine papillomavirus E1-related protein binds specifically to bovine papillomavirus DNA. J . V i r o l . 65, 5314-5322. Yang, L., Li, R., Mohr, I.J., Clark, R., and Botchan, M.R. (1 9 9 1 ). Activation of BPV-1 replication in vitro by the transcription factor E2. Nature 353, 628-632.

Sedman, J. and Stenlund, A. (1 9 9 6 ). The initiator protein E1 binds to the bovine papillomavirus origin of replication as a trimeric ring-like structure. EMBO J . 15, 50855092.

Yang, L., Mohr, I., Fouts, E., Lim, D.A., Nohaile, M., and Botchan, M. (1 9 9 3 ). The E1 protein of bovine papillomavirus 1 is an ATP-dependent DNA helicase. P r o c . N a t l . A c a d . S c i . U S A 90, 5086-5090.

Sedman, J. and Stenlund, A. (1 9 9 8 ). The papillomavirus E1 protein forms a DNA-dependent hexameric complex with ATPase and DNA helicase activities. J . V i r o l . 72, 68936897.

Yates, J.L., Warren, N., and Sugden, B. (1 9 8 5 ). Stable replication of plasmids derived from Epstein-Barr virus in various mammalian cells. Nature 313, 812-815.

Seo, Y.-S., Muller, F., Lusky, M., Gibbs, E., Kim, H.-Y., Phillips, B., and Hurwitz, J. (1 9 9 3 ). Bovine papilloma virus (BPV)-encoded E2 protein enhances binding of E1 protein to the BPV replication origin. P r o c . N a t l . Acad. Sci. USA 90, 2865-2869.

Zhou, J., Gissmann, L., Zentgraf, H., Muller, H., Picken, M., and Muller, M. (1 9 9 5 ). Early phase in the infection of cultured cells with papillomavirus virions. V i r o l o g y 214, 167-176.

Skiadopoulos, M.H. and McBride, A.A. (1 9 9 8 ). BPV1 viral genomes and the E2 transactivator protein are associated with cellular metaphase chromosomes. J V i r o l 72, 2079-2088. Spalholz, B.A., McBride, A.A., Sarafi, T., and Quintero, J. (1 9 9 3 ). Binding of bovine papillomavirus E1 to the origin is not sufficient for DNA replication. V i r o l o g y 193, 201-212. Spalholz, B.A., Yang, Y.C., and Howley, P.M. (1 9 8 5 ). Transactivation of a bovine papilloma virus transcriptional regulatory element by the E2 gene product. C e l l 42, 183-191. Steger, G., Ham, J., Lefebvre, O., and Yaniv, M. (1 9 9 5 ). bovine papillomavirus 1 E2 protein contains activation domains: one that interacts with TBP another that functions after TBP binding. EMBO J. 329-340.

The two and 14,

Stewart, A.-C., Jareborg, N., Simonsson, M., Alderborn, A., and Burnett, S. (1 9 9 4 ). Segregation properties of bovine papillomaviral plasmid DNA. J . M o l . B i o l . 236, 480490. Tan, S.H., Bartsch, D., Schwarz, E., and Bernard, H.U. (1 9 9 8 ). Nuclear matrix attachment regions of human papillomavirus type 16 point toward conservation of these genomic elements in all genital papillomaviruses. J . V i r o l . 72, 3610-3622. Ustav, M. and Stenlund, A. (1 9 9 1 a ). Transient replication of BPV-1 requires two viral polypeptides encoded by the E1 and E2 open reading frames. EMBO J. 10, 449-457. Ustav, M., Ustav, E., Szymanski, P., and Stenlund, A. (1 9 9 1 b ). Identification of the origin of replication of bovine papillomavirus and characterization of the viral origin recognition factor E1. EMBO J. 10, 4321-4329. Waldenstrom, M., Schenstrom, K., Sollerbrant, K., and Hansson, L. (1 9 9 2 ). Replication of bovine papillomavirus vectors in murine cells. Gene 120, 175181.

56

Gene Therapy and Molecular Biology Vol 3, page 57 Gene Ther Mol Biol Vol 3, 57-65. August 1999.

Gene transfer with adeno-associated virus 2 vectors: the growth factor receptor connection Review Article

Cathryn Mah 1-3, Keyun Qing1-3, Jonathan Hansen1-3, Benjawan Khuntirat1-3, Mervin C. Yoder4, and Arun Srivastava1-3,5 1

Department of Microbiology & Immunology, 2Walther Oncology Center, 5Division of Hematology/Oncology, Department of Medicine, 4 Herman B Wells Center for Pediatric Research and Department of Biochemistry & Molecular Biology, Indiana University School of Medicine, and 3Walther Cancer Institute, Indianapolis, IN 46202 __________________________________________________________________________________________________ Correspondence: Arun Srivastava, Ph.D., Department of Microbiology & Immunology, Indiana University School of Medicine, 635 Barnhill Drive, Medical Science Building, Room 257, Indianapolis, IN 46202-5120. Phone: (317) 274-2194; Fax: (317) 274-4090; E-mail: asrivast@iupui.edu Received: 30 September 1998; accepted: 10 October 1998

Summary Adeno-associated virus 2 (AAV)-based vectors have gained attention as a potentially useful alternative to the more commonly used retroviral and adenoviral vectors for human gene therapy. However, there are at least two major obstacles that limit high-efficiency transduction by AAV vectors. The first relates to the extent of expression of the cellular receptor for AAV, and the second concerns the rate-limiting step of the viral second-strand DNA synthesis. With reference to the first obstacle, although the ubiquitously expressed cell surface heparan sulfate proteoglycan (HSPG) has been reported to be a receptor AAV, HSPG alone is insufficient for AAV infection, and human fibroblast growth factor receptor 1 (FGFR1) has been identified as a co-receptor for successful viral entry into the host cell. With reference to the second obstacle, a cellular protein, designated the single-stranded D-sequence binding protein (ssD-BP), phosphorylated at tyrosine residues, has been identified which plays a crucial role in viral second-strand DNA synthesis. The ssD-BP is phosphorylated by the protein tyrosine kinase activity of the human epidermal growth factor receptor (EGFR). Thus, both FGFR1 and EGFR are crucial determinants in the life cycle of AAV, and further studies on the interaction between the FGFR and EGFR may yield new insights not only into its role in the host cell but also in the optimal use of AAV vectors in human gene therapy. therapy of cystic fibrosis (Flotte and Carter, 1997). Although AAV possesses a broad host-range that transcends the species barrier (Muzyczka, 1992), the efficiency of AAV-mediated transduction has been reported to vary widely. Recently, the ubiquitously expressed cell surface heparan sulfate proteoglycan (HSPG) was identified as a receptor for AAV (Summerford and Samulski, 1998), it has become increasingly clear that HSPG alone is insufficient for AAV infection. For example, our recent studies have documented a significant donor variation in terms of the ability of AAV vectors to transduce primary human bone marrow-derived CD34+ hematopoietic progenitor cells (Ponnazhagan et al., 1997). In these

I. Introduction The non-pathogenic nature of the adeno-associated virus 2 (AAV), a single-stranded DNA-containing human parvovirus (Srivastava et al., 1983), coupled with the remarkable site-specific integration of the wild-type (wt) AAV genome into the human chromosome 19 (Kotin et al., 1990, Samulski et al., 1991), generated a significant interest in the development of AAV vectors as a potentially useful alternative to the more commonly used retrovirus and adenovirus vectors in human gene therapy (Berns and Giraud, 1996). Indeed, AAV vectors have been successfully used for gene transfer in vitro as well as in vivo, and are currently in Phase II clinical trials for gene 57

Mah et al: AAV vectors and the EGFR connection studies, AAV-mediated transgene expression ranged between 15-80% of infected cells from approximately 50% of normal volunteer donors, whereas AAV failed to bind to CD34+ cells from approximately 50% of donors and consequently, cells from these donors could not be transduced. Similarly, we have reported that the efficiency of AAV transduction in permissive cells does not correlate with the receptor number, and that a cellular protein, designated as the single-stranded D-sequence-binding protein (ssD-BP), phosphorylated at tyrosine residues, plays a crucial role in the viral second-strand DNA synthesis (Qing et al., 1997; 1998), a rate-limiting step in AAV-mediated transgene expression (Fisher et al., 1996; Ferrari et al., 1996). Thus, the two obstacles encountered in attempting to obtain high-efficiency transduction by recombinant AAV vectors will be discussed briefly as follows.

(Bartlett and Samulski, 1998), could not bind FGF. Stable transfection with huFGFR1 cDNA alone allowed for a low-level of FGF binding, the extent of which was significantly higher when M07e were co-transfected with both HSPG and FGFR1 cDNAs. Interestingly, transfection with the muHSPG cDNA alone resulted in significant binding of FGF. These results suggest that M07e cells do indeed express the endogenous FGFR gene. As expected, mock-transfected Raji cells also failed to bind FGF as they lack both HSPG and FGFR. Only low levels of FGF binding were detected in Raji clones stably transfected with either the HSPG or FGFR1 expression plasmids alone, whereas in Raji cells co-expressing both, a significant binding of FGF occurred, further corroborating the requirement of both HSPG and FGFR1 for ligand binding. It was also interesting to note that the binding patterns of AAV to both the M07e and Raji cells coexpressing HSPG+FGFR1 genes closely resembled that of FGF binding. Taken together, these results strongly suggested that cell surface expression of both HSPG and FGFR1 is required for successful binding of AAV to the host cell (Qing et al., 1999).

II. The first obstacle A. Successful infection of cells by AAV requires fibroblast growth factor receptor 1 (FGFR1) as a cell surface co-receptor

In order to determine whether non-permissive cells could be rendered positive for AAV transduction following stable transfection with cDNAs encoding muHSPG, or huFGFR1, or both, individual clonal isolates from both cell types were either mock-infected or infected with a recombinant AAV vector under identical conditions and analyzed for transgene expression by fluorescence-activated cell-sorting (FACS). Whereas little transgene expression was seen in mock-infected M07e cells, as expected, it was evident that Mo7e cells expressing either HSPG alone, or both HSPG and FGFR1, but not FGFR1 alone, could be readily transduced by the recombinant AAV vector. Expression of the exogenous HSPG in M07e cells was sufficient to render the cells permissive to AAV infection because M07e cells express the endogenous FGFR gene. On the other hand, Raji cells failed to be transduced by recombinant AAV if only the exogenous HSPG or FGFR1 genes were expressed, but co-expression of both HSPG and FGFR1 conferred AAV infectivity to these cells, albeit at a relatively low-efficiency. Inclusion of additional individual clonal isolates from both cell types yielded very similar results. These studies establish that co-expression of both HSPG and FGFR1 is required both for binding and also entry of AAV into the host cell (Qing et al., 1999).

It has previously been demonstrated that all cell types which bind AAV can also be infected by AAV (Ponnazhagan et al., 1996, 1997; Qing et al., 1998; Summerford and Samulski. 1998; Bartlett and Samulski, 1998). For example, human cell lines such as HeLa, KB, and 293, which have been shown to be permissive for AAV infection, can bind AAV, whereas non-permissive cells, such as M07e, cannot. Interestingly, however, we noted that murine NIH3T3 cells, which could not be transduced by a recombinant AAV vector, could bind AAV quite efficiently. As NIH3T3 cells are known to express HSPG (Ledoux et al., 1992), this observation suggested that in addition to HSPG as a primary receptor for binding, AAV might require a putative cell surface co-receptor for efficient entry. Since fibroblast growth factor (FGF) has an absolute requirement for HSPG prior to efficient binding to the fibroblast growth factor receptor (FGFR) (Green et al., 1996), we reasoned that FGFR might be a potential candidate. We examined human cell types known to be non-permissive for AAV infection, such as the human megakaryocytic cell line M07e (Ponnazhagan et al., 1996), as well as those that do not express either HSPG or FGFR, such as the human lymphoblastoid cell line Raji (Kiefer et al., 1990; Lebakken and Rapraeger, 1996). These cell types were stably transfected with cDNA expression plasmids containing either the murine HSPG core protein (Syndecan-1) (Saunders et al., 1989), or the human FGFR1 (Johnson et al., 1990), or both, followed by the determination of radiolabeled FGF-binding, radiolabeledAAV binding, and recombinant AAV-mediated transgene expression. M07e cells, known to lack HSPG expression

B. FGFR autophosphorylation is not required for AAV-mediated transduction Since ligand binding to FGFR consequently leads to receptor dimerization followed by ion activation of the FGFR-associated protein tyrosine kinase (PTK), 58

Gene Therapy and Molecular Biology Vol 3, page 59 ultimately resulting in recruitment of intracellular signaling molecules (Rapraeger et al., 1991; Ledoux et al., 1992; Roghani and Moscatelli; 1992, Givol and Yayon, 1992; Kan et al., 1993), it was of interest to investigate whether FGFR PTK activity affected AAV-mediated transgene expression. To this end, cells permissive for AAV infection, human 293 and HeLa cells, were either mock-treated, or first treated with specific inhibitors of FGFR PTK (Mohammadi et al., 1997) followed by infection with a recombinant AAV vector under identical conditions and the extent of transgene expression was determined as described above. These experiments demonstrated that none of the FGFR PTK inhibitors tested had any significant effect on AAV-mediated transgene expression. From these studies, we conclude that FGFR PTK activity is not required for AAV-mediated transgene expression (Qing et al., 1999).

expression in the presence of FGF was not due to phosphorylation of the ssD-BP in 293 cells as these assays carried out with prior treatment with genistein also resulted in similar results (89% inhibition with FGF, 0% inhibition with EGF). Taken together, these results strongly suggest that HSPG-FGFR1 interaction is crucial not only for binding, but also for entry of AAV into the host cell. Based on all available information, we propose a model for the initial step in AAV infection, which is depicted in Figure 1. In this model, co-expression of cell surface HSPG and FGFR1 is required for successful AAV binding followed by viral entry (Panel A), both of which are blocked by FGF (Panel B) (Qing et al., 1999).

III. The second obstacle A. Inhibitors of epidermal growth factor receptor (EGFR) protein tyrosine kinase (PTK) activity increase the transduction efficiency of recombinant AAV

C. FGF treatment perturbs AAV binding to non-permissive as well as permissive cells, and abrogates viral entry into permissive cells

We have previously shown that inhibition of tyrosine phosphorylation of the ssD-BP by genistein, a specific inhibitor of all protein tyrosine kinases (Akiyama et al., 1987; Barnes and Peterson, 1995; Constantinou and Huberman, 1995; Carlo-Stella et al., 1996), increased transduction efficiency by recombinant AAV (Qing et al., 1997). To investigate which kinase may be responsible for tyrosine phosphorylation of the ssD-BP, we studied the effects of various kinase inhibitors such as apigenein (MAP kinase) (Kuo and Yang, 1995), herbimycin A (pp60c-src) (Fukazawa et al., 1991), LY294002 (PI 3kinase) (Vlahos et al., 1994), staurosporine (CaM kinase, MLC kinase, PK-A, PK-C, PK-G) (Couldwell et al., 1994), tyrphostin A48 (EGF-R PTK) (Gazit et al., 1989), wortmannin (MAP kinase, MLC kinase, PI 3-kinase, PI 4-kinase) (Okada et al., 1994), in addition to genistein, on the transduction efficiency of recombinant AAV. Following treatment with these reagents, cells were infected with a recombinant AAV vector, followed by staining with X-gal 48 hrs post-infection. The results indicated that in addition to genistein, treatment with tyrphostin A48, a specific inhibitor for EGFR PTK, caused an increase in the numbers of blue cells. These results suggest that EGF-R PTK may be involved in recombinant AAV-mediated transgene expression (Mah et al., 1998).

The following experiments further supported the contention that FGFR1 acts as a co-receptor for AAV biding and entry. First, we hypothesized that treatment of non-permissive cells such as NIH3T3 cells, and permissive cells such as 293 cells, with large excess of FGF would perturb the ability of AAV to bind to the host cell. Binding studies with NIH3T3 and 293 cells were carried out using radiolabeled AAV in the presence or absence of excess amounts of FGF, with additional controls including wt AAV or heparin (as positive controls) and EGF (as a negative control). The results of these experiments documented that AAV binding to NIH3T3 cells was inhibited by heparin, as expected (Summerford and Samulski, 1998), and FGF also inhibited AAV binding to a significant extent, whereas EGF had no effect under identical conditions. As expected, unlabeled wt AAV significantly inhibited binding of radiolabeled AAV to 293 cells. Likewise, excess FGF was also able to reduce AAV binding to 293 cells. On the other hand, as with the NIH3T3 cells, similar concentrations of EGF had no significant effect on AAV binding to 293 cells. Second, we reasoned that excess amounts of FGF might perturb AAV infection. To this end, equivalent numbers of 293 cells were infected with a recombinant AAV vector either in the absence or presence of excess FGF or EGF, under identical conditions. Forty-eight hrs post-infection, transgene expression was evaluated by X-gal staining as previously described (Ponnazhagan et al., 1996; 1997). The results of these experiments indicated that AAVmediated transduction of 293 cells was inhibited in the presence of FGF by approximately 89%, but in the presence of EGF by only 2%. The lack of transgene

In order to further investigate the role of EGFR PTK in recombinant AAV transduction efficiency, other inhibitors specific for EGFR PTK, tyrphostins 1, 23, 25, 46, 47, 51, 63, and AG1478 (Yaish et al., 1988; Gazit et al., 1989; Lyall et al., 1989; Levitzki, 1990, Levitzki et al., 1991) were tested for their effects on recombinant AAV 59

Mah et al: AAV vectors and the EGFR connection

Figure 1. A possible model for the role of cell surface HSPG and FGFR1 in mediating AAV binding and entry into the host cell. Coexpression of HSPG and FGFR1 is required for successful binding of AAV followed by viral entry into a susceptible cell (Panel A), both of which are perturbed by the ligand, FGF, which also requires HSPG-FGFR1 interaction (Panel B ) (Qing et al., 1999).

increase recombinant AAV transduction efficiency without causing deleterious effects (Mah et al., 1998).

transduction. For controls, tyrphostins specific for tumor necrosis factor ! (TNF-!) production, AG126, TNF-! cytotoxicity, AG1288 (Novogrodsky et al., 1994), platelet-derived growth factor receptor protein tyrosine kinase (PDGFR PTK), AG1295 and AG1296 (Kovalenko et al., 1994), were also used. It was evident that among all the inhibitors tested, treatment with tyrphostin 1 resulted in the greatest increase in recombinant AAV transduction efficiency (without causing significant cytotoxicity) followed by that of tyrphostins 23, 63, 25, 46, then 47. Again, these results emphasize the role EGFR PTK plays in recombinant AAV-mediated transgene expression. As expected, the control tyrphostins AG126, AG1288, AG1295, and AG1296 had no significant effect. In toxicity experiments, with reference to the mock-treated or solvent alone controls, both tyrphostin 1 and tyrphostin 23 are far less toxic than either genistein or hydroxyurea (HU), two reagents that have been previously shown to increase AAV transduction efficiency (Russell et al., 1995; Ferrari et al., 1996; Qing et al., 1997). Therefore, treatment of primary cells with tyrphostin may offer a physiological means to

We have previously demonstrated that recombinant AAV transduction efficiency correlates well with the phosphorylation state of the cellular ssD-BP (Qing et al., 1998). For example, in HeLa cells, the ssD-BP is predominantly in the phosphorylated form, and these cells are not readily transduced by recombinant AAV vectors. 293 cells, on the other hand, are very well transduced by recombinant AAV, and have been demonstrated to contain predominantly the dephosphorylated form of the ssD-BP. Following treatment of HeLa cells, all active tyrphostins caused a significant increase in the amount of dephosphorylated form of the ssD-BP, as determined by electrophoretic mobility-shift assays (EMSAs). Consistent with our previous data (Qing et al., 1998), the amount of dephosphorylated ssD-BP for each treatment corresponded with the level of increase in transduction efficiency for each of the compounds. That is, the greater the amount of dephosphorylated ssD-BP, the greater the increase in AAVmediated transgene expression. When 293 cells, either mock-treated or treated with EGF, were analyzed, the ssD-

60

Gene Therapy and Molecular Biology Vol 3, page 61 BP was present mostly in the dephosphorylated form in mock-treated cells as observed previously (Qing et al., 1997, 1998), whereas EGF treatment resulted in a significant increase in the amount of the phosphorylated form of ssD-BP. These results strongly suggest that EGFR PTK plays a direct role in the phosphorylation of the ssD-BP (Mah et al., 1998).

in these cells. To this end, equivalent numbers of A431 and H69 cells were either mock-treated or treated with EGF and analyzed by EMSA. As expected, EGF-treatment had no significant effect on the phosphorylation state of the ssD-BP in either cell type. In A431 cells, the ssD-BP was found to be predominantly in the phosphorylated form due to the high-levels of EGFR PTK expression. On the other hand, both phosphorylated and dephosphorylated forms of the ssD-BP were detected in H69 cells. Interestingly, however, treatment with either tyrphostin or genistein resulted in the conversion from the phosphorylated to the dephosphorylated form of the ssD-BP, consequently resulting in increased transduction in A431 cells. Under identical conditions, however, neither tyrphostin nor genistein treatments had any effect on the phosphorylation state of the ssD-BP in H69 cells, and these cells could not be transduced by AAV as they lack the cell surface receptor for AAV. Although it may not be readily apparent which cellular protein tyrosine kinase is responsible for the phosphorylation of the ssD-BP in H69 cells, these results are in agreement with the conclusion that the phosphorylation of the ssD-BP in A431 cells is catalyzed by the EGFR PTK (Mah et al., 1998).

B. Recombinant AAV transduction efficiency correlates inversely with the EGFR expression If EGFR PTK is responsible for catalyzing phosphorylation of the ssD-BP, then AAV-mediated transgene expression would be expected to be significantly lower in cells which express higher numbers of EGFRs than those which express fewer numbers of EGFRs. Therefore, AAV transduction efficiency would inversely correlate with the extent of EGF-R expression. To further investigate this hypothesis, equivalent numbers of cells known to express very high numbers of EGFRs, A431 cells (Giard et al., 1973), and cells known to express very low numbers, H69 cells (Gamou et al., 1987), in addition to HeLa and 293 cells, were infected with a recombinant AAV vector under identical conditions followed by X-gal staining 48 hrs post-infection. Consistent with previously published data (Qing et al., 1998), the transduction efficiency in HeLa and 293 cells was approximately 4% and 20%, respectively. As expected, the transduction efficiency in A431 cells was less than 1%. Contrary to our hypothesis, very little transduction (<1%) was also noted in the H69 cells. This apparent paradox was addressed by performing radiolabeled EGF and AAV binding assays. EGF binding assays demonstrated that A431 cells bound the greatest amounts of EGF, followed by HeLa, then 293 cells. H69 cells bound negligible amounts of EGF, as expected. It was evident from AAV binding assays that, similar to M07e cells, previously shown to lack AAV receptors (Bartlett and Samulski, 1998), H69 cells also do not express the cellular receptor for AAV. On the other hand, the low transduction efficiency seen in the A431 cells could not be attributed to a lack of expression of AAV receptors as these cells expressed far greater numbers of AAV receptors than HeLa or 293 cells.

C. Stable transfection of EGFR cDNA into 293 cells causes phosphorylation of the ssDBP and results in inhibition of AAVmediated transgene expression As 293 cells can be efficiently transduced by recombinant AAV vectors, since they contain predominantly the dephosphorylated from of the ssD-BP (Qing et al., 1997, 1998), we examined whether the deliberate over-expression of EGFR PTK in these cells would lead to phosphorylation of the ssD-BP, and consequently, result in the inhibition of AAV-mediated transgene expression. 293 cells were transfected with an EGFR cDNA expression plasmid and a number of stably transfected clones were used to determine the ratios of the dephosphorylated to the phosphorylated forms of the ssDBP and compared with that in control, untransfected 293 cells. Replicate cultures were also evaluated for the efficiency of recombinant AAV transduction, with or without pre-treatment with tyrphostin 1. These results indicated that in each of the transfected 293 cell clones, the ratio of dephosphorylated/phosphorylated ssD-BPs was reduced to an average of 0.45 from greater than 3.5 in control cells, which concomitantly led to a significant decrease in AAV transduction efficiency from approximately 18% in control 293 cells to an average of about 2% in the EGF-R-transfected 293 cell clones. These data strongly support the hypothesis that the EGFR-ssDBP interaction plays a crucial role in AAV-mediated transgene expression.

As the EGFR PTK appeared to catalyze the phosphorylation of the ssD-BP, it was next of interest to examine the effects of EGF, as well as tyrphostin- and genistein-treatments on A431 and H69 cells. Due to the high-levels of expression of EGFR in A431 cells, we hypothesized that the ssD-BP would be present in its phosphorylated form and that EGF treatment would have no effect on it's phosphorylation state. Similarly, it would be expected that H69 cells would also fail to respond to EGF treatment since little expression of the EGFR occurs 61

Mah et al: AAV vectors and the EGFR connection

Figure 2. A possible model for the role of the cellular EGFR PTK in AAV-mediated transgene expression. The phosphorylated ssDBP, which is phosphorylated by the EGFR PTK, binds to the single-stranded Dsequence within the AAV-ITR, and blocks the viral second-strand DNA synthesis. Co-infection with adenovirus, or expression of the Ad E4orf6 protein, or treatment with HU, genistein, or tyrphostins, leads to dephosphorylation of the ssD-BP, either via inhibition of the EGFR PTK, or via activation of a hitherto unknown cellular phosphotyrosine phosphatase, leading to some type of conformational change in the ssD-BP which, in turn, allows the viral second-strand DNA synthesis resulting in augmentation in transcription and translation of the transgene (Mah et al., 1998).

EGFR PTK. Based on all the available data, we propose a model for the subsequent steps in AAV-mediated transduction which is shown in Figure 2. In this model, the phosphorylated ssD-BP, which is phosphorylated by the EGFR PTK, binds to the single-stranded D-sequence within the AAV-ITR, and blocks the viral second-strand DNA synthesis. Co-infection with adenovirus, or expression of the Ad E4orf6 protein, or treatment with HU, genistein, or tyrphostins, leads to dephosphorylation of the ssD-BP, either via inhibition of the EGFR PTK, or

Subsequent in vitro phosphorylation assays performed with the commercially available purified EGFR PTK (McGlynn et al., 1992; Weber et al., 1984) and the affinity column-purified dephosphorylated form of the ssD-BP from 293 cells indicated that the ssD-BP was phosphorylated by the EGFR PTK and that this phosphorylation was abrogated in the presence of tyrphostin 1 and tyrphostin 23. These results provide direct evidence that the ssD-BP is a downstream target of the

62

Gene Therapy and Molecular Biology Vol 3, page 63 via activation of a hitherto unknown cellular phosphotyrosine phosphatase, leading to some type of conformational change in the ssD-BP which, in turn, allows the viral second-strand DNA synthesis resulting in augmentation in transcription and translation of the transgene (Mah et al., 1998).

The research in authors’ laboratory was supported in part by Public Health Service grants (HL-48342, HL53586, HL-58881, and DK-49218, Centers of Excellence in Molecular Hematology) from the National Institutes of Health, and a grant from the Phi Beta Psi Sorority. A.S. was supported by an Established Investigator Award from the American Heart Association.

IV. Conclusions and future prospects

References

The identification of FGFR1 as a co-receptor for AAV is an important step forward (Qing et al., 1999). Interestingly, however, although FGFRs have been shown to be expressed in every organ and tissue examined (Givol and Yayon, 1992), the relative abundance of their expression in skeletal muscle and in neuroblasts and glioblasts in the brain correlates particularly well with the documented high efficiency of AAV-mediated transduction in these tissues in vivo (Fisher et al., 1997; Kaplitt et al., 1994; Kessler et al., 1996; McCown et al., 1996; Xiao and Samulski, 1996). Since there are at least four distinct but related members in the FGFR family, viz FGFR1, FGFR2, FGFR3, and FGFR4 (Ledoux et al., 1992), it should be of interest to now systematically examine, both in vitro and in vivo, the relative involvement of each of these members in facilitating successful infection by AAV.

Akiyama T, Ishida J, Nakagawa S, Ogawara H, Watanabe S, Itoh N, Shibuya M, Fukami Y (1 9 8 7 ). Genistein, a specific inhibitor of tyrosine-specific protein kinases. J B i o l C h e m 262, 5592-5595. Barnes S, Peterson GT (1 9 9 5 ). Biochemical targets of the isoflavone genistein in tumor cell lines. S o c E x p B i o l Med 280, 103-109. Bartlett JS, Samulski RJ (1 9 9 8 ) Fluorescent viral vectors: A new technique for the pharmacological analysis of gene therapy. Nature Med 4, 635-637. Berns KI, Giraud C (1 9 9 6 ) Biology of adeno-associated virus. Curr Top Microbiol Immunol 218, 1-23. Carlo-Stella C, Regazzi E, Garau D, Mangoni L, Rizzo MT, Bonati A, Dotti G, Almici C, Rizzoli V (1 9 9 6 ) Effect of the protein tyrosine kinase inhibitor genistein on normal and leukaemic haemopoietic progenitor cells. Br J Haematol 93, 551-557.

The demonstration that the cellular EGFR PTK catalyzes phosphorylation of the ssD-BP, a crucial player in AAV-mediated transduction, this kinase should be an easy target for inhibition by low-toxicity compounds for their ability to significantly increase recombinant AAV transduction efficiency which may prove to be valuable for gene therapy. Although it is possible that other factors, in addition to the ssD-BP phosphorylation state, act in concert to influence the AAV transduction efficiency, it is noteworthy, however, that skeletal muscle and brain tissues, which have been shown to be extremely welltransduced by recombinant AAV vectors in vivo (Fisher et al., 1997; Kessler et al., 1996; Kaplitt et al., 1994; McCown et al., 1996; Xiao and Samulski, 1996), express little to no EGFR (Lim and Hauschka, 1984; Styren et al., 1993). Further studies on the interaction between FGFR and additional downstream target proteins, and the possible interaction between FGFR and EGFR should allow for a clearer understanding of molecular events involved in highefficiency AAV transduction which, in turn, should lead to improvements in the optimal use of AAV vectors in human gene therapy.

Constantinou A, Huberman E (1 9 9 5 ) Genistein as an inducer of tumor cell differentiation: possible mechanisms of action. S o c E x p B i o l M e d 203, 109-115. Couldwell WT, Hinton DR, He S, Chen TC, Sebat I, Weiss MH, Law RE (1 9 9 4 ) Protein kinase C inhibitors induce apoptosis in human malignant glioma cell lines. FEBS Lett 345, 43-46. Ferrari FK, Samulski T, Shenk T Samulski RJ (1 9 9 6 ) Secondstrand synthesis is a rate-limiting step for efficient transduction by recombinant adeno-associated virus vectors. J Virol 70, 3227-3234. Fisher KJ, Gao G-P, Weitzman MD, DeMatteo R, Burda JF, Wilson JM (1 9 9 6 ) Transduction with recombinant adenoassociated virus for gene therapy is limited by leadingstrand synthesis. J Virol 70, 520-532. Fisher KJ, Jooss K, Alston J, Yang Y, Haecker SE, High K, Pathak RE, Raper SE, Wilson JM. (1 9 9 7 ). Recombinant adeno-associated virus for muscle directed gene therapy. Nature Med 3, 306-312. Flotte TR, Carter BJ (1 9 9 7 ) In vivo gene therapy with adenoassociated virus vectors for cystic fibrosis. Adv Pharmacol 40, 85-101. Fukazawa H, Li PM, Yamamoto C, Murakami Y, Mizuno S, Uehara Y (1 9 9 1 ) Specific inhibition of cytoplasmic protein tyrosine kinases by herbimycin A in vitro. Biochem Pharmacol 42, 1661-1671.

Acknowledgments 63

Mah et al: AAV vectors and the EGFR connection Gamou S, Hunts J, Harigai H, Hirohashi H, Shimosato Y, Pastan I, Shimizu N (1 9 8 7 ) Molecular evidence for the lack of epidermal growth factor receptor gene expression in small cell lung carcinoma cells. Cancer R e s 47, 2668-2673.

Lebakken CS, Rapraeger AC (1 9 9 6 ) Syndecan-1 mediates cell spreading in transfected human lymphoblastoid (Raji) cells. J C e l l B i o l 132, 1209-1221. Ledoux D, Gannoun-Zaki L, Barritault D (1 9 9 2 ) Interactions of FGFs with target cells. Prog Growth Factor Res 4, 107-120.

Gazit A, Yaish P, Gilon C, Levitzki A (1 9 8 9 ) Tyrphostin I: Synthesis and biological activity of protein tyrosine kinase inhibitors. J Med Chem 32, 2344-2352.

Levitzki A (1 9 9 0 ) Tyrphostin - Potential anti-proliferative agents and novel molecular tools. Biochem Pharmacol 40, 913-918.

Giard, DJ, Aaronson SA, Todaro GJ, Arnstein P, Kersey JH, Dosik H, Parks WP (1 9 7 3 ) In vitro cultivation of human tumors: establishment of cell lines derived form a series of solid tumors. J Natl Cancer Inst 51, 1417-1423.

Levitzki A, Gazit A, Osherov N, Posner I, Gilon C (1 9 9 1 ) Inhibition of protein tyrosine kinases by tyrphostin. M e t h E n z y m o l 201, 347-361.

Givol D, Yayon A (1 9 9 2 ) Complexity of FGF receptors: Genetic basis for structural diversity and functional specificity. FASEB J 6, 3362-3369.

Lim RW, Hauschka SD (1 9 8 4 ) A rapid decrease in epidermal growth factor binding capacity accompanies the terminal differentiation of mouse myoblasts in vitro. J C e l l B i o l 98, 739-747.

Green PJ, Walsh FS, Doherty P (1 9 9 6 ) Promiscuity of fibroblast growth factor receptors. B i o E s s a y s 18, 639646.

Lyall RM, Zilberstein A, Gazit A, Gilon C, Levitzki A, Schlessinger J (1 9 8 9 ) Tyrphostin inhibit epidermal growth factor (EGF)-receptor tyrosine kinase activity in living cells and EGF-stimulated cell proliferation. J B i o l Chem 264, 14503-14509.

Johnson DE, Lee PL, Lu J, Williams LT (1 9 9 0 ) Diverse forms of a receptor for acidic and basic fibroblast growth factors. M o l C e l l B i o l 10, 4728-4736. Kan M, Wang F, Xiu J, Crabb JW, Hou J, McKeehan WL (1 9 9 3 ) An essential heparin-binding domain in the fibroblast growth factor receptor kinase. S c i e n c e 259, 1918-1921.

Mah C, Qing KY, Khuntirat B, Ponnazhagan S, Wang X-S, Kube DM, Yoder MC, Srivastava A (1 9 9 8 ) Adenoassociated virus type 2-mediated gene transfer: Role of epidermal growth factor receptor protein tyrosine kinase in transgene expression. J Virol 72, 9835-9843.

Kaplitt MG, Leone P, Samulski RJ, Xiao X, Pfaff DW, O'Malley KL, During MJ (1 9 9 4 ) Long-term gene expression and phenotypic correction using adenoassociated virus vectors in the mammalian brain. Nature Genet 8, 148-153.

McCown TJ, Xiao X, Li J, Breese GR, Samulski RJ (1 9 9 6 ) Differential and persistent expression patterns of CNS gene transfer by an adeno-associated virus (AAV) vector. Brain Res 713, 99-107.

Kessler PD, Podsakoff GM, Chen X, McQuiston SA, Colosi PC, Matelis LA, Kurtzman GJ, Byrne BJ (1 9 9 6 ) Gene delivery to skeletal muscle results in sustained expression and systemic delivery of a therapeutic protein. P r o c N a t l Acad Sci USA 93, 14082-14087.

McGlynn E, Becker M, Mett H, Reutner S, Cozens R, Lydon NB (1 9 9 2 ) Large scale purification and characterization of a recombinant epidermal growth factor receptor protein tyrosine kinase. Eur J Biochem 207, 265-275. Mohammadi M, McManhon G, Sun L, Tang C, Hirth P, Yeh BK Hubbard SR, Schlessinger J (1 9 9 7 ) Structures of the tyrosine kinase domain of fibroblast growth factor receptor in complex with inhibitors. S c i e n c e 276, 955960.

Kiefer MC, Stephans JC, Crawford K, Okino K, Barr PJ (1 9 9 0 ) Ligand-affinity cloning and structure of a cell surface heparan sulfate proteoglycan that binds basic fibroblast growth factor. Proc Natl Acad Sci USA 87, 6985-6989.

Muzyczka N (1 9 9 2 ) Use of adeno-associated virus as a general transduction vector for mammalian cells. Curr T o p M i c r o b i o l I m m u n o l 158, 97-129.

Kotin RM, Menninger JC, Ward DC, Berns KI (1 9 9 1 ) Mapping and direct visualization of a region-specific viral DNA integration site on chromosome 19q13-qter. G e n o m i c s 10, 831-834.

Novogrodsky A, Vanichkin A, Patya M, Gazit A, Osherov N, Levitzki A (1 9 9 4 ) Prevention of lipopolysaccharideinduced lethal toxicity by tyrosine kinase inhibitors. S c i e n c e 264, 1319-1322.

Kovalenko M, Gazit A, Bรถhmer A, Rorsman C, Rรถnnstrand L, Heldin C-H, Waltenberger J, Bรถhmer FO, Levitzki A (1 9 9 4 ). Selective platelet-derived growth factor receptor kinase blockers reverse sis-transformation. Cancer Res 54, 6106-6114.

Okada T, Sakuma L, Fukui Y, Hazeki O, Ui M (1 9 9 4 ). Blockage of chemotactic peptide-induced stimulation of neutrophils by wortmannin as a result of selective inhibition of phophatidylinositol 3-kinase. J B i o l Chem 269, 3563-3567.

Kuo ML, Yang NC (1 9 9 5 ) Reversion of v-H-ras-transformed NIH 3T3 cells by apigenin through inhibiting mitogen activated protein kinase and its downstream oncogenes. B i o c h e m B i o p h y s R e s C o m m 212, 767-75.

Ponnazhagan S, Mukherjee P, Wang X-S, Qing KY, Kube DM, Mah C, Kurpad C, Yoder MC, Srour EF, Srivastava A (1 9 9 7 ) Adeno-associated virus type 2-mediated transduction of primary human bone marrow-derived

64

Gene Therapy and Molecular Biology Vol 3, page 65 CD34 + hematopoietic progenitor cells: Donor variation and correlation of transgene expression with cellular differentiation. J Virol 71, 8262-8267.

Summerford C, Samulski RJ (1 9 9 8 ) Membrane-associated heparan sulfate proteoglycan is a receptor for adenoassociated virus type 2 virions. J V i r o l 72, 1438-1445.

Ponnazhagan S, Wang X-S, Woody MJ, Luo F, Kang LY, Nallari ML, Munshi NC, Zhou SZ, Srivastava A (1 9 9 6 ). Differential expression in human cells from the p6 promoter of human parvovirus B19 following plasmid transfection and recombinant adeno-associated virus 2 (AAV) infection: Human megakaryocytic leukaemia cells are non-permissive for AAV infection. J G e n V i r o l 77, 1111-1122.

Vlahos CJ, Matter WF, Hui KY, Brown RF (1 9 9 4 ) A specific inhibitor of phosphatidylinositol 3-kinase, 2-(4morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002) . J B i o l C h e m 269, 5241-5248. Weber W, Bertics PJ, Gill GN (1 9 8 4 ) Immunoaffinity purification of the epidermal growth factor receptor. J B i o l C h e m 259, 14631-14636. Xiao X, Li J, Samulski RJ (1 9 9 6 ) Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector. J V i r o l 70, 8098-8108.

Qing KY, Khuntirat B, Mah C, Kube DM, Wang X-S, Ponnazhagan S, Zhou SZ, Dwarki VJ, Yoder MC, Srivastava A (1 9 9 8 ) Adeno-associated virus type 2mediated gene transfer: Correlation of tyrosine phosphorylation of the cellular single-stranded D sequence-binding protein with transgene expression in human cells in vitro and murine tissues in vivo. J V i r o l 72, 1593-1599.

Yaish P, Gazit A, Gilon C, Levitzki A (1 9 8 8 ) Blocking of EGF-dependent cell proliferation by EGF receptor kinase inhibitors. S c i e n c e 242, 933-935.

Qing KY, Mah C, Hansen J, Zhou SZ, Dwarki VJ, Srivastava A (1 9 9 9 ). Human fibroblast growth factor receptor 1 is a co-receptor for infection by adeno-associated virus 2. Nature Med 5, in press. Qing KY, Wang X-S, Kube DM, Ponnazhagan S, Bajpai A, Srivastava A (1 9 9 7 ). Role of tyrosine phosphorylation of a cellular protein in adeno-associated virus 2-mediated transgene expression. Proc N a tl A c a d S c i USA 94, 10879-10884. Rapraeger AC, Krufka A, Olwin BB (1 9 9 1 ) Requirement of heparan sulfate for bFGF-mediated fibroblast growth and myoblast differentiation. S c i e n c e 252, 1705-1708. Roghani M, Moscatelli D (1 9 9 2 ) Basic fibroblast growth factor is internalized through both receptor-mediated and heparan sulfate-mediated mechanisms. J B i o l Chem 267, 22156-22162. Russell DW, Alexander IE, Miller AD (1 9 9 5 ) DNA synthesis and topoisomerase inhibitors increase transduction by adeno-associated virus vectors. Proc Natl Acad S c i USA 92, 5719-5723. Samulski RJ, Zhu X, Xiao X, Brooke JD, Houseman DE, Epstein N, Hunter LA (1 9 9 1 ) Targeted integration of adeno-associated virus (AAV) into human chromosome 19. EMBO J 10, 3941-3950 Saunders S, Jalkanen M, O'Farrell S, Bernfield M (1 9 8 9 ) Molecular cloning of syndecan, an integral membrane proteoglycan. J C e l l B i o l 1 0 8 , 1547-1556. Srivastava A, Lusby EW, Berns KI (1 9 8 3 ) Nucleotide sequence and organization of the adeno-associated virus 2 genome. J V i r o l 45, 555-564. Styren SD, DeKosky ST, Rogers J, Mufson EJ (1 9 9 3 ) Epidermal growth factor receptor expression in demented elderly: localization to vascular endothelial cells of brain, pituitary, and skin. Brain Res 615, 181-190.

65

Gene Therapy and Molecular Biology Vol 3, page 67 Gene Ther Mol Biol Vol 3, 67-74. August 1999.

Hepatocyte-specific gene expression by a recombinant adeno-associated virus vector carrying the apolipoprotein E enhancer and 1-antitrypsin promoter Research Article

Torayuki Okuyama 1,2, Motomichi Kosuga1,2, Satori Takahashi1, Kyoko Sasaki1, and Masao Yamada1 Department of Genetics, National Children’s Medical Research Center, Setagaya, Tokyo 154-8509 Japan, Department of Pediatrics, Keio University School of Medicine, Tokyo 160-8582 __________________________________________________________________________________ Correspondence: Torayuki Okuyama, M.D., Department of Genetics, National Children’s Medical Research Center, 3-35-31 Taishido Setagaya-ku Tokyo 154-8509, Japan. Phone: +81-3-3414-8121 ext. 2752; Fax: +81-3-3414-3208; E-mail: tora@nch.go.jp Key words: adeno-associated virus, liver-specific promoter, ! 1 -antitrypsin promoter, apolipoprotein E enhancer, gene therapy Received: 30 October 1998; accepted: 10 November 1998

Summary An adeno-associated virus vector was constructed to express exogenous genes to the liver. The original plasmid construct carried two expression units; a neomycin resistant gene and human 1antitrypsin cDNA under the control o f hepatocyte specific transcription elements. C e l l s were transfected with the constructed plasmid DNA with another packaging plasmid, and recombinant adeno-associated viruses (rAAV) were then recovered after adenovirus infection. Alternatively, rAAV were recovered by transduction of DNAs of the packaging plasmid and adenovirus into preselected cells carrying constructed proviral DNA. When the transducing abilities were evaluated b a s e d o n G 4 1 8 r e s i s t a n t c o l o n y f o r m a t i o n o n H e L a c e l l s , t h e l a t t e r m e t h o d w a s found t o give almost 10-fold more rAAV. We then isolated G418 resistant colonies and established several independent clones for the HeLa and Hepa1A cells infected with the rAAV. All of the eight clones derived from Hepa1A cells produced significant amounts of the human 1-antitrypsin protein. In contrast, none o f the five clones derived from HeLa c e l l s produced a detectable l e v e l o f 1antitrypsin. Our results suggest that liver-specific promoter and enhancer maintain the tissue specificity in the rAAV construct, and that the rAAV vector system would be useful in hepatocyte directed gene therapy.

administration because of the high immunogenicity (Jaffe et al. 1992; Okuyama et al. 1998). Retroviral vectors are also able to transduce an exogenous gene into hepatocytes, and a long term expression of the transduced gene has been identified in several experiments using rat or dog liver (Rettinger et al. 1994; Kay et al. 1992; Kay et al. 1993; Hafenrichter et al. 1994). However, the expression level is

I. Introduction Liver-directed gene therapy could revolutionize treatments for many genetic disorders such as phenylketonuria, familial hypercholesterolemia and hemophilia (Ledley 1993). Adenoviral vectors efficiently transduce a gene into hepatocytes, easily achieve its expression at a therapeutic level for many diseases, but do not allow a long-term expression and repetitive 67

Okuyama et al: Hepatocyte-specific gene expression using the apo E enhancer

II. Results

generally too low for therapeutic treatments of patients because the transducing efficiency is extremely low.

A. Generation of rAAV vector containing a liver-specific promoter and enhancer

We have previously demonstrated that a retroviral vector expressing an exogenous gene under the control of a human apolipoprotein E enhancer and ! 1-antitrypsin promoter as well as an original retroviral LTR promoter dramatically increase the level of protein production after administration into the rat liver (Okuyama et al. 1996). The apolipoprtein E enhancer has been detected through studies on a gene cluster of apoE/C-I/C-II in human chromosome 19. Studies on transgenic mice disclosed that a 154 bp region located 15 kb downstream of the apolipoprotein E gene was responsible for the high level of expression in hepatocytes (Shachter et al. 1993; Simonet et al. 1993). A 420 bp segment of the 5’ flanking region of the human ! 1-antitrypsin gene contains distinct HNF-1 and HNF-2 binding sites, and both sites are responsible for strong and tissue-specific expression of ! 1-antitrypsin (Li et al. 1988). Recently we found that rats administered the retroviral vector expressing human coagulation Factor X under the control of the 420 bp of ! 1-antitrypsin promoter produced a therapeutic level of functional Factor X (Le et al. 1997). These observations suggested that with the retroviral vectors designed to express an exogenous gene under the control of promoter-enhancer complex of apolipoprotein E and ! 1-antitrypsin, one is able to achieve significantly high levels of transgene expression. However, retroviral vectors can transduce foreign genes only into dividing cells, thus inducing the regeneration with partial hepatectomy which is essential for retroviral genetransduction into hepatocytes. Although the mortality for a 70% hepatectomy is relatively low and the procedure could be justified for life threatening genetic deficiencies,alternative methods that circumvent partial hepatectomy are desirable.

The structure of the rAAV vector TRNAEAT containing liver-specific transcriptional elements is shown in F i g . 1 . The vector contains the human ! 1-antitrypsin (hAAT) cDNA as a reporter gene downstream of the 420bp 5’ flanking sequence of its own gene. The enhancer region located in the apoE/C-I/C-II gene locus is necessary for high level expression of the apoE gene in livers of transgenic mice (Schacter et al. 1993). In an attempt to further enhance expression of the hAAT promoter from an rAAV vector, four copies of ApoE enhancer were placed upstream of the hAAT promoter (Simonet et al. 1993). In addition to these 2.8-kb hepatocyte-specific expression units of hAAT, an 1.5 kb expression cassette of neomycin resistant gene under the control of mouse phosphoglycerate kinase promoter (Soriano et al. 1991; Adra et al. 1987) was also introduced into the AAV vector plasmid pTR (+).

B. Comparison of the efficiency of rAAV generation We produced rAAV vectors via two different methods. Method I was a conventional co-transfection method described before (Zolotukhin et al. 1996). Briefly, HeLa cells were co-transfected with the vector plasmid pTRNAEAT and packaging plasmid pIM45 (Peel et al. 1997) at a molar ratio of 1:2 using the calcium-phosphate method, followed by the infection of human adenovirus 5 at MOI.2. For Method II, we established a stable HeLa cell line “HeLa-TRNAEAT” carrying proviral sequences of AAV-TRNAEAT by transfecting HeLa cells with pTRNAET following the selection in G418. HeLaTRNAEAT cells were transfected with pIM45 and infected with Ad5. Then, wild type HeLa cells were infected with the rAAV generated by these two methods and selected with G418 to determine the Neo-resistant titers. The rAAV titer of the viral stock from Method I was 0.8±0.4x104 cfu/ml, while those from Method II was 1.0±0.2x105 cfu/ml (F i g . 2 ). We were able to generate rAAV stocks with 10 fold higher titer using Method II. Moreover, since more consistent and reproducible results were obtainable with the latter method, we used Method II for generating rAAV for further analysis.

Adeno-associated virus is a replication-defective parvovirus that is being developed as a vector for human gene therapy (Laughlin et al. 1986). One advantage of AAV as a vector is that it can transduce genes into postmitotic cells like cells of the Central Nervous System (Kaplitt et al. 1994), lung epithelial cells (Flotte et al. 1993), or muscle fiber cells (Fisher et al. 1996). Since most hepatocytes in vivo are also in the growth arrested state, AAV vectors is expected to be suitable for in vivo hepatocyte-directed gene therapy. However, little is known about the transduction and expression efficiency of rAAV in hepatocytes in vitro as well as in vivo (Flotte et al. 1995; Fisher et al. 1997; Snyder et al. 1997). In this study, we generated a recombinant adeno-associated virus containing hepatocyte-specific expression unit, and evaluated its transduction efficiency, tissue specificity, and level of expression in gene-transduced cells of hepatocyte origin.

C. Testing the rAAV vectors for infectivity to human and rodent hepatoma cell lines We determined the NeoR titers of HepG2 and Hepa1A cells to evaluate the infectivity of the rAAV to cell lines of hepatocyte origin. The NeoR titers for HepG2 and Hepa-

68

Gene Therapy and Molecular Biology Vol 3, page 69

F i g u r e 1 . Schematic presentation of recombinant adeno-associated virus “TRNAEAT”. Two expression cassettes were introduced in an opposite direction to each other between two terminal inverted repeat sequences of AAV. Expression of neomysin resistant gene is under the control of phophorylgycerate kinase promoter, and expression of human ! 1 -antitrypsin cDNA is expected to be under the control of human ! 1 -antitrypsin promoter and apolipoprotein E enhancer. Tr, inverted terminal repeat sequence of adeno-associated virus; Neo-R, expression cassette for neomycin resistant gene; ApoEEn, 4 copies of apolipoprotein E enhancer; hAATcDNA, cDNA for human ! 1 -antitrypsin; A, polyA signal of SV40

Figure 3. Infectivity of rAAV in different cell lines. NeoR titers of the rAAV prepared with Method II were determined with 4 different cell lines, HeLa (Lane 1), NIH3T3 (Lane 2), HepG2 (Lane 3), and Hepa1A (Lane 4). Data is presented as the average ± SE.

1A cells were 0.8±0.2x105 cfu/ml and 0.6±0.1x105 cfu/ml, respectively. The same viral solution was used to calculate the titer in HeLa cells and NIH3T3 cells (F i g . 3 ). The relative infectivity of the rAAV in HepG2 and Hepa1A cells compared to those of HeLa cells were 0.75 and 0.67, respectively. These observations suggested that it is possible to transduce exogenous genes into cell lines of hepatocyte origin with similar efficiency as into HeLa or NIH3T3 cells using the rAAV gene transfer system.

F i g u r e 2 . Neo R titers of rAAV “TRNAET” in HeLa cells. The viral stocks of rAAV ”TRNAEAT” were prepared by two different methods, Method I and Method II (see details in text), and Neo R titers of each viral stock were determined in HeLa cells. Data is presented as the average ± SE.

69

Okuyama et al: Hepatocyte-specific gene expression using the apo E enhancer T a b l e 1 . Comparison of levels of human ! 1 -antitrypsin (hAAT) expression in Hepa1A cells infected with rAAV “TRNAEAT” and retroviral vector “ApoE(-)haat-LTR”. (Data is presented as the average (±SE) Vectors for hAAT gene transduction

hAAT expression (ng/ million cells / 24 h)

rAAV-TRNAEAT

103.8±23.5

retroviral vector “ApoE(-)hAATLTR”

71.0±6.6

D. Cell-type specific expression of cDNA for human 1 -antitrypsin in Hepa1A cells transduced with the rAAV “TRNAEAT” Hepa1A cells and HeLa cells infected with the rAAV were selected in G418 for 14 days, and eight Hepa1A and five HeLa clones were isolated. To verify that the cDNA region of hAAT was introduced into the cells, the 400 bp partial hAAT cDNA was amplified using purified genomic DNA as template. The expected DNA fragments were amplified in all HeLa and HepG2 clones, indicating that the expression unit of hAAT was integrated into the chromosomal DNA of the cells with the rAAV vector (F i g . 4 A and 4B). In HeLa and HepG2 cells, we identified faint 1.8-kb amplified DNA fragments, corresponding to the endogenous human hAAT gene. The hAAT protein secreted into the media of each clone was assayed by ELISA using human-specific antibody for ! 1antitrypsin. All eight clones of Hepa1A cells secreted hAAT into the media. The average amount of hAAT protein secreted from the rAAV-infected Hepa1A cells was 103.8 ng /106cells /24 hours (F i g . 5 A ). This represented 31% of hAAT secreted from one million cells of wild type HepG2, and was similar to the level of hAAT secreted from Hepa1A cells infected with retroviral vector “apoE(-) haat-LTR” carrying an identical liver-specific expression cassette for hAAT (Table 1). On the other hand, none of the HeLa clones obtained by the infection of rAAVTRNAEAT secreted detectable levels of hAAT protein in cultured media, although expression units for hAAT were administered into the host cell chromosome (F i g . 5B). These results suggested that rAAV-TRNAEAT was able to express exogenous genes exclusively in cells of hepatocyte origin.

Figure 4 . Detection of proviral genome sequence in Hepa1A and HeLa cells infected with rAAV ”TRNAET”. A 400 bp DNA region of human ! 1 -antitrypsin was amplified using purified genomic DNA of rAAV-infected cells as templates. (A) PCR results for the eight Hepa 1A clones. Template DNA samples of PCR reactions were as below, pTRNAET (Lane P), wild type Hepa1A (Lane 0), rAAV-infected Hepa1A clones No.1-No.8 (Lane 1-Lane 8), and wild type HepG2 (Lane G). Lane M stands for DNA size markers, HindI II digested Lambda DNA (left) and Sau3AI digested PUC 19 (right). The 400 bp amplified DNA fragments were identified in Lanes 1 to 8, indicating that proviral sequences of the rAAV were introduced into all of the eight clones. This signal was not identified in Lane 0 and Lane G, but a 1.8 kb signal was identified in Lane G instead. This corresponds to the DNA amplified from human genome for ! 1 -antitrypsin. (B ) PCR results for five independent HeLa cell clones. Template DNA samples of PCR reactions were wild type HeLa cells (Lane 0), and rAAVinfected HeLa clones No.1-No.5 (Lane 1-Lane5). Lane P, Lane G, and Lane M were same as for F i g u r e 4 A . PCR of the five HeLa clones and wild type HeLa cells resulted 1.8 kb DNA fragments amplified from human gene for ! 1 -antitrypsin, but only AAV-infected HeLa clones showed 0.4 kb fragments corresponding to the rAAV proviral sequence.

III. Discussion In this report we have shown that a recombinant adenoassociated virus vector is able to transfer an exogenous gene into human or rodent cells of hepatocyte origin with a similar efficiency as in HeLa cells or in NIH3T3 cells, and 70

Gene Therapy and Molecular Biology Vol 3, page 71 that it is possible to obtain hepatocyte specific transgene expression using a liver-specific promoter and enhancer. We previously generated a retroviral vector expressing ! 1-antitrypsin cDNA under the control of the enhancerpromoter complex of apolipoprotein E and ! 1-antitrypsin. This retroviral vector, apoE(-)haat-LTR, showed markedly increased expression of ! 1-antitrypsin in rat liver in vivo (Okuyama et al. 1996). The rAAV vector TRNAEAT contains the same liver-specific expression cassette, and the levels of transgene expression in Hepa1A cells infected with rAAV-TRNAEAT were similar to those in the same cells infected with the retroviral vector apoE(-)haat-LTR. These in vitro results suggest that a high level of expression could be expected in vivo using rAAVTRNAEAT, if similar gene transduction efficiency with retoroviral vectors is obtainable using the rAAV gene transfer system. In order to evaluate the level of expression in rat liver in vivo, however, it is necessary to prepare high titer viral stocks. It is difficult to generate high titer AAV viral particles in large scale with the conventional cotransfection method (Peel et al. 1997). To circumvent this problem, we established a HeLa cell line encoding the proviral genome sequence of the rAAV-TRNAEAT. Using this cell line, it was possible to obtain more than 10 fold higher titer viral stocks easily, and this result was highly reproducible (F i g . 2 ). Flotte et al. (1995) tried a similar approach using 293 cells, and were successful in generating rAAV with a 5 fold higher titer compared with the conventional co-transfection method. One of the potential advantages of rAAV for hepatic gene therapy is that it is possible to transduce genes into non-dividing cells (Podsakoff et al. 1994). Recently Snyder et. al. (1997) reported persistent transgene expression in mouse after a simple intraportal infusion of the rAAV expressing human Factor IX under control of MuLV LTR promoter/enhancer. This result suggests that the rAAV gene transfer system is promising for in vivo liver-directed gene therapy. However, one major disadvantage of in vivo hepatic gene transfer is that it is difficult to restrict gene transduction to hepatocytes, because there are many nonparenchymal cells, such as Kuppfer cells and sinusoidal endothelial cells, in the liver in addition to hepatocytes. Here we demonstrated that cell-type specific transgene expression was achievable by rAAV carrying liver-specific promoter enhancer sequences. The vector system described here has the potential advantage of eliminating the risk of miss-targeting, a problem encountered when rAAV vectors are used as an in vivo gene delivering vehicle.

Figure 5 . Quantification of human ! 1 -antitrypsin secreted from rAAV-infected Hepa 1A and HeLa cells. The amount of human ! 1 -antitrypsin protein secreted into the media was determined by ELISA. (A) The results of ELISA assay in wild type Hepa1A cells (Lane 0), in rAAV-infected Hepa1A clones No.1-No.8 (Lane1-Lane8), and in wild type HepG2 cells (Lane G). (B) The results of ELISA assay in wild type HeLa cells, rAAV-infected HeLa clones No.1-No.5 (Lane 1-Lane 5), and wild type HepG2 (Lane G). Data is presented as the average of three independent assays.

IV. Materals and Methods A. Plasmid construction The plasmid pIM45, encoding rep and cap genes of AAV, and pTR(+) for constructing vector plasmid of rAAV were generous gifts from Dr. Nick Muzyckzucka of the University of Florida. The structure of retroviral vector plasmid pAp(-) hAAT-LTR was described elsewhere (Okuyama et al. 1996). The plasmid pAp(-)hAAT-LTR was linealized at the unique BglII site, and then partially digested with BamHI. The 2.2 kb BglII-BamHI DNA fragment containing 4 copies of the 154 bp

71

Okuyama et al: Hepatocyte-specific gene expression using the apo E enhancer apolipoprotein E enhancer region, 400 bp of human ! 1 antitrypsin promoter sequence, and 1.2kb cDNA for human ! 1 antitrypsin was gel-isolated and ligated with BglII-digested pTR(+) generating the plasmid pTRAET.

G418 used for the selection: 400 µg/ml in NIH3T3 cells, 600 µg/ml in Hepa1A cells, and 800 µg/ml in HepG2 cells, respectively.

pTR(+) is a plasmid for constructing rAAV, using E.coli strain JC8111 (Deiss et al. 1990) as host cells for transformation. The plasmid pTRAET was again digested and linealized with BglII, blunt-ended with the Klenow fragment of E.coli DNA polymerase, and ligated with the 1.6 kb expression cassette of the neomycin resistant gene under the control of mouse phosphorylglycerol kinase promoter, isolated from another plasmid, pPGKNeo (Soriano et al. 1991). The plasmid pTRNAEAT was generated based on this cloning process.

D. Isolation of Hepa1A and HeLa cell clones infected with rAAV “TRNAEAT” Hepa 1A cells and HeLa cells were infected with rAAV “TRNAEAT” for 4 hours at MOI. 0.1, and two days after the infection, 600 µ g/ml of G418 was added to the media. About two weeks after the infection, several colonies were picked up, and further propagated. Finally we established eight Hepa1A clones, and five HeLa clones. Purified genomic DNA samples from these cells were used as templates of PCR reactions for detecting the 400bp DNA region of human ! 1 -antitrypsin cDNA. Forward and reverse primer sequences were 5’CACTCAGAAGCCTTCACTGTCA-3’, and 5’-ACCCAGCT GGACAGCTTCTT-3’. Thirty cycles of PCRs were performed at 1 minute of 95°C, 1 minute of 57°C, and 2 minutes of 72°C. Since the forward and reverse primers were synthesized based on the sequence of exon1 and 2 of human ! 1 -antitrypsin gene, this PCR reaction was expected to generate a 1.8kb DNA fragment covering the whole of intron 1 and the part of exon 1 and 2 of the human genomic DNA (Long et al. 1984).

B. Production of rAAV vector “TRNAEAT” HeLa cells were maintained with DMEM (GIBCO BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (Sanko Junyaku Co. Ltd., Tokyo Japan). The rAAV TRNAEAT was generated by two different methods. Method I involved conventional co-transfection. HeLa cells were transfected with pTRNAEAT and pIM45 at a molar ratio of 1:2 using calcium phosphate precipitation method described before (Chen and Okayama 1987). 24 hours after the transfection, the cells were infected with wild type adenovirus Ad5 with MOI.2 for 2 hours. Three days after the transfection, the cells were harvested, lysed by freezing and thawing 5 times, and incubated at 60 °C for one hour to inactivate co-existing adenoviruses. In Method II, we established the stable HeLa cell clone first. HeLa cells were transfected with the plasmid pTRNAEAT and selected with 600µg/ml G418 (GIBCO BRL) for 14 days, and established the HeLa cell clone HeLa-NAEAT carrying the proviral genome sequences in its chromosomes. To produce rAAV, HeLa-NAEAT was transfected with pIM45, and the transfected cells were treated in the same way as for Method I.

E. Quantification of human 1-antitrypsin produced from rAAV-infected HeLa and Hepa1A cells 24 hour-cultured media were used for ELISA assay to quantify the amounts of human ! 1 -antitrypsin secreted from the cells. The assay was performed in 96-well microtitration plates. Goat anti-human ! 1 -antitrypsin antibody, and peroxidase-conjugated goat anti-human ! 1 -antitrypsin antibody were purchased from Cappel (Durham, NC). After 2hour incubation at room temperature with goat anti-human ! 1 antitrypsin antibody (2 mg/well), non-specific binding was blocked by overnight incubation with 200 ml of 3% BSA and 0.02% sodium azide in PBS at 4°C. After rinsing with washing buffer, 200ml of the cultured media or control samples (purified human ! 1 -antitrypsin, Sigma, St. Louis MO) were added. The standard curve was made from 0 to 100 ng/ml. The microtitration plates were incubated for 3 hours at room temperature and washed four times with PBS. Then 200 ml of peroxidase-conjugated goat anti-human ! 1 -antitrypsin antibody (15mg/ml) was added to each well. After incubation for 2 hours at room temperature, the wells were rinsed five times and 200ml of substrate solution containing 10 mg ophenylendiamine hydrochloride (Sigma), 10ml 30% H2 O2 , and 25ml citrate-phosphate buffer pH5 was added. The reaction was stopped by the addition of 50ml of 3 M H2 SO4 .

C. Determination of neo R titers of the rAAV “TRNAEAT” HeLa and HepG2 cells were maintained with DMEM supplemented with 10% fetal bovine serum, NIH3T3 cells were maintained with DMEM with 10% calf serum (Sanko Junyaku Co. Ltd.), and Hepa1A cells, cells from a mouse hepatoma cell line, (Darlington 1987) were maintained with 75% MEM (GIBCO BRL), 25% Waymouth (GIBCO BRL), 10% fetal bovine serum. Serial dilutions of the viral stocks were made with DMEM, and certain amounts were added into the media of the plates culturing HeLa cells for 4 hours. Then cells were washed with PBS twice, and fed with fresh media for two more days. Two days after the infection, 600µg/ml of G418 (GIBCO BRL) was added to the media, and the culture was continued until distinct colonies were identified. The titer of each viral stock solution was calculated by counting the numbers of the G418 resistant colonies on the plates. The neoR titers of the rAAV to NIH3T3 cells, Hepa1A cells, and HepG2 cells were determined in the same way except for the concentrations of

Acknowledgements

72

Gene Therapy and Molecular Biology Vol 3, page 73 Kay MA, Rothenberg S, Landen CN, Bellinger DA, Leland F, Toman C, Finegold M, Thompson AR, Read MS, Brinkhous KM, Woo SLC (1 9 9 3 ) In vivo gene therapy of hemophilia B, sustained partial correction in factor IXdeficient dogs. S c i e n c e 262, 117-119

We thank Dr. Nick Muzyczka for recombinant adenoassociated virus construction, and Ms. K. Saito for editorial assistance. This work was supported in part by grants for pediatric research and gene therapy research from the Ministry of Health and Welfare of Japan.

Laughlin CA, Cardellichio CB, Coon HC (1 9 8 6 ). Latent infection of KB cells with adeno-associated virus type 2. J V i r o l 60, 515-524

References

Le M, Okuyama T, Cai SR, Kennedy SC, Bowling WM, Flye MW, Ponder KP (1 9 9 7 ) Therapeutic levels of functional human factor X in rats after retroviral-mediated hepatic gene therapy. B l o o d 89, 1254-1259

Adra CN, Boer PH, McBurney MW (1 9 8 7 ) Cloning and expression of the mouse pgk-1 gene and the nucleotide sequence of its promoter. Gene 60, 65-74 Chen C, Okayama H (1987) High-efficiency transformation of mammalian cells by plasmid DNA Mol Cell Biol 7, 27452752 Darlington GJ (1 9 8 7 ) Liver cell E n z y m o l o g y 151, 19-39

lines.

Methods

Ledley FD (1 9 9 3 ) Hepatic gene therapy, present and future. H e p a t o l o g y 18, 1263-1273 Li Y, Shen RF, Tsai SY, Woo SL (1 9 8 8 ) Multiple hepatic trans-acting factors are required for in vitro transcription of the human alpha-1-antitrypsin gene. M o l C e l l B i o l 8, 4362-4369

in

Deiss V, Tratschin JD, Weitz M, Siegl G (1 9 9 0 ) Cloning of the human parvovirus B19 genome and structural analysis of its palindromic termini. V i r o l o g y 175, 247-254

Long GL, Chandra T, Woo SL, Davie EW, Kurachi K (1 9 8 4 ). Complete sequence of the cDNA for human alpha 1antitrypsin and the gene for the S variant. B i o c h e m i s t r y 23, 4828-4837

Fisher KJ, Gao GP, Weitzman MD, DeMatteo R, Burda JF, Wilson JM (1 9 9 6 ) Transduction with recombinant adenoassociated virus for gene therapy is limited by leadingstrand synthesis. J Virol 70, 520-532

Okuyama T, Huber RM, Bowling W, Pearline R, Kennedy SC, Flye MW, Ponder KP (1 9 9 6 ) Liver-directed gene therapy, a retroviral vector with a complete LTR and the ApoE enhancer-alpha 1-antitrypsin promoter dramatically increases expression of human alpha 1-antitrypsin in vivo. Hum Gene Ther 7, 637-645

Fisher KJ, Jooss K, Alston J, Yang Y, Haecker SE, High K, Pathak R, Raper SE, Wilson JM (1 9 9 7 ) Recombinant adeno-associated virus for muscle directed gene therapy. Nat Med 3, 306-312 Flotte TR, Afione SA, Conrad C, McGrath SA, Solow R, Oka H, Zeitlin PL, Guggino WB, Carter BJ (1 9 9 3 ) Stable in vivo expression of the cystic fibrosis transmembrane conductance regulator with an adeno-associated virus vector. Proc Natl Acad Sci U S A 90, 10613-10617

Okuyama T, Li X-K, Funeshima N, Fujino M, Sasaki K, Kita Y, Kosuga M, Takahashi M, Saito H, Suzuki S, Yamada M. (1 9 9 8 ) Fas-mediated apoptosis is involved in the elimination of gene-transduced hepatocytes with E1/E3deleted adenoviral vectors. J G a s t r o e n t e r o l H e p a t o l 13, 5113-5118.

Flotte TR, Barraza-Ortiz X, Solow R, Afione SA, Carter BJ, Guggino WB (1 9 9 5 ). An improved system for packaging recombinant adeno-associated virus vectors capable of in vivo transduction. Gene Ther 2, 29-37

Peel AL, Zolotukhin S, Schrimsher GW, Muzyczka N, Reier PJ (1 9 9 7 ). Efficient transduction of green fluorescent protein in spinal cord neurons using adeno-associated virus vectors containing cell type-specific promoters. Gene Ther 4, 16-24

Hafenrichter DG, Wu X, Rettinger SD, Kennedy SC, Flye MW, Ponder KP (1 9 9 4 ) Quantitative evaluation of liverspecific promoters from retroviral vectors after in vivo transduction of hepatocytes. B l o o d 84, 3394-3404

Podsakoff G, Wong KK Jr, Chatterjee S (1 9 9 4 ). Efficient gene transfer into nondividing cells by adeno-associated virus-based vectors. J Virol 68, 5656-5666

Jaffe HA, Danel C, Longenecker G, Metzger M, Setoguchi Y, Rosenfeld MA, Gant TW, Thorgeirsson SS, StratfordPerricaudet LD, Perricaudet M, Pavirani A, Lecocq JP, Crystal RG (1 9 9 2 ) Adenovirus-mediated in vivo gene transfer and expression in normal rat liver. Nat Genet 1, 372-378

Rettinger SD, Kennedy SC, Wu X, Saylors RL, Hafenrichter DG, Flye MW, Ponder KP (1 9 9 4 ) Liver-directed gene therapy, quantitative evaluation of promoter elements by using in vivo retroviral transduction. P r o c N a t l A c a d S c i USA 91, 1460-1464

Kaplitt MG, Leone P, Samulski RJ, Xiao X, Pfaff DW, O'Malley KL, During MJ (1 9 9 4 ) Long-term gene expression and phenotypic correction using adenoassociated virus vectors in the mammalian brain. Nat Genet 8, 148-154

Shachter NS, Zhu Y, Walsh A, Breslow JL, Smith JD (1 9 9 3 ) Localization of a liver-specific enhancer in the apolipoprotein E/C-I/C-II gene locus. J L i p i d R e s 34, 1699-1707 Simonet WS, Bucay N, Lauer SJ, Taylor JM (1 9 9 3 ). A fardownstream hepatocyte-specific control region directs expression of the linked human apolipoprotein E and C-I genes in transgenic mice. J B i o l C h e m 268, 8221-8229

Kay MA, Li Q, Liu TJ, Leland F, Toman C, Finegold M, Woo SL (1 9 9 2 ) Hepatic gene therapy, persistent expression of human alpha 1-antitrypsin in mice after direct gene delivery in vivo. Hum Gene Ther 3, 641-647

73

Okuyama et al: Hepatocyte-specific gene expression using the apo E enhancer Snyder RO, Miao CH, Patijn GA, Spratt SK, Danos O, Nagy D, Gown AM, Winther B, Meuse L, Cohen LK, Thompson AR, Kay MA (1 9 9 7 ). Persistent and therapeutic concentrations of human factor IX in mice after hepatic gene transfer of recombinant AAV vectors. N a t G e n e t 16, 270-276 Soriano P, Montgomery C, Geske R, Bradley A (1 9 9 1 ). Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. C e l l 64, 693-702 Zolotukhin S, Potter M, Hauswirth WW, Guy J, Muzyczka N (1 9 9 6 ) A "humanized" green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J V i r o l 70, 4646-4654

74

Gene Therapy and Molecular Biology Vol 3, page 75 Gene Ther Mol Biol Vol 3, 75-78. August 1999.

Human cytomegalovirus (HCMV) nuclease: implications for new strategies in gene therapy Minireview

Elke Bogner Institute of Virology, Robert-Koch-Str. 17, 35037 Marburg, Germany __________________________________________________________________________________________________ Correspondence: Phone: +49-6421-285362; Fax: +49-6421-285482; E- mail: bogner@mailer.uni-marburg.de Received 9 October 1998 and in revised version 14 October; accepted 17 October 1998

I. Introduction Human cytomegalovirus (HCMV), one of eight human herpesviruses, can cause serious illness in neonates as well as in immunocompromised adults (Alford and Britt, 1993). Transplant and AIDS patients, e.g. may develop lifethreatening diseases as a consequence of primary infection or reactivation of latent infection. Additionally, HCMV infections are also associated with congenital neurological complications in approximately 7,500 newborns annually (Alford and Britt, 1993). The current drugs are toxic and cause additional complications including drug resistance. Since present therapeutical approaches are limited new strategies are needed that may result from a better understanding of viral molecular biology. The initial step of production of new virions is the packaging of newly synthesized, concatenated viral DNA into procapsids. As a consequence, blocking of this step will prevent production of viral progeny. Recently a new highly conserved gene product of ORF UL56, p130 (pUL56), was identified and partially characterized (Bogner et al., 1993). The homologous proteins of herpes simplex virus 1 (HSV-1), ICP 18.5 (UL28), and of pseudorabies virus (PrV) were reported to play an important role in DNA packaging. Viral mutants failed to cleave concatenated viral DNA which leads to an accumulation of naked nucleocapsid and uncleaved concatenated DNA in the nucleus (Addison et al., 1990; Mettenleiter et al., 1993). These reports suggested that UL56 may also play a role in virus assembly. The mechanism of DNA translocation into the procapsid and that of end formation by concatemer cutting at packaging sites (pac) are not well understood. Recently it was shown that HCMV p130 has the ability to interact with specific HCMV DNA packaging motifs and to cleave DNA bearing this motifs. Possible implications of this finding, its relation to the function of another HCMV 75

protein, pUL89, the so-called terminase, and to the bacteriophage system are discussed in this review.

II. HCMV p130 (pUL56) is a sequence specific nuclease Viral DNA-replication results in the formation of large head-to-tail DNA concatemers (Ben-Porat and Rixon, 1979), and maturation into unit-length molecules involves sitespecific cleavage at sequences (pac motifs) located within the a sequence (Spaete and Mocarski, 1985). Unit-length DNA is encapsidated in the nucleus and the DNAcontaining C-capsids bud into the cytoplasm through the nuclear membranes. The final event is the envelopment in the TGN and the release into the extracellular space. The process where newly synthesized viral DNA is cleaved and packaged into preformed capsids has long been of interest, because this is the initial step in viral assembly. The HCMV genomic a sequence is a short sequence located at both termini of the genome and in inverted orientation at the L-S junction (Mocarski et al., 1987; Tamashiro and Spector, 1986). The a sequence plays a key role in replication as a cis-acting signal for cleavage and packaging (Chou and Roizman, 1985). There is evidence that cleavage and packaging of DNA are linked processes (Ladin et al., 1980). The HCMV a sequence contains two conserved motifs, pac 1 and pac 2. These packaging motifs have an AT-rich core flanked by a GC-rich sequence. In the case of HCMV, pac sites are located on one side of the cleavage site, whereas in other herpesvirus genomes the cleavage site is between these motifs (Marks and Spector, 1988). We sugggest that during the initial step of viral packaging, the capsid-associated protein p130 may bind to the pac sequences and promote cleavage of the concatemer. Electrophoretic mobility shift assays with DNA probes

Bogner: HCMV nuclease in gene therapy spanning the region of the cis-acting pac elements demonstrated that recombinant baculovirus-infected HCMV p130 formed specific DNA-protein complexes. These data suggested that p130, as a putative cleavage and packaging protein, attaches to the pac sequence. Furthermore it is proposed that viral DNA is taken up into the capsid being scanned along the complex from the bound a sequence across the L- and S-component until an a sequence in an identical orientation is found. The final step requires nicking of both strands at signals on opposite sites of the sequence. Interestingly, by using circular plasmid DNA bearing a single a sequence as a substrate, purified baculovirus expressed HCMV p130 has an enzymatic activity that converts supercoiled plasmid DNA into open circular as well as linear molecules (Bogner et al., 1998; F i g . 1 ). This observation is comparable with the notion that HCMV p130 may be involved in cleavage of viral DNA. By using Apyrase, evidence was provided that in contrast to the adeno-associated virus origin-binding protein Rep68 (Im and Muzyczka, 1990) the reaction is independent of ATP. Interestingly, baculovirus-UL28 cell extracts, containing the HSV-1 homolog ICP18.5 (pUL28), also cleaved supercoiled plasmid DNA molecules bearing the a sequence (F i g . 1 ). It was then suggested that p130 and its herpesviral homologs are involved in cleaving of concatenated viral DNA and packaging into procapsids. It is currently under investigation, whether p130 operates with another viral protein, UL89.

III. Comparison with the bacteriophage system Herpesviruses share common features with respect to DNA maturation with dsDNA bacteriophages. DNA

F i g . 1 : Nuclease activity of the p130 homolog of HSV ICP18.5 (pUL28). Nuclease reactions were incubated for 1 h at 37°C and samples were treated with proteinase K for an additional hour at 37°C. Lane 1, plasmid pON205 in the absence of protein; 2, pON205 treated with the restriction enzyme HindIII; 3, pON205 incubated with rp130; 4, pON205 incubated with HSV UL28; 5, pON205 plus wild-type infected extracts; 6, pON205 incubated with mock-infected extracts. The arrows indicate three different plasmid DNA forms: open circular molecules (a); linear forms (b); and supercoiled molecules (c).

76

replication results in high molecular weight concatenated DNA and procapsids are assembled around a protein scaffold (Black, 1989; Murialdo and Becker, 1978). Translocation of the DNA to the procapsid bacteriophages is an ATPdependent process carried out by terminases (Feiss and Becker, 1983). Terminases are ATP-binding proteins which also bind and cleave concatenated DNA at cohesive (cos; e.g. phage ) or packaging (pac; e.g. phage P22) sites (Gold et al., 1983). In the case of bacteriophage T4, procapsids are apparently filled without sequence specificic cleavage by a ”headful mechanism” (Kalinski and Black, 1986; Streisinger et al., 1967). The HCMV UL89 gene product has some homology to the phage T4 gp17 terminase subunit. T4 terminase has two subunits. The large subunit, gp17, contains the ATP binding sites, the small subunit, gp16, is required for packaging of concatenated DNA (Rao and Black, 1988). There is preliminary evidence that baculovirus-UL89 cell extracts exhibit endonuclease activity (F i g . 2 ). HCMV UL89 is also able to nonspecifically cleave DNA in a manner reminiscent of phage T4 terminase (Black, 1986). Terminase can use either concatemeric or monomeric DNA of any sequence and prefers nonspecific ends in monomeric DNA over pac-containing concatemers (Serwer; 1986). Recently, Krosky et al. (1998) and Underwood et al. (1998) reported on a new drug, which is a derivative of benzimidazole ribonucleosides. HCMV mutants were selected by treatment of infected cultures with increasing amount of the drug. The mutations were mapped to the UL56 gene (Krosky et al., 1998) and to the gene product of pUL89 (Underwood et al., 1998). Based on the identical phenotypes of the drug resistant mutants, it is speculated that these viral proteins form the putative HCMV terminase.

Gene Therapy and Molecular Biology Vol 3, page 77

F i g . 2 : Nuclease activity of the T4 terminase homolog HCMV pUL89. Lanes: 1, pON205 alone; 2, pON205 treated with HindIII; 3, pON205 incubated with extract containing p130; 4, pON205 incubated with extracts containing HCMV UL89; 5, pON205 incubated with wild-type infected extracts; 6, pON205 incubated with mockinfected extracts. Open circular DNA molecules (a), linear (b) and supercoiled molecules (c) are indicated.

IV. Questions for the future Considering that the current drugs (ganciclovir, cidofovir and foscarnet) have limited effects and dosedependent toxicity, new antiviral therapeutics are needed. The mechanism of the current drugs is the inhibition of viral replication through an interaction with viral DNA polymerase (Erriksson et al., 1982; Ho et al., 1992; Mar et al., 1985). The inhibition of the cleavage and packaging of the viral DNA by a nuclease inhibitor may offer a potentially alternative therapy. Taken together, the reports by Addison et al. (1990), Tengelsen et al (1993) and Mettenleiter et al. (1993), demonstrating that the HSV-1 ICP18.5 (pUL28) gene product and the PRV homolog are necessary for cleavage and packaging of concatenated viral DNA, and the observation that HCMV p130 (pUL56) can interact with specific DNA packaging motifs and is able to cleave DNA bearing these motifs, provide a basis for understanding the herpesvirus DNA packaging process at the molecular level. Identification of the structure of the proteins involved is needed as a prerequisite for the development of new antivirals. Knowledge of the three dimensional protein structure is pertinent in revealing the catalytic domain for the enzymatic activity prior to anti-viral drug-design. Regarding that mammalian cell DNA replication does not involve cleavage of concatemeric DNA, drugs targeted to the viral nucleases should be safe and selective. Therefore, our findings may help to develop new nontoxic anti-HCMV 77

reagents for treatment of the immunocompromised patient population.

Acknowledgments I thank Fred Homa for kindly providing recombinant baculoviurs expressing HSV-1 UL28, Edward Mocarski for the a sequence containing plasmid pON205 and Mark Underwood for the baculovirus plasmid containing HCMV UL89. I am grateful to Klaus Radsak for critically reading the manuscript and to lab members for helpful comments. This work was supported by Deutsche Forschungsgesellschaft, Sonderforschungsbereich 286, Teilprojekt A3.

References Addison, C., F.J. Rixon and V.G. Preston ( 1 9 9 0 ) Herpes simplex virus type 1 UL28 gene product is important for the formation of mature capsids. J G e n V i r o l 71, 23772384. Alford, C.A. and W.J. Britt ( 1 9 9 3 ) Cytomegalovirus, p 227255. IN B.Roizman, RlJ. Whiteley and C. Lopez et al. (ed.), T h e h u m a n h e r p e s v i r u s e s . Raven Press, Ltd., New York. Ben-Porat, T. and F.J.Rixon ( 1 9 7 9 ) Replication of herpesvirus DNA. IV: Analysis of concatemers. V i r o l o g y 94, 61-70. Black, L.W. ( 1 9 8 6 ) In vitro packaging of bacteriophage T4

Bogner: HCMV nuclease in gene therapy DNA. V i r o l o g y 113, 336-344. Black, L.W. ( 1 9 8 8 ) DNA packaging in dsDNA bacteriophages. In: The B a c t e r i o p h a g e s , ed. R. Calendar, 2, 321-273. New York: Plenum. Black, L.W. ( 1 9 8 9 ) DNA packaging in dsDNA bacteriophages. A n n u R e v M i c r o b i o l 43, 267-292. Bogner, E., M. Reschke, B. Reis, T. Mockenhaupt and K. Radsak ( 1 9 9 3 ) Identification of the gene product encoded by ORF UL56 of human cytomegalovirus genome. V i r o l o g y 196, 290-293. Bogner, E., K. Radsak and M.F. Stinski ( 1 9 9 8 ) The gene product of human cytomegalovirus open reading frame UL56 binds the pac motif and has specific nuclease activity. J Virol 72 , 2259-2264. Chou, J. and B. Roizman ( 1 9 8 5 ) The isomerization of the herpes simplex virus 1 genome: identification of the cisacting and recombination sites within the domain of the a sequence. C e l l 41, 803-811. Erriksson, D., B. Oberg and B. Wahren ( 1 9 8 2 ) Pyrohphosphate analogs as inhibitors of DNA polymerases of cytomegalovirus, herpes simplex virus and cellular origin. B i o c h i m B i o p h y s A c t a 669, 115-123. Feiss, M. and A. Becker ( 1 9 8 3 ) DNA packaging and cutting, p 305-330. IN R.W. Hendrix, J.W. Roberts, F.W. Stahl and R.A. Weisberg (ed.), Lambda II. Cold Spring Harbor, New York. Gold, M. and A. Becker ( 1 9 8 3 ) The baceriophage ! terminase: partial purification and preliminary characterization of properties. J B i o l C h e m 258, 14619-14625. Ho, H.-T., K.L. Woods, J.J. Bronson, H. DeBoeck, J.C. Martin and M.J.M. Hitchcock ( 1 9 9 2 ) Intracellular metabolism of the antiherpes agent (S)-1-[3-hydroxy-2(phosphonylmethoxy)propyl] cytosine. Mol Pharmacol 41, 197-202. Kalinski, A. and L.W. Black ( 1 9 8 6 ) End structure and mechanism of packaging of bacteriophage T4 DNA. J V i r o l 58, 951-954. Krosky, P.M., M.R. Underwood, S.R. Turk, K. W.-H. Feng, R.K. Jain, R.G. Ptak, A.C. Westerman, K.K. Biron, L.B. Townsend and J.C. Drach (1 9 9 8 ) Resistance of human cytomegalovirus to benzimidazole ribonucleosides maps to two open reading frames: UL89 and UL56. J V i r o l 72, 4721-4728. Ladin, B.F., M.L. Blankenship and T. Ben-Porat ( 1 9 8 0 ) Replication of herpesviurs DNA. V. The maturation of concatemeric DNA of pseudorabies virus to genome length is related to capsid formation. J Virol 33, 1151-1164. Mar, E., J. Chiou, Y. Cheng and E. Huang ( 1 9 8 5 ) Inhibition

78

of cellular DNA polymerase alpha and human cytomegalovirus-induced DNA polymerase by triphosphates of 9-(2-hydroxymethyl)guanine and 9-(1,3dihydroxy-2-propoxymethyl)guanine. J V i r o l 53, 776780. Marks, J.R. and D.H. Spector ( 1 9 8 8 ) Replication of the murine cytomegalovirus genome: structure and role of the termini the generation and cleavage of concatemers. V i r o l o g y 162, 98-107. Mettenleiter, T.C., A. Saalm端ller and F.Weiland ( 1 9 9 3 ) Pseudorabies virus protein homologous to herpes simplex virus type1 ICP 18.5 is necessary for capsid maturation. J V i r o l 67, 1236-1245. Mocarski, E.S., A.C. Liu and R.R. Spaete ( 1 9 8 7 ) Structure and variability of the a sequence in the genome of human cytomegalovirus (Towne strain). J G e n V i r o l 68, 22232230. Murialdo, H. and A. Becker ( 1 9 7 8 ) Head morphogenesis of complex double-stranded deoxyribonucleic acid bacteriophages. M i c r o b i o l R e v 42, 529-576. Rao, V.B. and L.W. Black ( 1 9 8 8 ) Cloning, overexpression and purification of the terminase proteins gp16 and gp17 of bacteriophage T4: construction of a defined in vitro DNA packaging system using purified terminase proteins. J M o l B i o l 200, 475-488. Serwer, P. ( 1 9 8 6 ) Arrangement of double-stranded DNA packaged in bacteriophage capsids. J M o l B i o l 190, 509-512. Spaete, R.R. and E.S. Mocarski ( 1 9 8 5 ) The a sequence of the cytomegalovirus genome functions as a cleavage/packagung signal for herpes simplex virus defective genomes. J Virol 54, 817-824. Streisinger, G., J. Emrich and M.M. Stahl ( 1 9 6 7 ) Chromosome structure in phage T4. III. Terminal redundancy and lenght determination. Proc Natl Acad Sci USA 57, 292-295. Tamashiro, J.C. and D.H. Spector ( 1 9 8 6 ) Terminal structure and heterogeneity in human cytomegalovirus strain AD 169. J V i r o l 59, 591-604. Tengelsen LA, Pederson NE, Shaver PR, Wathen MW, Homa FL (1 9 9 3 ) Herpes simplex virus type 1 DNA cleavage and encapsidation require the product of the UL28 gene: isolation and characterization of two UL28 deletion mutants. J Virol 67, 3470-3480. Underwood, M.R., R.J. Harvey, S.C. Stanat, M.L. Hemphill, T. Miller, J.C. Drach, L.B. Townsend and K.K. Biron ( 1 9 9 8 ) Inhibition of human cytomegalovirus DNA maturation by a benzimidazole ribonucleoside is mediated through the UL89 gene product. J Virol 72, 717-725.

Gene Therapy and Molecular Biology Vol 3, page 79 Gene Ther Mol Biol Vol 3, 79-89. August 1999.

Application of recombinant Herpes Simplex Virus-1 (HSV-1) for the treatment of malignancies outside the central nervous system Review Article

George Coukos1, Stephen C. Rubin1, and Katherine L Molnar-Kimber2 Division of Gynecologic Oncology, Department of Obstetrics and Gynecology1; and Thoracic Oncology Laboratory, Department of Surgery2, University of Pennsylvania Medical Center, Philadelphia, PA 19104. __________________________________________________________________________________________________ Correspondence: Katherine L. Molnar-Kimber, Ph.D., Dept. of Surgery, 351 Stemmler Hall, 36th and Hamilton Walk, University of Pennsylvania, Philadelphia, PA 19104-6070, USA. Tel. 215-662-7898; Fax 215-573-2001; E-mail: molnark@mail.med.upenn.edu Key words: gene therapy, HSV-1, cancer Received: 19 November 1998; accepted: 20 November 1998

Summary Attenuated HSV-1 mutants are promising novel vectors for human gene therapy of cancer. In addition t o t h e i r e f f i c a c y i n t r e a t m e n t o f e x p e r i m e n t a l C N S t u m o r s , H S V m u t a n t s h a v e shown promise in treatment of extra-CNS tumors including mesothelioma, melanoma, breast cancer, epithelial ovarian carcinoma, colon carcinoma and non small c e l l lung carcinoma i n various animal models. HSV mutants which have been partially attenuated can function as direct oncolytic agents capable of proliferating within three-dimensional tumors and causing tumor cell death. A major advantage of these replication-restricted HSV mutants is that they can selectively replicate in tumor cells and thus, potentially express transgenes i n a higher percentage o f the tumor c e l l s . Alternatively, superattenuated HSV mutants and amplicons can function as efficient vectors for gene therapy and have the ability to host large and multiple transgenes. A multi-pronged strategy for HSV-based anti-tumor therapy is currently emerging, where multi-attenuated viruses or the oncolytic HSV mutants are used as gene therapy vectors for intratumoral delivery of immunomodulatory or chemotherapy sensitizing transgenes. HSV-based tumor therapy has been reported to induce an anti-tumor immune response in some animal models. These findings may be due t o the combination o f co-expression of immunomodulatory molecules, immunogenic properties of the virus, necrosis of the tumor tissue and subsequent tumor antigen presentation. Thus, HSV oncolytic agents and gene therapy vectors show great potential as anti-tumor therapies. Further studies are required to test the efficacy and safety of these agents in extra-CNS malignancies.

gene therapies are designed to deliver specific suicide genes, such as herpes simplex virus thymidine kinase or cytosine deaminase, into tumor cells (Singhal and Kaiser, 1998; Vile, 1998) which are rendered sensitive to the administration of prodrugs. The suicide gene converts the prodrug into toxic metabolites which can induce lysis in rapidly dividing cells. A third strategy involves the expression of immunomodulatory genes which may stimulate an antitumor response by the host immune system. These genes

I. Introduction Therapeutic strategies for the gene therapy of malignancies have been designed along three main pathways: corrective gene therapies entail the delivery of wild-type tumor suppressor genes to tumors which have been shown to display alterations in those genes. This approach can lead to restoration of normal tumor suppressor function and to tumor regression (Favrot et al., 1998). Secondly, suicide

79

Coukos et al: HSV-based oncolytic therapy include various cytokine genes (e.g. granulocyte/macrophagecolony stimulating factor (GM-CSF), interleukin (IL)-12, co-stimulatory molecules (e.g. B7.1) and allogeneic transplantation antigens (e.g. HLA-B7) (Pardoll, 1992). Combinations of the aforementioned strategies are also being investigated (Roth and Cristiano 1997).

suicide genes. In addition, these vectors can deliver cytokine genes, or costimulatory molecules to enhance tumor recognition and killing by the immune system. In the second line of investigation, oncolytic HSV mutants have been engineered by deletion of one or more genes to replicate poorly or not at all in normal host epidermal and neuronal tissues but to be able to replicate 30100 fold more efficiently in tumor cells. For example, viruses were initially attenuated by deletion of thymidine kinase or ribonucleotide reductase and were used as oncolytic agents of CNS malignancies. Since deletion of the thymidine kinase gene made the vector insensitive to the current antiherpetic drugs, acyclovir and ganciclovir, which is an important safety mechanism in case of inappropriate HSV spread, other strategies are being pursued. Ribonucleotide reductase deletion mutants have been efficacious in the treatment of malignant gliomas in immunocompromised and immunocompetent mice (Boviatsis et al., 1994). A third generation of viruses lacking both copies of ICP34.5 demonstrated efficacy in the treatment of several CNS tumors (Andreansky et al., 1997; Chambers et al., 1995; Kesari et al., 1995; Kramm et al., 1997; Yazaki et al., 1995). The HSV ICP34.5 mutants selectively replicated in tumor cells (McKie et al., 1996; Randazzo et al., 1997) and exhibited 105-106 fold attenuation in neurovirulence (Chou et al., 1990; MacLean et al., 1991; Valyi-Nagy et al., 1994). Several strategies have been pursued to further augment the efficacy of these mutants. The efficacy for treatment of experimental human glioma by R3616, an ICP 34.5 mutant, was augmented by radiation therapy in an immunodeficient model (Advani et al., 1998) and by co-expression of IL-4 (Andreansky et al., 1998). Concomitant deletions of the ICP34.5 genes and ribonucleotide reductase (Mineta et al., 1994; Mineta et al., 1995) or the uracil DNA glycosylase gene (Pyles et al., 1997) led to further attenuation but preserved oncolytic efficacy in the treatment of various CNS tumors.

Gene delivery remains one of the most important limitations in cancer gene therapy. The first generations of replication-incompetent adenoviral vectors, widely used in clinical trials for cancer gene therapy, may have limited therapeutic efficacy in bulky tumors, most likely due to localized gene delivery in three-dimensional tumors (Sterman et al., 1998). Replication-selective viral vectors may offer a suitable alternative. Among replication-selective vectors, recombinant Herpes Simplex Virus Type-1 (HSV-1) mutants represent potentially powerful tools for the treatment of cancer. HSV has a large genome of 152 Kb (Fink and Glorioso, 1998). It may be able to accommodate more than 30 Kb of transgene inserts, making it a suitable vector for large and/or multiple transgenes (Fink and Glorioso, 1998). Although HSV-1 is a common pathogen in humans, it very rarely induces serious complications. Attenuation of HSV will most likely augment its safety profile. Recombinant viruses have been engineered to lack specific genes necessary for neurovirulence or viral replication in quiescent cells, resulting in replication-restricted viral mutants that selectively or preferentially replicate in and lyse tumor cells. Thus, depending on the degree of attenuation, HSV-1 mutants can be used not only as vectors for gene therapy but also as direct oncolytic agents. HSV-1 mutants have been shown to be efficacious in the treatment of experimental malignancies localized within the central nervous system (CNS) (Andreansky et al., 1997; Chamberset al., 1995; Jia et al., 1994; Kesari et al., 1995; Kramm et al., 1997; Mineta et al., 1995; Pyles et al., 1997; Yazaki et al., 1995). Two main lines of investigation have been followed. In the first discussed strategy, multiattenuated viral vectors were engineered by deletion of multiple genes to be able to undergo at most one or two rounds of replication within cancer cells (Glorioso et al., 1997). Alternatively, HSV amplicons, which have additional deletions of essential HSV genes and require helper virus or complementation of many HSV functions to replicate in any cells, can be used to express various transgenes (Fraefel et al., 1996; Geller, 1993; Geller and Breakefield, 1991; Ho, 1994). Multi-attenuated viral vectors and amplicons were originally engineered for gene therapy of CNS hereditary conditions, such as neurodegenerative and neuromuscular diseases, based on their ability to express the transgene(s) but not HSV proteins in quiescent cells (Fink and Glorioso, 1998; Geller, 1993; Geller and Breakefield, 1991; Glorioso et al., 1997; Ho, 1994; Huard et al., 1997). These vectors may also be suitable for cancer gene therapy for transduction of

Recent evidence suggests that attenuated Herpes Simplex Virus-1 mutants can be also utilized for peripheral malignancies. The present review will offer a brief summary of HSV-1 mechanism of action, will provide the overall rationale for the utilization of mutant HSV-1 for treatment of malignancies in extra-CNS locations and summarize the evidence accumulated to date.

II. HSV-1 replication The replication cycle and epidemiology of HSV have recently been reviewed (Roizman and Sears, 1996; Whitley, 1996). HSV-1 is a DNA virus with a large genome of 152 Kb. To date, 80 HSV genes have been identified, but approximately 30 are non-essential for its replication in vitro in permissive Vero cells (Fink and Glorioso, 1998; 80

Gene Therapy and Molecular Biology Vol 3, page 81 McGeoch et al., 1988). In the immunocompetent human host, wild-type (wt) HSV-1 infects predominantly tissues of epidermal and neuronal origin (Whitley, 1996). Wt HSV infection of epidermal tissues results in a lytic infection and usually is accompanied by the induction of latency in peripheral neurons and the ganglia. Encephalitis, a lytic infection of the central nervous system, occurs only rarely. Briefly, the replication cycle begins with viral attachment to the cells, which is mediated by recognition of specific envelope glycoproteins, such as glycoprotein (g)B and gC to heparan sulfate (Laquerre et al., 1998; Spear et al., 1992). A cellular protein, EXT, can enhance the expression of heparan sulfate and has been shown to confer susceptibility of some cells to HSV infection (McCormick et al., 1998). In addition, gD can specifically bind to cells via the Herpes virus entry mediator (HVEM) protein (Montgomery et al., 1996) and by two additional, recently identified receptors (Geraghty et al., 1998; Whitbeck et al., 1997). Binding is followed by fusion of the viral envelope with the cell membrane of the infected host, partially mediated by viral gB, gD, and gH. The capsid is transported to the nucleus, where the viral DNA is released. During this process, VP16, a protein associated with the tegument, interacts with cellular transcription factors to activate transcription and expression of immediate early (!) genes ICP0, ICP4, ICP22, ICP27 and ICP47 (DeLuca and Schaffer, 1985; Fink and Glorioso, 1998; Honess and Roizman, 1974; Roizman and Sears, 1996). The viral early genes ("1 and "2 genes), which are mainly involved in nucleotide synthesis and viral DNA replication in quiescent cells, are then transcribed and translated. The late genes (# 1 and # 2) are subsequently expressed, resulting in the synthesis of the protein components of the capsid, tegument and viral envelope (Roizman and Sears, 1996; Subak-Sharpe and Dargan, 1998). There are some genes which are transcribed late as well as early and have been termed # 1 genes (Roizman and Sears, 1996 and ref. therein). Finally, the viral DNA is cleaved and packaged into capsids, and the DNA containing capsids appear to be enveloped at the nuclear membrane. The enveloped capsids transit through the cytoplasm in a multistep process still under investigation and get released from the cell. Along this process the infected cell dies (Roizman and Sears, 1996 and ref. therein).

al., 1997a). Although Ito et al. observed no change in frequency of apoptosis in non-activated cultures of T lymphocytes infected with HSV-1 vs. non-infected cells (Ito et al., 1997b), wt HSV-1 has been reported to induce apoptosis in non-activated human peripheral blood mononuclear cells (Tropea et al., 1995) as well as in other tissues (Irie et al., 1998). The HSV genes which induce apoptosis in the infected cell are being investigated. Since HSV-1 can induce apoptosis at several checkpoints (Galvan and Roizman, 1998), it is likely that HSV-1 encodes several genes which can induce apoptosis. HSV encodes early genes that destabilize cellular RNA, disrupt cellular transcription and degrade cellular DNA (Johnson et al., 1992; Kwong et al., 1988; Roizman and Sears, 1996) and are likely candidates. Additional genes, including the genes which are non-essential for its replication in vitro (McGeoch et al., 1988) may also be involved in induction of apoptosis in the infected host. Apoptosis of the HSV-infected cells can also occur in the absence of de novo protein synthesis, suggesting that proteins present in the virion may directly trigger some apoptotic pathways (Galvan and Roizman, 1998; Koyama and Adachi, 1997). Finally, oncolytic replication-restricted HSV-1 mutants lacking ICP34.5 may induce apoptosis (Chou et al., 1994; Chou and Roizman, 1992) due to the loss of the protective effect that ICP34.5 exerts on the premature shut-off of total protein synthesis in the infected host (Cassady et al., 1998a; Cassady et al., 1998b). HSV-1 infection can also inhibit apoptosis such as that induced by cytotoxic T lymphocytes (Jerome et al., 1998), hyperthermia (Galvan and Roizman, 1998; Leopardi and Roizman 1996), sorbitol treatment (Galvan and Roizman 1998; Koyama and Miwa 1997), anti-fas ligand (Galvan and Roizman, 1998), tumor necrosis factor alpha (TNF!) and C2 ceramide (Galvan and Roizman, 1998) in some cells. Wt HSV encodes at least two genes, ICP4 (Leopardi and Roizman, 1996) and Us3 (Leopardi et al., 1997), which have been shown to protect some infected cells from undergoing apoptosis (Koyama and Miwa, 1997). In addition, as mentioned above, ICP34.5 exerts a protective effect on the premature shut-off of total protein synthesis in the infected host (Cassady et al., 1998a; Cassady et al., 1998b). Although bcl-2 expression does not play a major role in regulation of apoptosis in HSV-1 infected activated T lymphocytes in vitro (Ito et al., 1997b), it may play a role in some systems (Geiger et al., 1997). The specific mechanisms by which apoptosis is regulated in the HSVinfected cells is the subject of current investigation.

The mechanism by which HSV infected cells die is still a matter of investigation. Galvan and Roizman (Galvan and Roizman, 1998) recently indicated that some HSV-infected cells undergo apoptosis, while other cells die of nonapoptotic death. The type of cell death was found to be celltype dependent (Galvan and Roizman, 1998). Normal proliferating cells, such as activated peripheral and cord blood derived T-lymphocytes, succumb to apoptosis when infected by wt HSV-1 (Ito et al., 1997a; Ito et al., 1997b) and this process is independent of the Fas/Fas ligand system (Ito et

III. HSV-1 mutants used as vectors for cancer gene therapy HSV-1 vectors have been engineered following two different strategies. Recombinant viral vectors are derived 81

Coukos et al: HSV-based oncolytic therapy directly from wtHSV-1, and contain deletion or insertional mutations in various genes. Many investigators have taken the approach of producing HSV mutants with multiple gene deletions, as a means to increase the insertion capacity of the vector and thus be able to host multiple transgenes (Fink and Glorioso, 1998; Johnson et al., 1994). For example, HSV mutants have been engineered with multiple mutations or deletions in genes which include ICP4, ICP27, ICP8, UL33, UL42 and gB and gH to attenuate viral replication (Breakefield and DeLuca, 1991; Glorioso et al., 1997). For example, HSV mutants with various combinations of deletions of ICP4, ICP22, ICP27 and ICP42 yield viral mutants with minimal cytotoxicity, due to their inability to replicate in normal cells (Huard et al., 1997; Johnson et al., 1992). Nevertheless, these vectors have been shown to achieve expression of transgenes in normal cells, in which the transgene is expressed with minimal expression of HSV genes. Recombinant multi-attenuated vectors have been utilized in experimental cancer gene therapy, and their use for suicide or immune gene therapy of extra-CNS malignancies is recently gaining interest (Glorioso et al., 1995). A multiattenuated HSV vector with alterations in ICP4, ICP22, ICP27 and ICP41 was utilized to transduce several ovarian cancer cell lines with the suicide gene HSV thymidine kinase, and was found to achieve high transduction efficiency (Wang et al., 1998). Further studies are needed to determine whether sufficient cells can be transduced to yield a clinical benefit. Rees et al. (1998) constructed a mutated HSV vector that could undergo a single round of viral replication and express murine granulocyte colony stimulating factor (mGCSF). This vector exhibited efficient transduction and achieved effective immunization in a murine syngeneic renal carcinoma model (Rees et al., 1998).

amplicon vector carrying IL-2 was found to achieve high therapeutic efficacy in treating intraperitoneal metastatic gastric carcinoma in nude mice and to increase the killing activity of splenocytes (Tsuburaya et al., 1998). Furthermore, subcutaneous murine lymphoma nodules were eradicated in approximately 85% of tumor-bearing mice by co-administration of HSV amplicon vectors expressing the chemokine RANTES and the T cell costimulatory ligand B7.1 (Kutubuddin et al., 1998).

IV. HSV-1 mutants used as direct oncolytic agents Molecular alterations in certain genes of the HSV genome have led to the engineering of replication-restricted HSV mutants, which maintain the ability to infect and rapidly kill proliferating cancer cells but still maintain low (or undetectable) replication rates in normal diploid cells. Several genes have been the target of alterations including the thymidine kinase (UL23) (Jia et al., 1994; Martuza et al., 1991; Sanders et al., 1982), the ICP6 gene (UL39) encoding the large subunit of HSV ribonucleotide reductase (RR) (Boviatsis et al., 1994; Idowu et al., 1992; Kramm et al., 1997), the uracil DNA glycosylase (UNG) gene (Pyles et al., 1997) and the ICP34.5 (Chambers et al., 1995; Kesari et al., 1995; Mineta et al., 1995). The thymidine kinase-negative HSV-1 mutant (Jia et al., 1994; Martuza et al., 1991) was shown to efficiently cause tumor growth inhibition after intraneoplastic inoculation of subcutaneously and subrenally implanted experimental human gliomas with minimal toxicity in immunodeficient mice. It may also be effective for treatment of other solid tumors localized in the periphery. Although HSVtk - mutants were sensitive to foscarnet and phosphonormal acid (Jia et al., 1994), a potential disadvantage of these strains relates to their resistance to commonly used anti-herpetic drugs such as acyclovir or ganciclovir and has spurred the engineering of alternate attenuated HSV vectors. HSV mutants lacking the ribonucleotide reductase through a deletion or mutations of ICP6 gene were also shown to be replication-restricted and demonstrated efficacy in CNS malignancies. The HSV-1 ribonucleotide reductase deficient (RR -) mutant hrR3, containing an E-coli LacZ gene insertion in the ICP6 gene, was recently tested in an experimental metastatic colon carcinoma with liver metastases in an immunodeficient mouse model (Carroll et al., 1996). This mutant displayed selectivity only for the intrahepatic tumors in vivo and did not spread to the surrounding normal liver after intrasplenic injection, supporting the notion that it replicated only in dividing cells, which provided RR in complementation (Carroll et al., 1996). HSV oncolytic agents have also been generated by mutations or deletions of the ICP34.5 genes, altering both copies in the HSV genome (Chambers et al.,

A second type of multi-attenuated vectors, the amplicon vectors, are engineered utilizing plasmids carrying the HSV DNA packaging signal, the HSV origin of DNA replication, expression cassettes regulating the transgenes of interest together with an E-coli origin of DNA replication and antibiotic resistance genes (Frenkel et al., 1994; Geller, 1993; Geller and Breakefield, 1991; Ho, 1994). Although propagation of amplicon vectors initially required coinfection with HSV helper virus (Frenkel et al., 1994; Geller, 1993; Geller and Breakefield, 1991; Ho, 1994), amplicons can now be propagated by complementation using plasmids (Fraefel et al., 1996). Amplicon HSV vectors have been utilized to rapidly transduce hepatoma cells from cultured cells or tissue explants with IL-2 or GM-CSF genes (Karpoff et al., 1997; Tung et al., 1996). Administration of these transduced cells into rats or mice, respectively induced an immune response to the hepatomas. Toda et al. (1998a) showed that co-expression of IL-12 by an HSV amplicon in the presence of an oncolytic G207 helper virus augmented the anti-tumor effect. Preliminary data indicated that an HSV

82

Gene Therapy and Molecular Biology Vol 3, page 83 1995; MacLean et al., 1991). Its protein product is implicated in the prevention of the protein synthesis premature shut-off in the infected host, through inhibition of the phosphorylation of the eukaryotic translation initiation factor eIF2! ( Cassady et al., 1998b), as well as in viral exit from the cell (Brownet al., 1994). ICP34.5 -/- mutants have proven efficient in treating several types of CNS malignancies in experimental rodent models (Andreansky et al., 1997; Chambers et al., 1995; Kesari et al., 1995; Kramm et al., 1997) and efficiently treat experimental tumors of melanoma (Randazzo et al., 1997) and mesothelioma origin (Kucharczuk et al., 1997). HSV-1716 was efficacious in the treatment of intraperitoneal (i.p.) human malignant mesothelioma in a severe combined immunodeficient (SCID) mouse model (Kucharczuk et al., 1997), reducing tumor burden and prolonging animal survival in a dose-dependent manner. Administration of the HSV-1716 yielded viral replication only within i.p. tumor nodules. There was no evidence of viral antigen (by immunohistochemistry) or DNA (by polymerase chain reaction analysis) in any mouse organs. The same virus was also used to treat experimental subcutaneous melanoma, yielding similar efficiency and minimal toxicity (Randazzo et al., 1997). Since mRNA for HVEM was readily detected in lung tissue (Montgomery et al., 1996), HSV mutants lacking ICP34.5 were investigated and demonstrated efficacy in vitro and in vivo against several human lung carcinoma lines (Abbas et al., 1998).

and did not spread to normal murine CNS but exerted a direct oncolytic activity in vitro and in vivo against human CNS tumor cell lines and brain tumor xenografts. Moreover, this mutant demonstrated a hypersensitivity to the anti-herpetic drug ganciclovir. G207 is also a derivative of the ICP34.5deleted mutant, R3616, in which "-galactosidase is inserted into ICP6 gene, which encodes the large subunit of the ribonucleotide reductase gene (Mineta et al., 1994). This mutant was also found to be efficacious in the treatment of various CNS tumors (Mineta et al., 1994; Mineta et al., 1995; Yazaki et al., 1995). Both these doubly deleted HSV mutants appear promising for extra-CNS applications. G207 demonstrated efficacy against some tumor cell lines of breast origin both in vitro and in vivo (Toda et al., 1998b). In our laboratory, a single i.p. administration of HSV-G207 to SCID mice bearing i.p. human ovarian carcinoma tumors (SKOV3 cell line) led to significant reduction in tumor volume four weeks later (Table 1). Immunostaining of tumors harvested from HSV-treated animals demonstrated the presence of HSV-1 antigens in multiple scattered areas throughout the tumor nodules, demonstrating the ability of the virus to replicate and penetrate in depth within the tumors (not shown). Extensive necrosis was observed adjacent to the areas that were positive for HSV particles. An emerging strategy for engineering replication selective HSV oncolytic agents involves replication-targeted HSV mutants, achieved through the insertion of tissue-specific promoters regulating HSV replication. To demonstrate the feasibility of this system, an expression cassette containing a heterologous eukaryotic promoter (albumin) regulating ICP4 expression was inserted into an ICP4- mutant (Miyatake et al., 1997). The authors observed that these viruses replicated 10-fold better in albumin-expressing hepatomas than in cells which did not express albumin.

A second generation of multi-attenuated viruses were engineered stemming from a parental ICP34.5-deleted virus, R3616, which is based on the wt HSV-F strain (Chambers et al., 1995). R3616UB was generated by interrupting the uracil DNA glycosylase (UNG) gene in the parental HSV-R3616 mutant (Pyles et al., 1997). This viral strain did not show any replication in primary human neuronal cultures in vitro

Tumor Weight

Pre-treatment

Control (Media)

HSV-G207

12.5±4 mg

278±45 mg

48±7 mg *

T a b l e 1 . To assess the efficacy of HSV-G207 in treating epithelial ovarian cancer in vivo, SCID mice (n=10/group) were administered a single intraperitoneal (i.p.) injection of 5x10 6 SKOV3 cells, which led to the establishment of i.p. tumors two weeks later. HSV-G207 was administered directly i.p. to a group of animals at that time. Control animals received media only. Animals from each group were sacrificed four weeks following treatment. A separate group of animals was sacrificed prior to viral administration at two weeks. Tumors were dissected and weighed. Weights are expressed in mg and values are expressed as the mean ± standard error (M±SE). (* =p<0.001 vs. control animals).

the immune response on the efficacy of HSV-based oncolytic or gene therapy in humans is an important issue. To address this issue, the effects of a pre-existing immunity to HSV-1 was tested in a syngeneic rat model. The presence of antiHSV primed immune response was found to dampen but not

V. HSV mutants used in the immune therapy of cancer Since HSV-1 and HSV-2 infections are highly prevalent in the adult human population (Whitley, 1996), the effects of

83

Coukos et al: HSV-based oncolytic therapy abolish gene transfer by an HSV vector (Herrlinger et al., 1998). However, it should be noted that the clinical significance of pre-existing immunity is still unknown in viral-based oncolytic or gene therapy. In fact, HSV-1 or HSV-2 recurrences occur commonly following a primary infection in the immunocompetent human (Whitley, 1996). Moreover, adenoviral-mediated gene transfer in a phase-1 clinical trial for the treatment of malignant mesothelioma was not blocked by significant anti-adenoviral neutralizing antibody titers or significant T cell proliferation (MolnarKimber et al., 1998). Thus, the effect of the immune response on the efficacy of viral therapies will have to be determined in clinical studies.

mutants represent suitable vectors for immunotherapy as they can accommodate large and multiple transgene inserts and efficiently deliver interleukin transgenes into tumors. The administration of a defective HSV vector containing tandem repeats of an amplicon plasmid encoding IL-12 together with a multi-attenuated HSV-1 mutant lacking ICP34.5 and RR (HSV G207) was followed by significant reduction in tumor growth in a syngeneic murine colon carcinoma model (Toda et al., 1998a). Importantly, IL-12 was expressed and secreted by infected tumor cells in vitro and in vivo. Unilateral inoculation of the virus and amplicon was accompanied by regression not only of the inoculated tumor but also of non-inoculated controlateral tumors. In addition, tumor reduction was significantly greater in animals receiving the amplicon plasmid encoding IL-12 compared to those receiving a control LacZ -expressing amplicon plasmid together with the HSV G207 helper. This effect was attributed to the enhancement of tumor-specific CTL activity (Toda et al., 1998a). Moreover, a replication-restricted HSV ICP34.5 -/- mutant encoding murine IL-4, but not IL-10, was shown to significantly prolong the survival of gliomabearing mice (Andreansky et al., 1998). Clearly, similar viruses encoding cytokines or immunostimulatory molecules appear very attractive for the treatment of non-CNS tumors as well. Additional support for the potential of HSV-based cytokine-mediated immunotherapy is provided by the observations that amplicons expressing RANTES and B7.1 (Kutubuddin et al., 1998) or IL-2 (Tsuburaya et al., 1998) or a multi-attenuated HSV vector expressing GM-CSF (Rees et al., 1998) were showed to augment the efficacy of treatment of lymphoma, metastatic gastric carcinoma or renal carcinoma, respectively, as mentioned above.

The interaction of the immune system with HSV-based therapeutic agents could potentially become advantageous. In fact, the utilization of HSV mutants as direct oncolytic agents or as vectors could generate or enhance an anti-tumor immune response. Infection of human cells by wild-type HSV induces an orchestrated immune response, which includes a cellular infiltrate, generation of cytotoxic T lymphocytes (CTL), release of cytokines and induction of an antibody response (Whitley, 1996) and ref. therein). Although ICP47 can decrease the expression of class I major histocompatibility antigens on the cell surface (York et al., 1994), tumor cell infection and death following infection by mutant HSV-1 will most likely induce intratumoral infiltration of lymphocytes and antigen-presenting cells and may lead to unmasking of tumor antigens, triggering an antitumor response. This strategy could become particularly advantageous in tumors that down-regulate the immune response or induce a predominant TH2-like response. Recent experimental evidence supports the concept that HSV-based oncolytic therapy may be followed by an adjuvant tumorspecific immune response (Toda et al., 1998a). In fact, intratumoral administration of HSV-G207 in immunocompetent animals bearing syngeneic tumors led to growth inhibition of distant non-inoculated tumors, likely mediated by an immune response (Toda et al., 1998a).

VI. Toxicity considerations Large amount of pre-clinical data has been accumulated in the rodent model on replication-selective attenuated HSV-1 ICP34.5 -/- mutants following intratumoral “stereotactic� inoculations of viral particles within the CNS. In both immunocompetent as well as immunodeficient mice, intracranial administration of viral particles did not lead to encephalitis (Andreansky et al., 1997; Carroll et al., 1996; Chambers et al., 1995; Kaplitt et al., 1994; Kesari et al., 1995; Mineta et al., 1994). HSV-1716 administered intracranially or intraocularly into SCID mice resulted in low or no virulence (Valyi-Nagy et al., 1994). Similarly, HSV1716 administered i.p. was found to be avirulent in SCID mice in contrast to rapid systemic spread of the wt HSV virus and death of the animals (Kucharczuk et al., 1997). No viral spread was detected beyond the tumor tissue (Kucharczuk et al., 1997). Administration of HSV-1716 to normal human skin in a murine xenograft model was accompanied by no toxicity, while administration of a wild-

Cytokines have been shown to enhance the anti-tumor immune response, but their systemic administration has been accompanied by significant side effects. Local administration of cytokines to tumors has led to decreased magnitude of side effects but may be technically challenging (Pardoll 1996). Recent evidence suggests that gene therapy with delivery of cytokine genes into tumors or the generation of cytokine gene-transduced cancer cell vaccines may represent a very powerful tool for augmenting anti-tumor immune responses (Pardoll, 1996). For instance, expression of interferon gamma (INF# ), tumor necrosis factor alpha (TNF!) or GMCSF in the milieu of the tumor has led to arrest of tumor growth in experimental models in vivo through stimulation of local inflammatory and immune responses (Andreansky et al., 1998; Pardoll, 1996; Tepper and Mule, 1994). HSV-1 84

Gene Therapy and Molecular Biology Vol 3, page 85

Acknowledgements

type HSV-1 led to rapid destruction of the xenograft (Randazzo et al., 1996). The replication selective hR3, a HSV-1 lacking RR expression, administered systemically (intrasplenic injections) was also found to infect only metastatic human colonic adenocarcinoma tumor nodules within the liver but not the surrounding murine normal liver tissue (Carroll et al., 1996). However, HSV-1 ICP34.5deleted mutants maintain their ability to infect ependymal cells in the CNS (Kesari et al., 1998; Markovitz et al., 1997). Severe intra-CNS inflammation was observed in some rodent strains after intracranial administration of HSV1716 which expressed LacZ (McMenamin et al., 1998). It is possible that additional mutations may significantly decrease any potential for HSV-1 neurotoxicity. Administration of G207, an ICP34.5-deleted/RR- mutated virus was found to be safe following administration to HSVsensitive primates (Markert et al., 1998). Sufficient toxicity data on the HSV ICP34.5 mutants, HSV-1716, and G207 (ICP34.5 -/- ,RR - ), has been presented to the regulatory bodies for initiation of phase I clinical trials. Preliminary results from the dose escalation phase I clinical trials employing HSV-1716 (ICP34.5-/- ) or G207 (ICP34.5-/- , RR -) utilizing intra-CNS administration for the treatment of malignant glioma have reported minimal side effects in humans (Brown et al., 1998; Markert et al., 1998).

We thank Ms. Carmen Lord for her editorial help in the preparation of this manuscript.

References Abbas, A., Caparrelli, D., Kang, E., Toyozumi, T., Albelda, S., Kaiser, L., Molnar-Kimber, K. (1 9 9 8 ). Replicationselective HSV-1 mutants are potential oncolytic agents for lung cancer. In AACR Proceedings; pp A3771. Advani, S., Sibley, G., Song, P., Hallahan, D., Kataoka, Y., Roizman, B., Weichselbaum, R. (1 9 9 8 ). Enhancement of replication of genetically engineered herpes simplex viruses by ionizing radiation, a new paradigm for destruction of therapeutically intractable tumors. Gene Ther. 5, 160165. Andreansky, S., He, B., van Cott, J., McGhee, J., Markert, J. M., Gillespie, G. Y., Roizman, B., Whitley, R. J. (1 9 9 8 ). Treatment of intracranial gliomas in immunocompetent mice using herpes simplex viruses that express murine interleukins. Gene Ther. 5, 121-130. Andreansky, S., Soroceanu, L., Flotte, E., Chou, J., Markert, J., Gillespie, G., Roizman, B., Whitley, R. (1 9 9 7 ). Evaluation of genetically engineered herpes simplex viruses as oncolytic agents for human malignant brain tumors. Cancer Res. 57, 1502-9. Boviatsis, E., Scharf, J., Chase, M., Harrington, K., Kowall, N., Breakefield, X., Chiocca, E. (1 9 9 4 ). Antitumor activity and reporter gene transfer into rat brain neoplasms inoculated with herpes simplex virus vectors defective in thymidine kinase or ribonucleotide reductase. Gene Ther. 1, 323331.

VII. Conclusions Attenuated HSV-1 mutants may represent an emerging powerful tool in human gene cancer therapy. HSV mutants are versatile in that, when partially attenuated, they can function as direct oncolytic agents capable of proliferating within three-dimensional tumors and causing tumor cell death. The advantage of replication-restricted HSV mutants is that they can selectively replicate in tumor cells and thus, potentially express transgenes in a higher percentage of the tumor cells. Alternatively, when super-attenuated or amplicons, they can function as efficient vectors for gene therapy. In that capacity these vectors have the potential to host large and multiple transgenes. A multi-pronged strategy for HSV-based anti-tumor therapy is currently emerging, where multi-attenuated viruses or the oncolytic HSV mutants are used as gene therapy vectors for intratumoral delivery of immunomodulatory or chemotherapy sensitizing transgenes. Based on experimental evidence, HSV-based tumor therapy may induce an anti-tumor “vaccine� effect. This may be due to the immunogenic properties of the virus, as well as to the tumor tissue necrosis. Thus, HSV oncolytic agents and gene therapy vectors show great potential as anti-tumor therapies. Further clinical studies are required to test the clinical efficacy and safety of these agents in extra-CNS malignancies.

Breakefield, X., DeLuca, N. (1 9 9 1 ). Herpes simplex virus for gene delivery to neurons. N e w . B i o l . 3, 203-218. Brown, S., MacLean, A., Aitken, J., Harland, J. (1 9 9 4 ). ICP34.5 influences herpes simplex virus type I maturation and egress from infected cells in vitro. J . G e n . V i r o l . 75, 3767-3686. Brown, S., Rampling, R., Cruikshank, G., McKie, E., MacLean, A., Harland, J., Mabbs, R. (1 9 9 8 ). A phase 1 dose escalation trial of intratumoral injection with ICP34.5-ve HSV1 into recurrent malignant glioma. In 23rd International Herpesvirus Workshop; York, UK pp A386. Carroll, N., Chiocca, E., Takahashi, K., Tanabe, K. (1 9 9 6 ). Enhancement of gene therapy. specificity for diffuse colon carcinoma liver metastases with recombinant herpes simplex virus. Annals Surg 224, 323-329. Cassady, K., Gross, M., Roizman, B. (1 9 9 8 a ). The Herpes Simplex Virus Us11 Protein effectively compensates for the gamma 1 34.5 gene if present before activation of protein kinase R by precluding its phosphorylation and that of the alpha subunit of eukaryotic translation initiation factor 2. J . V i r o l . 72, 8620-8626.

85

Coukos et al: HSV-based oncolytic therapy Cassady, K., Gross, M., Roizman, B. (1 9 9 8 b ). The second-site mutation in the herpes simplex virus recombinants lacking the gamma134.5 genes precludes shutoff of protein synthesis by blocking the phosphorylation of eIF-2alpha. J . V i r o l . 72, 7005-7011.

Geller, A., Breakefield, X. (1 9 9 1 ). A defective HSV-1 vector expresses Escherichia coli beta galactosidase in cultured peripheral neurons. S c i e n c e 241, 1667-1669. Geraghty, R., Krummenacher, C., Cohen, G., Eisenberg, R., Spear, P. (1 9 9 8 ). Entry of alphaherpesviruses mediated by poliovirus receptor-related protein 1 and poliovirus receptor. S c i e n c e 280, 1618-1620.

Chambers, R., Gillespie, G. Y., Soroceanu, L., Andreansky, S., Chatterjee, S., Chou, J., Roizman, B., Whitley, R. J. (1 9 9 5 ). Comparison of genetically engineered herpes simplex viruses for the treatment of brain tumors in a scid mouse model of human malignant glioma. P r o c . N a t l . A c a d . S c i . , U S A 92, 1411-1415.

Glorioso, J., Bender, M., Fink, D., DeLuca, N. (1 9 9 5 ). Herpes simplex virus vectors. M o l . C e l l B i o l . H u m . D i s . S e r . 5, 33-63. Glorioso, J., Goins, W., Schmidt, M., Oligino, T., Krisky, D., Marconi, P., Cavalcoli, J., Ramakrishnan, R., Poliani, P., Fink, D. (1 9 9 7 ). Engineering herpes simplex virus vectors for human gene therapy. Adv Pharmacol 40, 103-136.

Chou, J., Kern, E., Whitley, R., Roizman, B. (1 9 9 0 ). Mapping of Herpes Simplex Virus-1 Neurovirulence to g1 34.5, a gene nonessential for growth in culture. S c i e n c e 250, 12621265.

Herrlinger, U., Kramm, C., Aboody-Guterman, K., Silver, J., Ikeda, K., Johnston, K., Pechan, P., Barth, R., Finkelstein, D., Chiocca, E., Louis, D., Breakefield, X. (1 9 9 8 ). Preexisting herpes simplex virus 1 (HSV-1) immunity decreases, but does not abolish, gene transfer to experimental brain tumors by a HSV-1 vector. Gene Ther. 5, 809-819.

Chou, J., Poon, A., Johnson, J., Roizman, B. (1 9 9 4 ). Differential response of human cells to deletions and stop codons in the gamma 34.5 gene of herpes simplex virus. J V i r o l 68, 8304-8311. Chou, J., Roizman, B. (1 9 9 2 ). The gamma 34.5 gene of herpes simplex virus 1 precludes neuroblastoma cells from triggering total shut off of protein synthesis characteristics of programmed cell death. P r o c . N a t l . A c a d . S c i . U S A 89, 3266-3270.

Ho, D. (1 9 9 4 ). Amplicon-based herpes simplex virus vectors. M e t h o d s C e l l B i o l . 43 PtA, 191-210. Honess, R., Roizman, B. (1 9 7 4 ). Regulation of herpesvirus macromolecular synthesis. I. Cascade regulation of the synthesis of three groups of viral proteins. J . V i r o l . 14, 8-19.

DeLuca, N., Schaffer, P. (1 9 8 5 ). Activation of immediate-early, early, and late promoters by temperature-sensitive and wildtype forms of herpes simplex virus type 1 protein ICP4. M o l C e l l B i o l 5, 1997-2008.

Huard, J., Krisky, D., Oligino, T., Marconi, P., Day, C., Watkins, S., Glorioso, J. (1 9 9 7 ). Gene transfer to muscle using herpes simplex virus-based vectors. Neuromuscul. D i s o r d . 7, 299-313.

Favrot, M., Coll, J., Louis, N., Negoescu, A. (1 9 9 8 ). Cell death and cancer, replacement of apoptotic genes and inactivation of death suppressor genes in therapy. Gene Ther. 5, 728739.

Idowu, A., Fraser-Smith, E., Poffenberger, K., Herman, R. (1 9 9 2 ). Deletion of the herpes simplex virus type 1 ribonucleotide reductase gene alters virulence and latency in vivo. A n t i v i r a l R e s 17, 145-156.

Fink, D., Glorioso, J. (1 9 9 8 ). Engineering herpes simplex virus vectors for gene transfer to neurons. Nature Med. 3, 357-359. Fraefel, C., Song, S., Lim, F., Lang, P., Yu, L., Wang, Y., Wild, P., Geller, A. (1 9 9 6 ). Helper virus-free transfer of herpes simplex virus type 1 plasmid vectors into neural cells. J V i r o l . 70, 7190-7197.

Irie, H., Koyama, H., Kubo, H., Fukuda, A., Aita, K., Koike, T., Yoshimura, A., Yoshida, T., Shiga, J., Hill, T. (1 9 9 8 ). Herpes simplex virus hepatitis in macrophage-depleted mice, the role of massive, apoptotic cell death in pathogenesis. J . G e n . V i r o l . 79, 1225-1231.

Frenkel, N., Singer, O., Kwong, A. (1 9 9 4 ). Minireview, the herpes simplex virus amplicon--a versatile defective virus vector. Gene. Ther. 1, S40-46.

Ito, M., Koide, W., Watanabe, M., Kamiya, H., Sakurai, M. (1 9 9 7 a ). Apoptosis of cord blood T lymphocytes by herpes simplex virus type 1. J . G e n . V i r o l . 78, 1971-5.

Galvan, V., Roizman, B. (1 9 9 8 ). Herpes simplex virus 1 induces and blocks apoptosis at multiple steps during infection and protects cells from exogenous inducers in a cell-type-dependent manner. P r o c . N a t l . Acad. S c i . USA 95, 3931-6.

Ito, M., Watanabe, M., Kamiya, H., Sakurai, M. (1 9 9 7 b ). Herpes simplex virus type 1 induces apoptosis in peripheral blood T lymphocytes. J . I n f e c t . D i s . 175, 1220-1224. Jerome, K., Tait, J., Koelle, D., Corey, L. (1 9 9 8 ). Herpes simplex virus type 1 renders infected cells resistant to cytotoxic T-lymphocyte-induced apoptosis. J . V i r o l . 72, 436-441.

Geiger, K., Nash, T., Sawyer, S., Krahl, T., Patstone, G., Reed, J., Krajewski, S., Dalton, D., Buchmeier, M., Sarvetnick, N. (1 9 9 7 ). Interferon-gamma protects against herpes simplex virus type 1-mediated neuronal death. V i r o l . 238, 189-197.

Jia, W. W.-G., McDermott, M., Goldie, J., Cynader, M., Tan, J., Tufaro, F. (1 9 9 4 ). Selective Destruction of gliomas in immunocompetent rats by thymidine kinase defective herpes

Geller, A. (1 9 9 3 ). Herpesviruses, expression of genes in postmitotic brain cells. C u r r O p i n . G e n e t . D e v . 3, 8185.

86

Gene Therapy and Molecular Biology Vol 3, page 87 simplex virus type 1. J . N a t l . C a n c e r I n s t 86, 12091215.

Laquerre, S., Argnani, R., Anderson, D., Zucchini, S., Manservigi, R., Glorioso, J. (1 9 9 8 ). Heparan sulfate proteoglycan binding by herpes simplex virus type 1 glycoproteins B and C, which differ in their contributions to virus attachment, penetration, and cell-to-cell spread. V i r o l . 72, 6119-30.

Johnson, P., Miyanohara, A., Levine, F., Cahill, T., Friedmann, T. (1 9 9 2 ). Cytotoxicity of a replication-defective mutant of herpes simplex virus type 1. J . V i r o l . 66, 2952-2965. Johnson, P., Wang, M., Friedmann, T. (1 9 9 4 ). Improved cell survival by the reduction of immediate-early gene expression in replication-defective mutants of herpes simplex virus type 1 but not by mutation of the virion host shutoff function. J . V i r o l . 68, 6347-6362.

Leopardi, R., Roizman, B. (1 9 9 6 ). The herpes simplex virus major regulatory protein ICP4 blocks apoptosis induced by the virus or by hyperthermia. P r o c . N a t l . A c a d . S c i . USA 93, 9583-7. Leopardi, R., van Sant, C., Roizman, B. (1 9 9 7 ). The herpes simplex virus 1 protein kinase Us3 is required for protection from apoptosis induced by the virus. P r o c . N a t l . A c a d . S c i . U S A 94, 7891-7896.

Kaplitt, M., Tjuvajev, J., Leib, D., Berk, J., Pettigrew, K., Posner, J., Pfaff, D., Rabkin, S., Blasberg, R. (1 9 9 4 ). Mutant herpes simplex virus induced regression of tumors growing in immunocompetent rats. J . N e u r o o n c o l . 19, 137-47. Karpoff, H., D'Angelica, M., Blair, S., Brownlee, M., Federoff, H., Fong, Y. (1 9 9 7 ). Prevention of hepatic tumor metastases in rats with herpes viral vaccines and gammainterferon. J . C l i n . I n v e s t . 99, 799-804.

MacLean, M., Ul-Fareed, M., Roberson, L., Harland, J., Brown, S. (1 9 9 1 ). Herpes simplex virus type 1 deletion variant 1714 and 1716 pinpoint neurovirulence-related sequences in Glasgow strain 17+ between immediate early gene 1 and the 'a' sequence. J . G e n . V i r . 72, 63--639.

Kesari, S., Lasner, T., Balsara, K., Randazzo, B., Lee, V., Trojanowski, J., Fraser, N. (1 9 9 8 ). A neuroattenuated ICP34.5-deficient herpes simplex virus type 1 replicates in ependymal cells of the murine central nervous system. J . G e n . V i r o l . 79, 525-36.

Markert, J., Medlock, M., Martuza, R., Rabkin, S., Hunter, W. (1 9 9 8 ). Initial report of phase I trial of genetically engineered HSV-1 in Patients with malignant glioma. In 23rd International Herpesvirus workshop; York, UK pp A384.

Kesari, S., Randazzo, B., Valyi-Nagy, T., Huang, Q., Brown, S., MacLean, A., Lee, V., Trojanowski, J., Fraser, N. (1 9 9 5 ). Therapy of experimental human brain tumors using a neuroattenuated herpes simplex virus mutant. Lab. I n v e s t . 73, 636-48.

Markovitz, N., Baunoch, D., Roizman, B. (1 9 9 7 ). The range and distribution of murine central nervous system cells infected with the gamma(1)34.5- mutant of herpes simplex virus 1. J . V i r o l . 71, 5560-9. Martuza, R., Malick, A., Markert, J., Ruffner, K., Coen, D. (1 9 9 1 ). Experimental therapy of human glioma by means of a genetically engineered virus mutant. S c i e n c e 252, 854-856.

Koyama, A., Adachi, A. (1 9 9 7 ). Induction of apoptosis by herpes simplex virus type 1. J . G e n . V i r o l . 78, 29092912.

McCormick, C., Leduc, Y., Martindale, D., Mattison, K., Esford, L., Dyer, A., Tufaro, F. (1 9 9 8 ). The putative tumour suppressor EXT1 alters the expression of cell-surface heparan sulfate. Nature Genet 19, 158-61.

Koyama, A., Miwa, Y. (1 9 9 7 ). Suppression of apoptotic DNA fragmentation in herpes simplex virus type 1-infected cells. J . V i r o l . 71, 2567-71. Kramm, C. M., Chase, M., Herrlinger, U., Jacobs, A., Pechan, P. A., Rainov, N. G., Sena-Esteves, M., Aghi, M., Barnett, F. H., Chiocca, E. A., Breakefield, X. O. (1 9 9 7 ). Therapeutic efficiency and safety of a second-generation replication-conditional HSV1 vector for brain tumor gene therapy. Human Gene Ther. 8, 2057-68.

McGeoch, D., Dalrymple, M., Davison, A., Dolan, A., Frame, M., McNab, D., Perry, L., Scott, J., Taylor, P. (1 9 8 8 ). The complete DNA sequence of the long unique region in the genome of herpes simplex virus type 1. J . G e n . V i r o l . 69, 1531-74.

Kucharczuk, J. C., Randazzo, B., Elshami, A. A., Sterman, D. H., Amin, K. A., Molnar-Kimber, K. L., Brown, M. S., Litzky, L. A., Fraser, N. W., Albelda, S. M., Kaiser, L. R. (1 9 9 7 ). Use of a Replication-Restricted, Recombinant Herpes Virus to Treat Localized Human Malignancy. Cancer R e s . 57, 466-471.

McKie, E. A., MacLean, A. R., Lewis, A. D., Cruickshank, G., Rampling, R., Barnett, S. C., Kennedy, P. G., Brown, S. (1 9 9 6 ). Selective in vitro replication of herpes simplex virus type 1 (HSV-1) ICP34.5 null mutants in primary human CNS tumours--evaluation of a potentially effective clinical therapy. Br. J. Cancer 74, 745-752.

Kutubuddin, M., Federoff, H., Halterman, M., Atkinson, M., Planelles, V., Rosenblatt, J. (1 9 9 8 ). Eradication of established murine lymphoma using herpes amplicon vectors. AACR Proceedings A3777.

McMenamin, M., Byrnes, A., Charlton, H., Coffin, R., Latchman, D., Wood, M. (1 9 9 8 ). A gamma34.5 mutant of herpes simplex 1 causes severe inflammation in the brain. N e u r o s c i e n c e 83, 1225-1237.

Kwong, A., Kruper, J., Frenkel, N. (1 9 8 8 ). Herpes simplex virus virion host shutoff function. J . V i r o l . 62, 912-921.

Mineta, T., Rabkin, S., Martuza, R. (1 9 9 4 ). Treatment of malignant gliomas using ganciclovir-hypersensitive,

87

Coukos et al: HSV-based oncolytic therapy ribonucleotide reductase-deficient herpes simplex mutant. Cancer Res 54, 3963-3966.

viral

Singhal, S., Kaiser, L. (1 9 9 8 ). Cancer chemotherapy using suicide genes. Surg Oncol Clin N Am 7, 505-36.

Mineta, T., Rabkin, S., Yazaki, T., Hunter, W., Martuza, R. (1 9 9 5 ). Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. Nature Med 1, 938-943.

Spear, P., Shieh, M., Herold, B., WuDunn, D., Koshy, T. (1 9 9 2 ). Heparan sulfate glycosaminoglycans as primary cell surface receptors for herpes simplex virus. A d v . E x p . M e d . B i o l . 313, 341-53.

Miyatake, S.-I., Iyer, A., Martuza, R., Rabkin, S. (1 9 9 7 ). Transcriptional targetting of herpes simplex virus for cell specific replication. J V i r o l . 71, 5124-5132.

Sterman, D. H., Treat, J., Elshami, A. A., Amin, K., MolnarKimber, K., Coonrod, L., Recio, A., Wilson, J. M., Roberts, J. R., Litzky, L. A., Albelda, S. M., Kaiser, L. R. (1 9 9 8 ). Adenovirus mediated herpes simplex virus thymidine kinase/ganciclovir gene therapy. in patients with localized malignancy, results of a phase I clinical trial in malignant mesothelioma. Human Gene Ther. 9, 1083-1092.

Molnar-Kimber, K. L., Sterman, D. H., Chang, M., Kang, E. H., Elbash, M., Lanuti, M., Elshami, A., Wilson, J. M., Kaiser, L. R., Albelda, S. M. (1 9 9 8 ). Impact of pre-existing humoral and cellular immune responses induced by adenoviral-based gene therapy for localized mesothelioma. Human Gene Ther. 9, 2121-2133.

Subak-Sharpe, J., Dargan, D. (1 9 9 8 ). HSV molecular biology, general aspects of herpes simplex virus molecular biology. Virus Genes 16, 239-51.

Montgomery, R., Warner, M., Luro, B., Spear, P. (1 9 9 6 ). Herpes simplex virus-1 entry into cells mediated by a novel member of the TNF/NGF receptor family. C e l l 87, 427-436.

Tepper, R., Mule, J. (1 9 9 4 ). Experimental and clinical studies of cytokine gene-modified tumor cells. Hum. Gene Ther. 5, 153-64.

Pardoll, D. (1 9 9 2 ). Immunotherapy with cytokine genetransduced tumor cells, the next wave in gene therapy for cancer. Curr. Opin. Oncol. 4, 1124-9.

Toda, M., Martuza, R., Kojima, H., Rabkin, S. (1 9 9 8 a ). In situ cancer vaccination, an IL-12 defective vector/replicationcompetent herpes simplex virus combination induces local and systemic antitumor activity. J I m m u n o l 160, 445764.

Pardoll, D. (1 9 9 6 ). Cancer vaccines, a road map for the next decade. Curr. Opin. Immunol. 8, 619-21. Pyles, R. B., Warnick, R. E., Chalk, C., Szanti, B. E., Parysek, L. (1 9 9 7 ). A novel multiply mutated HSV-1 Strain for the treatment of Human brain tumors. Human Gene Ther. 8, 533-544.

Toda, M., Rabkin, S. D., Martuza, R. L. (1 9 9 8 b ). Treatment of Human Breast cancer in a brain metastatic model by G207, a replication competent multimutated Herpes Simplex virus 1. Human Gene Ther. 9, 2173-2185.

Randazzo, B., Bhat, M., Kesari, S., Fraser, N., Brown, S. (1 9 9 7 ). Treatment of experimental subcutaneous human melanoma with a replication-restricted herpes simplex virus mutant. J . I n v e s t . D e r m a t . 108, 933-7.

Tropea, F., Troiano, L., Monti, D., Lovato, E., Malorni, W., Rainaldi, G., Mattana, P., Viscomi, G., Ingletti, M., Portolani, M. e. a. (1 9 9 5 ). Sendai virus and herpes virus type 1 induce apoptosis in human peripheral blood mononuclear cells. E x p C e l l R e s 218, 63-70.

Randazzo, B. P., Kucharczuk, J. C., Litzky, L. A., Kaiser, L. R., Brown, S. M., MacLean, A., Albelda, S. M., Fraser, N. W. (1 9 9 6 ). Herpes simplex 1716--an ICP 34.5 mutant--is severely replication restricted in human skin xenografts in vivo. V i r o l o g y 223, 392-395.

Tsuburaya, A., Hattori, S., Yanoma, S., Kawamoto, S., Okuda, K., Amano, T., Noguchi, Y. (1 9 9 8 ). Treatment of peritoneal metastasis by a defective herpes simplex viral vector bearing interleukin-2. AACR P r o c e e d i n g s 39, A69.

Rees, R., Ali, S., McLean, C., Bourrsnell, M., Reedere, S., Sivasubramaniam, S., Entwisle, C., Blakeley, D., Shields, J. (1 9 9 8 ). Immunogenicity of murine renal carcinoma (RENCA) cells infected with a disabled infectious single cycle (DISC) herpes simplex vector carrying the mGM-CSF gene. AACR Proceedings 39, A49.

Tung, C., Federoff, H., Brownlee, M., Karpoff, H., Weigel, T., Brennan, M., Fong, Y. (1 9 9 6 ). Rapid production of interleukin-2- secreting tumor cells by herpes simplex virusmediated gene transfer, implications for autologous vaccine production. Human Gene Ther. 7, 2217-24.

Roizman, B., Sears, A. Herpes Simplex Viruses and Their Replication. In Fields Virology (1996), 3rd ed.B.Fields, D.H.Knipe, P.M. Howley. Philadelphia, Lippincott-Raven Publishers, 1996, , pp 2231-2296.

Valyi-Nagy, T., Fareed, M., O'Keefe, J., Gesser, R., MacLean, A., Brown, S., Spivak, J., Fraser, N. (1 9 9 4 ). The herpes simplex virus type 1 strain 17+ g-34.5 deletion mutant 1716 is avirulent in SCID mice. J . G e n . V i r . 75, 2059-2063.

Roth, J., Cristiano, R. (1 9 9 7 ). Gene Therapy for Cancer: What have we done and where are we going? J . N a t l . C a n c e r I n s t . 89, 21-39.

Vile, R. (1 9 9 8 ). Gene Therapy. C u r r . B i o l . 29, R73-5. Wang, M., Rancourt, C., Alvarez, R., Siegal, G., Marconi, P., Krisky, D., Glorioso, J., Curiel, D. (1 9 9 8 ). High efficiency of thymidine kinase gene transfer to ovarian cancer cell lines mediated by herpes simplex virus type 1 vector. In 29 th Annual Meeting of Society of Gynecologic Oncologists; Orlando , FL pp A61.

Sanders, P., Wilkie, N., Davison, A. (1 9 8 2 ). Thymidine kinase deletion mutants of herpes simplex virus type 1. J . G e n . V i r o l . 63, 277-95.

88

Gene Therapy and Molecular Biology Vol 3, page 89 Whitbeck, J., Peng, C., Lou, H., Xu, R., Willis, S., Ponce de Leon, M., Peng, T., Nicola, A., Montgomery, R., Warner, M., Soulika, A., Spruce, L., Moore, W., Lambris, J., Spear, P., Cohen, G., Eisenberg, R. (1 9 9 7 ). Glycoprotein D of herpes simplex virus (HSV) binds directly to HVEM, a member of the tumor necrosis factor receptor superfamily and a mediator of HSV entry. J . V i r o l . 71, 6083-93. Whitley, R. Herpes Simplex Viruses. In Fields Virology (1996), 3rd. ed. B.Fields, D.M. Knipe, P.M. Howley Philadelphia, PA, Lippincott-Raven Publishers, 2, pp 2297-2342. Yazaki, T., Manz, H., Rabkin, S., Martuza, R. (1 9 9 5 ). Treatment of human malignant meningiomas by G207, a replication competent multimutated herpes simplex virus 1. Cancer Res. 55, 4752-4756. York, I., Roop, C., Andrews, D., Riddell, S., Graham, F., Johnson, D. (1 9 9 4 ). A cytosolic herpes simplex virus protein inhibits antigen presentationto CD8+ T lymphocytes. C e l l 77, 525-535.

89

Gene Therapy and Molecular Biology Vol 3, page 91 Gene Ther Mol Biol Vol 3, 91-101. August 1999.

Transcriptional repression in cancer gene therapy: targeting HER-2/neu overexpression as an example Review Article

Mien-Chie Hung* and Shao-Chun Wang University of Texas M. D. Anderson Cancer Center, Department of Cancer Biology, Section of Molecular Cellular Biology, Box 79, 1515 Holcombe Boulevard, Houston, Texas 77030 . __________________________________________________________________________________________________ * To whom correspondence should be addressed. Phone: (713) 792-3668. Fax: (713) 794-4784. E-mail: mchung@odin.mdacc.tmc.edu Received: 16 October 1998; accepted: 9 November 1998

Summary Overexpression of the HER-2/neu oncogene has been well-documented as a frequent event in human cancers. In c l i n i c , overexpression o f HER-2/neu indicates a unfavorable prognosis and highly correlated with the l o w survival rate o f patients associated with breast and ovarian cancers. Downregulation of the HER-2/neu g e n e e x p r e s s i o n i n c a n c e r c e l l s b y attenuating the promoter a c t i v i t y o f t h e g e n e i s t h e r e f o r e a n a t t r a c t i v e s t r a t e g y t o r e v e r s e the transformation phenotype induced by HER-2/neu overexpression. We have identified a number o f cellular and viral transcriptional regulators, including the ets family member PEA3, the SV40 large T antigen, and the adenovirus type 5 E1A, which are able to repress the HER-2/neu gene expression. Expression of these transcriptional regulators resulted in downregulation of the HER-2/neu promoter activity and reversed the malignant phenotype of the transformed cells i n v i t r o . These observations were followed by a series o f studies t o investigate whether these HER-2/neu repressors can act therapeutically as tumor suppressor genes for cancers that overexpress HER-2/neu. T h e g r o w t h o f tumors derived from HER-2/neu-overexpressing cancer cells was inhibited by the transcriptional repressors, accompanied by decreased HER-2/neu e x p r e s s i o n i n t u m o r c e l l s . T h e r e s u l t s o f t h e s e preclinical studies clearly indicate that transcriptional repressors which downregulate HER-2/neu can be a promising regimen for cancer treatment in a gene therapy format.

transduction through all erbB receptor family members (Craus-Porta et al, 1997, Wallasch et al, 1995, Carraway et al, 1994, Sliwkowski et al, 1994, Plowman et al, 1993), due to the preference for the HER2/neu receptor as a heterodimerization partner for all erbB receptors. After ligand binding, EGFR, HER-3 (also known as erbB3), and HER-4 (also known as erbB4) can heterodimerize with HER2/neu, and can lead to the tyrosine phosphorylation of all of these receptors (Craus-Porta et al, 1997, Wallasch et al, 1995, Sliwkowski et al, 1994).

I. Introduction A. HER2/neu overexpression serves as a critical target for cancer gene therapy The HER-2/neu (also known as c-erbB2) gene encodes a receptor tyrosine kinase (p185) with significant structural and functional homology to the epidermal growth factor receptor (EGFR) (Bargmann et al, 1986a, Hung et al, 1986, Yamamoto et al, 1986). Each protein member of the erbB receptor family contains an extracellular domain, a transmembrane domain, and an intracellular domain with intrinsic tyrosine kinase activity. Although the ligand for the HER2/neu receptor has not been identified, the HER2/neu receptor is known to mediate lateral signal

The oncogenic property of the HER-2/neu protooncogene was originally demonstrated in the rat neu oncogene (Hung et al, 1989, Bargmann 1986b, Hung et al, 1986). As a matter of fact, the mutation-activated rat neu

91

Hung and Wang: Targeting HER-2/neu overexpression in cancer gene therapy oncogene, which contains a point mutation in the transmembrane domain of the protein resulting in a constitutive tyrosine kinase activity, was originally isolated from rat neuroblastoma due to its ability to transform mouse cells (Hung et al, 1989, Bargmann et al, 1986b, Hung et al, 1986). In human, the HER-2/neu proto-oncogene is frequently amplified or overexpressed in many types of cancers including breast (Gusterson et al, 1992, Toikkanen et al, 1992, Slamon et al, 1989, Slamon et al, 1987), ovarian (Slamon et al, 1989, Burchuck et al, 1991, Burchurk et al, 1990), lung (Shi et al, 1992, Weiner et al, 1990, Schneider et al, 1989), stomach (Yokota et al, 1988, Park et al, 1989), and oral (Xia et al, 1997) cancers, suggesting that HER-2/neu overexpression plays a critical role in the development of human cell malignancy. The overall survival rate of cancer patients whose tumors have HER-2/neu overexpression is significantly shorter than those patients whose tumor do not have HER-2/neu overexpression (Slamon et al, 1989, Slamon et al, 1987, Burchurk et al, 1990, Weiner et al, 1990, Xia et al, 1997). Furthermore, increased expression of the HER-2/neu gene has been shown to correlate with the number of lymph node metastases in breast cancer patients (Slamon et al, 1987), an observation consistent with many studies in that the mutation-activated neu gene induced metastatic potential in mouse 3T3 cells and that overexpression of the normal human HER2/neu gene enhanced metastatic potential in human breast, ovarian, and non-small-cell lung carcinoma (NSCLC) cells by promoting multiple steps in the metastatic cascade (Tan et al, 1997, Yu et al, 1994, Benz et al, 1993, Yu et al, 1992a, Chazin et al, 1992, Yu and Hung 1991a, Slamon et al, 1989, Slamon et al, 1987). In addition to metastasis of cancer cells, it is generally believed that HER-2/neu overexpression is correlated to chemoresistance of cancer cells. High level of HER-2/neu expression in human NSCLC appeared to result in enhanced resistance to a panel of chemotherapeutic agents (Tsai et al, 1995. Tsai et al, 1993). Similarly, overexpression of HER-2/neu in breast cancer cells induced chemoresistance to Taxol (Paclitaxel) (Yu et al, 1998, Yu et al, 1996). However, the expression level of HER-2/neu seems to be critical for the development of chemoresistance since in certain cell lines moderate p185 expression level is not accompanied with significant drug resistance (Pegram et al, 1997). It is likely that the HER-2/neu expression has to be higher than a threshold level to induce significant drug resistance. Furthermore, the chemoresistance developed in those HER2/neu-overexpressing breast cancer cells is limited to Paclitaxal and Taxotere but not to other drugs (Pegram et al, 1997; Yu et al, 1996; Yu, D. and Hung, M. -C., unpublished results), suggesting a selective mechanism of resistance. it is not yet clear why HER-2/neu overexpression-mediated drug resistance behaves differently between lung and breast cancer cells. However, in the case

92

of resistance to Paclitaxel by HER-2/neu overexpression in breast cancer cells, a molecular mechanism has recently been suggested (Yu et al, 1998a): upregulation of p21 by HER-2/neu overexpression inhibits cyclin B/cdc2 kinase activity in G2/M phase which is required for Paclitaxel induced apoptosis. This mechanism clearly indicates that HER-2/neu overexpression in breast cancer cells antagonizes Paclitaxel-induced apoptosis. Since the HER2-neu proto-oncogene overexpression significantly contributes to the malignant development of many types of human cancers in different aspects, molecular strategies which aim to down-regulate the HER2/neu gene expression have become highly attractive approaches to fight against human cancer.

B. Transcriptional repression as an effective means to downregulate HER2/neu expression in cancer cells HER-2/neu gene amplification can be detected in majority of breast tumor tissues with overexpression of the HER-2/neu-encoded p185 protein (Slamon et al, 1989). In established breast cancer cell lines, both gene amplification and transcriptional upregulation are common scenario accounting for the increased HER-2/neu gene expression in different breast cancer cells (Kraus et al, 1987, Millar et al, 1994, Bosher et al, 1996). Interestingly, it has been shown that in 10-20% of HER2/neu -overexpressing breast tumors and in virtually all HER2/neu-positive lung cancers the HER2/neu mRNA and protein expression can occur in the absence of increased gene copy number (Kameda et al, 1990, Kern et al, 1990, Slamon et al, 1989, King et al, 1989, Tandon et al, 1989, Berger et al, 1988). It is therefore likely that both gene amplification and transcriptional upregulation are involved in HER-2/neu overexpression in cancer cells. The promoter of the HER-2/neu gene has been well characterized. In the past few years, knowledge about the cis- and trans-acting elements regulating the transcription of the HER-2/neu proto-oncogene have been rapidly accumulated. A number of cis-acting motifs are distributed along the HER-2/neu promoter, including the binding sites of transcription factors Sp1, OTF1, AP2, E4TF1, and PEA3. Another 13-bp sequence in the promoter region has been identified as a positive element for HER-2/neu transactivation (Miller et al, 1994). The corresponding binding transcription factor(s), however, has not yet been identified. AP2 has been shown to be a strong activator of the HER-2/neu gene and is functionally activated in the HER-2/neu-overexpressing breast cancer cell lines such as MDA-MB-361, MDA-MB-175, ZR-75-1, BT-474, and SK-BR-3 (Hollywood et al, 1993). The high activity of AP2 in these cell lines has been correlated with the elevated HER-2/neu gene expression level in these cells.

Gene Therapy and Molecular Biology Vol 3, page 93 On the other hand, the HER-2/neu gene is subject to the negative regulation of a number of cellular or viral factors through different mechanisms. For example, PEA3, a member of the ets family (X. Xing, S. -C. Wang, and M. -C. Hung, unpublished results; Xing et al, 1997), and the retinoblastoma tumor suppressor (RB) (Yu et al, 1992b) can repress the HER-2/neu gene expression. Interestingly, in addition to the cellular factors, the HER-2/neu gene transcription can also be repressed by a number of viral transcription factors such as the simian virus 40 (SV40) large T antigen and the adenovirus type 5 E1A (Yan et al, 1991a, Yu et al, 1991b). These studies have indicated that repression of transcription is an effective way to reverse the malignant transformation mediated by HER-2/neu overexpression, and have demonstrated the potential application of transcriptional repressors as therapeutic agents targeting HER-2/neu-overexpressing cancer cells.

II. Tumor suppression effects of HER2/neu down-regulation mediated by genes encoding transcriptional regulators A. Tumor suppression by viral transcriptional regulators Both E1A and T antigen are viral proteins, and their ability to suppress HER-2/neu-mediated cell transformation is surely a surprising biological phenomenon. The adenovirus genome is about 36 kb in size. Among the proteins encoded by the adenovirus genome, E1A gene products are nuclear-localized phosphoproteins and have special regulatory role in the adenoviral life cycle (Berk 1986). E1A is the first region to be expressed after infection (Tooze 1981). Other late adenoviral genes can then be turned on by E1A proteins through interacting and modifying the host transcriptional apparatus. There are two types of adenovirus E1A. One is the transforming E1A carried by the adenovirus type 12. This type of E1A gene alone can transform normal cell lines (Schrier et al, 1983). The other type of E1A, such as the adenovirus type 2 or type 5 E1A, can not transform cells by itself alone. It is noteworthy that for the purpose of this review E1A refers to the type 5, non-transforming E1A. E1A was classified as an "immortalization oncogene" due to its ability to cooperate with the transforming ras or E1B genes to transform primary embryo cells (Byrd et al, 1988, Montell et al, 1984, Land et al 1983, Ruley 1983). However, expression of the E1A gene itself does not induce transforming phenotypes (Yu et al, 1992a). As a matter of fact, there are a number of studies indicated that E1A is associated with metastasis- or tumor-suppression activities (Pozzatti et al, 1988, Frisch 1991, van Groningen 1996). Recently, E1A has been shown to induce apoptosis under some conditions (Lowe and Ruley 1993, Rao et al, 1992). This property is similar

93

to the well-known tumor suppressor gene p53 that also has the ability to induce apoptosis (Subramanian et al, 1995, Symonds et al, 1994). All of these observations indicate that tagging E1A as an oncogene is a misconception. We have first discovered that the adenovirus 5 E1A gene can repress HER-2/neu overexpression through both transient transfection and adenovirus delivery systems (Yu et al, 1991b, Yu et al, 1990). Transfection of the E1A gene into the genomic rat neu oncogene transformed mouse embryo fibroblast cell lines virtually abolishes the tumorigenicity and metastatic potential induced by the HER-2/neu oncogene through repression of HER-2/neu gene expression (Yu et al, 1992a, Yu et al, 1991b). Reexpression of the HER-2/neu-encoded p185 protein in these E1A transfectants by transfection of a HER-2/neu cDNA construct driven by a promoter that cannot be inhibited by E1A recovered virtually all of the transforming phenotypes including tumorigenicity, the ability to grow in soft agar, and higher in vitro growth rate (Yu et al, 1993a, b). Interestingly, the ability to induce experimental metastasis (measured by lung colonization through i. v. injection of the tumor cells) was only partially recovered. The incomplete regeneration of metastatic potential could be accounted for by the fact that E1A inhibits gelatinolytic activity that was critical for invasive activity of metastatic cells. This result indicates that the suppression of metastasis by E1A is through multiple molecular mechanisms in addition to repressing the HER-2/neu gene expression (Yu et al, 1992a). We have also demonstrated that E1A can indeed function as a tumor suppressor in the HER-2/neu-overexpressing human ovarian cancer cell line by down-regulating the expression of the HER-2/neu mRNA and the p185 protein product (Yu et al, 1995, Yu et al, 1993a, b, Yu et al, 1991b, Yu et al, 1990). The E1A-expressing ovarian cancer cell line had reduced malignancy, including a decreased ability to develop tumors in nude mice. Therefore, for the HER2/neu-overexpressing transforming cells including fibroblasts and human cancer cells, E1A can function as tumor suppressor. And transcriptional repression of the HER-2/neu oncogene contributes to the tumor suppression function. However, since E1A is not a DNA-binding protein, the transcriptional repression of HER-2/neu by E1A has to be mediated through the targeting of other transcription factors. This is supported by our recent study demonstrating that E1A can abolish HER-2/neu overexpression by targeting the coactivator p300, which is required for efficient expression of HER-2/neu (Chen and Hung 1997). To further investigate whether the E1A gene can be used as a therapeutic agent for HER-2/neu-overexpressing human breast and ovarian cancers in living host, a tumorbearing mouse model was established and the E1A gene

Hung and Wang: Targeting HER-2/neu overexpression in cancer gene therapy was delivered by the cationic liposome DC-Chol or a recombinant replication-deficient adenovirus. E1A treatment was able to effectively reduce the mortality of tumor-bearing mice and, in 60-80 % of the treated mice, resulted in tumor-free survival, suggesting that E1A gene therapy is a promising therapeutic regimen for cancers that overexpress HER-2/neu . In addition, the number of mice with distant metastases was significantly reduced even though a local treatment protocol by mammary fat pad injection was used in the orthotopic breast cancer model (Lane and Crawford 1979, Zhang et al, 1995). In addition to the breast and ovarian cancer animal models, we also used a lung cancer animal model to test the therapeutic efficiency of E1A (Chang et al, 1996). In this case, the tumor-bearing mice were established through intratracheal inoculation of lung cancer cells and the E1A gene was delivered by an adenovirus vector through intravenous injection. A significant therapeutic efficacy was observed. Therefore the tumor suppression effect of E1A can be demonstrated through two independent gene delivery systems and three different animal models. Based on these results, a phase I clinical trial, using cationic liposome to deliver the E1A gene was initiated at the M. D. Anderson Cancer Center. Preliminary results suggested a downregulation of the HER-2/neu p185 oncoprotein concomitant with the detection of the E1A gene expression in treated breast and ovarian cancer patients. The simian virus 40 (SV40) large T antigen is a multifunctional protein required for the replication of the viral genome and for cell transformation (Lane and Crawford 1979, Linzer and Levine 1979). This viral protein contains transformation domains which can mediate binding to the retinoblastoma protein (pRb) and p53, respectively (Manfredi and Prive 1994). Our previous studies showed that a mutant SV40 large T antigen can repress rat neu transcription in mouse fibroblast NIH 3T3 cells (Matin and Hung 1993). The mutant large T antigen, named K1, contains a single amino acid change within the pRb-binding/transformation domain, which renders the viral protein unable to bind to pRb, and consequently failed to induce cell transformation (Kalderon and Smith 1984, Cherington et al, 1988, DeCaprio et al, 1988). Since the K1 mutant represses HER-2/neu expression as effectively as the wild-type counterpart (Matin and Hung 1993), we further tested whether K1 can function as a tumor suppressor for HER-2/neu-overexpressing ovarian cancer cells. K1 did suppress cancer cell growth, resulting in a significant therapeutic effect on mice with ovarian cancer with about 40% of treated mice were alive after one year (Xing et al, 1996). The autopsies showed that the mice from the control groups had larger volume of ascites and tumors within the peritoneal cavity or diaphragm or metastasis to the lungs. However, the mice that received K1-liposome complex had more locally distributed tumor

94

nodules in their peritoneal cavities. This difference indicates that K1 suppressed the growth of HER-2/neuoverexpressing tumor cells so that the tumors developed with longer latency. The K1-treated mice survived for one year were sacrificed and examined for residual tumors, but no tumors were observed in the peritoneal cavity. Our results indicate that both viral transcription factors, E1A and the large T antigen, can suppress tumor cell growth through a HER-2/neu-involved pathway. However, the possibility that E1A can mediate tumor suppression function through a HER-2/neu-independent mechanism should not be excluded.

B. Tumor suppression by cellular DNAbinding transcriptional factor, PEA3 The mouse PEA3 (Polyomavirus Enhancer Activator 3) gene and its human homologue were first cloned from cDNA expression libraries due to the binding ability to the sequence 5’-AGGAAG-3’ (the PEA3 binding motif) within the polyomavirus enhancer promoter element (Xin et al, 1992, Higashino et al, 1993). The PEA3 protein contains a stretch of about 85 amino acids with extensive sequence homology with the ETS domain, a conserved region shared by all ets family members that characteristically bind as monomers to the consensus core sequence GGAA by their ETS DNA-binding domains (Monte et al, 1994, Brown et al, 1992, Xin et al, 1992, Karim et al, 1990), and regulates the expression of target genes including genes involved in cell growth and differentiation (Ma et al, 1998, Taylor et al, 1997). The ets gene family currently contains at least 30 members present in a diverse spectrum of metazoan organisms (Degnan et al, 1993, Laudet et al, 1993) Subfamilies can be identified based on sequence/structure homology and the association with other accessory proteins for DNA binding. The PEA3 subfamily is composed of three members : PEA3 (Xin et al, 1992), ERM (Nakae et al, 1995, Monte et al, 1994), and ER81 (Brown et al, 1992). In addition to the ETS domain, members of this subfamily share significant sequence similarity at an N-terminal acidic transcriptional activation domain (Nakae et al, 1995, Wasylyk et al, 1993, Macleod et al, 1992, Seth et al, 1992, Karim et al, 1990). Expression of the PEA3 gene is ubiquitous in different species and can be identified in mouse, rat, monkey, and human cells. However, PEA3 RNA expression is tissue-specific with highest level detected in brain, and, to a lower level, in pancreas, lung, and mammary gland (Xin et al, 1992). Most members of the ets family express at high levels in hemotopoietic cells. Unlike other Ets proteins, PEA3 is the only member identified to date that is apparently not expressed in cells with hematopoietic origins (Xin et al, 1992). The significance of this tissue specific distribution is not clear.

Gene Therapy and Molecular Biology Vol 3, page 95 There have been a number of candidate PEA3-regulated genes reported mainly based on the occurrence of putative PEA3 binding motif in their promoter regions. Interestingly, a great portion of these candidates fall in the category of genes encoding matrix metalloproteinases (Higashino et al, 1995), such as collagenase (Gutman et al, 1990), stromelysin (Buttice et al, 1993), and the urokinase-type plasminogen activator (uPA), a serine proteinase (Nerlov et al, 1992). All these enzymes are believed to involve in the regulation of extracelluar proteolysis, both in the normal organisms and in certain pathological conditions including tumor invasion and metastasis (Matrisian 1994). Consistent with this correlation, exogenous expression of PEA3 in the breast cancer cell line MCF-7 resulted in enhanced tumor invasiveness and metastasis (Mitsunori et al, 1996). Caveats should be taken to interpret these results. It is possible that members of the Ets family can have the same specificity required for DNA binding and share the same binding motif (Xin et al, 1992). As a matter of fact, it has been shown that Ets-2, an Ets protein belonging to the Pointed subfamily (Klambt 1993), is critical for the phobol ester (TPA)-mediated induction of the human stromelysine gene expression through the PEA3 binding motif in the promoter (Buttice et al, 1993). A similar conclusion has been drawn for the promoter of the uPA gene (Pankov et al, 1994). In addition, whether the PEA3 protein directly binds to the putative PEA3 motif in the collagenase promoter is not clear due to the lack of appropriate anti-PEA3 antibodies to confirm the identity of the DNA-binding activity detected on the PEA3 motif (Gutman et al, 1990). The occurrence of the PEA3 binding motif is not limited to those genes which potentially can enhance invasion and metastasis. Two consensus PEA3 binding motifs, distal and proximal, have been identified in the upstream regulatory region of the tumor suppressor gene maspin (Zhang et al, 1997a, Zhang et al, 1997b). Both motifs are positive regulatory elements for expression of the gene; the proximal site is the major functioning motif of the gene in mammary epithelial cells while both sites are equally critical for maspin expression in prostate cells. Functional studies have demonstrated that maspin functions as a tumor suppressor by inhibiting tumor invasion, metastasis, as well as tumor growth (Sheng et al, 1995, Zou et al, 1994). Even though it is still not clear if the PEA3 protein binds to the PEA3 motif in the maspin promoter, these observations are consistent with the prospect that tumor metastasis may be the result of imbalance between enhancing and suppressing factors (Liotta et al, 1991). This point is especially noteworthy given the large number of the ets family members and the resemblance of their DNA-binding domains and the DNA sequences of their target DNA motifs.

95

It is interesting to investigate the role of PEA3 in HER-2/neu gene expression and HER-2/neu-mediated transformation since a consensus PEA3- binding motif, 5'AGGAAG-3', is present 26 nucleotide upstream from the major mRNA start site in the promoter of the human, rat, and mouse HER-2/neu gene (Tal et al, 1987). It has been reported that PEA3 can mediate induction of the HER2/neu gene expression through the PEA3 binding motif (Benz et al, 1997). These results, however, were derived from the experiment using the COS monkey cell line. As will be mentioned below, this cell line can be characteristically different from other laboratory human breast and ovarian cancer cell lines, for which the investigation of PEA3's functions would be more biologically relevant. Furthermore, the hypothesis of PEA3-mediated HER-2/neu induction would predict a causal relationship between elevated PEA3 expression and HER-2/neu overexpression in cancer cells. However, analysis of PEA3 gene expression in various breast cancer cells dose not support this hypothesis. In fact, decreased PEA3 RNA expression was detected in breast cancer cell lines with HER-2/neu overexpression (such as BT 474, SK-BR-3, MDA-MB-361), while there was no detectable PEA3 mRNA in other HER-2/neu-overexpressing cell lines (such as MDA-MB-453, ZR-75-1, and MDA-MB134-V) (Baert et al, 1997). Nevertheless, these results suggest a negative role of PEA3 in regulating HER-2/neu expression. This prospect was directly tested in our laboratory and the following results demonstrate that PEA3 is indeed a negative transregulator of the proto-oncogene HER-2/neu (Xing et al, 1997). (1) The purified GST-PEA3 fusion protein can specifically recognize and bind to the consensus PEA3 binding motif on the HER-2/neu promoter. (2) Based on the co-transfection experiments performed on HER-2/neu-overexpressing human cancer cell lines, the HER-2/neu promoter activity can be down-regulated by PEA3 in a dose-dependent manner. However, destruction of the PEA3-binding site on the HER-2/neu promoter by site-directed mutagenesis abolished the promoter activity, indicating that PEA3-induced trans-repression of the HER2/neu promoter might involve competition between PEA3 and another ets-related transcriptional activator(s), which contributes to the transformed phenotype of HER-2/neu . (3) PEA3 can suppress the focus forming ability of mouse embryonic fibroblast transformed by the genomic mutation-activated genomic rat neu. (4) Expression of PEA3 can suppress the growth of HER-2/neu-overexpressing human cancer cell lines in vitro, but not cell lines with basal level of HER-2/neu expression. Based on these results, the tumor suppression function

Hung and Wang: Targeting HER-2/neu overexpression in cancer gene therapy of PEA3 is emerging. Trimble et al, have reported that mammary tumors derived from the transgenic mice bearing the rat neu gene under the control of the mouse mammary tumor virus (MMTV) promoter expressed high level of PEA3 mRNA, suggesting that PEA3 may be required for tumorigenesis and metastasis in HER2/neu overexpressing cells (Trimble et al, 1993). However, the data is also consistent with the possibility that there may exist a negative regulatory loop pathway in which the overexpression of HER-2/neu would turn on the expression of PEA3 which then act as a transcriptional repressor of the HER-2/neu gene and resume the homeostatic balance. The rat neu gene in the transgenic mice setting was driven by the heterologous MMTV promoter which is very likely not subject to the negative control by PEA3. Expression levels of both PEA3 and HER-2/neu would be elevated under this situation. In addition to PEA3, other Ets proteins including ERF and Net have been reported to function as transcriptional repressors (Sgouras et al, 1995, Giovan et al, 1994). Other promoters negatively regulated by Ets binding sites have also been reported (Chen and Boxer 1995, Goldberg et al, 1994). Interestingly, the ets family member Ets2 has recently been reported to function as a tumor suppressor by reversing ras-mediated cellular transformation (Foos et al, 1998). To test whether PEA3 can be used as a therapeutic agent in vivo, tumors were induced in nude mice (nu/nu) with SK-OV-3-ip1, an ovarian cancer cell line derived from SK-OV-3 and has higher HER-2/neu expression. For mice treated with PEA3-DC-Chol complex, 50% of the mice were alive and healthy without palpable tumors after 12 months. The mice of the control group, however, developed tumors and ascites, and died within 6 months. The tumor suppression activity of PEA3 is correlated with HER-2/neu expression since another cell line 2774 c-10, an ovarian cancer cell line with basal level of HER-2/neu expressed, did not have response to PEA3 treatment and the mice died of tumor with 5 months. Tumor samples were examined for the expression of HER-2/neu with immunoblot analysis. The results confirmed that PEA3 delivered by the cationic liposome downregulated the expression of p185. The correlation between PEA3 expression and HER-2/neu downregulation was further demonstrated by immunohistochemical staining of the tumor samples obtained from the PEA3-treated, moribund mice. Approximately 30% of the cancer cells in the tumor were positive for PEA3 protein expression, while the p185 staining was negative for about 50% of cells. Similar level of PEA3 expression was observed for PEA3-treated 2774 c-10-derived tumors while no repression of p185 was detected in these tumors. These in vitro and in vivo data clearly demonstrate the tumor suppression activity of PEA3 and indicate the potential clinical application of

96

PEA3-cationic liposome targeting overexpressing cancer cells.

the

HER-2/neu

Even though PEA3 as well as the viral proteins E1A and SV40 large T can all suppress HER-2/neu transcription, they are very likely functioning through different mechanisms. Both E1A and SV40 large T may suppress HER-2/neu in an indirect manner. Association of E1A with the transcriptional co-activator CBP/p300 inhibits the p300 transactivation activity, which is required for efficient expression of the HER-2/neu gene (Chen and Hung, 1997). On the other hand, PEA3 down-regulates the HER-2/neu gene by directly binding to its cognate binding sequence on the promoter. This feature makes PEA3 a more attractive target for further molecular manipulation to develop therapeutic molecules with higher binding affinity and enhanced specificity.

III. Conclusions Overexpression of the proto-oncogene HER-2/neu can lead to cell transformation and tightly correlated with the development of malignant tumor growth in many tissue types. There are molecular approaches to target the promoter of HER-2/neu , which can downregulate the gene expression, reverses the malignant phenotype, and retards tumor growth in animal. The results of our in vivo and in vitro experiments demonstrate using viral or cellular transcriptional repressor genes transferred by safe and efficient molecular vehicles can result in significant therapeutic effects on cancer cells. Since gene overexpression is a common mechanism of cancer as well as other types of diseases such as AIDS, the therapeutic strategy discussed here can have a tremendous potential in clinical application. Finally, the studies of E1A- and PEA3-mediated HER-2/neu repression have unveiled new areas in cancer biology which is excitingly more complicated than what we used to expect. Studies of these areas would be critical for our understanding of cancer.

Acknowledgment The authors are supported by NCI RO1 CA 58880 and CA 77858 (to M.C.H.).

References Baert, J. L., Monte, D., Musgrove, E. A., Albagli, O., Sutherland, R. L., and Launoit, Y. (1 9 9 7 ). Expression of the PEA3 group of EST-related transcription factors in human breast cancer cells. Int. J. Cancer 70, 590-597. Bargmann, C. I. Hung, M. -C., and Weinberg, R. A. (1 9 8 6 a ). The neu oncogene encodes an epidermal growth factor receptor-related protein. Nature 319, 226-230. Bargmann, C. I. Hung, M. -C., and Weinberg, R. A. (1 9 8 6 b ).

Gene Therapy and Molecular Biology Vol 3, page 97 Multiple independent activations of the neu oncogene by a point mutation altering the transmembrane domain of p185. C e l l 45, 649-657. Benz, C. C., O’Hagan, R. C., Richter, B., et al, (1 9 9 7 ). HER2/Neu and the Ets transcription activator PEA3 are coordiately upregulated in human breast cancer. O n c o g e n e 15, 1513-1525. Benz CC. Scott GK. Sarup JC. Johnson RM. Tripathy D. Coronado E. Shepard HM. Osborne CK. (1 9 9 3 ). Estrogen-dependent, tamoxifen-resistant tumorigenic growth of MCF-7 cells transfected with HER2/neu. Breast Cancer Res. Treat. 24, 85-95. Berger, M. S., Locher, G. W., Saurer, S., et al, (1 9 8 8 ). Correlation of c-erbB-2 gene amplification and protein expression in human breast carcinoma with nodal status and nuclear grading. Cancer Res. 48, 1238-1243. Berk, A. J. (1 9 8 6 ). Adenovirus promoters and E1A transactivation. A n n . R e v . G e n e t . 2 0 , 45-79. Bosher, J. M., Totty, N. F., Hsuan, J. J., Williams, T., and Hurst, H. C. (1 9 9 6 ). A family of AP-2 proteins regulates c-erbB-2 expression in mammary carcinoma. O n c o g e n e 13, 1701-1707. Brown, T. A. and McKnight, S. L. (1 9 9 2 ). Specificities of protein-protein and protein-DNA interaction of GABP alpha and two newly defined ets related proteins. Ge ne s D e v . 6, 2502-2512. Burchuck, A, Rodriguez, G., Kinney, R., Soper, J., Dodge, R., Clark-Pearson, D., and Bast, R. (1 9 9 1 ). Overexpression of HER-2/neu in endometrial cancer is associated with advanced stage disease. Am. J . O b s t e t . G y n e c o l . 164, 15-21. Burchuck, A., Kamel, A., Whitaker, R., Kerns, B., Olt, G., Kinney, R., Soper, J., Dodge, R., Clark-Pearson, D., Marks, P., McKenzie, S., Yin, S., and Bast Jr. R. (1 9 9 0 ). Overexpression of HER-2/neu is associated with poor survival in advanced epithelial ovarian cancer. Cancer R e s . 50, 4087-4091. Buttice, G. and Kurkinen, M. (1 9 9 3 ). A polyomavirus enhancer A-binding protein-3 site and Ets-2 protein have a major role in the 12-O-tetradecanoylphorbol-13-acetate response of the human stromelysin gene. J . B i o l . C h e m . 268, 7196-7204. Byrd, P. J., Grand, R. J. A., and Gallimore, P. H. (1 9 8 8 ). Differential transformation of primary human embryo retinal cells by adenovirus E1A regions and combination of E1A (+) ras. O n c o g e n e 2, 477-484. Carraway, K. L., Sliwkowski, M. X., Akita, R. M., et al, (1 9 9 4 ). The erbB-3 gene product is a receptor for heregulin. J . B i o l . C h e m . 269, 14303-14306. Chang JY. Xia W. Shao R. Hung M. -C. (1 9 9 6 ). Inhibition of intratracheal lung cancer development by systemic delivery of E1A. O n c o g e n e 13, 1405-1412. Chang, J. Y., Xia, W., Shao, R., Sorgi, F., Hortobagyi, G. N., Huang, L., and Hung, M. -C. (1 9 9 7 ). The tumor suppression activity of E1A in HER-2/neu-overexpressing

97

breast cancer. O n c o g e n e 14, 561-568. Chazin VR. Kaleko M. Miller AD. Slamon DJ. (1 9 9 2 ). Transformation mediated by the human HER-2 gene independent of the epidermal growth factor receptor. O n c o g e n e 7, 1859-1866. Chen, H. and Hung M. -C. (1 9 9 7 ). Involvement of coactivator p300 in the transcriptional regulation of the HER-2/neu gene. J . B i o l . C h e m . 272, 6101-6104. Chen, H. M. and Boxer, L. M. (1 9 9 5 ). Pi 1 binding sties are negative regulator of bcl-2 expressionin pre-B cells. M o l . C e l l B i o l . 15, 3840-3847. Cherington V., Brown, M., Paucha, E., Louis, J., Spiegelman BM, and Roberts TM. (1 9 8 8 ). Separation of simian virus 40 large T antigen-transforming and origin-binding functions from the ability to block differentiation. M o l . C e l l B i o l . 8, 1380-1384. DeCaprio, J. A., Ludlow, J. W., Figge, J., Shew, J. -Y., Huang, C. -M., Lee, W. -H., Marsilis, E., Paucha, E., and Livingston, D. M. (1 9 8 8 ). SV40 large tumor antigen forms a specific complex with the product of the retinoblastoma susceptibility gene. C e l l 54, 275-283. Degnan, B. M., Degnan, S. M., Naganuma, T., and Morese, D. E. (1 9 9 3 ). The ets multigene family is conserved throughout the Metazoa. N u c l e i c A c i d R e s . 21, 34793484. Foos, G., Garcia-Ramirez, J. J., Galang, C. K., et al, (1 9 9 8 ). Elevated expression of Ets2 or distinct portions of Ets2 can reverse Ras-mediated cellular transformation. J . B i o l . C h e m . 273, 18871-18880. Frisch, S. M. (1 9 9 1 ). Antioncogenic effect of adenovirus E1A in human tumor cells. P r o s . N a t l . Acad. S c i . USA 88, 9077-9081. Giovan, A., Pintzas, A., Maira, S. M., et al, (1 9 9 4 ). Net, a new ets transcription factor that is activated by Ras. G e n e s D e v . 8, 1502-1513. Goldberg, Y., Treier, M., Ghysdael, J., et al, (1 9 9 4 ). Repression of AP-1-stimulated transcription by c-Ets-1. J . B i o l . C h e m . 269, 16566-16573. Graus-Porta, D., Beerli, R. R., Daly, J. M., et al, (1 9 9 7 ). ErbB-2, the preferred heterodimerazation partner of all ErbB receptors, is a mediator of lateral signalling. EMBO J . 16, 1647-1655. Gusterson, B., Gelber, R., Goldhirsch, A., Price, K., SaveSoderborgh, J., Anbazhagan, R., Styles, J., Rudenstam, C. -M., R., Reed, R., Martinez-Tello, F., Tiltman, A., Torhorst, J., Grigolato, P., Bettelheim, R., Neville, A., Burki, K., Castiglione, M., Collins, J., Lindtner, J., and Senn, H. -J. (1 9 9 2 ). Prognostic importance of c-erbB-2 expression in breast cancer. J . C l i n . O n c . 10, 10491056. Gutman, A. and Wasylyk, B. (1 9 9 0 ). The collagenase gene promoter contains a TPA and oncogene-responsive unit emcomapassing the PEA3 and AP-1 binding sites. EMBO J . 9, 2241-2246. Higashino, F., Yoshida, K., Noumi, T., Seiki, M., and

Hung and Wang: Targeting HER-2/neu overexpression in cancer gene therapy Fujinaga K. (1 9 9 5 ). Ets-related protein E1A-F can activate three different matrix metallo-proteinase gene promoters. O n c o g e n e 10, 1461-1463.

and Stehelin, D. (1 9 9 3 ). Evolution of the ets gene family. B i o c h e m . B i o p h y s . R e s . C o m m . 190, 814.

Higashino, F., Yoshida, K., Fujinaga, Y, et al, (1 9 9 3 ). Isolation of a cDNA encoding the adenovirus E1A enhancer binding protein: a new human member of the ets oncongene family. N u c l e i c A c i d R e s . 21, 547-553.

Linzer, D. I. H., and Levine, A. J. (1 9 7 9 ). Characterization of a 54K Dalton cellular SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal carcinoma cells. C e l l 17, 43-52.

Hollywood, D. P., and Hurst, H. C. (1 9 9 3 ). A novel transcription factor, OB2-1, is required for overexpression of the proto-oncogene c-erbB-2 in mammary tumor cell lines. EMBO J. 12, 2369-2375.

Liotta, L. A., Steeg, P. S., and Stetler-Stevenson, W. G. (1 9 9 1 ). Cancer metastasis and angiogenesis: imbalance of positive and negative regulation. C e l l 64, 327-336.

Hung, M. -C., Yan, D., and Xhao, X. (1 9 8 9 ). Amplification of the proto-neu gene facilitates oncogenic activation by a single point mutation. P r o c . N a t l . A c a d . S c i . U S A 86, 2545-2548. Hung, M. -C., Schechter, A. L., Chevray, P. L., Stern, D. F., and Weinberg, R. A. (1 9 8 6 ).Molecular cloning of the neu gene: absence of gross structural alteration in oncogenic alleles. P r o c . N a t l . A c a d . S c i . U S A 83, 261-264.

Lowe, S. W. and Ruley, H. E. (1 9 9 3 ). Stabilization of the p53 tumor suppressor is induced by adenovirus 5 E1A and accompanies apoptosis. G e n e s D e v . 7, 535-545. Ma, Y., Su, Q., and Tempst, P. (1 9 9 8 ). Differentiationstimulated activity binds an ETS-like, essential regulatory element in the human promyelocytic defensin-1 promoter. J . B i o l . C h e m . 273, 8727-8740. Macleod, K., Leprince, D., and Stehelin, D. (1 9 9 2 ). The ets gene family. Trends B i o c h e m . S c i . 17, 251-256.

Kalderon, D., and Smith, A. E. (1 9 8 4 ). In vitro mutagenesis of a putative DNA binding domain of SV40 large T. V i r o l o g y 39, 109-137.

Manfredi, J. J., and Prive, C. (1 9 9 4 ). The transformation activity of simian virus 40 large tumor antigen. B i o c h i m . B i o p h y s . A c t a . 1198, 65-83.

Kameda, T., Yasui, W., Yoshida, K., et al, (1 9 9 0 ). Expression of ERBB2 in human gastric carcinomas: relationship between p185ERBB2 expression and the gene amplification. Cacner Res. 50, 8002-8009.

Matin A., and Hung M. -C. (1 9 9 3 ). Negative regulation of the neu promoter by the SV40 large T antigen. C e l l Growth Differ. 4, 1051-1056.

Karim, F. D., Urness, L. D., Thummel, C. S., et al, (1 9 9 0 ). The ETS-domain: a new DNA-binding motif that recognizes a purine-rich core DNA sequence. Ge ne s D e v . 4, 1451-1453. Kern, J. A., Schwartz, D., Nordberg, J. E., et al, (1 9 9 0 ). p185neu expression in human lung adenocarcinomas predicts shortened survival. Cancer R e s . 50, 51845191. King, C. R., Swain, S. M., Porter, L., et al, (1 9 8 9 ). Heterogeneous expression of erbB-2 messenger RNA in human breast cancer. Cancer Res. 49, 4185-4191. Klambt, C. (1 9 9 3 ). The Drosophila gene pointed encodes two ETS-likeproteins which are involved in the development of the midline glial cells. D e v e l o p m e n t 117, 163-176. Kraus, M. H., Popescu, N. C., Ambaugh, S. C., King, C. R. (1 9 8 7 ). Overexpression of the EGF receptor-related proto-oncogene erbB-2 in human mammary tumor cell lines by different molecular mechanisms. E M B O J . 6 , 605-610. Land, H., Parada, L. F., and Weinberg, R. A. (1 9 8 3 ). Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes. Nature 304, 596-602. Lane, D. P., and Crawford, L. V. (1 9 7 9 ). T antigen is bound to a host protein in SV40-transformed cells. Nature 278, 261-262. Laudet, V., Niel, C., Deuterque-Coquillard, M., Le Prince, D.,

98

Matrisian, L. M. (1 9 9 4 ). Matrix metalloproteinase gene expression. A n n . N . Y . A c a d . S c i . U S A 732, 42-50. Miller, S. J., Suen, T. C., Sexton, T. B., Hung, M. -C. (1 9 9 4 ). Mechanisms of deregulated HER2/nue expression in breast cancer cell lines. I n t . J . O n c o l . 4, 599-608. Mitsunori, K., Yoshida, K., Higashino, F., Mitaka, T., Ishii, S., and Fujinaga, K. (1 9 9 6 ). A single ets-related transcription factor, E1AF, confers invasive phenotype on human cancer cells. O n c o g e n e 12, 221-227. Monte, D., Baert, J. L., Defossez, P. A., et al,, (1 9 9 4 ). Molecular cloning and characterization of human ERM, a new member of the Ets family closely related to mouse PEA3 and ER81 transcription factors. O n c o g e n e 9, 1397-1406. Montell, C., Courtois, G., Eng, C. et al, (1 9 8 4 ). Complete transformation by adenovirus 2 requires both E1A proteins. C e l l 36, 951-961. Nakae, K., Nakajima, K., Inazawa, J., Kitaoka, T., and Hirano, T. (1 9 9 5 ). ERM, a PEA3 subfamily of Ets transcription factors, can cooperate with c-Jun. J . B i o l . C h e m . 270, 23795-23800. Nerlov, C., De Cesare, D., Pergola, F., et al,, (1 9 9 2 ). A regulatory element that mediates co-operation between a PEA3-PA-1 element and an AP-1 stie is required for phorbol ester induction of urokinase enhancer activity in HepG2 hepatoma cells. EMBO J . 11, 4573-4582. Park, J. B., Rhim, J. S., Park, S. C., Kimm, S. W., and Kraus, M. H. (1 9 8 9 ). Amplification, overexpression, and rearrangement of the c-erbB-2 proto-oncogene in primary

Gene Therapy and Molecular Biology Vol 3, page 99 human stomach carcinomas. Cancer R e s . 49, 60056009. Pankov, R., Umezawa, A., Maki, R., et al, (1 9 9 4 ). Keratin 18 activation by Ha-ras is mediated through Ets and Jun binding sites. P r o s . N a t l . A c a d . S c i . U S A 91, 873877. Pegram, M. D., Finn, R. S., Arzoo, K., Beryt, M., Pietras, R. J., and Slamon, D. J. (1 9 9 7 ). The effect of HER-2/neu overexpression on chemotherapeutic drug sensitivity in human breast and ovarian cancer cells. O n c o g e n e 15, 537-547.

ovarian cancer. S c i e n c e 244, 707-712. Slamon, D. J., Clark, G. M., Wong, S. G., Levin, W. J., Ullrich A., and McGuire, W. L. (1 9 8 7 ). Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. S c i e n c e 235, 177-182. Sliwkowski, M. X, Schaefer, G., Akita, R. W., et al, (1 9 9 4 ). Coexpression of erbB2 and erbB3 proteins reconstitutes a high affinity receptor for heregulin. J . B i o l . C h e m . 269, 14661-14665.

Plowman, G. D., Grenn, J. M., Culouscou, J. M., et al, (1 9 9 3 ). Heregulin induces tyrosine phosphorylation of HER4/p180 erbB-4. Nature 366, 473-475.

Subramanian, T., Tarodi, B., and Chinnadurai, G. (1 9 9 5 ). p53-independent apoptotic and necrotic cell deaths induced by adenovirus infection: suppression by E1B 19K and Bcl-2 proteins. C e l l G r o w t h D i f f . 6, 131-137.

Pozzatti, R., McCormick, M., Thompson, M. A., and Khoury, G. (1 9 8 8 ). The E1A gene of adenovirus type 2 reduces the metastatic potential of ras-transformed rat embryo cells. M o l . C e l l B i o l . 8, 2984-2988.

Symonds, H., Krall, L., Remington, L., Saenz-Robles, M., Lowe, S., Jacks, T., and Van Dyke, T. (1 9 9 4 ). p53dependent apoptosis suppresses tumor growth and progression in vivo. C e l l 78, 703-711.

Rao, L., Debbas, M., Sabbatini, P., Hockenbery, D., Korsmeyer, S., and White, E. (1 9 9 2 ). The adenovirus E1A proteins induce apoptosis, which is inhibited by the E1B 19-kDa and Bcl-2 proteins. P r o s . N a t l . Acad. S c i . U S A 89, 7742-7746.

Tal, M., King, C. R., Kraus, M. H., Ullrich, A., Schlessinger, J,. Givol, D. (1 9 8 7 ). Human HER2 (neu) promoter: evidence for multiple mechanisms for transcriptional initiation. M o l . C e l l . B i o l . 7, 2597-601.

Ruley, H. E. (1 9 8 3 ). Adenovirus early region 1A enables viral and cellular transforming genes to transform primary cells in culture. Nature 304, 602-606. Schneider P. M., Hung, M. -C., Chiocca, S. M., Manning, J., Zhao, X. Y., Fang, K., and Roth, J. A. (1 9 8 9 ). Differential expression of the c-erbB-2 gene in human small cell and non-small cell lung cancer. C a n c e r R e s . 49, 4968-4971. Seth, A., Ascione, R., Fisher, R. J., et al, (1 9 9 2 ). The ets gene family. Cell Growth D i f f . 3 , 327-334. Sgouras, D. N., Athanasiou, M. A., Beal, G. J., Jr., et al, (1 9 9 5 ). ERF: an ETS domain protein with strong transcriptional repressor activity, can suppress etsassociated tumorigenesis and is regulated by phosphorylation during cell cycle and mitogenic stimulation. EMBO J. 14, 4781-4793. Sheng, S., Pemberton, P. A., and Sager, R. (1 9 9 5 ). Production, purification, and characterization of recombinant maspin proteins. J . B i o l . C h e m . 269, 30988-30993. Shi, D., He, G., Cao, S., Pan, W., Zhang, H. Z., Yu., D., and Hung, M. -C. (1 9 9 2 ). Overexpression of the c-erbB2/neu-encoded p185 protein in primary lung cancer. M o l . C a r c i n o g . 5, 213-218. Schrier, P. I., Bernards, R., Vaessen, R. T. M. J et al, (1 9 8 3 ). Expression of class I major histocompatibility antigens switched off by highly oncogenic adenovirus 12 in transformed rat cells. Nature 305, 771-775. Slamon, D. J., Godolphin, W., Jones, L. A., Holt, J. A., Wong, S. G., Keith, D. E., Levin, W. J., Stuart, S. G., Udove, J., Ullrich, A, and McGuire, W. L. (1 9 8 9 ). Studies of the HER-2/neu proto-oncogene in human breast and

99

Tan, M., Yao, J., and Yu, D. (1 9 9 7 ). Overexpression of the cerbB-2 gene enhanced intrinsic metastasis potential in human breast cancer cells without increasing their transformation abilities. Cancer Res. 57, 1199-1205. Tandon, A. K., Clark, G. M., Chamness, G. C., et al, (1 9 8 9 ). HER-2/neu oncogene protein and prognosis in breast cancer. J . C l i n . O n c o l . 7, 1120-1128. Taylor, J. M., Dupont-Versteegden, E. E., Davies, J. D., Hassell, J. A., Houle, J. D., Gurley, C. M., and Peterson, C. A. (1 9 9 7 ). A role for the ETS domain transcription factor PEA3 in myogenic differentiation. M o l . C e l l . B i o l . 17, 5550-5558. Toikkanen, S., Helin, H, Isola, J., and Joensuu, H. (1 9 9 2 ) Prognostic significance of HER-2 oncoprotein expression in breast cancer: a 30 year follow-up. J . C l i n . O n c . 10, 1044-1048. Tooze, J. (1 9 8 1 ). DNA tumor viruses. in Molecular biology of tumor viruses, 2nd Edition, ed.), Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Trimble, M. S., Xin, J. -H., Guy, C. T. et al, (1 9 9 3 ). PEA3 is overexpressed in mouse metastatic mammary adenocarcinomas. O n c o g e n e 8, 3037-3042. Tsai, C. M., Yu, D., Chang, K. T., Wu, L. H., Perng, R. P., Ibrahim, N. K., and Hung, M. -C. (1 9 9 5 ). Enhanced chemoresistance by elevation of the level of p185 neu i n HER-2/neu transfected human lung cancer cells. J . N a t l . Cancer Inst. 87, 682-684. Tsai, C. M., Chang, K. -T., Perng, R. -P., Mitsudomi, T., Chen, M. -H., and Gazdar, A. F. (1 9 9 3 ). Correlation of intrinsic chemoresistance of non-small-cell lung cancer cell lines with HER-2/neu gene expression but not with ras gene mutations. J . N a t l . C a n c e r I n s t . 85, 897901.

Hung and Wang: Targeting HER-2/neu overexpression in cancer gene therapy van Groningen, J. J., Cornelissen, I. M., van Muijen, G. N., et al, (1 9 9 6 ). Simultaneous suppression of progression marker genes in the highly malignant human melanoma cell line BLM after transfection with the adenovirus-5 E1A gene. B i o c h e m . B i o p h y s . R e s . Commun. 225, 808-816. Wallasch, C., Weiss, F. U., Niederfellner, G., et al, (1 9 9 5 ). Heregulin-dependent regulation of HER2/neu oncogenic signaling by heterodimerization with HER3. EMBO J . 14, 4267-4275. Wasylyk, B., Hahn, S. L., and Giovane, A. (1 9 9 3 ). The Ets family of transcription factors. E u r . J . B i o c h e m . 211, 7-18. Weiner, D. B., Nordberg, J., Robinson, R., Nowell, P. C., Gazdar, A, Green, M. I., Williams, W. V., Cohen, J. A., and Kern, J. A. (1 9 9 0 ). Expression of the neu geneencoded protein (p185 neu) in human non-small cell carcinomas of the lung. Cancer Res. 50, 421-425. Xia, W., Lau, Y. -K.,, Zhang, H. -Z., Liu, A. -R., Kiyokawa, N., Clayman, G. L., Katz, R., L., and Hung, M. -C. (1 9 9 7 ). Strong correlation between c-erbB-2 overexpression and overall survival of patients with oral squamous cell carcinoma. C l i n . C a n c e r R e s . 3, 3-9. Xin, J. H., Cowie, A., Lachance, P., and Hassell, J. A. (1 9 9 2 ). Molecular cloning and characterization of PEA3, a new member of the Ets oncogene family that is differentially expressed in mouse embryonic cells. Ge ne s D e v . 6, 481-496. Xing, X., Miller, S. J., Xia, W., and Hung, M. -C. (1 9 9 7 ) PEA3 as a therapeutic agent for HER-2/neuoverexpressing human cancers. Abstract of the Department of Defense Breast Cancar Research Program Meeting, Washington, DC, 1997. Volume II, 549-550. Xing, X., Matin, A., Yu, D., Xia, W., Sorgi, F., Huang, L., and Hung, M. -C. (1 9 9 6 ). Mutant SV40 large T antigen as a therapeutic agent for HER-2/neu-overexpressing ovarian cancer. Cancer Gene Therapy 3, 168-174. Yamamoto, T. M., Ikawa, S., Akjiyana, T., Semba, K., Normura, N., Miyajima, N., Saito, T., and Toyoshiman, K. (1 9 8 6 ). Similarity of protein encoded by the human cerbB-2 gene to the epidermal growth factor receptor. Nature 319, 230-234. Yan, D. H., Chang, L. S., Hung, M. -C. (1991). Repressed expression of the HER-2/neu/c-erbB-2 proto-oncogene by the adenovirus E1A gene products. O n c o g e n e 6, 343345. Yokota, J., Yamamoto, T., Miyajima, N., Toyoshima, K., Nomura, N., Sakamoto, H., Yoshida, T., Terada, M., and Sugimura, T. (1 9 8 8 ). Genetic alterations of the c-erbB-2 oncogene occur frequently in tubular adenocarcinoma of the stomach and are often accompanied by amplification of the v-erbA homologue. O n c o g e n e 2, 283-287. Yu, D., Liu, B., Jing, T., McDommell, T. J., Sun, D., ElDeiry, W. S., and Hung, M. -C. 1 9 9 8 a ). Overexpression of c-erbB2 blocks Taxol-induced apoptosis by

100

upregulation of p21 cip1 which inhibits p34 cdc2 kinase. M o l e c u l a r C e l l , in press. Yu, D., Liu, B., Jing, T., Sun, D., Price, J. E., Singletary, S. E., Ibrahim, N., Hortobagyi, G. N., and Hung, M. -C. (1 9 9 8 b ). Overexpression of both p185 c-erbB-2 and p170 mdr-1 renders breast cancer cells highly resistant to taxol. O n c o g e n e 16, 2087-2094. Yu, D., Liu, B., Tan, M., Li, J., Wang, S. -S., and Hung, M. C. (1 9 9 6 ). Overexpression of c-erbB-2/neu in breast cancer cells confers increased resistance to Taxol via mdr1-independent mechanisms. O n c o g e n e 13, 1359-1365. Yu, D., Matin, A., Xia, W., Sorgi, F., Huang, L., and Hung, M. -C. (1 9 9 5 ). Liposome-mediated E1A gene transfer as therapy for ovarian cancers that overexpress HER-2/neu. O n c o g e n e 11, 1383-1388. Yu. D, Wang, S. S., Dulski, K. M., Tsai, C. -M., Nicolson, G. L., and Hung, M. -C. (1 9 9 4 ). c-erbB2/neu overexpression enhances metastatic potential of human lung cancer cells by induction of metastasis-associated properties. Cancer Res. 54, 3260-3266. Yu, D., Shi, D., Scanlon, M., and Hung, M. -C. (1 9 9 3 a ). Reexpression of neu-encoded oncoprotein counteracts the tumor-suppressing activity of E1A. Cancer R e s . 53, 5784-5790. Yu, D., Wolf, J. K., Scanlon, M., Price, J. E., and Hung, M. C. (1 9 9 3 b ). Enhanced c-erbB-2/neu expression in human ovarian cancer cells correlates with more severe malignancy that can be suppressed by E1A. Cancer R e s . 53, 891-898. Yu, D., Hamada, J., Zhang, H., Nicolson, G. L., and Hung, M. -C. (1 9 9 2 a ). Mechanisms of c-erbB2/neu oncogeneinduced metastasis and repression of metastatic properties by adenovirus 5 E1A gene products. O n c o g e n e 6, 22632270. Yu, D., Matin, A., and Hung, M. -C. (1 9 9 2 b ). The retinoblastoma gene product suppresses neu oncogeneinduced transformation via transcriptional repression of neu*. J . B i o l . C h e m . 267, 10203-10206. Yu, D., and Hung, M. -C. (1 9 9 1 a ). Expression of activated rat neu oncogene is sufficient to induce experimental metastasis in NIH3T3 cells. O n c o g e n e 6, 1991-1996. Yu, D., Scorsone, K., and Hung, M. -C. (1 9 9 1 b ). Adenovirus Type 5 E1A products acts as transformation suppressors of the neu oncogene. M o l . C e l l B i o l . 11, 1745-1750. Yu, D., Suen T. C., Yan, D. H., Chang, L. S., and Hung, M. -C. (1 9 9 0 ). Transcriptional repression of the neu protooncogene by the Adenovirus 5 E1A gene products. P r o c . N a t l . A c a d . S c i . U S A 87, 4499-4503. Zhang, M., Maass, N., Magit, D., and Sager, R. (1 9 9 7 a ). Transactivation through Ets and Ap1 transcription sites determines the expression of the tumor-suppressing gene maspin. C e l l G r o w t h D i f f e r e n t i a t i o n 8, 179-86. Zhang, M., Magit, D., and Sager, R. (1 9 9 7 b ). Expression of maspin in prostate cells isregulated by a positive Ets

Gene Therapy and Molecular Biology Vol 3, page 101 element and a negative hormonal reponsive element site recognized by androgen receptor. P r o s . N a t l . Acad. S c i . U S A 94, 5673-5678. Zhang, Y. J., Yu, D. H., Xia, W. Y., and Hung, M. C. (1 9 9 5 ). HER-2/neu-targeting cancer therapy via adenovirusmediated E1A delivery in an animal model. O n c o g e n e 10, 1947-1954. Zou, Z., Anisowicz,A., Hendrix, M. J., et al, (1 9 9 4 ). Identification of a novel serpin with tumor suppressing activity in human mammary epithelial cells. S c i e n c e 263, 526-529.

101

Gene Therapy and Molecular Biology Vol 3, page 103 Gene Ther Mol Biol Vol 3, 103-112. August 1999.

Gene therapy targeting p53 Review Article

John Nemunaitis1,2 1

PRN Research, Inc., Dallas, Texas. 2Baylor University Medical Center, Dallas, Texas

__________________________________________________________________________________________________ Correspondence: John Nemunaitis, M.D., 3535 Worth Street, Collins Bldg., 5 th Floor, Dallas, Texas 75246. Phone: 214-8208799; Fax: 214-820-8497; E-mail: aepetro@prninc.com Received 13 October 1998; accepted: 17 October 1998

Summary The product of the p53 gene plays a critical role in the regulation of cell growth. Mutations of this gene are associated with transformation to a malignant phenotype. Correction of the gene defect through transfer of a wildtype p53 gene into malignant cells, or targeting malignant cells with oncolytic viruses (ONYX-015) genetically engineered to proliferate in cells containing mutant genes has been identified as a therapeutic approach by preclinical assessment. Initial clinical trials have confirmed functional activity and expression of the transgene product in Adp53-injected malignant tissue and tumor specific viral proliferation have been observed in patients receiving intratumoral injection of ONYX-015.

I. Introduction

II. p53 mutation

The most common genetic abnormality identified in human malignancy with an occurrence of approximately 60% involves the p53 gene, which is a tumor suppressor gene (Baker, 1990) located on chromosome 18. Disruption of p53 protein production or inhibition of its function is associated with abnormal cellular proliferation and differentiation.

Eighty percent of p53 mutations involving solid tumors are point mutations that result in a single amino acid substitution. At first glance, this may not appear to be a significant abnormality, given that the alteration involves less than 1% of the entire molecule. However, many of the aminoacid substitutions result in a charge change (i.e. positive to negative or vice versa), which dramatically alters the three-dimensional structure of the p53 protein. Once altered, receptor-binding affinity is disturbed. As a result, excess p53 protein is produced with accumulation within the nucleus. Normal cells have undetectable levels of p53 protein. Thus, elevated p53 protein expression often indicates the occurrence of a mutated p53 gene although not always (Barnes, 1992; Lehman, 1991).

Specific functions of the p53 gene product include upregulation of p21, which is a protein that inhibits cyclin-dependent kinase (CDK ), and is necessary for the G1 to S-phase transition. P53 protein also upregulates Bax (a positive regulator of apoptosis), MDM-2 (a negative regulator of p53 function), thrombospondin-1 (inhibitor of angiogenesis), GADD45 (role in DNA repair), and IGF-BP3 (growth regulator) (Harper, 1993; Miyashita, 1995; Dameron, 1994). Extensive analysis of tumors showing evidence of p53 gene dysfunction indicate that abnormal function correlates with poor prognosis in patients with malignancy (Drach, 1998; Horio, 1993; Thorlacius, 1993; Preudhomme, 1997; Lai, 1995).

Other molecules may also be produced by malignant cells which inhibit normal p53 function via binding to the p53 protein, enhancing degradation, or disruption of binding sites. One example of inactivation of p53, which may occur by interaction with another cellular protein, involves the murine double-minute-2 (MDM-2) protein which acts as a false binding site (Teoh, 1997). Another example involving induced degradation is seen in cervical cancer of the p53 protein (Caron de Fromentel, 1992; Vogelstein, 1992; Scheffner, 1992). The majority of

The purpose of this chapter is to describe data which identifies novel therapeutic approaches targeting correction of the p53 gene via transfection with a wildtype p53 gene using a replication defective adenoviral vector carrier and approaches utilizing oncolytic virus ONYX-015. 103

Nemunaitis: Gene therapy targeting p53 cervical cancers harbor the human papilloma virus (HPV), which enhance degradation of the p53 protein (Howley, 1991). Cervical cancer cells, which are HPV positive and contain the p53 mutation (less than 20%) are particularly aggressive, and such patients have an even more dismal prognosis.

the E1 and E3 regions provides empty space (~7KB) where the wildtype p53 gene sequence is inserted (Zhang, 1994). Transfection of several NSCLC cell lines and head and neck cancer cell lines reveal high expression of wildtype p53 protein (the transgene product). Optimal expression is observed at a multiplicity of infection (MOI) of 30-50 plaque-forming units (PFU) per cell (Zhang, 1995; Zhang, 1994). Maximal expression was observed 3 days after transfection and rapidly decreased over the next 5 days. Detection of the transgene product was still observed 15 days following transduction. Similar results were shown in vitro and in vivo. Transgene expression and normal function has been shown in cell lines of breast cancer, ovarian cancer, colorectal cancer, prostate cancer, the central nervous system, and bladder cancer (Harris, 1996; Wills, 1994; Lesoon-Wood, 1995; Blagosklonny, 1996; Bartek, 1990).

Poor survival prognosis has been observed in patients with cancer of the lung, colon, liver, breast, stomach, cervix, non-Hodgkin’s lymphoma, and multiple myeloma who have elevated p53 protein expression or a p53 DNA mutation detected from tumor samples prior to treatment (Drach, 1998; Horio, 1993; Thorlacius, 1993; Preudhomme, 1997). The development of p53 gene mutations may also involve environmental carcinogenic factors (Vogelstein, 1992). Malignant cells containing p53 mutations have an increased resistance to death in response to chemotherapeutic agents or ionizing radiation (Lee, 1993), and an increase in metastatic spread (Dutta, 1993). Twenty percent of patients with a p53 mutation have also been found to express antibodies to the mutant p53 protein, although it is unclear whether such patients have an altered prognosis (Crawford, 1982; Caron, 1987; Davidoff, 1992; Winter, 1992; Schlichtolz, 1992).

IV. Safety of the Advp53 vector The Adp53 vector is constructed from a serotype 5 adenovirus. A great deal of data has been accumulated suggesting the safety of this virus (Brandt, 1969). Eighty percent of adults have existing antibodies to adenovirus serotype 5 (Nicholson, 1993), but less 15% of exposed patients become clinically symptomatic. The most common symptoms of an adenoviral serotype 5 infection are flu-like in nature and include cough, gastroenteritis, conjunctivitis, cystitis, and rarely pneumonia. However, these symptoms are rarely seen even in immune compromised patients (Hierholzer, 1992). Oral adenoviral vaccines were given to thousands of military recruits in the 1960s without adverse effects or increase in cancer (Takafuji, 1979). Live adenovirus inocula was also given intratumorally and intra-arterially to patients with cervical carcinoma at the National Cancer Institute in the 1950s (Smith, 1956). No significant toxicities, other than transient fever and malaise, were observed even in subsets of patients treated with steroids and in those in which neutralizing adenovirus antibodies were not present.

In conclusion, an understanding of the p53 gene structure and protein function is important in developing therapeutic approaches, and may assist in the understanding of potential activity and toxicity to therapeutic approaches attempting to correct dysfunction of the p53 gene or protein.

III. Adp53 vector Preclinical studies have reported the introduction of the wildtype p53 gene into human tumor cells with a mutant p53 genotype using a variety of delivery methods including the retroviral vectors, lipid complexes, and adenoviral vectors (Harris, 1996; Wills, 1994; LesoonWood, 1995; Xu, 1997; Blagosklonny, 1996; Zhang, 1995; Nielsen, 1997; Nguyen, 1996). Results demonstrate that the expression of the transgene product provides a normal functioning wildtype p53 protein to the malignant cell, which has been shown to induce tumor regression and improve survival in animal models. Preclinical results also reveal enhanced activity when combined with chemotherapy (Nguyen, 1996; Fujuwara, 1994).

Work was conducted in animal models exploring the most significant serious clinical toxicity to live adenovirus (pneumonia). A unique strain of cotton rats (gigmodon hispidus) has been shown to consistently develop pulmonary infection in response to inoculation with adenovirus serotype 5 (Pacini, 1984). Pathogenicity was related to the dose of the viral inoculum. Additional safety testing has been conducted in mice and cotton rats in which high doses of adenovirus were injected locally and systemically. Animals developed minor histopathologic changes in several organs, but no pulmonary toxicity was observed (Pacini, 1984; Ginsberg, 1991). However, inflammatory infiltrates related to p53 have been observed in the lungs of animals given high doses of Adp53 directly

Vectors utilized for adenoviral introduction of the wildtype p53 gene involve wildtype adenovirus containing deletions of the E1 and E3 replication components (Zhang, 1993). Adenoviruses are single-stranded DNA viruses with genomes of approximately 35kB (Takahashi, 1989), which are easily propagated in human cells, and have been associated with minimal pathogenicity. The deletion of 104

Gene Therapy and Molecular Biology Vol 3, page 105 to the bronchial airway (Zhang, 1995; Ghosh-Choudjury, 1985; Englehardt, 1993; Rich, 1993; Ginsberg, 1990). The resulting inflammatory responses were characterized by interstitial infiltration of neutrophils, and monocytes within 1-2 days after exposure (Ginsberg, 1990; Prince, 1993). This early inflammatory process was felt to be mediated by local elaboration of various cytokines such as tumor necrosis factor, IL-1 and IL-6 (Prince, 1993). An additional inflammatory response also occurs within 3-7 days. At this time, peribronchial infiltration of lymphocytes is observed. Direct exposure of the lung with low concentrations of the adenovirus vector does not appear to be associated with pulmonary toxicity (Simon, 1993; Yei, 1994).

competent adenovirus. No replication competent adenovirus was detected, and elevated antibody formation did not inhibit gene expression with repeat injections (Tursz, 1996). Adenoviral vectors with E1 and E3 deletion containing the E-coli cytosine deaminase gene have also been administered to normal individuals to study immune response (Harvey, 1998). Six volunteers received intradermal injections of 106, 10 7, or 10 8 PFU (2 patients per group). Five of the 6 volunteers showed a rapid increase in anti-Ad5 neutralizing antibody titers above baseline. The peak antibody response occurred 2 weeks after vector injection. Erythema occurred at the site of injection with maximum induration of approximately 7mm by Day 3, and complete disappearance of induration by Day 10. Skin biopsies of the erythema revealed T-cell, B-cell and a macrophage infiltrate. Vector DNA was detected in biopsies of patients who received the 108 dose on Day 3, but no evidence of vector DNA was detected on Day 28. No systemic toxicity was observed in any of the normal volunteers (Harvey, 1998).

The possibility of adenoviral replication competency developing after vector injection also appears to be negligible, given the construction design of the vector (Zhang, 1995). However, complete inhibition of DNA replication solely from E1 deletion has not been 100% successful (Englehardt, 1993; Rich, 1993). This necessitates intense monitoring of the Adp53 clinical material for replication competency. Repeat sequencing of the product reveals that the wildtype p53 genotype does not undergo mutation changes during manufacturing. Expression of the transgene product also does not appear to be toxic. Studies performed in vitro looking at Adp53 transfection of non-malignant fibroblasts and human bronchial epithelial cells in comparison to malignant head and neck tumor cells indicate no change in p53 expression in non-malignant cells. These data suggest that normal cellular p53 expression is not altered by transfection with Adp53. The growth rate and morphology of the nonmalignant fibroblasts and bronchial epithelial cells was not altered following transfection with Adp53 (Zhang, 1995). Theoretical concerns regarding oncogenicity of adenoviruses are also unlikely. The life cycle of an adenovirus does not require integration into the host genome, thus, foreign genes delivered by adenoviral vectors are expressed episomally and have low genotoxicity (Zhang, 1995). DNA from thousands of human tumors have been analyzed for the presence of adenovirus DNA and no integrated viral DNA has been isolated from any human tumor (Green, 1979). Long- and short-term safety of adenoviral injection has been shown in several animal models (Lesoon-Wood, 1995; Zhang, 1995; Nielsen, 1997; Englehardt, 1993; Simon, 1993; Yei, 1994; Xu, 1998; Gomez-Foix, 1992; Le Gal La Salle, 1993).

Finally, if serious viral infection does develop, therapeutic approaches are available. Wildtype adenovirus dissemination has been seen in organ transplant recipients, however, in most cases, the viremia has been eliminated with the use of intravenous Ribavirine (Liles, 1979), although occasionally Ribavirine has not been successful (Mirza, 1994).

V. Preclinical studies with Adp53 Early preclinical studies with Adp53 vector in lung cancer initially utilized the H358 cell line. In one study, 50 mice received injections of 2 x 106 H358 cells, which had been previously transfected with Adp53 in vitro. Eighty percent of control animals developed tumors within 2-3 weeks; however, none of the p53 transfected cells evolved into malignant lesions 6 weeks after injection. Other work with the Adp53 vector involved the use of H326 cells which were derived from a highly aggressive squamous cell NSCLC lesion. This cell line contains a p53 point mutation (Zhang, 1994; Georges, 1993). Inoculation of 2 x 106 H326 cells into the trachea of mice followed by inoculation with Adp53 vector, control vector, or control vehicle, reveals that only 2 of the 8 Adp53treated mice developed tumors 6 weeks after treatment with a mean tumor volume of 8mm3, whereas 7 of 10 of the treated mice, and 8 of 10 of the control vector treated mice developed tumors where the mean volume exceeded 30mm3 within 6 weeks after inoculation. Subsequent approaches exploring the use of Adp53 in combination with Cisplatin revealed enhanced activity.

In humans, (-GAL vector injection was administered to patients with endobronchial lung cancer. Evidence of replication competent adenovirus was studied in caretaker staff samples. Specifically, 73 staff provided 78 blood samples, 272 urine samples, and 193 samples to study antibody formation or the presence of replication 105

Nemunaitis: Gene therapy targeting p53 Animal models have been designed to test whether transfection of head and neck cancer cells with Adp53 may alter response to radiation, chemotherapy or have direct effects. In one model, Adp53 was transfected into a radioresistent human cell line GSQ-3 (squamous cell carcinoma of head and neck). Wildtype p53 protein was shown to be expressed in high levels and have functional activity in the transfected cells (Xu, 1998). A dose of 108 PFU was shown to be sufficient to induce tumor regression without evidence of systemic toxicity (Liu, 1994; Yamamoto, 1998). Animal studies in other tumor xenograph models (ovarian, breast, prostate) have also shown activity following Adp53 injection (Sheikh, 1995; Eastham, 1995; Mujoo, 1996).

endothelial cells, airway epithelial cells, and mammary epithelial cells. Wildtype adenovirus showed cytopathic effects at a MOI as low as 0.01 virus particles within 8-10 days, whereas cytopathic effects of ONYX-015 virus were not observed until MOIs of >100 virus particles were achieved. Thus, safety and antitumor activity appear to be related to the dose of virus infused. Several studies involving oncolytic viruses other than ONYX-015 have been performed in vitro and in vivo in human patients without significant toxicity (Kenney, 1994; Russell, 1994; Asada, 1974; Smith, 1956). Unfortunately, the difficulties in characterizing viral load led to inconsistent results and there was no suggestion of efficacy. Preclinical studies with the ONYX-015 virus in vivo were performed to confirm direct tumor cell lysis through local injection and systemic infusion, and to determine whether or not tumor lysis is observed in response to viral replication (Yang, 1994).

VI. ONYX-015 preclinical studies P53 protein mediates cell cycle arrest via apoptosis if foreign DNA synthesis is occurring within a cell from viral replication (Debbas, 1993; Grand, 1994; Lowe, 1997). DNA tumor viruses such as certain adenoviruses, SV40 and human papilloma virus incode proteins which inactivate p53, thereby allowing their own replication (Debbas, 1993; Lechner, 1992; Gannon, 1987). Specifically, a 55dDa protein from the E1B region of adenovirus serotype 5 binds and inactivates p53 (Barker, 1987). Inability to block p53 function with deletion of the E1B region would enable the p53 protein to maintain its function thereby inhibiting viral replication. The ONYX-015 virus is a DNA adenovirus which was constructed with an E1B deleted region so that it no longer produces the 55kDa protein. In this manner, the virus would not be expected to proliferate in normal cells, but it would be expected to have extensive proliferative capacity in tumor cells which are either p53 mutant or have disrupted p53 function (Bischoff, 1996).

In animal human xenographt studies, intratumor injection of ONYX-015 virus has been tested in cervical cancer (C33 cervical carcinoma cells) and head and neck cancer (HLaC laryngeal carcinoma cells), both of which have a p53 functional deficiency (Heise, 1997). Significant tumor growth inhibition was observed compared to controls. Mice achieving a complete response remained disease-free for 4-6 months before sacrifice. U87 glioblastoma tumors, which do not have a p53 mutation, were not affected by injection with the ONYX-015 virus. Evidence of viral proliferation based on histochemical staining for adenovirus exon protein was confirmed in the sensitive tumors, but not in the U-87 tumors. Additional studies comparing vehicle versus chemotherapy (5-FU or Cisplatin), ONYX-015 alone, or ONYX-015 plus chemotherapy, were carried out (Heise, 1997). Median survival in mice receiving ONYX-015 plus 5-FU was further improved compared to control or ONYX-015 alone. Similar results were seen in combination with Cisplatin.

Initial studies testing the ONYX-015 virus involved incubation of virus with RKO human colon cancer cell lines which have normal p53 function and a subcloned line of RKO, which has a mutant p53 gene. The ONYX-015 virus replicated as efficiently as the wildtype adenovirus in the subclone lacking functional p53 protein, however, the cytopathic effects of ONYX-015 are reduced by 2 logs in the parent tumor line harboring normal p53 function (Bischoff, 1996). Cell lines resistant to ONYX-015 have also been made sensitive through transfection and expression of the E1B 55dDa gene (Bischoff, 1996). Cytopathic effects of ONYX-015 have also been shown in other malignant cells, which have abnormal p53 function, involving the breast, cervix, colon, central nervous system, liver, ovary, pancreas and head and neck region (Heise, 1997). Potential infectivity of ONYX-015 was tested against wildtype adenovirus by infecting nonmalignant (normal p53 functioning) human microvascular

Systemic injections of ONYX-015 at a dose of 108 PFU were also injected for 10 days into the tail vein of nude mice implanted with C33-a or HCT116 human xenographt tumors. Tumor growth was significantly inhibited in the C33-a tumors with ONYX-015 treatment by 55% compared with growth in mice injected with vehicle solution (p=0.004). Comparison of intravenous ONYX-015 virus (IV for 5 days) plus 5-FU (IP for 5 days) in mice showed that 6 of 7 mice had complete tumor regression following the combination, whereas only 2 of 7 mice achieved complete tumor regression following 5-FU treatment alone. The median tumor volume on day 40 was 93(L in the mice receiving ONYX-015 plus 5-FU. However, mice receiving 5-FU alone had a median tumor volume of 461(L, compared to ONYX-015 alone with a tumor volume of 671(L, and saline alone with a tumor volume of 748(L. No significant toxicity was observed. 106

Gene Therapy and Molecular Biology Vol 3, page 107 Results suggest that both intratumor and intra-venous infusion of ONYX-015 when combined with chemotherapy was safe and effective in inducing tumor regression and prolonging survival.

response (( 2-fold increase) was shown in 19 of 20 evaluable patients following course 1. Cytopathic effect assays (CPE) also revealed the presence of Adp53 vector in plasma within 30 minutes of intratumor injection in all 16 patients tested. Tumor biopsies collected 3 days posttreatment demonstrated p53 transgene expression by RTPCR in 10 of 17 (58%) patients receiving vector dose levels ( 3 x 1010 PFU, and only 8 of 27 (30%) patients who received the lower dose level. Toxicity attributed specifically to the vector was limited to transient fever and nausea. Cisplatin-related toxicity was not observed in any greater frequency than it would be expected when Adp53 gene vector was not combined with Cisplatin. Four patients fulfilled a definition of partial response (PR) (8%), 33 patients (64%) experienced stable disease for a transient period of time (minimum 1 month), 11 patients (20%) had progressive disease, and 4 patients (8%) were not evaluable for response (Nemunaitis, 1998; Swisher, in preparation; Nemunaitis, 1998). Overall, median survival was 149 days. The difference in survival between the patients who received Cisplatin or Adp53 + Cisplatin did not achieve statistical significance. Six of 12 patients with endobronchial-injected lesions had sufficient tumor regression to open obstructed airways.

VII. Human studies with Adp53 The first trial published to explore gene transduction of the p53 gene via intratumor injection in humans utilized a retroviral vector. In this trial, 9 patients (median age 68) with NSCLC were treated (Roth, 1996). Four received retrovector p53 gene via bronchoscopic injection, and 5 were treated via a percutaneous injection with CT guidance. Eight of the 9 patients treated had a point mutation, and 1 had a frame shift mutation of the p53 gene. Vector transduction was confirmed in 8 patients by PCR analysis, and 6 patients showed induction of apoptosis (TUNEL assay). Three patients showed evidence of tumor regression (all 3 of these patients received endobronchial injections). No toxic effects were attributed to the vector. Retroviral sequences were not detected in non-injected pulmonary tissue analyzed by PCR, and no evidence of replication competent retrovirus was detected. Unfortunately, low transduction efficiency associated with the retroviral vector was a major limiting factor.

The conclusion of this trial is that Adp53 endobronchial or CT-guided injections at a dose of 1011 PFU in patients with NSCLC are safe and well tolerated. The maximum tolerable dose of the vector has not been reached. This therapy can be administered monthly, alone or with Cisplatin with no increase in Cisplatin-related toxicity. Immune response to the Adp53 vector does not limit continued injections, and there is evidence of objective activity and clinical benefit.

Several studies with Adp53 were subsequently initiated. One Phase I trial investigating tolerability of Adp53 in NSCLC was recently completed. Fifty-two patients with advanced NSCLC who had previously failed conventional treatment were entered into trial (Swisher, 1998). Adp53 doses were escalated from 10 6 to 10 11 PFU and injected monthly into a single primary or metastatic tumor by bronchoscopy (12 patients) or computed tomographic (CT) guidance (40 patients). Patients were treated by direct assignment with or without Cisplatin (80mg/m 2) given IV over 2 hours prior to Adp53 injection. Each patient received up to 6 courses of treatment and median follow-up was 9.9 months. Vectorspecific deoxyribonucleic acid (DNA) was detected by PCR, and p53 transgene expression was determined by reverse transcriptase PCR and immunohistochemistry. Vector was present in plasma within 30 minutes of injection, and decreased in the next 60 minutes (Timmons, 1998). No replication competent adenovirus was detected in any body fluids tested. Antibody titers increased in patients receiving at least 2 doses and remained elevated for several months after completion of injections. In patients who received Cisplatin, the apoptotic index increase from 0.124 to 0.034 (p=0.011) when compared to baseline in samples harvested after the first course of Adp53 injection. The TUNEL assay showed an increase in the number of apoptotic cells in 11 of the 15 evaluable patients, a decrease in 2 patients, and no change in 2 patients (Nemunaitis, 1998). Anti-adenoviral type 5 IgG antibody

Additional work exploring the same Adp53 vector was done in head and neck cancer (Clayman, in press). In this trial, patients with recurrent or refractory squamous cell carcinoma of the head and neck region with a performance status of 0-2 were eligible for trial. Results of this trial concluded that repeated intratumoral injections of up 1011 PFU was safe and well tolerated. Transgene expression occurred despite evidence of adenovirus antibody response. Peri- and post-operative Adp53 injection had no adverse effect on surgical morbidity and/or wound healing. Evidence of activity based on tumor regression following injection of Adp53 was observed (1 CR, 2 PRs) (Clayman, in press; Wilson, 1998). Others have explored the use of Adp53 vectors in head and neck cancer and other tumor types such as colon cancer and ovarian cancer. In another Phase I trial using a different Adp53 vector (SCH-58500), 16 patients with head and neck cancer received escalated doses ranging from 7.5 x 109 PFU to 7.5 x 10 12 PFU (charts of patients received single or multiple intratumor injections). The median age of patients entered into this trial was 60.5 years. Ten of 16 107

Nemunaitis: Gene therapy targeting p53 patients had elevated serum IgG to p53 protein following injection, and p53 transgene expression was confirmed in a subset of patients. Toxicity attributed to the vector was limited to Grade 1/2 fever (11 patients) and injection pain (6 patients). One patient achieved a PR which correlated with the induction of apoptosis and transgene expression (Agarvala, 1998).

(Kirn, 1998). Injections were given throughout the perimeter of the tumor, and the volume of the injected medium was normalized to 30% of the target tumor volume. Neutralizing antibodies were found in 10 of 20 Phase II treated patients prior to injection, and the p53 gene sequence was mutated in 7 of 13 patients. There was also a suggestion of increased response in patients with tumor sized of (5cm in diameter. Thirty-seven percent of patients with tumor (5cm achieved a complete response or partial response compared to 0% of patients with tumor >5cm (n=30). The most frequent side effect observed in the Phase II trial was pain at the injection site and it occurred in 32% of patients. Transient fever and chills occurred in 28%, nausea in 8%, and confusion in 4% of patients. Despite these preliminary results and with the trial not yet completed, results are sufficient to determine that the ONYX-015 virus is well tolerated at a dose of 1010 PFU given to 5 consecutive day every 3 weeks. Subsequent studies exploring ONYX-015 virus (1 x 1010 PFU daily x 5 days every 3 weeks) combined with chemotherapy (Cisplatin 100mg/m2, IV on day 1; and 5FU 800-1,000mg/m2 by continuous infusion per day on days 1-5 every 3 weeks) were thus initiated. Patients with recurrent head and neck cancer who had not previously been exposed to chemotherapy or radiotherapy in the recurrent tumor setting were entered into trial. At the time of the preliminary analysis (Kirn, 1998), 10 patients had been treated and 9 of 10 patients achieved a partial response or complete response. Despite being preliminary, the data is very encouraging particularly when compared to expected response rates, in which similar patients receiving chemotherapy without ONYX-015 virus would be expected to achieve a 30-40% partial or complete response rate, and would not be expected to have a median survival >9 months.

Another trial utilizing SCH-58500 was performed in patients with colorectal cancer with liver metastasis. In this trial, 16 patients received hepatic arterial infusion of Adp53 vector. A single dose was administered prior to laparotomy. Patients received escalating dose levels ranging from 7.5 x 109 PFU to 2.5 x 10 12 PFU. Adverse events included fever in 15 of 16 patients, and headache in 3 of 16 patients. Transgene expression was confirmed in normal liver and tumor. No responses specifically attributed to the Adp53 therapy alone were observed, however, 12 patients subsequently received FUDR and 11 achieved a 50% reduction in disease, suggesting the potential for sequential therapeutic approaches to be considered in trial designs utilizing Adp53 (Agarvala, 1998). SCH58500 was also given to 18 patients with advanced NSCLC. Patients received escalating doses ranging from 107 to 10 10 PFU. No serious adverse events were observed. Only one patient required hospitalization for prolonged persistent flu-like symptoms. Transgene expression was confirmed in patients who received higher dose levels. In 4 of the 6 patients who showed evidence of wildtype p53 expression, progression of transient local disease was stabilized following injection with Adp53 (Schuler, 1998).

VIII. Human trials with ONYX-015 virus

These preliminary results suggest that ONYX-015 replicates in recurrent refractory head and neck cancer, and that ONYX-015 is well tolerated following intratumor injection alone, or when combined with chemotherapy.

Several trials with ONYX-015 virus in treatment of head and neck cancer were recently reported. These trials suggested that ONYX-015 is well tolerated except for transient low-grade fever and that antitumor activity is observed.

ONYX-015 is also being explored at escalating dose levels in patients with gastrointestinal tumors metastatic to liver (Bergsland, 1998). Patients with metastatic disease to the liver were administered intratumoral injections through CT guidance. The starting dose level was 1 x 108 PFU. Injections were given one time every 21 days. Patients not showing progressive disease were eligible for continued injections. A total of 16 patients and 29 injections had been administered at the time of this preliminary analysis, and the dose level of 1 x 108 PFU was reached without evidence of dose-limiting toxicity. Minor toxicities such as flu-like symptoms were observed in 11 patients, transient elevation and coagulation times were observed in 7 patients, lymphopenia in 5, and transient liver function enzyme elevations was observed in

Preliminary Phase I studies indicated that intratumor ONYX-015 injections are well tolerated and viral proliferation has been confirmed in malignant cells by electron microscopy. The duration of tumor response appeared to be greater in patients receiving multiple injections compared to a single injection per cycle (every 21 days). The optimal dose suggested for Phase II investigation was 1 x 1010 PFU given for 5 days every 21 days (unpublished results). Phase II studies performed in refractory head and neck patients utilized a dose of 1 x 10 10 PFU of ONYX-015 daily x 5 days every 3 weeks via intratumor injection 108

Gene Therapy and Molecular Biology Vol 3, page 109 4 patients. Response assessment after cycle 1 revealed 2 patients with minor responses, 9 patients with stable disease, and 4 patients with progressive disease. This is an ongoing trial in which patients are continuing to receive injections, and thus far it can be concluded that the treatment is well tolerated, although evidence of activity remains to be determined.

Overall, preliminary results of Phase I studies indicate that the p53 gene transfer through intratumoral injection using replication vectors is well tolerated, associated with antitumor activity at dose levels equal to and above 1 x 109 PFU. Data also suggest that administration of multiple injections and combination with chemotherapy or radiotherapy may enhance the overall antitumor effect. Phase II trials to determine efficacy are ongoing.

Others have also performed Phase I exploration of ONYX-015 in patients with unresectable carcinoma of the pancreas (Mulvihill, 1998). In another trial, escalating doses of ONYX-015 were administered to patients with to patients with unresectable pancreatic cancer. Sixteen patients received a total of 36 injections. At baseline, 5 of 13 tumors assessed contained mutant p53 gene sequences, and 9 of 10 patients had neutralizing antiadenoviral antibodies. All patients showed escalation of antiadenoviral antibodies following injection. One patient developed Grade 3 hyperbility rubimenia following the injection, otherwise no other Grade 3-4 toxicities were observed at dose levels up to 10 10 PFU. Grade 1-2 flu-like symptoms were reported in all patients. Four patients had minor regressions following the initial cycle of treatment with a 35-45% decrease in disease, 7 patients had stable disease, and 3 patients had progressive disease. Two patients reported a decrease in pain following injection. Preliminary conclusions are that the intratumor injection of ONYX-015 was well tolerated. Continued injections are ongoing.

Acknowledgment The author thanks Ana Petrovich for the manuscript preparation.

References Agarvala SS, van Osterom A, Petruzelli G, et al. (1 9 9 8 ). Phase I study of rad/p53 in patients with advanced head and neck cancer (HNC). Proc ASCO 7, 384a (abstr. 1470). Asada T. (1 9 7 4 ). Treatment of human cancer with mumps virus. Cancer 34, 1907-1928. Baker SJ, Markowitz S, Fearon E, et al. (1 9 9 0 ). Suppression of human colorectal carcinoma cell growth by wildtype p53. S c i e n c e 249, 912-915. Barker DD, Berk AJ. (1 9 8 7 ). Adenovirus proteins from both E1B reading frames are required for transformation of rodent cells by viral infection and DNA transfection. V i r o l 156, 107-121.

IX. Conclusions

Barnes DM, Hanby AM, Gillett CE, et al. (1 9 9 2 ). Abnormal expression of wildtype p53 protein in normal cells of a cancer family patient. Lancet 340, 259-263.

Results of clinical trials performed are encouraging and shown good tolerability to a variety of Adp53 vectors and confirm that the transgene product expressed from the transfected vector is functional and associated antitumor activity in small numbers of patients. Unfortunately, therapy at this time is limited to direct intratumor injection. If immunologic difficulties leading to vector neutralization can be overcome, safety data suggest that systemic infusion of Adp53 vector may be well tolerated. Studies to limit immunoreactivity to the Adp53 vector through inhibition of the immune response or alteration of the vector or other gene transfer vehicles are ongoing. For instance, using a ligand lyposome complex, wildtype p53 gene was efficiently delivered both in vitro and in vivo in murine squamous cell head and neck cancer models. Injection of the ligand/lyposome complex with the wildtype p53 gene was shown to be taken up in both head and neck and prostate tumors. Transfection was higher in malignant tissue than surrounding normal tissue. Furthermore, enhanced activity was shown following treatment with radiotherapy after ligand/lyposome encapsulated wildtype p53 injection or IV infusion (Pirollo, 1998), without significant toxicity (Joshi, 1998).

Bartek J, Iggo R, Gannon J, et al. (1 9 9 0 ). Genetic and immuno-chemical analysis of mutant p53 in human breast cancer cell lines. O n c o g e n e 5, 893-899. Bergsland E, Mani S, Kirn D, et al. (1 9 9 8 ). Intratumoral injection of ONYX-015 for gastrointestinal tumors metastatic to the liver, A Phase I trial. Proc ASCO 17, 211a (abstr. 814). Bischoff JR, et al. (1 9 9 6 ). An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. S c i e n c e 274, 373-376. Blagosklonny MV, El-Diry WS. (1 9 9 6 ). In vitro evaluation of a p53-expressing adenovirus as an anti-cancer drug. Int J Cancer 67, 386-392. Brandt CD, et al. (1 9 6 9 ). Infections in 18,000 infants and children in controlled study of respiration tract disease. Adenovirus pathogenicity in relation to serologic type and illness syndrome. Am J Epidem 90, 484-500. Caron de Fromentel C, May-Levin F, Mouriesse H, et al. (1 9 8 7 ). Presence of circulating antibodies against cellular protein p53 in a notable proportion of children with B-cell lymphoma. Int J Cancer 39, 185-189.

109

Nemunaitis: Gene therapy targeting p53 Caron de Fromentel C, Soussi T. (1 9 9 2 ). TP53 tumor suppressor gene, A model for investigating human mutagenesis. Genes Chromosome Cancer 4, 1-15.

Gomez-Foix AM, Coats WS, Baque S, et al. (1 9 9 2 ). Adenovirus-mediated transfer of the muscle glycogen phosphorylate gene into hepatocytes confers altered regulation of glycogen. J B i o l Chem 267, 2512925134.

Clayman G, El-Naggar A, Lippman S, et al. (i n p r e s s ). Adenoviral-mediated p53 gene transfer in patients with advanced recurrent head and neck squamous cell carcinoma. J Clin Oncol.

Grand RJ, Grant ML, et al. (1 9 9 4 ). Enhanced expression of p53 in human cells infected with mutant adenoviruses. V i r o l 203, 229-240.

Crawford LV, Pim DC, Bulbrook RD. (1 9 8 2 ). Detection of antibodies against the cellular protein p53 in sera from patients with breast cancer. Int J Cancer 30, 403-408.

Green M, et al. (1 9 7 9 ). Analysis of human tonsil and cancer DNAs and RNAs for DNA sequences in group C (serotypes 1, 2, 5 and 6) human adenoviruses. P r o c N a t l A c a d S c i USA 76, 6606-6610.

Dameron K, Vopert OV, Tainsky MA et al. (1 9 9 4 ). Control of angiogenesis in fibroblast by p53 regulation of thrombospondin-1. S c i e n c e 265, 1582-1584.

Harper JW, Adami GR, Wei N, et al. (1 9 9 3 ). The p21 Cdkinteracting protein Cip2 is a potent inhibitor of G1 cyclin-dependent kinase. C e l l 75, 805-816.

Davidoff AM, Inglehart JD, Marks JR. (1 9 9 2 ). Immune response to p53 is dependent upon p53/HSP70 complexes in breast cancers. P r o c N a t l A c a d S c i U S A 89, 34393442.

Harris MP, Sutjpto S, Wills KN, et al. (1 9 9 6 ). Adenovirusmediated p53 gene transfer inhibits growth of human tumor cells expressing mutant p53 protein. Cancer Gene Ther 2, 121-129.

Debbas M, White E. (1 9 9 3 ). Wildtype p53 mediates apoptosis by E1A, which is inhibited by E1B. Ge ne s D e v 7, 546-554.

Harvey BG, Worgall S, Ramirez M, et al. (1 9 9 8 ). Host responses to intradermal administration of a first generation replication deficient adenovirus vector to normal individuals. Am Soc Gene Ther Abstr 167 p, 43a.

Drach J, Ackermann J, Fritz E, et al. (1 9 9 8 ). Presence of p53 gene deletion in patients with multiple myeloma predicts for short survival after conventional-dose chemotherapy. B l o o d 3, 802-809.

Heise C, Sampson-Johannes A, Williams A, et al. (1 9 8 7 ). ONYX-015, an E1B gene-attenuated adenovirus, causes tumor-specific cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents.

Dutta A, Ruppert JM, Aster JC, et al. (1993). Inhibition of DNA replication factor PRA by p53. Nature 365, 79-82. Eastham JA, Hall SJ, Sehgal I, et al. (1 9 9 5 ). In vivo gene therapy with p53 or p21 adenovirus for prostate cancer. Cancer Res 55(22), 5151-5155.

Hierholzer JC. (1 9 9 2 ). Adenoviruses in the immunocompromised host. C l i n M i c r o b i o l R e v 5, 262-274.

Englehardt JF, Simon RH, Yang Y, et al. (1 9 9 3 ). Adenovirus mediated transfer of the CFRT gene to lung of non-human primates, Biological efficacy study. Human Gene Ther 4, 759-769.

Horio Y, Takahashi T, Kuroishi T, et al. (1 9 9 3 ). Prognostic significance of p53 mutations and 3p deletion in primary resected non-small cell lung cancer. Cancer Res 53, 1.

Fujuwara T, Grimm EA, Mukhopadhyay T, et al. (1 9 9 4 ). Induction of chemosensitivity in human lung cancer cells in vivo by adenovirus-mediated transfer of the wildtype p53 gene. Cancer Res 54, 2287-2291.

Howley PM. (1 9 9 1 ). Role of the human papilloma viruses in human cancer. Cancer Res 51, 5019s-5022s. Joshi US, Chen YQ, Kalemkerian GP, et al. (1 9 9 8 ). Inhibition of tumor cell growth by p21 WAF1 adenoviral gene transfer in lung cancer. Cancer Gene Ther 5(3), 183-191.

Gannon JV, Lane DP. (1 9 8 7 ). p53 and DNA polymerase alpha compete for binding to SV40 T antigen. Nature 329, 456-458.

Kenney S, Pagano JS. (1 9 9 4 ). Viruses as oncolytic agents, A new age for “therapeutic, viruses? J Nat Cancer I n s t 86, 1185-1186.

Georges RN, Mukhopadhyay T, Zhang Y, et al. (1 9 9 3 ). Prevention of orthotopic human lung cancer growth by intratracheal instillation of a retroviral antisense K-ras construct. Cancer Res 53, 1743-1746.

Kirn D, Nemunaitis J, Ganly M, et al. (1 9 9 8 ). A Phase II trial of intratumor injection with an E1B-deleted adenovirus, ONYX-015, in patients with recurrent refractory head and neck cancer. Proc ASCO 1, 391a (abstr. 1509).

Ghosh-Choudjury G, Haj-Ahmad Y, Brinkely P, et al. (1 9 8 5 ). Human adenovirus cloning vectors based on infectious bacterial plasmid. Gene 50, 161-171.

Lai JL, Preudhomme C, Zandecki M, et al. (1 9 9 5 ). Myelodysplastic syndromes and acute myeloid leukemia with 17p deletion. An entity characterized by specific dysgranulopoiesis and a high incidence of p53 mutations. Leukemia 9, 370.

Ginsberg HS, Horswood RL, Chanock RM, et al. (1 9 9 0 ). Role of early genes in the pathogenesis of adenovirus pneumonia. Proc Natl Acad Sci USA 87, 6191-6195. Ginsberg HS, Moldawer LL, Sehgal PB, et al. (1 9 9 1 ). A mouse model for investigating the molecular pathogenesis of adenovirus pneumonia. Proc Natl Acad Sci USA 88, 1651-1655.

Le Gal La Salle G, Robert JJ, Bernard S, et al. (1 9 9 3 ). An adenovirus vector for gene transfer into neurons and glia in the brain. S c i e n c e 259, 988-990.

110

Gene Therapy and Molecular Biology Vol 3, page 111 Lechner MS et al. (1 9 9 2 ). Human papilloma virus E6 proteins bind p53 in vivo and abrogate p53-mediated repression of transcription. EMBOJ 11, 3045-3052.

combination of sequential systemic Cisplatin and adenovirus-mediated p53 gene transfer. J Thor Cardiovasc Surg 112(5), 1372-1377.

Lee JM, Bernstein A. (1 9 9 3 ). P53 mutations increase resistance to ionizing radiation. P r o c N a t l A c a d S c i USA 90, 574205746.

Nicholson F. (1 9 9 3 ). Introduction to adenoviruses, overview of morphology, classification epidemiology. Eye (suppl.) 104.

An and

Lehman TA, Bennett WP, Metcalf RA, et al. (1 9 9 1 ). P53 mutations, ras mutations, and p53-heat shock 70 protein complexes in human lung carcinoma cell lines. Cancer R e s 51, 4090-4096.

Nielsen LL, Dell J, Maxwell E, et al. (1 9 9 7 ). Efficacy of p53 adenovirus-mediated gene therapy against human breast cancer xenografts. Cancer Gene Ther 4(2), 129-138. Pacini DL, Dubovi EJ, Clyde WA. (1 9 8 4 ). A new animal model for human respiratory tract disease due to adenovirus. J I n f e c t D i s e a s e 150, 92-97.

Lesoon-Wood LA, Kim WH, Kleinman HK, et al. (1 9 9 5 ). Systemic gene therapy with p53 reduces growth and metastases of a malignant human breast cancer in nude mice. Human Gene Ther 6, 395-405.

Pirollo KF, Xu L, Rait A, et al. (1 9 9 8 ). Tumor-targeted sensitization to radiotherapy by systemically delivered WTp53. Am Soc Gene Ther, Abstr. 379, p, 95a.

Liles WC, et al. (1 9 7 9 ). Severe adenoviral nephritis following bone marrow transplantation, Successful treatment with intravenous Ribavirine. B o n e Marrow Transplant 12, 409-412.

Preudhomme C, Fenaux P. (1 9 9 7 ). The clinical significance of mutations of the p53 tumor suppressor gene in hematological malignancies. Br J Haematol 98, 502.

Liu TJ, Zhang WW, Taylor DL, et al. (1 9 9 4 ). Growth suppression of human head and neck cancer cells by the introduction of a wildtype p53 gene via a recombinant adenovirus. Cancer Res 54, 3662-3667.

Prince GA, Porter DD, Jenson AB, et al. (1 9 9 3 ). Pathogenicity of adenovirus type 5 pneumonia in cotton rats (Sigmond hispidus). J Virol 67, 101-111.

Lowe SW, Ruley HE, Jacks T, et al. ( ). p53-dependent apoptosis modulated the cytotoxicity of anticancer agents.

Rich DP, Couture M, Cardoza LM, et al. (1 9 9 3 ). Development and analysis of recombinant adenoviruses for gene therapy of cystic fibrosis. Human Gene Ther 4, 461-473.

Mirza N, et al. (1 9 9 4 ). Adenovirus infections in adult bone marrow transplant (BMT) recipients. 33 rd ICAAC (Abstr.).

Roth JA, Nguyen D, Lawrence DD, et al. (1 9 9 6 ). Retrovirusmediated wildtype p53 gene transfer to tumors of patients with lung cancer. Nature Med 2(9), 985-999.

Miyashita T, Reed JC. (1 9 9 5 ). Tumor suppressor p53 is a direct transcriptional activator of the human bax gene. C e l l 80, 293-299.

Russell SJ. (1 9 9 4 ). Replicating vectors for gene therapy of cancer, Risks, limitations and prospects. Eur J Cancer 30A, 1165-1171.

Mujoo K, Maneval DC, Anderson SC, et al. (1 9 9 6 ). Adenoviral-mediated p53 tumor suppressor gene therapy of human ovarian carcinoma. O n c o g e n e 12, 16171623.

Scheffner M, Werness BA, Huibregtse JM, et al. (1 9 9 2 ). Mutations of p53 and ras gene in radon-associated lung cancer from uranium miners. Lancet 339, 576-580. Schlichtolz B, Legros Y, Gillet D, et al. (1 9 9 2 ). The immune response to p53 in breast cancer patients is directed against immuno-dominant epitopes unrelated to the mutational hot spot. Cancer Res 52, 6380-6384.

Mulvihill SJ, Warren RS, Fell S, et al. (1 9 9 8 ). A Phase I trial of intratumoral injection with and E1B-attenuated adenovirus, ONYX-015, into unresectable carcinomas of the exocrine pancreas. P r o c A S C O 17, 211a (abstr. 815).

Sheikh MS, Rochefort H, Garcia M. (1 9 9 5 ). Overexpression of p21 WAF1/CIP1 induces growth arrest, giant cell formation and apoptosis in human breast carcinoma cell lines. O n c o g e n e 11, 1899-1905.

Nemunaitis J, Swisher S, Connors D, et al. (1 9 9 8 ). Adenoviral-mediated p53 (Adp53) gene transfer in patients with advanced non-small cell lung cancer (NSCLC) and head and neck (H&N) cancer. Abstract presented at the Transplantation i n H e m a t o l o g y and O n c o l o g y – Gene Therapy S y m p o s i u m – Munster, Germany, May 10-12.

Simon RH, Engelhardt JF, Yang Y, et al. (1 9 9 3 ). Adenovirusmediated transfer of the CFRT gene to lung of non-human primates, Toxicity study. Human Gene Ther 4, 771780. Smith R. (1 9 5 6 ). Studies on the use of viruses in the treatment of carcinoma of the cervix. Cancer 9, 12111218.

Nemunaitis J, Swisher SG, Roth JA, et al. (1 9 9 8 ). Adenoviral mediated p53 gene transfer in sequence with Cisplatin to tumors of patients with non-small cell lung cancer (NSCLC). Abstr. sent to 7 th I n t e r n a t i o n a l Conference o n Gene Therapy o f Cancer, November 19-21.

Swisher SG, Roth JA, Nemunaitis J, et al. (1 9 9 8 ). Adenoviral-mediated p53 gene transfer in patients with advanced non-small cell lung cancer (NSCLC). Proc ASCO 17, 431a (abstr. #1659).

Nguyen DM, Spitz FR, Yen N, et al. (1 9 9 6 ). Gene therapy for lung cancer, Enhancement of tumor suppression by a

111

Nemunaitis: Gene therapy targeting p53 Swisher SG, Roth JA, Nemunaitis J, et al. Adenoviralmediated p53 gene transfer in advanced non-small cell lung cancer. Manuscript in preparation

Yang Y, et al. (1 9 9 4 ). Cellular immunity to viral antigens limits E1-deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA 91, 4407-4411.

Takafuji ET. (1 9 7 9 ). Simultaneous administration of live, enteric-coated adenovirus types 4, 7, and 21 vaccines, Safety and immunogenicity. J I n f e c t D i s e a s e 140, 4853.

Yei S, Mittereder N, Wert S, et al. (1 9 9 4 ). In vivo evaluation of the study of adenovirus-mediated transfer of the human cystic fibrosis transmembrane conductance of cDNA of the lung. Human Gene Ther 5, 731-744.

Takahashi T, Nau MM, Chiba I, et al. (1 9 8 9 ). p53, a frequent target for genetic abnormalities in lung cancer. S c i e n c e 246, 491-494.

Zhang W, Alemany R, Wang J, et al. (1 9 9 5 ). Safety evaluation of AdCMV-p53 in vitro and in vivo. Human Gene Ther 6, 155-164.

Teoh G, Urashima M, Ogata A, et al. (1 9 9 7 ). MDM2 protein overexpression promotes proliferation and survival of multiple myeloma cells. B l o o d 90, 1982.

Zhang WW, Fang X, Branch CD, et al. (1 9 9 3 ). Generation and identification of recombinant adenovirus by lyposome-mediated transfection and PCR analysis. B i o T e c h n i q u e s 15, 868-872.

Thorlacius S, Borresen AL, Eyfjord JE. (1 9 9 3 ). Somatic p53 mutations in human breast carcinomas in an Icelandic population, A prognostic factor. Cancer Res 53, 1637.

Zhang WW, Fang X, Mazur W, et al. (1 9 9 4 ). High efficiency gene transfer and high level expression of wildtype p53 in human lung cancer cells mediated by recombinant adenovirus. Cancer Gene Ther 1, 5-13.

Timmons T, Burt K, Chema D, et al. (1 9 9 8 ). Biodistribution of Adp53 (INGN 201) in a Phase I/II trials of gene therapy for non-small cell lung cancer (NSCLC). Proc ASCO 17, 431a (abstr. 1660). Tursz T, Le Cesne A, Baldeyrou P, et al. (1 9 9 6 ). Phase I study of a recombinant adenovirus-mediated gene transfer in lung cancer patients. J N a t l C a n c e r I n s t 88, 18571863. Vogelstein B, Kinzler KW. (1 9 9 2 ). Carcinogens leave fingerprints. Nature 355, 209-210. Vogelstein B, Kinzler KW. (1 9 9 2 ). P53 function and dysfunction. C e l l 70, 523-526. Wills KN, Maneval DC, Menzel P, et al. (1 9 9 4 ). Development and characterization of recombinant adenoviruses encoding human p53 for gene therapy of cancer. Human Gene Ther 5, 1079-1088. Wilson DR, Merritt JA, Clayman G, et al. (1 9 9 8 ). Clinical gene therapy strategies, Phase I/II results with adenoviral p53 (INGN 201) gene transfer in advanced head and neck and non-small cell lung cancer. A m S o c G e n e T h e r , Abstr. 380, p, 96a. Winter SF, Minna JD, Johnson BE, et al. (1 9 9 2 ). Development of antibodies against p53 in lung cancer patients appears to be dependent on the type of p53 mutation. Cancer Res 52, 4168-4174. Xu L, Pirollo KF, Chang EH. (1 9 9 7 ). Transferrin-liposomemediated p53 sensitization of squamous cell carcinoma of the head and neck to radiation in vitro. Human Gene Ther 8, 467-475. Xu M, Kumar D, Srinivas S, et al. (1 9 9 8 ). Parenteral gene therapy with p53 inhibits human breast tumor in vivo through a bystander mechanism without evidence of toxicity. Human Gene Ther 8, 177-185. Yamamoto T, Kamata N, Kawano H, et al. (1 9 9 8 ). High incidence of amplification of the epidermal growth factor receptor gene in human squamous carcinoma cell lines. Cancer Res 46, 414-416.

112

Gene Therapy and Molecular Biology Vol 3, page 113 Gene Ther Mol Biol Vol 3, 113-121. August 1999.

Targeted therapy of CEA-producing cells by combination of E. coli cd/HSV1-tk fusion gene and radiation Research Article

Dao-song Xu1 , 2 Xin-yao Wu 1 Yun-fei Xia 3 Ling-hua Wu 3 Chao-quan Luo 1 Yinhao Yang 1 Lu-qi Zhong 4 and Bin Huang4 1

Department of Biochemistry, and 3Department of Radiation of Tumor Hospital, and 4Experimental Animal Center at Sun Yat-sen University of Medical Sciences, Guangzhou, 510089, P. R. China __________________________________________________________________________________________________ 2

Present address: Heinrich-Pette-Institute for Experimental Virology and Immunology at the University of Hamburg, MartinistraĂ&#x;e 52, D-20251 Hamburg, Germany. Tel: 0049-40-4805 1212. E-mail: xu@hpi.uni-hamburg.de Correspondence: Xin-yao Wu, Ph.D. Professor, Department of Biochemistry, Sun Yat-sen University of Medical Sciences, Guangzhou, 510089, P.R. China. E-mail: xyaow@gzsums.edu.cn Received: 5 September 1998; revised and accepted: 23 October 1998

Summary To enhance the specific cytotoxic effects caused by the transfer of the E . c o l i cytosine deaminase (cd) and HSV1-tk to CEA (carcinoembryonic antigen)-producing cells, the expression of the cd-tk fusion gene, driven by the CEA promoter, was investigated followed by treatment with 5-FC and GCV in combination with radiation. The expression vector pCEAcd-tk, based on pcDNA3, was introduced into CEA-producing cells using liposomes. In CEA-producing cells, the CEA promoter could efficiently drive the expression of the fusion suicide gene. The expression activity of the E. coli cd gene driven by the CEA promoter was about three times higher than that driven by the CMV promoter in transfected LoVo cells. A combination of 5-FC and GCV could cause higher c y t o t o x i c i t y t o t h e c e l l s e x p r e s s i n g C D and TK than the use o f a s i n g l e prodrug alone. The cytotoxic effect after combining the two prodrugs with radiation was the highest among all treatments i n v i t r o . I n v i v o , the result of a subrenal capsule assay showed that the inhibition rates for 5-FC (0.5 mg/g) and GCV (0.1 mg/g) to GLC-82 cells transfected with pCEAcd-tk were 18.04% and 55.00%, respectively. A combination of the prodrugs at the same dose resulted in a 152.50% inhibition rate. In addition, the bystander effect exerted by the pCEAcd-tk/5-FC+GCV system in v i t r o was greater than that induced by cd/5-FC or tk/GCV alone.

CD, which can convert the nontoxic prodrugs, ganciclovir (GCV) and 5-fluorocytosine (5-FC), respectively, into metabolites highly toxic to the genetically-modified tumor cells. Experimental results showed that use of the E. coli cd/5-FC or of the HSV1-tk/GCV systems could inhibit the growth of CEA-producing tumor cells in vitro and in vivo (DiMaio et al., 1994; Richards et al., 1995; Lan et al., 1997). However, it has been observed that some tumor cells were resistant to E. coli cd/5-FC or HSV1-tk/GCV (Golumbek et al., 1992; Mullen et al., 1994; Bennedetti et al., 1997).

I. Introduction CEA (carcinoembryonic antigen)-positive tumors are common clinically. At present, there are no efficient therapeutic measures, especially for the patients who are in the mid- or final stages of this disease. Gene therapy may show its strength as an effective method for treating this carcinoma. The herpes simplex virus type I thymidine kinase (HSV1-tk) and the Escherichia coli cytosine deaminase (E. coli cd) genes are commonly used as suicide genes. The expression products of these two nonmammalian genes are two enzymes, HSV1-TK and E. coli 113

Xu et al: cd/tk fusion gene and radiation in cancer eradication The treatment efficiency of the suicide gene/prodrug system mostly depends on the expression efficiency of the introduced suicide gene in the tumor cells. Therefore, the promoter used to drive the expression of a suicide gene is very important. The most commonly used promoters are viral promoters. However, viral promoters are easily inactivated in mammalian cells, resulting in an unstable and low-efficiency expression of a suicide gene. In addition, viral promoters lack the cell-specific acitivities, which could repress expression of a suicide gene in normal cells. Using a retrovirus vector, high expression of the introduced gene was found only in a small subset of the transfected cells; most of the transfected cells did not display the expression product of the introduced gene (Mullen, 1994).

II. Results A. Enzymatic activities of CD and TK Using PCR, a single fragment of about 450-bp was observed on 2% agarose gels (Figure 1A). The same fragment could be amplified from different healthy donors. Sequencing analysis showed that there was only one base mismatch in the CEA promoter fragment of pCEA (Figure 1B), compared with the CEA promoter sequence published by others (Richards, et al., 1993). CEA quantitation in the cell lines was measured by RIA (radioimmuno assay). CEA concentrations were found to be different in different cell lines (Table 1): LoVo and HT-29 cells displayed the highest levels of CEA (581.4 and 316 nm/mg of cellular lysate, respectively). On the contrary, CEA was not detected in BEL-7402 cells. When cdc (E. coli complementary cd gene obtained by PCR, but using the antisense primer but leaving the native stop codon unchanged) was used as an indicator enzyme, the activity of CEA promoter driving its expression was 3.14 times higher than that of CMV promoter in LoVo cells (Table 2).

Over-expression of the CEA gene was a special feature for CEA-positive tumors, and the high level of CEA in physiological fluids has been used for early diagnosis and as a marker of treatment efficiency (Shively & Beatty, 1985; Thomas et al., 1990). Although there was some CEA expression in the normal epithelial cells of the colon, the level was very low (Baranov et al., 1994; Egan et al., 1977). The over-expression of CEA gene resulted from the activated CEA promoter and not from a mutation in the CEA promoter causing its upregulation (Schrewe et al., 1990; Jothy et al., 1993). The CEA promoter occupies a stretch of 420-bp upstream of the translation start site of the CEA gene (Chen et al., 1995; Richards et al., 1995). In CEA-positive cells, specific trans-acting elements are present which activate the CEA promoter. Because of these properties, the CEA promoter could be used to drive the expression of therapeutic genes only in CEA-positive tumor cells; in this context, the E. coli cd and HSV1-tk genes have been used (Osaki et al., 1994; Richards et al., 1995).

Through sequence analysis, our cd sequence was the same as the E. coli cd sequence of the GenBank No s56903 except that of the start and the stop codons which had been changed on purpose. Compared to the HSV1-tk of Genbank No. v00470, one base mismatch leads to the change of the 17th alanine to valine in the tk used in our experiment (data not shown). In transfected BEL-7402 cells, no CD activity was detected. In pCEAcd-tk transfected cells, the activities of CD and TK were measured respectively (T a b l e 3). The results indicate that all CEA-producing cells have higher enzymatic activities than the corresponding parental cells. In non-transfected BEL-7402 cells, there is a low relative activity of TK. It is the activity of cellular TK, not HSV1TK, because 3HdT was used as the substrate.

The mechanism of cell killing by radiation proceeds via damage of the strands of cellular DNA. Cells able to repair the damaged DNA will survive. Because the E. coli cd/5FC and HSV1-tk/GCV systems kill cells through inhibition of DNA synthesis, they could also be used as radiosensitizing agents. It was found that both the E. coli cd/5-FC and HSV1-tk/GCV systems could enhance the sensitivity of cells to radiation (Khil et al., 1996; Rogulski et al., 1997).

Table 1. Concentration of CEA protein in the tumor cell lines Cell line

Here we investigate whether a combination of E. coli cd/5-FC, HSV1-tk/GCV and radiation exert a greater cytotoxic effect to tumor cells, especially to CEAproducing tumor cells. Use of the CEA promoter can limit the expression of the fusion suicide gene in CEAproducing cells. Under these circumstances, treatment with the two prodrugs and application of a low-dose radiation had a much higher cytotoxicity to the tumor cells while minimizing side-effect to normal cells.

LoVo HT-29

Concentration of CEA (ng/mg of cellular lysate) 581.4 316.8

SGC-7901

60.6

GLC-82

80.2

BEL-7402

BT*

The quantity of CEA protein was measured by use of RIA (Radioimmuno assay) method. * below the threshold of 5 ng.

114

Xu et al: cd/tk fusion gene and radiation in cancer eradication

F i g u r e 1 . A . Amplification of CEA promoter. The PCR products were separated on 2% agarose gel. Lane 1, DNA molecular weight marker, Lambda DNA/EcoRI+HindIII; lane 2-9, PCR products from peripheral blood cell genomic DNA of healthy individuals. B . Comparion of the CEA promoter sequence with published CEA DNA sequence in the 5´non-translation region. Query: the sequence of CEA promoter used in the experiment; subject: part of the published CEA DNA sequence (Genbank No: z21818).

Table 2 . Enzymatic activities of CD in tumur cells (specific activity) Cell line

Parental

Tansfected with pCEAcdc

Transfected with pcDNA3cdc

LoVo

0

2748

875

HT-29

0

2034

-

SGC-7901

0

670

-

GLC-82

0

714

-

BEL-7402

0

0

-

Specific activity was defined as nmol of cytosine deaminated/min/mg protein. It was measured spectrophotometrically as a decrease in absorbance at 285 nm, in a 1-ml assay mixture containing cell extract in 50 mM Tris-HCl, pH 7.3, 0.5 mM cytosine. The product was estimated using a molar extinction coefficient 1.038 !10 litre/mol/cm. Table 3 . Enzymatic activites of CD and TK in the cells transfected with pCEAcd-tk Enzymatic activity Cell line LoVo

CD* 2516

TK# 238.1

HT-29

1603

149.2

SGC-7901

421

84.7

GLC-82

507

69.4

BEL-7402

0

10.8

*: specific activity; #: ralative activity to TK actvity in the pcDNA3tk-transfected cells.

114

Gene Therapy and Molecular Biology Vol 3, page 115

F i g u r e 2 . Additive cytotoxic effect of combined use of 5-FC and GCV to tumor cells expressing cd-tk. SEMs (standard error means) were presented by error bars(n=3).

F i g u r e 3 . In vitro bystander killing effect of pCEAcd-tk/5-FC+GCV. The pCEAcd-tk transfected LoVo cells were mixed with 3 non-transfected LoVo cell in different portions. The mixed cells were seeded on a 96-well plate at a density of 2 ¥10 cells per well. 24 hrs later, the cells were exposed to 0.5µM GCV and 100 µM 5-FC. After the cells were incubated for 72 hrs, the surviving rates were measured by MTT assay. Bars, SEMs (n=3).

treatment with 5-FC plus GCV produces 5.4% surviving rate. When the same dose of 5-FC or GCV was used alone, the survival rates were 40.2% and 56.7%, respectively. The subrenal capsule assay (SRCA) result indicates that the tumor inhibition rate is much higher when using a combination of the two prodrugs in nude mice (Table 5).

B. Cytotoxicity of 5-FC and GCV to tumor cells expressing CD and TK CEA-producing cells transfected with pCEAcd-tk become more sensitive to 5-FC and GCV than parental cells as deduced from growth inhibition in vitro measuring the IC50 (concentration of 50 % growth inhibition) (Table 4). Use of 5-FC in combination with GCV has a remarkable additive cytotoxic effect to CEA-producing cells expressing CD and TK (Figure 2). In addition, the in vitro pCEAcdtk/5-FC+GCV system has a higher bystander effect than cd/5-FC or tk/GCV (Figure 3). In the 20% group,

C. Radiosensitization of pCEAcd-tk/5FC+GCV When 100 µMol 5-FC or 0.5 µMol GCV is added to LoVo cells transfected with pCEAcd-tk, 6.5 Gy and 4.9 115

Gene Therapy and Molecular Biology Vol 3, page 113 Table 4 . The IC50 of tumor cells to 5-FC and GCV I C 5 0 SD( µM ) Cell line LoVo* LoVo HT-29* HT-29 SGC-7901* SGC-7901 GLC-82*

5-FC 67.2±

GCV

26.8

0.75±

9650.34± 563.00 162.70±

0.86±

9580.50± 762.58 287.57±

17.30± 12.83±

143.61

3360.80

58.88

977.56

37.78

51.50

59.12

74.09

1.22

1.15

5.16

890.91±231.73

81.34

GCV

0.24

840.70±125.42

40.48

10865.20± 481.82 232.10±

5-FC

0.16

25250.60±430.85

56.20

Ratio IC50#

7.51

GLC-82

13720.63±2407.38

950.60±124.30

BEL-7402*

9080.85± 375.31

678.59± 35.70

BEL-7402

11070.14±2512.17

780.41±147.20 3

Cells were seeded at a density of 2!10 cells/well on 96-well plates. Different concentrations of 5-FC, GCV were added. After 72hrs, the percentage of growth inhibition was measured by the MTT assay. The results represent mean±SD(n=3). IC50=the concentration of 50% growth inhintory rate. * cells transfected with pCEAcd-tk; # parental cell IC50/transfected cell IC50 to 5-FC or GCV.

Table 5 . In vivo growth inhibition of pCEAcd-tk transfected GLC-82 cells by 5-FC and GCV Drug

Dose (mg/gm)

Schedule

Do

Dn

D n-D o

Inhibition rate(%)

Control

0.04*

1/d!2

40.5±6.26

43.8±10.36

3.3± 9.01

----

5-FC

0.5

1/d!2

39.7±5.93

42.5±12.68

2.7±11.34

18.04

GCV

0.1

1/d!2

36.6±5.55

38.1±11.3

11.5± 9.44

55.00

5-FC+GCV

0.5+0.1

1/d!2

41.9±4.68

40.2±16.6

-1.7±16.91

152.50

The prodrugs were given by the intraperitonal injection. *: ml of 0.9% NaCl per gram; Do: tumor volume before transplantation, Dn: tumor volume after the animal was sucrificed.

difference, 109 A"G (Figure 1B). It was found that the essential part of CEA promoter was located between nucleotides 295-318 (Richards, et al, 1995). Through footprinting, Chen, et al (1995) and Hauck and Stanners (1995) found that there were 5 FP (footprinting) regions in the cis-acting sequence of the CEA promoter, in which FP1-4 represented the positive regulatory elements whereas FP5 (-568 to -560, where +1 is the start of translation) represented the negative regulatory elements. Sp1 and Sp1like factors could bind to Fp1, FP2 and FP3. The protein bound to FP4 was AP4. The only different base in the CEA promoter used in our experiments is in the FP4 region. Although the FP4 region was not an essential part of the CEA promoter, it may affect the activity of the CEA promoter. Using a common Taq DNA polymerase, an active CEA promoter could be obtained (DiMaio et al., 1994), but a low activity CEA promoter was observed (Osaki et al., 1994). It is not clear which bases play a key role in the CEA promoter activity. It is possible that a more efficient CEA promoter can be constructed by

Gy, respectively, were required to reduce the surviving fraction to 0.01 (Figure 4A). When a combination of the same dose of 5-FC plus GCV was used, only 4.2 Gy were required to obtain the same survival fraction. The pCEAcdtk/5-FC+GCV system had a similar effect on GLC-82 cells (Figure 4B).

III. Discussion It is possible that the mutation in the CEA promoter can affect its activity and cell-specificity. In our experiments, the CEA promoter is obtained using a high fidelity DNA polymerase. The CEA promoter shows a higher activity in CEA-producing cells. In LoVo cells its activity was 3.14 times higher than that of the CMV promoter (using the E. coli CD as the indicator). The sequence of the CEA promoter used in our experiments is almost identical to that of the CEA promoter sequence published before (Richards, et al., 1993) except one base

113

Xu et al: cd/tk fusion gene and radiation in cancer eradication changing some bases in the CEA promoter sequence, which may be much better suited for targeting expression of a suicide gene to CEA-producing tumor cells.

found that the bystander effect of the combined use is enhanced in vivo. Although HSV1-tk/GCV, E. coli cd/5-FC system could effectively kill tumor cells in vitro and in vivo, the efficiency between these two systems were different to some kinds of tumors. E. coli cd/5-FC therapy was more effective than HSV1-tk/GCV to pulmonary adenocarcinoma (Hoganson et al., 1996). In vivo, human colorectal carcinoma cells were more effectively eradicated by E. coli cd/GCV than HSV1-tk/GCV (Trinh et al., 1995). Most gastric-intestinal and lung carcinomas are CEA-positive. On the other hand, tumor microenvironment can determine the cell radiosensitivity, but the sensitivity of tumor cells to radiation also is dependent on intrinsic cellular factors. Both HSV1-tk/GCV and E. coli cd/5-FC could alter the cellular factors, and enhance the radiosensitivity (Kim et al., 1994, 1995; Khil et al., 1996; Rogulski et al., 1997). Most CEA-positive tumor cells, for example pulmonary adenocarcinoma cells, are not sensitive to radiation. Therefore, it is much more effective to use a combination of these two systems to kill these tumor cells.

Although the essential sequence for an active CEA promoter is known (Richards et al., 1995), the mechanism of activating CEA promoter is unclear. Our results indicate that the activities of CD and TK in different CEAproducing cell lines transfected with pCEAcd-tk are different. The enzymatic activity shows a positive relationship to the concentration of CEA in the cells. An active CEA promoter is determined by the interaction of a cis-acting sequence with trans-acting elements. We found that the nuclear proteins binding to the CEA promoter were different between LoVo and BEL-7402 using gel mobility shift assays (data not shown). The different enzymatic activities may reflect the different interactions involving these elements. Combined therapy to tumors can enhance the cytotoxicity and beneficial effect from each therapeutic regime. Using cotransfection of cells with HSV-tk and E. coli cd, Uckert, et al. (1998) found that the combination of the two genes was the most effective for killing tumor cells both in vitro and in vivo, and only this combination could cause complete eradication of tumors in vivo. Rogulski et al. (1997) revealed that the combined use of cd-tk/5-FC+GCV and radiation had a strong cytotoxic effect to 9L tumor cells. The best way to treat tumors is to kill only tumor cells without any severe damage to healthy cells. So it is important to limit the expression of a suicide gene only in tumor cells before using the prodrug. At present, two ways, targeting vectors and targeting transcription (see review by Miller & Whelan, 1997), can be used. We used the strategy of targeting transcription, and the pCEAcd-tk/5-FC+GCV system showed a strong cytotoxic effect to the CEA-producing cells. In addition, high concentration of GCV or 5-FC could cause remarkable nonspecific toxicity to nontransfected cells (Beck, et al., 1995, Cool, et al., 1996). Use of the combination of these two systems will reduce the dose of each prodrug, whereas the cytotoxic effect can be enhanced. In our experiments the doses of 5-FC and GCV are much lower than the “safe� concentrations of these in human blood. If higher doses of the prodrugs are used, the pCEAcd-tk/5-FC+GCV might kill tumor cells even more efficiently reducing the possibility of converting tumor cells to become resistant.

There were some limitations for treating pulmonary adenocarcinoma cells by use retrovirus-mediated HSV-tk gene transfer (Zhang et al, 1997). Song et al. (1997) found that injection of a pcDNA3-liposome mixture could cause the highest expression of an exogenous gene in mouse lungs. In addition, the most common reason for mortality of patients with colon carcinoma is hepatic metastases. In normal lung and liver tissues, the CEA gene is not expressed. If pCEAcd-tk is non-specifically transfected into these normal cells, the suicide gene will not be expressed since the CEA promoter is in an inactive state. After use of prodrug, no toxic metabolite of the prodrug will be produced in the normal cells, thus reducing the side effects of suicide gene/prodrug therapy to normal cells. The therapeutic system, pCEAcd-tk/5-FC+GCV accompanied with low dose of radiation, may become a useful tool for the eradication of CEA-producing tumors.

IV. Materials and methods A. Vector construction Two primers were used to amplify the CEA promoter from the genomic DNA of peripheral blood cells from healthy blood donors, 5'-GTA TCG CGA ATC ATC CCA CCT TCC CAG AG-3' (sense), 5'-GGG AAG CTT TGT CTG CTC TGT CCT CCT C-3' (antisense). The high-fidelity Pwo polymerase (Boehringer Mannheim Co.) was used to amplify a 438-bp CEA promoter. The amplified fragment was cut with NruI and HindIII, and then the CMV promoter in pcDNA3 (Invitrogen) was replaced with this fragment, resulting in the vector pCEA, in which the CEA promoter fragment was ensured by direct

The mechanisms of the bystander effect of E. coli cd/5FC and HSV1-tk/GCV are not completely clear, but clear differences between these two systems have been observed (Denning & Pitts, 1997). The combined use of the two systems could promote the bystander effect (Rogulski, et al., 1997, Uckert, et al., 1998). In agreement with this we

114

Xu et al: cd/tk fusion gene and radiation in cancer eradication

3

F i g u r e 4 . The radiative enhancing effect of pCEAcd-tk/5-FC+GCV. A: LoVo; B: GLC-82. 2 ¥10 cells/well were seeded on 96-well plates, and then 100 µM 5-FC and 0.5µM GCV were added. 72hrs later, the cells were irradiated with different doses of Xrays. After 6 days, the cell number in each well was estimated by MTT assay according to the standard calibration curves. SEMs (n=3) were omitted for clarity.

Military Medical University and Experimental Animal Center, Sun yat-sen University of Medical Sciences) were cultivated in RPMI 1640 (GIBCO-BRL) medium with 10% fetal calf serum, 100 units/ml penicillin and 100µg/ml streptomycin. No mycoplasma was detected by PCR. Cells were transfected with pCEAcd-tk by use of ESCORT transfection reagent (Sigma), and the positive clones were selected with G418(GIBCO-BRL) for fourteen days. These cells were used for measuring the enzymatic activities of E. coli CD and HSV1-TK, and for cytotoxicity assay.

dideoxynucleotide sequencing. The primers, (sense) 5'-GGG AAG CTT ACC ATG TCG AAT AAC GCTTTA C-3' (with a HindIII cut site in 5' end) and (antisense) 5'-CGC GGATCC TCC ACG TTT GTA ATC GAT GGC-3'(with a BamHI cut site in 5' end) were used to amplify the E. coli cd gene from chromosomal DNA of JM109 bacteria. In the sense primer, the initial context was changed into the Kozak sequence (Kozak, 1986). The stop codon (TGA) of E. coli cd was changed into GGA (encoding for glycine), leading to read through downstream HSV1-tk gene. The other two primers were used to amplify the HSV1-tk gene from the plasmid pHSV106 (GIBCO-BRL). The sense primer was 5'-CGC GGA TCC GGC GGG GGC GGT GGA GGA GGG GGT ATG GCT TCG TAC-3', in which there was a BamHI cut site and eight codons for glycine. The antisense primer was 5'-CGG GAA TTC CCT TCC GGT ATT GTC TCC TTC CGT-3'(with EcoRI cut site) (Rogulski et al., 1997). The ligation and identification of inserted fragments by using restriction enzyme analysis was carried out according to methods described (Sambrook et al., 1989). The amplified fragments were cut with relevant restriction enzymes, and then inserted into the MCS (multiple cloning site) of pCEA, resulting in the expression vector, pCEAcd-tk. Between the cd and tk, there was a linker which encoded ten glycines and one serine.

C. Enzymatic activities of CD, TK and cytotoxicity assay E. coli CD activity was measured according the method described (Austin & Huber, 1993). The buffer was 50 mM TrisHCl (pH7.3), 0.5 mM cytosine (Sigma). Specific activity was defined as nmol of cytosine deaminated/min/mg proteins. The 3 molar extinction coefficient was 1.038 ¥10 litre/mol/cm. TK activity was detected as follows: 25 µl of cell extract, 75 µl of reaction buffer contained 50 mM Tris-HCl (pH7.5), 10mM ATP, 10 mM MgCl2, 10 mM #-mercaptoethanol, 10 mM NaF, 3 50 µg/ml PMSF(Sigma) and 2µmol/L HdT (20ci/mmol). The mixture was incubated at 37 °C for 30 min, and then 100 µl of reaction mixture was dropped onto DE-81 filter paper (Whatman). The paper was washed with 95% ethanol three times, and then put in 5-ml scintillation liquid for measuring CPM. The relative activity of TK was defined as follows:

B. Cell culture and transfection The cell lines, LoVo, HT-29 (human colon carcinoma) and GLC-82 (human lung adenocarcinoma), SGC-7901 (human stomach carcinoma), BEL-7402 (human hepatoma) were used. LoVo, HT-29 (ATCC) and other cell lines (provided by the first

CPM/mg of proteins in the cells transfected with pCEAcd-tk ! 100% CPM/mg of proteins in the cells transfected with pcDNA3tk

114

Xu et al: cd/tk fusion gene and radiation in cancer eradication In pcDNA3tk, the CMV promoter drove the expression of HSV1-tk gene cut from pHSV106 with BglII and EcoRI.

cloning, sequencing, and expression of Escherichia coli cytosine deaminase. M o l . P h a r m a c o l . 43, 380-387.

The cytotoxicity assay was carried out by MTT (Sigma) 3 assay. In a 96-well culture plate, 2x10 cells/well were seeded, and the different concentrations of 5-FC (Sigma) and GCV (Roche) were added. After 72 hrs, 10 µl of MTT (5 mg/ml) was added into each well and incubated in 37 °C for 4hrs. The supernatant was discarded and 150 µl/well of DMSO was added. The absorbance (A) was measured at 570 nm. The survival rate=Atreated/Acontrol x100%. The experiment was performed three times.

Baranov, V., Yeung, M.M., and Hammarstrom, S. Y. (1 9 9 4 ). Expression of carcinoembryonic antigen and nonspecific cross-reacting 50-kDa antigen in human normal and cancerous colon mucosa: comparative ultrastructural study with monoclonal antibodies. Cancer R e s . 54(12), 3305-3314. Beck, C. Cayeux, S. Lupton, S.D. Dorken, B. and Blankenstein, T. (1 9 9 5 ). The thymidine kinase/ganciclovir-mediated "suicide" effect is variable in different tumor cells. Hum. Gene Ther. 6, 1525-1530.

The sensitivity of tumor cell expressing CD and TK was carried out according to the method described by Price & McMillan(1990). An X-ray instrument was used, and the dose rate was 106.82 cGR/min. The surviving fraction was calculated as follows:

Benedette, S., Dimeco, F., Pollo, B., Cirennei, N., Colombo B.M., Bruzzone, M.G., Cattaneo, E., Vescovi, A., Didonato, S., Colombo, M.P., and Finocchiaro, G. (1 9 9 7 ). Limited efficacy of the HSV-TK/GCV system or gene therapy of malignant gliomas and perspectives for the combined transduction of the interleukin-4 gene. Hum. Gene Ther. 8, 1345-1353.

(Cell number in the control well) divided by (Cell number in the radiated well or prodrug/radiation-treated well) x100%. The data and charts was processed and produced by the Department of Radiobiology, Tumor Hospital of China Academia.

Chen, C.J., Li, L.J, Maruya, A., and Shively, J.E. (1 9 9 5 ). In vitro and in vivo footprint analysis of the promoter of carcinoembryonic antigen in colon carcinoma cells: effects of interferon gamma treatment. C a n c e r R e s . 55, 3873-3882.

D. Radiosensitization

Cool V, Pirotte B, Gerard C, Dargent JL; Baudson N, Levivier M, Goldman S, Hildebrand J, Brotchi J and Velu T (1 9 9 6 ). Curative potential of herpes simplex virus thymidine kinase gene transfer in rats with 9L gliosarcoma. Hum. Gene Ther. 7, 627-635.

E . I n v i v o studies GLC-82 cells transfected with pCEAcd-tk were inoculated subcutaneously into 6-8 wk BALB/C-nu/nu mice, and tumors were allowed to grow for about one month. Afterwards, the tumor tissue was surgically removed and cut into 1-mm size fragments which were implanted under the renal capsules of BALB/C-nu/nu mice. 5-FC and GCV were delivered by intraperitoneal injection at days 2 and 3. 10 days later, the 2 3 animals were sucrificed. The tumor volume = (a xb ) /2 (mm ), where a is: the longest diameter of the tumor, and b: the shortest diameter. The tumor inhibition rate was calculated as follows: (Dn - Do in control) - (Dn - Do in subject )

Denning, C., and Pitts, J.D. (1 9 9 7 ). Bystander effects of different enzyme-prodrug systems for cancer gene therapy depend on different pathways for intercellular transfer of toxic metabolites, a factor that will govern clinical choice of appropriate regimes [see comments]. Hum. Gene Ther. 8, 1825-1835 Dimaio, J. M., Clary, B. M., Via, D.F., Coveney, E., Pappas, T.N., and Lyerly, H. K. (1 9 9 4 ). Directed enzyme pro-drug gene therapy for pancreatic cancer in vivo. Surgery. 116, 205-213.

x100%

Dn - Do in control

Egan, M. L., Pritchard, D. G., Todd, C.W., and Go,V.L. (1 9 7 7 ). Isolation andimmunochemical and chemical characterization of carcinoembryonic antigen-like substances in colon lavages of healthy individuals. Cancer Res. 37, 2638-2643.

Do: the tumor volume before translated into subrenal capsule; Dn: the tumor volume after the mouse was sacrificed.

Acknowledgements

Golumbek, P.T., Hamzek, F.M., Jaffee, E. M., Levitsky, H., Lietman, P.S., and Pardoll, D. M. (1 9 9 2 ). Herpes simplex-1 virus thymidine kinase gene is unable to completely eliminate live, nonimmunogenic tumor cell vaccines. J. Immunother. 12, 224-230.

We are indebted to Prof. Lin Lu and Drs. Yi-fang Chen for supplying vectors, and Department of Medicine of the First Military Medical University (Guangzhou, China) for supplying cell lines. This work is supported by the grants from China Fundation for Natural Sciences to X.Y. Wu, and from Research fundation of SUMS to D.S. Xu and X. Y. Wu.

Hoganson, D.K., Batra, R.K., Olsen, J.C., and Boucher, R.C. (1 9 9 6 ). Comparison of the effects of three different toxin genes and their level of experssion on cell growth and bystander effect in lung adenocarcinoma. C a n c e r R e s . 56, 1315-1323

References

Huber, B.E., Austin, E.A., Goog, S.S., Knic, V.C., Tibbels, S, and Richards, C.A. (1 9 9 3 ). In vivo antitumor activity of

Austin, E.A., and Huber, B.E. (1 9 9 3 ). A first step in the development of gene therapy for colorectal carcinoma:

114

Gene Therapy and Molecular Biology Vol 3, page 115 5-fluorocytosine on human colorectal carcinoma cells genetically modified to express cytosine deaminase. Cancer Res. 53, 4619-4626.

Richards,C.A., Wolberg, A.S. and Huber, B.E. (1 9 9 3 ). The transcriptional control region of the human carcinoembryonic antigen gene: DNA sequence and homology studies. DNA Seq. 4, 185-196

Jothy, S., Yuan, S.Y., and Shirota, K. (1 9 9 3 ). Transcription of carcinoembryonic antigen in normal colon and colon carcinoma. In situ hybridization study and implication for a new in vivo functional model. A m . J . P a t h o l . 143, 250-257.

Richards, C.A., Austin, E.A., and Huber, B.E. (1 9 9 5 ). Transcriptional regulatory sequences of carcinoembryonic antigen: identification and use with cytosine deaminase for tumor- specific gene therapy. Hum. Ge ne Ther. 6 , 881-893.

Khil, M.S., Kim, J.H., Mullen, C.A., Kim, S.H., and Freytag, S.O. (1 9 9 6 ). Radiosensitization by 5-fluorocytosine of human colorectal carcinoma cells in culture transfected with cytosine deaminase gene. C l i n . C a n c e r R e s . 2 , 53-57

Rogulski, K. R., Kim, J.H., Kim, S.H., and Freytag, S. O. (1 9 9 7 ). Glioma cells transduced with an Escherichia coli CD/HSV-1 TK fusion gene exhibit enhanced metabolic suicide and radiosensitivity. H u m . G e n e T h e r . 8, 7385.

Kim, J.H., Kim, S.H., Brown, S.L., Freytag, S.O, (1 9 9 4 ). Selective enhancement by an antiviral agent of the radiation-induced cell killing of human glioma cells transduced with HSV-tk gene. C a n c e r R e s . 54, 60536056.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1 9 8 9 ). M o l e c u l a r C l o n i n g (2nd ed ), Cold Spring Harbor Laboratory Press, P1.13-P1.85 Schrewe, H., Thompson, J., Bona, M., Hefta, L.J., Maruya, A., Hassauer,M., Shively, J.E., von-Kleist, S., and Zimmermann, W. (1 9 9 0 ). Cloning of the complete gene for carcinoembryonic antigen: analysis of its promoter indicates a region conveying cell type-specific expression. M o l . C e l l . B i o l . 10, 2738-2748.

Kim, J.H., Kim, S.H., Kolozsvary, A., Brown, S.L., Kim, O.B., and Freytag, S.O. (1 9 9 5 ). Selective enhancement of radiation response of herpes simplex virus thymidine kinase transduced 9L gliosarcoma cells in vitro and in vivo by antiviral agents. I n t J R a d i a t . O n c o l . B i o l . P h y s . 33, 861-868.

Shively, J.E., and Beatty, J.D. (1 9 8 5 ). CEA-related antigens: molecular biology and clinical significance. C r i t . R e v . O n c o l . H e m a t o l . 2, 355-399.

Kozak, M. (1 9 8 6 ). Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. C e l l . 44, 283-292.

Song, Y.K., Liu, F., and Liu, D. (1 9 9 7 ). Characterization of cationic liposome-mediated gene transfer in vivo by intravenous administration. Hum. Gene Ther. 8, 15851594.

Lan, K.H., Kanai, F., Shiratori, Y., Ohashi, M., Tanaka, T., Okudaira, T., Yoshida, Y., Hamada, H., and Omata, M. (1 9 9 7 ). In vivo selective gene expression and therapy mediated by adenoviral vectors for human carcinoembryonic antigen-producing gastric carcinoma. Cancer Res. 57, 4279-4284.

Thomas, P., Toth, C.A., Saini, K.S., Jessup, J.M., and Steele, G. Jr. (1 9 9 0 ). The structure, metabolism and function of the carcinoembryonic antigen gene family. B i o c h i m . B i o p h y s . A c t a . 1032, 177-189.

Miller, N., and Whelan, J. (1 9 9 7 ). Progress in transcriptionally targeted and regulatable vectors for genetic therapy. Hum. Gene Ther. 8, 803-815.

Trinh, Q.T., Austin, E.A., Murray, D.M., Knick, V.C., Huber, B.E. (1 9 9 5 ). Enzyme/prodrug gene therapy: comparison of cytosine deaminase/5-fluorocytosine versus thymidine kinase/ganciclovir enzyme/prodrug systems in a human colorectal carcinoma cell l ine. C a n c e r R e s . 55, 48084812.

Mullen, C.A. (1 9 9 4 ). Metabolic suicide genes in gene therapy. Pharmacol. Ther. 63, 199-207. Mullen, C.A., Coale, M. M., Lowe, R., and Blaese, R.M. (1 9 9 4 ). Tumors expressing the cytosine deaminase suicide gene can be eliminated in vivo with 5fluorocytosine and induce protective immunity to wild type tumor. Cancer Res. 54, 1503-1506.

Uckert, W., Kammertรถns, T., Haack, K., Qin, Z., Gebert, J., Schendel, D.J., and Blankenstein, T. (1 9 9 8 ). Double suicide gene(cytosine deaminase and herpes simplex virus thymidine kinase) but not single gene transfer allows reliable elimination of tumor cell in vivo. Hum. Gene Ther. 9, 855-865.

Osaki, T., Tanio, Y., Tachibana, I., Hosoe, S., Kumagai, T., Kawase, I., Oikawa, S., and Kishimoto, T. (1 9 9 4 ). Gene therapy for carcinoembryonic antigen-producing human lung cancer cells by cell type-specific expression of herpes simplex virus thymidine kinase gene. Cancer R e s . 54, 5258-5261.

Zhang, L., Wikenheiser, K.A., Whitsett, J.A. (1 9 9 7 ). Limitations of retrovirus-mediated HSV- tk gene transfer to pulmonary adenocarcinoma cells in vitro and in vivo. Hum. Gene Ther. 8, 563-574.

Price, P. and McMillan, T. J. (1 9 9 0 ). Use of the tetrazolium assay in measuring the response of human tumor cells to ionizing radiation. Cancer Res. 50, 1392-1396.

115

Gene Therapy and Molecular Biology Vol 1, page 123 Gene Ther Mol Biol Vol 3, 123-131. August 1999.

Efficacy of antiherpetic drugs in combined gene/chemotherapy of cancer is not affected by a specific nuclear or cytoplasmic compartmentation of herpes thymidine kinases Research Article

Bart Degrève1, Erik De Clercq1, Anna Karlsson2, and Jan Balzarini1 1

Rega Institute for Medical Research, Laboratory of Virology and Chemotherapy, B-3000 Leuven, Belgium

2

Karolinska Institute, Department of Immunology, Microbiology, Pathology and Infectious diseases, Division of Clinical Virology, S-141 86 Stockholm, Sweden __________________________________________________________________________________________________ C o r r e s p o n d i n g author: Jan Balzarini, Ph.D., Rega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven, Belgium. Tel: +32-16-337352; Fax: +32-16-337340; E-mail, jan.balzarini@rega.kuleuven.ac.be Abbreviations AA, amino acid; ACV, 9-(2-hydroxyethoxymethyl)guanine (acyclovir); araT, 1-!-D-(arabinofuranosyl)thymine; BCV, (R)-9[(3,4-dihydroxybutyl)guanine] (buciclovir); BVDC, (E)-5-(2-bromovinyl)-2’-deoxycytidine; BVaraU, (E)-5-(2-bromovinyl)-1!-D-arabinofuranosyluracil; BVDU, (E)-5-(2-bromovinyl)-2’-deoxyuridine; FIAC, 1-(2’-fluoro-2’-deoxy-!-D-arabinofuranosyl)5-iodocytosine; FMAU, 1-(2’-fluoro-2’-deoxy-1-!-D-arabinofuranosyl)-5-methyluracil; GCV, 9-(1,3-dihydroxy-2propoxymethyl)guanine (ganciclovir); GFP, green fluorescent protein; HSV-1 and HSV-2, herpes simplex virus type 1 and herpes simplex virus type 2; LBV, (R)-9-[2,3-bis(hydroxymethyl)cyclobutyl]guanine (lobucavir, cyclobut-G, BMS180194); NLS, nuclear localization signal; PCV, 9-[4-hydroxy-3-(hydroxymethyl)but-1-yl]guanine (penciclovir); S-BVDU, (E)-5-(2bromovinyl)-2’-deoxy-4’-thiouridine; VZV , varicella-zoster virus. Received: 30 October 1998; accepted 10 November 1998

Summary Introduction of the herpes simplex virus type 1 (HSV-1) thymidine kinase (TK) gene in tumor c e l l s , f o l l o w e d b y t r e a t m e n t o f t h e t r a n s f e c t e d t u m o r c e l l s w i t h a n antiherpes drug has shown promise in the treatment of solid tumors. We have recently shown that the HSV-1 TK fused to green fluorescent protein (GFP) was localized almost exclusively in the nuclei of HSV-1 TK-GFP fusion gene-transfected human osteosarcoma cells, due to the presence of a nuclear localization signal (NLS) at the N-terminus of the HSV-1 TK. A deletion mutant, lacking the N-terminal 34 amino acids [ (AA1-34)HSV-1 TK-GFP], was distributed throughout the cytoplasm and nucleus of transfected tumor cells. In addition, varicella-zoster virus (VZV) TK-GFP, which lacks the NLS and which i s uniformly distributed i n the nucleus and cytoplasm o f the VZV TK-GFP genetransfected tumor cells, could be specifically targeted to the nucleus by ligating the HSV-1 TK nuclear localization signal to the VZV TK-GFP sequence. Two pairs of osteosarcoma cell lines stably expressing HSV-1 TK-GFP or VZV TK-GFP either in the nucleus or throughout the cell were established and compared for their sensitivity t o the cytostatic effects o f a variety of antiherpetic nucleoside analogues. In addition, the efficacy of nucleoside analogues in contributing to the bystander effect (i.e., the killing of non-transfected tumor cells by neighbouring TK genetransfected c e l l s after gap junctional transfer o f phosphorylated nucleoside metabolites), was evaluated using the HSV-1 TK-GFP and (AA1-34)HSV-1 TK-GFP gene constructs. From our e x p e r i m e n t s i t i s i n f e r r e d t h a t t h e r e i s no difference i n cytostatic activity o f the antiherpetic 123

Degrève et al: Antiherpetic drugs in combined gene/chemotherapy of cancer nucleoside analogues against TK gene-transfected cells, whether the TK activity is solely localized in the nucleus or spread over the nucleus and cytosol. Also, the bystander killing effect of the antiviral compounds was independent of the nature of the intracellular compartment in which the HSV-1 TK-GFP fusion protein was expressed.

I. Introduction The broad substrate specificity of the thymidine kinase (TK) of most herpes viruses, including herpes simplex virus type 1 (HSV-1), herpes simplex virus type 2 (HSV2) and varicella-zoster virus (VZV), can be exploited in the treatment of herpesvirus infections (De Clercq, 1993; 1995). Exquisitely potent antiherpetic nucleoside analogues including (E)-5-(2-bromovinyl)-2’-deoxyuridine (BVDU) and ganciclovir (GCV) have been developed, which owe their selective antiviral activity to their specific phosphorylation by the herpetic TK but not by the mammalian TK (Figure 1). This concept of the herpetic TK-dependent cytostatic effect of otherwise non-toxic nucleoside analogues was later introduced in the field of anticancer research. Balzarini and coworkers reported on the highly selective cytostatic activity of the antiherpetic drugs GCV and BVDU, and various structurally related derivatives thereof, against murine mammary carcinoma (FM3A) cells transfected with the HSV-1 or HSV-2 TK gene (Balzarini et al., 1985, 1987, 1993, 1994). Differences were found in the cytostatic potency of the drugs depending on the nature of the suicide gene (i.e. HSV-1 or HSV-2 TK gene). In 1992, Culver and coworkers showed complete regression of established brain tumors in rats after in situ transduction with the HSV-1 TK gene and subsequent treatment with GCV (Culver et al., 1992). Several clinical trials, all utilizing the HSV-1 TK/GCV system, are underway to assess the safety and efficacy of this combined gene/chemotherapy treatment for cancer (Oldfield et al., 1993; Culver et al., 1994; Freeman et al., 1995; Kun et al., 1995; Raffel et al., 1994). Recently, we demonstrated that VZV TK and a variety of pyrimidine nucleoside analogues represent appropriate alternatives for the HSV-1 TK/GCV combination therapy (Degrève et al., 1997). We also recently studied the intracellular localization of HSV-1, HSV-2 and VZV TK (Degrève et al., 1998). The herpetic TKs were expressed as fusion proteins with the green fluorescent protein (GFP) (Chalfie et al., 1994; Rizzuto et al., 1995; Youvan et al., 1996) in human OstTK- cells and their intracellular localization was examined using a fluorescence microscope. HSV-1 TK fused with GFP was almost exclusively localized in the nuclei of HSV-1 TK-GFP gene-transfected tumor cells. In contrast, introduction of the HSV-2 TK-GFP fusion gene gave rise to predominant cytosolic fluorescence. VZV TKGFP showed a uniformly distributed fluorescence pattern. When the N-terminal 34 amino acids (AAs) were deleted 124

from the HSV-1 TK-GFP construct, the resulting mutant fusion protein lost its specific nuclear localization. We proved that this 34 amino acid stretch was also capable of targeting VZV TK-GFP and GFP to the nucleus of genetransfected OstTK- cells, indicating that a nuclear localization signal (NLS) was present in this N-terminal part of HSV-1 TK. By site-directed mutagenesis of each of the positively charged amino acids at the N-terminus of HSV-1 TK, we were able to identify a nonapeptide, 25R-RT-A-L-R-P-R-R33, which is strictly required for specific nuclear localization of HSV-1 TK (Degrève et al., 1998).

F i g u r e 1 . Structural formulae of 4 representative test compounds.

Gene Therapy and Molecular Biology Vol 1, page 125 The expression of HSV-1, HSV-2 and VZV TK in different intracellular localizations prompted us to investigate whether the intracellular localization of a particular TK would influence the cytostatic effects of antiherpetic nucleoside analogues. Therefore, a series of antiherpetic pyrimidine and purine nucleoside analogues were evaluated for their inhibitory activity on the proliferation of OstTK- cells expressing either the nucleustargeted (wild-type) HSV-1 TK fused to GFP, the uniformly distributed "(AA1-34)HSV-1 TK-GFP (lacking the first 34 amino acids that contain the nuclear targeting signal nonapeptide), the wild-type uniformly distributed VZV TK-GFP and the nucleus-targeted NLS-VZV TKGFP (containing the HSV-1 TK AA1-34 NLS). We have also recently explored the ability of a variety of purine and pyrimidine nucleoside analogues to exert a bystander killing effect in mixed tumor cell populations (Degrève et al., 1999), i.e. the potency of the compounds to kill TK tumor cells that are neighbouring HSV-1 TK-GFP genetransfected cells upon gap junctional transfer of the phosphorylated compounds. We showed that purine nucleoside analogues (represented by GCV) have a far more pronounced bystander killer effect than pyrimidine nucleoside analogues (represented by BVDU), regardless of their potent inhibitory potential against the HSV-1 TKGFP gene-transfected tumor cells. We have now evaluated the impact of intracellular compartmentation (ie nucleus or cytoplasm) of HSV-1 TK-GFP on the bystander effect of purine and pyrimidine nucleoside analogues. Our experimental data revealed that the intracellular localization of HSV-1 TK-GFP or VZV TK-GFP expression has no significant influence on either the cytostatic effect or the bystander effect of antiviral nucleoside analogues. These findings argue against the compartmentation of nucleotide pools in mammalian cells and suggest that phosphorylated nucleoside anabolites can rapidly equilibrate between the nuclear and cytosolic compartments of the cell.

II. Results A. Intracellular targeting of HSV-1 TKGFP and VZV TK-GFP constructs The HSV-1 TK-GFP and "(AA1-34)HSV-1 TK-GFP gene constructs were stably introduced in OstTK - cells and the fluorescence pattern was subsequently visualized under a fluorescence microscope. The OstTK-/HSV-1 TK-GFP+ cell line, as described earlier by Degrève et al. (1998), expresses the wild-type HSV-1 TK-GFP fusion protein, which is targeted to the nucleus of the transfected cells (Figure 2, panel A). In contrast, OstTK-/"(AA134)HSV-1 TK-GFP + cells express an N-truncated form of HSV-1 TK-GFP in both the nucleus and cytosol (panel B). Transfection of the VZV TK-GFP gene construct (panel C) 125

gave rise to a uniformly distributed fluorescence pattern. Finally, ligation of the HSV-1 TK nuclear localization signal to the VZV TK-GFP construct resulted in a nuclear fluorescence pattern (panel D).

B. Effect of intracellular localization of HSV-1 TK-GFP and VZV TK-GFP on the cytostatic activity of antiviral compounds The cytostatic activity of a series of the antiherpetic pyrimidine and purine nucleoside analogues was evaluated against OstTK - cells stably expressing either HSV-1 TKGFP, "(AA1-34)HSV-1 TK-GFP, VZV TK-GFP or NLSVZV TK-GFP. Non-transfected OstTK- cells were included as a control. The selection of the compounds was based on previous studies on HSV TK and VZV TK gene-transfected tumor cells in our laboratory (i.e. the protoype antiherpetic pyrimidine nucleoside analogue (E)-5-(2-bromovinyl)-2’deoxyuridine (BVDU) and its closely related derivatives SBVDU, BVaraU and BVDC, the antiherpetic thymidine and cytidine analogues araT, FMAU and FIAC, the acyclic guanosine analogue 9-(1,3-dihydroxy-2-propoxymethyl) guanine (ganciclovir, GCV) and its derivatives ACV, BCV, LBV and PCV (Balzarini et al., 1985, 1987, 1993, 1994; Degrève et al., 1997). The structural formulae of 4 representative antiherpes nucleoside analogues are shown in Figure 1. Results are summarized in T a b l e 1. The pyrimidine nucleosides BVDU, S-BVDU and BVaraU inhibited non-transfected OstTK- cell proliferation only at concentrations that exceeded 850µM. BVDC and araT showed 50% inhibitory concentrations (IC50) still above 200 µM, while FIAC and FMAU were more inhibitory to the proliferation of OstTK- cells (IC50 values of 9 and 17 µM, respectively). In sharp contrast, the pyrimidine nucleoside analogues became exquisitely inhibitory after transfection of the osteosarcoma cells with the HSV-1 TKGFP, "(AA1-34)HSV-1 TK-GFP, VZV TK-GFP and NLS-VZV TK-GFP genes. The IC50 values for the individual compounds were essentially comparable for the two HSV-1 TK-GFP constructs and the two VZV TKGFP constructs, except for FIAC which displayed a sixfold lower inhibitory effect against OstTK-/"(AA134)HSV-1 TK-GFP+ cells than against OstTK -/HSV-1 TKGFP + cells. BVDU and BVDC (IC50 values for TK-GFP gene-transfected cells ranging from 0.035 to 0.36 µM) exhibited 50% inhibitory concentrations that were approximately 10-fold higher than those for the other pyrimidine nucleoside analogues, which were in the lower nanomolar concentration range. The highest selectivity indices (i.e. the ratio of the IC 50 value for non-transfected cells versus the IC 50 value for TK-GFP gene-transfected cells) were observed for BVaraU (up to 250,000), S-BVDU (up to 150,000) and AraT (up to 100,000). BVDU was intermediate (selectivity index of 20,000), while BVDC, FIAC and FMAU were 1,000 to 6,000-fold more

Degrève et al: Antiherpetic drugs in combined gene/chemotherapy of cancer cytostatic to the various HSV-1 TK-GFP fusion genetransfected cells than to non-transfected OstTK- cells. The purine nucleoside analogues that were included in our study exhibited 50% inhibitory concentrations for the growth of non-transfected OstTK- cells ranging from 18 µM (LBV) to 231 µM (PCV) ( Table 1). GCV, BCV and PCV showed IC50 values in the nanomolar concentration range for OstTK-/HSV-1 TK-GFP+ and OstTK-/"(AA134)HSV-1 TK-GFP+ cells, that is at concentrations that

were 15,000 to 47,000-fold lower than the concentrations required to inhibit the proliferation of the wild-type OstTKcells. ACV, which displayed the highest IC50 value among all antiherpetic nucleoside analogues (up to 0.14 µM) and LBV (due to its stronger inhibitory effect against nontransfected OstTK- cells) ranked among the compounds with the lowest selectivity index (1,000 and 2,000, respectively). As proved to be the case with the pyrimidine

F i g u r e 2 . The HSV-1 TK-GFP and VZV TK-GFP fusion constructs (shown on top of each picture) were transfected into OstTKcells. After selection of stable transfectants, the fluorescence pattern was evaluated using a FITC filter-equipped fluorescence microscope. (A) HSV-1 TK-GFP; (B) "(AA1-34)HSV-1 TK-GFP; (C) VZV TK-GFP; (D) NLS-VZV TK-GFP.

126

Gene Therapy and Molecular Biology Vol 1, page 127 T a b l e 1 . C y t o s t a t i c a c t i v i t y o f n u c l e o s i d e a n a l o g u e s a g a i n s t w i l d - t y p e ( O s t T K -) and TK-GFP gene-transfected OstTK - c e l l s IC50 (µM)a

BVDU S-BVDU

OstTK-

OstTK- /HSV-1 TK-GFP+

862 ± 192 b b

911 ± 105 b

BVaraU

942 ± 47

BVDC

209 ± 35 b

OstTK- /"(AA1-34) HSV-1 TK-GFP+

OstTK- /VZV TKGFP+

OstTK- /NLS-VZV TK-GFP+

0.035 ± 0.006 b

0.038 ± 0.022

0.091 ± 0.055

0.36 ± 0.13

0.008 ± 0.004

b

0.006 ± 0.001

0.007 ± 0.000

0.028 ± 0.026

0.004 ± 0.001

b

0.004 ± 0.002

0.009 ± 0.017

0.029 ± 0.024

0.10 ± 0.00

1.6 ± 0.7c,d

-

0.059 ± 0.019 b

araT

231 ± 27

b

0.004 ± 0.0006

FIAC

9.1 ± 6.7

FMAU

b

c,d

0.002 ± 0.000

0.78 ± 0.41

0.002 ± 0.0001

0.012 ± 0.001

-

-

17 ± 0.5

0.006 ± 0.0001

0.004 ± 0.002

-

-

GCV

44 ± 22 b

0.001 ± 0.0005 b

0.003 ± 0.002

6.3 ± 7.3

ACV

b

73 ± 29

0.059 ± 0.015

BCV

173 ± 67

LBV

18 ± 0.4 b

PCV

231 ± 13

b

b

b

0.006 ± 0.0000

0.008 ± 0.0008 b 0.013 ± 0.0022

b

14 ± 4

48 ± 12

d

-

0.004 ± 0.001

57 ± 10

d

-

0.008 ± 0.0002

4.4 ± 1.0 d

0.14 ± 0.04 b

-

0.009 ± 0.001

27 ± 4

d

-

a The IC50 was defined as the drug concentration required to inhibit cell proliferation by 50%. Data are the mean value (± SD) for at least 3 independent experiments. b Data taken from Degrève et al. (1999). c Data taken from Degrève et al. (1997), where non-fused VZV TK gene-transfected OstTK- cells were evaluated.

nucleoside analogues, the IC50 values of the purine nucleoside analogues did not depend on the intracellular compartment in which the TK was localized (T a b l e 1). The poor cytostatic effect of ganciclovir against OstTKcells expressing VZV TK-GFP and NLS-VZV TK-GFP was not unexpected, since this drug has poor, if any, affinity for VZV TK (Degrève et al., 1997).

C. Bystander effect The bystander effect of two pyrimidine (BVDU and S-BVDU) and two purine (GCV and LBV) nucleoside analogues was evaluated. We have recently demonstrated the superior bystander effect of purine versus pyrimidine nucleoside analogues in mixed OstTK- and OstTK-/HSV-1 TK-GFP+ tumor cell populations (Degrève et al., 1999). Mixed tumor cell populations were cultured in the presence of 5-fold dilutions of the test compounds, after which the viable cell number was assessed using a colorimetric assay, as described in Materials and methods (Figure 3). The thick line in each graph represents the theoretically predicted values, in case no bystander effect is active (for

127

example, 25% non-transfected cells in the mixed tumor cell culture should result in 25% cell viability at the end of the 3-day incubation period in the presence of a lethal concentration of the nucleoside analogue). As shown in Figure 3, the inefficient bystander effect exerted by BVDU and S-BVDU was not enhanced by changing the intracellular HSV-1 TK-GFP localization. The very weak bystander effect of BVDU in OstTK-/HSV-1 TK-GFP+ cells was even completely absent in OstTK-/"(AA134)HSV-1 TK-GFP+ cells. For S-BVDU, the cell viability curves, obtained using a colorimetric assay, exactly reflected the percentages of OstTK- and HSV-1 TK-GFP gene-transfected tumor cells (Figure 3). In sharp contrast with the pyrimidine nucleosides, the guanosine nucleoside analogues GCV and LBV exhibited a pronounced bystander effect which was dose-dependent. LBV was not tested at 50µM because of profound inhibition of OstTK- cell growth at this concentration (IC50 value, 18µM). Even at the lowest concentration tested (2 µM), bystander killing was still observed with GCV and LBV. At a concentration

Gene Therapy and Molecular Biology Vol 1, page 128

F i g u r e 3 . Bystander effect of nucleoside analogues in mixed cell cultures. The thick line in each graph represents the theoretically predicted values, in case no bystander effect was noted. Concentrations tested, 50 µM (squares), 10 µM (triangles), 2µM (circles). OstTK - were mixed with OstTK- /HSV-1 TK-GFP+ cells (black symbols, data taken from Degrève et al., 1999) or OstTK- /"(AA1-34)HSV-1 TK-GFP+ cells (open symbols).

fusion proteins were localized in the nucleus of HSV-1 TK-GFP gene-transfected tumor cells, in the cytosol of HSV-2 TK-GFP gene-transfected tumor cells and in both the nucleus and the cytosol of VZV TK-GFP genetransfected tumor cells. The N-terminal 34 amino acids of HSV-1 TK, the deletion of which resulted in the loss of specific nuclear localization of HSV-1 TK-GFP, were also sufficient to target the otherwise uniformly distributed VZV TK-GFP to the nucleus of gene-transfected cells. In the experiments described in this report, we evaluated whether the intracellular localization of HSV-1 TK-GFP or VZV TK-GFP would influence the cytostatic potential and bystander effect of the antiherpetic nucleoside analogues in TK-GFP gene-transfected osteosarcoma cells. As shown in Table 1, all evaluated nucleoside analogues showed exquisite cytostatic properties against HSV-1 TK-GFP and

of 10µM, the bystander effect of LBV was 2- to 3-fold more pronounced than that of the prototype compound GCV. Abolishing the specific nuclear localization of HSV-1 TK-GFP by deleting the N-terminal NLS, had no significant impact on the bystander effect of GCV and LBV, that was observed in the tumor cell cultures that expressed the HSV-1 TK-GFP solely in the nucleus (Figure 3).

III. Discussion We recently reported the differential intracellular localization of the TKs of three herpesviruses, i.e. HSV-1, HSV-2 and VZV in TK-GFP fusion gene-transfected osteosarcoma cells (Degrève et al., 1998). The TK-GFP 128

Gene Therapy and Molecular Biology Vol 1, page 129 VZV TK-GFP expressing tumor cells, the lowest 50% inhibitory concentrations being in the lower nanomolar range. The pronounced cytostatic effect of pyrimidine nucleosides like S-BVDU and BVaraU makes them promising candidate compounds for the combined gene/chemotherapy treatment of cancer, with selectivity indices markedly higher than that of GCV, the current drug of choice for HSV-1 TK gene-mediated tumor cell killing. Moreover, S-BVDU and BVaraU are resistant to glycosidic bond cleavage by mammalian dThd phosphorylases, a major advantage compared to the cleavage-susceptible parent compound BVDU. However, one should also take the bystander effect into account. The bystander effect was described as the ability of a drug to kill non-transfected tumor cells that were in close contact with HSV-1 TK gene-transfected cells in mixed tumor cell populations. Complete tumor eradication has been demonstrated with GCV even when as few as 10% of the tumor cell inoculum was transfected with the HSV-1 TK gene (Culver et al., 1992; Ram et al., 1993; Freeman et al., 1993). The succes of the combined herpesviral TK gene/chemotherapeutic approach seems to depend on the bystander effect, since current gene therapy vectors are not capable of introducing the viral thymidine kinase gene in 100% of the cells of a particular tumor. Instead, getting 1% of the tumor cells transfected is a more realistic goal. Moreover, the low fraction of herpesviral TK gene-transfected tumor cells should be uniformly distributed in the tumor to yield an optimal bystander effect, which is virtually impossible to achieve. We recently demonstrated that the in vitro bystander effect of purine nucleoside analogues was superior to that of pyrimidine nucleoside analogues in mixed OstTK- and OstTK-/HSV-1 TK-GFP+ cell cultures. The bystander effect exerted by pyrimidine nucleoside analogues (i.e., BVDU and derivatives) proved to be very ineffective or even absent in most cases. In contrast, most of the evaluated purine nucleoside analogues (in particular GCV and LBV) displayed potent bystander killing potencies. Therefore, the lower selectivity index of purine nucleoside analogues like GCV and LBV is well compensated by their superior bystander killing as compared to pyrimidine nucleoside analogues. It is clear from Table 1 that the cytostatic activities of the evaluated compounds were generally independent of the compartmentation of the HSV-1 TK-GFP in the cell. We could also conclude from our experiments that the Nterminal 34 amino acids of the HSV-1 TK are not important for enzyme activity, since the "(AA1-34)HSV-1 TK-GFP fusion protein was fully catalytically active in gene-transfected tumor cells. These findings corroborate the observation of Halpern and Smiley (1984) that the Nterminal 45 amino acids are not required for the catalytic activity of HSV-1 TK. The slightly higher IC50 values (at 129

most 4-fold) obtained for OstTK-/NLS-VZV TK-GFP+ cells compared with OstTK -/VZV TK-GFP+ cells could be attributed to differences in the expression level of the VZV TK-GFP and the NLS-VZV TK-GFP fusion proteins, rather than to the different localization of VZV TK-GFP in the tumor cells. Indeed, the lower expression of NLS-VZV TK-GFP is in agreement with a weaker fluorescence signal (Figure 2, panel C and D). Thus, the intracellular localization of the VZV TK-GFP fusion protein is not a determining factor in the cytostatic potency of the antiviral nucleoside analogues. These findings are in full agreement with the observations of Johansson et al. (1997) on the cytostatic activities of nucleoside analogues against 2’-deoxycytidine kinase (dCK) expressing tumor cell lines. Until recently, it had generally been assumed that enzymes required for nucleic acid synthesis (i.e., nucleoside kinases) are localized in the cytosol (dCK and TK1) or in the mitochondria [i.e., 2’-deoxyguanosine kinase (dGK) and TK2] (Arnér and Eriksson, 1995). However, dCK, that shows substrate specificity for 2’-deoxycytidine, 2’deoxyadenosine and several clinically important nucleoside analogues, has now been found to be predominantly localized in the nuclear compartment (Johansson et al., 1997). Moreover, Johansson and collaborators identified a nuclear targeting signal in the primary structure of human dCK and showed that this signal was required for nuclear import of the protein. Irrespective of the intracellular localization of dCK, no marked differences in the cytostatic activity of 1-!-D-arabinofuranosylcytosine (araC), 2’,3’dideoxy-2’,3’-difluorocytidine (dFdC, gemcitabine) and 2chloro-2’-deoxyadenosine (CdA) were noted. These data indicate that the nucleus and the cytosol do not have separate deoxynucleotide pools, and that phosphorylated nucleoside analogues are rapidly equilibrated between the nuclear and cytosolic compartments of the cell. Since the localization of nucleoside kinases in the cell does not seem to have any determining role as to their function, it is currently unclear why certain nucleoside kinases are localized in the nucleus (dCK, HSV-1 TK), others in the cytosol (HSV-2 TK, mammalian cytosolic TK1) and still others spread over the nucleus and cytosol (VZV TK). In conclusion, we found that the inhibitory effects of antiviral nucleoside analogues against herpes TK-GFP gene-transfected cells were not significantly altered by changing the intracellular localization (either nucleus or throughout the cell) of the HSV-1 TK-GFP or VZV TKGFP fusion protein. Also, the bystander effect of the antiviral nucleoside analogues was not affected by the intracellular targeting of HSV-1 TK-GFP. Our experimental data indicate that phosphorylated nucleoside analogues can rapidly equilibrate between the nuclear and cytosolic compartments of the cell before exerting their potent cytostatic effect.

Degrève et al: Antiherpetic drugs in combined gene/chemotherapy of cancer

IV. Materials and methods A. Compounds BVDU and BVDC were synthesized by P. Herdewijn and A. Van Aerschot at the Rega Institute for Medical Research (Katholieke Universiteit Leuven, Leuven, Belgium). S-BVDU was provided by the late R.T. Walker (University of Birmingham, Birmingham, U.K.). BVaraU was a kind gift of H. Machida (Yamasa Shoyu Co., Choshi, Japan). AraT was from Sigma Chemical Co. (St. Louis, MO), and also a kind gift from M. Sandvold and F. Myhren (Norsk Hydro, Porsgrunn, Norway). FIAC and FMAU were a kind gift of J.J. Fox (SloanKettering Institute, New York). GCV was from Syntex (Palo Alto, CA), ACV from the former Wellcome Research Laboratories (Research Triangle Park, NC), BCV from Astra Läkemedel (Sodertälje, Sweden) and LBV from Bristol-Myers Squibb (Princeton, NJ). PCV was obtained from I. Winkler (Hoechst, Frankfurt, Germany).

B. Cell Culture Adherent human osteosarcoma cells deficient in cytosol TK (OstTK- , ATCC CRL-8303) and all TK-GFP genetransfected OstTK- cells were maintained at 37°C in a humidified CO2 -controlled atmosphere, in MEM culture medium (Gibco, Paisley, U.K.), supplemented with 10% heatinactivated fetal calf serum (Biochrom KG, Berlin, Germany), 2mM L-glutamine (Gibco), 0.075% (w/v) NaHCO3 (Gibco), 0.5µl/ml geomycine (Gentamycin#, 40mg/ml, ScheringPlough, Madison, NJ) and 0.5µl/ml Amphotericin B (Fungizone#, 5mg/ml, Bristol-Myers Squibb).

C. Plasmid construction The construction of the HSV-1 TK-GFP, "(AA1-34)HSV-1 TK-GFP, VZV TK-GFP and NLS-VZV TK-GFP expression vectors has been described elsewhere (Degrève et al., 1998). Briefly, the coding sequence for the full-length and Ntruncated HSV-1 TK (lacking the first 34 amino acids) were amplified by PCR from the pMCTK plasmid kindly provided by Dr. D. Ayusawa (Yokohama City University, Japan), and cloned in the multiple cloning site of the pEGFP-N1 NTerminal Protein Fusion Vector (CLONTECH, Palo Alto, CA). The VZV TK coding sequence, PCR-amplified from the pRc/CMV/VZV TK plasmid (kindly provided by Dr. J. Piette, University of Liège, Belgium) was ligated with (NLS-VZV TKGFP) or without (VZV TK-GFP) the PCR-amplified sequence encoding for AA1-34 of HSV-1 TK in the pEGFP-N1 vector.

D. Stable transfection of tumor cells The construction of the OstTK - /HSV-1 TK-GFP+, OstTK/ "(AA1-34)HSV-1 TK-GFP+, OstTK- /VZV TK-GFP+ and OstTK- /NLS-VZV TK-GFP+ cell lines has been described elsewhere (Degrève et al., 1998). Briefly, the herpes virus TKGFP fusion constructs were introduced into OstTK - cells via membrane fusion-mediated transfer using plasmid/ liposome complexes (LipofectAMINE$ Reagent, Gibco), as described by the supplier. Stable fusion gene transfectants were isolated by maintaining the cell cultures in the presence of HAT

130

medium (i.e. normal growth medium, supplemented with 100µM hypoxanthine, 0.4µM aminopterin and 16µM thymidine). Monoclonal transfected cell lines were obtained by plating the cells at clonal density in tissue culture plates (Corning, N.Y.), after which single colonies were isolated and expanded. A standard FITC filter-equipped fluorescence microscope was used to evaluate gene expression and GFP fusion protein localization.

E. Inhibition of tumor cell proliferation by antiherpetic compounds The cytostatic activity of antiviral nucleoside analogues against wild-type and herpes TK-GFP-expressing cells was evaluated as follows. 10 4 OstTK- , OstTK- /HSV-1 TK-GFP+, OstTK- /"(AA1-34)HSV-1 TK-GFP+, OstTK - /VZV TK-GFP+ or OstTK- /NLS-VZV TK-GFP+ cells/well were plated in 96-well microtiter plates (Falcon, Becton Dickinson, Franklin Lakes, NJ, USA) and subsequently incubated at 37°C, in a humidified CO2 -controlled atmosphere, in the presence of 5-fold dilutions (in normal growth medium) of the compounds. After 3 days, the number of cells was evaluated in a Coulter Counter (Coulter Electronics Ltd., Harpenden Hertz, U.K.). The IC50 was defined as the drug concentration required to inhibit tumor cell proliferation by 50%.

F. Bystander effect The procedure to evaluate the bystander effect of the compounds was as described elsewhere (Degrève et al., 1998). Briefly, OstTK- cells were mixed with HSV-1 TK-GFP genetransfected cells in percentages ranging from 0 to 100% (0, 0.2, 1, 5, 10, 25, 50, 75, 90 and 100%) transfected cells, and subsequently incubated in the presence of 5-fold dilutions (in 2% FCS-containing medium) of the compounds. After 3 days, i.e. the time needed by untreated cell cultures to reach confluency, cell viability was determined using the Cell Titer 96 Aqueous Non-radioactive MTT Cell Proliferation Assay (Promega, Madison, WI). Untreated cultures served as controls.

Acknowledgments We thank Christiane Callebaut for dedicated editorial help. This work was supported by Project 3-0180-95 from the Flemish “Fonds Voor Geneeskundig Wetenschappelijk Onderzoek”, Project 95/5 from the Belgian “Geconcerteerde Onderzoeksacties”, the Swedish Medical Research Council, the Medical Faculty of Karolinska Institute and the Harald and Greta Jeansson Foundation. Bart Degrève is recipient of an IWT fellowship from the “Vlaams Instituut voor de bevordering van het Wetenschappelijk-Technologisch onderzoek in de Industrie”.

References Arnér, E.S.J., and S. Eriksson. (1 9 9 5 ) Mammalian deoxyribonucleoside kinases. Pharmacol. Ther. 67,

Gene Therapy and Molecular Biology Vol 1, page 131 155-186.

cancer. Gene Ther. 4, 1107-1114.

Balzarini, J., C. Bohman, and E. De Clercq. (1 9 9 3 ) Differential mechanism of cytostatic effect of (E)-5-(2bromovinyl)-2’-deoxyuridine, 9-(1,3-dihydroxy-2propoxymethyl)guanine, and other antiherpetic drugs on tumor cells transfected by the thymidine kinase gene of herpes simplex virus type 1 or type 2. J . B i o l . C h e m . 268, 6332-6337.

Degrève, B., M. Johansson, E. De Clercq, A. Karlsson, and J. Balzarini. (1 9 9 8 ) Differential intracellular compartmentalization of herpetic thymidine kinases (TKs) in TK gene-transfected tumor cells. Molecular characterization of the nuclear localization signal of herpes simplex virus type 1 TK. J . V i r o l . 72, 95359543.

Balzarini, J., C. Bohman, R.T. Walker, and E. De Clercq. (1 9 9 4 ) Comparative cytostatic activity of different antiherpetic drugs against herpes simplex virus thymidine kinase gene-transfected tumor cells. M o l . P h a r m a c o l . 45, 1253-1258.

Freeman S.M., C.N. Abboud, K.A. Whartenby, C.H. Packman, D.S. Koeplin, F.L. Moolten, and G.N. Abraham. (1 9 9 3 ) The “bystander effect”, tumor regression when a fraction of the tumor mass is genetically modified. C a n c e r R e s . 53, 5274-5283.

Balzarini, J., E. De Clercq, A. Verbruggen, D. Ayusawa, K. Shimizu, and T. Seno (1 9 8 7 ) Thymidylate synthase is the principal target enzyme for the cytostatic activity of (E)5-(2-bromovinyl)-2’-deoxyuridine against murine mammary carcinoma (FM3A) cells transformed with the herpes simplex virus type 1 and type 2 thymidine kinase gene. M o l . P h a r m a c o l . 32, 410-416.

Freeman, S.M., C. McCune, W. Robinson, C.N. Abboud, G.N. Abraham, C. Angel, and A. Marrogi. (1 9 9 5 ) The treatment of ovarian cancer with a gene modified cancer vaccine, A phase I study. H u m . G e n e T h e r . 6, 927939.

Balzarini, J., E. De Clercq, D. Ayusawa, and T. Seno. (1 9 8 5 ) Murine mammary FM3A carcinoma cells transformed with the herpes simplex virus type 1 thymidine kinase gene are highly sensitive to the growth-inhibitory properties of (E)-5-(2-bromovinyl)-2’-deoxyuridine and related compounds. FEBS Lett. 185, 95-100. Chalfie, M., Y. Tu, G. Euskirchen, W.W. Ward, and D.C. Prasher. (1 9 9 4 ) Green fluorescent protein as a marker for gene expression. S c i e n c e 263, 802-805. Culver, K.W., J. Van Gilder, C.J. Link, T. Carlstrom, T. Buroker, W. Yuh, K. Koch, K. Schabold, S. Doornbas, and B. Wetjen. (1 9 9 4 ) Gene therapy for the treatment of malignant brain tumors with in vivo tumor transduction with the herpes simplex thymidine kinase gene/ganciclovir system. Hum. Gene Ther. 5, 343379. Culver, K.W., Z. Ram, S. Wallbridge, H. Ishii, E.H. Oldfield, and R.M. Blaese. (1 9 9 2 ) In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors. S c i e n c e 256, 1550-1552. De Clercq, E. (1 9 9 3 ) Antivirals for the treatment of herpesvirus infections. J . A n t i m i c r o b . C h e m o t h e r . 32, 121-132. De Clercq, E. (1 9 9 5 ) Trends in the development of new antiviral agents for the chemotherapy of infections caused by herpesviruses and retroviruses. R e v . M e d . V i r o l . 5, 149-164. Degrève, B., E. De Clercq, and J. Balzarini. (1 9 9 9 ) Bystander effect of purine nucleoside analogues in HSV-1 TK suicide gene therapy is superior to that of pyrimidine nucleoside analogues. Gene Ther. 6, in press. Degrève, B., G. Andrei, M. Izquierdo, J. Piette, K. Morin, E.E. Knaus, L.I. Wiebe, I. Basrah, R.T. Walker, E. De Clercq, and J. Balzarini. (1 9 9 7 ) Varicella-zoster virus thymidine kinase gene and antiherpetic pyrimidine nucleoside analogues in a combined gene/chemotherapy treatment for

131

Halpern, M.E., and J.R. Smiley. (1 9 8 4 ) Effects of deletions on expression of the herpes simplex virus thymidine kinase gene from the intact viral genome, the amino terminus of the enzyme is dispensable for catalytic activity. J . V i r o l . 5 0 , 733-738. Johansson, M., S. Brismar, and A. Karlsson. (1 9 9 7 ) Human deoxycytidine kinase is located in the cell nucleus. P r o c . N a t l . A c a d . S c i . USA 94, 11941-11945. Kun, L.E., A. Gajjar, M. Muhlbauer, R.L. Heideman, R. Sanford, M. Brenner, A. Walter, J. Langston, J. Jenkins, and S. Facchini. (1 9 9 5 ) Stereotactic injection of herpes simplex thymidine kinase vector producer cells (PA317G1Tk1SvNa.7) and intravenous ganciclovir for the treatment of progressive or recurrent primary supratentorial pediatric malignant brain tumors. Hum. Gene Ther. 6, 1231-1255. Oldfield, E.H. 1993. Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Clinical protocols. Hum. Gene Ther. 4, 36-69. Raffel, C., K. Culver, D. Kohn, M. Nelson, S. Siegel, F. Gillis, C.J. Link, and J.G. Villablanca. (1 9 9 4 ) Gene therapy for the treatment of recurrent pediatric malignant astrocytomas with in vivo tumor transduction with the herpes simplex thymidine kinase gene/ganciclovir system. Hum. Gene Ther. 5, 863-890. Ram Z., K.W. Culver, S. Walbridge, R.M. Blaese, E.H. Oldfield. (1 9 9 3 ) In situ retroviral-mediated gene transfer for the treatment of brain tumors in rats. C a n c e r R e s . 53, 83-88. Rizzuto, R., M. Brini, P. Pizzo, M. Murgia, and T. Pozzan. (1 9 9 5 ) Chimeric green fluorescent protein as a tool for visualizing subcellular organelles in living cells. Curr. B i o l . 5, 635-642. Youvan, D.C., and M.E. Michel-Beyerle. (1 9 9 6 ) Structure and fluorescence mechanism of GFP. Nature B i o t e c h n o l o g y 14, 1219-1220.

Degrève et al: Antiherpetic drugs in combined gene/chemotherapy of cancer

132

Gene Therapy and Molecular Biology Vol 1, page 133 Gene Ther Mol Biol Vol 3, 133-148. August 1999.

Glioblastoma multiforme: molecular biology and new perspectives for therapy Review Article

Giorgio Pal첫, Luisa Barzon, and Roberta Bonaguro Institute of Microbiology, University of Padova Medical School, Padova, Italy __________________________________________________________________________________________________ Corresponding Author: Giorgio Pal첫, MD, Institute of Microbiology, Via A. Gabelli 63, 35121 Padova, Italy. Tel: +39-049-8272350; Fax: +39-049-8272355; E-mail: gpalu@microb.unipd.it K e y w o r d s : Therapy, gene therapy, brain tumors, gliomas, glioblastoma multiforme, molecular biology, pathogenesis, immunotherapy, neoangiogenesis, oncogenes A b b r e v i a t i o n : GBM, glioblastoma multiforme Received: 23 November 1998; accepted 30 November 1998

Summary Pathogenic features of glioblastoma multiforme and of other gliomas are reviewed in the present article. Emphasis is given to those genetic alterations which are involved in oncogenesis, to the p r o c e s s o f t u m o r n e o a n g i o g e n e s i s a n d t o t h e r o l e p l a y e d b y t h e i m m u n e s y s t e m i n controlling neoplastic growth. Aspects which are relevant to therapeutic interventions are also dissected, and gene therapy in particular. A new gene therapy approach that combines tumor suicide, via enzymedirected prodrug activation, and cytokine-promoted immune rejection i s reported, together with results from the first application of this approach in humans.

of patients for tumor recurrence within 6-12 months from treatment.

I. Introduction The outcome of malignant gliomas remains extremely poor, in spite of aggressive use of currently available therapies. Recent advances in elucidating the molecular biology of gliomas have led to the development of innovative therapeutic strategies. The more promising approaches involve gene therapy, aiming at increasing tumor cell chemosensitivity and/or immunogenicity, by transfer of genes expressing cytokines and prodrug activating enzymes.

Glioblastoma multiforme (grade IV astrocytoma) is usually located in the cerebral hemispheres, though it occasionally appears at other sites, such as the cerebellum, the brain stem and the spinal cord. Histology shows marked cytological diversity, ranging from tumors composed of small cells with scant cytoplasm to those composed of multinucleated giant cells. The World Health Organization (WHO) classification recognizes two distinct subvariants of the tumor: (i ) giant cell glioblastoma, characterized by a predominance of enormous, multinucleated giant cells and, on occasion, an abundant stromal reticulin network; and (i i ) gliosarcoma, in which hyperplastic vascular elements have undergone sarcomatous transformation.

Glioblastoma multiforme (GBM) represents 15-20% of all intracranial tumors and 50% of gliomas (Russel and Rubistein, 1989). It affects 5,000 Americans and 1,000 Italians every year, and typically occurs in adults, with a peak incidence in the fifth and sixth decades of life. It is a very aggressive tumor, with a uniform and profound morbidity. Because of its morbidity it contributes to the cost of cancer on a pro capite basis more than any other tumor. Despite surgery, radiotherapy and/or chemotherapy, the prognosis is extremely poor and has not substantially changed over the last two decades, death resulting in 80%

Current therapies for malignant gliomas include surgical removal of the tumor mass, which is mandatory for precise diagnosis, and irradiation. Although surgery improves the prognosis (Levin et al, 1993), the infiltrative behavior of malignant gliomas precludes their complete

133

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme resection, and 90% of GBM recur within 2 cm of the primary site. Postoperative radiotherapy is therefore commonly administered, with a significant improvement in survival (Hochberg and Pruitt, 1980; Walker et al, 1980). Despite surgery and irradiation, however, only a few patients are alive two years after diagnosis. Results of chemotherapy trials are disappointing (Hosli et al, 1998). This is due both to the tumor intrinsic chemoresistance (Petersdorf and Berger, 1996) and to the tumor location within the central nervous system, which limits the penetration of drugs (Janzer and Raff, 1987; Mak et al, 1995). Among malignant gliomas, GBM is the least responsive to medical treatment. Available protocols include both monochemotherapy and polychemotherapy regimens. Nitrosoureas are the leading drugs in glioma chemotherapy, with response rates as single agents varying from 10% to 40% (Young et al, 1973; Fewer et al, 1972; Hoogstraten et al, 1972). Other drugs, evaluated in monochemotherapy (Forsyth and Cairncross, 1996), occasionally showed clinically and radiologically objective responses. Among these are vincristine (Smart et al, 1968), procarbazine (Rodriguez et al, 1989), paclitaxel (Chamberlain and Kormanik, 1995; Prados et al, 1996), and temozolomide (Newlands et al, 1996). However, methodological bias present in most studies raise doubts about the validity of these results. The most commonly used polychemotherapy regimens for gliomas are PVC (i.e. a combination of CCNU, procarbazine, and vincristine) and MOP (i.e. a combination of procarbazine, vincristine, and mechlorethamine). Response rates (complete or partial) of 17-37% have been reported for glioblastomas (Levin et al, 1980; Coyle et al, 1990). More recently, interesting results have been obtained in GBM patients with the ICE regimen (ifosfamide, carboplatin and etoposide), although in association with severe hematological toxicity (Sanson et al, 1996). The role of PVC as adjuvant chemotherapy is controversial (Fine et al, 1993), and, overall, there is no clear-cut evidence that survival of glioblastoma patients is improved by chemotherapy (Hosli et al, 1998).

insurmountable task that gene replacement, or gene suppression, should simultaneously involve a number of different genes, and should be applied to all tumor cells to reverse the malignant phenotype. Hence, corrective gene therapy seems to be quite difficult to propose as a single therapeutic approach.

A. Genetic alterations 1. Oncogenes Several members of the protein-tyrosine kinase receptor family are over-expressed by gene amplification in malignant gliomas, including the epidermal growth factor receptor (EGFR), the platelet-derived growth factor receptor-! (PDGFR!) and the c-met genes (Furnari et al, 1995). A high percentage of glioblastomas also have EGFR gene rearrangements that may lead to the expression of a truncated, constitutively activated receptor. The transfer of a mutant human EGFR gene into glioblastoma cells caused constitutive self phosphorylation and a pronounced enhancement tumorigenicity in nude mice

II. Molecular biology of glioblastoma multiforme and corrective gene therapy As for most cancers, brain tumors derive from a multistep process of successive alterations, including loss of cell cycle control, neoangiogenesis and evasion of immune control. Figure 1 summarizes the genetic alterations associated with the malignant transformation of astrocytes. Most of these changes involve the loss of putative tumor suppressor genes or activation of proto-oncogenes.

o

Gene therapy of cancer, in its most direct form, should aim at replacing a mutated gene with its correct form, or at suppressing the abnormal oncogenic function. At present, however, such a corrective gene therapy, faces the

f

F i g u r e 1 . Simplified representation of oncogenes and tumor suppressor genes contributing to malignant progression of astrocytic tumors.

134

Gene Therapy and Molecular Biology Vol 1, page 135 (Ekstrand et al, 1994; Nishikawa et al, 1994). Numerous strategies are currently being investigated to specifically inhibit EGFR using antibodies, immunoconjugates or antisense technology. The selective inhibition of EGFR in human GBM cells with kinase-deficient mutants inhibited cell proliferation and transforming efficiency in athymic mice (O'Rourke et al, 1997), and converted radioresistant human glioblastoma cells to a more sensitive phenotype (O'Rourke et al, 1998), providing a rationale for gene therapy applications.

cells with wild type p53 can significantly inhibit growth and neoangiogenesis, or can induce apoptosis in p53 mutant cells in several tumor models in vitro, including gliomas (Badie et al, 1995; Van Meir et al, 1995; GomezManzano et al, 1996). The presence of functional p53 has also been shown to modulate chemoresistance. Consequently, another possible advantage of the restoration of wild type p53 may be sensitization to chemotherapy and radiotherapy. Indeed, the combination of p53 gene transduction with radiation or chemotherapy (Lowe et al, 1994) has resulted in local tumor control superior to either therapy alone (Fujiwara T et al, 1994; Gjerset et al, 1995; Ngyuyen et al, 1996, Lang et al, 1998). This combined therapy is currently under investigation in clinical trials (Roth and Cristiano, 1997; Nielsen and Maneval, 1998).

Other dominant oncogenes, such as N-myc, fos, src, Hras or N-ras, and mdm2 are amplified and highly expressed in gliomas (Collins, 1993). GBM produce high levels of insulin-like growth factor I (IGF-I). When this alteration has been targeted by a vector expressing an antisense antiIGF-I gene, rejection of genetically altered rat C6 glioma cells was observed. Injection, even at a site distal to the tumor, caused regression of established brain GBM. Destruction of the tumor was mediated by a gliomaspecific T CD8+ (CTL) response (Trojan et al, 1993).

The cell cycle regulator genes provide an additional target for corrective gene therapy. The p105Rb product of the retinoblastoma tumor suppressor gene (Rb) is one of the most critical regulators of cellular proliferation. The Rb protein (pRb), when unphosphorylated, is responsible for arrest of cell cycle by inhibition of the activity of the E2F family of transcription factors. Normal cell cycle progression requires inactivation of Rb through phosphorylation by cyclin-dependent kinases (CDK). This process, in turn, is regulated by CDK inhibitors. Among these, p21 protein is induced directly by p53; p16 protein, and its homologue p15, specifically bind to and inhibit CDK4, and may therefore regulate Rb phosphorylation, and cell cycle progression. Dysregulation of cell cycle control is a frequent finding in malignant gliomas, like deletion or loss of expression of p16 and p15 tumor suppressor genes (Jen et al, 1994; Nishikawa et al, 1995), amplification of CDK4 (He et al, 1994), and deletion or mutation of the Rb tumor suppressor gene (Henson et al, 1994). Interestingly, both of the latter events take place when the p16 gene is intact and correctly expressed (He et al, 1995). Restoration of wild-type p16 gene in glioma cells through an adenoviral vector arrested cells in G0-G1 phases of the cell cycle (Fueyo et al, 1996) and suppressed glioma cells invasion in vitro (Chintala et al, 1997). Overexpression of p21 increases the susceptibility of glioblastoma cells to cisplatin-induced apoptosis (Kondo et al, 1996), whereas adenovirus-mediated transfer of exogenous E2F-1 protein induced massive apoptosis and suppressed glioma growth in vivo and in vitro (Fueyo et al, 1998). The possibility that E2F-responsive promoters may be more active in tumor cells relative to normal cells, because of loss of pRb function, has been exploited to design adenoviral vectors containing transgenes driven by the E2F-1 promoter for gene therapy of gliomas (Parr et al, 1997). These vectors showed tumor-selective gene expression in vivo and reduced toxicity of the normal tissue with respect to standard adenoviral vectors.

2. Genes associated to cell immortalization A role in cell immortalization has been proposed for telomerase, the RNA-protein complex that elongates telomeric DNA. Telomerase is expressed almost exclusively in cancer cells, but not in normal cells, suggesting the possibility that gene therapy may be applied to inhibit this function. A successful example of treatment via antisense oligonucleotides directed against human telomerase suppressed glioma cells growth and survival, both in vitro and in vivo, through the induction of apoptosis (Kondo et al, 1998). 3. Tumor suppressor genes Molecular and cytogenetic analyses of gliomas have shown frequent losses of genetic material, suggesting the inactivation of putative tumor suppressor genes. Loss of heterozygosity (LOH) has been described in chromosome 1p, 9p, 10p, 10q, 11p, 13q, 17p, 19q, and 22q, and in some cases the tumor suppressor gene involved in LOH has been identified. This is the case of the p53 tumor suppressor gene, which maps in 17p. Wild-type p53 protein is involved in G1 cell cycle arrest and apoptosis of DNA-damaged cells and is therefore crucial in preventing mutation or deletion of functional genes. Mutations of p53 seem to be an early event in glioma tumorigenesis, being frequently detected also in low grade astrocytomas. Along with p53 mutations, amplification of the mdm2 oncogene, whose product binds to and degrades p53, accounts for p53 inactivation in gliomas. Since p53 plays a key role in the pathogenesis of most cancers, it has raised great interest as a target for cancer gene therapy. Transduction of malignant

135

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme Inhibition or inactivation of genes/factors involved in DNA repair and/or cellular SOS response could represent a gene therapy approach that potentiates radiation therapy. In fact, inhibition of the RAD51 gene by antisense oligonucleotides enhanced the radiosensitivity of mouse malignant gliomas, both in vitro and in vivo, improving survival (Ohnishi et al, 1998). This gene, a homologue of the yeast RAD51 and E. coli RecA genes, is involved in repair of DNA double-strand breaks, in recombination repair, and in various SOS responses to DNA damage caused by gamma-irradiation and alkylating agents.

ribozymes against VEGF mRNA have been successfully employed to reduce VEGF expression in glioma cells (Ke et al 1998), once more suggesting a potential role for antiangiogenic gene therapy. Similarly, bFGF antisense cDNA decreased C6 glioma cells proliferation (Redekop and Naus, 1995). Besides inhibiting the production of angiogenic factors, a therapeutic intervention could also consist of providing tumors with antiangiogenic factors. Indeed, retroviral and adeno-associated viral vectors expressing a modified PF4 were reported to inhibit endothelial cells proliferation in vitro and the growth of intracerebrally implanted gliomas (Tanaka et al, 1997). Retroviral and adenoviral vectors transducing angiostatin gene increased apoptotic death of glioma tumor cells (Tanaka et al, 1998). Additionally, the intratumoral delivery of angiostatin gene by an adenoviral vector produced inhibition of tumor growth in vivo, suppression of neovascularization, and a marked increase of tumor cells apoptosis (Griscelli et al, 1998). Damage of tumor microvasculature was reported also in human malignant glioma xenografts, after gene therapy followed by radiotherapy. The treatment consisted of intratumoral injection of adenoviral vectors expressing tumor necrosis factor-! (TNF-!), under control of the Egr-1 promoter (Staba et al, 1998). The use of viral vectors containing radiation-inducible promoters, such as Egr-1, has the advantage of selectively, spatially, and temporally limiting the effects of the therapeutic gene in the radiation field. Recently, this strategy has yielded interesting results in rat 9L glioma cells (Manome et al, 1998).

Deletions of large regions or even of the entire copy of chromosome 10 are a genetic hallmark of GBM. At least two tumor suppressor genes located on chromosome 10 (one on each arm) have been demonstrated to participate to glial oncogenesis. A first candidate tumor suppressor gene, called PTEN (Phosphatase and tensin homologue deleted on chromosome TEN) was recently characterized (Li et al, 1997). The DNA region encoding PTEN is altered in glioblastoma multiforme, but not in lower grade astrocytic tumors (Tohma et al, 1998; Ichimura et al, 1998; Chiariello et al, 1998).

B. Neoangiogenesis Tumors may remain in a state of dormancy until they establish a blood supply for receiving oxygen and nutrients. The complex process of neoangiogenesis is regulated by numerous factors, some with angiogenic properties, i.e. vascular endothelial growth factor (VEGF), platelet derived growth factor (PDGF!), basic fibroblast growth factor (bFGF), epidermal growth factor (EGF), interleukin-8, and by endogenous inhibitors of angiogenesis, i.e. thrombospondin-1, platelet factor 4 (PF4), angiostatin, endostatin. VEGF, which binds to two specific tyrosine-kinase receptors, called Flk-1 and Flt-1, has been demonstrated to play a key role in angiogenesis of gliomas. Indeed, VEGF and its receptors are downregulated in the normal adult brain, whereas, VEGF is highly produced by GBM cells. Since both flt-1 and flk-1 genes are expressed by proliferating endothelial cells of gliomas, this leads to the establishment of a paracrine loop. Moreover, VEGF expression is higher around necrotic areas and seems to be stimulated by hypoxia.

III. Suicide gene therapy Suicide gene therapy operates by tumor transduction of genes converting a prodrug into a toxic substance; independently, the gene product and the prodrug are nontoxic. The prototype of this approach exploits the selective intracellular phosphorylation of ganciclovir (GCV), driven by the herpes simplex virus thymidine kinase gene product (HSV-TK). This activation generates a toxic drug metabolite that inhibits DNA synthesis, inducing cell death. For in vivo gene transfer of the HSV-TK gene to malignant cells, packaging cells that produces retroviral vectors expressing HSV-TK, have been injected directly into the tumor to transduce replicating cells. An interesting feature of the HSV-TK/GCV system is the bystander killing of nontransduced cells.

Glioblastoma multiforme is one of the most highly vascularized solid neoplasms; therefore, treatments that target neoangiogenesis would be of great interest in clinical practice. Co-injection of rat C6 glioma cells, either subcutaneously or intracerebrally in nude mice, together with cells producing retroviral vectors encoding a dominant-negative mutant of the Flk-1 receptor showed inhibition of neoangiogenesis, reduction of tumor growth, and survival improvement (Millauer et al, 1994; 1996). Antisense VEGF oligonucleotides (Saleh et al, 1996) and

The mechanisms that are responsible for this effect have not been fully defined, but are likely to include: (i ) transfer of non-diffusible, phosphorylated GCV to neighboring cells through gap junctions; (i i ) endocytosis by nontransduced cells of cellular debris containing toxic GCV; and (i i i ) stimulation of host antitumor immune response. The therapeutic efficacy of the HSV-TK/GCV

136

Gene Therapy and Molecular Biology Vol 1, page 137 system may be further increased by the use of adenoviral vectors, since these vectors can also transduce resting tumor cells. However, adenoviral vectors will infect also normal cells; hence, the inclusion of sequences able to restrict gene expression only in tumor cells can circumvent this problem. Selective tumor toxicity was obtained positioning the suicide gene under control of the E2Fresponsive promoter elements which are de-repressed in glioma cells (Parr et al, 1997).

have shown encouraging results both in vitro and in vivo (Khil et al, 1995; Kim et al, 1997). 9L glioma cells transduced with a retrovirus encoding a CD/HSV-TK fusion gene exhibited enhanced sensitivity to both GCV and 5-FC, as well as increased radiosensitivity (Rogulski et al, 1997). This experiment suggests the feasibility of a combined approach with two suicide genes associated with radiotherapy. Another prodrug activation system is represented by cytochrome P450 2B1 (the liver enzyme catalyzing cyclophosphamide and ifosfamide activation) gene transfer followed by cyclophosphamide or ifosfamide administration. Metabolites of these drugs produce interstrand DNA cross-linking in a cell cycle-independent fashion. C6 and 9L rat glioma cells, when stably transfected with the P450 2B1 gene, become highly sensitive to cyclophosphamide in in vitro and in vivo models (Wei et al, 1994; Manome et al, 1996). Rabbit cytochrome P450 isozyme CYP4B1, which converts the inert prodrugs 2-amonianthracene (2-AA) and 4-ipomeanol (4-IM) into highly toxic alkylating metabolites, also showed high antitumor effects, both in vitro and in vivo. The treatment had relatively low toxicity and was associated with a bystander effect, not requiring cell-to-cell contact (Rainov et al, 1998).

The effectiveness of suicide gene therapy has been explored for a variety of neoplasms, especially for refractory and localized diseases such as GBM. The HSVTK/GCV scheme has been demonstrated to be effective in animal models, by ex vivo and in vivo transduction with retroviral, adenoviral, or adenoviral-associated vectors expressing HSV-TK (Ezzeddine et al, 1991; Culver et al, 1992; Takamiya et al, 1992, 1993; Barba et al, 1993, 1994; Ram et al, 1993; Kim et al, 1994; Vincent et al, 1996; Maron et al, 1996; Mizuno et al, 1998). Novel gene delivery systems were also applied in a mouse model for gene therapy of meningeal gliomatosis. Liposomes coated with Sendai virus envelope protein were highly efficient in delivering the therapeutic gene in disseminated glioma cells (Mabuchi et al, 1997). A factor that may limit the effectiveness of HSVTK/GCV therapy is the GCV crossing through the bloodbrain barrier. This can be circumvented by the use of the bradykinin analogue and potent blood-brain barrier permeabilizer RMP-7, which, administered intravenously, increase the delivery of GCV into rat brain tumors, enhancing the cytotoxic and bystander effects of HSVTK/GCV (LeMay et al, 1998).

Another suicide gene system is based on E. coli purine nucleoside phosphorylase (PNP), which generates toxic purine nucleoside analogues intracellularly, either from 6methylpurine-2’-deoxyriboside or arabinofuranosyl-2fluoroadenine monophosphate. Significant antitumor activity and low systemic toxicity were reported in nude mice bearing human malignant D54MG glioma tumors expressing PNP (Parker et al, 1997).

Several clinical trials exploiting the HSV-TK/GCV system have been initiated. It is too early to estimate the effectiveness of these therapeutic procedures. Notwithstanding the evidence for growth-suppressive activities of HSV-TK plus GCV, cure rates are low. Explanation for lack of complete response in humans may reside in the different biological behavior of GBM cells when injected into animals (Sturtz et al, 1997).

Phosphorylation of the prodrug cytosine arabinoside (ara-C) by deoxycyticine kinase (dCK) is a limiting step for activation. Thus, ara-C, a potent antitumor agent for hematological malignancies, has only minimal activity against most solid tumors. Transduction of the dCK cDNA by retroviral and adenoviral vectors also resulted in marked sensitization of glioma cell lines to ara-C in vitro, and in significant antitumor activity in vivo (Manome et al, 1996).

The antitumor effects elicited by HSV-TK/GCV prompted to explore other prodrug activating enzymes. A promising system exploits the ability of E. coli cytosine deaminase (CD) to convert the relatively non toxic 5fluorocytosine (5-FC) to the chemotherapeutic agent 5fluorouracil (5-FU). Significant antitumor effects of CD/5FC were observed in nude mice bearing tumors derived from C6 glioma cells and transduced with CD. A "bystander" effect could also be demonstrated (Ge et al, 1997), suggesting a potential role for gene therapy of glioblastoma.

Unlike other prodrug activating enzymes, E. coli gpt sensitizes cells to the prodrugs 6-thioxanthine (6TX) and 6thioguanine (6-TG), and confers resistance to different regimens (mycophenolic acid, xantine, and hypoxanthine), providing a means to select for gpt-positive cells. Rat C6 glioma cells transduced with a retroviral vector expressing the gpt gene exhibited significant 6TX and 6GT susceptibility and a "bystander" effect in vitro. An antiproliferative effect was demonstrated in vitro and in vivo (Tamiya et al, 1996; Ono et al, 1997).

As already reported for p53, both the CD and HSV-TK systems sensitize cancer cells to radiation. Animal models 137

Palù et al: Perspectives for therapy of glioblastoma multiforme Tumor suicide can also be achieved by direct infection of tumor cells with a conditionally replicative virus, i.e. an infectious agent able to replicate and to kill only dividing cells. In the case of tumors highly proliferating in the context of a completely post-mitotic tissue, such as the brain, gene transfer can ideally be obtained by using neurotropic herpes viral vectors, which are rendered conditionally replicative after deletion of non-essential genes (Lachmann and Efstathiou, 1997).

Despite the location in the central nervous system (CNS), a long-believed “immunologically privileged site”, glioblastoma cells may interact with immune cells. These interactions are mediated by receptor-ligand recognition during cell to cell contact and by a plethora of cytokines. An imbalance in the tumor-host relationship, resulting in deficit in some components of the response, may explain the aggressive growth of malignant gliomas. Before discussing the designed strategies to increase the immune response against glioblastoma cells, we review the more recent acquisitions in the “dialogue” between these neoplastic cells and the immune system (Dietrich et al, 1997).

Tumor specific cell death has already been provided in animal glioma models by HSV vectors, deleted in neurovirulence genes, such as " 34.5, thymidine kinase and ribonucleotide reductase (Chambers et al, 1995; Mineta et al, 1995; Boviatsis et al, 1994; Miyatake et al, 1997; McKie et al, 1996; Andreansky et al, 1996). Their direct injection into gliomas produced tumor regression with minimal bystander effects on surrounding normal tissue. Enhancement of replication of defective HSV vectors lacking " 34.5 gene and a significant reduction of tumor mass was observed combining ionizing radiation (Advani et al, 1998).

A. Antigenicity of glioblastoma cells The presence of specific antigens at the surface of tumor cells to be recognized by cells of the immune system is essential for the generation of a specific antitumor immune response. At present, no tumor antigen able to elicit an immune response has been identified in glioblastoma in vivo. MAGE-1 (melanoma antigen) was the first tumor-specific antigen to be identified (Van der Bruggen et al, 1991), and MAGE family members are expressed by some glioblastoma cell lines (Rimoldi et al, 1993), but not in uncultured tumors (De Smet et al, 1994). This observation could be explained with different levels of DNA methylation induced by culture, where MAGE expression was regulated by methylation (De Smet et al, 1995).

Another way to treat malignant gliomas emerged from the discovery that these tumors often express functional Fas (CD95) (Weller et al, 1994). Fas is a transmembrane glycoprotein belonging to the nerve growth factor/TNF receptor superfamily: when activated, Fas can transduce an apoptotic signal through its cytoplasmic domain. Apoptosis is triggered by the binding of Fas to its natural ligand (FasL) or by cross-linking with anti-Fas antibodies. A high proportion of human glioma cell lines are sensitive to apoptosis mediated by anti-Fas antibodies in vitro. Some other cell lines are resistant, but may be rendered sensitive after stable transfection with human Fas cDNA (Weller et al, 1995). These results offer new possibilities for treating gliomas with anti-Fas antibodies or soluble FasL. One possible drawback of such an approach is that other Fas-positive cells may be affected, like infiltrating leukocytes. Thus their activity may be reduced, restricting strategies relying upon simultaneous immune response enhancement.

Proteins that are structurally altered during malignant transformation, or that contribute to this process, are possible tumor antigen candidates. The frequent alterations of p53 in the early stages of carcinogenesis, for example, may provide new antigenic peptides that could trigger an immune response. Consistent with this possibility, specific cytotoxic T-cell clones were generated in vitro against mutated p53 protein (Houbiers et al, 1993). Additionally, in vivo immunization with mutated p53 peptide was shown to induce specific CTL clones able to lyse MHC-matched tumor cells expressing the mutated p53 gene (Noguchi et al, 1994).

IV. Immunotherapy of cancer

The future identification of glioblastoma-specific antigens would aid new treatment strategies.

Glioblastoma multiforme has the propensity to microscopically infiltrate normal structures since its early stage of development. This characteristic makes therapeutic success rather difficult by any approach. Moreover, it becomes virtually impossible to obtain specific targeting of all tumor cells and sparing of normal ones. Therefore, inducing an efficient immune response against malignant cells becomes an attractive and essential treatment strategy. In this perspective, the natural circulatory properties of cells of the immune system offer also an important support for the recognition of secondary lesions.

B. Tumor-induced immunosuppression A high proportion of glioblastomas have shown to be infiltrated by lymphocytes, mostly T lymphocytes, but also B and NK cells. Differently from what reported for other tumors, the level of lymphocyte infiltration does not relate with a better prognosis. In fact, tumor-infiltrating lymphocytes (TIL) of gliomas appear to be functionally defective. Abnormalities range from abnormal 138

Gene Therapy and Molecular Biology Vol 1, page 139 hypersensitivity responses, depressed response to mitogens, decreased humoral responses (CD4+ T helper cells deficit?), and impaired T cell-mediated cytotoxicity. These functional alterations may be explained, at least in part, by a defective high-affinity IL-2 receptor (IL-2 R) (Roszman et al, 1991). It has been demonstrated that glioblastoma cells produce and release soluble factors that are responsible for a depressed immune response. T lymphocytes from normal individuals exhibit immunologic abnormalities when grown in presence of supernatant obtained from glioblastoma cell line cultures (Roszman et al, 1991). The most important soluble suppressing factor seems to be TGF-#2. This cytokine acts as a growth-inhibitory factor (Sporn et al, 1986; Sporn et al, 1987), and has a defined variety of immunoregulatory properties, including: inhibition of: (i ) T cell proliferation; (i i ) IL-2R induction (Kehrl et al, 1986); (i i i ) cytokine production (Espevik et al, 1987; Espevik et al, 1988); (i v ) natural killer cell activity (Rook et al, 1986); (v) cytotoxic T cell development (Jin et al, 1989; Ranges et al, 1987); (v i ) LAK cell generation (Espevik et al, 1987; Jin et al, 1989); and (v i i ) production of tumor-infiltrating lymphocytes (Kuppner et al, 1989). Most cells secrete TGF-# in a latent form (Sporn et al, 1986), but glioblastoma cells have also the capacity to convert it to an active form, through proteolytic cleavage. This was demonstrated by an experimental work in which T-cell suppression mediated by TGF-#2 was inhibited when proteolytic enzymes were blocked by protease inhibitors (Huber et al, 1992). Other soluble factors, namely prostaglandin E2 (PGE2), IL-1 receptor (IL-1 R) antagonist, and interleukin-10 (IL10) may be implicated in immunosuppression, although in vivo intervention has not been fully defined. A potential down regulation of some immune functions was shown for IL-10 cytokine, which was produced both by GBM cells and normal brain tissue (Nitta et al, 1994; Merlo et al, 1993). Furthermore, in different animal models, human and murine IL-10 was demonstrated to stimulate the acquisition of a specific and efficient antitumor immune response (Berman et al, 1996).

C. Costimulatory molecules A complete T-cell effector function needs not only antigen presentation, but also the delivering of activation signals to the T cell, which is mediated by the so-called costimulatory molecules. Unresponsiveness of T cells (anergy) may be due to absence of the second signal, essentially given by the B7-CD28 interaction (June et al, 1994). Two other members of the B7 family have been cloned, B7.1 and B7.2; their counter-receptors on T cells are CD28 and CTLA-4, respectively. CD28 mediates stimulatory effects, while CTLA-4 appears to be a negative

regulator of T cell responses. Glioblastoma cells and monocytes that infiltrate the tumor are not expressing B7 costimulatory molecules, while monocytes in the normal tissue that surrounds the tumor are B7-positive (Tada et al, 1996). This suggests the possible intervention of local mechanisms able to down-regulate B7 expression in glioblastomas, impeding efficient T-cell priming and favoring T-cell anergy. Moreover, B7-CD28 interactions in the CNS have been shown to be essential to generate a valid CTL response towards viral antigens (Kuendig et al, 1996). Hence, restoring B7 expression by gene transfer might become an interesting task to elicit a proper immune recognition of glioblastoma cells, and an appropriate immune response.

V. Restoring a proper immune response Many approaches have been tented to restore a proper immune response towards malignant gliomas. As previously stated, T lymphocytes play a major role in the antitumor response, and priming of T lymphocytes requires antigen recognition, with or without help from APC. Since no specific antigens have yet been identified for glioblastomas, a vaccination approach has been proposed by administration of genetically modified tumor cells. Moreover, tumor cells transfected to produce various cytokines have been used to enhance lymphocyte responsiveness in animal models. The most interesting results were obtained with cells of murine glioma transfected with an expression vector containing the murine interleukin 7 cDNA (Aoki et al, 1992). IL-7 transfected glioma cells were vigorously rejected by a CD8+ T-cell-mediated immune response, that was proportional to the level of IL-7 production. Moreover, the response was tumor-specific, since no effect was observed against other syngeneic tumor cells (melanoma and fibrosarcoma cells). IL-7 is a very interesting cytokine being able to increase IL-2R ! chain expression on CD4+ T lymphocytes and to inhibit TGF-# mRNA expression and production by murine macrophages (Dubinett et al, 1993). IL-12 can also be considered a promising agent to enhance the antitumor response, since it augments T-cell and natural killer-cell activities, induces IFN-" production, and promotes the differentiation of uncommitted T cells to Th1 cells (Hendrzak and Brunda, 1995). Vector-mediated delivery of IL-12 into established tumors suppresses tumor growth (Caruso et al, 1996) and can induce immune responses against challenge tumors (Bramson et al, 1996). Moreover, IL-12 has other nonimmune properties such as anti-angiogenic effects (Voest et al, 1995). IL-4, another cytokine with pleiotropic functions, increases T cell proliferation and cytotoxicity, and enhances eosinophil and B cell proliferation and differentiation. It

139

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme

VI. Combined gene therapy approach in humans

also exerts direct anti-proliferative effects in vitro against many tumor cell lines (Tepper, 1993). Antitumor effects induced by IL-4-transfected cells were reported in nude mice, suggesting T cell-independent mechanisms (Tepper, 1993; Yu et al, 1993). A strong recruitment of eosinophils and subsequent inhibition of the tumor growth was noted. Eosinophil depletion was not performed as a control; hence a direct anti-proliferative effect mediated by IL-4 cannot be excluded. In any respect, a possible non-T-cell-dependent mechanism could be an advantage in the glioblastoma setting, considering the various abnormalities of T cell function.

The strict localization of glioblastomas in the CNS, with only exceptional metastases, makes these tumors candidates for approaches of direct intra-tumoral gene delivery. Retrovirus-mediated gene therapy of GBM is particularly attractive, since these viral vectors transduce only mitotically active cells, sparing the normal neuronal tissue composed of non-replicating cells. Gene therapy of brain tumors by intra-tumoral injection of retroviral vector producing cells (RVPC) in human patients was initiated by Oldfield and colleagues in 1993 (Oldfield et al, 1993). The gene being transferred was that expressing the herpes simplex thymidine kinase (HSV-TK), which conferred sensitivity to the anti-herpes drug ganciclovir. This treatment has proved free of toxicity and safe for there was no evidence of systemic spread of the retroviral vector (Long et al, 1998). However, a clinical benefit was limited to very small tumors (1.5 ml), probably because only malignant cells adjacent to the RVPC were transfected (Ram et al, 1997).

In an attempt to overcome the local immunosuppression mediated by TGF-#2 Fakhrai et al conducted an experimental work on rats by antisense gene therapy. 9L gliosarcoma transfected cells inoculated subcutaneously became highly immunogenic and were able to induce eradication of an established wild-type tumor (Fakhrai et al, 1996). The enhancement of antigen presentation and T cell costimulation has also been considered and may be achieved with genes coding for cytokines, like GM-CSF, costimulatory molecules, like B7, or CIITA, a transcription factor playing a critical role in the regulation of MHC class II molecules. Encouraging results have been reported in a murine melanoma model located in the CNS, whereby an efficient antitumor response was induced by subcutaneous vaccination with irradiated, GM-CSF-producing tumor cells (Sampson et al, 1996). Vaccination with cells co-transfected with B7 and IL-2 was able to mediate rejection of established tumors (Gaken et al, 1997), suggesting a possible application of such an intervention for the treatment of glioblastomas.

A new treatment strategy combining two different modalities, enzyme-directed prodrug activation (tumor suicide) along with cytokine-promoted tumor rejection, has been recently devised to amplify the antitumor response, and proved to be efficacious in animal models (Castleden et al, 1997) (Figure 2). A bicistronic retroviral vector coexpressing HSV-TK and human interleukin-2 genes has been designed to pursue this new approach of cancer gene therapy in humans (Pizzato et al, 1998).

F i g u r e 2 . Structure of a bicistronic retroviral vector for transduction of genes coding for a cytokine and a prodrugactivating enzyme, expressed via a cap-dependent and an internal ribosome entry site (IRES)-dependent mechanism, respectively. A selectable marker (neomycin phosphotransferase gene, neo) is expressed under the control of a SV40 promoter.

140

Gene Therapy and Molecular Biology Vol 1, page 141

F i g u r e 3 . Gene therapy approach for treatment of glioblastoma multiforme, via intratumoral stereotactic injection of cells producing a triple gene retroviral vector.

F i g u r e 4 . Contrast-enhanced MRI sagittal images of left parietal GBM lesion before (l e f t ), and one month after completion of GCV treatment (t o t h e r i g h t ).

141

Gene Therapy and Molecular Biology Vol 1, page 142

F i g u r e 5 . Histology of stereotactic biopsies from patients treated by HSV/TK-IL-2 combined gene therapy. A) Toluidine blue staining - Evidence of a large number of infiltrating inflammatory cells; Immunostaining with B ) monoclonal antibodies marking CD3+ cells; C) Mac 387 antibodies recognizing young/activated macrophages; D) monoclonal antibodies marking CD1+ infiltrating cells.

trials with thymidine kinase (Ram et al, 1997; Ostertag and Chiocca, personal communications).

After in vitro characterization of efficacy and safety (Pizzato et al, 1998), the vector was employed in a pilot study to treat four patients with recurrent glioblastoma multiforme (Colombo et al, 1997; Pal첫 et al, 1998) (Figure 3). A significant and sustained reduction (>50% of the initial volume) of the tumor mass (80 ml) was demonstrated by magnetic resonance imaging (MRI) and computerized tomography (CT) in one patient (Figure 4). In this case, the objective response was associated with a dramatic clinical improvement. The other three patients showed areas of tumor necrosis (2 ml) around the site of stereotactic RVPC injection and stabilized disease for a long period of time (11-12 months).

Interestingly, endothelial cells stained positive for TK by in situ hybridization, indicating that the vector had targeted the neo-vascular component, a highly replicative population in glioblastomas. This is consistent with an anti-angiogenic effect of this therapeutic approach, that, in addition to direct tumor suicide and immune activation, may be relevant to the bystander phenomenon and to the clinical response. It is noteworthy that IL-2 was measurable in the cerebral-spinal fluid, even after GCV treatment. This cytokine might have derived from an autocrine-paracrine secretion of recruited infiltrating immune-inflammatory cells, after primary expression in transduced cells.

In stereotactic biopsies taken before ganciclovir administration, large tumor infiltrates of immuneinflammatory cells (T lymphocytes, mostly CD4+ but also CD8+ granzyme B-positive cells, activated macrophages, NK cells, neutrophils) were present, notwithstanding the standard steroid therapy (Figure 5). The observed inflammatory response has never been reported in previous

Efforts to achieve more efficient gene transfer systems are being sought for. These include the development of new generation retroviral vectors, produced at higher titres and characterized by higher transduction efficiency. Strategies involving envelope pseudotyping, use of new

142

Gene Therapy and Molecular Biology Vol 1, page 143 Bramson JL, Hitt M, Addison CL, Muller WJ, Gauldie J, Graham FL. (1 9 9 6 ) Direct intratumoral injection of an adenovirus expressing interleukin-12 induces regression and long-lasting immunity that is associated with highly localized expression of interleukin-12. Hum Gene Ther 7, 1995-2002

packaging cell lines of human origin, and substitution of promoter elements will contribute to the improvement of current available vectors. New therapeutic gene combinations should also be accomplished in order to promote a more generalized immune response. Genes for cytokines other than IL-2 (i.e., IL-4, IL-7, IL-12, GM-CSF) as well as genes targeting neoangiogenesis deserve further consideration for combined treatment approaches.

Caruso M, Pham-Nguyen K, Kwong Y-L, Xu B, Kosai K-I, Finegold M, Woo SLC, Chen SH. (1 9 9 6 ) Adenovirusmediated interleukin-12 gene therapy for metastatic colon cancer. Proc Natl Acad Sci USA 93, 11302-11306

The authors wish to acknowledge Fondazione Cassa di Risparmio di Padova e Rovigo and Regione Veneto for financial support.

Castleden SA, Chong H, Garcia-Ribas I, Melcher AA, Hutchinson G, Roberts B, Hart IR, Vile RG. (1 9 9 7 ) A family of bicistronic vectors to enhance both local and systemic antitumor effects of HSVTK or cytokine expression in a murine melanoma model. Hum Gene Ther 8, 2087-102.

References

Chamberlain MC, Kormanik P. (1 9 9 6 ) Salvage chemotherapy with paclitaxel for recurrent primary brain tumors. J C l i n O n c o l 13, 2316-21.

Advani SJ, Sibley GS, Song PY, Hallahan DE, Kataoka Y, Roizman B, Weichselbaum RR. (1 9 9 8 ) Enhancement of replication of genetically engineered herpes simplex viruses by ionizing radiation: a new paradigm for destruction of therapeutically intractable tumors. Gene Ther 5, 160-165.

Chambers R, Gillespie GY, Soroceanu L, Andreansky S, Chatterjee S, Chou J, Roizman B, Whitley RJ. (1 9 9 5 ) Comparison of genetically engineered herpes simplex viruses for the treatment of brain tumors in a scid mouse model of human malignant glioma. P r o c N a t l A c a d S c i USA 92, 1411-1415.

Andreansky SS, He B, Gillespie GY, Soroceanu L, Markert J, Chou J, Roizman B, Whitley RJ. (1 9 9 6 ) The application of genetically engineered herpes simplex viruses to the treatment of experimental brain tumors. P r o c N a t l A c a d Sci USA 93, 11313-11318.

Chen CY, Chang YN, Ryan P, Linscott M, McGarrity GJ, Chiang YL. (1 9 9 5 ) Effect of herpes simplex virus thymidine kinase expression levels on ganciclovirmediated cytotoxicity and the "bystander effect". Hum Gene Ther 6, 1467-1476.

Aoki T, Tashiro K, Miyatake S, Kinashi T, Nakano T, Oda Y, Kikuchi H, Honjo T. (1 9 9 2 ) Expression of murine interleukin 7 in a murine glioma cell line results in reduced tumorigenicity in vivo P r o c N a t l A c a d S c i U S A 89, 3850-3854.

Cheney IW, Johnson DE, Vaillancourt MT, Avanzini J, Morimoto A, Demers GW, Wills KN, Shabram PW, Bolen JB, Tavtigian SV, Bookstein R (1 9 9 8 ) Suppression of tumorigenicity of glioblastoma cells by adenovirusmediated MMAC1/PTEN gene transfer. C a n c e r R e s 58, 2331-2334.

Acknowledgements

Badie B, Drazan KE, Kramar MH, Shaked A, Black KL. (1 9 9 5 ) Adenovirus-mediated p53 gene delivery inhibits 9L glioma growth in rats. Neurol Res 17, 209-216.

Chiariello E, Roz L, Albarosa R, Magnani I, Finocchiaro G. (1 9 9 8 ) PTEN/MMAC1 mutations in primary glioblastomas and short-term cultures of malignant gliomas. O n c o g e n e 16, 541-545.

Barba D, Hardin J, Ray J, Gage FH. (1 9 9 3 ) Thymidine kinase-mediated killing of rat brain tumors. J Neurosurg 70, 175-82,.

Chintala SK, Fueyo J, Gomez Manzano C, Venkaiah B, Bjerkvig R, Yung WK, Sawaya R, Kyritsis AP, Rao JS. (1 9 9 7 ) Adenovirus-mediated p16/CDKN2 gene transfer suppresses glioma invasion in vitro. O n c o g e n e 15, 17, 2049-57.

Barba D, Hardin J, Sadelain M, Gage FH. (1 9 9 4 ) Development of antitumor immunity following thymidine kinase-mediated killing of experimental brain tumors. Proc Natl Acad Sci USA 91, 4348-4352. Berman RM, Suzuki T, Tahara H, Robbins PD, Narula SK, Lotze MT. (1 9 9 6 ) Systemic administration of cellular IL10 induces an effective, specific, and long-lived immune response against established tumors in mice. J Immunol 157, 231-238

Chiocca EA. (1 9 9 5 ) Brain tumor gene therapy in mice with a novel "suicide" gene, the cyclophosphamide-activating CYP2B1 gene. Clin Neurosurg 42, 370-82.

Boviatsis EJ, Scharf JM, Chase M, Harrington K, Kowall NY, Breakefield XO, Chiocca EA. (1 9 9 4 ) Antitumor activity and reporter gene transfer into rat brain neoplasms inoculated with herpes simplex virus vectors defective in thymidine kinase or ribonucleotide reductase. Gene Ther 1, 323-31.

Colombo F, Zanusso M, Casentini L, Cavaggioni A, Franchin E, Calvi P, Pal첫 G. (1 9 9 7 ) Gene stereotactic neurosurgery for recurrent malignant gliomas. Stereotact Funct Neurosurg 68, 245-251.

Collins VP. (1 9 9 3 ) Amplified genes in human gliomas. S e m i n C a n c e r B i o l 4, 27-32.

143

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme Coyle T, Baptista J, Winfield J. (1 9 9 0 ) Mechlorethamine, vincristine and procarbazine chemotherapy for recurrent high-grade glioma in adults, A phase II study. J C l i n O n c o l 8, 2014-2018.

Forsyth PA, Cairncross JG. (1 9 9 6 ) Chemotherapy of malignant gliomas. In Cerebral Gliomas. B a i l l e r e ' s Clin Neurol 5, 371-93. Fueyo J, Gomez Manzano C, Yung WK, Clayman GL, Liu TJ, Bruner J, Levin VA, Kyritsis AP. (1 9 9 6 ) Adenovirusmediated p16/CDKN2 gene transfer induces growth arrest and modifies the transformed phenotype of glioma cells. O n c o g e n e 12, 103-110.

Culver KW, Ram Z, Wallbridge S, Ishii I, Oldfield EH, Blaese RM. (1 9 9 2 ) In vivo gene transfer with retroviral vectorproducer cells for treatment of experimental brain tumors. S c i e n c e 256, 1550-1552. De Smet C, Courtois SJ, Faraoni I, Lurquin C, Szikora JP, De Backer O, Boon T. (1 9 9 5 ) Involvement of two Ets binding sites in the transcriptional activation of the MAGE-1 gene. I m m u n o n o g e n e t i c s 42, 282-290.

Fueyo J, Gomez Manzano C, Yung WK, Liu TJ, Alemany R, McDonnell TJ, Shi X, Rao JS, Levin VA, Kyritsis AP. (1 9 9 8 ) Overexpression of E2F-1 in glioma triggers apoptosis and suppresses tumor growth in vitro and in vivo. Nat Med 4, 685-690.

De Smet C, Lurquin C, Van Der Bruggen P, De Plaen E, Brasseur F, Boon T. (1 9 9 4 ) Sequence and expression pattern of the human MAGE 2 gene. I m m u n o g e n e t i c s 39, 121-129.

Fujiwara T, Grimm EA, Mukhopadhayay T, Zang WW, OwenSchaub LB, Roth JA. (1 9 9 4 ) Induction of chemosensitivity in human lung cancer in vivo by adenoviral mediated tranfer of the wild type p53 gene. Cancer Res 54, 2287-2291.

Dietrich P-Y, Walker PR, Saas P, de Tribolet N. (1 9 9 7 ) Immunobiology of Gliomas, New perspectives for therapy. Ann NY Acad Sci 824, 124-140.

Furnari FB, Su Huang HJ, Cavenee WK. (1 9 9 5 ) Genetics and malignant progression of human brain tumors. Cancer Surv 25, 233-275.

Dubinett SM, Huang M, Dhanani S, Wang J, Beroiza T. (1 9 9 3 ) Down-regulation of macrophage transforming growth factor-beta messenger RNA expression by IL-7. J Immunol 151, 6670-6680.

Gaken JA, Hollingsworth SJ, Hirst WJR, Buggins AGS, Galea J, Peakman M, Kuiper M, Patel P, Towner P, Patel PM, Collins MK, Mufti GJ, Farzaneh F, Darling DC. (1 9 9 7 ) Irradiated NC adenocarcinoma cells transduced with both B7.1 and interleukin-2 induce CD4+-mediated rejection of established tumors. Hum Gene Ther 8, 477-488.

Ekstrand AJ, Longo N, Hamid ML, Olson JJ, Liu L, Collins VP, James CD. (1 9 9 4 ) Functional characterization of an EGF receptor with a truncated extracellular domain expressed in glioblastomas with EGFR gene amplifacation. O n c o g e n e 9, 2313-20.

Ge K, Xu L, Zheng Z, Xu D, Sun L, Liu X. (1 9 9 7 ) Transduction of cytosine deaminase gene makes rat glioma cells highly sensitive to 5-fluorocytosine. Int J Cancer 71, 675-9.

Espevik T, Figari IS, Ranges GE, Palladino MA Jr. (1 9 8 8 ) Transforming growth factor-#1 (TGF-#1) and recombinant human tumor necrosis factor ! reciprocally regulate the generation of lymphokine-activated killer cell activity. Comparison between natural porcine platelet-derived TGF#1 and TGF-#2, and recombinant human TGF-#1. J Immunol 140, 2312-2316.

Gjerset RA, Turla ST, Sobol RE, Scalise JJ, Mercola D, Collins H, Hopkins PJ. (1 9 9 5 ) Use of wild type p53 to achieve complete treatment sensitization of tumor cells expressing endogenous mutant p53. M o l C a r c i n o g 14, 275-285. Gomez-Manzano C, Fueyo J, Kyritsis AP, Steck PA, Roth JA, McDouwell TJ, Steck KD, Levin VA, Yung WK. (1 9 9 6 ) Adenovirus-mediated tranfer of the p53 gene produces rapid and generalized death of human glioma cells via apoptosis. Cancer Res 56, 694-699.

Espevik T, Figari IS, Shalaby MR, Lackides GA, Lewis GD, Shepard HM, Palladino MA Jr. (1 9 8 7 ) Inhibition of cytokine production by cyclosporin A and transforming growth factor . J Exp Med 166, 571-576. Ezzeddine ZD, Martuza RL, Platika D, Short MP, Malick A, Choi B, Breakefield XO. (1 9 9 1 ) Selective killing of glioma cells in colture and in vivo by retrovirus transfer of the herpes simplex virus thymidine kinase gene. New B i o l 3, 608-614.

Griscelli F, Li H, Bennaceur Griscelli A, Soria J, Opolon P, Soria C, Perricaudet M, Yeh P, Lu H. (1 9 9 8 ), Angiostatin gene transfer, inhibition of tumor growth in vivo by blockage of endothelial cell proliferation associated with a mitosis arrest. Proc Natl Acad Sci USA 95, 11, 63676372.

Fakhrai H, Dorigo O, Shawler DL, Lin H, Mercola D, Black KL, Royston I, Sobol RE. (1 9 9 6 ) Eradication of established intracranial rat gliomas by transforming growth factor beta antisense gene therapy. Proc Natl Acad Sci USA 93, 2909-2914.

Hamel W, Magnelli L, Chiarugi VP, Israel MA. (1 9 9 6 ) Herpes simplex virus thymidine kinase/ganciclovir-mediated apoptotic death of bystander cells. Cancer R e s 56, 2697-2702.

Fewer D, Wilson CB, Boldrey EB. (1 9 7 2 ) Phase II study of CCNU in the treatment of brain tumors. Cancer Chem Rep, 56, 421-7.

He J, Allen JR, Collins VP, Allalunis-Turner MJ, Godbout R, Day RS 3rd, James CD. (1 9 9 4 ) CDK4 amplification is an alternative mechanism to p16 gene homozygous deletion in glioma cell lines. Cancer Res 54, 5804-5807.

Fine HA, Dear KBG, Loeffler JS. (1 9 9 3 ) Meta-analysis of radiation therapy with an without adjuvant chemotherapy for malignant glionas in adults. Cancer 71, 2585-97.

144

Gene Therapy and Molecular Biology Vol 1, page 145 He J, Olson JJ, James CD. (1 9 9 5 ) Lack of p16ink4 or retinoblastoma protein (pRb) or amplification-associated overexpression of cdk4 is observed in dinstinct subsets of malignant glial tumors and cell lines. C a n c e r R e s 55, 4833-4836.

Kerhl JH, Wakefield LM, Roberts AB, Jakowlew S, AlvarezMon M, Derynck R, Sporn MB, Fauci AS. (1 9 8 6 ) Production of transforming growth factor ( by human T lymphocytes and its potential role in the regulation of T cell growth. J Exp Med 163, 1037-1050.

Hendrzak JA and Brunda MJ. (1 9 9 6 ) Interleukin-12, biologic activity, therapeutic utility, and role in disease. Lab I n v e s t 72, 619-637.

Kim JH, Kim SH, Brown SL, Freitag SO. (1 9 9 4 ) Selective enhancement by an antiviral agent of the radiation induced cell killing of human glioma cells transduced with HSV-TK gene. Cancer Res 54, 6053-6056.

Henson JW, SchniTKer BL, Correa KM, von Dimling A, Fassbender F, Xu HJ, Benedict WF, Yandell DW, Louis DN. (1 9 9 4 ) The retinoblastoma gene is involved in malignat progression of astrocytomas. Ann Neurol 36, 714-721.

Kim JH, Kim SH, Kolozsvary A,Brown SL, Kim OB, Freytag SO. ( 1 9 9 5 ) Selective enhancement of radiation response of herpes simplex virus thymidine kinase transduced 9L gliosarcoma cells in vitro and in vivo by antiviral agents. I n t J R a d i a t O n c o l B i o l P h y s 33, 861-868.

Hochberg FH, Pruitt A. (1 9 8 0 ) Assumptions in the radiotherapy of glioblastoma. N e u r o l o g y 30, 907-911.

Kim SH, Kim JH, Kolozsvary A, Brown SL, Freytag SO. (1 9 9 7 ) Preferential radiosensitization of 9L glioma cells transduced with HSV-TK gene by acyclovir. J N e u r o o n c o l 33, 189-194.

Hoostraten B, Gottlieb JA, Caoili A. 1 9 7 3 CCNU in the treatment of cancer, A phase II study. Cancer 32, 38-43. Hosli P, Sappino AP, de Tribolet N, Dietrich PY. (1 9 9 8 ) Malignant glioma, Should chemotherapy be overthrown by experimental treatments? A n n O n c o l 9, 589-600.

Kondo S, Barna BP, Kondo Y, Tanaka Y, Casey G, Liu J, Morimura T, Kaakaji R, Peterson JW, Werbel B, Barnett GH. (1 9 9 6 ) WAF1/CIP1 increases the susceptibility of p53 non-functional malignant glioma cells to cisplatininduced apoptosis. O n c o g e n e 13, 1279-85.

Houbiers JG, Nijman HW, Van Der Burg SH, Drijfhout JW, Kenemans P, Van Der Velde CJ, Brand A, Momburg F, Kast WM, Melief CJ. (1 9 9 3 ) In vitro induction of human cytotoxic T lymphocyte responses against peptides of mutant and wild-type p53. Eur J Immunol 23, 20722077.

Kondo S, Kondo Y, Li G, Silverman RH, Cowell JK. (1 9 9 8 ) Targeted therapy of human malignant glioma in a mouse model by 2-5A antisense directed against telomerase RNA. O n c o g e n e , 16, 25, 3323-30.

Huber D, Philipp J, Fontana A. (1 9 9 2 ) Protease inhibitors interfere with the transforming growth factor-(-indipendent pathway of tumor cell-mediated immunosuppression. J Immunol 148, 277-284.

Kundig TM, Shahinian K, Kawai K, Mittruecker HW, Sebzda E, Bachmann MF, Mak TW, Ohashi PS. (1 9 9 6 ) Duration of TCR stimulation determines costimulatory requirement of T cells. Immunity, 5, 41-52

Ichimura K, Schmidt EE, Miyakawa A, Goike HM, Collins VP. (1 9 9 8 ) Distinct patterns of deletion on 10p and 10q suggest involvement of multiple tumor suppressor genes in the development of astrocytic gliomas of different malignancy grades. G e n e s C h r o m o s o m e s C a n c e r 22, 9-15.

Kuppner MC, Hamou M-F, Sawamura Y, Bodmer S, de Tribolet N. (1 9 8 9 ) Inhibition of lymphocyte function by glioblastoma-derived transforming growth factor #2. J Neurosurg 71, 211-217

Janzer RC, Raff MC. (1 9 8 7 ) Astrocytes induce blood brain barrier properties in e每dotelial cells. Nature 325, 253257.

Lachmann RH, Efstathiou S. (1 9 9 7 ) The use of herpes simplex virus-based vectors for gene delivery to the nervous system. Mol Med Today 3, 404-411.

Jen J, Harper JW, Bigner SH, Bigner DD, Papadopoulos N, Markowitz S, Willson JK, Kinzler KW, Vogelstein B. (1 9 9 4 ) Deletion of p16 and p15 genes in brain tumors. Cancer Res 54, 6353-6358.

Lang FF, Yung WK, Raju U, Libunao F, Terry NH, Tofilon PJ. (1 9 9 8 ) Enhancement of radiosensitivity of wild-type p53 human glioma cells by adenovirus-mediated delivery of the p53 gene. J Neurosurg 89, 1, 125-132.

Jin B, Scott JL, Vadas MA, Burns GF. (1 9 8 9 ) TGF-# down regulates TLiSA1 expression and ihibits the differentiation of precursor lymphocytes into CTL and LAK cells. I m m u n o l o g y 66, 570-576

LeMay DR, Kittaka M, Gordon EM, Gray B, Stins MF, McComb JG, Jovanovic S, Tabrizi P, Weiss MH, Bartus R, Anderson WF, Zlokovic BV. (1 9 9 8 ) Intravenous RMP-7 increases delivery of ganciclovir into rat brain tumors and enhances the effects of herpes simplex virus thymidine kinase gene therapy. Hum Gene Ther 9, 989-995.

June CH, Bluestone JA, Nadler LM, Thompson CB. (1 9 9 4 ) The B7 and CD28 receptor families. Immunol Today 15, 321-331

Levin VA, Edwards MS, Wright DC. (1 9 8 0 ) Modified procarbazine, CCNU and vincristine (PCV3) combination chomtherapy in the treatment of malignant brain tumors. Cancer Treat Rep 64, 237-241.

Ke LD, Fueyo J, Chen X, Steck PA, Shi YX, Im SA, Yung WK. (1 9 9 8 ) A novel approach to glioma gene therapy, downregulation of the vascular endothelial growth factor in glioma cells using ribozymes. Int J Oncol 12, 1391-6.

Levin VA, Gutin PH, Leibel S. Cancer, Principles and Practice of Oncology. In De Vita VT Jr, Hellman S, Rosemberg SA

145

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme (eds) (1 9 9 3 ) , N e o p l a s m s o f t h e C e n t r a l N e r v o u s S y s t e m , Philadelphia, JB Lippincot, 1679-737.

metastases and meningiomas suggests specific transcription patterns. Eur J Cancer 29A, 2118-2125.

Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J, Miliaresis C, Rodgers L, McCombie R, Bigner SH, Giovanella BC, Ittmann M, Tycko B, Hibshoosh H, Wigler MH, Parsons R. (1 9 9 7 ) PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. S c i e n c e , 275, 1943-1947.

Millauer B, Longhi MP, Plate KH, Shawer LK, Risau W, ulrich A, Strawn LM. (1 9 9 6 ) Dominant-negative inhibition of Flk-1 suppress the growth of many tumor types in vivo. Cancer Res 56, 1615-1620. Millauer B, Shawer LK, Plate KH, Risau W, Ulrich A. (1 9 9 4 ) Glioblastoma growth inhibited in vivo by a dominantnegative Flk-1 mutant. Nature, 367, 576-579.

Libermann TA, Nusbaum HR, Razon N, Kris R, Lax I, Soreq H, Whittle N, Waterfield MD, Ullrich A, Schlessinger J. (1 9 8 5 ) Amplification, enhanced expression and possible rearrangement of EGF receptor gene in primary human brain tumors of glial origin. Nature 313, 144-147.

Mineta T, Rabkin SD, Yazaki T, Hunter WD, Martuza RL. (1 9 9 5 ) Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. N a t M e d , 1 , 9 , 938-943.

Long Z, Li LP, Grooms T, Lockey C, Nader K, Mychkovsky I, Mueller S, Burimski I, Ryan P, Kikuchi G, Ennist D, Marcus S, Otto E, McGarrity G. (1 9 9 8 ) Biosafety monitoring of patients receiving intracerebral injections of murine retroviral vector producer cells. Hum Gene Ther 9, 1165-1172.

Miyatake S, Martuza RL, Rabkin SD. (1 9 9 7 ) Defective herpes simplex virus vectors expressing thymidine kinase for the treatment of malignant glioma. Cancer Gene Ther 4, 222-228. Mizuno M, Yoshida J, Colosi P, Kurtzman G. (1 9 9 8 ) Adenoassociated virus vector containing the herpes simplex virus thymidine kinase gene causes complete regression of intracerebrally implanted human gliomas in mice, in conjunction with ganciclovir administration. Jpn J Cancer Res, 89, 1, 76-80.

Lowe SW, Bodis S, McClatchey A, Remington L, Ruley HE, Fisher DE, Housman DE, Jacks T. (1 9 9 4 ) p53 status and the efficacy of cancer therapy in vivo. S c i e n c e 266, 807810. Mabuchi E, Shimizu K, Miyao Y, Kaneda Y, Kishima H, Tamura M, Ikenaka K, Hayakawa T. (1 9 9 7 ) Gene delivery by HVJ-liposome in the experimental gene therapy of murine glioma. Gene Ther 4, 768-72.

Newlands ES, O'Reilly SM, Glaser MG. (1 9 9 6 ) The Charing Cross hospital experience with Temozolomide in patients with gliomas. Eur J Cancer, 32A, 2236-41. Nguyen DM, Spitz FR, Yen N, Cristiano RJ, Roth JA. (1 9 9 6 ) Gene therapy for lung cancer, enhancement of tumor suppression by a combination of sequential systemic cisplatin and adenovirus-mediated p53 gene transfer. J Torac Cardiovasc Surg, 112, 1372-1377.

Mak M, Fung L, Strasser JF, Saltzman WM. (1 9 9 5 ) Distribution of drugs following controlled delivery to the brain interstitium. J N eu r o o n c o l 26, 91-102. Manome Y, Kunieda T, Wen PY, Koga T, Kufe DW, Ohno T. (1 9 9 8 ) Transgene expression in malignant glioma using a replication-defective adenoviral vector containing the Egr-1 promoter, activation by ionizing radiation or uptake of radioactive iododeoxyuridine. Hum Gene Ther 9, 1409-1417.

Nielsen LL, Maneval DC. (1 9 9 8 ) p53 tumor suppressor gene therapy for cancer. Cancer Gene Ther 5, 52-63. Nishikawa R, Furnari FB, Lin H, Arap W, Berger MS, Cavenee WK, Su Huang HJ. (1 9 9 5 ) Loss of p16ink4 expression is frequent in high-grade gliomas. C a n c e r R e s 55, 19411945.

Manome Y, Wen PY, Chen L, Tanaka T, Dong Y, Yamazoe M, Hirshowitz A, Kufe DW, Fine HA. (1 9 9 6 ) Gene therapy for malignant gliomas using replication incompetent retroviral and adenoviral vectors encoding the cytochrome P450 2B1 gene together with cyclophosphamide. Gene Ther 3, 513-520.

Nishikawa R, Ji XD, Harmon RC, Armon RC, Lazar CS, Gill GN, Cavenee WK, Huang HG. (1 9 9 4 ) A mutant epidermal growth factor receptor common in human gliomas confers enhanced tumorigenicity. P r o c Natl Acad S c i USA 91, 7727-7731.

Manome Y, Wen PY, Dong Y, Tanaka T, Mitchell BS, Kufe DW, Fine HA. (1 9 9 6 ) Viral vector transduction of the human deoxycytidine kinase cDNA sensitizes glioma cells to the cytotoxic effects of cytosine arabinoside in vitro and in vivo. Nat Med 2, 567-573.

Nitta T, Hishii M, Sato K, Okumura K. (1 9 9 4 ) Selective expression of interleukin-10 gene within glioblastoma multiforme. Brain Res, 649, 122-128. Noguchi Y, Chen YT, Old LJ. (1 9 9 4 ) A mouse mutant p53 product recognized by CD4+ and CD8+ T cells. P r o c N a t l Acad Sci 91, 3171-3175.

McKie EA, MacLean AR, Lewis AD, Cruickshank G, Rampling R, Barnett SC, Kennedy PG, Brown SM. (1 9 9 6 ) Selective in vitro replication of herpes simplex virus type 1 (HSV-1) ICP34.5 null mutants in primary human CNS tumours-evaluation of a potentially effective clinical therapy. B r J Cancer 74, 745-752.

Ohnishi T, Taki T, Hiraga S, Arita N, Morita T. (1 9 9 8 ) In vitro and in vivo potentiation of radiosensitivity of malignant gliomas by antisense inhibition of the RAD51 gene. B i o c h e m B i o p h y s R e s Commun 245, 319324.

Merlo A, Juretic A, Zuber M, Filgueira L, Luscher U, Caetano V, Ulrich J, Gratzl O, Heberer M, Spagnoli G. (1 9 9 3 ) Cytokine gene expression in primary brain tumors,

Oldfield EH, Ram Z, Culver KW, Blaese RM, De Vroom HL, Anderson WF. (1 9 9 3 ) Clinical Protocol, Gene therapy for

146

Gene Therapy and Molecular Biology Vol 1, page 147 the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase and intravenous ganciclovir. Hum Gene Ther 4, 39-69.

brain tumors by intratumoral implantation of retroviral vector-producing cells. Nature Med 3, 1354-1361 Ram Z, Culver KW, Wallbridge S, Blaese RM, Oldfield EH. (1 9 9 3 ) In situ retroviral-mediated gene transfer for the treatment of brain tumors in rats. Cancer Res 53, 83-88.

Ono Y, Ikeda K, Wei MX, Harsh GR 4th, Tamiya T, Chiocca EA. (1 9 9 7 ) Regression of experimental brain tumors with 6-thioxanthine and Escherichia coli gpt gene therapy. Hum Gene Ther 8, 2043-2055.

Ranges GE, Figari IS, Espevik T, Palladino MA Jr. (1 9 8 7 ) Inhibition of cytotoxic T cell development by transforming growth factor # and reversal by recombinant tumor necrosis factor !. J Exp Med 166, 991-998.

O'Rourke DM, Kao GD, Singh N, Park BW, Muschel RJ, Wu CJ, Greene MI. (1 9 9 8 ) Conversion of a radioresistant phenotype to a more sensitive one by disabling erbB receptor signaling in human cancer cells. Proc Natl Acad Sci USA 95, 10842-10847.

Redekop GJ, Naus CC. (1 9 9 5 ) Transfection with bFGF sense and antisense cDNA resulting in modification of malignant glioma growth. J Neurosurg 82, 83-90.

O'Rourke DM, Qian X, Zhang HT, Davis JG, Nute E, Meinkoth J, Greene MI. (1 9 9 7 ) Trans receptor inhibition of human glioblastoma cells by erbB family ectodomains. Proc Natl Acad Sci USA 94, 3250-3255.

Rimoldi D, Romero P, Carrel S. (1 9 9 3 ) The human melanoma antigen/encoding gene, MAGE-1, is expressed by other tumor cells of neuroectodermal origin such as glioblastomas and neuroblastomas. Int J Cancer 54, 527-528.

Pal첫 G, Cavaggioni A, Calvi P, Franchin E, Pizzato M, Boschetto R, Parolin C, Chilosi M, Ferrini S, Zanusso A, Colombo F. (1 9 9 8 ) Gene therapy of glioblastoma multiforme via combined expression of suicide and cytokine genes, a pilot study in humans. Gene Ther (in press)

Rodriguez LA, Prados M, Silver P. (1 9 8 9 ) Reevaluation of procarbazione for the treatmentof recurrent malignant central nervous system tumors. Cancer 64, 2420-2423. Rogulski KR, Kim JH, Kim SH, Freytag SO. (1 9 9 7 ) Glioma cells transduced with an Escherichia coli CD/HSV-1 TK fusion gene exhibit enhanced metabolic suicide and radiosensitivity. Hum Gene Ther 8, 73-85.

Parr MJ, Manome Y, Tanaka T, Wen P, Kufe D, Kaelin WG Jr, Fine HA. (1 9 9 7 ) Tumor-selective transgene expression in vivo mediated by an E2F-responsive adenoviral vector. Nature Med, 3, 1145-9.

Rook AH, Kerhl JH, Wakefield LM, Roberts AB, Sporn MB, Burlington DB, Lane HC, Fauci AS. (1 9 8 6 ) Effects of transforming growth factor ( on the functions of natural killer cells, depressed cytolytic activity and blunting of interferon responsiveness. J Immunol 136, 3916-3920.

Petersdorf SH, Berger MS. (1 9 9 6 ) Concepts in Neurosurgery, The molecular basis of neurosurgical disease. In Raffel C, Harsh IV GR (eds), Molecular Basis of Chemotherapy for Brain Tumors, Baltimore, Williams and Wilkins, chapter 12 (Vol 8), 198-210.

Rosemberg SA, Lotze MT, Muul LM, Chang AE, Avis FP, Leitman S, Linehan WM, Robertson CN, Lee RE, Rubin JT, Seipp CA, Simpson CG, White DE. (1 9 8 7 ) A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and interleukin-2 or high dose of interleukin-2 alone. New Engl J Med, 316, 889-905.

Pizzato M, Franchin E, Calvi P, Boschetto R, Ferrini S, Colombo M, Pal첫 G. (1 9 9 8 ) Production and characterization of a bicistronic Moloney-based retroviral vector expressing human interleukin 2 and herpes simplex virus thymidine kinase for gene therapy of cancer. Gene Ther 5, 1003-1007. Prados MD, Schold SC, Spence AM, Berger MS, Mc Allister LD, Mehta MP, Gilbert MR, Fulton D, Kuhn J, Lamborn K, Rector DJ, Chang SM. (1 9 9 6 ) Phase II study of paclitaxel in patients with recurrent malignant glioma. J C l i n O n c o l 14, 2316-2321.

Roszman T, Elliott L, Brooks W. (1 9 9 1 ) Modulation of Tcell function by gliomas. Immunol Today 12, 370-374.

Rainov NG, Dobberstein KU, Bahn H, Holzhausen HJ, Lautenschl채ger C, Heidecke V, Burkert W. (1 9 9 7 ) Prognostic factors in malignant glioma, influence of the overexpression of oncogene and tumor-suppressor gene products on survival. J N eu r o o n c o l 35, 13-28.

Russell DS, Rubistein LJ. (1 9 8 9 ) P a t h o l o g y o f T u m o r s o f t h e C e n t r a l N e r v o u s S y s t e m, 5th Edition, Edward Arnold, London.

Roth JA, Cristiano R. (1 9 9 7 ) Gene therapy for cancer, What we done and where we are going? J N a t l C a n c e r I n s t , 89, 21-39.

Saleh M, Stacker SA, Wilks AF. (1 9 9 6 ) Inhibition of growth of C6 glioma cells in vivo by expression of antisense vascular endothelial growth factor sequence. Cancer Res 56, 393-401.

Rainov NG, Dobberstein KU, Sena-Esteves M, Herrlinger U, Kramm CM, Philpot RM, Hilton J, Chiocca EA, Breakefield XO. (1 9 9 8 ) New prodrug activation gene therapy for cancer using cytochrome P450 4B1 and 2aminoanthracene/4-ipomeanol. Hum Gene Ther 9, 1261-73.

Sampson JH, Archer GE, Ashley DM, Fuchs HE, Hale LP, Dranoff G, Bigner DD. (1 9 9 6 ) Subcutaneous vaccination with irradiated, cytokine-producing tumor cells stimulates CD8+ cell mediated immunity against tumors located in the immunologically privileged central nervous system. Proc Natl Acad Sci USA 93, 10399-10404.

Ram Z, Culver KW, Oshiro EM, Viola JJ, De Vroom HL, Otto E, Long Z, Chiang Y, McGarrity GJ, Muul LM, Katz D, Blaese RM, Oldfield EH. (1 9 9 7 ) Therapy of malignant

147

Palù et al: Perspectives for therapy of glioblastoma multiforme Sanson M, Ameri A, Monjour A, Sahmoud T, Ronchin P, Poisson M, Delattre JY. (1 9 9 6 ) Treatment of recurrent malignant supratentorial gliomas with ifosfamide, carboplatin, and etoposide, A phase II study. Eur J Cancer 32A, 2229-2235.

Tepper RI. (1 9 9 3 ) The antitumor and proinflammatory actions of IL-4. Res Immunol 144, 633-637. Tohma Y, Gratas C, Biernat W, Peraud A, Fukuda M, Yonekawa Y, Kleihues P, Ohgaki H. (1 9 9 8 ) PTEN (MMAC1) mutations are frequent in primary glioblastomas (de novo) but not in secondary glioblastomas. J Neuropathol Exp Neurol 57, 684-689.

Smart CR, Ottomoan RE, Rochlin DB. (1 9 6 8 ) Clinical experience with vincristine with tumors of the central nervous system and other malignant diseases. Cancer Chem Rep 52, 733-741.

Trojan J, Johnson TR, Rudin SD, Ilan J, Tykocinski ML, Ilan J. (1 9 9 3 ) Treatment and prevention of rat glioblastoma by immunogenic C6 cells expressing antisense insulinlike growth factor I RNA. S c i e n c e 259, 94-97.

Sporn MB, Roberts AB, Wakefield LM, Assoian RK. (1 9 8 6 ) Transforming growth factor-(, biological fusion and chemical structure. S c i e n c e 233, 532-534

Van Der Bruggen P, Traversari C, Chomez P, Lurquin C, De Plaen E, Van Den Eynde B, Knuth A, Boon T. (1 9 9 1 ) A gene encoding an antigen recognized by cytolityc T lymphocytes on a human melanoma. S c i e n c e 254, 16431647

Sporn MB, Roberts AB, Wakefield LM, de Crombrugghe B. (1 9 8 7 ) Some recent advances in the chemistry and biology of transforming growth factor-beta. J C e l l B i o l 105, 1039-1045. Staba MJ, Mauceri HJ, Kufe DW, Hallahan DE, Weichselbaum RR. (1 9 9 8 ) Adenoviral TNF-alpha gene therapy and radiation damage tumor vasculature in a human malignant glioma xenograft. Gene Ther 5, 293-300.

Van Meir E, Roemer K, Diserens AC, Kikuchi T, Rempel SA, Haas M, Huang HG, Friedmann T, de Tribolet N, Cavenee WK. (1 9 9 5 ) Single cell monitoring of growth arrest and morphological changes induced by transfer of wild-type p53 alleles to glioblastoma cells. P r o c N a t l A c a d S c i USA 92, 1008-1012.

Sturtz FG, Waddell K, Shulok J, Chen X, Caruso M, Sanson M, Snodgrass HR, Platika D. (1 9 9 7 ) Variable efficiency of the thymidine kinase/ganciclovir system in human glioblastoma cell lines, implications for gene therapy. Hum Gene Ther 8, 1945-53.

Vincent AJ, Vogels R, Someren GV, Esandi MC, Noteboom JL, Avezaat CJ, Vecht C, Bekkum DW, Valerio D, Bout A, Hoogerbrugge PM. (1 9 9 6 ) Herpes simplex virus thymidine kinase gene therapy for rat malignant brain tumors. Hum Gene Ther 7, 197-205.

Tada M, Diserens AC, Hamou MF, Jaufeerally R, Van Meir E, de Tribolet N. (1 9 9 6 ) B r a i n T u m o r R e s T h e r , 327337.

Voest EE, Kenyon BM, O’Reilly MS, Truitt G, D’Amato RJ, Folkman J. (1 9 9 5 ) Inhibition ofangiogenesis in vivo by interleukin-12. J Natl Cancer Inst 87, 581-586.

Takamiya Y, Short MP, Ezzeddine ZD, Moolten FL, Breakefield XO, Martuza RL. (1 9 9 2 ) Gene therapy of malignant brain tumors, a rat glioma line bearing the herpes simplex virus type 1-thymidine kinase gene and wild type retrovirus kills other tumor cells. J N e u r o s c R e s 33, 493-503.

Walker MD, Green SB, Byar DP. (1 9 8 0 ) Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N E n g l J Med 303, 1323-1329.

Takamiya Y, Short MP, Moolten FL, Fleet C, Mineta T, Breakefield XO, Martuza RL. (1 9 9 3 ) An experimental model of retrovirus gene therapy for malignant brain tumors. J Neurosurg 79, 104-110.

Wei M, Tamiya T, Chase M, Bobiatsis EJ, Chang TK, Kowall NW, Hochberg FH, Waxman DJ, Breakefield XO, Chiocca EA. (1 9 9 4 ) Experimental tumor therapy in mice using the cyclophosphamide-activating cytochrome P450 2B1 gene. Hum Gene Ther 5, 969-978.

Tamiya T, Ono Y, Wei MX, Mroz PJ, Moolten FL, Chiocca EA. (1 9 9 6 ) Escherichia coli gpt gene sensitizes rat glioma cells to killing by 6-thioxanthine or 6thioguanine. Cancer Gene Ther 3, 3, 155-162.

Weller M, Malipiero U, Rensing Ehl A, Barr PJ, Fontana A. (1 9 9 5 ) Fas/APO-1 gene transfer for human malignant glioma. Cancer Res 55, 2936-2944.

Tamura M, Shimizu K, Yamada M, Miyao Y, Hayakawa T, Ikenaka K. (1 9 9 7 ) Targeted killing of migrating glioma cells by injection of HTK-modified glioma cells. Hum Gene Ther 8, 381-91.

Young RC, Walker MD, Canellos GP. 1 9 7 3 Initial clinical trials with methyl-CCNU 1-(2-chloroethyl)-3-(4-methyl cycloexyl)-I-nitrosourea (MeCCNU). Cancer 31, 11641169.

Tanaka T, Cao Y, Folkman J, Fine HA. (1 9 9 8 ) Viral vectortargeted antiangiogenic gene therapy utilizing an angiostatin complementary DNA. Cancer Res 58, 33623369.

Yu JS, Wei MX, Chiocca EA, Martuza RL, Tepper RI. (1 9 9 3 ) Treatment of glioma by engineered interleukin 4-secreting cells. Cancer Res 53, 3125-3128. Zerrouqi A, Rixe O, Ghoumari AM, Yarovoi SV, Mouawad R, Khayat D, Soubrane C. (1 9 9 6 ) Liposomal delivery of the herpes simplex virus thymidine kinase gene in glioma, improvement of cell sensitization to ganciclovir. Cancer Gene Ther 3, 385-392.

Tanaka T, Manome Y, Wen P, Kufe DW, Fine HA. (1 9 9 7 ) Viral vector-mediated transduction of a modified platelet factor 4 cDNA inhibits angiogenesis and tumor growth. Nat Med 3, 437-442.

148

Gene Therapy and Molecular Biology Vol 1, page 149

149

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme

tumor cell integrated viral vector

transduction LTR

IL-2

IRES

TK

SV

NEO

LTR

nucleus mRNA GCV IL-2

GCV-P

GCV-3P

TK

producer cells inoculation

TIL recruitment bystander effect

150

Gene Therapy and Molecular Biology Vol 1, page 151

LTR

$

cytokine IRES

TK

SV

regulatory elements therapeutic genes gene for positive selection

151

NEO

LTR

Gene Therapy and Molecular Biology Vol 1, page 133 Gene Ther Mol Biol Vol 3, 133-148. August 1999.

Glioblastoma multiforme: molecular biology and new perspectives for therapy Review Article

Giorgio Pal첫, Luisa Barzon, and Roberta Bonaguro Institute of Microbiology, University of Padova Medical School, Padova, Italy __________________________________________________________________________________________________ Corresponding Author: Giorgio Pal첫, MD, Institute of Microbiology, Via A. Gabelli 63, 35121 Padova, Italy. Tel: +39-049-8272350; Fax: +39-049-8272355; E-mail: gpalu@microb.unipd.it K e y w o r d s : Therapy, gene therapy, brain tumors, gliomas, glioblastoma multiforme, molecular biology, pathogenesis, immunotherapy, neoangiogenesis, oncogenes A b b r e v i a t i o n : GBM, glioblastoma multiforme Received: 23 November 1998; accepted 30 November 1998

Summary Pathogenic features of glioblastoma multiforme and of other gliomas are reviewed in the present article. Emphasis is given to those genetic alterations which are involved in oncogenesis, to the p r o c e s s o f t u m o r n e o a n g i o g e n e s i s a n d t o t h e r o l e p l a y e d b y t h e i m m u n e s y s t e m i n controlling neoplastic growth. Aspects which are relevant to therapeutic interventions are also dissected, and gene therapy in particular. A new gene therapy approach that combines tumor suicide, via enzymedirected prodrug activation, and cytokine-promoted immune rejection i s reported, together with results from the first application of this approach in humans.

of patients for tumor recurrence within 6-12 months from treatment.

I. Introduction The outcome of malignant gliomas remains extremely poor, in spite of aggressive use of currently available therapies. Recent advances in elucidating the molecular biology of gliomas have led to the development of innovative therapeutic strategies. The more promising approaches involve gene therapy, aiming at increasing tumor cell chemosensitivity and/or immunogenicity, by transfer of genes expressing cytokines and prodrug activating enzymes.

Glioblastoma multiforme (grade IV astrocytoma) is usually located in the cerebral hemispheres, though it occasionally appears at other sites, such as the cerebellum, the brain stem and the spinal cord. Histology shows marked cytological diversity, ranging from tumors composed of small cells with scant cytoplasm to those composed of multinucleated giant cells. The World Health Organization (WHO) classification recognizes two distinct subvariants of the tumor: (i ) giant cell glioblastoma, characterized by a predominance of enormous, multinucleated giant cells and, on occasion, an abundant stromal reticulin network; and (i i ) gliosarcoma, in which hyperplastic vascular elements have undergone sarcomatous transformation.

Glioblastoma multiforme (GBM) represents 15-20% of all intracranial tumors and 50% of gliomas (Russel and Rubistein, 1989). It affects 5,000 Americans and 1,000 Italians every year, and typically occurs in adults, with a peak incidence in the fifth and sixth decades of life. It is a very aggressive tumor, with a uniform and profound morbidity. Because of its morbidity it contributes to the cost of cancer on a pro capite basis more than any other tumor. Despite surgery, radiotherapy and/or chemotherapy, the prognosis is extremely poor and has not substantially changed over the last two decades, death resulting in 80%

Current therapies for malignant gliomas include surgical removal of the tumor mass, which is mandatory for precise diagnosis, and irradiation. Although surgery improves the prognosis (Levin et al, 1993), the infiltrative behavior of malignant gliomas precludes their complete

133

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme resection, and 90% of GBM recur within 2 cm of the primary site. Postoperative radiotherapy is therefore commonly administered, with a significant improvement in survival (Hochberg and Pruitt, 1980; Walker et al, 1980). Despite surgery and irradiation, however, only a few patients are alive two years after diagnosis. Results of chemotherapy trials are disappointing (Hosli et al, 1998). This is due both to the tumor intrinsic chemoresistance (Petersdorf and Berger, 1996) and to the tumor location within the central nervous system, which limits the penetration of drugs (Janzer and Raff, 1987; Mak et al, 1995). Among malignant gliomas, GBM is the least responsive to medical treatment. Available protocols include both monochemotherapy and polychemotherapy regimens. Nitrosoureas are the leading drugs in glioma chemotherapy, with response rates as single agents varying from 10% to 40% (Young et al, 1973; Fewer et al, 1972; Hoogstraten et al, 1972). Other drugs, evaluated in monochemotherapy (Forsyth and Cairncross, 1996), occasionally showed clinically and radiologically objective responses. Among these are vincristine (Smart et al, 1968), procarbazine (Rodriguez et al, 1989), paclitaxel (Chamberlain and Kormanik, 1995; Prados et al, 1996), and temozolomide (Newlands et al, 1996). However, methodological bias present in most studies raise doubts about the validity of these results. The most commonly used polychemotherapy regimens for gliomas are PVC (i.e. a combination of CCNU, procarbazine, and vincristine) and MOP (i.e. a combination of procarbazine, vincristine, and mechlorethamine). Response rates (complete or partial) of 17-37% have been reported for glioblastomas (Levin et al, 1980; Coyle et al, 1990). More recently, interesting results have been obtained in GBM patients with the ICE regimen (ifosfamide, carboplatin and etoposide), although in association with severe hematological toxicity (Sanson et al, 1996). The role of PVC as adjuvant chemotherapy is controversial (Fine et al, 1993), and, overall, there is no clear-cut evidence that survival of glioblastoma patients is improved by chemotherapy (Hosli et al, 1998).

insurmountable task that gene replacement, or gene suppression, should simultaneously involve a number of different genes, and should be applied to all tumor cells to reverse the malignant phenotype. Hence, corrective gene therapy seems to be quite difficult to propose as a single therapeutic approach.

A. Genetic alterations 1. Oncogenes Several members of the protein-tyrosine kinase receptor family are over-expressed by gene amplification in malignant gliomas, including the epidermal growth factor receptor (EGFR), the platelet-derived growth factor receptor-! (PDGFR!) and the c-met genes (Furnari et al, 1995). A high percentage of glioblastomas also have EGFR gene rearrangements that may lead to the expression of a truncated, constitutively activated receptor. The transfer of a mutant human EGFR gene into glioblastoma cells caused constitutive self phosphorylation and a pronounced enhancement tumorigenicity in nude mice

II. Molecular biology of glioblastoma multiforme and corrective gene therapy As for most cancers, brain tumors derive from a multistep process of successive alterations, including loss of cell cycle control, neoangiogenesis and evasion of immune control. Figure 1 summarizes the genetic alterations associated with the malignant transformation of astrocytes. Most of these changes involve the loss of putative tumor suppressor genes or activation of proto-oncogenes.

o

Gene therapy of cancer, in its most direct form, should aim at replacing a mutated gene with its correct form, or at suppressing the abnormal oncogenic function. At present, however, such a corrective gene therapy, faces the

f

F i g u r e 1 . Simplified representation of oncogenes and tumor suppressor genes contributing to malignant progression of astrocytic tumors.

134

Gene Therapy and Molecular Biology Vol 1, page 135 (Ekstrand et al, 1994; Nishikawa et al, 1994). Numerous strategies are currently being investigated to specifically inhibit EGFR using antibodies, immunoconjugates or antisense technology. The selective inhibition of EGFR in human GBM cells with kinase-deficient mutants inhibited cell proliferation and transforming efficiency in athymic mice (O'Rourke et al, 1997), and converted radioresistant human glioblastoma cells to a more sensitive phenotype (O'Rourke et al, 1998), providing a rationale for gene therapy applications.

cells with wild type p53 can significantly inhibit growth and neoangiogenesis, or can induce apoptosis in p53 mutant cells in several tumor models in vitro, including gliomas (Badie et al, 1995; Van Meir et al, 1995; GomezManzano et al, 1996). The presence of functional p53 has also been shown to modulate chemoresistance. Consequently, another possible advantage of the restoration of wild type p53 may be sensitization to chemotherapy and radiotherapy. Indeed, the combination of p53 gene transduction with radiation or chemotherapy (Lowe et al, 1994) has resulted in local tumor control superior to either therapy alone (Fujiwara T et al, 1994; Gjerset et al, 1995; Ngyuyen et al, 1996, Lang et al, 1998). This combined therapy is currently under investigation in clinical trials (Roth and Cristiano, 1997; Nielsen and Maneval, 1998).

Other dominant oncogenes, such as N-myc, fos, src, Hras or N-ras, and mdm2 are amplified and highly expressed in gliomas (Collins, 1993). GBM produce high levels of insulin-like growth factor I (IGF-I). When this alteration has been targeted by a vector expressing an antisense antiIGF-I gene, rejection of genetically altered rat C6 glioma cells was observed. Injection, even at a site distal to the tumor, caused regression of established brain GBM. Destruction of the tumor was mediated by a gliomaspecific T CD8+ (CTL) response (Trojan et al, 1993).

The cell cycle regulator genes provide an additional target for corrective gene therapy. The p105Rb product of the retinoblastoma tumor suppressor gene (Rb) is one of the most critical regulators of cellular proliferation. The Rb protein (pRb), when unphosphorylated, is responsible for arrest of cell cycle by inhibition of the activity of the E2F family of transcription factors. Normal cell cycle progression requires inactivation of Rb through phosphorylation by cyclin-dependent kinases (CDK). This process, in turn, is regulated by CDK inhibitors. Among these, p21 protein is induced directly by p53; p16 protein, and its homologue p15, specifically bind to and inhibit CDK4, and may therefore regulate Rb phosphorylation, and cell cycle progression. Dysregulation of cell cycle control is a frequent finding in malignant gliomas, like deletion or loss of expression of p16 and p15 tumor suppressor genes (Jen et al, 1994; Nishikawa et al, 1995), amplification of CDK4 (He et al, 1994), and deletion or mutation of the Rb tumor suppressor gene (Henson et al, 1994). Interestingly, both of the latter events take place when the p16 gene is intact and correctly expressed (He et al, 1995). Restoration of wild-type p16 gene in glioma cells through an adenoviral vector arrested cells in G0-G1 phases of the cell cycle (Fueyo et al, 1996) and suppressed glioma cells invasion in vitro (Chintala et al, 1997). Overexpression of p21 increases the susceptibility of glioblastoma cells to cisplatin-induced apoptosis (Kondo et al, 1996), whereas adenovirus-mediated transfer of exogenous E2F-1 protein induced massive apoptosis and suppressed glioma growth in vivo and in vitro (Fueyo et al, 1998). The possibility that E2F-responsive promoters may be more active in tumor cells relative to normal cells, because of loss of pRb function, has been exploited to design adenoviral vectors containing transgenes driven by the E2F-1 promoter for gene therapy of gliomas (Parr et al, 1997). These vectors showed tumor-selective gene expression in vivo and reduced toxicity of the normal tissue with respect to standard adenoviral vectors.

2. Genes associated to cell immortalization A role in cell immortalization has been proposed for telomerase, the RNA-protein complex that elongates telomeric DNA. Telomerase is expressed almost exclusively in cancer cells, but not in normal cells, suggesting the possibility that gene therapy may be applied to inhibit this function. A successful example of treatment via antisense oligonucleotides directed against human telomerase suppressed glioma cells growth and survival, both in vitro and in vivo, through the induction of apoptosis (Kondo et al, 1998). 3. Tumor suppressor genes Molecular and cytogenetic analyses of gliomas have shown frequent losses of genetic material, suggesting the inactivation of putative tumor suppressor genes. Loss of heterozygosity (LOH) has been described in chromosome 1p, 9p, 10p, 10q, 11p, 13q, 17p, 19q, and 22q, and in some cases the tumor suppressor gene involved in LOH has been identified. This is the case of the p53 tumor suppressor gene, which maps in 17p. Wild-type p53 protein is involved in G1 cell cycle arrest and apoptosis of DNA-damaged cells and is therefore crucial in preventing mutation or deletion of functional genes. Mutations of p53 seem to be an early event in glioma tumorigenesis, being frequently detected also in low grade astrocytomas. Along with p53 mutations, amplification of the mdm2 oncogene, whose product binds to and degrades p53, accounts for p53 inactivation in gliomas. Since p53 plays a key role in the pathogenesis of most cancers, it has raised great interest as a target for cancer gene therapy. Transduction of malignant

135

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme Inhibition or inactivation of genes/factors involved in DNA repair and/or cellular SOS response could represent a gene therapy approach that potentiates radiation therapy. In fact, inhibition of the RAD51 gene by antisense oligonucleotides enhanced the radiosensitivity of mouse malignant gliomas, both in vitro and in vivo, improving survival (Ohnishi et al, 1998). This gene, a homologue of the yeast RAD51 and E. coli RecA genes, is involved in repair of DNA double-strand breaks, in recombination repair, and in various SOS responses to DNA damage caused by gamma-irradiation and alkylating agents.

ribozymes against VEGF mRNA have been successfully employed to reduce VEGF expression in glioma cells (Ke et al 1998), once more suggesting a potential role for antiangiogenic gene therapy. Similarly, bFGF antisense cDNA decreased C6 glioma cells proliferation (Redekop and Naus, 1995). Besides inhibiting the production of angiogenic factors, a therapeutic intervention could also consist of providing tumors with antiangiogenic factors. Indeed, retroviral and adeno-associated viral vectors expressing a modified PF4 were reported to inhibit endothelial cells proliferation in vitro and the growth of intracerebrally implanted gliomas (Tanaka et al, 1997). Retroviral and adenoviral vectors transducing angiostatin gene increased apoptotic death of glioma tumor cells (Tanaka et al, 1998). Additionally, the intratumoral delivery of angiostatin gene by an adenoviral vector produced inhibition of tumor growth in vivo, suppression of neovascularization, and a marked increase of tumor cells apoptosis (Griscelli et al, 1998). Damage of tumor microvasculature was reported also in human malignant glioma xenografts, after gene therapy followed by radiotherapy. The treatment consisted of intratumoral injection of adenoviral vectors expressing tumor necrosis factor-! (TNF-!), under control of the Egr-1 promoter (Staba et al, 1998). The use of viral vectors containing radiation-inducible promoters, such as Egr-1, has the advantage of selectively, spatially, and temporally limiting the effects of the therapeutic gene in the radiation field. Recently, this strategy has yielded interesting results in rat 9L glioma cells (Manome et al, 1998).

Deletions of large regions or even of the entire copy of chromosome 10 are a genetic hallmark of GBM. At least two tumor suppressor genes located on chromosome 10 (one on each arm) have been demonstrated to participate to glial oncogenesis. A first candidate tumor suppressor gene, called PTEN (Phosphatase and tensin homologue deleted on chromosome TEN) was recently characterized (Li et al, 1997). The DNA region encoding PTEN is altered in glioblastoma multiforme, but not in lower grade astrocytic tumors (Tohma et al, 1998; Ichimura et al, 1998; Chiariello et al, 1998).

B. Neoangiogenesis Tumors may remain in a state of dormancy until they establish a blood supply for receiving oxygen and nutrients. The complex process of neoangiogenesis is regulated by numerous factors, some with angiogenic properties, i.e. vascular endothelial growth factor (VEGF), platelet derived growth factor (PDGF!), basic fibroblast growth factor (bFGF), epidermal growth factor (EGF), interleukin-8, and by endogenous inhibitors of angiogenesis, i.e. thrombospondin-1, platelet factor 4 (PF4), angiostatin, endostatin. VEGF, which binds to two specific tyrosine-kinase receptors, called Flk-1 and Flt-1, has been demonstrated to play a key role in angiogenesis of gliomas. Indeed, VEGF and its receptors are downregulated in the normal adult brain, whereas, VEGF is highly produced by GBM cells. Since both flt-1 and flk-1 genes are expressed by proliferating endothelial cells of gliomas, this leads to the establishment of a paracrine loop. Moreover, VEGF expression is higher around necrotic areas and seems to be stimulated by hypoxia.

III. Suicide gene therapy Suicide gene therapy operates by tumor transduction of genes converting a prodrug into a toxic substance; independently, the gene product and the prodrug are nontoxic. The prototype of this approach exploits the selective intracellular phosphorylation of ganciclovir (GCV), driven by the herpes simplex virus thymidine kinase gene product (HSV-TK). This activation generates a toxic drug metabolite that inhibits DNA synthesis, inducing cell death. For in vivo gene transfer of the HSV-TK gene to malignant cells, packaging cells that produces retroviral vectors expressing HSV-TK, have been injected directly into the tumor to transduce replicating cells. An interesting feature of the HSV-TK/GCV system is the bystander killing of nontransduced cells.

Glioblastoma multiforme is one of the most highly vascularized solid neoplasms; therefore, treatments that target neoangiogenesis would be of great interest in clinical practice. Co-injection of rat C6 glioma cells, either subcutaneously or intracerebrally in nude mice, together with cells producing retroviral vectors encoding a dominant-negative mutant of the Flk-1 receptor showed inhibition of neoangiogenesis, reduction of tumor growth, and survival improvement (Millauer et al, 1994; 1996). Antisense VEGF oligonucleotides (Saleh et al, 1996) and

The mechanisms that are responsible for this effect have not been fully defined, but are likely to include: (i ) transfer of non-diffusible, phosphorylated GCV to neighboring cells through gap junctions; (i i ) endocytosis by nontransduced cells of cellular debris containing toxic GCV; and (i i i ) stimulation of host antitumor immune response. The therapeutic efficacy of the HSV-TK/GCV

136

Gene Therapy and Molecular Biology Vol 1, page 137 system may be further increased by the use of adenoviral vectors, since these vectors can also transduce resting tumor cells. However, adenoviral vectors will infect also normal cells; hence, the inclusion of sequences able to restrict gene expression only in tumor cells can circumvent this problem. Selective tumor toxicity was obtained positioning the suicide gene under control of the E2Fresponsive promoter elements which are de-repressed in glioma cells (Parr et al, 1997).

have shown encouraging results both in vitro and in vivo (Khil et al, 1995; Kim et al, 1997). 9L glioma cells transduced with a retrovirus encoding a CD/HSV-TK fusion gene exhibited enhanced sensitivity to both GCV and 5-FC, as well as increased radiosensitivity (Rogulski et al, 1997). This experiment suggests the feasibility of a combined approach with two suicide genes associated with radiotherapy. Another prodrug activation system is represented by cytochrome P450 2B1 (the liver enzyme catalyzing cyclophosphamide and ifosfamide activation) gene transfer followed by cyclophosphamide or ifosfamide administration. Metabolites of these drugs produce interstrand DNA cross-linking in a cell cycle-independent fashion. C6 and 9L rat glioma cells, when stably transfected with the P450 2B1 gene, become highly sensitive to cyclophosphamide in in vitro and in vivo models (Wei et al, 1994; Manome et al, 1996). Rabbit cytochrome P450 isozyme CYP4B1, which converts the inert prodrugs 2-amonianthracene (2-AA) and 4-ipomeanol (4-IM) into highly toxic alkylating metabolites, also showed high antitumor effects, both in vitro and in vivo. The treatment had relatively low toxicity and was associated with a bystander effect, not requiring cell-to-cell contact (Rainov et al, 1998).

The effectiveness of suicide gene therapy has been explored for a variety of neoplasms, especially for refractory and localized diseases such as GBM. The HSVTK/GCV scheme has been demonstrated to be effective in animal models, by ex vivo and in vivo transduction with retroviral, adenoviral, or adenoviral-associated vectors expressing HSV-TK (Ezzeddine et al, 1991; Culver et al, 1992; Takamiya et al, 1992, 1993; Barba et al, 1993, 1994; Ram et al, 1993; Kim et al, 1994; Vincent et al, 1996; Maron et al, 1996; Mizuno et al, 1998). Novel gene delivery systems were also applied in a mouse model for gene therapy of meningeal gliomatosis. Liposomes coated with Sendai virus envelope protein were highly efficient in delivering the therapeutic gene in disseminated glioma cells (Mabuchi et al, 1997). A factor that may limit the effectiveness of HSVTK/GCV therapy is the GCV crossing through the bloodbrain barrier. This can be circumvented by the use of the bradykinin analogue and potent blood-brain barrier permeabilizer RMP-7, which, administered intravenously, increase the delivery of GCV into rat brain tumors, enhancing the cytotoxic and bystander effects of HSVTK/GCV (LeMay et al, 1998).

Another suicide gene system is based on E. coli purine nucleoside phosphorylase (PNP), which generates toxic purine nucleoside analogues intracellularly, either from 6methylpurine-2’-deoxyriboside or arabinofuranosyl-2fluoroadenine monophosphate. Significant antitumor activity and low systemic toxicity were reported in nude mice bearing human malignant D54MG glioma tumors expressing PNP (Parker et al, 1997).

Several clinical trials exploiting the HSV-TK/GCV system have been initiated. It is too early to estimate the effectiveness of these therapeutic procedures. Notwithstanding the evidence for growth-suppressive activities of HSV-TK plus GCV, cure rates are low. Explanation for lack of complete response in humans may reside in the different biological behavior of GBM cells when injected into animals (Sturtz et al, 1997).

Phosphorylation of the prodrug cytosine arabinoside (ara-C) by deoxycyticine kinase (dCK) is a limiting step for activation. Thus, ara-C, a potent antitumor agent for hematological malignancies, has only minimal activity against most solid tumors. Transduction of the dCK cDNA by retroviral and adenoviral vectors also resulted in marked sensitization of glioma cell lines to ara-C in vitro, and in significant antitumor activity in vivo (Manome et al, 1996).

The antitumor effects elicited by HSV-TK/GCV prompted to explore other prodrug activating enzymes. A promising system exploits the ability of E. coli cytosine deaminase (CD) to convert the relatively non toxic 5fluorocytosine (5-FC) to the chemotherapeutic agent 5fluorouracil (5-FU). Significant antitumor effects of CD/5FC were observed in nude mice bearing tumors derived from C6 glioma cells and transduced with CD. A "bystander" effect could also be demonstrated (Ge et al, 1997), suggesting a potential role for gene therapy of glioblastoma.

Unlike other prodrug activating enzymes, E. coli gpt sensitizes cells to the prodrugs 6-thioxanthine (6TX) and 6thioguanine (6-TG), and confers resistance to different regimens (mycophenolic acid, xantine, and hypoxanthine), providing a means to select for gpt-positive cells. Rat C6 glioma cells transduced with a retroviral vector expressing the gpt gene exhibited significant 6TX and 6GT susceptibility and a "bystander" effect in vitro. An antiproliferative effect was demonstrated in vitro and in vivo (Tamiya et al, 1996; Ono et al, 1997).

As already reported for p53, both the CD and HSV-TK systems sensitize cancer cells to radiation. Animal models 137

Palù et al: Perspectives for therapy of glioblastoma multiforme Tumor suicide can also be achieved by direct infection of tumor cells with a conditionally replicative virus, i.e. an infectious agent able to replicate and to kill only dividing cells. In the case of tumors highly proliferating in the context of a completely post-mitotic tissue, such as the brain, gene transfer can ideally be obtained by using neurotropic herpes viral vectors, which are rendered conditionally replicative after deletion of non-essential genes (Lachmann and Efstathiou, 1997).

Despite the location in the central nervous system (CNS), a long-believed “immunologically privileged site”, glioblastoma cells may interact with immune cells. These interactions are mediated by receptor-ligand recognition during cell to cell contact and by a plethora of cytokines. An imbalance in the tumor-host relationship, resulting in deficit in some components of the response, may explain the aggressive growth of malignant gliomas. Before discussing the designed strategies to increase the immune response against glioblastoma cells, we review the more recent acquisitions in the “dialogue” between these neoplastic cells and the immune system (Dietrich et al, 1997).

Tumor specific cell death has already been provided in animal glioma models by HSV vectors, deleted in neurovirulence genes, such as " 34.5, thymidine kinase and ribonucleotide reductase (Chambers et al, 1995; Mineta et al, 1995; Boviatsis et al, 1994; Miyatake et al, 1997; McKie et al, 1996; Andreansky et al, 1996). Their direct injection into gliomas produced tumor regression with minimal bystander effects on surrounding normal tissue. Enhancement of replication of defective HSV vectors lacking " 34.5 gene and a significant reduction of tumor mass was observed combining ionizing radiation (Advani et al, 1998).

A. Antigenicity of glioblastoma cells The presence of specific antigens at the surface of tumor cells to be recognized by cells of the immune system is essential for the generation of a specific antitumor immune response. At present, no tumor antigen able to elicit an immune response has been identified in glioblastoma in vivo. MAGE-1 (melanoma antigen) was the first tumor-specific antigen to be identified (Van der Bruggen et al, 1991), and MAGE family members are expressed by some glioblastoma cell lines (Rimoldi et al, 1993), but not in uncultured tumors (De Smet et al, 1994). This observation could be explained with different levels of DNA methylation induced by culture, where MAGE expression was regulated by methylation (De Smet et al, 1995).

Another way to treat malignant gliomas emerged from the discovery that these tumors often express functional Fas (CD95) (Weller et al, 1994). Fas is a transmembrane glycoprotein belonging to the nerve growth factor/TNF receptor superfamily: when activated, Fas can transduce an apoptotic signal through its cytoplasmic domain. Apoptosis is triggered by the binding of Fas to its natural ligand (FasL) or by cross-linking with anti-Fas antibodies. A high proportion of human glioma cell lines are sensitive to apoptosis mediated by anti-Fas antibodies in vitro. Some other cell lines are resistant, but may be rendered sensitive after stable transfection with human Fas cDNA (Weller et al, 1995). These results offer new possibilities for treating gliomas with anti-Fas antibodies or soluble FasL. One possible drawback of such an approach is that other Fas-positive cells may be affected, like infiltrating leukocytes. Thus their activity may be reduced, restricting strategies relying upon simultaneous immune response enhancement.

Proteins that are structurally altered during malignant transformation, or that contribute to this process, are possible tumor antigen candidates. The frequent alterations of p53 in the early stages of carcinogenesis, for example, may provide new antigenic peptides that could trigger an immune response. Consistent with this possibility, specific cytotoxic T-cell clones were generated in vitro against mutated p53 protein (Houbiers et al, 1993). Additionally, in vivo immunization with mutated p53 peptide was shown to induce specific CTL clones able to lyse MHC-matched tumor cells expressing the mutated p53 gene (Noguchi et al, 1994).

IV. Immunotherapy of cancer

The future identification of glioblastoma-specific antigens would aid new treatment strategies.

Glioblastoma multiforme has the propensity to microscopically infiltrate normal structures since its early stage of development. This characteristic makes therapeutic success rather difficult by any approach. Moreover, it becomes virtually impossible to obtain specific targeting of all tumor cells and sparing of normal ones. Therefore, inducing an efficient immune response against malignant cells becomes an attractive and essential treatment strategy. In this perspective, the natural circulatory properties of cells of the immune system offer also an important support for the recognition of secondary lesions.

B. Tumor-induced immunosuppression A high proportion of glioblastomas have shown to be infiltrated by lymphocytes, mostly T lymphocytes, but also B and NK cells. Differently from what reported for other tumors, the level of lymphocyte infiltration does not relate with a better prognosis. In fact, tumor-infiltrating lymphocytes (TIL) of gliomas appear to be functionally defective. Abnormalities range from abnormal 138

Gene Therapy and Molecular Biology Vol 1, page 139 hypersensitivity responses, depressed response to mitogens, decreased humoral responses (CD4+ T helper cells deficit?), and impaired T cell-mediated cytotoxicity. These functional alterations may be explained, at least in part, by a defective high-affinity IL-2 receptor (IL-2 R) (Roszman et al, 1991). It has been demonstrated that glioblastoma cells produce and release soluble factors that are responsible for a depressed immune response. T lymphocytes from normal individuals exhibit immunologic abnormalities when grown in presence of supernatant obtained from glioblastoma cell line cultures (Roszman et al, 1991). The most important soluble suppressing factor seems to be TGF-#2. This cytokine acts as a growth-inhibitory factor (Sporn et al, 1986; Sporn et al, 1987), and has a defined variety of immunoregulatory properties, including: inhibition of: (i ) T cell proliferation; (i i ) IL-2R induction (Kehrl et al, 1986); (i i i ) cytokine production (Espevik et al, 1987; Espevik et al, 1988); (i v ) natural killer cell activity (Rook et al, 1986); (v) cytotoxic T cell development (Jin et al, 1989; Ranges et al, 1987); (v i ) LAK cell generation (Espevik et al, 1987; Jin et al, 1989); and (v i i ) production of tumor-infiltrating lymphocytes (Kuppner et al, 1989). Most cells secrete TGF-# in a latent form (Sporn et al, 1986), but glioblastoma cells have also the capacity to convert it to an active form, through proteolytic cleavage. This was demonstrated by an experimental work in which T-cell suppression mediated by TGF-#2 was inhibited when proteolytic enzymes were blocked by protease inhibitors (Huber et al, 1992). Other soluble factors, namely prostaglandin E2 (PGE2), IL-1 receptor (IL-1 R) antagonist, and interleukin-10 (IL10) may be implicated in immunosuppression, although in vivo intervention has not been fully defined. A potential down regulation of some immune functions was shown for IL-10 cytokine, which was produced both by GBM cells and normal brain tissue (Nitta et al, 1994; Merlo et al, 1993). Furthermore, in different animal models, human and murine IL-10 was demonstrated to stimulate the acquisition of a specific and efficient antitumor immune response (Berman et al, 1996).

C. Costimulatory molecules A complete T-cell effector function needs not only antigen presentation, but also the delivering of activation signals to the T cell, which is mediated by the so-called costimulatory molecules. Unresponsiveness of T cells (anergy) may be due to absence of the second signal, essentially given by the B7-CD28 interaction (June et al, 1994). Two other members of the B7 family have been cloned, B7.1 and B7.2; their counter-receptors on T cells are CD28 and CTLA-4, respectively. CD28 mediates stimulatory effects, while CTLA-4 appears to be a negative

regulator of T cell responses. Glioblastoma cells and monocytes that infiltrate the tumor are not expressing B7 costimulatory molecules, while monocytes in the normal tissue that surrounds the tumor are B7-positive (Tada et al, 1996). This suggests the possible intervention of local mechanisms able to down-regulate B7 expression in glioblastomas, impeding efficient T-cell priming and favoring T-cell anergy. Moreover, B7-CD28 interactions in the CNS have been shown to be essential to generate a valid CTL response towards viral antigens (Kuendig et al, 1996). Hence, restoring B7 expression by gene transfer might become an interesting task to elicit a proper immune recognition of glioblastoma cells, and an appropriate immune response.

V. Restoring a proper immune response Many approaches have been tented to restore a proper immune response towards malignant gliomas. As previously stated, T lymphocytes play a major role in the antitumor response, and priming of T lymphocytes requires antigen recognition, with or without help from APC. Since no specific antigens have yet been identified for glioblastomas, a vaccination approach has been proposed by administration of genetically modified tumor cells. Moreover, tumor cells transfected to produce various cytokines have been used to enhance lymphocyte responsiveness in animal models. The most interesting results were obtained with cells of murine glioma transfected with an expression vector containing the murine interleukin 7 cDNA (Aoki et al, 1992). IL-7 transfected glioma cells were vigorously rejected by a CD8+ T-cell-mediated immune response, that was proportional to the level of IL-7 production. Moreover, the response was tumor-specific, since no effect was observed against other syngeneic tumor cells (melanoma and fibrosarcoma cells). IL-7 is a very interesting cytokine being able to increase IL-2R ! chain expression on CD4+ T lymphocytes and to inhibit TGF-# mRNA expression and production by murine macrophages (Dubinett et al, 1993). IL-12 can also be considered a promising agent to enhance the antitumor response, since it augments T-cell and natural killer-cell activities, induces IFN-" production, and promotes the differentiation of uncommitted T cells to Th1 cells (Hendrzak and Brunda, 1995). Vector-mediated delivery of IL-12 into established tumors suppresses tumor growth (Caruso et al, 1996) and can induce immune responses against challenge tumors (Bramson et al, 1996). Moreover, IL-12 has other nonimmune properties such as anti-angiogenic effects (Voest et al, 1995). IL-4, another cytokine with pleiotropic functions, increases T cell proliferation and cytotoxicity, and enhances eosinophil and B cell proliferation and differentiation. It

139

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme

VI. Combined gene therapy approach in humans

also exerts direct anti-proliferative effects in vitro against many tumor cell lines (Tepper, 1993). Antitumor effects induced by IL-4-transfected cells were reported in nude mice, suggesting T cell-independent mechanisms (Tepper, 1993; Yu et al, 1993). A strong recruitment of eosinophils and subsequent inhibition of the tumor growth was noted. Eosinophil depletion was not performed as a control; hence a direct anti-proliferative effect mediated by IL-4 cannot be excluded. In any respect, a possible non-T-cell-dependent mechanism could be an advantage in the glioblastoma setting, considering the various abnormalities of T cell function.

The strict localization of glioblastomas in the CNS, with only exceptional metastases, makes these tumors candidates for approaches of direct intra-tumoral gene delivery. Retrovirus-mediated gene therapy of GBM is particularly attractive, since these viral vectors transduce only mitotically active cells, sparing the normal neuronal tissue composed of non-replicating cells. Gene therapy of brain tumors by intra-tumoral injection of retroviral vector producing cells (RVPC) in human patients was initiated by Oldfield and colleagues in 1993 (Oldfield et al, 1993). The gene being transferred was that expressing the herpes simplex thymidine kinase (HSV-TK), which conferred sensitivity to the anti-herpes drug ganciclovir. This treatment has proved free of toxicity and safe for there was no evidence of systemic spread of the retroviral vector (Long et al, 1998). However, a clinical benefit was limited to very small tumors (1.5 ml), probably because only malignant cells adjacent to the RVPC were transfected (Ram et al, 1997).

In an attempt to overcome the local immunosuppression mediated by TGF-#2 Fakhrai et al conducted an experimental work on rats by antisense gene therapy. 9L gliosarcoma transfected cells inoculated subcutaneously became highly immunogenic and were able to induce eradication of an established wild-type tumor (Fakhrai et al, 1996). The enhancement of antigen presentation and T cell costimulation has also been considered and may be achieved with genes coding for cytokines, like GM-CSF, costimulatory molecules, like B7, or CIITA, a transcription factor playing a critical role in the regulation of MHC class II molecules. Encouraging results have been reported in a murine melanoma model located in the CNS, whereby an efficient antitumor response was induced by subcutaneous vaccination with irradiated, GM-CSF-producing tumor cells (Sampson et al, 1996). Vaccination with cells co-transfected with B7 and IL-2 was able to mediate rejection of established tumors (Gaken et al, 1997), suggesting a possible application of such an intervention for the treatment of glioblastomas.

A new treatment strategy combining two different modalities, enzyme-directed prodrug activation (tumor suicide) along with cytokine-promoted tumor rejection, has been recently devised to amplify the antitumor response, and proved to be efficacious in animal models (Castleden et al, 1997) (Figure 2). A bicistronic retroviral vector coexpressing HSV-TK and human interleukin-2 genes has been designed to pursue this new approach of cancer gene therapy in humans (Pizzato et al, 1998).

F i g u r e 2 . Structure of a bicistronic retroviral vector for transduction of genes coding for a cytokine and a prodrugactivating enzyme, expressed via a cap-dependent and an internal ribosome entry site (IRES)-dependent mechanism, respectively. A selectable marker (neomycin phosphotransferase gene, neo) is expressed under the control of a SV40 promoter.

140

Gene Therapy and Molecular Biology Vol 1, page 141

F i g u r e 3 . Gene therapy approach for treatment of glioblastoma multiforme, via intratumoral stereotactic injection of cells producing a triple gene retroviral vector.

F i g u r e 4 . Contrast-enhanced MRI sagittal images of left parietal GBM lesion before (l e f t ), and one month after completion of GCV treatment (t o t h e r i g h t ).

141

Gene Therapy and Molecular Biology Vol 1, page 142

F i g u r e 5 . Histology of stereotactic biopsies from patients treated by HSV/TK-IL-2 combined gene therapy. A) Toluidine blue staining - Evidence of a large number of infiltrating inflammatory cells; Immunostaining with B ) monoclonal antibodies marking CD3+ cells; C) Mac 387 antibodies recognizing young/activated macrophages; D) monoclonal antibodies marking CD1+ infiltrating cells.

trials with thymidine kinase (Ram et al, 1997; Ostertag and Chiocca, personal communications).

After in vitro characterization of efficacy and safety (Pizzato et al, 1998), the vector was employed in a pilot study to treat four patients with recurrent glioblastoma multiforme (Colombo et al, 1997; Pal첫 et al, 1998) (Figure 3). A significant and sustained reduction (>50% of the initial volume) of the tumor mass (80 ml) was demonstrated by magnetic resonance imaging (MRI) and computerized tomography (CT) in one patient (Figure 4). In this case, the objective response was associated with a dramatic clinical improvement. The other three patients showed areas of tumor necrosis (2 ml) around the site of stereotactic RVPC injection and stabilized disease for a long period of time (11-12 months).

Interestingly, endothelial cells stained positive for TK by in situ hybridization, indicating that the vector had targeted the neo-vascular component, a highly replicative population in glioblastomas. This is consistent with an anti-angiogenic effect of this therapeutic approach, that, in addition to direct tumor suicide and immune activation, may be relevant to the bystander phenomenon and to the clinical response. It is noteworthy that IL-2 was measurable in the cerebral-spinal fluid, even after GCV treatment. This cytokine might have derived from an autocrine-paracrine secretion of recruited infiltrating immune-inflammatory cells, after primary expression in transduced cells.

In stereotactic biopsies taken before ganciclovir administration, large tumor infiltrates of immuneinflammatory cells (T lymphocytes, mostly CD4+ but also CD8+ granzyme B-positive cells, activated macrophages, NK cells, neutrophils) were present, notwithstanding the standard steroid therapy (Figure 5). The observed inflammatory response has never been reported in previous

Efforts to achieve more efficient gene transfer systems are being sought for. These include the development of new generation retroviral vectors, produced at higher titres and characterized by higher transduction efficiency. Strategies involving envelope pseudotyping, use of new

142

Gene Therapy and Molecular Biology Vol 1, page 143 Bramson JL, Hitt M, Addison CL, Muller WJ, Gauldie J, Graham FL. (1 9 9 6 ) Direct intratumoral injection of an adenovirus expressing interleukin-12 induces regression and long-lasting immunity that is associated with highly localized expression of interleukin-12. Hum Gene Ther 7, 1995-2002

packaging cell lines of human origin, and substitution of promoter elements will contribute to the improvement of current available vectors. New therapeutic gene combinations should also be accomplished in order to promote a more generalized immune response. Genes for cytokines other than IL-2 (i.e., IL-4, IL-7, IL-12, GM-CSF) as well as genes targeting neoangiogenesis deserve further consideration for combined treatment approaches.

Caruso M, Pham-Nguyen K, Kwong Y-L, Xu B, Kosai K-I, Finegold M, Woo SLC, Chen SH. (1 9 9 6 ) Adenovirusmediated interleukin-12 gene therapy for metastatic colon cancer. Proc Natl Acad Sci USA 93, 11302-11306

The authors wish to acknowledge Fondazione Cassa di Risparmio di Padova e Rovigo and Regione Veneto for financial support.

Castleden SA, Chong H, Garcia-Ribas I, Melcher AA, Hutchinson G, Roberts B, Hart IR, Vile RG. (1 9 9 7 ) A family of bicistronic vectors to enhance both local and systemic antitumor effects of HSVTK or cytokine expression in a murine melanoma model. Hum Gene Ther 8, 2087-102.

References

Chamberlain MC, Kormanik P. (1 9 9 6 ) Salvage chemotherapy with paclitaxel for recurrent primary brain tumors. J C l i n O n c o l 13, 2316-21.

Advani SJ, Sibley GS, Song PY, Hallahan DE, Kataoka Y, Roizman B, Weichselbaum RR. (1 9 9 8 ) Enhancement of replication of genetically engineered herpes simplex viruses by ionizing radiation: a new paradigm for destruction of therapeutically intractable tumors. Gene Ther 5, 160-165.

Chambers R, Gillespie GY, Soroceanu L, Andreansky S, Chatterjee S, Chou J, Roizman B, Whitley RJ. (1 9 9 5 ) Comparison of genetically engineered herpes simplex viruses for the treatment of brain tumors in a scid mouse model of human malignant glioma. P r o c N a t l A c a d S c i USA 92, 1411-1415.

Andreansky SS, He B, Gillespie GY, Soroceanu L, Markert J, Chou J, Roizman B, Whitley RJ. (1 9 9 6 ) The application of genetically engineered herpes simplex viruses to the treatment of experimental brain tumors. P r o c N a t l A c a d Sci USA 93, 11313-11318.

Chen CY, Chang YN, Ryan P, Linscott M, McGarrity GJ, Chiang YL. (1 9 9 5 ) Effect of herpes simplex virus thymidine kinase expression levels on ganciclovirmediated cytotoxicity and the "bystander effect". Hum Gene Ther 6, 1467-1476.

Aoki T, Tashiro K, Miyatake S, Kinashi T, Nakano T, Oda Y, Kikuchi H, Honjo T. (1 9 9 2 ) Expression of murine interleukin 7 in a murine glioma cell line results in reduced tumorigenicity in vivo P r o c N a t l A c a d S c i U S A 89, 3850-3854.

Cheney IW, Johnson DE, Vaillancourt MT, Avanzini J, Morimoto A, Demers GW, Wills KN, Shabram PW, Bolen JB, Tavtigian SV, Bookstein R (1 9 9 8 ) Suppression of tumorigenicity of glioblastoma cells by adenovirusmediated MMAC1/PTEN gene transfer. C a n c e r R e s 58, 2331-2334.

Acknowledgements

Badie B, Drazan KE, Kramar MH, Shaked A, Black KL. (1 9 9 5 ) Adenovirus-mediated p53 gene delivery inhibits 9L glioma growth in rats. Neurol Res 17, 209-216.

Chiariello E, Roz L, Albarosa R, Magnani I, Finocchiaro G. (1 9 9 8 ) PTEN/MMAC1 mutations in primary glioblastomas and short-term cultures of malignant gliomas. O n c o g e n e 16, 541-545.

Barba D, Hardin J, Ray J, Gage FH. (1 9 9 3 ) Thymidine kinase-mediated killing of rat brain tumors. J Neurosurg 70, 175-82,.

Chintala SK, Fueyo J, Gomez Manzano C, Venkaiah B, Bjerkvig R, Yung WK, Sawaya R, Kyritsis AP, Rao JS. (1 9 9 7 ) Adenovirus-mediated p16/CDKN2 gene transfer suppresses glioma invasion in vitro. O n c o g e n e 15, 17, 2049-57.

Barba D, Hardin J, Sadelain M, Gage FH. (1 9 9 4 ) Development of antitumor immunity following thymidine kinase-mediated killing of experimental brain tumors. Proc Natl Acad Sci USA 91, 4348-4352. Berman RM, Suzuki T, Tahara H, Robbins PD, Narula SK, Lotze MT. (1 9 9 6 ) Systemic administration of cellular IL10 induces an effective, specific, and long-lived immune response against established tumors in mice. J Immunol 157, 231-238

Chiocca EA. (1 9 9 5 ) Brain tumor gene therapy in mice with a novel "suicide" gene, the cyclophosphamide-activating CYP2B1 gene. Clin Neurosurg 42, 370-82.

Boviatsis EJ, Scharf JM, Chase M, Harrington K, Kowall NY, Breakefield XO, Chiocca EA. (1 9 9 4 ) Antitumor activity and reporter gene transfer into rat brain neoplasms inoculated with herpes simplex virus vectors defective in thymidine kinase or ribonucleotide reductase. Gene Ther 1, 323-31.

Colombo F, Zanusso M, Casentini L, Cavaggioni A, Franchin E, Calvi P, Pal첫 G. (1 9 9 7 ) Gene stereotactic neurosurgery for recurrent malignant gliomas. Stereotact Funct Neurosurg 68, 245-251.

Collins VP. (1 9 9 3 ) Amplified genes in human gliomas. S e m i n C a n c e r B i o l 4, 27-32.

143

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme Coyle T, Baptista J, Winfield J. (1 9 9 0 ) Mechlorethamine, vincristine and procarbazine chemotherapy for recurrent high-grade glioma in adults, A phase II study. J C l i n O n c o l 8, 2014-2018.

Forsyth PA, Cairncross JG. (1 9 9 6 ) Chemotherapy of malignant gliomas. In Cerebral Gliomas. B a i l l e r e ' s Clin Neurol 5, 371-93. Fueyo J, Gomez Manzano C, Yung WK, Clayman GL, Liu TJ, Bruner J, Levin VA, Kyritsis AP. (1 9 9 6 ) Adenovirusmediated p16/CDKN2 gene transfer induces growth arrest and modifies the transformed phenotype of glioma cells. O n c o g e n e 12, 103-110.

Culver KW, Ram Z, Wallbridge S, Ishii I, Oldfield EH, Blaese RM. (1 9 9 2 ) In vivo gene transfer with retroviral vectorproducer cells for treatment of experimental brain tumors. S c i e n c e 256, 1550-1552. De Smet C, Courtois SJ, Faraoni I, Lurquin C, Szikora JP, De Backer O, Boon T. (1 9 9 5 ) Involvement of two Ets binding sites in the transcriptional activation of the MAGE-1 gene. I m m u n o n o g e n e t i c s 42, 282-290.

Fueyo J, Gomez Manzano C, Yung WK, Liu TJ, Alemany R, McDonnell TJ, Shi X, Rao JS, Levin VA, Kyritsis AP. (1 9 9 8 ) Overexpression of E2F-1 in glioma triggers apoptosis and suppresses tumor growth in vitro and in vivo. Nat Med 4, 685-690.

De Smet C, Lurquin C, Van Der Bruggen P, De Plaen E, Brasseur F, Boon T. (1 9 9 4 ) Sequence and expression pattern of the human MAGE 2 gene. I m m u n o g e n e t i c s 39, 121-129.

Fujiwara T, Grimm EA, Mukhopadhayay T, Zang WW, OwenSchaub LB, Roth JA. (1 9 9 4 ) Induction of chemosensitivity in human lung cancer in vivo by adenoviral mediated tranfer of the wild type p53 gene. Cancer Res 54, 2287-2291.

Dietrich P-Y, Walker PR, Saas P, de Tribolet N. (1 9 9 7 ) Immunobiology of Gliomas, New perspectives for therapy. Ann NY Acad Sci 824, 124-140.

Furnari FB, Su Huang HJ, Cavenee WK. (1 9 9 5 ) Genetics and malignant progression of human brain tumors. Cancer Surv 25, 233-275.

Dubinett SM, Huang M, Dhanani S, Wang J, Beroiza T. (1 9 9 3 ) Down-regulation of macrophage transforming growth factor-beta messenger RNA expression by IL-7. J Immunol 151, 6670-6680.

Gaken JA, Hollingsworth SJ, Hirst WJR, Buggins AGS, Galea J, Peakman M, Kuiper M, Patel P, Towner P, Patel PM, Collins MK, Mufti GJ, Farzaneh F, Darling DC. (1 9 9 7 ) Irradiated NC adenocarcinoma cells transduced with both B7.1 and interleukin-2 induce CD4+-mediated rejection of established tumors. Hum Gene Ther 8, 477-488.

Ekstrand AJ, Longo N, Hamid ML, Olson JJ, Liu L, Collins VP, James CD. (1 9 9 4 ) Functional characterization of an EGF receptor with a truncated extracellular domain expressed in glioblastomas with EGFR gene amplifacation. O n c o g e n e 9, 2313-20.

Ge K, Xu L, Zheng Z, Xu D, Sun L, Liu X. (1 9 9 7 ) Transduction of cytosine deaminase gene makes rat glioma cells highly sensitive to 5-fluorocytosine. Int J Cancer 71, 675-9.

Espevik T, Figari IS, Ranges GE, Palladino MA Jr. (1 9 8 8 ) Transforming growth factor-#1 (TGF-#1) and recombinant human tumor necrosis factor ! reciprocally regulate the generation of lymphokine-activated killer cell activity. Comparison between natural porcine platelet-derived TGF#1 and TGF-#2, and recombinant human TGF-#1. J Immunol 140, 2312-2316.

Gjerset RA, Turla ST, Sobol RE, Scalise JJ, Mercola D, Collins H, Hopkins PJ. (1 9 9 5 ) Use of wild type p53 to achieve complete treatment sensitization of tumor cells expressing endogenous mutant p53. M o l C a r c i n o g 14, 275-285. Gomez-Manzano C, Fueyo J, Kyritsis AP, Steck PA, Roth JA, McDouwell TJ, Steck KD, Levin VA, Yung WK. (1 9 9 6 ) Adenovirus-mediated tranfer of the p53 gene produces rapid and generalized death of human glioma cells via apoptosis. Cancer Res 56, 694-699.

Espevik T, Figari IS, Shalaby MR, Lackides GA, Lewis GD, Shepard HM, Palladino MA Jr. (1 9 8 7 ) Inhibition of cytokine production by cyclosporin A and transforming growth factor . J Exp Med 166, 571-576. Ezzeddine ZD, Martuza RL, Platika D, Short MP, Malick A, Choi B, Breakefield XO. (1 9 9 1 ) Selective killing of glioma cells in colture and in vivo by retrovirus transfer of the herpes simplex virus thymidine kinase gene. New B i o l 3, 608-614.

Griscelli F, Li H, Bennaceur Griscelli A, Soria J, Opolon P, Soria C, Perricaudet M, Yeh P, Lu H. (1 9 9 8 ), Angiostatin gene transfer, inhibition of tumor growth in vivo by blockage of endothelial cell proliferation associated with a mitosis arrest. Proc Natl Acad Sci USA 95, 11, 63676372.

Fakhrai H, Dorigo O, Shawler DL, Lin H, Mercola D, Black KL, Royston I, Sobol RE. (1 9 9 6 ) Eradication of established intracranial rat gliomas by transforming growth factor beta antisense gene therapy. Proc Natl Acad Sci USA 93, 2909-2914.

Hamel W, Magnelli L, Chiarugi VP, Israel MA. (1 9 9 6 ) Herpes simplex virus thymidine kinase/ganciclovir-mediated apoptotic death of bystander cells. Cancer R e s 56, 2697-2702.

Fewer D, Wilson CB, Boldrey EB. (1 9 7 2 ) Phase II study of CCNU in the treatment of brain tumors. Cancer Chem Rep, 56, 421-7.

He J, Allen JR, Collins VP, Allalunis-Turner MJ, Godbout R, Day RS 3rd, James CD. (1 9 9 4 ) CDK4 amplification is an alternative mechanism to p16 gene homozygous deletion in glioma cell lines. Cancer Res 54, 5804-5807.

Fine HA, Dear KBG, Loeffler JS. (1 9 9 3 ) Meta-analysis of radiation therapy with an without adjuvant chemotherapy for malignant glionas in adults. Cancer 71, 2585-97.

144

Gene Therapy and Molecular Biology Vol 1, page 145 He J, Olson JJ, James CD. (1 9 9 5 ) Lack of p16ink4 or retinoblastoma protein (pRb) or amplification-associated overexpression of cdk4 is observed in dinstinct subsets of malignant glial tumors and cell lines. C a n c e r R e s 55, 4833-4836.

Kerhl JH, Wakefield LM, Roberts AB, Jakowlew S, AlvarezMon M, Derynck R, Sporn MB, Fauci AS. (1 9 8 6 ) Production of transforming growth factor ( by human T lymphocytes and its potential role in the regulation of T cell growth. J Exp Med 163, 1037-1050.

Hendrzak JA and Brunda MJ. (1 9 9 6 ) Interleukin-12, biologic activity, therapeutic utility, and role in disease. Lab I n v e s t 72, 619-637.

Kim JH, Kim SH, Brown SL, Freitag SO. (1 9 9 4 ) Selective enhancement by an antiviral agent of the radiation induced cell killing of human glioma cells transduced with HSV-TK gene. Cancer Res 54, 6053-6056.

Henson JW, SchniTKer BL, Correa KM, von Dimling A, Fassbender F, Xu HJ, Benedict WF, Yandell DW, Louis DN. (1 9 9 4 ) The retinoblastoma gene is involved in malignat progression of astrocytomas. Ann Neurol 36, 714-721.

Kim JH, Kim SH, Kolozsvary A,Brown SL, Kim OB, Freytag SO. ( 1 9 9 5 ) Selective enhancement of radiation response of herpes simplex virus thymidine kinase transduced 9L gliosarcoma cells in vitro and in vivo by antiviral agents. I n t J R a d i a t O n c o l B i o l P h y s 33, 861-868.

Hochberg FH, Pruitt A. (1 9 8 0 ) Assumptions in the radiotherapy of glioblastoma. N e u r o l o g y 30, 907-911.

Kim SH, Kim JH, Kolozsvary A, Brown SL, Freytag SO. (1 9 9 7 ) Preferential radiosensitization of 9L glioma cells transduced with HSV-TK gene by acyclovir. J N e u r o o n c o l 33, 189-194.

Hoostraten B, Gottlieb JA, Caoili A. 1 9 7 3 CCNU in the treatment of cancer, A phase II study. Cancer 32, 38-43. Hosli P, Sappino AP, de Tribolet N, Dietrich PY. (1 9 9 8 ) Malignant glioma, Should chemotherapy be overthrown by experimental treatments? A n n O n c o l 9, 589-600.

Kondo S, Barna BP, Kondo Y, Tanaka Y, Casey G, Liu J, Morimura T, Kaakaji R, Peterson JW, Werbel B, Barnett GH. (1 9 9 6 ) WAF1/CIP1 increases the susceptibility of p53 non-functional malignant glioma cells to cisplatininduced apoptosis. O n c o g e n e 13, 1279-85.

Houbiers JG, Nijman HW, Van Der Burg SH, Drijfhout JW, Kenemans P, Van Der Velde CJ, Brand A, Momburg F, Kast WM, Melief CJ. (1 9 9 3 ) In vitro induction of human cytotoxic T lymphocyte responses against peptides of mutant and wild-type p53. Eur J Immunol 23, 20722077.

Kondo S, Kondo Y, Li G, Silverman RH, Cowell JK. (1 9 9 8 ) Targeted therapy of human malignant glioma in a mouse model by 2-5A antisense directed against telomerase RNA. O n c o g e n e , 16, 25, 3323-30.

Huber D, Philipp J, Fontana A. (1 9 9 2 ) Protease inhibitors interfere with the transforming growth factor-(-indipendent pathway of tumor cell-mediated immunosuppression. J Immunol 148, 277-284.

Kundig TM, Shahinian K, Kawai K, Mittruecker HW, Sebzda E, Bachmann MF, Mak TW, Ohashi PS. (1 9 9 6 ) Duration of TCR stimulation determines costimulatory requirement of T cells. Immunity, 5, 41-52

Ichimura K, Schmidt EE, Miyakawa A, Goike HM, Collins VP. (1 9 9 8 ) Distinct patterns of deletion on 10p and 10q suggest involvement of multiple tumor suppressor genes in the development of astrocytic gliomas of different malignancy grades. G e n e s C h r o m o s o m e s C a n c e r 22, 9-15.

Kuppner MC, Hamou M-F, Sawamura Y, Bodmer S, de Tribolet N. (1 9 8 9 ) Inhibition of lymphocyte function by glioblastoma-derived transforming growth factor #2. J Neurosurg 71, 211-217

Janzer RC, Raff MC. (1 9 8 7 ) Astrocytes induce blood brain barrier properties in e每dotelial cells. Nature 325, 253257.

Lachmann RH, Efstathiou S. (1 9 9 7 ) The use of herpes simplex virus-based vectors for gene delivery to the nervous system. Mol Med Today 3, 404-411.

Jen J, Harper JW, Bigner SH, Bigner DD, Papadopoulos N, Markowitz S, Willson JK, Kinzler KW, Vogelstein B. (1 9 9 4 ) Deletion of p16 and p15 genes in brain tumors. Cancer Res 54, 6353-6358.

Lang FF, Yung WK, Raju U, Libunao F, Terry NH, Tofilon PJ. (1 9 9 8 ) Enhancement of radiosensitivity of wild-type p53 human glioma cells by adenovirus-mediated delivery of the p53 gene. J Neurosurg 89, 1, 125-132.

Jin B, Scott JL, Vadas MA, Burns GF. (1 9 8 9 ) TGF-# down regulates TLiSA1 expression and ihibits the differentiation of precursor lymphocytes into CTL and LAK cells. I m m u n o l o g y 66, 570-576

LeMay DR, Kittaka M, Gordon EM, Gray B, Stins MF, McComb JG, Jovanovic S, Tabrizi P, Weiss MH, Bartus R, Anderson WF, Zlokovic BV. (1 9 9 8 ) Intravenous RMP-7 increases delivery of ganciclovir into rat brain tumors and enhances the effects of herpes simplex virus thymidine kinase gene therapy. Hum Gene Ther 9, 989-995.

June CH, Bluestone JA, Nadler LM, Thompson CB. (1 9 9 4 ) The B7 and CD28 receptor families. Immunol Today 15, 321-331

Levin VA, Edwards MS, Wright DC. (1 9 8 0 ) Modified procarbazine, CCNU and vincristine (PCV3) combination chomtherapy in the treatment of malignant brain tumors. Cancer Treat Rep 64, 237-241.

Ke LD, Fueyo J, Chen X, Steck PA, Shi YX, Im SA, Yung WK. (1 9 9 8 ) A novel approach to glioma gene therapy, downregulation of the vascular endothelial growth factor in glioma cells using ribozymes. Int J Oncol 12, 1391-6.

Levin VA, Gutin PH, Leibel S. Cancer, Principles and Practice of Oncology. In De Vita VT Jr, Hellman S, Rosemberg SA

145

Pal첫 et al: Perspectives for therapy of glioblastoma multiforme (eds) (1 9 9 3 ) , N e o p l a s m s o f t h e C e n t r a l N e r v o u s S y s t e m , Philadelphia, JB Lippincot, 1679-737.

metastases and meningiomas suggests specific transcription patterns. Eur J Cancer 29A, 2118-2125.

Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J, Miliaresis C, Rodgers L, McCombie R, Bigner SH, Giovanella BC, Ittmann M, Tycko B, Hibshoosh H, Wigler MH, Parsons R. (1 9 9 7 ) PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. S c i e n c e , 275, 1943-1947.

Millauer B, Longhi MP, Plate KH, Shawer LK, Risau W, ulrich A, Strawn LM. (1 9 9 6 ) Dominant-negative inhibition of Flk-1 suppress the growth of many tumor types in vivo. Cancer Res 56, 1615-1620. Millauer B, Shawer LK, Plate KH, Risau W, Ulrich A. (1 9 9 4 ) Glioblastoma growth inhibited in vivo by a dominantnegative Flk-1 mutant. Nature, 367, 576-579.

Libermann TA, Nusbaum HR, Razon N, Kris R, Lax I, Soreq H, Whittle N, Waterfield MD, Ullrich A, Schlessinger J. (1 9 8 5 ) Amplification, enhanced expression and possible rearrangement of EGF receptor gene in primary human brain tumors of glial origin. Nature 313, 144-147.

Mineta T, Rabkin SD, Yazaki T, Hunter WD, Martuza RL. (1 9 9 5 ) Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. N a t M e d , 1 , 9 , 938-943.

Long Z, Li LP, Grooms T, Lockey C, Nader K, Mychkovsky I, Mueller S, Burimski I, Ryan P, Kikuchi G, Ennist D, Marcus S, Otto E, McGarrity G. (1 9 9 8 ) Biosafety monitoring of patients receiving intracerebral injections of murine retroviral vector producer cells. Hum Gene Ther 9, 1165-1172.

Miyatake S, Martuza RL, Rabkin SD. (1 9 9 7 ) Defective herpes simplex virus vectors expressing thymidine kinase for the treatment of malignant glioma. Cancer Gene Ther 4, 222-228. Mizuno M, Yoshida J, Colosi P, Kurtzman G. (1 9 9 8 ) Adenoassociated virus vector containing the herpes simplex virus thymidine kinase gene causes complete regression of intracerebrally implanted human gliomas in mice, in conjunction with ganciclovir administration. Jpn J Cancer Res, 89, 1, 76-80.

Lowe SW, Bodis S, McClatchey A, Remington L, Ruley HE, Fisher DE, Housman DE, Jacks T. (1 9 9 4 ) p53 status and the efficacy of cancer therapy in vivo. S c i e n c e 266, 807810. Mabuchi E, Shimizu K, Miyao Y, Kaneda Y, Kishima H, Tamura M, Ikenaka K, Hayakawa T. (1 9 9 7 ) Gene delivery by HVJ-liposome in the experimental gene therapy of murine glioma. Gene Ther 4, 768-72.

Newlands ES, O'Reilly SM, Glaser MG. (1 9 9 6 ) The Charing Cross hospital experience with Temozolomide in patients with gliomas. Eur J Cancer, 32A, 2236-41. Nguyen DM, Spitz FR, Yen N, Cristiano RJ, Roth JA. (1 9 9 6 ) Gene therapy for lung cancer, enhancement of tumor suppression by a combination of sequential systemic cisplatin and adenovirus-mediated p53 gene transfer. J Torac Cardiovasc Surg, 112, 1372-1377.

Mak M, Fung L, Strasser JF, Saltzman WM. (1 9 9 5 ) Distribution of drugs following controlled delivery to the brain interstitium. J N eu r o o n c o l 26, 91-102. Manome Y, Kunieda T, Wen PY, Koga T, Kufe DW, Ohno T. (1 9 9 8 ) Transgene expression in malignant glioma using a replication-defective adenoviral vector containing the Egr-1 promoter, activation by ionizing radiation or uptake of radioactive iododeoxyuridine. Hum Gene Ther 9, 1409-1417.

Nielsen LL, Maneval DC. (1 9 9 8 ) p53 tumor suppressor gene therapy for cancer. Cancer Gene Ther 5, 52-63. Nishikawa R, Furnari FB, Lin H, Arap W, Berger MS, Cavenee WK, Su Huang HJ. (1 9 9 5 ) Loss of p16ink4 expression is frequent in high-grade gliomas. C a n c e r R e s 55, 19411945.

Manome Y, Wen PY, Chen L, Tanaka T, Dong Y, Yamazoe M, Hirshowitz A, Kufe DW, Fine HA. (1 9 9 6 ) Gene therapy for malignant gliomas using replication incompetent retroviral and adenoviral vectors encoding the cytochrome P450 2B1 gene together with cyclophosphamide. Gene Ther 3, 513-520.

Nishikawa R, Ji XD, Harmon RC, Armon RC, Lazar CS, Gill GN, Cavenee WK, Huang HG. (1 9 9 4 ) A mutant epidermal growth factor receptor common in human gliomas confers enhanced tumorigenicity. P r o c Natl Acad S c i USA 91, 7727-7731.

Manome Y, Wen PY, Dong Y, Tanaka T, Mitchell BS, Kufe DW, Fine HA. (1 9 9 6 ) Viral vector transduction of the human deoxycytidine kinase cDNA sensitizes glioma cells to the cytotoxic effects of cytosine arabinoside in vitro and in vivo. Nat Med 2, 567-573.

Nitta T, Hishii M, Sato K, Okumura K. (1 9 9 4 ) Selective expression of interleukin-10 gene within glioblastoma multiforme. Brain Res, 649, 122-128. Noguchi Y, Chen YT, Old LJ. (1 9 9 4 ) A mouse mutant p53 product recognized by CD4+ and CD8+ T cells. P r o c N a t l Acad Sci 91, 3171-3175.

McKie EA, MacLean AR, Lewis AD, Cruickshank G, Rampling R, Barnett SC, Kennedy PG, Brown SM. (1 9 9 6 ) Selective in vitro replication of herpes simplex virus type 1 (HSV-1) ICP34.5 null mutants in primary human CNS tumours-evaluation of a potentially effective clinical therapy. B r J Cancer 74, 745-752.

Ohnishi T, Taki T, Hiraga S, Arita N, Morita T. (1 9 9 8 ) In vitro and in vivo potentiation of radiosensitivity of malignant gliomas by antisense inhibition of the RAD51 gene. B i o c h e m B i o p h y s R e s Commun 245, 319324.

Merlo A, Juretic A, Zuber M, Filgueira L, Luscher U, Caetano V, Ulrich J, Gratzl O, Heberer M, Spagnoli G. (1 9 9 3 ) Cytokine gene expression in primary brain tumors,

Oldfield EH, Ram Z, Culver KW, Blaese RM, De Vroom HL, Anderson WF. (1 9 9 3 ) Clinical Protocol, Gene therapy for

146

Gene Therapy and Molecular Biology Vol 1, page 147 the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase and intravenous ganciclovir. Hum Gene Ther 4, 39-69.

brain tumors by intratumoral implantation of retroviral vector-producing cells. Nature Med 3, 1354-1361 Ram Z, Culver KW, Wallbridge S, Blaese RM, Oldfield EH. (1 9 9 3 ) In situ retroviral-mediated gene transfer for the treatment of brain tumors in rats. Cancer Res 53, 83-88.

Ono Y, Ikeda K, Wei MX, Harsh GR 4th, Tamiya T, Chiocca EA. (1 9 9 7 ) Regression of experimental brain tumors with 6-thioxanthine and Escherichia coli gpt gene therapy. Hum Gene Ther 8, 2043-2055.

Ranges GE, Figari IS, Espevik T, Palladino MA Jr. (1 9 8 7 ) Inhibition of cytotoxic T cell development by transforming growth factor # and reversal by recombinant tumor necrosis factor !. J Exp Med 166, 991-998.

O'Rourke DM, Kao GD, Singh N, Park BW, Muschel RJ, Wu CJ, Greene MI. (1 9 9 8 ) Conversion of a radioresistant phenotype to a more sensitive one by disabling erbB receptor signaling in human cancer cells. Proc Natl Acad Sci USA 95, 10842-10847.

Redekop GJ, Naus CC. (1 9 9 5 ) Transfection with bFGF sense and antisense cDNA resulting in modification of malignant glioma growth. J Neurosurg 82, 83-90.

O'Rourke DM, Qian X, Zhang HT, Davis JG, Nute E, Meinkoth J, Greene MI. (1 9 9 7 ) Trans receptor inhibition of human glioblastoma cells by erbB family ectodomains. Proc Natl Acad Sci USA 94, 3250-3255.

Rimoldi D, Romero P, Carrel S. (1 9 9 3 ) The human melanoma antigen/encoding gene, MAGE-1, is expressed by other tumor cells of neuroectodermal origin such as glioblastomas and neuroblastomas. Int J Cancer 54, 527-528.

Pal첫 G, Cavaggioni A, Calvi P, Franchin E, Pizzato M, Boschetto R, Parolin C, Chilosi M, Ferrini S, Zanusso A, Colombo F. (1 9 9 8 ) Gene therapy of glioblastoma multiforme via combined expression of suicide and cytokine genes, a pilot study in humans. Gene Ther (in press)

Rodriguez LA, Prados M, Silver P. (1 9 8 9 ) Reevaluation of procarbazione for the treatmentof recurrent malignant central nervous system tumors. Cancer 64, 2420-2423. Rogulski KR, Kim JH, Kim SH, Freytag SO. (1 9 9 7 ) Glioma cells transduced with an Escherichia coli CD/HSV-1 TK fusion gene exhibit enhanced metabolic suicide and radiosensitivity. Hum Gene Ther 8, 73-85.

Parr MJ, Manome Y, Tanaka T, Wen P, Kufe D, Kaelin WG Jr, Fine HA. (1 9 9 7 ) Tumor-selective transgene expression in vivo mediated by an E2F-responsive adenoviral vector. Nature Med, 3, 1145-9.

Rook AH, Kerhl JH, Wakefield LM, Roberts AB, Sporn MB, Burlington DB, Lane HC, Fauci AS. (1 9 8 6 ) Effects of transforming growth factor ( on the functions of natural killer cells, depressed cytolytic activity and blunting of interferon responsiveness. J Immunol 136, 3916-3920.

Petersdorf SH, Berger MS. (1 9 9 6 ) Concepts in Neurosurgery, The molecular basis of neurosurgical disease. In Raffel C, Harsh IV GR (eds), Molecular Basis of Chemotherapy for Brain Tumors, Baltimore, Williams and Wilkins, chapter 12 (Vol 8), 198-210.

Rosemberg SA, Lotze MT, Muul LM, Chang AE, Avis FP, Leitman S, Linehan WM, Robertson CN, Lee RE, Rubin JT, Seipp CA, Simpson CG, White DE. (1 9 8 7 ) A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and interleukin-2 or high dose of interleukin-2 alone. New Engl J Med, 316, 889-905.

Pizzato M, Franchin E, Calvi P, Boschetto R, Ferrini S, Colombo M, Pal첫 G. (1 9 9 8 ) Production and characterization of a bicistronic Moloney-based retroviral vector expressing human interleukin 2 and herpes simplex virus thymidine kinase for gene therapy of cancer. Gene Ther 5, 1003-1007. Prados MD, Schold SC, Spence AM, Berger MS, Mc Allister LD, Mehta MP, Gilbert MR, Fulton D, Kuhn J, Lamborn K, Rector DJ, Chang SM. (1 9 9 6 ) Phase II study of paclitaxel in patients with recurrent malignant glioma. J C l i n O n c o l 14, 2316-2321.

Roszman T, Elliott L, Brooks W. (1 9 9 1 ) Modulation of Tcell function by gliomas. Immunol Today 12, 370-374.

Rainov NG, Dobberstein KU, Bahn H, Holzhausen HJ, Lautenschl채ger C, Heidecke V, Burkert W. (1 9 9 7 ) Prognostic factors in malignant glioma, influence of the overexpression of oncogene and tumor-suppressor gene products on survival. J N eu r o o n c o l 35, 13-28.

Russell DS, Rubistein LJ. (1 9 8 9 ) P a t h o l o g y o f T u m o r s o f t h e C e n t r a l N e r v o u s S y s t e m, 5th Edition, Edward Arnold, London.

Roth JA, Cristiano R. (1 9 9 7 ) Gene therapy for cancer, What we done and where we are going? J N a t l C a n c e r I n s t , 89, 21-39.

Saleh M, Stacker SA, Wilks AF. (1 9 9 6 ) Inhibition of growth of C6 glioma cells in vivo by expression of antisense vascular endothelial growth factor sequence. Cancer Res 56, 393-401.

Rainov NG, Dobberstein KU, Sena-Esteves M, Herrlinger U, Kramm CM, Philpot RM, Hilton J, Chiocca EA, Breakefield XO. (1 9 9 8 ) New prodrug activation gene therapy for cancer using cytochrome P450 4B1 and 2aminoanthracene/4-ipomeanol. Hum Gene Ther 9, 1261-73.

Sampson JH, Archer GE, Ashley DM, Fuchs HE, Hale LP, Dranoff G, Bigner DD. (1 9 9 6 ) Subcutaneous vaccination with irradiated, cytokine-producing tumor cells stimulates CD8+ cell mediated immunity against tumors located in the immunologically privileged central nervous system. Proc Natl Acad Sci USA 93, 10399-10404.

Ram Z, Culver KW, Oshiro EM, Viola JJ, De Vroom HL, Otto E, Long Z, Chiang Y, McGarrity GJ, Muul LM, Katz D, Blaese RM, Oldfield EH. (1 9 9 7 ) Therapy of malignant

147

Palù et al: Perspectives for therapy of glioblastoma multiforme Sanson M, Ameri A, Monjour A, Sahmoud T, Ronchin P, Poisson M, Delattre JY. (1 9 9 6 ) Treatment of recurrent malignant supratentorial gliomas with ifosfamide, carboplatin, and etoposide, A phase II study. Eur J Cancer 32A, 2229-2235.

Tepper RI. (1 9 9 3 ) The antitumor and proinflammatory actions of IL-4. Res Immunol 144, 633-637. Tohma Y, Gratas C, Biernat W, Peraud A, Fukuda M, Yonekawa Y, Kleihues P, Ohgaki H. (1 9 9 8 ) PTEN (MMAC1) mutations are frequent in primary glioblastomas (de novo) but not in secondary glioblastomas. J Neuropathol Exp Neurol 57, 684-689.

Smart CR, Ottomoan RE, Rochlin DB. (1 9 6 8 ) Clinical experience with vincristine with tumors of the central nervous system and other malignant diseases. Cancer Chem Rep 52, 733-741.

Trojan J, Johnson TR, Rudin SD, Ilan J, Tykocinski ML, Ilan J. (1 9 9 3 ) Treatment and prevention of rat glioblastoma by immunogenic C6 cells expressing antisense insulinlike growth factor I RNA. S c i e n c e 259, 94-97.

Sporn MB, Roberts AB, Wakefield LM, Assoian RK. (1 9 8 6 ) Transforming growth factor-(, biological fusion and chemical structure. S c i e n c e 233, 532-534

Van Der Bruggen P, Traversari C, Chomez P, Lurquin C, De Plaen E, Van Den Eynde B, Knuth A, Boon T. (1 9 9 1 ) A gene encoding an antigen recognized by cytolityc T lymphocytes on a human melanoma. S c i e n c e 254, 16431647

Sporn MB, Roberts AB, Wakefield LM, de Crombrugghe B. (1 9 8 7 ) Some recent advances in the chemistry and biology of transforming growth factor-beta. J C e l l B i o l 105, 1039-1045. Staba MJ, Mauceri HJ, Kufe DW, Hallahan DE, Weichselbaum RR. (1 9 9 8 ) Adenoviral TNF-alpha gene therapy and radiation damage tumor vasculature in a human malignant glioma xenograft. Gene Ther 5, 293-300.

Van Meir E, Roemer K, Diserens AC, Kikuchi T, Rempel SA, Haas M, Huang HG, Friedmann T, de Tribolet N, Cavenee WK. (1 9 9 5 ) Single cell monitoring of growth arrest and morphological changes induced by transfer of wild-type p53 alleles to glioblastoma cells. P r o c N a t l A c a d S c i USA 92, 1008-1012.

Sturtz FG, Waddell K, Shulok J, Chen X, Caruso M, Sanson M, Snodgrass HR, Platika D. (1 9 9 7 ) Variable efficiency of the thymidine kinase/ganciclovir system in human glioblastoma cell lines, implications for gene therapy. Hum Gene Ther 8, 1945-53.

Vincent AJ, Vogels R, Someren GV, Esandi MC, Noteboom JL, Avezaat CJ, Vecht C, Bekkum DW, Valerio D, Bout A, Hoogerbrugge PM. (1 9 9 6 ) Herpes simplex virus thymidine kinase gene therapy for rat malignant brain tumors. Hum Gene Ther 7, 197-205.

Tada M, Diserens AC, Hamou MF, Jaufeerally R, Van Meir E, de Tribolet N. (1 9 9 6 ) B r a i n T u m o r R e s T h e r , 327337.

Voest EE, Kenyon BM, O’Reilly MS, Truitt G, D’Amato RJ, Folkman J. (1 9 9 5 ) Inhibition ofangiogenesis in vivo by interleukin-12. J Natl Cancer Inst 87, 581-586.

Takamiya Y, Short MP, Ezzeddine ZD, Moolten FL, Breakefield XO, Martuza RL. (1 9 9 2 ) Gene therapy of malignant brain tumors, a rat glioma line bearing the herpes simplex virus type 1-thymidine kinase gene and wild type retrovirus kills other tumor cells. J N e u r o s c R e s 33, 493-503.

Walker MD, Green SB, Byar DP. (1 9 8 0 ) Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N E n g l J Med 303, 1323-1329.

Takamiya Y, Short MP, Moolten FL, Fleet C, Mineta T, Breakefield XO, Martuza RL. (1 9 9 3 ) An experimental model of retrovirus gene therapy for malignant brain tumors. J Neurosurg 79, 104-110.

Wei M, Tamiya T, Chase M, Bobiatsis EJ, Chang TK, Kowall NW, Hochberg FH, Waxman DJ, Breakefield XO, Chiocca EA. (1 9 9 4 ) Experimental tumor therapy in mice using the cyclophosphamide-activating cytochrome P450 2B1 gene. Hum Gene Ther 5, 969-978.

Tamiya T, Ono Y, Wei MX, Mroz PJ, Moolten FL, Chiocca EA. (1 9 9 6 ) Escherichia coli gpt gene sensitizes rat glioma cells to killing by 6-thioxanthine or 6thioguanine. Cancer Gene Ther 3, 3, 155-162.

Weller M, Malipiero U, Rensing Ehl A, Barr PJ, Fontana A. (1 9 9 5 ) Fas/APO-1 gene transfer for human malignant glioma. Cancer Res 55, 2936-2944.

Tamura M, Shimizu K, Yamada M, Miyao Y, Hayakawa T, Ikenaka K. (1 9 9 7 ) Targeted killing of migrating glioma cells by injection of HTK-modified glioma cells. Hum Gene Ther 8, 381-91.

Young RC, Walker MD, Canellos GP. 1 9 7 3 Initial clinical trials with methyl-CCNU 1-(2-chloroethyl)-3-(4-methyl cycloexyl)-I-nitrosourea (MeCCNU). Cancer 31, 11641169.

Tanaka T, Cao Y, Folkman J, Fine HA. (1 9 9 8 ) Viral vectortargeted antiangiogenic gene therapy utilizing an angiostatin complementary DNA. Cancer Res 58, 33623369.

Yu JS, Wei MX, Chiocca EA, Martuza RL, Tepper RI. (1 9 9 3 ) Treatment of glioma by engineered interleukin 4-secreting cells. Cancer Res 53, 3125-3128. Zerrouqi A, Rixe O, Ghoumari AM, Yarovoi SV, Mouawad R, Khayat D, Soubrane C. (1 9 9 6 ) Liposomal delivery of the herpes simplex virus thymidine kinase gene in glioma, improvement of cell sensitization to ganciclovir. Cancer Gene Ther 3, 385-392.

Tanaka T, Manome Y, Wen P, Kufe DW, Fine HA. (1 9 9 7 ) Viral vector-mediated transduction of a modified platelet factor 4 cDNA inhibits angiogenesis and tumor growth. Nat Med 3, 437-442.

148

Gene Therapy and Molecular Biology Vol 3, page 149 Gene Ther Mol Biol Vol 3, 149-155. August 1999.

Gene-based vaccine strategies against cancer Review Article

Daniel Lee1, Ken Wang 1, Liesl K. Nottingham2, Jim Oh1, David B. Weiner1, and Jong J. Kim1 1

Department of Pathology and Laboratory Medicine; 2Department of Otolaryngology/Head and Neck Surgery

University of Pennsylvania, Philadelphia, PA 19104 __________________________________________________________________________________________________ Corresponding Author: Jong J. Kim, Ph.D., Department of Pathology and Laboratory Medicine, University of Pennsylvania, 505 Stellar-Chance, 422 Curie Blvd., Philadelphia, PA 19104. Tel: (215) 662-2352; Fax: (215) 573-9436; E-mail: jonger@seas.upenn.edu Received: 30 September 1998; accepted: 7 October 1998

Summary In recent years, the characterization of gene-based cancer vaccines has been an important step in the development of different treatment options for human carcinoma. These particular vaccines m a k e u s e o f p r o t e i n s t h a t a r e s p e c i f i c a l l y p r o d u c e d a t v e r y h i g h l e v e l s by tumor c e l l s . These tumor-associated antigens (TAAs) are not o n l y used i n diagnostic situations, but also i n the development of cancer vaccines. In this review we will focus on two well characterized TAAs, carcinoembryonic antigen (CEA) and prostate specific antigen (PSA). The two methods of i n v i v o delivery we will examine are recombinant vaccinia virus and nucleic acid immunization. The TAA g e n e c a n b e c l o n e d i n t o v a c c i n i a v i r u s a n d the viral infection stimulates an adequate immune response in the host. In the case of nucleic acid immunization, DNA constructs encoding for TAAs are directly injected into the host and are taken up by its cells. The cells express the specific encoded antigen upon which the immune system acts. The effects o f CEA recombinant vaccinia virus (rV-CEA) have been characterized i n rodents, macaques, and humans. It was shown that the vaccine induced both humoral and cellular immune responses in mice and monkey models. In a phase I clinical trial, a CEA-specific cytotoxic Tlymphocyte response was observed. The effects of a CEA DNA vaccine were investigated in both mice and dogs and both humoral and cellular immune responses were found as well. A recombinant vaccinia virus expressing PSA was tested in rhesus monkeys and induced a PSA-specific long term cellular immune response. Experiments were also performed injecting a PSA DNA construct into both mice and rhesus monkeys. PSA-specific humoral and cellular immune responses were observed in both cases. All these experimental approaches demonstrate the efficacy and advantages of genebased cancer vaccine strategies and support further clinical investigations.

Thus, researchers are continually investigating novel and more effective treatment strategies for various forms of cancer. Research, in recent years, has turned toward the use of vaccines to treat cancer. To this end, several proteins produced by tumor cells became a target for vaccine development. These tumor-associated antigens are predominantly expressed in a tissuespecific manner and are expressed at greatly increased levels in affected cells. Besides being important

I. Introduction Although advances in science have led to countless theories and methods designed to combat human carcinoma, the battle is far from being over. Surgical excision of tumors, drug therapies, and chemotherapy have been effective in certain cases but in other situations, particularly when the tumor has begun to metastasize, effective treatment is far more difficult and far less potent.

149

Lee et al: Gene-based vaccine strategies against cancer diagnostic aids, these antigens represent appropriate targets for the development of cancer vaccines (Sogn et al, 1993).

(Kaufman et al, 1991). Among its many advantages is that it greatly enhances the immune response when coupled with a weak immunogen such as a TAA. Through recombinant DNA technology, TAA genes can be cloned into the vaccinia viral vector and this recombinant vaccinia virus can be used to stimulate an effective immune response. Another advantage is that it can infect professional antigen presenting cells (APCs), such as dendritic cells or macrophages, and express the antigen along with MHC class I and/or class II complexes (Tsang et al, 1995). Finally, the stability and efficiency of vaccinia allows it to successfully incorporate fairly large inserts, which is advantageous in the context of cloning the genes for different TAAs (Kaufman et al, 1991). Potential disadvantages are toxicity effects, immunogenecity to the virus, and risk of viral reversion. Moreover, recombinant vaccinia viruses cannot be used to target specific cells.

Tumor-associated antigens (TAA) are proteins produced by tumor cells which can be presented on the cell surface in the context of major histocompatibility complexes (Kelley and Cole, 1998). Recently, these antigens have been the focus of study as a viable option for immunotherapy of various types of cancer. In this review we will examine the progress in the investigation of the immunological effects of two such TAAs, carcinoembryonic antigen (CEA) and prostate specific antigen (PSA).

II. Background The use of therapeutic cancer vaccines has several distinct advantages. The immune response can be directed against carcinomas with a high degree of specificity. They can also generate immunological memory, for continued protection. The immune response induced by the vaccine can be modified or enhanced with other forms of immunotherapy such as using cytokines and other cellular therapies (Jones and Mitchell, 1996). Gene-based cancer vaccine strategies have yielded promising results, and several different methods of in vivo delivery are currently being explored (Roth and Cristiano, 1997). Two such approaches are recombinant vaccinia virus and nucleic acid, or DNA immunization (Table 1).

DNA vaccination is a relatively new approach towards disease prophylaxis and/or treatment. DNA expression cassettes introduced in vivo can be taken up and expressed by host cells, leading to the production of specific foreign proteins. The presence of these foreign proteins can then elicit specific humoral and cellular immune responses against the foreign antigens (Wolff et al, 1990; Tang et al, 1992; Wang et al, 1993; Ulmer et al, 1993). This technique can be applied more widely than delivery through a recombinant vaccinia virus because there is no limitation on the size and type of nucleic acid used (Roth and Cristiano 1997). DNA vaccines are non-replicating, thereby minimizing the risk of any primary infections. It is also possible to alter or delete undesirable genes, such as those which may inhibit the immune response. More recently, the

Vaccinia virus is one of the most heavily investigated viral delivery vehicles; it is a type of pox virus which was used in the successful eradication of smallpox (Kantor et al, 1992a). It is extremely immunogenic and is capable of stimulating both humoral and cellular immune responses

Vaccinia

DNA

Advantages

Disadvantages

-highly immunogenic -infects APCs -induces both humoral and cellular responses -large insert size

-toxicity -risk of viral reversion -no targeting -induces vaccinia specific immune response

-possible to specifically target cells -low immunogenicity -no limit on size and type of nucleic acid -difficult to incorporate into cells in vivo -induces both humoral and cellular responses -non-replicating -able to genetically alter and enhance -use of molecular adjuvants to modulate response -repeated use without decrease in effect

T a b l e 1 . Comparison of recombinant vaccinia virus and nucleic acid immunization as in vivo delivery vehicles for gene-based cancer vaccine therapy.

use of molecular adjuvants such as cytokines and costimulatory molecules has proven to be effective in modulating and directing the desired immune responses 150

Gene Therapy and Molecular Biology Vol 3, page 151 (Kim et al, 1998). Nucleic acid immunization is promising in the development of vaccinations for a wide array of pathogens, including cancer (Kim et al, In Press). Using DNA expression cassettes, DNA sequences that encode certain cancer proteins, such as those found in colon cancer or prostate cancer, are introduced into host cells. These cells then synthesize the antigenic cancer proteins which can then elicit an immune response against those proteins. The first clinical studies for DNA vaccines tested the effects of the HIV-1 env/rev DNA vaccine in HIV-infected patients (MacGregor et al, 1998). Each patient in the trial received three injections each separated by ten weeks with increasing dosage (3 dosage groups of 5 subjects) of envelope vaccine. The clinical results reveal no significant clinical or laboratory adverse effects measured in all three dosage groups (30, 100, 300 Âľg). The immunized individuals developed increased antibody responses to envelope proteins and peptides after receiving the 100 Âľg dose of env/rev. Some increased cellular responses were also observed. These preliminary results demonstrate that the injection of even relatively low doses of a single immunogen DNA vaccine can augment both existing humoral and cellular immune responses in humans in a safe and tolerant manner.

III. Gene-based cancer vaccine strategies using CEA Human CEA is a 180-kDa glycoprotein expressed in elevated levels in 90% of gastrointestinal malignancies, including colon, rectal, stomach, and pancreatic tumors, 70% of lung cancers, and 50% of breast cancers (Zaremba et al, 1997, Kelley and Cole, 1998). CEA is also found in human fetal digestive organ tissue, hence the name carcinoembryonic antigen (Foon et al, 1995). It has been discovered that CEA is expressed in normal adult colon epithelium as well, albeit at far lower levels (Conry et al, 1996a). Sequencing of CEA shows that it is associated with the human immunoglobulin gene superfamily and that it may be involved in the metastasizing of tumor cells (Foon et al, 1995).

immunization were predominantly mediated by CEAspecific CD8+ T-cell response (Abrams et al, 1997). Splenocytes from rV-CEA immunized C57BL/6 mice were adoptively transferred to syngeneic immune deficient, tumor-bearing mice. They exhibited strong anti-tumor activity compared to splenocytes transferred from nonimmunized mice. Adoptive transfer of CD4+, but not CD8 + T cells did not show anti-tumor activity. However, transfer of CD8+, but not CD4 + T cells still showed some antitumor response, although this response was less compared to when both CD8+ and CD4+ cell populations are present. CD4+ cells therefore may play an important helper or regulatory role in anti-tumor responses. Immunization of mice with rV-CEA induced anti-tumor activity that was mediated mainly by CD8+ cells, but both CD8+ and CD4+ cells were necessary to acheive optimal anti-tumor responses (Abrams et al, 1997). The effects of rV-CEA vaccination were further characterized in experimental trials with non-human primates. After injection, the rhesus macaques of the experimental group showed both humoral and cellular immune responses to CEA. The immunization also resulted in toxic effects such as mild fever, irritation of the skin near the injection point, and lymphadenopathy (Kantor et al, 1992b). The results of this experiment along with the results from various rodent experiments demonstrated potential utility and limitations of the rVCEA vaccine. Additional information in this regard has been provided in the clinical setting. Tsang, et al. in conjunction with the National Cancer Institute, recently conducted a phase I clinical trial testing the effects of rV-CEA in 26 patients with advanced metastatic carcinoma (Tsang et al, 1995). Peripheral blood lymphocytes (PBLs) were taken from patients both before and after vaccination and analyzed for their response to specific CEA peptides with human leukocyte antigen (HLA) class I-A2 motifs. It was observed that CEA-specific MHC class I restricted cytotoxic T-lymphocyte response could be elicited (Tsang et al, 1995). However, following the first vaccination, there was an anti-vaccinia immune response which suppressed the effects of subsequent vaccinations (Kelley and Cole, 1998).

A. CEA recombinant vaccinia virus vaccine Recombinant vaccinia virus expressing the human CEA gene (rV-CEA) has been investigated as a potential therapy for colon and other gastrointestinal carcinomas. A number of groups have shown that immunization of these constructs into rodents induced both cellular and humoral responses. More importantly, immunization with rV-CEA led to antigen-specific inhibition of tumor growth in mice. Using an adaptive transfer experiment, Abrams, et. al. found that anti-tumor responses after rV-CEA

151

B. CEA DNA vaccine The immune response to nucleic acid vaccination using a CEA DNA construct was characterized in a murine model. The CEA insert was cloned into a vector containing the cytomegalovirus (CMV) early promoter/enhancer and injected intramuscularly. CEA spe-

Gene Therapy and Molecular Biology Vol 3, page 152

Humoral response

Cellular response

rV-PSA

+

+

PSA DNA

+

+

T a b l e 2 . Induction of PSA-specific immune responses in rhesus macaques.

human PSA (rV-PSA) were studied in rodent as well as in non-human primate models (Hodge et al, 1995). Hodge, et al. investigated the immunological effects of a recombinant vaccinia virus expressing human PSA (rVPSA) in rhesus monkeys. Because of the high degree of similarity between the rhesus and human prostate gland and PSA (>90%), this animal model was well suited to accurately assess the effects of rV-PSA. Murine and other models did not share this homology. A control group receiving high-dose V-Wyeth, a group receiving low-does rV-PSA and a group receiving high-dose rV-PSA were all given 3 injections at four week intervals. Before the initial injection, one monkey in each group was given a prostatectomy in order to mimic the situation of human patients who have undergone the same procedure. Following injection, the rhesus monkeys exhibited the expected low-grade fever and other symptoms of vaccinia infection. It was found that the monkeys receiving the high dose rV-PSA vaccination expressed long term cellular immune responses specific to PSA (Table 2). Also, there was no difference in the immune response of the monkeys who had their prostates removed (Hodge et al, 1995). Much like the experiments with rV-CEA, this experiment showed the effectiveness of rV-PSA in inducing an immune response in macaques.

cific humoral and cellular responses were detected in the immunized mice. These responses were comparable to the immune response generated by rV-CEA (Conry et al, 1994). The CEA DNA vaccine was also characterized in a canine model, where sera obtained from dogs injected intramuscularly with the construct demonstrated an increase in antibody levels (Smith et al, 1998). Cellular immune responses quantified using the lymphoblast transformation (LBT) assay also revealed proliferation of CEA-specific lymphocytes. Therefore a CEA nucleic acid vaccine was able to induce both arms of the immune responses (Smith et al, 1998). CEA DNA vaccines are currently being investigated in humans.

IV. Gene-based cancer vaccine strategies using PSA Prostate cancer is the most common form of cancer and the second most common cause of cancer related death in American men (Boring et al, 1994). The appearance of prostate cancer is much more common in men over the age of fifty (Gilliland and Keys, 1995). Three of the most widely used treatments are surgical excision of the prostate and seminal vesicles, external bean irradiation, and androgen deprivation. However, conventional therapies lose their efficacy once the tumor has metastasized, which is the case in more than half of initial diagnoses (Wei et al, 1997, Ko et al, 1996).

B. PSA DNA Vaccine

PSA is a serine protease and a human glandular kallikrein gene product of 240 amino acids which is secreted by both normal and transformed epithelial cells of the prostate gland (Wang et al, 1982; Watt et al, 1986). Because cancer cells secrete much higher levels of the antigen, PSA level is a particularly reliable and effective diagnostic indicator of the presence of prostate cancer (Labrie et al, 1992). PSA is also found in normal prostate epithelial tissue and its expression is highly specific (Wei et al, 1997).

The immune responses induced by a DNA vaccine encoding for human PSA has been investigated in a murine model. The vaccine construct was constructed by cloning a gene for PSA into expression vectors under control of a CMV promoter (Figure 1). The expression of 30 kD PSA protein was determined in vitro using immunoprecipitation following a transfection with the PSA construct (F i g u r e 1 ). In vivo expression of PSA was determined by intramuscularly injecting BALB/C mice with the DNA vaccine and performing an immunohistochemistry analysis on their quadriceps muscles (Figure 2).

A. PSA recombinant vaccinia virus vaccine

Following the injection of the PSA DNA construct (pCPSA), various assays were performed to measure both

Recombinant vaccinia virus

vaccines expressing

152

Gene Therapy and Molecular Biology Vol 3, page 153

Figure 1. Construction and in vitro expression of PSA DNA vaccine. The complete coding sequence of PSA was cloned into pCDNA3 vector. Expression of PSA was assayed by immunoprecipitation with !-PSA antibodies. The immunoprecipitated sample was analyzed by SDS-PAGE (12%).

Figure 2. Immunohistochemical assay for expression of PSA on muscle cells. Frozen muscle sections were prepared from DNA injected animals and stained with !-PSA antibody. Positive antigen expression is illustrated by PSAspecific staining and representative examples of in vivo expression are highlighted with black arrows. A) A slide from a leg immunized with PSA vaccine and stained with !-PSA antibody. B ) A slide from control plasmid immunized leg stained with !-PSA antibody.

153

Gene Therapy and Molecular Biology Vol 3, page 154 Improvement of hepatitis B virus DNA vaccines by plasmids coexpressing hepatitis B surface antigen and interleukin-2. J Virol 71, 169-178.

the humoral and cellular immune responses of the mice (Kim et al, In Press). PSA-specific immune responses induced in vivo by immunization were characterized by enzyme-linked immunosorbent assay (ELISA), T helper proliferation cytotoxic T lymphocyte (CTL), and flow cytometry assays. Strong and persistent antibody responses were observed against PSA for at least 180 days following immunization. In addition, a significant T helper cell proliferation was observed against PSA protein. Immunization with pCPSA also induced MHC Class I CD8+ T cell-restricted cytotoxic T lymphocyte response against tumor cell targets expressing PSA. The induction of PSA-specific humoral and cellular immune responses following injection with pCPSA was also observed in rhesus macaques (Table 2).

Conry RM, LoBuglio AF, Kantor J, Schlom J, Loechel F, Moore SE, Sumerel LA, Barlow DL, Abrams S, Curiel DT. (1 9 9 4 ) Immune response to a carcinoembryonic antigen polynucleotide vaccine. Cancer Res 54, 1164-1168. Conry RM, LoBuglio AF, Curiel DT. ( 1 9 9 6 a ) Polynucleotide-mediated immunization therapy of cancer. S e m i n O n c o l 23, 135-147. Conry RM, Widera G, LoBuglio AF, Fuller JT, Moore SE, Barlow DL, Turner J, Yang NS, Curiel DT. ( 1 9 9 6 b ) Selected strategies to augment polynucleotide immunization. Gene Ther 3, 67-74. Foon KA, Chakraborty M, John WJ, Sherratt A, Kohler H, Bhattacharya-Chatterjee M. ( 1 9 9 7 ) Immune response to the carcinoembryonic antigen in patients treated with an anti-idiotype antibody vaccine. J C l i n I n v e s t 96, 334342.

V. Conclusion Research involving different gene-based vaccines demonstrate that they can induce effective immune responses in a variety of animal models, including rodents and macaques as well as in humans. This effect was found in both methods of in vivo delivery, though differences remain between the two. Although recombinant vaccinia virus may produce more potent immune responses than DNA, it has many side effects such as eliciting an immune response against the virus itself. This immune response reduces the effectiveness of subsequent innoculations. DNA, while less immunogenic, can be used repeatedly with less adverse side effects. Furthermore, coadministration of molecular adjuvants with DNA vaccine constructs enhance the level of antigen-specific immune responses (Kim et al, 1997a,b; Conry et al, 1996b; Kim and Weiner, 1997; Chow et al, 1997; Sin et al, 1998).

Gilliland FD, Keys CR. ( 1 9 9 5 ) Male genital cancers. Cancer 75, 295-315. Hodge JW, Schlom J, Donohue SJ, Tomaszewski JE, Wheeler CW, Levine BS, Gritz L, Panicali D, Kantor JA. ( 1 9 9 5 ) A recombinant vaccinia virus expressing human prostatespecific antigen (PSA): safety and immunogenicity in a non-human primate. Int J Cancer 63, 231-237. Jones VE, Mitchell, MS. ( 1 9 9 6 ) Therapeutic vaccines for melanoma: progress and problems. T r e n d s B i o t e c h 14, 349-355. Kantor J, Irvine K, Abrams S, Kaufman H, DiPietro J, Schlom J. ( 1 9 9 2 a ) Antitumor activity and immune responses induced by a recombinant carcinoembryonic antigenvaccinia virus vaccine. J N a t l C a n c e r I n s t 84, 10841091. Kantor J, Irvine K, Abrams S, Snoy P, Olsen R, Greiner J, Kaufman H, Eggensperger D, Schlom J. ( 1 9 9 2 b ) Immunogenicity and safety of a recombinant vaccinia virus vaccine expressing the carcinoembryonic antigen gene in a nonhuman primate. Cancer R e s 52, 69176925.

Additional studies are warranted to optimize these strategies. Areas of future study could focus on controlling the immune responses induced by these therapies and further explore their effects on humans. It would be advantageous to modulate and refine the effects of these vaccines in order to gain optimal response. There are a number of ongoing clinical studies that will help ascertain how to best use gene-based therapies.

Kaufman H, Schlom J, Kantor J. ( 1 9 9 1 ) A recombinant vaccinia virus expressing human carcinoembryonic antigen (CEA). Int J Cancer 48, 900-907. Kelley JR, Cole DJ. ( 1 9 9 8 ) Gene therapy strategies utilizing carcinoembryonic antigen as a tumor associated antigen for vaccination against solid malignancies. Gene Ther M o l B i o l 2, 14-30.

References Abrams SI, Hodge JW, McLaughlin JP, Steinberg SM, Kantor JA, Schlom J. (1 9 9 7 ) Adoptive immunotherapy as an in vivo model to explore antitumor mechanisms induced by a recombinant anticancer vaccine. J Immunother 20, 4859.

Kim JJ and Weiner DB ( 1 9 9 7 ) DNA/genetic vaccination for HIV. Springer Sem Immunopathol 19, 174-195. Kim JJ, Bagarazzi ML, Trivedi N, Hu Y, Chattergoon MA, Dang K, Mahalingam S, Agadjanyan MG, Boyer JD, Wang B, Weiner DB ( 1 9 9 7 a ) Engineering of in vivo immune responses to DNA immunization via co-delivery of costimulatory molecule genes. N a t B i o t e c h 15, 641645.

Boring CC, Squires TS, Tong T. ( 1 9 9 4 ) Cancer statistics, 1994. Ca: A Cancer Journal for Clinicians 44, 726. Chow YH, Huang WL, Chi WK, Chu YD, Tao MH. ( 1 9 9 7 )

154

Gene Therapy and Molecular Biology Vol 3, page 155 Kim JJ, Ayyvoo V, Bagarazzi ML, Chattergoon MA, Dang K, Wang B, Boyer JD, Weiner DB. ( 1 9 9 7 b ) In vivo engineering of a cellular immune response by coadministration of IL-12 expression vector with a DNA immunogen. J . I m m u n o l . 158, 816-826. Kim JJ, Trivedi NN, Nottingham L, Morrison L, Tsai A, Hu Y, Mahalingam S, Dang K, Ahn L, Doyle NK, Wilson DM, Chattergoon MA, Chalian AA, Boyer JD, Agadjanyan MG, Weiner DB. ( 1 9 9 8 ) Modulation of amplitude and direction of in vivo immune responses by coadministration of cytokine gene expression cassettes with DNA immunogens. Eur J Immunol 28, 1089-1103. Kim JJ, Trivedi NN, Mahalingam S, Morrison L, Tsai A, Chattergoon MA, Dang K, Patel M, Ahn L, Chalian AA, Boyer JD, Kieber-Emmons T, Agadjanyan MG, Weiner DB. Molecular and immunological analysis of genetic prostate specific antigen (PSA) vaccine. O n c o g e n e In Press. Ko SC, Gotoh A, Thalmann GN, Zhau HE, Johnston DA, Zhang WW, Kao C, Chung LWK. ( 1 9 9 6 ) Molecular therapy with recombinant p53 adenovirus in an androgenindependent, metastatic human prostate cancer model. Hum Gene Ther 7, 1683-1691. Labrie F, Dupont A, Suburu R, Cusan L, Tremblay M, Gomez JL, Edmond J. ( 1 9 9 2 ) Serum prostate specific antigen as pre-screening test for prostate cancer. J U r o l 147, 84652. MacGregor RR, Boyer JD, Ugen KE, Lacy KE, Bagarazzi ML, Chattergoon MA, Baine Y, Higgins TJ, Ciccarelli RB, Coney LR, Ginsberg RS, Weiner DB ( 1 9 9 8 ) First human trial of a DNA-based vaccine for treatment of HIV-1 infection: safety and host response. J I n f D i s 178, 92100. Roth JA and Cristiano RJ. (1 9 9 7 ) Gene therapy for cancer: what have we done and where are we going? J Natl Cancer Inst 89, 21-39. Sin JI, Kim JJ, Boyer JD, Huggins C, Higgins T, Weiner DB. ( 1 9 9 8 ) In vivo modulation of immune responses and protective immunity against herpes simplex virus-2 infection using cDNAs expressing Th1 and Th2 Type Cytokines in gD DNA Vaccination. J Virol In Press. Smith BF, Baker HJ, Curiel DT, Jiang W, Conry RM ( 1 9 9 8 ) Humoral and cellular immune responses of dogs immunized with a nucleic acid vaccine encoding human carcinoembryonic antigen. Gene Ther 5, 865-868. Sogn JA, Finerty JF, Heath AK, Shen GLC, Austin FC. ( 1 9 9 3 ) Cancer vaccines: the perspective of the Cancer Immunology Branch, NCI. A n n a l s N Y A c a d S c i 690, 322-330. Tang D, DeVit M, Johnston S. ( 1 9 9 2 ) Genetic immunization is a simple method for eliciting an immune response. Nature 356, 152-154. Tsang KY, Zaremba S, Nieroda CA, Zhu MZ, Hamilton JM, Schlom J. (1 9 9 5 ) Generation of human cytotoxic T-cells specific for human carcinoembryonic antigen epitopes from patients immunized with recombinant vaccinia-CEA

155

vaccine. J Natl Cancer Inst 87, 982-990. Ulmer JB, Donnelly J, Parker SE, Rhodes GH, Felgner PL, Dwarki VL, Gromkowski SH, Deck R, DeVitt CM, Friedman A, Hawe LA, Leander KR, Marinez D, Perry H, Shiver JW, Montgomery D, Liu MA. ( 1 9 9 3 ) Heterologous protection against influenza by injection of DNA encoding a viral protein. S c i e n c e 259, 1745-1749. Wang B, Ugen KE, Srikantan V, Agadjanyan MG, Dang K, Refaeli Y, Sato A, Boyer J, Williams WV, Weiner DB. ( 1 9 9 3 ) Gene inoculation generates immune responses against human immunodeficiency virus type 1. P r o c . N a t l . A c a d . S c i . U S A 90, 4156-4160. Wang MC, Kuriyama M, Papsidero LD, Loor RM, Valenzyela LA, Murphy GP, Chu TM. ( 1 9 8 2 ) Prostate antigen of human cancer patients. Meth Cancer Res 19, 179-197. Watt KWK, Lee PJ, Timkulu TM, Chan WP, Loor R. ( 1 9 9 6 ) Human prostate-specific antigen: structural and functional similarity with serine proteases. P r o c N a t l A c a d S c i USA 83, 3166-3170. Wei C, Willis RA, Tilton BR, Looney RJ, Lord EM, Barth RK, Frelinger JG. ( 1 9 9 7 ) Tissue-specific expression of the human prostate-specific antigen gene in transgenic mice: Implications for tolerance and immunotherapy. Proc Natl Acad Sci USA 94, 6369-6374. Wolff JA, Malone RW, Williams P, Chong W, Acsadi G, Jani A, Felgner PL. ( 1 9 9 0 ) Direct gene transfer into mouse muscle in vivo. S c i e n c e 247, 1465-1468. Zaremba S, Barzaga E, Zhu M, Soares N, Tsang KY, Schlom J. ( 1 9 9 7 ) Identification of an enhancer agonist cytotoxic T lymphocyte peptide from human carcinoembryonic antigen. Cancer Res 57, 4570-4577.

Gene Therapy and Molecular Biology Vol 3, page 157 Gene Ther Mol Biol Vol 3, 157-165. August 1999.

Rational vaccine design through the use of molecular adjuvants Review Article

Jong J. Kim1, Liesl K. Nottingham2, Jim Oh1, Daniel Lee1, Ken Wang 1, Mera Choi3, Tzvete Dentchev 1, Darren Wilson, Devin M. Cunning2, Ara A. Chalian 2, Jean Boyer, Jeong I. Sin1, and David B. Weiner 1 1

Department of Pathology and Laboratory Medicine; 2Department of Otolaryngology/Head and Neck Surgery, University of Pennsylvania, Philadelphia, PA 19104, 3Bryn Mawr College _________________________________________________________________________________________________ Corresponding Author: Jong J. Kim, Ph.D., Department of Pathology and Laboratory Medicine, University of Pennsylvania, 505 Stellar-Chance, 422 Curie Blvd., Philadelphia, PA 19104. Tel: (215) 662-2352; Fax: (215) 573-9436; Email: jonger@seas.upenn.edu Received: 18 September 1998; accepted: 25 September 1998

Summary Nucleic acid immunization is an important vaccination strategy which delivers DNA constructs encoding for a specific immunogen into the host. These expression cassettes transfect the host cells, which become the i n v i v o protein source for the production of antigen. This antigen then is the focus of the resulting immune response. This vaccination technique is being explored as an immunization strategy against a variety o f infectious diseases as w e l l as cancer. The first generation DNA immunization experiments have shown that the DNA vaccines’ ability to elicit humoral and cellular responses i n v i v o i n a s a f e a n d well-tolerated manner i n various model systems, including humans. As we explore the next generation of DNA vaccines, our goal is to refine the current strategy t o elicit more clinically efficacious immune responses. A more c l i n i ca l l y ef fec t i ve v a cc i ne m a y need t o elicit a more specific immune response against the targeted pathogen. It would be a distinct advantage to design immunization strategies which can be “focused” according to the correlates of protection known for the particular pathogen. In order to focus the immune responses induced from DNA immunization, we have investigated the co-delivery o f genes for immunologically important molecules, such as costimulatory molecules and cytokines which play critical regulatory and signaling roles in immunity. We and others have shown that the use of these molecular adjuvants could enhance and modulate immune responses induced by DNA immunogens. Co-administration of costimulatory molecules (CD80 and CD86), proinflammatory cytokines (IL-1 , T N F - , and TNF- ), Th1 cytokines (IL-2, IL-12, IL-15, and IL-18), Th2 cytokine (IL-4, IL-5 and IL-10), and GM-CSF with DNA vaccine constructs led to modulation of the magnitude and direction (humoral or cellular) of the immune responses. These studies demonstrate the potential utility of molecular adjuvant strategy as an important tool for the development of more rationally designed vaccines.

with natural infection (Stasney et al, 1955; Paschkis et al, 1955; Ito, 1960). Nucleic acid or DNA inoculation is an important vaccination technique which delivers DNA constructs encoding specific immunogens directly into the host (Wolff et al, 1990; Tang et al, 1992; Wang et al, 1993; Ulmer et al 1993; Kim et al, 1997a; Agadjanyan et

I. Introduction Although the injection of DNA into tissues was originally reported in the 1950s, the technology has gained more attention in recent years as a safe means of mimicking in vivo protein production normally associated 157

Kim et al: Vaccine design using molecular adjuvants al, 1997, Tascon et al, 1996, Conry et al, 1996). This injection results in the subsequent expression of the foreign gene in that host and the presentation of the specific encoded proteins to the immune system. DNA vaccine constructs are produced as small circular vehicles or plasmids. These plasmids are constructed with a promoter site which starts the transcription process, an antigenic DNA sequence and a messenger RNA stop site containing the poly A tract necessary for conversion of the messenger RNA sequence into the antigen protein by the ribosomal protein manufacturing machinery (F i g u r e 1 ). This antigen then is the focus of the resulting immune response. This vaccination technique is being explored as an immunization strategy against cancer as well as a variety of infectious diseases including AIDS.

1997; Chattergoon et al, 1997). Since DNA vaccines are non replicating and the vaccine components are produced within the host cells, they can be constructed to function safely with the specificity of a subunit vaccine. However, DNA vaccine cassettes produce immunological responses that are more similar to live vaccine preparations. By directly introducing DNA into the host cell, the host cell is essentially directed to produce the antigenic protein, mimicking viral replication or tumor cell marker presentation in the host. This process has been reported to generate both antibody and cell mediated, particularly cytotoxic T cell-mediated, immunity (Figure 2). Unlike a live attenuated vaccine, conceptually there is little risk from reversion to a disease-causing pathogen from the injected DNA, and there is no risk for secondary infection as the material injected is not-living and not-infectious. In addition, genes which lead to undesired immunologic inhibition or cross-reactivity (autoimmunity) may be either altered or deleted altogether. Finally, DNA vaccines can be manipulated to present a particular genome of the pathogen or display specific tumor antigens in nonreplicating vectors (Figure 1).

II. Potential advantages of DNA vaccines Nucleic acid immunization may afford several potential advantages over traditional vaccination strategies such as whole killed or live attenuated virus and recombinant protein-based vaccines (Kim and Weiner,

Figure 1. Potential immunologic targets for DNA vaccination against HIV-1. These targets include env, gag, and pol genes as well as the four accessory genes.

158

Gene Therapy and Molecular Biology Vol 3, page 159

Figure 2. Induction of antigen-specific humoral and cellular immune responses.

III. Molecular adjuvants as a immune modulation strategy The overall objective of any immunization strategy is to induce specific immune responses which protect the immunized individual from a given pathogen over his or her lifetime. One major challenge in meeting this goal is that the correlates of protection from an individual pathogen vary from one infectious agent to the next. The first generation DNA immunization experiments have shown that the DNA vaccines’ ability to elicit humoral and cellular responses in vivo in a safe and well-tolerated manner in various model systems, including humans. As we explore the next generation of DNA vaccines, our goal is to refine the current strategy to elicit more clinically efficacious immune responses. A more clinically effective vaccine may need to elicit a more specific immune response against the targeted pathogen. It would be a distinct advantage to design immunization strategies which can be targeted according to the correlates of protection 159

known for the particular pathogen (Figure 3). Such refinement could be accomplished by co-delivering genes for immunologically important molecules, such as costimulatory molecules and cytokines which play critical regulatory and signaling roles in immunity (Kim and Weiner, 1997). These molecular adjuvant constructs could be co-administered along with immunogen constructs to modulate the magnitude and direction (humoral or cellular) of the immune responses induced (Figure 4). There has been several reports of immune modulation by protein delivered cytokines. However, the results in general appeared marginal. More recently, we and others have focused on analyzing immune responses induced to such gene delivery. Raz et al. observed that intramuscular injections of plasmids encoding IL-2, IL-4, or TGF-!1 modestly modulated immune responses to transferrin protein delivered at a separate site (Raz et al, 1993). IL-2 immunization resulted in an enhancement of antibody and T helper proliferative responses while TGF-!1 immunization reduced anti-transferrin responses.

Gene Therapy and Molecular Biology Vol 3, page 160

Figure 3. The potential utility of the molecular adjuvant network. Tailoring the induction of specific immune responses by vaccination programs against viral, bacterial, or parasitic diseases could be beneficial.

Figure 4. Cytokines as immune response regulators. Cytokines play critical roles in the immune and inflammatory responses. Based upon their specific function in the immune system these cytokines could be further grouped as proinflammatory, Th1, and Th2 cytokines. Along with costimulatory molecules, these cytokines also play important roles in the activation and proliferation of T and B cells.

160

Gene Therapy and Molecular Biology Vol 3, page 161 gD protein with the gene plasmids encoding for Th1-type (IL-2, 12, 15, 18) and Th2-type (IL-4, IL-10) cytokines in an effort to drive immunity induced by vaccination. We then analyzed the vaccine modulatory effects on resulting immune phenotype and on the mortality and the morbidity of the immunized animals following HSV lethal challenge. We observed Th1 cytokine gene coadministration not only enhanced survival rate, but also reduced the frequency and severity of herpetic lesions following intravaginal HSV challenge (Figure 6). On the other hand, co-injection with Th2 cytokine genes increased the rate of mortality and morbidity of the challenged mice. Again, among the Th1 type cytokine genes tested IL-12 was particularly a potent adjuvant for the gD DNA vaccination.

IV. Modulation of immune responses using cytokine molecular adjuvants In order to focus the immune responses induced from DNA immunization, we have investigated the co-delivery of molecular adjuvants. We first reported that coimmunization of GM-CSF genes with DNA vaccine constructs increases antigen-specific antibody and T helper cell proliferation responses while co-immunization with IL-12 genes results in weaker antibody responses and enhanced T helper cell proliferation (Kim et al, 1997b,c). In addition, IL-12 co-immunization resulted in a significant enhancement of CTL responses. Importantly, we observed a significant enhancement of CTL response in vivo with the co-administration of murine IL-12 genes with four different HIV-1 DNA immunogens (gag/pol, envelope, vif, and nef) which were CD8+ T cell- and MHC class I-restricted. In contrast, almost no effect on CTL induction was observed with the genes for GM-CSF in these studies. Moreover, Iwasaki et al. (1997) reported that GM-CSF and IL-12 co-delivery with DNA immunogen encoding for influenza NP resulted in enhanced cellular immune responses. Moreover, Agadjanyan et al. (1997) reported that co-administration of IL-12 genes with HIV-2 DNA immunogen resulted in a dramatic enhancement of both Th and CTL responses. Furthermore, coadministration of IL-12 genes with DNA immunogens strongly directed the antigen specific immune response towards a Th1 type immunity and induced delayed type hypersensitivity (DTH) to contact allergens as an in vivo model of the Th1 response (Kim et al, 1998a). In addition to these reports, Chow et al. reported that either injection of plasmid co-expressing hepatitis B surface antigen (HBsAg) and IL-2 or co-injection of IL-2 genes with plasmid expressing HBsAg resulted in the enhancement of both antibody and T helper cell responses (Chow et al, 1997).

V. Modulation of immune responses using costimulatory molecule adjuvants The generation of the T cell immune response is a complex process that requires the engagement of T cells with professional APCs such as dendritic cells, macrophages, and B cells. These professional APCs possess large surface areas for interaction with T cells. They also express high levels of MHC class I and II molecules, adhesion molecules, and costimulatory molecules which are critical for efficient antigen presentation and T cell activation. Professional APCs initiate T cell activation by binding antigenic peptideMHC complexes to T cell receptor molecules. In addition, the APCs provide secondary signals through the ligation of costimulatory molecules with their receptors (CD28/CTLA-4) present on T cells. These costimulatory signals are required for the clonal expansion and differentiation of T cells. The blocking of this additional costimulatory signal leads to T cell anergy (Schwartz et al, 1992). Among different costimulatory molecules, CD80 and CD86 have been observed to provide potent immune signals ( Lanier et al, 1995, Linsley et al, 1990).

More recently, we investigated the induction and regulation of immune responses from the co-delivery of proinflammatory cytokines (IL-1", TNF-", and TNF-!), Th1 cytokines (IL-2, IL-15, and IL-18), and Th2 cytokines (IL-4, IL-5 and IL-10) (Figure 5) (Kim et al, 1998b). We observed enhancement of antigen-specific humoral response with the co-delivery of Th2 cytokines IL-4, IL-5, and IL-10 as well as that of IL-2 and IL-18. A dramatic increase in antigen-specific T helper cell proliferation was seen with IL-2 and TNF-" co-injections. In addition, we observed a significant enhancement of the cytotoxic response with the co-administration of TNF-" and IL-15 genes with HIV-1 DNA immunogens. These increases in CTL response were both MHC class I-restricted and CD8+ T cell-dependent. We also investigated whether the Th1 or Th2-type immune responses are more important for protection from HSV-2 infection (Sin et al, 1998). We codelivered DNA expression construct encoding for HSV-2

The CD80 and CD86 molecules are surface glycoproteins and members of immunoglobulin superfamily which are expressed only on professional APCs (Lanier et al, 1995, Linsley et al, 1990, June et al,1994). Although both CD80 and CD86 molecules interact with either CD28 or CTLA-4 molecules on T cells, CD80 and CD86 expression seem to be differentially regulated. CD86 is constitutively expressed by the APCs whereas CD80 is expressed only after activation of these cells (Freeman et al, 1989; Azuma et al, 1993; Freedman et al, 1991). Thus, CD86 may be important in the early interactions between APCs and T cells during the induction phase of the immune response.

161

Gene Therapy and Molecular Biology Vol 3, page 162

Figure 5. Each cytokine gene was cloned into expression plasmids under the control of a CMV promoter.

Figure 6. Protection from lethal HSV-2 challenge. Each group of mice (n=10) was immunized with gD DNA vaccines (60 Âľg), and/or cytokine genes (40 Âľg) at 0 and 2 weeks. Three weeks after the second immunization, mice (n=8) were challenged i.v. with 200 x LD 50 of HSV2 strain 186 (7 x 10 5 pfu).

162

Gene Therapy and Molecular Biology Vol 3, page 163 candidates, such as chemokines, should be further developed and tested. Cumulatively, these studies demonstrate the potential utility of molecular adjuvant strategy as an important tool for the development of more rationally designed vaccines.

We recently reported that CD86 molecules play a prominent role in the antigen-specific induction of CD8+ cytotoxic T lymphocytes when delivered as vaccine adjuvants (Figure 7) (Kim et al, 1997a). Coadministration of CD86 cDNA along with DNA encoding HIV-1 antigens intramuscularly dramatically increased antigen-specific T-cell responses without a significant change to the level of the humoral response. This enhancement of cytotoxic T lymphocyte (CTL) response was both major histocompatibility complex (MHC) class I-restricted and CD8+ T cell-dependent. Similar results have been obtained by other investigators who also found that CD86, not CD80 co-expression results in the enhancement of T cell-mediated immune responses (Tsuji et al, 1997; Iwasaki et al, 1997). Accordingly, we speculate that engineering of non-professional APCs such as muscle cells to express CD86 costimulatory molecules could empower them to prime CTL precursors. On the other hand, the enhancement effect of CD86 co-delivery could also have been mediated through the direct transfection of a small number of professional APCs residing within the muscle tissue. Subsequently, these cells could have greater expression of costimulatory molecules and could in theory become more potent.

References Stasney, J., Cantarow, A., and Paschkis, K.E. (1 9 5 5 ). Production of Neoplams by Injection of Fractions of Mammalian Neoplasms. Cancer Res. 11, 775-782. Paschkis, K. E., Cantarow, A., Stasney, J. (1 9 5 5 ). Induction of Neoplasms by Injection of Tumor Chromatin. J . N a t l . Cancer Inst. 15:1525-1532. Ito, Y. (1 9 6 0 ). A Tumor-producing factor extracted by phenol from papillomatous tissue of cottontail rabbits. V i r o l o g y 12, 596-601. Wolff, J. A., R. W. Malone, P. Williams, W. Chong, G. Acsadi, A. Jani, P. L. Felgner. (1 9 9 0 ). Direct gene transfer into mouse muscle in vivo. S c i e n c e . 247, 14651468. Tang, D., M. DeVit, S. Johnston. (1 9 9 2 ). Genetic immunization is a simple method for eliciting an immune response. Nature. 356, 152-154. Wang, B., K. E. Ugen, V. Srikantan, M. G. Agadjanyan, K. Dang, Y. Refaeli, A. Sato, J. Boyer, W. V. Williams, D. B. Weiner. (1 9 9 3 ). Gene inoculation generates immune responses against human immunodeficiency virus type 1. P r o c . N a t l . A c a d . S c i . U S A 90, 4156-4160.

VI. Future directions As summarized in Figure 8, we observed that significant modulation was possible using molecular adjuvants. This cytokine gene adjuvant network underscores an important level of control in the induction of specific immune responses to tailor vaccination programs more closely to the correlates of protection which vary from disease to disease. This type of fine control of vaccine and immune therapies was previously very difficult to obtain. Controlling the magnitude and direction of the immune response could be advantageous in a wide variety of vaccine strategies. For instance, in a case where T cell mediated response is paramount, but the humoral response may not be needed or even be harmful, IL-12 genes could be chosen as the immune modulator to be co-delivered with a specific DNA immunogen. On the other hand, for building vaccines to target extracellular bacteria, for example, IL-4, IL-5 or IL-10 genes could be co-injected. Furthermore, in cases where both CD4+ T helper cells and antibodies play more important roles in protection, GM-CSF as well as IL-2 could be co-delivered. Lastly, in cases where all three arms of immune responses are critical, TNF-" could be co-injected to give a combined enhancement of antibody, T helper cell, and CTL responses. In this regard it will be important to examine combination delivery in the presence or the absence of costimulatory genes to further control the immune responses. Furthermore, additional molecular adjuvant

Ulmer, J. B., J. Donnelly, S. E. Parker, G. H. Rhodes, P. L. Felgner, V. L. Dwarki, S. H. Gromkowski, R. Deck, C. M. DeVitt, A. Friedman, L. A. Hawe, K. R. Leander, D. Marinez, H. Perry, J. W. Shiver, D. Montgomery, M. A. Liu. (1 9 9 3 ). Heterologous protection against influenza by injection of DNA encoding a viral protein. S c i e n c e 259, 1745-1749. Kim, J. J., M.L. Bagarazzi, N. Trivedi, Y. Hu, M.A. Chattergoon, K. Dang , S. Mahalingam, M.G. Agadjanyan, J.D. Boyer, B. Wang, D.B. Weiner. (1 9 9 7 a ). Engineering of In Vivo Immune Responses to DNA Immunization Via Co-Delivery of Costimulatory Molecule Genes. Nature Biotech. 15, 641-645. Agadjanyan, M. G., N.N.Trivedi, S. Kudchodkar, M. Bennett, W. Levine, A. Lin, J. Boyer, D. Levy, K. Ugen, J.J. Kim, D.B. Weiner. (1 9 9 7 ). An HIV-2 DNA vaccine induces cross reactive immune responses against HIV-2 and SIV. AIDS Human Retrov 13, 1561-1572. Tascon, R. E., M.J. Colston, S. Ragno, E. Stavropoulos, D. Gregory,, and D. B. Lowrie. (1 9 9 6 ). Vaccination against tuberculosis by DNA injection. Nature Med. 2, 888-92. Conry, R. M., G. Widera, A.F. LoBuglio, J.T. Fuller, S.E. Moore, D.L. Barlow, J. Turner, N.-S. Yang, D.T. Curiel. (1 9 9 6 ). Selected strategies to augment polynucleotide immunization. Gene Ther 3, 67-74.

163

Gene Therapy and Molecular Biology Vol 3, page 164

Figure 7. Co-expression of HIV-1 envelope gp120 protein with CD86 on muscle cells. Frozen muscle sections were prepared from DNA injected animals and stained with FITC-labeled (green) "-CD86 antibodies and Texas Red-labeled (red) "-gp120 antibodies. (A) A slide from a leg immunized with pCDNA3 (control vector) was stained with "-CD86 and "-gp120. (B ) A slide from a leg immunized with pCEnv+pCD86 was stained with "-CD86 and "-gp120 antibodies.

Figure 8. A summary of the each cytokine co-administration effects on antibody (y-axis), T helper (x-axis), and cytotoxic T lymphocyte responses (z-axis). Each cytokine is plotted on the 3-D axis according to its effects on the three modes of immune response.

164

Gene Therapy and Molecular Biology Vol 3, page 165 Kim, J. J. and D.B. Weiner. (1 9 9 7 ). DNA/genetic vaccination for HIV. S p r i n g e r S e m I m m u n o p a t h o l 19, 174-195.

June, C., J. A. Bluestone, L. M. Nadler, C. B. Thompson. (1 9 9 4 ). The B7 and CD28 receptor families. I m m u n o l . Today 15, 321-333.

Chattergoon, M., J. Boyer, , and D. B. Weiner. (1 9 9 7 ). Genetic immunization: a new era in vaccines and immune therapies. FASEB J. 11:753-763.

Freeman, G. J., A. S. Freedman, J. M. Segil, G. Lee, J. F. Whitman, L. M. Nadler. (1 9 8 9 ). B7 a new member of the Ig superfamily with unique expression on activated and neoplastic B cells. J . I m m u n o l. 143, 2714-2722.

Raz, E., A. Watanabe, S. M. Baird, R. A. Eisenberg, T. B. Parr, M. Lotz, T. J. Kipps, D. A. Carson. (1 9 9 3 ). Systemic immunological effects of cytokine genes injected into skeletal muscle. P r o c . N a t l . A c a d . S c i . USA 90, 4523-4527.

Azuma, M., D. Ito, H. Yagita, K. Okumura, J. H. Phillips, L. L. Lanier, C. Somoza. (1 9 9 3 ). B70 antigen is a second ligand for CTLA-4 and CD28. Nature 366, 76-79. Freedman, A. S., G. J. Freeman, K. Rhynhart, L. M. Nadler. (1 9 9 1 ). Selective induction of B7/BB-1 on interferon-#stimulated monocytes: a potential mechanism for amplification of T cell activation through the CD28 pathway. C e l l . I m m u n o l . 137, 429-437.

Kim, J. J., V. Ayyvoo, M. L. Bagarazzi, M. A. Chattergoon, K. Dang, B. Wang , J. D. Boyer, D. B. Weiner. (1 9 9 7 b ). In vivo Engineering of a Cellular Immune Response by Co-administration of IL-12 Expression Vector with a DNA Immunogen. J. Immunol. 158, 816-826.

Tsuji, T., K. Hamajima, N. Ishii, I. Aoki, J. Fukushima, K.Q. Xin, S. Kawamoto, S. Sasaki, K. Matsunaga, Y. Ishigatsubo, K. Tani, T. Okubo, K. Okuda. (1 9 9 7 ). Immunomodulatory effects of a plasmid expressing B7-2 on humanimmunodeficiency virus-1-specific cell-mediated immunity induced by a plasmid encoding the viral antigen. Euro. J. Immuno. 27, 782-787.

Kim, J. J., V. Ayyavoo, M.L. Bagarazzi, M. Chattergoon, J.D. Boyer, B. Wang, D.B. Weiner. (1 9 9 7 c ). Development of a Multi-component Candidate Vaccine for HIV-1. V a c c i n e 15, 879-883. Kim, J. J., H.C. Maguire, Jr., L.K. Nottingham, L.D. Morrison, A. Tsai, J.I. Sin, A.A. Chalian, D.B. Weiner. (1 9 9 8 a ). Co-administration of IL-12 or IL-10 expression cassettes drives immune responses towards a Th1 phenotype. J . I n t e r f . C y t o . R e s . 18, 537-547.

Iwasaki, A., B.J. Stiernholm, A.K. Chan, N.L. Berstein, B.H. Barber. (1 9 9 7 ). Enhanced CTL responses mediated by plasmid DNA immunogens encoding costimulatory molecules and cytokines. J . Immunol. 158, 45914601.

Chow, Y.-H., W. -L. Huang, W. -K Chi, Y. -D Chu, M. -H Tao. (1 9 9 7 ). Improvement of hepatitis B virus DNA vaccines by plasmids coexpressing hepatitis B surface antigen and interleukin-2. J . V i r o l . 71, 169-178. Kim, J. J., N. N. Trivedi, L. Nottingham, L. Morrison,A. Tsai, Y. Hu, S. Mahalingam, K. Dang, L. Ahn, N. K. Doyle, D. M. Wilson, M. A. Chattergoon, A. A. Chalian, J. D. Boyer, M. G. Agadjanyan, D. B. Weiner. (1 9 9 8 b ). Modulation of Amplitude and Direction of In Vivo Immune Responses By Co-Administration Of Cytokine Gene Expression Cassettes With DNA Immunogens. Eur. J. Immunol. 28, 1089-1103. Sin, J. I., J. J. Kim, J. D. Boyer, C. Huggins, T. Higgins, D. B. Weiner. (1 9 9 8 ). In Vivo Modulation of Immune Responses and Protective Immunity against Herpes Simplex Virus-2 Infection using cDNAs expressing Th1 and Th2 Type Cytokines in gD DNA Vaccination. J . V i r o l . In Press. Schwartz, R. H. (1 9 9 2 ). Costimulation of T lymphocytes: the role of CD29, CTLA-4, and B7/BB1 in interleukin-2 production and immunotherapy. C e l l 71, 1065-1068. Lanier, L. L., S. O’Fallon, C. Somoza, J. H. Phillips, P. S. Linsley, K. Okumura, D. Ito, M. Azuma. (1 9 9 5 ). CD80 (B7) and CD86 (B70) provide similar costimulatory signals for T cell proliferation, cytokine production, and generation of CTL. J . I m m u n o l. 154, 97-105. Linsley, P. S., E. A. Clark, J .A. Ledbetter. (1 9 9 0 ). The T cell antigen, CD28, mediates adhesion with B cells by interacting with activation antigen, B7/BB-1. P r o c . Natl. Acad. Sci. USA 87, 5031-5035.

165

Gene Therapy and Molecular Biology Vol 3, page 167 Gene Ther Mol Biol Vol 3, 167-177. August 1999.

In vivo production of therapeutic antibodies by engineered cells for immunotherapy of cancer and viral diseases Review Article

Mireia Pelegrin, Danièle Noël, Mariana Marin, Estanislao Bachrach, Robert M. Saller*, Brian Salmons*, and Marc Piechaczyk Institut de Génétique Moléculaire de Montpellier, UMR 5535, CNRS, 1919 route de Mende, 34293 Montpellier Cédex 05, France * Bavarian Nordic Research Institute, Fraunhoferstr. 18B, 82152 Martinsried, Germany __________________________________________________________________________________________________ Correspondence: Marc Piechaczyk, Ph.D. Tel: + (33) 4.67.61.36.68; Fax + (33) 4.67.04.02.45; E-mail: piechaczyk@jones.igm.cnrs-mop.fr Received: 18 November 1998; accepted: 25 November 1998

Summary Our recently developed ability to produce human monoclonal antibodies, together with that of reshaping antibody molecules, offers new tools for treating a number of human diseases. Direct injection of purified antibodies, or of antibody-related molecules, to patients would, however, not always be possible or desirable. This is especially true in the case of long-term therapies for at least two reasons. One is the high cost of antibodies certified for human use. The other is the possibility of neutralizing anti-idiotypic immune responses as a result of repeated injection of m a s s i v e d o s e s o f a n t i b o d y . I n v i v o production of therapeutic antibodies through either genetic modification of patients' cells or implantation of antibody-producing cells might overcome both of t h e s e h u r d l e s . S e v e r a l c e l l t y p e s s u i t a b l e f o r u s e i n c e l l / g e n e therapy protocols, such as skin fibroblasts, keratinocytes, myogenic cells and hepatocytes, are capable of producing monoclonal antibodies i n v i t r o upon gene transfer. Furthermore, the grafting of engineered myogenic cells permits the long-term systemic delivery o f recombinant antibodies i n immunocompetent mice. Importantly, antibodies produced both i n v i t r o and i n v i v o , retain the specificity and the affinity of the parental antibody and no anti-idiotypic response i s detected i n mice producing ectopic antibodies. Long-term systemic delivery of such antibodies into mice can also be achieved through the implantation o f antibody-producing c e l l s encapsulated into a new biocompatible material, cellulose sulphate. Importantly, no inflammation occurs at capsule implantation sites over periods as long as 10 months. Moreover, no anti-idiotypic response develops against antibodies released by encapsulated cells. Encapsulation of antibody-producing cells in immunoprotective devices should offer multiple advantages over genetic modification of patients' cells. These include protection against immune cells of treated individuals, the possibility of easy removal of implanted cells as w e l l as that o f implantation o f non-autologous c e l l s . Taken together, these observations demonstrate that long-term i n v i v o production and systemic delivery of monoclonal antibodies is technically feasible. Application o f this technology t o the treatment o f various viral and autoimmune diseases as well as that of cancer is currently underway.

I. Introduction Specific antibodies can be generated against virtually any type of molecule since antigens can be proteins, nucleic acids, lipids or glucids. They can also be self or

167

foreign. The potential of clinical applications for antibodies is thus enormous and concerns a wide range of diseases including cancer, viral infections, transplant rejection, autoimmunity, toxic shock, rheumatoid arthritis, and restenosis (Chester and Hawkins, 1995).

Pelegrin et al: Engineering cells to produce therapeutic antibodies for immunotherapy of cancer Since the discovery of monoclonal antibodies in 1975, various antibody-based therapies have been tested, mostly for treating patients suffering from cancer. However, the poor efficiency of the first monoclonal antibodies used in clinical trials, the development of neutralizing immune responses by patients against antibodies of animal origin and the long periods of time necessary for forming a proper view of the efficacy of treatments have momentarily tempered the initial enthusiasm raised by this technology. Nevertheless, the therapeutic successes obtained during the past years (Scott and Welt, 1997) and the rapid developments of antibody engineering have brought therapeutic monoclonal antibodies back to the fore. Among the therapeutic successes, one can mention a variety of anti-idiotypic antibodies for treating B lymphoma (White et al., 1996) and the now commercially available chimeric antibody ICED-C2B8, which is more efficient than conventional chemotherapy for treating nonHodgkin’s lymphomas (Maloney et al., 1997; Marwick, 1997). The main initial drawback met when administering monoclonal antibodies in human patients, namely the immunogenicity of murine antibodies, can now be overcome following several approaches (Figure 1). These include : (i ) the humanization of animal antibodies using site-directed mutagenesis possibly assisted by computerized molecular modeling (Wawrzynczak, 1995); (i i ) the generation of hybridomas from transgenic mice harboring the human immunoglobulins loci substituted for the mouse loci (Bruggemann and Taussig, 1997; Mendez et al., 1997); (i i i ) the construction of hybridomas from activated human B lymphocytes (Wawrzynczak, 1995); and

(i v ) the screening of bacteriophage libraries expressing human immunoglobulins at their surface (Marks and Marks, 1996; Rader and Barbas, 1997). In addition, gene engineering now allows both the improvement of intrinsic properties of antibodies, such as affinity and avidity, the grafting of new effector or enzymatic functions as well as the construction of new antibody-based molecules such as single chain Fv, bispecific antibodies (Chester and Hawkins, 1995; Wawrzynczak, 1995). In conclusion, molecular engineering of antibodies, together with the possibility of generating human monoclonal antibodies, provide us with new antibodies and antibody-related molecules which will, undoubtedly, find clinical therapeutical applications, especially in the field of gene therapy (Pelegrin et al., 1998).

II. A gene/cell therapy approach for the systemic delivery of therapeutic antibodies. In theory, the simplest mode of administration of therapeutic antibodies consists of repeated intraveinous injection. However, the high cost of antibodies produced under gmp (good manufacturing practice) conditions makes most monoclonal antibodies uneconomic for long-term treatments (several months to several years) on a large scale since numerous antibody-based therapies would involve several tens to several hundreds of mg of antibody per month and per patient. Therefore, clinical application of therapeutic antibodies in the long-term is limited by the necessity of finding financially acceptable delivery systems.

Generation of hybridomas from transgenic mice expressing human immunoglobulins genes

Humanisation of animal monoclonal antibodies

Production of monoclonal antibodies for long-term use in humans.

Screening of libraries of bacteriophages expressing human immunoglobulins

Generation of human hybridomas from human B lymphocytes

F i g u r e 1 . Generation of monoclonal antibodies suitable for long-term use in humans.

168

To solve this issue, a new gene/cell therapy based on the in vivo production of ectopic antibodies through either the genetic modification of patients' cells or the implantation of antibody-producing cells encapsulated within immunoprotective devices are currently being developed in the laboratory. These delivery systems should not only render long-term therapeutic antibody treatments cost-effective but should also provide an additional therapeutic benefit. Continuous and sustained delivery of antibodies at a low, but therapeutic, level should permit the suppression, or at least the delay, of neutralizing antiidiotypic immune responses which often develop when massive doses of purified immunoglobulins are repeatedly injected (see below).

III. Potential applications of the in vivo production of ectopic antibodies. A first and obvious clinical target for therapeutic antibodies produced in vivo is cancer. Long-term production of ectopic antibodies could, thus, be used in the context of surveillance treatments for preventing relapse after a primary treatment consisting of surgery, chemo- or radiotherapy. Providing the basis for future protocols, several antibodies, cytostatic or cytocidic for tumor cells, have already been characterised (Old, 1995; Riethm端ller et al., 1993; Scott and Welt, 1997; Vitetta and Uhr, 1994). Some of them have even been used with success in various clinical trials based on passive administration of purified immunoglobulins (Table 1) (Scott and Welt, 1997). A second target is life-threatening viral diseases, such as AIDS, for which no satisfactory treatment is available to date. The therapeutic antibodies could be virusneutralizing antibodies, antibodies toxic for virusproducing cells or antibodies specific for cell surface molecules required for viral infection. Supporting the notion that such treatments can be efficiently applied to the curing of viral diseases, transgenic mice expressing a neutralizing antibody are protected from lethal infection by the lymphocytic choriomeningitis virus (Seiler et al., 1998). Also supporting the view of the potential utility of such treatments, it was recently shown that blocking the entry of HIV into target cells by administration of a short peptide (T20) can provide potent inhibition of HIV replication in patients suffering from AIDS (Kilby et al., 1998). In addition, several monoclonal antibodies with a neutralizing effect on HIV, including primary virus isolates, are already available and might be used for passive immunoprophylaxy of AIDS in the future (Table 2) (Burton, 1997; Burton and Montefiori, 1997). These antibodies are directed against the envelope glycoprotein subunits, gp120 and gp41, and have originally been characterised in in vitro inhibition assays. Some of them can even synergise for inhibiting HIV (Mascola et al., 1997) and SHIV (Li et al., 1998) replication. Finally, some of these antibodies are able to inhibit HIV replication in SCID mice grafted with human peripheral

blood lymphocytes (Burton et al., 1994; Gauduin et al., 1997). A third therapeutic application would be the treatment of certain chronic inflammatory diseases such as rheumatoid arthritis. In vitro experiments and recent clinical data have shown that TNF-! is a critical inflammatory mediator of this autoimmune disease and might therefore represent a molecular target for specific immunotherapy (Maini et al., 1995). Indeed, it has been shown that administration of anti-TNF-! monoclonal antibodies causes an improvement in the health of rheumatoid arthritis-suffering patients, thus providing evidence that such antibodies might represent efficacious drugs for long-term treatments of this disease (Elliott et al., 1994; Maini et al., 1995). However, the elevated doses necessary for obtaining therapeutic effects as well as their high cost still restrict the use of these antibodies on a large scale. Besides therapy, in vivo production of monoclonal antibodies may also have applications in the laboratory. Although the construction of transgenic mice could, most often, be envisaged to reach the same goal, genetic modification of somatic cells of animals or implantation of antibody-producing cells are expected to represent more versatile and less time-consuming techniques, especially when production of combinations of antibodies is desired. A first example of this application would be the development of new animal models of human autoimmune diseases in which the humoral immune response is responsible for, or contributes to, the development of the pathology (Rose and Bona, 1993). Another interesting application would be continuous cell type-specific ablation for studying the biological role of certain cell lineages in living animals. According to this approach, cytotoxic antibodies recognizing specific cell surface markers would be delivered continuously into the bloodstream of living animals where they could kill cells immediately upon appearance (for example, after a differentiation step) of the cognate antigen at their surface. A third application, called "phenotypic knock-out", could be the systemic delivery of antibodies neutralizing the activity of circulating antigens. Demonstrating the relevance of this approach, expression in the central nervous system of transgenic mice of a monoclonal antibody directed against substance P was able to inhibit the activity of this neuropeptide and was shown to be useful for studying the mechanisms of action of the latter (Piccioli et al., 1995).

IV. In vivo production of antibodies by genetically modified cells. Plasmocytes are the terminally differentiated cells of the B lineage which are responsible for the production and the release of antibodies into the bloodstream (Piccioli et al., 1995). Because of their short life-span (several days to few weeks), they cannot be used for long-term gene/cell therapy. Moreover, they already produce an immunoglobu-

T a b l e 1 . A n t i b o d y and a n t i b o d y - b a s e d m o l e c u l e s used for immunotherapy o f cancer. This list is not exhaustive. (°) corresponds to radiolabelled antibodies and (*) corresponds to immunotoxins. Details of clinical trials are to be found in the indicated references. Agent

Antigen

Disease

Reference

° [131 I ]-anti-B1 (mouse Mab)

CD20

B-cell lymphoma

Kaminsky et al., 1996; Press et al., 1995

° [90 Y]-anti-CD20 (mouse Mab)

CD20

B-cell lymphoma

Knox, 1996

° [90 Y]-anti-idiotype (mouse Mab)

Idiotype

B-cell lymphoma

White et al., 1996

* IgG HD37-dgA [deglycosylated ricin A] (mouse Mab)

CD19

B-cell lymphoma

Stone et al., 1996

* RF84-dgA [deglycosylated ricin A] (mouse Mab)

CD22

B-cell lymphoma

Amlot et al, 1993

IDEC-C2B8 (human-mouse chimeric Mab)

CD20

B-cell lymphoma

Maloney et al. 1997

M195 (mouse humanised Mab)

CD33

Acute Myeloid Leukemia

Caron et al., 1994; Jurcic et al. 1995

° [131 I]-M195 (mouse humanised Mab)

CD33

Acute Myeloid Leukemia

Jurcic et al., 1995

CAMPATH-1H (humanised Mab)

CDw52

Chronic lymphocytic leukemia

Osterborg et al., 1997

17-1A (mouse Mab)

Epithelial membrane antigen (EMA)

Colorectal carcinoma

Riethmuller et al., 1994

° [125 I]-A33 (murine Mab)

A33

Colorectal carcinoma

Welt et al., 1996; Daghighian et al., 1996

Anti-Le y B3-liked to Pseudomonas exotoxin (murine Mab)

Ley -Antigen

Colorectal carcinoma

Pai, 1996

MFE-23 (scFv antibody)

carcinoembryonic antigen (CEA)

Colorectal carcinoma

Begent et al., 1996

rhuMabHER (humanised Mab)

p185 HER2

Breast cancer

Baselga et al., 1996

° [131 I]-cG250 (human-mouse chimeric Mab)

G250

Renal carcinoma

Surfus et al., 1996; Steffens et al., 1997

lin, the expression of which might interfere with the production of the therapeutic antibody. It has long been known that several eukaryotic cell types such as yeast and certain insect cells in addition to certain mammalian cell lines can produce antibodies upon appropriate genetic modification (for references, see Noël et al., 1997). Interestingly, this observation raised the possibility that a variety of non-plasmocytic cells could be used for production of immunoglobulins. Indeed, we have recently shown that a number of cell types amenable to genetic modification and suitable for graft to humans can secrete antibodies (Noël et al., 1997). These cells include myogenic cells, hepatocytes, keratinocytes and skin

fibroblasts. It is, however, likely that their number will increase in the near future. Furthermore, geneticallymodified myogenic cells (Noël et al., 1997) and fibroblasts (unpublished results) grafted to mice were shown to be capable to sustain systemic delivery of cloned antibodies for several months (Figure 2). Importantly, the antibodies expressed ectopically in vitro and in vivo retained the specificity and the kinetic and thermodynamic characteristics of the parental antibody secreted by lymphocytic cells, as assayed using the BIAcore technology (Noël et al., 1997). These data indicate that several (and possibly all) non-B cell types possess the machinery requi-

Table 2. HIV-neutralizing human monoclonal antibodies. Details can be found in the references indicated. Agent

Antigen

In vitro neutralisation

In vivo neutralisation

R e fe r e nces

2F5

gp41 (linear amino acid sequence ELDKWA)

potent neutralisation of a broad range of primary isolates of HIV

delayed seroconversion and decrease in the viral load of chimpanzees infected with primay isolates

Muster et al, 1994 D’Souza et al, 1997 Conley et al, 1996

IgGb12

gp120 (epitope overlapping the CD4 binding domain and the V2 loop)

potent neutralisation of a broad range of primary isolates of HIV

inhibition of primary isolates of HIV in hu-PBL/SCID mice

Burton et al., 1994 Gaudin et al., 1997 Kessler et al., 1997

2G12

gp120 (epitope overlapping the V3 loop and the V4 region)

potent neutralisation of a broad range of primary isolates of HIV

Trkola, 1996 D’Souza et al, 1997

694/98D

gp120 (V3 loop)

neutralisation of several laboratory strains of HIV, activation of complement

Gorny et al., 1993 Spears et al., 1993

F105

gp120 (CD4 binding domain)

neutralisation of several laboratory strains and primary isolates of HIV

Posner et al., 1993

red for both production and correct folding of antibodies. So far, the production of antibodies by engineered cells has proved weak as compared to the production by cells of the B lineage. However, it is very likely that poor production results, not from the inability of the various cell types to make and secrete antibodies, but rather from poor expression of the retroviral vectors used for gene transfer. Improvement of the latter will thus constitute a major step towards efficient antibody-based gene therapy.

Figure 2. Systemic production of cloned antibodies through grafting of genetically m o d i f i e d m y o g e n i c c e l l s . Primary myoblasts are isolated from mouse muscle biopsies and expanded ex vivo. Following retroviral gene transfer of the cloned monoclonal antibody, stably transduced cells producing the antibody are selected and amplified for implantation into recipient mice. Myogenic cells are grafted by simple injection into the tibialis anterior muscle of mice treated with cardiotoxin. The antibody produced is released into the bloodstream (For more details, see Noël et al., 1997; Pelegrin et al., 1998).

Pelegrin et al: Engineering cells to produce therapeutic antibodies for immunotherapy of cancer

V. In vivo production of antibodies by encapsulated cells. Systemic production of antibodies in mice implanted with cells encapsulated into various biocompatible materials has been achieved by several groups (Okada et al., 1997; Pelegrin et al., 1998; Savelkoul et al., 1994). In the context of gene therapy, capsules are interesting for at least two reasons. First, they constitute immunoprotective devices since the size of their pores can be adjusted in order to allow the diffusion of small molecules (such as nutrients and antibodies) through them but can prevent the passage of cells. In other words, encapsulated cells, which are efficiently retained within capsules, are protected from immune cells of the host which cannot enter the matrix. This property is important with regard to the versatility of the capsule approach since non-autologous, or even xenogenic, cells can potentially be implanted into individuals (Figure 3 ) . Second, capsules offer an advantage with respect to safety since, in contrast to grafted cells, they can easily be removed by simple surgery if, for any reason, the treatment needs to be terminated.

Several types of polymers have been used to encapsulate antibody-producing cells for implantation into mice. These include cellulose sulphate (Dautzenberg et al., in press; Pelegrin et al., 1998), alginate (Savelkoul et al., 1994) and alginate-poly(L)lysine-alginate (Okada et al., 1997). The various matrices used differ in their physical and mechanical properties with cellulose sulphate (Dautzenberg et al., in press) offering advantages over the other two which either rapidly deteriorate or induce an inflammatory response, respectively. Interestingly, cellulose sulphate capsules (Figure 4) implanted subcutaneously form neoorgans which are extensively vascularized within days and are stable for at least 10 months (Figure 5) (Pelegrin et al., 1998). This is certainly beneficial for two reasons. Antibody uptake by the blood is favored and a better supply of nutrients is achieved, thus favoring cell survival in the capsules. Alternatively, other biocompatible immunoprotective devices, such as polyethersulfone fibers, might be used to replace capsules for implantation of antibody-producing cells in vivo (DĂŠglon et al., 1996). So far, only encapsulation of cells with short life-span within capsules, such as hybridoma cells, has been tested for transient antibody production in vivo. It will thus be crucial to test whether long-lived primary cells or cell lines can also be used for long-term production. This seems possible since primary skin fibroblasts have already been shown to survive longer than one year in vivo when encapsulated in alginate-poly-L-lysine alginate (Tai and Sun, 1993). Work is currently underway to address this issue.

VI. Overcoming some of the possible hurdles. The possible development of an immune response against the ectopic antibody and/or the antibody-producing cells is a major threat for this therapeutic strategy. This response can thus potentially be cellular and/or humoral. In case of grafting engineered autologous cells, a cytotoxic response against antibody-producing cells is very unlikely to occur; this is because secreted antibodies are not foreign molecules, provided they are of human origin or of the species in which the experiments are being conducted. However, it cannot yet be ruled out that ectopic

Figure 3. Systemic production of antibodies by implantation of encapsulated antibody-producing c e l l s . Established cell lines validated for human use can be genetically modified to produce therapeutic antibodies upon gene transfer. Selected antibody-producing cells can be amplified and encapsulated in immunoprotective devices (see text and Dautzenberg et al., in press). Capsules are implanted subcutaneously by simple surgical treatment for systemic delivery of therapeutic antibodies.

172

F i g u r e 4 . C e l l u l o s e s u l p h a t e c a p s u l e s . A . Production of cellulose sulphate capsules. Cells are resuspended in a cellulose sulphate solution. Droplets of the suspension are generated and dropped into a solution of polymerization catalyst (PDADMAC). Capsules form within 90 second. After washing with the appropriate medium, they can be implanted immediately or kept in culture for several days to several weeks before use (Dautzenberg et al., in press). B . Cellulose sulphate capsules containing hybridoma cells. These capsules have an average diameter of 0.6 mm. The dark zones correspond to encapsulated cells

Figure 5. Neo-organ formation following implantation of e n c a p s u l a t e d c e l l s . Cellulose sulphate capsules containing antibody-producing cells are vascularised within a few days when implanted subcutaneously (Pelegrin et al., 1998). In this experiment, a group of 10 capsules (C) was implanted. Within 3 days they were wrapped in a pouch of loose connective tissue (CT) which rapidly underwent peripheral vascularization (PV). Later, blood vessels extended into the inner part (IV ) of this pouch for irrigation of the neo-organ.

antibodies can be degraded producing antigenic peptides presentable by MHC class I molecules when expressed in non-B cells. If this occurs a cytotoxic T cell response against such cells could be triggered. This issue merits thorough analysis. The situation is quite different in the case of capsule implantation : even if encapsulated cells are non-MHCmatched or xenogenic, they cannot be destroyed by host cytotoxic T cells because no physical contact is allowed between the two types of cells. However, xenogenic cells are sometimes killed by a mechanism involving complement-mediated lysis. An appropriate choice of xenogenic cells and/or adapted strategies for protecting cells from complement will, thus, be necessary. It is likely that cellular debris released from the capsules could trigger a cytotoxic T cell response directed against encapsulated cells. However, more than being a drawback, this response should present an advantage with respect to the safety: in case of accidental escape from capsules (after breakage, for example), antibody-producing cells released into the bloodstream would immediately be destroyed by circulating T cells. A more serious threat is the possible generation of humoral anti-idiotypic responses against therapeutic antibodies. Such immune responses were observed in patients treated with repeated injections of high doses of purified antibodies. Sometimes, but not always, they could even neutralize the treatment (Isaacs, 1990). It must, however, be emphasized, that in no clinical trial performed so far, were the antibodies of human origin : at best, they were humanized murine antibodies. Moreover, it is not clear whether the observed anti-idiotypic responses were primary responses against the idiotypes of the injected antibodies or just parts of responses directed against whole non-self proteins. In contrast with these observations, no detectable antiidiotypic response was observed in mice producing a model anti-human thyroglobulin monoclonal antibody upon grafting engineered myogenic cells (unpublished data) or upon implantation of cellulose sulphate capsules containing hybridoma cells (Pelegrin et al., 1998). Even though these first data are encouraging, such studies need to be extended to a number of other immunoglobulins to establish whether ectopic monoclonal antibodies produced in vivo are immunogenic or not. It is also possible that the concentration of antibody released systemically is crucial in triggering anti-idiotypic responses. In this case, determining the threshold levels of antibody required for the mounting of the immune response will be crucial for developing efficient long-term antibody-based gene therapies. Inducible expression systems, such as the tetracycline system of Bujard and co-workers, might reveal invaluable tools for adjusting the concentration of antibody delivered into the bloodstream of patients.

VII. Conclusions Using model systems, we have demonstrated the feasibility of the in vivo production and systemic delivery of antibodies by engineered cells. Our work thus sets up the technical basis for a new gene/cell therapy approach aimed at the long-term treatment of patients suffering from a variety of severe diseases such as cancer, viral diseases and various autoimmune diseases. The two major issues which must now be solved before application of this novel therapeutical strategy to humans are, (i ) the optimization of the in vivo production of antibodies and, (i i ), the validation of its therapeutical value in several animal models of human diseases. For optimization in antibody production, several approaches have already been considered. The use of cell lines certified for human use and amenable to encapsulation certainly constitutes a promising approach for various reasons including efficiency, cost-effectiveness and safety. Nevertheless, using in vivo injectable vectors, such as adenoviruses and AAV, for long-term production of monoclonal antibodies in vivo is also a promising approach. Finally, we have recently been able to protect mice from developing a lethal retroviral disease using systemic delivery of antibodies by antibody-producing cells, thus providing the first demonstration of the therapeutical potential of the approach. Extension of this study to other animal diseases is currently under investigation and should pave the way to human applications.

Acknowledgments. This work was supported by grants from the Centre National de la Recherche Scientifique, the Ligue Nationale contre le Cancer, the Association de Recherche contre le Cancer (ARC), the Agence Nationale de Recherche contre le Sida (ANRS) and the EC Biotech Program. We are grateful to Anna Oates for careful correction of the manuscript.

References. Amlot, P. L., Stone, M. J., Cunningham, D., Fay, J., Newman, J., Collins, R., May, R., McCarthy, M., Richardson, J., Ghetie, V., and et, a. l. (1 9 9 3 ). A phase I study of an anti-CD22-deglycosylated ricin A chain immunotoxin in the treatment of B-cell lymphomas resistant to conventional therapy. B l o o d 82, 2624-33. Baselga, J., Tripathy, D., Mendelsohn, J., Baughman, S., Benz, C. C., Dantis, L., Sklarin, N. T., Seidman, A. D., Hudis, C. A., Moore, J., Rosen, P. P., Twaddell, T., Henderson, I. C., and Norton, L. (1 9 9 6 ). Phase II study of weekly intravenous recombinant humanized antip185HER2 monoclonal antibody in patients with HER2/neu-overexpressing metastatic breast cancer [see comments]. J C l i n O n c o l 14, 737-44. Begent, R. H., Verhaar, M. J., Chester, K. A., Casey, J. L., Green, A. J., Napier, M. P., Hope, S. L., Cushen, N.,

Gene Therapy and Molecular Biology Vol 3, page 175 Keep, P. A., Johnson, C. J., Hawkins, R. E., Hilson, A. J., and Robson, L. (1 9 9 6 ). Clinical evidence of efficient tumor targeting based on single-chain Fv antibody selected from a combinatorial library. Nat Med 2, 97984. Bruggemann, M., and Taussig, M. J. (1 9 9 7 ). Production of human antibody repertoires in transgenic mice. Curr. O p i n . B i o t e c h n o l . 8, 455-458. Burton, D. R. (1 9 9 7 ). A vaccine for HIV type 1 : the antibody perspective. P r o c . N a t l . A c a d . S c i . U S A 94, 1001810023. Burton, D. R., and Montefiori, D. C. (1 9 9 7 ). The antibody response in HIV-1 infection [see comments]. Aids, S8798. Burton, D. R., Pyati, J., Koduri, R., Sharp, S. J., Thornton, G. B., Parren, P. W., Sawyer, L. S., Hendry, R. M., Dunlop, N., Nara, P. L., and et, a. l. (1 9 9 4 ). Efficient neutralization of primary isolates of HIV-1 by a recombinant human monoclonal antibody. S c i e n c e 266, 1024-7. Caron, P. C., Jurcic, J. G., Scott, A. M., Finn, R. D., Divgi, C. R., Graham, M. C., Jureidini, I. M., Sgouros, G., Tyson, D., Old, L. J., and et, a. l. (1 9 9 4 ). A phase 1B trial of humanized monoclonal antibody M195 (antiCD33) in myeloid leukemia: specific targeting without immunogenicity. B l o o d 83, 1760-8. Chester, K. A., and Hawkins, R. E. (1 9 9 5 ). Clinical issues in antibody design. T r e n d s B i o t e c h n o l 13, 294-300. Conley, A. J., Kessler, J. I., Boots, L. J., McKenna, P. M., Schleif, W. A., Emini, E. A., Mark, G. I., Katinger, H., Cobb, E. K., Lunceford, S. M., Rouse, S. R., and Murthy, K. K. (1 9 9 6 ). The consequence of passive administration of an anti-human immunodeficiency virus type 1 neutralizing monoclonal antibody before challenge of chimpanzees with a primary virus isolate. J V i r o l 70, 6751-8. D'Souza, M. P., Livnat, D., Bradac, J. A., and Bridges, S. H. (1 9 9 7 ). Evaluation of monoclonal antibodies to human immunodeficiency virus type 1 primary isolates by neutralization assays: performance criteria for selecting candidate antibodies for clinical trials. AIDS Clinical Trials Group Antibody Selection Working Group. J I n f e c t D i s 175, 1056-62. Daghighian, F., Barendswaard, E., Welt, S., Humm, J., Scott, A., Willingham, M. C., McGuffie, E., Old, L. J., and Larson, S. M. (1 9 9 6 ). Enhancement of radiation dose to the nucleus by vesicular internalization of iodine-125labeled A33 monoclonal antibody. J Nucl Med 37, 1052-7. Dautzenberg, H., Schuldt, U., Grasnick, G., Karle, P., Müller, P., Löhr, M., Pelegrin, M., Piechaczyk, M., vonRombs, K., Günzburg, W., Salmons, B., and Saller, R. (i n p r e s s ). Development of cellulose sulphate-based polyelectrolyte complex microcapsules for medical applications. A n n . N . Y . A c a d . S c i . Déglon, N., Heyd, B., Tan, S. A., Joseph, J. M., Zurn, A. D., and Aebischer, P. (1 9 9 6 ). Central nervous system delivery of recombinant ciliary neurotrophic factor by polymer encapsulated differentiated C2C12 myoblasts. Human Gene Ther. 7, 2135-2146.

175

Elliott, M. J., Maini, R. N., Feldmann, M., Kalden, J. R., Antoni, C., Smolen, J. S., Leeb, B., Breedveld, F. C., Macfarlane, J. D., Bijl, H., and et, a. l. (1 9 9 4 ). Randomised double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha (cA2) versus placebo in rheumatoid arthritis. Lancet 344, 1105-10. Elliott, M. J., Maini, R. N., Feldmann, M., Long, F. A., Charles, P., Bijl, H., and Woody, J. N. (1 9 9 4 ). Repeated therapy with monoclonal antibody to tumour necrosis factor alpha (cA2) in patients with rheumatoid arthritis. Lancet 344, 1125-7. Gauduin, M. C., Parren, P. W., Weir, R., Barbas, C. F., Burton, D. R., and Koup, R. A. (1 9 9 7 ). Passive immunization with a human monoclonal antibody protects hu-PBL- SCID mice against challenge by primary isolates of HIV-1. Nat Med 3, 1389-93. Gorny, M. K., Xu, J. Y., Karwowska, S., Buchbinder, A., and Zolla, P. S. (1 9 9 3 ). Repertoire of neutralizing human monoclonal antibodies specific for the V3 domain of HIV1 gp120. J Immunol 150, 635-43. Isaacs, J. D. (1 9 9 0 ). The antiimmunoglobulin response to therapeutic antibodies. S e m i n . I m m u n o l 2, 449-56. Jurcic, J. G., Caron, P. C., Miller, W. J., Yao, T. J., Maslak, P., Finn, R. D., Larson, S. M., Warrell, R. J., and Scheinberg, D. A. (1 9 9 5 ). Sequential targeted therapy for relapsed acute promyelocytic leukemia with all-trans retinoic acid and anti-CD33 monoclonal antibody M195. Leukemia 9, 244-8. Jurcic, J. G., Caron, P. C., Nikula, T. K., Papadopoulos, E. B., Finn, R. D., Gansow, O. A., Miller, W. J., Geerlings, M. W., Warrell, R. J., Larson, S. M., and et, a. l. (1 9 9 5 ). Radiolabeled anti-CD33 monoclonal antibody M195 for myeloid leukemias. Cancer Res 55, 5908s-5910s. Kaminski, M. S., Zasadny, K. R., Francis, I. R., Fenner, M. C., Ross, C. W., Milik, A. W., Estes, J., Tuck, M., Regan, D., Fisher, S., Glenn, S. D., and Wahl, R. L. (1 9 9 6 ). Iodine-131-anti-B1 radioimmunotherapy for Bcell lymphoma. J C l i n O n c o l 14, 1974-81. Kessler, J. n., McKenna, P. M., Emini, E. A., Chan, C. P., Patel, M. D., Gupta, S. K., Mark, G. r., Barbas, C. r., Burton, D. R., and Conley, A. J. (1 9 9 7 ). Recombinant human monoclonal antibody IgG1b12 neutralizes diverse human immunodeficiency virus type 1 primary isolates. Aids Res Hum Retroviruses 13, 575-82. Kilby, J. M., Hopkins, S., Venetta, T. M., DiMassimo, B., Cloud, G. A., Lee, J. Y., Alldredge, L., Hunter, E., Lambert, D., Bolognesi, D., Matthews, T., Johnson, M. R., Nowak, M. A., Shaw, G. M., and Saag, M. S. (1 9 9 8 ). Potent suppression of HIV-1 replication in humans by T20, a peptide inhibitor of gp41-mediated virus entry. Nat Med 4, 1302-7. Knox, S. J., Goris, M. L., Trisler, K., Negrin, R., Davis, T., Liles, T. M., Grillo-Lopez, A., Chinn, P., Varns, C., and Ning, S. C. (1 9 9 6 ). Yttrium-90-labelled anti-CD20 monoclonal antibody therapy of recurrent B-cell lymphoma. Clin Cancer Res 2, 457-470. Li, A., Katinger, H., Posner, M. R., Cavacini, L., Zolla, P. S., Gorny, M. K., Sodroski, J., Chou, T. C., Baba, T. W., and Ruprecht, R. M. (1 9 9 8 ). Synergistic neutralization of simian-human immunodeficiency virus SHIV- vpu+ by

Pelegrin et al: Engineering cells to produce therapeutic antibodies for immunotherapy of cancer triple and quadruple combinations of human monoclonal antibodies and high-titer anti-human immunodeficiency virus type 1 immunoglobulins. J Virol 72, 3235-40. Maini, R. N., Elliott, M. J., Brennan, F. M., and Feldmann, M. (1 9 9 5 ). Beneficial effects of tumour necrosis factoralpha (TNF-alpha) blockade in rheumatoid arthritis (RA) [published erratum appears in Clin Exp Immunol 1995 Nov;102(2):443]. Clin Exp Immunol 101, 207-12. Maini, R. N., Elliott, M. J., Brennan, F. M., Williams, R. O., Chu, C. Q., Paleolog, E., Charles, P. J., Taylor, P. C., and Feldmann, M. (1 9 9 5 ). Monoclonal anti-TNF alpha antibody as a probe of pathogenesis and therapy of rheumatoid disease. Immunol Rev 144, 195-223. Maloney, D. G., Grillo, L. A., White, C. A., Bodkin, D., Schilder, R. J., Neidhart, J. A., Janakiraman, N., Foon, K. A., Liles, T. M., Dallaire, B. K., Wey, K., Royston, I., Davis, T., and Levy, R. (1 9 9 7 ). IDEC-C2B8 (Rituximab) anti-CD20 monoclonal antibody therapy in patients with relapsed low-grade non-Hodgkin's lymphoma. B l o o d 90, 2188-95. Marks, C., and Marks, J. D. (1 9 9 6 ). Phage libraries - a new route to clinically useful antibodies. N ew . E n g l . J . Med. 335, 730-733. Marwick, C. (1 9 9 7 ). Monoclonal antibody lymphoma [news]. Jama 278, 616, 618.

to

treat

Mascola, J. R., Louder, M. K., VanCott, T. C., Sapan, C. V., Lambert, J. S., Muenz, L. R., Bunow, B., Birx, D. L., and Robb, M. L. (1 9 9 7 ). Potent and synergistic neutralization of human immunodeficiency virus (HIV) type 1 primary isolates by hyperimmune anti-HIV immunoglobulin combined with monoclonal antibodies 2F5 and 2G12. J Virol 71, 7198-206. Mendez, M. J., Green, L. L., Corvalan, J. R. F., Jia, X. C., Maynard-Curie, C. A., Yang, X. D., Gallo, M. L., Louie, D. M., Lee, D. V., Erickson, K. L., Luna, J., Roy, C. M., Abderrahim, H., Kirschenbaum, F., Nogushi, M., Smith, D. H., Fukushima, A., Hales, J. F., Finer, M. H., Zebo, C. G. D. K. M., and Jakobovitz, A. (1 9 9 7 ). Functional transplant of megabase human immunoglobulin loci recapitulates human antibody in mice. Nature G e n e t . 15, 146-156. Muster, T., Guinea, R., Trkola, A., Purtscher, M., Klima, A., Steindl, F., Palese, P., and Katinger, H. (1 9 9 4 ). Crossneutralizing activity against divergent human immunodeficiency virus type 1 isolates induced by the gp41 sequence ELDKWAS. J Virol 68, 4031-4. Noël, D., Pelegrin, M., Marin, M., Biard, P. M., Ourlin, J. C., Mani, J. C., and Piechaczyk, M. (1 9 9 7 ). In vitro and in vivo secretion of cloned antibodies by genetically modified myogenic cells. H u m . G e n e T h e r . 8, 121929. Okada, N., Miyamoto, H., Yoshioka, T., Sakamoto, K., Katsume, A., Saito, H., Nakagawa, S., Ohsugi, Y., and Mayumi, T. (1 9 9 7 ). Immunological studies of SK2 hybridoma cells microencapsulated with alginatepoly(L)lysine-alginate (APA) membrane following allogeneic transplantation. B i o c h e m B i o p h y s R e s Commun 230, 524-7. Old, L. J. (1 9 9 5 ). Immunotherapy for cancer. S c i . A m . Sept, 136-143.

176

Osterborg, A., Dyer, M. J., Bunjes, D., Pangalis, G. A., Bastion, Y., Catovsky, D., and Mellstedt, H. (1 9 9 7 ). Phase II multicenter study of human CD52 antibody in previously treated chronic lymphocytic leukemia. European Study Group of CAMPATH-1H Treatment in Chronic Lymphocytic Leukemia. J C l i n O n c o l 15, 1567-74. Pai, L. H., Wittes, R., Setser, A., Willingham, M. C., and Pastan, I. (1 9 9 6 ). Treatment of advanced solid tumors with immunotoxin LMB-1: an antibody linked to Pseudomonas exotoxin. Nat Med 2, 350-3. Pelegrin, M., Marin, M., Noël, D., and Piechaczyk, M. (1 9 9 8 ). Genetically engineered antibodies in gene transfer and gene therapy. Hum. Gene Ther. 9, 21652175. Pelegrin, M., Marin, M., Noël, D., DelRio, M., Saller, R., Stange, J., Mitzner, S., Günzburg, W. H., and Piechaczyk, M. (1 9 9 8 ). Systemic long-term delivery of antibodies in immunocompetent animals using cellulose sulphate capsules containing antibody-producing cells. Gene Ther. 5, 828-834. Piccioli, P., DiLuzio, A., Amann, R., Schuligoi, R., Surani, M. A., Donnerer, J., and Cattaneo, A. (1 9 9 5 ). Neuroantibodies : ectopic expression of a recombinant anti-substance P antibody in the central nervous system of transgenic mice. Neuron 15, 373-384. Posner, M. R., Cavacini, L. A., Emes, C. L., Power, J., and Byrn, R. (1 9 9 3 ). Neutralization of HIV-1 by F105, a human monoclonal antibody to the CD4 binding site of gp120. J Acquir Immune Defic Syndr 6, 7-14. Press, O. W., Eary, J. F., Appelbaum, F. R., Martin, P. J., Nelp, W. B., Glenn, S., Fisher, D. R., Porter, B., Matthews, D. C., Gooley, T., and et, a. l. (1 9 9 5 ). Phase II trial of 131I-B1 (anti-CD20) antibody therapy with autologous stem cell transplantation for relapsed B cell lymphomas. Lancet 346, 336-40. Rader, C., and Barbas, C. F. (1 9 9 7 ). Phage display of combinatorial antibody libraries. Curr. O p i n . B i o t e c h n o l . 8, 503-508. Riethmuller, G., Holz, E., Schlimok, G., Schmiegel, W., Raab, R., Hoffken, K., Gruber, R., Funke, I., Pichlmaier, H., Hirche, H., Buggisch, P., Witte, J., and Pichlmayr, R. (1 9 9 8 ). Monoclonal antibody therapy for resected Dukes' C colorectal cancer: seven-year outcome of a multicenter randomized trial. J C l i n O n c o l 16, 1788-94. Riethmüller, G., Schneider-Gädicke, E., and Johnson, J. P. (1 9 9 3 ). Monoclonal antibodies in cancer therapy. Current Opinion Immunol. 5, 732-739. Rose, N. R., and Bona, C. (1 9 9 3 ). Defining criteria for autoimmune diseases (Witebsky's postulates revisited). Immunol. Today 14, 426-430. Savelkoul, H. F., R., v., Vossen, A. C., Breedland, E. G., Coffman, R. L., and vanOudenaren, A. (1 9 9 4 ). Modulation of systemic cytokine levels by implantation of alginate encapsulated cells. J . I m m u n o l . M e t h o d s 170, 185-96. Scott, A. M., and Welt, S. (1 9 9 7 ). Antibody-based immunological therapies. Curr. Op. I m m u n o l . 9, 717-722.

Gene Therapy and Molecular Biology Vol 3, page 177 Seiler, P., Kalinke, U., Rulicke, T., Bucher, E. M., Bose, C., Zinkernagel, R. M., and Hengartner, H. (1 9 9 8 ). Enhanced virus clearance by early inducible lymphocytic choriomeningitis virus-neutralizing antibodies in immunoglobulin-transgenic mice. J . V i r o l . 72, 22532258. Spear, G. T., Takefman, D. M., Sullivan, B. L., Landay, A. L., and Zolla, P. S. (1 9 9 3 ). Complement activation by human monoclonal antibodies to human immunodeficiency virus. J Virol 67, 53-9. Steffens, M. G., Boerman, O. C., Oosterwijk, W. J., Oosterhof, G. O., Witjes, J. A., Koenders, E. B., Oyen, W. J., Buijs, W. C., Debruyne, F. M., Corstens, F. H., and Oosterwijk, E. (1 9 9 7 ). Targeting of renal cell carcinoma with iodine-131-labeled chimeric monoclonal antibody G250. J C l i n O n c o l 15, 1529-37. Stone, M. J., Sausville, E. A., Fay, J. W., Headlee, D., Collins, R. H., Figg, W. D., Stetler, S. M., Jain, V., Jaffe, E. S., Solomon, D., Lush, R. M., Senderowicz, A., Ghetie, V., Schindler, J., Uhr, J. W., and Vitetta, E. S. (1 9 9 6 ). A phase I study of bolus versus continuous infusion of the anti-CD19 immunotoxin, IgG-HD37-dgA, in patients with B-cell lymphoma. B l o o d 88, 1188-97. Surfus, J. E., Hank, J. A., Oosterwijk, E., Welt, S., Lindstrom, M. J., Albertini, M. R., Schiller, J. H., and Sondel, P. M. (1 9 9 6 ). Anti-renal-cell carcinoma chimeric antibody G250 facilitates antibody- dependent cellular cytotoxicity with in vitro and in vivo interleukin-2- activated

177

effectors. J Immunother Immunol 19, 184-91.

Emphasis

Tumor

Tai, I. T., and Sun, A. M. (1 9 9 3 ). Microencapsulation of recombinant cells: a new delivery system for gene therapy. F a s e b J . 7, 1061-9. Trkola, A., Purtscher, M., Muster, T., Ballaun, C., Buchacher, A., Sullivan, N., Srinivasan, K., Sodroski, J., Moore, J. P., and Katinger, H. (1 9 9 6 ). Human monoclonal antibody 2G12 defines a distinctive neutralization epitope on the gp120 glycoprotein of human immunodeficiency virus type 1. J Virol 70, 1100-8. Vitetta, E. S., and Uhr, J. W. (1 9 9 4 ). Monoclonal antibodies as agonists: an expanded role for their use in cancer therapy. Cancer Res 54, 5301-9. Wawrzynczak, E. J. (1 9 9 5 ). A n t i b o d y t h e r a p y (Oxford, UK: BIOS Scientific Publishers Ltd). Welt, S., Scott, A. M., Divgi, C. R., Kemeny, N. E., Finn, R. D., Daghighian, F., Germain, J. S., Richards, E. C., Larson, S. M., and Old, L. J. (1 9 9 6 ). Phase I/II study of iodine 125-labeled monoclonal antibody A33 in patients with advanced colon cancer. J C l i n O n c o l 14, 1787-97. White, C. A., Halpern, S. E., Parker, B. A., Miller, R. A., Hupf, H. B., Shawler, D. L., Collins, H. A., and Royston, I. (1 9 9 6 ). Radioimmunotherapy of relapsed B-cell lymphoma with yttrium 90 anti- idiotype monoclonal antibodies. B l o o d 87, 3640-9.

Gene Therapy and Molecular Biology Vol 3, page 179 Gene Ther Mol Biol Vol 3, 179-187. August 1999.

Use of DNA priming and vaccinia virus boosting to trigger an efficient immune response to HIV-1 gp120 Research Article

Dolores Rodríguez1, Juan Ramón Rodríguez 1, Mercedes Llorente2, Pilar Lucas2, Mariano Esteban1,3, Carlos Martínez-A.2 and Gustavo del Real2 1

2

Departments of Molecular & Cellular Biology and Immunology & Oncology , Centro Nacional de Biotecnología, CSIC, Campus Universidad Autónoma, E-28049 Madrid, Spain __________________________________________________________________________________________________ 3

Corresponding author: Tel: (+34) 91-585-4503; Fax: (+34) 91-585-4506; E-mail: mesteban@cnb.uam.es Received: 15 October 1998; accepted: 25 October 1998

Summary To enhance the efficiency of DNA vaccines to HIV-1, we immunized BALB/c mice sequentially with a gp120 DNA vector and a recombinant vaccinia virus (VV). We have also evaluated the effect of granulocyte macrophage colony stimulation factor (GM-CSF) expression by a DNA vector on both cellular and humoral immune responses when coadministered with the gp120-encoding DNA at priming. Our results show a significant enhancement of both arms of the immune system when the DNA prime/VV boost regime i s used, as compared with the immunization protocol based on priming and boosting with vector DNA. A 100-fold increase in the number of antigen-specific IFN-secreting CD8 + T c e l l s was observed i n splenocyte cultures from mice immunized with the combined vec tor DNA/VV protocol. The humoral immune response is also improved in animals receiving the vector DNA/VV combined vaccine, as shown by the increase in both env-specific antibody titers and HIV-1 neutralizing activity in sera. IgG1 was the predominant isotype detected i n sera from the immunized animals. T h i s , together with the IL-4 and IFN- production in splenocyte cultures from these animals, indicated that both Th1 and Th2 responses are induced by the combined immunization approach. Coadministration of a GM-CSF-expressing DNA vector in the priming step resulted in enhanced T cell proliferation rates, irrespective of whether the booster w a s g i v e n w i t h v e c t o r D N A o r r e c o m b i n a n t V V . I n a d d i t i o n , a s l i g h t i n c r e a s e i n the humoral immune response was also observed i n animals primed with gp120 and GM-CSF-expressing plasmid and boosted with recombinant V V . These findings describe a combinatorial priming/booster immunization approach that may be effective in the control of HIV-1 infection and of other pathogens.

contains epitopes that activate T cell memory, and against which neutralizing antibodies as well as anti-HIV-1 cytotoxic T cells (CTL) are directed, vaccination with recombinant env glycoprotein induces only modest antiHIV-1 CTL activity. DNA immunization represents a novel approach to vaccine development and immunotherapy aimed toward immune response control. DNA vaccination induces antigen-specific cellular and humoral immune responses through the delivery of non-replicating transcription units, which drive the synthesis of specific foreign proteins within the inoculated host. Inoculation with plasmid DNA has been shown to be protective in several viral disease

I. Introduction Since the mid-1980’s, when HIV was first isolated from patients, a variety of vaccine constructs and vaccination strategies have been explored to combat this virus, although none has yet been demonstrated to be effective in preventing in vivo HIV infection. Effective vaccines must stimulate the correct balance between cellular (Th1-driven) and humoral (Th2-driven) immune responses according to the particular infectious agent; strategies aimed toward triggering a both humoral and cellular responses are thus receiving considerable attention. Although the HIV-1 envelope (env) glycoprotein 179

RodrĂ­guez et al: Immunostimulation by HIV-1 env-GM-CSF models (Tighe et al. 1998). DNA vaccines are simple to manufacture, provide prolonged antigen expression, and allow manipulation of protein antigenicity at the cDNA level with no need for protein production and purification. Unlike conventional protein vaccines, plasmid DNA vaccination leads to antigen processing and presentation by MHC class I and class II molecules and thus more closely resembles a viral infection. DNA vectors are particularly effective in inducing a strong MHC class I-restricted CTL response. Virus-specific CTL have been induced by gene vaccination with plasmid DNA encoding influenza (Raz et al., 1994) and HIV-1 viral proteins (Fuller et al., 1994); high CTL activity levels have been detected more than one year after intradermal gene vaccination with plasmid DNA encoding influenza virus nucleoptrotein (Raz et al., 1994). Intradermal gene vaccination in mice induces antigenspecific Th1 cells secreting high IFN-! levels, and stimulates IgG2a isotype antibody production (Raz et al., 1996; Sato et al., 1996). The safety, feasibility and immunogenicity of DNA vaccines are currently under evaluation in clinical trials. Immunization with DNA coding for HIV-1 gp160 elicits neutralizing antibodies and CTL responses in mice and in primates (Wang et al., 1995), and HIV-1-infected chimpanzees immunized with DNA expressing env substantially decreased the HIV-1 viral load (Ugen et al., 1997). In general, the antibody response induced by gene vaccination is lower than that induced by protein vaccination, possibly due to the minute amount of gene product produced in vivo. The overall potency of naked DNA vaccines is less than that of recombinant vaccines; since DNA does not undergo replicative amplification as occurs with live recombinant viral and bacterial vaccines, the amount of antigen ultimately presented to the immune system may be limited. Although DNA vaccination appears to be a promising strategy, it may nonetheless require the use of immunomodulators, adjuvants such as cytokines, or specific immunization regimes to enhance immune responses. Several groups have recently shown enhancement of the response triggered by DNA vaccination by coinjection of cytokine-encoding plasmids (Xiang & Ertl, 1995). Codelivery of vectors encoding a cytokine such as GMCSF can augment antigen-specific responses following either intramuscular or epidermal plasmid DNA delivery (Xiang & Ertl, 1995). We have recently demonstrated that a recombinant vaccinia virus (VV) expressing a chimeric GM-CSF/gp120 fusion protein induces a significant increase in cellular immune responses against the HIV-1 env antigen (Rodriguez et al., 1998). It has also been shown that GM-CSF, given at priming in the form of a DNA/GM-CSF chimeric vaccine, increases the magnitude of the anamnestic response, irrespective of the antigen form used in the subsequent booster immunization (Gerloni et al., 1998). GM-CSF enhances viability and function of dendritic cells (DC) (Heufler et al., 1998; Witmer et al., 1987), activates cells of the dendritic lineage in vitro (Inaba et al., 1992), and potentiates antigen T a b l e 1 . Immunization regimes of mice

180

presentation in vivo (Disis et al., 1996). GM-CSF is neither a Th1 nor a Th2 cytokine and can exert its adjuvant effect without skewing the Th1/Th2 balance. We previously described a vaccination protocol based on priming with influenza virus expressing a CD8+ T cell epitope of circumsporozoite (CS) protein from Plasmodium yoelii, followed by boosting with a VV recombinant expressing the entire CS protein which induces high protection levels against malaria in the murine model system (Li et al., 1993; Rodrigues et al., 1994). Other recent experiments in mice have shown that priming with DNA and boosting with recombinant vaccinia virus (VV) expressing a circumsporozoite protein is associated with high immunogenicity and protective efficacy against P. berghei (Sedegah et al., 1998). Delivering recombinant VV in the boost rather than in the priming dose may elude development of host immunity to the viral portion of the immunogen, which would reduce the response to the booster. The critical event in priming may depend on the processing properties of the immunogen expressed (influenza or DNA better than VV), whereas the secondary response is dependent on the quantity of immunogen expressed (VV better than influenza or DNA). To improve the efficiency of a HIV-1 vaccine, we have implemented a vaccination regime consisting of priming with DNA vectors expressing gp120 and GM-CSF, followed by boosting with a recombinant VV expressing env. We also present compelling data that support the enhancement of both humoral as well as a cytotoxic T cell response by priming with DNA plasmids containing the appropriate antigen plus GM-CSF and boosting with a recombinant vaccinia virus containing the same antigen as that used for the initial immunization.

II. Results A. Enhanced humoral immune response to HIV-1 gp120 by the priming/booster approach: vector DNA/VV recombinant We have attempted to develop new strategies for triggering efficient immune responses by integrating current procedures with new techniques, using combinations of plasmids and live viruses. More specifically, we focused on the immune response to the HIV-1 env protein, using a DNA plasmid containing the env gp120 gene (pCdm7) to immunize mice intradermically; we then analyzed the antibody response to HIV-1 gp120. In a parallel set of experiments, we compared this response to that of mice injected intraperitoneally with a recombinant vaccinia virus encoding HIV-1 gp120. Finally, given the increase in immunogenicity of GM-CSF-linked antigens, we coimmunized mice with a plasmid DNA vector encoding murine GM-CSF. Balb/c mice were primed and boosted with combinations of DNA vectors expressing HIV-1 gp120 or

Gene Therapy and Molecular Biology Vol 3, page 181

pgp (50 µg) prime

boost

pgp (50 µg) i.d.

pgp (50 µg)

pGM (50 µg)

pgp (50 µg)

pGM (50 µg)

pGM (50 µg)

i.d.

i.d.

i.d.

i.d.

VVenv

VVenv

VVenv

pgp (50µg)

pgp (50 µg)

(5 x 107 pfu)

(5 x 107 pfu)

(5 x 107 pfu)

i.d.

i.d.

i.p.

i.p.

i.p.

Injections were intradermic (i.d.) for plasmid DNA or intraperitoneal (i.p.) for vaccinia virus. Booster injections were given two weeks after priming.

F i g u r e 1 . Humoral anti-HIV-1 gp120 immune response elicited in mice primed with vector DNA and boosted with DNA or recombinant VV. BALB/c mice (four per group) were primed by intradermal injection with the plasmid expressing gp120 (pgp), with the plasmid expressing GM-CSF (pGM) or with a combination of both (pgp+pGM). After 21 days, mice were boosted with pgp or with the recombinant VV-env as indicated in Table 1, and serum samples collected two weeks after the boost. (A) ELISA analysis of anti-gp120 reactivity in pooled serum samples. ( B ) ELISA determination of anti-gp120 antibody isotypes in pooled serum samples.

181

Gene Therapy and Molecular Biology Vol 3, page 182 unimmunized mice or mice immunized in conditions producing low antibody titers display no neutralizing capacity. In contrast, we found potent neutralizing activity in sera from mice primed with gp120-DNA+GM-CSFDNA and boosted with VV-env. Less potent antiviral activity was observed in sera from gp120-DNA/VV-envimmunized mice, and activity was insignificant in DNA/DNA-immunized and control mice.

GM-CSF and VV recombinants expressing env (Table 1). Serum samples were collected from mice two weeks after booster for quantitation of gp120-specific antibodies by titration of sera in gp120-coated ELISA plates (F i g . 1A). Sera from mice of the groups primed with plasmid DNA and boosted with VV-env showed higher antibody titers than those from animals primed and boosted with plasmid DNA alone, indicating that boosting with VV-env is an efficient system for triggering antibody responses. In addition, coinjection of the GM-CSF-DNA for priming resulted in a further increase in antibody titer in mice boosted with VV-env. Priming with gp120-DNA was essential for eliciting high specific antibody levels, since control animals primed with GM-CSF-DNA alone and boosted with VV-env showed low antibody levels. ELISA analysis of antibody isotypes showed production of both IgM and IgG following immunization. There is a predominance of IgG1, with lower IgG2a levels, in sera from mice receiving a VV boost (F i g . 1B). Concurring with the data presented in F i g . 1 A , the IgG1 titer increased when GM-CSF was used for priming. The gp120-specific antibody response in DNA/DNAimmunized mice was only of the IgM isotype, whereas control animals immunized with GM-CSF-DNA/VV-env showed reduced levels of all three isotypes, indicating that the presence of the appropriate antigen at priming is required to trigger antigen-specific responses. To determine the antiviral activity of sera from immunized mice, we performed neutralization assays in which activated human PBMC were challenged with cellfree HIV-1/NL4-3 virus preincubated with sera from mice immunized using the protocols described above. The amount of p24 present in culture supernatants was measured at ten days post-infection (F i g . 2 ). As expected,

B. Analysis of T cell proliferation and cytokine production triggered by DNA vaccination Activation and proliferation of T helper lymphocytes is critical in inducing humoral immune responses, via expansion of antigen-activated B cells, as well as cellular immune responses, via CD8+ cytotoxic T lymphocyte expansion. Exposure to antigen can stimulate at least two patterns of cytokine production by CD4+ T cells (Mosmann et al., 1986). Responses that result in secretion of interferon-! (IFN-! ) and interleukin 2 are classified as T helper 1 (Th1), whereas CD4+ T cells producing IL-4, IL-5 and IL-10 are classified as T helper 2 (Th2). Differentiation of CD4+ T cells into either Th1 or Th2 is influenced by the milieu in which antigen priming takes place. The identification of the conditions leading to a Th1 or Th2 response is critical, as under some circumstances the successful elimination of infectious + agents depends on expansion of the appropriate CD4 T cell subset. In general, Th1 cells are responsible for generating cellular immunity against intracellular pathogens and Th2 cells promote the development of humoral responses (Clerici et al., 1992).

F i g u r e 2 . In vitro anti-HIV-1 neutralization activity of sera from immunized mice. Purified cell-free HIV-1 (IIIB) virus was incubated with serial dilutions of sera from control and immunized mice. After 1 h at 37ยบC, the serum-virus mixture was used to infect PHA-activated human PBMC. At 10 days post-infection, the amount of p24 antigen in the culture medium was measured by ELISA. The mean of triplicate samples is represented.

182

Gene Therapy and Molecular Biology Vol 3, page 183

F i g u r e 3 . Cellular antiHIV-1 gp120 immune response induced after vaccination of mice with vector DNA followed by boosting with DNA or recombinant VV. Proliferative response of spleen cells from mice immunized as described for Fig. 1. Spleens were removed 14 days after boosting and spleen cells incubated with purified gp120 (1 Âľg/ml). [3 H]thymidine incorporation was measured as described in Methods. The Stimulation Index (SI) was determined as the ratio of 3 H cpm in gp120stimulated cultures/3 H cpm in unstimulated cultures.

Figure 4. Cytokine response in spleen cells from immunized mice. The gp120-stimulated splenocyte culture supernatants described above were collected after 48 h, and IFN-! and IL-4 levels determined as described in Methods. The mean of triplicate samples is represented.

whether the booster was given with vector DNA or VVenv (F i g . 3 ). To determine the type of immune response, we analyzed the pattern of cytokine responsiveness. IL-4 and IFN-! levels were measured in in vitro gp120-stimulated splenocytes from immunized animals (F i g . 4). IFN-! levels in DNA/VV-immunized mice were significantly higher than in DNA/DNA-immunized mice. There was a significant increase in IFN-! production when GM-CSFDNA was incorporated in the priming of mice immunized with DNA/DNA, but not when the priming/booster regime consisted of vector DNA/VV-env. In addition, IL-4 was only detectable in splenocyte cultures from DNA/VV-

Here we have analyzed the effect of different immunization protocols on triggering T cell activation. Two weeks after the last injection, splenocytes from immunized mice were tested for T cell proliferation. Antigen-specific T cell proliferation was determined after incubation of the splenocyte cultures with purified gp120. Cell proliferation was measured by addition of [3H]thymidine to cultures and determination of the [3H]thymidine incorporation ratio in gp120-stimulated cultures vs. unstimulated cultures, to derive the stimulation index (SI). The highest SI was obtained when the GM-CSFDNA vector was coinjected with the gp120 DNA vector in the priming; this enhancement was observed regardless of 183

RodrĂ­guez et al: Immunostimulation by HIV-1 env-GM-CSF env-immunized mice; coinjection of GM-CSF-DNA at priming induced increased production of this cytokine.

C. DNA vaccination activates efficient CD8 + T cells The ELISPOT assay uses peptides of defined MHC class I-restricted CTL epitopes to quantitate epitopespecific IFN-! release by individual CD8+ T cells in unstimulated splenocyte cultures, correlating with levels of cytotoxic activity (Hanke et al., 1998, Rodrigues et al, 1994). As detection of IFN-! -producing cells is more sensitive and quantitative than a 51Cr-release cytotoxicity assay, the ELISPOT assay is a useful method for evaluating the success of vaccination. The priming/boosting vaccination regime, using the DNA/VVenv, induces a dramatic stimulation of IFN-! -producing CD8+ cells, as compared to vaccination with the DNA/DNA approach (F i g . 5). A single VV-env inoculation was even more effective in inducing an envspecific CD8+ T cell response than the DNA/DNA double immunization regime, as seen by the larger number of IFN-! secreting cells among splenocytes from mice primed with GM-CSF-DNA and boosted with recombinant VV-env. In accordance with the observed increase in IFN-! production, there was a 10-fold increase in the CD8+ T cells number in mice primed with GM-CSF-DNA and the gp120-expressing DNA vector, and boosted with gp120DNA. This enhancing effect of GM-CSF was not, however, observed in animals treated with the DNA/VVenv approach.

F i g u r e 5 . HIV-1 specific IFN-!-secreting CD8+ T cells in splenocyte cultures from mice primed with vector DNA and boosted with DNA or recombinant VV-env. Spleens from mice, immunized as in Fig. 1, were removed 14 days after boosting, and spleen cells incubated with P815 cells coated with a synthetic peptide corresponding to the gp120 V3 loop. After stimulation of the splenocytes with V3 loop peptide-presenting APC, the number of IFN-!-secreting CD8+ T lymphocytes was quantitated by the ELISPOT assay following the protocol described in Methods. Each bar represents the mean of triplicate samples.

184

III. Discussion The ideal HIV-1 vaccine would probably elicit both humoral and cellular immune responses. Such a dual response would aid in clearing virus before persistent infection is established, as well as in eliminating infected cells by recognizing processed viral fragments associated with host-specific MHC class I antigens and presented on the infected cell surface. The induction of antibodies against the gp120 surface protein of HIV-1 would be important, as this protein is the principal viral determinant interacting with host receptors and the major antigenic determinant to which neutralizing antibodies are directed. Moreover, HIV-specific antibodies that mediate ADCC are found very early in acute infection and correlate well with a decline in plasma virus load (D´Souza et al., 1996). On the other hand, the dramatic decrease in HIV-1 viral load following the initial appearance of CTL after primary infection, and the temporal association between HIVspecific CTL activity and stable viral load or CD4+ cell counts during asymptomatic stages are the best indicators of CTL efficiency. Immunization with DNA vaccines induces protective CTL responses in many experimental models, and can prevent HIV-1 infection or trigger a neutralizing HIV-1 response in certain non-human systems (Wang et al., 1995; Ugen et al., 1997); the safety, feasibility and immunogenicity of DNA vaccines are currently under evaluation in clinical trials (unpublished data).

Gene Therapy and Molecular Biology Vol 3, page 185

and IL-4) in culture supernatants of splenocytes from mice receiving the combined immunization, whereas IFN-! but not IL-4 was present in DNA-immunized mice. It has been claimed that the effect of DNA is dominant, as preimmunization with plasmid DNA followed by boosting with protein in alum prevents the induction of the IgE antibody response or the activation of Th2 cells that would be expected in an alum-based vaccine (Raz et al., 1996). Our results do not confirm this observation, since boosting with gp120-expressing vaccinia virus induces a combined immune response characterized by IL-4 and IgG1 production, as well as by a large number of IFN-! producing CD8+ cells. The only reports of significant antigen-specific IL-4 and IgG1 production following DNA immunization are associated with the gene gun route, since direct intramuscular or intradermal inoculation of naked DNA results specifically in Th1 responses (Feltquate et al., 1997). Immunization with the chicken ovalbumin (OVA) gene induces primarily an IgG1 rather than an IgG2a response, although the OVA-specific T cell response includes IFN-! -secreting Th1 cells (Tighe et al., 1998). An unbiased increase in all immune responses was also observed by increasing the period between immunizations, with significant enhancement of IFN-! , IL-4, IgG1 and IgG2a production (Prayaga et al., 1997). Immune responses of this nature may be more desirable for protection against mucosally-transmitted viral diseases and certain inflammatory autoimmune disorders such as rheumatoid arthritis. Finally, we have observed that coimmunization of gp120-DNA with GM-CSF-DNA results in the enhancement of antigen-specific T cell proliferation rates in both DNA/DNA- and DNA/VV-immunized mice. The GM-CSF stimulation effect on the number of IFN-! producing CD8+ cells can clearly be seen in DNA/DNAimmunized mice, but not in DNA/VV-vaccinated mice, probably due to a saturation effect of the VV boost. A modest but significant increase was seen, however, in the humoral response when GM-CSF-DNA is included in the DNA/VV immunization protocol. Given the acceptability and potential safety of both DNA and VV vaccine vehicles in humans, it is reasonable to believe than this priming/booster regime based on these vectors may be effective not only for HIV, but also for other infectious agents and some forms of cancer.

Although DNA vaccines generally elicit strong cytotoxic responses, the antibody response generated by gene vaccination is lower than that induced by protein vaccination. An additional boost may be required for the generation of a protective humoral immune response, and combined prime-boost immunization approaches using different vector combinations are under evaluation. Priming with a DNA vector followed by a boost with an attenuated VV (MVA) recombinant has recently been shown to result in unexpectedly high levels of protection against P. berghei in mice (Schneider et al., 1998). The rationale for using two distinct vaccine vehicles for the same antigen in combined prime/boost regimes lies in the observation that sequential vaccination with the same vehicle complex decreases vaccine immunogenicity; for example, prior host exposure to VV has been observed to reduce the immunogenicity of VV-based vaccines. VV recombinants expressing HIV-1 envelope antigens were among the first prototype vaccine constructs generated against AIDS (Chakrabarty et al., 1986; Hu et al., 1986). In animal models, these constructs elicit both humoral and cellular immune responses, which are nonetheless unable to control viral infection after HIV challenge (Earl et al., 1989). A DNA prime-MVA boost regime has also been evaluated as a candidate HIV vaccine, given its ability to potentiate the induction of specific CTL (Hank et al., 1998); this study focuses, however, on cellular and not on antibody responses. Here we report the effect of the combined vaccine on both humoral and cellular responses, and take advantage of the immunostimulant properties of GM-CSF in the priming step. We and others have demonstrated that this cytokine increases the potency of immunization against tumor cells and protein antigens (Tao & Levy, 1993; Disis et al., 1996). Using a VV expressing the GM-CSF/gp120 fusion protein, we observed significant enhancement of the cellular immune response against the env antigen (Rodriguez et al., 1998). Here we show an increase in both humoral and cellular immune responses using the DNA-prime/VV-boost regime, as compared to vaccination with DNA alone. This enhancement is especially relevant in the induction of IFN! " producing CD8+ cells, for which a 100-fold increase was observed. The antibody response was also significantly augmented both in the titer of specific antibodies and in the viral neutralizing activity of the serum. Whereas sera from mice immunized with the DNA/DNA regime contained only IgM antibodies, IgG1 was the predominant isotype in sera from DNA/VV-immunized animals, although IgM and IgG2a were also detectable. This response appears to be paradoxical, since the induction of high CTL levels corresponds to a typical Th1-type response, whereas IgG1 antibody production is characteristic of a Th2-type response. In fact, we observed cytokines representative of both types of response (IFN-!

IV. Materials and methods A. Plasmids and recombinant virus Plasmid DNA expressing gp120 was kindly provided by Dr. Andreas BĂźltmann (Max von Pettenkofer-Institut, Genzentrum, Munich, Germany) (AndrĂŠ et al., 1998). The plasmid expressing GM-CSF was previously described (Rodriguez et al., 1998). Plasmids for immunization were purified using Qiagen Maxiprep Columns (Hilden, Germany).

185

Rodríguez et al: Immunostimulation by HIV-1 env-GM-CSF The VV-env recombinant virus has been previously described (Rodriguez et al., 1989).

B. Immunization of mice Six-week-old female BALB/c mice received 50 µg of DNA in PBS intradermically. Fifteen days later, they received an intraperitoneal injection of 5 x 10 7 plaque-forming units (pfu) of purified VV env.

C. Anti-gp120 ELISA Anti-gp120 levels in immunized mouse serum were quantitated by titering on ELISA plates (Maxi-sorb, Nunc) coated with gp120 (IIIB; Intracel) at 1 µg/ml in PBS (100 µl/well) overnight at 4ºC. Remaining protein binding sites were blocked with 0.5% BSA in PBS (200 µl/well, 60 min, 37ºC). After washing plates with distilled water, the diluted sera were added to the wells and incubated for 60 min at 37ºC, followed by a peroxidase (PO)-labeled goat anti-mouse immunoglobulin antibody (GAM-PO, Tago Inc., Burlingame, CA) and OPD (Sigma Chemical Co., St. Louis, MO). The reaction was terminated with 3N sulphuric acid and optical density determined at 492 nm. The titer is expressed as the highest serum dilution giving an absorbance value three times higher than that of the preimmune serum.

D. Viral neutralization assays HIV-1 strain IIIB cell-free virus (NL4-3 strain; 2 ng p24/10 6 cells) was incubated with serial dilutions of mouse antisera for 1 h at 37ºC. PHA-activated PBMC were incubated with the virus-antiserum mix for 2 h, then washed three times. Triplicate 0.5 ml cultures were tested for p24 production at 5, 7 and 10 days post-infection. p24 antigen levels were measured by ELISA (Coulter, Miami, FL).

E. T cell proliferation assay Spleens were removed from infected mice, single-cell suspensions prepared in complete medium (RPMI-1640 with 10% FCS, 2 mM L-glutamine and 10 µM 2-mercaptoethanol), and splenocytes (10 6 /well) dispensed into 96-well microtiter plates. Culture triplicates were challenged with 2 µg/ml of gp120 or concanavalin A (Con A, Sigma) and incubated for three days at 37ºC in 5% CO 2 , after which 1 µCi of [3 H]thymidine (5 Ci/mmol; Amersham) was added to each well. Cells were harvested after 16 h and [3 H]-thymidine incorporation into DNA measured by liquid scintillation counting. Stimulation index (SI) was determined as the ratio: experimental count (mean of [3 H]-thymidine incorporation in triplicate wells incubated with antigen) divided by the spontaneous count (mean of [3 H]-thymidine incorporation in triplicate wells incubated with medium alone).

E. Cytokine assays In vitro gp120-specific IFN-! and IL-4 production was measured using gp120-stimulated splenocytes as previously described (Rodriguez et al., 1998). Conditioned medium containing the secreted cytokines were collected from all cultures after 48 h.

186

F. ELISPOT assay The ELISPOT assay was used to detect epitope-specific IFN-!-secreting cells (Miyahira et al., 1995). Briefly, nitrocellulose-bottomed 96-well plates were coated with antimouse IFN-! mAb R4-6A2 (8 µg/ml, Pharmingen, San Diego, CA). After overnight incubation at room temperature, wells were washed threes times with RPMI 1640, then 100 µl of medium supplemented with 10% FCS were added to each well, and plates incubated at 37ºC for 1 h. Duplicate cultures were prepared with serial doubling dilutions of immunized splenocytes, beginning with 10 6 cells/well. P815 cells (H2 d ), used as antigen-presenting cells (APC), were pulsed with 10 -6 M of the synthetic peptide GPGRAFVTI, corresponding to the V3 loop of gp120, and treated with mitomycin C (30 µg/ml, Sigma). After several washes with culture medium, 105 P815 cells were added to each well. Control P815 cells were not pulsed with the peptide. Plates were incubated for 26-28 h at 37ºC, washed with PBS containing 0.05% Tween-20 (PBST) and incubated overnight at 4ºC with biotinylated antimouse IFN-! mAb XMG1.2 (2 µg/ml, Pharmingen) in PBS-T. Plates were washed with PBS-T and PO-labeled avidin (Sigma; 100 µl, 1/800 dilution in PBS-T) was added to each well. One hour later, wells were washed with PBS/T and PBS. Spots were developed by adding 50 mM Tris-HCl, pH 7.5 containing 1 mg/ml of 3,3'-diaminobenzidine tetrahydrochloride (Sigma) and 0.015% H 2 O2 . When the plates were completely dry, the number of spots was determined with the aid of a stereomicroscope.

Acknowledgments The authors are grateful to L. Gómez for help with animal procedures and to C. Mark for excellent editorial assistance. This work was supported in part by grants from the CICYT of Spain SAF95-0072 and the Comunidad de Madrid 08.6/0020/97 (to M.E.). D.R. and J.R.R. were recipients of contracts from the CSIC of Spain. The Department of Immunology and Oncology was founded and is supported by the Spanish Research Council (CSIC) and Pharmacia & Upjohn

References André, S., Seed, B., Eberle, J., Schraut, W., Bültmann, A. and Hass, J. ( 1 9 9 8 ) . Increased Immune response elicited by DNA vaccination with a synthetic gp120 sequence with optimized codon usage. J . V i r o l o g y 72, 1497-1503. Chakrarabarty, S., Robert-Guroff, M., Wong-Stall, F., Gallo, R.C. and Moss, B. ( 1 9 8 6 ) . Expression of the HTLV-III envelope gene by a recombinant vaccinia virus Nature 320, 535-537. Clerici, M., Giorgi, J.V., Chou, C.C., Gudeman, V.K., Zack, J.A., Gupta, P., Ho, H.N., Nishanian, P.G., Berzofsky, J.A. and Shearer, G.M. ( 1 9 9 2 ) . Cell mediated immune response to human immunodeficiency virus (HIV) type 1 in seronegative homosexual men with recent sexual exposure to HIV-1. J . I n f e c t . D i s . 165, 1012-1019. Disis, M.L., Bernhard, H., Shiota, F.M., Hand, S.L., Gralow,

Gene Therapy and Molecular Biology Vol 3, page 187 J.R., Huseby, E.S., Gillis, S. and Cheever, M.A. ( 1 9 9 6 ) . GM-CSF, and effective adjuvant for protein and peptide-based vaccines. B l o o d 88, 202-210. D’Souza, M.P. and Mathieson, B.J. ( 1 9 9 6 ) . Early phases of HIV-1 infection. AIDS Res. Human Retroviruses 12, 1-9. Feltquate, D.M., Heaney, S., Webster, R.G. and Robinson, H.L. ( 1 9 9 7 ) . Different T helper cell types and antibody isotypes generated by saline and gene DNA immunization J . I m m u n o l o g y . 158, 2278-2284. Fuller, D.H. and Haynes, J.R. ( 1 9 9 4 ) . A qualitative progression in HIV type 1 glycoprotein 120-specific cytotoxic cellular and humoral immune responses in mice receiving a DNA-based glycoprotein 120 vaccine. AIDS Res. Human Retroviruses 10, 1433-1441. Gerloni, M., Lo, D., Ballou R. and Zanetti M. ( 1 9 9 8 ) . Immunological memory after somatic transgene immunization is positively affected by priming with GMCSF and does not require bone marrow-derived dendritic cells. Eur. J. Immunol. 28, 1832-1838. Hanke, T., Blanchard,T.J., Schneider, J., Hannan, C.M., Becker, M. Gilbert, S.C., Hill, A.V.S., Smith, G.L. and McMichael, A. ( 1 9 9 8 ) . Enhancement of MHC class Irestricted peptide-specific T cell induction by a DNA prime/MVA boost vaccination regime. V a c c i n e 16, 439445. Heufler, C., Koch, F. and Schuler, G. ( 1 9 8 8 ) GM-CSF and IL-1 mediate the maturation of murine epidermal Langerhans cells into potent immunostimulatory dendritic cells. J . E x p . M e d i c i n e 167, 700-705. Hu, S.L., Kosowski, S.G. and Dalrymple, J.M. ( 1 9 8 6 ) . Expression of AIDS virus envelope gene in recombinant vaccinia viruses. Nature 320, 537-539. Inaba, K., Inaba, M., Romani, N., Aya, H., Deguchi, M., Ikehara, S., Muramatsu, S. and Steinmann, R.M. ( 1 9 9 2 ) . Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with GMCSF. J. Exp. Med. 176, 1693-1702. Li, S., Rodrigues, M., Rodríguez, D., Rodríguez, J.R., Esteban, M., Palese, P., Nussenzweig, R. and Zavala, F. ( 1 9 9 3 ) . Priming with recombinant influenza virus administration of recombinant vaccinia virus induces CD8+ T cell-mediated protective immunity against malaria. P r o c . N a t l . Acad. S c i . USA 90, 52145218. Miyahira,Y., Murata, K., Rodriguez, D., Rodríguez, J.R., Esteban, M., Rodrigues, M. and Zavala, F. ( 1 9 9 5 ) . Quantification of antigen specific CD8+ T cells using an ELISPOT assay. J . I m m u n o l . M e t h o d s 181, 45-54. Mosmann, T.R., Cherwinski, H. and Bond, M.W. ( 1 9 8 6 ) . Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. J.Immunol. 136, 2348-2353. Prayaga, S.K., Ford, M.J. and Haynes, J.R. ( 1 9 9 7 ) . Manipulation of HIV-1 gp120-specific immune responses elicited via gene gun-based DNA immunization. V a c c i n e 15, 1349-1352. Raz, E., Carson, D.A., Parker, S.E. ( 1 9 9 4 ) . Intradermal gene immunization, the possible role of DNA uptake in the induction of cellular immunity to viruses. P r o c . Natl. Acad. Sci. USA . 91, 9519-9523.

187

Raz, E., Tighe, H. and Sato, Y., Corr, M., Dudler, J.A., Roman, M., Swain, S.L., Spiegelberg, H.L., Carson, D.A. ( 1 9 9 6 ) . Preferential induction of a Th1 immune response and inhibition of specific IgE antibody formation by plasmid DNA immunization. P r o c . N a t l . Acad. Sci USA 93, 5141-5145. Rodrigues, M., Li, S., Murata, K., Rodríguez, D., Rodríguez, J.R., Bacik, I., Bennink, J.R., Yewdell, J.W., GarcíaSastre, A., Nussenzweig, R.S., Esteban, M., Palese, P. and Zavala, F. ( 1 9 9 4 ) . Influenza and vaccinia viruses expressing malaria CD8+ T and B cell epitopes: comparison of their immunogenicity and capacity to induce protective immunity. J . I m m u n o l o g y 153, 4636-4648. Rodríguez, D. Rodríguez, J.R., Rodríguez, J.F., Trauber, D. and Esteban, M. ( 1 9 8 9 ) . Highly attenuated vaccinia virus mutants for the generation of safe recombinant viruses. Proc. Natl. Acad. Sci USA 86, 1287-1291. Rodríguez, D., Rodríguez, J.R., Llorente, M., Vázquez, I., Lucas, P., Esteban, M. Martínez-A, C. and del Real, G. ( 1 9 9 8 ) . An HIV-1 env-GM-CSF fusion protein enhances the cellular immune response to env in a vaccinia virusbased vaccine. J . G e n . V i r o l . In press. Sato, Y., Roman, M., Tighe, H. et al. ( 1 9 9 6 ) . Immunostimulatory DNA sequences necessary for effective intradermal gene immunization. S c i e n c e 273, 352-354. Sedegah, M., Jones, T., Kaur, M., Hedstrom, R., Hobart, P., Tine, J.A. and Hoffman, S.L. ( 1 9 9 8 ) . Boosting with recombinant vaccinia increases immunogenicity and protective efficacy of malaria DNA vaccine. P r o c . N a t l . Acad. Sci USA 95, 7648-7653. Schneider, J., Gilbert, S.C., Blanchard, T.J., Hanke, T., Robson, K.J., Hannan, C.M., Becker, M., Sinden, R., Smith, G.L. and Hill, A.V.S. ( 1 9 9 8 ) . Enhanced immunogenicity for CD8 + T cell induction and complete efficacy of malaria DNA vaccination by boosting with modified vaccinia virus Ankara. Nature M e d i c i n e 4, 397-402. Tao, M.H. and Levy, R. ( 1 9 9 3 ) . Idiotype/granulocytemacrophage colony-stimulating factor fusion protein as a vaccine for B-cell lymphoma. Nature 362, 755-758. Tighe, H., Corr, M., Roman, M. and Raz, E. ( 1 9 9 8 ) . Gene vaccination, plasmid DNA is more than just a blueprint. Immunology Today 19, 89-96. Ugen, K.E., Boyer, J.D. and Wang, B. ( 1 9 9 7 ) . Nucleic acid immunization of chimpanzees as a prophylactic/ immunotherapeutic vaccination model for HIV-1, prelude to a clinical trial. Vaccine 15, 927-930. Wang, B., Boyer, J.D. and Srikantan, V. ( 1 9 9 5 ) . Induction of humoral and cellular immune responses to the human immunodeficiency type 1 virus in non-human primates by in vivo DNA inoculation. V i r o l o g y 211, 102-112. Witmer, P.M., Olivier, W., Valinsky, J., Schuler, G. and Steinman, R.M. (1 9 8 7 ) GM-CSF is essential for the viability and function of cultured murine epidermal Langerhans cells. J . E x p t l . M e d . 166, 1484-1498. Xiang, Z.Q. and Ertl, H.C. ( 1 9 9 5 ) . Manipulation of the immune response to a plasmid encoded viral antigen by coinoculation with plasmids expressing cytokines. Immunity 2, 129-135.

RodrĂ­guez et al: Immunostimulation by HIV-1 env-GM-CSF

188

Gene Therapy and Molecular Biology Vol 3, page 189 Gene Ther Mol Biol Vol 3, 189-196. August 1999.

Gene therapy approaches to the treatment of hemoglobinopathies Review Article

Linda Gorman and Ryszard Kole Lineberger Comprehensive Cancer Center and Department of Pharmacology, University of North Carolina, Chapel Hill, NC 27599. __________________________________________________________________________________________________ Correspondence: Ryszard Kole, Ph.D., University of North Carolina, Lineberger Comprehensive Cancer Center, CB #7295, Chapel Hill, NC 27599. Tel: (919) 966-1143; Fax: (919) 966-3015; E-mail: kole@med.unc.edu Received 25 September, 1998; accepted: 7 October 1998

Summary Hemoglobinopathies such as thalassemia and sickle cell anemia are potentially amenable to gene therapy. Applicable gene therapy strategies can be divided into four categories: those that replace t h e f a u l t y g e n e w i t h a c o m p l e t e t r a n s c r i p t i o n a l u n i t , those that activate transcription o f fetal hemoglobin genes, those that modify the endogenous gene itself, and those that attempt to repair the defective globin RNA transcripts transcribed from the gene. Before becoming valuable in the treatment of human patients, each of these methodologies must overcome obstacles in efficiency of delivery, level of effectiveness, and length of time the treatment remains effective.

normal red blood cells. In thalassemia, a decrease or absence of !# or "#globin synthesis leads to low levels of hemoglobin, causing anemia. In sickle cell anemia a point mutation leads to production of a mutant "-globin ("s) that polymerizes and accumulates in erythrocytes, resulting in changes in membrane morphology and properties, leading to vaso-occlusion (Platt, 1993). To correct these disorders, researchers utilized oligonucleotides, small nuclear RNAs (snRNAs), ribozymes, and other strategies to restore the production of correct globin mRNA and protein.

I. Introduction Gene therapy techniques show increasing promise for use in the treatment of genetic disease. Gene therapy has been tested in treatment of adenosine deaminase deficiency (Tolstoshev, 1993, Fenjves, 1997), cystic fibrosis (Knowles, 1995), and other genetic disorders (Acsadi, 1991, Dunbar, 1996) both in animal models and in the clinic. The therapeutic effects are normally accomplished by replacing the defective gene with the correct one or by expressing a transgene whose product substitutes for its defective counterpart. Although, in principle, gene therapy should be applicable to any gene-based disorder, the difficulties with vectors suitable for efficient delivery of large transgenes or providing sustained expression of the transfected genes in a tissue-specific, properly regulated manner (Byun, 1996, Shi, 1997) limit its clinical applicability.

II. Gene therapy strategies A. Gene replacement Replacement of defective "-globin genes with a functional transcription unit has been particularly difficult to accomplish (reviewed in Rivella, 1998). Although the "-globin gene is small, regulated expression from a transgene is difficult to achieve because it is controlled by a large locus control region (LCR) (Grosveld, 1998, Orkin, 1990). Since expression of "-globin is only useful if it occurs in erythroid precursor cells in concert with the !-globin genes, tight regulation of the "-globin transgene is particularly important in treatment of sickle cell anemia

An alternative approach to gene replacement is correction of the defect in an existing gene or gene product. This method has been particularly useful in the treatment of hemoglobinopathies such as thalassemia and sickle cell anemia. Correct formation of the !- and "-globin chains of hemoglobin is critical for the formation of hemoglobin in

189

Gorman and Kole: Gene therapy approaches for hemoglobinopathies or thalassemia. Effective gene replacement will require efficient transfer of the "-globin gene into hematopoietic stem cells along with sustained expression at an appropriate, developmentally regulated level. Good candidates for vectors will stably integrate into the cell’s chromosomal DNA or remain episomal. Possible viral vectors that could be useful include adeno-associated virus (AAV), adenovirus, and retroviral-, or simian virus 40 (SV40)-based vectors. Each of these vectors has advantages and disadvantages when factors such as insert size capacity, integration, and potential for long-term expression are considered. Recently, a construct, based on Epstein-Barr virus, which remains episomal in cell culture and is able to accommodate large DNA fragments containing the "globin gene and complete regulatory region, has been developed (Westphal, 1998). However, proper expression of the "-globin gene in hematopoietic cells have not yet been tested. Additional approaches include the replicationdeficient viral vectors or non-viral vectors as gene replacement carriers (Walsh, 1993, Gunzburg, 1996, Rivella, 1998). The latter are not limited by size, but more by difficulties in delivery into the nuclei of target cells and lack of chromosomal integration (Rivella, 1998). To circumvent the problem of the large size of the required "#globin insert several truncated constructs have been tested (Zhou, 1996, Ellis, 1997). "-globin transcripts modified by removal of introns and/or reorganization of the LCR showed improvement in stable proviral transmission (Sadelain, 1995, Leboulch, 1994, Takekoshi, 1995). One of the constructs was used in a mouse transplant model and showed some evidence of long-term, high level expression of human "-globin (Raftopoulos, 1997). Although this study represents significant progress toward somatic gene therapy, it will be necessary to achieve more consistent and high-level expression of the replacement genes before this approach can be tested in patients. This may involve the development of more efficient transfection protocols or improvement in "-globin constructs. Such constructs might contain different LCR components, stem cell targeting components, nuclear localization signals, or different promoters and/or enhancers (Raftopoulos, 1997).

aberrant 5’ splice sites and activate a common 3’ splice site upstream. 18-mer 2’-O-methyl-oligoribonucleoside phosphorothioates targeted to the aberrant splice sites restored the correct splicing pattern in a sequence specific and dose dependent manner by causing the splicing machinery to use only the correct splice sites. The correction of splicing was accompanied by translation of the resultant "-globin mRNA into full-length "-globin protein. A promising feature of this approach is that in patients, the antisense oligonucleotides would restore the correct splicing of pre-mRNA, properly transcribed from the "-globin gene which remains in its natural chromosomal environment. This precludes the possibility of overexpression or inappropriate expression of "-globin mRNA and protein, an important consideration in treatment of hemoglobinopathies. Note, that the oligonucleotides do not remove the mutation and would therefore require periodic, life-long administrations. This approach is, thus, more akin to a pharmacological treatment than to a gene therapy one.

C. Repair of defective splicing by small nuclear RNAs (SnRNAs) SnRNAs are small, capped RNA molecules that are located in the nucleus and participate in splicing and other RNA processing reactions. Many of the snRNAs contain sequences antisense to the target RNAs and perform their functions upon binding to their target (Birnstiel, 1988). Thus, by analogy to the oligonucleotides discussed above, they can be used as antisense reagents in gene therapy protocols.

B. Repair of defective splicing by oligonucleotides Work in this laboratory showed that antisense oligonucleotides may restore the production of normal "globin in cells expressing thalassemic "-globin genes (Figure 1). Three thalassemic mutations in intron 2 of the "-globin gene: IVS2-654, -705, and -745 (Dominski, 1993, Sierakowska, 1996, 1997, unpublished data) were studied. The RNAs transcribed from these genes are aberrantly spliced due to point mutations that create

F i g u r e 1 . Correction of splicing of IVS2-705 "-globin pre-mRNA by oligonucleotides or modified U7 snRNA. Boxes - exons, lines - introns. The dashed lines represent correct and aberrant splicing pathways. The oligonucleotides or modified U7 snRNA (U7.Hb) targeted to the IVS2-705 cryptic splice site (3') are depicted under the pre-mRNA (Sierakowska, 1996, Gorman, 1998).

190

Gene Therapy and Molecular Biology Vol 3, page 191

F i g u r e 2 . Structure of U7 snRNA constructs. Wild-type U7 snRNA (heavy line) includes a stem-loop structure, the U7specific Sm sequence (blue box) and a sequence antisense to the 3' end of histone pre-mRNA (green box). In anti-705 U7 snRNAs, the two sequences are replaced with the SmOPT sequence and with antisense sequences to the aberrant 3' or 5' splice sites in the "globin gene, respectively. The promoter and 3' end forming (termination) regions are indicated (Gorman, 1998).

of histone pre-mRNA during its 3' processing (Bond, 1991, Spycher, 1994). It seemed possible that upon replacement of the anti-histone sequence with a sequence complementary to the "-globin aberrant splice sites, the resulting U7 snRNA molecule would bind equally well to the new target sequences and correct aberrant splicing in a manner similar to antisense oligonucleotides. Indeed, it was found that the insertion of appropriate antisense sequences into the U7 snRNA prevented its function in histone mRNA processing and allowed it to modify alternative splicing of "-globin pre-mRNA (F i g u r e 1 ). Stable transfection of cells expressing thalassemic "globin gene with vectors carrying a modified U7 snRNA gene led to a permanent correction of the splicing pattern of the "-globin pre-mRNA. Levels of correction reached 65% in transient expression and 55% in stable cell lines. The treatment also resulted in the accumulation of significant amounts of "-globin protein (Figure 3) (Gorman, 1998).

There are several advantages to using snRNAs in this approach. SnRNAs are capped and associate tightly with proteins which protect them from ribonucleases present in the cellular milieu and making them much more stable than naked RNA. SnRNA genes are driven by strong promoters leading to high level of expression, up to 106 copies of the snRNA per cell. Since splicing occurs in the nucleus, the nuclear localization of snRNAs makes them ideal for use in correction of splicing defects frequent in thalassemia (Birnstiel, 1988). Utilization of an snRNA as a therapeutic agent involves replacement of the natural antisense sequence with that targeted to the desired RNA. The modified snRNA gene is incorporated into a plasmid and transfected into the cell. The relatively short insert carries the snRNA promoter and therefore exogenous promoters are not necessary. The transcribed snRNA migrates to the cytoplasm where it is complexed with specific proteins and subsequently returns to the nucleus where it can bind to the desired target. It is anticipated that snRNAs as antisense carriers will allow for long term, possibly permanent, expression of RNA antisense to its targets such as the aberrant thalassemic splice sites in "-globin pre-mRNA.

D. Removal of mutations by chimeric RNA-DNA oligonucleotides It has recently been shown in model cell culture systems that double stranded chimeric RNA-DNA oligonucleotides may induce site specific removal from the human "-globin gene of the mutation responsible for sickle cell anemia (Cole-Strauss, 1996). The "S allele is caused by an A to T mutation in the sixth codon of the "globin gene which leads to replacement of valine by glutamic acid. This point mutation in a coding sequence represents a good candidate for using the chimeric oligonucleotides as a potential treatment.

Anti-"-globin sequences were incorporated into the gene for murine U7 snRNA (Figure 2) (Gorman, 1998). U7 snRNA complexes with at least two U7 specific proteins and eight common Sm proteins (Smith, 1991), forming a ribonucleoprotein particle (U7 snRNP). U7snRNP is involved in the processing of the 3' end of histone pre-mRNAs (Galli, 1983, Birchmeier, 1984, Birnstiel, 1988). The first 18 nucleotides of this 62 nucleotide-long RNA function as a natural antisense sequence by hybridizing with the so-called spacer element

191

Gorman and Kole: Gene therapy approaches for hemoglobinopathies

Figure 3. Correlation of protein levels (A) with mRNA (B) expression in a stable cell line expressing modified U7 snRNA (U7.Hb) (A) Western blot of cell line expressing full length "-globin protein (Lane 1), IVS2-705 cell line (Lane 2), and stable cell line expressing U7.Hb snRNA (Lane 3). (B) RT/PCR products from RNA from human blood (Lane 1), IVS2-705 cell line (Lane 2), and stable cell line expressing U7.Hb snRNA (Lane 3) (Gorman, 1998).

F i g u r e 4 . Diagram of basic structure of chimeric oligonucleotide. Blue lines represent chimera DNA residues, red lines represent 2’-O-methyl RNA residues, yellow box represents target nucleotide, green lines represent target DNA (Cole-Strauss, 1996).

mutation. This single base mismatch is recognized by the endogenous cellular repair systems and either the oligonucleotide or the target is changed. Use of the chimeras resulted in almost equal amounts of "S and normal "- globin ("A) suggesting 50% correction of the mutation at the DNA level (Cole-Strauss, 1996).

The chimeric oligonucleotides are composed of a stretch of RNA and DNA residues in a duplex formation with double hairpin caps at the ends (Figure 4). The RNA residues were modified by 2’ O-methylation of the ribose increasing the oligonucleotide resistance to nuclease degradation. The sequences of these oligonucleotides align with the target sequence except at the position of the 192

Gene Therapy and Molecular Biology Vol 3, page 193

Figure 5. Ribozyme mediated repair strategy. Scheme to convert "S transcripts (yellow) to $-globin (red) expressing transcripts (Lan, 1998).

Krupple-like factor). Normally, high levels of "-globin are expressed when, in concert with other proteins that bind to locus control region sequences, EKLF binds to the "globin CACCC box. The %-globin gene promoter has a defective CACCC box that does not bind EKLF, which may be one reason for the low level of %-globin expression. Using a modified EKLF that binds to the defective %-globin gene promoter, increased levels of %globin were produced. Such an approach would lead to increased HbA2 (! 2%2) production. Since HbA2 has been shown to be an inhibitor of sickle cell HbS (! 2"S2) polymerization, transduction of erythroid stem cells with the modified EKLF gene, could be a useful treatment for sickle cell patients.

E. Ribozyme-mediated repair of mRNA An alternative treatment for sickle cell anemia utilizes ribozymes to repair the defective "-globin RNA transcripts (Lan, 1998). This work was based on the finding that the self-splicing intron from Tetrahymena thermophila mediates trans-splicing of RNA fragments in vitro (Inoue, 1986, Been, 1986). A shortened form of the ribozyme, L21 (Zaug, 1988), was able to repair defective lacZ transcripts in E. coli and mammalian cells (Sullenger, 1994, Jones, 1996). To test whether ribozymes could be used in a therapeutic manner they were designed to convert "S RNAs into $ -globin messages (Lan, 1998). $ -globin was selected since it was found that fetal, $ -globin containing, hemoglobin retards the polymerization of "S hemoglobin (Sunshine, 1978, Behe, 1979). The transsplicing group I ribozyme was modified to carry the 3’ portion of the $ -globin mRNA. When the ribozyme base paired with the mutant "-globin transcript upstream of the mutation, the transcript was cleaved, thereby releasing the portion containing the mutation, and subsequently the $ globin 3’ exon sequence was spliced in (Lan, 1998) (Figure 5). The ribozyme converted the "S globin RNA to RNA that encoded $ -globin not only in model cell lines but also in red blood cell precursors from human cord blood (Lan, 1998).

F. Activation of

G. -globin reduction The accumulation of excess !-globin in the red blood cell precursors of "#thalassemia patients results in premature cell death before the reticulocyte stage (Weatherall, 1972). Reducing the levels of !-globin can alleviate the imbalance between the !- and "-globin chains, leading to increased production of healthy reticulocytes. Ponnazhagen et al (1994) utilized recombinant adeno-associated virus 2-based antisense vectors to inhibit !-globin expression. Adeno-associated virus 2 (AAV) was selected since it is non-pathogenic and integrates in a site-specific manner into human chromosome 19. Different promoters (herpes virus thymidine kinase (TK) promoter, the SV40 early gene promoter, and the human !-globin gene promoter) were tested to achieve proper level of reduction of the !-globin

globin genes

Another alternative to gene replacement has been described by Donze et al. (Donze, 1996). In this approach the endogenous %-globin gene is activated with a modified erythroid-specific transcription factor, EKLF (Erythroid 193

Gorman and Kole: Gene therapy approaches for hemoglobinopathies mRNA. The observed levels of !-globin inhibition were 0, 29, and 91%, respectively, at the transcriptional level. Thus it may be possible to adjust the level of the !-globin to achieve the desired 1:1 ratio of !- and "#subunits. Complete inhibition of !-globin production would lead to a new imbalance, with excess "-globin chains building up in the cells with detrimental consequences.

Although further testing and improvements are clearly needed before most of the approaches can be used on patients there is hope that some of them will lead to successful treatment for patients with thalassemia and sickle cell anemia. For any of the gene therapy techniques described here, the main hurdles appear to be delivery, achievement of long-term effects of the treatment and proper expression limited to the small population of the target cells, the erythroid precursors. The effects should be, preferably, accomplished by systemic treatment, although ex vivo bone marrow treatment and autologous reimplantation can also be considered. Even if the permanent curative effects of the treatment were difficult to achieve the temporary effects may be of clinical value. The increase in the production of correct "- or other globins in thalassemia and sickle cell anemia should improve survival of the erythroblasts and promote their maturation into red blood cells. Since erythrocytes have a life span of approximately 120 days (Eadie, 1955), the treated cells would remain in the bloodstream for an extended time period.

H. Induced gene expression by drugs Currently, patients with "-thalassemia are treated with periodic blood transfusions and iron chelation therapy. Blood transfusions can lead to iron overload, the leading cause of death in thalassemic patients (Zurlo, 1989). Clinically relevant alteration of globin gene expression can be achieved by relatively simple pharmacological treatments. For example, hydroxyurea or butyric acid and its derivatives induce the expression of fetal hemoglobin (! 2$ 2) which partially compensates for the lack of correct "-globin expression in sickle cell anemia or thalassemia (Charache, 1995, Charache, 1996). In clinical trials, patients receiving hydroxyurea exhibited a decrease in crisis rates within the first three months of treatment (Charache, 1995, Charache, 1996). The lower crisis rates were accompanied by lower neutrophil and platelet counts, as well as higher mean corpuscular volume and mean corpuscular hemoglobin concentrations (Charache, 1995, Charache, 1996). Use of hydroxyurea in patients with sickle cell anemia was so successful that the trial was stopped early and the drug moved to the clinic.

References Acsadi, G., Dickson, G., Love, D.R., Jani, A., Walsh, F.S., Gurusinghe, A., Wolff, J.A., Davies, K.E. (1 9 9 1 ) Human dystrophin expression in mdx mice after intramuscular injection of DNA constructs. Nature 352, 815-818. Been, M., Cech, T.R. (1 9 8 6 ) One binding site determines sequence specificity of Tetrahymena pre-rRNA selfsplicing , trans-splicing, and RNA enzyme activity. C e l l 47, 207-216.

Sodium butyrate is used to treat urea-cycle disorder (Brusilow, 1991). It was found that one of the side effects from the drug was an increase in the patient’s fetal hemoglobin levels, although the exact mechanism involved is unknown. These findings led to attempts to use this drug for the treatment of "-thalassemia. These treatments elicited varying levels of success in clinical trials (Perrine, 1989, Charache, 1996, Sher, 1995, Collins, 1995). In one study, 36% of all patients or 50% of non-transfused patients exhibited an increase in fetal hemoglobin after treatment with sodium phenylbutyrate (Collins, 1995). The increase in hemoglobin was concomitant with an increase in the number of red blood cells and mean corpuscular volume (Collins, 1995). Unfortunately, until it can be determined which subset of "-thalassemic patients will respond to butyrate analogues, this therapy will remain of limited usefulness as a treatment.

Behe, M.J., Englander, S.W. (1 9 7 9 ) Mixed gelation theory. Kinetics, equilibrium and gel incorporation in sickle hemoglobin mixtures. J . M o l . B i o . 133, 137. Birchmeier, C., SchĂźmperli, D., Sconzo, D., Birnstiel, M.L. (1 9 8 4 ) 3' editing of mRNAs: sequencerequirements and involvement of a 60-nucleotide RNA in maturation of histone mRNA precursors. P r o c . N a t l . Acad. S c i . USA 81, 1057-1061. Birnstiel, M.L. and Schaufele, F. (1 9 8 8 ) in Structure and f u n c t i o n o f major and minor s m a l l nuclear r i b o n u c l e o p r o t e i n p a r t i c l e s , ed. Birnstiel, M.L., (Berlin) pp. 155-182. Bond, U., Yario, T.A., Steitz, J. (1 9 9 1 ) Multiple processingdefective mutations in a mammalian histone pre-mRNA are suppressed by compensatory changes in U7 RNA both in vivo and in vitro.G e n e s D e v . 5, 1709-1722. Brusilow, S.W. (1 9 9 1 ) in Treatment o f G e n e t i c D i s e a s e s , ed. Desnick, R.J, (New York) p. 79.

III. Conclusions

Byun, J., Kim, SH., Kim, JM., Yu, SS., Robbins, PD., Yim , J., Kim, S. (1 9 9 6 ) Analysis of the relative level of gene expression from different retroviral vectors used for gene therapy. Gene Ther. 3, 780-788.

Many possible treatments for hemoglobinopathies are currently explored. Strategies range from simple drug treatments to a wide range of gene therapy techniques.

Charache, S., Terrin, M.L., Moore, R.D., Dover, G.J., Barton,

194

Gene Therapy and Molecular Biology Vol 3, page 195 F.B., Eckert, S.V., McMahon, R.P., Bonds, D.R. (1 9 9 5 ) Effect of hydroxyurea on the frequency of painful crises in sickle cell anemia. N . E n g . J . M e d . 332, 1317-1322.

Gunzburg, W.H., Salmons, B. (1 9 9 6 ) Development of retroviral vectors as safe, targeted gene delivery systems. J . M o l . M e d . 74, 171-182.

Charache, S. Barton, F.B., Moore, R.D., Terrin, M.L., Steinberg, M.H., Dover, G.J., Ballas, S.K., McMahon, R.P,. Castro, O., Orringer, E.P. (1 9 9 6 ) Hydroxyurea and sickle cell anemia. Clinical utility of a myelosuppressive "switching" agent. The Multicenter Study of Hydroxyurea in Sickle Cell Anemia. M e d i c i n e 75, 300-326.

Inoue, T., Sullivan, F. X., Cech, T.R. (1 9 8 6 ) New reactions of the ribosomal RNA precursor of Tetrahymena and the mechanism of self-splicing. J . M o l . B i o . , 189, 143.

Cole-Strauss, A., Yoon, K., Xiang, Y., Byrne, B.C., Rice, M.C., Gryn, J., Holloman, W.K., Kmiec, E.B. (1 9 9 6 ) Correction of the Mutation Responsible for Sickle Cell Anemia by an RNA-DNA Oligonucleotide. S c i e n c e 273, 1386-1389.

Knowles, M.R., Hohneker, K.W., Zhou, Z., Olsen, J.C., Noah, T.L., Hu, P.C., Leigh, M.W., Engelhardt, J.F., Edwards, L.J., Jones, K.R. et al. (1 9 9 5 ) A controlled study of adenoviral-vector-mediated gene transfer in the nasal epithelium of patients with cystic fibrosis. New E n g l . J . M e d . 333, 823-831.

Jones, J.T., Lee, S., Sullenger, B.A. (1 9 9 6 ) Tagging ribozyme reaction sites to follow trans-splicing in mammalian cells. Nature Med. 2, 643-648.

Collins, A.F., Pearson, H.A., Giardina, P., McDonagh, K.T., Brusilow, S.W., Dover, G.J. (1 9 9 5 ) Oral sodium phenylbutyrate therapy in homozygous beta thalassemia: a clinical trial. B l o o d 85, 43-49.

Lan, N., Howrey, R.P., Lee, S., Smith, C.A., Sullenger, B.A. (1 9 9 8 ) Ribozyme-Mediated Repair of Sickle "-Globin mRNAs in Erythrocyte Precursors. S c i e n c e 280, 15931596.

Dominski, Z., Kole, R. (1 9 9 3 ) Restoration of correct splicing in thalassemic pre-mRNA by antisense oligonucleotides. P r o c . N a t l . A c a d . S c i . U S A 90, 8673-8677.

Leboulch, P., Huang, G.M., Humphries, R.K. Oh, Y.H., Eaves, C.J., Tuan, D.Y., London, I.M. (1 9 9 4 ) Mutagenesis of retroviral vectors transducing human betaglobin gene and beta-globin locus control region derivatives results in stable transmission of an active transcriptional structure. EMBO J. 13, 3065-3076.

Donze, D., Jeancake, P.H., Townes, T.M. (1 9 9 6 ) Activation of %-globin gene expression by EKLF. B l o o d 88, 40514057. Dunbar, C., Kohn, D. (1 9 9 6 ) Retroviral mediated transfer of the cDNA for human glucocerebrosidase into hematopoietic stem cells of patients with Gaucher disease. A phase I study. Hum. Gene Ther. 7, 231-253.

Perrine, S.P., Miller, B.A., Faller, D.V., Cohen, R.A., Vichinsky, E.P., Hurst, D., Lubin, B.H., Papayannopoulou, T. (1 9 8 9 ) Sodium butyrate enhances fetal globin gene expression in erythroid progenitors of patients with Hb SS and beta thalassemia. B l o o d 74, 454-459.

Eadie, G.S., Brown, I.W., Curtis, W.G. (1 9 5 5 ) The potential life span and ultimate survival of fresh red blood cells in normal healthy recipients as studied by simultaneous Cr51 tagging and differential hemolysis. J . C l i n . I n v e s t . 34, 629-636.

Orkin, S.H. (1 9 9 0 ) Globin gene regulation and switching: circa 1990. C e l l 63, 665-672. Platt, O.S., Dover, G.J. (1 9 9 3 ) in H e m a t o l o g y o f Infancy and Childhood, Nathan, D.G. and Oski, F.A. eds. Saunders, Philadelphia, 732-782.

Ellis, J., Pasceri, P., Tan-Un, K.C., Wu, X., Harper, A., Fraser, P., Grosveld, F. (1 9 9 7 ) Evaluation of beta-globin gene therapy constructs in single copy transgenic mice. N u c l e i c A c i d s R e s . 25, 1296-1302.

Ponnazhagen, S., Nallari, M.L., Srivastava, A. (1 9 9 4 ) Suppression of Human !-Globin Gene Expression Mediated by the Recombinant Adeno-associated Virus2based Antisense Vectors. J . E x p . M e d . 179, 733-738.

Fenjves, E.S., Schwartz, P.M., Blaese, R.M., Taichman, L.B. (1 9 9 7 ) Keratinocyte gene therapy for adenosine deaminase deficiency: a model approach for inherited metabolic disorders. Hum. Gene Ther. 8, 911-917.

Raftopoulos, H., Ward, M., Leboulch, P., Bank, A. (1 9 9 7 ) Long-term transfer and expression of the human betaglobin gene in a mouse transplant model. B l o o d 90, 3414-3422.

Galli, G., Hofstetter, H., Stunnenberg, H.G., Birnstiel, M.L. (1 9 8 3 ) Biochemical complementation with RNA in the Xenopus oocyte: a small RNA is required for the generation of 3' histone mRNA termini. C e l l 34, 823828.

Rivella, S., Sadelain, M. (1 9 9 8 ) Genetic treatment of severe hemoglobinopathies: the combat against transgene variegation and transgene silencing. S e m . in H e m a t o l . 35, 112-125.

Gorman, L., Suter, D., Emerick ,V., Schumperli, D., Kole, R. (1 9 9 8 ) Stable alteration of pre-mRNA splicing patterns by modified U7 small nuclear RNAs. P r o c . N a t l . A c a d . S c i . U S A 95, 4929-4934.

Sadelain, M., Wang, C.H., Antoniou, M. Grosveld, F., Mulligan, R.C. (1 9 9 5 ) Generation of a high-titer retroviral vector capable of expressing high levels of the human beta-globin gene. P r o c . N a t l . Acad. S c i . USA 92, 6728-6732.

Grosveld, F., de Boer, E., Dillon, N., Fraser, P., Gribnau, J., Milot, E., Trimborn, T., Wijgerde, M. (1 9 9 8 ) The dynamics of globin gene expression and gene therapy vectors. S e m . i n H e m a t o l . 35, 105-111.

Sher, G.D., Ginder, G.D., Little, J., Yang, S., Dover, G.J., Olivieri, N.F. (1 9 9 5 ) Extended therapy with intravenous

195

Gorman and Kole: Gene therapy approaches for hemoglobinopathies arginine butyrate in patients with betahemoglobinopathies. N . E n g . J . o f M e d . 332, 16061610.

globin/delta-globin hybrid gene linked to beta locus control region hypersensitive site 2 aimed at the gene therapy of sickle cell disease. P r o c . N a t l . A c a d . S c i . USA 92, 3014-3018.

Shi, Q., Wang, Y., Worton, R. (1 9 9 7 ) Modulation of the specificity and activity of a cellular promoter in an adenoviral vector. Hum. Gene Ther. 8, 403-410.

Tolstoshev, P. (1 9 9 3 ) Gene therapy, concepts, current trials and future directions. A n n . R e v . P h a r m . a n d T o x . 33, 573-596.

Sierakowska, H., Sambade, M.J., Agrawal, S., Kole, R. (1 9 9 6 ) Repair of thalassemic human beta-globin mRNA in mammalian cells by antisense oligonucleotides. P r o c . N a t l . A c a d . S c i . U S A 93, 12840-12844.

Walsh, C.E., Liu, J.M., Miller, J.L., Nienhuis, A.W., Samulski, R.J. (1 9 9 3 ) Gene therapy for human hemoglobinopathies. P r o c . S o c . E x p . B i o l . & Med. 204, 289-300.

Sierakowska, H., Montague, M., Agrawal, S., Kole, R. (1 9 9 7 ) Restoration of "-globin gene expression in mammalian cells by antisense oligonucleotides that modify the aberrant splicing patterns of thalassemic premRNAs. N u c l e o s i d e s & N u c l e o t i d e s 16, 11731182.

Weatherall, D.J., Clegg, J.B. (1 9 7 2 ) The T h a l a s s e m i a Syndromes. Blackwell Scientific Publications, Oxford, pg. 75-144. Westphal, E.M., Sierakowska, H., Livanos, E., Kole, R., Vos, J.M. (1 9 9 8 ) A system for shuttling 200kb BAC/PAC clones into human cells: stable extrachromosomal persistence and long-term ectopic gene activation. Hum. Gene Ther. 9, 1863-1873.

Smith, H.O., Tabiti, K., Schaffner, G., Soldati, D. Albrecht, U., Birnstiel, M.L. (1 9 9 1 ) Two-step affinity purification of U7 small nuclear ribonucleoprotein particles using complementary biotinylated 2'-O-methyl oligoribonucleotides. P r o c . N a t l . Acad S c i . USA 88, 9784-88.

Zaug, A.J., Grosshans, C.A., Cech, T.R. (1 9 8 8 ) Sequencespecific endoribonuclease activity of the Tetrahymena ribozyme: Enhanced cleavage of certain oligonucleotide substrates that form mismatched ribozyme substrates ribozyme-substrate complexes. B i o c h e m i s t r y 27, 8924-8931.

Spycher, C., Streit, A., Stefanovic, B., Albrecht, D. Konig, T.H., Sch端mperli, D. (1 9 9 4 ) 3' end processing of mouse histone pre-mRNA: evidence for additional base-pairing between U7 snRNA and pre-mRNA. N u c l e i c A c ids R e s . 22, 4023-4030.

Zhou, S.Z., Li, Q., Stamatoyannopoulos, G., Srivastava, A., (1 9 9 6 ) Adeno-associated virus 2-mediated transduction and erythroid cell-specific expression of a human betaglobin gene. Gene Ther. 3, 223-229.

Sullenger, B.A., Cech, T.R. (1 9 9 4 ) Ribozyme-mediated repair of defective mRNA by targeted trans-splicing. Nature 371, 619-622.

Zurlo, M.C., DeStafano, P., Borgna-Pignatti, C., DiPalma, A., Piga, A., Melevendi, C., DiGregorio, F., Buratini, M.G., Terzoli, S. (1 9 8 9 ) Survival and causes of death in thalassaemia major. Lancet 2, 27-30.

Sunshine, H.R., Hofrichter, W.A., Eaton, W.A. (1 9 7 8 ) Requirement for therapeutic inhibition of sickle haemoglobin gelation. Nature 275, 238. Takekoshi, K.J., Oh, Y.H., Westerman, K.W., London, I.M., Leboulch, P. (1 9 9 5 ) Retroviral transfer of a human beta-

196

Gene Therapy and Molecular Biology Vol 3, page 197 Gene Ther Mol Biol Vol 3, 197-206. August 1999.

Intramuscular injection of plasmid DNA encoding intracellular or secreted glutamic acid decarboxylase causes decreased insulitis in the nonobese diabetic mouse Research Article

Jingxue Liu1,2, Maria Filippova1, Omar Fagoaga 3, Sandra Nehlsen-Cannarella3, and Alan Escher 1,2 1

Center for Molecular Biology and Gene Therapy, 2Department of Microbiology and Molecular Genetics, and 3Immunology Center, Department of Pathology, School of Medicine, Loma Linda University, Loma Linda, California 92350, USA __________________________________________________________________________________________________ Correspondence: Alan Escher, Ph.D., Center for Molecular Biology and Gene Therapy, 11085 Campus Street, Loma Linda University, Loma Linda, California 92350, USA. Tel: (909) 824-0800 x81363; Fax: (909) 478-4177; E-mail, aEscher@ccmail.llu.edu A b b r e v i a t i o n s : NOD, non-obese diabetic; IDDM, insulin-dependent diabetes mellitus; GAD, glutamic acid decarboxylase; GABA, gamma-aminobutyric acid; aa , amino acids. Received: 28 October 1998; accepted: 11 November 1998

Summary O u r g o a l i s t o determine whether gene vaccination can be used for the treatment o f insulindependent diabetes mellitus (IDDM), an autoimmune disease. In this work, weanling non-obese d i a b e t i c ( N O D ) m i c e , a n a n i m a l m o d e l system for the study o f IDDM, received intramuscular injections of “naked� plasmid DNA encoding either intracellular or secreted human glutamic acid decarboxylase (GAD), one of the major autoantigens recognized during the onset of IDDM. Seven weeks later, each pancreas was scored for insulitis, an inflammation indicative of the disease. Mice treated with either type of gad gene-carrying plasmid showed a significant decrease in the severity of insulitis when compared to controls. These results suggest that vaccination using autoantigenencoding genes may provide a means of treating IDDM.

I. Introduction Insulin dependent diabetes mellitus (IDDM), or type I diabetes, is a disease with high morbidity and mortality that affects 1 in 300 persons in North America, with a prevalence ever increasing in small children (for a review see Mandrup-Poulsen, 1998). Although also called juvenile diabetes because it often affects young people, a similar disease has been diagnosed in patients 50 years of age and older (Molbak et al., 1994). IDDM is thought to be caused by both genetic and environmental factors, and is associated with the autoimmune destruction of insulin-producing beta cells found in the pancreatic islets of Langerhans. Loss of these insulin-secreting cells results in the inability to metabolize glucose, leading to hyperglycemia and

ketoacidosis, which in turn cause a variety of debilitating and life-threatening ailments such as blindness, kidney disease, heart attack, stroke, and neuropathy. Although injection of the hormone insulin can prolong life of IDDM patients, it does not provide a cure for the disease, likely due to lack of proper regulation of insulin levels within the body. A cure for IDDM could be achieved if the destruction of beta cells were averted. IDDM has been characterized as an autoimmune disease based on the observations that patients suffering from this illness have high levels of islet cell autoantibodies in their sera (Bottazo et al., 1974), and chronic mononuclear cell infiltration of their pancreatic islets (Gepts and Lecompte, 1981). Presence of autoantibodies can be detected years

197

Liu et al: Intramuscular injection of glutamic acid decarboxylase DNA and insulitis before the onset of symptoms, and is considered to be diagnostic for IDDM (Maclaren, 1988), although it does not always imply occurrence of the disease. In humans, the nature of these antibodies varies with age: autoantibodies against insulin and tyrosine phosphatase IA-2 are associated with early childhood, glutamic acid decarboxylase (GAD) and islet cell cytoplasmic protein autoantibodies with late childhood and adolescence, while late onset can be associated with other typical immune markers (MandrupPoulsen, 1998). Inflammatory infiltration of the islets (insulitis) and beta cell destruction are due mostly to T lymphocytes, both CD4+ helper and CD8+ cytotoxic (Itoh et al., 1993; Peakman et al., 1994), and result in loss of islet cell mass. When this cell mass drops below 10% of normal, hyperglycemia and ketosis develop. A large part of what is known about IDDM comes from studies of animal model systems, in particular the nonobese diabetic (NOD) mouse. The NOD mouse possesses most of the characteristics of human IDDM, such as genetic predisposition due to MHC II linkage, development of insulitis with infiltration of T lymphocytes selectively toxic to insulin-producing beta cells, and humoral reactivity to beta cells (for a review see Bowman et al., 1994). However, unlike humans, NOD mice have a strong gender bias in the appearance of the disease: 91% of females NOD/Lt mice manifest diabetes at 250 days of age, while only 21% of males show a similar symptom at that age (Baxter et al., 1991). Studies of NOD mice (Kaufman et al., 1993; Tisch et al., 1993) and patients (Baekkeskov et al., 1990) indicate that the GAD protein is a major autoantigen recognized during the onset of IDDM. GAD is an enzyme found mostly in neurons (Erlander et al., 1991) and pancreatic islet cells (Christgau et al., 1991), where it catalyzes the synthesis of gamma-aminobutyric acid (GABA). GABA is an inhibitory neurotransmitter in the central nervous system, and may be a paracrine signaling molecule in the pancreas. Two forms of GAD are encoded by different genes in mammals, a 65 kDa (previously called 64) and a 67 kDa (previously called 65) molecular weight form. GAD65 is a membrane-anchored intracellular protein, while GAD67 is found soluble in the cytosol (Christgau et al., 1991; Christgau et al., 1992). Both GAD65 and GAD67 are recognized by the immune system of IDDM patients (Baekkeskov et al., 1990; Honeyman et al., 1993). In addition, the first T cell response against beta cell antigens in 4-week old NOD mice is against GAD65 (Kaufman et al., 1993;Tisch et al., 1993), and both CD8+ cytotoxic (Panina-Bordignon et al., 1995) and CD4+ T helper 1 (Th1) (Tabata et al., 1998) lymphocytes specific for GAD65 can be found in patients suffering from IDDM. Together with the finding that adoptive transfer of GAD-reactive T cells can cause diabetes in NOD/SCID mice (Zekzer et al., 1998), these observations strongly indicate that GAD65

plays an important role as an autoantigen during onset of IDDM. The NOD mouse serves not only as a model to study IDDM, it is also an excellent system for the development of new methods for preventive transfer of this form of diabetes. Such therapies include immunosuppression, immunostimulation, tolerance induction, manipulation of hormonal/dietary milieu, and anti-inflammatory agents (Bowman et al., 1994). In this work, we have investigated whether gene vaccination could be used to prevent insulitis in the NOD mouse. Specifically, we have used intramuscular injection of “naked� plasmid DNA encoding human GAD65 and SGAD55, an engineered secreted form of this protein. We report that injection of DNA encoding these proteins resulted in significant decreases in insulitis, suggesting the possibility that this form of gene therapy might be useful to prevent clinical manifestation of IDDM.

II. Results A. Construction of a secreted form of human GAD65 Extracellular antigens can be used for tolerization or for suppression of MHC class II restricted Th1 inflammatory response, probably through a MHC class II restricted CD4+ Th2 lymphocyte response, as Th1 and Th2 responses appear to be mutually exclusive (Mosmann and Sad, 1996). Therefore, two genes encoding GAD proteins that had the potential of being secreted by mammalian cells were constructed. The first construct consisted of the leader peptide from human interleukin-2 (IL-2) protein fused to full-length human GAD65, generating a fusion protein encoded by the sgad65 gene. This leader sequence was previously shown to cause secretion by mammalian cells of normally intracellular proteins (Okano et al., 1990; Liu et al., 1997). However, because GAD65 is a membrane-anchored protein, the protein region responsible for the anchoring could have interfered with secretion. The sequence corresponding to approximately the first 100 amino acids (aa) of human GAD65 contains a Golgi-targeting sequence (Solimena et al., 1994), as well as cysteine residues that are palmitoylated and responsible for membrane anchoring (Christgau et al., 1992). In addition, this sequence is not recognized by autoantibodies from IDDM patients (Richter et al., 1993). The first 88 aa of the human GAD65 protein were therefore deleted, and the remainder of the protein was fused to the IL-2 leader sequence, generating a fusion protein (SGAD55) encoded by the sgad55 gene (F i g . 1 B ). Simian COS-7 cells were transiently transfected with the two gene constructs coding for these proteins, and immunoblot analysis of intracellular GAD protein was performed using a monoclonal human GAD65 antibody.

198

Gene Therapy and Molecular Biology Vol 3, page 199 Results confirmed the synthesis of SGAD65 (F i g . 2A, lane 3), and of the lower molecular weight SGAD55 (F i g . 2A, lane 4). To determine whether SGAD65 and SGAD55 were secreted by mammalian cells, proteins from COS-7 cells transiently transfected with the different gene constructs were labeled in vivo with 35S-methionine. Culture media from these cells were then used for immunoprecipitation using the same antibody used for

immunoblot analysis, and immunoprecipitates were fractionated using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Results showed that SGAD55 protein was immunoprecipitated from cell culture media (F i g . 2 B , lane 3). In contrast, no secreted SGAD65 protein could be detected (F i g . 2B, lane 4). The gene construct sgad55 was therefore selected for further use in animal experiments.

Figure 1 . Gene constructs used for intramuscular DNA injection. Three genes were placed under transcriptional control of the cytomegalovirus promoter (CMV) into an expression plasmid, gad65, encoding a wild type intracellular human GAD65 protein (A), sgad55 , encoding a fusion of the IL-2 leader sequence (IL2-LS) to a truncated human GAD65 protein (B ), and sruc3, encoding a secreted Renilla luciferase (C).

F i g u r e 2 . Detection of GAD proteins from lysates (A) and culture media (B ) of mammalian cells grown in vitro. A. Immunoblot analysis of simian COS-7 cells transiently transfected with different gad genes. Cells were transfected with plasmid vector only (lane 1), plasmid carrying gene gad65 (lane 2), sgad65 (lane 3), or sgad55 (lane 4). Lane 5 contains a truncated version of GAD65 isolated from Escherichia coli as control. Total cells lysates were fractionated using SDS-PAGE, transferred onto a membrane, and reacted with a mouse monoclonal antibody raised against wildtype human GAD65, and subsequently to a secondary antibody bound to alkaline phosphatase for chemiluminescent detection. B . Immunoprecipitation of 35 Smethionine-labeled proteins from culture media. Culture media from COS-7 cells transiently transfected with plasmid vector only (lane 1), plasmid carrying gene gad65 (lane 2), sgad55 (lane 3), or sgad65 (lane 4) were used for immunoprecipitation using the same antibody used in A. Immunoprecipitates were then fractionated using SDS-PAGE and exposed to X-ray film.

199

Gene Therapy and Molecular Biology Vol 3, page 200

F i g u r e 3 . Histopathological examination of pancreatic islets. Islets with score 0 (A), and 3 (B ) from a GAD65-treated mouse are shown for comparison with islets with score 5 (C) and 6 (D) from a control animal. Arrows point to T cell infiltration.

B. Effects of intramuscular injection of GAD-encoding genes on insulitis and cytokine profile Each of four groups of three-week old female NOD mice received injections of one of four plasmid DNAs, and injections were repeated after three days. Mice were injected with either plasmid vector only, plasmid vector carrying the sruc3 gene encoding a stable mutant (JL and AE, unpublished data) of a secreted soft coral luciferase (Liu et al., 1997), the human gad65 gene, or the sgad55 gene encoding secreted GAD protein. The sruc3 gene was used as control for possible non-specific effects of synthesis of a plasmid-encoded antigen on insulitis. Another group of NOD mice was used as a non-injected control (N.B. this group was kept in a different animal room at Loma Linda University, and at a different time). Mice were killed when 10 weeks old for histopathological analysis, insulitis scoring, and immune assay. None of the mice had developed diabetes, as determined by urine and blood glucose analysis (data not shown). Figure 3 shows representative islets illustrating the levels of insulitis that were observed. Figure 4A shows that a significant reduction in the severity of insulitis was detected in mice receiving injections of plasmids carrying the gad65 and sgad55 genes, when

compared to the three control groups. In addition, mice injected with these gad genes also had a higher percentage of insulitis-free islets (F i g . 4B). Cytokine profiles of GAD65-stimulated spleen lymphocytes (splenocytes) tended to support the histological findings. While the Th1-type cytokines (IFN! and IL-2) were not different between groups (F i g . 5 A and B), IL-4 production (Th2-type) was higher in the gene-vaccinated groups (F i g . 5 C , 3 and 4) than in the controls (F i g . 5 C , 1 and 2), when challenged in vitro with recombinant human GAD65 protein.

III. Discussion Gene vaccination consists of the introduction and expression of a gene into an organism, with the purpose of generating an immune response against its encoded product. The simplest way of achieving this purpose is to use the method of intramuscular or subcutaneous “naked” DNA injection, originally presented as a means of expressing plasmid-encoded genes after direct injection of DNA into mouse muscle (Wolff et al., 1990). This method has since been used to generate immune responses to a wide variety of antigens, such as human immunodefficiency virus 1 glycoproteins, and malarial circumsporozoite protein (for a review see Tighe et al., 1998). Although the majority of studies have focused on infectious diseases, “naked” DNA

200

Gene Therapy and Molecular Biology Vol 3, page 201

Figure 4. Insulitis scores of 10-week old female NOD mice. Severity of insulitis is presented based on a 0-6 scale (A), and as percentage of islets showing intra-, peri-, or no insulitis (B ). Insulitis was scored with untreated mice (group 1, n= 5), mice receiving injections of plasmid vector only (group 2, n=5), or vector carrying gene sruc3 (group 3, n= 6), gene gad65 (group 4, n=7), or gene sgad55 (group 5, n=7). Data are presented as the mean score Âą SEM. When individually compared to control groups 1 (*), 2 (* *), or 3 (* * *), group 4 or 5 showed statistically significant differences (P value <0.05). No statistically significant differences were found among groups 1, 2, 3, or groups 4 and 5.

F i g u r e 5 . Cytokine profile of GAD65 protein-stimulated splenocytes. Splenocytes from mice receiving injections with plasmid vector only (1), or plasmid carrying the sruc3 (2), gad65 (3), or sgad55 (4) gene were stimulated with 1.5 Âľg/mL of isolated recombinant human GAD65 protein. After 72 hrs, culture supernates were assayed for IFN! (A), IL-2 (B ), and IL-4 (C).

gene vaccination can also be applied to studies of alloimmunity (Geissler et al., 1994) and treatment of cancer (Condon et al., 1996). Recently, gene vaccination was used to suppress the symptoms of autoimmune

encephalomyelitis in rats through synthesis autoantigenic peptide (Lobell et al., 1998).

of

an

Glutamic acid decarboxylase (GAD) is thought to be a major autoantigen contributing to the onset of insulin-

201

Liu et al: Intramuscular injection of glutamic acid decarboxylase DNA and insulitis dependent diabetes mellitus (IDDM), and injection of GAD protein can delay the onset of the disease in NOD mice (Kaufman et al., 1993; Tisch et al., 1993; Elliot et al., 1994; Petersen et al., 1994; Sai et al., 1996). In addition, similar results are obtained with oral feeding of the protein (Ma et al., 1997; Ramiya et al., 1997). In this work, we investigated whether expression of genes encoding two forms of human GAD could cause reduction of insulitis in the NOD mouse, an inflammation of pancreatic islets which is characteristic of IDDM in this model. Two gad genes were used for expression in muscle tissue, a gad65 cDNA encoding human GAD65 protein, and sgad55, a gene construct based on gad65, encoding the interleukin-2 (IL-2) leader sequence fused to a GAD65 protein lacking its first 88 amino acids (SGAD55) (F i g .1). The N-terminal region of GAD65 was removed in the SGAD55 fusion protein because we suspected that it could interfere with its secretion, since this region contains a Golgi-targeting and membrane-anchoring sequence. This was confirmed by the finding that a fusion of the IL-2 leader sequence to full-length GAD65 could not be detected in the culture media of mammalian cells expressing its encoding gene (Figure 2B, lane 4). In contrast, SGAD55 protein was detected in culture media of cells expressing the sgad55 gene (F i g . 2 B , lane 3). Intracellular and secreted forms of GAD65 were used in this study because of the known differences in the type of immune response that intra- and extra-cellular antigens can generate. Intracellular antigens are presented by MHC class I molecules and generate a CD8+ cytotoxic T lymphocyte response, while extracellular antigens are presented by MHC class II molecules on the surface of antigen presenting cells, generating a CD4+ helper T lymphocyte response (Tighe et al., 1998). Although secreted proteins are synthesized within a cell, they appear to be less likely to be presented by MHC class I molecules than cytosolic proteins (Yewdell et al., 1998). Synthesis of intracellular GAD65 by muscle cells was not expected to effect insulitis. Considering the small number of muscle cells able to express injected genes, the levels of intracellular GAD65 protein found in injected and non-injected mice NOD should not have differed greatly enough to generate an immune response influencing T cell infiltration of islets. This supposition was corroborated by the finding that in NOD mice transgenic for murine gad65, only those mice showing the highest levels of transgene expression could exacerbate insulitis and diabetes (Geng et al., 1998). In contrast, secretion of SGAD55 could have caused either reduced insulitis (through an antiinflammatory Th2 response) or increased insulitis (through an inflammatory Th1 response), depending on the levels of extracellular antigens attained (Hosken et al., 1995). Our results show that injections of gad65-carrying

plasmids caused a reduction of insulitis similar, if not greater, to that resulting from injections of sgad55-carrying plasmid (F i g . 4). A variety of non-exclusive mechanisms are thought to lead to Th1 and Th2 immune responses after plasmid DNA injection, such as release of antigens from intact cells expressing the plasmid-carried gene, or from lysed cells after a cytotoxic T lymphocyte response, and direct transfection of antigen presenting cells (Davis et al., 1993; Xiang et al., 1994; Condon et al., 1996; Gregoriadis et al., 1997). Our results suggest that one (or both) of the latter two putative mechanisms was likely to be responsible for the similar reduction of insulitis after intramuscular injection of gad65 or sgad55 gene, since neither the intranor extra-cellular nature of the plasmid-encoded GAD antigens appeared to affect the extent of insulitis differently. To determine the nature of the immune response generated in plasmid-treated NOD mice, cytokine secretion by splenocyte was measured after challenge with recombinant human GAD65 protein. Splenocytes of all plasmid-injected mice secreted similar levels of Th1-specific IFN! (F i g . 5A) and IL-2 (F i g . 5B). However, cells from mice receiving injections of gad65 or sgad55 genes maintained higher levels of Th2-specific IL-4 than the controls when challenged with GAD65 autoantigen (F i g . 5C). No differences in splenic memory cell numbers (immunophenotyping) or blood levels of cytokines could be demonstrated between groups (data not shown). These results suggest that the reduced levels of insulitis observed after injection of gad65 or sgad55 genes could have been the result of a Th2-mediated response. This would be in accordance with the observation that suppression of insulitis is associated with elevated synthesis of IL-4 and IL-10 (for a review see Rabinovitch, 1998), and that suppression of the diabetogenic response in NOD mice after injection of GAD65 protein is mediated by the induction of GAD65-specific regulatory Th2 cells (Tisch et al., 1998). Failure to detect systemic changes in circulating cytokine levels (IFN! 75-1365 pg/mL) and splenic phenotypes (1315% CD62Lneg CD44pos T helper cells) was expected in light of the well-established organ-specificity of this disease. Further work on the cellular infiltrate should reveal more relevant information. Immunomodulatory gene therapy has been previously considered as a possible approach for the prevention of IDDM. In one study, islet-specific Th1 cells transduced with engineered retroviruses carrying a gene encoding the anti-inflammatory cytokine IL-10 were able to cause reduced insulitis and delayed onset of diabetes when injected into NOD mice (Moritani et al., 1996). In contrast, intramuscular injection of plasmid DNA encoding IL-10 did not cause reduced insulitis, but did result in delay of diabetes onset (Nitta et al., 1998). In another study, intramuscular injection of DNA encoding TGF-"1 caused both reduced insulitis and delayed onset of diabetes

202

Gene Therapy and Molecular Biology Vol 3, page 203 (Piccirillo et al, 1998). Our data suggest that intramuscular injection of DNA coding for an IDDM autoantigen could also be used for this purpose. Plasmid injection offers potentially both therapeutic and economical advantages. Injection of plasmid DNA could permit the development of plasmid “cocktails” encoding combinations of different autoantigens and immunomodulating cytokines. When compared to injection of isolated proteins, the availability, quality, and cost of these therapeutic proteins would not be a concern, since their synthesis would occur within the host. Clearly, injection of plasmid DNA is a promising approach for suppressing symptoms of IDDM or other autoimmune diseases in the future.

C. Immunoblot analysis Simian COS-7 cells were washed twice with cold PBS 48 hrs after transfection, and harvested in 100 µl hot 2x gel-loading buffer (100mM Tris.HCl, 4% SDS, 20% glycerol, 10% 2mercaptoethanol, 0.01% bromophenol blue) using a rubber policeman. Cells were lysed by sonication, boiled for 3 min, and lysates were centrifuged at 1000x g for 10 min to pellet cell debris. Twenty microliters from each sample was loaded on a 12% SDS-polyacrylamide gel for fractionation. Proteins were then transferred onto a nylon membrane by electroblotting, and a GAD65 monoclonal antibody was used to detected GAD protein using a method previously described (Liu et al., 1997).

D. Protein radiolabeling and immunoprecipitation

IV. Materials and Methods A. Gene and plasmid construction The sgad65 gene encodes a fusion of the leader peptide from human IL-2 to full-length human GAD65 protein. This gene was constructed by ligating an 89 base pair (bp) DNA fragment encoding the first 23 amino acids of IL-2 (isolated previously by PCR from human cell line A293 as described by Liu et al., 1997) in frame with a 1.8 Kilobase pair (Kb) NcoI-XhoI DNA fragment carrying a human GAD65 cDNA. The sgad55 gene encodes a fusion of the leader peptide from human IL-2 to a truncated version of human GAD65 with 88 aa deleted at its Nterminus. Two oligonucleotides were used to amplify the 89 bp DNA fragment encoding the IL-2 leader peptide from gene sgad65 , IL-01 (TTT TCT AGA ATG TAC AGG ATG CAA CTC CTG) and IL-03 (TTT ACG CGT AAG TAG GTG CAC TGT TTG TGA). IL-03 introduced an MluI site which was used to clone the PCR product in frame with the MluI-XhoI 1.5 Kb DNA fragment encoding GAD55, the truncated version of human GAD65. The identity of PCR products and gene fusion junctions were confirmed using automated DNA sequencing. For cell culture work, the gad65, sgad65, and sgad55 genes were cloned under transcriptional control of the cytomegalovirus (CMV) promoter into plasmid vector pLNCX (Miller and Rosman, 1989). For muscle injection, all genes were cloned under transcriptional control of the CMV promoter in plasmid pND-2, a vector known to provide high gene expression in muscle tissues (Gary Rhodes and Robert Malone, unpublished data).

B. Mammalian cell culture and transfection Simian COS-7 cells were grown in 60 mm tissue culture dishes containing 3 mL DMEM medium with 10% fetal bovine serum (FBS). Media were changed 3 hrs prior to transfection when cells were 70% confluent. Cell transfection was performed using the ProFection calcium phosphate system (Promega, Madison, WI) using 40 µg of plasmid DNA per plate. Cells were incubated with the DNA-calcium phosphate complex for 6 hours, washed twice with phosphate-buffered saline (PBS), and 3 mL DMEM medium +10% FBS was added into each plate. Culture plates were then incubated for 48 hrs before harvesting cells and media for analysis.

To detect secreted GAD proteins in cell culture media, 35 Smethionine (specific activity>1000 Ci/mmol, from DuPont NEN, Boston, MA) was used to label total cell protein from COS-7 cells. Media were removed after incubation with the DNA-calcium phosphate complex, and cells were rinsed twice with 1 x PBS and once with medium without methionine and serum. Cells were then incubated in 3 mL DMEM medium without methionine + 1% dialyzed FBS for 20 min to deplete intracellular pools of endogenous methionine. One hundred microcuries of 35 S-methionine was then added directly into media for protein labeling. Cells were incubated for 24 hrs before being harvested. Media were collected and concentrated using a Centricon spin column (15 kDa molecular weight cutoff, from Amicon) to a final volume of 500 µL. 35 S-labeled GAD protein was immunoprecipitated from media using the Protein A Immunoprecipitation Kit (Boehringer Mannheim, Indianapolis, IN) and a monoclonal antibody raised against human GAD65. Immunoprecipitates were fractionated in a 12% SDS-polyacrylamide gel, and protein bands were detected by exposure to X-ray film.

E. Isolation of plasmid DNA for muscle injection Plasmid DNA was amplified in Escherichia coli strain DH5#, using the alkaline-lysis method, and isolated by standard double-round cesium chloride purification (Maniatis et al., 1989). The quality and quantity of DNA was determined by U.V. spectrophotometry (A260 /A280 ratio greater than 1.8) and by agarose gel electrophoresis. Plasmid DNA was dissolved under sterile conditions in double distilled water at a final concentration of 2 µg/ µL, and stored at -20o C.

F. Intramuscular DNA injection Three-week old female NOD mice were purchased from Taconic Laboratories (Germantown, NY) and kept at Loma Linda University animal facilities. Mice were injected with DNA (200 µg/100 µL/leg) into each quadriceps muscle with a 27-gauge needle under general anaesthesia (Ketamine, 66 mg/Kg body weight, from Phoenix Scientific, St Joseph, MO; Oxylazine, 7.5 mg/Kg body weight, from LLOYD Laboratories,

203

Liu et al: Intramuscular injection of glutamic acid decarboxylase DNA and insulitis Shenandoa, IO; and Acepromazine Maleate, 1.5 mg/Kg body weight, from Fermenta Animal Health Co., MO), and injections were repeated three days later. Urine glucose levels were monitored weekly with Clinistix Reagent Strips for Urinalysis (Bayer Corporation, Elkhart, IN). Mice were killed for insulitis scoring at the age of 10 weeks, and blood glucose levels were checked with ACC.-CHEK Advantage (Boehringer Mannheim Corporation, Indianapolis, IN).

G. Histopathological analysis of insulitis Pancreatic tissues were fixed with 10% buffered Formalin, stained with hematoxylin, and counterstained with eosin, and an average of fifteen islets/mouse were scored. A 7-level semiquantitative scoring scale (Zhang et al., 1991) was used for insulitis scoring: 0, normal islet tissue without any detectable T cell infiltration; 1, focal peri-islet T cell infiltration with less than one-third of the peri-islet area; 2, more extensive peri-islet T cell infiltration with less than two-thirds of the peri-islet area; 3, peri-islet T cell infiltration with more than two-thirds of the peri-islet area; 4, intra-islet T cell infiltration with less than one-third of the islet area; 5, intra-islet T cell infiltration with less than two-thirds of the islet area; 6, severe intra-islet T cell infiltration with more than two-thirds of the islet area. Scoring of 1-3 indicated peri-insulitis, and scoring of 4-6 indicated intra-insulitis. Scoring was conducted using the double-blind method by two different scorers.

H. In vitro challenge of splenocytes Lymphocytes were flushed from splenic pulp and washed in complete media (RPMI, 10% FBS, 2% L-Glutamine, and 4 x 105 M 2-mercaptoethanol). In a 24-well plate, 1x106 cells in 1 mL complete media (unstimulated control) or 1 mL GAD65 (1.5 Âľg/mL) were cultured (37o C, 5% CO2 ) for 72 hrs. Cell culture supernatants and blood plasma (from terminal bleeds) were assayed by standard sandwich ELISA (Endogen, Woburn, MA) for IFN!, IL-2 and IL-4.

I. Immunophenotyping of splenocytes Since antigen-specific memory cells of the Th1-type T lymphocytes express CD44 and lose expression of CD62 ligand (CD62L) (Mocci and Coffman, 1997; Bradley et al., 1992), splenocytes were stained with three fluorochromeconjugated monoclonal antibodies (Becton Dickinson, Immunocytochemistry Systems, San Jose, CA) to CD4, CD44 and CD62L. After red cells lysis, the phenotypes were analyzed by 3-color flow cytometry. Phenotyping controls included untreated (autofluorescence) and isotype antibody-treated cells (nonspecific staining). CD4 T cells (phycoerythrin) were backgated and these were analyzed for the expression of CD44 (PerCP) and CD62L (FITC).

J. Statistical analysis Comparison between groups was done using a ONE-WAY ANOVA and Duncans post-hoc test for multiple comparisons.

Acknowledgements The authors would like to thank John Elliot for giving the GAD65 cDNA, isolated GAD protein, and monoclonal GAD65 antibody, Robert Malone for the gift of plasmid pND2, and William H.R. Langridge for helpful comments during preparation of the manuscript. This work was made possible by funds from Loma Linda University Medical School, and by a grant from the National Medical Technology Testbed to Loma Linda University. The view, opinions and/or findings contained in this report are those of the authors and should not be construed as a position, policy, decision or endorsement of the Federal Government the National Medical Technology Testbed Inc.

References Baekkeskov S, Aanstoot HJ, Christgau S, Reetz A, Solimena M, Cascalho M, Folli F, Richter-Olesen H, DeCamilli P. (1 9 9 0 ) Identification of the 64K autoantigen in insulindependent diabetes as the GABA-synthesizing enzyme glutamic acid decarboxylase. Nature 347, 151-156. Baxter AG, Koulmanda M, Mandel TE. (1 9 9 1 ). High and low diabetes incidence nonobese diabetic (NOD) mice, origins and characterisation. Autoimmunity 9, 61-67. Bottazzo GF, Florin-Christensen A, Doniach D. (1 9 7 4 ). Isletcell antibodies in diabetes mellitus with autoimmune polyendocrine deficiencies. Lancet 2, 1279-1283. Bowman MA, Leiter EH, Atkinson MA. (1 9 9 4 ). Prevention of diabetes in the NOD mouse, implications for therapeutic intervention in human disease. Immunol Today 15, 115-120 . Bradley LM, Atkins GG, Swain SL. (1 9 9 2 ). Long-term CD4 + memory T cells from the spleen lack MEL-14, the lymph node homing receptor. J Immunol 148, 324-331. Christgau S, Aanstoot HJ, Schierbeck H, Begley K, Tullin S, Hejnaes K, Baekkeskov S. (1 9 9 2 ). Membrane anchoring of the autoantigen GAD65 to microvesicles in pancreatic beta-cells by palmitoylation in the NH2-terminal domain. J C e l l B i o l 118, 309-320. Christgau S, Schierbeck H, Aanstoot HJ, Aagaard L, Begley K, Kofod H, Hejnaes K, Baekkeskov S. (1 9 9 1 ). Pancreatic beta cells express two autoantigenic forms of glutamic acid decarboxylase, a 65-kDa hydrophilic form and a 64-kDa amphiphilic form which can be both membrane-bound and soluble. J B i o l C h e m 266, 21257-21264. Condon C, Watkins SC, Celluzzi CM, Thompson K, Falo LD Jr. (1 9 9 6 ). DNA-based immunization by in vivo transfection of dendritic cells. Nat Med 2, 1122-1128. Davis HL, Demeneix BA, Quantin B, Coulombe J, Whalen RG. (1 9 9 3 ). Plasmid DNA is superior to viral vectors for direct gene transfer into adult mouse skeletal muscle. Hum Gene Ther 4, 733-740. Elliott JF, Qin HY, Bhatti S, Smith DK, Singh RK, Dillon T, Lauzon J, Singh B. (1 9 9 4 ). Immunization with the larger

204

Gene Therapy and Molecular Biology Vol 3, page 205 isoform of mouse glutamic acid decarboxylase (GAD67) prevents autoimmune diabetes in NOD mice. D i a b e t e s 43, 1494-1499.

Maniatis T, Fritsch EF, Sambrook J. (1 9 8 9 ). Molecular C l o n i n g . A Laboratory Manual, second edition. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Erlander MG, Tillakaratne NJ, Feldblum S, Patel N, Tobin AJ. (1 9 9 1 ). Two genes encode distinct glutamate decarboxylases. Neuron 7, 91-100.

Miller AD, Rosman GJ. (1 9 8 9 ). Improved retroviral vectors for gene transfer and expression. B i o t e c h n i q u e s 7, 980990.

Geissler EK, Wang J, Fechner JH Jr, Burlingham WJ, Knechtle SJ. (1 9 9 4 ). Immunity to MHC class I antigen after direct DNA transfer into skeletal muscle. J I m m u n o l 152, 413421.

Mocci S, Coffman RL. (1 9 9 7 ). The mechanism of in vitro T helper cell type 1 to T helper cell type 2 switching in highly polarized Leishmania major-specific T cell populations. J Immunol 158, 1559-1564.

Geng L, Solimena M, Flavell RA, Sherwin RS, Hayday AC. (1 9 9 8 ). Widespread expression of an autoantigen-GAD65 transgene does not tolerize non-obese diabetic mice and can exacerbate disease. P r o c N a t l A c a d S c i U S A 95, 10055-10060.

Molbak AG, Christau B, Marner B, Borch-Johnsen K, Nerup J. (1 9 9 4 ). Incidence of insulin-dependent diabetes mellitus in age groups over 30 years in Denmark. Diabet Med 11, 650-655.

Gepts W, Lecompte PM. (1 9 8 1 ). The pancreatic islets in diabetes. Am J Med 70, 105-115. Gregoriadis G, Saffie R, de Souza JB. (1 9 9 7 ). Liposomemediated DNA vaccination. FEBS Lett 402, 107-110. Honeyman MC, Cram DS, Harrison LC. (1 9 9 3 ). Glutamic Acid decarboxylase 67-reactive T cells, a marker of insulindependent diabetes. J Exp Med 177, 533-540. Hosken NA, Shibuya K, Heath AW, Murphy KM, O'Garra A. (1 9 9 5 ). The effect of antigen dose on CD4+ T helper cell phenotype development in a T cell receptor-alpha betatransgenic model. J Exp Med 182, 1579-1584. Itoh N, Hanafusa T, Miyazaki A, Miyagawa J, Yamagata K, Yamamoto K, Waguri M, Imagawa A, Tamura S, Inada M, et al. (1 9 9 3 ). Mononuclear cell infiltration and its relation to the expression of major histocompatibility complex antigens and adhesion molecules in pancreas biopsy specimens from newly diagnosed insulin-dependent diabetes mellitus patients. J C l i n I n v e s t 92, 23132322. Kaufman DL, Clare-Salzler M, Tian J, Forsthuber T, Ting GS, Robinson P, Atkinson MA, Sercarz EE, Tobin AJ, Lehmann PV. (1 9 9 3 ). Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulindependent diabetes. Nature 366, 69-72. Liu J, O'Kane DJ, Escher A. (1 9 9 7 ). Secretion of functional Renilla reniformis luciferase by mammalian cells. G e n e 203, 141-148. Lobell A, Weissert R, Storch MK, Svanholm C, de Graaf KL, Lassmann H, Andersson R, Olsson T, Wigzell H. (1 9 9 8 ). Vaccination with DNA encoding an immunodominant myelin basic protein peptide targeted to Fc of immunoglobulin G suppresses experimental autoimmune encephalomyelitis. J Exp Med 187, 1543-1548. Ma SW, Zhao DL, Yin ZQ, Mukherjee R, Singh B, Qin HY, Stiller CR, Jevnikar AM. (1 9 9 7 ). Transgenic plants expressing autoantigens fed to mice to induce oral immune tolerance. Nat Med 3, 793-796. Maclaren NK. (1 9 8 8 ). How, when, and why to predict IDDM. D i a b e t e s 37, 1591-1594 Mandrup-Poulsen T. (1 9 9 8 ). Diabetes. BMJ 316, 1221-1225.

Moritani M, Yoshimoto K, Ii S, Kondo M, Iwahana H, Yamaoka T, Sano T, Nakano N, Kikutani H, Itakura M. (1 9 9 6 ). Prevention of adoptively transferred diabetes in nonobese diabetic mice with IL-10-transduced islet-specific Th1 lymphocytes. A gene therapy model for autoimmune diabetes. J C l i n I n v e s t 98, 1851-1859. Mosmann TR, Sad S. (1 9 9 6 ). The expanding universe of T-cell subsets, Th1, Th2 and more. I mmunol T oday 17, 138146. Nitta Y, Tashiro F, Tokui M, Shimada A, Takei I, Tabayashi K, Miyazaki J. (1 9 9 8 ). Systemic delivery of interleukin 10 by intramuscular injection of expression plasmid DNA prevents autoimmune diabetes in nonobese diabetic mice. Hum Gene Ther 9, 1701-1707. Okano K, Aoki Y, Shimizu H, Naruto M. (1 9 9 0 ). Functional expression of human leukocyte elastase (HLE)/medullasin in eukaryotic cells. B i o c h e m B i o p h y s Res Commun 167, 1326-1332 . Panina-Bordignon P, Lang R, van Endert PM, Benazzi E, Felix AM, Pastore RM, Spinas GA, Sinigaglia F. (1 9 9 5 ). Cytotoxic T cells specific for glutamic acid decarboxylase in autoimmune diabetes. J Exp Med 181, 1923-7. Peakman M, Wen L, McNab GL, Watkins PJ, Tan KC, Vergani D. (1 9 9 4 ). T cell clones generated from patients with type 1 diabetes using interleukin-2 proliferate to human islet antigens. Autoimmunity 17, 31-39. Petersen JS, Karlsen AE, Markholst H, Worsaae A, Dyrberg T, Michelsen B. (1 9 9 4 ). Neonatal tolerization with glutamic acid decarboxylase but not with bovine serum albumin delays the onset of diabetes in NOD mice. D i a b e t e s 43, 1478-1484. Piccirillo CA, Chang Y, Prud'homme GJ. (1 9 9 8 ). TGF-beta1 somatic gene therapy prevents autoimmune disease in nonobese diabetic mice. J Immunol 161, 3950-3956. Rabinovitch A. (1 9 9 8 ). An update on cytokines in the pathogenesis of insulin-dependent diabetes mellitus. D i a b e t e s M e t a b R e v 14, 129-151. Ramiya VK, Shang XZ, Wasserfall CH, Maclaren NK. (1 9 9 7 ). Effect of oral and intravenous insulin and glutamic acid decarboxylase in NOD mice. Autoimmunity 26, 139151.

205

Liu et al: Intramuscular injection of glutamic acid decarboxylase DNA and insulitis Richter W, Shi Y, Baekkeskov S. (1 9 9 3 ). Autoreactive epitopes defined by diabetes-associated human monoclonal antibodies are localized in the middle and C-terminal domains of the smaller form of glutamate decarboxylase. Proc Natl Acad Sci USA 90, 2832-2836. Sai P, Rivereau AS, Granier C, Haertle T, Martignat L. (1 9 9 6 ). Immunization of non-obese diabetic (NOD) mice with glutamic acid decarboxylase-derived peptide 524-543 reduces cyclophosphamide-accelerated diabetes. C l i n E x p Immunol 105, 330-337. Solimena M, Dirkx R Jr, Radzynski M, Mundigl O, De Camilli P. (1 9 9 4 ). A signal located within amino acids 1-27 of GAD65 is required for its targeting to the Golgi complex region. J C e l l B i o l 126, 331-341. Tabata H, Kanai T, Yoshizumi H, Nishiyama S, Fujimoto S, Matsuda I, Yasukawa M, Matsushita S, Nishimura Y. (1 9 9 8 ). Characterization of self-glutamic acid decarboxylase 65-reactive CD4+ T-cell clones established from Japanese patients with insulin-dependent diabetes mellitus. Hum Immunol 59, 549-560. Tighe H, Corr M, Roman M, Raz E. (1 9 9 8 ). Gene vaccination, plasmid DNA is more than just a blueprint. Immunol Today 19, 89-97. Tisch R, Liblau RS, Yang XD, Liblau P, McDevitt HO. (1 9 9 8 ). Induction of GAD65-specific regulatory T-cells inhibits ongoing autoimmune diabetes in nonobese diabetic mice.

D i a b e t e s 47, 894-899. Tisch R, Yang XD, Singer SM, Liblau RS, Fugger L, McDevitt HO. (1 9 9 3 ). Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 366, 72-75. Wolff JA, Malone RW, Williams P, Chong W, Acsadi G, Jani A, Felgner PL. (1 9 9 0 ). Direct gene transfer into mouse muscle in vivo. S c i e n c e 247, 1465-1468. Xiang ZQ, Spitalnik S, Tran M, Wunner WH, Cheng J, Ertl HC. (1 9 9 4 ). Vaccination with a plasmid vector carrying the rabies virus glycoprotein gene induces protective immunity against rabies virus. V i r o l o g y 199, 132-140. Yewdell JW, Snyder HL, Bacik I, Anton LC, Deng Y, Behrens TW, Bachi T, Bennink JR. (1 9 9 8 ). TAP-independent delivery of antigenic peptides to the endoplasmic reticulum, therapeutic potential and insights into TAPdependent antigen processing. J Immunother 21, 127131. Zekzer D, Wong FS, Ayalon O, Millet I, Altieri M, Shintani S, Solimena M, Sherwin RS. (1 9 9 8 ). GAD-reactive CD4+ Th1 cells induce diabetes in NOD/SCID mice. J C l i n I n v e s t 101, 68-73. Zhang ZJ, Davidson L, Eisenbarth G, Weiner HL. (1 9 9 1 ). Suppression of diabetes in nonobese diabetic mice by oral administration of porcine insulin. P r o c N a t l A c a d S c i USA 88, 10252-10256.

206

Gene Therapy and Molecular Biology Vol 3, page 207 Gene Ther Mol Biol Vol 3, 207-221. August 1999.

Muscle-based tissue engineering for the musculoskeletal system Review Article

Douglas S. Musgrave, and Johnny Huard* Department of Orthopaedic Surgery and Molecular Genetics and Biochemistry, Children's Hospital of Pittsburgh and University of Pittsburgh, Pittsburgh, PA, 15261 USA _________________________________________________________________________________________________ *Correspondence: Johnny Huard, Ph.D., Asssistant Professor, Director, Growth & Development Laboratory, Department of Orthopaedic Surgery and Molecular Genetics and Biochemistry, Children's Hospital of Pittsburgh and University of Pittsburgh, Pittsburgh, PA 15261 USA. Tel: (412)-692-7807; Fax: (412)-692-7095; E-mail: jhuard+@pitt.edu Key Words: Gene therapy, muscle-derived cells, bone, Duchenne Muscular Dystrophy, ligament, meniscus, muscle injuries, dystrophin, bone morphogenetic protein, myoblast transplantation, muscle repair, bone healing, arthritis Abbreviations: DMD, Duchenne muscular dystrophy; AAV, adeno-associated virus; BMPs, bone morphogenetic proteins; rhBMP, recombinant human BMP; TGF- , transforming growth factor-!; ACL, anterior cruciate ligament Received: 3 December 1998; accepted: 12 December 1998

Summary Somatic gene therapy through the transfer of genes into a particular tissue to alleviate a biochemical deficiency has emerged as a novel and exciting form of molecular medicine. Due to a number of factors, muscle tissue has emerged as a promising target for muscle based gene therapy and tissue engineering. First, many muscle groups are readily accessible and tolerate delivery by injection well. Second, muscle is composed of multinucleated, postmitotic myofibers and may facilitate high and long term persistence of transgene expression. Third, muscle can be easily and repeatedly biopsied without compromising the health and function of human and animal subjects. Finally, muscle is very well vascularized, making systemic delivery through the bloodstream feasible. Based on these unique features of the skeletal muscle, we have described four different applications of muscle based gene therapy and tissue engineering: inherited muscle diseases, muscle injury and repair, bone healing and finally intra-articular disorders. Since the field of muscle based gene therapy and tissue engineering has expanded and matured over the last few years, we will review some hurdles facing the practical application of this technology as well as potential approaches to circumvent these limitations to eventually apply this technology to the treatment of pathologies and conditions of the musculoskeletal system.

for musculoskeletal disorders and discusses the focus of ongoing research.

I. Introduction The advent of gene therapy and tissue engineering has facilitated novel approaches to the treatment of musculoskeletal disorders. The delivery of growth factors, cells, and therapeutic genes promises to revolutionize a medical field historically limited to biomechanical approaches. Significant scientific contributions have been made in the last three decades toward the understanding of skeletal muscle biology and its potential therapeutic applications. However, despite the tremendous progress, many questions currently remain unanswered. This paper reviews the current status of muscle-based tissue engineering

As the molecular basis of an expanding number of inherited disorders has been discovered, increasing focus has been placed on gene therapy as a potential therapeutic approach. The transfer of a functional gene into a particular tissue has been explored in many disease systems using a variety of gene delivery approaches. Human inherited disorders of muscle are not uncommon diseases of childhood. Hence, skeletal muscle has been studied as a target tissue for the delivery of genes encoding proteins that may be therapeutic for inherited muscle disorders. However, since the multinucleated and post-mitotic myofibers in 207

Musgrave and Huard: Muscle-based tissue engineering for the musculoskeletal system skeletal muscle are capable of both long-term transgene expression and systemic delivery of proteins to the blood circulation, direct and ex vivo gene transfer to skeletal muscle has also been investigated as a means to create a tissue reservoir for the secretion of non-muscle proteins (Dhawan et al., 1991; Dai et al., 1992; Lynch et al., 1992; Jiao et al., 1993; Simonsen et al., 1996; Lau et al., 1996; Bosch et al., 1998; Musgrave et al., 1998).

1998). Based on these unique characteristics, the field of skeletal muscle-derived cells in muscle-based tissue engineering is burgeoning. In this chapter, we summarize muscle based tissue engineering applications for the musculoskeletal system including inherited muscle disease (Duchenne Muscular Dystrophy), muscle injuries and repair, bone healing, and intraarticular disorders.

The direct gene therapy approach, albeit technically straightforward, presents theoretical risks of in vivo genetic manipulation and possible reversion to pathogenicity of attenuated viral vectors. Furthermore, the direct approach does not provide for the introduction of cells capable of participating in the healing response. The ex vivo approach addresses these issues by limiting genetic manipulation of the cells to the culture flask thereby eliminating the potential risks of in vivo genetic manipulation and viral reversion. The ex vivo approach also allows for isolation and expansion of muscle-derived cells possibly capable of participating in the therapeutic process. Evidence exists that suggests musclederived cells can participate in both muscle and bone healing (Urist, 1965; Huard et al., 1992a,b; 1994a,b; Bosch et al., 1998). Finally, myoblast transplantation is a clinically feasible approach to delivering competent cells with complementary genomes to patients with inherited muscle diseases such as Duchenne muscular dystrophy (Huard et al, 1992a,b; 1994a,b). Therefore, current muscle-based tissue engineering approaches are aimed at both inherited and acquired musculoskeletal disorders.

II. Muscle based gene therapy for inherited muscle diseases The muscular dystrophies " which were the first target for gene therapy to skeletal muscle " are characterized by progressive muscle wasting and weakness. Duchenne muscular dystrophy (DMD), inherited on the X chromosome, is one of the most common and severe inherited myopathies. DMD is a devastating muscle disease characterized by a lack of dystrophin expression in the sarcolemma of muscle fibers (Hoffman et al., 1987; Arahata et al., 1988; Sugita et al., 1988; Zubryzcka-Gaarn et al., 1988). Dystrophin, one of the largest known human genes, has a high frequency of mutation affecting 1 in 3,500 males. Dystrophin appears to function in the maintenance of muscle membrane integrity. Its absence in DMD muscle causes damage to the membrane during muscle contraction, resulting in eventual muscle fiber necrosis (Bonilla et al., 1988; Watkins et al., 1988; Menke et al., 1991). There is no treatment, and affected children die in their late teens of cardiac and respiratory failures. Because genetic testing and counseling does not dramatically lower the incidence of this disorder, it is crucial to develop therapeutic approaches to alleviate the muscle weakness in DMD patients. The ultimate goal of therapy for DMD is to provide enough dystrophin to the membrane cytoskeleton of the majority of the DMD muscle fibers to be therapeutically effective. Various approaches have been explored to transfer dystrophin into skeletal muscle, including myoblast transplantation and gene delivery based on non-viral vectors (direct DNA injection, liposome) and viral (retrovirus, adenovirus and herpes simplex virus) vectors.

The theory behind muscle-based tissue engineering is predicated on the unique biology of skeletal muscle derived cells. First, as discussed below, skeletal muscle contains satellite cells. These cells are resting mononucleated precursor cells capable of fusing to form post-mitotic, multinucleated myotubes and myofibers. The post-mitotic, multinucleated myofibers are stable cells theoretically capable of persistent gene expression. Therefore, by focusing tissue engineering approaches on the satellite cell, one may be capable of maximizing the degree and persistence of gene expression. Second, as alluded to earlier, skeletal muscle may contain a population of mesenchymal stem cells. Mesenchymal stem cells are resting cells capable of differentiation into several different lineages (Caplan, 1991). In vitro (Katagiri et al., 1994; Young et al., 1995; Warejcka et al., 1996) and in vivo (Bosch et al., 1998) data suggest cells residing within skeletal muscle are capable of differentiation into several different tissue lineages. Consequently, muscle-derived cells may be capable of regenerating many different tissues. Tissue engineering based on these cells not only facilitates gene delivery but may also supply the needed stem cells for healing. Finally, muscle-derived cells are clinically accessible and reliably isolated. Skeletal muscle biopsies are of low morbidity and available on an outpatient basis. Furthermore, in vitro isolation of muscle-derived cells has been well described (Blau and Webster, 1981; Rando and Blau, 1994; Qu et al.,

A. Myoblast transplantation Myoblast transplantation (MT) consists of the implantation of normal myoblasts into dystrophic muscles to create reservoirs of muscle cells capable of dystrophin expression (Watt et al., 1982; 1984; Morgan et al., 1988; 1990;1993; Allamedine et al., 1989; 1990; Law et al., 1988; Karpati et al., 1989; Partridge et al., 1989; 1991). MT in animal models, as well as in DMD patients, is capable of delivering dystrophin and occasionally improving muscle strength, but is hindered by immune barriers, poor dispersion

208

Gene Therapy and Molecular Biology Vol 3, page 209 of the injected myoblasts, and rapid loss of the injected cells (Gussoni et al., 1992; Huard et al., 1992a,b; 1994a,b; Karpati et al., 1992; Tremblay et al., 1993a,b; Kinoshita et al., 1994; Vilquin et al., 1995a; Guerette et al 1997; Qu et al, 1998).

which permits the reintroduction of myoblasts expressing dystrophin, can be useful for DMD patients, especially for those over 10 years of age whose muscle regeneration has become inefficient due to a lack of viable satellite cells. This method was performed using adenovirus, retrovirus, and HSV-1 carrying reporter genes (!"galactosidase or luciferase) and showed that transduced myoblasts (isogenic myoblasts) fused and reintroduced the reporter genes into the injected muscle, demonstrating the feasibility of the ex vivo approach (Salvatori et al, 1993; Rando et al, 1994; Huard et al., 1994c, Booth et al, 1997, van Deutekom et al, 1998a,b). We have recently observed that ex vivo gene transfer can deliver dystrophin in mdx (dystrophin deficient) muscle, but the immune responses against the transduced cells remain (Floyd et al., 1997, 1998).

B. Gene therapy Since the efficiency of gene therapy (GT) using naked DNA has been very limited (Acsadi et al., 1991; Danko et al., 1993; Davies et al., 1993), a virus-mediated gene delivery system may provide a promising alternative for dystrophin gene delivery. However, gene transfer via recombinant viral vectors has also been limited by numerous technical problems. Current retroviral vectors have not been found to transduce muscle fibers since they require dividing cells for integration and expression (Dunckley et al., 1992). However, an intermediate level of retroviral transduction occurs in immature and adult regenerating muscles which is likely due to myoblast mediation (Dunckley et al, 1992, 1993, van Deutekom et al, 1998a,b). Although adenoviral vectors can deliver genes to post-mitotic cells including myoblasts and newborn muscle fibers, the efficiency of gene transfer to mature muscle fibers is severely reduced (Quantin et al., 1992; Ragot et al., 1993; Vincent et al., 1993; Acsadi et al., 1994a,b, 1995; Huard et al 1995a, van Deutekom et al, 1998a,b). Moreover, gene delivery mediated by a first generation adenoviral vector induces immune responses to the vector, leading to rejection of the transduced cells (Smith et al., 1993; Engelhardt et al., 1994a,b; Yang et al., 1994a,b; Vilquin et al., 1995b).

D. Barriers to viral gene transfer of mature myofibers Viral vectors which cannot transduce post-mitotic cells, such as retrovirus, are consequently incapable of directly infecting post-mitotic myofibers. However, adenovirus and HSV-1 can infect post-mitotic cells but still poorly transduce mature muscle fibers due to different mechanisms. Our hypothesis is that adenoviral transduction of both immature and mature myofibers is mediated at least in part by fusion of infected myoblasts. Neonatal muscle is efficiently transduced due to continued fusion of myoblasts during muscle growth, while mature myofibers are not efficiently transduced due to a lack of myoblast fusion. Our experiments suggest that adenovirus requires transduction of myoblasts prior to fusion with myotubes or myofibers in order to transduce these differentiated muscle cells (van Deutekom et al., 1998a,b). By using pure cultures of myoblasts and myotubes, we have observed that adenovirus efficiently infects myoblasts but poorly infects myotubes. However, adenovirus transduces large numbers of mononucleated cells remaining in the differentiated muscle cell cultures. We have also shown that irradiation of newborn muscles prior to transduction inactivates myoblasts in vivo and significantly decreases the level of adenovirus transduction in neonatal myofibers in vivo (van Deutekom et al., 1998a,b). Alternatively, we have used isolated mature myofibers as a model to evaluate the efficiency of viral gene delivery in vitro. We have shown that the maturationdependent loss of myofiber transducibility observed with adenovirus and HSV-1 is recapitulated in single muscle fibers in vitro, and thus is not solely due to host immune response (Feero et al, 1997). By using localized irradiation of muscle in vivo prior to isolation of myofibers, we observed that adenoviral infectivity of differentiated myofibers decreased significantly versus muscle fibers from non-irradiated muscles at the same stage of development.

It has been demonstrated that the replication-defective herpes simplex virus (HSV-1), which has been extensively used as a gene delivery vector to the central nervous system (Glorioso et al., 1992; 1994), can also be used as a gene delivery vector to skeletal muscle. HSV-1 efficiently infects myoblasts and myotubes in vitro. Furthermore, the intramuscular injection of the viral vector results in infection and transduction of a significant number of newborn mice muscle fibers and some adult mice muscle fibers (Huard et al., 1995b). However, limitations such as differential transducibility with HSV-1 throughout the maturation of muscle fibers, cytotoxicity, and immunological problems associated with HSV-1 (Huard et al., 1996, 1997a,b) have hindered the use of HSV-1 as a gene delivery vector to skeletal muscle.

C. Combination of myoblast transplantation and gene therapy The idea behind this approach involves the establishment of a primary myoblast cell culture from mdx mice or DMD patients. After an adequate transfection or transduction with a dystrophin cDNA, these transduced myoblasts are reinjected into the same host to bypass immunological problems against the injected myoblasts. This approach,

These results suggest that adenoviral transduction in myofibers depends, at least in part, on myoblasts to mediate 209

Musgrave and Huard: Muscle-based tissue engineering for the musculoskeletal system myofiber transduction. The myoblast content of skeletal muscle decreases in vivo as a function of age: thus, the documented dramatic decrease in adenovirus infectivity of skeletal muscle in the post-natal period may be a consequence of reduced myoblast availability and/or fusion.

retrovirus, and HSV-1 in mature myofibers. In support of this hypothesis, a higher level of viral transduction (adenovirus, retrovirus, HSV-1) has been observed in mature regenerating mdx muscle (Acsadi et al., 1994a, van Deutekom et al., 1998a,b). Different myonecrotic agents have been tested for their ability to specifically induce muscle regeneration and allow efficient viral transduction in mature muscle. We have observed that cardiotoxin treatment prior to adenoviral and retroviral transduction improves the efficiency of gene transfer in mature muscles (van Deutekom et al., 1998a,b). In order to determine whether the improved adenoviral transduction levels obtained in regenerating mature muscle were due to myoblast mediation, the presence of immature myofibers, or a combination of both, we irradiated regenerating muscle prior to adenoviral injection to inactivate the myoblasts. The irradiated muscles of the mice treated with cardiotoxin 2 and 3 days prior to adenoviral injection displayed a significantly decreased viral transduction in comparison to the non-irradiated muscles of the same mice (van Deutekom et al., 1998a,b). These low transduction levels suggested that the adenoviral transduction observed in the non-irradiated muscles of these mice was mainly due to myoblast-mediation. In contrast, the irradiated muscles of the mice treated with cardiotoxin 4 and 5 days prior to adenoviral injection did not show reduced transduction efficiencies, suggesting that the high adenoviral transduction levels were most likely due to the presence of immature myofibers.

Recently, genetically modified adenoviral (ADV) vectors have been developed (Wickham et al., 1996) that express heparan sulfate directed targeting peptides at the end of fiber proteins in assembled virions (ADV PK). These viruses no longer bind cells via the native attachment receptor, yet they retain the ability to enter cells via internalization receptors (#v !3/!5 integrin). The use of these new viruses will help to determine the role of the adenovirus’ attachment receptor in the maturation-dependent adenoviral transduction of muscle fibers, since it has been proposed that the gradual loss of viral receptors is involved in the maturation-dependent adenoviral transduction of skeletal myofibers (Acsadi et al, 1994b). Our results, obtained in newborn and adult mouse skeletal muscle, indicate that despite the enhanced attachment characteristics, ADV PK remains hindered by both the protective extracellular matrix and diminished myoblast mediation in mature muscle (van Deutekom et al., 1998c). On the other hand, HSV-1 is capable of infecting both myoblasts and myotubes with a similar efficiency in vitro. In addition, the irradiation of newborn muscle prior to HSV-1 infection does not significantly decrease HSV-1 transduction of myofibers in vitro and in vivo (van Deutekom et al., 1998a,b). Since HSV-1 is capable of transducing myotubes and newborn myofibers without myoblast transduction, we have performed experiments to characterize the cause(s) of the poor HSV-1 transduction in mature myofibers.

2. Permeability of the basal lamina Based on the hypothesis that the basal lamina acts as a physical barrier to viral injection in adult myofibers, we used agents that permeate the basal lamina prior to HSV-1 infection in an effort to achieve efficient transduction of adult myofibers. Different fenestrating agents, such as plasminogen activators and neutral proteases (streptokinase, urokinase), were tested to permeate the basal lamina and allow for HSV-1 to penetrate and transduce mature muscle fibers. This approach was first tested in vitro on mature muscle fibers isolated from adult mice, since in vitro muscle fibers represent a good model system for viral gene delivery to skeletal muscle (Feero et al., 1997). We observed that pretreatment with streptokinase (5 units) and urokinase/plasminogen activator (10 units each) for 60 min. prior to HSV-1 infection in isolated myofibers (2-month-old) enhanced the level of HSV-1 transduction in mature isolated myofibers (van Deutekom et al., 1998a,b).

Preliminary data suggests that the poor level of HSV-1 transduction in mature myofibers is the consequence of the basal lamina maturation causing a physical block to virus accessibility and penetration (Huard et al., 1996, 1997a,b). In order to support this hypothesis, we have shown increased transduction efficiencies in adult myofibers from dy/dy mice (Huard et al, 1996). Dy/dy mice have defective basal lamina due to merosin deficiency. Moreover, isolated myofibers from adult dy/dy muscles (Soleus and EDL) were also found less refractory to HSV-1 transduction in contrast to that observed with age-matched control (dy/+) normal adult myofibers in vitro (Feero et al., 1997).

E. Approaches to circumvent the maturation-dependent viral transduction of muscle fibers

3. Common approaches to improve adenovirus and HSV-1 transduction in mature muscle

1. Artificial induction of muscle regeneration Based on these results, it appears logical that artificial induction of muscle regeneration using agents which release satellite cells and promote myoblast proliferation and fusion may result in a higher level of transduction with adenovirus,

The maturation-dependent viral transduction with adenovirus and HSV-1 may be related at least in part by common mechanisms. Since we have observed that adenovirus displays a higher level of gene transfer in adult 210

Gene Therapy and Molecular Biology Vol 3, page 211 dy/dy mice, it is likely that adenoviral penetration and transduction in mature myofibers is hindered at least partially by the basal lamina which acts as a physical barrier. As mentioned above, approaches to fenestrate the basal lamina may consequently allow for a better adenoviral transduction in mature myofibers. Using new mutant HSV-1 vectors which display a reduction in cytotoxicity to muscle cells, we have observed that an intermediate level of HSV-1 transduction occurs in regenerating muscle (van Deutekom et al., 1998a,b). This observation suggests that approaches which artificially release myoblasts in mature muscle may help achieve efficient transduction of HSV-1 in mature muscle. In fact, artificial induction of muscle regeneration with cardiotoxin improves HSV-1 transduction in mature muscle (van Deutekom et al, 1998a,b).

by its restricted gene insert capacity (<5Kb). This is especially true in the field of DMD in which the dystrophin cDNA is 14 Kb. The identification of a new form of truncated dystrophin which displays a protection to the skeletal muscle fibers may eventually allow for its insertion into the AAV and consequently its delivery into dystrophic muscles. A schematic representation of the maturation-dependent viral transduction of skeletal muscle (adenovirus, retrovirus, and HSV-1) and the aforementioned approaches to improve the viral transduction of mature skeletal muscle are presented in Figure 1.

III. Muscle injury and repair Muscle injuries comprise a large percentage of recreational and competitive athletic injuries. Muscle injuries may result from both direct (contusions, lacerations) and indirect trauma (strains, ischemia and neurological injuries). Upon injury, satellite cells are released and activated in order to differentiate into myotubes and myofibers, thereby promoting muscle healing. However, this reparative process is usually incomplete and accompanied by a fibrous reaction producing scar tissue. This scar tissue limits the muscle’s potential for functional recovery (Hurme et al., 1991, 1992).

4. The myoblast-mediated ex vivo gene transfer approach The ex vivo gene transfer may circumvent the inability of viral vector to transduce mature myofibers. The ability of adenovirus, retrovirus, and HSV-1 to efficiently transduce myoblasts can be used in an ex vivo approach. This approach consists of first transducing myoblasts in vitro, then transplanting them intramuscularly in vivo. We have achieved an efficient level of adenovirus, retrovirus, and HSV-1 transduction in mature muscle fibers using the ex vivo approach (van Deutekom et al., 1998a,b). In fact, a higher level of gene transfer was observed using the ex vivo approach than with the direct gene transfer using the same amount of viral particles (Booth et al., 1997). Although the poor survival of the injected myoblasts limits the efficiency of the myoblast-mediated ex vivo gene transfer of viral vectors in mature muscle, it has recently been found that the poor survival of the injected myoblasts is related at least in part to inflammatory reactions (Guerette et al., 1997; Qu et al., 1998). In an effort to bypass this limitation, myoblasts engineered to express molecules capable of expressing antiinflammatory substances were used. Engineered myoblasts expressing interleukin 1 receptor antagonist protein (IRAP) were capable of improving the survival of the injected myoblast post-implantation (Qu et al., 1998). Furthermore, the use of specific populations of muscle-derived cells improves the cell survival after transplantation and consequently enhances the success of myoblast transplantation (Qu et al., 1998).

Investigations in animals identified possible clinical applications for muscle-based tissue engineering to treat muscle injuries (Garrett et al., 1984, 1990). Animal models of muscle laceration, contusion, and strain currently exist (Jarvinen and Sorvari, 1975; Carlson and Faulkner, 1983; Garrett et al., 1984, 1990; Nikolaou et al., 1987; Taylor et al., 1993; Crisco et al., 1994; Hughes et al., 1995). We have developed reproducible orthopaedic muscle injuries in mice: Laceration is performed by incising 75% of the width and 50% of the thickness of the gastrocnemius muscle (Menetrey et al, 1998a,b). Contusion is created by dropping a 16 gram iron ball from a height of 100 centimeters (cm) onto the gastrocnemius muscle (Kasemkijwattana et al, 1998a,b). Strain is created by elongating the muscle-tendon unit at a rate of 1 cm/min (Kasemkijwattana et al 1998a,c). Under these conditions, muscle myofiber regeneration is found at 7 and 10 days after injury, but begins to decrease at 14 days and continues decreasing until 35 days. Concomitantly, fibrosis is observed beginning at 14 days and gradually increases until 35 days (Kasemkijwattana et al., 1998a,b,c; Menetrey et al., 1998a,b). Fibrosis appears at the time muscle regeneration diminishes and, therefore, appears to hinder the healing response.

5. The use of new viral vectors More recently, recombinant adeno-associated viral vectors (rAAV) have been used as gene delivery vehicles for skeletal muscle cells. Although a high efficiency of gene transfer occurs in mature muscle and a long term transgene expression of up to 18 months has been observed in mouse skeletal muscle (Kessler et al., 1996; Xiao et al., 1996; Reed Clark et al., 1997), the use of this viral vector will be limited

Injured skeletal muscle releases numerous growth factors acting in autocrine and paracrine fashion to modulate muscle healing. These proteins activate satellite cells to proliferate

211

Gene Therapy and Molecular Biology Vol 3, page 212

Figure 1. Schematic representation of retroviral (RSV), herpes simplex viral (HSV), adenoviral (AV), and adenoassociated virus (AAV) transduction of mature skeletal muscle, as well as approaches (the permeating of the extracellular matrix, the induction of degeneration/ regeneration, and the ex vivo strategy) to improve the viral transduction of mature skeletal muscle.

and differentiate into myofibers (Hurme, 1992; Bischoff, 1994; Allamedine et al., 1989; Schultz, 1985, 1989). The delivery of exogenous growth factors, specifically selected to enhance myofiber regeneration, is an intuitive therapeutic approach to muscle injuries. In vitro experiments have identified several growth factors capable of enhancing myogenic proliferation and differentiation (Kasemkijwattana et al., 1998a; Menetrey et al, 1998b). Satellite cell activity in cell culture was assessed at 48 and 96 hours after incubation in prospective growth factors. Basic fibroblast growth factor (b-FGF), insulin-like growth factor-1 (IGF-1), and nerve growth factor (NGF) significantly enhanced myoblast proliferation, whereas b-FGF, acidic fibroblast growth factor (a-FGF), IGF-1, and NGF increased myoblast differentiation into myotubes. Consequently, bFGF, IGF-1, and NGF are the logical candidates for therapeutic applications to enhance muscle healing (Kasemkijwattana et al., 1998a; Menetrey, et al., 1998b).

The technique chosen to deliver prospective growth factors to injured muscle is of paramount importance to optimize therapeutic benefit. Options include direct injection of growth factors, direct gene therapy, ex vivo gene therapy, and myoblast transplantation. Individual direct injections of b-FGF, IGF-1, and NGF into injured muscle (laceration, contusion, and strain) can increase the number of regenerating myofibers in vivo and increase both muscle twitch and tetanic strength 15 days after injury (Kasemkijwattana et al., 1998a,b,c; Menetrey et al., 1998b). However, secondary to rapid clearance and short half-lives, the effect of direct growth factor injections is likely transient and suboptimal. Gene therapy provides a mechanism to achieve persistent protein production and, thereby, theoretically improved muscle healing. Direct gene therapy to deliver genes to skeletal muscle is possible using naked DNA, retrovirus, adenovirus, herpes simplex virus and adeno-associated virus (see Section II). Most of these vectors transduce relatively few adult myofibers. However, adenovirus is capable of tranducing a large number of regenerating muscle fibers, a condition present in injured 212

Gene Therapy and Molecular Biology Vol 3, page 213 muscle. Direct injection of adenovirus containing the betagalactosidase marker gene into lacerated, contused, and strained muscle results in many transduced myofibers at 5 days (Figure 2). Therefore, direct injection of adenovirus carrying growth factor genes (i.e. bFGF, IGF-1, NGF) should result in sustained protein production in injured muscle. Recent data shows that direct injection of adenoassociated virus (AAV) results in a high level of adult myofiber transduction in both injured and non-injured muscle (Pruchnic et al, 1998). AAV may be the preferred vector for direct gene delivery to mature skeletal muscle, although it is capable of carrying genes of only 1-4 Kb. Ex vivo gene therapy and myoblast transplantation are two closely related methods which require in vitro cell isolation and culture. Ex vivo techniques involve muscle biopsy and myogenic cell isolation (Rando and Blau, 1994; Qu et al, 1998). The isolated satellite cells are transduced in vitro with the desired gene carrying vector. The satellite cells are then reinjected into skeletal muscle, fuse to form post-mitotic

myotubes and myofibers, and begin growth factor production. This technique is feasible with adenoviral (Huard et al., 1994c), retroviral (Salvatori et al., 1993), and herpes simplex viral vectors (Booth et al., 1997). Ex vivo delivery of the !-galactosidase marker gene to injured muscle produces many !-galactosidase-positive myofibers (Figure 2). The ex vivo muscle cell-mediated approach provides not only an efficient method of delivering selected genes, but also provides cells capable of participating in the reparative process, similar to myoblast transplantation. However, myoblast transplantation lacks in vitro genetic manipulations. In addition to its application toward inherited muscle diseases, myoblast transplantation is shown to improve myofiber regeneration in muscle experimentally injured with myonecrotic agents (Huard et al., 1994b). Therefore, the closely related techniques of muscle cellmediated ex vivo gene therapy and myoblast transplantation are both applicable to muscle injuries.

Figure 2. Adenovirus mediated direct and ex vivo gene transfer of !-galactosidase in lacerated, contused, and strain-injured muscle. The direct (A, C, E) and ex vivo (B, D, F) gene transfer into contused (A, B), lacerated (C, D), and strain-injured muscle (E, F) lead to successful gene delivery of !-galactosidase marker gene in the injured site at 5 days post-injection. Magnification X10 for A-F.

213

Gene Therapy and Molecular Biology Vol 3, page 214 Kawasaki et al., 1998). Therefore, the in vitro data suggests that myogenic cells are capable of responding to rhBMP-2 and entering an osteogenic lineage. Primary rodent muscle-derived cells are capable of being engineered to produce intramuscular bone in vivo (Bosch et al., 1998). The ex vivo approach is utilized to transduce the primary muscle-derived cells with an adenovirus carrying the BMP-2 cDNA. Intramuscular injection of as little as 300,000 transduced cells produces bone in severe combined immune deficient (SCID) mice (Bosch et al., 1998). The bone produced contains osteoid and bone marrow elements as evidenced by hematoxylin and eosin (H&E) stain and von Kossa stain for mineralization (see Figure 3). Not only do the transduced muscle cells produce BMP-2, but strong evidence suggests that the injected cells also respond to BMP-2 by producing bone (Bosch et al., 1998). In addition to the ex vivo approach, an adenovirus mediated direct gene transfer of BMP-2 produces large amounts of intramuscular bone (Musgrave et al., 1998). Consequently, both the in vitro and in vivo data support the hypothesis that muscle cells may be engineered to become osteogenic cells. The ramifications of myogenic cells’ capabilities to form bone are immense. Muscle-based tissue engineering to produce bone may be applicable to multiple skeletal abnormalities. One such scenario is large bone defects resulting from trauma or oncologic resections. Muscle-derived cells capable of bone formation may be exploited to reconstruct the bone defect and minimize the use of autograft, allograft, and bone distraction. Currently, we are investigating whether a muscle flap can be engineered to produce bone and, thereby, reconstruct an experimental bone defect. Both ex vivo and in vivo gene therapy techniques are being applied in this model. Another approach is to transform muscle, restricted to the confines of a silicone mold, into bone of desired geometry such as a proximal femur or midshaft tibia (Khouri et al., 1991). The muscle-based approach to bone defect reconstructions is especially appealing in light of the often poor vascularity of traumatic and oncologic bone defects. The combination of vascularized muscle and de novo bone formation offers revolutionary possibilities worthy of further investigation.

Muscle-based tissue engineering offers exciting potential therapies for muscle disorders. A large number of recreational and professional athletic injuries involve skeletal muscle (Garrett, 1990). Therapies to improve functional recovery and shorten rehabilitation may both optimize performance and minimize morbidity. Further research is ongoing to refine these muscle-based tissue engineering applications. The results of such investigations may provide revolutionary treatments for these common muscle injuries.

IV. Bone healing Multiple surgical specialties, including orthopaedic, plastic, and maxillofacial, are concerned with bone healing augmentation. Physicians in these disciplines rely on bone augmentation techniques to improve healing of fracture nonunions, oncologic and traumatic bone defect reconstructions, joint and spine fusions, and artificial implant stabilizations. Unfortunately, current techniques of autograft, allograft, and electrical stimulation are often suboptimal. Therefore, tissue engineering approaches toward bone formation have immense implications. Intramuscular bone formation is a poorly understood phenomenon. It can be present in the clinically pathologic states of heterotopic ossification, myositis ossificans, fibrodysplasia ossificans progressiva, and osteosarcoma. Radiation therapy and the anti-inflammatory drug, indomethicin, can suppress myositis ossificans. However, neither the mechanism of formation nor suppression of ectopic bone is clearly understood. The first evidence toward the existence of growth factors capable of stimulating intramuscular bone was gathered 30 years ago (Urist, 1965). Now, a growing family of bone morphogenetic proteins (BMPs), members of the transforming growth factor-! (TGF-!) superfamily, is recognized. Human BMP-2 in recombinant form (rhBMP-2) and BMP-2 cDNA encoding plasmids induce bone formation when injected into skeletal muscle (Wang et al., 1990; Fang et al, 1996). Current applications focus on injecting rhBMP-2 directly into nonunions and bone defects. However, muscle-based tissue engineering has enormous promise in the arena of bone healing and may shed light on the physiologic mechanism of ectopic bone formation. Cells isolated from skeletal muscle are capable of responding to rhBMP-2 both in vitro and in vivo. Primary rodent myogenic cells in cell culture respond in a dose dependent fashion to rhBMP-2 by producing alkaline phosphatase, an osteogenic protein (Bosch et al., 1998). Furthermore, the purer the population of myogenic cells, as evidenced by desmin staining, the greater the alkaline phosphatase production (Bosch et al., 1998). Recombinant human BMP-2 inhibits myogenic differentiation as it stimulates osteoblastic differentiation of the muscle-derived cells (Yamaguchi et al., 1991; Katagiri et al., 1994;

V. Intraarticular disorders Degenerative and traumatic joint disorders are encountered frequently as our population becomes more active and lives longer. These disorders include arthritis of various etiologies, ligament disruptions, meniscal tears, and osteochondral injuries. Currently, the clinician’s tools consist primarily of surgical procedures aimed at biomechanically alterating the joint (anterior cruciate ligament [ACL] reconstructions, total knee replacement, menciscal repair or excision, cartilage debridement, etc.). Tissue engineering applied to these intraarticular disease states theoretically offers a more biologic and less disruptive

214

Gene Therapy and Molecular Biology Vol 3, page 215

Figure 3. Myoblast mediated gene transfer of bone morphogenic protein2 (BMP-2) leads to ectopic bone formation within skeletal muscle. The injection of adenovirally-transduced myoblasts to express BMP-2 in the gastrocnemius muscle of a scid mouse leads to ectopic bone formation, which is evidenced by H&E (A) and von Kossa (B) stains. Magnification: X10 for A, B.

intraarticular structures (Day et al., 1997). Tissues expressing !-galactosidase at 5 days after injection in the rabbit knee include the synovial lining, meniscal surface, and cruciate ligament (Day et al, 1997). In contrast, injection of transduced synovial cells results in !-galactosidase expression only in the synovium (Day et al., 1997). Likewise, injection of transduced immortalized myoblasts results in gene delivery to various intraarticular structures, including the synovial lining and patellar ligament surface. However, the purified immortalized myoblasts fused more readily and resulted in more de novo intraarticular myofibers than the primary myoblasts. This illustrates the importance of obtaining a pure population of myogenic cells, void of fibroblast and adipocyte contamination often seen in primary

reparative process. Both direct (Nita et al., 1996) and ex vivo (Bandara et al., 1993) gene therapy approaches to arthritis models have been reported. The synovial cell-mediated ex vivo approach, while offering advantages of ex vivo gene transfer such as the safety of in vitro genetic manipulation and precise cell selection, is hindered by a decline of gene expression after 5-6 weeks (Bandara et al., 1993). Due to its ability to form post-mitotic myotubes and myofibers, the skeletal muscle satellite cell offers theoretical advantages of longer term and more abundant protein production. Muscle cell-mediated ex vivo gene delivery to numerous intraarticular structures is possible. Intraarticular injection of primary myoblasts, transduced by adenovirus carrying the !galactosidase marker gene, results in gene delivery to many 215

Musgrave and Huard: Muscle-based tissue engineering for the musculoskeletal system myoblasts. Muscle cell-mediated ex vivo approaches are predicated on myoblast fusion to form myofibers, the multinuclear protein-producing factories. Intraarticular injection of transduced immortalized myoblasts into a severe combined immune deficient (SCID) mouse results in myotubes formation and transgene expression in multiple structures at 35 days. Therefore, intraarticular gene expression (for at least 35 days) resulting from muscle cellmediated tissue engineering is feasible in animal models. Based on this data, a muscle cell-mediated gene transfer approach may deliver genes to improve the healing of several intra-articular structures specifically to the ACL and meniscus.

Efficient and sustained delivery of desired growth factors may be best accomplished by gene delivery. Muscle cellmediated ex vivo gene delivery offers the possibility of sustained, high level gene expression. Investigations utilizing the muscle cell-mediated ex vivo approach to deliver marker genes and growth factors directly to the rabbit meniscus are currently underway. Such studies may lead to novel therapies for meniscal injuries, preventing significant morbidity from these chronically disabling injuries.

The ACL is the second most frequently injured knee ligament. Unfortunately, the ACL has a low healing capacity, possibly secondary to its encompassing synovial sheath or the surrounding synovial fluid. Because complete tears of the ACL are incapable of spontaneous healing, current treatment options are limited to surgical reconstruction using autograft or allograft. The replacement graft, often either patella ligament or hamstrings tendon in origin, undergoes ligamentization with eventual collagen remodeling (Arnosczky et al., 1982). Therefore, recent research is directed at augmentation of this ligamentization process using growth factors to affect fibroblast behavior. In vivo data suggests that platelet-derived growth factor (PDGF), transforming growth factor-! (TGF-!), and epidermal growth factor (EGF) promote ligament healing (Conti, 1993). Transient, low levels of these growth factors resulting from their direct injection into the injured ligament are unlikely to produce a significant response. Therefore, an efficient delivery mechanism is essential to the development of a clinically applicable therapy. Muscle cell-mediated ex vivo gene therapy offers the potential to achieve persistent local gene expression and subsequent growth factor delivery to the ACL. Investigations into the effect of muscle cellmediated ex vivo gene therapy to enhance the healing of torn ACLs, reconstructed ACLs, and the bone ligament interface are currently ongoing.

Muscle-based tissue engineering is a burgeoning new discipline with unknown possibilities. Data gathered thus far proposes to challenge traditional scientific beliefs at many levels, from basic muscle cell biology to clinical medicine. In addition to the characterization of possible skeletal muscle-derived mesenchymal stem cells, investigators must aggressively pursue potential clinical applications for muscle-based tissue engineering (see schematic representation in Figure 4). The development of musclebased tissue engineering approaches to inherited muscle diseases, acquired muscle injuries, bone healing, and intraarticular disorders is underway. Furthermore, investigations have been initiated into the utility of musclebased tissue engineering to heal cartilage defects, spinal injuries, and flexor tendon lacerations. An explosion of research, from basic science to clinical medicine, is mandated to fully elucidate the potential of muscle-based tissue engineering for musculoskeletal disorders (see Figure 4).

VI. Future directions

Acknowledgements The authors wish to thank Marcelle Pellerin and Ryan Pruchnic for their technical assistance, and Megan Mowry and Dana Och for assistance with the manuscript. This work was supported by grants to Dr. Johnny Huard from the National Institute of Health (NIH, #1P60 AR 44811-01) and the Pittsburgh Tissue Engineering Initiative (PTEI).

The knee meniscus plays a critical role in maintaining normal knee biomechanics. Primary functions of the meniscus include load transmission, shock absorption, joint lubrication, and tibiofemoral stabilization in the ACL deficient knee. The historical treatment of menisectomy for meniscal tears has been replaced by meniscal repair when tears involve the meniscus’ peripheral, vascular third. Growth factors, including platelet-derived growth factor (PDGF), are capable of enhancing meniscal healing (Spindler et al., 1995). In vitro data currently under review details numerous growth factors’ effects on fibroblast proliferation and collagen production (in preparation). Regardless of which growth factor is proven optimal for meniscal healing, the cardinal issue of protein delivery must be addressed. Direct intrameniscal growth factor injections are unlikely to produce sustained levels without the need for multiple injections, a scenario not clinically appropriate.

References Acsadi, G., Lochmueller, H., Jani, A., Huard, J., Massie, B., Prescott, S., Simoneau, M., Petrof, B., and Karpati, G. (1995) Dystrophin expression in muscles of mdx mice after adenovirus-mediated in vivo gene transfer. Hum. Gene Ther 7, 129-140. Acsadi, G.., Jani, A., Massie, B., Simoneau, M., Holland, P., Blaschuk, K., and Karpati, G. (1994a) A differential efficiency of adenovirus-mediated in vivo gene transfer into skeletal muscle cells of different maturity. Hum. Mol. Genet. 3, 579584. Allamedine, H.S., and Fardeau, M. (1990) Muscle reconstruction

216

Gene Therapy and Molecular Biology Vol 3, page 217

Figure 4. Schematic representation of the different applications of muscle based tissue engineering to various areas of the musculoskeletal system, including: muscle injuries and repair, bone defect, intra-articular structures, spinal injuries, and tendon repair.

by satellite cell graft. J. Neurol. Sci. 99, 126.

Bandara, G., Mueller, GM., Galea-Lauri, J., et al. (1993) Intraarticular expression of biologically active interleukin-1 receptor antagonist protein by ex vivo gene transfer. Proc Natl Acad Sci USA 90, 10764-10768.

Allamedine, H.S., Dehaupas, M., and Fardeau, M. (1989) Regeneration of skeletal muscle fiber from autologous satellite cells multiplied in-vitro. Muscle Nerve 12, 544-555.

Bischoff, R. (1994) The satellite cell and muscle regeneration. In: Engel AG, Franzini-Armstrong C. eds. Myology. 2nd ed. New York: McGraw-Hill, Inc. pp. 97-118.

Arahata, K., Ishiura, S., Ishiguro, T., Tsukahara, T., Suhara, Y., Eguchi, C., Ishihara, T., Nonaka, I., Ozawa, E., and Sugita, H. (1988) Immunostaining of skeletal and cardiac muscle surface membrane with antibody against Duchenne Muscular Dystrophy peptide. Nature 333, 861-863.

Blau, HM., and Webster, C. (1981) Isolation and characterization of human muscle cells. Proc Natl Acad Sci USA 78, 56235627.

Arnoczky, SP., Rarvin, GB., and Marshall, JL. (1982) Anterior cruciate ligament replacement using patellar tendon. J Bone Joint Surg 64A, 217-224.

Bonilla, E.C.E., Samitt, A.F., Miranda, A.P., Hays, G., Salviati, S., Dimauro, S., Kunkel, L.M., Hoffman, E.P., and Rowland L.P. Duchenne (1988) Muscular Dystrophy: deficiency of dystrophin at the muscle cell surface. Cell 54, 447-452.

Ascadi, G., Dickson, G., Love, D., Jani, A., Gurusinghe, A., Walsh, FS., Wolff, JA., and Davies, KE. (1991) Human dystrophin expression in mdx mice after intramuscular injection of DNA constructs. Nature 352, 815-818.

Booth, DK., Floyd, SS., Day, CS., Glorioso, JC., Kovesdi, I., and Huard, J. (1997) Myoblast mediated ex vivo gene transfer to mature muscle. J Tissue Eng 3, 125-133.

Ascadi, G., Jani, A., Huard, J., Blaschuk, K., Massie, B., Holland, P., Lochmuller, H., and Karpati, G. (1994b) Cultured human myoblasts and myotubes show markedly different transducibility by replication-defective adenovirus recombinants. Gene Ther 1, 338-340.

Bosch P, Musgrave DS, Shuler F, Ghivizzani SC, Evans C, Robbins PD, and Huard J: (1998) Bone formation by muscle derived stem cells. Submitted for publication in Nat. Biotechnol.

217

Musgrave and Huard: Muscle-based tissue engineering for the musculoskeletal system Caplan, AI: (1991) Mesenchymal stem cells. J Orthop Res 9, 641650.

Myology 7, 241-250. Floyd, SS., Clemens, PR., Ontell, MR., Kochanek, S., Day, CS., Hauschka, SD., Balkir, L., Morgan, JE., Moreland, MS., Feero, WG., Epperly, M., and Huard, J. (1998) Ex vivo gene transfer using adenovirus-mediated full length dystrophin delivery to dystrophic muscles. Gene Ther. 5, 19-30.

Carlson, BM., and Faulner, JA. (1983) The regeneration of skeletal muscle fibers following injury: a review. Med Sci Sports Exer 15, 187-196. Conti, NA., and Dahners, LE: (1993) The effect of exogenous growth factors on the healing of ligaments. Trans Orthop Res Soc 18, 60.

Garrett, W.E., Saeber, AV., Boswick, J., Urbaniak, JR., Goldner, L. (1984) Recoverey of skeletal muscle after laceration and repair. J. Hand Surg. (Am) 9A, 683-692.

Crisco, JJ., Jolk, P., Heinen, GT., Connell, MD., and Panjabi, MM. (1994) A muscle contusion injury model, biomechanics, physiology, and hisology. Am J Sports Med 22, 702-710.

Garrett, WE. (1990) Muscle strain injuries: clinical and basic aspects. Med Sci. Sports Exer 22, 436-443.

Dai, Y., Roman, M., Naviaux, RK., and Verma, IM (1992) Gene therapy via primary myoblasts: long term expression of factor IX protein following transplantation in vivo. Proc Natl Acad Sci USA 89, 10892-10895.

Glorioso, J.C., Goins, W.F., Fink, D.J., and DeLuca, N.A. (1994) Herpes Simplex virus vectors and gene transfer to brain. In Recombinant vectors in vaccine development. Brown F, editors. Dev Biol Stand. Basel, Karger 82, 79-87.

Danko, I., Fritz, J.D., Latendresse, J.S., Herweijer, H., Schultz, E., and Wolff, J.A. (1993) Dystrophin expression improves myofiber survival in mdx muscle following intramuscular plasmid DNA injection. Hum. Mol. Genet. 2, 2055-2061.

Glorioso, J.C., Sternberg, LR., Goins, W.F., and Fink, D.J. (1992) Development of Herpes Simplex virus as a gene transfer vector for the central nervous system. In Gene transfer and therapy in the nervous system. Gage, FH., Christen, Y, edi. SpringerVerlag Berlin Heidelberg, 133-145.

Davis, H., Demeneix, BA., Quantin, B., Coulombe, J., and Whalen, RG. (1993) Plasmid DNA is superior to viral vectors for direct gene transfer into adult mouse skeletal muscle. Hum. Gene Ther. 4, 733-740.

Guerette, B., Asselin, I., Skuk, D., Entman, M., Tremblay, J. (1997) Control of inflammatory damage by anti-LFA-1: Increase success of myoblast transplantation. Cell Transpl 6, 101-107.

Day, CS., Kasemkijwattana, C., Moreland, MS., Floyd, SS., and Huard, J. (1997) Muscle cells as a gene delivery vehicle to the joint. J Orthop Res 15, 894-903.

Gussoni, E., Pavlath, P.K., Lanctot, A.M., Sharma, K., Miller, R.G., Steinman, L., and Blau, H.M. (1992) Normal dystrophin transcripts detected in DMD patients after myoblast transplantation. Nature 356, 435-438.

Dhawan, J., et al.: (1991) Systemic delivery of human growth hormone by injection of genetically engineered myoblasts. Science 254, 1509-1512.

Hoffman, E.P., Brown, J., and Kunkel, L.M. (1987) Dystrophin: the protein product of the Duchenne Muscular Dystrophy locus. Cell 51, 919-928.

Dunckley MG, Wells DJ, Walsh FS, and Dickson G. (1993) Direct retroviral-mediated transfer of a dystrophin minigene into mdx mouse in muscle in vivo. Hum Mol Genet 2, 717-723.

Huard, J., Bouchard, J.P., Roy, R., Malouin, F., Dansereau, G., Labrecque, C., Albert, N., Richards, C.L, Lemieux, B., and Tremblay, J.P. (1992a) Human myoblast transplantation: preliminary results of 4 cases. Muscle Nerve 15, 550-560.

Dunckley, M.G., Davies, K.E., Walsh, F.S., Morris, G.E., and Dickson, G. (1992) Retroviral-mediated transfer of a dystrophin minigene into mdx mouse myoblasts in vitro. FEBS Lett 296, 128-134.

Huard, J., Roy, R., Bouchard, J.P., Malouin, F., Richards, C.L., and Tremblay, J.P. (1992b) Human Myoblast transplantation between immunohistocompatible donors and recipients produces immune reactions. Transpl. Proc .24, 3049-3051.

Engelhardt, J.F., Litzky, L., and Wilson, J.M. (1994a) Prolonged transgene expression in cotton rat lung with recombinant adenovirus defective in E2A. Hum. Gene Ther 5, 1217-1229.

Huard, J., Guerette, B., Verreault, S., Tremblay, G., Roy, R., Lille, S., and Tremblay, J.P. (1994a) Human myoblast transplantation in immunodeficient and immunosuppressed mice: Evidence of rejection. Muscle Nerve 17, 224-234.

Engelhardt, J.F., Ye, X., Doranz, B., and Wilson, J.M. (1994b) Ablation of E2A in recombinant adenoviruses improves transgene persistence and decreases inflammatory responses in mouse liver. Proc. Natl. Acad. Sci. USA 91, 6196-6200.

Huard, J., Verreault, S., Roy, R., Tremblay, M.,and Tremblay, J.P. (1994b) High efficiency of muscle regeneration following human myoblast clone transplantation in SCID mice. J. Clin. Invest. 93, 586-599.

Fang, J., Zhu, Y-Y., Smiley, E., Bonadio, J,. Rouleau, JP., Goldstein, SA., McCauley, LK., Davidson, BL., and Roessler, BJ. (1996) Stimulation of new bone formation by direct transfer of osteogenic plasmid genes. Proc Natl Acad Sci USA 93, 5753-5758.

Huard, J., Acsadi, G., Jani, A., Massie, B., and Karpati, G. (1994c) Gene transfer into skeletal muscles by isogenic myoblasts. Hum. Gene Ther 5, 949-958.

Feero, WG., Rosenblatt, JD., Huard, J., Watkins, SC., Epperly, M., Clemens, PR., Kochanek, S., Glorioso, JC., Partridge, TA., and Hoffman EP. (1997) Viral gene delivery to skeletal muscle: Insights on maturation-dependent loss of infectivity for adenovirus and herpes simplex type 1 viral vectors. Hum. Gene Ther. 8, 371-380.

Huard, J., Lochmuller, H., Acsadi, G., Jani, A, Holland, P., Guerin, C., Massie, B., and Karpati, G. (1995a) Differential short-term transduction efficiency of adult versus newborn mouse tissues by adenoviral recombinants. Exp Mol. Pathol. 62, 131-143. Huard, J., Goins, B., and Glorioso, J.C. (1995b) Herpes Simplex virus type 1 vector mediated gene transfer to muscle. Gene Ther 2, 1-9.

Floyd, SS., Booth, DK., van Deutekom, JCT., Day, CS., and Huard, J. (1997) Autologous myoblast transfer: A combination of myoblast transplantation and gene therapy. Basic Appl

Huard, J., Feero, WG., Watkins, SC., Hoffman, EP., Rosenblatt,

218

Gene Therapy and Molecular Biology Vol 3, page 219 DJ., and Glorioso, JC. (1996) The basal lamina is a physical barrier to HSV mediated gene delivery to mature muscle fibers. J. Virol 70 #11, 8117-8123.

systemic delivery of a therapeutic protein. Proc. Natl. Acad. Sci. USA 93, 14082-14087. Khouri, RK., Koudsi, B., and Reddi, H. (1991) Tissue transformation into bone in vivo: a potential practical application. JAMA 266, 1953.

Huard, J., Akkaraju, G., Watkins, SC., Cavalcoli, MP., and Glorioso, JC. (1997a) LacZ gene transfer to skeletal muscle using a replication defective Herpes Simplex virus type 1 mutant vector. Hum Gene Ther. 8, 439-452.

Kinoshita, I.,Vilquin, J.T., Guerette, B., Asselin, I., Roy, R., and Tremblay, J.P. (1994) Very efficient myoblast allotransplantation in mice under FK506 immunosuppression. Muscle Nerve 17, 1407-1415.

Huard, J., Krisky, D., Oligino, T., Marconi, P., Day, CS., Watkins, SC.,Glorioso, JC. (1997b) Gene transfer to muscle using herpes simplex virus-based vectors. Neuromusc Disord ----299- 313.

Lau, HT., Yu, M., Fontana, A., Stockert, CJ. (1996) Prevention of islet allograft rejection with engineered myoblasts expressing fasL in mice. Science 273, 109-111.

Hughes, C., Hasselman, CT., Best ,TM.,.Martinez, S., and Garrett, WE. (1995) Incomplete, intrasubstance strain injuries of the rectus femoris muscle. Am J Sports Med 23, 500-506.

Law, P.K., Goodwin, T.G., and Wang, M.G. (1988) Normal myoblast injections provide genetic treatment for murine dystrophy. Muscle Nerve 11, 525-533.

Hurme, T., Kalima, H., Lehto, H., and Jarvinen, M. (1991) Healing of skeletal muscle injury: an ultrastructural and immunohistochemical study. Med Sci Sports Exer 23, 801810.

Lynch, CM., Clowes, MM,. Osborne, WRA., Clowes, AW., and Miller, AD: (1992) Long-term expression of human adenosine deaminase in vascular smooth muscle cells of rats: a model for gene therapy. Proc Natl Acad Sci USA 89, 1138-1142.

Hurme, T., Kalimo, H. (1992) Activation of myoblast precursor cells after muscle injury. Med Sci Sports Exer 24, 197-205. Jarvinen, M, and Sorvari, T. (1975) Healing of a crush injury in rat striated muscle. Acta Path. Microbiol Scand Sect. A, 83, 259265.

Menetrey, J., Kasemkijwattana, C., Fu, FH., Moreland, MS., and Huard J. (1998a) Suturing versus immobilization of a muscle laceration: A morphological and functional study. Am J. Sports Med In press.

Jiao, S., Williams, P., Safda, N., Schultz, E., and Wolff, JA: (1993) Cotransplantation of plasmid-transfected myoblasts and myotubes into rat brains enables high levels of gene expression long-term. Cell Transpl 2, 185-192.

Menetrey, J., Kasemkijwattana, C., Day, CS., Bosch, P., Fu, FH., Moreland, MS., and Huard, J. (1998b) The potential of growth factors to improve muscle regeneration following injury. In revision J Bone Joint Surg (Am).

Karpati, G., and Worton, R.G. (1992) Myoblast transfer in DMD: problems and interpretation of efficiency. Muscle Nerve 15, 1209.

Menke, A., and Jokush, H. (1991) Decreased osmotic stability of dystrophin less muscle cells from the mdx mice. Nature 349, 69-71.

Karpati, G., Pouliot, Y., Zubrzycka-Gaarn, E.E., Carpenter, S., Ray, P.N., Worton, R.G., and Holland, P. (1989) Dystrophin is expressed in mdx skeletal muscle fibers after normal myoblast implantation. Am. J. Pathol 135, 27-32.

Morgan, J.E., Hoffman, E.P., and Partridge, T.A. (1990) Normal myogenic cells from newborn mice restore normal histology to degenerating muscle of the mdx mouse. J. Cell Biol. 111, 2437-2449.

Kasemkijwattana, C., Menetrey, J, Day, CS., Bosch, P., Buranapanitkit, B., Moreland, MS., Fu, FH., Watkins, SC., and Huard, J. (1998a) Biologic intervention in muscle healing and regeneration. Sports Med Arthro Rev 6, 95-102.

Morgan, J.E., Pagel, C.N., Sherrat,.T., and Partridge, T.A. (1993) Long-term persistence and migration of myogenic cells injected into pre-irradiated muscles of mdx mice. J. Neurol. Sci. 115, 191-200.

Kasemkijwattana, C., Menetrey, J., Somogyi, G., Moreland, MS., Fu, FH., Buranapanitkit, B., Watkins, SC., and Huard J. (1998b) Development of approaches to improve the healing following muscle contusion. Cell Transpl, In press.

Morgan, J.E., Watt, D.J., Slopper, J.C., Partridge, T.A. (1988) Partial correction of an inherited defect of skeletal muscle by graft of normal muscle precursor cells. J. Neurol. Sci. 86, 137147.

Kasemkijwattana, C., Menetrey, J, Bosch, P., Somogyi ,G., Moreland, MS, Fu, FH, Buranapanitkit, B., Watkins, SC, and Huard, J. (1998c) The use of growth factors to improve muscle healing following strain injury. In submission Clin. Orth. & Rel. Res.

Musgrave, DS, Bosch, P., Ghivazzani, SC., Robbins, PD., Evans, CH., and Huard, J. (1998) Adenovirus-mediated direct gene therapy with BMP-2 produces bone. In submission Bone. Nikolaou, PK., MacDonald, BL., Glisson, RR., Seaber, AV., and Garrett, WE. (1987) Biomechanical and histological evaluation of muscle after controlled strain injury. Am J Sports Med 15, 9-14.

Katagiri, T., Yamaguchi, A., Komaki, M., et al. (1994) Bone morphogenetic protein-2 converts the differentiation pathway of C2C12 myoblasts into the osteoblast lineage. J Cell Biol 127, 1755-1766.

Nita, I., Ghivizzani, SC., Galea-Lauri, J., Bandara, G., Georgescu, HI., Robbins, PD., and Evans, CH. (1996) Direct gene delivery to synovium: an evaluation of potential vectors in vitro and in vivo. Arthritis Rheum 39, 820-828.

Kawasaki, K., Aihara, M,. Honmo, J., et al.: (1998) Effects of recombinant human bone morphogenetic protein-2 on differentaion of cells isolated from human bone, muscle, and skin. Bone 23, 223-231.

Partridge, T.A. Myoblast transfer (1991) A possible therapy for inherited myopathies. Muscle Nerve 14, 197-212.

Kessler, PD., Podsakoff, GM., Chen, X., McQuiston, SA., Colosi, PC., Matelis, LA., Kurtzman, GJ., and Byrne, BJ. (1996) Gene delivery to skeletal muscle results in sustained expression and

Partridge, T.A.,.Morgan, J.E., Coulton, G.R., Hoffman, E.P., and Kunkel, L.M. (1989) Conversion of mdx myofibers from

219

Musgrave and Huard: Muscle-based tissue engineering for the musculoskeletal system dystrophin negative to positive by injection of normal myoblasts. Nature 337,176-179.

Collin, H., Rouche, A., Gilgenkrantz, S., Abbdi, N., and Tremblay, M. (1993a) Myoblast transplantation between monozygotic twin girl carriers of Duchenne Muscular Dystrophy. Neurom Disord 3, 583-592.

Pruchnic R., Cao BH., Qu Z., Xiao X., Li J., Samulski RJ., Epperly M., and Huard J. (1998) The use of adeno-associated virus to circumvent the maturation dependent viral transduction of muscle fiber. In submission Hum. Gene Ther. In press

Tremblay, J.P., Malouin, F., Roy, R., Huard, J., Bouchard, J.P., Satoh, A., and Richards, C.L. (1993b) Results of a blind clinical study of myoblast transplantations without immunosuppressive treatment in young boys with Duchenne Muscular Dystrophy. Cell. Transpl 2, 99-112.

Qu, Z., Balkir, L., van Deutekom, JCT., Robbins, PD., Pruchnic, R., and Huard, J. (1998) Development of approaches to improve cell survival in myoblast transfer therapy. J Cell Biol 142, 12257-12267.

Urist M. Bone: formation by autoinduction. Science 1965 150, 893-899.

Quantin, B., Perricaudet, L.D., Tajbakhsh, S., and Mandell, J.L. (1992) Adenovirus as an expression vector in muscle cells in vivo. Proc. Natl. Acad. Sci. USA 89, 2581-2584.

van Deutekom, JCT., Hoffman, EP., and Huard, J. 1998a. Muscle Maturation: implications for gene therapy. Mol. Med. Today 4, 214-220.

Ragot, T., Vincent, M., Chafey, P., Vigne, E., Gilgenkrantz, H., Couton, B., Cartaud, J., Briand, P., Kaplan, J.C., Perricaudet, M., and Kahn, A. (1993) Efficient adenovirus mediated gene transfer of a human mini-dystrophin gene to skeletal muscle of mdx mice. Nature 361, 647-650.

van Deutekom, JCT., Floyd, SS., Booth, DK., Oligino, T., Krisky, D., Marconi, P., Glorioso, J., and Huard, J. (1998b) Implications of maturation for viral gene delivery to skeletal muscle. Neurom Disord 8, 135-148.

Rando, TA., and Blau, HM: (1994) Primary mouse myoblast purification, characterization, and transplantation for cellmediated gene therapy. J Cell Biol 125, 1275-1287.

van Deutekom, JCT., Pruchnic, R.., Wickham, TJ., Kovesdi, I., and Huard, J. (1998c) Targeting of an adenoviral vector to heparancontaining receptors does not bypass the maturation-dependent transducibility od mouse skeletal muscle. In submission in Gene Ther.

Reed Clark, K., Sferra, TJ., Johnson, PR.. (1997) Recombinant adeno-associated viral vectors mediated long-term transgene expression in muscle. Hum. Gene Ther. 8, 659-669.

Vilquin, J.T., Wagner, E., Kinoshita, I., Roy, R., and Tremblay, J.T.. (1995a) Successful histocompatible myoblast transplantation in dystrophin-deficient mdx dystrophin. J. Cell Biol. 131, 975-988.

Salvatori, G., Ferrari, G., Messogiorno, A., Servidel, S., Colette, M,. Tonalli, P., Giarassi, R., Cosso, G., and Mavillo, F. (1993) Retroviral vector-mediated gene transfer into human primary myogenic cells leads to expression in muscle fibers in vivo. Hum Gene Ther 4, 713-723.

Vilquin, JT., Guerette, B., Kinoshita, I., Roy, B., Goulet, M., Gravel, C.., Roy, R., and Tremblay, JP. (1995b) FK506 immunosuppression to control the immune reactions triggered by first-generation adenovirus-mediated gene transfer. Hum Gene Ther 6, 1391-1401.

Schultz E. (1989) Satellite cells behavior during skeletal muscle growth and regeneration. Med Sci Sports Exer 21, 181-186. Schultz, E., Jaryszak, DL., and Valiere, CR. (1985) Response of satellite cells to focal skeletal muscle injury. Muscle Nerve 8, 217-222.

Vincent, M., Ragot, T., Gilgenkrantz, H., Couton, D., Chafey, P., Gregoire, A., Briand, P., Kaplan, J.C., Kahn, A., and Perricaudet, M. (1993) Long-term correction of mouse dystrophic degeneration by adenovirus-mediated transfer of a mini-dystrophin gene. Nat Genet. 5, 130-134.

Simonsen, GD., Groskreutz, DJ., Gorman, CM., and MacDonald, MJ: (1996) Synthesis and processing of genetically modified human pro-insulin by rat myoblast primary cultures. Hum Gene Ther 7, 71-78.

Wang, EA., Rosen, V., D’Assandro, JS., et al. (1990) Recombinant human bone morphogenetic protein induces bone formation. Proc Natl Acad Sci USA 87, 2220-2224.

Smith, T.A.G., Mehaffey, M.G., Kayda, D.B., Saunders, J.M.,.Yei, S., Trapnell, B.C., McClell, A., and Kaleko, M. (1993) Adenovirus mediated expression of therapeutic plasma levels of human factor 1X in mice. Nat Genet 5, 397-402.

Warejcka, DJ., Harvey, R., Taylor, BJ., Young, HE., Lucas, PA. (1996) A population of cells isolated from rat heart capable of differentiating into several mesodermal phenotypes. J Surg Res 62, 232-242.

Spindler, KP., Mayes, CE., Miller, RR., Imro, AK., and Davidson, JM. (1995) Regional mitogenic response to the meniscus to platelet-derived growth factor (PDGF-AB). J Orthop Res 13(2), 201-207.

Watkins, S.C., Hoffman, E.P., Slayter, H.S., and Kunkel L.M. (1988) Immunoelectron microscopic localization of dystrophin in myofibers. Nature 333, 863-866.

Sugita, H., Arahata, K., Ishiguro, T.,Sohara, Y.,Tsukahara, T., Ishiura, J., Eguchi, C., Nonaka, I., and Ozawa, E. (1988) Negative Immunostaining of Duchenne Muscular Dystrophy(DMD) and mdx muscle surface membrane with antibody against synthetic peptide fragment predicted from DMD cDNA. Proc. Japan. Acad. 64, 37-39.

Watt, D.J., Lambert, K., Morgan, J.E., Partridge, T.A., and Sloper, J.C. ( 1982) Incorporation of donor muscle precursor cells into an area of muscle regeneration in the host mouse. J. Neurol. Sci. 57, 319-331. Watt, D.J., Morgan, J.E., and Partridge, T.A.. (1984) Use of mononuclear precursor cells to insert allogenic genes into growing mouse muscles. Muscle Nerve 7, 741-750.

Taylor, DC., Dalton, JD., Seaber, AV., and Garrett, WE. (1993) Experimental muscle strain injury, early functional and structural deficits and the increased risk for reinjury. Am J Sports Med 21, 190-194.

Wickham, TJ., Roelvink, PW., Brough, DE., and Kovesdi, I.. (1996) Adenovirus targeted to heparin-containing receptors increases its gene delivery efficiency to multiple cell types. Nat

Tremblay, J.P., Bouchard, J.P., Malouin, F., Theau, D., Cottrell, F.,

220

Gene Therapy and Molecular Biology Vol 3, page 221 Biotechnol 14, 1570-1573. Xiao X, Li J, Samulski RJ. (1996) Efficient long term gene transfer into muscle tissue of immunocompetent mice by adenoassociated virus vector. J Virol 70, 8098-8108. Yamaguchi, A., Katagiri, T., Ideda, T, Wozney, JM., Rosen, V., Wang, EA., Kahn, AJ., Suda, T., and Yoshiki, S. (1991) Recombinant human bone morphogenetic protein-2 stimulates osteoblastic maturation and inhibits myogenic differentiation in vitro. J Cell Biol 113, 681-687. Yang, Y., Nunes, F.A., Berencsi, K., Furth, E.E., Gonczol, E., and Wilson, J.M. (1994a) Cellular immunity to viral antigens limits E1-deleted adenoviruses for gene therapy. Proc. Natl. Acad. Sci. USA 91, 4407-4411. Yang, Y., Nunes, F.A., Berencsi, K., Gonczol, E., Engelhardt, J.F., and Wilson, J.M. (1994b) Inactivation of E2A in recombinant adenoviruses improves the prospect for gene therapy in cystic fibrosis. Nat Genet 7, 362-369. Young, HE., Mancini, ML., Wright, RP., et al. (1995) Mesenchymal stem cells reside within the connective tissues of many organs. Dev Dyn 202, 137-144. Zubryzcka-Gaarn, E.E., Bulman, D.E., Karpati, G., Burghes, A.H.M., Belfall, B., Klamut, H.J., Talbot, J., Hodges, R.S., Ray, P.N., and Worton, R.G. (1988) The Duchenne Muscular Dystrophy gene is localized in the sarcolemma of human skeletal muscle. Nature 333, 466-469.

221

Gene Therapy and Molecular Biology Vol 3, page 223 Gene Ther Mol Biol Vol 3, 223-232. August 1999.

Helper-dependent adenoviral vectors as gene delivery vehicles Review Article

Manal A. Morsy*, Diane M. Harvey, and C. Thomas Caskey Department of Human Genetics, Merck & Co., Inc., Sumneytown Pike, West Point, PA 19486 __________________________________________________________________________________________________ * Correspondence: Manal A. Morsy, MD., Ph.D., Associate Director, Merck Research Laboratories, Department of Human Genetics, Gene Therapy, WP26A-3000, Sumneytown Pike, West Point, PA 19486. Phone: 215-652-9559; Fax: 215-652-2075; E-mail: morsy@merck.com Key words: adenoviral vectors, leptin, tetracycline-inducible, ob/ob mice, antiprogestin, GAL4 DNA-binding domain, gene switch, rapamycin-regulated gene expression Received: 9 December 1998 and in revised form 10 January 1999; accepted: 10 January 1999

Summary Adenoviral (Ad)-mediated in vivo gene transfer and expression is limited in part by cellular immune responses to viralencoded proteins. In an attempt to diminish these responses, we have previously developed and described helperdependent (HD) Ad vectors in which the viral protein coding sequences are completely deleted. These vectors provided efficient delivery, and greater safety which represents a significant advance over existing Ad vectors. In addition, the inherent enhanced insert capacity (up to ~ 37kb) allows for the insertion of large or multiple genes, including expression regulatory sequences. Several drug-regulated gene expression systems are now available for controlling target gene transcription through the use of small-molecule inducing compounds. While early experiments have demonstrated the utility of inducible systems in cell culture and transgenic mice, continued evaluation of such systems in viral gene therapy vectors should lead to discoveries and improvements which will make them amenable for use in a therapeutic context. The generation of a gene therapy approach that combines both safe and efficient vector delivery of one or multiple genes of interest and a small molecule-controlled gene expression system will provide a powerful tool for therapeutic intervention. complexes (cationic liposomes), DNA-polylysine conjugates or delivery of naked DNA (Morsy et al., 1993a). The adenoviral vectors in specific have seen a great deal of modifications over the years. These modifications ranged from rendering them replication deficient (E1 deleted vectors or 1st generation) to completely deleting all viral protein coding regions (helper-dependent). We will review some data comparing the efficacy of helper-dependent vectors to that of 1st generation in terms of safety, longevity of expression and efficiency of gene delivery in vivo.

I. Introduction Gene therapy is a rapidly evolving technology for therapeutic intervention which involves delivery and expression of nucleic acid in a target cell to complement a genetic defect or deliver a new protein. The diversity in the nature of the expressed product is wide and ranges from expressing cellular enzymes, cellular or circulating proteins, secreted hormones, cytokines or growth factors, to immunogens, ribozymes, or antisense oligonucleotides. The main components of a gene therapy agent are the vector or delivery vehicle and the expression cassette, which is composed of the gene(s) of interest and the promoter elements controlling expression. Recently, intense research has evolved and is ongoing to identify the most suitable vector(s) for gene delivery. Viral and non-viral vectors are continuously being modified, examined and compared for safety, persistence and efficacy in delivery and mediation of gene expression. Among the most extensively studied viral vectors are the retroviral, adenoviral and adeno-associated, and among the non-viral vectors are the DNA-lipid

The final section of this review will briefly summarize recent reports evaluating the potential for ligand-inducible control of gene expression in the context of gene therapy vectors. Use of drug-regulated systems may introduce several levels of control to current gene therapy vectors (Harvey and Caskey, 1998), as therapuetic gene expression will remain transcriptionally silent until activated by the addition of an inducing compound. Ideally, this can provide for temporal control in cases where constitutive gene expression may not be required or even preferred. In

223

Morsy et al: Helper-dependent adenoviral vectors as gene delivery vehicles addition, control of expression levels within their therapeutic range may be afforded by appropriate dosing of the inducer. In cases where expression of a drug-regulated tarnscriptional transactivator is driven by a tissue-specific promoter, some degree of spatial control over gene expression may also be achieved.

II. Adenoviral vectors Over the past few years adenoviruses (Ad) have taken a forefront position as gene delivery vehicles as a result of their numerous advantageous features. Their popularity as recombinant vectors is largely due to the successful and safe immunization of millions of US military recruits with enteric coated Ad4 and Ad7 to prevent against acute respiratory disease (ARD) outbreaks (Gaydos and Gaydos, 1995). In addition, the Ad genome is well characterized, and easily manipulated. Recombinant Ad vectors have been generated, with different deletions in their genome to render the vector replication deficient and to allow for insertion of foreign DNA sequences. Current generations of Ad vectors have insert capacity ranging from 7-9 Kb, and are deleted in one or more combinations of the early genes. Recombinant viruses are stable and stocks can be concentrated to titers higher than 1012 plaque forming units / ml. The virus has a broad cellular host range, its up take is not restricted to dividing cells, has natural tropism to liver, lung and intestine which is dictated by route of delivery (vascular, inhalation and oral, respectively) and persists as an episome in infected cells. It encodes for a cascade of polypeptides including the capsid structural proteins, the hexones, pentons (fiber and penton base). Recent studies have shown (Ad) vectors to be among the most efficient gene transfer vehicles for both in vitro and in vivo delivery. The general utilization, however, of current Ad vectors for many gene therapy applications is limited by the transient nature of transgene expression observed (Muzzin et al., 1996, Stratford-Perricaudet et al., 1990, Herz and Gerard, 1993, Morsy et al., 1993a, Morsy et al., 1993b, Morsy et al., 1996). Several factors have been shown to contribute to and modulate the duration of Admediated gene expression and the underlying immunogenicity of these vectors. These factors include “leaky” viral protein expression and/or the immuno-genicity of the transgene that is delivered (Yang, 1995, GaherySegard et al., 1997, Kaplan et al., 1997, Tripathy et al., 1996, Worgall et al., 1997). The development of Ad vectors that are deleted in all viral protein-coding sequences offers the prospect of a potentially safer, less immunogenic vector with an insert capacity of up to approximately 37 kb (Mitani et al., 1995, Kochanek et al., 1996, Clemens et al., 1996, Fisher et al., 1996, Kumar-Singh and Chamberlain, 1996, Hardy et al., 1997, Lieber et al., 1996, Parks et al., 1996, Schiedner et al., 1998, Haecker et al., 1996). This vector requires viral regulatory and structural proteins which, when supplied in trans, can support packaging and rescue, and is thus named 224

helper-dependent (HD) (Parks et al., 1996). It is noteworthy however, to emphasize that such modifications would modulate the toxicity and enhance the safety of the HD vehicle itself, yet may or may not have an effect on the impact or extent of transgene immunogenicity. A modified Ad vector has been generated such that it is completely devoid of viral protein encoding sequences (Morsy et al., 1998a). The new vector contains the ITR and packaging sequences, with an insert capacity of up to 37 kb. The propagation of this vector requires supplementation of the viral proteins in trans, which presently are supplied by co-propagation of a helper virus. Both viruses can be separated on a cesium gradient. Further more the helper virus is crippled by flanking the packaging signal sequence with lox sites that allow the excision of the intervening DNA, in the presence of cre, and thus the capacity of the helper virus to rescue itself as it propagates. This vector has been previously used to clone the full-length murine dystrophin cDNA (13.8 kb) under the control of the murine muscle specific creatinine kinease (6.5 kb) and a CMV promoter - E.coli LacZ gene cassette (4.6 kb) (Kochanek et al., 1996). This completely debilitated, recombinant virus propagated efficiently in 293 cells (which complement E1 functions) in the presence of a helper mutant Ad virus (SV5). The yield after cesium chloride density gradient banding was ~ 5x109 pfu obtained from 1.4x108 293 cells with about 1% contamination of helper virus as determined by a plaque forming unit (pfu )assay on 293 cells and by southern blot analysis. The recombinant HD vector was efficient in co-expressing the dystrophin protein and ß-gal in primary myoblasts derived from mdx mouse ( a genetic and biochemical model for human DMD disease) and in vivo (Kochanek et al., 1996, Clemens et al., 1996). In a more recent study we delivered the leptin cDNA using the HD virus, testing the hypothesis that elimination of the viral protein coding sequences would diminish the vector’s cellular immunogenicity and toxicity, and hence support its longevity in vivo (Morsy et al., 1998a-c). Since both the viral proteins and the transgene were factors implicated in the cellular immunogenicity of recombinant Ad viruses, we designed experiments to compare the HD and Ad vectors in ob/ob mice that are naive to leptin (in which the protein is potentially immunogenic), as well as in lean mice that normally express leptin. In this study, we showed that HD-leptin provided greater safety as reflected by absence of liver toxicity, cellular infiltrates, extended longevity of gene expression and stability of vector DNA in livers of treated mice over that observed with 1st generation Ad-mediated leptin treatment.

III. Safety of HD compared to 1st generation ad vectors

Gene Therapy and Molecular Biology Vol 3, page 225 Mice were treated with a single tail intravenous infusion of 1-2x10 11 particles of either HD-leptin, Ad-leptin, control Ad-ß-gal vector or an equal volume of control buffer. Toxicity was evaluated by measuring the levels of released liver enzymes in sera and by studying the histopathology of liver sections obtained from treated animals at successive intervals post treatment. Figure 1 shows the levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) in the sera of lean mice at one, two and four weeks post-treatment (similar results were observed in treated ob/ob mice - data not shown). Liver toxicity, as reflected by the significant elevation in AST and ALT serum levels over basal control levels, was observed only in mice treated with Ad-ß-gal and Ad-leptin, but not HD-leptin. Advector-associated toxicity observed in both the lean and ob/ob treated mice was most significant at one week, was present but to a less significant extent at two weeks, and was resolved by four weeks post-treatment. In contrast, HDtreatment was not associated with liver toxicity as reflected by the AST and ALT serum levels which were essentially indistinguishable from controls. Liver sections of HD-leptin-treated lean mice were histologically indistinguishable from control liver sections at all time points tested post treatment (Table 1). In contrast, Ad-leptin and Ad-ß-gal treated mice displayed hepatic pathology (hepatopathy) throughout the first 1-2 weeks posttreatment, which resolved by week 4.

IV. Efficacy of HD compared to 1st generation Ad vectors In the lean mice, treatment with Ad-leptin resulted in weight loss that lasted for only 7-10 days which was associated with a transient increase in serum leptin levels (Figure 2A and B). In contrast, treatment with HD-leptin resulted in approximately 20% weight loss that persisted at least two months and high serum leptin levels (6- to 10-fold over background) (Figure 2A and B). Weight loss in HDleptin-treated mice was associated with satiety that persisted over a longer period (2-3 weeks) than in those treated with Ad-leptin (5-7 days) (Figure 2C). Vector DNA in the livers of Ad-leptin treated mice was rapidly lost and fewer than 0.2 copies per cell were detected, compared to 1-2 copies per cell following HD-leptin treatment at 8 weeks post-injection (data not shown). These effects can be correlated with the duration of gene expression obtained with these two vector types. Gene expression mediated by Ad-leptin was transient and almost undetectable as early as 1 week post treatment as seen by northern blot analysis of total liver RNA, whereas that mediated by HD-leptin persisted for at least eight weeks (data not shown).

Figure 1. Mice were treated with Ad-ß-gal, Ad-leptin, HD-leptin or dialysis buffer (controls). Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels in the sera of lean control and treated mice are plotted at one, two and four weeks posttreatment.

225

Morsy et al: Helper-dependent adenoviral vectors as gene delivery vehicles Table 1. Table 1 is a summary of the liver histopathology findings. Histopathy refers to the displayed degenerative hepatic pathology found in livers of Ad-leptin and Ad-Ă&#x;-gal treated mice. This hepatic pathology was characterized by foci of round cell infiltration composed almost entirely (> 98%) of T-cells, individual liver cell necrosis, increased liver cell mitotic activity, and dissociation of hepatic cords. At two weeks posttreatment, Ad-leptin treated mice display a similar, but less pronounced hepatic pathology. The cellular infiltration observed resolved by four weeks post-treatment; there was almost an absence of lesions in the Ad-leptin treated mice, with only a trace of individual cell death present, which is within normal ranges. Examination of liver sections obtained from ob/ob mice reflected similar Ad-vector associated histopathology. Liver histology was indistinguishable between HD-leptin-treated and untreated control mice.

The ob/ob mice are naive to leptin and thus transgene immunogenicity is not an unexpected finding. In these animals, similar to what was observed in the lean mice, HDleptin was found to be more effective than the firstgeneration Ad-leptin vector. In the ob/ob mice treated with Ad-leptin, transient body weight loss of ~25%, followed by weight gain, two weeks after treatment was observed (Figure 3A and B). Associated, serum levels of leptin increased only for a short period during the first 4 days of treatment, returning to baseline levels within ten days postinjection (Figure 3C). Similar to the results obtained in lean mice, the Ad-leptin vector DNA (data not shown) was also rapidly lost (< 0.2 copies per cell were detected by 2 weeks post treatment, and undetectable by 8). In contrast, the ob/ob HD-leptin-treated mice had increased serum leptin levels up to ~15 days post-treatment, after which the levels gradually dropped to baseline over the subsequent 25 days (Figure 3C). The initial rise in leptin levels correlated with rapid weight reduction resulting in > 60% weight loss (reaching normal lean weight) by one month (Figure 3A). Weight loss was maintained for a period of 6-7 weeks post-treatment. The overall HD-leptin-mediated prolonged effect was also reflected in the accompanying phenotypic correction, which lasted longer than that seen in litter mates treated with Adleptin (6-7 versus 2-3 weeks) (Figure 3B ). As leptin levels dropped to baseline (Figure 3C), a gradual increase in body weight was observed (Figure 3A). Satiety was observed in association with increased leptin levels, and appetite suppression was sustained for a longer period (~1 month) compared to the short transient effect induced by Ad-leptin (~10 days) (Figure 3D). Leptin-specific antibodies were

226

detected in the sera of ob/ob Ad-leptin- and HD-leptintreated mice (data not shown) suggesting immunogenicity of leptin in these naive animals. Results of southern blot analysis showed greater stability of HD-vector DNA over Ad-vector DNA in livers of ob/ob treated mice compared at similar time points, the analysis revealed eventual loss of the HD-vector DNA over the 8 week time interval (data not shown). Approximately 75% less vector DNA was detected in the livers of HD-leptin-treated ob/ob mice at 4 and 8 weeks post-treatment compared to the persistent levels found in the livers of HD-leptin-treated lean litter mates at similar time points (data not shown). Gene expression in ob/ob Adleptin-treated mice correlated with the DNA findings, RNA levels were below the sensitivity level of detection at one week post-treatment, whereas in HD-leptin-treated mice, gene expression was detected up to four weeks postinjection and was undetectable at eight weeks (data not shown).

V. Future prospects for gene therapy: regulation of gene expression Development of gene therapy vectors allowing ligandinducible control of therapeutic gene expression will surely impact the safety and efficacy of future gene therapy protocols. Inducible systems capable of modulating gene expression in a ligand-dependent manner are being tested in a variety of different viral vectors with increasing frequency. Regulatable systems consist of a chimeric transactivator and its inducible promoter. In the examples which follow, the transactivators contain the following constituents:

Gene Therapy and Molecular Biology Vol 3, page 227 Figure 2. HD-leptin and Ad-leptin effects in lean mice. Animals were injected via the tail vein with a single dose of 1-2 x 1011 particles of HDleptin (n=5), Ad-leptin (n=10), Ad-Ă&#x;gal (n=10), or the equivalent volume of dialysis buffer (n=10). The time course shows (A) weight (grams) and the percent of maximum weight loss observed in Ad-leptin and HD-leptin treated mice (8% and 20%, respectively); (B) serum leptin levels, collected 2-3 times weekly (ng/ml) and the maximum fold increase in serum leptin levels above basal observed in the Ad-leptin and HD-leptin treated mice ( 6 fold and 10 fold, respectively); (C) the percentage of food intake relative to untreated control mice. The dashed line marks day 0 relative to day of injection.

VI. Tetracycline-inducible system

(i) functional elements(s) which interact with a small molecule inducing compound; (ii) a DNA-binding domain which does not exhibit cross reactivity with endogenous cellular sequences; and (iii) a transactivation domain. The inducible promoter consists of a minimal promoter (or TATA box) linked downstream to repeats of the transactivator recognition sequence. In the presence of inducer, the chimeric transcription factor should bind specifically to its DNA recognition sequence within the inducible promoter to transactivate target gene expression. Below is a brief review of the performance of the most promising drug-inducible gene expression systems in viral vectors.

A regulatable gene expression system utilizing the bacterial tetracycline repressor protein was originally described by Gossen and Bujard (1992). The tetracycline repressor was fused to a viral transactivation domain to form a tetracycline-controlled transactivator (tTA). This chimeric protein could activate transcription in mammalian cells from an inducible promoter consisting of tetracycline operator sequences fused to a minimal promoter. In the presence of tetracycline, protein -ligand interactions produced a conformational change in tTA so that it could no longer bind operator sequences and activate transcription, thus making this a tetracycline-repressible system.

227

Morsy et al: Helper-dependent adenoviral vectors as gene delivery vehicles

Figure 3. HD-leptin and Ad-leptin effects in ob/ob mice. Essentially as described in Fig. 2, animals were injected in the tail vein with a single dose of 1-2 x 1011 particles of HD-leptin (n=5), Ad-leptin (n=10), Ad-Ă&#x;-gal (n=10), or the equivalent volume of dialysis buffer (n=10). Lean control values are plotted for comparison . The time course shows (A) weight (grams) and the percent of maximum weight loss observed in Ad-leptin and HD-leptin treated mice (20% and 60%, respectively); (B) Phenotypic correction of HD-leptin-treated ob/ob mice. On the left is a representative ob/ob mouse treated with HD-leptin at day 54 post-treatment, next to a litter mate treated with Ad-leptin. The Ad-leptin-treated mouse initially lost weight during the first two weeks following the treatment, and subsequently gained weight. At 54 days post-Ad-leptin treatment, ob/ob mice are indistinguishable from untreated ob/ob control litter mates, whereas HD-leptin-treated mice remained indistinguishable from untreated lean control mice. Untreated ob/ob and lean control mice are shown for comparison as labeled. (C) serum leptin levels, collected 2-3 times weekly (ng/ml); (D) the percentage of food intake relative to untreated control mice. The dashed line marks day 0 relative to day of injection.

Figure 4. Reverse tetracyclineregulated gene expression. The reverse tetracycline-controlled transactivator (rtTA) consists of the reverse tetracycline repressor (rtetR) fused to a VP16 transactivation domain. In the presence of doxycycline (Dox), rtTA binds and transactivates transcription from the inducible promoter consisting of seven tet operator sites (tetO) located upstream of the human cytomegalovirus (hCMV) immediate early minimal promoter.

228

Gene Therapy and Molecular Biology Vol 3, page 229

Figure 5. Antiprogestin-regulated gene expression. The optimized transactivator GLVPc’ consists of a yeast GAL4 DNAbinding domain fused to a truncated human progesterone receptor ligand binding domain (hPR-LBD) and a VP16 transactivation domain. In the presence of RU486, GLVPc’ binds to an inducible promoter consisting of four GAL4 DNA-binding sites fused to the adenovirus E1B minimal promoter to transactivate gene expression.

The repressible system was later modified to an inducible system following the isolation of tetracycline repressor mutants conferring a reverse phenotype (Gossen et al., 1995). In contrast to the wild type protein, the reverse tetracycline repressor required tetracycline or a derivative such as doxycycline to bind operator sequences. Correspondingly, the reverse tetracycline transactivator (rtTA) could now activate gene expression in the presence of drug, rendering the system more suitable for therapeutic applications (Figure 4).

Comparable results were recently achieved using a recombinant AAV vector expressing rtTA and inducible Epo (Bohl et al., 1998). A single intramuscular injection of AAV into normal mice was sufficient to sustain Epo expression in a doxycycline-dependent manner over a 29-week period. Serum Epo levels were approximately 10-fold higher than basal concentrations.

Several retroviral vectors designed to evaluate tetracycline-inducible control of gene expression have been described (Bohl et al., 1997; Lindemann et al., 1997; Watsuji et al., 1997). Doxycycline-regulated control of erythropoietin (Epo) secretion was maintained for a period of 5 months in immunocompetent mice receiving implants of primary myoblasts transduced by a retroviral vector carrying rtTA and inducible Epo (Bohl et al., 1997). Although basal transcription under uninduced conditions was observed, overall induction of Epo secretion increased 70-fold. In similar cell implantation experiments, Lindemann et al. (1997) reported significant induction of doxycycline-regulated human growth hormone secretion in C3H mice for 46 days. Basal activity from the inducible promoter was apparent directly after implantation, but decreased over time. In vitro, an 1800-fold induction of chloramphenicol acetlytransferase (CAT) activity in response to doxycycline administration was observed in coinfection experiments utilizing recombinant adenoviral vectors expressing rtTA and an inducible CAT reporter construct (Molin et al., 1998). In vivo, intramuscular injection of these vectors into immunocompetent mice gave high levels of induction of CAT expression with variable basal activity.

The human progesterone receptor (hPR) is a member of the nuclear hormone receptor superfamily whose functional ligand binding domain (LBD) has been used to inducibly regulate gene expression. Original hPR-based systems utilized a human progesterone receptor with a truncated ligand binding domain (Wang et al., 1994; Delort and Cappechi, 1996). While the truncated receptor is no longer able to bind its natural ligand progesterone, it retains the ability to bind progesterone antagonists such as RU486 (Garcia et al., 1992; Vegeto et al., 1992). In developing an inducible system, a chimeric transactivator consisting of a viral transactivation domain fused to a yeast GAL4 DNAbinding domain and the truncated hPR LBD was constructed. The inducible promoter was composed of a series of GAL4 recognition sequences upstream of the adenovirus E1B TATA box.

VII. Antiprogestin-regulated gene switch

Subsequent modifications to the transactivator have increased the transcriptional activation potency and sensitivity of the system to inducing compound (Wang et al., 1997). Extension of the ligand binding domain and repositioning of the transactivation domain to the carboxy terminus produced an optimized transactivator capable of activating target gene expression at RU486 concentrations as low as 0.01nM (Figure 5).

229

Morsy et al: Helper-dependent adenoviral vectors as gene delivery vehicles Performance of the antiprogestin-regulated system has recently been tested in adenovirus and herpes simplex virus vectors. In vitro, approximately 600-fold induction of gene expression in response to RU486 was observed upon coinfection of cells with recombinant adenoviral transactivator and reporter vectors (Molin et al., 1998). In vivo, stereotactic injection of herpes simplex virus vectors containing the transactivator and an inducible lacZ reporter gene into rat hippocampus produced 150-fold activation of reporter expression following i.p. administration of RU486 (Oligino et al., 1998).

VIII. Dimerization-based gene regulation Development of a regulatable system based on chemical inducers of dimerization (CIDs) stems from earlier studies delineating the the mechanism of action of immunosuppressant compounds such as FK506, rapamycin, and cyclosporin A. These drugs functionally inhibit signalling pathways affecting T-cell activation and proliferation by mediating the dimerization and inactivation of endogenous cellular proteins involved in these processes. Rivera et al. (Rivera et al., 1998) have employed a dimerization-based strategy to develop a humanized system for inducing gene expression in vivo. In this system, a transcriptional transactivation domain is provided by the carboxy-terminal region of the NF!B p65 protein, which is

fused to the rapamycin-binding domain of FKBP12rapamycin associated protein (FRAP). The DNA-binding domain, termed ZFHD1, is a composite zinc fingerhomeodomain chimeric protein with novel DNA recognition specificity (Pomerantz et al., 1995) fused to a series of three repeats of the cellular protein FKBP12. These two proteins dimerize in the presence of rapamycin to form a functional transactivator which binds an inducible promoter containing ZFHD1 binding sites upstream of an hCMV minimal promoter (Figure 6). Although reports demonstrating the utility of such a system in viral vectors have not yet been published, the rapamycin-inducible system has been used to regulate hGH secretion in a cell implantation model in nude mice (Magari et al., 1997). Upon administering rapamycin to nude mice implanted with cells stably expressing hGH under control of the inducible promoter, peak levels of hGH secretion were observed within 24 hours with no detectable basal activity. Overall induction levels were found to be dependent on a number of parameters, including the dose of rapamycin received.

IX. Conclusions The leptin model used in the studies described above have provided a very instructive animal model to investigate

Figure 6. Rapamycin-regulated gene expression. The DNA-binding domain (ZFHD1) of the functional transactivator is a composite zinc finger/homeodomain chimera fused to three FKBP12 repeats (FKBP). The transactivation domain consists of the carboxy-terminal portion of the NF!B p65 protein (p65) fused to the FKBP12-rapamycin binding (FRB) domain from the cellular protein FRAP. NLS denotes nuclear localization signals. In the presence of rapamycin, DNA-binding and transactivation domains dimerize via FKBP and FRB to transactivate gene expression from an inducible promoter consisting of twelve ZFHD1 binding sites and an hCMV minimal promoter.

230

Gene Therapy and Molecular Biology Vol 3, page 231

References

the influence of both vector design and transgene product on the duration of expression after gene transfer. The HDvector system is a significant advance over existing Ad vectors with regards to safety, vector-mediated immunogenicity and insert capacity (up to 37kb). In addition to the gain of these valuable properties, the HD-vectors have not lost the features that contributed to the general attractiveness of Ad vectors which include: (i) efficient in vivo gene delivery, and (ii) high titer production. This system has come a long way in terms of development and ease of vector preparation and purification. Several studies involving the development of helper-dependent vectors were hindered by the complexity of the system (Mitani et al., 1995, Kochanek et al., 1996, Kumar-Singh and Chamberlain, 1996, Hardy et al., 1997, Lieber et al., 1996).

Bohl, D., Naffakh, N. and Heard, J. H. ( 1997) Long-term control of erythropoietin secretion by doxycycline in mice transplanted with engineered primary myoblasts. Nat Med, 3, 299-305. Bohl, D., Salvetti, A., Moullier, P. and Heard, J. M. (1998) Control of erythropoietin delivery by doxycycline in mice after intramuscular injection of adeno-associated vector. Blood, 92, 1512-1517. Clemens, P. R., Kochanek, S., Sunada, Y., Chan, S., Chen, H. H., Campbell, K. P. and Caskey, C. T. (1996) In vivo muscle gene transfer of full-length dystrophin with an adenoviral vector that lacks all viral genes. Gene Ther, 3, 965-972. Delort, J. P. and Cappechi, M. R. (1996) TAXI/UAS: A molecular switch to control expression of genes in vivo. Hum Gene Ther, 7, 809-820. Fisher, K. J., Choi, H., Burda, J., Chen, S. J. and Wilson, J. M. (1996) Recombinant adenovirus deleted of all viral genes for gene therapy of cystic fibrosis. Virology, 217, 11-22. Gahery-Segard, H., J. V., Gaston J., Lengagne R., Pavirani A., Boulanger P. and J.G., G. (1997) Humoral immune response to the capsid components of recombinant adenoviruses: routes of immunization modulate virus-induced Ig subclass shifts. Eur J Immunol, 27, 653-659. Garcia, T., Benhamou, B., Gofflo, D., Vergezac, A., Philibart, D., Chambon, P. and Gronemeyer, H. (1992) Switching agonistic, antagonistic, and mixed transcriptional responses to 11 betasubstituted progestins by mutation of the progesterone receptor. Mol Endocrinol, 6, 2071-2078. Gaydos, C. A. and Gaydos, J. C. (1995) Adenovirus vaccines in the U.S. military. Mil Med, 160, 300-304. Gossen, M. and Bujard, H. (1992) Tight control of gene expression in mammalian cells by tet-responsive promoters. Proc. Natl. Acad. Sci. USA, 89, 5547-5551. Gossen, M., Freundlieb, S., Bender, G., Muller, G., Hillen, W. and Bujard, H. (1995) Transcriptional activation by tets in mammalian cells. Science, 268, 1766-1769. Haecker, S. E., Stedman H.H., Balice-Gordon R.J., Smith D.B., Greelish J.P., Mitchell M.A., Wells A., Sweeney H.L. and J.M., W. ( 1996) In vivo expression of full-length human dystrophin from adenoviral vectors deleted of all viral genes. Hum Gen Ther, 7, 1907-1914. Hardy, S., Kitamura, M., Harris-Stansil, T., Dai, Y. and Phipps, M. L. (1997) Construction of adenovirus vectors through Cre-lox recombination. J Virol, 71, 1842-1849. Harvey, D. and Caskey, C. T. (1998) Inducible control of gene expression: prospects for gene therapy. Curr Opin Chem Biol, 2, 512-518. Herz, J. and Gerard, R. D. (1993) Adenovirus-mediated transfer of low density lipoprotein receptor gene acutely accelerates cholesterol clearance in normal mice. Proc Natl Acad Sci USA, 90, 2812-2816. Ingalls, A., Dickie, M. and Snell, G. (1950) Obese, a new mutation in the house mouse. J Hered, 41, 317-318. Kaplan, J. M., Armentano, D., Sparer, T. E., Wynn, S. G., Peterson, P. A., Wadsworth, S. C., Couture, K. K., Pennington, S. E., St George, J. A., Gooding, L. R. and A.E., S. (1997) Characterization of factors involved in modulating persistence of transgene expression from recombinant adenovirus in the mouse lung. Hum Gene Ther, 8, 45-56.

The characterization of size requirements for efficient packaging and the generation of crippled helper viruses greatly enhanced the prospects of these new vectors in becoming a promising tool for gene delivery (Parks and Graham, 1997, Parks et al., 1996). Further modifications and fine tuning are required to convert the HD vector system to an industrially scaleable system for clinical utility. Regarding the potential for drug-inducible gene expression in viral vectors, a number of requirements will ultimately need to be fulfilled before regulated systems such as those described here are included in human gene therapy protocols. First, components of the system (including the transactivator, inducible promoter, and inducing compound) should not interfere with normal cellular processes. Second, a strong induction profile of therapeutic gene expression in the presence of drug coupled with low basal activity in the uninduced state will be important for general applicability of the system. Third, the inducer will preferably be an orally bioavailable, physiologically inert compound that is cleared from body tissues within a reasonable length of time. Finally, due to potential host immune recognition, the transactivator protein(s) should ideally be nonimmunogenic. This is particularly important in cases where long-term correction of a genetic deficiency is desired. Although no inducible gene regulation system can claim to meet all of these requirements thus far, their continued evaluation in the context of gene therapy vectors will hopefully yield optimized systems capable of making exogenous control of gene regulation a reality.

Acknowledgement The authors would like to thank Mrs. Maria Grimsley for her excellent help and assistance in preparing this manuscript.

231

Morsy et al: Helper-dependent adenoviral vectors as gene delivery vehicles Kochanek, S., Clemens, P. R., Mitani, K., Chen, H. H., Chan, S. and Caskey, C. T. (1996) A new adenoviral vector: Replacement of all viral coding sequences with 28 kb of DNA independently expressing both full-length dystrophin and betagalactosidase. Proc Natl Acad Sci USA, 93, 5731-5736. Kumar-Singh, R. and Chamberlain, J. S. (1996) Encapsidated adenovirus minichromosomes allow delivery and expression of a 14 kb dystrophin cDNA to muscle cells. Hum Mol Genet, 5, 913-921. Lieber, A., He, C. Y., Kirillova, I. and Kay, M. A. (1996) Recombinant adenoviruses with large deletions generated by Cre-mediated excision exhibit different biological properties compared with first-generation vectors in vitro and in vivo. J Virol, 70, 8944-8960. Lindemann, D., Patriquin, E., Feng, S. and Mulligan, R. C. (1997) Versatile retrovirus vector systems for regulated gene expression in vitro and in vivo. Mol Med, 3, 466-476. Magari, S. R., Rivera, V. M., Luliucci, J. D., Gilman, M. and Cerasoli, F. Jr (1997) Pharmacologic control of a humanized gene therapy system implanted into nude mice. J Clin Invest, 100, 2865-2872. Mitani, K., Graham, F. L., Caskey, C. T. and Kochanek, S. (1995) Rescue, propagation, and partial purification of a helper virusdependent adenovirus vector. Proc Natl Acad Sci USA, 92, 3854-3858. Molin, M., Shoshan, M. C., Ohman-Forslund, K., Linder, S. and Akusjarvi, G. (1998) Two novel adenovirus vector systems permitting regulated protein expression in gene transfer experiments. J Virol, 72, 8358-8361. Morsy, M. A., Alford, E. L., Bett, A., Graham, F. L. and Caskey, C. T. (1993b) Efficient adenoviral-mediated ornithine transcarbamylase expression in deficient mouse and human hepatocytes. J Clin Invest, 92, 1580-1586. Morsy, M. A., and Caskey, C. T. (1998c) Helper dependent adenoviral vectors – improved safety and expression. Biogenic Amines, 14, 433-449. Morsy, M. A., Gu, M., Motzel, S., Zhao, J., Lin, J., Su, Q., Allen, H., Franlin, L., Parks, R. J., Graham, F. L., Kochanek, S., Bett, A. J. and Caskey, C. T. (1998a) An adenoviral vector deleted for all viral coding sequences results in enhanced safety and extended expression of a leptin transgene. Proc Natl Acad Sci USA, 95, 7866-7871. Morsy, M. A., Gu, M., Zhao, J. Z., Holder, D. J., Rogers, I. T., Pouch, W., Motzel, S. L., Klein, H. J., Gupta, S. K., Liang, X., Tota, M. R., Rosenblum, C. I. and Caskey, C. T. (1998b) Leptin gene therapy and daily protein administration: a comparative study in the ob/ob mouse. Gene Ther, 5, 8-18. Morsy, M. A., Mitani, K., Clemens, P. and Caskey, C. T. (1993a) Progress toward human gene therapy. Jama, 270, 2338-2345. Morsy, M. A., Zhao, J. Z., Warman, A. W., O'Brien, W. E., Graham, F. L. and Caskey, C. T. (1996) Patient selection may affect gene therapy success. Dominant negative effects observed for ornithine transcarbamylase in mouse and human hepatocytes. J Clin Invest, 97, 826-832. Muzzin, P., Eisensmith, R. C., Copeland, K. C. and Woo, S. L. C. (1996) Correction of obesity and diabetes in genetically obese mice by leptin gene therapy. Proc Nat Acad Sci USA, 93, 14804-14808. Oligino, T., Poliani, P. L., Wang, Y., Tsai, S. Y., O’Malley, B. W., Fink, D. J. and Glorioso, J. C. (1998) Drug inducible transgene

232

expression in brain using a herpes simplex virus vector. Gene Therapy, 5, 491-496. Parks, R. J. and Graham, F. L. (1997) A helper-dependent system for adenovirus vector production helps define a lower limit for efficient DNA packaging. J Virol, 71, 3293-3298. Parks, R. J., Chen, L., Anton, M., Sankar, U., Rudnicki, M. A. and Graham, F. L. (1996) A helper-dependent adenovirus vector system: removal of helper virus by Cre-mediated excision of the viral packaging signal. Proc Natl Acad Sci USA, 93, 13565-13570. Pomerantz, J. L., Sharp, P. A. and Pabo, C. O. (1995) Structurebased design of transcription factors. Science, 267, 93-96. Rivera, V. M., Clackson, T., Natesan, S., Pollock, R., Amara, J. F., Keenan, T., Magari, S. R., Phillips, T., Courage, N. L., Cerasoli, F. Jr, Holt, D. A. and Gilman, M. (1996) A humanized system for pharmacologic control of gene expression. Nat Med, 2, 1028-1032. Schiedner, G., Morral, N., Parks, R. J., Wu, Y., Koopmans, S. C., Langston, C., Graham, F. L., Beaudet, A. L. and Kochanek, S. (1998) Genomic DNA transfer with a high-capacity adenovirus vector results in improved in vivo gene expression and decreased toxicity. Nat Genet, 18, 180-183. Stratford-Perricaudet, L. D., Levrero, M., Chasse, J., Perricaudet, M. and Briand, P. (1990) Evaluation of the transfer and expression in mice of an enzyme-encoding gene using a human adenovirus vector. Hum Gene Ther, 1, 241-256. Tripathy, S. K., Black, H. B., Goldwasser, E. and Leiden, J. M. (1996) Immune responses to transgene-encoded proteins limit the stability of gene expression after injection of replicationdefective adenovirus vectors. Nat Med, 2, 545-550. Vegeto, E., Allan, G. F., Schrader, W. T., Tsai, M. J., McDonnell, D. P. and O’Malley, B. W. (1992) The mechanism of RU486 antagonism is dependent on the conformation of the carboxyterminal tail of the human progesterone receptor. Cell, 69, 703713. Wang, Y., O’Malley, B. W. Jr, Tsai, S. Y. and O’Malley, B. W. (1994) A regulatory system for use in gene transfer. Proc Natl Acad Sci USA, 91, 8180-8184. Wang, Y., Xu, J., Pierson, T., O’Malley, B. W. and Tsai, S. Y. (1997) Positive and negative regulation of gene expression in eukaryotic cells with an inducible transcriptional regulator. Gene Ther, 4, 432-441. Watsuji, T., Okamoto, Y., Emi, N., Katsuoka, Y. and Hagiwara, M. (1997) Controlled gene expression with a reverse tetracyclineregulated retroviral vector (RTRV) system. Biochem Biophys Res Comm, 234, 769-773. Worgall, S., Wolff, G., Falck-Pedersen, E. and Crystal, R. G. (1997) Innate immune mechanisms dominate elimination of adenoviral vectors following in vivo administration. Hum Gene Ther, 8, 37-44. Yang Y, Li Q, Ertl HC, Wilson JM, (1995) Cellular and humoral immune responses to viral antigens create barriers to lungdirected gene therapy with recombinant adenoviruses. J. Virol., 69, 2004-2015.

Gene Therapy and Molecular Biology Vol 3, page 233 Gene Ther Mol Biol Vol 3, 233-241. August 1999.

Gene transfer into muscle for the treatment of muscular dystrophy and haemophilia Review Article

Geoffrey Goldspink, Maria Skarli and Paul Fields The Royal Free and University College Medical School, University of London, UK, Royal Free Campus, Rowland Hill St. London NW3 2PF, UK __________________________________________________________________________________________________ Correspondence: Geoffrey Goldspink, Ph.D., Department of Anatomy and Developmental Biology, Royal Free and University College Medical School, Royal Free Campus, Rowland Hill Street, London NW3 2PF, England. Tel: +44-171-830 2410; Fax: +44171-830 2917; E-mail: goldspink@rfhsm.ac.uk Key words: intramuscular, gene therapy, vectors, muscular dystrophy, haemophilia Received: 27 November 1998; accepted: 19 December 1998

Summary M u s c l e h a s p r o v e n t o b e a n a p p r o p r i a t e e x p r e s s i o n s y s t e m f o r g e n e s , the product o f which i s required in the general circulation as well as for muscle genes per se. This review deals with the design o f the gene constructs including the vectors and the regulatory elements required for optimisation o f expression f o l l ow i ng introduction o f the relevant cDNA by intramuscular injection. The relative merits and problems associated with each type o f vector including the immunogenic responses they elicit are discussed. Duchenne muscular dystrophy is used to illustrate the problems associated with gene therapy for a disease in which a muscle protein is defective or missing whilst haemophilia is chosen as an example of how a systemic protein, Factor VIII or IX, may be produced at low constitutive levels in muscle rather than liver.

introducing the plasmid constructs into young muscle (Wells and Goldspink 1992). Plasmids are very useful for transfecting muscle by intramuscular injection. However, the possibilities of using other vectors are discussed. The review also deals with the immune response to the vectors as well as the gene product of the introduced cDNA as this is one of the crucial issues in gene therapy.

I. Introduction The available evidence shows that skeletal muscle is an appropriate target tissue for the transfer of DNA for the treatment of a number of diseases. These include those in which a systemic protein is absent or defective as well as muscle diseases per se such as Duchenne muscular dystrophy. It has also been shown that muscle is also an appropriate target tissue for the introduction of vaccine DNA although the requirements for effective gene therapy is not to elicit an immune response. This review deals, therefore, with the use of muscle as a target tissue for gene therapy for the muscle dystrophies in which the defect is myogenic and for the haemophilias in which there is a requirement to produce steady systemic levels of a particular clotting factor that is defective or missing. Direct intramuscular injection of plasmid DNA was shown by Jon Wolff’s group (Wolff et al, 1990) to result in a small percentage of the fibres taking up and expressing the cDNA. This expression has been shown to persist for a considerable time (Wolff et al, 1992). Levels of expression were improved by using different regulatory sequences to drive the expression of the introduced cDNA (Hansen et al, 1991; Novo et al, 1995; Skarli et al, 1998) and by

II. Design of vectors and gene constructs for appropriate expression in muscle In any gene therapy protocol one needs to consider the construction of the gene cDNA or genomic DNA to be transferred, the vector backbone, as well as host factors in determining immune responses. Consideration needs to be given to these individual components since any one of them can influence the outcome or therapeutic efficacy of a gene therapy approach. In this review these factors will be discussed. However, it is clear that additional factors such as the host immune status (Michou et al, 1997), dose of vector (Svenson et al, 1997), and mode of delivery all

233

Goldspink et al: Gene transfer into muscle for the treatment of muscular dystrophy and haemophilia affect the effector immune response but are beyond the scope of this review.

actual plasmid backbone acts as an adjuvant to the stimulation of a host immune response. It is clear that this immune response is both cell mediated and humoural. The immune response is associated with the presence of noncoding immunostimulatory sequences (ISS) within the plasmid backbone which are centred around hypomethylated CpG base pairs (Sato et al, 1996). These motifs are encountered frequently in eukaryotic DNA. Less than 5% of the cytosines in the CpG base pairs in prokaryotic DNA are methylated, but the frequency of methylated cytosines in CpG base pairs in eukaryotic DNA is 70-90% (Tighe et al, 1997). These hypomethylated motifs rapidly stimulate the innate immune response with production of IFN-" by NK cells and IFN-! and #, IL-12 and IL-18 by macrophages. This response is important phylogenetically as it is the host’s first line of defense against bacterial infection. In the adaptive immune response bacterial DNA favours the development of a TH1-driven response and secretion of IFN-" favours immunoglobulin class switching to the IgG2a subtype. Because of the strong and persistence cell mediated and humoural immune response to the plasmid backbone and encoded transgene, the use of DNA vaccines has an enormous potential against infectious diseases, allergy, and cancer. However, for use in gene transfer to generate a secretable protein this strong immune response is counter productive (see below). Therefore, plasmid vectors for gene therapy strategies should be designed to lack these ISS sequences. Nevertheless, for vaccine-based strategies the incorporation of these ISS endowed with adjuvant properties would appear beneficial.

A. Gene constructs for intramuscular injection For intramuscular injection consideration needs to be given to whether the desired product is secretable or nonsecretable. For exocytosis from the muscle fibres it is essential to include an appropriate signal sequence. Also the antigen presentation to the immune system of these responses will differ depending on which MHC pathway is involved. It is known that MHC class I will present intracellularly derived peptides whereas secreted proteins may be processed via MHC II pathways. Most of the early published studies were performed with non-secretable reporter genes such as lacZ whose gene product (ß galactosidase) is non-secretable. Another problem that hampered early gene therapy studies was the availability of species-specific transgenes; most of the available cloned transgenes for early studies were human and these were tested in non-human animal hosts. Recently, many more animal cDNA`s have been cloned which will allow appropriate testing of gene therapy strategies in the homologous animal.

B. Regulatory elements For effective gene transfer into muscle, tissue-specific expression of the introduced gene is desirable. This can be achieved by the use of promoters and enhancers specific for the tissue of interest. In the case of muscle, creatine kinase (Wang et al, 1997), !-actin (Bergsman et al, 1986; Draghia-Akli et al, 1997), and myosin heavy chain (Skarli et al, 1998) have been used. The necessity for the development of vector bearing muscle-specific regulatory elements is further emphasised by recent studies which have shown that intramuscular injection of genes results in the uptake of a proportion of the vector by the motoneurones (Keir et al, 1995; Ribota et al, 1997). Engineered genes that are under the control of myosin DNA regulatory elements have several potential advantages. Myosin isoforms are differentially regulated by physical activity and by hormones and this, therefore, offers the possibility for developing inducible vectors. Also muscle-specific elements used to drive DNA are more likely perceived as safer than viral promoters.

2. Viral delivery: adenoviral vectors The first vectors to be tested in gene therapy approaches were based on the adenovirus; adenoviral vectors still have advantages and are currently being used particularly in the field of haemophilia. Advantages include the ease with which they can be prepared in high titre, their wide host range and their ability to transduce non-dividing target cell such as liver and muscle. However, one of the early problems encountered with the use of these vectors was a rapid decline in gene expression although the early expression levels attained were high; re-administration of the adenoviral vector was precluded because of immunogenicity; expression of the transgene was finally lost. Subsequent experiments revealed that the decline in expression appears to arise from the elimination of the therapeutically-transduced cells (both muscle and liver) by the host immune response (Dai et al, 1995, Yang et al, 1994a, Yang et al, 1994b). This resulted from a CTL (cytotoxic T lymphocyte) response induced by the adenoviral proteins. The immune response also appears to mediate the failure of expression following repeated administration of the vector. Because of these problems a number of immunomodulatory strategies are being developed. Engineering of less immunogenic adenoviral vectors (so called “gutted” adenoviral vectors) is being developed so that the vectors contain only the minimal cis elements required for replication and packaging but are

C. The vector The choice of vector is crucial for the outcome of the immune response; construction of vectors that addressed these problems in early studies involved non-viral as well as viral delivery methods. 1 . Non-viral delivery: plasmid approach For non-viral vectors, such as injection of naked plasmid DNA, there is now evidence to suggest that the 234

Gene Therapy and Molecular Biology Vol 3, page 235 devoid of any regions encoding viral proteins (Fisher et al, 1996, Hardy et al, 1997). Immunomodulatory strategies have involved the use of immunosuppressive drugs such as cylophosphamide and cyclosporin, and the blocking of costimulatory pathways involved in T cell activation (Kay et al, 1995). These strategies have met with varying degrees of success but once optimisation can be reached it would be a major advance in adenovirus-mediated gene transfer. Because of the problems encountered with these two approaches, investigators have sought alternative vectors for use in gene transfer strategies in muscle. One such candidate vector for gene transfer is the adeno-associated virus (AAV). Significantly, AAV-mediated gene transfer in muscle is not highly immunogenic, as is the case for adenovirus.

Duchenne muscular dystrophy. Apart from the large size of the dystrophin gene and its cDNA, there have been difficulties due to the immune response generated by the introduced gene and/or the vector. When a reasonably mature individual lacks a particular protein and is then exposed to it, immune responses are generated. In patients, these proteins even though they are intracellular, are regarded as foreign. This was highlighted by a case in which a patient with Becker muscular dystrophy received a heart transplant and antibodies directed to parts of the dystrophin protein were detected in the blood shortly afterwards (Britter et al, 1995). Becker is a milder form of muscular dystrophy in which a truncated form of dystrophin is produced. However, there was a marked immune response to the domains in the normal dystrophin of the transplanted heart, which are not present in the endogenous dystrophin of this Becker patient. This emphasised that even particular domains of intracellular proteins will be regarded as foreign if they have never been produced in that individual. As with other gene therapy procedures there have been problems with the immunogenicity of the vectors used. Gene transfer for Duchenne muscular dystrophy poses additional difficulties. The disease affects every skeletal muscle in the body as well as the heart and the respiratory muscles. Therefore the transferred gene has to be taken up and expressed by a large number of muscles. Ideally, this would be achieved by systemic administration of DNA but this presents a number of problems. The rate of DNA uptake is different in different tissues and is probably different in different muscles depending on factors such as the amount of connective tissue present and vascular density. Other factors to be considered are DNA loss via liver uptake, and DNA uptake by other tissues. Most studies on systemic administration of DNA have used DNA-liposome complexes. Liposomes may increase the uptake but their lipid components may also be toxic especially to the kidney and heart (Wright et al, 1998). The dystrophin gene is very complex and produces a number of alternatively spliced transcripts that are translated into functional proteins. Transcription of different isoforms is regulated by eight promoters and is a developmentally regulated and tissue-specific process (Fabbrizio et al, 1994). There are three long isoforms specific to skeletal muscle, heart and Purkinje cells of the cerebellum where the cDNA is 14 kb long. Most gene transfer vectors have a DNA insert limit well below this size. This has led to efforts to transduce the muscle fibres with truncated forms of the gene, for example a dystrophin minigene that restores part of the function (see below). Two types of viral vectors have been mainly used in gene transfer studies for muscular dystrophies. Retroviral vectors have been initially used but their efficiency in vivo is extremely limited because their uptake by cells requires mitotic division. Therefore their use in muscular dystrophy gene therapy has been limited to myoblast transfer studies (Fassati et al, 1997). More recently research has focused on the use of adenoviral vectors which are taken up very efficiently by muscle fibres (Ragot et al, 1994). However,

III. Muscular dystrophies A. Background and difficulties Duchenne muscular dystrophy (DMD) is a sex linked hereditary disease which afflicts about 1 in 3200 young boys associated with progressive wasting of the muscles. The pathological defects in DMD, which are believed to result from dystrophin deficiency, are profound and widespread affecting in particular skeletal and cardiac muscle. DMD is lethal in the 2nd or 3rd decade of life but its debilitating effects are seen as early as 3 years of age. When affected boys become about 12 years old, their developing skeletal deformations and progressing muscle weakness confines them to the wheelchair. Subsequent heroic surgical procedures include spine immobilisation so that they can still sit up in a chair and thus improve their quality of life to some extent. The large dystrophin gene is susceptible to de novo mutations (>30%) and this makes complete disease prevention impossible. Therefore, effective therapy for DMD, remains a prime goal of research in this area. The autosomal dystrophies including limb girdle have recently been shown to arise from mutations in the genes encoding the other components of the dystrophin complex. These include the dystroglycans and sarcoglycans and also the extracellular proteins merosin, the form of laminin found in muscle and peripheral nerve (Sunada et al, 1994, Xu et al, 1994). Dystroglycans and sarcoglycans are linked to glycoproteins that in turn are attached to the dystrophin which is associated with the membrane of the muscle fibre. The fact that the dystrophin itself (Milner et al, 1993, Shemanko et al, 1995) is phosphorylated and its associated glycoproteins contain numerous potential phosphorylation sites, strongly suggests that the whole complex is involved in gene activation. This is likely to involve a growth factor gene that in turn uses an established pathway to activate the transcription of a range of structural genes. Possibilities exist for replacing the proteins that are defective or not expressed in each of the different types of dystrophy. Because of the severity of the disease virtually all the dystrophy gene therapy work has been aimed at 235

Goldspink et al: Gene transfer into muscle for the treatment of muscular dystrophy and haemophilia first-generation adenoviral vectors induced a cellular immune and inflammatory response that precluded long term expression of the gene (Stratford-Perricaudet et al, 1990; Li et al, 1993). This immune response was thought to arise from the presence of viral sequences in the vector. Recently a new adenoviral vector has been developed with a capacity for 28kb of foreign DNA. This vector is devoid of virtually all viral coding sequences and has been successfully used to transduce fibres in vivo (Clemens et al, 1996; Kochanek et al, 1996; Chen et al, 1997; Floyd et al, 1998). A number of other viral vectors have been used in gene transfer studies in skeletal muscle. Of these adenoassociated viral constructs have bee shown to be effective. However, their limited capacity for foreign DNA (about 4.5kb) makes them unsuitable for use with dystrophin. However they may be suitable for other forms of gene transfer as for example for haemophilia B. Herpes simplex virus has also been used in muscle but uptake by muscle fibres is very limited (for review see Huard et al, 1997). One way to increase uptake by muscle fibres is to damage the muscle by inducing necrosis and regeneration by chemical means. Various agents have been used such as bupivacaine (Davis et al, 1993), notexin, and barium chloride. It has been shown that regenerating muscle exhibits increased uptake of DNA which is taken up mainly by the newly-formed fibres. The use of viral vectors is also limited by safety factors. Since helper virus is used to produce the virus in the encapsulated form that can be taken up by cells, the construct to be transferred has to be highly purified. This is difficult to achieve especially on a large scale and there is always the possibility of contamination by helper virus. In addition recombinant events may occur which could lead to the activation of oncogenes; this is especially important when regenerating muscle models are used. Another approach for the treatment of muscular dystrophy has been the use of isolated satellite cells stably transformed with dystrophin genes that are then reintroduced into the host organism. This approach is hindered by the fact that there is limited integration between the introduced cells and the existing muscle fibres, by the development of immune responses and the limited penetration of satellite cells though muscle. There have been a number of studies on animal models (Rando and Blau, 1994; Rando et al, 1995) but clinical trials (Miller et al, 1997) have shown that there is a negligible therapeutic value in this approach.

form) and a marked amelioration of the symptoms. These included a significant reduction in serum CPK levels and in the histopathological changes normally associated with the form of dystrophy. Later, the group of Jeffrey Chamberlain (Cox et al, 1993; Corrado et al, 1994) introduced different lengths of the dystrophin cDNA into mdx mice to define targeting to the membrane and levels of expression. Apparently the full length cDNA is required for good expression and, therefore, it is difficult to see how the immunological response can be circumvented. Further transgenic experiments are being carried out by this group to determine which domains of the dystrophin protein are strategic and which are immunogenic. Another alternative strategy has been pursued in which there is an attempt to introduce a substitute protein that is related to dystrophin and which is expressed in muscle early in development. Recently a related protein called utrophin has been discovered which appears to have arisen during evolution by duplication of the same gene as dystrophin. From this point of view of gene therapy, the upregulation of utrophin seems to offer an alternative strategy particularly as it is expressed before dystrophin in dystrophic as well as normal muscle. Hence, the same immunological problems do not apply. Transgenic experiments in which utrophin has been over expressed in mdx mice (Tinsley et al, 1996) indicate that it may function as a substitute for dystrophin in protecting muscle fibres from accumulated cell damage and cell death. Utrophin is also associated with dystroglycans and seems to have a similar function as dystrophin. It is reasonable to predict that is also involved in the mechanosignalling that is required to prevent the consequences of microdamage i.e. cell death and in the case of muscle, permanent loss of muscle fibres.

C. Down stream treatment for muscular dystrophy It is now over 10 years since the dystrophin gene was identified. However, we still do not know what function its complex gene product serves apart from, perhaps, stiffening the plasma membrane. The dystrophin gene itself is of a complex structure, the expression of which depends on the cell type. For example it is spliced differently in neuronal cells than in muscle cells. However, the functions of the different length transcripts and gene products are not understood. In skeletal muscle where has received the most attention, the dystrophin protein is known to form part of an elaborate complex that at the N end terminal is attached to actin filaments. At the C terminal end dystrophin is attached to an elaborate array of sarcoglycans, dystroglycans as well as the extracellular matrix via merosin. Tyrosine kinase and nNOS moieties are also associated with the dystrophin complex. It seems inconceivable therefore that this elaborate structure has evolved merely to stiffen the membrane. As with other cytoskeletal systems (Ingber 1997) we believe it is involved in mechanosignalling and gene regulation.

B. Alternatives to introducing the full length dystrophin cDNA. Several laboratories, including our own, have used mini-dystrophin genes (Wells et al, 1992). Using transgenic biology methods we introduced a human Becker type minigene (England et al, 1990) into the mdx mouse which suffers from a dystrophin deficiency dystrophy. This resulted in the expression of dystrophin (albeit a truncated

236

Gene Therapy and Molecular Biology Vol 3, page 237 repair and maintain muscle and the progressive nature of the dystrophies.

The study of the underlying mechanisms via which cells respond to mechanical stimuli i.e. the link between the mechanical stimulus and gene expression represent a new and important area of cellular physiology (Goldspink and Booth 1992). Various mechanisms have been proposed for the way in which the genes involved in local tissue growth and repair are activated by mechanical signals. These include the production of autocrine growth factors. Because muscle is a mechanical tissue and a tissue in which there is no cell replacement, it is vitally important that local repair is initiated as soon as any microdamage appears. The hypothesis is that the dystrophies are diseases in which the mechanochemical signalling, and hence the local repair mechanisms, are defective. It has been known for some time that there are local factors as well as systemic factors that regulate tissue growth. The growth hormone /insulin growth factor-1 (GH/IGF-1) axis is the main regulator of tissue mass during early life. Our group (Yang SY et al, 1996) has cloned the cDNA of a splice variant of IGF-1 that is produced by active muscle that appears to be the factor that controls local tissue repair, maintenance and remodelling. From its sequence it can be seen that it is derived from the IGF-1 gene by alternative splicing but it has different exons to the liver isoforms. Unlike the liver isoforms it is not glycosylated, is therefore smaller, probably has a shorter half life and is thus suited for an autocrine/paracrine rather than a systemic mode of action (Yang SY et al, 1996). It has a 52 base insert in the E domain that alters the reading frame of the 3' end. Therefore, this splice variant of IGF-1 is likely to bind to a different protein, e.g. BP5, which only exists in the interstitial tissue spaces of muscle, neuronal tissue and bone. This would be expected to localise its action as it would be unstable in the unbound form which is important as its production would not disturb unduly the glucose homeostasis mechanism. This new growth factor has been called mechano growth factor (MGF) to distinguish it from the liver IGFs that have a systemic mode of action (Goldspink et al, 1996). We have also shown that, in contrast to normal muscle, the mRNA for MGF is not detectable in dystrophic mdx muscles even when subjected to stretch and stretch combined with electrical stimulation (Goldspink et al, 1996). The systemic levels of IGF-1s are mainly controlled by growth hormone in the blood system. Interestingly, it has recently been shown that during intensive exercise most of the circulating IGF-1 is actually derived from the active muscles. Also most of the IGF-1 circulating IGF-1 is actually utilised by the musculature (Brahm et al, 1997). With age, however, the circulating growth hormone and IGF-1 are known to decrease markedly particularly after the initial growth spurt (Rudman et al, 1981). Although IGF1s produced via growth hormone stimulation are important during early post-natal muscle development, it appears that IGF-1 produced by muscle during exercise becomes more important for the maintenance of muscle mass. The decline in the production by the liver and the inability to supplement it by locally produced IGF-1 is most probably one of the main factors for the progressive inability to

D. Functions of IGF-1 and the need to supplement levels of the autocrine isoform in muscular dystrophy. As mentioned, there is strong evidence that IGF-1 is important in determining muscle mass and preventing dystrophy as has been deciphered from transgenic mouse experiments. Transgenic mice produced by introduction of the human IGF-1 cDNA under the control of a chicken actin promoter showed elevated muscle but not systemic levels of IGF-1. They also exhibited muscle fibre hypertrophy but with no significant increase in body weight (Coleman et al, 1995). As far as dystrophy is concerned the knockout experiments are very important. These include those in which the IGF-1 gene (PowellBraxton et al, 1993, Baker et al, 1993) was truncated or the IGF-1 receptor(s) were knocked out (Ayling et al, 1989) and which resulted in early, severe muscular dystrophy. As a consequence these mice died at, or just after, birth. Also, recombinant IGF-1 has been shown to have a marked beneficial affect on murine muscular dystrophy of the dydy dystrophic mouse (Zdanowitz et al, 1995). If the autocrine isoform of IGF-1 (MGF) was used it would be likely to be much more effective in preventing dystrophic changes of loss of muscle tissue. In ongoing experiments, intramuscular injection of a MGF plasmid construct into normal and dystrophic muscles has resulted in an amelioration of the histopathological changes such as the number of central nuclei. Further experiments need to be carried out on younger mice to see if the muscle fibres can be rescued and the progressive nature of the disease halted.

IV. Haemophilias The haemophilias are used as an example of a disease where muscle tissue may be used to deliver a non-muscle protein systemically. Clotting factors are normally synthesised in the liver, but work by a number investigators has shown that biologically active factor IX can be produced in other cell types including myoblasts (Yao et al, 1992), fibroblasts (Palmer et al, 1989) and endothelial cells (Yao et al, 1991). Thus the choice of the target cell is not limited as long as the clotting factor protein can gain access to the circulation. Haemophilia is a common disease that occurs worldwide. Haemophilia B that is due to factor IX deficiency affects approximately 1 in 10,000 male births; haemophilia A that is caused by factor VIII deficiency is approximately three times more common than haemophilia B. Even in developed countries the disease may be regarded as life threatening (as it often results in intracranial haemorrhage) without prophylactic treatment. Also chronic morbidity may arise from repeated joint bleeding resulting in joint contractures and deformity. Highly purified serumderived factor VIII (i.e. free of HIV, B and C hepatitis and prions) is very expensive and for prophylactic treatment it 237

Goldspink et al: Gene transfer into muscle for the treatment of muscular dystrophy and haemophilia is prohibitively expensive (about $80,000 per year). No health service can afford to fund these patients except in emergency situations and 80% of the global population of haemophiliacs do not have access to even emergency treatment. Recombinant factor VIII and IX are now available but these are also extremely expensive and not freely available. Concerns still remain over the safety of plasma-derived products. Haemophilia has been widely regarded as a target for gene therapy because it is a wellcharacterised single gene disorder and only a small correction in clotting level is necessary to significantly improve the bleeding phenotype in severely affected patients. Another major advancement towards treating haemophilia by gene therapy has been the recent development of large and small animal models that accurately mirror the human disease. These models will prove very useful in working up gene therapy protocols in animals before clinical trials are approved for humans. Because of the knowledge accumulated about the molecular biology of the factor IX gene most of the early work in gene transfer for haemophilia has been performed using the factor IX gene. In the following section a non-viral gene delivery approach (plasmid-mediated) and viral delivery approaches in muscle are discussed.

viable approach, the efficiency of gene transfer is low. Therefore, although expression was demonstrable, plasmidmediated gene transfer was short lived due to the induction of the host immune response. New evidence suggested that although plasmid mediated-gene transfer is very inefficient for expression of a secretable protein, its immunostimulatory adjuvant effects are highly desirable in various vaccine strategies against infectious and malignant diseases.

B. Viral delivery approaches 1. Retroviral Many of the initial viral vectors tested for a gene therapy approach in haemophilia B were retroviral. This involved an ex vivo approach in transfecting cell lines and subsequently transplanting them back into rodent animal models. Some early success was achieved transfecting murine myoblasts with a retroviral construct expressing factor IX, with factor IX being expressed for up to six months, although at low levels (Dai et al, 1992, Yao et al, 1992). This was presumed to arise from inefficient expression cassettes (Dai et al, 1992) and a cross species specific transgene evoking a host immune response. Studies were then performed in large animal models but the results were less successful with only short-term expression at levels that were clinically insignificant (Verma et al, 1994). Although some limited success was derived from the use of retroviral vectors there were still many problems to be resolved linked to these vectors. These related to low viral copy numbers, requirement for target cell division, size limitation of the transferred gene, and the concern about potential long term safety of these vectors with respect to their tropism, random integration and oncogenic potential. Because of these early problems with retroviruses investigators began working with adenoviruses; some of the problems associated with adenoviruses were discussed above.

A. Non viral gene delivery: plasmidmediated approach Some of the early in vivo attempts using muscle as a target cell in our laboratory consisted of direct injection of a plasmid construct encoding the factor VII gene (this was chosen at the time because of availability and ease of manipulation in cloning and expression). The engineered construct consisted of the FVII cDNA under the control of a myosin heavy chain promoter from which negative regulatory elements were removed. A myosin enhancer was also included and this has later been shown to increase the level of expression several fold. The product of the internal gene was detected systemically and shown to have biological activity (at day 4) although was short lived and disappeared by day 7 (Miller et al, 1995). Subsequent follow up experiments revealed that although factor VII expression was detectable at a tissue level, the muscle sections revealed dense inflammatory infiltrates (consisting of CD4 cells, CD8 cells and macrophages) around the areas of tissue expression (Fields et al, 1998a). An antibody was also detected in the serum of the injected animals to the human factor VII antigen. Similar experiments were performed following injection of a plasmid containing the human Factor IX cDNA into mouse muscle and again this demonstrated little plasma elevation of systemic factor antigen but the presence of an antibody isotype IgG2a to the transgene. This would be in keeping with TH1 driven cell response occurring in the setting of plasmid mediated gene transfer for a secretable protein (Fields et al, 1998b). These experiments illustrated the importance of selecting a species-specific transgene in a gene therapy setting. These experiments also demonstrated that although the concept of in vivo gene therapy for haemophilia using muscle is a

2. Adeno-associated virus (AAV) vectors: a major recent advancement in gene therapy of haemophilia Recently, AAV vectors have shown great promise in a gene therapy setting of haemophilia using muscle as a target tissue both in small and large animal models of haemophilia B. Several studies have documented that AAV vectors can direct persistent expression of reporter genes in muscle fibres of immune competent animals (Xiao et al, 1996, Kessler et al, 1996, Fisher et al, 1997). One early report documented expression of therapeutic levels of erythropoietin following intramuscular injection of an AAV vector expressing erythropoietin (Xiao et al, 1996). AAV vectors have certain advantages that make them particularly attractive for muscle-directed gene therapy. These include their relative non-pathogenicity in normal individuals (up to 80% of humans are infected with parvovirus), their ability to infect non-dividing cells and a 238

Gene Therapy and Molecular Biology Vol 3, page 239 patient with Becker's muscular dystrophy. N e w E n g . J . Med. 333, 732-733

broad range of host recipients (primates, canine and murine models). Recombinant AAV vectors contain the engineered expression cassette for the transgene flanked only by the inverted terminal repeats and is therefore devoid of any viral coding sequences. Transduction of muscle with recombinant AAV is very efficient, and expression is stable and long-lived without eliciting the cellular immune responses characteristically seen in muscle after adenovirusmediated gene transduction (Xiao et al, 1996; Snyder et al, 1997). Various ex vivo approaches have targeted muscle in the past for factor IX gene expression; it is known that factor IX produced in myotubes in vitro is biologically active (Dhawan et al, 1991). With the use of AAV vectors it is now possible in an in vivo gene therapy setting to transduce muscle in a highly efficient way. Stable expression of therapeutic plasma levels of human factor IX (1 year) has been demonstrated after intramuscular injection of AAV-factor IX to Rag-1 immunodeficient mice; levels of 4-7% (200-350 ng/ml FIX Ag) of normal levels of FIX in a human being were attained (Herzog et al, 1997a). When this experiment was repeated in an immunocompetent mouse, no factor IX could be detected in mouse plasma due to the presence of an antibody to the human transgene. When this experiment was scaled up to a canine haemophilia B model a similar result was obtained (Monahan et al, 1998); furthermore data from the canine model indicated that antibody formation against the transgene could be avoided if species specific transgene boundaries are not transgressed (Herzog et al, 1997b). The results achieved so far with these vectors are very encouraging; optimisation of these gene vectors has been achieved in rodent models resulting in persistent expression of therapeutic plasma levels of clotting factor IX with reduced or absent cellular immune responses against the transduced muscle cells. Scale up of these approaches to larger animal models of haemophilia will form the basis for future human clinical trials.

Chen HH. Mack LM. Kelly R. Ontell M. Kochanek S. Clemens PR. (1 9 9 7 ) Persistence in muscle of an adenoviral vector that lacks all viral genes. P r o c N a t l A c a d S c i U S A 94, 1645-50 Clemens PR. Kochanek S. Sunada Y. Chan S. Chen HH. Campbell KP, Caskey CT. (1 9 9 6 ) In vivo muscle gene transfer of full-length dystrophin with an adenoviral vector that lacks all viral genes. Gene Therapy. 3, 965-72 Coleman ME DeMayo F, Yin KC Lee HM Geske R Montgomery C Schwartz RJ. (1 9 9 5 ) Myofenic vector expression of insulin-like growth factor I stimulates muscle cell differentiation and myofiber hypertrophy in transgenic mice. J . B i o l . C h e m . 270 12109-12116. Corrado K, Wills PL, Chamberlain JS. (1 9 9 4 ) Deletion analysis of the dystrophin-actin binding domain. FEBS L e t t . 344, 255-260. Cox GA, Cole NM, Matsumura K, Phelps SF, Hauschka SD, Campbell KP, Faulkner JA, Chamberlain JS. (1 9 9 3 ) Overexpression of dystrophin in transgenic mdx mice eliminates dystrophic symptoms without toxicity. Nature 364, 725-729. Dai Y, Roman M. Naviaux RK, Verma IM. (1 9 9 2 ) Gene therpy via primary myoblasts , longterm expression of factor IX protein following transplantation in vivo. Proc Natl Acad Sci USA 89, 10892-5 Dai Y, Schwarz EM, Gu D, Zhang WW, Verma I. (1 9 9 5 ) Cellular and humoural immune responses to adenoviral vectors containing factor IX gene. Tolerization of Factor IX and vector antigens allows for long-term expression. Proc Natl Acad Sci USA92, 401-1405 Davis HL. Demeneix BA. Quantin B. Coulombe J. Whalen RG. (1 9 9 3 ) Plasmid DNA is superior to viral vectors for direct gene transfer into adult mouse skeletal muscle. Hum.Gene. Ther. 4, 733-40 Dhawan J, Pan LC, Pavlath GK, Travis MA. Lanctot AM, Blau HM. (1 9 9 1 ) Systemic delivery of human growth hormone by injection of genetically engineered myoblasts. S c i e n c e 254, 1509-1512.

References Ayling CM, Moreland BH, Zanelli JM, Schulster D (1 9 8 9 ) Human growth hormone treatment of hypophysectomized rats increases the proportion of type-1 fibres in skeletal muscle. J Endocrinol 123, 429-435.

Draghia-Akli R, Li X, Schwartz RJ. (1 9 9 7 ). Enhanced growth by ectopic expression of growth hormone releasing hormone using an injectable myogenic vector. Nature B i o t e c h n o l . 15, 1285-1289.

Baker J, Liu JP, Robertson EJ, Efstratiadis A. (1 9 9 3 ) Role of insulin-like growth factors in embryonic and postnatal growth. C e l l 75, 73-82

England, S., Nicholson, L.V.B., Johnson, M.A., Forrest, S.M., Bundy, S., Zubrzycka-Gaarn, E., Love, D.R. and Davies, K.E. (1 9 9 0 ) Very mild muscular dystrophy associated with the deletion of 46% of dystrophin. Nature 343, 180-182.

Bergsman DJ, Grichnik JM, Gossett IM, Schwartz RJ. (1 9 8 6 ) Delimitation and characterization of cis-acting of DNA sequences required for the regulated expression and transcriptional control of the chicken skeletal !-actin gene. M o l . C e l l B i o l . 6, 2452-2475.

Fabbrizio E. Pons F Robert A Hugon G Bonet-Kerrache A Momet D. (1 9 9 4 ) The dystrophin superfamily, variability and complexity. J . Mucle R e s . C e l l M o t i l . 15 595-606

Brahm H, Piehl-Aulin K, Saltin B, Ljunghall S. (1 9 9 7 ) Net fluxes over working thigh of hormones, growth factors and biomarkers of bone metabolishm during lasting dynamic exercise. C a l c i f i e d T i s s u e 60, 175-89.

Fassati A. Wells DJ. Sgro Serpente PA. Walsh FS. Brown SC. Strong PN. Dickson G. (1 9 9 7 ) Genetic correction of dystrophin deficiency and skeletal muscle remodeling in adult MDX mouse via transplantation of retroviral producer cells. J . C l i n . I n v e s t . 100, 620-8

Britter R.E. et al, (1 9 9 5 ) Serum antibodies to the deleted dystophin sequence after cardiac transplantaion in a

239

Goldspink et al: Gene transfer into muscle for the treatment of muscular dystrophy and haemophilia systemic delivery of a therapeutic protein. Proc Natl Acad Sci USA 93, 14082-14087

Fields P, Herzog R, Arruda V, Hagstrom N, Pasi KJ , High K (1 9 9 8 b ) Immune response to a secretable protein following gene transfer into muscle by plasmid based, adenoviral or adeno-associated viral vectors. B l o o d 92, 147a.

Kochanek S. Clemens PR. Mitani K. Chen HH. Chan S. Caskey CT. (1 9 9 6 ) A new adenoviral vector, Replacement of all viral coding sequences with 28 kb of DNA independently expressing both full-length dystrophin and beta-galactosidase. Proc Natl Acad Sci USA 93, 5731-6,

Fields P, Murdoch P, Bayele H, Perry D, Wells D, Watt D, Goldspink G, Pasi J. (1 9 9 8 a ) Immune responses to direct plasmid injection in muscle , implications for transgene expression. K e y s t o n e S y m p o s i a , Molecular and cellular biology of gene therapy .

Li Q. Kay MA. Finegold M. Stratford-Perricaudet LD. Woo SL. (1 9 9 3 ) Assessment of recombinant adenoviral vectors for hepatic gene therapy. Hum. Gene Ther. 4, 403-409

Fisher K, Jooss K, Alston J, Yang Y, Haecker S, High K, Pathak R, Raper S , Wilson JM. (1 9 9 7 ) Recombinant Adeno-associated virus for muscle directed gene therapy. Nature Med 3, 3227-3234.

Michou A, Santoro L, Christ M, Jiullard V, Pavirani A, Mehtali M. (1 9 9 7 ) Adenovirus-mediated gene transfer influence of transgene, mouse strain and type of immune response on persistence of transgene expression. Gene Therapy 4, 473-482.

Fisher KJ, Choi H, Burda J, Chen S, Wilson JM. (1 9 9 6 ) Recombinant adenoviruses deleted of all viral genes for gene therapy of cystic fibrosis. V i r o l o g y 217, 11-22

Miller G, Steinbrecher R.A., Murdock P.J., Tuddenham E.G.D., Lee C.A., Pasi K.J. and Goldspink G. (1 9 9 5 ) Expression of factor VII by muscle cells in vitro and in vivo following direct gene transfer, modelling gene therapy for haemophilia. Gene Therapy 2, 736-742

Floyd SS Jr. Clemens PR. Ontell MR. Kochanek S. Day CS. Yang J. Hauschka SD. Balkir L. Morgan J. Moreland MS. Feero GW. Epperly M. Huard J. (1 9 9 8 ) Ex vivo gene transfer using adenovirus-mediated full-length dystrophin delivery to dystrophic muscles. Gene Therapy 5, 19-30

Miller RG. Sharma KR. Pavlath GK. Gussoni E. Mynhier M. Lanctot AM. Greco CM. Steinman L. Blau HM. (1 9 9 7 ) Myoblast implantation in Duchenne muscular dystrophy, the San Francisco study. Muscle & Nerve. 20, 469-478

Goldspink G & Booth F (1 9 9 2 ) General Remarks for Editorial Issue - Mechanical signals and gene expression in muscle. A m J P h y s i o l 262, R327-R328.

Milner, R.E., Busaan, J.L., Wang, J.H., Michalak, M. (1 9 9 3 ) Phosphorylation of dystrophin. The carboxy-terminal region of dystrophin is a substrate for in vitro phosphorylation by p34cdc2 protein kinase. J . B i o l . C h e m . 268, 21901-21905.

Goldspink G, Yang SY, Skarli M and Vrbova G (1 9 9 6 ) Local growth regulation is associated with an isoform of IGF-1 that is expressed in normal muscles but not in dystrophic muscles when subjected to stretch J . P h y s i o l 496P 10. Hansen, E., Goldspink, G., Butterworth, P.W. and Chang, KC. (1 9 9 1 ) Strong expression of some mammalian gene constructs in fish muscle following direct gene transfer. FEBS Letts. 290, 73-76

Monahan PE, Samulski RJ, Tazelaar J, Xiao X, Nichols TC, Bellinger DA , Read MS. (1 9 9 8 ) Direct intramuscular injection with recombinant AAV vectors results in sustained expression in a dog model of haemophilia. Gene Therapy 5, 40-49.

Hardy S, Kitamura , Harris T, Dai Y , Phipps L. (1 9 9 7 ) Construction of adenovirus through Cre Lox Recombination. J Virol . 71, 1842-1849.

Novo F.J. Kruszewski A.,MacDermot K.D.'Goldspink G and Gorecki, D.C. (1 9 9 5 ) Editing of human alphagalactosidase RNA resulting in a pyrimidine to purine conversion. N uc le ic A c ids R e s 23, 2636-2640.

Herzog R, Hagstrom J, Kung Z, Tai SJ, Wilson JM, Fisher K, High KA. (1 9 9 7 a ) Stable gene transfer and expresssion of human blood coagulation factor IX after intramuscular injection of recombinant adeno-associated virus. Proc Natl Acad Sci USA 94, 5804-5809

Palmer TD, Thompson AR, Miller AD. (1 9 8 9 ) Production of human factor IX in animals by genetically modified skin fibroblasts , Potential therapy for Haemophilia B. B l o o d 73, 438-445.

Herzog R, Hagstrom JN, Kung SZ, Yang EY, Couto LB, Kurtzmann GJ, High KA. (1 9 9 7 b ) Absence of antibodies against factor IX following IM injection of an AAV vector encoding a species specific transgene. B l o o d 90 , 1957

Powell-Braxton L et al, (1 9 9 3 ) IGF-1 is required for normal embryonic growth in mice. G e n e s D e v . 7, 2609-2617. Ragot T. Stratford-Perricaudet LD. Vincent N. Chafey P. Vigne E. Gilgenkrantz H. Couton D. Briand P. Kaplan JC. Kahn A. et al. (1 9 9 4 ) Adenovirus-mediated transfer of a human dystrophin gene to skeletal muscle of mdx mouse. Gene Therapy. 1 Suppl 1, S53-S54.

Huard J Krisky D Oligino T Marconi P Day CS Watkins SC Glorioso JC. (1 9 9 7 ) Gene transfer to muscle using herpes simplex-based vectors. Neuromusc. D i s o r d . 7 299313. Kay MA , Holterman AX, Meuse L, Gown A, Linsley P, Wilson CB. (1 9 9 5 ) Long term hepatic adenovirus mediated gene expression in mice following CTLAIg administration. Nat Genet 11, 191-197

Rando TA. Blau HM. (1 9 9 4 ) Primary mouse myoblast purification, characterization, and transplantation for cell-mediated gene therapy. J . C e l l B i o l . 125, 1275-1287,

Keir SD, Mitchell WJ, Feldman L, Martin JR. (1 9 9 5 ) Targeting and gene expression in spinal cord motoneurones following intramuscular inoculation of an HSV-1 vector. J . N e u r o v i r o l . 1, 259267.

Rando TA. Pavlath GK. Blau HM. (1 9 9 5 ) The fate of myoblasts following transplantation into mature muscle. E x p C e l l R e s 220, 383-389 Ribota MGY et al. (1 9 9 7 ) Prevention of motoneuron death by adenovirus-mediated neurotrophic factors. J . Neurosc R e s . 48, 281-285.

Kessler PD, Podsakoff GM, Chen X, McQiuston SA, Colosi PC, Kurtzmann G, Byrne B. (1 9 9 6 ) Gene delivery to skeletal muscle results in sustained expression and

240

Gene Therapy and Molecular Biology Vol 3, page 241 phenotype in transgenic mdx mice. Hum Mol Genet 1, 35-40.

Rudman DM, Kutner MH, Rogers CM, Lubin MF, Fleming GA, Brain RP. (1 9 8 1 ) Impaired growth hormone secretin in the adult population. J C l i n I n v 67, 1361-1369.

Wolff, J.A., Ludtke, J.L., Acsadi, G., Williams P. and Jani, A. (1 9 9 2 ) Long-term persistence of plasmid DNA and foreign gene expression in mouse muscle. Hum M o l Genet 1, 363-369.

Sato Y, Roman M, Tighe H, Lee D, Corr M, Nguyen M, Silverman G, Lotz M, Carson D, Raz E. (1 9 9 6 ) Immunostimulatory DNA sequences necessary for effective intradermal gene immunization. S c i e n c e 273, 352-354

Wolff, J.A., Malone, R.W., Williams, P., Chong, W., Acasadi, G., Jani, A. and Felgner. P.L. (1 9 9 0 ) Direct gene transfer into mouse muscle in vivo. S c i e n c e 247, 1465-1468.

Shemanko, C.S., Sanghera, J.S., Milner, R.E., Pelech, S., Michalak, M. (1 9 9 5 ) Phosphorylation of the carboxyl terminal region of dystrophin by mitogen-activated protein (MAP) kinase. M o l . C e l l . B i o c h e m . 152, 6370.

Wright MJ Rosenthal E Stewart L Withtman LM Miller AD Latchman DS Marber MS. (1 9 9 8 ) beta-galactosidase staining following intracoronary infusion of cationic liposomes in the in vivo rabbit heart is produced by microinfarcion rather than effective gene transfer, a cautionary tale. Gene Therapy 5, 301-308.

Skarli M. Kiri A, Vrbova G, Lee CA, Goldspink G. (1 9 9 8 ) Myosin regulatory elements as vectors for gene transfer by intramuscular injection. Gene Therapy 5, 514-520. Snyder R, Spratt S, Lgarde C, Bohl D, Kasper B, Sloan B, Cohen LK, Danos O. (1 9 9 7 ) Efficient and stable adenoassociated virus mediated transduction in the skeletal muscle of adult immunecompetent mice. Hum Gene Ther 8, 1891-1900.

Xiao X, Li J, Samulski RJ. (1 9 9 6 ) Efficient long-term gene transfer into muscle tissue of immunecompetent mice by adeno-associated virus vector. J Virol 70, 8098 -8108 Xu, H., Christmas, P., Wu, X-P., Wewer, U.M., Engvall, E. (1 9 9 4 ) Defective muscle basement membrane and lack of M-laminin in the dystrophic dy/dy mouse. P r o c . N a t l . A c a d . S c i . 91 5572-5576.

Stratford-Perricaudet LD. Levrero M. Chasse JF. Perricaudet M. Briand P. (1 9 9 0 ) Evaluation of the transfer and expression in mice of an enzyme-encoding gene using a human adenovirus vector. Hum Gene Ther 1, 241-56,

Yang SY, Alnaqeeb M, Simpson H and Goldspink G. (1 9 9 6 ) Cloning and characterisation of an IGF-1 isoform expressed in skeletal muscle subjected to stretch. J . M u s c l e R e s . C e l l M o t i l . 17, 487-495.

Sunada, Y., Bernier, S.M., Kozak, C.A., Yamada, Y., Campbell, K.P. (1 9 9 4 ) Deficiency of merosin in dystrophic dy mice and genetic linkage of laminin M chain gene to dy locus. J . B i o l . C h e m . 269, 1372913732.

Yang Y, Ertl HC, Wilson JM. (1 9 9 4 ) MHC class I restricted cytotoxic T lymphocytes to viral antigens create barriers to lung directed gene therapy with recombinant adenoviruses. Immunity 1, 433-442

Svenson E, Black H, Dugger D, Tripathy S, Goldwasser E, Hao Z, Chu L, Leiden J. (1 9 9 7 ) Long term erythropoietin expression in rodents and non human primates following intramuscular injection of a replication defective adenoviral vector. Hum Gene Ther 8, 797-1806.

Yang Y, Nunes FA, Berensci K, Furth EE, Gonzzol E, Wilson JM. ( 1 9 9 4 ) Cellular immunity to viral antigens limits E1 deleted adenovirus for gene therapy. Proc Natl Acad S c i USA 91, 4407-4411.

Tighe H, Corr M, Roman M,Raz E. (1 9 9 7 ) Gene vaccination, plasmid DNA is more than just a blueprint I m m u n o l . Today 19, 89-97

Yao SN , Kuracchi K. (1 9 9 2 ) Expression of human factor IX in mice after injection of genetically modified myoblasts. Proc Natl Acad Sci USA 89, 3357-3361

Tinsley JM, Potter AC, Phelps SR, Fisher R, Tricket JI. Davies KE. (1 9 9 6 ) Amelioration of the dystrophic phenotype of mdx mice using a truncated utrophin transgene. Nature 384, 349-353.

Yao SN, Wilson JM, Nabel EG, Kurachi K, Hachiya HL, Kurachi K. (1 9 9 1 ) Expression of human factor IX in rat capillary endothelial cells, Toward somatic gene therapy for Haemophilia B. Proc Natl Acad S c i USA 88, 8101-8105.

Verma IM, Dai Y. (1 9 9 4 ) Progress in gene therapy. Proc XXI Int Congr world federation of h a e m o p h i l i a , 15.

Zdanowitz, M.M., Moyse, J., Wingert Zahn, M.A., O'Connor, M., Teichberg, S., Slonim, A.E (1 9 9 5 ) Effect of insulinlike growth factor I in murine muscular dystrophy. E n d o c r i n o l o g y 136, 4880-4886.

Wang, JM Zheng H Blaivas M Kurachi K. (1 9 9 7 ) Persistent systemic production of human Factor IX in mice by skeletal myoblast-mediated gene transfer, feasibility of repeat application to obtain therapeutic levels. B l o o d 90 1075-1082 Wells DJ. Maule J. McMahon J. Mitchell R. Damien E. Poole A. Wells KE (1 9 9 8 ) Evaluation of plasmid DNA for in vivo gene therapy, factors affecting the number of transfected fibers. J . P h a r m . S c i . 87, 763-8. Wells, D. J. and Goldspink, G. (1 9 9 2 ) Age and sex influence expression of plasmid DNA directly injected into mouse skeletal muscle. FEBS Letts 306, 203-205. Wells, K.E., Wells, Walsh FS, Goldspink G, Love DR, ChanThomas, P., Dunckley M, Piper T., Dickson G. (1 9 9 2 ) Human dystrophin expression corrects the myopathic

241

Gene Therapy and Molecular Biology Vol 3, page 243 Gene Ther Mol Biol Vol 3, 243-248. August 1999.

Gene therapy for arthritis Review Article

Sherry Thornton and Raphael Hirsch Department of Rheumatology, Children's Hospital Research Foundation, Cincinnati, OH 45229 _________________________________________________________________________________________________ Correspondence: Raphael Hirsch, MD, Division of Rheumatology, Children’s Hospital Medical Center, Pavilion Building 2129, 3333 Burnet Avenue, Cincinnati, Ohio 45229. Tel: (513) 636-8027; Fax: (513) 636-4116; E-mail: hirschr@chmcc.org A b b r e v i a t i o n s : RA, rheumatoid arthritis; CIA, collagen-induced arthritis Key words: Gene therapy, in vivo gene transfer, rheumatoid arthritis, inflammation, gene expression Received: 10 October 1998; accepted: 15 October 1998

Summary Gene therapy for the treatment of arthritis is developing rapidly. The ability to deliver genes to local sites of inflammation decreases the possibility of systemic side effects, making arthritis a good candidate for gene therapy. Animal models of arthritis provide a means o f testing gene transfer strategies. Several issues still need to be addressed including which genes to deliver, how to deliver these genes, and how to regulate gene expression in vivo.

short half-life of these molecules necessitates frequent readministration. These naturally-produced molecules have the potential to be delivered via gene transfer, which may allow for a reduction in the requirement for frequent readministration of the drug.

I. Introduction Arthritis is a major health problem among working age people in the United States, with greater than 2 million men and greater than 3 million women reporting activity limitation (Yelin, 1992). The most prevalent form of arthritis, rheumatoid arthritis (RA), affects approximately 0.8% of most populations (Koopman, 1997). Arthritic symptoms have been reported for 55% of individuals !70 years of age. Among these elderly affected individuals, 3/4 were limited in physical actions and 1/3 were limited in daily living (Yelin, 1992). Therefore, the chronic symptoms of arthritis impact significantly on the quality of life.

II. Gene delivery strategies A. In vivo gene transfer Although RA often affects local joints, immunological responses observed in patients with RA demonstrate the presence of systemic components of arthritis. Therefore, the treatment of RA can be approached with either systemic or local therapies.

Until recently, the treatment of arthritis, particularly RA, involved the use of non-specific anti-inflammatory agents, such as nonsteroidal anti-inflammatory agents (NSAIDs), steroids and methotrexate. These anti-rheumatic drugs allow relief of many symptoms of the disease, but can exhibit harmful side effects and do not necessarily alter the natural course of the disease (Koopman, 1997). The cause of RA is unknown, and the mechanism of action of many of the drugs used to treat RA remains unknown. However, recent studies of molecular and cellular mechanisms that govern the pathophysiology of arthritis has led to the discovery of therapeutic biological agents that offer greater specificity in the treatment of arthritis. These biological agents are currently being delivered primarily at the protein level. The

Systemic gene delivery, such as by i.v. administration, has been demonstrated in animal models. These studies have examined mainly the short-term effects on arthritis and not the long-term systemic effects, including potential toxicity. Therefore, the delivery of genes directly into the bloodstream requires further investigation. For the treatment of arthritis, local gene delivery is an attractive therapeutic option. Since the target of arthritis is the synovium or cells contained within the affected joint, local therapies involving injection directly into the affected joint space, could potentially provide delivery of genes to a limited space and reduce toxic systemic effects. Local 243

Thornton and Hirsch: Gene therapy for arthritis injection of adenovirus encoding a reporter gene to inflamed joints of monkeys with CIA results in expression that is contained to the synovium and is not present in other tissue samples, indicating that gene transfer to synovial tissue may be safe in primates and may exhibit an ideal biodistribution (Goossens, 1997). However, local administration of adenovirus in other animal models has effects on distal joints, suggesting that local delivery of gene products may produce systemic effects that must be analyzed appropriately (Bakker, 1997; Ghivizzani, 1998; May, 1998).

B. Ex vivo gene transfer The ex vivo approach involves removal of synovial cells, culturing and infection of these cells with the appropriate virus, usually retrovirus, and subsequently returning the cells to the joint space. This procedure, while cumbersome and expensive, also provides for analysis and selection of the genetically altered cells before returning them to the joint space.

III. Gene transfer vectors A. Viral vectors Various gene therapy vectors have been utilized that can be grouped mainly into viral and non viral vectors. Since viruses naturally deliver genetic material to cells, the use of viruses is the basis for most gene delivery systems. Viruses are the most widely used means of delivering genes in arthritic animal models (Nita, 1996). Among viral vectors, retroviral and adenoviral vectors are primarily used for gene delivery, and both have particular characteristics that make them suitable for the delivery of genes in the treatment of arthritis. Adenoviruses are easily produced at high titres and infect non dividing synovial cells. Adenovirus delivery of the "galactosidase gene intra-articularly demonstrates that adenovirus can infect non dividing synovial cells and "galactosidase expression can last up to 21 days (Sawchuck, 1996). Intravenous administration of adenovirus encoding vIL-10 also indicates that vIL-10 can be detected up to 7 days after injection and can inhibit CIA (Apparailly, 1998; Ma, 1998). However, adenoviral vectors induce an inflammatory response, which may come from the viral proteins being expressed or the transgene product itself. In terms of gene expression, adenoviral encoded proteins are normally shortlived, which is thought to be due to this inflammatory process. Retroviruses are produced at relatively low titres, infect only dividing cells, and incorporate into the host genome. Retroviral vectors are primarily used ex vivo to transfect cultured synovial cells that divide, allowing for retroviral infection. Recent studies indicate that stimulation of cells 244

with TNF# in vitro allows retroviral transduction of cells (Jorgensen, 1997), and that inflamed synovium, which produces TNF#, may be more susceptible to retroviral uptake (Ghivizzani, 1997). These findings suggest that retroviral vectors might be delivered intra-articularly to target inflamed synovium. Long term gene expression is desirable for any gene therapy vector. Incorporation of the retrovirus into the host genome allows for long term gene expression; however, with this incorporation the risk of insertional mutagenesis exists. Unlike adenovirus-infected cells, retrovirus-infected cells have not been a target for destruction by the immune system (Evans, 1997). Another kind of viral vector, the lentivirus, is derived from retroviruses, but has the capability to infect non dividing cells (Naldini, 1996). This virus may have promise for targeting non dividing synovial cells in the treatment of arthritis.

B. Non viral vectors Various methods of non viral gene delivery include liposomal delivery, direct plasmid injection, and gene gun delivery. Non viral DNA delivery offers low toxicity, but most methods available are very inefficient at transfection of synovial cells. Gene transfer to rabbit and rat synovial cells by direct plasmid injection demonstrates that plasmid uptake resembles non specific endocytosis (Yovandich, 1995). The transient expression of the reporter plasmid corresponds with the degradation of plasmid DNA, indicating that intraarticular injection of plasmid DNA results in short-term gene expression. Long term gene expression of non viral plasmid DNA vectors has been achieved in muscle tissue (Tripathy, 1996). Expression of certain genes in skeletal muscle via plasmid injection has systemic effects on the immune system (Raz, 1993). Plasmid DNA encoding TGF-" delivered into thigh muscle of rats with streptococcal cell wall induced arthritis, suppressed the chronic disease and virtually eliminated subsequent inflammation and arthritis (Song, 1998). Therefore, intra-muscular injection of plasmid may be a less toxic way to systemically deliver anti-inflammatory products for the treatment of arthritis.

IV. Candidate genes Analysis of cytokine expression between arthritic and non-arthritic joints indicates an increase in a number of cytokines in arthritic joints. This information has led to two main strategies to reduce inflammation in arthritic joints. The first approach involves the use of natural inhibitors of pro-inflammatory cytokines. The second approach, immune deviation, involves administration of cytokines that naturally down regulate pro-inflammatory cytokine synthesis.

Gene Therapy and Molecular Biology Vol 3, page 245

A. Natural inhibitors of inflammatory cytokines TNF-# and IL-1 are major regulators of inflammation in arthritic joints. Inhibitors of these two cytokines reduce arthritis in both animal models of arthritis and in ongoing human trials. In collagen-induced arthritis (CIA), an animal model of RA, treatment with antibody to TNF# (Joosten, 1994; Thorbecke, 1992) or IL-1" (Geiger, 1993; Joosten, 1994; Joosten, 1996; Thorbecke, 1992) reduced disease severity. In human trials administration of cA2, an antibody specific for TNF#, dramatically suppressed symptoms of disease, although this effect required continual treatment (Elliott, 1994; Elliott, 1994; Elliott, 1993). 1. Interleukin-1 receptor antagonist (IL-1Ra) IL-1Ra regulates IL-1 activity in vivo by binding to IL-1 receptors. IL-1Ra, while inhibiting IL-1 from binding, itself does not stimulate activity through the IL-1 receptor. However, a 10-100 fold excess of IL-1Ra over IL-1 is necessary to block the effects of IL-1 activity in vivo (Dinarello, 1991; Hirsch, 1996). Continuous administration of high levels of IL-1Ra can block CIA (Joosten, 1996; Wooley, 1993). Transgenic mice overproducing IL-1Ra exhibit a reduction in the incidence and severity of CIA, and mice lacking IL-1Ra have a significantly earlier onset of CIA (Ma, 1998). Human trials aimed at determining the efficacy of administration of recombinant human IL-1Ra are still being assessed (Campion, 1996). These studies indicate that IL-1Ra is a good candidate gene for reduction of arthritis. Several animal models of arthritis have shown benefits after IL-1Ra gene delivery. Expression of human IL-1Ra in rabbits with antigen induced arthritis changed the course of arthritis and suppressed the effects of IL-1 (Bandara, 1993; Otani, 1996). Ex vivo retroviral transduction of primary synoviocytes grafted to ankle joints in rats with bacterial cell wall-induced arthritis showed a significantly suppressed severity of recurrence of arthritis and attenuated erosion of cartilage and bone (Makarov, 1996). Treatment of mice with CIA by ex vivo transduction of NIH/3T3 fibroblasts with retrovirus expressing human IL-1Ra prevented CIA in injected knee joints and the “draining” paws (Bakker, 1997). Rabbits treated with adenovirus expressing human IL-1Ra had both in vitro and in vivo effects, including inhibition of IL-1 activity and inhibition of induced prostaglandin E2 synthesis. Therefore, IL-1Ra shows great promise as a gene to deliver for the treatment of arthritis. Results from many of the above studies using IL-1Ra led the way to the first human gene therapy trials for RA which began in 1996. Using an ex vivo approach, cells removed from patients joints are transfected with retroviral vectors expressing IL-1Ra (McCarthy, 1996 and reviewed in Evan, 1998). The cells are tested for both IL-1Ra expression and for the presence of endotoxin and other agents. IL245

1Ra$transduced and untransduced cells are injected back into the joints, and removed at the time of joint replacement to determine whether expression of IL-1Ra was achieved. This human trial is the first step toward assessment of local gene therapy for RA. 2. Soluble TNF receptor (sTNFR) sTNFR is a natural inhibitor of TNF activity. Two receptors for TNF have been isolated, p55 and p75, that bind both TNF# and TNF" (Loetscher, 1990; Smith, 1990). Soluble forms of these receptors, which are extracellular and contain ligand binding domains, inhibit TNF activity (Mohler, 1993). The sTNFR administered in clinical trials of RA is comprised of the soluble portion of the p75 cell surface receptor fused to the Fc portion of human IgG1 (sTNFRFc). The IgG1 portion prolongs the half-life of the molecule (Mohler, 1993). sTNFRFc inhibits both CIA (Mori, 1996; Williams, 1995; Wooley, 1993) and can dramatically suppress the arthitic symptoms of RA, although again, continuous administration is required (Moreland, 1997). Recently, sTNFR marketed under the trade name Enbrel (Immunex Corporation) has received approval by the FDA for the treatment of RA as a subcutaneous injection administered twice weekly. Gene delivery of sTNFR in animal models has inhibitory effects on arthritis. In rats with CIA, systemic delivery of an adenoviral vector encoding sTNFR prior to or following the onset of arthritis, suppressed CIA. However, intra-articular administration of this vector induced an adenoviral synovitis, which was not overcome even by the expression of the sTNFR (Le, 1997). The transfer of CIA to SCID mice can also be inhibited by transducing DBA/1 spleen cells with retrovirus encoding sTNFR (Chernajovsky, 1995). In other gene delivery studies, Ghivizanni, et al., injected adenoviruses encoding either IL-1Ra or sTNFR, both separately and in combination, into rabbit’s knees (Ghivizzani, 1998). IL-1Ra reduced cartilage matrix degradation and white blood cell infiltration into the joint space. sTNFR by itself was not as effective as IL-1Ra. However, treatment with both IL-1Ra and sTNFR showed greater inhibition of white blood cell infiltration and cartilage breakdown with a considerable reduction in synovitis. In addtion, with both reagents, effects on contralateral control knees were also observed, suggesting that local intra-articular treatment may be used to treat systemic polyarticular arthritides.

B. Immune deviation An imbalance between the activities of Th1 and Th2 cells is thought to play a role in the pathophysiology of many autoimmune diseases, such as RA. Th1 cells secrete cytokines such as IL-2 and IFN-% , that normally mediate

Thornton and Hirsch: Gene therapy for arthritis pro-inflammatory immune responses, whereas Th2 cells secrete cytokines such as IL-4, lL-10, and IL-13 that can downregulate Th1 activity. Administration of IL-4, IL-10 and IL-13 proteins to CIA mice indicate that these cytokines can inhibit the disease process (Bessis, 1996; Hesse, 1996; Horsfall, 1997; Joosten, 1997; Tanaka, 1996; Walmsley, 1996). Another animal model using streptococcal cell wall fragments to induce arthritis in rats, also demonstrates that IL-4 administration can reduce pro-inflammatory cytokine production and can inhibit experimental arthritis (Allen, 1993). In human RA synovial cells, IL-4 and IL-10 also have inhibitory effects on pro-inflammatory cytokine production (Chomarat, 1995; Isomaki, 1996; Katsikas, 1994; van Roon, 1996). These studies indicate that the Th2 type cytokines IL-4, IL-10, IL13, which can inhibit pro-inflammatory cytokine production and the arthritic process in animal models, are good candidates for gene transfer. 1. Viral IL-10 (vIL-10) gene therapy vIL-10 is homologous to both mouse and human IL-10 and shares many of their immunosuppressive properties, but lacks their immunostimulatory properties (Go, 1990; MacNeil, 1990). Systemic administration of adenovirus encoding vIL-10 before the onset of CIA inhibited arthritis (Apparailly, 1998; Ma, 1998), but the effects were shortterm, probably due to the inflammatory response to the adenovirus. Local adminstration of vIL-10 in the footpad (Whalen, 1998) or intra-articularly into the knee (Ma, 1998) reduced the incidence of arthritis, indicating again that local gene expression can have systemic effects on disease. 2. Fas Ligand Other methods that eliminate proliferating synovial cells are also being investigated, even though the removal of synovium has not been a successful cure for arthritis. The transduction of synovial cells with adenovirus that expresses Fas ligand induced apoptosis of synovial cells producing Fas. Administration of the virus into inflamed joints ameliorated CIA in DBA/1 mice (Zhang, 1997).

V. Future gene therapy for arthritis Much progress has been made in recent years in the field of gene therapy for arthritis. Future efforts will be focused on determining which genes are the most promising for therapy, which vectors are the best for delivering these genes, and ultimately how to regulate expression of the genes being delivered.

References Allen JB, Wong HL, Costa GL, Bienkowski MJ, and Wahl SM (1 9 9 3 ). Suppression of monocyte function and differential regulation of IL-1 and IL-1ra by IL-4 contribute to resolution of experimental arthritis. J Immunol 151, 4344-4351. Apparailly F, Verwaerde C, Jacquet C, Auriault C, Sany J, and Jorgensen C (1 9 9 8 ). Adenovirus-mediated transfer of viral IL-10 gene inhibits murine collagen-induced arthritis. J . I m m u n o l . 160, 5213-5220. Bakker AC, Joosten LA, Arntz OJ, Helsen MM, Bendele AM, van de Loo FAJ, and van den Berg WB (1 9 9 7 ). Prevention of murine collagen-induced arthritis in the knee and ipsilateral paw by local expression of human interleukin-1 receptor antagonist protein in the knee. Arthritis Rheum 40, 893-900. Bandara G, Mueller GM, Galea-Lauri J, Tindal MH, Georgescu HI, Suchanek MK, Hung GL, Glorioso JC, Robbins PD, and Evans CH (1 9 9 3 ). Intra-articular expression of biologically active interleukin1-receptor antagonist protein by ex vivo transfer. P r o c . N a t l . A c a d . S c i . 90, 10764-10768. Bessis N, Boissier MC, Ferrara P, Blankenstein T, Fradelizi D, and Fournier C (1 9 9 6 ). Attenuation of collagen-induced arthritis in mice by treatment with vector cells engineered to secrete interleukin-13. J . I m m u n o l . 26, 2399-2403. Campion GV, Lebsack ME, Lookabaugh J, Gordon G, and Catalano M (1 9 9 6 ). Dose-range and dose-frequency study of recombinant human interleukin-1 receptor antagonist in patients with rheumatoid arthritis. The IL-1Ra Arthritis Study Group. Arthritis Rheum 39, 1092-1101. Chernajovsky Y, Adams G, Podhajcer OL, Mueller GM, Robbins PD, and Feldmann M (1 9 9 5 ). Inhibition of transfer of collagen-induced arthritis into SCID mice by ex vivo infection of spleen cells with retroviruses expressing soluble tumor necrosis factor receptor. Gene Ther 2, 731735. Chomarat P, Banchereau J, and Miossec P (1 9 9 5 ). Differential effects of interleukins 10 and 4 on the production of interleukin-6 by blood and synovium monocytes in rheumatoid arthritis. Arthritis Rheum. 38, 1046-1054. Dinarello CA (1 9 9 1 ). Interleukin-1 antagonism. B l o o d 77, 1627-1652.

and

interleukin-1

Elliott MJ, Maini RN, Feldmann M, Kalden JR, Antoni C, Smolen JS, Leeb B, Breedveld FC, Macfarlane JD, Bijl H, and et al. (1 9 9 4 ). Randomised double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha (cA2) versus placebo in rheumatoid arthritis. Lancet 344, 1105-1110. Elliott MJ, Maini RN, Feldmann M, Long-Fox A, Charles P, Bijl H, and Woody JN (1 9 9 4 ). Repeated therapy with monoclonal antibody to tumour necrosis factor alpha (cA2) in patients with rheumatoid arthritis. Lancet 344, 11251127. Elliott MJ, Maini RN, Feldmann M, Long-Fox A, Charles P, Katsikis P, Brennan FM, Walker J, Bijl H, Ghrayeb J, and et al. (1 9 9 3 ). Treatment of rheumatoid arthritis with chimeric

246

Gene Therapy and Molecular Biology Vol 3, page 247 monoclonal antibodies to tumor necrosis factor alpha. Arthritis Rheum 36, 1681-1690.

collagen-induced arthritis in DBA/1 mice. Arthritis Rheum. 39, 797-809.

Evans CH, and Robbins PD (1 9 9 7 ). Getting genes into human synovium [editorial; comment]. J R h e u m a t o l 24, 20612063.

Joosten LAB, Lubberts E, Durez P, Helsen MMA, Jacobs MJM, Goldman M, and van den Berg WB (1 9 9 7 ). Role of interleukin-4 and interleukin-10 in murine collagen-induced arthritis. Arthritis Rheum. 40, 249-260.

Evans CH, Whalen JD, Ghivizzani SC, and Robbins PD (1 9 9 8 ). Gene therapy in autoimmune diseases. A n n R h e u m D i s 57, 125-127. Geiger T, Towbin H, Cosenti-Vargas A, Zingel O, Arnold J, Rordorf C, Glatt M, and Vosbeck K (1 9 9 3 ). Neutralization of interleukin-1 beta activity in vivo with a monoclonal antibody alleviates collagen-induced arthritis in DBA/1 mice and prevents the associated acute-phase response. C l i n . E x p . R h e u . 11, 515-522. Ghivizzani SC, Lechman ER, Kang R, Tio C, Kolls J, Evans CH, and Robbins PD (1 9 9 8 ). Direct adenovirus-mediated gene transfer of interleukin 1 and tumor necrosis factor alpha soluble receptors to rabbit knees with experimental arthritis has local and distal anti-arthritic effects. Proc Natl Acad Sci USA 95, 4613-4618. Ghivizzani SC, Lechman ER, Tio C, Mule KM, Chada S, McCormack JE, Evans CH, and Robbins PD (1 9 9 7 ). Direct retrovirus-mediated gene transfer to the synovium of the rabbit knee: implications for arthritis gene therapy. Gene Ther 4, 977-982. Go NF, Castle BE, Barrett R, Kastelein R, Dang W, Mosmann TR, Moore KW, and Howard M (1 9 9 0 ). Interleukin 10, a novel B cell stimulatory factor: unresponsiveness of X chromosome-linked immunodeficiency B cells. J Exp Med 172, 1625-1631. Goossens P, Bout B, t'Hart B, Brok H, Breedveld FC, Valerio D, and Huizinga T (1 9 9 7 ). Possibility and safety of gene transfer to inflamed synovial tissue after intra-articular administration. Arthritis Rheum. S221. Hesse M, Bayrak S, and Mitchison A (1 9 9 6 ). Protective major histocompatibility complex genes and the role of interleukin-4 in collagen-induced arthritis. Eur J Immunol 26, 3234-3237. Hirsch E, Irikura VM, Paul SM, and Hirsh D (1 9 9 6 ). Functions of interleukin 1 receptor antagonist in gene knockout and overproducing mice. Proc Natl Acad S c i USA 93, 11008-11013. Horsfall AC, Butler DM, Marinova L, Warden PJ, Williams RO, Maini RN, and Feldmann M (1 9 9 7 ). Suppression of collagen-induced arthritis by continuous administration of IL-4. J . I m m u n o l . 159, 5687-5696. Isomaki P, Lukkainen R, Saario R, Toivanen P, and J. P (1 9 9 6 ). Interleukin-10 functions as an antiinflammatory cytokine in rheumatiod synovium. Arthritis Rheum. 39, 386-395. Joosten LAB, Helsen MMA, van de Loo FAJ, and van den Berg WB (1 9 9 4 ). Amelioration of established collagen-induced arthritis (CIA) with anti-IL-1. A g e n t s A c t i o n s S p e c i a l C o n f e r e n c e I s s u e 41, C174-C176. Joosten LAB, Helsen MMA, van de Loo FAJ, and van den Berg WB (1 9 9 6 ). Anticytokine treatment of established type II

247

Jorgensen C, Demoly P, Noel D, Mathieu M, Piechaczyc M, Gougat C, Bousquet J, and Sany J (1 9 9 7 ). Gene transfer to human rheumatoid synovial tissue engrafted in SCID mice [see comments]. J Rheumatol 24, 2076-2079. Katsikas PD, Chu C-Q, Brennan FM, Maini RN, and Feldmann M (1 9 9 4 ). Immunoregulatory role of interleukin 10 in rheumatoid arthritis. J . E x p . M e d . 179, 1517-1527. Koopman W. (1 9 9 7 ). Arthritis and Allied Conditions. A textbook of Rheumatology (Baltimore, MD: Williams and Wilkins). Le CH, Nicolson AG, Morales A, and Sewell KL (1 9 9 7 ). Suppression of collagen-induced arthritis through adenovirus-mediated transfer of a modified tumor necrosis factor alpha receptor gene. A r t h r i t i s R h e u m 40, 16621669. Loetscher H, Pan YC, Lahm HW, Gentz R, Brockhaus M, Tabuchi H, and Lesslauer W (1 9 9 0 ). Molecular cloning and expression of the human 55 kd tumor necrosis factor receptor. C e l l 61, 351-359. Ma Y, Thornton S, Boivin GP, Hirsh D, Hirsch R, and Hirsch E (1 9 9 8 ). Altered susceptibility to collagen-induced arthritis in transgenic mice with aberrant expression of interleukin-1 receptor antagonist. Arthritis Rheum. 41, 1798-1805. Ma Y, Thornton S, Duwell LE, Bluestone JA, and Hirsch R (1 9 9 8 ). Gene therapy with vIL-10 inhibits CIA. J . I m m u n o l . 161, 1516-1524. MacNeil IA, Suda T, Moore KW, Mosmann TR, and Zlotnik A (1 9 9 0 ). IL-10, a novel growth cofactor for mature and immature T cells. J Immunol 145, 4167-4173. Makarov SS, Olsen JC, Johnston WN, Anderle SK, Brown RR, Baldwin AS, Jr., Haskill JS, and Schwab JH (1 9 9 6 ). Suppression of experimental arthritis by gene transfer of interleukin 1 receptor antagonist cDNA. P r o c N a t l A c a d Sci USA 93, 402-406. McCarthy M (1 9 9 6 ). Gene therapy for rheumatoid arthritis starts clinical trials. Lancet 348, 323. Mohler KM, Torrance DS, Smith CA, Goodwin RG, Stremler KE, Fung VP, Madani H, and Widmer MB (1 9 9 3 ). Soluble tumor necrosis factor (TNF) receptors are effective therapeutic agents in lethal endotoxemia and function simultaneously as both TNF carriers and TNF antagonists. J I m m u n o l 151, 1548-1561. Moreland LW, Baumgartner SW, Schiff MH, Tindall EA, Fleischmann RM, Weaver AL, Ettlinger RE, Cohen S, Koopman WJ, Mohler K, Widmer MB, and Blosch CM (1 9 9 7 ). Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. N Engl J Med 337, 141-147.

Thornton and Hirsch: Gene therapy for arthritis Mori L, Iselin S, DeLibero G, and Lesslauer W (1 9 9 6 ). Attenuation of collagen-induced arthritis in 55-kDa TNF receptor type 1 (TNFR1)-IgG1-treated and TNFR1-deficient mice. J . I m m u n o l . 157, 3178-3182. Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH, Verma IM, and Trono D (1 9 9 6 ). In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector [see comments]. S c i e n c e 272, 263-267.

induction of collagen type II arthritis in mice. P r o c . N a t l . A c a d . S c i . 89, 7375-7379. Tripathy SK, Svensson EC, Black HB, Goldwasser E, Margalith M, Hobart PM, and Leiden JM (1 9 9 6 ). Long-term expression of erythropoietin in the systemic circulation of mice after intramuscular injection of a plasmid DNA vector. Proc Natl Acad Sci USA 93, 10876-10880.

Nita I, Ghivizzani SC, Galea-Lauri J, Bandara G, Georgescu HI, Robbins PD, and Evans CH (1 9 9 6 ). Direct gene delivery to synovium. Arthritis Rheum. 39, 820-828.

van Roon JAG, van Roy LAM, Gmelig-Meyling FHJ, Lafeber FPJG, and Bijlsma JWJ (1 9 9 6 ). Prevention and reversal of cartilage degradation in rheumatiod arthritis by interleukin10 and interleukin-4. Arthritis Rheum. 39, 829-835.

Otani K, Nita I, Macaulay W, Georgescu HI, Robbins PD, and Evans CH (1 9 9 6 ). Suppression of antigen-induced arthritis in rabbits by ex vivo gene therapy. J . I m m u n o l . 156, 3358-3562.

Walmsley M, Katsikis PD, Abney E, Parry S, Williams RO, Maini RN, and Feldmann M (1 9 9 6 ). Interleukin-10 inhibition of the progression of established collageninduced arthritis. Arthritis Rheum. 39, 495-503.

Raz E, Watanabe A, Baird SM, Eisenberg RA, Parr TB, Lotz M, Kipps TJ, and Carson DA (1 9 9 3 ). Systemic immunological effects of cytokine genes injected into skeletal muscle. Proc Natl Acad Sci USA 90, 4523-4527.

Whalen J, Lechman E, Robbins P, and Evans C. (1 9 9 8 ). Gene transfer of the viral IL-10 gene to the mouse footpad can prevent collagen-type II induced arthritis. In 44th Annual Meeting, Orthaepaedic Research Society (New Orleans, Louisiana), pp. 308.

Sawchuck S, Boivin GP, Duwel LE, Ball W, Bove K, Trapnell B, and Hirsch R (1 9 9 6 ). Anti T cell receptor monoclonal antibody prolongs transgene expression following adenovirus-mediated invivo gene transfer to the mouse synovium. Hum. Gene Ther. 7, 499-506 Smith CA, Davis T, Anderson D, Solam L, Beckmann MP, Jerzy R, Dower SK, Cosman D, and Goodwin RG (1 9 9 0 ). A receptor for tumor necrosis factor defines an unusual family of cellular and viral proteins. S c i e n c e 248, 1019-1023. Song XY, Gu M, Jin WW, Klinman DM, and Wahl SM (1 9 9 8 ). Plasmid DNA encoding transforming growth factor-beta1 suppresses chronic disease in a streptococcal cell wallinduced arthritis model. J C l i n I n v e s t 101, 2615-2621. Tanaka Y, Otsuka T, Hotokebuchi T, Miyahara H, Nakashima H, Kuga S, Nemoto Y, Niiro H, and Niho Y (1 9 9 6 ). Effect of IL-10 on collagen-induced arthritis in mice. I n f l a m m . R e s . 45, 283-288. Thorbecke GJ, Shah R, Leu CH, Kuruvilla AP, Hardison AM, and Palladino MA (1 9 9 2 ). Involvement of endogenous tumor necrosis factor # and transforming growth factor " during

248

Williams RO, Ghrayeb J, Feldmann M, and Maini RN (1 9 9 5 ). Successful therapy of collagen-induced arthritis with TNF receptor-IgG fusion protein and combination with anti-CD4. I m m u n o l o g y 84, 433-439. Wooley PH, Whalen JD, Chapman DL, Berger AE, Richard KA, Aspar DG, and Staite ND (1 9 9 3 ). The effect of an interleukin-1 receptor antagonist protein on type II collagen-induced arthritis and antigen-induced arthritis in mice. Arthritis Rheum. 36, 1305-1314. Yelin E (1 9 9 2 ). Arthritis. The cumulative impact of a common chronic condition. Arthritis Rheum. 35, 489-97. Yovandich J, O'Malley B, Jr., Sikes M, and Ledley FD (1 9 9 5 ). Gene transfer to synovial cells by intra-articular administration of plasmid DNA. Hum Gene Ther 6, 603610. Zhang H, Yang Y, Horton JL, Samoilova EB, Judge TA, Turka LA, Wilson JM, and Chen Y (1 9 9 7 ). Amelioration of collagen-induced arthritis by CD95 (Apo-1/Fas)-ligand gene transfer. J C l i n I n v e s t 100, 1951-1957.

Gene Therapy and Molecular Biology Vol 3, page 249 Gene Ther Mol Biol Vol 3, 249-256. August 1999.

Antisense gene therapy in the long-term control of hypertension Review Article

Craig H. Gelband1, Michael J. Katovich 2, Mohan K. Raizada1* 1

Departments of Physiology and 2Pharmacodynamics , Colleges of Medicine and Pharmacy, University of Florida, Gainesville, FL 32610 _________________________________________________________________________________________________ *Correspondence: Mohan K. Raizada, Ph.D., Professor of Physiology, University of Florida, College of Medicine, P.O. Box 100274, Gainesville, FL 32610, USA. Tel: 352-392-9299; Fax: 352-846-0270; E-mail: mraizada@phys.med.ufl.edu Key words: AT1 receptor antisense, hypertension, viral vectors, cardiac and renal pathophysiology, long-term prevention A b b r e v i a t i o n s: RAS, renin-angiotensin system; ACE, angiotensin converting enzyme; AT1 R, angiotensin II type 1 receptor; AT 1 R-AS angiotensin II type 1 receptor antisense; A n g , angiotensin; SHR, spontaneously hypertensive rat Received: 7 December 1998; accepted 10 December 1998

Summary Studies from the last two decades have established that both circulating and tissue renin-angiotensin system (RAS) are important. Their coordinated interaction is essential in the regulation of blood pressure and play a key role in the development, establishment and maintenance of hypertension. Interruption of the RAS pathway, either by preventing the formation of Ang II (i.e. ACE inhibitor) or by blocking its actions at the level of the receptor (i.e. AT 1 receptor antagonists), has been shown to reduce BP and protect against target-organ injury. Since there are problems associated with pharmacological control of high blood pressure, we developed a viral gene delivery approach to target hypertension. It was our intention to try and interrupt the RAS at the genetic level in order to achieve long term control of hypertension and reversal of pathophysiology associated with the disease. In general, delivery of antisense to the AT 1R was able to prevent (for up to 18 months) or reverse the elevated blood pressure, and the alterations in vascular calcium homeostasis, alterations in ion channel activity, and cardiac vascular ultrastructure. These results demonstrate that antisense gene delivery is useful in the long-term treatment of hypertension.

I. Current pharmacological treatment for hypertension A stepped care regimen, starting with drugs of lowest toxicity and adding drugs from other groups, is often used to manage hypertension. First line therapy is the use of diuretics including the thiazides. If response to the thiazides is inadequate to control the hypertension, a beta-adrenoceptor (!-blocker) would then be added to the regimen. If the antagonist response to the diuretic and the !-blocker is inadequate at tolerated doses then a direct vasodilator (calcium channel blocker) is generally added. Finally, if this combination does not work or is not

249

tolerated an ACE inhibitor is then substituted. In actuality, the ACE inhibitors are widely prescribed drugs of choice and have been beneficial in a wide groups of patients with primary hypertension. The reason for such a success using ACE inhibitors is that they not only attenuate vasoconstriction but have some important vasoprotective effects. These vasoprotective effects include: an antiatherogenic effect, an antiproliferative and antimigratory effect, improves/restores endothelial function, antiplatelet effect, enhances fribrinolysis, and improves arterial compliance (Lonn et al., 1994). Thus it is not surprising that emphasis has been placed in developing a strategy aimed at the RAS.

Gene Therapy and Molecular Biology Vol 3, page 250

Table 1. Pharmacological therapy for the treatment of angiotensin II-dependent hypertension

Jeunemaitre et al.,1992; Brunner et al., 1993; Bonnardeaux et al., 1994), it would only appear logical that AT1R is an important target in the intervention of high blood pressure and hypertension. Although major strides have been made in developing drugs which interfere with either Ang II formation or its action toward the management of Ang-dependent hypertension, there is neither long-term prevention nor a cure for this disease.

II. Renin-angiotensin system and its role in hypertension Primary human hypertension is characterized by normal cardiac output and an increase in total peripheral resistance (Khalil et al., 1990). Hypertension is one of the most important risk factors for stroke, congestive heart failure, myocardial infarction, end-stage renal diseases and peripheral vascular disease (Stamler et al., 1993; Kang et al., 1994; Whelton 1994). Studies from the last two decades have established that both circulating and tissue renin-angiotensin system (RAS) are important, that their coordinated interaction is essential in the regulation of blood pressure and they play a key role in the development, establishment and maintenance of hypertension (Brunner et al.,1993; Whelton 1994; Hsueh et al.,1995). The relevance of the RAS to blood pressure control is further supported by reports that various genes that encode renin, angiotensinogen, angiotensin converting enzyme (ACE), and the angiotensin II type 1 receptor (AT1R) have been associated with hypertension in both human and animal models (Kurtz et al.,1990; Jeunemaitre et al.,1992; Bonnardeaux et al., 1994). Interruption of the RAS pathway, either by preventing the formation of Ang II (i.e. ACE inhibitor) or by blocking its actions at the level of the peptide receptor (i.e. AT1 receptor antagonists), has been shown to reduce BP and protect against target-organ injury (Vogt et al.,1993; Kang et al., 1994; Kaneko et al., 1996 and Table 1). In fact, blockade of the RAS has become a well-accepted treatment for Ang-dependent hypertension and congestive heart failure (Vogt et al.,1993). Since ACE inhibition and AT1R blockade are standard means to treat hypertension and that AT1R encoding gene polymorphism is coupled with hypertension in both humans and in animal models of hypertension (Kurtz et al.,1990;

250

There are a number of limitations in the current pharmacological therapy to treat Ang-dependent forms of hypertension as summarized in Table 2. ACE inhibitors and AT1R antagonists must be administered chronically to achieve long term antihypertensive benefits. Required daily dosing and undesirable side effects such as sexual dysfunction, coughing, and lethargy, increased serum Ang II levels (with AT1R antagonists), and diminish patient compliance. Finally the attenuation or delay of non-hemodynamic pathophysiological impairments with these agents does not entirely reduce the risk to hypertensive patients (de Divitiis et al.,1993; Vogt et al.,1993). In other words current pharmacological therapies do not cure hypertension; only control the disease.

Table 2: Why administer gene therapy for the treatment of hypertension?

Gene Therapy and Molecular Biology Vol 3, page 251

Table 3. Gene therapy and hypertension sense approach.

Table 4. Gene therapy and hypertension antisense approach.

Therefore to circumvent the above problems associated with pharmacological control of high blood pressure, a number of research groups have used a gene delivery approach to target hypertension (Tables 3 and 4). Two approaches have been used to target hypertension using gene therapy; namely a sense and an antisense approach. Using the sense approach, Dr. Chao and her colleagues have been successful in over expressing genes relevant to vasodilatory effects. The genes have been delivered in hypertensive rats either in the form of naked DNA or by a viral mediated transduction system (T a b l e 3). For example, genes encoding kallikrein, ANP, eNOS and adrenomedullin have been successfully delivered and have had short-term reduction in high blood pressure and other beneficial effects on pathophysiological parameters associated with hypertension (Lin et al., 1995; 1997; Xiong et al., 1995; Chao et al., 1997; Yayama et al., 1998). The laboratories of Phillips 251

and Tomita independently have utilized an antisense oligodeoxynucloetide or naked DNA delivery approach to interrupting the RAS in order to target hypertension (T able 4; Gyurko et al., 1993; Wielbo et al., 1995; Tomita et al.,1995). However, the effects were short lived and did not present a major advance over the traditional pharmacological therapy. Later it was shown that viral delivery systems could extend the duration of antihypertensive action (Table 4, Phillips 1997). Although these studies did not produce desired long-term effects and thus were limited in scope, they were highly relevant in setting the stage indicating that a gene therapy strategy hold great potential in the treatment and cure of hypertension. Our objective has been to extend these concepts and investigate the feasibility of the antisense gene therapy approach in order to achieve long term control of hypertension and reversal of pathophysiology associated with the disease.

Gene Therapy and Molecular Biology Vol 3, page 252

Table 5. Protocol for AT1 -R-AS gene delivery.

III. Retroviral (LNSV)-mediated gene delivery system as a model to study AT1RAS therapy We chose the LNSV retrovirus because of its high infectivity, ability to effectively integrate into cells particularly, in slowly and rapidly diving cells, and its potential for long-term expression of an introduced gene (Lu and Raizada, 1995, Lu et al., 1995). In addition, the vector has been shown to influence non-dividing cells to a limited degree (Murata et al., 1998; personal communication N. Muzyczka, Univ. Florida). These properties, coupled with the fact that significant remodeling occurs during the development and establishment of hypertension, we argued that retroviruses may be an excellent vector for such purposes. With this in hand, our first attempts were to show that retroviral mediated gene delivery of AT1RAS in vitro would be successful. Astroglial cells in primary cultures were chosen first to demonstrate the efficiency of gene transduction mediated by this vector. The viral particles were able to infect >96% of the cells as evidenced by the detection of AT1R-AS transcript using RT- in situ PCR and Northern analysis. This was associated with a significant decrease in the number of AT1 receptors and AT1 receptor-mediated actions in these cells (Lu et al., 1995b). Next, primary neuronal cultures from hypothalamus and brainstem were used since neurons in culture have limited capacity to multiply and since neurons form the SHR show an increased expression of the AT1R gene, an increased Ang II-dependent norepinephrine (NE) uptake, increased stimulation of mRNA for c-fos and the NE transporter when compared to the normotensive control (Lu et al. 1995a, 1995b). Infection of neuronal cultures with the LNSV containing AT1R-AS resulted in decrease in AT1R number, an inhibition of AT1R-mediated stimulation of both

252

c-fos and NE transporter mRNA, as well as NE uptake in the SHR neurons (Lu et al., 1995a, 1995b). These data not only showed the retrovirally mediated delivery of AT1R-AS could be used to selectively control the actions of Ang II but laid the framework for the in vivo studies.

IV. Prevention of the development of high blood pressure and associated pathophysiology using in vivo AT 1R-AS gene delivery The first in vivo approach that we used was based on the hypothesis that interruption in the activity of the RAS at a “critical� stage in the development would prevent the onset of high blood pressure and other pathophysiology alterations associated with hypertension on a permanent basis. We used the spontaneously hypertensive rat (SHR) which is the most widely used animal model for studying human primary hypertension. As stated previously pharmacological intervention has been relatively successful in normalizing the elevation in blood pressure associated with hypertension in this model. However, the assumption that reduction of blood pressure will totally reverse hypertension-induced pathophysiological changes remains unclear.The protocol used for AT1R-AS gene delivery was to give a single intracardiac injection of the antisense into the ventricle of a 5 day old rat (T a b l e 5). This route of administration insured that the antisense was delivered through out the periphery . Indeed, using this route of administration, the AT1R-AS is expressed in a number of physiologically relevant tissue, including adrenals, heart, mesenteric arteries, kidney, and liver (Iyer et al., 1996). With the knowledge that the AT1R-AS is expressed in a number of different tissue types we next

Gene Therapy and Molecular Biology Vol 3, page 253 investigated whether the AT1R-AS had any effect on blood pressure and other cardiovascular pathophysiological alterations associated with the SHR (Figure 1 and Table 6). We first reported that AT 1R-AS can prevent the onset of high blood pressure for up to 90 days after a single injection (Figure 1, Iyer et al., 1996). We have extended those studies and now show an extension of up to 120 days (Figure 1, Martens et al., 1998), 210 days (Figure 1, Gelband et al., 1999) and 18 months (Reaves et al., 1999). This prevention of an increase in blood pressure is associated with a decrease in the specific binding of Ang II to the AT1R (Iyer et al.,1996). Similarly the AT1R-AS gene delivery prevented the Ang II dependent stimulation of blood pressure and the Ang II-stimulated increase in drinking in the SHR (Iyer et al.,1996). A number of cardiovascular pathophysiological alterations are exhibited in hypertension. These include altered renal resistance and arteriolar contractile sensitivity to circulating agents (i.e. Ang II and norepinephrine) as well as voltage dependent stimuli (KCl), endothelial dysfunction, increased smooth muscle cell Ca2+ current, increased Ca2+ release from the sarcoplasmic reticulum of smooth muscle cells, decreased smooth muscle cell voltage-dependent potassium channel (Kv) activity, increased left ventricular to body weight ratios and increased cardiac fibrosis. AT1R-AS gene delivery prevented all cardiovascular pathophysiological alterations associated with the disease mentioned above but caused no visible inflammatory response (Martens et al., 1998; Gelband et al., 1999).

V. Reversal of the development of high blood pressure and associated pathophysiology using in vivo AT 1R-AS gene delivery Although we have used this gene delivery approach to prevent the development of high blood pressure and cardiovascular pathophysiology in the developing SHR, the ultimate strategy would be the reversal of these actions in the adult SHR. Therefore we performed in vivo gene delivery studies in the adult SHR to determine if we could reverse the pathophysiology associated with hypertension. A similar protocol was used for gene delivery except the AT1R-AS was injected into the adult SHR six days in a row instead of a single injection (Gelband et al., 1998). This protocol resulted in a significant lowering of blood pressure for up to 45 days. At day 45 the blood pressure of the SHR treated with AT1R-AS was similar to the control SHR. In renal resistance arterioles the enhanced contractile response to KCl, norepinephrine, and angiotensin II as well as decreased endothelium-dependent relaxation was reversed in the SHR treated with AT1R-AS. Finally, the left ventricular weight to body weight ratio, an index of hypertension, was reversed in the adult SHR treated with AT 1R-AS. These results demonstrated the potential use of a similar gene transfer approach for long term reversal of

hypertension.

VI. Future directions Is antisense gene therapy targeting the RAS a therapeutic step forward? In short, the answer is yes. It results in the prevention and reversal of the increase in mean blood pressure and the associated pathophysiological impairments in hypertension. It also offers an alternative to the compliance problem and complications of vascular and target-organ injury. Finally, the AT1R-AS therapy does not produce a significant increase in plasma Ang II levels compared with losartan, the AT1R antagonist (Lu et al., 1997). Therefore, AT1R-AS gene delivery and therapy does have prolonged antihypertensive effects without the possible adverse side effects produced by traditional pharmacological therapies. Yet, there is a still question regarding the method of delivery. Conventional wisdom states that the LNSV retrovirus should only be successful in a population of cells undergoing cell division. Yet we find that there is an effect in the adult SHR. This leads to our first future direction and that is the development of a better viral gene delivery tool. The ideal viral vector should have the following characteristics for its successful use in a long-term reversal of hypertension: (i ) high titer should be achieved reproducibly and conveniently; (i i ) chromosome specific integration; (i i i ) long-term expression; (i v ) cell specificity and (v) no immune response. To date the ideal viral vector does not exist, but with genetic engineering it is only a matter of time before it is developed. At the present time the virus of choice may be a lenti or adeno-associated virus (AAV)-based vectors. A lentiviral based vector, for example, has the potential to be highly infective, can integrate into the host genome, has long term expression and little immune response. However, they are poorly defined at the present time. In contrast, AAV vectors are not highly infective but elicit a small immune response. In order for this approach to be successful for consideration in humans, it needs to demonstrate its effectiveness in many other forms of hypertension. Thus, our alternative direction would be to examine the feasibility of this approach in both non-genetic models of hypertension (such as the two kidney, one-clip Goldblatt model and the DOCA salt model of hypertension) as well as a monogenetic model of hypertension (such as the renin-transgenic rat). Other components of the RAS, such as antisense to ACE and angiotensinogen should also be tested in the prevention/reversal of hypertension. Antisense to ACE is of particular importance since ACE inhibitors have been shown to be beneficial not only as antihypertensive agents but also to play an important role in protecting against myocardial infarction, kidney failure, and the restenosis/remodeling that occurs after balloon injury in angioplasty. The latter would

253

Gene Therapy and Molecular Biology Vol 3, page 254

Figure 1. Time course of the change in blood pressure after antisense gene delivery. There is no change in the blood pressure in the control or antisense treated WKY rats. However there is a signifncat decrease in blood pressure in the SHRs that were treated with antisense. P<0.05, n> 8.

II blockade compared with other methods of inhibiting the renin-angiotensin system. J. Hyperten., 11, 553-558.

be clinically beneficial to those who are not only hypertensive but undergo coronary balloon angioplasty every year. Taken together gene therapy holds promise for a single dose, longterm treatment of hypertension and other potentially lethal cardiovascular disorders.

Chao J, Jin L, Lin KF, Chao L (1 9 9 7 ) Adrenomedullin gene delivery reduces blood pressure in spontaneously hypertensive rats. H y p e r t e n s . R e s . 20, 269-277. Christopherson KS, Bredt DS (1 9 9 7 ) Nitric oxide in excitable tissues, physiological roles and disease. J C l i n I n v e s t 100, 2424-2429.

References

de Divitiis O, Celentano A, De Simone G, Di Somma S, Galderisi M, Liguori V, de Divitiis M, Petitto M (1 9 9 3 ) Management of the patient with left ventricular hypertrophy. Eur Heart J Suppl D, 22-32.

Bonnardeaux, A., Davies, E., Jeunemaitre, X., Fery, I., Charru, A., Clauser, E., Tiret, L., Cambien, F., Corvol, P. and Soubrier, F. (1 9 9 4 ) Angiotensin II type 1 receptor gene polymorphism in human essential hypertension. H y p e r t e n s i o n 24, 63-69. Brunner, H.R., Nussberger, J. and Waeber, B. (1 9 9 3 ) Angiotensin

254

Gene Therapy and Molecular Biology Vol 3, page 255

Table 6. Effect of AT1 -RAS in developing rats

93, 9960-9965.

Gelband, C.H. Reaves, P.Y. Evans, J. Wang, H, Katovich, M.J. and Raizada M.K. Angiotensin II Type 1 Receptor Antisense Gene Therapy Prevents Altered Renal Vascular Calcium Homeostasis in Hypertension. H y p e r t e n s i o n (In Press.)

Jeunemaitre, X., Soubrier, F., Kotelevtsev, Y.V., Liffon, R.P., Williams, C.S., Charru, A., Hunt, S.C., Hopkins, P.N., Williams, R.R., Label, J.M. and Corvol, P. (1 9 9 2) Molecular basis of human hypertension, Role of angiotensinogen. C e l l 71, 169-180.

Gelband, C.H., Reaves, P.Y., Dang, H., Wang, H., Raizada, M.K., and Katovich, M.J. (1 9 9 8 ) Reversal of hypertension by retroviral-mediated (LNSV) delivery of angiotensin II type 1 receptor antisense (AT1 R-AS) in the adult spontaneously hypertensive rat (SHR). C i r c u l a t i o n 98, I-320.

Kaneko, K., Susic, D., Nunez, E. and Frohlich, E.D. (1 9 9 6 ) Losartan reduces cardiac mass and improves coronary flow reserves in the spontaneously hypertensive rat. J . Hypertens. 14, 645-653.

Gyurko R, Wielbo D, Phillips MI (1 9 9 3 ) Antisense inhibition of AT1 receptor mRNA and angiotensinogen mRNA in the brain of spontaneously hypertensive rats reduces hypertension of neurogenic origin. R e g . P e p . 49, 167-174.

Kang, P.M., Landau, A.J., Eberhardt, R.T. and Fishman, W.H. (1 9 9 4 ) Angiotensin II receptor antagonists, A new approach to blockade of renin angiotensin system. Am. Heart J. 127, 1388-1401.

Hsueh, W.A., Do, Y-S., Anderson, P.W., and Law, R.E. (1 9 9 5 ) Angiotensin II in cell growth and matrix production. IN, Tissue renin-angiotensin system. (Mukhopadhyay, A. and Raizada, M.K., eds) Plenum Press, New York, pp.217-223.

Khalil, R. A., Lodge, N. J., Gelband, C. H., and van Breemen, C. (1 9 9 0 ) in Hypertension, Pathophysiology, Diagnosis, and Management, eds. Laragh, J. H. & Brenner, B. M., (Raven Press), pp. 547-567.

Iyer, S.N., Lu, D., Katovich, M.J., Raizada, M.K. (1 9 9 6) Chronic control of high blood pressure in the spontaneouslyhypertensive rat by delivery of angiotensin type 1 receptor antisense. P r o c . N a t l . A c a d . S c i . U . S . A .

Kurtz, T.W., Simonet, L., Kabra, P.N., Wolfe, S., Chen, L. and HjeIIe, B.L. (1 9 9 0 ) Consegregation of the renin allele of the spontaneously hypertensive rat with an increase in blood

255

Gelband et al: Antisense gene therapy for hypertension pressure. J . C l i n . I n v e s t . 85, 1328-1332.

Wielbo D, Sernia C, Gyurko R, Phillips MI (1 9 9 5 ) Antisense inhibition of hypertension in the spontaneously hypertensive rat. H y p e r t e n s i o n 25, 314-319.

Lin KF, Chao J, Chao L (1 9 9 5 ) Human atrial natriuretic peptide gene delivery reduces blood pressure in hypertensive rats. H y p e r t e n s i o n 26, 847-853.

Xiong W, Chao J, Chao L (1 9 9 5 ) Muscle delivery of human kallikrein gene reduces blood pressure in hypertensive rats. H y p e r t e n s i o n 25, 715-719.

Lin KF, Chao L, Chao J (1 9 9 7) Prolonged reduction of high blood pressure with human nitric oxide synthase gene delivery. H y p e r t e n s i o n 30, 307-313.

Yayama K, Wang C, Chao L, Chao J (1 9 9 8 ) Kallikrein gene delivery attenuates hypertension and cardiac hypertrophy and enhances renal function in Goldblatt hypertensive rats. H y p e r t e n s i o n 31, 1104-1110.

Lonn EM, Yusuf S, Jha P, Montague TJ, Teo KK, Benedict CR, Pitt B (1 9 9 4 ) Emerging role of angiotensin-converting enzyme inhibitors in cardiac and vascular protection. C i r c u l a t i o n 90, 2056-2069. Lu, D., and Raizada, M.K. (1 9 9 5 ) Delivery of angiotensin type 1 receptor antisense inhibits angiotensin action in neurons from hypertensive rat brain. P r o c . N a t l . A c a d . S c i . U . S . A ., 92, 2914-2918. Lu, D., Raizada, M.K., Iyer, S., Reaves, P., Yang, H., Katovich, M.J. (1 9 9 7 ) Losartan vesus gene therapy, chronic control of high blood pressure in spontaneously hypertensive rats. H y p e r t e n s i o n 30, 363-370. Lu, D., Yu, K. and Raizada, M.K. (1 9 9 5 ) Retrovirus meidated transfer of an angiotensin type 1 receptor antisense sequence decreases AT1 -Rs and angiotensin II action in astroglial and neuronal cells in primary culture from the brain. P r o c . N a t l . A c a d . S c i . U . S . A ., 92, 1162-1166. Martens, J.R., Reaves, P.Y., Lu,D., Berecek, K.H. Bishop, S.P. Katovich, M.J., Raizada, M.K., and Gelband, C.H. (1 9 9 8 ) Prevention of cardiac and renovascular pathophysiological changes in hypertension by AT1 receptor antisense gene therapy. P r o c . N a t l . A c a d . S c i . U S A 95, 2664-2669. Murata, T., Hoffmann, S. Ishibashi, T., Spee, C., Gordon, E.M., Anderson, W.F., Hinton, D.R., and Ryan, S.J. (1 9 9 8 ) Retrovirus-mediated gene transfer targeted to retinal photocoagulation sites. D i a b e t o l o g i a 41, 500-506. Phillips MI. (1 9 9 7 ) Antisense inhibition and adeno-associated viral vector delivery for reducing hypertension. H y p e r t e n s i o n 29, 177-187. Reaves, P.Y. H. Wang, D. Lu, H. Yang, M. J. Katovich, M.K. Raizada and C.H. Gelband. (1 9 9 9 ) Permanent reversal of hypertension and altered renal vascular Ca 2+ homeostasis by angiotensin II type 1 receptor antisense (AT1 R-AS) gene therapy. B i o p h y s . J . (In Press). Stamler, J., Stamler, R., Neaton, J.D. (1 9 9 3 ) Blood pressure, systolic and diastolic and cardiovascular risks, US population data. Arch. Intern. Med., 153, 598-615. Tomita N, Morishita R, Higaki J, Aoki M, Nakamura Y, Mikami H, Fukamizu A, Murakami K, Kaneda Y, Ogihara T (1 9 9 5 ) Transient decrease in high blood pressure by in vivo transfer of antisense oligodeoxynucleotides against rat angiotensinogen. H y p e r t e n s i o n 26, 131-136. Vogt, M., Motz, W.H., Schwartzkopf, B., and Strauer, B. E. (1 9 9 3 ) Pathophysiology and clinical aspects of hypertensive hypertrophy. Eur. Heart. J. 14, 2-7. 12. Whelton, P.K. (1 9 9 4 ) Epidemiology of hypertension. Lancet 334, 101-106.

256

Gene Therapy and Molecular Biology Vol 3, page 257 Gene Ther Mol Biol Vol 3, 257-269. August 1999.

Construction and deployment of triple ribozymes targeted to multicatalytic proteinase subunits C3 and C9 Research Article

Ling Ren, Shani L. Schalles, Weihua Pan, Corinne E. Isom, Sarah E. Loy, JiaHai Lee, Catharine M. Benedict, Mary T. Pickering, James S. Norris1, and Gary A. Clawson* Departments of Pathology, Biochemistry and Molecular Biology, Cell and Molecular Biology Program, and the Gittlen Cancer Research Institute, The Pennsylvania State University, Hershey, PA; and 1Departments of Microbiology and Immunology and Medicine, The Medical University of South Carolina, Charleston, SC. _________________________________________________________________________________________________ *Correspondence: Gary A. Clawson, Ph.D., M.D., Gittlen Cancer Research Institute,C7768, M S Hershey Medical Center, Pennsylvania State University, 500 University Drive, Hershey, PA 17033. Phone (717) 531-5632; Fax (717) 531-5298; E-mail: gac4@email.psu.edu A b b r e v i a t i o n s : AOs, antisense oligonucleotides; Rz , ribozymes; TRz, triple ribozyme; ITRz, internal targeted ribozyme Key Words: antisense oligonucleotides, triple ribozyme, transfection Received: 12 October 1998; accepted: 23 October 1998

Summary We have been developing triple ribozyme constructs for modulating gene expression, which consist of 2 cis-acting ribozymes flanking an internal trans-acting ribozyme, which is targeted to a selected cellular R N A . The 2 cis-acting ribozymes function autocatalytically, resulting in liberation of the internal ribozyme with minimal nonspecific flanking sequences. Here, we test 2 triple ribozyme constructs targeted to the multicatalytic proteinase subunits C 3 and C 9 . The liberated internal ribozyme is 5-20 times more active in vitro than the same ribozyme contained with a double-G mutant (which cannot undergo autocatalytic processing) or contained within nonspecific flanking vector sequences. These triple ribozymes were placed within inducible expression vectors, which were used to produce stably transfected hepatocyte cell lines. Both of the constructs distributed between nucleus and cytoplasm, produced substantial growth inhibition in the cell lines, and their effectiveness was due to their catalytic activity and not to antisense effects, as demonstrated with catalytically inactive mutants. Thus, the triple ribozyme approach appears to represent a substantial improvement over conventional ribozymes.

1992). AOs are also often toxic at concentrations necessary to observe the desired effects, and in addition can trigger myriad non-specific effects (Krieg et al., 1995; Stein and Cheng, 1993; Storey et al., 1991; Wagner, 1994). Finally, abundant mRNAs, or RNAs with significant secondary structure, are unlikely to be modulated efficiently. "Hammerhead" ribozymes provide an alternative approach to downregulating specific gene products (Christoffersen and Marr, 1995; Haseloff and Gerlach, 1988; Ohkawa et al., 1995; Parker et al., 1992; Perreault et al., 1990). Ribozymes (Rz) can be conceptualized as 2 functional elements, a conserved stem-loop structure forming the catalytic core, and

I. Introduction Use of antisense oligonucleotides (AOs) has afforded the opportunity to delineate functions of specific genes (Helene and Toulme, 1990; Izant and Weintraub, 1985; Zamecnik and Stephenson, 1978), and it has proven possible to differentially modulate endogenous vs. exogenous mRNAs using AOs (Benedict and Clawson, 1996). However, in spite of widespread applicability, use of AOs suffers from a number of limitations. For example, AOs are inherently unstable in cells, necessitating modifications to phosphodiester linkages (Hoke et al., 1991; Iverson et al.,

257

Ren et al: C3- and C9-targeted triple ribozymes flanking sequences which are reverse complementary to sequences surrounding the target site in an RNA transcript (Goodchild and Kohli, 1991; Haseloff and Gerlach, 1988). Rz-mediated cleavage occurs 3' to a targeted nucleotide triplet NUX (N can be any nucleotide, whereas X can be A, C, or U) (Haseloff and Gerlach, 1988; Kawasaki et al., 1996; Koizumi et al., 1989; Ruffner et al., 1990): When the third nucleotide is G, cleavage cannot occur. Flanking sequences confer specificity, and extend on both sides of the target site selected. Relatively short flanking sequences (6-9 nt each) allow sufficient specificity for the cleavage reaction, while allowing ready dissociation from the target, which is typically rate-limiting for the catalytic cycle (Goodchild and Kohli, 1991; Parker, et al., 1992). However, in spite of the requirement for relatively short specifier/flanking sequences for efficient catalytic rates in vitro, such constructs are often relatively ineffective in vivo. For example, Rz targeted to HIV-1 RNAs were optimally active in vivo with flanking sequences ! 33 nt (Crisell et al., 1993), even though these Rz showed relatively poor catalytic activity in vitro. Parameters likely to contribute to this disparity between in vitro vs. in vivo activity are target site accessibility, cellular compartmentation, as well as potential deleterious effects of non-specific flanking sequences present within traditional Rz transcripts. With this in mind, we have been developing triple ribozyme (TRz) constructs which offer distinct advantages; these constructs consist of an internal targeted ribozyme (ITRz), which is flanked by two cis-acting ribozymes. The 2 cis-acting ribozymes are targeted to sites within the primary TRz construct, so that following transcription, they function autocatalytically liberating the ITRz. Our initial design is similar to one previously described by Taira and colleagues (Ohkawa et al., 1992; Ohkawa et al., 1993; Taira et al., 1991), and other self-trimming Rz have also been designed (Altschuler et al., 1992; Dzianott and Bujarski, 1989; Ruiz et al., 1997). We have previously characterized a TRz targeted to the retinoblastoma gene product (Rb) mRNA, and showed that it was efficiently liberated and functioned effectively in vivo (Benedict et al., 1998). Similar results were obtained with a TRz targeted to repetitive B2 transcripts, where we further showed that this approach results in a distribution of liberated ITRz between nucleus and cytoplasm (Crone et al., 1998). Here we have designed and produced TRz constructs targeted to the C3 and C9 subunits of the multicatalytic proteinase (MCP), and present results documenting their effectiveness both in vitro and in vivo. Their effectiveness in vivo is clearly due to their catalytic activity, since catalytically inactive mutant TRz do not produce analogous results. The TRz constructs were superior to the same ITRzs contained within TRz constructs which could not undergo self-liberation, or than the same ITRzs flanked by nonspecific vector sequences.

and were designed to function autocatalytically. We then chose target sites in MCP subunits C3 and C9 mRNAs, and constructed ITRz specific for the chosen target sites. These specific ITRz were then inserted into the parent double ribozyme construct to create the targeted TRz constructs (C3TRz and C9TRz). Catalytically inactive mutants (C3mTRz and C9mTRz) were created by changing 2 essential nucleotides. In addition, double-G mutants were created (designated C3GGm and C9GGm), in which the autocatalytic cleavage sites were changed from GUCs to GUGs, so that autocatalytic cleavage could not occur. Finally, “single” Rz were created (designated C3SRz and C9SRz), where the cis-acting Rz were replaced with nonspecific flanking vector sequences. All constructs were then placed into a vector allowing bidirectional transcription, as well as into suitable expression vectors. All constructs were sequenced to confirm identity. To document self-liberation, we employed pCRII constructs. For the C3TRz, "sense" transcripts were produced with Sp6 polymerase, and after transcription reactions the RNA was analyzed by PAGE (Figure 1A). About 80% of TRz transcripts underwent complete or partial processing during the transcription incubation. The expected 5', 3', and ITRz bands were observed, and additional bands were also observed for partially-processed 5' + ITRz and 3' + ITRz products (Figure 1A, as indicated). Enzyme "pausing" (premature termination) at the 5' autocatalytic site was generally low with Sp6 polymerase, although in some experiments it occurred in nearly 20% of transcripts (similar to that generally observed with T7 polymerase, see below). In further experiments, unprocessed or partially-processed transcripts were eluted from gels and incubated for various periods at 37°C. Approximately 75% of the full length transcripts underwent processing (complete or partial) during this subsequent incubation (Figure 1B). Autocatalytic cleavage occurred somewhat more efficiently at the 3' site compared with the 5' site under these conditions. With the partially processed transcripts, 85% of the 5' + ITRz transcripts underwent autocatalytic processing, while 55% of the 3' + ITRz transcripts were processed. Since no enzyme "pausing" complicates these experiments, the expected stoichiometry was observed for the various products. Quantitative analyses (for 3 separate preparations) indicate that >80% of C3TRz transcripts undergo complete processing, with consequent liberation of the ITRz, in the initial transcription reaction and the subsequent incubation period. For the C9TRz, transcription from the Sp6 promoter produced the "sense" TRz, which underwent self-cleavage (Figure 2A). Cleavage proceeded efficiently at both autocatalytic sites for 50% of the transcripts, with production of 5'-, 3'-, and liberated ITRz (as designated). A smaller proportion (roughly 20%) of transcripts showed selfcleavage at only one of the internal 5' or 3' sites (Figure 2). In addition, quantitative analyses indicated that the 5' Rz was overrepresented by 20% due to enzyme pausing between the 5' Rz and ITRz.

II. Results We prepared a double ribozyme cassette, consisting of 2 cis-acting ribozymes flanking a cloning site. These ribozymes were targeted to GUCs in the primary transcript, 258

Gene Therapy and Molecular Biology Vol 3, page 259

F i g u r e 1 . In vitro Characterization of the C3TRz. P a n e l A . Self-liberation. 32 P-labeled C3TRz was transcribed in the sense and antisense directions using Sp6 and T7 RNA polymerases (respectively), and the products were examined by PAGE and autoradiography. Transcription with Sp6 produced the active TRz, with consequent liberation of the ITRz (66 nt), the 5' cisacting Rz (151 nt), and the 3' cis-acting Rz (146 nt), the latter 2 of which remain associated with some vector sequences. Two less intense bands observed represent incompletely processed transcripts, where either the 5' or 3' autocatalytic cleavage did not occur (these are designated 5' + I and 3' + I, respectively). Panel B . Reincubation of unprocessed/partially processed transcripts. The residual unprocessed TRz, or the partially processed 5' + I and 3' + I transcripts were excised from gels, and incubated for an additional 30 min at 37 C, and then analyzed by PAGE and autoradigraphy. Lane 1 shows results with transcripts which were unprocessed during the initial transcription reaction, while lanes 2 and 3 depict results from incubation of the 5' + I and 3' + I transcripts (respectively). The band in lane 3 corresponding to the 5' transcript arose from contamination of the excised 3' + I band with 5' + I transcript during the gel-purification procedure (it was present in much greater quantity in this experiment). No self-liberation was observed when transcription was performed in the antisense (T7) direction (data not shown), or when incubations were performed in the presence of excess EDTA.

2B, lanes 2 and 3). “Zero-time” controls indicated that no processing of transcripts occurred after RNA extraction. These results (and extended time-course experiments) indicate that the ITRz is effectively liberated from >80% of the C9TRz transcripts. The balance may be contained in complexes (Groebe and Uhlenbeck, 1988), although quantitative liberation might also eventually occur.

When unprocessed (full-length) transcripts were eluted from the gel and incubated for 30 min at 37°C, a significant proportion (44%) of the transcripts underwent efficient selfcleavage, although a somewhat higher proportion (50%) of transcripts with only one self-cleavage was observed (Figure 2B). In addition, elution and subsequent incubation (30 min) of the partially processed 5' + ITRz and 3' + ITRz transcripts showed that efficient processing occurred (Figure

259

Ren et al: C3- and C9-targeted triple ribozymes

F i g u r e 2 . In vitro Characterization of the C9TRz. Pane l A. Self-liberation. Sense transcripts were obtained with Sp6 polymerase, and antisense transcripts were generated with T7 polymerase. Quantitative analysis of radioactivity in the respective bands (designated as in Figure 2) showed that two-thirds of the initial TRz transcripts underwent processing during the transcription reaction, and two-thirds of these were completely processed. No processing of antisense transcripts was observed, either during the initial transcription reaction or the subsequent incubation (Panel B). P a n e l B . Reincubation of unprocessed transcripts. Full length TRz transcripts which were not processed during the initial transcription reaction were excised from gels and further incubated for 30 min at 37 C. 40% underwent processing, and one-half of these were correctly liberated. In the reincubation reaction (where no enzyme pausing occurs) the 3' + I product was elevated, indicating that autocatalytic processing at the 3’ site may be somewhat less efficient than the 5' processing for this particular construct. One other round of excision and incubation (not shown) showed that subsequent processing of full-length TRz transcripts contributes another 10% of liberated ITRz. Similar analyses of the 5' + I and 3' + I transcripts suggest that about one-half of these transcripts subsequently undergo complete processing. Taken together, these figures indicate that a similar proportion of C9TRz undergo complete processing (with consequent liberation of the ITRz) as compared with the C3TRz, although the time required for this processing was somewhat longer.

occurred after addition of excess EDTA. These results thus confirm efficient, time-dependent liberation of the ITRz from our constructs in vitro. We then performed a series of experiments to document cleavage of the targeted RNAs in vitro. 32P-labeled target RNAs were synthesized in vitro. Unlabeled C9TRz and C3TRz were also synthesized. For comparative purposes, the C9-targeted ITRz was directly transcribed in vitro, as were

In further experiments, we also examined effects of extending the initial transcription reaction and/or subsequent incubations. When the initial transcription reactions were extended from 1 h to 2 h, 140% more ITRz was liberated. When Mg++ was increased to 50 mM for the second h following the 1 h transcription reaction (effectively terminating transcription), we observed a 50% increase in liberation of the ITRz, and when this incubation was extended to 2 h, we observed a 110% increase. No processing

260

Gene Therapy and Molecular Biology Vol 3, page 261 Figure 3. Target cutting in vitro with C3TRz and C9TRz. Panel A. Target cleavage by the C9TRz. A 32 P-labeled 365 nt partial C9 target was transcribed, gel-purified, and incubated with the C9TRz. Following 0.5 h (lane 2) or 2 h (lane 3) incubations, RNA was examined by PAGE and autoradiography as described. Cleavage produced 156 and 209 nt fragments, with cleavage of 40% of transcripts at 2 h. No cleavage occurred in the absence of Rz (lane 4), or in the presence of excess EDTA. Panel B. Target Cleavage by the C3TRz. A 32 Plabeled partial C3-target RNA was transcribed, gel-purified, and incubated with C3TRz at 37 C at a 1:1 molar ratio. Following incubation, RNA was examined by PAGE and autoradiography. Here, the target C3 RNA transcript was 238 nt, with cleavage products of 146 and 92 nt. Under the conditions used, no cleavage occurred in the absence of Rz or in the presence of excess EDTA (lane 2). 33% and 80% of transcripts were cleaved after 0.5 and 2 h incubations (lanes 3 & 4, respectively), while 100% cleavage was observed after 6 h (data not shown). Panel C. Lineweaver-Burke plot of C9TRz activity in vitro. 32 P-labeled C9 target and unlabeled C9TRz were prepared. Extensive time-course incubations were conducted at 37 C with multiple substrate concentrations. The extent of cleavage was ascertained by PAGE and phosphorimager analyses, and linear regression values were obtained. The calculated km (under these arbitrary conditions) is about 0.2 ÂľM, and the indicated Vmax is 10 cleavages/Rz-h. P a n el D. Comparison of in vitro Catalytic Activities of C9 Rz Constructs. C9TRz preparations, the liberated ITRz, and the C9GGm and C9SRz constructs were purified and incubated with a 32 Plabeled 305 nt partial C9 target at 37 C for 0.5 h (A) or 2 h (B). Cleavage produced 159 nt and 146 nt fragments. Following incubations, cleavage was assessed by PAGE and autoradiography.

compared the activity of the C9 ITRz with the C9GGm and C9SRz constructs. The ITRz was 5 times more active than C9SRz (which is the same ITRz flanked by nonspecific vector sequences) and 20 times more active than C9GGm (which cannot undergo autocatalytic processing (Figure 3D). Comparable results were obtained with the analogous C3 constructs.

the C9SRz and C9GGm Rz constructs. Unlabeled Rz and 32 P-labeled target RNAs were then incubated for various periods, and the reaction products were examined by PAGE (F i g u r e s 3 A and 3B). More extensive kinetic analyses with C9TRz preparations and target RNA at various concentrations indicated an apparent km of about 0.2 ÂľM, with an apparent Vmax of about 10 cleavages per h per ribozyme under these conditions (Figure 3C). We then

261

Ren et al: C3- and C9-targeted triple ribozymes

F i g u r e 4 . Reduction of C9 mRNA and protein in stably transfected clones expressing the C9TRz. CWSV1 cells were transfected with the C9TRz in the LacSwitch vector, and stably transfected clones were obtained by antibiotic selection. Four of 5 clones showed markedly reduced growth rates, whereas one clone was not expressing the C9TRz and therefore served as an additional control. Of the 4 clones showing growth inhibition, clone 3 was the fastest growing (with a 60% reduction in growth rate), and was used for subsequent analyses; the other 3 clones grew so slowly they were not easily amendable to further analyses. P a n e l A . Northern blot analysis of cellular RNA from control CWSV1 cells (lane 1), from an antibiotic-selected clone not expressing the TRz (lane 2), and from clone 3 (lane 3). The reduction of C9 mRNA in clone 3 was >65%, when corrected for slight differences in loading, as determined with 28 S rRNA probe (lower panel). Panel B. Immunoblot analysis of C9 protein. Cellular proteins were prepared from an antibiotic-selected clone not expressing the TRz (left), and from clone 3 (right), separated by SDS-PAGE, and probed with an antibody directed against C9. The membrane was then stripped and reprobed with antibodies to rat albumin (lower panel). Normalization to equivalent loading demonstrated that the actual reduction in C9 protein was > 80%.

doxycycline (allowing each clone to serve as its own control). Induction of expression of the C3-targeted TRz resulted in a marked reduction in growth of stably transfected clones (Figure 5A and 5B), consistent with previous results following downregulation of C3 mRNA with AOs (Benedict and Clawson, 1996). We consistently observed a 23 day delay before growth inhibition was manifested. A similar reduction was observed following induction of expression of the C9-targeted TRz in stably transfected clones (Figure 6A and 6B), although with all of these clones the lag period was consistently 4 days. Northern blot analysis was used to confirm reduction of target RNAs (Figures 7A and 7B). In contrast to the doxycycline-dependent growth inhibition observed with the C3TRz and C9TRz constructs, expression of the catalytically inactive mutant TRz (C3mTRz and C9mTRz) in a number of stably transfected clones did not produce any significant effects on cell growth (Figure 8 shows growth studies with 2 stably transfected cell lines expressing the catalytically inactive C9mTRz). A reduction in cell growth was observed with cells expressing the C3mTRz: Cell growth inhibition was about 30% with induction of expression of C3mTRz (but was not statistically significant), compared with the 60% reduction produced by expression of C3TRz (Figure 5B, and data not shown).

We next tested the TRz constructs in cell culture, first using the LacSwitch vector system. Transient transfection analyses were performed with CWSV1 cells, and cytoplasmic RNA was isolated 24 h later and examined by Northern blot analysis (equivalent loading was verified with GAPDH). We observed marked reductions in targeted RNAs, which were not observed with control cells transfected with the parent double ribozyme, and these reductions were contingent upon induction of TRz expression. With both the C9TRz and C3TRz, we observed 50% reductions in target RNAs in CWSV1 cells (data not shown). Since the transfection efficiency of this procedure is approximately 50%, this suggests very efficient cleavage of the targeted mRNAs within cells in these transient transfections. We next developed stably transfected CWSV1 clones expressing the C9TRz or C3TRz with the LacSwitch system. With the C9-targeted TRz, 4 clones expressing the TRz were obtained. All 4 clones showed significant growth inhibition; 3 of these clones grew so slowly that they were not easily amendable to further analyses, whereas one clone (#3) showed a 60% reduction in growth rate, and was further analyzed. Expression of the C9-targeted TRz in clone #3 produced a 65% reduction in C9 mRNA (Figure 4A), and this reduction was paralleled by a similar reduction in rC9 protein (Figure 4B). Stably transfected CWSV1 clones were also obtained with C9- and C3-targeted TRz with the pTet-On system, and their growth was compared in the presence or absence of

262

Gene Therapy and Molecular Biology Vol 3, page 263

Figure 5 ( ) . Effects of expression of the C3TRz in vivo. CWSV1 cells were transfected with the pTet-On construct, and stably transfected clones were obtained by antibiotic selection. One of these stably transfected clones which showed good expression levels was then transfected with the C3TRz (and a hygromycin resistance construct) and stably transfected clones were obtained by hygromycin/neomycin selection. Panels A and B. Results of growth studies with 2 individual clones + doxycycline, with each clone thus serving as its own control. In each case, growth of the clones was significantly inhibited (* p < 0.05; ** p < 0.01) when expression of the C3TRz was induced by addition of doxycycline.

Figure 6. Effects of Expression of the C9TRz in vivo. Individual clones were obtained as described in Figure 5 and growth studies were conducted. P a n e l s A and B show results with 2 individual clones + doxycycline. In both cases, growth of the clones was significantly inhibited when expression of the C9targeted TRz was induced by addition of doxycycline.

263

Gene Therapy and Molecular Biology Vol 3, page 264 F i g u r e 7 ( ) . Effects of Expression of C3TRz and C9TRz on target mRNAs in vivo. Panel A. Northern blot analysis of C3 mRNA. Cytoplasmic RNA was isolated (from the clone shown in Figure 5A) after 4 days of TRz expression, separated by PAGE, and examined by Northern blot analysis with 32 P-labeled C3 probe. A major reduction in C3 target RNA (upper panel) was observed in clones expressing the C3TRz (lane 1) compared to the same clone not expressing it (lane 2). The membrane was then stripped and equivalent loading was verified by reprobing with 18S rRNA (lower panel). C3TRz expression was 32 then examined using an P-labeled oligonucleotide antisense to the ITRz (data not shown). P a n e l B . Northern blot analysis of cytoplasmic RNA (from the clone shown in Figure 6A) isolated 4 days after induction of C9TRz expression. A major reduction in C9 target mRNA was observed when TRz expression was induced (lane 1) compared with the same clone not expressing the TRz (lane 2). The membrane was then stripped and reprobed with 18S rRNA for documentation of loading (lower panel), and with an 32 P-labeled oligonucleotide antisense to the ITRz (data not shown).

Figure 8. Effects of Expression of Catalytically Inactive C9mTRz in vivo. CWSV1 clones stably transfected with the C9mTRz construct (in the pTet-On system) were obtained, and growth studies were conducted as described + doxycycline (see Figure 5), except that initial platings were at 5 x 104 cells/plate. P a n e l s A and B show growth studies with 2 representative stably transfected clones, neither of which showed significant growth reduction upon induction of expression with doxycycline. Concurrent growth studies (Panel C) confirmed a doxycyclinedependent reduction in growth rate (significant at p < 0.005) of a stably transfected clone expressing the catalytically active C9TRz construct (clone C92; see Figure 6A).

264

Gene Therapy and Molecular Biology Vol 3, page 265 RNA represents approximately 10% of total cellular RNA, our results indicate that about 90-95% of the ITRz is found in the cytoplasm. These results parallel those previously reported for Rb- and B2-targeted TRz constructs (Benedict et al., 1998; Crone et al., 1998).

In preliminary further experiments (using the pIND vector system which is inducible with the ecdysone analog ponasterone A), the constructs unable to undergo selfliberation (C3GGm and C9GGm), or the ITRz flanked by nonspecific vector sequences (C3SRz and C9SRz) were also not effective in reducing growth rate (data not shown). Cytoplasmic and nuclear RNA was isolated from two of the stably transfected clones expressing C3TRz and C9TRz, and TRz expression and self-liberation was examined by RT/PCR using “inner” and “outer” primer pairs (the inner primer pair amplifies both processed and unprocessed TRz transcripts, whereas the outer primer pair amplifies only unprocessed transcripts). RT/PCR analyses confirmed doxycycline-dependent expression of the TRz constructs in cytoplasmic RNA, and showed that essentially all detectable transcripts were processed in vivo (Figure 9). Interestingly, doxycycline-induction of TRz expression was not observed in nuclear RNA (Figure 9C), even though inductions of 5- to 10-fold were characteristically observed in cytoplasmic RNA from the same clones (Figure 9A). Given that nuclear

III. Discussion This communication documents the general effectiveness of targeted TRz constructs, in which the ITRz (targeted to a cellular RNA) is flanked by 2 cis-acting ribozymes. The two cis-acting ribozymes are targeted to nucleotide sequences within the primary transcript, so that cleavage results in autocatalytic liberation of the ITRz . This design creates an uncapped 5' end for the relatively short liberated ITRz, and is based upon a design previously described by Taira and colleagues.

Figure 9. RT/PCR Detection and Processing of ITRz in vivo. Stably transfected clones expressing C9TRz (clone #2) or C3TRz (clone #5) were grown under inducing (+) or non-inducing (-) conditions, and cytoplasmic and nuclear RNA was prepared. 1 µg of RNA was used in RT/PCR analyses with the inner (I) or outer (O) primer pairs as described. Following reactions, products were separated on 8% polyacrylamide gels, and the amplified products were visualized by autoradiography. Panel A shows TRz expression with cytoplasmic RNA, and B shows results with nuclear RNA. Concurrent amplification of 18S rRNA was performed for standardization. Experiments using doubleG mutant RNA transcripts confirmed that products were obtained with both primer pairs, with the inner primer pair being 1.8X as efficient. With the cellular RNAs, products were observed only with the inner primer pair, indicating that essentially all (> 90%) of the ITRz was liberated in vivo.

265

Gene Therapy and Molecular Biology Vol 3, page 266

Thus far, we have obtained reasonably effective cleavage of targeted RNAs by choosing target sites based on RNA structural modeling using the mFold program, although optimal definition of target sites requires additional analyses, for example use of ribozyme expression libraries (Lieber and Strauss, 1995). In three extensive comparisons thus far, we have obtained 10-103-fold greater activity with libraryselected TRz vs. TRz designed based on mFold modeling. The liberated ITRz appears to effectively distribute between nuclear (about 10%) and cytoplasmic (about 90%) compartments (Benedict, et al., 1998; Crone, et al., 1998; Figure 9). This presumably reflects a competition between autocatalytic processing vs. nucleocytoplasmic transport of capped transcripts (see Benedict, et al., 1998; Crone, et al., 1998), and may alleviate potential problems resulting from physical separation of ribozyme and target within cells (Cotten and Birnstiel, 1989; Sullenger and Cech, 1993). Perhaps even more importantly, the liberated ITRz contains minimal non-specific flanking sequences, which may otherwise significantly hamper Rz catalytic activity in vivo. These reagents appear to provide a number of important advantages over AOs. First, given their catalytic activity, each ITRz could potentially cleave a considerable number of target transcripts in vivo during its life-time. This advantage is apparent in experiments where the catalytically inactive mutant TRzs did not significantly affect cell growth whereas the catalytically active TRz produced marked (and highly significant) growth inhibition. The lag period before effects of the TRz activity seen presumably reflects a build-up in ITRz concentration. Second, the liberated ITRz is not compromised by significant non-specific flanking sequences, which should be expected to hamper antisense transcripts. This may well underlie the obvious discrepancy between in vitro catalytic activities of Rz vs. their in vivo effectiveness (Crisell, et al., 1993). Additionally, use of TRz constructs in suitable expression systems provides continuous production, and is not likely to be complicated by problems with nonspecific toxicity or long-term stability, and verification of their efficiency is straightforward, with destruction of targeted RNAs and reduction in the corresponding proteins (where appropriate). Finally, the constructs described here represent “first generation” constructs. A number of improvements have now been introduced, including: A) Use of 2 contiguous trans-acting ITRz, whose activities do not adversely affect each other; B) Redesign of the cis-acting Rz flanking sequences, which markedly improves the autocatalytic selfliberation activity of the TRz constructs; C) Addition of a short hairpin loop and/or protein-binding domains, to the 3’ end of the double ITRz insert, to improve stability of the insert after liberation; and D) Development of a streamlined library selection procedure, which identifies target sites which yield TRz with 10-103 X greater catalytic activities. A number of these constructs are currently being tested in vivo, with the goal of developing suitable therapeutic reagents for clinical trials.

IV. Materials and Methods A. Cell lines An SV40-immortalized rat hepatocyte c e l l l i n e (Benedict et al., 1995; Woodworth and Isom, 1987), designated CWSV1, was used in these studies. The CWSV1 cells were maintained in chemically defined medium (Woodworth and Isom, 1987).

B. Vectors Vectors used in these experiments included: (i ) pCRII (Invitrogen). This vector contains Sp6/T7 RNA polymerase promoters for bidirectional transcription, and was used to produce RNA transcripts for in vitro studies. (i i ) LacSwitch system (pOPRSVICAT, from Stratagene), which was used for transfection analyses. This system contains a neomycin-resistance gene for antibiotic selection. When used in conjunction with the repressor vector, this system is inducible with IPTG. (i i i ) A tetracycline/doxycycline inducible system, which consists of the pBI-L and pTet-On vectors (Clontech). pTet-On contains the tetracycline transactivator, which is active in the presence of doxycycline, and stably transfected clones expressing pTet-On were obtained by neomycin selection. The TRz constructs were cloned into the multiple cloning site in pBIL. Clones stably transfected with pTet-On were then transfected with the TRz constructs in pBI-L (along with the pTKHyg vector to provide hygromycin resistance), and stably transfected clones were obtained by hygromycin selection. (i v ) The ecdysone-inducible expression system (InVitrogen). This system consists of the pIND expression vector and a vector (pVgRXR) expressing a heterodimeric ecdysone receptor. The heterodimer of the ecdysone receptor (VgEcR, modified to contain the VP16 transactivation domain) and the retinoid X receptor binds a modified ecdysone response element in pIND in the presence of the ecdysone analogs muristerone A or ponasterone A, thereby activating transcription. Stably transfected clones expressing the various ribozyme constructs were developed by antibiotic (zeocin and G418) selection.

C. Ribozyme synthesis The parent double ribozyme (designated pLSClip) was initially synthesized in the LacSwitch vector as previously described (Benedict, et al., 1998). Internal targeted ribozymes (ITRz) were then designed for MCP subunits C3 (EMBL locus TATC3AA, accession number J02897) and C9 (EMBL locus RNPTSC9, accession number X53304). Target sites were selected after analysis of predicted secondary structures generated using the mFold program, which is based upon previous analyses (Jaeger et al., 1989), and database searches were performed to check for collateral target redundancy. For C3, GUU22 was selected, and for C9 GUC 101 was selected. For the C3targeted ITRz, the 5’ and 3’ flanking sequences were UCGAAGCUGU and CCGCGUUGA respectively, and for the C9targeted ITRz, they were UCUUCGA and CAUGGCU, respectively. These ITRz were then synthesized using reverse complementary oligodeoxynucleotides, which were designed to

266

Gene Therapy and Molecular Biology Vol 3, page 267 include pre-cut BglII sites after annealing. Catalytically inactive ITRz were also created (Haseloff and Gerlach, 1988), where the catalytically essential G and A nucleotides were replaced with A and G, respectively. After annealing, product was ligated into BglII digested pLSClip at a 1:5 molar ratio. The identities of the final targeted TRz were verified by sequencing.

G. Transfection experiments For transfection experiments, CWSV1 cells were grown to mid-logarithmic phase and transfected with the various vectors, using either Lipofectin reagent (Gibco) as described (Benedict and Clawson, 1996; Benedict, et al., 1995) or electroporation. For electroporation, cells were resuspended in RPMI + 10% bovine calf serum. Each 0.4-cm gap cuvette contained 0.5 ml (5 x 10 6 cells) and 10 ug vector DNA. We used a BioRad Gene Pulser II, with capacitance 950 µF at 250 V/cm (t=20-25 msec). Transfection analyses using the green fluorescent protein construct (pEGFP-N1, from Clontech) indicated an efficiency of approximately 50% at 24 h. For development of stable transfectants, geneticin (at 0.5 mg/ml, from Sigma) was added after 48 h and selection was continued for approximately 4 weeks. Individual geneticin-resistant colonies were harvested using pipettes. As controls, clones stably transfected with the parent double ribozyme (designated Clip) or the catalytically inactive mutants were also produced. With the pTet-On expression system, cells stably transfected with pTet-On were obtained by geneticin selection. These cells were then transfected with the TRz constructs in pBI-L (along with pTKHyg), and stably transfected clones were obtained by hygromycin selection.

D. In vitro expression of ribozymes For in vitro expression, the targeted TRz were removed from pLSClip using Not I and inserted into the Not I site of the pCRII vector (Invitrogen). Forward and reverse M13 primers were used in PCR protocols to amplify the region of pCRII containing the TRz (and T7 and Sp6 promoters), which was then used as template in vitro , using the Riboprobe Transcription System (Promega) with 50 µCi 32 P-CTP. Based on sequence analysis, T7 or Sp6 RNA polymerase was used to produce transcripts in the sense direction, and the other polymerase was used to produce antisense transcripts. Reactions were incubated at 37o C for 1 or 2 h; they were then digested with DNase, extracted with phenol/chloroform, and RNA was precipitated with ethanol and resuspended in 10 µl H 2 O. RNA was incubated at 80 o C for 5 minutes in an equal volume of loading buffer (80% formamide, 100 mM EDTA, pH 8.0, 0.25% bromophenol blue, 0.25% xylene cyanol FF), and analyzed in a 6% urea/polyacrylamide gels, followed by blotting and autoradiography. In some experiments, unprocessed (full-length) or partially processed transcripts were purified by PAGE. The bands were excised, homogenized in buffer ( 20 mM Tris-HCl, pH 7.6, 250 mM NaCl) and then incubated for 2 h at 4°C and for 5 min at 65°C. Following centrifugation at 2,000 x g for 5 min, the supernate was removed and RNA was precipitated with ethanol. RNA was resuspended and incubated for various periods at 37°C, and then analyzed as described. Zero-time incubations of unprocessed or partially-processed transcripts showed that no processing occurred during the handling procedures.

H. Growth studies For growth studies, cells were generally plated at a density of 5 x 104 cells (or in some instances at 105 cells) and grown on 60 x 15mm tissue culture dishes. Plates were trypsinized and cells were counted in triplicate using a hemocytometer throughout a 78 day period.

I. Northern blotting For Northern blot analyses, RNA was isolated from cells by the guanidinium thiocyanate method (Ausubel et al., 1996). RNA was fractionated by agarose-formaldehyde gel electrophoresis using a 1.2% gel and then transferred to MagnaGraph nylon transfer membrane (MSI) by capillary transfer. Membranes were subsequently irradiated using a Stratalinker UV-Crosslinker (Stratagene) at 24 x 10 4 mjoules. Hybridization was at 42 o C overnight in 5 ml formamide hybridization buffer (Ausubel, et al., 1996), with 5 x 10 6 cpm/ml 32 P-DNA probe, which was labeled using the Prime-a Gene System (Promega), the cloned target inserts described above, and 50 µCi 32 P-dCTP. Following overnight hybridization, blots were washed extensively, with final rinses at 60 o C in 0.1X SSC with 0.1%SDS prior to autoradiography. Before reprobing, blots were stripped with two rinses for 2 h at 70 o C in 60% formamide, 50 mM Tris-HCl (pH 8.0), with 1% SDS, and stripping was verified by autoradiography.

E. Target cleavage reactions For target cleavage reactions, 32 P-labeled target RNAs were prepared as described above, using cloned partial target transcripts contained in pCRII. The C3 target used was 238 nt, yielding products of 146 and 92 nt after Rz cleavage, while the C9 target was 365 nt, giving 156 and 209 nt products after Rz cleavage. Following transcription, DNA template was removed by incubation with RNase-free DNase for 15 min at 37 o C, loading buffer was added, and the RNA was separated by PAGE on 6% polyacrylamide gels. The 32 P-labeled target RNA was then gel-purified as above.

F. Preparation of triple ribozymes (TRz) Unlabeled TRz were transcribed in vitro as above. Following transcription, the liberated ITRz were gel-purified, precipitated with ethanol, and resuspended in 50 µl H 2 O. Labeled target RNA was incubated at 37o C for various times with (or without) Rz in buffer (50 mM Tris-HCl, pH 7.6, 25 mM KCl, 20 mM MgCl2 ). In some cases, residual unprocessed and partially-processed TRz transcripts were also gel-purified and analyzed. The molar ratio of target:ribozyme generally used for standard analyses was 1:1, with final concentrations of about 50 nM. Reactions were terminated with an equal volume of stop buffer and the products were examined using an 6% urea/polyacrylamide gel followed by autoradiography.

J. TRz expression levels and RT/PCR amplifications An RT/PCR protocol was used for assessing TRz expression levels and the extent of self-liberation in vivo. These utilized “inner” and “outer” primer pairs as previously described (Benedict, et al., 1998; Crone, et al., 1998). Briefly, the inner primer pair utilized sites located internal to the autocatalytic cleavage sites (that is, an upstream region just 3’ to the 5’ autocatalytic cleavage site, and a downstream sequence reverse

267

Ren et al: C3- and C9-targeted triple ribozymes complementary to the region just 5’ to the 3’ autocatalytic cleavage site). This inner primer pair amplifies both unprocessed and processed (liberated ITRz) Rz transcripts. The outer primer pair utilizes sequences just external to the autocatalytic cleavage sites (i.e., sequence just 5’ to the 5’ autocatalytic cleavage site, and reverse complementary to the region just 3’ to the 3’ autocatalytic cleavage. The double-G mutants (which cannot undergo autocatalytic processing) were used to establish relative efficiencies of the primer pairs; the inner primer pair was 1.8X more efficient. The inner primer pair amplified a 69 nt product, whereas the outer primer pair amplified a 100 nt product. In some experiments, the 3’ inner primer was used with the 5’ outer primer for amplification of unprocessed TRz transcripts; results were analogous to those with the outer primer pair, producing an 86 nt product.

triple ribozyme targeted to repetitive B2 transcripts. Submitted. Dzianott, A., and Bujarski, J. (1 9 8 9 ) Derivation of an infectious viral RNA by autolytic cleavage of in vitro transcribed viral cDNAs. P r o c . N a t l . A c a d . S c i .(USA) 86, 4823-4827. Goodchild, J., and Kohli, V. (1 9 9 1 ) Ribozymes that cleave an RNA sequence from human immunodeficiency virus: The effect of flanking sequence on rate. Arch. B i o c h e m . B i o p h y s . 284, 386-391. Groebe, D., and Uhlenbeck, O. (1 9 8 8 ) Characterization of RNA hairpin loop stability. N u c l e i c A c i d s R e s . 16, 1172511735. Haseloff, J., and Gerlach, W. (1 9 8 8 ). Simple RNA enzymes with new and highly specific endoribonuclease activities. Nature 334, 585-591.

One of the primers was end-labeled with 32 P using T7 polynucleotide kinase, and 1.5 x 10 5 cpm was used in RT/PCR amplifications with 0.5 µg RNA. Following reactions (generally 25 cycles), products were separated by PAGE in 6% gels, and radioactivity in the bands was quantitated. To establish relative amounts of RNA used in the reactions, a 323 nt portion of 18S rRNA (nt 527-849) was amplified concurrently (15 cycles) and quantitated as above.

Helene, C., and Toulme, J. (1 9 9 0 ) Specific regulation of gene expression by antisense, sense, and antigene nucleic acids. B i o c h i m . B i o p h y s . A c t a 1049, 99-125. Hoke, G., Draper, K., Freier, S., Gonzalez, C., Driver, V., Zounes, M., and Ecker, D. (1 9 9 1 ) Effects of phosphorothioate capping on antisense oligonucleotide stability, hybridization, and antiviral efficacy versus herpes simplex virus infection. N u c l e i c A c i d s R e s . 19, 57435748.

Acknowledgements We wish to thank Dr. H. Isom for providing CWSV1 cells, and for antibodies specific for rat albumin. Supported by PHS grants CA21141 and CA40145 from the NIH/NCI, and funds from Hexal and the Gittlen Cancer Research Institute.

Iverson, P., Zhu, S., Meyer, A., and Zon, G. (1 9 9 2 ) Cellular uptake and subcellular distribution of phosphorothioate oligonucleotides into cultured cells. A n t i s e n s e R e s . D e v e l . 2, 211-222. Izant, J., and Weintraub, H. (1 9 8 5 ) Constitutive and conditional suppression of exogenous and endogenous genes by anti-sense RNA. S c i e n c e 229, 345-352.

References

Jaeger, J., Turner, D., and Zuker, M. (1 9 8 9 ) Improved predictions of secondary structures for RNA. P r o c . N a t l . Acad. Sci. USA 86, 7706-7710.

Altschuler, M., Tritz, R., and Hampel, A. (1 9 9 2 ) A method for generating transcripts with defined 5' and 3' termini by autolytic processing. Gene 122, 85-90.

Kawasaki, H., Ohkawa, J., Tanishige, N., Yoshinari, K., Murata, T., Yokoyama, K., and Taira, K. (1 9 9 6 ) Selection of the best target site for ribozyme-mediated cleavage within a fusion gene for Adenovirus E1A-associated 300 kDa protein (p300) and luciferase. N u c l e i c Acids R e s . 24, 30103016.

Ausubel, F., Brent, R., Kingston, R., Moore, D., Seidman, J., Smith, J., and Struhl, K. (1 9 9 6 ) “ C u r r e n t P r o t o c o l s i n M o l e c u l a r B i o l o g y . ” John Wiley & Sons, New York. Benedict, C., and Clawson, G. (1 9 9 6 ) Nuclear multicatalytic proteinase subunit RRC3 is important for growth regulation in hepatocytes. B i o c h e m i s t r y 35, 11612-11621. Benedict, C., Pan, W., Loy, S., and Clawson, G. (1 9 9 8 ) Triple ribozyme-mediated down-regulation of the retinoblastoma gene. C a r c i n o g e n e s i s 19, 1223-1230.

Koizumi, M., Hayase, Y., Iwai, S., Kamiya, H., Inoue, H., and Ohtsuka, E. (1 9 8 9 ) Design of RNA enzymes distinguishing a single base mutation in RNA. N u c l e i c A c i d s R e s . 17, 7059-7071.

Benedict, C., Ren, L., and Clawson, G. (1 9 9 5 ) Nuclear multicatalytic proteinase _ subunit RRC3: differential size, tyrosine phosphorylation, and susceptibility to antisense oligonucleotide treatment. B i o c h e m i s t r y 34, 9587-9598.

Krieg, A., Yi, A., Matson, S., Waldschmidt, T., Bishop, G., Tesdale, R., Koretzy, G., and Klinman, D. (1 9 9 5 ) CpG motifs in bacterial DNA trigger direct B-cell activation. Nature 374, 546-549.

Christoffersen, R., and Marr, J. (1 9 9 5 ) Ribozymes as human therapeutic agents. J. Med. Chem. 38, 2023-2037.

Lieber, A., and Strauss, M. (1 9 9 5 ) Selection of efficient cleavage sites in target RNAs by using a ribozyme expression library. M o l . C e l l . B i o l . 15, 540-551.

Cotten, M., and Birnstiel, M. (1 9 8 9 ) Ribozyme mediated destruction of RNA in vivo. EMBO J. 8, 3861-3866.

Ohkawa, J., Koguma, T., Kohda, T., and Taira, K. (1 9 9 5 ) Ribozymes: From mechanistic studies to applications in vivo. J . B i o c h e m . 118, 251-258.

Crisell, P., Thompson, S., and James, W. (1 9 9 3 ) Inhibition of HIV-1 replication by ribozymes that show poor activity in vitro. N u c l . A c i d s R e s . 21, 5251-5255.

Ohkawa, J., Yuyama, N., and Taira, K. (1 9 9 2 ) Activities of HIV-RNA targeted ribozymes transcribed from a 'shot-gun' type ribozyme-trimming plasmid. N u c l . Acids S y m p . Ser. 27, 15-16.

Crone, T., Schalles, S., Benedict, C., Pan, W., Ren, L., Loy, S., Isom, H., and Clawson, G. (1 9 9 8 ) Growth Inhibition by a

268

Gene Therapy and Molecular Biology Vol 3, page 269 Ohkawa, J., Yuyama, N., Takebe, Y., Nisikawa, S., Homann, M., Sczakiel, G., and Taira, K. (1 9 9 3 ) Multiple sitespecific cleavage of HIV RNA by transcribed ribozymes from shotgun-type trimming plasmid. N u c l e i c A c i d s S y m p . Ser. 29, 121-122. Parker, R., Muhlrad, D., Deshler, J., Taylor, N., and Rossi, J. (1 9 9 2 ). In “Gene R e g u l a t i o n : B i o l o g y of A n t i s e n s e R N A a n d D N A � (R. Erickson and J. Izant, eds.), pp. 55-70. Raven Press, New York. Perreault, J., Wu, T., Cousineau, S., Ogilvie, K., and Cedergren, R. (1 9 9 0 ) Mixed deoxyribo- and ribo-oligonucleotides with catalytic activity. Nature 344, 565-568. Ruffner, D., Stormo, G., and Uhlenbeck, O. (1 9 9 0 ) Sequence requirements of the hammerhead RNA self-cleavage reaction. B i o c h e m i s t r y 29, 10695-10702. Ruiz, J., Wu, C., Ito, Y., and Wu, G. (1 9 9 7 ) Design and preparation of a multimeric self-cleaving hammerhead ribozyme. BioTechiques 22, 338-345. Stein, C., and Cheng, Y. (1 9 9 3 ) Antisense oligonucleotides as therapeutic agents - is the bullet really magic? S c i e n c e 261, 1004-1012. Storey, A., Oates, D., Banks, L., Crawford, L., and Crook, T. (1 9 9 1 ) Anti-sense phosphorothioate oligonucleotides have both specific and non-specific effects on cells containing human papillomavirus type 16. N u c l e i c A c i d s R e s . 19, 4109-4114. Sullenger, B., and Cech, T. (1 9 9 3 ) Tethering ribozymes to a retroviral packaging signal for destruction of viral RNA. S c i e n c e 2 6 2 , 1566-1569. Taira, K., Nakagawa, K., Nishikawa, S., and Furukawa, K. (1 9 9 1 ) Construction of a novel RNA-transcript-trimming plasmid which can be used both in vitro in place of run-off and (G)-free transcriptions and in vivo as multi-sequences transcription vectors. N u c l e i c A c i d s R e s . 19, 51255130. Wagner, R. (1 9 9 4 ) Gene inhibition using oligodeoxynucleotides. Nature 372, 333-335.

antisense

Woodworth, C., and Isom, H. (1 9 8 7 ) Regulation of albumin gene expression in a series of rat hepatocyte cell lines immortalized by simian virus 40 and maintained in chemically defined medium. M o l . C e l l . B i o l . 7, 37403748. Zamecnik, P., and Stephenson, M. (1 9 7 8 ) Inhibition of Rous sarcoma virus replication and cell transformation by a specific oligodeoxynucleotide. P r o c . N a t l . A c a d . S c i . USA 75, 280-284.

269

Gene Therapy and Molecular Biology Vol 3, page 271 Gene Ther Mol Biol Vol 3, 271-280. August 1999.

Development of hammerhead ribozymes for HIV-1 gene therapy: principles and progress Review Article

A. Ramezani and Sadhna Joshi Department of Medical Genetics and Microbiology, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 3E2, Canada _________________________________________________________________________________________________ Corresponding Author: Sadhna Joshi, Ph.D., 150 College Street, #212, Department of Medical Genetics and Microbiology, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 3E2, Canada. Tel: (416) 978-2499; Fax: (416) 638-1459; E-mail: sadhna.joshi.sukhwal@utoronto.ca Key words: Hammerhead ribozymes, HIV-1 replication, gene therapy, multimeric ribozymes, retroviral vectors, HIV-1 Env Abbreviations: HIV-1, human immunodeficiency virus type-1; FCRs, flanking complementary regions; nts, nucleotides; TAR, trans-activation response; LTR, long terminal repeat; , retroviral RNA packaging signal Received: 18 December 1998; accepted: 22 December 1998

Summary Hammerhead ribozymes are small RNA molecules that can be designed to specifically recognize and cleave a target RNA. A single ribozyme can theoretically act in a catalytic manner thus cleaving more than one molecule of its target RNA. Because of their sequence specificity, ribozymes are being developed as therapeutic agents to eliminate unwanted cellular and viral RNAs. Ribozymes are being used to inhibit human immunodeficiency virus type-1 (HIV-1) replication. Promising results have been obtained by several groups using ribozymes targeted against various sites within the HIV-1 genome. This article reviews some of the factors relevant to the design of hammerhead ribozymes with a particular focus on their application in HIV-1 gene therapy.

applications, the ribozyme must cleave its target RNA in trans and in a catalytic manner. The hammerhead ribozyme could be adopted to cleave in trans by separating the ribozyme and substrate motifs (Feder and Uhlenbeck, 1992; Hasseloff and Gerlach, 1988; Uhlenbeck, 1987). The transcleaving hammerhead ribozyme (Fig. 1) contains a catalytic domain consisting of 11 conserved nts and two flanking complementary regions (FCRs). The catalytic domain consists of 11 nts that are highly conserved. This region also contains a stem-loop. The sequence and length of this stemloop can vary except for the innermost G-C base pair, which is conserved. Target RNA specificity is provided by the FCRs on either side of the catalytic domain. FCRs are designed to contain sequences that are complementary to those surrounding the cleavage site within the target RNA. These FCRs allow the ribozyme to recognize and hybridize with the target RNA. Cleavage of the target RNA occurs 3’ to the NUH sequence (where N is any nucleotide, and H is any nucleotide but G), although not all NUH sites are cleaved equally well (Foster and Symons, 1987).

I. Introduction Ribozymes are small RNA molecules with endoribonuclease activity. They can be designed to cleave RNAs in a sequence-specific manner. Ribozymes base-pair with nucleotides around the cleavage site of the target RNA and catalyze the hydrolysis of a specific phosphodiester bond. There are several different types of ribozymes: the self-splicing group I and II introns (Zaug et al., 1986), RNase P (Guerrier-Takada et al ., 1983), an RNA transcript from Neurospora mitochondrial DNA (Saville and Collins, 1990), the hepatitis delta virus (Wu et al., 1989), the hammerhead ribozyme (Foster and Symons, 1987), and the hairpin ribozyme (Haseloff and Gerlach, 1989). The hammerhead ribozyme is the smallest of the above ribozymes. It was first discovered in small circular pathogenic RNAs such as the satellite RNA of the Tobacco ringspot virus (Buzayan et al., 1986) and the Avocado sunblotch viroid (Hutchins et al., 1986). In its natural setting, the ribozyme motif has evolved to mediate a single turnover reaction in cis. However, to have therapeutic

271

Ramezani and Joshi: Hammerhead ribozymes for HIV-1 gene therapy

Fig. 1. Schematic representation of a transcleaving hammerhead ribozyme. The conserved nucleotides within the catalytic domain are indicated in boldface. The cleavage site within the target RNA is italicized. N represents non-conserved nucleotides within the 5’/3’ FCRs. The arrow points to the site of cleavage.

could decrease the rate of dissociation of the ribozyme from the cleaved target RNA and hence reduce the catalytic activity of the ribozyme.

Structural and functional characteristics have led to the development of hammerhead ribozymes that can cleave in trans. Factors which affect ribozyme-mediated cleavage of cellular or viral RNA have also been investigated. This review discusses factors affecting hammerhead ribozyme activity and provides the progress to date for ribozymemediated inhibition of HIV-1 replication.

Tabler et al. (1994) constructed ribozymes with asymmetric 5’/3’ FCRs against the leader-gag region of HIV-1. These ribozymes contained a long 3’ FCR (289 nts) and a short 5’ FCR which varied in length (0, 1, 2, 3, 5, 8, and 13 nts). In this study, as few as 3 nts in the 5’ FCR were found to be sufficient for effective cleavage of target RNA in vitro. Ribozymes with 3 and 5 nt-long 5’ FCRs were also shown to be effective at inhibiting HIV-1 replication in vivo. Although the 3’ cleavage product was shown to be released in vitro, the 5’ cleavage product could not be dissociated. Therefore, the design of asymetric ribozymes with long FCRs may decrease ribozyme catalytic activity.

II. Ribozyme design A. Flanking complementary regions (FCRs) In vitro studies suggest that the catalytic activity of hammerhead ribozymes with symmetrical FCRs is greatest if the 5’ and 3’ FCRs contain 6 to 8 nts (Fedor and Uhlenbeck, 1990; Goodchild and Kohli, 1991). However, short FCRs may not allow the ribozyme to specifically associate with the target RNA in vivo.

The optimal length of the 3’ FCR to be used with ribozymes containing the 3 nt-long 5’ FCR has been investigated using a set of ribozymes against the HIV-1 tat coding region (Hormes et al., 1997). These ribozymes contained a 3 nt-long 5’ FCR and variable length 3’ FCRs (20, 25, 31, 35, 41, 45, 51, 60, and 195 nts). All ribozymes were shown to be active in vitro. A minimum 51 nt-long 3’ FCR was found to be necessary to inhibit virus replication in a microinjection experiment where HIV-1 provirus DNA and ribozyme RNA were injected into the nucleus of human cells.

The effect of the length of the 5’ and 3’ FCRs on ribozyme activity have been studied and compared both in vitro and in vivo. Crisell et al. (1993) designed and tested a set of ribozymes targeted against the first coding exon of HIV-1 tat. These ribozymes were designed to contain either symmetrical 5’/3’ FCRs containing 9/9, 12/12, 15/15, 18/18, 21/21, 24/24, 27/27, 30/30, and 33/33 nts or asymmetrical 5’/3’ FCRs containing 45/70 and 45/564 nts. Optimum activity in vitro was observed with ribozymes containing 9/9 nt-long 5’/3’ FCRs, whereas the inhibition of HIV-1 replication was greatest with ribozymes containing ! 33/33 nt-long FCRs. Increasing the length of FCRs may have enhanced the ability of the ribozyme to “melt” the secondary structure of the target RNA in vivo. However, long FCRs

B. The catalytic domain and the stem-loop region Since most nucleotides in the single-stranded regions of the catalytic domain are highly conserved, efforts have been 272

Gene Therapy and Molecular Biology Vol 3, page 273 made to modify the stem-loop region and determine its effect on cleavage. While decreasing the length of this stem from the conventional 4 bps to 2 bps did not alter the cleavage activity, its further reduction significantly decreased the cleavage activity in vitro (Tuschl and Eckstein, 1993). Elongating the stem from 4 bps to 6, 10, 21 or 22 bps seemed to have no effect on in vitro cleavage activity of ribozymes, although this resulted in a reduction in their ability to inhibit HIV-1 inhibition (Homann et al., 1994). Therefore, except for the innermost G-C bp that must be conserved, the length or the composition of remaining bps in the stem-loop region does not seem to be critical for ribozyme activity (Tuschl and Eckstein, 1993).

replication defective. However, escape mutants may still arise from mutations within non-conserved regions adjacent to the cleavage site. Such mutations may alter target RNA structure and thereby prevent its accessibility to the ribozyme. To cleave target RNA, a ribozyme must also be able to properly associate with it. However, long RNAs contain complex secondary and tertiary structures involving significant intramolecular base pairing that may reduce the accessibility of a particular target site (Fedor and Uhlenbeck, 1990; Uhlenbeck et al., 1997). Single-stranded regions containing ribozyme target sites should therefore be mapped by enzymatic digestions or chemical modifications (Ehresmann et al., 1987). For example, HIV-1 RNA incubation with oligodeoxyribonucleotides containing potential ribozyme FCRs followed by RNase H digestion could locate target sites within the conserved regions that are likely to be accessible for ribozyme binding (Ho et al., 1996; Scherr and Rossi, 1998). It may be desirable to perform these analyses on the full-length target RNA similar to what is to be cleaved in vivo. Computer programs may also be used to predict the most probable RNA secondary structures with minimum free energy parameters (Jaeger et al., 1989), although the reliability of computer prediction for very large RNAs is still not high. Once the information on the target RNA secondary structure is available, target sites should be selected within the single-stranded regions that are accessible and highly conserved (Zhao and Lemake, 1998).

III. Target site selection The choice of HIV-1 RNA that must be targeted and cleaved is critical as inactivation of not all HIV-1 RNAs would have the same impact on virus replication. Furthermore, within any given HIV-1 RNA only conserved and accessible regions should be targeted. Finally, although hammerhead ribozymes could be designed to cleave HIV-1 RNA at any NUH cleavage site, the most efficiently cleaved sites should be targeted.

A. Target RNA HIV-1 provirus DNA transcription gives rise to over 20 distinct mRNA species in an infected host cell (Fig. 2). Although it has been reported that the unspliced HIV-1 RNA in the nucleus is the primary target of ribozymes (Paik et al., 1997), ribozyme activity may not be exclusively limited to the nucleus and may also occur within the cytoplasm. Ribozymes were also shown to be very effective at cleaving RNA within the progeny virus (Sullenger and Cech, 1993; Westaway et al., 1998). The choice of the target site within HIV-1 RNA only matters for cleavage of spliced mRNAs, as all sites are present within the unspliced HIV-1 RNA. Thus, to allow cleavage of all HIV-1 RNAs within the cell and within the progeny virus, ribozymes may be targeted against regions that are common to all HIV-1 RNAs (Joshi and Joshi, 1996).

C. Cleavage site With the exception of the AUA-cleaving satellite RNA of the barley yellow dwarf virus (Miller et al., 1991) and the GUA-cleaving lucerne transient streak virus (Foster and Symons, 1987), all naturally occurring hammerhead ribozymes cleave their target RNA at a GUC site (Bruening, 1990). Mutagenesis studies have been performed to determine cleavage sites that are best cleaved by the hammerhead ribozyme (Sheldon and Symons, 1989; Ruffner et al., 1990; Perriman et al ., 1992). Although initial studies led to the development of the general NUH rule (Koizumi et al., 1989), detailed kinetic analyses of a target RNA with all possible mutations at the cleavage site (Zoumadakis and Tabler, 1995) identified GUC as the most efficiently cleaved site. The influence of bases surrounding the cleavage triplet was demonstrated by Clouet-d’Orval and Uhlenbeck (1997) who analyzed a hammerhead ribozyme with a 10-fold higher cleavage rate than what was previously reported for hammerhead ribozymes. Mutational analyses demonstrated that the increased cleavage rate was due to the presence of an AU immediately after the GUC cleavage site within the target RNA.

B. Target site The most important factor that must be considered while selecting a ribozyme target site within HIV-1 RNA is the genetic variability. Fortunately, there are numerous regions throughout the viral genome that are highly conserved in all HIV-1 isolates within a given subtype. Some of these sites are also conserved within various subtypes of HIV-1. The ribozyme target sites should be selected from these highly conserved regions. This would decrease the emergence of escape mutants since mutations within these conserved regions are likely to be genetically attenuated and/or

Many attempts have been made to compare and identify ribozyme target sites within HIV-1 RNA that are best cleaved in vitro or that inhibit HIV-1 replication most

273

Ramezani and Joshi: Hammerhead ribozymes for HIV-1 gene therapy

Fig. 2. Ribozyme interference sites within the HIV-1 life cycle. Following entry inside the cell, HIV-1 RNA reverse transcribes and integrates within the cellular genome. Upon transcription, the full length 9.3 kb viral RNA is produced, which is differentially spliced to give rise to various HIV-1 mRNAs. The 2 kb RNAs then give rise to Tat which enhances gene expression and Rev which allows export of 4-5 and 9.3 kb HIV-1 RNAs. Translation of these RNAs then gives rise to various structural and maturation proteins. Virus assembly then takes place and recruits 2 copies of full length HIV RNA and cellular tRNA3Lys. Some of the steps taking place during the subsequent round of infection are also shown. Ribozymes ( ) may cleave HIV-1 RNA in the nucleus, cytoplasm, or progeny virus such that either no virus will be produced or virus produced will be non-infectious.

274

Gene Therapy and Molecular Biology Vol 3, page 275 Table 1 Summary of HIV-1 inhibition results obtained using monomeric hammerhead ribozymes. Target site

Target cell (pool/clone)

Expression

HIV-1 replication (compared to controls)

Reference

R

HeLa CD4/pool

Transient

Suppressed for 9 days

Dropulic and Jeang, 1994

U5

T-cell line/pool T-cell line/pool HeLa CD4/pool

Stable Stable Transient

Delayed for 18 days Suppressed for 5-7 days Suppressed for 10 days

Weerasinghe et al., 1991 Dropulic et al., 1992 Dropulic et al., 1992

T-cell line/clone

Stable

Suppressed for 12 days

Sun et al., 1994

RRE

HeLa CD4/pool

Transient

Suppressed for 6-9 days

Dropulic and Jeang, 1994

gag

HeLa CD4/pool T-cell line/pool

Transient Stable

Suppressed for 7 days Delayed for 9 days

Sarver et al., 1990 Ramezani and Joshi, 1996

pro

T-cell line/pool

Stable

Delayed for 15 days

Ramezani and Joshi, 1996

RT

T-cell line/pool

Stable

Delayed for 9 days

Ramezani and Joshi, 1996

tat

T-cell line/clone T-cell line/pool T-cell line/pool T-cell line/pool T-cell line/pool

Stable Stable Stable Stable Stable

delayed for 8 days Delayed for 6 days Delayed for 6-8 days Suppressed for 12 days Suppressed for 9 days

Lo et al., 1992 Crisell et al., 1993 Zhou et al., 1994 Sun et al., 1995 Wang et al., 1998

rev

T-cell line/clone T-cell line/pool

Stable Stable

Suppressed for 18 days Delayed for 6-8 days

Michienzi et al., 1998 Zhou et al., 1994

tat and rev

LTBMC/pool T-cell line/pool

Stable Stable

Suppressed Delayed for 6-8 days

Bauer et al., 1997 Zhou et al., 1994

env

T-cell line/pool

Stable

Delayed for 18 days

Ramezani et al., 1996

nef

T-cell line/clone

Stable

Delayed up to 14 days

Larsson et al., 1996

LTBMC: long-term bone marrow culture.

vectors are unable to transduce non-dividing cells such as hematopoietic stem cells, lentiviral vectors have recently been developed and successfully used to deliver genes into non-dividing cells (Naldini et al., 1996; Uchida et al., 1998). Among the cells transduced with HIV-1 based vectors are the human hematopoietic stem cells (Uchida et al., 1998), macrophages (Corbeau et al., 1998), and terminally differentiated neurons (Naldini et al., 1996). The restricted host range of HIV-1, which is limited to CD4+ cells, could be extended using the amphotropic envelope protein from the Moloney murine leukemia virus (MoMuLV) or the G protein from vesicular stomatitis virus (Naldini et al., 1996). The later envelope is also more stable and allows ultracentrifugal concentration of virions to high titers (Naldini et al., 1996; Reiser et al., 1996).

efficiently in vivo (Table 1). While targeting some sites has been found to be more effective at inhibiting virus replication than others, the in vitro cleavage results could not always be correlated with the results obtained in vivo (Dropulic and Jeang, 1994; Ramezani and Joshi, 1996). Even ribozymes with poor in vitro cleavage activities have been shown to significantly inhibit HIV-1 replication (Crisell et al., 1993; Ramezani and Joshi, 1996).

IV. Ribozyme delivery, expression, and localization A. Ribozyme delivery and expression vectors Retroviral vectors are commonly used for the delivery and expression of genes (Friedman, 1989). Since these

275

Ramezani and Joshi: Hammerhead ribozymes for HIV-1 gene therapy Table 2 In vitro selection studies aimed at improving/altering the cleavage activity of various ribozymes. Ribozyme

# of nts mutated/ mutation rate

Results (approximate improvement compared to unselected RNA)

Reference

Tetrahymena group I intron

140/5% per position

100-fold improved DNA-cleaving activity

Beaudry and Joyce, 1992

Tetrahymena group I intron

140/5% per position

170-fold improved catalytic activity utilizing an altered metal cation

Lehman and Joyce, 1993

RNase P

9/random

30-fold improved catalytic activity

Yuan and Altman, 1994

Hairpin

50/3 mutations per molecule

20-fold improved trans-cleavage activity

Joseph and Burket, 1993

Hammerhead

14/random

Consensus activity

Ishizaka et al., 1995

Hammerhead

4/random

Less efficient than consensus

Thomson et al., 1996

Hammerhead

10/random

Less efficient than consensus

Vaish et al., 1997

the ability of the ribozyme to better associate with the target RNA. Nuclear extracts have been shown to improve the hybridization of complementary RNAs (Portman and Dreyfuss, 1994) and ribozyme activity in vitro (Bertrand and Rossi, 1994; Heidenreich et al., 1995).

Ribozyme genes are expressed form the retroviral long terminal repeat (LTR) promoter and/or from internal promoters. However, internal promoters often function poorly when inserted downstream of LTR promoters due to transcriptional interference between promoters (Emerman and Temin, 1984). Optimum functional expression of ribozyme genes has been shown to occur when ribozymes are expressed as part of the long viral RNAs transcribed form the 5’ LTR promoter by RNA polymerase (pol) II, rather than as part of transcripts produced form internal pol II (CMV, U1 snRNA) or pol III (tRNA, U6 snRNA) promoters (Zhou et al., 1996; Bertrand et al., 1997). High level expression was obtained from pol III promoters when cloned within the 3’ LTR (Ilves et al., 1996). Upon reverse transcription, this design also resulted in gene duplication within both the 5’ and 3’ LTRs.

Alternatively, ribozymes could be designed so that they would be co-packaged with HIV-1 virion RNA. Cleavage of HIV-1 RNA within the progeny virus should prevent subsequent viral spread. Sullenger and Cech (1993) used the retroviral RNA packaging signal (") to develop a packagable ribozyme and demonstrated the feasibility of this approach in a MoMuLV-based system. A chimeric tRNA3Lys-ribozyme was also developed and shown to be packaged by HIV-1 (Westaway et al., 1995; 1998). As expected, infectivity of the progeny virus was shown to be significantly decreased.

B. Ribozyme localization

V. In vitro selection of ribozymes

Proper co-localization of the ribozyme with its target RNA is essential for ribozyme activity. Pol III-driven human tRNA or U6 snRNA as well as pol II-driven U1 snRNA have been shown to allow nuclear localization of ribozymes, while pol II promoters such as the Rous sarcoma virus LTR promoter mainly localized the ribozyme to the cytoplasm (Bertrand et al., 1997).

In vitro selection involves isolation of molecules with a desired phenotype (i.e. altered or improved catalytic activity) from a pool of partially or randomly mutated molecules. This strategy has been successfully applied (Table 2) to RNase P (Yuan and Altman, 1994), the hairpin ribozyme (Joseph and Burke, 1993), and the Tetrahymena group I ribozyme (Lehman and Joyce, 1993; Beaudry and Joyce, 1992; Tsang and Joyce, 1994). Larger ribozymes, such as the Tetrahymena group I intron, appear to have a very high potential for improvement. In contrast, it seems difficult to further improve upon the catalytic activity of smaller ribozymes, like the hammerhead ribozyme. Many unsuccessful attempts have been reported for selecting hammerhead ribozymes with improved catalytic activities (Nakamaye and Eckstein, 1994; Long and Uhlenbeck, 1994;

Anti-HIV ribozymes may be localized to the cytoplasm to inactivate the incoming virion RNA before it is reversetranscribed. However, hammerhead ribozymes expressed under the control of the pol II promoter were not found to inhibit the incoming HIV-1 RNA (Ramezani et al., 1997; Paik et al., 1997). In contrast, hammerhead ribozymes have been shown to cleave HIV-1 transcripts in the nucleus (Paik et al., 1997). Enhanced ribozyme activity in the nucleus may be due to certain RNA-protein interactions which increase

276

Gene Therapy and Molecular Biology Vol 3, page 277 Ishizaka et al., 1995; Vaish et al., 1997; Thomson et al., 1996; Ramezani and Joshi, unpublished results).

due to escape virus production, demonstrating the inability of monomeric ribozymes to completely inhibit virus replication.

Long and Uhlenbeck (1994) replaced the stem-loop region within the catalytic domain of a hammerhead ribozyme with 4 or 6 random nts and applied 3 rounds of in vitro selection to isolate active ribozymes. However, all selected ribozymes displayed poor catalytic activity. The most effective of these ribozymes contained 6 nts with a tetranucleotide loop and the consensus G-C base pair next to the single stranded regions of the catalytic domain.

B. Multimeric ribozymes The limited success at inhibiting HIV-1 replication using the existing monomeric ribozymes calls for new strategies to further improve on the intracellular activity of ribozymes. The efficiency of the ribozyme is largely dictated by the accessibility of the target site. However, it is unlikely that any given site will be available for cleavage at all times as RNA constantly folds and unfolds and is often masked by various cellular or viral proteins. Targeting various sites within a given RNA would significantly increase the possibility of having at least one site cleaved, which is sufficient for inactivation of the target RNA. Targeting multiple sites within HIV-1 RNA should also reduce the chance of developing escape mutants in clinical settings.

Thomson et al. (1996) replaced the consensus GAA sequence within the catalytic domain with 4 random nts and used 2 rounds of in vitro selection to isolate ribozymes with an improved catalytic activity. However, the most active ribozymes that could be selected contained the sequence HGAA (where H is A, C, or U) instead of GAA and possessed a reduced catalytic activity. Ishizaka et al. (1995) randomized all of the conserved nucleotides in the catalytic domain of a minizyme lacking the stem-loop region, and conducted in vitro selection to isolate active ribozymes. However, the ribozymes selected after 7 to 8 rounds of selection contained consensus sequences. Similar results were also reported by Vaish et al. (1997), suggesting that the naturally occurring ribozymes might have already evolved to optimum or near optimum levels.

Multimeric ribozymes may be designed to contain ribozymes targeted against various sites within the same or different HIV RNA molecules. Multitarget ribozymes could even be designed to target various subtypes of HIV-1 and/or HIV-2, which would potentially confer simultaneous resistance. Multimeric ribozymes may be expressed in tandem as part of a single RNA molecule or could be flanked by cisacting ribozymes such that individual ribozymes will be liberated (Chowrira et al., 1994; He et al., 1993; Ventura et al., 1993; Yuyama et al., 1992, 1994). Self-liberating multimeric ribozymes have been shown to be very effective at cleaving their target RNA in vitro, although the efficacy of these ribozymes is yet to be demonstrated in vivo (Ohkawa et al., 1993). These ribozymes may be more effective at inhibiting virus replication because of their smaller size and their ability to act simultaneously. However, issues related to the stability and compartmentalization of individual ribozymes remain to be addressed. Furthermore, the use of cis-acting ribozymes limits the application of this strategy as it can not make use of retroviral vectors for efficient gene delivery.

VI. The efficacy of anti-HIV-1 ribozymes A. Monomeric ribozymes A variety of sites within HIV-1 RNA have been targeted by monomeric ribozymes (Table 1). These include coding regions such as gag (Sarver et al., 1990; Ramezani and Joshi, 1996), pol (Dropuic and Jeang, 1994), pro (Ramezani and Joshi, 1996), RT (Ramezani and Joshi, 1996), tat (Lo et al. 1992; Crisell et al., 1993; Zhou et al., 1994; Sun et al., 1995; Ramezani and Joshi, 1996, Wang et al. 1998), tat/rev (Zhou et al., 1994), env (Ramezani and Joshi, 1996), and nef (Larsson et al., 1996), as well as non-coding regions such as the repeat (R) region (Dropulic and Jeang, 1994), the transactivation response (TAR) element (Ventura et al., 1994), the unique 5’ (U5) region (Weeraninghe et al., 1991; Dropulic et al., 1992, Westaway et al., 1995), the " region (Sun et al., 1994), and the Rev response element (RRE) (Dropulic and Jeang, 1994).

Chen et al. (1992) constructed several multimeric ribozymes (mono-, di-, tetra-, penta-, and nonameric ribozymes) targeted against various sites within the env coding region of HIV-1 RNA. In co-transfection experiments, these ribozymes were able to confer significant inhibition of virus replication. Paik et al. (1997) used a defective HIV-1 DNA to express the nonameric ribozyme in HeLa T4 cells. Compared to control cells, which were highly permissive for HIV-1 and HIV-2 replication, only HIV-1 replication was inhibited in multimeric ribozyme-expressing cells.

While the above studies have all demonstrated varying degrees of virus inhibition, there has not yet been any report of complete inhibition of virus replication using monomeric ribozymes. Virus replication eventually resumed despite the use of improved expression vectors or co-localization strategies. Partial inhibition was observed even in an extremely exaggerated experimental setting where a ribozyme was expressed as part of the HIV-1 RNA (Dropulic et al., 1992; Dropuic and Jeang, 1994). Interestingly, the break-through of virus production was not

We developed retroviral vectors expressing mono- or nonameric ribozymes against conserved sites within the HIV-1 Env RNA and compared their anti-HIV-1 efficacy in

277

Ramezani and Joshi: Hammerhead ribozymes for HIV-1 gene therapy a stably transduced T-cell line (Ramezani et al., 1997). The monomeric ribozymes could only delay virus replication, whereas almost no virus production could be detected from nonameric ribozyme-expressing cells for up to sixty days (the length of the experiment). We have also tested the mono- and nonameric ribozyme expression vectors in human peripheral blood lymphocytes (PBLs). When challenged with a primary, patient-derived isolate of HIV-1 (Ramezani and Joshi, 1998), the nonameric ribozyme-expressing PBLs were shown to delay virus replication more than the monomeric ribozyme-expressing PBLs.

HIV-1 replication – potential effectiveness against most presently sequenced HIV-1 isolates. Nucleic Acids Res 20, 4581-4589 Chowrira, B. M., Pavco, P. A., and McSwiggen, J. A. (1994). In vitro and in vivo comparison of hammerhead, hairpin, and hepatitis delta virus self-processing ribozyme cassettes. J Biol Chem 269, 25856-25864. Clouet-d’Orval, B., and Uhlenbeck, O. C. (1997). Hammerhead ribozymes with a faster cleavage rate. Biochemistry 36, 90879092. Corbeau, P., Kraus, G., and Wong-Staal, F. (1998). Transduction of human macrophages using a stable HIV-1/HIV-2-derivedgene delivery system. Gene Ther 5, 99-104.

VII. Prospects

Crisell, P., Thompson, S., and James, W. (1993). Inhibition of HIV1 replication by ribozymes that show poor activity in vitro. Nucleic Acids Res 21, 5251-5255.

The hammerhead ribozyme offers a potentially effective means of inhibiting HIV-1 replication. Although the catalytic activity of the hammerhead ribozyme might already be close to optimal, a number of other factors affecting ribozyme design and activity could improve ribozyme activity in vivo. Promising results have been obtained using multimeric ribozymes. Combination strategies could be used to achieve even better inhibition of HIV-1 replication.

Dropulic, B., and Jeang, K-T. (1994). Intracellular susceptibility to ribozymes in a tethered substrate-ribozyme provirus model is not predicted by secondary structure of HIV-1 RNAs in vitro . Antisense Nucleic Acid Drug Dev 4, 217-221. Dropulic, B., Lin, N. H., Martin, M. A., and Jeang, K-T. (1992). Functional characterization of a U5 ribozyme, Intracellular suppression of HIV-1 expression. J Virol 66, 1432-1441. Ehresmann, C., Baudin, F., Mougel, M., Romby, P., Ebel, J. P., and Ehresmann, B. (1987). Probing the structure of RNAs in solution. Nucleic Acids Res 15, 9109-9128.

Acknowledgments This work is supported by a grant from the Medical Research Council of Canada.

Emerman, M., and Temin, H. (1984). Genes with promoters in retrovirus vectors can be independently suppressed by an epigenetic mechanism. Cell 39, 459-467. Fedor, M. J., and Uhlenbeck O. C. (1990). Substrate sequence effects on hammerhead RNA catalytic efficiency. Proc Natl Acad Sci USA 87, 1668-1672.

References Bauer, G., Valdez, P., Kearns, K., Bahner, I., Wen, S. F., Zaia, J. A., Kohn, D. B. (1997). Inhibition of HIV-1 replication after transduction of granulocyte colony-stimulating factormobilized CD34+ cells from HIV-1-infected donors using retroviral vectors containing anti-HIV-1 genes. Blood 89, 2259-2267.

Fedor, M. J., and Uhlenbeck, O. C. (1992). Kinetics of intermolecular cleavage by hammerhead ribozymes. Biochemistry 31, 12042-12054. Foster, A. C., and Symons, R. H. (1987). Self-cleavage of plus and minus RNAs of a virusoid and a structural model for the active sites. Cell 49, 211-220.

Beaudry, A. A., and Joyce, G. F. (1992). Directed evolution of an RNA enzyme. Science 257, 635-641.

Friedman, T. (1989). Progress toward human gene therapy. Science 244, 1275-1292.

Bertrand, E., and Rossi, J. J. (1994). Facilitation of hammerhead ribozyme catalysis by the nucleocapsid protein of HIV-1 and the heterogeneous nuclear ribonucleoprotein A1. EMBO J 13, 2904-2912.

Goodchild, J., and Kohli, V. (1991). Ribozymes that cleave an RNA sequence form HIV, The effect of flanking sequence on rate. Arch Biochem Biophys 284, 386-391.

Bertrand, E., Castanotto, D., Zhou, C., Carbonnelle, C., Lee, N. S., Good, P., Chatterjee, S., Grange, T., Pictet, R., Kohn, D., Engelke, D., and Rossi, J. J. (1997). The expression cassette determines the functional activity of ribozymes in mammalian cells by controlling their intracellular localization. RNA 3, 7588.

Guerrier-Takada, C., Gardiner, k., Marsh, T., Pace, N., and Altman, S. (1983). The RNA moiety of ribonuclease P is the catalytic subunit of the enzyme. Cell 35, 849-857. Haseloff, J., and Gerlach, W. L. (1988). Simple RNA enzymes with new and highly specific endoribonuclease activities. Nature 334, 585-591.

Bruening, G. (1990). Compilation of self-cleaving sequences from plant virus satellite RNAs and other sources. Methods Enzymol. 180, 546-558.

Haseloff, J., and Gerlach, W. L. (1989). Sequences required for self-catalyzed cleavage of the satellite RNA of tobacco ringspot virus. Gene 82, 43-51.

Buzayan, J. M., Gerlach, W. L., and Bruening, G. (1986). Nonenzymatic cleavage and ligation of RNAs complementary to a plant virus satellite RNA. Nature 326, 349-353.

He, Y. K., Lu, C. D., and Qi, G. R. (1993). In vitro cleavage of HPV 16 E6 and E7 RNA fragments by synthetic ribozymes and transcribed ribozymes from RNA-trimming plasmids. FEBS Lett 322, 21-24.

Chen, C. J., Banerjee, A. C., Harmison, G. G., Haglund, K., and Schubert, M. (1992). Multitarget-ribozyme directed to cleave at up to nine highly conserved HIV-1 env RNA regions inhibits

278

Gene Therapy and Molecular Biology Vol 3, page 279 Heidenreich, O., Kang, S. H., Brown, D. A., Xu, X., Swiderski, P., Rossi, J. J., Eckstein, F., and Nerenberg, M. (1995). Ribozymemediated RNA degradation in nuclei suspension. Nucleic Acids Res 23, 2223-2228.

chimeric anti-HIV ribozymes in a human T lymphoblastoid cell line. Hum Gene Ther 9, 621-628. Miller, W.A., Hercus, T., Waterhouse, P. M., and Gerlach, W. L. (1991). A satellite RNA of barley yellow dwarf virus contains a novel hammerhead structure in the self-cleavage domain. Virology 183, 711-720.

Ho, S. P., Britton, D. H., Stone, B. A., Behrens, D. L., Leffet, L. M., Hobbs, F. W., Miller, J. A., and Trainor, G. L. (1996). Potent antisense oligonucleotides to the human multidrug resistance-1 mRNA are rationally selected by mapping RNAaccessible sites with oligonuceotide libraries. Nucleic Acids Res 24, 1901-1907.

Nakamaye, K. L., and Eckstein, F. (1994). AUA-cleaving hammerhead ribozymes: Attempted selection for improved cleavage. Biochemistry 33, 1271-1277. Naldini, L., Blomer, U., Gallay, P., Ory, D., Mulligan, R., Gage, F. H., Verma, I. M., and Trono, D. (1996). In vivo gene delivery and stable transduction of non-dividing cells by a lentiviral vector. Science 272, 263-267.

Homann, M., Tabler, M., Tzortzakaki, S., and Sczakiel, G. (1994). Extension of helix II of an HIV-1-directed hammerhead ribozyme with ling antisense flanks does not alter kinetic parameters in vitro but causes loss of the inhibitory potential in living cells. Nucleic Acids Res 22, 3951-3957.

Ohkawa, J., Yuyama, N., Takebe, Y., Nishikawa, S., and Taira, K. (1993). Importance of independence in ribozyme reactions: Kinetic behavior of trimmed and simply connected ribozymes with potential activity against HIV. Proc Natl Acad Sci USA 90, 11302-11306.

Hormes, R., Homann, M., Oelze, I., Marschall, P., Tabler, M., Eckstein, F., and Sczakiel, G. (1997). The subcellular localization and length of hammerhead ribozymes determine efficacy in human cells. Nucleic Acids Res 25, 769-775.

Paik, S-Y., Banerjea, A., Chen, C-J., Ye, Z., Harmison, G. G., and Schubert, M. (1997). Defective HIV-1 provirus encoding a multitarget-ribozyme inhibits accumulation of spliced and unspliced HIV-1 mRNAs, reduces infectivity of viral progeny, and protects the cells from pathogenesis. Hum Gene Ther 8, 1115-1124.

Hutchins, C. J., Rathjen, P. D., Forster, A. C., and Symons, R. H. (1986). Self-cleavage of plus and minus RNA transcripts of avocado sunblotch viroid. Nucleic Acids Res 14, 3627-3640 Ilves, H., Barske, C., Junker, U., Bohnlein, E., and Veres, G. (1996). Retroviral vectors designed for targeted expression of RNA polymerase III-driven transcripts, a comparative study. Gene 171, 203-208.

Perriman, R., Delves, A., and Gerlach, W. L. (1992). Extended target-site specificity for a hammerhead ribozyme. Gene 113, 157-163.

Ishizaka, M., Ohshima, Y., and Tani, T. (1995). Isolation of active ribozymes from an RNA pool of random sequences using an anchored substrate RNA. Biochem Biophys Res Commun 214, 403-409.

Portman, D. S., and Dreyfuss, G. (1994). RNA annealing activities in HeLa nuclei. EMBO J 13, 213-221. Ramezani, A., and Joshi, S. (1998). Inhibition of HIV-1 replication in human peripheral blood lymphocytes by retroviral vectors expressing monomeric and multimeric hammerhead ribozymes. 12t h World AIDS Conference 435-439.

Jaeger, J. A., Turner, D. H., and Zuker, M. (1989). Improved predictions of secondary structures of RNA. Proc Natl Acad Sci USA 86, 7706-7710. Joseph, S., and Burke, J. M. (1993). Optimization of an anti-HIV hairpin ribozyme by in vitro selection. J Biol Chem 268, 24515-24518.

Ramezani, A., and Joshi, S. (1996). Comparative analysis of five highly conserved target sites within the HIV-1 RNA for their susceptibility to hammerhead ribozyme-mediated cleavage in vitro and in vivo. Antisense Nucleic Acid Drug Dev 6, 229-235.

Joshi, S., and Joshi, R. L. (1996). Molecular biology of HIV-1. Transfus Sci 17, 351-378.

Ramezani, A., Ding, S. F., and Joshi, S. (1997). Inhibition of HIV-1 replication by retroviral vectors expressing mono- and multimeric hammerhead ribozymes. Gene Ther 4, 861-867.

Koizumi, M., Hayase, Y., Iwai, S., Kamiya, H., Inoue, H., and Ohtsuka, E. (1989). Design of RNA enzymes distinguishing a single base mutation in RNA. Nucleic Acids Res 17, 70597071.

Reiser, J., Harmison, G., Kluepfel-Stahl, S., Brady, R. O., Karlsson, S., and Schubert, M. (1996). Transduction of non-dividing cells using pseudotyped defective high titre HIV-1 particles. Proc Natl Acad Sci USA 93, 15266-15271.

Larsson, S., Hotchkiss, G., Su, J., Kebede, T., Andang, M., Nyholm, T., Johansson, B., Sonnerborg, A., Vahlne, A., Britton, S., and Ahrlund-Richter, L. (1996). A novel ribozyme target site located in the HIV-1 nef open reading frame. Virology 219, 161-169.

Ruffner, D. E., Stormo, G. D., and Uhlenbeck, O. C. (1990). Sequence requirements of the hammerhead RNA self-cleavage reaction. Biochemistry 29, 10695-10702.

Lehman, N., and Joyce, G. F. (1993). Evolution in vitro of an RNA enzyme with altered metal dependence. Nature 361, 182-185.

Sarver, N., Cantin, E. M., Chang, P. S., Zaia, J. A., Ladne, P. A., Stephens, D. A., and Rossi, J. J. (1990). Ribozymes as potential anti-HIV-1 therapeutic agents. Science 247, 1222-1225.

Lo, K. M. S., Biasolo, M. A., Dehni, G., Palu, G., and Haseltine, W. A. (1992). Inhibition of replication of HIV-1 by retroviral vectors expressing tat-antisense and anti-tat ribozyme RNA. Virology 190, 176-183.

Saville, B. J., and Collins, R. A. (1990). A site-specific selfcleavage reaction performed by a novel RNA I Neurospora mitochondria. Cell 61, 685.

Long, D. M., and Uhlenbeck, O. C. (1994). Kinetic characterization of intramolecular and intermolecular hammerhead RNAs with stem II deletions. Proc Natl Acad Sci USA 91, 6977-6981.

Scherr, M., and Rossi, J. J. (1998). Rapid determination and quantitation of the accessibility to native RNAs by antisense oligodeoxynucleotides in murine cell extracts. Nucleic Acids Res 26, 5079-5085.

Michienzi, A., Conti, L., Varano, B., Prislei, S., Gessani, S., and Bozzoni, I. (1998). Inhibition of HIV-1 replication by nuclear

279

Ramezani and Joshi: Hammerhead ribozymes for HIV-1 gene therapy Sheldon, C. C., and Symons, R. H. (1989). Mutagenesis analysis of a self-cleaving RNA. Nucleic Acids Res 17, 5679-5685. Sullenger, B. A., and Cech, T. R. (1993). Tethering ribozymes to a retroviral packaging signal for destruction of viral RNA. Science 262, 1566-1569.

Weerasinghe, M., Liem, S. E., Asad, S., Read, S. E., and Joshi, S. (1991). Resistance to HIV-1 infection in human CD4+ lymphocyte-derived cell lines using retroviral vectors expressing an HIV-1 RNA specific ribozyme. J Virol 65, 5531-5534.

Sun, L-Q., Wang, L., Gerlach, W. L., and Symonds, G. (1995). Target sequence-specific inhibition of HIV-1 replication by ribozymes directed to tat RNA. Nucleic Acids Res 23, 29092913.

Westaway, S. K., Larson, G. P., Li, S., Zaia, J. A., and Rossi, J. J. (1995). A chimeric tRNA3Lys-ribozyme inhibits HIV replication following virion assembly. Nucleic Acids Symp Ser 33, 194199.

Sun, L-Q., Warrilow, D., Wang, L., Witherington, C., and Macpherson, J. (1994). Ribozyme-mediated suppression of MoMuLV and HIV-1 replication in permissive cell lines. Proc Natl Acad Sci USA 91, 9715-9719.

Westaway, S. K., Cagnon, L., Chang, Z., Li, S., Li, H., Larson, G. P., Zaia, J. A., and Rossi, J. J. (1998). Virion encapsidation of tRNA3Lys-ribozyme chimeric RNAs inhibits HIV infection. Antisense Nucleic Acid Drug Dev 8, 185-197.

Tabler, M., Homann, M., Tzortzakaki, S., and Sczakiel, G. (1994). A three-nucleotide helix I is sufficient for full activity of a hammerhead ribozyme: advantages of an asymmetric design. Nucleic Acids Res 22, 3958-3965.

Wu, H. N., Lin, Y. J., Lin, F. P., Makino, S., Chang, M. F., and Lai, M. M. (1989). Human hepatitis delta virus RNA subfragments contain an autocleavage activity. Proc Natl Acad Sci USA 86, 1831-1835.

Thomson, J. B., Sigurdsson, S. T., Zeuch, A., and Eckstein, F. (1996). In vitro selection of hammerhead ribozymes containing a bulged nucleotide in stem II. Nucleic Acids Res 24, 44014406.

Yuan, Y., and Altman, S. (1994). Selection of guide sequences that direct efficient cleavage of mRNA by human ribonuclease P. Science 263, 1269-1273. Yuyama, N., Ohkawa, J., Koguma, T., Shirai, M., and Taira, K. (1994). A multifunctional expression vector for an anti-HIV-1 ribozyme that produces a 5’-and 3’-trimmed trans-acting ribozyme, targeted against HIV-1 RNA, and cis-acting ribozymes that are designed to bind to and thereby sequester trans-activator proteins such as Tat and Rev. Nucleic Acids Res 22, 5060-5067.

Tsang, J. and Joyce, G. F. (1994). Evolutionary optimization of the catalytic properties of a DNA-cleaving ribozyme. Biochemistry 33, 5966-5973. Tuschl, T., and Eckstein, F. (1993). Hammerhead ribozymes: importance of stem-loop II for activity. Proc Natl Acad Sci USA 90, 6991-6994.

Yuyama, N., Ohkawa, J., Inokuchi, Y., Shirai, M., Sato, A., Nishikawa, S., and Taira, K. (1992). Construction of a tRNAembedded-ribozyme trimming plasmid. Biochem Biophys Res Commun 186, 1271-1279.

Uchida, N., Sutton, R. E., Friera, A. M., He, D., Reitsma, M. J., Chang, W. C., Veres, G., Scollay, R., and Weissman, I. L. (1998). HIV, but not murine leukemia virus, vectors mediate high efficiency gene transfer into freshly isolated G0/G1 human hematopoietic stem cells. Proc Natl Acad Sci USA 95, 11939-11944.

Zaug, A. J., Been, M. D., and Cech, T. R. (1986). The Tetrahymena ribozyme acts like an RNA restriction endonuclease. Nature 324, 429-433.

Uhlenbeck, O. C. (1987). A small catalytic oligoribonucleotide. Nature 328, 596-600.

Zhao, J. J., and Lemake, G. (1998). Rules for ribozymes. Mol Cell Neurosc 11, 92-97.

Uhlenbeck, O. C., Pardi, A., and Feigon, J. (1997). RNA structure comes of age. Cell 90, 833-840.

Zhou, C., Bahner, I., Rossi, J. J., and Kohn, D. B. (1996). Expression of hammerhead ribozymes by retroviral vectors to inhibit HIV-1 replication: Comparison of RNA levels and viral inhibition. Antisense Nucleic Acid Drug Dev 6, 17-24.

Vaish, N. K., Heaton, P. A., and Eckstein, F. (1997). Isolation of hammerhead ribozymes with altered core sequences by in vitro selection. Biochemistry 36, 6495-6501.

Zhou, C., Bahner, I. C., Larson, G. P., Zaia, J. A., Rossi, J. J., and Kohn, D. B. (1994). Inhibition of HIV-1 in human Tlymphocytes by retrovirally transduced anti-tat and rev hammerhead ribozymes. Gene 149, 33-39.

Ventura, M., Wang, P., Franck, N., and Saragosti, S. (1994). Ribozyme targeting of HIV-1 LTR. Biochem Biophys Res Commun 203, 889-898. Ventura, M., Wang, P., Ragot, T., Perricaudet, M., and Saragosti, S. (1993). Activation of HIV-specific ribozyme activity by selfcleavage. Nucleic Acids Res 21, 3249-3255.

Zoumadakis, M., and Tabler, M. (1995). Comparative analysis of cleavage rates after systematic permutation of the NUX consensus target motif for hammerhead ribozymes. Nucleic Acids Res 23, 1192-1196.

Wang, L., Witherington, C., King, A., Gerlach, W. L., Carr, A., Penny, R., Cooper, D., Symonds, G., and Sun, L. Q. (1998). Preclinical characterization of an anti-tat ribozyme for therapeutic application. Hum Gene Ther 9, 1283-1291

280

Gene Therapy and Molecular Biology Vol 3, page 281 Gene Ther Mol Biol Vol 3, 281-291. August 1999.

Use of antisense oligonucleotides to study homeobox gene function Review Article

Olubunmi Afonja1,2, Takashi Shimamoto1,3, John E. Smith, Jr.4, Long Cui 1, and Kenichi Takeshita 1,5 1

Hematology Division, Department of Medicine, Kaplan Cancer Center and 2Department of Pediatrics, New York University Medical Center, New York, NY. 3First Department of Internal Medicine, Tokyo Medical College, Tokyo, Japan and 4Hematology-Oncology Division, Metropolitan Hospital, New York Medical College, New York, NY 5HematologyOncology Division, Brooklyn Hospital Center, Brooklyn, NY. __________________________________________________________________________________________________ Correspondence: Kenichi Takeshita, M.D., Hematology Division, Department of Medicine, Kaplan Cancer Center, New York University Medical Center, 550 First Avenue, New York, NY 10016. Tel: 1-212-263 5465; Fax: 1-212-263 8444; E-mail: kt15@is.nyu.edu Key Words: Homeobox genes, leukemia, antisense oligonucleotides, DLX7, apoptosis, hematopoietic differentiation Received: 12 October 1998; accepted: 25 October 1998

Summary Homeobox genes code for transcription factors known t o be important i n hematopoiesis and leukemogenesis. In order to understand the function of a homeobox gene in leukemia, we have used an antisense oligonucleotide technique to inhibit the expression of a homeobox gene, DLX7, which is expressed at high levels in some leukemia cell lines and patients. With careful design and use of the antisense oligonucleotide, we have found that the loss of DLX7 gene expression results in loss of expression of secondary genes, cell cycle arrest and apoptosis. These studies have led to an understanding of DLX7 gene function in leukemia c e l l growth and the identification o f genes regulated by DLX7. These studies also raise the possibility that DLX7 antisense oligonucleotides may be useful in the treatment of patients with DLX7-positive leukemia.

I. Introduction A. Homeobox genes as regulators of hematopoietic differentiation Homeobox genes are a family of genes coding for transcription factors originally identified in the Drosophila as genes responsible for homeosis (McGinnis and Krumlauf, 1992; Scott et al., 1989), a phenomenon where Drosophila mutants exhibit abnormalities in pattern formation and cell fate decisions during embryogenesis. Homeobox genes have a strikingly conserved 60 aminoacid “homeodomain� encoding a DNA-binding domain. We and many others have demonstrated that both lymphoid and non-lymphoid hematopoietic cells express homeobox genes (Baier et al., 1991; Crompton et al., 1992; Deguchi et al., 1993; Inamori et al., 1993; Kongsuwan et al., 1988; Petrini et al., 1992; Shen et al.,

1989; Takeshita et al., 1993). It has therefore been attractive to speculate that homeobox genes may also play important role(s) in the regulation of hematopoiesis and pathogenesis of hematologic malignancies. Several homeobox genes have been shown to have a function in normal hematopoiesis and in leukemia. Using over-expression strategies, several groups have shown that dysregulated expression of homeobox genes can result in increased cell proliferation (Sauvageau et al., 1995) and in some cases uncontrolled proliferation leading to leukemia. These studies have been instrumental in identifying some homeobox genes as leukemogenic genes in both mouse and human disease (Lawrence and Largman, 1992). The role of homeobox genes in leukemogenesis is corroborated by findings that some leukemia-associated chromosomal translocations affect homeobox genes.

Afonja et al: antisense oligonucleotides in homeobox gene function Over-expression or ectopic expression of homeobox genes have been shown to cause leukemia. The murine acute monocytic leukemia line WEHI-3B has been shown to carry a retrovirus-like insertion near the Hox B8 gene, resulting in its over-expression (Blatt et al., 1988; Kongsuwan et al., 1989), while in BXH-2 mice, another homeobox gene is affected by retrovirus insertion (Moskow et al., 1995; Nakamura et al., 1996). The WEHI-3B cell line also has an IAP insertion upstream of the IL-3 gene, and co-infection of normal bone marrow with retrovirus vectors carrying the cloned IL-3 and Hox B8 genes resulted in myeloid leukemia (Perkins et al., 1990). There are also examples in humans of aberrant homeobox gene expression causing leukemia. In pre-B ALL with t(1;19) translocation a fusion protein is created between the transcription factor E2A gene and the homeobox gene PBX (Kamps et al., 1990; Nourse et al., 1990). In T-cell ALL with t(10;14) translocation, the Hox 11 gene is deregulated (Hatano et al., 1991; Lu et al., 1991). In myeloid leukemia with t(7;11) translocation, the HoxA9 gene is rearranged (Borrow et al., 1996; Nakamura et al., 1996). Rearranged homeobox genes have also been found in solid tumors such as rhabdomyosarcoma (Barr et al., 1993); ectopic expression of homeobox genes results in focus formation in the NIH 3T3 transformation assay (Maulbecker and Gruss, 1994). Thus, in addition to the well documented function in specifying body parts during embryogenesis, a major function of many human homeobox genes appears to be directed towards control of hematopoietic cell proliferation. An alternative strategy to understand homeobox gene function has been to inhibit the expression of homeobox genes. The best experimental system to inhibit gene expression involves gene targeting to create engineered mutants in mice. Hoxa-9 mutant mice demonstrate decreases in the number of hematopoietic progenitor cells, although the peripheral blood counts are relatively normal (Lawrence et al., 1997). We have obtained similar findings in the Hoxc-8 knock-out mice (Shimamoto et al., 1998). Mice lacking genes known to control Hox genes, such as bmi-1 (vander Lugt et al., 1994) and mll (Yu et al., 1995) have more striking hematologic deficits. These knock-out mouse strategies indicate that some homeobox genes are important for hematopoiesis. However, these systems do not easily offer a detailed look on how the homeobox genes regulate hematopoiesis. In this article, we review our experience with the use of antisense oligonucleotides as a way to selectively inhibit the expression of a single homeobox gene in order to understand its function in hematologic cells (Shimamoto et al., 1998; Takeshita et al., 1993). The use of antisense oligonucleotides has been summarized in two recent reviews (Crooke, 1998; Gewirtz et al., 1998).

II. The DLX homeobox gene family We have focused on a member of the DLX gene family, the founding member being the distal-less gene in the Drosophila fruitfly (Cohen et al., 1989). There are 6 members of this homeobox gene family, DLX1, DLX2, DLX3, DLX5, DLX6 and DLX7. These genes exist as closely located pairs in the chromosome, located at 2q32, 7q22 and 17q23, with the gene transcription units pointing toward each other. Therefore, the 3' end of one gene is also the 3' end of the other gene (Nakamura et al., 1996; Weiss et al., 1994) (see Figure 1). At least four of the six DLX genes, DLX 1,2,3 and 7, are expressed in many leukemia cell lines. Since the known location of normal DLX gene expression is in the forebrain and the craniofacial structures, and since normal hematopoietic cells do not express or express only very low levels of DLX genes, expression of DLX genes in leukemia is likely to reflect either gene expression in progenitor cells or ectopic expression in malignant cells. Mutation in the DLX3 gene, a 4-bp deletion in the coding region of the DLX3 protein, has been identified on the tricho-dento-osseous syndrome, an autosomal dominant disorder characterized by kinky curly hair, enamel hypoplasia, and increased thickness and density of cranial bones (Price et al., 1998). The DLX7 gene was chosen for our study for two reasons. First, it is expressed at high levels in many leukemia cell lines of an erythroid phenotype and in about 20% of leukemia cells from patients with acute myelogenous leukemia, but is expressed at very low levels in the normal bone marrow. The normal sites of expression of this gene are the central nervous system and craniofacial structures during development (Weiss et al., 1994).

A. DLX7 gene cloning and structure DLX7 is expressed by leukemia cell lines of an erythroid phenotype at a high level (Figure 2, panel C). The major transcript is 2 kB, with a minor species of about 7 kB seen only in K562 erythroleukemia cells, which probably represents an alternative transcript, a phenomenon commonly seen in all DLX genes and other homeobox genes. Using the more sensitive RT-PCR, we have found that the gene is also expressed at a barely detectable level in normal bone marrow cells and in peripheral blood cells (Figure 2, panel B) and at a readily detectable level in about 30% of leukemia cell lines. High levels are seen in erythroleukemia cell line K562, as well as in TF1 cells treated with erythropoietin and in the human erythroleukemia cell line, HEL. Treatment of HEL cells with hemin increases DLX7 mRNA within 24 hr of hemin addition to a level equal to K562 (Figure 2, panel A) (Shimamoto et al., 1997).

Gene Therapy and Molecular Biology Vol 3, page 283

F i g u r e 1 . G e n o m i c o r g a n i z a t i o n o f D L X g e n e s . The diagram depicts the genomic organization of DLX genes (not drawn to scale). The red boxes indicate coding regions, with the homeodomain highlighted in green. Three pairs of DLX genes are located on human chromosomes 2, 7 and 17 (Nakamura et al., 1996). The transcription units are separated by an intergenic distance of 5-10 kilobases and point toward each other. In addition to the transcript which gives rise to the DLX homeoprotein, there is substantial evidence indicating that many if not all DLX genes give rise to non-coding transcripts and antisense transcripts of unknown function (Ding et al., 1997; Liu et al., 1997; McGuinness et al., 1996; Ryoo et al., 1997). Figure 2. Expression of DLX7 gene in various cancer cells, with a particular emphasis on leukemia cells (adapted from Shimamoto et al., 1997). Panel A. RT-PCR indicating expression in K562 human erythroleukemia and HEL human erythroleukemia cell lines, showing high expression in K562 and high expression in hemin treated HEL cells. Hemin treatment induces an "erythroid" phenotype in HEL cells (Martin and Papayannopoulou, 1982). Panel B. RT-PCR showing low expression of DLX7 in normal hematopoietic cells, bone marrow and peripheral blood, compared to K562 cells. Panel C. A northern blot showing expression in human cancer cell lines. Lane 1, HL60 human myeloblastic leukemia cell line; lane 2, HeLa cervical carcinoma; lane 3, K562 erythroleukemia cell line from chronic myelogenous leukemia; lane 4, MOLT-4 Tlymphoblastoid cell line; lane 5, Raji B-lymphoblastoid Burkitt cell line; lane 6, SW480 human colon adenocarcinoma; lane 7, A549 human lung adenocarcinoma; lane 8, G361 human melanoma.

Gene Therapy and Molecular Biology Vol 3, page 284

Figure 3. Design of antisense oligonucleotides directed against DLX7 gene (adapted from Shimamoto et al., 1997). Panel A. Location and sequence of oligonucleotides assayed for their ability to interfere with the expression of DLX7. The oligonucleotide exhibiting maximal effect is shown as a red line and has the sequence GACGGACAGTTTCATAAG. The thin line indicates oligonucleotides with minimal activity. P a n e l B . Effect of DLX7 antisense oligonucleotide on DLX7 mRNA level in K562 cells. RNA was extracted 3 hr after treatment with sense (lane S), antisense (lane AS), mutant (lane M) and mock treatment (lane no oligo), and analyzed by RT-PCR. Amplification of hypoxanthine phosphoribosyl transferase (HPRT) as a control is shown at the bottom. "Untreated" indicates cells not exposed to oligonucleotide. The oligonucleotide concentrations were 20 ÂľM.

B. Possible function of DLX7 gene in hematopoietic cells There is also a preliminary report that DLX7 is the protein which binds to BP1, a binding site located in the 5’ upstream region of the beta globin gene that represses adult beta globin gene (Berg et al., 1998). Erythroleukemia cell lines, such as K562, TF1 and hemin treated HEL, express the fetal globin genes but not the adult globin genes. We have overexpressed DLX7 in the IL-3 dependent lymphoid cell line Ba/F3 and found that DLX7 relieves IL3 dependence and induces ICAM gene expression (Shimamoto et al., 1998), suggesting roles for both cell proliferation and differentiation. Additional roles have been suggested through the use of antisense oligonucleotides, as described below.

III. Use of antisense oligonucleotides to inhibit gene expression The use of antisense oligonucleotides to inhibit gene expression has been reviewed by others. We have used the method as originally developed by Gewirtz and co-workers to inhibit gene expression in leukemia cells (Gewirtz and Calabretta, 1988; Szczylik et al., 1991). We have generally chosen antisense oligonucleotides based on

several randomly selected sequences near the translation start site. The oligonucleotides are generally 18 mers, with GC content of about 50% (Figure 3 panel A). For our purposes, it is particularly pertinent to note that questions have been raised about the specificity of biological effects seen with antisense treatment. Most of the reports of sequence-independent effects of oligonucleotides concern the phosphorothioate oligonucleotides, in which the sulfur entity is thought to result in a variety of biological effects (Abraham et al., 1997; Castier et al., 1998; Schobitz et al., 1997; Too, 1998; Wojcik et al., 1996; Yamaguchi et al., 1997). We have used phosphodiester oligonucleotides in all studies described here because it is cheaper and because the studies are based on cell lines where nuclease activity can be minimized. However, sequence independent effects have also been reported for phosphodiester oligonucleotides (Kabisch et al., 1994; Stull et al., 1993; Wu-Pong et al., 1994).

A. Inhibition of the target mRNA either at the protein level or at the mRNA level. We have performed the following studies to demonstrate sequence specificity of the oligonucleotides we

Gene Therapy and Molecular Biology Vol 3, page 285 have used. Antisense oligonucleotides are designed to bind to the target mRNA and cause either RNaseH-mediated degradation of the mRNA or block mRNA translation. However, in practice, many antisense oligonucleotides designed on the basis of the known mRNA sequence are ineffective (Figure 3, panel A). The reason for this is unclear but has been attributed to the complex secondary structure of the mRNA molecule which renders the mRNA inaccessible to the oligonucleotide.

B. Lack of non-specific toxicity In our experiments we have observed that some batches of oligonucleotide preparations obtained from many manufacturers and from our own institutional synthesizers give non-specific toxicity. The toxicity stems possibly from incomplete removal of many organic compounds used during the synthesis. Therefore, an initial concern is to demonstrate that any biological effect observed with an oligonucleotide is due to the oligonucleotide itself and not to impurities present in the oligonucleotide preparation. Several approaches are available for the purification of oligonucleotides, such as HPLC, gel electrophoresis, etc. We have also found that some manufacturers offer oligonucleotides which do not show non-specific toxicity, even at very high concentrations (>100 µM). These preparations do not appear to require additional specialized purification other than ethanol precipitation.

C. Demonstration of specificity by creating point mutations within the oligonucleotide sequence In order to demonstrate specificity of any effect observed with an antisense oligonucleotide, we have introduced point mutations within the oligonucleotide sequence to demonstrate that the mutation abolishes the observed biological effect. Exactly how many mismatches can result in the loss of the antisense oligonucleotide is unknown. Therefore, we generally have introduced 3 or 4 mismatches in an 18-mer to create a “mutated” oligonucleotide. The specificity of the DLX-7 antisense oligonucleotide is illustrated in F i g u r e 3 , panel B (Shimamoto et al., 1997). The specificity of mRNA inhibition by the antisense oligodeoxynucleotide used was demonstrated by measuring levels of DLX-7 mRNA in cells treated with sense, antisense, mutant (same sequence as the antisense except for 4 base changes), or no oligodeoxynucleotide. K562 erythroleukemia cells exposed to DLX7 antisense oligonucleotides showed a significant decrease in DLX-7 mRNA levels (Figure 3, panel B). In contrast, the negative control oligodeoxynucleotides, including the mutant oligo, had no effect (Shimamoto et al., 1997).

D. Are secondary changes seen with DLX7 antisense oligonucleotide treatment related to the binding of a DNA molecule with an mRNA, rather than specific inhibition of DLX7? Some have argued that the mere existence of DNARNA hybrid per se in the cell causes physiological changes in cells independent of any gene inhibition, for example RNase H activation, and that secondary effects are unrelated to the biological effects of DLX7 gene inhibition. Although we cannot exclude this possibility completely, we can present the following data which argue against this possibility. First, as we discuss below, we have observed secondary effects, such as down regulation of c-myc and GATA-1 genes, after DLX7 antisense treatment. In contrast an antisense oligonucleotide directed against MEIS1 homeobox gene does not cause secondary inhibition of c-myc or GATA-1, but instead inhibits a different set of oncogenes (JES unpublished). Second, there are tissue-specific differences in response to DLX7 antisense oligonucleotide. For example, the human lung cancer cell line A549, expresses the DLX7 the gene and shows proper down regulation of DLX7 mRNA in response to the antisense oligonucleotide. However, A549 does not show the secondary gene changes (e.g. cmyc gene inhibition seen in K562 leukemic cells).

IV. Cellular effects of antisense oligonucleotide mediated inhibition of DLX7 gene As reviewed above, there is good evidence suggesting that some homeobox genes may participate in leukemogenesis. In the mouse, it is possible to test such candidate leukemogenic homeobox genes using transgenic or retroviral systems. Usually after a latent period of weeks to months, leukemia develops in such mice. However, it is unclear whether such mouse leukemia accurately reflects events occurring in human leukemia because of the differences in the target cell that undergoes transformation. For example, genes isolated as giving rise to myeloid leukemia in humans have been noted to cause lymphoid leukemia in mice. In addition, the mechanisms leading to genetic dysregulation in leukemia are also different, namely retroviral gene activation in mice and chromosomal translocation in humans. There are also likely to be more potentially leukemogenic genes with no grossly apparent alterations in the gene structure. Thus, the antisense oligonucleotide approach is attractive in that genes can be inhibited in a perhaps more biologically meaningful setting.

Afonja et al: antisense oligonucleotides in homeobox gene function GATA-1, HOXC8 and c-myc mRNAs were decreased compared to control oligos (F i g . 4 ). GATA-1 and c-myc are well known regulators of hematopoiesis (Dubart et al., 1996; Mouthon et al., 1993; Weiss and Orkin, 1995). HOXC8 is a homeobox gene which also regulates hematopoiesis (Shimamoto et al., 1998); it is of note that the Drosophila versions of HOXC8 and DLX7 interact (Cohen et al., 1993; OHara et al., 1993; Panganiban et al., 1994). In contrast, GATA-2, SCL, c-myb, gamma-globin, c-abl, bcl-2 and bcl-xL mRNA were unaffected (Shimamoto et al., 1997). Preliminary data from nuclear run off studies indicate that the c-myc gene inhibition is mediated at the transcriptional level, with an associated alteration in the level of E2F activity, the transcription factor which regulates c-myc gene transcription. In contrast, the loss of GATA-1 gene expression is mediated at the mRNA level (data not shown).

B. Decreased colony forming efficiency after DLX-7 antisense treatment We emphasize that the changes in gene expression seen after antisense oligonucleotide treatment described above occurred at 3 hours after the treatment. During this time, no changes in the cell viability, morphology, etc., are seen.

F i g u r e 4 . Analysis of gene expression in K562 cells treated with DLX7 antisense oligonucleotide (adapted from (Shimamoto et al., 1997). RNA was extracted 3 hr after oligonucleotide treatment and analyzed by RT-PCR for the genes indicated.

A. c-myc and GATA-1 genes are down regulated following antisense oligonucleotide treatment Since DLX7 is a homeoprotein and thus likely to code for a transcription factor, we investigated the effects of inhibition of DLX7 gene expression on the expression of other genes. We selected several genes known to be important in hematopoiesis, cell proliferation or apoptosis. At 3 hours after DLX-7 antisense treatment,

However, we note that both GATA-1 and c-myc have been reported to be involved in apoptosis. The role of cmyc in cell proliferation is well known (Dubart et al., 1996; Mouthon et al., 1993; Weiss and Orkin, 1995). In the case of GATA-1, GATA-1 deficient cells exhibit apoptosis at the normoblast stage and inhibition of GATA-1 in erythroleukemia cells causes apoptosis (Blobel and Orkin, 1996; Dubart et al., 1996; Weiss and Orkin, 1995). Thus, the loss of these proteins might be expected to inhibit cell proliferation and cause apoptosis. To examine this hypothesis, we tested the viability of K562 cells after antisense oligo treatment by determining the plating efficiency of K562 cells treated with oligonucleotides in methylcellulose, a viscous culture medium which allows cells to grow as a colony. After a 30 min oligonucleotide treatment, the cells were plated in methylcellulose and the colonies counted after 7 days. 20 µM of DLX-7 antisense treatment suppressed the colony forming efficiency of K562 cells, whereas sense and mutant oligodeoxynucleotides had no effect on the plating efficiency, indicating essentially no non-specific toxicity of the oligonucleotide treatment (F i g . 5, panel A) (Shimamoto et al., 1997). Furthermore, the inhibitory effect of DLX-7 antisense oligodeoxynucleotide was dosedependent, ranging from 30% inhibition at 5 µM to 70% inhibition at 40 µM ( F i g . 5 , panel B) (Shimamoto et al., 1997).

Gene Therapy and Molecular Biology Vol 3, page 287

Figure 5. Effects of DLX7 antisense oligonucleotide treatment on K562 cell viability (adapted from (Shimamoto et al., 1997). Panel A. After exposure to the oligonucleotides, 500 cells were plated per well in quadruplicate and cultured for 7 days in a viscous methylcellulose-containing culture medium; after which the leukemia colonies were counted. The concentration of the antisense used was 20 ÂľM. Panel B. Dose response curve of the oligonucleotides on the colony formation by K562 leukemia cells. Studies were carried out as in panel A, with the oligonucleotide concentration varied as indicated.

C. Apoptosis after antisense oligonucleotide treatment To determine whether cells treated with antisense underwent apoptosis, K562 cells were stained by the TUNEL method, which detects in situ endonucleolytic cleavage characteristic of apoptosis. Apoptosis was assayed at 6 hr and at 12 hr after the oligonucleotide treatment. No apoptosis was seen at 6 hr. However, at 12 hr after the oligonucleotide treatment, apoptotic cells were observed (Figure 6) (Shimamoto et al., 1997). Additional studies indicate that the apoptosis is preceded by a block in the G1 to S progression in the cell cycle (data not shown), in agreement with the known site of action of c-myc and E2F.

V. Concluding remarks and observations In this review, we have outlined the uses of antisense oligonucleotide technology, primarily as a way to probe the function of the DLX7 gene. These data suggest that DLX7 homeobox gene controls directly or indirectly the expression of secondary genes such as c-myc and GATA-1, that the mechanism of control might be transcriptional or mRNA stability, and that the loss of DLX7 gene expression results in apoptosis. However, as physicians, we also believe that the real value of antisense oligonucleotides may come in the form of better treatment for leukemia and other disorders. Currently, several antisense oligonucleotides directed against other genes are in clinical trials. These include phase II and III clinical trials for antisense oligonucleotides

Afonja et al: antisense oligonucleotides in homeobox gene function directed against BCL2, c-myb, BCR-ABL, protein kinase A for a wide range of diseases, including hematologic malignancies, solid tumors, as well as immunologic diseases, such as ulcerative colitis, and infectious diseases. One antisense oligonucleotide has recently been approved by the United States Food and Drug Administration for the treatment of cytomegalovirus induced retinitis (Isis Co., 1998). Our intention is to continue to study the function of the DLX7 gene, with an emphasis on clinical applications.

Acknowledgments This work was supported by the Lauri Strauss Leukemia Foundation in memory of Michael J. Lopez. We also acknowledge the continued assistance by Migdalia Avila and support by the Marcia Slater Society for Research in Leukemia, the New York chapter of the American Heart Association, and the Kaplan Cancer Center (NIH P30CA16087). We are supported by the Florence Carter Fellowship in Leukemia Research of the American Medical Association (TS), NIH training grants T32CA09454 (JES) and T32-HL07151 (OA).

References Abraham, W. C., Logan, B., Thompson, V. L., Williams, J. M., and Tate, W. P. (1 9 9 7 ). Sequence-independent effects of phosphorothioated oligonucleotides on synaptic transmission and excitability in the hippocampus in vivo. Neuropharmacology 36 , 345-52. Baier, L. J., Hannibal, M. C., Hanley, E. W., and Nabel, G. J. (1 9 9 1 ). Lymphoid expression and TATAA binding of a human protein containing an Antennapedia homeodomain. B l o o d 78 , 1047-55. Barr, F. G., Galili, N., Holick, J., Biegel, J. A., Rovera, G., and Emanuel, B. S. (1 9 9 3 ). Rearrangement of the PAX3 paired box gene in the paediatric solid tumour alveolar rhabdomyosarcoma. Nature Genet 3, 113-7. Figure 6. In situ analysis for DNA fragmentation by the TUNEL assay. At 12 hr after oligonucleotide treatment, cytocentrifugation preparations of K562 leukemia cells were stained by the TUNEL assay (adapted from Shimamoto et al., 1997). P a n e l A , no oligonucleotide; p a n e l B , antisense oligonucleotide treatment; p a n e l C , sense oligonucleotide treatment. Apoptotic cells were identified by labeling with fluorescent dUTP. No apoptosis was observed at 3 hr and 6 hr after oligonucleotide treatment (not shown).

Berf., Fu, S., Karp, J., Ross, D., Williams, D., Hawkins, W.,Behm, F., Ruscetti, F., and Haga, S. (1998). Overexpression of BP!/DLX9 in acute myeloid leukemia. B l o o d 92 (suppl 1), 603a Blatt, C., Aberdam, D., Schwartz, R., and Sachs, L. (1 9 8 8 ). DNA rearrangement of a homeobox gene in myeloid leukaemic cells. EMBO J 7, 4283-90. Blobel, G. A., and Orkin, S. H. (1 9 9 6 ). Estrogen-induced apoptosis by inhibition of the erythroid transcription factor GATA-1. M o l C e l l B i o l 16 , 1687-94. Borrow, J., Shearman, A. M., Stanton, V. P., Becher, R., Collins, T., Williams, A. J., Dube, I., Katz, F., Kwong, Y. L., Morris, C., Ohyashiki, K., Toyama, K., Rowley, J., and Housman, D. E. (1 9 9 6 ). The t(7;11)(p15;p15) translocation in acute myeloid leukemia fuses the genes

Gene Therapy and Molecular Biology Vol 3, page 289 for nucleoporin NUP98 and class I homeoprotein Hoxa9. Nature Genet 12 , 159-167.

suppress Philadelphia-positive clonogenic cells. Acta H a e m a t o l o g i c a 92 , 190-6.

Castier, Y., Chemla, E., Nierat, J., Heudes, D., Vasseur, M. A., Rajnoch, C., Bruneval, P., Carpentier, A., and Fabiani, J. N. (1 9 9 8 ). The activity of c-myb antisense oligonucleotide to prevent intimal hyperplasia is nonspecific. J Cardiov Surg 39 , 1-7.

Kamps, M. P., Murre, C., Sun, X. H., and Baltimore, D. (1 9 9 0 ). A new homeobox gene contributes the DNA binding domain of the t(1;19) translocation protein in pre-B ALL. C e l l 60 , 547-55.

Cohen, B., Simcox, A. A., and Cohen, S. M. (1 9 9 3 ). Allocation of the thoracic imaginal primordia in the Drosophila embryo. D e v e l o p m e n t 117, 597-608.

Kongsuwan, K., Allen, J., and Adams, J. M. (1 9 8 9 ). Expression of Hox 2.4 gene directed by proviral insertion in myeloid leukemia. N u c l e i c A c i d s R e s 17 , 18811892.

Cohen, S. M., Bronner, G., Kuttner, F., Jurgens, G., and Jackle, H. (1 9 8 9 ). Distal-less encodes a homeodomain protein required for limb development in Drosophila. Nature 338, 432-434.

Kongsuwan, K., Webb, E., Housiaux, P., and Adams, J. M. (1 9 8 8 ). Expression of multiple homeobox genes within diverse mammalian haemopoietic lineages. EMBO J 7, 2131-8.

Crompton, M. R., Bartlett, T. J., MacGregor, A. D., Manfioletti, G., Buratti, E., Giancotti, V., and Goodwin, G. H. (1 9 9 2 ). Identification of a novel vertebrate homeobox gene expressed in hematopoietic cells. N u c l e i c A c i d s R e s 20 , 5661-5667.

Lawrence, H. J., Helgason, C. D., Sauvageau, G., Fong, S., Izon, D. J., Humphries, R. K., and Largman, C. (1 9 9 7 ). Mice bearing a targeted interruption of the homeobox gene HOXA9 have defects in myeloid, erythroid, and lymphoid hematopoiesis. B l o o d 89 , 1922-30.

Crooke, S. T. (1 9 9 8 ). An overview of progress in antisense therapeutics. A n t i s e n s e Nucl Acid Drug D e v 8, 115-22.

Lawrence, H. J., and Largman, C. (1 9 9 2 ). Homeobox genes in normal hematopoiesis and leukemia. B l o o d 80 , 244553.

Deguchi, Y., Yamanaka, Y., Theodossiou, C., Najfeld, V., and Kehrl, J. H. (1 9 9 3 ). High expression of two diverged homeobox genes, HB24 and HB9, in acute leukemias: molecular markers of hematopoietic cell immaturity. Leukemia 7, 446-51.

Liu, J. K., Ghattas, I., Liu, S., Chen, S., and Rubenstein, J. L. (1 9 9 7 ). Dlx genes encode DNA-binding proteins that are expressed in an overlapping and sequential pattern during basal ganglia differentiation. D e v D y n a m i c s 210, 498512.

Ding, M., Robel, L., James, A. J., Eisenstat, D. D., Leckman, J. F., Rubenstein, J. L., and Vaccarino, F. M. (1997). Dlx2 homeobox gene controls neuronal differentiation in primary cultures of developing basal ganglia. J M o l Neurosci 8, 93-113.

Lu, M., Gong, Z. Y., Shen, W. F., and Ho, A. D. (1 9 9 1 ). The tcl-3 proto-oncogene altered by chromosomal translocation in T-cell leukemia codes for a homeobox protein. EMBO J 10 , 2905-10.

Dubart, A., Romeo, P. H., Vainchenker, W., and Dumenil, D. (1 9 9 6 ). Constitutive expression of GATA-1 interferes with the cell-cycle regulation. B l o o d 87 , 3711-21. Gewirtz, A. M., and Calabretta, B. (1 9 8 8 ). A c-myb antisense oligonucleotide inhibits normal human hematopoiesis in vitro. S c i e n c e 242, 1303-1306. Gewirtz, A. M., Sokol, D. L., and Ratajczak, M. Z. (1 9 9 8 ). Nucleic acid therapeutics: state of the art and future prospects. B l o o d 92 , 712-36. Hatano, M., Roberts, C. W., Minden, M., Crist, W. M., and Korsmeyer, S. J. (1 9 9 1 ). Deregulation of a homeobox gene, HOX11, by the t(10;14) in T cell leukemia. S c i e n c e 253, 79-82. Inamori, K., Takeshita, K., Chiba, S., Yazaki, Y., and Hirai, H. (1 9 9 3 ). Identification of homeobox genes expressed in human T-lymphocytes. B i o c h e m B i o p h y s R e s Commun 196, 203-208.

Martin, P., and Papayannopoulou, T. (1 9 8 2 ). HEL cells: a new human erythroleukemia cell line with spontaneous and induced globin expression. S c i e n c e 216, 1233-5. Maulbecker, C. C., and Gruss, P. (1 9 9 4 ). The oncogenic potential of deregulated homeobox genes. C e l l G r o w t h D i f f 4, 431-441. McGinnis, W., and Krumlauf, R. (1 9 9 2 ). Homeobox genes and axial patterning. C e l l 68 , 283-302. McGuinness, T., Porteus, M. H., Smiga, S., Bulfone, A., Kingsley, C., Qiu, M., Liu, J. K., Long, J. E., Xu, D., and Rubenstein, J. L. (1 9 9 6 ). Sequence, organization, and transcription of the Dlx-1 and Dlx-2 locus. G e n o m i c s 35 , 473-85. Moskow, J. J., Bullrich, F., Huebner, K., Daar, I. O., and Buchberg, A. M. (1 9 9 5 ). Meis1, a PBX1-related homeobox gene involved in myeloid leukemia in BXH-2 mice. M o l C e l l B i o l 15 , 5434-43.

ISIS Co., press release (1 9 9 8 ). Press release: ISIS and Ciba Vision present statistically significant phase III efficacy results for antisense CMV retinitis drug.

Mouthon, M. A., Bernard, O., Mitlavila, M. T., Romeo, P. H., Vainchenker, W., and Mathieu-Mahul, D. (1 9 9 3 ). Expression of tal-1 and GATA-binding proteins during human hematopoiesis. B l o o d 81 , 647-655.

Kabisch, A., Perenyi, L., Seay, U., Lohmeyer, J., and Pralle, H. (1 9 9 4 ). Unmodified phosphodiester antisense oligodeoxynucleotides to the BCR-ABL junction do not

Nakamura, S., Stock, D. W., Wydner, K. L., Bollekens, J. A., Takeshita, K., Nagai, B. M., Chiba, S., Kitamura, T., Freeland, T. M., Zhao, Z., Minowada, J., Lawrence, J. B.,

Afonja et al: antisense oligonucleotides in homeobox gene function Weiss, K. M., and Ruddle, F. H. (1 9 9 6 ). Genomic analysis of a new mammalian distal-less gene: DLX7. G e n o m i c s 38 , 314-324. Nakamura, T., Largaespada, D. A., Shaughnessy, J. D., Jr., Jenkins, N. A., and Copeland, N. G. (1 9 9 6 ). Cooperative activation of Hoxa and Pbx1-related genes in murine myeloid leukaemias. Nature Genet 12 , 149-53. Nakamura, T., Largaspada, D. A., Lee, M. P., Johnson, L. A., Ohyashiki, K., Toyama, K., Chen, S. J., Willman, C. L., Chen, I. M., Feinberg, A. P., Jenkins, N. A., Copeland, N. G., and Shaughnessy, J. D. (1 9 9 6 ). Fusion of the nucleoporin gene NUP98 to HOXA9 by the chromosome translocation t(7;11)(p15;p15) in human myeloid leukemia. Nature Genet 12 , 154-158. Nourse, J., Mellentin, J. D., Galili, N., Wilkinson, J., Stanbridge, E., Smith, S. D., and Cleary, M. L. (1 9 9 0 ). Chromosomal translocation t(1;19) results in synthesis of a homeobox fusion mRNA that codes for a potential chimeric transcription factor. C e l l 60 , 535-45.

rats: occurrence of non-specific Pharmacol 331, 97-107.

effects.

Eur

J

Scott, M. P., Tamkun, J. W., and Hartzell, G. W. (1 9 8 9 ). The structure and function of the homeodomain. B i o c h i m B i o p h y s A c t a 989, 25-48. Shen, W. F., Largman, C., Lowney, P., Corral, J. C., Detmer, K., Hauser, C. A., Simonitch, T. A., Hack, F. M., and Lawrence, H. J. (1 9 8 9 ). Lineage-restricted expression of homeobox-containing genes in human hematopoietic cell lines. Proc Natl Acad Sci USA 86 , 8536-40. Shimamoto, T., Nakamura, S., Bollekens, J., Ruddle, F. H., and Takeshita, K. (1 9 9 7 ). Inhibition of DLX-7 expression causes decreased expression of GATA-1 and cmyc genes and apoptosis. P r o c N a t l A c a d S c i U S A 94 , 3245-3249. Shimamoto, T., Ohyashiki, K., and Takeshita, K. (1 9 9 8 ). Overexpression of the homeobox gene DLX7 inhibits apoptosis and alters expression of adhesion moelcules. B l o o d , 92 (suppl 1), 199a.

OHara, E., Cohen, B., Cohen, S. M., and McGinnis, W. (1 9 9 3 ). Distal-less is a downstream gene of Deformed required for ventral maxillary identity. D e v e l o p m e n t 117, 847-56.

Shimamoto, T., Tang, Y., Naot, Y., Nardi, M., Brulet, P., Bieberich, C. J., and Takeshita, K. (1 9 9 8 ). Hematopoietic progenitor cell abnormalities in hoxc8null mutant mice. J E xp Z ool 282, in press.

Panganiban, G., Nagy, L., and Carroll, S. B. (1 9 9 4 ). The role of the Distal-less gene in the development and evolution of insect limbs. Curr Biol 4, 671-5.

Stull, R. A., Zon, G., and Szoka, F. C., Jr. (1 9 9 3 ). Singlestranded phosphodiester and phosphorothioate oligonucleotides bind actinomycin D and interfere with tumor necrosis factor-induced lysis in the L929 cytotoxicity assay. A n t i s e n s e R e s D e v 3, 295-300.

Perkins, A., Kongsuwan, K., Visvader, J., Adams, J. M., and Cory, S. (1 9 9 0 ). Homeobox gene expression plus autocrine growth factor production elicits myeloid leukemia. Proc Natl Acad Sci USA 87 , 8398-402. Petrini, M., Quaranta, M. T., Testa, U., Samoggia, P., Tritarelli, E., Care, A., Cianetti, L., Valtieri, M., Barletta, C., and Peschle, C. (1 9 9 2 ). Expression of selected human HOX-2 genes in B/T acute lymphoid leukemia and interleukin-2/interleukin-1 beta-stimulated natural killer lymphocytes. B l o o d 80 , 185-93. Price, J. A., Bowden, D. W., Wright, J. T., Pettenati, M. J., and Hart, T. C. (1 9 9 8 ). Identification of a mutation in DLX3 associated with tricho-dento-osseous (TDO) syndrome. Hum Mol Genet 7, 563-9. Ryoo, H. M., Hoffmann, H. M., Beumer, T., Frenkel, B., Towler, D. A., Stein, G. S., Stein, J. L., van Wijnen, A. J., and Lian, J. B. (1 9 9 7 ). Stage-specific expression of Dlx-5 during osteoblast differentiation: involvement in regulation of osteocalcin gene expression. M o l E ndo c r i no l 11 , 1681-94. Sauvageau, G., Thorsteinsdottir, U., Eaves, C. J., Lawrence, H. J., Largman, C., Lansdorp, P. M., and Humphries, R. K. (1 9 9 5 ). Overexpression of HOXB4 in hematopoietic cells causes the selective expansion of more primitive populations in vitro and in vivo. G e n e s D e v 9, 175365. Schobitz, B., Pezeshki, G., Probst, J. C., Reul, J. M., Skutella, T., Stohr, T., Holsboer, F., and Spanagel, R. (1 9 9 7 ). Centrally administered oligodeoxynucleotides in

Szczylik, C., Skorski, T., Nicolaides, N. C., Manzella, L., Malaguarnera, L., Venturelli, D., Gewirtz, A. M., and Calabretta, B. (1 9 9 1 ). Selective inhibition of leukemia cell proliferation by BCR-ABL antisense oligodeoxynucleotides. S c i e n c e 253, 562-565. Takeshita, K., Bollekens, J. A., Hijiya, N., Ratajczak, M., Ruddle, F. H., and Gewirtz, A. M. (1 9 9 3 ). A homeobox gene of the Antennapedia class is required for human adult erythropoiesis. Proc Natl Acad Sci USA 90 , 3535-8. Too, C. K. (1 9 9 8 ). Rapid induction of Jak2 and Sp1 in T cells by phosphorothioate oligonucleotides. A n t i s e n s e Nucl Acid Drug Dev 8, 87-94. vander Lugt, N. M., Domen, J., Linders, K., vanRoon, M., Robanus-Maandag, E., te Riele, H., vander Valk, M., Deschamps, J., Sofroniew, M., and van Lohuizen, M. (1 9 9 4 ). Posterior transformation, neurological abnormalities, and severe hematopoietic defects in mice with a targeted deletion of the bmi-1 proto-oncogene. G e n e s D e v 8, 757-69. Weiss, K. M., Bollekens, J., Ruddle, F. H., and Takeshita, K. (1 9 9 4 ). Distal-less and other homeobox genes in the development of the dentition. J E xp Z ool 270, 273-84. Weiss, M. J., and Orkin, S. H. (1 9 9 5 ). Transcription factor GATA-1 permits survival and maturation of erythroid precursors by preventing apoptosis. Proc Natl Acad S c i USA 92 , 9623-7.

Gene Therapy and Molecular Biology Vol 3, page 291 Wojcik, W. J., Swoveland, P., Zhang, X., and Vanguri, P. (1 9 9 6 ). Chronic intrathecal infusion of phosphorothioate or phosphodiester antisense oligonucleotides against cytokine responsive gene-2/IP10 in experimental allergic encephalomyelitis of lewis rat. J Pharmacol Exp Therap 278, 404-10. Wu-Pong, S., Weiss, T. L., and Hunt, C. A. (1 9 9 4 ). Antisense c-myc oligonucleotide cellular uptake and activity. A n t i s e n s e R e s D e v 4, 155-63. Yamaguchi, K., Papp, B., Zhang, D., Ali, A. N., Agrawal, S., and Byrn, R. A. (1 9 9 7 ). The multiple inhibitory mechanisms of GEM 91, a gag antisense phosphorothioate oligonucleotide, for human immunodeficiency virus type 1. AIDS R e s Human R e t r o v 13 , 545-54. Yu, B., Hess, J., Horning, S., Brown, G., and Korsmeyer, S. (1 9 9 5 ). Altered Hox expression and segmental identity in Mll-mutant mice. Nature 378, 505-8.

Gene Therapy and Molecular Biology Vol 3, page 293 Gene Ther Mol Biol Vol 3, 293-300. August 1999.

Potential application of dominant negative retinoic acid receptor genes for ex vivo expansion of hematopoietic stem cells Research Article

Yoji Ogasawara1,3,4, Yutaka Hanazono1, Hiroshi Kodaira1,3, Masashi Urabe1,3, Hiroyuki Mano1,3, Akira Kakizuka5, Akihiro Kume1,3, Keiya Ozawa1,2,3 1

Division of Genetic Therapeutics, Center for Molecular Medicine, and 2Department of Hematology, Jichi Medical School, 3311-1 Yakushiji, Minamikawachi-machi, Kawachi-gun, Tochigi 329-0498, Japan. 3CREST, Japan Science and Technology Cooperation (JST), 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan. 4Second Department of Internal Medicine, Jikei University School of Medicine, 3-25-8 Nishi-Shinbashi, Minato-ku, Tokyo 105-0003, Japan. 5Osaka Bioscience Institute, 6-2-4 Furuedai, Suita, Osaka 565-0874, Japan. __________________________________________________________________________________________________ Correspondence: Keiya Ozawa, M.D., Ph.D., Division of Genetic Therapeutics, Center for Molecular Medicine, Jichi Medical School, 3311-1 Yakushiji, Minamikawachi-machi, Kawachi-gun, Tochigi 329-0498 Japan, Tel: +81-285-7402; Fax: +81-285-448675; E-mail: kozawa@ms.jichi.ac.jp Key words: gene therapy, hematopoietic stem cell, ex vivo expansion, retroviral vector, hematopoietic cytokine, dominant negative retinoic acid receptor, Cre recombinase, loxP, all-trans retinoic acid, differentiation block Received: 16 November 1998; accepted: 25 November 1998

Summary I t i s d i f f i c u l t t o expand hematopoietic stem c e l l s e x v i v o by stimulation with hematopoietic c y t o k i n e s , s i n c e any cytokine thus far tested induces differentiation as w e l l as growth. It i s therefore important to consider how to inhibit differentiation of hematopoietic stem cells during ex vivo culture. For this purpose, we have constructed retroviral vectors expressing dominant-negative retinoic acid receptor- (RAR ) genes. The immature hematopoietic cell lines 32D and FDCPmix, transduced with the dominant-negative RAR -expressing vectors, remained blastic or promyelocytic and continued to grow without differentiation even under the differentiation-inducing conditions. This block o f differentiation could be overcome by treatment with all-trans retinoic acid, suggesting that the transduced cells still retained their differentiating ability. This leads to the possible application of dominant-negative RAR genes for the ex vivo expansion of hematopoietic stem c e l l s i n combination with hematopoietic cytokines. For clinical applications, however, dominant-negative RAR g e n e s s h o u l d b e r e m o v e d f r o m h o s t c e l l s a f t e r e x v i v o e x p a n s i o n . W e have, therefore, incorporated two loxP sites on either side of the RAR gene in each vector so that removal of the integrated differentiation-blocking RAR genes from the transduced cells could be achieved using the Cre recombinase/loxP system (reversible integration of a gene of interest). We are investigating efficient ways to introduce the Cre recombinase into host cells.

combinations of hematopoietic cytokines. Most studies report increases in total cell number, colony-forming units (CFUs), or total CD34+ cells, probably inadequate surrogates for true stem cells. Clinical studies using ex vivo expanded cell populations have not proven maintenance or expansion of either short- or long-term

I. Introduction The possibility of ex vivo hematopoietic stem cell expansion is of interest for both gene therapy and transplantation applications (Dunbar and Young, 1996). Several recent papers describe the so-called “expansion� of human progenitor and stem cells with various 293

Ogasawara et al: RAR genes for ex vivo expansion of hematopoietic stem cells repopulating ability (Brugger et al., 1995; Holyoake et al., 1997). Gene marking studies in rhesus monkeys indicate that ex vivo expansion of mobilized peripheral blood cells for 10-14 days in the presence of interleukin (IL)-3, IL-6, stem cell factor (SCF), FLT-3 ligand, and stromal cells resulted in no increase in initial engraftment and diminished long-term engraftment (Tisdale et al., 1998). A recent report of graft failure after hematopoietic cell transplantation of ex vivo expanded cells supports the conclusion that “expansion� conditions may differentiate hematopoietic stem cells and damage engrafting cells and that committed progenitors do not contribute to even shortterm engraftment (Holyoake et al., 1997). From these studies, for true ex vivo expansion of hematopoietic stem cells, inhibition of differentiation seems to be required during ex vivo culture.

loxP sites as the first genes. Both RAR403 and RARE genes are dominant-negative mutants of the human RAR! gene. RAR403 is a C-terminus-truncated form which encodes a peptide of 403 amino acids (Tsai et al., 1993). RARE has a glycine to glutamic acid change at amino acid 303 (Saitou et al., 1994).

B. Expression of dominant-negative RAR genes in transduced cells F i g . 2 A shows the Northern blotting of FDCPmix cells transduced with the mock vector (MXLLneo) or the dominant-negative RAR!-expressing vector (MXL403Lneo). The MXL403Lneo vector expressed two forms of the RAR403 gene transcripts (a full length of the message and a spliced form) in the transduced FDCPmix cells, although the mock vector MXLLneo did not. F i g . 2B shows the Western blotting of the 32D cells transduced with the mock vector (MXLLneo) or the dominantnegative RAR!-expressing vector (MXL403Lneo or MXLELneo). MXL403Lneo expressed the RAR403 (Cterminus-truncated RAR) protein at the molecular weight of 50 kDa and MXLELneo expressed the RAR protein that has a single amino acid residue substituted at the same molecular weight as the intact human RAR! protein (55 kDa).

The retinoic acid receptors (RARs) are members of the steroid/thyroid hormone receptors that function as ligandinducible transcription factors (Evans, 1988). RARs (predominantly RAR!) are involved in regulating hematopoiesis. Retinoic acid (RA) induces the HL-60 human leukemia cell line to differentiate into mature neutrophils, and this process is mediated through RARs (Breitman et al., 1980; Collins et al., 1990). In most cases of acute promyelocytic leukemia, the gene of RAR! on chromosome 17 is translocated and fused with the PML gene on chromosome 15 (Alcalay et al., 1990; Borrow et al., 1990; De The et al., 1990), and the leukemic cells from these patients can be induced by RA to differentiate into mature neutrophils both in vitro and in vivo (Huang et al., 1988). In this paper, we show that immature hematopoietic cells transduced with dominant-negative RAR!-expressing retroviral vectors were not differentiated even under the differentiation-inducing conditions. For possible future applications, the mutant RAR! genes need to be removed from the integrated vector sequences after ex vivo culture. We incorporated loxP sites on the retroviral vectors so that the dominant-negative RAR! genes could be removed by treatment with the Cre recombinase (Dale et al., 1991; Bergemann et al., 1995).

C. Differentiation block of transduced 32D and FDCPmix cells 32D is a murine IL-3-dependent hematopoietic progenitor cell line. IL-3 stimulates growth of 32D cells but granulocyte colony-stimulating factor (G-CSF) stimulates differentiation of 32D cells into neutrophils. As shown in F i g . 3 A , growth of 32D cells by treatment with IL-3 was not affected whether they were transduced with any of the vectors, MXLLneo, MXL403Lneo, or MXLELneo. However, when 32D cells were transduced with the dominant-negative RAR!-expressing vectors (MXL403Lneo and MXLELneo), the 32D cells still continued to grow by treatment with the differentiationinducing cytokine G-CSF, although the cells transduced with the mock vector (MXLLneo) were differentiated into neutrophils and died by treatment with G-CSF within three weeks (F i g . 3B). The 32D cells transduced with the RARE-expressing vector showed less growth for the first two weeks but they grew well thereafter to the same levels as those cells transduced with the other dominant-negative RAR! (RAR403)-expressing vector.

II. Results A. Construction of retroviral vectors We have constructed three retroviral vectors (F i g . 1 ). All vectors are bicistronic and have the neomycin resistance gene (neo) as the second gene. An internal ribosome entry site (IRES) was used for bicistronic expression (Morgan et al., 1992). The first genes are laid between two loxP sites. MXLLneo is the mock vector only containing the neo gene and was used as the negative control vector. MXL403Lneo and MXLELneo have RAR403 and RARE genes, respectively, between two

FDCPmix is a murine multipotent hematopoietic cell line that is dependent on IL-3 for growth. IL-3 stimulates growth of FDCPmix cells but granulocyte-macrophage

294

Gene Therapy and Molecular Biology Vol 3, page 295

F i g . 1 Schematic structure of three retroviral vectors. All three vectors are bicistronic and have the neomycin resistant gene (neo) as the second gene. An internal ribosome entry site (IRES) was used for bicistronic expression. The first genes are laid between two loxP sites. MXLLneo is the mock vector only containing neo. MXL403Lneo and MXLELneo have the dominant-negative RAR! genes, RAR403 and RARE, respectively, between two loxP sites as the first genes.

F i g . 2 Expression of the dominantnegative RAR! genes in 32D and FDCPmix cells. (A , t o t h e l e f t ) Northern blotting of the FDCPmix cells transduced with either MXLLneo or MXL403Lneo vector. Total cellular RNA (20 Âľg/lane) was applied into an agarose gel, transferred to a membrane, and hybridized with a radiolabeled RAR! gene. A fulllength and a spliced form of the RAR403 gene were expressed in MXL403Lneo-transduced FDCPmix cells. (B , t o t h e r i g h t ) Immunoblotting of the 32D cells transduced with either MXLLneo, MXL403Lneo, or MXLELneo vector. Total cell lysates (10 Âľg/lane) were applied into SDS-PAGE, transferred to a membrane, probed with antiserum against the RAR! protein. The molecular weight of the RAR403 is 50 kDa and that of RARE is 55 kDa.

295

Ogasawara et al: RAR genes for ex vivo expansion of hematopoietic stem cells

F i g . 3 Growth of the transduced 32D cells after stimulation with the growth-inducing cytokine IL-3 (A) and the differentiationinducing cytokine G-CSF (B). Growth of the transduced FDCPmix cells after stimulation with the growth-inducing cytokine IL-3 (C) and the differentiation-inducing cytokine GM-CSF (D). Both cells transduced with the dominant-negative RAR!-expressing vectors (MXL403Lneo and MXLELneo) were not differentiated but continued to grow under the differentiation-inducing conditions, while cells transduced with the mock vector (MXLLneo) were differentiated into neutrophils and died within 3 weeks under the same conditions.

dominant-negative RAR!-expressing vectors (MXL403Lneo and MXLELneo) under the differentiationinducing conditions. The 32D and FDCPmix cells transduced with the mock vector were differentiated into neutrophils by day 15 and the cells died by day 24 under the differentiation-inducing conditions. However, the 32D and FDCPmix cells transduced with the dominant-negative RAR!-expressing vectors remained blastic or promyelocytic cells on day 24 and did not show neutrophilic differentiation even under the differentiationinducing conditions. There were no differences in morphological changes between the cells transduced with the vectors expressing the two forms of the dominantnegative RARs (RAR403 and RARE).

colony-stimulating factor (GM-CSF) stimulates differentiation of the cells into neutrophils. Similar growth curves were obtained to those of 32D cells. Growth of FDCPmix cells by treatment with IL-3 was not affected whether they were transduced with the mock vector (MXLLneo) or the dominant-negative RAR!-expressing vector (MXL403Lneo) as shown in F i g . 3 C . However, when FDCPmix cells were transduced with the dominantnegative RAR!-expressing vector, they continued to grow even in the presence of the differentiation-inducing cytokine GM-CSF, while FDCPmix cells transduced with the mock vector were differentiated into neutrophils and died by day 18 (F i g . 3 D ). F i g s . 4 A and 4B show the morphological changes of 32D and FDCPmix cells, respectively, which were transduced with the mock vector (MXLLneo) or the

296

Gene Therapy and Molecular Biology Vol 3, page 297 F i g . 4 (A) Morphology of transduced 32D cells after stimulation with the differentiation-inducing cytokine G-CSF. (B ) Morphology of FDCPmix cells after stimulation with the differentiation-inducing cytokine GM-CSF. Both cells transduced with the dominantnegative RAR!-expressing vectors (MXL403Lneo and MXLELneo) remained blastic or promyelocytic cells under the differentiation-inducing conditions, while cells transduced with the mock vector (MXLLneo) were differentiated into neutrophils and died by day 24 under the same conditions.

Fig.4A Fig. 4B

D. Effect of ATRA on differentiation block The block of differentiation of the transduced 32D cells into neutrophils could be overcome by treatment with alltrans retinoic acid (ATRA). F i g . 5 shows that ATRAinduced neutrophilic differentiation of 32D cells which were transduced with the dominant-negative RAR!–expressing vector (MXLELneo). The half-optimal concentration was about 1 µM and the optimal concentration was 10 µM. For comparison, the serum concentration of RA was estimated to be 1-10 nM.

III. Discussion Several groups have reported that dominant-negative RARs block differentiation of hematopoietic progenitor cells into neutrophils (Tsai et al., 1993; Saitou et al., 1994; Tsai et al., 1994). We have also confirmed that the dominant-negative mutants of the human RAR! gene inhibited the differentiation of 32D and FDCPmix cells into neutrophils by using bicistronic retroviral vectors. In these vectors, the mutant RAR! genes were laid between two loxP sites so that they could be removed by addition of the Cre recombinase, and a therapeutic gene could be placed as the second gene instead of neo. We have used in this paper two dominant-negative forms of the RAR! (RAR403 and RARE) genes. RAR403 is the prototype of the dominant-negative RAR!, which is the C-terminus-

297

Ogasawara et al: RAR genes for ex vivo expansion of hematopoietic stem cells Therefore, the strategy to use dominant-negative RARs might be applied to ex vivo expansion of hematopoietic stem cells in combination with some cytokines such as SCF, FLT-3 ligand, and thrombopoietin. These cytokines are able to stimulate the proliferation of immature hematopoietic cells but they also stimulate the differentiation of these cells to some degree. Dominantnegative RARs possibly inhibit differentiation but will not inhibit growth of the cells induced by these cytokines. However, this strategy needs to be examined in vivo, since there are no in vitro assays to determine exactly whether or not hematopoietic stem cells can be expanded. The differentiation ability of cells transduced with dominant-negative RAR!-expressing vectors should be restored after the ex vivo culture period for clinical applications. By treatment with ATRA, the differentiation block was overcome, leading to the production of mature hematopoietic cells as shown in F i g . 5 , suggesting that the transduced cells still retained the ability to differentiate. The vectors in this study were designed so that the mutant RAR! gene sequences could be removed from host cells after ex vivo culture. The vectors have loxP sites and the mutant RAR! genes can be eliminated by treatment with the Cre recombinase. We are now examining methods to remove the mutant RAR! genes from the integrated vector sequences in the transduced cells. For efficient removal of the genes inserted between loxP sites, high transfer efficiency of the Cre recombinase gene would be necessary although its transient expression would suffice.

F i g . 5 All-trans retinoic acid (ATRA)-induced neutrophilic differentiation of the transduced 32D cells. 32D cells were transduced with the dominant-negative RAR! (RARE)expressing vector to allow the cells to grow without differentiation even in the presence of the differentiationinducing cytokine G-CSF. These cells could be differentiated by treatment with ATRA. The horizontal axis shows the concentrations of ATRA and the vertical axis shows the ratio of neutrophils to all cells.

IV. Materials and Methods A. Cells and reagents deleted form (Tsai et al., 1993). RARE is another dominant-negative form which has a single amino acid residue substituted (Saitou et al., 1994). Both of them could inhibit the differentiation of 32D and FDCPmix cells into neutrophils. There were no significant differences between the two dominant-negative forms in terms of their ability to block differentiation.

Ecotropic packaging cell line BOSC23 cells were maintained in Dulbecco’s modified essential medium (Life Technologies, Rockville, MD) supplemented with 10% fetal calf serum (Filtron, Brooklyn, Australia). 32D cells were maintained in RPMI 1640 medium (Life Technologies) containing 10% fetal calf serum and 50 U/ml murine IL-3. FDCPmix cells were maintained in Fisher’s medium (Life Technologies) containing 20% horse serum and 100 U/ml murine IL-3. Human recombinant G-CSF was provided from Chugai Pharmaceutical (Tokyo, Japan). Murine GM-CSF was purchased from Life Technologies. Supernatant from C3H10T1/2 cells transfected with the mouse IL-3 expression plasmid was used as a source of murine IL-3 and it contains 10,000 U/ml murine IL-3. All-trans retinoic acid (ATRA) was purchased from Sigma (St. Louis, MO).

Tsai et al. showed that the expression of RAR403 in normal mouse bone marrow cells leads to a differentiation block in the neutrophil lineage at the promyelocytic stage (Tsai et al., 1993). They also showed that murine lymphohematopoietic progenitors, immortalized by a retroviral vector expressing RAR403, proliferate as an SCF-dependent clonal line that spontaneously generates pre-pro-B lymphocytes and myeloid progenitors. The developmental blocks imposed by the dominant-negative RAR! are mapped to the pre-CFU-GM as well as to the neutrophilic promyelocyte stages (Tsai et al., 1994). The block to CFU-GM formation may increase the probability of self-renewal of hematopoietic stem cells.

B. Retroviral plasmids The RAR403 gene (EcoRI-NheI fragment from pCMX403; Tsai et al., 1993) was subcloned into the EcoRI and SmaI site of pBS SK+ (pBS403). The EcoRI-BamHI fragment of pBS403 containing the RAR403 gene was inserted between two loxP sites of plox 2 (provided by Dr. J. D. Marth; Orban et al.,

298

Gene Therapy and Molecular Biology Vol 3, page 299 1992)(pL403L). The HindIII-SmaI fragment containing loxPRAR403-loxP of pL403L was inserted into the cloning site (HindIII-NotI site) of the retroviral plasmid pMX (provided by Dr. T. Kitamura; Onishi et al., 1996) (pMXL403L). The IRESneo sequence (XbaI-XhoI fragment of p1.1cIneo; Kodaira et al., 1998) was inserted into the SalI site of pMXL403L (pMXL403Lneo). pMXLLneo was obtained by removal of the RAR403 gene (BamHI-BamHI fragment) from pMXL403Lneo. pMXLELneo was obtained by inserting the RARE gene (EcoRI-BamHI fragment of pCMXRARE; Saitou et al., 1994) into the BamHI site of pMXLLneo. The dominant-negative RAR! (RAR403 and RARE) genes originated from the human RAR! gene. All restriction enzymes were purchased from Takara Shuzo (Shiga, Japan).

F. ATRA-induced differentiation 32D cells were transduced with the MXLELneo vector. When the cells grew in an exponential manner two or three weeks after transduction, ATRA was added to the culture medium at various concentrations. Forty-eight hours after treatment with ATRA, the cells were examined under a microscopy.

G. Antiserum The domain A (the N-terminal region) of the human RAR! gene was amplified by PCR. The primers were 5’-ATT GGA TCC ATG GCC AGC AAC AGC AGC TCC and 5’-TCA GAA TTC GGC TGG GGA TGG TGT GCT ATA. The PCR product was inserted into pGEX and the plasmid coding to the GST-fusion protein containing the domain A of the RAR! was transformed into BL21 strain of Escherichia coli. The resulting transformants were induced with isopropyl-1-thio-"-Dgalactopyranoside to produce a GST fusion protein. The bacteria were collected by centrifugation and resuspended in buffer containing 20 mM Tris-HCl pH 7.4, 50 mM ethylenediaminetetraacetic acid, 150 mM NaCl, and 1% Triton X-100. Vigorous sonication was performed followed by centrifugation. The GST fusion protein was purified through glutathion beads columns (Pharmacia, Piscataway, NJ). Rabbits were immunized against the GST-fusion protein and antiserum against the RAR! protein was prepared.

C. Retroviral vectors BOSC23 was transfected with retroviral plasmids by lipofectamine (Life Technologies) according to the manufacturer’s protocol. Two or three days after transfection, supernatants were harvested and filtered. They were used as ecotropic retroviral vectors.

D. Transduction Six-well plates were coated with retronectin (provided by Takara Shuzo) at the concentration of 96 µg/ml for 2 hr followed by blocking with 2% bovine serum albumin fraction V (Sigma) for 30 min. Cells were suspended in viral supernatants at the density of 2x10 5 /ml and 1 ml of cell suspension was added into each well. After 2-hour incubation at 37°C, another 1 ml of viral supernatants was added into each well. G418 selection (300 mg/ml active for FDCPmix cells and 800 µg/ml active for 32D cells) was started 24 hours after infection.

H. Immunoblotting Cells were lyzed with buffer containing 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 100 IU/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride. Ten µg of cell lysates was loaded into each well of 10% SDS-PAGE and electrotransferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA). Membranes were probed with antiserum against the RAR! protein using an ECL Western blotting detection kit (Amersham, Piscataway, NJ) according to the manufacturer’s protocol.

E. Growth assays and morphological examination Bulk 32D cells that were transduced and G418-resistant were resuspended in RPMI 1640 medium containing 10% fetal calf serum and 10 ng/ml human G-CSF for the differentiationinducing conditions, or in RPMI1640 medium containing 10% fetal calf serum and 50 U/ml murine IL-3 for the regular (proliferation-inducing) conditions. The cells were plated in 12-well dishes at the density of 1 x 10 5 cells per well. Cells were counted and replated at the density of 1 x 10 5 cells per well every three days. Bulk FDCP cells that were transduced and G418-resistant were suspended in Fisher’s medium containing 20% horse serum, 5 ng/ml murine GM-CSF and 1 U/ml murine IL-3 for the differentiation-inducing conditions, or in Fisher’s medium containing 20% horse serum and 100 U/ml murine IL-3 for the regular (proliferation-inducing) conditions. The cells were plated in 12-well dishes at the density of 5 x 10 4 cells per well. Cells were counted and replated at the density of 5 x 10 4 cells per well every three days. Cells were stained with Wright-Giemsa and were observed under a microscope.

I. Northern blotting Total cellular RNA was extracted with an RNA extraction kit Isogen (Nippon Gene, Tokyo, Japan). Twenty µg of RNA was loaded into each well of agarose gels and transferred onto HybondN+ (Amersham). Membranes were hybridized with a radiolabeled human RAR! gene (the EcoRI-BamHI fragment of pCMXhRAR!; Saitou et al., 1994). Radiolabeling of a probe was performed by using a DNA labeling kit (Amersham).

Acknowledgments We are grateful to Dr. Jamey D. Marth (Howard Hughes Medical Institute and Division of Cellular and Molecular Medicine, University of California San Diego, CA) for plox 2 and to Dr. Toshio Kitamura (Institute of

299

Ogasawara et al: RAR genes for ex vivo expansion of hematopoietic stem cells treatment of acute promyelocytic leukemia. B l o o d 72, 567-572, 1 9 8 8

Medical Science, University of Tokyo, Tokyo, Japan) for pMX. We also thank Takara Shuzo Co., Ltd. (Shiga, Japan) for supplying us retronectin. This work was supported in part by grants from the Ministry of Health and Welfare of Japan, and by grants-in-aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan.

Kodaira H, Kume A, Ogasawara Y, Urabe M, Kitano K, Kakizuka A, Ozawa K, Fas and mutant estrogen receptor chimeric gene, A novel suicide vector for tamoxifeninducible apoptosis. J p n J C a n c e r R e s 89, 741-747, 1998 Morgan RA, Couture L, Elroy-Stein O, Ragheb J, Moss B, Anderson WF, Retroviral vectors containing putative internal ribosome entry sites, Development of a polycistronic gene transfer system and applications to human gene therapy. N u c l e i c Acids R e s 20, 12931299, 1 9 9 2

References Alcalay M, Zangrilli D, Pandolfi P, Longo L, Mencarelli A, Giacomucci A, Rocchi M, Biondi A, Rambaldi A, Lo Coco F, Diverio D, Donti D, Donti E, Grignani F, Pelicci P, Translocation breakpoint of acute promyelocytic leukemia lies within the retinoic acid receptor a locus. P r o c N a t l Acad Sci USA 88, 1977-1981, 1 9 9 0

Onishi M, Kinoshita S, Morikawa Y, Shibuya A, Phillips J, Lanier LL, Gorman DM, Nolan GP, Miyajima A, Kitamura T, Application of retrovirus-mediated expression cloning. Exp Hematol 24, 324-329, 1 9 9 6

Bergemann J, Kuehlcke K, Fehse B, Ratz I, Ostertag W, Lother H, Excision of specific DNA-sequences from integrated retroviral vectors via site-specific recombination. N u c l e i c A c i d s R e s 23, 4451-4456, 1 9 9 5

Orban PC, Chui D, Marth JD, Tissue- and site-specific DNA recombination in transgenic mice. P r o c N a t l A c a d S c i USA 89, 6861-6865, 1 9 9 2

Borrow J, Goddard AD, Sheer D, Solomon E, Molecular analysis of acute promyelocytic leukemia breakpoints cluster region on chromosome 17. S c i e n c e 249, 15771580, 1 9 9 0

Saitou M, Narumiya S, Kakizuka A, Alteration of a single amino acid residue in retinoic acid receptor causes dominant-negative phenotype. J B i o l Chem 269, 19101-19107, 1 9 9 4

Breitman TR, Selonick SE, Collins SJ, Induction of differentiation of the human promyelocytic leukemia cell line (HL-60) by retinoic acid. Proc Natl Acad S c i USA 77, 2936-2940, 1 9 8 0

Tisdale JF, Hanazono Y, Sellers, SE, Agricola BA, Metzger, ME, Donahue RE, Dunbar CE, Ex vivo expansion of genetically marked rhesus peripheral blood progenitor cells results in diminished long-term repopulating ability. B l o o d 92, 1131-1141, 1 9 9 8

Brugger W, Heimfeld S, Berenson RJ, Mertelsmann R, Kanz L, Reconstitution of hematopoiesis after high-dose chemotherapy by autologous progenitor cells generated ex vivo. N Engl J Med 333, 283-287, 1 9 9 5

Tsai S, Bartelmez S, Sitnicka E, Collins S, Lymphohematopoietic progenitors immortalized by a retroviral vector harboring a dominant-negative retinoic acid receptor can recapitulate lymphoid, myeloid, and erythroid development. Genes D e v 8, 2831-2841, 1994

Collins SJ, Robertson K, Mueller L, Retinoic acid-induced granulocytic differentiation of HL-60 myeloid leukemia cells is mediated directly through the retinoic acid receptor (RAR!). M o l C e l l B i o l 10, 2154-2161, 1 9 9 0

Tsai S, Collins SJ, A dominant negative retinoic acid receptor blocks neutrophil differentiation at the promyelocyte stage. Proc Natl Acad S c i USA 90, 7153-7157, 1993

Dale EC, Ow DW, Gene transfer with subsequent removal of the selection gene from the host genome. P r o c N a t l A c a d Sci USA 88, 10558-10562, 1 9 9 1 De The H, Chomienne C, Lanotte M, Degos L, Dejean A, The t(15;17) translocation of acute promyelocytic leukemia fuses the retinoic acid receptor a gene to a novel transcribed locus. Nature 347, 558-561, 1 9 9 0 Dunbar CE, Young NS, Gene marking and gene therapy directed at primary hematopoietic cells. Curr Opin Hematol 3, 430-437, 1 9 9 6 Evans RM, The steroid and thyroid hormone receptor superfamily. S c i e n c e 240, 889-895, 1 9 8 8 Holyoake TL, Alcorn MJ, Richmond L, et al., CD34 positive PBPC expanded ex vivo may not provide durable engraftment following myeloablative chemoradiotherapy regimens. Bone Marrow Transplant 19, 1095-1101, 1997 Huang M-E, Ye Y-C, Chen S-R, Chai J-R, Lu J-X, Zhoa L, Gu HT, Wang Z-Y, Use of all-trans retinoic acid in the

300

Gene Therapy and Molecular Biology Vol 3, page 301 Gene Ther Mol Biol Vol 3, 301-310. August 1999.

Optimized expression of serotonin receptors in mammalian cells using inducible expression systems Research Article

Peter Vanhoenacker1*, Walter Gommeren2, Walter H.M.L. Luyten3, JosĂŠe E. Leysen2 and Guy Haegeman1 1

Unit of Eukaryotic Gene Expression and Signal Transduction, Department of Molecular Biology, University of Gent and Flanders Interuniversity Institute for Biotechnology, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium. 2 Department of Biochemical Pharmacology, 3 Department of Functional Genomics, 2,3 Janssen Research Foundation, Turnhoutseweg 30, B2340 Beerse, Belgium. __________________________________________________________________________________________________ * Corresponding author: Tel: 32.9.2645135; Fax: 32.9.2645304; E-mail: petervh@dmb.rug.ac.be A b b r e v i a t i o n s : 5-HT, 5-hydroxytryptamine; IFN, interferon; IL, interleukin; PK, protein kinase. Key words: serotonin receptors, interleukin-6 promoter, interferon, Mx promoter, tetracycline-inducible, interleukin-6, neurotransmitter receptors Received: 22 October 1998; accepted: 30 October 1998

Summary Efficient expression of neurotransmitter receptor proteins in a pure and active form has become an indispensable tool for modern pharmaceutical research. Different expression systems for producing receptor proteins have been used with varying success, but the expression levels are often low or not stable over a long period of time. We evaluated different inducible expression systems for the stable, high-level expression o f several serotonin receptors. U s i n g the human interleukin-6 promoter, which i s inducible by a variety o f biological and chemical agents, o n l y modest expression levels were obtained. Most likely, this is due to a down-regulation of the receptors by the inducing agents used. More successful was the type I interferon-inducible Mx promoter, with which high-level and stable expression o f four different serotonin receptors was obtained for several months. Finally, the tetracycline-inducible expression system was also tested and resulted in a still higher expression, with induction levels varying from 10- tot 700-fold.

distributed in the brain and the peripheral tissues, and which has been implicated in a wide variety of behavioral and physiological processes (Boess and Martin, 1994; Hoyer et al., 1994). Molecular cloning studies have shown the existence of 14 different genes, each encoding a distinct 5-HT receptor subtype (Lucas and Hen, 1995). In view of the development of highly selective and potent therapeutic agents, thousands of compounds need to be screened, not only for the effect on the receptor of interest, but also for their possible interactions with other related receptor subtypes. For this reason, efficient and long-term stable expression of these neurotransmitter receptor proteins in an active form has become an indispensable tool. As this goal is not always achieved with various receptor subtypes, most probably due to toxic effects and counter-selection on

I. Introduction In the last few years, molecular biology has had an enormous impact on the pharmaceutical industries. In their search of new pharmaceutics, the molecular target at which the drug is aimed, plays a pivotal role. Often, such targets are species-specific receptors, which are not readily available in sufficient quantity, especially those of human origin. Human receptors are preferred because homologous receptors from animals do not necessarily have the same characteristics. Therefore, cloned human receptors have become an essential and indispensable instrument in modern pharmaceutical companies (Luyten and Leysen, 1993). Serotonin (5-HT, 5-hydroxytryptamine) is a phylogenetically ancient neurotransmitter which is widely 301

Vanhoenacker et al: Expression of serotonin receptors using the Mx or the tetracycline-inducible promoter the producing host cells, we have addressed this problem using inducible expression systems.

activated by tyrosine phosphorylation and associates with ISGF-3" ; this complex then translocates to the nucleus, where it binds to the ISRE sequence, resulting in activation of transcription (reviewed in David, 1995).

Over the years, many regulatable expression systems have been developed and evaluated, ranging from heatshock- and heavy-metal-ion-inducible systems to the more recently developed tetracycline (tet)- and ecdysone-inducible systems.

C. The tetracycline-inducible system In the original tet system of Gossen and Bujard (1992), the E. coli tet repressor (TetR) has been fused to the activation domain of Herpes virus VP16, thus creating the transactivator protein tTA. Transcriptional activation results from binding of tTA to tet operator sequence elements (tetO), flanking a minimal RNA polymerase II promoter which drives the gene of interest. When the antibiotic tetracycline, or its analogs, binds to the TetR subunit, it abolishes DNA binding and hence activation by tTA. Correspondingly, only low background transcription was observed in the presence of tet, and a dramatic induction of reporter gene expression (up to 100,000-fold) was obtained upon withdrawal of tet (Gossen and Bujard, 1992). After chemical mutagenesis of the TetR, a mutant tTA has been created that displays the reverse properties of the original transactivator, i.e. rtTA, switching on the expression of reporter proteins upon addition of the antibiotic (Gossen et al., 1995). Hereby, three orders of magnitude of induction were obtained with a variety of tet analogs, of which doxycyline (dox) was found to be most efficient.

In this paper, we evaluate the efficiency, the advantages and the drawbacks of three different inducible systems: (i) the human interleukin-6 (IL-6) promoter; (ii) the murine Mx-promoter; and (iii) a tet-controlled expression system. Furthermore, and in contrast to recent review papers (Clackson, 1997; Saez et al., 1997; Burcin et al., 1998; Gingrich and Roder, 1998; Rossi and Blau, 1998), we have studied these different expression systems in one single cell line for the production in an active form of the same or comparable receptor proteins. This provides a more realistic and reliable comparison of the different systems.

A. The human interleukin-6 promoter IL-6 is a multifunctional cytokine that plays an active role in immunological responses, inflammation, bone metabolism, reproduction, neoplasia, and aging. It can be expressed in a variety of cell types, including epithelial cells and fibroblasts, T cells, monocytes, macrophages and some tumors (for a review, see Hirano et al., 1990). We have previously isolated the corresponding cDNA as well as a genomic clone, including a 1.2-kb fragment of the 5’flanking region, which contains all elements necessary for its induction (Haegeman et al., 1986; Ray et al., 1988; Dendorfer et al., 1994). The IL-6 gene can be activated by various agents, including cytokines such as tumor necrosis factor (TNF) and IL-1, lipopolysaccharide, bacteria and viruses, 12-O-tetradecanoyl-phorbol 1-3-acetate (TPA), dsI:C and/or cycloheximide (CHX) (Dendorfer et al., 1994; Vanhoenacker et al., 1994; Haegeman and Fiers, 1995).

II. Results A. Use of the human IL-6 promoter In previous studies, we have already shown that heterologous expression of SV40 T antigen under control of the IL-6 promoter was equally well induced as the endogenous IL-6 protein in the human cell lines MG63 and HeLa H21 (Vanhoenacker et al., 1994). Here, we have used the same human (h)IL-6 promoter fragment for the stable, tightly-regulated and high-level expression of serotonin receptors in the mouse fibrosarcoma cell line L929. To that end, L929 cells were transfected with the expression plasmids pIL6-5HT1A or pIL6-5HT2A, in which the cDNAs for the human 5-HT 1A and 5-HT2A receptors, respectively, were placed under the control of the hIL-6 promoter. From each transfection experiment, 24 individual G418-resistant colonies were selected, and tested for receptor expression, using radioligand binding assays, before and after induction with the combination of 110 IU/ml IL-1, 2 mM N6,2’-Odibutyryladenosine 3’:5’ cyclic monophosphate (dbcAMP) and 10 µM Ca 2+-ionophore (A23187). First, no measurable receptor expression could be demonstrated. This was not due to a failure of promoter stimulation, as the endogenous IL-6 gene was shown to be highly induced in the same

B. The murine Mx promoter The Mx protein is capable of mediating resistance in mice to influenza, measles and vesicular stomatitis viruses (Staeheli et al., 1986b; Meier et al., 1988; Pavlovic et al., 1990; Zürcher et al., 1992). Mx gene expression is strictly controlled at the transcriptional level by type I interferon (IFN) (Staeheli et al., 1986a; Horisberger et al., 1990). The promoter of the murine Mx1 gene, which was first characterized by Hug and coworkers (1988), contains an Sp1 binding site next to the TATA box, and a copy of the highly conserved “IFN stimulation response element” (ISRE) at position -131 to -120, relative to the start site of transcription (Staeheli et al., 1984; Reid et al., 1989). After binding of type I IFN to its specific cell-surface receptor, the IFN-stimulated gene factor ISGF-3! is 302

Gene Therapy and Molecular Biology Vol 3, page 303

B. Mx promoter-controlled expression of serotonin receptors

experiment. Furthermore, Southern-blot experiments confirmed the stable incorporation of the respective cDNAs in the genome, and for several clones the presence of receptor-specific mRNA could be demonstrated by RTPCR (data not shown). However, as it became evident from the literature, certain neurotransmitter receptors, like the human thyroid-stimulating-hormone receptor (hTSHR) and the human 5-HT1A receptor, could be desensitized by cAMP-dependent kinases (Harrington et al., 1994; Tezelman et al., 1994); for this reason we have attempted to circumvent this possible desensitization. To this end, various clones were re-induced with the combination of IL1/staurosporine/poly(rI).poly(rC) (110 IU/ml; 6 µM; 50 µg/ml). Measurable receptor expression could be demonstrated only for the h5-HT2A receptor (Table 1). The expression levels thus obtained were rather low; this could, most likely, be a consequence of the desensitizing kinase activities, mediated by the inducing agents used. Therefore, this possible negative regulatory effect was further investigated using Mx promoter-controlled expression of these receptors (see section C).

Using a 1,600-bp fragment of the Mx1 promoter, heterologous expression of human growth hormone has been obtained in VERO cells upon induction with human type I IFN (Lleonart et al., 1990). We have extended the usefulness and inducibility of this promoter for heterologous expression in the murine cell line L929 using the bacterial chloramphenicol acetyltransferase (CAT) as a reporter system; later on we have also achieved strictly IFN-controlled expression of five different human serotonin receptors (5-HT1A, 5-HT 2A, 5-HT 1B, 5-ht 1E and 5-ht1F) in this cell line (Vanhoenacker et al., 1997). Figure 1 shows the results of different selected clones, expressing the 5-HT 1B or the 5-HT 2A receptor. The expression levels for the 5-HT1B receptor varied from 300 fmol/mg protein up to 3,000 fmol/mg protein, with induction ratios of induced versus noninduced expression ranging from 6- to 40-fold. For the 5-HT2A receptor, the expression levels were similar, although here the induction ratios were slightly higher (7- to 49-fold). An overview of the maximum expression levels obtained for the different serotonin receptors, tested so far, is given in T a b l e 2. Bmax values varied from 700 fmol binding/mg protein for the 5-ht1F receptor, 3,100 fmol/mg protein for the 5-HT2A receptor, 3,300 fmol/mg protein for the 5HT1B receptor, 9,800 fmol/mg protein for the 5-ht 1E receptor, and up to 10,400 fmol/mg protein for the 5-HT1A receptor.

T a b l e 1 . IL-6 promoter-controlled expression of the human 5-HT2A receptor clone number

[ 125 I ] 5 - I - R 9 1 1 5 0 b i n d i n g (fmol/mg protein) -

+

1

2

63

2

17

92

3

15

85

4

15

166

5

19

115

6

35

115

mock

23

29

rat frontal cortex

T a b l e 2 . Overview of Mx promoter-controlled expression of serotonin receptors in L929 cells Receptor

Bmax (fmol/mg protein) (IFN-induced levels)

5-HT1A

10,400

5-HT1B

3,300

5-ht 1E

9,800

5-ht 1F

700

5-HT2A

3,100

512

L929 cells were cultured and induced as described in Materials and Methods; membrane preparation and radioligand binding studies were also carried out as indicated before. ‘-’ stands for noninduced; ‘+’ for induction with 110 IU/ml IL-1; 6 µM staurosporine; 50 µg/ml poly(rI).poly(rC) for a period of 24 hours at 37°C. Rat frontal cortex was used as a positive control.

_______________________________________________ Receptor-expressing L929 cells were cultured and induced for 24 hours with 1,000 U/ml mIFNb. Membrane preparation and radioligand binding studies were carried out as indicated before. Bmax values were derived as described in Leysen et al., 1996.

303

Gene Therapy and Molecular Biology Vol 3, page 304

F i g u r e 1 . Mx promoter-controlled expression of the 5-HT1B and 5-HT2A receptor subtypes. Parallel subconfluent monolayers of different G418-resistant transfectants, grown in 60 cm2 plates, were either left noninduced (blue bars) or were induced with 1,000 U/ml mIFN# (green bars) for 24 hr at 37째C. Receptor binding was measured as described before. The induction rate is shown by a number above the bars. M stands for mock-transfected cells.

these serotonin receptors are equal to or far better than values published in the literature (Hamblin et al., 1992; Zgombick et al., 1992; Van Huizen et al., 1993; Grotewiel and Sanders-Bush, 1994; Harrington et al., 1994; Langlois et al., 1996), we are persuaded that this system meets the requirements of the present-day pharmaceutical industry.

In our hands the expression levels remained stable for at least one year in continuous culture, and therefore we feel that these serotonin receptor-expressing cells are a reliable source of subtype-specific receptor material for characterizing the pharmacological profile of therapeutic agents as well as for functional studies. Taking into account that the expression levels obtained with most of

304

Gene Therapy and Molecular Biology Vol 3, page 305

C. Do dbcAMP and/or Ca-ionophore negatively affect serotonin receptor expression?

D. Tetracycline-inducible expression of neurotransmitter receptors As IFN is species-specific and not always readily available in large quantities, the Mx promoter-controlled expression system cannot easily be extrapolated to other cell types. Therefore, we have also evaluated the tetregulated system for the inducible expression of the serotonin 5-HT 1B receptor and the 5-ht1F receptor. To this end, a number of reasons led us to choose the recently developed “reverse” system in which the VP16 activation domain is fused to a mutant tetracycline repressor protein (rtTA), for direct induction by the antibiotic.

As the Mx promoter has turned out to be satisfactory for stable expression of serotonin receptors in L929 cells, we further investigated whether the conditions, used previously for the induction of the hIL-6 promoter, were deleterious for the expression of those receptors. Therefore, four different cell clones, expressing the 5-HT2A receptor under control of the Mx promoter, were induced with either IFN alone, or with IFN in combination with dbcAMP, or with dbcAMP and the Ca-ionophore together. The results of the radioligand binding studies are shown in Table 3. For all four clones, dbcAMP alone has only a small negative effect, while the addition of dbcAMP + Ca-ionophore leads to a huge reduction in radioligand binding as compared to the induction with IFN alone. The most plausible explanation is that this proceeds via a kinase-dependent down-regulation. PKA-dependent down-regulation has already been demonstrated for the #2-adrenergic receptor (Liggett et al., 1993) and a PKC-mediated down-regulation of the #1-adrenergic receptor and the ! 2A-adrenergic receptor has been described recently (Li et al., 1998; Liang et al., 1998). These results may thus explain why only a modest detection of receptors was obtained using the inducible hIL6 promoter, and studies to further investigate the presumed down-modulation are currently being performed.

First, we transfected the DNA coding region of rtTA into the cell line L929; selected cell clones were tested for rtTA expression by transient transfection with the vector pUHC13-3, in which the firefly luciferase gene is under control of a minimal CMV promoter, flanked by 7 tet operator sequences. Using this approach several wellregulated rtTA+ cell clones were retained; some of them were further stably transfected with the vectors pTet-5HT1B or pTet-5ht1F, in which the cDNA coding for the human 5HT1B or the 5-ht1F receptor, respectively, is positioned under the control of the minimal CMV promoter and the tet operator sequences. After selecting the appropriate colonies, cells were induced for 24 hours with 1 µg/ml dox and assayed for receptor expression by radioligand binding. The results are shown in Figure 2. For the 5-HT1B receptor, the expression levels varied between 4,000 and 20,000 fmol/mg protein with induction ratios of 50- to 200-fold; for the 5-ht1F receptor, expression levels of 9,500 fmol/mg protein were obtained with an induction ratio of 700-fold. These expression levels are higher than the values reported in the literature so far. Taking into account that the inducing agent is inexpensive and readily available, and considering the fact that this system may be less cell type-dependent, it can become a valuable alternative to the high-level production of neurotransmitter receptor proteins using Mx promoter-driven expression.

T a b l e 3 . Influence of dbcAMP and Ca-ionophore on IFNinduced expression of the human 5-HT2A receptor in L929 cells [ 125 I ] 5 - I - R 9 1 1 5 0 b i n d i n g (fmol/mg protein) mIFNb (1,000 U/ml)

+

+

+

dbcAMP (2mM)

-

+

+

Ca-ionophore ( 1 0 µM )

-

-

+

III. Discussion

clone 5

982

489

110

clone 13

2,270

1,875

353

clone 14

1,166

885

301

clone 22

1,345

976

262

Many different subtypes of (serotonin) receptors have been identified and cloned in the last few years; for the characterization of the pharmacological profile of candidate drug compounds, sufficient amounts of these receptors need to be available in a biologically active form. As tissue material is not always readily available and as it usually contains a variety of different receptor subtypes, heterologous expression of cloned receptor subtypes for screening programs has become a real necessity for a modern pharmaceutical company.

Different clones of Mx promoter-controlled 5-HT2A receptorexpressing cells were induced for 24 hours with the indicated reagents. Cells were then stored at -70°C and membrane preparation and radioligand binding studies were performed as described before.

Different constitutive expression systems have been used to produce receptor proteins, but with varying success, as in many cases expression levels are low and/or 305

Gene Therapy and Molecular Biology Vol 3, page 306

F i g u r e 2 . Dox-controlled expression of the 5-HT1B and 5-ht 1F receptor subtypes in a L929 rtTA+ cell line. Parallel subconfluent monolayers of different hygromycin-resistant transfectants, grown in 60 cm 2 plates, were either left noninduced (white bars) or were induced with 1µg/ml dox (black bars) for 24 hr at 37°C. Receptor binding was measured as described. The induction rate is shown by a number above the bars; in case the expression level of the noninduced cells was below the detection limit, the induction ratios (marked *) were obtained by taking the average expression level of the mock-transfected cells (M) as a background value.

disadvantages of three different inducible systems (i.e. the hIL-6 promoter, the murine Mx-promoter and the tetinducible system) for the expression of serotonin receptors in the murine cell line L929.

not stable over a long period of time (Zaworski et al., 1995). Therefore, we have addressed this problem by using ‘inducible’ expression systems as a possible valuable alternative and have evaluated the efficacy, advantages and 306

Gene Therapy and Molecular Biology Vol 3, page 307 First, we obtained no or only low level expression, as determined by radioligand binding assays, with the hIL-6 promoter, which was already successfully used for heterologous protein production in earlier studies (Vanhoenacker et al., 1994). During subsequent experiments using the Mx promoter, however, we demonstrated that some of the inducing agents used may be deleterious for serotonin receptor expression, probably by activating kinase pathways which may result in receptor phosphorylation and down-regulation. Currently, immunocytochemical studies are under investigation to further explore this phenomenon.

induction upon addition of dox. For L929 cells, for example, we have tested 48 G-418 resistant clones several times by transient transfection with pUHC13-3 in order to select at least several strictly regulatable and inducible cell clones. Although this extensive testing is labor-intensive and time-consuming, we feel that it is essential in order to obtain well-regulated expression of the gene of interest at a later stage. It is, however, not always possible: as with HEK293 cells, a more substantial leak in expression was obtained (data not shown). In summary, inducible expression has proven to be a good option to obtain high-level, stable expression of neurotransmitter receptors, which could be readily used for comparative binding studies and adequate drug screening.

On the other hand, the use of the murine Mx promoter, which is inducible by type I IFN, proved to be very successful. With this promoter system, we were able to generate five different, biologically active serotonin receptors; their expression levels, after induction, ranged from 700 fmol/mg protein for the 5-ht 1F receptor to up to 10,400 fmol/mg protein for the 5-HT1A receptor. In addition, the same promoter system was also found to be successful for the expression of two dopamine receptor variants, i.e. the human dopamine D3 receptor and dopamine D4 receptor (data not shown). As the expression levels measured remained stable for a long period of time (i.e. more than 1 year), Mx promoter-driven expression may be considered as a valuable and reliable system for the generation of proteins of pharmaceutical interest. Furthermore, it should be noted that this system has also been successfully used to express toxic compounds and carry out analytical studies (Vandevoorde et al., 1997; Boone et al., 1999). This system has, however, two major drawbacks. First, the inducing agent IFN is species-specific and thus not always readily available; extrapolation to other cell types is not obvious, although CHO and NIH3T3 cells were also found to be responsive to mouse (m)IFN#, regarding Mx promoter-driven expression of reporter genes (our unpublished results). Second, the inducing agent IFN is a cytokine and thus available in only limited amounts, which makes the system rather expensive and less attractive, if large scale industrial production is envisaged.

IV. Materials and Methods A. Cell lines MG63 (human osteogenic sarcoma), HeLa H21 (human cervix carcinoma) and L929 (mouse fibrosarcoma) cells were cultivated in a controlled environment (37째C, 5% CO 2 , 98% humidity) in DMEM supplemented with L-glutamine (2 mM), penicillin (100 U/ml), streptomycin (0.1 mg/ml) and 10% FCS or 5% FCS/5% NCS, respectively.

B. Recombinant DNA constructs For the construction of pIL6-5HT1A , the complete cDNA of the human 5-HT 1A receptor was isolated as a Tfi I (filled in with Klenow DNA polymerase)/BamHI fragment from pSP64H5HT1A (Vanhoenacker et al., 1997) and cloned into a XhoI (filled in with Klenow DNA polymerase)/XhoII opened pBLHIL6CAT vector (Vanden Berghe et al., 1993). Hereby the coding region for the 5-HT1A receptor was placed directly under control of the hIL-6 promoter. The construction of pIL6-5HT 2A was similar. A StuI/BamHI fragment from pUC18 /5HT2A (Vanhoenacker et al., 1997), containing the cDNA coding for the human 5-HT2A receptor, was ligated to the same pBLHIL6CAT fragment as used for the construction of pIL65HT1A . For the construction of pTet-5HT1B and pTet-5ht1F , a NheI/BamHI fragment containing the complete cDNA coding for the human 5-HT 1B and 5-ht 1F receptor, respectively, was cloned into a XhoI/BamHI opened pUHD10-3 vector (Gossen and Bujard, 1992).

Another valuable alternative is the reverse tet-inducible expression system, by which very high expression levels of the serotonin 5-HT1B and 5-ht1F receptors were obtained, reaching to 20,000 fmol/mg protein and 9,500 fmol/mg protein, respectively. Background expression was rather low for most of the clones tested, which resulted in induction ratios of up to 700-fold. Due to the simplicity and the low cost of the inducing agent used, no major problems are expected for industrial exploitation or for its application to other cell types. However, it should be taken into consideration that this expression system involves two-components and, thus, requires the establishment of stable rtTA-expressing clones to allow a profound

pPHT was constructed by inserting the hygromycin gene as an XhoI/XbaI fragment between the XhoI and XbaI sites of pPNT (Tybulewicz et al., 1991).

C. Transfection procedure Stable transfections were essentially performed as described previously (Vanhoenacker et al., 1994). The pSV2neo plasmid (Southern and Berg, 1982) and the pPHT plasmid (see above) provided a resistance gene, and

307

Vanhoenacker et al: Expression of serotonin receptors using the Mx or the tetracycline-inducible promoter transfectants were selected by G418 (400 µg/ml) or hygromycin-B (250 U/ml), respectively, for a period of three weeks. The selective medium was renewed every 7 days.

induction by prostaglandins, cyclic AMP, and lipopolysaccharide. M o l . C e l l . B i o l . 14, 4443-4454. Gingrich JR and Roder J ( 1 9 9 8 ) Inducible gene expression in the nervous system of transgenic mice. A n n u . R e v . N e u r o s c i . 21, 377-405.

D. Induction of promoters

Gossen M and Bujard H ( 1 9 9 2 ) Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. P r o c . N a t l . A c a d . S c i . U S A 89, 55475551.

For induction, L929 cells were plated at a density of 4 x 10 cells/ cm 2 . The inducing agents and conditions used for induction of the hIL-6 promoter and the murine Mx promoter were described previously (Vanhoenacker et al., 1994; Vanhoenacker et al., 1997). In case of the tet-system, cells were induced for 24 hours with 1µg/ml dox. After the induction period, cells were stored at -70°C for membrane preparation and assaying by radioligand receptor binding (Leysen et al., 1996). 4

Gossen M, Freundlieb S, Bender G, Müller G, Hillen W, and Bujard H ( 1 9 9 5 ) Transcriptional activation by tetracyclines in mammalian cells. S c i e n c e 268, 17661769. Grotewiel MS and Sanders-Bush E ( 1 9 9 4 ) Regulation of serotonin 2A receptors in heterologous expression systems. J. Neurochem. 63, 1255-1260.

E. Radioligand binding studies

Haegeman G and Fiers W ( 1 9 9 5 ) TNF-induced mechanisms for IL6 gene induction. N A T O A S I S e r i e s H92, 375382.

Binding experiments with membrane preparations of 5HT1A , 5-HT2A , 5-HT1B and 5-ht1E receptor-expressing cells were performed essentially as described previously (Vanhoenacker et al., 1997). The 5-ht1F expression levels were determined by radioligand binding with [ 3 H]5-HT; non-specific binding was measured in the presence of a 200-fold excess of 5-HT.

Haegeman G, Content J, Volckaert G, Derynck R, Tavernier J, and Fiers W ( 1 9 8 6 ) Structural analysis of the sequence coding for an inducible 26-kDa protein in human fibroblasts. E u r . J . B i o c h e m . 159, 625-632. Hamblin MW, Metcalf MA, McGuffin RW, and (1992) Molecular cloning and characterization of a human 5-HT 1B serotonin homologue of the rat 5-HT1B receptor with pharmacological specificity. B i o c h e m . R e s . C o m m u n . 184, 752-759.

Acknowledgements We wish to thank N. De Coussemaker and I. Van den Bogerd for technical assistance. P. Vanhoenacker was supported by the IWT (Vlaams Instituut voor de Bevordering van het Wetenschappelijk-Technologisch Onderzoek in de Industrie). G. Haegeman is a Research Director with the FWO (Fonds voor Wetenschappelijk Onderzoek-Vlaanderen). Research was supported by the IWT.

Karpells S functional receptor: a 5-HT1D -like Biophys.

Harrington MA, Shaw K, Zhong P, and Ciaranello RD ( 1 9 9 4 ) Agonist-induced desensitization and loss of high-affinity binding sites of stably expressed human 5-HT1A receptors. J. Pharmacol. Exp. Ther. 268, 1098-1106. Hirano T, Akira S, Taga T, and Kishimoto T ( 1 9 9 0 ) Biological and clinical aspects of interleukin 6. Immunol. Today 11, 443-449.

References

Horisberger MA, McMaster GK, Zeller H, Wathelet MG, Dellis J, and Content J ( 1 9 9 0 ) Cloning and sequence analyses of cDNAs for interferon- and virus-induced human Mx proteins reveal that they contain putative guanine nucleotide-binding sites: functional study of the corresponding gene promoter. J . V i r o l . 64, 1171-1181.

Boess FG and Martin IL ( 1 9 9 4 ) Molecular biology of 5-HT receptors. Neuropharmacology 33, 275-317. Boone E, Vandevoorde V, De Wilde G, and Haegeman G ( 1 9 9 9 ) Activation of p42/p44 mitogen-activated protein kinases (MAPK) and p38 MAPK by tumor necrosis factor (TNF) is mediated through the death domain of the 55-kDa TNF receptor. FEBS lett. in press.

Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylecharane EJ, Saxena PR, and Humphrey PP ( 1 9 9 4 ) International Union of Pharmacology classification of receptors for 5-hydroxytryptamine (Serotonin). P h a r m a c o l . R e v . 46, 157-203.

Burcin MM, O'Malley BW, and Tsai SY ( 1 9 9 8 ) A regulatory system for target gene expression. Frontiers i n B i o s c i e n c e 3, 1-7. Clackson T ( 1 9 9 7 ) Controlling mammalian gene expression with small molecules. C u r r . O p i n . C h e m . B i o l . 1 , 210-218. interferon

Hug H, Costas M, Staeheli P, Aebi M, and Weissmann C ( 1 9 8 8 ) Organization of the Murine Mx gene and characterization of its interferon- and virus-inducible promoter. M o l . C e l l . B i o l . 8, 3065-3079.

Dendorfer U, Oettgen P, and Libermann TA ( 1 9 9 4 ) Multiple regulatory elements in the interleukin-6 gene mediate

Langlois X, El Mestikawy S, Arpin M, Triller A, Hamon M, and Darmon M ( 1 9 9 6 ) Differential addressing of 5-HT1A and 5-HT1B receptors in transfected LLC-PK1 epithelial

David M ( 1 9 9 5 ) Transcription factors in signaling. Pharmacol. Ther. 65, 149-161.

308

Gene Therapy and Molecular Biology Vol 3, page 309 cells: a model of receptor N e u r o s c i e n c e 74, 297-302.

targeting

in

neurons.

Rossi FMV and Blau HM ( 1 9 9 8 ) Recent advances in inducible gene expression. C u r r . O p i n . B i o t e c h n o l . 9, 451-456.

Leysen JE, Gommeren W, Heylen L, Luyten WHML, Van de Weyer I, Vanhoenacker P, Haegeman G, Schotte A, Van Gompel P, Wouters R, and Lesage AS ( 1 9 9 6 ) Alniditan, a new 5-hydroxytryptamine 1D agonist and migraineabortive agent: ligand-binding properties of human 5hydroxytryptamine1Da , human 5-hydroxytryptamine 1Db , and calf 5-hydroxytryptamine 1D receptors investigated with [3 H]5-hydroxytryptamine and [3 H]alniditan. M o l . Pharmacol. 50, 1567-1580.

Saez E, No D, West A, and Evans RM ( 1 9 9 7 ) Inducible gene expression in mammalian cells and transgenic mice. C u r r . O p i n . B i o t e c h n o l . 8, 608-616. Southern PJ and Berg P ( 1 9 8 2 ) Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the SV40 early region promoter. J . M o l . A p p l . G e n e t . 1, 327-341. Staeheli P, Danielson P, Haller O, and Sutcliffe JG ( 1 9 8 6 a ) Transcriptional activation of the mouse Mx gene by type I interferon. M o l . C e l l . B i o l . 6, 4770-4774.

Li Z, Vaidya VA, Alvaro JD, Iredale PA, Hsu R, Hoffman G, Fitzgerald L, Curran PK, Machida CA, Fishman PH, and Duman RS ( 1 9 9 8 ) Protein Kinase C-mediated downregulation of #1 -adrenergic receptor gene expression in rat C6 glioma cells. M o l . P h a r m a c o l . 54, 14-21.

Staeheli P, Haller O, Boll W, Lindenmann J, and Weissmann C ( 1 9 8 6 b ) Mx protein: Constitutive expression in 3T3 cells transformed with cloned Mx cDNA confers selective resistance to influenza virus. C e l l 44, 147-158.

Liang M, Eason MG, Jewell-Motz EA, Williams MA, Theiss CT, Dorn II GW, and Liggett SB ( 1 9 9 8 ) Phosphorylation and functional desensitization of the a2A -adrenergic receptor by Protein Kinase C. M o l . Pharmacol. 54, 44-49.

Staeheli P, Horisberger MA, and Haller O ( 1 9 8 4 ) Mxdependent resistance to Influenza Viruses is induced by mouse interferons ! and # but not ". V i r o l o g y 132, 456461.

Liggett SB, Freedman NJ, Schwinn DA, and Lefkowitz RJ ( 1 9 9 3 ) Structural basis for receptor subtype-specific regulation revealed by a chimeric #3 /#2 -adrenergic receptor. P r o c . N a t l . Acad. S c i . USA 90, 36653669.

Tezelman S, Shaver JK, Grossman RF, Liang W, Siperstein AE, Duh Q-Y, and Clark OH ( 1 9 9 4 ) Desensitization of adenylate cyclase in Chinese Hamster Ovary cells transfected with human thyroid-stimulating hormone receptor. E n d o c r i n o l o g y 134, 1561-1569.

Lleonart R, N채f D, Browning H, and Weissmann C ( 1 9 9 0 ) A novel, quantitative bioassay for type I interferon using a recombinant indicator cell line. B i o / T e c h n o l o g y 8, 1263-1267.

Tybulewicz VLJ, Crawford CE, Jackson PK, Bronson RT, and Mulligan RC ( 1 9 9 1 ) Neonatal lethality and lymphopenia in mice with a homozygous disruption of the c-abl proto-oncogene. C e l l 65, 1153-1163.

Lucas JJ and Hen R ( 1 9 9 5 ) New players in the 5-HT receptor field: genes and knockouts. T r e n d s P h a r m a c o l . S c i . 16, 246-252.

Van Huizen F, Bansse M-T, and Stam NJ ( 1 9 9 3 ) Agonistinduced down-regulation of human 5-HT1A and 5-HT2 receptors in Swiss 3T3 cells. Neuroreport 4, 13271330.

Luyten WHML and Leysen JE ( 1 9 9 3 ) Receptor cloning and heterologous expression towards a new tool for drug discovery. T r e n d s B i o t e c h n o l . 11, 247-254.

Vanden Berghe W, Haegeman G, and Fiers W ( 1 9 9 3 ) Studies on the inducibility of the IL6 promoter using a reporter gene construction. Arch. Intern. Physiol. B i o c h i m . B i o p h y s . 101, B35

Meier E, F채h J, Grob MS, End R, Staeheli P, and Haller O ( 1 9 8 8 ) A family of interferon-induced Mx-related mRNAs encodes cytoplasmic and nuclear proteins in rat cells. J . V i r o l . 62, 2386-2393.

Vandevoorde V, Haegeman G, and Fiers W ( 1 9 9 7 ) Induced expression of trimerized intracellular domains of the human tumor necrosis factor (TNF) p55 receptor elicits TNF effects. J . C e l l B i o l . 137, 1627-1638.

Pavlovic J, Z체rcher T, Haller O, and Staeheli P ( 1 9 9 0 ) Resistance to influenza virus and vesicular stomatitis virus conferred by expression of human MxA protein. J . V i r o l . 64, 3370-3375.

Vanhoenacker P, Fiers W, and Haegeman G ( 1 9 9 4 ) Studies on the induction of the interleukin-6 promoter in cell lines of human and simian origin. E u r . C y t o k i n e N e t w . 5, 283-291.

Ray A, Tatter SB, May LT, and Sehgal PB ( 1 9 8 8 ) Activation of the human "#2-interferon/hepatocyte-stimulating factor/interleukin 6" promoter by cytokines, viruses, and second messenger agonists. P r o c . N a t l . A c a d . S c i . USA 85, 6701-6705.

Vanhoenacker P, Gommeren W, Luyten WHML, Leysen JE, and Haegeman G ( 1 9 9 7 ) Stable, high-level expression of human serotonin receptors in L929 cells using an inducible expression system. Receptors and C h a n n e l s 5, 125-137.

Reid LE, Brasnett AH, Gilbert CS, Porter ACG, Gewert DR, Stark GR, and Kerr IM ( 1 9 8 9 ) A single DNA response element can confer inducibility by both !- and "interferons. P r o c . N a t l . A c a d . S c i . U S A 86, 840844.

Zaworski PG, Evans DL, Lahti RA, and Gill GS ( 1 9 9 5 ) Growth of Chinese hamster ovary (CHO) cells expressing the 5-HT 2 serotonin receptor in suspension culture: an efficient method for large-scale acquisition of membrane

309

Vanhoenacker et al: Expression of serotonin receptors using the Mx or the tetracycline-inducible promoter protein for drug evaluation. J . N e u r o s c i . M e t h o d s 56, 169-175. Zgombick JM, Schechter LE, Macchi M, Hartig PR, Branchek TA, and Weinshank RL ( 1 9 9 2 ) Human gene S31 encodes the pharmacologically defined serotonin 5hydroxytryptamine1E receptor. M o l . Pharmacol. 42, 180-185. Z端rcher T, Pavlovic J, and Staeheli P ( 1 9 9 2 ) Mouse Mx2 protein inhibits vesicular stomatitis virus but not influenza virus. V i r o l o g y 187, 796-800.

310

Gene Therapy and Molecular Biology Vol 3, page 311 Gene Ther Mol Biol Vol 3, 311-325. August 1999.

Identification of a negative regulatory mechanism for the repair of U5 long terminal repeat DNA by the human immunodeficiency virus type 1 integrase DNA polymerase Research Article

Brian E. Udashkin, Andrea Acel, Avi Shtvi, Benjamin Alt, Henry Triller, Mark A. Wainberg2 and Emmanuel A. Faust1* Lady Davis Institute for Medical Research, SMBD-Jewish General Hospital and Departments of Medicine1 and Microbiology and Immunology2, and McGill AIDS Center, McGill University, Montreal, Quebec, Canada H3T 1E2 __________________________________________________________________________________________________ * Corresponding author: Tel: (514) 340-8260; Fax: (514) 340-7502; E-mail: mdef@musica.mcgill.ca Received: 2 October 1998; accepted: 13 November 1998 (Communicated by Allan Wolffe)

Summary The quasi-random integration of retroviral DNA elements into the chromosomes of infected cells is believed to proceed by a four-step mechanism. The 3'- ends of the long terminal repeats (LTRs) are processed by the endonucleolytic cleavage and removal, usually, of a GT dinucleotide (step 1); the 3'-processed DNA ends are inserted at staggered nicks in the host DNA via a DNA strand transfer r e a c t i o n , s i m u l t a n e o u s l y g e n e r a t i n g s h o r t g a p s a t t h e s i t e s o f i n s e r t i o n (step 2); the gaps are repaired by a DNA polymerase (step 3); and the 5'-ends of the viral DNA are joined to the host DNA (step 4). Human immunodeficiency virus type 1 (HIV-1) integrase was previously reported to p o s s e s s e n z y m a t i c a c t i v i t i e s c a p a b l e o f p e r f o r m i n g a t l e a s t t h e f i r s t 3 s t e p s o f t h e integration process including an intrinsic DNA - dependent polymerase activity capable of short gap repair (Acel et. al., 1998 J. Virol. 72: 2062-2071). In the present study, the behavior of the integrase DNA polymerase was examined i n a DNA end-repair assay i n which the frequency of polymerization on 3'-processed HIV-1 U5 LTRs was examined. The frequency of polymerization was negatively regulated by the 5'-AC sequence comprising the 2-nucleotide template and by the sequence of the adjacent conserved 5'TG/CA dinucleotide. Mutations within these DNA elements of the LTR enhanced the polymerization frequency on 2-nucleotide templates between 3- and 100-fold. In most cases, the integrase DNA polymerase added only one nucleotide to 3'-processed LTRs even t h o u g h t h e D N A w a s c o m p r i s e d o f 2 - n u c l e o t i d e t e m p l a t e - p r i m e r s . T h i s l e v e l o f regulation was controlled by a DNA binding and/or zinc finger domain in the integrase protein. By contrast, the integrase DNA polymerase behaved i n a processive manner with homopolymeric pyrimidine templates, extending nascent DNA chains up t o at least 2 0 nucleotides, whereas DNA polymerization with an oligo dA template exhibited a lower processivity of 1-7 nucleotides. The results suggest a model whereby an interaction between integrase and specific DNA elements in the HIV-1 LTRs prevents the repair of 3'-processed LTRs by the integrase DNA polymerase. Drugs with the ability to alter this regulatory aspect of integrase DNA polymerase function and thus induce repair of processed LTRs are predicted to block integration of HIV-1 DNA, and thus have potentially lethal consequences for HIV-1 replication. model systems for the study of the mechanisms of repair and recombination in mammalian cells are needed to provide a means to study DNA integration processes. Retroviruses provide accessible model systems to study the integration of DNA elements into mammalian

I. Introduction The success of gene therapy in treating disease will ultimately depend on the ability of the target tissue to integrate foreign DNA into its chromosomes and therefore 311

Udashkin et al: HIV-1 integrase prevents the repair of 3'-processed LTRs chromosomes. The process of integration, whereby a double stranded DNA copy of a retroviral genome is joined to host cell DNA, is typified in the human immunodeficiency virus type1 (HIV-1) replicative cycle (Ansari-Lari et al., 1995; Cannon et al., 1994; Cara et al., 1995; Engelman et al., 1995; Englund et al., 1995; Lafemina et al., 1992; Sakai et al., 1993; Shin et al., 1994; Taddeo et al., 1994; Wiskerchen and Muesing, 1995a; Wiskerchen and Muesing, 1995b). In this process, two nucleotides are cleaved from the ends of the U3 and U5 portions of the HIV-1 long terminal repeats (LTRs) by the HIV-1 encoded integrase protein (Engelman et al., 1991; Tramontano et al., 1998; Vink, 1991; Vink et al., 1991). Cleavage occurs next to an invariant CA dinucleotide within each copy of the LTR, thus creating recessed 3' ends and 5' overhangs (Bushman and Craigie, 1991; LaFemina et al., 1991; Leavitt et al., 1992; Sherman and Fyfe, 1990; Vink et al., 1991). The 3' ends of the cleaved preintegration intermediate are then joined pairwise in a concerted manner to the host cell DNA (Brown, 1990; Craigie, 1992; Grandgenett and Mumm, 1990). The mechanism of DNA strand transfer, sometimes referred to as 3' end joining, involves a nucleophilic attack of adenylate hydroxyls on phosphodiester bonds located 5 base pairs apart on either strand of the host DNA. This concerted transesterification reaction produces an intermediate in which the viral genome is flanked by 5nucleotide gaps and is linked to the host DNA by its 3' ends (F i g . 1). Gap repair and removal of adjoining mispaired nucleotides (the so-called 5'-joining reaction) may be carried out either by the integrase itself (Acel et al., 1998) or by host cell DNA repair enzymes, although experimental evidence to support the latter notion is lacking (Roe et al., 1997). Characteristic 5-bp repeats flank the inserted provirus (Ellison et al., 1990; Vink et al., 1990).

Cleavage and integration reactions are catalyzed in vitro by purified recombinant HIV-1 integrase produced in bacteria (Asante-Appiah and Skalaka, 1997). The in vitro reactions utilize synthetic oligonucleotide substrates homologous to terminal portions of the U3 and U5 regions of the HIV-1 LTR (Bushman and Craigie, 1991; Bushman et al., 1990; Carteau et al., 1993; LaFemina et al., 1991; Sherman and Fyfe, 1990; Vincent et al., 1993; Vink and Plasterk, 1993; Vink et al., 1991). Integrase also catalyzes bimolecular disintegration reactions using cross bone substrates indicating that integrase forms dimeric structures that are able to coordinate two substrate molecules perhaps by utilizing two appropriately positioned DNA binding sites (Chow and Brown, 1994a; Chow and Brown, 1994b; Mazumder et al., 1994). The crystal structure of a central catalytic core domain (amino acid residues 50-212) confirms the existence of dimeric contacts (F i g . 2) (Dyda et al., 1994; Goldgur et al., 1998; Maignan et al., 1998). There is a sequence requirement for approximately 5 nucleotides in line to the CA dinucleotide at which cleavage occurs (Balakrishnan and Jonsson, 1997; Katzman et al., 1989; LaFemina et al., 1991; Vink et al., 1991). Cleavage of LTR sequences is specific for the homologous IN protein so that HIV-1 IN does not cleave the LTR sequences of either the avian sarcoma virus or the murine leukemia virus (Katzman et al., 1989; LaFemina et al., 1991). Purified integrase exists as a homodimer in solution (Vincent et al., 1993) but likely functions as a multimer in which individual subunits provide separate functions for DNA binding, cleavage and integration (Engelman et al., 1993; Kalpana and Goff, 1993; van Gent et al., 1993).

F i g . 1 . S c h e m a t i c i l l u s t r a t i o n o f a g a p p e d r e t r o v i r a l D N A i n t e g r a t i o n i n t e r m e d i a t e . The diagram depicts the structure at the ends of the retroviral U3 and U5 LTRs after the 3'-ends of the viral DNAhave been joined to host cell DNA. Also shown are 5 - nucleotide gaps and unpaired 5' -AC tails, representative of HIV-1. In HIV-1 LTRs the base pair adjacent to the unpaired 5'-AC tails is 5' - TG/CA.

312

Gene Therapy and Molecular Biology Vol 3, page 313

Fig. 2. Ribbon diagram of the 3-dimensional structure of the HIV-1 integrase catalytic core domain ( D y d a e t a l . , 1 9 9 4 ) . The illustration depicts a dimeric structure in which aspartate and glutamate residues (D64, D116 and E152) form a pocket believed to be required for catalysis. The 3-dimensional structure has features in common with RNase H and other polynucleotidyl transferases.

C-terminal nonspecific DNA binding domain (Engelman et al., 1993; Vincent et al., 1993; Vink et al., 1993; Woerner et al., 1992; Woerner and Marcus-Sekura, 1993). A role for the zinc finger motif in the cleavage reaction has also been recognized (Asante-Appiah and Skalaka, 1997; Engelman and Craigie, 1992; Khan et al., 1991; Lafemina et al., 1992; Van Gent et al., 1992). Binding of zinc ions to the N-terminal domain promotes tetramerization of full length integrase (Lee et al., 1997). A highly conserved central domain (amino acids 50-212) contains the catalytic center upon which all cleavage and strand transfer reactions are dependent (Skalka, 1993). Indeed, there is an absolute requirement in catalysis for conserved Asp64, Asp116 and Glu152 residues which make up the so called D,D,(35)E region (Bushman et al., 1993; Drelich et al., 1993; Drelich et al., 1992; Engelman and Craigie, 1992; Kulkosky et al., 1992; Lafemina et al., 1992; Leavitt et al., 1993). These residues form a pocket which likely binds divalent cations. In the present study we sought to define in greater detail the role of the integrase DNA polymerase in the integration process as exemplified by HIV-1. In terms of

Dimeric, tetrameric and oligomeric forms of avian sarcoma virus integrase have been detected by crosslinking experiments and by gel exclusion chromatography (Andrake and Skalka, 1995).Only multimeric forms of HIV-1 integrase display DNA polymerase activity (Acel et al., 1998). Specific regions of integrase required for multimerization have been localized to the catalytic and Cterminal domains (Donzella et al., 1998). In vivo integrase may be part of a preintegration complex in which it interacts with other viral and cellular proteins (Carteau et al., 1997; Miller et al., 1997) . Detailed analysis of the effects of a variety of mutations in integrase on the in vitro cleavage and integration reactions has allowed the definition of distinct protein domains (Vink and Plasterk, 1993). Separate Nand C-terminal domains in the integrase molecule are responsible for DNA binding. The N-terminal domain contains an HHCC region resembling a zinc finger motif (Bushman et al., 1993; Cai et al., 1997; Woerner and Marcus-Sekura, 1993) which may play a role in positioning integrase on the viral DNA together with the

313

Udashkin et al: HIV-1 integrase prevents the repair of 3'-processed LTRs the current models for integration, it is assumed that multimeric forms of integrase are assembled at the ends of the viral DNA where they carry out the enzymatic reactions that are necessary to effect the integration process. In considering this model, it became clear that a mechanism would have to exist to regulate the repair of the viral DNA ends once they had been processed by the endonuclease activity of integrase. Model DNA substrates representing processed viral DNA ends were therefore used to examine the behavior of the integrase DNA polymerase to determine if there exists a regulatory mechanism governed strictly by the structure of HIV-1 integrase and/or the DNA sequence of the HIV-1 U5 LTR.

TG/CA dinucleotide known to be necessary for the processing reaction (see below).

II. Results A. Repair of a 3'-processed HIV-1 U5 LTR terminus by the integrase DNA polymerase is dependent on template sequence F i g . 3 . Diagram o f h a i r p i n 2 - n u c l e o t i d e template-primers used a s substrates for DNA repair. The hairpin stem region designated as U5 in the diagram is 30 nucleotides in length including the two 5' TG/CA terminal base pairs and its sequence is homologous to the terminal 30 nucleotides of the HIV-1 U5 LTR (see Materials and Methods). The diagram also indicates how the template sequences (A) or the internal dinucleotide sequences (B) were varied to produce different substrates for end repair reactions.

Template primers consisted of a U5 duplex DNA segment attached to a 2-nucleotide 5'-tail such as would exist in a DNA end that had undergone 3'-processing by integrase (F i g . 3 A ). All 16 2-nucleotide combinations were tested as substrates for the integrase DNA polymerase. Repair reactions were conducted in the presence of all 4 dNTPs and one radiolabeled dNTP. The latter was complementary to the first base next to the primer terminus and varied according to the sequence of the template - primer. Thus, we could ensure that if only the first base was utilized as a template by the integrase DNA polymerase, which was the case in many instances (see below), a radiolabeled product would still be produced. The results indicated that repair is controlled by the sequence of the DNA template. The frequency of polymerization observed for 13 out of the 16 template sequence combinations examined was increased up to 4-fold relative to the wild type sequence 5'-AC. (F i g . 4 , T a b l e 1 ). Analysis of the various reaction products by singlenucleotide resolution PAGE revealed that in many cases the integrase DNA polymerase appeared to add only a single nucleotide to the 2-nucleotide template (F i g . 4 ). In some cases the second nucleotide was added as well, but in these cases the enzyme paused frequently after the first nucleotide, yielding a doublet in the PAGE analysis (F i g . 4, lanes 8, 10, 12, 14, 15, 16). By contrast to this behavior of the integrase DNA polymerase, the Klenow fragment of E. coli DNA polymerase I and the HIV-1 reverse transcriptase copied both nucleotides almost quantitatively regardless of sequence (see below). It seemed unlikely that these results were due in part to the action of the integrase mediated 3'-processing activity since we have also observed this behavior of the integrase DNA polymerase with DNA substrates lacking the conserved

B. Effect of mutations in the conserved 5'TG/CA dinucleotide on the repair of 3'processed U5 LTRs All template primers used in this series of experiments had a 5'-AC unpaired tail and differed in the sequence of the adjacent conserved 5' -TG/CA dinucleotide (F i g . 3B). Repair by integrase DNA polymerase was conducted in the presence of all four dNTPs and radiolabeled dGTP. Thus, the first incorporated nucleotide would be radiolabeled in all DNA products. Mutations in the wild type TG/CA conserved dinucleotide had profound effects on the polymerization frequency occurring on the same 5'- AC unpaired tail (F i g . 5 , T a b l e 2 ). The wild type sequence 5'-ACTG exhibited a polymerization frequency that was between 3 and 100-fold lower than that obtained with 14 other sequences. The greatest polymerization frequency was obtained with the template-primer bearing the sequence, 5'ACAA (F i g . 5 , lane 9). Only one mutation, 5'-ACGC, reduced the polymerization frequency relative to the wild type sequence ACTG (F i g . 5 , lanes 1 and 7). Thus, the overwhelming overall effect of mutations in the conserved

314

Gene Therapy and Molecular Biology Vol 3, page 315 Fig. 4. Effect of template sequence on end repair by the wild type integrase DNA p o l y m e r a s e . Reaction mixtures (described in Materials and Methods) contained 4dNTPs and either [!-32P] dCTP (lanes 1-4), [!-32P] dGTP (lanes 5-8), [!-32P]TTP (lanes 9-12) or [!32P] dATP (lanes 13-16). The template sequence of each 2nucleotide template-primer is given at the top of each lane. The numbers to the right of the Fig. refer respectively to the incorporation of either 1 (+1) or 2 (+2) nucleotides into the DNA substrate.

Fig. 5 Effect of changes in the conserved dinucleotide on end repair by the wild type integrase DNA p o l y m e r a s e . The reaction mixtures as described in Materials and Methods contained all 4dNTPs and [!32 P] dGTP as the only radiolabeled nucleotide. The boxed 5' -AC at the top left of each section of the Fig. represents the 2-nucleotide template sequence in each of the 16 different templateprimers. The sequences of the dinucleotides adjacent to the 5'AC tail are given at the top of each lane of the gel. The complementary nucleotides to each of these sequences (not shown) were also included in the oligonucleotide templateprimers. The number (+1) to the right of the Fig. refers to the incorporation of 1 nucleotide into the 32 -nucleotide template-primers.

315

Gene Therapy and Molecular Biology Vol 3, page 316

Table 1. Relative frequency of polymerization by integrase DNA polymerase using 2-nucleotide NNTG templatesa,b Template sequence 5'"3'

AG

TG

GG

CG

AC

TC

GC

CC

AA

TA

GA

CA

AT

TT

GT

CT

Polymerization frequency c

0.5

2.7

3.4

1.0

1.0

2.8

1.9

1.2

1.6

4.2

0.1

2.2

3.8

2.3

3.0

1.0

Polymerization frequency d

0.2

1.0

2.2

0.3

1.0

3.1

1.4

2.8

0.9

4.0

0.7

1.7

2.4

1.6

0.9

0.6

__________________________________________________________________________________________ a Phosphorimager adjusted volumes normalized to 1.0; bDNA concentration - 0.02 Âľ g/ml; c Wild type; dCore domain.

Table 2. Relative frequency of polymerization by integrase DNA polymerase using 2-nucleotide ACN'N' templates a,b Template sequence 5'"3'

AC AG

AC TG

AC GG

AC CG

AC AC

AC TC

AC GC

AC CC

AC AA

Polymerization frequency c

5.5

1.0

6.7

9.0

25

9.7

0.2

10

Polymerization frequency d

1.2

1.0

1.4

1.4

2.6

1.6

.01

1.8

AC TA

AC GA

AC CA

AC AT

AC TT

AC GT

AC CT

100 12

8.7

9.5

18

12

3.3

17

5.9

1.0

0.1

2.0

1.0

1.1

1.9

1.0

__________________________________________________________________________________________ aPhosphorimager

adjusted volumes normalized to wild type = 1.0; b DNA concentration - 0.02 Âľ g/ml; cWild type; d Core

Domain

synthesis of relatively long 19-20 oligonucleotide DNA chains and the size distribution of the DNA products remained the same over an 8-fold range of enzyme concentration indicating that the reaction mechanism was processive (data not shown). Template pyrimidines therefore can be said to support DNA chain elongation largely by a processive mechanism. The final nucleotide of these homopolymeric templates was not utilized efficiently in either case, as reported previously (Acel et al., 1998), so that polymerization with relatively short DNA templates may be governed by different mechanisms as mentioned earlier. The results show clearly, that in contrast to the general inability to polymerize both nucleotides of a 2nucleotide template, the integrase DNA polymerase is capable of polymerizing relatively longer DNA chains when presented with the appropriate template-primer.

TG/CA dinucleotide was to enhance the polymerization frequency at the 5'-AC unpaired tail by as much as 100fold. Despite this significant increase in the polymerization frequency, only the first template nucleotide was consistently copied in every case.

C. The integrase DNA polymerase is processive with homopolymeric pyrimidine templates Next, the integrase DNA polymerase was incubated with a homopolymeric template-primer in which the primer stem consisted of a nonviral sequence and the template was oligo dC21 (F i g . 6 , lanes 3 and 4). The reaction products constituted a heterogeneous mixture of radiolabeled DNA molecules that formed a ladder after gel electrophoresis whose pattern was consistent with the addition to the primer terminus of between 1 and 21 deoxyguanylate residues. Although there was clear evidence of pausing at many positions, DNA products representing the addition of between 17 and 20 nucleotides to the DNA primer terminus predominated. When the template contained adenylate residues interspersed among oligo dC at 7 nucleotide intervals the integrase DNA polymerase appeared to pause less frequently before reaching the end of the template (F i g . 6 , lane 2). Integrase DNA polymerase activity with an oligo dT20 template, primed from an HIV1 U5 LTR hairpin primer, likewise supported the

D. Evidence for quasi-processive (distributive) chain elongation with a homopolymeric purine template When incubated with a homopolymeric oligo dA21 DNA template (F i g . 7, lane 2), the integrase DNA polymerase generally added between 1-7 nucleotides to this DNA as template - primer. Products consistent with the polymerization of up to 19 nucleotides, were not produced in significant amounts. There was a tendency to make longer DNA chains at intermediate DNA concentrations,

316

Gene Therapy and Molecular Biology Vol 3, page 317

F i g . 6 Behaviour o f t h e i n t e g r a s e DNA p o l y m e r a s e w i t h a h o m o p o l y m e r i c o l i g o dC t e m p l a t e . Reaction mixtures contained either dGTP alone or all 4 dNTPs together with [!-32 P] dGTP. The letters to the right and left of the Fig. represent the number of guanylate residues incorporated into DNA. G357CA and G357C21 oligonucleotides consist of the same nonviral hairpin stem of 30 nucleotides and 21 nucleotide templates as defined in Table 3.

F i g . 7 Effect o f t e m p l a t e p y r i m i d i n e s o n DNA c h a i n elongation by integrase D N A p o l y m e r a s e . The different template-primers used all were comprise of a common hairpin stem as described in Materials and Methods and a 21 nucleotide template sequence as designated at the top of each lane and defined in Table 3. A template consisting of oligodA 21 is listed in lane 2 and designated AAA. The numbers at the side of the Fig. refer to the number of nucleotides incorporated into DNA whereas the letters to the left and right of the Fig. indicate the identity of nucleotides incorporated into DNA in lanes 1 and 6, respectively.

of polymerization on a pyrimidine template (compare F i g . 6 , lanes 3 and 4 with F i g . 7 , lane 2). Hence, the processivity of the integrase DNA polymerase may be enhanced by pyrimidine tracts in the DNA template, a suggestion we explore further below.

albeit inefficiently, but this was abrogated at high DNA concentrations (not shown). The mechanism of chain elongation with an oligo dA template therefore appears to be either quasi-processive or distributive in nature. The longer DNA products obtained with an oligo dA21 template were comparatively in low abundance versus the products 317

Udashkin et al: HIV-1 integrase prevents the repair of 3'-processed LTRs

E. Utilization of mixed nucleotide sequence templates

F. Sequence specificity exhibited by a core domain integrase DNA polymerase

We explored the possibility that the processivity of the integrase DNA polymerase may be enhanced by pyrimidine tracts in the DNA template. When 3 dC residues were placed at 7-nucleotide intervals within an oligo dA18 template, processivity increased dramatically to a value of 20 (F i g . 7 , lane 1). Also, there appeared to be many pause sites with the strongest ones corresponding to the location of the dC residues in the template (F i g . 7 , lane 1). A parallel experiment in which Ts were used in place of dC yielded an almost identical effect on processivity while increasing the polymerization frequency several fold (F i g . 7, lane 6). Once again many pause sites were observed with the strongest ones located at A5 and A13, the position of these latter sites being apparently unrelated to the template T residues at positions 7 and 14 (F i g . 7 , lane 6). When the dinucleotides CT and GT were included in an oligodA15 template (Table 3) much the same results were obtained; the inclusion of these extra nucleotides enhanced the overall extent of DNA chain elongation (F i g . 7 , lanes 3 and 4). Finally, when 3 dG residues were included in the oligo dA 18 template (see Table 3), the extent of DNA chain elongation was relatively low and resembled the result obtained with an oligo dA21 template (F i g . 7 , lane 5). This result confirms the suspicion that template purines tend to contribute to low processivity by the integrase DNA polymerase. Nevertheless, the integrase DNA polymerase is clearly capable of chain elongation beyond one or two nucleotides, regardless of template sequence, as seen with 2-nucleotide template-primers.

Next we determined whether the amino or carboxyl terminus of integrase might play a role in the mechanism of sequence recognition at 2-nucleotide DNA templateprimers. We isolated a core domain segment of integrase that included the first 3 N-terminal amino acids, linked to amino acids 51-190 and termed IN1-3X451-190 . This truncated form of integrase displayed virtually the same specific activity for DNA polymerase as the wild type enzyme (data not shown). The core domain deletions had a profound effect on the relative polymerization frequencies observed for templateprimers with mutations in the conserved dinucleotide 5'TG/CA (F i g . 8 , T a b l e 2). The main effect was an overall tendency to reduce the variation in polymerization frequency among the various mutant sequences. Thus, while the wild type enzyme exhibited a relatively large variation in polymerization frequency, depending on the sequence of the dinucleotide adjacent to the 5'AC template, the core domain exhibited only a 2-5-fold variation in this regard. There were two exceptions to this, namely the 5'ACGC sequence which was not utilized at all by both enzymes and the 5' - ACCA sequence which exhibited a 50-fold reduction in polymerization frequency in the case of the core domain (F i g . 8 , lanes 10 and 16, Table 2). An amino terminal deletion construct IN1-3X4 51-288 missing specifically the zinc finger domain (amino acids 450) gave identical results to that obtained with the core domain indicating that the amino terminus is responsible for maintaining the large differences in polymerization frequency observed with the wild type enzyme.

G. Effect of core domain deletions on DNA chain elongation The polymerization frequency of the core domain was not affected differently from wild type integrase in respect to changes in the 5’-AC template sequence (F i g . 9, Table 1). The core domain deletions seemed to enhance the ability of the integrase DNA polymerase to copy both nucleotides of various 2-nucleotide template-primers (F i g . 9, lanes 3, 6 and 7). This capability of the core domain was further enhanced at higher enzyme concentrations but even under these conditions wherein the wild type enzyme also was more efficient in copying both nucleotides (data not shown) the core domain (and the wild type enzyme) paused frequently after inserting one nucleotide in the various 2-nucleotide template-primers. It is noteworthy that even at relatively high enzyme concentrations, the wild type template sequence 5'-AC was relatively inefficiently repaired either by the core domain or the wild type DNA polymerase (F i g . 9 , lane 5). By contrast, the HIV-1 reverse transcriptase readily repaired both nucleotides of all the 2-nucleotide template-primers examined even when the enzyme was present at a relatively low concentration (compare F i g u r e s 4 and 10). The results add to the evidence for an inherent regulatory

Table 3. List of oligonucleotide template sequences used in this study to examine DNA chain elongation by the integrase DNA polymerase a

___________________________________________

G357C21........cccccccccccccccccccc - 5' G357CA..........cccccaccccccaccccccac - 5' CAA...............aaaaaacaaaaaacaaaaaac - 5' AAA...............aaaaaaaaaaaaaaaaaaaaa - 5' CGA...............aaaaagcaaaaagcaaaaagc - 5' CTA................aaaaatcaaaaatcaaaaatc - 5' GAA...............aaaaaagaaaaaagaaaaaag - 5' TAA................aaaaaataaaaaataaaaaat - 5' ___________________________________________ a the sequences listed represent extensions at the 5' - end of a common hairpin stem whose sequence is given in Materials and Methods.

318

Gene Therapy and Molecular Biology Vol 3, page 319

incorporation of 1 nucleotide into the 32 -nucleotide template-primer. Fig. 8. Effect of changes in the conserved 5'-TG/CA dinucleotide on end repair by the integrase core domain. All reaction mixtures contained all 4 dNTPs and [!-32P] dGTP as the only radiolabeled nucleotide. The boxed 5' -AC at the top left of each section of the Fig. represents the 5'-AC 2nucleotide template sequence in each of the 16 different template-primers. The sequences of the dinucleotides adjacent to the 5'-AC tail are given at the top of each lane of the gel. The complementary nucleotides to each of these sequences (not shown) were also included in the oligonucleotide templateprimers. The number (+1) to the right of the Fig. refers to the

Fig. 9 Effect of template sequence on end repair by the integrase core d o m a i n . All reaction mixtures contained all 4 dNTPs and one [!-32P] dNTP as described in Materials and Methods and as indicated in Fig. 4. The template sequence of each 2nucleotide template-primer is given at the top of each lane. The numbers to the right of the Fig. refer respectively to the incorporation of either 1 (+1) or 2 (+2) nucleotides into the DNA substrate.

319

Gene Therapy and Molecular Biology Vol 3, page 320

F i g . 1 0 E f f e c t o f t e m p l a t e s e q u e n c e o n e n d r e p a i r b y H I V - 1 r e v e r s e t r a n s c r i p t a s e . All reaction mixtures contained all 4 dNTPs and one [!-32P] dNTP as described in Materials and Methods and as indicated in Fig. 4. The template sequence of each 2-nucleotide template-primer is given at the top of each lane. The numbers to the right of the Fig. refer respectively to the incorporation of either 1 (+1) or 2 (+2) nucleotides into the DNA substrate.

selectively poor ability to copy stretches of template purines (as compared to pyrimidines). The significance of the preference of the integrase DNA polymerase for template pyrimidines and their ability to increase the processivity of the enzyme is not clear at the present time. The influence of the conserved dinucleotide 5'-TG/CA at the end of the HIV-1 LTR on the frequency of polymerization of the unpaired 5'AC tail is most significant. The dinucleotide sequence has a profound negative regulatory effect on the integrase DNA polymerase since mutations in either of these two nucleotides of the HIV-1 U5 LTR yields a much greater polymerization frequency, albeit only with respect to the insertion of the first nucleotide. This negative regulatory mechanism appears to be ideal as a way of further preventing the repair of 3'-processed ends of the HIV-1 LTRs while permitting the integrase DNA polymerase to repair gaps fully at a later stage in the integration process. The conserved 5' - TG/CA dinucleotide plays an important role in the 3' - processing reaction as well; mutations of this sequence have the effect of reducing the efficiency of endonucleolytic cleavage substantially. Taken together with our results this effect would be consistent with a

mechanism intrinsic to the structure of the HIV-1 integrase.

III. Discussion The relatively poor ability of the HIV-1 integrase DNA polymerase to repair the ends of DNA molecules is one of the most striking observations presented in this study. This is in stark contrast to the behavior displayed by this enzyme with gapped DNA substrates, reported previously, where gaps of 1,2,5 or 7 nucleotides were efficiently and completely repaired, regardless of sequence (Acel et al., 1998). As far as we know, this differentiation of function is unique among DNA polymerases and would be consistent with a role for the integrase DNA polymerase in the repair of 5-nucleotide gaps that arise during the integration of HIV-1 DNA into cellular chromosomes. The integrase DNA polymerase therefore is likely designed to function exclusively as a repair enzyme. The integrase DNA polymerase clearly would be incapable of acting as a replicative enzyme based on the results presented here because of a strong tendency to pause frequently during DNA chain elongation and especially because of its 320

Gene Therapy and Molecular Biology Vol 3, page 321 model in which mutations in the LTR decrease 3' processing by the integrase endonuclease while at the same time, increase the frequency of repair by the integrase DNA polymerase. The combination of these effects would tend to maintain the integrity of the ends of the viral DNA and perhaps decrease the probability of integrating a defective viral genome. We have also reported here for the first time the effect of core domain deletions on the integrase DNA polymerase activity. The core domain retains all of the determinants required for the DNA polymerase activity of integrase including some of the regulatory determinants for DNA chain elongation. Some functions may however be lost such as the apparent negative regulatory effects of the conserved 5' TG/CA dinucleotide on polymerization frequency. Either the amino terminal zinc finger domain, or the C-terminal DNA binding domain may therefore play a regulatory role in this aspect of DNA polymerase function. In terms of mechanism then, what can explain the inability of the integrase DNA polymerase to repair short templates at the ends of DNA molecules? How is chain elongation abrogated and what determines the sequence specificity with regard to polymerization frequency? The answers to these questions have not been obtained in the present study. Possibilities include a high Km for certain template-primers leading to a low initial polymerization rate or in the absence of differences in the affinity of the enzyme for the DNA template, the results could be explained in terms of a very slow off rate after incorporation of the penultimate nucleotide. Further detailed kinetic experiments are necessary to distinguish among these possibilities. The regulation of end-repair by the integrase DNA polymerase emphasizes the importance of 3' - processing for the integration process and may have significance for the use of retroviruses as vectors or for the virus-free delivery of foreign genes into mammalian cells and into humans. The results presented here serve to emphasize the importance of preventing mutations in the LTRs in order to maintain a high efficiency of integration. It is possible to envision the design of artificial gene delivery systems which incorporate some features of the natural process of retroviral integration without the use of live viruses. Such systems might include the retroviral LTRs linked to a foreign therapeutic gene of interest in combination with an integrase expression system. Rudimentary designs of systems in which integrase, fused to another retroviral gene product, acts in trans to influence the properties of cotransfected LTR plasmids or mutant viral genomes missing the integrase gene, have been described (Faust et al., 1995; Liu et al., 1997; Wu et al., 1997). Integrase is the newest and perhaps most promising target for antiviral chemotherapy (Pommier et al., 1997). Although many compounds have been shown to inhibit integrase function in vitro it is not clear that a good in vivo antiviral agent targeting integrase has been developed as yet (Burke et al., 1995; Cherepanov et al., 1997; King

and Robinson, 1998; Mazumder et al., 1995; Neamati et al., 1997; Neamati et al., 1998; Robinson et al., 1996). Our study of end repair suggests a novel strategy for interfering with the integration process. Clearly, a mechanism exists preventing the integrase DNA polymerase from filling in the 3'-processed ends of the HIV-1 LTR. If the determinants on integrase that regulate this property of the enzyme could be clearly defined, then it might be possible to design drugs capable of 'flipping' the enzyme into a mode in which the regulation is lost, while maintaining the DNA polymerase activity intact. Under these circumstances the efficient repair of the 3'processed ends could be effected with lethal consequences to the virus due to a failure to complete the integration process.

IV. Materials and Methods A. Purification of the integrase DNA polymerase The expression plasmid pQE30#IN (Faust et al., 1995; Faust et al., 1996) was propagated in the Kan r E. coli strain M15 pREP(Qiagen) at 37 oC and integrase expression was induced by adding IPTG as described (Faust et al., 1995; Faust et al., 1996). All further steps of purification were carried out at room temperature unless otherwise specified. Frozen bacterial pellets derived from 1.2 L of bacterial culture were thawed and resuspended in 100 ml of B-Per solution (Pierce Chemical Co.) with the aid of a glass Dounce homogenizer. The suspension was centrifuged in a Beckman J2-21 centrifuge at 27,000g for 15 minutes using a Beckman JA-14 rotor. Pellets were resuspended in 100 ml B-Per solution by Dounce homogenization, lysozyme (Sigma) was added (200 micrograms/ml) and inclusion bodies were collected by centrifugation. The pellet was washed once more using a 1:10 dilution of the B-Per solution, left overnight at room temperature, and resuspended finally in 60 ml of buffer A ( 6M guanidine HCl, 0.1M Na-phosphate, 0.01M Tris-HCl pH 8.0) . Qiagen nickel nitrilotriacetate (Ni ++NTA) resin (6 ml of a 50% slurry) equilibrated in buffer A, was added to the protein solution and stirred for 1 hour at room temperature. The mixture was poured into a polypropylene column and the Ni++NTA beads allowed to pack under gravity. The resin bed was then washed with 120 ml of buffer A followed by 60 ml of 8M urea , 0.1M sodium phosphate, 0.01M Tris-HCl pH 8.0 and 18 ml each of the latter solution adjusted to pH 6.3 and pH 5.9. Integrase was eluted from the Ni++ NTA resin in 21 ml of 8M urea, 0.1M sodium phosphate, 0.01M Tris-HCl pH 4.5 and renatured at 4 o C by step-wise dialysis over a period of 3 days. Dialysis was done sequentially against 4M and 2M urea in 50mM Hepes-HCl pH 7.5, 1.0M NaCl and 1mM DTT. The sample was then dialyzed against final dialysis buffer (FDB) (50mM HEPES-HCl pH 7.5, 1.0M NaCl, 1mM DTT, 10% glycerol, 1mM CHAPS and 0.1mM EDTA) and then against 50mM imidazole, 1M NaCl, 50 mM HEPES-HCl pH 7.5, 10 mM $- mercaptoethanol, 1mM CHAPS and 10% glycerol. Dialyzed samples were applied to a 6 ml column of Ni++ NTA beads equilibrated in the same buffer. The flow through was collected and the column washed with an additional 3.5 ml of the equilibrating buffer. The wash was combined with the flow through fraction and the combined sample was concentrated to

321

Udashkin et al: HIV-1 integrase prevents the repair of 3'-processed LTRs a final volume of 0.5-1.0 ml using an Amicon Centricon Plus 80 filtration unit (80 ml capacity). The concentrated sample was applied immediately to a 92ml S-300 column (Pharmacia) 0.5x90cm that had been equilibrated in FDB. The column was developed at a rate of 5.5 ml/h in FDB, fractions of 1 ml were collected and the DNA polymerase activity was located using the trichloroacetic acid precipitation method as described previously (Acel et al., 1998). Fractions containing integrase DNA polymerase activity were pooled and stored at 4o C.

bovine serum albumin, DISPOL 17 DNA (0.25mg/ml) 5'T 20 ACTGCTAGAGATTTTAAAATCTCTAGCAGT 3' and 1µM dATP with 4µCi [!-32P] dATP (Mandel Scientific Co) in a total volume of 25 µl. Integrase (1 unit =1 pmol dAMP incorporated into DNA) was added in 1µl FDB. Reaction mixtures were incubated at 37o C for 1h. For 2-nucleotide repair reactions, synthetic oligonucleotides with a 5'-AC template sequence and a 15-base pair hairpin primer stem matching the U5 LTR were used and reactions were conducted in the presence of all four dNTPs under standard conditions. The oligonucleotide with a wild type U5 sequence was:

B. Zinc finger N-terminal deletion The amino terminal deletion mutant pQE 30 #%&#Z consisted of the following format in respect to integrase IN 13 X4 51-288 where X refers to ValValArgLeu amino acids derived from the insertion of a linker(see below) and the numbers refer to the position of wild type amino acids in the integrase protein. Thus, this construct has a deletion of integrase amino acids 4-50 inclusive. It was derived by isolating the large fragment from a partial Xba I digest of pQE30# IN, cutting the isolated fragment with NsiI and isolating the large fragment once again. The latter consists of pQE30# IN missing the XbaI/NsiI region. The deleted plasmid was recircularized in the presence of a synthetic oligonucleotide linker 5'- CTA GAC GTA GTC CGT CTG CA -3' hybridized with 3'- TG CAT CAG GCA G - 5' to produce the deleted plasmid pQE 30 # %&#'.

5'-ACTGCTAGAGATTTTCCGGAAAATCTCTAGCA-3'. Oligonucleotides used for end-repair reactions varied either in the sequence of first two nucleotides (5'-AC) or in the 4 nucleotides comprising the two adjacent base pairs 5'-TG/CA; in the former series of reactions the [!-32P] dNTP added (1µCi) was complementary to the penultimate 5' nucleotide (the first nucleotide adjacent to the primer stem) and in the latter series [!-32P] dGTP was added (1µCi) in all cases. Other oligonucleotides used in chain elongation studies consisted of a nonviral hairpin stem with the sequence 5'-GTAGCTCCGATCCGGTATATACCGGATCGGAGCTAC3' extended at the 5'-end by the template sequences listed in T a b l e 3 . DNA polymerase reactions conducted with these oligonucleotides contained all 4 dNTPs and 1µCi of either [!32 P] TTP or [!-32 P] dGTP depending on the sequence of the DNA template. Reaction products in this latter series of experiments were cleaved with MboI prior to PAGE analysis which was done at single-nucleotide resolution as described previously (Acel et al., 1998).

C. Carboxy-terminal deletion pCMV IN was cleaved with PstI and EcoRV. The linearized plasmid DNA was digested with exonuclease III and S1 nuclease using the Erase-a base kit (Promega) and circularized by ligation. The extent of deletion in the integrase gene was determined by DNA sequence analysis. A clone with a deletion end point at nucleotide 570 of the integrase coding region (includes amino acids 1-190) was cleaved with SmaI and recircularized with DNA ligase in the presence of HindIII linkers. The resulting plasmid was digested with HindIII and the small fragment was subcloned at the HindIII site of pQE30 # to produce pQE30 # %& 8.4.

Acknowledgements We gratefully acknowledge Michael Parniak for his generous gift of HIV-1 reverse transcriptase and for many helpful discussions. This work was supported by grants from the National Health and Research Development Project of Health and Welfare Canada, the Medical Research Council of Canada and the Canadian Foundation for AIDS Research.

D. Core domain construct The pQE 30# IN #Z plasmid was cleaved with BfrI and XhoI and the small fragment ligated to the large fragment derived from a BfrI/XhoI digest of pQE30 # IN 8.4. The resulting plasmid was comprised of the deleted N-terminus of the #Z plasmid and the deleted C-terminus of the 8.4 plasmid. Integrase was expressed from this construct as a core domain fusion protein that included the integrase amino acids IN1-3X451-190.

References Acel, A., Udashkin, B.E., Wainberg, M. and Faust, E.A. 1 9 9 8 Efficient gap repair catalyzed in vitro by an intrinsic DNA polymerase activity of human immunodeficiency virus type 1 integrase. J . V i r o l . 72, 2062-2071. Andrake, M.D. and Skalka, A.M. 1 9 9 5 Multimerization determinants reside in both the catalytic core and C terminus of avian sarcoma virus integrase. J B i o l C h e m 270(49), 29299-29306.

E. DNA polymerase reactions Unless stated otherwise standard DNA polymerase reactions were conducted using synthetic oligonucleotide template-primers with an oligodT20 template as described previously (Acel et al., 1998) and quantified by precipitation of nascent radiolabeled DNA in trichloroacetic acid followed by liquid scintillation counting. Reaction mixtures contained 10 mM Tris-HCl pH 7.5, 5mM MgCl 2 , 5mM DTT, 200µg/ml

Ansari-Lari, M.A., Donehower, L.A. and Gibbs, R.A. 1 9 9 5 Analysis of human immunodeficiency virus type 1 integrase mutants. V i r o l o g y 211(1), 332-335. Asante-Appiah, E. and Skalaka, A.M. 1 9 9 7 Molecular mechanisms in retrovirus DNA integration. A n t i v i r a l R e s . 36, 139-156.

322

Gene Therapy and Molecular Biology Vol 3, page 323 immunodeficiency virus type 1 integrase identified by using novel DNA substrates. J Virol 68, 3896-3907.

Balakrishnan, M. and Jonsson, C.B. 1 9 9 7 Functional identification of nucleotides conferring substrate specificity to retroviral integrase reactions. J . V i r o l . 71, 1025-1035.

Craigie, R. 1 9 9 2 Trends Genet. 8, 187 -190. Donzella, G.A., Leon, O. and Roth, M.J. 1 9 9 8 Implication of a central cysteine residue and the HHCC domain of moloney murine leukemia virus integrase protein in functional multimerization. J . V i r o l . 72, 1691-1698.

Brown, P.O. (1990) Integration of retroviral DNA, pp. 19-48. Vol. 157. C. Springer-Verlag, Berlin-Heidelberg. Burke, T.R., Jr., Fesen, M.R., Mazumder, A., Wang, J., Carothers, A.M., Grunberger, D., Driscoll, J., Kohn, K. and Pommier, Y. 1 9 9 5 Hydroxylated aromatic inhibitors of HIV-1 integrase. J Med Chem 38(21), 4171-4178.

Drelich, M., Haenggi, M. and Mous, J. 1 9 9 3 Conserved residues Pro-109 and Asp-116 are required for interaction of the human immunodeficiency virus type 1 integrase protein with its viral DNA substrate. J . V i r o l . 67, 50415044.

Bushman, F., Engelman, A., Palmer, I., Wingfield, P. and Craigie, R. 1 9 9 3 Domains of the integrase protein of human immunodeficiency virus type 1 responsible for polynucleotidyl transfer and zinc binding. P r o c . N a t l . Acad. Sci. USA 90, 3428-3432.

Drelich, M., Wilhelm, R. and Mous, J. 1 9 9 2 Identification of amino acid residues critical for endonuclease and integration activities of HIV-1 IN protein in vitro. V i r o l o g y 188, 459-468.

Bushman, F.D. and Craigie, R. 1 9 9 1 Activities of human immunodeficiency virus (HIV) integration protein in vitro: specific cleavage and integration of HIV DNA. P r o c . N a t l . A c a d . S c i . U S A 88, 1339-1343.

Dyda, F., Hickman, A.B., Jenkins, T.M., Engelman, A., Craigie, R. and Davies, D.R. 1 9 9 4 Crystal structure of the catalytic domain of HIV-1 integrase: Similarity to other polynucleotidyl transferases. S c i e n c e 266, 19811986.

Bushman, F.D., Fujiwara, T. and Craigie, R. 1 9 9 0 Retroviral DNA integration directed by HIV integration protein in vitro. S c i e n c e 249, 1555-1558.

Ellison, V., Abrams, H., Roe, T., Lifson, J. and Brown, P. 1 9 9 0 Human immunodeficiency virus integration in a cell-free system. J . V i r o l . 64, 2711-2715.

Cai, M., Zheng, R., Caffrey, M., craigie, R., Clore, G.M. and Gronenborn, A.M. 1 9 9 7 Solution structure of the Nterminal zinc binding domain of HIV-1 integrase. Nature s t r u c t u r a l b i o l o g y 4, 567-577.

Engelman, A., Bushman, F.D. and Craigie, R. 1 9 9 3 Identification of discrete functional domains of HIV-1 integrase and their organization within an active multimeric complex. EMBO J. 12, 3269-3275.

Cannon, P.M., Wilson, W., Byles, E., Kingsman, S.M. and Kingsman, A.J. 1 9 9 4 Human immunodeficiency virus type 1 integrase: Effect on viral replication of mutations at highly conserved residues. J . V i r o l . 68, 4768-4775.

Engelman, A. and Craigie, R. 1 9 9 2 Identification of conserved amino acid residues critical for human immunodeficiency virus type 1 integrase function in vitro. J . V i r o l . 66, 6361-6369.

Cara, A., Guarnaccia, F., Reitz, M.S., Jr., Gallo, R.C. and Lori, F. 1 9 9 5 Self-limiting, cell type-dependent replication of an integrase-defective human immunodeficiency virus type 1 in human primary macrophages but not T lymphocytes. V i r o l o g y 208, 242-248.

Engelman, A., Englund, G., Orenstein, J.M., Martin, M.A. and Craigie, R. 1 9 9 5 Multiple effects of mutations in human immunodeficiency virus type 1 integrase on viral replication. J Virol 69, 2729-2736.

Carteau, S., Batson, S.C., Poljak, L., Mouscadet, J.-F., de Rocquigny, H., Darlix, J.-L., Roques, B.P., Kas, E. and Auclair, C. 1 9 9 7 Human Immunodeficiency Virus Type 1 nucleocapsid protein specifically stimulates Mg2+dependent DNA integration in vitro. J . V i r o l . 71, 62256229.

Engelman, A., Mizuuchi, K. and Craigie, R. 1 9 9 1 HIV-1 DNA integration: Mechanism of viral DNA cleavage and DNA strand transfer. C e l l 61, 1211-1221. Englund, G., Theodore, T.S., Freed, E.O., Engelman, A. and Martin, M.A. 1 9 9 5 Integration is required for productive infection of monocyte-derived macrophages by human immunodeficiency virus type 1. J . V i r o l . 69, 32163219.

Carteau, S., Mouscadet, J.F., Goulaouiac, H., Subra, F. and Auclair, C. 1 9 9 3 Quantitative assay for human immunodeficiency virus deoxyribonucleic acid integration. Arch. B i o c h e m . B i o p h y s . 300, 756760.

Faust, E.A., Acel, A., Udashkin, B. and Wainberg, M.A. 1 9 9 5 Human immunodeficiency virus type 1 integrase stabilizes a linearized HIV-1 LTR plasmid in vivo. B i o c h e m M o l B i o l I n t 36, 745-758.

Cherepanov, P., Este, J.A., Rando, R.F., Ojwang, J.O., Reekmans, G., Steinfield, R., David, G., de ClercQ, E. and Debyser, Z. 1 9 9 7 Mode of interaction of G-quartets with the integrase of human immunodeficiency virus type 1. M o l . P h a r m a c o l o g y 52, 771-780.

Faust, E.A., Garg, A., Small, L., Acel, A., Wald, R. and Udashkin, B. 1 9 9 6 Enzymatic capability of HIS-tagged HIV-1 integrase using oligonucleotide disintegration substrates. J . B i o m e d . S c i . 3, 254-265.

Chow, S.A. and Brown, P.O. 1 9 9 4 a Juxtaposition of two viral DNA ends in a bimolecular disintegration reaction mediated by multimers of human immunodeficiency virus type 1 or murine leukemia virus integrase. J V i r o l 68, 7869-7878.

Goldgur, Y., Dyda, F., Hickman, A.B., Jenkins, T.M., Craigie, R. and Davies, D.R. 1 9 9 8 Three new structures of the core domain of HIV-1 integrase: An active site that binds magnesium. Proc Natl Acad Sci USA 95, 91509154.

Chow, S.A. and Brown, P.O. 1 9 9 4 b Substrate features important for recognition and catalysis by human

Grandgenett, D.P. and Mumm, S.R. 1 9 9 0 Unravelling retrovirus integration. C e l l 60, 3-4.

323

Udashkin et al: HIV-1 integrase prevents the repair of 3'-processed LTRs Kalpana, G.V. and Goff, S.P. 1 9 9 3 Genetic analysis of homomeric interactions of human immunodeficiency virus type 1 integrase using the yeast two-hybrid system. Proc Natl Acad Sci USA 90, 10593-10597.

Mazumder, A., Gazit, A., Levitzki, A., Nicklaus, M., Yung, J., Kohlhagen, G. and Pommier, Y. 1 9 9 5 Effects of tyrphostins, protein kinase inhibitors, on human immunodeficiency virus type 1 integrase. B i o c h e m i s t r y 34, 15111-15122.

Katzman, M., Katz, R.A., Skalka, A.M. and Leis, J. 1 9 8 9 The avian retroviral integration protein cleaves the terminal sequences of linear viral DNA at the in vivo sites of integration. J . V i r o l . 63, 5319-5327.

Miller, M.D., Farnet, C.M. and Bushman, F.D. 1 9 9 7 Human Immunodeficiency Virus Type 1 preintegration complexes: Studies of organization and composition. J . V i r o l . 71, 5382-5390.

Khan, E., Mack, J.P., Katz, R.A., Kulkosky, J. and Skalka, A.M. 1 9 9 1 Retroviral integrase domains: DNA binding and the recognition of LTR sequences. N u c l e i c A c i d s R e s . 19, 851-860.

Neamati, N., Hong, H., Sunder, S., Milne, G.W.A. and Pommier, Y. 1 9 9 7 Potent inhibitors of human immunodeficiency virus type 1 integrase: Identification of a novel four-point pharmacophore and tetracyclines as novel inhibitors. Mol Pharmacol 52, 1041-1055.

King, P.J. and Robinson, W.E., Jr. 1 9 9 8 Resistance to the anti-human immunodeficiency virus type 1 compound Lchicoric acid results from a single mutation at amino acid 140 of integrase [In Process Citation]. J Virol 72, 84208424.

Neamati, N., Mazumder, A., Sunder, S., Owen, J.M., Tandon, M., Lown, J.W. and Pommier, Y. 1 9 9 8 Highly potent synthetic polyamides, bisdistamycins, and lexitropsins as inhibitors of human immunodeficiency virus type 1 integrase. Mol Pharmacol 54, 280-90.

Kulkosky, J., Jones, K.S., Katz, R., Mack, J. and Skalka, A.M. 1 9 9 2 Residues critical for retrovoral integrative recombination in a region that is highly conserved among retroviral/retrotransposon integrases and bacterial insertion sequence transposases. M o l . C e l l B i o l . 12, 2331-2338.

Pommier, Y., Pilon, A.A., Bajaj, K., Mazumder, A. and Neamati, N. 1 9 9 7 HIV-1 integrase as a target for antiviral drugs. Antiviral Chem Chemother 8, 463-483. Robinson, J.W.E., Cordeiro, M., Abdel-Malek, S., Jia, Q., Chow, S.A., Reinecke, M.G. and Mitchell, W.M. 1 9 9 6 Dicaffeoylquinic acid inhibitors of human immunodeficiency virus integrase: Inhibition of the core catalytic domain of human immunodeficiency virus integrase. Mol Pharmacol 50, 846-855.

LaFemina, R.L., Callahan, P.L. and Cordingley, M.G. 1 9 9 1 Substrate specificity of recombinant human immunodeficiency virus integrase protein. J . V i r o l . 65, 5624-5630. Lafemina, R.L., Schneider, C.L., Robbins, H.L., Callahan, P.L., LeGrow, K., Roth, E., Schleif, W.A. and Emini, E.A. 1 9 9 2 Requirement of active human immunodeficiency virus type 1 integrase enzyme for productive infection of human T-lymphoid cells. J . V i r o l . 66, 7414-7419.

Roe, T., Chow, S.A. and Brown, P.O. 1 9 9 7 3'- End processing and kinetics of 5'-end joining during retroviral integration in vivo. J . V i r o l . 71, 1334-1340. Sakai, H., Kawamura, M., Sakuragi, J.-I., Sakuragi, R., shibata, R., Ishimoto, A., Ono, N., Ueda, S. and Adachi, A. 1 9 9 3 Integration is essential for efficient gene expression of human immunodeficiency virus type I. J . V i r o l . 67, 1169-1174.

Leavitt, A.D., Rose, R.B. and Varmus, H.E. 1 9 9 2 Both substrate and target oligonucleotide sequences affect in vitro integration mediated by human immunodeficiency virus type 1 integrase protein produced in Saccharomyces cerevisiae. J . V i r o l . 66, 2359-2368.

Sherman, P.A. and Fyfe, J.A. 1 9 9 0 Human immunodeficiency virus integration protein expressed in E. coli possesses selective DNA cleaving activity. P r o c . N a t l . Acad. S c i . U S A 87, 5119-5123.

Leavitt, A.D., Shuie, L. and Varmus, H.E. 1 9 9 3 Site-directed mutagenesis of HIV-1 integrase demonstrates differential effects on integrase functions in vitro. J . B i o l . C h e m . 268, 2113-2119.

Shin, C.G., Taddeo, B., Haseltine, W.A. and Farnet, C.M. 1 9 9 4 Genetic analysis of the human immunodeficiency virus type 1 integrase protein. J Virol 68, 1633-1642.

Lee, S.P., Xiao, J., Knutson, J.R., Lewis, M.S. and Han, M.K. 1 9 9 7 Zn++ promotes the self-association of human immunodeficiency virus type 1 integrase in vitro. B i o c h e m i s t r y 36, 173-180.

Skalka, A.M. 1 9 9 3 Retroviral DNA integration: lessons for transposon shuffling. Gene 135, 175-182.

Liu, H., Wu, X., Xiao, H., Conway, J.A. and Kappes, J.C. 1997 Incorporation of functional Human Immunodeficiency Virus Type 1 integrase into virions independent of the gag-pol precursor protein. J . V i r o l . 71, 7704-7710.

Taddeo, B., Haseltine, W.A. and Farnet, C.M. 1 9 9 4 Integrase mutants of human immunodeficiency virus type 1 with a specific defect in integration. J Virol 68, 8401-8405. Tramontano, E., LaColla, P. and Cheng, Y.-C. 1 9 9 8 Biochemical characterization of the HIV-1 integrase 3'processing activity and its inhibition by phosphorothioate oligonucleotides. B i o c h e m i s t r y 37, 7237-7243.

Maignan, S., Guilloteau, J.P., Zhou-Liu, Q., Clement-Mella, C. and Mikol, V. Crystal structures of the catalytic domain of HIV-1 integrase free and complexed with its metal cofactor: high level of similarity of the active site with other viral integrases. J M o l B i o l 282, 359-368.

Van Gent, D.C., Groeneger, A.A.M.O. and Plasterk, R.H.A. 1 9 9 2 Mutational analysis of the integrase protein of human immunodeficiency virus type 2. P r o c . N a t l . Acad. Sci. USA 89, 9598-9602.

Mazumder, A., Engelman, A., Craigie, R., Fesen, M. and Pommier, Y. 1 9 9 4 Intermolecular disintegration and intramolecular strand transfer activities of wild-type and mutant HIV-1 integrase. N u c l e i c A c i d s R e s 22, 10371043.

van Gent, D.C., Vink, C., Groeneger, A.M.O. and Plasterk, R.H.A. 1 9 9 3 Complementation between HIV integrase

324

Gene Therapy and Molecular Biology Vol 3, page 325 proteins mutated in different domains. EMBO J . 12, 3261-3267. Vincent, K.A., Ellison, V., Chow, S.A. and Brown, P.O. 1 9 9 3 Characterization of human immunodeficiency virus type 1 integrase expressed in Escherichia coli and analysis of variants with amino terminal mutations. J . V i r o l . 67, 425-437. Vink, C. 1 9 9 1 Site-specific hydrolysis and alcoholysis of human immunodeficiency viral DNA termini mediated by the viral integrase protein. N u c l e i c A c i d s R e s . 19, 6691-6698. Vink, C., Groeneger, A.A.M.O. and Plasterk, R.H.A. 1 9 9 3 Identification of the catalytic and DNA binding region of the human immunodeficiency virus type I integrase protein. N u c l e i c A c i d s R e s . 21, 1419-1425. Vink, C., Groenink, M., Elgersma, Y., Fouchier, R.A.M., Tersmette, M. and Plasterk, R.H.A. 1 9 9 0 Analysis of the junctions between human immunodeficiency virus type 1 proviral DNA and human DNA. J . V i r o l . 64, 56265627. Vink, C. and Plasterk, R.H.A. 1 9 9 3 The human immunodeficiency virus integrase protein. Trends Genet 9, 433-437. Vink, C., van Gent, D.C., Elgersma, Y. and Plasterk, R.H. 1 9 9 1 Human immunodeficiency virus integrase protein requires a subterminal position of its viral DNA recognition sequence for efficient cleavage. J . V i r o l . 65, 4636-4644. Wiskerchen, M. and Muesing, M.A. 1 9 9 5 a Human immunodeficiency virus type 1 integrase: Effects of mutations on viral ability to integrate, direct viral gene expression from unintegrated viral DNA templates, and sustain viral propagation in primary cells. J . V i r o l . 69, 376-386. Wiskerchen, M. and Muesing, M.A. 1 9 9 5 b Identification and characterization of a temperature-sensitive mutant of human immunodeficiency virus type 1 by alanine scanning mutagenesis of the integrase gene. J V i r o l 69, 597-601. Woerner, A.M., Klutch, M., Levin, J.G. and Marcus-Sekura, C.J. 1 9 9 2 Localization of DNA binding activity of HIV-1 integrase to the C-terminal half of the protein. AIDSRes-Hum-Retroviruses 8, 2433-2437. Woerner, A.M. and Marcus-Sekura, C.J. 1993 Characterization of a DNA binding domain in the Cterminus of HIV-1 integrase by deletion mutagenesis. N u c l e i c A c i d s R e s . 21, 3507-3511. Wu, X., Liu, H., Xiao, H., Conway, J.A., Hunter, E. and Kappes, J.C. 1 9 9 7 Functional RT and IN incorporated into HIV-1 particles independently of the gag/pol precursor protein. EMBO J 16, 5113-5122.

325

Gene Therapy and Molecular Biology Vol 3, page 327 Gene Ther Mol Biol Vol 3, 327-345. August 1999.

Molecular mechanisms of viral transcription and cellular deregulation associated with the HTLV-1 Tax protein Review Article

Brian A. Lenzmeier and Jennifer K. Nyborg* Department of Biochemistry and Molecular Biology, Colorado State University, Fort Collins, CO 80523-1870, USA _________________________________________________________________________________________________ *Corresponding Author: Tel: (970)-491-0420; Fax: (970)-491-0494; E-mail: jnyborg@vines.colostate.edu Abbreviations: HTLV-I, human T-cell leukemia virus type-I; ATL, adult T-cell leukemia; TSP/HAM, tropical spastic paraparesis/HTLV-1 associated myopathy; LTR, long terminal repeat; CREs, cyclic AMP response elements; CREB, cAMPresponse element binding protein; EMSA, electrophoretic mobility shift assays; bZIP, basic-leucine-zipper domain; bHLH , basic-helix-loop-helix domain; NF- B , nuclear factor kappa B; I- B , inhibitor kappa B; IKK, I-!B kinase; NIK , NF-!Binducing kinase. Received: 16 October 1998; accepted: 17 October 1998

Summary It is estimated that between 10 and 20 million people worldwide are infected with the human Tc e l l l e u k e m i a v i r u s , type I (HTLV-I). Since HTLV-I i s associated with a variety o f human diseases, including an aggressive lymphoproliferative disorder named adult T-cell leukemia, infection by HTLV-I has become increasingly recognized as an important public health concern. Malignant transformation associated with HTLV-I infection is linked with the synthesis o f a virally-encoded protein called Tax. In this review, we will highlight our current understanding of Tax protein function, both in its role as an activator of HTLV-I transcription and deregulator of cellular homeostasis. It is widely believed that Tax deregulation of cellular gene expression and cell cycle progression accounts for the pathogenicity associated with HTLV-I infection.

cell leukemia/lymphoma virus, suggesting a causal relationship between viral infection and the development of ATL (Popovic et al., 1982).

I. HTLV-1 discovery and associated diseases The human T-cell leukemia virus type-I (HTLV-I) was the first pathogenic human retrovirus isolated and characterized (Yoshida et al., 1982; Seiki et al., 1982; Chen et al., 1983). It was originally discovered in 1980 from two T-lymphoblast cell lines derived from a patient incorrectly diagnosed with cutaneous T-cell lymphoma (mycosis fungoides) (Poiesz et al., 1980). Both of these cell lines maintain the continuous release of mature and immature typical type C budding virus particles. Shortly after this original report, retroviral particles were also observed in a cell line obtained from a second patient who had been correctly diagnosed with adult T-cell leukemia (ATL) (Hinuma et al., 1981). Both of the original patients were eventually shown to be infected with the human T-

Since the publication of the first studies, HTLV-I has become widely accepted as the etiologic agent of ATL, a distinct disease entity (reviewed in Watanbe, 1997). ATL is characterized by clinical and hematological features that include unresponsiveness to radiation and chemotherapy, skin lesions due to infiltrating leukemic cells, lytic bone lesions, greater than 5% abnormal T-cells, and leukemic cells carrying the CD4+ phenotype (Poiesz et al., 1981; Robert-Guroff et al., 1982; Yoshida et al., 1984). Additional evidence for the causative role of HTLV-I in ATL comes from the observation that tumor cells from patients with ATL characteristically show monoclonal or oligoclonal integration of the HTLV provirus. A second disease, referred to as tropical spastic paraparesis/HTLV-1 327

Lenzmeier and Nyborg: Tax protein of HTLV-1 in adult T-cell leukemia associated myopathy (TSP/HAM), is also tightly linked with HTLV-1 infection (Gessain et al., 1985; Jacobson et al., 1988; Osame et al., 1986). TSP/HAM is a neurodegenerative disorder characterized by demylenation of the nerves of the spinal cord, resulting in paralysis of the lower extremities. TSP/HAM shares many similarities with multiple sclerosis, often confusing accurate diagnosis. The role HTLV-1 plays in the pathogenesis of TSP/HAM remains incompletely understood.

HTLV-I transcription. This strong activation by Tax leads to the synthesis of both the viral mRNAs that are translated into viral proteins and genomic RNA that is packaged into the virion. For these reasons, Tax transactivation plays an important role in the retroviral life cycle. Following the discovery of Tax, and its prominent role in the viral life cycle, researchers turned to an analysis of the sequences within the HTLV-I promoter, which confer responsiveness to Tax. Early studies found that nucleotide sequences within the U3 region of the HTLV-1 long terminal repeat (LTR) carried Tax-responsive transcriptional enhancer elements (Sodroski et al., 1984; Rosen et al., 1985; Fujisawa et al., 1985; Paskalis et al., 1986). The specific Tax-responsive elements in the HTLV1 promoter were originally identified by probing for deletions in the DNA, which resulted in abrogation of Tax transactivation. Sequence analyses quickly identified three imperfect direct repeats that were named the 21-base pair (bp) repeats (Seiki et al., 1982; Josephs et al., 1984; Rosen et al., 1985). Deletion mutagenesis of these elements identified them as the principal Tax-responsive sequences in the HTLV-1 promoter, with at least two of the three viral CREs required for efficient Tax function (Fujisawa et al., 1986; Paskalis et al., 1986; Shimotohno et al., 1986; Brady et al., 1987; Rosen et al., 1987; Jeang et al., 1988; Nakamura et al., 1989). In addition to the three 21-bp repeats, the HTLV-I promoter also carries a TATA element and several additional DNA elements which have been implicated in viral gene expression (Seiki et al., 1982; reviewed in Franklin and Nyborg, 1995).

The T-cell transforming properties of HTLV-1 are a major focus of research, as the virus provides an excellent model system for studying oncogenesis in humans. HTLV-1 is not an acute transforming retrovirus, nor does it rely on promoter insertion mechanisms for deregulation of cellular proto-oncogenes (Seiki et al., 1984). Only a small percentage of HTLV-1 infected individuals develop ATL, and disease onset follows a latency period of several decades (Kondo et al., 1987; Murphy et al., 1989; Kawano et al., 1985). The infrequency of ATL in infected individuals indicates that HTLV-1 is necessary, but not sufficient, for leukemic transformation. The development of ATL is hypothesized to occur as a consequence of a single oncogenic transformation event resulting from longterm autocrine T-cell proliferation (reviewed in Franchini, 1995) and/or virus-dependent disruption of the normal cellular processes outlined in this review. The regulatory protein Tax is thought to be the primary HTLV-1-encoded factor responsible for T-cell transformation (Grassmann et al., 1992; Grossman et al., 1995). Tax is required for HTLV-I replication, stimulating viral gene expression through enhancer elements in the HTLV-1 promoter (Chen et al., 1985; Cann et al., 1985; Felber et al., 1985; Sodroski et al., 1985). Tax is also very pleiotropic, as it deregulates host cell gene expression, influences apoptosis, and enhances cell-cycle progression. It is therefore not surprising that singularly, the HTLV-1 Tax protein has the ability to transform human T-lymphocytes in vitro and to promote tumorigenesis and leukomigenesis in vivo (Grassmann et al., 1992; Nerenberg et al., 1987; Grossman et al., 1995).

A. Characterization of cellular proteins that bind the Tax-responsive promoter elements Although the 21-bp repeats were identified as critical for Tax transactivation, DNase I footprinting studies suggested that Tax did not directly bind to the 21 bp repeats, but rather cellular proteins provided the primary recognition of these elements (Nyborg et al., 1988; Altman et al., 1988, Jeang et al., 1988). This led to efforts aimed at identifying the cellular proteins, as it seemed likely that these proteins would serve as mediators of Tax transactivation (Jeang et al., 1988; Giam and Xu, 1989; Nakamura et al., 1989; Tan et al., 1989; Nyborg and Dynan, 1990; Beimling and Moelling, 1990; Montagne et al., 1990; Poteat et al., 1990). A clue to their identity came shortly after the discovery of the ATF/CREB family of cellular transcription factors. Members of the ATF/CREB family are characterized by a DNA binding basic region that is located immediately adjacent to a leucine zipper dimerization domain (reviewed in Montminy, 1997).

II. The HTLV-1 promoter and Tax transactivation The Tax protein of HTLV-I is encoded downstream of the proviral structural genes, within a region originally termed X. Mutational analysis of the various open reading frames of HTLV-1, together with transient transfection assays using a Tax expression plasmid, confirmed that Tax was the virally encoded transactivator protein (Chen et al., 1983; Cann et al., 1985; Felber et al., 1985; Sodroski et al., 1985). The 353 amino acid Tax protein is synthesized from a doubly-spliced RNA transcript, and imported back into the nucleus where it functions as a potent activator of 328

Gene Therapy and Molecular Biology Vol 3, page 329

F i g u r e 1 . Nucleotide sequences of the three Tax-responsive viral CREs. The viral CREs are shown in blue, and a cellular CRE from the human chorionic gonadotropin gene promoter that is not responsive to Tax is shown in red. The CREB recognition element is indicated in b l a c k , and the conserved GC-rich viral CRE flanking sequences are underlined.

B. Tax enhancement of CREB binding to the viral CREs

These basic-leucine zipper (bZIP) proteins bind either as homodimers or heterodimers to specific palindromic DNA sequences, called cyclic AMP response elements, or CREs (reviewed in Montminy, 1997). Since each Tax-responsive 21 bp repeat carried a CRE-like sequence centered within the repeat, it seemed plausible that members of this family of cellular proteins interacted with the Tax-responsive HTLV-I promoter elements. Specific members of the ATF/CREB family of transcription factors have been confirmed to bind the 21 bp repeat elements, interact with Tax, and possibly mediate Tax transactivation (Yoshimura et al., 1990; Zhao and Giam, 1991; Tsujimoto et al., 1991; Beimling and Moelling, 1992; Franklin et al., 1993; Suzuki et al., 1993).

Although several members of the ATF/CREB family have been demonstrated to bind the viral CREs, the transcription factor CREB (cAMP-response element binding protein) appears to have the most prominent role in mediating Tax transactivation (Zhao and Giam, 1991; Beimling and Moelling, 1992; Franklin et al., 1993; Adya et al., 1994; Brauweiler et al., 1995; Adya and Giam, 1995; Yin et al., 1995a). The characterization of the CREB interaction with the viral CRE core led to an intense investigation on how Tax and CREB interact to enhance HTLV-1 transcription. Many researchers hypothesized that Tax transactivation resulted directly from an interaction between Tax and CREB, possibly resulting in enhancement in the transcriptional activation properties of CREB. In support of this hypothesis, several studies have shown that Tax increases the equilibrium binding affinity of CREB for the viral CRE (Zhao and Giam, 1992; Wagner and Green, 1993; Franklin et al., 1993; Anderson and Dynan, 1994; Baranger et al., 1995; Perini et al., 1995; Brauweiler et al., 1995; Yin and Gaynor, 1996a; Kwok et al., 1996). This increase in the apparent binding affinity derives, at least in part, from interactions between Tax and the bZIP segment of CREB. Consistent with this observation, Tax has also been found to enhance the DNA binding affinity of a number of related bZIP proteins (Armstrong et al., 1993; Wagner and Green, 1993; Low et al., 1994; Baranger et al., 1995; Perini et al., 1995). Although biochemical and genetic evidence suggest that Tax interacts with the leucine zipper domain, the basic DNA binding domain appears to be the primary Tax-recognition element within the CREB bZIP segment (Yin et al., 1995a, 1995b; Adya et al., 1994; Baranger et

Since the Tax-responsive 21 bp repeats carry CRE elements, they are now commonly referred to as viral CREs. Centered within each viral CRE is an 8-bp offconsensus CRE core sequence, which serves as the binding site for ATF/CREB proteins (viral CRE core). Immediately flanking the viral CRE core are short sequences rich in guanine and cytosine nucleotides (GCrich flanks). Both the viral CRE core and GC-rich flanks are critical for Tax transactivation in vivo and in vitro (Jeang et al., 1988; Fujisawa et al., 1989; Montagne et al., 1990; Numata et al., 1991; Paca-Uccaralertkun et al., 1994; Brauweiler et al., 1995; Giebler et al., 1997; Lenzmeier et al., 1998). The sequences of the three HTLVI viral CREs, as well as a cellular CRE that is not Taxresponsive, are shown in Figure 1.

329

Lenzmeier and Nyborg: Tax protein of HTLV-1 in adult T-cell leukemia al., 1995; Perini et al., 1995). This Tax-basic region interaction may stabilize the " helical structure of the parallel bZIP dimers, resulting in both enhanced DNA binding and dimerization of CREB (Baranger et al., 1995; Perini et al., 1995). These data are in agreement with other data suggesting that dimerization is tightly linked to DNA binding (Ellenberger et al., 1992; Wu et al., 1998). Furthermore, Tax increases the DNA binding affinity of a covalently cross-linked basic segment dimer, indicating that Tax interaction with just the basic amino acids is sufficient for increased binding affinity (Baranger et al., 1995). Together, these observations are consistent with a model where Tax contributes energetically to the stability of the CREB-DNA interaction.

activities of Tax remained elusive, as several studies had been unsuccessful in identifying a direct interaction between Tax and the viral CRE DNA (Nyborg et al., 1988; Altman et al., 1988; Jeang et al., 1988; Wagner and Green, 1993; Paca-Uccaralertkun et al., 1994). Although an inability to detect a Tax-DNA interaction did not preclude its existence, the absence of evidence for an interaction supported the widely held view that Tax does not bind DNA. Recently, high-resolution DNA footprinting techniques have provided evidence that Tax directly contacts the GCrich nucleotides within the viral CRE. Tax specifically expanded the cleavage protection pattern of CREB from the viral CRE core into the GC-rich flanking sequences (Lenzmeier et al., 1998; Lundblad et al., 1998). This Taxdependent expansion of the CREB footprint required the GC-rich sequences, as the expansion was not observed with a cellular CRE unresponsive to Tax (Lenzmeier et al., 1998). Protein-DNA cross-linking studies confirmed that Tax was intimately associated with the viral CRE flanking sequences, strengthening the likelihood that the changes in the cleavage protection pattern on the viral CRE were due to Tax (Lenzmeier et al., 1998; Kimzey and Dynan, 1998). Additional studies, using inosine substitution and GC-specific DNA binding drugs, provided corroborating evidence for a direct Tax interaction with the viral CRE DNA, and further showed that Tax contacted the minor groove of the DNA (Lundblad et al., 1998; Lenzmeier et al., 1998). The observation of a minor groove interaction perhaps accounts for the inability to detect the Tax-DNA interaction in previous studies. Together, these data provide a strong body of evidence supporting a direct interaction between Tax and the minor groove of the GC sequences adjacent to the CRE core, thereby defining a functional role for these viral promoter sequences.

The Tax-induced increase in the binding affinity of CREB indicates that Tax forms a stable ternary complex with CREB and the viral CRE. The physical incorporation of Tax into the complex has been observed in electrophoretic mobility shift assays (EMSA) where the mobility of the CREB-viral CRE complex is reduced in the presence of Tax (Zhao and Giam 1991, 1992; Brauweiler et al., 1995; Goren et al., 1995; PacaUccaralertkun et al., 1995). Many additional approaches have been utilized to demonstrate the presence of Tax in complexes containing CREB and the viral CRE (Zhao and Giam, 1992; Wagner and Green, 1993; Kwok et al., 1996; Yin and Gaynor, 1996a, 1996b; Lenzmeier et al., 1998; Kimzey and Dynan, 1998). Together, these studies provided support for a model where Tax transactivation resulted from the increased occupancy of the transcription factor CREB on the HTLV-I promoter.

C. Tax interaction with the viral CREs The molecular interactions between Tax and CREB strongly suggest that CREB plays a pivotal role in mediating Tax transactivation. However, the functional activity of Tax also requires the short stretches of GC-rich sequences which immediately flank the viral CRE core (see Figure 1). These GC-rich sequences are absolutely required for Tax transactivation in vivo and in vitro, as CREs lacking the GC-rich flanking sequences, are unresponsiveness to Tax (Jeang et al., 1988; Fujisawa et al., 1989; Montagne et al., 1990; Numata et al., 1991; Seeler et al., 1993; Paca-Uccaralertkun et al., 1994; Brauweiler et al., 1995; Yin and Gaynor, 1996a; Giebler et al., 1997; Lenzmeier et al., 1998). Consistent with the Tax functional studies, the flanking sequences are also required for Tax-CREB-viral CRE ternary complex formation and Tax-dependent CREB stabilization on a viral CRE (Zhao and Giam, 1992; Paca-Uccaralertkun et al., 1994; Brauweiler et al., 1995; Kwok et al., 1996; Yin and Gaynor, 1996a, 1996b; Brauweiler et al., 1995). Until recently, the role of the GC sequences in mediating the

D. Tax dimerization The apparent binding of Tax to both GC sequences flanking the viral CRE suggests that Tax exists as a dimer when incorporated into the ternary complex. In support of this idea, recent studies have provided additional evidence that Tax functions as a dimer (Tie et al., 1996; Jin et al., 1997a). Although the location of the putative Tax dimerization domain is controversial, point mutations have been identified which abrogate both Tax self-association and Tax entry into a stable Tax-CREB-viral CRE ternary complex (Tie et al., 1996; Jin et al., 1997a). Additionally, the yeast 2-hybrid assay revealed protein-protein interactions between Tax monomers in vivo (Tie et al., 1996; Jin et al., 1997b). Consistent with a functional role for Tax dimerization, Tax mutants deficient for selfassociation in vitro and in vivo were unable to activate

330

Gene Therapy and Molecular Biology Vol 3, page 331 transcription from the HTLV-1 promoter (Tie et al., 1996; Jin et al., 1997a). While there is good evidence that Tax exists as a dimer, the role of dimerization in Tax function is still somewhat controversial. Quantitative EMSA experiments indicate that only one Tax molecule is present in the viral CRE nucleoprotein complex (Giebler et al., 1997). Additionally, it has been established that only one of the two GC-rich flanking sequences needs to be present for Tax transactivation through the viral CREs, inferring that one Tax molecule per viral CRE is sufficient for Tax transactivation (Jeang et al., 1988; Fujisawa et al., 1989; Montagne et al., 1990; Numata et al., 1991; Seeler et al., 1993; Paca-Uccaralertkun et al., 1994). Although Tax monomers may be adequate, maximal Tax transactivation probably occurs when two Tax molecules are present on the viral CREs. Future studies which more rigorously address the stoichiometry of the viral CRE nucleoprotein complex, as well as further clarification of the nature of the Tax-Tax interaction, and its functional role in Tax transactivation, will help to solidify the role of Tax dimerization in Tax function.

F i g u r e 2 . Model for the Tax-CREB-viral CRE ternary complex. The DNA is shown in red and the bZIP domain of CREB is shown in y e l l o w . This model is based on the crystal structure of the yeast bZIP protein GCN4 (Ellenberger et al., 1992).

E. Coactivator recruitment by Tax A model showing Tax incorporated into the ternary complex with CREB and the viral CRE is presented in Figure 2. This model illustrates the interaction between Tax and the GC-rich DNA, as well as the interaction between Tax and the bZIP segment of CREB bound to the viral CRE core. The multiple protein-protein and proteinDNA contacts confer significant stability to the nucleoprotein complex, and are believed to be critical for Tax transactivation of HTLV-I gene expression. However, this model incompletely describes the precise molecular events that lead to strong transcriptional stimulation by Tax. Recent studies suggests that Tax stabilization of CREB on the viral promoter would not result in strong transcriptional activation, as the activity of CREB is largely dependent upon phosphorylation by cAMPdependent protein kinase A (PKA) (Gonzalez et al., 1989; reviewed in Montminy, 1997). Once CREB is phosphorylated, it directly recruits the cellular coactivator CREB binding protein (CBP) to promoters of CREBresponsive genes (Chrivia et al., 1993; Kwok et al., 1994). Since there is no evidence that Tax directly or indirectly phosphorylates CREB, a model proposing that Tax simply serves to stabilize CREB binding to the viral promoter insufficiently accounts for the strong HTLV-I transcriptional activation observed in the presence of Tax. The identification of CBP in mediating CREB-activated transcription, together with the recently discovered pleiotropic effects of CBP on many cellular transcription factor pathways, has led several groups to investigate

whether Tax may recruit CBP to the HTLV-I. In support of this hypothesis, published studies have now confirmed that Tax, in the context of the ternary complex, directly recruits CBP to the HTLV-I promoter to activate transcription (Kwok et al., 1996; Giebler et al., 1997). Tax interacts with CBP with relatively high affinity, recruiting CBP to the HTLV-I promoter to form a highly stable quaternary complex (Kwok et al., 1996; Yin and Gaynor, 1996b; Laurance et al., 1997; Giebler et al., 1997; Harrod et al., 1998). CBP recruitment is dependent upon CREB in the complex, although it is independent of the phosphorylation state of CREB (Kwok et al., 1996; Laurance et al., 1997; Giebler et al., 1997). In fact, the entire amino terminus of CREB (including the PKAphosphorylation domain), was dispensable for CBP recruitment, as the bZIP domain was sufficient for quaternary complex formation and Tax transactivation in vivo (Laurance et al., 1997; Giebler et al., 1997). The GCrich sequences of the viral CREs have also been demonstrated to be critical for efficient recruitment of CBP by Tax (Kwok et al., 1996; Giebler et al., 1997; Lenzmeier et al., 1998).

331

Gene Therapy and Molecular Biology Vol 3, page 332

F i g u r e 3 . Schematic illustration of the HTLV-1 promoter. For simplicity, only one of the three viral CREs is shown.

transcription machinery (Abraham et al., 1993; Kwok et al., 1994; Kee et al, 1996., Swope et al., 1996). Recent studies on Tax and CBP have begun to characterize the amino acids of Tax required for CBP recognition (Bex et al., 1998; Harrod et al., 1998), and the consequences of the Tax-CBP interaction on cellular gene expression (Colgin and Nyborg, 1998; Van Orden et al., submitted for publication). A model illustrating the interactions of CBP on the HTLV-1 promoter is shown in Figure 3.

Although CBP is a very large protein (2441 amino acids), Tax has been shown to interact with a small region of CBP called the KIX domain (Kwok et al., 1996; Giebler et al., 1997). Tax specifically binds KIX amino acids 588683, overlapping significantly with the region of KIX that is recognized by phosphorylated CREB (aa 586-665) (Yan et al., 1998; Radhakrishnan et al., 1997). These KIX amino acids fold into three "-helices which come together to form a hydrophobic protein-docking pocket (Radhakrishnan et al., 1997). Although the amino acids of KIX that bind Tax and CREB are very similar, specific point mutations within KIX distinguish between the binding of Tax and phosphorylated CREB, suggesting that the precise molecular recognition of KIX by Tax and CREB are distinct (Yan et al., 1998). This observation is supported by evidence showing that the phosphorylation of CREB increases the affinity of KIX for the Tax-containing ternary complex over 14-fold. These data support the idea that KIX can recognize both Tax and phosphorylated CREB simultaneously in the context of the Tax-CREBviral CRE complex (Giebler et al., 1997). This model is in agreement with previous studies showing that Tax transactivation was independent of, but augmented by, CREB phosphorylation (Poteat et al., 1989; Kadison et al., 1990).

III. Consequences of Tax expression on cellular gene expression In the HTLV-I infected cell, Tax expression appears to dysregulate various pathways of cellular gene expression, which may explain the tight link between Tax and malignant transformation. The classes of cellular gene products whose expression may be dysregulated by Tax are presented in Figure 4. These include: (i ) tumor suppressors (Jeang et al., 1990; Uittenbogaard et al., 1995; Feigenbaum et al., 1996); (i i ) apoptosis regulators (Brauweiler et al., 1997); (i i i ) transcription factors (Fujii et al., 1988; Alexandre and Verrier, 1991; Alexandre et al., 1991; Duyao et al., 1992); (i v ) extracellular signaling mitogens (Maruyama et al., 1987; Leung et al., 1988; Nimer et al., 1989; Kim et al., 1990; Grassman et al., 1992; reviewed in Franchini, 1995); (v) cell surface receptors (Inoue et al., 1986; Siekevitz et al., 1987; Cross et al., 1987); (v i ) signal transduction kinases (Uchiumi et al., 1992; Lemasson et al., 1997); (v i i ) cytoskeletal components (Lilienbaum et al., 1990); (v i i i ) DNA replication components (Ressler et al., 1997); and (i x ) 5S RNAs and tRNAs (Gottesfeld et al., 1996).

Together, the above data have established a critical role for CBP in Tax transactivation. Efficient HTLV-I transcription is likely dependent upon the stable Taxternary complex serving as a high affinity binding site for CBP. It is believed that the presence of CBP promotes transcriptional activation through its intrinsic and associated acetyltransferase activities (Bannister and Kouzarides, 1996; Ogryzko et al., 1996; Yang et al., 1996; Imhof et al., 1997), and through its link to the basal 332

Gene Therapy and Molecular Biology Vol 3, page 333

F i g u r e 4 . Overview of the cellular processes deregulated by the HTLV-I Tax protein.

T-cells (NFAT) (Good et al., 1996; Rivera et al., 1998), and serum response factor (SRF) (Alexandre and Verrier, 1991; Alexandre et al., 1991; Fujii et al., 1992). Interestingly, most of these transcriptional activators have also been shown to utilize the coactivators CBP and p300 (see Figure 5), suggesting that mechanisms of Tax deregulation of cellular gene expression may converge on alterations in the transcription properties of CBP. Of all the cellular transcription factors putatively targeted by Tax, NF-!B clearly plays the most significant role in Tax activation of cellular gene expression. The data associating NF-Y, Ets1, SP1, NFAT, and SRF with Tax transactivation is intriguing, but more studies aimed at defining the importance of these factors for HTLV-1 propagation and T-cell transformation are necessary before they are widely accepted as targets of Tax-deregulation. Additionally, although the ATF/CREB transcription factors are likely integral components in the mechanism of HTLV-1 transactivation by Tax, they have not been specifically implicated as targets of cellular gene deregulation by Tax.

While improper expression of each of these gene products alone might contribute to a transformation event, it is likely that the global effects of Tax on cellular gene expression are responsible for Tax-mediated transformation. Some cellular genes are activated by Tax, while others are repressed by Tax. The infrequent outcome of these pleiotropic properties of Tax appears to be promotion of uncontrolled T-cell growth. This section will focus on the mechanisms by which Tax both activates and represses cellular gene expression.

A. Mechanisms for Tax activation of cellular gene expression Many cellular transcription factors have been implicated as targets of deregulation by the HTLV-1 Tax protein. Among these are the ATF/CREB family (reviewed above), nuclear factor kappa B (NF-!B) (reviewed below), nuclear factor Y (NF-Y) (Pise-Masison et al., 1997), Ets1 and Sp1 (Dittmer et al., 1997), nuclear factor of activated

333

Gene Therapy and Molecular Biology Vol 3, page 334

F i g u r e 5 . Functional domains of the pleiotropic cellular coactivator CBP. The approximate amino acid locations of the various domains are indicated, as are the cellular proteins which interact with those respective domains.

in the cytoplasm by interactions with inhibitor kappa B (I!B), which binds to and masks its nuclear localization signal. Activation of the NF-!B pathway occurs following the phosphorylation of I-!B by a large multiprotein kinase complex (DiDanato et al., 1997; Mercurio et al., 1997; Zandi et al., 1997). This phosphorylation event leads to the ubiquitination and proteasomal degradation of I-!B (Baldi et al., 1996; Chen et al., 1995), concomitant with the release of NF-!B for nuclear localization and activation of NF-!B-responsive genes (reviewed in Baeuerle and Baltimore, 1996). Although HTLV-I studies have established that Tax increases the rate of I-!B degradation in the cytoplasm, and enhances NF-!B DNA binding by increasing the fraction of NF-!B found in the nucleus, little was known about how Tax activated this pathway until recently (Kanno et al., 1994; Lacoste et al., 1995; McKinsey et al., 1996; Good and Sun, 1996; Lilienbaum and Paulin, 1993; Pepin et al., 1994; Hirai et al., 1994; Suzuki et al., 1995; reviewed in Flint and Shenk, 1997).

This is probably because most cellular genes do not carry the appropriate GC-rich sequences immediately flanking the CREB binding sites. From these observations, it appears that Tax deregulation of the NF-!B transcription pathway may represent the most prominent pathway in Tax activation of cellular gene expression. In addition, since activate NF-!B proteins appear to inhibit apoptosis, Tax activation of this pathway may promote survival of the HTLV-I-infected cell, thus enhancing survival of the virus (Wang et al., 1996; reviewed in Baeuerle and Baltimore, 1996; Gilmore et al., 1996).

B. Tax activation of the NF- B pathway The NF-!B/rel transcription factors function normally in the cell to enhance expression of genes involved in mitogen-driven proliferation (reviewed in Baeuerle and Baltimore, 1996). Each member of the NF-!B/rel family of transcription factors contains a rel homology domain of approximately 300 amino acids, which is critical for NF!B dimerization, DNA binding, and nuclear localization (reviewed in Thanos and Maniatis, 1995). Regulation of NF-!B transactivation activity is by subcellular compartmentalization. The inactive form of NF-!B is kept

Three hypotheses have recently been proposed for the molecular events that lead to Tax-activation of NF-!Bresponsive genes. One study found that Tax directly interacts with two subunits of the proteasome, suggesting that proteasomal interactions are critical for maturation and 334

Gene Therapy and Molecular Biology Vol 3, page 335 nuclear localization of the active NF-!B dimer (Rousset et al., 1996). There is growing support for a second hypothesis which states that Tax induces I-!B phosphorylation, promoting release of NF-!B to the nucleus. Several groups have recently shown that Tax constituitively activates both of the I-!B kinases, IKK" and IKK#, resulting in direct phosphorylation of I-!B (Chu et al., 1998; Uhlik et al., 1998; Geleziunas et al., 1998). This observation has been extended to show that Tax also works immediately upstream of the IKKs by activating NF-!B-inducing kinase (NIK), a kinase capable of IKK"/# phosphorylation and activation (Uhlik et al., 1998; Geleziunas et al., 1998). Another group has found that Tax modulates I-!B phosphorylation upstream of the IKKs by functionally interacting with MEKK1, which directly phosphorylates and activates IKK#, but not IKK" (Yin et al., 1998). Together, these studies indicate that Tax expression increase the cytoplasmic activities of one or more of the kinases responsible for I-!B phosphorylation, leading to activation of the NF-!B pathway. Finally, a third hypothesis has been proposed where Tax also activates NF-!B-dependent transcription in the nucleus. CBP and the related coactivator p300 play important roles in NF-!B transcription (Gerritsen et al., 1997; Perkins et al., 1997; Zhong et al., 1998), and Tax has recently been shown to participate in the formation of distinct nuclear structures containing both NF-!B and p300 (Bex et al., 1997, 1998). These observations provide evidence that Tax may exert an effect on NF-!B transcription function directly at NF-!B-responsive promoters. A schematic outlining the mechanisms of Tax activation of cellular gene expression through the NF-!B pathway is presented in Figure 4. It is intriguing that Tax may function in both the cytoplasm and the nucleus to activate the same cellular transcription factor pathway.

promoters of the p53 gene, the Bax gene, and the lck gene (Uittenbogaard et al., 1995; Brauweiler et al., 1997; Lemasson et al., 1997). Although there is strong evidence that Tax represses transcription through bHLH proteins and E-box DNA elements, the mechanism of Tax repression remains elusive. There is evidence that the protein structure of the bHLH protein c-myc is altered in Tax-expressing cell, however, Tax does not appear to physically interact with any of the bHLH proteins (Semmes et al., 1996). These data support a model where Tax indirectly influences bHLH transcription factor activity, producing repression of bHLH-regulated cellular genes. It seems plausible that Tax repression of important genes like p53 and #-polymerase may be directly linked with the oncogenic properties of Tax.

D. Tax repression of cellular gene expression through CBP A new model for Tax repression of cellular gene expression is beginning to emerge. This model is based on the idea that Tax utilization of CBP for HTLV-I transcription reduces the concentration of available CBP for cellular transcription factor pathways. Although CBP appears to be ubiquitously expressed, the levels of intracellular CBP appear limiting (Petrij et al., 1995; Shi et al., 1998; Yao et al., 1998). Since CBP serves as a pleiotropic coactivator for a large number of structurally unrelated cellular transcription factors (see Figure 5), competition for limiting CBP can effectively abrogate CBP function (Arias et al., 1994; Kamei et al., 1996; Horvai et al., 1997). Although Tax interacts with the KIX domain of CBP, many cellular transcription factors also utilize this same region for recognition and recruitment of CBP. These include phosphorylated CREB, c-jun, c-myb and STAT1 (see F i g u r e 5 ; reviewed in Janknecht and Hunter, 1996; Shikama et al., 1997; Giles et al., 1998). Because Tax shares the same CBP-docking site as several important cellular transcription factors, it has been hypothesized that Tax binding to KIX might inhibit access of other transcription factors to CBP, thus altering patterns of cellular gene expression. We have recently found that Tax effectively represses c-jun (Van Orden et al., submitted for publication) and c-myb (Colgin and Nyborg, 1998) transcription activity in vivo, and reciprocally, overexpression of these cellular transcription factors represses the transcription function of Tax. The mechanism of repression is likely through a direct competition for limiting levels of intracellular CBP, as the binding of Tax and these two cellular transcription factors to the KIX domain of CBP is mutually exclusive in vitro (Colgin and Nyborg, 1998; Van Orden et al., submitted for publication). A general model for CBP competition as a mechanism of Tax repression is presented in F i g u r e 6

C. Tax repression of cellular gene expression through the bHLH proteins In 1990, it was reported for the first time that Tax represses expression of a cellular gene (Jeang et al., 1990). Transcription of the #-polymerase gene, which encodes a tumor suppressor important for host-cell DNA repair, was shown to be repressed in the presence of Tax. When the #polymerase study was published, the details of Tax repression was not known. Several years later, members of the cellular basic-helix-loop-helix (bHLH) family of transcription factors were implicated as the targets of Tax repression (Uittenbogaard et al., 1994; Semmes et al., 1996). A bHLH binding site, called an E-box, was identified in the promoter of the #-polymerase gene and shown to confer Tax repression to a heterologous promoter (Uittenbogaard et al., 1994). Tax has since been shown to utilize the cellular bHLH proteins to confer repression to

335

Gene Therapy and Molecular Biology Vol 3, page 336

F i g u r e 6 . Model for Tax repression of cellular gene expression through competition for the cellular coactivator CBP.

regulation of cell cycle and differentiation genes. It seems plausible that Tax binding to the KIX domain may in some way mimic the deregulation that is achieved following chromosomal translocations involving CBP, with both scenarios promoting malignant transformation.

These observations suggest that Tax and cellular transcription factors may compete for CBP utilization in the HTLV-I infected T-cell. However, the extent and the consequence of this competition may depend upon several criteria, including the abundance of the cellular transcription factors, their relative KIX-binding affinities, and the concentration of available CBP in the cell. Tax expression in an HTLV-I-infected cell is believed to be intermittent, but that during the brief periods of Tax expression, Tax protein levels are high (0.15% of total cell protein; Slamon et al., 1985). It seems likely that during these burst periods, Tax levels would exceed those needed for optimal proviral expression, and by mass action, the high concentrations of free Tax would bind to KIX, sequestering the limiting concentrations of intracellular CBP, and altering CBP-mediated cellular gene expression.

IV. Tax deregulation of cell cycle checkpoints There is also growing evidence that Tax may play a role in leukemigenesis by circumventing cell cycle checkpoints. By physically interacting with and modulating the activity of components that regulate cell division, Tax may deregulate cell cycle progression at the G1-S phase, M-phase, and DNA damage checkpoints. A great deal of progress has recently been made on understanding the molecular basis for Tax modulation of cell-cycle progression. These potential checkpoint interactions by Tax are summarized in Figure 4.

These Tax-dependent effects on CBP function may be directly linked with cellular transformation and adult T-cell leukemia, as a prominent role for CBP in hematopoetic malignancies is emerging (reviewed in Giles et al., 1998). Chromosomal translocations involving CBP are being identified with increasing frequency in patients with treatment-related acute and chronic myeloid leukemias and myelodysplastic syndrome (Borrow et al., 1996; Giles et al., 1997; Rowley et al., 1997; Sobulo et al., 1997; Taki et al., 1997; Ida et al., 1997; Satake et al., 1997). The molecular basis of CBP translocation-associated leukemogenesis is not known; however, the available evidence strongly suggests that chromosomal translocations involving CBP result in reduced and/or defective coactivator function. Because of the pleiotropic role for CBP in cellular gene expression, it is likely that alterations in CBP function promote inappropriate

Tax appears to promote cell cycle progression at the G1-S checkpoint by directly interacting with and functionally inactivating p16INK4a, a cyclin-dependent kinase (cdk) inhibitor (Suzuki et al., 1996; Low et al., 1997). The cdks bound by p16INK4a normally promote progression from G1 phase into DNA synthesis (S-phase) by phosphorylating key regulatory proteins, some of which are transcription factors (reviewed in Sherr and Roberts, 1995). Tax precludes the binding of p16INK4a to the cdks, thereby enabling cdk kinase activity to proceed unchecked. In addition to modulating mitosis progression via inhibiting p16INK4a function, it was also recently shown that Tax directly interacts with the cyclin D-cdk4 (or cdk6) complex in vitro and in vivo (Neuveut et al., 336

Gene Therapy and Molecular Biology Vol 3, page 337 1998). Although the molecular mechanism has not yet been defined, this interaction by Tax somehow increases the cyclin-cdk kinase activity and eventually leads to the hyperphosphorylation of the retinoblastoma (pRb) tumor suppressor protein (Neuveut et al., 1998), which has an established role in mitosis progression. Consistent with this observation, the E2F transcription factor, which is normally kept inactive by unphosphorylated pRB, is constituitively activated by Tax in a p16INK4a-independent manner (Lemasson et al., 1998). Together, the above experimental data suggest that both p16 INK4a and the cyclin-cdk complexes are targets of Tax.

p53 interactions with the basal transcription factor TFIID (TBP), thus inhibiting p53 transcription function (PiseMasison et al., 1998b). Future studies may help determine whether Tax modulation of p53 function is due to direct competition for CBP, or indirect manipulation of cellular signal transduction pathways.

V. Potential anti-viral therapeutic approaches The etiological link of adult T-cell leukemia (ATL) to HTLV-1 has created a field of research committed to understanding the pathogenesis associated with HTLV-1 infection. In particular, the observations that the HTLV-1encoded Tax protein is required for viral replication, and by itself has the ability to transform cells, has lead to intense studies addressing the molecular mechanisms of Tax function. A more complete understanding of how Tax activates HTLV-1 transcription, deregulates cellular gene expression through CBP, and circumvents cell cycle checkpoints will hopefully provide a foundation for the design and development of therapies aimed at the inhibition of viral replication and HTLV-1-mediated malignant transformation.

In addition to affecting mitotic events at the G1-S phase checkpoint, Tax was recently shown to deregulate mitosis progression by binding to and functionally inhibiting the centrosome binding protein MAD1 (Jin et al., 1998). Tax obstruction of MAD1 activity may be significant, as this causes the cell to skip the M-phase checkpoint for proper chromosomal alignment, and improper cytokinesis may proceed (Jin et al., 1998). Consistent with this hypothesis, aberrant chromosome segregation during cytokinesis, and multinucleated cells were detected in HTLV-1 transformed cells (Jin et al., 1998). These observations suggest that Tax manipulates cell cycle progression at two distinct points during mitosis.

Tax transactivation of HTLV-1 is necessary for efficient viral replication. Since viral transcription also drives Tax expression, the high-level Tax expression that is probably necessary for cellular transformation is also dependent upon this process. These two observations make abrogation of HTLV-1 transcription an attractive target for therapeutic approaches. HTLV-1 transcription is very complicated because it involves intricate protein-protein and protein-DNA interactions between viral and cellular components. Studies elucidating these interactions have been challenging, but also have exposed targets for antiTax therapies that may inhibit Tax transactivation and Tax-mediated malignant transformation. The discovery that minor groove binding drugs like chromomycin can preclude Tax-enhancement of CREB binding (Lenzmeier et al., 1998; Lundblad et al., 1998) and Tax recruitment of CBP (Lenzmeier et al., 1998) to the viral CREs, suggests that interruption of the Tax-DNA minor groove interaction may be a viable approach to inhibit HTLV-1 transcription. Unfortunately, the low sequence-specificity and pleiotropic effects of chromomycin (reviewed in Zimmer and Wahnert, 1996) make this molecule undesirable as an anti-Tax therapeutic. The recent development, however, of sequencespecific minor groove binding polyamides (Trauger et al., 1996; White et al., 1998; Kielhopt et al., 1998), which specifically alter gene expression in vivo (Gottesfeld et al., 1997), may provide a potentially viable therapeutic approach for inhibition of Tax transactivation through interruption of the Tax-DNA interaction.

Tax may also deregulate the cell cycle in the presence of DNA damage by functionally inactivating the tumor suppressor transcription factor p53 (Gartenhaus and Wang, 1995; Cereseto et al., 1996; Pise Masison et al., 1998a), which normally halts the cell-cycle in a DNA-damage dependent manner. There is no published evidence for a direct Tax-p53 interaction, indicating that Tax inactivation of p53 is likely indirect. Furthermore, although p53 is mutated in approximately 60% of human cancers, it is generally not mutated in HTLV-1 transformed cells. The mechanism by which Tax inhibits p53 activity is still unknown, and both transcription-dependent and transcription-independent mechanisms may be utilized. The observation that overexpression of p53 represses Tax transactivation of the HTLV-1 gene expression, indicates that Tax and p53 might compete for limiting amounts of a protein critical for both p53 and Tax transcription function (Mori et al., 1997; M. Gonzales and J.K. Nyborg unpublished data). Since p53 functionally utilizes CBP (Avantaggiati et al., 1997; Lill et al., 1997), it is plausible that Tax inhibition of p53 transcription function is simply due to competition for available CBP in the HTLV-I-infected cell. If this model is correct, Tax could promote progression through the cell cycle in the presence of DNA damage, and the resulting increase in genetic mutations could eventually lead to malignant transformation. Alternatively, the constitutive phosphorylation of p53 in HTLV-1 infected T-cells may preclude 337

Lenzmeier and Nyborg: Tax protein of HTLV-1 in adult T-cell leukemia

References

Mechanism of DNA binding enhancement by the HTLV-I transactivator Tax. Nature 376, 606-608.

Abraham, S.E., S. Lobo, P. Yaciuk, H.G. Wang, and E. Moran. 1 9 9 3 . p300, and p300-associated proteins, are components of TATA-binding protein (TBP) complexes. O n c o g e n e 8 , 1639-1647.

Beimling, P., and K. Moelling. 1 9 9 0 . Tax-independent binding of multiple cellular factors to Tax-response element DNA of HTLV-I. O n c o g e n e 5 , 361-368. Beimling, P., and K. Moelling. 1 9 9 2 . Direct interaction of CREB protein with 21 bp Tax-response elements of HTLV-I LTR. O n c o g e n e 7 , 257-262.

Adya, N., L.-J. Zhao, W. Huang, I. Boros, and C.-Z. Giam. 1 9 9 4 . Expansion of CREB’s DNA recognition specificity by Tax results from interaction with Ala-AlaArg at position 282-284 near the conserved DNA-binding domain of CREB. P r o c . N a t l . A c a d . S c i . U S A 9 1 , 5642-5646.

Bex, F., A. McDowall, A. Burny, and R.B. Gaynor. 1 9 9 7 . The human T-cell leukemia virus type 1 transactivator protein Tax colocalizes in unique nuclear structures with NF-kappaB proteins. J . V i r o l . 7 1 , 3484-3497.

Adya, N., and C.-Z. Giam. 1 9 9 5 . Distinct regions in human T cell lymphotropic virus type I Tax mediate interactions with activator protein CREB and basal transcription J . V i r o l . 6 9 , 1834-1841.

Bex, F., M.J. Yin, A. Burny, and R.B. Gaynor. 1 9 9 8 . Differential transcriptional activation by human T-cell leukemia virus type 1 Tax mutants is mediated by distinct interactions with CREB binding protein and p300. M o l . C e l l B i o l 1 8 , 2392-2405.

Alexandre, C., and B. Verrier. 1 9 9 1 . Four regulatory elements in the human c-fos promoter mediate transactivation by HTLV-1 Tax protein. O n c o g e n e 6 , 543551.

Borrow, J., V.P. Stanton, M. Andresen, R. Becher, F.G. Behm, R.S.K. Chaganti, C.I. Civin, C. Disteche, I. Dube, A.M. Frischauf, D. Horsman, F. Mitelman, S. Volinia, A.E. Watmore, and D.E. Housman. 1 9 9 6 . The translocation t(8;16)(p11;p13) of acute myeloid leukaemia fuses a putative acetyltransferase to the CREBbinding protein. Nature Genetics 14, 33-41.

Alexandre, C., P. Charnay, and B. Verrier. 1 9 9 1 . Transactivation of Krox-20 and Krox-24 promoters by the HTLV-I Tax protein through common regulatory elements. O n c o g e n e 6 , 1852-1857. Altman, R., D. Harrich, J.A. Garcia and R.B. Gaynor. 1 9 8 8 . Human T-Cell Leukemia Virus Types I and II exhibit different DNase I protection patterns. J . V i r o l . 6 2 , 1339-1346.

Brady, J., K.-T. Jeang, J. Duvall and G. Khoury. 1 9 8 7 . Identification of p40 x -responsive regulatory sequences within the Human T-cell Leukemia Virus Type I long terminal repeat. J . V i r o l . 6 1 , 2175-2181.

Anderson, M.G., and W.S. Dynan. 1 9 9 4 . Quantitative studies of the effect of HTLV-I Tax protein on CREB protein-DNA binding. N u c . A c i d s R e s . 2 2 , 3194-3201.

Brauweiler, A., P. Garl, A.A. Franklin, H.A. Giebler, and J.K. Nyborg. 1 9 9 5 . A molecular mechanism for HTLV-I latency and Tax transactivation. J . B i o l . C h e m . 2 7 0 , 12814-12822.

Arias, J., A.S. Alberts, P. Brindle, F.X. Claret, T. Smeal, M. Karin, J. Feramisco, and M. Montminy. 1 9 9 4 . Activation of cAMP and mitogen responsive genes relies on a common nuclear factor. Nature 370, 226-229.

Brauweiler, A., J.E. Garrus, J.C. Reed, and J.K. Nyborg. 1 9 9 7 . Repression of bax gene expression by the HTLV-1 Tax protein, implications for supression of apoptosis in virally infected cells. V i r o l o g y 2 3 1 , 135-140.

Armstrong, A.P., A.A. Franklin, M.N. Uittenbogaard, H.A. Giebler, and J.K. Nyborg. 1 9 9 3 . Pleiotropic effect of the Human T-cell Leukemia Virus Tax protein on the DNA binding activity of eucaryotic transcription factors. P r o c . N a t l . A c a d . S c i . U S A 9 0 , 7303-7307.

Cann, A.J., J.D. Rosenblatt, W. Wachsman, N.P. Shaw, and I.S.Y. Chen. 1 9 8 5 . Identification of the gene responsible for human T-cell leukaemia virus transcriptional regulation. Nature 318, 571-574.

Avantaggiati, M.L., V. Ogryzko, K. Gardner, A. Giordano, A.S. Levine, and K. Kelly. 1 9 9 7 . Recruitment of p300/CBP in p53-dependent signal pathways. C e l l 8 9 , 1175-1184. Baeuerle, P.A., and D. Baltimore. 1 9 9 6 . NF-kappa B, ten years after. C e l l 8 7 , 13-20.

Cereseto, A., F. Diella, J.C. Mulloy, A. Cara, P. Michieli, R. Grassmann, G. Franchini, and M.E. Klotman. 1 9 9 6 . p53 functional impairment and high p21waf1/cip1 expression in human T-cell lymphotropic/leukemia virus type Itransformed T cells. B l o o d 8 8 , 1551-1560.

Baldi, L., K. Brown, G. Franzoso, and U. Siebenlist. 1 9 9 6 . Critical role for lysines 21 and 22 in signal-induced, ubiquitin-mediated proteolysis of I kappa B-alpha. J . B i o l . C h e m . 2 7 1 , 376-379.

Chen, I.S.Y., J . McLaughlin, J.C. Gasson, S.C. Clark, and D.W. Golde. 1 9 8 3 . Molecular characterization of genome of a novel human T-cell leukaemia virus. N a t u r e 3 0 5 , 5 02-505.

Bannister, A.J., and T. Kouzarides. 1 9 9 6 . The CBP coactivator is a histone acetyltransferase. Nature 3 8 4 , 641-643.

Chen, I.S.Y., D.J. Slamon, J.D. Rosenblatt, N.P. Shah, S.G. Quan and W. Wachsman. 1 9 8 5 . The x gene is essential for HTLV replication. S c i e n c e 2 2 9 , 54-58.

Baranger, A.M., C.R. Palmer, M.K. Hamm, H.A. Giebler, A. Brauweiler, J.K. Nyborg, and A. Schepartz. 1 9 9 5 .

Chen, Z., J. Hagler, V.J. Palombella, F. Melandri, D. Scherer, D. Ballard, and T. Maniatis. 1 9 9 5 . Signal-induced sitespecific phosphorylation targets I kappa B alpha to the

338

Gene Therapy and Molecular Biology Vol 3, page 339 ubiquitin-proteasome pathway. G e n e s D e v . 9 , 15861597.

Transactivation by the Human T-cell Leukemia Virus Tax protein is mediated through enhanced binding of Activating Transcription Factor-2 (ATF-2) and cAMP Element-binding Protein (CREB). J . B i o l . C h e m . 2 6 8 , 21225-21231.

Chrivia, J.C., R.P. Kwok, N. Lamb, M. Hagiwara, M.R. Montminy, and R.H. Goodman. 1 9 9 3 . Phosphorylated CREB binds specifically to the nuclear protein CBP. Nature 365, 855-859.

Franklin, A.A. and J.K. Nyborg. 1 9 9 5 . Mechanisms of Tax regulation of Human T-Cell Leukemia Virus Type I gene expression. J . B i o m e d . S c i . 2 , 17-29.

Chu, Z.L., J.A. DiDonato, J. Hawiger, and D.W. Ballard. 1 9 9 8 . The Tax oncoprotein of human T-cell leukemia virus type 1 associates with and persistently activates IkappaB kinases containing IKKalpha and IKKbeta. J . B i o l . C h e m . 2 7 3 , 15891-15894.

Fujii, M., P. Sassone-Corsi, and I.M. Verma. 1 9 8 8 . c-fos promoter trans-activation by the Tax1 protein of human T-cell leukemia virus type I. P r o c . N a t l . A c a d . S c i . U S A 8 5 , 8526-8530.

Colgin, M., and J.K. Nyborg. 1 9 9 8 . The Human T-Cell Leukemia Virus Type 1 Oncoprotein Tax Inhibits the Transcriptional Activity of c-Myb through Competition for the CREB Binding Protein. J . V i r o l . 7 2 , 93969399.

Fujii, M., H. Tsuchiya, T. Chuhjo, T. Akizawa, and M. Seiki. 1 9 9 2 . Interaction of HTLV-I with p67 SRF Causes the Aberrant Induction of Cellular Immediate Early Genes Through CArG Boxes. G e n e s . D e v . 6 , 2066-2076.

Cross, S.L., M.B. Feinberg, J.B. Wolf, N.J. Holbrook, F. Wong-Staal, and W.J. Leonard. 1 9 8 7 . Regulation of the human interleukin-2 receptor a chain promoter, activation of a non-functional promoter by the transactivator gene of HTLV-1. C e l l 4 9 , 47-56.

Fujisawa, J.-I., M. Seiki, T. Kiyokawa, and M. Yoshida. 1 9 8 5 . Functional activation of the human T-cell leukemia virus type I by transacting factor. P r o c . N a t l . A c a d . S c i . U S A 8 2 , 2277-2281. Fujisawa, J.-I., M. Toita, M. Sato, and M. Yoshida. 1 9 8 6 . A transcriptional enhancer sequence of HTLV-1 is responsible for transactivation mediated by p40x of HTLV-1. EMBO J. 5, 713-718.

DiDonato, J.A., M. Hayakawa, D.M. Rothwarf, E. Zandi, and M. Karin. 1 9 9 7 . A cytokine-responsive IkappaB kinase that activates the transcription factor NF-kappaB. Nature 388, 548-554.

Fujisawa, J.-I., M. Toita, and M. Yoshida. 1 9 8 9 . A unique enhancer element for the transactivator (p40Tax) of human T cell leukemia virus type I that is distinct from cyclicAMP- and 12-O-tetradecanoylphorbal-13-acetateresponsive element. J . V i r o l . 6 3 , 3234-3239.

Dittmer, J., C.A. Pise-Masison, K.E. Clemens, K.S. Choi, and J.N. Brady. 1 9 9 7 . Interaction of human T-cell lymphotropic virus type I Tax, Ets1, and Sp1 in transactivation of the PTHrP P2 promoter. J . B i o l . C h e m . 2 7 2 , 4953-4958.

Gartenhaus, R.B., and P. Wang. 1 9 9 5 . Functional inactivation of wild-type p53 protein correlates with loss of IL-2 dependence in HTLV-I transformed human T lymphocytes. L e u k e m i a 9 , 2082-2086.

Duyao, M.P., D.J. Kessler, D.B. Spicer, C. Bartholomew, J.L. Cleveland, M. Siekevitz, and G.E. Sonenshein. 1 9 9 2 . Transactivation of the c-myc promoter by human T cell leukemia virus type 1 Tax is mediated by NF kappa B. J . B i o l . C h e m . 2 6 7 , 16288-16291. Ellenberger, T.E., C.J. Brandl, K. Struhl, and S.C. Harrison. 1 9 9 2 . The GCN4 basic region leucine zipper binds DNA as a dimer of uninterrupted alpha helices, crystal structure of the protein-DNA complex. C e l l 7 1 , 1223-1237.

Geleziunas, S., S. Ferrell, X. Lin, Y. Mu, E.T. Cunningham Jr., M. Grant, M.A. Connelly, J .E. Hambor, K.B. Marcu, and W.C. Greene. 1 9 9 8 . Human T-Cell Leukemia Virus Type 1 Tax Induction of NF-kB Involves Activation of the IkB Kinase (IKKa) and IKKb Cellular Kinases. M o l . C e l l . B i o l . 1 8 , 5157-5165.

Feigenbaum, L., K. Fujita, F.S. Collins, and G. Jay. 1 9 9 6 . Repression of the NF1 gene by Tax may explain the development of neurofibromas in human T-lymphotropic virus type 1 transgenic mice. J . V i r o l . 7 0 , 3280-3285.

Gerritsen, M.E., A.J. Williams, A.S. Neish, S. Moore, Y. Shi, and T. Collins. 1 9 9 7 . CREB-binding protein/p300 are transcriptional coactivators of p65. P r o c . N a t l . A c a d . S c i . U S A 9 4 , 2927-2932.

Felber, B.K., H. Paskalis, C. Kleinman-Ewing, F. WongStaal, and G.N. Pavlakis. 1 9 8 5 . The pX protein of HTLVI is a transcriptional activator of its long terminals repeats. S c i e n c e 2 2 9 , 675-679.

Gessain A., F. Barin, J.C. Vernant, O. Gout, L. Maurs, and A. de Calender. 1 9 8 5 . Antibodies to human T-lymphotropic virus type I in patients with tropical spastic paraparesis. Lancet 2, 407-410.

Flint, J., and T. Shenk. 1 9 9 7 . Viral Transactivating Proteins. A n n u . R e v . G e n e t . 3 1 , 177-212.

Giam, C.-Z., and Y.-L. Xu. 1 9 8 9 . HTLV-I Tax Gene Product Activates Transcription via Pre-existing Cellular Factors and cAMP Responsive Element. J . B i o l . C h e m . 2 6 4 , 15236-15241.

Franchini, G. 1 9 9 5 . Molecular Mechanisms of Human T-Cell Leukemia/Lymphotropic Virus Type I Infection. B l o o d 8 6 , 3619-3639.

Giebler, H.A., J.E. Loring, K. van Orden, M.A. Colgin, J.E. Garrus, K. Escudero, A. Brauweiler, and J.K. Nyborg. 1 9 9 7 . Anchoring of CBP to the HTLV-I Promoter, A

Franklin, A.A., M.F. Kubik, M.N. Uittenbogaard, A. Brauweiler, P. Utaisincharoen, M.-A.H. Matthews, W.S. Dynan, J.P. Hoeffler, and J.K. Nyborg. 1 9 9 3 .

339

Lenzmeier and Nyborg: Tax protein of HTLV-1 in adult T-cell leukemia Molecular mechanism of Tax Transactivation. M o l . C e l l . B i o l . 1 7 , 5156-5164.

and detection of antibodies to the antigen in human sera. P r o c . N a t l . A c a d . S c i . U S A 7 8 , 6476-6480.

Giles, R.H., J.G. Dauwerse, C. Higgins, F. Petrij, J.W. Wessels, G.C. Beverstock, H. Dohner, M. JotterandBellomo, and J.H. Falkenburg. 1 9 9 7 . Detection of CBP rearrangements in acute myelogenous leukemia with t(8;16). Leukemia 11, 2087-2096.

Hirai, H., T. Suzuki, J. Fujisawa, J. Inoue, and M. Yoshida. 1 9 9 4 . Tax protein of human T-cell leukemia virus type I binds to the ankyrin motifs of inhibitory factor kappa B and induces nuclear translocation of transcription factor NF-kappa B proteins for transcriptional activation. P r o c . N a t l . A c a d . S c i . U S A 9 1 , 3584-3588.

Giles, R.H., D.J. Peters, and M.H. Breuning. 1 9 9 8 . Conjunction dysfunction, CBP/p300 in human disease. T r e n d s . G e n e t . 1 4 , 178-183.

Horvai, A.E., L. Xu, E. Korzus, G. Brard, D. Kalafus, T.M. Mullen, D.W. Rose, M.G. Rosenfeld, and C.K. Glass. 1 9 9 7 . Nuclear integration of JAK/STAT and Ras/AP-1 signaling by CBP and p300. P r o c . N a t l . A c a d . S c i . U S A 9 4 , 1074-1079.

Gilmore, T.D., M. Koedood, K.A. Piffat, and D.W. White. 1 9 9 6 . Rel/NF-kappaB/IkappaB proteins and cancer. O n c o g e n e 1 3 , 1367-1378.

Ida, K., I. Kitabayashi, T. Taki, M. Taniwaki, K. Noro, K Yamamoto, M. Ohki, and Y. Hayashi.. 1 9 9 7 . Adenoviral E1A-associated protein p300 is involved in acute myeloid leukemia with t(11;22)(q23;q13). B l o o d 9 0 , 46994704.

Gonzalez, G.A., and M.R. Montminy. 1 9 8 9 . Cyclic AMP stimulates Somatostatin gene transcription by phosphorylation of CREB at Serine 133. C e l l 5 9 , 675680. Good, L., and S.C. Sun. 1 9 9 6 . Persistent activation of NFkappa B/Rel by human T-cell leukemia virus type 1 Tax involves degradation of I kappa B beta. J . V i r o l . 7 0 , 2730-2735.

Imhof, A., X.J. Yang, V.V. Ogryzko, Y. Nakatani, A.P. Wolffe, and H. Ge. 1 9 9 7 . Acetylation of general transcription factors by histone acetyltransferases. Curr. B i o l . 7 , 689-692.

Good, L., S.B. Maggirwar, and S.C. Sun. 1 9 9 6 . Activation of the IL-2 gene promoter by HTLV-I Tax involves induction of NF-AT complexes bound to the CD28-responsive element. EMBO J. 15, 3744-3750.

Inoue, J., M. Seiki, T. Taniguchi, S. Tsuru, and M. Yoshida. 1 9 8 6 . Induction of interleukin 2 receptor gene expression by p40 x encoded by human T-cell leukemia virus type I. EMBO J. 5, 2883-2888.

Goren, I., O.J. Semmes, K.-T. Jeang, and K. Moelling. 1 9 9 5 . The amino terminus of Tax is required for interaction with the cyclic AMP response element binding protein. J . V i r o l . 6 9 , 5806-5811.

Jacobson, S., C.S. Raine, E.S. Mingioli, and D.E. McFarlin. 1 9 8 8 . Isolation of an HTLV-I-like retrovirus from patients with tropical spastic paraparesis. N a t u r e 3 3 1 , 540-543.

Gottesfeld, J.M., D.L. Johnson, and J.K. Nyborg. 1 9 9 6 . Transcriptional activation of RNA polymerase IIIdependent genes by the human T-cell leukemia virus type 1 Tax protein. M o l . C e l l . B i o l . 1 6 , 1777-1785.

Janknecht, R., and T. Hunter. 1 9 9 6 . Transcriptional control, Versatile molecular glue. C u r r . B i o l . 6 , 951-954. Jeang, K.-T., I. Boros, J. Brady, M. Radonovich, and G. Khoury. 1 9 8 8 . Characterization of cellular factors that interact with the human T-cell leukemia virus type I p40 x -

Gottesfeld, J.M., L. Neely, J.W. Trauger, E.E. Baird, and P.B. Dervan. 1 9 9 7 . Regulation of gene expression by small molecules. Nature 387, 202-205.

responsive 21-base-pair sequence. J . V i r o l . 6 2 , 44994509.

Grassmann, R., S. Berchtold, I. Radant, M. Alt, B. Fleckenstein, J.G. Sodroski, W.A. Haseltine and U. Ramstedt. 1 9 9 2 . Role of human T-cell leukemia virus type I X region proteins in immortalization of primary human lymphocytes in culture. J . V i r o l . 6 6 , 45704575.

Jeang, K.-T., S.G. Widen, O.J. Semmes, and S.H. Wilson. 1 9 9 0 . HTLV-I Trans-Activator Protein, Tax, is a TransRepressor of the Human #-Polymerase Gene. S c i e n c e 2 4 7 , 1082-1084. Jin, D.Y., and K.T. Jeang. 1 9 9 7 a . HTLV-I Tax selfassociation in optimal trans-activation function. N u c l e i c . A c i d s . R e s . 2 5 , 379-388.

Grossman, W.J., J.T. Kimata, F.H. Wong, M. Zutter, T.J. Ley, and L. Ratner. 1 9 9 5 . Development of leukemia in mice transgenic for the Tax gene of human T-cell leukemia virus type I. P r o c . N a t l . A c a d . S c i . U S A 9 2 , 1057-1061.

Jin, D.Y., and K.T. Jeang. 1 9 9 7 b . Transcriptional activation and self-association in yeast, protein-protein dimerization as a pleiotropic mechanism of HTLV-I Tax function. L e u k e m i a 1 1 S u p p l . 3, 3-6.

Harrod, R., Y. Tang, C. Nicot, H.S. Lu, A. Vassilev, Y. Nakatani, and C.Z. Giam. 1 9 9 8 . An exposed KID-like domain in human T-cell lymphotropic virus type 1 Tax is responsible for the recruitment of coactivators CBP/p300. M o l . C e l l . B i o l . 1 8 , 5052-5061.

Jin, D.Y., F. Spencer, and K.T. Jeang. 1 9 9 8 . Human T cell leukemia virus type 1 oncoprotein Tax targets the human mitotic checkpoint protein MAD1. C e l l 9 3 , 81-91.

Hinuma, Y., K. Nagata, M. Hanaoka, M. Nakai, T. Matsumoto, K.I. Kinoshita, S. Shirakawa, and I. Miyoshi. 1 9 8 1 . Adult T-cell leukemia, antigen in an ATL cell line

Josephs, S.F., F. Wong-Staal, V. Manzari, R.C. Gallo, J.G. Sodroski, M..D Trus, D. Perkins, R. Patarca, and W.A. Haseltine. 1 9 8 4 . Long terminal repeat structure of an

340

Gene Therapy and Molecular Biology Vol 3, page 341 American isolate of type I human T-cell leukemia virus. V i r o l o g y 1 3 9 , 340-345.

Laurance, M.E., R.P. Kwok, M.S. Huang, J.P. Richards, J.R. Lundblad, R.H. Goodman. 1 9 9 7 . Differential activation of viral and cellular promoters by human T-cell lymphotropic virus-1 Tax and cAMP-responsive element modulator isoforms. J . B i o l . C h e m . 2 7 2 , 2646-2651.

Kadison, P., H.T. Poteat, K.M. Klein and D.V. Faller. 1 9 9 0 . Role of protein kinase A in Tax transactivation of the human T-cell leukemia virus type I long terminal repeat. J . V i r o l . 6 4 , 2141-2148.

Lemasson, I., V. Robert-Hebmann, S. Hamaia, M. Duc Dodon, L. Gazzolo, and C. Devaux. 1 9 9 7 . Transrepression of lck gene expression by human T-cell leukemia virus type 1encoded p40Tax. J . V i r o l . 7 1 , 1975-1983.

Kamei, Y., L. Xu, T. Heinzel, J. Torchia, R. Kurokawa, B. Gloss, S.C. Lin, R.A. Heyman, D.W. Rose, C.K. Glass, M.G. Rosenfeld. 1 9 9 6 . A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. C e l l 8 5 , 403-414.

Lemasson, I., S. Thebault, C. Sardet, C. Devaux, and J.M. Mesnard. 1998. Activation of E2F-mediated Transcription by Human T-cell Leukemia Virus Type I Tax Protein in a p16(INK4A)-negative T-cell Line. J . B i o l . C h e m . 2 7 3 , 23598-23604.

Kanno, T., K. Brown, G. Franzoso, and U. Siebenlist. 1 9 9 4 . Kinetic analysis of human T-cell leukemia virus type I Tax-mediated activation of NF-kappa B. M o l . C e l l . B i o l . 1 4 , 6443-6451. Kawano, F., K. Yamaguchi, H. Nishimura, H. Tsuda, and K. Takatsuki. 1 9 8 5 . Variation in the clinical courses of adult T-cell leukemia. Cancer 55, 851-856.

Lenzmeier, B.A., H.A. Giebler, and J.K. Nyborg. 1 9 9 8 . Human T-cell leukemia virus type 1 Tax requires direct access to DNA for recruitment of CREB binding protein to the viral promoter. M o l . C e l l . B i o l . 1 8 , 721-731.

Kee, B.L., J. Arias, M.R. Montminy. 1 9 9 6 . Adaptormediated recruitment of RNA polymerase II to a signaldependent activator. J . B i o l . C h e m . 2 7 1 , 2373-2375.

Leung, K., and G.J. Nabel. 1 9 8 8 . HTLV-I transactivator induces interleukin-2 receptor expression through an NF!B-like factor. Nature 333, 76-778.

Kielkopf, C.L., S. White, J.W. Szewczyk, J.M. Turner, E.E. Baird, P.B. Dervan, and D.C. Rees. 1 9 9 8 . A structural Basis for Recognition of AT and TA Base Pairs in the Minor Groove of B-DNA. S c i e n c e 2 8 2 , 111-115.

Lilienbaum, A., M. Duc Dodon, C. Alexandre, L. Gazzolo, and D. Paulin. 1 9 9 0 . Effect of human T-cell leukemia virus type I Tax proteinon activation of the human vimentin gene. J . V i r o l . 6 4 , 256-263.

Kim, S.J., J.H. Kehrl, J. Burton, C.L. Tendler, K.T. Jeang, D. Danielpour, C. Thevenin, K.Y. Kim, M.B. Sporn, and A.B. Roberts. 1 9 9 0 . Transactivation of the transforming growth factor beta 1 (TGF-1) gene by human Tlymphotropic virus type I Tax, a potential mechanism for the increased production of TGF-1 in adult T-cell leukemia. J . E x p . M e d . 1 7 2 , 121-129.

Lilienbaum, A., and D. Paulin. 1 9 9 3 . Activation of the human vimentin gene by the Tax human T-cell leukemia virus. I. Mechanisms of regulation by the NF-kappa B transcription factor. J . B i o l . C h e m . 2 6 8 , 2180-2188.

Kimzey, A.L., and W.S. Dynan. 1 9 9 8 . Specific regions of contact between human T-cell leukemia virus type I Tax protein and DNA identified by photocross-linking. J . B i o l . C h e m . 2 7 3 , 13768-13775.

Low, K.G., L.F. Dorner, D.B. Fernando, J. Grossman, KT Jeang, and MJ Comb. 1 9 9 7 . Human T-cell leukemia virus type 1 Tax releases cell cycle arrest induced by p16INK4a. J . V i r o l . 7 1 , 1956-1962

Kondo, T., H. Kono, H. Nonaka, N. Miyamoto, R. Yoshida, F. Bando, H. Inoue, I. Miyoshi, Y. Hinuma, and M. Hanaoka. 1 9 8 7 . Risk of adult T-cell leukaemia/lymphoma in HTLV-I carriers. Lancet 2 , 159.

Lundblad, J.R., R.P.S. Kwok, M.E. Laurance, M.S. Huang, J.P. Richards, R.G. Brennan, and R.H. Goodman. 1 9 9 8 . The human T-cell leukemia virus-1 transcriptional activator Tax enhances cAMP-responsive elementbinding protein (CREB) binding activity through interactions with the DNA minor groove. J . B i o l . C h e m . 2 7 3 , 19251-19259.

Lill, N.L., S.R. Grossman, D. Ginsberg, J. DeCaprio, and D.M. Livingston. 1 9 9 7 . Binding and modulation of p53 by p300/CBP coactivators. Nature 387, 823-827.

Kwok, R.P.S., J.R. Lundblad, J.C. Chrivia, J.P. Richards, H.P. B채chinger, R.G. Brennan, S.G.E. Roberts, M.R. Green, and R.H. Goodman. 1 9 9 4 . Nuclear protein CBP is a coactivator for the transcription factor CREB. Nature 3 7 0 , 223-226.

Maruyama, M., H. Shibuya, H. Haraa, M. Hatakeyama, M. Seiki, T. Fujita, J. Inoue, M. Yoshida, and T. Taniguchi. 1 9 8 7 . Evidence for aberrant activation of the interleukin2 autocrine loop by HTLV-I encoded p40 x and T3/Ti

Kwok, R.P.S., M.E. Laurance, J.R. Lundblad, P.S. Goldman, H.-M. Shih, L.M. Connor, S.J. Marriott, and R.H. Goodman. 1 9 9 6 . Control of cAMP-regulated enhancers by the viral transactivator Tax through CREB and the coactivator CBP. Nature 380, 642-646.

complex triggering. C e l l 4 8 , 343-350. McKinsey, T.A., J.A. Brockman, D.C. Scherer, S.W. AlMurrani, P.L. Green, and D.W. Ballard. 1 9 9 6 . Inactivation of IkappaB beta by the Tax protein of human T-cell leukemia virus type 1, a potential mechanism for constitutive induction of NF-kappaB. M o l . C e l l . B i o l . 1 6, 2083-2090.

Lacoste, J., L. Petropoulos, N. Pepin, and J. Hiscott. 1 9 9 5 . Constitutive phosphorylation and turnover of I kappa B alpha in human T-cell leukemia virus type I-infected and Tax-expressing T cells. J . V i r o l . 6 9 , 564-569.

341

Lenzmeier and Nyborg: Tax protein of HTLV-1 in adult T-cell leukemia Mercurio, F., H. Zhu, B.W. Murray, A. Shevchenko, B.L. Bennett, J. Li, D.B. Young, .M Barbosa, M. Mann, A. Manning, and A. Rao. 1 9 9 7 . IKK-1 and IKK-2, cytokineactivated IkappaB kinases essential for NF-kappaB activation. S c i e n c e 2 7 8 , 860-866.

p300 and CBP are histone acetyltransferases. C e l l 8 7 , 953-959. Osame, M., K. Usuku, S. Izumo, N. Ijichi, H. Amitani, A. Igata, M. Matsumoto, and M. Tara. 1 9 8 6 . HTLV-1 associated myelopathy, a new clinical entity. L a n c e t 1 , 1031-1032.

Montagne, J., C. BĂŠraud, I. Crenon, G. Lombard-Platet, L. Gazzolo, A. Sergeant, and P. Jalinot. 1 9 9 0 . Tax1 Induction of the HTLV-I 21 bp Enhancer Requires Cooperation Between Two Cellular DNA-Binding Proteins. EMBO J. 9, 957-964.

Paca-Uccaralertkun, S., .L.-J. Zhao, N. Adya, J.V. Cross, B.R. Cullen, I. Boros, and C.-Z. Giam. 1 9 9 4 . In vitro selection of DNA elements highly responsive to the Human T-Cell Lymphotropic Virus Type I transcriptional activator, Tax. M o l . C e l l . B i o l . 1 4 , 456-462.

Montminy, M. 1 9 9 7 . Transcriptional regulation by cyclic AMP. A n n u . R e v . B i o c h e m . 6 6 , 807-822. Mori, N., F. Kashanchi, D. Prager. 1 9 9 7 . Repression of transcription from the human T-cell leukemia virus type I long terminal repeat and cellular gene promoters by wildtype p53. B l o o d 9 0 , 4924-4932.

Paskalis, H., B.K. Felber, and G.N. Pavlakis. 1 9 8 6 . Cisacting sequences responsible for the transcriptional activation of human T-cell leukemia virus type I constitute a conditional enhancer. P r o c . N a t l . A c a d . S c i . 8 3 , 6558-6562.

Murphy, E.L., B. Hanchard, J.P. Figueroa, W.N. Gibb, W.S. Lofters, M. Campbell, J.J. Goedert, and WA Blattner. 1 9 8 9 . Modelling the risk of ATL in persons infected with HTLV-1. I n t . J . C a n c e r 4 3 , 250-253

Pepin, N., A. Roulston, J. Lacoste, R. Lin, and J. Hiscott. 1 9 9 4 . Subcellular redistribution of HTLV-1 Tax protein by NF-kappa B/Rel transcription factors. V i r o l o g y 2 0 4 , 706-716.

Nakamura, M., M. Niki, K. Ohtani, and K. Sugamura. 1 9 8 9 . Differential activation of the 21 base pair enhancer element of human T-cell leukemia virus type I by its own transactivator and cyclic AMP. N u c l . A c i d s R e s . 1 7 , 5207-5221.

Perini, G., S. Wagner, and M.R. Green. 1 9 9 5 . Recognition of bZIP proteins by the human T-cell leukemia virus transactivator Tax. Nature 376, 602-605. Perkins, N.D., L.K. Felzien, J.C. Betts, K. Leung, D.H. Beach, and G.J. Nabel. 1 9 9 7 . Regulation of NF-kappaB by cyclin-dependent kinases associated with the p300 coactivator. S c i e n c e 2 7 5 , 523-527.

Nerenberg, M., S. Hinrichs, R. Reynolds, G. Khoury, and G. Jay. 1 9 8 7 . The tat gene of human T-lymphotropic virus type I induces mesenchymal tumors in transgenic mice. S c i e n c e 2 3 7 , 1324-1329.

Petrij, F., R.H. Giles, H.G. Dauwerse, J.J .Saris, R.C.M. Hennekam, M. Masuno, N. Tommerup, G.-J.B. van Ommen, R.H. Goodman, D.J.M. Peters, and M.H. Breuning. 1 9 9 5 . Rubinstein-Taybi Syndrome caused by mutations in the transcriptional co-activator CBP. Nature 376, 348-351.

Neuveut, C., K.G. Low, F. Maldarelli, I. Schmitt, F. Majone, R. Grassmann, and K.T. Jeang. 1 9 9 8 . Human T-cell leukemia virus type 1 Tax and cell cycle progression, role of cyclin D-cdk and p110Rb. M o l . C e l l . B i o l . 1 8 , 3620-3632.

Pise-Masison, C.A., J. Dittmer, K.E. Clemens, and J.N. Brady. 1 9 9 7 . Physical and functional interaction between the human T-cell lymphotropic virus type 1 Tax1 protein and the CCAAT binding protein NF-Y. M o l . C e l l . B i o l . 1 7 , 1236-1243.

Nimer, S.D., J.C. Gasson, K. Hu, I. Smalberg, J.L. Williams, I.S.Y. Chen, and J.D. Rosenblatt. 1 9 8 9 . Activation of the GM-CSF promoter by HTLV-I and -II Tax proteins. O n c o g e n e 4 , 671-676. Numata, N., K. Ohtani, M. Niki, M. Nakamura, and K. Sugamura. 1 9 9 1 . Synergism between two distinct elements of the HTLV-I enhancer during activation by the trans-activator of HTLV-I. N e w B i o l . 3 , 896-906.

Pise-Masison, C.A., K.S. Choi, M. Radonovich, J. Dittmer, S.J. Kim, and J.N. Brady. 1 9 9 8 a . Inhibition of p53 transactivation function by the human T-cell lymphotropic virus type 1 Tax protein. J . V i r o l . 7 2 , 1165-1170.

Nyborg, J.K., W.S. Dynan, I.S.Y. Chen, and W.A. Wachsman. 1 9 8 8 . Binding of host-cell factor to DNA sequences in the long terminal repeat of T-cell leukemia virus type I, implications for viral gene expression. P r o c . N a t l . A c a d . S c i . 8 5 , 1457-1461.

Pise-Masison, C.A., M. Radonovich, K. Sakaguchi, Appella, and J.N. Brady. 1 9 9 8 b . Phosphorylation p53, a novel pathway for p53 inactivation in human cell lymphotropic virus type 1-transformed cells. V i r o l . 7 2 , 6348-6355.

Nyborg, J.K., and W.S. Dynan. 1 9 9 0 . Interaction of cellular proteins with the Human T-cell Leukemia Virus Type I transcriptional control region, Purification of cellular proteins that bind the 21-base pair repeat elements. J . B i o l . C h e m . 2 6 5 , 8230-8236.

E. of TJ.

Poiesz, B.J., F.W. Ruscetti, A.F. Gazdar, P.A. Bunn, J.D. Minna, and R.C. Gallo. 1 9 8 0 . Detection and isolation of type C retrovirus particles from fresh and cultured lymphocytes of a patient with cutaneous T-cell lymphoma. P r o c . N a t l . A c a d . S c i . U S A 7 7 , 74157419.

Ogryzko, V.V., R.L. Schiltz, V. Russanove, B.H. Howard, and Y. Nakatani. 1 9 9 6 . The transcriptional coactivators

342

Gene Therapy and Molecular Biology Vol 3, page 343 Poiesz, B.J., F.W. Ruscetti, M.S. Reitz, V.S. Kalyanaraman, and R.C. Gallo. 1 9 8 1 . Isolation of a new type-C retrovirus (HTLV) in primary uncultured cells of a patient with Sezary T-cell leukemia and evidence for virus nucleic acids and antigens in fresh leukemic cells. N a t u r e 2 9 4 , 268-271.

treatment-related hematologic disorders. B l o o d 9 0 , 535541. Satake, N., Y. Ishida, Y. Otoh, S. Hinohara, H. Kobayashi, A. Sakashita, N. Maseki, and Y. Kaneko. 1 9 9 7 . G enes C h r o m o s o m e s C a n c e r 2 0 , 60-63. Seeler, J.S., C. Muchardt, M. Podar, and R.B. Gaynor. 1 9 9 3 . Regulatory elements involved in Tax-mediated transactivation of the HTLV-I LTR. V i r o l o g y 1 9 6 , 442-450.

Popovic, M., M.S. Reitz Jr, M.G. Sarngadharan, M. RobertGuroff, V.S. Kalyanaraman, Y. Nakao, I. Miyoshi, J. Minowada, M. Yoshida, Y. Ito, and R.C. Gallo. 1 9 8 2 . The virus of Japanese adult T-cell leukaemia is a member of the human T-cell leukaemia virus group. N a t u r e 3 0 0 , 63-66.

Seiki, M., S. Hattori, and M. Yoshida. 1 9 8 2 . Human adult Tcell leukemia virus, molecular cloning of the provirus DNA and the unique terminal structure. P r o c . N a t l . A c a d . S c i . U S A 7 9 , 6899-6902.

Poteat, H.T., P. Kadison, K. McGuire, L. Park, J. Sodroski, and W. Haseltine. 1 9 8 9 . Response of the Human T-cell Leukemia Virus Type I long terminal repeat to cyclic AMP. J . V i r o l . 6 3 , 1604-1611.

Seiki, M., R. Eddy, T.B. Shows, and M. Yoshida. 1 9 8 4 . Nonspecific integration of the HTLV provirus genome into adult T-cell leukaemia cells. Nature 309, 640-642.

Poteat, H.T., F.-Y. Chen, P. Kaidson, J.G. Sodroski, and W. Haseltine. 1 9 9 0 . Protein kinase A-dependent binding of a nuclear factor of the 21-base-pair repeat of the Human Tcell Leukemia Virus Type I long terminal repeat. J . V i r o l . 6 4 , 1264-1270.

Semmes, O.J., J.F. Barret, C.V. Dang, and K.T. Jeang. 1 9 9 6 . Human T-cell leukemia virus type I Tax masks c-Myc function through a cAMP-dependent pathway. J . B i o l . C h e m . 2 7 1 , 9730-9738. Sherr, C.J., and J.M. Roberts. 1 9 9 5 . Inhibitors of mammalian G1 cyclin-dependent kinases. G e n e s . D e v . 9 , 1149-1163.

Radhakrishnan, I., G.C. Perez-Alvarado, D. Parker, H.J. Dyson, M.R. Montminy, and P.E. Wright. 1 9 9 7 . Solution structure of the KIX domain of CBP bound to the transactivation domain of CREB, a model for activator, coactivator interactions.C e l l 9 1 , 741-52.

Shi, Y., and C. Mello. 1 9 9 8 . A CBP/p300 homolog specifies multiple differentiation pathways in Caenorhabditis elegans. G e n e s . D e v . 1 2 , 943-955.

Ressler, S., G.F. Morris, and S.J. Marriott. 1 9 9 7 . Human Tcell leukemia virus type 1 Tax transactivates the human proliferating cell nuclear antigen promoter. J . V i r o l . 7 1 , 1181-1190.

Shikama, N., J. Lyon and N.B. La Thangue. 1 9 9 7 . The p300/CBP family, Integrating signals with transcription factors and chromatin. T r e n d s i n C e l l B i o l . 7 , 230236.

Rivera, I., E.W. Harhaj, and S.C. Sun. 1 9 9 8 . Involvement of NF-AT in type I human T-cell leukemia virus Tax-mediated fas ligand promoter transactivation. J . B i o l . C h e m . 2 7 3 , 22382-22388.

Shimotohno, K., M. Takano, T. Teruuchi, and M. Miwa. 1 9 8 6 . Requirement of multiple copies of a 21-nucleotide sequence in the U3 region of human T-cell leukemia virus type I and type II long terminal repeats for trans-acting activation of transcription. P r o c . N a t l . Acad. S c i . U S A 8 3 , 8112-8116.

Robert-Guroff, M., Y. Nakao, K. Notake, Y. Ito, A. Sliski, and R.C. Gallo. 1 9 8 2 . Natural antibodies to human retrovirus HTLV in a cluster of Japanese patients with Adult T Cell Leukemia. S c i e n c e 2 1 5 , 975-978.

Siekevitz, M., M.B. Feinberg, N. Holbrook, F. Wong-Staal, and W.C. Green. 1 9 8 7 . Activation of interleukin 2 and interleukin 2 receptor (Tac) promoter expression by the trans-activator (tat) gene product of human T-cell leukemia virus, type I. P r o c . N a t l . A c a d . S c i . U S A 8 4 , 53895393.

Rosen, C.A., J.G. Sodroski, and W.A. Haseltine. 1 9 8 5 . Location of cis-acting regulatory sequences in the human T-cell leukemia virus type I long terminal repeat. P r o c . N a t l . A c a d . S c i . U S A 8 2 , 6502-6506. Rosen, C.A., R. Park, J.G. Sodroski, and W.A. Haseltine. 1 9 8 7 . Multiple sequence elements are required for regulation of human T-cell leukemia virus gene expression. P r o c . N a t l . A c a d . S c i . U S A 8 4 , 49194923.

Slamon, D.J., M.F. Press, L.M. Souza, D.C. Murdock, M.J. Cline, D.W. Golde, J.C. Gasson, and I.S.Y. Chen. 1 9 8 5 . Studies of the putative transforming protein of the Type I Human T-cell Leukemia Virus. S c i e n c e 2 2 8 , 14271430.

Rousset, R., C. Desbois, F. Bantignies, and P. Jalinot. 1 9 9 6 . Effects on NF-kappa B1/p105 processing of the interaction between the HTLV-1 transactivator Tax and the proteasome. Nature 381, 328-331.

Sobulo, O.M., J. Borrow, R. Tomek, S. Reshmi, A. Harden, B. Schlegelberger, D. Housman, N.A. Doggett, J.D. Rowley, and N.J. Zeleznik-Le. 1 9 9 7 . MLL is fused to CBP, a histone acetyltransferase, in therapy-related acute myeloid leukemia with a t(11;16)(q23;p13.3). P r o c . N a t l . A c a d . S c i . U S A 9 4 , 8732-8737.

Rowley, J.D., S. Reshmi, O. Sobulo, T. Musvee, J. Anastasi, S. Raimondi, N.R. Schneider, J.C. Barredo, E.S. Cantu, B. Schlegelberger, F. Behm, N.A. Doggett, J. Borrow, and L.N. Zeleznik. 1 9 9 7 . All patients with the T(11;16)(q23;p13.3) that involves MLL and CBP have

Sodroski, J.G., C.A. Rosen, and W.A. Haseltine. 1 9 8 4 . Trans-acting Transcriptional Activation of the Long

343

Lenzmeier and Nyborg: Tax protein of HTLV-1 in adult T-cell leukemia Terminal Repeat of Human T-Cell Lymphotropic Viruses in Infected Cells. S c i e n c e 2 2 5 , 381-385.

IkappaB kinase participate in human T-cell leukemia virus I Tax-mediated NF-kappaB activation. J . B i o l . C h e m . 2 7 3 , 21132-21136.

Sodroski, J., C.A. Rosen, W.C. Goh, and W.A. Haseltine. 1 9 8 5 . A transcriptional activator protein encoded by the x-lor region of the human T-cell leukemia virus. S c i e n c e 2 2 8 , 1430-1434.

Uittenbogaard, M.N., A.P. Armstrong, A. Chiaramello, and J.K. Nyborg. 1 9 9 4 . HTLV-I Tax Protein Represses Gene Expression Through the bHLH Family of Transcription Factors. J . B i o l . C h e m . 2 6 9 , 22466-22469.

Suzuki, T., J.-I. Fujisawa, M. Toita, and M. Yoshida. 1 9 9 3 . The Transactivator Tax of Human T-Cell Luekemia Virus type I (HTLV-I) Interacts with cAMP-Response element (CRE) Binding and CRE Modulator Proteins that Bind to the 21-base pair Enhancer of HTLV-I. P r o c . N a t l . A c a d . S c i . U S A 9 0 , 610-614.

Uittenbogaard, M.N., H.A. Giebler, D. Reisman, and J.K. Nyborg. 1 9 9 5 . Transcriptional repression of p53 by human T-cell leukemia virus type I Tax protein. J . B i o l . C h e m . 2 7 0 , 28503-28506. VanOrden, K., J.P. Yan, A. Ullola, and J.K. Nyborg. 1 9 9 8 . Binding of the Human T-Cell Leukemia Virus Tax protein to the coactivator CBP interferes with CBP-mediated transcriptional control. Submitted for publication.

Suzuki, T., H. Hirai, T. Murakami, and M. Yoshida. 1 9 9 5 . Tax protein of HTLV-1 destabilizes the complexes of NFkappa B and I kappa B-alpha and induces nuclear translocation of NF-kappa B for transcriptional activation. O n c o g e n e 1 0 , 1199-1207.

Wagner, S., and M.R. Green. 1 9 9 3 . HTLV-I Tax Protein Stimulation of DNA Binding of bZIP Proteins by Enhancing Dimerization. S c i e n c e 2 6 2 , 395-399.

Suzuki, T., S. Kitao, H. Matsushime, and M. Yoshida. 1 9 9 6 . HTLV-1 Tax protein interacts with cyclin-dependent kinase inhibitor p16INK4A and counteracts its inhibitory activity towards CDK4. EMBO J. 15, 1607-1614.

Wang, C.Y., M.W. Mayo, and A.S. Baldwin Jr. 1 9 9 6 . TNFand cancer therapy-induced apoptosis, potentiation by inhibition of NF-kappaB. S c i e n c e 2 7 4 , 784-787.

Swope, D.L., C.L. Mueller, and J.C. Chrivia. 1 9 9 6 . CREBbinding protein activates transcription through multiple domains. J . B i o l . C h e m . 2 7 1 , 28138-28145.

Watanabe, T. 1 9 9 7 . HTLV-1-associated diseases. I n t . J . H e m a t o l . 6 6 , 257-278. White, S., J.W. Szewczyk, J.M. Turner, E.E. Baird, and P.B. Dervan. 1 9 9 8 . Recognition of the four Watson-Crick base pairs in the DNA minor groove by synthetic ligands. Nature 391, 468-471.

Taki, T., M. Sake, M. Tsuchida, and Y. Hayashi. 1 9 9 7 . The t(11;16)(q23;p13) translocation in myelodysplastic syndrome fuses the MLL gene to the CBP gene. B l o o d 8 9 , 3945-3950.

Wu, X., C. Spiro, W.G. Owen, C.T. McMurray. 1 9 9 8 . cAMP response element-binding protein monomers cooperatively assemble to form dimers on DNA. J . B i o l . C h e m . 2 7 3 , 20820-20827.

Tan, T.-H., M. Horikoshi, and R.G. Roeder. 1 9 8 9 . Purification and characterization of multiple nuclear factors that bind to the TAX-inducible enhancer within the Human T-cell Leukemia Virus Type 1 long terminal repeat. M o l . C e l l B i o l . 9 , 1733-1745. Thanos, D., and T. Maniatis. 1 9 9 5 . NF-kappa B, a lesson in family values. C e l l 8 0 , 529-532.

Yan, J.-P., J.E. Garrus, H.A. Giebler, L.A. Stargell, and J.K. Nyborg. 1 9 9 8 . Molecular interactions between the coactivator CBP and the human T-cell leukemia virus Tax protein. J . M o l . B i o l . 2 8 1 , 395-400.

Tie, F., N. Adya, W.C. Greene, and C.Z. Giam. 1 9 9 6 . Interaction of the human T-lymphotropic virus type 1 Tax dimer with CREB and the viral 21-base-pair repeat. J . V i r o l . 7 0 , 8368-8374.

Yang, X.J., V.V. Ogryzko, J. Nishikawa, B.H. Howard, and Y. Nakatani. 1 9 9 6 . A p300/CBP-associated factor that competes with the adenoviral oncoprotein E1A. Nature 3 8 2 , 319-324.

Trauger, J.W., E.E. Baird, and P.B. Dervan. 1 9 9 6 . Recognition of DNA by designed ligands at subnanomolar concentrations. Nature 382, 559-561.

Yao, T.P., S.P. Oh, M. Fuchs, N.D. Zhou, L.E. Ch'ng, D. Newsome, R.T. Bronson, E. Li, D.M. Livingston, R. Eckner. 1 9 9 8 . Gene dosage-dependent embryonic development and proliferation defects in mice lacking the transcriptional integrator p300. C e l l 9 3 , 361-372.

Tsujimoto, A., H. Nyunoya, T. Morita, T..Sato, and K. Shimotohno. 1 9 9 1 . Isolation of cDNAs for DNA-binding proteins which specifically bind to a Tax-responsive enhancer element in the long terminal repeat of Human Tcell Leukemia Virus Type I. J . V i r o l . 6 5 , 1420-1426.

Yin , M.-J., E.J. Paulssen, J.S. Seeler, and R.B. Gaynor. 1 9 9 5 a . Protein domains involved in both in vivo and in vitro interactions between human T-cell leukemia virus type I Tax and CREB. J . V i r o l . 6 9 , 3420-3432.

Uchiumi, F., K. Semba, Y. Yamanashi, J. Fujisawa, M. Yoshida, K. Inoue, K. Toyoshima, and T. Yamamoto. 1 9 9 2 . Characterization of the promoter region of the src family gene lyn and its trans activation by human T-cell leukemia virus type I-encoded p40Tax. M o l . C e l l . B i o l . 1 2 , 3784-3795.

Yin, M.-J., E.J. Paulssen, J.S. Seeler, and R.B. Gaynor. 1 9 9 5 b . Chimeric proteins composed of Jun and CREB define domains required for interaction with the Human Tcell Leukemia Virus Type 1 Tax protein. J . V i r o l . 6 9 , 6209-6218.

Uhlik, M., L. Good, G. Xiao, E.W. Harhaj, E. Zandi, M. Karin, and S.C. Sun. 1 9 9 8 . NF-kappaB-inducing kinase and

Yin, M.-J., and R.B. Gaynor. 1 9 9 6 a . Complex formation between CREB and Tax enhances the binding affinity of

344

Gene Therapy and Molecular Biology Vol 3, page 345 CREB for the human T-cell leukemia virus type 1 21-basepair repeats. M o l . C e l l . B i o l . 1 6 , 3156-3168. Yin, M.-J., and R.B. Gaynor. 1 9 9 6 b . HTLV-1 21 bp repeat sequences facilitate stable association between Tax and CREB to increase CREB binding affinity. J . M o l . B i o l . 2 6 4 , 20-31. Yin, M.-J., L.B. Christerson, Y. Yamamoto, Y.T. Kwak, S. Xu, F. Mercurio, M. Barbosa, M.H. Cobb, R.B. Gaynor. 1 9 9 8 . HTLV-I Tax protein binds to MEKK1 to stimulate IkappaB kinase activity and NF-kappaB activation. C e l l 9 3 , 875-884. Yoshida, M., I. Miyoshi, and Y. Hinuma. 1 9 8 2 . Isolation and characterization of retrovirus from cell lines of human adult T-cell leukemia and its implication in the disease. P r o c . N a t l . A c a d . S c i . U S A 7 9 , 2031-2035. Yoshida, M., M. Seiki, K. Yamaguchi, and K. Takatsuki. 1 9 8 4 . Monoclonal integration of human T-cell leukemia provirus in all primary tumors of adult T-cell leukemia suggests causative role of human T-cell leukemia virus in the disease. P r o c . N a t l . A c a d . S c i . U S A 8 1 , 25342537. Yoshimura, T., J.-I. Fugisawa, and M. Yoshida. 1 9 9 0 . Multiple cDNA clones encoding nuclear proteins that bind to the Tax-dependent enhancer of HTLV-1, All contain a leucine zipper structure and basic amino acid domain. EMBO J. 9, 2537-2542. Zandi, E., D.M. Rothwarf, M. Delhase, M. Hayakawa, and M. Karin. 1 9 9 7 . The IkappaB kinase complex (IKK) contains two kinase subunits, IKKalpha and IKKbeta, necessary for IkappaB phosphorylation and NF-kappaB activation. C e l l 9 1 , 243-252. Zhao, L.-J., and C.-Z. Giam. 1 9 9 1 . Interaction of the Human T-cell Lymphotrophic Virus Type I (HTLV-I) transcriptional activator Tax with cellular factors that bind specifically to the 21-base-pair repeats in the HTLV-I enhancer. P r o c . N a t l . A c a d . S c i . 8 8 , 11445-11449. Zhao, L.-J., and C.-Z. Giam. 1 9 9 2 . Human T-cell lymphotropic virus type I (HTLV-1) transcriptional activator, Tax, enhances CREB Binding to HTLV-1 21base-pair repeats by protein-protein interaction. P r o c . N a t l . A c a d . S c i . U S A 8 9 , 7070-7074. Zhong, H., R.E. Voll, and S. Ghosh. 1 9 9 8 . Phosphorylation of NF-kappaB p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300. M o l . C e l l 1 , 661-671. Zimmer, C., and U. Wahnert. 1 9 8 6 . Nonintercalating DNABinding Ligands, Specificity of the Interaction and Their Use as Tools in Biophysical, Biochemical, and Biological Investigations of the Genetic Material. P r o g . B i o p h y s . M o l e c . B i o l . 4 7 , 31-112.

345

Gene Therapy and Molecular Biology Vol 3, page 347 Gene Ther Mol Biol Vol 3, 347-354. August 1999.

What does acetylcholinesterase do in hematopoietic cells? Review Article

Roxanne Y.Y. Chan and Bernard J. Jasmin* Department of Cellular and Molecular Medicine, Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada K1H 8M5 __________________________________________________________________________________________________ * C o r r e s p o n d e n c e : Bernard.J. Jasmin, Ph.D., Department of Cellular and Molecular Medicine, Faculty of Medicine, University of Ottawa, 451 Smyth Road, Ottawa, Ontario, Canada K1H 8M5. Tel: (613) 562-5800 ext: 8383; Fax: (613) 562-5434; E-mail: bjasmin@danis.med.uottawa.ca Key words: acetylcholinesterase, cholinergic synapses, hematopoietic cells, tumor suppressor, myeloid leukaemia, apoptosis, proliferative disorders A b b r e v i a t i o n s : AchE, acetylcholinesterase; GPI , glycosyl-phosphatidylinositol; PI-PLC, phosphatidylinositol-specific phospholipase C Received: 19 November 1998; accepted: 10 December 1998

Summary Acetylcholinesterase (AChE) is an essential component of cholinergic synapses since it hydrolyzes acetylcholine released from presynaptic nerve terminals. However, it is well documented that AChE is also expressed in a variety of non-cholinergic tissues including hematopoietic cells. Despite the recent progress made in our understanding of the molecular mechanisms regulating expression of AChE, our knowledge o f the precise function o f this enzyme i n hematopoietic c e l l s s t i l l remains limited. Previous work has led to the notion that AChE may be involved in myelodysplastic syndromes as well as acute myeloid leukaemias since it may regulate hematopoiesis by acting as a tumor suppressor gene. In addition, recent studies have further demonstrated the involvement o f AChE i n the proliferation of multipotent stem cells, as well as in the mechanisms leading to apoptosis in cells undergoing erythroid and megakaryocytic differentiation. In this review, we first present an overview of the cellular and molecular biology of AChE and then, focus more specifically on the expression of AChE in hematopoietic cells. Finally, we also discuss the recent evidence linking AChE expression and the proliferative capacity of these cells. A better understanding of the functional significance of AChE in hematopoietic cells may be relevant for the future design of novel therapeutic strategies against proliferative disorders of hematopoietic tissues.

hippocampus and cerebellum express large amounts of AChE (see for example Landwehrmeyer et al., 1993; Legay et al., 1993a; Hammond et al., 1994 and refs therein). Furthermore, AChE has been shown to be homologous to the cell adhesion molecules glutactin and neurotactin (Krejci et al., 1991) as well as to neuroligins which are neuronal cell surface proteins (Ichtchenko et al., 1996). Such findings have led to the suggestion that AChE may perform additional, non-classical function in the nervous system (Robertson and Yu, 1993; Greenfield, 1995; Layer and Willbold, 1995).

I. Introduction Acetylcholinesterase (AChE; EC 3.1.1.7) is an essential component of cholinergic synapses in both central and peripheral nervous systems. Within these specialized structures, AChE is responsible for the rapid hydrolysis of acetylcholine released from presynaptic nerve terminals thereby ensuring precise temporal control of synaptic transmission (see for review MassouliĂŠ et al., 1993; Taylor and Radic, 1994). However, it is well documented that AChE is also expressed in a variety of non-cholinergic tissues. For example, non-cholinergic regions of the brain such as the

347

Gene Therapy and Molecular Biology Vol 3, page 348

F i g u r e 1 . Quaternary structures of AChE molecular forms. Homomeric forms consist of monomer G1 , dimer G2 , tetramer G4 and a glycophospholipid (GPI)-linked dimer. Heteromeric forms consist of the hydrophobic-tailed G4 form and the asymmetric forms containing a collagenic structural subunit. Only the asymmetric form A12 is shown.

Homomeric forms include monomer G1, dimer G2 and tetramer G4 as well as a glycophospholipid-linked (GPI) dimer. Heteromers on the other hand, consist of: (i ) amphiphilic tetramers G4 linked to a 20 kDa hydrophobic anchor; and (i i ) the asymmetric forms A4, A8 or A12 in which 1, 2 or 3 soluble tetramers attach to a collagenic subunit, respectively. The functional significance of such polymorphism remains to be established yet, it has been proposed that it allows the placement of AChE molecules at distinct cellular locations where they can assume site-specific functions.

Interestingly, AChE is also abundantly expressed in hematopoietic cells where its expression is even more puzzling. Although the presence of AChE in erythrocytes was detected more than 70 years ago (see Lawson and Barr, 1987), its role in blood cell physiology still remains unclear. In recent years however, there has been considerable interest in this issue, and there is now increasing evidence suggesting the existence of a link between AChE expression and the proliferation and differentiation of hematopoietic cells. In this brief review, we initially describe the cellular and molecular biology of AChE and then focus more specifically on the putative involvement of this enzyme in the development of blood cells and elements. Our main objective is to highlight some of the latest findings which should prove useful to design further experimentation dealing with the regulation and functional significance of AChE during hematopoiesis under normal and pathological conditions.

Previous studies have shown that AChE is encoded by a single gene (Rotundo et al., 1988; Maulet et al., 1990; Soreq et al., 1990; Li et al., 1993a; Chan and Jasmin, 1995). Although only one copy of the gene exists, several transcripts are produced by alternative splicing (Sikorav et al., 1987, 1988; Schumacher et al., 1988; Maulet et al., 1990) (Figure 2). In mammals, exon 1 is untranslated while exons 2, 3 and part of exon 4 appear in all AChE transcripts since they encode the common catalytic domain of the mature protein (Li et al., 1991, 1993a). The C-terminal region of the protein is variable and is encoded either by the rest of exon 4 (called R for readthrough), or by alternatively spliced exons 5 or 6 to yield the H (hydrophobic) or T (tail) transcript, respectively. Two polyadenylation signals have been identified which in the case of the T transcript for

II.The AChE molecular forms and splice variants AChE exists as a family of molecular forms which differ in their structures and hydrodynamic characteristics while displaying similar catalytic properties (Taylor, 1991; see MassouliĂŠ et al., 1993). The molecular forms may be classified as homomeric or heteromeric on the basis of their association with specialized structural subunits (Figure 1). 348

Gene Therapy and Molecular Biology Vol 3, page 349 example, give rise to two mRNAs of 2.4 and 3.2 kb (Rachinsky et al., 1990; Legay et al., 1993a).

porcine, bovine and rat erythrocytes, but not from human or murine erythrocytes. The presence of an additional acyl chain on the inositol ring in the latter two species is thought to confer the resistance of the GPI-linked AChE dimer to PIPLC treatment (Roberts et al., 1988a,b). Such addition effectively prevents the formation of a cyclic myo-inositol 1:2-monophosphate which is an intermediate product of the PI-PLC cleavage reaction (Wilson et al., 1985). Consequently, deacylation of AChE in human and murine erythrocytes by alkaline hydroxylamine treatment renders the enzyme susceptible again to cleavage by PI-PLC (Toutant et al., 1989; 1991). By contrast to the information already available on the expression of AChE in erythrocytes, only a few studies have examined AChE in lymphocytes and platelets most likely because of their limited quantities in circulating blood (Méflah et al., 1984; Bartha et al., 1987; Richier et al., 1992).

The H transcript encodes the AChE catalytic subunit that contains the signal for GPI addition (Li et al., 1991) which ultimately leads to the formation of GPI-linked dimers (Li et al., 1991; Legay et al., 1993b; Michaelson et al., 1994; Coussen et al., 1995). This GPI-linked AChE dimer is expressed in mature erythrocytes and T lymphocytes (Ott et al., 1982; Szelenyi et al., 1982; Rosenberry and Scoggin, 1984; Roberts et al., 1987). By contrast, the T transcript can give rise to all other AChE molecular forms when coexpressed with appropriate structural subunits (Duval et al., 1992). The T transcript is abundantly expressed in muscle and neuron where it accounts therefore, for the multiplicity of molecular forms found in these cell types (Li et al., 1991; Duval et al., 1992; Li et al., 1993a; Legay et al., 1993a). Finally, the R transcript encodes a secreted AChE monomer whose expression may be developmentally regulated (Li et al., 1993b; Legay et al., 1995; Chan et al., 1998).

Histochemical studies of early hematopoietic cells have revealed that AChE is present in several distinct subcellular compartments. In human bone marrow cultures, AChE is detected in both the nucleus and cytoplasm of erythroblasts (Koekebakker and Barr, 1988), as well as in the nucleus of immature megakaryocytes (Lev-Lehman et al., 1997). A close examination of AChE expression during murine erythroid cell maturation further indicates that the enzyme is widely distributed in the nuclear membrane, endoplasmic reticulum and Golgi apparatus at early stages of development, and that it becomes confined to the Golgi apparatus in orthochromatic nucleated red blood cells in accordance with the end of AChE biosynthesis at this particular stage of cellular differentiation (Keyhani and Maigne, 1981). In addition, it appears that AChE is also secreted from normoblasts (Keyhani and Maigne, 1981) and megakaryocytes (Paulus et al., 1981). Taken together, these results suggest that the subcellular distribution of AChE as well as the species of molecular forms that are expressed, vary with the stage of hematopoietic cell differentiation. Accordingly, these changes in AChE localization and expression may therefore reflect distinct roles for the different molecular forms at specific stages of cell maturation (see Chan et al., 1998).

III. Expression of AChE in hematopoietic cells Previous studies have shown that T lymphocytes (Szelenyi et al., 1982), platelets (Schukla, 1986; Koekebakker and Barr, 1988; Sánchez-Yagüe et al., 1990) and erythrocytes (Low and Finean, 1977; Massoulié and Bon, 1982; Toutant et al., 1989) express significant levels of AChE. Interestingly, there are notable differences between species concerning the distribution of the enzyme in blood cell lineages. Biochemical analysis of AChE in erythrocytes has shown that humans have the highest levels of enzyme activity while cats have none (Zajicek, 1957). However, platelets and megakaryocytes from cats contain large amounts of AChE whereas in humans, these cells express only low levels of enzyme activity. Other species such as guinea pigs, horses, rabbits, and rodents, fall in between these two extremes, but AChE activity tends to typically be found predominantly in one cell lineage (Zajicek, 1957). The significance of this variability between species and cell types remains currently unclear.

Previous studies have also determined the species of mRNAs expressed in various hematopoietic tissues. Analysis of rat fetal liver and spleen has shown for example that all three splice variants are present in these hematopoietic organs (Legay et al., 1993b). Studies using adult mouse bone marrow (Li et al., 1993a) and murine erythroleukemia (MEL) cells (Chan et al., 1998) have also revealed a similar pattern of expression thereby suggesting that hematopoietic cells from both embryonic and adult tissues are capable of expressing, albeit at different levels, the R, H and T transcripts of AChE.

Amongst blood cell lineages, erythrocytes have been the most thoroughly studied in terms of AChE expression. Mammalian erythrocytes express GPI-linked AChE dimers on the extracellular surface of their plasma membrane (Ott et al., 1982; Rosenberry and Scoggin, 1984) and it has been shown that the GPI moiety is particularly important for anchoring AChE molecules onto the cell membrane (Incardona and Rosenberry, 1996). Interestingly, a certain degree of variation has been observed in the structure of the inositol ring within the glycolipid anchor. For example, AChE has been shown to be released readily by phosphatidylinositol-specific phospholipase C (PI-PLC) in 349

Gene Therapy and Molecular Biology Vol 3, page 350

F i g u r e 2 . Structure of the mammalian AChE gene and alternative splicing of AChE mRNAs. The promoter (P), exons (dark boxes), introns (light boxes) and 3'UTR (hatched box) containing two polyA+ signals (A) are shown. 4' denotes part of exon 4 that is retained together with intron 4 in the splicing of the R transcript. Note that splicing from exon 4 to either exon 5 or 6 generate the H and T transcripts, respectively.

Willbold, 1995). In this context, there has been an increasing number of reports that have recently demonstrated that AChE can in fact regulate neuronal morphogenesis and differentiation independently of its catalytic activity (see for example Layer et al., 1993; Jones et al., 1995; Small et al., 1995, Dupree and Bigbee, 1996; Inestrosa et al., 1996; Beeri et al., 1997; Holmes et al., 1997; Koenigsberger et al., 1997; Robitzki et al., 1997; Srivatsan and Peretz, 1997; Sternfeld et al., 1998).

Since in our recent studies we have observed a preponderance of R and T transcripts in MEL cells (Chan et al., 1998; see also Li et al., 1993a) which basically correspond to erythroblasts and normoblasts (Friend et al., 1971), and since mature hematopoietic cells express significant amounts of GPI-linked dimers that are encoded by the H transcript (see Figure 2), it may thus be hypothesized that the splicing pattern of immature AChE mRNA changes during differentiation of hematopoietic cells hence, further supporting the notion that different AChE molecular forms, originating from different transcripts (see above), are required at distinct stages of hematopoiesis (Chan et al., 1998).

In hematopoietic cells, the presence of AChE remains an enigma but there is nonetheless, considerable interest in identifying the physiological role of AChE in these cells particularly since the AChE gene maps to 7q22 (Getman et al., 1992) which is considered a critical region of the genome involved in the development of myelodysplastic syndromes and acute myeloid leukemias (Kere et al., 1989; Baranger et al., 1990; Mufti, 1992; Green, 1993). Additional clinical observations have further supported a link between aberrations in the AChE gene and severe hematological disorders. For example, the AChE gene frequently undergoes incomplete somatic amplification (Lapidot-Lifson et al., 1989) and mutation (Zakut et al., 1992) in hematological proliferation disorders such as megakaryocytopoiesis and

IV. Function of AChE in hematopoietic cells The hypothesis that AChE is involved in physiological functions other than the termination of neurotransmission has received considerable attention particularly in the nervous system where these putative additional roles are collectively referred to as the non-cholinergic functions of AChE (Robertson and Yu, 1993; Greenfield, 1995; Layer and 350

Gene Therapy and Molecular Biology Vol 3, page 351 thrombopoiesis. Furthermore, organophosphates, which are potent inhibitors of AChE and key components of pesticides, are believed to represent causative agents in various forms of leukemias (Brown et al., 1990). Although the role of AChE in hematopoietic cells is still unclear, the location of the AChE gene in a region which may contain a novel myeloidspecific tumor suppressor gene (Le Beau et al., 1986; Neuman et al., 1992; Johansson et al., 1993; Rodrigues et al., 1996; Le Beau et al., 1996), has led to the suggestion that AChE may in fact function as a tumor suppressor by regulating proliferation, differentiation and apoptotic events during normal hematopoietic cell development (Soreq et al., 1994; Stephenson et al., 1996).

hematopoietic stem cells. Further confirmation of the role of AChE in regulating apoptosis in these cells may lead to the identification of additional regulatory mechanisms controlling programmed cell death in hematopoietic tissues and that the loss of this regulatory step may in fact be involved in the etiology of hematological disorders including leukemias.

V. Conclusions and perspectives The notion that AChE fulfils additional, non-cholinergic functions has received an increasing amount of attention. In this context, hematopoietic cells are of considerable interest since there is now ample evidence showing that AChE is expressed both in early hematopoietic progenitors as well as in mature blood cells and elements. Although the specific function of AChE in hematopoietic cells remains obscure, converging lines of evidence suggest the existence of a link between AChE levels and the proliferative capacity of these cells. Future experiments will therefore prove useful not only to further test this hypothesis directly, but also, to begin delineating the splice variants, the regions of the AChE molecule as well as the signal transduction pathways that may be involved in mediating these effects. Because of the postulated clinical relationship between AChE expression and hematological disorders, it may also be envisaged that studies focusing on the regulation and functional significance of AChE expression in hematopoietic cells may ultimately lead to the design of novel therapeutic strategies.

Over the last two decades, several laboratories have directly examined the role of AChE in hematopoietic cells by using distinct experimental approaches. Treatment of mice with the AChE inhibitor neostigmine, resulted in significant increases in colony forming unit-megakaryocytes in the humerus as well as in the percentage of progenitor cells undergoing DNA synthesis (Burnstein et al., 1980). Similarly, suppression of AChE expression using sequencespecific antisense oligonucleotides in cultures from mouse bone marrow cells led to enhanced proliferation of pluripotent stem cells committed to erythropoiesis, megakaryocytopoiesis and macrophage production (Soreq et al., 1994). Interestingly, and of particular relevance, normal apoptosis in these cells was significantly reduced in comparison to untreated cell cultures (Soreq et al., 1994). Based on these latter studies, it appears therefore that the functional role of AChE is to limit the proliferation of hematopoietic stem cells since its function is expected to be inversely related to the effects of the AChE antisense oligonucleotides.

Acknowledgement The financial support of the Medical Research Council of Canada and of the Faculty of Medicine, University of Ottawa, is gratefully acknowledged. We also wish to thank members of the Jasmin laboratory for fruitful discussion.

Additional studies performed by other laboratories including ours, have also examined the relationship that appears to exist between AChE and the proliferative capacity of hematopoietic cells. Using MEL cells in culture for example, we have demonstrated a large increase in both intracellular and secreted AChE activity during cellular differentiation which coincides with hemoglobinization and the concomitant loss of their proliferative capacity (Chan et al., 1998). Paoletti and co-workers have found that fastgrowing MEL cell clones express consistently lower levels of AChE enzyme activity as compared to slow-growing ones (Paoletti et al., 1992). In addition, treatment of these cells with exogenous AChE has been shown to lead to a marked decrease in cell growth (Paoletti et al., 1992). In our experiments, we have also recently observed following AChE addition to the growth media, significantly more cell death in MEL cells already committed to the differentiation program most likely as a result of apoptotic events (unpublished observation). Together these results suggest therefore that AChE can act as a negative regulator of cellular replication along the differentiation program of

References Baranger L, Baruchel A, Leverger G, Schaison G, Berger R. (1 9 9 0 ) Monosomy-7 in childhood hemopoietic disorders. Leukemia 4, 345-347 Bartha E, Rakonczay Z, KĂĄsa P, HollĂĄn S, GyĂŠvai A. (1 9 8 7 ) Molecular form of human lymphocyte membrane-bound acetylcholinesterase. L i f e S c i 41, 1853-1860 Beeri R, Le Novere N, Mervis R, Huberman T, Grauer E, Changeux JP, Soreq H. (1 9 9 7 ) Enhanced hemicholinium binding and attenuated dendrite branching in cognitively impaired acetylcholinesterase-transgenic mice. J Neurochem 69, 2441-2451 Brown LM, Blair A, Gibson R, Everett GD, Cantor KP, Schulman LM, Burmeister LF, Van Lier SF, Dick F. (1 9 9 0 ) Pesticide exposures and other agricultural risk factors for leukemia among men in Iowa and Minnesota. Cancer Res 50, 65856591

351

Chan and Jasmin: Acetylcholinesterase in hematopoietic cells Burstein SA, Adamson JW, Harker LA. (1 9 8 0 ) Megakaryocytopoiesis in culture: modulation by cholinergic mechanisms. J C e l l P h y s i o l 103, 201-208

Inestrosa NC, Alvarez A, Calderon F. (1 9 9 6 ) Acetylcholinesterase is a senile plaque component that promotes assembly of amyloid beta-peptide into Alzheimer's filaments. M o l P s y c h i a t r y 1, 359-361

Chan RYY, Jasmin BJ. (1 9 9 5 ) Regulatory elements and transcription of the acetylcholinesterase gene in adult rat skeletal muscle fibers. S o c f o r N e u r o s c i 21, 800 (abstr.)

Johansson B, Mertens F, Mitelman F. (1 9 9 3 ) Cytogenetic deletion maps of hematologic neoplasms:Circumstantial evidence for tumor suppressor loci. Genes Chrom Cancer 8, 205-218

Chan RYY, Adatia FA, Krupa AM, Jasmin BJ. (1 9 9 8 ) Increased expression of acetylcholinesterase T and R transcripts during hematopoietic differentiation is accompanied by parallel elevations in the levels of their respective molecular forms. J B i o l C h e m 273, 9727-9733

Jones SA, Holmes C, Budd TC, Greenfield SA. (1 9 9 5 ) The effect of acetylcholinesterase on outgrowth of dopaminergic neurons in organotypic slice culture of rat midbrain. C e l l T i s s u e R e s 279, 323-330

Coussen F, Bonnerot C, Massoulié‚ J. (1 9 9 5 ) Stable expression of acetylcholinesterase and associated collagenic subunits in transfected RBL cell lines: production of GPIanchored dimers and collagen-tailed forms. Eur J C e l l B i o l 67, 254-260

Kere J, Ruutu T, Davies KA, Robinson IB, Watkins PC, Winqvist R, Chapelle A de la. (1 9 8 9 ) Chromosome 7 long arm deletion in myeloid disorders: a narrow breakpoint region in 7q22 defined by molecular mapping. B l o o d 73, 230-234

Dupree JL, Bigbee JW. (1 9 9 6 ) Acetylcholinesterase inhibitor treatment delays recovery from axotomy in cultured dorsal root ganglion neurons. J N eu r o cy to l 25, 439-454

Keyhani E, Maigne J. (1 9 8 1 ) Acetylcholinesterase mammalian erythroid cells. J C e l l S c i 52, 327-339

in

Koekebakker M, Barr RD. (1 9 8 8 ) Acetylcholinesterase in the human erythron. I. Cytochemistry. Am J Hematol 28, 252-259

Duval N, Massoulié J, Bon S. (1 9 9 2 ) H and T subunits of acetylcholinesterase from Torpedo, expressed in COS cells, generate all types of globular forms. J C e l l B i o l 118, 641-653

Koenigsberger C, Chiappa S, Brimijoin S (1 9 9 7 ) Neurite differentiation is modulated in neuroblastoma cells engineered for altered acetylcholinesterase expression. J Neurochem 69, 1389-1397

Friend C, Scher W, Holland JG, Sato T. (1 9 7 1 ) Hemoglobin synthesis in murine erythroleukemia cells in vitro: stimulation of erythroid differentiation by dimethylsulfoxide. P r o c N a t l A c a d S c i U S A 68, 378382

Krejci E, Duval N, Chatonnet A, Vincens P, Massoulié J. (1 9 9 1 ) Cholinesterase-like domains in enzymes and structural proteins: functional and evolutionary relationships and identification of a catalytically essential aspartic acid. Proc Natl Acad Sci USA 88, 6647-6651

Getman DK, Eubanks JH, Camp S, Evans GA, Taylor P. (1 9 9 2 ) The human gene encoding acetylcholinesterase is located on the long arm of chromosome 7. A m J H u m G e n e t 51, 170-177

Landwehrmeyer B, Probst A, Palacios JM, Mengod G. (1 9 9 3 ) Expression of acetylcholinesterase messenger RNA in human brain: an in situ hybridization study. Neurosci 57, 615-634

Green AR. (1 9 9 3 ) Chromosomal deletions in haematological malignancies. Lancet 341, 1567-1568 Greenfield SA. (1 9 9 5 ) A non-cholinergic function for acetylcholinesterase. In: DM Quinn, AS Balasubramanian, BP Doctor, P Taylor (eds). Enzymes of the Cholinesterase Family, New York: Plenum Press, 415-421

Lapidot-Lifson Y, Prody CA, Ginzberg D, Meytes D, Zakut H, Soreq H. (1 9 8 9 ) Coamplification of human acetylcholinesterase and butrylcholinesterase genes in blood cells: correlation with various leukemias and abnormal megakaryocytopoiesis. Proc Natl Acad S c i USA 86, 4715-4717

Hammond R, Rao R, Koenigsberger C, Brimijoin S. (1 9 9 4 ) Regional variation in expression of acetylcholinesterase mRNA in adult rat brain analyzed by in situ hybridization. Proc Natl Acad Sci USA 91, 10933-10937

Lawson AA, Barr RD. (1 9 8 7 ) Acetylcholinesterase in red blood cells. Am J Haematol 26, 101-112

Holmes C, Jones SA, Budd TC, Greenfield SA. (1 9 9 7 ) Noncholinergic, trophic action of recombinant acetylcholinesterase on mid-brain dopaminergic neurons. J Ne ur o sc i Re s 49, 207-218

Layer PG, Weikert T, Alber R. (1 9 9 3 ) Cholinesterase regulate neurite growth in chick nerve cells in vitro by means of a non-enzymatic mechanism. C e l l T i s s u e R e s 273, 219226

Ichtchenko K, Nguyen T, Südhof TC. (1 9 9 6 ) Structures, alternative splicing, and neurexin binding of multiple neuroligins. J B i o l C h e m 271, 2676-2682

Layer PG, Willbold E. (1 9 9 5 ) Novel functions of cholinesterases in development, physiology and disease. P r o g H i s t o c h e m C y t o c h e m 29, 1-99

Incardona JP, Rosenberry TL. (1 9 9 6 ) Construction and characterization of secreted and chimeric transmembrane forms of Drosophila acetylcholinesterase: a large truncation of the c-terminal signal peptide does not eliminate glycoinositol phospholipid anchoring. M o l B i o l C e l l 7, 595-611

Le Beau MM, Albain KS, Larson RA, Vardiman JW, Davis EM, Blough RR, Golomb HM, Rowley JD. (1 9 8 6 ) Clinical and cytogenetic correlations in 63 patients with therapy-related myelodysplastic syndromes and acute nonlymphocytic leukemia: Further evidence for characteristic abnormalities of chromosomes nos. 5 and 7. J C l i n O n c o l 3, 325-345

352

Gene Therapy and Molecular Biology Vol 3, page 353 Le Beau MM, Espinosa R III, Davis EM, Eisenbart JD, Larson AA, Green ED. (1 9 9 6 ) Cytogenetic and molecular delineation of a region of chromosome 7 commonly deleted in malignant myeloid diseases. B l o o d 88, 1930-1953

Neuman WL, Rubin CM, Rios RB, Larson RA, Le Beau MM, Rowley JD, Vardiman JW, Schwartz JL, Farber R. (1 9 9 2 ) Chromosomal loss and deletion are the most common mechanisms for loss of heterozygosity from chromosomes 5 and 7 in malignant myeloid disorders. B l o o d 79, 15011510

Legay C, Bon S, Vernier P, Coussen F, Massoulié J. (1 9 9 3 a ) Cloning and expression of a rat acetylcholinesterase subunit: generation of multiple molecular forms and complementarity with a Torpedo collagenic subunit. J Neurochem 60, 337-346

Ott P, Lustig A, Brodbeck U, Rosenbusch JP. (1 9 9 2 ) Acetylcholinesterase from human erythrocytes membranes: dimers as functional units. FEBS Lett 138, 187-189

Legay C, Bon S, Massoulié J. (1 9 9 3 b ) Expression of a cDNA encoding the glycolipid-anchored form of rat acetylcholinesterase. FEBS Lett 31, 163-166

Paoletti F, Mocali A, Vannucchi AM. (1 9 9 2 ) Acetylcholinesterase in murine erythroleukemia (Friend) cells: evidence for megakaryocyte-like expression and potential growth-regulatory role of enzyme activity. B l o o d 79, 2873-2879

Legay C, Huchet M, Massoulié J, Changeux J-P. (1 9 9 5 ) Developmental regulation of acetylcholinesterase transcripts in the mouse diaphragm: Alternative splicing and focalization. Eur J Neurosci 7, 1803-1809

Paulus J-M, Maigne J, Keyhani E. (1 9 8 1 ) Mouse megakaryocytes secrete acetylcholinesterase. B l o o d 58, 1100-1106

Lev-Lehman E, Deutsch V, Eldor A, Soreq H. (1 9 9 7 ) Immature human megakaryocytes produce nuclear-associated acetylcholinesterase. B l o o d 89, 3644-3653

Rachinsky TL, Camp S, Li Y, Ekstrom TJ, Newton M, Taylor P. (1 9 9 0 ) Molecular cloning of mouse acetylcholinesterase: tissue distribution of alternatively spliced mRNA species. Neuron 5, 317-327

Li Y, Camp S, Rachinsky TL, Getman D, Taylor P. (1 9 9 1 ) Gene structure of mammalian acetylcholinesterase. Alternative exons dictate tissue-specific expression. J B i o l Chem 266, 23083- 23090

Richier P, Arpagaus M, Toutant JP. (1 9 9 2 ) Glycolipid-anchored acetylcholinesterases from rabbit lymphocytes and erythrocytes differ in their sensitivity to phosphatidylinositol-specific phospholipase C. B i o c h i m B i o p h y s A c t a 1112, 83-88

Li Y, Camp S, Taylor P. (1 9 9 3 a ) Tissue-specific expression and alternative mRNA processing of the mammalian acetylcholinesterase gene. J B i o l C h e m 268, 5790-5797

Roberts WL, Kim BH, Rosenberry TL. (1 9 8 7 ) Bovine brain acetylcholinesterase primary sequence involved in intersubunit disulfide linkages. Proc Natl Acad Sci USA 84, 7817-7821

Li Y, Camp S, Rachinsky TL, Bongiorno C, Taylor P. (1 9 9 3 b ) Promoter elements and Transcriptional control of the mouse acetylcholinesterase gene. J B i o l C h e m 268, 3563-3572 Low MG, Finean JB. (1 9 7 7 ) Non-lytic release of acetylcholinesterase from erythrocytes by a phosphatidylinositol-specific phospholipase C. FEBS Lett 82, 143-146

Roberts WL, Myher JJ, Kuksis A, Low MG, Rosenberry TL. (1 9 8 8 a ) Structural characterization of the glycoinositol phospholipid membrane anchor of human erythrocyte acetylcholinesterase by fast atom bombardment mass spectrometry. J B i o l C h e m 263, 18766-18775

Massoulié J, Bon S. (1 9 8 2 ) The molecular forms of cholinesterase and acetylcholinesterase in vertebrates. Ann Re v Ne ur o sc i 5, 57-106

Roberts WL, Santikarn S, Reinhold VR, Rosenberry TL. (1 9 8 8 b ) Lipid analysis of the glycoinositol phospholipid membrane anchor of human erythrocyte acetylcholinesterase. J B i o l C h e m 263, 18776-18784

Massoulié J, Pezzementi L, Bon S, Krejci E, Vallette FM. (1 9 9 3 ) Molecular and cellular biology of cholinesterases. P r o g N e u r o b i o l 13, 31-91

Robertson RT, Yu T. (1 9 9 3 ) Acetylcholinesterase and neural development: new tricks for an old dog? N e w s P h y s i o l S c i 8, 266-268

Maulet Y, Camp S, Gibney G, Rachinski T, Ekstrom TJ, Taylor P. (1 9 9 0 ) Single gene encodes glycophospholipidanchored and asymmetric acetylcholinesterase forms: alternative coding exons contain inverted repeat sequences. Neuron 4, 289-301

Robitzki A, Mack A, Hoppe U, Chatonnet A, Layer PG. (1 9 9 7 ) Regulation of cholinesterase gene expression affects neuronal differentiation as revealed by transfection studies on reaggregating embryonic chicken retinal cells. Eur J Neurosci 9, 2394-2405

Méflah K, Bernard S, Massoulié J. (1 9 8 4 ) Interactions with lectins indicate differences in the carbohydrate composition of the membrane-bound enzymes acetylcholinesterase and 5'-nucleotidase in different cell types. B i o c h i m i e 66, 5969

Rodrigues Pereira Velloso E, Michaux L, Ferrant A, Hernandez M, Mesus P, Dierlamm J, Criel A, Louwagie A, Verhoef G, Booaerts M, Michaux JL, Bosly A, Mecucci C, Van den Berghe H. (1 9 9 6 ) Deletions of the long arm of chromosome 7 in myeloid disorders: loss of band 7q32 implies worst prognosis. Br J Haematol 92, 574-581

Michaelson S, Small DH, Livett BG. (1 9 9 4 ) Expression of dimeric and tetrameric acetylcholinesterase isoforms on the surface of cultured bovine adrenal chromaffin cells. J C e l l B i o c h e m 55, 398-407 Mufti GJ. (1 9 9 2 ) Chromosomal deletions in myelodysplastic syndrome. Leukemia Res 16, 35-41

Rosenberry TL, Scoggin DM. (1 9 8 4 ) Structure of human erythrocyte acetylcholinesterase. Characterization of

the

353

Chan and Jasmin: Acetylcholinesterase in hematopoietic cells intersubunit disulfide bonding and detergent interaction. J B i o l C h e m 259, 5643-5652

Szelenyi JG, Bartha E, Hollan SR. (1 9 8 2 ) Acetylcholinesterase activity of lymphocytes: an enzyme characteristic of Tcells. Brit J Haematol 50, 241-245

Rotundo RL, Gomez AM, Fernandez-Valle C, Randall WR. (1 9 8 8 ) Allelic variants of acetylcholinesterase: genetic evidence that all acetylcholinesterase forms in avian nerves and muscles are encoded by a single gene. Proc Natl Acad Sci USA 85, 7805-7809

Taylor P. (1 9 9 1 ) The cholinesterases. J B i o l Chem 266, 4025-4028 Taylor P, Radic Z. (1 9 9 4 ) The cholinesterases: from genes to proteins. Ann Rev Pharmacol T o x i c o l 34, 281-320

Sánchez-Yagüe J, Cabezas JA, Llanillo M. (1 9 9 0 ) Subcellular distribution and characterization of acetylcholinesterase activities from sheep platelets: relationships between temperature-dependence and environment. B l o o d 76, 737744

Toutant JP, Roberts WL, Murray NR, Rosenberry TL. (1 9 8 9 ) Conversion of human erythrocyte acetylcholinesterase from an amphiphilic to a hydrophilic form by phosphatidylinositol-specific phospholipase c and serum phospholipase D. Eur J Biochem 180, 503-508

Schukla SD. (1 9 8 6 ) Action of phosphatidylinositol specific phospholipase C on platelets: nonlytic release of acetylcholinesterase, effect on thrombin and PAF induced aggregation. L i f e S c i 38, 751-755

Toutant JP, Krall JA, Richards MK, Rosenberry TL. (1 9 9 1 ) Rapid analysis of glycolipid anchors in amphiphilic dimers of acetylcholinesterases. C e l l B i o l N e u r o b i o l 11, 219230

Schumacher M, Maulet Y, Camp S, Taylor. (1 9 8 8 ) Multiple mRNA species give rise to structural diversity in cholinesterase. J B i o l C h e m 263, 18979-18987

Wilson DB, Bross TE, Sherman WR, Berger RA, Majerus PW. (1 9 8 5 ) Inositol cyclic phosphates are produced by cleavage of phosphatidylphosphoinositols (polyphosphoinositides) with purified sheep seminal vesicle phospholipase c enzymes. Proc Natl Acad Sci USA 82, 4013-4017

Sikorav JL, Krejci E, Massoulié J. (1 9 8 7 ) cDNA sequences of Torpedo marmorata acetylcholinesterase: primary structure of the precursor of a catalytic subunit; existence of multiple 5' untranslated regions. EMBO J 6, 1865-1873

Zajicek J. (1 9 5 7 ) Studies on the histogenesis of blood platelets megakaryocytes. A c t a P h y s i o l Scand 40, Suppl 138

Sikorav JL, Duval N, Anselmet A, Bon S, Krejci E, Legay C, Osterlund M, Reimund B, Massouli‚ J. (1 9 8 8 ) Complex alternative splicing of acetylcholinesterase in Torpedo electric organ. EMBO J 7, 2983-1873

Zakut H, Lapidot-Lifson Y, Beeri R, Ballin A, Soreq H. (1 9 9 2 ) In vivo gene amplification in non-cancerous cells: cholinesterase genes and oncogenes amplify in thrombocytopenia associated with lupus erythematosus. Mutat Res 276, 275-284

Small DH, Gullveig R, Whitefield B, Nurcombe V. (1 9 9 5 ) Cholinergic regulation of neurite outgrowth from isolated chick sympathetic neurons in culture. J N e u r o s c i 15, 144151 Soreq H, Ben-Aziz R, Prody CA, Seidman S, Gnatt A, Neville L, Lieman-Hurwitz J, Lev-Lehman E, Ginzberg D, (1 9 9 0 ) Lapidot-Lifson Y, Zakut H. Molecular cloning and construction of the coding region for human acetylcholinesterase reveals a G+C-rich attenuating structure. Proc Natl Acad Sci USA 87, 9688-9692 Soreq H, Patinkin D, Lev-Lehman E, Grifman M, Ginzberg D, Eckstein F, Zakut H. (1 9 9 4 ) Antisense oligonucleotide inhibition of acetylcholinesterase gene expression induces progenitor cell expansion and suppresses hematopoietic apoptosis ex vivo. P r o c N a t l A c a d S c i U S A 91, 79077911 Srivatsan M, Peretz B. (1 9 9 7 ) Acetylcholinesterase promotes regeneration of neurites in cultured adult neurons of Aplysia. Neurosci 77, 921-931 Stephenson J, Czepulkowski B, Hirst W, Mufti GJ. (1 9 9 6 ) Deletion of the acetylcholinesterase locus at 7q22 associated with myelodysplastic syndromes (MDS) and acute myeloid leukaemia (AML). Leukemia Res 20, 235-241 Sternfeld M., Ming G-l, Song H-j, Sela K, Timberg R, Poo M-m, Soreq H. (1 9 9 8 ) Acetylcholinesterase enhances neurite growth and synapse development through alternative contributions of its hydrolytic capacity, core protein, and variable c termini. J Neurosci 18, 1240-1249

354

Gene Therapy and Molecular Biology Vol 3, page 355 Gene Ther Mol Biol Vol 3, 355-371. August 1999.

The ETS-domain transcription factors: lessons from the TCF subfamily. Review Article 1,3

1,3

2

Shen-Hsi Yang , Paula R. Yates , Yi Mo and Andrew D. Sharrocks

1

1

Department of Biochemistry and Genetics, The Medical School, University of Newcastle upon Tyne, Newcastle upon Tyne, NE2 4HH, UK. 2

The Wistar Institute, Department of Chemistry, University of Pennsylvania. Philadelphia, Pennsylvania 19104, USA.

3

These authors contributed equally to this work. __________________________________________________________________________________________________ Correspondence: A.D. Sharrocks, Ph.D.; Tel: +44-191 222 8800; Fax: +44-191 222 7424; E-mail: a.d.sharrocks@ncl.ac.uk Key Words: ETS-domain, ternary complex factor, TCF, transcription factors, MAP kinases. Received: 3 October 1998; accepted: 10 November 1998

Summary The ternary complex factors (TCFs) represent a subfamily of ETS-domain transcription factors. In recent years, several significant advances have been made in our understanding of their function at the molecular level. Many aspects of TCF function are conserved with other ETS-domain proteins and with transcription factors from other families, therefore these studies contribute more generally t o our understanding o f transcription factor function. Structural and biochemical studies have furthered our understanding of protein-DNA recognition by the TCFs. Furthermore, significant insights into the regulation of TCF activity by phosphorylation have been achieved. The TCFs have developed as a paradigm for nuclear targets of MAP kinase pathways and further advances have been made into understanding the spe cificity of kinase action towards the TCFs. The TCFs are k n o w n t o b i n d t o m u l t i p l e p r o t e i n p a r t n e r s w h i c h h a v e r o l e s i n up- and down-regulating their activity, recruitment to DNA and transmission of transcriptional activation and repression signals. F i n a l l y , o u r u n d e r s t a n d i n g o f t h e r o l e o f T C F s i n specific biological processes i s starting to become clearer, with roles defined in immediate-early gene regulation in a neuronal context.

I. Introduction

Members of the ETS-domain transcription factor family have been implicated in a variety of developmental processes in organisms that include, fruit flies, worms, fish, frogs and mice. In mammals, many ETS-domain proteins have also been associated with the adult immune system (reviewed in Bassuk and Leiden, 1996). One common theme appears to be that these transcription factors act as nuclear targets of the MAP kinase signal transduction pathways and thereby convert extracellular cues into nuclear responses (reviewed in Sharrocks et al., 1997; Graves and Petersen, 1998). Their importance is further emphasised by their implication in a variety of diseases in mammals, including cancer, in which individual proteins are either inappropriately expressed or expressed as

ETS-domain transcription factors are found in all metazoan organisms investigated to date (Degnan et al., 1993; Laudet et al., 1993). However, they appear to be absent in yeast and plant species, indicating that they function in novel regulatory cascades which are specific to the metazoan lineage. The ETS-domain proteins are characterised by the presence of a conserved DNA-binding domain (the ETS-domain; Karim et al., 1990) and are further subclassified on the basis of sequence similarities within this domain and the presence of additional conserved domains (reviewed in Sharrocks et al., 1997; Graves and Petersen, 1998).

355

Yang et al: The ETS-domain transcription factors fusions with other proteins. Such deregulated ETS-domain proteins act to subvert the normal tight cellular controls over the signalling pathways which are required to prevent deregulated cell growth and hence tumourigenesis (reviewed in Dittmer and Nordheim, 1998). In this review we will summarise recent advances in our understanding of the mechanism of action of the ternary complex factor (TCF) subfamily of ETS-domain transcription factors and relate these to our understanding of the function of other family members and transcription factors in general (for earlier review on the TCFs, see Treisman, 1994). In particular we will focus on how the specificity of promoter targeting is achieved by a combination of protein-DNA and protein-protein interactions. We also discuss the mechanisms by which protein kinase cascades are targeted to the TCFs and how phosphorylation affects their function. Finally, we will review the current understanding of how the TCFs function in a physiological context to convert extracellular signals into specific nuclear responses. Other recent reviews provide a more comprehensive account of the ETS-domain family in general (Bassuk and Leiden, 1996, Sharrocks et al., 1997, Graves and Petersen, 1998, Dittmer and Nordheim, 1998).

II. The modular structure of the TCFs. The name of the TCF subfamily is derived from the fact that they form ternary DNA-bound complexes with a second transcription factor, serum response factor (SRF),

and serum response elements (SREs) such as the one found in the c-fos SRE. Three human TCFs (Elk-1, SAP-1 and SAP-2) have been identified (Rao et al., 1989; Dalton and Treisman, 1992, Price et al., 1995). Murine Elk-1 and SAP-2 (ERP/Net) homologues also exist (Giovane et al., 1994, Lopez et al., 1994) and their conservation in vertebrates is demonstrated by the cloning of a cDNA encoding a TCF homologue (zTCF-1) from zebrafish (Brown and Sharrocks, unpublished data). In addition to the N-terminal ETS DNA-binding domain, the TCFs share three regions of sequence similarity (F i g . 1). This sequence similarity is also reflected in the functional similarity exhibited by these domains. The B-box mediates protein-protein interactions with SRF (Shore and Sharrocks, 1994), the D-domain serves as a docking site for MAP kinases (Yang et al., 1998a, Yang et al., 1998b) and the C-domain is a transcriptional activation domain which also serves as a regulatory domain whose activity is modified upon phosphorylation by MAP kinases (reviewed in Treisman, 1994, Whitmarsh and Davis, 1996; Gille et al., 1995; Kortenjann et al., 1994; Janknecht et al., 1994; Whitmarsh et al., 1995; Shore et al., 1996, Whitmarsh et al., 1997, Cavigelli et al., 1995). An additional functional domain has been identified in SAP-2 which acts as an autoinhibitory motif (NID) that regulates DNA binding and also mediates transcriptional repression (Maira et al., 1996). SAP-1 and zTCF-1 exhibit some similarity with this motif but this region is not conserved in Elk-1.

F i g u r e 1 . The domain structure of the TCFs. Each TCF contains four conserved domains (ETS-, B-, D- and C-), indicated by red boxes. SAP-2 contains an additional domain (NID), which is partially conserved in SAP-1 (45% identity) and zTCF-1 (40% identity) (indicated by a blue box). The lengths of the inter-domain linker regions vary, therefore additional dotted lines are included between the domains when a shorter linker is present.

356

Gene Therapy and Molecular Biology Vol 3, page 357 In the context of complex promoters, it is unlikely that autonomous DNA binding will be the sole determinant of binding specificity. Instead, the formation of complexes with partner proteins is also likely to contribute to the affinity and specificity of promoter binding by ETSdomain proteins (see section V). These additional proteinprotein interactions likely compensate for the loss of affinity associated with the more degenerate binding sites typically found in promoters. Indeed, in ternary complexes with SRF, Elk-1 recognises a relaxed version of its consensus binding site (RC/aC/aGGAA/tRT/c) which more closely resembles the consensus sequence selected by SAP-1 (Treisman et al., 1992). Significantly, the etsmotif within the c-fos SRE represents a lower affinity TCF binding site and requires concomitant binding of SRF to permit high affinity binding. This combinatorial binding might act to enhance promoter-specific recognition and/or allow a tighter regulation of TCF DNA binding by MAP kinase phosphorylation (see section III-C). Similarly, binding of Elk-1 and SAP-2 to the ets-motif in the mb-1 promoter requires binding of Pax-5 to the adjacent site (Fitzsimmons et al., 1996). Autonomous binding of SAP-1 to the c-fos SRE is also possible, albeit with much reduced affinity compared to SAP-1 binding in ternary complexes with SRF (Dalton and Treisman, 1992, Shore and Sharrocks, 1995, Masutani et al., 1997). However, although the physiological significance of this observation is questionable, it raises the possibility that binding sites might exist at which the TCFs act as the major determinants of DNA binding specificity such as is seen with the N10 promoter (Latinkic et al., 1996).

Each domain in the TCFs is functionally separable from the rest of the protein and each has been shown to act in an autonomous fashion. The ETS-domain is sufficient for DNA binding (Janknecht and Nordheim, 1992), the Bbox is sufficient for binding to SRF (Shore and Sharrocks, 1994), the D-domain is sufficient for binding to ERK2 (Yang et al., 1998a) and the C-domain represents an independent transcriptional activation domain (reviewed in Treisman, 1994). However, it is becoming apparent that within the context of the full-length proteins, significant intramolecular crosstalk occurs between domains. For example, the D-domain is required for efficient phosphorylation of the C-domain (Yang et al., 1998a, Yang et al., 1998b) and phosphorylation of the C-domain elicits a conformational change elsewhere in the protein which alters inhibitory interactions with the N-terminal part of the protein (see section III-C). Therefore, although the TCFs are clearly modular in structure, their normal function requires the integrated functioning of their constituent domains.

III. DNA binding properties A. DNA binding sites The recognition of DNA target sites is one of the major mechanisms responsible for maintaining the specificity of transcription factor function and is mediated by structural modules referred to as DNA-binding domains. ETS-domain transcription factors share a highly conserved DNA-binding domain of approximately 85 amino acids that mediates binding to DNA target sites known as etsmotifs which harbour a central GGA trinucleotide motif. The DNA binding affinity and specificity for individual family members is further defined by the DNA base-pairs that flank this central core motif (reviewed in Sharrocks et al., 1997 and Graves and Petersen, 1998). However, although the consensus binding sequences for ETS-domain proteins are very similar, it should be emphasised that the binding efficiency of proteins to individual sites may vary considerably. This is evident from the DNA binding properties of the TCFs (Table 1, Shore et al., 1996, Brown and Sharrocks, unpublished data). These studies show that sequences corresponding to the site, ACCGGAAGTR, represent high affinity binding sites for the TCFs. However, SAP-1 shows a more relaxed binding site selectivity and can tolerate several nucleotide changes thereby allowing efficient binding to a greater spectrum of sites than Elk-1. SAP-2 also differs and exhibits an intermediate DNA binding specificity (Shore et al., 1996; Brown and Sharrocks unpublished data). Moreover, the binding specificity of the recently cloned zebrafish TCF (zTCF-1) closely mirrors that exhibited by Elk-1 (Brown and Sharrocks, unpublished data).

B. Structure of the ETS-domain The ETS-domain transcription factors all contain a common DNA-binding motif which is conserved both in the primary sequence and in the secondary and tertiary structure elements. The three-dimensional structures of the ETS-domains of the family members, Fli-1, Ets-1, PU.1/Spi-1, (reviewed in Graves and Petersen 1998; Sharrocks et al., 1997) and GABP! (Batchelor et al., 1998) have been determined. More recently, the structures of the SAP-1 ETS-domain bound to the high affinity E74 site and the ets motif in the c-fos SRE DNA have also been elucidated (F i g . 2 ; Mo et al., 1998). Inspection of the tertiary structure provides insights into the roles of highly-conserved residues within the ETS-domain and the roles of non-conserved residues in mediating the unique activities of each protein. The ETS-domain of SAP-1, like Fli-1, Ets-1, PU.1 and GABP!, forms a winged helixturn-helix topology and the protein assumes an !/" fold with three !-helices and four anti-parallel "-strands on opposite faces of the protein which pack against one another to form an extensive hydrophobic core.

357

Yang et al: The ETS-domain transcription factors

DNA binding properties Consensus binding site

Elk-1

AACCGGAAG TG/a

Phosphorylation properties

3 comple x formati on (TCFSRFSRE)

Autoinhibitory domain

Regulation of DNA binding by phosphorylati on

Enhanced transcriptio nal activation

b

YES

B-box and C-terminal TAD

YES

YES

ERKs/ JNKs/p38!,"2 ,#,$

C-terminal TAD B-box and C-terminal regions

YES

YES

ERK2/p38!

ERK

?

?

YES

YES

ERK/p38!,"2 /unknown IL-1 & CSF1 induced kinases

ERK/p38!,"2

?

S381 /S387

NID and Cterminal regions

NO

YES

ERK2/p38!

?

?

e

a

Activating kinases

Kinases targeted by D-domain binding

Phosphoacceptor sites

c

S383/S389(ERK2) T363/T368(JNK2)

ERK/JNK

preferred in vivo sites

regulate transcription activation and DNA binding

d

S383 /S389

T417/S422(p38!)

RC/aC/a GGA A/tRT/c

zTCF-1

Elk-1-like

YES

SAP-1

NACCGGAA/t G/aT/cN

YES

SAP-2

SAP-1-like

Weak

S357 /S363

T a b l e 1 . Summary of DNA binding and phosphorylation properties of the TCFs. See text for details and references. a Site-selection carried out in ternary complexes with SRF. b These kinases are implicated from in vivo studies using overexpressed kinase cascade components or "natural" inducers and pharmacological substrates. c ERK2 and JNK2 require different D-domain residues for optimal binding. d These sites were identified in a single comparative studies of the three TCFs (Price et al., 1996). Other studies have analysed the sites phosphorylated by individual kinases. e Phosphorylation at these sites has no apparent effect on DNA binding.

Most of the conserved residues function to stabilise this hydrophobic core and to mediate protein-DNA interactions. The DNA recognition helix (!3) is embedded in the DNA major groove and additional DNA contacts are made by the 'wing' (between the "-strands 3 and 4) and the loop separating helices 2 and 3. These additional binding regions flank the DNA recognition helix and are anchored to opposite strands in the DNA minor groove. An overall bend of 11째 is induced in the DNA which facilitates the insertion of the recognition helix into the major groove. Interestingly, the binding site adopts partial A-form character. This may have some biological significance as this requirement would reduce the DNA binding affinity in vivo, thereby reducing promiscuous DNA binding by ETSdomain proteins in the absence of appropriate protein partners. The replacement of two residues (D38 and D69) in Elk1 by the corresponding residues from SAP-1 (Q37 and V68 respectively) is sufficient to confer the SAP-1 DNAbinding specificity on Elk-1 (Shore et al., 1996). V68 is located after the C-terminus of recognition helix whereas Q37 is located between "2 and !2. The locations of these residues suggest that neither is in a position to make direct hydrophobic or H-bond contacts with the DNA (Mo et al., 1998). Thus, the effect of these residues must be indirect 358

and be mediated by altering DNA contacts made by other amino acids. More recently, further insights into this recognition process have been achieved from structural studies on the Elk-1-DNA complex (Mo and Marmorstein, unpublished data). Preliminary studies indicated that the structure of the Elk-1 ETS-domain would not differ significantly from other family members (Shore et al., 1995; Bisset, Virden and Sharrocks, unpublished data). Indeed, the overall structure and DNA contacts made by the ETS-domain are very similar to SAP-1. However, there are several key differences which may help to explain the different DNA binding specificities of the TCFs. One of these is that Tyr66 in the recognition helix no longer makes a base contact and instead makes a H-bond contact to a backbone phosphate. The presence of D69 (one of the key specificity determinants) appears to be responsible for this change in interactions by Tyr66. The second residue involved in determining the unique binding specificities of the TCFs (D38) may act to alter the overall orientation of the recognition helix. Hence, a combination of biochemical studies and the structure of the SAP-1/DNA complex demonstrates that these non-conserved residues within a conserved structural framework have a profound effect on DNA binding specificity.

Gene Therapy and Molecular Biology Vol 3, page 359 F i g u r e 2 . Model of the ternary SAP1-SRF-SRE complex. The structure of the SAP-1 ETS-domain bound to the c-fos SRE (Mo et al., 1998) and SRF bound to an SRE-like site (Pellegrini et al., 1995) have been determined independently and the two structures subsequently combined (Mo et al., 1998). The secondary structure elements in the SAP-1 ETS-domain are shown in dark blue whereas a hypothetical structure of the B-box and linker region joining this to the ETS-domain are shown in light blue. Mutagenic studies indicate that the B-box is likely to form an a-helix and that several hydrophobic residues (indicated) play essential roles in the interaction with SRF (Ling et al., 1997). The core DNA-binding domain of SRF is shown in red as a space filling model. Residues which have been shown to form the interaction surface with the Elk-1 Bbox are indicated in white. The yellow arrow indicates that the B-box interacts with this region of SRF. The DNA is shown in grey and the nucleotides comprising the central GGA of the etsmotif are shown in yellow.

allosteric change mediated by cooperating C- and Nterminal modules from outside the ETS-domain (reviewed in Graves and Petersen, 1998). Part of the regulatory mechanism in Ets-1 involves structural changes in the inhibitory elements themselves which may be transduced into changes in the overall protein conformation. Such conformational changes may be activated by phosphorylation and/or coupling with other cofactors and would act to relieve the auto-inhibition, although the regulatory 'switch' within Ets-1 is still unknown. Moreover, these mechanisms may also achieve tighter regulation over transcriptional activation by reciprocal effects on the structure and/or accessibility of the transcriptional activation domain. The TCFs are also regulated by autoinhibitory mechanisms (Table 1; reviewed in Sharrocks et al., 1997 and Graves and Petersen, 1998). In Elk-1, DNA binding is inhibited by a combination of the B-box and C-terminal transcriptional activation domain (Janknecht et al., 1994; Yang and Sharrocks, unpublished data). Recently, several advances in our understanding of this inhibitory process have been made (Yang and Sharrocks, unpublished data). Direct

A comparison of the structures of the SAP-1/DNA and Elk-1/DNA complex and the analogous complexes containing the mutant (D38Q/D69V) Elk-1 ETS-domain will provide further insights into this molecular recognition process.

C. Regulation of DNA binding Many ETS-domain transcription factors are subject to auto-inhibitory mechanisms which regulate their DNAbinding activity (Sharrocks et al., 1997 and Graves and Petersen, 1998). This may function to prevent promiscuous DNA-binding because of their relatively nonstringent DNA-binding specificities. However, the autoinhibitory regions and mechanisms of action vary amongst family members. Recently, the concept of transcription factor modularity has been challenged by the observation that DNA can act as an allosteric activator, thereby implying structural coupling between the DNA-binding and transcriptional activation domains (reviewed in Lefstin and Yamamoto, 1998). In the case of the ETS-domain protein Ets-1, the inhibitory mechanism involves such an 359

Yang et al: The ETS-domain transcription factors binding of the ETS-domain to the B-box and binding of the N-terminal half of the protein (containing the ETS-domain and B-box) to the C-terminal TAD can be demonstrated, indicating that intramolecular interactions may serve to regulate DNA binding. Partial proteolysis experiments, CD and fluorescence absorption spectra have clearly demonstrated that phosphorylation by ERK2 induces a significant conformational change in Elk-1 which is thought to relieve these inhibitory interactions. Moreover, deletions and disruptive point mutations in the B-box block the induction of DNA binding and concomitant conformational structural changes observed upon phosphorylation. This suggests that this motif acts to couple C-terminal phosphorylation and relief of inhibitory interactions with the N-terminal ETS-domain. Furthermore, experiments with phosphorylated peptide effector molecules (which encompass the key phosphoacceptor motifs, Ser383/Ser389) indicate that phosphorylation triggers an allosteric change in Elk-1 which activates DNA binding. An elegant model emerges in which phosphorylation of the Elk-1 TAD plays a

pivotal role in the activation of its DNA binding activity (F i g . 3 ). Inhibitory interactions are relieved by an intramolecular switch which triggers a conformational changes via the B-box to the ETS DNA-binding domain. In addition to the effects on DNA binding, the conformational changes triggered by phosphorylation are also likely to play a pivotal role in enhancing the transcriptional activation potential of Elk-1 (see section VC). SAP-1 is also regulated by an autoinhibitory mechanism which involves both B-box and C-terminal regions although these studies are less extensive (Dalton and Treisman, 1992). In contrast, the autoinhibition of DNA binding by SAP-2 appears to be regulated differently to the other TCFs by a unique inhibitory domain (NID) which is located C-terminal to the B-box and is predicted to form a helix-loop-helix (HLH)-like structure (Maira et al., 1996). This domain of SAP-2 is also sufficient for mediating transcriptional repression by SAP-2 (see section V-C). The NID exhibits sequence similarity with the HLH domain in the Id proteins.

F i g u r e 3 . Activation of Elk-1 DNA binding by phosphorylation. In the unphosphorylated state, Elk-1 exists in an inhibited conformation. Upon phosphorylation, Elk-1 undergoes a conformational change (arrows indicate changes in the accessibility of the inter-domain linkers to proteolytic cleavage; Yang and Sharrocks, unpublished data) which is accompanied by a relief of intramolecular inhibition of the ETS-domain. Elk-1 can then efficiently bind to DNA, either on its own in binary complexes (2 0 ) or in conjunction with SRF in ternary complexes (30 ). The major regulatory sites (Ser383 and Ser389) involved in the activation of DNA binding by MAP kinase-induced phosphorylation are indicated. Phosphorylation of these residues is proposed to trigger a local change in the C-terminal transcriptional activation domain (TAD) which in turn triggers the conformational change and subsequent activation of DNA binding.

360

Gene Therapy and Molecular Biology Vol 3, page 361

F i g u r e 4 . Sequence conservation in the C-terminal domains of the TCFs. The sequences of the D-domain, the C-domain, the linker separating these two domains and the C-terminal tail of Elk-1, SAP-1, SAP-2 and zTCF-1 are shown. The numbers of the amino acids at the ends of each segment with respect to the full-length proteins are shown. Residues which are identical in at least three of the four TCFs are shown in red. Asterisks indicate that the residues are conserved in all four TCFs. Potential MAP kinase sites conserved in all four TCFs are highlighted in blue.

Interestingly, the Ids interact with the TCFs via their ETS-domains (see section V-D) and cause their dissociation from low affinity ets-motifs such as the one contained in the c-fos SRE (Yates et al, 1999). Thus helix-loop-helixlike motifs can act both in cis and in trans to inhibit DNA binding by the TCFs.

IV. Regulation by signal transduction pathways A. Activation of the TCFs by phosphorylation The MAP kinase pathways play major roles in converting extracellular signals such as mitogens, growth factors, cytokines and stress into nuclear responses (reviewed in Karin, 1994; Treisman, 1996). The activation of MAP kinase signal transduction pathways results in the phosphorylation of transcription factors by the terminal kinases in these cascades. The activity of many ETSdomain transcription factors is regulated by MAP kinase pathways (reviewed in Sharrocks et al., 1997) and may therefore contribute to the generation of certain types of cancer which result from inappropriate triggering of these cascades. 361

TCFs have been used as a paradigm to study the activation of transcription factors by signal-induced phosphorylation. Phosphorylation of the TCFs, Elk-1 and SAP-1, occurs at multiple conserved carboxy-terminal S/TP motifs and leads to enhanced DNA binding and TCFmediated transcriptional activation (Table 1; reviewed in Treisman, 1994, Whitmarsh and Davis, 1996; Gille et al., 1995; Kortenjann et al., 1994; Janknecht et al., 1994; Whitmarsh et al., 1995; Shore et al., 1996, Whitmarsh et al., 1997, Cavigelli et al., 1995). However, phosphorylation only affects the transactivation potential of SAP-2 and does not appear to stimulate its DNA binding activity (Lopez et al., 1994; Giovane et al., 1994; Price et al., 1996). It has also been demonstrated that the phosphorylation enhances the formation of a quaternary complex at the c-fos SRE containing two Elk-1 molecules (Gille et al., 1996), although the physiological relevance of this observation remains unclear. Five potential MAP kinase sites are conserved amongst all four family members (F i g . 4 ). Additional sites exist which are either unique to individual proteins or conserved in a subset of the TCFs. However, the role of the individual phoshoacceptor motifs is still unknown. It is also unclear as to whether these motifs are phosphorylated

Yang et al: The ETS-domain transcription factors in a specific or random order and whether phosphorylation occurs cooperatively. However, it is evident that phosphorylation of Ser383 in Elk-1 is critical for transcriptional activation and efficient ternary complex formation with DNA-bound SRF (reviewed in Treisman, 1994). Similarly, phosphorylation of the corresponding serine residues is essential for maximal stimulation of the transcriptional activity of SAP-1 and SAP-2 (Giovane et al., 1994; Strahl et al., 1996; Janknecht and Hunter, 1997). However, it should be emphasised that whilst other sites play less critical roles when analysed individually, in combination, they play important roles in regulating the activity of the TCFs (Treisman, 1994). Although the TCFs are closely related and exhibit significant sequence conservation in the C-terminal activation domain (F i g . 4 ; Price et al., 1995), they appear to respond differently to MAP kinase signalling pathways. In humans, at least three parallel MAP kinase pathways exist and several distinct MAP kinases have been identified in each class of pathway. The ERK subclass contains ERK1 and ERK2, the JNK subclass contains JNK1, JNK2 and JNK3, and the p38 subclass contains p38!, p38", p38# and p38$ (reviewed in Whitmarsh and Davis, 1996, Robinson and Cobb, 1997, Cohen, 1997, Wang et al., 1997). Current evidence suggests that Elk-1 is a target for all three classes of MAP kinase pathways, ERK, JNK and p38!/"2/# /$ (reviewed in Treisman, 1994, Whitmarsh and Davis, 1996, Kortenjann et al., 1994, Janknecht et al., 1994; Gille et al., 1995, Whitmarsh et al., 1995; Cavigelli et al., 1995; Enslen et al., 1998; Yang and Sharrocks, unpublished data). However, SAP-1 and SAP-2 appear to only be targeted efficiently by the ERK and p38!/"2 pathways (Price et al., 1996; Strahl et al., 1996; Whitmarsh et al., 1995; Whitmarsh et al., 1997; Yang, Galanis and Sharrocks, unpublished data). In comparison, SAP-1 and SAP-2 appear to be a poor JNK substrates (Price et al., 1996; Strahl et al., 1996; Whitmarsh et al., 1995, 1997; Yang and Sharrocks, unpublished data) although when overexpressed, JNKs can phosphorylate and activate SAP-1 to some extent (Janknecht and Hunter, 1997a). Similarly, zTCF-1 appears to be differentially targeted by different classes of MAP kinases as it is only efficiently activated by ERK and p38! but is a poor substrate for JNK isoforms both in vitro and in vivo (Yang and Sharrocks, unpublished data). However, as many of these studies have been carried out in the presence of overexpressed kinase cascade components and/or transcription factor substrates, it is possible that normal specificity determinants are bypassed (e.g. section IV-B). Thus care has to be taken in interpreting these results in a physiological context. Nevertheless, it is clear that multiple MAP kinase pathways can converge on the TCFs. One of the future challenges will be to determine which of these act under normal physiological conditions.

362

The existence of novel SAP-1 and zTCF-1-specific kinase(s) has been suggested from transient transfection assays. In CHO cells, IL-1 stimulation of Elk-1 is mediated by the JNK pathway. However, a dominantnegative form of MKK4 and overexpression of JIP-1 only have small inhibitory effect on IL-1-induction of reporters regulated by GAL4-SAP-1 and GAL4-zTCF fusions, indicating that IL-1 signalling to SAP-1 and zTCF-1 is mediated by a JNK-independent kinase pathway (Whitmarsh et al., 1997; Brown, Yang and Sharrocks, unpublished data). In the case of SAP-1, this activation is also independent of the p38 pathway (Whitmarsh et al., 1997). Interestingly, the existence of additional signalling pathways to SAP-1 has also been implicated from studies on CSF-1 stimulation of macrophages. In this case, growth factor activation of SAP-1 occurs independently of the Ras/ERK signalling cascade (Hipskind et al., 1994a). Similarly, the existence of novel Elk-1 kinases (other than ERK1/2) which are activated in response to FGF signalling and overexpression of activated Raf-1 proteins has been demonstrated (Chung et al., 1998). Thus further complexities in signalling to the TCFs are likely to be uncovered as the identities of these kinases become known. The differential utilisation of TCFs and MAP kinase signalling pathways represents a potential mechanism for the determination of cell-type-specific responses to extracellular stimuli. However, cross-cascade signals involving Rac1 and cdc42 (activating the JNK pathway) and Raf-1 co-operate to activate ERK and lead to a synergistic increase in TCF-dependent transcriptional activation (Frost et al., 1997). Similarly, the ERK and p38 cascades cooperate to activate Elk-1 in response to UV stimulation (Price et al., 1996). The TCFs may therefore also allow integration of distinct extracellular signals via a single nuclear effector molecule.

B. MAP kinase targeting motifs Due to the limited nature of the consensus sequence of the phosphoacceptor motifs, initial binding of the MAP kinases to transcription factors has been proposed as a mechanism to enhance substrate specificity over simple recognition of the local context of the phosphorylation motif (Kallunki et al., 1996). For example, in the case of phosphorylation of c-Jun by the JNK MAPKs, the local context of the phosphoacceptor motifs and the presence of a kinase docking domain on the transcription factor, combine to enhance its substrate specificity (Derijard et al., 1994; Kallunki et al., 1994, 1996; Sluss et al., 1994; Dai et al., 1995; Gupta et al., 1996). Recently, the D-domain homology region has been demonstrated to play an analogous role in the TCFs. This motif is distinct from the conserved phosphoacceptor motifs (F i g . 4 ) and in the case of Elk-1 is responsible for targeting by the ERK and

Gene Therapy and Molecular Biology Vol 3, page 363 JNK MAP kinases and is required for its efficient phosphorylation and activation. In contrast, this domain does not appear to play a role in phosphorylation by the p38 subclass (Yang et al., 1998a; Yang et al., 1998b). This suggests that Elk-1 phosphorylation by p38 may differ from that mediated by the ERK and JNK subclasses, and that the p38 MAPKs may phosphorylate Elk-1 in a more constitutive manner which does not require rapid and efficient kinase targeting. This is consistent with the observation that p38 is insufficient for mediating Elk-1 activation in response to UV and requires cooperation with the ERK pathway is required (Price et al., 1996). Phosphorylation of both SAP-1 and zTCF-1 by ERK also appears to be facilitated by the presence of the D-domain (Galanis, Yang and Sharrocks, unpublished data). However, in contrast to Elk-1, SAP-1 appears to be targeted by p38! and p38"2 via the D-domain (Galanis and Sharrocks, unpublished data) which is consistent with the robust activation of SAP-1 observed with the p38 MAP kinases (Janknecht and Hunter, 1997b). Thus, although the Ddomains are highly conserved amongst the TCFs, sequence differences within this motif must be responsible for this differential targeting. In addition to kinase binding, it should be emphasised that the local context of the phosphoacceptor motifs probably plays a role in determining substrate specificity although this has not been investigated. Indeed, each class of MAP kinase exhibits preferences for a different subset of the potential phosphoacceptor sites (Table 1; Price et al., 1996). It is unclear how widespread the docking of kinases to specific motifs on transcription factors is. However, one possible implication is that under physiological conditions, these interactions are an essential component of the specificity determining mechanism and hence explain how different MAP kinases could elicit different effects on highly related transcription factors. Future studies should help resolve the generality and importance of this phenomenon.

role of this phosphatase in vivo. In addition, a novel regulator of MAP kinase signalling, the kinase suppression of Ras (KSR), blocks EGF-induced Elk-1 activity without directly altering the activity of the ERK pathway and may promote the accumulation of dephosphorylated Elk-1 by activating the PP2B phosphatase (Sugimoto et al., 1997). Previously, the okadaic acid-sensitive phosphatases PP1A and PP2A have been implicated in the downregulation of c-fos by the ternary TCF-SRF-SRE complex. However, it is unclear whether these phosphatases directly target the TCF component (Hipskind et al., 1994b). Further studies are required to determine which phosphatases act upon the different TCFs in vivo following activation by each of the MAPK cascades.

D. "Physiological" Elk-1 phosphorylation To date, most of the studies on Elk-1 phosphorylation have been carried out in fibroblast-derived cell lines with overexpressed proteins. However, there is growing evidence linking Elk-1 phosphorylation with MAP kinase activation in neuronal cells. In the brain for example, Elk-1 is expressed exclusively in neuronal cell types where it is localised to both the cytoplasm and nucleus (see section VI-B) (Sgambato et al., 1998). Phosphorylated (ie activated) Elk-1 can be detected in both these compartments and hyperphosphorylation of ERK and Elk-1 spatiotemporally correlates with c-fos mRNA induction in these cells, providing a link between Elk-1 phosphorylation and immediate-early gene induction in vivo (Sgambato et al., 1998). Furthermore, cAMP can activate Elk-1 (via cAMP%B-Raf%Rap1%ERK%Elk-1) in a cell type-specific manner and induce neuronal differentiation (Vossler et al., 1997). Thus, Elk-1 appears to play an important role in tissue-specific regulation of cell growth and differentiation via the ERK signal pathway in neuronal cells.

V. Protein-protein interactions.

C. Downregulation of TCF activity by dephosphorylation Whilst the activation of TCFs by MAP kinases is relatively well understood, negative regulators such as protein phosphatases have not been studied in detail. In mammalian cells, several subfamilies of serine-threonine phosphatases exist, including PP1A, PP2A, PP2B (calcineurin) and PP2C. Elk-1 phosphatase activity in cell extracts is inhibited by cyclosporin A (a PP2B inhibitor), but not by okadaic acid (PP1 and PP2A inhibitors) in vitro indicating that PP2B may be a physiological Elk-1 phosphatase (Sugimoto et al., 1997). Furthermore, cyclosporin A also significantly enhances EGF-induced Ser383 Elk-1 phosphorylation in COS cells following activation of the ERK pathway, providing evidence for a 363

The TCFs have been implicated in interactions with a wide range of proteins including transcription factors, regulatory proteins and coactivators. Such interactions serve to connect the TCFs to upstream regulatory pathways, modify their activities and mediate their effect on the transcriptional apparatus. In addition, these interactions serve to provide an additional level of complexity which enhances their specificity of action. This is particularly important in promoter recognition where protein-protein interactions supplement the DNA-protein interactions mediated by the ETS-domain (see section III).

A. SRF and Pax-5

Yang et al: The ETS-domain transcription factors Multi-component transcription factor complexes play important roles in integrating extracellular signals into promoter-specific transcriptional responses. The TCFs can form complexes on two types of composite binding sites, which contain ets-motifs and either Pax- or SRF-(CArG boxes) binding motifs. The ternary complex at the c-fos SRE between TCFs and SRF has developed as a paradigm for the latter type of complex. This ternary complex plays a pivotal role in the activation of immediate-early gene expression upon induction by a series of mitogenic and stress stimuli (reviewed in Cahill et al., 1996; see section VI-A). Direct protein-protein interactions between the TCFs and SRF play an essential role in the formation of the ternary complex. Each TCF possesses a 21 amino acid conserved segment known as the B-box which is required, in addition to the ETS-domain, for complex formation with SRF and the c-fos SRE (reviewed in Treisman, 1994, Sharrocks et al., 1997). Recently, Net-b, a naturally occurring splice variant of the murine SAP-2 homologue, has been cloned and found to represent one of the major components of the ternary complexes found in vivo (Maira et al., 1996). Net-b represents a dominant-negative form of the TCFs as it contains the ETS-domain and the B-box but lacks the C-terminal transcriptional activation domain. It can therefore still bind to SRF-SRE complexes but is unable to activate transcription. Furthermore, as Net-b also retains the NID, it has the potential to act as an "active" repressor in addition to acting as a competitive inhibitor protein. The 'grappling hook' model was proposed to describe the mechanism of ternary complex formation in which the B-box is tethered to the ETS-domain by a flexible linker which permits interactions with SRF from variable distances (Treisman et al., 1992). Recent molecular modelling studies of a SAP-1/SRF/DNA complex provide evidence to support this model (F i g . 2 ; Mo et al., 1998). In the model, the DNA-binding domains of SRF and SAP1 bind to opposite sides of the major groove, positioning the C-terminus of the SAP-1 ETS-domain and the B-box interaction surface on SRF more that 45Ă… apart. However, the linker region between the ETS-domain and the B-box in SAP-1 would be long enough to accommodate this distance in an extended conformation. The B-box is necessary and sufficient to interact with SRF in the absence of DNA and acts when bound to other proteins as a "portable" protein-protein interaction motif (Hill et al., 1993; Shore and Sharrocks, 1994). In the Elk1 B-box, five predominantly hydrophobic, residues are thought to constitute a hydrophobic face of an inducible !-

364

helix which forms the interaction surface for SRF (F i g . 2, F i g . 5 ; Ling et al., 1997). Such short motifs with significant hydrophobic content have been increasingly implicated in interactions between other eukaryotic transcription factors. Recently, the cognate binding surface for the B-box on SRF has been found to be small surfaceexposed hydrophobic patch (Ling et al., 1998), indicating that the interactions between the two proteins are likely to be predominantly hydrophobic in nature. The yeast protein MAT!2 interacts with the SRFrelated protein Mcm1 by using a short motif which shows some sequence similarity to the B-box. Hydrophobic residues in this motif play a key role in these interactions (Mead et al., 1996). The crystal structure of the ternary complex formed between Mcm1/MAT!2 on DNA was recently solved (Tan and Richmond, 1998) and revealed that the otherwise flexible N-terminal extension of the MAT!2 homeodomain, forms a "-hairpin that grips the Mcm1 surface through parallel "-strand hydrogen bonds and closely packed, predominantly hydrophobic, side chains. Several parallels therefore exist with the TCF-SRF interaction with the use of hydrophobic residues and a similar interaction surface on SRF/Mcm1. One key difference is the prediction of an !-helical binding motif on the TCFs but a "-strand structure on MAT!2. Interestingly, the crystal structure suggests that the interaction motif on MAT!2 can also adopt a mixed !helical/"-strand structure, raising the possibility that these short motifs might adopt different conformations in different contexts. Future structural and mutagenic studies on the TCF-SRF complex will resolve these issues. Another interesting implication from the model of the DNA-bound SAP-1/SRF complex is that the SAP-1 ETSdomain might make direct contacts with SRF. The Cterminal region of the !3 recognition helix of the SAP-1 ETS-domain is in position to interact with the N-terminus of the recognition helix of SRF. This interaction has been proposed to affect protein-DNA contacts mediated by the SAP-1 ETS-domain (Mo et al., 1998). Thus, this part of the ETS-domain appears to play a key role in mediating the biochemical properties of the TCFs, being involved in interactions with Pax-5 (see below), modifying their DNA binding specificity (see section III-B) and potentially, directly binding to SRF. Several ETS-domain proteins, including the TCFs Elk1 and SAP-2 can be recruited by Pax-5 into ternary complexes on the mb-1 promoter (Fitzsimmons et al., 1996). In contrast to the TCF-SRF complex, ternary complex assembly requires only the ETS DNA-binding domains and the Pax-5 paired box DNA-binding domain.

Gene Therapy and Molecular Biology Vol 3, page 365

F i g u r e 5 . Protein-protein interactions involving TCFs. (A) Interactions with SRF and MAP kinases. The sequences of the SRF-interaction motif (B-box) and MAP kinase interaction motif (D-domain) are shown. Key hydrophobic residues involved in these interactions are shown in red. (B) Interactions with Ids. The Ids bind to the TCFs via the ETS-domain and inhibit binding to the c-fos SRE. (C) Elk-1 and SAP-1 bind to CBP. CBP can subsequently affect downstream events by mechanisms such as interacting with the basal transcription machinery or by acetylating histones, coactivator proteins or other transcription factors.

365

Gene Therapy and Molecular Biology Vol 3, page 366 SAP-1 is not recruited by Pax-5, indicating that a specificity-determining mechanism exists. This differential ability to assemble in complexes with Pax-5 on the mb-1 promoter is defined by a single amino acid in the ETSdomain (V68 in SAP-1 and D69 in Elk-1) (Fitzsimmons et al., 1996). This amino acid also contributes significantly to the differential DNA-binding specificities of the TCFs Elk-1 and SAP-1 (see section III-B), indicating a key role for this amino acid in determining the specificity of complex formation by modifying either protein-DNA or protein-protein interactions.

B. Protein kinases. The TCF D-domain has been demonstrated to act as a binding site for different classes of MAP kinases (Fig. 5B; see section VI-B). In common with the B-box, this is a very short domain (18-20 amino acids; Fig. 4) acts as a portable protein-protein interaction motif and can direct the binding of MAP kinases to heterologous substrates. Interestingly, this binding site exhibits sequence similarity with several other transcription factors and scaffold proteins which bind to MAP kinases including ATF-2, ATFa, MEF2C, TCFs, Spi-B, c-Jun, NFAT4 and JIP-1 (see Yang et al., 1998b). One common motif found in all these sites is the central "LXL" motif. Indeed, in the case of the TCFs, the two leucine residues appear to be the most critical amino acids for binding to MAP kinases (Yang et al., 1998a, Yang et al., 1998b). This suggests that in common with the B-box motif, hydrophobic interactions may be major determinants of the interface with the MAP kinases. The situation is however clearly more complex as different TCFs are targeted by different kinases, suggesting a role for other residues in specificity determination. For example, the JNK binding epitope within the Elk-1 Ddomain overlaps the ERK binding site but extends further downstream (Yang et al., 1998). Further studies are however required to identify these specificity determinants. Hence, the D-domain does not represent a promiscuous MAPK targeting motif but allows discrimination between different classes of MAPKs.

C. Interactions with coactivators and corepressors. The conserved C-domain of the TCFs represents a phosphorylation-dependent transcriptional activation domain (reviewed in Treisman, 1994; see section IV-A). It is currently unclear how this domain mediates transcriptional activation. Moreover, the role of phosphorylation in activating this process is still unknown although one attractive hypothesis would be recruitment of coactivator proteins in a phosphorylation-dependent manner, either by contributing to or by unmasking an interaction surface. Support for the latter mechanism is 366

provided by the observation that Elk-1 undergoes a phosphorylation-dependent conformational change which might unmask an interaction surface (see section III-C). An increasing number of transcription factors use the coactivator, CBP/p300 to mediate the transactivation of RNA polymerase II (reviewed in Janknecht and Hunter, 1996). Indeed, both SAP-1 and Elk-1 bind directly to the N-terminus of CBP (between amino acids 451-721) in a phosphorylation-independent manner. However, functional cooperation between the two proteins requires the TCFs to become phosphorylated (Janknecht and Nordheim, 1996a, Janknecht and Nordheim, 1996b). Nucleosome positioning occurs on the c-fos promoter in vivo at a site between -90 and -280 relative to the transcriptional start site (Herrera et al., 1997). It is possible that the recruitment of CBP affects either the general transcription machinery or the nucleosomal structure, either within or downstream from the c-fos promoter and thereby facilitates gene activation (F i g . 5 C ). SAP-2 has been shown to act as a repressor in some contexts (Maira et al., 1996). In this case, the NID is sufficient to mediate this activity but it is currently unknown how this occurs and what relevance the binding of bHLH proteins has to its activity as a repression domain.

D. Helix-loop-helix proteins. Members of the TCF subfamily have recently been shown to interact with two different classes of proteins containing helix-loop-helix motifs. Firstly, SAP-2 has been shown to bind to bHLH proteins via the NID motif (Maira et al., 1996). However, the functional consequences of this interaction are unclear. Secondly, the Id HLH proteins can bind to all three TCFs (Yates et al, 1999). Id proteins are characterised by their ability to bind and subsequently inhibit the DNA binding activity of bHLH proteins and thereby inhibit differentiation (reviewed in Norton et al., 1998). In contrast to the bHLH-SAP-2 interaction, the HLH of Id2 interacts with the ETS DNA-binding domain of the TCFs. Binding of the Ids causes the dissociation of TCFs from low affinity ets motifs and from ternary complexes, thereby inactivating ternary complex-mediated transcription of immediate-early genes such as c-fos (see section VI-A). Regulation of immediate-early genes in this way may implicate the Ids in resetting or ‘dampening down’ the activity of the SRE in a negative feedback mechanism once cell cycle progression has been initiated. Indeed, consistent with this hypothesis, the Id proteins are themselves expressed from immediate-early responsive promoters and but their expression in response to serum stimulation is slightly delayed in comparison to other genes such as c-fos (Yates et al, 1999; reviewed in Norton et al., 1998).

Gene Therapy and Molecular Biology Vol 3, page 367

VI. Biological Roles. A. Immediate-early gene regulation and Cancer. The precise regulation of the basic cellular processes of proliferation, differentiation and migration is a fundamental aspect of development and malignant disease is associated with loss of control over such events. Hence, an understanding of the role that key proteins play in coordinating such behaviour during embryogenesis, as well as the processes by which these proteins are regulated will provide important insights into the molecular basis of cancer. Many of the ETS-domain transcription factors have been shown to have roles in embryonic development and have also been implicated in tumourigenesis (reviewed in Sharrocks et al., 1997 and Dittmer and Nordheim, 1998). MAP kinase cascades mediate cellular responses by the transduction of mitogenic and stress stimuli into the activation of immediate-early genes such as c-fos and egr-1. The resultant transcription factors can direct an appropriate cellular response via proliferative or differentiation pathways. As major players in the transduction of signals via the SRE, activated TCFs upregulate c-fos expression (reviewed in Treisman, 1994) and increasing evidence suggests a role for TCFs in the upregulation of other immediate-early genes such as egr-1 (Lim et al., 1998, Watson et al., 1997) and Pip92 (Chung et al., 1998). Permanent activation of the proto-oncoprotein ras, as seen in a high percentage of human tumours and implicated in tumour angiogenesis, is likely to lead to the constitutive activation of TCFs and other ETS proteins by MAPKs (reviewed in Dittmer and Nordheim, 1998). Thus, the TCFs are likely to play an important role in oncogenesis. The recent observation that SREs containing both SRF- and ets-binding motifs are activated during gastrulation represents the first example of the SRE being directly targeted by MAP kinase signalling pathways during development (Panitz et al., 1998). Activation via this composite element, and hence by implication through complexes containing ETS-domain proteins and SRF, was shown to be necessary and sufficient for Xegr-1 expression in response to MAP kinase pathway activation in Xenopus embryos. It is currently unknown whether TCFs or other ETS-domain proteins are directly involved in this process. However, based on work carried out in mammalian systems and the observation that TCF homologues are present in other lower vertebrates (Brown and Sharrocks, unpublished data), it is likely that homologues also exist in Xenopus and participate in the activation of Xegr-1 via the SRE.

B. Neuronal gene regulation 367

Whilst the expression levels of SRF mRNA and protein are similar in most cell lines tested, levels of the TCFs SAP-1 and Elk-1 are less homogeneous with high levels of one TCF being accompanied by low levels of the other (Price et al., 1995, Magnaghi-Jaulin et al., 1996). The expression of SAP-2 also differs according to cell-type (Price et al., 1995; Giovane et al., 1997). Interestingly, the relative expression levels of the murine SAP-2 homologue Net and its alternative splice form Net-b vary significantly in different cell lines and tissues. As the products of the two splice forms react differently to Ras-signalling pathways (see section V-A), this alternative splicing likely represents a key regulatory mechanism of TCF activity. In situ hybridisation studies have revealed that Elk-1 mRNA is expressed in various structures of the adult rat brain but is restricted to neuronal cell types, suggesting a role in the regulation of neuronal function (Sgambato et al., 1998). The Elk-1 protein is found in several intracellular compartments including the soma, dendrites and axon terminals in addition to the nucleus. In contrast, in fibroblasts, Elk-1 appears to be mainly nuclear when overexpressed (Janknecht and Nordheim, 1994) suggesting the possibility that a limiting amount of a cytoplasmic anchoring protein might exist which becomes saturated by Elk-1. Significantly, stimulation of glutamate receptors in the CNS induces ERK expression and a recent in vivo model of immediate-early gene induction has linked the stimulation of ERK to the phosphorylation and activation of Elk-1 and c-fos mRNA induction (Sgamboto et al., 1998). Immediate-early genes, and in particular c-fos, are also induced in astrocytes during early postnatal stages and at adult age in response to severe injury (axotomy, hypoxia and hypoglycaemia) (reviewed in Arenander and De Vellis, 1995) although Elk-1 is barely detectable in astrocytes in vivo (Sgamboto et al., 1998). As Elk-1 is strongly expressed in astrocytoma cell lines (Zinck et al., 1995) it is therefore possible that it may be transiently induced in astrocytes and play a role in immediate-early gene activation in this cell-type.

VII. Summary and perspectives. Studies of the TCFs have provided considerable insights into how other transcription factors belonging to the ETS-domain and other transcription factor families function at the molecular level. For example, the specificity of target promoter binding is determined by a combination of DNA-protein and protein-protein interactions. In the case of ETS-domain proteins, binding in combination with partner proteins appears to be the major mechanism of achieving promoter-specific binding. The elucidation of the structure of the DNA binding domains of several ETS-domain proteins, including SAP-

Yang et al: The ETS-domain transcription factors 1, has enhanced our knowledge of their mechanisms of DNA binding. Future studies of higher order complexes (eg TCF-SRF-DNA) will provide further significant insights into their molecular function. It is becoming increasingly clear that post-translational modifications such as phosphorylation and acetylation play major roles in regulating the activities of transcription factors. Again, studies on the TCFs have provided insights into how phosphorylation can affect DNA binding and transcriptional activation mediated by these proteins. In particular, the observation that phosphorylation triggers conformational changes which are transmitted intramolecularly between different domains, is likely to be a common mechanism shared by many transcription factors. Another emerging theme is that MAP kinase pathways actually represent compact modules in which the kinases are maintained in complexes to increase the specificity and speed of signal transduction. The observation that transcription factors such as the TCFs possess distinct kinase docking sites which contribute to the specificity of kinase action, further supports the hypothesis that proteinprotein interactions play a major role in determining the specificity of signal transduction in the cell. Furthermore, the ability of TCFs to be phosphorylated by different classes of MAP kinases, provides a mechanism by which different signals can be integrated into a single nuclear response by using a common substrate. Subsequent differences in the profiles of gene expression induced by mitogenic and stress stimulation must presumably result from their ability to phosphorylate other nuclear substrates which respond uniquely to different MAP kinase cascades. Finally, one of the major challenges of the future is to discover which genes and biological processes are regulated by individual transcription factors. This is particularly important in transcription factor families where different family members have presumably evolved to carry out specific tasks. The availability of a large body of data on the molecular action of the TCFs and numerous reagents is likely to ensure that future studies on the TCFs are likely to continue to contribute significantly to our understanding of eukaryotic transcription factor function.

Acknowledgements We are grateful to members of our laboratories for stimulating discussions and Amanda Greenall for comments on the manuscript. We would also like to thank Ronen Marmorstein for the discussion of data prior to publication. The work in the corresponding Author's laboratory is supported by grants from the Cancer Research Campaign [CRC], the Wellcome Trust, and the BBSRC. ADS is a Research Fellow of the Lister Institute of Preventative Medicine. 368

References Arenander, A. and De Vellis, J. (1 9 9 5 ) Early response gene expression in glial cells. In: Neurologia (Kettenman, H. and Ransom, B.R. eds), pp510-522. New Yok: Oxford UP. Bassuk, A. and Leiden, J. (1 9 9 6 ) The role of Ets transcription factors in the development and function of the mammalian immune system. Adv. Immunol. 64, 65-104. Batchelor, A.H., Piper, D.E., de la Brousse, F.C., McKnight, S.L. and Wolberger, C. (1 9 9 8 ) The structure of GABPa/b: an ETS-domain-ankyrin repeat heterodimer bound to DNA. S c i e n c e 279, 1037-1041. Cahill, M.A., Janknecht, R and Nordheim, A. (1 9 9 6 ) Signalling pathways: Jack of all cascades. Current B i o l o g y 6, 16-19. Cavigelli, M., Dolfi, F., Claret, F.X. and Karin, M. (1 9 9 5 ) Induction of c-fos expression through JNK-mediated TCF/Elk-1 phosphorylation. EMBO. J. 14, 5957-5964. Chung, K-C., Gomes, I., Wang, D., Lau, L.F. and Rosner, M.R. (1 9 9 8 ) Raf and fibroblast growth factor activate the serum response element of the immediate early gene pip92 by mitogen-activated protein kinase-independent as well as -dependent signalling pathways. M o l . C e l l . B i o l . 18, 2272-2281. Cohen, P. (1 9 9 7 ) The search for physiological substrates of MAP and SAP kinases in mammalian cells. Trends In C e l l . B i o l . 7, 353-361. Dai, T., Rubie, E., Franklin, C.C., Kraft, A., Gillespie, D.A.F., Avruch, J., Kyriakis, J.M. and Woodgett, J.R. (1 9 9 5 ) Stress-activated protein kinases bind directly to the $ domain of c-Jun in resting cells: implications for repression of c-jun function. O n c o g e n e 10, 849-855. Dalton, S. and Treisman, R. (1 9 9 2 ) Characterisation of SAP1, a protein recruited by serum response factor to the c-fos serum response element. C e l l 68, 597-612. Degnan, B.M., Degnan, S.M., Naganuma, T. and Morese, D.E. (1 9 9 3 ) The ets multigene family is conserved throughout the Metazoa. N u c l e i c A c i d s R e s . 21, 3479-3484. DĂŠrijard, B., Hibi, M., Wu, I-H., Barrett, T., Su, B., Deng, T., Karin, M. and Davis, R.J. (1 9 9 4 ) JNK1: A protein kinase stimulated by UV light and Ha-Ras that binds and phosphorylates the c-jun activation domain. C e l l 76 , 1025-1037. Dittmer, J. and Nordheim, A. (1 9 9 8 ) Ets transcription factors and human disease. Biochim. B i o p h y s . A c t a 1377, F1F11. Enslen, H., Raingeaud, J., and Davis, R.J. (1 9 9 8 ) Selective activation of p38 MAP kinase isoforms by the MAP kinase kinases MKK3 and MKK6. J . B i o l . C h e m . 273, 1741-1748. Fitzsimmons, D., Hodsdon, W., Wheat, W., Maira, S-M., Wasylyk, B. and Hagman, J. (1 9 9 6 ) Pax-5 (BSAP) recruits Ets proto-oncogene family proteins to form

Gene Therapy and Molecular Biology Vol 3, page 369 functional ternary complexes on a promoter. G e n e s D e v . 10, 2198-2211.

B-cell-specific

Frost, J.A., Steen, H., Shapiro, P., Lewis, T., Ahn, N., Shaw, P.E. and Cobb, M.H. (1 9 9 7 ) Cross-cascade activation of ERKs and ternary complex factors by Rho family proteins. EMBO J. 16, 6426-6438. Gille, H., Kortenjann, M., Strahl, T. and Shaw, P.E. (1 9 9 6 ) Phosphorylation-dependent formation of a quaternary complex at the c-fos SRE. M o l . C e l l . B i o l . 16, 10941102.

Janknecht, R. and Hunter, T. (1 9 9 7 b ) Activation of the SAP1a transcription factor by the c-Jun N-terminal kinase (JNK) mitogen-activated protein kinase. J . B i o l . Chem. 272, 4219-4224. Janknecht, R. and Nordheim, A. (1 9 9 2 ) Elk-1 protein domains required for direct and SRF-assisted DNA-binding. N u c l e i c A c i d s R e s . 20, 3317-3324. Janknecht, R. and Nordheim, A. (1 9 9 6 a ) Regulation of the cfos promoter by the ternary complex factor SAP-1a and its coactivator CBP. O n c o g e n e 12, 1961-1969.

Gille, H., Kortenjann, M., Thomae, O., Moomaw, C., Slaughter, C., Cobb, M.H. and Shaw P.E. (1 9 9 5 a ) Phosphorylation-dependent formation of a quaternary complex at the c-fos SRE. EMBO J. 14, 951-962.

Janknecht, R. and Nordheim, A. (1 9 9 6 b ) MAP kinasedependent transcriptional coactivation by Elk-1 and its cofactor CBP. B i o c h e m . B i o p y s . R e s . C o m m . 228, 831-837.

Gille, H., Strahl, T. and Shaw, P.E. (1 9 9 5 b ) Activation of ternary complex factor Elk-1 by stress-activated protein kinases. C u r r e n t B i o l o g y 5 , 1191-1200.

Janknecht, R., Zinck, R., Ernst, W.H. and Nordheim, A. (1 9 9 4 ) Functional dissection of the transcription factor Elk-1. O n c o g e n e 9, 1273-1278.

Giovane, A., Pintzas, A., Maira, S-M., Sobieszcuk, P. and Wasylyk, B. (1 9 9 4 ) Net, a new ets transcription factor that is activated by Ras. G e n e s D e v . 8, 1502-1513.

Kallunki, T., Deng, T., Hibi, M. and Karin, M. (1 9 9 6 ) c-Jun can recruit JNK to phosphorylate dimerization partners via specific docking interactions. C e l l 87, 929-939.

Giovane, A., Sobieszczuk, P., Ayadi, A., Maira, S-M. and Wasylyk, B. (1 9 9 7 ) Net-b, a Ras-insensitive factor that forms ternary complexes with the serum response element of the fos promoter. M o l . C e l l . B i o l . 17, 5667-5678.

Kallunki, T., Su, B. Tsigelny,I., Sluss, H.K., DĂŠrijard, B., Moore, G., Davis, R.J., and Karin, M. (1 9 9 4 ) JNK2 contains a specificity-determining region responsible for efficient c-jun binding and phosphorylation. G enes D e v . 8 , 2996-3007.

Graves, B.J. and Petersen, J.M. (1 9 9 8 ) Specificity within the ets family of transcription factors. In Adv. Cancer Res., G. van de Woude and G. Klein, Eds. (Academic Press, San Diego, 1998). Gupta, S., Barrett, T., Whitmarsh, A.J., Cavanagh, J., Sluss, H.K., DĂŠrijard, B. and Davis, R.J. (1 9 9 6 ) Selective interaction of JNK protein kinase isoforms with transcription factors. EMBO J. 15, 2760-2770. Herrera, R.E., Nordheim, A. and Stewart, A.F. (1 9 9 7 ) Chromatin structure analysis of the human c-fos promoter reveals a centrally positioned nucleosome. Chr o m o so m a 106, 284-292. Hill, C.S., Marais, R., John, S., Wynne, J., Dalton, S. and Treisman, R. (1 9 9 3 ) Functional analysis of a growth factor-responsive transcription factor complex. C e l l 73, 395-406. Hipskind, R.A., Baccarini, M. and Nordheim, A. (1 9 9 4 b ) Transient activation of RAF-1, MEK and ERK2 coincides kinetically with TCF phosphorylation and IEG promoter activity in vivo. M o l . C e l l . B i o l . 14, 6219-6231. Hipskind, R.A., Buscher, D., Nordheim, A. and Baccarini, M. (1 9 9 4 a ) Ras/MAP kinase-dependent and -independent signalling pathways target distinct ternary complex factors. G e n e s D e v . 8, 1803-1816. Janknecht, R. and Hunter, T. (1 9 9 6 ) A growing coactivator network. Nature 383, 22-23. Janknecht, R. and Hunter, T. (1 9 9 7 a ) Convergence of MAP kinase pathways on the ternary complex factor SAP-1a. EMBO J. 16, 1620-1627.

369

Karim, F.D., Urness, L.D., Thummel, C.S., Klemsz, M.J., McKercher, S.R., Celada, A., Van Beueren, C., Maki, R.A., Gunther, C.V., Nye, J.A. and Graves, B.J. (1 9 9 0 ) The ETS-domain: a new DNA-binding motif that recognises a purine-rich core DNA sequence. G e n e s D e v . 4, 1451-1453. Karin, M. (1 9 9 4 ) Signal transduction from the cell surface to the nucleus through the phosphorylation of transcription factors. C u r r . O p i n . C e l l B i o l . 6, 415-424. Kortenjann, M., Thomae, O. and Shaw, P.E. (1 9 9 4 ) Inhibition of v-raf-dependent c-fos expression and transformation by a kinase-defective mutant of the mitogen activated protein kinase Erk2. M o l . C e l l . B i o l . 14, 4815-4824. Latinkic, B.V., Zeremski, M. and Lau, L.F. (1 9 9 6 ) Elk-1 can recruit SRF to form a ternary complex upon the serum response element. N u c l e i c A c i d s . R e s . 24, 13451351. Laudet, V., Niel, C., Duterque-Coquillaud, M., Leprince, D. and Stehelin, D. (1 9 9 3 ) Evolution of the ets gene family. B i o c h e m . B i o p h y s . R e s . C o m m . 190, 8-14. Lefstin, J.F. and Yamamoto. K.R. (1 9 9 8 ) Allosteric effect of DNA on transcriptional regulator. Nature 392, 885-888. Lim, C.P., Jain, N. and Cao, X. (1 9 9 8 ) Stress-induced immediate-early gene, egr-1, involves activation of p38/JNK1. O n c o g e n e 16, 2915-2926. Ling, Y., Lakey, J.H., Roberts, C.E. and Sharrocks, A.D. (1 9 9 7 ) Molecular characteristics of the B-box proteinprotein interaction motif of the ETS-domain transcription factor Elk-1. EMBO. J. 16, 2431-2440.

Yang et al: The ETS-domain transcription factors Ling, Y., West, A.G., Roberts, E.C., Lakey, J.H. and Sharrocks, A.D. (1 9 9 8 ) Interaction of transcription factors with SRF: identification of the Elk-1 binding surface. J . B i o l . C h e m . 273, 10506-10514. Lopez, M., Oettgen, P., Akbarali, Y., Fendorter, U. and Liberman, T.A. (1 9 9 4 ) ERP, a new member of the ets transcription factor/oncoprotein family: cloning, characterization, and differential expression during Blymphocyte development. M o l . C e l l . B i o l . 14, 32923309.

vivo expression and regulation of Elk-1, a target of the extracellular-regulated kinase signalling pathway, in the adult rat brain. J . N e u r o s c i e n c e 18, 214-226. Sharrocks, A.D., Brown, A.L., Ling, Y. and Yates, P.R. (1 9 9 7 ) The ETS-domain transcription factor family. I n t . J . B i o c h e m . C e l l B i o l . , 29, 1371-1387. Shore, P. and Sharrocks, A.D. (1 9 9 5 ) The ETS-domain transcription factors Elk-1 and SAP-1 exhibit differential DNA binding specificities. N u c l e i c Acids R e s . 23, 4698-4706.

Magnaghi-Jaulin, L., Masutani, H., Lipinski, M. and HarelBellan, A. (1 9 9 6 ) Analysis of SRF, SAP-1 and Elk-1 transcripts and proteins in human cell lines. FEBS Lett. 391, 247-251.

Shore, P. and Sharrocks, A.D. (1 9 9 4 ) The transcription factors Elk-1 and serum response factor interact by direct protein-protein contacts mediated by a short region of Elk-1. M o l . C e l l . B i o l . 14, 3283-3291.

Maira, S-M., Wurtz, J-M. and Wasylyk, B. (1 9 9 6 ) Net (ERP/SAP2), one of the Ras-inducible TCFs, has a novel inhibitory domain with resemblance to the helix-loophelix motif. EMBO J. 15, 5849-5865.

Shore, P., Bisset, L., Lakey, J., Waltho, J.P., Virden, R. and Sharrocks, A.D (1 9 9 5 ) Characterization of the Elk-1 ETS DNA-binding domain. J . B i o l . Chem. 270, 58055811.

Masutani, H., Magnaghi-Jaulin, L., Ali, S.A.S., Groisman, R., Robin, P. and Harel-Bellan, A. (1 9 9 7 ) Activation of the c-fos SRE through SAP-1a. O n c o g e n e 15, 16611669.

Shore, P., Whitmarsh, A.J., Bhaskaran, R., Davis, R.J., Waltho, J.P. and Sharrocks, A.D. (1 9 9 6 ) Determinants of DNA-binding specificity of ETS-domain transcription factors. M o l . C e l l . B i o l . 16, 3338-3349.

Mead, J., Zhong, H., Acton, T.B. and Vershon, A.K. (1 9 9 6 ) The yeast !2 and Mcm1 proteins interact through a region similar to a motif found in homeodomain proteins of higher eukaryotes. M o l . C e l l . B i o l . 16, 2135-2143.

Sluss, H.K., Barrett, T., Derijard, B. and Davis, R.J. (1 9 9 4 ) Signal transduction by tumor necrosis factor mediated by JNK protein kinases. M o l . C e l l . B i o l . 14 , 83768384.

Mo, Y., Vaessen, B., Johnston, K. and Marmorstein, R. (1 9 9 8 ) Structures of SAP-1 bound to DNA sequences from the E74 and c-fos promoters provide insights into how ETS proteins discriminate between related DNA targets. M o l e c u l a r C e l l 8, 210-212.

Strahl, T., Gille, H. and Shaw, P.E. (1 9 9 6 ) Selective response of ternary complex factor Sap-1a to different mitogenactivated protein kinase sub-groups. P r o c . N a t l . A c a d . S c i . 93, 11563-11568.

Norton, J.D., Deed, R.W., Craggs, G. and Sablitzky, F. (1 9 9 8 ) Id helix-loop-helix proteins in cell growth and differentiation. T r e n d s C e l l . B i o l . 8, 58-65.

Sugimoto, T., Stewart, S. and Guan, K-L. (1 9 9 7 ) The calcium/calmodulin-dependent protein phosphatase calcineurin is the major Elk-1 phosphatase. J . B i o l . Chem. 272, 29415-29418.

Panitz, F., Krain, B., Hollemann, T., Nordheim, A. and Pieler, T. (1 9 9 8 ) The Spemann organiser-expressed zinc finger gene Xerg-1 responds to MAP kinase/Ets-SRF signal transduction pathway. EMBO J. 17, 4414-4425.

Tan, S. and Richmond, T.J. (1 9 9 8 ) Crystal structure of the yeast MATa2/MCM1/DNA ternary complex. Nature 391, 660-666

Price, M.A., Cruzalegui, F.H. and Treisman, R. (1 9 9 6 ) The p38 and ERK MAP kinase pathways co-operate to activate ternay complex factors and c-fos transcription in response to UV light. EMBO. J. 15, 6552-6563. Price, M.A., Rogers, A.E. and Treisman, R. (1 9 9 5 ) Comparative analysis of the ternary complex factors Elk1, SAP-1a and SAP-2 (ERP/NET). E M B O J . 14, 25892601. Rao, V.N., Huebner, K., Isobe, M., Ab-Rushdi, A., Croce, C.M. and Reddy, E.S.P. (1 9 8 9 ) elk, tissue-specific etsrelated genes on chromosomes X and 14 near translocation breakpoints. S c i e n c e 244, 66-70. Robinson, M.J. and Cobb, M.H. (1 9 9 7 ) Mitogen-activated protein kinase pathways. C u r r e n t O p i n i o n i n C e l l B i o l o g y 9, 180-186. Sgamboto, V., Vanhoutte, P., Pages, C., Rogard, M., Hipskind, R., Besson, M-J. and Caboche, J. (1 9 9 8 ) In

370

Treisman, R. (1 9 9 4 ) Ternary complex factors: growth regulated transcriptional activators. C u r r . O p i n . G e n . D e v . 4, 96-101. Treisman, R. (1 9 9 6 ) Regulation of transcription by MAP kinase cascades. C u r r . O p i n . C e l l B i o l . 8, 205-215. Treisman, R., Marais, R. and Wynne, J. (1 9 9 2 ) Spatial flexibility in complexes between SRF and its accessory proteins. EMBO J. 11, 4631-4640. Vossler, M.R., Yao, H., York, R.D., Pan, M-G., Rim, C.S. and Stork, P.J.S. (1 9 9 7 ) cAMP activates MAP kinase and Elk-1 through a B-Raf- and Rap1-dependent pathway. C e l l 89, 73-82. Wang, X.S., Diener, K., Manthey, C.L., Wang, S-W., Rosenzweig, B., Bray, J., Delaney, J., Cole, C.N., ChanHui, P-Y., Mantlo, N., Lichenstein, H.S., Zukowski, M. and Yao, Z. (1 9 9 7 ) Molecular cloning and characterization of a novel p38 mitogen-activated protein kinase. J . B i o l . C h e m . 272, 23668-23674.

Gene Therapy and Molecular Biology Vol 3, page 371 Watson, D.K., Robinson, L., Hodge, D.R., Kola, I., Papas, T.K. and Seth, A. (1 9 9 7 ) FLI1 and EWSFLI1 function as ternary complex factors and ELK1 and SAP1a function as ternary and quaternary complex factors on the Egr1 promoter serum response elements. O n c o g e n e . 14, 213221.

Yang, S-H., Whitmarsh, A.J., Davis, R.J. and Sharrocks, A.D. (1 9 9 8 b ) Differential targeting of MAP kinases to the ETS-domain transcription factor Elk-1. EMBO J. 17, 1740-1749.

Whitmarsh, A.J. and Davis, R.J. (1 9 9 6 ) Transcription factor AP-1 regulation by mitogen-activated protein kinase signal transduction pathways. J . M o l . M e d . 74, 589607.

Yates PR, Atherton GT, Deed RW, Norton JI and Sharrocks AD (1 9 9 9 ) Id helix-loop-helix proteins inhibit nucleoprotein complex formation by the TCF ETS-domain transcription factors. EMBO J. In press.

Whitmarsh, A.J., Shore, P., Sharrocks, A.D. and Davis, R.J. (1 9 9 5 ) Integration of MAP kinase signal transduction pathways at the serum response element. S c i e n c e 269, 403-407.

Zinck, R., Cahill, M.A., Kracht, M., Sachsenmaier, C., Hipskind, R.A. and Nordheim, A. (1 9 9 5 ) Protein synthesis inhibitors reveal differential regulation of mitogen-activated protein kinase and stress-activated protein kinase pathways that converge on Elk-1. M o l . C e l l . B i o l . 15, 4930-4938.

Whitmarsh, A.J., Yang, S-H., Su, M. S-S., Sharrocks, A.D. and Davis, R.J. (1 9 9 7 ) Role of p38 and JNK MAP kinases in the activation of ternary complex factors. M o l . C e l l . B i o l . 17, 2360-2371. Yang, S-H., Whitmarsh, A.J., Davis, R.J. and Sharrocks, A.D. (1 9 9 8 a ) The Elk-1ETS-domain transcription factor

371

contains a MAP kinase targeting motif. M o l . C e l l . B i o l . 18, 710-720.

Gene Therapy and Molecular Biology Vol 3, page 373 Gene Ther Mol Biol Vol 3, 373-378. August 1999.

Transcriptional activation of the ras oncogenes and implications of BRCA1 in the cell cycle regulation through p53 checkpoint Review article

D.A. Spandidos, G. Sourvinos, S. Miyakis Laboratory of Virology, Medical School, University of Crete, Heraklion 71409, Crete, Greece __________________________________________________________________________________________________ Corresponding author: Professor D.A. Spandidos, Medical School, University of Crete, PO Box 1393, Heraklion 71409, Crete, Greece. Tel/Fax: +301 7226469; E-mail: spandido@hol.gr Key words: ras genes, transcriptional regulation, p53, steroid hormone receptors,BRCA1, cell cycle regulation. Received: 8 August 1998; accepted: 15 August 1998

Summary Aberrant expression of ras genes has been recognized in several human cancers and is associated with the development of the disease. Thus, revealing the mechanisms that regulate the expression o f ras genes is critical for understanding their role in the process of tumorigenesis. Transcriptional regulation of the H-ras gene, involves nuclear factors recognizing elements in the promoter region of the gene and hormones; so far, a glucocorticoid response element and a p53 element have been identified. Alternative splicing and specific methylation patterns may regulate the expression of ras genes as well. Altered expression of ras genes has been detected in a variety of human tumours. Differential expression of the ras f a m i l y g e n e s i n b r e a s t c a n c e r h a s s h o w n o v e r e x p r e s s i o n o f a l l three members of ras genes. A significant correlation of overexpression of ras genes and stage of the disease was also observed suggesting that aberrant expression of the ras genes may be an initial event in breast cancer oncogenesis. Overexpression of Ras p21 protein has been detected in human endometrial and ovarian tumours, due to elevated p53 protein binding on the p53 element of the c-H-ras gene, suggesting that p53 protein participates i n the development o f human gynecological neoplasias through aberrant transcriptional regulation of the H-ras proto-oncogene. I n v e s t i g a t i o n o f t h e B R C A 1 e x p r e s s i o n l e v e l s i n r e l e v a n c e w i t h t h e expression l e v e l s o f p53, mdm-2 and p21 WAF1/CIP1 genes, implicated in cell cycle progression, revealed combined alterations of these genes in sporadic breast cancer specimens, indicating that loss of function of B R C A 1 m a y arrest the cell cycle through p53 checkpoint.

exogenous signals that are essential for the regulation of vital cell functions (Lowy and Willumsen, 1991). The interchange of the p21 proteins between “on” (GTP-bound) and “off” (GDP-bound) position allows them to operate as switches in the cytoplasmic relay of external growth and differentiation signals (Hwang and Cohen, 1997). Interaction of p21 with the Raf oncoprotein results in activation of a cascade of serine/threonine kinases. The intensity and duration of this event strongly contributes to the regulation of cell differentiation and division (Avruch et al., 1994, Marschall, 1995).

I. Introduction The ras family genes are among the most well studied and frequently detected genes participating in oncogenesis of human tumors. Three ras proto-oncogenes have been so far identified in the mammalian genome: H-ras 1, K-ras 2 and N-ras (Barbacid, 1987). They all encode similar GTPbinding proteins of the same molecular weight (21kDa), termed p21 proteins. These cellular components are associated with the inner face of the plasma membrane, playing, thus, a major role in the transduction of

373

Spandidos et al: Transcriptional regulation of ras genes and BRCA1-dependent cell cycle progression

II. Activation of the ras family genes

correlated with increased levels of intranuclear wild-type p53 and with elevated p53 binding to the H-ras element. These findings provide evidence for implication of the Hras gene in gynaecological cancer through aberrant regulation of the p53 protein.

ras proto-oncogenes are normally expressed in most human tissues. Mechanisms of activation of these genes are frequently observed in human tumors and mainly include point mutations as well as overexpression of wildtype p21. Point mutations, occurring mainly at codons 12, 13 and 61 of the three ras genes, lead to the transformation of the proto-oncogene to an activated oncogene. The mutant p21 loses its ability to become inactivated and, therefore, stimulates cell growth and differentiation constitutively (Kiaris and Spandidos 1995). A plausible explanation for the tendency of ras mutations to affect selectively distinct genetic sites is that the cells bearing codons 12, 13 and 61 mutations have a proliferative advantage (Barbacid, 1987). These genetic alterations, consequently, are selected within the cell population as compared to other mutations in different sites of the ras genes.

On the other hand, mutated p53 protein was unable to bind to the H-ras element. In some of these cases overexpression of the ras p21 protein was detected, as well as in certain tumors showing similar levels of p53 binding to the H-ras, compared to normal tissue. It has been suggested that regulation of H-ras expression in such cases is effected by alternative mechanisms. Elevated steroid receptor binding to corresponding elements of the H-ras gene has been observed in more than 90% of gynaecological tumor-normal pairs tested (Zachos et al., 1996). Correlation between ras p21 overexpression and estrogen receptor levels in ovarian tumors was also demonstrated (Scambia et al., 1993), moreover, estrogens participate in regulation of the p53 levels (Hurd et al., 1997). Thus, it has been suggested that regulation of H-ras expression by steroid hormone receptors and modulation of the H-ras gene transcription levels by p53 are inter-related factors in the activation of the gene in human hormonedependent tumors.

Aberrant expression of the ras genes has been recognized in several human cancers and is associated with the development of the disease (Zachos and Spandidos 1997). It is the result of alterations in transcriptional regulation of the genes which quantitatively contribute to the malignant phenotype. In vitro experiments have shown that overproduction of even the normal Ras protein is sufficient to give a transforming potential to cultured cells (Spandidos and Wilkie, 1984).

IV. Transcriptional activation of ras genes in human breast cancer

The mechanisms of regulation of ras oncogene expression have been widely studied in H-ras protooncogene. These comprise regulatory elements in the promoter region, regulation by of H-ras expression by intronic sequences or by sequences in the 3’ end of the gene, as well as interaction of H-ras with the p53 tumorsuppressor protein or with steroid hormone receptors. Furthermore, DNA methylation and alternative splicing have been demonstrated to affect H-ras gene expression. The proposed models for the regulation of the expression of these genes have been recently reviewed (Zachos and Spandidos 1998).

Breast cancer is the most common type of cancer in women. It has been suggested that the participation of ras genes in breast carcinogenesis consists mainly of expressional activation, since ras mutations have been reported infrequently in breast cancer (Rochlitz et al., 1989). Elevated levels of the p21 proteins - encoded by ras genes - compared to the respective normal tissues have been detected by immunohistochemical methods in 6571% of cases (Rochlitz et al., 1989; Spandidos et al., 1992). Recently we examined 27 human sporadic breast cancer specimens analysing the expression levels of tumor ras mRNA, compared to respective adjacent normal tissue, using the reverse transcription-polymerase chain reaction (RT-PCR) technique (our unpublished results). Eighteen of the 27 (67%) tumors examined showed transcriptional activation of at least one of the ras family genes. N-ras exhibited overexpression in 10 (37%), K-ras in 9 (33%) and H-ras in 10 (37%) of the 27 tumor samples examined. Nine cases (33%) did not present overexpression of any member of the ras family genes. Moreover, six (22%) did not show mutational activation of ras genes. On the other hand, three samples (11%) exhibited transcriptional activation of all the three ras genes, while ten tumors (37%) overexpressed only one, and five (18.5%) overexpressed two of the ras genes.

III. Transcriptional regulation of the Hras gene by the p53 tumor-suppressor protein and by steroid hormone receptors It has been shown that the H-ras proto-oncogene contains in the first intron a p53 binding element that acts as a transcriptional enhancer in vitro in the presence of a wild-type p53 protein (Spandidos et al., 1995). Elevated binding of overexpressed wild-type p53 protein has been detected in more than 40% of endometrial and ovarian tumors, compared to the respective normal tissue (Zachos and Spandidos 1998b). Furthermore, overexpression of the ras p21 protein in tumor versus the adjacent normal tissue

374

Gene Therapy and Molecular Biology Vol 3, page 375 Our experimental data confirm the high incidence of ras overexpression reported previously for this type of malignancy (Miyakis et al, submitted). Furthermore, it is shown that overproduction of p21 ras is not due to the activation of only one member of the ras family, but all three ras genes are activated, in various combinations. This ascertainment is enhanced by the lack of correlation between the expression levels of any particular ras gene and the clinicopathological parameters of the patients.

On the other hand, ras overexpression is considered as a favourable marker for neuroblastomas (Tanaka et al., 1991, Spandidos et al., 1992). Increased levels of ras gene expression have been detected in precursors of gastrointestinal neoplasias (Barret’s mucosa, colorectal polyps) probably contributing to the malignant transformation of these lesions (Abdelatif et al., 1991, Spandidos et al., 1994). The exhibition of higher Ras p21 protein levels in 30% of a series of thyroid adenomas, compared to normal tissue (Papadimitriou et al., 1988) implicates that elevated ras expression may be involved in the conversion of these lesions to carcinomas.

Our finding, that ras mRNA overexpression is associated with tumors at an earlier stage, is in agreement with observations from related studies on other types of cancer (Kiaris et al., 1995, Vageli et al., 1996). Therefore, aberrant expression of ras genes may be an initial event in the breast cancer oncogenetic process. Despite the fact that such alterations seem to play an important role in the primary stages of the disease, ras expression is not necessarily required for the maintenance of the transformed phenotype, since mRNA levels tend to decrease in tumors of a more advanced stage. This is further supported by the rather heterogeneous staining of ras p21, encountered previously in metastatic breast cancer tissue (Fromowitz et al., 1987).

The three ras genes exhibit a high incidence of altered expression in numerous human tumors. In addition, a variety of correlations between these genetic changes and major clinicopathological parameters of most types of malignancies have been reported. These features could become useful tool in prognosis and -sometimes- in early diagnosis of human cancer. Understanding the regulatory mechanisms of transcription in these genes, creates new perspectives in the future development of effective molecular strategies for therapy.

VI. Loss of function of BRCA1 may activate the p53 checkpoint

Clinical data available were insufficient for accurate correlation of ras expression with the patients’ outcome. Immunohistochemically detected ras overexpression has not been found to be significantly associated with time to progression and overall survival (Archer et al., 1995). Nevertheless it has been postulated that oncogene coexpression may serve as a prognostic correlate for recurrence and survival (Bland et al., 1995, Jiang et al., 1997).

Extensive studies have revealed underlying mechanisms of p53 growth suppression and cell cycle regulation. The acidic domain in the amino-terminal region of p53 has transactivation activity (Farmer et al., 1992, Fields et al., 1990). In addition, p53 binds preferentially to specific DNA sequences (El-Deiry et al., 1992, Funk et al., 1992). p53 may also exert growth suppression by binding to the MDM-2 protein. The interaction between MDM-2 and p53 may modulate the activity of p53 (Momand et al., 1992). p53 suppresses growth by transcriptional activation of p21WAF1/CIP1 which inhibits Cdk2, a cell division cyclindependent kinase (El-Deiry et al., 1993). Inhibition of Cdk2 stops cell division and inhibits DNA synthesis. These data demonstrate the link between p53 and the cell cycle and suggest a possible pathway which may be altered during carcinogenesis (Harper et al., 1993).

V. Transcriptional aberrations of ras genes in other types of human cancer. Apart from its involvement in the oncogenesis of hormone-related human neoplasias overexpression of ras genes has been reported in various human tumors. Quantitative molecular biology methods are being applied for the detection of gene expression; these are accomplished at the RNA level using the RT-PCR technique or the RNA spot hybridization analysis, while Western blotting and immunohistochemistry are able to define intracellular levels of the p21 protein. The frequency of the detection of ras overexpression varies widely with the stage in the oncogenetic process in which these genetic alterations are believed to be involved (Zachos and Spandidos 1997).

Several known tumor suppressor genes interact with or negatively regulate the cell cycle machinery (Sherr et al., 1995); BRCA1 may play an important role in this process. Several properties of BRCA1 and p53 suggest that these two proteins may functionally interact. Both p53 and BRCA1 are tumor suppressor genes that have been implicated in DNA damage response and repair pathways (Levine 1997, Scully et al., 1997, Brugarolas et al., 1997). Both p53 and BRCA1 are physically altered by the cellular response to DNA damage: p53 by stabilization and BRCA1 by hyperphosphorylation (Scully et al., 1997, Kastan et al., 1991). Both p53 and BRCA1 can activate

High incidence of augmented ras gene expression has been recorded for head and neck carcinomas, as well as for lung and endometrial tumors; ras overexpression appears in a later oncogenetic stage in lung and endometrial tumors.

375

Spandidos et al: Transcriptional regulation of ras genes and BRCA1-dependent cell cycle progression p21WAF1/CIP1 as a common target gene (El-Deiry et al., 1993, Somasundaram et al., 1997).

breast cancer and response to first line hormonal therapy. Br J Cancer 72, 1259-1266.

Investigation of the BRCA1 expression levels in relation to the expression levels of p53, mdm-2 and p21WAF1/CIP1 genes implicated in the cell cycle progression, revealed combined alterations of these genes in sporadic breast cancer specimens (Sourvinos and Spandidos, 1998). Specimens expressing BRCA1 up to 2.7-fold lower than normal tissues, overexpressed p21 and mdm-2 at the same time, whereas specimens expressing more than 2.7-fold reduced BRCA1 mRNA levels expressed p21 at high levels and mdm-2 was unchanged. These results indicate that certain levels of BRCA1, even reduced levels of BRCA1, are sufficient to upregulate p21, when p53 activity is inhibited by its negative regulator, the mdm-2. p53 expression levels were unaffected, although expression of mdm-2, a gene coding for a negative regulator of p53 activity, was elevated in some cases. The latter indicates a critical role for p53, not at the expression level but in the activity of the gene. Furthermore, specimens exhibiting more than 2.7-fold reduced BRCA1 levels overexpressed p21 while mdm-2 expression was normal, suggesting that p21 transcriptional activation is due to p53 activity in cases with dramatically decreased BRCA1 expression.

Avruch J, Zhang X and Kyriakis JM (1 9 9 4 ) Raf meets Ras: completing the framework of a signal transduction pathway. TIBS 19, 279-283. Barbacid M (1 9 8 7 ) ras genes. A n n u R e v B i o c h e m 56, 779-827. Bland KI, Konstadoulakis MM, Vezeridis MP and Wanebo HJ (1 9 9 5 ) Oncogene protein co-expression. Value of Ha-ras , c-myc , c-fos , and p53 as prognostic discriminants for breast carcinoma. Ann Surg 221, 706-720. Brugarolas J and Jacks T (1 9 9 7 ) Double indemnity: p53, BRCA and cancer. Nature Med 3, 721-722. El-Deiry WS, Kern SE, Pietenpol JA, Kinzler KW and Vogelstein B (1 9 9 2 ) Definition of a consensus binding site for p53. Nat Genet 1, 45-49,. El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW and Vogelstein B (1 9 9 3 ) WAF1, a potential mediator of p53 tumor suppressioin. C e l l 75, 817-825. Farmer G, Bargonetti J, Zhu H, Friedman P, Prywes R and Prives C ( 1 9 9 2 ) Wild-type p53 activates transcription in vitro. Nature 358, 83-86. Fields S and Jang SK (1 9 9 0 ) Presence of a potent transcription activating sequence in the p53 protein. S c i e n c e 249, 1046-1049.

These findings, obtained in human sporadic breast tumours, support the model proposed here which has been derived from experiments in mouse embryos during embryogenesis; according to this model BRCA1, BRCA2 and Rad51 act as a complex to repair damaged DNA (F i g . 1). Abrogation of BRCA1 function, either through mutation or via decreased expression could lead to the accumulation of DNA damage and the subsequent activation of a checkpoint mechanism, resulting in p53 activation and the upregulation of the p53-responsive gene, p21. Increased p21 levels inhibit cyclin-dependent kinases, resulting in cell cycle arrest. In case of overexpression of the MDM-2, the negative regulator of p53, p21 can be transcriptionally activated, directly, by BRCA1, when it is present in sufficient amounts.

Fromowitz FB, Viola MV, Chao S, Oravez S, Mishriki Y, Finkel G, Grimson R and Lundy J (1 9 8 7 ) ras p21 expression in the progression of breast cancer. Hum P a t h o l 18, 1268-1275. Funk WD, Pak DT, Karas RH, Wright WE and Shay JW (1 9 9 2 ) A transcriptionally active DNA-binding site for human p53 protein complexes. M o l C e l l B i o l 12, 28662871. Harper JW, Adami GR, Wei N, Keyomarsi K and Elledge SJ (1 9 9 3 ) The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. C e l l 75, 805816. Hurd C, Khattree N, Dinda S, Alban P and Moudgil VK (1 9 9 7 ) Regulation of tumor suppressor proteins, p53 and retinoblastoma, by estrogen and antiestrogens in breast cancer cells. O n c o g e n e 15, 991-995.

References

Hwang D-Y and Cohen JB (1 9 9 7 ) A splicing enhancer in the 3’-terminal c-H-ras exon influences mRNA abundance and transforming activity. J Virol 71, 6416-6426.

Abdelatif OM, Chandler FW, Mills LR, McGuire BS, Pantazis CG and Barrett JM (1 9 9 1 ) Differential expression of cmyc and H-ras oncogenes in Barrett's epithelium. A study using colorimetric in situ hybridization. A r c h P a t h o l Lab Med 115, 880-885.

Jiang M, Shao ZM, Wu J, Lu JS, Yu LM, Yuan JD, Han QX, Shen ZZ and Fontana JA (1 9 9 7 ) p21/waf1/cip1 and mdm2 expression in breast carcinoma patients as related to prognosis. Int J Cancer 74, 529-534.

Archer SG, Eliopoulos A, Spandidos DA, Barnes D, Ellis IO, Blamey RW, Nicholson RI and Robertson JFR (1 9 9 5 ) Expression of ras p21, p53 and c-erb B-2 in advanced

376

Gene Therapy and Molecular Biology Vol 3, page 377

Figure 1. Activation of the p53 checkpoint in response to loss of function of BRCA1.

Kastan MB, Onyekwere O, Sidransky D, Vogelstein B and Craig RW (1 9 9 1 ) Participation of p53 protein in the cellular response to DNA damage. Cancer R e s 51, 6304-6311.

Levine AJ (1 9 9 7 ) p53, the cellular gatekeeper for growth and division. C e l l 88, 323-331.

Kiaris H and Spandidos DA (1 9 9 5 ) Mutations of ras genes in human tumours (Review). Int J Oncol 7, 413-421.

Marshall MS (1 9 9 5 ) Ras target proteins in eucaryotic cells. FASEB J 10, 625-630.

Kiaris H, Spandidos DA, Jones AS, Vaughan ED and Field JK (1 9 9 5 ) Mutations, expression and genomic instability of the H-ras proto-oncogene in squamous cell carcinomas of the head and neck. Br J Cancer 72, 123-128.

Momand J, Zambetti GP, Olson DC, George D and Levine AJ (1 9 9 2 ) The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. C e l l 69, 1237-1245.

Lowy DR and Willumsen BM (1 9 9 3 ) Function and regulation of ras. A n n u R e v B i o c h e m 62, 851-891.

377

Spandidos et al: Transcriptional regulation of ras genes and BRCA1-dependent cell cycle progression Miyakis S, Sourvinos G and Spandidos DA (1 9 9 8 ) Differential expression and mutations of the ras family genes in human breast cancer. B i o c h e m B i o p h y s R e s Commun 251, 609-612.

oncogene by the p53 tumor suppressor. I n t J O n c o l 7, 1029-1034. Spandidos DA, Karaiossifidi H, Malliri A, Linardopoulos S, Vassilaros S and Field JK (1 9 9 2 ) Expression of Ras Rb1 and p53 proteins in human breast cancer. A n t i c a n c e r R e s 12, 81-90.

Papadimitriou K, Yiagnisis M, Tolis G and Spandidos DA (1 9 8 8 ) Immunohistochemical analysis of the ras oncogene protein product in human thyroid neoplasms. Anticancer Res 8, 1223-1227.

Tanaka T, Slamon DJ, Shimada H, Shimoda H, Fujisawa T, Ida N and Seeger RC (1 9 9 1 ) A significant association of Haras p21 in neuroblastoma cells with patient prognosis. A retrospective study of 103 cases. Cancer 15, 1296-1302.

Rochlitz CF, Scott GK, Dodson JM, Liu E, Dollbaum C, Smith HS and Benz CC (1 9 8 9 ) Incidence of activating ras oncogene mutations associated with primary and metastatic human breast cancer. C a n c e r R e s 49, 357360.

Vageli D, Kiaris H, Delakas D, Anezinis P, Cranidis A. and Spandidos DA (1 9 9 6 ) Transcriptional activation of Hras, K-ras, and N-ras proto-oncogenes in human bladder tumors. Cancer Lett 107, 241-247.

Scambia G, Catozzi L, Panici PB, Ferrandina G, Coronetta F, Barazzi R, Baiocchi G, Uccelli L, Piffanelli A and Mancuso S (1 9 9 3 ) Expression of ras oncogene p21 protein in normal and neoplastic ovarian tissues: correlation with histopathologic features and receptors for estrogen, progesterone and epidermal growth factor. Am J Obstet G y n e c o l 168, 71-78.

Zachos G and Spandidos DA (1 9 9 7 ) Expression of ras protooncogenes: regulations and implication in the development of human tumors. Crit R e v O n c o l Hematol 26, 65-75. Zachos G and Spandidos DA (1 9 9 8 ) Transcriptional regulation of the H-ras1 proto-oncogene by DNA binding proteins: mechanisms and implications in human tumorigenesis. Ge ne T he r M ol B iol 1, 629-639.

Scully R, Chen J, Ochs RL, Keegan K, Hoekstra M, Feunteun J, Livingston DM (1 9 9 7 ) Dynamic changes of BRCA1 subnuclear location and phosphorylation state are initiated by DNA damage. C e l l 90, 425-435.

Zachos G and Spandidos DA (1 9 9 8 b ) Transcriptional regulation of the c-H-ras1 gene by the p53 protein is implicated in the development of human endometrial and ovarian tumours. O n c o g e n e 16, 3013-3017.

Sherr CJ and Roberts JM (1 9 9 5 ) Inhibitors of mammalian G1 cyclin-dependent kinases. G e n e s D e v 9, 1149-1163. Sobol H, Stoppa-Lyonnet D, B De Paillerets, Peyrat JP, Guinebretiere J-M, Jacquemier J, Eisinger F and Birnbaum D (1 9 9 7 ) BRCA1-p53 relationship in hereditary breast cancer. Int J Oncol 10, 349-353.

Zachos G, Varras M, Koffa M, Ergazaki M, and Spandidos DA (1 9 9 6 ) Glucocorticoid and estrogen receptors have elevated activity in human endometrial and ovarian tumors as compared to the adjacent normal tissues and recognize sequence elements of the H-ras proto-oncogene. J a p n J Cancer Res 87, 916-922.

Somasundaram K, Zhang H, Zeng Y-X, Hourvas Y, Peng Y, Zhang H, Wu GS, Licht JD, Weber BL and El-Deiry W (1 9 9 7 ) Arrest of the cell cycle by the tumour-suppressor BRCA1 requires the CDK-inhibitor p21 WAF1/CiPI. Nature 389, 187-190. Sourvinos G and Spandidos DA (1 9 9 8 ) Decreased BRCA1 expression levels may arrest the cell cycle through activation of p53 checkpoint in human sporadic breast tumours. B i o c h e m B i o p h y s R e s C o m m u n 245, 7580. Spandidos DA and Kerr IB (1 9 9 4 ) Elevated expression of the human ras oncogene family in premalignant and malignant tumours of the colorectum. B r J C a n c e r 49, 681-688. Spandidos DA and Wilkie NM (1 9 8 4 ) Malignant transformation of early passage rodent cells by a single mutated human oncogene. Nature 310, 469-475. Spandidos DA, Arvanitis D and Field JK (1 9 9 2 ) ras p21 expression in neuroblastomas and ganglioneuroblastomas: correlation with patients’ prognosis. Int J Oncol 1, 53-58. Spandidos DA, Zoumpourlis V, Zachos G, Toas SH and Halazonetis TD (1 9 9 5 ) Specific recognition of a transcriptional element within the human H-ras proto-

378

Gene Therapy and Molecular Biology Vol 3, page 379 Gene Ther Mol Biol Vol 3, 379-385. August 1999.

Nuclear receptor coactivators as potential therapeutical targets: the HATs on the mouse trap Review Article

Arndt Benecke and Hinrich Gronemeyer Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, BP 163, 67404 ILLKIRCH Cedex, France __________________________________________________________________________________ Corresponding author: Arndt Benecke, Tel: +33 (0)3 8865 3457; Fax: +33 (0)3 8865 3201; E-mail: arndt@igbmc.u-strasbg.fr K e y w o r d s : histone acetyl transferase, retinoic acid receptor, nuclear receptor, drug discovery, transcriptional regulation, coactivator A b b r e v i a t i o n s : HAT, histone acetyl transferase; NR, nuclear receptor; LBD, ligand binding domain; RAR, retinoic acid receptor; RXR, retinoic X receptor; PPAR, peroxisome proliferator activated receptor; SRC1, steroid receptor coactivator 1; NIDDM, non-insulin dependent diabetes mellitus; AF-2, activation function-2; R I P 1 4 0 , receptor interacting protein 140; TIF2, transcription intermediary factor 2; RAC3, receptor associated coactivator 3; A R A 70 , androgen receptor activator 70; GRIP1, glucocorticoid receptor interacting protein 1; TR, thyroid receptor; ER, estrogen receptor; AIB1, amplified in breast cancer 1; CBP, CREB binding protein; P/CAF, p300/CBP associated factor; GNAT , GCN5 related N-acetyltransferase; AcCoA, acetyl-coenzyme A. Received: 17 November 1998; accepted: 17 November 1998 (contributed by Hinrich Gronemeyer)

Summary The recent past has seen an immense burst i n our understanding o f nuclear receptor (NR) signaling. Key achievements have been the structure determination of the ligand binding domain and the identification of coregulators which mediate the transcriptional effects of NRs. Both types of studies have now converged on the description of the NR coactivator interface at the atomic l e v e l , w h i c h , t o g e t h e r w i t h t h e e l u c i d a t i o n o f t h e structure o f two coactivator related histone acetyl transferases (HATs) points towards previously unknown targets for drug design potentially leading to novel types of non-ligand antagonists of NR function.

the treatment of several endocrine cancers or as contraceptives, and recent improvements in the specificity and efficacy predict that NR ligands will acquire an even broader spectrum of applications in medicine, like for example in the treatment of osteoporosis via differential regulation of the activity of the estrogen receptors alpha and/or beta. Furthermore, through the insights into orphan nuclear receptor signaling, novel targets for synthetic ligands emerge. For instance it may be possible to treat obesity and NIDDM (non-insulin dependent diabetes mellitus) via regulating the activity of the nuclear receptor PPAR (Mukherjee et al., 1997), and there is still a great number of orphan receptors with potentially very important physiological functions (Mangelsdorf et al., 1995; Chambon, 1996) which have not yet been analysed in detail on a functional level. NR activity is negatively regulated by corepressors and positively by coactivators. A common mechanism of NR coregulator action seems to be the covalent modification of

I. Introduction Nuclear receptors are ligand inducible transcription factors that are implicated in virtually any genetic program, such as development, differentiation, control of proliferation, homeostasis and apoptosis (for recent review on NR signaling see Mangelsdorf et al., 1995; Chambon, 1996; Heine and Gronemeyer, 1998). They act directly or indirectly on the expression of a variety of target genes by (i ) modifying the chromatin environment of the promoter, (i i ) altering the activity of the basal transcription machinery, and (i i i ) mutually interfering with other transcriptional signaling pathways. As key regulatory molecules they have attracted much attention for therapeutic treatment, also in view of the fact that misregulation of nuclear receptor signaling is apparently directly related to the generation of a number of diseases (Carapeti et al., 1998; Chen et al., 1997; Fenaux et al., 1997; Taki et al., 1997). Synthetic ligands that partially or completely block NR activity are already routinely used in 379

Benecke and Gronemeyer: The HATs on the mouse trap histones (reviewed in: Torchia et al., 1998), although other non-histone targets may exist as well (Imhof and Wolffe, 1998; Imhof et al., 1997; Gu and Roeder 1997) (Figure 1). NR-recruited coactivator complexes contain HAT activities that acetylate the nucleosomal template thereby overcoming a barrier to enhanced transcription, while corepressor complexes are capable to reverse this effect by deacetylating and consequently condensing the chromatin template (Kuo and Allis, 1998; Imhof and Wolffe, 1998). Through direct and indirect, often ligand-independent, interactions with basal transcription factors, NRs are thought to modulate the activity of the pre-initiation complexes (Mangelsdorf et al., 1995; Chambon, 1996). Positive or negative interference with other signal transduction pathways, referred to as signal transduction crosstalk, is still poorly understood but may result from several different mechanisms such as coregulator sharing/squelching, direct interactions between the involved transcription factors and regulation of posttranslational modification. The cloning and characterization of NR coregulators has shed some light on the mechanisms by which NRs exert their multiple effects on chromatin and, possibly, the basal transcription machinery (Chambon, 1996; Heine and Gronemeyer, 1998). With detailed mechanistic information at hand new dimensions of specific therapeutic interference into the highly complex phenomenon of NR transcription regulation emerge. The very recent structural definition of the NR coactivator interface and the insights we gain form the elucidation of the structures of histone acetyl transferases will undoubtedly nucleate new drug development approaches.

(1998). Residues adjacent to the core LxxLL motif indeed encode domains with specificity for different NRs and the definition of specificity-conferring residues, directly Nterminal to the LxxLL motif (M. Parker, pers. communication), substantiates this observation. The emerging picture here points towards two levels of NR coactivator specificity. First, highly related coactivators from the TIF2/SRC1/RAC3 family display NR preferences, since for example the mouse homologue of TIF2 binds well to the androgen receptor, whereas SRC1 binds very poorly (Ding et al. 1998). Second, different LxxLL motifs found on one coactivator are only partially redundant for their binding to a specific NR. The NR box 2 of TIF2 is probably the major contact site for the estrogen receptor since mutations in this site have the greatest effects on transcriptional stimulation and binding (Voegel et al., 1998; Leers et al., 1998). Though the other NR boxes are able to compensate in vitro to some extent loss of a critical motif (Voegel at al., 1998; Leers et al., 1998), it is conceivable to assume that in vivo redundancy is much less important. Another level of complexity is introduced by the fact that different isoforms of the same coactivator (SRC1a vs. SRC1e) differ in their ability to activate a single NR (Kalkhoven et al. 1998). These isoform-specific effects are attributable to a fourth NR box found in SRC1a which is not present in the differentially spliced form SRC1e, or other coactivators from this family. In turn, SRC1e is the major SRC1 isoform to mediate thyroid hormone response (Hayashi et al. 1997). Concluding, it seems very likely that the redundancy of LxxLL boxes reflects the need to accommodate in vivo different NRs with different variations of the same theme. Thinking along these lines it is even tempting to speculate that coactivators are not at all promiscuous, as suggested by initial in vitro binding studies, in their choice of NRs. This view is supported by the cloning of an apparently androgen receptor specific coactivator (ARA70, Yeh and Chang, 1996), and helps to explain the multiplicity of different coactivators for NRs. If coactivators have NR specificity, why then do they need several LxxLL motifs? An appealing answer to this question comes from the structure of the complex between the PPAR" LBD homodimer and a SRC1 peptide encompassing two of the three NR boxes (Nolte et al. 1998). The complex has a molar composition of two PPAR" LBDs per SRC1 peptide molecule, and the crystal structure reveals that each LxxLL motif participates in the binding to one of the LBDs. Both motifs make identical contacts to the hydrophobic clefts of the respective monomers (Nolte et al. 1998). Although there is yet no confirmation that these data indeed reflect simultaneous binding of entire coactivators to both subunits of a NR homo- or heterodimer, the crystallographic data are supported by in vitro binding studies indicating that, in solution, SRC1 is also able to bind both partners in a retinoic acid receptor (RAR)-retinoic X receptor (RXR) heterodimer (Westin et al., 1998).

II. NR coactivator specificity The cloning and analysis of NR coregulators was followed by definition of the short and structurally defined coactivator signature LxxLL (where x is any amino acid) motifs (or NR boxes) embedded in a short !-helical peptide, which are necessary and sufficient for ligand dependent interaction with the transcriptional activation function-2 (AF-2) located in the ligand binding domain (LBD) of NRs (Torchia et al. 1998). The LBDs of different NRs share a common fold that has been compared to a mouse trap since the binding of a specific ligand results in a conformational change (springs the trap) involving repositioning of several helices to form the coactivator binding site (see Moras and Gronemeyer, 1998). The fact that some coactivators contain multiple LxxLL motifs (up to nine in RIP140), all of which appear to be functional at least in terms of in vitro binding to NRs, had brought up the intriguing question of whether this multiplicity reflects redundancy or a means of conferring specificity to the interface. Initial observations that different coactivators display NR preferences (Voegel et al., 1998; Ding et al., 1998; Kalkhoven et al., 1998; Hayashi et al., 1997; Leers et al., 1998) have gained ground through the studies on the binding of NR boxes to holo LBDs by Darimont et al. 380

Gene Therapy and Molecular Biology Vol 3, page 381

Figure 1. A schematic representation of a nuclear receptor homo- or hetero-dimer bound to DNA with a coactivator or coactivator-complex and potential interactions with the basal machinery and the nucleosomal template. Note that either one or both partners in the dimer might be ligand-bound and contribute to coactivator recruitment. The LxxLL nuclear receptor interaction motifs in the coactivator are indicated, as well as a histone acteyltransferase active center, which might target histones, general transcription factors or other non-histone targets. For simplicity direct interactions between the nuclear receptors and the basal machinery as well as the possibility of multiple coactivator/-complexes have been omitted. For further explanations refer to the text.

III. The structure of the interface

The view that NR coactivators might contact both partners in NR homo- or hetero-dimers, which represent the biologically active states of NRs, helps to explain remaining obstacles associated with NR transactivation. Westin et al. (1998) and Nolte et al. (1998) for example provide with their findings a possible explanation of how RXR ligands can potentiate the effect of RAR ligands in the RAR-RXR heterodimer, namely by cooperative recruitment of coactivators. Here, the RXR partner, in presence of a specific ligand, further stabilizes the interaction with the RAR-recruited coactivator. However, it has to be kept in mind, that there is no in vivo evidence for cooperative recruitment of coactivators to heterodimers where both partners are ligand-bound. The possibility remains that RXR recruits a second coactivator to the heterodimer. Furthermore, Westin et al. (1998) still assume in their model that RXR in the heterodimer is neither able to bind its cognate ligand nor to recruit a coactivator unless RAR is also ligand-bound. This is in disagreement with several studies that unequivocally demonstrate that RXR is indeed able to bind its ligand and subsequently recruit a coactivator to a heterodimer with an unliganded RAR in vitro (Chen et al., 1998; Kersten et al., 1996; AB, unpublished).

Until recently, it was unclear whether the holo (ligandbound) LBD surface provides multiple coactivator binding sites allowing cooperative binding of two or more NR boxes present in one single coactivator (di-, tri-partite interface) or even the cooperative recruitment of two coactivator molecules at the time. Three recently solved crystal structures of holo-NR LBD coactivator peptide complexes [SRC1-PPAR" , (Nolte et al. 1998); TIF2/GRIP1-TR#, (Darimont et al. 1998); and TIF2/GRIP1-ER!, (Geoffrey Greene, pers. Communication)] display a ratio of one NR box per LBD, making the hypothesis of a multipartite interface between NRs and coactivators unlikely. This view is further supported by scanning surface mutagenesis of the TR# LBD, and subsequent monitoring of coactivator (in this case SRC1 and GRIP1, the mouse homologue of TIF2) binding and transactivation properties (Feng et al. 1998). This study delineates a hydrophobic cleft on the holo LBD that is in part composed of residues from helix H12 which had already been known to undergo a major transitional relocation upon ligand binding (“springing the trap�) and in numerous mutational studies been implicated 381

Benecke and Gronemeyer: The HATs on the mouse trap in transactivation (Moras and Gronemeyer 1998). In this case ligand binding is directly linked to coactivator binding since the repositioning of helix H12 on the one hand is induced by the ligand which makes contacts to several residues in this helix, and on the other hand then contributes with two more residues to the hydrophobic cleft that accommodates the coactivator. From the mutational analysis it is furthermore apparent, that the hydrophobic cleft is flanked by charged residues at either side making up a so-called “charge clamp� which probably contributes to (i ) defining the orientation of the NR box towards the LBD and (i i ) making contacts to residues outside of the core LxxLL motif to achieve the above mentioned specificity for different NR boxes (Feng et al. 1998). In both co-crystal structures the LxxLL motifs are accommodated in a hydrophobic cleft on the respective LBDs that is very similar to the one defined by the surface scanning mutagenesis, highlighting once more the validity of the mouse trap model based on the common fold of NR LBDs (Moras and Gronemeyer 1998). The fact that the LBD cavity is almost completely filled by one coactivator NR box argues against the possibility that multiple NR box LBD contacts are made. Furthermore, it is obvious from both structural studies that the charged residues at either end of the hydrophobic cleft are contacting additional residues in the NR box. In this respect it will be very interesting to see whether a good correlation between the N-terminal specificity encoding amino acids (Malcom Parker, pers. communication) and the identity of the charged residues at the corresponding parts on the LBD surface can be made. In good agreement with the mutational studies defining the LxxLL motif (reviewed in: Torchia et al. 1998), is also the fact that the leucine residues of the NR box indeed contribute in two ways to its identity. First, they confer structural identity to the encompassing peptide resulting in the formation of this amphipathic !-helix, and second, they form the hydrophobic surface that matches the hydrophobic cleft on the NR LBD. It is also worth to note that only residues in the immediate vicinity of the NR boxes make contacts to the LBDs in the PPAR" co-crystal (Nolte et al. 1998). This reduces the importance of other residues in the nuclear receptor interacting domains to providing an overall structure that is promiscuous to the formation of the interface. Noteworthy, the intervening residues between the two NR boxes in the PPAR" SRC1 co-crystal seem to have little structural identity (Nolte et al. 1998), which is in keeping with the low sequence conservation found between members of this coactivator family. In contrast, the spacing between the different NR boxes is rather conserved, making the finding that SRC1 can bind to both homo- or hetero-dimeric NRs likely to be a general feature of this coactivator family.

synthetic NR ligands. For the efficient development of synthetic NR ligands two new aspects should be considered: First, although it has been known for some time that the natural ligands for NRs make, among others, also direct contacts to the helix H12 of the LBD, the fact that helix H12 contributes directly to the coactivator binding surface, and that this surface potentially displays coactivator specificity, should be incorporated in the rational of ligand design. For instance, one approach would be to screen for synthetic partial agonists that influence only slightly helix H12 positioning; these molecules, instead of abrogating coactivator binding altogether, would change coactivator affinity. In contrast to classical antagonists which are non-permissive for the relocation of helix H12 on the LBD surface, and therefore for the formation of the coactivator binding site, such ligands might shift the relative affinities for different coactivators. This might prove effective for disorders where a complete block of NR activity is not feasible since it generates secondary effects that might promote disease but where it is desirable to down-regulate NR activity. An example could be to target breast cancer with an amplification/ overexpression of the AIB1/RAC3 coactivator (Anzick et al. 1997) with such ligands. If high amounts of AIB1 trigger unusual estrogen receptor activity in these cells, it would certainly be advantageous to selectively lower the affinity of estrogen receptor for AIB1, while preserving other functions performed by the receptor in combination with other coactivators, e.g. the antiproliferative effects probably exerted via the general mediator CBP (reviewed in Heine and Gronemeyer, 1998). The ultimate aim would be to obtain synthetic ligands that selectively impair specific combinations of NR coactivator pairs. This would allow to overcome a limitation that classical synthetic ligands for NRs have: they are specific for one receptor but restricted in their action to the receptor molecule itself and are not coactivator-specific. Second, a search for combinatorial ligands could be prompted by the fact that one coactivator molecule seems to contact both partners in a heterodimer composed of receptors with different natural ligands. It will be of interest to develop combinations of ligands that do not have major effects on a single NR activity, but when administered together block coactivator association to a specific heterodimer combination; for example, such drugs could block RAR-RXR heterodimer signaling but not RXR homodimers or RXR heterodimers with partners other than RAR. Both strategies for ligand development account for the increasing combinatorial complexity in NR signaling by aiming to restrict their action to very specific NR functions. Interestingly, considerable progress has been made on the development of RXR dimer-selective ligands that specifically affect either PPAR-RXR heterodimers (Mukherjee et al. 1997) or RXR homodimers (Lala et al. 1996).

IV. Implications for NR ligand design The structural clues obtained in these studies will certainly have an impact on the development of new 382

Gene Therapy and Molecular Biology Vol 3, page 383 1998; Bayle and Crabtree 1997), are of high importance. As mentioned in the Introduction, NRs recruit HAT activities in order to render the chromosomal target promoter prone to transcriptional activation. Recently, the structures of two histone N-acetyltransferase enzymes that are highly related to the nuclear receptor coactivators P/CAF and GCN5 have been solved (Dutnall et al., 1998; Wolf et al., 1998). Both the yeast HAT1 and the Sarratia marcescene aminoglycoside 3-N-acetyltransferase share common features encoded in a canonical GCN5-related Nacetyltransferase (GNAT) core motif (Dutnall et al., 1998; Wolf et al. 1998; Neuwald and Landsman, 1997). This motif confers cofactor (coenzyme A) binding, encodes the active center of the enzymes and contributes to substrate recognition and binding. Differences between members of the HAT family in the core region reflect probably different substrate specificities (reviewed by Kuo and Allis, 1998). The alignment of the GNAT motif with the structures obtained in both crystallographic approaches allows now a rigid analysis of structure & function relationships of different residues within the enzymatic core of different HATs. Together with biochemical analysis this will allow a more precise definition of HAT substrates and their mode of recognition. Since acetyl coenzyme A (AcCoA) is an abundant metabolic key intermediate it might be difficult to specifically interfere with HAT function at the level of competitive AcCoA inhibitors. Efforts could be directed to developing either allosteric effectors of HAT activities, or suicide inhibitors that covalently modify the active center. The structural information will be of great help to accomplish these tasks. However, most promising seems to be the approach of blocking specific HAT activities by targeting the substrate recognition site with competitive inhibitors. Based on their data as well as molecular modeling Dutnall and coworkers suggested a complementary fit model for the recognition of histone tails by the HAT enzyme (Dutnall et al., 1998). As discussed above for the NR coactivator interface, peptidomimetics might also lead to the development of synthetic inhibitors of HATs with high bioactivity able to selectively associate with the substrate recognition site of the enzyme thereby blocking its function. Of special interest is the fact that different HAT enzymes have different substrates (Kuo and Allis, 1998), some of nonhistone nature (Bayle and Crabtree 1997), and display considerable substrate specificity (Kuo and Allis, 1998; Imhof and Wolffe, 1998). For example, p53 is regulated in it's DNA binding activity through acetylation by p300 (Gu and Roeder 1997) and the general transcription factors TF IIE# and TF IIF can be acetylated by P/CAF or p300 (Imhof et al. 1997). Again, in view of the differential effects that closely related coactivator-HATs like CBP and p300 (Yao et al. 1998; Kawasaki et al. 1998) or CBP and P/CAF (Puri et al. 1998) have, it will be of prime importance to define drugs that are capable of selectively interfering with a specific function of a given HAT rather than blocking the whole enzyme family.

V. The NR coactivator interface as drug target The recent gain in understanding of NR coactivator function at the molecular level sets the grounds for new strategies of pharmacological interference within NR signaling pathways. The fact that the interface between NR and coactivator is composed of very defined features, namely an amphipathic !-helical chain containing the LxxLL motif of the coactivator and a hydrophobic cleft plus “charge clamp� on the surface of the LBD of the NR (Darimont et al. 1998; Nolte et al. 1998; Feng et al. 1998), together with the fact that the coactivator !-helix is structured in solution (as little as 8 amino acids are sufficient for ligand-dependent interaction with NRs, reviewed in Torchia et al., 1998) raise the possibility to disrupt such interactions with small synthetic molecules. On basis of the structural information that we have now, it is feasible to screen combinatorial peptides containing the core LxxLL motif for high affinity binding to the hydrophobic groove on the LBD. These peptides might prove to be effective in inhibiting the interaction between NRs and coactivators in the cell and therefore abrogate NR transactivation. In this respect, since the LxxLL motif confers structural identity to the encompassing peptide, it is tempting to envision hybrid proteins containing LxxLL motifs that act as dominant negative coactivator mimicries while being more stable than peptides in vivo. Alternatively, peptidomimetics might open a path leading to overcoming obstacles associated with the delivery and stability of peptides in the organism (Kieber-Emmons et al. 1997). The fact that combinatorial compound libraries can be created by medicinal chemistry and molecular biology make it feasible to attempt generating drugs that mimic the structure of the LxxLL peptides, and therefore prove effective in blocking the NR coactivator interface. In this respect it is interesting to note that peptidomimetics have already been used very successfully to mimic a peptide hormone (Livnah et al. 1996).

VI. The shape of the HAT Thinking further along these lines, one could ask whether it is feasible to interfere with the activity of a given coactivator for NRs rather than regulating the activity of the NR itself. Given that NR coactivators seem to play a pivotal role also in non-NR driven transcription e.g. the general mediator and nuclear receptor coactivator CBP/p300 (Yao et al. 1998), inactivating the whole molecule e.g. by antisense approaches will have deleterious side effects on other signaling pathways. Hence, it seems reasonable to concentrate on either blocking specific interaction domains (as discussed in the paragraph above), or blocking the endowed enzymatic activities that most coactivators posses. In this respect obviously the histone acetyl transferase (HAT) activities, that stimulate transcription (Martinez-Balbas et al. 1998; Zhang et al. 1998) not only by acetylating histones but also non-histone protein targets (reviewed in Kuo and Allis

383

Benecke and Gronemeyer: The HATs on the mouse trap

VII. Concluding remarks

References Anzick, S. L., Kononen, J., Walker, R. L., Azorsa, D. O., Tanner, M. M., Guan, X. Y., Sauter, G., Kallioniemi, O. P., Trent, J. M. and Meltzer, P. S. (1 9 9 7 ). AIB1, a steroid receptor coactivator amplified in breast and ovarian cancer. S c i e n c e . 277, 965-958.

The importance of nuclear receptors in cell fate has been elucidated and understood for a long time. The fact that nuclear receptors represent “master genes� makes them attractive targets for drug research in disease therapy. Since their mode of action is highly complex the emerging details from investigations on nuclear receptor coregulators not only decipher an amazing transcription network that controls spatial and temporal expression of target genes, but also promote the identification of potential candidate functions for pharmacological interference. To this end several new options have been sketched here, focussing on the coactivator rather than on the nuclear receptor itself. If such approaches prove to be effective this might mark the beginning of a post-ligand era for NR drugs.

Bayle, J. H. and Crabtree, G. R. (1 9 9 7 ). Protein acetylation: more than chromatin modification to regulate transcription. C h e m B i o l . 4, 885-8. Carapeti, M., Aguiar, R. C., Goldman, J. M. and Cross, N. C. (1 9 9 8 ). A novel fusion between MOZ and the nuclear receptor coactivator TIF2 in acute myeloid leukemia. B l o o d . 91, 3127-33. Chambon, P. (1 9 9 6 ). A decade of molecular biology of retinoic acid receptors. FASEB J. 10, 940-54. Chen, Z. P., Iyer, J., Bourguet, W., Held, P., Mioskowski, C., Lebeau, L., Noy, N., Chambon, P. and Gronemeyer H. (1 9 9 8 ). Ligand- and DNA-induced dissociation of RXR tetramers. J M o l B i o l . 275, 55-65.

Acknowledgements The authors wish to apologize to those colleagues whose work could not be cited due to space constraints, and thank their colleagues at IGBMC for helpful discussions. A.B. is recipient of a Marie-Curie long term fellowship from the European Commission (ERBFMBICT961269). Work at the Gronemeyer laboratory is supported by grants from CNRS, INSERM, HUS, the EC Biomed programme (BMH4-96-0181) and Bristol-Myers-Squibb.

Chen, Z., Wang, Z. Y. and Chen, S. J. (1 9 9 7 ). Acute promyelocytic leukemia: cellular and molecular basis of differentiation and apoptosis. Pharmacol Ther. 76, 141-149. Darimont, B. D., Wagner, R. L., Apriletti, J. W., Stallcup, M. R., Kushner, P. J., Baxter, J. D., Fletterick, R. J. and Yamamoto, K. R. (1 9 9 8 ). Structure and Specificity of Nuclear Receptor-Coactivator interactions. G e n e s D e v . 12, 3343-3356. Ding, X. F., Anderson, C. M., Ma, H., Hong, H., Uht, R. M., Kushner, P. J. and Stallcup, M. R. (1 9 9 8 ). Nuclear receptor-binding sites of coactivators glucocorticoid receptor interacting protein 1 (GRIP1) and steroid receptor coactivator 1 (SRC-1): multiple motifs with different binding specificities. M o l E n d o c r i n o l . 12, 302-313.

Note added in proof: While this manuscript was in press the work cited as Darimont et al, 1998 has been published. The full citation is as follows:

Dutnall, R. N., Tafrov, S. T., Sternglanz, R. and Ramakrishan, V. (1 9 9 8 ). Structure of the Histone Acetyltransferase Hat1: A Paradigm for the GCN5-related N-acetyltransferase Superfamily. C e l l . 94, 427-438.

Darimont BD, Wagner RL, Apriletti JW, Stallcup MR, Kushner PJ, Baxter JD, Fletterick RJ, Yamamoto KR (1 9 9 8 ) Structure and specificity of nuclear receptor-coactivator interactions. G e n e s D e v 12, 3343-3356.

Fenaux, P., Chomienne, C. and Degos, L. (1 9 9 7 ). Acute promyelocytic leukemia: biology and treatment. S e m i n O n c o l . 24, 92-102.

In the same issue another study dealing with the NR coactivator interface has been published, that substantiates and extends the idea of specificity determining residues adjacent to the LxxLL motif:

Feng, W., Ribeiro, R. C., Wagner, R. L., Nguyen, H., Apriletti, J. W., Fletterick, R. J., Baxter, J. D., Kushner, P. J. and West, B. L. (1 9 9 8 ). Hormone-dependent coactivator binding to a hydrophobic cleft on nuclear receptors. S c i e n c e . 280, 1747-1749.

McInerney EM, Rose DW, Flynn SE, Westin S, Mullen TM, Krones A, Inostroza J, Torchia J, Nolte RT, Assa-Munt N, Milburn MV, Glass CK, Rosenfeld MG (1 9 9 8 ) Determinants of coactivator LXXLL motif specificity in nuclear receptor transcriptional activation. G e n e s D e v 12, 3357-3368.

Gu, W. and Roeder, R. G. (1 9 9 7 ). Activation of p53 sequence-specific DNA binding by acetylation of the p53 C-terminal domain. C e l l . 90, 595-606.

Furthermore, another non-histone target for the acetyltransferase activity of CBP has been identified, further substantiating the suggestions made in the final paragraph of this paper:

Hayashi, Y., Ohmori, S., Ito, T. and Seo, H. (1 9 9 7 ). A splicing variant of Steroid Receptor Coactivator-1 (SRC1E): the major isoform of SRC-1 to mediate thyroid hormone action. B i o c h e m B i o p h y s R e s C o m m u n . 236, 83-87.

Munshi N, Merika M, Yie J, Senger K, Chen G, Thanos D (1 9 9 8 ) Acetylation of HMG I(Y) by CBP turns off IFN beta expression by disrupting the enhanceosome. M o l C e l l 2, 457-467 .

Heine, M. J. S. and Gronemeyer, H. (1 9 9 8 ). Nuclear receptors as targets for drug design: new options and old challenges. In: Molecular Basis of Sex Hormone Receptor Function, Ernst Schering Research Foundation. Editors:

384

Gene Therapy and Molecular Biology Vol 3, page 385 Gronemeyer, H., Fuhrmannn, U., Parczyk, K., Springer P r e ss , Springer Verlag Berlin Heidelberg, pp2-41.

Neuwald, A. F. and Landsman, D. (1 9 9 7 ). GCN5-related histone N-acetyltransferases belong to a deverse superfamily that includes the yeast SPT10 protein. T r e n d s B i o c h e m S c i . 22, 154-155.

Imhof, A. and Wolffe, A. P. (1 9 9 8 ). Gene control by targeted histone acetylation. Curr Biol. 8, 422-424. Imhof, A., Yang, X. J., Ogryzko, V. V., Nakatani, Y., Wolffe, A. P. and Ge, H. (1 9 9 7 ). Acetylation of general transcription factors by histone acetyltransferases. Curr B i o l . 7, 689-692.

Nolte, R. T., Wisely, G. B., Westin, S., Cobb, J. E., Lambert, M. H., Kurokawa, R., Rosenfeld, M. G., Willson, T. M., Glass, C. K., Milburn, M. V. (1 9 9 8 ) Ligand binding and co-activator assembly of the peroxisome proliferatoractivated receptor ". Nature. 395, 137-143.

Kalkhoven, E., Valentine, J. E., Heery, D. M. and Parker, M. G. (1 9 9 8 ). Isoforms of steroid receptor co-activator 1 differ in their ability to potentiate transcription by the oestrogen receptor. Embo J. 17, 232-243.

Puri, P. L., Sartorelli, V., Yang, X.-J., Hamamori, Y., Ogryzko, V. V., Howard, B. B., Kedes, L., Wang, J. Y. J., Graessmann, A., Nakatani, Y. and Levrero, M. (1 9 9 7 ). Differential Roles of p300 and PCAF Acetyltransferases in Muscle Differentiation. M o l C e l l . 1, 35-45.

Kawasaki, H., Eckner, R., Yao, T. P., Taira, K., Chiu, R., Livingston, D. M. and Yokoyama, K. K. (1 9 9 8 ). Distinct roles of the co-activators p300 and CBP in retinoic-acidinduced F9-cell differentiation. Nature. 393, 284-289.

Taki, T., Sako, M., Tsuchida, M. and Hayashi, Y. (1 9 9 7 ). The t(11; 16)(q23; p13) translocation in myelodysplastic syndrome fuses the MLL gene to the CBP gene. B l o o d . 89, 3945-3950.

Kersten, S., Dawson, M. I., Lewis, B. A. and Noy, N. (1 9 9 6 ). Individual subunits of heterodimers comprised of retionoc acid and retinoid X receptors interact with their ligands independently. B i o c h e m i s t r y 35, 3816-3824.

Torchia, J., Glass, C. and Rosenfeld, M. G. (1 9 9 8 ). Coactivators and co-repressors in the integration of transcriptional responses. C u r r O p i n C e l l B i o l 10, 373-383.

Kieber-Emmons, T., Murali, R. and Greene, M. I. (1 9 9 7 ). Therapeutic peptides and peptidomimetics. Curr Opin B i o t e c h n o l . 8, 435-441.

Voegel, J. J., Heine, M. J., Tini, M., Vivat, V., Chambon, P. and Gronemeyer, H. (1 9 9 8 ). The coactivator TIF2 contains three nuclear receptor-binding motifs and mediates transactivation through CBP binding-dependent and -independent pathways. EMBO J. 17, 507-519.

Kuo, M.-H. and Allis, C. D. (1 9 9 8 ). Roles of histone acetyltransferases and deacetylases in gene regulation. B i o E s s a y s . 20, 615-626. Lala, D. S., Mukherjee, R., Schulman, I. G., Koch, S. S., Dardashti, L. J., Nadzan, A. M., Croston, G. E., Evans, R. M. and Heyman, R. A. (1 9 9 6 ). Activation of specific RXR heterodimers by an antagonist of RXR homodimers. Nature. 383, 450-453.

Wang, Y., Krushel, L. A. and Edelman, G. M. (1 9 9 6 ). Targeted DNA recombination in vivo using an adenovirus carrying the cre recombinase gene. Pr oc N atl Acad Sci USA 93, 3932-3936.

Leers, J., Treuter, E. and Gustafsson, J.-A. (1 9 9 8 ). Mechanistic Principles in NR Box-Dependent Interaction between Nuclear Hormone Receptors and the Coactivator TIF2. M o l e c u l a r a n d C e l l u l a r B i o l o g y . 18, 60016013.

Westin, S., Kurokawa, R., Nolte, R. T., Wisely, G. B., McInerney, E. M., Rose, D. W., Milburn, M. V., Rosenfeld, M. G. and Glass, C. K. (1 9 9 8 ). Interactions controlling the assembly of nuclear-receptor heterodimers and co-activators. Nature. 395, 199- 202.

Livnah, O., Stura, E. A., Johnson, D. L., Middleton, S. A., Mulcahy, L. S., Wrighton, N. C., Dower, W. J., Jolliffe, L. K. and Wilson, I. A. (1 9 9 6 ). Functional mimicry of a protein hormone by a peptide agonist: the EPO receptor complex at 2.8 A. S c i e n c e . 273, 464-471.

Wolf, E., Vassilev, A., Makino, Y., Sali, A., Nakatani, Y. and Burley, S. K. (1 9 9 8 ). Crystal Structure of a GCN5-Related N-acetyltransferase: Serratia marcescens Aminoglycoside 3-N-acetyltransferase. C e l l . 94, 439-449.

Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schutz, G., Umesono, K., Blumberg, B., Kastner, P., Mark, M., Chambon, P. and et al. (1 9 9 5 ). The nuclear receptor superfamily: the second decade. C e l l 83, 835839.

Yao, T. P., Oh, S. P., Fuchs, M., Zhou, N. D., Ch’ng, L. E., Newsome, D., Bronson, R. T., Li, E., Livingston, D. M. and Eckner, R. (1 9 9 8 ). Gene dosage-dependent embryonic development and proliferation defects in mice lacking the transcriptional integrator p300. C e l l . 93, 361-372.

Martinez-Balbas, M. A., Bannister, A. J., Martin, K., HausSeuffert, P., Meisterernst, M. and Kouzarides, T. (1 9 9 8 ). The acetyltransferase activity of CBP stimulates transcription. EMBO J. 17, 2886-2893.

Yeh, S. and Chang, C. (1 9 9 6 ). Cloning and characterization of a specific coactivator, ARA70, for the androgen receptor in human prostate cells. P r o c N a t l A c a d S c i USA 93, 5517-5521.

Moras, D. and Gronemeyer, H. (1 9 9 8 ). The nuclear receptor ligand-binding domain: structure and function. Curr O p i n C e l l B i o l 10, 384-391.

Zhang, W., Bone, J. R., Edmondson, D. G., Turner, B. M. and Roth, S. Y. (1 9 9 8 ). Essential and redundant functions of histone acetylation revealed by mutation of target lysines and loss of the Gcn5p acetyltransferase. EMBO J. 17, 3155-3167.

Mukherjee, R., Davies, P. J., Crombie, D. L., Bischoff, E. D., Cesario, R. M., Jow, L., Hamann, L. G., Boehm, M. F., Mondon, C. E., Nadzan, A. M., Paterniti, J. R., Jr. and Heyman, R. A. (1 9 9 7 ). Sensitization of diabetic and obese mice to insulin by retinoid X receptor agonists. Nature 386, 407-410.

385

Benecke and Gronemeyer: The HATs on the mouse trap

386

Gene Therapy and Molecular Biology Vol 3, page 387 Gene Ther Mol Biol Vol 3, 387-395. August 1999.

High mobility group protein HMGI-C: a molecular target in solid tumor formation Review Article

Erik Jansen* , Marleen M.R. Petit, Eric F.P.M. Schoenmakers, Torik A.Y. Ayoubi, and Wim J.M. Van de Ven Laboratory for Molecular Oncology, Center for Human Genetics, University of Leuven, and Flanders Interuniversity Institute for Biotechnology, Herestraat 49, B-3000 Leuven, Belgium __________________________________________________________________________________ *

Correspondence: Erik Jansen, Tel +(32) 16-34.60.76; Fax +(32) 16-34.60.73; E-mail: Erik.Jansen@med.kuleuven.ac.be

Key words: solid tumor formation; high mobility group I proteins; LPP Received: 17 November 1998; accepted: 19 November 1998

Summary The high mobility group protein HMGI-C is a non-histone chromosomal protein characterized by its capacity to bind in the narrow minor groove of AT-rich DNA. It is thought to act as a so-called architectural transcription factor and t o modulate the expression o f target genes through the formation of stereospecific complexes on the regulatory regions of these targets. Towards defining its function, there is now increasing evidence for a critical role of HMGI-C in the regulation of cell growth and proliferation. A direct role for HMGI-C in tumorigenesis has been demonstrated by the finding that the HMGI-C gene on human chromosome 12q15, is rearranged in a variety of solid tumors, resulting in ectopic expression. In lipomas, the LPP g e n e o n c h r o m o s o m e 3 q 2 7 - 2 8 i s t h e preferential translocation partner. It was shown t o encode a n o v e l , proline-rich, LIM domain containing protein and HMGI-C/LPP fusion transcripts have been identified in lipomas. Moreover, in overexpression experiments, the tumor-derived aberrant HMGI-C protein products are able to induce neoplastic transformation. In complementary studies, i t was shown that targeted inactivation of Hmgi-c i n m i c e di s r upt s bo t h pre - a nd po s t na t a l g ro w t h, r e s ul t i ng i n t he p y g m y phenotype. When cultured i n v i t r o , pygmy mouse-derived embryonic fibroblasts display a lower growth rate. In addition, anti-sense HMGI-C e x p r e s s i o n h a s b e e n s h o w n t o inhibit retrovirally induced transformation.

Malignant tumors involve liposarcomas (Berner et al, 1997), osteosarcomas (Kools and van de Ven, 1996; Berner et al., 1997), and aggressive angiomyxomas (Kazmierczak et al., 1995). The clinical relevance is maybe best illustrated by the fact that uterine leiomyoma is the most common pelvic neoplasm in women, occurring with an incidence of up to 77% of all women of reproductive age. Although most patients with these steroid-dependent tumors (Kawaguchi, 1989 and references therein) are asymptomatic, leiomyomas can be associated with abnormal uterine bleeding, pelvic pain, urinary dysfunction, spontaneous abortions, premature delivery, and infertility (Quade, 1995). The high incidence of this benign smooth muscle tumor constitutes a major public

I. Introduction. Chromosome 12q13-15 rearrangements in solid tumors Rearrangements involving chromosome region 12q13q15 are associated with a wide variety of human solid tumors, both malignant and benign. Among benign neoplasms, these recurrent chromosomal aberrations are observed in e.g. uterine leiomyomas (smooth muscle) (Heim et al., 1988; Vanni and Lecca, 1988), endometrial polyps (Walter et al, 1989; Vanni et al., 1993), lipomas (Sreekantaiah et al, 1991) pleiomorphic adenomas of the salivary glands (Bullerdiek et al, 1993), pulmonary chondroid hamartomas (Kazmierczak et al., 1995a), and fibroadenomas of the breast (Staats et al., 1996). 387

Jansen et al: HMGI- as a molecular target in solid tumor formation health problem, leading to over 200,000 hysterectomies performed annually in the USA, on the basis of the diagnosis of myomatous uterus (Cramer 1990; Carlson, 1993).

Only very recently, we have been able to molecularly define the uterine leiomyoma cluster region on chromosome 14q23-24 (ULCR14) to an interval of approximately 1 Mb. Subsequent gene-identification experiments within this region, revealed the presence of a single transcript, the exons of which are dispersed over an interval of approximately 0.9 Mb. FISH analysis of tumor-derived material revealed that all genomic breakpoints indeed mapped within the span of this gene. Furthermore, removal of the last exon seems to be the minimal tumor-associated event. By means of 3’-RACE and RT-PCR experiments we were able to confirm the direct, reciprocal involvement of this ULCR14 gene, since they led to the isolation of various HMGI-C-containing fusion transcripts (Schoenmakers et al., 1999).

The above-mentioned anomalies of chromosome 12 segment q13-15 seem to be one of the most frequent chromosomal abnormalities associated with mesenchymal neoplasms in humans. Recently, it has been shown that the HMGI-C gene is affected by these chromosomal aberrations (Ashar et al., 1995; Schoenmakers et al., 1995; Kazmierczak et al., 1996). The structure of the human HMGI-C gene and its protein product is shown in Figure 1 and will be discussed in more detail in one of the next sections. The gene consists of five exons and spans about 175 kb of genomic DNA. Many breakpoints were found to be clustered within the large (>140 kb) third intron, resulting in the loss of exons 4 and 5. By means of RTPCR analysis, we have shown that the corresponding transcripts encode a truncated version of the HMGI-C protein in which the carboxy-terminal domains are deleted. In addition, some breakpoints were found in the 5’- and 3’flanking regions of the HMGI-C gene. In lipomas and osteosarcomas a specific amplification of a rearranged HMGI-C gene was detected, resulting in ectopic expression of HMGI-C (Kools and van de Ven, 1996; Berner et al., 1997). Many chromosomes have been found as translocation partner of HMGI-C. By means of RT-PCR analysis, various aberrant fusion transcripts were detected (Schoenmakers, 1997). So far, only a few fusion partner genes of HMGI-C have been extensively characterized, i.e. the FHIT (Geurts et al., 1998) and NFIB (Geurts et al., 1997) genes in pleomorphic adenomas of the salivary glands, the mitochondrial aldehyde dehydrogenase gene (ALDH2) in a uterine leiomyoma (Kazmierczak et al., 1995b). Moreover, this review will discuss recent data on the preferential translocation partner genes on chromosomes 3q27-28 in lipomas (Turc-Carel et al., 1986) and 14q23-24 in uterine leiomyomas (Mitelman, 1991).

II. The HMGI-C protein, DNA binding, and search for target genes The HMGI-C protein belongs to the high mobility group (HMG) of DNA binding proteins. This class of nuclear proteins can be divided into three distinct families. The HMG-1/2, HMGI, and HMG14/17 families (reviewed in Bustin and Reeves, 1996). The HMGI family consists of HMGI-C, HMGI, and HMGY. The HMGI proteins are characterized by the presence of three DNA binding domains, the AT-hooks, containing a short stretch of basic amino acids, and a carboxy-terminal highly acidic domain. A prominent feature of these proteins is their ability to bind in the narrow minor groove of AT-rich DNA. HMGI and HMGY are derived by alternative splicing of transcripts of a single gene on chromosome 6p21 (Johnson et al., 1989; Friedman et al., 1993). Interestingly, the HMGI/Y gene was also found to be subject of chromosomal rearrangements in e.g. lipomas and pulmonary chondroid hamartomas (Kazmierczak et al., 1996; Tkachenko et al., 1997). In some cases, fusion transcripts and resultant chimeric proteins were detected. As is the case for HMGIC, ectopic expression is observed in various tumor types (Ram et al., 1993; Tamimi et al., 1993).

In lipomas, we have identified and characterized the chromosome 3-derived translocation partner gene of HMGIC, which we have designated LPP (Lipoma P referred P artner gene) (Petit et al., 1996, 1998). We were able to detect aberrant transcripts encoding HMGI-C/LPP fusion proteins in both primary lipomas and lipoma cell lines. Apart from lipomas, HMGI-C/LPP fusion transcripts were also recently detected in a cytogenetic subgroup of pulmonary chondroid hamartomas with chromosome (3;12)(q27;q14-15) translocations (Rogalla et al, 1998). A detailed description of the resultant chimeric proteins and wild-type LPP will be provided in one of the next sections.

The AT-hooks of HMGI-C are each encoded by separate exons (exons 1-3). Exons 4 and 5 encode the spacer domain and highly acidic carboxy-terminal tail, respectively, (Figure 1) and therefore translocations involving chromosomal breakage within intron 3, result in the synthesis of aberrant HMGI-C proteins in which the latter domains are deleted. Apart from an 11 amino acid segment only present within HMGI-C, the HMGI/Y and HMGI-C proteins display a high degree of sequence identity, especially within the three DNA-binding domains (Figure 2).

388

Gene Therapy and Molecular Biology Vol 3, page 389

Figure 1. Structure of the human HMGI-C gene and its protein product. On the gene map, the exons are depicted as boxes, with the 5’- and 3’-unstranslated regions represented as shaded areas. The numbers below the map indicate the intron sizes in kb. The dashed lines indicate which regions of the HMGI-C protein are encoded by the individual exons and the amino acid numbering above the protein map mark the boundaries of the various DNA-binding and acidic domains.

Figure 2. Human HMGI-C amino acid sequence aligned with human HMG-I and HMG-Y.

could act as a classical transcription factor with activating and/or silencing properties, several domains of HMGI-C were fused to the DNA binding domain of the GAL4 protein and the resultant fusion proteins were assayed for transcriptional effects in a standard assay using chimeric

The overall structure of HMGI proteins is reminiscent of the modular structure of a classical transcription factor containing activation and/or silencing domain(s) spatially separated form the DNA binding domain (Sadowski et al., 1988). Therefore, in order to investigate whether HMGI-C 389

Gene Therapy and Molecular Biology Vol 3, page 390

Figure 3. To investigate potential transcription activation or silencing activities associated with the HMGI-C protein itself, the various domains of it were fused to the DNA-binding domain of GAL4. In transfection experiments, we found no significant effect of the GAL4-HMGI-C hybrids on transcriptional activity of 5x-GAL4-luciferase reporter constructs.

Figure 4. DNA binding activity of purified, recombinant human HMGI-C protein and its reactivity with anti-HMGI-C antibodies. In this EMSA experiment, a synthetic, double-stranded oligonucleotide containing high affinity binding sites was used as a probe. Lane 1, binding mixture with probe only, no HMGI-C added. Lane 2, 10 ng of HMGI-C added to the binding reaction. Lane 3, as lane 2, and anti-HMGI-C antibody also included. The antibody is raised against HMGI-C-specific amino acid residues. It does not interfere with DNA binding, but gives rise to a supershift of the existing protein-DNA complex.

concert with other DNA binding proteins in order to specifically regulate target promoter activity. It is now believed that HMGI-C and HMGI/Y act as so-called architectural transcription factors (Wolffe, 1994) and thereby regulate the affinity and activity of other transcription factors by altering local chromatin structure. As a result of this, they modulate the expression of their

promoter-luciferase reporter constructs containing GAL4 binding sites. As can be deduced from the results (Figure 3), both wild-type domains and the aberrant HMGI-C proteins as observed in tumors do not have transcription activating or silencing activities. These data are in agreement with the current model that the HMGI proteins do not regulate transcription on their own, but act in 390

Gene Therapy and Molecular Biology Vol 3, page 391 respective target genes through the formation of stereospecific complexes on the regulatory regions. This model is supported by experimental data of which the formation of a multi-protein complex, called “enhanceosome” at the interferon-ß promoter is most extensively described (Thanos and Maniatis, 1995; Yie et al., 1997). In addition, several other target genes for HMGI/Y have been identified in which HMGI/Y enhances or represses promoter activity (a detailed list is provided in Cmarik et al., 1998).

exons (dispersed over about 400 kb) with the predicted start codon in exon 3 and the stop codon in exon 11. All three LIM domains are encoded by separate exons. The function of LPP is not known, but database searches revealed that the LPP protein is a member of a new family of proteins. Apart from LPP, two other family members are known to date: TRIP6 and zyxin. TRIP6 was identified in a yeast two-hybrid assay as a Thyroid Receptor Interacting Protein (Yi et al., 1998). Zyxin (Beckerle, 1997), the founding member of this family, is found at sites of cell adhesion to the extracellular matrix (adhesion plaques), and was recently shown to enhance actin organizing activity in mammalian cells (Golsteyn et al., 1997). In addition, it was shown that this protein is able to shuttle to the nucleus (Nix and Beckerle, 1998).

To aid the identification of HMGI-C target genes, of which none are known at the moment, we started with defining low and high affinity binding sites among all possible combinations of A/T-rich DNA stretches. This was done employing a PCR-assisted DNA binding site selection strategy (CASting) (Ko and Engel, 1993) using purified recombinant HMGI-C and anti-HMGI-C antibodies. DNA binding and antibody reactivity was evaluated in EMSA experiments (Figure 4). Preliminary results revealed a preference for homopolymeric A- or Tstretches, which were infrequently disrupted by C- or Gresidues, respectively (Jansen et al., in prep). With the obtained set of data we are currently selecting potential target genes for transcriptional analysis. In addition, using a PCR-based subtractive cloning technique (Sagerström et al., 1997), we are in the process of isolating genes expressed in one cell population but not in another, e.g. HMGI-C knockout versus wild-type.

III. The HMGI-C/LPP fusion in lipomas As mentioned above, we discovered the chromosome 3derived translocation partner gene of HMGI-C in lipomas, which was designated LPP (Lipoma P referred P artner gene). By Northern blot analysis, the LPP gene was shown to encode an mRNA of over 13 kb that is expressed in a variety of human tissues. We have constructed a cDNA contig encompassing the complete coding region of LPP. Nucleotide sequence analysis revealed that the gene encodes a novel protein of 612 amino acids that is unusually proline-rich, in its amino-terminal region, and has three LIM domains, in its carboxy-terminal region (Petit et al., 1996). LIM domains are cysteine-rich, zinc-binding protein sequences that are found in a growing number of proteins with diverse functions, including transcription regulators, proto-oncogene products, and adhesion plaque constituents. Many of the LIM family members have been postulated to play a role in cell signaling and control of cell fate during development and it was demonstrated that LIM domains can act as protein binding interfaces (Schmeichel and Beckerle, 1994). Extensive studies on the genomic organization of the human LPP gene resulted in the identification of 11 LPP

391

Using RT-PCR in the analysis of a number of lipoma cell lines, primary lipomas, a parosteal lipoma, and pulmonary chondroid hamartoma, it appeared that LPP is frequently rearranged, also in cases without a cytogenetically detectable involvement of 3q27-28 (Petit et al., 1996, 1998; Rogalla et al., 1998). In lipomas, two alternative HMGI-C/LPP hybrid transcripts have been detected encoding the three DNA binding domains of HMGI-C followed by: 1) part of the proline rich domain and all three LIM domains of LPP or, most frequently observed: 2) the two most carboxy-terminal LIM domains (LIM 2-3) of LPP (Figure 5). The subcellular localization and transcriptional regulatory properties of wild-type LPP and the lipoma-derived HMGI-C/LPP fusion proteins are subject of present studies in our lab. This will provide further insight in the mechanism of lipoma formation.

IV. Critical role for HMGI-C in growth and development There is now substantial evidence for the involvement of HMGI-C in the regulation of cell growth and proliferation. Moreover, as outlined above, a first clue came from the finding that translocations disrupting the human HMGI-C gene result in ectopic expression of HMGI-C in a variety of solid tumors (Ashar et al., 1995; Schoenmakers et al., 1995). A regulatory function of HMGI-C in growth and proliferation is further substantiated by the observation that targeted disruption of the Hmgi-c gene in mice results in a pygmy phenotype (Zhou et al., 1995). When cultured in vitro, pygmy mouse-derived embryonic fibroblasts display a lower growth rate (Zhou et al., 1995; and our unpublished observations). Interestingly, Hmgi-c inactivation does not affect the growth hormone/insulin-like growth factor endocrine pathway. In addition, anti-sense HMGI-C expres-

Jansen et al: HMGI- as a molecular target in solid tumor formation

Figure 5 . Schematic representation of the HMGI-C/LPP fusion proteins which arise from the chromosome 3;12translocation. This event is observed in lipomas and pulmonary chondroid hamartomas and affects the proteins encoded by the HMGI-C gene on 12q15 and the LPP gene on 3q27-28.

as a delayed early response gene passing proliferation/differentiation control signals to responsive target genes (Figure 6).

sion has been shown to inhibit retrovirally induced transformation resulting in reversal of the transformed phenotype (Berlingieri et al., 1995) and high expression of HMGI-C is only observed in early embryogenesis, in cell lines of embryonic origin and in malignantly transformed cells.

V. Concluding remarks and future prospects

The expression of the HMGI-C gene itself is under control of various growth-associated factors which were shown to act at the level of transcription initiation (Ayoubi et al., in prep.) and HMGI-C is identified as a secondary response gene to oncogenic Raf-1 activation (Li et al., 1997). The obtained results let us to define HMGI-C

The experimental data mentioned in this review, clearly point towards an essential function of HMGI-C in the control of cell growth and proliferation. HMGI-C behaves like an oncogene and is therefore potentially a target for therapeutic treatment of tumors which overexpress the gene. Due to the role of HMGI-C in tumor development 392

Gene Therapy and Molecular Biology Vol 3, page 393 and its deregulation in tumors, it is important to understand how HMGI-C expression is regulated and which signal transduction pathways are involved in the growthassociated expression of HMGI-C. In addition, the exact mechanism by which the HMGI-C protein itself exerts its regulatory effects is not known so far and, therefore, the identification of target genes and functional characterization of the various tumor-derived chimeric HMGI-C fusion proteins may provide insight in the molecular mechanism of solid tumor formation and may be instrumental in the development of potential therapeutic agents interfering with tumor formation.

Acknowledgements This work was supported by the "Fonds voor Wetenschappelijk Onderzoek, Vlaanderen" (FWO), the "Geconcerteerde Onderzoekacties 1997-2001", the ASLK "Programma voor Kankeronderzoek". The authors would like to thank Anne-Fleur Jansen for inspiration. Marleen M.R. Petit is a Research Assistant of the FWO ("Kom op tegen Kanker", Belgium).

Figure 6. Signal transduction model in which the potential activities of HMGI-C and LPP in the regulation of proliferation and differentiation are indicated.

393

Jansen et al: HMGI- as a molecular target in solid tumor formation

References

translocation, t(12;14)(q14-q15;q23-q24), characterizes a subgroup of uterine leiomyomas. Cancer Genet C y t o g e n e t 32, 13-17.

Ashar HR, Fejzo MS, Tkachenko A, Zhou X, Fletcher JA, Weremowicz S, Morton CC, Chada K. (1 9 9 5 ) Disruption of the architectural factor HMGI-C: DNA-binding AT hook motifs fused in lipomas to distinct transcriptional regulatory domains. C e l l 82, 57-65.

Johnson KR, Lehn DA, Reeves R. (1 9 8 9 ) Alternative processing of mRNAs encoding mammalian chromosomal high-mobility-group proteins HMG-I and HMG-Y. M o l C e l l B i o l 9, 2114-2123.

Beckerle MC. (1 9 9 7 ) Zyxin: zinc fingers at sites of cell adhesion. B i o e s s a y s 19, 949-957.

Kawaguchi K, Fujii S, Konishi I, Nanbu Y, Nonogaki H, Mori T. (1 9 8 9 ) Mitotic activity in uterine leiomyomas during the menstrual cycle. A m J O b s t e t G y n e c o l 160, 637641.

Berlingieri MT, Manfioletti G, Santoro M, Bandiera A, Visconti R, Giancotti V, Fusco A. (1 9 9 5 ) Inhibition of HMGI-C protein synthesis suppresses retrovirally induced neoplastic transformation of rat thyroid cells. M o l C e l l B i o l 15, 1545-1553. Berner JM, Meza-Zepeda LA, Kools PF, Forus Schoenmakers EFPM, Van de Ven WJM, Fodstad Myklebost O. (1 9 9 7 ) HMGIC, the gene for architectural transcription factor, is amplified rearranged in a subset of human sarcomas. O n c o g e n e 2935-2941.

Kazmierczak B, Hennig Y, Wanschura S, Rogalla P, Bartnitzke S, Van de Ven W, Bullerdiek J. (1 9 9 5 b ) Description of a novel fusion transcript between HMGI-C, a gene encoding for a member of the high mobility group proteins, and the mitochondrial aldehyde dehydrogenase gene. Cancer R e s 55, 6038-6039.

A, O, an and 14,

Kazmierczak B, Rosigkeit J, Wanschura S, Meyer-Bolte K, Van de Ven WJM, Kayser K, Krieghoff B, Kastendiek H, Bartnitzke S, Bullerdiek J. (1 9 9 6 ) HMGI-C rearrangements as the molecular basis for the majority of pulmonary chondroid hamartomas: a survey of 30 tumors. O n c o g e n e 12, 515-521.

Bullerdiek J, Wobst G, Meyer-Bolte K, Chilla R, Haubrich J, Thode B, Bartnitzke S. (1 9 9 3 ) Cytogenetic subtyping of 220 salivary gland pleomorphic adenomas: correlation to occurrence, histological subtype, and in vitro cellular behavior. Cancer Genet Cytogenet 65, 27-31.

Kazmierczak B, Wanschura S, Meyer-Bolte K, Caselitz J, Meister P, Bartnitzke S, Van de Ven WJM, Bullerdiek J. (1 9 9 5 ) Cytogenic and molecular analysis of an aggressive angiomyxoma. Am J Pathol 147, 580-585.

Bustin M, Reeves R. (1 9 9 6 ) High-mobility-group chromosomal proteins: architectural components that facilitate chromatin function. P r o g N u c l e i c A c i d R e s M o l B i o l 54, 35-100.

Kazmierczak B, Wanschura S, Rommel B, Bartnitzke S, Bullerdiek J. (1 9 9 6 ) Ten pulmonary chondroid hamartomas with chromosome 6p21 breakpoints within the HMG-I(Y) gene or its immediate surroundings. J N a t l Cancer Inst 88, 1234-1236.

Carlson KJ, Nichols DH, Schiff I. (1 9 9 3 ) Indications for hysterectomy. N Engl J Med 328, 856-860. Cmarik JL, Li Y, Ogram SA, Min H, Reeves R, Colburn NH. (1 9 9 8 ) Tumor promoter induces high mobility group HMG-Y protein expression in transformation-sensitive but not -resistant cells. O n c o g e n e 16, 3387-3396.

Kazmierczak B, Wanschura S, Rosigkeit J, Meyer-Bolte K, Uschinsky K, Haupt R, Schoenmakers EFPM, Bartnitzke S, Van de Ven WJM, Bullerdiek J. (1 9 9 5 a ) Molecular characterization of 12q14-15 rearrangements in three pulmonary chondroid hamartomas. Cancer R e s 55, 2497-2499.

Cramer SF, Patel A. (1 9 9 0 ) The frequency of uterine leiomyomas. Am J Clin Pathol 94, 435-438. Friedmann M, Holth LT, Zoghbi HY, Reeves R. (1 9 9 3 ) Organization, inducible-expression, and chromosome localization of the human HMG-I(Y) nonhistone protein gene. N u c l e i c A c i d s R e s 21, 4259-4267.

Ko LJ, Engel JD. (1 9 9 3 ) DNA-binding specificities of the GATA transcription factor family. M o l C e l l B i o l 13, 4011-4022.

Geurts JMW, Schoenmakers EFPM, Rรถijer E, Astrรถm A-K, Stenman G, van de Ven WJM. (1 9 9 8 ) Identification of NFIB as recurrent translocation partner gene of HMGIC in pleomorphic adenomas. O n c o g e n e 16, 865-872.

Kools PF, Van de Ven WJM. (1 9 9 6 ) Amplification of a rearranged form of the high-mobility group protein gene HMGIC in OsA-CI osteosarcoma cells. Cancer Genet C y t o g e n e t 91, 1-7.

Geurts JMW, Schoenmakers EFPM, Rรถijer E, Stenman G, Van de Ven WJM (1 9 9 7 ) Expression of reciprocal hybrid transcripts of HMGIC and FHIT in a pleomorphic adenoma of the parotid gland. Cancer Res 57, 13-17.

Li D, Lin HH, McMahon M, Ma H, Ann DK. (1 9 9 7 ) Oncogenic raf-1 induces the expression of non-histone chromosomal architectural protein HMGI-C via a p44/p42 mitogen-activated protein kinase-dependent pathway in salivary epithelial cells. J B i o l Chem 272, 2506225070.

Golsteyn RM, Beckerle MC, Koay T, Friederich E. (1 9 9 7 ) Structural and functional similarities between the human cytoskeletal protein zyxin and the ActA protein of Listeria monocytogenes. J C e l l S c i 110 ,1893-1906.

Mitelman F. (1 9 9 1 ) C a t a l o g o f c h r o m o s o m e aberrations in cancer , 4 th ed. (Wiley-Liss, New York).

Heim S, Nilbert M, Vanni R, Floderus U-M, Mandahl N, Liedgren S, Lecca U, Mitelman F. (1 9 8 8 ) A specific

Nix DA, Beckerle MC. (1 9 9 7 ) Nuclear-cytoplasmic shuttling of the focal contact protein, zyxin: a potential mechanism

394

Gene Therapy and Molecular Biology Vol 3, page 395 for communication between sites of cell adhesion and the nucleus. J C e l l B i o l 138, 1139-1147.

Tamimi Y, van der Poel HG, Denyn MM, Umbas R, Karthaus HF, Debruyne FM, Schalken JA. (1 9 9 3 ) Increased expression of high mobility group protein I(Y) in high grade prostatic cancer determined by in situ hybridization. Cancer Res 53, 5512-5516.

Petit MMR, Mols R, Schoenmakers EFPM, Mandahl N, Van de Ven WJM. (1 9 9 6 ) LPP, the preferred fusion partner gene of HMGIC in lipomas, is a novel member of the LIM protein gene family. G e n o m i c s 36, 118-129.

Thanos D, Maniatis T. (1 9 9 5 ) Virus induction of human IFN beta gene expression requires the assembly of an enhanceosome. C e l l 83, 1091-1100.

Petit MMR, Swarts S, Bridge JA, Van de Ven WJM (1 9 9 8 ) Expression of reciprocal fusion transcripts of the HMGIC and LPP genes in parosteal lipoma. Cancer Genet C y t o g e n e t 106, 18-23.

Tkachenko A, Ashar HR, Meloni AM, Sandberg AA, Chada KK. (1 9 9 7 ) Misexpression of disrupted HMGI architectural factors activates alternative pathways of tumorigenesis. Cancer Res 57, 2276-2280.

Quade BJ. (1 9 9 5 ) Pathology, cytogenetics and molecular biology of uterine leiomyomas and other smooth muscle lesions. Curr Opin Obstet Gynecol 7, 35-42

Turc-Carel C, Dal Cin P, Rao U, Karakousis C, Sandberg AA. (1 9 8 6 ) Cytogenetic studies of adipose tissue tumors. I. A benign lipoma with reciprocal translocation t(3;12)(q28;q14). C a n c e r G e n e t C y t o g e n e t 23, 283289.

Ram TG, Reeves R, Hosick HL. (1 9 9 3 ) Elevated high mobility group-I(Y) gene expression is associated with progressive transformation of mouse mammary epithelial cells. Cancer Res 53, 2655-2660.

Vanni R, Dal Cin P, Marras S, Moerman Ph, Andrai M, Valdes E, Deprest J, Van den Berghe H. (1 9 9 3 ) Endometrial polyp: another benign tumor characterized by 12q13-15 changes. Cancer Genet Cytogenet 68, 32-33.

Rogalla P, Kazmierczak B, Meyer-Bolte K, Tran KH, Bullerdiek J. (1 9 9 8 ) The t(3;12)(q27;q14-q15) with underlying HMGIC-LPP fusion is not determining an adipocytic phenotype. G e n e s C h r o m o s o m e s C a n c e r 22, 100-104.

Vanni R, Lecca U. (1 9 8 8 ) Involvement of the long arm of chromosome 12 in chromosome rearrangement of uterine leiomyoma. Cancer Genet Cytogenet 32, 33-34.

Sadowski I, Ma J, Triezenberg S, Ptashne M. (1 9 8 8 ) GAL4VP16 is an unusually potent transcriptional activator. Nature 335, 563-564. Sagerstrรถm CG, Sun BI, Sive HL. (1 9 9 7 ) Subtractive cloning: past, present, and future. A n n u R e v B i o c h e m 66, 751783.

Walter TA, Fan SX, Medchill MT, Berger CS, Decker HJH, Sandberg AA. (1 9 8 9 ) Inv(12)(p11.2q13) in an endometrial polyp. C a n c e r G e n e t C y t o g e n e t 41, 99103.

Schmeichel KL, Beckerle MC. (1 9 9 4 ) The LIM domain is a modular protein-binding interface. C e l l 79, 211-219.

Wolffe AP. (1 9 9 4 ) Architectural transcription factors. S c i e n c e 264, 1100-1101.

Schoenmakers EFPM (1 9 9 7 ) The molecular b a s i s o f benign solid tumors: discovery of a common genetic denominator on the long arm of human c h r o m o s o m e 1 2 . PhD thesis, University of Leuven, Belgium.

Yi J, Beckerle MC. (1 9 9 8 ) The human TRIP6 gene encodes a LIM domain protein and maps to chromosome 7q22, a region associated with tumorigenesis. G e n o m i c s 49, 314-316. Yie J, Liang S, Merika M, Thanos D. (1 9 9 7 ) Intra- and intermolecular cooperative binding of high-mobilitygroup protein I(Y) to the beta-interferon promoter. M o l C e l l B i o l 17, 3649-3662.

Schoenmakers EFPM, Wanschura S, Mols R, Bullerdiek J, Van den Berghe H, Van de Ven WJM. (1 9 9 5 ) Recurrent rearrangements in the high mobility group protein gene, HMGI-C, in benign mesenchymal tumours. Nat Genet 10, 436-444.

Zhou X, Benson KF, Ashar HR, Chada K. (1 9 9 5 ) Mutation responsible for the mouse pygmy phenotype in the developmentally regulated factor HMGI-C. N a t u r e 3 7 6 , 771-774.

Schoenmakers EFPM, Huysmans C, Van de Ven WJM. (1 9 9 9 ) Allelic knockout of novel splice variants of human recombination repair gene RAD51B in t(12;14) uterine leiomyomas. Cancer Res. 59, 19-23. Sreekantaiah C, Leong SLP, Karakousis CP, McGee DL, Rappaport WD, Villar HV, Neal D, Fleming S, Wankel A, Herrington PN, Carmona R, Sandberg AA. (1 9 9 1 ) Cytogenetic profile of 109 lipomas. Cancer R e s 51, 422-433. Staats B, Bonk U, Wanschura S, Hanisch P, Schoenmakers EFPM, Van de Ven WJM, Bartnitzke S, Bullerdiek J. (1 9 9 6 ) A fibroadenoma with a t(4;12) (q27;q15) affecting the HMGI-C gene, a member of the high mobility group protein gene family. Breast Cancer R e s Treat 38, 299-303.

395

Gene Therapy and Molecular Biology Vol 3, page 397 Gene Ther Mol Biol Vol 3, 397-412. August 1999.

Replication of simple DNA repeats Review Article

Maria M. Krasilnikova, George M. Samadashwily and Sergei M. Mirkin* Department of Molecular Genetics, University of Illinois at Chicago, Chicago IL 60607 __________________________________________________________________________________________________ Corresponding Author: Phone:(312)996-9610; Fax: (312)413-0353; E-mail:mirkin@uic.edu Key Words: DNA repeats, inverted repeats, DNA replication, repeat length polymorphism, replication attenuation A b b r e v i a t i o n s : WC, Watson-Crick; IR , inverted repeats; MR, mirror repeats; DTR , direct tandem repeats; S-DNA, slippedstranded DNA Received: 7 October 1998; accepted: 16 October 1998

Summary This chapter presents an overview of studies on the replication of simple DNA repeats conducted in our laboratory during the last seven years. The recent massive increase in available DNA sequences has led to the clear understanding that natural DNAs, particularly in eukaryotes, are extraordinarily e n r i c h e d i n d i f f e r e n t r e p e a t s ( S c h r o t h a n d H o , 1 9 9 5 ; C o x a n d M i r k i n , 1 9 9 7 ) . T h i s leads t o an o b v i o u s q u e s t i o n : w h a t a r e t h e b i o l o g i c a l functions ( i f any) o f these repeated elements? This problem is currently the subject of very intense studies in many laboratories all over the world. We came to this question after we realized that many repeated DNA sequences constitute a major obstacle t o DNA polymerization i n v i t r o (Dayn e t . a l . , 1 9 9 2 ; Samadashwily e t . a l . , 1993; Samadashwily and Mirkin, 1994; Krasilnikov et. al., 1997). Subsequently, we found that several such repeats attenuate DNA replication i n v i v o as well (Samadashwily et. al., 1997; Krasilnikova et. al., 1998). Based on our data, we conclude that there are at least three mechanisms by which different repeats inhibit replication. We believe that this may reflect a potentially important role of repeated DNA as punctuation marks for major genetic processes in DNA texts. Repeat-caused replication attenuation might also contribute to the mechanisms of repeat length polymorphism seen in many human diseases.

I. Repeat types, structures and frequencies Based on sequence arrangement and symmetry, three major types of simple DNA repeats are usually considered (F i g . 1): inverted repeats, mirror repeats, and direct tandem repeats. Inverted repeats (IR) are DNA sequences in which DNA bases that are equidistant from the symmetry center in a DNA strand are Watson-Crick (WC) complements to each other. Mirror repeats (MR) are also symmetrical, but here equidistant DNA bases are identical to each other. Finally, direct tandem repeats (DTR) are simple, uninterrupted iterations of a core repeat unit along the DNA strand. The distinction between the different repeat types is not absolute. There are DNA sequences that meet criteria for all three repeat types. A well studied example is the d(A-T)n. d(T-A)n sequence which is simultaneously an inverted, mirror and direct tandem repeat. The secondary structure of repeated DNA often differs 397

dramatically from canonical B-DNA. The exact conformation of a repeated DNA depends on its symmetry, base composition, DNA supercoiling, ambient conditions, etc. Below, we will briefly summarize the best characterized structures formed by different DNA repeats. Inverted repeats are capable of forming cruciform structures in double stranded DNA or hairpins in single stranded DNA (F i g . 2). While hairpin formation in single-stranded DNA is generally energetically favorable, cruciform formation in a double-stranded DNA is only favorable under the influence of negative supercoiling (Lilley, 1980; Panayotatos and Wells, 1981; Mizuuchi, et. al., 1982). Indeed, to convert a duplex DNA segment into a cruciform state, one would need to at least partially unwind it in order to allow for self-pairing by each DNA strand. Since initial unwinding of a DNA duplex is energy consuming, this stage represents an energetic barrier for cruciform formation (reviewed in Sinden, 1994). In addition, the resultant cruciform structure contains singlestranded bases in central loops and energetically costly

Krasilnikova et al:Replication of Simple DNA Repeats cruciform is topologically equivalent to unwound DNA, its formation under torsional stress release negative supercoils, compensating for the high nucleation cost (Vologodskii and Frank-Kamenetskii, 1982). In fact, simple energetics calculations show that the probability of cruciform extrusion must increase exponentially with inverted repeat length (Benham, 1982; Vologodskii and Frank-Kamenetskii, 1982). In vivo formation of cruciform structures was observed in several studies. The most direct evidence was obtained by chemical probing of plasmids in E. coli cells. AT-rich inverted repeats were shown to adopt cruciform conformation when intracellular supercoiling increased due to certain changes in environmental conditions (McClellan, et. al., 1990; Dayn, et. al., 1991; Zheng, et. al., 1991) or as a consequence of transcriptional activation (Dayn, et. al., 1992). Formation of a cruciform-like structure in the enhancer of the enkephalin gene was suggested based on chemical probing of human intracellular DNA followed by ligation-mediated PCR (Spiro et. al., 1995). While mirror repeats of arbitrary composition are common in natural DNA (Schroth and Ho, 1995; Cox and Mirkin, 1997), only one type of them, i.e. homopurinehomopyrimidine mirror repeats (H-palindromes), has been shown to adopt a non-B conformation. These sequences can adopt an intramolecular triplex called H DNA (Lyamichev, et. al., 1986). To form this structure (F i g . 3), a DNA strand from one half of the repeat folds back, forming a triplex with the duplex half of the repeat, while its complement remains single-stranded (Mirkin et. al., 1987). Depending on the chemical nature of the strand donated to the triplex, either pyrimidine- or purine-rich, the resultant structures are called H-y or H-r, respectively (F i g . 3 ). The H-y form is built from TA*T and CG*C+ triads (F i g . 4 A ), where pyrimidines from the third strand are situated in the major groove, forming Hoogsteen hydrogen bonds with the purines of the duplex (Hoogsteen, 1963).The extingency for cytosine protonation makes this structure preferred under acidic pH (Mirkin and FrankKamenetskii, 1994). The H-r form can be built of CG*G, TA*A and, unexpectedly, TA*T triads (Kohwi and KohwiShigematsu, 1988; Beal and Dervan, 1991; Dayn, et. al., 1992). In this case, DNA bases of the third strand form reverse Hoogsteen hydrogen bonds with the purines of the duplex (F i g . 4B) (Hoogsteen, 1963). These triads are stable at physiological pH, but are greatly stabilized in the presence of divalent cations (Kohwi, 1989; Bernues, et. al., 1990; Beltran, et. al., 1993; Malkov, et. al., 1993; Martinez-Balbas and Azorin, 1993). Like cruciform structures, a H DNA is topologically equivalent to the completely unwound DNA and its formation requires substantial duplex unwinding (Lyamichev, et. al., 1985). Thus it is favored in negatively supercoiled DNA, and its formation depends exponentially on repeat length (reviewed in Mirkin and Frank-Kamenetskii, 1994). Cloned homopurine-homopyrimidine repeats were shown to adopt H conformation in E. coli cells when the intracellular supercoiling increased due to mutations in topoisomerase

F i g . 1 . Different repeat types. A: inverted repeats; B : mirror repeats; C: direct tandem repeats. Arrows of the same color represent symmetrical sequences. Complementary sequences differ in color.

F i g . 2 . Cruciform and hairpin structures formed by inverted repeats in double- and single-stranded DNA, respectively. Complementary halves of an inverted repeat are red and g r e e n , single-stranded segments are o r a n g e , and the surrounding DNA is purple.

junctions between the cruciform and the adjacent duplex DNA (4-way junctions). Altogether, this leads to a high nucleation energy for cruciform formation, approaching 20 kcal mol -1 (reviewed in Vologodskii, 1992). Since a 398

Gene Therapy and Molecular Biology Vol 3, page 399

F i g . 3 . H DNA structure. Both H-y and H-r forms are shown. R e d l i n e: homopurine and gre e n l i n e : homopyrimidine strands of an H palindrome, respectively. P u r p l e l i n e s : adjacent DNA. B l a c k l i n e s : Watson-Crick hydrogen bonds; st a r s : Hoogsteen hydrogen bonds.

F i g . 4 . Triplex forming triads. A: Hoogsteen triads forming H-y DNA. B : reverse Hoogsteen triads forming H-r DNA.

genes, chloramphenicol treatment, or transcriptional activation (Kohwi, et. al., 1992; Kohwi and Panchenko, 1993; Ussery and Sinden, 1993). Antibodies against triplex DNA were also shown to interact with chromosomes of permeabilized mammalian cells (Agazie, et. al., 1996).

Direct tandem repeats (DTRs) can adopt a variety of conformations. As is obvious from the above discussion, some DTRs can form cruciforms or H DNA as long as they happened to be inverted repeats or H-palindromes, respectively. Another structure, called a G-quartet (F i g . 5 ) can be formed by DTRs containing tandemly arranged runs of guanines (Gellert, et. al., 1962; Zimmerman, et. al., 399

Krasilnikova et al:Replication of Simple DNA Repeats

F i g . 6 . Schematic representation of slipped-stranded DNA structure (S-DNA). Red and green lines represent complementary strands of a DTR. Purple lines - surrounding DNA.

al., 1979; reviewed in Rich, et. al., 1984). In linear DNA, this structure is only possible under rather exotic conditions (such as very high ionic strength) (Peck, et. al., 1982; Singleton, et. al., 1982). In superhelical DNA, by contrast, it is extremely favorable under physiological conditions, since its releases twice as many supercoils per DNA base as unwound DNA, cruciforms or H DNA (Singleton, et. al., 1982). Z-DNA was detected in bacterial cells after an increase in DNA supercoiling due to environmental changes or transcription (Haniford and Pulleyblank, 1983; Jaworski, et. al., 1989; Rahmouni and Wells, 1992). In permeabilized mammalian cells anti-Z antibodies specifically interact with chromosomes, targeting upstream parts of actively transcribed genes (Wittig, et. al., 1992; Wolfl, et. al., 1996). Finally, DTRs of various base compositions can adopt a structure called slipped-stranded DNA (S-DNA) (reviewed in Sinden, 1994). This structure (F i g . 6) utilizes the multiply repeated nature of the sequence: upon denaturing and renaturing, the complementary repeats can mispair, resulting in a peculiar combination of double-helical stretches intervened by single-stranded loops. In linear DNA, this conformation is thermodynamically unfavorable but can be trapped kinetically. In superhelical DNA, it might become favorable given the release of substantial torsional tension. It is worth noting, that for core repeated units of certain base compositions, the loops can be additionally stabilized by hydrogen bonds of both WC and non-WC nature (Pearson and Sinden, 1996). This would certainly make S-DNA more favorable. For example, formation of S-DNA was suggested for expandable (CXG)n trinucleotide repeats in linear or superhelical DNA upon denaturing/renaturing (Pearson and Sinden, 1996; Chen, et. al., 1998; Mariappan, et. al., 1998; Pearson, et. al., 1998;

F i g . 5 . Quadruplex DNA structure. A. General overview. Black line - DNA strand, purple rectangles - stacked Gquartets. B - Chemical structure of a G-quartet.

1975; Sen and Gilbert, 1988; Sundquist and Klug, 1989). It is built from stacked G 4 blocks that are additionally stabilized in the presence of monovalent ions (Pinnavaia, et. al., 1978; Williamson, et. al., 1989; Murchie and Lilley, 1994; Weitzmann, et. al., 1997). This structure is definitely formed by single-stranded G-rich DTRs, but there are also indications that it can exist in superhelical DNA (Ahmed, et. al., 1994). Formation of G-quartets in vivo has never been directly demonstrated. DTRs consisting of regularly alternating purines and pyrimidines can adopt left-handed Z DNA conformation (Mitsui, et. al., 1970; Pohl and Jovin, 1972; Wang, et. 400

Gene Therapy and Molecular Biology Vol 3, page 401 Pearson, et. al., 1998). In this case, the loops are likely to be stabilized by CG base pairs and some non-WC pairs such as GG. Although there is some indirect evidence for S-DNA in vivo, especially during DNA replication (reviewed in Pearson and Sinden, 1998), direct proof is still lacking. The recent availability of large genomic texts of many different organisms has allowed their detailed computer analysis. This analysis (Trifonov, et. al., 1985; Karlin, 1986; Morris, et. al., 1986; Manor, et. al., 1988; Smillie and Bains, 1990; Lagercrantz, et. al., 1993; Han, et. al., 1994; Schroth and Ho, 1995; Karlin and Burge, 1996; Cox and Mirkin, 1997; Raghavan, et. al., 1997; Saunders, et. al., 1998) as well as numerous experimental approaches, including pattern matching (Galas, et. al., 1985), word frequency counting (Karlin and Burge, 1995) and basic linguistic techniques (Pevzner, et. al., 1989; Pevzner, et. al., 1989), has revealed that simple repeated sequences are remarkably abundant in natural DNAs, particularly in eukaryotic genomes. We have recently carefully evaluated the representation of different repeat types in prokaryotes, eukaryotes and bacteria, and compared those values with the expected frequencies based on the local DNA base composition (Cox and Mirkin, 1997). This analysis led to several important conclusions. It became evident that simple DNA repeats of substantial length (>24 bp-long) occur in genomes with much higher frequency than it would be statistically predicted. However, genomes belonging to different kingdoms of life (prokarya, eukarya and archaea) are enriched in different types of repeats. Eukaryotic genomes showed the enrichment of all three types of simple repeats. Of all repeats, mirror repeats, and particularly H-palindromes, were the most overrepresented, reaching 109 over the chance value. Bacterial genomes and organelles have a substantial overrepresentation of inverted repeats and sometimes direct tandem repeats. In contrast, in archae none of the repeats were abundant. The enrichment in different repeats shows an interesting length dependence. The chance frequency, calculated taking into account local GC-content, exponentially decreases with the length of repeats. The actual frequency of repeat occurrence also decreased exponentially but at a much slower rate. As a result, the normalized frequencies of overrepresented repeats showed almost perfect exponential increasing lengths (F i g . 7). This is particularly interesting since, as discussed above, the structure forming ability of a repeat also increases exponentially with length. One might speculate that the abundance of long repeats may indicate an evolutionary advantage conferred by unusual DNA structures. While these questions are at the focus of numerous studies, we were specifically interested in the mechanisms of simple repeat replication. Those studies, which started from the effects of H DNA on DNA polymerization in vitro and expanded into the analysis of replication of different repeats in vivo, are outlined below.

F i g . 7 . Ratio of observed to expected frequencies of H palindromes for pro- and eukaryotic genomes. R e d c i r c l e s : H. sapiens genome; green squares: E. coli genome.

II. Effects of simple DNA repeats on DNA polymerization in vitro It has long been known that simple DNA repeats affect DNA polymerization in vitro, presumably via the unusual conformation of the DNA template. Many instances of inverted repeats slowing down different DNA polymerases, most likely due to hairpin formation, have been described (Sherman and Gefter, 1976; Chalberg and Englund, 1979; Huang and Hearst, 1980; Kaguni and Clayton, 1982; Weaver and DePamphilis, 1982; Bedinger, et. al., 1989). Tetraplex-forming repeats also inhibit DNA synthesis carried out by many different DNA polymerases (Woodford, et. al., 1994; Usdin and Woodford, 1995; Weitzmann, et. al., 1997). The polymerization arrest was K+-dependent, strongly indicating the role of intrastranded G-quartets. Finally, numerous homopurinehomopyrimidine stretches are known to impede DNA polymerization, presumably due to triplex formation (Lapidot, et. al., 1989; Baran, et. al., 1991; Dayn, et. al., 1992; Samadashwily, et. al., 1993; Mikhailov and Bogenhagen, 1996; Krasilnikov, et. al., 1997). However, the detailed mechanisms of repeat-caused polymerization blockage were largely unknown until recently. Our interest to the effects of repeated DNA on polymerization arose from the pioneering studies of Manor and colleagues (Lapidot, et. al., 1989; Baran, et. al., 1991). They found that polymerization by Klenow or Taq polymerases on single-stranded DNA templates was partially blocked within d(GA)n or d(CT)n tracts. Since polymerization halted in the middle of those stretches, they suggested that when the newly synthesized DNA strand reached the center of a stretch its remaining part folded

401

Krasilnikova et al:Replication of Simple DNA Repeats supercoiled templates containing these structures. The location of the termination sites differed for various isoforms but always coincided with triplex boundaries as defined by chemical probing (F i g . 8 C ). We concluded, therefore, H DNA prevents DNA polymerization. Subsequently, we analyzed DNA polymerization through H-forming repeats in double-stranded open circular DNAs, where the triplex structure did not exist prior to polymerization (Samadashwily, et. al., 1993). DNA polymerization stopped almost completely at the center of those sequences but only when the homopyrimidine strand served as a template. Mutations that destroyed H-forming potential of a repeat abolished polymerization blockage, while compensatory mutations restoring H-forming potential, restored polymerization arrest as well. We concluded that the formation of H-r DNA during DNA polymerization was responsible for the observed polymerization arrest. During DNA synthesis on a doublestranded template, the DNA polymerase must displace the non-template DNA strand. When the displaced segment contains the purine-rich half of an H-motif, it can fold back to form an intramolecular triplex downstream of the polymerase, which, in turn, blocks polymerase progression (F i g . 8 B ). The severity of the triplex-caused polymerization blockage led us to wonder about the mechanisms of their inhibitory effects. Several possibilities should be considered. First, under polymerization conditions, triplexes may be so much more stable than duplexes that the triplex blockage of polymerization is a simple reflection of their persistence. Second, the kinetics of polymerase passage through triplexes may be much slower than through duplexes, simulating polymerization blockage. Finally, DNA polymerases, while capable of dismantling duplexes, may be unable to do so with triplexes. Our recent study (Krasilnikov, et. al., 1997) distinguished between these possibilities. We used singlestranded DNA templates containing intrastranded H-r triplexes or control duplexes and studied the efficiency of Vent DNA polymerase passage at different temperatures and time intervals. In parallel, the stability of different triplex and duplex structures was determined in DNA melting experiments. At physiological temperatures, we found that triplexes completely block polymerization, but duplexes just slow it down several fold. Melting temperature curves showed that triplexes were only slightly more stable than the corresponding duplexes. Such small differences are unlikely to account for the dramatic differences in temperatures (up to 40째C) at which the polymerase traverses these structures. Projection of polymerase passage temperatures onto melting curves for different structures revealed that the polymerase passes triplex barriers at temperatures where they start to dissociate, whereas duplexes are overcome far below their dissociation temperatures (F i g . 9 ). This shows that DNA polymerase can slowly untangle duplexes in DNA templates, but not triplex structures.

F i g . 8 . Models of triplex-caused polymerization arrest. A. polymerization on single-stranded template; B . polymerization on linear double-stranded template; C. polymerization on supercoiled template. R e d r e c t a n g l e s : homopurine halves of an H-palindrome; green r e c t a n g l e s : homopyrimidine halves of an Hpalindrome; p u r p l e l i n e s : growing DNA strands; purple arrows: polymerization direction; b l a c k l i n e s : parent DNA strands; purple stars: polymerization stop sites.

back to form a triplex (F i g . 8A). As a result, the polymerase is trapped and is unable to continue elongation. This hypothesis was supported by the characteristic pH dependence for d(CT)n tracts, and the reversal of termination by substituting dGTP by deazaGTP which is incapable of forming Hoogsteen hydrogen bonds. We first analyzed DNA polymerization on superhelical DNA templates containing different isoforms of H-r DNA (Dayn, et. al., 1992). We found that DNA polymerase terminates at specific sites on both DNA chains within 402

Gene Therapy and Molecular Biology Vol 3, page 403 by RNase H, and DNA is unwound as evident from the release of supercoils (Reaban, et. al., 1994; Grabczyk and Fishman, 1995). The exact nature of this complex remains unknown, and several possibilities are currently considered. One idea is that RNA polymerization generates negative supercoiling upstream of the enzyme, provoking transient H DNA formation. This structure might become kinetically trapped if an RNA transcript binds to its singlestranded portion (Grabczyk and Fishman, 1995). Formation of such a trapped complex immediately upstream of the RNA polymerase might attenuate its propagation. Another hypothesis is that transcription through a homopurine-homopyrimidine sequence could create an unusually long and stable R-loop (Reaban, et. al., 1994). The non-template DNA strand could collapse onto this R-loop, forming some hydrogen bonds with either the DNA or RNA strand as possible (collapsed Rloop) (Reaban, et. al., 1994). Future studies are needed to understand the structure of this RNA/DNA complex and how it affects RNA polymerization.

III. Effects of simple DNA repeats on DNA replication in vivo.

F i g . 9 . Comparison of melting curves for a triplex and duplex with the temperatures of DNA polymerase passage through these structures. B l u e c i r c l e s : triplex melting data; r e d c i r c l e s : duplex melting data; arrows: temperatures of polymerase passage (blue: triplex template; red: duplex template).

Because DNA polymerases cannot efficiently pass structured parts of DNA templates, one might envision problems during DNA replication in vivo. It is well documented that a portion of the lagging strand DNA template (of an Okazaki fragment size) must be singlestranded in order to pursue coordinated synthesis of both DNA strands (reviewed in Kornberg and Baker, 1992). This several hundred bp-long single stranded piece can adopt a plethora of different conformations, potentially serving as roadblocks for DNA polymerase, unless accessory replication proteins, including single-stranded DNA binding proteins and DNA helicases, helped to remove them. In some cases, however, even they may not be sufficient, as indicated by observations that the replication fork as a whole stalls within some simple DNA repeats in vivo (Rao, et. al., 1988; Brinton, et. al., 1991; Rao, 1994). This consideration encouraged us to study the mode of replication fork progression through simple DNA repeats in vivo (Samadashwily, et. al., 1997; Krasilnikova, et. al., 1998). As discussed above, these repeats are widespread in natural DNAs, and can be cloned and maintained in many model systems including bacteria and yeast. This clearly shows that they are able to replicate in vivo. However, one might expect that the rate of replication fork progression through the repeated DNA is slower. Unfortunately, this is a difficult problem to study, since the normal replication rate is very fast, ranging from 1000 bp/sec in bacteria to several hundreds bp/sec in eukaryotes (reviewed in Kornberg and Baker, 1992). For example, given a 100 bplong repeat in pBR322 slows the replication fork progres-

Based on these results, it is plausible to speculate that the elongating DNA polymerase is equipped to sense the structure of the DNA template ahead of it. Single-stranded DNA is an optimal template, double-helical segments represent an obstacle which can be slowly unwound by DNA polymerase, and unusual template conformations, such as triplexes or quadruplexes, represent steady roadblocks. It is highly likely that other enzymes of DNA metabolism might experience similar problems while tracking along DNA. One important question is whether different RNA polymerases are similarly sensitive to the conformation of a DNA template. This question is less studied, but there are some provocative data suggesting that this is the case. RNA polymerase was shown to stall within or immediately after several H-palindromes (Reaban and Griffin, 1990; Reaban, et. al., 1994; Grabczyk and Fishman, 1995; Kiyama and Oishi, 1996). This stalling profoundly depended on the repeat's orientation in the transcription unit. In most cases, it occurred when the transcript carried an oligopurine stretch, though for d(A)n. d(T)n repeat, it happened for the oligo(U) transcript (Kiyama and Oishi, 1996). It was suggested that stalling occurs upon formation of a three-stranded complex between RNA and DNA strands corresponding to the H-palindrome. In this complex, RNA is resistant to RNase A but cleaved

403

Krasilnikova et al:Replication of Simple DNA Repeats plasmids in vivo using an approach called 2-dimensional neutral/neutral electrophoresis of replication intermediates. This technique was developed for mapping of the replication origins (Brewer and Fangman, 1987; Huberman, et. al., 1987) but lately has become instrumental in defining replication termination sites as well (MacAllister, et. al., 1990; Zhu, et. al., 1992; Little, et. al., 1993). Bacterial plasmids were chosen for two reasons: (i ) they replicate unidirectionally which unequivocally determines leading and lagging strands during DNA replication; (i i ) they replicate very efficiently which allows an easy isolation and analysis of replication intermediates. The idea of electrophoretic analysis of replication intermediates applied to unidirectional replication is presented in F i g . 10A. Intermediate products of plasmid replication are !-shaped. Upon cleaving these intermediates with a restriction enzyme upstream of the replication origin, they convert into bubble-shaped molecules, where the size of the bubble correlates with the duration of replication. Bubble intermediates differ in their molecular mass (ranging from 1 to 2 plasmid masses) and shape. They are separated in two dimensions: first by mass (low percentage agarose) and second by mass and shape (high percentage agarose with ethidium bromide). Southern blotting hybridization with the radioactive plasmid probe reveals a so-called bubble arc. If there are no roadblocks during DNA replication, this arc is smooth. Stalling of the replication fork at a specific DNA repeat, however, leads to the accumulation of an intermediate of a given size and shape, generating a bulge on the arc. The ratio of the signal of this bulge to the signal of the corresponding area of a smooth replication arc (relative s top s trength, RSS) is an index of replication fork retardation by the repeat. Using this approach we found that different simple DNA repeats, including d(CGG)n. d(CCG)n, d(CTG)n. d(CAG)n, d(G)n. d(C)n, d(G-A)n. d(T-C)n, etc., block the replication fork progression. The typical picture of such repeat-caused blockage is presented in F i g . 1 0 B for the d(G)32. d(C)32 repeat. In this case the RSS is "30, i.e. this repeat slows the replication down 30-fold. Notably, in all cases, longer repeats caused more profound replication stops than the shorter ones. To prove that replication stops coincides with those repeats, we used a modified version of the electrophoretic analysis of replication intermediates (Friedman and Brewer, 1995). After the first dimension of electrophoresis, replication intermediates were digested with a restriction enzyme in the gel. The enzymes selected for this analysis cut the plasmid either upstream or downstream of the repeat. As a result, a fraction of bubble-shaped intermediates converted into identical y-shaped intermediates (F i g . 1 1 A ). In the second dimension of electrophoresis, these intermediates migrate similarly and can be detected as a horizontal line upon hybridization with a probe adjacent to the replication ori. As is clear from F i g . 1 1 A , restriction cleavage downstream of the repeat (relative to the ori) would leave the bulge on the bubble-

F i g . 1 0 . Detection of repeat-caused replication blocks by 2-dimensional gel-electrophoresis. A. Schematic representation of our approach. Upper panel shows the structure of the linearized plasmid DNA. The green triangle corresponds to the replication origin, the red box corresponds to cloned repeated DNA. The lower left panel shows the shapes of different replication intermediates. The red intermediate corresponds to the one which preferentially accumulates due to repeat-caused replication blockage. The lower right panel shows the replication arc. The red circle corresponds to the replication stop site. B . Actual electrophoregram of replication intermediates of a plasmid containing a d(G)32 . d(C)32 repeat. The red arrow points to the replication stop site.

sion 10-fold, the overall plasmid replication would only be slowed from 5 sec to 6 sec. Therefore, most conventional methods of DNA replication analysis are not applicable to this problem. To solve this problem we decided to analyze the effects of different DNA repeats on the replication of bacterial 404

Gene Therapy and Molecular Biology Vol 3, page 405

F i g . 1 1 . Mapping of the replication stop sites. A. Schematic representation of 2-D gel-electrophoresis upon restriction cleavage after the first dimension. The red square shows the d(G)32 . d(C)32 insert. Purple vertical lines show HindIII and EcoRI restriction sites located upstream and downstream from the insert, respectively. The stalled replication intermediate is shown in red. Upon EcoRI digestion, this stalled intermediate should remain bubble-shaped and, thus, remain on the arc after the second dimension of electrophoresis (right panel). In contrast, upon HindIII digestion, this intermediate should become y-shaped and move onto the line after the second dimension of electrophoresis (left panel). B . Actual figures of electrophoretic separation of replication intermediates. Left panel - HindIII digestion together with the hypothetical structure of an underreplicated stalled intermediate, right panel - EcoR1 digestion. Red arrows point to replication stop sites.

arc, while upstream cleavage shifts the bulge from the bubble-arc onto the horizontal line. F i g . 11B shows a characteristic example of such mapping for the (G)32. (C)32 repeat (Krasilnikova, et. al., 1998). One can see that cleavage of the replication intermediates downstream from the repeat leaves the bulge on the bubble arc. By contrast, cleavage upstream of the repeat shifts the bulge away from the bubble arc. Thus, the replication fork is indeed stalled within the (G)32. (C)32 stretch. Note, however, that after cleavage upstream of the repeat, the bulge does not co-migrate with the horizontal line, but migrates to a point in between the bubble-arc and the horizontal line. Thus, the shape of this intermediate is

less compact than the y-shape but more compact than the bubble. To explain this migration pattern, one must assume that a portion of the lagging strand around the HindIII site in stalled replication intermediates was not yet synthesized. This will lead to an incomplete HindIII digestion and the appearance of butterfly-like DNA molecules (shown in the diagram). If this assumption is correct, we detect the underreplication of the lagging strand within the d(G)n. d(C)n sequences. Different DNA repeats mentioned above gave phenomenologically similar results in the electrophoretic analysis of the replication intermediates: (i ) they caused replication blockage; (i i ) the efficiency of replication blockage increased with repeat length; and (i i i ) the lagging 405

Krasilnikova et al:Replication of Simple DNA Repeats strand at the repeated DNA segment was underreplicated. We have found, however, that there are at least three different mechanisms responsible for the replication fork blockage by different repeats. The first mechanism applies to the expandable trinucleotide repeats such as (CGG)n. (CCG)n, . (CTG)n (CAG)n (Samadashwily, et. al., 1997). These repeats attracted very broad attention, since more than a dozen human neurological disorders were attributed to their length expansion (reviewed in Ashley and Warren, 1995; McMurray, 1995; Wells, 1996). Trinucleotide repeats expand with a length-dependent probability. In normal individuals carrying 5-to-30 repeats, expansion is highly unlikely. Individuals with repeat numbers exceeding a threshold of n"30 can transmit expanded repeats to their progeny. In the following generations, expansions become more frequent, and each subsequent expansion has a higher probability than the previous one. The latter phenomenon is likely to account for the anticipation in the inheritance of these disorders (Caskey, et. al., 1992; Bates and Lehrach, 1994) The length dependence of repeat expansion suggests the involvement of an unusual DNA secondary structure(s) (Cox and Mirkin, 1997). Supporting this, it was demonstrated that these repeats in a single-stranded state fold into imperfect hairpins stabilized by both WC and non-WC base pairs (Chen, et. al., 1995; Gacy, et. al., 1995; Yu, et. al., 1995; Petruska, et. al., 1996; Zheng, et. al., 1996; Yu, et. al., 1997). Moreover, the threshold length for expansion is similar to the threshold energy of hairpin formation (Gacy, et. al., 1995). The mechanisms of repeat expansion remain unknown, but most data implicate replication in this process. Trinucleotide repeats stall in vitro DNA polymerization (Kang, et. al., 1995; Usdin and Woodford, 1995; Ohshima and Wells, 1997). This blockage can facilitate a misalignment between the newly synthesized and the template DNA strand (Ohshima and Wells, 1997), potentially leading to expansion. In vivo, expansion of different trinucleotide repeats occurs preferentially on their 3'-ends, implying that it could be due to the miscoordination between the leading and lagging strand synthesis (Jodice, et. al., 1994; Kunst and Warren, 1994; Snow, et. al., 1994). This hypothesis is additionally supported by observations that in bacterial and yeast models, the equilibrium between the repeats' expansions or contractions depends on their positioning with regard to the replication origin (Kang, et. al., 1995; Kang, et. al., 1996; Shimizu, et. al., 1996; Freudenreich, et. al., 1997). To obtain direct data on trinucleotide repeats replication in vivo, we analyzed the replication fork movement through these repeats within bacterial plasmids using the electrophoretic approach discussed above (Samadashwily, et. al., 1997). We found that (CGG)n. (CCG)n and (CTG)n. (CAG)n repeats blocked replication fork progression, and the efficiency of blockage increased with the repeat length so that the length responsible for signifi-

F i g . 1 2 . Quantitative analysis of the replication stop strength (RSS). Purple squares: RSS for plasmids carrying the d(CGG)n -insert in the lagging strand template; red squares: RSS for plasmids carrying the d(CCG)n -insert in the lagging strand template.

F i g . 1 3 . Model of replication blockage caused by trinucleotide repeats. The structure prone strand of a repeat is depicted by a red line, while its complementary strand is shown by a green line. The purple lines show the neighboring DNA. Arrows show the 3'-ends of the growing DNA strands. When the structure-prone DNA sequence is in the lagging strand template, it can form a hairpin-like structure which might prevent the lagging strand synthesis. Since synthesis of both DNA strands is coordinated during DNA replication, this results in stalling of the whole replication fork.

cant replication stalling ("5-fold) was similar to the threshold length for expansion (F i g . 1 2 ). The inhibitory effect didn't depend on whether the repeated segment was situated in the transcribed or non-transcribed part of the

406

Gene Therapy and Molecular Biology Vol 3, page 407

F i g . 1 4 . Model for replication blockage caused by transcription through d(G) n . d(C)n repeats. Stalled RNA polymerase is shown by an orange oval. The (G)n stretches in DNA and RNA chains are depicted as red lines, while the d(C)n stretch is depicted as green line. DNA adjacent to the repeat is shown in purple. The RNA chain is depicted by a gray line except for the r(G)n stretch, shown in red. Arrows show the 3'-ends of the newly synthesized DNA and RNA chains. Transcriptional stall is believed to be caused by the formation of a stable complex between the G-rich RNA chain and its DNA template. The exact structure of this three-stranded complex remains to be established. The replication fork stops upon encountering the stalled transcription complex.

plasmid. However, it depended on the repeat's orientation relative to the replication origin. Specifically, when structure-prone strands of the repeated DNA, such as (CGG)n or (CTG) n, were in the lagging strand template, the replication blockage was the most prominent. We believe, therefore, that the unusual structure of repeated DNA in the lagging strand template is responsible for the replication blockage (F i g . 1 3 ). The second mechanisms appears to be responsible for the replication blockage caused by the d(G)n. d(C)n repeat (Krasilnikova, et. al., 1998). Similarly to the (CGG)n. (CCG)n repeats, d(G)n. d(C)n blocks replication in a length dependent manner, except much stronger: for n=30 replication is slowed down "30-fold. Unlike the trinucleotide repeats, however, replication blockage relied exclusively on the repeats' transcription so that when the d(C)n sequence served as the transcriptional template the replication was severely impaired. This led us to study transcription through d(G)n. d(C)n repeat in vivo. We found that when the d(C)n sequence served as the transcriptional template, transcription was stalled, as was detected by the accumulation of a truncated transcript. This truncated transcript contained an oligo(G) stretch. We conclude, therefore, that transcription is stalled within or immediately after the d(G)n. d(C)n repeat, and this is likely caused by the formation of a multistranded complex between the G-rich transcript and its DNA template (F i g . 1 4 ). The replication fork, in turn, can not progress through this stalled ternary complex of the RNA polymerase, the DNA template, and the r(G)n transcript (F i g . 1 4 ).

The third mechanism applies to the d(G-A)n. d(T-C)n repeat. In this case, there is also a length-dependent replication blockage. However, it depends neither on repeat's transcription, nor on its orientation relative to the replication origin (Krasilnikova et al., unpublished results). In order to clarify this situation, we studied the replication of this repeat in bacterial cells treated with chloramphenicol. This protein synthesis inhibitor is necessary for bacterial chromosome replication but is dispensable for the replication of ColE1-type plasmids. Thus, in the presence of chloramphenicol, plasmid DNA becomes profoundly amplified, while the protein content of the cell remains at best stagnant (Clewell, 1972). We found that replication stops in plasmids containing d(GA)n. d(T-C)n inserts completely disappeared under chloramphenicol treatment (Krasilnikova et al., unpublished results). It is plausible to speculate that the inhibitory effect of this repeat on replication is due to a protein binding to this repeat. The length-dependence of the d(G-A)n. d(T-C)n -caused replication blockage could be explained by assuming cooperative protein binding to the repeated DNA (F i g . 1 5 ). This situation is markedly different from the trinucleotide repeats and the d(G)n. d(C)n repeat. In the latter cases, chloramphenicol treatment does not abolish replication blockage but rather enhances it.

IV. Conclusions The fact that very different types of simple DNA repeats impede the replication fork progression in vivo 407

Gene Therapy and Molecular Biology Vol 3, page 408

F i g . 1 5 . Model for replication blockage caused by transcription through d(G-A)n . d(T-C)n repeats. DNA strands of the repeated segment are depicted by red and green lines. Purple lines show surrounding DNA. Gray ovals show cooperatively repeatbound protein molecules. The replication fork stalls upon encountering a protein/DNA complex.

might have profound biological implications. First, and most important, this may explain the remarkable length polymorphism observed for simple DNA repeats in genomic DNAs. Indeed, to bypass a roadblock involving a DNA repeat, the replication fork could either jump past it (which might cause contractions or deletions) or pull back and try again (which might case expansions). Recently, several groups have suggested models detailing the above explanation for repeat expansions and contractions (Kang, et. al., 1995; McMurray, 1995; Gordenin, et. al., 1997; Tishkoff, et. al., 1997). Note, however, that the current knowledge on the mechanisms of repeat replication is insufficient to choose between those models. Second, the interplay between transcription and replication blockage may play a role in several biological processes. In notoriously long eukaryotic genes, the collision of the replication and transcription machinery is almost inevitable. Our proposed mechanism might prevent the replication of genes that undergo active transcription. Another provocative opportunity is that stalling of the replication fork caused by transcription of repeated DNA generates DNA ends that are potentially highly recombinogenic. This may contribute to the welldocumented stimulation of genetic recombination by transcription (reviewed in Gangloff, et. al., 1994), as well as to the recombinational hotspot activity of some DNA repeats (Schon, et. al., 1989; Wahls, et. al., 1990; Weiller, et. al., 1991; Sumegi, et. al., 1997; Boan, et. al., 1998) Future studies will undoubtedly contribute to a better understanding of both the replication of simple DNA repeats and the consequences of their replication peculiarities for different processes of DNA metabolism.

Acknowledgments

We thank the current and former members of our lab Andrey Dayn, Randal Cox, Andrey Krasilnikov and Gordana Raca for their invaluable contribution for studying the effects of simple sequence repeats on DNA replication and many helpful discussions, and Randal Cox for critical reading of this manuscript. Supported by grants from the National Institutes of Health (GM54247), the National Science Foundation (MCB-9723924) and the Council for Tobacco Research (CTR-4468) to S.M.M. M.M.K. was in part supported by the Office of International Affairs of the National Cancer Institute.

References Agazie, Y. M., Burkholder, G. D., and Lee, J. S. ( 1 9 9 6 ) . Triplex DNA in the nucleus: direct binding of triplexspecific antibodies and their effect on transcription, replication and cell growth. B i o c h e m . J . 316, 461466. Ahmed, S., Kintanar, A., and Henderson, E. ( 1 9 9 4 ) . Human telomeric C-strand tetraplexes. Nature Struct. B i o l . 1, 83-88. Ashley, C., Jr., and Warren, S. T. ( 1 9 9 5 ) . Trinucleotide repeat expansion and human disease. Annu. R e v . G e n e t . 29, 703-728. Baran, N., Lapidot, A., and Manor, H. ( 1 9 9 1 ) . Formation of DNA triplexes accounts for arrests of DNA synthesis at d(TC)n and d(GA)n tracts. P r o c . N a t l . Acad. S c i . USA 88, 507-511. Bates, G., and Lehrach, H. ( 1 9 9 4 ) . Trinucleotide repeat expansions and human genetic disease. B i o e s s a y s 16, 277-284. Beal, P. A., and Dervan, P. B. ( 1 9 9 1 ) . Second structural motif for recognition of DNA by oligonucleotide-directed triple-helix formation. S c i e n c e 251, 1360-1363.

408

Gene Therapy and Molecular Biology Vol 3, page 409 P r o c . N a t l . A c a d . S c i . U S A 89, 11406-11410.

Bedinger, P., Munn, M., and Alberts, B. M. ( 1 9 8 9 ) . Sequence-specific pausing during in vitro DNA replication on double stranded DNA templates. J . B i o l . C h e m . 264, 16880-16886.

Freudenreich, C. H., Stavenhagen, J. B., and Zakian, V. A. ( 1 9 9 7 ) . Stability of a CTG/CAG trinucleotide repeat in yeast is dependent on its orientation in the genome. M o l . C e l l . B i o l . 17, 2090-2098.

Beltran, R., Martinez-Balbas, A., Bernues, J., Bowater, R., and Azorin, F. ( 1 9 9 3 ) . Characterization of the zincinduced structural transition to *H-DNA at a d(GA. CT)22

Friedman, K. L., and Brewer, B. J. ( 1 9 9 5 ) . Analysis of replication intermediates by two-dimensional agarose gel electrophoresis. M e t h . E n z y m o l . 262, 613-627.

sequence. J . M o l . B i o l . 230, 966-978.

Gacy, A. M., Goellner, G., Juranic, N., Macura, S., and McMurray, C. T. ( 1 9 9 5 ) . Trinucleotide repeats that expand in human disease form hairpin structures in vitro. C e l l 81, 533-540.

Benham, C. J. ( 1 9 8 2 ) . Stable cruciform formation at inverted repeat sequences in supercoiled DNA. B i o p o l y m e r s 21, 679-696. Bernues, J., Beltran, R., Casasnovas, J. M., and Azorin, F. ( 1 9 9 0 ) . DNA-sequence and metal-ion specificity of the formation of *H-DNA. N u c l e i c A c i d s R e s . 18, 40674073.

Galas, D. J., Eggert, M., and Waterman, M. S. ( 1 9 8 5 ) . Rigorous pattern-recognition methods for DNA sequences. Analysis of promoter sequences from Escherichia coli. J . M o l . B i o l . 186, 117-128.

Boan, F., Rodriguez, J. M., and Gomez-Marquez, J. ( 1 9 9 8 ) . A non-hypervariable human minisatellite strongly stimulates in vitro intramolecular homologous recombination. J . M o l . B i o l . 278, 499-505.

Gangloff, S., Lieber, M. R., and Rothstein, R. ( 1 9 9 4 ) . Transcription, topoisomerases and recombination. E x p e r i e n t i a 50, 261-269.

Brewer, B. J., and Fangman, W. L. ( 1 9 8 7 ) . The localization of replication origins on ARS plasmids in S. cerevisiae. C e l l 51, 463-471.

Gellert, M., Lipsett, M. N., and Davies, D. R. ( 1 9 6 2 ) . Helix formation by guanylic acid. P r o c . N a t l . A c a d . S c i . USA 48, 2013-2018.

Brinton, B. T., Caddle, M. S., and Heintz, N. H. ( 1 9 9 1 ) . Position and orientation-dependent effects of a eukaryotic Z-triplex DNA motif on episomal DNA replication in COS-7 cells. J . B i o l . C h e m . 266, 5153-5161.

Gordenin, D. A., Kunkel, T. A., and Resnick, M. A. ( 1 9 9 7 ) . Repeat expansion - all in a flap? Nature G e n e t . 16, 116-118. Grabczyk, E., and Fishman, M. C. ( 1 9 9 5 ) . A long purinepyrimidine homopolymer acts as a transcriptional diode. J . B i o l . C h e m . 270, 1791-1797.

Caskey, C. T., Pizzuti, A., Fu, Y.-H., Fenwick Jr., R. G., and Nelson, D.L. ( 1 9 9 2 ) . Triplet repeat mutations in human disease. S c i e n c e 256, 784-789.

Han, J., Hsu, C., Zhu, Z., Longshore, J. W., and Finley, W. H. ( 1 9 9 4 ) . Over-representation of the disease associated (CAG) and (CGG) repeats in the human genome. N u c l e i c A c i d s R e s . 22, 1735-1740.

Chalberg, M. D., and Englund, P. T. ( 1 9 7 9 ) . The effect of template secondary structure on vaccinia DNA polymerase. J . B i o l . C h e m . 254, 7820-7826.

Haniford, D. B., and Pulleyblank, D. E. ( 1 9 8 3 ) . Facile transition of poly[d(TG). d(CA)] into a left-handed helix in physiological conditions. J . B i o m o l . Struct. D y n a m . 1, 593-609.

Chen, X., Mariappan, S. V., Catasti, P., Ratliff, R., Moyzis, R. K., Laayoun, A., Smith, S. S., Bradbury, E. M., and Gupta, G. ( 1 9 9 5 ) . Hairpins are formed by the single DNA strands of the fragile X triplet repeats: structure and biological implications. P r o c . N a t l . Acad. S c i . USA 92, 5199-5203.

Hoogsteen, K. ( 1 9 6 3 ) . The crystal and molecular structure of a hydrogen-bonded complex between 1 methylthymine and 9 methyladenine. A c t a C r y s t . 16, 907-916.

Chen, X., Mariappan, S. V., Moyzis, R. K., Bradbury, E. M., and Gupta, G. ( 1 9 9 8 ) . Hairpin induced slippage and hyper-methylation of the fragile X DNA triplets. J . B i o m o l . S t r u c t . D y n . 15, 745-756.

Huang, C. C., and Hearst, J. E. ( 1 9 8 0 ) . Pauses at positions of secondary structure during in vitro replication of single-stranded fd bacteriophage DNA by T4 DNA polymerase. A n a l . B i o c h e m . 103, 127-139.

Clewell, D. B. ( 1 9 7 2 ) . Nature of Col E1 plasmid replication in Escherichia coli in the presence of chloramphenicol. J . B a c t e r i o l . 110, 667-676.

Huberman, J. A., Spotila, L. D., Nawotka, K. A., El-Assouli, S. M., and Davis, L. R. ( 1 9 8 7 ) . The in vivo replication origin of the yeast 2 microns plasmid. C e l l 51, 473-481.

Cox, R., and Mirkin, S. M. ( 1 9 9 7 ) . Characteristic enrichment of DNA repeats in different genomes. P r o c . N a t l . A c a d . S c i . U S A 94, 5237-5242.

Jaworski, A., Blaho, J. A., Larson, J. E., Shimizu, M., and Wells, R. D. ( 1 9 8 9 ) . Tetracycline promoter mutations decrease non-B DNA structural transitions, negative linking differences and deletions in recombinant plasmids in Escherichia coli. J . M o l . B i o l . 207, 513-526.

Dayn, A., Malkhosyan, S., Duzhy, D., Lyamichev, V., Panchenko, Y., and Mirkin, S. ( 1 9 9 1 ) . Formation of (dA-dT)n cruciforms in Escherichia coli cells under different environmental conditions. J . B a c t e r i o l . 173, 2658-2664.

Jodice, C., Malaspina, P., Persichetti, F., Novelletto, A., Spadaro, M., Giunti, P., Morocutti, C., Terrenato, L., Harding, A. E., and Frontali, M. ( 1 9 9 4 ) . Effect of trinucleotide repeat length and parental sex on phenotypic variation in spinocerebellar ataxia I. Am. J . Hum. G e n e t . 54, 959-965.

Dayn, A., Malkhosyan, S., and Mirkin, S. M. ( 1 9 9 2 ) . Transcriptionally driven cruciform formation in vivo. N u c l e i c A c i d s R e s . 20, 5991-5997. Dayn, A., Samadashwily, G. M., and Mirkin, S. M. ( 1 9 9 2 ) . Intramolecular DNA triplexes: unusual sequence requirements and influence on DNA polymerization.

Kaguni, L. S., and Clayton, D. A. ( 1 9 8 2 ) . Template-directed pausing in in vitro DNA synthesis by DNA polymerase #

409

Krasilnikova et al:Replication of Simple DNA Repeats from Drosophila melanogaster embryos. P r o c . N a t . Acad. Sci. USA 79, 983-987.

(dG-dA)n tracts arrest single stranded DNA replication in vitro. N u c l e i c A c i d s R e s . 17, 883-900.

Kang, S., Jaworski, A., Ohshima, K., and Wells, R. D. ( 1 9 9 5 ) . Expansion and deletion of CTG repeats from human disease genes are determined by the direction of replication in E.coli. Nature Genet. 10, 213-218.

Lilley, D. M. ( 1 9 8 0 ) . The inverted repeat as a recognizable structural feature in supercoiled DNA molecules. P r o c . N a t l . A c a d . S c i . U S A 77, 6468-6472. Little, R. D., Platt, T. H., and Schildkraut, C. L. ( 1 9 9 3 ) . Initiation and termination of DNA replication in human rRNA. M o l . C e l l . B i o l . 13, 6600-6613.

Kang, S., Ohshima, K., Jaworski, A., and Wells, R. D. ( 1 9 9 6 ) . CTG triplet repeats from the myotonic dystrophy gene are expanded in Escherichia coli distal to the replication origin as a single large event. J . M o l . B i o l . 258, 543-547.

Lyamichev, V. I., Mirkin, S. M., and Frank-Kamenetskii, M. D. ( 1 9 8 5 ) . A pH-dependent structural transition in the homopurine-homopyrimidine tract in superhelical DNA. J . B i o m o l . S t r u c t . D y n . 3, 327-338.

Karlin, S. ( 1 9 8 6 ) . Significant potential secondary structures in the Epstein-Barr virus genome. P r o c . N a t l . A c a d . S c i . U S A 83, 6915-6919.

Lyamichev, V. I., Mirkin, S. M., and Frank-Kamenetskii, M. D. ( 1 9 8 6 ) . Structures of homopurine-homopyrimidine tract in superhelical DNA. J . B i o m o l . S t r u c t . D y n . 3, 667-669.

Karlin, S., and Burge, C. ( 1 9 9 5 ) . Dinucleotide relative abundance extremes: a genomic signature. Trends G e n e t . 11, 283-290.

MacAllister, T., Khatri, G. S., and Bastia, D. ( 1 9 9 0 ) . Sequence-specific and polarized replication termination in vitro: complementation of extracts of tus- Escherichia coli by purified Ter protein and analysis of termination intermediates. P r o c . N a t l . Acad. S c i . USA 87, 2828-2832.

Karlin, S., and Burge, C. ( 1 9 9 6 ) . Trinucleotide repeats and long homopeptides in genes and proteins associated with nervous system disease and development. P r o c . N a t l . Acad. Sci. USA 93, 1560-1565. Kiyama, R., and Oishi, M. ( 1 9 9 6 ) . In vitro transcription of a poly(dA). poly(dT)-containing sequence is inhibited by interaction between the template and its transcripts. N u c l e i c A c i d s R e s . 24, 4577-4583.

Malkov, V. A., Voloshin, O. N., Soyfer, V. N., and FrankKamenetskii, M. D. ( 1 9 9 3 ) . Cation and sequence effects on stability of intermolecular pyrimidine-purine-purine triplex. N u c l e i c A c i d s R e s . 21, 585-591.

Kohwi, Y. ( 1 9 8 9 ) . Cationic metal-specific structures adopted by the poly(dG) region and the direct repeats in the chicken adult $A globin gene promoter. N u c l e i c A c i d s R e s . 17, 4493-4502.

Manor, H., Sridhara-Rao, B., and Martin, R. G. ( 1 9 8 8 ) . Abundance and degree of dispersion of genomic d(GA)n. d(TC)n sequences. J . M o l . E v o l . 27, 96-101.

Kohwi, Y., and Kohwi-Shigematsu, T. ( 1 9 8 8 ) . Magnesium ion-dependent triple-helix structure formed by homopurine-homopyrimidine sequences in supercoiled plasmid DNA. P r o c . N a t l . A c a d . S c i . U S A 85, 37813785.

Mariappan, S. V., Silks, L. A. 3., Chen, X., Springer, P. A., Wu, R., Moyzis, R. K., Bradbury, E. M., Garcia, A. E., and Gupta, G. ( 1 9 9 8 ) . Solution structures of the Huntington's disease DNA triplets, (CAG)n. J . B i o m o l . S t r u c t . D y n . 15, 723-744.

Kohwi, Y., Malkhosyan, S. R., and Kohwi-Shigematsu, T. ( 1 9 9 2 ) . Intramolecular dG. dGdC triplex detected in Escherichia coli cells. J . M o l . B i o l . 223, 817-822.

Martinez-Balbas, A., and Azorin, F. ( 1 9 9 3 ) . The effect of zinc on the secondary structure of d(GA. TC)n DNA sequences of different length: a model for the formation *H-DNA. N u c l e i c A c i d s R e s . 21, 2557-2562.

Kohwi, Y., and Panchenko, Y. ( 1 9 9 3 ) . Transcriptiondependent recombination induced by triple-helix formation. G e n e s D e v . 7, 1766-1778.

McClellan, J. A., Boublikova, P., Palecek, E., and Lilley, D. M. J. ( 1 9 9 0 ) . Superhelical torsion in cellular DNA responds directly to environmental and genetic factors. P r o c . N a t l . A c a d . S c i . U S A 87, 8373-8377.

Kornberg, A., and Baker, T. ( 1 9 9 2 ) . DNA Replication - 2nd. ed. (New York: W. H. Freeman and Co).

McMurray, C. T. ( 1 9 9 5 ) . Mechanisms of DNA expansion. C hr omos oma 104, 2-13.

Krasilnikov, A. S., Panyutin, I. G., Samadashwily, G. M., Cox, R., Lazurkin, Y. S., and Mirkin, S. M. ( 1 9 9 7 ) . Mechanisms of triplex-caused polymerization arrest. N u c l e i c A c i d s R e s . 25, 1339-1346.

Mikhailov, V. S., and Bogenhagen, D. F. ( 1 9 9 6 ) . Termination within oligo(dT) tracts in template DNA by DNA polymerase gamma occurs with formation of a DNA triplex structure and is relieved by mitochondrial singlestranded DNA-binding protein. J . B i o l . C h e m . 271, 30774-30780.

Krasilnikova, M. M., Samadashwily, G. M., Krasilnikov, A. S., and Mirkin, S. M. ( 1 9 9 8 ) . Transcription through a simple DNA repeat blocks replication elongation. EMBO J . 17, 5095-5102.

Mirkin, S. M., and Frank-Kamenetskii, M. D. ( 1 9 9 4 ) . HDNA and related structures. Annu. R e v . B i o p h y s . B i o m o l . S t r u c t . 23, 541-576.

Kunst, C. B., and Warren, S. T. ( 1 9 9 4 ) . Cryptic and polar variation of the fragile X repeat could result in predisposing normal alleles. C e l l 77, 853-861.

Mirkin, S. M., Lyamichev, V. I., Drushlyak, K. N., Dobrynin, V. N., Filippov, S. A., and Frank-Kamenetskii, M. D. ( 1 9 8 7 ) . DNA H form requires a homopurinehomopyrimidine mirror repeat. Nature 330, 495-497.

Lagercrantz, U., Ellegren, H., and Andersson, L. ( 1 9 9 3 ) . The abundance of various polymorphic microsatellite motifs differs between plants and vertebrates. N u c l e i c A c i d s R e s . 21, 1111-1115.

Mitsui, Y., Langridge, R., Grant, R. C., Kodama, M., Wells, R. D., Shortle, B. E., and Cantor, C. R. ( 1 9 7 0 ) . Physical

Lapidot, A., Baran, N., and Manor, H. ( 1 9 8 9 ) . (dT-dC)n and

410

Gene Therapy and Molecular Biology Vol 3, page 411 and enzymatic studies ion poly(dI-dC)•poly(dI-dC), an unusual double helical DNA. Nature 228, 1166-1169.

Pohl, F. M., and Jovin, T. M. (1 9 7 2 ) . Salt induced cooperative conformational changes of a synthetic DNA: Equilibrium and kinetic studies with poly (dG-dC). J . M o l . B i o l . 67, 375-396.

Mizuuchi, K., Mizuuchi, M., and Gellert, M. ( 1 9 8 2 ) . Cruciform structures in palindromic DNA are favored by DNA supercoiling. J . M o l . B i o l . 156, 229-243. Morris, J., Kushner, S. R., and Ivarie, R. ( 1 9 8 6 ) . The simple repeat poly(dT-dG) . poly(dC-dA) common to eukaryotes is absent from eubacteria and rare in protozoans. M o l . B i o l . E v o l . 3, 343-355.

Raghavan, S., Burma, P. K., and Brahmachari, S. K. (1 9 9 7 ). Positional preferences of polypurine/polypyrimidine tracts in Saccharomyces cerevisiae genome: implications for cis regulation of gene expression. J . M o l . E v o l . 45, 485-498.

Murchie, A. I., and Lilley, D. M. ( 1 9 9 4 ) . Tetraplex folding of telomere sequences and the inclusion of adenine bases. EMBO J. 13, 993-1001.

Rahmouni, A. R., and Wells, R. D. (1 9 9 2 ). Direct evidence for the effect of transcription on local DNA supercoiling in vivo. J . M o l . B i o l . 223, 131-144.

Ohshima, K., and Wells, R. D. ( 1 9 9 7 ) . Hairpin formation during DNA synthesis primer realignment in vitro in triplet repeat sequences from human hereditary disease genes. J . B i o l . C h e m . 272, 16798-16806.

Rao, B. S. (1 9 9 4 ). Pausing of simian virus 40 DNA replication fork movement in vivo by (dG-dA)n . (dT-dC)n tracts. Gene 140, 233-237. Rao, S., Manor, H., and Martin, R. G. (1 9 8 8 ). Pausing in simian virus 40 DNA replication by a sequence containing (dG-dA)27. (dT-dC)27. N u c l e i c A c i d s R e s . 16, 80778094.

Panayotatos, N., and Wells, R. D. ( 1 9 8 1 ) . Cruciform structures in supercoiled DNA. Nature 289, 466-470. Pearson, C. E., Eichler, E. E., Lorenzetti, D., Kramer, S. F., Zoghbi, H. Y., Nelson, D. L., and Sinden, R. R. ( 1 9 9 8 ) . Interruptions in the triplet repeats of SCA1 and FRAXA reduce the propensity and complexity of slipped strand DNA (S-DNA) formation. B i o c h e m i s t r y 37, 27012708.

Reaban, M. E., and Griffin, J. A. (1 9 9 0 ). Induction of RNAstabilized DNA conformers by transcription of an immunoglobulin switch region. Nature 348, 342-344. Reaban, M. E., Lebowitz, J., and Griffin, J. A. (1 9 9 4 ). Transcription induces the formation of a stable RNA. DNA

Pearson, C. E., and Sinden, R. R. ( 1 9 9 6 ) . Alternative structures in duplex DNA formed within the trinucleotide repeats of the myotonic dystrophy and fragile X loci. B i o c h e m i s t r y 35, 5041-5053.

hybrid in the immunoglobulin # switch region. J . B i o l . C h e m . 269, 21850-21857. Rich, A., Nordheim, A., and Wang, A. H. (1 9 8 4 ). The chemistry and biology of left-handed Z-DNA. Annu. R e v . B i o c h e m . 53, 791-846.

Pearson, C. E., and Sinden, R. R. ( 1 9 9 8 ) . Trinucleotide repeat DNA structures: dynamic mutations from dynamic DNA. C u r r . O p i n . S t r u c t . B i o l . 8, 321-330.

Samadashwily, G. M., Dayn, A., and Mirkin, S. M. (1 9 9 3 ). Suicidal nucleotide sequences for DNA polymerization. EMBO J. 12, 4975-4983.

Pearson, C. E., Wang, Y. H., Griffith, J. D., and Sinden, R. R. ( 1 9 9 8 ) . Structural analysis of slipped-strand DNA (SDNA) formed in (CTG)n. (CAG)n repeats from the myotonic dystrophy locus. N u c l e i c Acids R e s . 26, 816-823.

Samadashwily, G. M., and Mirkin, S. M. (1 9 9 4 ). Trapping DNA polymerases using triplex-forming oligodeoxyribonucleotides. Gene 149, 127-136.

Peck, L. J., Nordheim, A., Rich, A., and Wang, J. C. ( 1 9 8 2 ) . Flipping of cloned d(pC-pG)n•d(pG-pC)n DNA sequences from right- to left handed helical structure by salt, Co(III), or negative supercoiling. P r o c . N a t l . Acad. S c i . USA 79, 4560-4564.

Samadashwily, G. M., Raca, G., and Mirkin, S. M. (1 9 9 7 ). Trinucleotide repeats affect DNA replication in vivo. Nature Genet. 17, 298-304. Saunders, N. J., Peden, J. F., Hood, D. W., and Moxon, E. R. (1 9 9 8 ). Simple sequence repeats in the Helicobacter pylori genome. M o l . M i c r o b i o l . 27, 1091-1098.

Petruska, J., Arnheim, N., and Goodman, M. F. ( 1 9 9 6 ) . Stability of intrastrand hairpin structures formed by the CAG/CTG class of DNA triplet repeats associated with neurological diseases. N u c l e i c A c i d s R e s . 24, 19921998.

Schon, E. A., Rizzuto, R., Moraes, C. T., Nakase, H., Zeviani, M., and DiMauro, S. (1 9 8 9 ). A direct repeat is a hotspot for large-scale deletion of human mitochondrial DNA. S c i e n c e 244, 346-349.

Pevzner, P. A., Borodovsky, M., and Mironov, A. A. (1 9 8 9 ). Linguistics of nucleotide sequences. I: The significance of deviations from mean statistical characteristics and prediction of the frequencies of occurrence of words. J . B i o m o l . S t r u c t . D y n . 6, 1013-1026.

Schroth, G. P., and Ho, P. S. (1 9 9 5 ). Occurrence of potential cruciform and H-DNA forming sequences in genomic DNA. N u c l e i c A c i d s R e s . 23, 1977-1983. Sen, D., and Gilbert, W. (1 9 8 8 ). Formation of parallel fourstranded complexes by guanine-rich motifs in DNA and its implications for meiosis. Nature 334, 364-366.

Pevzner, P. A., Borodovsky, M., and Mironov, A. A. (1 9 8 9 ). Linguistics of nucleotide sequences. II: Stationary words in genetic texts and the zonal structure of DNA. J . B i o m o l . S t r u c t . D y n . 6, 1027-1038.

Sherman, L. A., and Gefter, M. L. (1 9 7 6 ). Studies of the mechanism of enzymatic DNA elongation of by Escherichia coli DNA polymerase II. J . M o l . B i o l . 103, 61-76.

Pinnavaia, T. J., Marshall, C. L., Metterl, C. M., Fisk, C. L., Miles, T., and Becker, E. D. (1 9 7 8 ). Alkali metal ion specificity in the solution ordering of a nucleotide, 5'guanosine monophosphate. J . A m . C h e m . S o c . 100, 3625-3627.

Shimizu, M., Gellibolian, R., Oostra, B. A., and Wells, R. D. (1 9 9 6 ). Cloning, characterization and properties of plasmids containing CGG triplet repeats from the FMR-1

411

Krasilnikova et al:Replication of Simple DNA Repeats

Sinden, R. R. (1 9 9 4 ). DNA structure and function. (San Diego: Academic Press, Inc.).

gene. J . M o l . B i o l . 258, 614-626.

Schweyen, R. J. (1 9 9 1 ). A GC cluster repeat is a hotspot for mit- macro-deletions in yeast mitochondrial DNA. M o l . G e n . G e n e t . 226, 233-240.

Singleton, C. K., Klysik, J., Stirdivant, S. M., and Wells, R. D. (1 9 8 2 ). Left-handed Z DNA is induced by supercoiling in physiological ionic conditions. Nature 299, 312316.

Weitzmann, M. N., Woodford, K. J., and Usdin, K. (1 9 9 7 ). DNA secondary structures and the evolution of hypervariable tandem arrays. J . B i o l . C h e m . 272, 9517-9523.

Smillie, F., and Bains, W. (1 9 9 0 ). Repetition structure of mammalian nuclear DNA. J . T h e o r . B i o l . 142, 463471.

Wells, R. D. (1 9 9 6 ). Molecular basis of genetic instability of triplet repeats. J . B i o l . C h e m . 271, 2875-2878. Williamson, J. R., Raghuraman, M. K., and Cech, T. R. (1 9 8 9 ). Monovalent cation-induced structure of telomeric DNA: the G-quartet model. C e l l 59, 871-880.

Snow, K., Tester, D. J., Kruckeberg, K. E., Schaid, D. J., and Thibodeau, S. N. (1 9 9 4 ). Sequence analysis of the fragile X trinucleotide repeat: implications for the origin of the fragile X mutation. H u m . M o l . G e n e t . 3, 1543-1551.

Wittig, B., Wolfl, S., Dobric, T., Vahrson, W., and Rich, A. (1 9 9 2 ). Transcription of human c-myc in permeabilized nuclei is associated with formation of Z-DNA in three discrete regions of the gene. EMBO J. 11, 4653-4663.

Spiro, C., Bazett-Jones, D. P., Wu, X., and McMurray, C. T. (1 9 9 5 ). DNA structure determines protein binding and transcriptional efficiency of the proenkephalin cAMPresponsive enhancer. J . B i o l . C h e m . 270, 2770227710.

Wolfl, S., Martinez, C., Rich, A., and Majzoub, J. A. (1 9 9 6 ). Transcription of the human corticotrophin-releasing hormone gene in NPLC cells is correlated with Z-DNA formation. P r o c . N a t l . A c a d . S c i . U S A 93, 36643668.

Sumegi, A., Birko, Z., Szeszak, F., Vitalis, S., and Biro, S. (1 9 9 7 ). A short GC-rich sequence involved in deletion formation of cloned DNA in E. coli. A c t a B i o l . H u n g . 48, 275-279.

Woodford, K. J., Howell, R. M., and Usdin, K. (1 9 9 4 ). A novel K(+)-dependent DNA synthesis arrest site in a commonly occurring sequence motif in eukaryotes. J . B i o l . C h e m . 269, 27029-27035.

Sundquist, W. I., and Klug, A. (1 9 8 9 ). Telomeric DNA dimerizes by formation of guanine tetrads between hairpin loops. Nature 342, 825-829. Tishkoff, D. X., Filosi, N., Gaida, G. M., and Kolodner, R. D. (1 9 9 7 ). A novel mutation avoidance mechanism dependent on S. cerevisiae RAD27 is distinct from DNA mismatch repair. C e l l 88, 253-263.

Yu, A., Barron, M. D., Romero, R. M., Christy, M., Gold, B., Dai, J., Gray, D. M., Haworth, I. S., and Mitas, M. (1 9 9 7 ). At physiological pH, d(CCG)15 forms a hairpin containing protonated cytosines and a distorted helix. B i o c h e m i s t r y 36, 3687-3699.

Trifonov, E. N., Konopka, A. K., and Jovin, T. M. (1 9 8 5 ). Unusual frequencies of certain alternating purinepyrimidine runs in natural DNA sequences: relation to ZDNA. FEBS Lett. 185, 197-202.

Yu, A., Dill, J., Wirth, S. S., Huang, G., Lee, V. H., Haworth, I. S., and Mitas, M. (1 9 9 5 ). The trinucleotide repeat sequence d(GTC)15 adopts a hairpin conformation. N u c l e i c A c i d s R e s . 23, 2706-2714.

Usdin, K., and Woodford, K. J. (1 9 9 5 ). CGG repeats associated with DNA instability and chromosome fragility from structures that block DNA synthesis in vitro. N u c l e i c A c i d s R e s . 23, 4202-4209.

Zheng, G., Kochel, T., Hoepfner, R. W., Timmons, S. E., and Sinden, R. R. (1 9 9 1 ). Torsionally tuned cruciform and ZDNA probes for measuring unrestrained supercoiling at specific sites in DNA of living cells. J . M o l . B i o l . 221, 107-129.

Ussery, D. W., and Sinden, R. R. (1 9 9 3 ). Environmental influences on the in vivo level of intramolecular triplex DNA in Escherichia coli. B i o c h e m i s t r y 32, 62066213.

Zheng, M., Huang, X., Smith, G. K., Yang, X., and Gao, X. (1 9 9 6 ). Genetically unstable CXG repeats are structurally dynamic and have a high propensity for folding. An NMR and UV spectroscopic study. J . M o l . B i o l . 264, 323336.

Vologodskii, A. (1 9 9 2 ). Topology and physics of circular DNA. (Boca Raton: CRC Press). Vologodskii, A. V., and Frank-Kamenetskii, M. D. (1 9 8 2 ). Theoretical study of cruciform states in superhelical DNA. FEBS Lett. 143, 257-260.

Zhu, l. J., Newlon, C. S., and Huberman, J. A. (1 9 9 2 ). Localization of a DNA replication origin and termination zone on chromosome III of Saccharomyces cerevisiae. M o l . C e l l . B i o l . 12, 4733-4741.

Wahls, W. P., Wallace, L. J., and Moore, P. D. (1 9 9 0 ). Hypervariable minisatellite DNA is a hotspot for homologous recombination in human cells. C e l l 60, 95103.

Zimmerman, S. B., Cohen, G. H., and Davies, D. R. (1 9 7 5 ). X-ray fiber diffraction and model-building study of polyguanylic acid and polyinosinic acid. J . M o l . B i o l . 92, 181-192.

Wang, A. H. J., Quigley, G. J., Kolpak, F. J., Crawford, J. L., van Boom, J. H., Van der Marel, G., and Rich, A. (1 9 7 9 ). Molecular structure of a left-handed double helical DNA fragment at atomic resolution. Nature 282, 680-686. Weaver, D. T., and DePamphilis, M. L. (1 9 8 2 ). Specific sequences in native DNA that arrest synthesis by DNA polymerase alpha. J . B i o l . C h e m . 257, 2075-2086. Weiller, G. F., Bruckner, H., Kim, S. H., Pratje, E., and

412

Gene Therapy and Molecular Biology Vol 3, page 413 Gene Ther Mol Biol Vol 3, 413-422. August 1999.

Separation of the DNA replication and transactivation activities of EBNA1, the origin binding protein of Epstein-Barr virus Research Article

Derek F.J. Ceccarelli1 and Lori Frappier2* 1

Department of Biochemistry, McMaster University, 1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5 Department of Medical Genetics and Microbiology, University of Toronto, 1 Kings College Circle, Toronto, Ontario, Canada M5S 1A8 __________________________________________________________________________________________________ 2

* Correspondence: Lori Frappier, Ph.D., Phone: 416-946-3501; Fax: 416-978-6885; E-mail: lori.frappier@utoronto.ca Key words: Epstein-Barr virus, Epstein-Barr nuclear antigen 1 (EBNA1), DNA replication, origin of replication, transactivation, activation of transcription, transactivation domain. Received: 28 September 1998; accepted: 16 October 1998

Summary During latent infection of human B-lymphocytes, Epstein-Barr virus (EBV) genomes are stably maintained as DNA episomes that replicate once per cellular S phase. The replication and s e g r e g a t i o n o f t h e E B V e p i s o m e s r e q u i r e s t h e l a t e n t o r i g i n o f r e p l i c a t i o n , oriP , and one viral protein, Epstein-Barr Nuclear Antigen 1 (EBNA1). EBNA1 also activates the transcription of other latent viral genes and some cellular genes. EBNA1 f u l f i l l s all o f i t s functions by directly interacting with EBV sequences, but the functional role of EBNA1 residues outside of the DNA binding domain is not well understood. We have explored the contribution of EBNA1 regions to transactivation activity. Our results indicate that the C-terminal acidic tail of EBNA1, that was previously implicated i n transactivation, i s not required for this function. Rather, the transactivation activity resides primarily in an internal arginine-rich region (amino acids 325-376) that was previously shown t o mediate interactions at a distance between DNA-bound EBNA1 molecules as well as interactions with at least two cellular factors. An EBNA1 mutant, lacking amino acids 325-376, supports the transient replication of oriP plasmids at wild type levels but is severely impaired for transcriptional enhancement. Therefore, our results indicate that the replication and transactivation functions of EBNA1 can be separated.

I. Introduction

The latent origin of replication, oriP, was identified as an 1800 bp fragment of the EBV genome that supported the replication and stable maintenance of plasmids in EBVinfected cells (Sugden et al., 1985; Yates et al., 1984). Subsequently it was shown that only one viral protein, Epstein-Barr Nuclear Antigen 1 (EBNA1), was required for the replication and stable maintenance of oriP plasmids in dividing cells (Lupton and Levine, 1985; Yates et al., 1985). OriP contains two essential elements, the family of repeats (FR) and the dyad symmetry (DS) element, which contain 20 and 4 EBNA1 binding sites, respectively (Rawlins et al., 1985). Bidirectional replication of oriP containing plasmids initiates at or near the DS element (Gahn and Schildkraut, 1989; Niller et al., 1995) but, in most cell lines, additional EBNA1 sites, either tandem DS

Epstein-Barr virus (EBV) is a human herpes virus that establishes a latent infection in peripheral B-lymphocytes, inducing them to proliferate (reviewed in Kieff, 1996; Rickinson and Kieff, 1996). During latency, only a fraction of the viral genes are expressed and infectious virions are not produced. Multiple copies of the 172 Kb viral genome persist as double-stranded circular DNA episomes within the cell nucleus. The viral DNA replicates once per cellular S phase, in concert with the host cell chromosomes, and segregates efficiently to the daughter cells, so that a constant copy number of EBV genome per cell is maintained (Adams, 1987; Yates and Guan, 1991; Yates et al., 1985).

413

Ceccarelli and Frappier: DNA replication and transactivation activities of EBNA1 elements or the FR element is required for efficient replication (Harrison et al., 1994; Reisman et al., 1985; Wysokenski and Yates, 1989). Recombinant plasmids with more than one DS element are not amplified (Kirchmaier and Sugden, 1995), suggesting that oriP replication is regulated in a similar manner as human chromosomal origins. The FR element of oriP activates replication and is located approximately 1 Kb from the DS element. The intervening sequences do not appear to be important for oriP function since large deletions and insertions in these sequences do not affect the replication activity of oriP (Reisman et al., 1985). The FR also governs the stable segregation of oriP plasmids to daughter cells (Chittenden et al., 1989; Reisman et al., 1985). Plasmid segregation requires 8 of the 20 EBNA1 binding sites and can occur in the absence of a DS element (Chittenden et al., 1989; Middleton and Sugden, 1994). The FR performs a third function as a transcriptional enhancer of viral latency promoters (Gahn and Sugden, 1995; Langle-Rouault et al., 1998; Pugeilli et al., 1996; Reisman and Sugden, 1986; Sugden and Warren, 1989). Transactivation assays with reporter gene constructs have shown that only 6-7 of the FR EBNA1 binding sites are required for enhancer activity and that the FR activates transcription when positioned upstream or downstream of a promoter (Reisman and Sugden, 1986; Wysokenski and Yates, 1989). The three functions associated with oriP, namely DNA replication, DNA segregation and transactivation, require EBNA1. EBNA1 binds as a dimer to each of its recognition sites in the FR and DS (Ambinder et al., 1991; Frappier and O'Donnell, 1991a; Rawlins et al., 1985) and occupies oriP throughout all or most of the cell cycle (Hsieh et al., 1993). The mechanism by which EBNA1 activates replication from the DS element is not yet clear, but likely involves changes in the DNA structure of the DS and the recruitment of cellular factors. EBNA1 binds cooperatively to the multiple sites within the DS element (Harrison et al., 1994; Summers et al., 1996) causing localized distortion of the DNA (Bochkarev et al., 1996; Frappier and O'Donnell, 1992; Hearing et al., 1992; Hsieh et al., 1993; Summers et al., 1997). Transient replication of oriP plasmids requires two EBNA1 recognition sites separated by 3 bp suggesting that specific contact between the neighbouring EBNA1 dimers is important for initiation (Harrison et al., 1994). EBNA1 does not appear to possess any enzymatic activities (Frappier and O'Donnell, 1991a; Middleton and Sugden, 1992) suggesting that recruitment of cellular factors to oriP DNA through EBNA1 interactions is important for origin activation. DNA-bound EBNA1 has been shown to interact with replication protein A (RPA), the human replicative single-strand DNA binding protein, and this interaction may be important for the initiation of DNA replication (Zhang et al., 1998). The mechanism by which EBNA1 mediates the segregation of EBV episomes and oriP plasmids is thought to involve “piggybacking� on the host chromosomes. This

hypothesis stems from the observation that the EBV genome, oriP plasmids and EBNA1 all localize to the host metaphase chromosomes (Delecluse et al., 1993; Harris et al., 1985; Petti et al., 1990; Simpson et al., 1996). Finally, EBNA1 can act as both an enhancer and repressor of transcription. Enhanced expression of EBV latent promoters and reporter gene constructs occurs upon EBNA1 interaction with the FR element (Gahn and Schildkraut, 1989; Reisman and Sugden, 1986). Repression occurs when EBNA1 binds to the two recognition sites present in the Bam HI-Q region of the EBV genome (Rawlins et al., 1985; Sample et al., 1992). The latter interaction negatively regulates the expression of EBNA1 from the Qp promoter (Sample et al., 1992). EBNA1 consists of 641 amino acids and is shown schematically in Figure 1. A number of functional domains of EBNA1 have been identified but the contribution of many EBNA1 regions is still unknown. The large internal repeat of glycine and alanine residues (amino acids 101-325) is not required for EBNA1 replication, transactivation or segregation functions (Yates and Camiolo, 1988; Yates et al., 1985), but enables EBNA1 to evade cytotoxic T-lymphocyte responses (Levitskaya et al., 1995). The adjacent region of EBNA1 (325-376) is a glycine and arginine-rich domain, termed the looping domain, that mediates homotypic interactions at a distance between FR- and DS-bound EBNA1 molecules resulting in looped or linked DNA complexes (AvolioHunter and Frappier, 1998; Frappier et al., 1994; Goldsmith et al., 1993; Laine and Frappier, 1995; Mackey et al., 1995; Mackey and Sugden, 1997) and heterotypic interactions with some cellular proteins (Shire et al., submitted; Wang et al., 1997). This domain of EBNA1 has also been suggested to bind RNA (Snudden et al., 1994). The looping domain is followed by a sequence of basic amino acids (379-386) that functions as a nuclear localization signal for EBNA1 (Ambinder et al., 1991). The DNA binding and dimerization domains of EBNA1 colocalize to amino acids 459-607 (Ambinder et al., 1991; Summers et al., 1996) and high resolution structures of these EBNA1 domains have been determined (Bochkarev et al., 1996; Bochkarev et al., 1995). The extreme Cterminus of EBNA1 contains an aspartate and glutamaterich region termed the acidic tail. The acidic tail was suggested to be a transactivation domain (Ambinder et al., 1991), but assignment of a functional role for this region has not been conclusive (Kirchmaier and Sugden, 1997; Polvino-Bodnar et al., 1988; Polvino-Bodnar and Schaffer, 1992; Yates and Camiolo, 1988). Plasmids that contain oriP and express EBNA1 provide useful gene delivery vectors for gene therapy strategies due to their stable maintenance in extrachromosomal form in human cells (Franken et al., 1996; Judde et al., 1996; Robertson et al., 1996). However, a potential drawback to introducing EBNA1 into mammalian cells was revealed by the induction of B-cell neoplasia in EBNA1 transgenic mice (Wilson et al., 1996) and the induction of

414

Gene Therapy and Molecular Biology Vol 3, page 415

F i g u r e 1 . EBNA1 truncation and internal deletion mutants. EBNA1 amino acids present in each mutant are shown. The position of the DNA binding and dimerization domain, looping domain, nuclear localization signal (NLS) and other features of the EBNA1 polypeptide are shown. Residues 101-324 of the glycine-alanine repeat are not present an any of the EBNA1 proteins used in this study. The average transactivation activity of each mutant relative to wild type EBNA1 (Âą standard deviation) determined in this study are also shown.

looping domain (amino acids 325-376) is critical for the transactivation activity but not the replication activity of EBNA1, and that the C-terminal acidic tail does not contribute to transactivation.

recombinase activating genes, Rag1 and Rag2, in EBNA1expressing human peripheral B-lymphocytes (Srinivas and Sixbey, 1995). In order for the EBNA1-oriP system to be useful for human gene therapy, the transactivation activity of EBNA1 needs to be more carefully defined and disabled. To identify the transactivation domain, we have tested a series of truncated and internally deleted EBNA1 proteins for the ability to activate transcription of a reporter gene in human cells. Our results indicate that the arginine-rich

II. Results In order to determine the regions of EBNA1 that contribute to transactivation functions, we constructed a 415

Ceccarelli and Frappier: DNA replication and transactivation activities of EBNA1 series of EBNA1 truncation and internal deletion mutants (Figure 1). Mutations were designed to specifically target two regions of EBNA1, the acidic tail and the looping domain, while maintaining the integrity of the DNA binding and dimerization domains. The transactivation activity of each EBNA1 protein in human cells was determined using chloramphenicol acetyl transferase (CAT) reporter assays. C33A cells were transfected with a plasmid expressing EBNA1 or EBNA1 mutants from the cytomegalovirus (CMV) promoter and a second plasmid (pFRTKCAT) containing the CAT reporter gene under control of the oriP FR element (Reisman et al., 1985). 24 hours post transfection, lysates were prepared and CAT assays were performed using equal amounts of total protein. The percentage of acetylated chloramphenicol was monitored and plotted relative to the reaction time in order to determine the acetylation rate for each mutant (Figure 2). Expression of all mutant proteins was confirmed by Western blotting (data not shown) using antisera directed against the DNA binding domain (kindly provided by Jaap Middledorp). We first compared the transactivation activity of EBNA1 truncation mutants that lacked the C-terminal acidic tail (1-607), the N-terminal 376 amino acids including the looping domain (377-641), or both (377607) with that of wild type EBNA1. As shown in F i g u r e s 2 a n d 3 , removal of the acidic tail had no significant effect on the transactivation activity of EBNA1, while removal of the N-terminal 376 amino acids severely reduced transactivation. The small amount of transactivation activity observed with 377-641 was similar to that of the DNA binding and dimerization domain (452641) and was not significantly decreased by the removal of the acidic tail. Transactivation results from multiple experiments are summarized in Figure 1. Our results indicate that the acidic tail is not a transactivation domain and that residues between 1-376 mediate transactivation by EBNA1.

extracts (data not shown). Data from multiple experiments, summarized in Figures 1 and 3, showed that less than 1% of wild type transactivation activity was observed for the !325-376 looping domain mutant. Small deletions within the looping domain (!356-362 and !367-376) resulted in levels of CAT activity comparable to wild type EBNA1 (Figures 2 and 3). These results are consistent with our previous findings that small deletions in the looping domain do not abrogate the protein-protein interactions mediated by this region (Avolio-Hunter and Frappier, 1998; Laine and Frappier, 1995; Shire et al., 1998).

B. The looping domain is dispensable for transient replication of oriP plasmids Previous findings have suggested that the EBNA1 domains that contribute to replication and transactivation overlap since EBNA1 mutants that are deficient in only one of these two activities have not been isolated (Polvino-Bodnar and Schaffer, 1992; Yates and Camiolo, 1988). Therefore, we were interested in determining whether the !325-376 mutant that is defective for transactivation was able to support the transient replication of oriP plasmids in human cells. To this end, C33A cells were transfected with a plasmid containing both the EBNA1 expression cassette and oriP. 72 hours posttransfection, the plasmids were isolated, linearized and digested with Dpn I to remove plasmids that had not undergone replication in the human cells. A Southern blot of the recovered and replicated plasmid DNA is shown in Figure 4. Comparison of the amount of replicated plasmid DNA recovered with !325-376 and wild type EBNA1 indicated that !325-376 supports the transient replication of oriP plasmids. The average replication activity of !325376 observed in four different experiments was 95% ( Âą 45%) of the wild type replicative activity. The fact that !325-376 is functional for replication indicates that its lack of transactivation activity is not due to complete misfolding of the protein or due to an inability to enter the nucleus. The identification of an EBNA1 mutant that is defective for transactivation but active for replication suggests that these functions of EBNA1 are not entirely overlapping.

A. Transactivation is mediated by the looping domain The glycine and arginine-rich looping domain of EBNA1 is located within the N-terminal residues important for transactivation. This domain has been shown to mediate interactions between DNA-bound EBNA1 molecules (Avolio-Hunter and Frappier, 1998; Frappier et al., 1994; Goldsmith et al., 1993; Laine and Frappier, 1995; Mackey et al., 1995;), as well as with some cellular factors (Shire et al., 1998; Wang et al., 1997). To investigate the contribution of the looping domain to transactivation, we constructed internal deletions that lacked all (!325-376) or part (!356-362 and !367-376) of the looping domain and tested their ability to activate the FR-CAT reporter construct. Comparison of the acetylation rates with wild type EBNA1 (Figure 2) indicated that the removal of the fifty amino acids (!325-376) comprising the looping domain resulted in a loss of transactivation activity. All EBNA1 mutants were expressed at similar levels as determined by Western blot of transfected cell

III. Discussion We have identified a region of EBNA1 that is necessary for transactivation but not required for replication. This region maps to the EBNA1 looping domain (amino acids 325-376) which has previously been shown to mediate protein-protein interactions (Avolio-Hunter and Frappier, 1998; Frappier et al., 1994; Goldsmith et al., 1993; Laine and Frappier, 1995; Mackey et al., 1995; Shire et al., 1998; Wang et al., 1997). The EBNA1 mutant lacking the looping domain demonstrated less than 1% of the wild type transactivation activity but was functional in transient replication assays, indicating that the protein was properly folded, present in the nucleus, and capable of interacting 416

Gene Therapy and Molecular Biology Vol 3, page 417

F i g u r e 2 . Activation of a CAT reporter gene by EBNA1 mutants. C33A cells were transfected with a plasmid (pFRTKCAT) containing the CAT reporter gene under control of the oriP FR element and a plasmid expressing EBNA1 or EBNA1 mutants. Equal amounts of protein from cell lysates were tested for CAT activity and aliquots were removed at 5, 20 and 60 minute time intervals. Acetylated and unacetylated chloramphenicol was separated by thin layer chromatography and quantified by phosphorimager analysis (Molecular Dynamics).

F i g u r e 3 . Relative transactivation ability for EBNA1 and EBNA1 mutants. Transactivation rates for EBNA1 mutants were determined for each experiment and expressed as a percentage of wild type EBNA1 activity. The results displayed for each EBNA1 mutant represent the average of multiple experiments (error bars,Âą standard deviation).

417

Gene Therapy and Molecular Biology Vol 3, page 418

replication. EBNA1 mutants containing small deletions within the looping domain (!356-362 and D367-376) exhibited wild type levels of transactivation (this study), were functional for oriP plasmid replication (Shire et al., submitted) and mediated both homotypic and heterotypic protein-protein interactions (Avolio-Hunter and Frappier, 1998; Shire et al., submitted). The looping domain consists of six imperfect repeats of an eight amino acid sequence (Laine and Frappier, 1995) and the tolerance of this domain to small deletions suggests a degree of functional redundancy within this region. We have also shown that the acidic tail of EBNA1 is neither required nor sufficient for transactivation. This finding is in contrast to a previous study by Ambinder et al. (1991) that suggested the acidic tail was important for transactivation, but is in agreement with the results of Yates and Camiolo (1988), Polvino-Bodnar and Schaffer (1992) and Kirchmaier and Sugden (1997). The latter study demonstrated that an EBNA1 fragment containing the DNA binding domain and acidic tail functions as a dominant negative inhibitor of the transactivation and replication activities of wild type EBNA1 (Kirchmaier and Sugden, 1997). The looping domain of EBNA1 has a propensity to mediate both homotypic and heterotypic protein-protein interactions. Homotypic interactions occur between EBNA1 molecules bound to the FR or DS elements of oriP, resulting in the formation of looped (when interactions occur within a single oriP molecule) or linked (when interactions occur between different oriP molecules) DNA molecules (Avolio-Hunter and Frappier, 1998; Frappier et al., 1994; Frappier and O'Donnell, 1991b; Goldsmith et al., 1993; Laine and Frappier, 1995; Mackey et al., 1995; Middleton and Sugden, 1992; Su et al., 1991). Similar homotypic interactions mediated by the EBNA1 looping domain are observed when this domain is fused to the DNA binding domain of GAL4 (Laine and Frappier, 1995; Mackey et al., 1995). The looping domain of EBNA1 has also been shown to mediate heterotypic interactions with two cellular proteins (Wang et al., 1997; Shire et al., submitted). Wang et al. (1997) showed that the P32/TAP protein, previously shown to interact with a wide variety of proteins, interacts with residues 40-60 and 325-376 of EBNA1. Based on the findings that the latter EBNA1 region is important for transactivation and that a fragment of P32/TAP activates transcription when fused to the GAL4 DNA binding domain, it has been suggested that the P32/TAP interaction may be important for EBNA1mediated transactivation. A second cellular protein that interacts with residues 325-376 of EBNA1 was recently identified and called EBP2 (Shire et al., submitted). Functional assays with EBNA1 mutants show a correlation between the ability to bind EBP2 and the ability to mediate oriP plasmid segregation and transactivation. Therefore the EBNA1-EBP2 interaction

Figure 4 . Transient replication of oriP plasmids in human cells expressing EBNA1 or !325-376. C33A cells were transfected with oriP plasmids expressing EBNA1, !325-376 or no EBNA1 (pc3oriP). Plasmids were isolated 72 hours posttransfection and linearized with Xho I. 90% of each sample was further digested with Dpn I to remove unreplicated plasmid DNA (Xho I/ Dpn I). The products were separated on a 1% agarose gel and visualized by Southern blotting. Replicated plasmids were quantified by phosphorimager analysis (Molecular Dynamics).

with oriP. Our finding that the looping domain is not required for the EBNA1 replication function is consistent with the results of Kim et al. (1997) who showed that a similar EBNA1 mutant (!328-374) supported transient 418

Gene Therapy and Molecular Biology Vol 3, page 419 may be important for either or both of these EBNA1 functions. One of the reasons for defining regions within EBNA1 that mediate specific functions is to design oriP-based vectors suitable for use in gene therapy. Plasmids that contain oriP and express EBNA1 are useful because they are replicated and stabily maintained in human cells (Franken et al., 1996; Judde et al., 1996; Robertson et al., 1996). However, a negative aspect of this system stems from the observation that EBNA1 transactivates the expression of some cellular genes (Srinivas and Sixbey, 1995). The EBNA1-oriP system would be more useful for human gene therapy if a mutant of EBNA1 was utilized that was inactive for transactivation but active for DNA replication and segregation functions. While we successfully identified an EBNA1 mutant that lacked transactivation activity and replicated oriP plasmids, the mutant was not able to maintain the plasmids in long-term culture (Shire et al., submitted). A better understanding of how the EBNA1 looping domain and interacting cellular factors contribute to transactivation and segregation may facilitate the development of a safe EBV-based vehicle for the stable delivery of therapeutic genes into human cells.

IV. Materials and Methods A. Cell culture C33A cells (human papillomavirus negative cervical carcinoma) were grown in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 2 mM glutamine and penicillin/streptomycin.

B. Mammalian expression plasmids The plasmids used for mammalian transfections were derived from pcDNA3 (Invitrogen, Carlsbad Ca.). The plasmid pcDNA3 was digested with Hind III, treated with mung bean nuclease to remove the 5’ extensions and digested with Bam HI. DNA fragments encoding EBNA1 lacking the Gly-Ala repeat region (amino acids 101-324) or EBNA1 mutants additionally lacking amino acids 608-641 (EBNA 1-607), 1376 (EBNA 377-641), both 1-376 and 608-641 (EBNA 377607), or 1-451 (EBNA 452-641) were generated by PCR amplification from p205 (Yates et al., 1985) using an Nterminal primer containing an Nde I site and a C-terminal primer containing a Bam HI site. These DNA fragments were digested with Nde I, filled in with the Klenow fragment of DNA polymerase I, then digested with Bam HI. An EBNA1 mutant lacking amino acids 325-376 (!325-376) was PCR amplified from pVLE!325-376 (Avolio-Hunter and Frappier, 1998) using an N-terminal primer containing an Nco I site and a Cterminal primer containing a Bam HI site. The DNA fragment was digested with Nco I, filled in with Klenow, then digested with Bam HI. DNA fragments containing the EBNA1 mutants lacking amino acids 356-362 (!356-362) and 367-376 (!367-376) were excised by digesting pVLE!356-362 and pVLE!367-376 (Laine and Frappier, 1995; Shire et al., 1998) with Eco RI, filling in the 5’ overhang with Klenow then digesting with Bam HI. DNA fragments encoding EBNA1 or the EBNA1 mutants described above were ligated into

419

pcDNA3. Plasmids used for transient replication experiments were modified by the addition of EBV oriP DNA sequences. A DNA fragment encoding oriP was excised from pGEMoriP (Frappier and O'Donnell, 1991b) by digestion with Bam HI and Rsa I and inserted between the Bgl II and Nru I sites of pcDNA3 containing EBNA1 or EBNA1 mutants to generate pc3oriP-EBNA1.

C. Transcription enhancement assays C33A cells were seeded in 60 mm dishes at a density of 1 x 10 6 cells/dish and grown for 24 hours prior to transfection by the calcium phosphate/DNA coprecipitation method. Five micrograms of pcDNA3 plasmids encoding EBNA1 or EBNA1 mutants were combined with 2 µg of pFRTKCAT and 2.5 µg of herring sperm DNA in 0.25 ml of 0.25 M CaCl 2 then added dropwise to 0.25 ml of 2x HBS pH 6.95 (50 mM HEPES, 280 mM NaCl, 1.5 mM Na 2 HPO4 ) with vortexing. After a 30 minute incubation at room temperature, the precipitate was added dropwise to the cells and incubated for 12-16 hours at 37˚C. The cells were then washed twice with PBS, supplemented with fresh medium and incubated for 24 hours at 37˚C. After the cells were harvested, a small portion of the sample was set aside for protein analysis while the remaining cells were lysed by three rounds of freezing and thawing. The supernatant was tested for CAT reporter expression levels. CAT assays contained 50 µg of protein extract, 0.25 M TrisHCl pH 7.5, 0.25 mM acetyl CoA and 3-6 pmol of C14 chloramphenicol (NEN) in a 150 µl reaction volume . The reactions were incubated at 37˚C and 50 µl aliquots were removed at various time points. Acetylated and unacetylated products were separated on thin layer chromatography plates (Whatman) in a chloroform/methanol (95:5) mixture and quantified by phosphorimager analysis using ImageQuant software (Molecular Dynamics).

D. Transient replication assays C33A cells were plated in 10 cm dishes at 2.5 x 10 6 cells/dish and grown 24 hours prior to transfection. Transfections were performed as described for transcription assays except that 10 µg of pc3 oriP-EBNA1 plasmid DNA and 10 µg of herring sperm DNA was used and the reaction volumes were doubled. Following removal of the DNA precipitate, cells were washed in PBS, split into 150 mm dishes and grown for 72 hours. Cells from each plate were collected and lysed in 700 µl of 10 mM Tris-HCl pH 7.5, 10 mM EDTA, 0.6% SDS. High molecular weight DNA was precipitated by the addition of NaCl to 0.83 M and incubated overnight at 4˚C. Low molecular weight DNA in the supernatant was extracted with phenol:chloroform (1:1), ethanol precipitated and resuspended in TE pH 8.0. Half of each sample was linearized with Xho I and 90% of the linearized samples was subsequently digested with Dpn I (4 units) for 2 hours at 37˚C. DNA fragments from the restriction digests were separated on a 1% agarose gel, transferred to Gene Screen Plus (NEN Research Products) and probed with pc3oriP that had been labeled with P32 -dCTP by random primer extension. Radiolabelled bands were visualized by autoradiography and quantified by phosphorimager analysis using ImageQuant software (Molecular Dynamics).

Ceccarelli and Frappier: DNA replication and transactivation activities of EBNA1 increase expression of the Epstein-Barr virus LMP gene. J . V i r o l . 69, 2633-2636.

Acknowledgements We gratefully acknowledge Dr. Bill Sugden for the pFRTKCAT reporter plasmid. This work was supported by a grant from the National Cancer Institute of Canada. LF is a Medical Research Council of Canada Scientist.

Goldsmith, K., Bendell, L., and Frappier, L. (1 9 9 3 ). Identification of EBNA1 amino acid sequences required for the interaction of the functional elements of the EpsteinBarr virus latent origin of DNA replication. J . V i r o l . 67, 3418-3426. Graham, F. L., and van der Eb, A. J. (1 9 7 3 ). A new technique for the assay of infectivity of human adenovirus 5 DNA. V i r o l o g y 5, 456-467.

References Adams, A. (1 9 8 7 ). Replication of latent Epstein-Barr virus genomes in Raji cells. J . V i r o l . 61, 1743-1746.

Harris, A., Young, B. D., and Griffin, B. E. (1 9 8 5 ). Random association of Epstein-Barr virus genomes with host cell metaphase chromosomes in Burkitt's lymphoma-derived cell lines. J . V i r o l . 56, 328-332.

Ambinder, R. F., Mullen, M., Chang, Y. N., Hayward, G., and Hayward, S. D. (1 9 9 1 ). Functional domains of EpsteinBarr virus nuclear antigen EBNA-1. J . V i r o l . 65, 14661478.

Harrison, S., Fisenne, K., and Hearing, J. (1 9 9 4 ). Sequence requirements of the Epstein-Barr virus latent origin of DNA replication. J . V i r o l . 68, 1913-1925.

Avolio-Hunter, T. M., and Frappier, L. (1 9 9 8 ). Mechanistic studies on the DNA linking activity of the Epstein-Barr nuclear antigen 1. N u c l . A c i d s R e s . 26, 4462-4470.

Hearing, J., Mulhaupt, Y., and Harper, S. (1 9 9 2 ). Interaction of Epstein-Barr virus nuclear antigen 1 with the viral latent origin of replication. J . V i r o l . 66, 694-705.

Bochkarev, A., Barwell, J. A., Pfuetzner, R. A., Furey Jr., W., Edwards, A. M., and Frappier, L. (1 9 9 5 ). Crystal structure of the DNA-binding domain of the Epstein-Barr virus origin-binding protein EBNA1. C e l l 83, 39-46.

Hirt, B. (1 9 6 7 ). Selective extraction of polyoma DNA from infected mouse cell culture. J . M o l . B i o l . 26, 365-369.

Bochkarev, A., Barwell, J. A., Pfuetzner, R. A., Bochkareva, E., Frappier, L., and Edwards, A. (1 9 9 6 ). Crystal structure of the DNA-binding domain of the Epstein-Barr virus origin-binding protein, EBNA1, bound to DNA. C e l l 84, 791-800.

Hsieh, D. J., Camiolo, S. M., and Yates, J. L. (1 9 9 3 ). Constitutive binding of EBNA1 protein to the EpsteinBarr virus replication origin, oriP, with distortion of DNA structure during latent replication. E M B O J . 12, 49334944.

Chittenden, T., Lupton, S., and Levine, A. J. (1 9 8 9 ). Functional limits of oriP, the Epstein-Barr virus plasmid origin of replication. J . V i r o l . 63, 3016-3025.

Judde, J. G., Spangler, G., Magrath, I., and Bhatia, K. (1 9 9 6 ). Use of Epstein-Barr virus Nuclear Antigen 1 in targetted therapy of EBV-associated neoplasia. Human Gene Ther. 7, 647-653.

Delecluse, H.-J., Bartnizke, S., Hammerschmidt, W., Bullerdiek, J., and Bornkamm, G. W. (1 9 9 3 ). Episomal and integrated copies of Epstein-Barr virus coexist in Burkitt lymphoma cell lines. J . V i r o l . 67, 1292-1299.

Kieff, E. (1 9 9 6 ). Epstein-Barr virus and its replication. p. 2343-2396 in F i e l d s V i r o l o g y , B. N. Fields, D. M. Knipe, and P. M. Howley, (eds.), third ed., LippincottRaven Publishers, Philadelphia.

Franken, M., Estabrooks, A., Cavacini, L., Sherburne, B., Wang, F., and Scadden, D. T. (1 9 9 6 ). Epstein-Barr virus driven gene therapy for EBV-related lymphomas. Nature M e d i c i n e 2, 1379-1382.

Kim, A. L., Maher, M., Hayman, J. B., Ozer, J., Zerby, D., Yates, J. L., and Lieberman, P. M. (1 9 9 7 ). An imperfect correlation between DNA replication activity of EpsteinBarr virus nuclear antigen 1 (EBNA1) and binding to the nuclear import receptor, Rch1/Importin ". V i r o l o g y 239, 340-351.

Frappier, L., and O'Donnell, M. (1 9 9 1 a ). Overproduction, purification and characterization of EBNA1, the origin binding protein of Epstein-Barr virus. J . B i o l . C h e m . 266, 7819-7826.

Kirchmaier, A. L., and Sugden, B. (1 9 9 5 ). Plasmid maintenance of derivatives of oriP of Epstein-Barr virus. J . V i r o l . 69, 1280-1283.

Frappier, L., and O'Donnell, M. (1 9 9 1 b ). Epstein-Barr nuclear antigen 1 mediates a DNA loop within the latent replication origin of Epstein-Barr virus. P r o c . N a t l . Acad. Sci. USA 88, 10875-10879.

Kirchmaier, A. L., and Sugden, B. (1 9 9 7 ). Dominantnegative inhibitors of EBNA-1 of Epstein-Barr virus. J . V i r o l . 71, 1766-1775.

Frappier, L., and O'Donnell, M. (1 9 9 2 ). EBNA1 distorts oriP, the Epstein-Barr virus latent replication origin. J . V i r o l . 66, 1786-1790.

Laine, A., and Frappier, L. (1 9 9 5 ). Identification of EpsteinBarr virus nuclear antigen 1 protein domains that direct interactions at a distance between DNA-bound proteins. J . B i o l . C h e m . 270, 30914-30918.

Frappier, L., Goldsmith, K., and Bendell, L. (1 9 9 4 ). Stabilization of the EBNA1 protein on the Epstein-Barr virus latent origin of DNA replication by a DNA looping mechanism. J . B i o l . C h e m . 269, 1057-1062. Gahn, T. A., and Schildkraut, C. L. (1 9 8 9 ). The Epstein-Barr virus origin of plasmid replication, oriP, contains both the initiation and termination sites of DNA replication. C e l l 58, 527-535.

Langle-Rouault, F., Patzel, V., Benavente, A., Taillez, M., Silvestre, N., Bompard, A., Sczakiel, G., Jacobs, E., and Rittner, K. (1 9 9 8 ). Up to 100-fold increase of apparent gene expression in the presence of Epstein-Barr virus oriP sequences and EBNA1: Implications of the nuclear import of plasmids. J . V i r o l . 72, 6181-6185.

Gahn, T. A., and Sugden, B. (1 9 9 5 ). An EBNA-1 dependent enhancer acts from a distance of 10 kilobase pairs to

Levitskaya, J., Coram, M., Levitsky, V., Imreh, S., Steigerwald-Mullen, P. M., Klein, G., Kurilla, M. G., and

420

Gene Therapy and Molecular Biology Vol 3, page 421 Masucci, M. G. (1 9 9 5 ). Inhibition of antigen processing by the internal repeat region of the Epstein-Barr virus nuclear antigen-1. Nature 375, 685-688. Lupton, S., and Levine, A. J. (1 9 8 5 ). Mapping of genetic elements of Epstein-Barr virus that facilitate extrachromosomal persistence of Epstein-Barr virusderived plasmids in human cells. M o l . C e l l . B i o l . 5 , 2533-2542.

Sambrook, J., Fritsch, E. F., and Maniatis, T. (1 9 8 9 ). Vol. 3 in Molecular C l o n i n g : A Laboratory Manual, Cold Spring Harbor Laboratories Press, Cold Spring Harbour. Sample, J., Henson, E. B. D., and Sample, C. (1 9 9 2 ). The Epstein-Barr nuclear antigen I promoter active in type I latency is autoregulated. J . V i r o l . 66, 4654-4661.

Mackey, D., Middleton, T., and Sugden, B. (1 9 9 5 ). Multiple regions within EBNA1 can link DNAs. J . V i r o l . 69, 6199-6208.

Shire, K., Ceccarelli, D. F.J, Avolio-Hunter, T. M., and Frappier, L. EBP2, a human protein that interacts with sequences of the Epstein-Barr nuclear antigen 1 important for plasmid maintenance. (submitted).

Mackey, D., and Sugden, B. (1 9 9 7 ). Studies on the mechanism of DNA linking by Epstein-Barr virus nuclear antigen 1. J . B i o l . C h e m . 272, 29873-29879.

Simpson, K., McGuigan, A., and Huxley, C. (1 9 9 6 ). Stable episomal maintenance of yeast artificial chromosomes in human cells. M o l . C e l l . B i o l . 16, 5117-5126.

Middleton, T., and Sugden, B. (1 9 9 2 ). EBNA1 can link the enhancer element to the initiator element of the EpsteinBarr virus plasmid origin of DNA replication. J . V i r o l . 66, 489-495.

Snudden, D. K., Hearing, J., Smith, P. R., Grasser, F. A., and Griffin, B. E. (1 9 9 4 ). EBNA1, the major nuclear antigen of Epstein-Barr virus, resembles 'RGG' RNA binding proteins. EMBO. J. 13, 4840-4847.

Middleton, T., and Sugden, B. (1 9 9 4 ). Retention of plasmid DNA in mammalian cells is enhanced by binding of the Epstein-Barr virus replication protein EBNA1. J . V i r o l . 68, 4067-4071.

Srinivas, S. K., and Sixbey, J. W. (1 9 9 5 ). Epstein-Barr virus induction of recombinase-activating genes RAG1 and RAG2. J . V i r o l . 69, 8155.

Niller, H. H., Glaser, G., Knuchel, R., and Wolf, H. (1 9 9 5 ). Nucleoprotein complexes and DNA 5'-Ends at oriP of Epstein-Barr virus. J . B i o l . C h e m . 270, 1286412868.

Su, W., Middleton, T., Sugden, B., and Echols, H. (1 9 9 1 ). DNA looping between the origin of replication of Epstein-Barr virus and its enhancer site: Stabilization of an origin complex with Epstein-Barr nuclear antigen 1. P r o c . N a t l . A c a d . S c i . U S A 88, 10870-10874.

Petti, L., Sample, C., and Kieff, E. (1 9 9 0 ). Subnuclear localization and phosphorylation of Epstein-Barr virus latent infection nuclear proteins. V i r o l o g y 176, 563574.

Sugden, B., Marsh, K., and Yates, J. (1 9 8 5 ). A vector that replicates as a plasmid and can be efficiently selected in Blymphoblasts transformed by Epstein-Barr virus. M o l . C e l l . B i o l . 5, 410-413.

Polvino-Bodnar, M., Kiso, J., and Schaffer, P. A. (1 9 8 8 ). Mutational analysis of Epstein-Barr virus nuclear antigen 1 (EBNA1). N u c l e i c A c i d s R e s . 16, 3415-3435.

Sugden, B., and Warren, N. (1 9 8 9 ). A promoter of EpsteinBarr virus that can function during latent infection can be transactivated by EBNA-1, a viral protein required for viral DNA replication during latent infection. J . V i r o l . 63, 2644-2649.

Polvino-Bodnar, M., and Schaffer, P. A. (1 9 9 2 ). DNA binding activity is required for EBNA1-dependent transcriptional activation and DNA replication. V i r o l o g y 187, 591-603. Pugeilli, M. T., Woisetschlaeger, M., and Speck, S. H. (1 9 9 6 ). OriP is essential for EBNA gene promoter activity in Epstein-Barr virus-immortalized lymphoblastoid cell lines. J . V i r o l . 70, 5758-5768. Rawlins, D. R., Milman, G., Hayward, S. D., and Hayward, G. S. (1 9 8 5 ). Sequence specific DNA binding of the EpsteinBarr virus nuclear antigen (EBNA-1) to clustered sites in the plasmid maintenance region. C e l l 42, 859-868. Reisman, D., and Sugden, B. (1 9 8 6 ). Transactivation of an Epstein-Barr viral transcriptional enhancer by the Epstein-Barr viral nuclear antigen 1. M o l . C e l l . B i o l . 6, 3838-3846. Reisman, D., Yates, J., and Sugden, B. (1 9 8 5 ). A putative origin of replication of plasmids derived from EpsteinBarr virus is composed of two cis-acting components. M o l . C e l l . B i o l . 5, 1822-1832. Rickinson, A. B., and Kieff, E. (1 9 9 6 ). Epstein-Barr Virus. p. 2397-2446 in F i e l d s V i r o l o g y , B. N. Fields, D. M. Knipe, and P. M. Howley, (eds.), third ed., LippincottRaven Publishers, Philadelphia. Robertson, E. S., Ooka, T., and Kieff, E. D. (1 9 9 6 ). EpsteinBarr virus vectors for gene delivery to B-lymphocytes. P r o c . N a t l . A c a d . S c i . U S A 93, 11334-11340.

421

Summers, H., Barwell, J. A., Pfuetzner, R. A., Edwards, A. M., and Frappier, L. (1 9 9 6 ). Cooperative assembly of EBNA1 on the Epstein-Barr virus latent origin of replication. J . V i r o l . 70, 1228-1231. Wang, Y., Finan, J. E., Middeldorp, J. M., and Hayward, S. D. (1 9 9 7 ). P32/TAP, a cellular protein that interacts with EBNA-1 of Epstein-Barr virus. V i r o l o g y 236, 18-29. Wilson, J. B., Bell, J. L., and A. J. Levine. (1 9 9 6 ). Expression of Epstein-Barr virus nuclear antigen-1 induces B-cell neoplasia in transgenic mice. E M B O J . 15, 3117-3126. Wysokenski, D. A., and Yates, J. L. (1 9 8 9 ). Multiple EBNA1-binding sites are required to form an EBNA1dependent enhancer and to activate a minimal replicative origin within oriP of Epstein-Barr virus. J . V i r o l . 63, 2657-2666. Yates, J., Warren, N., Reisman, D., and Sugden, B. (1 9 8 4 ). A cis-acting element from the Epstein-Barr viral genome that permits stable replication of recombinant plasmids in latently infected cells. P r o c . N a t l . A c a d . S c i . U S A 81, 3806-3810. Yates, J. L., and Camiolo, S. M. (1 9 8 8 ). Dissection of DNA replication and enhancer activation functions of EpsteinBarr virus nuclear antigen 1. C anc e r C e lls 6, 197-205.

Ceccarelli and Frappier: DNA replication and transactivation activities of EBNA1 Yates, J. L., and Guan, N. (1 9 9 1 ). Epstein-Barr virus derived plasmids replicate only once per cell cycle and are not amplified after entry into cells. J . V i r o l . 65, 483-488. Yates, J. L., Warren, N., and Sugden, B. (1 9 8 5 ). Stable replication of plasmids derived from Epstein-Barr virus in various mammalian cells. Nature 313, 812-815. Zhang, D., Frappier, L., Gibbs, E., Hurwitz, J., and O'Donnell, M. (1 9 9 8 ). Human RPA (hSSB) interacts with EBNA1, the latent origin binding protein of Epstein-Barr virus. N u c l e i c A c i d s R e s . 26, 631-637.

422

Gene Therapy and Molecular Biology Vol 3, page 423 Gene Ther Mol Biol Vol 3, 423-435. August 1999.

The activation of the chicken lysozyme locus in development is a cooperative process Research Article

Matthias C. Huber 1 and Constanze Bonifer1,2 1

Institut fĂźr Biologie III, Schänzlestr.1, 79104 Freiburg, Germany. 2Molecular Medicine Unit, University of Leeds, St James University Hospital, Leeds LS9 7TF, UK __________________________________________________________________________________________________ C o r r e s p o n d i n g a u t h o r : Constanze Bonifer, Ph.D. Molecular Medicine Unit, University of Leeds, St James’s University Hospital, Leeds LS9 7TF, UK. Tel: +44-113-2065676, Fax: +44-113-2444475; email: c.bonifer@leeds.ac.uk Received: 25 July 1998; accepted 16 October 1998

Summary T h e c h i c k e n l y s o z y m e g e n e i s a marker for the myelomonocytic lineage o f the hematopoietic system. In early experiments we demonstrated that correct activation of the chicken lysozyme locus in macrophages of transgenic mice requires the complete set of cis-regulatory elements. Different cis-elements are activated at distinct developmental stages and their chromatin structure i s differentially remodelled. We have shown that the early onset of transcriptional activation of the chicken lysozyme locus is entirely dependent on enhancer elements which are structurally activated early in development (-6.1 kb and -3.9 kb early enhancers). However, the structural reorganization o f t h e early enhancers requires the presence o f promoter sequences. We concluded from these experiments that the early enhancers and the promoter cooperate in order to activate the lysozyme l o c u s . Subsequently, we performed experiments aimed at elucidating the cis-regulatory requirements of chromatin rearrangement at the early enhancers. The - 6 . 1 kb enhancer i s well characterized at the molecular level and all transcription factors contributing to its activity in transfection studies are known. We have placed this element into a new sequence context on the lysozyme locus by deleting extended flanking regions and analyzed this construct in transgenic mice. Surprisingly, its chromatin rearrangement ability as judged from DNaseI hypersensitive site formation was impaired. We conclude from this experiment that the cooperation of enhancer core and flanking sequences is necessary for enhancer activity. We hypothesize that all sequences of a gene locus serve a purpose in the developmental control of its activation.

a number of different cis-regulatory elements distributed over large distances. In addition it has been demonstrated that, depending on the developmental stage, different combinations of transcription factors can occupy the same cis-regulatory element (Roque et al., 1996; Gualdi et al., 1996). It is known that different members of the same transcription factor family can be differentially expressed in development. Thus, some transcription factors will only transiently occupy cis-regulatory elements. At present it is not clear, whether only one type or different factor family members can occupy the same binding site within a larger protein complex at the various developmental stages.

I. Introduction One of the key objectives in developmental biology is to understand how the regulatory information from the genome is translated into the controlled expression of different genes. The gradual change in the activity of individual gene loci is the basis for the various steps in the commitment of a differentiating cell towards the terminally differentiated state. A large number of gene expression studies have been performed to describe cellular differentiation processes. However, due to the difficult nature of the experiments involved, the molecular details of cell differentiation at the level of the genome still remain to be elucidated. Eukaryotic genes are regulated by

Another open question concerns the order of transcription factor assembly. Many cis-regulatory 423

Huber and Bonifer: Activation of the lysozyme locus in development elements are organized in positioned nucleosomes (Richard -Foy and Hager, 1987; Straka and Hörz, 1991). Since not all transcription factors are capable of binding when their recognition sequences are organized in a nucleosomal core (Blomquist et al., 1996b; Taylor et al., 1991; Pina et al., 1990) this, in turn, implies that a certain order of factor complex assembly may be necessary for correct regulation. In addition, factors capable of recruiting chromatin modifying enzyme like histone acetylases might have to pave the way for other factors joining the transcription complex later in the assembly process (reviewed in: Grunstein, 1997; Peterson and Tamkun, 1995).

full set of regulatory elements (Bonifer et al., 1990). Deletion of one enhancer region abolishes position independence of expression, indicating that for position independent transgene expression the cooperative action of all cis-regulatory elements is necessary (Bonifer et al., 1994b). Repression of gene expression by genomic position effects is correlated with suppression of DHS formation and with an inefficient reorganization of nucleosomes in the cis-regulatory regions (Huber et al., 1994; Huber et al., 1996), indicating that active chromatin formation and transcriptional activity are closely linked. The structural activation of the lysozyme locus takes place in several steps. Accordingly, the individual enhancer elements of the lysozyme locus can be categorized into early or late enhancers. The early enhancers at –6.1kb and –3.9kb and the promoter become DNaseI hypersensitive at the myeloblast stage when the gene is also transcriptionally activated. A low level of gene expression is observed. The DHS at the silencer element is still present. The DHS at the late –2.7kb enhancer appears only later in differentiation. Simultaneously, the -2.4kb silencer disappears (Huber et al., 1995). Each cis-regulatory element shows an distinct structural organization, with transcription factor binding sites specifically arranged with respect to nucleosomes. Transcriptional activation leads to a rearrangement of chromatin structure in an elementspecific fashion (Huber et al., 1996). The results of our structural studies suggest that the correct alignment of transcription factor binding sites with respect to the position of nucleosomes may be essential for their interaction and thus for position independent expression.

Last, but not least, the role of sequences flanking cisregulatory elements has been elusive. The cores of cisregulatory elements are very often separated by long arrays of flanking DNA which on first sight seem to serve no purpose. However, for genes whose role in different species have been conserved, like the globin or the Hox gene clusters, we also find a conservation of gene order and spacing, suggesting a regulatory role of such sequences. It is therefore tempting to speculate that a eukaryotic gene locus is not just a collection of cis-regulatory elements separated by “junk” DNA, but that there is more to it than that.

II. The lysozyme locus as a model for gene locus activation In order to answer these questions we have been studying the molecular basis of transcriptional activation of the chicken lysozyme locus in the myeloid lineage of the mammalian hematopoietic system. The lysozyme locus is small (21 kb), thus facilitating the manipulation of individual cis-regulatory elements within the context of an entire genomic locus. Expression of the lysozyme gene is regulated by a combination of several cis-regulatory elements located in the 5’-half of the locus. Three enhancers, 6.1kb-, 3.9 kb- and 2.7 kb upstream of the transcriptional start site as well as a silencer element at 2.4 kb and a complex promoter (Baniahmad et al., 1987; Grewal et al., 1992; Hecht et al., 1988; Theisen et al., 1986; Steiner et al., 1987; Baniahmad et al., 1990; Luckow and Schutz, 1989; Sippel et al., 1987a) have been identified. All active cis-regulatory elements colocalize with DNaseI hypersensitive sites (DHSs) in chromatin (Fritton et al., 1984; Fritton et al., 1987; Sippel et al., 1988; Huber et al., 1995; Sippel et al., 1996).

III. The activation of the chicken lysozyme locus in development requires the interaction of a subset of enhancer elements with the promoter Together with the promoter, each enhancer is capable of activating the gene locus specifically in mature macrophages of transgenic mice (Bonifer et al., 1994b; Jägle et al., 1997). However, the temporal regulation of their activity is different, since the early enhancers and the promoter are sufficient to activate the chicken lysozyme gene at the correct, early developmental stage, whereas a deletion of the early -6.1kb enhancer leads to a delay in gene activation, indicating that this element plays a crucial role in the initial activation process (Jägle et al., 1997). The presence of the -3.9 kb enhancer alone is insufficient for the early onset of transcription.

In order to be able to draw relevant conclusions regarding the contribution of each cis-element to lysozyme gene regulation, we first determined in transgenic mice which of several modified constructs was specifically expressed in the right cell type (macrophages) and was unaffected by chromosomal position effects. This holds true for the complete chicken lysozyme locus carrying the

We subsequently examined the role of the promoter in lysozyme locus activation by analyzing a construct carrying the complete lysozyme locus with an internal deletion of the promoter in transgenic mice (Huber et al., 1997). Transcription from this construct was completely 424

Gene Therapy and Molecular Biology Vol 3, page 425 abolished. However, the formation of a DHS at the -2.4 kb silencer element was unaffected and in macrophages, the DHS at the -2.7 kb enhancer element was formed. In contrast, DHS formation and chromatin remodelling at the early –6.1 kb and -3.9 kb enhancers were abolished. Taken together our data indicate that in the initial activation of the lysozyme locus the early enhancers have to interact with the promoter.

and thus the spatial arrangement of the regulatory elements. The TR/RXR heterodimer is able to bind its recognition sites within chromatin and to repress or activate transcription in the absence or presence of thyroid hormone (TH) (Wong et al., 1995). However, since the chromatin reorganization in the –2.4 kb/–2.7 kb region is cell differentiation dependent and can be induced in cultured cells solely by induction with LPS (Huber et al., 1995; Faust et al., 1997), we assume that it is independent of the presence of a TR/RXR ligand and thus is at least partially driven by transcription factors recognizing the enhancer element in a differentiation dependent fashion. We could indeed demonstrate a LPS stimulated binding of members of the C/EBP transcription factor family to the -2.7 kb enhancer (Faust N, Bonifer C, and Sippel AE, submitted). Whether the formation of the DHS at the silencer element is the structural prerequisite for the promoter-independent DHS formation of the -2.7 kb enhancer or whether the factors binding to the enhancer are capable of disrupting chromatin by themselves remains to be tested.

How this interaction takes place and which transcription factors are involved is at present unknown. However, since our initial promoter deletion removed defined sequence elements from the complete lysozyme locus without any further changes, we are now in the position to replace defined sequence elements in the promoter and thus analyze transcription factor interaction truly involved in gene locus activation.

IV. The –2.4 kb silencer and the -2.7 kb enhancer are able to reorganize chromatin in the absence of a promoter The -2.4kb silencer is inactive in mature, lysozymeexpressing macrophages and is active in all other cell types analyzed. It extends from -2310bp to -2410bp and carries a recognition sequence for thyroid (TR) or retinoic acid (RXR) hormone receptors. The second identified protein is the chicken homologue of factor CTCF (NeP1) (Köhne et al., 1993; Baniahmad et al., 1990; Arnold et al., 1996; Burcin et al., 1997). Our transgenic mouse experiments now show that the silencer element is capable of forming a DHS in any cell type also in the absence of a promoter. To our surprise, in macrophages of such mice not only the -2.4kb DHS but also the –2.7kb DHS was formed. This type of chromatin rearrangement is normally correlated with maximal transcriptional activity of the lysozyme gene at late macrophage differentiation stages.

V. Results and Discussion A. Enhancer promoter interaction in lysozyme locus activation Our earlier experiments have demonstrated the existence of two different classes of cis-regulatory elements of the lysozyme locus. The early enhancers at -6.1 kb and 3.9 kb need a promoter in order to remodel chromatin, whereas the silencer element at -2.4 kb and the enhancer at -2.7 kb do not (Figure 2). In turn, our functional experiments showed that the correct timing of activation of the lysozyme locus is dependent on the presence of the 6.1 kb enhancer. Figure 1 depicts our present model of the various cell differentiation dependent regulatory states of the lysozyme locus as deduced from our structural and functional experiments. In our structural experiments we have employed retrovirally-transformed myeloid cell lines representing various fixed differentiation states.

We could show that the presence or absence of the silencer element has no influence on the time course of activation of the chicken lysozyme gene in developing macrophages (Jägle et al., 1997). It is therefore possible that this element is repressing the activity of the –2.7kb enhancer at early stages of macrophage differentiation. In both, chicken and mouse macrophages, the increase in MNase- and DNaseI accessibility at the enhancer parallels a decrease in accessibility at the silencer (Huber et al., 1995; Huber et al., 1996; Sippel et al., 1988). Both elements are located on adjacent positioned nucleosomes (Huber et al., 1996), whereby the factor binding sites possibly face the same nucleosomal side. Such a spatial arrangement suggests that silencer and enhancer are an integrated cisregulatory element, with factor binding at both subelements being mutually exclusive. DNA bending, shown to be mediated by the active silencer complex on this site (Arnold et al., 1996), may influence nucleosome phasing

In differentiating primary cells we envisage the structural reorganization of the complete lysozyme locus as a dynamic process which involves the assembly and cooperative action of a large number of transcription factors as well as chromatin modifying enzymes. In order to understand the molecular details of this process we have to fully characterize the transcription factor composition of each cis regulatory element. However, here we encounter a problem: as with most other gene loci, most cis-regulatory elements of the lysozyme locus have been defined and analyzed the classical way, that is, by transient transfection assays as well as using in vitro and in vivo DNA binding studies.

425

Gene Therapy and Molecular Biology Vol 3, page 426

Figure 1. Model of the various steps in the activation of the chicken lysozyme locus during macrop h a g e d i f f e r e n t i a t i o n . (A) In lysozyme non-expressing cells we only find a DNaseI hypersensitive chromatin site (DHS) at the silencer complex located -2.4 kb upstream of the transcription start. (B ) In myeloblasts, which represent a cell type still capable of differentiating to granulocytes and macrophages, DHS form at the promoter and at the early (-6.1 kb and -3.9 kb) enhancers. The presence of promoter sequences is essential for DHS formation at the early enhancers.The early enhancers and the promoter are sufficient to activate the lysozyme locus at the correct developmental stage. (C) In terminally differentiated and activated macrophages the silencer complex is inactivated and the third enhancer becomes active leading to a 20-fold increase in mRNA levels. The position of the different cis-regulatory elements are indicated by black bars, the coding region with the first two exons is depicted by striped boxes, and the various cis-regulatory protein complexes are shown as differently patterned shapes. The transcriptional activity of the lysozyme gene is indicated by arrows of different sizes. Note that enhancer complexes are extending over larger regions than the actual core sequences.

transfection assays in macrophage cell lines (Steiner et al., 1987). Consequently, only the proteins binding to this fragment have been identified as being necessary for promoter function (Altschmied et al., 1989; DÜlle and Strätling, 1990) (Figure 3A). Using a similar strategy, the functional elements of the -6.1 kb enhancer have been assigned to reside on a sequence array of approximately 200 bp within a 596 bp BamHI - HindIII fragment encompassing the DHS at this position (Figure 3B). The minimal enhancer fragment showing full enhancer activity in transient transfection assays was assigned to an even smaller fragment (Grewal et al., 1992). In neither case was it known whether the minimal elements, as defined by transient transfection assays, are sufficient for their correct activation during ontogeny.

Figure 3 shows what we know about the functional elements of the -6.1 kb enhancer and the promoter based on those experiments. In our initial promoterless construct we had deleted sequences from -1 bp to -830 bp. The promoter is a complex element, since it contains three start sites for the RNA polymerase as well as a large collection of ubiquitous and cell-specific transcription factor binding sites. Transcription factor composition of the promoter in oviduct and macrophages is different, which is reflected in a difference of DHS fine structure in the two tissues (Sippel et al., 1988). In macrophages, an additional pair of DHS at -0.7 kb is formed. However, promoter activity seems to entirely reside on a fragment of approximately 250 bp in length; further upstream, an element has been identified which exerts a negative effect on reporter gene expression in transient 426

Gene Therapy and Molecular Biology Vol 3, page 427

Figure 2. Chromatin structure of the wt and promoter-less chicken lysozyme constructs in transgenic m i c e . Summary of chromatin structure analyses of different constructs in different cell types of transgenic mice as indicated on the left. At the top of each panel the 5´- region of the chicken lysozyme constructs is depicted. The promoter deletion is indicated as a black triangle. The main transcription start is indicated by a horizontal arrow. Exons 1 and 2 are symbolized by grey boxes. Black arrows: DHS displayed at wild type strength irrespective of the chromosomal location of the transgene. Grey arrows: –2.4 kb- and –2.7 kb DHS displaying changes in intensity according to the developmental stage of the cells. Striped arrows: DHS forming with variable efficiency depending on the chromosomal position of the transgene. (A): construct carrying the full set of cis-regulatory elements; (B ): construct carrying a promoter deletion.

line, also in the lung, a tissue which can contain are large number of macrophage cells.

B. Deletion of flanking sequences abolishes DHS formation at the -6.1 kb enhancer

The wild type lysozyme locus construct encompassing the complete chromatin domain is expressed also in the brain (Bonifer et al., 1990; Bonifer et al., 1994b). Expression is truly ectopic, but cell-specific and copy number-dependent, indicating that the trans-species transgene is activated correctly in a brain cell type fortuitously containing the correct transcription factor combination (Bonifer et al., 1994a). Brain-specific expression is lost when the -6.1 kb enhancer is deleted (Bonifer et al., 1994b). Interestingly, as opposed to the wild type locus, no ectopic expression of the transgene in the brain is observed with the deletion construct described here, indicating, that the -6.1 kb enhancer is not active in the brain of these mice.

Our original deletion of the -6.1 kb enhancer which abolished the early onset of transcriptional upregulation had removed a large fragment encompassing sequences from -5400 bp up to -8700 bp. We were now interested to examine whether the 596 bp BamHI - HindIII fragment carried all the information for its interaction with the promoter and, hence, for its chromatin remodelling activity. We generated two transgenic mouse lines which carried a deletion construct as described above but where the 596 bp fragment carrying all the transcription factor binding sites found to be necessary for enhancer activity had been reinserted, thus removing large flanking sequence arrays and placing the -6.1 kb core enhancer element into a new sequence context (F i g u r e 5 B ). The deletion also removed a non-tissue specific weak DHS at -7.9 kb (see Figure 5 lane 15) of no apparent cis-regulatory function. As expected, the deletion did not affect the tissue specificity of expression.

Structural studies confirmed these findings. We isolated macrophages from transgenic mice carrying either the full set of cis-regulatory elements (XS.0b mice) or the above described deletion construct (BH596.1 and BH596.3 mice) and analyzed the DHS pattern in the -6.1 kb- and the -3.9 kb enhancer region (F i g u r e 4 ). We also examined the nucleosomal pattern around the -2.4 kb silencer/ -2.7 kb enhancer region using MNase digestion of nuclei of lysozyme non-expressing embryonic fibroblasts and macrophages of XS.0b and BH 596 mice.

As in the transgenic mouse lines carrying the complete locus, the lysozyme gene is expressed in a tissue specific fashion (Figure 4). Expression is only observed in hematopoietic tissues and, in case of the BH 596.3 mouse

427

Huber and Bonifer: Activation of the lysozyme locus in development

Figure 3. DNA - Protein interactions at the Promoter (A) and at the early -6.1 kb enhancer (B). Summary of in vitro - and in vivo DNA protein interactions at the promoter (A) and at the -6.1 kb enhancer (B) of the chicken lysozyme locus. Transcription factor binding sites as are indicated as differently patterned boxes, the nature of the presumed transcription factor binding to this sequence is indicated above the sequence. The data represent in (A) a compilation of the data from the following references: (Altschmied et al., 1989; DÜlle and Strätling, 1990; Grez et al., 1981). LS defines linker scan mutations leading to a inhibition of promoter activity as assayed by (Luckow and Schutz, 1989). The arrows in (B) indicate the position of point mutations leading to a loss in enhancer activity as described in (Grewal et al., 1992). The data in (B) represent a compilation of the following references: (Grewal et al., 1992; Sippel et al., 1987a; Borgmeyer et al., 1984; Sippel et al., 1987b).

428

Gene Therapy and Molecular Biology Vol 3, page 429

F i g u r e 4 . T i s s u e s p e c i f i c e x p r e s s i o n p a t t e r n o f B H 5 9 6 m i c e . mRNA expression analysis of BH596 mouse lines. Expression of the chicken lysozyme gene in different tissues of two independently derived BH596 transgenic mouse lines. Total RNA (20Âľg) was analyzed in an S1 protection assay with probes specific for chicken lysozyme ( upper panel ) or mouse Ă&#x;-actin (l o w e r p a n e l) as described (Bonifer et al., 1990). Abbreviations below the lanes indicate the investigated tissues / cell types. L: liver; H: heart; K: kidney; Lg: lung; S: spleen; B: bone marrow; T: thymus; M: peritoneal macrophage. Lane (-): no RNA; HD11: RNA prepared from HD11 cells stimulated with LPS. The numbers at the right indicate the positions of the three major start sites at the lysozyme promoter (Grez et al., 1981).

Embryonic fibroblasts show a regular MNase pattern indicative of a phased nucleosome over both elements which is perturbed after the activation of the -2.7 kb enhancer in macrophages (Figure 6). This pattern is identical in both mouse lines indicating that the deletion does not affect the nucleosomal organization of these elements. In addition, no difference between the mouse lines was observed with respect to the formation of the DHS at the -3.9 kb enhancer. We have already shown that the presence or absence of the -6.1 kb enhancer has no effect on the formation of an active promoter structure (Huber et al., 1996). However, the deletion of flanking sequences strongly affected DHS formation at the -6.1 kb enhancer. In contrast to the mice carrying the complete lysozyme locus which show a strong DHS at the position of the -6.1 kb enhancer, no DHS is formed at the position of reinserted BamHI ! Hind III fragment. This indicates that flanking regions are required for DHS formation at the -6.1 kb enhancer. The DNA region required to stabilize a functional enhancer complex at -6.1 kb is obviously much larger than previously anticipated.

organized in a phased nucleosome which is remodelled by enhancer activation (Huber et al., 1996). It is therefore possible that this preset chromatin structure is disturbed in case of the enhancer deletion, which would in turn lead to a disturbance of enhancer activation. One of the crucial transcription factors involved in enhancer activity is nuclear factor I (NFI) which by itself in unable to bind to DNA organized in a nucleosome irrespective of nucleosome positioning (Pina et al., 1990; Blomquist et al., 1996a). In vivo, this factor requires assistance to be able to bind to its recognition sequence in chromatin (Perlmann and Wrange, 1988; Truss et al., 1995). In our case, this assistance might be prohibited by a change in chromatin architecture.

At present we do not fully understand the molecular basis of our finding. Several explanations are possible which most likely are not mutually exclusive. Firstly, we know that in its inactive state the -6.1 kb enhancer is

Thirdly, the constitutive hypersensitive site at -7.9 kb may be involved in stabilizing the -6.1 kb enhancer complex.

Secondly, it is possible that earlier in vivo and in vitro DNA-binding studies have failed to detect low affinity binding sites for factors binding to flanking sequences recruited by the factors binding to the enhancer cores. In vivo a large complex might be formed which is too fragile to be reconstituted in vitro using conventional extract preparation and assembly technology.

429

Gene Therapy and Molecular Biology Vol 3, page 430

Figure 5. The deletion of flanking sequences around the core of the -6.1 kb enhancer abolishes DNAseI h y p e r s e n s i t i v e s i t e f o r m a t i o n a t t h e - 6 . 1 k b e n h a n c e r b u t n o t a t t h e - 3 . 9 k b e n h a n c e r . (A) DHS mapping with macrophages of mouse line XS.0b (lanes 2 - 5) which expresse the lysozyme gene in a position independent manner and mouse lines BH596.3 and BH596.1 (lanes 5 - 9 and lanes 10 - 13, respectively). Lane 15: Nuclei prepared from chicken HD11 promacrophage cells. M: size marker. Genomic DNA was prepared, restricted with SphI, transferred to a nylon membrane and hybridized with probe DS indicated at the map at the right. The map on the right indicates the position of SphI restriction sites and the position of the -6.1 kb enhancer DHS in the wt- (grey circle) and the deletion construct (white circle). Note that in the BH596 construct one SphI site is deleted. The position of the -3.9 kb DHS is indicated by a grey oval bar. (B ) Map of the wild type locus (upper panel) and the BH596 deletion construct (lower panel). The coding region is indicated by the white box with the exon sequences drawn as black bars and the transcriptional start site as horizontal arrow. The positions of the DHSs mapped in macrophages are shown as vertical arrows, constitutive DHSs are indicated as smaller arrows. The position of the upstream enhancer region and the medial enhancer region are indicated as stippled boxes. The nature of the cis regulatory elements and their position relative to the transcriptional start site are shown in the lowest panel. E: enhancer element; S: silencer element; P: promoter elements. The position and nature of the transcription factors binding to the 596 bp BamHI - HindIII fragment are indicated below the line.

430

Gene Therapy and Molecular Biology Vol 3, page 431

Figure 6. The deletion of upstream sequences does not affect nucleosomal remodelling at the -2.7 kb enha nc e r . MNase analysis of the –2.4 kb silencer/–2.7 kb enhancer region in the BH596.3 mouse line. Lanes 1 - 6: MNase digestion pattern of nuclei prepared from mouse line XS.0b (restricted with SphI - SacI); lanes 1-3: analysis of MNase digestion pattern in the chromatin of transgenic mouse macrophages; lanes 4 - 6: embryonic fibroblasts. Lane 7: DHS pattern of HD11 nuclei in the analyzed region (symbolized by small grey circles). Lanes 8 - 15: MNase analysis of nuclei prepared from macrophages (lanes 8 - 11) or embryonic fibroblasts (lane 12 - 15) of mouse line BH596.3. The probe used for indirect endlabelling (SpS) is indicated by a stippled box on the map depicted at the right. Prominent MNase cleavage sites are indicated by arrows, cleavage sites only observed in macrophages are indicated by striped arrows. M: Size markers.

deacetylase. It is therefore tempting to speculate that the DNaseI-resistant chromatin of the domain borders has spread into the newly inserted -6.1 kb enhancer sequences thus rendering this element inactive. Taken together, these explanations suggest that with our present technology in identifying cis-regulatory elements we are missing out essential regulatory features of eukaryotic gene loci which are crucial for locus activation in a developmentallyregulated system. It also demonstrates that it is important to analyse the composition of cis-regulatory elements in their natural sequence context.

Finally, it is possible that the deletion of sequences upstream of the -6.1 kb have brought this element too close to the boundaries of the DNaseI sensitive domain. It has been demonstrated that these regions have insulator properties (Stief et al., 1989). Recently, it could be shown that they contain high affinity binding sites for the chicken homologue of the methyl-binding-protein MeCP2, which acts as a transcriptional repressor (Weitzel et al., 1997). The repressor activity of this protein can be explained by the finding that it is capable of recruiting a histone

431

Huber and Bonifer: Activation of the lysozyme locus in development

C. All sequences of a eukaryotic locus are part of a functional unit

however, points to the presence of important transcription factor binding sites. It may also be possible that structural information is present which, at the moment, we are unable to identify.

The activation of the chicken lysozyme locus during cell differentiation is a stepwise process. Our analysis of the developmental activation of the lysozyme locus has demonstrated a definitive requirement for all cis-regulatory elements of the gene locus to cooperate. The chicken lysozyme locus harbors no single element with dominant chromatin opening function. Although an element exists which is able to stably reconfigure chromatin in the absence of promoter elements, it acts later in cell differentiation and its chromatin reorganizing capacity is limited to its site. The differentiation-dependent reorganization and activation of the lysozyme locus is mediated by the interplay of separate cis-regulatory elements with distinct abilities to generate or maintain transcription competent chromatin structures. Our results support the concept that all essential cis-regulatory elements (enhancer and promoter elements) have to be integrated into one functional entity to perform locus activation in transgenic mice.

What we can detect with our classical techniques like DNaseI hypersensitive site mapping, in vitro DNA binding studies and transient transfection studies may be nothing but the “tip of the iceberg� in terms of clusters of high affinity transcription factor binding sites. We might be unable to detect scattered single or low affinity factor binding sites. Taken together it is obvious that gene locus activation is a cooperative process. What emerges from our work and the above described experiments is the concept that this cooperative process may involve all sequences of a eukaryotic gene locus, some of which span hundreds of kilobases. The elucidation of the type of information encoded in these sequences and the way this information is translated into the enormous complexities of developmentally-controlled gene expression will be a major challenge for developmental biologists in the next years.

In the study described here, we demonstrate in addition that the characterization of the early -6.1 kb enhancer by transient transfection assays has been incomplete and that not only a promoter but also enhancer flanking sequences are required for the developmentally-controlled chromatin remodelling activity of this element. Several laboratories have independently detected a cis-regulatory role for enhancer core flanking regions. The results of our experiments are reminiscent of a study in which the activity of one of the major control elements of the human adenosine deaminase (ADA) gene was examined. This control region is located in the first intron and is essential for the correct activation of the ADA locus in transgenic mice (Aronow et al., 1995). It has also been demonstrated in this system that a core enhancer region is insufficient for the activation of the enhancer in transgenic mice. Flanking regions are required which have no enhancer activity on their own.

Acknowledgements The authors thank Gudrun KrĂźger for expert technical assistance and Dr. Louise Coletta, Molecular Medicine Unit, for critically reading the manuscript. This work was supported by a grant from the Deutsche Forschungsgemeinschaft to C.B

VI. Materials and Methods A. Construction of pIIIilysBH596 The pIIIilysBH596 plasmid was constructed by cloning the BamHI ! HindIII fragment carrying the -6.1 kb enhancer (Theisen et al., 1986) in sense orientation into the Asp718 site (at -5.4 kb upstream of the trasncription start) of plasmid PIIIilysdXK (Bonifer et al., 1994b) which carried a deletion of a fragment between -5.4 kb and -8.7 kb.

In another study the role of sequences outside the cores of the DNaseI hypersensitive sites of the "-globin LCR has been examined (Jackson et al., 1996). Also here, a cisregulatory role of these regions could be established. It was found that a strong synergism in transcriptional stimulation was observed when sequences outside the cores were present and the natural spacing between the hypersensitive sites was preserved. A sequence comparison between "-globin LCRs of different mammals indeed revealed a high level of sequence conservation of certain core flanking sequences (Slightom et al., 1997), indicating an important role of these sequences in LCR function. At present it is unknown which type of information is encoded in enhancer core flanking regions. The fact that individual sequence motifs are evolutionary conserved,

B. Transgenic mice, cell culture and mRNA expression analysis Production of the BH596 transgenic mouse lines by pronuclear injection of DNA was essentially performed as described in Hogan et al. (1994). First-generation heterozygous mice from the founders BH959.1 and BH593.3 were examined for intact integration and construct integrity by Southern blotting. Copy-numbers were calculated from Southern blots as described by Bonifer et al. (1990) and phosphorimager analysis. BH 596.1 mice carried 2 copies and BH596.3 mice carried 18 copies of the lysozyme locus. Expression and chromatin analysis were performed with homozygous progeny. Transgenic mouse lines carrying the XS construct (Bonifer et al., 1994b) were kept as homozygous lines in our mouse colony. Primary macrophages were

432

Gene Therapy and Molecular Biology Vol 3, page 433

References

prepared from the peritoneal cavity of transgenic mice as described (Bonifer et al., 1990). For each transgenic mouse line, cells from 12 - 20 mice were taken in culture in standard Iscove´s medium supplemented with 10% fetal calf serum (FCS) and 10% L-cell conditioned medium for 16 hours. Embryonic fibroblasts were prepared from mouse embryos 12 days after fertilization as described earlier (Huber et al., 1994). HD11 cells were grown in standard Iscove´s medium containing 8% FCS and 2% chicken serum. Preparation of mRNA and the S1 protection analyses were performed as described in (Bonifer et al., 1990).

Altschmied, J., Müller, M., Baniahmad, S., Steiner, S., and Renkawitz, R. (1 9 8 9 ). Cooperative interaction of chicken lysozyme enhancer sub-domains partially overlapping with a steroid receptor binding site. N u c l e i c A c i d s R e s . 17, 4975-4991. Arnold, R., Burcin, M., Kaiser, R., Muller, M., and Renkawitz, R. (1 9 9 6 ). DNA bending by the silencer protein NeP1 is modulated by TR and RXR. N u c l e i c A c i d s R e s . 24, 2640-2647. Aronow, B., Ebert, C.A., Valerius, M.T., Potter, S.S., Wiginton, D.A., Witte, D.P., and Hutton, J.J. (1 9 9 5 ). Dissecting a locus control region: facilitation of enhancer function by extended enhancer-flanking sequences. M o l . C e l l . B i o l . 15, 1123-1135.

C. Nuclei preparation Nuclei were prepared by homogenizing cultured cells on ice with a Dounce homogenizer in buffer 1 (0.15mM spermine, 0.5mM spermidine, 15mM Tris-HCl pH 7.5, 60mM KCl, 15mM NaCl, 2mM EDTA, 0.5mM EGTA, 500mM Sucrose, 1mM PMSF) followed by centrifugation for 5 min at 1000g at 4°C. Nuclei were washed once in buffer 2 (buffer 1 + 0.5% Triton X-100), followed by a wash in buffer 3 (buffer 1 but with 350mM instead of 500mM Sucrose). After this wash nuclei were centrifuged for 5 min at 600g at 4°C.

Baniahmad, A., Müller, M., Steiner, C., and Renkawitz, R. (1 9 8 7 ). Activity of two different silencer elements of the chicken lysozyme gene can be compensated by enhancer elements. EMBO J. 6, 2297-2303. Baniahmad, A., Müller, M., Steiner, C., and Renkawitz, R. (1 9 9 0 ). Modular structure of chicken lysozyme silencer: involvement of an unusual thyroid receptor binding site. C e l l 61, 729-740.

D. DNaseI and MNase digestion analysis

Blomquist, P., Li, Q., and Wrange, O. (1 9 9 6 a ). The affinity of nuclear factor I for its DNA site is drastically reduced by nucleosome organization irrespective of its rotational or translational position. J . B i o l . C h e m . 271, 153-159.

Aliquots of 2 x 10 7 t o 1 x 1 0 8 nuclei in 100-200µl of buffer 3 were centrifuged for 5 min at 600g and 4°C and thereafter resuspended in buffer 4 (0.15mM spermine, 0.5mM spermidine, 15mM Tris-HCl pH 7.5, 60mM KCl, 15mM NaCl, 0.2mM ETDA, 0.2mM EGTA). DNaseI digestions were performed in 500µl buffer 4. To 2x10 7 nuclei 0, 4, 10, 20 and 40 Units/ml DNaseI (Pharmacia) were added. HD11 nuclei were digested with 24 Units/ml DNaseI. Digestion was started by adding 4mM MgCl 2 and 2mM CaCl 2 . Incubations (15 min, 4°C) were stopped by adding 10µl 0.5M EDTA. MNase1 digestions were performed in 200µl buffer 4. To 2x10 7 nuclei 0, 15, 80 Units MNase (Pharmacia) were added. Digestion was started by adding 10µl CaCl 2 (100mM) and stopped after incubation (5 min, 25°C) by the addition of 10µl 0.5M EDTA.

Blomquist, P., Li, Q., and Wrange, Ö. (1 9 9 6 b ). The affinity of nuclear factor I for its DNA site is drastically reduced by nucleosome organisation irrespective of its rotational or translational position. J . B i o l . C h e m . 271, 153-159. Bonifer, C., Vidal, M., Grosveld, F., and Sippel, A.E. (1 9 9 0 ). Tissue specific and position independent expression of the complete gene domain for chicken lysozyme in transgenic mice. EMBO J. 9, 2843-2848. Bonifer, C., Bosch, F., Faust, N., Schuhmann, A., and Sippel, A.E. (1 9 9 4 a ). Evolution of gene regulation as revealed by differential regulation of the chicken lysozyme transgene and the endogenous mouse lysozyme gene in mouse macrophages. Eur. J. Biochem . 226, 227-235.

Digestion of naked genomic DNA with MNase was performed in 150µl 10mM Tris-HCl pH7.5 with 0.2 - 6.4 Units/ml MNase. Incubations (15 min, 25°C) were started by adding 15µl CaCl 2 (10mM) and stopped with 15µl 50mM EDTA. After DNaseI or MNase digestion nuclei were lysed in 500µl Tris-HCl pH 8.0, 2mM ETDA, 0.2% SDS, 0.5mg/ml Proteinase K and incubated overnight at 37°C. RNase A (0.2mg/ml) was then added and after a further incubation at 37°C for 1h the DNA was precipitated three times with ethanol. Digested DNA was cleaved with restriction enzymes for indirect end-labelling analysis and 7 - 30µg of fragmented DNA were loaded on 3mm thick vertical 1% agarose gels (DNaseI analysis) or 10mm thick vertical 1.5% agarose gels (MNase analysis). The DNA was transferred to Biodyne B membrane and the filter was hybridized with an appropriate probe (a SphI-SpeI fragment from -3163 to -2906bp) for indirect endlabelling.

Bonifer, C., Yannoutsos, N., Krüger, G., Grosveld, F., and Sippel, A.E. (1 9 9 4 b ). Dissection of the locus control function located on the chicken lysozyme gene domain in transgenic mice. N u c l e i c A c i d s R e s . 22, 4202-4210. Borgmeyer, U., Nowock, J., and Sippel, A.E. (1 9 8 4 ). The TGGCA-binding protein: A eucaryotic nuclear protein recognizing a symmetrical sequence in double stranded linear DNA. N u c l e i c A c i d s R e s . 12, 4295-4311. Burcin, M., Arnold, R., Lutz, M., Kaiser, B., Runge, D., Lottspech, F., Filipova, G.N., Lobanenkov, V.V., and Renkawitz, R. (1 9 9 7 ). Negative protein 1, which is required for the function of the chicken lysozyme gene silencer in conjunction with hormone receptors is identical to the multivalent zinc finger repressor CTCF. M o l . C e l l . B i o l . 17, 1281-1288.

433

Huber and Bonifer: Activation of the lysozyme locus in development Dölle, A. and Strätling, W.H. (1 9 9 0 ). Genomic footprinting of proteins interacting with the chicken lysozyme promoter. Gene 95, 187-193.

Huber, M.C., Krüger, G., and Bonifer, C. (1 9 9 6 ). Genomic position effects lead to an inefficient reorganization of nucleosomes in the 5`-regulatory region of the chicken lysozyme locus in transgenic mice. N u c l e i c Aci ds R e s . 24, 1443-1453.

Faust, N., Huber, M.C., Sippel, A.E., and Bonifer, C. (1 9 9 7 ). Different macrophage populations develop from embryonic/fetal and adult hematopoietic tissues. E x p . H e m a t o l o g y 25, 432-444.

Jackson, J.D., Petrykowskaja, H., Philipsen, S., Miller, W., and Hardison, R. (1 9 9 6 ). Role of DNA sequences outside the cores of DNase hypersensitive sites (HSs) in function of the ß-globin locus control region. J . B i o l . C h e m . 271, 11871-11876.

Fritton, H.P., Igo-Kemenes, T., Nowock, J., Strech-Jurk, U., Theisen, M., and Sippel, A.E. (1 9 8 4 ). Alternative sets of DNase I-hypersensitive sites characterize the various functional states of the chicken lysozyme gene. Nature 311, 163-165.

Jägle, U., Müller, A.M., Kohler, H., and Bonifer, C. (1 9 9 7 ). Role of positive and negative cis-regulatory elements regions in the regulation of transcriptional activation of the lysozyme locus in developing macrophages of transgenic mice. J . B i o l . C h e m . 272, 5871-5879.

Fritton, H.P., Igo-Kemenes, T., Nowock, J., Strech-Jurk, U., Theisen, M., and Sippel, A.E. (1 9 8 7 ). DNase Ihypersensitive sites in the chromatin structure of the lysozyme gene in steroid hormone target and non-target cells. B i o l . C h e m . H o p p e - S e y l e r 368, 111-119.

Köhne, A.C., Banaihmad, A., and Renkawitz, R. (1 9 9 3 ). A ubiquitous transcription factor synergizes with v-erbA in transcriptional silencing. J . M o l . B i o l . 232, 747-755.

Grewal, T., Theisen, M., Borgmeyer, U., Grussenmeyer, T., Rupp, R.A.W., Stief, A., Qian, F., Hecht, A., and Sippel, A.E. (1 9 9 2 ). The -6.1-kilobase chicken lysozyme enhancer is a multifactorial complex containing several cell-type-specific elements. M o l . C e l l . B i o l . 12, 2339-2350.

Luckow, B. and Schütz, G. (1 9 8 9 ). Cell-type specificity of regulatory elements identified by linker scanning mutagenesis in the promoter of the chicken lysozyme gene. N u c l e i c A c i d s R e s . 17, 8451-8462. Perlmann, T. and Wrange, Ö. (1 9 8 8 ). Specific glucocorticoid receptor binding to DNA reconstituted in a nucleosome. EMBO J. 7, 3073-3079.

Grez, M., Land, H., Giesecke, K., Schütz, G., Jung, A., and Sippel, A.E. (1 9 8 1 ). Multiple mRNAs are generated from the chicken lysozyme gene. C e l l 25, 743-752. Grunstein, M. (1 9 9 7 ). Histone acetylation in chromatin structure and transcription. Nature 389, 349-352.

Peterson, C.L. and Tamkun, J.W. (1 9 9 5 ). The SWI-SNF complex: a chromatin remodeling machine? TIBS 20, 143-156.

Gualdi, R., Bossard, P., Zheng, M., Hamada, Y., Coleman, J.R., and Zaret, K.S. (1 9 9 6 ). Hepatic specification of the gut endoderm in vitro: cell signaling and transcriptional control. G e n e s D e v 10, 1670-1682.

Pina, B., Brüggemeier, U., and Beato, M. (1 9 9 0 ). Nucleosome positioning modulates accessibility of regulatory proteins to the mouse mamary tumor virus promoter. C e l l 60, 719-731.

Hecht, A., Berkenstam, A., Strömstedt, P.-E., Gustafsson, J.A., and Sippel, A.E. (1 9 8 8 ). A progesterone responsive element maps to the far upstream steroid dependent DNase hypersensitive site of the chicken lysozyme gene. EMBO J. 7, 2063-2073.

Richard -Foy, H. and Hager, G.L. (1 9 8 7 ). Sequence-specific positioning of nucleosomes over the steroid-inducible MMTV promotor. EMBO J. 6, 2321-2328. Roque, M.C., Smith, P.A., and Blasquez, V.C. (1 9 9 6 ). A developmentally modulated chromatin structure at the mouse immunoglobilin 3´ enhancer. M o l . C e l l . B i o l . 16, 3138-3155.

Hogan, B., Beddington, R., Constantini, F., and Lacy, E. (1 9 9 4 ). Manipulating the mouse embryo - A Laboratory Manual (CSHL Press).

Sippel, A.E., Borgmeyer, U., Püschel, A.W., Rupp, R.A.W., Stief, A., Strech-Jurk, U., and Theisen, M. (1 9 8 7 a ). Multiple nonhistone protein-DNA complexes in chromatin regulate the cell- and stage-specific activity of a eukaryotic gene. In Results and Problems in Cell Differentiation 14. Structure and F u n c t i o n o f Eukaryotic C h r o m o s o m e s . W. Hennig, ed. (Heidelberg: Springer Verlag), pp. 255-269.

Huber, M.C., Bosch, F., Sippel, A.E., and Bonifer, C. (1 9 9 4 ). Chromosomal position effects in chicken lysozyme gene transgenic mice are correlated with suppression of DNaseI hypersensitive site formation. N u c l e i c A c i d s R e s . 22, 4195-4201. Huber, M.C., Graf, T., Sippel, A.E., and Bonifer, C. (1 9 9 5 ). Dynamic changes in the chromatin of the chicken lysozyme gene domain during differentiation of multipotent progenitors to macrophages. DNA C e l l . B i o l 14, 397-402.

Sippel, A.E., Borgmeyer, U., Püschel, A.W., Rupp, R.A.W., Stief, A., Strech-Jurk, U., and Theisen, M. (1 9 8 7 b ). Multiple nonhistone protein-DNA complexes in chromatin regulate the cell- and stage-specific activity of an eucaryotic gene. In R e s u l t s and P r o b l e m s i n C e l l D i f f e r e n t i a t i o n . W. Hennig, ed. (Berlin Heidelberg: Springer-Verlag), pp. 255-269.

Huber, M.C., Jägle, U., Krüger, C., and Bonifer, C. (1 9 9 7 ). The developmental activation of the chicken lysozyme locus in transgenic mice requires the interaction of a subset of enhancer elements with the promoter. N u c l e i c A c i d s R e s . 25, 2992-3000.

434

Gene Therapy and Molecular Biology Vol 3, page 435 Sippel, A.E., Saueressig, H., Huber, M.C., Hoefer, H.C., Stief, A., Borgmeyer, U., and Bonifer, C. (1 9 9 6 ). Identification of cis-acting elements as DNaseI hypersensitive sites in lysozyme chromatin. M e t h o d s i n E n z y m o l o g y 274, 233-246. Sippel, A.E., Theisen, M., Borgmeyer, U., Strech-Jurk, U., Rupp, R.A.W., Püschel, A.W., Müller, A., Hecht, A., Stief, A., and Grussenmeyer, T. (1 9 8 8 ). Regulatory function and molecular structure of DNaseI-hypersensitive elements in the chromatin domain of a gene. In The A r c h i t e c t u r e o f Eukaryotic Ge n e s . G. Kahl, ed. (Weinheim: Verlagsgesellschaft Chemie(VHC)), pp. 355369. Slightom, J.L., Bock, J.H., Tagle, D.A., Gumucio, D.L., Goodman, M., Stojanovic, N., Jackson, J., Miller, W., and Hardison, R. (1 9 9 7 ). The complete sequence of the Galago and Rabbit ß-globin locus control regions: extended sequence and functional conservation outside the cores of DNaseI hypersensitive sites. G e n o m i c s 39, 9094. Steiner, C., Müller, M., Baniahmad, A., and Renkawitz, R. (1 9 8 7 ). Lysozyme gene activity in chicken macrophages is controlled by positive and negative regulatory elements. Nuc le ic A c id s R e s . 15, 4163-4178. Stief, A., Winter, D.M., Strätling, W.H., and Sippel, A.E. (1 9 8 9 ). A nuclear attachment element mediates elevated and position independent gene activity. Nature 341, 343-345. Straka, C. and Hörz, W. (1 9 9 1 ). A functional role for nucleosomes in the repression of a yeast promoter. EMBO J 14, 1727-1736. Taylor, I.C.A., Workman, J.L., Schuetz, T.J., and Kingston, R.E. (1 9 9 1 ). Facilitated binding of GAL4 and heat shock factor to nucleosomal templates: differential function of DNA-binding domains. G e n e s D e v . 5, 1285-1298. Theisen, M., Stief, A., and Sippel, A.E. (1 9 8 6 ). The lysozyme enhancer: cell-specific activation of the chicken lysozyme gene by a far-upstream element. EMBO J. 5, 719-724. Truss, M., Bartsch, J., Schelbert, A., Hache, R.J.G., and Beato, M. (1 9 9 5 ). Hormone induces binding of receptors and transcription factors to a rearranged nucleosome on the MMTV promoter in vivo. EMBO J. 14, 1737-1751. Weitzel, J.M., Buhrmester, H., and Strätling, W.H. (1 9 9 7 ). Chicken MAR-binding protein ARBP is homologous to rat methyl-CpG-binding protein MeCP2. M o l . C e l l . B i o l . 17, 5656-5666. Wong, J., Shi, Y.-B., and Wolffe, A.P. (1 9 9 5 ). A role for nucleosome assembly in both silencing and activation of the Xenopus TRßA gene by the thyroid hormone receptor. G e ne s D e v . 9, 2696-2711.

435

Gene Therapy and Molecular Biology Vol 3, page 437 Gene Ther Mol Biol Vol 3, 437-445. August 1999.

Mechanisms involved in regulation of the estrogenresponsive pS2 gene Review Article

Ann M. Nardulli*, Jongsook Kim, Jennifer R. Wood, and Lorene E. Romine Molecular and Integrative Physiology, University of Illinois, 524 Burrill Hall, 407 South Goodwin, Urbana, IL 61801 __________________________________________________________________________________________________ *Correspondence: Ann M. Nardulli, Ph.D. Phone: 217 244-5679; Fax: 217 333-1133; E-mail: anardull@uiuc.edu Abbreviations: ER, estrogen receptor; EREs, estrogen response elements; DMS, dimethylsulfate; LMPCR, ligation mediated polymerase chain reaction Received: 10 October 1998; accepted: 19 October 1998

Summary Estrogen is a hormone of critical importance in the development and maintenance of normal reproductive tissues and has been implicated in initiation of mammary carcinogenesis. Estrogen's actions are mediated through an intracellular estrogen receptor (ER), which interacts with estrogen response elements (EREs) to bring about changes in transcription of estrogen-responsive genes. Although it is clear that the ER-ERE interaction is a critical link in the chain of events that lead to transcription activation, the mechanisms by which transcriptional changes occur remain unclear. We present evidence that ERE sequence and ligand act as allosteric modulators of ER conformation and that these conformational changes most likely play a role in regulating transcription of estrogen-responsive genes. 1976; Wasterberg 1980) and has resulted in a 40% reduction in breast cancer recurrence (Early Breast Cancer Trialists' Collaborative Group 1992) as well as favorable effects on lipid profiles and bone mineral density (Love et al., 1991, 1992). Tamoxifen is also being tested for its ability to prevent breast cancer in women who are at risk for developing this disease (Davidson 1992; Powles 1998). Raloxifene has positive effects on bone mineral density and cardiovascular health without stimulating proliferation in the endometrium and may be useful in breast cancer prevention (Delmas et al., 1997; Gustafsson 1998). Clinical trials have been instituted to test the ability of ICI 182,780 to prevent breast cancer recurrence and treat tamoxifen-resistant tumors (DeFriend et al., 1994; Howell et al., 1995; Wakeling et al., 1991). Although the importance of estrogens and antiestrogens in human health is indisputable, the mechanisms by which these compounds bring about their effects are unclear.

I. Introduction The importance of estrogen action in the development and maintenance of normal reproductive function in the female has been well established. Estrogen has been extensively used in birth control pills to regulate ovulation and avoid pregnancy. More recently, it has become clear that estrogen's effects extend well beyond reproductive tissues. Estrogen plays an important role in maintenance of cardiovascular health (Mendelsohn and Karas 1994; Stevenson et al., 1994; Subbiah 1998), bone mineral density (Smith et al., 1994; Yang et al., 1996), and neural function (Toran-Allerand 1996; Wickelgren 1997). Estrogen's ability to confer these very positive effects has led to the widespread use of estrogen replacement therapy in women with postmenopausal symptoms (Stanford and Colditz 1996). Surprisingly, it is now clear that estrogen plays a crucial role in the male reproductive tract and is required for production of viable sperm (Hess, Bunick, and Bahr 1995; Hess et al., 1997). Studies in ER-deficient mice indicate that estrogen may also influence behavior in both males and females (Ogawa et al., 1996a,b, 1997, 1998).

The actions of estrogen and antiestrogens are mediated by the intracellular estrogen receptor (ER). Upon binding to ligand, the ER interacts with estrogen response elements (EREs) present in target genes and initiates changes in transcription. To better understand how hormone and antihormone regulate gene expression, we have examined

Antiestrogens have also been used in a number of clinical situations. Tamoxifen has been used in breast cancer therapy in women with ER-positive tumors (Tormey et al., 437

Nardulli et al: Regulation of the estrogen-responsive pS2 gene the ER-ERE interaction in in vitro and in vivo assays and considered the ability of different ERE sequences to activate transcription.

The ability of DNA to induce changes in protein conformation has been reported for a number of different transcription factors (Alber, 1993; Frankel and Kim, 1991; Pabo and Sauer, 1992; Spolar and Record, 1994; Steitz, 1990; von Hippel, 1994). More importantly, NMR and crystal structure studies carried out with the ER and glucocorticoid receptor DBDs have demonstrated that a dimerization interface is induced upon binding to DNA (Hard et al., 1990; Luisi et al., 1991; Schwabe et al., 1993a,b). Work with mutant GRs provide further support for this hypothesis (Lefstin et al., 1994; Starr et al., 1996).

II. Different estrogen response elements activate transcription to different extents Although the consensus ERE (Klein-Hitpass et al., 1988) found in the Xenopus laevis vitellogenin A2 gene is a high affinity binding site for the ER and the most thoroughly characterized ERE, the overwhelming majority of ERE sequences identified in endogenous genes differ from the consensus sequence by one or more nucleotides. To understand how different EREs regulate gene expression, we examined the ability of the ER to activate transcription of simple promoters containing either the Xenopus laevis vitellogenin A2 ERE (GGTCAnnnTGACC, Ref. KleinHitpass et al., 1988), which contains a palindromic, consensus ERE sequence, the Xenopus laevis vitellogenin B1 ERE2, which contains a one base pair change in the 5' end of the half palindrome (AGTCAnnnTGACC, Ref. Walker et al., 1984), or the human pS2 ERE (GGTCAnnnTGGCC, Ref. Nunez et al., 1987), which contains a one base pair mismatch in the 3' ERE half site. When Chinese Hamster Ovary cells were transfected with reporter plasmids containing either the A2, B1, or pS2 ERE separated from the TATA sequence by 2.6 helical turns, exposure to 10 nM 17Ă&#x;-estradiol increased transcription 12.7-, 2.4-, and 3.8-fold, respectively (Fig. 1). Increasing the spacing between the ERE and TATA sequence to 3 helical turns decreased the ability of the A2 ERE to activate transcription by 55%, increased the ability of the pS2 ERE to activate transcription by 35%, but had no significant effect on B1 ERE activity. Further increasing the distance between the ERE and TATA sequence to 3.6 helical turns restored the activity of promoters containing the A2 and pS2 EREs to the original levels, but decreased the activity of the promoter containing the relatively weak B1 ERE. Thus, in spite of the fact that the promoters were identical, except for the ERE sequence, these three promoters had very different abilities to activate transcription. Furthermore, their abilities to effectively activate transcription were dependent upon the phasing of the ERE and TATA sequences.

III. Differential interaction of ERspecific antibodies with the A2 and pS2 ERE-bound ER suggest differences in receptor conformation To determine if there were differences in epitope availability when the intact receptor was bound to the A2 or pS2 ERE, antibody supershift experiments were carried out.

From these findings we hypothesized that individual EREs might serve as allosteric modulators of ER conformation and that these DNA-induced changes in ER conformation may in turn influence ER-protein interactions and lead to changes in transcription activation. Thus it seemed possible that the A2 and the pS2 ERE sequences might elicit unique changes in ER conforma-tion, which fostered the interaction of the receptor with different sets of transcription factors and resulted in different levels of transcription activation.

Fig. 1 The A2, B1, and pS2 ERE activate transcription to different extents. CHO cells were transfected with CAT reporter plasmid, human ER expression plasmid, !-galactosidase expression plasmid, and pTZ nonspecific DNA using the calcium phosphate coprecipitation method. Cells were treated with ethanol vehicle (E2, open bars) or 10 nM 17Ă&#x;-estradiol (+E2, solid bars). Each reporter plasmid was included in at least 6 independent experiments. Values are presented as the mean + S.E. (From Nardulli et al, 1996).

438

Gene Therapy and Molecular Biology Vol 3, page 439 Fig. 2. Antibodies to various ER epitopes can detect differences in conformation of the A2 and pS2 ERE-bound receptor. A. Schematic representation of the epitopes for ER-specific antibodies utilized. B. Partially purified, yeastexpressed ER was combined with A2 ERE-containing DNA fragments (odd lanes) or pS2 ERE-containing DNA fragments (even lanes). After a short incubation, antibodies (Ab) were added to the binding reactions as indicated and the complexes were fractionated on a non-denaturing acrylamide gel. The complexed DNA and free probe were visualized by autoradiography. (From Wood et al, 1998).

(compare lane 2 with 4 and 20). The other antibodies tested (H226, ER6, ER1, D547, h151 and H222) supershifted both the A2 and pS2 ERE-containing complexes in a similar manner. The differential interaction of three ER-specific antibodies with A2 and pS2 ERE-bound ER implied that there were differences in ER epitope availability not only in the DBD, but also in the amino and carboxy termini of the receptor.

A series of monoclonal and polyclonal antibodies, which had been made against several different ER regions (Fig. 2A), were utilized with partially purified yeast-expressed ER and 32 P-labeled A2 or pS2 ERE-containing DNA fragments. Partially purified ER and DNA fragments containing either the A2 or pS2 ERE were incubated with or without antibodies made against different ER epitopes. The proteinDNA complexes were fractionated on non-denaturing polyacrylamide gels. The most striking difference in epitope availability was observed with the monoclonal antibody P1A3, which was made against purified Xenopus laevis DBD. P1A3 enhanced the ER-A2 ERE interaction approximately six fold (Fig. 2B, compare lanes 1 and 7) and strongly inhibited the ER-pS2 ERE interaction (compare lanes 2 and 8), but failed to supershift either the A2 or pS2 ERE-ER complex. Two other antibodies also discriminated between the A2 and pS2 ERE-bound ER. ER21, which binds to the amino terminal region, and D75, which binds to the carboxy terminal region, both enhanced the supershifted ERA2 ERE complex formation approximately two-fold (compare lanes 1 with 3 and 19), but did not alter (ER21) or decreased (D75) formation of the ER-pS2 ERE complex

IV. Protease sensitivity assays detect differences in the A2 and pS2 ERE-bound ER To more directly assess possible differences in receptor conformation, protease sensitivity assays of A2 and pS2 ERE-bound ER were carried out. These assays utilize limited proteolysis of a DNA-bound protein to produce a pattern of digestion based upon amino acid accessibility and provides information about native protein conformation (Schreiber et al., 1988; Tan and Richmond, 1990). 32Plabeled DNA fragments containing the A2 or pS2 ERE were combined with partially purified ER.

439

Nardulli et al: Regulation of the estrogen-responsive pS2 gene Fig.3. Distinct protease digestion patterns of A2 and pS2 ERE-bound ER provide evidence for ERE-mediated differences in receptor conformation. • Partially purified, estrogenoccupied ER was combined with A2 or pS2 ERE-containing DNA fragments. After a short incubation, 0, 0.05, 0.5, 1.25, 2.5, 3.75, or 5 ng chymotrypsin was added to the binding reaction. ER-DNA complexes and free DNA were fractionated on a nondenaturing acrylamide gel and the gel was dried and subjected to autoradiography. The undigested ER-DNA complex (C0) and ER-DNA complexes formed with chymotrypsin-proteolyzed receptor (C1 - C5) are indicated. •

Partially purified ER and A2 or pS2 ERE-containing DNA fragments were combined as in panel A except that 0, 0.05, 0.5, 1.25, 2.5, 3.75, or 5 ng trypsin was added to the binding reactions. The undigested ERDNA complex (T0) and ERDNA complexes formed with trypsin-proteolyzed receptor (T1 - T4) are indicated. (From Wood et al, 1998).

difference in digestion patterns observed with these two EREs was not due to differences in ER or DNA concentrations, chymotrypsin concentrations, association of different proteins with the ER, or increased dissociation of the ER from the pS2 ERE as discussed previously (Wood, Greene, and Nardulli 1998). Proteolysis of the A2 and pS2 ERE-bound ER with trypsin also produced very distinct digestion patterns (Fig. 3B). The pS2 ERE-bound receptor appeared to be particularly susceptible to trypsin cleavage as evidenced by the loss of ER-DNA complex at higher trypsin concentration. Therefore, we believe that the different digestion patterns we observed with the A2 and pS2 EREbound ER resulted from differences in receptor conformation and that the conformation was dictated by the ERE sequence.

Increasing amounts of chymotrypsin were added to the reactions and the resulting protein-DNA complexes were fractionated on a non-denaturing, acrylamide gel. The differences in the digestion patterns observed with the A2 ERE-bound ER and the pS2 ERE-bound ER were striking (Fig. 3A). Limited chymotrypsin digestion of the A2 EREbound ER produced a larger stable ER-DNA complex (Fig. 3A, C3) than that observed with the pS2 ERE-bound ER (C5). The number of intermediate ER-DNA complexes observed with A2 and pS2 ERE-bound ER was also quite distinct. While chymotrypsin digestion of the A2 EREbound receptor produced several ER-DNA complexes of intermediate size, digestion of the pS2 ERE-bound ER produced few intermediate sized ER-DNA complexes. The

440

Gene Therapy and Molecular Biology Vol 3, page 441 MCF-7 cells were treated with either control vehicle or 10 nM E2 and then exposed to DNase I. DNase I-treated genomic DNA was used in ligation mediated polymerase chain reaction (LMPCR) footprinting analysis (Mueller and Wold, 1992) to examine the pS2 ERE, which is located from -393 to -405 relative to the transcription start site of the pS2 gene (Nunez et al., 1989). When in vitro-treated DNA was compared to in vivo-treated DNA, nucleotides within and adjacent to the consensus ERE half site appeared to be occupied in the absence of hormone (Fig. 4, Compare V t and -). Exposure of MCF-7 cells to E2 resulted in a more extensive pattern of protection, which included both the imperfect and consensus ERE half sites (Fig. 4, E2). The occupation of the pS2 ERE in the presence and in the absence of hormone suggests that the ERE may be involved in silencing as well as activation of the pS2 gene. It should also be noted that other DNA regions 3' of the ERE were protected by proteins suggesting that the ER-occupied receptor may recruit other proteins to the ERE.

VI. In vivo footprinting suggests that unoccupied, estrogen-occupied and antiestrogen-occupied ER recruit different sets of transcription factors to the pS2 5' flanking region It has become clear from various studies that ligand binding alters receptor conformation (Beekman et al., 1993; Bourguet et al., 1995; Hansen and Gorski, 1986; Renaud et al., 1995; Wagner et al., 1995). Recent crystal structure studies demonstrate that E2 and raloxifene induce distinct differences in the conformation of the ligand binding domain (Brzozowski et al., 1997). To determine whether ligandinduced changes in ER conformation are involved in recruitment of different proteins to the ERE, in vivo footprinting experiments were carried out to examine the endogenous pS2 gene in MCF-7 cells that had or had not been exposed to hormone and subjected to DMS treatment. LMPCR footprinting analysis (Mueller and Wold, 1992) indicates that distinct differences were apparent in the footprinting patterns when MCF-7 cells had or had not been exposed to E2. When MCF-7 cells were maintained in an estrogen-free environment, the footprinting pattern observed was very similar to that of in vitro DMS-treated naked, genomic DNA, except that three adenine residues (Fig. 5, Compare G and -), one of which was located in the consensus ERE half site, displayed an increased sensitivity to DMS methylation. When cells were exposed to E2, one guanine residue in the imperfect ERE half site was protected and the adenine residue in the consensus ERE half site again displayed increased sensitivity to DMS methylation (Fig. 5, E2). Even more striking was that the pattern of protection extended to include sequences flanking both sides of the ERE and multiple regions 3' of the ERE. These findings sup-

Fig. 4. DNase I in vivo footprinting of the pS2 ERE. MCF-7 cells, which had been maintained on serum-free medium for six days, were exposed to either control vehicle (-) or 10 nM E2 (E 2) for 24 hours and then treated with lysolecithin and DNase I. Genomic DNA was isolated and used in in vivo footprinting. Naked genomic DNA samples, which had been treated in vitro with either DNase I (V t) or DMS (G), were included as references. Consensus and imperfect ERE half site locations are indicated.

V. Both the unoccupied and estrogenoccupied receptor may be involved in regulating estrogen-responsive genes Our long term goal is not to understand how the ER interacts with supercoiled plasmids in transfection assays or linear DNA fragments in acrylamide gels, but rather to understand how estrogen-responsive genes are regulated in target cells. To define how the ERE is involved in regulating endogenous genes in living cells, we have used in vivo footprinting to examine the endogenous pS2 ERE in intact MCF-7 breast cancer cells. 441

Nardulli et al: Regulation of the estrogen-responsive pS2 gene sites and adjacent nucleotide sequences were protected. Numerous changes in protein-DNA interactions were also observed at multiple sites 3' of the ERE. Thus, the two antiestrogens tested, one a partial agonist/antagonist and the other a pure antagonist, produced very different footprinting patterns. The divergent footprinting patterns observed with control-, estrogen-, and antiestrogen-treated MCF-7 cells suggest that unoccupied, estrogen-occupied, and antiestrogen-occupied ER associate with different sets of coactivator and/or corepressor proteins and that these proteins in turn form an interconnected protein-DNA complex, which serves to modulate gene expression. The extensive pattern of protection flanking the ERE suggests that this region is intimately associated with a number of proteins. Numerous coactivators and corepressors associated with the steroid receptors have been identified (Horwitz et al., 1996). Recent studies have also identified coactivator and corepressor proteins with histone acetylase and deactylase activities, respectively (Heinzel et al., 1997; Nagy et al., 1997; Pazin and Kadonaga, 1997). Association of ER with these coregulators could be important in modulating chromatin structure and the accessibility of transcription factor binding sites in native chromatin. Estrogen treatment could release corepressor proteins and promote interaction of the receptor with coactivators resulting in changes in local chromatin structure.

VII. The basal promoter is poised for transcription even in the absence of hormone

Fig. 5. DMS in vivo footprinting of the pS2 ERE. MCF-7 cells were exposed to either control vehicle (-), 10 nM E2 (E 2), 100 nM 4-hydroxytamoxifen (T), or 100 nM ICI 182,780 (I) for 2 hours and treated with DMS. Genomic DNA was isolated and used in in vivo footprinting. Naked genomic DNA that had been treated in vitro with DMS (G) was included for reference. Consensus and imperfect ERE half site locations are indicated.

TATA and CAAT boxes are often present in the proximal promoters of inducible genes and are involved in formation of the basal transcription complex. The footprinting patterns in the TATA and CAAT regions of the pS2 gene (Jeltsch et al., 1987) were quite similar when MCF-7 cells had been treated with either control vehicle or E2 and then exposed to DNase I (Fig. 6). Interestingly, DNase I hypersensitive sites (*) were observed flanking the TATA and CAAT sequences in both control vehicle- and E 2treated cells. The presence of hypersensitive sites in these regions suggests that protein-induced conformational changes in DNA structure brought about by binding of transcription factors to these regions may enhance the susceptibility of specific nucleotides to DNase I cleavage (Suck, 1994). The ability of the TATA binding protein to induce DNA bending has been demonstrated (Kim et al., 1993) and is consistent with this hypothesis. Since the TATA and CAAT sequences were flanked by hypersensitive sites before and after hormone treatment, it appears that the basal promoter is accessible and poised for transcription even in the absence of hormone.

port the idea that the E2-occupied, ERE-bound receptor was involved in recruiting numerous other proteins to the pS2 5' flanking region. When MCF-7 cells were treated with 4hydroxytamoxifen, the footprinting pattern observed was strikingly similar to that of in vitro DMS-treated naked DNA, except that a guanine residue in the consensus ERE half site (Fig. 5, T) and two guanines in more distant regions 3' of the ERE were protected. Thus, 4-hydroxytamoxifen treatment of MCF-7 cells resulted in minimal changes in the protection of this region of the pS2 5' flanking region. When MCF-7 cells were treated with ICI 182,780, a very different and distinct footprinting pattern was observed. Guanine residues in the consensus (Fig. 5, I) and imperfect ERE half 442

Gene Therapy and Molecular Biology Vol 3, page 443

VIII. Changes in ER conformation induced by ligand and ERE sequence may play a role in regulating estrogenresponsive genes A number of studies have demonstrated that the activity of many ERE-containing promoters is cell-type specific (Berry et al., 1990; Metzger et al., 1995; Montano et al., 1996; Tora et al., 1989; Tzukerman et al., 1994). It is generally thought that these tissue-specific effects are brought about by restricting the expression of required regulatory cofactors to target cells. A more versatile way of differentially regulating gene expression would be to provide the receptor with a large repertoire of functional surfaces that can be formed and serve as contact points for other cellular proteins. The presentation of these functional surfaces and the selection of ER-associated proteins would provide tremendous regulatory versatility to a single cell harboring multiple estrogen-responsive genes. Thus, it appears that the estrogen receptor may be subject to two ligands: hormone and DNA. Binding of either ligand could induce changes in receptor conformation that could lead to recruitment of different sets of proteins to the 5' flanking region of estrogen-responsive genes in target cells and ultimately result in differential gene expression.

Acknowledgements We are grateful to Geoffrey Greene (The Ben May Institute, Chicago, IL) for partially purified, yeast-expressed human ER. We also thank ICI Pharmaceuticals and Besins Iscovesco for ICI 182,780 and 4-hydroxytamoxifen, respectively. This work was supported by NIH grant HD31299 and DOD grants DAMD17-94-J-4273 and DAMD17-97-1-7201. J.R.W. was supported by NIH Reproductive Training Grant PHS 2732 HD 0728.

References Alber, T. (1993) How GCN4 binds DNA. Curr Biol 3, 182-184. Beekman, Johanna M., George F. Allan, Sophia Y. Tsai, Ming-Jer Tsai, and Bert O'Malley. (1993) Transcriptional activation by the estrogen receptor requires a conformational change in the ligand binding domain. Mol Endocrinol 7, 1266-1274. Berry, Meera, Daniel Metzger, and Pierre Chambon. (1990) Role of the two activating domains of the oestrogen receptor in the celltype and promoter-context dependent agonistic activity of the anti-oestrogen 4-hydroxytamoxifen. EMBO J 9, 2811-2818. Bourguet, W, M Ruff, P Chambon, H Gronemeyer, and D Moras. (1995) Crystal structural of the ligand-binding domain of the human nuclear receptor RXR-alpha. Nature 375, 377-382.

Fig. 6. DNase I in vivo footprinting of the pS2 TATA and CAAT sequences. MCF-7 cells were treated with control vehicle () or 10 nM E2 (E2) and then exposed to DNase I. Genomic DNA was isolated and used in in vivo footprinting. Naked genomic DNA samples, which had been treated in vitro with either DNase I (V t) or DMS (G), were included as references. Regions of DNase I hypersensitivity are indicated (*). TATA and CAAT sequence positions are noted.

Brzozowski, A.M., A.C.W. Pike, Z. Dauter, R.E. Hubbard, T. Bonn, O. Engstrom, L. Ohman, G.L. Greene, J. Gustafsson, and M. Carlquist. (1997) Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389, 753-758. Davidson, NE. (1992) Tamoxifen - panacea or Pandora's box? New

443

Nardulli et al: Regulation of the estrogen-responsive pS2 gene Engl J Med 326, 885-886.

Nucleic Acids Res 16, 647-663.

DeFriend, DJ, A Howell, RI Nicholson, E Anderson, M Dowsett, RE Mansel, RW Blamey, NJ Bundred, JF Robertson, C Saunder, M Baum, P Walton, F Sutcliffe, and AE Wakeling. (1994) Investigation of a new pure antiestrogen (ICI 182,780) in women with primary breast cancer. Cancer Res 54, 408414.

Lefstin, Jeffrey A., Jay R. Thomas, and Keith R. Yamamoto. (1994) Influence of a steroid receptor DNA-binding domain on transcriptional regulatory functions. Genes Dev 8, 2842-2856. Love, R R, R B Mazess, H S Barden, S Epstein, P A Newcomb, C V Jordan, P P Carbone, and D L DeMets. (1992) Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. New Engl J Med 326, 852-856.

Delmas, P.D., N.H. Bjarnason, B.H. Mitlas, A-C. Ravoux, A.S. Shah, W.J. Huster, M. Draper, and C. Christiansen. (1997) Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. New Engl J Med 337, 1641-1647.

Love, R R, D A Wiebe, P A Newcomb, L Cameron, H Leventhal, C V Jordan, J Teyzi, and D L De Mets. (1991) Effects of tamoxifen on cardiovascular risk factors in postmenopausal women. Annals Int Med 115, 860-864.

Early Breast Cancer Trialists' Collaborative Group. (1992) Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy, 133 reandomised trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Lancet 339, 1-5.

Luisi, B.F., W. X. Xu, Z. Otwinowski, L. P. Freedman, K. R. Yamamoto, and P. B. Sigler. (1991) Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature 352, 497-505. Mendelsohn, M.E. and R.H. Karas. (1994) Estrogen and the blood vessel wall. Curr Opin Cardiol 9, 619-626.

Frankel, A. D. and P. S. Kim. (1991) Modular structure of transcription factors, implications for gene regulation. Cell 65, 717-719.

Metzger, Daniel, Simak Ali, Jean-Marc Bornert, and Pierre Chambon. (1995) Characterization of the amino-terminal transcriptional activation function of the human estrogen receptor in animal and yeast cells. J Biol Chem 270, 95359542.

Gustafsson, J-A. (1998) Raloxifene, magic bullet for heart and bone? Nature Med 4, 152-153. Hansen, Jeffrey C. and Jack Gorski. (1986) Conformational transitions of the estrogen receptor monomer. J Biol Chem 261, 13990-13996.

Montano, Monica M, Kirk Ekena, Kristopher D Krueger, Anne L Keller, and Benita S Katzenellenbogen. (1996) Human estrogen receptor ligand activity inversion mutants, receptors that interpret antiestrogens as estrogens and estrogens as antiestrogens and discriminate among different antiestrogens. Mol Endocrinol 10, 230-242.

Hard, Torleif, Edwin Kellenbach, Rolf Boelens, Bonnie A Naler, Karin Dahlman, Leonard P. Freedman, Jan Carlstedt-Duke, Keith R. Yamamoto, Jan-Ake Gustafsson, and Robert Kaptein. (1990) Solution structure of the glucocorticoid receptor DNAbinding domain. Science 249, 157-160.

Mueller, Paul R. and Barbara Wold. (1992) Ligation-mediated PCR for genomic sequencing and footprinting. In Current Protocols in Molecular Biology, ed. F Ausubel, R Brent, R Kingston, D Moore, J Seidman, J Smith, and K Struhl:15.5.115.5.26, John Wiley & Sons, Inc.

Heinzel, T, R M Lavinsky, T M Mullen, M Soderstrom, C D Laherty, J Torchia, W M Yang, G Brard, S D Ngo, R N Eisenman, D W Rose, C K Glass, and M G Rosenfeld. (1997) A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature 386, 43-48.

Nagy, Laszio, Hung-Ying Kao, Debabrata Chakravarti, Richard J Lin, Christian A Hassig, Donald E Ayer, Stuart L Schreiber, and Ronald M Evans. (1997) Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell 89, 373-380.

Hess, R.A., D. Bunick, and J.M. Bahr. (1995) Sperm, a source of estrogen. Env Health Persp 103, 59-62. Hess, R.A., D. Bunick, K-H. Lee, J. Bahr, J.A. Taylor, K.S. Korach, and D.B. Lubahn. (1997) A role for oestrogens in the male reproductive system. Nature 390, 509-512.

Nunez, A.-M., M. Berry, J.-L. Imler, and P. Chambon. (1989) The 5' flanking region of the pS2 gene contains a complex enhancer region responsive to oestrogens, epidermal growth factor, a tumour promoter (TPA), the c-Ha-ras oncoprotein and the cjun protein. EMBO J 8, 823-829.

Horwitz, K.B., T.A. Jackson, D.L. Bain, J.K. Richer, G.S. Takimoto, and L. Tung. (1996) Nuclear receptor coativators and corepressors. Mol Endocrinol 10, 1167-1177. Howell, A., D. DeFriend, J. Robertson, R. Blamey, and P. Walton. (1995) Response to a specific antioestrogen (ICI 182780) in tamoxifen-resistant breast cancer. Lancet 345, 29-30.

Nunez, Anne-Marie, Sonia Jakowlev, Jean-Paul Briand, Mireille Gaire, Andree Krust, Marie-Christine Rio, and Pierre Chambon. (1987) Characterization of the estrogen-induced pS2 protein secreted by the human breast cancer cell line MCF-7. Endocrinology 121, 1759 - 1765.

Jeltsch, J. M., M. Roberts, C. Schatz, J. M. Garnier, A. M. C. Brown, and P. Chambon. (1987) Structure of the human oestrogen-responsive gene pS2. Nucleic Acids Res 15, 1401 1414.

Ogawa, S., J.D. Gordan, J. Taylor, D. Lubahn, K. Korach, and D.W. Pfaff. (1996a) Reproductive functions illustrating direct and indirect effects of genes on behavior. Hormones and Behavior 30, 487-494.

Kim, Youngchang, James H. Geiger, Steven Hahn, and Paul B. Sigler. (1993) Crystal structure of a yeast TBP/TATA-box complex. Nature 365, 512-520.

Ogawa, S., D.B. Lubahn, K.S. Korach, and D.W. Pfaff. (1996b) Aggressive behaviors of transgenic estrogen-receptor knockout male mice. Annals NY Acad Sci 794, 384-385.

Klein-Hitpass, L., G.U. Ryffel, E. Heitlinger, and A.C.B. Cato. (1988)A 13 bp palindrome is a functional estrogen responsive element and interacts specifically with estrogen receptor.

Ogawa, S., D.B. Lubahn, K.S. Korach, and D.W. Pfaff. (1997)

444

Gene Therapy and Molecular Biology Vol 3, page 445 Behavioral effects of estrogen receptor gene disruption in male mice. Proc Natl Acad Sci USA 94, 1476-1481.

PRTF. Cell 62, 367-377. Tora, Laszlo, John White, Christel Brou, Diane Tasset, Nicholas Webster, Elisabeth Scheer, and Pierre Chambon. (1989) The human estrogen receptor has two independent nonacidic transcriptional activation functions. Cell 59, 477-487.

Ogawa, S., S. Inoue, T. Watanabe, H. Hiroi, A. Orimo, T. Hosoi, Y. Ouchi, and M. Muramatsu. (1998) The complete primary structure of human estrogen receptor ! (hER!) and its heterodimerization with ER" in vivo and in vitro. Biochem Biophys Res Commun 243, 122-126.

Toran-Allerand, C.D. (1996) The estrogen/neurotrophin connection during neural development, is co-localization of estrogen receptors with the neurotrophins and their receptors biologically relevant? Developmental Neuroscience 18, 36-48.

Pabo, C. O. and R. T. Sauer. (1992) Transcription factors, structural families and principles of DNA recognition. Annu Rev Biochem 61, 1053-1095.

Tormey, D.C., R.M. Simon, M.E. Lippman, J.M. Bull, and C.D. Myers. (1976) Evaluation of tamoxifen dose in advanced breast cancer, a progress report. Cancer Treat Rep 60, 1451-1459.

Pazin, Michael M and James T Kadonaga. (1997) What's up and down with histone deacetylation and transcription? Cell 89, 325-328.

Tzukerman, Maty T, Abby Esty, Dolores Santiso-Mere, Paul Danielian, Malcolm G Parker, Robert B Stein, J Wesley Pike, and Donald P McDonnell. (1994) Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8, 21-30.

Powles, T.J. ( 1998) Status of antiestrogen breast cancer prevention trials. Oncology 12, 28-31. Renaud, J-P, N Rochel, M Ruff, V Vivat, P Chambon, H Gronemeyer, and D Moras. (1995) Crystal structure of the RAR- gamma ligand-binding domain bound to all-trans retinoic acid. Nature 378, 681-689.

von Hippel, P. H. (1994) Protein-DNA recognition, new perspectives and underlying themes. Science 263, 769-770.

Schreiber, Edgar, Patrick Matthias, M. M端ller, and Walter Schaffner. (1988) Identification of a novel lymphoid specific octamer binding protein (OTF-2B) by proteolytic clipping bandshift assay (PCBA). EMBO J 7, 4221-4229.

Wagner, RL, JW Apriletti, ME McGrath, BL West, JD Baxter, and RJ Fletterick. (1995) A structural role for hormone in the thyroid hormone receptor. Nature 378, 690-697.

Schwabe, J. W. R., L. Chapman, J. T. Finch, and D. Rhodes. (1993a) The crystal structure of the estrogen receptor DNAbinding domain bound to DNA, how receptors discriminate between their response elements. Cell 75, 567-578.

Wakeling, AE, M Dukes, and J Bowler. (1991) A potent specific pure antiestrogen with clinical potential. Cancer Res 51, 38673873. Walker, P., J.-E. Germond, M. Brown-Luedi, F. Givel, and W. Wahli. (1984) Sequence homologies in the region preceding the transcription initiation site of the liver estrogen-responsive vitellogenin and apo-VLDLII genes. Nucleic Acids Res 12, 8611-8626.

Schwabe, J. W. R., L. Chapman, J. T. Finch, D. Rhodes, and D. Neuhaus. (1993b) DNA recognition by the oestrogen receptor, From solution to the crystal. Structure 1, 187-204. Smith, E.P., J. Boyd, G.R. Frank, H. Takahashi, R.M. Cohen, B. Specker, T.C. Williams, D.B. Lubahn, and K.S. Korach. (1994) Estrogen resistance caused by a mutation in the estrogenreceptor gene in a man. New Engl J Med 331, 1056-1061.

Wasterberg, H. (1980) Tamoxifen and fluoxymesterone in advanced breast cancer, a controlled clinical trial. Cancer Treat Rep 64, 117-122.

Spolar, R. S. and M. T. Record. ( 1994) Coupling of local folding to site-specific binding of proteins to DNA. Science 263, 777784.

Wickelgren, I. (1997) Estrogen, a new weapon against alzheimer's? Science 276, 676-678. Wood, Jennifer R., Geoffrey L. Greene, and Ann M. Nardulli. (1998) Estrogen response elements function as allosteric modulators of estrogen receptor conformation. Mol Cell Biol 18, 1927-1934.

Stanford, Janet L and Graham A Colditz. (1996) Controversies in science, hormone replacement therapy. J NIH Res 8, 40-41. Starr, D. Barry, William Matsui, Jay R. Thomas, and Keith R. Yamamoto. (1996) Intracellular receptors use a common mechanism to interpret signaling information at response elements. Genes Dev 10, 1271-1283.

Yang, N.N., H.U. Bryant, S. Hardikar, M. Sato, R.J.S. Galvin, A.L. Glasebrook, and J.D. Termine. (1996) Estrogen and raloxifene stimulate transforming growth factor-!3 gene expression in rat bone, a potential mechanism for estrogen- or raloxifenemediated bone maintenance. Endocrinology 137, 2075-2084.

Steitz, T. A. (1990) Structural studies of protein-nucleic acid interaction, The sources of sequence-specific binding. Quarterly Rev Biophys 23, 205-280. Stevenson, J.C., D. Crook, I.F. Godsland, P. Collins, and M.I. Whitehead. (1994) Hormone replacement therapy and the cardiovascular system. Nonlipid effects. Drugs 47, 35-41. Subbiah, M.T.R. (1998) Mechanisms of cardioprotection by estrogens. Proc Soc Exp Biol Med 217, 23-29. Suck, Dietrich. (1994) DNA recognition by DNase I. J Mol Recognition 7, 65 - 70. Tan, Song and Timothy J. Richmond. (1990) DNA binding-induced conformational change of the yeast transcriptional activator

445

Gene Therapy and Molecular Biology Vol 3, page 447 Gene Ther Mol Biol Vol 3, 447-453. August 1999.

Biological function of the USF family of transcription factors Review Article

Michèle Sawadogo*, Xu Luo + , Mario Sirito, Tao Lu, Preeti M. Ismail, Yibing Qyang, and Marilyn N. Szentirmay Department of Molecular Genetics, University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030. _______________________________________________________________________________________________ *

Correspondence: Michèle Sawadogo, Tel: (713) 794-1281; Fax: (713) 7944295; E-mail: msawadog@mdanderson.org

+

Present address: Howard Hughes Medical Institute, University of Texas Southwestern Medical Institute at Dallas, Texas 77235. Received: 13 October 1998; accepted: 23 October 1998

Summary U S F i s a family o f ubiquitous transcription factors that are structurally related t o the Myc oncoproteins and also share with Myc a common DNA-binding specificity. While the structure and DNA-binding properties of the USF transcription factors are well characterized, their biological function i s o n l y beginning t o emerge. Experiments i n cultured c e l l s suggest that U S F can antagonize the activity of Myc in cellular proliferation and transformation. The phenotype of USFdeficient mice indicates an additional and essential role of USF in embryonic development as well as pleiotropic functions in adult animals.

control.

I. Introduction Although USF was one of the first gene-specific transcription factors to be identified in eukaryotes, its biological function has, until recently, remained quite elusive. For this type of transcription factor, biological function can be revealed by the relationships linking their various target genes. However, many other transcription factors, including all the members of the TFE3 and Myc families, recognize the same DNA-binding sites as USF (Beckmann et al., 1990; Fisher et al., 1991; Blackwood and Eisenman, 1991; Ayer et al., 1993). This redundancy greatly complicates the identification of genes that are regulated by USF. Consequently, and despite the development of dominant negative mutants of USF (Meier et al., 1996; Krylov et al., 1997), few cellular genes can be unambiguously classified as bona fide USF targets. Nevertheless, recent studies, summarized here, are beginning to shed light on the biological function of this important family of transcriptional regulators and its essential role in embryonic development and growth

II. The USF family of transcription factors The USF proteins were first identified through in vitro transcription studies as an activity that stimulated expression of the adenovirus major late promoter (Sawadogo and Roeder, 1985; Carthew et al., 1985, Miyamoto et al., 1985). Purification of USF to homogeneity from HeLa cell nuclear extracts indicated that this transcription factor was composed of two different polypeptides with molecular masses of 43- and 44-kDa (Sawadogo et al., 1988). Cloning of the corresponding genes, respectively called Usf1 and Usf2, revealed that the USF proteins belong to the same basic-helix-loop-helixleucine zipper (bHLH-zip) group of transcriptional regulators as the Myc oncoproteins (Murre et al., 1989; Gregor et al., 1990; Sirito et al., 1992).

447

Gene Therapy and Molecular Biology Vol 3, page 448

Figure 1 General structure of the USF proteins. Locations of the USF-specific region (USR), basic-helix-loop-helix (bHLH) domain and leucine zipper (LZ) are shown. The role and evolutionary conservation of the different USF domains are also indicated.

of the mRNAs (Duret et al., 1993). The genomic structure of both Usf genes is characterized by multiple exons, many of which correspond precisely to discrete functional domains of the transcription factors (Lin et al., 1994, Henrion et al., 1996).

USF cDNA clones have now been isolated from several other species, including mouse, Xenopus, and sea urchin (Kozlowski et al., 1991; Kaulen et al., 1991; Sirito et al., 1994). Amino acid comparisons have revealed a strong evolutionary conservation of the bHLH domain needed for dimerization and DNA binding (Figure 1).

The USF1 and USF2 polypeptides are both ubiquitously expressed and the USF1. USF2 heterodimers represent the major USF species in most tissues and cell types. USF1 homodimers are expressed at lower concentration, while the USF2 homodimers are usually scarce, except in B cell lines (Sirito et al., 1994; Viollet et al., 1996). The existence of differentially spliced USF messages has been reported for both the Usf1 and Usf2 genes, but the contribution of minor isoforms to the biological function of USF remains unclear (Gregor et al., 1990; Sirito et al., 1994; Viollet et al., 1996).

The C-terminal leucine zipper is also conserved in vertebrate USFs, where it plays an important role in dimerization specificity. Also extremely conserved in all USF family members is a small domain, located just upstream of the basic region, that has no homologies in other bHLH transcription factors (Figure 1). This USFspecific region or USR is necessary and sufficient for transcriptional activation by USF of promoters containing both a TATA box and an initiator element. The USR also contains an atypical nuclear localization signal that can function independently of a second nuclear localization signal present in the basic region (Luo and Sawadogo, 1996b).

III. Dimerization and DNA binding properties of USF

Outside of the USR and bHLH-zip domains, the sequences of USF1 and USF2 diverge considerably, suggesting that the two proteins can establish different interactions with other transcription factors and thus regulate different sets of genes. Interestingly, the regions of the Usf1 and Usf2 genes that are highly homologous in human and mouse extend outside of the coding region to also include the 5' and 3' untranslated regions of the mRNAs (Sirito et al., 1994; Henrion et al., 1996). This unusual feature indicates that the expression of the Usf genes is controlled by posttranscriptional mechanisms (e.g., translational regulation or message stability) that are conserved between species and involve untranslated regions

The USF proteins exist in solution and also bind DNA as dimers. Efficient dimerization requires both the bHLH domain and the adjacent leucine zipper (Beckmann and Kadesh, 1991; Sirito et al., 1992). By stabilizing the interaction between subunits, the leucine zipper of USF controls the specificity of dimerization and prevents dimerization with other bHLH proteins. Consequently, the USF proteins are excluded from the class of bHLH transcription factors whose activity can be regulated by formation of DNA binding-deficient dimers with members of the Id family of proteins (Sun et al., 1991).

448

Gene Therapy and Molecular Biology Vol 3, page 449

F i g u r e 2 : USF and Myc have very similar DNA-binding specificities. Shown are the complete consensus sequences determined for the two transcription factors, with the most important residues in capital letters and the common core motif boxed. Also shown is the sequence of the adenovirus major late E box that is known to bind both transcription factors in vitro as well as in vivo (Li et al., 1994).

USF or Myc (Li et al., 1994; Reisman and Rotter, 1993; Roy et al., 1994). Together, these observations suggest that the two families of transcription factors may have both specific and common target genes.

The structure of the dimeric bHLH domain of USF1 in a cocrystal with DNA has been solved. Like that of Max, the DNA-binding partner of Myc, the USF bHLH is characterized by a parallel, left-handed four-helix bundle, with the basic regions contacting the DNA in the major groove (Ferré-D'Amaré et al., 1994). However, the structure of USF may be quite different in solution. Indeed, there are strong indications that major conformational changes are required for a stable interaction of USF with the DNA. For example, the basic region undergoes a random coil to alpha-helix folding transition upon specific DNA recognition (Fisher et al., 1993; Ferré-D'Amaré et al., 1994). The presence of the leucine zipper greatly stabilizes the conformation of USF dimers (Bresnick and Felsenfeld, 1994; Lu and Sawadogo, 1994). Therefore, protein-protein interactions that would either favor or hinder essential conformational changes in the USF proteins may well contribute to the regulation of USF function. The formation of tetrameric USF species have also been implicated in the ability of the transcription factor to simultaneously interact with two DNA-binding sites (Sawadogo, 1988; Sha et al., 1995).

IV. Antagonism between USF and Myc in cellular transformation The important role of the Myc proteins, and in particular the ubiquitous c-Myc, in promoting cellular proliferation and preventing differentiation is well documented. Furthermore, overexpression of c-Myc, whether due to gene amplification or translocation or to increased message stability, is an important parameter in cancer progression (reviewed in Marcu et al., 1992; Koskinen and Alitalo, 1993). The transforming ability of c-Myc is best exemplified by its ability to elicit the complete transformation of primary cells when cotransfected with a second oncoprotein such as activated Ras (Land et al., 1983). The effect of the USF proteins on cellular transformation was also investigated by focus formation assay in primary embryonic fibroblasts and is summarized in Table 1.

USF1 and USF2 display identical dimerization and DNA binding specificities. Like the Myc and TFE3 family members, all USF dimers recognize palindromic E boxes characterized by a central CACGTG or CACATG sequence (Blackwell et al., 1990; Kerkhoff et al., 1991; Halazonetis and Kandil, 1991; Bendall and Molloy, 1994). Outside the core sequence, there are differences in the USF and Myc consensus binding sites (Figure 2). Most notably, T and A residues on each side of the CACGTG core sequence are essential for high USF binding affinity (M. N. Szentirmay, unpublished observation), while Myc prefers G and C residues at these locations. Nevertheless, a number of sequences, including the E boxes present in the adenovirus major late and p53 promoters, can bind either

Cotransfected expression vectors

Cellular transformation

Ras alone

No

Ras + c-Myc

Yes

Ras + USF

No

Ras + c-Myc + USF

No

Table 1: Effect of USF and c-Myc on cellular transformation as monitored by focus formation assay in primary embryo fibroblasts.

449

Sawadogo et al: The USF family of transcription factors Deleted gene

USF1 expression

USF2 expression

Total USF level

Phenotype

Usf1

None

Increased

Unchanged

Mild

Usf2

Decreased

None

Decreased

Growth defect

Usf1 + USF2

None

None

(None)

Embryonic lethal

Table 2. Phenotype of USF-deficient mice.

Cotransfection of either USF1 or USF2 with Ras did not result in the appearance of foci of morphologically transformed cells, demonstrating that the function of USF in transformation was clearly different from that of c-Myc. Instead, cotransfection of USF was found to abolish cellular transformation mediated by c-Myc and activated Ras (Luo and Sawadogo, 1996a). This inhibition of cellular transformation by USF requires not only its DNAbinding domain but also domains involved in transcriptional activation, indicating that the effect is not a simple DNA-binding competition with Myc. Rather, it seems that the activity of USF can antagonize the transforming ability of Myc. The inhibitory activity of USF1 in the focus formation assay was specific to the Myc pathway since USF1 overexpression had no effect on the cellular transformation of embryonic fibroblasts mediated by E1A and Ras. In contrast, USF2 overexpression inhibited focus formation mediated by a variety of oncogenes. However, it is unclear whether this strong antiproliferative effect of USF2 affects in all cases transformation per se, or whether it simply prevents the subsequent proliferation of the transformed cells (Luo and Sawadogo, 1996a).

Direct interactions between USF proteins and other cell cycle regulators of the basic-leucine zipper family have also been reported (Blanar and Rutter, 1992; Pognonec et al., 1997). Such interactions are likely to contribute to the regulation of USF function. Finally, it is interesting to note that many of the suspected targets of USF, including the genes encoding p53, cyclin B1, and transforming growth factor-!2, are themselves involved in proliferation or cell cycle control (Reisman and Rotter, 1993; Cogswell et al., 1995; Scholtz et al., 1996)

VI. Early lessons from the USF knockout mice Mutant mice lacking either USF1 or USF2 have been constructed by individually targeting the Usf1 and Usf2 genes by homologous recombination in embryonic stem cells. These experiments have yielded essential information regarding the role of the USF proteins in both embryos and adult animals (Vallet et al., 1997; Sirito et al., 1998; Vallet et al., 1998). When analyzing the phenotype of the USF-deficient mice, it is important to remember that the major USF species normally present in most tissues and cell types is the USF1 . USF2 heterodimer. Thus, phenotypic traits common to the USF1 and USF2 mutants may be caused by the absence of the heterodimers. Similarly, specific phenotypic traits in the single mutants could result either from the absence of the corresponding homodimer or the resulting increase in the other homodimer. Finally, genes that seem unaffected by either mutation may still be controlled by USF if there is a significant overlap between the functions of USF1 and USF2.

V. Involvement of USF in the control of cellular proliferation Many independent observations are consistent with a role of USF in the control of cellular proliferation. First, the expression levels and the transcriptional activities of the USF proteins are both tightly regulated during the cell cycle (T. Lu and M. Sawadogo, unpublished observation) and the activity of USF is induced in response to mitogens (Zhang et al., 1998; Berger et al., 1998). Second, ectopic expression of USF in general, and USF2 in particular, causes strong growth inhibition in certain transformed cell lines (Luo and Sawadogo, 1996a; Aperlo et al., 1996). Third, a number of cancer cell lines contain USF proteins that are active in DNA binding but completely inactive in transcription activation (Y. Qyang, X. Luo, P.M. Ismail, T. Lu and M. Sawadogo, unpublished observations). This loss of USF function, just like Myc overexpression, may well play an important role in triggering the rapid and uncontrolled proliferation of cancer cells.

Major findings reported so far with the single and double USF1/USF2 mutants are summarized in Table 2. A very interesting result was the nature of the crosstalk between the Usf1 and Usf2 genes. Analysis in embryonic fibroblasts demonstrated the existence in USF1-null cells of a compensatory increase in USF2 expression. In sharp contrast, USF2-null fibroblasts exhibited strongly decreased USF1 expression (Sirito et al., 1998). This asymmetrical cross-regulation indicates that one of the roles of USF1 may be to prevent overexpression of the

450

Gene Therapy and Molecular Biology Vol 3, page 451

F i g u r e 3 : Abundance of the various USF dimers in wild-type cells. Asymmetrical cross-regulation between the two Usf genes explains the relatively low level of USF2 homodimers observed in most cell types

VII. Conclusion

more potent USF2 protein. Note that this feedback mechanism accounts perfectly for the low concentration of USF2 homodimers present in wild-type cells (Figure 3).

Analysis of the biological role of the USF proteins is complicated by the existence of two genes with partially overlapping functions. However, these ubiquitous transcription factors are clearly essential and their involvement in growth control has now been demonstrated both at the cellular and whole organism levels. A more complete understanding of the downstream targets of USF will be necessary to further delineate the importance of the different USF species in various developmental and regulatory pathways. Hopefully, the availability of the different USF-deficient mice will soon allow unambiguous determination of genes that are specific targets of either USF1, USF2, or both. By providing tissues and cell lines with different levels of USF1 and USF2 expression, these animals should also prove useful in defining the role of USF in cellular proliferation and differentiation.

The fact that the USF1-null mice appear normal is perfectly understandable if the increased USF2 expression can, for the most part, compensate for the absence of the heterodimers and USF1 homodimers. These animals were found to be both viable and fertile and display only mild behavioral abnormalities (Sirito et al., 1998). In contrast, the USF2-null mice, where the total USF activity is greatly diminished, display a much stronger phenotype, including an obvious growth defect during embryonic development. At birth, these animals are 20-40% smaller than their wild-type or heterozygous littermates and many of them die in the first few hours. Those that survive subsequently develop in an apparently normal fashion, but remain proportional dwarfs. They also demonstrate other abnormalities, including metabolic defects and male infertility (Vallet et al., 1997; Sirito et al., 1998). The double USF1/USF2 mutants, as well as the mutants containing a single Usf1 allele, are embryonic lethal (Sirito et al., 1998; Vallet et al., 1998). Taken together, these results demonstrate an overlapping and essential role of the USF proteins in embryonic development and pleiotropic functions in adult animals.

The antagonism between the cellular functions of the USF transcription factors and of the c-Myc oncoprotein may lead to a better understanding of cancer progression. In particular, the loss of USF transcriptional activity in several cancer cell lines suggests the existence of a cofactor that regulates both USF1 and USF2. Thus, complete loss of USF function can be brought about by the inactivation of a single gene and this event may play a similar role as the overexpression of c-Myc in triggering uncontrolled cellular proliferation.

One common feature observed in USF-deficient mice of various genotypes is their propensity to spontaneous epileptic seizures (Sirito et al., 1998 and unpublished observations). In mice, overexpression of c-Myc in oligodendrocytes causes severe neurological disturbances (Jensen et al., 1998). It is therefore tempting to link these related observations in whole animals to the antagonism demonstrated by USF and Myc functions in cultured cells.

Acknowledgments Work in our laboratory is supported by Grants G-1195 from the Robert A. Welch foundation, CA79578 from the National Institutes of Health, and DMAD17-96-1-6221 451

Sawadogo et al: The USF family of transcription factors regulation of gene expression. N u c l e i c A c i d s R e s . 21, 2314-2322.

from the Department of the Army. T.L. was supported by a postdoctoral fellowship from the National Cancer Institute Training Grant CA09299.

FerrĂŠ-D'AmarĂŠ, A.R, P. Pognonec, R.G. Roeder & S.K.Burley (1 9 9 4 ) Structure and function of the b/HLH/Z domain of USF. EMBO J. 13, 180-189.

References

Fisher, D.E., C.S. Carr, L.A. Parent & P.A. Sharp (1 9 9 1 ) TFEB has DNA-binding and oligomerization properties of a unique helix-loop-helix/leucine-zipper family. G enes D e v . 5, 2342-2352.

Aperlo, C., K.E. Boulukos & P. Pognonec (1 9 9 6 ) The basic region/helix-loop-helix/leucine repeat transcription factor USF interferes with Ras transformation. Eur. J . B i o c h e m . 241, 249-253.

Fisher, D.E., L.A. Parent & P.A. Sharp (1 9 9 3 ) High affinity DNA-binding Myc analogs: recognition by an helix. C e l l 72, 467-476.

Ayer, D.E, L. Kretzner & R.N Eisenman (1 9 9 3 ) Mad: a heterodimeric partner for Max that antagonizes Myc transcriptional activity. C e l l 72, 211-222.

Gregor, P.D., M. Sawadogo & R.G. Roeder (1 9 9 0 ) The adenovirus major late transcription factor USF is a member of the helix-loop-helix group of regulatory proteins and binds to DNA as a dimer. G e n e s D e v . 4 , 1730-1740.

Beckmann, H. & T. Kadesch (1 9 9 1 ) The leucine zipper of TFE3 dictates helix-loop-helix dimerization specificity. G e n e s D e v . 5, 1057-1066. Beckmann, H., L.K. Su & T. Kadesch (1 9 9 0 ) TFE3, a helixloop-helix protein that activates transcription through the immunoglobulin enhancer muE3 motif. G e n e s D e v . 4, 167-179.

Halazonetis, T. & A.N. Kandil (1 9 9 1 ) Determination of the cMyc DNA-binding site. P r o c . N a t l . A c a d . S c i . U S A 88, 6162-6166.

Bendall, A.S. & P.L. Molloy (1 9 9 4 ) Base preference for DNA binding by the bHLH-Zip protein USF: effects of MgCl2 on specificity and comparison with binding of Myc family members. N u c l e i c A c i d s R e s . 22, 2801-2810.

Henrion, A.A., S. Vaulont, M. Raymondjean & A. Kahn (1 9 9 6 ) Mouse USF1 gene cloning: comparative organization within the c-myc gene family. Mamm. Genome 7, 803-809.

Berger, A., C.M. Cultaro, S. Segal & S. Spiegel (1 9 9 8 ) The potent lipid mitogen sphingosylphosphocholine activates the DNA binding activity of upstream stimulating factor (USF), a basic helix-loop-helix-zipper protein. B i o c h i m . B i o p h y s . A c t a 1390, 225-236.

Jensen, N.A., K.M. Pedersen, J.E. Celis & M.J. West (1 9 9 8 ) Failure of central nervous system myelination in MBP/cmyc transgenic mice: evidence for c-myc toxicity. O n c o g e n e 16, 2123-2129. Kaulen, H.P, P. Pognonec, P.D. Gregor & R.G. Roeder (1 9 9 1 ) The Xenopus B1 factor is closely related to the mammalian activator USF and is implicated in the developmental regulation of TFIIIA gene expression. M o l . C e l l . B i o l . 11, 412-424.

Blackwell, T.K., L. Kretzner, E.M. Blackwood, H. Weintraub & R.N. Eisenman (1 9 9 0 ) Sequence-specific DNA binding by the c-Myc protein. S c i e n c e 250, 1149-1151. Blackwood, E.M. & R.N. Eisenman (1 9 9 1 ) Max: a helixloop-helix zipper protein that forms a sequence-specific DNA-binding complex with Myc. S c i e n c e 251, 12111217.

Kerkhoff, E., K. Bister & K.H. Klempnauer (1 9 9 1 ) Sequencespecific DNA binding by Myc proteins. P r o c . N a t l . Acad. Sci. USA 88, 4323-4327.

Blanar, M.A. & W.J. Rutter (1 9 9 2 ) Interaction cloning: identification of a helix-loop-helix zipper protein that interacts with c-Fos. S c i e n c e 256, 1014-1018.

Koskinen, P.J. & K. Alitalo (1 9 9 3 ) Role of myc amplification and overexpression in cell growth, differentiation and death. S e m i n . C a n c e r B i o l . 4, 312.

Bresnick,E.H. & G. Felsenfeld (1 9 9 4 ) The leucine zipper is necessary for stabilizing a dimer of the helix-loop-helix transcription factor USF but not for maintenance of an elongated conformation. J . B i o l C h e m . 269, 2111021116.

Kozlowski, M.T., L. Gan, J.M. Venuti, M. Sawadogo & W.H. Klein (1 9 9 1 ) Sea urchin USF: a helix-loop-helix protein active in embryonic ectoderm cells. D e v . B i o l . 148, 625-630.

Carthew, R.W, L.A. Chodosh & P.A Sharp (1 9 8 5 ) An RNA polymerase II transcription factor binds to an upstream element in the adenovirus major late promoter. C e l l 43, 439-448.

Krylov, D., K. Kasai, D.R. Echlin, E.J. Taparowsky, H. Arnheiter & C. Vinson (1 9 9 7 ) A general method to design dominant negatives to B-HLHZip proteins that abolish DNA binding. P r o c . N a t l . A c a d . S c i . U S A 94, 12274-12279.

Cogswell, J.P., M.M. Godlevski, M. Bonham, J. Bisi & L. Babiss (1 9 9 5 ) Upstream stimulatory factor regulates expression of the cell cycle-dependent cyclin B1 gene promoter. M o l . C e l l . B i o l . 15, 2782-2790.

Land, H., L.F. Parada & R.A. Weinberg (1 9 8 3 ) Tumorigenic conversion of primary fibroblasts requires at least two cooperating oncogenes. Nature 304, 596-602.

Duret, L., F. Dorkeld & C. Gautier (1 9 9 3 ) Strong conservation of non-coding sequences during vertebrates evolution: potential involvement in post-transcriptional

Li, L.H., C. Nerlov, G. Prendergast, D. MacGregor & E.B. Ziff (1 9 9 4 ) c-Myc represses transcription in vivo by a novel mechanism dependent on the initiator element and Myc

452

Gene Therapy and Molecular Biology Vol 3, page 453 box III. EMBO J. 13, 4070-4079.

USF. J . B i o l . C h e m . 263, 11994-12001.

Lin, Q., X. Luo & M. Sawadogo (1 9 9 4 ) Archaic structure of the gene encoding transcription factor USF. J . B i o l . C h e m . 269, 23894-23903.

Sawadogo, M., M.W. Van Dyke, P.D. Gregor & R.G. Roeder (1 9 8 8 ) Multiple forms of the human gene-specific transcription factor USF. I. Complete purification and identification of USF from HeLa cell nuclei. J . B i o l . C h e m . 263, 11985-11993.

Lu, T. & M. Sawadogo (1 9 9 4 ) Role of the leucine zipper in the kinetics of DNA binding by transcription factor USF. J . B i o l . C h e m . 269, 30694-30700.

Scholtz, B., M. Kingsley-Kallesen & A. Rizzino (1 9 9 6 ) Transcription of the transforming growth factor-b2 gene is dependent on an E-box located between an essential cAMP response element/Activation transcription factor motif and the TATA box of the gene. J . B i o l . C h e m . 271, 32375-32380.

Luo, X. & M. Sawadogo (1 9 9 6 a ) Antiproliferative properties of the USF family of helix-loop-helix transcription factors. P r o c . N a t l . A c a d . S c i . U S A 93, 1308-1313. Luo, X. & M. Sawadogo (1 9 9 6 b ) Functional domains of the transcription factor USF2: atypical nuclear localization signals and context-dependent transcriptional activation domains. M o l . C e l l . B i o l . 16, 1367-1375.

Sha, M., A.R. FerrĂŠ-D'AmarĂŠ, S.K. Burley & D.J. Goss (1 9 9 5 ) Anti-cooperative biphasic equilibrium binding of transcription factor upstream stimulatory factor to its cognate DNA monitored by protein fluorescence changes. J . B i o l . C h e m . 270, 19325-19329.

Marcu, K.B., S.A. Bossone & A.J. Patel (1 9 9 2 ) Myc function and regulation. A n n u . R e v . B i o c h e m . 61, 809-860. Meier, J.L., X. Luo, M. Sawadogo & S.E. Straus (1 9 9 4 ) The cellular transcription factor USF cooperates with varicella-zoster virus immediate-early protein 62 to symmetrically activate a bidirectional viral promoter. M o l . C e l l . B i o l . 14, 6896-6906.

Sirito, M., Q. Lin, J.M. Deng, R.R. Behringer & M. Sawadogo (1 9 9 8 ) Overlapping roles and asymmetrical cross-regulation of the USF proteins in mice. P r o c . N a t l . A c a d . S c i . U S A 95, 3758-3763. Sirito, M., Q. Lin, T. Maity & M. Sawadogo (1 9 9 4 ) Ubiquitous expression of the 43- and 44kDa forms of transcription factor USF in mammalian cells. N u c l e i c A c i d s R e s . 22, 427-433.

Miyamoto, N.G., V. Moncollin, J.M. Egly & P. Chambon (1 9 8 5 ) Specific interaction between a transcription factor and the upstream element of the adenovirus-2 major late promoter. EMBO J. 4, 3563-3570.

Sirito, M., S. Walker, Q. Lin, M.T. Kozlowski, W.H. Klein & M. Sawadogo (1 9 9 2 ) Gene Expr. 2, 231-240.

Murre, C., P.S. McCaw & D. Baltimore (1 9 8 9 ) A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD and myc proteins. C e l l 56, 777-783.

Sun, X. H., N.G. Copeland, N.A. Jenkins & D. Baltimore (1 9 9 1 ) Id proteins Id1 and Id2 selectively inhibit DNA binding by one class of helix-loop-helix proteins. M o l . C e l l . B i o l . 11, 5603-5611.

Pognonec, P., K.E. Boulukos, C. Aperlo, M. Fujimoto, H, Ariga, A. Nomoto & H. Kato (1 9 9 7 ) Cross-family interaction between the bHLHZip USF and bZip Fra1 proteins results in down-regulation of AP1 activity. O n c o g e n e 14, 2091-2098.

Vallet, V.S., A.A. Henrion, D. Bucchini, M. Casado, M. Raymondjean, A. Kahn & S. Vaulont (1 9 9 7 ) Glucosedependent liver gene expression in upstream stimulatory factor 2 -/- mice. J . B i o l . C h e m . 272, 21944-21949.

Reisman, D. and V. Rotter (1 9 9 3 ) The helix-loop-helix containing transcription factor USF binds to and transactivates the promoter of the p53 tumor suppressor gene. N u c l e i c A c i d s R e s . 21, 345-350.

Vallet, V.S., M. Casado, A.A. Henrion, D. Bucchini, M. Raymondjean, A. Kahn & S. Vaulont (1 9 9 8 ) Differential roles of upstream stimulatory factors 1 and 2 in the transcriptional response of liver genes to glucose. J . B i o l . C h e m . 273, 20175-20179.

Roy, B., J. Beamon, E. Balint & D. Reisman (1 9 9 4 ) Transactivation of the human p53 tumor suppressor gene by c-Myc/Max contributes to elevated mutant p53 expression in some tumors. M o l . C e l l . B i o l . 14, 7805-7815.

Viollet, B., A.M. Lefrancois-Martinez, A. Henrion, A. Kahn, M. Raymondjean & A. Martinez. 1 9 9 6 . Immunochemical characterization and transacting properties of upstream stimulatory factor isoforms. J . B i o l . C h e m . 271, 1405-1415.

Sawadogo, M. & R.G. Roeder (1 9 8 5 ) Interaction of a genespecific transcription factor with the adenovirus major late promoter upstream of the TATA box region. C e l l 43, 165-175.

Zhang, Z.C., H. Nechushtan, J. Jacob-Hisch, D. Avni, O. Meyuhas & E. Razin (1 9 9 8 ) Growth-dependent and PKCmediated translational regulation of the upstream stimulating factor-2 (USF2) mRNA in hematopoietic cells. O n c o g e n e 16, 763-769.

Sawadogo, M. (1 9 8 8 ) Multiple forms of the human genespecific transcription factor USF. II. DNA binding properties and transcriptional activity of the purified Hela

453

Gene Therapy and Molecular Biology Vol 3, page 455 Gene Ther Mol Biol Vol 3, 455-464. August 1999.

The role of chromatin in the establishment of enhancer function during early mouse development Review Article

Luca Rastelli and Sadhan Majumder* University of Texas MD Anderson Cancer Center, Departments of Neuro-Oncology and Molecular Genetics, 1515 Holcombe Boulevard, Box 100, Houston, TX 77030, USA __________________________________________________________________________________________________ *Corresponding author: Sadhan Majumder, Ph.D. Phone: (713) 792-8920; Fax: (713) 792-6054; E-mail: majumder@mdanderson.org Key words: transcription, enhancer function, chromatin, mammalian embryonic development A b b r e v i a t i o n s : ZGA, zygotic gene activation Received: 14 October 1998; accepted 25 October 1998

Summary In mammals, enhancers are believed to stimulate transcription from RNA polymerase II promoters primarily by relieving their chromatin-mediated repression. Interactions responsible for enhancer function are developmentally acquired. Factors responsible for this repression are not present in the paternal pronuclei of 1-cell embryos, making them impervious to enhancer function. Although such chromatin-mediated repression is observed in oocytes, maternal pronuclei of 1-cell embryos, and the 2-cell embryos, the enhancer function first appears in 2-cell embryos, a stage in development that corresponds to the beginning of major zygotic gene expression (ZGE). The lack of enhancer function prior to 2-cell stage is also not due to the absence of specific activation proteins, but appears to be due to the absence of an essential coactivator activity. The coactivator activity first appears in 2-cell embryos, and i s required for enhancers driven by different classes o f transcription factors. The absence of the coactivator activity and the corresponding enhancer function appears to be unique to oocytes and 1-cell embryos, suggesting that it provides a safeguard against premature activation of genes prior to ZGE.

facilitate formation of an active transcription complex with enhancers providing additional transcription factors that can function at a greater distance from the mRNA start site (Carey et al. 1990; Schatt et al. 1990). This model is encouraged by the fact that the same transcription factors frequently can function as components of either promoters or enhancers. In the second model, the primary role of an enhancer is distinct from that of a promoter; whereas promoters facilitate formation of an active complex, enhancers stimulate weak promoters by relieving chromatin mediated repression of promoters, a process that might involve histone acetylation (Grunstein 1997; Pazin and Kadonaga 1997; Roth and Allis 1996; Vermaak and Wolffe 1998; Wolffe and Pruss 1996). Support for this model comes from analyses of gene expression and of chromatin structure in mouse preimplantation embryos and transcription in cell free systems.

I. Introduction Transcription by RNA polymerase II is controlled primarily by two elements: promoters and enhancers. Promoters determine where transcription begins, and they function upstream and proximal to the initiation site. Enhancers stimulate weak promoters in a tissue specific manner, and they function distal to the initiation site from either an upstream or downstream position. Enhancers can also function as components of origins of DNA replication (DePamphilis 1997) where their activity depends on binding specific transcription factors that can interact with replication proteins (He et al. 1993). However, whereas the primary function of promoters and replication origins is to facilitate assembly of an active initiation complex, the primary function of enhancers remains controversial. Two models are generally considered. In the first model, enhancers are simply extensions of promoters; enhancers and promoters both 455

Rastelli and Majumder: Chromatin in enhancer function during development

F i g u r e 1 . Schematic representation of events at the beginning of mouse development. Upper Panel: Morphological and cell cycle events are indicated as a function of time after injection of human chorionic gonadotropin (post-hCG), a hormone that stimulates ovulation. The paternal pronucleus is indicated by vertical striations, oocyte and maternal pronucleus are denoted by horizontal striations, and the zygotic nuclei are represented by filled circles. Addition of aphidicolin to 1-cell embryos prior to the appearance of pronuclei insures their morphological arrest as they enter S-phase. Lower Panel: Events in gene expression are divided into two phases: maternal and zygotic gene expression. Periods of DNA replication, transcription and mRNA translation are indicated by bars. The ability of oocytes and embryos to utilize enhancers encoded in plasmid DNA injected into cell nuclei is indicated as the general capacity of enhancers to stimulate promoter activity. The resumption of transcription in embryos is delayed by a time-dependent biological clock mechanism (zygotic clock).

cell embryo containing two diploid "zygotic" nuclei, each with a set of paternal and a set of maternal chromosomes. While translation of maternally inherited mRNAs occurs continuously in mature eggs and 1-cell embryos, 1-cell embryos transcribe endogenous genes at a very low rate (Aoki et al. 1997). They can also transcribe microinjected plamids (Ram and Schultz 1993), with the male pronucleus being 4 to 5 more active than the female (Henery et al. 1995; Majumder 1997; Wiekowski et al. 1993). Major zygotic gene activation (ZGA; !-amanitin sensitive protein synthesis) begins in 2-cell embryos. Prior to S-phase only a small number of proteins are expressed (Christians et al. 1995; Davis et al. 1996); after S-phase both the number and overall rate of protein synthesis

II. Early mouse development Relevant features of early mouse development are diagrammed in Figure 1. Growing oocytes, arrested at diplotene of the first meiotic prophase, transcribe and translate many of their own genes. Transcription stops during meiotic maturation, when unfertilized eggs arrest in metaphase of the second meiotic division. Fertilization triggers completion of meiosis and formation of a 1-cell embryo containing a haploid paternal pronucleus derived from the sperm and a haploid maternal pronucleus derived from the oocyte. Each pronucleus then undergoes DNA replication before entering the first mitosis to produce a 2-

456

Gene Therapy and Molecular Biology Vol 3, page 457 increases significantly (Latham et al. 1991). Concurrent with these events is a rapid degradation of maternally inherited mRNAs, reflecting a switch from maternal to zygotic control of embryonic development (Telford et al. 1990). If 1-cell embryos are arrested in S-phase using inhibitors of DNA replication, they remain morphologically 1-cell embryos, but ZGA still occurs at the same time postfertilization that it normally occurs in developing embryos (i.e. the time when 2-cell embryos appear) (Aoki et al. 1997; Conover et al. 1991; Majumder and DePamphilis 1995; Wiekowski et al. 1991). Therefore, initiation of ZGA is governed by a biological clock ("zygotic clock", F i g . 1) that delays transcription until a specified time after fertilization. Other mammals also exhibit a zygotic clock, although the developmental stage at which transcription begins is species specific. Zygotic gene expression begins in at the 2-cell stage in mice and hamsters, the 4-cell stage in pigs, the 4 to 8-cell stage in humans, and the 8 to 16-cell stage in sheep, rabbits and cows (Braude et al. 1988; Ferrer et al. 1995; Schultz 1993; Seshagiri et al. 1992).

to the same signals that regulate endogenous DNA replication and gene expression. Injected DNA undergoes replication and transcription (i ) only when unique eukaryotic regulatory sequences are present, and (i i ) only in cells that are competent for that function (Majumder and DePamphilis 1994). For example, mouse oocytes are arrested in meiosis and therefore do not replicate their own DNA. Accordingly, oocytes do not replicate plasmid DNA injected into their nucleus, even if the injected DNA contains a viral origin and is provided with the appropriate viral replication proteins (Chalifour et al. 1986; Wirak et al. 1985). In contrast, plasmid DNA does replicate when injected into 1-cell or 2cell embryos that undergo S-phase (Wirak et al. 1985), but only when the plasmid contains a functional viral origin and only when the cognate origin recognition protein is present (Martinez-Salas et al. 1988). On the other hand, oocytes do express some of their own genes, and they also can express genes encoded by plasmids if an appropriate promoter is present (Chalifour et al. 1987; Dooley et al. 1989). For example, the oocyte-specific promoter for zona pellucida protein-3 provides oocytespecific expression when present on injected plasmid DNA (Lira et al. 1990; Millar et al. 1991; Schickler et al. 1992). DNA transcription can be analyzed in 1-cell embryos that are arrested as they enter S-phase by including aphidicolin in the culture medium, because the zygotic clock still triggers zygotic gene activation at the normal time post-fertilization. Thus, genes injected into a pronucleus of S-phase"arrested 1cell embryos remain inactive for as long as 20 hours until the first zygotic genes are expressed (Martinez-Salas et al. 1989; Wiekowski et al. 1991). In contrast, plasmid-encoded genes are expressed immediately when injected into 2-cell embryos that are actively expressing their own cellular genes, regardless of whether these embryos are cultured in the presence or absence of aphidicolin. Thus, we conclude that the response of plasmid DNA injected into mammalian oocytes and embryos is not an artifact of the experimental protocol, but accurately reflects the cell's capacity for carrying out transcription or replication in vivo.

III. Injection of DNA as a method for studying DNA transcription and replication in early mammalian embryos The fact that mammalian embryos are available in limited quantities places a serious roadblock in front of any effort to identify cis-acting sequences and trans-acting factors that are required for DNA transcription or replication at the beginning of mammalian development. One solution to this problem has been to inject plasmid DNA into the nuclei present in oocytes, 1-cell embryos, or 2-cell embryos (F i g . 1) and then identify sequences that are required to either replicate the plasmid or express an encoded reporter gene. These transient assays, like those used following transfection of cultured mammalian cells, reveal the capacity of cells to replicate or express genes, their ability to respond to specific cis-acting sequences, and their ability to utilize specific transacting factors provided by co-injecting expression vectors (Majumder 1997). A variety of promoters and enhancers or transactivators have been examined in mice or rabbits. These include promoters for the early genes of SV40 (Bonnerot et al. 1991; Chalifour et al. 1987; Chalifour et al. 1986; Delouis et al. 1992; Ram and Schultz 1993) and polyomavirus (Melin et al. 1993), herpes simplex virus thymidine kinase gene (Majumder et al. 1993; MartinezSalas et al. 1989; Wiekowski et al. 1991), adenovirus EIIa (Dooley et al. 1989), mouse ZP3 (Lira et al. 1990; Millar et al. 1991), hypoxanthine phosphoribosyl transferase, !-actin and hydroxymethyl glutaryl CoA reductase (Bonnerot et al. 1991; Delouis et al. 1992; Vernet et al. 1992), TATA-box and Sp1-dependent promoters (Majumder et al. 1993). The most active transcription factor identified so far in growing oocytes (Chalifour et al. 1987; Chalifour et al. 1986) to cleavage-stage embryos (Majumder et al. 1993) is Sp1. The responses of DNA injected into embryonic cell nuclei reflect the normal conditions at the beginning of mammalian development, because the injected DNA responds

IV. Enhancers relieve promoter repression: a lesson from 1-cell and 2-cell embryos Microinjection experiments using plasmid DNA containing a reporter gene linked to a promoter and/or an enhancer revealed that promoters injected into the paternal pronucleus of S-phase"arrested 1-cell embryos are highly active and their activity is not increased by linking them to enhancers. In contrast, promoters injected in the zygotic nuclei of 2-cell embryos are strongly repressed and this repression can be relieved by linking them to an embryo specific enhancer. For example, the F101 enhancer can stimulate various RNA pol II promoters from 20 fold to more than 300 fold in 2-cell embryos depending on promoter strength and the amount of DNA injected, but does not stimulate the same promoter in 1-cell embryos (Majumder and DePamphilis, 1995). 457

Rastelli and Majumder: Chromatin in enhancer function during development butyrate/trichostatin also correlates with composition and modification status of histones present in the embryos as discussed below. Thus, taken together, these results indicate that the primary function of enhancers is to relieve promoter repression that is observed in 2-cell embryos, and not in 1cell embryos. Enhancers presumably perform this function by preventing chromatin structure from interfering with assembly of an initiation complex. To accomplish this task, transcription factors that activate enhancers must fulfill two criteria: (i ) they must successfully compete with chromatin structure, and (i i ) they must interact with at least one of the transcription factors that constitute a promoter. Transcription factors that cannot compete with chromatin structure would constitute weak promoters that require an enhancer for full activity, whereas when transcription factors that can compete with chromatin constitute part or all of a promoter, the need for an enhancer to activate that promoter would decrease accordingly. This model does not exclude a secondary role of enhancers in facilitating formation of the initiation complex. Such a role could account for the 2 to 3-fold stimulation of promoters in mouse 2-cell embryos by the F101 PyV enhancer above the level observed in 1-cell embryos (Majumder et al., 1993) as well as low levels of enhancer activity sometimes observed in vitro in the absence of chromatin assembly.

Why are the activities of promoters so much lower in 2cell embryos than in the paternal pronucleus of 1-cell embryos? It is not due to changes in the requirements for transcription factors, because site specific mutations affecting individual transcription factor binding sites in the tk promoter show that 1-cell and 2-cell mouse embryos utilize the same promoter elements (Majumder et al. 1993). It is also not due to a decreased ability to utilize promoters, because arrested 2-cell and 4-cell embryos are as effective as arrested 1-cell embryos in utilizing a variety of promoters, if the promoters are linked to the F101 PyV enhancer (Majumder et al. 1993). Therefore, the transcriptional capacity of 2-cell embryos is equivalent to that of 1-cell embryos. This is more directly seen by comparing the amount of a specific transcription factor, such as Sp1, that limits the activity of the tk promoter in both 1-cell and 2cell embryos. Although the tk promoter is at least 20-fold more active in the paternal pronucleus of arrested 1-cell embryos than in the zygotic nuclei of arrested 2-cell and 4cell embryos, the amount of Sp1 in 2-cell embryos is about 6-fold greater than in 1-cell embryos (Majumder et al. 1993; Worrad et al. 1994). Therefore, the reduction in the capacity of embryos to utilize promoters observed upon formation of a 2-cell embryo must result from the appearance of a repressor rather than the loss of positively acting transcription factors. This repression does not likely result from proteins that bind to specific sequences, because it affects different promoters and replication origins that bear little sequence homology and that interact with different initiation factors. This repression also does not depend on the presence of a zygotic nucleus, because it occurs in 2-cell embryos constructed to contain diploid or haploid maternal or paternally derived nuclei (Wiekowski et al. 1993) and in paternal pronuclei that have been transplanted to 2-cell embryos (Henery et al. 1995). Instead, this repression appears to be mediated by changes in chromatin structure that result, at least in part, from factors that are present in the cytoplasm of 2-cell embryos but absent from the cytoplasm of 1-cell embryos. Treatment of 2-cell embryos with butyrate or trichostatin strongly stimulates promoter activity, relieving repression and reducing the need for enhancers (Majumder et al. 1993; Wiekowski et al. 1993). Butyrate has been found to stimulate promoters in mammalian cells by altering the acetylated state of chromatin structure (Grunstein 1997; Kamakaka and Thomas 1990; Tazi and Bird 1990). Butyrate rapidly blocks histone deacetylase, thus increasing the fraction of acetylated core histones that makes the DNA more accessible to transcription factors and reduces the ability of nucleosomes to interact with histone H1 (Turner 1991). Two lines of evidence confirm that the effect of butyrate on relieving repression is at the level of chromatin structure rather than increasing the levels of transcription factors. Butyrate has opposite effects on maternal and paternal pronuclei in 1-cell embryos (discussed below), and does not change the pattern of endogenous protein synthesis in 1-cell and 2-cell embryos (Wiekowski et al. 1993). The effect of

V. Enhancer function requires a specific coactivator: a lesson from oocytes and 2cell embryos In summary, the studies discribed above revealed that transcription promoters and replication origins introduced into the nuclei of 2-cell mouse embryos undergo repression, regardless of the origin or ploidy of the nucleus, and regardless of whether or not cells are arrested in S-phase or allowed to continue cell division. Furthermore, this repression can be relieved either by linking the promoter or origin-core to an embryo responsive enhancer (e.g. F101 PyV enhancer), providing a general transactivator (e.g. HSV ICP4 protein), or by treating the cells with butyrate. Surprisingly, the same DNA sequences are not repressed when injected into the paternal pronucleus of S-phase arrested 1-cell embryos; their activities are high and they are not stimulated further by enhancers, transactivators or butyrate. Since transplantation of the injected paternal pronucleus to a 2-cell embryo returns the injected promoter to a repressed state that can be relieved by enhancers, the absence of repression is not unique to paternal pronuclei, but to the cytoplasm of 1-cell embryos. The environment of the paternal pronucleus in a mouse 1-cell embryo is analogous to cell free systems in which enhancers are no longer needed to activate promoters or replication origins unless the DNA substrate is first assembled into chromatin. Thus, the question remained: what happens when the same sequences are injected into maternal pronucleus of a 1-cell embryo and its precursor, the maternal nucleus (germinal vesicle) of an oocyte? 458

Gene Therapy and Molecular Biology Vol 3, page 459 the presence of saturating amounts of GAL4-VP16 (Table 1) (Majumder et al. 1993; Majumder et al. 1997). Therefore, the missing factor is a co-activator protein(s) that mediates the activity of GAL4:VP16 protein with the transcription complex that binds to the promoter (F i g . 2). This coactivator protein is specific for enhancer function, because the same series of Gal4 DNA binding sites located proximal to a TATA box (i.e. Gal4-dependent promoter) functions efficiently under all conditions. mRNA isolated from mouse embryonic stem (ES) cells and preinjected into the cytoplasm of mouse oocytes, was able to confers enhancer function. ES cells, like cleavage stage embryos, efficiently utilize the F101 enhancer (Melin et al., 1993), and therefore provided a convenient source of mRNA encoding the putative enhancer coactivator. Whether this activity is brought in by a single molecule or a family of molecules is not known However, our preliminary experiments show that a fraction of the nuclear extract from HeLa cells can restore enhancer activity in oocytes (Rastelli, Zhao and Majumder, unpublished observation). It appears that the coactivator activity mediates protein-protein interaction between factors bound at the enhancer site and the transcription complex bound at the promoter site. Since the chromatin structure and enhancer function are intimately connected, the coactivator activity might act by remodeling chromatin structure (Felsenfeld 1996; Kingston et al. 1996; Pazin and Kadonaga 1997; Struhl 1998).

Since oocytes do not replicate DNA, only promoters were injected into the maternal nucleus of an oocyte, parthenogenetically activated egg, or fertilized egg (1-cell embryo). Under these conditions, promoters are repressed from 80 to 96% relative to the paternal pronucleus of S-phase arrested 1-cell embryos (Table 1), (Majumder et al. 1993; Majumder et al. 1997; Martinez-Salas et al. 1989; Wiekowski et al. 1993). As with 2-cell embryos, repression can be relieved by treating cells with butyrate, raising the level of promoter activity in maternal nuclei to that observed in paternal pronuclei. This suggests that the mechanism of repression in maternal nuclei of oocytes and 1-cell embryos is the same as in 2-cell embryos. In fact, the composition of nascent histones in mouse oocytes are indistinguishable from those in mouse fibroblasts, and the slight increase of promoter activity observed in maternal pronuclei (Table 1) corresponds to increased hyperacetylation of histone H4 (Adenot et al. 1997; Wiekowski et al. 1997). Since the cytoplasm of 1-cell embryos does not contain repressor activity, maternal pronuclei must retain some of the repressor that is produced in oocytes. This may be histone H1 that is associated with cellular chromatin, but can transfer easily to chromatin assembled onto plasmid DNA. However, in contrast to 2-cell embryos, linking the promoter to the F101 enhancer could not relieve repression of promoter activity in the maternal nuclei of oocytes and 1-cell embryos. Therefore, some factor is missing in oocytes and 1-cell embryos that is required for enhancer function. This missing factor is not the transcription factor that must bind to the enhancer sequence, because oocytes and 1cell embryos cannot utilize enhancers even when the appropriate enhancer activation protein is provided. In the presence of sufficient GAL4-VP16 protein to drive a GAL4dependent promoter at its maximum rate, a GAL4-dependent enhancer located 600 bp upstream of the tk promoter stimulates promoter activity only 1.5-fold in the maternal nucleus of oocytes, 2-fold in the maternal pronucleus and none in the paternal pronucleus of S-phase arrested 1-cell embryos. In contrast, the same enhancer stimulates promoter activity at least 30-fold when injected into 2-cell embryos in

VI. The appearance of promoter repression correlates with changes in chromatin structure Repression of promoter activity correlates with a decrease in histone H4 hyperacetylation and the concurrent appearance of histone H1 and H2A and H2B. Metaphase II oocyte chromatin and sperm chromatin do not contain hyperacetylated H4 histones, as revealed by antibody staining

Table 1. Properties of mouse oocytes and preimplanation embryos that affect gene expression. Stage Nuclear origin Promoter Chromatin Histone H1,H2A Enhancer repression hyperacetilation and H2B synthesis stimulation Oocyte

Maternal

+++

-

++

-

Putative enhancerspecific coactivator -

1-cell embryo

Maternal

+

+ early

-

-

-

1-cell embryo

Paternal

-

++ early

-

-

-

2-cell embryo

Zygotic

++

+/-

+

+

+

4-cell embryo

Zygotic

+++

-

++

++

++

This table summarizes all the data on oocyte and preimplantation embryo properties that are relevant to explain the difference in transcription activity among the various stages.

459

Gene Therapy and Molecular Biology Vol 3, page 460

VII. A Role for DNA replication in relieving promoter repression

(Adenot et al. 1997), or by labeling of nascent H4 (Wiekowski et al. 1997). Adenot et al. showed that upon sperm entry and throughout most of G1, the paternal chromatin in 1-cell embryos has higher level of hyperacetylated H4. Since transcriptionally active eukaryotic genes are associated with acetylated core histones (Turner 1991), the authors propose that hyperacetylated paternal chromatin can preferentially recruit transcription factors at this stage to explain the higher promoter activity. But, the levels of hyperacetylated H4 are similar in both pronuclei during S/G2 (Adenot et al. 1997) and remain high in 2 cell embryos (Wiekowski et al. 1997) and therefore cannot explain the promoter repression that occurs with the formation of the 2 cell embryos. However, the beginning of promoter repression also correlates with the beginning of histones H1, H2A and H2B synthesis in late 1-cell embryos (Wiekowski et al. 1997). The synthesis of these histones is independent of DNA replication, DNA transcription and cell cleavage indicating that they are translated from maternally inherited mRNAs. Antibodies directed against somatic H1 do not detect any H1 until the late 4-cell stage (Clarke et al. 1992), suggesting that this early form of histone H1 may represent a novel histone H1 subtype as seen in other organisms (Ohsumi and Katagiri 1991; Smith et al. 1988). Thus, it appears that promoter repression correlates with the presence of histone H1 and the absence of acetylation (other modification?) of core histones (Table 1). Studies on promoter activity in vitro further support the hypothesis that the primary role of enhancers is to prevent repression by chromatin structure. Enhancers have little, if any, effect on DNA replication (Prives et al. 1987) or transcription discussed in (Majumder et al. 1993) when they are assayed in cell free systems that do not assemble chromatin, but enhancers can stimulate promoters in cell free systems when the DNA is packaged into chromatin (Paranjape et al. 1994; Sheridan et al. 1995). Nucleosome assembly can interfere with the activity of some, but not all, proteins that are required for initiation of transcription or DNA replication (Workman et al. 1991). Although it appears that histone H1 and acetylation of core histone are involved in chromatin-mediated repression their relative contribution in the process is yet unclear . In particular, the role of H1 on general transcription activity is unknown. In-vitro binding of histone H1 to chromatin in stoichiometric conditions does not have repressive effects (Howe et al. 1998; Sandaltzopoulos et al. 1994). At the same time, in vivo overproduction of mouse histone variant H1(0) results in repression of transcript levels of all polymerase II genes tested. (Brown et al. 1997). It is therefore possible that the maternal H1 variant is responsible for promoter repression in 2 cell embryos. In later stages, repression could be maintained because the level of hyperacetylated H4 drops starting with the 4-cell embryos (Wiekowski et al. 1997).

Two types of repression have been observed in 2-cell embryos: the "reversible" repression described above that can be relieved by enhancers when DNA is injected into 2-cell embryos, and an "irreversible" repression that occurs when DNA is injected into either pronucleus of a normally developing 1-cell embryo, and the injected embryo then undergoes mitosis to form a 2-cell embryo. Under these conditions, the injected promoter or replication origin is "irreversibly" repressed in that its activity cannot be restored either with butyrate or enhancers (Wiekowski et al., 1993; Henery et al., 1995). This phenomenon is unique to the first cell cycle since plasmid DNA injected into 2-cell embryos undergo reversible repression whether or not the injected 2cell embryos is arrested in S-phase or continues cell division to produce 4-cell and then 8-cell embryos. What happens to DNA between completion of S-phase in a 1-cell embryo and formation of a 2-cell embryo that results in "irreversible" repression of injected plasmid DNA, but not of endogenous cellular DNA? One possibility is that the injected DNA is lost to the cytoplasm when the pronucleus, but not a zygotic nucleus, undergoes mitosis. This does not appear to be the case since transplantation of the injected pronucleus to a 2cell embryo that then undergoes mitosis exhibits reversible repression of the plasmid encoded reporter gene. Instead, we suggest that DNA in early 1-cell embryos is subjected to repression before enhancer activation factors become available when zygotic genes are expressed. Once repression has been established, DNA replication may be required in order to displace repressor (histones?) so that transcription factors can bind to promoter and enhancer sequences (Wolffe 1991) (F i g . 2 ). Plasmid expression vectors, such as the ones used in these experiments, do not replicate when injected into mouse embryos (DePamphilis 1997), presumably because they lack a suitable replication origin. However, the 1-cell embryo's genome undergoes at least one round of replication prior to any zygotic gene transcription, an event that may be required to restore the zygotic genome to a transcriptionally competent state. DNA in 2-cell embryos competes for binding to both repressor and enhancer specific proteins (F i g . 2 ); the result of this competition determines the fraction of molecules that are transcriptionally active. Thus, enhancer stimulated promoter activity is greater when plasmid DNA is injected into late 2-cell embryos that progress to the 4-cell stage than in early 2-cell embryos that are arrested in S-phase (Henery et al. 1995) (Table 1). In a recent paper Forlani et al. (Forlani et al. 1998) found that when plasmid DNA containing the intronic sequences of the human HPRT gene as an enhancer element attached to a “Pytk” promoter driving a reporter gene is microinjected in 1-cell embryos arrested by aphidicolin at a time that corresponds to post-DNA replication, the HPRT sequence can stimulate the “Pytk” promoter 4-5 fold as compared to the microinjected construct without the HPRT sequence. This observation led them to propose that acquisition of enhancer function requires the first round of DNA replication. However, the “Pytk” promoter activity seen in 460

Gene Therapy and Molecular Biology Vol 3, page 461

F i g u r e 2 . A working model showing the repression of promoters and replication origins by chromatin structure and the role of enhancers. Core histones and transcription/replication proteins (including enhancer activation proteins) compete to bind to the plasmid DNA microinjected into mouse oocytes and early embryos. Depending on their relative affinities for DNA, there is a dynamic equilibrium between DNA bound to core histones and DNA bound to various transcription/replication factors. In the presence of histone H1, DNA bound to core histones can then interact with them, resulting in a condensed chromatin structure and a repressed state. The repressive action of histone H1 can be blocked by acetylation of core histones. Sodium butyrate or trichostatin are known to inhibit histone deacetylases, and thus increase the fraction of acetylated core histones, causing stimulated transcription. On the other hand, the equilibrium can be shifted to the other direction where DNA bound to transcription/replication factors and enhancer activation proteins can interact with the enhancer specific coactivator resulting in the prevention of repression and the formation of an active state. DNA replication at each cell division may provide the cell with a chance to re-establish the equilibrium between repressed and unrepressed states.

is necessary for the acquisition of the ability to utilize enhancers in mammals, suggesting that one or more enhancer activation proteins, like the enhancer specific coactivator, is produced at this time.

1-cell embryos arrested after DNA replication is similar in magnitude, when compared to promoter activity seen in 1cell embryos arrested before DNA replication. Thus, the low level of the HPRT sequence dependent stimulation appears not to be due to the release of promoter repression (primary role of enhancer function), but perhaps due to other secondary roles of enhancers as discussed above. This observation could also be explained by the effect of the additional transcription factor binding sites provided by the HPRT sequence on the promoter strength. Furthermore, it is also interesting to note that the experiments described in this paper use “Pytk� promoter that contains the tk promoter and the polyoma F101 enhancer (Py). Thus the Pytk promoter possibly represents not the promoter activity alone but the combined effect of promoter and enhancer activities. Thus, taken together, these results indicate that zygotic gene expression in 2-cell embryos and not the first round of DNA replication

VIII. The lack of enhancer function is unique to oocytes and fertilized eggs The lack of enhancer function in oocytes raises the question whether the absence of enhancer function is a unique property of oocytes or a general property of other terminally differentiated cells. To explore this question, transcription activity was examined in terminally differentiated hNT neurons that ceased cell-division like oocytes, and their precursor, undifferentiated NT2 stem cells. The results showed that both NT2 and hNT cells could utilize Gal4VP16- and Sp1-dependent enhancers as well as 461

Rastelli and Majumder: Chromatin in enhancer function during development promoters (Lawinger et al. 1998). Thus, the lack of enhancer specific coactivator activity, and the corresponding lack of enhancer function, appears to be unique to oocytes and fertilized eggs, suggesting that it provides a safeguard against premature activation of genes prior to ZGE. How does chromatin mediated repression and enhancer utilization help to regulate gene expression at the beginning of mammalian development? The onset of transcription during mouse development is regulated by a time dependent mechanism (zygotic clock), and takes place about 40 hours postfertilization, a time when a normally developing embryo is at the 2-cell stage. This stage of development also coincides with the onset of major chromatin repression of promoters (Majumder and DePamphilis, 1995). The paternal genome in sperm comes with protamines, whereas the maternal genome in eggs comes with a normal complement of core histones (Zirkin et al., 1989; Nonchev and Tsanev, 1990). After fertilization, they undergo chromatin remodeling to establish the zygotic genome at the 2-cell stage. This process of remodeling probably generates DNA that is not complexed with either histones or protamines (Rodman et al., 1981), and exposes promoters to transcription factors. Thus the zygotic clock may provide a mechanism to ensure no spurious transcription during the remodeling period. On the other hand, after the zygotic remodeling, the chromatin mediated repression of most promoters in 2-cell embryos may provide a mechanism for enhancer-mediated tissue specific transcription of genes during development and growth. Delaying expression of the enhancer specific coactivator prior to zygotic gene expression provides an additional mechanism for preventing inappropriate transcription of genes destined for expression in specific cell types. The same mechanisms of transcriptional control that initiates mouse development also seem to occur in other animals. In mammals other than mice, transcription is delayed until the 2-cell or 16-cell stage, presumably by the same zygotic clock mechanism. Thus, the zygotic gene expression begins at the 2-cell stage in hamsters, the 4-cell stage in pigs, the 4- to 8-cell stage in humans, and the 8- to 16-cell stage in sheep, rabbits and cows (Telford et al., 1990; Seshagiri et al., 1992; Schulz, 1993). Whether or not repression of promoter activities appears at the 2-cell stage in these mammals, or is delayed until the same stage that transcription begins, remains to be seen. The S-phase of a 2cell mouse embryo appears equivalent to the 6th cleavage stage in Xenopus where synthesis of heterogeneous, nonribosomal mRNA is first detected. The G2-phase of a 2-cell mouse embryo appears equivalent to the 12th cleavage stage in Xenopus where the major onset of RNA polymerase II and III transcription occurs (the midblastula transition, MBT, Kimelman et al., 1987; Shiokawa et al., 1989). The activity of promoter/enhancer sequences injected into Xenopus eggs is generally delayed until the MBT although they appear to exhibit a low but constant rate of gene expression per cell prior to the MBT (Shiokawa et al., 1990). Activation of transcription at the MBT can require specific enhancers (Krieg and Melton, 1987), analogous to the need for an

enhancer to activate promoters in 2-cell mouse embryos. The MBT also marks the appearance of histone H1 mediated repression of oocyte specific genes such as 5S RNA (Wolffe, 1989; Ohsumi and Katagiri, 1991), analogous to the repression observed upon formation of 2-cell mouse embryos. Furthermore, analogous stage-specific acquisition of specific transcriptional coactivators for enhancer function may also occur at the MBT (Xu et al., 1994).

Acknowledgements This work was supported in part by grants to S. M. from the Pediatric Brain Tumor Foundation of the US, Association for Research of Childhood Cancer, and the National Institutes of Health (GM53454). L.R. is supported by the American Brain Tumor Association 25th Anniversary Translational Grant.

References Adenot, P. G., Mercier, Y., Renard, J. P., and Thompson, E. M. (1 9 9 7 ). “Differential H4 acetylation of paternal and maternal chromatin precedes DNA replication and differential transcriptional activity in pronuclei of 1-cell mouse embryos.” D e v e l o p m e n t 124, 4615-4625. Aoki, F., Worrad, D. M., and Schultz, R. M. (1 9 9 7 ). “Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryo.” D e v B i o l 181, 296-307. Bonnerot, C., Vernet, M., Grimber, G., Briand, P., and Nicolas, J. F. (1 9 9 1 ). “Transcriptional selectivity in early mouse embryos: a qualitative study.” N u c l e i c Acids R e s 19, 7251-7257. Braude, P., Bolton, V., and Moore, S. (1 9 8 8 ). “Human gene expression first occurs between the four- and eight-cell stages of preimplantation development.” Nature 332, 459461. Brown, D. T., Gunjan, A., Alexander, B. T., and Sittman, D. B. (1 9 9 7 ). “Differential effect of H1 variant overproduction on gene expression is due to differences in the central globular domain.” N u c l e i c A c i d s R e s 25, 5003-5009. Carey, M., Leatherwood, J., and Ptashne, M. (1 9 9 0 ). “A potent GAL4 derivative activates transcription at a distance in vitro.” S c i e n c e 247, 710-712. Chalifour, L. E., Wirak, D. O., Hansen, U., Wassarman, P. M., and DePamphilis, M. L. (1 9 8 7 ). “cis- and trans-acting sequences required for expression of simian virus 40 genes in mouse oocytes.” G e n e s D e v 1, 1096-1106. Chalifour, L. E., Wirak, D. O., Wassarman, P. M., and DePamphilis, M. L. (1 9 8 6 ). “Expression of simian virus 40 early and late genes in mouse oocytes and embryos.” J V i r o l 59, 619-627. Christians, E., Campion, E., Thompson, E. M., and Renard, J. P. (1 9 9 5 ). “Expression of the HSP 70.1 gene, a landmark of early zygotic activity in the mouse embryo, is restricted to the first burst of transcription.” D e v e l o p m e n t 121, 113122. Clarke, H. J., Oblin, C., and Bustin, M. (1 9 9 2 ). “Developmental regulation of chromatin composition

462

Gene Therapy and Molecular Biology Vol 3, page 463 during mouse embryogenesis: somatic histone H1 is first detectable at the 4-cell stage.” D e v e l o p m e n t 115, 791799.

Lira, S. A., Kinloch, R. A., Mortillo, S., and Wassarman, P. M. (1 9 9 0 ). “An upstream region of the mouse ZP3 gene directs expression of firefly luciferase specifically to growing oocytes in transgenic mice.” P r o c N a t l A c a d S c i U S A 87, 7215-7219.

Conover, J. C., Temeles, G. L., Zimmermann, J. W., Burke, B., and Schultz, R. M. (1 9 9 1 ). “Stage-specific expression of a family of proteins that are major products of zygotic gene activation in the mouse embryo.” D e v B i o l 144, 392-404.

Majumder, S. (1 9 9 7 ). “Mouse pre-implantation embryos as an in vivo system to study gene expression.” Microinjection and Transgenesis: Strategies and Protocols, A. Cid-Arregui and A. Garcia-Carranca, eds., Springer-Verlag, Heidelberg, 323-349.

Davis, W., Jr., De Sousa, P. A., and Schultz, R. M. (1 9 9 6 ). “Transient expression of translation initiation factor eIF-4C during the 2-cell stage of the preimplantation mouse embryo: identification by mRNA differential display and the role of DNA replication in zygotic gene activation.” D e v B i o l 174, 190-201.

Majumder, S., and DePamphilis, M. L. (1 9 9 4 ). “Requirements for DNA transcription and replication at the beginning of mouse development.” J C e l l B i o c h e m 55, 59-68.

Delouis, C., Bonnerot, C., Vernet, M., and Nicolas, J. F. (1 9 9 2 ). “Expression of microinjected DNA and RNA in early rabbit embryos: changes in permissiveness for expression and transcriptional selectivity.” E x p C e l l R e s 201, 284-291.

Majumder, S., and DePamphilis, M. L. (1 9 9 5 ). “A unique role for enhancers is revealed during early mouse development.” B i o e s s a y s 17, 879-889. Majumder, S., Miranda, M., and DePamphilis, M. L. (1 9 9 3 ). “Analysis of gene expression in mouse preimplantation embryos demonstrates that the primary role of enhancers is to relieve repression of promoters [published erratum appears in EMBO J 1993 Oct;12(10):4042].” E m b o J 12, 1131-1140.

DePamphilis, M. L. (1 9 9 7 ). “The search for origins of DNA replication.” M e t h o d s 13, 211-219. Dooley, T. P., Miranda, M., Jones, N. C., and DePamphilis, M. L. (1 9 8 9 ). “Transactivation of the adenovirus EIIa promoter in the absence of adenovirus E1A protein is restricted to mouse oocytes and preimplantation embryos.” D e v e l o p m e n t 107, 945-956.

Majumder, S., Zhao, Z., Kaneko, K., and DePamphilis, M. L. (1 9 9 7 ). “Developmental acquisition of enhancer function requires a unique coactivator activity.” E m b o J 16, 17211731.

Felsenfeld, G. (1 9 9 6 ). “Chromatin unfolds.” C e l l 86, 13-19. Ferrer, F., Garcia, C., Villar, J., and Arias, M. (1 9 9 5 ). “Ultrastructural study of the early development of the sheep embryo.” A n a t H i s t o l E m b r y o l 24, 191-196.

Martinez-Salas, E., Cupo, D. Y., and DePamphilis, M. L. (1 9 8 8 ). “The need for enhancers is acquired upon formation of a diploid nucleus during early mouse development.” G e n e s D e v 2, 1115-1126.

Forlani, S., Bonnerot, C., Capgras, S., and Nicolas, J. F. (1 9 9 8 ). “Relief of a repressed gene expression state in the mouse 1-cell embryo requires DNA replication [In Process Citation].” D e v e l o p m e n t 125, 3153-3166.

Martinez-Salas, E., Linney, E., Hassell, J., and DePamphilis, M. L. (1 9 8 9 ). “The need for enhancers in gene expression first appears during mouse development with formation of the zygotic nucleus.” Genes and D e v e l o p m e n t 3, 14931506.

Grunstein, M. (1 9 9 7 ). “Histone acetylation in chromatin structure and transcription.” Nature 389, 349-352.

Melin, F., Miranada, M., Montreau, N., DePamphilis, M. L., and Blangy, D. (1 9 9 3 ). “Transcription enhancer factor-1 (TEF1) DNA binding sites can specifically enhance gene expression at the beginning of mouse development.” EMBO J. 12, 4657-4666.

He, Z., Brinton, B. T., Greenblatt, J., Hassell, J. A., and Ingles, C. J. (1 9 9 3 ). “The transactivator proteins VP16 and GAL4 bind replication factor A.” C e l l 73, 1223-1232. Henery, C. C., Miranda, M., Wiekowski, M., Wilmut, I., and DePamphilis, M. L. (1 9 9 5 ). “Repression of gene expression at the beginning of mouse development.” D e v . B i o l . 169, 448-460.

Millar, S. E., Lader, E., Liang, L. F., and Dean, J. (1 9 9 1 ). “Oocyte-specific factors bind a conserved upstream sequence required for mouse zona pellucida promoter activity.” M o l C e l l B i o l 11, 6197-6204.

Howe, L., Iskandar, M., and Ausio, J. (1 9 9 8 ). “Folding of chromatin in the presence of heterogeneous histone H1 binding to nucleosomes.” J B i o l Chem 273, 1162511629.

Ohsumi, K., and Katagiri, C. (1 9 9 1 ). “Occurrence of H1 subtypes specific to pronuclei and cleavage-stage cell nuclei of anuran amphibians.” D e v B i o l 147, 110-120.

Kamakaka, R. T., and Thomas, J. O. (1 9 9 0 ). “Chromatin structure of transcriptionally competent and repressed genes.” Embo J 9, 3997-4006.

Paranjape, S. M., Kamakaka, R. T., and Kadonaga, J. T. (1 9 9 4 ). “Role of chromatin structure in the regulation of transcription by RNA polymerase II.” Annu R e v B i o c h e m 63, 265-297.

Kingston, R. E., Bunker, C. A., and Imbalzano, A. N. (1 9 9 6 ). “Repression and activation by multiprotein complexes that alter chromatin structure.” G e n e s D e v 10, 905-920.

Pazin, M. J., and Kadonaga, J. T. (1 9 9 7 ). “What's up and down with histone deacetylation and transcription?” C e l l 89, 325-328.

Latham, K. E., Garrels, J. I., Chang, C., and Solter, D. (1 9 9 1 ). “Quantitative analysis of protein synthesis in mouse embryos. I. Extensive reprogramming at the one- and twocell stages.” D e v e l o p m e n t 112, 921-932.

Prives, C., Murakami, Y., Kern, F. G., Folk, W., Basilico, C., and Hurwitz, J. (1 9 8 7 ). “DNA sequence requirements for replication of polyomavirus DNA in vivo and in vitro.” M o l C e l l B i o l 7, 3694-3704.

Lawinger, P., Rastelli, L., Zhao, Z., and Majumder, S. (1 9 9 8 ). “Lack of a coactivator activity required for enhancer function is unique to oocytes and fertilized eggs.” submitted.

463

Rastelli and Majumder: Chromatin in enhancer function during development Ram, P. T., and Schultz, R. M. (1 9 9 3 ). “Reporter gene expression in G2 of the 1-cell mouse embryo.” D e v B i o l ,156, 552-556.

Vermaak, D., and Wolffe, A. P. (1 9 9 8 ). “Chromatin and chromosomal controls in development.” D e v G e n e t 22, 16.

Roth, S. Y., and Allis, C. D. (1 9 9 6 ). “Histone acetylation and chromatin assembly: a single escort, multiple dances?” C e l l 87, 5-8.

Vernet, M., Bonnerot, C., Briand, P., and Nicolas, J. F. (1 9 9 2 ). “Changes in permissiveness for the expression of microinjected DNA during the first cleavages of mouse embryos.” M e c h D e v 36, 129-139.

Sandaltzopoulos, R., Blank, T., and Becker, P. B. (1 9 9 4 ). “Transcriptional repression by nucleosomes but not H1 in reconstituted preblastoderm Drosophila chromatin.” Embo J 13, 373-379.

Wiekowski, M., Miranda, M., and DePamphilis, M. L. (1 9 9 1 ). “Regulation of gene expression in preimplantation mouse embryos: effects of the zygotic clock and the first mitosis on promoter and enhancer activities.” D e v B i o l 147, 403414.

Schatt, M. D., Rusconi, S., and Schaffner, W. (1 9 9 0 ). “A single DNA-binding transcription factor is sufficient for activation from a distant enhancer and/or from a promoter position.” Embo J 9, 481-487.

Wiekowski, M., Miranda, M., and DePamphilis, M. L. (1 9 9 3 ). “Requirements for promoter activity in mouse oocytes and embryos distinguish paternal pronuclei from maternal and zygotic nuclei.” D e v B i o l 159, 366-378.

Schickler, M., Lira, S. A., Kinloch, R. A., and Wassarman, P. M. (1 9 9 2 ). “A mouse oocyte-specific protein that binds to a region of mZP3 promoter responsible for oocyte-specific mZP3 gene expression.” M o l C e l l B i o l 12, 120-127.

Wiekowski, M., Miranda, M., Nothias, J. Y., and DePamphilis, M. L. (1 9 9 7 ). “Changes in histone synthesis and modification at the beginning of mouse development correlate with the establishment of chromatin mediated repression of transcription.” J C e l l S c i 110, 1147-1158.

Schultz, R. M. (1 9 9 3 ). “Regulation of zygotic gene activation in the mouse.” B i o e s s a y s 15, 531-538. Seshagiri, P. B., McKenzie, D. I., Bavister, B. D., Williamson, J. L., and Aiken, J. M. (1 9 9 2 ). “Golden hamster embryonic genome activation occurs at the two-cell stage: correlation with major developmental changes.” Molecular Reproduction and Development 32, 229-235.

Wirak, D. O., Chalifour, L. E., Wassarman, P. M., Muller, W. J., Hassell, J. A., and DePamphilis, M. L. (1 9 8 5 ). “Sequencedependent DNA replication in preimplantation mouse embryos.” M o l C e l l B i o l 5, 2924-2935.

Sheridan, P. L., Sheline, C. T., Cannon, K., Voz, M. L., Pazin, M. J., Kadonaga, J. T., and Jones, K. A. (1 9 9 5 ). “Activation of the HIV-1 enhancer by the LEF-1 HMG protein on nucleosome- assembled DNA in vitro.” Ge ne s D e v 9, 2090-2104.

Wolffe, A. P. (1 9 9 1 ). “Implications of DNA replication for eukaryotic gene expression.” J C e l l S c i 99, 201-206.

Smith, R. C., Dworkin-Rastl, E., and Dworkin, M. B. (1 9 8 8 ). “Expression of a histone H1-like protein is restricted to early Xenopus development.” G e n e s D e v 2, 1284-1295.

Workman, J. L., Taylor, I. C., and Kingston, R. E. (1 9 9 1 ). “Activation domains of stably bound GAL4 derivatives alleviate repression of promoters by nucleosomes.” C e l l 64, 533-544.

Wolffe, A. P., and Pruss, D. (1 9 9 6 ). “Targeting chromatin disruption: Transcription regulators that acetylate histones.” C e l l 84, 817-819.

Struhl, K. (1 9 9 8 ). “Histone acetylation and transcriptional regulatory mechanisms.” G e n e s D e v 12, 599-606.

Worrad, D. M., Ram, P. T., and Schultz, R. M. (1 9 9 4 ). “Regulation of gene expression in the mouse oocyte and early preimplantation embryo: developmental changes in Sp1 and TATA box- binding protein, TBP.” D e v e l o p m e n t 120, 2347-2357.

Tazi, J., and Bird, A. (1 9 9 0 ). “Alternative chromatin structure at CpG islands.” C e l l 60, 909-920. Telford, N. A., Watson, A. J., and Schultz, G. A. (1 9 9 0 ). “Transition from maternal to embryonic control in early mammalian development: a comparison of several species.” Mol Repr Dev 26, 90-100. Turner, B. M. (1 9 9 1 ). “Histone acetylation and control of gene expression.” J C e l l S c i 99, 13-20.

464

Gene Therapy and Molecular Biology Vol 3, page 465 Gene Ther Mol Biol Vol 3, 465-474. August 1999.

Molecular mechanisms that regulate hyaluronan synthesis Review Article

Paraskevi Heldin Department of Medical Biochemistry and Microbiology, Biomedical Center, Box 575, S-751 23 Uppsala, Sweden. _________________________________________________________________________________________________ Correspondence: Paraskevi Heldin, Ph.D., Phone: 46 18 4714261; Fax: 46 18 4714975; E-mail: Paraskevi.Heldin@medkem.uu.se Key words: Hyaluronan, hyaluronan synthase, HAS1, cytokine, extracelular matrix A b b r e v i a t i o n s : HAS, hyaluronan synthase; PDGF, platelet-derived growth factor; TGF-ß, transforming growth factor-ß; TNF , tumor necrosis factor; IL, interleukin; IFN, interferon Received: 24 December 1998; accepted: 31 December 1998

Summary Hyaluronan i s an ubiquitous glycosaminoglycan found i n almost all tissues o f the body in vertebrates and i n the extracellular capsule o f certain pathogenic bacteria. Hyaluronan i s biologically active and affects cell migration and proliferation. The amount of hyaluronan in the extracellular matrix increases during inflammation, wound healing and certain forms of cancer. Hyaluronan i s synthesized by a membrane-bound enzyme which uses UDP-sugar nucleotides precursors (UDP-glucuronic acid and UDP-N-acetylglucosamine). Three mammalian hyaluronan synthase genes (HAS1, HAS2 and HAS3) have been identified. The encoded proteins share 56-71% sequence similarities. The synthesis o f hyaluronan i s carefully regulated i n a c e l l specific manner. Certain growth factors and cytokines lead to an increased activity of existing synthase molecules as well as induce the synthesis of new enzyme molecules.

Annau, 1953; Laurent and Reed, 1991). Excessive amounts of hyaluronan are found during the first stage in the generation of extracellular matrix. This stage is followed by a decrease in the hyaluronan content as cell growth and differentiation ensues. In normal healthy tissues the amount of hyaluronan is maintained in equilibrium by a balance between synthesis and degradation. However, during inflammation, wound healing and certain forms of cancer the amount of hyaluronan increases and hyaluronan becomes the dominant glycosaminoglycan in the extracellular matrix. Accumulation of hyaluronan leads to retention of fluid which most likely contributes to the swelling of tissues characterizing inflammation (Laurent, 1998; Knudson, 1998). However, during restoration of damaged tissue hyaluronan-rich matrix has been shown to have beneficial effects; increased amounts of hyaluronan in the matrix forms a favorable environment which promotes cell migration and proliferation. For example, treatment with

I. Introduction Hyaluronan was first isolated by Meyer and Palmer (1934) from the vitreous body of the eye and therefore named hyaluronic acid (from the Greek word hyalos, which means glass). Its architectural construction of repeating disaccharide units [D-glucuronic acid (1-!-3) N-acetyl-Dglucosamine(1-!-4)]n was established twenty years later by Weissmann and Meyer (1954). Hyaluronan is an unbranched linear polysaccharide with a molecular mass ranging from about 200 kDa in blood to 5000 kDa in synovial fluid. The largest molecules have a chain length in the average of 10 µm. The chains form in solution kinked coils which immobilize water within their domains (Laurent and Fraser, 1992; Fraser and Laurent, 1996) (Figure 1). Hyaluronan is found virtually in every tissue and body fluid in vertebrates and in the capsules of certain bacterial pathogens, e.g Gram-positive group A and C streptococci and type A Pasteurella multocida (Carter and 465

Heldin: Molecular mechanisms that regulate hyaluronan synthesis

466

Gene Therapy and Molecular Biology Vol 3, page 467 Figure 1. Schematic drawing of cell surface associated hyaluronan chains and hyaluronan chains asociated with extracelular matrix proteins.

Figure 2. Hyaluronan containing coat surrounding a normal human mesothelial cell. The coat is visualized by the exclusion of formalin fixed erythrocytes. The picture was kindly provided by Dr. H책kan Pertoft (Uppsala, Sweden).

467

Gene Therapy and Molecular Biology Vol 3, page 467 based products in areas such as drug delivery and tissue repair (Balazs and Laurent, 1998).

hyaluronan gel led to improved healing with less scar formation of damaged tendons and the tympanic membrane of the ear (Balazs and Denlinger, 1989; Laurent et al, 1986). The biological effects of hyaluronan in tumor tissues have not yet been clarified. Hyaluronan may facilitate tumor invasion (Lesley et al, 1997; Knudson, 1998) or may be involved in a host defense response (Teder et al, 1995).

Hyaluronan forms a capsule around group A and group C Streptococci. This is a perfect camouflage which allows these bacteria to be more successful pathogens by evading host defense (Schmidt et al, 1996; Husmann et al, 1997). Acapsular mutants of group A Streptococci exhibit considerable losses in virulence (Wessels et al, 1994).

Hyaluronan stabilizes the architecture of extracellular matrix through its interactions with specific matrix proteins and receptors, collectively referred to as hyaladherins (Toole, 1990). Moreover, it is involved in many biological processes modulating cell behavior during embryogenesis, inflammation and tumorigenesis (Knudson and Knudson, 1993; Sherman et al, 1994). These effects may be mediated through hyaluronan containing coats which constitute the immediate cellular environment (Figure 2). These coats are three-dimensional gels formed around some cells, and can be visualized in vitro by their exclusion of formalin-fixed erythrocytes (Clarris and Fraser, 1968; Knudson et al, 1993; Heldin and Pertoft, 1993). The most thoroughly studied hyaladherins are the link protein and aggrecan of cartilage (Hardingham and Muir, 1972) which have analogues in other tissues such as versican in connective tissues (Zimmermann and Ruoslahti, 1989), and hyaluronectin in the brain (Delpech et al, 1989). Hyaladherins are also found in certain body fluids, e.g. tumor necrosis factor (TNF)-stimulated gene-6 (TSG-6) found in synovial fluids of patients with arthritis and in sera of patients with different inflammatory or autoimmune disorders. TSG-6 forms a complex with the plasma protein inter-"-trypsin inhibitor that is also a hyaluronan binding protein (Day and Parkar, 1998). Cellassociated hyaluronan receptors are not only confined to connective tissue cells. Hyaluronan binding to cells of non mesodermal origin was demonstrated by the aggregation of lymphoma cells but not normal lymphocytes after addition of the polysaccharide (Wasteson et al, 1973). A putative hyaluronan receptor is the lymphocyte homing receptor gp90Hermes (CD44). The interaction between CD44 and hyaluronan leads to internalization of hyaluronan by lung macrophages and chondrocytes, adhesion of lymphocytes to the endothelium, formation of pericellular matrices around chondrocytes and increased invasion of certain tumors (Knudson and Knudson, 1993; Sherman et al, 1994). Another hyaluronan receptor is the receptor for hyaluronanmediated motility (RHAMM) which affects cell locomotion and proliferation (Hall et al, 1994, 1995; Savani et al, 1995). Furthermore, the liver endothelial cells carry specific receptors for clearance of hyaluronan from the blood (Smedsrod et al, 1990). The non-immunogenic character of hyaluronan and its ability to bind specifically to cell surface receptors has led to the use of hyaluronan-

II. Hyaluronan biosynthesis The molecular mechanisms that regulate the biosynthetic pathway of hyaluronan are yet unclear. Hyaluronan biosynthesis in mammalian cells differs from that of other glycosaminoglycans which are synthesized in the endoplasmic reticulum/Golgi and then transported to the cell surface. The key enzymes in hyaluronan biosynthesis are hyaluronan synthases (HAS), a family of enzymes located in plasma membrane, which have beeen identified both in bacteria and vertebrates (Sugahara et al, 1979; Prehm, 1984; Philipson and Schwartz, 1984). During the synthesis process the hyaluronan chains are extruded into the intercellular space. This unique route of biosynthesis may be facilitated through interactions between the hydrophobic patches of hyaluronan chains and plasma membrane phospholipids (Ghosh et al, 1994); continuous extrusion through the membrane may be necessary because of the large size of hyaluronan chains. Hyaluronan synthesis requires intracellular sources of UDP-glucuronic acid (UDP-GlcA) and UDP-N-acetylglucosamine (UDP-GlcNAc), which are generated in the glycolytic pathway (O´Regan et al, 1994).

A. Growth factor and cytokine regulation of hyaluronan biosynthesis The tissue hyaluronan content is increased during inflammation, wound healing, certain malignancies, e.g. mesotheliomas and Wilm´s tumor, and in other destructive injuries such as reumatoid arthritis. Some pathways on the molecular mechanisms of regulation of hyaluronan synthesis during cancer cell metastasis, wound healing and inflammation have recently been deciphered. Initial studies by us and other laboratories revealed that exogenously added growth factors, conditioned media, body fluids, as well as tumor promotors, lead to stimulation of hyaluronan synthesis in cultures of mesenchymal origin, e.g. fibroblasts (Heldin et al, 1989; Westergren-Thorsson G, 1990; Suzuki et al, 1995), normal human mesothelial cells (Honda et al, 1991; Heldin et al, 1992; Heldin and Pertoft, 1993), and hepatic stellate cells (Heldin et al, 1991; Gressner and Haarmann, 1988; Vrochides et al, 1996). Among these stimulators, platelet-derived growth factor (PDGF)-BB, transforming growth factor-ß (TGF-ß)

467

Heldin: Molecular mechanisms that regulate hyaluronan synthesis and the tumor promotor TPA exhibited the highest activity. PDGF-BB and TGF-Ă&#x; mediated their stimulatory activity on hyaluronan synthesis at least in part through protein kinase C (PKC), since the PKC inhibitor calphostin C inhibited most of the hyaluronan synthesis induced by the growth factors. Direct activation of PKC by phorbol esters also stimulated hyaluronan production, and the combination of either PDGF or TGF-Ă&#x; and TPA gave an increased effect. The stimulations by PDGF and TGF-Ă&#x; were dependent on protein synthesis since cycloheximide inhibited their effects. In contrast, the effects mediated by TPA were not (Suzuki et al, 1995) (F i g u r e 3 ). Thus, these results indicate that PKC is involved in the transduction of the effects of growth factors on hyaluronan biosynthesis in foreskin fibroblasts and normal human mesothelial cells, and that the effects involve direct or indirect activation of existing HAS molecules, as well as induction of new enzyme molecules or other factors involved in the transduction of growth factor signaling (Suzuki et al, 1995). The insulin-growth factor (IGF)-I and epidermal growth factor (EGF) also enhance the production of hyaluronan in mesothelial cell cultures derived from pericardium through a receptor tyrosine-kinase-involved transmembrane signaling process (Honda et al, 1991).

to cytokines, e. g. tumor necrosis factor (TNF)-", interleukin (IL)-1", interferon (IFN)-# , and leukoregulin (Elias et al, 1988; Butler et al, 1988; Smith et al, 1995; Sampson et al, 1992). For example, treatment of human orbital fibroblasts (Smith et al, 1995) by leukoregulin, a cytokine released by activated T lymphocytes, resulted in a 3-fold higher stimulation of hyaluronan synthesis compared to leukoregulin stimulation of dermal fibroblasts (Mauviel et al, 1991). The leukoregulin-induced accumulation of hyaluronan in orbital fibroblasts was dependent on ongoing protein synthesis and was not mediated through the PKC pathway (Smith et al, 1995). However, prostaglandin E2 stimulated hyaluronan synthesis in mesothelial cell cultures obtained from pericardium through a cAMP signaling pathway (Honda et al, 1993). Further in vitro studies revealed that combinations of IFN# and TNF-" augmented the synthesis of high molecular mass hyaluronan whereas combination of IL-1 and TNF-" induced the production of hyaluronan chains of low molecular mass (Sampson et al, 1992). These findings may be important for our understanding of the presence of abnormally low molecular weight hyaluronan chains (about 50 kDa) in several inflammatory disorders, such as in reumatoid arthritis where the hyaluronan found in joints has a low viscosity (Dahl and Husby, 1985). The differencies in the ability to synthesize hyaluronan may be due both to differences in intracellular signaling pathways (Figure 3) as well as to variations in the regulation of the expression of the three HAS isoforms (see below).

In inflammed tissues, hyaluronan accumulation is induced following infiltration by inflammatory cells which secrete growth factors and cytokines, many of which are capable of stimulating hyaluronan synthesis. In vitro studies on hyaluronan biosynthesis in human fibroblasts derived from various organs, revealed differential responses

Figure 3. Schematic depiction of mechanisms that regulate hyaluronan synthesis.

468

Gene Therapy and Molecular Biology Vol 3, page 468

Figure 4. Comparison of amino acid sequencies between vertebrate HAS enzymes (mouse, mHAS1, mHAS2, mHAS3; frog, DG42) and bacteria enzymes (Streptococcus pyogenes, spHAS; Streptococcus equisimilis, seHAS). Conserved residues are boxed.

mechanisms involved in their inhibitory action on hyaluronan synthesis are not known.

B. Bacterial and vertebrate hyaluronan synthases

Non-steroidal anti-inflammatory drugs, such as indomethacin, have been shown to inhibit in vitro synthesis of hyaluronan by skin fibroblasts (August et al, 1994). Glucocorticoid and thyroid hormones also inhibit hyaluronan synthesis in skin fibroblast cultures (Smith et al, 1982; Smith, 1984). The glucocorticoid-dependent prevention of hyaluronan synthesis has physiological implications since it has been demonstrated that glucocorticoid therapy induces skin atrophy due to depletion of hyaluronan and proteoglycans which leads to closer packing of collagen fibrils (Lehman et al, 1983). In addition to these naturally occuring hormones, carboxylic acids, such as n-butyrate, have also been demonstrated to inhibit hyaluronan synthesis (Smith, 1987). The

Although hyaluronan was isolated about 60 years ago, the genes involved in its biosynthesis were identified and characterized only recently. Their identification has been approached by isolating stable acapsular mutants of Streptoccocci via transposon mutagenesis and subsequent identification of the DNA flanking the inserted transposon (Dougherty and van de Rijn, 1994; DeAngelis et al, 1993a,b). These studies led to the identification of group A Streptococci genetic locus for hyaluronan synthesis termed has (hyaluronic acid synthesis). The has operon is composed of three genes, hasA, hasB, and hasC. The hasA gene codes for a hyaluronan synthase designated spHAS for S. pyogenes HAS (Weigel et al., 1997). The hasB gene

468

Heldin: Molecular mechanisms that regulate hyaluronan synthesis encodes UDP-glucose dehydrogenase (converts UDPglucose to UDP-GlcA) and hasC gene exhibits homology to UDP-glucose pyrophosphorylase (converts glucose 1phosphate and UTP to UDP-glucose). The mechanisms that regulate the expression of the has operon are yet not known. However, the has operon is adjacent to genes involved in DNA repair (DeAngelis and Weigel, 1995). Operons similar to the has operon in Streptococcus pyogenes are also found in other Gram-positive bacteria such as Streptococcus pneumonia; the Cap3B glycosyltransferase enzyme which produces a cellobiuronic acid is encoded in an operon flanked by UDP-glucose dehydrogenase and UDP-glucose pyrophosphorylase (Arrecubieta et al, 1996). The spHAS protein has a molecular weight of 42 kDa (DeAngelis and Weigel, 1994), possesses 4 transmembrane domains and one inner membrane-associated domain (Weigel et al, 1997; TlapakSimmons et al, 1998) with a large intracellular loop which contains the catalytic activity of the enzyme. However, its exact membrane topology has not yet been experimentally derived. The spHAS protein shows about 10% similarity to the rizobium nodC gene product (a fungal chitin synthase) (Atkinson and Long, 1992), and 30% similarity to the Xenopus laevis protein DG42 which was expressed transiently in frog embryos (Developmentally expressed during Gastrulation) (Rosa et al, 1988; DeAngelis et al, 1993a). Recently the hyaluronan synthase from the highly encapsulated strain of group C Streptococcus equisimilis, seHAS, was also cloned (Kumari and Weigel, 1997). The two bacterial synthases exhibit about 70% amino acid sequence similarity and are of similar size (Figure 4). There is, however, a difference in the rate of hyaluronan synthesis; seHAS polymerizes the growing hyaluronan chains 2-fold faster. More recently a new hyaluronan synthase from the bacterial pathogen type A Pasteurella multocida, PmHAS, was cloned (DeAngelis et al, 1998). PmHAS is twice the size of other bacterial HAS (972 amino acids versus 417 residues) and shows higher amino acid similarity to other bacterial glycosyltransferases, such as galactose transferase, than to the known HAS molecules from streptococci. Moreover, PmHAS is predicted to have only two transmembrane domains and possibly both termini are intracellular; this implies that a large part of the enzyme is located outside of the cell. It is a challenge to elucidate the mechanisms through which different bacterial HAS polymerize the same product.

enzyme found which had hyaluronan synthase activity. However, other studies have shown that DG42 synthesizes chitin even more efficiently (Semino et al, 1996; Varki, 1996). In 1996, almost simultaneously, several groups reported the molecular cloning of two mammalian HAS genes designated HAS1 and HAS2 (Watanabe and Yamaguchi, 1996; Itano and Kimata, 1996; Shyjan et al, 1996; Spicer et al, 1996). One year later a third HAS gene, HAS3, was identified (Spicer et al, 1997) which was found to exhibit 57%, 71%, 56% and 28% amino acid sequence similarities to HAS1, HAS2, DG42, and spHAS, respectively (Figure 4). The similarities between the same HAS gene from different mammalian species are larger than 95%. An important question raised is of course how the expression and enzymatic activity of HAS are regulated. The mammalian HAS gene family (Spicer et al, 1997; Spicer and McDonald, 1998) has most likely arisen by sequencial gene duplication and divergence. Family members share characteristics such as similarity in flanking genes, location within the genome and location of exon-intron boundaries (Ruddle et al, 1994; Brown et al, 1995; Aruga et al, 1996). Mammalian HAS are located in different chromosomes, HAS1 on human chromosome 19q13.3-q13.4, HAS2 on 8q24.12, and HAS3 on 16q22.1. However, similarities in exon-intron boundaries as well as similarities in the expression pattern of HAS genes in developing mouse embryos (Spicer et al, 1997; Spicer and McDonald, 1998) support the notion that mammalian HAS isoforms belong to an ancient gene family and that the related genes appeared through sequential gene duplication events. Surprisingly, a glycosyltransferase with the ability to synthesize hyaluronan has recently been described in Chlorella virus PBCV-1 that infects a chlorella-like green algae (DeAngelis et al, 1997). This is the first report where hyaluronan is found outside vertebrates and pathogens. It is important to point out here that the PBCV-1 genome possess genes that encode, in similarity with A Streptococci, UDP-Glc dehydrogenase and fructose-6phosphate amidotransferase that is involved in the UDPGlcNAc metabolic pathway. Therefore it is likely that the HAS operon in group A Streptococci evolved in context with other capsule polysacharide operons in procaryotic organisms as well as in some virus genomes. Further studies are needed in order to understand the evolution of HAS enzymes.

The similarity of spHAS with DG42, whose function at that time was unknown, led to a series of experiments, e.g infection of mammalian cells with a DG42 vaccinia virus construct (Meyer and Kreil, 1996), as well as transfection of recombinant enzyme in yeast cells (DeAngelis and Achyuthan, 1996; Pummill et al, 1998) showing that the gene product exhibited hyaluronan synthesizing activity. Thus, DG42 was the first vertebrate

The expression pattern of mRNAs for the three HAS isoforms in response to growth factors differs, indicating different functional roles of the synthases (Sugiyama et al, 1998). Moreover, studies in our laboratory suggest that the three eukaryotic HAS isoforms are expressed differentially, and possess different intrinsic properties in their abilities to polymerize hyaluronan and in their interactions with other 470

Gene Therapy and Molecular Biology Vol 3, page 471 proteins (unpublished observations). Recently, Yamada and his collegues (1998) have reported the structural organization of the mouse HAS1 gene; a CCAAT box is located in the promoter region of the HAS1 gene upstream of the transcription initiation site, as well as binding sites for AP-2 (activated in response to cAMP and phorbol ester), MyoD (regulatory gene for skeletal myogenesis), SPY and Sox-5 genes (expressed during embryogenesis), and IRF-1, IRF-2, and p53. Further studies by site-directed mutations in the HAS1 gene and expression in transgenic mice may elucidate the regulation of HAS1 expression. Yet, nothing is known about the structural organization of the HAS2 and HAS3 genes.

production by indomethacin and mefenamic acid. Cancer Lett 82, 49-54. Balazs EA, Denlinger JL. (1 9 8 9 ). Clinical uses of hyaluronan: In T h e B i o l o g y o f H y a l u r o n a n . Evered D, and Whelan J, eds. (Chichester, Wiley & Sons Ltd) vol. 143, pp. 265-280. Balazs EA, Laurent TC.(1 9 9 8 ) Round table discussion: new applications for hyaluronan: In The C h e m i s t r y , Biology and Medical applications of H y a l u r o n a n a n d i t s D e r i v a t i v e s . Laurent TC, ed. (London, Portland Press) vol. 72, pp. 325-336 Brown NP, Whittaker AJ, Newell WR, Rawlings CJ, Beck S (1 9 9 5 ). Identification and analysis of multigene families by comparison of exon fingerprints. J . M o l . B i o l . 249, 342-359.

III. Future prospects Hyaluronan has preserved its simple primary structure throughout the evolution in contrast to the diversification seen in other macromolecules of the extracellular matrix. In inflammatory conditions, such as lung fibrosis, myocardial infarctions, as well as in certain invasive tumors, increased levels of hyaluronan are observed in tissues and exudates. This accumulation of hyaluronan often leads to organ dysfunction and increased tumor invasion and, therefore, the excess hyaluronan has to be eliminated. Increased knowledge about the molecular mechanisms which regulate the activities of hyaluronan synthases will make it possible to design specific inhibitors for hyaluronan synthases which may be of clinical value.

Butler DM, Vitti GF, Leizer T, Hamilton JA (1 9 8 8 ). Stimulation of the hyaluronic acid levels of human synovial fibroblasts by recombinant human tumor necrosis factor ", tumor necrosis factor ! (lymphotoxin), interleukin-1", and interleukin-1!. A r t h r i t i s R h e u m . 31, 1281-1289. Carter GR, Annau E (1 9 5 3 ). Isolation of capsular polysaccharides from colonila variants of Pasteurella Multicida. A m . J . V e t . R e s . 14, 475-478. Clarris BJ, Fraser JRE (1 9 6 8 ). On the pericellular zone of some mammalian cells in vitro. E x p . C e l l R e s . 49, 181-193. Dahl INS, Husby G (1 9 8 5 ). Hyaluronic acid production in vitro by synovial lining cells from normal and rheumatoid joints. A n n . R h e u m . D i s . 44, 647-657.

Acknowledgement

Day AJ, Parkar AA. (1 9 9 8 ) The structure of the link module: a hyaluronan-binding domain: In The C h e m i s t r y , Biology and Medical applications of H y a l u r o n a n a n d i t s D e r i v a t i v e s . Laurent TC, ed. (London, Portland Press) vol. 72, pp. 141-147

I would like to thank prof. Torvard Laurent for constructive criticism of this review.

References Arrecubieta C, Lopez R, Garcia E (1 9 9 6 ). Type 3-specific synthase of Streptococcus pneumoniae (Cap3B) directs type 3 polysaccharide biosynthesis in Escherichia coli and in pneumococcal strains of different serotypes. J . E x p . M e d . 184, 449-455.

DeAngelis PL, Achyuthan AM (1 9 9 6 ). Yeast-derived recombinant DG42 protein of xenopus can synthesize hyaluronan in vitro. J . B i o l . C h e m . 271, 2365723660. DeAngelis PL, Jing W, Drake RR, Achyuthan AM (1 9 9 8 ). Identification and molecular cloning of a unique hyaluronan synthase from Pasteurella multocida. J . B i o l . C h e m . 273, 8454-8458.

Aruga J, Nagai T, Tokuyama T, Hayashizaki Y, Okazaki Y, Chapman VM, Mikoshiba K (1 9 9 6 ). The mouse zic gene family. Homologues of the Drosophila pair-rule gene oddpaired. J . . B i o l . C h e m . 271, 1043-1047.

DeAngelis PL, Jing W, Graves MV, Burbank DE, Van Etten JL (1 9 9 7 ). Hyaluronan synthase of chlorella virus PBCV-1. S c i e n c e 278, 1800-1803.

Atkinson EM, Long SR (1 9 9 2 ). Homology of Rhizobium meliloti NodC to polysaccharide polymerizing enzymes. M o l . P l a n t . M i c r o b e I n t e r a c t . 5, 439-442.

DeAngelis PL, Papaconstantinou J, Weigel PH (1 9 9 3 a ). Isolation of a Streptococcus pyogenes gene locus that directs hyaluronan biosynthesis in acapsular mutants and in heterologous bacteria. J . B i o l . C h e m . 268, 1456814571.

August EM, Nguyen T, Malinowski NM, Cysyk RL (1 9 9 4 ). Non-steroidal anti-inflammatory drugs and tumor progression: inhibition of fibroblast hyaluronic acid

471

Heldin: Molecular mechanisms that regulate hyaluronan synthesis DeAngelis PL, Papaconstantinou J, Weigel PH (1 9 9 3 b ). Molecular cloning, identification, and sequence of the hyaluronan synthase gene from group A Streptococcus pyogenes. J . B i o l . C h e m . 268, 19181-19184.

Hardingham TE, Muir H (1 9 7 2 ). The specific interaction of hyaluronic acid with cartillage proteoglycans. B i o c h i m . B i o p h y s . A c t a . 279, 401-405. Heldin P, Asplund T, Ytterberg (1 9 9 2 ). Characterization of involved in the activation of platelet-derived growth factor B i o c h e m . J . 283, 165-170.

DeAngelis PL, Weigel PH (1 9 9 4 ). Immunochemical confirmation of the primary structure of Streptococcal hyaluronan synthase and synthesis of high molecular weight product by the recombinant enzyme. B i o c h e m i s t r y 33, 9033-9039.

D, Thelin S, Laurent TC the molecular mechanism hyaluronan synthetase by in human mesothelial cells.

Heldin P, Laurent TC, Heldin C-H (1 9 8 9 ). Effect of growth factors on hyaluronan synthesis in cultured human fibroblasts. B i o c h e m . J . 258, 919-922.

DeAngelis PL, Weigel PH (1 9 9 5 ). Characterization of the recombinant hyaluronic acid synthase from streptococcus pyogenes: In Genetics of Streptococci, E n t e r o c o c c i and L a c t o c o c c i . Ferretti JJ, Gilmore MS, Klaenhammer TR, Brown F, eds. (Karger, Dev. Biol Stand. Basel) vol. 85, pp. 225-229

Heldin P, Pertoft H (1 9 9 3 ). Synthesis and assembly of the hyaluronan-containing coats around normal human mesothelial cells. E x p . C e l l R e s . 208, 422-429. Heldin P, Pertoft H, Nordlinder H, Heldin C-H, Laurent TC (1 9 9 1 ). Differential expression of platelet-derived growth factor "- and !-receptors on fat-storing cells and endothelial cells of rat liver. E x p . C e l l R e s . 193, 364369.

Delpech B, Delpech A, Bruckner G, Girard N, Maingonnat C. (1 9 8 9 ) Hyaluronan and hyaluronectin in the nervous system: In T h e B i o l o g y o f H y a l u r o n a n . Evered D, and Whelan J, eds. (Chichester, Wiley & Sons Ltd) vol. 143, pp. 208-220.

Honda A, Noguchi N, Takehara H, Ohashi Y, Asuwa N, Mori Y (1 9 9 1 ). Cooperative enhancement of hyaluronic acid synthesis by combined use of IGF-I and EGF, and inhibition by tyrosine kinase inhibitor genistein, in cultured mesothelial cells from rabbit pericardial cavity. J . C e l l S c i . 98, 91-98.

Dougherty BA, van de Rijn I (1 9 9 4 ). Molecular characterization of hasA from an operon required for hyaluronic acid synthesis in group A streptococci. J . B i o l . C h e m . 269, 169-175. Elias JA KR, Freundlich B, Sampson PM. (1 9 8 8 ). Regulation of human lung fibroblast glycosaminoglycan production by recombinant interferons, tumor necrosis factor, and lymphotoxin. J . C l i n . I n v e s t . 81, 325-333.

Honda A, Sekiguchi Y, Mori Y (1 9 9 3 ). Prostaglandin E2 stimulates cyclic AMP-mediated hyaluronan synthesis in rabbit pericardial mesothelial cells. B i o c h e m . J . 292 (Pt 2), 497-502.

Fraser JRE, Laurent TC. (1 9 9 6 ) Hyaluronan: In Extracellular matrix. Comper WD, ed. (Amsterdam, Harwood Academic) vol. 2, pp. 141-199.

Husmann LK, Yung DL, Hollingshead SK, Scott JR (1 9 9 7 ). Role of putative virulence factors of Streptococcus pyogenes in mouse models of long-term throat colonization and pneumonia. I n f e c t . Immun. 65, 1422-1430.

Ghosh P, Hutadilok N, Adam N, Lentini A (1 9 9 4 ). Interactions of hyaluronan (hyaluronic acid) with phospholipids as determined by gel permeation chromatography, multi-angle laser-light- scattering photometry and 1H-NMR spectroscopy. I n t . J . B i o l . M a c r o m o l . 16, 237-244.

Itano N, Kimata K (1 9 9 6 ). Expression cloning and molecular characterization of HAS protein, a eukaryotic hyaluronan synthase. J . B i o l . C h e m . 271, 9875-9878.

Gressner AM, Haarmann R (1 9 8 8 ). Hyaluronic acid synthesis and secretion by rat liver fat storing cells (Perisinisoidal Lipocytes) in culture. B i o c h e m . B i o p h y s . R e s . Commun. 151, 222-229.

Knudson CB, Knudson W (1 9 9 3 ). Hyaluronan-binding proteins in development, tissue homeostasis, and disease. FASEB J. 7, 1233-1241. Knudson W. (1 9 9 8 ) Hyaluronan in malignancies: In The Chemistry, Biology and Medical applications o f H y a l u r o n a n a n d i t s D e r i v a t i v e s . Laurent TC, ed. (London, Portland Press) vol. 72, pp. 169-179.

Hall CL, Wang C, Lange LA, Turley EA (1 9 9 4 ). Hyaluronan and the hyaluronan receptor RHAMM promote focal adhesion turnover and transient tyrosine kinase activity. J . C e l l B i o l . 126, 575-588.

Knudson W (1 9 9 8 ). The role of CD44 as a cell surface hyaluronan receptor during tumor invasion of connective tissue. F r o n t i e r s B i o s c i 3, d604-615.

Hall CL, Yang B, Yang X, Zhang S, Turley M, Samuel S, Lange LA, Wang C, Curpen GD, Savani RC, et al. (1 9 9 5 ). Overexpression of the hyaluronan receptor RHAMM is transforming and is also required for H-ras transformation. C e l l 82, 19-26.

Knudson W, Bartnik E, Knudson CB (1 9 9 3 ). Assembly of pericellular matrices by COS-7 cells transfected with CD44 lymphocyte-homing receptor genes. P r o c . N a t l . A c a d . S c i . U S A 90, 4003-4007.

472

Gene Therapy and Molecular Biology Vol 3, page 473 Kumari K, Weigel PH (1 9 9 7 ). Molecular cloning, expression, and characterization of the authentic hyaluronan synthase from group C Streptococcus equisimilis. J . B i o l . C h e m . 272, 32539-32546.

gradient pattern during Xenopus embryogenesis. D e v . B i o l . 129, 114-123. Ruddle FH, Bentley KL, Murtha MT, Risch N (1 9 9 4 ). Gene loss and gain in the evolution of the vertebrates. D e v Suppl 155-161.

Laurent C, Hellstrom S, Stenfors LE (1 9 8 6 ). Hyaluronic acid reduces connective tissue formation in middle ears filled with absorbable gelatin sponge: an experimental study. A m . J . O t o l a r y n g o l . 7, 181-186.

Sampson PM, Rochester CL, Freundlich B, Elias JA (1 9 9 2 ). Cytokine regulation of human lung fibroblast hyaluronan (hyaluronic acid) production. J . C l i n . I n v e s t . 90, 1492-1503.

Laurent TC.(1 9 9 8 ) Hyaluronan as a clinical marker of pathological processes: In T h e C h e m i s t r y , B i o l o g y and Medical a p p l i c a t i o n s o f Hyaluronan and i t s D e r i v a t i v e s . Laurent TC, ed. (London, Portland Press) vol. 72, pp. 305-313

Savani RC, Wang C, Yang B, Zhang S, Kinsella MG, Wight TN, Stern R, Nance DM, Turley EA (1 9 9 5 ). Migration of bovine aortic smooth muscle cells after wounding injury. The role of hyaluronan and RHAMM. J . C l i n . I n v e s t . 95, 1158-1168.

Laurent TC, Fraser JRE (1 9 9 2 ). Hyaluronan. F A S E B J . 6 , 2397-2404.

Schmidt KH, Gunther E, Courtney HS (1 9 9 6 ). Expression of both M protein and hyaluronic acid capsule by group A streptococcal strains results in a high virulence for chicken embryos. Med. M i c r o b i o l . I m m u n o l . ( B e r l ) 184, 169-173.

Laurent UBG, Reed RK (1 9 9 1 ). Turnover of hyaluronan in the tissues. A d v . D r u g D e l i v e r y R e v i e w s 7, 237-256. Lehman P, Zheng P, Lavker RM, Kligman AM (1 9 8 3 ). Corticosteroid atrophy in human skin. A study by light, scanning and transmission electron microscopy. J . I n v e s t . D e r m a t o l . 81, 169-176.

Semino CE, Specht CA, Raimondi A, Robbins PW (1 9 9 6 ). Homologs of the Xenopus developmental gene DG42 are present in zebrafish and mouse and are involved in the synthesis of Nod-like chitin oligosaccharides during early embryogenesis. P r o c . N a t l . Acad. S c i 93, 45484553.

Lesley J, Hyman R, English N, Catterall JB, Turner GA (1 9 9 7 ). CD44 in inflammation and metastasis. G l y c o c o n j u g a t e J . 14, 611-622. Mauviel A RF, Hartmann DJ, Pujol J-P, Evans CH. (1 9 9 1 ). Modulation of human dermal fibroblast extracellular matrix metabolism by the lymphokine leukoregulin. J C e l l B i o l 113, 1455-1462.

Sherman L, Sleeman J, Herrlich P, Ponta H (1 9 9 4 ). Hyaluronate receptors: key players in growth, differentiation, migration and tumor progression. Curr. O p i n . C e l l B i o l . 6, 726-733.

Meyer K, Palmer JW (1 9 3 4 ). The polysaccharide of the viteous humor. J . B i o l . C h e m . 107, 629-634.

Shyjan AM, Heldin P, Butcher EC, Yoshino T, Briskin MJ (1 9 9 6 ). Functional cloning of the cDNA for a human hyaluronan synthase. J . B i o l . C h e m . 271, 2339523399.

Meyer MF, Kreil G (1 9 9 6 ). Cells expressing the DG42 gene from early Xenopus embryos synthesize hyaluronan. P r o c . N a t l . A c a d . S c i . 93, 4543-4547.

Smedsrod B, Pertoft H, Gustafson S, Laurent TC (1 9 9 0 ). Scavenger functions of the liver endothelial cell. B i o c h e m J . 266, 313-327.

O´Regan M, Martini I, Crescenzi F, De Luka C, Lansing M (1 9 9 4 ). Molecular mechanisms and genetics of hyaluronan biosynthesis. I n t . J . B i o l . M a c r o m o l . 16, 283-286.

Smith TJ (1 9 8 4 ). Dexamethasone regulation of glycosaminoglycan synthesis in cultured human skin fibroblasts. Similar effects of glucocorticoid and thyroid hormones. J . C l i n . I n v e s t . 74, 2157-2163.

Philipson LH, Schwartz NB (1 9 8 4 ). Subcellular localization of hyaluronate synthetase in oligodendroglioma cells. J . B i o l . C h e m . 259, 5017-5023. Prehm P (1 9 8 4 ). Hyaluronate is synthesized at plasma membranes. B i o c h e m . J . 220, 597-600.

Smith TJ (1 9 8 7 ). n-butyrate inhibition of hyaluronate synthesis in cultured human fibroblasts. J . C l i n . I n v e s t . 79, 1493-1497.

Pummill PE, Achyuthan AM, DeAngelis PL (1 9 9 8 ). Enzymological characterization of recombinant xenopus DG42, a vertebrate hyaluronan synthase. J . B i o l . C h e m . 273, 4976-4981.

Smith TJ, Murata Y, Horwitz AL, Philipson L, Refetoff S (1 9 8 2 ). Regulation of glycosaminoglycan synthesis by thyroid hormone in vitro. J . C l i n . I n v e s t . 70, 10661072.

Rosa F, Sargent TD, Rebbert ML, Michaels GS, Jamrich M, Grunz H, Jonas E, Winkles JA, Dawid IB (1 9 8 8 ). Accumulation and decay of DG42 gene products follow a

Smith TJ, Wang HS, Evans CH (1 9 9 5 ). Leukoregulin is a potent inducer of hyaluronan synthesis in cultured human orbital fibroblasts. A m . J . P h y s i o l . 268, C382-388.

473

Heldin: Molecular mechanisms that regulate hyaluronan synthesis Spicer AP, Augustine ML, McDonald JA (1 9 9 6 ). Molecular cloning and characterization of a putative mouse hyaluronan synthase. J . B i o l . C h e m . 271, 2340023406.

hyaluronan by nonparenchymal liver cells during liver regeneration. H e p a t o l o g y 23, 1650-1655. Wasteson Å, Westermark B, Lindahl U, Pontén J (1 9 7 3 ). Aggregation of Feline lymphoma cells by hyaluronic acid. 12, 169-178.

Spicer AP, McDonald JA (1 9 9 8 ). Characterization and molecular evolution of a vertebrate hyaluronan synthase gene family. J . B i o l . C h e m . 273, 1923-1932.

Watanabe K, Yamaguchi Y (1 9 9 6 ). Molecular identification of a putative human hyaluronan synthase. J . B i o l . C h e m . 271, 22945-22948.

Spicer AP, Olson JS, McDonald JA (1 9 9 7 ). Molecular cloning and characterization of a cDNA encoding the third putative mammalian hyaluronan synthase. J . B i o l . C h e m . 272, 8957-8961.

Weigel PH, Hascall VC, Tammi M (1 9 9 7 ). Hyaluronan synthases. J . B i o l . C h e m . 272, 13997-14000. Weissmann B, Meyer K (1 9 5 4 ). The structure of hyalobiuronic acid and of hyaluronic acid from umbilical cord. J . A m . C h e m . S o c . 27, 1753-1757.

Spicer AP, Olson JS, McDonald JA (1 9 9 7 ). Molecular Cloning and Characterization of a cDNA Encoding the Third Putative Mammalian Hyaluronan Synthase. J . B i o l . C h e m . 272, 8957-8961.

Wessels MR, Goldberg JB, Moses AE, DiCesare TJ (1 9 9 4 ). Effects on virulence of mutations in a locus essential for hyaluronic acid capsule expression in group A Streptococci. I n f e c t i o n I m m u n . 62, 433-441.

Spicer AP, Seldin MF, Olsen AS, Brown N, Wells DE, Doggett NA, Itano N, Kimata K, Inazawa J, McDonald JA (1 9 9 7 ). Chromosomal localization of the human and mouse hyaluronan synthase genes. G e n o m i c s 41, 493-497.

Westergren-Thorsson G SB, Fransson L-Å, Malmström A (1 9 9 0 ). TGF-! enhances the production of hyaluronan in human lung but not in skin fibroblasts. E x p . C e l l R e s . 186, 192-195.

Sugahara K, Schwartz NB, Dorfman A (1 9 7 9 ). Biosynthesis of hyaluronic acid by Streptococcus. J . B i o l . C h e m . 254, 6252-6261.

Yamada Y, Itano N, Zako M, Yoshida M, Lenas P, Niimi A, Ueda M, Kimata K (1 9 9 8 ). The gene structure and promoter sequence of mouse hyaluronan synthase 1 (mHAS1). B i o c h e m . J . 330, 1223-1227.

Sugiyama Y, Shimada A, Sayo T, Sakai S, Inoue S (1 9 9 8 ). Putative hyaluronan synthase mRNA are expressed in mouse skin and TGF- beta upregulates their expression in cultured human skin cells [published erratum appears in J Invest Dermatol 1998 Jun;110(6):991]. J . I n v e s t . D e r m a t o l . 110, 116-121.

Zimmermann DR, Ruoslahti E (1 9 8 9 ). Multiple domains of the large fibroblast proteoglycan, versican. EMBO J. 8, 2975-2981.

Suzuki M, Asplund T, Yamashita H, Heldin C-H, Heldin P (1 9 9 5 ). Stimulation of hyaluronan biosynthesis by platelet-derived growth factor-BB and transforming growth factor-!1 involves activation of protein kinase C. B i o c h e m . J . 307, 817-821. Teder P, Bergh J, Heldin P (1 9 9 5 ). Functional hyaluronan receptors are expressed on squamous cell lung carcinoma cell line but not on other lung carcinoma cell lines. Cancer Res. 55, 3908-3914. Tlapak-Simmons VL, Kempner ES, Baggenstoss BA, Weigel PH (1 9 9 8 ). The active streptococcal hyaluronan synthases (HASs) contain a single HAS monomer and multiple cardiolipin molecules. J . B i o l . C h e m . 273, 26100-26109. Toole BP (1 9 9 0 ). Hyaluronan and its binding proteins, the hyaladherins. C u r r . O p i n . C e l l B i o l . 2, 839-844. Varki A (1 9 9 6 ). Does DG42 synthesize hyaluronan or chitin?: A controversy about oligosaccharides in vertebrate development. P r o c . N a t l . A c a d . S c i . 93, 4523-4525. Vrochides D, Papanikolaou V, Pertoft H, Antoniades AA, Heldin P (1 9 9 6 ). Biosynthesis and degradation of

474

Gene Therapy and Molecular Biology Vol 3, page 475

475