Colonic Drug Targeting

0 1998 OPA (Overseas Publishers Association) N.V.

Journal of Drug Targeting, 1998, Vol. 6. No. 2. pp. 129-149 Reprints available directly from the publisher Photocopying permitted by license only

Published by license under the Hanvood Academic Publishers imprint, part of T h e Gordon and Breach Publishing Group.

Printed in Malaysia.

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

Review Article

Colonic Drug Targeting RENAAT KINGET, WILLBRORD KALALA, LIESBETH VERVOORT and GUY VAN DEN MOOTER* Laboratorium voor Farmacotechnologie en Biofarmacie. Katholieke Universiteit Leuven, Campus Gasthuisberg 0 -t N , Herestraat 49, 3000 Leuven, Belgium (Received 4 February 1997; Revised 9 May 1997; In final form 14 October 1997)

Specific targeting of drugs to the colon is recognized to have several therapeutic advantages. Drugs which are destroyed by the stomach acid and/or metabolized by pancreatic enzymes are slightly affected in the colon, and sustained colonic release of drugs can be useful in the treatment of nocturnal asthma, angina and arthritis. Treatment of colonic diseases such as ulcerative colitis, colorectal cancer and Crohn’s disease is more effective with direct delivery of drugs to the affected area. Likewise, colonic delivery of vermicides and colonic diagnostic agents require smaller doses.

This article is aimed at providing insight into the design considerations and evaluation of colonic drug delivery systems. For this purpose, the anatomy and physiology of the lower gastrointestinal tract are surveyed. Furthermore, the biopharmaceutical aspects are considered in relation to drug absorption in the colon and hence various approaches to colon-specific drug delivery are discussed.

Keywords: Biodegradable polymers, Colonic targeting, pH-dependent polymers, Prodrugs, Time-dependent release

delivering drugs to the colon (colon-targeting) are recognized to have important therapeutic advantages. A number of colonic diseases could be treated-more efficiently when the drug is delivered locally, such as Crohn’s disease, ulcerative colitis, constipation, colorectal cancer, spastic colon and irritable bowel syndrome. Vermicides and diagnostic agents placed directly in the affected area would prove more effective and a smaller dose would be required. Likewise, colonic delivery

INTRODUCTION Targeting of drugs to a specific diseased organ has many advantages (Rubinstein, 1990). A smaller dose is required, which inevitably results in reduced incidence of undesirable systemic adverse reactions. Moreover, the drug is maintained in its intact form as close to the target as possible and can be supplied to the biophase only when required. Specific delivery systems or carriers for

*Corresponding author. Tel.: +32 16 345830. Fax: +32 16 345996. E-mail: guy.vandenmooter@med.kuleuven.ac.be.

129

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

130

R.KINGET et al.

would additionally be valuable when a delay in absorption is desired from a therapeutical point of view in the treatment of diseases that have peak symptoms in the early morning, such as nocturnal asthma, angina or arthritis. With the rapid advancement of biotechnology and genetic engineering resulting into availability of peptides and proteins at reasonable costs, there has been an increased interest in utilizing the colon as site for drug absorption. The potential candidates in this respect include analgesic peptides, contraceptive peptides, oral vaccines, growth hormone, insulin, interferons, erythropoietin and interleukins (Saffran et al., 1988; Mackay and Tomlinson, 1993). The only route to administer this group of therapeutic agents hitherto, is the parenteral route, which is both painful and inconvenient. Due to negligible activity of brush-border membrane peptidase activity, and much less activity of pancreatic enzymes, the colon is considered to be more suitable for delivery of peptides and proteins in comparison to the small intestine (Lee, 1991; Ikesue et al., 1991). However, a substantial amount of research still needs to be conducted in order to find out to what extent these molecules can be absorbed after oral administration.

Colonic delivery can be accomplished by oral or rectal administration. Rectal delivery forms (suppositories and enemas) are not always effective since a high variability in the distribution of these forms is observed (Wood et al., 1985). Suppositories are only effective in the rectum because of the confined spread (Jay et al., 1985), and enema solutions can only offer topical treatment to the sigmoid ahd descending colon (Hardy, 1989).

Therefore, oral administration is preferred, but for this purpose, many physiological barriers have to be overcome. Absorption or degradation of the active ingredient in the upper part of the GI tract is the major obstacle and must be circumvented for successful colonic drug delivery.

There are several ways in which colon-specific drug delivery has been attempted. Prodrugs, pHsensitive polymers, bacterial degradable polymers, hydrogels and matrices and recently multicoating

time dependent delivery systems, are an array of such attempts. These are discussed hereunder in detail.

BRIEF SURVEY OF THE ANATOMY AND PHYSIOLOGY OF THE GASTROINTESTINAL TRACT RELEVANT TO COLON TARGETING The GI tract is divided into the stomach and the small and large intestine. The large intestine extending from the ileocecal junction to the anus, is divided into three main parts. These are the colon, the rectum and the anal canal. The cecum, the colon ascendens, the colon transversale, the colon descendens and the rectosigmoid colon make up the colon. It is about 1.5 m long, the transverse colon being the longest and most mobile part (Meschan, 1975), and has an average diameter of about 6.5 cm, although it varies in diameter from approximately 9 cm in the cecum to 2 cm in the sigmoid colon (Mrsny, 1992). The colonic wall has four layers: the serosa, which is made of a single layer of squamous mesothelial cells, the muscularis externa, the submucosa and the mucosa (Samuel et al., 1968).The muscularis is composed of circular fibres which form a thin layer inwards and longitudinal fibres on the outside. The longitudinal fibres are arranged in three bands called taeniu coli and are more pronounced in the ascending and transverse colon. The configuration of the taenia and their state of contractions are responsible for the formation of the haustra, which refers to sacculations or protrusions of the bowel wall between the taenia, giving the colon its characteristic appearance. The submucosa is a connective tissue containing blood vessels and lymphatic vessels and is separated from the mucosa by a fine layer of muscularis mucosa. The mucosa lines the lumen of the colon and is divided into epithelium, lamina propria, and muscularis mucosa, which is made of a thin layer of circular smooth muscles and longitudinal fibres. The lamina propria provides a space in which are contained blood capillaries,

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING

lymphatic lacteals as well as macrophages, eosinophils, neutrophils, lymphocytes, plasma cells and antibodies. The presence of lymphoid tissue in this area permits an important immune function of the colon (Cooper er al., 1966). Many of the cells move through the epithelial barrier to the lumen in response to pathological conditions. The lamina propria also supports the epithelium which is a single layer of cells lining the internal surface of the mucosa. In contrast to the small intestine, the colon lacks villi. However, due to the presence of the plicae semilunares, which are crescentic folds, the intestinal surface of the colon is increased to approximately I300 cm2 (Mrsny, 1992). Within the mucosal epithelium are columnar absorptive cells, enteroendocrine cells, and goblet cells which secrete mucus, thereby facilitating the movement of faeces. Colonic secretions are stimulated by contact with faecal material.

The colon is well supplied with blood (Roslyn and Zinner, 1991). The cecum, ascending colon, hepatic flexure and proximal part of the transverse colon derive their blood from ileocolic, right colic and middle colic branches of the superior mesenteric artery. The inferior mesenteric artery supplies blood to the distal transverse colon, splenic flexure, descending colon and sigmoid via the left colic artery and branches of the sigmoid and superior haemorrhoidal vessels, while the middle and inferior haemorrhoidal arteries supply the rectum.

The colon serves four major functions: (1) Creation of suitable environment for the growth of colonic microorganisms; (2) Storage reservoir of faecal contents; (3) Expulsion of the contents of the colon at an appropriate time and (4) Absorption of potassium and water from the lumen, concentrating the faecal content, and secretion and excretion of potassium and bicarbonate. In vivo and in vitro electrophysiologicalstudies have demonstrated that the mechanism of sodium absorption is an active transport (Binder et al., 1987; Hawker et al., 1978).

Furthermore, it is influenced by mucosal cyclic adenosine monophosphate level, pH, osmolarity and compounds such as fatty acids, and bile acids.

Water is absorbed passively, while potassium and

131

bicarbonate are actively secreted by colonic epithelium. The active secretion of potassium is stimulated by mineralocorticoids (Gingell et al., 1968).

For the purpose of colonic drug delivery, there are two important physiological factors to be considered. These are pH and GI transit time.

The pH of the gastrointestinal tract is subject to both inter- and intra-subject variations. The absence or presence of food in the stomach determines the gastric pH range; this can vary from approximately 1.5 in the fasted state, to approximately 4 or 5 in the fed state (Wilson and Washington, 1989). The steepest gradient in the luminal pH of the gastrointestinal tract is encountered in the proximal part of the duodenum.

Beyond the first few centimeters, bile and pancreatic secretions mix with the duodenal content and neutralize the chyme (Phillips, 1993). Passing from the jejunum through the mid small bowel and the ileum, pH rises slightly from approximately 6.6 to 7.5; but falls to approximately 6.4 in the right colon. The mid and left colon have pH values of approximately 6.6 and 7.0, respectively (Meldrum et al., 1972; Bown et al., 1974; Evans et al., 1988).

Interspecies variability in pH is a major concern when developing and testing colon-specific delivery systems in animals and applying the information to humans. Table I gives an overview of pH of the GI tract.

The gastrointestinal transit time of food and drugs shows great variability and may be a constraint on colonic drug delivery. Factors such as age, disease and drug interactions also contribute TABLE I Average pH in the GI tract (Wilson and Washington, 1989) Location Oral cavity Oesophagus Stomach Small intestine Large intestine

PH 6.2-7.4 5.0-6.0 Fasted condition: 1.5-2.0 Fed condition: 3.0-5.0 Jejunum: 5.0-6.5 Ileum: 6.0-7.5 Right colon: 6.4 Mid colon and left colon: 6.0-7.6

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

132

R. KINGET er al.

to the variability of transit time. The arrival of an oral dosage form at the colon is determined, to a large extent, by the rate of gastric emptying. Proper appreciation of the role that gastric emptying plays in the delivery of drugs requires an understanding of gastrointestinal motor function during fasting and after meals. Between meals, mammals exhibit a characteristic cycle of gut motility. In man, a brief (2- 10 min) but intense muscle contraction occurs every 90min, beginning in the stomach and progressing to the distal ileum. This is followed by a period of resting and later, a period of intermittent milder contractions. Then the cycle begins again. These movements have been designated as phase I (quiescence), phase I1 (irregular activity) and phase I11 (intense bursts of contractile activity). After food ingestion, the cycle is inhibited and replaced by irregular activity lasting one to several hours, depending on the type of meal (Phillips, 1993). However, the stomach does not participate in all the cycles. Indeed, should a drug be ingested after the phase that involves the '0stomach, its gastric retention time ma longed. Drugs taken before meals usual1 1ut of the stomach within 1 h, but when taken atter a meal, the drug can take up to 10 h in the stomach (Hardy, 1989). In man, small particles, less than 7mm in diameter can pass out of the fed stomach regardless of its emptying time (Davis, 1989). In contrast, the transit time of the small intestine is reasonably constant, ranging between 3 and 4 h (Gruber et al., 1987) regardless of various conditions such as physical state, size of dosage form or presence of food in the stomach (Davis et al., 1986; Malagelada et al., 1984). More time may be spent at the ileocaecal junction which acts as a mechanical valve (Gruber et al., 1987; Davis, 1989). Here, small pellets which may be distributed in the small intestine tend to regroup, and will be redistributed in the ascending colon (Davis, 1989). When chyme reaches the colon, it is delayed and transformed into more concentrated faecal material. Segmentational contractions mix the colonic contents, making absorption easy (Hagikara and Griffin, 1977), while peristaltic waves propel the faecal

material in the anal direction (Connel, 1968). The most important of colonic motility are the massive contractions which result in emptying the colon (Misiewics, 1975). The motility is mediated through the myenteric plexus. The movement of faecal material in the colon is slow as compared to the movement of chyme in the small intestine. The transit time of drugs in the large intestine ranges between 20 and 30 h, although transit times of a few hours to more than two days have been reported (Hardy, 1989). Adkin et al. (1993) demonstrated that larger tablets (9- 12 mm) move faster through the proximal and mid colon than small tablets (3-6mm), but the transit time through the ileo-cecal junction was unaffected by size. Compared with tablets, pellets move faster through the ascending colon, and therefore, with respect to colonic drug absorption, pellets may be more favourable than tablets (Hardy et al., 1985).

Colonic transit time is only slightly affected by food (Khosla and Davis, 1989), and under stress, the transit time may be shorter (Enck et al., 1989).

Drugs which act on the parasympathetic or sympathetic nervous system affect the propulsive motor activity, influencing the transit time accordingly (Haeberlin and Friend, 1992). Finally, transit time can be affected by disease states which result in diarrhoea or constipation, although in most disease conditions, transit time appears to remain reasonably constant (Hardy et al., 1988; Cann et al., 1983).

DRUG ABSORPTION IN THE COLON

The colon may not be the best part for drug absorption since the colonic mucosa lacks well defined villi as found in the small intestine and this drastically reduces the absorptive surface area, despite the large diameter. Indeed, the large diameter even reduces the exposure of drug molecules to the site of absorption. Moreover, the viscosity of colonic contents is high, especially after the hepatic flexure when chyme is processed into feces which makes it even more difficult for a drug to diffuse

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING

from the lumen to the site of absorption. Further, there is reduced permeability to polar compounds due to tight junctions between cells (Madara and Dharmsathaphorn, 1985).

Many factors can impede colonic drug absorption and were reviewed by Mrsny (1992) in an excellent paper on colonic drug absorption: (1) Specific and non-specific drug binding can occur in the lumen when drugs interact with dietary components or products released from bacteria.

This in turn, might result in facilitated enzymatic or environmental degradation by increasing the time that the drug remains in the colon. (2) Interaction between the negatively charged mucus layer and drug molecules can result in either drug-mucus binding or drug-mucus repulsion. Penicillins, cephalosporins and aminoglycosides are examples of drugs that bind to mucus (Niibuchi et af., 1986).

(3) At the epithelium, the lipid bilayer of the individual colonocytes, and the occluding junctional complex between these cells, provide a physical barrier to the absorption of drugs.

Although significantly less than observed in the small intestine, there are enzymatic activities associated with the colonocytes. Since carrier mediated uptake of drugs across the epithelial cell barrier is not extensive, lipid solubility, the ability of a drug molecule to partition between an aqueous and a lipid environment, the degree of ionization, and the pH at the site of absorption are very important factors. The unstirred water layer between mucus layer and epithelial surface area can also produce a diffusional barrier for the absorption of drugs especially the very lipophilic ones (Rahman et af., 1986). Table I1 lists factors that affect absorption of drugs from the colon.

TABLE I1 Factors affecting drug absorption from the colon ( I ) Physical characteristics of drug (pKa, degree of ionization) (2) Colonic residence time as dictated by GI tract motility (3) Degradation by bacterial enzymes and by-products (4) Selective and non-selective binding to mucous (5) Local physiological action of drug (6) Disease state (7) Use of chemical absorption enhancers, enzyme inhibitors or bioadhesives

133

The positive aspect for drug absorption is the prolonged residence time in the colon. The residence time depends on motility of the colon which in turn is affected by stress, disease state and drugs.

Drugs which produce excessive muscular activity of the colon could result in rapid elimination of drugs, and conversely, the residence time is prolonged for drugs which inhibit muscular activity.

In spite of the unfavourable conditions for absorption, Fara (1989) concluded that a large variety of drugs are well absorbed from the colon.

These include theophylline, ibuprofen, nifedipine, metoprolol, diclofenac, pseudoephedrine, brompheniramine, isosorbide dinitrate, and oxprenolol.

Oral delivery of low molecular weight peptides and peptide-like drugs is often claimed to be more efficient if these molecules could be targeted to the large intestine. A lot of research remains to be conducted to find out to what extent these molecules can be absorbed after oral administration. Kim et af. (1994) studied the absorption of two ACE inhibitors, CGS 16617 and the prodrug benazepril to determine the potential for colonic delivery of small peptides. The uptake rate of CGS16617 and benazepril was respectively 3.5 and 2.5 times higher from the jejunum than from colonic rings. The uptake of both drugs from colonic rings was purely passive. Langguth et af. (1994) showed that metkephamid, a pentapeptide, was absorbed from the large intestine. The peptide concentration in the portal vein increased over-proportionally with increasing perfusate concentration, suggesting a saturable transport or metabolism. In addition the peptide was found to be relatively stable in pig cecal contents, the half-life being 14.9h. Lee (1992) reported that the upper GI tract was the preferential site of absorption for octreotide, whereas the duodenum and ileocecal region was the preferential site for desmopressin. Also cyclosporin is preferentially absorbed in the duodenal region, but tetragastrin in both the upper GI tract as well as the rectum. It is far beyond the scope of this article to review peptide and protein absorption, but it must be clear that there is evidence that peptides can be absorbed after oral administration, but the

R. KINGET et al.

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

134

region where absorption occurs is variable.

Whether or not a peptide drug will benefit from colon targeting should be evaluated in a case by case study.

The permeability of the epithelium to drugs can be modified through the use of chemical enhancers, which are compounds that promote absorption. Intestinal absorption of polar drugs has been reported to be promoted by several compounds with widely differing chemical structures: chelating agents, non-steroidal anti-inflammatory drugs (NSAIDs), surfactants and mixed micelles, fatty acids and other substances (Mrsny, 1992; Muranishi et al., 1979; Taniguchi et al., 1980; Murakami et al., 1986; Sakai et al., 1986; Fix, 1987; Muranishi, 1987; Tomita et al., 1988).

Table I11 lists the compounds that posses colonicabsorption enhancing activity. Although many penetration enhancers have been found to improve oral bioavailability, a lot of them suffer from nonspecificity in action, propensity to elicit local irritation, and irreversibility in action (Lee et al., 1991). A succesfull approach will be the result of the balance between the efficiency of absorption and the effect of the enhancing agents on cells and tissues. The interested reader is referred to excellent reviews on this timely topic (Muranishi, 1990; Fix, 1996).

The absorption of drugs from the colon can be studied by a variety of techniques. Pharmacokinetic data can be used to indirectly evaluate the TABLE I11 Chemical enhancers of colonic drug absorption Group Chelating agents Non-steroidal anti-inflammatory agents (NSAIDS) Surfactants Bile salts Fatty acids Mixed micelles

Example

absorption (Lee et al., 1991), while the direct methods include colonoscopy and intubation (Gleiter et al., 1985) and the use of a high frequency capsule (Macleod and Tozer, 1992). Gamma scintigraphy (Davis et al., 1992) is nowadays the preferred method to study GI transit behaviour.

COLONIC FLORA The GI tract has a variety of microorganisms that contribute to its physiology and functions and participate in the metabolism of ingested material (Haeberlin and Friend, 1992). The upper part (the stomach, duodenum and proximal ileum) only contains a small number of bacteria, mainly Grampositive facultative bacteria (Gorbach, 197 1 ; Simon and Gorbach, 1986). Whereas the stomach and the proximal small bowel have less than lo4 colony-forming units/ml (CFUlml) (Simon and Gorbach, 1986), the colon contains 10'1-10'2 CFU/ml (Moore and Holdeman, 1975), mostly anaerobic bacteria. Bacteroides, Bfidobacterium and Eubacterium are the predominant species. Also present are the anaerobic Gram-positive cocci as well as Clostridium Enterococci, and various species of Enterobacteriaceae (Hill and Drasar, 1975) as depicted in Table IV. The growth and overgrowth of the intestinal microflora is regulated by gastric acid, peristaltic TABLE IV Bacterial flora of the lower intestine in man (Hill and Drasar, 1975) Species

EDTA; Citric acid Indomethacin; Salicylates

Polyoxyethylenelauryl ether; Saponins Taurocholate;Glycocholate Sodium caprate; Sodium caprilate; Sodium laurate; Sodium oleate Monoolein taurocholate;Oleic acid taurocholate;Oleic acid; Polyoxyethylene hydrogenated caster

Enterobacteria Enterococci Clostridia Lactobacilli Bacteroides Gram + ve non-sporing anaerobes (Eubacteria and bifidobacteria) Number of samples studied

Intestinal material (Mean log 10 viable count/gram) Terminal ileum

Caecum

Faeces

3.3 2.2 <2 <2

5.8

6.2 3.6 3.0 6.4 7.8 8.4

7.4 5.6 5.4 6.5 9.8 10.0

6

2

10

5.7

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING

activity and microbial interaction including bacterial metabolic byproducts. Administration of antibiotics can adversely affect the balance of intestinal flora (Edlund et al., 1987; Lindsey et al., 1990; Nord and Heimdahl, 1986). Furthermore, the metabolic activity of the intestinal microflora can be affected by disease and age (Rowland, 1988).

Hydrolytical and redox reactions are the predominant metabolic conversions triggered by the intestinal microflora. The main hydrolytic enzymes produced by the intestinal bacteria are P-glucuronidase, P-xylosidase, a-L-arabinosidase and P-galactosidase, while the reductive enzymes include nitroreductase, azoreductase, deaminase and urea dehydroxylase (Scheline, 1973). Although significantly less important than in the small intestine, proteolysis can also occur in the colon, and this must be kept in mind when targeting peptides and proteins to the large intestine. In a number of studies, the group of Macfarlane (Macfarlane et al., 1986; 1988; 1989; Gibson et al., 1989) reported on the proteolytic effect of colonic bacteria. Not only was it demonstrated that the proteolytical activity was significantly higher in the small intestine than in faeces, but there was also a qualitative difference. Indeed, bovine serum albumin could only be degraded by the faecal protease activity, but not by that of the small intestine. More recently, they reported on the possibility that bacterial fermentation of proteins in humans could account for approximately 17% of the short chain fatty acids in the cecum and for 38% in the sigmoid and rectum (Macfarlane et al., 1992). Ingested compounds can thus be activated upon by these enzyme systems, and this is the rationale of some approaches to colonic drug delivery as will be discussed in following sections.

COLON-SPECIFIC DRUG DELIVERY SYSTEMS Several approaches have been studied for the purpose of delivering drugs specifically to the colon. These are discussed hereunder in detail.

135

Prodrugs Prodrugs are usually designed to circumvent chemical, physical or physiological problems, and are therapeutic agents that have to undergo biotransformation before exerting a pharmacological action (Stella, 1975).Prodrug-based systems have been extensively exploited in targeting drugs to the colon. Once in the colon, the drugs are acted upon by enzymes produced by colonic resident bacteria to release the active moiety. Sulphasalazine (SAS) which is used in the treatment of ulcerative colitis, Crohnâ&#x20AC;&#x2122;s disease, and rheumatoid artritis, is a well known colon-specific prodrug. It consists of 5-aminosalicylic acid (5-ASA) linked via an azo bond to sulphapyridine (SP). When orally administered, approximately 12% is absorbed in the small intestine, but the main part reaches the colon intact, where bacterial azoreductase cleaves the azo bond, thereby releasing 5-ASA and SP. It has been demonstrated that 5-ASA is the active moiety while sulphapyridine is only a carrier which might be responsible for the side effects (Azad Khan et al., 1977), and therefore, Olsalazine (Dipenturn@) was developed for the delivery of 5ASA to the colon avoiding SP. It consists of two molecules of 5-ASA linked by an azo bond (Van Hozegard, 1988). It was demonstrated that olsalazine was as effective as 5-ASA in the treatment of ulcerative colitis, but it did not show a proportionally greater effect, suggesting a maximum effective delivery level of 5-ASA (Willoughby et al., 1988).

Other approaches to deliver 5-ASA to the colon resulted in the development of ipsalazide and balsalazide, where 5-ASA is linked to 4-aminobenzoyl glycine and 4-aminobenzoyl P-alanine, respectively (Chan et al., 1983). However, only balsalazide has been developed further. Fig. 1 shows the chemical structures of sulphasalazine, olsalazine and 5-ASA. 5-ASA has also been azo-linked with polymeric materials (Brown et al., 1983). A comparable total release of 5-ASA and metabolites in rats has been reported for SAS and the polymeric prodrug that consists of polysulfonamidoethylene as carrier

R. KINGET er al.

136

COOH

Ill

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

FIGURE 1 Chemical structures of sulphasalazine (I), Olsalazine (11) and 5-ASA (111).

molecule (Polyasa@), but the latter was more effective in reducing inflammation in the guinea pig ulcerative colitis model. Schacht and co-workers (1996) demonstrated that the release of 5-ASA from polymeric prodrugs was dependent upon the structure of the polymeric backbone. When 5-ASA was coupled with glycine or aminocaproic acid as a spacer group to either poly( 1-vinyl-2-pyrrolidoneco-maleic anhydride), poly(N-(2-hydroxyethyl))DL-aspartamide, or dextran, they found that the release of 5-ASA from SAS was initially faster than from the polymeric prodrugs, but after incubation of 5h in a human colonic fermentation model, release of the parent drug from the dextran derivative was comparable with SAS. The release from the other polymers was considerably slower, and it was suggested that bacterial degradation of dextran contributed to the higher drug release rate.

To compensate for a slow release at the target site, the approach of bioadhesive polymeric prodrugs has been proposed (Kopeckova and Kopecek, 1990; Rihova et al., 1992). 5-ASA polymeric prodrugs containing fucosylamine were prepared (Fig. 2), since it was suggested that glucose- and fucose-specific adhesion of bacteria to colonic cells of guinea pigs resides in the host cells. The rate of azo bond cleavage was comparable to that of low molecular weight compounds. Addition of the redox mediator benzylviologen increased the rate of reduction and of 5-ASA release, which was found to be zero order. However, it was found later that hydrophobic binding of the aromatic side chains in the polymeric backbone to the colonic mucosa may have a greater influence than the

FIGURE 2 Polymeric prodrug of 5-ASA (Kopeckova and Kopecek, 1990; Rihova er al., 1992).

specific recognition of the fucosylamine groups (Kopeckova et al., 1994). The setback of coupling 5-ASA to polymeric materials is the large amount of drug that is needed to be taken orally. The dose of 5-ASA ranges from 0.5 to 3 g daily but since the drug can only constitute approximately 20% of the total weight of the prodrug, the total amount would not be acceptable.

Another approach that has been exploited involves prominent bacterial enzymes such as glycosidases and polysaccharidases. Friend and Chang (1984; 1985) developed prodrugs by coupling a hydrophilizing promoiety, glucose or galactose, to steroids such as dexamethasone, prednisolone, hydrocortisone and fludrocortisone via a P-glycosidic bond. When taken orally, the prodrugs being polar, undergo minimal absorption in the small intestine, in the colon however, the bacterial glycosidases cleave the polar moiety releasing the steroid. Such selective delivery of corticosteroids is highly desirable in the treatment of disorders of the large intestine, since side effects can be minimized due to decreased absorption of polar compounds in the small intestine as well as dose reduction. It was demonstrated that

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING

dexamethasone and dexamethasone-P-glucoside in a concentration of 1.3pmol/kg and 1.3 or 0.65 pmol/kg, respectively, significantly reduced the total number of ulcers in the guinea pig. While there was no significant difference between drug and prodrug treatment, the results indicated that a lower dose of dexamethasone, given as a prodrug, was equally effective to the higher dose of the parent drug (Friend and Tozer, 1992). This approach has been further extended with the development of glycoside prodrugs of budesonide and menthol (Friend, 1995), and with dexamethasonepoly(L-aspartate) (Leopold and Friend, 1994).

A class of macromolecules that has been the focus of increasing interest is one that includes dextran prodrugs. The first attempt was carried out

by Harboe and co-workers (1989) who conjugated naproxen to dextran by an ester linkage. The release of naproxen was up to 17 times higher in the cecum and in colon homogenates of the pig than in control medium or homogenates of the small intestine. Furthermore, MacLeod and coworkers (1994) developed dextran prodrugs of methylprednisolone and dexamethasone. In both cases the results were promising, with most of the drugs being released in the contents of the large intestine, while only minor chemical hydrolysis occurred in the upper G I tract.

Recent developments involve the utilization of cyclodextrins (CD), which are known to be absorbed only in minor quantities from the small intestine, whereas they are fermented by colonic bacteria. In a recent paper, Hirayama et al. (1996) described the use of biphenyl acetic acid conjugates of p-CD. The ester conjugate released the drug preferentially when incubated with rat cecal contents and almost no release was observed on incubation with contents of the stomach or small intestine.

These promising results can make CD-based colonic delivery systems an area for further research.

intestine. For the purpose of targeting drugs to the colon, it is plausible therefore to coat tablets, capsules or pellets with a pH-dependent polymer which is insoluble at low pH but soluble at neutral or alkaline pH. This would thus release the drug in the distal part of the small intestine or in the colon. The problems with this strategy arise from the fact that the intestinal pH is not predictably stable, being influenced by diet, disease and presence of fatty acids, carbon dioxide and other fermentation products (Rubinstein, 1990). Moreover, the intra- and inter-individual difference in GI tract pH is considerable and is a major drawback in successful, i.e. reproducible drug delivery to the large intestine.

Widely used polymers include methacrylic resins, commercially available as Eudragit@(Rohm Pharma; Weiterstadt, Germany), and were developed by Lehmann (1975). Different types of Eudragit@, whether water insoluble or water soluble are used for colon targeting (Fig. 3).

Whereas Eudragit L is soluble at pH 6 or more, Eudragit S dissolves at pH 7 or more due to a higher amount of esterified groups in relation to carboxylic groups. The former is used for enteric coating while the latter is used for colon targeting.

However, Eudragit S showed poor site specificity when its role in colonic drug targeting was investigated (Ashford et al., 1992; 1993a). The solubility and rate of dissolution of Eudragit@ has been investigated in comparison with other coatings used for gastro-resistance (Spitael and Kinget, 1977a).

R,

= CH,;

R 2 = CH,.

pH-Sensitive Polymers The pH of the GI tract is low in the stomach but increases when passing to the small and large

137

H CH3CH2-

R, = -COOH (Eudragit@

L and S)

-COOCH2CH28(CH3),CP(EudragitB

RL and RS)

FIGURE 3 Chemical structures of Eudragit@'.

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

138

R.KINGET et al.

It was clear that the pH at which dissolution of the polymer starts only has a relative value. More interesting is the dissolution rate. Polymers with non-esterified phthalic acid groups dissolve much faster and at a lower pH as those with acrylic or methacrylic acid groups. The dissolution rate of Eudragit@ remains rather low. The presence of plasticizer (Spitael and Kinget, 1979) and the nature of the salts (Spitael and Kinget, 1977b; Spitael et al., 1980) in the dissolution medium affected the dissolution rate.

Despite the relative unsteadiness of the site where disintegration starts, Eudragit S and Eudragit L have been widely used to deliver mesalazine (trade name of 5-ASA), a drug of proven value in the treatment of inflammatory bowel disease. It is unstable in the stomach and absorbed in the small intestine, and must be protected against conditions encountered in the upper GI tract. At present, mesalazine is commercially available as an oral dosage form, coated with Eudragit@L 100 (tradenames Claversal@,Mesazal@,and Colitofalk@),or Eudragit@S (tradename Asacol@).

Apart from mesalazine, Eudragit has been utilized to deliver enteral targeted insulin (Touitou and Rubinstein, 1986) and prednisolone (Thomas et af., 1985).

A recent development in this field has been the partially methylation of the free carboxylic groups of Eudragit S. As a result, it was found by Peeters and Kinget (1993) that the modified product dissolves in water at a slightly higher pH value than the original polymer. In earlier in vivo scintigraphic studies with human volunteers the effectiveness of this product as a colon-specific coating material was established (Peeters, 1990).

Time-dependent Formulations

The pH-dependence phenomenon has been extended to include a time factor. Time-based positioning of a drug in the colon is designed in such a way as to resist the acidic environment of the stomach and to undergo a silent phase of a predetermined time duration, after which release of the

drug takes place. The silent phase in this case is the transit time from the mouth to the terminal ileum.

The first formulation of this type was Pulsincap@(MacNeil and Stevens, 1990). It consists of a capsule half of which is enteric coated and the other half is non-disintegrating. The enteric coat dissolves on entering the small intestine revealing a hydrogel plug, which is made up of a crosslinked copolymer of polyethylene oxide and polyurethane, stoppering the non-disintegrating part, which then swells. The swelling is pH-independent, and after a predetermined lag time, which is goverened by the length of the hydrogel plug, it is swollen so much that it is ejected from the bottom half of the capsule thereby releasing the drug. Fig. 4 depicts the different stages in drug release from Pulsincap. Wilding et af. (1992) studied the oral absorption of captopril in human volunteers. The mean in vivo pulse time was 327min (range 246389 min), which corresponded well with the in vitro pulse time of 5h, indicating that the device performed as expected. In six out of eight volunteers, the plug was not ejected before entering the colon. Due to intestinal instability of the peptide, measurable blood concentrations were only found in three volunteers. Imaging studies by Binns et al.

(1994) confirmed the in vivo performance of the system. In six volunteers, it was shown that the device was positioned in the colon ascendens at the time of plug ejection. The mean value of the pulse time was 230 min (range 275-208 min).

Comparable to the Pulsincap system is the design of an ethylcellulose capsule (Niwa et al., 1995).

The capsule is composed of four parts: a drug container, low substituted hydroxypropyl cellulose and a capsule body and cap made up of ethylcellulose. As water penetrates through micropores, which are made at the bottom of the capsule, the polymer swells, and forces the ethylcellulose cap to disintegrate, thereby releasing the drug. The release time of the drug appeared to be ruled by the thickness of the cap. A fair correlation was found between thickness of the cap and the in vitro release of fluorescein as well as with the time of maximal plasma fluorescein concentration in beagle dogs.

COLONIC DRUG TARGETING

ENTERIC-COATED PULSINCAP ORAL ADMINISTRATION *la

uata

h#lUbl.

Drug

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

farmulotlon

STOMACH EMPTYING

INTESTINAL FLUID 0 0 6

FIGURE 4 Schematic representation of the Pulsincap@ system.

Recently, Gazzaniga and co-workers (1994; 1995) developed an oral dosage form consisting of a core coated with two polymeric layers. The outer layer dissolves at pH > 5 and serves as enteric coating. The inner layer consists of hydroxypropylmethylcellulose (HPMC) and acts as a retarding agent to give a lag time in order for a drug to be released at a predetermined time. The thickness of the inner layer dictates how long it takes before the drug is released. Preliminary in vivo experiments in rats indicated that all the tested systems start

139

releasing in the colon between 5 and 10h, thus confirming the expected release behaviour. Pozzi and co-workers (1994) developed a pulsed system known as Time-Clock@-System. The system is made up of a solid dosage form, coated with a hydrophobic surfactant layer to which a hydrosoluble polymer is added to improve adhesion to the core. The outer layer redisperses in the aqueous environment in a time proportional to the thickness of the film. The core is then available for dispersion. One of the advantages of this system is the fact that it can be manufactured using common technology with excipients known to the pharmaceutical industry. In a study with human volunteers, it was shown that the lag time was independent of gastric residence time, and hydrophobic film redispersion did not appear to be influenced by the presence of intestinal digestive enzymes, nor by mechanical action of the stomach.

Furthermore, pharmacokinetic data indicated that the absorption of salbutamol from the Time-Clock system concurred with previous reported data (Goldstein et al., 1987), indicating that the absorption of the drug was not influenced by the in vivo behaviour of the Time-Clock system. Using the same principle as in Time-Clock@, Ueda and coworkers (1994) developed the Time-Controlled Explosion (TES) drug delivery system. It has a four layered spherical structure, consisting of a core with the drug, a swelling agent, and a water insoluble polymer membrane made up of ethylcellulose, and it is characterized by rapid drug release with a programmed lag time. When water penetrates through the polymer membrane, the swelling agent expands, leading to destruction of the membrane with subsequent drug release. An in vitro study using metoprolol as a model drug showed that the release was not influenced by the pH, but that the lag time was controlled by the thickness of the outer membrane layer. An alternative approach using a core drug tablet enclosed in a hollow cylindrical polymeric matrix, and coated at the top with a copolymer of ethylenevinyl acetate has been proposed by Vandelli et al.

(1996). Due to the water impermeable coating, the

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

140

R. KINGET et al.

drug can only be released after a lag time which is related to the wall thickness of the hollow matrix.

Release of isosorbide-5-nitrate, used as a model drug, occurred as a consequence of diffusion through the swollen polymer matrix and polymer erosion, and was related to the polymer to drug ratio and to the exposed surface area. Although this system is inventive, it can encounter difficulties to manufacture. Indeed, incorrect positioning of the hole will produce a non-uniform width of the hollow matrix, decreasing the path length of water penetration and the time needed for polymer erosion. Also, a non-homogeneous application of the coating could modify the coating rigidity, leading to non-desirable infiltration of the aqueous medium.

A remark that applies to the majority of the time-dependent systems is that only if modification in the design and appropriate technology is developed and available, large scale manufacturing is possible.

Bacterial Degradable Polymer Coatings

The colon is inhabited by a large number and variety of bacteria which inevitably secrete many enzymes. Of the many metabolic reactions known to be carried out by bacteria, the reduction of the azo bond has been extensively exploited to study colon-specific drug release. The azo reductase activity is responsible for cleaving the azo bond and is non-specific (Scheline, 1973). Polymers containing azo bonds have been used to coat drugs so as to protect them against degradation or absorption in the stomach and the small intestine.

Upon reaching the large intestine, the azo polymer coat is acted upon by bacterial enzymes and subsequently releases the drug for absorption or local action. The approach of coating a drug with a biodegradable material for colon targeting is advantageous in that one can administer a large amount of drugs, and the nature of drug release is unique and selective because it is the bacterial enzymes rather than host’s enzymes which are responsible.

Saffran and co-workers (1986; 1991) reported the release of insulin and vasopressin in the colon of rats and dogs from oral dosage forms coated with copolymers of styrene and hydroxyethylmethacrylate (HEMA), crosslinked with 4,4’-divinylazobenzene and N,N‘-bis(P-styrene sulphonyl)-4,4/-diaminoazobenzene.The coating was degraded in the colon by bacterial azoreductase, with subsequent release of the drug from the capsules. However, problems due to variability in absorption rates were encountered when administering the coated peptide drugs. The variations observed were probably due to intra- and intersubject differences in microbial degradation of the coating, which might be caused by the fact that the polymer that is applied is insufficiently hydrophilic.

An extreme large variation in blood glucose levels was also found by Cheng et al. (1994) after administration of insulin with azo polymer coated capsules to beagle dogs. The difference was also explained in terms of difference in GI transit and microbial digestion in different dogs. In a number of publications Van den Mooter and co-workers (1992; 1993; 1994; 1995) reported on the relation between hydrophylicity and biodegradable properties of copolymers of HEMA and methylmethacrylate (MMA) and terpolymers of HEMA, MMA and methacrylic acid (MA) which were synthesized in the presence of azo bifunctional compounds such as divinylazobenzene, N,N’-bis[(methacryloyloxyethyl)oxy(carbonylamino)]azobenzene, and N,N’bis(methacryloylaminoazobenzene) (Fig. 5).

The azo polymers were film forming and used to coat capsules. In vitro and in vivo release studies in rats confirmed the effectiveness of azo reductase in breaking down the azo bonds to release drugs such as theophylline and ibuprofen. However, they found that only polymers with a sufficient degree of hydrophilicity can be degraded within an acceptable period of time, but too high a degree of hydrophylicity will lead to premature drug release before the colon is reached. The spacer length of the incorporated azo agent appeared to be of minor importance. The same research group introduced azo polymers with the incorporation of

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING

141

FIGURE 5 Chemical structures of the azo polymers, developed by Van den Mooter er al. (1992; 1993; 1994; 1995).

(I) = Divinylazobenzene,(11) = N,Nbis(methacryloylamino-azobenzene),and (111) = N,N'-bis[(methacryloyloxyethyl)oxy(carbonylamino)]azobenzene.

the azo group in the main chain, and included ethoxy groups to compensate for the presence of hydrophobic C- 12 or C- 16 chains (Van den Mooter et al., 1995; Samyn et al., 1995; Kalala et al., 1996).

In vitro, the polymers were degraded to a remaining azo concentration of less than 60% on average after 48 h in rat cecal content, and the release of ibuprofen from coated capsules showed a significantly higher rate in the reductive medium than in a control medium. However, although these polymers contained a higher azo concentration, the degradation rate was much lower than methacrylic based azo polymers, stressing the necessity to have hydrophilic, and swellable polymers which allow

better penetration of azo reduction mediators into the polymer network.

It has always been taken for granted that the azo function in azo polymers is reduced in the same way as in small azo molecules such as azo dyes, and the resulting degradation products were thought to be aromatic amines. However, Kimura and coworkers (1992) found that the azo bonds in segmented polyurethanes were reduced to hydrazo intermediates after incubation with human faeces, since no decrease in the molecular weight was observed. It was then postulated that drug release from pellets coated with these azo polymers was the result of a conformational change and a breakdown

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

142

R. KINGET et al.

of the film structure. In this respect, interesting observations were done by Schacht et al. (1996) with azo containing polyamides. Dependent upon the hydrophilic/hydrophobic nature, the polymers were reduced to the hydrazo intermediate or were completely degraded into aromatic amines. Similar observations were reported by Ueda et al. (1996).

To date, there is still no foregone conclusion on toxicity of azo polymers. Although small azo molecules such as azo dyes include several potential cancerogenic and/or mutagenic agents, it has not been established if this is also the case for polymers containing azo aromatic bonds. In order to avoid possible azo toxicity related issues, some biodegradable natural substances capable of forming coatings which degrade in the colon have also been studied. Poor film forming properties or water solubility of these natural polymers are often encountered and therefore, they might be mixed with other, synthetic polymers in order to obtain a film forming mixture, or they have to be derivatized to decrease water solubility.

The ability of the microflora to degrade substrates of natural origin was already exploited by Zeitoun and Brisard (1988), who describe in a patent a coating consisting of microcrystalline cellulose (Avicel@) and ethylcellulose. Lehmann and Dreher (1991) used a suspension of natural polygalactomannans in polymethacrylate solutions to form degradable coatings. The polygalactomannans form a swellable layer around the drug core, thus delaying the release of the drug in the small bowel. They are destroyed enzymatically in the colon, and consequently the drug is released.

Polymethacrylate copolymers were used to compensate for the insufficient film forming properties of polygalactomannans.

It is generally accepted that inulin can resist degradation in the upper GI tract, but in the colon it is fermented by Bijidobacteria and Bacteroides.

Taking this into consideration, Vervoort and Kinget ( 1996), incorporated inulin in Eudragit films. When incubated for 24h in a degradation medium from faecal origin, a decrease in pH was observed, probably as the result of the bacterial

conversion of inulin to fatty acids. Also the permeability of isolated polymer membranes increased significantly after incubation in the degradation medium. After 8, 16 and 24 h of incubation, the permeability coefficient increased with a factor 6, 20 and 70 respectively. Similarly, in order to overcome manufacturing problems when using high methoxy pectin as biodegradable excipient, Wakerly et al. (1996) developed film coatings consisting of ethylcellulose and pectin. I n vitro degradation studies indicated that the release was controlled by the ratio of ethylcellulose towards pectin.

Although both galactomannan, inulin and pectin were incorporated and somehow coated with a water insoluble polymer, the polysaccharides clearly retained their susceptibility to be degraded by bacterial glycosidase. At first sight, this might look surprising since it might be expected that the physical structure of the polysaccharides is altered by the manufacturing process. A clear example of altered physical structure leading to different hydrolytical susceptibility was described for amylose. Indeed, amylose when prepared under appropriate conditions, is not only able to produce films, it was also found that the microstructure of the film is resistant to the action of pancreatic a-amylase, but is digested by the colonic flora (Allwood et al., 1989). This was later confirmed by Milojevic et al. (1996). However, incorporation of ethylcellulose was necessary to control swelling in order to prevent premature drug release through simple diffusion. I n vitro release of 5-ASA from pellets coated with a mixture of amylose-ethylcellulose (ratio 1 :4) was complete after 4 h in a colonic fermenter, while it took more than 24h to release only 20% of the drug in conditions mimicking the stomach and the small intestine.

The glycolytic activity of colonic bacteria on galactomannans was also exploited by Sarlikiotis and Bauer (1992). By synthesizing block copolymers of polyurethanes with ethylated or acetylated galactomannan segment, materials were obtained which were film forming and water insoluble. A

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING

significant change in the mechanical properties of these films was observed when they were incubated in a suspension of human feces and pig cecal content, indicating that the derivatized polysaccharides were still fermentable.

Interesting observations were made with lauroyl dextran esters (Bauer and Kesselhut, 1995). Esters based on dextran with a molecular weight of approximately 250 000 and a degree of substitution ranging from 0.11 to 0.3 were found to be film forming and stable between pH 1 and 7.4. Tablet cores with theophyllin were coated using conventional equipment with a dispersion of 4% lauroyl dextran to theoretical polymer weights of 5 , 10 and 15 mg/cm2 (Hirsch et al., 1997). Theophylline dissolution was monitored for 4 h in a buffer of pH 5.5, after which the passage to the cecum was simulated by the addition of dextranase. Almost linear dissolution was observed during the first 4 h.

The release rate was inversely proportional to the amount of ester applied on the coatings. After the addition of dextranase, the coatings were degraded, leading to complete release of the drug in less than 2 h after the addition of the enzyme. These results are very promising, not only because the degradation is fast, but also because the polymers are film forming, and since they are based on natural products which are derivatized by conventional methods using acceptable reactants, they are expected to have an acceptable toxicological profile, although this must be confirmed.

Bacterial Degradable Matrix and Hydrogel Systems

An alternative approach to the manufacture of a colon-specific dosage form involves the compression of blends of active agent, a degradable polymer, and additives to form a monolithic or multiparticulate solid dosage form; the drug is embedded in the matrix core of the degradable polymer.

Bioerodible matrix systems of cross-linked chondroitin sulphate with different level of crosslinking were developed by Rubinstein and co-workers

143

(1992a,b) for the delivery of indomethacin. When the matrix was incubated in phosphate-buffered saline (PBS) and inoculated with rat cecal content, a significant difference in release of the drug was observed after 14h, as compared with release in PBS alone. Furthermore a linear relationship was found between the degree of cross-linking of the polymer and the amount of drug released in rat cecal content. Better results were obtained when the cecal content was sonicated prior to the release experiments.

A more promising matrix system was developed by the same research group (Rubinstein and Radai, 1991; Rubinstein et al., 1993) by mixing pectic salt with indomethacin which was compressed into matrices. The difference between the indomethacin levels measured in the rat cecal medium and in the control medium was significant at each timepoint of the experiment. In a more recent study it was found that highly compressed matrices, either in the form of plain tablets or in the form of compression coated tablets, were able to retain indomethacin in simulated gastric and intestinal juice, prior to degradation in a medium containing pectinolytic enzymes (Rubinstein and Radai, 1995). A colonic drug delivery system based on pectin which has been 70% methoxylated on the acid groups has been investigated by Ashford and co-workers (1993). In vitro experiments demonstrated that this high methoxyl pectin, when applied as a compression coat, proved capable of protecting a core tablet during conditions mimicking mouth-to-colon transit, and was susceptible to enzymatic attack. In vivo gamma-scintigraphic studies confirmed the in vitro findings. As expected, pectins with a low degree of methoxylation were more susceptible to degradation, and it was assumed that the presence of calcium increased the susceptibility to enzymatic attack (Ashford et al., 1994).

A drawback of every monolithic or single unit system is that they may exhibit a delay at the ileocecal junction, leading to drug loss prior to entry in the colon. This may be avoided when using a multiparticulate dosage form which passes freely

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

144

R. KINGET et al.

through the ileocecal junction. Such a multiparticulate system based upon amidated pectin, a low methoxyl pectin in which some of the carboxylic groups are amidated, has been investigated recently (Munjeri et al., 1997). Coating of the amidated pectin beads with chitosan significantly reduced the release of sulfamethoxazole and indomethacin in simulated gastric and intestinal juice as compared with non-coated beads. The cationic polymer chitosan forms a polyelectrolyte complex around the bead, thereby altering the properties of the amidated pectin beads. In simulated colonic medium, both drugs were released within approximately 2 h.

An interesting approach was introduced by the group of Prof. Kopecek. Several types of hydrogels based on N-substituted (meth)acrylamides, N-tertbutylacryl amide and acrylic acid were synthesized in the presence of different azo crosslinking agents (crosslinking polymerization) (Brondsted and Kopecek, 1991; 1992). Due to the incorporation of acidic monomers, the hydrogels showed a pHdependent degree of swelling, which was highest in the colonic environment. In vitro and in vivo biodegradation studies in rat cecal contents and rats, respectively, showed that the degree of degradation was to a high extent related to the equilibrium degree of swelling of the hydrogels, the degradability being inversely proportional to the crosslinking density. It also appeared that hydrogels with a larger azo crosslinking agent, but having the same crosslinking density, were degraded faster. However, the enzymatic degradation of the azo crosslinks occurred over a period of a few days. Alternatively, azo polymeric hydrogels were prepared by crosslinking of polymeric precursors, or by polymer-polymer reactions (Kopecek et al., 1992; Yeh et al., 1994; 1995; Ghandehari et al., 1997). By manipulation of the stoichiometry of the monomers in the feed, a variety of copolymer compositions was achieved, and although biodegradation of the gels in vivo was consistent with the in vitro data, the degradation rate was faster in vivo. A comparative study was performed to differentiate between the degra-

dation rate of the azo functionality present in methyl orange, a linear, soluble azo polymer, and a hydrogel that was synthesized from the linear azo polymer. The azo dye was completely reduced in 1h, while the soluble polymer was reduced up to 70%. The degradation rates were 23.76 and 17.55pmol/g*h for the azo dye and the polymer, respectively. The difference was attributed to reduced azo accessibility due to the presence of bulky polymer chains, and to inter- and intramolecular aggregates formed by the hydrophobic side chains. As expected, the reduction rate decreased dramatically for the hydrogel; the degradation rate was 0.14 pmol/g*h, or 125 times lower than for the soluble azo polymer. It was also found that hydrogels prepared by crosslinking polymerization, but having the same polymer composition and crosslink structure followed predominantly a bulk-degradation like process, while hydrogels prepared by crosslinking of polymeric precursors, or by a polymer-polymer reaction followed predominantly a surface erosion process when having a low degree of crosslinking, and a bulk-degradation like process when the degree of crosslinking increased.

Instead of incorporating the azo aromatic functionality in the crosslinks, Shanta et al. (1995) prepared hydrogels by copolymerisation of 2hydroxyethyl methacrylate with 4-methacryloyloxyazobenzene. In vitro release of 5-fluorouracil from the hydrogels was faster in human faecal medium compared to the release in simulated gastric and intestinal fluids, and followed zero order release up to 4h. Approximately 80% of the drug was released in that period. Although the degradable azo bonds were not located in crosslinks, cleavage of them seem to cause loosening of the polymeric matrix, leading to release of 5-fluorouracil. A serious drawback of this system is that microbial degradation of the hydrogel will result in the release of aniline.

Hydrogels based on natural products are considered to be more acceptable with respect to toxicity, and are therefore more preferable than azo polymeric materials. However, care must be

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING

taken when derivatization is carried out, since chemical modifications of natural, biodegradable products might possibly lead to products which are less prone to degradation. Indeed, altering chemical structure or configuration might lead to sites which are no longer recognizable to the enzyme system responsible for biodegradation. It is also preferable that the enzymes can penetrate the polymer network, leading to bulk degradation instead of surface degradation, and consequently a faster release of the entrapped drug. Hovgaard and Brondsted (1995) formed hydrogels of dextran using diisocyanate as the crosslinking agent. The gels were still biodegradable, but from a study of Simonsen et al. (1995) it was shown that, although the gels were fermented completely to short chain fatty acids in a human colonic fermentation model, it took more than 24h before the gels started to degrade. Clearly, such delivery systems are only capable to release a drug in the dista! part of the colon, where the conditions for absorption are dramatically reduced. On the other hand, it was found that the release of hydrocortisone from dextran hydrogels was successfully triggered after the addition of dextranase; in less than 2 h 100% of the hydrophobic drug was released. It must, however, be reminded that the experimental conditions did not completely reflect the in vivo situation, since the experiments were performed at pH 5.4.

Recently, Rubinstein and Gliko-Kabir (1995) reported on the biodegradable properties of guar gum which was crosslinked with borax. In vitro degradation of the modified guar gum showed that the rate of degradation by galactomannase form Aspergillus niger was the same as for not crosslinked guar gum. The molecular weight decreased with more than a factor 10 after only 45min of incubation. Taking into consideration the time required for hydrogel degradation, this system seems to be able to release drugs in the proximal colon. In order to confirm these promising findings, the experiments should be performed with bacterial enzymes, or a comparison should be made between the activity of the fungal enzyme and the enzymes present in the human microflora.

145

Although it seems to be possible to design crosslinked systems with reduced aqueous solubility, but with unaltered biodegradable properties, the matrices and hydrogel systems have one major disadvantage, since only a limited amount of drug can be incorporated. Thus when a large amount of drug is required at the target site, as in the case of 5-ASA, this may not be the best vehicle.

CONCLUSION During the past fifteen years, a considerable amount of work has been carried out in order to design several delivery systems for targeting drugs specifically to the large intestine. The large interand intra-subject variation in GI pH, makes the approach of delivery systems based upon pH dependent polymers less suitable. Nevertheless, marketed products of 5-ASA rely on the use of Eudragit. Despite the fairly constant small intestinal transit time, the possible â&#x20AC;&#x153;hold upâ&#x20AC;? time at the ileocecal valve is a serious drawback in the use of delivery systems which release drugs after a predetermined lag time. Undoubtly, the preferred systems are those which rely on conditions which are only encountered in the colon, since these systems will give true site-specific delivery. In this respect, systems which can be degraded by colonic bacteria are very attractive and promising. For the moment, no conclusive data are available to assign the same toxicological profile to azo polymeric delivery systems as to azo dyes, which are known to be potential cancerogenic/mutagenic agents.

Although every new excipient has to undergo a whole series of safety assessement tests, delivery systems based on natural polymers such as dextran, pectin, galactomannan, inulin, etc. are more favorable with respect to safety. However, the disadvantage of most naturally occurring polymers is the inherent water solubility, which has to be decreased by chemical derivatization, but which can lead to a decreased biodegradability.

Local therapy of pathologies of the large intestine and reduced drug availability due to

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

146

R. KINGET er al.

degradation by digestive or mucosal enzymes, can benefit from colon delivery. Still, a substantial amount of research remains to be conducted in order to find out to what extent drugs, and more specifically peptides and proteins can be absorbed, since it has often been claimed that colon targeting is one of the best alternatives for the oral administration of peptides and proteins.

A major problem in comparing different delivery systems is the fact that degradation studies are performed under different conditions. In this respect, the use of the simulated human intestinal microbial system, developed by Molly et al. (1993), might offer the possibility to standardize drug release studies and will permit to correlate the results with in vivo situations.

References Adkin, D.A., Davis, S.S., Sparow, R.A. and Wilding, I.R. (1993) Colonic transit of different sized tablets in healthy subjects. J. Controlled Rel., 23, 147-153.

Alwood, M.C., Archer, D.B. and Ring, S.G. (1989) Delayed release formulations. Eur. Pat. Appl. 0343993Al. Ashford, M., Fell, J.T., Attwood, D. and Woodhead, P.J. (1992) An in vitro investigation into the suitability of pH-dependent polymer for colonic targeting. Int. J. Pharm., 94, 241-245.

Ashford, M., Fell, J.T., Attwood, D., Sharma, H. and Woodhead, P.J. (1993a) An in vivo investigation into the suitability of pH-dependent polymers for colon targeting. Inr. J. Pharm., 95, 193-199.

Ashford, M., Fell, J.J., Attwood, D., Sharma, H.L. and Woodhead, P.J. (1993b) An evaluation of pectin as carrier for drug targeting to the colon. J. Controlled Rel., 26, 213-220.

Ashford, M., Fell, J., Attwood, D., Sharma, H. and Woodhead, P. (1994) Studies on pectin formulations for colonic drug delivery. J. Controlled Rel., 30, 225-232.

Azad Khan, A.K., Pins, J. and Truelove, S.C. (1977) An experiment to determine the active therapeutic moiety of sulphasalazine. The Lancer, 2,895-895.

Bauer, K.H. and Kesselhut, J.F. (1995) Novel pharmaceutical excipients for colon targeting. S.T.P. Pharma Sci., 5, 54-59.

Binder, H.J., Foster, E.S., Budinger, M.E. and Hayslett, J.P. (1987) Mechanism of electroneural sodium chloride absorption in distal colon of the rat. Gastroenterology, 93,449-455.

Binns, J.S., Stevens, H.N.E., Bakhshaee, M. and Wilson, C.G. (1994) Colon targeted release using the Pulsincap delivery system. Proc. Intern. Symp. Control. Rel. Bioact. Muter., 21, 260-261.

Bown, R.L., Gibson, J.A., Sladen, G.E., Hicks, B. and Dawson, A.M. (1974) Effects of lactulose and other laxatives on ileal and colonic pH as measured by a radiotelemetry device. Gur, 15,999-1004.

Brondsted, H. and Kopecek, J. (1991) Hydrogels for site-specific oral drug delivery: Synthesis and characterization. Biomaterials, 12, 584-592.

Brondsted, H. and Kopecek, J. (1992) Hydrogels for site-specific drug delivery to the colon: In vitro and in vivo degradation.

Pharm. Res., 9, 1540-1545.

Brondsted, H., Hovgaard, I. and Simonsen, I. (1995) Dextran hydrogels for colon-specific drug delivery IV. Comparative release study of hydrocortisone and prednisolone phosphate.

STP Pharma Sciences, 5, 65-69.

Brown, J.P., McGarraugh, G.V., Parkinson, T.M., Wingerd, R.E. Jr. and Oderndonck, A.B. (1983) A polymeric drug for treatment of inflammatory bowel disease. J. Med. Chem., 26, 1300-1307.

Cann, P.A., Read, N.W., Brown, C., Hobson, N. and Holdsworth, (1983) Imtable bowel syndrome: Relationship of disorders in the transit of a single solid meal to symptom patterns. Gut, 24, 405-41 1.

Chan, R.P., Pope, D.J., Gilbert, A.P., Sacra, P.J., Baron, J.,H. and Lennard-Jones, J.E. (1983) Studies of two novel sulphasalazine analogs, ipsalazide and balsalazide. Dig. Dis. Sci., 28, 609-615.

Cheng, C.L., Gehrke, St. H. and Ritschel, W.A. (1994) Development of an azopolymer based colonic release capsule for delivering proteins/macromolecules. Meth. Find. Exp. Clin. Pharmacol., 16, 271-278.

Connel, A.M. (1968) Motor action of the large bowel. In Coda, C.F. and Heidel, W. (ed.), Handbook of Physiology, Sect 6, Vol. IV (Washington D.C.: American Physiology Society), pp. 66-92.

Cooper, M.D., Perey, D.Y., McKnealiky, M.F., Gabrielsa, A.E., Sutherland, D.E.R. and Good, R.A. (1966) A mammalian equivalent of the avian Bursa of fabricus. Lancet, l , 1388- 1391.

Davis, S.S. (1989) Assessment of gastrointestinal transit and drug absorption. In Prescott L.F. and Nimmo S.W. (ed.)

Novel Drug Delivery and its Therapeutic Application (Chichester: Wiley), pp. 89-101.

Davis, S.S., Hardy, J.G. and Fara, J.W. (1986) Transit of pharmaceutical dosage forms through the small intestine. Gut, 27,886-892.

Davis, S.S., Hardy, J.G., Newman, S.P. and Wilding, I.R. (1992) Gamma scintigraphy in the evaluation of pharmaceutical dosage forms. Eur. J. Nucl. Med., 19, 971-986.

Edlund, C., Lidbeck, A., Kager, L. and Nord, C.E. (1987) Comparative effects of enoxacin and norfloxacin on human colonic microflora. Antimicrob. Agents Chemother., 31, 1846-1848.

Enck, P., Merlin, V., Erckenbrecht, J.F. and Wienbeck, M. (1989) Stress effects on gastrointestinal transit in the rat.

Gut, 30,455-459.

Evans, D.F., Pye, G., Bramley, R., Clark, A.G., Dyson, T.J. and Hardcastle, J.D. (1988) Measurement of gastrointestinal pH profiles in normal ambulant human subjects. Gut, 29, 1035-1041.

Fara, J.W. (1989) Colonic drug absorption and metabolism. In Prescott, L.F. and Nimmo, W.S. (ed.) Novel Drug Delivery and Therapeutics (Chichester: Wiley), pp. 103-1 12.

Fix, J.A. (1987) Absorption enhancing agents for gastrointestinal system. J. Controlled Rel., 6, 151-162.

Fix, J.A. (1996) Strategies for delivery peptides utilizing absorption-enhancing agents. J. Pharm. Sci., 85, 1282- 1285.

Friend, D.R.and Chang, G.W. (1984) A colon-specific drug delivery system based on drug glycosides and glycosidases of colonic bacteria. J . Med. Chem., 27, 261-266.

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

COLONIC DRUG TARGETING Friend, D.R. and Chang, G.W. (1985) Drug glycosides: Potential prodrugs for colon-specific drug delivery. J . Med. Chem., 28, 51-57.

Friend, D.R. and Tozer, T.N. (1992) Drug glycosides in oral colon-specificdrug delivery. J . Controlled Rel., 19, 109- 120.

Friend, D.R. (1995) Glycoside prodrugs: novel pharmacotherapy for colonic diseases. S.T.P. Pharma Sci., 5, 70-76.

Gazzaniga, A,, Bussetti, C., Moro, L., Sangali, M.E. and Giordano, F. (1995) Time dependent oral delivery systems for colon targeting. S.T.P. Pharma. Sci., 5, 83-88.

Gazzaniga, A,, Iamartino, P., Maffione, G. and Sangalli, M.E. (1994) Oral delayed release system for colonic specific delivery. Int. J. Pharm., 108, 77-83.

Ghandehari, H., Kopeckova, P. and Kopecek, J. (1997) In vitro biodegradation of pH-sensitive hydrogels containing aromatic azo bonds. Biomaterials, 18, 861-872.

Gibson, S.A.W., Mcfarlan, C., Hay, S. and Macfarlane, G.T. (1989) Significance of microflora in proteolysis in the colon.

Appl. Environm. Microbiol., 55, 679-683.

Gingell, J.C., Davies, M.W. and Shields, R. (1968) Effect of synthetic gastrin-like pentapeptide upon the intestinal transport of sodium, potassium and water. Gut, 9, 111-115.

Gleiter, C.H., Antonin, K.H., Bieck, P., Godbillon, J., Schonleber, W. and Malchow, H. (1985) Colonoscopy in the investigation of drug absorption in healthy volunteers.

Gastrointest. Endosc., 31, 71-73.

Goldstein, D.A., Tan, Y.K. and Soldin, S.J. (1987) Pharmacokinetics and absolute bioavailability of salbutamol in healthy adult volunteers. Eur. J . Clin. Pharmacol., 32, 631-634.

Gorbach, S.L. (1971) Intestinal microflora, Gastroenterology, 60, I 1 10-1 129.

Gruber, P., Longer, M.A. and Robinson, J.R. (1987) Some biological issues in oral, controlled drug delivery. Adv. Drug Del. Rev., 1, 1-18.

Haeberlin, B. and Friend, D.R. (1992) Anatomy and physiology of the gastrointestinal tract. In Friend, D.R. (ed.) Oral ColonSpecipc Drug Delivery (Boca Raton: CRC Press), pp. 1-43.

Hagikara, P.F. and Griffin, W.O.J. (1977) Physiology of the colon and the rectum. Surg. Clin. North. Am., 52, 797-805.

Harboe, E.. Larsen, C., Johansen, M. and Olesen, H.P. (1989) Macromolecular prodrugs XV. Colon targeted delivery bioavailability of naproxen from orally administered dextran-naproxen prodrugs varying in molecular size in the pig.

Pharm. Res., 6, 919-923.

Hardy, J.G. (1989) Colonic transit and drug delivery. In Hardy, J.G., Davis, S.S. and Wilson, C.G. (ed.) Drug Delivery to the Gastrointestinal Tract (Chichester: Ellis Horwood), pp. 75-8 1.

Hardy. J.G., Healy, J.N.C. and Reynolds, J.R. (1987) Evaluation of an enteric coated delayed release 5-aminosalicylic acid tablet in patients with inflammatory bowel disease. Alimentary Pharmacol. Therap., 1,273-280.

Hardy, J.G., Davis, S.S., Khosla, R. and Robertson, C.S. (1988) Gastrointestinal transit of small tablets in patients with ulcerative colitis. Int. J. Pharm., 48, 79-82.

Hardy, J.G., Wilson, C.G. and Wood, E. (1985) Drug delivery to the proximal colon. J . Pharm. Pharmacol., 37,874-877.

Hawker, P.C., Mashier, K.E. and Turnberg, L.A. (1978) Mechanism of transport of sodium chloride and potassium in the human colon. Gastroenterology, 74, 1241- 1247.

Hill, M.J. and Drasar, B.S. (1975) The normal colonic bacterial flora. Gut, 16, 318-323.

Hirayama, F., Minami, K. and Uekama, K. (1996) In vitro evaluation of biphenyl acetic acid-p-cyclodextrin conjugates as colon targeting prodrugs: drug release behaviour in rat biological media. J. Pharm. Pharmacol., 48,27-31.

147

Hirsch, S., Binder, V., Kolter, K., Kesselhut, J.F. and Bauer, K.H. (1997) Lauroyldextran as a coating material for sitespecific drug delivery to the colon: in vitro dissolution of coated tablets. Proc. Int. Symp. Control. Rel. Bioact. Muter., 24, 379-380.

Hovgaard, L.and Bronsted, H. (1995) Dextran hydrogels for colon-specific drug delivery. J . Controlled Rel., 36, 159- 166.

Ikesue, K., Kopeckova, P. and Kopecek, J. (1991) Degradation of proteins by enzymes of the gastrointestinal tract. Proc.

Intern. Symp. Control. Rel. Bioact. Muter., 18, 580-581.

Jay, M., Beihn, R.M., Digenis, G.A., Deland, F.H., Caldwell, L. and Mlodozeniec, A.R. (1985) Disposition of radiolabelled suppositories in humans. J. Pharm. Pharmacol., 37,266-268.

Kalala, W., Kinget, R., Van den Mooter, G. and Samyn, C. (1996) Colonic drug targeting: in vitro release of ibuprofen from capsules coated with poly(ether-ester) azo polymers.

Znt. J . Pharm., 139, 187-195.

Khosla, R. and Davis, S.S. (1989) Gastric emptying and small and large bowel transit of non-disintegrating tablets in fasted subjects. Znt. J. Pharm., 52, 1-10.

Kim, J.S., Oberle, R.L. , Krummel, D.A., Dressman, J.B. and Fleisher, D. (1994) Absorption of ACE inhibitors from small intestine and colon. J . Pharm. Sci., 83, 1350-1356.

Kimura, Y., Makita, Y., Kumagai, T., Yamane, H., Kitao, T., Sasatani, H. and Kim, S.I. (1992) Degradation of azo containing polyurethane by the action of intestinal microflora. Polymer, 33, 5294-5299.

Kopecek, J., Kopeckova, P., Brondsted, H., Rathi, R., Rihova, B., Yeh, P.Y. and Ikesue, K. (1992) Polymers for colonspecific drug delivery. J . Controlled Rel., 19, 121 - 130.

Kopeckova, P. and Kopecek, J. (1990) Release of 5-aminoâ&#x20AC;&#x2122; salicylic acid from bioadhesive N-(2-hydroxypropyl)methacrylamide copolymers by azoreductases in vitro. Makromol.

Chem., 191, 2037-2045.

Kopeckova, P., Rathi, R., Takada, S., Rihova, B., Berenson, M.M. and Kopecek, J. (1994) Bioadhesive N-(2-hydroxypropy1)methacrylamide copolymers for colon-specific drug delivery. J. Controlled Rel., 28, 21 1-222.

Langguth, P., Breves, G., Stokli, A., Merkle, H.P. and Wolffram, S. (1994) Colonic absorption and bioavailability of the pentapeptide metkephamid in the rat. Pharm. Res., 11, 1640- 1645.

Lee, V.H.L. (1991) Changing needs in drug delivery in the era of peptide and protein drugs. In Lee, V.H.L. (ed.) Peptide and Protein Drug Delivery (New York: Marcel Dekker), pp. 1-56.

Lee, V.H.L., Doddo-Kashi, S., Grasss, G.M. and Rubas, W. (1991) Oral route of peptide and protein drug delivery. In Lee, V.H.L (ed.) Peptide and Protein Drug Delivery (New York: Marcel Dekker), pp. 691-738.

Lee, V.H.L. (1992) Oral route of peptide and protein drug delivery. BioPHARM, 5, 39-50.

Lehmann, K. (1975) Magensaftresistente und retardierende Arzneimitteluberziige aus wassrigen Acrylharzdispersionen.

Acta Pharm. Techn., 21, 255-260.

Lehmann, K.O.R. and Dreher, K.D. (1991) Methacrylategalactomannan coating for colon-specific drug delivery. Proceed. Intern. Symp. Control. Rel. Bioact. Muter., 18, 331-332.

Leopold, C.S. and Friend, D.R. (1994) Poly (L-aspartic acid) as drug carrier for colon-specific drug delivery. An in vitro assesment. Proc. Intern. Symp. Control. Rel. Bioact. Muter., 21,258-259.

Lindsey, J.T., Smith, J.W., McCluggage, S.G.J. and Nicholas, R.L. (1990) Effects of commonly used bowel preparations on the large bowel mucosa-associated and luminal microflora in the rat model. Dis. Col. Rectum, 33, 554-560.

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

148

R. KINGET et al.

Macfarlane, G.T., Cummings, J.H. and Allison, C. (1986) Protein degradation by human intestinal bacteria. J. Gen. Biol., 132, 1647-1656.

Macfarlane, G.T., Allison, C., Gibson, S.A.W. and Cummings, J.H. (1988) Contribution of the microflora to proteolysis in the human large intestine. J. Appl. Bacteriol., 64, 37-46.

Macfarlane, G.T., Hay, S. and Gibson, G.R. (1989) Influence of mucin on glycosidase, protease and arylamidase activities of human gut bacteria grown in a 3-stage continuous culture system. J. Appl. Bacteriol., 66, 407-417.

Macfarlane, G.T., Gibson, G.R., Beatty, J.H. and Cummings, J.H. (1992) Estimation of short chain fatty acid production from protein by human intestinal bacteria based on branchedchain fatty acid measurements. FEMS Microbiol. Ecol., 101, 81-88.

Mackay, M. and Tomlinson, E. (1993) Colonic delivery of therapeutic peptides and proteins. In Biek, P.R. (ed.) Colonic Absorption and Metabolism (New York: Marcel Dekker), pp. 159-176.

Macleod, A.D. and Tozer, T.N. (1992) Kinetic perspectives in colonic drug delivery. In Friend, DR. (ed.) Oral ColonSpecific Drug Delivery (Boca Raton: CRC Press), pp. 85-1 14.

Macleod, A.D., Tolentino, L. and Tozer, T.N. (1994) Glucocorticoid dextran conjugates as potential prodrugs for colonspecific delivery: Steady state pharmacokinetics in the rats.

Biopharm. Drug Disp., 15, 151-161.

MacNeil, M.E. and Stevens, H.N.E. (1990) Patent W090/09168.

Madara, J.L. and Dharmsathaphorn, K. (1985) Occluding junction structure-function relationships in cultured epithelial monolayer. J. Cell Biol., 101, 2124-2133.

Malagelada, J.R., Robertson, J.S., Brown, M.L., Remington, M., Duenes, J.A., Thomforde, G.M. and Carryer, P.W. (1984) Intestinal transit of solid and liquid components of a meal in health. Gastroenterology, 87, 1255-1263.

Meldrum, S.J., Watson, B.W., Riddle, H.C., Bown, R.L. and Sladen, G.E. (1972) pH profile of gut as measured by radiotelemetry capsule. Br. Med. J., 2, 104-106.

Meschan, I. (1975) Small intestine, colon and biliary tract. In Meschan, I. An Atlas of Basic Anatomy to Radiology (Philadelphia: W.B. Saunders Co.), pp. 843-925.

Milojevic, S., Newton, J.M., Cummings, J.H., Gibson, G.R., Botham, R.L., Ring, S.G., Stockham, M. and Allwood, M.C. (1996) Amylose as a coating for drug delivery to the colon: preparation and in vitro evaluation using 5-aminosalicylic acid pellets. J. Controlled Rel., 38, 75-84.

Misiewics, J.J. (1975) Colonic motility. Gut, 16, 311-314.

Molly, K., Vande Woestyne, M. and Verstraete, W. (1993) Development of a 5-step multi chamber reactor as a simulation of the human intestinal microbial ecosystem. Appl.

Microbiol. Biotechnol., 39, 254-258.

Moore, W.E.C. and Holdeman, L.V. (1975) Discussion of current bacteriologic investigations of the relationship between intestinal flora, diet and colon cancer. Cancer Res., 35, 3418-3420.

Mrsny, R.J. (1992) The colon as a site for drug delivery.

J. Controlled Rel., 22, 15-34.

Munjeri, O., Collet, J.H. and Fell, J.T. (1997) Hydrogel beads based on amidated pectins for colon-specific drug delivery: the role of chitosan in modifying drug release. J. Controlled Rel., 46, 273-278.

Murakami, M., Yoshikawa, H., Takada, K. and Muranishi, S. (1986) Effect of oleic acid vesicles on intestinal absorption of carboxyfluorescein in rats. Pharm. Res., 3, 35-40.

Muranishi, S. (1987) Potentials and limitations of absorption enhancers. Proceed. Int. FIP Sat. Symp., p. L18.

Muranishi, S. (1990) Absorption enhancers. Crit. Rev. Ther.

Drug Car. Sys., 7 , 1-33.

Muranishi, S., Muranishi, N. and Sezaki, K. (1979) Improvement of absolute bioavailability of normally poorly absorbed drug: Inducement of intestinal absorption of streptomycin and gentamycin by lipid-bile salt mixed micelles in rat and rabbit. h i . J. Pharm., 2, 101-109.

Niibuchi, J.J., Aramaki, Y. and Tchusiya, S. (1986) Binding of antibiotics to rat intestinal mucin. Int. J. Pharm., 30,181187.

Niwa, K., Takaya, T., Morimoto, T. and Takada, K. (1995) Preparation and evaluation of a time-controlled release capsule made of ethylcellulose for colon delivery of drugs.

J. Drug Target., 3, 83-89.

Nord, C.E. and Heimdahl, A. (1986) Impact of orally administered antimicrobial agents on human oropharyngeal and colonic microflora. J. Antimicrob. Chemother., 18, Suppl. C , 159-164.

Ovesen, L., Bendsten, F., Tage-Jensen, K., Pedersen, N.T., Gram, B.R. and Rune, S.J. (1986) Intraluminal pH in the stomach, duodenum and proximal jejenum in normal subjects and patients with exocrine pancreatic insufficiency. Gastroenterology, 90, 258-262.

Peeters, R. (1990) Studie over de ontwikkeling van een colonspecifieke artsenijvorm. Doctoral Thesis K.U. Leuven.

Peeters, R. and Kinget, R. (1993) Film-forming polymers for colonic drug delivery. I. Synthesis and physical and chemical properties of methyl derivatives of Eudragit S. Int. J. Pharm., 94, 125-134.

Phillips, S.F. (1993) Gastrointestinal physiology and its relevance to targeted drug delivery. In Current Status on Targeted Drug Delivery to the Gastrointestinal Tract (Capsugel Symposia Series), pp. 8-18.

Pozzi, F., Furlani, P., Gazzaniga, A,, Davis, S.S. and Wilding, I.R. (1994) The TlME-CLOCK@ system: a new oral dosage form for fast and complete release of drug after a predetermined lag time. J. ConfroNed Rel., 31, 99-108.

Rahman, A,, Barrowman, J.A. and Rahimtula, A. (1986) The influence of bile on the bioavailability of polynuclear aromatic hydrocarbons from rat intestines. Can. J. Physiol. ' Pharmacol., 64, 1214-1218.

Rihova, B., Rathi, R.C., Kopeckova, P. and Kopecek, J. (1992) In vitro bioadhesion of carbohydrate-containing N-(2-hydroxypropy1)methacrylamide copolymers to the G I tract of guinea pigs. Int. J. Pharm., 87, 105-1 16.

Roslyn, J.J. and Zinner, M.Y. (1991) Surgical anatomy and operative procedures. In Sabiston, D.L. (ed.) Textbook of Surgery (Philadelphia: W.B. Saunders Co.), pp. 899-901.

Rowland, I.R. (1988) Factors affecting metabolic activity of the intestinal microflora. Drug Metabol. Rev., 19, 243-261.

Rubinstein, A. (1990) Microbially controlled drug delivery to the colon. Biopharm. Drug Dispos., 11,465-475.

Rubinstein, A. and Gliko-Kabir, I. (1995) Synthesis and swelling-dependent enzymatic degradation of borax-modified guar gum for colonic delivery purposes. S.T .P. Pharma Sci., 5, 41 -46.

Rubinstein, A,, Nakar, D. and Sintov, A. (1992a) Colonic drug delivery enhanced release of indomethacin from cross-linked chondroitin matrix in rat caecal contents. Pharm. Res., 9, 276-278.

Rubinstein, A,, Nakar, D. and Sintov, A. (1992b) Chondroitin sulfate: a potential biodegradable carrier for colon-specific drug delivery. Int. J. Pharm., 84, 141- 150.

Rubinstein, A., Radai, R., Ezna, M. and Pathnack, S. (1993) In vitro evaluation of calcium pectate: A potential colonspecific drug delivery carrier. Pharm. Res., 10,258-263.

Rubinstein, A. and Radai, R. (1995) In vitro and in vivo analysis of colon specificity of calcium pectinate formulations. Eur. J. Pharm. Biopharm., 41,291-295.

COLONIC DRUG TARGETING

Touitou, E. and Rubinstein, A. (1986)Targeted enteral delivery of insulin to rats. Int. J. Pharm., 30,95-99.

Ueda, S., Ibuki, R., Kawamuza, A,, Murata, S., Takahashi, T., Kimuza, S. and Hata, T. (1994)Development of a novel drug delivery system, Time-Controlled-Explosion System (TES).

J. Drug Targeting., 2, 133-140.

Ueda, T., Yamaoka, T., Miyamoto, M., Kimura, Y., Sasatani, H. and Kim, S.I. (1996)Bacterial reduction of azo compounds Saffran,M.,Field,J.B.,Pena,J.,Jones,R.H.andOkuda,Y.(1991) as a model reaction for the degradation of azo-containing polyurethane by the action of intestinal flora. Bull. Chem. SOC. Oral insulin in diabetic dogs. J. Endocrinol., 131: 267-278.

Jpn., 69, 1139-1 142.

Sakai, K., Kutsuna, T.M., Nishino, T., Fujihara, Y. and Yata, Vandelli, M.A., Leo, E., Forni, F. and Bernabei, M.T. (1996) N. (1986) Contribution of calcium ion sequestration by In vitro evaluation of a potential colonic delivery system that polyoxyethylated nonionic surfactants to the enhanced releases drug after a controllable lag-time. Eur. J . Pharm. colonic absorption of p-aminobenzoic acid. J. Pharm. Sci., Biopharm., 43, 148-151.

75,387-390.

Van den Mooter, G., Offringa, M., Kalala, W., Samyn, C. and Samuel, P.G., Sypol, G.M., Meilman, E., Mosbach, E.H. and Kinget, R. (1995) Synthesis and evaluation of new linear Chafizadeh, M. (1968)Absorption of bile acids from large azo-polymers for colonic targeting. S. T.P. Pharma Sci., 5, bowel in man. J. Clin. Invest., 42,2070-2078.

36-40.

Samyn, C., Kalala, W., Van den Mooter, G. and Kinget, R. Van den Mooter, G., Samyn, C. and Kinget, R. (1992)Azo (1995)Synthesis and in vitro biodegradation of poly(etherpolymers for colon-specific drug delivery. Int. J. Pharm., 87, ester) azo polymers designed for colon targeting. Int. J. 31-46.

Pharm., 121,211-216.

Van den Mooter, G., Samyn, C. and Kinget, R. (1993)Azo Sarlikiotis, A.W. and Bauer, K.H. (1992) Synthese und polymers for colon-specific drug delivery. Part 11: Influence of unterschung von polyurethanen mit galactomannan-segmenthe type of azo polymer on the degradation by intestinal ten als hilfsstoffe zur freizetzung von peptid-arzneistoffen im microflora. Int. J. Pharm., 97, 133-139. dickdarm. Pharm. Ind., 54,873-886.

Van den Mooter, G., Samyn, C. and Kinget, R. (1994) The Schacht, E., Gevaert, A., Kenawy, E.R., Molly, K., Verstraete, relationship between swelling properties and enzymatic W., Adriaensens, P., Carleer, R. and Gelan, R. (1996) degradation of azo polymers designed for colon-specific drug Polymers for colon specific drug delivery. J . Controlled Rel., delivery. Pharm. Res., 11, 1737-1741.

39,327-338.

Van den Mooter, G., Samyn, C. and Kinget, R. (1995)In vivo Scheline, R.R. (1973) Metabolism of foreign compounds evaluation of a colon-specific drug delivery system: An by gastrointestinal microorganisms. Pharmacol. Rev., 25, absorption study of theophylline from capsules coated with 451-523. azo polymers in rats. Pharm. Res., 12,244-247.

Shanta, K.L., Ravishandran, P. and Rao, K.P. (1995) Azo Van Hozegard, R.A. (1988)Pharmacokinetics of olsalazine and polymeric hydrogels for colon targeted drug delivery. Eioits metabolites. Scand. J. Gastroenterol., 23 Suppl. 148,17-20. materials, 16,1313- 1318.

Vervoort, L. and Kinget, R. (1996) In vitro degradation by Simon, G.I. and Gorbach, S.L. (1986)The human intestinal colonic bacteria of inulin H P incorporated in Eudragit films. microflora. Dig. Dis. Sc., 31, 147s-162s.

Int. J. Pharm., 129, 185-190.

Simonsen, L., Hovgaard, L., Mortensen, P.B. and Brondsted, Wakerly, Z., Fell, J.T., Attwood, D. and Parkins, D. (1996) H. (1995)Dextran hydrogels for colon-specific drug delivery.

Pectin/ethylcellulose film coating formulations for colonic V. Degradation in human intestinal incubation models. Eur. drug delivery. Pharm. Res., 13, 1210-1212.

J. Pharm. Sci., 3,329-337.

Wilding, I.R., Davis, S.S., Stevens, H.N.E., Bakhshaee, M., Spitael, J. and Kinget, R. (1977a) Solubiliti et vitesse de Sparrow, R.A. and Brennan, J. (1992)Gastrointestinal transit dissolution de diffirents polymires entiriques. Lab. Pharm., and systemic absorption of captopril from a pulsed release 25,847-849. formulation. Pharm. Res., 9,654-657.

Spitael, J. and Kinget, R. (1977b) Factors affecting the Willoughby, C.P., Cowan, R.E., Gould, S.R., Machell, R.J. and dissolution rate of enteric coatings. Pharm. Ind., 39,502-505.

Stewart, J.B. (1988) Double-blind comparison of olsalazine Spitael, J. and Kinget, R. (1979) Solubility and dissolution and sulphasalazine in active ulcerative colitis. Scand. J. rate of enteric polymers. Acta Pharm. Technol., 25 (Suppl 7), Gastroenterol., 23 Suppl. 148,40-44.

163- 168.

Wilson, C.G. and Washington, N. (1989) Physiological PharSpitael, J., Kinget, R. and Naessens, K. (1980)Dissolution rate maceutics, Biological Barriers to Drug Absorption. Ellis Horof cellulose acetate pthalate and the Bronsted catalysis law. wood Ltd., Chichester, UK. Pharm. Ind., 42,846-849.

Wood, E., Wilson, C.G. and Hardy, J.G. (1985)The spreading Stella, V. (1975) Pro-drugs: an overview and definition. In of foam and solution enemas. Int. J. Pharm., 25, 191-197.

Higuchi, T. and Stella, V. (eds.) Pro-drugs as Novel Drug Yeh, P.Y., Kopeckova, P. and Kopecek, J. (1994)BiodegradDelivery Systems (Washington D.C.: ACS), pp. 1-1 15. able and pH-sensitive hydrogels: synthesis by crosslinking of Taniguchi, K., Muranishi, S. and Sezaki, H. (1980)Enhanced N,N-dimethylacrylamide copolymer precursors. J. Polymer intestinal permeability to macromolecules. 11. Improvement Sci. Part A Polymer Chem., 32, 1627-1637. of the large intestinal absorption of heparin by lipidYeh, P.Y., Berenson, M.M., Samowitz, W.S., Kopeckova, P. surfactant mixed micelles in rat. Int. J. Pharm., 4,219-228. and Kopecek, J. (1995) Site-specific drug delivery and Thomas, P., Richard, A., Rogers, L., Evans, B.K., Dew, M.J. penetration enhancement in the gastrointestinal tract. J. and Rhodes, J. (1985) Absorption of delayed-release preControlled Rel., 36, 109-124. dnisolone in ulcerative colitis and Crohnâ&#x20AC;&#x2122;s disease. J. Pharm.

Zeitoun, P. and Brisard, P. (1988)Im colon zelfallende tabletten Pharmacol., 37,757-758. und ihre herstellung, Patent Rousel-Uclaf, Puientschrgi DE Tomita, M., Shiga, M., Hayashi, M. and Awazu, S. (1988) 3046559 C2.

Enhancement of colonic drug absorption by paracellular permeation route. Pharm. Res., 5,431-435.

Saffran, M., Kumar, G.S., Savariar, C., Burnham, J.C., Williams, S. and Neckers, D.C. (1986)A new approach to the oral administration of insulin and other peptide drugs.

Science, 233, 1081-1084.

Saffran, M.,Kumar, G.S., Neckers, D.G., Williams, F., Field, J.B., Jones, R.H. and Panepinto, L. (1988)New approaches to the oral administration of peptide drugs. Pharm. Week Sci. Ed., 10,43-44.

Journal of Drug Targeting Downloaded from informahealthcare.com by University of Montreal For personal use only.

149