Mechanisms of disease pathogenesis of crohns disease and ulcerative colitis sartor et al, nature cli

REVIEW www.nature.com/clinicalpractice/gasthep

Mechanisms of Disease: pathogenesis of Crohn’s disease and ulcerative colitis R Balfour Sartor

INTRODUCTION

S U M M A RY Crohn’s disease and ulcerative colitis are idiopathic, chronic, relapsing, inflammatory conditions that are immunologically mediated. Although their exact etiologies remain uncertain, results from research in animal models, human genetics, basic science and clinical trials have provided important new insights into the pathogenesis of chronic, immunemediated, intestinal inflammation. These studies indicate that Crohn’s disease and ulcerative colitis are heterogeneous diseases characterized by various genetic abnormalities that lead to overly aggressive T-cell responses to a subset of commensal enteric bacteria. The onset and reactivation of disease are triggered by environmental factors that transiently break the mucosal barrier, stimulate immune responses or alter the balance between beneficial and pathogenic enteric bacteria. Different genetic abnormalities can lead to similar disease phenotypes; these genetic changes can be broadly characterized as causing defects in mucosal barrier function, immunoregulation or bacterial clearance. These new insights will help develop better diagnostic approaches that identify clinically important subsets of patients for whom the natural history of disease and response to treatment are predictable.

KEYWORDS animal models, bacteria, Crohn’s disease, pathogenesis, ulcerative colitis

REVIEW CRITERIA This review emphasizes studies performed since 2003, and puts these results into a clinical context. PubMed was searched in September 2005 for papers and abstracts published in English-language journals, using the following keywords, alone and in combination: “Crohn”, “colitis”, “gene”, “animal model”, “bacteria”, “mouse”, “dendritic cell”, “Paneth cell”, and “epithelial cell”. The reference list was updated in May 2006.

RB Sartor is the Margaret W and Lorimer W Midgette Distinguished Professor in the Departments of Medicine, Microbiology and Immunology at the University of North Carolina, Chapel Hill, NC, USA, where he is the Director of the Multidisciplinary Inflammatory Bowel Disease Center, Co-Director of the Center for Gastrointestinal Biology and Disease, and Associate Chief for Research of the University of North Carolina Division of Gastroenterology and Hepatology.

Correspondence University of North Carolina, Department of Medicine/Division of Gastroenterology & Hepatology, CB #7032, Room 7309 Biomolecular Building (MBRB), Chapel Hill, NC 27599-7032, USA rbs@med.unc.edu Received 21 October 2005 Accepted 19 April 2006 www.nature.com/clinicalpractice doi:10.1038/ncpgasthep0528

390 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

Ulcerative colitis and Crohn’s disease are chronic, relapsing, immunologically mediated disorders that are collectively referred to as inflammatory bowel diseases (IBD). The prevalence of IBD rapidly increased in Europe and North America in the second half of the twentieth century and is becoming more common in the rest of the world as different countries adopt a Western lifestyle.1 Such epidemiologic observations indicate that there are strong environmental influences on IBD: their influence is confirmed by the relatively low concordance rate in identical twins (~50% for Crohn’s disease, and ~10% for ulcerative colitis).2 These twin studies and the increased incidence of IBD in first-degree relatives of probands with either disease indicate that genetic factors are also integrally involved. Even with this knowledge, the etiologies of these diseases remain an enigma. In the past few years, however, work in animal models, human genetics, basic science and clinical trials, have provided new insights into the pathogenesis of these diseases.

The most widely held hypothesis on the pathogenesis of IBD is that overly aggressive acquired (T cell) immune responses to a subset of commensal enteric bacteria develop in genetically susceptible hosts, and environmental factors precipitate the onset or reactivation of disease. This complex theory involves four separate components that must intersect in multiple ways for disease to become clinically apparent (Figure 1).

A convenient approach to understanding the pathogenesis of IBD considers the possible mechanisms by which ulcerative colitis and Crohn’s disease might occur (Box 1). This approach addresses the possibility of a specific infectious etiology and the various ways that commensal bacteria can induce chronic, immune-mediated inflammation. In this review, I discuss the evidence underlying the disparate theories and attempt to reconcile them into a coherent hypothesis based on experimental data obtained in the past few years.

JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

GENETICS

Advances have occurred in understanding the genetics of human IBD, from studies based on single nucleotide polymorphism (SNP) and candidate gene approaches, and from studies in mouse experimental colitis that used transgenic and deletion (knockout) techniques.3,4 Work in these two independent systems has produced partially overlapping results, with several genes implicated in both IBD and experimental colitis (Table 1). The common threads are that the implicated genes regulate several important biologic functions, including immunoregulation, mucosal barrier integrity and microbial clearance and/or homeostasis.

The CARD15 gene

The first gene to be associated with Crohn’s disease was CARD15 (caspase recruitment domain family member 15, formerly known as NOD2).5,6 There are three mutations—causing amino-acid substitutions Arg702Trp and Gly908Arg and the frameshift 1007fs—found within the region of CARD15 that encodes a leucine-rich repeat, which is responsible for bacterial recognition. At least one of these mutations is present in 25–35% of Crohn’s disease patients of European ancestry, but not in Asian or African American Crohn’s disease patients.3 Mutations in CARD15 are associated with distal ileal Crohn’s disease in particular, and have been found in some patients with stricturing disease.

The leucine-rich repeat region of CARD15 binds muramyl dipeptide (MDP), which is the biologically active moiety of peptidoglycan, a ubiquitous cell-wall polymer found in almost all bacteria.7 The binding of MDP by dimerized CARD15 activates nuclear factor (NF)κB, which forms part of a central signaling pathway that stimulates the transcription of multiple genes that encode both proinflammatory and protective molecules. The mutations causing Arg702Trp, Gly908Arg and 1007fs cause defective MDP binding, but studies report conflicting consequences of having such mutations.

Mutant CARD15 fails to clear Salmonella from epithelial cells,8 and clearance of invasive bacteria is dependent on NFκB activation via the cell-death regulatory protein GRIM-19.9 It is also possible that defective CARD15 results in increased luminal bacterial populations, particularly within the crypts. CARD15 is constitutively expressed in Paneth cells,10 the source of secreted antimicrobial peptides such as the α-defensins.

Targeted deletion of Card15 in mice decreases

JULY 2006 VOL 3 NO 7 SARTOR

Luminal microbial antigens and adjuvants

Genetic susceptibility

IBD

Immune response

Environmental triggers

Figure 1 Interaction of various factors contributing to chronic intestinal inflammation in a genetically susceptible host. Genetic susceptibility is influenced by the luminal microbiota, which provide antigens and adjuvants that stimulate either pathogenic or protective immune responses. Environmental triggers are necessary to initiate or reactivate disease expression.

Abbreviation: IBD, inflammatory bowel diseases.

Box 1 Etiologic theories of inflammatory bowel diseases ■

Persistent specific infection

■

Dysbiosis (abnormal ratio of beneficial and detrimental commensal microbial agents)

■

Defective mucosal barrier function

■

Defective microbial clearance

■

Aberrant immunoregulation

α-defensin production and enhances susceptibility to experimental Listeria monocytogenes infection after oral, but not systemic (intraperitoneal), challenge.11 These results are consistent with the decrease in α-defensin production seen in Crohn’s disease patients, particularly those with CARD15 mutations.12 In addition, Paneth cells are selectively expressed in the ileum, perhaps accounting for the distal ileal involvement of Crohn’s disease in patients with CARD15 mutations.

Finally, Strober and colleagues have attempted to reconcile the observed activation of NFκB in patients with active Crohn’s disease, rather than decreased activity predicted by a lossof-function mutation. Their investigations

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 391 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

Table 1 Genes with functions associated with inflammatory bowel diseases and experimental colitis.

Gene

Chromosome (human)

Function

CARD15

16

NFκB activation and/or regulation, killing of intracellular pathogens, Paneth-cell function, (α-defensin production)

Crohn’s disease

SLC22A4 & SLC22A5

5

Organic cation, carnitine transporters, possibly transport xenobiotic substances

DLG5

10

Epithelial scaffolding protein

PPARG

3

Intracellular inhibitor of NFκB and cellular activation

7

Efflux transporter for drugs and, possibly, xenobiotic compounds

Ulcerative colitis MDR1

Abbreviations: CARD15, caspase recruitment domain family, member 15 (formerly NOD2); DLG5, discs large homolog 5 (Drosophila); MDR1; multidrug resistance 1; PPARG, peroxisome proliferative-activated receptor gamma; SLC22A4 and SLC22A5, solute carrier family 22 (organic cation transporter), members 4 and 5 (formerly OCTN1 and OCTN2).

demonstrated that, in CARD15-defective cells, Toll-like receptor 2 (TLR2) was unable to downregulate NFκB activation.13 These abnormalities could result in defective downregulation of the innate immune response to bacterial adjuvant stimulation, ineffective clearance of intracellular bacterial infection and proliferation of both luminal and mucosally adherent commensal bacteria. Each of these situations has been documented in Crohn’s disease patients.14 SLC22A4 and SLC22A5

Two functional variants of the organic cation transporters OCTN1 and OCTN2 have been associated with Crohn’s disease in association with CARD15 mutations.15 Mutations in the transcribed region of SLC22A4, which encodes OCTN1, and the promoter region of SLC22A5, which encodes OCTN2, affect the transcription and function of these carnitine and organic cation transporters. These variants are most actively expressed in the intestinal epithelium, macrophages and T cells, and cause decreased carnitine transport. Although many studies have associated the region of chromosome 5 that contains SLC22A4 and SLC22A5 with Crohn’s disease, some investigators are hesitant to identify the mutations in these genes as causative of Crohn’s disease because of the tight linkage disequilibrium that exists between multiple genes in this chromosomal region.16 The DLG5 gene

Two haplotypes of DLG5, which encodes a scaffolding protein that helps to maintain epithelial

392 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

integrity, have been associated with Crohn’s disease and combined ulcerative colitis and Crohn’s disease populations.17 Like the OCTN1 and OCTN2 variants, the 113G>A substitution in DLG5 is associated with CARD15 mutations in patients with Crohn’s disease. The association of the 113A DLG5 allele with Crohn’s disease has been confirmed.

The MDR1 gene

The multidrug resistance gene MDR1 encodes P-glycoprotein 170, a transporter that governs efflux of drugs and possibly xenobiotic compounds from cells. P-glycoprotein 170 might also function as a ‘flippase’ that moves amphipathic substrates from the inner to the outer leaflet of the cell membrane. MDR1 variants have been associated with ulcerative colitis18 and Crohn’s disease.19 MDR1 is of particular interest, because it has been associated with treatmentrefractory IBD,20 and because mice in which Mdr1 has been deleted develop colitis.21 Bone-marrowtransplantation studies have implicated epithelial and/or mesenchymal cells in the pathogenesis of colitis in Mdr1-deficient mice.

The PPARG gene

PPARG (peroxisome proliferative-activated receptor γ) variants have been linked with susceptibility in the SAMP1/YitFc mouse model of spontaneous chronic ileitis, and rare PPARG polymorphisms were found to be associated with human Crohn’s disease.22 PPARγ is a nuclear receptor that inhibits NFκB activity: its expression is decreased in patients with active ulcerative colitis23 and its expression is upregulated by

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

5-aminosalicylic acid.24 In addition to a potential role in protecting against intestinal inflammation, treatment with the PPARγ ligand rosiglitazone was effective in an open-label trial involving ulcerative colitis patients25 as well as in mouse experimental colitis.26 Summary

To date, four genes have been associated with Crohn’s disease and one with ulcerative colitis; these data have been replicated. Strong associations with other chromosomal regions and genes (e.g. NFκB1, TLR5) have yet to be replicated, but such associations make it highly likely that many additional genes will be implicated in the pathogenesis of IBD, while others will be associated with extraintestinal disease (e.g. HLA-B27 and HLA-DR0103 human leukocyte antigen haplotypes, etc.) and with responses to pharmacologic treatment (i.e. pharmacogenomics). The genes associated with the pathogenesis of IBD regulate innate immune responses, mucosal barrier function and bacterial killing.

IMMUNE RESPONSE

Both Crohn’s disease and ulcerative colitis patients have activated innate (macrophage, neutrophil) and acquired (T and B cell) immune responses and loss of tolerance to enteric commensal bacteria.27,28 Tolerance, in normal hosts, is mediated by regulatory T cells, B lymphocytes, natural killer T cells and dendritic cells that secrete transforming growth factor (TGF)-β and interleukin (IL)-10, interferon (IFN)-α/β and prostaglandin J2. Antibody-neutralization studies have implicated tumor necrosis factor (TNF) and IL-12 p40 in the pathogenesis of Crohn’s disease,29,30 while T cells have been linked to ulcerative colitis by the effectiveness of T-cell-ablative therapies,31 ciclosporin and tacrolimus.32 Innate immune responses

Macrophages and dendritic cells in the lamina propria are increased in absolute number and have an activated phenotype in both forms of IBD, but have been studied in greater detail in Crohn’s disease. Production of proinflammatory cytokines and chemokines is enhanced in IBD (Table 2), and expression of adhesion molecules and co-stimulatory molecules is increased.33 As shown in Table 2, cells involved in innate immune responses are activated and the expression of most proinflammatory cytokines and

JULY 2006 VOL 3 NO 7 SARTOR

Table 2 Cytokines associated with inflammatory bowel diseases.

Cytokine

Crohn’s disease

Ulcerative colitis

Innate immune response IL-1β TNF IL-6 IL-8a IL-12

N

IL-18 IL-23

N

IL-27

N

T-cell response N

IFN-γ IL-5

N

IL-13

N

IL-17

N

IL-21

N

aRepresentative of a large number of chemokines.

Abbreviations: arrow indicates increase; IL, interleukin; N, normal.

chemokines is upregulated in both Crohn’s disease and ulcerative colitis. Type 1 T-helper lymphocyte (TH1) and TH17-related cytokines involved in innate immunity (e.g. IL-12, IL-23 and IL-27) are, however, selectively activated in Crohn’s disease.

Adhesion molecules such as intercellular cell adhesion molecule 1 (ICAM1) are necessary for circulating cells to be able to stick to the activated endothelium, which is the first step in the extravasation of mononuclear cells and polymorphonuclear cells into the inflammatory focus. In addition, adhesion molecules mediate migration of the extravasated immune cells through the stroma to the source of maximal chemokine production, as well as through the epithelium to the lumen, where they produce crypt abscesses.34 Proinflammatory molecules are preferentially produced by monocytes and polymorphonuclear cells that have migrated to the inflammatory focus, rather than by resident macrophages.

Resident intestinal macrophages have a limited capacity to respond to bacterial adjuvants owing

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 393 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

dsRNA

Lipopeptides Peptidoglycan IL-1

Lipopolysaccharide HSP60 TLR4/ MD2 Flagellin

TLR3

TLR2

IL-1R

TLR5

CpG DNA

TNF TNFR

TLR9

MAPK ERK1/2

DAP

NFκB p38

p50

p65

CARD4

JNK CARD15 MDP

Transcription

Figure 2 Binding of microbial adjuvants to extracellular and intracellular pattern-recognition receptors.

Toll-like receptors on the cell membrane selectively bind to various bacterial, viral or fungal components.

This ligation activates conserved signaling pathways that activate NFκB and mitogen-activated protein kinases. These transcription factors stimulate the expression of a number of proinflammatory and antiinflammatory genes. Homologous intracellular receptors, CARD4 (formerly NOD1) and CARD15 (formerly NOD2), bind to diaminopimelic acid and muramyl dipeptide, respectively, to activate NFκB as discussed in the text. CARD15 can modulate NFκB activation following ligation of TLR2. Abbreviations: CARD, caspase recruitment domain family member; CpG DNA, DNA containing cytosine–guanine repeats linked by phosphodiester bonds; DAP, diaminopimelic acid; dsRNA, double-stranded RNA; ERK, extracellular signalregulated kinase; HSP60, heat shock protein 60; IL-1, interleukin-1, IL-1R, interleukin 1 receptor; JNK, c-Jun amino-terminal kinase; MAPK, mitogen-activated protein kinase; MDP, muramyl dipeptide; NFκB, nuclear factor κB; P38, mitogen-activated protein kinase 1; P50, subunit of NFκB that forms a heterodimer with P65; P65, NFκB subunit; TLR, Toll-like receptor; TLR4/MD2, complex of Toll-like receptor 4 and MD2, a molecule that confers responsiveness to lipopolysaccharide; TNF, tumor necrosis factor; TNFR, TNF receptor. With permission from reference 32 © (2005) Elsevier.

to downregulation of their bacterial recognition receptors, such as TLR and CD14, the co-ligands for lipopolysaccharide.35 Similarly, intestinal epithelial cells normally have low levels of TLRs, which allows epithelial cells to reside in the high bacterial concentration of the distal ileum and colon. TLR molecules are expressed on the surface of various effector cells of the innate immune response.36 Like CARD15, these pattern-recognition receptors selectively bind to specific microbial adjuvants and initiate signaling through NFκB (Figure 2). Although each type of TLR binds a specific bacterial adjuvant (i.e. TLR4 and CD14 bind lipopolysaccharide, and TLR2 binds peptidoglycan), these signals all converge on a single pathway via myeloid differentiation primary response protein MyD88, which activates NFκB. Activation of NFκB stimulates expression of numerous molecules relevant to the pathogenesis

394 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

of IBD. These include molecules involved in the inflammatory response, such as IL-1β, TNF, IL-6, IL-8 and other chemokines, ICAM1 and other adhesion molecules, and co-stimulatory molecules, including CD40, CD80, CD86 and the inducible T-cell co-stimulator ICOS. Expression of each of these proinflammatory molecules is increased in active IBD.33 By contrast, cytokines that induce TH1 and TH17 responses are selectively upregulated in active Crohn’s disease but not ulcerative colitis (Table 2).

Moreover, blockade of TNF by neutralizing monoclonal antibodies treats active Crohn’s disease and ulcerative colitis,29,37 and antibodies to IL-12 p40 treat Crohn’s disease.30 Selective inhibition of most of these cytokines attenuates the onset of experimental colitis.33 In addition to inducing the expression of inflammatory genes, NFκB simultaneously stimulates the expression of various protective molecules,

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

such as TNF-induced protein 3 (formerly A20), CARD15, cyclo-oxygenase 2, β defensins, PPARγ and its own inhibitor, IκBα, that inhibit inflammatory responses.

NFκB is activated in the tissues of IBD patients and its inhibition can attenuate experimental colitis;38 accordingly, the NFκB pathway was thought to have predominantly proinflammatory activities. Selective genedeletion studies have, however, shown that NFκB has both beneficial and detrimental effects on inflammation, with strikingly different functions in different cell types. For example, NFκB is essential to epithelial homeostasis but is also integrally involved in the pathogenesis of bone-marrow-derived, mononuclear-cellmediated intestinal inflammation induced by various stimuli.39,40 Similarly, in mice, deletion of Myd88 exacerbates the colitis induced by dextran sodium sulfate, presumably by blocking epithelial NFκB activation by commensal enteric bacteria.41,42 T-cell responses

In contrast to innate immune responses, which are similarly activated in all forms of IBD, T-cell profiles are disparate in Crohn’s disease and ulcerative colitis, and so are considered separately.

Crohn’s disease The TH1 cytokine profile, which includes IFN-γ and IL-12 p40, is dominant in patients with Crohn’s disease. Traditional TH1 responses are mediated by IFN-γ, the production of which is stimulated by IL-12, produced by antigen-presenting cells (APCs). Most experimental colitis models also have a dominant TH1 response,43 although in several models TH1 responses can change into TH2 (type 2 T-helper lymphocyte) responses as the inflammatory process matures.44,45 How we think about TH1 responses has been influenced by the discovery of an additional TH17 pathway. IL-17 mediates TH17 responses.46 The production of this cytokine is stimulated by the production of IL-6, TGFB and IL-23 by innate immune cells and APCs, especially dendritic cells. Bacterial colonization stimulates IL-23 expression by ileal dendritic cells.47 The levels of both IL-23 and IL-17 are increased in Crohn’s disease tissues and most forms of experimental colitis.48–50 Of pathogenic importance, the IL-12–IFN-γ and IL-23–

JULY 2006 VOL 3 NO 7 SARTOR

IL-17 pathways seem to be mutually exclusive, since IFN-γ suppresses IL-17, and vice versa.46 The levels of an IL-12-related protein, IL-27, are also increased in patients with Crohn’s disease.49 In addition, production of IL-21 is induced by IL-12 and is selectively increased in Crohn’s disease.51 Like IL-12, IL-21 stimulates T-bet, an intracellular transcription factor that is key to TH1 cell differentiation and activation.52 Ulcerative colitis The T-cell profile of ulcerative colitis has been more difficult to characterize. This disease was considered to have a TH2 profile, but the concentrations of IL-4 and IL-5, which are normally elevated in TH2 responses, have been variable in ulcerative colitis tissues.53 On the basis of results obtained from studies of the oxazalone colitis model, one of the few models to exhibit a TH2 profile, Fuss and colleagues have suggested that ulcerative colitis has an atypical TH2 response, mediated by natural killer T cells that secrete IL-13.54 These natural killer T cells are activated by APCs that express the nonclassical major histocompatibility complex (MHC) molecule, CD1d, which presents lipid rather than protein antigens to T cells. These unique observations could explain some of the discrepancies of past studies, but will require confirmation before being accepted. If true, blockade of IL-13 could provide an exciting new approach to ulcerative colitis treatment.

T-cell subsets are stimulated by APCs, most notably dendritic cells. Dendritic cells have the unique capacity to activate naive T cells.

Dendritic cells are found in a resting (inactive) state in the lamina propria and Peyer’s patches in the normal intestine. When they encounter a microbial adjuvant, they take on an activated phenotype and migrate to the mesenteric or caudal lymph nodes, where they stimulate T cells. APCs interact with T cells in several ways, including by binding to co-stimulatory molecules (CD40–CD40 ligand, CD80–CTLA4, ICOS, etc.), by presenting an antigen on the surface of the MHC, which is recognized by the appropriate T-cell receptor, and by secretion of cytokines such as IL-6, IL-12, IL-23, IL-10 or TGFβ. Each of these signals contributes to the T-cell phenotype and state of activation.

Dendritic cells are present in increased numbers in the inflamed mucosa of patients with active IBD and in experimental ileocolitis.55 Moreover, co-stimulatory molecules seem to be attractive

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 395 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

targets for therapeutic intervention, on the basis of promising results in animal models.56

recruitment seem to be similar in ulcerative colitis and Crohn’s disease.

Cell trafficking

Summary

Circulating effector and regulatory cells enter the intestine through a highly selective mechanism that involves interaction with the vascular endothelium, diapedesis through the vessel wall and migration to the lamina propria.57 Circulating T cells that bear integrin-α4β7 bind to the endothelium of the colonic and small intestinal postcapillary venules that selectively express mucosal vascular addressin cell adhesion molecule (MAdCAM). MAdCAM expression is enhanced in the inflamed intestine, leading to increased entry of gut-specific T cells. Endothelial cells in the small intestine also selectively express CCL25 (formerly known as thymus-expressed chemokine; TECK), an adhesion molecule that binds to T cells that express the chemokine receptor CCR9.58 The inflamed ileum contains CD4+ T cells that express the chemokine receptor CCR2, which binds to CCL2 (formerly known as macrophage chemotactic factor 1).59 The selective expression of vascular adhesion molecules provides a mechanism for targeted recruitment of T-cell subsets to the small intestine rather than the colon.

Inflammatory cytokines such as TNF, IL-1β and IL-6 upregulate local endothelial expression of vascular cell adhesion molecule 1 (VCAM1), very late antigen 4 (VLA4) and ICAM1 that cause circulating neutrophils and monocytes to adhere to the inflamed endothelium.

Chemokines secreted by activated cells of the innate immune response in the lamina propria provide a chemotactic gradient to recruit new effector cells to the inflammatory focus to sustain and potentiate the inflammatory response. These newly recruited innate immune cells are particularly sensitive to activation by bacterial adjuvants like lipopolysaccharide, peptidoglycan and flagellin.35 This better understanding of cellular trafficking in the intestine has provided new molecular targets for blocking inflammation in IBD patients. Monoclonal antibody to integrin-α4 (natalizumab), which binds both intergrin-α4β7 (the MAdCAM1 ligand) and integrin-α4β1 (the VLA4 ligand), is effective in treating Crohn’s disease,60 while a humanized anti-integrinα4β7 antibody has been used to treat active ulcerative colitis.61 Mechanisms of effector-cell

Crohn’s disease and ulcerative colitis are characterized by enhanced recruitment and retention of effector macrophages, neutrophils and T cells into the inflamed intestine, where they are activated and release proinflammatory cytokines. Accumulation of effector cells in the inflamed intestine is a result of enhanced recruitment as well as prolonged survival caused by decreased cellular apoptosis. Crohn’s disease is a predominantly TH1- and TH17mediated process, while ulcerative colitis seems to be an atypical TH2 disorder. There is no direct evidence of defective regulatory T-cell function in either disease, but evidence is accumulating that innate immune responses are deficient in Crohn’s disease. Therapy can be directed towards blocking effector activation, blocking biologic activity of proinflammatory cytokines and their receptors, inhibiting T cell–APC interactions, selectively blocking effector cell entry, inducing apoptosis of activated effector cells or enhancing regulatory cell activity.

396 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

COMMENSAL MICROBIAL STIMULANTS

Enteric microflora can stimulate immune responses either by functioning as adjuvants or antigens. As adjuvants they activate innate immune responses, including dendritic cells and other APCs, and as antigens they stimulate the clonal expansion of T cells that selectively recognize the antigen through their T-cell receptor.

Adjuvants

Numerous bacterial adjuvants, most notably lipopolysaccharide, peptidoglycan, flagellin and nonmethylated DNA (CpG motif), can bind selectively to various TLRs on innate immune cells, intestinal epithelial cells and mesenchymal cells (Figure 2).7 Ligation of these TLRs activates NFκB and the mitogen-activated protein kinases, which stimulate the transcription of a host of proinflammatory and regulatory genes.

Activation of macrophages from susceptible individuals, by enteric lipopolysaccharide, peptidoglycan, flagellin or CpG, stimulates the production of IL-1β, TNF, IL-6, IL-8 and other chemokines, IL-12 p40 (and thus IL-12 and IL-23), adhesion molecules, IL-18, reactive oxygen species, nitric oxide and leukotrienes,

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

which can all participate in the inflammatory response. In addition, NFκB activation of APCs (including dendritic cells) by microbial adjuvants induces the expression of MHC class II antigens, co-stimulatory molecules, IL-12 and IL-23, which can activate TH1 and TH17 cells, respectively, if the appropriate antigen is present.

Lipopolysaccharide can stimulate IL-12 p40 production by bone-marrow-derived dendritic cells in IL-10-deficient mice, and colonization of previously germ-free rodents with various commensal bacterial species can induce ICAM1 and IL-6 expression by intestinal epithelial cells.62,63 In vivo studies show that commensal bacteria selectively activate IL-12 p40 in dendritic cells of the distal ileum.47 Bacterial flagellin is both an antigen and adjuvant. A form of flagellin was shown to be a dominant antigen in experimental colitis and to induce serum antibody production in 50% of Crohn’s disease patients.64,65 Flagellin binds to TLR5 to activate NFκB. In addition to their proinflammatory properties, bacterial adjuvants can induce protective anti-inflammatory responses. For example, lipopolysaccharide stimulates IL-10 production in dendritic cells from normal mice,66 and certain CpG preparations can prevent the onset of experimental colitis by inducing the production of type 1 IFN (IFN-α/β) in plasmacytoid dendritic cells, via TLR9 ligation.67 Antigens

Genetically susceptible hosts, including IBD patients, and genetically engineered rodents and mice with spontaneous mutations (C3H/ HeJBir and SAMP1/YitFc), have aggressive T-cell responses to luminal commensal bacteria.27,64,68,69 Although B-cell responses to enteric microbial constituents are exaggerated in Crohn’s disease,28,64,70 ulcerative colitis70 and experimental intestinal inflammation,69,71,72 antibodies are not necessary to transmit disease in experimental colitis, and B-lymphocytes seem to be regulatory rather than pathogenic.73,74 Antibodies to bacteria have, therefore, been primarily useful as part of diagnostic tests and could potentially identify clinically important subsets of IBD patients who have selective responses to therapeutic agents and have predictable natural histories.

Very few studies of antigen specificity have been performed in IBD patients; however, results

JULY 2006 VOL 3 NO 7 SARTOR

of animal model experiments show that disease can be transferred to immunodeficient hosts, such as T-cell-deficient SCID (severe, combined immunodeficient) and Rag–/– mice and nude rats, via T cells that respond to commensal enteric bacteria. Cong et al. demonstrated that T-cell clones that respond to cecal bacterial lysates transfer colitis to SCID mice, but that nonspecifically activated T cells do not.75 Since then, the same group have demonstrated that flagellin-specific T-cell clones can transfer colitis to SCID mice.64 It has been shown that persistent luminal antigen stimulation is necessary for the transfer of colitis.76 CD4+ T cells from rodents with colitis induced colitis when they were transferred into T-cell-deficient recipients colonized with specific-pathogen-free commensal bacteria, but did not cause disease in germ-free (sterile) recipients. After transfer, T cells in the recipients responded to luminal bacterial products. There is both bacterial and host specificity in these TH1 responses to bacterial antigens.66 Selective T-cell responses have been shown in IL-10–/– mice selectively colonized (monoassociated) with either Escherichia coli or Enterococcus faecalis. T cells from E. colimonoassociated mice with colitis produced IFN-γ in response to E. coli but not E. faecalis lysates, and T cells from E. faecalis-monoassociated mice with colitis only responded to E. faecalis.68 HLAB27TG rats monoassociated with either E. coli or E. faecalis failed to develop colitis—instead, colitis and T-cell responses were elicited by Bacteroides vulgatus.77 Several groups have used molecular approaches to identify the dominant antigens in experimental colitis models. Lodes et al. identified a specific form of flagellin from commensal Clostridium species as the dominant antigen in C3H/HeJBir mice and transferred colitis via T-cell clones that reacted to flagellin.64 Of considerable importance, approximately half of Crohn’s disease patients, but not ulcerative colitis patients or normal controls, have a selective serologic response to the same form of flagellin,65 showing the clinical relevance of these experimental model studies.

Tannock et al. used denaturing gradient gel electrophoresis to show that Bifidobacterium animalis, Clostridium cocleatum, Bacteroides and Enterococcus species were selectively expanded in IL-10–/– mice with colitis.78 Preliminary studies have show that B. animalis monoassociation

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 397 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

of gnotobiotic IL-10–/– mice (i.e. mice whose microfauna and microflora are known in their entirety) induces a mild colitis and specific TH1 immune responses.79 Finally, Bacteroides vulgatus heat-shock protein 60 has been identified as a dominant antigen in HLA-B27TG rats monoassociated with B. vulgatus.80 These results are important because both heat-shock protein 60 and flagellin are strong adjuvants as well as antigens by virtue of their binding to TLR4 and TLR5, respectively.

Evidence of commensal microbial stimulation of intestinal inflammation

There is abundant evidence that commensal bacteria are involved in the pathogenesis of human IBD and in experimental colitis.81 The most convincing mechanistic studies have been done in gnotobiotic animal models.

Animal models In at least 11 different animal models, colitis and immune activation fail to develop in the absence of commensal bacteria,81 and multiple animal models of colitis respond to antibiotics and probiotics. Studies have demonstrated both bacterial species and host specificity for the induction of experimental colitis.

Monoassociation of IL-10–/– mice with the commensal bacteria E. faecalis and E. coli induced phenotypically distinct forms of colitis.68 E. faecalis-induced colitis was slow in onset (10–12 weeks after bacterial colonization) and involved the distal colon, with severe transmural colitis accompanied by dysplasia and duodenal obstruction after 24 weeks of monoassociation. By contrast, E. coli monoassociation led to relatively early (3 weeks) onset of a mild-to-moderate inflammation that was at its most severe in the cecum. Dual association with both commensal bacterial species rapidly (by 1 week) induces severe pancolitis with dysplasia after 5 weeks.82 Preliminary studies show that Klebsiella monoassociation induces moderate pancolitis81 and Bifidobacterium animalis monoassociation causes distal colonic and duodenal inflammation.79 These results demonstrate that even a traditionally probiotic bacterial species can induce inflammation in a susceptible host, raising concern over the safety of probiotic therapy in some patients. An exception to the requirement of bacteria to induce intestinal inflammation is provided by the

398 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

dextran-sodium-sulfate-induced colitis model, which (uniquely) worsens in the absence of commensal bacteria.42 This acute epithelial injury model demonstrates the protective role of NFκB in epithelial cell homeostasis.41,42 Variable—and at times totally opposite— responses to three different bacterial species, in three different models of colitis, demonstrate the requirement for host specificity:77 E. faecalis and E. coli both induced colitis in monoassociated IL-10–/– mice, but Bacteroides vulgatus did not.

By contrast, B. vulgatus, but not E. coli nor E. faecalis, induced colonic inflammation in HLA-B27 transgenic rats.77 None of the three bacterial species caused disease in bone-marrowtransplanted CD3ε transgenic mice.77 Similarly, various probiotic species have variable results in the same host, and different hosts respond variably to the same probiotic species.84,85 These studies indicate that different bacterial species can cause different disease phenotypes in a single host, that different bacterial species provide the dominant stimuli for disease, and that different probiotic species can provide selective treatment in hosts with different genetic backgrounds. These observations have clear relevance for therapeutic responses to antibiotics and probiotics.

Human studies IBD patients can respond favorably to antibiotic and probiotic treatment.86 Several pathophysiologically relevant messages come from these studies. First, antibiotics are effective for Crohn’s colitis but not isolated ileitis, except in the postoperative state in which loss of the ileocecal valve probably changes the luminal microenvironment. Second, antibiotics are not effective in ulcerative colitis, but reproducibly treat pouchitis. Third, probiotics can prevent the relapse of chronic, relapsing pouchitis and ulcerative colitis, but (with the exception of a single uncontrolled trial in mild-to-moderate active ulcerative colitis patients87) are not useful in treatment of active disease. Four, different probiotic preparations produce variable responses in the same clinical setting.88,89 These therapeutic responses suggest that the dominant bacterial stimuli are different in ileal and colonic Crohn’s disease, that commensal bacteria probably have a more important role in Crohn’s disease and pouchitis than in ulcerative colitis, and that subsets of patients will selectively respond to individual treatments.

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

ENVIRONMENTAL TRIGGERS

Studies have implicated several environmental factors in the pathogenesis of IBD.1 These factors include smoking, which is protective in ulcerative colitis but detrimental in Crohn’s disease, diet, the use of antibiotics and nonsteroidal anti-inflammatory drugs (NSAIDs), stress and infection. Unfortunately, the mechanisms by which these factors initiate the onset of disease or reactivate quiescent IBD are not well understood. From a broad perspective, these triggering factors alter mucosal barrier integrity, immune responses, or the luminal microenvironment, each of which have an impact on susceptibility to inflammation.

Infection and NSAIDs can transiently initiate nonspecific inflammation, break the mucosal barrier and activate innate immune responses.

These events could then lead to enhanced uptake of commensal bacterial antigens and adjuvants that stimulate protracted T-cell-mediated intestinal inflammation in the genetically susceptible host. An example of this is induction of chronic colitis in IL-10–/– mice, by exposure to the NSAID piroxicam for 2 weeks.90 Although IBD patients routinely implicate diet and stress in their disease, roles for these factors in IBD and potential mechanisms by which they might act are very poorly understood. Dietary additives such as aluminum and iron have a well-described adjuvant activity and stimulate bacterial virulence.91 Stress can alter mucosal permeability, mucosal blood flow, epithelial electrolyte and water secretion and expression of cytokines and neuropeptides.

Smoking is perhaps the most thoroughly documented environmental contributor to IBD, but its opposite effect on Crohn’s disease and ulcerative colitis, respectively, is difficult to understand. Nicotine, carbon monoxide and hypoxia have all been suggested to be mediators of the effects of smoking on IBD.92,93 SPECIFIC THEORIES

As stated in the introduction, several theories attempt to explain the persistent, relapsing inflammation seen in patients with Crohn’s disease or ulcerative colitis. There have been new developments for each theory and these are briefly discussed below.

Persistent pathogenic infection

Although numerous infectious explanations for Crohn’s disease and ulcerative colitis have

JULY 2006 VOL 3 NO 7 SARTOR

been postulated since they were first described, there is reasonable experimental support for only two pathogens, in the development of Crohn’s disease.

Mycobacterium avium subspecies paratuberculosis

The granulomatous inflammation of Crohn’s disease resembles that of intestinal tuberculosis or the spontaneous enteritis caused by Mycobacterium avium subspecies paratuberculosis (MAP) in ruminants, which is known as Johne’s disease.94 MAP was first experimentally implicated in the pathogenesis of Crohn’s disease in 1984, when the organism was cultured from three resected intestinal specimens.95 Noncontrolled studies have reported that up to 84% of patients respond to treatment with combinations of antibiotics effective against MAP.94,96 The controversy intensified when MAP was reported to be present in commercial pasteurized milk and in human breast milk, providing a mode of transmission.97,98 The controversy has continued with the finding that viable MAP can be cultured from the blood of patients with Crohn’s disease (55%) and ulcerative colitis (22%), but not from controls.99 A recent study detected the MAP DNA insertion sequence (IS)900 in resected gut tissue of 52% of Crohn’s disease patients, versus 2% of ulcerative colitis patients and 5% of controls.100 These results are consistent with most studies that detect MAP more frequently in Crohn’s disease patients than in ulcerative colitis patients or controls, although results are highly variable and MAP detection rates have ranged from 0–100%.

There is the potential for zoonotic infection with MAP. In North America, Europe and Australia, there is widespread MAP carriage in dairy cattle.101 MAP can be recovered from the ground and water surrounding infected herds,102 and it is shed into the milk of infected cows, where MAP can resist pasteurization because of its intracellular location.103 This theory has weaknesses, however, since histochemical evidence of tissue infection is absent and the burden of MAP infection is extremely low in Crohn’s disease.94 Tissue damage from a paucibacillary infection is most likely the result of a cell-mediated immune response to MAP, yet none has been detected in Crohn’s disease patients. Moreover, a mycobacterial disease would be expected to

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 399 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

intensify with immunosuppression caused by corticosteroids, anti-TNF treatment or concomitant HIV infection. Crohn’s disease patients clinically improve with immunosuppression rather than worsen. A definitive 2-year, blinded, placebo-controlled trial with triple antibiotic therapy designed to clear mycobacterial infection has been conducted in Australia139. Preliminary results show a 33% response rate in patients treated with antibiotics versus 18% in controls (P <0.05), but there was no persistence of response 1 year after treatment (19% versus 12%; not significant). It is possible that a subset of genetically susceptible Crohn’s disease patients, particularly those with a defect in their ability to clear an intracellular infection (for example those with CARD15 polymorphisms), have a persistent MAP infection that causes their disease, or that this organism potentiates cellular immune responses to commensal enteric bacteria. MAP DNA IS900 is not preferentially found in the ileum, however, which would be expected if MAP was associated with CARD15 polymorphisms. It is more likely that this relatively common environmental agent selectively colonizes or lodges in the ulcerated mucosa of Crohn’s disease patients but does not cause disease.

Enteroadherent and invasive E. coli Darfeuille-Michaud and Colombel have recovered a virulent strain of E. coli from 22% of mucosal biopsies from Crohn’s disease patients with chronic, postoperative, recurrent inflammation in the neoterminal ileum, 36% of those with early recurrence and only 6% of controls.104 These E. coli adhere to epithelial cells via type 1 pilli, invade macrophages by an active, monofilamentdependent and microtubule-dependent mechanism, and persist and replicate within phagocytes. Infection of macrophages induces TNF production, but no cellular apoptosis. Of interest, following infection with this E. coli strain, monocytes of patients with CARD15 polymorphisms had decreased TNF and IL-10 production relative to cells from normal controls.105 These results are consistent with reports of immunohistochemical evidence of increased E. coli mucosal adherence, invasion of ulcers and fistulae, and their presence within lamina propria macrophages in Crohn’s disease patients,14,106 and with the concept of defective clearance of intracellular infections of patients with CARD15 polymorphisms.8

400 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

Dysbiosis

An altered balance of beneficial versus aggressive microbial species could lead to a proinflammatory luminal milieu that drives chronic intestinal inflammation in a susceptible host. Numerous studies have implicated several commensal organisms, such as E. coli, Bacteroides, Enterococcus and Klebsiella species, in the pathogenesis of experimental intestinal inflammation and human IBD.81 By contrast, various Lactobacillus and Bifidobacterium species have predominantly protective effects and have been used therapeutically as probiotics.81 Several groups have documented alterations in luminal or adherent microbial commensal flora in patients with Crohn’s disease, ulcerative colitis and pouchitis.14,81 An alternative means of changing the microenvironment in such a way as to stimulate aggressive immune responses is the acquisition of virulence factors by commensal bacteria.

As discussed above, enteroadherent and invasive E. coli have been found in the neoterminal ileum of patients with postoperative recurrence of Crohn’s disease.107 Flagellin from Clostridium subphylum cluster XIVa organisms has been shown to be a dominant antigen in experimental colitis, and to elicit serologic responses in approximately 50% of Crohn’s disease patients.64 The presence of superoxide dismutase activity alters the pathogenicity of E. faecalis.108 Dietary components can alter the composition and virulence of enteric commensal bacteria, providing one potential explanation for the marked increase in the incidence of IBD in Western countries in the second half of the twentieth century, and more recently in Eastern countries, as they adopt Western dietary practices.1 Nonabsorbed carbohydrates (prebiotics) such as inulin and fructose oligosaccharides enhance the growth of Bifidobacterium and Lactobacillus species, and provide a substrate for the production of short-chain fatty acids by these bacterial species. Short-chain fatty acids, especially butyrate, are the preferred metabolic substrates of colonocytes, and can stimulate various mucosal barrier functions.86 Iron stimulates growth and virulence of intracellular bacteria, whereas aluminum is an adjuvant for bacterial stimulation of immune responses.91 Both iron and aluminum are ubiquitous food additives in Western diets and are processed identically by mammalian and bacterial

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

acquisition and storage proteins. Dietary iron and aluminum can potentiate experimental colitis.109,110 The hygiene hypothesis offers an alternative explanation for the increased incidence of IBD, asthma and autoimmune disorders such as rheumatoid arthritis and type I diabetes in Western society. This theory suggests that exposure to pathogens or parasites, especially early in life, stimulates protective immunity that prevents later aggressive immunologic processes. Weinstock et al. have proposed that elimination of helminths by public health measures has increased the incidence of IBD and have demonstrated a therapeutic effect of Tricuris suis, the pig whipworm, in ulcerative colitis, Crohn’s disease and experimental colitis models.111–113 Defective mucosal barrier function

A defect in mucosal barrier integrity could lead to increased uptake of luminal antigens and/ or adjuvants that overwhelm the net suppressive tone of the mucosal immune system.

Alternatively, a defect in epithelial repair could potentiate damage by environmental triggers such as NSAIDs or infections that cause only transient damage in normal hosts. The net effect of either pathway is constant stimulation of innate and acquired mucosal immune responses by luminal adjuvants and antigens, respectively. As mentioned in the section on genetics, each of the genes associated with IBD, including CARD15, the genes encoding OCTN1 and OCTN2, DLG5 and MDR1 are involved in epithelial function.4 In addition, MDR1 seems to mediate excretion of xenobiotic, possibly bacterial, molecules from epithelial cells, while CARD15 might mediate bacterial clearance and production of antimicrobial α-defensins by Paneth cells.8,11 Defective microbial clearance

The association of CARD15 with Crohn’s disease and the therapeutic activity of granulocytemacrophage colony-stimulating factor (GMCSF) support the novel hypothesis that Crohn’s disease is the result of defective bacterial killing.

As discussed above, CARD15 is constitutively expressed in small-intestinal Paneth cells, and seems to stimulate α-defensin and cryptdin expression,11 and in addition might mediate intracellular bacterial killing.8 Oral challenge of CARD15–/– mice with Listeria monocytogenes leads to greater mortality

JULY 2006 VOL 3 NO 7 SARTOR

Antigen presenting cells, which process and present bacterial antigens are activated by bacterial adjuvants

APC

APC IL-12 IL-18 IL-23 IL-27

CD40

MHC

CD80 CD86

Ag CD40L

MHC IL-10 TGF-β

TCR

Ag

IL-10 IFN-α/β TCR

CD28

IFN-γ IL-17

CD80 CD86

CTLA4 CD25

TR1 T H3

TH1

Figure 3 The interaction between antigen-presenting cells and naive T cells triggers T-cell activation and differentiation. These interactions consist of cell–cell interactions and secreted products. Cellular interactions are mediated by ligation of co-stimulatory molecules, for example CD80 and CD28, or CD80 and CTLA4, and via antigen presentation by the major histocompatibility complex (MHC) on antigen-presenting cells to the T-cell receptor (TCR) complex on T cells. Secreted IL-12, IL-18, IL-23 and IL-27 stimulate TH1 lymphocytes, while IL-10 and IFN-α/β stimulate regulatory T cells. Conversely, both effector and regulatory T cells secrete cytokines that influence antigen-presenting cells. TH1 products such as interferon-γ stimulate IL-12 p40 and MHC class II expression, while regulatory T cells secrete IL-10 or transforming growth factor beta and suppress antigen-presenting cell activity. Abbreviations: Ag, antigen; CD, cluster designation; CTLA4, cytotoxic T-lymphocyte-associated protein 4; IFN, interferon; IL, interleukin; L, ligand; MHC, major histocompatibility complex; TCR, T-cell receptor; TGF-β, transforming growth factor beta; TH1, type 1 T-helper lymphocytes; TR1, type 1 T-regulatory lymphocytes. With permission © American Gastroenterological Association Institute, Bethesda, MD.

compared with normal mice; no such increased mortality is seen after systemic inoculation.11 The authors of this study attributed this increased mortality on oral challenge to deficient cryptdin production with CARD15 deficiency, although it is possible that defective bacterial clearance by epithelial cells was a contributory factor. In addition, Swidsinski et al. have shown dramatic increases in mucosally adherent enteric bacteria in patients with active Crohn’s disease and ulcerative colitis.14 The results of these experiments in mice are consistent with the independent observations of Wehkamp et al., who demonstrated defective α-defensin and β-defensin expression in Crohn’s disease

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 401 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

Table 3 Characteristics of T-cell subsets.

Designation

Origin

Antigen

Marker

Mechanisms of action

TREG

Thymus

Self antigen (possibly)

FoxP3, CD25+

Cell–cell contact, membrane-bound TGF-β

TR1

Intestine

Luminal bactieria

None

Secretion of IL-10 > TGF-β

TH3

Intestine

Oral protein

None

Secretion TGF-β > IL-10

Abbreviations: CD, cluster designation; FoxP3, forkhead box P3; IL, interleukin; TGF, transforming growth factor; TH3, type 3 T-helper lymphocyte; TREG, regulatory T cell; TR1, type 1 regulatory T cell.

patients, with maximal alterations in patients with CARD15 polymorphisms.12 Korzenik and Dieckgraefe have postulated that Crohn’s disease results from defective killing of bacteria by the innate immune system, based on pathogenic similarities of Crohn’s disease and the enterocolitis associated with chronic granulomatous disease and with glycogen-storage disease type IIb.114 The persistence of bacteria in tissues of patients with Crohn’s disease has been detected by immunohistochemistry and by culturing mesenteric lymph nodes.81 In a randomized, blinded clinical trial, daily administration of recombinant GM-CSF induced remission in 40% of patients with moderate-to-severe Crohn’s disease.115 In addition to activating microbial killing by macrophages, monocytes and neutrophils, GM-CSF might affect the functions of plasmacytoid dendritic cells, Paneth cells and epithelial cells, because these cells express receptors for this growth factor.

Aberrant immunoregulation

Animal models convincingly demonstrate that chronic intestinal inflammation can result from defective immunosuppression, hyperactive TH1 cell function or cytokine hypersecretion.43,116–119 The primary mediators of immunosuppression are IL-10 and TGF-β, which functionally interact.120 Most human studies, however, indicate that immune activation is driven by aggressive innate or T-cell responses, rather than defective immunoregulatory function. IBD patients have decreased oral tolerance.121 A potential mechanism of acquired defects in regulatory function has been suggested by MacDonald, who reported induction of SMAD7 expression in patients with active disease.122 SMAD 7 protein blocks TGF-β activity by inhibiting SMAD 2/4 phosphorylation.122 Overexpression of proinflammatory TH1 and TH17 cytokines such as TNF, IL-6, IL-12, IL-17, IL-23 and IFN-γ in Crohn’s disease,48,49 TH2 402 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

cytokines, IL-4 and IL-13 in ulcerative colitis,54 and the innate immune products IL-1β, TNF, IL-6 and chemokines33 are well documented.

Induction of effector versus regulatory T cells depends on the interaction of APCs and T cells via cell–cell interactions mediated by co-stimulatory molecules and MHC–T-cell-receptor ligation, as well as through signals from secreted cytokines (Figure 3). Three broad phenotypes of T regulatory cells are characterized, each with a unique origin, antigens and function (Table 3).

T regulatory and TR1 cells can prevent the onset of intestinal inflammation,75,123 and CD4+CD25+ T regulatory cells can reverse established experimental colitis.124 Preliminary studies show that APCs determine the phenotype of T cell (effector versus regulatory) in the IL-10–/– mouse model of colitis.125 CD4+ T cells from mesenteric lymph nodes of IL-10-deficient and wild-type mice were incubated with APCs from a pulsed lysate of cecal bacteria, from either IL-10–/– or wildtype mice.125 IFN-γ production correlated with exposure to IL-10–/– APCs, with induction of IFN-γ production by wild-type CD4+ T cells incubated with IL-10–/– APCs, and suppression of IL-10–/– T cells that had been incubated with wild-type APCs. Similar findings were observed in T-cell transfer studies in IL-10–/– or wild-type Rag2–/– (T-cell-deficient) mice.126 Loss of tolerance to commensal bacteria and induction of colitis was seen in IL-10–/– mice that received wild-type CD4+ T cells, whereas IL-10–/– CD4+ T cells induced only mild colitis in wild-type recipients.

These studies demonstrate the importance of APCs in regulating T-cell function. Several studies have now shed light on innate immunity and dendritic-cell function in intestinal inflammation.

CARD15 can inhibit cellular NFκB activation and IL-12 p40 secretion following stimulation by a TLR2 ligand,13 identifying a potential mechanism for the observed activation of NFκB in Crohn’s

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

CONCLUSIONS

Chronic intestinal inflammation can develop via many different mechanisms, which strongly suggests that Crohn’s disease and ulcerative colitis are heterogeneous diseases with similar final common pathways. This heterogeneity exists at genetic, phenotypic, immunologic, bacteriologic and therapeutic levels. It is highly likely that each IBD subtype will have unique responses to various

JULY 2006 VOL 3 NO 7 SARTOR

Tolerance Acute injury Environmental trigger Normal gut

Luminal bacteria

(infection, NSAID) Acute inflammation

Tolerance controlled inflammation

→

disease. Similarly enhanced NFκB activation was seen with overexpression of truncated Card15 in mice (truncated Card15 is the result of a mutation that is associated with a Crohn’s disease phenotype).127 Bacterial DNA (CpG motif), which prevents and treats experimental colitis in numerous models,128 inhibits NFκB activation by inducing IFN-α/β in dendritic cells.67 Finally, GM-CSF can stimulate regulatory functions in plasmacytoid dendritic cells, by inducing IFN-α/β expression,129 which provides an alternative mechanism for GM-CSF activity beyond its role in stimulating the innate immune system.

Conversely, defective APC function, as measured by decreased innate responses to bacterial ligands, has been implicated in enhanced pathogenic TH1 responses, possibly through genetic alterations in NFκB function.130 The number of inflammatory cells in the intestine is determined by cellular recruitment, proliferation and death by necrosis or apoptosis.

Data indicate that T cells in Crohn’s disease patients are resistant to apoptosis, which could lead to an expanded population of activated effector TH1 cells.130–132 Molecular mechanisms underlying this resistance are provided by altered activity of proapoptotic and antiapoptotic molecules130 and by the binding of IL-6 and soluble IL-6 receptor complexes to membrane-bound glycoprotein 130.132 Transsignaling by IL-6 induces the antiapoptotic genes BCL2 and BCL2L1 (formerly BCL-XL), while anti-IL-6-receptor antibody, or glycoprotein 130 fusion protein, induce apoptosis and attenuate colitis.132 Several therapies that are effective in the treatment of IBD, including corticosteroids, sulfasalazine, azathioprine, 6-mercaptopurine, infliximab and anti-IL-12 antibody, induce apoptosis of activated T cells and in some cases monocytes.30,133–137 Their respective ability to induce apoptosis of activated cells expressing membrane-bound TNF might explain the different therapeutic activities of anti-TNF agents in Crohn’s disease.138

immunoregulation, failure of repair or bacterial clearance

Complete healing

Genetically susceptible host

Chronic inflammation

Figure 4 Different responses to transient intestinal injury in genetically susceptible versus genetically resistant hosts. After nonspecific injury from an environmental trigger, such as an infection or exposure to a nonsteroidal anti-inflammatory drug (NSAID), normal hosts rapidly repair the mucosal defect and downregulate innate and T-cell immune responses with no residual tissue damage. By contrast, individuals in whom immunoregulation, epithelial barrier function or bacterial killing is defective, develop chronic inflammation that is mediated by aggressive T-cell responses to commensal bacterial antigens.

Chronic inflammation is perpetuated by continued uptake of luminal antigens.

With permission © American Gastroenterological Association Institute, Bethesda, MD.

treatments, explaining the lack of a universal therapeutic response to any single agent, but raising the possibility that treatment tailored for a specific subtype is highly likely to succeed.

Chronic intestinal inflammation results from the interactions of genetic, immunologic, microbial and environmental factors (Figure 1).

I propose that IBD results from the failure to appropriately downregulate nonspecific inflammation initiated by an environmental trigger, such as an acute, self-limited infection or NSAID use (Figure 4). Normal hosts quickly clear infections of invasive enteric bacteria, downregulate innate immune responses and heal the injured mucosa without stimulating effector T-cell responses.

By contrast, genetically susceptible hosts who are unable to clear an invading pathogen and/ or generate tolerogenic immune response to commensal microbial agents—by mounting appropriate innate immunity, downregulating immune responses or healing the mucosal barrier—subsequently activate pathogenic T-cell responses to commensal bacteria and proceed

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 403 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

to chronic, relapsing intestinal inflammation.

Resistance to T-cell apoptosis, lack of response to downregulatory signals and continuous exposure to luminal antigens and adjuvants help sustain this inflammatory response.

Many environmental factors can influence mucosal immune responses and enteric bacteria composition, including diet, smoking, stress, altered microenvironment and NSAID exposure.

Although I postulate that self-limited, nonspecific infections can initiate the onset of chronic inflammation and reactivate quiescent disease, it is possible that a persistent pathogen could cause disease in individuals unable to clear infections (i.e. those with certain CARD15 polymorphisms), or that the commensal bacteria of some patients could acquire virulence factors (e.g. toxins, adherence and/or invasion properties) that might cause chronic intestinal inflammation. Our current lack of complete understanding of the precise molecular mechanisms of disease pathogenesis makes this topic seem to be exceedingly complex; however, it is highly likely that further insights will allow us to conceptually characterize IBD into a spectrum of related, but distinct, subgroups that have clear etiologies, pathogenic processes and predictable responses to customized therapy.

3

4

5

6

7

8

9

10

11

12

13

14 15

KEY POINTS ■

Genes involved in Crohn’s disease and experimental ileocolitis regulate innate immune responses, bacterial killing, immune responses to endrogenous microbial antigens and epithelial function

16

■

Chronic intestinal inflammation requires the presence of commensal enteric bacteria and activated T lymphocytes

18

■

Patients with inflammatory bowel diseases and rodents with chronic intestinal inflammation exhibit loss of immunologic tolerance to normal enteric bacteria

19

■

■

17

20

Some enteric bacteria are detrimental and some are protective

21

Defective innate immune responses can lead to lack of clearance of invading bacteria and to the activation of pathogenic T cells

22

References 1 Loftus EV Jr (2004) Clinical epidemiology of inflammatory bowel disease: incidence, prevalence, and environmental influences. Gastroenterology 126: 1504–1517 2 Halfvarson J et al. (2003) Inflammatory bowel disease in a Swedish twin cohort: a long-term follow-up of concordance and clinical characteristics.

Gastroenterology 124: 1767–1773

404 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

23

24

Newman B and Siminovitch KA (2005) Recent advances in the genetics of inflammatory bowel disease. Curr Opin Gastroenterol 21: 401–407 Schreiber S et al. (2005) Genetics of Crohn disease, an archetypal inflammatory barrier disease. Nat Rev Genet 6: 376–388 Hugot JP et al. (2001) Association of NOD2 leucinerich repeat variants with susceptibility to Crohn’s disease. Nature 411: 599–603 Ogura Y et al. (2001) A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease.

Nature 411: 603–606 Cario E (2005) Bacterial interactions with cells of the intestinal mucosa: Toll-like receptors and NOD2. Gut 54: 1182–1193 Hisamatsu T et al. (2003) CARD15/NOD2 functions as an anti-bacterial factor in human intestinal epithelial cells. Gastroenterology 124: 993–1000 Barnich N et al. (2005) GRIM-19 interacts with nucleotide oligomerization domain 2 and serves as downstream effector of anti-bacterial function in intestinal epithelial cells. J Biol Chem 280: 19021–19026 Lala S et al. (2003) Crohn’s disease and the NOD2 gene: a role for Paneth cells. Gastroenterology 125: 47–57 Kobayashi KS et al. (2005) Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 307: 731–734 Wehkamp J et al. (2005) Reduced Paneth cell α-defensins in ileal Crohn’s disease. Proc Natl Acad Sci USA 102: 18129–18134 Watanabe T et al. (2004) NOD2 is a negative regulator of Toll-like receptor 2-mediated T helper type 1 responses. Nat Immunol 5: 800–808 Swidsinski A et al. (2002) Mucosal flora in inflammatory bowel disease. Gastroenterology 122: 44–54 Peltekova VD et al. (2004) Functional variants of OCTN cation transporter genes are associated with Crohn disease. Nat Genet 36: 471–475 Noble CL et al. (2005) The contribution of OCTN1/2 variants within the IBD5 locus to disease susceptibility and severity in Crohn’s disease.

Gastroenterology 129: 1854–1864 Stoll M et al. (2004) Genetic variation in DLG5 is associated with inflammatory bowel disease. Nat Genet 36: 476–480 Ho GT et al. (2005) Allelic variations of the multidrug resistance gene determine susceptibility and disease behavior in ulcerative colitis. Gastroenterology 128: 288–296 Brant SR et al. (2003) MDR1 Ala893 polymorphism is associated with inflammatory bowel disease. Am J Hum Genet 73: 1282–1292 Farrell RJ et al. (2000) High multidrug resistance (P-glycoprotein 170) expression in inflammatory bowel disease patients who fail medical therapy.

Gastroenterology 118: 279–288 Panwala CM et al. (1998) A novel model of inflammatory bowel disease: mice deficient for the multiple drug resistance gene, Mdr1a, spontaneously develop colitis. J Immunol 161: 5733–5744 Sugawara K et al. (2005) Linkage to peroxisome proliferator-activated receptor-γ in SAMP1/YitFc mice and in human Crohn’s disease. Gastroenterology 128: 351–360 Dubuquoy L et al. (2003) Impaired expression of peroxisome proliferator-activated receptor γ in ulcerative colitis. Gastroenterology 124: 1265–1276 Rousseaux C et al. (2005) Intestinal anti-inflammatory effect of 5-aminosalicylic acid is dependent on peroxisome proliferator-activated receptor-γ. J Exp Med 201: 1205–1215

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

25 Lewis JD et al. (2001) An open-label trial of the PPAR-γ ligand rosiglitazone for active ulcerative colitis. Am J Gastroenterol 96: 3323–3328 26 Su CG et al. (1999) A novel therapy for colitis utilizing PPAR-γ ligands to inhibit the epithelial inflammatory response. J Clin Invest 104: 383–389 27 Duchmann R et al. (1995) Tolerance exists towards resident intestinal flora but is broken in active inflammatory bowel disease (IBD). Clin Exp Immunol 102: 448–455 28 Mow WS et al. (2004) Association of antibody responses to microbial antigens and complications of small bowel Crohn’s disease. Gastroenterology 126: 414–424 29 Targan SR et al. (1997) A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor α for Crohn’s disease. Crohn’s Disease cA2 Study Group. N Engl J Med 337: 1029–1035 30 Mannon PJ et al. (2004) Anti-interleukin-12 antibody for active Crohn’s disease. N Engl J Med 351: 2069–2079 31 Sawada K et al. (2005) Leukocytapheresis in ulcerative colitis: results of a multicenter double-blind prospective case–control study with sham apheresis as placebo treatment. Am J Gastroenterol 100: 1362–1369 32 Lichtiger S et al. (1994) Cyclosporine in severe ulcerative colitis refractory to steroid therapy. N Engl J Med 330: 1841–1845 33 Sartor RB and Hoentjen F (2005) Proinflammatory cytokines and signaling pathways in intestinal innate immune cells. In Mucosal Immunology, 681–701 (Eds Mestecky J et al.) Philadelphia: Elsevier 34 Reaves TA et al. (2005) Neutrophil transepithelial migration: role of Toll-like receptors in mucosal inflammation. Mem Inst Oswaldo Cruz 100 (Suppl 1): 191–198 35 Smythies LE et al. (2005) Human intestinal macrophages display profound inflammatory anergy despite avid phagocytic and bacteriocidal activity.

J Clin Invest 115: 66–75 36 Iwasaki A and Medzhitov R (2004) Toll-like receptor control of the adaptive immune responses. Nat Immunol 5: 987–995 37 Jarnerot G et al. (2005) Infliximab as rescue therapy in severe to moderately severe ulcerative colitis: a randomized, placebo-controlled study.

Gastroenterology 128: 1805–1811 38 Neurath MF et al. (1996) Local administration of antisense phosphorothioate oligonucleotides to the p65 subunit of NF-κB abrogates established experimental colitis in mice. Nat Med 2: 998–1004 39 Egan LJ et al. (2004) IκB-kinaseβ-dependent NF-κB activation provides radioprotection to the intestinal epithelium. Proc Natl Acad Sci USA 101: 2452–2457 40 Chen LW et al. (2003) The two faces of IKK and NF-κB inhibition: prevention of systemic inflammation but increased local injury following intestinal ischemia– reperfusion. Nat Med 9: 575–581 41 Araki A et al. (2005) MyD88-deficient mice develop severe intestinal inflammation in dextran sodium sulfate colitis. J Gastroenterol 40: 16–23 42 Rakoff-Nahoum S et al. (2004) Recognition of commensal microflora by Toll-like receptors is required for intestinal homeostasis. Cell 118: 229–241 43 Sartor RB (2004) Animal models of intestinal inflammation. In Kirsner’s Inflammatory Bowel Diseases, 120–137 (Eds Sartor RB and Sandborn WJ) Philadelphia: Elsevier 44 Spencer DM et al. (2002) Distinct inflammatory mechanisms mediate early versus late colitis in mice.

Gastroenterology 122: 94–105

JULY 2006 VOL 3 NO 7 SARTOR

45 Bamias G et al. (2005) Proinflammatory effects of TH2 cytokines in a murine model of chronic small intestinal inflammation. Gastroenterology 128: 654–666 46 Kolls JK and Linden A (2004). Interleukin-17 family members and inflammation. Immunity 21: 467–476 47 Becker C et al. (2003) Constitutive p40 promoter activation and IL-23 production in the terminal ileum mediated by dendritic cells. J Clin Invest 112: 693–706 48 Fujino S et al. (2003) Increased expression of interleukin 17 in inflammatory bowel disease. Gut 52: 65–70 49 Schmidt C et al. (2005) Expression of interleukin-12related cytokine transcripts in inflammatory bowel disease: elevated interleukin-23p19 and interleukin27p28 in Crohn’s disease but not in ulcerative colitis.

Inflamm Bowel Dis 11: 16–23 50 Yen D et al. (2006) IL-23 is essential for T-cell mediated colitis and promotes inflammation via IL-17 and IL-6. J Clin Invest 116: 1310–1316 51 Monteleone G et al. (2005) Interleukin-21 enhances T-helper cell type I signaling and interferon-γ production in Crohn’s disease. Gastroenterology 128: 687–694 52 Neurath MF et al. (2002) The transcription factor T-bet regulates mucosal T cell activation in experimental colitis and Crohn’s disease. J Exp Med 195: 1129–1143 53 Fuss IJ et al. (1996) Disparate CD4+ lamina propria (LP) lymphokine secretion profiles in inflammatory bowel disease. Crohn’s disease LP cells manifest increased secretion of IFN-γ, whereas ulcerative colitis LP cells manifest increased secretion of IL-5.

J Immunol 157: 1261–1270 54 Fuss IJ et al. (2004) Nonclassical CD1d-restricted NK T cells that produce IL-13 characterize an atypical TH2 response in ulcerative colitis. J Clin Invest 113: 1490–1497 55 Hart AL et al. (2005) Characteristics of intestinal dendritic cells in inflammatory bowel diseases.

Gastroenterology 129: 50–65 56 Totsuka T et al. (2003) Ameliorating effect of anti-inducible co-stimulator monoclonal antibody in a murine model of chronic colitis. Gastroenterology 124: 410–421 57 Williams IR (2004) Chemokine receptors and leukocyte trafficking in the mucosal immune system.

Immunol Res 29: 283–292 58 Hosoe N et al. (2004) Demonstration of functional role of TECK/CCL25 in T lymphocyte-endothelium interaction in inflamed and uninflamed intestinal mucosa. Am J Physiol Gastrointest Liver Physiol 286: G458–G466 59 Connor SJ et al. (2004) CCR2 expressing CD4+ T lymphocytes are preferentially recruited to the ileum in Crohn’s disease. Gut 53: 1287–1294 60 Ghosh S et al. (2003) Natalizumab for active Crohn’s disease. N Engl J Med 348: 24–32 61 Feagan BG et al. (2005) Treatment of ulcerative colitis with a humanized antibody to the α4β7 integrin.

N Engl J Med 352: 2499–2507 62 Haller D et al. (2002) IKKβ and phosphatidylinositol 3-kinase/Akt participate in non-pathogenic Gram-negative enteric bacteria-induced RelA phosphorylation and NF-κB activation in both primary and intestinal epithelial cell lines. J Biol Chem 277: 38168–38178 63 Ruiz PA et al. (2005) IL-10 gene-deficient mice lack TGF β/Smad signaling and fail to inhibit proinflammatory gene expression in intestinal epithelial cells after the colonization with colitogenic Enterococcus faecalis. J Immunol 174: 2990–2999 64 Lodes MJ et al. (2004) Bacterial flagellin is a dominant antigen in Crohn’s disease. J Clin Invest 113: 1296–1306

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 405 ©2006 Nature Publishing Group

REVIEW www.nature.com/clinicalpractice/gasthep

65 Targan SR et al. (2005) Antibodies to CBir1 flagellin define a unique response that is associated independently with complicated Crohn’s disease.

Gastroenterology 128: 2020–2028 66 Hoentjen F et al. (2005) STAT3 regulates NF-κB recruitment to the IL-12 p40 promoter in dendritic cells. Blood 105: 689–696 67 Katakura K et al. (2005) Toll-like receptor 9-induced type I IFN protects mice from experimental colitis.

J Clin Invest 115: 695–702 68 Kim SC et al. (2005) Variable phenotypes of enterocolitis in interleukin 10-deficient mice monoassociated with two different commensal bacteria. Gastroenterology 128: 891–906 69 Sellon RK et al. (1998) Resident enteric bacteria are necessary for development of spontaneous colitis and immune system activation in interleukin-10deficient mice. Infect Immun 66: 5224–5231 70 Macpherson A et al. (1996) Mucosal antibodies in inflammatory bowel disease are directed against intestinal bacteria. Gut 38: 365–375 71 Brandwein SL et al. (1997) Spontaneously colitic C3H/HeJBir mice demonstrate selective antibody reactivity to antigens of the enteric bacterial flora.

J Immunol 159: 44–52 72 Mizoguchi E et al. (1997) Antineutrophil cytoplasmic antibodies in T-cell receptor alpha-deficient mice with chronic colitis. Gastroenterology 113: 1828–1835 73 Mizoguchi A (1997) Suppressive role of B cells in chronic colitis of T cell receptor alpha mutant mice.

J Exp Med 186: 1749–1756 74 Velazquez P et al. (2005) Surveillance B lymphocytes and mucosal immunoregulation. Springer Semin Immunopathol 26: 453–462 75 Cong Y et al. (2002) Bacterial-reactive T regulatory cells inhibit pathogenic immune responses to the enteric flora. J Immunol 169: 6112–6119 76 Veltkamp C et al. (2001) Continuous stimulation by normal luminal bacteria is essential for the development and perpetuation of colitis in Tg(ε26) mice. Gastroenterology 120: 900–913 77 Kim SC et al. (2005) Different host genetic backgrounds determine disease phenotypes induced by selective bacterial colonization [abstract].

Gastroenterology 128: A512 78 Bibiloni R et al. (2005) Analysis of the large bowel microbiota of colitic mice using PCR/DGGE. Lett Appl Microbiol 41: 45–51 79 Moran JP et al. (2006) Bifidobacterium animalis causes mild inflammatory bowel disease in interleukin-10 knockout mice [abstract].

Gastroenterology 130: A6 80 Holt L et al. (2005) Heat shock protein (HSP) 60 from Bacteroides vulgatus stimulates proinflammatory cytokine production by rat splenocytes [abstract].

Gastroenterology 128: A512 81 Sartor RB (2004) Microbial influences in inflammatory bowel disease: role in pathogenesis and clinical implications. In Kirsner’s Inflammatory Bowel Diseases, 138–162 (Eds Sartor RB and Sandborn WJ) Philadelphia: Elsevier 82 Kim SC et al. (2004) Dual-association of gnotobiotic IL-10–/– mice with two nonpathogenic commensal bacterial species accelerates colitis [abstract].

Gastroenterology 126: A291 83 Tsang J and Sartor RB (2005) Effect of diet on colitis in IL-10–/– mice [abstract]. Gastroenterology 128: A210 84 Dieleman LA et al. (2003) Lactobacillus GG prevents recurrence of colitis in HLA-B27 transgenic rats after antibiotic treatment. Gut 52: 370–376 85 Schultz M et al. (2002) Lactobacillus plantarum 299V in the treatment and prevention of spontaneous colitis

406 NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY ©2006 Nature Publishing Group

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

in interleukin-10 deficient mice. Inflamm Bowel Dis 8: 71–80 Sartor RB (2004) Therapeutic manipulation of the enteric microflora in inflammatory bowel diseases: antibiotics, probiotics and prebiotics.

Gastroenterology 126: 1620–1633 Bibiloni R et al. (2005) VSL#3 probiotic mixture induces remission in patients with active ulcerative colitis. Am J Gastroenterol 100: 1539–1546 Kuisma J et al. (2003) Effect of Lactobacillus rhamnosus GG on ileal pouch inflammation and microbial flora. Aliment Pharmacol Ther 17: 509–515 Gionchetti P et al. (2000) Oral bacteriotherapy as maintenance treatment in patients with chronic pouchitis: a double-blind, placebo-controlled trial.

Gastroenterology 119: 305–309 Berg DJ et al. (2002) Rapid development of colitis in NSAID-treated IL-10-deficient mice.

Gastroenterology 123: 1527–1542 Perl DP et al. (2004) Bacterial–metal interactions: the potential role of aluminum and other trace elements in the etiology of Crohn’s disease. Inflamm Bowel Dis 10: 881–883 Birrenbach T and Bocker U (2004) Inflammatory bowel disease and smoking: a review of epidemiology, pathophysiology, and therapeutic implications. Inflamm Bowel Dis 10: 848–859 Cosnes J (2004) Tobacco and IBD: relevance in the understanding of disease mechanisms and clinical practice. Best Pract Res Clin Gastroenterol 18: 481–496 Sartor RB (2005) Does Mycobacterium avium subspecies paratuberculosis cause Crohn’s disease?

Gut 54: 896–898 Chiodini RJ et al. (1984) Possible role of mycobacteria in inflammatory bowel disease. I. An unclassified Mycobacterium species isolated from patients with Crohn’s disease. Dig Dis Sci 29: 1073–1079 Shafran I et al. (2002) Open clinical trial of rifabutin and clarithromycin therapy in Crohn’s disease. Dig Liver Dis 34: 22–28 Millar D et al. (1996) IS900 PCR to detect Mycobacterium paratuberculosis in retail supplies of whole pasteurized cows’ milk in England and Wales. Appl Env Microbiol 62: 3446–3452 Naser SA et al. (2000) Isolation of Mycobacterium avium subsp paratuberculosis from breast milk of Crohn’s disease patients. Am J Gastroenterol 95: 1094–1095 Naser SA et al. (2004) Culture of Mycobacterium avium subspecies paratuberculosis from the blood of patients with Crohn’s disease. Lancet 364: 1039–1044 Autschbach F et al. (2005) High prevalence of Mycobacterium avium subsp. paratuberculosis IS900 DNA in gut tissues from individuals with Crohn’s disease. Gut 54: 944–999 Jubb TF and Galvin JW (2004) Effect of a test and control program for bovine Johne’s disease in Victorian dairy herds 1992–2002. Aust Vet J 82: 228–232 Raizman EA et al. (2004) The distribution of Mycobacterium avium ssp. paratuberculosis in the environment surrounding Minnesota dairy farms. J Dairy Sci 87: 2959–2966 Lund BM et al. (2002) Pasteurization of milk and the heat resistance of Mycobacterium avium subsp. paratuberculosis: a critical review of the data.

Int J Food Microbiol 77: 135–145 Darfeuille-Michaud A et al. (2004) High prevalence of adherent-invasive Escherichia coli associated with ileal mucosa in Crohn’s disease. Gastroenterology 127: 412–421 Peeters H et al. CARD15 variants determine a disturbed response of monocytes to adherent-

SARTOR JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/gasthep

106

107

108

109

110

111

112 113

114

115 116

117

118 119

120

121

122

123

invasive Escherichia coli LF82 in Crohn’s disease [abstract]. Ann NY Acad Sci, in press Liu Y et al. (1995) Immunocytochemical evidence of Listeria, Escherichia coli, and Streptococcus antigens in Crohn’s disease. Gastroenterology 108: 1396–1404 Darfeuille-Michaud A (2002) Adherent-invasive Escherichia coli: a putative new E. coli pathotype associated with Crohn’s disease. Int J Med Microbiol 292: 185–193 Huycke MM et al. Superoxide production by Enterococcus faecalis upregulates NTN1 oncogene in the colonic epithelium and alters inflammationassociated gene expression [abstract]. Ann NY Acad Sci, in press Erichsen K et al. (2005) Low-dose oral ferrous fumarate aggravated intestinal inflammation in rats with DSS-induced colitis. Inflamm Bowel Dis 11: 744–748 Lerner A et al. (2006) Bacterial–aluminum interactions.

The role of aluminum in bacterial-induced colitis in IL-10 deficient mice [abstract]. Gastroenterology 130: A362 Summers RW et al. (2005) Trichuris suis therapy for active ulcerative colitis: a randomized controlled trial. Gastroenterology 128: 825–832 Summers RW et al. (2005) Trichuris suis therapy in Crohn’s disease. Gut 54: 87–90 Elliott DE et al. (2004) Heligmosomoides polygyrus inhibits established colitis in IL-10-deficient mice.

Eur J Immunol 34: 2690–2698 Korzenik JR and Dieckgraefe BK (2000) Is Crohn’s disease an immunodeficiency? A hypothesis suggesting possible early events in the pathogenesis of Crohn’s disease. Dig Dis Sci 45: 1121–1129 Korzenik JR et al. (2005) Sargramostim for active Crohn’s disease. N Engl J Med 352: 2193–2201 Powrie F (2004) Immune regulation in the intestine: a balancing act between effector and regulatory T cell responses. Ann NY Acad Sci 1029: 132–141 Elson CO et al. (2004) Gene disruption and immunity in experimental colitis. Inflamm Bowel Dis 10 (Suppl 1): S25–S28 Allez M and Mayer L (2004) Regulatory T cells: peace keepers in the gut. Inflamm Bowel Dis 10: 666–676 Bouma G and Strober W (2003). The immunological and genetic basis of inflammatory bowel disease.

Nat Rev Immunol 3: 521–533 Fuss IJ et al. (2002) The interrelated roles of TGF-β and IL-10 in the regulation of experimental colitis.

J Immunol 168: 900–908 Kraus TA et al. (2004) Failure to induce oral tolerance to a soluble protein in patients with inflammatory bowel disease. Gastroenterology 126: 1771–1778 Monteleone G et al. (2001) Blocking Smad7 restores TGF-β1 signaling in chronic inflammatory bowel disease. J Clin Invest 108: 601–609 Powrie F et al. (1994) Regulatory interactions between CD45RBhigh and CD45RBlow CD4+ T cells are important for the balance between protective and

JULY 2006 VOL 3 NO 7 SARTOR

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

pathogenic cell-mediated immunity. J Exp Med 179: 589–600 Mottet C et al. (2003) Cutting edge: cure of colitis

by CD4+CD25+ regulatory T cells. J Immunol 170: 3939–3943 Albright C et al. (2002) Endogenous IL-10 inhibits APC stimulation of T lymphocyte responses to luminal bacteria [abstract]. Gastroenterology 122: A270 Liu B et al. (2006) IL-10 production by both antigen presenting cells and CD4+ T lymphocytes maintains mucosal homeostasis [abstract]. Gastroenterology 130: A232 Maeda S et al. (2005) Nod2 mutation in Crohn’s disease potentiates NF-κB activity and IL-1β processing. Science 307: 734–738 Rachmilewitz D et al. (2004) Toll-like receptor 9 signaling mediates the anti-inflammatory effects of probiotics in murine experimental colitis.

Gastroenterology 126: 520–528 Anant S et al. Plasmacytoid DCs are critical elements in the response to GM-CSF in the DSS colitis model [abstract]. Ann NY Acad Sci, in press Beckwith J et al. (2005) Cdcs1, a major colitogenic locus in mice, regulates innate and adaptive immune response to enteric bacterial antigens.

Gastroenterology 129: 1473–1484 Itoh J et al. (2001) Decreased Bax expression by mucosal T cells favours resistance to apoptosis in Crohn’s disease. Gut 49: 35–41 Atreya R et al. (2000) Blockade of interleukin 6 trans signaling suppresses T-cell resistance against apoptosis in chronic intestinal inflammation: evidence in Crohn’s disease and experimental colitis in vivo.

Nat Med 6: 583–588 Peppelenbosch MP and van Deventer SJ (2004) T cell apoptosis and inflammatory bowel disease.

Gut 53: 1556–1558 Tiede I et al. (2003) CD28-dependent Rac1 activation is the molecular target of azathioprine in primary human CD4+ T lymphocytes. J Clin Invest 111: 1133–1145 Doering J et al. (2004) Induction of T lymphocyte apoptosis by sulphasalazine in patients with Crohn’s disease. Gut 53: 1632–1638 Lugering A et al. (2001) Infliximab induces apoptosis in monocytes from patients with chronic active Crohn’s disease by using a caspase-dependent pathway. Gastroenterology 121: 1145–1157 ten Hove T et al. (2002) Infliximab treatment induces apoptosis of lamina propria T lymphocytes in Crohn’s disease. Gut 50: 206–211 Van den Brande JM et al. (2003) Infliximab but not etanercept induces apoptosis in lamina propria T-lymphocytes from patients with Crohn’s disease.

Gastroenterology 124: 1774–1785 Selby W (2005) Antibiotic therapy for Crohn's disease. Presented at 9th Annual Coolum Update in Gastroenterology and Hepatology: 2005 June 11–13; Coolum

Acknowledgments Original research was supported by NIH RO1 grants DK40249 and DK 53347. The author thanks S May for her expert secretarial assistance.

Competing interests The author has declared associations with the following companies/ organizations: Berlex, Danone/Yakult, Procter & Gamble, Salix Pharmaceuticals and VSL Pharmaceuticals. See the article online for full details of these relationships.

NATURE CLINICAL PRACTICE GASTROENTEROLOGY & HEPATOLOGY 407 ©2006 Nature Publishing Group