Intestinal homeostasis and its breakdown in inflammatory bowel disease

Journal name:
Nature
Volume:
474,
Pages:
298–306
Date published:
DOI:
doi:10.1038/nature10208
Published online

Abstract

Intestinal homeostasis depends on complex interactions between the microbiota, the intestinal epithelium and the host immune system. Diverse regulatory mechanisms cooperate to maintain intestinal homeostasis, and a breakdown in these pathways may precipitate the chronic inflammatory pathology found in inflammatory bowel disease. It is now evident that immune effector modules that drive intestinal inflammation are conserved across innate and adaptive leukocytes and can be controlled by host regulatory cells. Recent evidence suggests that several factors may tip the balance between homeostasis and intestinal inflammation, presenting future challenges for the development of new therapies for inflammatory bowel disease.

At a glance

Figures

  1. Bacterial sensing and cellular stress pathways in intestinal homeostasis.
    Figure 1: Bacterial sensing and cellular stress pathways in intestinal homeostasis.

    a, Bacterial sensing and handling pathways cooperatively maintain the intestinal epithelial barrier. In IECs, basal sensing of microbial pathogen-associated molecular patterns by PRRs such as TLRs maintains epithelial barrier function by stimulating AMP expression (green arrows), and by fortifying tight junctions and inducing the release of protective cytokines such as IL-18 (red arrows). In addition, intrinsic PRR signals in IECs stimulate anti-apoptotic and proliferative responses (black arrows). The autophagy pathway, induced by PRR signals or by the ER stress response (blue arrows), cooperates with PRRs to promote the secretion of AMP and mucins (green arrows), and also constitutes an important cell-intrinsic defence mechanism in IECs for bacterial clearance. Therefore, defects in PRR sensing, the UPR or the autophagy pathways can result in impaired barrier function, leading to increased bacterial colonization and translocation and eliciting an exacerbated inflammatory response. b, Aberrant PRR signals in haematopoeitic cells drive chronic inflammation in IBD. PRR signals in dendritic cells and macrophages drive chronic inflammatory responses in the gut through the activation of NF-κB-dependent pro-inflammatory cytokines (such as IL-23, IL-6 and TNF-α) and caspase-1 (CASP1)-mediated induction of IL-1β, IL-18 and pyroptosis (red arrows). The magnitude of these inflammatory responses may be tempered by NOD2 suppression of TLR signals (green line) and stimulation of the autophagy pathway by NOD2, TLRs or ROS induction (blue arrows). Autophagy may attenuate ROS production and CASP1 activation (black lines), thus limiting inflammatory responses. Therefore, defects in NOD2 or autophagy pathways may contribute to the excessive inflammatory responses observed in IBD.

  2. Conserved innate and adaptive immune effector modules in the gut.
    Figure 2: Conserved innate and adaptive immune effector modules in the gut.

    IECs and intestinal dendritic cells sense distinct infectious agents, leading to the production of factors that direct different effector responses (black arrows). Local innate leukocyte populations can rapidly produce these effector cytokines to restrict pathogen growth until specific adaptive responses have been induced. Cells and molecules associated with distinct effector profiles are colour-coded; TH1 type (blue), TH17 type (red), mixed TH1 and TH17 (brown), TH2 type (purple), and aberrant or prolonged expression of any of these modules may contribute to chronic intestinal inflammation. Regulatory T-cell circuits (green) can suppress all types of inflammatory effector response and may enhance the production of protective secretory IgA (sIgA) antibodies. 17+γ+, IL-17- and IFN-γ-secreting CD4+ T cell; ILC2, type-2 ILC; NK, natural killer; RA, retinoic acid; TSLP, thymic stromal lymphopoietin.

  3. A multihit model of IBD pathogenesis.
    Figure 3: A multihit model of IBD pathogenesis.

    The induction and perpetuation of chronic intestinal pathology may require the convergence of many abnormalities that affect several overlapping layers of immune homeostasis in the intestine. Defects in one layer are unlikely to precipitate IBD in the absence of further pathogenic lesions (panels on left). These layers are in turn controlled by a range of cell-intrinsic and cell-extrinsic regulatory modules that function in an interrelated way to maintain homeostasis (panels on right). Defects in these homeostatic modules may predispose people to the development of IBD by affecting several layers of immune homeostasis (filled panels). The dashed line denotes the threshold at which the level of inflammation manifests as IBD.

Inflammatory bowel disease (IBD) refers to chronic inflammatory disorders that affect the gastrointestinal tract1. There are two main clinical forms of IBD — Crohn's disease, which can affect any part of the gastrointestinal tract, and ulcerative colitis, in which pathology is restricted to the colonic mucosa1. The precise aetiology of IBD remains unclear, but several factors that make a major contribution to disease pathogenesis have been identified1. These fall into three distinct categories: genetic factors, the host immune system, and environmental factors such as the gut microbiota, which is dominated by intestinal bacteria2.

On a cellular level, the dynamic crosstalk between intestinal epithelial cells (IECs), intestinal microbes and local immune cells represents one of the fundamental features of intestinal homeostasis3, 4. These interactions are not only important for the pathogenesis of IBD, but also essential for maintaining normal intestinal homeostasis and for mounting protective immunity to pathogens. In this Review, we summarize recent findings from disease models and clinical samples that illustrate key interactions and pathways that regulate intestinal homeostasis. We discuss how defects in epithelial barrier function, innate immune recognition or immune regulatory circuits may precipitate the aberrant expression of pathological inflammatory responses in IBD. Finally, we offer some perspectives on future challenges for developing therapies for IBD.

Regulation by the epithelial barrier

The intestinal epithelium represents a huge surface area of approximately 100 m2 that is lined by a single layer of columnar IECs, which form a stout physical barrier. IECs, however, form much more than a simple physical barrier that processes and absorbs dietary nutrients. They perform several other functions that are crucial for intestinal homeostasis3. These include secretion of compounds that influence microbial colonization, sampling of the intestinal microenvironment, sensing of both beneficial and harmful microbes, and induction and modulation of immune responses. To fulfil such diverse functions, the intestinal epithelium has unique anatomical and cellular adaptations, and IECs comprise several specialized cell types with distinct functions3, 4. In addition, IECs do not regulate intestinal homeostasis in a solely intrinsic fashion, but instead function as part of a coordinated response to signals provided by the commensal microbiota and from local leukocyte populations3, 4.

The mucosal surfaces of the body, including the intestine, are coated by a thick mucus gel comprising an outer layer of secreted mucins overlying a dense inner glycocalyx of membrane-anchored mucins that is inaccessible to most bacteria3, 4. In addition to providing a biophysical barrier, mucus forms a matrix that allows the retention of high concentrations of antimicrobial molecules, such as defensins and secretory IgA, close to the epithelial surface. The mucus layer has a crucial role in intestinal homeostasis, as decreased levels of goblet cells, leading to reduced mucin secretion, are a hallmark of human IBD, and mice lacking the major mucin protein MUC2 develop spontaneous colitis1, 3, 4.

IECs also regulate the colonization and penetration of the epithelium by luminal microbes through the secretion of antimicrobial peptides (AMPs), which include lysozymes, defensins, cathelicidins, lipocalins and C-type lectins such as RegIIIγ5. Although some AMPs are produced constitutively by many IECs, others are secreted in an inducible fashion by Paneth cells — a type of specialized IEC located at the base of the intestinal crypts of the small intestine. In mice, Paneth cells have been implicated in protection against intestinal pathogens, as well as in limiting colonization by commensal bacteria6. Several lines of evidence suggest that Paneth cell dysfunction and impaired defensin secretion may contribute to IBD susceptibility. Patients with ileal Crohn's disease or those with NOD2 (also known as CARD15)-susceptibility alleles had reduced α-defensin expression, and genetic variants in the transcription factor TCF4, which is involved in Paneth cell maturation and function, have recently been associated with ileal Crohn's disease5.

Autophagy and ER stress control epithelial homeostasis

Paneth cell abnormalities have also been reported in patients with Crohn's disease who are homozygous for the T300A disease-risk allele of the autophagy gene Atg16l1, and in mice rendered hypomorphic for the ATG16L1 protein7. Autophagy is a fundamental process that controls the catabolism of intracellular constituents in response to stress or infection, characteristically involving the formation of autophagosomes that target cargo for degradation by the lysosomal machinery8. Autophagy affects host immune responses on several levels and has a vital role in cell-intrinsic defence against intracellular infections8. Defects in ATG16L1 led to the accumulation of morphologically abnormal granules in Paneth cells, suggesting that the secretory granule pathway is impaired, although ATG16L1-deficient Paneth cells also expressed higher levels of pro-inflammatory mediators7.

Similarly, several recent studies have identified a link between endoplasmic reticulum (ER) stress and the consequent unfolded protein response (UPR) and IBD9. The maintenance of functional secretory cells requires coordinated protein folding and trafficking of secretory proteins by the ER–Golgi network. The UPR is elicited by the accumulation of unfolded or misfolded proteins in the ER, thus highly secretory cells, such as goblet cells and Paneth cells, are very susceptible to ER stress, and a functional UPR is required to maintain epithelial homeostasis in the gut9. Mice containing a genetic deletion in the IEC-restricted isoform of the UPR effector enzyme IRE1 showed increased susceptibility to colitis induced by dextran sulphate sodium (DSS) administration. Activation of IRE1 results in splicing and activation of the transcription factor XBP1, and mice with a conditional Xbp1 deletion in IECs developed spontaneous enteritis that showed many characteristic features of human IBD10. Deletion of Xbp1 in mouse IECs led to a loss of Paneth cells, a significant reduction in goblet cells and hyper-responsiveness of IECs to pro-inflammatory signals10. Other genetic lesions that result in increased ER stress in the intestinal epithelium also predispose individuals to intestinal inflammation9.

The ER stress pathway is also relevant to human gastrointestinal diseases, as increased ER stress is observed in the intestinal epithelium of patients with IBD, and polymorphisms within ER stress response genes, including XBP1, AGR2 and ORMDL3, are associated with susceptibility to both Crohn's disease and ulcerative colitis9. The degree of ER stress within the intestinal epithelium may be modulated by environmental factors derived from the host or from the intestinal microbiota. Pro-inflammatory conditions and cytokines such as tumour necrosis factor-α (TNF-α) exacerbate ER stress, whereas the anti-inflammatory cytokine interleukin-10 (IL-10) reduces it9. Furthermore, there is evidence to suggest that the ER acts as a source for the autophagosome membrane and that the UPR can activate autophagy9, indicating that these processes cooperate during infection or stress of the intestinal epithelium. Taken together, these studies indicate that defects in specialized secretory IEC populations, or aberrant ER stress or autophagy responses in IECs, greatly predispose people to the development of intestinal inflammation.

IECs may also influence intestinal homeostasis through the secretion of conditioning cytokines that affect adaptive responses primed by intestinal dendritic cells3. In the healthy intestine, these conditioning factors help to maintain a state of hyporesponsiveness towards commensal bacteria. For example, cytokines constitutively expressed by IECs, such as thymic stromal lymphopoietin and IL-25, limit dendritic-cell production of the p40 subunit of IL-12 and IL-23 and promote IL-10 secretion, impeding the priming of T helper 1 (TH1)-cell responses and instead favouring the induction of T regulatory (Treg)-cell and TH2-cell responses3, 11. Conversely, after sensing pathogenic invasion or damage, IECs can elaborate the secretion of pro-inflammatory chemokines, such as IL-8 (also known as CXCL8), which have an important role in alerting the immune system to microbial attack3, 12. IECs also exert a strong influence on local antibody responses by producing factors such as transforming growth factor-β (TGF-β), B-cell activating factor (BAFF, also known as TNFSF13B) and a proliferation-inducing ligand (APRIL, also known as TNFSF13), which promote class-switching of B cells towards the production of IgA13. IECs mediate the transport of secretory IgA into the mucus layer, where it has a complementary role to innate defences in limiting the penetration of commensal bacteria across the epithelium13, 14.

Pattern recognition receptors and intestinal homeostasis

Accumulating evidence indicates that microbial sensing through pattern recognition receptors (PRRs) drives complementary functions in IECs and haematopoeitic cells, which together control intestinal homeostasis12, 15, 16. The context of PRR activation is crucial. In the healthy intestine, basal PRR activation maintains barrier function and commensal composition, but aberrant PRR signalling may be a central contributor to the pathophysiology of IBD. The latter point is emphasized by genetic-association studies linking PRR genes, including NOD2, NLRP3 and various Toll-like receptor (TLR) genes, with IBD susceptibility, although the mechanisms responsible remain unclear12, 16.

Tonic PRR signals maintain a healthy epithelium

The importance of TLR signalling in regulating epithelial barrier function has been shown by studies using the DSS-colitis mouse model, in which DSS administration results in chemical destruction of the IEC layer and penetration of commensal bacteria, leading to acute colitis followed by restitution and repair of the epithelial barrier. Mice that lack specific TLRs, such as Tlr2, Tlr4, Tlr5 or Tlr9, or that are deficient in the shared TLR signalling adaptor protein MyD88 show increased susceptibility to DSS colitis, characterized by defective tissue repair and/or increased mortality12, 16. TLR signals drive intrinsic protective effects in IECs by inducing several proliferative and anti-apoptotic factors and by promoting epithelial restitution and fortifying intercellular tight junctions12, 16. Intrinsic TLR signals in IECs also have a central role in limiting bacterial colonization and translocation by stimulating IEC production of AMPs, such as defensins and RegIIIγ6, 15. An elegant study in which MyD88 expression was selectively limited to Paneth cells showed that these cells sense commensal bacteria directly through TLRs, and that this sensing induced AMP production that limited bacterial translocation across the intestinal mucosa6.

The activation of cytosolic NOD-like receptors (NLRs) may also be important in maintaining barrier function, as mice lacking Nod1 or Nod2, or bearing a Crohn's-disease-associated mutant allele of human NOD2, showed defects in defensin secretion and increased susceptibility to DSS colitis17, 18. Recent studies have indicated that NLR-mediated inflammasome activation also contributes to protection after damage of the epithelium, because mice deficient in NLRP3, its adaptor ASC or caspase-1 showed enhanced colitis and mortality after DSS administration19, 20, 21. Inflammasomes are multimolecular complexes that activate caspase-1. They are formed after the activation of cytosolic NLRs, which then associate with caspase-1, often through interactions with adaptor proteins such as ASC22. Activated caspase-1 has a central involvement in the processing and secretion of two key pro-inflammatory cytokines, IL-1β and IL-18, which in turn bind to receptors that use MyD88 as a signal-transduction adaptor22. The administration of exogenous IL-18 attenuated DSS colitis in Casp1−/− mice (which lack caspase-1)19, 20, and Il18−/− mice and Il18r1−/− mice (which lack the IL-18 receptor) showed increased susceptibility to DSS colitis, highlighting a role for IL-18 in epithelial barrier integrity and repair23, 24.

Sustained PRR signals drive intestinal inflammation

In contrast to the protective barrier responses elicited by tonic PRR signals in IECs, studies in mouse models of chronic colitis have demonstrated that sustained PRR activation drives chronic intestinal inflammation. Thus, MyD88 signals were required for spontaneous colitis development in Il10−/− mice12, 16. A key role for MyD88 signals in haematopoietic cells was indicated by experiments in which selective ablation of MyD88 rendered mice refractory to intestinal inflammation induced by Helicobacter hepaticus15. Although these results do not exclude a contribution from PRR signals in IECs to inflammatory responses in the gut, they indicate that such responses alone are insufficient to drive chronic inflammatory pathology. Similarly, recent studies have demonstrated that T-cell-intrinsic MyD88 signals are not essential for the expression of pathogenic effector or regulatory functions in the intestine15, whereas MyD88 signals in dendritic cells during the sensing of intestinal microbiota were shown to be essential for T-cell proliferation and intestinal pathology25.

Patients with IBD, particularly those with ulcerative colitis, have a greatly increased risk of developing colitis-associated cancer. Studies of the role of PRR signals in colitis-associated cancer, however, have produced conflicting results26, 27. Although sustained PRR-driven inflammatory responses can exacerbate intestinal tumorigenesis, their role in epithelial barrier maintenance and repair is protective against the development of intestinal tumours26, 27. It is also worth noting that a few studies have reported protective effects of PRR signals against chronic intestinal inflammation, again emphasizing that some PRR signalling can be beneficial in the gut. For example, bacterial polysaccharide A derived from the human commensal bacterium Bacteroides fragilis was able to protect mice from T-cell-mediated colitis in a TLR2-dependent manner through the induction of Treg cells28.

Integration of bacterial handling and stress responses

Microbes trigger a diverse range of PRRs and cellular stress responses that do not operate in isolation, but are integrated by the cell to direct appropriate effector responses. Thus, defects in one PRR pathway may influence other PRR signalling cascades, as well as affecting other processes implicated in intestinal homeostasis, such as autophagy and ER stress.

For example, Crohn's-disease-associated mutations in the NOD2 gene are mainly located in the leucine-rich-repeat region that mediates sensing of the peptidoglycan motif muramyl dipeptide (MDP), resulting in reduced activation of nuclear factor-κB (NF-κB)17. It has been proposed that NOD2 could act by attenuating TLR signalling that drives excessive activation of dendritic cells and TH1-cell responses17. Further evidence that NOD2 can regulate TLR signalling in the gut came from a mouse model of necrotizing enterocolitis, in which activation of NOD2 by MDP inhibited TLR4 signalling in IECs and ameliorated the condition29.

There is an increasing appreciation that PRR signals intersect with other bacterial-handling and cellular-stress processes to coordinate protective and inflammatory responses. For example, TLR signals can induce autophagy, and this has been reported to enhance the clearance of microbes8. Recent studies have found that NOD2 stimulates autophagy by interacting directly with ATG16L1, which allows the recruitment of ATG16L1 to sites of bacterial entry30, 31. Furthermore, dendritic cells expressing Crohn's-disease-associated mutant forms of NOD2 or ATG16L1 showed reduced autophagy in response to MDP, and this led to impaired antigen presentation and bacterial killing31.

The regulation of inflammasome activation by autophagic pathways has been reported, with selective ablation of ATG16L1 in mouse haematopoietic cells leading to increased inflammasome activation and IL-1β secretion in response to lipopolysaccharide32. These mice were also highly susceptible to DSS colitis, which was reversed by the neutralization of IL-1 and IL-18 (ref. 32). The mechanism involved is not clear, but autophagy has been reported to inhibit the generation of reactive oxygen species (ROS), especially by dysfunctional mitochondria, which have been shown to trigger the activation of NLRP3 inflammasomes33. Autophagy may also inhibit pyroptosis, a highly inflammatory form of caspase-1-dependent cell death that has been observed in myeloid cells infected with intracellular pathogens8. In addition, ROS generation has been shown to stimulate autophagy that restricted the replication of Salmonella enterica serovar Typhimurium (S. Typhimurium) in IECs, suggesting that ROS-induced autophagy may act as a negative-feedback mechanism to limit caspase-1-driven inflammatory circuits, while providing a complementary mechanism of bacterial defence22. Recent studies have also reported interactions between TLR and ER stress pathways, although both inhibition and activation of distinct arms of the UPR have been observed after TLR stimulation9, 34.

Taken together, these studies indicate that diverse PRR signals interact with the autophagy and ER stress pathways to coordinate bacterial handling and inflammatory responses, and suggest that deficiencies or perturbations of these networks could contribute to IBD pathogenesis (Fig. 1). A better understanding of how these networks function in IECs and leukocytes in the healthy and inflamed intestine may give rise to new therapeutic avenues for IBD, and could also reveal strategies for boosting mucosal barrier defences in immune-suppressed individuals.

Figure 1: Bacterial sensing and cellular stress pathways in intestinal homeostasis.
Bacterial sensing and cellular stress pathways in intestinal homeostasis.

a, Bacterial sensing and handling pathways cooperatively maintain the intestinal epithelial barrier. In IECs, basal sensing of microbial pathogen-associated molecular patterns by PRRs such as TLRs maintains epithelial barrier function by stimulating AMP expression (green arrows), and by fortifying tight junctions and inducing the release of protective cytokines such as IL-18 (red arrows). In addition, intrinsic PRR signals in IECs stimulate anti-apoptotic and proliferative responses (black arrows). The autophagy pathway, induced by PRR signals or by the ER stress response (blue arrows), cooperates with PRRs to promote the secretion of AMP and mucins (green arrows), and also constitutes an important cell-intrinsic defence mechanism in IECs for bacterial clearance. Therefore, defects in PRR sensing, the UPR or the autophagy pathways can result in impaired barrier function, leading to increased bacterial colonization and translocation and eliciting an exacerbated inflammatory response. b, Aberrant PRR signals in haematopoeitic cells drive chronic inflammation in IBD. PRR signals in dendritic cells and macrophages drive chronic inflammatory responses in the gut through the activation of NF-κB-dependent pro-inflammatory cytokines (such as IL-23, IL-6 and TNF-α) and caspase-1 (CASP1)-mediated induction of IL-1β, IL-18 and pyroptosis (red arrows). The magnitude of these inflammatory responses may be tempered by NOD2 suppression of TLR signals (green line) and stimulation of the autophagy pathway by NOD2, TLRs or ROS induction (blue arrows). Autophagy may attenuate ROS production and CASP1 activation (black lines), thus limiting inflammatory responses. Therefore, defects in NOD2 or autophagy pathways may contribute to the excessive inflammatory responses observed in IBD.

Pathological effector modules in the gut

There are a multitude of animal models of IBD that either arise spontaneously or are induced by various experimental manipulations, which reproduce distinct features of human IBD. However, there is no perfect experimental model, because patients with IBD present a heterogeneous spectrum of pathological features that reflect the participation of a diverse range of innate and adaptive immune effectors. This heterogeneity is further underscored by the recent observations that around 100 distinct genetic loci may contribute to IBD susceptibility (see page 307), and the key target of these aberrant immune responses, the gut microbiota, is unique to each individual2. It is therefore likely that there will be several aetiologies of human IBD, and these may reflect aberrant expression of distinct immune modules. In this context, an immune module is used to define an effector response coordinated by a group of cytokines that may be produced by innate and/or adaptive leukocytes. These distinct immune modules evolved to protect against the different types of challenge posed by diverse pathogens that target the gastrointestinal tract, and some immune pathology associated with their expression is one 'cost' of a functional immune system (Fig. 2).

Figure 2: Conserved innate and adaptive immune effector modules in the gut.
Conserved innate and adaptive immune effector modules in the gut.

IECs and intestinal dendritic cells sense distinct infectious agents, leading to the production of factors that direct different effector responses (black arrows). Local innate leukocyte populations can rapidly produce these effector cytokines to restrict pathogen growth until specific adaptive responses have been induced. Cells and molecules associated with distinct effector profiles are colour-coded; TH1 type (blue), TH17 type (red), mixed TH1 and TH17 (brown), TH2 type (purple), and aberrant or prolonged expression of any of these modules may contribute to chronic intestinal inflammation. Regulatory T-cell circuits (green) can suppress all types of inflammatory effector response and may enhance the production of protective secretory IgA (sIgA) antibodies. 17+γ+, IL-17- and IFN-γ-secreting CD4+ T cell; ILC2, type-2 ILC; NK, natural killer; RA, retinoic acid; TSLP, thymic stromal lymphopoietin.

TNF-α and IL-6

The central role of innate myeloid cells in IBD is mirrored by the potent pro-inflammatory effects of the cytokines that they secrete, particularly TNF-α and IL-6. The successful application of anti-TNF-α antibody treatment signified a major breakthrough in the treatment of IBD35, and resulted directly from convergent data in animal IBD models indicating a role for TNF-α in chronic intestinal inflammation1. IL-6 is increased in the inflamed intestinal mucosa, and blockade of IL-6 signalling ameliorated colitis in mouse models and also had beneficial effects in a clinical trial of patients with Crohn's disease1. Despite the success of anti-TNF-α biologics in IBD, approximately one-third of patients do not respond to anti-TNF-α treatment, and many others eventually lose responsiveness or become intolerant to these agents35. In addition, patients treated with anti-TNF-α show an increased incidence of severe infections and malignancies, emphasizing the need for further therapies that may target intestinal inflammation more selectively35.

IL-23, TH1 and TH17 responses

Early studies on the production of T-cell-derived cytokines suggested a role for IL-13-producing natural killer T (NKT) cells in ulcerative colitis, and the differential activation of IL-12 p40 and TH1-cell responses was associated with Crohn's disease1. The colitis-attenuating effects observed in mice lacking the Il12b (also known as Il-12p40) gene, or given neutralizing antibodies directed against IL-12 p40 or interferon-γ (IFN-γ), also emphasized a key role for TH1-cell responses in intestinal inflammation1. However, the discovery that another IL-12-p40-containing heterodimeric cytokine, IL-23, was the central driver in several autoimmune pathologies prompted analysis of the role of IL-23 in intestinal inflammation. Studies in several mouse IBD models have used selective targeting of the IL-23 p19 subunit to demonstrate that IL-23 plays a key part in chronic intestinal pathology36. These findings were quickly followed by genome-wide association studies (GWAS) that reported strong associations of polymorphisms in the IL23R and IL12B gene loci with Crohn's disease and ulcerative colitis (see page 307). IL-23 is induced by PRR stimulation and is constitutively expressed in a small population of dendritic cells present in the lamina propria of the terminal ileum, although in patients with Crohn's disease, CD14+ intestinal macrophages have also been reported to secrete large amounts of IL-23 (ref. 37). The factors that determine whether an activated dendritic cell will preferentially produce IL-23 or IL-12 are not clear, but a recent study showed that ER stress and activation of the UPR can synergize with TLR signals to selectively increase IL-23 expression by dendritic cells38. Although IL-23 was initially linked to the preferential expression of TH17 responses, it can promote a wide range of pathological responses in the intestine, mediated either by T cells or by excessive innate immune activation36, 39. IL-23-mediated enhancement of TH1 and TH17 responses is consistent with the increased levels of IFN-γ, IL-17 and IL-22 observed in the chronically inflamed intestine36, 39. Transcription factors that direct TH1-cell or TH17-cell responses — such as T-bet (also known as T-box protein 21 and TBX21) or retinoic-acid-receptor-related orphan receptor-γt (RORγt), respectively — were shown to be essential for T-cell-mediated colitis1, 40. T-cell-intrinsic IL-23R signals favour the expression of pathogenic pro-inflammatory T-cell responses in several ways, including enhanced proliferation of effector T cells, reduced differentiation of FOXP3+ Treg cells and the emergence of IL-17+IFN-γ+CD4+ T cells39. The precise origins, stability and pathogenic properties of these IL-17+IFN-γ+CD4+ T cells remain to be determined. Notably, IL-17+IFN-γ+CD4+ T cells have been isolated from the inflamed lamina propria of patients with Crohn's disease, and were shown to respond to IL-23 and to be derived from a discrete subset of CD161+CD4+ T cells that express chemokine receptor 6 (CCR6) and the nuclear receptor RORγt41, 42. In phase II clinical trials, anti-IL-12-p40 monoclonal antibodies have shown clinical efficacy in a subset of patients with Crohn's disease, particularly in those who had not responded to anti-TNF-α therapy, suggesting that TNF-α and IL-12 (or IL-23) drive distinct pathways of immune pathology35.

The relative enrichment of TH17 cells at mucosal sites, together with the increased levels of TH17 cytokines in the inflamed gut, has fuelled interest in their potential role in IBD pathogenesis36, 39. TH17 cells produce several cytokines, including IL-17A, IL-17F, IL-21 and IL-22 (ref. 36). Many studies have focussed on the roles of IL-17A and IL-17F, which are known to have pro-inflammatory effects in tissues such as the lung and brain, through the elaboration of cytokines and chemokines, particularly those that promote neutrophil recruitment36. Analyses of IL-17A and IL-17F in mouse colitis models have produced conflicting results. In acute DSS colitis, IL-17A has a protective role, whereas IL-17F seems to exacerbate disease43. By contrast, the neutralization of IL-17A attenuated chronic colitis in mice with Stat3-deficient Treg cells44 and decreased innate immune colitis after H. hepaticus infection45. Studies in T-cell-transfer colitis models suggest that IL-17A and IL-17F can have redundant pro-inflammatory effects in the gut40. TH17 responses were also recently implicated in a model of colitis-associated cancer, because IL-17A depletion reduced colitis and tumour development46. The microbiota has an important role in the preferential localization of TH17 cells in the gut, as the colonization of germ-free mice with segmented filamentous bacteria led to the marked accumulation of TH17 cells in the intestinal lamina propria47, 48.

Another TH17 cytokine that is highly dependent on IL-23 is IL-22, which enhances the innate immunity of tissues. Expression of the IL-22R complex is restricted to non-haematopoietic cells, especially epithelial cells in the skin, gut and lungs49. IL-22 signalling in IECs drives the production of AMPs and also promotes epithelial regeneration and healing by activating the transcription factor STAT3 (ref. 50). Consistent with this epithelial-protective role, IL-22 administration attenuated disease severity in the DSS and T-cell receptor-α (Tcra−/−) mouse IBD models, by restoring goblet cells and mucus production49. By contrast, other studies support a pathogenic role for IL-22 in IBD, as its expression is increased in patients with Crohn's disease, and high serum IL-22 levels correlate with increased disease activity and susceptibility-associated IL23R polymorphisms49. A recent study of bleomycin-induced airway inflammation showed that IL-22 could mediate either tissue-protective or pathogenic functions, depending on the absence or presence of IL-17A, respectively51. Thus, further studies are required to determine whether IL-17A or other pro-inflammatory mediators can have similar modulating effects on IL-22 activity in the gut. Although less extensively studied, IL-21 may also regulate intestinal inflammation, through effects on TH17 cells and the production of matrix metalloproteinases, which are involved in tissue remodelling36. In summary, although it is clear that TH17 cytokines are important in many aspects of intestinal homeostasis and protection from mucosal pathogens, their role in IBD pathogenesis remains ambiguous, and further investigations are necessary to clarify their potential for therapeutic intervention.

NLRs and inflammasome-associated cytokines

Recent advances identifying a central role for inflammasomes and NLRs in autoinflammatory diseases22 — together with the association of IBD with polymorphisms in NLRP3 and IL18RAP — have rekindled interest in the potential roles of IL-1β and IL-18 in IBD. Levels of IL-1β and IL-18 are increased in IBD1, 23, and Il18−/− mice were resistant to colitis induced by trinitrobenzene sulphonic acid24, suggesting that IL-1β and IL-18 enhance chronic intestinal pathology. This hypothesis is consistent with the ability of IL-1β and IL-18 to promote TH17 and TH1 responses, respectively23, 52, and with studies indicating that ATG16L1 and α-defensins negatively regulate IL-1β expression32. Furthermore, this inflammatory axis is important in responses to gut pathogens, as ASC-mediated IL-1β production has an essential role in Clostridium difficile toxin-induced intestinal pathology53, and both IL-1β and IL-18 were required for the induction of intestinal inflammation after infection with S. Typhimurium54. Thus, inflammasome-forming NLRs can contribute to intestinal pathology through IL-1β and IL-18, and further studies are required to define their roles in IBD23.

Conserved innate and adaptive effector modules

Although predominantly attributed to CD4+ TH cells, many innate leukocyte populations, including γδ T cells, NKT cells and natural killer cells, can secrete TH1 and TH17 cytokines such as IFN-γ, IL-17A and IL-22 (refs 36, 49, 55–57). In particular, γδ T cells can express the TH17-associated transcription factors RORγt and the aryl hydrocarbon receptor (AHR), as well as the homing receptor CCR6, and can secrete IL-17 and IL-22 in response to IL-23 and IL-1β58, 59.

However, recent studies have converged on the identification of several new innate lymphoid cell (ILC) populations present in the gut that can produce these pro-inflammatory cytokines55, 56, 60. Although many functionally heterogeneous ILC populations have been described, their phenotypic characteristics suggest that they are related to natural killer cells and lymphoid tissue inducer (LTI) cells55, 56, 60. Indeed, a recent cell-fate-mapping study suggested that several natural-killer-like and LTI-like ILC subsets were derived from a common RORγt+ precursor61, and functional specialization is coordinated by cytokines and microbiota-dependent signals that direct the expression of distinct transcription factors55, 56, 60. In terms of IBD, we identified a population of CD90+CD4 ILCs that accumulated in the inflamed colons of Rag−/− mice infected with H. hepaticus45. These cells expressed high levels of IL-23R and RORγt and produced IFN-γ, IL-17A and IL-22 in response to IL-23, and their depletion with an anti-CD90 monoclonal antibody led to the attenuation of typhlocolitis45. A similar population of IL-23-responsive CD90+CD4+ LTI cells was recently shown to be an essential source of protective innate IL-22 during the initial phase of Citrobacter rodentium infection62. Other studies have identified further populations of innate leukocytes in the gut and mesenteric adipose tissues that secrete large amounts of the TH2 signature cytokines IL-4 and IL-13 (refs 63–65). These type-2 ILCs are RORγt and were rapidly activated in response to IEC-derived IL-25 and IL-33 after infection with parasitic helminth worms, an infection in which TH2 cytokines are key contributors to protective immunity64, 65. Although the precise relationships of these cells to one another, and to the RORγt+ ILC populations discussed earlier, remain to be determined, these results further emphasize that various populations of intestinal innate leukocytes can rapidly respond to different types of infection by producing appropriate effector cytokines.

The emerging data on shared innate and adaptive cytokine profiles suggest that these conserved immune modules evolved before adaptive lymphocytes, and that protective immunity can be mediated by sentinel tissue-resident innate leukocytes early after infection, whereas subsequent T-cell responses add memory and specificity to the relevant protective axis (Fig. 2).

Adaptive regulation of intestinal inflammation

The intestine contains an extensive network of dendritic cells and macrophages that has an important role in shaping adaptive immunity in response to intestinal environmental cues66, 67. Under homeostatic conditions, both dendritic cells and macrophage populations have specific adaptations that promote tolerance. During infection, however, responses shift to a more inflammatory nature, which can lead to immune pathology when dysregulated.

Antigen-presenting myeloid cells

Intestinal myeloid antigen-presenting cell (APC) populations are heterogeneous in terms of phenotype, function, developmental origin and anatomical location66, 67. Recently, two major populations of intestinal dendritic cells have been identified on the basis of differential expression of the integrin subunit CD103 and the chemokine receptor CX3CR1. CD11chigh CD103+ dendritic cells share developmental origins with lymphoid tissue dendritic cells and are derived from pre-dendritic cells without a monocyte intermediate68, 69. By contrast, monocytes give rise to intestinal CD11chigh CD103 CX3CR1+ dendritic cells, suggesting a close relationship between these cells and CX3CR1+ intestinal macrophages. CD103+ dendritic cells are dispersed throughout the lamina propria and in organized lymphoid structures. In the small intestine, they act as important sentinel cells as they can take up pathogenic and commensal bacteria, as well as innocuous antigens or apoptotic IECs. After maturation, CD103+ dendritic cells migrate to the draining mesenteric lymph node (MLN), where they initiate adaptive responses69, 70 focused on the intestine, including upregulation of the gut-homing receptors CCR9 and α4β7-integrin on activated T cells, and IgA class-switch recombination by intestinal B cells66, 67. These properties depend on the production of the vitamin A dietary metabolite retinoic acid, and CD103+ dendritic cells express the retinal-metabolizing enzyme genes Aldh1a1 and AlDh1a2, although it is not known whether CD103+ dendritic cells are an essential functional source of retinoic acid in vivo66.

CD103+ dendritic cells preferentially induce tolerance pathways, including FOXP3+ Treg cells in the draining MLN by a TGF-β-dependent and retinoic-acid-dependent mechanism66. The intestinal pathways that influence CD103+ dendritic cell function in vivo are poorly understood, but they do not seem to depend on the microbiota71. A recent study showed that ablation of β-catenin expression in mouse dendritic cells led to reduced frequencies of Treg cells and higher frequencies of effector TH1 and TH17 cells in the intestinal lamina propria72. This correlated with reduced messenger RNA levels of Il10, Tgfb, Aldh1a1 and Aldh1a2, resulting in enhanced susceptibility to DSS colitis. How β-catenin expression is regulated in intestinal dendritic cells is not known.

The functional properties of CD103+ dendritic cells are not hardwired, and they acquire inflammatory properties during intestinal inflammation such as the ability to produce IL-6 and drive TH1 responses73. Thus, migratory CD103+ dendritic cells can promote both tolerogenic and effector T-cell responses, and further work is required to identify quantitative and qualitative factors that drive intestinal dendritic cell conditioning and the effect of these factors on adaptive effector pathways.

CX3CR1+CD103 APCs comprise a heterogeneous population of dendritic cells and macrophages. CD11c+CX3CR1+ dendritic cells are present adjacent to the intestinal epithelium and can extend processes through the epithelium to sample antigens and bacteria66, 67. However, CD11c+CX3CR1+ dendritic cells do not seem to migrate to the MLN and fail to prime naive T cells, suggesting that their main role may be to modulate local adaptive intestinal responses70. CX3CR1+ APCs accumulate in response to microbiota-derived signals71, and a CD70high subset promotes colonic TH17 responses in response to commensal-derived ATP67. Colonic macrophages contribute to intestinal homeostasis in several ways. As highly phagocytic cells, they clear apoptotic cells and debris and contribute to wound repair of the epithelium74. Intestinal macrophages have adaptations to prevent excessive inflammatory responses towards the intestinal flora, including expression of inhibitors of NF-κB signalling that permit bactericidal activity in the absence of TLR-driven pro-inflammatory cytokine production74. Recent evidence suggests that these cells also promote tolerance in part through IL-10 production and maintenance of FOXP3 among colonic Treg cells75. A similar APC population in the small intestine induced FOXP3+ Treg cells in vitro66 and promoted Treg-cell proliferation in a CX3CR1-dependent manner76.

In IBD and experimental colitis, there is an increase in dendritic cell and macrophage populations that may contribute to intestinal pathology through pro-inflammatory cytokine production67. Although it is not yet fully established whether this reflects changes in resident myeloid cell populations or the accumulation of newly recruited cells, there is evidence to support the latter. Thus, acute and chronic mouse colitis models were associated with a marked increase in recruited monocyte-derived dendritic cells that produced IL-12, IL-23 and TNF-α and showed enhanced TLR responsiveness68, 77. A similar population of inflammatory macrophages that promote colonic inflammation has also been described78. The data are consistent with a model in which sustained pro-inflammatory cytokines and chemokines promote myelopoiesis — the mobilization of monocytes from the bone marrow to the blood and the recruitment of inflammatory monocytes to the inflamed intestine. These recruited myeloid cells lack gut-specific adaptations associated with tolerance and instead mediate inflammatory responses after microbial challenge. Further understanding of the factors that control the recruitment and function of myeloid cells in intestinal inflammation may provide new therapeutic targets.

Regulatory T-cell populations

Although various T-cell populations have anti-inflammatory functions, FOXP3+ Treg cells and FOXP3 IL-10-secreting CD4+ T cells are particularly important in the intestine79. Most of the former acquire FOXP3 expression in the thymus and represent a functionally distinct population that has a non-redundant role in controlling immune homeostasis. Deletion or loss-of-function mutations in the gene encoding FOXP3 result in a fatal inflammatory disease in mice, and in immune dysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome in humans, which is often accompanied by intestinal inflammation79. FOXP3+ Treg cells are abundant in the small intestine and colon, where they control potentially deleterious responses to dietary and microbial stimuli79. In addition to thymic-derived Treg cells, the intestine is also a preferential site for TGF-β-dependent induction of FOXP3+ Treg cells from naive CD4+ T-cell precursors79. Such antigen-induced Treg cells are expanded in models of oral tolerance and can control local and systemic antigen-induced hypersensitivity responses. Little is known about the antigen specificity of intestinal Treg cells that control microbiota-driven responses. However, Treg-cell accumulation in the colon is reduced in germ-free mice and can be increased by particular indigenous bacteria, suggesting a role for the microbiota in promoting intestinal Treg-cell responses80.

Induced Treg-cell and TH17-cell populations seem to be reciprocally regulated in the intestine. Although TGF-β is required for the differentiation of both populations, the presence of STAT3-mediated signals (such as IL-6 or IL-23) promotes TH17 cells at the expense of FOXP3+ Treg cells39, 81. Such a mechanism allows the inflammatory response to override Treg-cell induction in the presence of pro-inflammatory stimuli, promoting intestinal effector T-cell responses and host defence. Recent evidence suggests that bacterial components differentially affect this balance, providing potential therapeutic strategies to influence tolerance and immunity in the gut28, 82.

An important component of Treg-cell-mediated control of intestinal homeostasis is their ability to survive and compete with effector T cells in the intestinal niche81. This has recently been shown to involve the expression of co-stimulatory pathways and transcription factor modules associated with the colitogenic response44, 83. For example, mice with a Stat3 deletion in FOXP3+ Treg cells develop aggressive colitis owing to uncontrolled TH17 responses44. In addition to STAT3, Treg cells can express several transcription factors associated with particular effector responses, including T-bet, IRF4 and GATA3 (ref. 81). Under homeostatic conditions these allow Treg-cell-mediated control of distinct effector modules. However, the system is delicately poised and can sometimes lead to Treg-cell instability. For example, high-level T-bet expression in the presence of acute intestinal infection drives Treg cells into an inflammatory IFN-γ-secreting phenotype84.

The immune-suppressive modules TGF-β and IL-10

TGF-β is present at high concentrations in the intestine and has a crucial involvement in modulating the immune response85. Deletion of Tgfb1 in mice leads to a fatal inflammatory disease similar to FOXP3 deficiency. Cell-type-specific targeting of TGF-β or its receptor has shown a key role for T cells in both the production and responsiveness to TGF-β that is required to maintain immune homeostasis85. T cells that cannot respond to TGF-β escape Treg-cell-mediated control, and T cells from patients with IBD are refractory to the anti-inflammatory actions of TGF-β through expression of the negative regulator of TGF-β signalling SMAD7 (ref. 86). Whether this is a primary or secondary event is not known, but it suggests that restoring TGF-β responsiveness may have therapeutic benefit in IBD.

TGF-β is produced as an inactive precursor that needs to be post-translationally modified to become biologically active. This is a tightly controlled process that has recently been shown to involve expression of the αvβ8-integrin molecule on intestinal dendritic cells and macrophages66, as well as expression of the proprotein convertase furin on T cells87. αvβ8-Integrin has also been implicated in myeloid-cell uptake of apoptotic cells, a process that has been linked to TGF-β production88. Because IECs undergo apoptosis under physiological conditions, this may provide a source of TGF-β that promotes tolerance under homeostatic conditions. Enhanced apoptosis of IECs accompanies infection and inflammation, and under these circumstances TGF-β in the presence of pro-inflammatory cytokines such as IL-6 promotes the development of inflammatory TH17 responses89. Lastly, a recent report suggests that TGF-β-dependent stimulation of intestinal IgA responses is another mechanism through which Treg cells can reinforce intestinal homeostasis90.

IL-10 is produced by a wide range of leukocytes, including T cells, B cells and myeloid cells, and its deletion in mice leads to colitis development79. CD4+ T-cell-produced IL-10 is required to prevent intestinal inflammation, with functional contributions from both FOXP3+ and FOXP3 CD4+ cells79. The intestine contains large numbers of CD4+IL-10+ cells; in the colon, these are mainly FOXP3+, whereas both FOXP3+ and FOXP3 IL-10+ cells are present in the small intestine79. Intestinal bacteria can promote the activity of colonic Treg cells by inducing IL-10 production, and recent evidence suggests a specific role for particular Clostridium species in this process80. Unlike FOXP3-expressing cells, FOXP3IL-10+CD4+ cells may represent a more heterogeneous mix, because most effector TH-cell subsets, including TH1, TH2 and TH17 cells, produce IL-10 after chronic immune stimulation91. Myeloid sources of IL-10 are important in some settings, as IL-10 production by intestinal macrophages promoted FOXP3 Treg-cell function in an adoptive transfer model of colitis75. IL-10 controls chronic intestinal inflammation partly through direct anti-inflammatory effects on myeloid cells79. Evidence for the role of IL-10 in human IBD comes from findings that mutations in the IL-10 receptor genes IL10RA and IL10RB lead to severe early-onset IBD92. GWAS have also identified single nucleotide polymorphisms in IL10 associated with susceptibility to Crohn's disease and ulcerative colitis (see page 307). Together, both genetic and functional studies highlight the importance of IL-10 in intestinal homeostasis and suggest that the ability of intestinal bacteria to induce IL-10 may be an important facet of host–commensal mutualism.

A multihit model of IBD

As noted above, in some cases single gene defects in crucial regulatory circuits, such as the IL-10 pathway92, can trigger severe IBD in infants. However, the heterogeneity of IBD and the low disease penetrance in individuals carrying disease-susceptibility alleles suggest that, in most patients, several host and environmental factors interact to cause IBD. There is evidence to suggest that Crohn's disease stems from an immunodeficiency of macrophages that results in defective acute inflammatory responses and impaired clearance of commensal bacteria, leading to the subsequent expression of chronic granulomatous inflammation93. Pathogenic infections may act as triggers or contributing factors for IBD, and adherent-invasive Escherichia coli are frequently present in close association with the ileal mucosa in patients with Crohn's disease1.

It is not yet clear whether the presence of E. coli is a cause or effect of colitis, as recent studies have highlighted that intestinal inflammation can confer a selective growth advantage to certain pathogens, including S. Typhimurium94. A growing body of evidence suggests that IBD is associated with an imbalance in the composition of the intestinal bacterial microbiota, termed dysbiosis2, 95. Patients with IBD, particularly those with Crohn's disease, have alterations in the gut microbiota, with reduced diversity in major phyla, such as Firmicutes and Bacteroidetes, and increased numbers of Enterobacteriaceae2, 95. A key unresolved issue is whether dysbiosis represents a primary or secondary predisposing factor for IBD, as it may be related to, or compounded by, other defects. Recent studies have indicated that dysbiosis is influenced both by the host genotype, such as the presence of NOD2- or ATG16L1-susceptibility alleles96, and by IBD phenotype, with patients with ileal Crohn's disease showing the most pronounced changes97. It is interesting that core commensals belonging to the Clostridiales order, such as Faecalibacterium and Roseburia, were significantly reduced in patients with ileal Crohn's disease96, 97. These genera are potent sources of short-chain fatty acids, such as butyrate, that have been shown to have protective effects in mouse colitis models98. In addition, clostridial groups IV (which includes Faecalibacterium) and XIVa were recently shown to promote the accumulation of FOXP3+ Treg cells in the mouse colon80. Dietary factors may also affect microbiota composition, leading to alterations in intestinal immune homeostasis98.

Taken together, these studies are beginning to illuminate some of the complex interactions between different host genetic and environmental factors that can predispose patients to the development of IBD. The induction and perpetuation of chronic intestinal pathology may require additive lesions that affect several layers of immune regulation in the gut (Fig. 3). Early animal model studies showed that an interaction between the intestinal flora and host factors is required for the development of intestinal inflammation. For example, colitis in IL-10-deficient mice requires the presence of triggering bacteria such as H. hepaticus infection79. Two recent studies further illustrate how multiple lesions may interact to elicit intestinal pathology. The first of these found that the Paneth cell abnormalities present in mice carrying a hypomorphic mutant of the Atg16l1 autophagy gene (Atg16lHM) were triggered by persistent infection with an enteric mouse norovirus strain (MNV CR6)99. In addition, Atg16lHM mice infected with MNV CR6 developed exacerbated pathology after the administration of DSS, with characteristics of human Crohn's disease, including blunting of villi in the ileum, that depended on IFN-γ, TNF-α and the presence of commensal bacteria99. Thus, Crohn's-disease-like pathology required persistent viral infection of a genetically susceptible host together with environmental factors and commensal bacteria. The second study examined a model of transmissible ulcerative-colitis-like disease that arises in T-bet−/−Rag2−/− (also known as Tbx21−/−Rag2−/−) mice and is associated with defective colonic barrier function and hyperactivation of inflammatory dendritic cells100. Transmission of a milder degree of colitis to co-housed wild-type mice correlated with the presence of the Enterobacteriaceae species Proteus mirabilis and Klebsiella pneumoniae, but also required the presence of an endogenous microbiota100. Thus, maximal colitis involved barrier defects, hyperactivated innate immunity, an absence of Treg cells and alterations in the microbiota composition.

Figure 3: A multihit model of IBD pathogenesis.
A multihit model of IBD pathogenesis.

The induction and perpetuation of chronic intestinal pathology may require the convergence of many abnormalities that affect several overlapping layers of immune homeostasis in the intestine. Defects in one layer are unlikely to precipitate IBD in the absence of further pathogenic lesions (panels on left). These layers are in turn controlled by a range of cell-intrinsic and cell-extrinsic regulatory modules that function in an interrelated way to maintain homeostasis (panels on right). Defects in these homeostatic modules may predispose people to the development of IBD by affecting several layers of immune homeostasis (filled panels). The dashed line denotes the threshold at which the level of inflammation manifests as IBD.

Perspectives

Recent advances in mapping the genetic basis of disease susceptibility, coupled with rapid improvements in characterization of the microbiota in healthy and diseased individuals, offer great hope for the continued development of new IBD treatments. However, several key issues need to be understood better. These include distinguishing between individuality in IBD aetiology and commonality in pathogenic effector modules, so that therapies may be tailored to appropriate patient subgroups, such that distinct responses may be either suppressed or enhanced to restore homeostasis. The influence of microbiota-derived molecules on local and systemic immune responses is an area of great promise, but it will also be important to determine how immune responses feed back into shaping the composition of the microbiota and how different members of the microbiota interact within different environments in the gut, as well as to determine how to stably manipulate the gut microbiota. The real extent of effector T-cell plasticity in vivo and whether innate effector responses are similarly malleable needs to be investigated further, to establish whether the stable conversion of deleterious responses into beneficial ones may be achieved. Accomplishing these goals will require the cooperation of scientists working across several disciplines, an improved characterization of the pathophysiology of disease models and application of new technical approaches to clinical samples from patients with IBD.

References

  1. Kaser, A., Zeissig, S. & Blumberg, R. S. Inflammatory bowel disease. Annu. Rev. Immunol. 28, 573621 (2010).
  2. Hill, D. A. & Artis, D. Intestinal bacteria and the regulation of immune cell homeostasis. Annu. Rev. Immunol. 28, 623667 (2010).
  3. Artis, D. Epithelial-cell recognition of commensal bacteria and maintenance of immune homeostasis in the gut. Nature Rev. Immunol. 8, 411420 (2008).
  4. Hooper, L. V. & Macpherson, A. J. Immune adaptations that maintain homeostasis with the intestinal microbiota. Nature Rev. Immunol. 10, 159169 (2010).
  5. Koslowski, M. J., Beisner, J., Stange, E. F. & Wehkamp, J. Innate antimicrobial host defense in small intestinal Crohn's disease. Int. J. Med. Microbiol. 300, 3440 (2010).
  6. Vaishnava, S., Behrendt, C. L., Ismail, A. S., Eckmann, L. & Hooper, L. V. Paneth cells directly sense gut commensals and maintain homeostasis at the intestinal host–microbial interface. Proc. Natl Acad. Sci. USA 105, 2085820863 (2008).
  7. Cadwell, K. et al. A key role for autophagy and the autophagy gene Atg16l1 in mouse and human intestinal Paneth cells. Nature 456, 259263 (2008).
  8. Deretic, V. & Levine, B. Autophagy, immunity, and microbial adaptations. Cell Host Microbe 5, 527549 (2009).
  9. Kaser, A., Martinez-Naves, E. & Blumberg, R. S. Endoplasmic reticulum stress: implications for inflammatory bowel disease pathogenesis. Curr. Opin. Gastroenterol. 26, 318326 (2010).
  10. Kaser, A. et al. XBP1 links ER stress to intestinal inflammation and confers genetic risk for human inflammatory bowel disease. Cell 134, 743756 (2008).
    This report links ER stress in IECs to the development of intestinal inflammation in both mice and humans.
  11. Taylor, B. C. et al. TSLP regulates intestinal immunity and inflammation in mouse models of helminth infection and colitis. J. Exp. Med. 206, 655667 (2009).
  12. Abreu, M. T. Toll-like receptor signalling in the intestinal epithelium: how bacterial recognition shapes intestinal function. Nature Rev. Immunol. 10, 131144 (2010).
  13. Fagarasan, S., Kawamoto, S., Kanagawa, O. & Suzuki, K. Adaptive immune regulation in the gut: T cell-dependent and T cell-independent IgA synthesis. Annu. Rev. Immunol. 28, 243273 (2010).
  14. Slack, E. et al. Innate and adaptive immunity cooperate flexibly to maintain host–microbiota mutualism. Science 325, 617620 (2009).
  15. Asquith, M. J., Boulard, O., Powrie, F. & Maloy, K. J. Pathogenic and protective roles of MyD88 in leukocytes and epithelial cells in mouse models of inflammatory bowel disease. Gastroenterology 139, 519529 (2010).
  16. Cario, E. Toll-like receptors in inflammatory bowel diseases: a decade later. Inflamm. Bowel Dis. 16, 15831597 (2010).
  17. Strober, W., Murray, P. J., Kitani, A. & Watanabe, T. Signalling pathways and molecular interactions of NOD1 and NOD2. Nature Rev. Immunol. 6, 920 (2006).
  18. Chen, G. Y., Shaw, M. H., Redondo, G. & Nunez, G. The innate immune receptor Nod1 protects the intestine from inflammation-induced tumorigenesis. Cancer Res. 68, 1006010067 (2008).
  19. Dupaul-Chicoine, J. et al. Control of intestinal homeostasis, colitis, and colitis-associated colorectal cancer by the inflammatory caspases. Immunity 32, 367378 (2010).
  20. Zaki, M. H. et al. The NLRP3 inflammasome protects against loss of epithelial integrity and mortality during experimental colitis. Immunity 32, 379391 (2010).
  21. Allen, I. C. et al. The NLRP3 inflammasome functions as a negative regulator of tumorigenesis during colitis-associated cancer. J. Exp. Med. 207, 10451056 (2010).
  22. Schroder, K. & Tschopp, J. The inflammasomes. Cell 140, 821832 (2010).
  23. Siegmund, B. Interleukin-18 in intestinal inflammation: friend and foe? Immunity 32, 300302 (2010).
  24. Salcedo, R. et al. MyD88-mediated signaling prevents development of adenocarcinomas of the colon: role of interleukin 18. J. Exp. Med. 207, 16251636 (2010).
  25. Feng, T., Wang, L., Schoeb, T. R., Elson, C. O. & Cong, Y. Microbiota innate stimulation is a prerequisite for T cell spontaneous proliferation and induction of experimental colitis. J. Exp. Med. 207, 13211332 (2010).
  26. Asquith, M. & Powrie, F. An innately dangerous balancing act: intestinal homeostasis, inflammation, and colitis-associated cancer. J. Exp. Med. 207, 15731577 (2010).
  27. Saleh, M. & Trinchieri, G. Innate immune mechanisms of colitis and colitis-associated colorectal cancer. Nature Rev. Immunol. 11, 920 (2011).
  28. Round, J. L. & Mazmanian, S. K. Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota. Proc. Natl Acad. Sci. USA 107, 1220412209 (2010).
  29. Richardson, W. M. et al. Nucleotide-binding oligomerization domain-2 inhibits Toll-like receptor-4 signaling in the intestinal epithelium. Gastroenterology 139, 904917 (2010).
  30. Travassos, L. H. et al. Nod1 and Nod2 direct autophagy by recruiting ATG16L1 to the plasma membrane at the site of bacterial entry. Nature Immunol. 11, 5562 (2010).
  31. Cooney, R. et al. NOD2 stimulation induces autophagy in dendritic cells influencing bacterial handling and antigen presentation. Nature Med. 16, 9097 (2010).
  32. Saitoh, T. et al. Loss of the autophagy protein Atg16L1 enhances endotoxin-induced IL-1β production. Nature 456, 264268 (2008).
    References 31 and 32 describe a link between key Crohn's disease susceptibility factors by showing that NOD2 can stimulate autophagy and that this constitutes an important bacterial handling mechanism.
  33. Zhou, R., Yazdi, A. S., Menu, P. & Tschopp, J. A role for mitochondria in NLRP3 inflammasome activation. Nature 469, 221225 (2011).
  34. Martinon, F., Chen, X., Lee, A. H. & Glimcher, L. H. TLR activation of the transcription factor XBP1 regulates innate immune responses in macrophages. Nature Immunol. 11, 411418 (2010).
  35. Melmed, G. Y. & Targan, S. R. Future biologic targets for IBD: potentials and pitfalls. Nature Rev. Gastroenterol. Hepatol. 7, 110117 (2010).
  36. Maloy, K. J. & Kullberg, M. C. IL-23 and Th17 cytokines in intestinal homeostasis. Mucosal Immunol. 1, 339349 (2008).
  37. Kamada, N. et al. Unique CD14+ intestinal macrophages contribute to the pathogenesis of Crohn disease via IL-23/IFN-γ axis. J. Clin. Invest. 118, 22692280 (2008).
  38. Goodall, J. C. et al. Endoplasmic reticulum stress-induced transcription factor, CHOP, is crucial for dendritic cell IL-23 expression. Proc. Natl Acad. Sci. USA 107, 1769817703 (2010).
  39. Ahern, P. P. et al. Interleukin-23 drives intestinal inflammation through direct activity on T cells. Immunity 33, 279288 (2010).
  40. Leppkes, M. et al. RORγ-expressing Th17 cells induce murine chronic intestinal inflammation via redundant effects of IL-17A and IL-17F. Gastroenterology 136, 257267 (2009).
  41. Cosmi, L. et al. Human interleukin 17-producing cells originate from a CD161+CD4+ T cell precursor. J. Exp. Med. 205, 19031916 (2008).
  42. Kleinschek, M. A. et al. Circulating and gut-resident human Th17 cells express CD161 and promote intestinal inflammation. J. Exp. Med. 206, 525534 (2009).
  43. Yang, X. O. et al. Regulation of inflammatory responses by IL-17F. J. Exp. Med. 205, 10631075 (2008).
  44. Chaudhry, A. et al. CD4+ regulatory T cells control TH17 responses in a Stat3-dependent manner. Science 326, 986991 (2009).
  45. Buonocore, S. et al. Innate lymphoid cells drive interleukin-23-dependent innate intestinal pathology. Nature 464, 13711375 (2010).
  46. Wu, S. et al. A human colonic commensal promotes colon tumorigenesis via activation of T helper type 17 T cell responses. Nature Med. 15, 10161022 (2009).
  47. Ivanov, I. I. et al. Induction of intestinal Th17 cells by segmented filamentous bacteria. Cell 139, 485498 (2009).
  48. Gaboriau-Routhiau, V. et al. The key role of segmented filamentous bacteria in the coordinated maturation of gut helper T cell responses. Immunity 31, 677689 (2009).
    References 47 and 48 document a strong link between segmented filamentous bacteria colonization and TH17 cells and, together with reference 80, provide compelling evidence that colonization with distinct types of commensal bacterium results in the accumulation of different effector T cells in the intestine.
  49. Wolk, K., Witte, E., Witte, K., Warszawska, K. & Sabat, R. Biology of interleukin-22. Semin. Immunopathol. 32, 1731 (2010).
  50. Pickert, G. et al. STAT3 links IL-22 signaling in intestinal epithelial cells to mucosal wound healing. J. Exp. Med. 206, 14651472 (2009).
  51. Sonnenberg, G. F. et al. Pathological versus protective functions of IL-22 in airway inflammation are regulated by IL-17A. J. Exp. Med. 207, 12931305 (2010).
  52. Chung, Y. et al. Critical regulation of early Th17 cell differentiation by interleukin-1 signaling. Immunity 30, 576587 (2009).
  53. Ng, J. et al. Clostridium difficile toxin-induced inflammation and intestinal injury are mediated by the inflammasome. Gastroenterology 139, 542552.e3 (2010).
  54. Muller, A. J. et al. The S. Typhimurium effector SopE induces caspase-1 activation in stromal cells to initiate gut inflammation. Cell Host Microbe 6, 125136 (2009).
  55. Colonna, M. Interleukin-22-producing natural killer cells and lymphoid tissue inducer-like cells in mucosal immunity. Immunity 31, 1523 (2009).
  56. Cua, D. J. & Tato, C. M. Innate IL-17-producing cells: the sentinels of the immune system. Nature Rev. Immunol. 10, 479489 (2010).
  57. Park, S. G. et al. T regulatory cells maintain intestinal homeostasis by suppressing γδ T cells. Immunity 33, 791803 (2010).
  58. Martin, B., Hirota, K., Cua, D. J., Stockinger, B. & Veldhoen, M. Interleukin-17-producing γδ T cells selectively expand in response to pathogen products and environmental signals. Immunity 31, 321330 (2009).
  59. Sutton, C. E. et al. Interleukin-1 and IL-23 induce innate IL-17 production from γδ T cells, amplifying Th17 responses and autoimmunity. Immunity 31, 331341 (2009).
  60. Spits, H. & Di Santo, J. P. The expanding family of innate lymphoid cells: regulators and effectors of immunity and tissue remodeling. Nature Immunol. 12, 2127 (2011).
  61. Sawa, S. et al. Lineage relationship analysis of RORγ+ innate lymphoid cells. Science 330, 665669 (2010).
  62. Sonnenberg, G. F., Monticelli, L. A., Elloso, M. M., Fouser, L. A. & Artis, D. CD4+ lymphoid tissue-inducer cells promote innate immunity in the gut. Immunity 34, 122134 (2011).
    References 45 and 62 identify new populations of LTI-like ILCs that secrete TH1 and TH17 pro-inflammatory cytokines in response to IL-23, and these contribute to intestinal pathology and host defences against intestinal pathogenic bacteria.
  63. Moro, K. et al. Innate production of TH2 cytokines by adipose tissue-associated c-Kit+Sca-1+ lymphoid cells. Nature 463, 540544 (2010).
  64. Saenz, S. A. et al. IL25 elicits a multipotent progenitor cell population that promotes TH2 cytokine responses. Nature 464, 13621366 (2010).
  65. Neill, D. R. et al. Nuocytes represent a new innate effector leukocyte that mediates type-2 immunity. Nature 464, 13671370 (2010).
    References 63–65 describe various ILC populations that secrete TH2 cytokines (ILC type 2) in response to IL-25 and IL-33 and that can contribute to defence against intestinal helminth infection.
  66. Coombes, J. L. & Powrie, F. Dendritic cells in intestinal immune regulation. Nature Rev. Immunol. 8, 435446 (2008).
  67. Varol, C., Zigmond, E. & Jung, S. Securing the immune tightrope: mononuclear phagocytes in the intestinal lamina propria. Nature Rev. Immunol. 10, 415426 (2010).
  68. Varol, C. et al. Intestinal lamina propria dendritic cell subsets have different origin and functions. Immunity 31, 502512 (2009).
  69. Bogunovic, M. et al. Origin of the lamina propria dendritic cell network. Immunity 31, 513525 (2009).
  70. Schulz, O. et al. Intestinal CD103+, but not CX3CR1+, antigen sampling cells migrate in lymph and serve classical dendritic cell functions. J. Exp. Med. 206, 31013114 (2009).
  71. Niess, J. H. & Adler, G. Enteric flora expands gut lamina propria CX3CR1+ dendritic cells supporting inflammatory immune responses under normal and inflammatory conditions. J. Immunol. 184, 20262037 (2010).
  72. Manicassamy, S. et al. Activation of β-catenin in dendritic cells regulates immunity versus tolerance in the intestine. Science 329, 849853 (2010).
  73. Laffont, S., Siddiqui, K. R. & Powrie, F. Intestinal inflammation abrogates the tolerogenic properties of MLN CD103+ dendritic cells. Eur. J. Immunol. 40, 18771883 (2010).
  74. Smith, P. D. et al. Intestinal macrophages and response to microbial encroachment. Mucosal Immunol. 4, 3142 (2011).
  75. Murai, M. et al. Interleukin 10 acts on regulatory T cells to maintain expression of the transcription factor Foxp3 and suppressive function in mice with colitis. Nature Immunol. 10, 11781184 (2009).
  76. Hadis, U. et al. Intestinal tolerance requires gut homing and expansion of FoxP3+ regulatory T cells in the lamina propria. Immunity 34, 237246 (2011).
    This study shows that oral tolerance requires homing and expansion of Treg cells in the intestine.
  77. Siddiqui, K. R., Laffont, S. & Powrie, F. E-cadherin marks a subset of inflammatory dendritic cells that promote T cell-mediated colitis. Immunity 32, 557567 (2010).
  78. Platt, A. M., Bain, C. C., Bordon, Y., Sester, D. P. & Mowat, A. M. An independent subset of TLR expressing CCR2-dependent macrophages promotes colonic inflammation. J. Immunol. 184, 68436854 (2010).
  79. Izcue, A., Coombes, J. L. & Powrie, F. Regulatory lymphocytes and intestinal inflammation. Annu. Rev. Immunol. 27, 313338 (2009).
  80. Atarashi, K. et al. Induction of colonic regulatory T cells by indigenous Clostridium species. Science 331, 337341 (2011).
    This paper describes a link between Clostridium spp. and Treg cells in the gut and, together with references 47 and 48, provides compelling evidence that colonization with distinct commensal bacteria results in the accumulation of different effector T cells in the intestine.
  81. Littman, D. R. & Rudensky, A. Y. Th17 and regulatory T cells in mediating and restraining inflammation. Cell 140, 845858 (2010).
  82. Hall, J. A. et al. Commensal DNA limits regulatory T cell conversion and is a natural adjuvant of intestinal immune responses. Immunity 29, 637649 (2008).
  83. Griseri, T., Asquith, M., Thompson, C. & Powrie, F. OX40 is required for regulatory T cell-mediated control of colitis. J. Exp. Med. 207, 699709 (2010).
  84. Oldenhove, G. et al. Decrease of Foxp3+ Treg cell number and acquisition of effector cell phenotype during lethal infection. Immunity 31, 772786 (2009).
  85. Li, M. O. & Flavell, R. A. Contextual regulation of inflammation: a duet by transforming growth factor-β and interleukin-10. Immunity 28, 468476 (2008).
  86. Fantini, M. C. et al. Smad7 controls resistance of colitogenic T cells to regulatory T cell-mediated suppression. Gastroenterology 136, 13081316.e3 (2009).
  87. Pesu, M. et al. T-cell-expressed proprotein convertase furin is essential for maintenance of peripheral immune tolerance. Nature 455, 246250 (2008).
  88. Perruche, S. et al. CD3-specific antibody-induced immune tolerance involves transforming growth factor-β from phagocytes digesting apoptotic T cells. Nature Med. 14, 528535 (2008).
  89. Torchinsky, M. B., Garaude, J., Martin, A. P. & Blander, J. M. Innate immune recognition of infected apoptotic cells directs TH17 cell differentiation. Nature 458, 7882 (2009).
  90. Cong, Y., Feng, T., Fujihashi, K., Schoeb, T. R. & Elson, C. O. A dominant, coordinated T regulatory cell–IgA response to the intestinal microbiota. Proc. Natl Acad. Sci. USA 106, 1925619261 (2009).
  91. Saraiva, M. & O'Garra, A. The regulation of IL-10 production by immune cells. Nature Rev. Immunol. 10, 170181 (2010).
  92. Glocker, E. O. et al. Inflammatory bowel disease and mutations affecting the interleukin-10 receptor. N. Engl. J. Med. 361, 20332045 (2009).
  93. Smith, A. M. et al. Disordered macrophage cytokine secretion underlies impaired acute inflammation and bacterial clearance in Crohn's disease. J. Exp. Med. 206, 18831897 (2009).
  94. Winter, S. E. et al. Gut inflammation provides a respiratory electron acceptor for Salmonella. Nature 467, 426429 (2010).
  95. Sartor, R. B. Microbial influences in inflammatory bowel diseases. Gastroenterology 134, 577594 (2008).
  96. Frank, D. N. et al. Disease phenotype and genotype are associated with shifts in intestinal-associated microbiota in inflammatory bowel diseases. Inflamm. Bowel Dis. 17, 179184 (2011).
  97. Willing, B. P. et al. A pyrosequencing study in twins shows that gastrointestinal microbial profiles vary with inflammatory bowel disease phenotypes. Gastroenterology 139, 18441854.e1 (2010).
  98. Maslowski, K. M. & Mackay, C. R. Diet, gut microbiota and immune responses. Nature Immunol. 12, 59 (2011).
  99. Cadwell, K. et al. Virus-plus-susceptibility gene interaction determines Crohn's disease gene Atg16L1 phenotypes in intestine. Cell 141, 11351145 (2010).
  100. Garrett, W. S. et al. Enterobacteriaceae act in concert with the gut microbiota to induce spontaneous and maternally transmitted colitis. Cell Host Microbe 8, 292300 (2010).
    References 99 and 100 provide illustrative examples of how several host and environmental factors may act together to precipitate chronic intestinal inflammation.

Download references

Acknowledgements

We thank M. Asquith, H. Uhlig, P. Ahern and M. Barnes for review and G. Song-Zhao and O. Harrison for help with the figures. We apologize to those whose work was not cited owing to space constraints. K.J.M. and F.P. are supported by grants from the Wellcome Trust, Cancer Research UK and the European Union (FP7, INFLAMMACARE).

Author information

Affiliations

  1. Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford OX1 3RE, UK.

    • Kevin J. Maloy &
    • Fiona Powrie
  2. Translational Gastroenterology Unit, Experimental Medicine Division – NDM, University of Oxford, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.

    • Fiona Powrie

Competing financial interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to:

Reprints and permissions information is available at http://www.nature.com/reprints.

Author details

Additional data