Article Text

Download PDFPDF

Immune regulation and colitis: suppression of acute inflammation allows the development of chronic inflammatory bowel disease
  1. B Eksteen,
  2. L S K Walker,
  3. D H Adams
  1. Liver Research Laboratories and MRC Centre for Immune Regulation, Institute for Biomedical Research, Queen Elizabeth Hospital, University of Birmingham, Birmingham, UK
  1. Correspondence to:
    Professor D H Adams
    Liver Research Laboratories, Institute for Biomedical Research, Queen Elizabeth Hospital, University of Birmingham, Birmingham B15 2TT, UK;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Persistent colitis is the result of a balance between local inflammation and regulatory networks. Regulatory T cells have potent anti-inflammatory effects and are likely to be important in the pathogenesis of chronic inflammatory bowel disease

The success of the gastrointestinal immune system depends on a balance between mounting effective immune responses to pathogenic antigens while suppressing potentially damaging responses against commensal organisms or food antigens. Both the innate and acquired immune system contribute to fighting pathogens. The innate immune system, which includes phagocytes, dendritic cells (DCs), and natural killer (NK) cells, does not require previous exposure to a pathogen and instead relies on evolutionarily ancient pathways such as Toll-like receptors (TLRs) to recognise molecular patterns associated with harmful pathogens.1 Different TLRs are able to recognise bacterial products or motifs in viral RNA or DNA. TLR activation triggers an immediate response resulting in the activation of phagocytic mechanisms and the production of cytokines and costimulatory signals that activate the cognate immune response.

The cognate or acquired immune system developed in higher vertebrates to provide a more sophisticated response to a wide variety of antigens. It involves lymphocytes that recognise specific antigens from pathogens processed and presented by specialised antigen presenting cells called DCs. A crucial feature of the cognate immune system is immunological memory whereby a subsequent exposure to the same antigen leads to a more potent and sustained immune response. Antigens from pathogens that penetrate the mucosal barrier are taken up by local DCs and carried via lymphatics to draining mesenteric lymph nodes.2 In addition, DCs situated in gut associated lymphoid tissue such as Peyer’s patches sample luminal antigens either directly, by extending finger-like processes into the lumen, or via specialised epithelial cells (M cells) that actively transport luminal antigens to the underlying DCs.3 Antigen containing DCs then migrate to secondary lymphoid tissues where they interact with naïve lymphocytes with the appropriate antigenic specificity to generate primed effector lymphocytes expressing the receptors CCR9 and α4β7 that direct their migration back to gut tissues.4–6 Once the antigen is cleared, most effector cells die, leaving a small cohort of long lived memory cells that can rapidly augment immunity to the antigen if encountered again.

The importance of our ability to develop immunity and appropriate responses to gut derived antigens is highlighted by, on one hand, immunodeficient states where persistent gut infections often dominate and translocation of gut pathogens can lead to overwhelming sepsis, and on the other, the fact that inappropriate activation of the intestinal immune system leads to uncontrolled inflammation and inflammatory bowel disease (IBD).7 Recent studies suggest that damaging inflammation in IBD is not only the consequence of inappropriate stimulation of effector responses but is also due to a failure of the normal immunosuppressive mechanisms that have evolved to control inflammation in the gut.

The immune system has developed several ways to suppress responses to harmless self or non-pathogenic environmental antigens. Central tolerance is regulated in the thymus where T cells are selected for survival based on the affinity of their T cell receptor (TCR) for antigens expressed on thymic epithelium and those T cells showing high affinity for self antigens are deleted. In addition, thymic selection leads to generation of a distinct subset of regulatory T cells (Tregs) which when activated by antigen can suppress effector responses generated by both NK and T cells in the periphery.8 Thymic Tregs, which comprise approximately 10% of CD4+ T cells in the mouse circulation, display several cell surface receptors, including the interleukin (IL)-2 receptor, CD25, and the glucocorticoid induced tumour necrosis factor (TNF) receptor (GITR). They are characterised by expression of the transcription factor Foxp3 which is critical for Treg function, as demonstrated by experiments in which retroviral transfer of Foxp3 to naïve T cells converts them into functional Tregs whereas its deletion results in loss of regulatory function and the development of autoimmunity.7,9,10 Tregs mediate contact dependent suppression of effector cells in vitro, although in vivo the situation is more complex with IL-10 and/or transforming growth factor β (TGF-β) being required for suppression in many circumstances. Indeed, it has been reported that Tregs express a membrane bound form of the TGF-β cytokine.11 In addition to thymic Tregs, inducible Tregs can be generated in the periphery as a consequence of activation of naïve T cells by immature DCs or activation in the presence of specific cytokines. Thus generation of inducible Tregs depends critically on the local microenvironment in which activation takes place. Inducible Tregs secrete IL-10 and stimulate the local secretion of TGF-β, both of which are potent regulators of inflammation capable of suppressing the proliferation of effector cells.8,12

Powrie et al originally defined the importance of Tregs in gut inflammation, demonstrating that adoptively transferred Tregs suppress colonic inflammation in experimental colitis and that this suppression depends on both TGF-β13 and IL-10.14 The subsequent demonstration that IL-10 can promote TGF-β secretion in the setting of experimental colitis provides a further clue as to how these cytokines may be cooperating in vivo.15 The importance of IL-10 is underscored by the fact that deficiency in either IL-1016 or in the ability of macrophages/neutrophils to respond to IL-10 (as a result of targeted stat-3 deficiency)17 is sufficient to trigger gut pathology. Likewise, removal or inhibition of Tregs at either the central or peripheral level is associated with autoimmunity,18 a break in tolerance, and intestinal inflammation in animal models.

Much of the decision making that leads to generation of peripheral regulatory networks falls upon DCs and suppression and tolerance are the consequences of lymphocyte activation by immature DCs that produce IL-10.19,20 Further control of inflammation in the periphery is afforded by the susceptibility of effector lymphocytes to programmed cell death or apoptosis.18 Mechanisms that have evolved to control the expansion of antigen specific effector cells are critical for normal immune homeostasis to prevent the uncontrolled proliferation of effector cells. IL-2 is a survival signal for activated lymphocytes which is produced in large amounts during inflammation. When the antigen has been cleared, IL-2 production falls, resulting in apoptosis of most effector cells except for a small population of long lived memory lymphocytes.21 In addition, effector lymphocytes express the death receptor Fas and its ligand Fas-L and once cytolytic T cells have destroyed Fas bearing target cells, effector T cells can kill each other by the same Fas/FasL interactions allowing inflammation to resolve. Curiously, IL-2 potentiates Fas mediated killing, illustrating the complex interplay that exists between different peripheral tolerance mechanisms. The importance of the Fas pathway is demonstrated by the development of lymphomas in mice that lack either Fas or FasL.

Mechanisms that lead to uncontrolled chronic inflammation in IBD are poorly understood but are likely to involve many if not most of the above mechanisms.22 The paper by Westendorf and colleagues23 in this issue of Gut, provides further important insights into the mechanisms of chronic inflammation in IBD (see page 60). The authors used a transgenic model in which gut inflammation was triggered by overexpression of a single foreign antigen, influenza HA, in enterocytes. Crossing these mice with transgenic animals that expressed αβ-T cells specific for HA resulted in an animal with autoreactive T cells that recognised HA as a “self” antigen restricted to the gut. They reported that these animals developed autoimmune colitis and chronic inflammation demonstrating that expression of a self antigen on enterocytes in sufficient to trigger colitis. However, they found that although their model induced chronic inflammation, colitis was far less severe than in other colitis models, suggesting that the inflammation is partially controlled by regulatory mechanisms. The authors then studied the cytokine profiles secreted by mucosal lymphocytes from their transgenic animals in response to antigen stimulation. They found that whereas secretion of the classical Th1 effector cytokines interferon γ and IL-2 was reduced, secretion of the proinflammatory cytokines TNF-α, monocyte chemoattractant protein 1, and IL-6 was increased, indicating that while T cells were capable of responding to antigen, the nature of the response was markedly altered. This raises the possibility that a balance has developed between regulatory and inflammatory mechanisms giving rise to the generation of chronic persistent inflammation. When the authors pursued this hypothesis further they found that autoreactive lamina propria lymphocytes and intraepithelial lymphocytes expressed increased levels of the anti-inflammatory cytokine IL-10 and several genes associated with the development of regulatory T cells. While some of these genes would also be highly expressed in activated T cells (for example, OX40, GITR), others such as neuropilin-1 are thought to specifically identify regulatory cells.24 The authors concluded that, in their double transgenic mouse, enterocyte specific CD4+ T cells are sufficient to induce colitis which is neither acute and self limiting (which might be expected if regulatory mechanisms were dominant) nor acute and fatal (if they were absent) but rather chronic and persistent. They propose that this outcome is the result of a balance between local inflammatory and regulatory networks and infer that regulatory T cells may be important in the pathogenesis of chronic inflammation.

Several important conclusions regarding the pathogenesis of IBD can be drawn from the study and there are parallels between the VILLIN-HA×TCR-HA transgenic model of gut inflammation and IBD in humans. Recognition of an autoantigen by either breakdown of central tolerance in the thymus or by acquired cross reactivity to an external antigen provides a potent immune response that is able to establish clinical disease. However, the pattern of disease may well be determined by the nature of the local tolerogenic networks. This is turn will be affected by (1) the genetic background of the individual, which will determine whether they generate strong regulatory or inflammatory responses, and (2) the local microenvironment, including the nature of the bacterial flora. The paper demonstrates the importance of effective thymic selection in deleting autoreactive T cells. This is not a new concept—loss of thymic regulation has been shown to trigger a variety of autoimmune conditions, including colitis—but is nevertheless important as these animals had no obvious defects in the thymus. The authors concede that the transgenic TCR might interfere with thymic selection but it is also plausible that the large numbers of potentially autoreactive lymphocytes in this transgenic model allow some to escape deletion. It is not known if this phenomenon occurs in humans with IBD.

A second important finding is the pattern of inflammation induced in the VILLIN-HA×TCR-HA mice. Models of colitis induced by agents such as dextran sulphate result in acute inflammation and require repeated administration to mimic chronic gut inflammation whereas the authors’ model of autoimmune intestinal inflammation is sustained by the persistent presence of the gut antigen. However, the real novelty of the present study is the suggestion that the chronicity of the inflammation, which in many ways resembles clinical IBD, is the result of a balance between pro and anti-inflammatory pathways involving regulatory T cells. This balance permits continuing inflammation while preventing uncontrolled progression of destructive colitis. Factors underlying this regulatory response are at present unknown although the data in the paper suggest testable theories. The cytokine data suggest that local IL-10 is able to prevent acute destruction of the gut but not chronic intestinal inflammation, leading to persistence of inflammation and chronic disease. Local intestinal DCs are a potential source of IL-10 and may induce the development of IL-10 secreting T cells. Because the transgenic model expresses local antigen in the absence of an inflammatory “danger” signal, local DCs may be only partially activated, leading to generation of immune responses dominated by IL-10 secretion. The gene expression data suggest that regulatory T cells are involved, and the nature, localisation, and function of these cells will be important to determine. For instance, are these thymic Tregs or Tregs induced locally by high levels of IL-10 and immature DCs? Is this regulatory network driven by persistent antigen and what is the role if any of gut epithelial cells in maintaining local Tregs? The concept that regulatory T cells are required for the development of chronic inflammation is intriguing and suggests that these cells are more than simple anti-inflammatory agents. The signals that lead to the development and persistence of chronic inflammation are poorly understood and the involvement of regulatory T cells adds another element to the pathogenesis of chronic inflammatory disease (fig 1).

Figure 1

 Simplified pathways involved in the generation of effector and regulatory lymphocyte responses. Naïve lymphocytes that have been selected and matured in the thymus are able to enter secondary lymphoid tissue such as lymph nodes and Peyer’s patches where they interact with dendritic cells (DC) that have assimilated antigens in the gut. Interactions with DCs subsequently prime the naïve lymphocytes to the antigen and activate their differentiation into effector lymphocytes. Part of this differentiation is induction of tissue specific homing molecules that direct the effector cells back to the tissue where the antigen was found. During this process memory lymphocytes are also generated to allow for rapid expansion of effector cells if the specific antigen is encountered again. Regulation occurs at several levels with autoreactive lymphocytes being deleted in the thymus and generation of central/ thymic regulatory T cells (Tregs). Peripheral or induced Tregs are also generated locally by DC-lymphocyte interactions. Both sets of Tregs are able to control inflammation by the production of anti-inflammatory cytokines such as transforming growth factor β (TGF-β) and interleukin 10 (IL-10).

Persistent colitis is the result of a balance between local inflammation and regulatory networks. Regulatory T cells have potent anti-inflammatory effects and are likely to be important in the pathogenesis of chronic inflammatory bowel disease



  • Conflict of interest: None declared.

Linked Articles