The portion of the immune system resident within the intestine faces significant challenges. A single layer of epithelium separates the largest population of immune cells in the body from a massive number of bacteria. It is therefore probably not surprising that the mediation and control of intestinal immunity follows rules quite distinct from those governing systemic immune reactivity.
The overall challenges faced by the intestine not only include achieving efficient nutrient absorption, but also maintaining tolerance toward dietary antigens and the enteric microbiota, while retaining the ability to react vigorously to intestinal pathogens. Such balance of immunologic responses is made possible by the depth of interaction between the ancient innate immune system and the evolutionarily more recent adaptive immune system. The footprints of evolution are clearly seen within the immune system of the intestine: Different cells that first arose in distinct evolutionary eras work together within the human intestine. This has led to the addition of control mechanisms over time, rather than simple replacement of more archaic cell types by evolutionarily more modern successors. Dysfunction of cell types that have arisen relatively late in evolutionary history (for example regulatory T-cell subsets) can induce significant disturbance of intestinal immune homeostasis, although the effector mechanisms of the more ancient elements of the mucosal immune system function perfectly well.
An important question is why intestinal inflammation is not more common. The gut lumen contains 10 times as many bacterial cells as there are human cells in the entire body (10 14 vs. 10 13 ), while the majority of the body’s immune cells reside in the intestine, separated from these bacteria by a single epithelial layer. Each person ingests complex dietary antigens, which if injected parenterally would evoke systemic reactions. Thus there has necessarily been established a number of mechanisms that inhibit potential reactivity to both dietary antigens and the gut flora. Intestinal inflammation often occurs because of breakdown of these mechanisms.
There has been a great deal of study attempting to dissect such mechanisms. Many of the proof-of-principle studies have been in mice, and relatively less is known of human mucosal immunology. The same broad principles do, however, appear to apply, as evidenced by diseases occurring in people with genetic mutations that affect immune function. The mucosal immune system is undoubtedly very complex, with multiple cell types and mechanisms involved. This review attempts to steer a path between unhelpful oversimplification and bewildering overcomplexity. References are provided to review articles that will provide more in-depth detail.
First, this chapter provides an overview of important components of the intestinal environment that contribute to the maintenance of immune tolerance in such a potentially inflammatory environment. Later in the chapter, more detail is given about individual elements and mechanisms.
The Components of Gut Immune Responses
Many of the cell populations that cause tissue damage and inflammation are of innate immune origin, such as macrophages, neutrophils, eosinophils, mast cells, and dendritic cells. These cells may respond rapidly, but they do not directly exhibit immune memory. Products of some of these cell types may cause epithelial disruption, tissue breakdown, and vascular thrombosis. Some innate immune cells may respond directly to invading bacteria without prior involvement of adaptive immune cells (e.g., T cells and B cells). However, these effector cell types are most commonly recruited by induced chemotactic cytokine (chemokine) expression and may be activated by secreted T-cell products and/or immune complexes, reflecting a downstream effector response.
Innate immune responses may be shaped by adaptive immunity through products of either B cells or T cells. B cells themselves are shifted in their isotype from a default immunoglobulin M (IgM) polarity, dependent on the local cytokine environment and cell–cell contact with T cells. In general, the IgA responses characteristic of the intestinal mucosa protect against inflammation, whereas intestinal IgG responses are more proinflammatory. Deviation toward allergic IgE responses may also promote inflammation by disrupting epithelial barrier and neural function.
Among T cells, CD4-expressing helper (T H ) cells produce cytokines to alter function of other cells, while CD8-expressing cytotoxic (T C ) cells are capable of directly killing other cells. Among T helper cells, three major populations can drive different forms of intestinal immune reactions and inflammation—T H 1 cells (which produce the cytokines interferon γ [IFN-γ] and interleukin 2 [IL-2]), T H 2 cells (producing IL-4, IL-5, and IL-13), and T H 17 cells (producing IL-17). These cell types are discussed in more detail later in the chapter.
The commitment to such lineage and the functional state of T cells depend critically on input from the innate immune system, notably antigen-presenting cells (APCs). These are thus the most upstream part of the gut immune hierarchy. Sensing of bacterial luminal contents by dendritic cells is critical in this process ( Figure 5-1 ), as is the local cytokine environment that shapes dendritic cell–lymphocyte interactions. Thus T H 1 cells are generated by dendritic cells producing IL-12; T H 2 cells in response to IL-4; T H 17 in response to transforming growth factor β (TGF-β), IL-23, and IL-6; and T regulatory (T REG ) cells in response to TGF-β or IL-10. Consequently sensitization, rather than tolerance, may occur if pathogens induce local cytokine production at the time of initial priming.
The Generation of Inflammation
Pathogens may break immune tolerance by disrupting the epithelial barrier and/or inducing secretion of proinflammatory cytokines by resident subepithelial macrophages. In addition, they may induce expression of chemokines by the epithelium, leading to recruitment of other inflammatory cells. These newly recruited cells may in turn respond to other antigens that are penetrating the breached epithelial barrier, or self-antigens liberated from tissues as a consequence of tissue damage. Providing there is adequate repair of the epithelial barrier and clearance of the initiating pathogen or antigen, such inflammatory responses are normally damped down by regulatory immune responses, which are discussed in more detail later. The triggering of chronic inflammatory disorders by pathogens represents a failure of regulatory responses, or of epithelial barrier repair.
Mechanisms That Prevent Inflammatory Reactions to Gut Luminal Contents
Epithelial Integrity
The epithelium plays a very important role in mucosal immune responses. Epithelial barrier function is critical in preventing immune reactions to the gut flora and antigens. First, bacterial ingress is minimized by secretion of mucus by goblet cells and antibacterial peptides (such as α- and β-defensins) by Paneth cells. Paneth cell α-defensin production in fact shapes the composition of the bacterial flora, thus indirectly regulating mucosal T-cell responses. Two mechanisms may lead to disruption of this coordinated Paneth cell response to the normal flora—defects in either bacterial autophagy (a process of intracellular bacterial digestion and consequent immune presentation) or intracellular bacterial response (through loss-of-function polymorphism in the NOD2 pattern-recognition receptor). Both lead to suboptimal immune response to bacteria and are strongly associated with the development of Crohn’s disease.
Second, tight junction integrity limits penetration of antigens past the epithelium via the paracellular route, where they might be taken up by APCs. In health, peptide chains longer than 11 amino acids are normally prevented from penetrating; these are too short to evoke effective T-cell activation. Results of experimental studies of animals with leaky intestinal epithelium (caused by mutated cell adhesion genes) have confirmed that epithelial leakiness alone is sufficient to drive inflammation in response to the normal flora. Human genetic disorders with impaired gut epithelial adhesion (such as epidermolysis bullosa) are also characterized by inflammation. At the population level, increase in paracellular permeability is associated with nutritional failure, intestinal inflammation, and overall mortality in children in the developing world. Pathogens may directly cause such leakiness through epithelial damage, or they may trigger local production of cytokines such as tumor necrosis factor α (TNF-α) and IFN-γ, which promote paracellular leakiness. Their action is opposed by local production of the cytokine-transforming growth factor β, which in addition to promoting tight junction integrity also plays numerous roles in maintaining intestinal immune tolerance. Thus either infections or local inflammatory reactions may impair epithelial barrier function and thus promote secondary inflammatory or sensitizing events.
The epithelium also functions to regulate mucosal lymphocyte populations through constitutive secretion of chemokines such as CCL25 (TECK) in the small intestine and CCL28 (MEC) in the colon. These chemokines allow retention of B cells and T cells that have been primed within mucosal lymphoid follicles, following their circulation via the thoracic duct and subsequent homing to the mucosa. The epithelium also produces mediators that induce local adaptation of retained cells toward a regulatory, noninflammatory type. However, when epithelium is stressed or activated, it produces different chemokines that attract polymorphonuclear neutrophils (IL-8), monocytes (MIP-1α), T cells (CCL20), or eosinophils (eotaxins), depending on the initiating stimulus. One important and consistent feature of intestinal immune regulation is the different responses made by such newly recruited cells compared to the locally adapted populations. Thus the epithelium may play a critical role in determining the overall status of mucosal immune responses.
Finally, epithelial cells may play a role in antigen presentation that may promote tolerance, by presenting absorbed antigens to lymphocytes in an inherently non-sensitizing manner, because these cells do not express the costimulatory ligands required for full T-cell activation. T cells primed in this way may be rendered anergic and thus fail to respond to their cognate antigen.
IgA Production
IgA is generated in response to the gut flora and other luminal antigens, following their uptake by APCs and transport to lymph nodes within the gut wall. These mesenteric lymph nodes appear to be highly important in segregating mucosal from systemic immune responses and regulating intestinal tolerance mechanisms. IgA-producing plasma cells generated in the mesenteric lymph nodes then home to the gut from the circulation and go on to secrete specific IgA beneath the epithelium. This secreted IgA is in turn taken up and transported through the epithelial cells into the lumen ( Figure 5-2 ). This has two effects—secreted IgA adheres to bacteria and minimizes their invasiveness, and down-regulates the transcellular absorption of antigen through the epithelial cell (enterocyte). In contrast, secreted IgE accelerates antigen uptake by the enterocyte (and may induce tight junction leakiness through triggering of subepithelial mast cells). Thus there appears to be dynamic balance between IgA and IgE with respect of sensitization potential and maintenance of immune tolerance.
Regulatory Lymphocytes
These cells represent a critical component of the gut’s antiinflammatory repertoire and are discussed in more detail later in the chapter. Broadly, there are several types of regulatory lymphocytes, functioning in similar manner to inhibit inflammatory responses but differentiated by their pattern of surface molecule expression (e.g., CD4+ CD25+ T cells) or their cytokine production (e.g., TGF-β–producing T H 3 cells, IL-10 producing TR1 cells). One molecule is critical in generation of these regulatory cell types—the transcription factor forkhead box P3 (FOXP3). Mutations in FOXP3 cause a severe inflammatory autoimmune disorder, affecting the intestine and other organs (IPEX syndrome), confirming the importance of regulatory lymphocytes in preventing gut inflammation.
There is some evidence that mucosal IgA production and regulatory T-cell generation may function as a coordinated system, with regulatory T cells providing the major help for IgA responses, ensuring immunologic tolerance to the enteric flora.
Coordinated Immune Responses
It is not only the ability to make regulatory responses that inhibits inflammation. Animals deficient in a wide variety of immunologic molecules or cell types will spontaneously develop inflammation in response to the normal gut flora. This may relate to an inability to regulate the composition of the flora, allowing overgrowth of pathogens, or occur because the immunodeficiency predisposes to a pathologically skewed response to normal bacteria. A balanced immune response thus appears critical in preventing skewed and damaging intestinal inflammation. Infants with a variety of inborn immunodeficiency disorders develop intestinal inflammation, often remitting only after correction of the underlying disorder by gene therapy or bone marrow transplantation.
The Gut Flora
Normal mucosal tolerance cannot be established in the absence of a gut flora—animals maintained germ-free do not exhibit normal tolerance responses. However, it is important to note that different bacteria have different specific effects on this process. Thus, it is unclear how much the changes in the early gut bacterial composition of children that have been identified since the introduction of antibiotics in the last 50 years have contributed to the increased incidence of allergic and inflammatory diseases.
Generation of regulatory lymphocytes within the gut is at least partly dependent on the gut flora. This depends on signals from the epithelium and directly through dendritic cells, which require input from gut bacteria via pattern-recognition molecules (such as Toll-like receptors [TLRs]) to provide appropriate inductive signals to the T cells. The cytokine critical in induction of T REG is TGF-β, which is also important in developing IgA responses and maintaining epithelial integrity.
It is becoming clear that specific components of the flora, rather than the overall bacterial load, may be critical in the development of normal mucosal immune responses. Although much of the literature on probiotics has focused on the properties of lactobacilli and Bifidobacteria , other bacterial types appear much more important in maturation of the mucosal and systemic immune systems. A carbohydrate produced by Bacteroides fragilis induced both mucosal and systemic immune shift away from T H 2 toward T H 1 responses. Segmented filamentous bacteria and clostridia appear critical in maturing mucosal T helper cell and IgA responses in mice. It thus appears that, among the myriad bacterial species found in the gut, only relatively few have shaped host mucosal immune responses during evolution.
Micronutrients
Generation and function of regulatory T cells are also dependent on specific micronutrients—notably zinc, vitamin A, and vitamin D. Vitamin A is also important in the maintenance of epithelial integrity and the generation of gut-homing plasma cells within gut-associated lymphoid tissue (GALT). The consequence of micronutrient deficiency in intestinal inflammatory or allergic states may therefore be an inability to restore normal regulatory responses, and thus an exaggerated inflammatory response.
Organization of the Mucosal Immune System
GALT is organized within three compartments—diffusely scattered through the lamina propria beneath the intestinal epithelium, within the epithelial compartment itself, and in organized lymphoid follicles such as Peyer’s patches ( Figure 5-3 ).
- 1.
The diffuse lymphoid tissue of the intestinal lamina propria is dominated by plasma cells, most of which (in health) are IgA producing, although in early infancy IgM-producing cells are more common. T lymphocytes within this compartment are more commonly CD4+ rather than CD8+. These CD4+ cells may be subdivided functionally into T effector (T helper—T H ) and T regulatory (T REG ) cells. The T REG cells are particularly important in maintaining immune homeostasis within the intestine.
In addition, the mucosal lamina propria contains numerous dendritic cells and macrophages, most of which are locally adapted to their antigen-rich environment. During inflammatory responses, increased expression of chemotactic cytokines (chemokines) and other proinflammatory mediators leads to recruitment of additional T and B cells, monocyte/macrophages, and other cell types such as polymorphonuclear neutrophils, eosinophils, and mast cells. The pattern of cellular recruitment will depend on the polarity of T-cell responses that are induced following antigen presentation to the T cells by dendritic cells or macrophages. These cells from the innate immune system are finely attuned to local microbiologic influences, and therefore components of the gut flora may have a profound effect on overall immune responses within the intestine.
- 2.
The intraepithelial compartment contains populations of lymphocytes that are uncommon elsewhere in the immune system. Among T cells, around three-fourths of the intraepithelial lymphocytes (IELs) are CD8+ (i.e., cytotoxic T cells). Minority T-cell populations (type b IELs), the true function of which is uncertain in humans, include cells expressing neither CD4 nor CD8 (CD4-CD8-T cells), cells expressing CD8 with two α chains rather than the usual αβ combination (CD8αα cells), and cells with the T-cell receptor composed of γ and δ chains (γδ cells) rather than the α and β chains usually found in circulating T cells (αβ cells). There is also a significant population of natural killer (NK) and NK-T cells in this compartment. They may be involved in distinct mechanisms of antigen presentation based on enterocyte expression of nonclassical major histocompatibility complex (MHC) molecules.
Both T cells and NK cells jointly provide a surveillance role for the intestinal epithelium, and may be induced to cause cell death of enterocytes in circumstances of infection or local production of the cytokine IL-15.
- 3.
Organized lymphoid follicles occur throughout the intestine. They are most numerous in the terminal ileum, where they cluster to form macroscopically visible aggregates called Peyer’s patches, after Johan Conrad Peyer who reported them in 1677 in his Exercitatio anatomico-medica de glandulis intestinorum earumque usu et affectionibus , although thinking them to be glands producing digestive juices. The follicles do not have afferent lymphatics, and are notable for unusually permeable overlying epithelium, due to the presence of M cells (so called because of their ultrastructural appearance of microfolds). This permeability ensures penetration of luminal antigens into a subepithelial pocket containing large numbers of antigen-presenting dendritic cells. These lymphoid follicles are therefore able to sample and respond to a wide variety of luminal antigens, both bacteriologic and dietary. Efferent lymphatics from the Peyer’s patches drain to the mesenteric lymph nodes, where immune responses are further amplified.
The appendix is a specialized intestinal region with dense aggregation of lymphoid follicles. It appears to be important in mucosal immune priming, as appendectomy protects against later development of ulcerative colitis, and neonatal appendectomy prevented later development of colitis in mutant mice. The appendix is now thought to function as an immune-mediated reservoir for the indigenous host flora, allowing repopulation of the colon after infection. The bacteria adhere to biofilms, enriched in mucus and defensins from the innate immune system and IgA from the adaptive immune system. The similarity of such biofilms in mammals and nonmammalian vertebrates, including frogs, suggests an ancient origin for immune support of indigenous bacterial species. The mucous biofilm is capable of excluding bacteria from the colonic epithelial surface in health, and may indeed deliver tolerogenic signals to the mucosal immune system, although this barrier becomes defective in intestinal inflammation. Blind outpockets of the distal gut, similar to the appendix, have arisen by convergent evolution across many unrelated species, suggesting a more important function than had previously been ascribed to the appendix.
Bacterial translocation into organized mucosal lymphoid follicles has been studied in resected appendix tissues from human infants. This provides insight into the initial reactions to early colonizing bacteria within mucosal lymphoid tissue. Bacterial translocation within the appendiceal mucosa was identified in all specimens from infants older than 2 weeks, with whole bacteria identified beneath follicular epithelium, within follicles, and in efferent lymphatics. Few lymphoid follicles were present at birth, increasing rapidly on colonization, with germinal centers identifiable by 4 weeks. IgM plasma cells increased rapidly from 2 weeks, declining from 6 weeks as IgA plasma cells began to dominate, and reaching their peak at around 10 weeks.
Components of the Mucosal Immune System
Innate Immunity within the Intestine
It is particularly important to recognize that the gut is an organ of huge evolutionary longevity—indeed well-developed gastrointestinal tracts can be identified in fossils of organisms from the Cambrian period 600 million years ago. Thus, immunologic tolerance of gut luminal contents must have been established prior to the development of any adaptive immune responses. Many products of innate immunity, in addition to defensins, including C-type lectins, surfactants, and cathelicidins, contribute to shape the host’s immune response to the flora and indeed the composition of the flora itself.
Several cells of innate immune lineage play roles in presenting antigens to lymphocytes of the adaptive immune system. Their own pattern of activation helps to shape subsequent adaptive immune responses. Professional APCs, such as dendritic cells, macrophages, and B cells, can efficiently take up antigen (by phagocytosis or specific receptor-mediated uptake) and then present fragments of that antigen complexed with class II MHC molecules to naive T cells. The consequent antigen-specific T-cell response will be shaped by both expression of costimulatory molecules and secretion of cytokines by the APC.
Dendritic cells are the most efficient activators of T cells because of their constitutive expression of costimulatory molecules such as B7-1 (CD80) or B7-2 (CD86). There is important functional heterogeneity within populations of professional APCs—thus both dendritic cells and macrophages may exist as locally adapted resident populations or represent recently recruited cells, derived from bone marrow, which exhibit a more proinflammatory phenotype. Such local adaptation is one of the key mechanisms underpinning the maintenance of immune tolerance within the intestine.
The intestinal epithelium can contribute to antigen presentation by processing ingested antigen and presenting using both classical and nonclassical MHC molecules. However, the enterocyte does not express costimulatory molecules such as B7-1 or B7-2, so this form of antigen presentation does not activate lymphocytes but may render them anergic—incapable of proliferation and activation.
Other cell types in the intestinal mucosa, including fibroblasts and vascular endothelial cells, may act as nonprofessional APCs. Their interactions with lymphocytes may become functionally important in inflammatory states, but are unlikely to play a role in the normal maintenance of immune tolerance. This review thus focuses first on the primary interactions between innate and adaptive immune cells in establishing and maintaining tolerance to dietary antigens and the enteric flora ( Table 5-1 ).
Recognition Element | Microbial Component | Effect Transduced |
---|---|---|
TLR-2 | Peptidoglycans | NF-κB response |
TLR-4 | Lipopolysaccharides | NF-κB response |
TLR-5 | Flagellins | NF-κB response |
TLR-9 | Bacterial DNA | NF-κB response |
NOD-1, NOD-2 | Bacterial molecules | NF-κB response |
Mannose receptor | Bacterial carbohydrates | ↑ Ag presentation |
Complement components | O- and N-linked glycans | Opsonization, possible regulatory response |
Mannan-binding lectin | Bacterial carbohydrates | Complement activation |
Surfactant proteins A and D | O- and N-linked glycans | ↑ Phagocytosis, regulate T-cell and macrophage activation |
Dendritic Cells within the Intestine
Dendritic cells play a central role in the maintenance of immunologic tolerance within the intestine, through their primary role of taking up antigens and presenting them to lymphocytes. They provide an important means of sampling luminal contents—both microbial and dietary in origin. Three distinct mechanisms have been identified by which such sampling may be effected.
First, dendritic cells cluster in the subepithelial region of organized lymphoid follicles, such as Peyer’s patches. Specialized epithelial cells in the surface epithelium, so-called microfold or M cells, are much more permeable to luminal antigens than are normal epithelial cells. Such focal epithelial leakiness allows ingress of luminal antigens of all kinds. However, there may be some specificity in uptake, because M cells express the lectin glycoprotein-2, which allows selective adherence and uptake of fimbriated bacteria. Bacterial or dietary components crossing the M cells are then in turn taken up by dendritic cells. These may present processed antigen to T cells within the local area of the lymphoid follicle (see Figure 5-1 ). In addition, it has been demonstrated that dendritic cells in Peyer’s patches may phagocytose live bacteria that have penetrated through M cells and may then migrate to the regional draining mesenteric lymph nodes. It is in this site that fundamental adaptive immune responses may occur, including generation of antigen-specific IgA.
Second, it is now known that subepithelial dendritic cells, situated away from organized lymphoid follicles, may insinuate processes between adjacent enterocytes to sample luminal contents. This appears to be a coordinated mechanism involving induced focal breakdown of the mechanisms that normally maintain tight junction integrity.
Third, antigen may be transported through the enterocyte following uptake either by IgG, which is shuttled back and forth across the epithelium by the neonatal Fc receptor for IgG, or by IgE, which is taken up by induced luminal expression of the low-affinity IgE receptor CD23 (see Figure 5-2 ). This antibody-mediated uptake will thus be antigen-specific, rather than the less-selective uptake across Peyer’s patches or by intraluminal dendritic cell sampling.
Conserved Pattern-Recognition Receptors and Dendritic Cell Function
Dendritic cells do not present antigen in isolation from the massive numbers of enteric bacteria situated so close to them across the epithelial barrier. Indeed, these bacteria induce profound changes in the behavior of the entire enteric immune system, and may even shape systemic immune responses away from the intestine. The effects of the enteric flora on the behavior of dendritic cells are mediated through a number of highly conserved pattern-recognition molecules. These may be situated on the cell surface or may be expressed intracellularly. Pattern-recognition molecules in both extracellular and intracellular sites signal through shared proinflammatory pathways, converging on nuclear transcription factor-κB (NF-κB). On the cell surface, various TLRs recognize conserved sequences in bacteria, viruses, fungi, and protozoa. Similarly, within the cell, NOD1 and NOD2 recognize sequences in bacterial cell walls. Binding of the conserved microbial sequence by these pattern-recognition receptors transmits a signal through NF-κB that induces nuclear transcription of cytokines such as TNF-α. This has the effect of altering the interaction between the APC and any lymphocytes with which it interacts.
Subgroups of Dendritic Cells
Dendritic cells may be subdivided functionally into myeloid (monocyte-like) or plasmacytoid (plasma cell-like). Myeloid dendritic cells produce predominantly the cytokine IL-12 and plasmacytoid cells IFN-α, which may affect the behavior of cells in their vicinity and the subsequent polarization of lymphocytes to which they present antigen.
In comparison to dendritic cells from the spleen, intestinal dendritic cells tend to produce more of the regulatory cytokine IL-10, which may contribute to the maintenance of immune tolerance in such a highly antigen-challenged site. Within Peyer’s patches, a subset of dendritic cells expressing the CD11b molecule promotes a more T H 2 skewed response among T cells, whereas subgroups that do not express this molecule (CD11b − ) induce a more T H 1 skewed response. Expression of CD103 (αE integrin) by Peyer’s patch dendritic cells is associated with a tendency toward T H 2 or regulatory cell polarization. The factors that determine expression of markers such as CD11b and CD103 are not well understood in humans, and there appear to be a number of other subsets with different surface marker expression and functions ( Table 5-2 ).
Cell Type | Identifying Markers | Effects |
---|---|---|
Peyer’s patch dendritic cells | CD11b+ | T H 2 skewed response T REG response |
CD11b− | T H 1 skewed response | |
CD103+ | T H 2 skewed response T REG response | |
CD103− | T H 1 skewed response | |
Lamina propria dendritic cells | CD103+ resident cells | T H 2 skewed response T REG response |
CD103− newly recruited | T H 1 skewed response | |
Resident macrophages | CD14− | Reduced LPS response |
Newly-recruited macrophages | CD14+ | Full LPS response (TNF-α, etc.) |
Polymorph neutrophils | CD11b/CD18, CD66b (activated) | Release proteases, free radicals, G-CSF, IL-8, etc. |
Mast cells | Mast cell tryptase, c-kit | Release tryptase, histamine, 5-HT, TNF-α |
Eosinophils | Eosinophil peroxidase, CD66b (activated) | Release ECP, IL-4, histamine, leukotrienes |
Basophils | CD63 (activated), CCR3 | Release IL-4, histamine, leukotrienes |
Natural killer cells | CD16, CD56 | Induced apoptosis |
Although the field is complex, the overall pattern is that cells that have become locally adapted to the intestinal lamina propria generally inhibit the development of delayed-type hypersensitive reactions, in a manner not seen among splenic or Peyer’s patch lymphocytes, thus promoting immune tolerance. In contrast, newly arrived dendritic cells recently derived from the bone marrow exhibit unrestricted responses to antigens and bacterial products in the intestinal microenvironment.
The gut flora may play an important role in the conditioning of dendritic cells within the intestine to such local adaptation. This important change in the functional properties of these cells depends in part on molecules released by intestinal epithelial cells on bacterial exposure, including thymic stromal lymphopoietin (TSLP) and retinoic acid (a vitamin A derivative). Other cytokines that contribute to this process include IL-10 and TGF-β, which may be produced by a number of cells within the microenvironment, including other locally adapted dendritic cells.
Dendritic Cells and Induction of Immune Tolerance within the Intestine
A central mechanism for maintenance of tolerance within the intestinal environment is the induction of a regulatory phenotype in T cells that interacts with the locally conditioned dendritic cells. As discussed, the transcription factor FOXP3 and the cytokine TGF-β are critical components in the transition of a naive T cell to a regulatory phenotype (T REG ). Subgroups of locally adapted Peyer’s patch and lamina propria dendritic cells (expressing CD103) and lamina propria macrophages are particularly effective in inducing FOXP3 expression in naive T cells.
In addition, dendritic cells may alter the homing potential of T cells with which they interact by inducing expression of specific integrins that favor homing back to the gut, following passage from efferent lymphatics to the thoracic duct and back into the circulation. Finally, Peyer’s patch and mesenteric lymph node dendritic cells play a role in the isotype shift of B cells toward IgA, which dominates intestinal immunoglobulin production in health and contributes to maintenance of intestinal homeostasis.
Dendritic Cells and Effector Immune Responses to Pathogens
Dendritic cell function is clearly not restricted to the induction of tolerance in all circumstances. This would be inappropriate in the case of pathogens, which require prompt responses from the mucosal immune system. This response may follow recruitment of new dendritic cells and macrophages, which have not undergone local conditioning. The response of epithelial cells to pathogen-induced damage includes expression of chemokines such as IL-8 and MIP-3α, which induce cell recruitment, and cytokines such as IL-1, IL-15, and TNF-α, which may activate or prime locally recruited cells. The consequences will be an appropriate proinflammatory response and the generation of effector and memory T cells, polarized toward appropriate immune responses upon future challenge.
As mentioned previously, micronutrient status is particularly important in dendritic cell function, and thus the establishment and maintenance of immune tolerance. In particular, vitamins A and D and zinc are essential factors in the ability of intestinal dendritic cells to induce regulatory T cells and IgA responses. Thus treatment of established micronutrient deficiency in enteropathy or other inflammatory states may be clinically important.
Intestinal Macrophages
Macrophages are highly important effector cells, capable of producing more than 100 mediators upon activation. Among these mediators, the molecules TNF-α, IL-1β, and IL-6 have important proinflammatory effects. Excess production of TNF-α and IL-1β has been particularly associated with intestinal inflammatory conditions, and therapeutic inhibition of these molecules by biologic therapies has had profound effects on complex inflammatory responses in vivo. As with dendritic cells, macrophages express an extensive range of bacterial pattern-recognition receptors, notably TLRs. The response of macrophages to TLR ligation is a much more potent proinflammatory response than that seen in dendritic cells mediated through the transcription factor NF-κB, in which cytokines such as TNF-α, free oxygen radicals, proteases, and nitric oxide are released. In addition, the secretion of enzymes by macrophages, such as matrix metalloproteases, may have important effects on extracellular matrix integrity and their release of endothelins on vascular supply.
Similar to the pattern seen in dendritic cells, there is evidence of important local adaptation among macrophages. Intestinal macrophages do not proliferate, and their numbers are continually replenished by blood-derived monocytes, which in turn become locally adapted. As with other regulatory mucosal responses, the cytokine TGF-β plays a critical role in the transformation from newly recruited monocytes to locally adapted macrophages.
Locally conditioned intestinal macrophages do not have a full reaction to bacterial lipopolysaccharides, as they have down-regulated expression of CD14, a molecule critical to the function of TLR-4 in its inflammatory response to bacterial LPS. In addition, resident lamina propria macrophages show down-regulated expression of receptors for IgG and IgA, while retaining strong phagocytic and bactericidal activity. Resident macrophages also contribute significantly to normal tolerance to the flora by depleting the lamina propria environment of tryptophan, necessary for full T-cell activation, through expression of the enzyme indoleamine 2,3-dioxygenase (IDO). Locally adapted macrophages may also play an important immunomodulatory role during gut inflammation by secreting IL-10, which in turn induces a local regulatory T-cell response.
Lamina propria macrophages have important effector roles in host defense against invading microorganisms. They kill most ingested bacteria more efficiently than can unadapted monocytes, despite their relative lack of proinflammatory response. They are also able to neutralize viruses of many kinds, thereby functioning as effective gatekeepers to the lamina propria. However, when large-scale influx of newly recruited monocytes occurs in response to chemokine expression during inflammatory responses, these newly recruited former monocytes produce large amounts of proinflammatory cytokines and may thus potently amplify mucosal inflammation. Within the inflamed mucosa in Crohn’s disease, around one-third of mucosal macrophages express CD14, and are thus recently recruited cells able to make an uninhibited response to bacterial LPS. Important in this influx is a subgroup of cells that show characteristics of both macrophages and dendritic cells, both presenting antigen and promoting both T H 1 and T H 17 responses.
Polymorphonuclear Neutrophils
Polymorph neutrophils do not play a significant role in intestinal antigen presentation, and their most important contribution is in the proinflammatory response to pathogens. Activation of intestinal epithelial cells by pathogens induces secretion of the chemokine IL-8, which leads to enhanced neutrophil recruitment. Neutrophils then become involved in immediate responses to invading pathogens, and may damage tissue through release of proteases, cytokines, and reactive oxygen and nitrogen radicals.
Although their best-recognized role in host defense is in immediate proinflammatory responses, the role of neutrophils within the intestinal microenvironment is more complex and nuanced. This is demonstrated by the development of intestinal inflammation in disorders of neutrophil function, such as chronic granulomatous disease or glycogen storage disease 1b. Impaired neutrophil function has been linked more generally to the development of inflammatory bowel disease (IBD), and enhancement of neutrophil function by stimulatory factors such as granulocyte colony-stimulating factor (G-CSF), may have an antiinflammatory effect in Crohn’s disease.
Eosinophils, Basophils, and Mast Cells
There is overlap of function between these cell types, all of which are involved in T H 2 type immune responses within the intestine. All appear important in host defense against helminth infection, and may have effects on intestinal motility. Upon activation, which frequently occurs in the context of IgE-mediated intestinal reactions, these cell types produce an overlapping array of cytokines and proinflammatory mediators that induce vascular permeability and promote antigen penetration. Activation of these cell types may also directly affect intestinal neural function. Mast cells are closely situated to enteric nerves and share a common origin (dependent on c-kit ligand) with the interstitial cells of Cajal that function as pacemaker cells within the myenteric plexus. Eosinophil- and mast cell–dominated gut disorders are characterized by dysmotility and enhanced pain sensation (visceral hyperalgesia).
Recruitment of eosinophils is particularly dependent on the T H 2 group cytokine IL-5 and the eotaxin subfamily of cytokines. Commitment of precursor cells within the bone marrow to the eosinophil lineage is dependent on the transcription factor GATA-1.
Eosinophils are constitutively present at low density in most of the gastrointestinal tract, with the exception of the esophagus where they are usually absent in health. In addition to effector functions during inflammatory reactions, eosinophils can also function as APCs, inducing antigen-specific T-cell stimulation.
For reasons currently unclear, there has been rapid temporal increase in eosinophilic gut disorders, in particular eosinophilic esophagitis. In such disorders, there is frequently an increase in tissue mast cell and basophil density, pointing toward a coordinated immune response. This is likely to represent a conserved mechanism for combating intestinal helminth infection, which has been almost ubiquitous throughout evolutionary history. Whether the relative absence of helminth infection in privileged modern societies actually contributes to dysregulation of this coordinated response through lack of normal induction and priming is the subject of much interest.
Adaptive Immunity within the Intestine
Adaptive immune responses in the gut are mediated by cells of both T-cell and B-cell lineage. The earlier rather simple differentiation of T-cell populations into CD4 (helper) and CD8 (cytotoxic) cell types and functional subdifferentiation into T H 1/T c 1 and T H 2/T c 2 cells based on cytokine secretion patterns, now appears to represent a gross underestimate of a highly varied grouping of many cell types, each capable of modulating antiinfective or inflammatory responses. Much of the data on such subpopulations come from murine study and must be interpreted with caution. However, there is no doubt that the intestinal mucosa hosts a large array of different lymphocyte subpopulations and that there may be highly complex levels of control that are only partly understood.
Archaic Lymphocyte Populations
The intestine is unusual in that it maintains relatively high expression of cells that arose much earlier in evolution than classical T and B cells. Some of these cells function on the borderline between innate and adaptive immunity, maintaining the ability to provide rapid response to newly encountered pathogens, while also demonstrating some elements of immune adaptation. It may not be coincidental that these cells are highly represented in the epithelial compartment, where exposure to luminal organisms and pathogens may he highest. They are known as type b IELs.
NK-T cells show some overlap of function with NK cells of the innate immune system, but differ in their ability to produce high levels of cytokines such as IL-2 and IFN-γ. All are restricted in their antigen recognition by the nonclassical MHC molecule CD1d, whereas some possess an invariant T-cell receptor α chain (Vα24 NK-T cells), which responds to lipid antigens presented by the nonclassical MHC molecule CD1d expressed by the epithelium. This is important in host defense against mycobacterial glycolipids but may be subverted in allergy, and allergic responses to dietary lipid antigens may be mediated in this manner.
Most γδ T cells within the intestine are of a type (Vδ1) that is uncommon in peripheral blood. They express receptors that are more akin to NK-T cells than conventional αβ T cells. They particularly recognize stress-induced molecules (MICA, MICB) on epithelium, and are thus thought to play a particular role in surveillance of epithelial integrity. Overall, γδ cells thus appear to protect the epithelium, possibly by elimination of stressed or infected cells. Although best recognized for their increase within the epithelium in celiac disease, there is experimental evidence to suggest that lack of γδ cells may cause an amplification of tissue damage in intestinal infection or inflammation. However, in other circumstances, γδ cells may contribute to inflammatory damage.
B-Lymphocyte Populations
Intestinal B cells also show important differences from circulating B-cell populations. There is overrepresentation of an archaic cell type unusual in the circulation (B1 cells), which arose earlier in evolution than conventional B cells (B2 cells). Although B1 cells can produce antibody and present antigen, they do not mature into memory cells. Most intestinal B1 cells express CD5, a molecule involved in B-B cell interaction. They predominantly produce IgM of broad specificity (natural antibody), binding particularly to bacterial carbohydrates. B1 cells migrate to the intestine from the peritoneal cavity, and may undergo isotype shift to IgA within the mucosa, although this remains controversial. B1 cells form a first line of defense against bacterial invasion from the gut lumen by contributing to immunoglobulin coating of bacteria within the lumen.
The isotype of conventional B2 cells is also skewed compared to elsewhere in the body, with great predominance of IgA-producing cells generated within Peyer’s patches and mesenteric lymph nodes. Around 80% of human plasma cells are located in the gut, with almost all producing secretory IgA, leading to excretion of approximately 3 g daily. Within the small intestine, as in plasma, IgA1 (specific for protein antigens) is the dominant secretory isoform, whereas in the colon, IgA2 (specific for bacterial LPS and lipoteichoic acid) dominates.
Shift in immunoglobulin isotype from the default IgM occurs under the influence of local cytokines, but it is also dependent on direct cell-cell contact with T cells, through the CD40-CD40 ligand interaction. As required for induction of a regulatory phenotype in T cells, the generation of IgA-producing plasma cells appears to be dependent on the normal flora and the cytokine TGF-β.
Although most circulating IgA is monomeric, most intestinal luminal IgA is of the secretory type, consisting largely of dimers and tetramers, joined by a polypeptide J-chain and stabilized by a molecule called secretory component that provides resistance to proteolysis. The complex is taken up by the polymeric Ig receptor on enterocytes, and then shuttled across the enterocyte to be secreted into the lumen. In addition to protecting secretory IgA from proteolysis, this receptor may itself play a role in immune responses by direct antimicrobial effects and by inhibiting pathogen and antigen ingress through the epithelium.
Luminally secreted IgA performs a number of functions that tend to diminish inflammation, including reducing uptake of particulate antigens, neutralizing biologically active molecules, inhibiting bacterial adherence, and enhancing activity of innate immune factors such as lactoferrin. Within the enterocyte, IgA can retard transfer of pathogens including HIV and can aid elimination of immune complexes, while within the mucosa IgA has antiinflammatory activities including complement inhibition and contributing to bacterial opsonization. Thus IgA-deficient individuals show increased uptake of food antigens and may demonstrate low-grade enteropathy.
Homing and Recruitment of B Lymphocytes
Common to both B1 cells and conventional (B2) cells is the ability to home to the mucosal surface. This is mediated in the high endothelial venules of GALT and mesenteric lymph nodes by expression of mucosal addressin cell adhesion molecule 1 (MadCAM-1), which interacts with L-selectin on lymphocytes, followed by specific binding of those expressing the mucosal integrin α4β7. Following recruitment of lymphocytes by this mechanism, they are held within the intestine by local chemokine expression. Within the small intestine, epithelial production of the chemokine CCL25 (TECK) induces retention in the lamina propria of both T and B cells expressing the chemokine receptor CCR9. Regional variation within the intestine of chemokine production by the epithelium induces homing of specific subgroups of T and B cells, so that colonic tropism is mediated by interaction between epithelial CCL28 (MEC) and lymphocyte CCR10.
Induction of Mucosal IgG and IgE Responses
Immunoglobulin class switching within the intestine is not always or entirely directed toward IgA. In the presence of cytokines other than TGF-β, isotype shift toward IgG or IgE may occur. Thus during inflammatory or pathogen-induced reactions, the production of T H 1 or T H 2 cytokines by T cells within the lymphoid follicle may ensure that naive B cells are committed toward IgG2 (IFN-γ) or IgE (IL-4), so that they mature into gut-homing IgG2- or IgE-producing plasma cells. These would be retained within the lamina propria by chemokine interactions, as above. However, their interaction with antigens would induce a distinct immunologic consequence relative to that of IgA.
T-Cell Populations in the Intestine
As discussed, intraepithelial T cells are usually of the CD8+ (cytotoxic) type, whereas lamina propria T cells are more commonly CD4+ (helper) cells. There is functional subdivision of T-cell responses based on the pattern of cytokines that these cells produce on activation ( Table 5-3 ). In contrast to previous dogma, there is emerging evidence that some T helper cells can alter lineage commitment within the gut, particularly between T H 17 and T REG phenotype, depending on local environmental inputs. Long-lived populations of both CD4+ and CD8+ cells provide important immunologic memory within the lamina propria.