Professor Dr A. U. Dignass, Charité Medical School — Virchow Hospital, Department of Medicine, Division of Hepatology and Gastroenterology, Augustenburger Platz 1, D-13353 Berlin, Germany. E-mail: firstname.lastname@example.org
Although the aetiopathogenesis of Crohn's disease and ulcerative colitis, remains unsolved, current evidence indicates that defective T-cell apoptosis and impairment of intestinal epithelial barrier function play important roles in the pathogenesis of both conditions. Without appropriate control of T-cell proliferation and death during an immune response, an inappropriate accumulation of T cells and subsequent intestinal inflammation may occur. Differences in T-cell responses between Crohn's disease and ulcerative colitis have been identified, with mucosal T-cell apoptosis being defective in Crohn's disease, but not in ulcerative colitis. Furthermore, cell cycling is considerably faster, with a vigorous clonal expansion, in Crohn's disease, whereas, in ulcerative colitis, T cells cycle normally, but have a remarkably reduced capacity to divide and expand. The elimination of excessive T cells therefore seems to be a reasonable approach to restore the gut to a physiological state or, at least, a controlled state of inflammation. The tumour necrosis factor-α blocker, infliximab, exerts its beneficial effects, at least in part, by the induction of apoptosis in lamina propria T cells and monocytes. In addition, repeated damage and injury of the intestinal surface is a hallmark of inflammatory bowel disease and may facilitate the entry of luminal antigens into the mammalian organism and the initiation and perpetuation of both nonspecific and specific immune responses. A better understanding of and enhancement of intestinal repair mechanisms may thus provide future approaches for the treatment of inflammatory bowel disease.
The surface epithelium of the alimentary tract creates an essential barrier to a broad spectrum of potentially immunogenic and noxious factors within the intestinal lumen. The epithelial interface separates the lumen, with its residential and nonresidential flora, food components and secretory products (from the salivary glands, stomach, small and large intestine, and pancreas), from the specific mucosa-associated immune system. Intestinal inflammation is a common response mounted by the mammalian organism against infectious agents or other insults. The type of antigen generally initiates the type of inflammatory response that is most appropriate for elimination. This explains the diversity of intestinal inflammatory conditions observed in clinical practice. The intrinsic genetic make up of the host also plays a key role in determining the duration and final outcome of the inflammatory response. Acute inflammation induced by common bacterial agents usually involves an immediate and relatively simple response orchestrated by polymorphonuclear cells that eliminate the invading microbe. In contrast, in chronic inflammation of unknown origin, more diverse types of immune cells are involved, which interact in complex ways to sustain and prolong inflammation. The latter scenario occurs in inflammatory bowel disease. In general, immune responses can be nonspecific (innate immunity) or highly specific (specific acquired or adaptive immunity).
The intestinal surface barrier
The surface of the gastrointestinal tract is covered by polarized epithelial cells that constitute an efficient physical barrier and also facilitate an exchange between nutrients and the systemic circulation. Mucosal defence mechanisms can be categorized into three key components: pre-epithelial, epithelial and postepithelial, the latter being represented by the lamina propria1–3(Figure 1). The pre-epithelial mucus barrier is composed of mucin associated with other proteins, such as trefoil peptides and lipids, and forms a continuous gel into which a bicarbonate-rich fluid is secreted, maintaining a neutral pH at the epithelial surface. Mucus is secreted by intestinal epithelial cells and intestinal immune cell populations, and consists of trefoil peptides and various glycoproteins, phospholipids, secretory immunoglobulin A and a glycocalyx.2–5 The tight adherence of mucin to the apical surfaces of the epithelia results from a specific complex between mucin oligosaccharides and a mucin-binding protein on the apical mucosal membrane.5 The hydrophobic lining of the luminal surface has an important functional role: it prevents microorganisms from adhering to the plasma membrane and, furthermore, protects the mucosal epithelium against chemical and mechanical injuries.6
Epithelial cells provide the second line of the mucosal defence system. Whereas, in the upper digestive tract, this layer consists of a stratified epithelium, the stomach and small and large intestine are covered with a simple single epithelial layer sealed by tight junctions.1,7 In the healthy organism, the uptake of antigens and microorganisms through this layer is restricted by the luminal cell surface structures. The mucosal surface epithelium provides a protective barrier to a broad spectrum of physiological factors (intestinal flora, nutrients, etc.), and to noxious and immunogenic substances present in the lumen of the gastrointestinal tract (pathogenic microorganisms, toxic products of digestive processes, proteases, etc.) and the underlying mucosa-associated immune system. However, this barrier is not impermeable and the possibility of flux through the epithelium via both transcellular and paracellular pathways exists. Interestingly, the barrier properties of the gastrointestinal mucosa vary in different parts of the intestinal tract.8 In addition to intestinal epithelial cells in toto, various tight junction proteins represent a series of discrete membrane contacts of adjacent cells. They consist of a cluster of protein species and, although additional constituents of this complex remain to be identified, they include the cytoplasmic proteins, zonula occludens ZO-1, ZO-2, ZO-3, claudins, cingulin and 7H6, as well as the transmembrane proteins, occludin and cadherins.9,10 Occludins and claudins may serve as the major sealing proteins. The permeability of tight junctions may be regulated by modulating the expression of tight junction components, especially certain claudins, and, possibly, the phosphorylation of certain tight junction proteins.
The third line of the mucosal defence system, the postepithelial defence, is formed by various constituents of the lamina propria and will be discussed later.
Mechanisms of intestinal epithelial healing
In general, the enormous regenerative capability of the mucosal surface epithelium ensures that the integrity of the intestinal mucosal surface barrier is rapidly re-established, even after extensive destruction. Rapid resealing of the surface epithelium is accomplished by epithelial cell migration, also termed epithelial restitution, followed by subsequent epithelial cell proliferation, differentiation and maturation. Damage to the intestinal barrier is a hallmark of inflammatory bowel disease and may permit an increased penetration and absorption of toxic and immunogenic factors, leading to or perpetuating inflammation and an uncontrolled immune response. Thus, rapid resealing of the epithelial surface barrier following injuries or physiological damage is essential to the preservation of normal homeostasis. The intestinal tract is able to rapidly re-establish the continuity of the surface epithelium after extensive destruction, and this occurs by at least three distinct mechanisms.2,11,12 First, epithelial cells adjacent to the injured surface migrate into the wound to cover the denuded area. Those epithelial cells that migrate into the wound defect de-differentiate, form pseudopodia-like structures, re-organize their cytoskeleton to extend into the wound and then re-differentiate after closure of the wound defect. This process has been termed epithelial restitution. It does not require cell proliferation and has been shown to occur within minutes to hours, both in vivo and in vitro. Subsequent epithelial cell proliferation is necessary in order to replenish the decreased cell pool and, finally, maturation and differentiation of undifferentiated epithelial cells are needed to maintain the functional activities of the mature mucosal epithelium. In vivo, these three wound healing processes overlap. Deeper lesions or penetrating injuries require additional repair mechanisms that involve inflammatory processes, degradation and the formation of extracellular matrix components and various nonepithelial cell populations, such as immunocytes, myofibroblasts, platelets and others. Inflammatory processes interfere with epithelial cell migration and proliferation, and thus modulate intestinal epithelial healing.
Modulation of intestinal epithelial wound healing
The repair of the intestinal surface epithelium is regulated by a complex network that includes a broad spectrum of structurally distinct regulatory peptides, nonpeptide factors, extracellular matrix factors and direct cell–cell interactions. Regulatory peptides, conventionally designated growth factors or cytokines, have been shown to play an essential role in regulating differential epithelial cell functions to preserve normal homeostasis and integrity of the intestinal mucosa. In addition, nonpeptide molecules, including phospholipids, short-chain fatty acids, adenine nucleotides, trace elements and pharmacological agents, have been demonstrated to modulate intestinal epithelial repair mechanisms.2,11 Intestinal epithelial healing is even further complicated by direct cell–cell interactions, which build another dimension to a seemingly never-ending interactive network within the alimentary tract. The identification and characterization of numerous factors involved in intestinal repair have led to the recognition of a network of interrelated factors within the intestine2,11(Figure 2). Various regulatory peptides have been demonstrated to modulate intestinal wound healing. The key activities with respect to intestinal wound healing of some important regulatory peptides are summarized in Table 1.
Table 1. Functional activities of regulatory peptides with respect to wound healing within the intestinal mucosa
A significant number of regulatory peptides, including transforming growth factor-α, epidermal growth factor, transforming growth factor-β, Activin A, hepatocyte growth factor, fibroblast growth factor, interleukin-1, interleukin-2 and interferon-γ, have been demonstrated to enhance epithelial cell restitution.2,11,13,14 The above-mentioned growth factors and cytokines are assumed to act from the basolateral site of the epithelial surface and seem to stimulate intestinal epithelial restitution through a common transforming growth factor-β-dependent pathway (Figure 3). However, various members of the trefoil factor family (TFF peptide family) appear to stimulate epithelial restitution, in conjunction with mucin glycoproteins, through a transforming growth factor-β-independent mechanism from the apical pole of the epithelium.15,16
In addition to their potent effects on epithelial restitution, a number of regulatory peptide factors also act as potent modulators of epithelial cell proliferation.2,11,17,18 The most important modulators of intestinal epithelial cell proliferation include epidermal growth factor and transforming growth factor-α, both of which act as potent stimulators of intestinal epithelial cell proliferation, and transforming growth factor-β, which inhibits intestinal epithelial cell proliferation and plays an important counterbalancing role in the regulation of this process.17,19 Transforming growth factor-β is the most potent inhibitor of intestinal epithelial cell proliferation, overriding the stimulatory effects of other stimulatory factors.
A broad spectrum of nonpeptide factors has been demonstrated to modulate the repair of intestinal injury (Table 2). These encompass unrelated factors, such as phospholipids, nutrients, adenine nucleotides, polyamines, short-chain fatty acids, products of the intestinal microflora, trace elements, pharmacological agents and other factors.2,11 Some of these factors are released by injured or dying mucosal cell populations (e.g. adenine nucleotides, phospholipids), whereas others reach the intestinal mucosa via the intestinal lumen or the bloodstream. These nonpeptide factors may exert growth factor-like activities and have potent effects on cell growth and differentiation in different cell populations, including fibroblasts, vascular smooth muscle cells, endothelial cells and keratinocytes.2,3,11,20–24 As some of these nonpeptide factors are stable within the gastrointestinal tract despite high concentrations of acid, bile salts, proteases and microorganisms, they may serve as potential future targets to improve the armamentarium for the healing of mucosal epithelial injury. Interestingly, drugs that are used for the treatment of inflammatory bowel disease may also interfere with wound healing processes. It has been recognized for a long time that surgical intervention, with concomitant therapy with corticosteroids in therapeutic doses, is often complicated by impaired intestinal wound healing and insufficiency of intestinal anastomoses.25 Recently, it has been suggested that impaired intestinal epithelial wound healing with corticosteroid therapy may be caused by the inhibition of intestinal epithelial cell restitution and proliferation, as indicated by in vitro studies.26
Table 2. Selected nonpeptide factors with relevance for intestinal epithelial wound healing
Decrease bursting strength of intestinal anastomoses
Cellular components of innate and specific immunity
The mucosal immune system is composed of a considerable number of different cell populations, whose timing of action and class of activity can be promiscuous or extremely selective (Figure 4). At present, the normal immune response is viewed as being formed by two main components: innate immunity and specific immunity.27–29 Innate immunity is activated when invading antigens first make contact with the host. At this time, epithelial barriers physically prevent their entry, phagocytic cells, such as neutrophils, monocytes and macrophages, engulf and destroy the antigens, and natural killer cells spontaneously kill cells bearing foreign antigens. These events occur quickly and the innate immune response acts within hours.
Recently, defensins have been identified as playing an important role in mucosal innate immunity.30–32 Defensins can be classified as α or β. An increasing number of different peptides with varying antimicrobial properties have been identified. They are distributed widely in humans, and organ-specific expression patterns have been observed. Homologous peptides have been found in other mammals, vertebrates, invertebrates, insects and plants. The identification of α-defensins and their murine counterparts, cryptdins, in the small intestine prompted intensive research into epithelial antimicrobial defence. Defensins comprise a class of cationic antimicrobial peptides with a molecular weight of 3–5 kDa. They can be divided into constitutive forms, e.g. HBD-1 with its widespread stable distribution, and inducible peptides, such as HBD-2.31,32 The mechanisms of activation are currently under investigation. A cytokine-driven induction, for example by interleukin-1β and tumour necrosis factor-α, has been demonstrated, in addition to a direct response to bacterial components such as lipopolysaccharides and lipoproteins. Possible signalling pathways involve Toll-like receptors, especially TLR2 and TLR4, eventually leading to nuclear factor κB-mediated activation of transcription.28,33,34NOD2/CARD15 as an intracellular lipopolysaccharide receptor induces nuclear factor κB which, in turn, triggers HBD-2 transcription.
If the antigens are not eliminated, a second phase of the immune response takes over, that is specific immunity, also called adaptive or acquired immunity.35,36 This phase is long-lasting, developing during several days, and involves the participation of B and T lymphocytes, which are responsible for humoral (antibody-mediated) and cell-mediated immunity, respectively (Figure 4). In an early recognition phase, antigen-presenting cells (classically macrophages and dendritic cells) present processed antigens to immunologically naive lymphocytes. These are presented in the context of major histocompatibility complex class II antigens on the antigen-presenting cell surface, with the help of costimulatory (CD28-CTLA-4/B7.1–2) and adhesion molecule (intercellular adhesion molecule-1/leukocyte functional antigens (LFA)-1) pairs. On antigen priming, the lymphocytes are activated and mature into antibody-producing plasma cells or effector cytolytic or helper T-cells, all of which carry out the elimination of the antigen. After this occurs, most of the effector cells are no longer needed and they die through a process of programmed cell death (apoptosis) to terminate the immune response. However, a tiny subset of long-lasting memory cells is left behind, which can be reactivated in the future if the immune system re-encounters the same antigen.
In addition to differences in mediator cells and timing of action, the innate and specific branches of immunity display several other distinguishing features (Table 3). In innate immunity, the specificity of the mediator cells is restricted to nonself, whereas specific immunity permits the recognition and handling of a much wider range of antigens. Thus, receptor diversity is limited in innate immunity, but is practically endless for lymphocytes, being estimated to range between 1014 and 1018.
Table 3. Differences between innate and specific immunity
Inflammatory bowel diseases and abnormalities of the intestinal immune system
Although the pathogenesis of inflammatory bowel disease is still unclear and no single agent or mechanism can explain all aspects of Crohn's disease and ulcerative colitis, it is clear that distinct immune abnormalities play a major role in the initiation and perpetuation of both diseases. Amongst the numerous immune cell populations within the intestinal mucosa, intestinal T cells are assumed to play a key role in the pathogenesis of inflammatory bowel disease.35 The function of T cells is dependent on their activation through various receptors, but is limited on termination of their life by apoptosis, which can be initiated by various factors. T-cell activation leads to cell proliferation, a process necessary not only to expand the respective immune cell population, but also for cytokine production and cell differentiation.
A deleterious series of events appears to occur in Crohn's disease, and an intrinsic hyperreactivity of mucosal T cells to growth signals has been described.35 In the majority of cases, cell proliferation is increased, but cell death is also decreased in Crohn's disease lamina propria T cells. In Crohn's disease patients, lamina propria T cells cycle faster through the cell cycle and replicate their DNA in a significantly shorter time than in controls or ulcerative colitis patients and, consequently, expand to a greater extent, indicating a diverse pattern of cell cycle progression in Crohn's disease and ulcerative colitis.37,38 In Crohn's disease, p53 and p27 protein levels are substantially lower than in equally stimulated control and ulcerative colitis lamina propria T cells, indicating less restricted cell cycling in Crohn's disease. Additional data suggest that increased telomerase activity may result in an increased expansion of Crohn's disease lamina propria T cells. In ulcerative colitis, telomerase activity was not detected, explaining the limited capacity to expand, and further suggesting defective T-cell cycling in ulcerative colitis.
As the proliferation of lamina propria T cells is increased in Crohn's disease, appropriate control and up-regulation of apoptosis are crucial to preserve immune homeostasis; otherwise, the immune response will be dysregulated. This may result in an inappropriate accumulation of T cells which may cause intestinal inflammation. A decreased susceptibility to apoptotic signals and decreased cell death are characteristic features of Crohn's disease.38 Proliferation and apoptosis are linked to the cell cycle. However, cell cycle progression does not seem to be essential for activation-induced apoptosis in inflammatory bowel disease, where the increased cell cycle progression in Crohn's disease is combined with a reduced apoptosis rate. In contrast, in ulcerative colitis, delayed cell cycle progression is accompanied by increased apoptosis. This unique pattern indicates that the balance between cell cycle progression and cell death is lost in inflammatory bowel disease, showing an unrestrained cell cycle progression in Crohn's disease, but a defective clonal expansion in ulcerative colitis. It seems reasonable to assume that the elimination of excessive T cells could restore the gut to a normal state of controlled inflammation. Strong evidence for this effect has been provided by animal models, where experimental colitis was abrogated by the induction of increased T-cell apoptosis with interleukin-12 antibodies, blockade of interleukin-6 trans-signalling or the deletion of CD44v7+ cells.39–41 Recent data have provided evidence that infliximab, a tumour necrosis factor-α blocker used for the treatment of severe Crohn's disease, may act at least in part by the induction of apoptosis in lamina propria T cells and monocytes.42,43 These findings may explain the rapid and sustained therapeutic effects of infliximab in Crohn's disease.