At least three pieces of evidence exist indicating a major role for innate immune cells in the perpetuation of human chronic intestinal inflammation. First, the strongest genetic association to date with small bowel Crohn disease is the loss-of-function polymorphisms in the bacterial sensing gene CARD15/NOD2. Whether the ultimate pathogenic dysfunction rests in Paneth cells [37] or monocytes [38], it appears certain that dysregulation in the detection of and/or responsiveness to enteric bacteria by innate immune cells promotes chronic inflammation in this subgroup of patients. The second piece of evidence rests in the efficacy of immune therapy targeted against pro-inflammatory cytokines produced by innate immune cells. Activated macrophages produce large amounts of TNF-α and IL-12, two cytokines responsible for the recruitment and activation of pathogenic effector T cells. The efficacy of anti-TNF-α therapy is established [39], and promising anti-IL-12 studies are currently ongoing [40]. The third piece of evidence is the presence of a defective acute inflammatory response to injury and bacterial products in Crohn disease patients, demonstrated by reduced neutrophilic infiltration, IL-8 production and vascular flow [41], and corroborated by the clinical improvement provided by the administration of GM-CSF, supposedly by improving macrophage and neutrophilic function [42]. This failure of clearance of bacteria and inflammatory debris indicates a disruption in the autophagy pathway. Autophagy, (process of cell self-digestion), has received significant investigative interest recently as many of the 90 Crohn’s associated genetic loci fall generally within the category of autophagy homeostatic function (i.e., ATG16L1 and IRGM) [43]. This avenue of research holds great promise in the near future.

Within the lamina propria of the intestine CD4+ and CD8+ T cells bearing the conventional αβ TCR are roughly equally represented. The intraepithelial space contains unusual and enigmatic T cell populations bearing the γδ TCR and CD8+ cells with αα homodimeric expression of TCR (discussed briefly earlier). As current understanding of both human and experimental IBD emphasizes the primary importance of activated CD4+ TCR αβ T cells to disease pathogenesis, the discussion later will focus on activation of CD4+ T cells and the homing pattern that ensues. Upon maturation in the thymus, naïve T cells (cells which have yet to experience antigen exposure) circulate the lymphatic tissues in search of its cognate MHC–peptide complex. Constitutive expression of the selectin CD62L (l-selectin) and the chemokine receptor 7 (CCR7) ensures that naïve T cells bind to glycosylated CD34 and the chemokine ligand 21 expressed on endothelial cells of high endothelial venules [44]. The interaction of CD62L and CD34 (among other glycosylated endothelial molecules) promotes a rolling action of the T cell across the endothelial surface. The T cell surface adhesion molecule LFA-1 binds to ICAM-1 and ICAM-2 promoting firm adhesion and crossing of the endothelial lining into the lymphoid tissue. Within the underlying lymphatics await DCs loaded with antigen in variable states of activation. Naïve T cells search the DC selection for a recognizable MHC–peptide complex. If none is found the cells exit the lymph node, return to the circulation, and reenter other nodes to repeat this process.

Eventually, a naïve T cell finds its specific antigenic mate and receives the appropriate activation signals upon ligation of its TCR (Fig. 2.2). A full T cell activation requires two signals: signal 1 is delivered by TCR stimulation and signal 2 is a co-stimulatory signal provided by secondary accessory pathways. Among the latter, co-stimulation through CD28 by B7.1 or B7.2 results in T cell activation; in contrast, co-stimulation through CTLA-4 by B7.1 or B7.2 results in T cell inhibition. In addition, TCR signaling in the absence of co-stimulation results in anergy, defined as a lack of response upon re-exposure to the same antigen in the future. A key consequence resulting from activation of CD4+ T cells is cytokine secretion, and the particular pattern of cytokines secreted orchestrates the type of the ensuing immune response (i.e., directed against intracellular pathogens, parasites, etc.). The phenotype of the mature effector T cell depends upon in large part the cytokine milieu available to the T cell while undergoing the activation and differentiation program. While insight into these complex events is in constant evolution, current understanding is as follows (Fig. 2.3). Activation of CD4+ T cells in the presence of IL-12 and absence of IL-4 results in a T helper 1 (Th1) phenotype, resulting in IFNγ-producing cells effective in the control of intracellular pathogens. Cells activated in the presence of IL-4 acquire a T helper 2 (Th2) phenotype, generating IL4-, IL5-, and IL13-producing cells effective in allergic responses and clearance of parasitic infections. Finally, activation in the presence of IL6, TGFβ and IL-23 results in cells expressing the recently described Th17 phenotype, i.e., IL17 and IL6-producing T cells responsible for acute inflammation and recruitment of granulocytes [45, 46]. Full activation of T cells takes 4–5 days and is accompanied by clonal expansion and remarkable changes in the homing behavior of these cells.

Once activated, these effector T cells leave the lymph node returning to the circulation, thus allowing a locally generated immune response to promote defense at distance throughout the gastrointestinal tract [47]. Activated T cells express unique adhesion molecules that direct the cells to sites of inflammation, and indeed T cells primed by mucosal DCs are destined to return to the gut through the upregulation of two surface molecules, α4β7 integrin and the chemokine receptor CCR9 [22]. The integrin α4 binds to the vascular adhesion molecule VCAM-1, a molecule expressed on vascular endothelium only at sites of inflammation. The α4β7 integrin recognizes the more ubiquitous vascular endothelial molecule MAdCAM-1, directing migration to the intestinal lamina propria [48, 49], while αEβ7 is uniquely expressed on T cells destined to the intestinal intraepithelial space. MAdCAM-1 has been shown to be upregulated during exacerbations of IBD, and thus represents a promising target for therapeutic intervention [49].

Effector T cells appear to enter nearly all tissues in limited numbers. Recognition of cognate antigen within the tissue results in T cell cytokine production. Pro-inflammatory cytokines, such as TNF-α, stimulate endothelial cells to upregulate adhesion molecules such as E-selectin (that recruits monocytes and neutrophils), and VCAM-1 and ICAM-1 (both of which recruit activated T cells). TNF-α and IFN-γ likewise act to alter the vascular permeability, endothelial cell shape, and blood flow, resulting in enhanced infiltration of inflammatory cells into the tissue. These inflammatory cascades set in motion by activated T cells and the significant structural alterations they cause in the inflamed tissue eventually trigger the signs and symptoms characteristic of active IBD.

Thus, naïve T cells circulate nearly all tissues due to the expression of CD62L, CCR7, and LFA-1. Activation of T cells in the GALT leads to expression of α4β7 and CCR9 expression, two molecules that produce selective homing to the intestine through endothelial expression of MAdCAM-1 and CCL-25. Mucosal inflammation results in upregulation of MAdCAM-1, in addition to other endothelial ligands, and enhanced recruitment of pro-inflammatory T cells to the inflamed organ.

It is now well established that subpopulations of CD4+ T cells with regulatory potential exist in both mice and humans. Current models describe three major and not necessarily mutually exclusive subtypes of Tregs: the “naturally occurring” CD4+ CD25+ cell, the IL-10- and TGF-β-producing Tr1 cell, and the TGF-β-producing Th3 cell derived through oral tolerance [50]. While accurate description of this rapidly evolving field is a moving target, the more studied and better understood Treg cell type is the CD4+ CD25+ T cell expressing the transcription factor FOXP3. The importance of this cell type to murine colitis and its capacity to suppress mucosal inflammation are well established [51, 52]. However, although this type of Treg has also been described in IBD [53], the function of these cells in the periphery and the gut of patients with IBD has yet to contribute to a better understanding of IBD pathogenesis.

B cells of the GALT work in close collaboration with the epithelium to export secretory IgA (sIgA) and to some extent secretory IgM (sIgM) to enhance mucosal defense from intestinal pathogens. Mucosal B cells exhibit a predominant IgA class switch and, upon differentiation to mature plasma cells, produce approximately 3 g of sIgA per day [54]. Over 80 % of all human plasma cells are found in the gut, and nearly all these plasma cells produce IgA [55]. Mucosal plasma cells produce primarily dimeric or polymeric forms of IgA. The joining or “J” chain of polymeric IgA spontaneously interacts with the polymeric Ig receptor expressed on the basolateral surface of epithelial cells facilitating exportation of the sIgA to the gastrointestinal lumen [56]. Once in the lumen, sIgA coats commensal and potentially pathogenic bacteria. This coating may promote M cell-mediated bacterial uptake and target presentation to intestinal DCs and macrophages, providing a barrier to bacterial penetration as well as a positive feedback loop to enhance secretory immunity [57].

Lymphocytes begin to populate the lamina propria by 12 weeks of gestation (after the thymus matures) and are primarily CD4+ cells [58]. Organized lymphoid tissue of the small bowel (Peyer’s patches) consisting of B cell follicles, and interfollicular CD4+ and CD8+ T cells are detected by 16–18 weeks of gestation [58]. Dendritic cells are detected at 19 weeks, representing the last of the required cellular components of an adaptive immune response [59]. Neither the T cells nor the B cells of the Peyer’s patch exist in an activated state within the fetal intestine. Indeed, the number of Peyer’s patches increases from 80 to 120 at birth to 250 by adolescence [60]. Without antigenic stimulation, there are no IgA secreting cells; yet, within 2 weeks of birth IgA and IgM secreting plasma cells are present within the lamina propria [61].