The proinflammatory capacity of effector cells is kept in balance by regulatory T cells. These cells are classically hypoproliferative, hyporesponsive, and potent in their suppressive capacity to inhibit proliferation and activation of non-regulatory T cell counterparts. Regulatory T cells are classically divided into at least two distinct populations (summarized in Fig. 8.1): naturally occurring regulatory T cells (nTregs) that are generated in the thymus and “induced” regulatory T cells (iTregs) that are generated in peripheral tissues or are generated in vitro in the presence of certain anti-inflammatory cytokines. Both populations are likely essential, with distinct and nonredundant functions in maintaining mucosal homeostasis [14–16].

In addition to CD4⁺ T cells , other T lymphocyte subsets have potential effector or regulatory properties. In the lamina propria, CD8⁺ cells possess regulatory function and have been noted to be decreased in IBD tissues compared to controls [62, 63]. Likewise, in the intraepithelial compartment, TCRαα⁺ CD8⁺ T cells have been shown to be protective in animal models of colitis using adoptive cell transfer [64, 65] whereas TCRγδ⁺ cells appear to limit the extent of injury and promote tissue healing in a chemically induced model of colitis [66]. However, others have demonstrated a potential pathogenic role of TCRγδ⁺ cells given absence of these cells in a transgenic colitis model leads to attenuated disease [67]. These findings may have implications for the pathogenesis of ulcerative colitis given higher frequency of these TCRγδ⁺ cells has been noted in colonic tissue from UC patients [67]. Another less common subset of T cell is NK-T cells , which express surface markers and possess certain functional properties typical of both NK cells and T cells. In addition to expressing components of the T cell receptor complex, CD3, they express semi-invariant CD1d-restricted αβ TCRs recognizing glycolipid antigens. Some regard these cells as potentially pathogenic in inflammatory bowel disease. In fact, Fuss et al. demonstrated circumstantial evidence for NK-T cells being the source of high IL-13 levels noted in samples from patients with active UC, suggestive of a pathogenic role of these cells [4]. Likewise, in the oxazolone-induced murine model of ulcerative colitis, depletion of NK-T cells led to protection against disease [68]. However, in a chemically induced model of colitis (with dextran sodium sulfate), NK-T cells have been demonstrated to possess a regulatory function [69]. A recent murine study showed that microbial exposure during early life leads to the generation of mucosal invariant (i)-NKT cell tolerance. In absence of such tolerogenic commensal exposure in germ-free mice leads to CXCL16-mediated iNKT cell driven intestinal inflammation in oxazolone-induced colitis model [70]. Although these less conventional T cell subsets typically have reduced frequencies when compared with the more conventional CD4⁺ T cells, their pathogenic/regulatory properties may contribute, at least in part, to overall mucosal homeostasis [70].

T cells have been long recognized to be central to the mucosal inflammatory process and thought early on to even be the instigator of disease given the marked expansion of activated CD4⁺ T cells in the lamina propria of both inflamed mouse and human intestinal tissues [1, 2, 10]. CD4⁺ T cells from intestinal samples isolated from Crohn’s disease and ulcerative colitis patients were found to be resistant to apoptosis [71], which may be one mechanism for such seemingly inadequately controlled expansion of T cells. The ability to transfer colitis by adoptively transferring CD4⁺ T cells from a colitic mouse to an immunodeficient host in many animal models also implicated the colitogenic potential of CD4⁺ T cells [6–8]. Even transfer of CD4⁺ T cells depleted of CD25⁺ regulatory T cells from normal animals could lead to disease in a lymphocyte-deficient host [27], demonstrating the pro-inflammatory nature of effector T cells and its usual tight regulation by regulatory T cells during homeostasis.

Similar to that seen in a variety of murine models of IBD, CD4⁺ T cells isolated from patients with Crohn’s disease or ulcerative colitis express large quantities of pro-inflammatory cytokines [3, 5, 11, 72]. Interferon-(γ)gamma , tumor necrosis factor-(α)alpha and IL-2 have been demonstrated repeatedly to be elevated in Crohn’s disease mucosa, whereas IL-5 and IL-13 have been associated with ulcerative colitis. Early studies demonstrating an abundance of Th1 type cytokines in Crohn’s disease suggested IFN-γ⁺ Th1 cells to be a major mediator of disease, while ulcerative colitis was suggested in some studies to be mediated primarily by excessive Th2 responses. Interestingly, despite an elevated level of IL-5 and IL-13 associated with ulcerative colitis, there is evidence of surprisingly reduced expression of IL4⁺ Th2 cells [3, 4, 10]. More recent reports demonstrating increased population of IL-5 and IL-13 expressing atypical natural killer T (NKT) cells in ulcerative colitis provided one explanation for this apparent anomaly [4, 10]. More recent studies have shown that tissues from both Crohn’s disease and ulcerative colitis patients have a massive infiltration of Th17 cells and Th17-related cytokines [10, 73, 74]. Apart from activating cellular targets like epithelium, endothelium, monocytes/macrophages, and neutrophils, IL-17 is known to induce proinflammatory cytokines (TNF-α, IL-1B, GM-CSF, G-CSF, IL-6), chemokines (CXCL8, CXCL9, CXCL10), and metalloproteases [74–76]. Interestingly, several signature cytokines involved in Th17 differentiation and expansion, including IL-23R, IL-12B, JAK2, STAT3, CCR6, and TNFSF15, have been identified as susceptibility genes of Crohn’s disease and ulcerative colitis in genome-wide association studies [77–79].

On the basis of the observations that IFNγ, IL-17 , IL-3 may play a key role in the pathogenesis of Crohn’s disease and ulcerative colitis, both in mouse models and in humans, these cytokines have been considered as potential targets for the treatment of IBD. However the outcome in several studies has not been met with much success. Several groups have studied the effect of humanized antibody against IFN-γ, fontolizumab, in patients with Crohn’s disease; however, the efficacy of anti-IFN-γ treatment remains unclear [80–82]. Blockade of IL-17 pathway was also not effective in Crohn’s disease: secukinumab , a fully humanized anti-IL-17A monoclonal antibody, failed to control symptoms and was associated with severe adverse effects [83]. Finally, two anti-IL-13 monoclonal antibodies , tralokinumab [84] and anrunkinzumab [85], recently failed in phase II clinical trials in ulcerative colitis patients [86]. The failure of these agents, which target adaptive immune responses, has led to some rethinking of the central role of these cytokines in the pathogenesis of both Crohn’s disease and ulcerative colitis.

Recent data has suggested that the initial aberrant signal may derive from defects in the innate immune system (such as an epithelial cell, macrophages, dendritic cells, NK cells, etc.). Polymorphisms in genes (e.g., NOD2, ATG16L1, and IRGM) that code for proteins important in recognizing and handling intracellular microbes, processes that occur in innate immune cells, have been recognized to be associated with Crohn’s disease risk in genome-wide association studies [87]. Data from animal studies have also implicated professional antigen presenting cells or epithelial cells in playing a more direct role in colitis induction [37, 88, 89]. Nonetheless, T cells are still thought to be an important mediator of disease, contributing to either the initiation or perpetuation of disease [1]. In this latter context, once activated by soluble factors and/or by cell–cell contact, T cells are able to expand and secrete chemokines and cytokines that recruit and activate, respectively, other immune cell subsets to perpetuate an inflammatory signal.

In order to damage the gut, effector T cells need to adhere and migrate into intestinal mucosa, a process which depends on the recognition between adhesion receptors (CCR9 and α4β7 ) expressed on the T cell surface and their ligands (MAdCAM-1 ) on the gut vascular endothelial surface [90]. CCR9 and α4β7 are the two most well characterized gut-homing receptors on T cells, and blocking the interaction between these receptors and their ligands on endothelial cells prevents T cells from entering the gut. This serves as the basis for the development of biologic therapies for Crohn’s disease such as natalizumab , which is a monoclonal antibody against the integrin α4, and vedolizumab , a monoclonal antibody against α4β7. Both agents interfere with T cell homing to the gut and consequently decrease inflammatory activities leading to clinical efficacy in both Crohn’s disease and ulcerative colitis [91–94]. Additional anti-adhesion molecule strategies including the anti-β7 antibody etrolizumab [95] and anti-MAdCAM-1 antibody targeting the α4β7 ligand [96] are currently being tested in IBD [97]. Moreover, a CCR9 chemokine receptor small-molecule antagonist targeting the CCR9-CCL25 interaction has recently failed in a large [96–98]. Other agents blocking such signals crucial to T cell gut homing are also under development.

One may also hypothesize that T cell suppression is diminished in IBD either from a qualitative or quantitative defect in Treg function. However, a defect in Treg numbers appears not to be a principal cause of IBD, since their proportions are noted to be increased in inflamed IBD tissue when compared to non-inflamed samples from IBD patients or samples from healthy control patients [99]. In fact, during periods of active disease, fewer Tregs are seen in the peripheral blood and more are found in the mucosa, suggesting an increase (through homing) or retention of Treg to sites of inflammation. Moreover, peripheral blood, mesenteric lymph node, and mucosal Tregs when removed from IBD patients retain in vitro suppressive function [99–101]. Thus, Tregs from IBD patients have the capacity for functional activity. However, the increase in Treg proportions in inflamed mucosa from IBD patients appear to be quantitatively less than that seen in non-IBD inflammatory control samples (those with diverticulitis), suggesting the increase in Treg numbers seen in IBD is not as robust as seen in other inflammatory disorders. Nonetheless, despite the presence of regulatory T cells in IBD associated mucosa these cells are not sufficient in situ to control inflammation.

Given the potent suppressive properties of regulatory T cells, one potential therapeutic option for IBD is quantitative enhancement of Treg numbers and/or functional capacity [21]. One theoretical approach is to isolate Tregs from a patient’s peripheral blood, followed by augmentation of cell function/numbers in vitro, and subsequent infusion of these cells into the patient. However, this idea has been hindered by difficulty in Treg expansion given their tendency for hypoproliferation, poorly defined surface markers, and uncertain stability and longevity in vivo. One potential way of overcoming these difficulties is to generate and expand inducible Tregs in vitro from naïve T cells, such as iTr35 cells , Tr1 cells , or regulatory CD8⁺ cells [47, 52, 63]. Even then, it is difficult to purify some of these populations due to the lack of specific cell surface markers. In addition, this technique will still require peripheral blood cell isolation; immediate processing or cryopreservation; in vitro culture and stimulation; vigorous testing for function, purity, and viability; and reinfusion of newly induced Tregs, posing technical challenges to avoid infection and maintain quality control. Recently Canavan and colleague have shown that Tregs can be expanded from the blood of patients with Crohn’s disease. They demonstrated that the expanded cells were epigenetically stable and home to human small bowel in a C.B-17 severe combined immune deficiency (SCID) xenotransplant model [102]. In a phase 1/2a study 40 % of the patients showed a clinical response after a single injection of ovalbumin-specific Treg cells in 20 Crohn’s disease patients [97, 103]. In this study autologous Treg cells were expanded in vitro, cloned by limiting dilution, and selected for IL-10 production in response to ovalbumin before transfer [15, 103]. An alternative option is to pharmacologically expand the already existing pool of Tregs or induction of an iTreg population. The latter could be done with an agent such as oral anti-CD3 antibody, which has been shown to expand a particular type of Tregs and has been shown to suppress EAE, SLE, and diabetes in mice [104, 105]. In another study Koreth et al. showed that daily administration of low-dose interleukin-2 in patients with active chronic GVHD that was refractory to glucocorticoid therapy was associated with preferential sustained Treg cell expansion in vivo and amelioration of the manifestations of chronic GVHD in a substantial proportion of patients [106]. Phase 1 studies investigating the role of oral OKT3 (anti-CD3) and low-dose IL2 have been initiated for ulcerative colitis (ClinicalTrials.gov: NCT02200445; ClinicalTrials.gov: NCT01287195). Taken together therapies targeting the immunomodulatory potential of Treg cells seem to be a promising approach for treatments of IBD.

T cells have been clearly implicated as a mediator and propagator of disease in many animal models of IBD. In patient samples, T cells secrete an abundance of pro-inflammatory cytokines suggesting a central role of T cells in the pathogenesis of disease. Despite the large body of knowledge regarding the role of T cells in mucosal homeostasis, much still remains to be understood with regard to how these T cell subsets interact with each other and with other immune cell subsets to lead to IBD and how these interactions can be repaired to restore intestinal homeostasis.