Regulatory T Cells as Potential Targets for Immunotherapy in Inflammatory Bowel Disease

KEYWORDS::

• Crohn‘s disease
• IL-17
• inflammation
• inflammatory bowel disease
• intestinal mucosa
• regulatory T cell
• TGF-β

The gut mucosal immune system is the largest lymphoid organ in the body. This site undergoes continuous antigenic challenges from food antigens, antigens of the abundant normal bacterial flora and pathogens. Despite this constant antigenic stimulation, controlled inflammatory responses and suppression of inflammation appear to be the rule. Apparently, the gut immune system effectively differentiates the potentially harmful antigenic signals from the high background noise of food and bacterial antigens. This tight regulation required to maintain homeostasis is achieved through multiple nonimmune and immune factors [1].

Regulatory pathways mediated by regulatory T cells (Tregs) are essential homeostatic mechanisms. Distinct Treg subsets coexist in the intestinal mucosa and mesenteric lymph nodes. These include the ‘natural‘ and ‘adaptive‘ CD4⁺ forkhead box p3 (FoxP3⁺) Tregs, as well as Tr1 and Th3 cells. Tregs that develop in the thymus are referred to as ‘natural‘ Tregs (nTregs). In addition to their natural counterparts, Tregs can develop in response to stimuli in the periphery and these have been termed ‘adaptive‘ Tregs. A subset of adaptive Tregs express FoxP3, making them difficult to distinguish from their natural counterparts. In murine CD4⁺ T cells, FoxP3 can be induced in response to activation of the T-cell receptor in the presence of TGF-β and confers suppressor function. TGF-β is also necessary for the induction of Th17 effector cells, but this requires IL-6 [2]. Interestingly, inducible Tregs (iTregs), but not nTregs, are resistant to Th17 conversion by IL-6 [3]. The intestinal environment appears to be an important site for the generation of Tregs, as dendritic cells (DCs) from the mesenteric lymph nodes and small intestine are particularly efficient for the peripheral conversion of Tregs via TGF-β- and retinoic acid-dependent mechanisms. Tregs that do not express FoxP3 but are functionally suppressive both in vitro and in vivo include Tr1 cells, which are defined by the production of IL-10 and can be induced by oral administration of antigen, in vitro culture with IL-10 or exposure to immature DCs. Another subset of Tregs, Th3 cells, suppresses immune responses in a TGF-β-dependent fashion. Recent data have suggested that Th3 cells may function by inducing de novo FoxP3⁺ Tregs via TGF-β [4]. Indeed, we have reported that inhibition of TGF-β1 interferes with the induction of mucosal Tregs [5]. Disturbances in Treg number and function are associated with immune-mediated disorders.

The inflammatory bowel diseases (IBDs) include ulcerative colitis and Crohn‘s disease, and are chronic diseases characterized by infiltration of inflammatory cells into the lamina propria of the intestinal tract. Data from patients and animal models suggest that IBD results from an aberrant immune response to the intestinal microflora [6]. In support of this hypothesis, genetic susceptibility to IBD in humans is associated with variations in genes that encode proteins relevant to both innate and adaptive immunity, including NOD2, ATG16L1, IL-12B, STAT3 and the IL-23 receptor (IL-23R) [7]. In addition, data from animal models demonstrate that intact intestinal flora are necessary for the development of IBD, and antibiotics and probiotics have been demonstrated to be beneficial under certain circumstances in the treatment of patients with IBD [8].

As Tregs are thought to play a critical role in limiting inflammation in response to these nonpathogenic antigens, defects in this T-cell subset have been implicated in the pathogenesis of IBD. One of the first murine models to convincingly and reproducibly demonstrate the existence of Tregs was the T-cell transfer model of colitis. In this model, the transfer of naive T cells expressing high levels of CD45RB (CD45RB^hi) induces Th1-mediated colitis when transferred into an immunodeficient host. Concomitant transfer of T cells with low expression of CD45RB (CD45RB^lo) prevents the development of colitis. It has now been demonstrated that the regulatory capacity of CD45RB^lo T cells is present in the CD4⁺CD25⁺FoxP3⁺ fraction and that transfer of these Tregs not only prevents colitis initiation, but can also reverse established disease. In addition to the T-cell transfer model of colitis, multiple murine models have demonstrated a crucial role for Tregs in the control of intestinal inflammation, and elucidate the mechanisms of their generation, migration, interactions and function [4].

Characterization of the role of Tregs in IBD has been hampered by the fact that potentially causal antigens have not been identified. Nonetheless, it has been reported that peripheral blood of IBD patients contains increased numbers of Th17 cells combined with decreased Treg numbers [9]. In addition, expression of both FoxP3 and IL-17a is increased in IBD intestinal mucosa. This may indicate that Tregs are sequestered within the gut mucosa, where a proinflammatory cytokine environment high in IL-1β and IL-6 restricts Tregs activity and promotes the continual differentiation and development of Th17 cells. Hence, therapeutic approaches that aim to re-establish homeostasis by increasing the number of Tregs may prove effective in the treatment of IBD [9].

Protocols to expand Tregs ex vivo have been published that rely on repetitive stimulation via the T-cell receptor combined with cytokine exposure [10]. In addition, gene transfer of FoxP3 into naive murine T cells has been successful in producing a large homogeneous population of functional Tregs [11]. There has also been excitement over protocols that generate Tregs from peripheral CD4⁺ T cells in the presence of TGF-β. Scurfy mice treated neonatally with such induced murine Tregs are protected from autoimmune disease [12].

Increasing the lifespan of Tregs may reduce the number required for transfer, and circumvent some of the difficulties in expanding or generating Tregs ex vivo. Transduction of murine Tregs with a stabilized β-catenin results in prolonged survival via upregulation of Bcl-xL and significantly reduces the number of transferred Tregs required to protect against colitis induced by CD4⁺CD45RB^hi cells [13].

The identification of the important roles that intestinal bacteria play in normal development and regulation of the mammalian immune system, and particularly in the induction of Tregs, provides a rationale for using therapeutic agents based on bacterial-derived signals in preventative or therapeutic endeavors. Live biologics given as supplements to confer some benefit to the host are referred to as probiotics. Several studies in animal models have investigated the beneficial effects of probiotic bacteria, including Lactobacillus spp. and others. Probiotic bacteria, including lactobacilli, induce IL-10 production by DCs and inhibit the subsequent generation of Th1 cells in vitro, and lactobacillus and bifidobacterium strains prevent experimental colitis while expanding IEL-γδ T cells and Tregs. A significant number of experimental studies have shown that some nonpathogenic microorganisms, including lactobacilli, bifidobacteria and helminths, are particularly effective for the induction of regulatory cytokines and pathways [14]. The effect of probiotics on the induction of Tregs probably relates to the interactions between probiotics and DCs that favor Treg development. Induction of human monocyte-derived DCs in the presence of Lactobactillus rhamnosus leads to cells that have a reduced capacity to support T-cell production of cytokines [15]. It seems that the phagocytotic uptake of specific probiotic bacteria by bone marrow-derived DCs influences the ability of these cells to produce cytokines. These studies suggest that the DCs exposed to probiotics become regulatory cell-inducing antigen-presenting cells. The fact that some probiotic strains have these stimulatory effects on DCs and others do not suggests that a unique interaction between DCs and T cells in the gut is necessary for the probiotic induction of regulatory cells [16].

A number of pharmacologic agents with immunosuppressive properties have also been demonstrated to promote Treg number or function and may be of benefit for the treatment of IBD. One of the best characterized is anti-CD3 monoclonal antibody, which has been used extensively in humans and mice for the treatment of autoimmune diabetes and has recently shown efficacy for the treatment of ulcerative colitis. Anti-CD3 therapy has been demonstrated to promote the development of adaptive TGF-β-producing Tregs in a murine model of diabetes. In addition, data from an IL-10 reporter mouse demonstrated that anti-CD3 monoclonal antibodies lead to the accumulation of IL-10-expressing FoxP3^- (Tr1) cells in the small intestine and IL-10-expressing FoxP3⁺ T cells in the colonic lamina propria. It has recently been suggested that following anti-CD3 therapy, ingestion of apoptotic T cells by macrophages and immature DCs results in production of TGF-β, potentially through IFN-γ-dependent induction of indoleamine 2,3-dioxygenase, and this may be responsible for the observed de novo generation of Tregs following therapy [17]. This local noninflammatory milieu influences the phenotypic and functional characteristics of antigen-presenting cells. DCs that present specific autoantigens express fewer costimulatory molecules but deliver more inhibitory signals involving molecules such as programmed cell death ligand 1 or ICOS ligand. These DCs can inhibit autoantigen-specific pathogenic T cells and promote the expansion of adaptive CD4⁺FoxP3⁺ Tregs in the presence of TGF-β. In addition, in the presence of such tolerogenic DCs and TGF-β, naive autoreactive CD4⁺CD25^- T cells can be converted into adaptive Tregs. Together, these events contribute to the immunoregulatory process that perpetuates the long-term maintenance of self-tolerance [18].

However, in the presence of TGF-β plus IL-6 or IL-21, the Treg developmental pathway is abrogated, and instead, T cells develop into Th17 cells. There is a reciprocal relationship between Tregs, and Th17 cells and IL-6 plays a pivotal role in dictating whether the immune response is dominated by proinflammatory Th17 cells or protective Tregs. It has been proposed that the unactivated immune system at a steady state (in the absence of inflammatory stimuli) produces TGF-β, which induces the generation of iTregs, which together with nTregs keep activated/effector memory cells in check. IL-6, induced during an acute-phase response, inhibits the function of nTregs and prevents the generation of iTregs but instead induces Th17 cells. Thus, IL-6 plays a pivotal role in dictating the balance between the generation of Tregs and Th17 cells. We conclude that IL-6 acts directly to promote the development of Th17 by activating the T-cell gp130–STAT3 pathway, but has a minimal effect on Treg development, at least in the steady state in vivo[19]. Therefore, blockade of the IL-6 pathway in CD4⁺ T cells could be a good target for controlling unwanted Th17-mediated immune responses, including autoimmune diseases. Collective studies have emphasized a central role for IL-6 in governing inflammation and highlight the therapeutic potential of targeting IL-6 as a strategy for the treatment of chronic inflammatory diseases. Indeed, MRA, a humanized anti-IL-6R monoclonal antibody (also known as atalizumab) that blocks both soluble and membrane-bound IL-6R, is also highly effective in the management of Crohn‘s disease and appears to be well tolerated, with no adverse reports of infection or toxicity [20].

Another class of pharmacologic agents, the histone/protein deacetylase inhibitors (HDACs), has recently been demonstrated to improve Treg function and increase Treg numbers in mice. HDACs regulate chromatin remodeling and affect gene expression by modifying histones via acetylation. HDAC9 is preferentially expressed in FoxP3⁺ Tregs, and HDAC9-deficient mice have increased numbers of Tregs and are less susceptible to dextran sulfate sodium-induced colitis. HDAC inhibitor therapy increases the number of Tregs in wild-type mice and diminishes dextran sulfate sodium colitis in a Treg-dependent manner. These results suggest that HDACs may be an important target to alter Treg function, although more specific inhibitors may be necessary for safe use in humans [11].

Finally, gene therapy approaches may induce an anti-inflammatory environment through the induction of Tregs. These include attempts to directly transduce the gut mucosa in order to increase the production of regulatory cytokines, as well as the administration of ex vivo-engineered cells. We and others have reported that colonic inflammation in experimental trinitrobenzene sulfonic acid colitis in rats, mice or IL-10-deficient mice can be prevented by systemic administration of recombinant adenovirus 5 encoding IL-10 [21–23]. Adoptive transfer of IL-10-transduced T cells has recently been successfully used to treat colitis in the murine CD4⁺CD45RB^high severe combined immunodeficiency transfer model. Since the disappointing therapeutic effect of systemic injection of recombinant IL-10 in patients with active Crohn‘s disease has been attributed to a lack of mucosal bioavailability, it is interesting to speculate whether the local delivery of immunoregulatory T cells reliably producing IL-10 would result in a better outcome [24].

In summary, the ability to alter regulatory pathways may be a critical avenue for achieving long-term remission in patients, and several results from animal models suggest that Treg expansion and Treg transfer in IBD patients may be beneficial [25].

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.