Glycans for good

Abstract

The gut commensal bacterium Bifidobacterium bifidum promotes immune tolerance by facilitating the induction of colonic regulatory T cells.

Colonization of mucosal barriers by commensal microbes is important for the development and conditioning of the host immune system. In the mammalian gastrointestinal tract, resident bacterial communities can also support host metabolism through the digestion of complex carbohydrates, synthesis of essential vitamins and nutrients, and the biotransformation of bile acids. More recently, it has been appreciated that some bacterial components and metabolites can strongly influence the inflammatory potential of immune cells and profoundly affect in vivo phenotypes (1). Although considerable efforts have been made to isolate particular mucosa-associated microbes that contribute to host protection or disease susceptibility, the effects of individual bacterial species remain largely unknown (2). In this issue of Science Immunology, Verma et al. explore the mechanism by which the gut commensal bacterium Bifidobacterium bifidum induces peripheral regulatory T (T_reg) cells within the mouse gastrointestinal tract (Fig. 1) (3). Using an ex vivo screening approach to test for the ability of individual probiotic strains to promote the differentiation of T_reg cells from naïve CD4⁺ T cell precursors, the authors identify B. bifidum as a highly immunomodulatory microbe that potently induces T_reg cells both in vitro and in vivo. In subsequent experiments that used a number of germ-free and gnotobiotic animal models, the authors report the molecular cross-talk between B. bifidum and host immune cells that dampens inflammation and prevents colitis.

Fig. 1. Induction of intestinal T_reg cells by colonization with B. bifidum.

Neutral CSGG of B. bifidum are recognized by TLR2 on CD103⁺ intestinal DCs. TLR2 promotes an rDC phenotype, marked by up-regulation of tolerogenic cytokines. Among these, TGF-β production by DCs induces Foxp3⁺RORγt⁺ T cells with broad antigenic specificity.

Over the past 15 years, many studies have demonstrated that the administration of specific strains of probiotic bacteria can promote the induction of intestinal T_reg cells and suppress inflammation (4, 5). One of the earliest publications in this area described a role for Bacteroides fragilis polysaccharide A (PSA) in promoting the differentiation of mouse intestinal T_reg cells (5). Verma et al. build on this growing body of literature to show that molecules isolated from B. bifidum can also influence mucosal immunity. Interestingly, in comparison with T_reg cells generated during long-term monocolonization with B. fragilis, the authors observed a robust expansion of colonic T_reg cells within just 3 weeks after monocolonization of germ-free mice with B. bifidum. In addition to expressing high levels of the immunosuppressive cytokine interleukin-10 (IL-10), a substantial proportion of these newly expanded Foxp3⁺ T_reg cells express RAR-related orphan receptor-γ (RORγt), which is indicative of a subset of T_reg cells whose development is dependent on the detection of discrete microbial signals. T_reg cells induced in the context of B. bifidum colonization also exhibit low expression of Helios and NRP1, two markers believed to be specifically expressed by thymically derived, as opposed to peripherally induced, T_reg cells. Together, these data support the notion that the presence of B. bifidum within the intestine regulates peripheral T_reg cell development.

Higher frequencies of colonic T_reg cells may arise from their increased differentiation from naïve CD4⁺ T cells or by the expansion of preexisting intestinal T_reg cell populations. To tease apart these possibilities, the authors used an OT-II T cell receptor (TCR)–transgenic T cell transfer model in which T cells that recognize the food antigen ovalbumin (OVA) contribute to oral tolerance. Delivery of naïve, congenic, OT-II T cells into recipient mice who were fed OVA and monocolonized with B. bifidum resulted in increased numbers of differentiated T_reg cells, as compared with that of mice colonized with a control bacterium, Lactobacillus paracasei. This biased differentiation of naïve OT-II T cells into T_reg cells indicated that the presence of B. bifidum within the colonic microbiota can support de novo induction of peripheral T_reg cells.

T_reg cells generated in the context of B. bifidum colonization have a highly diverse TCR repertoire and recognize a wide range of antigenic targets besides B. bifidum itself. The broad tolerogenic capacity of B. bifidum-induced T_reg cells was further demonstrated by using a CBir TCR transgenic model of bacteria-induced colitis, in which transgenic T cells are activated in response to bacterial flagellin. Delivery of naïve CBir transgenic T cells into Rag1^−/− recipients that lack adaptive T and B cells resulted in unrestricted inflammatory responses against the commensal microbiota. Notably, recipients gavaged with B. bifidum were protected from colitis, highlighting the beneficial role of B. bifidum in promoting tolerance to a variety of different antigens in vivo. Inappropriate immune responses to gut commensals are a feature of intestinal inflammatory disorders. Identifying targetable pathways that dampen these responses through the induction of T_reg cells may bring about long-term improvement in patients who suffer from inflammatory bowel disease.

Diving deeper into the molecular interactions between B. bifidum and intestinal leukocytes, the authors identified specific B. bifidum–derived ligands that recapitulate the immunological changes that were observed after treatment with intact bacterial preparations. Subfractionation and nuclear magnetic resonance analysis of cell surface polysaccharides revealed a mixture of neutral glycans capable of inducing T_reg cells in vitro, with evidence for β-1-6-glucan being primarily responsible. Administration of B. bifidum neutral cell surface β-glucan/galactan polysacchaccharides (CSGG) alone ameliorated progression of inflammatory colitis, indicating CSGG to be a functional component of B. bifidum in promoting immune tolerance.

Several reports have identified that particular bacterial-derived polysaccharides and short-chain fatty acids generated by bacterial metabolism of complex carbohydrates can promote T_reg cell differentiation (6, 7). Gut-resident dendritic cells (DCs) are key antigen-presenting cells that shape the intestinal T cell compartment through activation of the TCR and production of polarizing cytokines (8). Toll-like receptor (TLR) ligands, including evolutionarily conserved pathogen-associated molecular patterns, promote DC activation and elicit inflammatory programs that influence T cell lineage commitment. In the setting of B. fragilis colonization, TLR2 expression on both DC and T cells has been shown to contribute to PSA-driven T_reg induction (7, 9). Dasgupta et al. identified that recognition of PSA by plasmacytoid DCs (pDCs) is a critical step that defines a pro-tolerance phenotype, and differentiation of T_reg cells occurred via reciprocal inducible costimulator (ICOS)–ICOS ligand (ICOS-L) interactions between regulatory pDCs and cognate T cells (9). In a similar vein, previous studies in the Im laboratory identified a mixture of probiotic bacteria that supports the differentiation of intestinal DCs that exhibit a “regulatory phenotype,” marked by high expression of IL-10, transforming growth factor-β (TGF-β), Cox-2, and indoleamine 2,3-dioxygenase (10).

In this context, the generation of T_reg cells with diverse specificities after colonization with B. bifidum prompted Verma et al. to examine whether bacterial products directly influenced T_reg cell differentiation or whether T_reg cell induction occurred indirectly by activating regulatory DCs (rDCs). In support of their earlier work and that of others in the field, the authors demonstrate that surface glycans of B. bifidum act to modify DC gene expression through a partially TLR2-dependent mechanism. Although incubation of naïve T cells with bacterial products was unable to promote T_reg cell differentiation, delivery of the neutral fraction of surface glycans (CSGGs) to DCs enhanced rDC-associated gene expression, including up-regulation of Pdcd1, Tgfb1, Csf2, Ptgs2, Il10, and Il27. Increased production of TGF-β by conventional DCs (cDCs) as a result of stimulation with CSGGs was primarily responsible for the generation of T_reg cells because antibody-mediated neutralization of TGF-β almost entirely prevented their induction. CSGG treatment of cDCs isolated from TLR2^−/− mice or animals deficient for the downstream signaling adaptor MyD88 exhibited a partial but important reduction in their ability to support T_reg cell induction. Signaling through TLR3, TLR6, TLR7, and TLR9 as well as through the C-type lectin receptors Dectin1 and Dectin2 had no observable role in CSGG-mediated T_reg cell induction. These results describe a molecular mechanism by which DC-mediated detection of specific B. bifidum–derived glycans induces transcriptional changes that, in turn, alter T cell differentiation. This is in contrast to B. fragilis PSA, which, in addition to promoting regulatory pDCs, has also been reported to bias T_reg cell differentiation by directly activating TLR2 expressed by T cells.

The identification of B. bifidum–derived molecules capable of drastically altering DC inflammatory gene signatures provides a method for the manipulation of adaptive immune responses within the gastrointestinal tract. Co-opting pathways that are influenced by gut microbes to modulate immune cells may provide powerful opportunities to bolster protective immunity or limit systemic inflammation. The present study circumvents the need for administering nonspecific probiotics by providing a small subset of polysaccharides capable of achieving T_reg differentiation in vivo while highlighting β-1-6-glucan as a potentially key effector within the mixture. Future studies may probe the use of these immunomodulatory polysaccharides in the treatment of inflammatory bowel disorders as well as investigate the role of B. bifidum in altering response rates to existing therapeutic interventions.