Metabolite Signals from the Microbiota: Instructing T Cell Fate and Function

The microbiota has a major impact on host health, and research suggests that the enzymatic capacity of the microbiota is important to mediate many of its effects on the host. Short-chain fatty acids (SCFAs) have recently been demonstrated to promote regulatory and effector T cell differentiation in the GI tract, but germ-free mouse studies suggest that additional signals from the microbiota are necessary in addition to SCFAs in order to recapitulate T cell lineage prevalence seen in colonized mice. We recently determined that a number of tryptophan (Trp) metabolites are present in colonized mice at high concentrations similar to SCFAs but not in germ-free mice. Therefore, we tested the ability of Trp metabolites to regulate T cell differentiation and investigated the interaction between Trp metabolites and SCFAs. We found that the Trp metabolites indole and 5-hydroxyindole have anti- and pro- inflammatory effects on T cell differentiation, respectively, and some of these effects are dependent on the aryl hydrocarbon receptor. Additionally, a combination of indole, 5-hydroxyindole, and SCFAs produced unexpected outcomes during T cell differentiation, specifically promoting tolerogenic T cell differentiation without inhibition or augmentation of inflammatory T cell differentiation. The powerful effects we observed in vivo on T cells conditioned with Trp metabolites in vitro suggests that these metabolites could be a useful conditioning agent for cellular therapeutics. In addition, the level of Trp metabolites and SCFAs being produced by the microbiota could be a useful marker for clinical manipulation of the microbiota during pathology such as Inflammatory Bowel Disease. To support the clinical implications of this work, we verified our major results in human T cells. Additionally, we established a protocol for generating a predictive model of microbiota metabolite interaction during T cell differentiation. We believe this tool will become increasingly useful as additional active microbiota-derived metabolites are identified and empirical determination of interaction between all types of active metabolites becomes increasingly cumbersome