A few microbiome-derived metabolites have been implicated in human disease, but the microbiota produce hundreds of additional metabolites that are not well studied. Moreover, many of these chemicals possess disease connections which have not yet been explored in detail. In unpublished collaborative studies with Project 1 using untargeted metabolomics as a discovery platform, we uncovered two such metabolite classes with members whose levels are correlated with the risk of developing cardiometabolic disease: aryl sulfates and secondary bile acids. Aryl sulfates, the focus of Aim 1, including indoxyl sulfate (IS) and p-cresol sulfate (pCS), derive exclusively from gut microbiota generated metabolites and their plasma levels correlate with cardiovascular disease (CVD) risk. Through untargeted metabolomics in subjects with preserved renal function we identified new (previously unknown) Aryl sulfates in plasma that are microbial in origin and associated with CVD. Secondary bile acids, Aim 2: The bile acid pool is remarkably concentrated and consists almost entirely (98%+) of microbiome-derived bile acids. Because bile acids are present in high concentrations, a compound that makes up even 1% of this pool is present at biologically relevant concentrations. In screening studies with Projects 1 & 2 (Hazen & Brown) among T2DM patients we discovered plasma levels of two bile acid sub- classes strikingly correlate with disease risks, lithocholic acid (directly correlated with CVD) and isoDCA/isoLCA (inversely correlated with T2DM). To date, it has been difficult to directly test the role of gut microbiota generated metabolites like aryl sulfates and secondary bile acids due to the inability to `toggle' individual metabolites on/off within a host. Leveraging our expertise in metabolic pathway discovery and microbiome gene editing we have enabled studies of causality and mechanism for two key classes of gut microbiota-derived molecules, aryl sulfates and secondary bile acids. We take a parallel approach in Aims 1 and 2 that starts with powerful human metabolomics data that links microbiome-derived metabolites to CVD and T2DM (Aims 1a/2a). We will predict biosynthetic pathways for metabolites that emerge from these analyses (Aims 1b/2b), create gain- and loss-of-function mutants in these pathways, and validate them in vitro (Aims 1c/2c). We will then validate the pathways in vivo by colonizing germ-free mice with strain or community pairs that differ only in the production of the metabolite of interest (Aims 1d/2d). We will then use these carefully controlled mice to study the role of the microbial enzyme and metabolite of interest in phenotypes relevant to CVD and T2DM (Aims 1e-f/2e-f).
