Depletion of key gut bacteria predicts disrupted bile acid metabolism in inflammatory bowel disease

The gut microbiome plays a key role in bile acid (BA) metabolism, where a diversity of metabolic products contribute to human health and disease. In particular, Inflammatory Bowel Disease (IBD) is characterized by a low concentration of secondary bile acids (SBAs), whose transformation from primary bile acids (PBAs) is an essential function performed solely by gut bacteria. BA-transformation activity mediated by the bile acid inducible (bai) operon has been functionally characterized in the genus <i>Clostridium</i>, and homologous bai gene sequences have been found in metagenome-assembled genomes (MAGs) belonging to other taxa in the human gut, but it is unclear which species of bai-carrying bacteria perform physiologically significant amounts of bile acid transformation in healthy and sick individuals. Here, we analyzed hundreds of stool samples with paired metagenomic and metabolomic data from IBD patients and controls and found that the abundance of the bai operon in metagenomic samples was highly predictive of that sample's high- or low-SBA metabolic state. We further found that bai genes from the <i>Clostridium</i> species best characterized as BA transformers were more prevalent in IBD patients than in non-IBD controls, while bai genes from uncharacterized taxa known only from MAGs were much more physiologically relevant in non-IBD samples. These un-isolated clades of BA-transforming bacteria merit further research; as beyond their prevalence in the human population, we found some cases in which they engrafted in IBD patients who had undergone fecal microbiota transplantation and experienced a clinical response.IMPORTANCEIn this paper, we identify specific bacteria that perform an important metabolic function in the human gut and demonstrate that in the guts of a large subset of patients with IBD, these bacteria are missing and the function is defective. This is a rare example where the correlation between the absence of specific bacteria and the dysfunction of metabolism is directly observed, not in mice nor in the lab, but in physiologic microbial communities in the human gut. Our results point to a path for studying how a small but important set of bacteria is affected by conditions in the IBD gut and perhaps to the development of interventions to mitigate the loss of these bacteria in IBD.