UC Berkeley

Abstract: The genomic catalogue of the human microbiota has expanded dramatically in recent years, and insights derived from human microbiota genomics has vast potential to generate treatments for human diseases. However, predictably harnessing the microbiota for beneficial outcomes is currently limited by our lack of understanding of the physiology of the constituent bacteria. For instance, the functions of most of their genes are not known. Here, we systematically measure mutant phenotypes for genes from the gut commensalBacteroides thetaiotaomicron. Using a barcoded transposon mutant library, we measured the fitness of 4,055B. thetaiotaomicrongenes across 492 experiments, including growth on 45 carbon substrates and in the presence of 57 stress-inducing compounds. Our data is in strong agreement with previous studies, and more importantly also uncovers the biological roles of poorly annotated genes. We identified 497 genes with a specific phenotype in only one or a handful of conditions, thus enabling informed predictions of gene function for a subset of these genes. For example, we identified a glycoside hydrolase important for growth on type I rhamnogalacturonan, a DUF4861 protein for glycosaminoglycan utilization, a DUF1080 protein for disaccharide utilization, and a tripartite multidrug resistance system specifically important for bile salt tolerance. Our approach can be applied to other members of the human microbiota to experimentally characterize their genes.