A MULTI-OMICS APPROACH TO CHARACTERIZING THREE HEALTH RELEVANT FUNCTIONS OF THE HUMAN GUT MICROBIOME

Abstract

The human gut is home to trillions of microorganisms that routinely interact with their human host in both beneficial and detrimental ways. The advent of next-generation sequencing and high-throughput “omics” technologies has created new opportunities to examine the role that the human gut microbiome plays on human health, especially in regard to gastrointestinal diseases such as Inflammatory Bowel Disease and colorectal cancer. In my dissertation, I utilize genomic, transcriptomic, metabolomic, and protein sequence datasets to characterize three health-relevant functions of the human gut microbiome. First, I performed a multi-omic, bioinformatic analysis to identify the bacterial enzyme, bilirubin reductase. While bilirubin reduction to urobilinogen and stercobilinogen is a well-known function of the human gut microbiome, the enzyme(s) responsible for the conversion of bilirubin to non-toxic reduced products have yet to be fully characterized. In this chapter, I review how I leveraged publicly available metabolomic, metagenomic, and metatranscriptomic data to explore over 2 million putative reductase genes and identify a candidate operon encoding bilirubin reductase. Second, I examined sources of microbial hydrogen sulfide (H2S) production by bacteria of the human gut microbiome. H2S is a sulfuric gas produced by various bacterial phyla of the human gut microbiome and is implicated in the etiology of gastrointestinal diseases such as Inflammatory Bowel Disease and colorectal cancer. In this chapter, I show via bioinformatic analysis that the capacity to produce H2S via cysteine degradation is ubiquitous in the human gut. Third, I explored bacterial prodrug activation required for the activation of immune system modulators such as sulfasalazine. After curating amino acid sequences of known azoreducing genes and performing a protein sequence search across the Unified Human Gastrointestinal Genomes (UHGG) collection containing 4,644 genomes, I identified putative azoreducing and non-azoreducing bacterial strains to be experimentally validated. Together, these results highlight a successful mult-omic approach to characterizing three diverse but health-relevant functions of the human gut microbiome.