Engineering Bacterial Two Component Sensors to Study Gut Inflammation

The vast majority of bacteria in a healthy gut produce their energy by fermentation of complex polysaccharides and amino acids. However, a minority of bacteria found in the gut, including pathogenic enterobacteria such as E. coli and S. typhimurium, have a suite of enzymes, called reductases, which enable them to produce energy by coupling ATP production to the transfer of electrons from a donor molecule to the specific electron acceptor used by the reductase. Reactive oxygen and nitrogen species produced during intestinal inflammation result in the production of oxidized species that can be used by bacteria as electron acceptors during anaerobic respiration. These energy sources, only accessible to bacteria with reductases, allow enterobacteria to thrive during inflammatory conditions and out-compete the normal microbial flora, resulting in an altered composition of the microbiome. Enterobacterial blooms have been shown to exacerbate inflammation and are associated with inflammatory bowel diseases and colorectal cancer. To identify inflammatory conditions that can lead to enterobacterial blooms, we seek to detect inflammation-generated electron acceptors directly at the site of inflammation. Enterobacteria can use many terminal electron acceptors during anaerobic respiration including nitrate, nitrite, tetrathionate, TMAO, DMSO, and thiosulfate, all of which are produced in the gut during inflammation. Bacteria have naturally evolved to sense and respond to these molecules in the environment using two-component systems (TCSs), based upon membrane-associated histidine kinases. We have developed a computational and experimental TCS mining methodology and used it to identify and engineer the first ever tetrathionate and thiosulfate sensors. Our sensors show a strong response to and a high specificity for the desired ligand with no unwanted cross-regulation in contrast to a previously reported tetrathionate sensor, which is cross-regulated by oxygen. We have optimized our sensors to work in vitro in a non-pathogenic lab strain of E. coli as well as a probiotic gut-colonizing strain of E. coli for in vivo testing in mouse models of intestinal inflammation that lead to the development of colon cancer.