Identifying the drivers of ESX1 evolution in Mycobacterium tuberculosis

Project Summary Even in trial settings, 5-10% of patients with uncomplicated, drug sensitive TB fail treatment. Data from our lab and others suggest that there are bacterial determinants of treatment response, in addition to high level drug resistance, which contribute to the variability in treatment response]. Population genomic analyses have defined Mtb genes under positive selection, providing a genetic roadmap to the differences between Mtb strains that may contribute to treatment failure. These include known drug resistance genes but also a few canonical virulence genes including multiple components of the ESX1 secretion system. Multiple structural components of the ESX1 system as well as two key ESX1 regulators show evidence of genetic diversification and positive selection to a similar degree and following similar timeline as known drug resistance genes. The cadence of ESX1 mutation suggests that they are driven by a strong selective pressure that began relatively recently, which we hypothesize to be drug. We will test the model that Mtb strains alter drug responses in people by altering their ESX1-dependent interactions with the host. We term this concept “virulence associated drug resistance”. In this R21, we propose two Aims to acquire foundational data to test this model. In the first Aim, we will use bacterial genetic approaches to ask what are the effects of clinically prevalent variants on ESX1 function in vitro? And in the second Aim, we will use mouse and macrophage models to test the effects of clinically prevalent variants on virulence and treatment outcomes in vivo. We anticipate that the results of these studies will provide data to direct more extensive, mechanistic dissection of ESX1 evolution in Mtb in future work.