Host-pathogen interactions and carbon processing in glycerol kinase deficient Mycobacterium tuberculosis

Tuberculosis (TB) is the second leading cause of death due to infectious disease worldwide and is notoriously difficult to treat. This is partially due to the emergence of drug-tolerant populations of Mycobacterium tuberculosis (Mtb), the causative agent of TB. Drug tolerance, defined as a non-heritable phenotypic state of reduced growth induced by stressors such as antibiotic treatment, differs from drug resistance, which is permanent and heritable. The transient nature of drug tolerant populations provides a major barrier to identifying drug targets that kill Mtb. Our lab identified a key glycerol kinase gene (glpK) as a major determinant of drug tolerance in Mtb driven by reversible mutations in a homopolymeric region. These mutants are found in clinical samples and are associated with poor clinical outcomes. Identifying novel drug targets against these clinically significant glpK mutants may significantly improve the outcomes of TB treatment. However, little is known about the bacterial and host factors that confer drug tolerance to glpK mutants in vivo. To study this clinically important phenotype, we generated a H37RV glpK strain with a permanent drug tolerant phenotype. We found that glpK strain mimics classic phenotypes of drug tolerant Mtb; glpK strain forms small colony variants, exhibits drug tolerance, overexpresses genes related to dormancy, differentially expresses carbon metabolism genes, and accumulates intracellular lipids. Given that glpK strain lacks a functional glycerol kinase, it is likely that glpK strain depends on carbon sources other than glycerol for drug tolerance. However, most experiments are performed in glycerol-containing media. As drug tolerance in Mtb is often associated with cholesterol accumulation and Mtb persistence requires host cholesterol, we selected this biologically relevant condition to test the role of lipid metabolism on drug tolerance in glpK strain. Furthermore, preliminary data suggest that macrophage infection can induce glpK strain drug tolerance, but not all macrophage environments are equally likely to have a high drug tolerance induction phenotype. We hypothesize that drug tolerance in Mycobacterium tuberculosis glpK mutants is connected to alterations in lipid metabolism, which occur in vitro as well as in infected macrophages. The goal of this project is to identify new drug targets in Mtb glpK mutants under biologically significant conditions to improve translatability and unlock novel therapeutics that enhance the overall efficacy of TB treatment. To identify novel drug targets under biologically significant conditions, we propose two models: a cholesterol- containing media model (Aim 1) and a macrophage infection model (Aim 2). We will probe the essentiality of relevant genetic pathways and genes within Mtb that are required for drug tolerance and then use specific macrophage mutants as well as a library of genetically diverse macrophages to study how specific host environments as well as a variable host genetic background affect bacterial drug tolerance. Together, these studies will identify within pathogen, within host, and host-pathogen interactions, including genetic and metabolic cross-talk, that regulate Mtb survival and drug tolerance, leading to novel drug targets against drug tolerant Mtb.