Investigating the RegX3-Dependent Regulation of Protein Release from Mycobacterium tuberculosis

Abstract

Mycobacterium tuberculosis (Mtb) is the causative agent of tuberculosis. As a facultative intracellular pathogen, Mtb is well-adapted to survive within the host and establishes a persistent infection in humans despite a strong immune response. As part of its survival strategy, Mtb releases a wide array of proteins that aid in nutrient acquisition and immune evasion. Broadly, the work presented in this thesis focuses on how an Mtb response regulator, RegX3, regulates protein export by the bacteria, and how export of these proteins impacts Mtb viability. In Chapter 2 we describe that a ∆pstA1 mutant, in which RegX3 is constitutively activated, hyper-secretes a variety of proteins. In previous work, we found that RegX3 induces secretion through the specialized ESX-5 secretion system. We also found hyper-secretion of LpqH, a lipoprotein associated with membrane vesicles (MV) produced by Mtb. Further investigation into this phenotype revealed that ∆pstA1 bacteria release significantly more MV than wild-type (WT) bacteria. We showed that this is a RegX3-dependent phenotype, suggesting a gene(s) within the RegX3 regulon controls MV production. Furthermore, MV release does not depend on ESX-5 activity, so RegX3 induces protein export by at least two different mechanisms. Work in Chapter 3 builds upon the observation that ∆pstA1 bacteria produce more MV. Because this strain constitutively activates RegX3, we leverage it as a tool to identify factors that drive MV release. We hypothesized that the deletion of a specific RegX3-induced gene in the ∆pstA1 background would restore MV production to WT levels, and that the product of this gene would provide information about the physical process of MV release through the Mtb cell wall. We demonstrate that only the deletion of lppF or whiB3 reduce MV production. While WhiB3 was previously reported to transcriptionally regulate the production of cell wall lipids, we found no evidence to support this regulatory mechanism in our ∆whiB3 strains. We conclude that no single gene drives the increased MV production by the ∆pstA1 strain; activation of WhiB3 is responsible for a portion of the phenotype and additional RegX3-dependent factors also contribute. Chapter 4 focuses on the biological functions of ESX-5 activity, specifically to determine how ESX-5 affects Mtb growth and what proteins are secreted by the system. Creating strains harboring tetracycline-repressible copies of ESX-5 components allowed us to circumvent ESX-5 essentiality while still controlling its activity. Using these strains, we found that Mtb requires ESX-5 activity for growth when either glycerol or glucose was the sole carbon source. Additionally, ESX-5 activity was required for optimal growth and survival in both resting and interferon-gamma activated murine macrophages in vitro. It remains to be investigated whether this intracellular survival is connected to ESX-5-dependent nutrient acquisition. These strains also allowed us to perform the first full proteomic analysis to identify protein substrates secreted via ESX-5. Our results uncovered that ESX-5 may play a role in the release of a wider range of proteins than was previously appreciated. We report that release of proteins associated with the cell membrane/cell wall and also proteins secreted via other systems rely on ESX-5 activity to reach the culture filtrate. Taken together, the results presented in Chapter 4 uncover new roles of the ESX-5 system and highlight the importance of this system to Mtb viability.