Human tuberculosis (TB) is a frequently fatal lung disease mostly caused by Mycobacterium tuberculosis (M. tb), a slow-growing intracellular pathogen. M. tb utilizes the ESX-1 secretion system, classified as type VII, to export the virulence factors needed for host infection and pathogenesis. A functional ESX-1 pathway largely involves the proteins encoded by two gene clusters, esx-1 and espACD. These proteins are diverse in functions, such as cytolysis, immunogenicity, pore forming, stabilization and energization. Although many of the ESX-1 associated proteins were characterized individually, the overall mechanism of ESX-1 secretion is still far from clear. The main objective of my thesis is to better understand the structure and function of the ESX-1 secretion system in M. tb. To achieve this, an integrated approach which combines biochemistry, biophysics and genetics was applied. The experimental results are organized into three parts, in which I investigated EspC, EccCb1 and the effect of calcium on ESX-1 secretion. Firstly, the focus of this thesis is EspC, a secreted protein encoded in the espACD operon which is not linked to the esx-1 locus, but mediates ESX-1-associated virulence. Based on what I observed from a series of experiments, we propose a novel working model of the ESX-1 apparatus, in which EspC interacts with EspA in the cytosol and then translocates across the membrane to assemble into a channel in the outer membrane and spans the capsular layer of M. tb. This model could explain the mechanism by which the major virulence factors EsxA and EsxB (also named ESAT-6 and CFP-10) rely on EspC for secretion. Moreover, EspC is a potential outer membrane pore protein which has not been previously identified in M. tb. Secondly, the putative ATPase EccCb1 is another protein that was studied. EccCb1 is predicted as an ATPase that targets the EsxA/EsxB heterodimer during translocation. EccCb1 was overexpressed and purified in the form of a NusA- EccCb1 fusion protein due to its low stability. The protein was characterized as a hexamer with an ATPase activity. This result confirms that EccCb1 belongs to the FtsK/SpoIIIE ATPase family. Thirdly, we observed that ESX-1 secretion can be inhibited by a high concentration of calcium (~500 ÎŒM) in the culture medium. Moreover, bacterial persistence, rather than growth rate, decreased with the elevated calcium. RNA- seq was performed to evaluate the changes of global gene expression of M. tb in the presence of high calcium levels. Surprisingly, calcium dramatically repressed hypoxia response genes in the DevRS regulon, leading to reduced persistence. I also include a section that describes the additional work I was involved with, in which we investigated the activation of cGAS-dependent host response by ESX-1. Taken together, these findings provide new insights into the structural and functional aspects of ESX-1 secretion system, and will contribute to discovery of ESX-1 inhibitors with potential applications in TB therapy.