EspA Acts as a Critical Mediator of ESX1-Dependent Virulence in Mycobacterium tuberculosis by Affecting Bacterial Cell Wall Integrity

Mycobacterium tuberculosis (Mtb) requires the ESX1 specialized protein secretion system for virulence, for triggering cytosolic immune surveillance pathways, and for priming an optimal CD8+ T cell response. This suggests that ESX1 might act primarily by destabilizing the phagosomal membrane that surrounds the bacterium. However, identifying the primary function of the ESX1 system has been difficult because deletion of any substrate inhibits the secretion of all known substrates, thereby abolishing all ESX1 activity. Here we demonstrate that the ESX1 substrate EspA forms a disulfide bonded homodimer after secretion. By disrupting EspA disulfide bond formation, we have dissociated virulence from other known ESX1-mediated activities. Inhibition of EspA disulfide bond formation does not inhibit ESX1 secretion, ESX1-dependent stimulation of the cytosolic pattern receptors in the infected macrophage or the ability of Mtb to prime an adaptive immune response to ESX1 substrates. However, blocking EspA disulfide bond formation severely attenuates the ability of Mtb to survive and cause disease in mice. Strikingly, we show that inhibition of EspA disulfide bond formation also significantly compromises the stability of the mycobacterial cell wall, as does deletion of the ESX1 locus or individual components of the ESX1 system. Thus, we demonstrate that EspA is a major determinant of ESX1-mediated virulence independent of its function in ESX1 secretion. We propose that ESX1 and EspA play central roles in the virulence of Mtb in vivo because they alter the integrity of the mycobacterial cell wall.