The Ernst Lab studies mechanisms of immune evasion and subversion in tuberculosis. Mycobacterium tuberculosis causes a chronic infection in humans and experimental animals, and the bacteria are not eliminated despite development of a robust cellular immune response. The Ernst Lab has recently discovered that, while M. tuberculosis-infected humans and mice generate appropriately-differentiated antigen-specific CD4+ and CD8+ T cells that traffic to the site of infection, only a small fraction of antigen-specific effector T cells recognize antigens at the site of infection in TB. In addition, they have also discovered that the M. tuberculosis epitopes known to be recognized by CD4+ and CD8+ T cells from humans with tuberculosis are hyperconserved in the M. tuberculosis genome. This finding implies an evolutionary benefit to the bacteria from human T cell recognition, and has important implications for vaccine design.