Joel Ernst, MD

Our laboratory studies the mechanisms of protective and pathological immunity to Mycobacterium tuberculosis, the bacteria that cause tuberculosis (TB).  Even though TB can be treated and potentially cured, it still causes more human mortality than any other infectious disease, including HIV.  Our goal is to inform development of effective TB vaccines, and for this we use mouse models and human translational immunology studies.

We use mouse models to discover and characterize mechanisms of immune evasion employed by M. tuberculosis to prevent immune responses from clearing the infection and establishing sterile immunity and memory.  Among the mechanisms that we have discovered and characterized are:

1. The mechanisms that account for the long delay (in humans, 6 weeks; in mice, ~2 weeks) between initial infection and the development of T cell responses to M. tuberculosis antigens.  M. tuberculosis is transmitted by aerosol, and fewer than 10 bacteria are sufficient to establish infection.  We discovered that after aerosol infection of mice, live M. tuberculosis bacteria are transported by dendritic cells from the lungs to a local lymph node, where antigen produced by the bacteria in the lymph node primes naive CD4 T cells.  We also discovered that the rate limiting step in initiating antigen-specific CD4 T cell responses in TB is the acquisition of live bacteria by lung dendritic cells and that this process is accelerated by neutrophils. 
2. The mechanisms required for priming of antigen-specific CD4 T cells in TB.  We and others have discovered that M. tuberculosis inhibits MHC class II antigen presentation by infected cells, and this posed the question how CD4 T cell priming occurs at all.  We serendipitously discovered that M. tuberculosis-infected dendritic cells (which present bacterial antigens poorly) export bacterial protein antigens for uptake, processing, and presentation by uninfected dendritic cells in the lymph node, thereby bypassing the inhibition of antigen presentation by infected cells.
3. Mechanisms that prevent naturally-occuring T cell responses from being able to eliminate M. tuberculosis.  We first discovered that CD4 T cells must directly recognize M. tuberculosis-infected cells in the lungs to control intracellular bacterial replication; ‘bystander activation’ of CD4 cells and diffusion of soluble cytokines is insufficient.  This indicated that any mechanism that limits T cell recognition of infected cells can contribute to persistence of M. tuberculosis infection.  This led to our paradoxical finding that the same antigen export mechanism that facilitates CD4 T cell priming in the lymph node also limits CD4 T cell control of M. tuberculosis in the lungs, by limiting antigen presentation by infected cells.  In a related study, we confirmed that suboptimal MHC class II antigen presentation is a virulence mechanism that allows M. tuberculosis to evade elimination by T cell responses, indicating that interventions that improve antigen presentation by M. tuberculosis-infected cells can promote clearance of infection.  We recently completed a medium-throughput RNAi screen and identified multiple genes whose products are implicated in export of mycobacterial antigens from infected cells – an ongoing project is to characterize the roles of each of those genes in distinct steps of antigen export and to determine whether blockade of antigen export can make CD4 T cell responses more efficacious against M. tuberculosis.  In another recent study, we discovered that, while CD4 T cells are recruited to the lungs of M. tuberculosis-infected mice, they are only rarely in contact with infected cells there, indicating that suboptimal spatial positioning of CD4 T cells relative to infected cells can also contribute to persistent infection.
4. Mechanisms employed by M. tuberculosis to evade elimination, in addition to evading CD4 T cells.  We recently combined high-speed flow sorting in a Biosafety Level 3 facility with RNA-Seq on sorted myeloid cell populations, to discover two novel cell-intrinsic mechanisms that we hypothesize contribute to persistence of infection with M. tuberculosis.  We infected mice with fluorescent protein-expressing bacteria, and sorted infected from uninfected cells from the lungs, and discovered that in certain myeloid cell subsets, most of the intracellular bacteria are dead (we term these host cells ‘resistant’), while in one abundant myeloid subset, there were multiple live bacteria per cell (we term these ‘susceptible’ host cells).  In follow up studies, we used the sorted cell subsets and RNA-Seq to discover: 1) downregulation of a specific cytokine receptor in the susceptible cell subset, and this is accompanied by underexpression of genes regulated by that cytokine signalling pathway; 2) disruption of the biosynthetic pathway for a cellular organelle essential for control of multiple intracellular pathogens; and 3) activation of genes that are classically considered to be specific for epithelial cells, pointing to a transition state for the susceptible cells from macrophages to epithelioid cells.  Our efforts are now focused on discovering the underlying host and bacterial mechanisms that account for each of these affected pathways, and on discovering small molecules that modulate the underlying mechanisms to convert susceptible cells to resistant cells and improve control and potentially eliminate M. tuberculosis.

Our studies of immunity to TB in humans began with our discovery that nearly all of the  M. tuberculosis epitopes recognized by human T cells are highly conserved, indicating that human T cell recognition of those epitopes does not exert evolutionary pressure to select escape mutants.  Following this, we combined comparative bacterial genomics and experimental immunology to discover novel M. tuberculosis antigens that do show evidence of diversifying selection, suggesting that human T cell recognition of those antigens is more detrimental to the bacteria than is T cell recognition of the conserved antigens.  This has led us to test the general hypothesis that human T cells specific for different antigens exhibit distinct patterns of differentiation, trafficking, and functions, and that some of these provide superior protection against TB compared with others.

Our studies of human immunity to TB are currently based on a cohort in Ethiopia, where we collaborate closely with investigators at the Armauer Hansen Research Institute to identify antigen-specific T cell signatures that are associated with differential outcomes of M. tuberculosis infection (long-term control vs. progression to active TB disease) that can be targeted by vaccine strategies to reduce the global burden of TB.