Investigating Immune Correlates of Protection to Tuberculosis Using an Ultra-Low Dose Infection in a Mouse Model

BACKGROUND: Mycobacterium tuberculosis (M.tb) is the leading infectious cause of mortality; however, there is no vaccine that confers lasting protection. To investigate the immune response to M.tb, we have developed an ultra-low dose (ULD) infection model in mice that better reflects the heterogeneous outcomes of human infection. Additionally, we have identified a blood transcriptional signature, taken at day 24, that predicts future bacterial burden after day 70. METHODS: Mice were infected with an ULD (1–3 CFU) of H37Rv M.tb. Control mice were infected with 50–100 CFU. Blood was drawn at day 24 for RNA signature. At early (day 34–35), and late (day 82–83) time points, mice were injected with an intravascular label. Individual lungs were then assessed for bacterial burden, phenotypic and spatial analysis of immune cells by flow cytometry, and histocytometry, a type of immunohistochemistry that allows for imaging of >12 colors RESULTS: At day 35, the following cell populations were correlated with an RNA signature score predicting disease progression: ESAT-6 tetramer+ CD4+ T cells (R(2) = 0.35, P < 0.01), TB10.4 tetramer+ CD8+ T cells (R(2) = 0.34, P < 0.01), and B cells (R(2) = 0.28, p = 0.01) within the lung parenchyma, as well as CD11b+ cells, negative for CD64, Ly6c, Ly6g and MHCII (R(2) = 0.38, p < 0.01) within the lung vasculature. These same populations were correlated with elevated CFU at day 83, as well as dendritic cells (R(2) = 0.53, p < 0.01). No populations were correlated with a protective RNA score. We have observed the complex spatial organization of granulomas while optimizing our histocytometry panel. This includes infected macrophages (Fig 1) interdigitated with B cell aggregates, associated with naïve T cells, interspersed with CD44+ T cells, with diffuse staining for Ki67, suggestive of tertiary lymphoid structures (Figure 2). CONCLUSION: This model replicates heterogeneity of TB seen in humans, while also providing a way to correlate differences in the immune response to future outcome. We have associated distinct immune cell subsets with the failure to control TB. With a larger sample size and data from histocytometry, we will have improved resolution to discern protective elements of the immune response to TB, which we can then test mechanistically in our model. DISCLOSURES: All authors: No reported disclosures.