Spatial multi-omics to predict granuloma trajectory and bacterial restriction during vaccination

About This Grant

Project Summary The TB vaccine field has struggled to identify predictive correlates of protection against TB, especially following vaccination. Granulomas are immunological structures that have many features that overlap with tumors and the tumor microenvironment. A fundamental gap in knowledge in the TB vaccine field is what predicts whether granulomas are permissive or restrictive to Mtb growth after infection. Much like the tumor microenvironment, Mtb infected macrophages are similarly found in the complex granuloma microenvironment (GME). While TB specific T cells are found in granulomas, studies in monkeys suggest that very few gain access to the granuloma core where Mtb infected macrophages are found. Less is known about how the antigen-specificity, numbers, and proximity of T cells to Mtb infected macrophages determine whether T cells can promote Mtb killing. Profiling T cell interactions, phenotypes, and their spatial relationship to Mtb infected macrophages in lung and draining lymph nodes under conditions of natural infection and during vaccination will provide foundational knowledge about spatially relevant cell-cell interactions during the adaptive immune response to TB. Building upon predictive spatial modeling of tumor responsiveness to check-point inhibitor blockade, we propose to develop methodology to predict localized restriction of bacterial growth during vaccination by evaluating the GME as a novel way to assess vaccine efficacy. We will use spatial proteomic and transcriptional profiling, to comprehensively map immune cell infiltrates granulomas in naïve and vaccinated monkeys receiving unique live attenuated vaccines (LAV). We will then develop a computational pipeline to define GME frequency and distribution that predicts granuloma restriction of Mtb and correlate GME responses to changes in draining versus peripheral lymph node—an accessible tissue that can be surveyed longitudinal to better understand vaccine responses. Methods to combine and integrate spatial datasets with measures of clinical disease and peripheral immunological responses after vaccination are lacking. We will determine spatial correlates of vaccine efficacy by 1) mapping of cell networks and immunological synapses across granulomas during the adaptive immune response to Mtb and 2) generating multimodal modeling that incorporates both clinicopathological and spatial features including granuloma microenvironments to predict vaccine regimens and modalities that induce tissue resident memory and effective bacterial control. Assessment of unique spatial correlates of vaccine protection that complement traditional immunological assessments of vaccine immunogenicity may drastically alter the ways in which we evaluate vaccine efficacy and shift TB vaccine development priorities. Successful development of computational models of vaccine efficacy that incorporate spatial biology has future applications for many other infectious diseases of global health importance where vaccine development has struggled, including schistosomiasis, borreliosis, and malaria.