PROJECT SUMMARY/ABSTRACT The Marburg virus (MARV) and Ravn virus (RAVV), which both belong to the genus Marburgvirus of the family Filoviridae, cause the severe disease in humans, with case fatality rates up to 90%. There are no licensed vaccines against marburgviruses. Clinical trials of vaccine candidates against Ebola virus (EBOV), which belongs to the genus Ebolavirus of the family Filoviridae, demonstrated that the high vaccine doses that are required to induce an immune response at the protective level result in toxic effects associated with their principal component EBOV glycoprotein (GP). We and others have recently isolated and characterized human monoclonal antibodies (mAbs) to MARV and EBOV and defined the principal antigenic determinants for neutralization and protection on filovirus GP. We have demonstrated the successful protection of non-human primates against MARV by passively transferred mAbs. The central hypothesis of this study is that the epitopes of naturally-occurring human protective antibodies from survivors of a MARV infection can be used as templates for optimal rationally-designed structure-based vaccines. This hypothesis is supported by our recent extensive progress in the isolation of protective mAbs from survivors in conjunction with the recent advances in computational immunology techniques. We propose the rational design of structure-based vaccine MARV candidates that present the immunogenic determinants on GP. As in the wild-type (wt) GP, antigenic elements on the protein are obscured by glycosylation, the glycan cap and the mucin-like domain. The designed antigens are expected to better present protective determinants than the wt GP. We also propose that a vaccine based on conserved GP epitopes will be protective against both MARV and RAVV. The vaccine will use a highly innovative vaccine delivery platform based on pseudouridin-modified RNA delivered in a lipid nanoparticle formulation. The proposal is based on an interdisciplinary approach with a diverse team of experts in computational modeling, antibody and antigen discovery, filovirus virology, immunology and vaccinology. The Meiler computational group will use the ROSETTA software platform techniques to design novel structure- based vaccine candidate antigens, using high-resolution structures of antigen-antibody complexes in the GP receptor-binding domain. The Crowe laboratory will generate recombinant antigens and antibodies, validate proper structure and function of the constructs, and determine the fine details of their biomolecular interaction. Moderna Therapeutics will provide the innovative mRNA vaccine platform. The Bukreyev and Geisbert laboratories will test the vaccine constructs expressing the designed antigens in rodent and non-human primate models of marburgviruses, and the Bukreyev laboratory will perform in-depth characterization of the immune response. The completion of this proposal will result in the development of a universal and safe next...