Structure-based design of coronavirus subunit vaccines

Project SummaryViral subunit vaccines are safe and convenient, but generally suffer low efficacy. Our overall hypothesis is thatan intrinsic limitation is associated with subunit vaccine designs in which artificially exposed surfaces of subunitvaccines contain epitopes unfavorable for vaccine efficacy. The receptor-binding domain (RBD) of acoronavirus spike protein consists of a core subdomain that serves as the structural scaffold and a receptor-binding motif (RBM) that binds the receptor and contains neutralizing epitopes. The RBDs are primecandidates for subunit vaccine designs. In preliminary studies, we identified epitopes on the core subdomain ofMERS coronavirus (MERS-CoV) RBD that were buried in the full-length spike protein but became artificiallyexposed in recombinant RBDs. We further showed that these epitopes severely reduce vaccine efficacy byinducing strong non-neutralizing immune responses and distracting the host immune system from reacting tothe neutralizing epitopes on the RBM. This novel finding reveals an intrinsic limitation of viral subunit vaccinesthat the vaccine field had been unaware of. In this proposal, we aim to characterize this intrinsic limitation andestablish novel approaches to overcome it. We use the RBDs from highly pathogenic coronaviruses, includingMERS-CoV and SARS coronavirus (SARS-CoV), as the model system. This proposal contains three majordesign approaches for coronavirus RBD vaccines. First, we will identify and characterize the artificiallyexposed unfavorable epitopes on the core subdomain of coronavirus RBDs. To this end, we introduce a novelconcept, neutralizing immunogenicity index (NII), to evaluate the contribution of each epitope to the overallvaccine efficacy. We will mask the negative epitopes on the core subdomain through glycan shielding orresurfacing. This design enhances the efficacy of the individually optimized RBD vaccines. Second, we willconstruct chimeric RBDs containing the core subdomain from one coronavirus RBD as the structural scaffoldand the RBM from another coronavirus RBD as the immunogenic sites. The unfavorable epitopes on the coresubdomain should have been silenced from the first design approach. The interface of the core subdomain andRBM will be optimized to maximize the stability of the chimeric RBD vaccines. This design prepares us for theemergence of highly pathogenic coronaviruses in the future. Third, we will construct nanoparticle-carriedcoronavirus RBD vaccines in a way that artificially exposed unfavorable epitopes on the core subdomain arere-buried at the molecular interfaces to enhance the RBD vaccine's efficacy. We will use mice to evaluate theimmunogenicity of the above engineered RBD vaccines and will use animal models (including hDPP4-knock-in(KI)) mice and non-human primates) to assess the selected RBD vaccines against live coronavirus challenge.Overall, this research establishes the artificially exposed unfavorable epitopes as the intrinsic limitation of viralsubunit vaccines and finds novel approaches to overcome it. Therefore, this research holds the promise ofmaking subunit vaccines a more successful and widely used strategy in combating virus infections.