Improving mRNA vaccines with extracellular vesicle-associated immunogens

Abstract The central hypothesis of this proposal is that the efficacy of mRNA vaccines that deliver membrane-anchored immunogens can be improved by localizing the immunogen to extracellular vesicles (EVs, small membrane- limited structures shed by eukaryotic cells). Our rationale is that EVs provide a natural scaffold for immunogen multimerization while also enabling membrane-bound antigens to access antigen presenting cells, both local to the site of injection, and in the draining lymph node. To localize immunogens to EVs and promote EV shedding we propose two complimentary approaches. In Aim 1, we will append a viral "late domain" to the carboxy terminus of our immunogen. Viral late domains are small protein domains, usually associated with a matrix or capsid protein, used by enveloped viruses to facilitate budding and egress. We have found that these domains can act out of context; fusing a late domain from feline immunodeficiency virus Gag to a SARS-CoV-2 spike protein immunogen caused the immunogen to re- localize to EVs and improved its immunogenicity nearly two-fold. We will expand this work by testing late domains from other viruses for their ability to promote EV localization and/or production. We will thoroughly characterize these EVs to determine correlates of vaccine immunogenicity. In Aim 2, we will modify our immunogens to overcome the activity of the host anti-viral restriction factor BST-2 (a.k.a. tetherin). Tetherin inhibits viral egress by "tethering" budding enveloped viruses to the host cell membranes and also inhibits the release of EVs by the same mechanism. Therefore, we will explore strategies for antagonizing tetherin in order to promote release of our immunogen-laden EVs. Enveloped viruses have evolved different strategies for tetherin evasion that we will attempt to incorporate into our immunogen designs. Indeed, we have identified a portion of the SARS-CoV-2 spike protein that we suspect is responsible for tetherin antagonism. Incorporating this S protein domain into our immunogen dramatically increases the amount of immunogen recovered from EV fractions of tissue culture supernatants. We will also explore similar strategies based on tetherin resistance mechanisms from other viruses. Finally, in Aim 3, promising immunogen design strategies in the context of different viral envelope protein immunogens (SARS-CoV-2, influenza A virus, HIV) will be compared in mice. These tests will allow us to establish correlations between the behavior of our vaccine immunogens in tissue culture (quantity and characteristics of the EVs, cytoxicity, etc.) and performance of the vaccine in vivo and determine if our modifications universally improve vaccine efficacy, or if particular immunogen designs are better suited for specific viral antigens.