Computationally designed anchor scaffolds for elicitation of broadly neutralizing influenza antibodies

Project Summary Influenza viruses remain a global health burden due to yearly epidemics and their pandemic potential. Therefore, understanding immunity to these viruses and further research on the development of improved vaccines is of high importance. The influenza hemagglutinin (HA) and neuraminidase (NA) proteins are the major targets of protective antibodies. Long-term protection to influenza viruses remains a challenge due to high mutation rates caused by a low-fidelity RNA polymerase as well as reassortment events of HA and NA with zoonotic influenza viruses, and this necessitates annual vaccination for protection against circulating strains. However, vaccine efficacy varies year to year due to mismatches between circulating strains and vaccine strains. This variability highlights the importance of developing improved influenza vaccines. We and others have recently discovered a class of antibodies targeting a conserved membrane-proximal epitope on the H1N1 influenza HA protein, termed the anchor epitope. The overall goal of this proposal is to elicit a robust immune response of anchor- specific and broadly neutralizing antibodies to influenza virus. As a main tool to achieve this goal, we will be scaffolding the anchor epitope to test the hypothesis that the scaffold will increase the prevalence of anchor- specific antibodies alone or together with a soluble recombinant HA as part of a prime-boost regimen. In Aim 1, we will utilize a computational strategy to scaffold the anchor epitope and rapidly screen constructs using previously isolated anti-anchor antibodies. Scaffolds will be generated using a combination of Rosetta and machine learning-based approaches to design and predict the stability and folding of the novel proteins. The top candidate proteins will be recombinantly expressed and tested for antigenicity using a panel of anchor-targeting antibodies as well as for thermal stability and monodispersion. In Aim 2, we will determine the efficacy of the anchor scaffolds in the mouse vaccination and infection models for the elicitation of broadly neutralizing antibodies. The top three candidates will be tested for immunogenicity and the ability to elicit anchor-like antibodies in mice using several combinations alone and in a prime-boost regimen with influenza HA protein. The top candidate will then be tested in a mouse challenge model with two different H1N1 pandemic viruses. This R21 proposal is high risk as we will develop new scaffolding strategies and vaccine candidates, but it is high reward as our approach has the potential to redirect strain-specific antibody responses toward a highly conserved, and broadly protective epitope. Overall, our proposal will develop new approaches for scaffolding broadly neutralizing protein epitopes, which could be applied to additional influenza and other viral glycoprotein epitopes.