Rationally designed immunogens enable immune focusing following SARS-CoV-2 spike imprinting

Abstract

The emergence of SARS-CoV-2 (SARS-2) has highlighted the health risk posed by novel pathogens for which there is limited protective population immunity. Though vaccinations were developed and used rapidly across much of the world, SARS-2 continues to circulate and is evolving antigenically to escape host immunity. Several related sarbecoviruses that use the same human angiotensin converting enzyme 2 (ACE2) receptor to enter cells have been identified in animals. This suggests the possibility that additional novel sarbecoviruses may emerge. The combination of ongoing SARS-2 antigenic evolution and the possibility of zoonotic spillover underscore the need for pan-sarbecovirus vaccines that provide protection against both currently circulating variants and emerging viruses. Because a large proportion of people have previously been infected with or vaccinated against SARS-2, pan-sarbecovirus vaccine candidates must be evaluated in the context of pre-existing SARS-2 immunity. In this thesis, we evaluated structure-guided, rational immunogen design approaches to focus the humoral immune response to conserved epitopes on the SARS-2 receptor binding domain. We have pursued this objective in the following ways.

We demonstrated that boosting with receptor binding domain-based immunogens, rather than the full SARS-2 spike protein, elicits a memory B cell response that is biased towards the receptor binding domain. However, both boosting approaches generate broad serum neutralization across related sarbecoviruses. We leveraged hyperglycosylation and scaffolding design approaches to engineer immunogens that could focus to conserved epitopes on the receptor binding domain. Boosting with these engineered immunogens in the context of prior SARS-2 immunity focused the murine serum antibody response to conserved, broadly neutralizing epitopes. We also pursued structural characterization of several immunogen-elicited broadly neutralizing antibodies to help understand the basis of this broad serum neutralization. We refined the design of one of the hyperglycosylated immunogens to focus to the epitope targeted by these broadly neutralizing antibodies. Boosting with this immunogen in the context of prior immunity to SARS-2 in mice resulted in immune focusing. Additionally, this immune focusing significantly improved serum neutralization of the antigenically distant Omicron variant. We also characterized the cross-reactive human SARS-2 antibody response and identified broadly neutralizing antibodies that targeted a similar epitope. Our data support the use of hyperglycosylation and scaffolding immunogen design approaches to focus humoral responses to epitopes on the SARS-2 receptor binding domain in context of prior SARS-2 immunity. These tools may provide a framework for a future pan-sarbecovirus vaccine.