Engineering HIV viral variants as immunogens to stimulate Broadly Neutralizing Antibodies

Abstract

Southern Africa is currently the most affected region of the world in terms of the HIV/AIDS pandemic and despite decades of research into this virus, a vaccine that is adequately effective is yet to be developed. However, it is widely believed that broadly neutralizing antibodies (bNAbs) are likely to be required for protection. Eliciting bNAbs by vaccination has been challenging for several reasons, one of which is the failure of most viral proteins to bind the germline versions of bNAbs (bNAb precursors). For vaccine design, it will be necessary to either select or engineer suitable immunogens that bind the inferred bNAb precursors in vitro, which is the first step in driving such responses towards breadth. This study focused on an unusual set of six previously identified viral escape mutations (collectively referred to as CS-Mut) that enhance neutralization by bNAbs to the membrane proximal external region (MPER). These mutations were used to engineer and test a variety of viral constructs for their potential as vaccine candidates. We first tested the effect of each individual cleavage site mutation on viral entry and MPER exposure. We showed that individual mutations resulted in reduced viral entry potential, but not to the same extent as all six mutations combined in the CS-Mut. We next tested the effect of the individual mutations on neutralization by MPER bNAbs. We showed that most of the single mutations enhanced viral sensitivity to MPER bNAbs, with three of the six mutations most promising in terms of MPER exposure. However, the MPER enhancement was less marked than the combined set of six mutations, suggesting further improvement was needed. We therefore next combined the three most promising mutations into a single construct and showed that enhancement of MPER sensitivity was improved for this triple mutant compared to the individual mutants. Indeed, for some MPER bNAbs, the triple mutant was more sensitive to MPER bNAbs than the matched virus containing all six mutations. Moreover, the entry capacity of the triple mutant was significantly improved over the six mutant construct, making it more amenable to incorporation into virus-like particles for vaccine design. Lastly, we engineered dual germline targeting immunogens by simultaneously adding the MPER enhancing mutations to five viruses that were previously shown to have a

high probability of binding V2 bNAb precursors (V2 “special strains”). The “special strain” CS-Mut viruses exhibited varying degrees of reduction in viral entry potential, with infectivity abrogated for two. For the remaining three “special strain” CS-Mut viruses, enhancement in MPER sensitivity was observed. Increased sensitivity was largely specific for bNAbs targeting the MPER epitope or interface, as bNAbs and plasma to other viral epitopes showed no enhanced neutralization. However, despite enhanced sensitivity to mature MPER bNAbs, no neutralization was observed using germline reverted MPER bNAbs, the best available approximate of MPER precursors. In conclusion, the cleavage site mutations resulted in favorable exposure of the MPER epitope, a trait that is promising for immunogen design. However, the fitness cost that results from the addition of the cleavage site mutations is problematic for use of these constructs in vaccines using virus-like particles. In future studies, a balance between viral entry potential and enhancement of MPER neutralization needs to be determined to optimize immunogen candidates. The CAP256 SU CS-Mut construct showed promise as a dual germline targeting immunogen, exhibiting limited reduction in entry potential while favorably exposing the MPER epitope, with minimal disruption to the native envelope trimer structure.