Impact of Glycan Depletion, Glycan Debranching and Increased Glycan Charge on HIV-1 Neutralization Sensitivity and Immunogenicity

Abstract

Modifying HIV-1’s envelope glycoprotein glycans can impact its neutralization sensitivity. The use of the knock out cell line GnT1-prevents the elaboration of complex-type glycans, and opens up the glycan shield, increasing bNAb vulnerability. Some bNAb precursors can bind to GnT1-trimers, supporting their use in vaccine priming. However, GnT1-trimers express poorly and exhibit very low infectious counts. Here, we describe two other potentially vaccine-relevant glycoengineered trimers, 1) To truncate complex glycans, we used of a cocktail of glycosidases, termed “NGAF3” (Neuraminidase, β-Galactosidase, N-Acetylglucosaminidase and endoglycosidaseF3). Like GnT1-trimers, NGAF3 reduced glycan clashes and increased bNAb potency while retaining a closed trimer conformation; 2) Modified by β-1,4-galactosyltransferase 1 (B4GalT1) and β-galactoside α-2,6 sialyltransferase 1 (ST6Gal1) during Env biosynthesis. Glycan mass spectrometry revealed that NGAF3 removed glycan heads of 3 of 7 positions on the trimer that are largely occupied by complex glycans. It also revealed a novel B4GalT1 activity to favor glycan precursor conversion to hybrid glycans rather than complex glycans. A comparison to monomeric gp120 revealed that B4GalT1’s new activity depends on tight glycan spacing. B4GalT1 affected more glycans than NGAF3 (6 out of 7 glycans), perhaps due to greater accessibilityduringEnv folding rather thanafterfolding. Surprisingly, the N611 glycan was unaffected by either modification. B4GalT1 and ST6Gal1 cooperatively increased the abundance of hybrid glycans and α-2,6 hypersialylated termini. This occurred largely by amplifying the abundance of a limited number of hybrid glycan structures that are also present on unmodified trimers. In rabbit vaccinations, B4GalT1+ST6GalT1-modified virus-like particles reduced the frequency and titers of serum NAbs that showed a modest preference for modified glycans. Conversely, chronically HIV-1-infected donor plasma neutralizing antibody titers were 1.7- to 10.8- fold higher against B4GalT1+ST6GalT1-modified pseudovirus. Overall, our data provide tools for heterologous prime, boost and polishing vaccine regimens using modified glycans.AUTHOR SUMMARYAn HIV-1 vaccine remains one of the most significant biomedical challenges today. Vaccines often work by triggering virus fighting antibodies to stave off infection. For HIV-1, this has been exceedingly difficult because the target, called the HIV-1 envelope glycoprotein (Env) is extremely variable and carries a thick sugar coat, protecting it from all but a few rare broadly reactive antibodies (termed bNAbs) that sometimes develop during natural HIV-1 infection. Given these challenges, it is reasonable to propose that any successful vaccine will need to be rationally designed to trigger the rare bNAbs. To meet this goal might require more 3 different vaccine components. First, a vaccine “prime” to trigger rare bNAb proliferation, followed by heterologous “boosts” and “polishing” immunogens to select for bNAbs. In this study, we evaluated two potential immunogens. In one approach, we used cocktails of enzymes to strip Env’s sugar coat, in the hope of reducing the barrier to stimulate bNAb precursors. In another, we used excess enzymes to build particular structures on Env’s coat. We checked how well these two Env variants were recognized by antibodies. We also checked in fine detail how the sugars were changed at the molecular level. Finally, we immunized rabbits. Our data enrich the number of strategies available to further explore the concept of priming, boosting and polishing vaccine shots.