SARS-CoV-2 mRNA VACCINATION EXPOSES LATENT HIV to Nef-SPECIFIC CD8+ T CELLS

ABSTRACT

Background: The kinetics and functional profiles (granzyme-B production) of HIV-specific T-cell responses support that those targeting the early viral gene product Nef disproportionately recognize residual antigen expression during long-term antiretroviral therapy (ART). Here, we leveraged this insight to test whether SARS-CoV2 mRNA vaccines-which activate TLR and inflammatory signaling pathways-would reactivate latent HIV, stimulating T-cell responses with these characteristics. Methods: T-cell responses to individual HIV gene products were measured by IFN-g or granzyme B ELISPOT, and by activation induced marker (AIM) assays at baseline and ∼2 weeks after SARS-CoV-2 mRNA vaccine prime and boost, in 13 long-term ART treated adults. Total and unspliced HIV mRNA, as well as intact and defective (IPDA) HIV DNA were measured in parallel by digital droplet PCR (ddPCR). Results: We observed transient increases Nef-specific T-cell responses following vaccine prime by granzyme B ELISPOT (3.1-fold increase, p=0.002) and a trend by AIM assay (1.5-fold increase, p=0.06). Such increases were not observed in granzyme B responses to late gene products nor in any IFN-g responses. Both unspliced and total HIV mRNA decreased significantly across the study, unspliced-1.6-fold decrease p = 0.03;total-1.5-fold decrease p = 0.05. Changes in total HIV mRNA correlated inversely with Nef-specific granzyme B-producing (spearman's ρ =-0.73, p = 0.006) and Nef-specific CD8+ AIM T-cell responses (ρ =-0.76, p = 0.006) following vaccine prime. These reductions in HIV RNA were not accompanied by significant changes in total or intact HIV DNA. Conclusion: Consistent with our hypothesis, a restricted profile of HIV-specific T-cell responses showed significant increases following SARS-CoV-2 vaccine prime, each of which were then correlated with reductions in HIV RNA. This supports that vaccination promoted productive interactions between Nef-specific CTL and HIV-infected cells in vivo. We propose three scenarios for why this was not reflected in reductions in intact or total HIV DNA i) meaningful depletions in inducible proviruses occurred but were lost against the background of non-inducible proviruses ii) interactions with CTL involved only a fraction of inducible proviruses, or iii) substantive proviral depletions occurred, but were counterbalanced by clonal expansion of HIV-infected cells.