HIV post-treatment controllers (PTCs) represent a natural model of sustained HIV remission, but these individuals are rare and little is known about their viral reservoir. We performed near-full length proviral sequencing to assess the proviral reservoir landscape of a cohort of PTCs and post-treatment non-controllers (NCs). The results were also used to assess the relationship between different proviral species, levels of intracellular HIV RNA expression and host immune responses.
PBMCs were obtained from the pre-treatment interruption (on-ART) time point for 10 PTCs (controlled for ≥24 weeks) and 15 NCs identified through ACTG. Near-full length proviral sequencing was performed by Illumina next-generation single-genome sequencing (NG-SGS). Unspliced CA-RNA levels were quantified by qPCR. T and NK cell phenotypes were assessed by flow cytometry, and T cell intracellular cytokine staining was performed on PBMCs stimulated with an HIV [i]gag[/i] peptide pool.
We obtained a total of 1193 proviral genomes. PTCs had approximately 6-fold lower levels of total proviral genomes (TPG, PTCs vs. NCs: median 1.9 vs. 11.2 copies/106 PBMCs, P<0.001) and intact proviral genomes (IPG, 0.1 vs. 0.6 copies/106 PBMCs, P=0.01). 80% of NCs vs. 0% of PTCs had TPG <5 copies/106 PBMCs (P<0.001). There were no significant differences between PTCs and NCs in the percent of proviruses that were intact, harboring deletions, or identical sequences detected more than once (i.e., clonally expanded). Amongst all participants, a median 97% of proviral genomes were defective, most of which harbored large deletions. CA-RNA correlated with levels of defective proviral genomes (DPG, Spearman r=0.51, P=0.02), but not with levels of IPG. Higher CD38+ NK cell percentages correlated with lower levels of DPG (r=-0.48, P=0.02), as were higher levels of HIV-specific CD4+ cells expressing IFN-γ (r=-0.59, P=0.01) and CD8+ cells expressing CD107 (r=-0.56, P=0.02).
Prior to treatment interruption, the median total proviral genomes in PTCs was 6-fold lower than in NCs, but the proportions of intact, defective and clonally expanded viral genomes were similar in both groups. The association of defective proviral genomes with levels of CA-RNA, NK activation, and HIV-specific T cell activity support the concept that defective HIV genomes lead to viral antigen production and interact with both the innate and adaptive immune responses.