Abstract Body

Background: CMV replication is frequent in the genital tract of HIV infected men, and is associated with increased immune activation and HIV disease progression. We hypothesized that CMV-associated immune activation could influence the size of the HIV DNA reservoir during suppressive antiretroviral therapy (ART). Methodology: Paired blood and semen samples from 53 HIV infected men on long-term ART with undetectable HIV RNA in blood plasma were studied. Levels of CMV DNA were measured in seminal plasma by RT-PCR. Total HIV DNA (pol), 2-long terminal repeat (2-LTR), and cellular HIV RNA (unspliced [us] gag and multiply-spliced [ms] encoding for tat/rev) were measured in peripheral blood mononuclear cells (PBMC) by droplet digital PCR. Levels of CD4+ and CD8+ T-cells immune activation (CD38+HLA-DR+) and proliferation (Ki67+) were determined from PBMC by multicolor flow-cytometry. Associations between levels of proviral HIV DNA, cellular HIV RNA, immune activation and proliferation in blood, time on ART, CD4+ T-cells counts and the presence of CMV replication in semen were investigated using univariate and multivariate non-parametric statistics. Results: Detectable CMV DNA was found in 44% of seminal samples, and was associated with higher levels of proliferating (Ki67+) and activated (CD38+HLA-DR+) CD4+T-cells (P<0.01), and a trend towards higher levels of proliferating CD8+ T-cells (P=0.1) in blood. Subjects with detectable seminal CMV had higher levels of proviral HIV DNA (2.84 versus 2.11 log10 copies/million CD4+, P=0.05) and higher levels of cellular [ms] HIV RNA (P=0.04) compared to those without seminal CMV. Levels of [us] HIV RNA, 2-LTR and average cellular transcription (i.e. cellular HIV RNA/DNA ratio) were not different between subjects with and without CMV. There was a weak correlation between seminal CMV DNA and CD4-associated HIV DNA levels (P=0.06, r= 0.26). Among subjects with detectable CMV in semen, there was a significant correlation between HIV DNA levels and frequency of activated (P<0.01, r=0.6) and proliferating CD4+ T-cells (P<0.01, r=0.64). In multivariate analysis, presence of seminal CMV (P=0.04), and detectable 2-LTR (P<0.01) were independent predictors of higher HIV DNA in PBMC. Conclusions: Asymptomatic seminal CMV replication is associated with higher levels of systemic CD4+T-cell immune activation and proliferation and with a greater size of the HIV DNA reservoir in PBMC. Average HIV transcription was not increased in CMV positive samples, suggesting that CMV is not directly associated with low-level HIV replication but more likely with homeostatic proliferation of HIV-infected CD4+ T-cells. Interventions aimed at reducing seminal CMV replication and associated immune activation could be important for HIV curative strategies.