Abstract
HIV integrates into the host genome, creating a viral reservoir of latently infected cells that persists despite effective antiretroviral treatment. CD4-positive (CD4+) T cells are the main contributors to the HIV reservoir. CD4+ T cells are a heterogeneous population, and the mechanisms of latency establishment in the different subsets, as well as their contribution to the reservoir, are still unclear. In this study, we analyzed HIV latency establishment in different CD4+ T cell subsets stimulated with interleukin 15 (IL-15), a cytokine that increases both susceptibility to infection and reactivation from latency. Using a dual-reporter virus that allows discrimination between latent and productive infection at the single-cell level, we found that IL-15-treated primary human CD4+ T naive and CD4+ T stem cell memory (TSCM) cells are less susceptible to HIV infection than CD4+ central memory (TCM), effector memory (TEM), and transitional memory (TTM) cells but are also more likely to harbor transcriptionally silent provirus. The propensity of these subsets to harbor latent provirus compared to the more differentiated memory subsets was independent of differential expression of pTEFb components. Microscopy analysis of NF-κB suggested that CD4+ T naive cells express smaller amounts of nuclear NF-κB than the other subsets, partially explaining the inefficient long terminal repeat (LTR)-driven transcription. On the other hand, CD4+ TSCM cells display similar levels of nuclear NF-κB to CD4+ TCM, CD4+ TEM, and CD4+ TTM cells, indicating the availability of transcription initiation and elongation factors is not solely responsible for the inefficient HIV gene expression in the CD4+ TSCM subset. IMPORTANCE The formation of a latent reservoir is the main barrier to HIV cure. Here, we investigated how HIV latency is established in different CD4+ T cell subsets in the presence of IL-15, a cytokine that has been shown to efficiently induce latency reversal. We observed that, even in the presence of IL-15, the less differentiated subsets display lower levels of productive HIV infection than the more differentiated subsets. These differences were not related to different expression of pTEFb, and modest differences in NF-κB were observed for CD4+ T naive cells only, implying the involvement of other mechanisms. Understanding the molecular basis of latency establishment in different CD4+ T cell subsets might be important for tailoring specific strategies to reactivate HIV transcription in all the CD4+ T subsets that compose the latent reservoir.