Background: The “shock and kill” approach to eliminating the latent HIV reservoir depends on efficient reactivation of HIV provirus. Multiple studies have observed variability in the levels of HIV reactivation after treatment of latently infected CD4+ T cells with the histone deacetylase inhibitor SAHA. In this study, we hypothesized that the heterogeneity in HIV reactivation is partly related to donor-to-donor differences that are reflected in host gene expression. Therefore, we sought to identify genes displaying a high level of donor-to-donor variability upon treatment with different doses of SAHA.
Methods: Primary CD4+ T cells from six healthy seronegative volunteers were incubated with 0.34, 1.0, 3.0 or 10.0 µM SAHA for 24 hours, or left untreated, after which RNA was extracted and gene expression was measured using Illumina HT-12 v4 BeadChips. To identify donor-to-donor variation specific to the SAHA treatment, we initially selected only those genes that were expressed at the same level in the untreated condition for each donor, and then ranked genes according to highest variation between donors upon SAHA treatment.
Results: The 100 most variable genes between donors showed higher donor-to-donor variation at the higher SAHA concentrations, 1, 3 and 10 µM compared to 340 nM as determined by Pearson’s correlation analysis. Protein interaction network analysis identified STAT1 (interferon-responsive transcriptional activator) and AKT2 (serine/threonine kinase) as “hub” genes within the 100 most variable genes, and gene ontology analysis revealed a cluster of genes whose functions are relevant to HIV latency. Interesting genes identified were JMJD1A, a lysine demethylase involved in hormone-dependent transcriptional activation; EGR2, a zinc-finger protein shown to interact with HIV Tat; SIN3B, a transcriptional repressor of MYC; L3MBTL3 a putative Polycomb group (PcG) protein and chromatin-interacting transcriptional repressor; and NAE1, which has been shown to interact with HIV-1 Vpr. A separate study of 4 donors treated with 1.0 µM SAHA confirmed similar inter-individual variability at the protein level.
Conclusions: It appears that SAHA has different effects in different donors at the level of gene and protein expression. Future work is required to determine if such differences lead to variation in the ability of SAHA to activate HIV in different donors. Future clinical trials with SAHA may need to be tailored to HIV infected individuals likely to respond favorably to this activating compound.