Background: Small molecules targeting bromodomain (BRD) proteins, a well-conserved class of transcriptional regulators distinguished by the presence of bromodomains, have recently emerged as novel epigenetic therapeutics in hematological and virological disease. Bromodomains are small helical interaction modules that specifically bind acetylated lysines in histones and nonhistone proteins. We recently tested the effect of bromodomain inhibitors JQ1, I-BET, I-BET151 and MS417 on HIV latency. We and others showed that these compounds effectively reactivate HIV from latency in cell culture cells and select primary T cell models of latency. We hypothesize that bromodomain inhibitors target BRD proteins to neutralize their suppressive action and activate HIV from latency. Methodology: We performed a comprehensive lentiviral shRNA screen of BRD proteins in J-Lat cells to identify which BRD protein is involved in the regulation of HIV latency. We also performed co-immunoprecipitation experiments to test for interaction of P-TEFb and BRD proteins. Results: We tested two independent shRNAs against BRD1-9 in J-Lat A2 and A72 cells and tested reactivation from latency by flow cytometry of GFP with and without JQ1. Knockdown of BRD8 resulted in activation of the HIV LTR, similar to activation observed with BRD2 knockdown, however, this reactivation was still responsive to JQ1. In contrast, knockdown of BRD3 resulted in a decrease of basal HIV transcription, and the activatory potential of JQ1 was unchanged. Co-treatment of cells with JQ1 and DRB, a CDK9 inhibitor, resulted in a decreased ability of the cells to respond to JQ1 treatment. Similar results were observed when Cyclin T1 was knocked down with two independent Cyclin T1 shRNAs, indicating that the JQ1 effect in latent cells requires intact P-TEFb. In co-immunoprecipitation experiments we found that BRD2, which lacks a bone-fide P-TEFb interacting domain and has not been associated with P-TEFbbinding, efficiently co-immunoprecipitated with the P-TEFb component Cyclin T1, an interaction that was enhanced in the presence of JQ1. This interaction also depended on intact acetylation sites in Cyclin T1, indicating that BRD2 interacts with P-TEFb via acetylated Cyclin T1 and the BRD2 bromodomain. Conclusions: Our data indicate that bromodomain inhibitors activate HIV latency by a Tat-independent mechanism and implicate a thus far unrecognized bromodomain family member BRD8 in the establishment and/or maintenance of HIV latency. Further, we confirm BRD2 as a candidate target of bromodomain inhibitors in the reversal of HIV latency.