In Vitro Exposure to Prostratin but Not Bryostatin-1 Improves Natural Killer Cell Functions Including Killing of CD4+ T Cells Harboring Reactivated Human Immunodeficiency Virus

In the attempt of purging the HIV-1 reservoir through the "shock-and-kill" strategy, it is important to select latency-reversing agents (LRAs) devoid of deleterious effects on the antiviral function of immune effector cells. Here, we investigated two LRAs with PKC agonist activity, prostratin (PRO) and bryostatin-1 (BRY), for their impact on the function of natural killer (NK) cells, the major effectors of innate immunity whose potential in HIV-1 eradication has emerged in recent clinical trials. Using NK cells of healthy donors, we found that exposure to either PRO or BRY potently activated NK cells, resulting in upmodulation of NKG2D and NKp44 activating receptors and matrix metalloprotease-mediated shedding of CD16 receptor. Despite PRO and BRY affected NK cell phenotype in the same manner, their impact on NK cell function was diverse and showed considerable donor-to-donor variation. Altogether, in most tested donors, the natural cytotoxicity and antibody-dependent cellular cytotoxicity (ADCC) of NK cells were either improved or maintained by PRO, while both activities were impaired by BRY. Moreover, we analyzed the effect of these drugs on the capacity of treated NK cells to kill autologous latently infected CD4(+) T cells reactivated via the same treatment. First, we found that PRO but not BRY increased upmodulation of the ULBP2 ligand for NKG2D on reactivated p24(+) cells. Importantly, we showed that clearance of reactivated p24(+) cells by NK cells was enhanced when both targets and effectors were exposed to PRO but not to BRY. Overall, PRO had a superior potential compared with BRY as to the impact on key NK cell functions and on NK-cell-mediated clearance of the HIV-1 reservoir. Our results emphasize the importance of considering the effects on NK cells of candidate "shock-and-kill" interventions. With respect to combinative approaches, the impact on NK cells of each LRA should be re-evaluated upon combination with a second LRA, which may have analogous or opposite effects, or with immunotherapy targeting NK cells. In addition, avoiding co-administration of LRAs that negatively impact ADCC activity by NK cells might be essential for successful application of antibodies or vaccination to "shock-and-kill" strategies.