While combination antiretroviral therapy maintains undetectable viremia in People Living With HIV (PLWH), a life-long treatment is necessary to prevent viremic rebound after therapy cessation. This rebound seemed mainly caused by long lived HIV-1 latently infected cells reversing to a viral productive status. Reversing latency and elimination of these cells by the so-called shock and kill strategy is one of the main investigated leads to achieve an HIV-1 cure. Small molecules referred as latency reversal agents (LRAs) proved to efficiently reactivate latent CD4 + T cells. However, LRAs impact on de novo infection or HIV-1 production in productively infected macrophages remain elusive. Nontoxic doses of bryostatin-1, JQ1 and romidepsin were investigated in human monocyte-derived macrophages (MDMs). Treatment with bryostatin-1 or romidepsin resulted in a downregulation of CD4 and CCR5 receptors respectively, accompanied by a reduction of R5 tropic virus infection. HIV-1 replication was mainly regulated by receptor modulation for bryostatin-1, while romidepsin effect rely on upregulation of SAMHD1 activity. LRA stimulation of chronically infected cells did not enhance neither HIV-1 production nor gene expression. Surprisingly, bryostatin-1 caused a major decrease in viral production. This effect was not viral strain specific but appears to occur only in myeloid cells. Bryostatin-1 treatment of infected MDMs led to decreased amounts of capsid and matrix mature proteins with little to no modulation of precursors. Our observations revealed that bryostatin-1-treated myeloid and CD4 + T cells are responding differently upon HIV-1 infection. Therefore, additional studies are warranted to more fully assess the efficiency of HIV-1 eradicating strategies. Importance HIV-1 persists in a cellular latent form despite therapy that quickly propagates infection upon treatment interruption. Reversing latency would contribute to eradicate these cells, closing a gap to a cure. Macrophages are an acknowledged HIV-1 reservoir during therapy and are suspected to harbor latency establishment in vivo . Yet, the impact of latency reversal agents (LRAs) on HIV-1 infection and viral production in human macrophages is poorly known but nonetheless crucial to probe the safety of this strategy. In this in vitro study, we discovered encouraging anti-replicative features of distinct LRAs in human macrophages. We also described a new viral production inhibition mechanism by protein kinase C agonists which is specific to myeloid cells. This study provides new insights on HIV-1 propagation restriction potentials by LRAs in human macrophages and underline the importance of assessing latency reversal strategy on all HIV-1 targeted cells.

Introduction HIV‐1 eradication is hindered by the presence of inducible long‐lived reservoirs of latently infected cells which rapidly disseminate viral particles upon treatment interruption. Eliminating these reservoirs by the so‐called shock and kill strategy represents a crucial concept toward an HIV‐1 cure. Several molecules called latency‐reversing agents (LRAs) are under intensive investigations to reactivate virus gene expression. These studies are mainly conducted on CD4 + T cells where LRAs are well tolerated and did not induce global cellular activation. However, despite their broad spectrum, the putative impact of LRAs on other cellular reservoirs such as macrophages is still ill‐defined. Methods We investigated the impact of the protein kinase C (PKC) activator bryostatin‐1, bromodomain inhibitor JQ1 and histone deacetylase inhibitor romidepsin used either alone or in combination on human primary monocyte‐derived macrophages (MDMs). Results We demonstrate that bryostatin‐1, JQ1, and romidepsin or their combinations are not toxic at nanomolar concentrations but induce metabolic and morphologic alterations of MDMs. Bryostatin‐1 triggered the secretion of pro‐inflammatory cytokines, while JQ‐1 decreased it. Phagocytosis and endocytosis were modestly impaired upon bryostatin‐1 treatment whereas efferocytosis was markedly downregulated by romidepsin. Despite its pro‐inflammatory profile, bryostatin‐1 did not induce classically activated macrophage markers. Finally, we reveal that conditioned medium from bryostatin‐1‐treated macrophages did not potentiate its reactivation feature. Conclusions Our study reveals that LRAs can diversely impact basic physiologic features of human primary macrophages and could potentially decrease reactivation of nearby CD4 + T cells latently infected with HIV‐1. Our observations further stress the need to include different cell populations when assessing HIV‐1 cure strategies.

No abstract