Eradication of HIV-1 with highly active antiretroviral therapy (HAART) is not possible due to the persistence of long-lived, latently infected resting memory CD4 + T cells. We now show that HIV-1 latency can be established in resting CD4+ T cells infected with HIV-1 after exposure to ligands for CCR7 (CCL19), CXCR3 (CXCL9 and CXCL10), and CCR6 (CCL20) but not in unactivated CD4 + T cells. The mechanism did not involve cell activation or significant changes in gene expression, but was associated with rapid dephosphorylation of cofilin and changes in filamentous actin. Incubation with chemokine before infection led to efficient HIV-1 nuclear localization and integration and this was inhibited by the actin stabilizer jasplakinolide. We propose a unique pathway for establishment of latency by direct HIV-1 infection of resting CD4 + T cells during normal chemokine-directed recirculation of CD4 + T cells between blood and tissue.

Background: Eradication of HIV cannot be achieved with combination antiretroviral therapy (cART) because of the persistence of long-lived latently infected resting memory CD4 + T cells. We previously reported that HIV latency could be established in resting CD4+ T cells in the presence of the chemokine CCL19. To define how CCL19 facilitated the establishment of latent HIV infection, the role of chemokine receptor signalling was explored. Results: In resting CD4+ T cells, CCL19 induced phosphorylation of RAC-alpha serine/threonine-protein kinase (Akt), nuclear factor kappa B (NF-κB), extracellular-signal-regulated kinase (ERK) and p38. Inhibition of the phosphoinositol-3-kinase (PI3K) and Ras/Raf/Mitogen-activated protein kinase/ERK kinase (MEK)/ERK signalling pathways inhibited HIV integration, without significant reduction in HIV nuclear entry (measured by Alu-LTR and 2-LTR circle qPCR respectively). Inhibiting activation of MEK1/ERK1/2, c-Jun N-terminal kinase (JNK), activating protein-1 (AP-1) and NF-κB, but not p38, also inhibited HIV integration. We also show that HIV integrases interact with Pin1 in CCL19-treated CD4 + T cells and inhibition of JNK markedly reduced this interaction, suggesting that CCL19 treatment provided sufficient signals to protect HIV integrase from degradation via the proteasome pathway. Infection of CCL19-treated resting CD4 + T cells with mutant strains of HIV, lacking NF-κB binding sites in the HIV long terminal repeat (LTR) compared to infection with wild type virus, led to a significant reduction in integration by up to 40-fold (range 1-115.4, p = 0.03). This was in contrast to only a modest reduction of 5-fold (range 1.7-11, p > 0.05) in fully activated CD4 + T cells infected with the same mutants. Finally, we demonstrated significant differences in integration sites following HIV infection of unactivated, CCL19-treated, and fully activated CD4 + T cells. Conclusions:HIV integration in CCL19-treated resting CD4 + T cells depends on NF-κB signalling and increases the stability of HIV integrase, which allow subsequent integration and establishment of latency. These findings have implications for strategies needed to prevent the establishment, and potentially reverse, latent infection.