Comparison of HDAC inhibitors in clinical development: effect on HIV production in latently infected cells and T-cell activation

Abstract

Objective: We aimed to compare the potential for inducing HIV production and the effect on T-cell activation of potent HDAC inhibitors undergoing clinical investigation.

Design: In vitro study

Results: The various HDAC inhibitors displayed significant potency differences in stimulating HIV-1 expression from the latently infected cell lines with panobinostat>givinostat ≈belinostat>vorinostat>valproic acid. Panobinostat was significantly more potent than all other HDAC inhibitors and induced virus production even in the very low concentration range 8-31 nM. The proportion of primary T-cells expressing the early activation marker CD69 increased moderately in all HDAC inhibitor-treated cells compared with untreated cells. Finally, proof was obtained that panobinostat, givinostat and belinostat induce virus production in latently infected primary cells at therapeutic concentrations with panobinostat being the most potent stimulator.

Methods: The latently infected cell lines ACH2 and U1 were treated with the HDAC inhibitors panobinostat, givinostat, belinostat, vorinostat and valproic acid. Viral induction was estimated by p24 production. Peripheral blood mononuclear cells from uninfected donors were treated with the HDAC inhibitors and the expression of activation markers on T-cell phenotypes was measured using flow cytometry. Finally, the ability of givinostat, belinostat and panobinostat to reactivate latent HIV-1 expression in primary T-cells was investigated employing a CCL19-induced latent primary CD4+ T cell infection model.

Conclusion: At therapeutic concentrations panobinostat stimulate HIV-1 expression in latently infected cells with greater potency than other HDAC inhibitors undergoing clinical investigation. These findings warrant further investigation and panobinostat is now being advanced into clinical testing against latent HIV infection.

Keywords: HIV; HIV cure; HIV eradication; histone deacetylase inhibitors.

Figures

Figure 1. HIV-1 expression in U1 and ACH2 cells treated with HDAC inhibitors. HIV-1 expression in U1 and ACH2 cells treated with panobinostat (LBH589), belinostat (PXD101), givinostat (ITF2357), vorinostat (SAHA) and valproic acid (VPA) for 48 h; median +/− IQR p24 concentrations of 15 (ACH2) and 8 (U1) separate experiments in each cell line are depicted in graphs A and B. Comparison of median +/− IQR induced viral production with indicated concentrations of LBH589 or SAHA is shown in C and D. Capped lines above error bars indicate the range of therapeutic plasma levels for each HDAC inhibitor. Dotted line indicates p24 levels in untreated cells.

Figure 2. Toxicity in lymphocytes and U1 cells treated with HDAC inhibitors. PBMCs from healthy donors were treated with panobinostat (LBH589) (2–62.5nM), givinostat (ITF2357), belinostat (PXD101), vorinostat (SAHA) (all 62.5–500nM) and valproic acid (VPA) (0.5mM). Panel A shows cell death in lymphocytes evaluated with flow cytometry using viability stain. Only data from the highest concentration of HDAC inhibitor is shown in figure. Toxicity of panobinostat was evaluated in U1 cells using a similar experimental setup. Panel B shows cell death in U1 cells treated with panobinostat. DMSO was used as negative control and PMA as positive control.

Figure 3. CD69 expression in CD4+ T-cells treated with HDAC inhibitors. PBMCs from healthy donors were treated 16 (CD69, CD45RA, CCR7, CD27) or 40 (CCR5, CXCR4) hours with panobinostat (LBH589) (2–62.5nM), givinostat (ITF2357), belinostat (PXD101), vorinostat (SAHA) (all 62.5–500nM) and valproic acid (VPA) (0.5mM). Expression of the early activation marker CD69 (n = 5), the phenotype markers CD45RA, CD27 and CC7 (n = 5) and the HIV co-receptors CCR5 and CXCR4 (n = 4) were assessed by flow cytometry. Mean +/− SEM percentage CD69 expression in CD4+ T-cells across the full range of concentrations used for panobinostat (LBH589), givinostat (ITF2357), belinostat (PXD101) and vorinostat (SAHA) (2–500nM is shown in Fig. 2A with data from 5 separate experiments. Dotted line indicates CD69 expression in untreated cells. CD69 expression in CD4+ central memory T-cells and T-cell subsets are shown in Figure 2B andC with data from one selected experiment. CD69 expression was analyzed in the CD4+ T-cell subsets: naive T-cells (CD45RA+, CD27+, CCR7+), central memory T-cells (CD45RA-, CD27+, CCR7+; Tcm), transitional memory T-cells (CD45RA-, CD27+, CCR7-; Ttm) and effector memory T-cells (CD45RA-, CD27-, CCR7-; Tem). Alterations in CCR5 expression are shown in Figure 2D, expressed as mean +/− SEM percentage of untreated cells with data from 4 separate experiments.

Figure 4. HIV-1 expression in latently infected primary CD4+ T cells. Panel A displays HIV-1 expression in latently infected resting CD4 T cells treated with panobinostat (LBH589), belinostat (PXD101), givinostat (ITF2357) and PMA/PHA for 72 h; mean +/− SEM RT/cpm. Panel B shows the percentage stimulation relative to the induction observed with PMA/PHA panobinostat (15 nM), belinostat (500 nM) and givinostat (500 nM); mean +/− SEM. Results are from two donors with each stimulation performed in duplicate.