HIV Articles  
The novel histone deacetylase inhibitors metacept-1 and metacept-3 potently increase HIV-1 transcription in latently infected cells (HIV eradication)
24 September 2009
Shehu-Xhilaga, Miranda; Rhodes, David; Wightman, Fiona; Liu, Hong B; Solomon, Ajantha; Saleh, Suha; Dear, Anthony E; Cameron, Paul U; Lewin, Sharon R aInfectious Diseases Unit, Alfred Hospital, Australia
bDepartment of Medicine, Monash University, Australia
cAvexa Ltd., Richmond, Australia
dAustralian Centre for Blood Diseases, Monash University, Australia
eCentre for Virology, Burnet Institute, Melbourne, Victoria, Australia.
We investigated the ability of several novel class I histone deacetylase inhibitors to activate HIV-1 transcription in latently infected cell lines. Oxamflatin, metacept-1 and metacept-3 induced high levels of HIV-1 transcription in latently infected T cell and monocytic cells lines, were potent inhibitors of histone deacetylase activity and caused preferential cell death in transcriptionally active cells. Although these compounds had potent in-vitro activity, their cytotoxicity may limit their use in patients.
Latent infection of resting central memory CD4+ T cells is considered to be the major reservoir leading to long-term HIV-1 persistence in individuals receiving antiretroviral therapy (ART) [1,2]. One potential strategy for eliminating this reservoir may be to stimulate HIV-1 transcription in latently infected CD4+ T cells in patients receiving ART [3-5]. Subsequent rounds of HIV-1 replication would be blocked by ART, and the transcriptionally active infected cell would then die as occurs in infected activated T-cells.
Activation and silencing of HIV-1 transcription in latently infected cells is governed by acetylation and methylation. Acetylation is counteracted by deacetylation, a process mediated by histone deacetylases (HDACs). HDACs can be subdivided into three distinct groups, known as class I, class II and class III [6]. Drugs that inhibit class I HDACs (trichostatin A, TSA; valproic acid; and suberoylanilide hydroxamic acid, SAHA) induce HIV-1 transcription in latently infected T-cell lines and latently infected resting CD4+ T-cells from HIV-1-infected patients [3,7-11]. We have recently synthesized two novel HDAC inhibitors (HDACi), metacept-1 (MCT-1) and metacept-3 (MCT-3) [12], which are conformational analogues of oxamflatin, a class I HDACi [13] that inhibits the deacetylation of H3 [14] and belongs to the same biochemical class as SAHA. We assessed the effect of these new HDACi on the induction of HIV-1 transcription in latently infected cell lines.
Treatment of the chronically infected T-cell line, ACH2, with oxamflatin, MCT-1 and MCT-3 led to the production of reverse transcriptase in a dose-dependent manner and at a higher level than valproic acid when used at the same concentration (Fig. 1ai,ii) [15-17]. The virus released in the supernatant of ACH2 cells following treatment with HDACi was fully infectious (Fig. 1aiii). An increase in HIV-1 production was also seen following treatment of the 8E5 T-cell line and the chronically infected monocytic cell line U1 (data not shown). We observed an increase in expressed green fluorescent protein (EGFP) following treatment of the J-Lat cell line [a Jurkat cell line stably infected with HIV-1 that contains a deletion of the Env and Nef genes and encodes for EGFP under the control of the HIV-1 long terminal repeat (LTR)] [18] confirming that HIV-1 production following treatment with these compounds occurred via activation of the HIV-1 LTR (Fig. 1aiii).
Treatment of tumor cell lines with most HDACi also induces apoptosis [19,20]. We quantified cytotoxicity of these novel HDACi with the mitochondrial tetrazolium (MTT) cell proliferation assay as previously described [21]. We found a significant correlation between the concentration of HDACi and MTT expression in the A301 T-cell line and primary resting CD4+ T-cells (data not shown). In contrast, minimal toxicity was observed following incubation with valproic acid at similar concentrations. The EC50 in the infected ACH2 and Jurkat cell line for oxamflatin, MCT-1 and MCT-3 was 3.5, 3.3 and 4.3 µmol/l, respectively.
The potency of these novel HDACi in inducing HIV-1 transcription was similar to other new HDACi such as SAHA [11,22] and TSA [23], both of which induce HIV-1 transcription in the micromolar range. SAHA also induces cell cycle arrest with an EC50 ranging from 0.75 µmol/l in ovarian cancer cells to 20 µmol/l in MCF-7 cells, a human breast cancer line [24,25]. The potent effect on HIV-1 transcription by oxamflatin, MCT-1 and MCT-3 was in contrast to that of valproic acid, which had only a modest effect on HIV-1 production and minimal effect on cell viability when used at the same concentration. Valproic acid did increase HIV-1 production but only when we used substantially higher concentrations of more than 0.5 mmol/l (data not shown), which was associated with significant toxicity (80% cell death at 1 mmol/l), as previously reported [22,26]. These data demonstrate the wide differences in HDACi, with respect to their relative potency in inducing HIV-1 transcription in vitro, cytotoxicity and likely clinical tolerability. Therefore, further evaluation of cytotoxicity in animal models will be a critical step in the clinical development of these compounds.
To determine whether cell death may preferentially occur in infected or transcriptionally active cells, we next investigated whether there was any correlation between HIV-1 production and cell death. Using the J-Lat cell line, we found that almost 100% of EGFP-positive (transcriptionally active) cells were also caspase 3-positive, regardless of the drug concentration (Fig. 1b). In contrast, in the EGFP-negative cells, the proportion of caspase 3-positive cells ranged from 10% to 90%, depending on the concentration of the drug used. Similar findings were observed when we measured expression of propidium iodide and annexin V (data not shown). When J-Lat cells were treated with phorbol-myristate acetate (PMA), we saw high levels of EGFP expression (>10% of total cells), but cell viability remained high (approximately 100% viability). These data suggested that transcriptionally active cells (EGFP-positive) were more likely to undergo cell death than transcriptionally inactive (EGFP-negative) cells. We do not believe that preferential cell death occurred secondary to HIV-1 replication as: (i) PMA treatment of J-Lat cells resulted in a significant increase in EGFP expression but minimal cell death and (ii) HDACi other than MCT-1 and MCT-3, such as SAHA and oxamflatin, are known to induce cell cycle arrest and apoptosis in a number of tumor-induced cell lines via activation of intrinsic death pathways unrelated to HIV-1 production [27].
We found a significant reduction in HDAC activity and an increase in acetylation of H3 in ACH2 cells treated with oxamflatin, MCT-1 and MCT-3 (Fig. 1ci-iii). The reduction of HDAC activity and acetylation of H3 were significantly less in cells treated with valproic acid at the same concentration (Fig. 1ci-iii). The data support and extend our previous findings for leukemia cells [12] and suggest that acetylation of H3 may be required to induce HIV-1 transcription from latently infected cells. However, it is possible that other histones may be involved. For example, oxamflatin, depsipeptide and valproic acid lead to acetylation of both H3 and H4 [28,29]. Further work is needed to fully define the specific histones and HDACs altered by MCT-1 and MCT-3 and their relationship with HIV-1 transcription.
We have demonstrated that oxamflatin, MCT-1 and MCT-3 inhibit HDAC activity, cause hyperacetylation of H3 and induce HIV-1 production in chronically infected cell lines. We also demonstrated significant cytotoxicity and preferential cell death in transcriptionally active cells. Although these compounds are potent inducers of HIV-1 transcription in vitro in latently infected cell lines, their narrow therapeutic range and significant cytotoxicity may limit their use in patients.
  iconpaperstack view older Articles   Back to Top