A New Small-Molecule Compound, Q308, Silences Latent HIV-1 Provirus by Suppressing Tat- and FACT-Mediated Transcription

ABSTRACT

Eliminating the latent HIV reservoir remains a difficult problem for creating an HIV functional cure or achieving remission. The “block-and-lock” strategy aims to steadily suppress transcription of the viral reservoir and lock the HIV promoter in deep latency using latency-promoting agents (LPAs). However, to date, most of the investigated LPA candidates are not available for clinical trials, and some of them exhibit immune-related adverse reactions. The discovery and development of new, active, and safe LPA candidates for an HIV cure are necessary to eliminate residual HIV-1 viremia through the block-and-lock strategy. In this study, we demonstrated that a new small-molecule compound, Q308, silenced the HIV-1 provirus by inhibiting Tat-mediated gene transcription and selectively downregulating the expression levels of the facilitated chromatin transcription (FACT) complex. Strikingly, Q308 induced the preferential apoptosis in HIV-1 latently infected cells, indicating that Q308 may reduce the size of the viral reservoir and thus further prevent viral rebound. These findings highlight that Q308 is a novel and safe anti-HIV-1 inhibitor candidate for a functional cure.

INTRODUCTION

AIDS is a highly disabling clinical syndrome caused by human immunodeficiency virus (HIV), which is a virus that progressively destroys the immune system. Combination antiretroviral therapy (cART) can effectively reduce the morbidity and mortality associated with HIV and shift AIDS from a fatal to a controllable chronic disease (1–4). Unfortunately, one of the limitations of cART is that it cannot completely eliminate HIV-1 due to the persistence of a latent HIV reservoir expressing low-level viral RNAs or proteins (5). Latent proviruses in the latent HIV reservoir are refractory to the immune response and antiviral drugs (5, 6), resulting in a rebound of the viral load after cART interruption (7). Therefore, approaches to clear the HIV reservoir are vital for achieving an HIV cure or remission (8, 9).

Given the undetectable level of the latent HIV-1 reservoir, there is considerable focus on HIV-1 gene transcription, which is an essential step in the viral life cycle and the only stage when viral genome amplification occurs (10). One functional cure strategy, called “block-and-lock,” has recently gained substantial attention. This therapeutic approach aims to permanently silence all proviruses even after treatment interruption by using latency-promoting agents (LPAs) to steadily suppress the transcription of the viral reservoir and lock the HIV promoter in deep latency (11). LPAs are used commonly to reduce residual viremia during therapy and block viral rebound upon treatment interruption (12, 13). Many viral and cellular regulatory proteins or factors related to HIV transcription and silencing are potential targets of future LPAs (14–16). To date, some small-molecule LPAs have been reported to achieve the long-term silencing of the latent provirus in infected cells by interfering with different factors of HIV transcription, which is consistent with the block-and-lock strategy (16–18).

The virally encoded HIV Tat protein, which is essential for efficient transcription and plays a central role in sustaining a high level of viral replication, has received considerable attention as a block-and-lock strategy (10, 19). Tat represents an interesting target because it can be expressed upon infection and has no cellular homolog. One of the most promising LPAs, didehydro-cortistatin A (dCA), is a Tat inhibitor that blocks Tat-dependent transcription and induces a repressive epigenetic landscape that inhibits latent HIV reactivation upon treatment interruption and prevents viral rebound (19). The small-molecule inhibitors LEDGINs interact with both HIV integrase (IN) and the cellular chromatin-tethering factor LEDGF/p75 and might be useful in a block-and-lock strategy by inhibiting viral integration and reactivation (20). An anticancer compound named curaxin CBL0100 blocks HIV replication and reactivation by inhibiting the facilitation of the chromatin transcription (FACT) complex (21). Moreover, heat shock protein 90 (HSP90) inhibitors might cause long-term remission and could potentially be developed into a functional cure for HIV infection (18). However, to date, most of the investigated LPAs are still in the early phase of development, and clinical data are lacking. As many new, active, and safe LPA candidates as possible are needed to facilitate the development of an effective therapy to eliminate residual HIV-1 viremia through the block-and-lock strategy.

Our group previously identified that a small-molecule bromodomain and extraterminal (BET) inhibitor, apabetalone (RVX-208), activated latent HIV-1 through the Tat-dependent P-TEFb pathway and had the potential to be developed as a latency-reversing agent (LRA) candidate. Apabetalone has been evaluated in phase III clinical trials in which patients with high-risk cardiovascular disorders, dyslipidemia, and low high-density lipoprotein (HDL) cholesterol were enrolled. However, due to the weak reactivation of apabetalone, we initially designed and synthesized a series of 2-substituted quinazoline-4(3H)-one derivatives based on the structure of apabetalone to screen new LRA candidates with high reactivation and low toxicity. Interestingly, we discovered accidentally that one derivative, i.e., 6,7,8-trimethoxy-2-(1H-pyrrol-2-yl)quinazolin-4(3H)-one (Q308), effectively suppresses the reactivation of latent HIV-1-infected cells by enhancing the proteasomal degradation of the viral Tat protein and inhibiting the activity of the FACT protein complex. Importantly, Q308 also induces preferential apoptosis of HIV-1 chronically infected cells, which indicates that Q308 may also be able to reduce the size of the viral reservoir and suppress viral rebound. These results suggest that Q308 is compatible with the block-and-lock strategy and might be developed as a potential anti-HIV-1 inhibitor candidate for a functional HIV-1 cure.

RESULTS

Q308 effectively suppresses the reactivation of HIV-1 latently and chronically infected cell lines.

A series of 2-substituted quinazoline-4(3H)-one derivatives were designed and synthesized from the structure of apabetalone to evaluate their potential reactivation of latent HIV-1 expression as LRAs in our laboratory. Unfortunately, no compounds induced a strong enough reactivation of HIV-1 latently infected cell lines. Interestingly, we found accidentally that some compounds effectively inhibited HIV-1 reactivation by suppressing the increase in the basal percentage of GFP-positive cells (GFP⁺%) in latently infected J-Lat 10.6 cells. The most active compound Q308 exhibited promising suppression of HIV-1 reactivation after phorbol 12-myristate 13-acetate (PMA)-induced latency reversal, thus maintaining latency in J-Lat 10.6 cells (Fig. 1a). Q308 at 10 μM reduced the GFP⁺% induced by PMA from 65.0% to 6.5%, which almost completely eliminated the reactivation efficiency of PMA and resulted in a GFP⁺% similar to the background level of 1.2%. The chemical structure of Q308 is shown in Fig. 1b. In addition to Q308, we noted that compounds 2-(furan-2-yl)-6,7,8-trimethoxyquinazolin-4(3H)-one (Q301) and 6-chloro-2-(1H-pyrrol-2-yl)quinazolin-4(3H)-one (Q508) could also modestly suppress HIV-1 reactivation (see Fig. S1 in the supplemental material). In contrast, the 3 compounds described above (Q201, Q206, and Q210), which maintained 5,7-dimethoxy quinazolin-4(3H)-one in a similar manner as apabetalone, could synergistically enhance PMA-induced HIV-1 activation (Fig. S1). Based on those results, 5,7-dimethoxyquinazolin-4(3H)-one might be a key pharmacophore for the reactivation of latent HIV-1. These data suggest that Q308 had a specific HIV-suppressive effect in J-Lat cells and was not simply a panassay interference compound.

FIG 1.

The small-molecule compound Q308 suppresses the reactivation of HIV-1 latently or chronically infected cells. (a) The inhibitory effects of a series of small-molecule compounds on PMA-induced HIV-1 reactivation in J-Lat 10.6 cells. Cells were cotreated with PMA (10 ng/ml) and individual compounds (10 μM) for 24 h (labeled 3 to 48). Cells and PMA only were used as negative and positive controls (labeled 1 and 2). GFP⁺% was measured by flow cytometry. (b) Chemical structure of Q308. Q308 effectively suppresses the reactivation of HIV-1 latency in PMA-induced J-Lat 10.6 (c), J-Lat A2 (d), ACH2 (e), and U1 (f) cells, and the corresponding cell viability examined by a CellTiter-Glo assay is shown. (g and h) Latently infected J-Lat 10.6 (g) and J-Lat A2 (h) cells were cultured with prostratin (0.5 μM), SAHA (0.5 μM), or CPI-203 (2 μM) in the presence/absence of Q308 (10 μM), and the corresponding cell viability examined by eBioscience fixable viability dye eFluor 780 is shown. GFP⁺% and HIV-1 p24 antigen levels in the supernatant were measured and analyzed by flow cytometry and ELISA. The data represent the means ± SDs of three independent experiments. Statistical significance was determined by a one-way ANOVA followed by Dunnett’s multiple comparison post hoc test (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Here, the basal level of HIV-1 transcription in the presence or absence of Q308 was first tested in J-Lat 10.6 cells by flow cytometry. As shown in Fig. S2 in the supplemental material, Q308 at all tested concentrations slightly inhibited the basal level of HIV-1 transcription. We further investigated the suppression effect of Q308 on PMA-induced latent HIV-1 reactivation in both J-Lat 10.6 and J-Lat A2 cells. As shown in Fig. 1c, compared with PMA-induced cells alone (GFP⁺%, 73.1%), the GFP⁺% of the PMA-treated J-Lat 10.6 cells was decreased from 24.4% to 10.7% when the concentration of Q308 was increased from 5 μM to 20 μM. In the J-Lat A2 cells, Q308 at concentrations ranging from 5 μM to 20 μM decreased the GFP⁺% induced by PMA from 34.2% to 7.5%, while the GFP⁺% of the PMA-treated cells alone was 56.7% (Fig. 1d). These results were further confirmed in two HIV-1 chronically infected cell lines, i.e., ACH2 and U1 cells. The production of the HIV-1 p24 antigen in the supernatants of the PMA-treated cells in the presence and absence of Q308 was detected by enzyme-linked immunosorbent assay (ELISA). The p24 levels in the supernatant were decreased from 15.9 ng/ml to 1.7 ng/ml in the PMA-treated ACH2 cells (Fig. 1e) and from 13.0 ng/ml to 3.8 ng/ml in the PMA-treated U1 cells (Fig. 1f) when the concentration of Q308 was increased from 5 μM to 20 μM. Furthermore, Q308 strongly inhibited HIV-1 reactivation induced by conventional chemical latency activators, including prostratin, SAHA, and CPI-203, in both J-Lat 10.6 (Fig. 1g) and J-Lat A2 cells (Fig. 1h). Q308 at the tested concentrations had no or low effects on cell viability in all used cell lines infected latently or chronically (Fig. 1c to h).

Q308 inhibits HIV-1 reactivation in primary CD4⁺ T cells.

We further examined the inhibition effect of Q308 on HIV-1 reactivation using the primary CD4⁺ T cell model of HIV-1 latency described previously by the Lewin group with slight modifications (22). This primary cell in vitro model was acquired by sorting nonpolarized cells, and then the cells were infected with HIV-1_{81A and NL4-3} viruses for 7 days (Fig. 2a). HIV-1 reactivation was induced by anti-CD3/CD28 antibodies for another 3 days, and the p24 levels of HIV-1 were tested by ELISA. The results showed that Q308 strongly inhibited both the anti-CD3/CD28-stimulated HIV-1 reactivation and the basal level of the transcription of HIV-1 proviruses in primary CD4⁺ T cells (Fig. 2b). As shown in Fig. 3a, Q308 at the tested concentrations had no or low effects on CD4⁺ T cellular toxicity even after treatment for 7 days by a CellTiter-Glo assay. In addition, Q308 demonstrated no cytotoxicity in human peripheral blood mononuclear cells (PBMCs) even at a concentration of 160 μM by using fixable viability dyes (FVDs) in flow cytometric analyses (Fig. 3b). Q308 did not affect the cell viability in both human CD4⁺ T cells and PBMCs at the working concentrations suppressing HIV-1 activity. Taken together, these results show that Q308 suppresses the reactivation of latent HIV-1-infected cells and could be developed as a potential anti-HIV-1 inhibitor candidate for an HIV-1 functional cure.

FIG 2.

Q308 inhibits HIV-1 reactivation in primary CD4⁺ T cells. (a) Scheme used to test the effect of Q308 on HIV-1 reactivation by using a primary CD4⁺ T cell model (Lewin model) of HIV-1 latency. (b) Primary CD4⁺ T cells with HIV-1 latency were treated with anti-CD3/CD28 antibodies (+) or mock treated with PBS (−) in the presence of 10 μM Q308 or 0.1% DMSO for 3 days. The p24 antigen levels in the supernatant were determined by ELISA and are presented relative to DMSO (−).

FIG 3.

Cell viability of Q308 on human T cells. Cell viability effect of Q308 on CD4⁺ T cells (a) and PBMCs (b) was measured by a CellTiter-Glo assay (a) or a eBioscience fixable viability dye eFluor 780 assay (b) and normalized to that in the control group (drug untreated). **, P < 0.01; ***, P < 0.001; ns, not significant.

Q308 does not induce the expression of T cell phenotypes and activation markers.

We used flow cytometry to evaluate the effects of Q308 on the expression of HIV-1 cell surface receptors and coreceptors (CD4 and CXCR4/CCR5) and activation markers associated with HIV infection and replication. The expression of the activation markers (CD25, CD38, CD69, and HLA-DR) was comparable between the solvent control and Q308 treatment, and Q308 (10 μM) was slightly downregulated after the treatment of 24 h (Fig. 4a) and 72 h (Fig. 4b). In addition, the stimulation by Q308 did not induce the expression of the HIV coreceptors CCR5 and CXCR4 and even slightly downregulated such expression in a dose-dependent manner, suggesting that this compound might not pose a risk of increasing PBMC susceptibility to HIV infection (Fig. 4c). However, Q308 could not increase or decrease susceptibility to HIV infection in PBMCs (Fig. 4d). It is worth noting that prostratin also could not decrease susceptibility to HIV infection even though it obviously downregulated the expression of a CD4 receptor/coreceptor. Some extensive animal studies might be more suitable to evaluate the infection of naive cells. To assess the impact of Q308 on T cells more broadly, the expression of an array of genes associated with T cell phenotypes and functions in PBMCs following Q308 treatment or no treatment was examined. The data showed that Q308 did not induce the expression of these genes (interleukin-6 [IL-6], IL-1β, tumor necrosis factor alpha [TNF-α], CXCL-10, IL-8, and interferon beta [IFN-β]) in PBMCs (Fig. 4e), indicating that Q308 did not induce global changes in human T cells.

FIG 4.

Q308 does not induce global activation in human T cells and downregulates the cell surface expression of HIV-1 receptor/coreceptors. Normal human PBMCs were treated with Q308 at the indicated concentrations for 24 h (a) and 72 h (b). Expression of activation markers (CD25, CD38, CD69, and HLA-DR) on T cells was detected by flow cytometry. (c) Q308 downregulates the expression of HIV-1 receptor/coreceptors. PBMCs were treated with the indicated concentrations of Q308 for 48 h. The surface expression profiles of CD4, CXCR4, and CCR5 on T cells were determined by flow cytometry. Cell-only (blank) and prostratin treatment were used as controls. (d) Isolated PBMCs pretreated with Q308 (10 μM) or prostratin (1 μM) for 48 h were infected with HIV-1_{81A and NL4-3} for another 2 h and further washed with PBS, replenished with complete medium, and cultured for 48 h. The p24 level in the supernatants was determined by ELISA. (e) Isolated PBMCs were treated with Q308 (10 μM), or the solvent control. After 48 h, cellular RNAs were extracted, and the expression of genes associated with the T cell phenotype was measured by qPCR. **, P < 0.01; ***, P < 0.001; ns, not significant.

Q308 exhibits inhibitory activities against HIV-1 infection in vitro.

To investigate the inhibitory activities of Q308, different HIV-1 infectious clones were used to infect TZM-bl cells. Our results demonstrated that Q308 exhibited potent inhibitory activities against HIV-1 infection in all tested HIV-1 infectious clones, including HIV-1_NL4-3 (X4), HIV-1_SF162 (R5), and HIV-1_{81A and NL4-3} (X4R5), with 50% inhibitory concentration (IC₅₀) values of 6.212 ± 0.952 μM, 2.265 ± 0.667 μM, and 2.610 ± 0.274 μM, respectively (Table 1). The positive control zidovudine (AZT; a reverse transcriptase inhibitor) had strong inhibitory activities against all tested HIV-1 infectious clones (Table 1). These results indicate that Q308 displayed potent anti-HIV activities without significant coreceptor tropism and that Q308 alone blocked both acute HIV-1 replication and the reactivation of latent HIV-1 proviruses in vitro.

TABLE 1.

The anti-HIV-1 activity of Q308 in vitro^a

Compound	Inhibitory activity (IC50) of:			Cytotoxicity (CC50) of:
	HIV-1NL4-3 (X4)	HIV-1SF162 (R5)	HIV-181A and NL4-3 (X4R5)	TZM-bl cells
Q308 (μM)	6.212 ± 0.952	2.265 ± 0.667	2.610 ± 0.274	68.627 ± 3.148
AZT (nM)	9.434 ± 3.064	1.964 ± 0.660	3.602 ± 0.402	>1,000

^aAll values are mean ± SD.

Q308 inhibits Tat-mediated gene transcription.

To explore the potential mechanism by which Q308 suppresses HIV-1 transcription, we first determined the effect of Q308 on HIV-1 transcription by real-time quantitative PCR (RT-qPCR) assay as described previously (23). Here, we used initiation primers (Ini) targeting 10 to 59 bp of HIV-1 transcription, proximal primers targeting 29 to 180 bp of HIV-1 transcription, intermediate primers targeting 836 to 1015 bp of HIV-1 transcription, distal primers targeting 2341 to 2433 bp of HIV-1 transcription, and GAPDH primers for normalization. The numbers denoting the amplification region of HIV-1_NL4-3 transcription is determined from the transcriptional start site. The primer sequences were designed as reported previously (24) and shown in Table 2. As shown in Fig. 5a, Q308 potently inhibited transcriptional initiation and elongation in a dose-dependent manner in PMA-treated J-Lat 10.6 cells. Interestingly, Q308 could not inhibit the elongation of HIV-1 transcripts at intermediate and distal sites in PMA-treated J-Lat A2 cells (Fig. 5b). Notably, both J-Lat 10.6 and J-Lat A2 cells are used widely in HIV latency lymphocyte cell models derived from human Jurkat T cells containing a latent and transcriptionally competent HIV provirus, while J-Lat A2 cells express the Tat protein under the control of HIV long terminal repeat (LTR). These data suggest that Q308 inhibits the initiation and elongation of HIV-1 LTR-driven transcription and that the effect of Q308 on transcriptional elongation might depend on the Tat protein.

TABLE 2.

RT-PCR primer sequences

Primer name	Direction	Sequence
GAPDH	Forward	CTCTGCTCCTCCTGTTCGAC
	Reverse	AGTTAAAAGCAGCCCTGGTGA
Tat	Forward	ATGGAGCCAGTAGATCCTAGACT
	Reverse	CGCTTCTTCCTGCCATAGGA
SUPT16H	Forward	CGG GCA GCA TTA CTT ACA GA
	Reverse	TTC AGT CAA TCG CCT CTT TG
SSRP1	Forward	ATT CAA CCC AGG TGA AGA GG
	Reverse	GTT TCC GCT TCT TCT CAT CC
IL-8	Forward	ATAAAGACATACTCCAAACCTTTCCAC
	Reverse	AAGCTTTACAATAATTTCTGTGTTGGC
CXCL-10	Forward	GTGGCATTCAAGGAGTAGCTC
	Reverse	GCCTTCGATTCTTGGATTCAG
IL-1β	Forward	CAACAGGCTGCTCTGGGATT
	Reverse	GGGCCATCAGCTTCAAAGAAC
IFN-β	Forward	TGCCTGGACCATAGTCAGAGTG
	Reverse	CAGTTTCGGAGGTAACCTGTAAGTC
TNF-α	Forward	GCCCATGTTGTAGCAAACCC
	Reverse	GGACCTGGGAGTAGATGAGGT
IL-6	Forward	AATAACCACCCCTGACCCAAC
	Reverse	TGCTACATTTGCCGAAGAGC
Initiation	Forward	GTT AGA CCA GAT CTG AGC CT
	Reverse	GTG GGT TCC CTA GTT AGC CA
Proximal	Forward	TGG GAG CTC TCT GGC TAA CT
	Reverse	TGC TAG AGA TTT TCC ACA CTG A
Intermediate	Forward	GTA ATA CCC ATG TTT TCA GCA TTA TC
	Reverse	TCT GGC CTG GTG CAA TAG G
Distal	Forward	GAG AAC TCA AGA TTT CTG GGA AG
	Reverse	AAA ATA TGC ATC GCC CAC AT

FIG 5.

Q308 inhibits HIV-1 Tat-mediated transcription. The inhibitory effect of Q308 on transcriptional initiation and elongation in PMA-induced J-Lat 10.6 (a) and J-Lat A2 (b) cells. The relative HIV-1 mRNA levels of HIV-1 initiated or elongated (proximal [Prox], intermediate [Int], and distal [Dis]) transcripts were measured by qPCR. (c and d) The overexpression efficiency of TZM-bl cells. TZM-bl cells stably transfected with a plasmid expressing Tat (+) or an empty vector (−) were determined by a WB analysis (c) and qPCR (d). The quantification of the Tat protein and mRNA levels is shown. The lines appearing in c are artifacts, which are potentially caused by the scanning quality of the images. (e) The inhibitory effect of Q308 on Tat-mediated LTR transcription. The luciferase activities in each treated group were normalized to those in the control group (normal TZM-bl cells). The data represent the means ± SDs of three independent experiments. **, P < 0.01; ***, P < 0.001; ns, not significant.

To confirm this hypothesis, TZM-bl cells stably transfected with a Tat expression vector (pEZ-Lv242-Tat-Flag) were used to demonstrate that the effect of Q308 on HIV-1 gene transcription is closely related to the Tat protein. The Tat overexpression efficiency of the TZM-bl cells on both protein levels and mRNA levels was shown in Fig. 5c and d, respectively. The luciferase activity in the Tat-overexpressing TZM-bl cells was approximately 41-fold higher than that in the negative-control TZM-bl cells (Fig. 5e). Our results show that Q308 induced a significant reduction (30- to 37-fold) in luciferase activity in the Tat-overexpressing TZM-bl cells compared with that in the untreated cells, suggesting that the inhibitory effect of Q308 on HIV-1 LTR-driven transcription might be attributed to the regulation of the expression of the Tat protein (Fig. 5e). Accordingly, Q308 at the tested concentrations could not affect HIV LTR-driven transcription in normal TZM-bl cells. As expected, a typical HIV LRA that allows Tat to bind p-TEFb, JQ1 (a bromodomain inhibitor), caused an almost 28-fold increase in reporter activity in Tat-overexpressing TZM-bl cells compared with that in the untreated cells (Fig. 5e).

Q308 promotes the degradation of Tat through the proteasomal pathway.

Since Q308 inhibits Tat-mediated gene transcription, we sought to determine whether it directly affects the expression of the Tat protein. Our results confirmed that the treatment with Q308 for 24 h and 48 h significantly decreased the mRNA expression level of Tat in both J-Lat 10.6 and ACH2 cells without an obvious dose dependency, suggesting that this finding might be a result of transcriptional suppression (see Fig. S3 in the supplemental material). Subsequently, we measured the effect of Q308 on endogenous Tat protein expression levels in ACH2 cells by a Western blot (WB) analysis. As shown in Fig. 6a, Q308 (10 μM) markedly decreased the expression level of the Tat protein, whereas the expression levels of the BRD4 and Cyclin T1 proteins were unchanged. We further confirmed the above conclusion by detecting the expression levels in HEK293T cells after transient transfection with 0.5 μg of the Tat-expressing plasmid (cytomegalovirus promoter [CMV]). To exclude the possibility that Q308 acts against the transcriptional activity of the CMV promoter, we also constructed plasmids expressing the Gag protein as a control. After the transfection, the expression level of Gag was not reduced in the presence of Q308 (Fig. 6c). However, this compound led to a dramatic dose-dependent decrease in the Tat protein expression level (Fig. 6b) without changes in the Tat mRNA abundance (Fig. 6d), suggesting that Q308 does not affect CMV promoter activity and specifically reduces the Tat protein levels.

FIG 6.

Q308 promotes the degradation of Tat by the proteasomal pathway. The effects of Q308 on the expression of endogenous (a) and exogenous (b) Tat protein. (c) The effects of Q308 on Gag-overexpressing HEK293T cells transfected with the pcDNA3.1-Gag-Flag plasmid. The protein expression level of Tat was determined by a WB analysis. The expression level of β-actin was used as a loading control. (d) The effect of Q308 on Tat mRNA levels. HEK293T cells were transfected with pEZ-Lv242-Tat-Flag in the presence of increasing concentrations of Q308. The cells were harvested 48 h posttransfection. The steady-state mRNA levels of Tat and GAPDH were examined using a semiquantitative reverse transcription-PCR analysis. Tat-overexpressing HEK293T cells were treated in the presence or absence of Q308 at 10 μM for 21 h (e) or 12 h (f). Before being harvested, the cells were treated with 20 μM MG-132 and cultured for an additional 3 h or treated with bafilomycin A1 (20 μM) or Z-VAD-FMK (50 μM) for an additional 12 h. The protein expression levels were determined by a WB analysis. The data represent the means ± SDs of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Generally, most cellular protein degradation occurs via both lysosomal and/or proteasome pathways. As reported in previous studies, the unassociated Tat cellular protein is prone to degradation through the proteasomal pathway in a ubiquitin-independent manner (25). Here, we sought to determine whether the effect of Q308 on Tat protein expression might influence the degradation of the Tat protein by these pathways. HEK293T cells transiently transfected with the pEZ-Lv242-Tat plasmid and treated with Q308 in the presence or absence of the proteasomal inhibitor MG132 were probed by WB analysis. Our results show that Q308 alone decreased the level of the Tat protein by approximately 70% compared with the solvent control, while the effect was partially inhibited by MG132 with a 40% decrease in the level of the Tat protein (Fig. 6e). However, the Q308-induced degradation of the Tat protein was unaffected by the bafilomycin A1 (a lysosomal inhibitor) or Z-VAD-FMK (a pancaspase inhibitor) treatment (Fig. 6f). It is worth mentioning that bafilomycin A1 (channel 3) slightly increased the basal level of the Tat protein, demonstrating that lysosomes might also influence Tat degradation. However, bafilomycin A1 did not significantly block the effect of Q308 on Tat protein degradation. Furthermore, we analyzed the gray values of all channels. We found that bafilomycin A1 alone increased the Tat level from 100% (basal level, blank control) to 142%, while combined with Q308, it decreased the Tat level to 26%, which is similar to the effect of Q308 alone (18%) (Fig. 6f). These results indicate that Q308 promoted the degradation of the Tat protein mainly through the proteasomal pathway rather than the lysosomal or caspase-dependent apoptosis pathway.

Q308 inhibits HIV-1 reactivation through the FACT signaling pathway.

Previous research shows that the protein components of the FACT complex, i.e., SUPT16H and SSRP1, are significant host factors that negatively regulate HIV-1 replication and that the complex is associated with Tat (26). Therefore, SUPT16H and SSRP1 may play an important role in suppressing HIV-1 transcription and promoting viral latency and, therefore, could serve as promising targets in the development of novel LPAs. Here, we tested the effect of Q308 on the expression of FACT and histone deacetylase (HDAC), as we know that FACT affects transcriptional elongation by interacting with histones in the histone deacetylase signaling pathway. However, Q308 reduced the expression of all protein components of the FACT complex, including SUPT16H and SSRP1, in a dose-dependent manner, but it did not affect histone acetylation (Fig. 7a). Q308 (10 μM) also reduced the protein expression of SUPT16H and SSRP1 in a time-dependent manner (Fig. 7b). To further investigate whether this compound decreases the protein levels of the FACT complex by reducing its mRNA stability, we assessed the steady-state mRNA level of the FACT complex in the presence or absence of Q308 using a reverse-transcription PCR analysis. Notably, the treatment with Q308 did not significantly alter the abundance of FACT mRNA (Fig. 7c). Because the FACT complex could interact with Tat, we speculated whether its downregulation by Q308 also involves the proteasomal pathway. The WB results showed that MG-132 could rescue the decreased expression of the FACT complex and Tat protein caused by Q308 (Fig. 7d). Therefore, these results indicate that Q308 might specifically reduce the FACT protein steady-state levels by enhancing the proteasomal degradation of the FACT complex.

FIG 7.

Q308 suppresses HIV-1 transcription by reducing the expression of the FACT complex. Q308 reduced the expression of SUPT16H and SSRP1 in a dose-dependent manner (a) and a time-dependent manner (b) in J-Lat 10.6 cells. The expression levels of SUPT16H, SSRP1, acetyl-H4K8, and acetyl-H3K14 were determined by a WB analysis. The expression level of GAPDH was used as a loading control. The SUPT16H and SSRP1 expression levels were quantified. (c) The effect of Q308 on the FACT mRNA levels in J-Lat 10.6 cells at 48 h. (d) The inhibitory effect of Q308 on the expression of the FACT protein in the presence or absence of MG-132. J-Lat 10.6 cells were treated with/without 10 μM Q308 for an additional 21 h. Before being harvested, the cells were treated with 20 μM MG-132 and cultured for an additional 3 h. The protein expression levels were determined by WB analysis. The effect of Q308 on the occupancy of SUPT16H (e) and p-Rpb1 (f) at nuc-1. Mock-treated and PMA-stimulated J-Lat 10.6 cells were subjected to ChIP-qPCR assays using anti-SUPT16H and anti-p-Rpb1 antibodies or normal rabbit IgG. PCR primers specific for the LTR promoter were used to amplify DNA isolated from the immunoprecipitated chromatin as described in the Materials Methods. Each ChIP assay was repeated three times to confirm the reproducibility of the results. Real-time quantitation of the fold change relative to that in the mock or PMA group is shown. **, P < 0.01; ***, P < 0.001; ns, not significant.

To confirm the suppressive effect of Q308 on HIV-1 through the FACT signaling pathway, the effect of Q308 on the occupancy of SUPT16H at nuc-1 in both mock-treated and PMA-stimulated J-Lat 10.6 cells was determined by a chromatin immunoprecipitation-quantitative PCR (ChIP-qPCR) assay. Our results show that the Q308 treatment (10 μM) significantly decreased SUPT16H occupancy at nuc-1 in PMA-stimulated latent HIV-1-infected cells (Fig. 7e). There were no obvious changes in SUPT16H occupancy at nuc-1 in the presence or absence of Q308 (10 μM) in the mock-treated J-Lat 10.6 cells. Since the loss of the FACT complex interferes with elongation driven by RNA polymerase II, we further determined the effect of Q308 on the occupancy of p-Rpb1 at nuc-1 in PMA-stimulated J-Lat 10.6 cells using a ChIP-qPCR assay. Similarly, the Q308 treatment (10 μM) significantly reduced the association between p-Rpb1 and the HIV-1 LTR promoter stimulated by PMA (Fig. 7f). Altogether, these results demonstrate that Q308 inhibits HIV-1 reactivation through the FACT signaling pathway.

Q308 preferentially induces apoptosis in latent HIV-1 chronically infected cell lines.

With the expansion and/or repopulation of the latent HIV reservoir, recent efforts have focused on developing novel therapies to accelerate the decay of the latent HIV reservoir, which might be beneficial for the block-and-lock strategy. To explore whether Q308 could reduce the size of the latent HIV-1 reservoir, apoptosis in HIV-1 chronically infected cells was assessed by flow cytometry. Here, we chose CEM7/U937 cells and their subclones ACH2 cells (CEM7 cells with an integrated proviral copy of LAV)/U1 (U937 cells chronically infected with HIV-1). An anticancer agent, obatoclax, was used as a positive control. Intriguingly, the Q308 treatment significantly induced apoptosis in the ACH2 and U1 cells in a dose-dependent manner (Fig. 8a and b). However, the Q308 treatment had only a background effect on apoptosis in the CEM7 and U937 cells at concentrations up to 20 μM. In comparison, obatoclax (1 μM) induced apoptosis in both cell types and showed no selectivity. Furthermore, we evaluated the induction of apoptosis by Q308 in CD8⁺ T cells. However, Q308 at 10 μM did not induce apoptosis of CD8⁺ T cells and even decreased the background level of cell apoptosis (Fig. 8c and d). In addition, as shown in Fig. 1c to f and Fig. 3, the cytotoxicity of Q308 in latently infected cells was higher than that of Q308 in HIV-1 target cells. These results indicate that the induction of cell death by Q308 is more specific for HIV-latent cells than uninfected cells, which may help eliminate the latent HIV-1 reservoir.

FIG 8.

Q308 preferentially induces apoptosis in chronically infected HIV-1 cells. (a) Summary of apoptotic effects of Q308 in chronically infected HIV-1 cells are presented as histograms. (b) The percentage of apoptotic cells as determined by annexin V staining. ACH2 and U1 cells or their parental CEM7 and U937 cells were treated with the indicated concentrations of Q308 for 48 h. Obatoclax at 1 μM was used as a positive control. The data in all panels are presented as the means ± SDs. *, P < 0.05; **, P < 0.01. (c) Q308 does not induce apoptosis in CD8⁺ T cells. PBMCs were treated without or with Q308 (10 μM) for 48 h and then labeled with anti-APC-CD3, anti-BV786-CD8, annexin V-FITC, and propidium iodide (PI) to assess apoptosis in CD8⁺ T cells. The effects of Q308 on early and late apoptosis are presented as histograms in d.

DISCUSSION

Although cART effectively suppresses detectable viremia, there is still a barrier to cure HIV-1 due to the existence of a latent viral reservoir. A strategy named “shock-and-kill” was proposed to reactivate HIV-1 gene transcription in latently infected cells using LRAs and then promote their elimination by viral cytopathic effects and/or HIV-1-specific cytolytic T lymphocytes (CTLs) (27). Despite extensive efforts targeting the shock-and-kill strategy, the currently available LRA candidates have failed to reduce the size of the latent reservoir and have low efficacy, high toxicity, and/or poor clinical outcomes (12, 28).

The limited success of the shock-and-kill strategy has caused scientists to look for new strategies to achieve an HIV cure. The so-called block-and-lock strategy was proposed to achieve the long-term silencing of the latent provirus in infected cells and thus prevent viral rebound (9, 29, 30). Although this strategy was proposed only recently, numerous studies have shown that HIV transcription can be significantly downregulated. The discovery and development of new and novel LPAs with high efficacy and safety are needed urgently.

Our group focused previously on the development of new small-molecule LRA candidates to advance HIV-1 eradication (23, 31). Accidentally, we found that a 2-substituted quinazoline-4(3H)-one derivative, Q308, could inhibit the reactivation caused by cellular self-activation in HIV-1 latently infected cells. Further studies confirmed that Q308 effectively inhibited HIV-1 reactivation and induced HIV-1 suppression in a dose-dependent manner in different PMA-stimulated latently or chronically infected cells (Fig. 1c to f). In addition, we found that Q308 significantly inhibited HIV-1 reactivation induced by multiple types of well-known LRAs (Fig. 1g and h). Importantly, the same effect of Q308 on HIV-1 suppression was detected in the primary CD4⁺ T cell model of HIV-1 latency (Fig. 2b). These results indicate that Q308 might be developed into a potential anti-HIV-1 inhibitor because of its effect on HIV-1 transcription to support the block-and-lock strategy.

Many viral and cellular proteins involved in HIV transcription and silencing may represent potential targets of LPAs. HIV is generally silenced at the transcriptional level, which is predominantly directed by a promoter in the 5′ LTR of the integrated provirus (30). Our results clearly show that Q308 significantly decreased the viral mRNA levels of HIV-1 LTR-driven transcription in J-Lat 10.6 cells, while it did not inhibit the elongation of HIV-1 transcription at intermediate or distal sites in Tat-dependent latency in J-Lat A2 cells transfected with the LTR-Tat-internal ribosome entry site (IRES)-green fluorescent protein (GFP) plasmid (Fig. 5a and b). During HIV infection, the Tat protein binds the transactivation response (TAR) RNA stem-loop in the HIV LTR promoter, thereby stimulating transcriptional elongation. Further results confirmed that Q308 inhibited Tat-transactivated LTR-driven transcription in Tat-overexpressing TZM-bl cells (Fig. 5e). Tat-dependent transcriptional inhibition by Q308 might play a key role in suppressing HIV reactivation upon treatment interruption. Notably, the most advanced LPA candidate, dCA, is a Tat inhibitor that silences HIV transcription. We found that Q308 preferentially decreased the expression of Tat at both endogenous and exogenous protein levels (Fig. 6a and b). Generally, most cellular protein degradation occurs via both lysosomal and/or proteasome pathways. A recent study showed preliminarily that the degradation of viral Tat protein is achieved through the proteasomal pathway. Our results show that Q308 reduced the expression of the Tat protein and decreased Tat-mediated LTR promoter transactivation mainly through the proteasomal pathway (Fig. 6e) rather than the lysosomal and caspase-dependent apoptosis pathways (Fig. 6f).

Here, we found that Q308 inhibits not only transcriptional elongation but also transcriptional initiation. Therefore, Q308 likely has other target(s) in addition to the Tat protein during HIV-1 suppression. Understanding the host factors that suppress HIV-1 transcription and promote its latency could help identify new gene targets for block-and-lock therapy. The two protein components of the FACT complex SUPT16H and SSRP1 are considered essential for both HIV transcriptional initiation and elongation and increase HIV-1 intracellular replication (21, 32). The nucleosome is the major target of FACT activity. The FACT protein affects the assembly and disassembly of nucleosomes by interacting with histones H2A/H2B during transcriptional initiation and elongation (26, 32). The FACT protein is a transcriptional regulator that binds all core histones or at least the oligomers of H2A/H2B and H3/H4.

In the current study, the results indicate that Q308 specifically reduced the expression of the two protein components of the FACT complex with no effects on acetylated histones (Fig. 7a and b) and decreased the occupancy of SUPT16H at nuc-1 induced by PMA (Fig. 7e). The suppressive effect of Q308 on transcription might be associated mainly with the FACT pathway rather than the HDAC pathway. Furthermore, we investigated whether Q308 decreases the protein levels of the FACT complex by reducing its mRNA stability, but the results did not support this hypothesis (Fig. 7c). Based on the interaction between Tat and FACT (26), we hypothesized that the FACT complex might be a Tat-associated cellular protein and, thus, degraded along with Tat as a complex dependent on the proteasomal pathway. As expected, MG-132 rescued the decreased the FACT complex and Tat protein expression caused by Q308, implying that the degradation of the FACT complex was also associated with the proteasome pathway (Fig. 7d). The associated Tat protein can be degraded as a whole complex with other proteins, such as CycT1-U7, in an ubiquitin-dependent proteasome pathway (33). Thus, although we demonstrated that the degradation of the FACT complex is associated with the proteasome pathway, it could also be interesting to investigate whether it is related to ubiquitination in the future. The FACT protein complex is a well-studied histone chaperone that removes the H2A-H2B dimer to facilitate RNA polymerase II-driven transcription. Q308 efficiently decreased the occupancy of phosphorylated RNA polymerase II largest subunit (p-Rpb1) at nuc-1 induced by PMA (Fig. 7e). Collectively, these results indicate that Q308 suppressed HIV-1 transcription by interfering with FACT activity and thus inhibited transcriptional elongation mediated by RNA polymerase II.

In general, interventions in HIV-1 transcription also affect HIV-1 replication. As expected, Q308 exhibited good antiviral activity against different tested HIV-1 strains in a dose-dependent manner with low cytotoxicity in HIV-1 target cells (Table 1). Q308 at the tested concentrations (0 to 20 μM) had no or low effects on CD4⁺ T cellular toxicity even after treatment for 7 days (Fig. 3a). In addition, notably, Q308 could not induce global T-cell activation (Fig. 4a and b), the expression of HIV entry receptors (Fig. 4c), or cytokine release (Fig. 4e). To better evaluate its safety, some extensive animal studies might be necessary for evaluating cytokine release and immune activation upon Q308 treatment in the future. Notably, LPAs maintain latent HIV-1 silencing, and cARTs are crucial for reducing the viral load. The combination of LPAs with cARTs might be useful for the long-term silencing of all HIV-1 variants in a future block-and-lock functional cure strategy. Thus, further studies are needed to evaluate the comprehensive capacity of the combination of Q308 with cARTs in this strategy.

As mentioned above, the block-and-lock strategy seems to delay the emergence of viral rebound in concert with cARTs and maintain the repression of the viral reservoir (29). However, the size of the viral reservoir cannot be reduced and could even possibly be increased. In general, better diagnostic tests are needed to evaluate the size of the HIV reservoir. The expansion of the pool of HIV-infected cells may be a major reason for the slow decay of the latent HIV reservoir. LPAs may contribute to delaying or halting the replenishment of the latent viral reservoir, but the pool of HIV-infected cells inevitably expands under the influence of interleukin-7 (IL-7) or antigenic stimulation (34). Thus, shrinking and clearing the latent reservoir are still particularly important for an HIV-1 cure. The induction of HIV-1-specific apoptosis through the processing of viral proteins or regulator protein B-cell lymphoma 2 (Bcl-2) using their inhibitors may be an encouraging strategy for depleting the viral reservoir (35, 36). Importantly, Cummins et al. (37) proved that ixazomib, a proteasome inhibitor, blocks the degradation of the Casp8p41-Bcl2 complex, increases the abundance of Casp8p41, induces preferential apoptosis in patient-derived CD4⁺ T cells, and decreases the viral DNA load ex vivo. Ixazomib is the first oral proteasome inhibitor tested in a phase I clinical trial in patients receiving cART (ClinicalTrials.gov registration no. NCT02946047). Furthermore, similar effects have also been demonstrated by using the Bcl-2 antagonist venetoclax, which prevents Casp8p41 from binding to Bcl-2 and, thus, allows Casp8p41 to bind Bak and cause infected cells to die (34). These findings provide powerful support for the applicability and necessity of HIV reservoir reduction. Surprisingly, our results show that Q308 preferentially induced apoptosis in HIV-1 chronically infected cells (Fig. 8a and b), which completely differs from the effect on CD8⁺ T cells (Fig. 8c and d) and may be conducive to preventing the expansion of latent pools and further minimalizing latent reservoirs. However, both ACH2 and U1 cells are chronically HIV-infected cell lines, and it is necessary to determine whether Q308 can actually reduce the size of HIV reservoirs by using a primary CD4⁺ T cell model of latency and animal models (e.g., HIV-infected humanized mice or SIV-infected macaques) in the future.

In addition to selective apoptosis in the postdrug effect of Q308, the long negative transcriptional regulation of post-Q308 treatment on latently infected cells was also meaningful to investigate; otherwise, a block-and-lock drug that requires continuous dosing would essentially imply substituting one class of drugs with another. The small molecules ZL0580 and dCA, which induce epigenetic HIV-1 suppression in vitro and ex vivo, are superior to other LPAs in terms of the block-and-lock approach, even though none of them can eventually end viral rebound (16, 19). Thus, additional experiments evaluating the kinetics of Q308-induced transcriptional blockade should be performed in the future, although none of the investigated LPAs led to complete, long-term viral suppression in all cell and/or animal models. Furthermore, in the future, it will also be meaningful to design and synthesize more potent small-molecule compounds by the modification of the Q308 structure to overcome the above problems.

In this study, we provided strong evidence suggesting that Q308 is a promising anti-HIV-1 inhibitor that can silence the provirus and inhibit both transcriptional initiation and elongation by interfering with both FACT and Tat activity associated with the proteasome pathway. In addition, Q308 appears to be a promising reasonable choice for preventing the rebounding of the latent HIV-1 reservoir in the context of cART and inducing the preferential apoptosis in HIV-1 latently infected cells. These results suggest that Q308 is a promising anti-HIV-1 inhibitor candidate that could be developed for use in a block-and-lock approach.

MATERIALS AND METHODS

Cell lines and culture.

The latently infected (J-Lat A2 and J-Lat 10.6) and chronically infected (U1 and ACH2) cell lines were kindly provided by Shibo Jiang of Fudan University, China. The U937, CEM7, HEK293T, HeLa, and TZM-bl cells were purchased from ATCC (Manassas, VA). All cell lines were cultured in RPMI 1640 medium or Dulbecco’s modified Eagle’s medium (DMEM) (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin (Invitrogen, Carlsbad, CA, USA) in a 37°C incubator containing 5% CO₂.

Materials.

Phorbol 12-myristate 13-acetate (PMA) and prostratin were purchased from Sigma-Aldrich (St. Louis, MO, USA). CPI-203 was purchased from Selleck (Boston, MA, USA). JQ1 and SAHA were purchased from MedChemExpress (MCE; Vantaa, Finland), and TNF-α was purchased from R&D Systems (Minneapolis, MN, USA). MG-132 was purchased from Topscience (Shanghai, China). The glyceraldehyde-3-phosphate dehydrogenase (GAPDH), acetyl-H4K8, and acetyl-H3K14 antibodies were purchased from Cell Signaling Technology (CST; Beverly, MA, USA). The specific antibodies against Tat, SSRP1, and SUPT16H were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The human anti-CD4, anti-CCR5, and anti-CXCR4 antibodies were obtained from BD Biosciences (San Jose, CA, USA).

Synthesis.

A series of 2-substituted quinazoline-4(3H)-one derivatives, including Q308, was designed and synthesized to screen potential LPAs and latency-reversing agents (LRAs) at the Chemical Biology Laboratory of Southern Medical University. These chemicals were dissolved in dimethyl sulfoxide (DMSO) at a concentration of 20 mM and stored at −20°C. The detailed descriptions of the synthesis of Q308 are available in the supplemental material.

Flow cytometry.

HIV-1 latently infected J-Lat A2 and J-Lat 10.6 cells (5 × 10⁵ cells) were seeded into the wells of a 48-well plate and cotreated with PMA (10 ng/ml) and the indicated concentrations of compounds for 24 h. Cells treated with DMSO or PMA (10 ng/ml) alone were used as a blank or negative control. After cells were washed three times with PBS, green fluorescent protein (GFP) expression representing the reactivation of latent HIV was measured by a flow cytometric analysis using a BD FACS Canto II flow cytometer (San Jose, CA, USA). The percentage of GFP-positive cells within the entire population was analyzed using FlowJo software (Treestar, San Carlos, CA, USA).

HIV-1 p24 antigen assay.

ACH2 and U1 cells (5 × 10⁵ cells) were seeded into 48-well plates and incubated with different concentrations of Q308 and 10 ng/ml PMA for 48 h. After the cells were centrifuged, the supernatants were collected and mixed with an equal volume of 5% Triton X-100 overnight. The viral lysates were analyzed to determine the p24 antigen levels in the supernatant by ELISA as described previously (23).

Cell viability assay.

The cell viability of Q308 on all tested cells was evaluated using the CellTiter-Glo luminescent cell viability assay (Promega) or eBioscience fixable viability dye (FVD) eFluor 780 (Invitrogen). A CellTiter-Glo assay determines the number of viable cells in a well based on the quantification of ATP, which is an indicator of metabolically viable cells. FVD can be used to irreversibly label dead cells prior to cryopreservation, fixation, and/or permeabilization procedures and has been used in flow cytometric analyses.

HIV-1 latency model using primary CD4⁺ T cells.

To determine the effect of Q308 on the reactivation of latent HIV-1 in primary cells, we utilized the Lewin model, with slight modifications. Briefly, human peripheral blood mononuclear cells (PBMCs) from healthy donors were isolated by Ficoll density gradient centrifugation using Histopaque-1077 (Sigma-Aldrich), and then primary CD4⁺ T cells were isolated from the PBMCs by negative selection using a CD4⁺ T cell isolation kit (Miltenyi Biotec). The CD4⁺ T cells were activated with 29 nM CCL19 for 3 days and then spinoculated with HIV-1_{81A and NL4-3} (10 ng of p24 per 10⁶ cells) at 2,000 × g for 2 h at 37°C. Then, the cells were incubated in complete media containing IL-2 (30 UI/ml) for 7 days to establish HIV-1 latency. Finally, the cells were stimulated with anti-CD3/CD28 antibodies (1 μg/ml of each) or PBS in the presence of Q308 (10 μM) for another 3 days. The level of HIV-1 reactivation was evaluated by ELISA.

Detection of T cell phenotypes and activation markers.

To test the effects of Q308 on the expression of T cell phenotypes and activation markers, human PBMCs isolated from healthy individuals (1 × 10⁶ per well) were seeded in 6-well plates and incubated with Q308 for 48 h. Then, the cells were incubated with CD4-fluorescein isothiocyanate (FITC)/CXCR4-APC, CD4-FITC/CCR5-APC, CD69-FITC, CD38-FITC, CD25-FITC, and HLA-DR-FITC antibodies (BD Biosciences) at 4°C for an additional 30 min in the dark and analyzed by flow cytometry. Prostratin (1 μM) was used as a positive control.

Anti-HIV-1 activity.

HIV-1 target TZM-bl cells were seeded at a density of 2 × 10⁴ cells per well in 96-well cell culture plates and then incubated with 100 50% tissue culture infective doses (TCID₅₀) of HIV-1 virus in the presence or absence of Q308 at graded concentrations at 37°C. At 48 h postinfection, the supernatants were removed, the cells were lysed, and then the luciferase activity in the cells was determined using a ONE-Glo luciferase assay system (Promega, Madison, WI, USA). The inhibitory concentration achieving 50% inhibition (IC₅₀) was calculated by Calcusyn software, which was kindly provided by T. C. Chou (Sloan-Kettering Cancer Center, New York, NY). AZT was chosen as a positive control.

Real-time quantitative PCR (RT-qPCR) analysis.

The effects of Q308 on gene transcription and gene expression were analyzed by RT-qPCR as described previously (23). Briefly, J-Lat 10.6 and J-Lat A2 cells were treated with PMA (10 ng/ml) in either the presence or absence of graded concentrations of Q308 for 24 h to analyze the effects on gene transcription. In addition, J-Lat 10.6 cells or PBMCs were incubated with different concentrations of Q308 for 48 h to test the effects of gene expression on FACT or T cell phenotypes. After incubation, the relative gene expression levels were assessed by RT-qPCR. In brief, the cells were collected, the total RNA was extracted using TRIzol (Invitrogen), and cDNA was synthesized by a PrimeScript RT reagent kit (TaKaRa, Japan). RT-qPCR was performed using SYBR select master mix (Applied Biosystems, Foster City, CA) on a LightCycler 480 system (Roche, Switzerland). The primer sequences are shown in Table 2. GAPDH primers were used as a reference. All of the primers were synthesized by Sangon Biotech (Shanghai, China).

Western blot (WB) analysis.

After the compound treatments, the cells were collected and lysed in radioimmunoprecipitation assay (RIPA) buffer (containing protease and phosphatase inhibitors). Approximately 50 to 100 μg of cell lysates was separated by SDS-PAGE, transferred to a polyvinylidene difluoride (PVDF) membrane (Roche, Indianapolis, IN, USA), and blocked with defatted milk for 1.5 h. Then, the membrane was incubated with primary antibodies and horseradish peroxidase-conjugated secondary antibodies. Subsequently, enhanced chemiluminescence (ECL) substrates (Cell Signaling Technology) were added, and the immunoreactive signals were detected by an autoradiographic film.

Cell transfection.

The ectopic expression of HIV-1 Tat was obtained by cloning HIV-1 Tat cDNA into the pEZ-Lv242 lentivector. For the lentivirus generation and infection, HEK293T cells were seeded into a 6-well plate at 2 × 10⁵ cells/well for 24 h and then cotransfected with the glycoprotein-expression vector vesicular stomatitis virus G (VSV-G; 500 ng), lentiviral packaging construct psPAX2 (500 ng), and pEZ-Lv242-Tat-Flag-expressing lentiviral construct (1 μg) using polyjet (Invitrogen) according to the manufacturer’s instructions. At 48 h posttransfection, the viral supernatants were collected and centrifuged at 1,500 rpm for 15 min. Then, TZM-bl cells were spin-infected with pEZ-Lv242-Tat-Flag-expressing lentivirus, cultured, and selected by puromycin (Sigma-Aldrich). Three days postselection, the medium of the infected cells was changed to fresh medium, and the cells were cultured for another 2 to 7 days. The overexpression efficiency of the TZM-bl cells was authenticated by a WB analysis. The TZM-bl cells were further treated with the indicated compounds for another 48 h, and luciferase activity was measured as described previously (31).

For the transient transfection, HEK293T cells were plated at 2 × 10⁵ cells/well in a 12-well culture plate for 24 h before the transfection, and then fully adhered cells were transfected with pEZ-Lv242-Tat-Flag, pcDNA3.1-Gag-Flag, or an empty vector plasmid using polyjet (Invitrogen) as described above. Twenty-four hours posttransfection, the cells were mock treated or incubated with the indicated compounds for an additional 24 h, and then RT-qPCR or a WB analysis was performed.

Annexin V labeling.

To test the preferential proapoptotic effect of Q308 on latent HIV-1 cells, chronically infected HIV-1 cells (U1 and ACH2) and their parental cells (U937 and CEM7) were seeded in 6-well plates at a density of 1 × 10⁶ cells per well and incubated with Q308 (5, 10, and 20 μM) at 37°C for 48 h. After three washes with PBS, apoptosis was detected using a FITC annexin V apoptosis detection kit (BD Biosciences) according to the manufacturer’s instructions. The extent of apoptosis is represented by the percentage of annexin V-positive cells. The antitumor drug obatoclax (1 μM) was chosen as a positive control.

Chromatin immunoprecipitation (ChIP).

The ChIP assays were performed as described previously (31). Briefly, J-Lat 10.6 cells (1 × 10⁷ cells) were treated with Q308 for 24 h. Then, the cells were fixed with 1% formaldehyde for 10 min at room temperature, and cross-linking was quenched by the addition of a 125 mM glycine-PBS solution for an additional 10 min. The cross-linked cells were resuspended in SDS lysis buffer and sonicated to obtain DNA fragments of 200 to 500 bp. The DNA fragments were diluted with 1× ChIP dilution buffer and then incubated with immunoglobulin G (IgG), SUPT16H, and p-Rpb1 (CST) antibodies at 4°C overnight; subsequently, the immune complexes were retrieved by incubation with protein A/G beads for another 8 h. The beads were collected and washed with each of the following buffers in sequence: low salt buffer, high salt buffer, and LiCl buffer. After two wash steps with 1× Tris-EDTA (TE) buffer, chromatin was eluted from the protein A/G magnetic beads. DNA was extracted, and RT-qPCR was performed as described above with forward (5′-AGCTTGCTACAAGGGACTTTCC-3′) and reverse (5′-GTGGGTTCCCTAGTTAGCCAGAG-3′) primers. The results of the fragments obtained after incubation with different antibodies were normalized against input DNA.

Statistical analysis.

The experimental data are representative of at least three independent experiments and are expressed as the means ± SDs. The statistical analysis of the differences between groups was performed by ordinary one-way ANOVAs with Dunnett’s multiple-comparison correction using GraphPad Prism 7.0 (San Diego, CA, USA). The differences were deemed significant at P values of < 0.05, < 0.01, and < 0.001.