Epigenetic Metabolite Acetate Inhibits Class I/II Histone Deacetylases, Promotes Histone Acetylation, and Increases HIV-1 Integration in CD4⁺ T Cells

ABSTRACT

In this study, we investigated the effect of acetate, the most concentrated short-chain fatty acid (SCFA) in the gut and bloodstream, on the susceptibility of primary human CD4⁺ T cells to HIV-1 infection. We report that HIV-1 replication is increased in CD3/CD28-costimulated CD4⁺ T cells upon acetate treatment. This enhancing effect correlates with increased expression of the early activation marker CD69 and impaired class I/II histone deacetylase (HDAC) activity. In addition, acetate enhances acetylation of histones H3 and H4 and augments HIV-1 integration into the genome of CD4⁺ T cells. Thus, we propose that upon antigen presentation, acetate influences class I/II HDAC activity that transforms condensed chromatin into a more relaxed structure. This event leads to a higher level of viral integration and enhanced HIV-1 production. In line with previous studies showing reactivation of latent HIV-1 by SCFAs, we provide evidence that acetate can also increase the susceptibility of primary human CD4⁺ T cells to productive HIV-1 infection.

IMPORTANCE Alterations in the fecal microbiota and intestinal epithelial damage involved in the gastrointestinal disorder associated with HIV-1 infection result in microbial translocation that leads to disease progression and virus-related comorbidities. Indeed, notably via production of short-chain fatty acids, bacteria migrating from the lumen to the intestinal mucosa could influence HIV-1 replication by epigenetic regulatory mechanisms, such as histone acetylation. We demonstrate that acetate enhances virus production in primary human CD4⁺ T cells. Moreover, we report that acetate impairs class I/II histone deacetylase activity and increases integration of HIV-1 DNA into the host genome. Therefore, it can be postulated that bacterial metabolites such as acetate modulate HIV-1-mediated disease progression.

INTRODUCTION

Short-chain fatty acids (SCFAs) have been reported to play key roles in the development of many conditions and diseases, including HIV-1 infection (1–7). These carboxylic acids, which contain an aliphatic tail of 6 carbon atoms or less, are mainly synthesized in the cecum and proximal colon by anaerobic commensal bacteria composing the gut microbiota via a dynamic process alternating between the glycolytic pathway, the pentose phosphate pathway, and the Wood-Ljungdahl pathway (8–18). It is known that gut bacteria can transform indigestible food components (e.g., fiber, polysaccharides, and oligosaccharides) into saturated aliphatic organic acids carrying one to six carbons, such as formate (C₁), acetate (C₂), propionate (C₃), butyrate (C₄), valerate (C₅), and caproate (C₆), through various metabolic pathways (8–10, 14, 17, 19–24). The three major SCFAs, acetate, propionate, and butyrate, are found in an approximate molar ratio of 60:25:15 in the gut (8). This ratio, as well as the overall production of SCFAs, is influenced by many factors, including microbiota diversity, substrate source, and gut transit time. However, independently of this, it was established that acetate always remains the most concentrated SCFA in the colon and is the most important one found in the systemic circulation (25). Active transport mechanisms mediated by MCT-1 and SMCT-1 receptors allow SCFAs to reach the peripheral circulation via the hepatic and portal venous systems (8, 9, 26–28). The capacity of SCFAs to modulate and regulate numerous cell functions, such as ion transport, proliferation, differentiation, and gene expression, is initiated by their potential to ligate different surface receptors (10, 15). It has been demonstrated that the G protein-coupled receptors (GPCRs) FFAR2 (also called GPR43) and FFAR3 (also called GPR41) mediate signal transduction following their binding with different SCFAs (29). Like, all GPCRs, FFAR2 (which preferentially binds acetate and propionate) and FFAR3 (which binds to propionate, butyrate, and valerate) mainly signal via heterotrimeric G proteins (8, 30–33). In addition to these, MCT-1 or SMCT-1 transporters are also implicated in SCFA-mediated signal transduction events in many cell types, such as in diverse lymphocyte populations (34). In contrast to GPCRs, which respond to SCFAs through G-protein signal transduction cascades, these receptors promote SCFA uptake into the cell to modulate biological responses (9, 34). Once they have reached the cytoplasm, SCFAs are known to directly regulate gene expression by competitively inhibiting histone deacetylases (HDACs) at the hydrophobic cleft of their active site (35–37).

In the past few years, several groups have demonstrated that HDACs prevent gene transcription by removing histone-bound acetyl groups, a process leading to a closed and repressed chromatin structure (2, 38–40). Inhibitors of class I/II HDAC activity (such as SCFAs) block this epigenetic regulation and are involved in the development of different diseases. Recent publications have highlighted the role of HDACs in the repression of HIV-1 regulatory elements also called long-terminal repeats (LTR). In this case, negative transcription factors, such as Yin-Yang-1 (YY-1), activator protein-4 (AP-4), and the Sp1-myc complex, have been shown to recruit HDAC-1 to LTR or the TATA box and to restrain transcription from the viral promoter to maintain viral latency (2, 41–45). Interestingly, it has been reported that butyrate-producing bacteria residing in the mouth, vaginal, and gut mucous membranes are implicated in AIDS progression by inducing reactivation of HIV-1 proviral DNA in latently infected cell lines and mononuclear cells via inhibition of HDAC activity (2, 3, 6, 7, 36, 38, 41, 45–48).

Given the inherent ability of SCFAs to affect diverse biological cell functions and as HIV-1 infection leads to profound and early alterations of the permeability of the gut epithelium, we assessed the overall impact of acetate on the susceptibility of primary human CD4⁺ T cells to productive HIV-1 infection. Our studies were focused on acetate based on the idea that it is the most concentrated SCFA in the gut. Our results suggest that acetate treatment, when combined with CD3/CD28 costimulation to mimic antigen presentation, reduces class I/II HDAC activity, increases histone acetylation, and enhances HIV-1 integration into the host chromosomal DNA. It can thus be proposed that bacterial metabolites, such as acetate, could exert an influence on HIV-1-mediated disease progression.

RESULTS

Acetate increases HIV-1 replication in CD3/CD28-costimulated CD4⁺ T cells.

In the last few years, several studies have monitored the impact of SCFAs produced by bacteria on HIV-1 biology with particular emphasis on reactivation of virus gene expression (2–7, 49). Here, to characterize the potential modulatory effect of acetate on HIV-1 replication, quiescent primary human CD4⁺ T cells were purified by magnetic separation from peripheral blood mononuclear cells (PBMCs) from different healthy donors and were subjected to anti-CD3/CD28 costimulation in the absence or presence of acetate. The cells were next inoculated with replication-competent NL4.3Balenv virus (i.e., NL4.3 carrying Env glycoproteins of the R5-using Bal isolate), and the extracellular p24 contents were quantified either at the peak of virus production (i.e., 5 days postinfection [dpi]) or at various time points postinfection. The results displayed in Fig. 1A show that acetate enhances virus production in CD3/CD28-costimulated primary human CD4⁺ T cells. There was no virus replication in acetate-treated primary human CD4⁺ T cells, which were not subjected to anti-CD3/CD28 costimulation (data not shown). Experiments were also performed with the infectious reporter virus NL4.3Bal-IRES-HSA to measure the effect of acetate on HIV-1 replication at the single-cell level. An acetate-mediated enhancement in the percentage of heat-stable antigen (HSA)-expressing cells (indicative of the number of cells productively infected with HIV-1) was seen at 3 dpi (Fig. 1B). Our data also demonstrate that acetate augments the percentage of HIV-1-infected (i.e., HSA⁺) cells and not the mean fluorescence intensity (MFI) per individual cell, thus suggesting that acetate increases the overall sensitivity of primary human CD4⁺ T cells to productive HIV-1 infection without affecting virus gene expression at the single-cell level. Representative fluorescence-activated cell sorter (FACS) analyses are shown in Fig. 1C. We carried out experiments with increasing concentrations of acetate ranging from 2 to 40 mM. Cell viability was not affected by the acetate concentrations studied (Fig. 2A), and we detected an acetate-dependent, dose-dependent increase in HIV-1 infection (as monitored by estimating the percentage of HSA⁺ cells) (Fig. 2B). We selected the 20 mM concentration of acetate for the next series of experiments based on previous studies indicating that this level of acetate can be reached in the proximal and distal human colon (12). Together, these results suggest that acetate promotes productive HIV-1 infection in CD3/CD28-costimulated primary human CD4⁺ T cells.

FIG 1.

HIV-1 replication in CD3/CD28-costimulated CD4⁺ T cells is augmented by acetate. Purified primary human CD4⁺ T cells were costimulated with anti-CD3 and anti-CD28 MAbs in the absence or presence of acetate (20 mM). (A) The cells were next inoculated with replication-competent NL4.3Balenv viruses, and virus production was estimated either at 6 dpi (left) (6 donors are shown) or at the indicated time points postinfection (right) (a representative donor is shown) by measuring the p24 contents in cell-free supernatants. (B) In some experiments, untreated and acetate-treated cells were incubated with the NL4.3Bal-IRES-HSA reporter virus for 3 days before assessing HSA expression by flow cytometry. Percentages of HSA⁺ cells are depicted on the left, while the MFI is shown on the right. Each symbol represents a different donor, with the horizontal lines depicting the means of six donors tested. Statistical analyses were done using ratio-paired Student's t test. The asterisks denote statistically significant data (**, P ≤ 0.01; ***, P ≤ 0.001). (C) The gating strategy used in flow cytometry analyses to estimate the percentage of cells productively infected with HIV-1 (HSA⁺ as defined with an allophycocyanin [APC]-conjugated anti-HSA MAb) for each experimental condition consisted of live lymphocyte gating based on size and complexity on a forward scatter (FSC)/side scatter (SSC) plot (left), followed by doublet discrimination on an FSC-height (H)/FSC-width (W) plot (center), to finally gate HSA⁺ cells on an FSC-H/APC plot (right). Mock-infected cells were used as negative controls for HSA staining. The number in the plots indicates the percentage of cells within the gate.

FIG 2.

Acetate does not affect cell viability but induces a dose-dependent increase in HIV-1 replication. Purified primary human CD4⁺ T cells were costimulated with anti-CD3 and anti-CD28 MAbs in the absence or presence of increasing concentrations of acetate. (A) Cell viability was monitored by flow cytometry at day 6 following acetate treatment. (B) Purified primary human CD4⁺ T cells were first treated as described for panel A and then incubated with the NL4.3Bal-IRES-HSA reporter virus for 3 days before quantifying the percentages of HSA⁺ cells by flow cytometry. Each symbol represents a different donor, with the horizontal lines depicting the means of five donors tested. Statistical analyses were done using ratio-paired Student's t tests. The asterisks denote statistically significant data (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001).

Cell proliferation and activation profiles are affected differently by acetate treatment.

Cellular proliferation and activation are known to have an impact on the susceptibility of CD4⁺ T cells to HIV-1 infection. For example, cellular proliferation plays an important role in the HIV-1 life cycle by promoting virus dissemination, which helps to maintain viral reservoirs (50, 51), whereas cell activation allows the translocation of host transcription factors to the nucleus, where they trigger genes implicated in immune response and virus production (52, 53). Thus, cell proliferation was evaluated by the use of a dilution assay that is based on the fluorescent cell staining dye carboxyfluorescein succinimidyl ester (CFSE). We also studied the cell activation status by measuring the surface expression of some activation markers (i.e., CD25, CD69, and CD154) by flow cytometry. As expected, cell proliferation was induced in a statistically significant manner upon CD3/CD28 costimulation at the two time points tested (Fig. 3A). However, proliferation of CD4⁺ T cells was significantly decreased at the earliest time point by acetate treatment. Surface expression of the activation-associated receptors CD25, CD69, and CD154 was significantly induced following CD3/CD28 costimulation compared to untreated CD4⁺ T cells, while CD69 expression was the only surface marker to be further increased upon treatment with acetate (Fig. 3B). These observations demonstrate that acetate exhibits differential effects with respect to cell proliferation and activation.

FIG 3.

Acetate exerts differential effects on cell proliferation and activation markers. Purified primary human CD4⁺ T cells were first treated as described in the legend to Fig. 1. (A) A CFSE-based dilution assay was performed by flow cytometry to evaluate cell proliferation following acetate treatment for 3 or 6 days. Representative proliferation profiles are depicted on the left, whereas division indices are shown on the right. (B) Surface expression of some T cell activation markers (CD25, CD69, and CD154) was evaluated by flow cytometry following acetate treatment for 3 days. The data shown were obtained from CD4⁺ T cell preparations isolated from the peripheral blood of 4 (A) or 6 (B) distinct healthy donors. Each symbol represents a different donor, and the horizontal lines depict the means of all donors tested. Statistical analyses were done using one-way ANOVA, followed by a Dunnett multiple-comparison test. The asterisks denote statistically significant data (*, P ≤ 0.05; **, P ≤ 0.01; ****, P ≤ 0.0001).

Acetate does not modulate surface levels of CD4 but decreases CCR5 expression and virus entry into CD3/CD28-costimulated CD4⁺ T cells.

We aimed at elucidating the mechanism by which acetate promotes HIV-1 replication in CD3/CD28-costimulated primary human CD4⁺ T cells. To this end, we analyzed its impact on the surface expression of HIV-1 receptor and coreceptor, CD4 and CCR5. Although surface expression of both CD4 and CCR5 are significantly enhanced upon CD3/CD28 costimulation, acetate treatment does not alter surface levels of CD4 (Fig. 4A) but reduces CCR5 expression (Fig. 4B). Therefore, the acetate-mediated increase in HIV-1 production is not due to higher expression of cell surface receptors known to directly participate in the early events in the virus life cycle, such as attachment and/or entry. We performed a virus entry test to define whether the decrease in CCR5 surface levels affects the early steps in the HIV-1 life cycle. The results depicted in Fig. 5 demonstrate that HIV-1 entry was slightly lower in acetate-treated target cells.

FIG 4.

CCR5 surface expression in CD3/CD28-costimulated CD4⁺ T cells is diminished by acetate. Purified primary human CD4⁺ T cells were treated as specified in the legend to Fig. 1. Surface expression of CD4 (A) and CCR5 (B) in the total CD4⁺ T cell population was evaluated by flow cytometry. The data shown were obtained from CD4⁺ T cell preparations isolated from the peripheral blood of 5 (A) and 4 (B) distinct healthy donors. Each symbol represents a different donor, and the horizontal lines depict the means of all the donors tested. Statistical analyses were done using one-way ANOVA, followed by a Dunnett multiple-comparison test. The asterisks denote statistically significant data (*, P ≤ 0.05; **, P ≤ 0.01).

FIG 5.

Acetate decreases virus entry into CD3/CD28-costimulated CD4⁺ T cells. Purified primary human CD4⁺ T cells were subjected to CD3/CD28 costimulation in the absence or presence of acetate. Virus entry was estimated by quantifying the intracellular p24 contents by an enzyme-linked immunosorbent assay (ELISA) in the total cell population as described in Materials and Methods. The data shown represent the means and standard errors of the mean (SEM) of triplicate samples. Statistical analysis was performed using ratio-paired t tests (*, P < 0.05).

Acetate decreases class I/II HDAC activity in CD3/CD28-costimulated CD4⁺ T cells.

It is now well established that SCFAs, at physiological levels, can inhibit HDAC activity. Indeed, many groups have recently revealed the dominant role of this epigenetic regulatory mechanism in different contexts, such as neutrophil apoptosis and human colon cancer (40, 54, 55). Moreover, several publications have demonstrated the positive effect of butyrate on HIV-1 gene expression. Those data suggested that this SCFA promotes chromatin modification via histone acetylation and leads to HIV-1 reactivation in latently infected cells (2–7, 41, 46, 56). Based on the hypothesis that other SCFAs may induce epigenetic regulatory mechanisms such as histone acetylation to modulate HIV-1 replication, we investigated the ability of acetate to also act as a class I/II HDAC inhibitor in CD3/CD28-costimulated primary human CD4⁺ T cells. We found that acetate decreases class I/II HDAC activity in CD3/CD28-costimulated primary human CD4⁺ T cells (Fig. 6). As expected, the antifungal antibiotic trichostatin A (TSA), which is known to selectively inhibit class I/II mammalian HDACs, completely abrogates class I/II HDAC activity.

FIG 6.

Class I/II HDAC activity in CD3/CD28-costimulated CD4⁺ T cells is reduced by acetate. Purified primary human CD4⁺ T cells were treated as specified in the legend to Fig. 1. Some cells were also subjected to CD3/CD28 costimulation before treatment with TSA. The relative activity of class I/II HDACs in the total CD4⁺ T cell population was measured with the HDAC-Glo I/II assay according to the manufacturer's instructions. Each symbol represents a different donor, and the horizontal lines depict the means of 5 donors tested. Statistical analyses were done using one-way ANOVA, followed by a Dunnett multiple-comparison test. The asterisks denote statistically significant data (*, P ≤ 0.05).

Acetate induces acetylation of histones H3 and H4 in CD3/CD28-costimulated CD4⁺ T cells.

To establish a relationship between the acetate-mediated enhancement of HIV-1 infection in primary human CD4⁺ T cells and its HDAC-inhibitory activity, we performed intracellular flow cytometry analyses to quantify histone lysine acetylation events. The results from our studies reveal that acetate causes acetylation of both histones H3 (K9) and H4 (K5/8/12/16) in primary human CD4⁺ T cells (Fig. 7).

FIG 7.

Acetate mediates histone H3 and H4 acetylation. Primary human CD4⁺ T cells were first subjected to CD3/CD28 costimulation for 72 h in the absence or presence of acetate. Control cells were further stimulated for 4 h with the HDAC inhibitor TSA. Histone acetylation was detected by flow cytometry using MAbs specific for H3K9 (A) and H4 K5/8/12/16 (B). Each bar represents the mean percentages and standard deviations (SD) of positive cells from 5 distinct donors. Flow cytometry data obtained with cells costimulated with CD3 and CD28 MAbs were set at 100%. Statistical analyses were performed using ratio-paired t tests. The asterisks denote statistically significant data (*, P ≤ 0.05; **, P ≤ 0.01).

Acetate enhances integration of HIV-1 DNA in CD3/CD28-costimulated CD4⁺ T cells.

With its capacity to open chromatin by inhibiting class I/II HDAC activity, acetate could augment HIV-1 infection by facilitating provirus integration into the host genome (57, 58). To validate this hypothesis, we performed a combined Alu-HIV-1 PCR (where Alu-HIV-1 refers to Alu elements [a short stretch of DNA originally characterized by the action of the Arthrobacter luteus/Alu restriction endonuclease] and adjacent LTR/gag sequences [M667/M661]) and quantitative real-time PCR assay on purified primary CD4⁺ T cells costimulated with anti-CD3 and anti-CD28 monoclonal antibodies (MAbs) and either left untreated or treated with acetate. Moreover, we also measured HIV-1 completed reverse transcripts to further refine the mechanism of action of acetate (i.e., its modulatory effect on reverse transcription or integration). To discriminate between an effect of acetate on virus integration at the single-cell level versus an influence on cell-to-cell propagation, virus infection was allowed to proceed for 24 h only. Acetate caused a more significant increase in the number of copies of proviral DNA compared to the number of completed reverse transcripts (i.e., 1.57- versus 4.82-fold increase) (Fig. 8).

FIG 8.

Acetate exerts a more potent effect on HIV-1 integration events. Primary human CD4⁺ T cells from 6 distinct donors were first subjected to CD3/CD28 costimulation for 72 h in the absence or presence of acetate. Next, the cells were infected for 24 h with NL4.3Balenv viruses. The cells were processed to determine the number of copies of completed reverse transcripts (A) and proviral DNA (B) by a quantitative real-time PCR test. Each symbol represents a different donor, and the horizontal lines depict the means of 6 donors tested. Statistical analyses were done using ratio-paired Student's t tests. The asterisks denote statistically significant data (*, P ≤ 0.05; **, P ≤ 0.01).

DISCUSSION

Resulting from the fermentation of indigestible food components by anaerobic commensal bacteria composing the gut microbiome (8–10, 14, 17, 19–24), SCFAs have been suggested to affect HIV-1-mediated disease progression (1–7, 42). Previous studies have focused on the impact of the SCFA butyrate on reactivation of the provirus in cells latently infected with HIV-1. It has been suggested that butyrate acts as an inhibitor of class I HDACs, leading to histone hyperacetylation and enhancement of HIV-1 gene transcription (2, 3). Indeed, several lines of evidence indicate that HDACs repress HIV-1 LTR activity, whereas their inhibition increases the expression of viral genes (43–45). As enteric bacteria composing the gut microbiota are known to generate SCFAs as products of degradation of dietary fibers and these microorganisms are rapidly translocated to mucosal tissues following HIV-1 infection (59–64), the presence of these SCFAs could impact the overall permissiveness of CD4⁺ T cells to virus infection. The present work thus focused on acetate, the SCFA with the highest concentration in the colon and the major one found in the systemic circulation (8, 26). For instance, total acetate levels in humans range from a concentration of 63.4 mmol/kg of body weight in the ascending large intestine to 50 mmol/kg in the sigmoid/rectum anatomic segment (65). Moreover, the molar ratios of acetate to proprionate to butyrate are 91:5:4 in the peripheral blood (65). Importantly, our work was performed with a physiological concentration of acetate (i.e., 20 mM) based on previous studies indicating that the total concentrations of SCFAs vary from 70 to 140 mM in the proximal colon to 20 to 70 mM in the distal colon (12). Moreover, acetate, proprionate, and butyrate are found in an approximate molar ratio of 60:20:20 in the colon and stool (25, 66, 67).

First, we explored the impact of acetate on the replication of two R5-tropic NL4.3-based viruses in CD3/CD28-costimulated primary human CD4⁺ T cells. In contrast to previous studies, which focused on the effect of butyrate on reactivation of HIV-1 in latently infected cells (6), the present results underscore the impact of acetate exposure on the permissiveness of primary human CD4⁺ T cells to productive HIV-1 infection. Our observations revealed that acetate enhances the susceptibility of primary human CD4⁺ T cells to productive HIV-1 infection (i.e., a greater proportion of HIV-1-infected cells) without affecting virus gene expression at the single-cell level (Fig. 1 and 2).

It is now well established that CD4⁺ T cell activation is essential for the efficient replication of HIV-1 in these cells (68). While it is not required for viral entry, cellular activation is important for subsequent steps of the viral cycle, such as reverse transcription, integration of the provirus into the cell genome, and transcription of the viral genome (69). The influence of acetate treatment on cell activation was evaluated by assessing the expression of known surface activation markers (i.e., CD69, CD25, and CD154) by flow cytometry. As shown in Fig. 3B, the presence of acetate increases the expression of the early activation marker CD69 while surface expression of CD25 and CD154 are not affected. As CD25 and CD154 are, respectively, intermediate and late activation markers (70), it is possible that a stimulation period of 3 days is too brief to significantly modulate CD25 and CD154 among all the donors tested. Nonetheless, the increased expression of CD69 implies a positive effect of acetate on the activation profile of CD3/CD28-costimulated CD4⁺ T cells. As an enhanced activation state suggests greater susceptibility of CD4⁺ T cells to HIV-1 infection, acetate treatment could possibly promote HIV-1 infection via this mechanism.

Furthermore, it is now well established that cellular activation leads to a variety of responses in CD4⁺ T cells, such as their proliferation (71). Proliferation of infected cells increases their pool and contributes to the maintenance of reservoir cells (50, 51). As acetate treatment modulates the early activation marker CD69, we performed cell proliferation studies. Surprisingly, acetate treatment significantly decreased the average number of cell divisions at day 3 poststimulation. However, the acetate-mediated reduction in cell proliferation was no longer seen at a later time point (i.e., 6 days poststimulation). Considering the important role of the K⁺ channels in controlling T-cell proliferation (72–75) and that butyrate has been shown to downregulate calcium-activated K⁺ channels in human airway epithelial cells (76), it is possible that the observed delay of cellular proliferation could be partially caused by a similar effect of acetate on K⁺ channels in CD3/CD28-costimulated CD4⁺ T cells. Thus, despite having a positive impact on the cell surface activation marker profile, acetate delays cellular proliferation. Enhanced proliferation of virus-infected CD4⁺ T cells therefore cannot explain the effect of acetate on the greater permissiveness of CD3/CD28-costimulated CD4⁺ T cells to HIV-1 infection.

HIV-1 interacts with different constituents at the surfaces of its host cells. Two of them, the primary receptor CD4 and the predominant coreceptor CCR5, act sequentially in concert with the viral envelope protein gp120 to promote fusion of viral and cellular membranes and allow viral entry (77). As modulation of their expression could affect cellular permissiveness to HIV-1, we investigated the impact of acetate treatment on the expression of these two proteins at the surfaces of CD3/CD28-costimulated CD4⁺ T cells. As illustrated in Fig. 4, CD4 expression was significantly increased following CD3/CD28 costimulation but was not affected further by acetate treatment. However, while CCR5 expression was also significantly augmented at the surfaces of CD3/CD28-costimulated CD4⁺ T cells, concomitant exposure to acetate returned CCR5 surface expression to basal levels, near those found on the surfaces of quiescent cells. The influence of acetate on CCR5 expression is reminiscent of the action of ITF2357, another HDAC inhibitor, which was shown to inhibit the transcription of the CCR5 gene in PBMCs (78). As cell surface expression of CCR5 is correlated with the susceptibility of target cells to R5-tropic HIV-1 infection (79–81), our observations indicating that the process of virus entry is reduced by acetate are not unexpected (Fig. 5).

The mechanism by which acetate mediates a positive effect on HIV-1 replication in activated CD4⁺ T cells despite its capacity to decrease virus entry appears to be a paradox. Following multiple lines of studies demonstrating that HDAC inhibitors can drive HIV-1 gene expression in various experimental model systems of HIV-1 latency, we assessed the capacity of acetate to abrogate HDAC activity in CD3/CD28-costimulated primary human CD4⁺ T cells. We found that acetate decreased class I/II HDAC activity in such cells. Indeed, CD3/CD28 costimulation significantly increased class I/II HDAC activity at day 3 poststimulation, whereas treatment with acetate reduced HDAC activity to levels almost similar to those found in quiescent cells (Fig. 6). Importantly, we provide evidence that acetate induces acetylation of both histones H3 and H4 in primary human CD4⁺ T cells (Fig. 7). These results are highly significant considering that acetylation of the lysine residues at the N termini of histone proteins (primarily histones H3 and H4) results in decondensation of heterochromatin (i.e., the transcriptionally silent form of chromatin) into the more relaxed euchromatin structure (i.e., the transcriptionally active form of chromatin). Moreover, it is now established that HIV-1 preferentially integrates into the euchromatin based on in vitro (82, 83) and in vivo studies (83, 84).

Considering that inhibition of such chromatin modifiers facilitates a switch from repressive heterochromatin to permissive euchromatin (2), inhibition of class I/II HDAC activity could promote the integration of proviral DNA into host euchromatic domains. The results from HIV-1 integration assays, which combine an Alu-HIV-1 PCR and a quantitative real-time PCR test (85), showed that acetate can potently increase integration of HIV-1 DNA into the host's genome (Fig. 8). Interestingly, acetate displays a more significant impact on virus integration than completion of reverse transcripts. These results suggest that acetate, by acting as a class I/II HDAC inhibitor, could relieve epigenetic controls restricting retroviral integration into the host DNA, possibly by allowing the provirus to access open and active regions of genomic DNA where viral integration is strongly favored (86, 87).

Based on data described in the present study, we hypothesize that the bacterial metabolite acetate could play a role in the early stages of HIV-1 infection, such as initial spreading and reservoir establishment, by enhancing cellular activation and viral integration events. Considering that other SCFAs also act as inhibitors of class I/II HDAC activity (7, 35–37), their effect could be compounded by the action of acetate in a physiological setting. Therefore, further studies are required to provide a better understanding of the putative impact of bacterial metabolites such as SCFAs during HIV-1 pathogenesis.

MATERIALS AND METHODS

Ethics statement and cell culture.

The current study was approved by the Bioethics Committee of the Centre Hospitalier Universitaire de Québec-Université Laval. Samples were collected from healthy donors in accordance with the guidelines of the Institutional Bioethics Committee. All the donors provided written informed consent. To obtain purified primary human CD4⁺ T cells, PBMCs were isolated by Ficoll-Hypaque gradient centrifugation from healthy donors and kept frozen at −80°C at a final concentration of 100 × 10⁶ cells/ml in freezing medium (90% fetal bovine serum [FBS] and 10% dimethyl sulfoxide [DMSO]) for a maximum of 3 months. Purified CD4⁺ T cells were isolated from thawed PBMCs by an immunomagnetic negative-selection procedure according to the manufacturer's instructions (StemCell Technologies, Vancouver, BC, Canada) and cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% FBS (Thermo Fisher Scientific) at a final concentration of 2 × 10⁶ cells/ml.

Human embryonic kidney (HEK) 293T cells were kindly provided by Warner C. Greene (The J. Gladstone Institutes, San Francisco, CA), and TZM-bl indicator cells were obtained from John C. Kappes, Xiaoyun Wu, and Tranzyme Inc. via the AIDS Research Reagent Program (Germantown, MD) (88–92). The cell lines were cultured in Dulbecco's modified Eagle medium (DMEM) (Thermo Fisher Scientific) supplemented with 10% FBS.

Antibodies, reagents, and infectious virus constructs.

Allophycocyanin (APC)-conjugated anti-mouse CD24 (also known as HSA) was purchased from eBioscience (San Diego, CA). Fluorescein isothiocyanate (FITC)-conjugated anti-human CD4, APC-Cy7-conjugated anti-human CD195 (CCR5), phycoerythrin (PE)-conjugated anti-human CD25, FITC-conjugated anti-human CD69, and FITC-conjugated anti-human CD154 were all obtained from BD Biosciences (Mississauga, ON, Canada). The CellTrace CFSE Cell proliferation kit was purchased from Thermo Fisher Scientific. Hybridomas producing 183-H12-5C and 31-90-25 MAbs, which recognize different epitopes of the HIV-1 major viral core protein p24 Gag, were supplied by the AIDS Research Reagent Program and the ATCC (Manassas, VA), respectively. The rabbit anti-acetyl-histone H3 (K9) MAb (Alexa Fluor488 conjugate) was purchased from Cell Signaling Technology (Danvers, MA), while the anti-acetyl-histone H4 MAb (K5/8/12/16) was purchased from EMD Millipore (Etobicoke, ON, Canada). Mix-n'-Stain kit CF405M was purchased from Sigma-Aldrich (Saint Louis, MO) and used to stain anti-acetyl-histone H4 MAb following the manufacturer's protocol. The optimal concentration of the anti-acetyl-histone H4 MAb was determined by titration on TSA-stimulated and control cells before experiments were performed. A 200 mM stock solution of acetate was made by diluting glacial acetic acid (99.8% purity; Merck, Germany) in phosphate-buffered saline (PBS) and adjusted to a pH of 7.4 using NaOH. This neutral acetate solution was then filtered on a 0.22-μm syringe filter. Controls for acetate consisted of sterile PBS (pH 7.4). The infectious molecular vectors pNL4.3Balenv and pNL4.3Bal-IRES-HSA are described below.

Production of virus stocks.

The NL4.3Balenv vector codes for R5-tropic Bal envelope (Env) glycoproteins inserted within the NL4.3 backbone and generates fully infectious HIV-1 particles (93). The NL4.3Bal-IRES-HSA plasmid produces R5-using replication-competent reporter viruses and carries all the viral genes, as well as a reporter gene coding for the murine HSA protein, which is expressed on the membranes of productively infected cells via a glycosylphosphatidylinositol anchor (94). Virus preparations were made by calcium phosphate transfection of the infectious molecular clones pNL4.3Balenv and pNL4.3Bal-IRES-HSA. Briefly, HEK293T cells were transfected with the expression vectors, and cell-free supernatants were collected after 48 h. Newly produced viral particles were obtained following an ultracentrifugation step (100,000 × g for 45 min). The pellet was suspended in 2 ml of endotoxin-free PBS and frozen at −80°C until needed. Quantification of infectious virus titers harvested after ultracentrifugation was performed according to the Spearman-Karber method and following infection of TZM-bl indicator cells.

Stimulation and virus infection assays.

Purified primary human CD4⁺ T cells (1 × 10⁶ cells per experimental condition) were cultured in RPMI 1640 medium supplemented with 10% FBS at a final concentration of 2 × 10⁶ cells/ml, seeded in 48-well flat-bottom tissue culture plates, and either left untreated or treated with anti-CD3 (clone OKT3; 5 μg/ml) and anti-CD28 (clone 9.3; 2 μg/ml) MAbs in the presence or absence of acetate (at a final concentration of 20 mM for most experiments). After 72 h of incubation at 37°C under a 5% CO₂ atmosphere, the cells were either left uninfected to analyze surface expression of CD4, CCR5, CD25, CD69, and CD154 by flow cytometry or inoculated with HIV-1. In some studies, cellular viability was assessed by flow cytometry using the fixable viability dye eFluor 780 (Thermo Fisher Scientific). Virus infection experiments were performed using NL4.3Balenv and NL4.3Bal-IRES-HSA virus preparations added to cells at a final multiplicity of infection (MOI) of 0.1 for 2 h, after which the cells were washed three times with 2 mM PBS-EDTA/0.5% bovine serum albumin (BSA), resuspended in RPMI 1640 medium supplemented with 10% FBS, and incubated at 37°C under a 5% CO₂ atmosphere. Virus replication was estimated either by measuring the p24 content at different days postinfection with an in-house enzymatic assay (95) (for cells infected with NL4.3Balenv) or by quantifying the percentages of HSA⁺ cells at 3 dpi (for cells infected with NL4.3Bal-IRES-HSA).

Virus entry assay.

Primary human CD4⁺ T cells were either left untreated or treated as described above. Next, the cells were inoculated with NL4-3 Balenv (50 ng of p24 per 1 × 10⁵ cells) for 2 h at 37°C. The cells were washed twice with an equal volume of PBS (pH 7.4), resuspended in 100 μl 2.5% trypsin-EDTA (Thermo Fisher Scientific), and incubated for 10 min at 37°C to remove unadsorbed/uninternalized virions. Finally, the cells were washed three times with PBS (pH 7.4) and lysed using 500 μl of disruption buffer (20 mM HEPES [pH 7.4], 150 mM sodium chloride, and 0.5% Triton X-100). Samples were kept at −80°C until they were assayed by the homemade p24 enzymatic test.

Flow cytometry studies.

Surface expression of HSA, CD4, CCR5, CD25, CD69, and CD154 on the total population of purified CD4⁺ T cells was analyzed by flow cytometry. Briefly, for each experimental condition, 2 × 10⁵ cells were resuspended in 100 μl wash/stain buffer (PBS-2 mM EDTA-0.5% BSA) and stained with the appropriate MAb at optimal concentrations for 15 min at room temperature under dark conditions. The cells were washed twice with wash/stain buffer and resuspended in 2% paraformaldehyde (Thermo Fisher Scientific) prior to flow cytometry analysis (BD FACSCanto; BD Biosciences. Mississauga, ON, Canada). Cellular proliferation assays were performed with the CellTrace CFSE cell proliferation kit (Thermo Fisher Scientific). Briefly, primary human CD4⁺ T cells were stained with the CFSE dye as specified in the experimental protocol provided with the kit and then stimulated for 72 h with anti-CD3 (clone OKT3; 5 μg/ml) and anti-CD28 (clone 9.3; 2 μg/ml) MAbs in the absence or presence of acetate (20 mM). Fluorescence emission was measured by flow cytometry (BD FACSCanto; BD Biosciences, Mississauga, ON, Canada) to determine the effect of the acetate treatment on cell proliferation.

Analysis of class I/II HDAC activity.

Class I/II HDAC activity was monitored using the HDAC-Glo I/II commercial assay (Promega, Madison, WI). This luminescent assay measures the relative activity of class I/II HDAC from cell extracts. The test uses an acetylated, live-cell-permeant, luminogenic peptide substrate that can be deacetylated by HDAC activities. Deacetylation of the peptide substrate is measured using a coupled enzymatic system in which a protease in the developer reagent cleaves the peptide from aminoluciferin, which is quantified in a reaction using Ultra-Glo recombinant luciferase. In brief, cells were stimulated with anti-CD3 (clone OKT3; 5 μg/ml) and anti-CD28 (clone 9.3; 2 μg/ml) MAbs for 72 h in the absence or presence of acetate (20 mM) or TSA (200 nM) (Promega, Madison, WI), resuspended in PBS-2 mM EDTA (1 × 10⁴ cells/well), and incubated with HDAC-Glow I/II reagent supplemented with 1% Triton X-100 (Sigma-Aldrich, St. Louis, MO) for 30 min at room temperature. Luminescence was measured using a Varioskan Flash multimode reader (Thermo Fisher Scientific). Controls consisted of TZM-bl nuclear extracts because this HeLa-derived cell line contains high endogenous class I/II HDAC activity (unpublished observations).

Quantification of HIV-1 completed reverse transcripts and proviral DNA copies.

Primary human CD4⁺ T cells (3 × 10⁶ cells/condition) were cultured in RPMI 1640 medium supplemented with 10% FBS at a concentration of 2 × 10⁶ cells/ml, seeded in 48-well flat-bottom tissue culture plates, and treated for 72 h with anti-CD3 (clone OKT3; 5 μg/ml) and anti-CD28 (clone 9.3; 2 μg/ml) MAbs in the absence or presence of acetate (20 mM). NL4.3Balenv virus stocks were treated with DNase I (40 μg/ml; Roche, Bâle, Switzerland) in the presence of MgCl₂ (10 mM) and EGTA for 45 min at room temperature prior to their use. Controls consisted of cells treated for 4 h with TSA (200 nM). Next, the cells were infected with HIV-1 for 2 h before being washed three times with PBS-2 mM EDTA-0.5% BSA. The cells were resuspended in RPMI 1640 medium supplemented with 10% FBS and incubated for 24 h to minimize cell-to-cell virus propagation. Genomic DNA was subsequently extracted using a Nucleospin tissue DNA purification kit (Macherey-Nagel, Duren, Germany). Finally, integrated proviral DNA copies were quantified using a combined Alu-HIV-1 PCR and quantitative real-time PCR assay as described previously (85, 96), whereas quantification of completed reverse transcripts was achieved by a quantitative real-time PCR test in a 10-μl reaction mixture containing 20 ng of DNA, TaqMan Fast advanced master mix (Thermo Fisher Scientific), 1 μM (each) of HIV-1-specific sense M667 and antisense M661 primers, and 0.3 μM of the TaqMan probe HIV-5′-carboxyfluorescein (97). Normalization of quantitative real-time PCR data was achieved using the β-globin housekeeping gene. In brief, from every diluted DNA sample, β-globin was quantified using 1 μM of sense (TGGTCTATTTTCCCACCCT) and antisense (TGGCAAAGGTGCCCTTGA) specific primers and 0.3 μM of the TaqMan probe 5′-β-globin-VIC-TCTGTCCACTCCTGATGCTG-NFQ-MGB-3′.

Measurement of histone H3 and H4 acetylation by flow cytometry.

Primary human CD4⁺ T cells were either left untreated or treated as described above. Next, the cells were washed twice in PBS-2 mM EDTA-0.5% BSA and fixed in 100 μl of 4% formaldehyde for 15 min at room temperature. Controls consisted of cells stimulated with TSA (200 nM) for 4 h before formaldehyde fixation. The cells were washed twice with PBS-2 mM EDTA-0.5% BSA and permeabilized in 100 μl of BD perm/wash buffer for 45 min at 4°C in the dark. The cells were washed twice with PBS-2 mM EDTA-0.5% BSA and stained for 45 min in the dark at room temperature using a mixture of H3 and H4 MAbs. Finally, the cells were washed twice and resuspended in PBS-2 mM EDTA before monitoring the fluorescence intensity by flow cytometry.

Statistical analysis.

Statistical analysis was performed using GraphPad Prism version 6. Ratio-paired Student's t test and one-way analysis of variance (ANOVA) with corrections for multiple comparisons (Dunnett) were performed to define the statistical significance of our results. A threshold P value of ≤0,05 was considered statistically significant, whereas P values of ≤0,01, ≤0,001, and ≤0,0001 were considered highly significant.