Research ArticlesPhysiology

Resveratrol stimulates the metabolic reprogramming of human CD4+ T cells to enhance effector function

See allHide authors and affiliations

Sci. Signal.  17 Oct 2017:
Vol. 10, Issue 501, eaal3024
DOI: 10.1126/scisignal.aal3024

Resveratrol and lymphocyte responses

Resveratrol is a polyphenolic plant compound, which has attracted much interest as a pharmacological agent because of its potential therapeutic effects against cancer, aging, and inflammation. However, many studies have produced conflicting evidence of the effects of resveratrol in different contexts. Craveiro et al. found that high doses of resveratrol inhibited the responses of human CD4+ T cells to antigens. However, low doses of the drug reprogrammed the metabolism of the cells to make them more responsive to antigens and produce increased amounts of the inflammatory cytokine interferon-γ. These data suggest that the use of resveratrol to treat various pathologies should be carefully assessed, especially in an autoimmune setting.

Abstract

The polyphenol resveratrol activates the deacetylase Sirt1, resulting in various antioxidant, chemoprotectant, neuroprotective, cardioprotective, and anti-inflammatory properties. We found that at high concentrations of resveratrol, human CD4+ T cells showed defective antigen receptor signaling and arrest at the G1 stage of the cell cycle, whereas at low concentrations, cells were readily activated and exhibited enhanced Sirt1 deacetylase activity. Nevertheless, low-dose resveratrol rapidly stimulated genotoxic stress in the T cells, which resulted in engagement of a DNA damage response pathway that depended on the kinase ATR [ataxia telangiectasia–mutated (ATM) and Rad3-related], but not ATM, and subsequently in premitotic cell cycle arrest. The concomitant activation of p53 was coupled to the expression of gene products that regulate cell metabolism, leading to a metabolic reprogramming that was characterized by decreased glycolysis, increased glutamine consumption, and a shift to oxidative phosphorylation. These alterations in the bioenergetic homeostasis of CD4+ T cells resulted in enhanced effector function, with both naïve and memory CD4+ T cells secreting increased amounts of the inflammatory cytokine interferon-γ. Thus, our data highlight the wide range of metabolic adaptations that CD4+ T lymphocytes undergo in response to genomic stress.

INTRODUCTION

Resveratrol (3,5,4′-trihydroxy-trans-stilbene) is a natural polyphenolic compound that is produced by plants in response to environmental stress, providing them with protection from microbial infections (14). Resveratrol appears to mimic the effects of caloric restriction, increasing life span in lower organisms (5). Furthermore, this pharmacological agent has elicited much interest because of its potential to modulate a diverse array of pathological conditions, and it is associated with anticancer, antiaging, and anti-inflammatory properties (610). On the basis of the promising data emerging from ex vivo studies and preclinical animal models, resveratrol has been tested in more than 30 clinical trials involving more than 1000 individuals. Nevertheless, the specific pathologies in which resveratrol has substantial clinical benefits are not yet clear (1116).

The pharmacological properties of resveratrol have been attributed, at least in part, to its activation of the nicotinamide adenine dinucleotide (NAD+)–dependent silent information regulator 2 (Sir2) deacetylase (17) both in vitro (18, 19) and in vivo (20). Overexpression of the mammalian Sir2 homolog sirtuin-1 (Sirt1) in mice extends their life span (21, 22) and protects them from a diverse array of diseases (2328). Conversely, knocking out Sirt1 is associated with autoimmunity (2932). However, the effects of Sirt1 are likely to be complex. Although Sirt1 attenuates murine T cell signaling and effector function (29, 30, 3335), it also promotes the differentiation of naïve CD4+ T cells into T helper 17 (TH17) effector cells in mice (36). Furthermore, physiological modifications of Sirt1 function in human T cell subsets have thus far not been evaluated.

T cell activity is of great importance in a wide range of pathophysiological conditions for which resveratrol activity is being clinically evaluated. Hence, elucidating the potential on-target and off-target effects of resveratrol on T lymphocytes is critical. T cells present a complex target because their cellular metabolism is altered after activation by a cognate antigen. The capacity of T lymphocytes to respond to stimulation by antigen depends on an extensive proliferative response, a process that requires new energetic and biosynthetic components that are supplied, at least in part, through a metabolic shift from oxidative phosphorylation (OXPHOS) toward glycolytic and glutaminolytic pathways (3739). This shift from OXPHOS contrasts with the activity of resveratrol, a compound that generally increases mitochondrial activity and associated OXPHOS (4042). However, note that resveratrol leads to a wide range of effects, including decreased, stabilized, and enhanced T cell effector functions (4346).

Disparate effects of resveratrol on genomic stability have also been reported. In some studies, resveratrol contributes to genomic stability and reduces tumorigenesis by reducing the amount of reactive oxygen species (ROS), which leads to oxidative damage (4751). However, in other studies, resveratrol mediates DNA damage, facilitating antitumor treatments (47, 5259). One possible reason for these discrepancies could be that resveratrol has distinct effects on quiescent cells versus proliferating cells. In this regard, T lymphocytes present a challenging target. Although they are generally quiescent, exposure to foreign antigen rapidly stimulates cell cycle entry and cellular proliferation. A coordinated response to genotoxic stress is regulated by the kinases ATM (ataxia telangiectasia–mutated) and ATR (ATM and Rad3-related) (60, 61). Of interest are reports that resveratrol activates one or both of these kinases in different cellular contexts (5355). Here, we report that resveratrol rapidly stimulates the ATR-dependent damage pathway in antigen-stimulated human CD4+ T cells, with activation of the tumor suppressor p53. This genotoxic stress response links a metabolic reprogramming to an enhanced CD4 T cell effector function characterized by increased production of the cytokine interferon-γ (IFN-γ).

RESULTS

Low-dose resveratrol increases the activity of the NAD+-dependent deacetylase Sirt1 in primary human CD4+ T cells

To gain insight into the role of resveratrol in modulating Sirt1 function in human CD4+ T lymphocytes, we first examined its expression profile in response to T cell receptor (TCR) stimulation. We found that TCR engagement resulted in a substantial increase in Sirt1 abundance, with augmented nuclear localization and aggregation (Fig. 1A). Low-dose resveratrol (20 μM) further increased the mean fluorescence intensity (MFI) of Sirt1 staining by about twofold (Fig. 1A and fig. S1A). However, high-dose resveratrol (100 μM) attenuated the TCR-mediated increase in Sirt1 abundance, and these CD4+ lymphocytes did not undergo blast formation (Fig. 1A and fig. S2A). This differed markedly from treatment with low-dose resveratrol, which augmented blast size (fig. S2A). Separating subsets of TCR-stimulated CD4+ T cells based on their forward scatter (FSC) and side scatter (SSC) profiles demonstrated that Sirt1 abundance paralleled increases in cell size and granularity (Fig. 1B). As expected from these data, low-dose resveratrol substantially increased Sirt1 activity in TCR-stimulated lymphocytes, as monitored by the generation of O-acetyl–adenosine diphosphate–ribose (OAADPr), a reaction product of the Sirt-catalyzed, NAD+-dependent deacetylation of target proteins (Fig. 1C). Thus, TCR stimulation combined with low-dose, but not high-dose, resveratrol augments Sirt1 activity in human T lymphocytes.

Fig. 1 Resveratrol modulates the TCR-stimulated activity of the NAD+-dependent deacetylase Sirt1 and blast formation in human CD4+ T cells.

(A) Left: The presence of Sirt1 in freshly isolated quiescent human CD4+ T cells was assessed by staining with a Sirt1 polyclonal antibody (green) and the nuclear stain 4′,6-diamidino-2-phenylindole (DAPI) (blue). The presence and localization of Sirt1 were also assessed after 1 (D1) or 3 (D3) days under nonstimulating conditions (NS) or after TCR stimulation (TCR) in the absence or presence of 20 or 100 μM resveratrol (RVT). Images are representative of 50 cells in three independent experiments. Right: Sirt1 was also monitored by flow cytometric analysis of quiescent human CD4+ T cells and cells that were stimulated for 3 days through the TCR in the absence or presence of 20 or 100 μM resveratrol. Data are representative of four independent experiments. (B) Left: Sirt1 abundance in CD4+ T cells treated with 20 μM resveratrol was monitored by flow cytometry as a function of both FSC and SSC. The four numbered populations of cells were distinguished on the basis of their FSC-SSC characteristics (dot plot), and Sirt1 abundance in the indicated populations was further analyzed. Right: Histograms are representative of 10 independent experiments. Bottom: Sirt1 abundance and localization in cells from the indicated populations were also analyzed by immunofluorescence staining. Images are representative of three experiments. (C) Sirt1 deacetylase activity was monitored as a function of the generation of OAADPr generation, a reaction product of Sirt1-catalyzed NAD+-dependent protein deacetylation. CD4+ T cells were untreated or were stimulated through the TCR in the absence or presence of 20 or 100 μM resveratrol for 3 days before the amount of OAADPr in each sample was determined. Data are means ± SEM of six independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.0001 by one-way analysis of variance (ANOVA) and Tukey’s post hoc test.

Low- and high-dose resveratrol stimulate distinct types of G1 and premitotic cell cycle arrest in TCR-stimulated CD4+ T cells

In light of our findings that high-dose resveratrol inhibited TCR-mediated increases in T cell size and granularity, it was of interest to determine how different doses of resveratrol affected T cell proliferation and cell cycle progression. Although cell viability was not affected by resveratrol (fig. S2A), high-dose resveratrol inhibited the entry of T lymphocytes into the G1 phase of the cell cycle, as assessed by the detection of reduced amounts of total RNA (Fig. 2A and fig. S1B). On the other hand, T cells exposed to low-dose resveratrol exhibited a cell cycle entry and progression that was equivalent to that observed in control TCR-stimulated cells, with about 40% of cells having entered into S phase by day 3 of stimulation (Fig. 2A and fig. S1B). However, note that low-dose resveratrol almost completely abrogated TCR-mediated cellular proliferation (Fig. 2B and fig. S1C), an effect that was not ameliorated by the addition of exogenous interleukin-2 (IL-2) nor by extended time in culture (fig. S2, B and C).

Fig. 2 Low- and high-dose resveratrol block TCR-mediated cell cycle progression at distinct stages of the cell cycle.

(A) Cell cycle entry after TCR stimulation at day 1 (D1; top) and day 3 (D3; bottom) in the presence or absence of resveratrol was monitored by simultaneous staining of DNA and RNA with 7-aminoactinomycin D and pyronin Y, respectively. Representative dot plots from five experiments of nonstimulated and TCR-activated CD4 T cells, in the absence or presence of resveratrol, are shown. The percentages of cells in G0-G1A phase (lower left quadrant), G1B phase (lower right quadrant), and S, G2, and M phases (upper right quadrant) are indicated. (B) T cell proliferation under the indicated conditions was monitored by carboxyfluorescein diacetate succinimidyl ester (CFSE) labeling, and dilution of the fluorescent dye was assessed at 72 hours. The number of division peaks is indicated in each histogram. Data are representative of six experiments. (C) Schematic representation of cell cycle regulators that are altered upon TCR-mediated cell cycle entry. Cell cycle progression requires the expression of cyclins and Cdks, the F-box protein Skp2-dependent and ubiquitin-mediated degradation of the p27Kip1 Cdk inhibitor, and Cdk-mediated hyperphosphorylation of the pRb tumor suppressor and the related p130 pocket protein. Cdk1 activity and mitotic entry are regulated by the kinase Wee1 and the phosphatase of Cdc25. (D) The abundances of the cyclins-Cdks that regulate cell cycle entry, including cyclins D2, E1, A2, and B1, and Cdk4, Cdk6, Cdk2, and Cdk1 were monitored by Western blotting analysis on days 1 and 3 of activation. Data are representative of three independent experiments. The arrow indicates the hyperphosphorylated Cdk1 isoform. (E) The abundances of the Ki-67 proliferation marker, cell division inhibitors (pRb, p130, and p27), and the p27 regulator Skp2 under the indicated conditions were monitored by Western blotting analysis. Data are representative of three independent experiments. Arrows indicated hyperphosphorylated p130 and phosphorylated Skp2. Quantification of all panels is shown in fig. S1.

Cell cycle entry and progression are tightly controlled processes involving the action of cyclin-dependent kinases (Cdks) and cyclins (Fig. 2C). To understand the molecular bases underlying the cell cycle arrest caused by different doses of resveratrol, we investigated the regulation of components of the cell cycle machinery. Cyclins D2, E1 and A2, and B1, as well as their cognate kinases, Cdk4/Cdk6, Cdk2, and Cdk1, respectively, were increased in abundance upon T cell activation. Although this increased abundance was not altered by low-dose resveratrol, it was significantly attenuated by high-dose resveratrol under conditions in which high amounts of the Cdk inhibitor p27Kip1 were maintained (P < 0.05 at day 1 and P < 0.005 at day 3; Fig. 2, D and E, and fig. S1D). However, by day 3 of activation, high-dose resveratrol-treated cells showed increased cyclin D2 and cyclin E1, but these cells did not progress into S phase (Fig. 2A). Cdk2, cyclin A2, and Cdk1 were not detectable, and phosphorylation of the pocket proteins pRb (retinoblastoma protein) and p130, hallmarks of S-phase progression, was also not observed (Fig. 2, D and E, and fig. S1E). Furthermore, the Cdks regulating the G1-S phase progression (Cdk4, Cdk6, Cdk2, and Cdk1) were not increased in abundance (Fig. 2D). We found that this was likely because of the reduced abundance and phosphorylation of the F-box protein Skp2 (Fig. 2E, upper band, and fig. S1E), the rate-limiting component responsible for p27Kip1 ubiquitination and degradation (62, 63). The premitotic cell cycle arrest that was triggered by low-dose resveratrol did not result in senescence, as shown by the enhanced phosphorylation of pocket proteins, increased amounts of cyclins A2 and B1 and the proliferation marker Ki-67, as well as decreased p27Kip1 abundance (Fig. 2, D and E, and fig. S1, D and E). Furthermore, accumulation of hyperphosphorylated Cdk1 (Fig. 2D, arrow) suggests that resveratrol blocks the G2-M transition of the cell cycle by abrogating the Cdc25-mediated activation of Cdk1 (Fig. 2C).

Low-dose resveratrol does not alter TCR-stimulated proximal and distal signaling cascades

To determine whether the cell cycle blockade mediated by low-dose resveratrol was due to defective initiation of the TCR signaling cascade, we first assessed proximal signaling intermediates. The kinase ZAP-70, which is associated with the TCRζ chain in activated lymphocytes, was phosphorylated by 1 min after TCR engagement and was not affected by either low- or high-dose resveratrol (Fig. 3A). Further downstream signaling was monitored as a function of extracellular signal–regulated kinase 1/2 (ERK1/2) and AKT phosphorylation. Neither ERK1/2 nor AKT phosphorylation was altered by low-dose resveratrol, and phosphorylation was only marginally decreased in the presence of high-dose resveratrol (Fig. 3, A and B, and fig. S1F).

Fig. 3 TCR signaling is attenuated by high-dose, but not low-dose, resveratrol.

(A) Top: Phosphorylation of ZAP-70 and ERK after TCR stimulation of human CD4+ T cells in the presence of 20 or 100 μM resveratrol was monitored by flow cytometry. Representative histograms at 1 min after stimulation are presented. Bottom: Quantification of the fold increase in MFIs of the indicated proteins in stimulated relative to nonstimulated CD4+ T cells. Data are means ± SEM of three independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.005 by one-way ANOVA and Tukey’s post hoc test. (B) The relative amounts of total and phosphorylated Akt in T cells 5 and 30 min after activation under the indicated conditions were determined by Western blotting analysis. Data are representative of three experiments. (C) The extent of phosphorylation of TSC2, mTOR, and S6 in CD4+ T cells 24 hours after stimulation under the indicated conditions was monitored by Western blotting analysis. Data are representative of three experiments. (D) Left: The cell surface expression of the CD69, IL-2Rα (CD25), and transferrin receptor (CD71) activation markers on CD4+ T cells stimulated under the indicated conditions were assessed by flow cytometry. Histograms are representative of three independent experiments. Right: Quantification of the percentages of positive cells under each condition. **P < 0.01 and ****P < 0.0001 by two-way ANOVA with Bonferroni’s post hoc test. Quantification data are shown in fig. S1.

TCR signaling also activates mammalian target of rapamycin (mTOR) (6466), a serine/threonine protein kinase that integrates environmental cues such as nutrients, growth factors, and stress signals into an “optimal” cellular response (67, 68). Sirt1 activity generally inhibits mTOR signaling (6971), but in murine T cells, ectopic Sirt1 has not been shown to alter this cascade (35). mTOR complex 1 (mTORC1) activity is negatively regulated by tuberous sclerosis complex 1/2 (TSC1/2), which serves as a hub for both positive and negative cues for signaling kinases. Phosphorylation of TSC2 at T1462 by Akt leads to the activation of mTORC1, and in human CD4+ T cells, low-dose resveratrol did not affect the TCR-mediated phosphorylation of this signaling molecule. Furthermore, neither phosphorylation of mTOR itself nor S6 ribosomal protein, a downstream mTOR substrate, was altered by low-dose resveratrol. Note that phosphorylation was significantly decreased in the presence of 100 μM resveratrol (P < 0.01; Fig. 3C and fig. S1G). Thus, mTOR signaling in TCR-stimulated T cells is attenuated by high-dose resveratrol, whereas at low doses, the activity resulting from TCR engagement is maintained.

We next assessed whether distal TCR signaling was altered by resveratrol, monitored as a function of the cell surface abundance of the CD69, CD25 (IL-2Rα subunit), and CD71 (transferrin receptor) activation markers. Surface abundance of CD69, due to the translocation of intracellular stores to the cell membrane without a requirement for protein synthesis (72, 73), was increased in most of the activated cells, irrespective of the presence of resveratrol (Fig. 3D). In marked contrast, induction of CD25 and CD71, both of which are dependent on de novo protein synthesis, was significantly attenuated by high-dose resveratrol but was unaffected by low doses of the polyphenol (P < 0.001; Fig. 3D). Thus, only high-dose resveratrol impedes mTOR and distal TCR signaling cascades.

Low-dose resveratrol stimulates a replication stress response in TCR-stimulated CD4+ T cells

The experiments performed thus far demonstrated that low-dose resveratrol inhibits CD4+ T cell division under conditions in which TCR and mTOR signaling responses are maintained. To further explore this phenomenon and to determine the origin of the cell cycle arrest, we focused on the effects of low-dose resveratrol on genomic integrity. Resveratrol has been found to both positively and negatively affect genome integrity in cancer cells (57, 7478), but its function in primary human T cells has not been elucidated. To specifically address this point in T lymphocytes, we monitored histone H2AX phosphorylation (γH2AX). This modification identifies DNA damage foci as well as stalled replication forks that promote the concentration of repair proteins (79, 80).

TCR engagement of CD4+ T cells did not result in the augmentation of γH2AX (Fig. 2, A and D), at least at time points before entry into S phase (2 to 24 hours; Fig. 4A). However, in low-dose resveratrol, γH2AX+ cells reached significantly higher percentages by 24 hours (45%, P < 0.005; Fig. 4A); these percentages were similar to those detected in the presence of aphidicolin, an inhibitor of replication polymerases that stalls replication forks and results in a late G1-phase arrest (81). High-dose resveratrol had a distinct effect, increasing the abundance of H2AX foci in 7 to 10% of cells, irrespective of the kinetics or TCR stimulation. The lower amount of H2AX phosphorylation in cells treated with high-dose resveratrol may be due to their attenuated response to TCR stimulation (Fig. 3). Notably, stimulation with the homeostatic cytokine IL-7 significantly increased γH2AX in cells treated with high, but not low, doses of resveratrol (P < 0.005; Fig. 4B), suggesting that resveratrol effects on genomic integrity are likely to be dependent on the nature of the activation signal. Whereas it is not known how IL-7 signaling affects the potential of resveratrol to alter genomic integrity or its response to this stress, note that the addition of IL-2 did not alter resveratrol-driven H2AX phosphorylation in TCR-stimulated T cells (Fig. 4C).

Fig. 4 Low-dose resveratrol stimulates H2AX phosphorylation in TCR-stimulated CD4+ T lymphocytes.

(A) Top: Freshly isolated quiescent CD4+ T cells were either left nonstimulated or were TCR-stimulated in the presence of resveratrol (20 and 100 μM) or aphidicolin. The amount of H2AX phosphorylation (γH2AX) was assessed at 24 hours by flow cytometry. Data are representative of four independent experiments. Bottom left: The percentages of γH2AX-positive cells were quantified after 2, 6, 12, and 24 hours of stimulation. Data are representative of four independent experiments. Bottom right: Means ± SEM of γH2AX-positive cells from three independent experiments. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc test. ***P < 0.005. (B) Left: CD4+ T cells were cultured with the homeostatic cytokine IL-7 (10 ng/ml) in the absence or presence of resveratrol (20 and 100 μM). H2AX phosphorylation (γH2AX) was assessed at 24 hours by flow cytometry. Dot plots are representative of three experiments. Right: Means ± SEM of three independent experiments. Data were analyzed by one-way ANOVA with Tukey’s post hoc test. (C) CD4+ T cells were TCR-stimulated in the presence or absence of resveratrol (20 and 100 μM) and in the presence or absence of IL-2 (50 U/ml). H2AX phosphorylation was assessed by flow cytometry as described earlier, and the increase in phosphorylation relative to that in cells stimulated by TCR engagement alone is shown. Data are means ± SEM of three experiments with statistical significance determined by one-way ANOVA with Tukey’s post hoc test. **P < 0.01 and ***P < 0.005.

Resveratrol stimulates p53 phosphorylation and ATR-mediated cell cycle arrest

In light of the data shown earlier, it was critical to monitor activation of the p53 tumor suppressor, a protein that is phosphorylated in response to a wide range of genotoxic insults (60). At day 1 of TCR stimulation, low-dose resveratrol resulted in the accumulation and phosphorylation of p53, similar to that detected under conditions of aphidicolin-mediated stalled replication. Furthermore, by day 3, p53 phosphorylation was also increased under high-dose resveratrol conditions (P < 0.005; Fig. 5A and fig. S1H). Several hypotheses could account for the phosphorylation of p53: first, activation of the adenosine monophosphate–activated protein kinase (AMPK) pathway (8284); second, activation of the ATM pathway in response to double-stranded DNA breaks (60, 85); and third, activation of the ATR pathway in response to single-stranded breaks and replication stress due to stalled fork progression or DNA synthesis (61). Regarding the first hypothesis, glucose deprivation and AMPK activation have been specifically shown to stimulate p53 phosphorylation (82, 83). However, an AMPK inhibitor (compound C) did not significantly alter resveratrol-stimulated p53 phosphorylation, although it decreased AMPK phosphorylation by a mean of 75% (Fig. 5B and fig. S1I). These data suggest that AMPK signaling does not drive p53 phosphorylation in resveratrol-treated CD4+ T lymphocytes.

Fig. 5 Resveratrol stimulates p53 phosphorylation and the ATR-mediated arrest of the cell cycle.

(A) Phosphorylation of p53 at Ser15 and total p53 were monitored by Western blotting analysis of CD4+ T cells activated in the absence or presence of resveratrol (20 and 100 μM) or aphidicolin (1 μM). Representative blots at days 1 and 3 of activation and loading controls (LC) are shown. Data are representative of three experiments; quantification data are shown in fig. S1. (B) Phosphorylation of AMPK and p53 was assessed by Western blotting analysis of CD4+ T cells at day 1 after treatment as indicated with resveratrol and the AMPK inhibitor (compound C; 1 μM). Representative blots showing phosphorylated and total proteins under the indicated conditions are shown. Data are representative of three experiments; quantification data are shown in fig. S1. (C) Schematic model of the ATR and ATM signaling cascades culminating in cell cycle arrest. The former results in Chk1 activation, whereas the latter proceeds through Chk2 activation and the p53-mediated expression of p21. p21 directly inhibits Cdks, whereas Chk1 blocks cell cycle progression by activating Wee1 and preventing the Cdc25-mediated dephosphorylation of Cdk1. (D) The abundance and phosphorylation of Chk1 and Chk2 and the cell cycle regulators Wee1, Cdk1, Cdk2, p21, and Mcm2 were monitored by Western blotting analysis of cells treated under the indicated conditions. Blots of samples at days 1 and 3 are representative of three experiments. The arrow indicates hyperphosphorylated Cdk1. Quantification data are shown in fig. S1. (E) Phosphorylation of Mcm2 and p53 in cells was assessed by Western blotting analysis with the appropriate phosphospecific antibodies at day 1 after treatment with resveratrol and the ATR inhibitor (VE-821; 1 and 5 μM). The amounts of total p53 and γ-tubulin were assessed. Data are representative of three independent experiments. Quantification data are shown in fig. S1. (F) Top: H2AX phosphorylation (γH2AX) in response to resveratrol treatment was assessed at 24 hours after stimulation in the absence or presence of VE-821 (5 μM). Plots show the percentages of γH2AX-positive cells and are representative of three independent experiments. Bottom: Quantification of the MFI of γH2AX staining is shown as means ± SEM of three experiments. *P < 0.05 and **P < 0.01 by two-way ANOVA with Bonferroni’s post hoc test.

To address the second and third hypotheses, we assessed the implication of the ATM and ATR pathways in the response of CD4+ T cells to resveratrol. Classically, ATR is stimulated by replication stress or single-stranded DNA breaks, resulting in the activation of checkpoint kinase 1 (Chk1) (86) and its downstream target, the kinase Wee1, a negative mitotic regulator (Fig. 5C) (85). Although Chk1 activation is often underestimated because of its transient phosphorylation (87), resveratrol stimulated both Chk1 and Wee1 phosphorylation (Fig. 5D and fig. S1J). Chk1 and Wee1 block Cdk1 activation by inhibiting Cdc25 phosphatase and hyperphosphorylating Cdk1, respectively (Fig. 5C). Accordingly, ATR signaling in response to low-dose resveratrol was associated with a defective dephosphorylation of Cdk1 that is required for Cdk1 activation and mitotic entry (Fig. 5D, arrow). In contrast, the ATM pathway was not activated by low-dose resveratrol because neither phosphorylation of Chk2 nor the downstream Cdk inhibitor p21Waf1/Cip1 (p21) was detected (Fig. 5D and fig. S1J). As positive controls, we assessed the capacity of bleomycin and aphidicolin to efficiently activate ATM/ATR and ATR pathway intermediates, respectively (Fig. 5, C and D).

These data suggested that the early H2AX phosphorylation in resveratrol-treated CD4+ T cells was due to a replication stress–like insult rather than an ATM signaling cascade stimulated by double-stranded DNA breaks. To test this possibility, we assessed the phosphorylation of the minichromosome maintenance (MCM) helicase complex, a key component of the prereplication complex that is specifically phosphorylated by ATR at Ser108 in response to multiple forms of DNA damage (8890). Before the onset of S phase (24 hours), Mcm2 phosphorylation at S108 was augmented by low-dose resveratrol to a similar extent to that detected in response to aphidicolin and bleomycin (Fig. 5D and fig. S1J). Furthermore, although T cells exposed to high-dose resveratrol never progressed to S phase (Fig. 2, A and D), p53, Chk1, Wee1, and Mcm2 phosphorylation were also detected in these cells by 72 hours after stimulation (Fig. 5D). Together, these data suggest that resveratrol activates the ATR, but not ATM, signaling cascade in stimulated CD4+ T cells.

We therefore assessed whether the resveratrol-stimulated phosphorylation of p53 was directly regulated by ATR signaling in resveratrol-treated T cells. To this end, we tested the effects of VE-821, a potent ATR inhibitor (91, 92). As expected, VE-821 decreased Mcm2 phosphorylation (Fig. 5E and fig. S1K). Moreover, p53 phosphorylation was attenuated by a mean of 90% in the presence of VE-821, and the global amount of p53 decreased (P < 0.001; Fig. 5E and fig. S1K), demonstrating that the ATR stress response pathway regulates the resveratrol activation of p53 in TCR-stimulated CD4+ T cells. We also found that the ATR pathway is directly implicated in the formation of γH2AX foci (61) as VE-821 inhibited resveratrol activation of H2AX phosphorylation (P < 0.05; Fig. 5F). Together, these data reveal a critical role for the ATR cascade in mediating a p53-associated stress response in response to resveratrol.

CD4+ T cells exhibit increased expression of p53-dependent target genes and undergo a metabolic switch in response to low-dose resveratrol

To elucidate potential molecular mechanisms associated with resveratrol treatment of CD4+ T cells, we performed an array analysis of genes involved in DNA damage signaling responses. Whereas the expression of multiple genes was altered after treatment with high-dose resveratrol or aphidicolin, only 1 of 84 assessed genes, Bbc3 [PUMA (p53 up-regulated modulator of apoptosis)], was consistently increased in expression in low-dose resveratrol–treated CD4 T cells (4.1- to 6.4-fold, n = 3; Fig. 6A and fig. S3). Furthermore, it was even more highly expressed at high-dose resveratrol (8- to 22-fold, n = 3; fig. S3). Notably, PUMA is a proapoptotic gene whose transcription is directly regulated by p53 (93). Because the p53 pathway has also been linked to a metabolic reprogramming (9496), at least in part through the regulation of metabolic genes, we assessed whether expression of these genes is altered by resveratrol. Notably, expression of TIGAR (TP53-induced glycolysis and apoptosis regulator), PGM (phosphoglycerate mutase), GLS2 (glutaminase 2), and SCO2 (synthesis of cytochrome c oxidase 2) was significantly altered by resveratrol (Fig. 6B). p53-mediated induction of TIGAR, a fructose-2,6-bisphosphatase, and attenuated amounts of PGM, would both be expected to decrease glycolysis (9799). Moreover, we found that whereas TCR engagement significantly increased the cell surface abundance of the glucose transporter Glut1, the amount was significantly lower in the presence of low-dose resveratrol and was abrogated by high-dose resveratrol (P < 0.05 and P < 0.001; Fig. 6C and fig. S1L). The changes correlated directly with glucose uptake and glycolysis, as monitored by the production of lactate and extracellular acidification (P < 0.001; Fig. 6D).

Fig. 6 TCR-stimulated CD4+ T cells exhibit increased transcription of p53-dependent metabolic target genes and an altered T cell metabolism after exposure to resveratrol.

(A) The expression of 84 genes involved in DNA damage signaling pathways was evaluated in CD4+ T cells activated by TCR engagement alone as compared to TCR engagement in the presence of resveratrol (20 μM) using a polymerase chain reaction (PCR) array profile (Qiagen; fig. S3). Representative data in one of three samples at day 1 of stimulation are shown with only Bbc3 (PUMA) significantly induced in the latter conditions (4.2- to 6.4-fold induction, n = 3). (B) Transcripts of p53 metabolic target genes including TIGAR, PGM, GLS2, and SCO2 were monitored by quantitative reverse transcription PCR analysis of cells under the indicated conditions (day 3) and were normalized against the abundance of 18S ribosomal RNA. Data are means ± SEM of four independent experiments. *P < 0.05 and **P < 0.01 by paired t test. (C) Glut1 surface expression was monitored by flow cytometry, and representative histograms from three experiments under the indicated conditions (day 3) are shown, with quantifications shown in fig. S1. (D) 2-Deoxy-d[1-3H]glucose (2DG) uptake (n = 3), lactate production (n = 5), and extracellular acidification (n = 5) were monitored in cells under the indicated conditions at day 3. Data are means ± SEM of the indicated number of experiments. *P < 0.05, **P < 0.01, and ***P < 0.005 by one-way ANOVA with Tukey’s post hoc test. (E) Left: ASCT2 surface expression was monitored by flow cytometry, and representative histograms are shown. Middle: The ratio ± SEM of ASCT2 to Glut1 abundance in cells stimulated through the TCR in the presence or absence of resveratrol are presented as a function of the MFI of both transporters (n = 3; paired t test). Right: Uptake of l-2,3,4-[3H]glutamine was performed for 10 min, and mean counts per minute ± SEM for triplicate samples from three independent experiments at day 3 are presented. Data were analyzed by one-way ANOVA with Tukey’s post hoc test. **P < 0.01. (F) Left: Cellular respiration was monitored on a Seahorse XF-24 analyzer, and OCRs of triplicate samples under basal conditions and in response to the indicated mitochondrial inhibitors are presented for cells on day 3 of activation. Mean basal consumption rates (OCR; picomoles/min per 106 cells) ± SEM of triplicate samples from three independent experiments are shown (upper right). Right: ATP was measured in cells under the same conditions by luminescent detection, and mean intracellular amounts ± SEM from data obtained in 10 independent experiments are presented. Data were analyzed by one-way ANOVA with Tukey’s post hoc test. Mitochondrial superoxide anion was monitored using MitoSOX Red reagent, and the MFI ± SEM of triplicate samples are presented. Data were analyzed by paired t test. FCCP, carbonyl cyanide p-trifluoromethoxyphenylhydrazone; Ant.A, antimycin A; Rot., rotenone. *P < 0.05, **P < 0.01, and ***P < 0.005.

Glucose metabolism was decreased in the presence of resveratrol, but lymphocyte metabolism can also be fueled by other nutrients such as glutamine. Notably, the ASCT2 glutamine transporter, recently shown to be critical for T cell activation (100), was augmented in the presence of low-dose, but not high-dose, resveratrol. Moreover, the ratio of surface ASCT2/Glut1 was significantly increased by low-dose resveratrol as compared to TCR activation alone (P < 0.01; Fig. 6D). The potential importance of resveratrol-linked glutaminolysis in CD4+ T cells was also suggested by the induction of the p53-dependent GLS2 gene (Fig. 6B), catalyzing the hydrolysis of glutamine to glutamate (101103). This link was corroborated by an increased glutamine uptake in activated T cells treated with low-dose resveratrol (Fig. 6E).

The increase in glutamine entry and the enzymes involved in the first steps of glutaminolysis suggested that p53 activation might augment OXPHOS in these T lymphocytes. In tumor cells, p53 decreases glycolysis while enhancing OXPHOS (104, 105). One of the mechanisms through which p53 augments tricarboxylic acid (TCA) cycling from glutamine intermediates (anaplerosis) is through the expression of SCO2, a protein that catalyzes the transfer of reducing equivalents from cytochrome c to molecular oxygen in mitochondria (103). To assess the extent of OXPHOS in human CD4+ T cells, we directly monitored the oxygen consumption rate (OCR) and found that TCR stimulation significantly increased the basal cellular respiration of CD4+ T cells (P < 0.005; Fig. 6F). Low-dose resveratrol further enhanced respiration, but TCR-stimulated respiration was almost completely abrogated by high-dose resveratrol (P < 0.05; Fig. 6F). Increased oxygen consumption would also be expected to result in increased ROS, with mitochondria serving as the major intracellular source of ROS. Superoxide anion—the predominant ROS in mitochondria—was significantly augmented by resveratrol, as monitored by MitoSOX staining (P < 0.05; Fig. 6F). These metabolic parameters correlated directly with the bioenergetic profile of the CD4+ T cells; intracellular adenosine triphosphate (ATP) was significantly increased in low-dose resveratrol–treated cells but was attenuated in lymphocytes treated with high-dose resveratrol (P < 0.005; Fig. 6F). Thus, resveratrol significantly modulated the expression of p53 target genes, concordant with changes in the metabolic state of these lymphocytes.

Altered metabolism in resveratrol-treated naïve and memory CD4+ T cells enhances their effector function

In recent years, the critical importance of metabolism (and most specifically glucose metabolism) in T cell effector function has been established (37, 106108). However, entry of glutamine via the ASCT2 amino acid transporter (SLC1A5) has been shown to be a sine qua non for the effector function of murine CD4+ T cells (100, 109). Furthermore, although it has not been specifically shown for T cells, a heightened need for biosynthetic intermediates is often associated with a metabolic switch resulting in a disproportionate dependency on glutamine, undergoing anaplerotic reactions to form α-ketoglutarate for use in the TCA cycle (110112). It was therefore of interest to assess whether CD4+ T cell effector function is altered in the presence of resveratrol wherein biosynthetic intermediates were not required for proliferation, due to the block in cell cycle, and a switch to an oxidative metabolism was associated with increased intracellular ATP stores (Fig. 6).

T cell effector function is tightly linked to the differentiation state of the lymphocyte, and different subsets of memory T helper cells can be distinguished on the basis of chemokine receptor markers [reviewed in (113)]. It was therefore important to determine whether resveratrol potentially changed the survival of sorted naïve and memory T cell subsets (fig. S4A) or, rather, whether it altered the effector profile of memory T helper cells. We therefore monitored the surface abundance of CXCR3, CCR4, and CCR6 on memory CD4+ T cells (fig. S4B). Although CXCR3+ and CCR4+CCR6+ profiles are associated with TH1 and TH17 subsets, respectively, we found that CXCR3 and CCR4 were increased in most of the CD4+ T cells after TCR stimulation (fig. S4B). Notably, the relative percentage of CXCR3+CCR6+ cells within the CD4+ T cell subset was 1.7 ± 0.2–fold higher in the presence of low-dose resveratrol (P < 0.01; fig. S4B), suggesting a higher cytokine secretion potential. The percentages of naïve and memory CD4+ T cells secreting IFN-γ after TCR stimulation (day 6) was greater than four- and twofold higher in the presence of low-dose resveratrol, respectively (P < 0.005; Fig. 7A). Furthermore, most of the T cells that secreted IFN-γ also produced IL-2 (P < 0.005; Fig. 7A). The increased capacity of resveratrol-treated T cells to secrete cytokines was also detected in the total CD4+ population (fig. S4C). Thus, low-dose resveratrol stimulates an ATR-mediated cell cycle arrest in antigen receptor–stimulated CD4+ T lymphocytes that is coupled to a metabolic reprogramming and augmented effector function.

Fig. 7 IFN-γ secretion by TCR-stimulated naïve and memory CD4+ T cells is markedly enhanced by low-dose resveratrol.

(A) Left: Naïve and memory CD4+ T cells were isolated, as described in fig. S4A, and were stimulated through the TCR in the presence or absence of resveratrol (20 and 100 μM). Secretion of IL-2 and IFN-γ was monitored by intracellular staining at day 6, and representative dot plots of three representative experiments are shown. Right: Quantification of the mean percentages ± SEM of IFN-γ–secreting and double IL-2/IFN-γ–secreting cells are shown. Data were analyzed by two-way ANOVA with Bonferroni’s post hoc test. (B) Proposed model showing the effects of resveratrol on TCR-stimulated CD4+ T cells. Left: Sirt1 and p53 are interrelated, regulating mTOR signals and metabolic networks (115), and both are activated in response to TCR stimulation of human CD4+ T cells. The integration of TCR signals stimulates intracellular glycolysis and glutaminolysis, resulting in proliferation and effector function. Middle: In response to low-dose resveratrol, TCR-engaged CD4+ T cells undergo a genomic stress response, which results in an ATR- and Chk1-mediated S-G2 cell cycle arrest. Moreover, ATR-mediated p53 signaling decreases glycolysis and increases glutaminolysis. Under these conditions, wherein cell cycle progression is blocked and OXPHOS is augmented, there is a substantial increase in IFN-γ secretion. Right: In response to high-dose resveratrol, TCR-mediated mTOR and Sirt1 signaling pathways are markedly attenuated, leading to p27-mediated G1 cell cycle arrest.

DISCUSSION

Our study demonstrates that resveratrol modulates the potential of human CD4+ T lymphocytes to respond to antigen receptor stimulation. Although resveratrol exerts effects in both Sirt1-dependent and Sirt1-independent manners (20, 114), we found that high-dose resveratrol (100 μM) inhibited the TCR-induced expression of Sirt1 in T lymphocytes. In this condition, this was likely caused by an attenuation of mTOR and distal TCR signaling. In contrast, low-dose resveratrol (20 μM) markedly increased TCR-stimulated Sirt1 activity. Furthermore, p53—a tumor suppressor whose activity is coordinated by and coordinates that of Sirt1 (115)—was highly phosphorylated in response to low-dose resveratrol. We determined that this p53 phosphorylation was mediated by the kinase ATR, a key regulator of the genotoxic stress response pathway. Concordant with ATR and p53 signaling, T cells exposed to low-dose resveratrol underwent Chk1- and Wee1-mediated premitotic cell cycle arrest and induced expression of p53-dependent metabolic target genes, resulting in a metabolic shift with increased OXPHOS. Note that these conditions, which promoted an enhanced bioenergetic profile, endowed CD4+ T cells with a substantially enhanced cytokine secretion potential (Fig. 7B).

The role of resveratrol in protecting against carcinogenesis has been the subject of intense study, and multiple reports showed that resveratrol functions by preventing DNA damage formation as well as by improving DNA damage repair (47, 50, 116). Resveratrol affects multiple aspects of DNA metabolism, including DNA replication, recombination, repair, and telomere maintenance, as well as the redox state, thereby promoting the integrity of genomic DNA. However, in vitro, resveratrol mediates DNA cleavage in a process requiring DNA-bound copper [Cu(II)] ions (49, 57, 59, 117119). On the basis of diverse studies, it is nearly impossible to draw clear-cut conclusions about the effects of sirtuins or resveratrol on genomic stability. Both reduction of DNA breaks and an inhibition of replicative senescence (57, 7478), as well as the generation of DNA breaks with associated senescence (57, 76, 77), have been reported (120). Here, we found that low-dose resveratrol triggered a marked DNA damage response in TCR-stimulated T cells, as shown by the presence of γH2AX in 20 to 50% of cells. Furthermore, DNA damage, or more precisely the response to genotoxic stress, was linked to the activation state of the T cell. Low-dose resveratrol was associated with H2AX phosphorylation in TCR-stimulated cells, whereas high-dose resveratrol resulted in H2AX phosphorylation in T cells exposed to the homeostatic cytokine IL-7.

The ATR pathway is activated under conditions of single-stranded DNA breaks or instability of replication forks (85). We find that low-dose resveratrol activated the ATR pathway within 24 hours of the treatment of stimulated CD4+ T cells, well before S-phase entry. Under these conditions, ATR rapidly phosphorylated histone H2AX and Mcm2 at Ser108, with the latter potentially stabilizing prereplication complexes in response to DNA damage (121). In addition, at later time points, we found that resveratrol stimulated the ATR-mediated phosphorylation of Chk1 and Wee1, blocking Cdc25-mediated Cdk1 activation and mitotic entry (Fig. 7B). Although both ATR and ATM phosphorylate Ser15 of p53 (60), we found no evidence that resveratrol activated an ATM-Chk2-p53-p21 pathway, a pathway that is generally activated in response to double-stranded DNA breaks. Thus, our data suggest that resveratrol triggers a replication stress–like response rather than classical DNA damage. In this state, p53 appears to serve as a node between upstream stress signaling cascades and downstream DNA repair pathways (Fig. 7B) (122).

In CD4+ T cells, both low- and high-dose resveratrol induced transcription of the p53 proapoptotic target, PUMA. This was the only gene in an 84-gene DNA damage signaling pathway array to exhibit increased expression in response to low-dose resveratrol (fig. S3), suggesting that many of the effects of low-dose resveratrol occur at a posttranscriptional level. However, low-dose resveratrol altered the expression of all p53-directed metabolic gene targets that we assessed. Although other transcription factors, such as c-Myc and HIF-1α, also regulate cell metabolism and are produced in TCR-stimulated human CD4+ T cells, their abundance was not substantially modulated by resveratrol (fig. S5). Thus, we focused on p53, a tumor suppressor whose function has paradoxically been found to protect tumor cells from modest amounts of stress through metabolic reprogramming (95, 123). The p53-mediated decreases in Glut1 and PGM abundance, together with increased TIGAR, attenuate aerobic glycolysis, whereas increased SCO2 and GLS2 abundance drives glutamine-driven OXPHOS (97, 101103, 124), which was the case that we observed in resveratrol-treated CD4+ T cells. Note that cell cycle arrest and senescence act as signals for a cell to undergo metabolic reprogramming, decreasing glycolysis and increasing TCA cycle usage (96, 125127). Thus, our data suggest that the ATR-mediated cell cycle arrest initiated by low-dose resveratrol in CD4+ T cells was coupled to a metabolic shift, adjusting the balance between glycolysis and OXPHOS (Fig. 7B).

The bioenergetic profile of resveratrol-treated T cells was altered by this skewing of metabolism away from glycolysis and toward a setting characterized by an increased ASCT2-to-Glut1 ratio, with an augmented glutamine transport, substantially increased mitochondrial ROS production, and increased OXPHOS. How this would affect T cell effector function is unclear because glycolysis can increase and decrease the potential of T cells to secrete effector cytokines, such as IFN-γ and IL-17 (128133). Note that amino acid metabolism is essential for effector T cell differentiation (100, 109, 134, 135), and memory cells rely more on OXPHOS than on glycolysis (38, 130). In both naïve and memory CD4+ T cells, low-dose resveratrol markedly augmented the amount of IFN-γ secreted, but this increase was even higher for naïve cells than for memory cells (means of 10- and 2-fold, respectively). Furthermore, in both cell types, resveratrol substantially increased the number of cells that produced both IL-2 and IFN-γ. Thus, our data suggest that resveratrol is an agent that, by altering the metabolic fitness of T lymphocytes, enhances their cytokine effector potential.

Adjusting the balance between glycolysis and OXPHOS can also have substantial effects in other cell types. Decreasing OXPHOS in mice expressing a mutant p53 markedly attenuates tumorigenesis (136). Thus, generating a context that is the converse of that shaped by resveratrol, that is, inhibiting a p53-mediated shift to mitochondrial metabolism, may be beneficial for individuals with an increased risk of developing cancers, such as Li-Fraumeni syndrome patients with germline mutations in the TP53 gene. The potential use of resveratrol as a therapy for the treatment of neurological, cardiovascular, hepatic, and metabolic pathologies therefore necessitates a critical evaluation of its effect on T lymphocytes in vivo, especially in an autoimmune setting. The data shown here reveal a complex network of resveratrol-stimulated changes in cell cycle progression and metabolism, altering the potential of T lymphocytes to respond to foreign antigens.

MATERIALS AND METHODS

T cell isolation and culture

CD4+ T cells were isolated from adult peripheral blood, obtained from healthy donors after informed consent. Cells were purified using negative-selection Rosette tetramers (STEMCELL Technologies), and the purity of the cell population was monitored on a FACSCanto II (BD Biosciences). Purities were always greater than 94%. Naïve and memory CD4+ T cells were sorted on a FACSAria after staining with anti-CD4, anti-CD45RA, anti-CD45RO, CD62L, CD127, and CD25 antibodies (fig. S4A). Lymphocytes (1 × 106 per well in a 24-well plate) were cultured in RPMI 1640 + GlutaMAX (Gibco, Life Technologies) supplemented with 10% fetal calf serum (FCS) and 2% penicillin/streptomycin (Gibco, Life Technologies). For TCR stimulation, 24-well plates were coated with anti-CD3 (clone OKT3, BioLegend) and anti-CD28 (clone 9.3, provided by C. June) monoclonal antibodies (mAbs) at a concentration of 1 μg/ml, and recombinant IL-2 (rIL-2) (50 U/ml) was added as indicated. T cells were also cultured in the presence of rIL-7 (10 ng/ml). As indicated, resveratrol (20 or 100 μM; Sigma-Aldrich), compound C1 (1 μM; Sigma-Aldrich), VE-821 (1 and 5 μM; Euromedex), aphidicolin (1 μM), and bleomycin (1 μM) were added to T cell cultures 1 hour before TCR stimulation.

Immunofluorescence

Cells were collected and coated on poly-l-lysine–treated slides. Cells were fixed in a 4% paraformaldehyde (PFA) solution [phosphate-buffered saline (PBS), 4% PFA] at 37°C for 15 min, permeabilized in PBS containing 3% bovine serum albumin (BSA)/0.1% saponin for 10 min, and blocked for nonspecific protein binding with 10% FCS. Staining with primary anti-Sirt1 antibody (Ab) (Santa Cruz Biotechnology) and a secondary Alexa Fluor 488–coupled anti-rabbit immunoglobulin G (Invitrogen) was performed in PBS containing 3% BSA for 1 hour at room temperature. Nuclei were then labeled by DAPI staining for 10 min at room temperature.

Flow cytometric analyses

To detect cell surface markers, cells were incubated with the appropriate fluorochrome-conjugated mAbs, and expression was monitored in comparison with isotype controls. Antibodies against CD4, CD25, CD69, and CD71 were from Beckman Coulter. Y319-phosphorylated ZAP-70 (BD Biosciences), T202/Y204-phosphorylated ERK1/2 (BD Biosciences), and phosphorylated H2AX (BioLegend) were detected after cell fixation and permeabilization. Surface Glut1 and ASCT2 were detected by binding to their respective retroviral envelope ligands fused to enhanced green fluorescent protein or recombinant rabbit fragment crystallizable (rFc) (Metafora Biosystems), as previously described (72, 137139). The presence of mitochondrial superoxide was assessed by staining with MitoSOX Red indicator (1 μM; Invitrogen). Proliferation was monitored as a function of carboxyfluorescein diacetate succinimidyl ester (Invitrogen) or violet proliferation dye (Invitrogen) dilution. Before staining for intracellular IFN-γ and IL-2 (BD Biosciences), cells were activated with phorbol 12-myristate 13-acetate (100 ng/ml; Sigma-Aldrich) and ionomycin (1 μg/ml; Sigma-Aldrich) in the presence of brefeldin A (10 μg/ml; Sigma-Aldrich) for 3.5 to 4 hours at 37°C. Cell cycle analysis was performed by simultaneous staining for DNA and RNA using 7-aminoactinomycin D (20 μM; Sigma-Aldrich) and pyronin Y (5 μM; Sigma-Aldrich), respectively. Cells were assessed on a FACSCanto II or BD LSR II Fortessa (BD Biosciences), and data were analyzed using FACSDiva (BD Biosciences) or FlowJo (Tree Star) software.

Metabolic assays

OCRs were measured on an XF-24 Extracellular Flux Analyzer (Seahorse Bioscience). TCR-stimulated T cells with and without low-dose resveratrol (20 μM) were seeded at a concentration of 1.5 × 106 cells, whereas nonstimulated and high-dose resveratrol-stimulated cells were seeded at a concentration of 2.0 × 106 cells in XF medium (nonbuffered Dulbecco’s modified Eagle’s medium containing 2.5 mM glucose, 2 mM l-glutamine, and 1 mM sodium pyruvate). Oxygen consumption was monitored under basal conditions and in response to oligomycin (1 μM), carbonyl cyanide p-trifluoromethoxyphenylhydrazone (1.5 μM), rotenone (100 nM), and antimycin A (1 μM; Sigma-Aldrich). The basal respiration rate was calculated as the difference between basal OCR and the OCR after inhibition of mitochondrial complexes 1 and 3 with rotenone and antimycin A, respectively. ATP and l-lactate were measured according to the standard procedures of the ATPlite kit (PerkinElmer) and l-lactate kit (Eton Bioscience), respectively. Extracellular pH was measured immediately after harvesting of medium using a standard pH meter.

Glucose and glutamine uptake assays

Cells (2 × 106) were starved by incubation at 37°C in serum and glucose- or glutamine-free RPMI 1640 for 30 min. Radiolabeled 2-deoxy-d-[1-3H] glucose or glutamine-l-[3,4-3H(N)] (PerkinElmer) was added to a final concentration of 0.1 mM (2 μCi/ml). Cells were incubated for 10 min at room temperature, washed in cold serum/glucose/glutamine-free RPMI 1640, and solubilized in 500 μl of 0.1% SDS. Radioactivity was measured by liquid scintillation.

Total protein extraction and analyses

Cells were lysed in lysis buffer containing 20 mM Hepes (pH 7.6), 100 mM KCl, 0.1 mM EDTA, 1 mM NaF, 1 mM sodium orthovanadate, 1% Triton X-100, 0.5% NP-40, 0.15 mM spermine, 0.5 mM spermidine, 1 mM dithiothreitol, and a protease inhibitor cocktail. After a 30-min incubation on ice, extracts were centrifuged, and supernatants were harvested. Extracts (20 μg) were resolved on SDS–polyacrylamide gel electrophoresis gels (8.5 to 12%) and transferred electrophoretically onto polyvinylidene difluoride (PVDF) membranes. PVDF membranes were incubated with the indicated antibodies (table S1) for 1 hour at room temperature or overnight at 4°C and with horseradish peroxidase–conjugated anti-goat, anti-rabbit, or anti-mouse secondary Abs, and immunoreactive proteins were visualized using enhanced chemiluminescence (ECL, Amersham) according to the manufacturer’s instructions. Proteins were quantified with ImageJ software and normalized to amido black–stained proteins (loading controls) or tubulin as indicated.

Gene expression analysis by PCR array

Total RNA was isolated with the RNeasy Mini Kit (Qiagen) and was reverse-transcribed with QuantiTect Reverse Transcription Kit (Qiagen). SYBR Green–based (SYBR Green I Master, Roche) real-time quantitative PCR (qPCR) for TIGAR, PGM, GLS2, SCO2, HIF1A, and RNA18S was performed with the LightCycler 480 Real-Time PCR System (Roche), and all primers are shown in table S2. To determine relative expression, samples for each experimental condition were run in duplicate and were normalized to 18S. Primer sequences used for amplification were designed with Primer3 and NetPrimer (PREMIER Biosoft) software packages. The expression of 84 genes involved in DNA damage signaling pathways was analyzed with the RT2 Profiler PCR Array (SABiosciences, Qiagen) according to the manufacturer’s instructions. Briefly, RNA was prepared from 5 million cells after 24 hours in culture under the indicated conditions using the Qiagen RNeasy kit. First-strand complementary DNA was then prepared and used in the PCR array in combination with SYBR Green qPCR master mixes on a Roche Light Cycler 480. Data were normalized to ACTB (actin beta), B2M2-microglobulin), and RPLP0 using the SABiosciences DNA template analysis software.

Statistical analyses

Data were analyzed with GraphPad Software (GraphPad Prism), and P values were calculated by one-way ANOVA (Tukey’s post hoc test), two-way ANOVA (with Tukey’s post hoc or Bonferroni’s tests), or paired t tests, as indicated.

SUPPLEMENTARY MATERIALS

www.sciencesignaling.org/cgi/content/full/10/501/eaal3024/DC1

Fig. S1. Quantification and statistical analyses of main data panels.

Fig. S2. Effects of resveratrol on the formation of CD4+ T cell blasts and cell counts in response to TCR engagement.

Fig. S3. Effect of resveratrol treatment on the expression of DNA damage signaling pathway genes.

Fig. S4. Sorting of naïve and memory CD4+ T cells for the assessment of phenotype and cytokine secretion profiles.

Fig. S5. TCR-mediated induction of HIF1A and c-Myc in human CD4+ cells is not altered by low-dose resveratrol.

Table S1. Antibody list.

Table S2. Primer sequences.

REFERENCES AND NOTES

Acknowledgments: We thank all members of our laboratories for discussions, scientific critique, and continual support. We are grateful to Montpellier RIO Imaging for support in cytometry experiments and the RAM (Animal House Network of Montpellier) animal facility of the Institut de Génétique Moléculaire de Montpellier. We are indebted to T. Gostan of the Service d'Analyse de Données Biologiques Complexes platform for data and biostatistical analyses and to M. Sitbon and A. Singer for critical discussions and reading of the manuscript. Funding: M.C. was supported by a fellowship from the Portuguese Foundation for Science and Technology. G.C. was supported by a fellowship from the Ligue Contre le Cancer. M.I.M. received funding from the French Ministry of Education. C.M., V. Dardalhon, V.S.Z. and V. Dulić are supported by CNRS. N.T. is supported by INSERM. This work was supported by funding from the Association de la Recherche contre le Cancer, French national (Agence Nationale de la Recherche) research grants (PolarATTACK and GlutStem), Institut National du Cancer, the European Community (contracts LSHC-CT-2005-018914 “ATTACK” and PIRG5-GA-2009-249227 “T cell homeostasis”), and the French laboratory consortiums (Labex) EpiGenMed and GR-Ex and support by the Labex EpiGenMed and GR-Ex programs, reference ANR-10-LABX-12-01 and ANR-11-LABX-0051. Author contributions: M.C., G.C., C.M., V. Dulić, and N.T. conceived the project and designed the experiments. M.C., G.C., C.M., M.I.M., V.S.Z., V. Dardalhon, and V. Dulić executed the experiments, and all authors contributed to data analyses and interpretations of the results. M.C., G.C., C.M., V. Dulić, and N.T. wrote the manuscript, and all authors added critical reviews and modifications. Competing interests: The authors declare that they have no competing interests.
View Abstract

Navigate This Article