Pharmacological preconditioning with resveratrol: role of CREB-dependent Bcl-2 signaling via adenosine A3 receptor activation

Samarjit Das, Gerald A. Cordis, Nilanjana Maulik, Dipak K. Das

Abstract

Recent studies demonstrated that resveratrol, a grape-derived polyphenolic phytoalexin, provides pharmacological preconditioning (PC) of the heart through a NO-dependent mechanism. Because adenosine receptors play a role in PC, we examined whether they play any role in resveratrol PC. Rats were randomly assigned to groups perfused for 15 min with 1) Krebs-Henseleit bicarbonate buffer (KHB) only; 2) KHB containing 10 μM resveratrol; 3) 10 μM resveratrol + 1 μM 8-cyclopentyl-1,3-dimethylxanthine (CPT; adenosine A1 receptor blocker); 4) 10 μM resveratrol + 1 μM 8-(3-chlorostyryl)caffeine (CSC; adenosine A2a receptor blocker); 5) 10 μM resveratrol + 1 μM 3-ethyl-5-benzyl-2-methyl-4-phenylethynyl-6-phenyl-1,4-(±)-dihydropyridine-3,5-dicarboxylate (MRS-1191; adenosine A3 receptor blocker); or 6) 10 μM resveratrol + 3 μM 2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride [LY-294002, phosphatidylinositol (PI)3-kinase inhibitor], and groups perfused with adenosine receptor blockers alone. Hearts were then subjected to 30-min ischemia followed by 2-h reperfusion. The results demonstrated significant cardioprotection with resveratrol evidenced by improved ventricular recovery and reduced infarct size and cardiomyocyte apoptosis. CPT and MRS 1191, but not CSC, abrogated the cardioprotective abilities of resveratrol, suggesting a role of adenosine A1 and A3 receptors in resveratrol PC. Resveratrol induced expression of Bcl-2 and caused its phosphorylation along with phosphorylation of cAMP response element-binding protein (CREB), Akt, and Bad. CPT blocked phosphorylation of Akt and Bad without affecting CREB, whereas MRS 1191 blocked phosphorylation of all compounds, including CREB. LY-294002 partially blocked the cardioprotective abilities of resveratrol. The results indicate that resveratrol preconditions the heart through activation of adenosine A1 and A3 receptors, the former transmitting a survival signal through PI3-kinase-Akt-Bcl-2 signaling pathway and the latter protecting the heart through a CREB-dependent Bcl-2 pathway in addition to an Akt-Bcl-2 pathway.

  • Akt
  • cAMP response element-binding protein
  • Bad
  • apoptosis

preconditioning (PC)-induced tolerance against myocardial ischemia-reperfusion injury has been the subject of intensive investigation (13, 34). PC may be achieved by preexposing the heart to a repetitive short exposure of ischemia or hypoxia followed by reperfusion or reoxygenation (ischemic or hypoxic PC) (36, 12), heat stress (30), oxidative stress (7), or chemical agents (pharmacological PC) (28, 43).

3,4′,5-Trihydroxy-trans-stilbene (resveratrol), a naturally occurring polyphenolic compound found abundantly in grape skins and seeds and in wines, has been found to pharmacologically precondition the heart against ischemia-reperfusion injury (22, 35). Recently, it was found that resveratrol protects a variety of vital organs including kidney, heart, lung, and brain from ischemia-reperfusion injury (14, 18). Resveratrol possesses diverse biochemical and physiological actions for cellular protection, which include estrogenic, antiplatelet, and anti-inflammatory properties (3, 17).

Adenosine is a potent intermediate of myocardial preservation. It shows its effects during the ischemia as well as reperfusion phases through receptor-mediated actions (9). There are four adenosine receptors, A1, A2a, A2b, and A3 (26). Among these, A1 and/or A3 receptors have been found to play a crucial role in cardioprotection by virtue of their abilities to precondition the heart (31, 42). There are conflicting results in the literature when agonists of adenosine A1 and A2a receptors are used to precondition the heart (1, 44). More positive results were found with adenosine A3 receptors. Activation of adenosine A3 receptors reduces the degree of apoptosis produced by ischemia-reperfusion injury in the brain (11) and in the myocardium (39).

The objective of this study was to examine whether the cardioprotective effects of resveratrol are realized through adenosine receptor activation. The results of the present study are consistent with many previous reports that indicate the role of adenosine A1 and A3 receptors, but not A2a receptor, in ischemic PC (33). Initial studies with adenosine receptors implicated a role of adenosine A1 receptor in PC (9). Subsequent studies showed conflicting results, some of them demonstrating negative findings (1, 44). However, recent studies clearly demonstrate involvement of adenosine A3 receptors in PC (33). The results indicate that resveratrol preconditions the heart through the activation of adenosine A1 and A3 receptors, the former transmitting a survival signal through phosphatidylinositol (PI)3-kinase-Akt-Bcl-2 signaling pathway and the latter protecting the heart through a cAMP response element-binding protein (CREB)-dependent Bcl-2 pathway in addition to an Akt-Bcl-2 pathway.

MATERIALS AND METHODS

Resveratrol.

Resveratrol, a natural phytoalexin, was obtained from Sigma (St. Louis, MO). Adenosine A1 receptor antagonist 8-cyclopentyl-1,3-dimethylxanthine (CPT), adenosine A2a receptor antagonist 8-(3-chlorostyryl)caffeine (CSC), and the highly specific blocker of the adenosine A3 receptor 3-ethyl-5-benzyl-2-methyl-4-phenylethynyl-6-phenyl-1,4-(±)-dihydropyridine-3,5-dicarboxylate (MRS- 1191) were also purchased from Sigma. 2-(4-Morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride (LY-294002), a PI3-kinase inhibitor, was obtained from Calbiochem. The drugs were dissolved in DMSO, and the aliquots were frozen at 4°C. Control experiments used the vehicle (DMSO) only.

Animals.

All animals used in this study received humane care in compliance with the principles of laboratory animal care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences and published by the National Institutes of Health (Pub. No. NIH85–23, revised 1985). Sprague-Dawley male rats weighing between 250 and 300 g were fed regular rat chow ad libitum with free access to water until the start of the experimental procedure. The rats were randomly assigned to one of the following groups (Fig. 1) perfused for 15 min with Krebs-Henseleit bicarbonate buffer (KHB) with 1) vehicle only, 2) 1 μM CPT, 3) 1 μM CSC, 4) 1 μM MRS-1191, 5) 3 μM LY-294002, 6) KHB containing 10 μM resveratrol, 7) 10 μM resveratrol + 1 μM CPT, 8) 10 μM resveratrol + 1 μM CSC, 9) 10 μM resveratrol + 1 μM MRS-1191, or 10) 10 μM resveratrol + 3 μM LY-294002. All hearts were then subjected to 30-min ischemia followed by 2-h reperfusion. Control experiments were those performed with vehicle (DMSO) only or CPT, CSC, and MRS-1191 only.

Fig. 1.

Experimental protocol. KHB, Krebs-Henseleit bicarbonate buffer; CPT, 8-cyclopentyl-1,3-dimethylxanthine; CSC, 8-(3-chlorostyryl)caffeine; MRS-1191, 3-ethyl-5-benzyl-2-methyl-4-phenylethynyl-6-phenyl-1,4-(±)-dihydropyridine-3,5-dicarboxylate; LY-294002, 2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride. CREB, cAMP response element-binding protein.

Isolated working heart preparation.

Rats were anesthetized with pentobarbital sodium (80 mg/kg ip; Abbott Laboratories, North Chicago, IL) and administered the anticoagulant heparin sodium (500 IU/kg iv; Elkins-Sinn, Cherry Hill, NJ). After sufficient depth of anesthesia was ensured, thoracotomy was performed and hearts were perfused in the retrograde Langendorff mode at 37°C at a constant perfusion pressure of 100 cmH2O (10 kPa) for a 5-min washout period. The perfusion buffer used in this study consisted of a modified KHB (in mM: 118 NaCl, 4.7 KCl, 1.7 CaCl2, 25 NaHCO3, 0.36 K2HPO4, 1.2 MgSO4, and 10 glucose). The Langendorff preparation was switched to the working mode after the washout period as previously described (12).

At the end of 10 min, after the attainment of steady-state cardiac function, baseline functional parameters were recorded. The circuit was then switched back to the retrograde mode, and hearts were perfused with KHB with vehicle or any of the adenosine receptor blockers (control), resveratrol at a concentration of 10 μM, or a combination of resveratrol and adenosine receptors for a duration of 15 min. This was followed by a 5-min washout with KHB buffer, and then the hearts were subjected to global ischemia for 30 min and then 2 h of reperfusion. Reperfusion was in the retrograde mode for the first 10 min to allow for postischemic stabilization and thereafter in the antegrade working mode to allow for assessment of functional parameters, which were recorded at 10, 30, 60, and 120 min of reperfusion (R-10, R-30, R-60, and R-120).

Cardiac function assessment.

Aortic pressure was measured with a Gould P23 XL pressure transducer (Gould Instrument Systems, Valley View, OH) connected to a sidearm of the aortic cannula; the signal was amplified with a Gould 6600 series signal conditioner and monitored on a CORDAT II real-time data acquisition and analysis system (Triton Technologies, San Diego, CA) (11). Heart rate (HR), left ventricular developed pressure (LVDP; defined as the difference of the maximum systolic and diastolic aortic pressures), and the first derivative of developed pressure (dP/dt) were all derived or calculated from the continuously obtained pressure signal. Aortic flow (AF) was measured with a calibrated flowmeter (Gilmont Instrument, Barrington, IL), and coronary flow (CF) was measured by timed collection of the coronary effluent dripping from the heart.

Infarct size estimation.

At the end of reperfusion, a 10% (wt/vol) solution of triphenyltetrazolium chloride in phosphate buffer was infused into the aortic cannula for 20 min at 37°C (3). The hearts were excised and stored at −70°C. Sections (0.8 mm) of frozen heart were fixed in 2% paraformaldehyde, placed between two coverslips, and digitally imaged with a Microtek Scan Maker 600z. To quantify the areas of interest in pixels, NIH Image 5.1 (a public-domain software package) was used. The infarct size was quantified and expressed in pixels.

Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling assay for assessment of apoptotic cell death.

Immunohistochemical detection of apoptotic cells was carried out with terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) (32). The sections were incubated again with mouse monoclonal antibody recognizing cardiac myosin heavy chain to specifically recognize apoptotic cardiomyocytes. The fluorescence staining was viewed with a confocal laser microscope. The number of apoptotic cells was counted and expressed as a percentage of total myocyte population.

Western blot analysis.

Left ventricles from the hearts were homogenized in a buffer containing (in mM) 25 Tris·HCl, 25 NaCl, 1 mM orthovanadate, 10 mM NaF, 10 mM pyrophosphate, 10 mM okadaic acid, 0.5 mM EDTA, and 1 mM PMSF (36). One hundred micrograms of protein of each heart homogenate was incubated with one microgram of antibody against the phospho-Akt, CREB, Bcl-xl/Bcl-2-associated death promoter (Bad), or Bcl-2 (Santa Cruz Biotechnology, Santa Cruz, CA) for 1 h at 4°C. The immune complexes were precipitated with protein A-Sepharose, and immunoprecipitates were separated by SDS-PAGE and immobilized on polyvinylidene difluoride membrane. The membrane was immunoblotted with PY20 to evaluate the phosphorylation of the compounds. The membrane was stripped and reblotted with specific antibodies against Akt, CREB, Bad, and Bcl-2. The resulting blots were digitized and subjected to densitometric scanning with a standard NIH Image program.

Statistical analysis.

The values for myocardial functional parameters, total and infarct volumes and sizes, and cardiomyocyte apoptosis are all expressed as means ± SE. Analysis of variance was first carried out to test for any differences between the mean values of all groups. If between-group differences were established, the values of the treated groups were compared with those of the control group by a modified t-test. The results were considered significant if P < 0.05.

RESULTS

Effects of resveratrol on myocardial function.

There were no differences in baseline function among 10 groups. In general, there were no significant differences between resveratrol vs. control and also resveratrol + A1, resveratrol + A2a, or resveratrol + A3 versus resveratrol alone in HR and CF (Table 1). As expected, on reperfusion, the absolute values of all functional parameters were decreased in all groups compared with the respective baseline values. The group perfused with resveratrol alone displayed significant recovery of postischemic myocardial function. The cardioprotective effects of resveratrol were evidenced by significant differences in the LVDP from R-30 onward (Table 1); the difference was especially apparent at R-60 and at R-120 and also in dP/dt at R-120. Aortic flow was markedly higher in the resveratrol group from R-30 onward. With the use of adenosine A1 and A3 receptor inhibitors (CPT and MRS-1191, respectively), resveratrol lost its cardioprotective effects, which was evidenced by significant differences in the postischemic period of LVDP from R-30 onward (the decrease was prominent at R-60 and R-120) and also by the significant decrease of dP/dt at R-120. This was also confirmed from the AF value, which was markedly lower throughout the whole reperfusion period (Table 1).

View this table:
Table 1.

Effects of resveratrol and inhibitors of adenosine receptors and PI3-kinase on ventricular function

LY-294002, when used with resveratrol, partially abolished the cardioprotective effect of resveratrol, which was evidenced by LVDP from R-30 onward; the decrease was more prominent at R-60 and also R-120. The same result was also evidenced by dp/dt from R-60 onward; the decrease was more prominent at R-120.

Effects of resveratrol on myocardial infarct size.

Infarct size (% of infarct vs. total area at risk) was noticeably reduced in the resveratrol group compared with the control group (18.17 ± 2.08% vs. 33.79 ± 2.74%). This infarct zone was increased significantly when resveratrol was used along with CPT or MRS-1191 but not with CSC (20.1 ± 1.6%). Thus the infarct size was significantly higher in resveratrol + CPT and resveratrol + MRS-1191 groups compared with the resveratrol-alone group (27.9 ± 2.4% and 26.33 ± 2.45%, respectively, vs. 18.17 ± 2.08%), as shown in Fig. 2. Inhibition of PI3-kinase with LY-294002 also increased infarct size (24.9 ± 2.3%) compared with the resveratrol-alone group.

Fig. 2.

Effects of CPT, MRS-1191, CSC, and LY-294002 on resveratrol-mediated lowering of myocardial infarct size. Isolated rat hearts were perfused with KHB in the absence or presence of different inhibitors for 15 min, followed by 30-min ischemia and 2-h reperfusion. At the end of the experiments, the tissues were processed to determine infarct size by the triphenyltetrazolium chloride (TTC) staining method as described in materials and methods. Left to right: 1st bar, isolated hearts made globally ischemic for 30 min followed by 2 h of reperfusion; 2nd bar, isolated hearts pretreated for 15 min with 1 μM CPT followed by 30-min ischemia and 2-h reperfusion; 3rd bar, isolated hearts pretreated for 15 min with 1 μM CSC followed by 30-min ischemia and 2-h reperfusion; 4th bar, isolated hearts pretreated for 15 min with 1 μM MRS-1191 followed by 30-min ischemia and 2-h reperfusion; 5th bar, isolated hearts pretreated for 15 min with 1 μM LY-294002 followed by 30-min ischemia and 2-h reperfusion; 6th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol followed by 30-min ischemia and 2-h reperfusion; 7th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol + 1 μM CSC followed by 30-min ischemia and 2-h reperfusion; 8th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol + 1 μM CPT followed by 30-min ischemia and 2-h reperfusion; 9th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol + 1 μM MRS-1191 followed by 30-min ischemia and 2-h reperfusion; 10th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol + 1 μM LY-294002 followed by 30-min ischemia and 2-h reperfusion. The results are shown as means ± SE of at least 6 different animals per group. *P < 0.05 vs. ischemia-reperfusion (I/R); †P < 0.05 vs. resveratrol alone.

Effects of resveratrol on cardiomyocyte apoptosis.

The percentage of apoptotic cardiomyocytes was significantly reduced in the resveratrol-alone group compared with the control (3.7 ± 1.2% vs. 22.7 ± 1.5%). This apoptotic cell death was increased significantly when resveratrol was used along with CPT or MRS 1191 but not with CSC. Thus apoptosis was significantly higher in resveratrol + CPT and resveratrol + MRS 1191 groups compared with the resveratrol-alone group (11.3 ± 1.7% and 23.3 ± 1.6%, respectively, vs. 3.7 ± 1.2%), as shown in Fig. 3. Inhibition of PI3-kinase with LY-294002 also increased the number of apoptotic cells significantly (17.4 ± 1.2%) compared with the resveratrol-alone group.

Fig. 3.

Effects of CPT, MRS-1191, CSC, and LY-294002 on resveratrol-mediated lowering of cardiomyocyte apoptosis. Isolated rat hearts were perfused with KHB in the absence or presence of different inhibitors for 15 min, followed by 30-min ischemia and 2-h reperfusion. At the end of the experiments, the tissues were processed to determine cardiomyocyte apoptosis by double-antibody staining method as described in materials and methods. Left to right: 1st bar, isolated hearts made globally ischemic for 30 min followed by 2-h reperfusion; 2nd bar, isolated hearts pretreated for 15 min with 1 μM CPT followed by 30-min ischemia and 2-h reperfusion; 3rd bar, isolated hearts pretreated for 15 min with 1 μM CSC followed by 30-min ischemia and 2-h reperfusion; 4th bar, isolated hearts pretreated for 15 min with 1 μM MRS-1191 followed by 30-min ischemia and 2-h reperfusion; 5th bar, isolated hearts pretreated for 15 min with 1 μM LY-294002 followed by 30-min ischemia and 2-h reperfusion; 6th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol followed by 30-min ischemia and 2-h reperfusion. 7th bar, isolated hearts were pretreated for 15 min with 10 μM resveratrol + 1 μM CSC followed by 30-min ischemia and 2-h reperfusion; 8th bar, isolated hearts were pretreated for 15 min with 10 μM resveratrol + 1 μM CPT followed by 30-min ischemia and 2-h reperfusion. 9th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol + 1 μM MRS 1191 followed by 30-min ischemia and 2-h of reperfusion; 10th bar, isolated hearts pretreated for 15 min with 10 μM resveratrol + 1 μM LY-294002 followed by 30-min ischemia and 2-h reperfusion. The results are shown as means ± SE of at least 6 different animals per group. *P < 0.05 vs. I/R; †P < 0.05 vs. resveratrol alone.

Effects of resveratrol and adenosine receptor antagonists on expression of Akt, Bcl-2, Bad, and CREB.

Resveratrol significantly enhanced the phosphorylation of Akt, Bcl-2, Bad, and CREB. As shown in Figs. 47, phosphorylation of Akt was increased by 10- to 12-fold, Bcl-2 by 8- to 10-fold, Bad by 5-fold, and CREB by 6- to 7-fold. In addition, induction of Bcl-2 expression was also increased significantly by about eightfold. The resveratrol-mediated induction of Bcl-2 and its subsequent phosphorylation were reduced slightly, but significantly, by CPT and dramatically by MRS-1191.

Fig. 4.

Western blot analysis of Akt and its phosphorylated product. The results are shown as means ± SE of 6 different experiments per group. Representative Western blots are shown at bottom. *P < 0.05 vs. baseline (BL) or I/R. †P < 0.05 vs. resveratrol alone.

Fig. 5.

Western blot analysis of Bcl-2 and its phosphorylated product. The results are shown as means ± SE of 6 different experiments per group. Representative Western blots are shown at bottom. Filled bars, phosphorylated Bcl-2; dotted bars, nonphosphorylated Bcl-2. *P < 0.05 vs. BL or I/R; (GAPDH was used as an internal standard; not shown); †P < 0.05 vs. resveratrol alone.

Fig. 6.

Western blot analysis of Bad and its phosphorylated product. The results are shown as means ± SE of 6 different experiments per group. Representative Western blots are shown at bottom. *P < 0.05 vs. BL or I/R; †P < 0.05 vs. resveratrol.

Fig. 7.

Western blot analysis of CREB and its phosphorylated product. The results are shown as means ± SE of 6 different experiments per group. Representative Western blots are shown at bottom. *P < 0.05 vs. BL or I/R; †P < 0.05 vs. resveratrol.

DISCUSSION

In the present study, we show that resveratrol exerts PC-like effects on the ischemic myocardium as evidenced by improved postischemic ventricular function, reduced myocardial infarct size, and decrease in cardiomyocyte apoptosis. CPT and MRS-1191, but not CSC, abrogated the cardioprotective abilities of resveratrol, suggesting a role of adenosine A1 and A3 receptors in resveratrol PC. Resveratrol induced the expression of Bcl-2 and caused its phosphorylation along with phosphorylation of CREB, Akt, and Bad. CPT blocked the phosphorylation of Akt and Bad without affecting CREB, whereas MRS-1191 blocked phosphorylation of all compounds, including CREB. LY-294002 partially blocked the cardioprotective abilities of resveratrol. The results indicate that resveratrol preconditions the heart through the activation of adenosine A1 and A3 receptors, the former transmitting a survival signal through a PI3-kinase-Akt-Bcl-2 signaling pathway and the latter protecting the heart through a CREB-dependent Bcl-2 pathway in addition to an Akt-Bcl-2 pathway.

A large number of studies exist in the literature indicating cardioprotective role of resveratrol (2123, 35). Several studies including our own have indicated the ability of resveratrol to pharmacologically precondition the heart (5, 21, 23). Ischemic PC refers to the paradoxical mechanism by which cyclic episodes of brief, reversible ischemia, each followed by another brief period of reperfusion, render the heart tolerant to subsequent ischemia-reperfusion injury (12, 13, 34, 36). PC is a complex phenomenon, which occurs through multiple interrelated cascades of events. A variety of neurohumoral factors are released during the onset of PC that include, among many intracellular mediators, adenosine (1, 9, 11, 22, 26, 31, 42, 44). Both adenosine A1 and A3 receptors have been implicated in PC-mediated cardioprotection (9, 26, 31, 39, 42). The same adenosine was recently been implicated in resveratrol PC (5). In addition, NO, which is a powerful mediator of PC (8, 4), is also involved in resveratrol PC (5, 21, 23). Resveratrol activated both inducible nitric oxide synthase (iNOS) and endothelial nitric oxide synthase (eNOS), which presumably contributed toward the ability of resveratrol to provide cardioprotection. Resveratrol failed to precondition mouse hearts devoid of any copies of iNOS gene, suggesting a crucial role of iNOS in resveratrol PC (23).

The results of the present study are consistent with many previous reports that indicate the role of adenosine A1 and A3 receptors, but not A2a receptor, in ischemic PC (31). Initial studies with adenosine receptors implicated a role of adenosine A1 receptor in PC (9). Subsequent studies showed conflicting results, some of them demonstrating negative findings (1, 44). However, recent studies clearly demonstrate involvement of adenosine A3 receptors in PC (31). The results of the present study reveal that both A1 and A3 receptors are involved in resveratrol PC and both use a PI3-kinase-Akt signaling pathway. It appears that adenosine A1 receptor is at least partly responsible for resveratrol PC, because inhibition of this receptor with CSC partially blocked both anti-infarct and antiapoptotic effects of resveratrol. In contrast, MRS 1191 only partially blocked anti-ischemic effects of resveratrol but almost completely eliminated its antiapoptotic properties. The reason for this was not clearly understood; however, it can be speculated that diverse survival signaling pathways exist for necrosis and apoptosis. It is well established that necrosis and apoptosis are independent contributors to myocardial infarction. Thus it is possible that adenosine A3 receptor significantly contributes (greater than the contribution by adenosine A1 receptor) to resveratrol-mediated inhibition of necrotic cell death such that the anti-infarction effects (comprising effects due to necrosis + apoptosis) of A1 and A3 receptors remain comparable.

The paradoxical nature of the data supporting complete dependence of the antiapoptotic actions of A3 receptors and the observed reduction in this response with an A1 antagonist may be due to nonselective inhibition of A3 receptors with CPT, as suggested by previous studies of even more selective A1 antagonists (25). There are reports that A1 receptors can indeed contribute to the antiapoptotic effects, even more effectively than A3 receptors (20), which is in direct contrast to the current data supporting greater effects of A3 versus A1 receptors on apoptosis. Thus the resveratrol effect may be unique. Another possible explanation could be the inability of the TUNEL method to distinguish apoptosis from necrosis, thus possibly leading to overestimation of apoptosis. TUNEL detects single-strand and double-strand DNA breaks with free 3′-OH termini and is thus nonspecific; apoptosis may occur without DNA fragmentation, and cells can exhibit both internucleosomal DNA fragmentation and membrane damage typical of apoptosis and necrosis. It may be that these limitations and lack of distinction lead to an overestimation of apoptosis by TUNEL and therefore that the apparent A1 receptor dependence stems from the antinecrotic actions of the A1 receptor.

Akt is a critical regulator of PI3-kinase-mediated cell survival (28). A large number of studies have demonstrated in various cell types that constitutive activation of Akt is sufficient to block cell death induced by a variety of apoptotic stimuli (33). Akt is activated by PC as a result of activation of PI3-kinase leading to the activation of PKC and eNOS (19). Because of the suggested role of Akt in cell survival through antiapoptotic pathways that directly antagonize mitochondria-directed apoptosis, and because Akt is reported to be regulated by PC, we investigated the expression of Akt in resveratrol-treated hearts. The results revealed resveratrol-mediated increased Akt phosphorylation at the serine 478 site. The increased Akt phosphorylation was blocked by MRS-1191 and CPT, suggesting the involvement of both A1 and A3 receptors in Akt signaling. Interestingly, LY-294002 abolished the cardioprotective effects of resveratrol, indicating PI3-kinase as the upstream signaling molecule for resveratrol PC. Several recent studies showed phosphorylation of Akt as a result of adenosine A3 receptor activation. For example, A3 adenosine receptor activation triggered phosphorylation of PKB/Akt and protected rat basophilic leukemia 2H3 mast cells from apoptosis (16). Low concentrations of ethanol activate cell survival-promoting PI3-kinase/Akt pathway in endothelial cells by an adenosine receptor-dependent mechanism (29). In another study, adenosine receptor was found to regulate insulin-induced activation of PI3-kinase/PKB in rat adipocytes (40).

Several downstream targets of Akt have been recognized to be apoptosis-regulatory molecules, including Bcl-2 family member Bad (24) and CREB (6). We therefore examined whether resveratrol could regulate Bad and CREB expression. Our results showed that resveratrol could induce the expression of Bcl-2, which was inhibited by A1 and A3 receptor antagonism. Additionally, the downstream target molecules of Bcl-2, Bad, and CREB were phosphorylated with resveratrol. CSC and MRS-1191 significantly inhibited the phosphorylation of Bad, indicating that resveratrol-mediated Akt-Bad survival signal was regulated by both A1 and A3 adenosine receptors. It has been shown that Akt can inhibit caspase-mediated cell death through the phosphorylation of the death agonist Bad releasing Bcl-2 family members (3). Akt-mediated Bad phosphorylation presumably blocks apoptotic cell death by promoting binding of Bad to the 14-3-3 protein, thereby sequestering BAD from Bcl-2. Our results indicated that resveratrol inhibits translocation of Bad in response to ischemia-reperfusion-induced loss of Bcl-2 from the mitochondria by phosphorylated Bad via Akt. These findings are consistent with previous findings in other cell lines, in which Akt acts at the level of mitochondria to release cytochrome c via Bcl-2 in adult cardiomyocytes. PI3-kinase and Akt signaling pathways were also found to play a critical role in the prevention of apoptotic cell death by adenosine A3 receptor activation (37).

CREB, a major nuclear transcription factor that transduces cAMP activation of gene transcription, is another regulatory downstream target molecule of Akt (38). CREB has been recognized as an important nuclear factor for cell survival. Overexpression of a dominant-negative CREB transgene induced apoptosis in T cells (2). A recent study showed that CREB contributes to cell survival in response to growth factor stimulation (10). Our results showed simultaneous induction of CREB and Bcl-2 in response to resveratrol treatment. The promoter region of Bcl-2 contains a cAMP-response element site, and the transcription factor CREB has been recognized as a positive regulator of Bcl-2 expression. Like NF-κB, CREB is also a target for several signaling pathways mediated by a variety of stimulation. For example, IGF-I-mediated CREB phosphorylation was decreased by wortmannin, an inhibitor of PI3-kinase, suggesting a role of Akt in CREB activation.

Recently, an alternative survival pathway via CREB that may bypass PI3-kinase-Akt signaling was described. CREB phosphorylation was found to occur through the activation of the MAP kinase pathway via activation of p90rsk (15). In a recent study, relaxin activated CREB through a Akt-independent signaling pathway (45), In this case, CREB may be phosphorylated via a MEK/MAPK/p90rsk/CREB or cAMP-PKA signaling pathway or both (41).

In summary, the results of this study demonstrate that resveratrol PC is mediated by both adenosine A1 and A3 receptors (Fig. 8). Whereas activation of A1 and A3 receptors is clearly linked with the survival signal mediated by the Akt-Bcl-2-Bad signaling pathway, the A3 receptors also trigger another survival pathway mediated via CREB. Whether this survival pathway uses Akt (pathway 1 in Fig. 8) or not (pathway 2 in Fig. 8) was not clear from the present study. However, the activation of CREB only through adenosine A3 receptor, and not via A1 receptor, suggests that such activation may have occurred independent of Akt. Nevertheless, it appears that simultaneous activation of A1 and A3 receptors is required for resveratrol PC, suggesting that there may be cross-talk between these two receptors during resveratrol PC of the heart.

Fig. 8.

Resveratrol preconditions the heart through 2 different signaling pathways: by activating adenosine A1 and A3 receptors, which triggers a survival pathway through phosphatidylinositol (PI)3-kinase-Akt-Bad/Bcl-2 (pathway 1) and by activating adenosine A3-CREB-Bcl-2 signaling (pathway 2). NOS, nitric oxide synthase.

GRANTS

This study was supported in part by National Heart, Lung, and Blood Institute Grants HL-22559, HL-33889, HL-34360, and HL-56803.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract