Augmentation by resveratrol of the inhibitory effect of ethanol on platelet aggregation

Abstract Ethanol and resveratrol have been shown to inhibit platelet aggregation. The aim of this study was to determine whether resveratrol has an additional effect on ethanol-induced inhibition of platelet aggregation. Ca2+ entry and subsequent aggregation of human platelets were measured by the fluorescence method and light transmittance method, respectively. Thromboxane B2 concentrations in media containing platelets were measured by using the enzyme-linked immunosorbent assay. Platelet aggregation induced by thrombin (0.025 U/ml) was significantly inhibited by preincubation of platelets with ethanol (0.5%). Preincubation with resveratrol (3.125 µM), which did not affect thrombin-induced platelet aggregation, significantly augmented the inhibitory effect of ethanol on platelet aggregation. Similar synergic effects of ethanol and resveratrol were found on aggregatory responses to collagen (2 µg/ml) and arachidonic acid (0.25 mM). On the other hand, the thrombin-induced increase in intracellular Ca2+ concentration ([Ca2+]i) was not affected by ethanol alone, resveratrol alone or both ethanol and resveratrol together. In nominally Ca2+-free medium, arachidonic acid (0.75 mM) caused a potent platelet aggregation, which was not affected by the presence of ethanol alone, resveratrol alone, or both of them together. Thromboxane B2 formation induced by thrombin was significantly inhibited by ethanol (0.5%) alone and resveratrol (3.125 µM) alone, and these inhibitory effects were significantly augmented in the presence of both ethanol and resveratrol together. Resveratrol shows an additive effect on ethanol-induced inhibition of platelet aggregation. This effect by resveratrol is partly explained by its inhibitory action on thromboxane A2 production in platelets. In addition, both ethanol and resveratrol attenuate platelet aggregation through acting on the Ca2+-dependent intra-platelet pathway after an increase in [Ca2+]i induced by thrombin.

• Light-to-moderate alcohol consumption has been shown to reduce the risk of ischemic heart disease in the general population.• Alcohol (Ethanol) is known to inhibit platelet aggregation.
• Red wine has been suggested to be more beneficial than other alcohol beverages for prevention of cardiovascular disease.• Resveratrol, a representative polyphenol compound contained in red wine, has an antioxidant activity, which is thought to be involved in prevention of cardiovascular disease.
• It remains to be clarified whether resveratrol has an additional effect on alcohol-induced inhibition of platelet aggregation.
What is new?
• Preincubation with resveratrol at a low concentration, which did not affect thrombin-induced platelet aggregation, significantly augmented the inhibitory effect of ethanol on platelet aggregation.
• Similar synergic effects of ethanol and resveratrol were found on aggregatory responses to collagen and arachidonic acid.

Introduction
The mortality rate from ischemic heart disease is lower in France than in other European countries despite the fact that dietary fat intake is not low in France [1].This evidence is known as the French paradox, and polyphenol compounds, represented by resveratrol, contained in red wine have been suggested to be responsible for the lower risk of cardiovascular disease in French people [2,3].Resveratrol has an antioxidant activity, which is thought to be involved in prevention of cardiovascular disease through inhibiting LDL oxidation and increasing formation and release of nitric oxide from vascular endothelial cells [4].Light-to-moderate alcohol consumption has been shown to reduce the risk of ischemic heart disease in the general population [5,6].The major reason for this preventive effect of alcohol on cardiovascular diseases is alcohol-induced elevation of blood HDL cholesterol level [7][8][9].Blood coagulability and the risk of thrombotic diseases have been shown to be lower in drinkers than in nondrinkers [10].In addition to lower levels of blood coagulation factors including fibrinogen, von Willebrand factor and coagulation factor VII in light-to-moderate drinkers than in nondrinkers [11], ethanol-induced inhibition of platelet aggregation has been suggested to be involved in the lower risk of cardiovascular disease in light-to-moderate drinkers than in nondrinkers [12].Resveratrol has also been shown to inhibit platelet aggregation [13].Therefore, both alcohol and resveratrol are thought to display antithrombotic effects in red wine drinkers.However, it remains to be clarified whether resveratrol has an additional effect on alcohol-induced inhibition of platelet aggregation.
As the mechanisms for inhibiting platelet aggregation, ethanol has been reported to inhibit store-operated Ca 2+ entry [14] and phospholipase A 2 activity [15] in platelets, which cause an increase in intracellular Ca 2+ concentrations ([Ca 2+ ] i ) and thromboxane A 2 production, respectively.Nonetheless, thrombin-induced increase in [Ca 2+ ] i is not affected by ethanol because Ca 2+ entry through another pathway, diacylglycerolactivated Ca 2+ entry, induced by thrombin is increased by ethanol, resulting in no change in total Ca 2+ entry following stimulation of platelets by thrombin [16].Resveratrol has also been reported to inhibit store-operated Ca 2+ entry into platelets [17].However, it remains to be determined whether ethanol and resveratrol together affect increases in [Ca 2+ ] i and phospholipase A 2 activity following stimulation of platelets.
In this study, we therefore investigated effects of ethanol and resveratrol together on thrombin-induced platelet aggregation, transplasmalemmal Ca 2+ entry into platelets and thromboxane A 2 production by platelets.Ethanol and resveratrol at low concentrations that are close to their threshold concentrations for inhibition of platelet aggregation were used to investigate additive effects of resveratrol on ethanol-induced inhibition of platelet aggregation and the related responses.

Preparation of a washed-platelet suspension
Blood was obtained from healthy donors who had been medication-free for at least 10 days prior to the experiments.This study was approved by the Ethics Committee of Hyogo College of Medicine (No. 1799), and the experimental procedures were in accordance with the Helsinki Declaration.Blood (36 ml) was rapidly transferred to a plastic tube containing 4 ml of 3.2% sodium citrate and mixed.The blood was then centrifuged at 150 × g for 10 min, and the supernatant was obtained as plateletrich plasma (PRP).PRP was subsequently mixed with 40 ml of Ca 2+ -and Mg 2+ -free Tyrode solution buffered by Hepes (NaCl 150 mM, KCl 5 mM, glucose 10 mM, HEPES 10 mM) (pH 7.4) and containing 1 mM EGTA, and the mixture was centrifuged at 150 × g for 10 min.After the supernatant had been further centrifuged at 400 × g for 5 min, the obtained pellet was suspended with 40 ml of the above Tyrode-Hepes solution and then further centrifuged at 400 × g for 5 min.The pellet was suspended with 4 ml of Ca 2+ -free Tyrode solution (NaCl 150 mM, KCl 5 mM, MgCl 2 1 mM, glucose 10 mM, and HEPES 10 mM) (pH 7.4), and the resulting platelet suspension was used for the experiments within 2 hours after blood collection.The concentration of platelets in the suspension used for each experiment was adjusted to be approximately 10 5 /µl.
Fluorescence measurements were carried out with a dualwavelength spectrofluorimeter (F-2500 Fluorescence Spectrophotometer, Hitachi High-Technologies Corporation, Tokyo, Japan) using a 0.4-ml cuvette maintained at 37°C.The wavelengths used for excitation were 340 and 380 nm, and the wavelength used for emission was 510 nm.Fractional changes in [Ca 2+ ] i were determined by using a ratio (R) of fluorescence intensity (F) of F340/F380.The fluorescence intensities after sequential additions of 0.25% Triton X-100 and EGTA (5 mM) to the platelet suspension provided the maximum fluorescence ratio and minimum fluorescence ratio, respectively.[Ca 2+ ] i was calculated using the method proposed by Grynkiewicz et al. [18].Ca 2+ entry induced by thrombin was expressed as the net increase in [Ca 2+ ] i calculated by subtraction of the basal [Ca 2+ ] i level from the maximum [Ca 2+ ] i level after stimulation.

Measurement of platelet aggregation in a washed-platelet suspension
Aggregation of platelets suspended in Ca 2+ -free Tyrode solution buffered by Hepes (NaCl 150 mM, KCl 5 mM, MgCl 2 1 mM, glucose 10 mM, and HEPES 10 mM) (pH 7.4) was evaluated by using an aggregometer (PRP313 M, TAIYO Instruments INC., Osaka, Japan) that measures increases in light transmission through a cuvette (0.2 ml) containing a stirred platelet suspension.The light transmission through a washed-platelet suspension without any treatment and that through a suspended buffer not containing platelets were considered as 0% and 100%, respectively.
The percentage of aggregation during the course of each experiment was calculated.The experimental conditions were the same as those for [Ca 2+ ] i measurement except for the volume of the cuvettes.

Protocols for experiments to assess Ca 2+ entry and platelet aggregation in a washed-platelet suspension
As mentioned below, we first determined the threshold concentrations of ethanol and resveratrol by using concentration-dependent inhibitory effects of ethanol and resveratrol on thrombin-induced platelet aggregation (Figure 1).In the further experiments, platelets were stabilized in nominally Ca 2+ -free medium in the cuvette for 3 min and then pretreated with ethanol (0.5%) alone, resveratrol (3.125 µM) alone, ethanol (0.5%), and resveratrol (3.125 µM) together, or a vehicle.Platelets were then stimulated with thrombin (0.025 U/ml), collagen (2 µg/ml) or arachidonic acid (0.25 mM).At 1 min after the addition of each stimulant, Ca 2+ entry and aggregation were induced by adding CaCl 2 (0.5 mM) to the cuvette.Mean levels of platelet aggregation and Ca 2+ entry were evaluated at 10 and 5 minutes, respectively, after addition of CaCl 2 .Because one of the aims of this study was to determine the effects of resveratrol and/ or ethanol on transplasmalemmal Ca 2+ entry, the effects of resveratrol and/or ethanol were observed under the non-physiological conditions, in which platelets were stimulated with thrombin in Ca 2+free buffer and then platelet aggregation was induced by the addition of Ca 2+ .Arachidonic acid (0.75 mM)-induced aggregation was observed in the nominally Ca 2+ -free solution.
Figure 1.Effects of different concentrations of resveratrol (A) and ethanol (B) on platelet aggregation induced by thrombin.In the experiments using ethanol, after stabilization, platelets were further incubated for 3 min.Then, thrombin (0.025 U/ml) was added to the platelet suspension, and at 1 min after stimulation with thrombin, each concentration of ethanol or a vehicle was added.After further incubation for 1 min, CaCl 2 (0.5 mM) was added to the platelet suspension to induce platelet aggregation.In the experiments using resveratrol, after stabilization, platelets were pretreated with each concentration of resveratrol or a vehicle for 3 min.Then, thrombin (0.025 U/ml) was added to the platelet suspension, and at 1 min after stimulation with thrombin, CaCl 2 (0.5 mM) was added to the platelet suspension to induce platelet aggregation.Symbols indicate significant differences from aggregation of platelets pretreated with the vehicle (control) (*, P < .05;**, P < .01).N.S., no significant difference.N = 5-6.

Evaluation of thromboxane A 2 production by platelets
Thromboxane A 2 (TXA 2 ) is produced from arachidonic acid in platelets and causes irreversible platelet aggregation; since TXA 2 is rapidly hydrolyzed non-enzymatically to thromboxane B 2 (TXB 2 ), TXB 2 concentration in washed platelets was measured as an indicator of TXA 2 production by platelets.TXB 2 was measured by the enzyme-linked immunosorbent assay using a commercial kit (Cayman Chemical Company, USA).Washed platelets just after measurement of platelet aggregation was centrifuged at 400 × g for 5 minutes, and the collected supernatant was used as a sample for TXB 2 measurement.In preliminary experiments, we confirmed that resveratrol did not affect the standard curve of TXB 2 measurement by using the above kit and thus did not interfere with the measurement.

Preparation of whole blood
Blood (18 ml) was rapidly transferred to a plastic tube containing 2 ml of sodium citrate (3.2%) and mixed.Since stabilization of platelets for 1 hour after blood collection is needed in order to avoid the effects of prostacyclin included in whole blood before experiments, the experiments using whole blood were done from 1 hour to 2 hours after blood collection.

Measurement of platelet aggregation in whole blood
Platelet aggregation in whole blood was measured by the screen filtration pressure method using a whole-blood aggregometer (WBA-Neo, ISK, Tokyo, Japan).While constantly stirring at 37°C, the reaction was started by addition of a solution containing thrombin and Ca 2+ .At 5 min after stimulation, the absorbing pressure of aggregated whole blood was measured through a micro sieve with 30 × 30 µm windows, and negative pressures of −130 mmHg and −6 mmHg were defined as 100% and 0% aggregation, respectively, the latter deviation from 0 mmHg being designated because of the viscosity of unstimulated whole blood.Thrombin at final concentrations of 0.05, 0.1, 0.2 and 0.4 U/ml and Ca 2+ at a final concentration of 0.5 mM were simultaneously added to whole blood.Thrombin stimulation caused formation of platelet aggregates, which were trapped in the mesh filter and resulted in an increase of its absorbing pressure in the wholeblood aggregometer.

Drugs
Resveratrol (Sigma-Aldrich, St Louis, MO, USA), arachidonic acid (Tokyo Chemical Industry, Tokyo, Japan), 1-oleoyl-2-acetylglycerol (OAG, Sigma-Aldrich) and fura-2/AM (Dojindo Laboratories, Kumamoto, Japan) were dissolved in dimethylsulfoxide to make stock solutions of 50 mM, 250 mM, 100 mM and 5 mM, respectively, and were stored at −80°C and diluted with distilled water at the time of use.Bovine thrombin (FUJIFILM Wako Pure Chemical Corporation Osaka, Japan) was dissolved in distilled water to make a stock solution of 1 U/µl and was stored at −80°C.Collagen (DS Medical Co., Ltd.Tokyo, Japan) was dissolved in distilled water to make a stock solution of 100 µg/ml and was stored at 4°C.Aspirin (acetylsalicylic acid, FUJIFILM Wako Pure Chemical Corporation) was prepared at time of use by dissolving in ethanol to make a solution of 100 mM.

Statistical analysis
Data are presented as means ± standard deviations.Statistical analysis was performed using analysis of variance followed by Scheffé F-test.P values less than 0.05 were regarded as significant.

Concentration-dependent effects of ethanol and resveratrol on thrombin-induced platelet aggregation
Figure 1 shows effects of different concentrations of ethanol and resveratrol on thrombin-induced platelet aggregation.Thrombininduced platelet aggregation was significantly inhibited by resveratrol at 6.25 µM or higher (Figure 1A) and by ethanol at 0.5% or higher (Figure 1B) but was not significantly affected by ethanol at 0.25% (data not shown).Then, in further experiments for testing an additive effect of resveratrol on ethanol-induced inhibition of platelet aggregation, we used ethanol at 0.5%, which was its threshold concentration for inhibiting the aggregation, and resveratrol at 3.125 µM, which was close to but was slightly lower than the threshold concentration for inhibiting the aggregation.

Effects of resveratrol, ethanol, and both of them on thrombin-induced platelet aggregation
Figure 2A shows a representative time course of thrombininduced platelet aggregation in a washed platelet suspension under different experimental conditions.Platelets were incubated in a nominally Ca 2+ -free medium for 180 sec and then stimulated with thrombin, which induced a small aggregation.At 120 sec after stimulation with thrombin, CaCl 2 was added to the platelet suspension, which caused a large aggregatory response.Platelet aggregation was significantly inhibited in the presence of ethanol (0.5%) alone compared with the control, while preincubation of platelets with resveratrol (3.125 µM) alone did not affect platelet aggregation.In the presence of both resveratrol and ethanol, platelet aggregation was significantly lower than that in the presence of ethanol alone (Figure 2A).
The effects of resveratrol and/or ethanol on thrombin-induced aggregation was also investigated in the Ca 2+ -containing medium.As shown in Figure 2B, thrombin-induced aggregation in the presence of extra-platelet Ca 2+ was significantly inhibited by ethanol (0.5%) alone, and this inhibitory effect was augmented in the presence of resveratrol at a low concentration (3.125 µM).

Effects of resveratrol, ethanol, and both of them on thrombin-induced increase in [Ca 2+ ] i
[Ca 2+ ] i was monitored under the same condition as that for the above experiments on platelet aggregation.As shown in Figure 3, stimulation with thrombin caused a slight increase in [Ca 2+ ] i , and addition of CaCl 2 thereafter induced a large increase in [Ca 2+ ] i .Incubation of platelets with resveratrol, ethanol, or both of them did not significantly affect the thrombin-induced increase in [Ca 2+ ] i (Figure 3).

Effects of resveratrol, ethanol, and both of them on 1-oleoyl-2-acetylglycerol (OAG)-induced increase in [Ca 2+ ] i
OAG-induced Ca 2+ entry into platelets was slightly but significantly augmented by ethanol but was not significantly affected by resveratrol.Resveratrol did not affect OAG-induced Ca 2+ entry in the presence of ethanol (Figure 4).

Effects of resveratrol, ethanol, and both of them on collagen-induced platelet aggregation
Figure 5A shows effects of ethanol and resveratrol on collageninduced platelet aggregation.Collagen-induced platelet aggregation was inhibited in the presence of ethanol alone or resveratrol alone.In the presence of both ethanol and resveratrol, collageninduced platelet aggregation was significantly less than that in the presence of ethanol alone or resveratrol alone.Effects of ethanol and/or resveratrol on collagen-induced aggregation were also examined in the presence of aspirin (1 mM).As shown in Figure 5B, collagen-induced platelet aggregation was much less in the presence of aspirin than in the absence of aspirin and was significantly inhibited by resveratrol alone, ethanol alone and both together, although a synergic inhibitory action of resveratrol and ethanol was not observed for collagen-induced aggregation in the presence of aspirin.

Effects of resveratrol, ethanol, and both of them on arachidonic acid-induced platelet aggregation in the presence and absence of Ca 2+ in the medium
Figure 6A shows the effects of resveratrol and ethanol on platelet aggregation induced by arachidonic acid at a lower concentration (0.25 mM).Platelet aggregation induced by arachidonic acid in the presence of Ca 2+ in the medium was significantly attenuated by ethanol alone and by resveratrol alone.The inhibitory effects of Figure 2. (A) Effects of resveratrol alone, ethanol alone, and resveratrol and ethanol together on platelet aggregation induced by thrombin.After stabilization, platelets were pretreated with resveratrol (3.125 μM) or a vehicle for 3 min.Then, thrombin (0.025 U/ml) was added to the platelet suspension, and at 1 min after stimulation with thrombin, ethanol (0.5%) or a vehicle was added.After further incubation for 1 min, CaCl 2 (0.5 mM) was added to the platelet suspension to induce platelet aggregation.Symbols indicate significant differences from aggregation of platelets pretreated with the vehicle (control) (**, P < .01)and from aggregation of platelets pretreated with ethanol alone (#, P <.05).N = 6-8.(B) Effects of resveratrol alone, ethanol alone, and resveratrol and ethanol together on platelet aggregation induced by thrombin in Ca 2+containing medium.After stabilization, platelets were pretreated with CaCl 2 (0.5 mM), resveratrol (3.125 μM) or a vehicle for 3 min.Then, ethanol (0.5%) or a vehicle was added, and after further incubation for 1 min, thrombin (0.025 U/ml) was added to the platelet suspension.Symbols indicate significant differences from aggregation of platelets pretreated with the vehicle (control) (**, P < .01)and from aggregation of platelets pretreated with ethanol alone (#, P < .05).N.S., no significant difference.N = 6-7.(2 μg/ml) was added to the platelet suspension to induce platelet aggregation.Symbols indicate significant differences from aggregation of platelets pretreated with the vehicle (control) (**, P < .01).N = 6.(B) Effects of resveratrol alone, ethanol alone, and resveratrol and ethanol together on platelet aggregation induced by collagen in the presence of aspirin.Platelets are pretreated with aspirin (1 mM) for 45 min.After washing twice with Ca 2+ -free Tyrode solution and stabilization, platelets were pretreated with CaCl 2 (0.5 mM), resveratrol (3.125 μM) or a vehicle for 3 min.Then, ethanol (0.5%) or a vehicle was added, and after further incubation for 1 min, collagen (2 μg/ml) was added to the platelet suspension to induce platelet aggregation.Symbols indicate significant differences from aggregation of platelets pretreated with the vehicle (control) (*, P < .05;**, P < .01).N = 5.
ethanol alone and resveratrol alone on the arachidonic acid-induced aggregation were significantly augmented by resveratrol and ethanol, respectively.
Figure 6B shows platelet aggregation induced by arachidonic acid at a higher concentration (0.75 mM) in a nominally Ca 2+ -free medium.Arachidonic acid induced a rapid and large aggregatory response, which was not affected by the presence of resveratrol alone, ethanol alone or both of them together.

Effects of resveratrol, ethanol, and both of them on thrombin-induced platelet aggregation in whole blood
Platelet aggregation induced by different concentrations of thrombin was tested using whole blood.As shown in Figure 7, platelet aggregation induced by thrombin at 0.2 U/ml was significantly inhibited in the presence of both resveratrol and ethanol but not in the presence of ethanol alone or resveratrol alone.Platelet aggregation induced by thrombin at 0.4 U/ml was significantly inhibited in the presence of resveratrol alone, ethanol alone and both of them.The degree of the inhibition tended to be stronger in the presence of both resveratrol and ethanol than in the presence of resveratrol alone and ethanol alone.

Effects of resveratrol, ethanol, and both of them on thromboxane B 2 formation stimulated by thrombin and arachidonic acid
Table I shows the results of comparison of thromboxane B 2 formation under different conditions.Thrombin-induced thromboxane B 2 formation was significantly inhibited in the presence of resveratrol alone, ethanol alone, and both of them.The degree of inhibition of thromboxane B 2 formation in the presence of both ethanol and resveratrol was significantly stronger than that in the presence of resveratrol alone and ethanol alone.Thromboxane B 2 formation in the presence of arachidonic acid was slightly inhibited by resveratrol alone and was not significantly affected by ethanol alone but was strongly attenuated in the presence of both ethanol and resveratrol together (Table I).

Discussion
Aggregation of platelets stimulated with thrombin was inhibited by ethanol, and this inhibition was augmented in the presence of a low concentration (3.125 µM) of resveratrol that had no effect on platelet aggregation.Thus, resveratrol potentiated the inhibitory effect of ethanol on thrombininduced platelet aggregation.Similar synergic inhibitory effects of ethanol and resveratrol were found on aggregatory responses induced by collagen and arachidonic acid.To the best of our knowledge, this is the first study showing an additive inhibitory effect of resveratrol on ethanol-induced inhibition of platelet aggregation.Moreover, interestingly, the synergic inhibitory effect was also found in experiments under a more physiological condition using whole blood.The inhibitory effect of the combination of resveratrol and ethanol on platelet aggregation in whole blood was more pronounced with a higher dose of thrombin than with a lower dose (0.4 Ethanol, resveratrol and platelet aggregation 7 versus 0.2 U/ml).Although the exact reason for this difference is unknown, one possible reason is a larger variability of the results of experiments with a lower dose of thrombin, which was reflected by higher standard deviations of the distribution of platelet aggregation levels (Figure 7).Since resveratrol is a major component of polyphenol compounds contained in red wine, platelet aggregability is expected to be lower in red wine drinkers than in other alcohol beverage drinkers.This may partly explain the reason for the beneficial preventive effect of red wine on cardiovascular disease since proneness to thrombosis resulting from increased blood coagulability is deeply involved in the pathogenesis of thrombo-atherosclerotic diseases.Further in vivo and ex vivo studies are needed to confirm this hypothesis.
On the other hand, ethanol has been shown not to affect thrombin-induced Ca 2+ entry into platelets [16].The effects of alcohol on platelet function are complicated.Transplasmalemmal Ca 2+ entry is increased when platelets are activated by stimulants [19].are two major pathways of transplasmalemmal Ca 2+ entry into platelets, capacitative Ca 2+ entry and diacylglycerol-activated Ca 2+ entry [20].Thrombin induces platelet aggregation through activating both of these Ca 2+ entry pathways.Ethanol at a concentration of 0.5% or higher has been shown to inhibit the former pathway and to augment the latter pathway (diverse effects) [16].Therefore, it is reasonable that thrombin-induced Ca 2+ entry was not affected by ethanol (0.5 ~ 2%) [16], which was confirmed in the present study (Figure 3).Diacylglycerol-activated Ca 2+ entry in platelets is mediated via TRPC6 [21] and is less than capacitative Ca 2+ entry, resulting in the possibility of diacylglycerol-activated Ca 2+ entry being masked in thrombin-induced platelet aggregation.Then, we investigatedeffects of resveratrol and ethanol on Ca 2+ entry induced by OAG, an activator of TRPC6.However, resveratrol did not affect OAGinduced Ca 2+ entry in the absence or presence of ethanol.Therefore, TRPC6 is not a target molecule for the synergic inhibitory action of ethanol and resveratrol on platelet activation.The above findings suggest that inhibition of thrombin-induced aggregation of platelets by ethanol is independent of an increase in [Ca 2+ ] i .Resveratrol at a concentration of 6.25 µM or higher has been shown to inhibit thrombin-induced platelet aggregation and capacitative Ca 2+ entry but not to affect diacylglycerol-activated Ca 2+ entry [17].In the present study, we used a lower concentration (3.125 µM) of resveratrol that did not significantly affect thrombininduced aggregation (Figure 1A) and Ca 2+ entry.Resveratrol at this concentration significantly augmented the inhibitory effect of ethanol on thrombin-induced aggregation (Figure 2) but did not significantly affect the corresponding increase in [Ca 2+ ] i (Figure 3).Therefore, the enhancing effect of resveratrol on inhibition by ethanol of thrombin-induced aggregation of platelets is thought to be independent of an increase in [Ca 2+ ] i .However, interestingly, platelet aggregation induced by arachidonic acid in a nominally Ca 2+ -free medium was not affected by ethanol or resveratrol (Figure 6B).Therefore, the intraplatelet action point of ethanol and resveratrol for inhibition of platelet aggregation is speculated to be Ca 2+ sensitivity of cytoskeletal proteins including actomyosin in platelets, which remains to be elucidated in future studies.
In our preliminary experiments, in the medium not containing Ca 2+ , arachidonic acid at a high concentration (0.75 mM) induced potent platelet aggregation, which was not affected by ethanol at concentrations of up to 2.0% and by resveratrol at concentrations of up to 25 µM (data not shown).Thus, the inhibitory effects of ethanol (0.5%) and resveratrol (3.125 µM) on platelet aggregation in the present study were not due to their nonspecific cytotoxic effects on the aggregation pathway of platelets including cytoskeletal proteins, the final molecules of the pathway that are activated by an increase in [Ca 2+ ] i .In the present study, thrombin-induced platelet aggregation was attenuated by resveratrol and ethanol, while thrombin-induced elevation of [Ca 2+ ] i was not affected by resveratrol or ethanol.Elevation of [Ca 2+ ] i is prerequisite for activation of actomyosin in platelet aggregation.It is known that the sensitivity to Ca 2+  of the contractile apparatus is increased by several signaling pathways such as protein kinase C, RhoA/Rho kinase and reactive oxygen species (ROS) in vascular smooth muscle cells [22].Therefore, one possible mechanism for the attenuation of aggregation by ethanol and resveratrol is inhibition of the intra-platelet signal transduction pathway(s) that regulates Ca 2+ sensitivity of actomyosin in platelets.In fact, ethanol has been reported to lower RhoA/Rho-kinase-mediated Ca 2+ sensitivity of contractile apparatus in lung parenchymal tissues [23].ROS, an intracellular signal regulating Ca 2+ sensitivity of the actomyosin, is known to be inhibited by polyphenols represented by resveratrol [24].Future studies are therefore needed to clarify the intracellular molecular mechanism following an increase in [Ca 2+ ] i for the synergic inhibitory action of ethanol and resveratrol on platelet aggregation.
Another possible target molecule for inhibition of platelet aggregation is phospholipase A 2 since its activity in platelets was shown to be inhibited by ethanol [25], and it was reported that resveratrol bound to phospholipase A 2 and inhibited its activity [26].We therefore investigated the effects of ethanol and resveratrol on thromboxane A 2 production in platelets.Through the mechanisms mentioned above, thrombin induced an elevation of [Ca 2+ ] i , which increases activity of phospholipase A 2 , resulting in an increase of thromboxane A 2 production.In this study, the thrombininduced increase in thromboxane B 2 levels was significantly inhibited by ethanol alone and resveratrol alone, and these inhibitions were augmented in the presence of ethanol and resveratrol together.Therefore, phospholipase A 2 is a target molecule for inhibition of platelet aggregation by ethanol and resveratrol.Although thrombin-induced thromboxane B 2 formation was inhibited by ethanol and resveratrol, they did not affect thrombin-induced elevation of [Ca 2+ ] i , which increases phospholipase A 2 activity.Therefore, ethanol and resveratrol are thought to inhibit phospholipase A 2 through a Ca 2+independent mechanism.
Resveratrol and ethanol displayed synergistic inhibitory effects on thrombin-induced platelet aggregation (Figure 2) and thrombin-induced thromboxane B 2 formation (Table I).As shown in Table I, arachidonic acid-induced thromboxane B 2 formation was inhibited by resveratrol alone but was not inhibited by ethanol alone.Since arachidonic acid-induced thromboxane A 2 production in platelets is mediated via cyclooxygenase-1 but not via phospholipase A 2 , cyclooxygenase-mediated thromboxane A 2 synthesis is thought to be inhibited by resveratrol but not by ethanol.Consistently, resveratrol was reported to inhibit the in vitro activity of cyclooxygenase-1 purified from ram seminal vesicles [27].Therefore, one possible mechanism for the augmenting action of resveratrol on ethanol-induced inhibition of platelet aggregation stimulated by thrombin is inhibition of cyclooxygenase-1.The synergic inhibitory effect of resveratrol and ethanol on collagen-induced platelet aggregation was observed in the absence of aspirin but not in the presence of aspirin.This finding also supports that the synergic action is at least partly mediated via inhibition of the cyclooxygenase-1 pathway.
When platelets were stimulated by arachidonic acid at a higher concentration (0.75 mM), combination of resveratrol and ethanol considerably reduced thromboxane B 2 formation but did not affect aggregation.Although inhibited by about 60% in the presence of resveratrol and ethanol, arachidonic acid-induced thromboxane B 2 formation level in the presence of resveratrol and ethanol was much higher than thrombininduced thromboxane B 2 formation level in the absence of resveratrol and ethanol (197.57± 98.14 vs. 8.74 ± 2.77 ng/ ml).Therefore, residual 40% of thromboxane A 2 production induced by arachidonic acid in the presence of resveratrol and ethanol was thought to be sufficient for causing platelet aggregation at a maximum level, resulting in comparable levels of arachidonic acid-induced platelet aggregation in the absence and presence of resveratrol and/or ethanol (Figure 6B).
There is also an inconsistency in the findings of this study regarding thrombin-induced platelet aggregation and thromboxane A 2 production: Resveratrol (3.125 µM) inhibited thrombin-induced thromboxane B 2 formation (Table I) but not platelet aggregation (Figure 2).In fact, thromboxane B 2 formation was reduced by 67% in the presence of resveratrol.This means that residual 33% or less of thromboxane A 2 production was sufficient for causing platelet aggregation when stimulated with thrombin at the concentration used.These results agree with the results of a previous study showing that thrombin-induced aggregation was not affected by 13-azaprostanoic acid, a specific antagonist of thromboxane A 2 /endoperoxide receptors [28].Further studies are therefore needed to clarify how thromboxane A 2 is involved in platelet aggregation using different concentrations of thrombin and different kinds of agonists of platelets.
In conclusion, the inhibitory effect of ethanol on thrombininduced aggregation of platelets was enhanced by resveratrol at a low concentration, which alone did not affect platelet aggregation and corresponding [Ca 2+ ] i entry.Resveratrol also augmented the ethanol-induced inhibition of thromboxane A 2 production in platelets stimulated with thrombin.Thus, resveratrol has additive inhibitory effects on platelet aggregation and thromboxane A 2 production in the presence of ethanol.This may in part explain the greater preventive effect of red wine than that of other alcohol beverages on thrombotic cardiovascular disease [29].

Figure 3 .
Figure 3. Effects of resveratrol alone, ethanol alone, and resveratrol and ethanol together on [Ca 2+ ] i of platelets stimulated with thrombin.After stabilization, platelets were pretreated with resveratrol (3.125 μM) or a vehicle for 3 min.At 1 min after stimulation with thrombin (0.025 U/ml), ethanol (0.5%) or a vehicle was added.After further incubation for 1 min, CaCl 2 (0.5 mM) was added to the platelet suspension to induce platelet aggregation.N.S., no significant difference.N = 5-6.

Figure 6 .
Figure 6.(A) Effects of resveratrol alone, ethanol alone, and resveratrol and ethanol together on platelet aggregation induced by arachidonic acid.After stabilization, platelets were pretreated with CaCl 2 (0.5 mM), resveratrol (3.125 μM) or a vehicle for 3 min.Then, ethanol (0.5%) or a vehicle was added, and after further incubation for 1 min, arachidonic acid (0.25 mM) was added to the platelet suspension.Symbols indicate significant differences from aggregation of platelets pretreated with the vehicle (control) (*, P < .05;**, P < .01).N = 6-7.(B) Effects of resveratrol alone, ethanol alone, and resveratrol and ethanol together on platelet aggregation induced by arachidonic acid in nominally Ca 2+free solution.After stabilization, platelets were pretreated with resveratrol (3.125 μM) or a vehicle for 3 min.Then, ethanol (0.5%) or a vehicle was added to the platelet suspension, and after further incubation for 1 min, arachidonic acid (0.75 mM) was added to induce platelet aggregation.N.S., no significant difference.N = 8-9.