Platelet Rebound Effect of Alcohol Withdrawal and Wine Drinking in Rats
Relation to Tannins and Lipid Peroxidation
Abstract We investigated in rats fed a purified diet for 2 and 4 months whether wine drinking was associated with the rebound effect on thrombin-induced platelet aggregation observed after alcohol withdrawal. With 6% ethanol drinking or its equivalent in red or white wine, platelet aggregation was reduced similarly by 70% when the animals drank the alcoholic beverages up to the venipuncture. Depriving the rats of alcoholic beverages for 18 hours was associated with an increase in the platelet response of 124% in those receiving 6% ethanol, of 46% with white wine but a decrease of 59% in those with red wine. The protective effect of red wine on platelets could be reproduced by tannins (procyanidins) extracted from grape seeds or red wine and added to 6% ethanol, but not by glycerol or wine without alcohol. That was related to inhibition of the alcohol-induced lipid peroxidation as shown by the lowering of conjugated dienes, lipid peroxides, and the increase in vitamin E in plasma. Owing to tannins, the platelets of rats drinking red wine did not exhibit the rebound effect observed hours after alcohol drinking, eventually associated with sudden death and stroke in humans.
- Received August 9, 1994.
- Accepted October 31, 1994.
Alcoholic beverages have been associated with protective effects against coronary heart disease (CHD) in a large number of ecological, case-control, and prospective studies.1 St Leger et al2 have postulated that this protective effect could be due essentially to wine. It has been further confirmed that wine drinking may eventually explain at least part of the French paradox, ie, a low mortality rate from CHD despite a high consumption of saturated fat3 and smoking habits similar to those of other countries.4 In a recent large prospective study, it has been shown that wine preference was associated with a 30% to 40% lower risk for CHD death compared with liquor or beer.5 Perhaps even more surprising, for oropharyngeal cancers6 a moderate intake of wine in both men and women was related to a protective effect of 20% to 50% that was not shared by other alcoholic beverages.
Wine, in addition to alcohol, contains glycerol and tannins. In tannins, there are powerful antioxidants such as resveratrol and polyphenols (procyanidins).7 The phenolic substances have been shown recently to protect low-density lipoprotein (LDL) from peroxidation in vitro more effectively than vitamin E.8 Because peroxidized lipoproteins could be the main atherogenic lipoproteins,9 the antioxidant effects of wine combined with the high-density lipoprotein (HDL) cholesterol–promoting effect of alcohol10 may offer an explanation for the apparent advantages of wine.
In addition to its effect on lipoproteins, alcohol has been shown to impede all tests of platelet aggregation after infusion or ingestion in humans,11 especially secondary aggregation to ADP.12 13 In a recent study on 1600 subjects in Caerphilly,14 primary and secondary aggregation to ADP were inhibited in a dose-related manner, to the same degree and by the same level of alcohol consumption3 known to protect against CHD.15 By contrast, there was a 1.5-fold to threefold increase in thrombin-induced aggregation, depending on the level of alcohol and fat consumption.14 Although the primary effect of alcohol is to decrease aggregation to thrombin, as shown in humans11 and in rats,16 the alcohol-withdrawal rebound effect has been described in alcoholic subjects17 and is associated with increased risk of thrombosis, sudden death, and stroke.18 The subjects from Caerphilly14 were not alcoholics but were consumers of primarily beer and spirits who had been deprived of food and alcohol for at least 12 hours when platelet reactivity was examined. Because the rebound effect on thrombin-induced aggregation was not observed in French farmers19 who drank primarily wine, the purpose of the present study in rats was to determine whether wine drinking was associated with the rebound effect on thrombin-induced aggregation that has been observed with alcohol drinking. Our results showed that enhanced platelet aggregation was not observed in rats drinking wine, after withdrawal. This protection of wine against the platelet rebound effect associated with alcohol drinking seems to be due essentially to the tannins it contains, which counteract the lipid peroxidation associated with alcohol drinking.
A total of 144 male Sprague-Dawley rats (Charles River, Cléon, France) of an initial body weight of 150 to 175 g was used for these studies. The animals were housed two per cage in a room with controlled temperature (23°C to 24°C). The animals were maintained in compliance with the INSERM policy on animal care and use, which is similar to the policy expressed in the National Research Council guidelines (NRC 1985).
Diet and Beverages
Rats were offered free access to their beverage and diet. After being equally distributed according to body weight into four groups (series 1 and 2) or three groups (series 3 and 4), they were given the following purified diet ad libitum: (grams per 100 g/body wt) casein 23, vitamin diet mixture (US Biochemical Corporation [USB]) 2, salt mixture Wesson (modification USB) 1.5, sawdust 6, sucrose 23, glucose 23, butter 21.5. This diet contained only a saturated fat because previous studies had shown that the effect of alcohol on platelet aggregation was observed primarily in animals16 and men14 on a diet high in saturated fat.
In series 1 (12 rats per group, except group 4 with only 6 rats), group 1 received commercial mineral water (Volvic; [mg/L] Ca 9.9, Mg 6.1, Na 9.4, K 5.7, HCO3 65.3) only, which was also used for diluting alcohol or wine. Group 2 received a 6% ethanol solution in the same water as group 1; group 3, a red wine solution in water to obtain a 6% ethanol content; and group 4, white wine, also diluted to 6% ethanol.
In series 2 (12 rats per group), group 1 received a 6% ethanol solution in the mineral water used in series 1; group 2, red wine diluted to 6% alcohol; group 3, a 6% ethanol solution containing 0.4% glycerol; and group 4, a 6% ethanol solution containing 0.025% of a grape-seed extract.
In series 3 (6 rats per group), group 1 received a 6% ethanol solution as in series 2; group 2, a red wine solution diluted to 6% alcohol; and group 3, a phenolic extract (0.03%) of the wine used in group 2 added to a 6% ethanol solution.
In series 4 (12 rats per group), group 1 received only the mineral water; group 2, red wine diluted to 6% alcohol adjusted for its content in glycerol to the level of the wine without alcohol; and group 3, wine without alcohol diluted to the same extent as the wine with alcohol of group 2. For preservation of the wine without alcohol, a higher level of glycerol was required.
To confirm the similarity of the tannin content (polyphenol) of the red wines used and of the alcohol solutions, spectrophotometry analysis of the samples was performed at 280 nm; concentration was adjusted to obtain the same optical density of the total phenols.
Wine with and without alcohol and wine extracts, which contained the phenolic compounds of the same wine used in series 3 and 4, were obtained from Institut National de Recherche Agronomique (Pech rouge, Narbonne, France). These wines were produced from Cabernet-Sauvignon vines.
Grape-seed extracts were obtained from Centre d’Experimentation Pharmaceutique (Leognan, France; Professor J. Masquelier) using the following technique. In brief, the polyphenols (procyanidins) from white grape seeds were extracted by acetone in water (30:70, vol/vol), with the water phase saturated by NaCl to remove high-molecular tannins. After filtration, the procyanidins were extracted by ethyl acetate, precipitated by chloroform, and dried in a vacuum.
Collection of Samples
After series 1, venipuncture was performed in animals not deprived and, 2 weeks later, in the same animals deprived of alcoholic beverages for 18 hours but given water. All animals were in a fasting condition. In series 2 to 4, the studies were performed only in animals that had alcoholic beverages withdrawn for 18 hours before blood removal and replaced by water.
Platelet aggregation was studied in platelet-rich plasma (PRP) adjusted to 600 000 platelets per microliter with platelet-poor plasma as previously described.20 Only aggregation to thrombin (from human plasma, Sigma Chemical Co) was performed by adding 100 μL of thrombin (0.08 U/mL PRP, final concentration) to 400 μL of PRP diluted to 500 μL, 1 minute before adding thrombin, by incomplete tyrode containing 148.9 mmol/L sodium chloride, 5.5 mmol/L glucose, 9.5 mmol/L sodium hydrogen carbonate, 2.7 mmol/L potassium chloride, 0.5 mmol/L sodium dihydrogen orthophosphate, and distilled water to 1000 mL, adjusted to pH 7.4, immediately before use.
Vitamins A and E were analyzed in plasma by high-performance liquid chromatography (Thermo Separation Products) on a spherisorb ODS2 (100×4.6 mm) column with a 3-μm particle size (Interchim). Serum levels of vitamin A and E were determined by a modification of the method of Vuilleumier et al.21 Serum proteins were precipitated by ethanol containing the internal standard (tocopherol acetate), and serum lipids were extracted by heptane. The lipid extract was evaporated under nitrogen, dissolved in methanol, and eluted by methanol with a flow rate of 1.5 mL/min. Response factors were calculated by external standardization with a mixture containing retinol, α-tocopherol, and tocopherol acetate standards.
Evaluation of Peroxidative Status
Serum peroxidative status was assessed by measuring 4-hydroxyalkenals and conjugated dienes.
4-Hydroxyalkenals were determined using a new kit from Bioxytech. It has been recognized that measurement of such aldehydes provides a satisfactory index of lipid peroxides. The method combines one molecule of 4-hydroxyalkenal with two molecules of reagent, yielding a stable chromophore with maximal absorbance at 586 nm.
Conjugated dienes were evaluated according to the methods of Quintanilha and Packer22 in citrated plasma and kept at −40°C for a few days until analysis.
Plasma lipid levels were determined in citrated plasma samples using enzymatic kits from BioMérieux for total and HDL cholesterol, an enzymatic kit from Merck (Merckotest) for triglycerides, and an enzymatic kit from Sigma for glycerol.
The data were analyzed using statview 512 (Brain Power Inc) computer program. Results are mean±SE. One-way ANOVA was used to examine the significance assigned to P<.05.
At the level of 6% alcohol, all alcoholic beverages were consumed by the rats in exactly the same amount as water alone after the switch from water to alcohol even in the first 24 hours. This indicates that the animals were not reluctant to consume the alcoholic beverages they were offered.
All determinations were performed after 2 and 4 months of diet and beverage consumption. Because the results were very similar, they have been pooled in the corresponding graphs and tables.
Series 1: Effects of Drinking Alcohol and Wine
As shown in Fig 1⇓, when the fasted animals were not deprived of their alcoholic beverage, thrombin-induced aggregation was reduced at about the same rate by alcohol, red wine, and white wine. By contrast, when the same fasted animals were deprived of their alcoholic beverage for 18 hours, a marked rebound effect (124% increase in response) was observed with alcohol, a moderate effect (46% increase) with white wine, but no rebound at all with red wine because the aggregation was still reduced by 59% compared with water.
Concerning the plasmatic variables (Table 1⇓), triglycerides were significantly increased with wine but only in animals not deprived of the alcoholic beverages, and the level of vitamin E decreased only with drinking alcohol compared with red wine. Conjugated dienes were significantly higher in the group drinking alcohol.
Series 2: Comparative Effects of Glycerol and Grape-Seed Extracts Added to Alcohol
The main two components of red wine in addition to alcohol are glycerol and tannins. In the present study, the tannins used were extracted from grape seed and added to alcohol (0.025%) at approximately the level contained in red wine as determined by spectrophotometry. Glycerol was also added to alcohol (0.4%).
Fig 2⇓ indicates that the inhibitory effect of red wine on thrombin-induced aggregation can be completely reproduced by adding the tannins extracted from grape seed to alcohol. Glycerol added to alcohol did not appear to change the response of platelets to aggregation observed with alcohol alone.
Nevertheless, addition of glycerol to alcohol decreased the level of lipid peroxides (Table 2⇓) and increased the ratio of vitamin E to triglycerides to about the same extent as red wine or tannins added to alcohol. In addition, only red wine and grape-seed extracts added to alcohol were significantly associated with a lower level of conjugated dienes.
Series 3: Comparative Effects of Red Wine and Wine Extracts Added to Alcohol
The purpose of this series was to determine whether the tannins directly extracted from wine had a similar effect in our study as the tannins extracted from grape seeds. The results in Fig 3⇓ indicate that tannins extracted from wine had at least the same protective effect on platelets as the tannins from grape seeds. As shown in Table 3⇓, both the red wine and the wine extracts (added to alcohol) similarly reduced the level of lipid peroxides and conjugated dienes compared with alcohol alone.
Series 4: Comparative Effects of Red Wine With and Without Alcohol
The aim of this study was to determine whether wine without alcohol could have inhibitory effects on platelet reactivity and beneficial effects on peroxidation. As shown in Fig 4⇓, only the wine with alcohol was associated with a decrease in platelet aggregation compared with water. In contrast, the wine without alcohol increased the response of platelets to aggregation, a result significant only when compared with results of wine with alcohol.
The only blood level variable that was different in this series (Table 4⇓) was cholesterol, which was significantly lower in the group receiving wine without alcohol.
The main findings of the present study in rats are that the drinking of red wine, when compared with alcohol, is not associated with the rebound effect on platelet reactivity that has been observed by several investigators in humans who drink spirits or beer.13 14 17 In humans, the platelet rebound effect in alcoholics has been attributed to parallel changes in the ratio of LDL cholesterol to HDL cholesterol.17 In the present study in rats, with somewhat moderate drinking, there was no change in the level of HDL or cholesterol, except in the group given wine without alcohol. The protective effect of red wine appears to be essentially associated with tannins (polyphenols) extracted either from grape seeds or from wine itself. This protection was related to an inhibition of the known increased lipid peroxidation observed with alcohol drinking,23 as shown here by the higher level of lipid peroxides and conjugated dienes in animals drinking alcohol compared with that of those drinking water. By contrast, the animals drinking red wine exhibited levels of lipid peroxides, conjugated dienes, and vitamin E similar to those of the animals drinking water. Nevertheless, with red wine (or tannins added to alcohol), the inhibitory effect of alcohol on platelet reactivity observed after alcohol drinking was preserved for a much longer time. This may explain why in certain studies in humans deprived of alcohol for 24 hours, only red wine was associated with inhibition of platelet reactivity.24
The present study demonstrates also that in vivo, in animals, the tannins from grapes and red wine exert antioxidant properties. Nevertheless, the sparing effect of tannins on the plasma level of vitamin E, also observed recently in France25 mostly in moderate wine drinkers, was not noted in rats drinking red wine without alcohol. In addition, the wine without alcohol was associated with a significant increase in platelet aggregability, which was not explained in the present study by the variables determined. Whether the tannins were not absorbed in the absence of alcohol, or whether the wine was altered by the process that removed alcohol, has to be investigated in further studies.
The evaluation of white wine in our study was done for only six animals and one sample of white wine. Thus, additional studies are required to evaluate more extensively the potential effect of white wine. In the present study, its effect was intermediate between that of alcohol and of red wine.
Further studies are needed also to determine which substances in tannins are responsible for their antioxidant properties. Procyanidins preserved in the extraction process used here have been shown to exert oxygen free radical scavenger capacities26 and seem to be good candidates.
The present study, if reproduced in humans, could explain at least part of the long-term advantage of red wine over other alcoholic beverages. Because higher responses of platelets to thrombin and ADP-induced aggregation have been associated with prevalent cases of myocardial infarction,27 the rebound effect associated with alcohol drinking might be related to the increased risk of binge drinkers for CHD, sudden death,28 29 and stroke.18
We wish to thank Professor J. Masquelier for supplying grape-seed extract and Professors S. Brun and M. Bourzeix for the supply of wine extracts and the wine with and without alcohol used in series 3 and 4 of the present study.
Renaud S, Criqui MH, Farchi G, Veenstra J. Alcohol drinking and coronary heart disease. In: Verschuren PV, ed. Health Issues Related to Alcohol Consumption. Washington, DC: ILSI Press; 1993:81-124.
Klatsky AL, Armstrong MA, Friedman GD. Alcohol and mortality. Ann Intern Med. 1992;117:646-654.
Blot WJ, McLaughlin K, Winn DM, Austin DF, Greenberg RS, Preston-Martin S, Bernstein L, Schoenberg JB, Stemhagen A, Fraumani JF Jr. Smoking and drinking in relation to oral and pharyngeal cancer. Cancer Res. 1988;48:3282-3287.
Sharp D. Coronary disease: when wine is red. Lancet. 1993;343:27-28.
Suh IL, Shaten BJ, Cutler JA, Kuller LH. Alcohol use and mortality from coronary heart disease: the role of high-density lipoprotein cholesterol. Ann Intern Med. 1992;116:881-887.
Meade TW, Vickers MV, Thompson SG, Stirling Y, Haines AP, Miller GJ. Epidemiological characteristics of platelet aggregability. Br Med J. 1985;290:428-432.
Renaud S, Beswick AD, Fehily AM, Sharp DS, Elwood PC. Alcohol and platelet aggregation: the Caerphilly prospective heart disease study. Am J Clin Nutr. 1992;55:1012-1017.
Desai K, Owen JS, Wilson DT, Hutton RA. Platelet aggregation and plasma lipoproteins in alcoholics during alcohol withdrawal. Thromb Haemost. 1986;56:173-177.
Hillbom ME. What supports the role of alcohol as a risk factor for stroke? Acta Med Scand. 1987;717(suppl):93-106.
Renaud S, Dumont E, Godsey F, Supplisson A, Thevenon C. Platelet functions in relation to dietary fats in farmers from two regions of France. Thromb Haemost. 1978;40:518-531.
Seigneur M, Bonnet J, Dorian B, Benchimol D, Drouillet F, Gouverneur G, Larrue J, Crockett R, Boisseau MR, Ribereau-Gayon P, Bricaud H. Effect of the consumption of alcohol, white wine and red wine on platelet function and serum lipids. J Appl Cardiol. 1990;5:215-222.
Leconte E, Herbeth B, Pirollet P, Chancerelle Y, Arnaud J, Musse N, Paille F, Siest G, Artur Y. Effect of alcohol consumption on blood antioxidant nutrients and oxidative stress indicators. Am J Clin Nutr. 1994:60:255-261.
Elwood PC, Renaud S, Sharp DS, Beswick AD, O’Brien J, Yarnell JWG. Ischaemic heart disease and platelet aggregation: the Caerphilly collaborative heart disease study. Circulation. 1991;83:38-44.
Peterson B. Analysis of the role of alcohol in mortality, particularly sudden unwitnessed death, in middle-aged men in Malmö, Sweden. Alcohol Alcohol. 1988;23:259-263.