Inverse correlation between alcohol consumption and lymphocyte ...

14 downloads 0 Views 103KB Size Report
Mean levels of total alcohol for the European Prospective Investigation into Cancer and intake and of types of alcoholic beverages consumed. Nutrition (EPIC; a ...
Carcinogenesis vol.22 no.6 pp.885–890, 2001

Inverse correlation between alcohol consumption and lymphocyte levels of 8-hydroxydeoxyguanosine in humans

Franca Bianchini, Andro Jaeckel1, Paolo Vineis2, Carmen Martinez-Garcia´3, So¨lve Elmstahl4, Anne-Linda van Kappel, Heiner Boeing1, Hiroshi Ohshima, Elio Riboli and Rudolf Kaaks5 International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon, Cedex 08, France, 1Deutsches Institut fu¨r Erna¨hrungsforschung, Arthur Scheunert Allee 114-116, 14558 Potsdam-Rehbru¨cke, Germany, 2Servizio di Epidemiologia dei Tumori, Via Santena 7, 10126 Turin, Italy, 3Granada Cancer Registry, Escuela Andaluza de Salud Publica, E-18080 Granada, Spain and 4Department of Community Medicine, Lund University, Malmo ¨ University Hospital, S-20502 Malmo¨, Sweden 5To

whom correspondence should be addressed Email: [email protected]

In a cross-sectional study of 115 premenopausal nonsmoking women, we examined the relationship between lymphocyte levels of 8-hydroxy-2⬘-deoxyguanosine (8oxodGuo) and habitual alcohol consumption. The study was conducted in four different regions of Europe, including Potsdam (Germany), Turin (Italy), Malmo¨ (Sweden) and Granada (Spain). Mean 8-oxodGuo levels differed significantly across study centres (P ⍧ 0.001), with the highest levels in Granada [2.17 8-oxodGuo⍥10–6 2⬘-deoxyguanosine (95% confidence interval 1.27–4.40)] and lowest levels in Turin [1.19 (0.36–4.29)]. Mean levels of total alcohol intake and of types of alcoholic beverages consumed (wine, fortified wines, beer and cider) also differed across the study centres (P < 0.05), with the highest total alcohol consumption in Turin, and the lowest intake in Granada. When combining all the data, but adjusting for study centre, individual 8-oxodGuo level correlated inversely with alcohol intake. This inverse association remained unaltered after further adjustment for Quetelet Index, fruit and vegetable consumption, and plasma carotenoid levels. Furthermore, the inverse association was also observed for each of the study centres separately, and for different beverage types, with the exception of Granada, where the majority of women were non-drinkers and where alcohol intakes were also very low for the consumers. Finally, on a group level, mean levels of 8-oxodGuo and alcohol intake were also inversely associated between the four study centres. The finding of a relationship between alcohol consumption and 8-oxodGuo in lymphocytes was unexpected and not based on a prior hypothesis. This finding consequently requires confirmation from a randomized intervention study.

Oxidative damage to macromolecules, including DNA, proteins and lipids, has been implicated in the etiology of aging and various age-related diseases, including cardiovascular diseases Abbreviations: dG, 2⬘-deoxyguanosine; 8-oxodGuo, 8-hydroxy-2⬘-deoxyguanosine; EPIC, European Prospective Investigation into Cancer and Nutrition; GLM, generalized linear regression models. © Oxford University Press

and cancer (1). Several studies have shown that these forms of oxidative damage can be modulated by environmental and dietary factors. For example, cigarette smoking has been found to have a pro-oxidant effect, and was positively associated with levels of 8-hydroxy-2⬘-deoxyguanosine (8-oxodGuo) in lymphocyte DNA (2), and with urinary levels of 8-oxodGuo (3) and 5-hydroxymethyluracil (4). Among dietary factors, a rise in dietary fat intake was found to increase leukocyte levels of 5-hydroxymethyl-2⬘-deoxyuridine (5), whereas supplementation with antioxidants (6,7), carrot juice (8) or brussel sprouts (9) decreased levels of oxidative damage measured in lymphocytes or urine. Nevertheless, there have also been paradoxical findings. For example, we recently found that lymphocyte levels of 8-oxodGuo were not lower, but higher in a group of women consuming a typical Mediterranean diet, rich in natural antioxidants such as vitamin C, carotenoids and tocopherols compared with women with a Northern European dietary intake pattern (10). To investigate further the relationships between habitual diet and oxidative DNA damage, we extended this study to two additional research centres, Potsdam (Germany) and Turin (Italy). In this paper we report some unexpected findings, relating habitual moderate alcohol consumption to reduced lymphocyte levels of 8-oxodGuo. Premenopausal women, aged 45–50, previously recruited for the European Prospective Investigation into Cancer and Nutrition (EPIC; a multi-centre prospective cohort study on diet and cancer), were selected from four of the EPIC study centres, in Germany (Potsdam), Italy (Turin), Sweden (Malmo¨) and Spain (Granada). All the women selected were nonsmokers, and none of them had used any vitamin supplements, or had any form of chronic inflammatory disease, hypertension or diabetes. Information on habitual diet was obtained by previously validated food frequency questionnaires (11). The amount (ml) of alcohol consumed with different alcoholic beverages was computed assuming 11 vol% alcohol in wine, 16 vol% in fortified wines, 4 vol% in beer and cider and 40 vol% in spirits and brandy. A fasting blood sample, containing heparin as anticoagulant, was collected for laboratory analyses. Immediately after the blood draw, plasma and lymphocytes were prepared by centrifugation on Lymphoprep tubes (Nycomed Pharma AS, Oslo, Norway) and stored at –80°C. Lymphocyte levels of 8-oxodGuo were measured in a single laboratory at IARC by HPLC with electrochemical detection, as previously described (10). The analyses were performed in batches of 12 DNA samples per day; each batch included in random order an equal number of subjects either from Malmo and Granada, or from Potsdam and Turin. The mean storage times before analyses were, respectively, 30.34 ⫾ 2.49 and 29.70 ⫾ 2.38 months for Malmo¨ and Granada, and 18.09 ⫾ 1.69 and 11.31 ⫾ 1.75 months for Potsdam and Turin, respectively. In order to avoid artefactual oxidation of guanine, especially during the DNA extraction steps (12), 0.1 mM deferoxamine was added to all buffers. In addition, samples were always processed on ice and all centrifugation 885

F.Bianchi et al.

Table I. Subject characteristics Potsdam (n ⫽ 27) Age (years)a Height (cm)a Weight (kg)a Quetelet Indexa (kg/m2) 8-OxodGuob (⫻ 10–6 dg)

47.4 164.0 68.1 25.34 1.29

⫾ 1.6c ⫾ 4.8c,d ⫾ 11.7c ⫾ 4.3d,e (0.37–6.23)

Turin (n ⫽ 36) 47.7 161.2 61.7 23.7 1.19

⫾ 1.8c ⫾ 6.7d ⫾ 8.2d ⫾ 2.9e (0.36–4.29)

Malmo¨ (n ⫽ 24) 47.8 164.7 73.0 26.8 1.31

⫾ 1.6c ⫾ 6.2c ⫾ 15.1c ⫾ 4.8c,d (0.41–4.30)

Granada (n ⫽ 28) 47.9 157.8 68.8 27.6 2.17

⫾ 1.6c ⫾ 4.6e ⫾ 9.3c ⫾ 3.1c (1.27–4.40)

⫾ SD. mean, numbers in brackets are the minimum and maximum values. c,d,eMean values of centres with different alphabetical letters are significantly different (P ⬍ 0.05). aMeans

bGeometric

steps were performed at 4°C. In some samples 8-oxodGuo was not detectable, either because the amount of extracted DNA was not sufficient for the chemical analysis or because the chromatographic peak of 8-oxodGuo presented some interference. These subjects were excluded from statistical analyses. These exclusions, however, did not change significantly the average subject characteristics of age, height, body weight or mean levels of alcohol consumption. Carotenoids and α-tocopherol were measured in plasma using HPLC with a diode array detector (13). The data were analysed by generalized linear regression models (GLM), with 8-oxodGuo levels as a dependent variable, and ‘centre’, storage time and total alcohol consumption as predictor variables. Additional variables added to the models for further adjustments included Quetelet Index (weight/height2), total fruit and vegetable consumption, and plasma total carotenoids (the sum of α-carotene, β-carotene, β- cryptoxanthin, canthaxanthin, lutein, lycopene, zeaxanthin and retinol), and plasma α- and γ-tocopherols. All analyses were performed on 8-oxodGuo values that had been log-transformed to normalize their population frequency distributions. Basic characteristics of study participants and means and ranges of 8-oxodGuo are in Table I. Women had the same average age in all four centres, but had a lower Quetelet Index in Turin and Potsdam than in Granada (P ⬍ 0.0004), and were taller in Potsdam, Turin and Malmo¨ than in Granada (P ⬍ 0.0001). Levels of 8-oxodGuo ranged from 0.36 to 6.23 8-oxodGuo⫻10–6 2⬘-deoxyguanosine (dG), and thus showed high inter-individual variation, consistent with previously reported data (2,14,15). On an absolute scale, these levels were relatively low compared with those measured in other laboratories by similar techniques (14–17). These relatively low absolute levels suggest that there was not considerable artefactual formation of 8-oxodGuo during sample processing and analysis. Geometric means in the four study centres were 1.29 8-oxodGuo⫻10–6 dG (95% confidence interval 0.96–1.72) for Potsdam, 1.19 (0.96–1.48) for Turin, 1.31 (1.01–1.70) for Malmo¨ and 2.17 (1.92–2.46) for Granada. Analysis of variance by GLM models (see also below) indicated that the mean levels differed significantly between the four centres (P ⍧ 0.001), with the strongest contrast between Granada and the other three centres. Average sample storage times differed between the centres, but within centres showed no significant correlation with 8-oxodGuo levels. Mean alcohol intake calculated for all subjects (either drinkers or non-drinkers), was highest in Turin, where alcohol intake was mainly derived from wine, whereas intake was particularly low in Granada (Table II). The low average alcohol consumption in Granada was mainly due to a relatively large 886

proportion of non-drinkers: only 12 out of 28 women in Granada reported drinking alcoholic beverages; nine reported drinking wine and only four reported drinking beer. On the contrary, all subjects in Potsdam were drinkers and only five out of 36 and seven out of 24 were non-drinkers in Turin and Malmo, respectively. In Potsdam and Malmo¨ levels of total alcohol consumption were quite similar, and in these two centres beer consumption contributed to a higher percentage of total alcohol intake than in Turin. Differences between the four centres in the consumption levels of total alcohol were highly significant (P ⫽ 0.001), while there were also clear differences in the consumption levels of specific beverages. The relationship between alcohol intake and 8-oxodGuo level was analysed both at the level of individuals, with adjustment for study centre, and at the level of study centres by comparing average alcohol intakes and mean 8-oxodGuo levels for each of the four study populations. At the individual level, a negative relationship between 8-oxodGuo and total alcohol intake was observed in three centres (Malmo¨ , Potsdam and Turin) where most women reported regular alcohol consumption (Figure 1). In addition, when comparing mean alcohol intakes and mean 8-oxodGuo values between the four study centres an inverse association on a group level appeared (Figure 2). Multivariate analysis by GLM, all study subjects combined (n ⫽ 115), showed a significant effect of the variable ‘study centre’ (P ⫽ 0.001), which explained about 13% of variation in 8-oxodGuo levels (model R-square of 0.134) (Table III). Entering total alcohol intake into the model diminished the effect of ‘study centre’ (partial R-square of 0.063), but showed a significant effect of alcohol itself (P ⫽ 0.02), with a partial R-square of 0.043 adjusted for the centre effect. For alcohol intake (a continuous variable), the square root of this partial R-square indicates the average correlation within study centres of 8-oxodGuo with alcohol intake (–0.21). The inverse association between alcohol intake and 8-oxodGuo levels remained unaffected after further adjustments for sample storage time, Quetelet Index, total fruit and vegetable consumption, and plasma levels of total carotenoids and tocopherols. It is worth noting that in the latter models sample storage time was unrelated to 8-oxodGuo levels; however, total plasma carotenoids and tocopherols were positively associated with 8-oxodGuo levels, confirming our earlier observations for Malmo¨ and Granada only (10). Considering different beverage types, 8-oxodGuo levels were inversely related to levels of consumption of all types of beverage combined, the negative correction being more important, although not always significant for beer and spirits in Potsdam, wine and fortified wines in Turin, wine in Malmo¨ and beer in Granada (Table

Alcohol consumption and 8-oxodGuo levels

Table II. Daily consumption (ml) of alcohol from different alcoholic beverages in Potsdam, Turin, Malmo¨ and Granada (mean ⫾ SD) Alcoholic beverage

Potsdam (n ⫽ 27)

Total Wine Fortified wines Beer, cider Spirits, brandy

9.9 6.7 0.05 3.0 0.2

a,b,cMean

⫾ ⫾ ⫾ ⫾ ⫾

9.4a 9.1a,b 0.06b 4.9a 0.6b

Turin (n ⫽ 36)

Malmo¨ (n ⫽ 24)

13.1 ⫾ 15.7a 11.4 ⫾ 15.2a 0.2 ⫾ 0.3a 0.7 ⫾ 1.2b 0.7 ⫾ 1.4a,b

9.7 4.5 – 4.0 1.2

⫾ 11 a ⫾ 6.3b,c ⫾ 4.6a ⫾ 3.3a

Granada (n ⫽ 28) 1.0 ⫾ 2b 0.6 ⫾ 1.5c – 0.4 ⫾ 1.2b –

values of centres with different alphabetical letters are significantly different (P ⬍ 0.05).

Fig. 1. Relationships between an individual’s alcohol intake and 8-oxodGuo levels in lymphocyte DNA, within each of the four study centres.

IV). This suggests that the inverse association of alcohol intake with 8-oxodGuo levels was independent of beverage type. In summary, we observed a negative relationship between level of alcohol intake and oxidative DNA damage, as measured by 8-oxodGuo levels in lymphocytes. This finding was unexpected, and was not based on a prior hypothesis. The inverse relationship was most evident in three of the four centres (Potsdam, Turin and Malmo¨ ) where the proportions of alcohol drinkers were high. In addition, mean alcohol intakes were correlated inversely with mean 8-oxodGuo levels across the four study populations. Interestingly, this inverse association between alcohol intake and oxidative DNA damage appeared to be independent of beverage type. The mechanisms that may underlie our findings are not clear. On the one hand, ethanol is known as a potent hydroxyl radical scavenger in vitro (18). On the other hand, several experimental studies in vivo have shown pro-oxidant effects of pure ethanol. In rats, ethanol feeding increased the production of reactive oxygen species, lipid peroxidation and

DNA fragmentation in the gastrointestinal tract (19), increased 8-oxodGuo levels in hepatic mitochondrial DNA (20,21) and induced DNA strand breaks in brain tissue (22). Likewise, in mice, intragastric doses of ethanol increased lipid peroxidation and degradation of hepatic mitochondrial DNA (23). In contrast, a moderate consumption of beer did not affect plasma antioxidant status in rats, but increased their resistance to lipid peroxidation (24), whereas another study showed no difference in the oxidative status of rats supplemented with wine (25). Studies in humans on the effect of alcohol consumption on oxidative damage also produced contradictory results. A positive correlation between alcohol intake and levels of 8oxodGuo in lymphocytes was found in subjects deficient in aldehyde dehydrogenase-2 (14), an enzyme that metabolizes the alcohol-derived acetaldehyde. In another study, alcohol consumption was also associated with increased DNA damage as measured by the COMET assay (26). In contrast, supplementation with moderate amounts of red wine decreased leukocyte 8-oxodGuo and increased total antioxidant capacity 887

F.Bianchi et al.

Fig. 2. Association between geometric mean 8-oxodGuo levels and mean alcohol consumption between the four study centres.

Table III. Partial R-squares, in multivariate linear regression models for 8-oxodGuoa

(1) Centre (2) Centre, alcohol (3) Centre, alcohol, storage time (4) Centre, alcohol, storage time, Quetelet Index (5) Centre, alcohol, storage time, Quetelet Index, fruit & vegetables (6) Centre, alcohol, storage time, Quetelet Index, fruit & vegetables, total carotenoids, tocopherols

Centre

Alcohol

0.134 P ⫽ 0.001 0.063 P ⫽ 0.04 0.042 P ⫽ 0.14 0.042 P ⫽ 0.15 0.038 P ⫽ 0.18 0.025 P ⫽ 0.33

0.043 (-0.21) P ⫽ 0.02 0.044 (-0.21) P ⫽ 0.018 0.040 (-0.20) P ⫽ 0.025 0.039 (-0.20) P ⫽ 0.03 0.035 (-0.19) P ⫽ 0.03

Sample storage time

Quetelet index

Fruits & vegetables

Plasma carotenoids & tocopherols

0.00017 (0.01) P ⫽ 0.88 0.00013 (0.01) P ⫽ 0.90 0.00085 (0.03) P ⫽ 0.74 0.00078 (0.03) P ⫽ 0.74

0.0012 (0.03) P ⫽ 0.70 0.00069 (0.03) P ⫽ 0.77 0.0065 (0.08) P ⫽ 0.36

0.015 (0.12) P ⫽ 0.39 0.027 (0.16) P ⫽ 0.16

0.054 (0.23) P ⫽ 0.03

aThe

partial R-square (R2) represents the percent of variation that is explained by a variable or subset of variables, adjusting for variation attributable to other variables in the model. 8-OxodGuo measurements were log-transformed before analyses. Data between parentheses are the square roots of the R2 estimates; these indicate the partial correlation of 8-oxodGuo with each given (continuous) variable, after adjustment for the other variables in the model.

in a group of volunteers consuming a Mediterranean diet (27), and still other studies showed no correlation at all between alcohol intake and 8-oxodGuo in lymphocytes (15) or between alcohol drinking and 8-oxodGuo in leukocytes from buffy coat (28). The discrepancies between these previously reported findings might be due to differences in the methods of 8-oxodGuo measurement, the types of alcoholic beverages considered and the level and duration of consumption. Furthermore, differences in genetic susceptibility—for example, polymorphisms in aldehyde dehydrogenase (14) or DNA repair genes (29,30)—might also explain some of the discrepancies. 888

There are arguments in favour of wine, especially red wine, being superior to other alcoholic drinks in protecting against oxidative damage, because of a high content of polyphenolic compounds (flavonoids and non-flavonoid components) (31). These phytochemicals are antioxidants and free radical scavengers and inhibit lipid peroxidation processes (32). Administration of wine complex polyphenols and tannins have been shown to reduce 8-oxodGuo in rat liver DNA (33) and phenolic substances in red wine have demonstrated a potent antioxidant effect on low-density lipoproteins (34). Interestingly, however, our results indicate an inverse correlation

Alcohol consumption and 8-oxodGuo levels

Table IV. Pearson’s correlation coefficient between 8-oxodGuo and consumption of different alcoholic beverages Alcoholic beverage

Total# (n ⫽ 115)

Potsdam (n ⫽ 27)

Turin (n ⫽ 36)

Malmo¨ (n ⫽ 24)

Granada (n ⫽ 28)

Total Wine Fortified wines Beer, cider Spirits, brandy

⫺0.21b ⫺0.13 ⫺0.12 ⫺0.25a ⫺0.12

⫺0.22 0.03 ⫺0.03 ⫺0.46b ⫺0.25

⫺0.25 ⫺0.23 ⫺0.23 ⫺0.19 ⫺0.11

⫺0.24 ⫺0.27 – ⫺0.09 ⫺0.18

⫺0.12 0.16 – ⫺0.37b –

#Adjusted for aP ⬍ 0.01. bP

study centre.

⬍ 0.05.

between level of 8-oxodGuo and alcohol intake, regardless of the type of alcoholic beverage. This suggests a possible protective effect due to alcohol itself, although it cannot be ruled out that such effect was due to other anti-oxidant components present in these different beverage types. To our knowledge, this cross-sectional study is the largest conducted so far, and it shows a remarkable consistency of findings across study centres, both at the level of the individual and at the group level. However, although our findings persisted after adjustments for the possible confounding effects of total fruit and vegetable consumption and plasma carotenoid levels, our study still has all the usual limitations of cross-sectional studies, in that our results could still be confounded by other, unidentified factors. Such confounding effects may only be overcome entirely by randomized intervention studies, where subjects are given different amounts and types of alcoholic beverages for various periods of time. Our unexpected findings may provide a strong incentive for the conduct of such intervention studies. Without confirmation from intervention studies and before having more detailed insight into the mechanisms of possible effects of moderate alcohol intake on oxidative stress parameters, it would be unwise to advocate the consumption of alcohol to diminish the risk of chronic diseases potentially related to oxidative processes. Acknowledgements We gratefully acknowledge the assistance of Carine Biessy with all statistical analyses, and the financial support of the World Cancer Research Fund (grant no. 97A55) and Europe Against Cancer Programme of the European Commission.

References 1. Ames,B.N., Shigenaga,M.K. and Hagen,T.M. (1993) Oxidants, antioxidants, and the degenerative diseases of aging. Proc. Natl Acad. Sci. USA, 90, 7915–7922. 2. Asami,S., Hirano,T., Yamaguchi,R., Tomioka,Y., Itoh,H. and Kasai,H. (1996) Increase of a type of oxidative DNA damage, 8-hydroxyguanine, and its repair activity in human leukocytes by cigarette smoking. Cancer Res., 56, 2546–2549. 3. Loft,S., Vistisen,K., Ewertz,M., Tjonneland,A., Overvad,K. and Poulsen,H.E. (1992) Oxidative DNA damage estimated by 8-hydroxydeoxyguanosine excretion in humans: influence of smoking, gender and body mass index. Carcinogenesis, 13, 2241–2247. 4. Bianchini,F., Donato,F., Faure,H., Ravanat,J.L., Hall,J. and Cadet,J. (1998) Urinary excretion of 5-(hydroxymethyl) uracil in healthy volunteers: effect of active and passive tobacco smoke. Int. J. Cancer, 77, 40–46.

5. Djuric,Z., Heilbrun,L.K., Reading,B.A., Boomer,A., Valeriote,F.A. and Martino,S. (1991) Effects of a low-fat diet on levels of oxidative damage to DNA to human peripheral nucleated blood cells. J. Natl Cancer Inst., 83, 766–769. 6. Duthie,S.J., Ma,A., Ross,M.A. and Collins,A.R. (1996) Antioxidant supplementation decreases oxidative DNA damage in human lymphocytes. Cancer Res., 56, 1291–1295. 7. Howard,D.J., Ota,R.B., Briggs,L.A., Hampton,M. and Pritsos,C.A. (1998) Oxidative stress induced by environmental tobacco smoke in the workplace is mitigated by antioxidant supplementation. Cancer Epidemiol. Biomarkers Prev., 7, 981–988. 8. Pool-Zobel,B.L., Bub,A., Muller,H., Wollowski,I. and Rechkemmer,G. (1997) Consumption of vegetables reduces genetic damage in humans: first results of a human intervention trial with carotenoid-rich foods. Carcinogenesis, 18, 1847–1850. 9. Verhagen,H., Poulsen,H.E., Loft,S., van Poppel,G., Willems,M.I. and van Bladeren,P.J. (1995) Reduction of oxidative DNA-damage in humans by brussels sprouts. Carcinogenesis, 16, 969–970. 10. Bianchini,F., Elmstahl,S., Martinez-Garcia,C., van Kappel,A.L., Douki,T., Cadet,J., Ohshima,H., Riboli,E. and Kaaks,R. (2000) Oxidative DNA damage in human lymphocytes: correlations with plasma levels of αtocopherol and carotenoids. Carcinogenesis, 21, 321–324. 11. Kaaks,R., Slimani,N. and Riboli,E. (1997) Pilot phase studies on the accuracy of dietary intake measurements in the EPIC project: overall evaluation of results. European Prospective Investigation into Cancer and Nutrition. Int. J. Epidemiol., 26 (suppl. 1), S26–S36. 12. Cadet,J., Douki,T. and Ravanat,J.L. (1997) Artifacts associated with the measurement of oxidized DNA bases. Environ. Health Perspect., 105, 1034–1039. 13. Steghens,J.P., van Kappel,A.L., Riboli,E. and Collombel,C. (1997) Simultaneous measurement of seven carotenoids, retinol and α-tocopherol in serum by high-performance liquid chromatography. J. Chromatogr. B Biomed. Sci. Appl., 694, 71–81. 14. Nakajima,M., Takeuchi,T., Takeshita,T. and Morimoto,K. (1996) 8Hydroxydeoxyguanosine in human leukocyte DNA and daily health practice factors: effects of individual alcohol sensitivity. Environ. Health Perspect., 104, 1336–1338. 15. van Zeeland,A.A., de Groot,A.J., Hall,J. and Donato,F. (1999) 8Hydroxydeoxyguanosine in DNA from leukocytes of healthy adults: relationship with cigarette smoking, environmental tobacco smoke, alcohol and coffee consumption. Mutat. Res., 439, 249–257. 16. Chen,L., Bowen,P.E., Berzy,D., Aryee,F., Stacewicz-Sapuntzakis,M. and Riley,R.E. (1999) Diet modification affects DNA oxidative damage in healthy humans. Free Radic. Biol. Med., 26, 695–703. 17. Lenton,K.J., Therriault,H., Fulop,T., Payette,H. and Wagner,J.R. (1999) Glutathione and ascorbate are negatively correlated with oxidative DNA damage in human lymphocytes. Carcinogenesis, 20, 607–613. 18. Cohen,G. (1977) A novel route for the metabolism of ethanol: the oxidation of ethanol by hydroxyl free radicals. In Thurman,R.G., Drott,H., Williamson,J.R. and Chance,B. (eds) Alcohol and Aldehyde Metabolizing Systems. Academic Press, New York, p. 403. 19. Bagchi,D., Carryl,O.R., Tran,M.X., Krohn,R.L., Bagchi,D.J., Garg,A., Bagchi,M., Mitra,S. and Stohs,S.J. (1998) Stress, diet and alcohol-induced oxidative gastrointestinal mucosal injury in rats and protection by bismuth subsalicylate. J. Appl. Toxicol., 18, 3–13. 20. Cahill,A., Wang,X. and Hoek,J.B. (1997) Increased oxidative damage to mitochondrial DNA following chronic ethanol consumption. Biochem. Biophys. Res. Commun., 235, 286–290. 21. Cahill,A., Stabley,G.J., Wang,X. and Hoek,J.B. (1999) Chronic ethanol consumption causes alterations in the structural integrity of mitochondrial DNA in aged rats. Hepatology, 30, 881–888. 22. Renis,M., Calabrese,V., Russo,A., Calderone,A., Barcellona,M.L. and Rizza,V. (1996) Nuclear DNA strand breaks during ethanol-induced oxidative stress in rat brain. FEBS Lett., 390, 153–156. 23. Mansouri,A., Gaou,I., De Kerguenec,C., Amsellem,S., Haouzi,D., Berson,A., Moreau,A., Feldmann,G., Letteron,P., Pessayre,D. and Fromenty,B. (1999) An alcoholic binge causes massive degradation of hepatic mitochondrial DNA in mice. Gastroenterology, 117, 181–190. 24. Gasbarrini,A., Addolorato,G., Simoncini,M., Gasbarrini,G., Fantozzi,P., Mancini,F., Montanari,L., Nardini,M., Ghiselli,A. and Scaccini,C. (1998) Beer affects oxidative stress due to ethanol in rats. Dig. Dis. Sci., 43, 1332–1338. 25. Cestaro,B., Simonetti,P., Cervato,G., Brusamolino,A., Gatti,P. and Testolin,G. (1996) Red wine effects on peroxidation indexes of rat plasma and erythrocytes. Int. J. Food Sci. Nutr., 47, 181–189.

889

F.Bianchi et al. 26. Ye,X.B., Fu,H., Zhu,J.L., Ni,W.M., Lu,Y.W., Kuang,X.Y., Yang,S.L. and Shu,B.X. (1999) A study on oxidative stress in lead-exposed workers. J. Toxicol. Environ. Health, 57, 161–172. 27. Leighton,F., Cuevas,A., Guasch,V., Perez,D.D., Strobel,P., San Martin,A., Urzua,U., Diez,M.S., Foncea,R., Castillo,O., Mizon,C., Espinoza,M.A., Urquiaga,I., Rozowski,J., Maiz,A. and Germain,A. (1999) Plasma polyphenols and antioxidants, oxidative DNA damage and endothelial function in a diet and wine intervention study in humans. Drugs Exp. Clin. Res., 25, 133–141. 28. Lodovici,M., Casalini,C., Cariaggi,R., Michelucci,L. and Dolara,P. (2000) Levels of 8-hydroxydeoxyguanosine as a marker of DNA damage in human leukocytes. Free Radic. Biol. Med., 28, 13–17. 29. Minowa,O., Arai,T., Hirano,M., Monden,Y., Nakai,S., Fukuda,M., Itoh,M., Takano,H., Hippou,Y., Aburatani,H., Masumura,K., Nohmi,T., Nishimura,S. and Noda,T. (2000) Mmh/Ogg1 gene inactivation results in accumulation of 8-hydroxyguanine in mice. Proc. Natl Acad. Sci. USA, 97, 4156–4161. 30. Boiteux,S. and Radicella,J.P. (1999) Base excision repair of 8-

890

hydroxyguanine protects DNA from endogenous oxidative stress. Biochimie, 81, 59–67. 31. van Golde,P.H., Sloots,L.M., Vermeulen,W.P., Wielders,J.P., Hart,H.C., Bouma,B.N. and van de Wiel,A. (1999) The role of alcohol in the anti low density lipoprotein oxidation activity of red wine. Atherosclerosis, 147, 365–370. 32. Rice-Evans,C.A. and Miller,N.J. (1996) Antioxidant activities of flavonoids as bioactive components of food. Biochem. Soc. Trans., 24, 790–795. 33. Casalini,C., Lodovici,M., Briani,C., Paganelli,G., Remy,S., Cheynier,V. and Dolara,P. (1999) Effect of complex polyphenols and tannins from red wine (WCPT) on chemically induced oxidative DNA damage in the rat. Eur. J. Nutr., 38, 190–195. 34. Frankel,E.N., Kanner,J., German,J.B., Parks,E. and Kinsella,J.E. (1993) Inhibition of oxidation of human low-density lipoprotein by phenolic substances in red wine. Lancet, 341, 454–457. Received October 16, 2000; revised February 19, 2001; accepted February 28, 2001