Oxidative DNA damage measured in human lymphocytes: large differences between sexes and between countries, and correlations with heart disease mortality rates

^a Rowett Research Institute, Bucksburn, Aberdeen AB21 9SB, United Kingdom
^b Clínica Puerta de Hierro, Servicio de Nutrición, 28035-Madrid, Spain
^c Institute of Food Research, Norwich Research Park, Colney, Norwich NR4 7UA, United Kingdom

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS REFERENCES

 
The `antioxidant hypothesis' proposes that vitamin C, vitamin E, carotenoids, and other antioxidants occurring in fruit and vegetables afford protection against heart disease and cancer by preventing oxidative damage to lipids and to DNA, respectively. To test elements of this hypothesis, we have measured blood levels of dietary antioxidants, and 8-oxodeoxyguanosine (8-oxo-dG) concentrations in lymphocyte DNA, in healthy men and women from five European countries: France, Ireland, The Netherlands, Spain, and the U.K. Volunteers, aged 25–45, all nonsmokers, gave blood samples before and after a 12-wk carotenoid supplementation regime. Vitamin C was measured in plasma and vitamin E and carotenoids were measured in serum by high-performance liquid chromatography (HPLC). 8-oxo-dG was assayed by HPLC (with coulometric detection) in DNA isolated from lymphocytes from the same blood samples. Mean values were calculated for groups of volunteers at each sampling time according to country, sex, and supplementation (between 9 and 24 individual samples contributing to each mean). We found that 8-oxo-dG levels in lymphocyte DNA vary significantly according to sex and country. A low mean 8-oxo-dG concentration is seen in DNA of women from all five countries, and of men from France and Spain. 8-oxo-dG is significantly higher (up to about threefold) in lymphocyte DNA from men in Ireland and the U.K. Oxidative DNA damage is not significantly affected by carotenoid supplementation; nor is there any association with mean baseline levels of antioxidants, which are generally similar in the five countries. The five countries sampled lie on an axis from northern to southern Europe with a steep gradient in terms of premature heart disease. There is a strong association between premature coronary heart disease mortality in men and the mean levels of 8-oxo-dG for the five countries (r = 0.95, P < 0.01). Women have low coronary heart disease mortality rates, which do not correlate with 8-oxo-dG. In terms of cancer deaths, only colorectal cancer in men shows a significant positive correlation (r = 0.91, P < 0.05), and stomach cancer in women is negatively correlated with DNA oxidation (r = -0.92, P = 0.01).—Collins, A. R., Gedik, C. M., Olmedilla, B., Southon, S., Bellizzi, M. Oxidative DNA damage measured in human lymphocytes: large differences between sexes and between countries, and correlations with heart disease mortality rates. FASEB J. 12, 1397–1400 (1998)

Key Words: molecular epidemiology • oxidative DNA damage • coronary heart disease • cancer

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS REFERENCES

 
PROTECTION BY FRUIT and vegetables against coronary heart disease (CHD)² and cancer has been documented in a large number of epidemiological studies (1, 2). Free radical damage to lipids and DNA has been invoked as a cause of both these diseases, and it has been proposed (3) that the factor in fruits and vegetables responsible for the protective effect may be the natural antioxidants such as vitamin C, vitamin E, carotenoids, and flavonoids. The free radicals may arise exogenously (e.g., in tobacco smoke) as a consequence of inflammation or as a by-product of normal respiration. Oxidative damage has been detected in the form of oxidized bases in the DNA from normal human subjects. We have measured 8-oxo-deoxyguanosine (8-oxo-dG) in samples of lymphocyte DNA from men and women in different European countries (France, Ireland, The Netherlands, Spain, and the U.K.), and find large differences that are not readily explained in terms of variations in concentrations of dietary antioxidants in the blood.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS REFERENCES

 
Human subjects
Volunteers in this study were participating in a blinded vitamin E/carotenoid supplementation trial carried out in five centers (Grenoble, France; Cork, Ireland; Madrid, Spain; Zeist, The Netherlands; and Coleraine, U.K.). Apparently healthy, nonsmoking male and female volunteers aged 25–45 were recruited, excluding anyone consuming a diet not representative of the region, consuming supplements, or taking prescribed medication. The vounteers were checked for normal biochemical blood profile. Mean ages were not significantly different between sexes in any center, and ranged between 31 and 36 between centers. Volunteers were randomly sorted into four groups (of each sex) at each center and received supplements over a period of 20 wk, as shown in Fig. 1. Vitamin E (-tocopherol) was taken at 100 mg per day and carotenoids at 15 mg per day. `Palm oil carotenes' is a mixture of - and ß-carotene (30 and 66.5%, respectively) and `lycopene' contains about 10% of ß-carotene. The study was approved by all relevant local ethics committees.

View larger version (12K): [in this window] [in a new window]   Figure 1. Carotenoid intervention trial protocol.

Sample preparation
Fasting venous blood samples were taken at weeks 0 and 16; lymphocytes were isolated by density gradient sedimentation (4) and slowly frozen in 90% fetal calf serum, 10% dimethylsulfoxide, to -80°C before storage in liquid nitrogen. Because the amount of blood available for this part of the investigation was insufficient to measure individual levels of 8-oxo-dG, lymphocytes from three or four subjects of the same sex from the same country and within the same supplementation group were combined; DNA was isolated by a high salt protein precipitation method (5). The DNA was hydrolyzed with deoxyribonuclease I, alkaline phosphatase, and phosphodiesterases I and II (6), and analyzed by high-performance liquid chromatography (HPLC) on a 15 x 0.46 cm Apex C18 3 µm column (Jones, Hengoed, Wales) with a 2 x 0.4 cm guard column containing Perisorb RP18 (Anachem, Luton, England). The mobile phase was 50 mM potassium phosphate, pH 5.5, with 7.5% methanol and the flow rate was 0.8 ml/min. Deoxyguanosine was measured using a Gilson Holochrome UV detector set at 254 nm. Detection of 8-oxo-dG required an electrochemical detector (Coulochem 5100H) with a 5021 conditioning cell and 5010 analytical cell. Three determinations were made on each hydrolyzed pooled sample, one of them including a spike of standard 8-oxo-dG, and the average of the two unspiked replicates was used in the calculation of mean values shown in the tables.

Antioxidant determinations
HPLC was used to measure vitamin C in acidified plasma (7) and vitamin E and carotenoids on nonacidified serum (8) from the same blood samples used for lymphocyte isolation.

Statistical analysis
Each set of data, for weeks 0 and 16, was subjected to analysis of variance (ANOVA) with factors `country', `sex', and (for week 16) `supplement'. Student's t test (two-tailed, assuming unequal variance) was used to compare means. Premature CHD mortality rates and cancer mortality rates (used to calculate correlations with 8-oxo-dG levels in Table 4) were obtained from the WHO Database on Health Indicators, 1991.

RESULTS
Lymphocytes were available from France, Ireland, and Spain for analysis of DNA damage at week 0. ANOVA indicated a significant effect of sex (P<0.05) and of country (P<0.01). Similar levels of 8-oxo-dG occur in DNA from men and women from France and Spain and from women in Ireland, but there is significantly more in the samples from Irish men compared with women ( Table 1). At week 16, after 12 wk of supplementation with carotenoid or placebo, we analyzed samples from France, Ireland, The Netherlands, and the U.K. ANOVA showed no effect of supplementation (P=0.74), but a significant effect of sex (P=0.01) with an interaction with country (P=0.06). Therefore the data were analyzed without reference to supplementation group ( Table 2). Concentrations of 8-oxo-dG in men from Ireland and the U.K. are significantly elevated compared with those from France. No significant differences are seen between mean values for women in the four countries. In samples from Ireland, there is a particularly large difference in DNA damage between the sexes.

Concentrations of antioxidants are shown in Table 3. Overall, vitamin C concentrations are higher in women than in men (P<0.05). Concentrations of `total carotenoids' (i.e., lutein, zeaxanthin, lycopene, ß-cryptoxanthin, - and ß-carotene) in serum from week 0 do not vary significantly between sexes or between countries. At week 16, mean serum carotenoid concentrations in France and Ireland (the two countries with pre- and postsupplementation samples) have increased as a result of 12 wk of supplementation, and are higher in women than in men (P<0.01). Serum vitamin E (not shown) does not vary significantly between countries or sexes at either week 0 or week 16.

DISCUSSION
This study represents the first international comparison of 8-oxo-dG levels as well as the first attempt to compare the sexes. As discussed elsewhere (9), estimates of the level of 8-oxo-dG in the DNA of human white blood cells have given a wide range of values, depending on the assay method as well as the precautions taken to avoid spurious oxidation of the sample DNA. The procedures used by us give low values for background levels of 8-oxo-dG (9). Differences in 8-oxo-dG between countries could conceivably result from subtle variations in the method of isolating cells (but not from different methods of isolating or hydrolyzing DNA, as this was done on all samples in one center). We believe this explanation to be unlikely, as a standard protocol was used by all participants in the study. Differences between men and women from the same center cannot be so explained, as the coded samples were treated in exactly the same way throughout.

It is evident that the highest levels of 8-oxo-dG are seen in the lymphocytes of men from northern Europe. Results for women in all five countries are indistinguishable. There is no obvious correspondence between levels of 8-oxo-dG and dietary antioxidant concentrations in blood. In the crucial case of the most extreme variation—between men and women in Ireland at week 16—vitamin C, vitamin E, and total carotenoid concentrations do not differ. Intrinsic antioxidant defences (glutathione, superoxide dismutase, and glutathione peroxidase) were also measured, and results will be published elsewhere; there were no consistent differences between the sexes and no association with levels of DNA damage.

On the basis of the supposed link between oxidative stress and certain human diseases, we looked for variations in patterns of incidence of various diseases that might reflect the variations in oxidative DNA damage. Figure 2 illustrates the striking association between 8-oxo-dG concentrations and premature deaths from CHD (i.e., deaths in the age range of 0–64 years) in the five countries. The correlation coefficient (r) overall is 0.90. Taken separately, data for men have a correlation coefficient of 0.95, whereas mortality rates for women show no association with 8-oxo-dG levels. This is not surprising in view of the limited variation seen in the 8-oxo-dG concentrations for women. Note, too, that the measurements of DNA damage were made in premenopausal women (with very low CHD risk), whereas mortality figures reflect mainly postmenopausal CHD. On the basis of these data, we cannot say anything about the possibility of a link between CHD and oxidative stress in women; however, the ranking of countries in terms of CHD risk is the same for women as for men, suggesting that qualitatively similar factors operate in both sexes.

View larger version (10K): [in this window] [in a new window]   Figure 2. Relationship between mean 8-oxo-dG in lymphocyte DNA and early deaths from CHD in men () and women () in the five countries (each country is identified by its first letter).

Correlations with cancer as well as CHD death rates are shown in Table 4. Overall cancer mortality is not associated with oxidative damage to DNA. Colorectal cancer in men is significantly positively correlated with 8-oxo-dG levels. However, the only other significant correlation is a negative one between stomach cancer and 8-oxo-dG in women. The hint of a difference in etiology of these diseases between men and women is intriguing.

High mean levels of oxidative DNA damage in lymphocytes from men in northern Europe do not reflect particularly low mean levels of plasma vitamin C, serum vitamin E, or carotenoids at the national level, as would have been predicted by the antioxidant hypothesis. Yet, as will be reported elsewhere, when individual values of DNA damage (measured by the more sensitive `comet assay') were assessed, we found significant negative correlations between DNA base oxidation and concentrations of carotenoids.

8-oxo-dG serves as a useful biomarker of oxidative stress. It may even have predictive value for CHD, but extensive studies of 8-oxo-dG concentrations in individuals (rather than mean values for population groups as estimated here) will be needed to determine the extent of inter- and intra-individual variability and to look for correlations with other risk factors. 8-oxo-dG appears less promising as an indicator of cancer risk.

	ACKNOWLEDGMENTS

 
Henk van den Berg, Bernice Corridan, Isabelle Hininger, and David Thurnham supervised collection of samples at the centers in Zeist, Cork, Grenoble, and Coleraine, respectively. We are grateful for the expert technical assistance of Sharon Wood, Sue Tsang, Inmaculada Blanco, Fernando Granado, and Enrique Gil-Martinez, for statistical advice from Graham Horgan of BioSS, and for valuable comments on the manuscript from Susan Duthie. Financial support was provided by the EU (AIR2-CT93–0888), the Scottish Office Agriculture, Environment and Fisheries Department, the Biotechnology and Biological Sciences Research Council (U.K.), and the Comisíon Interministerial de Ciencia y Tecnología (CICYT, Spain, SAF96–1552,CE).

	FOOTNOTES

 
¹ Correspondence: DNA Instability Group, Rowett Research Institute, Greenburn Road, Bucksburn, Aberdeen AB21 9SB, U.K.

² Abbreviations: CHD, coronary heart disease; ANOVA, analysis of variance; 8-oxo-dG, 8-oxo-deoxyguanosine; HPLC, high-performance liquid chromatography.

Received for publication January 15, 1998. Accepted for publication May 27, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS REFERENCES

 

1. Renaud, S., and Delorgeril, M. (1994) Nutrition, atherosclerosis and coronary heart disease. Reprod. Nutr. Dev. 34, 599–607
2. Block, G., Patterson, B., and Subar, A. (1992) Fruit, vegetables and cancer prevention: a review of the epidemiological evidence. Nutr. Cancer 18, 1–29[Medline]
3. Ames, B. N. (1983) Dietary carcinogens and anticarcinogens. Science 221, 1256–1264[Abstract/Free Full Text]
4. Duthie, S. J., Ross,M., and Collins, A. R. (1995) The influence of smoking and diet on the hypoxanthine phosphoribosyltransferase (hprt) mutant frequency in circulating T lymphocytes from a normal human population. Mutat. Res. 331, 55–64[Medline]
5. Collins, A. R., Duinská, M., Gedik, C. M., and tetina, R. (1996) Oxidative damage to DNA: do we have a reliable biomarker? Environ. Health Perspect. Suppl. 3 104 465–469
6. Richter, C., Park, J.-W., and Ames, B. N. (1988) Normal oxidative damage to mitochondrial and nuclear DNA is extensive. Proc. Natl Acad. Sci. USA 85, 6465–6467[Abstract/Free Full Text]
7. Finglas, P. M., Bailey, A. L., Walker, A., Wright, A. J. A., and Southon, S. (1993) Vitamin C intake and plasma ascorbic acid concentration in adolescents. Br. J. Nutr. 69, 563–576[Medline]
8. Olmedilla, B., Granado, F., Gil-Martinez, E., Blanco, I., and Rojas-Hidalgo, E. (1997) Reference values for retinol, tocopherol and main carotenoids in serum of control and insulin-dependent diabetic Spanish subjects. Clin. Chem. 43, 1066–1071[Abstract/Free Full Text]
9. Collins, A., Cadet, J., Epe, B., and Gedik, C. (1997) Problems in the measurement of 8-oxoguanine in human DNA. Report of a workshop, DNA Oxidation, held in Aberdeen, U.K. Carcinogenesis 18 1833–1836[Abstract/Free Full Text]