Association between carcinogen-DNA adducts in white blood cells and lung cancer risk in the physicians health study.

In this matched case-control study nested within the prospective Physicians' Health Study, we evaluated whether DNA damage in blood samples collected at enrollment significantly predicted risk, consistent with our hypothesis that cases have greater biological susceptibility to polycyclic aromatic hydrocarbons and other aromatic tobacco carcinogens. The subjects were 89 cases of primary lung cancer and 173 controls, all males, matched on smoking, age, and duration of follow-up. Aromatic-DNA adducts were measured in WBCs by the nuclease P1-enhanced (32)P-postlabeling method that primarily detects smoking-related adducts. Among current smokers, but not former or nonsmokers, there was a significant increase in mean adduct levels of cases compared with controls (11.04 versus 5.63; P = 0.03). "Healthy" current smokers who had elevated levels of aromatic DNA adducts in WBCs were approximately three times more likely to be diagnosed with lung cancer 1-13 years later than current smokers with lower adduct concentrations (odds ratio, 2.98; 95% confidence interval, 1.05-8.42; P = 0.04). We were not able to discern case-control differences in former smokers and nonsmokers. The findings are of interest because they suggest that individuals who become cases have greater biological susceptibility to tobacco carcinogens, a biological difference, which manifests most clearly while exposure is ongoing.

DNA adducts in cervical tissue of smokers and non-smokers.

Cervical biopsy samples were taken from 40 women, aged between 31 and 72, undergoing hysterectomies. Twenty-two of the women were smokers, four were ex-smokers and 14 were non-smokers. DNA was isolated and analysed using 32P-postlabelling, after butanol extraction or nuclease P1 digestion enhancement of the adducts. Resolution of the adducts was by thin-layer chromatography on polyethyleneimine (PEI)-cellulose. The pattern of adducts seen was similar to smoking-related adducts detected in other tissues and consisted mainly of a diagonal zone of radioactivity. With the butanol extraction enrichment method, the levels of adducts in DNA from the 22 smokers ranged from 1.65 to 6.04 adducts/10(8) nucleotides (mean = 3.70, SD = 1.36), in DNA from non-smokers from 1.16 to 3.98 (mean = 2.04, SD = 0.77) and in samples from ex-smokers from 2.57 to 3.35 (mean = 2.86, SD = 0.37). The increase in adduct levels in smokers compared with non-smokers was highly significant (Mann-Whitney test p = 0.0005, two-tailed). When analysed by the nuclease P1 digestion enhancement method, total adduct levels in samples from smokers (mean = 2.95, SD = 1.77) were not significantly different (p = 0.3, two-tailed) from levels in non-smokers (mean = 2.34, SD = 0.96). However, the level of a minor discrete adduct spot was significantly lower (p = 0.02, two-tailed) in smokers (mean = 0.19, SD = 0.36) than in non-smokers (mean = 0.39, SD = 0.41). The results indicate that some of the DNA adducts detected in cervical epithelium correlate with tobacco smoking and support the hypothesis that smoking-related cervical cancer results from exposure to genotoxic components of cigarette smoke that become activated to DNA-binding products in this tissue.

Influence of dietary fat on DNA binding by 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) in the mouse liver.

Female BALB/c mice were fed either a low (1%)-fat or one of three high-fat diets (containing an additional 25% (w/w) beef fat, hydrogenated vegetable oil or non-hydrogenated vegetable oil) for 4 wk. They were then orally treated with 10 mg 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx)/kg body weight and killed 6 hr later. Consumption of the hydrogenated vegetable oil was accompanied by increased DNA adduct formation in mice. The abilities of hepatic S-9 preparations from mice fed the various diets to convert MeIQx to an active bacterial mutagen was assessed using Salmonella typhimurium TA98. Preparations from mice fed the high-fat diets exhibited significantly greater capacity to activate MeIQx than did those from low-fat-fed mice. The greatest increases were seen with S-9 from animals fed either beef fat or hydrogenated vegetable oil.

Evidence for the involvement of a bis-diol-epoxide in the metabolic activation of dibenz[a,h]anthracene to DNA-binding species in mouse skin.

Dibenz[a,h]anthracene (DB[a,h]A) and its microsomal metabolites, trans-3,4-dihydro-3,4-dihydroxydibenz[a,h]anthracene (DBA-3,4-diol), trans,trans-3,4:8,9-tetrahydro-3,4:8,9-tetrahydroxydibenz[a,h]anth racene, trans,trans-3,4:10,11-tetrahydro-3,4:10,11-tetrahydroxydibenz[a,h] - anthracene (DBA-3,4,10,11-bis-diol) and trans,trans-3,4:12,13-tetrahydro-3,4:12,13- tetrahydroxydibenz[a,h]anthracene were each applied topically to mouse skin and the epidermal DNA isolated 24 h later. 32P-postlabeling analysis of each of the DNA samples was performed. DNA from mice treated with DB[a,h]A produced an adduct map on TLC consisting of one major and three minor adduct spots. A similar pattern of spots was produced by DBA-3,4-diol. No detectable DNA adducts were produced by trans,trans-3,4:12,13-tetrahydro-3,4:12,13-tetrahydroxy- dibenz[a,h]anthracene, although a single, minor adduct spot was produced by trans,trans-3,4:8,9-tetrahydro-3,4:8,9-tetrahydroxydibenz[a,h]- anthracene. However, DBA-3,4,10,11-bis-diol was found to produce a major single adduct that comigrated on thin layer chromatography with the major adduct produced by both DB[a,h]A and DBA-3,4-diol. In addition, this adduct was present at a level 10 times higher than the corresponding adduct produced by treatment with the parent hydrocarbon. Coelution of the major adducts formed from DB[a,h]A and DBA-3,4-diol with that formed from DBA-3,4,10,11-bis-diol was also demonstrated on reverse-phase high performance liquid chromatography. Thus, we propose that, in mouse skin, the major pathway of DB[a,h]A activation to DNA binding products is via a 3,4-diol to the 3,4,10,11-bis-diol and ultimately to a bis-diol-epoxide (potentially the 3,4,10,11-bis-dihydrodiol-1,2-oxide).

Detection and characterization by 32P-postlabelling of DNA adducts induced by a Fenton-type oxygen radical-generating system.

Reactive oxygen species can give rise to numerous modifications of DNA. We have investigated the formation of such modifications using the nuclease P1 digestion method of the 32P-postlabelling procedure for the detection of DNA damage. Analysis of DNA that had been treated with a Fenton-type system of copper (or iron) ions and H2O2 resulted in the detection of up to ten discrete 32P-labelled spots, displaying chromatographic characteristics similar to aromatic adducts, on PEI-cellulose TLC. Maximum total levels equivalent to 28 adducts/10(8) nucleotides were achieved after 15 min of treatment with Cu2+/H2O2. The formation of adducts was 1.5 times greater if single-stranded rather than double-stranded DNA was employed, suggesting an intrastrand effect. Experiments with 3'-deoxyribonucleotides demonstrated that the adducts detected did not represent base modifications such as 8-hydroxydeoxyguanosine or thymidine glycols. However, treatment of specific dinucleotides (dApdG and dApdA) was found to produce two major adducts that were chromatographically identical by TLC and HPLC to the two major adducts formed in DNA. It is proposed that these species with aromatic adduct-like characteristics are the result of the intrastrand linking of specific adjacent bases in DNA.

The metabolic activation of dibenz[a,h]anthracene in mouse skin examined by 32P-postlabelling: minor contribution of the 3,4-diol 1,2-oxides to DNA binding.

Dibenz[a,h]anthracene (DB[a,h]A) and the related 3,4-diol and anti- and syn-3,4-diol 1,2-oxides were applied to the shaved dorsal skin of groups of four C57Bl/CB1 mice. Twenty-four hours later the mice were killed, DNA isolated from the treated skin, hydrolysed and examined for the presence of aromatic adducts using the nuclease P1 modification of the 32P-postlabelling technique. Autoradiography of the maps obtained by chromatography on polyethyleneimine-cellulose plates showed that six DNA adduct spots that were derived from DB[a,h]A were also present in the DNA of skin treated with the DBA 3,4-diol and that, whilst four of these adduct spots were also seen in maps prepared from the DNA of skin treated with the anti-3,4-diol-1,2-oxide, they were not present in DNA from skin to which the syn-isomer had been applied. The identity of these adduct spots was confirmed by their coincidence when mixtures of different DNA hydrolysates were chromatographed together. Quantitatively, the highest levels of mouse skin modification were obtained with the diol-epoxides and the lowest with DB[a,h]A. The results suggest that most of the DNA adducts formed in DB[a,h]A-treated mouse skin arise through metabolism of the hydrocarbon to the related 3,4-diol and that some may be formed following the conversion of this diol to the bay-region anti-3,4-diol-1,2-oxide.

Tissue distribution of DNA adducts in CDF1 mice fed 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) and 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ).

Male and female CDF1 mice were administered a single oral dose of 3 mumol of the food mutagens 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) or 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ) and killed 24 h later. DNA was isolated from the livers, lungs, kidneys, colon and forestomach and analysed by 32P-postlabelling for the presence of IQ and MeIQ adducts. Several adduct-enrichment procedures were investigated, including ATP-deficient labelling conditions, butanol extraction and nuclease P1 digestion, and only the ATP-deficient procedure was found to produce the same adduct pattern on polyethyleneimine--cellulose TLC as the standard procedure. Up to nine adduct spots were detected in liver DNA from IQ-treated mice, two of which were not detected in other tissues. The levels of binding in both male and female mice were in the order liver greater than kidney greater than colon greater than forestomach greater than lung. Analysis of DNA from MeIQ-treated mice revealed the presence of up to seven adducts, one of which was detected in liver but not in other tissues. The relative order of DNA binding was kidney greater than liver greater than or equal to colon greater than forestomach greater than lung. As dietary feeding of IQ induces liver, lung and forestomach tumours, and MeIQ induces liver and forestomach tumours in this mouse strain, these binding levels do not correlate with the susceptibility of the organs to carcinogenesis induced by these compounds; the results may indicate the importance of additional factors in determining organ specificity of carcinogenicity.