Compounds of nickel(II) and cadmium(II) are carcinogenic to humans and to experimental animals. One frequently discussed mechanism involved in tumor formation is an increase in reactive oxygen species by both metals with the subsequent generation of oxidative DNA damage. In the present study we used human HeLa cells to investigate the potential of nickel(II) and cadmium(II) to induce DNA lesions typical for oxygen free radicals in intact cells and the effect on their repair. As indicators of oxidative DNA damage, we determined the frequencies of DNA strand breaks and of lesions recognized by the bacterial formamidopyrimidine-DNA glycosylase (Fpg protein), including 7,8-dihydro-8-oxoguanine (8-hydroxyguanine), a pre-mutagenic DNA base modification. Nickel(II) caused a slight increase in DNA strand breaks at 250 microM and higher, while the frequency of Fpg-sensitive sites was enhanced only at the cytotoxic concentration of 750 microM. The repair of oxidative DNA lesions induced by visible light was reduced at 50 microM and at 100 microM nickel(II) for Fpg-sensitive sites and DNA strand breaks, respectively; the removal of both types of lesions was blocked nearly completely at 250 microM nickel(II). In the case of cadmium(II), DNA strand breaks occurred at 10 microM and no Fpg-sensitive sites were detected. However, the repair of Fpg-sensitive DNA lesions induced by visible light was reduced at 0.5 microM cadmium(II) and higher, while the closure of DNA strand breaks was not affected. Since oxidative DNA damage is continuously induced during aerobic metabolism, an impaired repair of these lesions might well explain the carcinogenic action of nickel(II) and cadmium(II).

No abstract

The lysosomal enzyme MPO is found in neutrophils and monocytes. Exposure to cigarette smoke stimulates the recruitment of neutrophils into human lung tissue, resulting in local release of MPO. For its microbicidal activity, MPO produces hypochlorous acid, a strong oxidant that can attack nucleic acids, proteins and unsaturated lipids. 1,2 Through the concomitant release of reactive oxygen species, MPO can also participate in the activation of carcinogens in tobacco smoke, such as BP to BPDE, its ultimate carcinogen. 3 A frequently occurring polymorphism in Caucasians in the promoter region of the MPO gene is a -463 G3 A transition, which is located in the consensus binding site of the SP1 transcription factor. 4 The MPO G wild-type allele confers about 25 times higher transcriptional activation compared to the -463 A variant in vitro, 5 and the former has been associated with increased MPO mRNA and protein levels in myeloid leukemia cells. 6 In our previous study, we showed that the in vivo formation of BPDE-DNA adducts in the skin of patients treated with a similar dose of coal tar was significantly affected by MPO genotypes. Individuals with the MPO -463 (G/AϩA/A) variant had 4-fold lower BPDE-DNA adduct levels compared to those with the wild-type MPO. 7 In contrast, Brockstedt et al. 8 reported a significantly higher level of DNA adducts in subjects carrying at least 1 A-463 allele.In 6 independent studies on Caucasians from different geographic regions, 9 -14 the presumed decreased MPO activity in the variant A allele carriers was associated with a lower lung cancer risk. However, the results were not entirely consistent because it is not clear whether the effect was due to the A allele or the (AA) genotype alone. Some studies reported protection only for the (AA) genotype, 9,14 the (GA) genotype 10,11 or the 2 combined (GAϩAA). 10,11,13 In 2 further studies, 15,16 no evidence for an overall association between lung cancer risk and MPO genotype was found.SCC and SCLC are etiologically strongly associated with cigarette smoking, whereas adenocarcinoma, the most common tumor type seen in women and nonsmokers, is less strongly associated with tobacco smoking. 17 Therefore, our aims were (i) to resolve the observed discrepancies in study results by investigating whether the MPO genotype differentially affects the risk of developing adenocarcinomas, SCC and SCLC in Caucasian smokers and (ii) to verify whether the relation of MPO genotype frequency and lung cancer risk is affected by smoking status and age, as reported in earlier studies. To achieve much higher power, our study includes nearly twice as many cases than the largest previously published studies. MATERIAL AND METHODS Study subjects and sample collectionCaucasian lung cancer patients (n ϭ 625) and hospital controls (n ϭ 340) were selected from a larger case-control study. Analyses of overlapping smaller subgroups from this study have previously been reported, investigating polymorphisms in N-acetyltransferases, 18 glutathione-S-transferases 19 and huma...

CYP3A isozymes are involved in tobacco carcinogen- and steroid-metabolism, and are expressed in human lung tissue showing interindividual variation in expression and activity. The CYP3A4*1B allele has been associated with a two-fold higher promoter activity and with high-grade prostate cancers. The very frequent intron 3 polymorphism in the CYP3A5 gene (CYP3A5*3) results in decreased CYP3A5 protein levels. A case-control study was conducted in 801 Caucasian lung cancer patients that included 330 adenocarcinomas, 260 squamous cell carcinomas, 171 small cell lung cancers (SCLC) and 432 Caucasian hospital-based controls. CYP3A-genotyping was performed by capillary polymerase chain reaction followed by fluorescence-based melting curve analysis. A significantly increased SCLC risk for CYP3A4*1B allele carriers [odds ratio (OR) 2.25, 95% confidence interval (CI) 1.11-4.55, P=0.02] was found. After dividing cases and controls by gender, an increased lung cancer risk for CYP3A4*1B carriers (OR 3.04, 95% CI 0.94-9.90, P=0.06) for women but not for men (OR 1.00, 95% CI 0.56-1.81) was revealed. Heavier smoking men (> or =20 pack-years) with the CYP3A4*1B allele had a significant OR for lung cancer of 3.42 (95% CI 1.65-7.14, P=0.001) compared to *1A/*1A carriers with lower tobacco exposure (<20 pack-years). For women, the respective OR was 8.00 (95% CI 2.12-30.30, P=0.005). Genotype frequencies were generally in Hardy-Weinberg equilibrium, except for CYP3A5 where a greater than expected number of CYP3A5*1 homozygotes was observed among cases (P=0.006). In addition, we observed linkage disequilibrium of CYP3A4 and CYP3A5 (P<0.00001), but a non-significantly increased lung cancer risk was only found for homozygous CYP3A5*1 allele carriers (OR 5.24, 95% CI 0.85-102.28, P=0.14) but not for heterozygotes. To confirm our observation that the CYP3A4*1B allele increases SCLC risk and modifies the smoking-related lung cancer risk in a gender-specific manner, further studies, including CYP3A haplotype analysis, will be necessary.