DNA adduct burden and tobacco carcinogenesis - Nature

ª

Oncogene (2002) 21, 7376 – 7391 2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00 www.nature.com/onc

DNA adduct burden and tobacco carcinogenesis John K Wiencke*,1 1

Laboratory for Molecular Epidemiology, Department of Epidemiology and Biostatistics, University of California San Francisco, San Francisco, California, CA 94143-0560, USA

DNA adducts associated with tobacco smoking could provide a marker of biologically effective dose of tobacco carcinogens and improve individual cancer risk prediction. A significant number of clinical and epidemiologic studies have reported associations of increased DNA adduct levels with the occurrence of the prevalent tobacco related cancers including cancer of the lung, head and neck, and bladder. The inducibility of DNA adducts following in vitro treatments using blood lymphocytes also appears to be a risk factor in the development of lung and head and neck cancer. Corroborative evidence pointing to the importance of DNA adducts in tobacco carcinogenesis include numerous studies showing associations of tobacco smoke exposure with the induction of DNA adducts in humans in vivo. Further effort is necessary, however, to more fully characterize the dose – response relationship between smoking and DNA adducts in exposed target and surrogate tissues. The relationship between gene polymorphisms thought to modify tobacco-related cancer risk and DNA adduct levels is complex. Results of some DNA adduct studies (both in vitro and in vivo) appear inconsistent with the epidemiologic findings. This is evident for polymorphisms involving both carcinogen metabolism (e.g. GSTP1) and DNA repair (e.g. XRCC1). Molecular studies of human tumors suggest associations of p53 mutation with DNA adducts and have revealed correlations of DNA adduct levels with somatic alterations (e.g. 3p21 LOH) that are thought to occur at the very earliest stages of tobacco carcinogenesis. More research is needed to assess the relationship between endogenous sources of DNA adducts and tobacco smoke exposure and the relative oncogenic effects of chemically stable versus unstable DNA adducts. Many potentially fruitful new avenues of cancer research are emerging that integrate DNA adduct analyses with assessments of smoking, genetics, diet and ambient air quality. These investigations aim to understand the multifactorial nature of interindividual variability in response to tobacco carcinogens. As these trends continue a variety of innovative study designs and approaches will become important in human populations. Oncogene (2002) 21, 7376 – 7391. doi:10.1038/sj.onc. 1205799 Keywords: DNA adducts; tobacco; risk assessment; molecular epidemiology

*Correspondence: JK Wiencke; E-mail: [email protected]

Background Introduction DNA adducts are physical complexes formed between reactive chemical species and sites within the DNA molecule. DNA adducts have been proposed as potential markers of ‘biologically effective dose’ from exposure to tobacco carcinogens that may help to provide an integrated measure of carcinogen exposure relevant to individual cancer risk assessment. The emphasis of this review is on human in vivo studies except where model in vitro or animal studies are relevant to a specific issue. This review emphasizes literature appearing since 1998; below are cited reviews of earlier literature. The field of investigation that addresses DNA modifications in human cancer is expansive. This review is restricted to epidemiologic and clinical studies that address five areas bearing on the etiologic role of DNA adducts in tobacco carcinogenesis: (1) The evidence that genetic polymorphisms implicated in tobacco related cancers modify DNA adduct concentrations in human tissues; (2) The evidence that DNA adducts are risk factors in the development of tobacco related cancers; (3) The association of DNA adducts with molecular alterations involving oncogenes and tumor suppressor genes in human tumors; (4) The potential importance of endogenous and chemically unstable DNA adducts in tobacco carcinogenesis; and (5) Methodologic issues that affect the validity of investigations into the role of DNA adducts in tobacco carcinogenesis. Recent reviews Several recent reviews have addressed the use of DNA adducts in risk assessment and toxicology studies (van Delft et al., 1998a; Vainio, 1998; Timbrell, 1998; Kriek et al., 1998; Garner, 1998; Carmichael, 1998; Hemminki et al., 2001). Other reviews emphasize the chemistry of adduct formation and laboratory methodology (Phillips et al., 2000; Reddy, 2000; Szeliga and Dipple, 1998). Although most recent studies have employed only a few analytical approaches (i.e. 32P-postlabeling or immunologically based assays), new technologies and improved applications have also been reviewed (Turteltaub and Dingley, 1998, accelerated mass

DNA adduct burden and tobacco carcinogenesis JK Wiencke

7377

spectrometry); (Marzilli et al., 2001, capillary-electrophoresis-mass spectrometry); (Shinozaki et al., 1998, flow cytometry); (Ni et al., 1998, electrospray tandem mass spectrometry); (Petruzzelli et al., 1998, serum BPDE-DNA antibodies by ELISA); (Bucci et al., 1998, competitive ELISA); (Xing et al., 2001, capillary electrophoresis); (Tan et al., 2001, immunoelectrophoresis with laser induced fluorescence). Genetic susceptibility Previous reviews of genetic polymorphisms have assessed the potential for adducts to provide an intermediary tool for genotyping studies (Hemminki et al., 2000). The relationship of genotype to adduct formation in determining risk for lung, esophagus and oral cancer have been evaluated (Bartsch et al., 1998). An overall weight of evidence approach concluded that under conditions giving rise to detectable PAH – DNA adducts levels in tissues, most studies reviewed reported a significant role for GSTM1 deletion either alone or in combination with CYP1A1 variants on variations in adduct levels and urinary metabolites of genotoxic substances (Pavanello and Clonfero, 2000). Metabolic polymorphisms Glutathione S-transferases and cytochrome P4501A1 Although early studies suggested a role of GSTM1 deficiency in aromatic-hydrophobic DNA adduct levels (Ryberg et al., 1994; Grinberg-Funes et al., 1994; Shields et al., 1993; Kato et al., 1995) these studies were criticized (Cuzick, 1995) for not defining cigarette smoking exposure adequately and for not having sufficient size to adjust for potential statistical artifacts. Continued work on the GSTs has appeared (Table 1) and more recent studies have included polymorphism of the GSTT1 and GSTP1 genes. The GSTP1 isoforms could be particularly relevant in lung carcinogenesis as GSTP1 is highly expressed in lung and shows high activity in detoxifying diol epoxides of PAHs. Human GSTP1 expression in cell culture has been shown to be very effective in preventing the formation of BPDE – DNA adducts produced in vitro (Fields et al., 1998). Two common polymorphisms in GSTP1 give rise to four genotypic variants in human populations. Although allelic variants (i.e. A313G; Ile105Val) show reduced enzyme activity toward the model substrate 1-chloro-2,4-dinitrobenzene (Watson et al., 1998) when transfected into human HepG2 cells these variants were shown to provide greater protection against the formation of (+)-anti-BPDE – DNA adducts (Hu et al., 1999) compared with the wildtype GSTP1 isoform. The wildtype GSTP1 contains isoleucine at position 105. Catalytic properties of both recombinant and native variants of GSTP1 enzymes were compared with wildtype forms and the variants were shown to display greater catalytic efficiency towards carcinogenic PAHs (Coles et al., 2000a). Lung, bladder and esophageal cancer risk, however, have

been associated with GSTP1 genotypes presumed to result in greater not less PAH detoxification (Ryberg et al., 1997; Kihara and Noda, 1999; Harries et al., 1997; van Lieshout et al., 1999). DNA adduct studies thus far have not fully resolved this apparent anomaly. In studies of newborns, PAH – DNA adducts were marginally higher among subjects with the GSTP1 Ile/Val and Ile/Ile genotypes compared with the GSTP1 Val/Val genotype (Whyatt et al., 2001). Similarly, the lowest levels of DNA adducts were observed among non-smoking women carrying the GSTP1 Val/Val genotype (Grzybowska et al., 2000) compared to others, but in occupationally exposed populations no effect of the GSTP1 polymorphism by itself has been reported on variations in PAH – DNA adduct levels (Zhang et al., 2000; Viezzer et al., 1999; Grzybowska et al., 2000; Schoket et al., 2001). The GSTP1 polymorphism was also not found to be not predictive of PAH – DNA adduct levels in bronchial tissues of lung cancer patients (Ozawa et al., 1999). There have been suggestions of interactions between GSTP1 and other polymorphisms, but again the picture is far from clear. Whereas PAH – DNA adducts in occupationally exposed workers were elevated in GSTM1 null subjects and in persons carrying a combined GSTM1 null and GSTP1 variants (Schoket et al., 2001), another study of cigarette smokers found evidence of the GSTM1 non-null genotype interacting with GSTP1 variants and being associated with elevated DNA adducts (Butkiewicz et al., 2000). A suggested interaction of the MspI variant of CYP1A1 with the GSTP1 Val/Val genotype on adduct levels in newborns was reported in one study (Whyatt et al., 2001). It has been proposed (Coles et al., 2000b) that GSTP1 like GSTM1 is coordinately regulated with the GSTM3 locus. However, GSTM3 genotypes were not found to affect the relationship of GSTP1 with lung cancer in a recent study of 389 lung cancer cases and 353 controls in Germany (Risch et al., 2001). A re-examination of GSTP1 genotype – phenotype relationships (Sundberg et al., 2002) has suggested that the GSTP1 enzyme containing valine at position 105 may actually have less than, or at most equal activity in reducing, PAH – DNA adduct formation compared with the wildtype GSTP1 Ile105/Ala114 isoform. The authors caution against extrapolating catalytic information from pure enzymes to the complex situation within the intact cell. Notably, they found only about 1 – 2% of the expected rate of BPDE conjugation to be observed in vivo in cells, which the authors postulate may be a result of the reduced concentrations of lipophilic substrates available for conjugation by the soluble GSTs (Sundberg et al., 2002). Among coke oven workers the GSTM1 null genotype was associated with adducts among highly PAH exposed groups but GSTT1 null subjects, who are deficient in GSTT1 mediated GSH conjugation, demonstrated lower not higher adduct burdens (Viezzer et al., 1999). Negative studies have also appeared in which adduct levels in non-tumorous lung or blood MNCs were not modified by GSTM1 (Wiencke et al., Oncogene


7378 Table 1 Genetic polymorphisms and DNA adducts levels

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Oncogene

Endpoint/Methods

Tissue

Gene Loci

Study design/population

Bulky-DNA adducts

Bronchial tissues

GSTM1 CYP1A1

Lung cancer case series (n=124) Referent non-malignant lung conditions (n=26)

Results

Reference

Current smokers and short term Schoket et al., exsmokers had higher adduct 1998 32P-postlabeling levels than former smokers (1 yr quit); no association with genotype, cig/day BPDE-DNA adducts Non-tumorous lung GSTM1 Lung cancer case series Non detectable (low) adduct levels Rojas et al., tissue CYP1A1 (n=30), occupational PAH associated with GSTM1 non-null 1998 HPLC/fluorometry exposed coke oven workers genotypes; gene interactions WBCs (n=20) suggested Aromatic-DNA adducts Total WBC CYP1A1 Occupational case referent No association of cumulative Pan et al., 32P-postlabeling (China) exposure and adducts; among 1998 25 coke oven male workers, CYP1A1 variants adducts 25 referent male subjects correlated with 1-hydroxyprene Urinary 1-OHPY levels Bulky aromatic-DNA Total WBC and GSTM1 Occupational case referent NAT2 slow acetylators had higher Binkova, adducts lymphocytes NAT2 adduct levels no effect of GSTM1 et al., 1998 68 male PAH exposed coke Urinary cotinine associated with 32P-postlabeling oven workers adduct levels Urinary continine 56 non exposed machine Plasma vitamin levels workers PAH-DNA adducts Oral mucosa GSTM1 Exposure case referent No association of GSTM1 or Motykiewicz CYP1A1 Polish non-cmoking women CYP1A1 with adducts, adducts et al., 1998 Immunohistochemical 67 PAH-exposed 72 referent higher in PAH exposed group, subjects evidence of seasonal variation BPDE-DNA adducts (in T47D cells GSTP1 In vitro Cell Culture 70 – 90% reduction in DNA adducts Fields et al., vitro MCF-7 cells expression in T47D cells highly expressing 1998 only BPDE (30 – 600 nM) GSTP1 [3H]BPDE binding PAH-DNA adducts Placental tissue CYP1A1 PAH exposure case referent Neither CYP1A1 expression nor Whyatt et al., Msp1 smoking associated with adduct 1998 Immunochemical (Genotype) 70 exposed newborns levels, higher placental adducts in 90 referent newborns CYP1A1 variant infants Infant plasma cotinine CYP1A1 mRNA (expression) PAH-DNA adducts Maternal and GSTM1 Cross-sectional Maternal smoking associated with Whyatt et al., umbilical white blood CYP1A1 70 mothers and newborns mothers adduct levels; higher 1998 Immunochemical cells adducts among newborns with CYP1A1 variants PAH-DNA adducts Blood lymphocytes GSTM1 Cross-sectional Smokers had higher adduct levels Wang et al., GSTT1 compared with nonsmokers; 1998b 32P-postlabeling CYP1A1 158 subjects CYP1A1 variants and GSTM1 null NAT2 subjects had higher adduct levels; (plasma beta-carotene, ALDH2 interactions with plasma alpha-tocopherol) antioxidants PAH-DNA adducts Blood lymphocytes GSTM1 Cross-sectional DNA adduct levels positively Soni et al., correlated with daily cigarette 1998 32P-postlabeling 21 smokers consumption and inversely with GSTM1 status Bulky hydrophobic – Blood MNCs GSTM1 Lung cancer case series Among former smokers Wiencke et al., DNA adducts Non tumorous lung CYP1A1 polymorphisms were not 1999 tissue 143 lung associated with DNA adduct 32P-postlabeling 54 blood levels; early age of smoking initiation associated with increased adduct levels Aromatic/hydrophobic – Nontumorous Expression Case series Female cases had higher smoking Mollerup, DNA adducts Lung tissues only Lung cancer patients adjusted adduct levels; adducts 1999 CYP1A1 29 females were positively correlated with 32P-postlabeling 93 males CYP1A1 expression PAH-DNA adducts, Blood MNCs GSTM1 Occupational cross-sectional Positive interaction of high PAH Viezzer et al., GSTT1 70 male coke oven workers, exposure and GSTM1 null 1999 32P-postlabeling GSTP1 Italian steel workers (mean genotype, GSTT1 null had lower mEPHX age 41.3 yrs) adduct levels urinary 1-pyrenol PAH exposure subgroups by urinary 1-pyrenol BPDE-DNA adducts Total WBC CYP1A1 Occupational Increased adduct levels in GSTM1 Rojas et al., GSTM1 case-referent null but not GSTT1 null subjects 2000 HPLC/fluorometry GSTT1 89 coke oven Significant effects of CYP1A1 (35 smokers, 36 exsmokers, genotypes on adduct levels Continued


7379

Table 1 (Continued ) Endpoint/Methods

15

PAH-DNA adducts

Tissue

Maternal and Newborn WBCs

Immunoassay

16

PAH-DNA adducts

Total WBC

32P-postlabeling Urinary cotinine 17 Aromatic-DNA adducts 32P-postlabeling 18

PAH-DNA adducts

Total WBC

Bronchoaleveolar Macrophages

32P-postlabeling 19 20

21

PAH-DNA adducts

Blood MNCs

32P-postlabeling Hydrophobic – DNA adducts 32P-postlabeling

WBC Buffy coat

PAH-DNA adducts 32P-postlabeling (w/ butanol)

Peripheral blood Lymphocytes

Urinary 1-OHPY 22 Aromatic-DNA adducts Isolated blood MNC

Gene Loci

4-ABP DNA adducts

MNCs

GC-MS 23

23

4-ABP-DNA adducts 32P-postlabeling p53 mutation PAH-DNA adducts

Bladder biopsy

Peripheral blood Lymphocytes,

32P-postlabeling Bronchial tissue Urinary 1-OHPY 24

4-ABP-DNA adducts

Bladder cancer biopsies

GC/MS 25 Aromatic-DNA adducts 32P-postlabeling

WBC

26 Aromatic-DNA adducts 32P-postlabeling

Pancreas tissue

8-oxo-dG: HPLC-EC

Results

18 nonsmokers) 44 power plant workers (all smokers) CYP1A1 Cross Sectional Among mothers no association of GSTP1 160 Polish mothers and their adduct levels and gene newborns polymorphisms; Among newborns CYP1A1 MSPI variants had higher adducts, positive interaction with GSTP1 variants CYP1A1 Occupational Case referent Marginal association of adducts GSTM1 65 coke oven workers (China) with high versus low PAH GSTP1 exposure, significant associations NAT2 of adducts with urinary cotinine, P53 CYP1A1 and p53 among subgroups CYP1A1 Occupational Nonsignificant differences in CYP2D6 133 nonsmoking Polish combined subgroups of GSTM1 women (ages 35 – 45 years) genotypes; no clear seasonal GSTP1 associations Expression Smoking Case-referent threefold higher adduct levels in only smokers; no detectable CYP1A1 CYP1A1 31 smokers protein or mRNA CYP1B1 16 non smokers CYP3A GSTP1 Cross sectional Smokers with combine GSTP1 GSTM1 GSTM1 positive had higher adduct 170 healthy volunteers levels; seasonal variation GSTM1 Cross-sectional GSTT1 null associated with higher GSTT1 309 subjects adduct levels and suggested CYP1A1 interaction of GSTT1 and GSTM1 NAT2 GSTM1 Occupational case referent No effects of polymorphisms on GSTP1 adducts. Smoking associated with 32 Dutch coke oven workers Increased DNA adducts and 33 nonexposed subjects urinary 1-OHPY excretion GSTM1 NAT2

32P-postlabeling 23


GSTM1 GSTT1 NAT1 NAT2 GSTM1 NAT2

Case Control 170 cases 146 controls (Stockholm Sweden) Cross Sectional 67 Smokers Bladder cancer Case series

Among current smokers, NAT2 slow acetyltors show higher adducts; postive interaction with GSTM1 null No overall effect of genotypes on Adduct levels

Reference

Whyatt et al., 2000

Zhang et al., 2000

Grzybowska et al., 2000 Piipari et al., 2000

Butkiewicz et al., 2000 Palli et al., 2000 Van Delft, et al., 2001

Hou et al., 2001 Godschalk et al., 2001

4ABP-DNA adducts higher in Martone et al., smokers of air cured tabacco; no 1998 association with NAT2

CYP1A1 Occupationally DNA adducts correlated with Schocket et al., CYP1B1 Exposed aluminum plant urinary 1-OHPY in GSTM1 null 2001 CYP2C9 workers, lung cancer patients workers; interaction of CYP2C GSTM1 and P1 suggested NQO1 GSTM1 GSTP1 GSTM1 Case series No association of GSTM1 or NAT2 Martone NAT2 45 bladder cancers and adducts; adducts higher in et al., 1998 current smokers of air cured tobacco NAT2 Badder cancer NAT2 genotype associated with Peluso et al., Case (n=114) control (n=46) adducts; adducts associated with 1998; See also case status Peluso et al., 2000 GSTM1 Cross sectional Cyp1A1 var/var (n=3) showed Li et al., CYP1A1 30 pancreas cancer higher adduct levels compared 2002 CYP2E1 24 organ donors with hets and wildtype (n=61) NAT1 6 non-pancreas cancer NAT2 5 chronic pancreatitis MNSOD hOGG1 Continued Oncogene


7380

Table 1 (Continued ) Endpoint/Methods 27 Oxidative Stress related DNA adducts

Tissue

Gene Loci


Results

Reference

Pancreas tissue

GSTM1 GSTT1 NQO1

Smoking case referent

Neither GSTM1 nor NQO1 polymorphisms associated with adduct levels

Kadlubar et al., 1998

15 smokers 8-oxo-dG: HPLC-electro15 non-smokers chemical detection m1G:GC-MS edA, edC:immunoaffinity/ 32P-postlabeling 28 Polyphenol-related Isolated Blood MNCs XRCC1 Cross sectional analysis Positive gene dosage effect for Duell et al., DNA adducts ERCC2 Controls is lung cancer case – detectable DNA adducts with the 2000 control study XRCC1 399GIn allele; positive 32P-postlabeling n=61 (median age 65 yrs) interaction with age 29 Bulky DNA adducts WBC XRCC1 Cross sectional Higher adduct levels in nonMatullo et al., XRCC3 n=302 Italian subjects smoking XRCC1 399GIn 2001b 32P-postlabeling XPD EPIC Study homozygotes; no effect on smoking adducts 30 Bulky-DNA adducts WBC XRCC1 Case control Higher adduct levels in XRCC3 Matullo et al., XRCC3 124 bladder ca cases, 85 non smokers, Slow NAT2 2001a 32P-postlabeling XPD hospital controls acetlyators showed higher adduct NAT2 levels 31 Bulky-DNA adducts Peripheral blood XPD Occupational XPDLys751GIn Palli et al., Leukocytes Case-referent Associated with higher adduct 2001 32P-postlabeling levels; twofold increase with traffic Traffic exposed workers exposure; (n=114) Significant seasonal variation (n=100) general population

1999). Although several analytic factors have been discussed in regards to discordant results (Phillips and Castegnaro, 1999), some studies point to factors in the diet that may influence these types of comparisons. For example, GSTM1 null smokers were reported to have higher adduct levels compared with non-null smokers, and these associations were found to be modified by plasma antioxidants (Wang et al., 1998b). DNA adducts were found higher in GSTM1 null subjects in the EPIC – Italy study, and were inversely associated with the consumption of fresh fruit, vegetables, olive oil and antioxidants (Palli et al., 2000). Earlier studies suggested an important role for dietary antioxidants on DNA adduct results in smokers (Grinberg-Funes et al., 1994; Mooney et al., 1997). Given the significant role dietary constituents are now thought to play in lung cancer (Williams and Sandler, 2001) further work in this area is needed. Such information may help to improve preventive strategies involving anti-oxidants and dietary interventions (Saha et al., 2001; Jacobson et al., 2000). Expression of CYP1A1 as assessed by mRNA levels in lung tissue was shown to be positively correlated with variations in hydrophobic-DNA adduct levels in currently smoking lung cancer patients (Mollerup et al., 1999). Earlier work showed a strong correlation of microsomal aryl hydrocarbon hydroxylase activity in lung with BPDE – DNA adduct levels (Alexandrov et al., 1992). The importance of CYP1A1 gene polymorphisms and other CYP related gene polymorphisms in adduct formation is more ambiguous. Polymorphisms at the CYP1A1 locus were not associated with variations in adduct levels in occupationally exposed workers (Schoket et al., 2001; Grzybowska et al., 2000; Zhang et al., 2000; Pan et al., 1998; Motykiewicz et al., Oncogene

1998), nor in lung cancer patients (Schoket et al., 1998; Wiencke et al., 1999). One small study of adducts in lung cancer patients, however, did suggest an interaction of the GSTM1 null genotype with homozygosity of the CYP1A1 MspI variant (Rojas et al., 1998). Several studies have reported CYP1A1 variants to be associated with higher adduct levels in newborn blood and placental tissues (Whyatt et al., 1998, 2001). Despite years of study, the role of CYP1A1 gene polymorphisms in PAH activation remains unclear. The notion that DNA adduct studies in vivo among smokers and occupationally exposed populations or patients with smoking related cancers could help to elucidate the toxicologic functionality of CYP polymorphisms has not yet been proven valid. The unexpected and paradoxical finding of increased hepatic DNA adducts of BP in CYP1A1 knockout mice also suggests that further study is warranted on the kinetics of metabolic clearance of PAHs and its impact on DNA adduct formation (Uno et al., 2001). N-acetylation and aromatic amines After cancer of the lung and bronchi, bladder cancer is the most prevalent tobacco related malignancy. A series of earlier studies suggested that differences in carcinogen-adduct levels can correlate closely to differences in bladder cancer risks that are determined both by carcinogen exposure intensity and metabolic genotype (Vineis and Martone, 1996; Vineis and Ronco, 1992). Generally these studies support the idea that adduct levels correlate with smoking related risk, particularly those differences in risk associated with smoking black (air cured) versus blonde (flue cured) tobacco. Protein – carcinogen adduction has been the


7381

focus of most of the research. Some potential inconsistencies were raised, however, in studies of DNA adducts. For example, no evidence of case control differences in urothelial cell DNA adduct levels were recorded in bladder biopsies (Talaska et al., 1994), nor were associations observed between 4aminobiphenyl-DNA adduct levels and p53 mutations in bladder tumors (Martone et al., 1998). In urine, the major adduct forming species associated with smoking black tobacco was suggested to be PhIP (Peluso et al., 1991). More relevant to the more widely consumed flu cured tobacco is the identification of the putative N(deoxyguanosin-8-yl)-4-aminobiphenyl DNA adduct in exfoliated bladder cells, the concentration of which was found to be related to the number of cigarettes smoked per day (Talaska et al., 1991). Recently, in a case control study of bladder cancer, NAT2 slow acetylator genotypes were associated with increased aromatic DNA adducts in white blood cells from buffy coats obtained from cancer cases, but no such association was evident among the controls (Peluso et al., 2000). Researchers also reported that dietary fruits and vegetables were inversely related to adduct levels. Curiously, smoking was not found to be associated with the DNA adducts levels measured (Peluso et al., 2000). In the aforementioned case series of bladder cancer biopsies no association of NAT2 genotypes was observed with 4-ABP DNA adduct measurements although current smoking and smoking air cured tobacco were positively associated with increased adduct levels (Martone et al., 1998). NAT2 genotypes associated with slow acetylator phenotypes were associated with higher aromatic-DNA adduct levels (Hou et al., 2001; Matullo et al., 2001a). No effect of either NAT1 or NAT2 polymorphisms was found in an analysis of aromatic amine-DNA adducts in another study (Godschalk et al., 2001). These mixed results relating adduct measurements and NAT2 genotypes should be considered in light of the overall modest increase in risk for bladder cancer associated with polymorphisms at this locus indicated in a recent meta-analysis of epidemiologic studies (Marcus et al., 2000). Quinone reductases and oxidative DNA damage Only a few studies have examined DNA adducts in pancreatic tissues of smokers and non-smokers. One of these concluded that neither GSTM1 nor NQO1 modified DNA adducts related to oxidative stress (Kadlubar et al., 1998). This study must be considered preliminary with regard to the less prevalent genotypes (e.g. NQO1 and GSTT1 homozygous variant individuals), and more studies could be valuable. Cigarette smoking is one of only a few identified environmental risk factors for pancreatic cancer. A recent study (Duell et al., 2002) indicated an increased risk of GSTT1 null subjects among heavy smokers and perhaps a higher risk among women who were GSTT1 null and who smoked heavily. A strong mechanistic rationale implicates metabolic polymorphisms in the formation of smoking related

DNA adducts, however, the evidence supporting genetic modification of DNA adducts from cigarette smoking in vivo in man must be considered equivocal. Strong associations of CYP1A1 expression and enzyme activity with PAH – DNA adducts in vivo and negative or inconsistent findings with CYP1A1 genetic variants argues against the functional importance of the known CYP1A1 polymorphisms. Indeed, the relationships between the common genetic variations and enzyme inducibility is still unclear. Extensive characterization of the catalytic properties of GSTP1 variants and the weak or inconsistent relationships in vivo require further evaluation. Hemoglobin carcinogen adducts appear a better marker of biological effective dose for bladder cancer compared with DNA adducts. DNA repair polymorphisms XRCC1, XCRCC3, XPD, ERCC2, hOGG1, p53 While several epidemiological studies have implicated polymorphisms in DNA repair enzymes in cancers of the lung (Ratnasinghe et al., 2001; Park et al., 2002; Divine et al., 2001; David-Beabes and London, 2001; Butkiewicz et al., 2001; Cheng et al., 2000a; Sugimura et al., 1999), head and neck (Olshan et al., 2002; Sturgis et al., 1999) and bladder (Stern et al., 2001), several of these reports have indicated protective effects of alleles that are presumed to compromise DNA repair activities (David-Beabes and London, 2001; Olshan et al., 2002; Stern et al., 2001). One possible explanation for this apparent inconsistency, drawn from recent investigations of ultraviolet radiation related basal and squamous skin cancer, is that there are different effects of DNA damage in cells with intact apoptotic mechanisms versus those that have already lost the apoptotic response to DNA damage stimuli. Lower repair capacity could thus be protective for cancer, in that deficient cells would be targeted for cell death rather than survive and harbor deleterious mutations (Nelson et al., 2002). Such a model predicts interactions of DNA repair polymorphisms with tobacco carcinogen dose and/or dose intensity. Further study is obviously required, as both positive (Park et al., 2002) and negative interactions (David-Beabes and London, 2001) of heavy smoking and the XRCC1 Gln allele have already been recorded for lung cancer risk. Although these epidemiologic observations make it problematic to formulate a priori hypotheses regarding DNA adducts and repair polymorphisms, several reports indicate increased adduct burdens in individuals who carry DNA repair variants. In a study of blood MNC DNA adducts with chromatographic properties similar to adducts detected in cells treated with simple phenols and quinones, researchers observed a significant association of the XRCC1 Gln allele with detectable DNA adduct levels. Adduct levels also showed a positive interaction with blood donor age but appeared unrelated to cigarette smoking (Duell et al., 2000). No association was observed for polymorphisms in the ERCC2 gene and phenol related Oncogene


7382

adducts. Higher levels of bulky hydrophobic – DNA adducts were observed in white blood cells of nonsmokers who were homozygous for the XRCC1 399Gln allele (Matullo et al., 2001b). In another study, XRCC3 variant alleles were associated with higher adduct levels in non-smokers (Matullo et al., 2001a). Among traffic exposed workers the XPD 751Gln allele was associated with higher adduct levels (Palli et al., 2001). A polymorphism involving the endonuclease 8-oxoguanine DNA glycosylase I (hOGG1;[C1245G;Exon7,Ser326Cys]) was not found to affect aromatic DNA adducts in pancreatic tissues (Li et al., 2002). No association of 8-oxo-dG levels and hOGG1 were found in lung tumor and normal lung tissue (Hardie et al., 2000). Recent work has questioned the functional importance of the hOGG1 Ser326Cys variant in repair of 8-oxo-dG lesions (Janssen, 2001, #399) and in lung cancer pathogenesis itself (Wikman et al., 2000). Finally, PAH – DNA adduct levels in white blood cells were suggested to be associated with the occurrence of a P53 polymorphism (Zhang et al., 2000). These early studies suggest that DNA repair polymorphisms may help us to understand endogenous or age related DNA adduct formation and removal. More comprehensive studies of smokers are required to address the role of these genetic variants in smokinginduced DNA damage. Studying interactions among gene loci and carcinogen dose will require more sophisticated models and statistical analyses. At this time it is difficult to predict whether high risk genotypes will be associated with increased or decreased adduct level. The relationship may be different at early phases of a chronic exposure compared with later times and the establishment of genetically altered tissues. New candidate gene polymorphisms Additional polymorphic candidate genes with potential to modify tobacco related cancer risk via modulation of DNA adduct levels, include the phenol sulfotransferases such as SULT1A1 (Nowell et al., 2000), myeloperoxidase (Petruska et al., 1992; London et al., 1997; Le Marchand et al., 2000; Schabath et al., 2000; Misra et al., 2001; Rojas et al., 2001; Kantarci et al., 2002), dihydronicotinamide riboside (NRH)-quinone oxidoreductase 2 (NQQ2) (Jaiswal et al., 1999; Long and Jaiswal, 2000), UDP-glucuronosyl transferases (Ren et al., 2000; Grant and Bell, 2000; Yueh et al., 2001), and the transcription factor Nrf2, which is essential for inducible and constitutive expression of a group of detoxification and antioxidant enzymes (Aoki et al., 2001).

DNA adducts, cancer risk associated with the induction of DNA adducts in biopsied cells, and correlative studies of adduct levels with exposures to known smoking carcinogens in target or surrogate tissues. The latter indirectly support a causal relationship of DNA adduct formation with cancer risk, Table 2. DNA adducts associated with lung cancer Studies of DNA adducts in lung cancer patients have indicated higher adduct levels in lung tissue of cancer cases (Cheng et al., 2000b, 2001) and in their peripheral white blood cells (Vulimiri et al., 2000; Perera et al., 1989) compared with controls. Higher adduct levels were reported in lung tissue from women compared with men (Cheng et al., 2001). In a prospective study, researchers reported that increased adduct levels in white blood cells were associated with lung cancer risk among those who were current smokers at the time of blood sampling (Tang et al., 2001). A small cohort study of heavy smokers reported that high adduct levels in bronchoalveolar cells were associated with higher cancer mortality (Bonassi et al., 2001), though not specifically lung cancer. DNA adducts induced in vitro have been associated with lung cancer risk in case – control studies (Li et al., 2001b; Wei et al., 2000). DNA adducts and cancers of the bladder, head and neck, pancreas and uterine cervix DNA adduct levels in white blood cells were significantly associated with bladder cancer risk (Peluso et al., 1998, 2000). A series of studies have shown that smoking is associated with smoking related adducts in the oral cavity and larynx (Banaszewski et al., 2000; Szyfter et al., 1999a,b; Nath et al., 1998). In a case – control study, induced DNA adduct levels were found to be associated with risk of head and neck cancer (Li et al., 2001a). Particularly striking was the finding of specific adducts in gingival tissue derived from unsaturated aldehydes in cigarette smoke (Nath et al., 1998). A small study of pancreatic cancer patients reported higher levels of several DNA adduct species in nontumorous pancreatic tissues from cases compared with control tissues (Wang et al., 1998a). Smoking is a risk factor in cervical cancer and DNA adducts related to smoking have been observed in cervical cells of smokers (Melikian et al., 1999a). Interestingly, HPV infection appeared to enhance induction of DNA adducts by BPDE in cervical epithelial cells (Melikian et al., 1999b) suggesting a mechanism for the interaction of HPV and smoking in cervical cancer risk. DNA-adducts and exposure to lung carcinogens

DNA adducts and cancer risk Three lines of epidemiologic evidence implicate DNA adducts in the etiology of smoking related cancers. These include: case – control and cohort studies showing associations of cancer occurrence with smoking related Oncogene

Studies of highly occupationally exposed populations have generally reported DNA adducts to be related to airborne PAH exposure but sorting out the independent effects of occupational and smoking related contributions to total PAH burdens has been problematic. Some exposures postulated to pose an increased


7383 Table 2 Studies of DNA adducts in tabacco related cancer risk Hypothesis/question

Adduct

Tissue

Study design/Population

Results

Reference

1

Is in vivo adduct level a risk factor for lung cancer?

PAH-DNA adducts

WBC

Is in vivo adduct level a risk factor for lung cancer? Is in vivo adduct level in lung a risk factor for lung cancer Is in vivo adduct level in lung associated with lung cancer, are there gender differences?

Aromatic-DNA adducts

WBC

Aromatic/ hydrophopic – DNA adducts PAH-DNA adducts

Non-tumorous lung tissue Non-tumorous lung

No overall association of adducts with lung cancer; suggested interaction of adducts and GSTM1 genotype on case-control status Among current smokers mean adduct levels were greater among cases compared with controls Lung cancer patients had greater DNA adduct levels compared with controls Lung cancer patient had higher adduct levels measured by two techniques, women had higher adducts compared with men

Tang, et al., 1998

2

Aromatic-DNA adducts

Blood lymphocytes

Hospital based lung cancer case cotnrol study Case (n=136) Control (n=115) Prospective lung cancer study Nested case (n=89) control (n=173) study (all males) Hospital based Case (n=73) Control (n=33) Lung cancer case-referent (Taiwan) 62 non-smoking lung cancer patients, 20 non-cancer controls Hospital based Case (n=43) Control (n=47)

Higher adduct levels in cases compared with controls; includes minority populations

Vulimiri et al., 2000

Cohort 40 heavy smokers

Smokers dying from cancer had twofold higher levels than smokers still alive at follow-up Cases were 2.15-fold more likely to have induced adduct levels above the median control level; positive dose response of induced adduct level and risk Cases were 3.7-fold more likely to have detectable DNA adducts compared with controls; suggested interaction with low fruit/vegetable consumption; no association of adducts with tobacco smoking Cases were 2.2-fold more likely to have adduct levels above the median control level; highest induced adduct levels observed in current smokers Adduct levels higher among current smokers; adduct levels greater in mouth floor cells compared with buccal mucosal cells Adduct levels positively correlated with cigarettes smoked per day, patient age, male gender and TNM stage. Adduct levels higher in normal versus tumor tissues. Adducts higher in low risk anatomic area Total levels of cyclic adducts (AdG and CdG) 4.4-fold greater in smokers compared with non-smokers. One CdG adduct increased 8.8-fold in smokers compared with non-smokers PAH-DNA adducts associated with smoking and increased with daily cigarette consumption; results not affected by diet, infection, alsohol Smoking associated adducts detected in non-tumorous tissues; higher adduct levels detected in nontumor tissues from cases compared with noncancer tissues No correlation between adduct level and PAH air concentrations was observed

Bonassi et al., 2001

3

4

5

Is adduct level in blood a risk factor for lung cancer?

6

Is cancer mortality associated with adduct level? Is induction of DNA adducts a risk factor for lung cancer?

7

8-oxo-dG PAH-DNA adducts BPDE-DNA adducts

Bronchoalveolar Cells In vitro treated blood lymphocytes

Hospital based case control Case n=221) Control (n=229)

8

Is the detection of DNA adducts a risk factor for bladder cancer?

Total aromatic DNA adducts

White blood cells

Bladder cancer case control Case (n=162) Controls (n=104)

9

Is induced DNA adduct level a risk factor in head and neck cancer?

BPDE-DNA adducts

In vitro treated blood lymphocytes

Hospital based head and neck case control study

10

Are adducts in oral tissues related to smoking and oral cancer risk?

PAH-DNA adducts

Mouth floor and buccal mucosal cells

11

Are adducts correlated with cigarette smoking in laryngeal tissues?

PAH-DNA Adducts

Laryngeal tissues, Squamous cell tumors

12

Does smoking induce &, &-unsaturated aldehyde related DNA adducts in gingival tissue? Does smoking induce DNA adducts in oral mucosal cells?

1,N2-propanodeoxyguanosine derived from acrolein (AdG) crotonaldehyde (CdG)

Gingical tissue

Smoking case referent 11 smokers 12 non-smokers

PAH-DNA adducts

Oral epithelial cells


14

Are in vitro DNA adducts associated with the occurence of pancreas cancer?

Aromatic and lipid peroxidation-related DNA adducts

15

Is working a carbon electrode manufacturing plant a risk factor for increased DNA adduct levels? Is working in a carbon electrode manufacturing plant a risk factor for increased DNA adduct levels? Do cooking oil fumes induce DNA adducts?

BP – DNA adducts

13

16

17

Case (n=91) Control (n=115) Smoking case referent Smokers (n=26) Non-smokers (n=22) Case series 33 primary laryngeal cancer patients (5 women) (28 men)

Pancreatic tumors Pancrease cancer case series, and normal case-referent pancreas tissues Cases (13 normal, 20 tumor) Non-cancer (n=19 organ donors) WBC Cross-sectional study of 17 workers in a plant producing carbon electrodes

PAH-DNA adducts

Blood lymphocytes

Cross sectional study of workers exposed to 5200 mg PAH/m3 in air

BPDE-DNA adducts

In vitro treatment of lung adenocarcinoma cell lines

In vitro study

High PAH exposure increased urinary 1-OHPY excretion but not DNA adducts; tobacco smoking was associated with increases in DNA adducts Cooking oil fumes implicated in lung cancer among nonsmokers induce BPDE-DNA adducts

Tang et al., 2001 Cheng et al., 2000b Cheng et al., 2001

Li et al., 2001

Peluso et al., 2000 See also Peluso et al., 1998 Li et al., 2001a

Besarati Nia, et al., 2000

Banaszewski et al., 2000 See also Szyfter et al., 1999b

Nath et al., 1998

Ramano, et al., 1999

Wang et al., 1998

Arnould et al., 1999

Van Delft et al., 1998b

Yang et al., 2000

Continued Oncogene


7384

Table 2 (Continued )

18

Hypothesis/question

Adduct

Tissue

Study design/Population

Results

Reference

Does smoking lead to detectable DNA adducts in sputum?

Lipophilic-DNA adducts

Inhaled saline inhaled sputum

Smoking case referent 20 smokers 24 healthy non-smokers

Adduct levels greater in sputum from smokers compared with non-smokers; enrichment by the P1 nuclease method more sensitive thn butanol to smoking BPDE-DNA adducts in blood positively correlated urinary 1-OHPY

Nia, et al., 2000

19

Does occupational PAH exposure correlated with adduct levels in blood? 20 Do smoking related adducts in sputum correlated levels in peripheral lymphocytes?

BPDE-DNA adducts

White blood cells Cross-sectional study of coke oven workers

Lipophilic-DNA adducts

Inhaled saline induced sputum Blood lymphocytes Lung tissues obtained at autopsy

21

Do PAH levels correlate with PAH-DNA adduct levels in lung tissues?

BPDE-DNA levels

22

Are adducts associated with age?

BPDE-DNA adducts

Peripheral blood lymphocytes

23

Are DNA adduct levels associated with moderate to low levels of PAH exposure from urban air pollution?

PAH-DNA adducts

Isolated peripheral blood lymphocytes

24

Are blood MNCs a good surrogate for detection of smoking related DNA adducts? Are blood MNCs a good surrogate for detection of smoking related DNA adducts?

Aromatic DNA adducts

Bulky hydrophobicDNA adducts

Isolated MNCs, broncho-alveolar lavage cells, white cell subtypes Isolated MNCs lung tissue

Does high PAH exposure increase DNA adduct level In blood cells?

BPDE-DNA adducts

Isolated MNCs

28 Does smoking induce DNA adducts in cells of the uterine cervix? 29 Is smoking associated with increased levels of DNA adducts in cervical tissue?

PAH-DNA adducts

Cervical cells

25

27

30

Does HPV-16 infection potentiate adduct formation from smoking carcinogens?

BPDE-DNA adducts Cervical epithelial and stromal tissue samples BPDE-DNA adducts

In vitro treated epethelial cervical cells


BPDE-DNA adduct levels correlated with BP concentration in tissues and were higher in smokers compared with non-smokers Cross-sectional study of Smokers with more than 20 packsmokers and non-smokers years had increased adduct levels; adducts positively correlated with age but not daily cigarette consumption Population-based comparison Despite higher urban air PAH of 194 young adult subjects exposure, adduct levels in subjects living in two Greek cities from Athens were not lower than subjects (Halkida, Athens) from Halkida; adduct levels greater in men compared with women Cross-sectional study of 78 Adduct levels in MNCs or BAL were smokers correlated with daily cigarette tar exposure using the P1 nuclease method but not butanol extraction Cross-sectional study of 147 Blood MNC adduct levels highly lung cancer patients correlate with lung adduct levels; cigarettes per day a predictor of adduct levels in current smokers, early age of smoking predicts adduct level in former smokers Occupational PAH case Occupational (coke oven) but not referent medicinal (coal tar) exposure 26 psoriasis patients associated with increased adduct (coal tar), levels; neither charcol grilled meat 15 coke oven workers, nor smoking associated with adduct 19 chimney sweeps, 36 levels non-PAH exposed Smoking case referent Mean adduct level higher in epithelial 16 smokers cervical cells of smokers compared 16 non-smokers with non-smokers Case series of tissues from 17 Mean BPDE-adduct levels in women undergoinf surgery epithelial but not stromal tissues for non-malignant disease higher in smokers compared with non-smokers In vitro cell culture Greater levels of BPDE adducts experiment induced in HPV-16 immortalized cells compared with normal cells

risk for lung cancer have failed to lead to detectable increases in adduct levels. For example, among workers exposed to PAHs in carbon electrode manufacturing no correlations were found between DNA adducts and occupational PAH exposures (Arnould et al., 1999; van Delft et al., 1998b). In another study, and one of the only population based investigations on DNA adducts, researchers did not find that PAH – DNA adducts in blood lymphocytes correlated with differences in PAH exposure related to urban air pollution (Kyrtopoulos et al., 2001). As indicated above, adduct levels in white blood cells have been associated with tobacco related cancer risk. Direct correlations among surrogate tissue and target tissues, however, are few in number. An early important observation was that smoking effects on adducts were stronger when measurements were made using DNA isolated from blood MNC compared with Oncogene

Cross-sectional autopsy study

Current smoking was associated with increased adduct levels in both tissues; higher levels in sputum; stable adduct levels over time

Arnould, et al., 2000 Nia, 2000

Lodovici et al., 1998

Shinozaki, et al., 1999

Kyrtopoulos et al., 2001

Godschalk et al., 1998

Wiencke et al., 1999

Pavenello, et al., 1999

Vineis et al., 2000 Melikian et al., 1999a

Melikian et al., 1999b

the DNA from the short-lived granulocyte fraction of WBC (Savela and Hemminki, 1991). In a subsequent study, hydrophobic-DNA adducts in blood MNCs were found to be a reliable surrogate of lung DNA adduct levels (Wiencke et al., 1995). Another researcher used induced sputum to assess smoking related DNA adducts (Nia et al., 2000). White blood cell adduct levels were shown to provide a suitable surrogate for DNA adducts in bronchoalveolar cells (Godschalk et al., 1998). The lung retains PAHs, and higher concentrations can be found in men compared with women, and perhaps among high risk ethnic minority groups (Seto et al., 1993; Goldman et al., 2001). Interestingly, concentrations of PAHs measured directly in lung tissue at autopsy were found to correlate well with PAH – DNA adduct levels measured in the same tissues (Lodovici et al., 1998). This observation suggests that exposed lung may constitute


7385

a tissue reservoir of unmetabolized PAHs. Immune cells from the peripheral blood or bronchoalveolar cells that migrate and circulate through lung tissue may be exposed to such accumulated PAHs. This may help explain the correlation of peripheral blood or bronchoalveolar cell DNA adduct levels with those measured in lung parenchyma and airways. Dose response for DNA adducts and cigarette smoking The nature of the dose – response for cigarette smoking and DNA adduct formation has significant implications in assessing cancer risks associated with tobacco use (Vineis et al., 2000). It is significant, therefore, that it is still somewhat unclear which measures of tobacco smoke exposure are most highly correlated with smoking related DNA adduct levels. Several studies now support the idea that current smokers have higher adduct levels compared with former smokers, and that daily cigarette consumption (cigarettes per day) is the most predictive measure of variations in adduct levels in target (Wiencke et al., 1999; Szyfter et al., 1999a; Banaszewski et al., 2000; Romano et al., 1999) and surrogate tissues of current smokers (Nia et al., 2000; Hou et al., 2001). In one study the daily cigarette tar consumed, but not the number of cigarettes, was found to be predictive of adduct levels in blood MNCs (Godschalk et al., 1998). Daily cigarette consumption in current smokers also predicted PAH – DNA adduct levels in cardiac tissue from patients undergoing open heart surgery (van Schooten et al., 1998). Nonetheless some investigations have reported higher levels of aromatic-DNA adducts in former smokers (Vulimiri et al., 2000). Earlier suggestions (Phillips et al., 1988) that DNA adducts may reflect lifetime cumulative tobacco smoke exposure as estimated by the pack – year variable do not appear to have been substantiated. Few studies have examined predictors of adduct levels among former smokers specifically. One of the earliest studies of lung PAH – DNA adducts in former smokers reported that adduct levels were similar to those observed among never smokers after about 5 years of smoking cessation (Phillips et al., 1988). Later, a biphasic loss of PAH – DNA adducts was proposed that included a rapid initial phase followed by a second slower phase of adduct loss (Schoket et al., 1993). Studies of subjects in a smoking cessation program indicated an estimated half-life of PAH – DNA adducts of about 9 – 13 weeks in blood leukocytes (Mooney et al., 1995). The half-life of PAH – DNA adducts in the lung is probably much more protracted. One study of lung cancer patients indicated a half-life of approximately 1.7 years (Schoket et al., 1998) among short-term quitters. In a study of lung tissues from former smokers that included long-term quitters (410 years), several statistical models were applied that suggested an approximate 10% decline in adducts per year for about 9 years and a much slower residual decline after 9 years (Thurston et al., 2000). Among these former smokers early age of smoking initiation was the most

important predictor of interindividual variations in lung DNA adduct levels after adjusting for other measures of cigarette consumption (Wiencke et al., 1999). Aromatic-DNA adducts were also found to be increased in smokers who started smoking at an early age in another study (Vulimiri et al., 2000). Time elapsed since quitting smoking would seem a logical inverse predictor of DNA adducts in former smokers but this has not been reported. Although animal models have been used to address these questions (Helleberg et al., 2001; Izzotti et al., 1999) they do not take into account the protracted exposures typical in humans, nor the cellular and physiologic changes induced in tissues undergoing chronic exposure, which may modify adduct induction and repair. As indicated below, widespread accumulations of genetically altered clones of cells is known to occur in smokers chronically exposed to tobacco carcinogens (Franklin et al., 1997). DNA adducts and somatic mutation in tumors DNA adducts and p53 A review several years ago cited the need for human clinical studies of DNA adducts and the mutational spectra of cancer-relevant genes, and noted the paucity of studies in which mutations as a consequence of DNA damage at specific genes had been investigated in vivo (Kriek et al., 1998). This conclusion is still relevant today. The important mechanistic hypotheses on p53 mutational spectrum and smoking (see Hussain in this volume) raise significant challenges for clinical and epidemiological investigators to translate these insights into human populations. Early relatively small studies suggested that aromatic-DNA adduct levels were increased in tissues from patients with p53 mutant lung cancer compared with cases bearing non-p53 mutant tumors (Kure et al., 1996; Ryberg et al., 1997). Another study reported only a weak association between PAH – DNA adduct levels in lung tissue and p53 mutations (Andreassen et al., 1996). Our group found that lung cancer patients with high hydrophobic – DNA adduct levels in non-tumorous lung tissue (above the median adduct level) were threefold more likely to have a tumor containing a p53 mutation (Wiencke et al., 1998). This association was significant, even after adjusting for smoking status. In contrast, in a study of bladder cancer, 4-aminbiphenyl – DNA adducts were not found to be associated with the p53 mutational status of tumors (Martone et al., 1998). Larger studies with carefully collected smoking and lifestyle histories are necessary to define the relationship of DNA adduct burden in vivo and the p53 mutational status and spectra of human lung cancer. DNA adducts and LOH at 3p21 An event more common than p53 mutation in lung cancer is the LOH and deletion of regions of the shortarm of chromosome 3 (e.g. 3p14, 3p12). LOH in these Oncogene


7386

regions is strongly associated with smoking history (Nelson et al., 1998; Hirao et al., 2001; Zienolddiny et al., 2001) and is thought to be among the earliest events in lung carcinogenesis (Wistuba et al., 1999, 2000; Park et al., 1999; Endo et al., 2000; Mitsudomi et al., 1996). These early somatic events may be the molecular basis for what Slaughter first termed ‘field cancerization’. According to this idea, grossly normal epithelium can be ‘preconditioned’ by carcinogen exposure. Such preconditioning may result in early tobacco-induced somatic mutations that are propagated by enhanced cell replication into a clonal ‘field’ of genetically altered progenitor cells. Consistent with these observations, we found a strong correlation of early age of smoking initiation with LOH at a loci near the hMLH1 gene on chromosome 3p21 in surgically resected lung cancer patients (Hirao et al., 2001). Interestingly, our investigations also revealed a link between these events and increased hydrophobic – DNA adduct levels in lung tissue from the same patients (Hirao et al., 2001). Another group, using microsatellite markers close to hMLH1, reported a similar increase in PAH – DNA adduct levels in lung tissues of lung cancer patients whose tumors contained LOH at 3p21 (Zienolddiny et al., 2001), although the question of age at smoking initiation was not addressed in the latter study. In a study of lymphocytes from lung cancer patients and controls, the highest aromatic DNA adduct levels were observed among cases who started smoking before 15 years of age (Vulimiri et al., 2000). These links between DNA adducts and somatic alterations in the lung and early smoking initiation should be considered in light of the epidemiologic observations that early smoking initiation is an independent risk factor for lung cancer (Hegmann et al., 1993). Taken together these associations lead to a novel ‘critical period’ hypothesis for lung cancer susceptibility (Wiencke and Kelsey, 2002): growth and development of the lungs during adolescence may set up a critical period of susceptibility to tobacco-related DNA damage. This idea is consistent with the burst of lung growth that is known to occur in the early teen years. Exposure to tobacco carcinogens during this period may lead to the propagation of fields of altered epithelial cells, particularly in the airway, that later evolve into frank malignancy. Exposure after the critical growth period may have a much lower probability of inducing large, clonal outgrowths with the same malignant potential. Emerging issues in tobacco carcinogenesis and DNA adducts Endogenous adducts, oxidative DNA damage and indirect action It has been suggested that damage to DNA from endogenous exposures may be orders of magnitude greater than those from exogenous sources including exposure to tobacco smoke (Povey, 2000). This does Oncogene

not imply, however, that endogenous agents do not play a role in tobacco carcinogenesis, because tobacco exposures can directly and indirectly increase oxidative stress. In animals, cigarette smoke exposures were shown to enhance endogenous DNA adducts by several fold in different tissues (Gupta et al., 1999; Zhou et al., 2000). Significant progress has been made in characterizing carbonyl-containing products of lipid peroxidation (e.g. malondialdehyde, 4-hydroxy-2-alkenals) and the adducts they form with DNA (Leuratti et al., 1998; Chen et al., 1998; Moller and Wallin, 1998; Yi et al., 1998) as well as the contributions of physiologic mediators such as nitric oxide in these reactions (Nair et al., 1998). The DNA adducts formed in reactions with carcinogenic aldehydes may produce structurally unique adducts (Hecht et al., 2001); these hold promise as sensitive markers of oral cancer risk associated with tobacco use (Nath et al., 1998). Some endogenously derived DNA adducts are thought to increase with age (Josyula et al., 2000; Yang et al., 1998). Several reviews have been presented on endogenous DNA adducts and cancer risk (Burcham, 1998; Bartsch and Nair, 2001; Blair, 2001). Future studies of genetic susceptibility studies should consider the potential overlap between endogenous and exogenous pathways and sources of DNA adducts. For example, ethylene oxide, which is an endogenous metabolite of ethylene, is also present in cigarette smoke. Studies of blood hemoglobin adducts formed from ethylene oxide indicate that genetic deficiency in GSTT1 leads to higher adduct levels in smokers (Fennell et al., 2000) and non-smokers (Thier et al., 2001). DNA adducts in blood leukocytes were also reported to be higher in GSTT1 null subjects in the EPIC – Italy study (Palli et al., 2000). Chemically unstable DNA adducts The majority of studies of tobacco-related cancer and DNA adducts have focused on chemically stable modifications of DNA, like those produced by the reaction of BPDE with the exocyclic amino group of guanine. PAHs present in tobacco smoke, including BP, can also be metabolized via radical cation intermediates to electrophiles that bind to DNA bases and destabilize the N-glycosyl bond, causing rapid depurination or depyrimidination of the adducted bases (Bodell et al., 1989; Chen et al., 1996; Devanesan et al., 1996). In some experimental systems (e.g. mouse skin treated with BP) these unstable DNA adducts are much more prevalent than the stable DNA adducts. Understanding the relative importance of unstable versus stable adducts in tobacco carcinogenesis in humans and the role of DNA repair mechanisms that act on these different lesions will require further investigation (Melendez-Colon et al., 1997, 1999; Chakravarti et al., 2001). Future epidemiologic studies may be the only way to accurately assess the importance of depurinating adducts in tobacco carcinogenesis. This will require improvements in the methods for detecting such compounds in biological


7387

specimens. One sensitive method for detecting unstable DNA adducts in urine (i.e. using capillary electrophoresis – fluorescence line narrowing spectroscopy) has been developed (Roberts et al., 2000). Initial studies using this approach found BP adducted bases (e.g. BP6 N7Gua) in the urine of three of seven cigarette smokers and three of seven women exposed to coal smoke, but none were detected in urine from 13 control subjects (Casale et al., 2001).

Alternatives to case – control or cross-sectional studies include those employing prospective and interventional designs. For example, in the SELECT study (Selenium and Vitamin E Cancer Prevention Trial) the goal is to enroll 32 400 men aged 55 years and older and follow them for a 12 year study period (Hoque et al., 2001). Blood samples collected at baseline could be used for DNA adduct and candidate gene studies; an expected 800 lung cancer cases are predicted to occur within the study period.

Factors affecting the validity of DNA adduct studies Bias caused by seasonal variation in DNA adducts Exposure misclassification, statistical analysis, study design Others have reviewed some of the statistical and analytical issues that can compromise the validity of DNA adduct studies (Cuzick, 1995). The literature reviewed here alludes to a divergence in adduct results in current versus former smokers. This has important implications for the design of future investigations into the role of DNA adducts in tobacco-related cancer. At any single point in time the levels of tobacco-related DNA adducts in a human tissue represent a dynamic process that is dependent on a number of factors; these include the intensity and recency of exposure to tobacco smoke, the metabolic balance of activation and detoxification mechanisms and the repair or removal of adducts by DNA repair and/or cell turnover. Therefore, to observe relationships between putative modifiers of DNA adduct concentrations, it may be important to stratify on smoking status (current, former, never) in all statistical analyses or to employ multivariate statistical techniques that carefully address potential interactions of modifiers and smoking status. It may be expected, for example, that genetic modification of adduct levels arising through a metabolic mechanism (adduct formation) may be more evident among current smokers. Genetic factors affecting repair and tissue renewal (adduct removal) may be more evident in comparisons involving former smokers. In addition, many of the case – control studies cited above can be criticized, on the basis that DNA adducts may be linked to cancer occurrence through their association with tobacco smoke exposure. As DNA adducts are thought to be part of the causal pathway in tobacco-related cancer they are not formally considered to be confounders. Nonetheless, these considerations raise concerns about the clinical value of DNA adducts as independent predictors of tobaccorelated cancer risk. To mitigate against these concerns, the question of adducts as risk factors could be addressed in studies that carefully match cases and controls on cigarette smoking exposure. Case – control studies matching on smoking may also provide a more robust way to examine the role of candidate genes as modifiers of DNA adduct formation or repair. Because of the difficulty in recruiting suitable controls, however, such studies are more difficult to perform than traditional case – control studies.

Finally, a question still unresolved is whether studies of DNA adducts can be significantly biased by seasonal variations in adduct levels. Several studies have reported statistically significant variations in DNA adduct levels according to the season of study (Butkiewicz et al., 2000; Grzybowska et al., 1993; Moller et al., 1996; Motykiewicz et al., 1998; Palli et al., 2001; Perera et al., 1992; Tang et al., 1995; Topinka et al., 2000). Many of these reported higher ‘background’ adduct levels in winter compared with summer, and these variations appear to be related to the airborne concentrations of PAHs and nitro-PAHs, particularly for studies carried out in industrialized and polluted areas of Poland and the Czech Republic (Moller et al., 1996; Topinka et al., 2000). These observations suggest that seasonal variation in DNA adducts may not be a generalized phenomenon but is rather an issue that is regionally specific to those areas where ambient air quality varies widely with the season of the year. However, recent research also indicates the complexities of these ecological associations. For example, concentrations of PAH – DNA adducts levels in WBCs were found to peak in Northern Italy (Turin, Varese) in winter and in summer in Central and Southern Italy (Florence, Naples, Ragusa), which suggested opposite seasonal trends according to latitude (Palli et al., 2000). The authors speculate that in summer photochemical pollution may be higher at lower latitudes and play a relatively greater role in the induction of adducts compared to the winter months. In another study, blood PAH – DNA adducts in non-smokers were found to be higher in the small Greek city of Halkida compared with adduct levels in non-smokers living in Athens, despite the far greater air pollution burden that exists in Athens (Kyrtopoulos et al., 2001). Consequently, it would appear prudent to consider seasonal information and ambient air quality in planning future studies to avoid potential biases that may arise from regional or seasonal variations in DNA adducts in human populations.

Abbreviations GSTM1, GSTT1, GSTP1, Glutathione S-transferase M1, T1, P1; NAT1, NAT2, N-acetyltransferase type I and type II; NQQ1 NAD(P)H:quinone oxidoreductase type I; BP, benzo(a)pyrene; BPDE, 7,8-dihydro-7,8-dihydroxybenzo(a)pyrene 9, 10-oxide; MNC, mononuclear cells; PAH, Oncogene


7388

polynuclear aromatic hydrocarbons; GSH, glutathione; HPLC, high performance liquid chromatography; 1-OHPY, 1-hydroxypyrene; WBC, white blood cells; GS/MS, gas chromatography-mass spectrometry; eA, 1,N6-ethenodeoxyadenosine; eC, 3,N4-etheneodeoxycytidine; M1G, 3-(2-deoxyb-D-erythro-pentofuranosyl)pyrimido[1,2-a]purin-10(3H)-one; BP-6 N7Gua, 7-(benzo[a]pyren-6-yl)guanine; ELISA, enzyme linked immunosorbent assay; 8-oxo-dG, 8-hydroxy-2’-deox-

yguanosine; PhIP, 2-amino-1-methyl-6-phenylimidazao[4,5b]pyridine; HPV, human papilloma virus; EPIC, European Prospective Investigation into Cancer and Nutrition

Acknowledgements This work was supported by NIH ES06717.

References Alexandrov K, Rojas M, Geneste O, Castegnaro M, Camus AM, Petruzzelli S, Giuntini C and Bartsch H. (1992). Cancer Res., 52, 6248 – 6253. Andreassen A, Kure EH, Nielsen PS, Autrup H and Haugen A. (1996). Mutat. Res., 368, 275 – 282. Aoki Y, Sato H, Nishimura N, Takahashi S, Itoh K and Yamamoto M. (2001). Toxicol. Appl. Pharmacol., 173, 154 – 160. Arnould JP, Kubiak R, Belowski J, Belegaud J and Szczeklik J. (2000). Pathol. Bio. (Paris), 48, 548 – 553. Arnould JP, Pfohl-Leszkowicz A, Bach V, Libert JP and Belegaud J. (1999). Hum. Exp. Toxicol., 18, 314 – 321. Banaszewski J, Szmeja Z, Szyfter W, Szyfter K, Baranczewski P and Moller L. (2000). Eur. Arch. Otorhinolaryngol., 257, 149 – 153. Bartsch H and Nair J. (2001). Adv. Exp. Med. Biol., 500, 675 – 686. Bartsch H, Rojas M, Alexandrov K and Risch A. (1998). Recent Results Cancer Res., 154, 86 – 96. Beserati Nia A, Van Straaten HW, Godschalk RW, Van Zandwijk N, Balm AJ, Kleinjans JC and Van Schooten FJ. (2000). Environ. Mol. Mutagen, 36, 127 – 133. Binkova B, Topinka J, Mrackova G, Gajdosova D, Vidova P, Stavkova Z, Peterka V, Pilcik T, Rimar V, Dobias L, Farmer PB and Sram RJ. (1998). Muta. Res., 416, 67 – 84. Blair IA. (2001). Exp. Gerontol., 36, 1473 – 1481. Bodell WJ, Devanesan PD, Rogan EG and Cavalieri EL. (1989). Chem. Res. Toxicol., 2, 312 – 315. Bonassi S, Neri M and Puntoni R. (2001). Mutat. Res., 480 – 481, 349 – 358. Bucci F, Galati R, Zito R, Falasca G, Federico A and Verdina A. (1998). Anticancer Res., 18, 2669 – 2674. Burcham PC. (1998). Mutagenesis, 13, 287 – 305. Butkiewicz D, Grzybowska E, Phillips DH, Hemminki K and Chorazy M. (2000). Environ. Mol. Mutagen., 35, 99 – 105. Butkiewicz D, Rusin M, Enewold L, Shields PG, Chorazy M and Harris CC. (2001). Carcinogenesis, 22, 593 – 597. Carmichael PL. (1998). Arch. Toxicol., 20 (Suppl): 199 – 205. Casale GP, Singhal M, Bhattacharya S, RamaNathan R, Roberts KP, Barbacci DC, Zhao J, Jankowiak R, Gross ML, Cavalieri EL, Small GJ, Rennard SI, Mumford JL and Shen M. (2001). Chem. Res. Toxicol., 14, 192 – 201. Chakravarti D, Mailander PC, Li KM, Higginbotham S, Zhang HL, Gross ML, Meza JL, Cavalieri EL and Rogan EG. (2001). Oncogene, 20, 7945 – 7953. Chen HJ, Zhang L, Cox J, Cunningham JA and Chung FL. (1998). Chem. Res. Toxicol., 11, 1474 – 1480. Chen L, Devanesan PD, Higginbotham S, Ariese F, Jankowiak R, Small GJ, Rogan EG and Cavalieri EL. (1996). Chem. Res. Toxicol., 9, 897 – 903. Cheng L, Spitz MR, Hong WK and Wei Q. (2000a). Carcinogenesis, 21, 1527 – 1530. Cheng YW, Chen CY, Lin P, Huang KH, Lin TS, Wu MH and Lee H. (2000b). Eur. J. Cancer, 36, 1381 – 1388. Oncogene

Cheng YW, Hsieh LL, Lin PP, Chen CP, Chen CY, Lin TS, Su JM and Lee H. (2001). Environ. Mol. Mutagen., 37, 304 – 310. Coles B, Yang M, Lang NP and Kadlubar FF. (2000a). Cancer Lett., 156, 167 – 175. Coles BF, Anderson KE, Doerge DR, Churchwell MI, Lang NP and Kadlubar FF. (2000b). Cancer Res., 60, 573 – 579. Cuzick J. (1995). J. Natl. Cancer Inst., 87, 861 – 862. David-Beabes GL and London SJ. (2001). Lung Cancer, 34, 333 – 339. Devanesan PD, Higginbotham S, Ariese F, Jankowiak R, Suh M, Small GJ, Cavalieri EL and Rogan EG. (1996). Chem. Res. Toxicol., 9, 1113 – 1116. Divine KK, Gilliland FD, Crowell RE, Stidley CA, Bocklage TJ, Cook DL and Belinsky SA. (2001). Mutat. Res., 461, 273 – 278. Duell EJ, Holly EA, Bracci PM, Liu M, Wiencke JK and Kelsey KT. (2002). J. Natl. Cancer Inst., 94, 297 – 306. Duell EJ, Wiencke JK, Cheng TJ, Varkonyi A, Zuo ZF, Ashok TD, Mark EJ, Wain JC, Christiani DC and Kelsey KT. (2000). Carcinogenesis, 21, 965 – 971. Endo C, Sato M, Fujimura S, Sakurada A, Aikawa H, Takahashi S, Usuda K, Saito Y and Sagawa M. (2000). Surg. Today, 30, 695 – 699. Fennell TR, MacNeela JP, Morris RW, Watson M, Thompson CL and Bell DA. (2000). Cancer Epidemiol. Biomarkers Prev., 9, 705 – 712. Fields WR, Morrow CS, Doss AJ, Sundberg K, Jernstrom B and Townsend AJ. (1998). Mol. Pharmacol., 54, 298 – 304. Franklin WA, Gazdar AF, Haney J, Wistuba II, La Rosa FG, Kennedy T, Ritchey DM and Miller YE. (1997). J. Clin. Invest., 100, 2133 – 2137. Garner RC. (1998). Mutat. Res., 402, 67 – 75. Godschalk RW, Dallinga JW Wikman H, Risch A, Kleinjans JC, Bartsch H and Van Schooten FJ. (2001). Pharmacogenetics, 11, 389 – 398. Godschalk RW, Maas LM, Van Zandwijk N, van’t Veer LJ, Breedijk A, Borm PJ, Verhaert J, Kleinjans JC and van Schooten FJ. (1998). Carcinogenesis, 19, 819 – 825. Goldman R, Enewold L, Pellizzari E, Beach JB, Bowman ED, Krishnan SS and Shields PG. (2001). Cancer Res., 61, 6367 – 6371. Grant DJ and Bell DA. (2000). Mol. Carcinog., 29, 198 – 204. Grinberg-Funes RA, Singh VN, Perera FP, Bell DA, Young TL, Dickey C, Wang LW and Santella RM. (1994). Carcinogenesis, 15, 2449 – 2454. Grzybowska E, Butkiewicz D, Motykiewicz G and Chorazy M. (2000). Mutat. Res., 469, 271 – 277. Grzybowska E, Hemminki K, Szeliga J and Chorazy M. (1993). Carcinogenesis, 14, 2523 – 2526. Gupta RC, Arif JM and Gairola CG. (1999). Mutat. Res., 424, 195 – 205. Hardie LJ, Briggs JA, Davidson LA, Allan JM, King RF, Williams GI and Wild CP. (2000). Carcinogenesis, 21, 167 – 172.


7389

Harries LW, Stubbins MJ, Forman D, Howard GC and Wolf CR. (1997). Carcinogenesis, 18, 641 – 644. Hecht SS, McIntee EJ and Wang M. (2001). Toxicology, 166, 31 – 36. Hegmann KT, Fraser AM, Keaney RP, Moser SE, Nilasena DS, Sedlars M, Higham-Gren L and Lyon JL. (1993). Epidemiology, 4, 444 – 448. Helleberg H, Xu H, Ehrenberg L, Hemminki K, Rannug U and Tornqvist M. (2001). Mutagenesis, 16, 333 – 337. Hemminki K, Koskinen M, Rajaniemi H and Zhao C. (2000). Regul. Toxicol. Pharmacol., 32, 264 – 275. Hemminki K, Koskinen M and Zhao C. (2001). Int. J. Cancer, 92, 923 – 926. Hirao T, Nelson HH, Ashok TD, Wain JC, Mark EJ, Christiani DC, Wiencke JK and Kelsey KT. (2001). Cancer Res., 61, 612 – 615. Hoque A, Albanes D, Lippman SM, Spitz MR, Taylor PR, Klein EA, Thompson IM, Goodman P, Stanford JL, Crowley JJ, Coltman CA and Santella RM. (2001). Cancer Causes Control, 12, 627 – 633. Hou SM, Falt S, Yang K, Nyberg F, Pershagen G, Hemminki K and Lambert B. (2001). Cancer Epidemiol. Biomarkers Prev., 10, 133 – 140. Hu X, Herzog C, Zimniak P and Singh SV. (1999). Cancer Res., 59, 2358 – 2362. Izzotti A, Bagnasco M, D’Agostini F, Cartiglia C, Lubet RA, Kelloff GJ and De Flora S. (1999). Carcinogenesis, 20, 1499 – 1505. Jacobson JS, Begg MD, Wang LW, Wang Q, Agarwal M, Norkus E, Singh VN, Young TL, Yang D and Santella RM. (2000). Cancer Epidemiol. Biomarkers Prev., 9, 1303 – 1311. Jaiswal AK, Bell DW, Radjendirane V and Testa JR. (1999). Pharmacogenetics, 9, 413 – 418. Josyula S, Schut HA, Diwan BA, Anver MR and Anderson LM. (2000). Int. J. Oncol., 17, 811 – 818. Kadlubar FF, Anderson KE, Haussermann S, Lang NP, Barone GW, Thompson PA, MacLeod SL, Chou MW, Mikhailova M, Plastaras J, Marnett LJ, Nair J, Velic I and Bartsch H. (1998). Mutat. Res., 405, 125 – 133. Kantarci OH, Lesnick TG, Yang P, Meyer RL, Hebrink DD, McMurray CT and Weinshenker BG. (2002). Mayo Clin. Proc., 77, 17 – 22. Kato S, Bowman ED, Harrington AM, Blomeke B and Shields PG. (1995). J. Natl. Cancer Inst., 87, 902 – 907. Kihara M and Noda K. (1999). Cancer Lett., 137, 53 – 60. Kriek E, Rojas M, Alexandrov K and Bartsch H. (1998). Mutat. Res., 400, 215 – 231. Kure EH, Ryberg D, Hewer A, Phillips DH, Skaug V, Baera R and Haugen A. (1996). Carcinogenesis, 17, 2201 – 2215. Kyrtopoulos SA, Georgiadis P, Autrup H, Demopoulos N, Farmer P, Haugen A, Katsouyanni K, Lambert B, Ovrebo S, Sram R, Stefanou G and Topinka J. (2001). Mutat. Res., 496, 207 – 228. Le Marchand L, Seifried A, Lum A and Wilkens LR. (2000). Cancer Epidemiol. Biomarkers Prev., 9, 181 – 184. Leuratti C, Singh R, Lagneau C, Farmer PB, Plastaras JP, Marnet LJ and Shuker DE. (1998). Carcinogenesis, 19, 1919 – 1924. Li D, Firozi PF, Chang P, Wang LE, Xiong P, Sturgis EM, Eicher SA, Spitz MR, Hong WK and Wei Q. (2001a). Int. J. Cancer, 93, 436 – 440. Li D, Firozi PF, Wang LE, Bosken CH, Spitz MR, Hong WK and Wei Q. (2001b). Cancer Res., 61, 1445 – 1450. Li D, Firozi PF, Zhang W, Shen J, DiGiovanni J, Lau S, Evans D, Friess H, Hassan M and Abbruzzese JL. (2002). Mutat. Res., 513, 37 – 48.

Lodovici M, Akpan V, Giovannini L, Migliani F and Dolara P. (1998). Chem. Biol. Interact., 116, 199 – 212. London SJ, Lehman TA and Taylor JA. (1997). Cancer Res., 57, 5001 – 5003. Long 2nd DJ and Jaiswal AK. (2000). Chem. Biol. Interact., 129, 99 – 112. Marcus PM, Vineis P and Rothman N. (2000). Pharmacogenetics, 10, 115 – 122. Martone T, Airoldi L, Magagnotti C, Coda R, Randone D, Malaveille C, Avanzi G, Merletti F, Hautefeuille A and Vineis P. (1998). Int. J. Cancer, 75, 512 – 516. Marzilli LA, Koertje C and Vouros P. (2001). Methods Mol. Biol., 162, 395 – 406. Matullo G, Guarrera S, Carturan S, Peluso M, Malaveille C, Davico L, Piazza A and Vineis P. (2001a). Int. J. Cancer, 92, 562 – 567. Matullo G, Palli D, Peluso M, Guarrera S, Carturan S, Celentano E, Krogh V, Munnia A, Tumino R, Polidoro S, Piazza A and Vineis P. (2001b). Carcinogenesis, 22, 1437 – 1445. Melendez-Colon VJ, Luch A, Seidel A and Baird WM. (1999). Carcinogenesis, 20, 1885 – 1891. Melendez-Colon VJ, Smith CA, Seidel A, Luch A, Platt KL and Baird WM. (1997). Proc. Natl. Acad. Sci. USA, 94, 13542 – 13547. Melikian AA, Sun P, Prokopczyk B, El-Bayoumy K, Hoffman D, Wang X and Waggoner S. (1999a). Cancer Lett., 146, 127 – 134. Melikian AA, Wang X, Waggoner S, Hoffmann D and ElBayoumy K. (1999b). Oncol. Rep., 6, 1371 – 1376. Misra RR, Tangrea JA, Virtamo J, Ratnasinghe D, Andersen MR, Barrett M, Taylor PR and Albanes D. (2001). Cancer Lett., 164, 161 – 167. Mitsudomi T, Oyama T, Nishida K, Ogami A, Osaki T, Sugio K, Yasumoto K, Sugimachi K and Gazdar AF. (1996). Clin. Cancer Res., 2, 1185 – 1189. Moller L, Grzybowska E, Zeisig M, Cimander B, Hemminki K and Chorazy M. (1996). Carcinogenesis, 17, 61 – 66. Moller P and Wallin H. (1998). Mutat. Res., 410, 271 – 290. Mollerup S, Ryberg D, Hewer A, Phillips DH and Haugen A. (1999). Cancer Res., 59, 3317 – 3320. Mooney LA, Bell DA, Santella RM, Van Bennekum AM, Ottman R, Paik M, Blaner WS, Lucier GW, Covey L, Young TL, Cooper TB, Glassman AH and Perera FP. (1997). Carcinogenesis, 18, 503 – 509. Mooney LA, Santella RM, Covey L, Jeffrey AM, Bigbee W, Randall MC, Cooper TB, Ottman R, Tsai WY and Wazneh L et al. (1995). Cancer Epidemiol. Biomarkers Prev., 4, 627 – 634. Motykiewicz G, Michalska J, Pendzich J, Malusecka E, Strozyk M, Kalinowska E, Butkiewicz D, Mielzynska D, Midro A, Santella RM and Chorazy M. (1998). Toxicol. Lett., 96 – 97, 195 – 202. Nair J, Gal A, Tamir S, Tannenbaum SR, Wogan GN and Bartsch H. (1998). Carcinogenesis, 19, 2081 – 2084. Nath RG, Ocando JE, Guttenplan JB and Chung FL. (1998). Cancer Res., 58, 581 – 584. Nelson HH, Kelsey KT, Mott LA and Karagas MR. (2002). Cancer Res., 62, 152 – 155. Nelson HH, Wiencke JK, Gunn L, Wain JC, Christiani DC and Kelsey KT. (1998). Cancer Res., 58, 1804 – 1807. Ni J, Liu T, Kolbanovskiy A, Krzeminski J, Amin S and Geacintov NE. (1998). Anal. Biochem., 264, 222 – 229. Nia AB, Maas LM, Brouwer EM, Kleinjans JC and Van Schooten FJ. (2000). Carcinogenesis, 21, 1335 – 1340.

Oncogene


7390

Nowell S, Ambrosone CB, Ozawa S, MacLeod SL, Mrackova G, Williams S, Plaxco J, Kadlubar FF and Lang NP. (2000). Pharmacogenetics, 10, 789 – 797. Olshan AF, Watson MA, Weissler MC and Bell DA. (2002). Cancer Lett., 178, 181 – 186. Ozawa S, Schoket B, McDaniel LP, Tang YM, Ambrosone CB, Kostic S, Vincze I and Kadlubar FF. (1999). Carcinogenesis, 20, 991 – 995. Palli D, Russo A, Masala G, Saieva C, Guarrera S, Carturan S, Munnia A, Matullo G and Peluso M. (2001). Int. J. Cancer, 94, 121 – 127. Palli D, Vineis P, Russo A, Berrino F, Krogh V, Masala G, Munnia A, Panico S, Taioli E, Tumino R, Garte S and Peluso M. (2000). Int. J. Cancer, 87, 444 – 451. Pan G, Hanaoka T, Yamano Y, Hara K, Ichiba M, Wang Y, Zhang J, Feng Y, Shujuan Z, Guan D, Gao G, Liu N and Takahashi K. (1998). Carcinogenesis, 19, 1963 – 1968. Park IW, Wistuba II, Maitra A, Milchgrub S, Virmani AK, Minna J and Gazdar AF. (1999). J. Natl. Cancer Inst., 91, 1863 – 1868. Park JY, Lee SY, Jeon HS, Bae NC, Chae SC, Joo S, Kim CH, Park JH, Kam S, Kim IS and Jung TH. (2002). Cancer Epidemiol. Biomarkers Prev., 11, 23 – 27. Pavanello S and Clonfero E. (2000). Mutat. Res., 463, 285 – 308. Pavanello S, Favretto D, Brugnone F, Mastrangelo F, Dal Pra G and Clonfero E. (1999). Carcinogenesis, 20, 431 – 435. Peluso M, Airoldi L, Armelle M, Martone T, Coda R, Malaveille C, Giacomelli G, Terrone C, Casetta G and Vineis P. (1998). Cancer Epidemiol. Biomarkers Prev., 7, 341 – 346. Peluso M, Airoldi L, Magagnotti C, Fiorini L, Munnia A, Hautefeuille A, Malaveille C and Vineis P. (2000). Carcinogenesis, 21, 183 – 187. Peluso M, Castegnaro M, Malaveille C, Friesen M, Garren L, Hautefeuille A, Vineis P, Kadlubar F and Bartsch H. (1991). Carcinogenesis, 12, 713 – 717. Perera F, Brenner D, Jeffrey A, Mayer J, Tang D, Warburton D, Young TI, Wazneh L, Latriano L and Motykiewicz G et al. (1992). Environ. Health Perspect., 98, 133 – 137. Perera F, Mayer J, Jaretzki A, Hearne S, Brenner D, Young TL, Fischman HK, Grimes M, Grantham S, Tang MX, Tsai WY and Santella RM. (1989). Cancer Res., 49, 4446 – 4451. Petruska JM, Mosebrook DR, Jakab GJ and Trush MA. (1992). Carcinogenesis, 13, 1075 – 1081. Petruzzelli S, Celi A, Pulera N, Baliva F, Viegi G, Carrozzi L, Ciacchini G, Bottai M, Di Pede F, Paoletti P and Giuntini C. (1998). Cancer Res., 58, 4122 – 4126. Phillips DH and Castegnaro M. (1999). Mutagenesis, 14, 301 – 315. Phillips DH, Farmer PB, Beland FA, Nath RG, Poirier MC, Reddy MV and Turteltaub KW. (2000). Environ. Mol. Mutagen., 35, 222 – 233. Phillips DH, Hewer A, Martin CN, Garner RC and King MM. (1988). Nature, 336, 790 – 792. Piipari R, Savela K, Nurminen T, Hukkanen J, Raunio H, Hakkola J, Mantyla T, Beaune P, Edwards RJ, Boobis AR and Anttila S. (2000). Int. J. Cancer, 86, 610 – 616. Povey AC. (2000). Toxicol. Pathol., 28, 405 – 414. Ratnasinghe D, Yao SX, Tangrea JA, Qiao YL, Andersen MR, Barrett MJ, Giffen CA, Erozan Y, Tockman MS and Taylor PR. (2001). Cancer Epidemiol. Biomarkers Prev., 10, 119 – 123. Oncogene

Reddy MV. (2000). Regul. Toxicol. Pharmacol., 32, 256 – 263. Ren Q, Murphy SE, Zheng Z and Lazarus P. (2000). Drug Metab. Dispos., 28, 1352 – 1360. Risch A, Wikman H, Thiel S, Schmezer P, Edler L, Drings P, Dienemann H, Kayser K, Schulz V, Spiegelhalder B and Bartsch H. (2001). Pharmacogenetics, 11, 757 – 764. Roberts KP, Lin CH, Singhal M, Casale GP, Small GJ and Jankowiak R. (2000). Electrophoresis, 21, 799 – 806. Rojas M, Alexandrov K, Cascorbi I, Brockmoller J, Likhachev A, Pozharisski K, Bouvier G, Auburtin G, Mayer L, Kopp-Schneider A, Roots I and Bartsch H. (1998). Pharmacogenetics, 8, 109 – 118. Rojas M, Cascorbi I, Alexandrov K, Kriek E, Auburtin G, Mayer L, Kopp-Schneider A, Roots I and Bartsch H. (2000). Carcinogenesis, 21, 35 – 41. Rojas M, Godschalk R, Alexandrov K, Cascorbi I, Kriek E, Ostertag J, Van Schooten FJ and Bartsch H. (2001). Carcinogenesis, 22, 1015 – 1018. Romano G, Sgambato A, Boninsegna A, Flamini G, Curigliano G, Yang Q, La Gioia V, Signorelli C, Ferro A, Capelli G, Santella RM and Cittadini A. (1999). Cancer Epidemiol. Biomarkers Prev., 8, 91 – 96. Ryberg D, Hewer A, Phillips DH and Haugen A. (1994). Cancer Res., 54, 5801 – 5803. Ryberg D, Skaug V, Hewer A, Phillips DH, Harries LW, Wolf CR, Ogreid D, Ulvik A, Vu P and Haugen A. (1997). Carcinogenesis, 18, 1285 – 1289. Saha BK, Bishayee A, Kanjilal NB and Chatterjee M. (2001). Cell Mol. Life Sci., 58, 1141 – 1149. Savela K and Hemminki K. (1991). Carcinogenesis, 12, 503 – 508. Schabath MB, Spitz MR, Zhang X, Declos GL and Wu X. (2000). Carcinogenesis, 21, 1163 – 1166. Schoket B, Kostic S and Vincze I. (1993). IARC Sci. Publ., 124, 315 – 319. Schoket B, Papp G, Levay K, Mrackova G, Kadlubar FF and Vincze I. (2001). Mutat. Res., 482, 57 – 69. Schoket B, Phillips DH, Kostic S and Vincze I. (1998). Carcinogenesis, 19, 841 – 846. Seto H, Ohkubo T, Kanoh T, Koike M, Nakamura K and Kawahara Y. (1993). Arch. Environ. Contam. Toxicol., 24, 498 – 503. Shields PG, Bowman ED, Harrington AM Doan VT and Weston A. (1993). Cancer Res., 53, 3486 – 3492. Shinozaki R, Inoue S and Choi KS. (1998). Cytometry, 31, 300 – 306. Shinozaki R, Inoue S, Choi KS and Tatsuno T. (1999). Arch. Environ. Health, 54, 79 – 85. Soni M, Madurantakan M and Krishnaswamy K. (1998). Toxicology, 126, 155 – 162. Stern MC, Umbach DM, van Gils CH, Lunn RM and Taylor JA. (2001). Cancer Epidemiol. Biomarkers Prev., 10, 125 – 131. Sturgis EM, Castillo EJ, Li L, Zheng R, Eicher SA, Clayman GL, Strom SS, Spitz MR and Wei Q. (1999). Carcinogenesis, 20, 2125 – 2129. Sugimura H, Kohno T, Wakai K, Nagura K, Genka K, Igarashi H, Morris BJ, Baba S, Ohno Y, Gao C, Li Z, Wang J, Takezaki T, Tajima K, Varga T, Sawaguchi T, Lum JK, Martinson JJ, Tsugane S, Iwamasa T, Shinmura K and Yokota J. (1999). Cancer Epidemiol. Biomarkers Prev., 8, 669 – 674. Sundberg K, Dreij K, Seidel A and Jernstrom B. (2002). Chem. Res. Toxicol., 15, 170 – 179.


7391

Szeliga J and Dipple A. (1998). Chem. Res. Toxicol., 11, 1 – 11. Szyfter K, Banaszewski J, Jaolszynski P, Pabiszczak M, Szyfter W and Szmeja Z. (1999a). Acta Biochim. Pol., 46, 275 – 287. Szyfter K, Szmeja Z, Szyfter W, Hemminki K, Banaszewski J, Jaskula-Sztul R and Louhelainen J. (1999b). Mutat. Res., 445, 259 – 274. Talaska G, al-Juburi AZ and Kadlubar FF. (1991). Proc. Natl. Acad. Sci. USA, 88, 5350 – 5354. Talaska G, Schamer M, Casetta G, Tizzani A and Vineis P. (1994). Cancer Lett., 84, 93 – 97. Tan WG, Carnelley TJ, Murphy P, Wang H, Lee J, Barker S, Weinfeld M and Le XC. (2001). J. Chromatogr. A, 924, 377 – 386. Tang D, Phillips DH, Stampfer M, Mooney LA, Hsu Y, Cho S, Tsai WY, Ma J, Cole KJ, She MN and Perera FP. (2001). Cancer Res., 61, 6708 – 6712. Tang D, Santella RM, Blackwood AM, Young TL, Mayer J, Jaretzki A, Grantham S, Tsai WY and Perera FP. (1995). Cancer Epidemiol. Biomarkers Prev., 4, 341 – 346. Tang DL, Rundle A, Warburton D, Santella RM, Tsai WY, Chiamprasert S, Hsu YZ and Perera FP. (1998). Carcinogenesis, 19, 1949 – 1953. Thier R, Balkenhol H, Lewalter J, Selinski S, Dommermuth A and Bolt HM. (2001). Mutat. Res., 482, 41 – 46. Thurston SW, Wand MP and Wiencke JK. (2000). Biometrics, 56, 139 – 144. Timbrell JA. (1998). Toxicology, 129, 1 – 12. Topinka J, Schwarz LR, Wiebel FJ, Cerna M and Wolff T. (2000). Mutat. Res., 469, 83 – 93. Turteltaub KW and Dingley KH. (1998). Toxicol. Lett., 102 – 103, 435 – 439. Uno S, Dalton TP, Shertzer HG, Genter MB, Warshawsky D, Talaska G and Nebert DW. (2001). Biochem. Biophys. Res. Commun., 289, 1049 – 1056. Vainio H. (1998). Toxicol. Lett., 102 – 103, 581 – 519. van Delft JH, Baan RA and Roza L. (1998a). Crit. Rev. Toxicol., 28, 477 – 510. van Delft JHG, Steenwinkel MS, van Asten JG, de Vogel N, Bruijntjes-Rozier TC, Schouten T, Cramer P, Maas L, van Herwijnen MH, van Schooten F and Hopmans PM. (2001). Ann. Occup. Hyg., 45, 395 – 408. van Delft JH, Steenwinkel MJ, van Asten JG, van Es J, Kraak A and Baan RA. (1998b). Ann. Occup. Hyg., 42, 105 – 114. van Lieshout EM, Roelofs HM, Dekker S, Mulder CJ, Wobbes T, Jansen JB and Peters WH. (1999). Cancer Res., 59, 586 – 589. Van Schooten FJ, Hirvonen A, Maas LM, De Mol BA, Kleinjans JC, Bell DA and Durrer JD. (1998). FASEB J., 12, 1409 – 1417. Viezzer C, Norppa H, Clonfero E, Gabbani G, Mastrangelo G, Hirvonen A and Celotti L. (1999). Mutat. Res., 431, 259 – 269. Vineis P, Kogevinas M, Simonato L, Brennan P and Boffetta P. (2000). Mutat. Res., 463, 103 – 110. Vineis P and Martone T. (1996). Ann. Ist. Super Sanita, 32, 21 – 27. Vineis P and Ronco G. (1992). Environ. Health Perspect., 98, 95 – 99. Vulimiri SV, Wu X, Baer-Dubowska W, de Andrade M, Detry M, Spitz MR and DiGiovanni J. (2000). Mol. Carcinog., 27, 34 – 46.

Wang M, Abbruzzese JL, Friess H, Hittelman WN, Evans DB, Abbruzzese MC, Chiao P and Li D. (1998a). Cancer Res., 58, 38 – 41. Wang Y, Ichiba M, Iyadomi M, Zhang J and Tomokuni K. (1998b). Ind. Health, 36, 337 – 346. Watson MA, Stewart RK, Smith GB, Massey TE and Bell DA. (1998). Carcinogenesis, 19, 275 – 280. Wei Q, Cheng L, Amos CI, Wang LE, Guo Z, Hong WK and Spitz MR. (2000). J. Natl. Cancer Inst., 92, 1764 – 1772. Whyatt RM, Bell DA, Jedrychowski W, Santella RM, Garte SJ, Cosma G, Manchester DK, Young TL, Cooper TB, Ottman R and Perera FP. (1998). Carcionogenesis, 19, 1389 – 1392. Whyatt RM, Perera FP, Jedrychowski W, Santella RM, Garte S and Bell DA. (2000). Cancer Epidemiol. Biomarkers Prev., 9, 207 – 221. Whyatt RM, Jedruchowski W, Hemminki K, Santella RM, Tsai WY, Yang K and Perera FP. (2001). Cancer Epidemiol. Biomarkers Prev., 10, 581 – 588. Wiencke JK, Kelsey KT, Varkonyi A, Semey K, Wain JC, Mark E and Christiani DC. (1995). Cancer Res., 55, 4910 – 4914. Wiencke JK, Nelson H, Wain J, Mark E, Christiani DC and Kelsey KT. (1998). Association of increased PAH – DNA adducts and p53 mutations in lung cancer. Proceedings of the 89th Annual Meeting of the American Association for Cancer Research 39: 1998, p 562. Wiencke JK and Kelsey KT. (2002). Teen smoking, ‘field cancerization’ and a critical period hypothesis for lung cancer susceptibility Env. Health Perspect, June 2002. Wiencke JK, Thurston SW, Kelsey KT, Varkonyi A, Wain JC, Mark EJ and Christiani DC. (1999). J. Natl. Cancer Inst., 91, 614 – 619. Wikman H, Risch A, Klimek F, Schmezer P, Spiegelhalder B, Dienemann H, Kayser K, Schulz V, Drings P and Bartsch H. (2000). Int. J. Cancer, 88, 932 – 937. Williams MD and Sandler AB. (2001). Cancer Treat. Res., 105, 31 – 52. Wistuba II, Behrens C, Virmani AK, Mele G, Milchgrub S, Girard L, Fondon 3rd JW, Garner HR, McKay B, Latif F, Lerman MI, Lam S, Gazdar AF and Minna JD. (2000). Cancer Res., 60, 1949 – 1960. Wistuba II, Bryant D, Behrens C, Milchgrub S, Virmani AK, Ashfaq R, Minna JD and Gazdar AF. (1999). Clin. Cancer Res., 5, 991 – 1000. Xing JZ, Carnelley T, Lee J, Watson WP, Booth E, Weinfeld M and Le XC. (2001). Methods Mol. Biol., 162, 419 – 428. Yang SC, Weinfeld M, Jenq SN, Kang ZC and Lee H. (2000). Chem. Res. Toxicol., 13, 1046 – 1050. Yang K, Fang JL and Hemminki K. (1998). Mutat. Res., 422, 285 – 295. Yi P, Sun X, Doerge DR and Fu PP. (1998). Chem. Res. Toxicol., 11, 1032 – 1041. Yueh MF, Nguyen N, Famourzadeh M, Strassburg CP, Oda Y, Guengerich FP and Tukey RH. (2001). Carcinogenesis, 22, 943 – 950. Zhang J, Ichiba M, Feng Y, Pan G, Hanaoka T, Yamano Y, Hara K, Takahashi K and Tomokuni K. (2000). Int. Arch. Occup. Environ. Health, 73, 127 – 135. Zhou G, Hernandez NS, Randerath E and Randerath K. (2000). Mutat. Res., 447, 137 – 147. Zienolddiny S, Ryberg D, Arab MO, Skaug V and Haugen A. (2001). Br. J. Cancer, 84, 226 – 231.

Oncogene