Polymorphism of Xenobiotic-Metabolizing Genesand Breast Cancer Susceptibility in North Indian Women

Virendra Singh,1 Ghanshyam Upadhyay,1 Neeraj Rastogi,2 Kalpana Singh,3 and Mahendra Pratap Singh1

NAD(P)H:quinone oxidoreductase 1 (NQO1) and cytochrome P450 1A2 (CYP1A2) are involved in the metab-olism of estrogens. Genetic polymorphisms in these genes may lead to interindividual variation in breast cancersusceptibility. This study was undertaken to investigate the association of NQO1 exon 6 proline187serine(C609T) and CYP1A2 exon 2 phenylalanine21leucine (C63G) polymorphisms with breast cancer susceptibility inNorth Indian women. Polymorphisms were analyzed by polymerase chain reaction amplification of the desiredsegment of NQO1 and CYP1A2 genes followed by restriction fragment length polymorphism. NQO1 mRNAexpression was analyzed by semiquantitative reverse transcription–polymerase chain reaction and its enzymeactivity was estimated spectrofluorophotometrically. Odds ratios for NQO1 C609T heterozygous and homo-zygous variants were 0.66 (95% confidence interval: 0.39–1.13; p-value: 0.141) and 1.07 (95% confidence interval:0.46–2.46; p-value: 0.976). All cases and controls were monomorphic for the CYP1A2 exon 2 phenylalanine21-leucine (C63G) genotype. NQO1 mRNA expression and its catalytic activity among wild-type genotype, ho-mozygous variant, and heterozygous variant were not significantly altered, except for catalytic activity of theNQO1 homozygous variant, which was observed extremely low. The results of the study suggest that NQO1exon 6 proline187serine (C609T) and CYP1A2 exon 2 phenylalanine21leucine (C63G) polymorphisms do notplay a significant role in breast cancer susceptibility in North Indian women.

Introduction

Xenobiotic-metabolizing enzymes are involved in theactivation and detoxification of environmental estrogens

and mammary carcinogens and contribute to breast cancersusceptibility (Brockstedt et al., 2002). Various genetic, meta-bolic, and lifestyle factors are reported to alter the metabolismof environmental estrogens and determine breast cancer risk(Bradlow and Sepkovic, 2004; Singh et al., 2007, 2008). Geneticpolymorphisms in xenobiotic-metabolizing genes that alterenzymatic activity may lead to interindividual variations in theextent of DNA damage and cancer susceptibility (Brockstedtet al., 2002). Estrogens are normally beneficial to the body,sexual development, and fertility and are eliminated from thebody by metabolic conversion to estrogenically inactive me-tabolites, which are excreted via urine and/or feces (Bradlowand Sepkovic, 2004; Tsuchiya et al., 2005). Estrogens may act ascarcinogen facilitators in certain conditions and their effects areat least partially regulated by estrogen-metabolizing enzymes(Bradlow and Sepkovic, 2004; Long et al., 2007).

NAD(P)H:quinone oxidoreductase 1 (NQO1) metabolizesnitrosamines and heterocyclic amines, protects against free

radicals, and contributes to carcinogenesis (Cavalieri et al.,1997, 2000; Zhang et al., 2003; Lyn-Cook et al., 2006). On theother hand, cytochrome P450 1A2 (CYP1A2) catalyzes 2-hydroxylation of estrogens into catechol derivatives in hepaticand extra-hepatic tissues, such as breast (Hayes et al., 1996;Zhu and Conney, 1998a; Tsuchiya et al., 2005). Catechol de-rivatives are further metabolized into semiquinones andquinones and may undergo redox cycling, leading to thegeneration of reactive oxygen species, which may stably bindto DNA causing DNA damage (Cavalieri et al., 1997, 2000;Liehr, 2000). The NQO1 gene is localized in the p arm ofchromosome 16, consists of six exons and five introns, andspans 17,230 base pairs (Fowke et al., 2004). The CYP1A2 genelocalizes in the q arm of chromosome 15, consists of sevenexons and six introns, and spans 7758 base pairs. Geneticpolymorphisms in exon 6 of the NQO1 (C609T) and exon 2 ofthe CYP1A2 (C63G) genes are reported to play critical roles inbreast cancer susceptibility (Huang et al., 1999; Fowke et al.,2004). Epidemiological findings on NQO1 and CYP1A2 genepolymorphisms and their association with breast cancer riskhave been found to be inconsistent (Menzel et al., 2004;Sarmanova et al., 2004; Le Marchand et al., 2005; Long et al.,

1Indian Institute of Toxicology Research (Council of Scientific and Industrial Research), Lucknow, India.2Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow, India.3Sarojini Naidu Medical College, Agra, India.

GENETIC TESTING AND MOLECULAR BIOMARKERSVolume 15, Number 5, 2011ª Mary Ann Liebert, Inc.Pp. 343–349DOI: 10.1089/gtmb.2010.0197

343

2006; Hong et al., 2007). A NQO1 serine homozygous variant(C609T at codon 187) exhibits null enzyme activity; however,the CYP1A2 leucine homozygous variant (C63G at codon 21)exhibits altered enzyme activity (Moran et al., 1999; Sachseet al., 2003).

Although associations of single-nucleotide polymorphisms(SNPs) in NQO1 and CYP1A2 with breast cancer suscepti-bility are reported in some populations, despite high preva-lence of the disease, such association studies have not yet beendone in Indian women. This study was undertaken to inves-tigate the association of SNPs in the NQO1 and CYP1A2 genesand to assess their impact at the level of mRNA expressionand catalytic activities to determine their actual role in breastcancer susceptibility in Indian women.

Materials and Methods

Materials

Acrylamide, agarose, N,N0-methylene bisacrylamide, boricacid, bromophenol blue, chloroform, dicoumarol, ethylenediamine tetraacetic acid, magnesium chloride, phenol, resor-ufin ethyl ether, sucrose, tri-BD reagent, trisodium citrate, andxylene cyanol were obtained from Sigma Aldrich. Deoxyr-ibonucleotide triphosphates, proteinase K, forward and re-verse primers, and Taq DNA polymerase were purchasedfrom Bangalore Genei. Restriction enzymes (Hinf1 and MboII)and reverse transcription (RT)–polymerase chain reaction(PCR) kits were obtained from Fermentas. Acetic acid, ethylalcohol, and other chemicals were of analytical grade andpurchased locally from Sisco Research Laboratory.

Sample collection

The medical ethics committees of the Indian Institute ofToxicology Research, Lucknow, and the Sanjay GandhiPostgraduate Institute of Medical Sciences, Lucknow, ap-proved the study. A written informed consent was obtainedprior to blood collection from all individuals involved in thisstudy. The blood samples were collected from controls andbreast cancer cases by expert clinicians at the Sanjay GandhiPostgraduate Institute of Medical Sciences. A total of 200 fe-male controls and 200 female breast cancer patients were re-cruited for blood collection. Cases and controls were residentsof Lucknow or its adjacent localities in north. The patients andcontrols were from similar localities to avoid differences dueto environmental factors. The controls did not possess anytype of tumor/cancer and also did not exhibit any geneticdisorder or visible disease. The patients with clinically definedsymptoms of breast cancer were considered for this study(Tables 1 and 2). Patients and controls were classified intopremenopausal and postmenopausal groups based on thecase history and information given by them.

DNA isolation

Blood was collected in 3.8% sodium citrate in the ratio of9:1, mixed well, and stored at �208C. DNA was extractedaccording to the method previously described (Miller et al.,1998; Singh et al., 2007, 2008; Yadav et al., 2009). The purityand integrity of genomic DNA were checked on agarose gel(0.8%) and by calculating the ratio of absorbance at 260/280 nm (1.8–2.0). The DNA content was quantified by ab-sorbance at 260 nm (1.00 optical density was considered

equivalent to 50 mg/mL genomic DNA) before PCR ampli-fication.

Genotyping

Using appropriate primers and PCR amplification condi-tions, restriction fragment length polymorphism of exon 6 ofNQO1 gene-containing C609T sequence polymorphism wasperformed (Fig. 1) based on the information reported else-where (Fowke et al., 2004). Genotyping of CYP1A2*2(C63G)was performed as described elsewhere (Sachse et al., 2003).

RNA isolation, cDNA synthesis, and RT-PCR

Blood was collected in sodium citrate (3.8%, 9:1) and mixedwell. RNA was isolated from blood using tri-BD reagentand reverse transcribed into cDNA (Singh et al., 2007). NQO1and glyceraldehyde-3-phosphate dehydrogenase mRNA ex-pressions were analyzed concurrently using RT-PCR as de-scribed elsewhere (Kemp et al., 2003; Lyn-Cook et al., 2006).Relative expressions were calculated separately with respectto glyceraldehyde-3-phosphate dehydrogenase.

Isolation of white blood cells, cell lysis,and protein estimation

White blood cells were isolated from whole blood as pre-viously described (Singh et al., 2007). The viability of cells wastested by trypan blue exclusion test and was never less than

Table 1. Demographics Such as Age, Smoking

Habit, and Menopausal States of Cases

and Controls in This Study

Basic informationPatients

(n¼ 200)Controls(n¼ 200)

Age in years (mean� SE)Premenopausal 39.85� 0.65 32.39� 0.56Postmenopausal 54.66� 0.69 52.22� 0.76

Smoking habit, n (%)Smoker 6 (3.00%) 7 (3.50%)Nonsmoker 194 (97.00%) 193 (96.50%)

Menopausal status, n (%)Premenopausal 90 (45.00%) 117 (58.50%)Postmenopausal 110 (55.00%) 83 (41.50%)

Table 2. Information on Characteristics

Such as Stage, Histological Classification,

and Nodal Status of Tumor of the Patients

Recruited in This Study

Stage at which the tumor was diagnosed n (%)

�Stage I 8 (4.00)Stage II 102 (51.00)�Stage III 90 (45.00)

Histological classification of tumorTumor of duct 187 (93.50)Tumor of lobule 7 (3.50)Unclassified/mixed 6 (3.00)

Nodal status of the tumorþ 131 (65.50)� 69 (34.50)

344 SINGH ET AL.

95%. The cells were lysed in phosphate–potassium chloridebuffer (100 mM, pH 7.4) by sonication (Singh et al., 2007). Thereaction mixture was centrifuged and protein content wasestimated in the supernatant. A standard curve for bovineserum albumin was developed using the Bradford methodand the protein content was calculated by extrapolation of theoptical density of biological samples (Bradford, 1976).

Enzymatic activity of NQO1

NQO1 enzyme activity was measured spectrofluoro-photometrically by dicoumarol-mediated inhibition of resor-ufin reduction (Lee et al., 2005). Varying amounts of cell lysatewere mixed with NADPH (100 mM) and the volume was ad-justed with phosphate-buffered saline (200 mL, pH 7.4). Thereaction was initiated by the addition of resorufin stock so-lution (10 mL of 10mM; final concentration: 500 nM) and di-methyl sulfoxide (10%). The reduction of resorufin wasanalyzed spectrofluorophotometrically at 522 nm excitationand 590 nm emission spectra at room temperature for 2 min inthe absence and presence of dicoumarol. The rate of resorufinreduction was determined by comparing the change of fluo-rescence as a function of time, relative to the fluorescence of aknown amount of resorufin.

Statistical analysis

The data are expressed as means� standard error of themeans. The statistical analysis was performed using Epi Info-5software. The unadjusted odds ratio (OR) was calculatedseparately with 95% confidence interval (CI) for genotypefrequencies in total, premenopausal, and postmenopausal

breast cancer patients, when compared with respective controlsusing 2�2 contingency table. The statistical significance for ORwas calculated using the chi-square test; however, Fisher’sexact test was used where expected cell frequencies were <5.Hardy–Weinberg equilibrium for each SNP in cases and con-trols was calculated using Hardy–Weinberg calculator (www.oege.org/software/hardy-weinberg.shtml). One-way analy-sis of variance was used to analyze the significance level of thechange in mRNA expression and enzymatic activity.

Results

Allelic frequencies of SNP C609T of NQO1

The NQO1 C609T heterozygous genotype was found morefrequently in both patients and controls than the wild-type orhomozygous variant genotypes. Although a statistically sig-nificant difference in genotypic frequency between cases andcontrols was not found, the frequency of the heterozygous ge-notype was higher in total, premenopausal, and postmeno-pausal controls in comparison with respective patients. The ORfor heterozygous and homozygous variant genotypes were 0.66(95% CI: 0.39–1.13; p-value: 0.141) and 1.07 (95% CI: 0.46–2.46;p-value: 0.976) in total, 0.85 (95% CI: 0.40–1.81; p-value: 0.778)and 1.92 (95% CI: 0.54–6.97; p-value: 0.394) for premenopausalwomen, and 0.51 (95% CI: 0.22–1.16; p-value: 0.117) and 0.58(95% CI: 0.17–1.92; p-value: 0.461) for postmenopausal women(Table 3). The genotypic frequency of the T/T genotype washigher but not statistically significant in patients than respectivecontrols in premenopausal women (OR: 1.92; 95% CI: 0.54–6.97;p-value: 0.394). Similarly, the NQO1 genotype did not exhibitassociation with any pathological features of cancer (Table 4).

Allelic frequencies of SNP C63G of CYP1A2

Interestingly, all cases and controls studied exhibited onlythe wild-type genotype.

Hardy–Weinberg equilibrium for NQO1C609T polymorphism

The observed genotypic frequencies for NQO1 C609T werefound to be significantly different from the expected ones in allgroups of controls and patients. For NQO1 C609T wild-type,heterozygous, and homozygous variant genotypes, the ob-served frequencies versus expected frequencies were 45, 131,and 24 versus 61.05, 98.9, and 40.05, respectively, in total cases(w2: 21.07; p-value <0.001) and 34, 149, and 17 versus 58.86,99.28, and 41.86, respectively, in total controls (w2: 50.17;p-value <0.001). Similarly, the observed frequencies versusexpected frequencies were 18, 61, and 11 versus 26.14, 44.73,and 19.14 for premenopausal cases (w2: 11.91; p-value <0.001)and 22, 88, and 7 versus 37.23, 57.54, and 22.23 for premeno-pausal controls (w2: 32.79; p-value <0.001). The observed fre-quencies versus expected frequencies were 27, 70, and 13versus 34.95, 54.11, and 20.95 for postmenopausal cases (w2:9.49; p-value<0.005) and 12, 61, and 10 versus 21.76, 41.48, and19.76 for postmenopausal controls (w2: 18.39; p-value <0.001).

NQO1 mRNA expression

No significant change in mRNA expression was observedin controls and patients carrying different NQO1 genotypes(Fig. 2).

FIG. 1. A representative diagram of polymerase chain re-action amplicons and their restriction fragment lengthpolymorphism byproducts of NQO1 gene in polyacrylamidegel. Lanes 1, 2, 3, and 4 indicate polymerase chain reactionamplicon, wild-type C/C genotype, homozygous variantT/T genotype, and heterozygous C/T genotype. NQO1,NAD(P)H:quinone oxidoreductase 1.

XENOBIOTIC-METABOLIZING GENES AND BREAST CANCER 345

NQO1 catalytic activity

The NQO1 homozygous variant (T/T) showed almostnegligible activity in controls and patients in comparison withthe wild-type C/C genotype (Fig. 3).

Discussion

Several enzymes are known to participate in the metabo-lism of estrogens. SNPs in NQO1 and CYP1A2 were studied,as these are the more commonly studied genes in manypopulations. The hydroxylation of estrogens into water-soluble metabolites is an important elimination step. Catecholderivatives of estrogens are inactivated by catechol-O-methyltransferase (COMT). Because of incomplete inactiva-tion, 3,4-semiquinone gets synthesized and reacts with DNAto form adducts that cause DNA damage. If it is not repaired,it may lead to mutation and alters the susceptibility to breastcancer (Cavalieri et al., 1997, 2000; Zhu and Conney, 1998b).SNPs in these genes may serve as biomarkers for breast cancerrisk, as they can account for altered metabolic pathways.

NQO1 functions as a two-electron donor and protects DNAfrom free radicals and quinones by reducing quinones tocatechol derivatives. Catechol derivatives are inactivated byO-methylation, catalyzed by COMT, and might be associated

Table 3. Allele and Genotype Frequencies of NAD(P)H:Quinone Oxidoreductase 1 C609Tin Controls and Patients

Patients Case controls Odds ratio (95% CI) p-Value

Total women n¼ 200 n¼ 200Allele frequency (total number

of alleles)C 0.55 (221) 0.54 (217)T 0.45 (179) 0.46 (183)

Genotypic frequency (total numberof genotypes)C/C 0.22 (45) 0.17 (34) 1.0 (reference) –C/T 0.66 (131) 0.75 (149) 0.66 (0.39–1.13) 0.141T/T 0.12 (24) 0.08 (17) 1.07 (0.46–2.46) 0.976C/T þ T/T 0.78 (155) 0.83 (166) 0.71 (0.42–1.19) 0.209

Total premenopausal women n¼ 90 n¼ 117Allele frequency (total number

of alleles)C 0.54 (97) 0.56 (132)T 0.46 (83) 0.44 (102)

Genotypic frequency (total numberof genotypes)C/C 0.20 (18) 0.19 (22) 1.0 (reference) –C/T 0.68 (61) 0.75 (88) 0.85 (0.40–1.81) 0.778T/T 0.12 (11) 0.06 (07) 1.92 (0.54–6.97) 0.394C/T þ T/T 0.80 (72) 0.81 (95) 0.93 (0.44–1.96) 0.969

Total postmenopausal women n¼ 110 n¼ 83Allele frequency (total number

of alleles)C 0.56 (124) 0.51 (85)T 0.44 (96) 0.49 (81)

Genotypic frequency (total numberof genotypes)C/C 0.25 (27) 0.14 (12) 1.0 (reference) –C/T 0.63 (70) 0.74 (61) 0.51 (0.22–1.16) 0.117T/T 0.12 (13) 0.12 (10) 0.58 (0.17–1.92) 0.461C/T þ T/T 0.75 (83) 0.86 (71) 0.52 (0.23–1.16) 0.121

CI, confidence interval.

Table 4. Polymorphism in NQO1 Gene

and Clinical Demographics of Breast

Cancer Patients in North Indian Women

NQO1 C609T genotype

C/C C/T T/T C/TþT/T

Stage atdiagnosis, n (%)�Stage I 3 (06.67) 5 (03.81) 0 5 (03.23)Stage II 22 (48.89) 60 (45.80) 20 (83.33) 80 (51.61)�Stage III 20 (44.44) 66 (50.38) 4 (16.67) 70 (45.16)

Lymph nodestatus, n (%)þ 29 (64.44) 89 (67.94) 13 (54.17) 102 (65.81)� 16 (35.56) 42 (32.06) 11 (45.83) 53 (34.19)

Tumorhistology, n (%)Ductal 43 (95.56) 124 (94.66) 20 (83.33) 144 (92.90)Lobular 2 (04.44) 3 (02.29) 2 (08.33) 5 (03.23)Other 0 4 (03.05) 2 (08.33) 6 (03.87)

346 SINGH ET AL.

with cancer predisposition (Cavalieri et al., 1997; Rauth et al.,1997; Zhu and Conney, 1998b). The C/C genotype of C609Tpolymorphism shows total activity, whereas heterozygousand homozygous variants show decreased and null activity,respectively (Moran et al., 1999). A higher but statisticallyinsignificant change in the frequency of homozygous variantgenotype in premenopausal patients when compared withrespective controls was observed in the present study. Thefrequency of heterozygous variant genotype showed lack ofsignificant difference in cases and controls. NQO1 protectsagainst estrogen-mediated carcinogenesis, and loss of NQO1heterozygosity has been observed in ductal breast cancerpatients (Rebbeck et al., 1996). NQO1 C609T polymorphism

was found to be not associated with postmenopausal Amer-ican, Chinese, and Japanese women as in Indian women(Hamajima et al., 2002; Fowke et al., 2004; Hong et al., 2007);however, significant associations of this polymorphism withbreast cancer risk in Caucasian populations have been re-ported (Menzel et al., 2004; Sarmanova et al., 2004). The fre-quency of the 609T allele was found to be lower in Indianpopulation as observed in Caucasians but not in Japanese,Korean, and Chinese populations (Hamajima et al., 2002).Probably, NQO1 alone does not play a critical role in breastcancer susceptibility in the Indian population. Although theNQO1 C609T variant leads to a reduced activity of NQO1,it does not show a significant effect on the levels of DNA

FIG. 2. Differential expression of NQO1mRNA in the white blood cells of case controlsand breast cancer patients having different ge-notypes of NQO1 C609T polymorphism (A). Bardiagram showing differential expression ofNQO1 mRNA in the white blood cells of casecontrols and patients having different genotypesof NQO1 C609T polymorphism. The data areexpressed in means� standard error of themeans (B). Lanes 1–3 represent case controlsand lanes 4–6 represent patients.

FIG. 3. NQO1 catalytic activity in thewhite blood cells of case controls and breastcancer patients having different genotypesof NQO1 C609T polymorphism. The dataare expressed in means� standard errorof the means and significant changes(**p< 0.01) are shown compared with C/Cgenotype. Lanes 1–3 represent case controlsand lanes 4–6 represent patients.

XENOBIOTIC-METABOLIZING GENES AND BREAST CANCER 347

adducts in breast tissue of NQO1 C609T wild-type, hetero-zygote, or homozygote variant genotype carriers, suggestingthat DNA adduct formation leading to carcinogenesis is notlinked with NQO1 and does not support the direct involve-ment of the NQO1 C609T genotype in breast cancer onset(Brockstedt et al., 2002).

In the present study, the association between CYP1A2C63G polymorphism and breast cancer incidence to assess thecontribution of CYP1A2 C63G polymorphism in Indian wo-men was investigated. All cases and controls exhibited similargenotype and were monomorphic. CYP1A2 is one of the mostimportant enzymes involved in 2-hydroxylation of estrogens(Zhu and Conney, 1998a). CYP1A2 is an inducible enzymeand its activity varies with sex, age, race, smoking status,consumption of coffee and alcohol, and exposure to variouspollutants and contaminants (Le Marchand et al., 1997). InIndian women, the relationship between smoking, coffee oralcohol consumption, and breast cancer risk with this genewas not investigated, as only 2–3 women were smokers andcoffee/alcohol drinkers and, most importantly, all cases andcontrols were monomorphic. Although CYP1A2 polymor-phism was found to modify breast cancer susceptibility inThai women (Sangrajrang et al., 2009), it was not found as amajor determinant of breast cancer susceptibility in NorthIndian women. The presence of the monomorphic form ofCYP1A2 C63G genotype in Indian women clearly showedthat it has nothing to do with breast cancer susceptibility inthe North Indian population, as reported in some populations(Le Marchand et al., 1997). CYP1A2 and NQO1 participate inestrogen metabolism at entry and final levels, respectively,and NQO1 protects DNA from reactive oxygen speciesand quinones by reducing quinones to catechol estrogens,which are inactivated by O-methylation catalyzed by COMT(Cavalieri et al., 1997, 2000; Rauth et al., 1997; Zhu and Con-ney, 1998b; Liehr, 2000; Mitrunen and Hirvonen, 2003). Inpostmenopausal women, mostly circulating estrogens arederived from conversion of androgen to estrogen in the breastand the metabolism is mainly catalyzed by CYP1A2 (Mi-trunen and Hirvonen, 2003). Exposure to chemicals inducesNQO1, which stabilizes and transiently activates p53-inde-pendent MDM2-mediated ubiquitination and protects thecells from adverse effects of stress conditions (Gong et al.,2007). In the present study, cases and controls were found tocarry the same CYP1A2 genotype and were monomorphic,showing that the studied CYP1A2 polymorphism is not as-sociated with breast cancer susceptibility in North Indianwomen. Despite the established roles of NQO1 C609T andCYP1A2 C63G polymorphisms in several populations, theseare unlikely to play any critical role in determining the breastcancer susceptibility in North Indian women.

Acknowledgments

The authors sincerely thank the University Grant Com-mission, New Delhi, and the Council of Scientific and In-dustrial Research, New Delhi, for providing researchfellowships to Virendra Singh and Ghanshyam Upadhyay,respectively. The IITR communication number of this article is2724.

Author Disclosure Statement

The authors report that there are no conflicts of interest.

References

Bradford MM (1976) A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal Biochem 72:248–254.

Bradlow HL, Sepkovic DW (2004) Steroids as procarcinogenicagents. Ann NY Acad Sci 1028:216–232.

Brockstedt U, Krajinovic M, Richer C, et al. (2002) Analyses ofbulky DNA adduct levels in human breast tissue and geneticpolymorphisms of cytochromes P450 (CYPs), myeloperox-idase (MPO), quinone oxidoreductase (NQO1), and glutathi-one S-transferases (GSTs). Mutat Res 516:41–47.

Cavalieri E, Frenkel K, Liehr JG, et al. (2000) Estrogens as en-dogenous genotoxic agents-DNA adducts and mutations. JNatl Cancer Inst Monogr 27:75–93.

Cavalieri EL, Stack DE, Devanesan PD, et al. (1997) Molecularorigin of cancer: catechol estrogen-3,4-quinones as endogenoustumor initiators. Proc Natl Acad Sci USA 94:10937–10942.

Fowke JH, Shu XO, Dai Q, et al. (2004) Oral contraceptive useand breast cancer risk: modification by NAD(P)H:quinoneoxoreductase (NQO1) genetic polymorphisms. Cancer Epide-miol Biomarkers Prev 13:1308–1315.

Gong X, Kole L, Iskander K, et al. (2007) NRH:quinone oxido-reductase 2 and NAD(P)H:quinone oxidoreductase 1 protecttumor suppressor p53 against 20s proteasomal degradationleading to stabilization and activation of p53. Cancer Res67:5380–5388.

Hamajima N, Matsuo K, Iwata H, et al. (2002) NAD(P)H: qui-none oxidoreductase 1 (NQO1) C609T polymorphism and therisk of eight cancers for Japanese. Int J Clin Oncol 7:103–108.

Hayes CL, Spink DC, Spink BC, et al. (1996) 17-beta-estradiolhydroxylation catalyzed by human cytochrome P450 1B1. ProcNatl Acad Sci USA 93:9776–9781.

Hong CC, Ambrosone CB, Ahn J, et al. (2007) Genetic variabilityin iron-related oxidative stress pathways (Nrf2, NQ01, NOS3,and HO-1), iron intake, and risk of postmenopausal breastcancer. Cancer Epidemiol Biomarkers Prev 16:1784–1794.

Huang JD, Guo WC, Lai MD, et al. (1999) Detection of a novelcytochrome P-450 1A2 polymorphism (F21L) in Chinese. DrugMetab Dispos 27:98–101.

Kemp TJ, Causton HC, Clerk A (2003) Changes in gene ex-pression induced by H2O2 in cardiac myocytes. BiochemBiophys Res Commun 307:416–421.

Le Marchand L, Donlon T, Kolonel LN, et al. (2005) Estrogenmetabolism-related genes and breast cancer risk: the multi-ethnic cohort study. Cancer Epidemiol Biomarkers Prev 14:1998–2003.

Le Marchand L, Franke AA, Custer L, et al. (1997) Lifestyle andnutritional correlates of cytochrome CYP1A2 activity: inverseassociations with plasma lutein and alpha-tocopherol. Phar-macogenetics 7:11–19.

Lee YY, Westphal AH, de Haan LH, et al. (2005) HumanNAD(P)H:quinone oxidoreductase inhibition by flavonoids inliving cells. Free Radic Biol Med 39:257–265.

Liehr JG (2000) Is estradiol a genotoxic mutagenic carcinogen?Endocr Rev 21:40–54.

Long JR, Cai Q, Shu XO, et al. (2007) Genetic polymorphisms inestrogen-metabolizing genes and breast cancer survival.Pharmacogenet Genomics 17:331–338.

Long JR, Egan KM, Dunning L, et al. (2006) Population-basedcase-control study of AhR (aryl hydrocarbon receptor) andCYP1A2 polymorphisms and breast cancer risk. Pharmaco-genet Genomics 16:237–243.

Lyn-Cook BD, Yan-Sanders Y, Moore S, et al. (2006) Increasedlevels of NAD(P)H: quinone oxidoreductase 1 (NQO1) in

348 SINGH ET AL.

pancreatic tissues from smokers and pancreatic adenocarci-nomas: a potential biomarker of early damage in the pancreas.Cell Biol Toxicol 22:73–80.

Menzel HJ, Sarmanova J, Soucek P, et al. (2004) Association ofNQO1 polymorphism with spontaneous breast cancer in twoindependent populations. Br J Cancer 90:1989–1994.

Miller SA, Dykes DD, Polesky HF (1998) A simple salting outprocedure for extracting DNA from human nucleated cells.Nucleic Acids Res 16:1215.

Mitrunen K, Hirvonen A (2003) Molecular epidemiology ofsporadic breast cancer. The role of polymorphic genes in-volved in oestrogen biosynthesis and metabolism. Mutat Res544:9–41.

Moran JL, Siegel D, Ross D (1999) A potential mechanism un-derlying the increased susceptibility of individuals with apolymorphism in NAD(P)H:quinone oxidoreductase 1 (NQO1)to benzene toxicity. Proc Natl Acad Sci USA 96:8150–8155.

Rauth AM, Goldberg Z, Misra V (1997) DT-diaphorase: possibleroles in cancer chemotherapy and carcinogenesis. Oncol Res9:339–349.

Rebbeck TR, Godwin AK, Buetow KH (1996) Variability in lossof constitutional heterozygosity across loci and among indi-viduals: association with candidate genes in ductal breastcarcinoma. Mol Carcinog 17:117–125.

Sachse C, Bhambra U, Smith G, et al. (2003) Polymorphisms inthe cytochrome P450 CYP1A2 gene (CYP1A2) in colorectalcancer patients and controls: allele frequencies, linkage dis-equilibrium and influence on caffeine metabolism. Br J ClinPharmacol 55:68–76.

Sangrajrang S, Sato Y, Sakamoto H, et al. (2009) Genetic poly-morphisms of estrogen metabolizing enzyme and breast can-cer risk in Thai women. Int J Cancer 125:837–843.

Sarmanova J, Susova S, Gut I, et al. (2004) Breast cancer: role ofpolymorphisms in biotransformation enzymes. Eur J HumGenet 12:848–854.

Singh V, Rastogi N, Mathur N, et al. (2008) Association ofpolymorphism in MDM-2 and p53 genes with breast cancerrisk in Indian women. Ann Epidemiol 18:48–57.

Singh V, Rastogi N, Sinha A, et al. (2007) A study on the asso-ciation of cytochrome-P450 1A1 polymorphism and breastcancer risk in north Indian women. Breast Cancer Res Treat101:73–81.

Tsuchiya Y, Nakajima M, Yokoi T (2005) Cytochrome P450-mediated metabolism of estrogens and its regulation in hu-man. Cancer Lett 227:115–124.

Yadav S, Singhal NK, Singh V, et al. (2009) Association of singlenucleotide polymorphisms in CYP1B1 and COMT genes withbreast cancer susceptibility in Indian women. Dis Markers27:203–210.

Zhang J, Schulz WA, Li Y, et al. (2003) Association of NAD(P)H:quinone oxidoreductase 1 (NQO1) C609T polymorphism withesophageal squamous cell carcinoma in a German Caucasianand a northern Chinese population. Carcinogenesis 24:905–909.

Zhu BT, Conney AH (1998a) Functional role of estrogen me-tabolism in target cells: review and perspectives. Carcino-genesis 19:1–27.

Zhu BT, Conney AH (1998b) Is 2-methoxyestradiol an endoge-nous estrogen metabolite that inhibits mammary carcinogen-esis? Cancer Res 58:2269–2277.

Address correspondence to:Mahendra Pratap Singh, M.Sc., Ph.D.Indian Institute of Toxicology Research

(Council of Scientific and Industrial Research)Mahatma Gandhi Marg

Post Box-80Lucknow 226 001

India

E-mail: singhmahendrapratap@rediffmail.com

XENOBIOTIC-METABOLIZING GENES AND BREAST CANCER 349