Association of Xenobiotic Metabolizing Enzymes Genetic Polymorphisms With Esophageal Cancer in Kashmir Valley and Influence of Environmental Factors

June 10, 2010 19:14 801xml HNUC_A_461099

HNUC #461099, VOL 62, ISS 6

Association of Xenobiotic Metabolizing Enzymes GeneticPolymorphisms With Esophageal Cancer in Kashmir Valleyand Influence of Environmental Factors

Manzoor Ahmad Malik, Rohit Upadhyay, Rama Devi Mittal, Showket Ali Zargar,and Balraj Mittal

QUERY SHEETThis page lists questions we have about your paper. The numbers displayed at left can be found in the text of the paper for reference.In addition, please review your paper as a whole for correctness.

Q1. Au: Please verify the definitions of GST and CYP or correct.Q2. Au: Please note that gene names are not italicized in this journal.Q3. Au: PCR-RFLP was defined and the abbreviation deleted here, as it is not used again in the abstract text (journal style).Q4. Au: I found the Fermentas U.S. location to be in Glen Burnie, MD. Please advise.Q5. Au: The figure must be called out (mentioned) in the article text. Please indicate the appropriate place to do so its a dummy

citation.Q6. Au: 1997 was listed in the beginning of Ref. 7; please indicate the correct year.Q7. Au: Please give only one year for Ref. 33.

TABLE OF CONTENTS LISTING

The table of contents for the journal will list your paper exactly as it appears below:

Association of Xenobiotic Metabolizing Enzymes Genetic Polymorphisms With Esophageal Cancer in Kashmir Valleyand Influence of Environmental FactorsManzoor Ahmad Malik, Rohit Upadhyay, Rama Devi Mittal, Showket Ali Zargar, and Balraj Mittal

June 10, 2010 19:14 801xml HNUC_A_461099

Nutrition and Cancer, 62(6), 1–9Copyright C© 2010, Taylor & Francis Group, LLCISSN: 0163-5581 print / 1532-7914 onlineDOI: 10.1080/01635581003605904

Association of Xenobiotic Metabolizing Enzymes GeneticPolymorphisms With Esophageal Cancer in Kashmir Valleyand Influence of Environmental Factors

Manzoor Ahmad Malik, Rohit Upadhyay, and Rama Devi MittalSanjay Gandhi Post-Graduate Institute of Medical Sciences, Lucknow, India5

Showket Ali ZargarSher-i-Kashmir Institute of Medical Sciences, Kashmir, India

Balraj MittalSanjay Gandhi Post-Graduate Institute of Medical Sciences, Lucknow, India

10

The Kashmir Valley has an elevated incidence rate of esophagealcancer (EC). Several environmental and genetic factors have beensuspected for development of EC. A case-control study was per-formed in 135 EC patients and 195 healthy controls to analyzeassociation of polymorphisms in glutathione S-transferase (GST)15mu (GSTM1), GST theta (GSTT1), GST pi (GSTP1), GSTM3, cy-tochrome P (CYP)1A1, and CYP2E1 genes with susceptibility toQ1EC as well as their interaction with environmental factors suchas smoking and high consumption of salted tea in Kashmir valley.Q2All subjects were genotyped through polymerase chain reaction20restriction fragment length polymorphism. Data was statisticallyQ3analyzed using the chi-square test and logistic regression model.Results showed that GSTP1313 valine (val)/val and CYP2E1c1c2genotypes imparted risk for esophageal squamous cell carcinoma(ESCC) and esophageal adenocarcinoma [EADC; odds ratio (OR)25= 3.24, 95% confidence interval (CI) = 1.30–8.05; OR = 4.20,95% CI = 1.65–10.70], respectively. GSTM3AB genotype/B allelewas found to be associated with low risk for EC. Tobacco smokingthrough Hukka (water pipe) and consumption of salted tea itselfwere high risk factors for developing EC (OR = 21.44, 95% CI =3011.63–39.54; OR = 14.86, 95% CI = 8.41–26.24), and the risks weremodulated through the interaction of GSTM3AB, GSTP1val/valgenotypes. In conclusion, GSTP1val/val and CYP2E1c1c2 geno-types/c2 allele increased the risk of ESCC and EADC, respec-tively, in the Kashmiri population; whereas GSTM3AB genotype35imparted lower risk for both ESCC and EADC.

Submitted 14 March 2009; accepted in final form 9 December 2009.Address correspondence to Balraj Mittal, Department of Ge-

netics, Sanjay Gandhi Post-Graduate Institute of Medical Sciences,Raebareilly Road, Lucknow-226014, India+91-522-2668973, 004-8,Ext. 2322 +91-522-2668017, 2668074. E-mail: [email protected];bml [email protected]

INTRODUCTION

Among human cancers, esophageal carcinogenesis appearsto be a complex multistep process with multifunctional etiolo-gies in which environmental, geographical, and genetic factors 40have been attributed to play major roles in causation of cancers(1). Esophageal cancer (EC) is the eighth most commonly occur-ring cancer in the world (2). In the Kashmir Valley, northern partof India, EC has been reported to exceed 40% of all cancers, andincidence is 3 to 6 times higher than various metropolis cancer 45registries in India. However, very few reports have associatedthis malignancy with specific risk factors prevalent in the area(3). Some of the environmental factors including dietary habitssuch as hot noon chai (salted tea), smoking, use of pickled andpreserved foods rich in salt, nitrate, and preformed nitrosocom- 50pounds have been associated with an increased risk of EC in theKashmir Valley (4). A recent study from northern Iran, whichis a part of the EC belt, also reported a large intake of hot tea asa risk factor for EC (5).

Individual susceptibility to cancer may be partly due to ex- 55posure to environmental risk factors, which may be explainedby genetic variability in metabolic activities related to phase Iand phase II detoxification enzyme pathways. In phase I, cy-tochrome P (CYP) 450 enzymes represent a large multigenefamily with different substrate specificities that catalyze the 60first oxidative step in the metabolism of polycyclic aromatichydrocarbons (PAHs) such as those found in tobacco smoketo phenolic products and epoxides that are known carcinogens(6). The CYP-450 enzymes are induced by exposure to agentssuch as dioxin, benzo (α) pyrene, and other aromatic hydrocar- 65bons (7). Among phase II enzymes, glutathione S-transferases(GSTs) are a family of multifunctional cytosolic enzymes thatdetoxify reactive chemical species such as PAH epoxides by cat-alyzing their conjugation to glutathione, to make them less toxic,

1

June 10, 2010 19:14 801xml HNUC_A_461099

2 M. A. MALIK ET AL.

thus forming a protective mechanism against chemical carcino-70genesis. GST gene family encoding GST mu (GSTM1), GSTtheta (GSTT1), and GST pi (GSTP1) are polymorphic. GSTM1and GSTT1 have null alleles that result in homozygous absenceof these genes in 10%–60% of different ethnic populations (8).Single nucleotide polymorphisms such as GSTP1313 isoleucine75(ile) > valine (val; rs1695), CYP1A16235T>C (rs4646903),and CYP2E1Rsa1–1091C>T (rs2031920) have been reportedto alter the expression of their respective genes, which furthermodulate the enzyme activities (9–11). GSTM3 3bp deletionin intron 6 (rs1799735) can both repress and enhance its ex-80pression (12). In several case-control studies, polymorphismsin genes encoding phase I and phase II xenobiotic metabolizingenzymes have been linked to modulation of risk for EC (13–15).However, to date, no study has been carried out to evaluate therole of these polymorphisms in relation to higher prevalence of85EC in the Kashmir Valley. Therefore, the aim of this study wasto access the influence of genetic polymorphisms in CYP1A1,CYP2E1, GSTM1, GSTT1, GSTP1, and GSTM3 genes andtheir interaction with environmental factors such as smokingand salted tea intake in conferring susceptibility to EC in the90Kashmir Valley.

MATERIALS AND METHODSThis study comprised untreated histopathologically con-

firmed cases with EC (135) and healthy controls (195) from the95Kashmir Valley. All subjects were unrelated ethnic Kashmiriresidents, referred from the Departments of Gastroenterology,Sher-i-Kashmir Institute of Medical Sciences, Srinagar, fromMay 2006 to August 2008. Patients were excluded if they hadnonmalignant conditions such as corrosive esophageal injury,100Achalasia injury, Barrett’s esophagus, gastro-esophageal refluxdisease (GERD), and nonulcer dyspepsia, but were included ifthey had histopathologically proven malignancy of esophagus.At the same time, age- and gender-matched individuals whocame for their routine health checkup and medical staff at Sher-105i-Kashmir Institute of Medical Sciences, Srinagar, were selectedas controls. Selection criterion for controls included no evidenceof malignancy as well as no prior family history of cancer. All theindividuals were personally interviewed for their age, occupa-tion, demographic features, usage of hot noon chai (salted tea),110drinking alcohol, and smoking (Hukka) habits. Tobacco usagemainly included smoking of Hukka (water pipe). Hukka smok-ing is a type of tobacco smoking in which tobacco smoke passesthrough a water pipe, and it is very common in the Kashmiri pop-ulation. The hot salted tea taken by Kashmiri people is prepared115by using baking soda (sodium bicarbonate) and common salt(sodium chloride) and boiled for few hours before consuming.Written informed consent was obtained from all participants inthe study. The research protocol was approved by the ethicalcommittee of Sanjay Gandhi Postgraduate Institute of Medical120Sciences, Lucknow (28th Ethics Committee Meeting on May 14,2004). Sample collection, storage, and transport were complied

with guidelines of the committee. Blood samples were collectedin EDTA and the genomic DNA was extracted from peripheralblood leukocytes pellet using the standard salting-out method 125(16). The quality and quantity of DNA was checked by gel elec-trophoresis and spectrophotometry using Nanodrop Analyser(ND-1000) spectrophotometer (Nano Drop Technologies, Inc.,Wilmington, DE). The ratio of absorbance at 260 and 280 nmof DNA was around 1.7 to 1.9. The isolated DNA was stored at 130–70◦C.

Genotype AnalysisNull alleles of GSTM1 and GSTT1 were determined by using

the multiplex polymerase chain reaction (PCR) with CYP1A1gene as an internal positive control (17). Briefly, a 215 bp region 135between exon 4 and 5 of the GSTM1 gene and 480 bp productsfor GSTT1 were amplified along with a 312 bp size productof CYP1A1. The PCR products were electrophoresed on a 2%agarose gel. The absence of 480 and 215 bp bands indicated ho-mozygous null genotypes of GSTM1 and GSTT1, respectively. 140In GSTM3, a 3 bp deletion polymorphism is present in intron 6.Primer sequences were used as described earlier by Loktionovet al. (18). Allele A and B of the GSTM3 gene showed 79 bp and76 bp PCR products, respectively, and were separated on 20%polyacrylamide gel. A GSTP1 polymorphism is present in exon 1455 at nucleotide 313 due to transition of base G>A (19). The PCRproduct of 176 bp was digested with 5 U of Alw261 restrictionenzyme (Fermentas, Inc., MA) at 37◦C in total volume of 10 µl, Q4

and the products were visualized on 15% polyacrylamide gelafter ethidium bromide staining. The genotypes were classified 150as ile/ile (176 bp), ile/val (176 bp, 91 bp, and 85 bp), and val/val(91 bp and 85 bp).

CYP1A16235 T>C polymorphism involves substitution ofCTGG to CCGG in the Msp1 site at base 264 from the additionalpolyadenylation signal in the 3’ flanking region. The region of 155interest was amplified by PCR using primer sequences describedby Kawajiri et al. (20) CYP1A1 T and C alleles were determinedby presence or absence of Msp1 restriction site through differ-ent band patterns on 2% agarose gel. The wild-type genotype(CYP1A1TT) showed a single band of 360 bp. The variant geno- 160type (CYP1A1CC) resulted in two fragments of 220 and 140 bp,whereas the heterozygous genotype (CYP1A1TC) showed threebands of 360, 220, and 140 bp. Genotyping for CYP2E1 5’UTRRsa1–1091C>T polymorphism was detected using PCR, RFLP(21). After digestion of 410 bp PCR product by RsaI restriction 165enzyme (Fermentas, Inc.), c1c1 genotype was represented bytwo bands (360 bp and 50 bp), c1c2 genotype produced threebands of 410 bp, 360 bp, and 50 bp, whereas c2c2 genotypeshowed a single band of 410 bp.

Ten percent of samples from patients and controls including 170samples of each genotype were re-genotyped by other labora-tory personnel. No discrepancy was also found after sequencingrandomly selected 5% samples.

June 10, 2010 19:14 801xml HNUC_A_461099

XENOBIOTIC METABOLIZING ENZYMES GENETIC POLYMORPHISMS WITH ESOPHAGEAL CANCER 3

TABLE 1Characteristics of Esophageal cancer patients and healthy individuals of the Kashmir Valleya

Characteristic EC Patientsb Healthy IndividualscOR (95% CI),

P (EC vs. Control)

Mean age ± SD 60.38 yr ± 8.402 57.98 yr ± 12.664 P > 0.05Sex

Male 92 (68.1%) 139 (71.3%) P > 0.05Female 43 (31.9%) 56 (28.7%) P > 0.05

HistologyAdenocarcinoma 32 (23.7%)Squamous cell

carcinoma103 (76.3%)

Smokingd

Smokers (Hukka) 106 (84.1%) 38 (20.5%) 21.44 (11.63–39.54),0.0001

Salted tea intaked

≤4 cups daily) 36 (28.6%) 159 (85.9%) 14.86 (8.41–26.24),0.0001

>4 cups daily) 90 (71.4%) 26 (14.1%)Alcohol consumption Nil Nil

aAbbreviations are as follows: EC, esophageal cancer; OR, odds ratio; CI, confidence interval. OR is age and genderadjusted. Significant values shown in bold.

bn = 135.cn = 195.dData missing.

Statistical AnalysisDescriptive statistics of patients and controls were presented175

as mean and SDs for continuous measures, whereas frequenciesand percentages were used for categorical measures. Effectivesample sizes were calculated by Quanto software version 1.2(22). The chi-square (χ2) goodness-of-fit test was used for anydeviation from Hardy Weinberg Equilibrium. Statistical signif-180icance of differences in genotype frequencies between patientsand controls was estimated by the χ2 test. Gene–environmentinteractions were examined between genotypes and environ-mental risk factors in all esophageal cancer cases as well asin controls. Binary logistic regression was used for all anal-185ysis variables to estimate risk as odds ratio (OR) with 95%confidence intervals (CIs) using age and sex as covariates. Allstatistical analyses were performed using SPSS software ver-sion 15.0 (SPSS, Chicago, IL), tests of statistical significancewere 2-sided, and differences were taken as significant when190the P value was less than 0.05. All ORs were adjusted for ageand gender.

RESULTSThe mean age of controls and patients with EC was 57.98

yr ± 12.66 and 60.38 yr ± 8.40, respectively (t-test P value195= NS). Most of the patients that suffered from EC were male(68.1%) as compared to female (31.9%). Histopathologically,the malignancies were diagnosed as esophageal squamous cell

carcinoma (ESCC) and esophageal adenocarcinoma (EADC). Inpatients with EC, ESCC histopathology was common (76.3%) as 200compared to EADC (23.7%). Most of the patients were smok-ers (106/135; 84.3%), and all were hukka smokers. Smokinghabit (hukka) showed significantly higher risk in EC (OR =21.44, 95% CI = 11.63–39.54; P = 0.0001) patients. Individ-uals consumed salted tea in a range of 2 to 8 cups/day; and 205median consumption of tea was 4 cups/day. Thus we groupedindividuals in to ≤4 cups/day or >4 cups/day; and individualswho consumed salted tea >4 cups/day were regarded as highsalted tea consumers. Higher consumption of salted tea was alsofound to be associated with increased risk of EC (OR = 14.86, 21095% CI = 8.41–26.24; P value = 0.0001; Table 1). None ofpatients or controls reported consumption of alcohol, so inter-action of alcohol intake with genetic variations could not beanalyzed. Sample size calculations revealed that inclusion of135 patients was sufficient for getting 80% power for 4 out of 6 215polymorphisms (GSTM1, GSTP1313ile>val, GSTM3intron 63bp deletion, and CYP1A1 6235T>C) taken in the study.

Association of Genetic Variations With Susceptibilityto EC

Individuals with GSTP1313val/val, CYP2E1c1c2 genotype 220were found to be imparted more than twofold risk for EC (ORs= 2.48 and 2.29, respectively; P < 0.05). However, at allelelevel, only the CYP2E1c2 allele was found to be significantly

June 10, 2010 19:14 801xml HNUC_A_461099

4 M. A. MALIK ET AL.

TABLE 2Frequency distribution of CYP1A1, GSTM1, GSTT1, GSTP1, GSTM3, and CYP2E1 and risk assessment in esophageal

cancer patients and controla

Controlb Esophageal Cancerc

Genotype n % n % OR (95% CI), P

GSTM1present 116 58.5 67 49.6 1 (Reference)GSTM1null 79 41.5 68 50.4 1.50 (0.96–2.35), 0.075GSTT1present 146 74.9 110 81.5 1 (Reference)GSTT1null 49 25.1 25 18.5 0.68 (0.39–1.17), 0.164

GSTP1313ile > val (rs1695)ile/ile 111 56.9 72 53.3 1 (Reference)ile/val 75 38.5 48 35.6 1.00 (0.62–1.6), 0.991val/val 9 4.6 15 11.1 2.48 (1.03–6.02), 0.044ile 297 76.2 192 71.1 1 (Reference)val 93 23.8 78 28.9 1.29 (0.91–1.84), 0.158

GSTM3 intron 6 3 bp deletion (rs1799735)AA 147 75.4 120 88.9 1 (Reference)AB 45 23.1 13 9.6 0.35 (0.18–0.68), 0.002BB 3 1.5 2 1.5 0.77 (0.73–4.71), 0.778A 339 86.9 253 93.7 1 (Reference)B 51 13.1 17 6.3 0.44 (0.25–0.78), 0.005

CYP1A16235T > C (rs4646903)TT 95 48.7 76 56.3 1 (Reference)TC 88 45.1 52 38.5 0.72 (0.45–1.14), 0.157CC 12 6.2 7 5.2 0.70 (0.26–1.87), 0.471T 278 71.3 204 75.6 1 (Reference)C 112 28.7 66 24.4 0.78 (0.55–1.12), 0.182

CYP2E1Rsa1–1091C > T (rs2031920)c1c1 177 90.8 109 80.7 1 (Reference)c1c2 17 8.7 25 18.5 2.29 (1.18–4.48), 0.015c2c2 1 0.5 1 0.7 1.42 (0.09–23.00), 0.806c1 371 95.1 243 90.0 1 (Reference)c2 19 4.9 27 10.0 2.06 (1.11–3.80), 0.021

aAbbreviations are as follows: CYP, cytochrome P; GST, glutathione S-transferase; OR, odds ratio; CI, confidence interval; GSTM, GST mu;GSTT, GST tau; GSTP, GST pi; bp, base pair; ile, isoleucine; val, valine. OR is age and gender adjusted. Significant values shown in bold.

bN = 195.cN = 135.

associated with EC (OR = 2.06, 95% CI = 1.11–3.80; P =0.021; Table 2).225

GSTM3A/B genotype and GSTM3B allele were significantlyassociated with low risk in EC, respectively (OR = 0.35, P =0.002; OR = 0.44, P = 0.002), but variants of GSTM1, GSTT1,and CYP1A16235T > C gene polymorphisms were not foundto be associated with EC development (Table 2).230

Association of Genotypes With Tumor HistopathologyWhen tumor histopathologies were analyzed, GSTP1val/val

genotype was found to be significantly associated with morethan threefold risk for ESCC. The GSTM3A/B genotype was

significantly associated with reduced risk for both ESCC and 235EADC (Table 3).

Interaction of Genotypes With Smoking Habit and HighSalted Tea Consumption

On analyzing the interaction of genotypes with tobacco habit,smokers with GSTM3AB genotype were found to be signifi- 240cantly at low risk for developing EC (OR = 0.48, 95% CI =0.25–0.90; P = 0.022; Table 4). Salted tea intake is commonamong Kashmiris, but patients with EC had higher intake ofthe salted tea (71.4% vs. 14.1%). The interaction of genotypeswith high salted tea intake showed increased risk for developing 245

June 10, 2010 19:14 801xml HNUC_A_461099

XENOBIOTIC METABOLIZING ENZYMES GENETIC POLYMORPHISMS WITH ESOPHAGEAL CANCER 5

TABLE 3Association of genotypes with tumor histopathology and risk of esophageal cancera

Genotype Control ESCC OR∗(95% CI), P Value EADC OR (95% CI), P Value

GSTP1313ile>val (rs1695)ile/ile 111 (56.9%) 53 (51.5%) 1 (Reference) 19 (59.4%) 1 (Reference)ile/val 75 (38.5%) 36 (35.0%) 1.03 (0.62–1.74), 0.903 12 (37.5%) 0.94 (0.43–2.05), 0.870val/val 9 (4.6%) 14 (13.6%) 3.24 (1.30–8.05), 0.011 1 (3.1%) 0.63 (0.07–5.30), 0.669

GSTM3 intron 6 3 bp deletion (rs1799735)AA 147 (75.4%) 89 (86.4%) 1 (Reference) 31 (96.9%) 1 (Reference)AB 45 (23.1%) 12 (11.7%) 0.43(0.21–0.86)0.017 1 (3.1%) 0.11 (0.01–0.80), 0.029BB 3 (1.5%) 2 (1.9%) 1.03 (0.17–6.34), 0.971 0 NC

CYP2E1Rsa1–1091C>T (rs2031920)c1c1 177 (90.8%) 76 (83.5%) 1 (Reference) 18 (72.0%) 1 (Reference)c1c2 17 (8.7%) 15 (16.5%) 1.81 (0.86–3.79), 0.119 6 (24.0%) 4.20 (1.65–10.70), 0.003c2c2 1 (0.5%) 0 NC 1 (4.0%) 7.56 (0.45–127.14), 0.160

aAbbreviations are as follows: ESCC, esophageal squamous cell carcinoma; OR, odds ratio; CI, confidence interval; EADC,esophageal adenocarcinoma; GSTP, glutathione S-transferase (GST) pi; ile, isoleucine; val, valine; GSTM, GST mu; bp, base pair;CYP, cytochrome P; NC, not calculated. OR is age and gender adjusted. Significant values shown in bold.

EC in GSTP1val/val genotype carriers (OR = 3.60, 95% CI =1.44–8.97; P = 0.006; Table 4).

DISCUSSIONEsophageal cancer is the most common gastrointestinal tract

malignancy in the Kashmir Valley (3). The wide geographical250variation in the incidence of esophageal cancer suggests thatthere are major genetic and environmental influences in the

development of this cancer. Xenobiotic metabolizing enzymessuch as GSTs and CYPs play important roles in metabolism ofcarcinogenic chemicals. These enzymes act as a cellular guard 255by detoxifying polycyclic aromatic hydrocarbon carcinogensafter glutathione conjugation. This is the first study in whichwe examined the link between polymorphisms of CYP1A1,GSTM1, GSTT1, GSTP1, GSTM3, and CYP2E1 with suscep-tibility to esophageal cancer and their interaction with environ- 260mental and dietary factors in the Kashmiri population.

TABLE 4Interaction of genotypes with smoking and/or high salted tea consumption in risk assessment of ECa

Smoking Habit (Smokers)b

Genotype Nonsmokerc Smokerd OR (95% CI), P

GSTM3 intron 6 3bp deletion (rs1799735)AA 130 (77.8%) 123 (85.4%) 1 (Reference)AB 36 (21.6%) 17 (11.8%) 0.48 (0.25–0.90), 0.022BB 1 (0.6%) 4 (2.8%) 3.91 (0.43–35.83), 0.228

High salted tea consumption (>4 cups daily)b

Genotype ≤4 cups dailye >4 cups dailyf OR (95% CI), PGSTP1313ile>val (rs1695)

ile/ile 112 (57.4%) 61 (52.6%) 1 (Reference)ile/val 75 (38.5%) 39 (33.6%) 0.95 (0.58–1.57), 0.851val/val 8 (4.1%) 16 (13.8%) 3.60 (1.44–8.97), 0.006

aAbbreviations are as follows: EC, esophageal cancer; OR, odds ratio; CI, confidence interval; GSTM, glutathione S-transferase (GST)mu; GSTP, GST pi. OR is age and gender adjusted. Significant values shown in bold.

bData missing in some subjects.cN = 155.dN = 133.eN = 181.fN = 107.

June 10, 2010 19:14 801xml HNUC_A_461099

6 M. A. MALIK ET AL.

TABLE 5Ethnic differences among various populations for xenobiotic metabolizing enzyme gene polymorphismsa

Minor Allele Frequencies in Controls

Population GSTM1null GSTT1null GSTP1val GSTM3B CYP1A1C CYP2E1c2 Reference

Indian (Kashmir;present study)

41.5% 25.1% 23.8% 13.1% 28.7% 4.9% —

Indian (exceptKashmir)

37.2% 27.0% 27.0% 10.0% 32%,35.5%

— (33–40)

Caucasian 50% 28% 33.2% 15%–24% 19% 1.6% (23,25,27,29,30,32)Asian (Chinese,

Koreans, Japanese)45% 60%–64% 9.1%–17.8% — 33%–41% 17%–26% (23,25,31,32,36)

African American 33% 22% 40.5% 68% 17.9% 2%–8% (25,28,30,32)European 52.3%–54.8% 13.9%–22.2% 37.0%–39.2% — 8%–11% 2.1%–5.9% (24–26,32,35)

aAbbreviations are as follows: GSTM, glutathione S-transferase (GST) mu; GSTT, GST tau; GSTP, GST pi; CYP, cytochrome P.

Ethnic differences of GSTM1null, GSTT1null,GSTP1313ile>val, GSTM3 intron6 3bp deletion,CYP1A16235T>C, and CYP2E1 Rsa1–1091C>T ge-netic polymorphisms among various populations have been265observed in various previous reports (Table 5) (23–32).Comparison of the Kashmir population with other Indianpopulations has shown minor differences in allelic frequenciesof polymorphisms studied (Table 5) (33–40). Our study showedno significant association between the GSTM1 and GSTT1270genotypes and the risk for esophageal cancer, which is inagreement with previous studies in esophageal cancer fromIndian (33) and other populations with different incidences ofcancer (24,25,26,40,41).

GSTP1313ile>val substitution creates the ile/val polymor-275phism at codon 105 that leads to expression of enzymes withreduced activity. The val/val genotype associated with low en-zyme activity has also been reported to confer high risk for blad-der, testicular, and prostate cancers (42). In this study, higherfrequency of GSTP1val/val genotype was observed among EC280patients, which is similar to other studies (24,43).

GSTM3 3 bp deletion in intron 6 generates a site for the bind-ing of transcription factor Yin Yang 1 (YY1). It is well knownthat YY1 transcription factor can both repress and activate tran-scription; however, molecular mechanisms controlling its be-285havior are complex and still not understood (9). There are sev-eral studies that have shown no association between GSTM3BBgenotype and various cancers (44,45). A previous study fromIndia (other than Kashmir) showed higher risk for developmentof EC (34). However, in our study, both GSTM3AB genotype290and GSTM3B allele showed reduced risk for the developmentof EC. Thus, GSTM3 3 bp deletion polymorphism seems toconfer inconsistent results among different populations.

We did not find any significant association between CYP1A1polymorphism with susceptibility of EC. Previous reports also295show no association of CYP1A16235T > C polymorphism withEC in Indian (39) and American populations (10,41). Thus, our

findings show that CYP1A16235T>C polymorphism is not asusceptibility marker for EC.

Our results suggest that CYP2E1c1/c2 genotype and c2 al- 300lele were significantly associated with EC. An increased riskof gastric cancer has also been reported for the CYP2E1c2/c2genotype in Taiwan and Japan (46). Marchand et al. (47) showedthat CYP2E1c2 genotype expresses reduced enzyme activity invivo compared with CYP2E1c1 genotype, and this observation 305further adds to the fact that c2 allele of CYP2E1Rsa1–1091C>Tpolymorphism may be a risk factor for carcinogenesis.

Previous reports have shown different etiologies and geneticrisk factors for two histological types of esophageal cancer. Ourstudy also shows a different pattern of susceptibility in ESCC 310and EADC histopathologies in the Kashmiri population. In EC,GSTM3AB genotype was found to be significantly associatedwith reduced risk for both ESCC and EADC. GSTP1val/valgenotype was found to confer increased risk of ESCC. Someother studies have shown positive association of GSTP1val/val 315with ESCC (43,48), whereas CYP2E1c1/c2 was associated withincreased susceptibility to EADC histopathology. The differen-tial pattern in the two histopathologies probably represents dif-ferent etiologies and genetic involvements in ESCC and EADC.

Several previous studies have attributed high incidence of 320EC in Kashmir to considerable amount of nitroso compoundsin raw foodstuffs and use of hot salted tea (3,49). It is sus-pected that the salts might cause thermal injury to esophagealepithelium (3). Salted tea used in the Kashmir Valley has con-siderable amounts of N-nitrosoproline (NPRO) (360 µg/kg) and 325N-nitrosopipecolic acid (NPIC) (5,870 µg/kg), which may im-part risk for EC in this area (50). Although high consumption ofsalted tea (>4 cups/day) was independently associated with in-creased high risk for EC, the interaction of genetic variations an-alyzed in this study showed that the cancer risk further increased 330only in GSTP1val/val genotype carriers. Our results show sig-nificant association of smoking (Hukka) with EC (OR = 21.44;P = 0.0001). However, when the interaction of smoking habit

June 10, 2010 19:14 801xml HNUC_A_461099

XENOBIOTIC METABOLIZING ENZYMES GENETIC POLYMORPHISMS WITH ESOPHAGEAL CANCER 7

FIG. 1. Representative gel picture of glutathione S-transferases (GSTs) and cytochrome Ps (CYPs) genotypes. a: Lane 1, GST mu (M)1 (GSTM1) present/GSTtau (T)1 (GSTT1) null; Lane 2, GSTM1 null/GSTT1 present; Lane 3, 50 bp ladder; Lane 4, GSTM1present/GSTT1present. b: Lane 1, 50 bp ladder; Lane 2,GSTP1313ile/ile; Lane 3, GSTP1313ile/val; Lane 4, GSTP1313val/val. c: Lane 1, 50 bp ladder; Lane 2, GSTM3AA; Lane 3, GSTM3AB; Lane 4, GSTM3BB. d:Lane 1, CYP1A16235TT; Lane 2, CYP1A16235TC; Lane 3, CYP1A16235CC; Lane 4, 100 bp ladder. e: Lane 1, 100 bp ladder; Lane 2, CYP2E1Rsa1–1091c1c2;Lane 3, CYP2E1Rsa1–1091c1c1; Lane 4, CYP2E1Rsa1–1091c2c2.

was analyzed with genotypes, we found that GTM3AB geno-type was significantly associated with reduced risk of EC (OR335= 0.48; P = 0.022), which implies that GSTM3 may not beinvolved in tobacco-related higher risk for EC in the KashmirValley. Therefore, there is a need to look for association of othercandidate genes in conferring risk of esophageal cancer. Pre-viously, Li et al (51) reported that esophageal cancer related340gene-1(ECRG1) 290Gln variant allele may have a role in ECsusceptibility and interaction with smoking. In addition, otherclasses of genes such as CYP1B1, MAPK14, ATF4, SOD2, andSERPINB2 may also be explored in response to environmentalexposure in EC patients (52).345

The limitation in our study was a smaller sample size. As itis the first report of genetic susceptibility of EC due to poly-morphisms in xenobiotic metabolizing enzyme encoding genesin the Kashmiri population, there is a definite need to performsimilar studies in a larger sample size. Furthermore, functional350assays to determine conclusive role of these polymorphisms inthe same population are also warranted.

In conclusion, GSTP1val/val genotype and CYP2E1c1c2genotype/c2 allele increased risk of ESCC and EADC, respec-tively, in the Kashmir population, whereas GSTM3AB genotype355seems to confer lower risk for both ESCC and EADC. Gene en-vironment interaction showed that carriage of GSTP1val/val

with high salted tea intake enhanced the EC risk, but GTM3ABcarriers were associated with the lower risk of EC due to tobaccosmoking through a water pipe. 360

ACKNOWLEDGMENTSThe work was supported by a research grant from the Indian

Council of Medical Research (ICMR), New Delhi. The authorsthank Bashir Ahmad, Ghulam Mustafa, and faculty members ofGastroenterology Department, SKIMS, Srinagar for their help in 365sample collection. Manzoor Ahmad and Malik Rohit Upadhyayboth contributed equally. Q5

REFERENCES1. Tsigris C, Chatzitheofylaktou A, Xiromeritis C, Nikiteas N, and Yannopou-

los A: Genetic association studies in digestive system malignancies. Anti- 370cancer Res 27, 3577–3587, 2007.

2. Parkin DM, Bray F, Ferlay J, and Pisani P: Global cancer statistics, 2002.CA Cancer J Clin 55, 74–108, 2005.

3. Khuroo MS, Zargar SA, Mahajan R, and Banday MA: High incidence ofoesophageal and gastric cancer in Kashmir in a population with special 375personal and dietary habits. Gut 33, 11–15, 1992.

4. Siddiqi M, Tricker AR, and Preussmann R: The occurrence of preformedN nitroso compounds in food samples from a high risk area of esophagealcancer in Kashmir, India. Cancer Lett 39, 37–43, 1988.

June 10, 2010 19:14 801xml HNUC_A_461099

8 M. A. MALIK ET AL.

5. Islami F, Pourshams A, Nasrollahzadeh D, Kamangar F, Fahimi S, et al.:380Tea drinking habits and oesophageal cancer in a high risk area in northernIran: population based case-control study. BMJ 26, 338–b929, 2009. Gon-zalez FJ: The molecular biology of cytochrome P450s. Pharmacol Rev 40,243–288, 1988.

6. Shi SF, Xie QX, and Li XB: The advance process in relation to the suscep-385tibility of tumor. Guowai Yixue Zhongliuxue Fence 4, 304–307, 1988.Q6

7. Cai L, Yu SZ, and Zhang ZF: Glutathione S-transferases M1, T1 genotypesand the risk of gastric cancer: a case-control study. World J Gastroenterol7, 506–509, 2001.

8. Ali-Osman F, Akande O, Antoun G, Mao JX, and Buolamwini J: Molecular390cloning, characterization, and expression in Escherichia coli of full-lengthcDNAs of three human glutathione S-transferase Pi gene variants. Evidencefor differential catalytic activity of the encoded proteins. J Biol Chem 272,10004–10012, 1997.

9. Spink DC, Spink BC, Cao JQ, De Pasquale JA, Pentecost BT, et al.: Differ-395ential expression of CYP1A1 and CYP1B1 in human breast epithelial cellsand breast tumor cells. Carcinogenesis 19, 291–298, 1998.

10. Hayashi S, Watanabe J, and Kawajiri K: Genetic polymorphisms in the 5′flanking region change transcriptional regulation of the human cytochromeP450IIE1 gene. J Biochem 110, 559–565, 1991.400

11. Shi Y, Lee JS, and Galvin KM: Everything you have ever wanted to knowabout Yin Yang 1. Biochim Biophys Acta 1332, 49–66, 1997.

12. Wideroff L, Vaughan TL, Farin FM, Gammon MD, Risch H, et al.: GST,NAT1, CYP1A1 polymorphisms and risk of esophageal and gastric adeno-carcinomas. Cancer Detect Prev 31, 233–236, 2007.405

13. Abbas A, Delvinquiere K, Lechevrel M, Lebailly P, Gauduchon P, et al.:GSTM1, GSTT1, GSTP1, and CYP1A1 genetic polymorphisms and sus-ceptibility to esophageal cancer in a French population: different pattern ofsquamous cell carcinoma and adenocarcinoma. World J Gastroenterol 10,3389–3393, 2004.410

14. Martınez C, Martın F, Fernandez JM, Garcıa-Martın E, Sastre J, et al.:Glutathione S-transferases mu 1, theta 1, pi 1, alpha 1 and mu 3 geneticpolymorphisms and the risk of colorectal and gastric cancers in humans.Pharmacogenomics 7, 711–718, 2006.

15. Miller SA, Dykes DD, and Polesky HF: A simple salting out procedure for415extracting DNA from human nucleated cells. Nucleic Acids Res 16, 1215,1988.

16. Setiawan VW, Zhang ZF, Yu GP, Li YL, Lu ML, et al.: GSTT1 andGSTM1 null genotypes and the risk of gastric cancer: a case-control studyin a Chinese population. Cancer Epidemiol Biomarkers Prev 9, 73–80,4202000.

17. Loktionov A, Watson MA, Gunter M, Stebbings WS, SpeakmanCT, et al.: Glutathione-S-transferase gene polymorphisms in colorec-tal cancer patients: interaction between GSTM1 and GSTM3 allelevariants as a risk modulating factor. Carcinogenesis 22, 1053–1060,4252001.

18. Toruner GA, Akyerli C, Ucar A, Aki T, Atsu N, et al.: Polymorphisms ofglutathione S-transferase genes (GSTM1, GSTP1 and GSTT1) and bladdercancer susceptibility in the Turkish population. Arch Toxicol 75, 459–464,2001.430

19. Kawajiri K, Nakachi K, Imai K, Yoshii A, Shinoda N, et al.: Identifi-cation of genetically high risk individuals to lung cancer by DNA poly-morphisms of the cytochrome P450IA1 gene. FEBS Lett 263, 131–133,1990.

20. Hayashi S, Watanabe J, and Kawajiri K: Genetic polymorphisms in the 5′435flanking region change transcriptional regulation of the human cytochromeP450IIE1 gene. J Biochem 110, 559–565, 1991.

21. Gauderman WJ and Morrison JM: QUANTO 1.1. A computer programfor power and sample size calculations for genetic-epidemiology studies.Available from: http://hydra.usc.edu/gxe, 2006.440

22. Nelson HH, Wiencke JK, Christiani DC, Cheng TJ, Zuo ZF, et al.: Eth-nic differences in the prevalence of the homozygous deleted genotype ofglutathione S-transferase theta. Carcinogenesis 16, 1243–1245, 1995.

23. Shao G,Su Y, Huang G, and Wen B: Relationship between CYP1A1,GSTM1 genetic polymorphisms and susceptibility to esophageal squamous 445cell carcinoma. Zhonghua Liuxing Bingxue Zazhi 21, 420–423, 2000.

24. Boccia SLa, Torre G, Gianfagna F, Mannocci A, and Ricciardi G: Meta-analysis of GSTT1 and gastric cancer. Mutagenesis 3, 1432–1435, 2006.

25. Tan W, Song N, Wang GQ, Liu Q, Tang HJ, et al.: Impact of geneticpolymorphisms in cytochrome P450 2E1 and glutathione S-transferases 450M1, T1, and P1 on susceptibility to esophageal cancer among high-riskindividuals in China. Cancer Epidemiol Biomarkers Prev 9, 551–556, 2000.

26. Jourenkova-Mironova N, Voho A, Bouchardy C, Wikman H, Dayer P, et al.:Glutathione S-transferase GSTM3 and GSTP1 genotypes and larynx cancerrisk. Cancer Epidemiol Biomarkers Prev 8, 185–188, 1999. 455

27. Park LY, Muscat JE, Kaur T, Schantz SP, Stern JC, et al.: Comparison ofGSTM polymorphisms and risk for oral cancer between African-Americansand Caucasians. Pharmacogenetics 10, 123–131, 2000.

28. Rannug A, Alexandrie AK, Persson I, and Ingelman-Sundberg M: Geneticpolymorphism of cytochromes P450 1A1, 2D6 and 2E1: regulation and 460toxicological significance. J Occup Environ Med 37, 25–36, 1995.

29. Liu S, Park JY, Schantz SP, Stern JC, and Lazarus P: Elucidation of CYP2E15’ regulatory RsaI/Pstl allelic variants and their role in risk for oral cancer.Oral Oncol 37, 437–445, 2001.

30. De SE, Boffetta P, Carzoglio J, Mendilaharsu S, and Deneo-Pellegrini H: 465Tobacco smoking and alcohol drinking as risk factors for stomach cancer:a case-control study in Uruguay. Cancer Cause Control 9, 321–329, 1998.

31. Available from: http://www.ncbi.nlm.nih.gov/projects/SNP/32. Jain M, Kumar S, Rastogi N, Lal P, Ghoshal UC, et al.: GSTT1, GSTM1

and GSTP1 genetic polymorphisms and interaction with tobacco, alcohol, 470and occupational exposure in esophageal cancer patients from North India.Cancer Lett 242, 60–67, 1995, 2006. Q7

33. Jain M, Kumar S, Lal P, Tiwari A, Ghoshal UC, et al.: Role of GSTM3 poly-morphism in the risk of developing esophageal cancer. Cancer EpidemiolBiomarkers Prev 16, 178–181, 2007. 475

34. Sikdar N, Paul RR, and Roy B: Glutathione S-transferase M3 (A/A) geno-type as a risk factor for oral cancer and leukoplakia among Indian tobaccosmokers. Int J Cancer 109, 95–101, 2004.

35. Sikdar N, Mahmud SA, Paul RR, and Roy B: Polymorphism in CYP1A1and CYP2E1 genes and susceptibility to leukoplakia in Indian tobacco 480users. Cancer Lett 195, 33–42, 2003.

36. Sobti RC, Sharma S, Joshi A, Jindal SK, and Janmeja A: Genetic poly-morphism of the CYP1A1, CYP2E1, GSTM1 and GSTT1 genes and lungcancer susceptibility in a north Indian population. Mol Cell Biochem 266,1–9, 2004. 485

37. Babu KA, Reddy NG, Deendayal M, Kennedy S, and Shivaji S: GSTM1,GSTT1, and CYP1A1 detoxification gene polymorphisms and their rela-tionship with advanced stages of endometriosis in South Indian women.Pharmacogenet Genomics 15, 167–172, 2005.

38. Jain M, Kumar S, Ghoshal UC, and Mittal B: CYP1A1 Msp1 T/C poly- 490morphism in esophageal cancer: no association and risk modulation. OncolRes 16, 437–443, 2007.

39. Tripathi S, Ghoshal U, Ghoshal UC, Mittal B, Krishnani N, et al.: Gastriccarcinogenesis: possible role of polymorphisms of GSTM1, GSTT1, andGSTP1 genes. Scand. J. Gastroenterol 43, 431–439, 2008. 495

40. Morita S,Yano M, Shiozaki H, Tsujinaka T, Ebisui C, et al.: CYP1A1,CYP2E1 and GSTM1 polymorphisms are not associated with susceptibilityto squamous-cell carcinoma of the esophagus. Int J Cancer 71, 192–195,1997.

41. Harries LW, Stubbins MJ, Forman D, Howard GC, and Wolf CR: Identifi- 500cation of genetic polymorphisms at the glutathione S-transferase Pi locusand association with susceptibility to bladder, testicular and prostate cancer.Carcinogenesis 18, 641–644, 1997.

42. van-Lieshout EM, Roelofs H, Dekker S, Mulder CJ, Wobbes T, et al.:Polymorphic expression of the glutathione S-transferase P1 gene and its 505susceptibility to Barrett’s esophagus and esophageal carcinoma. CancerRes 59, 586–589, 1999.

June 10, 2010 19:14 801xml HNUC_A_461099

XENOBIOTIC METABOLIZING ENZYMES GENETIC POLYMORPHISMS WITH ESOPHAGEAL CANCER 9

43. Lan Q, Chow WH, Lissowska J, Hein DW, Buetow K, et al.: GlutathioneS-transferase genotypes and stomach cancer in a population-based case-control study in Warsaw, Poland. Pharmacogenetics 11, 655–661, 2001.510

44. Nakachi K, Imai K, Hayashi S, Watanabe J, and Kawajiri K: Genetic sus-ceptibility to squamous cell carcinoma of the lung in relation to cigarettesmoking dose. Cancer Res 51, 5177–5180, 1991.

45. Wu MS, Chen CJ, Lin MT, Wang HP, Shun CT, et al.: Genetic polymor-phisms of cytochrome p450 2E1, glutathione S-transferase M1 and T1,515and susceptibility to gastric carcinoma in Taiwan. Int J Colorectal Dis 17,338–343, 2002.

46. Marchand LL, Wilkinson GR, and Wilkens LR: Genetic and dietary pre-dictors of CYP2E1 activity: a phenotyping study in Hawaii Japaneseusing chlorzoxazone. Cancer Epidemiol Biomarkers Prev 8, 495–500,5201999.

47. Rossini A, Rapozo DC, Soares, Lima SC, Guimaraes DP, et al.: Polymor-phisms of GSTP1 and GSTT1, but not of CYP2A6, CYP2E1 or GSTM1,

modify the risk for esophageal cancer in a western population. Carcinogen-esis 28, 2537–2542, 2007. 525

48. Siddiqi M, Kumar R, Fazili Z, Spiegelhalder B, and Preussmann R: In-creased exposure to dietary amines and nitrate in a population at high riskof oesophageal and gastric cancer in Kashmir (India). Carcinogenesis 13,1331–1335, 1992.

49. Siddiqi M, Tricker AR, and Preussmann R: Formation of N-nitroso com- 530pounds under simulated gastric conditions from Kashmir foodstuffs. CancerLett 39, 259–265, 1988.

50. Li Y, Zhang X, Huang G, Miao X, Guo L, et al.: Identification of a novelpolymorphism Arg290Gln of esophageal cancer related gene 1 (ECRG1)and its related risk to esophageal squamous cell carcinoma. Carcinogenesis 53527, 798–802, 2006.

51. van Leeuwen DM, van Agen E, Gottschalk RW, Vlietinck R, Gielen M,et al.: Cigarette smoke-induced differential gene expression in blood cellsfrom monozygotic twin pairs. Carcinogenesis 28, 691–697, 2007.