Tobacco Documents Online

Int. J. Cancer: 71, 192-195 (1997) © 1997 Wiley-Liss, Inc. Publication of the International Union Against Cancer Publication de I'Union Internation~le Contre le Cancer CYP1A1, CYP2E1 AND GSTM1 POLYMORPHISMS ARE NOT ASSOCIATED WITH SUSCEPTIBILITY TO SQUAMOUS-CELL CARCINOMA OF THE ESOPHAGUS Shunji MoRrr~,~*, Masahiko YANO~, Hitoshi SHIOZAKZ~, Toshimasa TsuJ~rq~,KA1, Chikara EBtsU0, Takashi MORIMOTO~, Masanod KtSHIBUTIl, Junya FUJITA1, Atsuhiro OGAWA~, Masaaki TANIGUCHIl, Masatoshi INOUE~, Shigeyuki TAMURA1, Keiji YAMAZAKI3, Nobuteru KIKKAWA4, Sumio MIZUr~OYA5 and Morito MO~DE~1 t Department of Surgery II, Osaka Universi~. Medical School, Suim, Japan 2Department of Surgery, Toyonaka Municipal Hospital, Toyonaka, Japan 3Department of Surgery, Kansai Rosai Hospital, Amagasaki, Japan 4Departtnent of Surgery, Osaka National Hospital, Osaka, Japan 5Department of Surgery, Kinki Central Hospital, Itarni, Japan We investigated the genetic polymorphisms of CYPIAI, CYP2EI and GSTNII in Japanese esophageal cancer patients (n = 53) with a histological diagnosis of squamous-cell carci- noma, to determine whether susceptibility to esophageal cancer is associated with these polymorphisms. There were no significant differences in the frequency distribution of any one of the 3 polymorphisms between esophageal cancer patients and 132 healthy Japanese controls. The genotype distributions in tobacco smokers or alcohol drinkers were also quite similar for male patients and male controls. The age at onset of esophageal cancer was also similar for patients with any genotype of the 3 polymorphisms. We conclude that the 3 polymorphisms are unlikely to be associated with esophageal cancer susceptibility. Int. J. Cancer 71 : 192-195, 1997. © 1997 W~ley-Liss, Inc. Recent advances in molecular biology have revealed that several genetic mutations or deletions in cancer-related genes cause carcinogenesis. At the same time, about 65% of human cancers are associated with diet and tobacco smoking (Doll and Peto, 1981), and this suggests that the main causes of human cancers are genetic changes in cancer-related genes produced by carcinogens in the environment. Most carcinogens are not carcinogenic per se, but require metabolic activation to become carcinogenic, for example by drug-metabolizing enzymes such as cytochromes P450 (CYPs). Recently, many investigators have reported an association between the polymorphisms of CYPs and cancer susceptibility. Lung cancer is one of the most extensively investigated cancers and has been reported to be associated with the polymorphisms of CYP1A1, 2E1 and 2D6 and with the gluthathione S-transferase M 1 gene (GSTM1) (Raunio etal., 1995). These CYPs are involved in the metabolic activation of procarcinogens contained in tobacco smoke into ultimate carcinogens, which bind to DNA covalently to form DNA adducts, while GSTs are involved in the detoxification of the activated carcinogens. Therefore, the genotypes with high phase-! enzyme activity and low phase-II enzyme levels are considered to pose a high risk cancer development (Hayashi et al., 199 la; Kihara et al., 1995). Anttila et al. (1991) reported a correlation between aromatic lung DNA adducts and CYPIA1 activity in smokers, while Kato et al. (1995) have shown an association between the polymorphisms of CYP2D6, CYP2E1 and GSTM1 and higher DNA adduct levels in lung tissue. The esophagus is similar to the lung in terms of being frequently exposed to xenobiotics, including tobacco smoke and many kinds of food. Epidemiological studies have shown that an increased risk of developing squamous-cell carcinoma of the esophagus is associated with smoking and the consumption of alcoholic beverages, indicating that CYPIAI and CYP2E1 may be involved in carcinogenesis of esophageal cancer, because 2El is an ethanol-inducible form of CYP (Perrot etaL, 1989). This report examines the association between the polymor- phisms of CYPIA1, 2E1 and GSTMI and susceptibility to esopha- geal carcinoma. MATERIAL AND METHODS Patients and controls The patients and controls enrolled in this study are listed in Table I. The former consisted of 53 Japanese patients with esophageal cancer histol6gically diagnosed as squamous-cetl carcinoma, who underwent esophagectomy between'September, 1992 and Novem- ber, 1995 at Osaka University Medical School and 3 associated hospitals in Osaka and Hyogo Prefectures. The normal esophageal mucosa of these patients was kept frozen immediately after removal and stored at -80°C until use. The mucosa samples were ascertained as normal by iodine staining and by H. and E. staining of the surrounding esophagus. Blood samples were collected from 132 healthy Japanese controls during their periodical general health check-up in Osaka in March, 1996. Information concerning sex, age, tobacco, alcohol use and past history was obtained from clinical records for patients and from interviews for controls. Those controls with any history of malignancy were excluded from the study. DNA extraction Genomic DNA was extracted from normal esophageal tissues and from white blood cells of controls by using a DNA separation kit, SepaGene (Sanko-Jyunyaku, Tokyo, Japan). The amount of DNA was determined spectrophotometrically and an aliquot (1 lag of DNA) from each sample was used for polymerase chain reaction (PCR). PCR analysis ofCYPIAI gene polymorphism The A to G transition polymorphism in exon 7 of the CYP1A1 gene was analyzed for each subject as described previously (Katoh etal., 1995). Briefly, 1 lag of the DNA sample was amplified by using the primers 5'GTCTCCCTCTGGTTACAGGA and 5' GAAAGACCTCCCAGCGGTCA in buffer [10 mM Tris-HC1, 50 mM KC12, 2 mM MgC12] with 8 laM of each deoxynucleoside tfiphosphate and Taq polymerase (Takara Shuzo, Kyoto, Japan). Initial denaturation was performed at 95°C for I0 rain, followed by 35 cycles of annealing for 1 rain at 55°C, extension for 1 min at 72°C and denaturation for 1 rain at 95°C with a DNA Thermal Cycler PJ2000 (Perkin-Elmer Cetus, Norwalk, CT). The PCR products were digested with HincI[ (Fermantas-MB[, Vilnius, Abbreviations: CYP, cytochrome P450; GST, glutathione S-transferase; PCR, polymerase chain reaction. Contract grant sponsor: Ministry of Health and Welfare of Japan, contract grant number 7-24. *Correspondence to: Department of Surgery [[, Osaka University Medical School, 2-2 Yamadaoka, Suita City 565, Osaka, Japan. Fax: 81 6 879 3259. Received 7 October 1996; revised 18 December 1996 q

CYP POLYMORPHISMS AND ESOPHAGEAL CANCER 193 TABLE 1 - BACKGROUND DATA OF PATIENTS AND CONTROLS Patients n z 53 Controls n z 132 p valuet Male/female 45/8 112/20 >0.99 Age' 62.2 ± 7.9 50.5 ± 9.1 <0.0001 (46-79) (29-74) Smokers 30 56 0.083 Drinkers 40 66 0.0014 tStatistical comparison between patients and controls was done by Xz test or Student's t-test.-2Mean ± SD (range).~X2 = 3.1.-4X2 = 10.0. Lithuania) at 37°C overnight, subjected to electrophoresis on a 4.0% NuSieve GTG agarose gel (Takara Shuzo, Kyoto, Japan) and visualized by ethidium bromide staining. PCR analysis resulted in the following genotype classification: a predominant homozygote (lie/lie), a heterozygote (Ile/Val) and a rare homozygote (Val/Val). PCR analysis ofCYP2E 1 gene polymorphism The 5'-flanking region polymorphism of the CYP2E1 gene was analyzed by procedures described previously (Hayashi et al., 1992). Briefly, PCR was performed using the primers 5'CCAGTC- GAGTCTACATTGTCA and 5'TTCATTCTGTCTTCTAACTGG. Initial denaturation was performed at 95°C for 2 min, followed by 25 thermal cycles consisting of denaturation for 1 min at 95°C, annealing for I min at 55°C, extension for I min at 72°C and a final extension for 10 min at 72°C. The PCR products were digested with RsaI (Life-Technologies, Gaithersburg, MD) or PstI (Takara Shuzo) at 37°C overnight and subjected to electrophoresis on a 2.0% agarose gel (Life-Technologies). The genotypes of CYP2E1 were classified as follows: a predominant homozygote (cl/cl), a heterozygote (cl/c2) and a rare homozygote (c2/c2). PCR analysis ofGSTM1 gene deletion The GSTM1 gene deletion was determined as described else- where (Groppi et aL, 1991) with some modification. Briefly, PCR was performed by using the primers 5' GAAGGTGGCCTCCTCCT- TGG and 5' AATTCTGGATTGTAGCAGAT in the presence of a separate pair of primers to co-amplify the 5'-flanking region of CYP2E1 described earlier. PCR conditions were the same as those for CYP2E1 gene analysis. On the basis of the results obtained from this procedure, GSTM1 genotypes were divided into GSTM1 (+) and GSTMI (-), representing respectively the homozygous/ heterozygous positive genotype and the null genotype. Statistical analysis The X2 test was used to examine the differences in genotype distribution between patients and controls. Student's t-test was used to examine differences in continuous variables. To exclude interference of background factors among patients and controls, logistic regression analysis and analysis of variance were also used. RESULTS Although age-matching of patients and controls was not pos- sible, because almost all those who had a general health check-up were younger than the esophageal cancer patients (Table I), the polymorphisms we studied featured genetically determined charac- teristics and should therefore remain unchanged with age, so that we assumed the age distribution would have no effect on the genotype frequency of either population. In fact, none of the genotypes of the controls showed any correlation with age, thus supporting our assumption (data not shown). In addition, the genotype distributions of controls were comparable with those previously reported for Japanese subjects (Hayashi et aL, .1992; Kihara et aL, 1995; Hayashi et al., 199 la). The polymorphism of CYP1AI in exon 7 has been reported to correlate with the inducibility phenotype (Cosma et al., 1993; Taioli et al., 1995) and the polymorphism in the 5'-flanking region of the CYP2EI to affect transcriptional regulation of the gene, resulting in different expression levels of the CYP2EI mRNA (Hayashi et al., 1991a," Watanabe et al., 1994). We therefore divided the genotypes of CYP1A1 and CYP2E1 into 2 groups: low-inducibility genotype (Ile/Ile or cl/cl) and high-inducibility genotype (Ile/Val, Val/Val, cl/c2 or c2/c2). No significant differ- ence was obtained between esophageal cancer cases and healthy controls in terms of frequency distribution of any of the genotypes in any single gene. The odds ratios of high-inducibility genotypes of CYP1A1 and CYP2E1 or GSTMI null genotype were approxi- mately 1.0 (Table II). Next, the frequency distributions of com- bined genotypes of the 3 polymorphisms were assessed, and again similar genotype distributions were seen for cases and controls. The odds ratios of the combined genotypes compared with the Ile/lle, cl/cl, GSTMI(+) gentStype were 0.8-1.7, none of which significantly differ from 1.0 (Table III). In order to examine the effect of tobacco and alcohol consumption on cancer susceptibility, the genetic analysis was also carded out for smokers and drinkers. We defined smokers as daily smokers and drinkers as daily drinkers. Patients showed a significantly greater proportion of drinkers and tended towards a higher percentage of smokers than did controls, as shown by epidemiological studies (Table I). The analysis of smokers and drinkers was done only for males to avoid the possibility that gender might affect these habits. Even when "smoking and drinking were taken into account, none of the genotype distributions showed any significant differences with male controls (Table IV). Finally, we compared the age at onset of cancer between low- and high-inducibility genotypes or between GSTM1 positive and null genotypes. The onset ages (mean - SD) of low- and high-inducibility genotypes were 63.1 ___ 6.8 and 60.8 - 9.4 for CYPIA1, 62.4 +-- 8.4 and 61.7 -- 7.2 for CYP2E1, and 62.1 +- 7.3 and 62.3 - 8.5 for GSTMI positive and GSTM1 null genotypes, respectively. Once again, there was no significant difference between the 2 groups. DISCUSSION Epidemiological studies have revealed the association of alcohol consumption and tobacco smoking with the risk of development of esophageal cancer. Tobacco smoke includes many kinds of carcino- gen, including polycyclic aromatic hydrocarbons and nitrosocom- pounds, which are known to be metabolized into their activated forms by CYPs. CYP1AI and CYP2E1 are known to play an important role in this process. N-nitrosonornicotine in particular, which is one of the most abundant nitrosocompounds in tobacco smoke, is well known to induce esophageal cancer in animals. Furthermore, CYP2E 1 has been shown to be induced by administra- tion of ethanol (Perrot etaL, 1989). For these reasons, we suspected that the inducible genotypes of CYPIAI or CYP2EI or the GSTMI null genotype might contribute to esophaget cancer development. Recently, many investigators have reported an association be- tween polymorphisms in carcinogen-activating and inactivating enzymes and cancer susceptibility (reviewed by Raunio et aL, 1995). A significant correlation was found in a Japanese population between susceptibility to lung cancer and the polymorphisms in exon 7 and in the 3'-flanking region of CYPIA1 recognized by MspI restriction. Such a correlation could not be established for Caucasian populations. Polymorphism in the 5'-flanking region of CYP2E1 has also been reported to be associated with susceptibility to lung cancer in a Swedish population, but not in a Japanese population. Individuals lacking GSTM1 reportedly could be at greater risk of developing lung cancer, bladder cancer, skin cancer and colon cancer. Furthermore, several reports have revealed a greater risk for developing lung cancer when the susceptible genotypes of CYP1A1 and GSTMI co-exist. The esophagus is similar to the lung inasmuch as both are frequently exposed to xenobiotics, including tobacco smoke, which is a major risk factor for the development of cancer in both organs. However, this study could not identify any correlation at all between esophageal cancer susceptibility and these polymor- phisms. Though patients and controls had different background features, we were able to ascertain the results by multivariate

194 MORITA ETAL. TABLE II - FREQUENCY DISTRIBUTIONS OFTHE POLYMORPHISMS OF CYPIAI, CYP2EI AND GSTMI Genotype CYPIA 1 CYP2EI GSTMI Ileflle lle/Val ValAtal el/el cl/e2 e2/e2 (+) (-) Patients n = 53 Controls n = 132 32 20 1 34 18 I 30 23 (60.4)~ (37.7) (1.9) (64.2) (34.0) (1.9) (56.6) (43.4) 80 49 3 85 42. 5 77 55 (60.6) (37. I) (2.3) (64.4) (31.8) (3.8) (58.3) (41.7) OR (95% CI)2 1.0 1.0 (0.5-t.9)3 1.0 1.0 (0.5-2.0)3 1.0 I. 1 (0.6-2.0) tFigures in parentheses indicate percentages.AOdds ratio (95% confidence interval).~Odds ratio of the heterozygous and homozygous variants combined. TABLE Ill - COMBINED GENOTYPES OF PATIENTS AND CONTROLS CYPIAI CYP2EI " GSTMI Patients n = 53 Controls n ~ 132 OR (95% CI)~ Ileflle el/el (+) 1.1 (20.8)2 33 (25.0) 1.0 (-) 10 (t8.9) 19 (14.4) 1.6 (0.6-4.4) ci/c2 or c2/c2 (+) 7 (13.2) 13 (9.9) 1.6 (0.5-5.1) (-) 4 (7.6) 15 (11.4) 0.8 (0.2-2.9) cl/cl (+) 7 (13.2) 22 (16.7) 1.0 (0.3-2.8) (-) 6 (11.3) 11 (8.3) 1.6 (0.5-5.5) c 1/c2 or c2/c2 (+) 5 (9.4) 9 (6.8) 1.7 (0.5-6.0) (-) 3 (5.7) 10 (7.6) 0.9 (0.2-3.9) Ile[V'al or VaLtVal ~Odds ratio (95% confidence interval) of each combination of genotypes compared with IIedlle, c l/c 1, GSTMI (+) genotype.-2Figures in parentheses indicate percentages. q TABLE IV - COMPARISON OF THE FREQUENCY DISTRIBUTIONS OF THE THREE POLYMORPHISMS Genotyp~ FOR MALE CONTROLS AND MALE PATIENTS: SMOKERS AND DRINKERS CYP I A 1 CYP2E I GSTM I lie/lie [le/Va[ or Val/Val c11¢[ el/c2 or c2/e2 (+) (-) Smokers Male patients 17 13 17 13 16 1~ n = 30 (56.7)~ (43.3) (56.7) (43.3) (56.6) (43.4) Male controls 31 25 37 19 36 20 n = 56 (55.4) (44.6) (66.1) (33.9) (64.3) (35.7) OR (95% CI)z 1.0 0.9 (0.4-2.3) 1.0 1.5 (0.6-3.7) 1.0 1.6 (0.6-3.9) Drinkers Male patients 26 14 27 13 21 19 n = 40 (65.0) (35.0) (67.5) (32.5) (52.5) (47.5) Male controls 37 29 43 23 40 26 n = 66 (56.1) (43.9) (65.2) (34.8) (60.6) (39.4) OR (95% CI)2 1.0 0.9 (0.4-2.1) 1.0 0.7 (0.3-1.5) 1.0 1.4 (0.6-3. [) ~Figures in parentheses indicate percentages.~Odds ratio (95% confidence interval). analysis to exclude the interference of the background factors; logistic regression analysis indicated that only age (p < 0.01), alcohol consumption (p < 0.01), and sex (p = 0.03) had a signifi- cant association with cancer susceptibility, while none of the 3 polymorphisms or tobacco smoking did. In addition, the analysis of variance, using the 3 polymorphisms as "factor variables with the covariates of age, alcohol consumption and sex, also proved that there was no association between the polymorphisms of CYPIA1 (p = 0.95), CYP2EI (p = 0.96) and GSTM1 (p = 0.95) and cancer susceptibility. We could not evaluate the single risk of the rare homozygotes of CYPIAI and CYP2E1 because of the small population. However, such an evaluation would require a very large number of patients and controls, since our results suggest only a low risk contribution by these polymorphisms, if it exists at all. For example, on the assumption that the rare homozygosity frequency of Japanese is 0.04 and the number of controls is twice that of patients, more than 200 patients and 400 controls would be needed to detect an odds ratio of 2.0 which is significantly (p < 0.05) greater than 1.0. If our hypothesis is correct, that individual st~sceptibility to esophageal carcinogenesis is genetically determined by the recta- bolic potential for activation or inactivation of carcinogens in cigarette smoke, the 3 following possibilities must be considered. First, drug-metabolizing enzyme(s) other than CYP 1A l, CYP2E 1 and GSTM1 may play a major role in the metabolism of carcino- gens in the esophagus. A recent Western-blot analysis study has revealed the existence of CYP 1A2/l, CYP2E l, CYP2 B6, CYP2C l 1 and CYP3A4/3 in human esophagus (Nakajima et aL, 1996). However, what type of CYP metabolizes the carcinogens contained in tobacco smoke in the human esophagus has not yet been clarified. There might be other forms of CYP or phase-II enzymes more closely related to the activation and detoxification of carcino- gens in the esophagus. Second, the polymorphisms of CYPIA1 and CYP2EI examined in this study may not reflect individual inducibility. In fact, some recent reports have failed to show any association between these CYP polymorphisms and their function. Zhang et al. (1996) observed that the point mutation in exon 7 of CYP1A1 did not substantially affect the function of the CYPIAI protein. Wedlund et al. (1994) also failed to find any absolute correlation between the CYP1A1 polymorphism and the CYPIAI inducibility phenotype. Kim et al. (1995) reported that the polymorphism in the 5'-flanking

CYP POLYMORPHISMS AND ESOPHAGEAL CANCER region of CYP2E1 was not associated with differences in tl~e disposition of chlorzoxazone. In contrast to preliminary reports, the exact correlation between these polymorphisms and their function now remains controversial. CYPIA1 has 2 other polymorphisms, one called the Msp I polymorphism (Kawajiri et al., 1990) and the other the African-American polymorphism (Crofts et al., 1993). The Mspl polymorphism of CYP1A1 has been reported to be closely linked to the polymorphism in exon 7 (Hayashi et al., 199 Ib) and the African-American polymorphism has been found in African Americans but not in Caucasians or Asians (Crofts et al., 1993). CYP2EI also has another potymorphism, DraI polymor- phism, which has a polymorphic site at the second intron of CYP2E1 and is reportedly associated with lung cancer in Japanese patients (Uematsu et al., 1991), but it is not clear whether this polymorphism affects the inducibility of CYP2E1. Because CYP is one of the most polymorphic genes, there might be other unknown genetic polymorphisms of CYPIA1 and CYP2E1 which have a stronger effect on inducibility. 195 Finally, other factors such as repair genes, for example O~- methylguanine-DNA methyltransferase, might determine the indi- vidual susceptibility to esophageal carcinogenesis. We conclude that the 3 polymorphisms investigated in this study are unlikely to be correlated with esophageal cancer susceptibility. However, we still consider cytochrome P450 to be one of the most likely risk factors that determine individual esophageal-cancer susceptibility, because esophageal carcinoma clearly shows the characteristics of chemical carcinogenesis. Further investigation is therefore needed to ascertain the role of cytochrome P450 in esophageal carcinogenesis. ACKNOWLEDGEMENTS We thank Associate Prof. H. Kusuoka (Division of Tracer Kinetics, Biomedical Research Center, Osaka University Medical School) for his advice on statistical analysis. This work was supported in part by Grant-in-Aid for Cancer Research 7-24 from the Ministry of Health and Welfare. REFERENCES ANTTILA, S., HIETANEN, E., VAL\'IO. H., CAMUS, A.-M., GELBOIN, H.V., PARK, S.S., HEmKILX, L.. KAP.JAgAINEN. A. and BAR'rSCH, H., Smoking and peripheral type of cancer are related to high levels of pulmonary cyto- chrome P450IA in lung cancer patients, hu. J. Cancer, 47, 681-685 (1991 ). COSMA, G., Cgo~rs, E, TAtOLt, E.. TONtOLO, P. and GAg'rE, S., Relationship between genotype and function of the human CYP1A1 gene. J. Toxicol. environ. Hlth, 40, 309-316 (1993). CROFTS, E, COSt~A. G.N., CuP,~E. D.. TA~OLI, E., TONIOLO, E and GARTE, S.J., A novel CYP IA 1 gene polymorphism in Afdcan-Americ .arts. Carcino- genesis, 14, 1729-1731 (1993). DOLL, R. and PETO. R., The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J. nat. Cancer Inst., 66, 1191-1308 (1981). GROPPI, A., COUTELLE, C., FLEURY. B., IRON, A., BEGUERET, J. and COUZtGOU, P., Glutathione S-transferase class Ia in French alcoholic cirrhotic patients. Hum. Genet.. 87, 628-630 ( 1991). HAVASHL S.-I.. WA'rANABE. J. and KANAME, K., High susceptibility to lung cancer analyzed in terms of combined genotypes of P450IA 1 and mu-class glutathione S-transferase genes..lap. J. Cancer Res., 83, 866-870 (1992). HAVASHL S.-I., WA'rANAnE, J. and KAWAJIPd, K., Genetic polymorphisms in the Y-flanking region change transcriptional regulation of the human cytochrome P4501IE 1 gene. J. Biochem., 110, 559-565 (199 la). HAYASHL S.-I., WATANABE. J., NAKACHI, K. and KAWAJmt, K., Genetic linkage of lung cancer-associated MspI polymorphisms with amino acid replacement in the heme binding region of the human cytochrome P450IA 1 gene. J. Biochem., 110, 407--411 (1991b). KATO, S., BOWMAN, E.D.. HARRINGTON, A.M., BLOMEKE, B. and SHIELDS, P.G., Human lung carcinogen-DNA adduct levels mediated by genetic polymorphisms h~ vivo. J. nat. Cancer b)st., 87, 902-907 (1995). KATOH. T., INATOMI, H.. NAGAOKA. A. and SuGrrA, A:, Cytochrome P4501AI gene polymorphism and homozygous deletion of the glutathione S-transferase MI gene in umthelial cancer patients. Carcinogenesis, 16, 655-657 (1995). KAWAJIRI, K., NAKACHI. K., LMAI. K., YOSHIL A., SHINODA, N. and WATANABE, J., Identification of genetically high risk individuals to lung cancer by DNA polymorphisms of the cytochrome P450IAI gene. FEBS Lett., 263, 131-133 (1990). KIHARA, M., KIHARA, M. and NODA, K., Risk of smoking for squamous and small-cell carcinomas of the lung modulated by combinations of CYPIA1 and GSTMI gene polymorphisms in a Japanese population. Carcinogenesis, 16, 2331-2336 (1995). K~M, R.B., O'SHEA, D. and WILKINSON, G.R., Interindividual variability of chlorzoxazone 6-hydroxylation in men and women and its relationship to CYP2E1 genetic polymorphisms. Clin. Pharmacol. Ther., 57, 645-655 (1995). NAKAJIMA, T., WANG, R.-S., NIMURA, V., PIN, Y.-M., HE, M., VAINIo, H., MURAYAMA, N., AOYAMA, T. and tlDA, F., Expression of cytochrome P450s and glutathione S-transferases in human esophagus with squamous-cell carcinomas. Carcinogenesis, 17, 1477-1481 (1996). PEa.go'r, N., NALt'AS, B., YANO, C.S. and BEAUNE, P.H., Modulation of cytochrome P450 isozymes in human liver, by ethanol and drug intake. Europ. J. din. Invest., 19, 549-555 (1989). RAUNIO, H., HUSGAFVEL-PURSlAINEN, K., ANTTILA, S., HIETANEN, E., HIRVONEN, A. and PELKONEN, O.. Diagnosis of polymorphisms in carcinogen- activating and inactivating enzymes and cancer susceptibility--a review. Gene, 159, 113-t21 (1995). TAIOLL E., CROFS, F., TRACHMAN, J., BAYO, S., TONIOLO. P. and GAgTE, S.J., Racial differences in CYPIAI genotype and function. Toxicol. Lett., 77, 357-362 (1995). UEMATSU, F., KIKUCH1, H., MOTOMIYA, M., TATSUYA, A., SAGAMI, L, OHMACHI. T., WAKUI, A., KANAMURA, R. and WATANABE, M., Association between restriction fragment length polymorphism of the human cyto- chrome P450IIIEI gene and susceptibility to lung cancer. Jap?J. Cancer Res., 82, 254-256 (1991). WATANABE, J., HAYASHI, S.-I. and KAWAJIRI, K.. Different regulation and expression of the human CYP2E1 gene due to the RasI polymorphism in the Y-flanking region. J. Biochem., 116, 321-326 (1994). WEDLUND, P.J., KIMURA, S., GONZALEZ. F.J. and NEBERT, D.W., 1462V mutation in the human CYPIA1 gene: lack of correlation with either the Mspl 1.9 kb (M2) allele or CYPIA1 inducibility in a three-generation family of east Mediterranean descent. Pharmacogenetics. 4, 21-26 (1994). ZHANG, Z.-Y., FASCO, M.J., HAUNG, L., GUENGERICH, F.P. and KAMINSKY, L.S., Characterization of purified human recombinant cytochrome P4501A l- Ile~62 and -Val : assessment of a roIe for the rare allele in carcinogenesis. Cancer Res., 56, 3926-3933 (1996).