Germ line polymorphisms in cytochrome-P450 1A1 (C4887 CYP1A1 ...

Carcinogenesis vol.18 no.12 pp.2307–2311, 1997

ACCELERATED PAPER

Germ line polymorphisms in cytochrome-P450 1A1 (C4887 CYP1A1) and methylenetetrahydrofolate reductase (MTHFR) genes and endometrial cancer susceptibility Manel Esteller1,2, Angel Garcia3, Josep Maria MartinezPalones4, Jordi Xercavins4 and Jaume Reventos1,5 1Unitat

de Recerca Biomedica, Centre d’Investigacions en Bioquimica i Biologia Molecular Vall d’Hebron and Departments of 3Pathology and 4Gynecology and Obstetrics, Hospitals Vall d’Hebron, Barcelona, Spain 2Current

address: Tumor Biology, The Johns Hopkins Oncology Center, 424 North Bond Street, Baltimore, MD 21213, USA

5To

whom correspondence should be addressed at: Unitat de Recerca Biomedica, Centre d’Investigacions en Bioquimica i Biologia Molecular Vall d’Hebron, 14th Floor, Hospital Universitari Materno-Infantil Vall d’Hebron, Pg. Vall d’Hebron 119–129, 08035, Barcelona, Spain

We describe here a case-control study to identify associations between polymorphisms at the methylenetetrahydrofolate reductase (MTHFR) and cytochrome P-450 1A1 (CYP1A1) genes and susceptibility to endometrial cancer. Accordingly, genotype frequencies in 80 endometrial carcinoma patients were compared with frequencies in 60 controls. DNA analysis suggest a significantly increased endometrial cancer risk with an alanine to valine substitution at nucleotide 677 of MTHFR gene with an odds ratio of 2.8 (95% confidence interval: 1.36–6.14, P J 0.002). Moreover, the tumors from patients with the valine allele were more undifferentiated (P J 0.03). On the other hand, a recently described mutation in exon 7 of CYP1A1 gene (threonine exchanged to asparagine in codon 461) showed a strong association with endometrial cancer risk with an odds ratio of 6.36 (95% confidence interval: 1.99–26.5, P J 0.0004). Thus, this study suggests that polymorphisms at MTHFR and a novel CYP1A1 variant could influence susceptibility to endometrial cancer, although larger sample sizes would be required to corroborate these findings. Introduction Classical genetic approaches for identifying susceptibility genes, although successful for cancer with strong familial links, have not yet identified corresponding genes for sporadic cancers. For example, the BRCA1 and BRCA2 genes in breast cancer and several DNA mismatch repair genes in hereditary nonpolyposis colon cancer, although strongly associated with familial cancer, accounts only for ,10% of nonfamilial malignancies (1,2). In addition, phenotype variation between individuals carrying the same mutation in a high-penetrance gene has been described, i.e. the severity of duodenal polyposis in carriers of identical APC gene germ line mutations may be influenced for a locus on chromosome 1p35-p36 (3). Such observations have led to a number of studies attempting to identify those low-penetrance genes that modify an indi*Abbreviations: MTHFR, methylenetetrahydrofolate reductase; CYP1A1, cytochrome P-450 1A1; CYPs, phase I cytochromes P450; GSTs, phase II glutathione-S-transferases; Thr, threonine; Asn, asparagine; Ala, alanine; Val, valine; PCR-RFLP, polymerase chain reaction and restriction fragment length polymorphism; OR, odds ratio; CI, confidence intervals; HNPCC, hereditary non-polyposis colorectal carcinoma; PAHs, polycyclic aromatic hydrocarbons. © Oxford University Press

vidual’s risk of cancer. Because sporadic cancers result from mutations in transforming genes, and carcinogen detoxification influences the mutational events in these key genes, several polymorphic carcinogen-metabolism genes are potentially useful candidates. In this sense, three supergene families have attracted interest: phase I cytochromes P450 (CYPs*) and phase II glutathione-S-transferases (GSTs) and N-acetyltransferases. In particular, certain variants at CYP1A1, CYP2D6, CYP2E1, GSTM1, GSTT1, NAT1 and NAT2 genes have been related with altered risk of various cancers (4). Three polymorphisms had been described in the human CYP1A1 gene: a Msp I RFLP in the 39-noncoding region (5), an adenine to guanine transition in the heme-binding domain of exon 7 (Ile .Val exchange in residue 462) (6), and an African-American-specific RFLP in intron 7 (7). The first two have been found overrepresented among lung cancer patients in Japan (8), but reports in Caucasians are inconclusive (4). Moreover, both have been related with a slightly elevated risk of colon and postmenopausal breast cancer (9–11). In addition, the Msp I polymorphism has been found associated with breast cancer in African-American women (12). Recently, we have found an enhanced endometrial cancer risk associated with the 39-end and exon 7 CYP1A1 germ line polymorphisms (13). Finally, a new polymorphism in exon 7 of human CYP1A1 (C4887) has been recently described resulting in a threonine (Thr) to asparagine (Asn) exchange in codon 461 in the heme binding region of the protein (14). The significance of this variant is still unknown, although the exchange of the adjacent amino acid residues, isoleucine to valine, results in a change in enzyme activity (15). On the other hand, germ line polymorphisms in the same oncogenes and tumor suppressor genes may also account for the different cancer susceptibility. In this sense, rare alleles of HRAS1 and p53 may confer an increased risk of certain types of cancer, including breast, ovarian and endometrial cancer (13,16–18). A disproportionally high rate of CpG . TpG transitions is frequently observed in tumor suppressor genes in various types of carcinomas (19) potentially due to deamination of 5-methylcytosine. In addition, methyldeficient diets may cause imbalances in the pools of nucleotide precursors leading to DNA strand breaks and mutations (20,21). Moreover, due to methylenetetrahydrofolate reductase (MTHFR) gene being a critical enzyme in the regulation of folate and methionine metabolism, both of which are important factors in DNA methylation and synthesis, germ line polymorphisms in the MTHFR gene could influence susceptibility to cancer. In this sense, an alanine (Ala) . valine (Val) (nucleotide 677: C .T) germ line polymorphism of the MTHFR gene been found to be related to the risk of colorectal cancer (22). Although endometrial carcinoma is a common female malignancy, relatively little attention has been given to genetic susceptibility factors. The present study was undertaken to examine the recently described CYP1A1 and MTHFR germ line polymorphisms as potential molecular markers of endometrial carcinoma susceptibility. 2307

M.Esteller et al.

Fig. 1. MTHFR gene polymorphism analyzed by PCR. The polymorphism at position 677 in the MTHFR gene was studied by PCR followed by Hinf I restriction enzyme digestion. Lane 1, mol. wt marker (pGEM/Hinf I, Rsa I and Sin I); lane 2, water control; lane 3, mutant homozygote; lanes 4, 5 and 8, wild-type homozygotes; lanes 6 and 7, mutant heterozygotes. Positions of the 198- and 175-base pair polymorphic DNA fragments are shown in the right margin.

Material and methods Study subjects We analyzed DNA extracted from 80 unrelated Caucasian patients with histologically proven diagnosis of endometrial carcinoma recruited in the Department of Obstetrics and Gynecology at the Hospital Universitari MaternoInfantil Vall d’Hebron of Barcelona. The protocol was approved by the Institutional Review Board and informed consent was obtained from all the patients involved in the study. None of the patients had received radiation therapy or hormonal treatment prior to surgery. Their ages ranged from 45– 82 years. Only five patients were premenopausal and the remaining 75 were postmenopausal. The stage distribution of the 80 patients according to the International Federation of Gynecology and Obstetrics (F.I.G.O.) staging system was Stage Ia (11 cases), Stage Ib (17 cases), Stage Ic (7 cases), Stage IIa (14 cases), Stage IIb (11 cases), Stage IIc (6 cases), Stage IIIa (9 cases), Stage IIIa (3 cases) and Stage IIIc (2 cases). Histologically, 74 of the 80 patients had endometrioid-type carcinomas whereas the remainder were four clear cell carcinomas and two papillary serous carcinomas. Among all surgically collected endometrial carcinomas, 40 were well-differentiated (G1), 23 were moderately differentiated (G2) and 17 were poorly differentiated (G3). The prevalence of the MTHFR and CYP1A1 polymorphisms studied were compared to those observed in a control group comprising 60 unrelated women from the same region and with the same ethnic background, attending to the Hospital Universitari Materno-Infantil Vall d’Hebron of Barcelona in the annual gynecological cancer screening program. Controls were randomly selected from those women who were free of clinical or histological malignancy. In addition, none had any personal history of cancer. Their ages ranged from 44–76 years. No differences were observed between cases and controls according to age distribution (P . 0.05). All endometrial cancer patients and control subjects were selected at Vall d’Hebron Hospital of Barcelona from January 1993 to December 1995. DNA was extracted from fresh endometrial tissue by proteinase K digestion and phenol/chloroform extraction (23). Genotyping The analysis of C.T transversion at position 677 of MTHFR, which results in the replacement of Ala by Val, was performed using polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP) as described by Frosst and co-workers (24). The Hinf I RFLP analysis for MTHFR polymorphism is illustrated in Figure 1. A single undigested band at 198 base pair represents a homozygous wild-type allele, two bands at 198and 175-base pairs represents the heterozygous genotype and a single band at 175 base pair represents a homozygous rare mutant allele. The analysis of C.T transversion at position 4887 in exon 7 of CYP1A1, which results in the replacement of Thr by Asn at residue 461 in the heme binding region of the enzyme, was performed using PCR-RFLP as described by Cascorbi and co-workers (14). The Bsa I RFLP analysis for CYP1A1 polymorphism is illustrated in Figure 2. A single undigested band at 204 base pair represents a homozygous rare mutant allele, two bands at 204- and 139base pairs represent the heterozygous genotype and a single band at 139 base pair represents a homozygous wild-type allele. Statistical analysis The odds ratio (OR) and 95% confidence intervals (CI) were calculated as a measure of the association between genotypes and endometrial cancer. The StatXact-Turbo statistical package was used to obtain exact P-values. Expected genotype frequencies were calculated by the Hardy-Weinberg equation (15 p2 1 2pq 1 q2) from the allele frequencies. Homozygous and heterozygous patients for the MTHFR Val variant where

2308

Fig. 2. CYP1A1 gene polymorphism analyzed by PCR. The polymorphism at position 4887 in exon 7 of the CYP1A1 gene was studied by PCR followed by Bsa I restriction enzyme digestion. Lane 1, mol. wt marker (pGEM/Hinf I, Rsa I and Sin I); lane 2, water control; lanes 3, 5, 6, 7 and 8, wild-type homozygotes; lane 4, mutant heterozygote. Positions of the 204- and 139-base pair polymorphic DNA fragments are shown in the right margin.

Table I. Association between germ line polymorphisms in CYP1A1 (Thr–Val) and MTHFR (Ala–Val) and endometrial cancer Genotype CYP1A1 Thr/Thr Thr/Val Val/Val MTHFR Ala/Ala Ala/Val Val/Val aHeterozygous

Case no. (%)

Control no. (%)

55 (68.7) 21 (26.2) 4 (5) ORa 6.36 (1.99–26.5)

56 (93.3) 4 (6.7) – P 5 0.0004

25 (31.2) 43 (53.7) 12 (15) ORa 2.88 (1.36–6.14)

34 (56.6) 20 (33.3) 6 (10) P 5 0.002

and homozygous mutant genotypes combined.

collapsed because both genotypes have significantly lower sp. act., lower residual activity after heating and enhanced homocysteine levels compared to the Ala homozygous genotype (24).

Results Table I shows the frequency of MTHFR and CYP1A1 genotypes in controls and in patients with endometrial carcinoma and Table II shows the distribution of both rare alleles according to clinicopathologic characteristics in endometrial carcinoma. Within the 60 individuals tested in the control group, all observed CYP1A1 and MTHFR genotype frequencies matched exactly the expected percentages as calculated from the allele frequencies. The MTHFR Val allele, collapsing the heterozygous and homozygous categories, was significantly associated with endometrial cancer with an OR of 2.8 (95% confidence interval: 1.36–6.14, P 5 0.002). The MTHFR Val allele distribution in endometrial cancer patients according to age at onset, F.I.G.O. stage and histological type of the tumors was not significantly different (Table II). However, when the results from the endometrial cancer patients were divided into well-differentiated (G1) versus moderately and poorly differentiated tumors (G21G3), a statistically significant association between the MTHFR Val allele and an undifferentiated cellular grade in endometrial carcinoma was found (P 5 0.03). Finally, a statistically strong significant association was found between endometrial carcinoma and replacement of threonine by asparagine at residue 461 in the heme binding region of the CYP1A1 gene (nt 4887). The OR and 95% CI of endometrial cancer risk for the combined genotypes of heterozygous and homozygous rare mutant alleles was 6.36

CYP1A1 and MTHFR genes and endometrial cancer Table II. Characteristics of patients with endometrial cancer in relation to polymorphisms of the CYP1A1 and MTHFR genes CYP1A1 polymorphism Bsa I genotypes w/w

w/m 1 m/m

23 57

Age 13 15

10

74 6

50 5

35 45

24 31

40 40

31 24

Variables

No. of patients

Case

80

ø60 .60 Histology Endometrioid Clear cell/UPSC F.I.G.O. stage I II 1 III Cellular grade G1 G21G3

MTHFR polymorphism Hinf 1 genotypes Pa

w/w

w/m 1 m/m

Pa

6 38 NS 22 4 NS 14 11 NS 8 17

17

NS

NS 19 NS 24 1 NS 11 14 NS 9 16

52 2 21 34 32 23

P 5 0.03

was calculated by χ2 test. w: wild-type allele; m: mutant allele; NS: Not significant (P .0.05); UPSC: Uterine Papillary Serous Carcinoma.

aP

(CI 1.99–26.5, P 5 0.0004). No significant differences in the distribution of CYP1A1 rare mutant allele in endometrial cancer patients according to age at onset, F.I.G.O. stage, cellular differentiation grade, and histological type of the tumors were found (Table II). Composite genotype analysis of the codon 461-Val variant with the codon 462 and Msp I CYP1A1 gene variants previously studied (13), reveals that the simultaneous carriers of codon 461-codon 462 variants possess the higher endometrial cancer risk (P 5 0.001). In addition, the simultaneous carriers of MTHFR-Val and CYP1A1-codon 461 Val risk variants show a higher endometrial cancer risk (OR 5 16, CI 95%, 2.61–163.29, P 5 0.0003) compared with those that possess neither of them. Discussion Although endometrial carcinoma is the most frequently diagnosed neoplasm of the female genital tract, little is known about genetic factors in the etiology of the disease. Several studies have shown endometrial carcinoma to be a significant component in a dominantly inherited cancer syndrome known as hereditary non-polyposis colorectal carcinoma (HNPCC). The genetic background of HNPCC involves mutations in several genes including MSH2, MLH1, PMS1 and PMS2 (1). In addition, epidemiologic studies have shown that a woman who is the first-degree relative of a patient affected with endometrial cancer has a significantly increased cancer risk. For example, analysis of the Cancer and Steroid Hormone Study data indicated that mothers and sisters of endometrial cancer patients had 2.7 times the risk for endometrial cancer than did controls (25). Some may harbor a strong hereditary predisposition to cancer, although its identification is obscure. To our knowledge, this is the first report of an association of a MTHFR germ line polymorphism with endometrial cancer risk. In this study, we found an association between the C.T transversion at position 677 of MTHFR gene, which results in the replacement of alanine by valine, and endometrial cancer risk with an OR of 2.8 (95% confidence interval: 1.36–6.14, P 5 0.002). Moreover, expanding our first finding of enhanced endometrial cancer risk associated with two known CYP1A1 polymorphisms (13), a stronger significant association was found between endometrial carcinoma risk and a recently described new genetic polymorphism in CYP1A1 gene (14), a replacement of threonine by asparagine at residue 461 in the

heme binding region in exon 7, with an OR of 6.36 (CI 1.99– 26.5, P 5 0.0004). Germ line variants of the MTHFR gene could be involved in endometrial cancer risk by altering DNA methylation or by influencing the rate of DNA mutation. The MTHFR mutation studied (677C.T Ala-to-Val) causes reduced enzyme activity (24). A decrease in the product of the MTHFR gene, 5-methylenetetrahydrofolate as (5-methylTHF), could contribute to carcinogenesis 5-methylTHF provides the methyl group for de novo methionine synthesis and DNA methylation (26). Abnormalities of DNA methylation are one of the most consistent molecular changes in human cancers. The alterations consist, simultaneously, of increased potential capacity for DNA methylation, widespread genomic hypomethylation, and more regional areas of hypermethylation; in this sense, selective growth and transformation of cells can result from DNA hypomethylation of protooncogenes or hypermethylation of tumor suppressor genes (27). Particularly for endometrial cancer, atypical distribution and increased levels of protein carboxyl methyltransferase (involved in methylation of ras and other GTP-binding proteins) (28,29) and abnormal methylation of estrogen receptor gene (30) have been described. In addition, an imbalance of the MTHFR substrate, 5,10methylenetetrahydrofolate (5,10-methyleneTHF), interferes with thymidylate biosynthesis, leading to accumulation of deoxyuridylate in DNA (31). Removal of this abnormal base might labilize DNA to strands breaks (32) and chromosome breaks appear to be important in nearly all human cancers (33.). Finally, our finding that a recently described CYP1A1 genetic polymorphism is associated with endometrial cancer risk, could be related with the known involvement of estrogens in both initiation and promotion of endometrial cancer. In premenopausal women, persistent anovulation due to the polycystic ovary syndrome (34) and ovarian neoplasia (granulosa cell tumor, thecal cell tumor or adrenocortical hyperplasia) often causes an estrogen-predominant milieu associated with the occurrence of endometrial cancers; in addition, adipose tissues contribute to the formation of extraglandular estrogen (mainly estrone), which is relevant for tumor risk factor, especially in perimenopause (35). Due to 2309

M.Esteller et al.

the fact that estrogen metabolism is partially determined by cytochrome P450 activity and is under the genetic control of both the CYP1A1 and CYP1A2 genes, the new CYP1A1 genetic polymorphism studied may influence the production of estrogen 2-hydroxylated metabolites (36) and therefore individual susceptibility to endometrial cancer. In a parallel way, the individuals who inherit the rare CYP1A1 genotypes could suffer alterations in the metabolism of polycyclic aromatic hydrocarbons (PAHs) (known human carcinogens found in tobacco, smoke, food and combustion fumes) to reactive intermediates with mutagenic activity (37), contributing perhaps also to the higher endometrial cancer risk detected. In this sense, PAHs have a high capacity for adduct formation in human cancer cells (38) and benzo[a]pyreneDNA adducts have been identified in human endometrium (39). In addition, PAHs metabolites have been shown to activate the H-ras protooncogene in vitro (40), a gene closely related to K-ras, which is found to be mutated in about one quarter of all sporadic endometrial carcinomas (41) and which is thought to be altered early in the progression from endometrial hyperplasia to endometrial carcinoma (42,43). Finally, we should point out that PAHs are lipophilic and stored in adipose tissue, and obesity is an epidemiological classic risk factor for endometrial carcinoma (44). In conclusion, our preliminary data are in agreement with a genetic susceptibility to endometrial cancer associated with a MTHFR germ line polymorphism and a recently described rare CYP1A1 variant. In addition to suggest the contribution of polymorphisms in DNA methylation-related genes and carcinogen-metabolism genes to an enhanced cancer risk, this study could provide a link with the epidemiological association between estrogen exposure, obesity and endometrial cancer. Therefore, studies in larger populations of sporadic endometrial cancer cases, in families with endometrial cancer aggregation with different phenotypes, and in functional assesment of the genotypes described, are in progress. Acknowledgements This work was supported in part by the Institut Catala de la Salut and Fondo de Investigaciones Sanitarias (Grant FIS 95/0501). Manel Esteller is a fellow of Spanish Ministerio de Educacion y Ciencia, Universitat Rovira i Virgili. We thank Dr Lluis Armadans for his help with the statistical analysis of the data and Mrs T.Berry for correction of the manuscript.

References 1. Marra,G. and Boland,C.R. (1995) Hereditary nonpolyposis colorectal cancer: the syndrome, the genes and historical perspectives. J. Natl Cancer Inst., 87, 11–25. 2. Blackwood,M.A. and Weber,B.L. (1996) Recent advances in breast cancer biology. Curr. Opin. Oncol., 8, 449–454. 3. Tomlinson,I.P., Neale,K., Talbot,I.C., Spigelman,A.D., Williams,C.B., Phillips,R.K. and Bodmer,W.F. (1996) A modifying locus for familial adenomatous polyposis may be present on chromosome 1p35-p36. J. Med. Genet., 33, 268–273. 4. Raunio,H., Husgafvel-Pursiainen,K., Anttila,S., Hietanen,E., Hirvonen,A. and Pelkonen,O. (1995) Diagnosis of polymorphisms in carcinogenactivating and inactivating enzymes and cancer susceptibility—a review. Gene, 159, 113–121. 5. Kawajiri,K., Nakachi,K., Imai,K., Yoshii,A., Shinoda,N. and Watanabe,J. (1990) Identification of genetically high risk individuals to lung cancer by DNA polymorphisms of the cytochrome P4501A1 gene. FEBS Lett., 263, 131–133. 6. Hayashi,S., Watanabe,J., Nakachi,K. and Kawajiri,K. (1991) Genetic linkage of lung cancer-associated Msp I polymorphism with amino acid replacement in the heme binding region of the human cytochrome P4501A1 gene. J. Biochem., 110, 407–411. 7. Crofts,F., Cosma,G., Currie,D., Taioili,E., Toniolo,P. and Garte,S.J. (1993)

2310

A novel CYP1A1 gene polymorphism in African-Americans. Carcinogenesis, 14, 829–831. 8. Nakachi,K., Imai,K., Hayashi.,S., Watanabe,J. and Kawajiri,K. (1991) Genetic susceptibility to squamous cell carcinoma of the lung in relation to cigarette smoking dose. Cancer Res., 51, 5177–5180. 9. Ambrosone,C.B., Freudenheim,J.L., Graham,S. et al. (1995) Cytochrome P4501A1 and glutathione S-transferase (M1) genetic polymorphisms and postmenopausal breast cancer risk. Cancer Res., 55, 3483–3485. 10. Sivaraman,L., Leatham,M.P., Yee,J., Wilkens,L.R., Lau,A.F. and Marchand,L. (1994) CYP1A1 genetic polymorphisms and in situ colorectal cancer. Cancer Res., 54, 3692–3695. 11. Reebeck,T.R., Rosvold,E.A., Duggan,D.J., Zhang,J. and Buetow,K.H. (1994) Genetics of CYP1A1: coamplification of specific alleles by polymerase chain reaction and association with breast cancer. Cancer Epid. Biomarkers Prev., 3, 511–514. 12. Taioli,E., Trachman,J., Chen,X., Toniolo,P. and Garte,S.J. (1995) A CYP1A1 restriction fragment length polymorphism is associated with breast cancer in African-American women. Cancer Res., 55, 3757–3758. 13. Esteller,M., Garcia,A., Martinez-Palones,J.M., Xercavins,J. and Reventos,J. (1997) Susceptibility to endometrial cancer: influence of allelism at p53, glutathione S-transferase (GSTM1 and GSTT1) and cytochrome P-450 (CYP1A1) loci. Br. J. Cancer, 75, 1385–1388. 14. Cascorbi,I., Brockmoller,J. and Roots,I. (1996) A C4887A polymorphism in exon 7 of human CYP1A1: population frequency, mutation linkages, and impact on lung cancer susceptibility. Cancer Res., 56, 4965–4969. 15. Kawajiri,K., Nakachi,K., Imai,K., Watanabe,J. and Hayashi,S. (1993) The CYP1A1 gene and cancer susceptibility. Crit. Rev. Oncol. Hematol., 14, 77–87. 16. Krontiris,T., Devlin,B., Karp,D., Robert,N. and Risch,N. (1993) An association between the risk of cancer and mutations in the HRAS1 minisatellite locus. N. Engl. J. Med., 329, 517–523. 17. Runnebaum,I.B., Tong,X.W., Konig,R., Hong,Z., Korner,K., Atkinson,E.N., Kreienberg,R. and Kieback,D.G. (1995) p53-based blood test for p53PIN3 and risk for sporadic ovarian cancer. Lancet, 345, 994. 18. Phelan,C.M., Rebbeck,T.R., Weber B.L. et al. (1996) Ovarian cancer risk in BRCA1 carriers is modified by the HRAS1 variable number of tandem repeats (VNTR) locus. Nat. Genet., 12, 309–312. 19. Greenblatt,M.S., Bennett,W.P., Hollstein,M. and Harris,C.C. (1994) Mutations in the p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res., 54, 4855–4878. 20. James,S.J., Basnakian,A.G. and Miller,B.J. (1994) In vitro folate deficiency induces deoxynucleotide pool imbalance, apoptosis, and mutagenesis in Chinese hamster ovary cells. Cancer Res., 54, 5075–5080. 21. Blount,B.C. and Ames,B.N. (1995) DNA damage in folate deficiency. Baillieres Clin. Haematol., 8, 461–478. 22. Chen,J., Giovannucci,E., Kelsey,K., Rimm,E.B., Stampfer,M.J., Colditz,G.A., Spiegelman,D., Willet,W.C. and Hunter,D.J. (1996) A methylenetetrahydrofolate reductase polymorphism and the risk of colorectal cancer. Cancer Res., 56, 4862–4864. 23. Esteller,M., Garcia,A., Martinez-Palones,J.M., Xercavins,J. and Reventos,J. (1995) Detection of c-erbB-2/neu and fibroblast growth factor-3/int-2 but not epidermal growth factor receptor gene amplification in endometrial cancer by differential polymerase chain reaction. Cancer, 75, 2139–2145. 24. Frosst,P., Blom,H.J., Milos,R. et al. (1995) A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat. Genet., 10, 111–113. 25. Schildkraut,J.M., Risch,N. and Thompson,W.D. (1989) Evaluating genetic association among ovarian, breast and endometrial cancer: evidence for a breast/ovarian cancer relationship. Am. J. Hum. Genet., 45, 521–529. 26. Selhub,J. and Miller,J.W. (1992) The pathogenesis of homocysteinemia: interruption of the coordinate regulation by S-adenosylmethionine of the remethylation and transsulfuration of homocysteine. Am. J. Clin. Nutr., 55, 131–138. 27. Laird,P.W. and Jaenisch,R. (1996) The role of DNA methylation in cancer genetics and epigenetics. Annu. Rev. Genet., 30, 441–464. 28. Ben-Baruch,G., Solomon,R., Menczer,J. and Kloog,Y. (1988) High levels of protein carboxyl methyltransferase in well-differentiated human endometrial carcinoma. FEBS Lett., 228, 277–280. 29. Kleinz,Z., Ben-Baruch,G., Marciano,D., Solomon,R., Altaras,M. and Kloog,Y. (1994) Characterization of the prenylated protein methyltransferase in human endometrial carcinoma. Biochim. Biophys. Acta, 3, 330–336. 30. Piva,R., Kumar,V.L., Hanau,S., Rimondi,A.P., Pansini,S., Mollica,G. and del Senno,L. (1989) Abnormal methylation of estrogen receptor gene and reduced estrogen receptor RNA levels in human endometrial carcinomas. J. Steroid. Biochem., 32, 1–4.

CYP1A1 and MTHFR genes and endometrial cancer 31. Blount,B.C. and Ames,B.N. (1994) Analysis of uracil in DNA by gas chromatography-mass spectrometry. Anal. Biochem., 219, 195–200. 32. Reidy,J.A. (1988) Role of deoxyuridine incorporation and DNA repair in the expression of human chromosomal fragile sites. Mutat. Res., 200, 215–220. 33. Weinberg,R.A. (1988) The genetic origins of human cancer. Cancer, 61, 1963–1968. 34. Nisker,J.A., Ramzy,I. and Collins,J.A. (1978) Adenocarcinoma of the endometrium and abnormal ovarian function in young women. Am. J. Obstet. Gynecol., 130, 9–13. 35. Lippman.,M.E. and Swain,S.M. (1992) Endocrine-responsive cancers of human. In Wilson,D.J. and Foster,D.W. (eds) Textbook of Endocrinology, W.B.Saunders, Philadelphia, pp. 1577–1598. 36. Schneider,J., Huh,M.M., Bradlow,L.H. and Fishman,J. (1984) Antiestrogen action of 2-hydroxyestrone on MCF-7 human breast cancer cells. J. Biol. Chem., 259, 4840–4845. 37. Nebert,D.W. (1991) Role of genetics and drug metabolism in human cancer risk. Mutat. Res., 247, 267–281. 38. Calaf,G. and Russo,J. (1993) Transformation of human breast epithelial cells by chemical carcinogens. Carcinogenesis, 15, 483–492. 39. Kulkarni,M.S., Calloway,K., Irigaray,M.F. and Kaufman,D.G. (1986) Species differences in the formation of benzo(a)pyrene-DNA adducts in rodent and human endometrium. Cancer Res., 46, 2888–2891. 40. Marshall,C.J., Vousden,K.H. and Phillips,D.H. (1984) Activation of c-Haras-1 proto-oncogene by in vitro modification with a chemical carcinogen, benzo(a)pyrene diol-epoxide. Nature, 310, 586–589. 41. Berchuck,A. and Boyd,J. (1995) Molecular basis of endometrial cancer. Cancer, 76, 2034–2040. 42. Sasaki,H., Nishii,H., Takahashi,H., Furusato,M., Terashima,Y. and Siegal,G.P. (1993) Mutation of the Ki-ras protooncogene in human endometrial hyperplasia and carcinoma. Cancer Res., 53, 1906–1910. 43. Esteller,M., Garcia,A., Martinez-Palones,J.M., Xercavins,J. and Reventos,J. (1997) Detection of clonality and genetic alterations in endometrial pipelle biopsy and its surgical specimen counterpart. Lab. Invest., 76, 109–116. 44. Kurman,R.J., Zaino,R.J. and Norris, H.J. (1994) Endometrial carcinoma. In Kurman,R.J. (ed) Blaustein’s Pathology of the Female Genital Tract. Springer-Verlag, New York, pp. 439–486. Received on March 21, 1997; revised on July 4, 1997; accepted on August 19, 1997

2311