Genetic Polymorphisms of CYP2E1 and Risk of Colorectal Cancer: The Fukuoka Colorectal Cancer Study Free

The hepatic cytochrome P450 2E1 (CYP2E1) enzyme plays an important role in the metabolic activation of nitrosamines and other carcinogenic compounds (1, 2). Of the many known genetic polymorphisms in the CYP2E1 gene, the RsaI variant corresponding to a C-1054T substitution (rs 2031920) and the 96-bp insertion in the 5′-flanking region have drawn much interest with regard to having potential functionality. According to the conventional nomenclature, the RsaI wild-type allele (commonly called c1 allele) and the variant c2 allele correspond to CYP2E1*5A and CYP2E1*5B, respectively. The insertion allele is named CYP2E1*1D, whereas the noninsertion allele is CYP2E1*1C. In vitro studies have suggested an enhanced transcription of the CYP2E1 gene associated with the variant c2 allele (3, 4). In contrast, some human studies showed that individuals with the variant c2 allele had lower activity of the enzyme (5) and lower inducibility by chronic alcohol ingestion (6), whereas others observed no clear variation in the activity of the enzyme according to RsaI genotypes (7-9). Similarly, a greater transcriptional activity has been shown for the variant 96-bp insertion allele (10), with individuals with the insertion showing an increased induction of the enzyme in relation to obesity and high alcohol consumption (11).

Few studies have assessed the association between the RsaI polymorphism and colorectal cancer risk, and their findings are rather inconsistent. Individuals with the RsaI c2 allele tended to have decreased risks of colon and rectal cancers in a population-based study in Hawaii (12), whereas a statistically significant increase in colorectal cancer risk was reported for individuals with the RsaI c2 allele in Hungary (13) and for those homozygous for the c2 allele in China (14). No association was detected between the RsaI polymorphism and the risk of colorectal cancer in Australia (15) and The Netherlands (16). The RsaI c2 allele is rather rare in Caucasians, and results from studies of Caucasians may have been subject to chance. Only one previous study, the case-control study in Hawaii (12), examined the relation between the 96-bp insertion polymorphism and colorectal cancer. This study showed a statistically significantly increased risk of rectal cancer, but not of colon cancer, for individuals harboring at least one allele of the 96-bp insertion. In a recent Japanese study on colorectal adenomas (17), the 96-bp insertion allele was associated with an increased prevalence odds of large adenomas (≥5 mm in size), but not of small adenomas (<5 mm), whereas RsaI c2 allele did not show a clear association with either large or small adenomas.

We investigated the association of CYP2E1 RsaI and 96-bp insertion polymorphisms with colorectal cancer risk in the Fukuoka Colorectal Cancer Study, a community-based case-control study in Japan (18), with particular attention to the effect of these polymorphisms on the relation of alcohol use, obesity, and red meat intake to colorectal cancer. These lifestyle factors have consistently been related to increased risk of colorectal cancer (19, 20). CYP2E1 is induced by alcohol drinking and obesity (21-23), and the induction may vary by the RsaI (6) and 96-bp insertion polymorphism (11). A possible interaction between the 96-bp insertion polymorphism and red meat intake has been suggested (12).

The Fukuoka Colorectal Cancer Study aims to examine the relation of lifestyle factors and genetic susceptibility to the risk of colorectal cancer among residents living in Fukuoka City and three adjacent areas. The study protocol was approved by the ethics committees of the Kyushu University Faculty of Medical Sciences and the participating hospitals. Details of the methods have been reported previously (18).

Cases comprised a consecutive series of patients with histologically confirmed incident colorectal adenocarcinomas who were admitted to two university hospitals or six affiliated hospitals for surgical treatment during the period October 2000 to December 2003. Eligible cases were ages 20 to 74 years at time of diagnosis; lived in the study area; had no prior history of partial or total removal of the colorectum, familial adenomatous polyposis, or inflammatory bowel disease; and were mentally competent to give informed consent and to complete the interview. Of the total 1,053 eligible cases, 840 (80%) cases participated in the interview, and 685 gave informed consent for the genotyping. Cases of colon and rectal cancers numbered 384 and 290, respectively; the remaining 11 cases had cancers at both sites.

Eligibility criteria for controls were the same as described for the cases, except for two conditions: no prior diagnosis of colorectal cancer and ages 20 to 74 years at the time of selection. A total of 1,500 persons living in 15 geographic areas were selected as control candidates by a two-stage random sampling and were invited to participate in the study by mail. Of these, 833 persons participated in the survey, and 778 gave an informed consent for genotyping. The participation rate for the interview was 60% (833 of 1,382), with 118 persons excluded in the denominator for the following reasons: death (n = 7), migration from the study area (n = 22), undelivered mail (n = 44), mental incompetence (n = 19), history of partial or total removal of the colorectum (n = 21), and diagnosis of colorectal cancer after the survey (n = 5).

Research nurses interviewed cases and controls in person regarding smoking, alcohol use, physical activity, and other factors using a standardized questionnaire. The index dates were the date of onset of symptom or screening for cases and the time of interview for controls. Anthropometric questions inquired about height (cm), and body weight (kg) currently and 10 years earlier. Body mass index (BMI; kg/m²) 10 years earlier was used in the analysis because current BMI was unrelated to risk (24). Body weight 10 years earlier was not ascertained from 2 cases and 10 controls and was substituted with the current body weight. Habitual alcohol consumption 5 years before the index date was ascertained. Individuals reported the average number of days per week that alcohol was consumed and the average amount of alcohol per day of drinking. The amount of alcohol was expressed using the conventional Japanese units; one go (180 mL) of sake, one large bottle (633 mL) of beer, and half a go (90 mL) of shochu were each expressed as one unit, and one drink (30 mL) of whisky or brandy and one glass (100 mL) of wine were each converted to half a unit. The reproducibility of the questionnaire was tested on 29 control subjects (14 men and 15 women) with an interval of ∼1 year, and the reported alcohol intake was highly reproducible (Spearman's r = 0.82).

With regard to smoking, ever-smokers were asked about duration of smoking in years and numbers of cigarettes smoked per day for each decade of life from the second to the last decade. We calculated the cumulative exposure to cigarette smoking until the beginning of the previous decade of age. Questions on physical activities elicited type of job (sedentary or standing work, work with walking, labor work, hard labor work, and no job), activities in commuting and housework shopping, and leisure time activities 5 years before. As described in detail previously (24), leisure time physical activity (including activities in commuting and housework shopping) was expressed as a sum of metabolic equivalents (MET) multiplied by hours of weekly participation in each activity (in MET-h/wk).

Consumption frequencies and portion sizes of 148 food/beverage items were ascertained by computer-assisted interview, and quantitative intake estimates were obtained for selected nutrients and food groups (25). Individuals were asked to report their usual consumption over the 1 year before onset of symptoms for cases or interview for controls. In a validation study in 28 control study participants who recorded their diets over a period of 7 days in four consecutive seasons, Pearson correlation coefficients of energy-adjusted intakes of beef/pork and processed meat (ham/sausages) assessed by the food frequency questionnaire and diet record were 0.70 and 0.57, respectively. Beef/pork and processed meat were combined as red meat in the present study. The consumption of mutton is rare in Japan, and it was not included in the dietary interview.

DNA was extracted from the buffy coat using a commercial kit (Qiagen). The following genotyping procedures used 1 μL template DNA with a concentration of ∼50 to 150 ng/μL. Genotyping for the CYP2E1 RsaI polymorphism was carried out by PCR-RFLP as described by Le Marchand et al. (5) using primers 5′-CCAGTCGAGTCTACATTGTCA-3′ (forward) and 5′-TTCATTCTGTCTTCTAACTGG-3′ (reverse). The PCR product of 413 bp was digested with RsaI, resulting in fragments of 352 and 61 bp for the c1 allele. The 96-bp insertion genotype was determined by the PCR method as described by Fritsche et al. (26) using primers 5′-GTGATGGAAGCCTGAAGAACA-3′ (forward) and 5′-CTTTGGTGGGGTGAGAACAG-3′ (reverse). The PCR product of the 96-bp insertion allele is 729 bp in length, and the noninsertion allele is 633 bp in length. Details of the PCR conditions and electrophoresis have been described previously (17).

Comparison of means, medians, and proportions between cases and controls was done by t test, Wilcoxon rank-sum test, and χ² test, respectively. The association of CYP2E1 polymorphisms with colorectal cancer was examined in terms of odds ratio (OR) and 95% confidence interval (95% CI), which were obtained from a logistic regression analysis. The 95% CI was derived from the SE of the logistic regression coefficient. Statistical adjustment was made for 5-year age class (starting with the lowest class of <45 years), sex, residency area (Fukuoka City or the adjacent areas), BMI 10 years ago (<22.5, 22.5-24.9, 25.0-27.4, or ≥27.5 kg/m²), smoking (0, 1-399, 400-799, or ≥800 cigarettes per year), alcohol intake (0, 0.1-0.9, 1.0-1.9, or ≥2.0 units/d), type of job (sedentary work and no job, work with walking, or labor work), leisure time physical activity (0, 1-15.9, or ≥16 MET-h/wk), and parental history of colorectal cancer. Possible gene and environment interactions were evaluated by examining the association with alcohol use, BMI, and red meat intake according to genotypes. In this analysis, alcohol consumption was categorized into 0, 0.1 to 1.9, and ≥2 units/d, BMI into <22.5, 22.5 to 24.9, and ≥25.0 kg/m², and red meat intake into three levels using tertile cutpoints based on the distribution of cases and controls combined. Regarding red meat consumption, 4 cases and 2 controls with total calorie intake >5,000 kcal/d were excluded, and the intake was adjusted to 2,000 kcal/d by the regression residual method. Trends in the OR were evaluated with ordinal values assigned to levels of the variable under study. Interactions between genotype and lifestyle variables were statistically tested using the likelihood ratio test, comparing a model including interaction terms with one that only included the main effects. Ordinal variables were used for the lifestyle variables in the statistical test for interaction.

Deviation from the Hardy-Weinberg equilibrium was evaluated by χ² test with 1 df. Lewontin's D′ was calculated for assessing linkage disequilibrium based on the maximum likelihood estimates of the haplotype frequencies. Statistical significance was declared if a two-sided P < 0.05 or if the 95% CI did not include unity. Statistical analyses were carried out using SAS version 8.2 (SAS Institute), except for the analyses of Hardy-Weinberg equilibrium and linkage disequilibrium, which used SAS/Genetics 9.1 (SAS Institute).

Selected characteristics of colorectal cancer cases and controls are shown in Table 1. Although the two groups did not differ by sex and residence area, cases were slightly older than controls. Alcohol consumption and BMI 10 years earlier were greater in the cases as was the frequency of parental history of colorectal caner. There was no significant difference between cases and controls in cigarette smoking, physical activity, and red meat intake.

The frequency of the RsaI c2 allele was 0.226 in cases and 0.236 in controls. The corresponding frequencies for the 96-bp insertion allele were 0.245 and 0.217, respectively. The distributions of the CYP2E1 RsaI and 96-bp insertion genotypes were each in agreement with Hardy-Weinberg equilibrium in both cases and controls (RsaI: P = 0.80 for cases and 0.89 for controls; 96-bp insertion: P = 0.77 for cases and 0.25 for controls). The two polymorphisms were in almost complete linkage disequilibrium with a Lewontin's D′ of 0.943.

No association was observed with colorectal cancer for the RsaI c1/c2 and c2/c2 genotypes compared with the c1/c1 genotype (Table 2). However, a statistically significant increase in the adjusted OR was noted for individuals homozygous for the 96-bp insertion allele compared with those having no insertion.

Associations with these two polymorphisms differed by location of the colorectal cancer (Table 2). The c2 allele was associated with a significant decrease in the risk of rectal cancer, with the OR progressively decreasing with increasing number of c2 allele (P_trend = 0.02). As for the 96-bp insertion polymorphism, a statistically significant increase in the OR for rectal cancer was seen among individuals with one or two copies of the insert. In addition, a statistically significant 2.28-fold increase in the OR for colon cancer was observed among those having two alleles of the 96-bp insertion.

Table 3 shows the association between alcohol use and colorectal cancer for individuals without and with RsaI c2 allele as well as for those without and with 96-bp insertion allele. Statistically significant positive associations with alcohol consumption were observed for colorectal and colon cancers in those with the RsaI c1/c1 genotype and in those without the insertion allele, although the interaction test failed to reach the statistical significance. There was also a suggestion that an association between alcohol and rectal cancer was only seen for subjects with the c1/c1 genotype or the noninsertion genotype. The association between BMI and colorectal cancer did not differ by RsaI or 96-bp insertion polymorphism genotype whether the analysis was done for colorectal cancer as a whole or subsite-specific cancer (data not shown).

Red meat intake showed no clear association with overall or subsite-specific colorectal cancer risk whether individuals were classified by the presence of RsaI c2 allele or 96-bp insertion allele (Table 4). However, there was a statistically significant interaction between red meat intake and the insertion allele on the risk of colon cancer (P = 0.03).

The present study showed a modest, statistically significant decrease in the risk of rectal cancer in individuals harboring the RsaI c2 allele and an increased risk of rectal cancer associated with 96-bp insertion allele. Individuals homozygous for the 96-bp insertion allele were at increased risks of colorectal and colon cancers. Another notable finding was that increased risks of colorectal cancer with alcohol use and red meat intake were observed for specific genotypes of the two polymorphisms. Alcohol was associated with an increased risk of colorectal cancer among individuals without the variant c2 allele and in those with no insertion allele, whereas red meat seemed to be related to an increased risk of colon cancer in subjects with one or two insertion alleles.

Only one other case-control study has examined the relation of RsaI polymorphism to colon and rectal cancer separately (12). In that study from Hawaii, the OR of rectal cancer for the c1/c2 and c2/c2 genotypes versus c1/c1 genotype were 0.9 (95% CI, 0.6-1.3) and 0.4 (95% CI, 0.1-1.3), respectively, and the OR for the combined genotype was 0.8 (95% CI, 0.6-1.3). The OR of colon cancer for the c1/c2, c2/c2, and combined genotypes were 0.8 (95% CI, 0.6-1.1), 1.2 (95% CI, 0.6-2.2), and 0.8 (95% CI, 0.6-1.1), respectively. Furthermore, the study in Hawaii first addressed the relation between the 96-bp insertion polymorphism and colorectal cancer and reported an increased risk of rectal cancer, but not of colon cancer, associated with this polymorphism; the reported OR for those with at least one 96-bp insertion allele were 1.6 (95% CI, 1.1-2.5) for rectal cancer and 1.1 (95% CI, 0.8-1.5) for colon cancer (12). Thus, the present findings for rectal cancer are consistent with the results from Hawaii with respect to both RsaI and the 96-bp insertion polymorphisms. It is not clear why the findings on the 96-bp insertion polymorphism and colon cancer were somewhat different between the two studies. Moreover, 96-bp insertion allele was reported to be associated with an increased risk of large (≥5 mm), but not of small, colorectal adenomas in a study of Japanese men (17). The past findings for adenomas are also consistent with the present study.

Alcohol consumption has consistently been related to increased risk of colorectal cancer (19, 27), but the underlying mechanisms remain uncertain. In the present study, an increased risk of colorectal cancer associated with high intake of alcohol was limited to individuals having no RsaI c2 or 96-bp insertion variant allele. CYP2E1 is known to be involved in the metabolism of alcohol (1). However, the observed genotype-specific association between alcohol and colorectal cancer is difficult to interpret. Whereas RsaI c2 allele seems to be associated with lower activity and lower alcohol-related inducibility of the enzyme (5, 6), alcohol-related induction of the enzyme was reported to be greater in individuals with the 96-bp insertion allele (11). CYP2E1 is predominantly expressed in the liver especially after alcohol ingestion (1), but the enzyme is also detected in human colon mucosa (28) and induced in the colon of rodents administered with alcohol (29). Additional studies on the mechanisms by which these two polymorphisms may modify enzyme activity and inducibility are clearly needed.

We reported previously that red meat intake was unrelated to the risk of colorectal cancer in the Fukuoka Colorectal Cancer Study (30). The present findings indicated that high intake of red meat was related to increased risk of colon cancer in individuals with the 96-bp insertion polymorphism. In the Hawaii study (12), an increased risk of rectal cancer, rather than colon cancer, associated with the 96-bp insertion allele was more evident in individuals with a high consumption of red meat. Higher transcriptional activity has been reported for the 96-bp insertion allele (10), and this allele may be associated with increased bioactivity of meat carcinogens. The present findings add to evidence that N-nitroso compounds bioactivated by CYP2E1 may be an underlying cause for the increased risk of colorectal cancer observed with high intake of red meat. Red meat has been shown to increase local production of N-nitroso compounds in the large bowel (31, 32).

The use of community controls and the large number of subjects are strengths of the present study. Both the RsaI c2 allele and the 96-bp insertion allele are fairly common in Asian populations compared with Caucasians. In the controls of the present study, the RsaI c2 and 96-bp insertion alleles were present at frequencies of 24% and 22%, respectively. These frequencies are very similar to those reported previously for Japanese elsewhere. Frequencies of the RsaI c2 allele were 23% in Japanese, 4% in Caucasians, and 15% in Hawaiians in Hawaii (12). The frequency for the 96-bp insertion has been reported to be 15% in Taiwanese, 10% in African Americans, and 2% in Caucasians (26). The frequency of the 96-bp insertion allele was reported to be 23% in Hawaiian Japanese (12).

A limitation of the present study was that the participation rate in terms of the genotyping was not optimal. However, the participation was relatively similar in cases (65%) and controls (56%), and it appears unlikely that participation would be associated with genotype and thus that selection bias occurred with respect to the CYP2E1 polymorphisms. As reported previously (33), although older persons and women were less likely to give consent for the genotyping, there was no difference between those who gave consent and those who did not in terms of residency area, smoking, and alcohol use. Another limitation was the retrospective assessment of alcohol consumption and red meat intake. We assessed alcohol consumption 5 years before. We have no data as to how valid the recalled information of past alcohol consumption was in the study, although it was found to be highly reproducible (see above). Consumption of red meat during the previous year may not have reflected the consumption during the period relevant to the development of colorectal cancer, which can be as long as 30 years. Finally, although the total number of subjects was fairly large, the sample size was much smaller in the interaction analyses by alcohol and red meat. Thus, statistical power was necessarily low in these analyses. For example, with a two-sided significance level of 0.05, the power of detecting an 1.5-fold increase in the OR of colorectal cancer for those with the variant allele versus those without was >95%, but the power ranged from 46% to 63% for the comparison between the highest and the lowest categories of alcohol use and red meat intake in the interaction analyses.

In summary, this large case-control study in Japan showed that the risk of rectal cancer was decreased in individuals with the CYP2E1 RsaI c2 allele and was increased in individuals with the CYP2E1 96-bp insertion allele. Individuals homozygous for the 96-bp insertion polymorphism were also at increased risk for colorectal and colon cancer. Increased risks of colorectal cancer associated with alcohol consumption and red meat were limited to specific genotypes of the two polymorphisms. These findings suggest that variation in activity and inducibility of CYP2E1 may contribute to the development of colorectal cancer and add to the evidence that compounds metabolized by CYP2E1, such as nitrosamines and alcohol, may be involved in colorectal carcinogenesis.

No potential conflicts of interest were disclosed.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Emeritus Professor Keizo Sugimachi, Profs. Seiyo Ikeda, Takayuki Shirakusa, and Sumitaka Arima, and Drs. Motonori Saku, Yoichi Ikeda, Soichiro Maekawa, Kazuo Tanoue, Kinjiro Sumiyoshi, and Shoichiro Saito for support in conducting the survey of cases; the following physicians who kindly supervised the survey of controls at their clinics: Drs. Hideaki Baba, Tomonori Endo, Hiroshi Hara, Yoichiro Hirokata, Motohisa Ikeda, Masayoshi Ishibashi, Fumiaki Itoh, Yasuhiro Iwanaga, Hideki Kaku, Shoshi Kaku, Minoru Kanazawa, Akira Kobayashi, Ryunosuke Kumashiro, Shinichi Matsumoto, Soukei Mioka, Umeji Miyakoda, Osamu Nakagaki, Nobuyoshi Nogawa, Nobuyuki Ogami, Toyoaki Okabayashi, Hironao Okabe, Nishiki Saku, Masafumi Tanaka, Masahiro Ueda, Bunichi Ushio, and Koheisho Yasunaga; and Hiroko Mizuta, Masumi Koga, and Kumiko Arie for technical assistance.