2016 Volume 39 Issue 1 Pages 84-89
Several of the procarcinogens inhaled in tobacco smoke, the primary risk factor for bladder cancer, are activated by CYP2A6. The association between the whole-gene deletion of CYP2A6 (CYP2A6*4) and a reduced risk of bladder cancer was suggested in Chinese Han smokers. However, there is no evidence for association between the risk of bladder cancer and CYP2A6 genotypes in the Japanese population. Using genomic DNA from smokers of the Japanese population (163 bladder cancer patients and 116 controls), we conducted a case-control study to assess the association between CYP2A6 polymorphisms and the risk of bladder cancer. Determination of CYP2A6 genotypes was carried out by amplifying each exon of CYP2A6 using polymerase chain reaction (PCR) and Sanger sequencing. The CYP2A6*4 allele was identified by an allele-specific PCR assay. Bladder cancer risk was evaluated using the activity score (AS) system based on CYP2A6 genotypes. The odds ratios (95% confidence interval) for the AS 0, AS 0.5, AS 1.0, and AS 1.5 groups were 0.46 (0.12–1.83), 0.43 (0.15–1.25), 0.86 (0.40–1.86), and 1.36 (0.60–3.06), respectively. In conclusion, although decreased CYP2A6 AS tended to reduce the risk of bladder cancer in Japanese smokers, no significant association was recognized in this population. However, given the relatively small size of the sample, further study is required to conclude the lack of a statistically significant association between CYP2A6 genotypes and the risk of bladder cancer.
Bladder cancer is one of the most common cancers worldwide, with an estimated 386300 new cases and 150200 deaths reported in 2008.1) The age-adjusted bladder cancer mortality rates of Japanese male and female patients in 2013 were approximately 3.7 per 100000 and 1.0 per 100000, respectively.2) Bladder cancer is much more common among men, and age-standardized bladder cancer death rates have increased in recent years.3) Furthermore, ethnicity and smoking are known to play a role in bladder cancer development.4) In particular, environmental exposure to tobacco smoke is a primary risk factor for bladder cancer, and smokers are at least three times as likely to develop bladder cancer as non-smokers.5,6) When smokers inhale the procarcinogens contained in tobacco smoke, some of these get absorbed from the lungs and enter the bloodstream. A large number of these procarcinogens are activated by several enzymes in the liver and extrahepatic tissues.7,8) From the blood, the carcinogens activated by enzymes are filtered by the kidneys and concentrated in the urine where they induce damage to the urothelial cells, which increases the chance of developing bladder cancer.4)
Human CYP2A6 is a phase I enzyme responsible for the metabolism of drugs and chemical compounds such as tegafur, nicotine, and coumarin.9) CYP2A6 is also an important enzyme involved in the metabolic activation of procarcinogens such as 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and N-nitrosonornicotine in tobacco smoke.7,8) Thus, it is reasonable to hypothesize that CYP2A6 activity may be related to the susceptibility of developing bladder cancer among smokers.
Thus far, numerous allelic variants of CYP2A6 have been identified, and most of these are derived from single nucleotide polymorphisms (SNPs) in coding and regulatory regions (http://www.cypalleles.ki.se/cyp2a6.htm). Of these, 23 variants (CYP2A6*6, CYP2A6*7, CYP2A6*9–CYP2A6*13, CYP2A6*15, CYP2A6*17–CYP2A6*19, CYP2A6*22–CYP2A6*24, CYP2A6*26, CYP2A6*27, CYP2A6*35, CYP2A6*39, CYP2A6*40, and CYP2A6*42–CYP2A6*45) are currently known to lead to reduced enzymatic activities, and 5 variants (CYP2A6*2, CYP2A6*4, CYP2A6*5, CYP2A6*20, and CYP2A6*41) for which CYP2A6 enzymatic activities were shown to be completely abolished.9–13) The whole-gene deletion variant CYP2A6*4 is one of the most common variants in the Japanese, Korean, and Chinese populations, whereas it is a minor variant in French and Brazilian populations.14,15) Overall, two deletion mutants of the CYP2A6, CYP2A6*4A (22.3%) and CYP2A6*4B (0.24%), have been reported in the Japanese population.16–18)
A number of studies have reported an association between CYP2A6 polymorphisms and the risk of various types of cancer.19–24) In particular, the CYP2A6*4 allele has been widely studied with respect to its association with bladder cancer and lung cancer.19–24) Although the association between the risk of bladder cancer and CYP2A6 genotypes in Japanese populations has not been investigated directly, a CYP2A6 gene defect was reported to be associated with a reduced risk of bladder cancer in Chinese smokers.19) The association between genetic polymorphisms of CYP2A6 and the risk of lung cancer has also been reported in different ethnic groups. A study of lung cancer conducted in Japanese subjects suggested that the presence of CYP2A6*4 decreased lung cancer risk, whereas a study conducted in Chinese subjects suggested that the presence of at least one CYP2A6*4 allele increased lung cancer risk.20–24)
We conducted a replicated case-control study to clarify the association between CYP2A6 polymorphisms and the risk of bladder cancer in Japanese smokers, including 163 bladder cancer cases and 116 healthy controls. No non-smokers were included in the study. To achieve a more comprehensive analysis to evaluate the relationship between bladder cancer risk and CYP2A6 activity, we classified all of the CYP2A6 genotypes, including minor alleles, using whole-exon sequencing.
This case-control study consisted of 163 patients with bladder cancer and 116 age- and gender-matched healthy controls. A total of 122 patients with bladder cancer and 38 of the control subjects were recruited from Tohoku University Hospital, Japan. The bladder cancer patients were defined as newly diagnosed from November 2005 to August 2007. An additional 41 patients with bladder cancer, defined as newly diagnosed from June 2009 to July 2014, were treated at Hamamatsu University School of Medicine. The other 78 control subjects were recruited from Tokyo Metropolitan Geriatric Medical Center, Tokyo, Japan. The participants provided written informed consent according to the protocols approved by the ethical review board of Tohoku University Hospital, the Ethics Committees of Tokyo Metropolitan Geriatric Hospital, and Hamamatsu University School of Medicine. Genomic DNA of cases and controls was isolated from peripheral blood samples. In order to achieve a statistical power greater than 80%, the sample size needed in this study to achieve this level of power was estimated using R version 3.0.3 (R core team (2014; R foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org/).
CYP2A6 GenotypingThe presence of CYP2A6*4A was detected by a nested allele-specific polymerase chain reaction (PCR) assay performed according to a previously described method.25,26) CYP2A6*4B genotyping was performed based on a previous report with minor modifications.27) PCR was performed with the genomic DNA samples (>10 ng), 2× AmpliTaq Gold 360 Master Mix (Applied Biosystems, Foster City, CA, U.S.A.), and 0.5 µM each primer (*4B-S, 5′-GCA CAA TAG GGT GAA TGT AGT TAA CA-3′; *4B-AS, 5′-GGA ATA ACT GAA TTT CCT TAA GG-3′). PCR conditions involved an initial denaturation step at 95°C for 10 min, followed by 30 cycles of denaturation at 95°C for 30 s, annealing at 56°C for 30 s, extension at 72°C for 1 min 30 s, and a final extension at 72°C for 7 min. Ten microliters of the PCR product was incubated with 5 U of ClaI (New England Biolabs Japan, Tokyo, Japan) and 5 U of BspHI (New England Biolabs Japan, Tokyo, Japan) in a final volume 20 µL at 37°C ranging from several hours to overnight. The samples were then analyzed on a 1% agarose gel by staining with ethidium bromide. The genotypes were classified as CYP2A6*1 (ClaI+ and BspHI+: 1082, 1066, 701, 195, and 186 bp), CYP2A6*4A (ClaI+ and BspHI+: 1082, 195, and 186 bp), and CYP2A6*4B (ClaI+ and BspHI−: 1080, 894, and 186 bp).
To identify the CYP2A6 haplotypes, Sanger sequencing of the CYP2A6 gene for the amplification of all nine exons and exon/intron junctions was performed according to a previously reported method.25,26,28) We detected the SNP 567C>T of the CYP2A6 gene, which is not registered as a CYP2A6 variant allele in The Human Cytochrome P450 Allele Nomenclature Database, without any other exonic SNPs in the same allele.29) The genes for subjects that were heterozygous for two SNPs (6558T>C, 6600G>T) were TOPO-cloned to determine the haplotype. The fragments amplified by PCR were subcloned into the pENTR/D-TOPO vector (Invitrogen; Carlsbad, CA, U.S.A.). The plasmids were purified and Sanger sequencing was performed.
Statistical AnalysesThe risk of bladder cancer was evaluated using the activity score (AS) system.30) Based on the known enzymatic activity of CYP2A6 variants, a value of 1.0 was assigned to the same degree of enzymatic activity resulting from the CYP2A6*1 allele, 0.5 was assigned to alleles resulting in reduced, but not abolished, enzymatic activity, and 0 was assigned to non-functional alleles. Namely, CYP2A6*4, CYP2A6*5, and CYP2A6*41, which have been reported to abolish the enzymatic activity, was allocated a value of 0.13,31–33) The alleles CYP2A6*7, CYP2A6*9–*11, and CYP2A6*18 were assigned a value of 0.5 based on previous studies.31,34) CYP2A6*13 and CYP2A6*15 were also given a value of 0.5, because Nakajima et al. reported that these alleles decreased enzymatic activity in vivo.10) The SNP 567C>T was assigned a value of 0 due to its mutation at the Arg101Stop codon. CYP2A6 genotypes of each subject were classified into five groups: AS 2.0, AS 1.5, AS 1.0, AS 0.5, and AS 0.
All statistical analyses were performed with IBM SPSS Statistics 22.0 (IBM; New York, NY, U.S.A.). Power calculation was performed using R version 3.0.3. For comparisons with respect to age and smoking behavior (cigarettes per day), the distribution of characteristics was estimated using an unpaired t-test. The chi-squared test was used to compare the sex ratio between cases and controls. Hardy–Weinberg equilibrium in CYP2A6 genotypes was tested using chi-squared test in the cases and controls. The frequency of each genotype in cases and controls was assessed using the chi-squared test. AS values between cases and controls were assessed using binomial logistic regression to calculate the odds ratio (OR) and 95% confidence interval (CI). The genotypes between cases and controls were also evaluated using binomial logistic regression to compare the differences between classification based on the AS system and typing by CYP2A6*4 or non*4. A two-tailed p-value <0.05 indicated statistical significance.
As shown in Table 1, there were no statistically significant differences in the sex ratio between cases and controls (94.5% males in cases versus 91.4% males in controls, p=0.440). The mean ages were 68.4±10.5 and 71.0±12.0 years for the case and control groups, respectively (p=0.052). Smoking behavior variables in the case and control groups were 22.8±12.6 and 23.2±15.4 packs per day, respectively (p=0.782).
| Variable | Cases | Controls | p-Value |
|---|---|---|---|
| Total, n | 163 | 116 | |
| Gender, n (%) | |||
| Male | 154 (94.5) | 106 (91.4) | 0.440a) |
| Female | 9 (5.5) | 10 (8.6) | |
| Mean age±S.D. | 68.4±10.5 | 71.0±12.0 | 0.052b) |
| Cigarettes per day±S.D. | 22.8±12.6 | 23.2±15.4 | 0.782b) |
a) Chi-squared test. b) Unpaired t-test.
In our study, 11 CYP2A6 allelic variants registered on The Human Cytochrome P450 Allele Nomenclature Database and a SNP, 567C>T, were detected. The genotype distributions of CYP2A6 in bladder cancer patients and controls are shown in Table 2, which did not deviate from Hardy–Weinberg equilibrium (χ2=25.06 and 55.19, p=1.000 and 0.467, respectively). The presence of the CYP2A6*4B allele was confirmed whether the fragment of 894 bp in length detected, although we did not detect this fragment in the study population.
| Genotype | Cases (n=163) | Controls (n=116) | Crude ORa) (95%CI) | p-Value |
|---|---|---|---|---|
| *1/*1 | 27 (16.6) | 18 (15.5) | ||
| *1/*4 | 32 (19.7) | 23 (19.8) | 0.93 (0.42–2.07) | 1.000 |
| *1/*7 | 23 (14.2) | 5 (4.3) | 3.07 (0.98–9.55) | 0.085 |
| *1/*9 | 18 (11.1) | 13 (11.2) | 0.92 (0.36–2.34) | 1.000 |
| *1/*10 | 4 (2.5) | 7 (6.0) | 0.38 (0.10–1.49) | 0.282 |
| *1/*11 | 2 (1.2) | 0 (0) | ||
| *1/*13 | 1 (0.6) | 0 (0) | ||
| *1/*15 | 0 (0) | 1 (0.9) | ||
| *1/*18 | 1 (0.6) | 0 (0) | ||
| *1/*41 | 1 (0.6) | 0 (0) | ||
| *1/567C>T | 1 (0.6) | 1 (0.9) | 0.67 (0.04–11.36) | 1.000 |
| *4/*4 | 5 (3.1) | 7 (6.0) | 0.48 (0.13–1.74) | 0.418 |
| *4/*5 | 0 (0) | 1 (0.9) | ||
| *4/*7 | 4 (2.5) | 2 (1.7) | 1.33 (0.22–8.06) | 1.000 |
| *4/*9 | 8 (4.9) | 14 (12.0) | 0.38 (0.13–1.09) | 0.119 |
| *4/*10 | 3 (1.8) | 0 (0) | ||
| *4/*11 | 1 (0.6) | 0 (0) | ||
| *4/*13 | 1 (0.6) | 1 (0.9) | 0.67 (0.04–11.36) | 1.000 |
| *4/*15 | 1 (0.6) | 1 (0.9) | 0.67 (0.04–11.36) | 1.000 |
| *4/*18 | 0 (0) | 1 (0.9) | ||
| *4/567C>T | 1 (0.6) | 0 (0) | ||
| *7/*7 | 2 (1.2) | 1 (0.9) | 1.33 (0.11–15.82) | 1.000 |
| *7/*9 | 14 (8.6) | 8 (6.9) | 1.17 (0.09–3.35) | 0.984 |
| *7/*10 | 3 (1.8) | 0 (0) | ||
| *7/*11 | 2 (1.2) | 2 (1.7) | 0.67 (0.09–5.17) | 1.000 |
| *7/*13 | 0 (0) | 1 (0.9) | ||
| *9/*9 | 3 (1.8) | 9 (7.7) | 0.22 (0.05–0.93) | 0.067 |
| *9/*10 | 2 (1.2) | 0 (0) | ||
| *9/*13 | 2 (1.2) | 0 (0) | ||
| *9/*15 | 1 (0.6) | 0 (0) |
a) Chi-squared test.
We assessed the risk of bladder cancer using the AS system. The predicted CYP2A6 activities were classified as shown in Table 3. The risk of developing bladder cancer was lower in the AS 1.0 group (adjusted OR=0.86, 95% CI=0.40–1.86, p=0.707), AS 0.5 group (adjusted OR=0.43, 95% CI=0.15–1.25, p=0.119), and AS 0 group (adjusted OR=0.46, 95% CI=0.12–1.83, p=0.270). In contrast, the AS 1.5 group showed a higher risk of bladder cancer than the AS 2.0 group (adjusted OR=1.36, 95% CI=0.60–3.06, p=0.463). However, bladder cancer risk did not differ significantly in each AS group when compared to the AS 2.0 group (p>0.05, Table 4). As shown in Table 5, there were no significant differences in the frequency of CYP2A6non*4/*4 (adjusted OR=0.69, 95% CI=0.41–1.36, p=0.167) or CYP2A6*4/*4 (adjusted OR=0.46, 95% CI=0.12–1.83, p=0.144). Therefore, CYP2A6 polymorphisms did not show a significant association with developing bladder cancer in this population of Japanese smokers.
| Activity score (AS) | CYP2A6 genotype |
|---|---|
| AS 2.0 | *1/*1 |
| AS 1.5 | *1/*7, *1/*9, *1/*10, *1/*11, *1/*13, *1/*15, *1/*18 |
| AS 1.0 | *1/*4, *1/*41, *1/567C>T, *7/*7, *7/*9, *7/*10, *7/*11, *7/*13, *9/*9, *9/*10, *9/*13, *9/*15 |
| AS 0.5 | *4/*7, *4/*9, *4/*10, *4/*11, *4/*13, *4/*15, *4/*18 |
| AS 0 | *4/*4, *4/*5, *4/567C>T |
| Group | Cases n (%) n=163 | Controls, n (%) n=116 | OR (95%CI)a) | p-Value | Power (%) |
|---|---|---|---|---|---|
| AS 2.0 | 27 (16.6) | 18 (15.5) | |||
| AS 1.5 | 49 (30.0) | 26 (22.4) | 1.36 (0.60–3.06) | 0.463 | 19.0 |
| AS 1.0 | 63 (38.7) | 45 (38.8) | 0.86 (0.40–1.86) | 0.707 | 9.3 |
| AS 0.5 | 18 (11.0) | 19 (16.4) | 0.43 (0.15–1.25) | 0.119 | 60.4 |
| AS 0 | 6 (3.7) | 8 (6.9) | 0.46 (0.12–1.83) | 0.270 | 29.5 |
a) Logistic regression analysis adjusted by age, gender, and smoking status.
| Group | Cases n (%) n=163 | Controls, n (%) n=116 | OR (95%CI)a) | p-Value | Power (%) |
|---|---|---|---|---|---|
| non*4/non*4 | 108 (66.2) | 67 (57.8) | |||
| non*4/*4 | 50 (30.7) | 42 (36.2) | 0.69 (0.41–1.36) | 0.167 | 30.0 |
| *4/*4 | 5 (3.1) | 7 (6.0) | 0.46 (0.12–1.83) | 0.144 | 26.6 |
a) Logistic regression analysis adjusted by age, gender, and smoking status.
Smoking is recognized as a risk factor for bladder cancer.5) Various CYP enzymes play important roles in metabolically activating procarcinogens, including tobacco smoke.35) In particular, CYP2A6 mainly metabolizes many species of tobacco-specific procarcinogens such as NNK.7,8) In the present study, we investigated the association between CYP2A6 genetic polymorphisms and the risk of developing bladder cancer in Japanese smokers. Our results suggested that CYP2A6 genetic polymorphisms have no significant association with bladder cancer risk in Japanese smokers.
Song et al. first suggested that a defect in the CYP2A6 gene defect may be associated with a reduced risk of bladder cancer in Chinese Han smokers (129 cases and 87 controls), but not non-smokers.19) The association between CYP2A6 genetic polymorphisms and the risk of lung cancer has also been reported in many case-control studies.25) In case-control studies of lung cancer conducted in Japan and Bangladesh, CYP2A6*4 was suggested to reduce the risk of lung cancer by decreasing the level of carcinogens in affected organs, whereas Tan et al. provided evidence that a CYP2A6 gene defect may promote the development of squamous cell lung cancer in the Chinese population.20,21,23,24,36) In other studies conducted in Japan, China, and Canada, lung cancer risk was neither elevated nor diminished with respect to analyses of major mutant alleles specific to each population.37–39) The lack of a consistent relationship between CYP2A6 genotypes and cancer risk may be due to the mix of subjects that are smokers and never smokers, ethnic differences, and insufficient genotyping without consideration of minor alleles. We recently reported that CYP2A6 genetic polymorphisms were related to the susceptibility of squamous cell lung cancer in Japanese smokers using haplotype-based genotyping methods.26) Thus, the degree of association between CYP2A6 genetic polymorphisms and the risk of developing cancer may depend on the ethnicity of the population considered.
In general, xenobiotic metabolic enzymes can be categorized into phase I and II enzymes. Phase I enzymes are related to the activation of procarcinogens, whereas phase II enzymes centrally perform detoxification.40) We focused on CYP2A6, a phase I enzyme, which plays a role in metabolizing several procarcinogens contained in tobacco smoke, because of the high frequency of CYP2A6 allelic variants in the Japanese population, especially CYP2A6*4, CYP2A6*7, and CYP2A6*9 (22.3%, 10.7%, and 26.6%, respectively).18,21,29) However, CYP2A6 genetic polymorphisms did not show an effect on the risk of bladder cancer when CYP2A6 genotypes were stratified according to the AS system. This suggests that the risk of bladder cancer cannot be explained in terms of CYP2A6 genetic polymorphisms alone. Several other enzymes have been reported to be associated with bladder cancer risk.19,41–44) The phase II enzymes glutathione S-transferases (GSTs) and N-acetyltransferases (NATs) are associated with an increased risk of bladder cancer.19,42–44) In particular, defective GSTM1 alleles increase the risk of bladder cancer in relation to tobacco smoking.19,42) In a meta-analysis, the NAT2 slow acetylator was also found to increase the risk of bladder cancer.43,44) We also previously reported that CYP4B1 genotypes were associated with bladder cancer development.41) In other words, the risk of bladder cancer increased in smokers with the CYP4B1*1/*2 or CYP4B1*2/*2 genotypes.41) Other variations in CYP isoforms, such as CYP1A1 rs2472299 or rs2198843, CYP1A2 −2467T/delT, and one copy of the variant CYP2D6 allele, have been reported to be associated with the development of bladder cancer.45) Therefore, it is possible that, in addition to CYP2A6 genotypes, the GSTs, NAT2, and CYP4B1 genes, as well as other CYP isoform genotypes, intricately affect the risk of bladder cancer.
There is limitation of our study that must be mentioned. Each phenotype was classified according to the insufficient AS system. To date, the consequence for enzymatic activity of many CYP2A6 allelic variants has been evaluated, and although most of these were found to result in reduced metabolic capacities, the decrease in activity level was variable.10–13) In our AS system, CYP2A6 allelic variants were classified into three AS values (1.0, 0.5, and 0), which could be an insufficient classification system for accurate prediction of the CYP2A6 phenotype. For example, there is no evidence of the phenotypes resulting from the variants CYP2A6*1/*4 and CYP2A6*7/*9 in the AS 1.0 group, which would help to achieve higher prediction accuracy.
In conclusion, our results do not support a major role for CYP2A6 polymorphisms in the risk of bladder cancer in Japanese smokers. However, the results should be interpreted with caution because the sample size was relatively small. In order to achieve a statistical power of 80% for analyses of these data, approximately 500 cases and 500 controls subjects would be needed. Therefore, further studies with larger sample sizes should be conducted to confirm the present findings.
We thank the Biomedical Research Core of Tohoku University Graduate School of Medicine for technical support. This study was supported by Grants from the Smoking Research Foundation, the Ministry of Health, Labour and Welfare (MHLW) of Japan (“Initiative to facilitate the development of innovative drugs, medical devices, and cellular & tissue-based products”), and the Japan Research Foundation for Clinical Pharmacology.
The authors declare no conflict of interest.
