O R I G I N A L P A P E R
Polymorphisms in one-carbon metabolism pathway genes, urinary
arsenic profile, and urothelial carcinoma
Chi-Jung Chung• Yeong-Shiau Pu•
Chien-Tien Su• Hui-Wen Chen•Yung-Kai Huang• Horng-Sheng Shiue•Yu-Mei Hsueh
Received: 7 January 2010 / Accepted: 18 May 2010 / Published online: 8 June 2010 Ó Springer Science+Business Media B.V. 2010
Abstract
Background Gene polymorphisms in the one-carbon metabolism pathway could contribute to arsenic methylation capability through plasma folate and homocysteine metabo-lism, thereby increasing the susceptibility to urothelial carci-noma (UC) risk.
Objectives The goal of our study was to evaluate the roles of gene polymorphisms in the one-carbon metabolism pathway in the carcinogenesis of UC.
Methods A hospital-based case–controlled study was conducted. The urinary arsenic profile was examined using high-performance liquid chromatography and hydride gen-erator-atomic absorption spectrometry. Folate levels were
measured using a competitive immunoassay kit. Genotyping was conducted using polymerase chain reaction-restriction fragment length polymorphism technique.
Results Patients with UC had higher urinary total arsenic, inorganic arsenic percentage (InAs%) and monomethylar-senic acid percentage (MMA%), and lower dimethylarmonomethylar-senic acid percentage (DMA%), plasma folate and homocysteine levels than controls. The correlations between folate and DMA%, and folate and homocysteine, were significant according to Pearson’s correlation coefficients. Subjects carrying the 5,10-methylenetetrahydrofolate reductase (MTHFR) CT or TT genotype had a lower DMA% and lower folate levels than those carrying the CC genotype. Partici-pants with the methionine synthase (MS) AA genotype had higher homocysteine levels than those with the AG or GG genotype. However, neither MTHFR nor MS gene poly-morphisms were associated with UC risk.
Conclusions Environmental factors played a more important role in UC carcinogenesis than MTHFR or MS gene polymorphism.
Keywords Urothelial carcinoma
One-carbon metabolism Urinary arsenic MTHFR MS CBS Polymorphism SNP
Introduction
Bladder cancer is the most common malignancy of the urinary tract and the eighth most common cancer of men in Taiwan. It was estimated that 1,985 new cases would be diagnosed and 681 deaths would occur in Taiwan in 2006 (Department of Health, the Executive Yuan, 2009). Inor-ganic arsenic in drinking water is one of the most important risk factors for bladder cancer and accounted for an
C.-J. Chung H.-W. Chen Y.-M. Hsueh
School of Public Health, Taipei Medical University, Taipei, Taiwan
Y.-S. Pu
Department of Urology, National Taiwan University Hospital, Taipei, Taiwan
C.-T. Su
Department of Family Medicine, Taipei Medical University Hospital, Taipei, Taiwan
Y.-K. Huang
Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taipei, Taiwan
H.-S. Shiue
Department of Chinese Medicine, Chang Gung Memorial Hospital, Taipei, Taiwan
Y.-M. Hsueh (&)
Department of Public Health, School of Medicine, Taipei Medical University, No. 250 Wu-Hsing Street, Taipei 110, Taiwan
e-mail: ymhsueh@tmu.edu.tw DOI 10.1007/s10552-010-9589-3
increased risk of urothelial carcinoma (UC) [1, 2]. In humans, methylation of inorganic arsenic (iAs) to mo-nomethylarsenic acid (MMA?5) and dimethylarsenic acids (DMA?5) is by the one-carbon metabolism pathway, with S-adenosylmethionine (SAM) serving as the methyl donor [3]. Arsenic metabolism has generally been a detoxification pathway; however, increasing evidence reveals that the trivalent methylated arsenic intermediate may be the most toxic arsenic species [4, 5]. Epidemiology studies suggest that people with higher inorganic arsenic age (InAs%), higher monomethylarsenic acid percent-age (MMA%), or lower dimethylarsenic acid percentpercent-age (DMA%) have an increased risk of cancers, including UC and bladder cancer, skin lesions, and vascular diseases [6–10].
The levels of folate and homocysteine in one-carbon metabolism could affect the concentration of urinary arsenic profiles by indirectly mediating the transfer of a methyl group from SAM to arsenic [11]. Studies from Bangladesh have indicated that people with folate defi-ciency or hyperhomocysteinemia were associated with decreased arsenic methylation and had increased risk of skin lesions [12]. Similar results were also observed in other studies of cancers [13,14]. Based on these findings, the one-carbon metabolism pathway may play an important role in arsenic methylation.
The polymorphisms of many different enzymes affect-ing arsenic metabolism in one-carbon metabolism have been found in 5,10-methylenetetrahydrofolate reductase (MTHFR), methionine synthase (MS), and cystathionine b-synthase (CBS), including MTHFR C677T, MS A2756G, and CBS 644 insertion 68 bp [15,16]. MTHFR may cata-lyze the conversion of 5,10-methylenetetrahydrofolate to 5-methylenetetrahydrofolate, which provides the methyl group for methionine synthesis through homocysteine remethylation. Further, remethylation of homocysteine to form methionine is catalyzed by MS. CBS catalyzes the conversion of homocysteine to cystathionine through the transsulfuration pathway [17–19]. Studies indicated that these genetic variations of MTHFR C677T, MS A2756G, and CBS 644 insertion 68 bp were associated with altered enzymatic activity. Individuals with the MTHFR 677TT or 677CT genotype had lower enzymatic activity than those with the 677CC genotype [20]. A higher concentration of homocysteine was observed in carriers of the MS 2756A allele than in those with the MS 2756G allele [21]. Fur-thermore, the CBS 644 insertion 68-bp variation was associated with decreased homocysteine levels [15]. Pre-vious studies found that people with the MTHFR TT genotype or the MS GG genotype had a higher MMA% and a lower DMA% than those with either the MTHFR CC genotype or the MS AA genotype, respectively [22,23]. In addition, Steinmaus et al. found that subjects with the
MTHFR TT genotype excreted a significantly higher inorganic arsenic percentage (InAs%) and a lower DMA percentage (DMA%) [24].
Although there were some findings showing the asso-ciation between one-carbon metabolism gene polymor-phisms and urinary arsenic profiles, few studies have explored the relationship between these genetic variants and UC risk. Therefore, we wanted to evaluate whether these gene polymorphisms modified the UC risk by affecting folate and homocysteine levels; we also explored the interaction of folate, homocysteine, urinary arsenic profiles, and gene polymorphism of MTHFR, MS and CBS on UC risk.
Materials and methods Study participants
A hospital-based case–controlled study was constructed. The study design has been described previously [9]. Briefly, from September 2002 to May 2004, we recruited 150 patients with UC as cases and 300 healthy participants as controls from the Medical Center that includes National Taiwan University Hospital and Taipei Municipal Wan Fang Hospital. All cases of UC (or transitional cell carci-noma) were diagnosed by histological confirmation. None of the cases presented other histology, such as squamous cell carcinoma, adenocarcinoma, sarcoma, lymphoma, or benign lesions. Healthy controls were frequency matched with UC cases in terms of age ± 3 years, as well as gender, and had no prior history of cancer. No study participants had lived in the blackfoot disease area of Chaiyi city or drank arsenic-contained artesian well water. We randomly collected and measured the arsenic concentration of tap water from the homes of 37 UC cases. The arsenic concentration of tap water ranged from 11.33 to 27.25 lg/l, and the mean and standard error were 17.14 and 0.55 lg/l, respectively. All participants provided their informed consent forms before a questionnaire interview and biological specimen collection. The Research Ethics Committee of the National Taiwan University Hospital, Taipei, Taiwan, approved the study and it was consistent with the World Medical Association Dec-laration of Helsinki.
Questionnaire interview
Well-trained interviewers collected information based on a structured questionnaire through a face-to-face interview. The questionnaire included demographics and socioeco-nomic characteristics, lifestyle issues such as cigarette smoking and alcohol consumption, environmental expo-sure, and personal and family disease histories.
Biological specimen collection
Twenty ml of spot urine samples was collected at the time of recruitment and immediately transferred to a -20°C freezer until required for urinary arsenic species analyses. Concurrently, peripheral blood specimens of 5–8 ml were collected with ethylene-diamine-tetraacetic acid (EDTA)-vacuumed syringes and then separated into buffy coat and plasma to be frozen at -80°C for the measurements of plasma folate and homocysteine, as well as DNA extraction.
Urinary arsenic species measurement
The analytical methods of urinary arsenic species have been described previously [25]. Briefly, the concentrations of iAs3?, iAs5?, MMA5?, and DMA5?were evaluated by high-performance liquid chromatography (HPLC), equip-ped with a hydride generator and atomic absorption spec-trometer. Recovery rates for iAs3?, DMA5?, MMA5?, and iAs5?ranged from 93.8 to 102.2%, with detection limits of 0.02, 0.08, 0.05, and 0.07 lg/l, respectively. Considering the stability of urinary arsenic species, the urine sample assay of arsenic species was performed within 6 months of sample collection [26].
Plasma folate determination
Plasma folate levels were measured using a competitive immunoassay kit (Diagnostic Products Corporation, Los Angeles, CA) in accordance with the manufacturer’s instructions. All plasma samples were operated under dim yellow light. For 23 pairs of replicate plasma samples, the mean coefficient of variation (CV) was 8.8% [14]. Plasma homocysteine determination
An enzymatic assay for plasma homocysteine was described by Chan et al. [27]. Briefly, the assay uses crude lysate of E. coli containing a different recombinant enzyme, methi-onine c-lyase, a commercially available chromophore. It can be performed manually, using 96-well microtiter plates, or by automation, using the TECAN analyzer. The CVs for within-individual and between-individual precision were \10%. Close correlation (r [ 0.9) was observed between results by the enzymatic method and a reference HPLC procedure [27].
Genotyping
Genomic DNA was extracted from blood specimens using proteinase K digestion following phenol and chloroform extraction. Genotyping for single nucleotide polymorphisms
(SNPs) in MTHFR C677T (Ala222Val), MS A2756G (Asp919Glu), and CBS 844ins68 was performed with a polymerase chain reaction-restriction fragment length polymorphism (PCR–RFLP) technique [16,21,28]. Briefly, the following primers were used to amplify 198-, 189-, 252-(insertion), and 184-bp (no insertion) PCR products: 50-TGAAGGAGAAGGTGTCTGCGGGA-30(forward) and 50-AGGACGGTGCGGTGAGAGTG-30 (backward) for MTHFR, 50-CATGGAAGAATATGAAGATATTAGAC-30 and 50-GAACTAGAAGACAGAAATTCTCTA-30for MS, and 50-CTGGCCCTGGCCTTGAGCCCTGAAAC-30 and 50-GGCCGGGCTCTGGACTC-30 for CBS. PCR products were obtained in a total volume of 30 ll, consisting of an 80 ng sample DNA, 109 PCR buffer, 2.5 mM dNTP, 2 lM of each primer, and 2 U Taq polymerase. After initial denaturation for 4 min at 94°C, 35 cycles were performed of 94°C for 40 s (denaturation), 53°C for 30 s (annealing), and 72°C for 30 s (extension) for MTHFR, 94°C for 30 s, 53°C for 30 s, and 72°C for 30 s for MS, and 94°C for 1 min, 60°C for 1 min, and 72°C for 2 min for CBS, followed by a final step at 72°C for 5 min. The amplified products were visu-alized by electrophoresis in a 2% agarose gel. PCR products were digested with HinfI ([12 h, at 37°C) for MTHFR, and HaeIII ([12 h, at 37°C) for MS. The products were analyzed by electrophoresis on 3% agarose gels. A random 5% of the samples were repeated with a concordance of 100% for quality control.
Statistical analysis
Hardy–Weinberg equilibrium was tested by the goodness of fit v2 test. All genotype distributions of MTHFR, MS, and CBS fit Hardy–Weinberg equilibrium. Due to the low frequency of variants for CBS644 polymorphism, it was not included in the data analysis. Status of cigarette smoking habits included never, former, and current smokers. Former smokers were defined as those who had quit smoking at the time of recruitment and current smokers as those who were still smoking at the time of the interview. Information on the duration (years) and frequency (packs per day) of smoking habits was also collected to calculate the cumu-lative exposure of cigarette smoking (packs–years). Habits of alcohol drinking included never drinking, occasional drinking, and regular drinking. The sum of iAs3?, iAs5?, MMA5?, and DMA5?was defined as total urinary arsenic concentration (lg/g creatinine). The InAs%, MMA per-centage (MMA%), and DMA% were calculated by divid-ing the concentration of each species (iAs3?? iAs5?, MMA5?, or DMA5?) by the total arsenic concentration. Continuous values of urinary arsenic profiles were used in all model analyses. After log-transformation of non-normalized variables, Student’s t test and v2test were used to compare the differences in related risk factors between
UC cases and controls. We performed multivariate linear regression models to estimate the p-values of independent variables (UC status, smoking, alcohol consumption, and polymorphisms of MTHFR and MS) on relevant variables (urinary arsenic profiles, plasma folate, and homocysteine) after adjustment for age, gender, and education levels. In addition, we performed multivariate logistic regression models to estimate the odds ratios (ORs) and 95% confi-dence intervals (CIs) to calculate the risks of genotypes of MTHFR and MS on UC after adjustment for age, gender, education levels, and other relevant variables. Pearson’s correlations were adopted to estimate the relationships among plasma folate, homocysteine, and urinary arsenic profiles. The joint effects were evaluated by estimating the synergy index [29]. Finally, we used a stepwise logistic regression model to construct model selection. All analyses were conducted using SAS statistical package (SAS, ver-sion 8.0, Cary, NC).
Results
The distributions of sociodemographic characteristics and lifestyle risk factors, as well as MTHFR and MS genotypes, are shown in Table1. Healthy controls had higher educa-tion levels than UC cases. The distribueduca-tion of cumulative exposure of cigarette smoking (pack–years) and alcohol consumption was significantly different between UC patients and controls. No gene polymorphisms were asso-ciated with UC risk, even after adjustment for age, gender, and other risk factors (data not shown).
The differences among urinary arsenic profiles, folate, and homocysteine levels were compared by cancer status, cumulative exposure of cigarette smoking status (0 vs. [0 pack-years), alcohol consumption status, and gene poly-morphisms (Table2). UC cases had significantly higher total arsenic, InAs%, and MMA%, and lower DMA% than healthy controls. Further, UC cases had significantly lower folate levels and higher homocysteine levels than controls. There were no differences in the urinary arsenic profile, folate levels, or homocysteine levels by cumulative exposure of cigarette smoking status (pack-years) or alcohol con-sumption status in either UC patients or controls. Healthy controls with the MTHFR CT or TT genotype had a signif-icantly lower DMA% and lower folate concentration than those with the CC genotype after adjustment for age, gender, and education. In addition, healthy controls and UC cases with the MS AA genotype had significantly higher homo-cysteine levels than those with the AG or GG genotype.
The relationships between plasma folate and homocys-teine levels and between plasma folate and urinary arsenic profiles are shown in Fig.1. There were significant corre-lations between folate and DMA% and between folate and
homocysteine levels. However, there were no associations between homocysteine levels and urinary arsenic profiles (data not shown).
Analyses of all relevant variables for UC risk through multivariate logistic regressions are shown in Table3. Urinary total arsenics, plasma folate, and homocysteine were significantly associated with UC risk after adjusting for age, gender, educational level and occasional alcohol consumption. Similar results were observed when we examined InAs%, or MMA% or DMA% in place of total arsenics (data not shown). The significant variables in Table3 yielded similar results through model selection by stepwise logistic regressions (data not shown).
Discussion
To our knowledge, this is the first study to simultaneously evaluate the associations among the one-carbon metabo-lism pathway of MTHFR and MS gene polymorphisms, plasma folate and homocysteine as well as urinary arsenic profiles. Further, we explored the impact of MTHFR and MS genotypes on the susceptibility to UC risk. In the present study, participants who carried either the MTHFR CT or TT genotypes had lower DMA% and lower folate levels than those who carried the CC genotype. In addition, those who carried the MS AA genotype had higher homocysteine levels than those who carried the AG or GG genotype. It appears that people with high homocysteine or with low folate levels had an increased UC risk.
Our previous studies showed that participants with higher MMA% or lower DMA% had an increased risk of UC compared to those with lower MMA% or higher DMA% in the blackfoot disease-endemic area and in the non-arsenic-exposure area of Taiwan [8, 9]. Factors affecting arsenic methylation capability included age, gender, habits of cig-arette smoking and alcohol consumption, nutritional status, and gene susceptibility [30]. Generally, the elderly, males, cigarette smokers, alcohol drinkers, and those with poor nutritional status (low urinary selenium, low serum alpha-tocopherol, or low plasma folate) had decreased arsenic methylation capability [10, 14, 31]. In addition, we also found that arsenic metabolism-related genes including AS3MT, PNP, and GSTO2 regulate the arsenic methylation process and might influence the susceptibility to cancer risk [32].
Besides the factors mentioned above, folate, a water-soluble B vitamin, plays an important role in the one-car-bon metabolism pathway and provides one-carone-car-bon groups to DNA methylation, DNA synthesis and arsenic methyl-ation [11,33,34]. One cross-sectional study of Banglade-shi measured urinary arsenic metabolites for a subset of 300 individuals and found urinary DMA% was positively
associated with plasma folate (r = 0.14, p = 0.02) and negatively associated with total homocysteine (r = -0.14, p = 0.01). Conversely, MMA% was negatively associated with folate (r = -0.12, p = 0.04) and positively associ-ated with homocysteine (r = 0.21, p = 0.0002) [13]. Fur-ther, some folic acid supplementation trials in Bangladesh indicated that folic acid supplementation of arsenic-exposed adults would increase arsenic methylation [35,36]. These data suggest that higher folate and lower homocysteine levels could facilitate arsenic methylation through SAM as the methyl donor, leading to decreased MMA% and increased DMA%. Our previous study found significant joint effects of plasma folate and urinary arsenic metabolites (total arsenic, or MMA%, or DMA%) on UC risk [14]. The joint effects of folate and urinary arsenic profiles on healthy hazards have been shown in other studies [37–39]. Further, in the present study, we showed a significant negative correlation between folate and homo-cysteine levels (Fig.1). There is a significant dose– response relationship between the joint effects of folate and
homocysteine levels and UC risk (trend p-value \ 0.05; data not shown). Participants with lower folate and higher homocysteine levels had a significantly higher UC risk than those with higher folate and lower homocysteine levels, but their synergistic effects were insignificant (data not shown). In addition, participants with higher folate levels had significantly increasing urinary DMA%. Possibly, participants with higher folate levels had a lower UC risk through increasing urinary DMA% than those with lower folate levels.
Further, we defined the role of one-carbon metabolism pathway genes (MTHFR and MS) in folate and homocys-teine metabolism and explored the association between the gene polymorphisms in one-carbon metabolism pathway and UC risk. The MTHFR T allele frequency in our group was 33%, which was in accordance with those in China and in Taiwan [40,41]. Control participants with the MTHFR CT or TT genotype had lower folate levels and lower DMA% than those with the CC genotype in this study. Weisberg et al. had documented that the MTHFR CT or TT
Table 1 Sociodemographic characteristics, lifestyle risk factors, and MTHFR as well as MS genotypes of 150 UC patients and 300 controls UC patients (n = 150) Controls (n = 300) p-Valuea
Age (years) 65.32 ± 1.08 66.2 ± 0.73 0.55
Male (%) 103 (68.67) 206 (68.67) 1.00
Education (%)c 0.001
Elementary school 60 (40.00) 91 (30.43) Junior high school 60 (40.00) 94 (31.44) College or above 30 (20.00) 114 (38.13)
Cumulative exposure of cigarette smoking (pack-years)d 81 (54.00) 122 (40.67) 0.0014
0 71 (50.71) 179 (62.81) 0–22 20 (14.29) 52 (18.25) [22 49 (35.00) 54 (18.95) Alcohol consumption (%) 0.004 No 94 (62.67) 157 (52.33) Occasional 25 (10.67) 94 (31.33) Frequent 31 (20.67) 49 (16.33) MTHFR C677Tb 0.36 CC 80 (53.33) 141 (47.00) CT 57 (38.00) 123 (41.00) TT 13 (8.67) 36 (12.00) CT ? TT 70 (46.67) 159 (53.00) MS A2756Gb 0.44 AA 121 (80.67) 244 (81.33) AG 29 (19.33) 53 (17.67) GG 0 (0.00) 3 (1.00) AG ? GG 39 (19.33) 56 (18.67)
a p Values were evaluated by Student’s t test or v2test b p[ 0.05 were evaluated by Hardy–Weinberg equilibrium test c Education levels were unknown for 1 control
genotype was associated with decreased enzymatic activity [20]. Hence, decreased enzymatic activity could not effi-ciently convert 5-methylenetetrahydrofolate to tetrahydro-folate (THF), an active form of tetrahydro-folate, and could not catalyze the conversion of homocysteine to methionine. According to this hypothesis, participants with the MTHFR CT or TT genotype might have higher homocysteine and lower THF levels, leading to inefficient arsenic methylation and, therefore, increased UC risk. There was a slightly negative association between MTHFR polymorphism and UC risk (0.05 \ p \ 0.1), in spite of the fact that it seems to be associated with lower folate levels and lower urinary DMA%, which would be an interesting finding. However, we do not have potential explanation at this point. On the
other hand, the MS G allele frequency in our group was 10%, which was similar to 9.5% of Han population in China but lower than other populations [42, 43]. In this study, people with the MS AA genotype had higher homocysteine levels than those with the AG or GG geno-type, which was in accordance with the findings of Chen et al. [21]. However, the direct association between MS polymorphism and UC risk was not presented in this study. Considering all risk factors of UC through stepwise logistic regression model (Table 3), the gene polymorphism of MTHFR or MS played a less important role in UC risk than environmental factors, including cumulative cigarette smoking, occasional alcohol consumption, urinary arsenic profiles, plasma folate, and homocysteine levels.
Table 2 Comparison of urinary arsenic profile, plasma folate, and homocysteine (mean ± stander error) by UC, smoking as well as alcohol drinking status and polymorphisms of MTHFR and MS
Categories Number Total arsenic (lg/g creatinine)
InAs (%) MMA (%) DMA (%) Folatea(ng/ml) Homocysteine (lM)
UC cases 150 39.21 ± 3.36* 7.13 ± 0.58* 12.49 ± 1.03* 80.38 ± 1.23* 7.00 ± 0.50* 11.15 ± 0.58* Controls 300 25.79 ± 1.03* 5.37 ± 0.36* 8.13 ± 0.46* 86.50 ± 0.61* 12.82 ± 0.35* 9.17 ± 0.27* In cases
Cumulative exposure of cigarette smoking (pack–years)
0 71 42.59 ± 5.50 6.48 ± 0.70 11.37 ± 1.52 82.16 ± 1.66 6.76 ± 8.46 11.05 ± 0.81 [0 69 36.17 ± 4.56 7.70 ± 0.96 14.32 ± 1.50 77.98 ± 1.86 7.21 ± 0.59 11.41 ± 0.93 Alcohol consumption (%) No 94 41.54 ± 5.09 6.68 ± 0.63 11.87 ± 1.23 81.45 ± 1.40 7.55 ± 0.60 10.24 ± 0.49 Occasional 31 35.12 ± 4.14 8.29 ± 1.46 12.24 ± 1.78 79.46 ± 1.98 5.26 ± 0.64 13.68 ± 2.25 Frequent 25 35.51 ± 3.83 7.38 ± 1.76 15.11 ± 3.49 77.51 ± 4.58 7.36 ± 1.96 11.41 ± 0.86 MTHFR CC 80 33.22 ± 2.24 7.30 ± 0.70 12.78 ± 1.47 79.91 ± 1.70 7.40 ± 0.76 10.91 ± 0.52 CT ? TT 70 46.05 ± 6.66 6.93 ± 0.94 12.15 ± 1.44 80.92 ± 1.80 6.52 ± 0.64 11.41 ± 1.09 MS AA 121 41.25 ± 4.10 7.47 ± 0.67 12.93 ± 1.22 79.60 ± 1.44 6.67 ± 0.57 11.53 ± 0.68* AG ? GG 29 30.68 ± 2.60 5.72 ± 1.01 10.63 ± 1.66 83.65 ± 2.07 8.39 ± 1.06 9.56 ± 0.93* In controls
Cumulative exposure of cigarette smoking (pack–years)
0 179 26.76 ± 1.47 4.98 ± 0.50 7.47 ± 0.59 87.55 ± 0.83 13.13 ± 0.44 9.50 ± 0.39 [0 106 24.13 ± 1.45 6.04 ± 0.58 9.32 ± 0.81 84.64 ± 0.98 12.33 ± 0.57 10.46 ± 0.38 Alcohol consumption (%) No 157 26.98 ± 1.55 5.17 ± 0.53 7.88 ± 0.68 86.95 ± 0.91 12.77 ± 0.43 8.86 ± 0.43 Occasional 49 22.83 ± 2.08 7.05 ± 0.98 8.57 ± 1.02 84.38 ± 1.34 12.59 ± 0.99 9.81 ± 0.68 Frequent 94 25.34 ± 1.70 4.83 ± 0.55 8.31 ± 0.77 86.87 ± 1.00 13.03 ± 0.65 9.35 ± 0.32 MTHFR CC 141 27.29 ± 1.66 4.86 ± 0.47 7.93 ± 0.69 87.21 ± 0.95* 13.59 ± 0.53* 9.00 ± 0.26 CT ? TT 159 24.47 ± 1.26 5.82 ± 0.54 8.30 ± 0.62 85.88 ± 0.78* 12.19 ± 0.46* 9.32 ± 0.46 MS AA 244 26.07 ± 1.16 5.36 ± 0.41 8.20 ± 0.53 86.43 ± 0.70 12.58 ± 0.38 9.36 ± 0.32** AG ? GG 56 24.58 ± 2.22 5.41 ± 0.82 7.79 ± 0.92 86.81 ± 1.16 13.77 ± 0.85 8.33 ± 0.43** * p \ 0.05 or ** 0.05 \ p\0.1 through multivariate linear regression model after adjusting for age, sex, and education
There are some limitations of this study that need to be considered when interpreting the results. First, the UC cases are prevalent cases, and the exact time from diagnosis until
sample collection is not known. We cannot exclude the possibility that the levels of folate, homocysteine, and uri-nary arsenic methylation have changed since UC diagnosis. MMA% r = -0.18 p = 0.10 DMA% r = 0.14 p = 0.04 Total arsenic r = 0.05 p = 0.74 InAs % r = -0.11 p = 0.26 Homocysteine r = -0.11 p = 0.04 (A) (B) (D) (C) (E)
Fig. 1 Pearson’s correlation between plasma folate levels (ng/ml) (X-axis) and urinary arsenic profiles in control subjects (n = 300). a total arsenic (lg/g creatinine), b InAs%, c MMA%, d DMA%, and e homocysteine
Second, the sample size of 150 UC cases was too small to evaluate the association of CBS gene polymorphism with UC risk; therefore, we could not explore the role that CBS polymorphisms play in the one-carbon metabolism pathway. In summary, this study shows that high homocysteine and low folate levels may be responsible for an increased UC risk. The MTHFR CT or TT genotype was associated with decreased urinary DMA% and decreased folate levels. The MS AA genotype was associated with increased homocys-teine levels. Considering these variables, environmental factors (urinary arsenic profiles, plasma folate, and homo-cysteine levels) played a more important role in UC carci-nogenesis than MTHFR or MS gene polymorphism.
Acknowledgments The study was supported by grants from the National Science Council of the ROC (NSC 86-2314-B-038-038, NSC 87-2314-B-038-029, NSC-88-2314-B-038-112, NSC-89-2314-B038-049, SC-89-2320-B038-013, NSC-90-2320-B-038-021, NSC91-3112-B-038-0019, NSC92-3112-B-038-001, NSC93-3112-B-038-001, NSC94-2314-B-038-023, NSC-95-2314-B-038-007, NSC-96-2314-B038-003, NSC B-038 -015 -MY3 (1-3), and NSC 97-2314-B-038 -015 -MY3 (2, 3)). References
1. Chiou HY, Chiou ST, Hsu YH et al (2001) Incidence of transitional cell carcinoma and arsenic in drinking water: a follow-up study of 8, 102 residents in an arseniasis-endemic area in northeastern Taiwan. Am J Epidemiol 153:411–418
2. Steinmaus C, Yuan Y, Bates MN, Smith AH (2003) Case–control study of bladder cancer and drinking water arsenic in the western United States. Am J Epidemiol 158:1193–1201
3. Marafante E, Vahter M (1984) The effect of methyltransferase inhibition on the metabolism of [74As]arsenite in mice and rab-bits. Chem Biol Interact 50:49–57
4. Styblo M, Drobna Z, Jaspers I, Lin S, Thomas DJ (2002) The role of biomethylation in toxicity and carcinogenicity of arsenic: a research update. Environ Health Perspect 110:767–771 5. Wang TC, Jan KY, Wang AS, Gurr JR (2007) Trivalent arsenicals
induce lipid peroxidation, protein carbonylation, and oxidative DNA damage in human urothelial cells. Mutat Res 615:75–86 6. Ahsan H, Chen Y, Kibriya MG et al (2007) Arsenic metabolism,
genetic susceptibility, and risk of premalignant skin lesions in Bangladesh. Cancer Epidemiol Biomarkers Prev 16:1270–1278 7. Chen YC, Su HJ, Guo YL et al (2003) Arsenic methylation and
bladder cancer risk in Taiwan. Cancer Causes Control 14:303– 310
8. Huang YK, Huang YL, Hsueh YM et al (2008) Arsenic exposure, urinary arsenic speciation, and the incidence of urothelial carci-noma: a twelve-year follow-up study. Cancer Causes Control 19:829–839
9. Pu YS, Yang SM, Huang YK et al (2007) Urinary arsenic profile affects the risk of urothelial carcinoma even at low arsenic exposure. Toxicol Appl Pharmacol 218:99–106
10. Tseng CH, Huang YK, Huang YL et al (2005) Arsenic exposure, urinary arsenic speciation, and peripheral vascular disease in blackfoot disease-hyperendemic villages in Taiwan. Toxicol Appl Pharmacol 206:299–308
11. Li L, Ekstrom EC, Goessler W et al (2008) Nutritional status has marginal influence on the metabolism of inorganic arsenic in preg-nant Bangladeshi women. Environ Health Perspect 116:315–321 12. Pilsner JR, Liu X, Ahsan H et al (2009) Folate deficiency,
hyperhomocysteinemia, low urinary creatinine, and hypomethy-lation of leukocyte DNA are risk factors for arsenic-induced skin lesions. Environ Health Perspect 117:254–260
13. Gamble MV, Liu X, Ahsan H et al (2005) Folate, homocysteine, and arsenic metabolism in arsenic-exposed individuals in Ban-gladesh. Environ Health Perspect 113:1683–1688
Table 3 Multivariate logistic regression analysis of lifestyle risk factors, total arsenics, plasma folate, and homocysteine and MTHFR as well as MS genotypes for UC risk
Variables Odds ratio (95% CI)
Model 1 p Model 2 p
Cumulative exposure of cigarette smoking (continuous scale)
1.01 (0.99–1.02) 0.39 1.01 (0.99–1.02) 0.36 Alcohol consumptions No 1.00 1.00 Occasional 0.25 (0.12–0.51) \0.01 0.24 (0.12–0.50) \0.01 Frequent 0.51 (0.23–1.13) 0.09 0.50 (0.23–1.11) 0.09 Total arsenic 1.04 (1.02–1.06) \0.01 1.04 (1.03–1.06) \0.01 Folate 0.81 (0.76–0.86) \0.01 0.81 (0.77–0.86) \0.01 Homocysteine 1.05 (1.00–1.11) 0.05 1.05 (1.00–1.10) 0.05 MTHFR CC 1.00 CT ? TT 0.60 (0.35–1.03) 0.06 MS AA 1.00 AG ? GG 1.21 (0.62–2.39) 0.58
14. Huang YK, Pu YS, Chung CJ et al (2008) Plasma folate level, urinary arsenic methylation profiles, and urothelial carcinoma susceptibility. Food Chem Toxicol 46:929–938
15. Dekou V, Whincup P, Papacosta O et al (2001) The effect of the C677T and A1298C polymorphisms in the methylenetetrahy-drofolate reductase gene on homocysteine levels in elderly men and women from the British regional heart study. Atherosclerosis 154:659–666
16. Frosst P, Blom HJ, Milos R et al (1995) A candidate genetic risk factor for vascular disease: a common mutation in methylenete-trahydrofolate reductase. Nat Genet 10:111–113
17. Botto LD, Yang Q (2000) 5, 10-Methylenetetrahydrofolate reductase gene variants and congenital anomalies: a HuGE review. Am J Epidemiol 151:862–877
18. Maron BA, Loscalzo J (2006) Homocysteine. Clin Lab Med 26:591–609
19. Selhub J (1999) Homocysteine metabolism. Annu Rev Nutr 19:217–246
20. Weisberg I, Tran P, Christensen B, Sibani S, Rozen R (1998) A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 64:169–172
21. Chen J, Stampfer MJ, Ma J et al (2001) Influence of a methionine synthase (D919G) polymorphism on plasma homocysteine and folate levels and relation to risk of myocardial infarction. Ath-erosclerosis 154:667–672
22. Engstrom KS, Broberg K, Concha G, Nermell B, Warholm M, Vahter M (2007) Genetic polymorphisms influencing arsenic metabolism: evidence from Argentina. Environ Health Perspect 115:599–605
23. Lindberg AL, Kumar R, Goessler W et al (2007) Metabolism of low-dose inorganic arsenic in a central European population: influence of sex and genetic polymorphisms. Environ Health Perspect 115:1081–1086
24. Steinmaus C, Moore LE, Shipp M et al (2007) Genetic Poly-morphisms in MTHFR 677 and 1298, GSTM1 and T1, and Metabolism of Arsenic. J Toxicol Environ Health, Part A 70: 159–170
25. Hsueh YM, Huang YL, Huang CC et al (1998) Urinary levels of inorganic and organic arsenic metabolites among residents in an arseniasis-hyperendemic area in Taiwan. J Toxicol Environ Health 54:431–444
26. Chen YC, Amarasiriwardena CJ, Hsueh YM, Christiani DC (2002) Stability of arsenic species and insoluble arsenic in human urine. Cancer Epidemiol Biomarkers Prev 11:1427–1433 27. Chan EC, Chang PY, Wu TL, Wu JT (2005) Enzymatic assay of
homocysteine on microtiter plates or a TECAN analyzer using crude lysate containing recombinant methionine gamma-lyase. Ann Clin Lab Sci 35:155–160
28. Harmon DL, Shields DC, Woodside JV et al (1999) Methionine synthase D919G polymorphism is a significant but modest
determinant of circulating homocysteine concentrations. Genet Epidemiol 17:298–309
29. Hosmer DW, Lemeshow S (1992) Confidence interval estimation of interaction. Epidemiology 3:452–456
30. Tseng CH (2009) A review on environmental factors regulating arsenic methylation in humans. Toxicol Appl Pharmacol 235:338–350
31. Hsueh YM, Ko YF, Huang YK et al (2003) Determinants of inorganic arsenic methylation capability among residents of the Lanyang Basin, Taiwan: arsenic and selenium exposure and alcohol consumption. Toxicol Lett 137:49–63
32. Chung CJ, Hsueh YM, Bai CH et al (2009) Polymorphisms in arsenic metabolism genes, urinary arsenic methylation profile and cancer. Cancer Causes Control 20:1653–1661
33. Stanger O, Semmelrock HJ, Wonisch W, Bos U, Pabst E, Wascher TC (2002) Effects of folate treatment and homocysteine lowering on resistance vessel reactivity in atherosclerotic sub-jects. J Pharmacol Exp Ther 303:158–162
34. Suh JR, Herbig AK, Stover PJ (2001) New perspectives on folate catabolism. Annu Rev Nutr 21:255–282
35. Gamble MV, Liu X, Ahsan H et al (2006) Folate and arsenic metabolism: a double-blind, placebo-controlled folic acid-sup-plementation trial in Bangladesh. Am J Clin Nutr 84:1093–1101 36. Gamble MV, Liu X, Slavkovich V et al (2007) Folic acid sup-plementation lowers blood arsenic. Am J Clin Nutr 86:1202–1209 37. Hall M, Gamble M, Slavkovich V et al (2007) Determinants of arsenic metabolism: blood arsenic metabolites, plasma folate, cobalamin, and homocysteine concentrations in maternal-new-born pairs. Environ Health Perspect 115:1503–1509
38. Pilsner JR, Liu X, Ahsan H et al (2007) Genomic methylation of peripheral blood leukocyte DNA: influences of arsenic and folate in Bangladeshi adults. Am J Clin Nutr 86:1179–1186
39. Wu MM, Chiou HY, Hsueh YM et al (2006) Effect of plasma homocysteine level and urinary monomethylarsonic acid on the risk of arsenic-associated carotid atherosclerosis. Toxicol Appl Pharmacol 216:168–175
40. Chou YC, Wu MH, Yu JC et al (2006) Genetic polymorphisms of the methylenetetrahydrofolate reductase gene, plasma folate levels and breast cancer susceptibility: a case–control study in Taiwan. Carcinogenesis 27:2295–2300
41. Wilcken B, Bamforth F, Li Z et al (2003) Geographical and ethnic variation of the 677C[T allele of 5, 10 methylenete-trahydrofolate reductase (MTHFR): findings from over 7000 newborns from 16 areas world wide. J Med Genet 40:619–625 42. Dekou V, Gudnason V, Hawe E, Miller GJ, Stansbie D,
Humphries SE (2001) Gene-environment and gene-gene inter-action in the determination of plasma homocysteine levels in healthy middle-aged men. Thromb Haemost 85:67–74
43. Matsuo K, Suzuki R, Hamajima N et al (2001) Association between polymorphisms of folate- and methionine-metabolizing enzymes and susceptibility to malignant lymphoma. Blood 97:3205–3209
