• 沒有找到結果。

Polymorphisms in arsenic metabolism genes, urinary arsenic methylation profile and cancer.

N/A
N/A
Protected

Academic year: 2021

Share "Polymorphisms in arsenic metabolism genes, urinary arsenic methylation profile and cancer."

Copied!
9
0
0

加載中.... (立即查看全文)

全文

(1)

O R I G I N A L P A P E R

Polymorphisms in arsenic metabolism genes, urinary arsenic

methylation profile and cancer

Chi-Jung ChungÆ Yu-Mei Hsueh Æ Chyi-Huey Bai Æ Yung-Kai HuangÆ Ya-Li Huang Æ Mo-Hsiung Yang Æ Chien-Jen Chen

Received: 17 January 2009 / Accepted: 28 July 2009  Springer Science+Business Media B.V. 2009

Abstract Arsenic-metabolism-related genes can regulate the arsenic methylation process and may influence suscep-tibility to cancer. We evaluated the roles of arsenic metabolism genes on urinary arsenic profiles of repeated measurement with 15-year follow-up (1988–2004) through general linear model (GLM) and assessed the effect of the changed extent of urinary arsenic profiles on cancer risk. Questionnaire information and blood samples and two urines (1988 and 2004) were collected from 208 subjects in an arseniasis hyperendemic area in Taiwan. Profiles for concentrations of urinary arsenic were determined using HPLC-HG-AAS. The relative proportion of each arsenic species was calculated by dividing the concentration of each arsenic species by the total arsenic concentration. Geno-typing was done using the 50nuclease allelic discrimination (Taqman) assay. The incidence of cancer was identified

through linking to the National Cancer Registry Systems. The Cox proportional hazards model and survival curves were used in the analyses. After a 15-year follow-up, baseline monomethylarsonic acid percentage (MMA%) and change in MMA% exhibited a significant dose–response relationship with cancer risk. Individuals with a higher baseline MMA% and a lower change in MMA% had the earliest cancer incidence (statistically significant). Through GLM, significant gene effects of arsenic (?3 oxidation state)-methyltransferase (AS3MT) on MMA%, dimethyl-arsinic acid percentage (DMA%) and DMA/MMA, purine nucleoside phosphorylase (PNP) on DMA% and glutathi-one S-transferase omega 2 (GSTO2) on inorganic arsenics (InAs%) were found. Our results show that MMA% might be a potential predictor of cancer risk. The change in MMA% was linked to individual cancer susceptibility related to AS3MT rs3740393.

Keywords Urinary arsenic methylation profile Arsenic metabolism AS3MT  PNP  GSTO1 

GSTO2 Polymorphism  Cancer  Repeated measurement

Introduction

Arsenic has induced skin, lung, bladder and internal cancers as well as other severe health problems in the residents chronically exposed to arsenic-contaminated drinking water in the black-foot disease (BFD) endemic area of Taiwan [1–4]. Even though we found that their total urinary arsenic concentrations were reduced after a tap water supply system was installed in the 1970s [5], arsenic-related diseases still occur in these residents. Differences in individual genetic susceptibility might affect arsenic metabolism and toxicity; to date, arsenic (?3 oxidation state)-methyltransferase

C.-J. Chung

Graduate Institute of Public Health, Taipei Medical University, Taipei, Taiwan

Y.-M. Hsueh (&)  Y.-K. Huang  Y.-L. Huang

Department of Public Health, School of Medicine, Taipei Medical University, No. 250 Wu-Hsing Street, Taipei 110, Taiwan

e-mail: ymhsueh@tmu.edu.tw

C.-H. Bai

Section of Neurology, Shin Kong Wo Ho-Su Memorial Hospital, Taipei, Taiwan

M.-H. Yang

Department of Nuclear Science, National Tsing-Hua University, Hsinchu, Taiwan

C.-J. Chen

Genomics Research Center, Academia Sinica, Taipei, Taiwan DOI 10.1007/s10552-009-9413-0

(2)

(AS3MT), purine nucleoside phosphorylase (PNP), gluta-thione S-transferase omega 1 (GSTO1) and GSTO2 have been proposed to participate in the arsenic metabolism pathway, including reduction and oxidative methylation.

AS3MT in arsenic biotransformation could catalyze the transfer of a methyl group using S-adenosyl-methionine as the methyl donor, leading to the methylation of trivalent arsenic species (arsenite; iAs3? and MMA3?) [6, 7]. A variety of cell line studies have indicated that AS3MT is capable of methylation of iAs3? or MMA3?, yielding mono- or dimethylated As3?- and As5?-containing metab-olites [8–10]. The human AS3MT gene, located on chro-mosome 10q24, contains 11 exons and several single nucleotide polymorphisms (SNPs), including three exonic SNPs of Arg173Trp (exon6), Met287Thr (exon9) and Thr306Ile (exon10) as well as other SNPs in the 50UTR and intronic regions [11]. Enzymes involved in reducing the pentavalent arsenic species iAs5?, MMA5? and DMA5? were PNP, GSTO1 and GSTO2, respectively. Genes encoding these enzymes mapped to chromosomes 14q11.2 and 10q24.3, respectively. GSTO1 and GSTO2 are 7.5 kb apart [12, 13]. The MMA5? reducing activity seen in the liver cytosol in GSTO1 gene knockout mice was only 20% of that found in wild-type mice after a single injection of iAs3?(4.16 mg As/kg body weight) [14]. Using polymerase chain reaction (PCR) and sequencing, Hernandez et al. (2008) found nine polymorphisms in 50 Chilean-copper-smelting workers exposed to arsenic with 152.8 ± 19.2 lg/ l of total arsenic in the urine. Among these SNPs, subjects with the variant genotypes promoter-114 G/C, 50UTR GC/AT and Met287Thr had more MMA% in the urine than those with wild-type genotypes (p \ 0.01) [15]. Among 104 indigenous women from northern Argentina, exposed to approximately 200 lg/l of arsenic in drinking water, carriers with two non-exonic polymorphisms of rs3740400 and rs7085104 in AS3MT had lower MMA% and higher DMA% [16]. Further, three intronic polymorphisms in AS3MT (G12390C, C14215T and G35991A) were associ-ated with a lower MMA% and a higher DMA% in urine [17]. Several SNPs were also reported in PNP, GSTO1 and GSTO2 genes. Yu et al. (2003) found 33 polymorphic sites during sequencing of the GSTO1 gene in 22 European and 24 American ancestries. These included 6 exonic SNPs. In these populations, the minor allele frequency of SNPs in GSTO1, including the non-conservative amino acid sub-stitution Ala140Asp (A140D), was 0–45% [13]. In addition, 66 polymorphisms, including the G to A substitution at codon 142 in exon 4 (Asn142Asp) in GSTO2, were iden-tified in four populations, with variant allele frequencies of 12–26% [12]. A genetic association study of 135 subjects in Mexico was performed by testing 23 polymorphic sites in AS3MT, PNP and GSTO1. Among these, only AS3MT 30585 was strongly associated with an altered ratio of

urinary DMA to MMA in children (7–11 years) but not in adults (18–79 years) [18].

Although there is scant of evidence for the function of PNP, GSTO1 and GSTO2 on arsenic metabolism, the reduction steps of arsenic metabolism catalyzed by PNP, GSTO1 and GSTO2 played an important role in detoxifi-cation pathway of arsenics [14].

Considering the genotype frequency of AS3MT, PNP, GSTO1 and GSTO2 in Asian populations based on the lit-erature [19,20], along with a review of the Single Nucle-otide Polymorphism database (http://www.ncbi.nlm.nih.gov/ sites/entrez), we chose rs3740393 of AS3MT, rs1760940 and rs1713420 of PNP, rs4925 of GSTO1 (Ala140Asp) and rs156697 of GSTO2 (Asn142Asp) for further exploration. In addition, the residents of the BFD region ceased drinking arsenic-contaminated water about 30 years ago, correspond-ing to the latency period of arsenic-induced cancer develop-ment [21]. The functions of AS3MT, PNP, GSTO1 and GSTO2 may influence on the arsenic metabolism, and their polymorphisms may affect individual susceptibility toward arsenic toxicity. Our goal in this study was to explore if SNPs in the arsenic metabolism genes affect the urinary arsenic profile and to ascertain a possible association between SNPs and development of cancer in subjects from an arseniasis hyperendemic area in Taiwan.

Materials and methods Study area and participants

We conducted a cohort study and recruited study partici-pants from Homei, Funshin and Hsinming villages in the Putai Township of Chiayi County, which had the highest prevalence of BFD in Taiwan. The study design has been described previously [22]. Briefly, the BFD cohort was conducted between September 1988 and June 1989. Out of 1,571 eligible subjects who lived in the BFD area 5 or more days a week and received interviews, 1,081 participants completed a health examination. The median arsenic con-centration of the artesian well water ranged from 0.7 to 0.93 mg/l before the implementation of a tap water supply system in the early 1960s. The median value of concen-tration of total urinary arsenic for a total of 1,078 residents with urine samples was from 50 to 80 lg/l [5]. The Insti-tutional Review Board of National Taiwan University approved this study. In August 2004, we further invited the participants of the BFD cohort to attend a community-based mass screening, in cooperation with the Bureau of Health Promotion of the Chayi County Government of Taiwan, by postcard. Three hundred and five participants were exclu-ded due to death, 529 due to withdrawal and 36 due to incomplete interviews. Finally, a sub-cohort included 208

(3)

participants who had completed questionnaire interviews, blood specimens and two urine samples (1988 and 2004). Follow-up of cancer incidence

The incidence of all cancers among study participants was identified through the subjects’ unique national identifica-tion numbers, linked to the Naidentifica-tional Cancer Registry Systems in Taiwan. From 1 January 1989 to 31 December 2004, 17 out of 208 participants developed malignant neoplasms, including one of the thymus (ICD 9, 164.0), one of the eyelid (ICD 9, 173.1), one of the skin of the ear and external auditory canal (ICD 9, 173.2), one of the scalp and skin of neck (ICD 9, 173.4), two of the trunk, except scrotum (ICD 9, 173.5), one of the lower limbs (ICD 9, 173.7), one of the female breast (ICD 9, 174.4), two of the cervix uteri (ICD 9, 180.9), three of the bladder (ICD 9, 188.9), one of the renal pelvis (ICD 9, 189.1), two of the ureter (ICD 9, 189.2) and one of the thyroid gland (ICD 9, 193). Overall, 17 cases of cancer and 191 cancer-free residents were analyzed in the present study.

Questionnaire interview and biological specimen collection

Two well-trained interviewers collected detailed informa-tion on demographics and socioeconomic characteristics, lifestyle-related risk factors such as cigarette smoking and alcohol consumption, residential and occupational history and history of drinking well water, through a structured questionnaire [22]. Two urine samples were collected; one was at the time of recruitment in 1988–1989 (baseline), and the other one was in 2004 (second). Urine samples were immediately transferred to a -20C freezer until required for urinary arsenic profile analyses. In addition, blood specimens from 1988 to 1989 (the time of recruitment) were collected and frozen at -80C for DNA extraction. Urinary arsenic profiles assessment

Levels of urinary arsenic profiles, including iAs3?, iAs5?, MMA5?and DMA5?were measured by high-performance liquid chromatography (HPLC), equipped with a hydride generator and atomic absorption spectrometer (HG-AAS). Details of the experimental protocol are reported in Hsueh et al. [23]. Freeze-dried SRM 2670 urine obtained from the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA) contained 480 ± 100 lg/l arsenic and was analyzed together with urine samples from subjects as a control. Arsenic from these SRM 2670 con-trols measured at 507 ± 17 lg/l (n = 4). Recovery rates for iAs3?, DMA5?, MMA5? and iAs5?were from 93.8 to 102.2%, with detection limits of 0.02, 0.08, 0.05 and

0.07 lg/l, respectively. Taking into consideration the sta-bility of urinary arsenic profiles, the arsenic assays of two urine samples in 1988–1989 and 2004 were performed within 6 months postcollection, respectively [24].

Genotyping

Genotyping for SNPs in AS3MT, PNP, GSTO1 and GSTO2 was performed with the ABI PRISM 7300 Real-time PCR System (Applied Biosystems Inc., Foster City, CA) using the 50nuclease allelic discrimination (Taqman) assay. In brief, the specific wild type and variant probes (MGB probes) were 50labeled with the VIC reporter dye and the 6-FAM reporter dye, respectively. Assays were run with the predesigned primer and probe. SNP Genotyping Assays: C__25804 287_10 (AS3MT), C___8921562_10 (PNP), C___8921 603_10 (PNP), C__11309430_10 (GSTO1) and C___32231 36_1_ (GSTO2) (www.appliedbiosystems.com) using 29 ABI TaqMan Universal PCR Master Mix, 209 Probe/primer assay Mix and 20 ng of genomic DNA in a total reaction volume of 20 ll through the default fast cycling conditions in 96-well plates. Thermocycling parameters started with an initial denaturation step of 95C for 10 min followed by 40 cycles of 92C for 15 s and 60C for 1 min. The com-pletion time was about 1.5 hours. A postamplification plate read was used for allelic discrimination using the Sequence Detection Software (v1.3) supplied with the instrument. Random repeats of 10% of the 208 samples genotyped yielded 100% reproducibility, validating genotyping proce-dures. However, DNA quality was insufficient for C__2580 4287_10 (AS3MT) genotyping in four participants, C___89 21562_10 (PNP) genotyping in four participants, C___8921 603_10 (PNP) genotyping in five participants, C__11309 430_10 (GSTO1) genotyping in nine participants and C___3223136_1_ (GSTO2) genotyping in nine participants. Statistical analysis

All data analyses were performed using the SAS statistical package (SAS, version 8.0, Cary, NC). Total arsenic con-centration (lg/l) was determined as the sum of iAs3?, iAs5?, MMA5? and DMA5?. The relative proportion of each arsenic species (InAs%, MMA% and DMA%) was calculated by dividing the concentration of each arsenic species by the total arsenic concentration. The methylation ratio (MR) of arsenic metabolism was calculated by dividing the sum of MMA and DMA by the total arsenic concentration. Then, the secondary methylation index (SMI) was calculated as DMA/MMA. A log10 -transfor-mation was applied to the urinary arsenic concentration approximating the data to a normal distribution before statistical analyses. The changed extent (D) of each urinary arsenic profile was the difference of second estimate (2004)

(4)

minus baseline estimate (1988–1989). Student’s t-test or chi-square test was used to compare the differences in continuous or categorical variables between cases and controls. Paired t-test was adapted to evaluate intra-indi-vidual variability of urinary arsenic profiles over the 15 years. The frequency distribution of polymorphisms of AS3MT, PNP, GSTO1 and GSTO2 were tested among controls to evaluate the Hardy–Weinberg equilibrium. We initially carried out the analyses under a co-dominant inheritance model (three genotypes separated) for all analyses of gene association. Subjects of heterozygous and homozygous variant genotypes for PNP rs1760940 and rs1713420 were few; so, we adopted a dominant model (heterozygous grouped with the homozygous for the minor allele) in the following analysis. We calculated the cumu-lative incidence of the variant genotypes for the follow-up cohort. Person-years at risk were calculated as the sum of follow-up times for all subjects. We used Cox proportional hazards models to estimate the multivariate hazard ratios (HR) and 95% confidence intervals (CI) of cancer risk associated with arsenic metabolism related gene polymor-phisms. All models were adjusted for age, sex and educa-tion level. For the joint effect analysis, the cutoff points of urinary arsenic profiles and the changed extent (D) of each urinary arsenic profile were respective medians of the controls. Furthermore, Kaplan–Meier curves and the log-rank test were performed to compare the cancer incidences of the different groups. Finally, we utilized general linear models (GLM), including the fixed effect, to account for the relationship between polymorphisms of AS3MT, PNP, GSTO1, GSTO2 and different urinary arsenic profiles.

Results

Of 208 residents enrolled in our study, during a median follow-up period of 15.74 years (range: 2.87–19.38 years), 17 residents contracted cancers (8%) and 191 (92%) were cancer-free controls. Sociodemographic data, genotypes and hazard ratios of 208 subjects are shown in Table1. Elderly subjects and subjects with lower educational levels had a higher cancer risk than younger subjects and those with higher educational levels (data not shown). Subjects with variant allele frequency of AS3MT, PNP rs1760940, PNP rs1713420, GSTO1 and GSTO2 were 29, 8, 8, 18 and 22%, respectively. Overall, the variant GSTO1 AA geno-type was associated with increased cancer risk, even when adjusted by baseline urinary arsenic species value and the changed extent of total arsenic concentration. How-ever, polymorphisms of AS3MT, PNP rs1760940, PNP rs1713420 and GSTO2 did not show a significant effect on the cancer risk (data not shown). Distribution of different urinary arsenic profiles in all subjects and the comparison

between baseline and second urinary arsenic species mea-surements are shown in Table2. We found a significant intra-individual decrease in urinary total arsenic, InAs% and MMA%, and increase in DMA% and MR as well as DMA/ MMA during the 15-year follow-up interval. Subjects with higher baseline urinary MMA% and lower changed extent of MMA% had approximately 3-fold cancer risk than those with lower baseline MMA% and higher changed extent of MMA%, although it was borderline statistically significant (Table3). The joint effects of other urinary arsenic species indices and their change value were not observed. Figure1 illustrates that the changed extent in MMA% was signifi-cantly related to the cancer incidences but that of baseline MMA% was not. Subjects with higher baseline MMA% values in combination with lower changed MMA% values contracted cancer earlier than other groups. Similarly, we did not find any significant differences in cancer incidence for other combinations of urinary arsenic species indices

Table 1 Sociodemographic data and genotype information of 17

newly diagnosed cancer patients and 191 controls from an arseniasis hyperendemic area All cancer patients (n = 17) Controls (n = 191) p value

Age (years) (mean ± SE) 53.18 ± 1.47 45.90 ± 0.66 B0.01

Male (%) 6 (35.29) 69 (36.13) 0.95

Education (%) B0.01

Elementary school 10 (58.82) 42 (21.99)

Junior high school or above 7 (41.17) 149 (78.01) Cigarette smokers (%) 1 (5.88) 35 (18.32) 0.19 Alcohol drinkers (%) 4 (23.53) 22 (11.52) 0.15 As3MT rs3740393 0.69 GG 8 (47.06) 89 (47.59) GC 7 (41.18) 86 (45.99) CC 2 (11.76) 12 (6.42) PNP rs1760940 0.37 AA 15 (93.75) 161 (85.64) AC or CC 1 (6.25) 27 (14.36) PNP rs1713420 0.73 AA 13 (81.25) 158 (84.49) AG or GG 3 (18.75) 29 (15.50) GSTO1 rs4925 B 0.01 CC 10 (66.67) 118 (64.13) CA 3 (20.00) 65 (35.33) AA 2 (13.33) 1 (0.54) GSTO2 rs156697 0.21 GG 9 (60.00) 111 (60.33) GA 4 (26.67) 66 (35.87) AA 2 (13.33) 7 (3.80) SE standard error

(5)

and their changed levels (data not shown). The comparison of different urinary arsenic profiles stratified according to polymorphisms of AS3MT, PNP, GSTO1 and GSTO2 are shown in Table4. We further adapted repeated measure-ment analysis in the association between arsenic metabo-lism gene polymorphisms and urinary arsenic profiles. Through GLM models, AS3MT GC genotype carriers had significantly lower MMA% and higher DMA/MMA than GG genotype carriers for both the baseline value and the extent of changes. Furthermore, AS3MT GC genotype carriers had significantly higher baseline values for DMA% than GG genotype carriers. Significant differences in

Table 2 Distribution of urinary arsenic species in all subjects, and the comparison between baseline and second measurements for different urinary arsenic profiles Total (lg/L) InAs% MMA% DMA% MR DMA/MMA Baseline level 78.16 ± 4.65 (61.38; 4.06–52.90) 7.48 ± 0.52 (5.82; 0–82.08) 13.01 ± 0.62 (10.99; 0–73.23) 79.51 ± 0.88 (82.30; 8.13–98.27) 92.52 ± 0.52 (94.18; 17.92–100) 9.41 ± 0.58 (7.29; 0.11–75.10) Second level 57.15 ± 2.98 (44.28; 6.00–290.91) 3.04 ± 0.19 (2.36; 0–15.66) 5.64 ± 0.33 (5.14; 0–27.75) 91.32 ± 0.41 (92.22; 64.61–100) 96.96 ± 0.19 (97.64; 84.34–100.00) 49.39 ± 17.48 (15.04; 2.33–3,091.52) D -21.01 ± 5.49 (-13.07; -415.22–250.93) -4.44 ± 0.56 (-3.21; -82.08–7.90) -7.37 ± 0.71 (-6.03; -69.95–12.46) 11.81 ± 0.94 (9.27; -10.29–90.66) 4.44 ± 0.56 (3.21; -7.90–82.08) 40.32 ± 17.87 (6.82; -66.69–3,080.90) p value \ 0.01 \ 0.01 \ 0.01 \ 0.01 \ 0.01 \ 0.01 All data are shown mean ± standard error (median; minimum–maximum) p value was estimated through paired t-test after log-transformation D : The changed extent of each urinary arsenic species was second value (2004) minus baseline value (1988–1989)

Table 3 Hazard ratios for cancer risk and combination of baseline

value, and the changed extent of repeated measurements for different urinary arsenic profiles

Baseline value D Number of case/control HR (95% CI) Total arsenic \59.2 C-11.78 3/79 1.00 \-11.78 3/17 3.64 (0.71, 18.58) C59.2 C-11.78 3/17 3.63 (0.72, 18.29) \-11.78 8/78 2.54 (0.66, 9.81) InAs% \5.88 C-3.23 9/79 1.00 \-3.23 2/17 0.55 (0.11, 2.71) C5.88 C-3.23 1/17 0.49 (0.06, 3.96) \-3.23 5/78 0.44 (0.14, 1.39) MMA% \10.78 C-5.17 3/75 1.00& \-5.17 2/21 1.65 (0.26, 10.37) C10.78 C-5.17 1/21 1.76 (0.18, 17.26) \-5.17 11/74 3.21 (0.86, 12.05)# DMA% C82.4 \8.67 5/77 1.00 C8.67 3/18 2.62 (0.60, 11.53) C82.4 \8.67 0/19 – C8.67 9/77 1.55 (0.49, 4.83) MR C94.13 \3.23 9/79 1.00 C3.23 2/16 0.60 (0.12, 2.94) \94.13 \3.23 1/17 0.49 (0.06, 3.95) C3.23 5/79 0.43 (0.14, 1.35) DMA/MMA C7.39 \6.32 4/62 1.00 C6.32 1/31 0.37 (0.04, 3.42) \7.39 \6.32 3/48 0.82 (0.18, 3.87) C6.32 9/50 2.24 (0.64, 7.81)

All models were adjusted for age, sex and education level

D: The changed extent of each urinary arsenic species was second value (2004) minus baseline value (1988–1989)

(6)

DMA% were seen in different PNP genotypes (rs1760940 vs. rs1713420). In addition, for GSTO2, the gene effect on InAs% was only observed using GLM models. GSTO2 AG genotype carriers had significantly lower baseline value of InAs% than GG genotype carriers.

Discussion

To our knowledge, the present study is the first to simul-taneously evaluate the impact of AS3MT, PNP, GSTO1 and GSTO2 gene polymorphisms on cancer susceptibility in an arseniasis hyperendemic area. Based on our data, we found a positive association between the GSTO1 Ala140Asp polymorphism and cancer. Further, we discovered signifi-cant gene effects of AS3MT on MMA%, DMA% and DMA/MMA, PNP on DMA%, and GSTO2 on InAs% using GLM models.

Compared to 870 subjects who were losing to follow-up, 208 participants were younger, lower percentage of male and smokers, and higher education level. Considering uri-nary arsenic profiles, there were no differences of all indices between 208 participants and 870 subjects, except for the DMA/MMA ratio. It seemed that the healthy status of 208 participants were much better than 870 subjects. In addition, all cancer incidence (17/208) of 208 participants was smaller

than that of 870 subjects (160/870). However, we still found the important role of MMA% on cancer risk. Our previous data revealed that subjects with higher baseline values of DMA% had a lower incidence of urothelial carcinoma (UC) at a 12-year follow-up, and significant associations of cumulative arsenic exposure (CAE) and MMA% (or and DMA%) on UC risk were also found [5]. This suggests that urinary arsenic, reflecting a short-term internal dose, still played an important role in cancer incidence, even though these subjects had ceased exposure to arsenic for more than three decades. During the 15-year follow-up, concentration of total arsenic, InAs% and MMA% for all subjects was indeed significantly reduced, and levels of DMA%, MR and DMA/MMA were significantly increased. The intra-individual changed levels for urinary arsenic species appeared to be involved in individual susceptibility to arsenic biotrans-formation. There are few studies to explore the importance of the changed levels of urinary arsenic species on arsenic-related health disorders. In this study, we first estimated the changed extent of urinary arsenic species on all cancers, and found that the change in MMA% significantly affected cancer incidence, regardless of whether baseline MMA% values were high or low. Further, we observed that subjects with MMA% baseline value C 10.78% and changed extent of MMA% B 5.17% had a 3.21-fold increased cancer risk compared to those with MMA% baseline value B 10.78%

Fig. 1 Cumulative cancer

incidence for combination of baseline MMA% and change in MMA% in repeated measurement. (a) — MMA% C 10.78%; … MMA% \ 10.78%. (b) — Change in MMA% \ 5.17%;… Change in MMA% C 5.17%. (c) — MMA% C 10.78% and change in MMA% \ 5.17%; … Combined

MMA% \ 10.78% and change in MMA% C 5.17%,

MMA% \ 10.78% and change in MMA% \ 5.17% and MMA% C 10.78% and change in MMA% C 5.17% three groups

(7)

Table 4 The association between urinary arsenic profiles and polymorphisms of AS3MT, PNP, GSTO1 and GSTO2 genes Total InAs% MMA% DMA% MR DMA/MMA Baseline Second Baseline Second Baseline Second Baseline Second Baseline Second Baseline Second As3MT rs3740393 GG 74.61 ± 63.98 51.95 ± 3.23 8.17 ± 9.63 3.38 ± 0.31 14.71 ± 10.59 6.52 ± 0.52 77.12 ± 15.09 90.10 ± 0.68 91.83 ± 9.63 96.62 ± 0.31 8.72 ± 9.25 28.65 ± 6.19 GC 87.58 ± 74.29 61.39 ± 5.22 6.51 ± 4.16 2.87 ± 0.28 11.57 ± 7.29 5.08 ± 0.46 81.92 ± 9.51 92.05 ± 0.53 93.49 ± 4.16 97.13 ± 0.28 10.18 ± 7.58 37.73 ± 6.25 CC 51.51 ± 21.04 58.16 ± 11.88 7.61 ± 7.54 1.90 ± 0.50 11.18 ± 4.87 3.89 ± 1.05 81.22 ± 9.59 94.21 ± 1.19 92.39 ± 7.54 98.10 ± 0.50 9.14 ± 5.24 305.67 ± 278.64 GLM p value 0.02 0.01 0.01 PNP rs1760940 AA 80.12 ± 70.97 56.31 ± 3.22 7.78 ± 7.86 3.10 ± 0.22 13.75 ± 9.31 5.80 ± 0.37 78.47 ± 13.01 91.10 ± 0.47 92.22 ± 7.86 96.90 ± 0.22 8.80 ± 8.19 52.86 ± 20.40 A C or CC 68.74 ± 37.21 58.90 ± 7.29 4.89 ± 3.63 2.63 ± 0.44 8.55 ± 5.00 4.80 ± 0.79 86.57 ± 6.85 92.57 ± 0.84 95.11 ± 3.63 97.37 ± 0.44 13.24 ± 8.16 30.57 ± 7.26 GLM p value 0.01 PNP rs1713420 AA 81.62 ± 71.82 56.07 ± 3.27 7.79 ± 7.97 3.12 ± 22 13.41 ± 9.43 5.79 ± 0.38 78.79 ± 13.28 91.08 ± 0.48 92.21 ± 7.97 96.88 ± 0.22 9.30 ± 8.67 54.56 ± 21.23 AG or GG 63.86 ± 32.95 58.68 ± 6.91 5.35 ± 3.44 2.59 ± 0.42 10.94 ± 6.23 5.04 ± 0.66 83.71 ± 7.76 92.36 ± 0.77 94.65 ± 3.44 97.41 ± 0.42 10.10 ± 6.11 25.94 ± 4.04 GLM p value 0.04 GSTO1 rs4925 CC 81.38 ± 58.23 57.87 ± 4.19 7.37 ± 6.05 3.22 ± 0.26 13.00 ± 9.32 5.49 ± 0.44 79.62 ± 12.57 91.29 ± 0.56 92.63 ± 6.05 96.78 ± 0.26 9.71 ± 8.79 61.79 ± 27.92 CA 75.64 ± 84.28 54.87 ± 3.74 7.36 ± 9.91 2.68 ± 0.34 12.85 ± 8.68 6.10 ± 0.56 79.79 ± 13.31 91.22 ± 0.66 92.64 ± 9.91 97.32 ± 0.34 9.02 ± 7.53 31.18 ± 8.85 AA 52.57 ± 17.78 46.49 ± 13.31 6.91 ± 4.84 3.59 ± 0.85 20.50 ± 5.55 3.37 ± 2.19 72.59 ± 10.20 93.04 ± 1.77 93.09 ± 4.84 96.41 ± 0.85 3.87 ± 1.81 23.91 ± 11.93 GSTO2 rs156697 AA 82.43 ± 69.81 60.06 ± 4.48 7.38 ± 5.89 3.16 ± 0.26 13.62 ± 9.94 5.81 ± 0.43 79.00 ± 12.75 91.03 ± 0.54 92.62 ± 5.89 96.84 ± 0.26 9.35 ± 9.04 33.89 ± 4.72 AG 76.47 ± 68.85 50.19 ± 3.45 7.07 ± 10.01 2.87 ± 0.34 12.02 ± 7.50 5.81 ± 0.61 80.91 ± 13.14 91.32 ± 0.75 92.93 ± 10.01 97.13 ± 0.34 9.62 ± 7.44 30.95 ± 9.06 GG 50.93 ± 24.47 60.79 ± 11.90 9.38 ± 4.54 2.60 ± 0.73 14.30 ± 8.65 3.69 ± 1.05 76.32 ± 9.64 93.71 ± 1.11 90.62 ± 4.54 97.40 ± 0.73 8.71 ± 6.76 32.08 ± 11.33 GLM p value 0.05 GLM models were adjusted for age, sex and education level, respectively

(8)

and changed MMA% value C 5.17% (borderline signifi-cance; 0.05 \ p \ 0.1). Hence, our data revealed that MMA% seemed to be a potential biomarker of urinary arsenic profiles that reflect arseniasis. This finding was consistent with the association between urinary MMA% and cancers in some studies [4,25,26].

The intra-individual changes in urinary arsenic species might result from differences in individual genetic sus-ceptibility, especially for arsenic-related metabolic genes. The findings from Drobna et al. (2005, 2006) pointed out that the AS3MT enzyme is required for conversion of inorganic arsenic to its corresponding methylated products [8, 10]. The functional impact of the rs3740393 SNP in human AS3MT on arsenic methylation is not yet known. However, in this study we found that subjects with an AS3MT GC genotype had lower MMA%, higher DMA% and DMA/MMA than GG genotype carriers. Subjects with the GC genotype appear to have better arsenic methylation capability. Up to now, the frequency of a Met/Thr amino acid change at position 287 of AS3MT, the most relevant variant, was more pronounced in non-Asian than in Asian populations [19]. To explore this, we picked out high occurrences of the SNP of AS3MT rs3740393 in Asians from the NCBI databases. It was nevertheless hard to compare the findings to those of other studies.

PNP is an important enzyme for the reduction of arsenate to arsenite in mammalian systems [27]. In the present study, individuals with heterozygous and homozygous PNP rs1760940 or 1713420 genotypes exhibited significant gene effects on DMA% during a 15-year follow-up. The minor allele frequencies of PNP rs1760940 and 1713420 were both 8%. We further analyzed the association between polymor-phisms and iAs3?% (or iAs5?%), but we did not find any relationship between them (data not shown). De et al. (2008) found that three exonic polymorphisms of PNP, including His20His, Gly51Ser and Pro57Pro, were significantly asso-ciated with arsenicism [28]. Other polymorphisms in exons like Thr65Ala and Ala174Ala have also been found. Perhaps future studies will clarify the roles of exonic polymorphisms on urinary arsenic profiles in the future.

For GSTO1 rs4925 and GSTO2 rs156697 polymor-phisms, the minor frequency seen in subjects in this study was in accordance with that seen in some populations [29, 30] but was different from those in several other popula-tions [13, 31, 32]. In the present study, the association between GSTO1 rs4925 polymorphisms and urinary arsenic profiles was not found and it was similar to the results seen in other studies [33,34]. In addition, out of five SNPs, only the GSTO1 frequency did not fit Hardy– Weinberg equilibrium. This violation might be due to small sample size, especially considering the loss of data from two cases and seven controls during the process of GSTO1 genotyping. We found that the GSTO2 rs156697 genotype

had a significant gene effect on InAs%. However, our data do not indicate an important role for GSTO2 in cancer risk, which is in agreement with the findings for skin cancer as well as for colon and breast cancer [28,29,35].

To date, only discordant screening of SNP locations for arsenic-related genes has been done, and few works focused on the association between these SNPs and urinary arsenic profiles. Our study is the first to assess whether SNPs of arsenic-related enzyme genes affect the changed extent of urinary arsenic profiles from an arseniasis hyperendemic area in Taiwan. However, in drawing conclusions, certain aspects of our study should be taken into consideration. First, survival bias might exist in the traditional cohort study; especially, for long term follow-up. The 208 indi-viduals in the sub-cohort seemed to have a better health condition than those in the original cohort, suggesting that they had a better arsenic methylation capacity, enhancing survival. Therefore, we might underestimate the association among polymorphisms of arsenic-related genes, urinary arsenic profiles and cancer. However, we still found that individuals with higher MMA% in baseline value and lower change in MMA% contracted cancer earlier than other groups. Second, we combined all kinds of cancer types to increase statistical power, because the number of varied cancer patients was relatively small. Further, we could not examine the interaction between gene polymorphisms and changed extents of urinary arsenic species on cancer risk. In addition, there were only two measurements of urinary arsenic profiles in the repeated measurement analysis, with 15 years between them. In spite of this limitation, this represents the first attempt to address the effect of change in urinary arsenic profiles on cancer risk.

In summary, our findings showed that the combined effects of baseline MMA% and change in MMA% were strong predictors of cancer incidence. The effect of MMA% on cancer risk might be influenced to some extent by the varied genotype of AS3MT rs3740393. In the future, large-scale, whole-genome-based studies would be of benefit to further explore the role of arsenic-metabolism-related enzymes on arsenic-induced carcinogenesis.

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, NSC 94-2314-B-038-023, NSC-95-2314-B-038-007, NSC- 96-2314-B038-003 and NSC 97-2314-B-038-015-MY3). References

1. Chen CJ, Chen CW, Wu MM, Kuo TL (1992) Cancer potential in liver, lung, bladder and kidney due to ingested inorganic arsenic in drinking water. Br J Cancer 66:888–892

(9)

2. Chen CL, Hsu LI, Chiou HY et al (2004) Ingested arsenic, cig-arette smoking, and lung cancer risk: a follow-up study in arseniasis-endemic areas in Taiwan. JAMA 292:2984–2990 3. Chiou HY, Hsueh YM, Liaw KF et al (1995) Incidence of internal

cancers and ingested inorganic arsenic: a seven-year follow-up study in Taiwan. Cancer Res 55:1296–1300

4. Hsueh YM, Chiou HY, Huang YL et al (1997) Serum beta-car-otene level, arsenic methylation capability, and incidence of skin cancer. Cancer Epidemiol Biomarkers Prev 6:589–596

5. 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

6. Lin S, Shi Q, Nix FB et al (2002) A novel S-adenosyl-L-methi-onine: arsenic (III) methyltransferase from rat liver cytosol. J Biol Chem 277:10795–10803

7. Thomas DJ, Li J, Waters SB et al (2007) Arsenic (?3 oxidation state) methyltransferase and the methylation of arsenicals. Exp Biol Med (Maywood) 232:3–13

8. Drobna Z, Waters SB, Devesa V, Harmon AW, Thomas DJ, Styblo M (2005) Metabolism and toxicity of arsenic in human urothelial cells expressing rat arsenic (?3 oxidation state)-methyltransferase. Toxicol Appl Pharmacol 207:147–159 9. Drobna Z, Waters SB, Walton FS, LeCluyse EL, Thomas DJ,

Styblo M (2004) Interindividual variation in the metabolism of arsenic in cultured primary human hepatocytes. Toxicol Appl Pharmacol 201:166–177

10. Drobna Z, Xing W, Thomas DJ, Styblo M (2006) shRNA silencing of AS3MT expression minimizes arsenic methylation capacity of HepG2 cells. Chem Res Toxicol 19:894–898 11. Wood TC, Salavagionne OE, Mukherjee B et al (2006) Human

arsenic methyltransferase (AS3MT) pharmacogenetics: gene resequencing and functional genomics studies. J Biol Chem 281: 7364–7373

12. Mukherjee B, Salavaggione OE, Pelleymounter LL et al (2006) Glutathione S-transferase omega 1 and omega 2 pharmacoge-nomics. Drug Metab Dispos 34:1237–1246

13. Yu L, Kalla K, Guthrie E, Vidrine A, Klimecki WT (2003) Genetic variation in genes associated with arsenic metabolism: glutathione S-transferase omega 1–1 and purine nucleoside phosphorylase polymorphisms in European and indigenous Americans. Environ Health Perspect 111:1421–1427

14. Chowdhury UK, Zakharyan RA, Hernandez A, Avram MD, Kopplin MJ, Aposhian HV (2006) Glutathione-S-transferase-omega [MMA(V) reductase] knockout mice: Enzyme and arsenic species concentrations in tissues after arsenate administration. Toxicol Appl Pharmacol 216:446–457

15. Hernandez A, Xamena N, Surralles J et al (2008) Role of the Met(287)Thr polymorphism in the AS3MT gene on the metabolic arsenic profile. Mutat Res 637:80–92

16. Schlawicke EK, Nermell B, Concha G, Stromberg U, Vahter M, Broberg K (2008) Arsenic metabolism is influenced by poly-morphisms in genes involved in one-carbon metabolism and reduction reactions. Mutat Res (Article in Press)

17. 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

18. Meza M, Gandolfi AJ, Klimecki WT (2007) Developmental and genetic modulation of arsenic biotransformation: a gene by environment interaction? Toxicol Appl Pharmacol 222:381–387

19. Fujihara J, Kunito T, Agusa T et al (2007) Population differences in the human arsenic (?3 oxidation state) methyltransferase (AS3MT) gene polymorphism detected by using genotyping method. Toxicol Appl Pharmacol 225:251–254

20. Fujihara J, Soejima M, Koda Y, Agusa T, Kunito T, Takeshita H (2008) Asian specific low mutation frequencies of the M287T polymorphism in the human arsenic (?3 oxidation state) meth-yltransferase (AS3MT) gene. Mutat Res 654:158–161

21. Tapio S, Grosche B (2006) Arsenic in the aetiology of cancer. Mutat Res -Rev Mutat Res 612:215–246

22. Chen CJ, Hsueh YM, Lai MS et al (1995) Increased prevalence of hypertension and long-term arsenic exposure. Hypertension 25: 53–60

23. 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

24. 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 25. Steinmaus C, Bates MN, Yuan Y et al (2006) Arsenic

methyla-tion and bladder cancer risk in case–control studies in Argentina and the United States. J Occup Environ Med 48:478–488 26. Yu RC, Hsu KH, Chen CJ, Froines JR (2000) Arsenic

methyla-tion capacity and skin cancer. Cancer Epidemiol Biomarkers Prev 9:1259–1262

27. Radabaugh TR, Sampayo-Reyes A, Zakharyan RA, Aposhian HV (2002) Arsenate reductase II. Purine nucleoside phosphorylase in the presence of dihydrolipoic acid is a route for reduction of arsenate to arsenite in mammalian systems. Chem Res Toxicol 15:692–698

28. De CS, Ghosh P, Sarma N et al (2008) Genetic variants associ-ated with arsenic susceptibility: study of purine nucleoside phosphorylase, arsenic (?3) methyltransferase, and glutathione s-transferase omega genes. Environ Health Perspect 116:501–505 29. Marahatta SB, Punyarit P, Bhudisawasdi V, Paupairoj A,

Wongkham S, Petmitr S (2006) Polymorphism of glutathione S-transferase omega gene and risk of cancer. Cancer Lett 236:276–281

30. Whitbread AK, Tetlow N, Eyre HJ, Sutherland GR, Board PG (2003) Characterization of the human Omega class glutathione transferase genes and associated polymorphisms. Pharmacoge-netics 13:131–144

31. Arning L, Jagiello P, Wieczorek S, Saft C, Andrich J, Epplen JT (2004) Glutathione S-Transferase Omega 1 variation does not influ-ence age at onset of Huntington’s disease. BMC Med Genet 5:7–10 32. Morari EC, Lima AB, Bufalo NE, Leite JL, Granja F, Ward LS (2006) Role of glutathione-S-transferase and codon 72 of P53 genotypes in epithelial ovarian cancer patients. J Cancer Res Clin Oncol 132:521–528

33. 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 34. 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

35. Leite JL, Morari EC, Granja F, Campos GM, Guilhen AC, Ward LS (2007) Influence of the glutathione s-transferase gene poly-morphisms on the susceptibility to basal cell skin carcinoma. Rev Med Chil 135:301–306

數據

Table 3 Hazard ratios for cancer risk and combination of baseline value, and the changed extent of repeated measurements for different urinary arsenic profiles
Fig. 1 Cumulative cancer incidence for combination of baseline MMA% and change in MMA% in repeated measurement

參考文獻

相關文件

Although oral malodour is mostly associated with poor oral hygiene and the presence of gingivitis or even periodontitis, evidence suggests that anaerobic microorganisms present in

42 Thus, the prevention of oral cancer and its associated morbidity and mortality, hinges upon the early detection of oral precancerous lesions, allowing for histological evaluation

Our case highlights an enigmatic presentation of oral submucous fibrosis and its coexistence with oral cancer presenting with unusual neurological disturbance of the inferior

-For both root filled teeth and non-root filled teeth, the baseline PAI score was the most important predictive factor.. A high baseline PAI score increased the risk for an

RNA in situ hybridization (ISH) for detection of high-risk HPV E6/E7 mRNA was performed in the Translational Research Laboratory within the Peter MacCallum Cancer Centre of

Papillary thyroid carcinoma, the commonest histopathologic variant of thyroid cancer, has minimal potential for distant metastasis, and most reported metastatic thyroid carcinomas

Distant Metastases Staging (M0) and Detection of Synchronous

In the fourth quarter of 2012, total value of merchandise export (MOP2.01 billion) and merchandise import (MOP18.82 billion) increased by 4.8% and 5.3% respectively