Science of the Total Environment

(1)

Urinary arsenic methylation capability and carotid atherosclerosis risk in subjects living in arsenicosis-hyperendemic areas in southwestern Taiwan☆

Ya-Li Huang

^a

, Yu-Mei Hsueh

^a,

⁎ , Yung-Kai Huang

^b

, Ping-Keung Yip

^c

, Mo-Hsiung Yang

^d

, Chien-Jen Chen

^e,f

aDepartment of Public Health, School of Medicine, Taipei Medical University, Taipei, Taiwan

bGraduate Institute of Medical Sciences, School of Medicine, Taipei Medical University, Taipei, Taiwan

cDepartment of Neurology, College of Medicine, National Taiwan University, Taiwan

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

eGenomics Research Center, Academia Sinica, Taipei, Taiwan

fGraduate Institute of Epidemiology, College of Public Health, National Taiwan University Taipei, Taiwan

a b s t r a c t a r t i c l e i n f o

Article history:

Received 4 June 2008

Received in revised form 18 December 2008 Accepted 22 December 2008

Available online 1 February 2009

Keywords:

Arsenite

Monomethylarsonic acid Dimethylarsinic acid Arsenate

Carotid atherosclerosis Cumulative arsenic exposure

Long-term exposure to inorganic arsenic from artesian drinking well water is associated with carotid atherosclerosis in the Blackfoot Disease (BFD)-hyperendemic area in Taiwan. The current study examined the arsenic methylation capacity and its risk on carotid atherosclerosis. A total of 304 adults (158 men and 146 women) residing in the BFD-hyperendemic area were included. The extent of carotid atherosclerosis was assessed by duplex ultrasonography. Chronic arsenic exposure was estimated by an index of cumulative arsenic exposure (CAE) and the duration of artesian well water consumption. Urinary levels of inorganic arsenite [As(III)], arsenate [As(V)], monomethylarsonic acid [MMA(V)] and dimethylarsinic acid [DMA(V)] were determined by high performance liquid chromatography linked on-line to a hydride generator and atomic absorption spectrometry (HPLC-HG-AAS). The percentage of arsenic species, primary methylation index [PMI= MMA(V) / (As(III) + As(V)]

and secondary methylation index [SMI = DMA(V) / MMA(V)] were calculated and employed as indicators of arsenic methylation capacity. Results showed that women and younger subjects had a more efficient arsenic methylation capacity than did men and the elderly. Carotid atherosclerosis cases had a significantly greater percentage of MMA(V) [%MMA(V)] and a lower percentage of DMA [%DMA (V)] compared to controls. Subjects in the highest two tertiles of PMI with a median of CAEN0 mg/L-year had an odds ratio (OR) and a 95% confidence interval (CI) of carotid atherosclerosis of 2.61 and 0.98–6.90 compared to those in the highest two tertiles of PMI with a CAE = 0 mg/L-year. We conclude that individuals with greater exposure to arsenic and lower capacity to methylate inorganic arsenic may be at a higher risk to carotid atherosclerosis.

1. Introduction

Arsenic is a ground water contaminant in various areas worldwide (Nordstrom, 2002). Blackfoot Disease (BFD), a rare peripheral vascular disease, was identiﬁed in the endemic area along the southwestern coast of Taiwan and has been correlated with the exposure to inorganic arsenic through drinking artesian well water (Tseng, 1989).

Long-term exposure to arsenic through the ingestion of drinking artesian well water has also been found to be associated with an

increased risk of diabetes mellitus (DM) (Lai et al., 1994), hypertension (Chen et al., 1995), ischemic heart disease (Chen et al., 1996), cerebral infarction (Chiou et al., 1997b) and carotid atherosclerosis (Wang et al., 2002) in a dose-dependent manner.

Arsenic in drinking water is usually found in the form of inorganic arsenate [As(V)] or arsenite [As(III)] (Shraim et al., 2002).

The arsenic methylation pathway [As(V)→As(III)→MMA(V)→MMA (III)→DMA(V)→DMA(III)] was previously considered as a detoxiﬁca- tion mechanism because monomethylarsonic acid [MMA(V)] and dimethylarsinic acid [DMA(V)], which are the endpoint products in humans, have relatively low toxicity (Yamauchi and Fowler, 1994) and are rapidly excreted in the urine (Gebel, 2002). The methylation of arsenic is catalyzed by an S-adenosylmethionine dependent arsenic methyltransferase, a cyt19 (AS3MT) gene encoded in the mouse and human genome (Aposhian et al., 2006; Thomas et al., 2004), glutathione-S-transferase omega (Aposhian et al., 2004), and purine nucleoside phosphorylases (Radabaugh et al., 2002), but the mechanisms are not completely understood. Recent in vivo and in vitro studies, however, have conﬁrmed the existence of trivalent intermediates and

☆ The study was supported by grants NSC-86-2314-B-038-038, NSC-87-2314-B-038- 029, NSC-88-2314-B-038-112, NSC-88-2318-B-038-002-M51, NSC-89-2320-B-038-013, NSC-89-2318-B-038-M51, NSC-89-2314-B-038-049, NSC-90-2320-B-038-021, NSC-91- 3112-B-038-0019, NSC-92-3112-B-038-001, NSC-92-2321-B-038-004, NSC-93-3112-B- 038-001, NSC- 94- 2314-B038-023, NSC- 95-2314-B038-007 and NSC-96-2314-B-002- 311 from the National Science Council of the ROC.

⁎ Corresponding author. Department of Public Health, School of Medicine, Taipei Medical University, No. 250 Wu-Hsing Street, Taipei 110, Taiwan. Tel.: +886 2 27361661 x 6513; fax: +886 2 27384831.

E-mail address:ymhsueh@tmu.edu.tw(Y.-M. Hsueh).

doi:10.1016/j.scitotenv.2008.12.061

Contents lists available atScienceDirect

Science of the Total Environment

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i to t e n v

(2)

products of monomethylarsonous acid [MMA (III)] and dimethylarsinous acid [DMA(III)], which are more toxic than inorganic arsenite (Thomas et al., 2001; Kitchin, 2001; Styblo et al., 2002). Evaluation of arsenic methylation efﬁciency is primarily based on the relative amounts of the different metabolites present in urine. Our recent studies have documented that subjects with greater cumulative arsenic exposure (CAE) and a greater urinary MMA(V) percentage or lower urinary DMA(V) percentage suffered from a higher risk of peripheral vascular disease (Tseng et al., 2005) and hypertension (Huang et al., 2007). Whether the metabolism of arsenic could have an effect on the risk of developing carotid atherosclerosis is an interesting issue that has not been studied before in the BFD-hyperendemic area. In this area, a contaminated water supply was present that was corrected over a 30 year period. The present study evaluated the impact of the CAE and urinary arsenic species on the development of carotid atherosclerosis among residents in the BFD-hyperendemic area.

2. Materials and methods 2.1. Study area and subjects

The study area and subjects were previously described in detail (Chen et al., 1995). In brief, the study area included Homei, Fuhsing and Hsinming Villages in the Putai Township of Chiayi County located along the southwestern coast of Taiwan. A total of 2258 subjects, 30 years of age or older, registered in the household registration office in these villages were considered. Among them, 1571 (70%) were eligible and lived in these villagesfive or more days a week. A total of 1081 (69%) of the eligible subjects were interviewed from September to December 1988. All 1081 subjects were invited to participate in the first health examination during January and February 1989, and 941 (87%) subjects participated. Biannual health examinations were then conducted. An ultrasonographic assessment of the extracranial carotid artery (ECCA) atherosclerosis was performed during the sixth examination in 1996. All subjects provided informed consent forms before specimen collection and the questionnaire interview. During this examination, a total of 451 cohort members were invited, and 397 (88%) completed the ultrasonographic assessment of ECCA, and their urinary arsenic species were measured. The Institutional Review Board of National Taiwan University approved this study.

2.2. Questionnaire interview and arsenic exposure index

Two well-trained public health nurses carried out the standardized personal interview based on a structured questionnaire. Information obtained from the interview included the history of high-arsenic artesian well water consumption, residential history, socioeconomic and demographic characteristics, alcohol consumption, smoking behavior, consumption frequency of various dietary items, and personal and family histories of hypertension, diabetes and cardio- vascular diseases.

Some subjects moved from one village to another, and the arsenic concentrations in the artesian well water of these villages were known to differ. As both the duration of exposure and the arsenic levels in the artesian well water may be associated with risk of CAI, an index of cumulative arsenic exposure (CAE) was derived to reﬂect the overall exposure to arsenic for each subject. The detailed residential history and duration of consuming high-arsenic artesian well water were used to derive the cumulative arsenic exposure (CAE) from consuming artesian well water for each study subject. The arsenic levels in the artesian well water of villages where subjects had lived were obtained from previous reports carried out in the 1960s. These water samples were collected from 155 wells located in the 42 villages where BFD is endemic (Kuo, 1964). The median arsenic concentration of artesian well water was between 0.70 and 0.93 mg/L in the early 1960s (Kuo, 1964). Arsenic levels of the artesian well water in the study area were

found to be reasonably constant in two surveys carried out by the Taiwan Provincial Institute of Environmental Sanitation (Wu et al., 1961). Because these wells have been permanently sealed, it was impossible to determine whether the arsenic levels in the artesian well water has changed. One limitation of this study was that water consumption could not be estimated for each individual. We therefore employed the CAE index in mg/L-year for a given subject, deﬁned by the following formula: CAE =Σ(Ci×Di), where Ci is the median arsenic concentration of artesian well water in mg/L in the village where the subject lived and Di is the duration of artesian well water consumption in years while residing in the village. The CAE for a given subject was considered to be unknown if the median arsenic concentration of artesian well water in any of the villages where the subject had lived was unknown. The CAE was not calculable for 93 of the 397 subjects (23.4%) in this study. The odds ratio for those without CAE data was between the odds ratio for the lowest and greatest arsenic exposure groups (data not shown). The unavailable of arsenic data for some study subjects did not appear to affect the association between arsenic exposure and carotid atherosclerosis risk.

2.3. Assessment of carotid atherosclerosis

An ultrasonographic assessment of ECCA atherosclerosis was performed with a Hewlett-Packard Sono 1000 equipped with a 7.5- MHz frequency in B-mode and 5.6-MHz frequency in pulsed Doppler mode. The duplex scanning was performed by a neurologist and a cardiologist from the Taiwan University Medical Center. Study subjects were in the supine position with the head slightly extended and rotated 45° away from the side being examined during the 6th health examination in 1996 (Wang et al., 2002). Transverse and longitudinal scans were performed to assess the common carotid artery, carotid bifurcation, and internal and external carotid arteries. Plaque of the carotid artery was deﬁned as an irregular surface, luminal encroach- ment, signiﬁcant wall thickening ≥50% of the adjacent intima media (IMT), and/or structural heterogeneity, such as an acoustic shadow.

The IMT was measured in the far wall of three segments of bilateral ECCAs. Three segments were examined on each side: the distal 1.0 cm of the common carotid artery proximal to the bifurcation, the bifurcation itself and the proximal 1.0 cm of the internal carotid artery. The IMT was automatically estimated as the mean of these six measurements. The three carotid atherosclerosis indices (CAIs) were deﬁned as follows: CAI-1, presence of plaque or IMT ≥1.0 mm; CAI-2, IMT≥1.0 mm; and CAI-3, presence only of plaque.

2.4. Specimen collection and laboratory examinations

Fasting blood samples of 10 mL were collected from the subjects and centrifuged at 3000 rpm for 15 mins at room temperature. The obtained serum used for the determination of cholesterol and triglycerides. The total cholesterol and triglycerides concentrations were measured with an autoanalyzer (Hitachi 737, USA) with reagents obtained from Boehringer Mannheim Diagnostics (Indianapolis, IN, USA).

2.5. Urine collection and the determination of arsenic species present

Urinary samples were collected from 397 subjects. To collect samples, subjects were instructed to pass urine before 7:00 PM, which was discarded, and 7:00 PM to theﬁrst void urine upon awakening the next morning was collected. The samples were collected and drawn into a 1% nitric acid rinsed PE bottle and stored at−20 °C without any additive. The samples were retrieved for the determination of urinary arsenic species within six months after collection in 1989.

The urine sample was thawed at room temperature, mixed ultrasonically andﬁltered through a Sep-Pak C18column. The analytical methods used for As(III), As(V), MMA(V), and DMA(V) were previously described (Hsueh et al., 1998). A total of 200μL of the ﬁltered urine

(3)

sample was used to separate As(III), As(V), MMA(V), and DMA(V) by high performance liquid chromatography (HPLC; Waters 501; Waters Associates, Milford, MA, USA) equipped with an anion column (Phenomenex, Nucleosil 10sB, Torrance, CA, USA) and linked on-line to a hydride generator and atomic absorption spectrometry to quan- tify the levels of various species of inorganic arsenic and its metabolites.

Recovery rates of the four arsenic species were estimated according to the following calculation: [(sample containing a standard solution of known concentration)−sample concentration]/standard solution concentration × 100. Recovery for As(III), DMA(V), MMA(V) and As(V) ranged between 93.8% and 102.2% with detection limits of 0.02, 0.06, 0.07 and 0.10μg/L, respectively. The percentage of subjects below the limits of detection for As(III), AS(V), MMA(V) and DMA(V) were 7.0% (28/

397), 3.5% (14/397), 0.3% (1/397) and 0%, respectively. The half level of the limit of detection for the three arsenic species was replaced in the data analysis. A standard arsenic reference material, SRM 2670, containing 480 ± 100μg/L arsenic was obtained from the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA) and used as a quality and internal quantity standard to check the procedural valid- ity. The mean arsenic value of SRM 2670 determined by our system was 507 ± 17 (SD)μg/L (n=4). The intra- and inter-day coefﬁcients of variation

for each of the arsenic species were used to test the reliability and wasb5% for all experiments. A limitation in this study was that the creatinine levels of the urine samples did not determine the time for the measurement of urinary arsenic species. The percentage of arsenic species was therefore used to avoid being inﬂuenced by the dilution of urine volume.

The total arsenic level was calculated as the sum of As(III), As(V), MMA(V) and DMA(V). In addition to expressing as the percentage of arsenic species, the arsenic methylation efficiency index was also defined by two other indices: the primary methylation index (PMI), which was defined as the ratio between the MMA(V) level and InAs (inorganic arsenic, As(III) + As(V)) levels; the secondary methylation index (SMI) was defined as the ratio between the DMA(V) and MMA (V) levels.

2.6. Statistical analysis

Data was analyzed by SAS 8.2 software. A total of 304 subjects that had both cumulative arsenic exposure and urinary arsenic species data were analyzed. The indices of arsenic species are presented as the mean ± standard error. Because the values were not normally

Table 1

Comparison of urinary arsenic proﬁle based on gender, age, smoking behavior, alcohol drinking, lipid proﬁle, hypertension and diabetes (DM) status

Variable No. Total urinary arsenic

content (μg/L)

Inorganic arsenic percentage

MMA(V) percentage

DMA(V) percentage

PMI SMI

Gender

Male 158 89.1 ± 5.4 7.3 ± 0.5 15.5 ± 0.7 77.1 ± 0.9 3.0 ± 0.2 7.9 ± 0.7

Female 146 69.2 ± 3.7 7.1 ± 0.7 12.1 ± 0.7 80.9 ± 1.02 3.6 ± 0.5 19.1 ± 5.6

p value 0.012 0.115 b0.001 b0.001 0.063 b0.001

Age

≤46 67 77.6 ± 7.8 9.0 ± 1.4 10.5 ± 0.8â 80.6 ± 1.7â 2.4 ± 0.4â,b 28.2 ± 12.0â

47–57 98 83.6 ± 6.8 6.5 ± 0.5 13.8 ± 1.0 79.8 ± 1.2 3.0 ± 0.4^b 10.1 ± 1.4

N57 139 77.7 ± 4.2 6.9 ± 0.4 15.6 ± 0.7â 77.6 ± 0.9â 3.9 ± 0.5â 8.3 ± 0.8â

p value 0.560 0.371 b0.001 0.007 b0.001 b0.001

Smoking behavior

No 238 77.9 ± 3.6 7.3 ± 0.5 13.6 ± 0.6 79.1 ± 0.8 3.4 ± 0.3 14.7 ± 3.5

Yes 66 85.5 ± 8.8 6.9 ± 0.5 14.8 ± 1.0 78.4 ± 1.3 2.9 ± 0.3 8.0 ± 0.8

p value 0.684 0.526 0.248 0.4536 0.568 0.307

Alcohol drinking

No 259 75.2 ± 3.1 7.3 ± 0.5 13.7 ± 0.5 78.9 ± 0.7 3.4 ± 0.3 14.08 ± 3.2

Yes 45 104.5 ± 13.9 6.5 ± 0.6 14.6 ± 1.3 79.0 ± 1.7 2.8 ± 0.3 9.2 ± 1.2

p value 0.140 0.590 0.698 0.928 0.620 0.816

Total cholesterol (μg/dL)

b200 126 74.7 ± 3.8 7.0 ± 0.5 13.1 ± 0.7 79.9 ± 0.9 2.9 ± 0.3 13.6 ± 2.5

≥200 178 83.0 ± 5.1 7.4 ± 0.6 14.4 ± 0.7 78.3 ± 0.9 3.6 ± 0.4 13.0 ± 4.3

p value 0.974 0.606 0.361 0.287 0.186 0.275

Triglyceride (μg/dL)

b150 214 83.1 ± 4.4 6.8 ± 0.5 13.4 ± 0.9 79.1 ± 0.8 3.4 ± 0.3 14.6 ± 3.8

≥150 90 71.2 ± 4.3 8.1 ± 0.8 14.1 ± 0.6 78.5 ± 1.2 3.0 ± 0.7 10.1 ± 1.5

p value 0.318 0.089 0.581 0.690 0.091 0.720

Hypertension

No 219 82.3 ± 4.4 6.9 ± 0.5 13.7 ± 0.6 79.5 ± 0.8 3.3 ± 0.3 15.0 ± 3.7

Yes 85 72.5 ± 4.4 8.0 ± 0.8 14.6 ± 0.9 77.4 ± 1.3 3.4 ± 0.5 8.9 ± 1.5

p value 0.598 0.474 0.156 0.100 0.523 0.113

DM

No 192 83.4 ± 4.4 7.1 ± 0.5 13.4 ± 0.6 79.5 ± 0.8 3.0 ± 0.2 15.8 ± 4.3

Yes 65 66.7 ± 4.7 7.6 ± 0.8 15.0 ± 1.0 77.4 ± 1.3 3.9 ± 0.6 7.2 ± 0.6

p value 0.033 0.471 0.143 0.136 0.163 0.143

Data are mean ± SE; Total arsenic = As(III) + As(V) + MMA(V) + DMA(V).

Inorganic arsenic percentage = ((arsenite + arsenate) / total urinary arsenic) × 100.

DMA (V) percentage = (DMA(V)/total urinary arsenic) × 100.

MMA(V) percentage = (MMA(V)/total urinary arsenic) × 100.

PMI = primary methylation index = MMA(V) / (arsenite + arsenate).

SMI = secondary methylation index = DMA(V) / MMA(V).

aAge≤46 compared with age N57, Pb0.05;^bage≤46 compared to age 47–57, Pb0.05.

(4)

distributed, the Wilcoxon Mann–Whitney U test was used to analyze the differences in indices of urinary arsenic species between males and females, smokers and non-smokers, alcohol consumers and non- consumers, hypertension and normotension, diabetes (DM) and non- DM, normal and abnormal total cholesterol, normal and abnormal triglycerides, and presence or absence of carotid atherosclerosis. For comparing urinary arsenic indices among the three age strata, the urinary arsenic indices were ranked and analyzed by analysis of variance (ANOVA) followed by Scheffe's post hoc test. Linear regression was used to elucidate the relationship between internal arsenic exposure (urinary arsenic capability) and CAE, IMT and the duration of consuming artesian well water. The strength of association between urinary arsenic capability and CAE, IMT and the duration of consuming artesian well water is expressed by Spearmen correlation coefﬁcient.

The logistic regression analyses were further used to estimate the relative risk (in terms of odds ratio; OR) and their 95% confidence intervals (CIs) for any individual subgroup of subjects having a particular exposure dosage of arsenic and urinary arsenic profile in relation to the referent subgroup with an odds ratio of 1.00. The cut-off points for the PMI or SMI were thefirs tertile of those indices. The cut- off point for the CAE was 0. When CAE wasN0 in healthy residents, they were averagely separated into two strata.

3. Results

A comparison of the urinary arsenic proﬁle among subgroups for gender, age, smoking behavior and alcohol drinking is presented in Table 1. A greater urinary total arsenic and %MMA(V) was observed for men and a greater %DMA(V) and SMI for women, indicating lower arsenic exposure or a more complete methylation capacity of arsenic to DMA(V) in women. The %MMA(V) and PMI increased with age and was accompanied by a lower %DMA(V) and SMI. Subjects with high

triglyceride had borderline, yet signiﬁcantly greater %InAs and lower PMI compared to those with low triglyceride. Subjects with DM had lower urinary total arsenic than non-DM subjects. The indices of arsenic species were not signiﬁcantly different between cigarette smokers and non-smokers, alcohol drinkers and non-drinkers, hypertension and normal tension subjects, DM and non-DM subjects or high and low total cholesterol subjects.

The comparison of urinary arsenic profiles between carotid atherosclerosis patients and controls defined by different CAIs is shown inTable 2. Carotid atherosclerosis patients, defined by CAI-1, CAI-2 or CAI-3, had a greater %MMA(V) and PMI than controls.

Patients deﬁned as CAI-2 had a lower SMI than controls. Furthermore,

%DMA(V) was significantly lower, %MMA(V) was slightly but significantly higher and SMI was lower in CAI-1 and CAI-3 groups compared to controls after adjusting for age, gender, hypertension, DM, smoking behavior, total cholesterol and total urinary arsenic content. These results suggest that carotid atherosclerosis patients had insufficient arsenic methylation capacity.

There was a signiﬁcant positive relationship between %MMA(V) and CAE and a negative relationship between %DMA(V), SMI and CAE after adjusting for age, gender, hypertension, DM, smoking behavior, total cholesterol and total urinary arsenic content (data not shown).

These results are consistent with our previous data (Huang et al., 2007). The %MMA(V) was positively associated while the %DMA(V) and SMI were negatively associated with the amount of time spent consuming artesian well water and the IMT. After adjusting for age, gender, hypertension, DM, smoking behavior, total cholesterol and total urinary arsenic content, the signiﬁcance disappeared in the IMT (data not shown). These results suggest that study subjects with longer arsenic exposure, showing arsenic accumulation or higher CAE, had a lower arsenic methylation capacity.

The multivariate-adjusted ORs for CAI are presented inTable 3. The association between the arsenic profile and the CAI risk was not significant (data not shown), indicating that methylation capability alone did not predict CAI risk. Compared to a PMI≤1.28, however, a significant dose–response relationship between CAE and CAI risk was observed in the subgroup of subjects who had a PMIN1.28 after Table 2

Comparison of urinary arsenic proﬁle between patients to carotid atherosclerosis and controls deﬁned by different carotid atherosclerosis indices (CAIs)

Variable No. Total urinary arsenic content (μg/L)

Inorganic arsenic percentage

MMA^V percentage

DMA^V percentage

PMI SMI

CAI-1

Controls 183 74.7 ± 3.4 7.4 ± 0.6 13.1 ± 0.6 79.5 ± 0.8 2.9 ± 0.3 16.0 ± 4.5 Patients 121 86.9 ± 6.7 6.9 ± 0.5 15.1 ± 0.9 78.1 ± 1.1 3.9 ± 0.6 9.1 ± 1.1

p value^a 0.437 0.297 0.086 0.325 0.007 0.129

p value^b 0.854 0.191 0.106 0.078 0.783 0.099

CAI-2

p value^a 0.608 0.883 0.05 0.126 0.037 0.060

p value^b 0.375 0.594 0.147 0.233 0.746 0.172

CAI-3

p value^a 0.466 0.286 0.094 0.343 0.008 0.140

p value^b 0.929 0.166 0.084 0.070 0.813 0.078

Data are mean ± SE.

CAI-1 indicates presence of plaque or intima media thickness (IMT)≥1.0 mm; CAI-2, IMT≥1.0 mm; CAI-3, presence of plaque.

Inorganic arsenic percentage = ((arsenite + arsenate) / total urinary arsenic) × 100.

DMA(V) percentage = (DMA(V)/total urinary arsenic) × 100.

MMA(V) percentage = (MMA(V)/total urinary arsenic) × 100.

aWilcoxon Mann–Whitney U test.

b Arsenic species indices were ranked and tested by a linear regression model and adjusted for age, gender, hypertension, diabetes, smoking behavior, total cholesterol and total urinary arsenic content.

Table 3

Multivariate analysis for carotid atherosclerosis by cumulative arsenic exposure and arsenic methylation capability

Variables Odds ratio^a

(95% conﬁdence interval)

Odds ratio^b

(95% conﬁdence interval) PMI≤1.28

CAE = 0 1.00 1.00

0bCAE≤12.1 1.45 (0.21–10.32) 0.65 (0.03–14.43)

CAEN12.1 0.80 (0.09–7.06) 0.50 (0.03–7.68)

PMIN1.28

CAE = 0 1.00^c,⁎⁎ 1.00^c,⁎

0bCAE≤12.1 0.90 (0.32–2.51) 0.77 (0.12–4.12)

CAEN12.1 3.61 (1.15–8.98)⁎⁎ 2.61 (0.98–6.90)⁺

SMI≤5.18

CAE = 0 1.00^c,+ 1.00

0bCAE≤12.1 1.07 (0.24–4.83) 0.71 (0.12–4.12)

CAEN12.1 2.80 (0.81–9.66) 2.11 (0.52–8.55)

SMIN5.18

CAE = 0 1.00^c,⁎ 1.00

0bCAE≤12.1 0.97 (0.32–3.00) 1.02 (0.27–3.84)

CAEN12.1 2.64 (0.90–7.74)⁺ 2.17 (0.65–7.20)

CAE = cumulative arsenic exposure (mg/L-year).

aAge and gender were adjusted.

b Age, gender, hypertension, diabetes, smoking behavior, total cholesterol and total urinary arsenic content were adjusted.

c Trend test⁺0.05bpb0.1 ⁎pb0.05 ⁎⁎pb0.01.

(5)

adjusting for DM, hypertension, smoking behavior, total cholesterol and total urinary arsenic content. Conversely, a CAI risk was related to an increase in CAE either in the subgroup of subjects who had an SMIN5.18 or an SMI≤5.18 after adjusting for DM, hypertension, smoking behavior, total cholesterol and total urinary arsenic content.

These results suggested subjects with a low arsenic methylation capability (i.e., a PMIN1.28) and their CAI risk was signiﬁcantly increased when they had a high cumulative arsenic exposure. The OR and 95% CI of the CAI in subjects with a PMIN1.28 and a CAEN12.1 mg/

L-year was 2.61 (0.98–6.90) compared to subjects with a PMIN1.28 and a CAE = 0 mg/L-year.

4. Discussion

Our previous study in the BFD hyperendemic area described a dose–response relationship between CAE and carotid atherosclerosis risk diagnosed by CAI-1, CAI-2 or CAI-3 (Wang et al., 2002). Using CAE as an exposure index may reﬂect the cumulative dosage of long term exposure to arsenic in individual subjects. A limitation for this study, however, was that the arsenic exposure assessment was unable to be based on the nearest well water to the place of residence instead of the median village levels and also could not be based on individual water consumption. The urinary arsenic species metabolically related to inorganic arsenic are better biomarkers of exposure to inorganic arsenic than total arsenic. This includes the high levels of organic arsenic derived from dietary intake of sea food (Buchet et al., 1980). An individuals' capacity to metabolize and detoxify the ingested inorganic arsenic is believed to have an important role on disease development (Tseng, 2007).

The capacity to metabolize inorganic arsenic differs among individuals and inﬂuences the biological effects on various organ systems. In general, 10–30% of inorganic arsenic, 10–20% of MMA(V) and 60–80% of DMA(V) are excreted in the urine, but the arsenic metabolic proﬁle varies among different animal species (Vahter, 1999).

The urinary arsenic species can provide useful insight into the arsenic methylation capacity in different animal species and humans (Francesconi and Kuehnelt, 2004). The detection of the transient metabolites of MMA(III) and DMA(III), however, depends on the conditions of sample storage and their concentration in the urine, which was beyond the analytical method at the time in 1989. Levels of trivalent methylated metabolites in the urine are expected to be very low, since these metabolites have short half-lives and are therefore considered not to be suitable markers for arsenic methylation at the present time (Del Razo et al., 2001; Francesconi et al., 2004). Further investigations that focus on the association between these highly toxic arsenic metabolic intermediates and clinical diseases would be potentially meaningful. In the arsenic methylation process, MMA(V) must be reduced to MMA(III) and methylated to DMA(V), which is the endpoint product in humans. The rat is the only species that excretes signiﬁcant amounts of trimethylarsine oxide (TMAO) (Cohen et al., 2006). It cannot be excluded, however, that the higher MMA(V) in the urine is a reﬂection of higher MMA(III) in the blood or inside the cells, where the injury was induced by arsenic. The MMA(III) and DMA(III), however, are unstable, and they rapidly oxidize to MMA(V) and DMA (V), respectively (Gong et al., 2001).

Because of the high salinity of shallow well water, residents had drunk artesian well water for more than 50 years. Tap water was implemented in the early 1960s, but the coverage remained low until early 1970s. Artesian well water was no longer used for drinking after the mid-1970s. The arsenic concentration allowance in public water supplies in Taiwan was 50μg/L, and a new standard of 10 μg/L was announced in 2000. The total urinary arsenic content of the residents in this area was higher than residents who lived in Taipei city (Hsueh et al., 2002). Arsenic may come from other potential sources, as at least 12 arsenic metabolites were biotransformed by arsenosugar ingestion in humans (Francesconi et al., 2002) and a novel thioarsenic

metabolite was found in human urine after ingestion of an arsenosugar (Raml et al., 2005). The data for seafood ingestion was not available in the questionnaire during this study. Our previous study, however, showed the frequencies offish, shellfish, and seaweed dietary intake were not significantly correlated with urinary arsenic species. In addition, the frequency of seaweed ingestion was not related to urinary DMA(V) (Hsueh et al., 2002), a result that appears contradictory to the observation from Japanese volunteers after consumption of seaweed (Ma and Le, 1998). Any variation of seaweed intake might be biased in the identification of urinary %DMA(V) in both cases of CAI and healthy residents, thereby causing a non- differential misclassification and displacement of the relative risk toward the null.

In this study, we demonstrated that patients with carotid atherosclerosis diagnosed by CAI-1 or CAI-3 had lower %InAs and % DMA(V) and greater %MMA(V) and PMI compared to normal subjects, indicating an efficient primary methylation capacity and an inefficient secondary methylation capacity related to CAI development. Our previous studies found that CAE was significantly related to the total urinary arsenic level and a lower capacity of methylate inorganic arsenic to DMA(V). These patients were found to have a greater risk of hypertension (Huang et al., 2007) and peripheral vascular disease (Tseng et al., 2005) in patients from the BFD hyperendemic area in Taiwan. Here, we also found that longer durations of consuming artesian well water, indicating increased arsenic accumulation, were associated with more inefficient arsenic methylation capacity. In addition, CAE was positively related to %MMA(V) and inversely related to %DMA(V) and SMI. These results suggest that individual variability for metabolizing inorganic arsenic is related to arsenic exposure.

Previous findings have demonstrated that traditional CAI risk factors, such as DM and hypertension (Hara et al., 2007; Taylor et al., 2007), may play an important role in the development of CAI. We, however, did notfind that DM and hypertension has any influence in the impact of arsenic methylation capacity. By adjusted these traditional CAI risk factors, however, we found that in subjects with low arsenic methylation capability (PMIN1.28 or SMI≤5.18), the CAI risk was slightly significantly increased as the cumulative arsenic exposure increased.

The mechanisms of arsenic-induced atherosclerosis are not fully understood. The possible mechanisms of arsenic-induced atherosclerosis were proposed to induce cellular redox alteration (Pi et al., 2002), impair nitric oxide homeostasis (Pi et al., 2003), enhance coagulation activity (Simeonova and Luster, 2004), elevate homo- cysteinemia in individuals with high levels of MMA% in urine (Wu et al., 2006) and increase expression of inﬂammatory molecules (Wu et al., 2003). Studies that have explored the mechanism of methylated arsenic-induced atherosclerosis, however, are rare.

Several factors related to general health may affect arsenic methylation ability. This study found signiﬁcantly greater total arsenic and %MMA(V) and lower %DMA(V) in males compared to females.

These results suggest that females have lower arsenic exposure and/or have an influence of sex steroids, leading to a more efficient methylation capability compared to males as previously reported (Pu et al., 2007; Steinmaus et al., 2007; Lindberg et al., 2007). We found that the elderly had a less efficient arsenic methylation capacity compared to younger people, an observation consistent with our previousfindings (Hsueh et al., 1998). An Inner Mongolia study found that children had a greater capacity for the secondary methylation of arsenic compared to adults exposed to the same arsenic levels in drinking water (Sun et al., 2007). Thesefindings suggest that arsenic metabolism can be influenced by age. A more efficient arsenic methylation was observed in overweight or obese females compared to normal weight males (Lindberg et al., 2007). In this study, however, we did notfind a relationship between the body mass index and arsenic methylation capacity (data not shown). Our previous study reported that high urinary selenium may reflect a greater dietary

(6)

intake of selenium and may also alter the arsenic methylation (Hsueh et al., 2003). Subjects were recruited from western Nevada and Kings County in California regions that had arsenic levels of nearly 100μg/L in their drinking water. Subjects with lower dietary protein, iron, zinc and niacin intake excreted a greater amount of inorganic arsenic and had a greater %MMA and a lower %DMA compared to subjects with higher intakes of these nutrients (Steinmaus et al., 2005). Another study in Bangladesh adults showed that dietary intakes of cysteine, methionine, niacin, vitamin B-12 and choline modulated arsenic metabolism (Heck et al., 2007). A clinical trial demonstrated that folic acid supplementation in subjects with low plasma folate enhanced arsenic methylation (Gamble et al., 2006). Another study reported that fasting for a period of 12 hours resulted in a signiﬁcant increase in the

%MMA (Brima et al., 2007). The urine collection protocol required 8 hours of fasting in this study. The identification of urinary arsenic profile might therefore be biased in both cases and controls, resulting in non-differential misclassification and a displaced odds ratios toward the null.

Genetic polymorphisms may also contribute to individual variability in the biotransformation of arsenic. Subjects with the GST M1 null genotype had an increased percentage of inorganic arsenic in their urine, while those with the GST T1 null genotype had an elevated

%DMA (Chiou et al., 1997a). Recently, a report from Central Europe found that the M287T polymorphism in the AS3MT gene and A222V polymorphism in the MTHFR gene influenced arsenic metabolism (Lindberg et al., 2007). The intronic polymorphisms in AS3MT (G12390C, C14215T, and G35991A) were associated with a lower % MMA and a greater %DMA in the urine of Argentinean subjects (Schlawicke et al., 2007), and that GSTM1, GSTT1, MTR and MTHFR polymorphisms are also responsible for inter-individual variability in arsenic metabolism. Subjects with the MTHFR 677TT and 1298AA variant genotypes excreted a significantly greater proportion of inorganic arsenic and a lower proportion of DMA. The null genotype of GSTM1 excreted a significantly greater percentage of MMA compared to those with an active genotype (Steinmaus et al., 2007).

In addition, due to individual variability in arsenic metabolism, there may be difference in disease susceptibility. The interaction between the GSTT1 wild type and the secondary methylation index modiﬁed the risk of skin lesions among individuals exposed to arsenic in Bangladesh (McCarty et al., 2007). Another report from Bangladesh found a combined effect of either an MTHFR 677TT/1298AA and 677CT/1298AA diplotype or a GSTω1 diplotype coupled with a greater

%MMA increased the risk of skin lesions (Ahsan et al., 2007). It seems plausible to suggest that both genetic and acquired factors inﬂuence arsenic biotransformation and induce arsenic related health consequences.

Wu and colleagues found that either an increased efficiency of methylation from arsenite to MMA(V) or an increased methylation efficiency from MMA(V) to DMA(V) may be related to the influence of S-adenosylmethionine consumption and plasma homocysteine levels on the development of carotid atherosclerosis (Wu et al., 2006). Low arsenic exposure was positively related to high pulse pressure, especially among those with a low intake of nutrient-B vitamins (Chen et al., 2007). Another study demonstrated that genetic polymorphisms of GSTP1 and P53 modified the risk of carotid atherosclerosis when subjects were exposed to arsenic (Wang et al., 2007). It is possible that a trimethylated arsenic metabolite inhibits insulin-stimulated glucose uptake, impairs glucose tolerance (Paul et al., 2007) or generates reactive oxygen species and free radicals (Kitchin and Ahmad, 2003). Based on these studies, genetic susceptibility, homocysteine, S-adenosylmethionine consumption or low intakes of nutrients may influence the arsenic methylation capability.

Inefﬁcient methylation capability, however, may produce oxidative stress that induces carotid atherosclerosis. The precise mechanism for arsenic metabolite-related atherosclerosis requires further investiga- tion. In conclusion, theﬁndings of this study suggest that individuals

with a greater arsenic exposure and a lower capacity to methylate inorganic arsenic to DMA(V) may have a higher risk of carotid atherosclerosis.

Acknowledgements

The authors wish to thank all the participants in the study for their assistance during the study and data collection. The study was supported by grants NSC-86-2314-B-038-038, NSC-87-2314-B-038- 029, NSC-88-2314-B-038-112, NSC-88-2318-B-038-002-M51, NSC-89- 2320-B-038-013, NSC-89-2318-B-038-M51, NSC-89-2314-B-038-049, NSC-90-2320-B-038-021, NSC-91-3112-B-038-0019, NSC-92-3112-B- 038-001, NSC-92-2321-B-038-004, NSC-93-3112-B-038-001, NSC-94- 2314-B038-023, NSC-95-2314-B038-007 and NSC-96-2314-B-002-311 from the National Science Council of the ROC.

References

Ahsan H, Chen Y, Kibriya MG, Slavkovich V, Parvez F, Jasmine F, Gamble MV, Graziano JH.

Arsenic metabolism, genetic susceptibility, and risk of premalignant skin lesions in Bangladesh. Cancer Epidemiol Biomarkers Prev 2007;16:1270–8.

Aposhian HV, Zakharyan RA, Avram MD, Sampayo-Reyes A, Wollenberg ML. A review of the enzymology of arsenic metabolism and a new potential role of hydrogen peroxide in the detoxication of the trivalent arsenic species. Toxicol Appl Pharmacol 2004;198:327–35.

Aposhian HV, Aposhian MM, Aposhian HV, Aposhian MM. Arsenic toxicology:ﬁve questions. Chemical Research in Toxicology 2006;19:1-15.

Brima EI, Jenkins RO, Lythgoe PR, Gault AG, Polya DA, Haris PI. Effect of fasting on the pattern of urinary arsenic excretion. J Environ Monit 2007;9:98-103.

Buchet JP, Lauwerys R, Roels H. Comparison of several methods for the determination of arsenic compounds in water and in urine. Their application for the study of arsenic metabolism and for the monitoring of workers exposed to arsenic. Int Arch Occup Environ Health 1980;46:11–29.

Chen CJ, Hsueh YM, Lai MS, Shyu MP, Chen SY, Wu MM, Kuo TL, Tai TY. Increased prevalence of hypertension and long-term arsenic exposure. Hypertension 1995;25:53–60.

Chen CJ, Chiou HY, Chiang MH, Lin LJ, Tai TY. Dose–response relationship between ischemic heart disease mortality and long-term arsenic exposure. Arterioscler Thromb Vasc Biol 1996;16:504–10.

Chen Y, Factor-Litvak P, Howe GR, Graziano JH, Brandt-Rauf P, Parvez F, van Geen A, Ahsan H. Arsenic exposure from drinking water, dietary intakes of B vitamins and folate, and risk of high blood pressure in Bangladesh: a population-based, cross- sectional study. Am J Epidemiol 2007;165:541–52.

Chiou HY, Hsueh YM, Hsieh LL, Hsu LI, Hsu YH, Hsieh FI, Wei ML, Chen HC, Yang HT, Leu LC, Chu TH, Chen-Wu C, Yang MH, Chen CJ. Arsenic methylation capacity, body retention, and null genotypes of glutathione S-transferase M1 and T1 among current arsenic-exposed residents in Taiwan. Mutat Res 1997a;386:197–207.

Chiou HY, Huang WI, Su CL, Chang SF, Hsu YH, Chen CJ. Dose–response relationship between prevalence of cerebrovascular disease and ingested inorganic arsenic. Stroke 1997b;28:1717–23.

Cohen SM, Arnold LL, Eldan M, Lewis AS, Beck BD. Methylated arsenicals: the implications of metabolism and carcinogenicity studies in rodents to human risk assessment. Crit Rev Toxicol 2006;36:99-133.

Del Razo LM, Styblo M, Cullen WR, Thomas DJ. Determination of trivalent methylated arsenicals in biological matrices. Toxicol Appl Pharmacol 2001;174:282–93.

Francesconi KA, Kuehnelt D. Determination of arsenic species: a critical review of methods and applications, 2000–2003. Analyst 2004;129:373–95.

Francesconi KA, Tanggaar R, McKenzie CJ, Goessler W, Francesconi KA, Tanggaar R, McKenzie CJ, Goessler W. Arsenic metabolites in human urine after ingestion of an arsenosugar. Clinical Chemistry 2002;48:92-101.

Gamble MV, Liu X, Ahsan H, Pilsner JR, Ilievski V, Slavkovich V, Parvez F, Chen Y, Levy D, Factor-Litvak P, Graziano JH. Folate and arsenic metabolism: a double-blind, placebo-controlled folic acid-supplementation trial in Bangladesh. Am J Clin Nutr 2006;84:1093–101.

Gebel TW. Arsenic methylation is a process of detoxiﬁcation through accelerated excretion. Int J Hyg Environ Health 2002;205:505–8.

Gong ZL, Lu XF, Cullen WR, Le XC. Unstable trivalent arsenic metabolites, monomethylarsonous acid and dimethylarsinous acid. J Anal At Spectrom 2001;16:1409–13.

Hara A, Ohkubo T, Kikuya M, Shintani Y, Obara T, Metoki H, Inoue R, Asayama K, Hashimoto T, Harasawa T, Aono Y, Otani H, Tanaka K, Hashimoto J, Totsune K, Hoshi H, Satoh H, Imai Y. Detection of carotid atherosclerosis in individuals with masked hypertension and white-coat hypertension by self-measured blood pressure at home: the Ohasama study. J Hypertens 2007;25:321–7.

Heck JE, Gamble MV, Chen Y, Graziano JH, Slavkovich V, Parvez F, Baron JA, Howe GR, Ahsan H. Consumption of folate-related nutrients and metabolism of arsenic in Bangladesh. Am J Clin Nutr 2007;85:1367–74.

Hsueh YM, Huang YL, Huang CC, Wu WL, Chen HM, Yang MH, Lue LC, Chen CJ. Urinary levels of inorganic and organic arsenic metabolites among residents in an arseniasis- hyperendemic area in Taiwan. J Toxicol Environ Health A 1998;54:431–44.

Hsueh YM, Hsu MK, Chiou HY, Yang MH, Huang CC, Chen CJ. Urinary arsenic speciation in subjects with or without restriction from seafood dietary intake. Toxicol Lett 2002;133:83–91.

(7)

Hsueh YM, Ko YF, Huang YK, Chen HW, Chiou HY, Huang YL, Yang MH, Chen CJ.

Determinants of inorganic arsenic methylation capability among residents of the Lanyang Basin, Taiwan: arsenic and selenium exposure and alcohol consumption.

Toxicol Lett 2003;137:49–63.

Huang YK, Tseng CH, Huang YL, Yang MH, Chen CJ, Hsueh YM. Arsenic methylation capability and hypertension risk in subjects living in arseniasis-hyperendemic areas in southwestern Taiwan. Toxicol Appl Pharmacol 2007;218:135–42.

Kitchin KT. Recent advances in arsenic carcinogenesis: modes of action, animal model systems, and methylated arsenic metabolites. Toxicol Appl Pharmacol 2001;172:249–61.

Kitchin KT, Ahmad S. Oxidative stress as a possible mode of action for arsenic carcinogenesis. Toxicol Lett 2003;137:3-13.

Kuo TL. Arsenic content of artesian well water in endemic area of chronic arsenic poisoning. Taiwan: Institute of Pathology, National Taiwan University; 1964. p. 7-13.

Lai MS, Hsueh YM, Chen CJ, Shyu MP, Chen SY, Kuo TL, Wu MM, Tai TY. Ingested inorganic arsenic and prevalence of diabetes mellitus. Am J Epidemiol 1994;139:484–92.

Lindberg AL, Kumar R, Goessler W, Thirumaran R, Gurzau E, Koppova K, Rudnai P, Leonardi G, Fletcher T, Vahter M. Metabolism of low-dose inorganic arsenic in a central European population: inﬂuence of sex and genetic polymorphisms. Environ Health Perspect 2007;115:1081–6.

Ma M, Le XC. Effect of arsenosugar ingestion on urinary arsenic speciation. Clin Chem 1998;44:539–50.

McCarty KM, Chen YC, Quamruzzaman Q, Rahman M, Mahiuddin G, Hsueh YM, Su L, Smith T, Ryan L, Christiani DC. Arsenic methylation, GSTT1, GSTM1, GSTP1 polymorphisms, and skin lesions. Environ Health Perspect 2007;115:341–5.

Nordstrom DK. Public health. Worldwide occurrences of arsenic in ground water.

Science 2002;296:2143–5.

Paul DS, Harmon AW, Devesa V, Thomas DJ, Styblo M. Molecular mechanisms of the diabetogenic effects of arsenic: inhibition of insulin signaling by arsenite and methylarsonous acid. Environ Health Perspect 2007;115:734–42.

Pi J, Horiguchi S, Sun Y, Nikaido M, Shimojo N, Hayashi T, Yamauchi H, Itoh K, Yamamoto M, Sun G, Waalkes MP, Kumagai Y. A potential mechanism for the impairment of nitric oxide formation caused by prolonged oral exposure to arsenate in rabbits.

Free Radic Biol Med 2003;35:102–13.

Pi J, Yamauchi H, Kumagai Y, Sun G, Yoshida T, Aikawa H, Hopenhayn-Rich C, Shimojo N.

Evidence for induction of oxidative stress caused by chronic exposure of Chinese residents to arsenic contained in drinking water. Environ Health Perspect 2002;110:331–6.

Pu YS, Yang SM, Huang YK, Chung CJ, Huang SK, Chiu AW, Yang MH, Chen CJ, Hsueh YM.

Urinary arsenic proﬁle affects the risk of urothelial carcinoma even at low arsenic exposure. Toxicol Appl Pharmacol 2007;218:99-106.

Radabaugh TR, Sampayo-Reyes A, Zakharyan RA, Aposhian HV. 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 2002;15:692–8.

Raml R, Goessler W, Traar P, Ochi T, Francesconi KA. Novel thioarsenic metabolites in human urine after ingestion of an arsenosugar, 2′,3′-dihydroxypropyl 5-deoxy-5- dimethylarsinoyl-beta-D-riboside. Chem Res Toxicol 2005;18:1444–50.

Schlawicke EK, Broberg K, Concha G, Nermell B, Warholm M, Vahter M. Genetic polymorphisms inﬂuencing arsenic metabolism: evidence from Argentina. Environ Health Perspect 2007;115:599–605.

Shraim A, Sekaran NC, Anuradha CD, Hirano S. Speciation of arsenic in tube-well water samples collected from West Bengal, India, by high-performance liquid chromato-

graphy-inductively coupled plasma mass spectrometry. Appl Organomet Chem 2002;16:202–9.

Simeonova PP, Luster MI. Arsenic and atherosclerosis. Toxicol Appl Pharmacol 2004;198:444–9.

Steinmaus C, Carrigan K, Kalman D, Atallah R, Yuan Y, Smith AH. Dietary intake and arsenic methylation in a U.S. Population. Environ Health Perspect 2005;113:1153–9.

Steinmaus C, Moore LE, Shipp M, Kalman D, Rey OA, Biggs ML, Hopenhayn C, Bates MN, Zheng S, Wiencke JK, Smith AH. Genetic polymorphisms in MTHFR 677 and 1298, GSTM1 and T1, and metabolism of arsenic. J Toxicol Environ Health A 2007;70:159–70.

Styblo M, Drobna Z, Jaspers I, Lin S, Thomas DJ. The role of biomethylation in toxicity and carcinogenicity of arsenic: a research update. Environ Health Perspect 2002;110 (Suppl 5):767–71.

Sun G, Xu Y, Li X, Jin Y, Li B, Sun X. Urinary arsenic metabolites in children and adults exposed to arsenic in drinking water in Inner Mongolia, China. Environ Health Perspect 2007;115:648–52.

Taylor AJ, Zhu D, Sullenberger LE, Lee HJ, Lee JK, Grace KA. Relationship between glycemic status and progression of carotid intima-media thickness during treatment with combined statin and extended-release niacin in ARBITER 2. Vasc Health Risk Manag 2007;3:159–64.

Thomas DJ, Styblo M, Lin S. The cellular metabolism and systemic toxicity of arsenic.

Toxicol Appl Pharmacol 2001;176:127–44.

Thomas DJ, Waters SB, Styblo M. Elucidating the pathway for arsenic methylation.

Toxicol Appl Pharmacol 2004;198:319–26.

Tseng CH. Arsenic methylation, urinary arsenic metabolites and human diseases: current perspective. J Environ Sci Health C Environ Carcinog Ecotoxicol Rev 2007;25:1-22.

Tseng WP. Blackfoot disease in Taiwan: a 30-year follow-up study. Angiology 1989;40:547–58.

Tseng CH, Huang YK, Huang YL, Chung CJ, Yang MH, Chen CJ, Hsueh YM. Arsenic exposure, urinary arsenic speciation, and peripheral vascular disease in blackfoot disease- hyperendemic villages in Taiwan. Toxicol Appl Pharmacol 2005;206:299–308.

Vahter M. Methylation of inorganic arsenic in different mammalian species and population groups. Sci Prog 1999;82(Pt 1):69–88.

Wang CH, Jeng JS, Yip PK, Chen CL, Hsu LI, Hsueh YM, Chiou HY, Wu MM, Chen CJ.

Biological gradient between long-term arsenic exposure and carotid atherosclerosis. Circulation 2002;105:1804–9.

Wang YH, Wu MM, Hong CT, Lien LM, Hsieh YC, Tseng HP, Chang SF, Su CL, Chiou HY, Chen CJ. Effects of arsenic exposure and genetic polymorphisms of p53, glutathione S-transferase M1, T1, and P1 on the risk of carotid atherosclerosis in Taiwan.

Atherosclerosis 2007;192:305–12.

Wu HY, Chen KP, Tseng WP, Hsu CL. Epidemiologic studies on Blackfoot disease:

I. Prevalence and incidence of the disease by age, sex, occupation and geographical distribution. Mem College Med Natl Taiwan Univ 1961;7:33–50.

Wu MM, Chiou HY, Ho IC, Chen CJ, Lee TC. Gene expression of inﬂammatory molecules in circulating lymphocytes from arsenic-exposed human subjects. Environ Health Perspect 2003;111:1429–38.

Wu MM, Chiou HY, Hsueh YM, Hong CT, Su CL, Chang SF, Huang WL, Wang HT, Wang YH, Hsieh YC, Chen CJ. Effect of plasma homocysteine level and urinary monomethylarsonic acid on the risk of arsenic-associated carotid atherosclerosis. Toxicol Appl Pharmacol 2006;216:168–75.

Yamauchi H, Fowler BA. Toxicity and metabolism of inorganic and methylated arsenicals. In: Nriagu JO, editor. Arsenic in the environment, part 2: human health and ecosystem effects. New York: John Wiley and Sons; 1994. p. 33–43.