Received on May 17, 2007
Accepted on October 11, 2007

Drug Monitoring and Toxicology

Arsenic Speciation Analysis in Human Saliva

¹ Analytical and Environmental Toxicology, Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Alberta, T6G 2G3, Canada; School of Environmental Sciences and Engineering, North China Electric Power University, Baoding 071003, Hebei Province, P. R. China
² Analytical and Environmental Toxicology, Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Alberta, T6G 2G3, Canada
³ Inner Mongolia Center for Endemic Disease Control and Research, Huhhot 010020, Inner Mongolia, P. R. China
⁴ National Health and Environmental Effects Research Laboratory, Human Studies Division, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711, U.S.A.

^* To whom correspondence should be addressed. E-mail: xc.le{at}ualberta.ca.

BACKGROUND: Determination of arsenic species in saliva is potentially useful for biomonitoring of human exposure and studying arsenic metabolism. Arsenic speciation in saliva has not been reported previously.

METHODS: We separated arsenic species in saliva using liquid chromatography (LC) and quantified them by inductively coupled plasma mass spectrometry. We further confirmed the identities of arsenic species by LC coupled with electrospray ionization tandem mass spectrometry. These methods were successfully applied to the determination of arsenite (As^III), arsenate (As^V), and their methylation metabolites, monomethylarsonic acid (MMA^V), and dimethylarsinic acid (DMA^V), in >300 saliva samples collected from people who were exposed to varying concentrations of arsenic.

RESULTS: The mean (range) concentrations (µg/L) in the saliva samples from 32 volunteers exposed to background levels of arsenic were As^III 0.3 [not detectable (ND) to 0.7], As^V 0.3 (ND to 0.5), MMA^V 0.1 (ND to 0.2), and DMA^V 0.7 (ND to 2.6). Samples from 301 people exposed to increased concentrations of arsenic in drinking water showed detectable As^III in 99%, As^V in 98%, MMA^V in 80%, and DMA^V in 68% of samples. The mean (range) concentrations of arsenic species in these saliva samples were (in µg/L) As^III 2.8 (0.1–38), As^V 8.1 (0.3–120), MMA^V 0.8 (0.1–6.0), and DMA^V 0.4 (0.1–3.9). Saliva arsenic correlated with drinking water arsenic. Odds ratios for skin lesions increased with saliva arsenic concentrations. The association between saliva arsenic concentrations and the prevalence of skin lesions was statistically significant (P < 0.001).

CONCLUSIONS: Speciation of As^V, As^III, MMA^V, and DMA^V in human saliva is a useful method for monitoring arsenic exposure.