Technical and biological factors driving inter-individual body burden of arsenic species in murine models of human arsenic exposure.

Abstract

Arsenic is one of the most important environmental toxicants, requiring advanced analytical techniques to resolve individual species. There is little consensus on arsenic speciation methodology for in vivo studies. The objectives of this study were to generate a robust framework for arsenic speciation in murine models of human exposure and evaluate factors influencing the levels of arsenobetaine, inorganic arsenite, dimethylarsinate, monomethylarsonate, and inorganic arsenate resolved by high-performance liquid chromatography (HPLC)-inductively coupled plasma mass spectrometry (ICPMS). Enzyme-assisted digestion by papain vs. pepsin and maceration by bead beating vs. mechanical homogenization were evaluated using chemical standard spiking experiments. Dose-controlled mouse exposures to inorganic arsenite were conducted, and species detected in urine and bladder tissue were compared. Species in stool, liver, and bladder were compared between groups of mice eating a standard vs. purified diet, fasted vs. unfasted mice, and conventional vs. germ-free mice. Finally, between-lab differences in HPLC-ICPMS instrumentation/quantification procedures were evaluated. These comparisons led to several important conclusions, including: Significant conversion of inorganic arsenate to arsenite by papain, significant inorganic arsenate background in bead-beating lysing matrix, significant arsenobetaine in mice eating standard but not purified chow, significant correlation between species detected in urine and bladder, significant correlation of results between laboratories that differed in absolute quantification, and large inter-individual variability between mice of the same treatment group. Finally, diet type and the presence of a microbiome had the largest effect on arsenic species levels. Our results provide a benchmark for evaluating arsenic species in murine models, including adequate sample sizes for powering studies to avoid erroneous conclusions.