Arsenic speciation methods reveal significant shifting in murine models of human arsenic exposure

Abstract

Background: Arsenic is an environmental toxicant associated with adverse health effects in humans. Its toxicity depends largely on its chemical form, which requires advanced biochemical techniques to resolve. To date, there is little consensus on speciation methods or murine model design for human-equivalent arsenic exposure. Objectives: The study objectives were to generate a robust methodological framework for arsenic speciation of mouse tissues and address current limitations and challenges associated with murine models of human arsenic exposure. Methods: Arsenic species (arsenobetaine, inorganic arsenite, inorganic arsenate, dimethylarsinate, and monomethylarsonate) were quantified using HPLC-ICPMS against commercial standards. Enzyme-assisted digestion of tissue by papain vs. pepsin and tissue maceration by bead beating vs. mechanical homogenization were evaluated using liver samples spiked with standards. Between-lab HPLC-ICPMS data reproducibility was evaluated. Live animal exposures were conducted using C57BL/6 mice orally exposed to 1 micrograms of inorganic arsenite per gram body weight. Arsenic species in stool, liver, and bladder were compared between mice eating a standard vs. purified diet, between fasted vs. unfasted mice, and between conventional vs. germ-free mice. Background levels of arsenic in food, water, and processing reagents were also quantified. Results: Papain digestion, although more efficient, exhibited significant conversion of inorganic arsenite to arsenate. Bead beating and mechanical homogenization of tissue performed similarly, but the lysing matrix of the beads contributed a detectible amount of inorganic arsenate background. HPLC-ICPMS results between laboratories were highly correlated but differed slightly in absolute abundance. Significant levels of arsenobetaine were observed in typical mouse chow samples, highlighting the need to consider dietary impacts on speciation analysis. Large inter-individual variability was observed within and between cages of the same treatment group, which was minimally affected by diet type, fasting, or microbiome status. Microbiome status had a significant effect on arsenical speciation of all sample types. Conclusion: Improving the accuracy and reliability of murine models of human arsenic exposure is critical for understanding factors that influence disease risk. Our results provide a benchmark for evaluating and interpreting data from murine models. Importantly, experiments considering low sample sizes may not adequately account for inter-individual variability, leading to low power and erroneous conclusions.