You are here:
Development of a high throughput PBTK model for co-exposure as part the software package (httk)
Citation:
Schacht, C., M. Evans, AND J. Wambaugh. Development of a high throughput PBTK model for co-exposure as part the software package (httk). SOT, Salt Lake City, UT, March 10 - 14, 2024. https://doi.org/10.23645/epacomptox.25395520
Impact/Purpose:
Presentation to the Society of Toxicology (SOT) 63rd Annual Meeting and ToxExpo March 2024
Description:
Mixtures research addresses real exposure scenarios for humans exposed to multiple classes of chemicals present in the environment. The main goal of this research is to examine how interactions between co-exposed chemicals might affect their mutual Absorption, Distribution, Metabolism, and Excretion (ADME). Concurrent exposure to multiple chemicals can alter their toxicities by changing ADME, particularly metabolism. This work seeks to develop a co-exposures PBTK model for the R software package httk, a software tool that predicts TK scenarios for numerous chemicals. One way we can understand how target tissue concentrations are altered through co-exposure is by quantifying mechanisms of toxicokinetic interaction. Chemical mixtures can result in ADME synergism due to inhibition or induction of metabolic enzymes. Metabolic inhibition decreases clearance, resulting in a higher concentration of parent chemical, but lower production of metabolites. If the parent chemical is the cause for toxicity, then inhibition increases parent concentration, leading to a potential increase in toxicity. If the metabolites lead to toxicity, then inhibition decreases overall toxicity by decreasing metabolic production. Alternatively, metabolic induction can increase toxicity by converting relatively innocuous parent chemicals to highly toxic metabolites. As such, we have developed a framework for identifying inducer or inhibitor chemicals and applied it to a generic PBTK model for binary mixtures. This model uses mathematical descriptors of metabolic inhibition (competitive, non-competitive, uncompetitive) or induction, to rapidly model co-exposure toxicokinetics. The modeling predictions can glean information about the magnitude of effects that can be expected from binary co-exposure. Our approach focuses on chemicals metabolized by CYP 2E1, a P450 family enzyme that metabolizes volatile organic chemicals (VOCs) and certain drugs. Utilizing the knowledge that there exists relationships between logP, (as a marker lipophilicity) and metabolic inhibition, we can filter pairs of chemicals and classify them as inhibitors. Similarly, we can relate molecular weight in classes of chemicals with induction and filter these chemicals as inducers. The changes in metabolic rate can be described by modifications to the Michaelis-Menten equation described by its two parameters: maximum metabolic rate and affinity for each compound. This allows not only for predictions of tissue concentrations but for predictions of the degree to which the co-exposure produces synergism or inhibition. We verify our model with in vivo data from 48 binary chemical dose combinations of 13 VOCs. We observe that, generally, tissue concentrations of mixtures can be well predicted with a log(RMSE) of 0.71. Furthermore, we assess the degree to which the co-exposed chemical induced or inhibited the substrate by comparing the relative errors between the tissue concentrations from the chemical alone and that of the mixture and found that the model generally predicts magnitude of behavior well with a log(RMSE) of 0.16 . We further asses our model with an example of in vivo co-exposure of carbon tetrachloride and trichloroethylene, two volatile organic chemicals for which experimental data show metabolic induction at low doses of carbon tetrachloride but inhibition at higher doses. Evaluating our model with data, we were able to quantify the effects of such a case of metabolic interaction, which can help inform the process to extrapolate these methods to other similar chemicals. (This abstract does not reflect US EPA policy).
URLs/Downloads:
DOI: Development of a high throughput PBTK model for co-exposure as part the software package (httk)
POSTER.PDF (PDF, NA pp, 1284.769 KB, about PDF)