You are here:
A Pipeline for Prediction of Developmental Toxicants, Screening in Zebrafish and Correlation to HTS Data
Citation:
Bondesson, M., S. Thrikawala, R. Mirchandani, C. McCollum, N. Nguyen, M. Sobieski, R. Judson, AND C. Stephan. A Pipeline for Prediction of Developmental Toxicants, Screening in Zebrafish and Correlation to HTS Data. Society of Toxicology 59th Annual Meeting, Anaheim, CA, March 15 - 19, 2020.
Impact/Purpose:
Abstract submitted to the Society of Toxicology 59th Annual Meeting in March 2020. This study looked at how medium throughput screening data from an in vivo model in combination with modeling of ToxCast high throughput in vitro data can be used to produce testable hypotheses on modes of action of chemical exposures, and to predict apical end-point specific effects of chemical exposures.
Description:
The zebrafish embryo is an effective alternative in vivo model to rodents for studying chemically-induced perturbations because of its small size and rapid development. In addition, the availability of transgenic embryos that express fluorescence in specific tissues or organs allows for detection of tissue-specific structural malformations (apical end-points). We performed screening for skeletal and angiogenic disruption in transgenic zebrafish embryos caused by exposure to Phase I chemicals from EPA's Toxicity Forecaster (ToxCast) program. Around 300 chemicals were screened and we identified 38 skeletal and 10 angiogenic disruptors. The identified sets of skeletal and angiogenic disruptors were used to identify ToxCast assays that were affected by the same chemicals using univariate correlation analysis. Identified assays included those that measure vitamin D3 metabolism and dopamine transporter activity, and oxidative stress and nuclear receptor signaling for skeletal and angiogenic disruption, respectively. We further used the identified assays to make predictions of other chemicals, tested in the Tox21 screening project, but not tested for skeletal and angiogenic disruption in zebrafish embryos, using the Toxicological Priority Index (ToxPi) program. This program ranks compounds based on their AC50 value for the selected assays. Predicted disruptors were tested in the laboratory, and several new skeletal and angiogenic disruptors were identified. We conclude that medium throughput screening data from an in vivo model in combination with modeling of ToxCast high throughput in vitro data can be used to produce testable hypotheses on modes of action of chemical exposures, and to predict apical end-point specific effects of chemical exposures. This abstract does not necessarily represent U.S. EPA policy.