Citation:

RABINOWITZ, J. R., S. LITTLE, AND M. GOLDSMITH. Computational Molecular Modeling Methods Applied to Screening for Toxicity. Presented at Society of Toxicology Annual Meeting, Baltimore, MD, March 15 - 19, 2009.

Impact/Purpose:

Each chemical has been tested under the same experimental conditions for competitive binding to rat estrogen receptors. Only 15 of these chemicals bind to the estrogen receptor, 3 – 5 orders of magnitude more weakly than estrogen. Computational docking methods that would be applicable to larger chemical data bases interacting with large numbers of targets were employed to investigate the same library of chemicals interacting with the agonist and antagonist modes of both the  and rat estrogen receptor. These methods consider the potential ligands as flexible but not the target. All active chemicals were in the top 26% of those screened. Our analysis of these results considers the importance of decreasing the false positive rate while eliminating false negatives. When a pharmacophore filter based on two hydrogen bonds was employed to restrict the space explored, a significant decrease in the false positive rate was observed. All actives are found in the top 11% of the chemicals. This approach is applicable to many of the proteins in our library of potential targets.

Description:

The risk to human health and the environment of chemicals that result from human activity often must be evaluated when relevant elements of the preferred data set are unavailable. Therefore, strategies are needed that estimate this information and prioritize the outstanding data requirements. The rate determining step of many relevant mechanisms of action requires the chemical or one of its biotransformation products (the toxicant) to interact with a specific biological macromolecule (the target). A library of potential protein targets for chemical toxicity has been developed from the Protein Data Bank (www.rcsb.org/pdb). Molecular modeling methods have been used to determine the interaction between these targets and two small libraries of environmental chemicals, the Laws 281 and the Toxcast 320 (www.epa.gov/ncct/dsstox). The Laws 281 contains small organic chemicals.

Record Details: