You are here:
Application of molecular target homology-based approaches to predict species sensitivities to two pesticides, permethrin and propiconozole
Citation:
LaLone, C., D. Villeneuve, L. Burgoon, C. Russom, J. Berninger, Joe Tietge, H. Helgen, M. Severson, J. Cavallin, AND G. Ankley. Application of molecular target homology-based approaches to predict species sensitivities to two pesticides, permethrin and propiconozole. Presented at Society for Environmental Toxicology and Chemistry, November 11 - 15, 2012.
Impact/Purpose:
Not applicable
Description:
In the U.S., registration of pesticide active ingredients requires a battery of intensive and costly in vivo toxicity tests which utilize large numbers of test animals. These tests use a limited array of model species from various aquatic and terrestrial taxa to represent all plants and animals potentially at risk. Predictive methods that systematically and quantitatively assess molecular target homology across species, i.e., at the molecular initiating event level of an adverse outcome pathway, show promise for identifying and ranking species most likely to respond to chemical perturbations of these protein targets. The advent and refinement of these strategies could lead to more focused and integrated approaches to testing and assessment, utilizing the most relevant species with mode of action (MOA) specific toxicity tests, thereby reducing cost and animal use. To further understand and demonstrate the capabilities of protein homology-based species sensitivity predictions, we conducted case studies with two pesticides with known MOAs: permethrin a common insecticide that targets voltage gated para-like sodium channels, and propioconozole a fungicide that inhibits sterol 14 alpha-demethylase (CYP51). Primary amino acid sequence and conserved functional domain analyses were conducted and non-target species were ranked according to their predicted relative sensitivity to the pesticides. We then compared the results of the homology analyses with empirical toxicity data as a means to demonstrate the appropriate domain(s) of applicability for such predictive methods. This presentation will describe these analyses, and additional progress made in the development of an automated web-tool for such protein-based assessments.