Structure, function and ligand interactions of the ecdysone receptor from Daphnia magna

Abstract

Background: Endocrine disruptors are of an increasing concern to the global environment due to their ability to modulate endocrine processes and cause adverse apical effects. Invertebrates are important species in the aquatic environment and are a central subject for toxicological testing, but the effects of endocrine disruptors (EDs) in invertebrates are limited due to less knowledge concerning the endocrine systems. A molecular modeling approach can be used for high-throughput screening of potential active compounds to predict binding affinity towards a target. Homology modeling, docking and scoring studies can play an important role in risk assessment of EDs. Two homology models of the ecdysone receptor (EcR) in D. magna were constructed from resolved X-ray structures of Bemisia tabacil and Heliothis virescens EcR ligand binding domain. The models were evaluated by docking studies and an in vitro two-hybrid reporter assay as an attempt to support the constructed models, identify a possible ED target and identify ED chemicals. This project is as a part of the research council of Norway (RCN) funded and NIVA-led project EDRISK, which main goal is to develop and evaluate adverse outcome pathways of EDs in the crustacean D. magna for potential inclusion in hazard and risk assessment of EDs. Results: Docking scores of presumed active binders were good for both models. In vitro data of the presumed active binders, ponasterone A and 20-hydroxyecdysone were verified to act as agonists in the reporter assay supporting the docking results. In vitro data of TFOA, triclosan and diethyl phthalate showed that the compounds were not able to bind to the EcR, partially opposing the predicted scores. Conclusion: Theoretical studies predicted model II to be a more accurate representation of the EcR in D. magna than model I and results of experimental testing supported this prediction. The experimental testing of the selected compounds was not sufficient to fully support the predicted models since too few compounds were tested.