Binding mode prediction of imidazothiadiazoles against insect ecdysone receptors via a combination of AlphaFold3 and molecular simulations.

Imidazothiadiazoles (ITDs) are a group of non-steroidal ecdysone agonists that possess potent binding affinity against lepidopteran ecdysone receptors (EcRs). We previously conducted quantitative structure-activity relationship (QSAR) analyses and molecular docking simulations of ITDs; however, the detailed molecular mechanism of the ligand-EcR interaction remained unclear. In this study, AlphaFold3, a deep learning model predicting biomolecular interactions, was used to elucidate the binding mode of ITDs against the receptor. AlphaFold3 predicted a range of binding mode candidates, which were screened using a series of computational methods, including clustering, molecular dynamics simulations, and binding free-energy calculations. The identified binding mode reasonably accounted for the receptor binding affinity of ITDs, including those synthesized de novo in this study, and was compatible with our previous QSAR results. The results provide robust evidence for the biological significance of the current binding model. The ITD-EcR interaction involved an induced-fit mechanism, which is hardly accessible with standard docking algorithms. The present methodology will serve as a foundation for understanding ligand-protein interactions of other pest control agents that are difficult to resolve experimentally. © 2025 Society of Chemical Industry.