Reverse chemical ecology is an innovative chemical ecology approach that consists in targeting olfactory receptors to accelerate the identification of new semiochemicals active on the behavior of crop pest insects. We recently demonstrated that machine learning approaches guided the discovery of novel natural ligands for the noctuid moth Spodoptera littoralis Olfactory Receptors 24 and 25 (SlitOR24 and SlitOR25). However, the applicability domain of the models limits the chemical space to explore to molecules similar to the training set. To overcome this limitation, a structure-based virtual screening (SBVS) protocol was optimized as follows. First, we collected experimental data of two recent studies and set up a database of 184 odorants for SlitOR24 and SlitOR25. Then the structure of the receptors has been obtained with Alphafold. The ligand binding site has been deduced from the experimental structure of the insect olfactory receptor MhOR5 (from Machilis hrabei) in complex with eugenol. Docking simulations have been performed using Vina and optimized by testing various parameters and rescoring functions. The performance of the SBVS (AUC ~0.75 for both receptors) seems sufficient to expand the chemical space of potentially active odorants.
Funding: This work has been funded by INRAE, Sorbonne Université and the French National Research Agency (ANR-16-CE21-0002). We thank BECAL and the National Council of Science and Technology of Paraguay for the doctoral fellowships awarded to G.C-V. We also benefited from funding from the French government, through the UCAJEDI “Investments in the Future” project managed by the ANR grant No. ANR-15-IDEX-01. The authors are grateful to the OPAL infrastructure from Université Côte d’Azur and the Université Côte d’Azur’s Center for High-Performance Computing for providing resources and support.
References: [1] G. Caballero-Vidal et al. Cell. Mol. Life Sci., 2021, 78, 6593-6603. [2] G. Caballero-Vidal et al. Sci. Rep, 2020, 1655.