Virtual Screening of CB2 Receptor Agonists from Bayesian Network and High-Throughput Docking: Structural Insights into Agonist-Modulated GPCR Features

The relevance of CB2-mediated therapeutics is well established in the treatment of pain, neurodegenerative and gastrointestinal tract disorders. Recent works such as the crystallization of class-A G-protein-coupled receptors in a range of active states and the identification of specific anchoring sites for CB2 agonists challenged us to design a reliable agonist-bound homology model of CB2 receptor. Docking-scoring enrichment tests of a high-throughput virtual screening of 140 compounds led to 13 hits within the micromolar affinity range. Most of these hits behaved as CB2 agonists, among which two novel full agonists emerged. Although the main challenge was a high-throughput docking run targeting an agonist-bound state of a CB2 model, a prior 2D ligand-based Bayesian network was computed to enrich the input commercial library for 3D screening. The exclusive discovery of agonists illustrates the reliability of this agonist-bound state model for the identification of polar and aromatic amino acids as new agonist-modulated CB2 features to be integrated in the wide activation pathway of G-protein-coupled receptors.