Cannabinoid receptor subtype 2 (CB2) is a class-A family G protein-coupled receptor (GPCR), located primarily in immune-associated tissues but also in specific regions of the brain, and implicated in several inflammatory diseases and addiction. Drugs targeting CB2 are attractive treatment alternatives for chronic neurological pain and neuroinflammatory autoimmune diseases since they avoid deleterious psychotropic effects that are associated with CB1. While drug development efforts have been primarily focused on small molecules targeting the orthosteric site, limitations of poor selectivity, lack of efficacy, and development of resistance have hampered such effort. At present, there is great interests in identifying GPCR allosteric modulators that either enhance (positive allosteric modulators, or PAMs) or inhibit (negative allosteric modulators, or NAMs) agonist-induced receptor activity. PAMs/NAMs often exhibit improved subtype selectivity and spatiotemporal sensitivity, as well as potential biased signaling properties compared to orthosteric ligands. We have recently reported a 3.2 Å cryo-EM structure of the agonist-bound human CB2-Gi complex. Based on such progress, the overall goals of this proposal are to obtain a structural understanding of CB2 allosteric modulation and use our integrated computational and experimental medicinal chemistry/biology approaches to design and synthesize novel allosteric modulators for the development of CB2-specific small-molecules with potential to treat CB2-associated maladies. Thus, we first propose to elucidate the structural basis for the action of CB2 allosteric modulators by cryo-EM and X-ray crystallography approaches. To achieve the goal, we will advance our established methods for structural studies on CB2 to obtain structure of CB2 with known PAMs or NAMs. Subsequently, we plan to perform in silico design of novel CB2 allosteric modulators by our established molecular fingerprint machine-learning (ML) computing algorithms and receptor docking approaches, on basis of our reported chemogenomics cannabinoid molecular information database (CBID) and 3D CB2-Gi cryo-EM structure; a virtual allosteric modulator library will be constructed using our fragment-based design (FBD) method and our established ML-classifiers and features-ranking will be applied for selection of virtual hits. Results will be correlated with CB2 structure-based modulator design via adapting the structural information obtained from our recent CB2-Gi cryo-EM structure and our novel molecular complex characterizing system (MCCS) algorithm. Finally, we will carry out medicinal chemistry synthesis of CB2 PAM and NAM ligands and validate them by radiometric binding and cellular functional assays. With the proof-of-evidence of our recent discovery of a putative CB2 NAM, successful completion of these Aims will provide unprecedented structural information on CB2 allosteric pockets, identify promising new CB2 allosteric modulators, and help to...