A universal cannabinoid CB1 and CB2 receptor TR-FRET kinetic ligand binding assay.

Abstract

INTRODUCTIONThe kinetics of ligand binding to G protein-coupled receptors (GPCRs) is an important determining factor in the preclinical evaluation of a molecule. Therefore, efforts should be made to measure this property as part of any drug development plan. The original assays used to assess ligand binding kinetics were developed using radioligands. However, these types of assays are very labor-intensive, limiting their application to the later phases of the drug discovery process. Recently, fluorescence-based ligand binding assays have been developed for multiple GPCRs, demonstrating their superiority through a homogeneous format and continuous data acquisition capabilities. The overriding aim of this study was to develop a fluorescence-based homogeneous ligand binding assay to profile the kinetics of compounds binding to human cannabinoid type 1 and 2 receptors (CB1R and CB2R). METHODSWe designed and synthesized D77, a novel universal tracer based on the lower affinity non-selective naturally occurring psychoactive cannabinoid, {Delta}8-THC. Using the TR-FRET (time-resolved Forster resonance energy transfer) technique to develop an assay to study the kinetics of ligand binding to CB1R and CB2R at physiological temperature. To establish a CB1R construct suitable for this assay, it was necessary to truncate the first 90 amino acids of the flexible CB1R N-terminal domain, in order to reduce the FRET distance between the terbium cryptate (donor) and the fluorescent ligand (acceptor), while the full length CB2R construct remained functional due to its shorter N-terminus. We then used the Motulsky-Mahan competition binding model to study the binding kinetics of non-fluorescent ligands. RESULTSD77 tracer displayed affinity for the truncated human CB1R (CB1R91-472) and full length CB2R (CB2R1-360) in the nanomolar range, and competitive binding behavior with orthosteric ligands. Crucially, D77 displayed fast dissociation kinetics from both CB1R and CB2R, comparable to those of the most rapidly dissociating reference compounds tested. This unique property of D77 proved pivotal to accurately determining the on- and off-rates of the fastest dissociating compounds. Using D77, we successfully determined the kinetic binding properties of a series of CB1R and CB2R agonists and antagonists at 37{degrees}C, including rimonabant, which was marketed for the treatment of obesity but later withdrawn due to serious neurological side effects. DISCUSSIONThe kon values of molecules binding CB1R showed a difference of three orders of magnitude from the slowest associating compound, HU308 to the most rapid, rimonabant. Interestingly, we found a strong correlation between kon and affinity for compounds binding to CB1R, suggesting that the association rate is the main parameter determining the affinity of compounds binding to CB1R. For compounds binding to CB2R, both kon and koff parameters contributed as affinity determinants. However, in contrast to CB1R, a stronger correlation was found between the dissociation constant rate parameter and the affinity of these molecules, suggesting that a combination of kon and koff dictates the overall affinity of compounds binding to CB2R. Ultimately, exploring the kinetic parameters of potential cannabinoid drug candidates could help future drug development programs targeting these receptors.