Biased Signaling at the CB1 Cannabinoid Receptor: Functional Amino Acids and Allosteric Modulators

Abstract

The CB1 cannabinoid receptor is a G-protein coupled receptor highly expressed throughout the central nervous system, that has been suggested as a target for the treatment of various disorders, including anxiety, pain and neurodegeneration. Despite the wide therapeutic potential of CB1, development of potential drug candidates has long been hindered by concerns about adverse effects, rapid tolerance development and abuse potential. Ligands that produce biased signaling have been proposed as a strategy to dissociate therapeutic and adverse effects for a variety of G-protein coupled receptors. Biased signaling involves selective activation of a signaling transducer in detriment of another, mainly involving selective activation of G-protein signaling or b-arrestin signaling. However, biased signaling at the CB1 receptor is poorly understood due to the lack of strongly biased agonists. The development of biased agonists would be aided by understanding the molecular mechanism that leads to biased signaling. Although the structure of CB1 has been resolved in the inactive state and in the canonical active state, which allows G-protein signaling, little is known about the alternative active state that allows b-arrestin biased signaling. Therefore, we set out to investigate molecular and pharmacological tools that could shed light on the mechanism of CB1 biased signaling and to characterize novel allosteric ligands with a biased signaling profile. Using molecular dynamics stimulation of CB1 bound to a ORG27569, an allosteric ligand that stimulates b-arrestin signaling and inhibits G-protein signaling, we proposed single amino acid mutations that were predicted to impact b-arrestin signaling, and expressed wild-type and mutated CB1 receptor in HEK293 cells to measure signaling through different signaling transducers. We found that N7.49 and Y7.53, two amino acids in the highly conserved NPXXY motif, were essential for b-arrestin recruitment and signaling, but mutating them to Ala and Phe, respectively, did not impact G-protein signaling. We also found that I2.43, a functionally conserved amino acid on transmembrane helix 2, negatively regulates a switch in the rotameric position of Y7.53, as mutating I2.43 to Ala reduced steric hindrance upon Y7.53 and enhanced b-arrestin1 recruitment and signaling, while mutating it to Thr, a polar residue that would further hinder Y7.53, partially inhibited b-arrestin recruitment. Therefore, we concluded that N7.49 and Y7.53 form a hydrogen bond network along with D2.50 that is essential for the alternative active state that stimulates b-arrestin biased signaling. N7.49 acts as a fulcrum on which transmembrane helix 7 can bend, and Y7.53 acts as a rotamer toggle switch, stabilizing conformational changes on the intracellular end of transmembrane helix 7. This is the first record of a molecular mechanism for CB1 b-arrestin biased signaling involving the NPXXY motif. Due to the highly conserved character of these residues, it is possible that this mechanism can also be applied to other class A G-protein coupled receptors. In addition, we characterized novel biased allosteric ligands that stimulate or inhibit b-arrestin1 signaling. Two ORG27569 analogs were found to enhance orthosteric agonist induced b-arrestin1 recruitment and extracellular-signal regulated kinase 1/2 phosphorylation (pERK), with no effect on G-protein signaling. Two pregnenolone analogs absent of the steroid scaffold were found to inhibit pERK signaling independent of Gprotein signaling, indicating that they hinder b-arrestin dependent signaling. Since these analogs are believed to mediate their effects via stimulation or inhibition of conformational changes on transmembrane helix 7, our findings support a role for this domain on the alternative active state of CB1. In contrast, a GAT211 analog, GAT1601, had no effect on recruitment of b-arrestin1, but stimulated G-protein signaling and slightly enhanced barrestin2 recruitment. This compound binds to an allosteric site, where it stimulates the canonical active state of CB1 by facilitating the outward movement of transmembrane helix 6. Altogether, the results presented in this dissertation suggest that CB1 b-arrestin biased signaling is regulated by the NPXXY motif, which stimulates conformational changes on the transmembrane helix 7/helix 8 elbow, and that stimulating or hindering these conformational changes can enhance or disrupt CB1 b-arrestin biased signaling. However, facilitating the movement of transmembrane helix 6 favors G-protein biased signaling. Our findings provide molecular and pharmacological tools that will be of great importance to structure guided drug design and to future studies on the functional consequences of biased signaling at the CB1 receptor.