Characterization of structurally novel G protein biased CB1 agonists: Implications for drug development.

The human cannabinoid subtype 1 receptor (hCB1R) is highly expressed in the CNS and serves as a therapeutic target for endogenous ligands as well as plant-derived and synthetic cannabinoids. Unfortunately, acute use of hCB1R agonists produces unwanted psychotropic effects and chronic administration results in development of tolerance and dependence, limiting the potential clinical use of these ligands. Studies in β-arrestin knockout mice suggest that interaction of certain GPCRs, including μ-, δ-, κ-opioid and hCB1Rs, with β-arrestins might be responsible for several adverse effects produced by agonists acting at these receptors. Indeed, agonists that bias opioid receptor activation toward G-protein, relative to β-arrestin signaling, produce less severe adverse effects. These observations indicate that therapeutic utility of agonists acting at hCB1Rs might be improved by development of G-protein biased hCB1R agonists. Our laboratory recently reported a novel class of indole quinulidinone (IQD) compounds that bind cannabinoid receptors with relatively high affinity and act with varying efficacy. The purpose of this study was to determine whether agonists in this novel cannabinoid class exhibit ligand bias at hCB1 receptors. Our studies found that a novel IQD-derived hCB1 receptor agonist PNR-4-20 elicits robust G protein-dependent signaling, with transduction ratios similar to the non-biased hCB1R agonist CP-55,940. In marked contrast to CP-55,940, PNR-4-20 produces little to no β-arrestin 2 recruitment. Quantitative calculation of bias factors indicates that PNR-4-20 exhibits from 5.4-fold to 29.5-fold bias for G protein, relative to β-arrestin 2 signaling (when compared to G protein activation or inhibition of forskolin-stimulated cAMP accumulation, respectively). Importantly, as expected due to reduced β-arrestin 2 recruitment, chronic exposure of cells to PNR-4-20 results in significantly less desensitization and down-regulation of hCB1Rs compared to similar treatment with CP-55,940. PNR-4-20 (i.p.) is active in the cannabinoid tetrad in mice and chronic treatment results in development of less persistent tolerance and no significant withdrawal signs when compared to animals repeatedly exposed to the non-biased full agoinst JWH-018 or Δ9-THC. Finally, studies of a structurally similar analog PNR- 4-02 show that it is also a G protein biased hCB1R agonist. It is predicted that cannabinoid agonists that bias hCB1R activation toward G protein, relative to β-arrestin 2 signaling, will produce fewer and less severe adverse effects both acutely and chronically.