The enigmatic pharmacology of GPR55

Preliminary data presented at conferences and in the patent literature introduced the possibility the orphan receptor GPR55 might account for some of the well-documented non-CB₁, non-CB₂ effects reported for certain cannabinoid ligands. Several peer-reviewed publications have recently emerged in which the pharmacology of the cannabinoids at GPR55 has been probed in more depth. Despite this, the classification of GPR55 as a cannabinoid receptor remains a contentious issue. The weight of evidence points to GPR55 as a receptor that is activated by certain cannabinoid ligands and by the bioactive lipid l-α-lysophosphatidylinsoitol. It couples to G₁₂ proteins, activates RhoA and mobilizes intracellular Ca²⁺, possibly in an agonist- and tissue-dependant manner, thus displaying ‘agonist functional selectivity’. Here, I review the recent literature in an effort to glean the key controversies and outstanding questions surrounding the interaction between cannabinoids and this orphan receptor.

Introduction

In 2005, both meeting presentations and the patent literature revealed that GPR55 might be activated by several cannabinoid ligands (for review, see Ref. [1]). GPR55 has the highest amino acid sequence homology with GPR23, GPR35, P2Y5 and CCR2 [2]. mRNA for this receptor has been detected in regions of the central nervous system (CNS) including the caudate nucleus, putamen, hippocampus, thalamus, pons, cerebellum, frontal cortex and hypothalamus 2, 3 and in the adrenal glands and gastrointestinal tract [3]. The endocannabinoid system comprises two established cannabinoid receptors (CB₁ and CB₂); GPR55 shares only 13.5% sequence identity with the CB₁ receptor and 14.4% with the CB₂ receptor. The levels of GPR55 mRNA expression are notably between three- and tenfold lower than those reported for the cannabinoid CB₁ receptor [3]. To put this finding in perspective, it is important to note that CB₁ receptor expression in the CNS is considerably higher than other G-protein-coupled receptors (GPCRs) [4]. Within the brain, the distribution of CB₁ receptors is heterogeneous; they are found predominantly on nerve terminals where they attenuate neurotransmitter release [4]. The CB₂ receptor is predominantly expressed on immune cells, although there is evidence for neuronal localization in the brainstem [5].

There is a family of endogenous cannabinoid ligands (endocannabinoids) and specific molecular machinery for the synthesis and inactivation of these ligands 6, 7. The most studied endocannabinoids are arachidonoylethanolamide (anandamide) and 2-arachidonoyl glycerol (2-AG), both of which are synthesized on demand [8]. 2-AG acts as a ‘retrograde’ signalling molecule in the CNS; it is synthesized postsynaptically and acts presynaptically to inhibit neurotransmitter release. The endocannabinoid system is involved in the pathophysiology of obesity, osteoporosis, nicotine addiction, mental illness, neuropathic and inflammatory pain, cardiovascular disorders and liver disease 6, 8. The most abundant constituents of cannabis (phytocannabinoids) include Δ⁹-tetrahydrocannabinol (Δ⁹THC), which is psychoactive, and (−)-cannabidiol (CBD), which is non-psychoactive. These compounds interact with various elements of the endocannabinoid system 6, 8.

The pharmacology of both the endocannabinoids and the phytocannabinoids is complex; their actions are mediated by cannabinoid CB₁ and CB₂ receptors and by putative non-CB₁, non-CB₂ receptors. One such novel target has been implicated in endothelium-dependent vasodilatation in isolated resistance vessels, in haemodynamic responses and in modulation of endothelial, microglial and immune-cell migration (for review, see Refs 9, 10, 11, 12). Such non-CB₁, non-CB₂ effects have been observed with a range of cannabinoid ligands including anandamide and virodhamine (which are eicosanoids) and the phytocannabinoid-like compounds O-1602 and the CBD analogue known as ‘abnormal-CBD’; these effects are antagonized by CBD and its analogue O-1918 and by micromolar concentrations of the CB₁ receptor antagonist SR141716A (Rimonabant®). This novel target has been referred to as: the abnormal-CBD receptor, the CB_x receptor or the endothelial anandamide receptor (for reviews, see Refs 10, 11). It is increasingly likely that more than one non-CB₁, non-CB₂ target might exist [9]. Clearly, the pharmacology surrounding the abnormal-CBD receptor might, in fact, encompass several novel hitherto unrecognized ‘cannabinoid’ receptors, one of which might be GPR55 (Box 1).

In the past two years, several peer-reviewed studies have been published in which the pharmacology of cannabinoids at GPR55 was probed in more depth 3, 13, 14, 15, 16, 17, 18. Despite this, GPR55 pharmacology remains a contentious issue 19, 20. The concept of ‘agonist functional selectivity’ is particularly pertinent to the debate surrounding GPR55 (for reviews, see Refs 21, 22). Canonical pharmacology implied that ligands interact with GPCRs in such way that, irrespective of the ligand structure or the measured functional response, this union culminates in the same endpoint. However, the scenario seems to be less simplistic and recent data demonstrate that the downstream signalling of a given GPCR might vary in both a ligand- and a system-dependant manner. Clearly, this might underlie the contradictory nature of the pharmacological findings surrounding GPR55 (Table 1).

Here, the recent literature is reviewed in an effort to glean the key controversies and questions surrounding the interaction of the cannabinoids with this orphan receptor. I will consider the GPR55 pharmacology of various classes of compound including the endocannabinoids, the bioactive lipid l-α-lysophosphatidylinsoitol (LPI), the phytocannabinoids and synthetic cannabinoid ligands. The characteristics of GPR55 downstream signalling and the agonist-specific nature of this signalling is reviewed. Furthermore, the debate surrounding the classification of the GPR55 as a ‘cannabinoid receptor’ will also be explored.