Pharmacology of GPR55 in yeast and identification of GSK494581A as a mixed-activity glycine transporter subtype 1 inhibitor and GPR55 agonist

GPR55 is a G protein-coupled receptor activated by L-α-lysophosphatidylinositol and suggested to have roles in pain signaling, bone morphogenesis, and possibly in vascular endothelial cells. It has affinity for certain cannabinoids (molecules that interact with the cannabinoid CB(1) and CB(2) receptors), but investigation of its functional role in cell-based systems and in tissue has been limited by a lack of selective pharmacological tools. Here, we present our characterization of GPR55 in the yeast Saccharomyces cerevisiae and in human embryonic kidney (HEK293) cells. We describe GSK494581A (1-{2-fluoro-4-[1-(methyloxy)ethyl]phenyl}-4-{[4'-fluoro-4-(methylsulfonyl)-2-biphenylyl]carbonyl}piperazine), a selective small-molecule ligand of GPR55 identified through diversity screening. GSK494581A is one of a series of benzoylpiperazines originally identified and patented as inhibitors of the glycine transporter subtype 1 (GlyT1). The structure-activity relationship between GPR55 and GlyT1 is divergent across this series. The most GPR55-selective example is GSK575594A (3-fluoro-4-(4-{[4'-fluoro-4-(methylsulfonyl)-2-biphenylyl]carbonyl}-1-piperazinyl)aniline), which is approximately 60-fold selective for GPR55 (pEC(50) = 6.8) over GlyT1 (pIC(50) = 5.0). Several exemplars with activity at GPR55 and GlyT1 have been profiled at a broad range of other molecular targets and are inactive at cannabinoid receptors and all other targets tested. The benzoylpiperazine agonists activate human GPR55 but not rodent GPR55, suggesting that the relatively low level of sequence identity between these orthologs (75%) translates to important functional differences in the ligand-binding site.

The putative cannabinoid receptor GPR55 plays a role in mechanical hyperalgesia associated with inflammatory and neuropathic pain

It has been postulated that the G protein-coupled receptor, GPR55, is a third cannabinoid receptor. Given that the ligands at the CB(1) and CB(2) receptors are effective analgesic and anti-inflammatory agents, the role of GPR55 in hyperalgesia associated with inflammatory and neuropathic pain has been investigated. As there are no well-validated GPR55 tool compounds, a GPR55 knockout (GPR55(-/-)) mouse line was generated and fully backcrossed onto the C57BL/6 strain. General phenotypic analysis of GPR55(-/-) mice revealed no obvious primary differences, compared with wild-type (GPR55(+/+)) littermates. GPR55(-/-) mice were then tested in the models of adjuvant-induced inflammation and partial nerve ligation. Following intraplantar administration of Freund's complete adjuvant (FCA), inflammatory mechanical hyperalgesia was completely absent in GPR55(-/-) mice up to 14 days post-injection. Cytokine profiling experiments showed that at 14 days post-FCA injection there were increased levels of IL-4, IL-10, IFN gamma and GM-CSF in paws from the FCA-injected GPR55(-/-) mice when compared with the FCA-injected GPR55(+/+) mice. This suggests that GPR55 signalling can influence the regulation of certain cytokines and this may contribute to the lack of inflammatory mechanical hyperalgesia in the GPR55(-/-) mice. In the model of neuropathic hypersensitivity, GPR55(-/-) mice also failed to develop mechanical hyperalgesia up to 28 days post-ligation. These data clearly suggest that the manipulation of GPR55 may have therapeutic potential in the treatment of both inflammatory and neuropathic pain.

The novel endocannabinoid receptor GPR55 is activated by atypical cannabinoids but does not mediate their vasodilator effects

BACKGROUND AND PURPOSE

Atypical cannabinoids are thought to cause vasodilatation through an as-yet unidentified 'CBx' receptor. Recent reports suggest GPR55 is an atypical cannabinoid receptor, making it a candidate for the vasodilator 'CBx' receptor. The purpose of the present study was to test the hypothesis that human recombinant GPR55 is activated by atypical cannabinoids and mediates vasodilator responses to these agents.

EXPERIMENTAL APPROACH

Human recombinant GPR55 was expressed in HEK293T cells and specific GTPgammaS activity was monitored as an index of receptor activation. In GPR55-deficient and wild-type littermate control mice, in vivo blood pressure measurement and isolated resistance artery myography were used to determine GPR55 dependence of atypical cannabinoid-induced haemodynamic and vasodilator responses.

KEY RESULTS

Atypical cannabinoids O-1602 and abnormal cannabidiol both stimulated GPR55-dependent GTPgammaS activity (EC50 approximately 2 nM), whereas the CB1 and CB2-selective agonist WIN 55,212-2 showed no effect in GPR55-expressing HEK293T cell membranes. Baseline mean arterial pressure and heart rate were not different between WT and GPR55 KO mice. The blood pressure-lowering response to abnormal cannabidiol was not different between WT and KO mice (WT 20+/-2%, KO 26+/-5% change from baseline), nor was the vasodilator response to abnormal cannabidiol in isolated mesenteric arteries (IC50 approximately 3 micro M for WT and KO). The abnormal cannabidiol vasodilator response was antagonized equivalently by O-1918 in both strains.

CONCLUSIONS

These results demonstrate that while GPR55 is activated by atypical cannabinoids, it does not appear to mediate the vasodilator effects of these agents.

Changes in dorsal root ganglion CGRP expression in a chronic inflammatory model of the rat knee joint: differential modulation by rofecoxib and paracetamol

Neuropeptide-expressing small diameter sensory neurones are thought to be vital in generating inflammatory hyperalgesic responses. Within the dorsal root ganglion (DRG), both the levels of the neuropeptide calcitonin gene-related peptide (CGRP) and the numbers of CGRP-immunoreactive (CGRP-IR) DRG neurones have been shown to increase in a number of acute adjuvant-induced inflammatory pain models. The aim of this study was to look specifically at changes in numbers of CGRP-IR DRG neurones in a chronic model of inflammatory joint pain following complete Freund's adjuvant (CFA) injection into the rat knee. In this model, there were significant increases in the number of ipsilateral CGRP-IR small DRG neurones at days 1, 16 and 35 following intra-articular CFA, compared to saline-injected sham animals. This correlated with the behavioural readouts of hypersensitivity and knee joint inflammation at the same time points. There was also a significant increase in the number of ipsilateral CGRP-IR medium DRG neurones and contralateral CGRP-IR small DRG neurones at day 1. Following dosing of CFA-injected rats with rofecoxib (Vioxx) or paracetamol, there was a significant decrease in the number of ipsilateral CGRP-IR small and medium DRG neurones in rofecoxib- but not paracetamol-treated rats. These data also correlated with behavioural readouts where hypersensitivity and knee joint inflammation were significantly reduced by rofecoxib but not paracetamol treatment. In conclusion, these data show that changes in ipsilateral CGRP expression within small DRG neurones are consistent with behavioural readouts in both time course, rofecoxib and paracetamol studies in this model of chronic inflammatory pain.

Mitogen and stress response kinase-1 (MSK1) mediates excitotoxic induced death of hippocampal neurones

Activation of the mitogen-activated protein kinase (MAPK/ERK) signal transduction pathway may mediate excitotoxic neuronal cell death in vitro and during ischemic brain injury in vivo. However, little is known, of the upstream regulation or downstream consequences of ERK activation under these conditions. Magnesium removal has been described to induce hyperexcitability and degeneration in cultured hippocampal neurones. Here, we show that neurotoxicity evoked by Mg2+ removal in primary hippocampal neurones stimulates ERK, but not p38, phosphorylation. Removal of Mg2+ also resulted in induction of the MAPK/ERK substrate mitogen- and stress-response kinase 1 (MSK1) and induced phosphorylation of the MSK1 substrate, the transcription factor cAMP response element binding protein (CREB). Neuronal death and phosphorylation of components in this cascade were inhibited by the Raf inhibitor SB-386023, by the MEK inhibitor U0126, or by the MSK1 inhibitors H89 and Ro318220. Importantly, this form of cell death was inhibited in hippocampal neurones cultured from MSK1-/- mice and inhibitors of Raf or MEK had no additive neuroprotective effect. Together, these data indicate that MSK1 is a physiological kinase for CREB and that this activity is an essential component of activity-dependent neuronal cell death.

Role of group III metabotropic glutamate receptors in excitotoxin-induced cerebellar granule cell death

The neuronal effects of the metabotropic glutamate receptor agonist (1S,3R)-aminocyclopentane-1,3-dicarboxylic acid have been studied in cultured rat cerebellar granule cells, and compared with those of the endogenous excitotoxin glutamate, and the dietary excitotoxin beta-N-methylamino-L-alanine. Glutamate, beta-N-methylamino-L-alanine, and (1S,3R)-aminocyclopentane-1,3-dicarboxylic acid all caused concentration-dependent cerebellar granule cell death over a 24-h exposure period. The metabotropic antagonist (RS)-alpha-methyl-4-carboxyphenylglycine reduced glutamate-, beta-N-methylamino-L-alanine-, and (1S,3R)-aminocyclopentane-1,3-dicarboxylic acid-induced death by 50, 37, and 90%, respectively. (1S,3R)-Aminocyclopentane-1,3-dicarboxylic acid-induced death was unaffected by the group I antagonist (RS)-1-aminoindan-1,5-dicarboxylic acid, increased by the group II antagonist ethylglutamic acid, and markedly decreased by the group III antagonist (RS)-alpha-methylserine-O-phosphate. Neither (1S,3R)-aminocyclopentane-1,3-dicarboxylic acid nor the group I agonist (RS)-3,5-dihydroxyphenylglycine caused an increase in intracellular free calcium levels. The group III agonist L-(+)-2-amino-4-phosphonobutyric acid also induced concentration-dependent cerebellar granule cell death, and so it was suggested that the group III metabotropic glutamate receptors were responsible for (1S,3R)-aminocyclopentane-1,3-dicarboxylic acid-induced death. Blocking these receptors with (RS)-alpha-methylserine-O-phosphate also prevented a proportion of glutamate- and beta-N-methylamino-L-alanine-induced death.

The dietary excitotoxins beta-N-methylamino-L-alanine and beta-N-oxalylamino-L-alanine induce necrotic- and apoptotic-like death of rat cerebellar granule cells

The neurotoxic properties of the dietary excitotoxins beta-N-methylamino-L-alanine and beta-N-oxalylamino-L-alanine have been studied in rat cerebellar granule cells and compared with those of glutamate. Glutamate caused dose-dependent death of cerebellar granule cells after a 30-min exposure when viability was assessed 24 h later. Beta-N-methylamino-L-alanine and beta-N-oxalylamino-L-alanine, however, were toxic only after 24 or 48 h of exposure. The neurotoxic effects of beta-N-methylamino-L-alanine were blocked by D(-)-2-amino-5-phosphonopentanoic acid, and those of beta-N-oxalylamino-L-alanine were blocked by kynurenic acid, which demonstrated that these excitotoxins caused cerebellar granule cell death through the activation of glutamate receptors. The features of this death were examined morphologically (fluorescent dyes, electron microscopy) and biochemically (conventional agarose gel electrophoresis, effect of aurintricarboxylic acid). Characteristics of apoptosis were identified by transferring cerebellar granule cells from a high K+ (30 mM)- to a low K+ (10 mM)-containing medium. In cerebellar granule cells exposed to beta-N-methylamino-L-alanine or beta-N-oxalylamino-L-alanine (3 mM), hallmarks of necrotic- and apoptotic-like death were observed at various time points over a 72-h period. Therefore, in cerebellar granule cells, beta-N-methylamino-L-alanine and beta-N-oxalylamino-L-alanine induce death over 12-72 h of exposure via a mechanism that involves both necrotic- and apoptotic-like cell death.