The orphan G protein-coupled receptors, GPR55 and GPR18 as vascular targets for endocannabinoids and related lipids

C. T. Choy, E. E. McNaughton, C. E. Sudlow, V. Ho. St George's University of London, London, UNITED KINGDOM

Introduction: Previous studies have suggested that the endocannabinoid N-arachidonoyl ethanolamine (AEA) induces arterial relaxation through a novel target distinct from the cannabinoid CB1 and CB2 receptors1. In some assays, GPR55 and GPR18 receptors, both G protein-coupled, are activated by AEA but their role in vascular tone regulation remains speculative.

Method: Male wild-type (WT; C57BL/6J) and GPR55 knockout (KO) mice were killed by cervical dislocation and mesenteric arteries were isolated for isometric tension recording. Relaxant responses are shown as mean±sem (n=4-8) and analysed by two-way/one-way analysis of variance followed by post-hoc tests, or Student t-test, where appropriate. Quantitative RT-PCR (n=3) was used to probe receptor expression in arteries.

Results: The endocannabinoids AEA and 2-arachididonoyl glycerol (2-AG) induced relaxation (AEA: pEC50=5.8±0.2, relaxation at 30µM=66±5%; 2-AG: pEC50=5.5±0.2, relaxation at 30µM=91±2%). Relaxation to AEA but not 2-AG were reduced by GPR55 KO, or the GPR55 antagonists O-1918 (3µM) and ML191 (10µM) (20-30% less relaxation at 30µM; P<0.05). Transient Receptor Potential Vanilloid type 1 receptor desensitisation by capsaicin (10µM) had no effect on relaxation to both endocannabinoids in KO mice (+capsaicin; AEA: pEC50=5.3±0.2, relaxation at 30µM=49±2%; 2-AG: pEC50=5.6±0.3, relaxation at 30µM=80±5%). The GPR55 agonist L-lysophosphotidylinositol (LPI) also induced relaxation that was inhibited by ML191 or GPR55 KO (at 30µM, WT: 66±7%; +ML191: 12±15%, P<0.01; KO: 19±13%, P<0.05) but this response was sensitive to the contractile agents used (data not shown). Interestingly, prior exposure of vessels to LPI (up to 30µM) attenuated AEA-induced relaxation, possibly due to GPR55 desensitisation (data not shown). Relaxation to AEA and LPI were also reduced by blockage of large conductance Ca2+-activated K+ channels (+ 50nM iberiotoxin; AEA: pEC50=5.3±0.2, relaxation at 30µM=54±12%, P<0.01; 30µM LPI: 5±8%, P<0.01). By comparison, the putative GPR18 agonist, N-arachidonoyl glycine (NAGly) evoked no relaxation up to 30µM, whereas abnormal-cannabidiol, a GPR55/GPR18 agonist (WT: pEC50=4.9±0.1, relaxation at 30µM=78±5%; KO: pEC50=5.2±0.1, relaxation at 30µM=70±9%) and cannabidiol, a GPR55 antagonist and GPR18 partial agonist (at 10µM, WT: 89±1%; KO: 90±4%) induced comparable relaxation in WT and KO mice. Quantitative RT-PCR analysis demonstrated mRNA expression of GPR55 and GPR18 in WT arteries (mRNA for GPR55 but not GPR18 was absent in KO arteries; data not shown).

Conclusion: AEA and LPI evoke GPR55-mediated relaxation in mouse mesenteric arteries. Some GPR18 agonists are vasorelaxants but the role of GPR18 in endocannabinoid responses require further investigation.(1) Bondarenko AI (2014). Endothelial atypical cannabinoid receptor: do we have enough evidence? Br J Pharmacol 171: 5573-5588