Functional Selectivity at CB1 Cannabinoid Receptors

RS Priestley1, SA Nickolls2, SPH Alexander1, DA Kendall1. 1University of Nottingham, Nottingham, UK, 2Pfizer Neusentis, Cambridge, UK

Cannabinoid type 1 (CB1) receptors have great potential as a therapeutic target for the treatment of pain, anxiety, obesity and addition1. They are recognised by a variety of structurally distinct ligands, and mediate their effects via activation of multiple signalling pathways2. There is increasing evidence that CB1 receptors, along with other GPCRs, mediate ligand-selective patterns of intracellular signalling, a process referred to as functional selectivity or biased agonism. The aim of the present study, therefore, was to examine functional responses to a variety of cannabinoid agonists acting at the CB1 receptor, in a number of in vitro models, in an effort to identify any agonist bias.

Chinese hamster ovary (CHO) and human embryonic kidney (HEK 293) cells stably transfected with the human recombinant CB1 receptor and murine Neuro 2a (N2a) cells without transfection, were exposed to a number of cannabinoid agonists, including the endogenous agonist anandamide, the phytocannabinoid Δ9-THC, and several synthetic ligands. cAMP accumulation and extracellular signal-regulated kinase (ERK) phosphorylation were measured using In-Cell Western and immuno-based assays3. Using concentration/response data, agonist bias was analysed using bias plots (equimolar comparison), and by comparison of intrinsic relative activities for all agonists, calculated relative to the high potency agonist HU-210 (equiactive comparison)4.

All three cell lines showed CB1- and Gi/o protein-dependent activation of ERK and inhibition of forskolin-stimulated cAMP accumulation: however, the magnitude of these responses differed between the three cell lines, with the responses in the Neuro 2a cells being markedly smaller than in the recombinant cell lines (e.g. pERK, HU-210 Rmax CHO - 172±18, HEK - 372±49, N2a - 113±17 % above basal response; cAMP, HU-210 Rmax CHO - 47±4. HEK - 71±5%, N2a - 10±1% % inhibition FSK response). The synthetic agonists WIN 55,212-2 and ACEA both exhibited bias towards ERK activation, in comparison to inhibition of cAMP accumulation, in both the CHO (10.4 and 9.8-fold, respectively; p<0.05; t-test) and HEK (5.3 and 13.4-fold respectively; p<0.05; t-test) cell lines. Furthermore, the endogenous ligand anandamide exhibited the same bias in CHO cells only (9.1-fold; p<0.05; t-test), while the phytocannabinoid Δ9-THC exhibited the same bias in HEK cells only (30.3-fold; p<0.05; t-test). While concentration/response curves were obtained in N2a cells, quantification of any bias was complicated by the relatively small response magnitude and increased data variance. Adding further complexity to these findings, Gi/o-independent ERK activation and enhancement of cAMP accumulation were observed in HEK cells treated with PTX (500 ng/ml), but not in CHO or N2a cells, in which WIN 55212-2 exhibited modest bias towards Gi/o-independent ERK activation, in comparison with Gi/o-dependent ERK signalling (2.3-fold p<0.05; t-test).

In summary, CB1 receptors do indeed mediate functionally selective signalling, raising the possibility of designing therapeutics with preferred signalling specificity, e.g. drugs biased towards ERK, providing enhanced coupling to cell proliferation/differentiation and plasticity. Our comparison of results from various in vitro models demonstrates the key role which the cellular environment plays in facilitating agonist bias and highlights the importance of selecting appropriate model cell lines.

References

1 – Mackie, (2006), Annu. Rev. Pharmacool. Toxicol. 46:101-122

2 – Alexander & Kendall, (2007), Br. J. Pharmacol. 152(5):602-623

3 – Chen et al., (2005), Anal. Chem. 338:136-142

4 – Kenakin & Christopoulos, (2011) Nat. Rev. Drug. Discov. 12:205-216