Investigating ligand-directed signalling at the cannabinoid receptor 2

The cannabinoid receptor 2 (CB2) is a promising therapeutic target due to its potential to manage various types of conditions including neuroinflammatory and neurodegenerative diseases without eliciting psychotropic effects that are associated with CB1 receptor activation. Receptors have been identified to be able to adopt multiple conformational states and to signal through multiple, apparently independent pathways. It has been proposed that particular ligands might only stabilize a specific conformational state which may elicit a particular response, without leading to the activation of all potential signalling pathways that an alternative ligand might evoke. This distinct activation of signalling is referred to as “functional selectivity” also known as biased agonism. A potential exploitation of this phenomenon is that agonists may be identified which promote beneficial effects of receptor activation at the expense of the undesirable effects, suggesting new strategies for the design of more effective drugs targeting particular G protein-coupled receptors, such as the CB2 receptor. The aim of this project was to investigate functional selectivity mediated by the CB2 cannabinoid receptor in vitro. Different cannabinoid ligands from various chemical structure classes were used in this study. These include; the classical cannabinoids, both natural and synthetic (Δ9-THC, cannabinol, and HU210, JWH133), the nonclassical (CP55940), the aminoalkylindoles (JWH015, WIN55212-2), the eicosanoids (2-AG, AEA) and the peroxisome proliferator-activated receptor α (PPAR α) agonist prodrug, fenofibrate. Chinese hamster ovary (CHO) cell line stably expressing human CB2 receptors was used as a model to characterize these ligands measuring binding affinity, [35S]-GTPγS stimulation, intracellular calcium elevation, ERK (extracellular signal-regulated kinase), PI3K (phosphatidylinositol 3-kinase)/Akt (Protein Kinase B), mTOR (mechanistic target of rapamycin) phosphorylation and cAMP accumulation. These were conducted using the techniques of radioligand binding, calcium mobilization using a FlexStation scanning fluorimeter, immunofluorescence (In-Cell Western) and [3H]-adenine pre-labelling and dual column chromatography. In this study, we demonstrated that there is a distinction in the rank order of potencies among cannabinoid ligands depending on the signalling cascade under study. Data derived from the concentration-effect curves were fitted to the Black-Leff operational model of agonism to provide log(τ/KA) ratios, normalized transduction coefficients Δlog(τ/KA) ratios were determined for test compounds in each assay by comparison with the endogenous ligand 2-AG as the reference agonist. All cannabinoid ligands showed a preference away from calcium or Akt pathways when compared to other signalling pathways. Analysis of bias showed no significant bias for any agonist between ERK and calcium, cAMP or Akt, Akt and calcium, cAMP and GTPγS or mTOR. However, all cannabinoid agonists exhibited bias towards mTOR over ERK or Akt. CP55940 was the only ligand that showed significant bias towards [35S]-GTPγS binding over calcium, whereas JWH133 was the only ligand that showed significant bias towards cAMP over Akt. The aminoalkylindoles JWH015 and WIN55212-2 were biased towards mTOR over [35S]-GTPγS, whereas CP55940 and the classical cannabinoids JWH133 and HU210 were biased towards [35S]-GTPγS over ERK or Akt. The next aim was to examine the functional expression of the CB2 receptor in breast cancer cell lines, including MCF-7, MDA-MB-231, MDM-MB-468 and SKBR3 cells, with the intention of examining agonist bias in CB2-positive cells. In all examined breast cancer cell lines, however, cannabinoid treatment did not affect [35S]-GTPγS binding, phosphorylated levels of ERK1/2, calcium mobilization or cell viability, demonstrating the absence of functional cannabinoid receptors in these breast cancer cell lines. In summary, these findings provided evidence of biased agonism at CB2 receptors in recombinant expression in CHO-hCB2 cells. These results suggest the possibility of new strategies for the design of biased ligands targeting CB2 receptors with a modified signalling response which might lead to an improvement in their efficacy and safety