Studies on the biased signalling of some novel delta opioid receptor agonists

The delta opioid receptor (DOR) is a G protein-coupled receptor (GPCR) which is important in the regulation of neuronal function, predominantly via coupling to heterotrimeric Gi/0 proteins. The receptor has been shown to be a potential target for the treatment of chronic pain and affective disorders. Although a large number of opioid agonists exist, their properties vary widely, at least partly due to their differential coupling to post-receptor signalling systems, a phenomenon referred to as ligand-biased signalling or functional selectivity. The aim of the current project was to examine the signalling properties of a set of established and novel DOR agonists in an attempt to identify compounds that have biased signalling profiles. It was hypothesized that DOR agonists with partial efficacy regarding β-arrestin recruitment would be less liable to induce receptor internalization and desensitization of G protein-mediated signalling than full agonists. Chinese hamster ovary (CHO) cells, stably transfected with GFP-tagged human (h)DOR and CHO-K1 and U2OS cells over-expressing hDOR were exposed to a number of novel selective DOR agonists compared with the commercially available agonists, SNC80, ADL5859 and DADLE. The compounds’ potencies and efficacies were measured in four different assay systems; 1. Inhibition of forskolin-stimulated cyclic AMP accumulation, 2. Extracellular signal-regulated kinase (ERK1/2) phosphorylation using an immunocytochemical In-cell Western (I-CW) assay, 3. β-arrestin 2 recruitment and 4. hDOR internalization. The latter two assays employed DiscoverX Enzyme Fragment Complementation technology. An attempt to develop a secreted placental alkaline phosphatase (SPAP) reporter gene assay to measure DOR-mediated cyclic AMP inhibition was not successful. All of the ligands were nearly full agonists in relation to cyclic AMP inhibition although some were less efficacious than the standard SNC80 regarding ERK1/2 activation. Their absolute potencies and rank orders of potency in inhibiting cyclic AMP and activating ERK1/2 were quite different, although both signalling systems were apparently Gi/o- protein mediated. In contrast, the agonists exhibited a full range of efficacies and potencies in both β-arrestin 2 recruitment and hDOR internalization assays and there was a significant correlation between the maximum efficacies of the compounds in the two assays. A potential relationship between β-arrestin 2 recruitment/ hDOR internalization and desensitization of agonist-induced cyclic AMP accumulation was explored. Responses to the highly arrestin-recruiting agonists SNC80 and DADLE desensitized fully after extended exposure, whereas the novel partial agonists PN6047 and OPD00003 resisted desensitization. Bias factors were calculated for the agonist set and both PN6047 and OPD00003 were found to be significantly biased towards G protein-mediated cyclic AMP inhibition. In conclusion, this study reports, for the first time, a detailed characterization of signalling bias for a set of selective DOR agonists in cells over-expressing human DORs. The findings suggest that it is potentially possible to predict wanted and unwanted properties of agonists by determining post-receptor signalling profiles in vitro which will facilitate the discovery and development of novel therapeutics based on the DOR