Quantitative analysis of D1 dopamine receptor interaction with β -arrestin1 and 2 using bimolecular fluorescence complementation (BiFC)

RE Satchell1, ND Holliday2, RAJ Challiss1. 1Department of Cell Physiology and Pharmacology, University of Leicester, Leicester LE1 9HN, UK, 2School of Biomedical Sciences, University of Nottingham Medical School, Queen's Medical Centre, Nottingham NG7 2UH, UK

β-arrestins are important molecular adaptors, which regulate signalling from GPCRs. D1 dopamine receptor (D1R) stimulation has been demonstrated to promote translocation of both β-arrestin1 and 2 to the plasma membrane (Kim et al., 2004). Bimolecular fluorescence complementation (BiFC) is a technique for detecting protein-protein interactions, which has previously been used to investigate recruitment of β-arrestins to a number of different G protein-coupled receptor (GPCR) subtypes (Kilpatrick et al., 2010; Watson et al., 2012). Here we have used BiFC to quantify agonist-stimulated associations between D1R and β-arrestin1 and 2. D1R-stimulated signalling was investigated by measuring cAMP responses. Concentration-response parameters for D1R-β-arrestin BiFC and cAMP accumulation were compared for a range of dopaminergic agonists.

HEK293T cell-lines stably expressing either β-arrestin1 or 2 tagged with venus-YFP N-fragment (2-173) were additionally transfected with FLAG-tagged D1R-venus-YFP C-fragment (155-238) cDNA. Confluent cells on Greiner 96-well plates were incubated for 30 min at 37oC in HBSS, before agonist additions for 60 min. Cells were fixed with 3% PFA, and nuclei were stained with H33342. Images were acquired automatically using an IX Ultra plate-reader (Molecular Devices). Average intensity of 3-18 μm diameter vesicles was quantified by granularity analysis (Kilpatrick et al. 2010). Concentration-response data were normalized to plate minima and maxima (vehicle/dopamine (100 μM)) and log(agonist) vs. response curves were fitted to pooled data by non-linear regression using GraphPad Prism, with Hill slope constraints. To measure cAMP accumulation in D1R-β-arrestin2 cells, confluent cells on Costar 24-well plates were incubated for 20 min at 37oC in KHB containing IBMX (500 µM), and then agonist added for 10 min. cAMP concentration was determined as previously described (Brown et al., 1971). Pooled data were fitted with log(agonist) vs. response curves (GraphPad Prism).

A strong correlation between pEC50 values for BiFC YFP fluorescence in D1R-β-arrestin1 versus D1R-β-arrestin2 cells was found (r2 = 0.86; p = 0.001). However, a comparison of pEC50 values for BiFC YFP fluorescence and cAMP accumulation in D1R-β-arrestin2 cells produced a weak correlation (r2 = 0.36; p = 0.118). In general, agonist potency was greater for cAMP responses, with the exception of SKF 83822, where equipotency was observed, and SKF 83959, which exhibited a 2.9-fold greater potency for BiFC YFP fluorescence.

The current findings provide insights into potential ligand bias of some D1R agonists with respect to β-arrestin- and cAMP-dependent pathways.

We thank the Medical Research Council for financial support.

Brown, B.L. et al. (1971). Biochem J 121, 561-562.

Kim, O.-J. et al. (2004). J Biol Chem 279, 7999-8010.

Kilpatrick, L.E. et al. (2010). Br J Pharmacol 160, 892-906.

Watson, S.J. et al. (2012). Mol Pharmacol 81, 631-642.

Table 1. Agonist efficacy (Emax) and potency (pEC50) values for D1R-stimulated β-arrestin1 and 2 BiFC and cAMP accumulation. Data shown are mean values (±S.E.M.) from 3-4 independent experiments performed in duplicate, where the response to dopamine (100 µM) is defined as 100%.