Interaction of the µ-Opioid Receptor with G-proteins and GRK2

Introduction: The initial step in the homologous desensitization of G-protein-coupled receptors is their phosphorylation by one of the G-protein-coupled receptor kinases (GRKs). We demonstrate here measurement of the interaction of GRK2, a ubiquitously expressed GRK, with the µ-opioid receptor (µOR) by FRET and its dependence on agonist efficacy.

Methods: FRET measurements were performed between the µOR tagged with YFP and Gβ, Gγ or GRK2 labelled with Cerulean or mTurquoise as well as between YFP-labelled Gαi1 subunit and Cerulean-labelled Gβ subunit. All measurements were performed in transiently transfected HEK293 cells which were superfused with 10 µM [D-Ala2, N-MePhe4, Gly-ol]-enkephalin (DAMGO) or 30 µM morphine.

Results: HEK293T cells transfected with YFP-tagged µOR, mTurquoise-tagged GRK2 and non-fluorescent hete-rotrimeric Gi,proteins showed a robust increase in FRET upon superfusion with 10 µM DAMGO which was reversible upon agonist washout. The partial agonist morphine (30 µM) also caused an increase in FRET but the amplitude of the FRET signal was only 20.0% (± 3,4%) of that of the prior DAMGO stimu-lation, compared to 84,5% (±3,4%) for DAMGO. GRK2 binds G-protein βγ (Gβγ) subunits, and therefore we aimed to find out how cotransfection of GRK2 affected the interaction of Gβγ with the µOR. However, we could not measure any DAMGO-induced FRET changes between YFP-tagged µOR and Cerulean-tagged Gβ1 or CFP-tagged Gγ2 in the presence of non-fluorescent Gαi1 and Gγ2 or Gβ1 respectively. This was unexpected because we had previously successfully determined interactions between Gβγ and the α2A-adrenergic receptor or the M3 muscarinic acetylcholine receptor. This lack of FRET was not due to an inability of the tagged Gβγ to interact with the µOR as we could measure DAMGO-induced FRET changes between YFP-tagged Gαi1 and Gβ1γ2 (Gβ tagged with Cerulean) in the presence of non-fluorescent µOR. Moreover, we could pick up FRET between the µOR and Gβγ in the presence of non-fluorescent GRK2. Comparison of the on- and off-kinetics of the µOR-GRK2 interaction with that of the µOR-Gβγ interaction in the presence of GRK2 revealed similar time constants both for the on- and off-kinetics (GRK2: kon 0.16 s-1; koff 0.087 s-1; Gβγ in the presence of GRK2: kon 0.23 s-1, koff 0.069 s-1).

Conclusion: The results suggest that, in the absence of GRK2, the orientation of the two fluorophores on the µOR and Gβγ may be unfavorable to produce an appreciable FRET signal. In the presence of GRK2, however, Gβγ changes its position relative to the µOR in a way that allows the interaction of the GRK2-Gβγ complex with the µOR to be detected by FRET. Similarly, we measured FRET between Gβγ and the α2A-adrenergic receptor or the M3 muscarinic acetylcholine receptor in the absence and presence of GRK2 and compared the kinetics with the kinetics of GRK2 binding and unbinding to these receptors. In both cases, we found that GRK2 and Gβγ in the presence of GRK2 associate and dissociate from these receptors with comparable kinetics. Our results suggest that ligand efficacy for µORs is already apparent on the level of receptor-GRK interaction.