Expression of the regulator of G-protein signalling (RGS) proteins results in rapid deactivation of G-protein activated inwardly rectifying K⁺ (GIRK) channels, consistent with their catalytic function as a GTPase-activating protein on active Gα subunits and their role as negative regulators of other G-protein pathways. Unexpectedly RGS proteins also accelerate receptor-stimulated activation kinetics of GIRK channels, by an unknown mechanism. The effects of neuronal RGS8 were examined in HEK-293 cell lines, stably expressing GIRK channel subunits (Kir3.1+3.2A) plus one of the following GPCRs: A₁ adenosine, α_2A adrenergic, D₂ dopamine, M₄ muscarinic, or the GABA-B_1b/2 heterodimeric receptor.

Whole-cell K⁺ currents in symmetrical K⁺ (140 mM) were recorded at a membrane potential of -60 mV. Currents were activated by fast application and removal of agonists (delay time approx. 0.25s): GABA_B, 100 µM baclofen; M₄, 100 µM carbachol; D₂, 10 µM dopamine; A₁, 1 µM adenosine; α_2A, 3 µM noradrenaline. The kinetics of receptor-stimulated currents were characterised by measuring the lag and time to peak current amplitude (lag+TTP) for activation, and by fitting a single exponential time constant (t{special}_deact) to the deactivation phase. Data are given as means ± S.E.M. and were reciprocated prior to statistical analysis with Student’s unpaired t test.

The main findings were that transfection of RGS8 (tagged with the yellow fluorescent protein, RGS8-YFP) significantly accelerated current deactivation in each of the receptor cell lines (compared to control recordings) but a receptor-selective effect was seen on activation kinetics whereby activation of currents via the D₂, GABA_B and M₄ receptors was significantly accelerated whilst activation via A₁ and α_2A receptors was not affected by RGS8-YFP expression (GABA_B, lag+TTP(s): 2.6 ± 0.7, n = 13; + RGS8: 1.3 ± 0.4, n = 11, P < 0.001; A₁, lag+TTP: 1.1 ± 0.01, n = 14; +RGS8: 1.2 ± 0.01, n = 15, NS). The lag TTP data for all receptors studied is shown in Fig. 1. These effects were dependent on the RGS catalytic domain since an N-terminally deleted construct exhibited identical kinetic effects to RGS8-YFP. The receptor selectivity of RGS8-YFP was not due to preferential regulation of certain Gα-isoforms because, when signalling was constrained to transfected PTx-insensitive GoαA(CŒ{special}G) subunits in the GABA_B and A₁ cell lines (treated with PTx at 100 ng ml^-1 for >16 h), RGS8-YFP accelerated GABA_B-mediated activation (lag+TTP(s): +GoαA: 3.3 ± 0.6; +GoαA + RGS8: 1.2 ± 0.1, P < 0.001; n = 8) but not A₁-mediated activation of currents (lag+TTP: +GoαA: 2.2 ± 0.3; +GoαA + RGS8: 2.5 ± 0.3; n = 6, NS). Deactivation was significantly accelerated in both lines. Our data support current ideas of kinetic scaffolding and indicate that this occurs in a receptor-dependent fashion.

This work was funded by the Wellcome Trust and BHF.