Elsevier

Cellular Signalling

Volume 9, Issue 8, December 1997, Pages 551-573
Cellular Signalling

Signalling Via the G Protein-Activated K+ Channels

https://doi.org/10.1016/S0898-6568(97)00095-8Get rights and content

Abstract

The inwardly rectifying K+ channels of the GIRK (Kir3) family, members of the superfamily of inwardly rectifying K+ channels (Kir), are important physiological tools to regulate excitability in heart and brain by neurotransmitters, and the only ion channels conclusively shown to be activated by a direct interaction with heterotrimeric G protein subunits. During the last decade, especially since their cloning in 1993, remarkable progress has been made in understanding the structure, mechanisms of gating, activation by G proteins, and modulation of these channels. However, much of the molecular details of structure and of gating by G protein subunits and other factors, mechanisms of modulation and desensitization, and determinants of specificity of coupling to G proteins, remain unknown. This review summarizes both the recent advances and the unresolved questions now on the agenda in GIRK studies.

Introduction

The G protein-activated K+ channels (GIRK, or Kir3) present one out of the six groups that constitute the family of inwardly rectifying K+ channels, Kir. The six groups differ in the extent of homology and in gating properties [reviews: 38, 45]. As implied by their name, the GIRKs are activated by G proteins, which in itself is not remarkable: many ion channels are modulated by G proteins, most via second messenger cascades that include the production of water-soluble second messengers, and some via membrane-delimited mechanisms that do not involve cytosolic messengers. In the latter case, activation or inhibition of the channel by a direct interaction with a G protein subunit is often assumed [for reviews, see [12], [15], [37], [67], [68], [168], [195], [236], [237]]. However, only in two cases have such interactions been unequivocally demonstrated: some voltage dependent Ca2+ channels are inhibited by a direct interaction with the βγ subunits of G proteins, Gβγ 34, 63, 78, 249, whereas GIRKs are the only ion channels known to be activated by direct binding of Gβγ. In fact, the cardiac GIRK, termed also “muscarinic K+ channel” or K(ACh), was the first G protein-activated ion channel 14, 122, 123, 182 and the first Gβγ effector [141] discovered. It still remains one of the best Gβγ effectors studied; hence the popularity of this protein for exploring G protein-effector interactions.

GIRKs fulfill important physiological roles. In the heart, activation of K(ACh) by acetylcholine (ACh) via the muscarinic m2 receptor and a pertussis toxin (PTX)-sensitive G protein mediates the vagal (parasympathetic) negative chronotropic effect i.e., slowing of the heart rate 14, 122, 182. Adenosine and ATP via purinergic receptors, and calcitonin gene-related peptide act similarly 50, 123, 100. In the brain, inwardly rectifying K+ channels, presumably of the GIRK family, are activated via PTX-sensitive G proteins by a number of neurotransmitters: GABA, serotonin, opioids, dopamine, etc. The wide distribution of GIRK subunits in the brain (see below) and the potential for regulation of neuronal excitability by mediating many of the actions of major inhibitory neurotransmitters suggest important physiological roles for GIRKs in the central nervous system, although the specific functions in different brain areas and cell types remain to be elucidated 15, 68, 84, 168, 170, 236. RNAs of GIRK subunits are also found in pancreas, where their role is unknown [47].

Despite the intensive research on GIRKs during the last decade, much remains unknown. Unresolved problems include the molecular details of structure and of gating by Gβγ and other factors, mechanisms of modulation and desensitisation, and determinants of specificity of coupling to G proteins. One of the reasons is that the first representative of the group, GIRK1 (KGA, or Kir3.1), was cloned only in 1993 31, 118, shortly after the cloning of the first two representatives of the Kir family, ROMK1 (Kir1.1 [70]) and IRK1 (Kir2.1 [117]). The other members of the GIRK group (GIRK2-GIRK5) were cloned between 1994 and 1996. The cloning boosted the research and brought great progress. This review complements previous extensive reviews concerning G-protein activated K+ channels 10, 121, 129, 236 by presenting much of the information accumulated since the cloning of GIRK1 and especially the other subunits. The focus is on GIRKs, but they may not be the only inward rectifier K+ channels directly gated by G protein βγ subunits. At least one representative of the IRK (Kir2) class found mainly in the brain, Kir2.3 (but not Kir2.1 or Kir2.2), is inhibited by Gβγ, probably by a direct binding to the N-terminus [29]; and an 18 pS inward rectifier (single channel conductance typical for Kir2 family) is inhibited by GTPγS in a membrane-delimited fashion in rat brain oligodendro- cytes [97].

Section snippets

Primary Structure

The building blocks (subunits) of all Kirs share a common structural design as predicted from the primary structure 31, 70, 117, 118, characterised by two transmembrane domains M1 and M2, a pore region (P, or H5) similar to that of voltage-dependent channels, and cytoplasmic N and C termini (Fig. 1) 38, 45, 136. These predictions are now being confirmed by various methods with different representatives of the family; the conclusions are usually projected onto the other members' structure. This

Distribution and function

As noted above, ACh, CGRP and adenosine activate GIRK in atrial and pacemaker sinoatrial (SA) node cells. This is believed to be the main mechanism of slowing of the heart rate by these substances [see 121]. The K(ACh) channel not only mediates the inhibitory neurotransmitter or hormone responses but may play an important part in determining the resting potential. In the pacemaker cells of SA node, most or all of the basal K+ conductance is due to the K(ACh) 87, 93, 190. Only a minor portion

Activation by Agonists

The general scheme has been extensively reviewed 25, 121, 165, 236 and is now considered common knowledge. An agonist binds to a 7-helix receptor which, in turn, binds to a GDP-bound heterotrimeric G protein composed of α, β and γ subunits, enabling GDP-GTP exchange at the α subunit and leading to dissociation of GTP-bound α subunit (Gα-GTP) from the Gβγ dimer [52]. Both free (“activated”) Gα-GTP and Gβγ can activate or inhibit effectors [165]. Termination of activation is achieved by

Permeation, rectification, and voltage-dependent gating

The gating of GIRKs is very complex. In addition to their activation by agonists via Gβγ and by ATP and Na+, the GIRKs show strong inward rectification which depends on intracellular cations, membrane voltage, external K+ and K+ equilibrium potential (EK), and display voltage- and time-dependent activation upon hyperpolarisation. Understanding of these processes continues to give clues to many aspects of their structure and function.

Single Channel Kinetics

Analysis of single channel kinetics can provide outstanding insights into the molecular details of ion channel function [191]. Unfortunately, so far, single channel analysis of GIRK activity has been hampered by its high density in the cells used for most studies, i.e., atrial and pacemaker cells in the heart and hippocampal or basal ganglia neurons. The patches usually contain more than one channel, which precludes accurate measurement of closed and especially interburst times. It is expected

Specificity of g protein—girk coupling

A major open question in GIRK gating is the specificity of activation by G proteins. Since Gβγ is released from any heterotrimeric G protein when it is activated by the appropriate receptor, why do only receptors that are coupled to PTX-sensitive G proteins activate the channel under normal physiological conditions? Do all PTX-sensitive Gα subunits contribute equally? A few hints are available regarding the identity of the Gα that preferentially mediates GIRK activation in vivo, so far mainly

Desensitisation

Desensitisation is an important negative feedback mechanism for prevention of excessive stimulation. Desensitisation of K(ACh) cardiac currents was described already in 1980 [90]. Later it has been shown that, in the continuous presence of agonist, the IK(ACh) decay shows two phases: a fast one (a 20–40% current decay within 10–20 s) and a slow one, over several minutes 17, 100, 125, 250.

The fast desensitisation is heterologous between ACh, adenosine, and a novel activator of GIRK, phospholipid

Modulation

Since most of the researchers' attention was focused on G protein regulation of GIRKs, little is known about modulation of GIRK channels by phosphorylation, Ca2+ and other intracellular factors. Many of the modulatory effects are discussed in detail by Kurachi et al. [129].

It is known that the basal activity of cardiac GIRK is modulated by Na+ and ATP (see above); it is not certain that ATP and Na+ have the same bold effect on Gβγ-evoked activity. Another intracellular ion that may modulate

The weaver pathology

Weaver is an inherited neurological disorder in mouse characterised by degeneration of cerebellar granule cells and dopaminergic neurons in substantia nigra, which results in a severe motor deficiency. Surprisingly, a single site mutation in GIRK2 underlies this complex and specific phenotype [for review, see [64]]. The expression of the mutated subunit (GIRK2wv) in cerebellar granule cells, hippocampus and substantia nigra is disrupted over the first few weeks after birth. Importantly, the

Conclusions

G protein-activated inward rectifier K+ channels (GIRKs) are widely distributed in the brain where they mediate inhibitory actions of many neurotransmitters, and are highly concentrated in the atria where their activation by acetylcholine is the primary cause of slowing of the heart rate by vagal nerve stimulation. They are known to underlie at least one neurological disorder in the mouse. It is clear that the structure and gating mechanisms of these channels are very intricate. Normally the

Acknowledgements

The author is grateful to H. Lester, P. Kofuji, Y. Kurachi, D. Kim, C. Chavkin and G. Breitwieser for sharing preprints and unpublished results, and I. Lotan, W. Schreibmayer, E. Reuveny, S. Silverman and H. Lester for critical reading of the manuscript. During writing of this review, the author was supported by grants from Human Frontiers Science Foundation and USA-Israel Binational Science Foundation.

References (252)

  • G.K. Aghajanian et al.

    Brain Res.

    (1986)
  • J. Aiyar et al.

    J. Biol. Chem.

    (1996)
  • A. Aleksandrov et al.

    Biophys. J.

    (1996)
  • R. Andrade

    Eur. J. Pharmacol.

    (1991)
  • B.P. Bean

    Trends Pharmacol. Sci.

    (1992)
  • I. Ben-Efraim et al.

    Biophys. J.

    (1997)
  • W.A. Catterall

    Trends Neurosci.

    (1993)
  • K.W. Chan et al.

    J. Biol. Chem.

    (1997)
  • Y. Chen et al.

    J. Biol. Chem.

    (1994)
  • N.A. Cohen et al.

    Neuron

    (1996)
  • A.M. De Paoli et al.

    Mol. Cell Neurosci.

    (1994)
  • A.C. Dolphin

    Trends Neurosc.

    (1996)
  • C.A. Doupnik et al.

    Current Opin. Neurobiol.

    (1995)
  • C.A. Doupnik et al.

    Neuropharmacology

    (1996)
  • F. Duprat et al.

    Biochem. Biophys. Res. Commun.

    (1995)
  • B. Fakler et al.

    FEBS Lett.

    (1994)
  • J. Ferrer et al.

    J. Biol. Chem.

    (1995)
  • M. Fink et al.

    FEBS Lett.

    (1996)
  • A.P. Gadbut et al.

    J. Biol. Chem.

    (1994)
  • K.E. Hedin et al.

    Cell. Signal.

    (1993)
  • K.E. Hedin et al.

    Neuron

    (1996)
  • D.J. Henry et al.

    Neurosci. Lett.

    (1995)
  • E.J. Hess

    Neuron

    (1996)
  • J.B. Higgins et al.

    Cell. Signal.

    (1996)
  • B. Hille

    Neuron

    (1992)
  • C.-L. Huang et al.

    Neuron

    (1995)
  • A. Inanobe et al.

    Biochem. Biophys. Res. Commun.

    (1995)
  • C. Karschin et al.

    FEBS Lett.

    (1994)
  • A. Karschin et al.

    Neurosci. Lett.

    (1995)
  • R. Andrade et al.

    Science

    (1986)
  • R. Andrade et al.

    J. Physiol. (Lond.)

    (1987)
  • C.M. Armstrong

    J. Gen. Physiol.

    (1969)
  • C.M. Armstrong et al.

    J. Gen. Physiol.

    (1977)
  • S.B. Bausch et al.

    Recept. Chann.

    (1995)
  • G. Breitwieser

    J. Membr. Biol.

    (1996)
  • G.E. Breitwieser et al.

    Nature

    (1985)
  • G.E. Breitwieser et al.

    J. Gen. Physiol.

    (1988)
  • D.A. Brown

    Annu. Rev. Physiol.

    (1990)
  • M. Bunemann et al.

    J. Physiol. (Lond.)

    (1996)
  • E. Carmeliet et al.

    J. Physiol. (Lond.)

    (1986)
  • K.W. Chan et al.

    J. Gen. Physiol.

    (1996)
  • K.W. Chan et al.

    Proc. Natl. Acad. Sci. USA

    (1996)
  • J. Chen et al.

    Science

    (1995)
  • S. Ciani et al.

    J. Membr. Biol.

    (1978)
  • D.E. Clapham et al.

    Nature

    (1993)
  • E.A. Clark et al.

    Science

    (1995)
  • G.B. Cohen et al.

    Cell

    (1996)
  • N.A. Cohen et al.

    J. Biol. Chem.

    (1996)
  • N. Dascal et al.

    Proc. Natl. Acad. Sci. USA

    (1993)
  • N. Dascal et al.

    Proc. Natl. Acad. Sci. USA

    (1993)
  • Cited by (0)

    View full text