Unifying Mechanism of Controlling Kir3 Channel Activity by G Proteins and Phosphoinositides

Diomedes E. Logothetis1; Rahul Mahajan; Scott K. Adney; Junghoon Ha; Takeharu Kawano; Xuan-Yu Meng2; Meng Cui    Department of Physiology and Biophysics, School of Medicine, Virginia Commonwealth University, Richmond, Virginia, USA
1 Corresponding author: email address: delogothetis@vcu.edu
2 Present Addresses: School for Radiological and Interdisciplinary Sciences (RAD-X), Collaborative Innovation Center of Radiation Medicine of Jiangsu Higher Education Institutions, and Jiangsu Provincial Key Laboratory of Radiation Medicine and Protection, Soochow University, Suzhou 215123, China.

Abstract

The question that started with the pioneering work of Otto Loewi in the 1920s, to identify how stimulation of the vagus nerve decreased heart rate, is approaching its 100th year anniversary. In the meantime, we have learned that the neurotransmitter acetylcholine acting through muscarinic M2 receptors activates cardiac potassium (Kir3) channels via the ß? subunits of G proteins, an important effect that contributes to slowing atrial pacemaker activity. Concurrent stimulation of M1 or M3 receptors hydrolyzes PIP2, a signaling phospholipid essential to maintaining Kir3 channel activity, thus causing desensitization of channel activity and protecting the heart from overinhibition of pacemaker activity. Four mammalian members of the Kir3 subfamily, expressed in heart, brain, endocrine organs, etc., are modulated by a plethora of stimuli to regulate cellular excitability. With the recent great advances in ion channel structural biology, three-dimensional structures of Kir3 channels with PIP2 and the Gß? subunits are now available. Mechanistic insights have emerged that explain how modulatory control of activity feeds into a core mechanism of channel-PIP2 interactions to regulate the conformation of channel gates. This complex but beautiful system continues to surprise us for almost 100 years with an apparent wisdom in its intricate design.

Keywords

Kir3 channels

GIRK channels

PIP2

Phosphatidylinositol bisphosphate

Gß?

G proteins

G-loop gate

Phosphorylation

Computational model

Crystal structure

1 Introduction

In this review, we will present our current understanding of the control of Kir3 channel activity by phosphoinositides (PIPs) aided by other molecules, such as G proteins, with a historical perspective of its development.

When the vagus nerve is stimulated, chemical transmission of neural signals occurs via release of acetylcholine (ACh) to slow heart rate (Loewi, 1921; Loewi & Navratil, 1926). An increase in K+ conductance, KACh, found in atrial myocytes and cardiac pacemaking cells, underlies the hyperpolarization effect of ACh that precedes the decrease in heart rate (Burgen & Terroux, 1953; Trautwein & Dudel, 1958). With the advent of the patch-clamp technique (Hamill, Marty, Neher, Sakmann, & Sigworth, 1981), KACh single-channel characteristics were determined, showing short open times (1-2 ms) and a conductance of ~ 40 pS (Sakmann, Noma, & Trautwein, 1983). A year later it was shown that the effect of ACh on KACh channels is mediated locally, in a membrane-delimited fashion, and not by diffusible second messengers (Soejima & Noma, 1984). In their cell-attached experiments, Soejima and Noma perfused the inside of the pipette with solutions containing ACh to demonstrate that KACh activation would only occur when ACh was perfused in the patch pipette and not in the bath (Fig. 1). The ability to unequivocally identify KACh has led to hundreds of publications revealing multiple ways of controlling the activity of these important channels directly or allosterically through interactions of the channel protein with other regulatory proteins, metal ions, ethanol, and membrane lipids. In this review, we will focus on how fundamental protein-lipid interactions lie at the core of controlling the ion channel gates and how other modulatory signals feed onto this central control mechanism to adjust channel activity.

2 Regulation of KACh by G Proteins

In 1985, several reports established the involvement of GTP-binding (G) proteins in transducing the ACh signal to KACh (Breitwieser & Szabo, 1985; Endoh, Maruyama, & Iijima, 1985; Pfaffinger, Martin, Hunter, Nathanson, & Hille, 1985; Sorota, Tsuji, Tajima, & Pappano, 1985). The involvement of G proteins to stimulate KACh in response to ACh was also demonstrated using excised membrane patches from atrial cells with ACh or adenosine on the external side of the patch and GTP on the cytoplasmic side, an effect prevented by prior pertussis toxin treatment of the cells (Kurachi, Nakajima, & Sugimoto, 1986). Purified G proteins applied to inside-out patches of atrial cell membranes could induce KACh activity (Yatani, Codina, Brown, & Birnbaumer, 1987). In fact, the ß? subunits of G proteins (Gß?) were found to be responsible for KACh activation, identifying the KACh channel as the first effector protein to be activated by nanomolar concentrations of the Gß? subunits (Logothetis, Kurachi, Galper, Neer, & Clapham, 1987). Later in the same year, it was reported that picomolar concentrations of active Ga subunits, activated with a nonhydrolyzable GTP analog (GTP?S), could induce KACh activation (Codina, Yatani, Grenet, Brown, & Birnbaumer, 1987), igniting a controversy that took several years to settle in favor of Gß?, which is now accepted as the stimulatory component of G proteins (Ito et al., 1992; Logothetis, Kim, Northup, Neer, & Clapham, 1988, Reuveny et al., 1994; Wickman et al., 1994; reviewed by Stanfield, Nakajima, & Nakajima, 2002). Yet, a role for Ga subunit in the regulation of KACh activity is not fully understood and has not been definitively excluded, as activated Ga has been reported by other labs to exhibit weak stimulatory (Ito et al., 1992; Logothetis et al., 1988) or inhibitory effects (Schreibmayer et al., 1996; Vivaudou et al., 1997). Chapter "The Roles of Gß? and Ga in Gating and Regulation of GIRK Channels" by Dascal and Kahanovitch explores in detail the role of Ga in regulating G protein-sensitive K+ channels.

3 Molecular Constituents of Kir3 Channels and Physiological Roles

In 1993, the first molecular component of KACh, Kir3.1 (also referred to as GIRK1), was cloned from heart cDNA libraries, either through PCR, using degenerate primers based on the related Kir2.1 (Kubo, Baldwin, Jan, & Jan, 1993) or by expression cloning in Xenopus oocytes (Dascal et al., 1993). Two additional cDNAs from a brain library were obtained by low stringency hybridization using Kir3.1 as the probe (Lesage et al., 1994). Kir3.2 (or GIRK2) and Kir3.3 (or GIRK3) are expressed mainly in the CNS (Jelacic, Sims, & Clapham, 1999; Lesage et al., 1994, 1995). In 1995, Kir3.4 (also referred to as GIRK4 or CIR) was identified as the second molecular component of KACh (Krapivinsky et al., 1995). Although Kir3.4 are expressed as homomers in heart (e.g., Bender et al., 2001; Corey & Clapham, 1998), it is their coexpression into heteromers with Kir3.1 that gives rise to a conductance that behaves like KACh both biophysically and in response to G protein subunit modulation. In the brain, Kir3 channels can exist as homomers (e.g., Kir3.2a and Kir3.2c: Inanobe et al., 1999). However, several neurons coexpress several Kir3 subunits (Kir3.1-Kir3.4) (e.g., Chen, Ehrhard, Goldowitz, & Smeyne, 1997; Cruz et al., 2004; Karschin, Dissmann, Stuhmer, & Karschin, 1996) and form multiple heteromeric combinations (e.g., Kir3.1/3.2: Inanobe et al., 1999; Liao, Jan, & Jan, 1996; Kir3.2/3.3: Cruz et al., 2004; Inanobe et al., 1999; Jelacic, Kennedy, Wickman, & Clapham, 2000; Kir3.2/3.4: Lesage et al., 1995; Spauschus et al., 1996; reviewed by Hibino et al., 2010). Although Kir3.1 expression alone yields nonfunctional channels mostly trapped in the endoplasmic reticulum, heteromeric combinations of Kir3.2 and Kir3.4 with Kir3.1 produce greatly enhanced currents compared to homomeric assemblies (Chan, Sui, Vivaudou, & Logothetis, 1996). Single point mutations on the pore helices of Kir3 channels have yielded high...