![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
q/11-Mediated Modulation of Ion Channels in Central Nervous System Neurons
Department of Pharmacology and Toxicology, Medical College of Georgia, Augusta, Georgia
Received September 16, 2005; accepted December 20, 2005
Abstract
Slow synaptic potentials are generated when metabotropic G-protein-coupled receptors activate heterotrimeric G-proteins, which in turn modulate ion channels. Many neurons generate excitatory postsynaptic potentials mediated by G-proteins of the G
q/11 family, which in turn activate phospholipase C-
. Accessory GTPase-activating proteins (GAPs) are thought to be required to accelerate GTP hydrolysis and rapidly turn off G-proteins, but the involvement of GAPs in neuronal Gq/11 signaling has not been examined. Here, we show that regulator of G-protein signaling (RGS) proteins provide necessary GAP activity at neuronal Gq/11 subunits. We reconstituted inhibition of native 2-pore domain potassium channels in cerebellar granule neurons by expressing chimeric G subunits that are activated by Gi/o-coupled receptors, thus bypassing endogenous Gq/11 subunits. RGS-insensitive variants of these chimeras mediated inhibition of potassium channels that developed and recovered more slowly than inhibition mediated by RGS-sensitive (wild-type) chimeras or native Gq/11 subunits. These changes were not accompanied by a change in agonist sensitivity, as might be expected if RGS proteins acted primarily as effector antagonists. The slowed recovery from potassium channel inhibition was largely reversed by an additional mutation that mimics the RGS-bound state. These results suggest that endogenous RGS proteins regulate the kinetics of rapid Gq/11-mediated signals in central nervous system neurons by providing GAP activity.
subunit of a G
heterotrimer. This exchange promotes the dissociation of G-GTP and G subunits, both of which can directly or indirectly modify the function of ion channels. Many receptors couple exclusively to one family of G subunits and thereby activate a specific set of effector molecules. For example, Gi/o proteins often activate inwardly rectifying potassium channels, whereas Gq/11 proteins often inhibit potassium channels or activate nonselective cation channels (Hille, 1994
). In neurons, these transduction pathways are responsible for metabotropic inhibitory and excitatory postsynaptic potentials (EPSPs), respectively.
G-protein signaling terminates when G-GTP is hydrolyzed to G-GDP and the heterotrimer reassociates. Although G subunits possess intrinsic GTPase activity, this activity is too slow to account for the rapid termination of many G-protein-mediated signals. The timely termination of brief physiological signals thus depends on the activity of GTPase-activating proteins (GAPs), which bind to GTP-bound G subunits and greatly accelerate the rate of GTP hydrolysis (Ross and Wilkie, 2000). Regulator of G-protein signaling (RGS) proteins are ubiquitous proteins that have GAP activity at Gi/o,Gt, and Gq/11 proteins (Hepler, 1999; Neubig and Siderovski, 2002). As such, these proteins can regulate the strength and kinetics of a variety of G-protein signals. RGS proteins are the only known GAPs at Gi/o and Gt proteins, and rapid termination of events mediated by these subunits clearly depends on RGS proteins (Doupnik et al., 1997; Saitoh et al., 1997; Chen et al., 2000). In contrast, the GTPase activity of Gq/11 proteins can be accelerated not only by RGS proteins but also by the effector molecule phospholipase C- (PLC) (Berstein et al., 1992; Hepler et al., 1997; Chidiac and Ross, 1999). Thus, it is not clear to what extent RGS proteins are essential for terminating transient Gq/11-mediated signals such as slow EPSPs. In addition, RGS proteins could negatively regulate Gq/11 signals by serving as "effector antagonists", competing with the effector PLC for binding to active G-proteins (Hepler et al., 1997).
We tested the hypothesis that endogenous RGS proteins regulate Gq/11-mediated signals by providing GAP activity in CNS neurons. Several previous studies of endogenous RGS protein function in intact cells have relied on functional replacement of native G subunits with RGS-insensitive (RGSi) mutants (Chen and Lambert, 2000; Jeong and Ikeda, 2000, 2001). We adopted a similar strategy by expressing RGS-sensitive and RGSi variants of G subunit chimeras consisting of Gq with nine amino acids at the carboxy terminus substituted for those found in Gi (Conklin et al., 1993). These chimeric (Gqi9) subunits are activated by receptors that normally activate Gi/o subunits but couple to downstream effectors that normally interact with Gq/11 subunits. Using this system, we reconstituted inhibition of "leak" two-pore domain potassium (K2P) channels with RGS-sensitive and RGSi G subunits. We found that signals mediated by RGSi subunits develop and decay more slowly than those mediated by endogenous Gq/11 or RGS-sensitive Gqi9 subunits. These results suggest that endogenous RGS proteins are essential for the rapid kinetics of Gq/11-mediated synaptic potentials.
Materials and Methods
The coding sequence for the chimeric G-protein subunit Gqi9 (provided by Dr. Bruce R. Conklin, University of California, San Francisco, San Francisco, CA) was subcloned into pcDNA 3.1+ (Invitrogen, Carlsbad, CA). The plasmid coding for the 2 adrenoreceptor (2AR) was obtained from the Guthrie Research Institute cDNA resource (www.cdna.org). A plasmid coding for the enhanced green fluorescent protein (EGFP)-RGS2 fusion protein was provided by Dr. Peter Chidiac (University of Western Ontario, London, ON, Canada). QuikChange (Stratagene, La Jolla, CA) mutagenesis was carried out according to the manufacturer's instructions, and mutagenesis was confirmed by automated oligonucleotide sequencing.
Dissociated cultures of cerebellar granule neurons (CGNs) were prepared from 5- to 8-day-old Sprague-Dawley rats as described previously (Chen et al., 2004). CGNs were maintained in minimal essential medium supplemented with 5% fetal bovine serum, 25 mM KCl, 2% B-27 (Invitrogen, Carlsbad, CA), 0.1% MITO serum extender (BD Collaborative, San Diego, CA), 0.6% glucose, 1 mM pyruvate, 50 IU/ml penicillin, and 50 µg/ml streptomycin in 5% CO2 at 37°C.
Neurons were transfected after 6 days in culture using polyethylenimine as described previously (Chen et al., 2004). Neurons were cotransfected with a plasmid coding for pEGFP-N1 (Clontech, Mountain View, CA) and were identified for recording by fluorescence microscopy. Control neurons were transfected with vector (pcDNA 3.1+) and marker (pEGFP-N1) only. Transfection efficiency using this method was low (<5%) but provided ample green fluorescent protein-positive cells for consistent recordings. Control experiments were consistent with the idea that the amount of protein expressed in individual cells was related to the amount of DNA added during transfection, although we made no attempt to quantify the levels of protein expressed in individual cells. Electrophysiological recordings were carried out 14 to 48 h after transfection.
Recordings were carried out at room temperature (24°C) on the stage of an inverted fluorescence microscope. Neurons were constantly perfused with a solution containing 150 mM NaCl, 2.5 mM KCl, 10 mM HEPES, 10 mM glucose, 0.5 mM CaCl2, and 0.5 mM MgCl2, pH 7.8; osmolality, 
310 mOsm/kg H2O). Solution changes were made by switching between perfusion reservoirs using a series of pinch valves (Warner Instruments, Hamden, CT). All recordings were made using the whole-cell perforated patch configuration. Recording pipettes were filled with a solution containing 70 mM K+-gluconate, 70 mM KCl, 0.2 mM EGTA, 10 mM HEPES, and 0.03% amphotericin B, pH 7.3; 295 mOsm/kg H2O). Recordings began 5 min after seal formation, at which time series resistance reached a stable minimum because of patch perforation. Neurons were held in voltage-clamp mode at -20 mV and were stepped to -110 mV for 20 ms every 450 ms. Standing outward K+ current (IKSO) was measured as the absolute holding current at -20 mV. Currents were digitized and recorded using a multifunction I/O board (National Instruments, Austin, TX) and WinWCP software (provided by Dr. J. Dempster, University of Strathclyde, Glasgow, Scotland).
For imaging EGFP-RGS2 translocation, HEK 293 cells were cotransfected with plasmids coding for EGFP-RGS2 and various Gqi9 subunits using LipofectAMINE (Invitrogen) according to the manufacturer's instructions. After at least 16 h, cells were imaged using a Zeiss LSM 510 confocal microscope (Carl Zeiss Inc., Thornwood, NY). Single optical sections through the center of each cell were analyzed using ImageJ software (http://rsb.info.nih.gov/ij/) by drawing a profile normal to and centered on the plasma membrane and plotting the fluorescence intensity along this profile. Because of the limits of diffraction and imperfect centering of these profiles, some fluorescence seems to originate on the extracellular side of the plasma membrane (Fig. 2A).
|
subunits and RGS-sensitive chimeras could be fitted adequately with a single exponential function, recovery mediated by RGSi chimeras began after a lag and an overshoot and thus could not be fitted by a simple function. We presume that different mechanisms rate-limit response recovery with and without RGS function. Concentration-response curves were fitted to a logistic equation using Origin version 6.1 software (OriginLab Corp., Northampton, MA). Numerical values are expressed as mean ± S.E.M. Statistical comparisons were made by ANOVA and unpaired t tests.
Results
Reconstitution of IKSO Inhibition in Cerebellar Granule Neurons with Gqi9 Chimeras. Cerebellar granule neurons express "leak" potassium channels of the tandem- or K2P family (Watkins and Mathie, 1996). These channels are constitutively open and mediate an IKSO that contributes to the resting membrane potential of these cells (Watkins and Mathie, 1996). Activation of Gq/11-coupled receptors such as the endogenous m3 muscarinic acetylcholine receptors expressed by these cells inhibits this current. Because inhibition of similar channels underlies slow EPSPs in many neurons (Hille, 1994), we chose to use these cells as a model to study the involvement of endogenous RGS proteins in neuronal Gq/11 signaling. As described previously (Watkins and Mathie, 1996), application of the nonselective cholinergic agonist carbachol (50 µM) reversibly inhibited IKSO in cultured CGNs, as indicated by a decrease in the outward holding current at a membrane potential of -20 mV and a decrease in whole-cell conductance (Fig. 1A). Previous work has shown that inhibition of K2P channels in these and other cells is mediated by pertussis toxin-insensitive Gq/11-proteins and activation of PLC (Czirjak et al., 2001; Chemin et al., 2003). We confirmed these observations by applying pertussis toxin (100 ng/ml overnight), which had no effect on IKSO inhibition, and by transiently transfecting a constitutively active mutant of Gq (Gq Q209L), which tonically inhibited IKSO and occluded inhibition by carbachol (data not shown). After carbachol was removed from the bath, IKSO recovered with a t1/2 of 5.6 ± 0.9 s (n = 7). Control experiments in which IKSO was inhibited by application of low pH medium documented that the rate of solution exchange using our perfusion system was considerably faster (t1/2 2 s) than the rate of IKSO recovery, suggesting that the latter accurately reflected the termination of Gq/11 signaling. When recovery from carbachol-induced IKSO inhibition was fitted to a single exponential function the corresponding rate constants ranged from 0.09 to 0.27 s-1, values that are approximately 8- to 30-fold faster than the reported intrinsic rate of GTP hydrolysis by purified Gq (0.012/s) (Mukhopadhyay and Ross, 1999; Ross and Wilkie, 2000).
|
subunits. To accomplish this, we used G-protein chimeras consisting largely of Gq except for the nine amino acids at the carboxy terminus, which were identical to those found in the C terminus of Gi. These Gqi9 subunits are activated by receptors that normally activate Gi/o-proteins, but in turn activate effectors that are normally activated by Gq/11 subunits such as PLC (Conklin et al., 1993). As such, we expected that expressing Gqi9 would allow us to inhibit IKSO in CGNs by activating Gi/o-coupled receptors, thus bypassing endogenous Gq/11 subunits. Indeed, we found that activation of native or transiently transfected Gi/o-coupled receptors reversibly inhibited IKSO in neurons that were transfected with Gqi9 but not those that were transfected with a control vector or receptor alone. For example, when CGNs were cotransfected with plasmid vectors coding for the 2AR and Gqi9 application of 10 µM norepinephrine (NE) reversibly inhibited IKSO in a manner similar to that produced by application of carbachol in untransfected or control cells (Fig. 1B). Inhibition of IKSO by NE under these conditions recovered with a t1/2 of 6.7 ± 0.5 s (n = 5), which was not significantly different from recovery from carbachol-induced inhibition in parallel controls (P = 0.37). IKSO was not inhibited by NE in neurons that were transfected only with the 2AR, suggesting that inhibition of IKSO by these receptors was mediated by coexpressed Gqi9 rather than endogenous G subunits. In early experiments, we activated Gqi9 with coexpressed A1 adenosine receptors instead of 2ARs. However, in these experiments, the recovery from adenosine-induced IKSO inhibition was substantially slower than recovery from inhibition by native muscarinic receptors (t1/2 25 s; data not shown) (Chen et al., 2004). This result is consistent with a recent report by Benians et al. (2003) that ligand unbinding may rate limit the recovery of physiological responses mediated by some receptors, including A1 adenosine receptors. Because this limitation could mask any effect of endogenous RGS proteins, we used 2ARs for all of our experiments with RGSi Gqi9 mutants.
IKSO Inhibition Mediated by RGS-Insensitive Gqi9 Chimeras. Having reconstituted IKSO inhibition with exogenous Gqi9 subunits, we introduced point mutations to render this subunit insensitive to endogenous RGS proteins (RGSi). In the first set of experiments, we tested two mutations using three different constructs: Gqi9 G188S (GS), Gqi9 S211D (SD), and a double mutant Gqi9 G188S:S211D (GS:SD). These mutations are located in the highly conserved switch regions of the G subunit and have been shown by several studies to greatly inhibit RGS protein binding without affecting endogenous GTPase activity, nucleotide exchange rates, or effector activation (DiBello et al., 1998; Lan et al., 1998; Natochin and Artemyev, 1998). Although these mutations have been characterized well in previous physiological studies (Chen and Lambert, 2000; Jeong and Ikeda, 2000, 2001), we confirmed the RGSi phenotype of the GS and SD mutants in the context of the Gqi9 chimera by testing the ability of constitutively active (R183C mutant) versions of each to translocate EGFP-RGS2 from the nucleus to the plasma membrane of HEK 293 cells (Heximer et al., 2001; Roy et al., 2003). As expected, the RGSi mutants did not recruit EGFP-RGS2 to the plasma membrane, whereas constitutively active Gqi9 R183C efficiently translocated EGFP-RGS2 (Fig. 2). The nuclear/cytosolic ratio of EGFP-RGS2 was also unchanged by the RGSi mutants, whereas constitutively active Gqi9 R183C recruited EGFP-RGS2 from the nucleus (data not shown).
As was the case with "wild-type" (WT) Gqi9, transfection of any of the three RGSi mutants enabled 2ARs to inhibit IKSO in transfected CGNs (Fig. 3A). However, recovery from NE-induced inhibition was significantly slower for all three of the RGSi mutants compared with parallel WT Gqi9 controls (t1/2 = 8.6 ± 0.1 s; n = 21). This slowed recovery was consistently more pronounced for the GS single mutant (t1/2 = 20.8 ± 1.8 s; n = 21; P < 0.01) than the SD single mutant (t1/2 = 12.5 ± 1.7 s; n = 16; P < 0.05; Fig. 3B). There was a synergistic effect of the two mutations, because inhibition of IKSO mediated by the GS:SD double mutant recovered with a t1/2 of 40.9 ± 2.5 s (n = 13; P < 0.01). Slow recovery from inhibition mediated by RGSi mutants was accompanied by a change in the shape of the recovery time course. Recovery from IKSO inhibition mediated by WT Gqi9 could be fitted well with a single exponential function and began with minimal delay after solution exchange. In contrast, recovery from IKSO inhibition mediated by the most affected RGSi Gqi9 mutants (GS and GS:SD) began after a delay of several seconds (Fig. 3D). These results are consistent with the hypothesis that endogenous RGS proteins in CGNs accelerate the termination of Gq signaling by acting as GAPs.
|
i/o-mediated events (Doupnik et al., 1997; Saitoh et al., 1997; Jeong and Ikeda, 2000, 2001). We observed a similar effect of endogenous RGS proteins on Gqi9-mediated inhibition of IKSO in CGNs. The rate of onset of IKSO inhibition mediated by two of the three RGSi mutants was slower than that mediated by WT Gqi9 (Fig. 3B). Changes in onset kinetics paralleled changes in recovery kinetics. Thus, the GS mutant was more affected than the SD mutant, and the double GS:SD mutant was the most affected. The time to reach half-maximal inhibition (onset t1/2) of IKSO was 2.3 ± 0.1 s for WT Gqi9 and 2.4 ± 0.1 s for the SD mutant. These values were not significantly different (P > 0.05) and may have been limited by the speed of solution exchange. However, the onset t1/2 values for the GS (5.0 ± 0.3 s) and GS:SD (6.7 ± 0.8 s) mutants were significantly different from that of WT Gqi9 (P < 0.05). The slow onset of IKSO inhibition mediated by RGSi Gqi9 subunits suggests that endogenous RGS proteins may facilitate Gq/11 activation by an unknown mechanism, as may also be the case for Gi/o subunits (see Discussion).
The maximal extent of IKSO inhibition at steady state was similarly effected by RGSi mutations. IKSO was inhibited 62 ± 2% by WT Gqi9, 66 ± 4% by the SD mutant (P > 0.05 versus WT Gqi9), 46 ± 4% by the GS mutant (P < 0.05 versus WT Gqi9), and 28 ± 3% by the GS:SD mutant (P < 0.05 versus WT Gqi9). Increasing the amount of plasmid DNA transfected into CGNs failed to increase the magnitude of steady-state inhibition by RGSi chimeras, even though control experiments using G subunits to buffer G subunits indicated more G protein was expressed (data not shown). One possible explanation for the decreased steadystate inhibition of IKSO by RGSi mutants is that these mutants are constitutively active and thus produce tonic inhibition of IKSO, which would occlude agonist-induced inhibition. However, the amplitude of IKSO before agonist application was 239 ± 20 pA (n = 9) in neurons transfected with WT Gqi9 compared with 255 ± 42 pA in neurons transfected with Gqi9 GS:SD (n = 10; P = 0.69); thus, no tonic inhibition was evident. Moreover, after partial inhibition mediated by RGSi Gqi9, IKSO could be inhibited further by activation of endogenous muscarinic receptors, presumably acting via endogenous Gq/11 subunits. These results suggest that RGSi Gqi9 subunits are not constitutively active and are unable to fully activate downstream effector molecules.
Because RGS proteins are negative regulators of G-protein signaling, they have been shown to decrease the agonist sensitivity of some Gi/o-mediated events (Jeong and Ikeda, 2000). Endogenous RGS proteins could also decrease the agonist sensitivity of Gq/11-mediated events by acting either as GAPs or effector antagonists (Hepler et al., 1997). We asked whether endogenous RGS proteins regulated the agonist sensitivity of Gq/11-mediated signals in a similar manner by constructing concentration-response curves for NE-induced inhibition of IKSO. If endogenous RGS proteins predominantly enhanced the rate of GTP hydrolysis by Gqi9 subunits, it was predicted that RGSi mutants would show a leftward shift in the NE concentration-response curve. However, as shown in Fig. 4 the concentration-response curves for RGSi mutants were slightly right shifted compared with WT Gqi9, with EC50 values increased by 2- to 5-fold in this series of experiments. A similar shift was not observed in a second, identical series of experiments (see below); thus, we conclude that steady-state agonist sensitivity is not regulated by RGS proteins in these cells.
|
A Mutation That Mimics the Actions of RGS Proteins Speeds Recovery from IKSO Inhibition Mediated by RGSi Gqi9 Chimeras. A recent mutational analysis of the switch regions of Gi1 by Thomas et al. (2004) identified a highly conserved glycine residue (G202 in Gi1) that when mutated to alanine greatly increased this mutant's rate of GTP hydrolysis. The GTPase activity of Gi1 G202A was increased only marginally by added RGS4, even though the affinity of this mutant for RGS4 was increased. Therefore, this mutation seemed to stabilize the pretransition state of the G-protein in a manner similar to RGS protein binding. We reasoned that introducing the corresponding mutation into RGSi Gqi9 chimeras could have a similar effect and might thus reverse the changes brought about by disrupting RGS binding. Therefore, we constructed Gqi9 chimeras containing this RGS-mimicking (RGSm) mutation (G207A) in the RGSi Gqi9 G188S and G188S:S211D backgrounds. As was the case with the other chimeras we tested, Gqi9 constructs containing the G207A mutation allowed activation of cotransfected 2ARs to reversibly inhibit IKSO. As shown in Fig. 5, recovery from NE-induced inhibition of IKSO mediated by the RGSi/RGSm Gqi9 G188S:G207A (GS:GA) and GS: SD:GA chimeras was not significantly different from that mediated by WT Gqi9, whereas recovery from inhibition mediated by their RGSi counterparts was significantly slower, as observed previously (P < 0.05; ANOVA followed by Bonferroni comparison of means). Thus, this RGSm mutation was largely able to reverse the effect of the RGSi mutations on IKSO recovery kinetics. In contrast, the RGSm mutation failed to restore rapid onset kinetics, because all RGSi and RGSi/RGSm mutants mediated IKSO inhibition with significantly slower onset than WT Gqi9 (Fig. 5C). However, the onset of IKSO inhibition mediated by the GS:SD:GA chimera was significantly faster than that mediated by the RGSi GS:SD mutant (Fig. 5C; P < 0.05), suggesting that this mutation partially restored onset kinetics. Peak steady-state inhibition was decreased for all of the RGSi chimeras, and the addition of the RGSm mutation failed to alter this change (Fig. 5D). However, steady-state agonist sensitivity was unaltered for any of the RGSi or RGSi/RGSm chimeras, as indicated by unchanged EC50 values (n = 6-13; P > 0.05; ANOVA). Together, these results suggest that the slow response recovery characteristic of RGSi chimeras could be counteracted by a mutation that restored GAP activity, and the slow response onset mediated by RGSi chimeras could be partially reversed by restoring GAP activity.
|
Overexpression of RGS16-EGFP Accelerates Recovery from IKSO Inhibition. The above-mentioned results suggest that slowing the rate of GTP hydrolysis by interfering with endogenous RGS protein GAP activity can slow the termination of Gq/11 signals. If timely termination of these signals depends on RGS GAP activity, then it might be possible to accelerate the termination of events mediated by endogenous Gq/11 proteins by increasing the availability of RGS proteins. To test this possibility, we transfected cerebellar granule neurons with RGS2-EGFP and RGS16-EGFP, because both of these RGS proteins are known to act as GAPs at Gq/11 subunits. However, RGS2-EGFP was not expressed efficiently in these cells, as indicated by the absence of detectable green fluorescence. Cotransfection with dsRed indicated that neurons were actually transfected (data not shown), and transfection of HEK cells under identical conditions resulted in robust RGS2-EGFP expression (Fig. 2). In contrast, RGS16-EGFP expressed well in cerebellar granule neurons, producing substantial fluorescent signal located in the cytosol, extending into neuronal processes (Fig. 6A, inset). We tested the ability of RGS16-EGFP to alter IKSO inhibition mediated by endogenous muscarinic receptors and Gq/11 proteins by comparing responses to the agonist carbachol in neurons transfected with RGS16-EGFP and EGFP alone as a control. As shown in Fig. 6, RGS16-EGFP overexpression had no significant effect on the onset or peak magnitude of IKSO inhibition. However, RGS16-EGFP overexpression significantly accelerated recovery from IKSO inhibition from 6.1 ± 0.6 to 3.9 ± 0.3 s (n = 17; P < 0.05). RGS16-EGFP did not alter the steady-state sensitivity of IKSO inhibition mediated by native receptors and G-proteins, because EC50 values were not significantly different (7.0 ± 1.1 µM for EGFP; 8.8 ± 1.8 µM for RGS16-EGFP; P > 0.05). This result supports our conclusion that the kinetics of Gq/11 signals in these neurons are regulated by RGS GAP activity and suggests that the availability of RGS proteins may limit the degree of this regulation.
|
The GAP activity of RGS proteins has been widely studied using in vitro assays, and the function of these proteins in intact cells has been well established using heterologous expression systems. Genetic approaches have also documented the role of endogenous RGS proteins in physiological events in vivo (Chen et al., 2000; Oliveira-Dos-Santos et al., 2000). However, relatively little is known about the roles of endogenous RGS proteins at the cellular level. This deficit largely reflects the absence of effective methods to disable RGS proteins in native cells. Much of what is known about endogenous RGS function has been learned through the use of RGS-insensitive mutants of Gi/o subunits (Chen and Lambert, 2000; Jeong and Ikeda, 2000, 2001). These studies have confirmed that endogenous RGS proteins regulate Gi/o-mediated signals much as would be predicted from studies in heterologous systems.
Our aim was to apply this strategy to Gq/11-mediated signals in CNS neurons. Endogenous RGS proteins could regulate GTP-bound Gq/11 by acting as GAPs or by binding to active Gq/11 and blocking access to effectors (Hepler et al., 1997). A significant difference between Gi/o subunits and Gq/11 subunits is that RGS proteins are the only known GAPs for the former, whereas both RGS proteins and PLC are GAPs at the latter (Berstein et al., 1992). A second difference between Gi/o subunits and Gq/11 subunits is that RGS proteins could serve as effector antagonists at the latter by binding to activated Gq/11 subunits and thus preventing binding of PLC. Rapid signals mediated by Gi/o subunits are generally carried by G dimers and thus are not susceptible to effector antagonism. The possibility that RGS proteins regulate Gq/11 subunits in CNS neurons is an important question, because these subunits mediate synaptic potentials generated by several neurotransmitters.
We found that RGSi mutants of Gqi9 mediated inhibition of IKSO that recovered more slowly than inhibition produced by WT Gqi9 or native Gq/11 subunits. Our results are thus consistent with the hypothesis that endogenous RGS proteins function as GAPs at GTP-bound Gq/11 subunits in CNS neurons, as has been suggested previously in pancreatic acinar cells (Luo et al., 2001). This conclusion depends on the assumption that the RGSi mutations we introduced into Gqi9 interfered with RGS GAP activity rather than (for example) the intrinsic rate of GTP hydrolysis. Likewise, interpretation of our experiments using the G207A mutation depends on extrapolating biochemical results obtained with Gi1 (Thomas et al., 2004). Previous biochemical characterization of the RGSi mutations has suggested that the intrinsic hydrolysis rate is not changed (DiBello et al., 1998; Lan et al., 1998; Natochin and Artemyev, 1998), but we would emphasize that these mutants have not been biochemically characterized in the Gqi9 background, have not been biochemically characterized in combination and have not been challenged with a wide range of RGS proteins. Despite that all of the mutated residues are located in highly conserved regions of the G sequence, it is possible that functional characteristics other than those intended could have been altered in our experiments (see below). Nonetheless, the results we obtained are entirely consistent with previous studies of endogenous RGS function at Gi/o subunits.
Our results suggest that the kinetics of slow EPSPs can be governed by RGS proteins, and it is therefore possible that changes in the availability of RGS proteins could result in changes in slow synaptic transmission. Indeed, increasing the availability of RGS proteins modestly accelerated recovery from inhibition of IKSO. This result suggests that the rate of RGS-accelerated GTP hydrolysis limits the decay rate of Gq/11-mediated signals. Of course, the time course of Gq/11-mediated signals might be dictated by different factors in different cells. For example, muscarinic inhibition of the M-current in sympathetic neurons recovers over the course of hundreds of seconds, even though termination of Gq/11 signaling in these cells seems to be substantially faster (Delmas et al., 2002; Suh and Hille, 2002). Inhibition of the M-current and K2P channels by Gq/11-coupled receptors is thought to be mediated by depletion of phosphatidylinositol 4,5-bisphosphate (Czirjak et al., 2001; Suh and Hille, 2002; Chemin et al., 2003; Zhang et al., 2003); thus, it is possible that slow recovery from M-current inhibition in sympathetic neurons is rate limited by phosphatidylinositol 4,5-bisphosphate resynthesis rather than GTP hydrolysis.
Because PLC is also a GAP at Gq subunits, it is important to consider the possibility that the mutations we introduced into Gqi9 altered response kinetics by interfering with the GAP activity of PLC. This possibility is unlikely for several reasons. First, there is little structural homology between PLC and RGS proteins; thus, the regions of the G subunit contacted by the two GAPs are likely to be different. This idea is supported by the structure of RGS4 complexed to Gi1 (Tesmer et al., 1997), which indicates steric and/or electrostatic interactions between the residues we mutated (G183 and S206 in Gi1) and RGS4. In contrast, these residues do not directly contact PLC in a structural model of Gq docked to the PLC C-terminal dimer (J. Sondek and A. Singer, personal communication) (Singer et al., 2002). This assertion is consistent with biochemical characterization of Gq G188S, which indicates that this mutant responds to the GAP activity of PLC well, with an unchanged EC50 and a maximal effect that reaches 50% of the wild-type protein. In contrast, this mutant responds to the GAP activity of RGS4 very poorly, with a highly right-shifted EC50 (E. Ross and S. Nayak, personal communication). Second, RGSi Gqi9 subunits mediated responses with slowed onset kinetics. Changes in onset kinetics have been consistently observed in studies where RGS proteins are heterologously overexpressed or with RGSi Gi/o subunits (Chen and Lambert, 2000; Jeong and Ikeda, 2000, 2001). Our observations thus imply there is a common (albeit unknown) mechanism whereby RGS proteins participate in G-protein activation (see below). Finally, Jeong and Ikeda (2000) have previously studied the calcium channel inhibition mediated by the GS, SD, and GS:SD mutants of GoA, which is not subject to PLC GAP activity. In that study and the present study, these three mutants differed with respect to the degree of kinetic slowing in an analogous manner: the SD mutant was less effected than the GS mutant, and the double GS:SD mutant was the most effected. Given the structural dissimilarity between RGS proteins and PLC, it would be surprising if two point mutations known to interfere with RGS protein binding also interfered with PLC GAP activity and did so to the same relative degree. Although we cannot rule out an effect of these RGSi mutations on PLC GAP activity, we conclude that the effects we observed largely reflect disruption of endogenous RGS protein function.
Finally, although the slowing of IKSO recovery kinetics is relatively easy to interpret in light of the known GAP activity of RGS proteins at GTP-bound Gq/11 subunits, the slowing of the onset of IKSO inhibition is more difficult to explain. Because the time for a response to reach equilibrium is inversely proportional to the sum of the activation and deactivation rates, part of the slowed response onset with RGSi Gqi9 subunits could be explained by disruption of GAP activity. However, the fact that maximal response amplitude is decreased with RGSi Gqi9 subunits implies that there is an additional positive effect of endogenous RGS proteins on Gq/11 signaling. This situation mirrors several previous studies of the regulation of Gi/o response kinetics in heterologous and native systems (Doupnik et al., 1997; Saitoh et al., 1997; Chen and Lambert, 2000; Jeong and Ikeda, 2000, 2001). As yet, there is no general agreement regarding the mechanism whereby RGS proteins accelerate Gi/o response onset. One possibility is that RGS proteins serve a scaffold function by binding to receptors, G-proteins, and/or effectors. Interactions between receptors and RGS proteins have been demonstrated and invoked to explain receptor-specific RGS actions (Zeng et al., 1998; Zhang et al., 2002), including receptors that couple to Gq/11 (Bernstein et al., 2004). More recently, Tinker and colleagues have suggested that RGS proteins form stable complexes with inactive G-protein heterotrimers and thus participate in a quaternary complex (Benians et al., 2005). A competing hypothesis suggests that RGS proteins serve as kinetic rather than physical scaffolds (Ross and Wilkie, 2000; Zhong et al., 2003) by promoting rapid cycling of active and inactive G-protein. Although our experiments did not directly address this issue, our results are more consistent with the former explanation. A mutation that restored GTPase activity (G207A) to RGSi mutants failed to fully reverse the slow onset of IKSO inhibition. Because kinetic scaffolding would depend entirely on GAP activity, restoring GAP activity should have restored normal onset kinetics if this were the sole mechanism. In either case, our results indicate that acceleration of response onset is not limited to responses mediated by Gi/o subunits.
In summary, we have shown that RGSi Gqi9 chimeras mediate signals with slowed onset and recovery kinetics in primary neurons. These results are consistent with the idea that endogenous RGS proteins provide essential GAP activity for the regulation of slow synaptic events mediated by Gq/11 subunits.
Acknowledgements
We thank John Sondek, Alex Singer, Elliot Ross, and Surendra Nayak for sharing unpublished results. We thank Bruce Conklin and Peter Chidiac for supplying plasmids encoding Gqi9 and EGFP-RGS2, respectively. We thank John Dempster for providing WinWCP software.
Footnotes
This work was supported by National Institutes of Health grant NS36455.
Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org.
ABBREVIATIONS: EPSP, excitatory postsynaptic potential; GAP, GTPase-activating protein; RGS, regulator of G-protein signaling; PLC, phospholipase C-; CNS, central nervous system; RGSi, regulator of G-protein signaling-insensitive; 2AR, 2 adrenoreceptor; EGFP, enhanced green fluorescent protein; CGN, cerebellar granule neuron; IKSO, standing outward K+ current; HEK, human embryonic kidney; ANOVA, analysis of variance; NE, norepinephrine; GS, Gqi9 G188S; SD, Gqi9 S211D; GS:SD, double mutant Gqi9 G188S:S211D; WT, wild-type; RGSm, regulator of G-protein signaling-mimicking.
Address correspondence to: Dr. Nevin Lambert, Department of Pharmacology and Toxicology, Medical College of Georgia, Augusta, GA 30912-2300. E-mail: nlambert{at}mcg.edu
References
Benians A, Leaney JL, and Tinker A (2003) Agonist unbinding from receptor dictates the nature of deactivation kinetics of g protein-gated K+ channels. Proc Natl Acad Sci USA 100: 6239-6244.
Benians A, Nobles M, Hosny S, and Tinker A (2005) Regulators of G-protein signalling form a quaternary complex with the agonist, receptor and G-protein: a novel explanation for the acceleration of signalling activation kinetics. J Biol Chem 280: 13383-13394.
Berstein G, Blank JL, Jhon DY, Exton JH, Rhee SG, and Ross EM (1992) Phospholipase C-beta 1 is a GTPase-activating protein for Gq/11, its physiologic regulator. Cell 70: 411-418.[CrossRef][Medline]
Bernstein LS, Ramineni S, Hague C, Cladman W, Chidiac P, Levey AI, and Hepler JR (2004) RGS2 binds directly and selectively to the M1 muscarinic acetylcholine receptor third intracellular loop to modulate Gq/11 signaling. J Biol Chem 279: 21248-21256.
Chemin J, Girard C, Duprat F, Lesage F, Romey G, and Lazdunski M (2003) Mechanisms underlying excitatory effects of group I metabotropic glutamate receptors via inhibition of 2P domain K+ channels. EMBO (Eur Mol Biol Organ) J 22: 5403-5411.[CrossRef][Medline]
Chen CK, Burns ME, He W, Wensel TG, Baylor DA, and Simon MI (2000) Slowed recovery of rod photoresponse in mice lacking the GTPase accelerating protein RGS9-1. Nature (Lond) 403: 557-560.[CrossRef][Medline]
Chen H, Clark MA, and Lambert NA (2004) Endogenous RGS proteins regulate presynaptic and postsynaptic function: functional expression of RGS-insensitive Galpha subunits in central nervous system neurons. Methods Enzymol 389: 190-204.[Medline]
Chen HM and Lambert NA (2000) Endogenous regulators of G protein signaling proteins regulate presynaptic inhibition at rat hippocampal synapses. Proc Natl Acad Sci USA 97: 12810-12815.
Chidiac P and Ross EM (1999) Phospholipase C- 1 directly accelerates GTP hydrolysis by Gq and acceleration is inhibited by G subunits. J Biol Chem 274: 19639-19643.
Conklin BR, Farfel Z, Lustig KD, Julius D, and Bourne HR (1993) Substitution of three amino acids switches receptor specificity of Gq alpha to that of Gi alpha. Nature (Lond) 363: 274-276.[CrossRef][Medline]
Czirjak G, Petheo GL, Spat A, and Enyedi P (2001) Inhibition of TASK-1 potassium channel by phospholipase C. Am J Physiol 281: C700-C708.
Delmas P, Wanaverbecq N, Abogadie FC, Mistry M, and Brown DA (2002) Signaling microdomains define the specificity of receptor-mediated InsP3 pathways in neurons. Neuron 34: 209-220.[CrossRef][Medline]
DiBello PR, Garrison TR, Apanovitch DM, Hoffman G, Shuey DJ, Mason K, Cockett MI, and Dohlman HG (1998) Selective uncoupling of RGS action by a single point mutation in the G protein -subunit. J Biol Chem 273: 5780-5784.
Doupnik CA, Davidson N, Lester HA, and Kofuji P (1997) RGS proteins reconstitute the rapid gating kinetics of Gbetagamma-activated inwardly rectifying K+ channels. Proc Natl Acad Sci USA 94: 10461-10466.
Hepler JR (1999) Emerging roles for RGS proteins in cell signalling. Trends Pharmacol Sci 20: 376-382.[CrossRef][Medline]
Hepler JR, Berman DM, Gilman AG, and Kozasa T (1997) RGS4 and GAIP are GTPase-activating proteins for Gq alpha and block activation of phospholipase C beta by gamma-thio-GTP-Gq alpha. Proc Natl Acad Sci USA 94: 428-432.
Heximer SP, Lim H, Bernard JL, and Blumer KJ (2001) Mechanisms governing subcellular localization and function of human RGS2. J Biol Chem 276: 14195-14203.
Hille B (1994) Modulation of ion-channel function by G-protein-coupled receptors. Trends Neurosci 17: 531-536.[CrossRef][Medline]
Jeong SW and Ikeda SR (2000) Endogenous regulator of G-protein signaling proteins modify N-type calcium channel modulation in rat sympathetic neurons. J Neurosci 20: 4489-4496.
Jeong SW and Ikeda SR (2001) Differential regulation of G protein-gated inwardly rectifying K+ channel kinetics by distinct domains of RGS8. J Physiol (Lond) 535: 335-347.
Lan KL, Sarvazyan NA, Taussig R, Mackenzie RG, DiBello PR, Dohlman HG, and Neubig RR (1998) A point mutation in Go and Gi1 blocks interaction with regulator of G protein signaling proteins. J Biol Chem 273: 12794-12797.
Luo X, Popov S, Bera AK, Wilkie TM, and Muallem S (2001) RGS proteins provide biochemical control of agonist-evoked [Ca2+]i oscillations. Mol Cell 7: 651-660.[CrossRef][Medline]
Mukhopadhyay S and Ross EM (1999) Rapid GTP binding and hydrolysis by Gq promoted by receptor and GTPase-activating proteins. Proc Natl Acad Sci USA 96: 9539-9544.
Natochin M and Artemyev NO (1998) Substitution of transducin Ser202 by Asp abolishes G-protein/RGS interaction. J Biol Chem 273: 4300-4303.
Neubig RR and Siderovski DR (2002) Regulators of G-protein signalling as new central nervous system drug targets. Nat Rev Drug Discov 1: 187-197.[CrossRef][Medline]
Oliveira-Dos-Santos AJ, Matsumoto G, Snow BE, Bai D, Houston FP, Whishaw IQ, Mariathasan S, Sasaki T, Wakeham A, Ohashi PS, et al. (2000) Regulation of T cell activation, anxiety and male aggression by RGS2. Proc Natl Acad Sci USA 97: 12272-12277.
Ross EM and Wilkie TM (2000) GTPase-activating proteins for heterotrimeric G proteins: regulators of G protein signaling (RGS) and RGS-like proteins. Annu Rev Biochem 69: 795-827.[CrossRef][Medline]
Roy AA, Lemberg KE, and Chidiac P (2003) Recruitment of RGS2 and RGS4 to the plasma membrane by G proteins and receptors reflects functional interactions. Mol Pharmacol 64: 587-593.
Saitoh O, Kubo Y, Miyatani Y, Asano T, and Nakata H (1997) RGS8 accelerates G-protein-mediated modulation of K+ currents. Nature (Lond) 390: 525-529.[CrossRef][Medline]
Singer AU, Waldo GL, Harden TK, and Sondek J (2002) A unique fold of phospholipase C-beta mediates dimerization and interaction with G alpha q. Nat Struct Biol 9: 32-36.[CrossRef][Medline]
Suh BC and Hille B (2002) Recovery from muscarinic modulation of M current channels requires phosphatidylinositol 4,5-bisphosphate synthesis. Neuron 35: 507-520.[CrossRef][Medline]
Tesmer JJG, Berman DM, Gilman AG, and Sprang SR (1997) Structure of RGS4 bound to AlF4-activated Gi1: stabilization of the transition state for GTP hydrolysis. Cell 89: 251-261.[CrossRef][Medline]
Thomas CJ, Du X, Li P, Wang Y, Ross EM, and Sprang SR (2004) Uncoupling Conformational change from GTP hydrolysis in a heterotrimeric G protein alpha-subunit. Proc Natl Acad Sci USA 101: 7560-7565.
Watkins CS and Mathie A (1996) A non-inactivating K+ current sensitive to muscarinic receptor activation in rat cultured cerebellar granule neurons. J Physiol (Lond) 491: 401-412.
Zeng WZ, Xu X, Popov S, Mukhopadhyay S, Chidiac P, Swistok J, Danho W, Yagaloff KA, Fisher SL, Ross EM, et al. (1998) The N-terminal domain of RGS4 confers receptor-selective inhibition of G protein signaling. J Biol Chem 273: 34687-34690.
Zhang H, Craciun LC, Mirshahi T, Rohacs T, Lopes CM, Jin T, and Logothetis DE (2003) PIP2 Activates KCNQ channels and its hydrolysis underlies receptor-mediated inhibition of M currents. Neuron 37: 963-975.[CrossRef][Medline]
Zhang Q, Pacheco MA, and Doupnik CA (2002) Gating properties of GIRK channels activated by Go- and Gi-coupled muscarinic M2 receptors in Xenopus oocytes: the role of receptor precoupling in RGS modulation. J Physiol (Lond) 545: 355-373.
Zhong H, Wade SM, Woolf PJ, Linderman JJ, Traynor JR, and Neubig RR (2003) A spatial focusing model for G protein signals. Regulator of G protein signaling (RGS) protein-mediated kinetic scaffolding. J Biol Chem 278: 7278-7284.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
