| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Journal of Neurophysiology Vol. 79 No. 1 January 1998, pp. 379-391
Copyright ©1998 by the American Physiological Society
1 Department of Molecular Physiology and Biophysics, Vanderbilt University School of Medicine, Nashville, Tennessee 37203; 2 Mécanismes Moléculares des Communications Cellulaires, Centre National de la Recherche Scientifique-Institut National de la Santé et de la Recherche Médicale de Pharmacolgie-Endocrinologie, Monpelier Cedex 5, France; and 3 SIBIA Neurosciences, La Jolla, California 92037
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
McCool, Brian A., Jean-Phillipe Pin, Michael M. Harpold, Paul F. Brust, Kenneth A. Stauderman, and David M. Lovinger. Rat group I metabotropic glutamate receptors inhibit neuronal Ca2+ channels via multiple signal transduction pathways in HEK 293 cells. J. Neurophysiol. 79: 379-391, 1998. We have shown previously that metabotropic glutamate receptors with group I-like pharmacology couple to N-type and P/Q-type calcium channels in acutely isolated cortical neurons using G proteins most likely belonging to the Gi/Go subclass. To better understand the potential mechanisms forming the basis for group I mGluR modulation of voltage-gated calcium channels in the CNS, we have examined the ability of specific mGluRs to couple to neuronal N-type (
1B-1/2
/
1b) and P/Q-type (1A-2/2/1b) voltage-gated calcium channels in an HEK 293 heterologous expression system. Using the whole cell patch-clamp technique where intracellular calcium is buffered to low levels, we have shown that group I receptors inhibit both N-type and P/Q-type calcium channels in a voltage-dependent fashion. Similar to our observations in cortical neurons, this voltage-dependent inhibition is mediated almost entirely by N-ethylmaleimide (NEM)-sensitive heterotrimeric G proteins, strongly suggesting that these receptors can use Gi/Go-like G proteins to couple to N-type and P/Q-type calcium channels. However, inconsistent with the apparent NEM sensitivity of group I modulation of calcium channels, modulation of N-type channels in group I mGluR-expressing cells was only partially sensitive to pertussis toxin (PTX), indicating the potential involvement of both PTX-sensitive and -resistant G proteins. The PTX-resistant modulation was voltage dependent and entirely resistant to NEM and cholera toxin. A time course of treatment with PTX revealed that this toxin caused group I receptors to slowly shift from using a primarily NEM-sensitive G protein to using a NEM-resistant form. The PTX-induced switch from NEM-sensitive to -resistant modulation was also dependent on protein synthesis, indicating some reliance on active cellular processes. In addition to these voltage-dependent pathways, perforated patch recordings on group I mGluR-expressing cells indicate that another slowly developing, calcium-dependent form of modulation for N-type channels may be seen when intracellular calcium is not highly buffered. We conclude that group I mGluRs can modulate neuronal Ca2+ channels using a variety of signal transduction pathways and propose that the relative contributions of different pathways may exemplify the diversity of responses mediated by these receptors in the CNS.
The regulation of intracellular calcium is a cornerstone for many cellular responses to extracellular signals in the CNS. The best examples of this are in presynaptic terminals where the influx of calcium through voltage-gated calcium channels is absolutely required for active release of neurotransmitter or in dendritic processes where the rising Ca2+i underlies the signal transduction events leading to long-term alterations of synaptic efficacy. These essential second messenger roles for Ca2+i may be contrasted with the apparent neurotoxic effects of failed Ca2+i homeostasis during ischemic insult in many brain regions. A knowledge of Ca2+ influx/release in the CNS therefore has served not only to further our understanding of the physiology underlying proper neuronal function but to better comprehend the role of Ca2+i in the pathophysiology of the CNS.
Cell culture and transfection
Isolation and characterization of the HEK 293 cell line stably expressing the human Isolation of superior cervical ganglion neurons
Superior cervical ganglion (SCG) neurons were acutely isolated from adult female Sprague-Dawley (250-275 g) rats using slight modifications of a previously reported procedure (Ikeda et al. 1995 Solutions and drugs
The internal pipette solution for whole cell recordings consisted of (in mM) 125 N-methylglucamine (NMG), 20 tetraethylammonium (TEA)-OH; 14 Tris2-phosphocreatine, 10 HEPES, 11 ethylene glycol-bis( Electrophysiology
All recordings were performed at ambient room temperature with standard patch-clamp techniques (Hamil et al. 1981 Voltage-dependent modulation of calcium channels by group I receptors
Receptor pharmacology
The pharmacological characterization of mGluR1a and mGluR5a coupled to N-type calcium channels in this HEK 293 cell system is depicted in Fig. 1, E and F. For both receptors, inhibition by each agonist was characterized by a rapid onset, was readily reversible, and did not attenuate with repeated or prolonged exposures (1-2 min) to high concentrations (>100 µM). The data for each concentration of agonist plotted in Fig. 1, E and F, were normalized to a maximally efficacious concentration of L-glutamate (100 µM) in each cell to facilitate comparisons among agonists. For mGluR1a-expressing cells, EC50 values were 0.5 µM, 3.2 µM, and 110 µM for DHPG, L-glutamate, and t-ACPD, respectively. Similarly, for mGluR5a-expressing cells, EC50 values for DHPG, L-glutamate, and t-ACPD were 0.9, 0.7, and 9.2 µM, respectively. The rank order of agonist potency for both receptors, DHPG Modulation by group I mGluRs is mediated primarily by Gi/Go-like G proteins
G protein
PTX treatment reveals an additional signal transduction pathway for N-type channel modulation
To confirm the signal transduction pathway used by group I mGluRs to inhibit N-type calcium channels involves Gi/o-like G proteins, we treated transiently transfected cells overnight with PTX (0.5 µg/ml; 12-18 h), a treatment paradigm that completely eliminates modulation of calcium channels by group II mGluRs in HEK 293 cells (McCool et al. 1996
Characterization of the PTX-resistant pathway
Modulation by mGluR1a and mGluR5a using this PTX/NEM-resistant pathway did not appear to be the result of some mechanistically novel signal transduction pathway. The modulation still was relieved partially after a large, depolarizing prepulse, indicating that the PTX-resistant pathway was voltage dependent (Fig. 4, A and B). In fact, the relative contribution of voltage dependent and -independent inhibition was similar regardless of the types of G proteins (NEM sensitive vs. NEM resistant) or the types of mGluR (group I vs. group II) coupled to channel inhibition. For the data shown in Fig. 4C, "voltage-independent" modulation was defined as the inhibition remaining after a depolarizing prepulse, whereas "voltage-dependent" modulation was calculated as the difference between total percent inhibition and voltage-independent inhibition. In mGluR5a-expressing cells, voltage-independent modulation represented 49% of the total inhibition. Similarly, for mGluR5a-expressing cells treated 18-24 h with PTX (0.5 µg/ml), it is 48% of the total inhibition. Likewise, for the strictly PTX-sensitive coupling between mGluR2 and these channels (McCool et al. 1996
Group I receptors initiate irreversible inhibition of N-type channels in perforated patch recordings
Group I receptors have been shown previously to couple to signal transduction pathways that stimulate phosphoinositol metabolism and release of intracellular calcium in both native (Schoepp et al. 1990
Group I metabotropic glutamate receptors have been long thought of as the "PI-coupled"/"Gq-coupled" branch of the mGluR family. However, group I-like receptors in various preparations initiate physiological responses not classically associated with PTX-resistant G proteins, like the voltage-dependent inhibition of voltage-gated calcium channels (Choi and Lovinger 1996
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
1, 2, and ). Although the function of the 2 subunit is not well understood (but see Brust et al. 1993
), the subunit is an essential ancillary subunit required for high expression levels of neuronal calcium channels and alters the biophysical properties of the calcium current (Oclese et al. 1994). The 1 subunit forms the ion conducting pore and provides the binding sites for pharmacological agents that have been used to identify, along with the channel's biophysical properties, each channel subtype. For the studies presented below, the 1B- (or 
-conotoxin GVIA-sensitive) and the 1A- (or the -agatoxin IVA-sensitive) containing channels were chosen for examination because they mediate the vast majority of excitation/secretion coupling in the CNS (Lovinger et al. 1994; Takahashi and Momiyama 1993). These channel subtypes also are modulated by a large diversity of G-protein-coupled receptors in an ever-expanding list of central and peripheral nervous tissues (Beech et al. 1992; Choi and Lovinger 1996; Swartz and Bean 1992).
), whereas group II and III mGluRs inhibit adenylyl cyclase via activation of PTX-sensitive G proteins (Tanabe et al. 1992).
), modulation of resting membrane currents (Guérineau et al. 1994; Zheng and Gallagher 1995), inhibition of voltage-gated calcium channels via activation of Gi/Go-like G proteins (Choi and Lovinger 1996; Hay and Kunze 1994), and modulation of synaptic transmission via presynaptic autoreceptors (Gereau and Conn 1995). These physiological responses underlie the apparent role of group I mGluRs in the regulation of neuronal excitability (Desai and Conn 1991; McBain et al. 1994) and in both long- and short-term changes of synaptic efficacy in various brain regions (Bashir et al. 1993; Glaum et al. 1992). The ability of a single class of receptors to couple to these widely divergent signal transduction pathways is a poorly understood aspect of group I mGluR function in the CNS.
i/o, s, q, and 13) have been detected (Toth et al. 1996). Our results show that, unlike group II receptors expressed in this HEK 293 cell heterologous system (McCool et al. 1996), group I receptors can use several distinct signal transduction pathways to inhibit voltage-gated calcium channel function. These studies may help to further define the multifunctional nature group I mGluRs in the CNS.
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
1B-1, 2b/, 1b calcium channel subunits has been described previously (McCool et al. 1996). Stable cell lines expressing the human 1A-2-subunits in place of the 1B-1 subunit were isolated using similar techniques. Cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM; GIBCO/BRL, Grand Island, NY) supplemented with bovine calf serum (5.5%, Hyclone Labs, Salt Lake City, UT), 100 U/ml penicillin G/100 µg/ml streptomycin (GIBCO/BRL), and 0.5 mg/ml G418 (Sigma Chemical, St. Louis, MO, or CellGro) using a 37°C/5% CO2 incubator.
) and involved a CaPO4-mediated protocol. Expression constructs containing group I receptor cDNAs were under the direction of cytomegaloviros (CMV)-based promoters and have been described elsewhere (see Joly et al. 1995). Transiently transfected cells were used 36-48 h after the introduction of the expression plasmid. The N-type channel-expressing cell line was used without any special culturing techniques. However, after transfection, the P/Q-type channel-expressing cell line was incubated at 37°C for 10-12 h to allow for receptor expression then shifted to 28°C >6 h before electrophysiology to maximize channel expression. Although we do not fully understand the apparent need for this exposure to lower temperatures to maximize channel expression, the most likely explanation would involve the interaction between nonnative protein folding pathways available to the channel at higher temperatures and overexpression of the channel protein, a potential hazard in any heterologous system.
). Briefly, ganglia were isolated and incubated in DMEM (Sigma Chemical) containing 10 mM N-2-hydroxyethylpiperazine-N
-2-ethanesulfonic acid (HEPES; pH 7.4 with NaOH), 1 mg/ml collagenase type D (Boehringer Mannheim Biochemicals, Indianapolis, IN), 0.3 mg/ml trypsin (Sigma Chemical), and 0.1 mg/ml DNase I (Sigma Chemical) at 37°C for 1 h. Neurons then were dispersed by vigorous shaking and washed twice with plating media, which contained DMEM, 10% fetal bovine serum (FBS; Hyclone Labs), and 1% penicillin-streptomycin solution (GIBCO/BRL). Dispersed neurons then were plated onto poly-L-lysine-coated dishes and used for recordings within 24 h.
-aminoethyl ether)-N,N,N,N-tetraacetic acid, 1 CaCl2, 4 Mg-ATP, and 0.3 Tris-GTP; pH 7.2 with methane sulfonic acid and HCl (final [Cl] was 10 mM); adjust to 300-310 mmol/kg with sucrose. For perforated-patch recordings, the internal pipette solution contained (mM) 140 CsCl, 10 HEPES, and 2 MgCl2; pH 7.2 with CsOH. Amphotericin B (Sigma Chemical) was added from a concentrated stock in dimethyl sulfoxide to the CsCl internal at a final concentration of 50 µg/ml; this solution was sonicated vigorously immediately before use to ensure a relatively homogeneous suspension.
S,2S-(carboxycyclopropyl)glycine (L-CCGI; Tocris Cookson) were diluted fresh daily from frozen stocks. PTX and cholera toxin (Research Biochemicals International) were added to final concentrations of 500 ng/ml and 1 µg/ml, respectively, from concentrated stocks made according to the manufacturer's instructions.
) using the axopatch-1D amplifier (Axon Instruments, Foster City, CA) in the voltage-clamp mode. Gigaohm seals were formed using polished patch pipettes made from borosilicate glass (World Precision Instruments, Sarasota, FL). For the whole cell patch-clamp recording configuration, patch pipettes typically had input resistances of 0.5-2 M
using the NMG/TEA internal solution; whole cell capacitance (typically 10-40 pF) and series resistance (typically <10 M) were compensated manually after opening the cell. For perforated patch recordings, patch pipettes had input resistances of 0.25-1 M using the CsCl/amphotericin B internal; and whole cell capacitance and series resistance were generally 10-25 pF and <20 M, respectively. Whole cell currents were typically on-line leak-subtracted using a p/n protocol; and all currents were low-pass filtered (3-pole Bessel filter) at 1-5 kHz with >70% compensation. Depolarizing test pulses to potentials where inward barium current was maximal were given at 0.25 Hz to prevent prolonged channel inactivation. Data were digitized at 
10 kHz with a Labmaster DMA (Axon Instruments), stored on a 486 microprocessor, and analyzed off-line using pClamp software (Axon Instruments). Current amplitudes for whole cell and perforated-patch traces were measured from cursers placed immediately before the initiation of a depolarizing test pulse and at the peak of the calcium current, typically 15 ms after initiation of the test pulse. To calculate percent inhibition, the average of amplitudes before and after agonist application were compared with the average current amplitude during agonist application (typically 30 s or 2 episodes). Numerical analysis was performed using the QuatroPro software package (v 5.00; Borland International, Scotts Valley, CA). Current traces, other graphs, and Hill plots were generated using GraphPad Prism (v 1.03; GraphPad Software, San Diego, CA). Statistical analysis was performed using InStat (v 2.05; GraphPad Software).
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
1B-expressing (N-type) cells transiently transfected with mGluR1a and mGluR5a were chosen for study based on cell morphology, with receptor-expressing cells generally being fairly large (30-40 pF) and having a "granular" appearance. Although the specific reasons for this unique appearance are unknown, these characteristics seem independent of the subtype of mGluR cDNA transiently expressed (group I vs. group II) and may be related generally to overexpression as similar results are seen with other kinds of proteins (McCool, unpublished observations). Twenty-six of 111 mGluR1a-transfected cells (50.4 ± 2.6% inhibition; mean ± SE) and 29 of 67 mGluR5a-transfected cells (41.8 ± 2.4% inhibition) responded to application of a saturating concentration of L-glutamate (100 µM) with robust and reversible inhibition of whole cell N-type barium currents. Similar results were obtained for the splice variants of both mGluR1 and mGluR5. Specifically, mGluR1b inhibited whole cell N-type barium currents (data not shown) by 41.4 ± 2.8% in four of eight cells tested, whereas mGluR1c modulated calcium channels (data not shown) in two of eight cells tested with an average of 30.5% inhibition. Likewise, mGluR5b-mediated inhibition (38.6 ± 8.5%, data not shown) was found in 5 of 10 cells tested. With regard to P/Q-type channels, modulation was generally smaller in the 1A-expressing (P/Q-type) stable cell line. Nonetheless, 6 of 28 mGluR1a transfected cells (25.2 ± 5.2% inhibition), and 6 of 14 mGluR5a transfected cells (25.8 ± 3.3% inhibition) 1A-expressing cells responded to 100 µM glutamate with robust and reversible inhibition. Mock transfected (vector alone) or untransfected 1B-expressing cells (McCool et al. 1996) and 1A-expressing cells (data not shown, n = 5) did not respond to application of any mGluR agonist, indicating that these cells do not express endogenous glutamate receptors capable of inhibiting barium currents. Glutamate application to mGluR1a- or mGluR5a-expressing cells did not cause any changes in resting membrane currents when using either whole cell or perforated-patch (see further text) recording configurations. The relatively low percentage of responding cells (20-50% of cells tested) was highly dependent on the specific expression construct but was within the dish-to-dish variability seen during transient expression studies with CMV-driven green fluorescent protein constructs (data not shown).
. In this protocol, two test pulses to moderately depolarized potentials at which inward barium currents were maximal (approximately 
10 mV from a holding potential of 90 mV) are separated by a large membrane depolarization (+80 mV). Voltage dependence is represented by a "relief" from inhibition in the second test pulse relative to the first test pulse. For both mGluR1a and mGluR5a coupled to either N-type or P/Q-type channels, the large intervening depolarization partially relieved inhibition (Fig. 1, A-D), indicating that modulation was indeed voltage dependent. In all cases, modulation in the first test depolarization was characterized by an apparent slowing of macroscopic activation kinetics (see Fig. 1, A-D, left); however, "slowing" appeared more pronounced for modulation of human P/Q-type channels than for human N-type channels. Together, these properties indicate that group I modulation of N-type and P/Q-type calcium channels in HEK 293 cells is mediated by a signal transduction pathway similar to the membrane-delimited, voltage-dependent pathway previously described for group II mGluR modulation of N-type calcium channels in HEK 293 cells (McCool et al. 1996) and for mGluR2 modulation of N-type channels when transiently expressed in sympathetic neurons (Ikeda et al. 1995).
View larger version (28K):
[in a new window]
FIG. 1.
Group I mGluR inhibition of N-type (A and B) and P/Q-type (C and D) calcium channels in HEK 293 cells is relieved partially by a depolarizing prepulse. mGluR1a (left) and mGluR5a (right) modulation of these neuronal voltage-gated calcium channels was assessed by separating 2 test voltages ( 10 mV) with a large depolarization (+80 mV). In all cases, inhibition was reduced after the large depolarization: A, 57% inhibition for the 1st test pulse vs. 34% inhibition for the 2nd test pulse; B, 43% vs. 16%; C, 42% vs. 17%; and D, 32% vs. 12%. Voltage protocol is shown at bottom and was taken from Ikeda (1991). Calibration bars for each set of traces are as follows: x = 25 ms (A-D); A, y = 0.5 nA; B, y = 0.25 nA; C, y = 0.1 nA; and D, y = 0.3 nA. E and F: pharmacology of mGluR1a- and mGluR5a-mediated inhibition of N-type calcium channels. Responses to various agonist concentrations were normalized to 100 µM L-glutamate in each cell to facilitate comparisons between cells and among agonists. Normalized data were plotted against the log10 of agonist concentration and fitted using a standard Hill (variable-slope) equation. EC50 values for L-glutamate (
), (R,S)-3,5-dihydroxyphenyl glycine (DHPG; 
), and (±)-1-aminocyclopentane-trans-1,3-dicarboxylic acid (t-ACPD; 
) were 3.2 µM (n = 5-6), 0.5 µM (n = 3-4), and 110 µM (n = 6-7) for mGluR1a (E) and 0.7 µM (n = 4-5), 0.9 µM (n = 3-6), and 9.2 µM (n = 4) for mGluR5a (F). Hill slopes of curves fit for L-glutamate, DHPG, and t-ACPD were 2.3, 1.8, and 1.2 for mGluR1a and 1.9, 1.7, and 1.3 for mGluR5a, respectively. Due to limited solubility, the highest t-ACPD concentration tested was 1 mM.
L-glutamate 
t-ACPD, reported here is roughly similar to values reported elsewhere (reviewed in Conn and Pin 1997). Interestingly, t-ACPD clearly appeared to be a partial agonist for activation of mGluR1a (see also Pickering et al. 1993) but not for mGluR5a. Modulation mediated by mGluR5a also was activated fully by low concentrations of quisqualate (0.1 µM) and partially activated (~70% of maximal) by high concentrations of L-CCGI (10 µM), consistent with group I receptor pharmacology (Conn and Pin 1997).
subunits generally can be classified based on the effectors to which they couple and their relative sensitivity to inactivation or down regulation by naturally occurring bacterial toxins, such as cholera toxin and PTX. Recently, it has been shown that low concentrations of the sulfhydryl alkylating agent, NEM, can preferentially inactivate voltage-dependent calcium channel modulation by Gi/Go subunit-containing, PTX-sensitive G proteins (Kasai 1991). This is presumably due to direct alkylation of the G protein alpha subunit as NEM can prevent ADP-ribosylation of purified Gi and Go alpha subunits (Asano and Ogasawara 1986) and modifies specific amino acid residues, one of which is the substrate for ADP-ribosylation, on purified Go (Hoshino et al. 1990). As there are no alterations of the biochemical properties of Gi/o heterotrimers after treatment with NEM (Asano and Ogasawara 1986), G-protein/receptor uncoupling is presumed to be the major effect of NEM, similar to the actions of PTX.
, and may indicate minimal effects of NEM on the channels themselves.
View larger version (36K):
[in a new window]
FIG. 2.
Group I mGluR modulation of neuronal voltage gated calcium channels in HEK 293 cells is mediated by Gi/Go-like heterotrimeric G proteins. For this and subsequent figures, the numbers in parentheses indicate the number of cells. A and B: after identification of a cell as expressing either mGluR1a or mGluR5a (based on an inhibitory response to agonist), the cell was exposed to minimal concentrations of the sulfhydryl alkylating agent, N-ethylmaleimide (NEM; 50 µM for 2 min), during the recording and subsequently retested with agonist (100 µM L-glutamate). For both receptors coupled to either N-type (A) or P/Q-type (B) channels, modulation was almost entirely NEM sensitive, indicating that group I mGluRs use Gi/Go to modulate neuronal calcium channels in HEK 293 cells. For N-type channels (A), mGluR1a inhibition was reduced from 59.0 ± 4.0% to 7.1 ± 3.0% after NEM treatment, and mGluR5a was reduced from 47.6 ± 3.5% to 9.5 ± 3.8%. Likewise, for P/Q-type channels (B), mGluR1a inhibition fell from 25.3 ± 4.7% to 5.1 ± 1.6% after exposure to NEM, whereas mGluR5a inhibition fell from 25.8 + 3.3% to 8.6 + 2.0%. C and D: NEM selectively interferes with modulation of N-type calcium channels by Gi/Go-coupled ( 2-adrenergic) (Shapiro et al. 1994a) but not Gs-coupled [vasoactive intestinal peptide (VIP)] (Zhu and Ikeda 1994), receptors in acutely isolated superior cervical ganglia neurons. Modulation of whole cell barium currents by the 2-adrenergic receptor agonist UK14304 (C) was reduced from 33.6 ± 6.3% to 9.6 ± 2.5% after exposure to NEM. Conversely, inhibition mediated by VIP (D) is insensitive to NEM, with inhibition being 44.2 ± 3.0% before exposure and 40.7 ± 3.9% after exposure.
2-adrenergic receptors via a voltage-dependent, primarily PTX-sensitive pathway (Beech et al. 1992) was also acutely sensitive to exposure to NEM (Fig. 2C), consistent with previous reports (Shapiro et al. 1994a). Conversely, voltage-dependent modulation of N-type channels in SCG neurons by vasoactive intestinal peptide (VIP) receptors was resistant to NEM (Fig. 2D). These results are consistent with previous reports that VIP inhibition of N-type channels in SCG neurons is mediated by a cholera toxin (CTX)-sensitive G protein (Zhu and Ikeda 1994) (see also Fig. 3). Additionally, it has been shown previously that muscarinic receptor modulation of M-type potassium channels in SCG neurons is also resistant to NEM (Choi et al. 1995), consistent with this modulation being mediated by the Gq/11-class of G proteins (Caulfield et al. 1994). Together, these results indicate that inhibition mediated by mGluR1a and mGluR5a in both the HEK 293 cell lines was mediated almost entirely by Gi/Go-like G proteins and, as exemplified by the insensitivity of VIP N-type calcium channel modulation to NEM, N-ethylmaleimide most likely does not disrupt steps in the voltage-dependent pathway downstream of the G protein.
View larger version (37K):
[in a new window]
FIG. 3.
Treatment of group I-expressing cells with pertussis toxin (PTX) reveals an additional signal transduction pathway for inhibition of N-type channels. A and B: modulation by mGluR1a (A) or mGluR5a (B) is only partially sensitive to PTX, with inhibition being reduced from 54.1 ± 3.8% to 22.4 ± 2.1% for mGluR1a and from 44.2 ± 2.7% to 23.8 ± 1.9% for mGluR5a using a saturating concentration of L-glutamate (100 µM). This PTX-resistant modulation is also resistant to NEM (25.6 ± 2.9% for mGluR1a and 24.3 ± 3.5% for mGluR5a) and to cholera toxin (CTX; 21.5% for mGluR1a and 23.5 ± 5.0% for mGluR5a). C: a comparison of NEM-resistant modulation without PTX treatment (see Fig. 2A) to that after PTX treatment (see A and B) demonstrates the significantly larger amount of NEM-resistant inhibition remaining after PTX treatment when compared with control modulation. For this comparison, the mean "control" values (100 µM L-glutamate columns, untreated cells) from the N-type channel experiments in Fig. 2A and from A and B were used. For mGluR1a, NEM-resistant modulation was 12.1 ± 5.3% of the total control (i.e., untreated cells) inhibition without PTX treatment (see Fig. 2A) and increased to 47.3 ± 5.4% of control inhibition after PTX treatment (see A). Likewise, NEM-resistant modulation by mGluR5a increased from 19.9 ± 7.9% of control (Fig. 2A) to 55.0 ± 6.4% of control (see B) after treatment with PTX. D: CTX used in A and B inactivates VIP-mediated modulation (see Zhu and Ikeda 1994 ) of N-type channels in superior cervical ganglion (SCG) neurons (decrease from 49.7 ± 2.1% to 6.4 ± 1.7% inhibition) while having very little effect on modulation by the 2-adrenergic agonist, UK14304 (57.5 ± 6.3% inhibition for controls, 49.6 ± 5.1% inhibition for CTX-treated cells).
) and in neuronal expression systems (Ikeda et al. 1995). PTX treatment of mGluR1a- (Fig. 3A) and mGluR5a (Fig. 3B)-transfected cells failed to remove all of the group I modulation of N-type channels, in apparent contradiction to the results obtained using NEM. Specifically, using data taken from three to five independent experiments, inhibition was reduced from 54.1 ± 3.8% (n = 11) and 44.2 ± 2.7% (n = 10) in control mGluR1a- and mGluR5a-expressing cells, respectively, to 22.4 ± 2.1% (n = 8) and 23.8 ± 1.9% (n = 7) after overnight PTX treatment (Fig. 3, A and B). This decrease in inhibition was not due to a shift in maximal efficacy of L-glutamate as the concentration used here (100 µM) was still saturating (not shown). mGluR2-transfected cells tested at the same time using the same concentration of PTX and time of exposure showed complete loss of glutamate inhibition after PTX treatment (not shown, see McCool et al. 1996). Because expression of the P/Q-type channel required incubation of cells at nonphysiological temperatures (see METHODS) and such temperatures may seriously effect the actions of PTX and other cellular processes, we have not yet used this cell line to characterize any additional signal transduction pathways.
2-adrenergic receptor-mediated modulation in these same cells (Fig. 3D), confirming previous observations by Zhu and Ikeda (1994).
View larger version (18K):
[in a new window]
FIG. 4.
PTX/NEM-resistant modulation by mGluR1a and mGluR5a is relieved partially by a large, depolarizing prepulse. A and B: voltage protocol for these traces is given in Fig. 1, bottom. mGluR1a traces (A) were obtained from a cell treated overnight with both PTX and CTX and after acute treatment with NEM. Modulation in the 1st test pulse (27% inhibition) was relieved partially in the 2nd test pulse (17% inhibition) by a large, intervening depolarization. Similarly, mGluR5a modulation from a cell treated overnight with PTX and after acute treatment with NEM (B) also was relieved partially (39 vs. 23% inhibition). Calibration bars for each set of traces are x = 25 ms for all; y = 0.5 nA (A); y = 1.4 nA (B). Unlike the traces in Fig. 1, these traces have not been leak subtracted (···, zero current level) to emphasize the absence of any change in resting membrane current in the presence of the agonist. C: voltage-dependent modulation by PTX-insensitive G proteins and by PTX-sensitive G proteins is similar relative to voltage-independent modulation. Voltage-independent modulation is referred to as the inhibition remaining after a depolarizing prepulse (see Fig. 1). Voltage-dependent modulation was calculated by subtracting the percent voltage-independent inhibition for the total inhibition in the 1st test pulse (A and B, inward currents on left). For mGluR5a in control cells (i.e., using primarily NEM-sensitive G proteins, left), voltage-dependent inhibition was 30.2 ± 3.4%, whereas voltage-independent inhibition was 25.9 ± 2.9% (representing 46% of the total inhibition). Similarly, mGluR5a coupled to NEM-resistant G proteins (i.e., after PTX treatment, middle) inhibited barium currents by 21.3 ± 2.5% voltage dependently and by 18.5 ± 3.2% in a voltage-independent manner (again representing 46% of total). For comparison, mGluR2, which uses entirely PTX/NEM-sensitive G proteins (McCool et al. 1996 ), inhibited barium currents in a voltage-dependent manner by 27.0 ± 4.7% and by 20.5 ± 2.7% using a voltage-independent pathway (representing 43% of the total inhibition).
), voltage-independent inhibition is 43% of the total modulation. This is the first demonstration of a voltage-dependent modulation of N-type calcium channels by PTX/CTX/NEM-insensitive G proteins.
View larger version (21K):
[in a new window]
FIG. 5.
A and B: a time course of PTX treatment indicates that the relative contributions of NEM-sensitive (denoted NEMSensitive) and NEM-resistant (NEMResistant) components to N-type channel modulation are differentially affected. NEMSensitive component was calculated by subtracting theNEM-resistant modulation from the total amount of inhibition; the NEMResistantcomponent is the amount of inhibition after treatment with NEM (50 µM for 2 min). For mGluR1a (A), the percent NEMSensitive inhibition was 52.3 ± 4.0%, 11.0 ± 2.3%, 0 + 5.2%, and 0 ± 2.4% at 0, 2, 5 h, and overnight (ON), respectively; whereas the percent NEMResistant inhibition was 7.7 ± 2.7% (see Fig. 2), 12.3 ± 2.5%, 16.3 ± 6.0%, and 25.6 ± 2.9% at 0, 2, 5 h, and ON, respectively. Similarly for mGluR5a (B), percent NEMSensitive inhibition was 34.8 ± 4.8%, 21.3 ± 6.4%, 2.6 ± 2.4%, and 0 ± 3.1% at 0, 2, 5 h, and ON; whereas percent NEMResistant inhibition was 8.5 ± 3.0%, 10.3 ± 2.5%, 15.4 ± 3.0%, and 24.3 ± 3.5% at these same times. C: development of the NEM/PTX-resistant modulation by mGluR5a is dependent on protein synthesis. Total inhibition ( 
) was compared with NEM-resistant modulation () after PTX treatment (0.5 µg/ml overnight) in the presence or absence of the protein synthesis inhibitor cycloheximide (CHX; 50 µg/ml). Total inhibition values are 45.6 ± 6.4% for control; 35.2 ± 4.6% for CHX alone; 26.3 ± 5.3% for PTX alone; and 9.5 ± 1.9% for PTX + CHX. NEM-resistant inhibition values are 6.3 ± 1.1% for control; 7.8 ± 2.4% for CHX alone; 27.2 ± 6.2% for PTX alone; and 7.9 ± 1.0% for PTX + CHX. Although PTX only partially removes inhibition by 100 µM L-glutamate (PTX-alone percent inhibition is ~58% of control inhibition; ), coincubation with CHX makes the modulation by mGluR5a relatively more PTX sensitive (PTX + CHX percent inhibition is only ~27% of the inhibition after CHX treatment alone; ). This sensitivity to CHX is most evident when comparing the percent NEM-resistant inhibition () after PTX treatment with and without cotreatment with CHX.
). Incubation with CHX modestly and nonsignificantly [P > 0.05, analysis of variance (ANOVA)] decreased the amount of modulation by mGluR5a in cells treated with CHX alone compared with controls (percent inhibition: control, 45.6 ± 6.4%, n = 8; +CHX, 35.2 ± 4.6%, n = 5; 23% reduction in modulation). In contrast, the accumulation of the PTX-resistant modulation (Fig. 5C, open bars) was prevented by incubation with cycloheximide (percent inhibition: PTX treated, 26.3 ± 5.3%, n = 6; CHX and PTX treated, 9.5 ± 1.9%, n = 5; 64% reduction in modulation; significance at P < 0.05, ANOVA). The effects of CHX are even more apparent when one compares the relative levels of NEM-resistant modulation (Fig. 5C, closed bars) with percent inhibition values dropping significantly (P < 0.01, ANOVA) from 27.2 ± 6.2% in cells treated with PTX alone to 7.9 ± 1.0% in PTX + CHX-treated cells, a level of modulation not significantly different from NEM-resistant modulation in control (6.3 ± 1.1%) and CHX-only (7.8 ± 2.1%) cells. These results strongly suggest that development of the NEM-resistant modulation after PTX treatment requires new protein synthesis and more generally that PTX treatment may itself alter signal transduction components.
) and heterologous systems (Kingston et al. 1995; Pin et al. 1992), including HEK 293 cells (Flor et al. 1996). We examined the effect of group I receptor activation on N-type calcium channels using a recording configuration, the perforated patch, that should interfere only minimally with subtle changes in the intracellular environment. After forming a gigaohm seal, depolarizing test voltages were initiated only after maximal electrical access to the cell interior had been achieved as evidenced by saturation of the capacitive transients during small depolarizing steps. Inward barium currents through N-type channels recorded using the perforated-patch configuration were similar to those observed in the whole cell configuration and were stable for many minutes (Fig. 6A). Conversely, in cells transfected with mGluR1a (not shown) or mGluR5a (Fig. 6B), barium currents were modulated by two distinct processes on exposure to agonist (100 µM L-glutamate). These distinct modulatory events differed considerably with regard to the time to onset after agonist exposure and the duration of inhibition.
View larger version (18K):
[in a new window]
FIG. 6.
Perforated-patch recordings indicate that group I receptors can modulate N-type channels using multiple signal transduction pathways that differ in time to onset and duration. Traces were not leak subtracted;- - -, zero current level. A: stability of this recording configuration is exemplified by the sample control traces in which current amplitude varied by <10% during the recording; a = 1 min, b = 2 min, c = 4 min, and d = 7 min after initiation of the recording. Calibration bars are x = 10 ms and y = 0.25 nA. B: in contrast, agonist application to a mGluR5a-expressing cell caused a rapid, reversible modulation [compare trace a (or control), b (or 100 µM L-glutamate, 24% inhibition), and c (or wash; 90% of control trace amplitude)] superimposed on a slowly developing and long-lasting inhibition [d (74% of control trace), e (30% of control), and f (21% of control)]. Traces shown are 2.25 min (a), 2.5 min (b), 3.0 min (c), 3.5 min (d), 6.5 min (e), and 9.5 min (f) after initiation of the recording. Calibration bars are x = 10 ms and y = 0.75 nA.
View larger version (25K):
[in a new window]
FIG. 7.
Irreversible inhibition in the perforated patch configuration is specific to group I receptors and dependent on the agonist-induced release of intracellular calcium. A: a time course of absolute current amplitudes for the mGluR5a-expressing cell shown in Fig. 6B illustrates both the rapid, reversible inhibition and the slowly developing, irreversible inhibition. B: in contrast, a time course for a mGluR2-expressing cell indicates the presence of only the rapid, reversible pathway. For both A and B, agonist (100 µM L-glutamate) application is indicated by 
, and the percent inhibition for each application also is included. ···, linear regression line derived using the "baseline" current amplitudes of each recording (i.e., prior to the 1st agonist application); ×, "expected" current amplitudes based on predictions from the linear regression. Relative differences between the predicted or expected amplitudes and the observed amplitudes are expressed as an expected/observed ratio in C. Generally, receptors initiating the irreversible pathway should give expected/observed ratios significantly more positive than unity. C: summary of perforated patch experiments for group I (mGluR1a and 5a) and a group II (mGluR2) receptors indicate that only group I receptors have ratios significantly different from unity. Furthermore, treatment of mGluR5a-expressing cells with bis-(o-aminophenoxy)-N,N,N,N-tetraacetic acid (BAPTA)-AM (500 µM) prevented the irreversible inhibition, indicating the ability of group I receptors to release Ca2+i in HEK 293 cells may be an essential component of the "irreversible" inhibition. Ratio values are: 2.67 ± 0.51 for mGluR1a; 2.99 ± 0.59 for mGluR5a; 1.43 ± 0.20 for mGluR5a + BAPTA-AM; 0.99 ± 0.02 for mGluR2.
). Furthermore, group I mGluRs stimulate a rise of intracellular calcium from the HEK 293 cells stably expressing N-type calcium channels, even in the absence of extracellular Ca2+ (McCool, unpublished observations). To test whether the irreversible inhibition initiated by group I receptors was indeed related to a rise intracellular calcium, mGluR5a-expressing cells were treated with thecell-permeant analogue of bis-(o-aminophenoxy)-N,N,N,Ntetraacetic acid (BAPTA), BAPTA-AM (500 µM, 37°C for 20 min) and subjected to perforated-patch recordings. BAPTA-loaded cells expressing mGluR5a exhibited significantly less irreversible inhibition (P < 0.05, 2-tailed Student's t-test) than control cells (ratios were ~3.5 for controls and 1.6 for BAPTA-AM-treated cells). These results imply that the release of intracellular calcium plays some critical role in the slowly developing, irreversible modulatorypathway.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
) and the inhibition of synaptic transmission (Gereau and Conn 1995). These findings led us to the possibilities that either additional receptors with a pharmacology similar to group I receptors but signal transduction properties more like group II or III receptors remained to be isolated or the currently identified group I receptors are actually quite "promiscuous" in their G protein selectivity. Hints of the latter phenomenon were revealed by findings that group I receptors may use PTX-sensitive G proteins to stimulate phosphatidylinositol metabolism and arachidonic acid release in various mammalian heterologous systems (Aramori and Nakanishi 1992; Pickering et al. 1993) and to stimulate phosphatidylinositol metabolism (Suzdak et al. 1993) and Ca2+i release (Milani et al. 1993) in cerebellar granule cells. We have shown clearly that the currently known group I receptor clones, when transiently expressed in HEK 293 cells, may use Gi/Go-mediated pathways to modulate rapidly both N-type and P/Q-type calcium channels in a voltage-dependent fashion. However, previous studies using heterologous expression of mGluR1a in SCG neurons (Ikeda et al. 1995) did not indicate a strong coupling between this receptor, PTX-sensitive G proteins, and N-type calcium channels. Nonetheless, the modulation by group I receptors in HEK 293 cells is essentially identical to that seen for quisqualate-mediated inhibition in acutely isolated deep-layer cortical (Choi and Lovinger 1996) and CA3 hippocampal (Swartz and Bean 1992) neurons. Furthermore, the pharmacology of mGluR1a and mGluR5a are consistent with that reported for these receptors in other heterologous and native systems (reviewed by Conn and Pin 1997). Although our result appear to conflict somewhat with the results of Ikeda et al. (1995), coupling of group I receptors to N-type calcium channels in SCG neurons has been accomplished recently using methodologies that allow for consistently high levels of receptor expression (S. Ikeda, personal communication). Together, these observations support the notion that group I receptors can couple to numerous signal transduction pathways in a variety of heterologous as well as native systems using a variety of G proteins.
). Although we have not specifically ruled out the possibility that NEM may directly influence mGlu receptors, the fact that the PTX-resistant modulation by group I mGluRs is also NEM-resistant tends to strengthen the idea that NEM does not significantly alter the properties of group I receptors.
subunits and are replaced only slowly by receptors coupled to PTX-resistant G proteins. Although we have not addressed the latter possibility, our results with the translation inhibitor CHX are consistent with new protein synthesis being an essential component involved in the development of NEM-resistant modulation after PTX treatment. Although not necessary to explain our observations, the contributions of previously existing receptor pools remain to be addressed directly.
). Additionally, modulation of N-type channels by angiotensin II AT1 receptors in sympathetic neurons (Shapiro et al. 1994b) is only partially PTX sensitive while being relatively more NEM sensitive. The results of this study therefore may represent a PTX-induced switch in the types of G proteins used by this receptor similar to that reported here. We currently are examining the possibility that PTX may directly influence the kinds of G proteins used by group I mGluR and other receptors in native neuronal systems. More generally, our results indicate that partially PTX-sensitive responses should be reexamined using independent measures (like NEM). Although it is true that NEM is by no means as specific in its effects at the cellular level as PTX, it also should be realized that PTX may not merely be a benign indicator of Gi/Go involvement in a given receptor-driven response but itself may alter the signal transduction components used by a given receptor. In addition to our results, neurochemical studies of cultured cerebellar granule cells (Cullen et al. 1994) where PTX treatment "induces" a CHX-sensitive increase in glutamate release also support this notion.
1B (N-type) channel subunit is known to be a target for Ca2+i-sensitive kinases, like protein kinase C and Ca2+/calmodulin-dependent protein kinase II, in vitro (Hell et al. 1994). Furthermore, the phosphorylation state of the N-type channel can dramatically influence its biophysical properties (Werz et al. 1993). The time course of such processes and whether they can be initiated by G protein-coupled receptors in HEK 293 cells have yet to be determined.
). Furthermore, the irreversible/Ca2+-dependent pathway in HEK 293 cells appears to associate with the ability of a given receptor to couple to phospholipase C (PLC) and/or PTX-resistant G proteins because both mGluR1a and mGluR5a are capable of initiating this pathway, whereas mGluR2, a receptor that appears to couple only to PTX-sensitive G proteins in this HEK 293 cell system, does not. Because group I mGluRs can activate the PLC/Ca2+i signal transduction pathway in HEK 293 cells (Flor et al. 1996) and activation of this pathway is PTX-resistant (Pin, unpublished observations), our results suggest that the activation of phospholipase C may lie somewhere between the group I mGluRs and the releasable pool of intracellular calcium. Additional experiments will be necessary to unequivocally assign a role for a rise in intracellular calcium in the irreversible pathway and to define the components of this modulatory process both upstream and downstream from intracellular calcium release.
| |
ACKNOWLEDGEMENTS |
|---|
This work was supported by National Institute of Neurological Disorders and Stroke Grants NS-30470 to D. M. Lovinger and National Research Service Award NS-09719 to B. A. McCool.
| |
FOOTNOTES |
|---|
Address for reprint requests: B. A. McCool, Dept. Medical Pharmacology and Toxicology, Texas A&M University Health Science Center, 368 Reynolds Medical Bldg., College Station, TX 77843.
Received 25 April 1997; accepted in final form 4 September 1997.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
-aminobutyric acid B receptors from GTP-binding proteins by N-ethylmaleimide: effect of N-ethylmaleimide on purified GTP-binding proteins.
Mol. Pharmacol.
29: 244-249, 1986.[Abstract] 
q/11 and -adrenoreceptor inhibition of Ca2+ current via Go in rat sympathetic neurons.
J. Physiol. (Lond.)
477: 415-422, 1994. and mGluR5a subtypes.
Neuropharmacology
34: 887-894, 1995.[Medline] -conotoxin GVI-sensitive calcium channel in HEK 293 cells.
Mol. Pharmacol.
50: 912-922, 1996.[Abstract] subunit sets rates of channel inactivation independently of the subunit's effect on activation.
Neuron
13: 1433-1438, 1994.[Medline] This article has been cited by other articles:
|
|
 |
|
  F. Ferraguti, L. Crepaldi, and F. Nicoletti Metabotropic Glutamate 1 Receptor: Current Concepts and Perspectives Pharmacol. Rev., December 1, 2008; 60(4): 536 - 581. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. McDavid and K. P. M. Currie G-Proteins Modulate Cumulative Inactivation of N-Type (CaV2.2) Calcium Channels J. Neurosci., December 20, 2006; 26(51): 13373 - 13383. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Zhang and D. Sulzer Glutamate Spillover in the Striatum Depresses Dopaminergic Transmission by Activating Group I Metabotropic Glutamate Receptors J. Neurosci., November 19, 2003; 23(33): 10585 - 10592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Liu and A. R. Rittenhouse Arachidonic acid mediates muscarinic inhibition and enhancement of N-type Ca2+ current in sympathetic neurons PNAS, January 7, 2003; 100(1): 295 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Tempia, M. E. Alojado, P. Strata, and T. Knopfel Characterization of the mGluR1-Mediated Electrical and Calcium Signaling in Purkinje Cells of Mouse Cerebellar Slices J Neurophysiol, September 1, 2001; 86(3): 1389 - 1397. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Kammermeier, B. Xiao, J. C. Tu, P. F. Worley, and S. R. Ikeda Homer Proteins Regulate Coupling of Group I Metabotropic Glutamate Receptors to N-Type Calcium and M-Type Potassium Channels J. Neurosci., October 1, 2000; 20(19): 7238 - 7245. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Hermans, R. Saunders, J. V. Selkirk, R. Mistry, S. R. Nahorski, and R. A. J. Challiss Complex Involvement of Pertussis Toxin-Sensitive G Proteins in the Regulation of Type 1alpha Metabotropic Glutamate Receptor Signaling in Baby Hamster Kidney Cells Mol. Pharmacol., August 1, 2000; 58(2): 352 - 360. [Abstract] [Full Text]
|
|
|
|
|
|
A. Francesconi and R. M. Duvoisin Opposing effects of protein kinase C and protein kinase A on metabotropic glutamate receptor signaling: Selective desensitization of the inositol trisphosphate/Ca2+ pathway by phosphorylation of the receptor-G protein-coupling domain PNAS, May 23, 2000; 97(11): 6185 - 6190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Varney, N. D. P. Cosford, C. Jachec, S. P. Rao, A. Sacaan, F.-F. Lin, L. Bleicher, E. M. Santori, P. J. Flor, H. Allgeier, et al. SIB-1757 and SIB-1893: Selective, Noncompetitive Antagonists of Metabotropic Glutamate Receptor Type 5 J. Pharmacol. Exp. Ther., July 1, 1999; 290(1): 170 - 181. [Abstract] [Full Text]
|
|
|
|
|
|
F. Tempia, M. C. Miniaci, D. Anchisi, and P. Strata Postsynaptic Current Mediated by Metabotropic Glutamate Receptors in Cerebellar Purkinje Cells J Neurophysiol, August 1, 1998; 80(2): 520 - 528. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
