G Protein Modulation of N-type Calcium Channels Is Facilitated by Physical Interactions between Syntaxin 1A and G bg *

The direct modulation of N-type calcium channels by G protein bg subunits is considered a key factor in the regulation of neurotransmission. Some of the molecular determinants that govern the binding interaction of N-type channels and G bg have recently been identified (see, i.e. , Zamponi, G. W., Bourinet, E., Nelson, D., Nar-geot, J., and Snutch, T. P. (1997) Nature 385, 442–446); however, little is known about cellular mechanisms that modulate this interaction. Here we report that a protein of the presynaptic vesicle release complex, syntaxin 1A, mediates a crucial role in the tonic inhibition of N-type channels by G bg . When syntaxin 1A was coexpressed with (N-type) a 1B 1 a 2 - d 1 b 1b channels in tsA-201 cells, the channels underwent a 18 mV negative shift in half-inactivation potential, as well as a pronounced tonic G protein inhibition as assessed by its reversal by strong membrane depolarizations. This tonic inhibition was dramatically attenuated following incubation with botulinum toxin C, indicating that syntaxin 1A expression was indeed responsible for the enhanced G protein modulation. However, when G protein bg subunits were concomitantly coexpressed, the toxin became ineffective in removing G protein inhibition, suggesting that syntaxin 1A optimizes, rather than being required for G protein modulation of N-type channels. We also demonstrate that G bg physically binds to syntaxin 1A, and that syntaxin 1A can simultaneously 20, 2 m M EDTA, and 0.1% b -mercaptoethanol and lysed by two passages through a French press (Spectronic Instruments Inc.). The fusion protein was recovered by binding of the GST domain to gluta- thione-agarose beads (Sigma). The fusion protein beads were washed extensively and finally resuspended in 0.5% Triton X-100, 20 m M MOPS (pH 7.0), 4.5 m M Mg(CH 3 COO) 2 , 150 m M KCl, and 0.5 m M PMSF. Immunoblotting for Syntaxin 1A-G Protein Binding— Proteins were transferred electrophoretically at constant voltage from polyacrylamide gels to nitrocellulose (0.2 m m) in 20 m M Tris, 150 m M glycine, 12% methanol. Transferred proteins were visualized by staining with Ponceau S. Nitrocellulose membranes were blocked for nonspecific binding using 5% milk, 0.15% Tween 20, PBS solution (137 m M NaCl, 2.7 m M KCl, 4.3 m M Na 2 HPO 4 , 1.4 m M KH 2 PO 4 (pH 7.3)) and incubated over- night with primary antibody (1:1000). The membranes were washed four times in the above milk/Tween 20/PBS solution and incubated for 30 min with goat anti-rabbit or goat anti-mouse IgG-coupled horserad-ish peroxidase. Antigen was detected using chemiluminescent horse- radish peroxidase substrate (ECL, Amersham Pharmacia Biotech). Im-munoreactive bands were visualized following exposure of the membranes to Amersham Hyperfilm-MP. (v/v) four times three buffers

The modulation of presynaptic calcium channel activity by intracellular messenger pathways, including protein kinase C, and G protein ␤␥ subunits (21)(22)(23)(24)(25), has been well documented. Upon activation of G protein-coupled seven-helix transmembrane receptors, both N-type and P/Q-type channels undergo a pronounced voltage-dependent inhibition (26 -30). This inhibition is likely caused by direct binding of G protein ␤␥ subunits to the calcium channel ␣ 1 subunit with 1:1 stoichometry, which results in a stabilization of the deep closed states of the channel (31)(32)(33). This effect can be reversed by strong membrane depolarizations, resulting in an apparent facilitation of calcium currents after application of depolarizing prepulses (22,27,32). Recent molecular biological evidence indicates that the G protein ␤␥ subunits interact with the calcium channel ␣ 1 subunit at multiple points, in particular, the I-II linker and the carboxyl tail regions (34 -36). It has been reported that cleavage of the synaptic vesicle release protein syntaxin 1A with botulinum toxin C1 (BTC1) 1 abolishes the ability of N-type channels to undergo G protein modulation (37). Syntaxin is a cytoplasmically oriented membrane protein that is involved in synaptic vesicle release triggered by the influx of calcium through voltagedependent calcium channels. Proteolysis by BTC1 completely and irreversibly inhibits synaptic release. Because syntaxin 1A is known to physically bind to these channels (19,20), Stanley and Mirotznik (37) suggested that syntaxin 1A might facilitate the co-localization of G␤␥ and the channel, but no direct evidence for this has been presented. Furthermore, because G protein modulation of transiently expressed N-type channels does not require coexpression with syntaxin 1A (22,34), it is * This work was supported in part by an operating grant (to G. W. Z.) from the Medical Research Council of Canada (MRC) and through a scholarship award (to G. W. Z.) from the EJLB Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶Holder of an Alberta Heritage Foundation for Medical Research (AHFMR) studentship.
ʈ unclear why the cleavage of syntaxin 1A in the native neuronal environment would lead to the loss of G protein modulation.
Here, we present evidence that syntaxin 1A augments, but is not required per se for G protein modulation of transiently expressed calcium channels. Human embryonic kidney cells do not endogenously express syntaxin 1A; however, coexpression of N-type calcium channels with G␤ 1 ␥ 2 resulted in a pronounced inhibition of N-type currents. Upon coexpression of syntaxin 1A, N-type channels underwent a large tonic G protein inhibition even in the absence of exogenous G protein ␤␥ subunits. This inhibition was sensitive to botulinum toxin C1. However, the effect of the toxin was abolished upon overexpresssion of exogenous G␤␥ subunits. Biochemical data show that syntaxin 1A is able to physically interact concomitantly with both G␤␥ and the calcium channel. Overall, our data are consistent with a model in which syntaxin 1A physically localizes free endogenous G protein ␤␥ subunits into the vicinity of the N-type calcium channel ␣ 1 subunit, thus increasing the effective local G␤␥ concentration near the channel and thereby facilitating the interactions between the channel and the G protein. P/Q-type calcium channels regulate syntaxin 1A expression (51), which would in turn modulate tonic inhibition of N-type calcium channels by G␤␥. In that context, our data suggest a potential feedback mechanism by which expression of one calcium channel type regulates the activity of another.

Transient Transfection of HEK Cells
Human embryonic kidney tsA-201 cells were grown in standard DMEM medium, supplemented with 10% fetal bovine serum and penicillin-streptomycin. The cells were grown to 85% confluence, split with trypsin-EDTA, and plated on glass coverslips at 10% confluence 12 h before transfection. Immediately prior to transfection, the medium was exchanged and a standard calcium phosphate protocol was used to transfect the cells with cDNAs encoding for calcium channel subunits (␣ 1B ϩ ␣ 2 -␦ ϩ ␤ 1b ), green fluorescent protein (EGFP; CLONTECH), and as appropriate with G␤ 1 , G␥ 2 , and/or syntaxin 1A. After 12 h, the cells were washed with fresh DMEM and allowed to recover for 12 h. Subsequently the cells were incubated at 28°C in 5% CO 2 for 1-3 days prior to recording. For experiments examining protein levels of syntaxin 1A and G␤, an analogous protocol was used; however, the cells were plated directly on the bottom of the culture dish.

Patch Clamp Recordings
Immediately prior to recording, individual coverslips were transferred to a 3-cm culture dish containing recording solution comprised of 20 mM BaCl 2, 1 mM MgCl 2 , 10 mM HEPES, 40 mM tetraethylammonium chloride (TEA-Cl), 10 mM glucose, 65 mM CsCl (pH 7.2 with TEA-OH). Whole cell patch clamp recordings were performed using an Axopatch 200B amplifier (Axon Instruments, Foster City, CA) linked to a personal computer equipped with pCLAMP v 6.0. Patch pipettes (Sutter borosilicate glass, BF150 -86-15) were pulled using a Sutter P-87 microelectrode puller, fire polished using a Narashige microforge, and showed typical resistances of 3-4 megaohms. The internal pipette solution contained 108 mM CsMS, 4 mM MgCl 2 , 9 mM EGTA, 9 mM HEPES (pH 7.2). Data were filtered at 1 kHz and recorded directly onto the hard drive of the computer. Data were analyzed using Clampfit (Axon Instruments) and fitted using Sigmaplot 4.0 (Jandel Scientific). Steady state inactivation curves were fitted with the Boltzman equation I peak (normalized) ϭ 1/(1 ϩ exp((V Ϫ V h )z/25.6)) where V and V h are, respectively, the conditioning and the half-inactivation potential, and z is a slope factor. Unless stated otherwise, all error bars are standard errors, numbers in parentheses displayed in the figures reflect numbers of experiments, and p values given reflect Student's t tests.

Molecular Biology and Biochemistry
Cloning of Syntaxin 1A-A forward primer containing a 5Ј NotI restriction site (5Ј-GCGGCCGCATGAAGGACCGAACCCAGGAGCTC-CGC-3Ј) and a reverse primer containing a 5Ј KpnI restriction site (5Ј-GGTACCTTTCTATCCAAAGATGCCCCCGATGGTGG-3Ј) directed against the 5Ј and 3Ј sequences of rat syntaxin 1A were purchased from University Core DNA services (University of Calgary, Calgary, Alberta, Canada). Rat hypothalamus cDNA was kindly provided by Bob Wink-fein. All PCR reagents were purchased from Life Technologies, Inc. unless otherwise stated. The PCR reaction solution, in a volume of 50 l, consisted of 20 mM Tris-HCl, pH 8.4, 50 mM KCl, dNTPs (0.2 mM each), 1.5 mM MgCl 2 , 2.5 units of platinum Taq DNA polymerase, 20 pmol of each primer, and 50 ng of cDNA. Using a PTC-100HB thermal cycler (MJ Research, Watertown, MA), the reaction was hot-started, and held at 94°C for 2 min. 34 cycles were conducted, which consisted of denaturation for 30 s at 94°C, annealing for 45 s at 65°C, and extension for 1.25 min at 72°C. The resultant syntaxin 1A DNA product was run on a 0.8% agarose gel, extracted, and purified using QIAquick Gel Extraction (Quigen, Mississauga, Ontario, Canada), ligated into a pGEM T-Easy vector (Promega, Madison, WI), and sequenced to rule out PCR errors. The syntaxin 1A-T-Easy construct was digested by KpnI and NotI, and the syntaxin 1A DNA was ligated into pMT2SX for subsequent expression.
Subcloning of G Protein Subunits-cDNAs coding for bovine brain G␤ 1 and G␥ 2 in pBluescript II KS(-) were kindly provided by Dr. T. P. Snutch. The G␤ 1 -pBS construct was digested by SmaI and XbaI to excise G␤ 1 , which was then ligated into similarly digested pSL1180. This G␤ 1 -pSL1180 construct was then digested by KpnI and SalI, and the G␤ 1 fragment was then ligated into the polycloning site of pMT2SX. The G␥ 2 -pBS construct was digested by EcoRV and XbaI to liberate G␥ 2 , which was then ligated into similarly digested pSL1180. This G␤ 1 -pSL1180 construct was then digested by KpnI and SalI, and the G␥ 2 fragment was then ligated into the polycloning site of pMT2SX. The pMT2 constructs containing G␤ 1 and G␥ 2 were used for transient transfection of HEK cells.
Construction of 6ϫHis-Synprint Fusion Proteins-A forward primer containing an XhoI restriction site (5Ј-CACTCGAGAGGAGTTGAC-CAAGGATGAAGAGGAGATGG-3Ј) and a reverse primer (5Ј-CTATCT-TGCGCGACGCTCGCCCTTGGG-3Ј) were used in PCR to amplify a segment of recombinant ␣ 1B cDNA corresponding to the synprint site (amino acids 718 -963) using the same conditions as described for the PCR of syntaxin 1A. The resultant synprint DNA product was run on a 0.8% agarose gel, extracted and purified using QIAquick Gel Extraction (Qiagen), ligated into a pGEM T-Easy vector (Promega), and sequenced. The synprint-T-Easy construct was then digested with XhoI and EcoRI and ligated into similarly digested bacterial expression vector pTrcHis C (Invitrogen, San Diego, CA). This construct was transformed into TOP10 Escherichia coli (Invitrogen). Induction and preparation of a cell lysate were performed using conditions adapted from the manufacturer.
Immunoblots for Syntaxin 1A and G␤␥ in HEK Cells-tsA-201 cells were transfected, as described above. Following the 2-day incubation at 28°C, the DMEM media was washed off and 2 ml of trypsin-EDTA (Life Technologies, Inc.) was added and the cells were incubated for 3 min. 8 ml of PBS buffer (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na 2 HPO 4 -7H 2 O, 1.4 mM KH 2 PO 4 ) was added, and the cells were pelleted by centrifugation at 3000 ϫ g for 2 min. The supernatant was removed, and the pellet was resuspended in 3 ml of homogenization buffer (50 mM HEPES-Tris, pH 7.3, 8% sucrose, 100 mM NaCl, 1 mM EDTA, 10 mM dithiothreitol, 0.1 mM PMSF, 30 g of leupeptin, 30 g of pepstatin, 3 g of aprotonin, 0.1% Triton X-100), followed by three rapid freeze-thaw cycles between a dry ice-methanol slurry and 37°C water. The preparation was centrifuged at 5000 ϫ g for 5 min to remove cell debris, and this supernatant was then spun at 100,000 ϫ g for 1 h. The pellet was then treated with 2ϫ Laemmli sample buffer (100 mM Tris-HCl, pH 6.8, 200 mM ␤-mercaptoethanol, 20% glycerol, 4% SDS, 0.2% bromphenol blue), and shaken for 2 h at 4°C. The proteins were denatured at 95°C for 1 min and loaded for SDS-PAGE.
Proteins were transferred from the gel to Hybond ECL nitrocellulose membrane (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom) at 30 V for 10 h. Transfer buffer was removed by two washes with PBS, and the membrane was blocked in 5% skim milk powder in PBST (PBS ϩ 0.1% Tween 20) for 2 h. The volume of the blocking solution was reduced to 10 ml, and the primary antibody (mouse antisyntaxin 1A (Stressgen, Victoria, British Columbia, Canada) or rabbit anti-G␤␥ (Calbiochem, La Jolla, CA)) was added at a concentration of 1:1000. After 2 h of incubation, the membrane was washed twice with PBST for 10 min and twice with PBS for 10 min, and subsequently incubated with the secondary antibody (horseradish peroxidase-conjugated goat anti-mouse, or goat anti-rabbit, Amersham Pharmacia Biotech) in 2% skim milk powder in PBST at a concentration of 1:5000 for 2 h. The membrane was again washed twice with PBST for 10 min and twice with PBS for 10 min. Following the final wash, the blot was subjected to chemiluminescence analysis using ECL plus (Amersham Pharmacia Biotech) and detected on film, or via Storm Scan 860.
Preparation of Rat Hippocampal Homogenate-Rat hippocampi were hand homogenized with a Teflon-coated homogenizer in 0.32 M sucrose, 10 mM HEPES KOH (pH 7.0), 1 mM EGTA, 0.1 mM EDTA, 0.5 mM PMSF, protease inhibitor mixture (Roche Molecular Biochemicals), 1 M microcystin, 1 M okadaic acid, and 1 mM sodium orthovanadate (2 ml/hippocampus). The homogenate was centrifuged for 10 min at 500 ϫ g and the supernatant collected and subsequently centrifuged for 20 min at 20,000 ϫ g (4°C). The pellet containing the synaptic proteins was resuspended in 1% Triton X-100, 20 mM MOPS (pH 7.0), 4.5 mM Mg(CH 3 COO) 2 , 150 mM KCl, and 0.5 mM PMSF, protease inhibitor mixture (Roche Molecular Biochemicals), 1 M microcystin, 1 M okadaic acid, 1 mM sodium orthovanadate and incubated for 30 min at 37°C. Following solubilization, large membrane fragments were removed by centrifugation at 1000 ϫ g for 5 min. The resulting supernatant is a crude hippocampal homogenate containing synaptic proteins. Protein concentration was determined by Bio-Rad protein assay using bovine serum albumin as the standard.
Preparation of SNARE Fusion Proteins-Glutathione S-transferase (GST) fusion proteins of syntaxin 1A and VAMP2 were prepared as described previously (49). Briefly, GST fusion proteins encoding the cytoplasmic portions of syntaxin 1A and VAMP2 were constructed in the vector pGEX-KG (50) and expressed in AB1899 strain of E. coli. After induction of expression with 100 M isopropyl-␤-D-thiogalactopyranside for 5 h, the bacteria were suspended in PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 2 mM KH 2 PO 4 ) supplemented with 0.05% Tween 20, 2 mM EDTA, and 0.1% ␤-mercaptoethanol and lysed by two passages through a French press (Spectronic Instruments Inc.). The fusion protein was recovered by binding of the GST domain to glutathione-agarose beads (Sigma). The fusion protein beads were washed extensively and finally resuspended in 0.5% Triton X-100, 20 mM MOPS (pH 7.0), 4.5 mM Mg(CH 3 COO) 2 , 150 mM KCl, and 0.5 mM PMSF.
Immunoblotting for Syntaxin 1A-G Protein Binding-Proteins were transferred electrophoretically at constant voltage from polyacrylamide gels to nitrocellulose (0.2 m) in 20 mM Tris, 150 mM glycine, 12% methanol. Transferred proteins were visualized by staining with Ponceau S. Nitrocellulose membranes were blocked for nonspecific binding using 5% milk, 0.15% Tween 20, PBS solution (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na 2 HPO 4 , 1.4 mM KH 2 PO 4 (pH 7.3)) and incubated overnight with primary antibody (1:1000). The membranes were washed four times in the above milk/Tween 20/PBS solution and incubated for 30 min with goat anti-rabbit or goat anti-mouse IgG-coupled horseradish peroxidase. Antigen was detected using chemiluminescent horseradish peroxidase substrate (ECL, Amersham Pharmacia Biotech). Immunoreactive bands were visualized following exposure of the membranes to Amersham Hyperfilm-MP.
Alternatively, for the synprint-syntaxin 1A-G␤␥ binding assay, bacterial cell lysate of the 6ϫHis-synprint peptide was batch-bound to 50% (v/v) ProBond Ni 2ϩ -agarose beads (Invitrogen) and washed four times with three buffers of pH levels 7.8, 6.0, and 5.5, containing 20 mM Na 3 PO 4 and 500 mM NaCl to remove nonspecifically bound proteins. The beads were then washed and equilibrated with TBST-Ca (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Triton X-100, 15 M CaCl 2 ). Bound 6ϫHis fusion protein from a fraction of the beads was eluted and quantified by a Bradford assay using bovine serum albumin as a standard. The 6ϫHis fusion protein bound to beads was incubated for 2 h at 4°C with 40 g of rat hippocampal homogenate; bound proteins were subsequently eluted and immunoassayed for both G␤␥ and syntaxin 1A. Before loading on SDS-PAGE, protein samples were treated with 2ϫ Laemmli sample buffer, rotated end-over-end for 2 h at 4°C, denatured at 95°C for 1 min, and loaded. Transfer from SDS-PAGE to Hybond ECL nitrocellulose (Amersham Pharmacia Biotech) and subsequent immunodetection was performed as described above.

G Protein Modulation of N-type Channels Occurs
Independently of Syntaxin 1A-It is well established that N-type calcium channels transiently expressed in human embryonic kidney tsA-201 (HEK) cells are subject to potent inhibition by G protein ␤␥ subunits upon either activation of seven helix transmembrane receptors (33), cytoplasmic application of purified G␤␥ (32,34), or via transient overexpression of G␤␥ (38,39). One characteristic feature of this type of inhibition is its reversal following a strong membrane depolarization (termed facilitation). Such an experiment is illustrated in Fig. 1A. In the absence of exogenous G␤␥, application of a strong depolarizing prepulse (ϩ150 mV, 50 ms) to N-type (␣ 1B ϩ ␤ 1b ϩ ␣ 2 -␦) channels has little effect on peak current amplitude, or current waveform. However, upon cotransfection with G␤ 1 ␥ 2 , peak current amplitude is increased by almost 2-fold subsequent to application of the prepulse, thus reflecting relief of the tonic inhibition of the channel induced by G␤ 1 ␥ 2 . The experiments of Stanley and Mirotznik (37) suggested that syntaxin 1A might be a prerequisite for G protein modulation of N-type calcium channels, and if so, this would imply that HEK cells should endogenously express syntaxin 1A. To examine this possibility, we carried out an immunoblot for syntaxin 1A in HEK cells under several different conditions. As seen in Fig. 1B, syntaxin 1A could only be detected when transfected, indicating that syntaxin 1A is not present endogenously in HEK cells. In Fig.  1B, we also see that syntaxin 1A expression is not triggered by overexpression of G␤␥ or secondarily by expression of N-type calcium channels as described recently for P/Q-type calcium channels (51). Under the same experimental conditions, we could not detect any expression of nSec-1 or SNAP-25, even following transfection with syntaxin 1A (data not shown), indicating that HEK cells lack many of the proteins responsible for vesicle release in neurons, and that the exogenous expression of syntaxin 1A does not result in the concomitant expression of other proteins important for vesicle release. Shown in Fig. 1C, syntaxin 1A transfection did not mediate a detectable change in the amount of endogenous G␤␥ present in HEK cells. However, this result needs to be viewed cautiously, because high endogenous G␤␥ levels could mask small increases in exogenous G␤␥, and Western blot analysis does not evaluate FIG. 1. A, whole cell current traces obtained upon transient expression of ␣ 1B ϩ ␤ 1b ϩ ␣ 2 -␦ in tsA-201 cells. Currents were elicited by stepping from a holding potential of Ϫ100 mV to ϩ20 mV. In the absence (left) of exogenous G␤␥, application of a strong depolarizing prepulse (ϩ150 mV, 50 ms) has little effect on current waveform. Upon coexpression of G␤ 1 ␥ 2 (right traces), the channels undergo a large tonic inhibition, which is reversed by the prepulse. B, immunoblot for syntaxin 1A in tsA-201 cells transfected with various exogenous proteins.
, syntaxin 1A (c), G␤ 1 ϩ G␥ 2 (d), EGFP (e) or sham-transfected (f), and immunoblotted for syntaxin 1A. No endogenous syntaxin 1A was detected (lane f), nor was production induced by exogenous expression of ␣ 1B , ␤ 1b , ␣ 2 -␦, G␤ 1 , G␥ 2 , or EGFP (lanes a, b, d, and e). Syntaxin 1A expression was limited only to the batch of cells in which exogenous syntaxin 1A was transiently expressed (lane c). C, immunoblot for G␤␥ in tsA-201 cells transfected with sham (control), syntaxin 1A, and G␤ 1 ␥ 2 . The bands were visualized and analyzed using a Molecular Dynamics Storm Scan 860 system (equipped with Scanner control version 4.0 and ImageQuaNT version 4.2). This method yields a linear signal that allows quantification of the chemiluminescent signal from ECLϩ. After background subtraction, there was no detectable increase in endogenous G␤␥ levels upon transfection with syntaxin 1A, whereas total G␤␥ levels were increased by about 50% upon transfection with exogenous G␤ 1 ␥ 2 .
how much G␤␥ is complexed with G␣ subunits in the cell. Overall, the data shown in Fig. 1 indicate that the presence of syntaxin 1A, or SNAP-25 and nSec-1, is not required for N-type channels to undergo G protein inhibition.
Syntaxin 1A Promotes G Protein Inhibition of N-type Channels-Although syntaxin 1A is not a prerequisite for G protein modulation of N-type channels, it is possible that syntaxin 1A does modulate the effects of G␤␥ on N-type calcium channels. To examine this possibility, we cloned syntaxin 1A from rat brain and investigated its effect on transiently expressed Ntype channels. N-type channels coexpressed with syntaxin 1A underwent a hyperpolarizing shift in steady state inactivation from Ϫ44.2 mV (n ϭ 11) to Ϫ61.9 mV (n ϭ 7), similar to what has been previously reported for N-type channels expressed in Xenopus ocytes (40) and thereby confirming functional syntaxin 1A expression in our experiments. However, as evident upon examination of the current records in Fig. 2A (top left), an additional effect of syntaxin 1A was that the current waveform exhibited dramatically slowed activation and inactivation kinetics, and the average peak current amplitude decreased by nearly 1 order of magnitude. Qualitatively, the current waveform in the presence of syntaxin 1A was reminiscent of that obtained upon coexpression of the channel with exogenous G␤␥ (compare with Fig. 1A), suggesting that syntaxin 1A might secondarily mediate a tonic G protein modulation of the channel. Consistent with this idea, application of a strong depolarizing prepulse resulted in a 2.2 Ϯ 0.2-fold (n ϭ 10) increase in peak current amplitude. The magnitude of the prepulse effect was independent of the type of calcium channel ␤ subunit coexpressed (i.e. ␤ 1b , ␤ 2a , or ␤ 3 ) and of the presence of the ancillary ␣ 2 -␦ complex (data not shown). The prepulse facilita-tion was reduced to a 1.2 Ϯ 0.04-fold (n ϭ 12) enhancement when cells were incubated with BTC1 for 12 h prior to recording, indicating that our observations were due to the presence of the syntaxin 1A protein rather than the preceding transcription events. To ensure that BTC1 did not directly interfere with G protein modulation of the channel, we investigated the effects of syntaxin 1A on N-type channels coexpressed with both syntaxin 1A and G␤ 1 ␥ 2 . As seen from the current records in Fig. 2A (bottom left), the effects of syntaxin 1A and G␤ 1 ␥ 2 were not additive, and more importantly, BTC1 was ineffective in removing the tonic G protein inhibition (bottom right). This is also reflected in Fig. 2B in form of bar graphs. Coexpression of the N-type channels with syntaxin 1A or G␤ 1 ␥ 2 plus syntaxin 1A resulted in a degree of G protein inhibition that did not differ significantly from that obtained upon coexpression of G␤ 1 ␥ 2 alone, and BTC1 selectively removed the tonic G protein inhibition induced by syntaxin 1A overexpression. Overall, these data are consistent with a mechanism by which syntaxin 1A promotes tonic G protein modulation of N-type calcium channels.
Syntaxin 1A Physically Interacts with G␤␥-One possible explanation for our observations is a mechanism by which syntaxin 1A binding to the channel increases the sensitivity of the channel to free G␤␥ subunits, i.e. those that are not part of the G␣␤␥ trimers associated with seven helix transmembrane receptors. In this scenario, the cleavage of the syntaxin 1A protein would attenuate the enhancing effect, and the overexpression of exogenous G␤␥ would mask the effect. In principle, there are two possible mechanisms by which this could occur. Binding of syntaxin 1A to the synprint site on the channel (19) could allosterically enhance G protein binding to the channel. Alternatively, syntaxin 1A might serve as an anchoring mechanism by which G protein ␤␥ subunits are confined to the vicinity of their site of action on the channel protein. To discriminate between these two possibilities, we carried out a binding assay involving a GST-syntaxin 1A fusion protein. As shown in Fig. 3A, upon incubation with rat hippocampal homogenate, the syntaxin 1A fusion proteins were able to precipitate G protein ␤␥ subunits in a syntaxin 1A concentration-dependent manner. Similarly, increasing amounts of nSec-1, a protein known to tightly associate with syntaxin 1A, were precipitated from the homogenate as the GST-syntaxin 1A concentration was increased. On the other hand, GST-VAMP-2 beads were unable to associate with either of G␤␥ or nSec-1, indicating that nonspecific binding of G␤␥ to the GST beads did not occur. While these data are consistent with a direct interaction between syntaxin 1A and the G protein, this experiment does not permit us to rule out the possibility that G␤␥ interacts with syntaxin 1A indirectly via one or more additional proteins. To investigate this possibility, we examined the ability of GSTsyntaxin 1A to interact with purified G␤␥ subunits in the absence of other proteins. As shown in Fig. 3B, the recombinant GST-syntaxin 1A was able to bind the purified G␤␥ subunits, indicating that there is indeed a direct physical interaction between G␤␥ and syntaxin 1A, independent of any intermediary proteins. In summary, our data suggest a mechanism by which syntaxin 1A physically binds to G␤␥, bringing G␤␥ into close vicinity of its target site on N-type calcium channel ␣ 1 subunits and thereby optimizing G␤␥ modulation of channel activity.
If our hypothesis is correct, then syntaxin 1A should be able to concomitantly bind to the N-type channel and to G␤␥. To confirm this, we carried out an assay examining the interactions of syntaxin and G␤␥ with fusion proteins directed against the synprint site. As seen in Fig. 3C, the 6ϫHis-synprint site bound syntaxin 1A from rat hippocampal homogenate, but FIG. 2. Effect of syntaxin 1A expression on transiently expressed N-type channels. A, current records obtained with ␣ 1B ϩ ␤ 1b ϩ ␣ 2 -␦ as described in Fig. 1. Expression of syntaxin 1A mediates a tonic G protein inhibition of the channels as defined by the prepulse relief. The effect of syntaxin 1A is removed upon cleavage with botulinum toxin C1 (12-h incubation with 1 g of BTC1/ml of DMEM). Overexpression of exogenous G protein ␤␥ subunits diminishes the effect of the toxin. B, bar graphs reflecting the degree of peak current amplitude enhancement following a strong depolarizing prepulse as a measure of tonic G protein inhibition. The values reflected in bars 2, 3, 5, and 6 did not differ significantly from each other (p Ͼ 0.6). more importantly, G␤␥ was also detected and reflected the increase in syntaxin 1A binding as the amount of homogenate was increased. We have previously shown that G␤␥ does not bind directly to the N-type channel domain II-III linker (34) consistent with the idea that the G␤␥ detected in Fig. 3C is coupled to the synprint site via syntaxin. Overall, these data further support the idea that syntaxin 1A may mediate a physical role in enhancing tonic G protein modulation of N-type calcium channels, and it is likely that distinct regions on the syntaxin 1A molecule interact with the calcium channel and G␤␥.

DISCUSSION
It has been known for almost two decades that N-type calcium channels are subject to potent inhibition upon activation of seven helix transmembrane receptors (27,30,41,42). Recently, it has been shown that this inhibition is due to direct binding of G protein ␤␥ subunits to the N-type channel domain I-II linker (34,35,38,39), although several other regions of the channel have also been implicated (43)(44)(45). The key characteristics of this type of inhibition are its reversal upon application of strong depolarizing prepulses, as well as an apparent slowing of activation and inactivation kinetics, which can be attributed to an increase in first latency to opening (30 -32). G protein inhibition of N-type channels is subject to modulation by a number of factors, including the calcium channel ␤ subunit (22,46) and protein kinase C (24, 25, 33, 34). More recently, Stanley and Mirotznik (37) suggested that G protein modula-tion of calcium channels in chick ciliary ganglion might require the presence of syntaxin 1A; however, the underlying molecular mechanisms for this observation remained unknown. Here, we have presented novel evidence that syntaxin 1A may serve to optimize tonic G protein modulation of N-type calcium channels, but is not required.
Syntaxin 1A does not appear to be a requirement for G protein modulation of N-type calcium channels transiently expressed in tsA-201 cells, as these cells lack endogenous syntaxin 1A, and yet, the channels are subject to G protein inhibition either upon transient overexpression of G␤␥, or upon activation of endogenous somatostatin receptors (33). Nonetheless, in our system, syntaxin 1A appeared to modulate the G protein-mediated inhibition of the channel. In the absence of exogenous G protein ␤␥ subunit, the transient expression of syntaxin 1A affected N-type channels in two ways. First, consistent with previous observations by Bezprozvanny et al. (40), syntaxin 1A mediated an 18-mV negative shift in half-inactivation potential, which is likely mediated by direct binding of syntaxin 1A to the synprint site of the channel (19). Second, upon coexpression with syntaxin 1A, the current waveform exhibited slowed activation and inactivation kinetics, and the peak current amplitude could be increased 2-fold upon application of depolarizing prepulse. While it is possible that these effects are directly due to simple binding of syntaxin 1A to the channel, the notion that such effects were not observed in the Xenopus oocyte expression system (40), and the lack of additivity of the effects of exogenously expressed G␤␥ and syntaxin 1A would argue against a direct syntaxin 1A effect. Instead, the observed effects exhibited the key characteristics of G␤␥-mediated inhibition of N-type channels, suggesting that syntaxin 1A expression more likely results in a tonic inhibition of the channel by G␤␥ subunits in addition to the previously reported effects of syntaxin 1A on steady state inactivation (40).
In principle, the expression of syntaxin 1A could boost the endogenous levels of free G␤␥, thereby resulting in a tonic level of G protein inhibition. We could not observe such an increase in G␤␥ levels upon transfection of tsA-201 cells with syntaxin 1A. However, we have no way of knowing if syntaxin 1A expression alters the relative proportion between endogenous G␣ and G␤␥ subunits, thereby perhaps mediating an increase in free G␤␥ that would be able to inhibit the channel. Furthermore, despite the notion that exogenous expression of G␤ 1 ␥ 2 mediated a potent inhibition of N-type currents, we only detected a ϳ50% increase in total G␤␥ levels following transfection with G␤ 1 ␥ 2 , and thus it is possible that we may not be able to resolve any putative syntaxin 1A-mediated increases in free endogenous G␤␥. A more convincing argument can be based on our observation with botulinum toxin C1, which cleaves the syntaxin 1A protein. If syntaxin 1A were to mediate an upregulation of endogenous G␤␥ levels, the increased levels should be still be maintained after the 12-h incubation with BTC1 (particularly because syntaxin 1A is continuously being expressed during that period); yet, after application of BTC1, the tonic G protein inhibition became dramatically attenuated. Together with our observations that syntaxin 1A and G␤␥ have the propensity to form a physical complex, we favor a mechanism by which syntaxin 1A facilitates coupling of G␤␥ to the channel protein.
Overall, our data are consistent with a mechanism by which syntaxin 1A binding to the channel enhances its susceptibility to G protein modulation. While we cannot completely rule out the possibility that syntaxin 1A binding allosterically increases the affinity of the channel for G␤␥, the existence of a physical interaction between G␤␥ and syntaxin 1A leads us to favor a mechanism in which syntaxin 1A mediates a co-localization of FIG. 3. A, binding of hippocampal G␤ and nSec-1 to recombinant syntaxin 1A and VAMP2-GST fusion proteins immobilized on glutathione-agarose. Hippocampal homogenates (160 g) were incubated with syntaxin 1A-GST or VAMP2-GST fusion proteins immobilized on glutathione-agarose as indicated. The beads were washed and bound proteins were eluted in sample buffer, fractionated electrophoretically on a 12% SDS-PAGE gel, transferred to nitrocellulose, stained with Ponceau S, and then processed by Western blotting. The lanes are as follows: G protein ␤␥ subunits and the N-type calcium channel. Fig. 4 illustrates how this might occur. Syntaxin 1A is known to bind at the synprint site of the N-type channel domain II-III linker (19,20). The interaction of G␤␥ subunits on the N-type channel ␣ 1 subunit occurs at the domain I-II linker (33,34,35,38) as well as the carboxyl-terminal region (43,45). We hypothesize that G␤␥ subunits physically couple to the syntaxin 1A/synprint complex, thereby encouraging a more effective interaction between G␤␥ and its target region on the channel. Via this mechanism, the endogenous levels of free G␤␥ would become effective in modulating channel activity when syntaxin 1A is present. BTC1 treatment results in the loss of this effect due to cleavage and removal of the syntaxin 1A protein. Upon exogenous expression of G␤␥ or by activation of seven helix transmembrane receptors, the concentration of free G␤␥ would become sufficiently high to modulate the channel independently of the enhancing effect of syntaxin 1A, thus rendering BTC1 ineffective in removing G protein inhibition. The observation that recombinant syntaxin 1A was able to interact with G␤␥ from rat hippocampal homogenate, together with the observations of Stanley and Mirotznik (37) in chick ciliary ganglion calyces suggests that this effect is not limited to our expression system, but rather, is a physiological feature of neurons. The regulation of neurotransmission by voltage-dependent calcium channels is a highly dynamic process which is modulated by complex interactions between cytoplasmic messenger molecules, SNARE proteins, and multiple types of voltage-dependent calcium channels (47). For example, G protein inhibition of N-type calcium channels due to the activation of seven helix transmembrane receptors inhibits synaptic transmission, whereas protein kinase C-dependent phosphorylation of the N-type channel ␣ 1 subunit increases channel activity, and antagonizes G protein and syntaxin 1A binding to the channel (20,33,34). Syntaxin 1A binding to the channel is critically dependent on the amount of free calcium (48), and results in inhibition of channel activity by shifting the steady state inactivation curve toward more negative potentials (Ref. 40; see also "Results"). In addition, the two major calcium channel subtypes found at presynaptic nerve terminals (i.e. P/Q-type and N-type channels) are differentially modulated by G proteins and protein kinase C (22,23,34). More recently it has been reported that entry of calcium ions through certain (P/Qtype) ␣ 1A calcium channel isoforms triggers the expression of syntaxin 1A via calcium-dependent gene transcription (51). In that context, the notion of a syntaxin 1A induced tonic G protein inhibition of N-type calcium channels could perhaps provide a novel mechanism by which expression of certain P/Q-type splice variants could regulate the activity of N-type channels and thereby contribute to the fine tuning of neurotransmission.
FIG. 4. Schematic representation of a possible model explaining the role of syntaxin 1A in mediating tonic G␤␥ inhibition of N-type calcium channels. The calcium channel ␣ 1 and ␤ subunits are depicted, respectively, in light blue and purple; syntaxin 1A (partially inserted into the plasma membrane) is indicated in orange; and G␤ and G␥ (partially inserted into the plasma membrane) are depicted, respectively, in dark and light green. Syntaxin 1A binds to the synprint motif of the cytoplasmic region connecting domains II and III of the N-type channel ␣ 1 subunit. G␤␥ modulates N-type channel activity by binding to the cytoplasmic linker between domains I and II at a region that partially overlaps with the calcium channel ␤ subunit interaction site. Syntaxin 1A is hypothesized to facilitate G protein modulation of the channel by aiding the targeting of G␤␥ to its binding site on the channel.