HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ilouz, N. Articles by Linial, M. Search for Related Content PubMed PubMed Citation Articles by Ilouz, N. Articles by Linial, M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 41, 29519-29528, October 8, 1999

Depolarization Affects the Binding Properties of Muscarinic Acetylcholine Receptors and Their Interaction with Proteins of the Exocytic Apparatus^*

Nili Ilouz, Leora Branski, Julia Parnis, Hanna Parnas^§¶, and Michal Linial^§

From the  Department of Biological Chemistry and the ^§ Otto Loewi Center for Molecular and Cellular Neurobiology, Alexander Silberman Institute of Life Sciences, The Hebrew University, Jerusalem 91904, Israel

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Membrane depolarization is the signal that triggers release of neurotransmitter from nerve terminals. As a result of depolarization, voltage-dependent Ca²⁺ channels open, level of intracellular Ca²⁺ increases. and release of neurotransmitter commences. Previous study had shown that in rat brain synaptosomes, muscarinic acetylcholine (ACh) receptors (mAChRs) interact with soluble NSF attachment protein receptor proteins of the exocytic machinery in a voltage-dependent manner. It was suggested that this interaction might control the rapid, synchronous release of acetylcholine. The present study investigates the mechanism for such a voltage-dependent interaction. Here we show that depolarization shifts mAChRs, specifically the m2 receptor subtype, to a low affinity state toward its agonists. At resting potential, mAChRs are in a high affinity state (K_d of ~20 nM) and they shift to a low affinity state (K_d of tens of µM) upon membrane depolarization. In addition, interaction between m2 receptor subtype and the exocytic machinery increases with receptor occupancy. Both phenomena are independent of Ca²⁺ influx. We propose that these results may explain control of ACh release from nerve terminals. At resting potential the exocytic machinery is clamped due to its interaction with the occupied mAChR and depolarization relieves this interaction. This, together with Ca²⁺ influx, enables release of ACh to commence.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Neurotransmitter release from nerve terminals is a unique instance of a secretory process where external stimuli must be handled with extreme speed and accuracy. Many proteins have been implicated to be involved in neurotransmitter release, but only a few have been found to be essential (1-4). The minimal essential core common to all secretory processes consists of the SNARE¹ (SNAP receptors) proteins (5): VAMP, syntaxin, and SNAP-25. In nerve terminals the essential core contains also voltage-dependent Ca²⁺ channels and synaptotagmin (reviewed in Refs. 1, 2, 6, and 7). This set of proteins comprises the exocytic apparatus, henceforth called ExA. The ExA proteins have a combinatorial large repertoire of interactions that dynamically change throughout the process of neurotransmitter release (1, 6-8). Such interactions are modulated by a variety of ions (9, 10), second messengers and signaling cascades (11-16), and membrane potential (17).

Despite the overwhelming amount of accumulated data, a major challenge still remains to explain how the depolarization signal is transduced into molecular interactions that eventually lead to a rapid synchronous release of neurotransmitter.

We have recently addressed this question and identified a group of membrane receptors not previously considered part of the control machinery of neurotransmitter release. We showed that, in rat brain synaptosomes, at Ca²⁺-depleted solution, presynaptic muscarinic ACh receptors (mAChRs) directly interact with the ExA, specifically with syntaxin and SNAP-25, and that this interaction is voltage-dependent (18). A related, large body of work has documented cross-talk between presynaptic G-protein-coupled receptors, including mAChR, and Ca²⁺ channels (e.g. Refs. 19-27). The observed role of these receptors is in reducing Ca²⁺ influx via an activation of their associated G-proteins. These studies do not consider, however, a physical interaction between the receptor and the voltage-dependent channel. Consequently, these studies concern the receptor's role in modulating release by determining the Ca²⁺ levels. The findings of Linial et al. (18) suggest an additional role to mAChRs, namely that they may serve as voltage-dependent gatekeepers of ACh release in nerve terminals. At resting potential, the release apparatus is clamped due to its tight interaction with mAChR. Upon depolarization, presumably in parallel to the influx of Ca²⁺, the clamp is removed and release may commence.

The present study explores a mechanism that may underlie the voltage-dependent interaction between mAChRs and the ExA. Experiments were performed on rat brain synaptosomes in Ca²⁺-free solution and the effects of depolarization per se could thus be discerned. We show that depolarization shifts mAChRs, specifically the m2 subtype, from a high to a low affinity state toward its agonists. We also found that interaction of the m2 subtype receptor with the SNARE protein syntaxin increases with receptor occupancy. We suggest that the depolarization-induced shift in mAChR affinity affects its interaction with ExA. Relief from this interaction, together with Ca²⁺ influx, evokes release of ACh from nerve terminals.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Immunoprecipitation and Immunoblot Analysis-- Fresh synaptosomes were prepared from rat brains (28) and were used within 3 h of preparation. In all experiments the physiological state of the synaptosomes was monitored by a glutamate release assay as described previously (18). Immunoprecipitation (IP) was performed as described (18). Briefly, antibodies were prebound to Protein G-Sepharose or Protein A-Sepharose beads (Zymed Laboratories Inc.) in HKA buffer (50 mM Hepes-KOH, pH 7.4, 140 mM potassium acetate, 1 mM MgCl₂, and 0.1 mM EGTA), supplemented with 0.1% gelatin and 0.1% bovine serum albumin. Aliquots of synaptosomes (150 µg) were incubated for 30 min at 25 °C in Ca²⁺-free physiological buffer (BSS; 10 mM Hepes/NaOH, pH 7.4, 128 mM NaCl, 2.4 mM KCl, 1.2 mM MgCl₂, 1.2 mM KH₂PO₄, 10 mM D-glucose). For depolarization, NaCl was replaced by KCl (BSS-depolarization buffer). Synaptosomes were gently washed (twice), and solubilization was performed for 1 h (4 °C) in IP buffer containing HKA buffer with the respective levels of KCl and 2% freshly prepared CHAPS (Roche Molecular Biochemicals). Protease inhibitors (10 µg/ml aprotonin, leupeptin, and pepstatin; Roche Molecular Biochemicals) and 10 mM AEBSF (Calbiochem) were added to the IP buffer. Following overnight incubation of the prebound beads (4 °C) with solubilized synaptosomes, the bound proteins were thoroughly washed (IP buffer with only 0.2% CHAPS), separated by SDS-PAGE, and Western blot analysis was performed using the ECL detection system (Amersham Pharmacia Biotech). Special precautions were taken to avoid unspecific interactions of mAChR with syntaxin. Syntaxin is prone to artificial interactions due to its strong adhesion to Protein A/G-Sepharose beads. Including gelatin during the experiment and 5% glycerol during a final wash step minimizes syntaxin's adhesion. The intensity of nonspecific immunoreactive signal for syntaxin on Protein G-Sepharose did not exceed 5% of the signal obtained by including the relevant antibody. The amounts of m2-syntaxin complex were insensitive to varying protein concentrations during the IP experiment (range from 1.5 to 0.1 mg/ml). Most IP experiments were done at a concentration of 0.15 mg/ml. [³H]quinuclidinyl benzilate ([³H]QNB), 45 Ci/mmol (NEN Life Science Products) at 2 nM was incubated (25 °C) with the synaptosomes for 45 min prior to solubilization and throughout the rest of the procedure. The level of [³H]QNB that had been immunoprecipitated with the antibodies was identical irrespective of KCl concentration (in the range of 5-120 mM KCl). After overnight incubation, bound proteins were filtered through GF/C filters using washing buffer and radioactivity was monitored in a scintillation counter. Acetyl cholinesterase (AChE, Sigma), muscarine (Sigma), methoctramine, and pirenzepine (RBI) were incubated with the synaptosomes for 30 min (25 °C) before and during the depolarization step, and were also maintained (excluding AChE) throughout membrane solubilization. The antibodies used to detect proteins of the ExA are: polyclonal anti-syntaxin 1A (Alomone, Israel), monoclonal anti-HPC-1 (Sigma), polyclonal antibodies against SNAP-25 and VAMP-2 (kindly provided by M. Takahashi), monoclonal anti-synaptotagmin I (Signal Transduction), and anti-synaptophysin (Roche Molecular Biochemicals). Secondary antibodies were goat anti-rabbit or goat anti-mouse peroxidase (Jackson Laboratories). Protein concentration was measured by the Bicinchoninic acid assay (BCA, Pierce). Immunoreactivity was measured by densitometry of a luminescence reaction within the linear range of the reaction. Data were analyzed using a National Institutes of Health imaging program (NIH Image, version 1.59). All IP experiments were repeated in Ca²⁺-depleted buffer (2 mM EGTA) with no measurable differences to the results obtained in Ca²⁺-free buffers.

Antibody Specificity Assays-- For mAChRs, we used monoclonal m2 antibody (Chemicon) and polyclonal m1-m5 antibodies (kindly provided by E. Heldman, Nes-Ziona, Israel). The titer of the antibodies was determined by enzyme-linked immunosorbent assay using predetermined amounts of mAChR antigens obtained from CHO-K1-transfected cells. Cells were grown in Dulbecco's modified Eagle's medium (Beit-hemeek, Israel) buffer containing 10% fetal calf serum and 400 ng/ml neomycin (Promega). Cells were harvested, and membranes were prepared (2 mg/ml) for binding assays (see below). The concentration of each antibody used for immunoprecipitation was optimized for efficient precipitation (>70% efficiency). The level of nonspecific binding, as determined by including 1 µM atropine in the immunoprecipitation, did not exceed 10%. For blocking experiments, purified GST fusion proteins containing the i3 loop of m1 and m2 receptors were prepared and purified as described previously (29). These peptides (20 µg/ml) were included in the immunoprecipitation experiments. Membranes prebound with 2 nM of N-[methyl-³H]scopolamine ([³H]NMS, 84 Ci/mmol, 1 mCi/ml; Amersham Pharmacia Biotech) were treated with either GTPS (Sigma) or GDPS (Sigma) at concentrations of 10-100 µM. Immunoprecipitation was performed to quantify the efficiency of the different antibodies to precipitated receptors coupled or uncoupled to their associated G-proteins.

Syntaxin recombinant peptide was prepared from a GST fusion construct (amino acids 4-267). The GST fusion protein was purified on glutathione-agarose beads (Sigma). Syntaxin was purified following thrombin (1:500 molar ratio) cleavage, and the amount of the peptide was quantified by micro-BCA procedure (Pierce).

Agonists and Antagonist Binding Assays-- Binding of radiolabeled agonists was performed as described (30), using [³H]ACh (20 Ci/mmol, 90% ethanol; ARC Inc.). Briefly, small aliquots of labeled ACh were kept at 70 °C. For each experiment, ACh was dried under N₂ flow and resuspended in BSS buffer. We used both BSS and BSS-Ca²⁺-depleted (BSS containing 2 mM EGTA and 3.4 mM MgCl₂) in binding experiments. In BSS buffer the synaptosomes were found to be physiologically stable for longer periods and a smaller variability among data points was observed. Synaptosomes were first washed, to eliminate contamination by endogenous ACh, in 10 volumes of physiological BSS buffer and then depolarized for 30 min at 25 °C in BSS-depolarization buffer. AChE inhibitor (phospholine iodide, 46 µM) was added to all reaction mixtures. Specified concentration of [³H]ACh was added to the depolarized synaptosomes in siliconized glass tubes (to a total volume of 40-200 µl) and incubated with gentle shaking (45 min, 25 °C). Synaptosomes were subsequently filtered through GF/C filters, washed four times with ice-cold depolarization buffer (within less than 30 s, 2 ml each). Extended washing procedure resulted in a marked reduction in the cpm values. Filters were dried, and radioactivity was determined using scintillation counter. Identical experiments were performed on lysed synaptosomes. Lysis was achieved by physical shearing (using Polytron) in cold, hypotonic buffer (5 mM Tris, 0.1 mM EDTA, pH 8.0). To ensure complete lysis, the procedure was repeated three times, and, between lysis, membranes were collected by centrifugation (100,000 × g, TLA 100.2, Beckman). All binding determinations were done in duplicates, with less than 10% variation between the duplicates. Nonspecific binding was determined in the presence of 1 µM freshly prepared atropine. To construct the nonspecific curve, we fitted the data points (each in duplicate) at concentration of ACh up to 40 nM and extrapolated it to the entire range of ACh concentration. In experiments where mAChR antagonists (pirenzepine, methoctramine) or Ca²⁺ channel blockers (calcicludine, -conotoxin GVIA, and -agatoxin IVA; Alomone Laboratories) were included, they were incubated with the synaptosomes in BSS buffer (30 min) prior and throughout depolarization or prior to the addition of agonist. Binding of the radiolabeled antagonist [³H]NMS (84 Ci/mmol, 1 mCi/ml; Amersham Pharmacia Biotech) to synaptosomes was performed as described above, but [³H]NMS concentrations were 0.1-20 nM and binding was performed in BSS buffer for 45 min at 25 °C. For accurate measurements of [³H]ACh and [³H]NMS, an aliquot from each experimental point was withdrawn for scintillation counting and the actual concentration was determined. Binding curves were fitted assuming a Michaelis-Menten equation with a single binding site. All binding experiments were repeated in BSS-Ca²⁺-depleted buffers with no measurable differences in results in comparison to BSS-Ca²⁺-free buffers.

Displacement Experiments-- [³H]NMS (0.6-1 nM) was incubated with 100 µg of protein from fresh synaptosomes following depolarization at 90 mM KCl (45 min at 25 °C) and was displaced by the presence of unlabeled agonists (muscarine, carbachol, or ACh; 10¹⁰ to 10² M). ACh stock solutions were prepared freshly before the experiments to minimize ACh decomposition during storage. Samples were filtered and counted as in the binding assays. Total number of binding sites were calculated at each experiment from the [³H]NMS counts obtained prior to the addition of the agonists. Displacement experiments in the presence of Veratridine (Calbiochem) were performed by preincubation of the drug for 10 min before adding [³H]NMS or the different concentrations of unlabeled agonists. Displacement of [³H]oxotremorine (7-10 nM) with unlabeled antagonists was performed similarly but the experiments were performed in the dark (due to the sensitivity of oxotremorine to light). Fitting of the data was accomplished using competitive inhibition equation with two affinity states and a K_d for [³H]NMS of 0.2 nM (31). The value of the high affinity K_d for carbachol was fixed to 100 nM. This value is based on the corresponding K_d for ACh (~20 nM, see Fig. 3) and the affinity for carbachol, which is 5.7-fold lower (32, 33). Data analysis was done using GraphPad Prism 1.03 software package. Statistical significance was determined by Student's t test comparing two means.

Determining Membrane Potential-- For each experiment, the membrane potential was determined by the level of accumulation of [³H]TPP⁺ (tetraphenylphosphonium bromide, 31.4 Ci/mmol, ethanol, NEN Life Science Products) within the synaptosomes, as described previously (34). Briefly, synaptosomes (at a concentration of 5 mg/ml) were incubated at 25 °C, 20 min with gentle shaking in BSS-depolarization buffer, in the presence of 0.3 µM [³H]TPP⁺. Each sample (in duplicates) was filtered through a GF/C filter, and identical samples were filtered while being washed four times with ice-cold water, in order to lyse the synaptosomes. The values of the lysed synaptosomes were taken as the nonspecific background and subtracted from the experimental data. Values for specific binding at 5-120 mM KCl ranged between 85,000-110,000 and 12,000-15,000 cpm, respectively. The membrane potential was calculated according to a Nernst equation referring to 150 mM KCl (assumed to reflect the intracellular [K⁺] in mammals) as zero membrane potential. Resting potential was defined in 5 mM KCl (rather than at the physiological level of 2.4 mM KCl), as we found that the synaptosomes could not maintain a fixed potential for long periods (>2 h) at KCl concentration below 5 mM. The calculated resting potential we obtained for most of our synaptosomal preparations ranged between 70 and 60 mV. Preparations whose initial membrane potential was below 55 mV were not used. Each one of the added compounds (e.g. channel blockers, mAChRs agonists, and antagonists) were checked for their effect on membrane potential.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Muscarinic ACh Receptor of Subtype m2 and m4 Interact with SNARE Proteins-- Five different subtypes of mAChRs, referred to as m1-m5 (35-37), were identified. These subtypes are localized in distinct brain areas and are distributed differently at pre- and postsynaptic regions (38, 39). The cumulative experimental data suggest that the m1 receptor is the major postsynaptic muscarinic receptor and that m2/m4 predominate presynaptically and serve as autoreceptors in the cholinergic pathway (29, 40). In order to discern the types of mAChRs involved in the interaction with the ExA, we used a panel of mAChRs subtype-selective antibodies (41). The efficiency of the antibodies in immunoprecipitation was determined as follows; CHO-K1 cells transfected with individual receptors (m1-m5) were used to determine the specificity of each antibody. Total amounts of muscarinic receptor binding sites were determined by performing saturation binding curves using [³H]QNB, and the B_max value for each cell line was extracted. Immunoprecipitation experiments from all five CHO-K1-transfected cells using five antibodies (raised against the i3 loop of each of the receptors; Ref. 41) were performed on solubilized membranes prelabeled by [³H]QNB. The efficiency of immunoprecipitation of the solubilized receptors was high for all five receptors. Antibodies against m1 and m4 precipitated about 75% of the sites, and immunoprecipitation efficiency of the other receptors (m2, m3, and m5) with their appropriate antibodies reached 95% (Fig. 1A). Specificity of the antibodies is further demonstrated by the absence of cross-reactivity between the m1 and m2 receptor subtypes (Fig. 1A).

View larger version (34K): [in this window] [in a new window]   Fig. 1.   m2/m4 subtype receptors interact with proteins of the exocytic release apparatus. A, the efficiency of antibody-specific subtypes in rat cortical synaptosomes is determined by quantification of bound [3H]QNB. CHO-K1 subclones stable transfected with m1-m5 receptors were used and maximal binding of [3H]QNB was referred to 100%. Bmax values are as follows (in pmol/mg membranes): m1, 2.0, m3, 1.7; m2, 0.6; m4, 1.2; m5, ~0.8. Polyclonal antibodies against m1-m5 mAChR subtypes were used for IP. N.S. refers to an identical amount of rabbit IgG used as a control. Specificity is determined by the lack of cross-reactivity in using m1 and m2 antibodies in IP with the non-related cell membranes. Values for nonspecific binding are determined by including atropine (1 µM) during IP and are subtracted from each experiment. All antibodies used immunoprecipitated their appropriate receptors with efficiency ranging from 75% to 95%. Average results of three experiments are presented. B, proteins co-precipitated by m1-m5-specific antibodies were separated by SDS-PAGE, blotted, and detected by the antibodies indicated: synp, synaptophysin (SY38), syx, syntaxin (anti-HPC-1); sn25, SNAP-25. Total, 5 µg of solubilized synaptosomes. Combinations of antibodies included in the IP experiments are indicated. C, a Coomassie Blue staining of GST-i3-m2 fusion protein (lane 1) and following cleavage by thrombin (lane 2). A Western blot of IP experiments without (lane 3) and with (lane 4) the addition of the purified i3 -m2 peptide (20 µg/ml). Addition of the purified peptide largely reduced syntaxin immunoreactivity (compare lanes 3 and 4).

In our previous study, we demonstrated a specific interaction between mAChRs and various proteins of the exocytic apparatus by using a membrane-permeable cross-linking reagent (dithiobis-succinimidyl propionate) in intact synaptosomes. We demonstrated that protein complexes of mAChRs and syntaxin are present in situ, before solubilization (18). We wished to determine which of the various mAChR subtypes participate in such interactions. To this end, we used immunoblot analysis to monitor syntaxin co-precipitating with the various mAChR subtype-specific antibodies. Fig. 1B shows that over 85% of syntaxin interacting with mAChRs co-precipitates with m2 and m4 receptor subtypes, and it is the m2 subtype that predominates in this interaction (Fig. 1B, bottom). The relative amounts of mAChR receptors in the synaptosomal preparation were evaluated by pharmacological and immunological means. Taking into account the efficiency of the immunoprecipitation (as seen in Fig. 1A), we estimated m1 (35%) and m2 and m4 (about 30% each) to account for the majority of mAChRs in rat brain synaptosomes. Both m3 and m5 are present in relatively minute quantities. Accordingly, we concluded that it is mainly the m2 (and to a lesser extent m4) which is responsible for the interaction with syntaxin. The prevalence of the m2 (and m4) receptors in the immunoprecipitation was also demonstrated for the other proteins of the ExA (Fig. 1B and data not shown).

In a previous study (18), we used a mixture of mAChR antibodies to show that the interaction of mAChRs and syntaxin (and also SNAP-25) is voltage-dependent. In the present study, therefore, we examine whether the interaction of the m2 subtype, by itself, with syntaxin and with other proteins of the ExA is voltage-dependent. To this end we have used a monoclonal antibody raised against the i3 loop of m2 receptor that was optimized for quantitative immunoprecipitation as determined by using m2-CHO-K1-transfected cells. Table I shows that the antibody is specific to m2 receptor subtype and no cross-reactivity is detected with other mAChR subtypes. Addition of excess m2-i3 expressed peptide to the reaction competed efficiently with the binding of the antibody to m2 subtype receptor while addition of saturated amounts of expressed m1-i3 peptide had a negligible effect on the amount of immunoprecipitated receptor. In addition, supplementing GTPS or GDPS to the reaction does not affect the immunoprecipitation efficiency indicating that the antibody is insensitive to the level of association of the receptor with G-proteins (Table I). We thus conclude that the antibody is suitable for quantitative immunoprecipitation experiments under varying experimental conditions. We have used this m2 antibody to reconfirm the specific interaction of m2 receptor with syntaxin. As seen, addition of m2-i3 expressed peptide greatly reduced the amount of co-immunoprecipitated syntaxin with m2 receptor (Fig. 1C).

To discern the specific role of membrane potential on m2-syntaxin interaction, we performed the following experiments with fresh synaptosomes depolarized to various levels (by exposure to varying concentrations of KCl) in Ca²⁺-free solutions. Fig. 2A shows that maximal binding between the m2 subtype and syntaxin 1A/1B occurs at 5 mM KCl (referred to as resting potential, see "Experimental Procedures") and that interaction is gradually weakened as depolarization rises (30 and 60 mM KCl). We determined the amounts of the m2 receptor immunoprecipitated in each experiment by measuring the [³H]NMS following IP (Fig. 2B). As seen, identical amounts of the receptor were measured in all depolarization levels. These results are consistent with the identical level m2 receptor antibodies included in each experiment (Fig. 2A). To ensure that the changes in the strength of the interaction between m2 receptor and syntaxin are not a result of the solubilization step and the duration of the IP procedure, we included exogenous syntaxin (overexpressed without its carboxyl-terminal transmembrane domain, amino acids 4-267) during depolarization and throughout the procedure. Fig. 2C shows that the endogenous syntaxin is not detected in the IP experiments following depolarization. The results with or without the addition of exogenous recombinant syntaxin were indistinguishable. We thus conclude that the results in Fig. 2 reflect changes in protein-protein interactions that occur due to varying the physiological conditions. Similar depolarization-dependent interaction with m2 receptor was found for SNAP-25 (data not shown). Other presynaptic proteins such as SV2, n-Sec1, and synaptophysin were not detected at any level of depolarization (Fig. 2 and Ref. 18). Identical results (as in Fig. 2A) were obtained when experiments were performed in the presence of a Ca²⁺ chelator.

View larger version (36K): [in this window] [in a new window]   Fig. 2.   Depolarization-dependent interaction between the m2 receptor subtype and syntaxin. A, proteins co-precipitated by m2 monoclonal antibody, separated by SDS-PAGE, blotted, and detected by syntaxin antibodies (anti HPC-1). Immunoreactivity of the antibodies (heavy chain, HC) which were used in the IP experiments presents an internal control. The level of KCl (in mM) in the depolarization solution is indicated (5-60). Note that synaptophysin is not precipitated with the m2 antibody at any level of depolarization. Each IP experiment was repeated three times. The experiments shown are typical. B, IP experiments were done as in A. Reactions were split and incubation was carried out in the presence of 2 mM [3H]NMS throughout the IP. At the end, in parallel to separating the results on SDS-PAGE, the reaction was filtrated and radioactivity was determined. The data are average of triplicates. The data between experiments were normalized for the values obtained at resting conditions. Nonspecific binding refers to the IP reaction (at 5 mM KCl) in the presence of 1 µM atropine. C, in an experiment similar to that shown in A, purified recombinant syntaxin 1A (amino acids 4-267, 2.5 µg) was added before solubilization and throughout the experiments at 5 and 60 mM KCl conditions. The recombinant syntaxin (30 kDa) was not incorporated in the protein complexes following IP with m2 mAChR antibody.

Membrane Potential Affects Binding Properties of mAChRs-- Having found the interaction of mAChRs with the exocytic apparatus to be dependent on membrane potential, we considered the possibility that membrane potential affects the binding properties of mAChRs. To explore this possibility, we used fresh synaptosomes to determine [³H]ACh (in the range of 5-150 nM) binding to mAChRs at different levels of depolarization. All binding experiments were done in Ca²⁺-free solutions, under conditions similar to the immunoprecipitation experiments. Under such conditions no transmitter release was encountered above basal level even at a high depolarization. Fig. 3 (A and B) shows that the maximal binding of [³H]ACh to synaptosomes varies as a function of depolarization. Scatchard analysis of the data of Fig. 3A yields a B_max of 397 ± 15.7 fmol/mg protein and a K_d of 19.8 ± 2.7 nM at 5 mM KCl. At 60 mM KCl the B_max dropped to 105 ± 12.1 fmol/mg protein while the K_d was found to be rather similar (12.2 ± 4.3 nM). Moreover, a close examination of Fig. 3 (A and B) shows that the half-saturation level was not significantly altered, as depolarization varied, while maximal binding was reduced severalfold. These results, together, indicate that depolarization does not affect the rate constants of agonist binding and dissociation, but rather causes a reduction in the number of high affinity binding sites for ACh. This reduction is gradual, as evident from Fig. 3B, where in the same experiment several levels of depolarization were induced. Depolarization-dependent binding is specific to agonist binding, as no voltage dependence was observed in binding of mAChR antagonists such as [³H]NMS (Fig. 3C).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Depolarization-dependent binding properties of mAChRs. A, aliquots of synaptosomes (100 µg of protein) were incubated in [3H]ACh binding solution containing either 5 () or 60 () mM KCl. The data points are from three independent experiments with overlapping [3H]ACh concentrations. Identical [3H]ACh concentrations with 1 µM atropine were used in each experiment to determine nonspecific binding. B, [3H]ACh binding curves, as in A, at varying depolarization levels (, 5; , 20; , 40; , 60 mM KCl), from a single typical experiment. C, [3H]NMS binding at resting potential (, 5 mM KCl) and following depolarization (, 60 mM KCl). Curves were fitted with regression values (R2) of 0.975 and 0.997 for 5 and 60 mM KCl, respectively.

To examine whether the reduction in high affinity mAChRs binding sites is a direct result of KCl-mediated depolarization, we performed the binding experiments using synaptosomes that were first lysed by hypotonic shock and physical shearing. Under these conditions, depolarization did not affect B_max (Fig. 4A). Moreover, the B_max value itself was low and comparable to that obtained under high depolarization (60 mM KCl, compare Fig. 4A and Fig. 3A). Synaptosome membrane potential was determined by [³H]TPP⁺ uptake (Fig. 4B). Indeed, the lysed synaptosomes are already depolarized at 5 mM KCl and almost invariant to changes in KCl concentrations (Fig. 4B).

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Depolarized synaptosomes exhibit low binding properties of mAChR. A, lysed synaptosomes were used in binding assays, as in Fig. 3 in 5 () or 60 () mM KCl. B, membrane potential of fresh () or lysed () synaptosomes and those treated with 20 µM veratridine () was determined by [3H]TPP accumulation. 150 mM KCl was considered as 0 mV. C, binding of [3H]oxotremorine to synaptosomes at resting potential and following incubation with 20 µM veratridine. Veratridine at lower concentration (<5 µM) had negligible effect on the membrane potential and on [3H]oxotremorine binding. Data are the means and S.E. from three experiments using the same synaptosomal preparation. Statistical significance between veratridine-treated and untreated synaptosomes was determined by Student's t test. *, <0.1; ***, <0.005.

To ensure that the reduction in affinity reflects the number of the high affinity binding sites and is not an artifact of the high level of KCl we repeated the binding experiments but used several concentrations of veratridine (between 1 and 100 µM) to depolarize the synaptosomes. Veratridine induces membrane depolarization by a persistent activation of the Na⁺ channels. In the presence of 20 µM veratridine added under conditions of resting potential (5 mM KCl), a depolarization of 30 mV was measured (Fig. 4B). We examined the effect of veratridine (20 µM) on the binding properties of mAChR using oxotremorine as an agonist. Binding of oxotremorine is also voltage-dependent with maximal binding observed at resting potential (data not shown). Binding of [³H]oxotremorine is reduced in the presence of veratridine (20 µM), and the reduction in binding is more pronounced as the concentration of [³H]oxotremorine is elevated (Fig. 4C). Addition of veratridine (20 µM) to already depolarized synaptosomes did not show any additive effect (data not shown). We thus conclude that it is the depolarization that causes reduction in the number of high affinity mAChRs.

Depolarization-mediated Shift of mAChRs from a High Affinity State to a Low Affinity State-- To examine whether the observed reduction in the number of high affinity binding sites can be attributed to depolarization-mediated shift of the receptors affinity states, we conducted displacement experiments. Carbachol (a stable mAChR agonist) was used to displace [³H]NMS. From the displacement curves we could extract the relative fraction of high and low affinity receptor populations and the K_d values for the low affinity state, using conventional competitive inhibition equations.

An example of a displacement experiment is shown in Fig. 5. The data were best fitted using a two-state model. The K_d for the low affinity state (K_L) is calculated to be 110 µM. Following depolarization (Fig. 5B), the population of high affinity sites decreased from 17.5% to less than 1%. Thus, depolarization caused a shift of the receptors from a high to a low affinity state. The K_L calculated from four experiments varied between 84 and 110 µM, and depolarization (60 mM KCl) induced a shift of 15-25% of the receptors from a high to a low affinity state.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Displacement curves for carbachol at different levels of depolarization. Fresh synaptosomes were incubated with AChE (3 units/ml) with 0.8 nM [3H]NMS. Synaptosomes were kept in 5 mM KCl (, A) or depolarized using 90 mM KCl (, B), and carbachol was added at the indicated concentrations. A typical experiment is presented. Each point is a mean of triplicate with variation <15%. The results of the experiments under the different conditions were normalized, and binding of [3H]NMS without adding of the agonist is considered as 100%. Total number of binding sites varies (up to 25%) at different synaptosomal preparations. Four experiments were done. Nonspecific binding (with 1 µM atropine) from each data point was subtracted. The fitting curve is based on a two-affinity state model. Goodness of fit (R2) for A and B are 0.932 and 0.981, respectively. The presented experiment is a representative of four experiments using carbachol and muscarine (two experiments).

The m2 Receptor Subtype Predominates in the Depolarization-mediated Shift in Affinity-- In order to determine which receptor subtype(s) dominates the depolarization-dependent shift in affinity, we performed displacement experiments where methoctramine, a selective antagonist of the m2 receptor subtype, was used to displace [³H]oxotremorine. [³H]Oxotremorine (10 nM) was displaced by methoctramine (10⁹ to 10⁵ M) at different depolarization conditions (Fig. 6). The observed depolarization-dependent shift in [³H]oxotremorine displacement curves indicates that the shift in binding affinity (as determined by oxotremorine) can be attributed mostly to the m2 subtype. At high concentration of methoctramine (>10⁴), the displacement was complete. We repeated the experiment as above, and no depolarization-dependent shift in displacement curves was seen when pirenzepine, a selective antagonist of m1 subtype (m1 > m4 > m2) (35) (at concentration of 10⁹-10⁵ M) was included. We thus conclude that it is mostly the m2 receptor subtype that is responsible for the shift in mAChRs from a high to a low affinity state.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Shift in agonist binding affinity of mAChRs is specific to m2 subtype receptor. Fresh synaptosomes were incubated with 7.5 nM [3H]oxotremorine. Synaptosomes were kept in 5 mM KCl () or depolarized using 90 mM KCl (), and methoctramine was added at the indicated concentrations. Each point is a mean of triplicate with variation <20%. Binding of [3H]oxotremorine without adding of the antagonist is considered as 100%. Three experiments were done. Nonspecific binding (with 1 µM atropine) from each data point was subtracted. The fitting curve is based on a two-affinity state model. Fraction of receptors in high affinity (FH) is 41.5% and 11.4%, at resting and depolarized conditions, respectively. The calculated KL is 7 µM.

The Level of Receptor Occupancy Determines the Strength of the Interaction between m2 and SNAREs-- We next examined the effect of the level of the m2 receptor occupancy on the magnitude of its interaction with the exocytic machinery. The results in Fig. 6 imply that at low depolarization, the m2 receptor is occupied already at rather low concentrations of ACh (in the tens of nanomolar range). In contrast, a much higher concentration of ACh (in the micromolar range) is needed to achieve a high level of occupancy at high depolarization. Based on the above considerations and recalling that the interaction of mAChRs with syntaxin becomes weaker as depolarization rises, we next examined whether experimental conditions that affect the level of receptor occupancy also affect the degree of interaction between the m2 receptor subtype and proteins of the ExA, e.g. syntaxin. To this end, we quantified the co-immunoprecipitation experiments by using a m2-specific antibody (as described in Fig. 2).

We first examined the effect of preincubation of the synaptosomes with AChE. It can be seen that with the excess of AChE that hydrolyzes the endogenous ACh, the magnitude of m2-syntaxin interaction at resting potential (5 mM KCl) was reduced in comparison to control conditions where no AChE was added. Furthermore, this overall weaker interaction was voltage-independent (Fig. 7). In the following experiments we replaced ACh by the AChE non-hydrolyzable agonist, muscarine, and the synaptosomes were pretreated with AChE to hydrolyze endogenous ACh. Fig. 8A shows that, at resting potential, addition of 0.5 µM muscarine strengthened (over 3-fold) the m2-syntaxin interaction, while methoctramine (1 µM), largely reversed the effect of muscarine alone. In contrast, pirenzepine (10 µM), had only a minor effect on the strength of interaction recorded with muscarine alone. An increased level of interaction obtained by addition of muscarine, which is reversed by methoctramine, was demonstrated not only for syntaxin but also for SNAP-25, VAMP-2, and synaptotagmin I (Fig. 8B).

View larger version (49K): [in this window] [in a new window]   Fig. 7.   Co-IP of syntaxin by m2 antibodies in the presence of AchE. Analysis of the co-IP using the m2 mAChR subtype was performed as in Fig. 1. Syntaxin immunoreactivity was quantified following the addition of AChE (5 units/ml). The data are averages taken from three independent experiments, each performed in duplicates. All samples of each experiment were loaded on the same immunoblot and processed simultaneously. Statistical significance was calculated using the Student's t test. Asterisks (***) indicate the significance (p < 0.01) of results at 5 mM KCl with respect to 30 and 60 mM KCl. Intensity units are the pixel numbers measured by densitometry of the immunoreactivity bands using exposures within the linear range of detection.

View larger version (43K): [in this window] [in a new window]   Fig. 8.   Co-IP of syntaxin by m2 antibodies in the presence of mAChR agonist and antagonists. A, Experiment as in Fig. 7, with the following supplements: muscarine (mus) at 0.5 µM, methoctramine (met) and pirenzepine (piz) at 1 and 10 µM, respectively. Data from three experiments at 5 mM KCl are presented, each done in duplicate. Both experiments were normalized according to immunoreactivity intensity value in the first bar (100%). Statistical significance was calculated as in Fig. 7. B, a representative immunoblot presenting the various proteins analyzed. The experiment was performed at 5 mM KCl without AChE, with the addition of muscarine (0.5 µM, lane 2) or methoctramine (1 µM, lane 3). An enhanced interaction induced by muscarine is also seen by comparing synaptosomes without pretreatment with AChE (compare lanes 1 and 2). The abbreviations and antibodies used are as in Fig. 1, with the addition of vamp and syt (indicating antibodies against VAMP-2 and synaptotagmin I, respectively).

Finally, we examined the effect of muscarine concentration on the m2-syntaxin interaction at three levels of depolarization (Fig. 9). As shown, the magnitude of the interaction is reduced as depolarization increases at both levels of muscarine concentration (0.1 and 10 µM). Furthermore, consistent with the qualitative finding shown in Fig. 8, at 5 mM KCl, 0.1 µM muscarine enhanced the m2-syntaxin interaction about 3-fold, and higher muscarine levels (10 µM) had no additional effect. At higher depolarization of 30 mM KCl, the degree of m2-syntaxin interaction in the presence of muscarine is lower than at 5 mM KCl and the interaction is somewhat strengthened at the higher muscarine concentration. At still higher depolarization (60 mM KCl), 0.1 µM muscarine only very slightly strengthened the interaction above control but 10 µM muscarine strengthened the m2-syntaxin interaction to a similar level to that seen in 30 mM KCl. All these results are consistent with the finding seen in Figs. 3 and 5 that depolarization reduces the fraction of high affinity mAChR receptors.

View larger version (52K): [in this window] [in a new window]   Fig. 9.   Co-IP of syntaxin by m2 antibodies under different muscarine concentrations and different depolarization levels. Quantified intensity of m2-syntaxin interaction, at varying depolarization levels. AChE was supplemented in all experiments. Experiments were performed at various KCl concentrations with the addition of 0.1 µM (+) and 20 µM (++) muscarine. Immunoblot was performed with syntaxin antibodies. The control bar (mus ) represents average of IP experiments performed without the addition of muscarine. Experiments were performed in triplicates and identical exposure time (following ECL) were taken for all data points. Averages and variation within the triplicates are shown.

Neither the Shift in mAChR Affinity nor Its Interaction with Syntaxin Depend on Ca²⁺ Influx-- The results shown in Figs. 3-5 were obtained in a Ca²⁺-free solution. Although unlikely, it is still possible that some accumulation of Ca²⁺ in the bathing solution did occur. If so, a depolarization-induced Ca²⁺ influx could occur and potentially affect the binding properties of mAChR. To test this possibility, we repeated the binding experiments following incubation of the synaptosomes with a mixture of voltage-dependent Ca²⁺ channel blockers (calcicludine, or -conotoxin GVIA combined with -agatoxin IVA). At concentration of 1 µM for each -conotoxin GVIA and -agatoxin IVA, a block of the N- and P/Q-type is complete (42). Calcicludine (1 µM) completely blocked the L-, N-, and P/Q-type channels (43). Fig. 10A shows that the effect of depolarization on the binding properties of mAChR is fully retained even when the voltage-dependent Ca²⁺ channels are blocked. The slight increase in the maximal binding capacity at resting potential in the presence of voltage-dependent Ca²⁺ channel blockers is due to a change toward a more negative potential imposed by the blockers (as determined by [³H]TPP⁺ accumulation).

View larger version (49K): [in this window] [in a new window]   Fig. 10.   Effect of depolarization on the binding properties of mAChRs and on the m2-syntaxin interaction is independent of Ca2+ entry. A, specific [3H]ACh binding to fresh synaptosomes at resting potentials (5 mM KCl) and following depolarization (60 mM KCl). Aliquots of synaptosomes (100 µg of protein) were incubated in [3H]ACh binding solution containing either 5 or 60 mM KCl. Experiments were done with different [3H]ACh concentration as in Fig. 3. Binding at 50 nM [3H]ACh is shown (control), with addition of 1 µM -conotoxin GVIA and -agatoxin IVA (-con, -aga, respectively) or with 1 µM calcicludine (calcicl). Each point is an average of duplicates with less than 15% deviation. Experiments were repeated four times. The functionality of the different channel blockers was evaluated by measuring inhibition of release following depolarization in Ca2+-containing solutions. B, IP experiments were performed as in Fig. 2B, except that the fresh synaptosomes were preincubated with calcicludine (1 µM) before and during depolarization. KCl concentrations (in mM) are indicated. The experiment was repeated three times with less than 20% deviation. Quantification of syntaxin immunoreactivity is shown from one representative experiment.

We next examined the effect of the Ca²⁺ channel blockers on the m2-syntaxin interaction. Fig. 10B shows the quantification of such immunoprecipitation experiments at different levels of depolarization. It can be seen that the depolarization-mediated reduction in m2-syntaxin interaction is retained in the presence of the Ca²⁺ channel blockers.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Here we showed that syntaxin is specifically associated with the m2 mAChR subtype (and to a lesser extent with m4 receptor). Based on pharmacological and on immunological tools used in this study, an apparent stoichiometry of 1:1 was calculated between m2-mAChR and the associated proteins in the immunoprecipitation experiments. An apparent 1:1 ratio was also found between the m2 receptor and the N-type voltage-dependent Ca²⁺ channels following immunoprecipitation.² Based on these results we assumed that m2 receptor, proteins on the ExA and the voltage-dependent Ca²⁺ channel are all in a preassembled complex. The molar ratio between m2 subtype receptor and syntaxin in the synaptosomes can be extracted based on the following considerations: (i) the amounts of mAChRs, specifically the m2 receptor, as determined by antagonist binding experiments and by m2-specific antibodies; (ii) the amount of syntaxin present in synaptosomes (syntaxin accounts for nearly 0.3% of the P2 fraction as determined by a purified syntaxin recombinant protein); (iii) the fact that cholinergic terminals occupy only about 10% of all nerve terminals; (iv) quantifying the amounts, in moles, of individual proteins (by using recombinant syntaxin, synaptotagmin, and SNAP-25) and considering their relative molecular mass. Based on the above considerations, we calculated the molar ratio between m2 and syntaxin to be close to 1:20. Such ratio is in accord with the limited amounts of N-type Ca²⁺ channels relative to syntaxin and SNAP-25 in nerve terminals (e.g. Refs. 44-46). Accordingly, we propose that the m2 receptor associated with the exocytic complex and with the Ca²⁺ channels provide a molecular configuration for controlling ACh release from cholinergic nerve terminals upon arrival of the depolarizing signal.

We have demonstrated, in rat brain synaptosomes, two voltage-dependent phenomena associated with mAChRs, in particular the m2 subtype. One involves a depolarization-mediated shift of mAChRs from a high affinity state to a low affinity state. The other phenomenon is associated with the m2-syntaxin (and other SNARE proteins) interaction. This interaction becomes weaker as depolarization rises (see also Ref. 18). We ensured that neither of these voltage-dependent phenomena is mediated by Ca²⁺ influx. All our experiments were conducted in Ca²⁺-free solutions. Furthermore, the effect of membrane potential on both the agonist binding properties of mAChR and on the m2-syntaxin interaction was fully retained in the presence of a mixture of voltage-dependent Ca²⁺ channel blockers (Fig. 10). These two phenomena, therefore, are induced directly by the membrane potential itself. Indeed, neither phenomenon exhibited membrane potential dependence once the synaptosomes were lysed and lost their sensitivity to KCl. Furthermore, veratridine induced similar effects to those achieved by KCl depolarization.

It could be argued that the two phenomena are indeed voltage-dependent but result from a nonspecific effect of depolarization. For example, prolonged depolarization induced depletion of the -subunit of G_i-proteins (47). In such a scenario, coupling of mAChR to its G-protein may be impaired, leading to an accumulation of low affinity receptors (48). We believe that this was not the case in our experiments. No difference in results was recorded between naive synaptosomes and those previously subjected to two complete cycles (30 min each) of depolarization (60 mM KCl) and repolarization (5 mM KCl). Moreover, identical binding curves were obtained whether depolarization was as short as 5 min or as long as 1 h. An additional nonspecific effect of prolonged depolarization may involve membrane endocytosis. In such scenario, lowering in B_max values (Fig. 3, A and B) will reflect the reduction in total number of exposed receptors at high depolarization levels. This possibility can be excluded, as binding of antagonists was not affected at all by depolarization (e.g. Fig. 3C). Moreover, in direct binding experiments (as in Fig. 3), the effect of mAChR affinity shift is duplicated when synaptosomes were incubated throughout the procedure at 10 °C, conditions in which endocytosis is prevented. We may therefore conclude that the two phenomena, the effect of membrane potential on agonist binding properties of mAChR and on the m2-syntaxin interaction reflect a genuine specific effect of membrane potential on mAChR properties.

At this stage of the research, it is not clear which molecule is the voltage sensor that transduces the depolarization signal to the release machinery. The change in membrane potential may induce a change in the receptor itself (as proposed in Ref. 49) or in its associated G-protein (50). Recently, a link between the integrity of syntaxin and a G-protein-mediated effect on the voltage-gated Ca²⁺ channels was shown (51). A physical and functional association between voltage-gated N-type Ca²⁺ channels and the exocytic machinery, specifically syntaxin, SNAP-25, and synaptotagmin was shown (e.g. Refs. 15, 17, 46, and 52). Our observation that a protein complex of m2 receptor subtype with ExA also includes the N-type calcium channel³ is in accord for a role for these channels as voltage sensors. While blocking the N-type Ca²⁺ channels with -conotoxin did not abolished any of the voltage-dependent effects reported in this study (Fig. 10), the channel may still serve as a voltage sensor. As such, it may alter the interaction with the m2 subtype receptor associated with the exocytic machinery.

In an attempt to resolve a direct interaction between the m2-mAChR receptor and the individual components of the exocytic machinery we transfected mAChR-m2 and mAChR-m1 expressing CHO cells with several of the proteins that co-precipitated with mAChRs from brain synaptosomes. Following transient transfection of CHO-m2 expressing cells with syntaxin, we detected high level of expression at various cellular membranes including substantial amounts at the plasma membrane. Still, no significant binding (between the receptor and expressed syntaxin) was recorded by immunoprecipitation experiment using either syntaxin antibodies or m2-mAChR antibodies following agonist and antagonist manipulation of the transfected cells.⁴ Our assumption is that the interactions reported here are mediated (or stabilized) by the presence of a linker protein, which is by itself a syntaxin-binding protein. Best candidates are synaptotagmin and the N-type Ca²⁺ channel but other candidates may still be involved. We are currently exploring these possibilities.

We may now ask whether the two voltage-dependent phenomena (altered agonist binding properties of mAChR and altered interaction of the m2 subtype with syntaxin) are linked, and if so, what could be their physiological role in the release of neurotransmitter. We propose that the two phenomena are indeed linked and exhibit a causal relationship. We suggest that the occupancy of mAChR is determined by its affinity state which is controlled by membrane potential. When occupied, mAChR, specifically the m2 subtype, interacts with member(s) of the exocytic apparatus. Several observations support this idea: (i) In the presence of the endogenous concentration of ACh, mAChR exhibits a voltage-dependent interaction with ExA, strong at resting potential and weak upon depolarization. (ii) Upon addition of AChE, the interactions were weaker overall and the voltage dependence of the interaction nearly disappeared. (iii) At resting potential, when a larger proportion of mAChRs are in the high affinity state, addition of muscarine, even at low concentration (0.1 µM), enhanced the m2-syntaxin interaction. At high depolarization (60 mM KCl) where most of the mAChRs are already at their low affinity state, only high concentrations of muscarine (20 µM) had measurable effect on this interaction. However, at 30 mM KCl, where a fraction of high affinity mAChRs still exists, but is reduced (Fig. 3B), addition of muscarine (0.1 µM and 20 µM) strengthened the m2-syntaxin interaction.

How could these observations be related to release of neurotransmitter from nerve terminals? We can speculate that at resting potential the presynaptic mAChRs exhibit high affinity for ACh. As a result, the low concentration of neurotransmitter present in the synaptic cleft ensures that receptors are in an occupied state. The occupied receptors, in turn, interact with the exocytic machinery and consequently keep it blocked. When an action potential reaches the nerve terminal, there is an abrupt depolarization, and the receptors shift into a low affinity state. The neurotransmitter dissociates and the unoccupied receptors no longer interact with the exocytic machinery. The free exocytic machinery can now be engaged in the processes needed for fusion to commence. Presumably, parallel to these processes Ca²⁺ that enters via voltage-dependent channels enables the necessary Ca²⁺-dependent reactions to take place. These two parallel processes together enable release of neurotransmitter to occur.

	ACKNOWLEDGEMENTS

We are grateful to E. Heldman for the contribution of CHO-K1 receptor-specific transfected cells and mAChR antibodies, and for valuable suggestions and comments; H. Boschwitz for expert technical assistance; and Y. Bledi for analyzing mAChRs in CHO cells. We thank M. Takahashi (Misubishi, Tokyo, Japan) for the kind contribution of antibodies.

	FOOTNOTES

^* This work was supported in part by grants from the National Institute for Psychobiology in Israel (to M. L) and the Israeli Ministry of Health (to M. L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ Supported by a Sonderforschungsbereich grant.

To whom correspondence should be addressed. Tel.: 972-2-6585425; Fax: 972-2-6586448; E-mail: michall@leonardo.ls.huji.ac.il.

² L. Branski and M. Linial, manuscript in preparation.

³ L. Branski, Y. Bledi, and M. Linial, in preparation.

⁴ Y. Bledi and M. Linial, unpublished data.

	ABBREVIATIONS

The abbreviations used are: SNARE, soluble NSF attachment protein receptor; NSF, N-ethylmaleimide-sensitive factor; ExA, exocytic apparatus; mAChR, muscarinic acetylcholine receptor; IP, immunoprecipitation; SNAP-25, synaptosomal associated protein of 25 kDa; PAGE, polyacrylamide gel electrophoresis; VAMP, vesicle-associated membrane protein; CHO, Chinese hamster ovary; GST, glutathione S-transferase; AChE, acetylcholine esterase; ACh, acetylcholine; QNB, quinuclidinyl benzilate; NMS, N-[methyl]scopolamine; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; GDPS, guanosine 5'-O-(2-thiodiphosphate); GTPS, guanosine 5'-O-(3-(thiotriphosphate).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES