Ca2+ regulates the interaction between synaptotagmin and syntaxin 1.

While there is compelling evidence that the synaptic vesicle protein synaptotagmin serves as the major Ca2+ sensor for regulated exocytosis, it is not known how Ca2+ binding initiates membrane fusion. Here we report that Ca2+ increases the affinity, by approximately 2 orders of magnitude, between synaptotagmin and syntaxin 1, a component of the synaptic fusion apparatus. This effect is specific for divalent cations which can stimulate exocytosis of synaptic vesicles (Ca2+ > Ba2+, Sr2+ Mg2+). The Ca2+-dependence of the interaction was composed of two components with EC50 values of 0.7 and 180 μM Ca2+. The interaction is mediated by the carboxyl-terminal region of syntaxin 1 (residues 194-288) and is regulated by a novel Ca2+-binding site(s) which does not require phospholipids and is not disrupted by mutations that abolish Ca2+-dependent phospholipid binding to synaptotagmin. We propose that this interaction constitutes an essential step in excitation-secretion coupling.

Exocytosis of synaptic vesicles is strictly controlled by Ca 2ϩ ions (Katz, 1969;Augustine et al., 1987). Presumably, Ca 2ϩ ions initiate conformational changes in proteins which ultimately catalyze membrane fusion. The Ca 2ϩ binding properties of the synaptic vesicle protein synaptotagmin are consistent with the requirements for the exocytotic Ca 2ϩ receptor (Brose et al., 1992;reviewed by DeBello et al. (1993), Popov and Poo (1993), Chapman and Jahn (1994b)). Indeed, gene disruption (Geppert et al., 1994;Broadie et al., 1994;Littleton et al., 1994;Nonet et al., 1994) and microinjection experiments (Elferink et al., 1993;Bommert et al., 1993) have provided strong evidence that synaptotagmin functions as the major Ca 2ϩ sensor in regulated exocytosis. However, the mechanism by which Ca 2ϩ binding to synaptotagmin triggers membrane fusion has yet to be elucidated. To address this issue, efforts have been directed at identifying downstream effectors of Ca 2ϩ -synaptotagmin action. Synaptotagmin was shown to form Ca 2ϩ -independent complexes with neurexins (Petrenko et al., 1991), a family of neuronal cell surface proteins, and the adaptor protein AP-2 (Zhang et al., 1994). Synaptotagmin also interacts with syntaxin 1, an abundant neuronal plasma membrane protein associated with N-type Ca 2ϩ channels (Bennett et al., 1992;Yoshida et al., 1992;Sheng et al., 1994). This property suggests that syntaxin may physically link the calcium receptor to the site of Ca 2ϩ influx.
Syntaxin has been shown recently to form a complex with the synaptic vesicle protein synaptobrevin (VAMP) and the synaptic plasma membrane protein SNAP-25. This complex serves as the membrane receptor for the soluble proteins NSF 1 (N-ethylmaleimide-sensitive fusion protein) and SNAPs (soluble NSF attachment proteins), factors required for intracellular membrane fusion. Therefore, syntaxin, SNAP-25, and synaptobrevin have been designated as SNAREs (SNAP receptors (Söllner et al., 1993a)). In addition, each of the SNAREs is selectively proteolyzed by clostridial neurotoxins, potent inhibitors of exocytosis (Schiavo et al., 1992;Link et al., 1992;Blasi et al., 1993aBlasi et al., , 1993b. Furthermore, disruption of the syntaxin 1 gene in Drosophila abolishes evoked neurotransmission (Schulze et al., 1995). Thus, the complex containing syntaxin, SNAP-25, and synaptobrevin is thought to comprise the core of the exocytotic fusion machine. Consequently, the interaction between syntaxin 1 and synaptotagmin is of particular interest, since it could provide a direct link between the Ca 2ϩ -sensor and the fusion apparatus. In the present study we have characterized the interaction between synaptotagmin and syntaxin 1 and report that it is regulated by Ca 2ϩ ions.

EXPERIMENTAL PROCEDURES
Immunoprecipitation-All manipulations were carried out on ice. Synaptosomes were prepared by homogenizing one to two rat brains in 30 ml of 320 mM sucrose with 10 strokes at 900 rpm using a Teflon glass homogenizer. The homogenate was centrifuged at 5000 rpm for 2 min in an SS34 rotor, and the crude synaptosomes were collected by centrifugation of the supernatant at 11,000 rpm for 12 min in an SS34 rotor. Synaptosomes were solubilized for 45 min at a detergent to protein ratio of 10:1 (w:w) with 1% Triton X-100 in 50 mM HEPES, pH 7.2, 100 mM NaCl, and protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 2 g/ml pepstatin, 20 g/ml aprotinin). Insoluble material was removed by centrifugation at 50,000 rpm in a TLA 100.3 rotor for 15 min, and samples were then supplemented with 2 mM EGTA or 0.5 mM CaCl 2 . Immunoprecipitations were carried out by incubating aliquots of the detergent extract (1 mg of protein) with 15 l of ascites containing monoclonal antibodies directed against synaptophysin (Cl 7.3) or syntaxin 1 (HPC-1 or Cl 78.2) for 2 h followed by mixing with 30 l of protein G-Sepharose fast flow (Pharmacia Biotech Inc.) for 1 h. HPC-1 recognizes both syntaxin 1A and 1B and has been described previously (Barnstable et al., 1985). Cl 78.2 is a newly generated monoclonal antibody raised against recombinant full-length syntaxin 1A that also recognizes syntaxin 1A and 1B and will be described in detail elsewhere. Immunoprecipitates were washed four times with the immunoprecipitation buffer and subjected to SDS-PAGE and immunoblot analysis as described (Chapman and Jahn, 1994a). Immunoreactive bands were visualized with 125 I-protein A.
Binding of recombinant proteins was also carried out by immunoprecipitation of syntaxin. Recombinant full-length syntaxin 1A and C2AB were incubated at the indicated concentrations in 20 mM Tris, pH 7.4, 150 mM NaCl, 0.5% Triton X-100 supplemented with EGTA or divalent * 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.
‡ To whom correspondence should be addressed. cations for 4 h. Syntaxin 1 was immunoprecipitated by incubating the samples with affinity purified HPC-1 IgG (7 g) for 2 h and 12 l of protein G-Sepharose fast flow (Pharmacia) for 1 h. The immunoprecipitates were washed three times and analyzed by SDS-PAGE and immunoblotting as described above.
Recombinant Proteins-A soluble form of full-length rat syntaxin 1A was prepared as described (Chapman et al., 1994) which contains a His 6 -tag at the amino terminus. Immobilized full-length and truncated syntaxins were prepared by amplifying the indicated regions of syntaxin 1A, using the polymerase chain reaction, and subcloning the products into pGEX-2T (Pharmacia). The resulting GST fusion proteins were expressed in Escherichia coli, purified, and immobilized using glutathione-Sepharose as described (Chapman and Jahn, 1994a). The cytoplasmic domain of rat synaptotagmin I, designated C2AB (residues 97-421; Chapman and Jahn, 1994a) was also prepared as a GST fusion protein by subcloning into pGEX-2T. A motif crucial for Ca 2ϩ -dependent phospholipid binding to the first C2-domain (SDPYVK--L, residues 177-185 (Chapman and Jahn, 1994a;Davletov and Sü dhof, 1993)) was deleted from C2AB and designated C2A⌬B. For comparison, the corresponding motif (amino acids 308 -316) was also deleted in the second C2-domain of C2AB (referred to as C2AB⌬). Mutagenesis was performed as described (Chapman and Jahn, 1994a) and confirmed by DNA sequencing. These deletion mutants were prepared as GST fusion proteins by subcloning into pGEX-2T as described above. Soluble C2AB, C2A⌬B, and C2AB⌬ were generated by thrombin cleavage as described (Chapman et al., 1994).
Miscellaneous-Phospholipid binding assays were carried out as described previously (Chapman and Jahn, 1994a) by immobilizing GST-C2AB, GST-C2A⌬B, and GST-C2AB⌬ on glutathione-Sepharose. The total radiolabeled phospholipid binding data are plotted in Fig. 6B. Quantitation of the immunoblots was carried out using a Molecular Dynamics PhosphorImager and ImageQuant software. For the Ca 2ϩ dose-response analysis, samples were buffered with 1 mM EGTA and the total Ca 2ϩ added to yield the indicated free Ca 2ϩ concentrations was determined as described previously (Chapman and Jahn, 1994a).

RESULTS
To determine the effect of Ca 2ϩ ions on the interaction between syntaxin and synaptotagmin, we immunoprecipitated syntaxin 1 from synaptosomal detergent extracts, using two distinct monoclonal antibodies directed against syntaxin 1, in the presence of Ca 2ϩ or excess Ca 2ϩ chelator and examined the precipitates for the presence of synaptotagmin. As shown in Fig. 1 (top panel), Ca 2ϩ dramatically increased the level of synaptotagmin associated with syntaxin 1. We did not observe the calcium-dependent recruitment of additional proteins to the syntaxin immunoprecipitates by either protein staining or immunoblot analysis using antibodies directed against other synaptic proteins (data not shown). In addition, the amount of synaptobrevin and SNAP-25, components of the SNARE complex, bound to syntaxin was not affected by Ca 2ϩ ( Fig. 1 and data not shown). Neither synaptotagmin nor syntaxin 1 was detected in control immunoprecipitations using anti-synaptophysin antibodies (Fig. 1, left lanes).
To determine whether the Ca 2ϩ -dependent coprecipitation is due to a direct interaction between syntaxin 1 and synaptotagmin, we repeated the syntaxin immunoprecipitations using purified recombinant proteins (Fig. 2, left panel). Full-length syntaxin 1A was mixed with the cytoplasmic portion of synaptotagmin I (amino acids 97-421). This domain contains two repeats of a region homologous to the C2-domains found in Ca 2ϩ -dependent isoforms of protein kinase C (Perin et al., 1990) and is designated C2AB. As shown in Fig. 2 (left panel), addition of Ca 2ϩ results in a 6-fold increase (determined by phosphorimage analysis) in the level of C2AB which coprecipitates with syntaxin 1. Surprisingly, this increase did not require, and was not enhanced by, negatively charged phospholipids (Fig. 2, left panel). Since Ca 2ϩ binding to isolated synaptotagmin requires the presence of negatively charged phospholipids (Brose et al., 1992), these findings suggest that a novel Ca 2ϩ binding site regulates the synaptotagmin ⅐ syntaxin 1 interaction (discussed in more detail below).
We next compared the effects of different divalent cations on the syntaxin 1-synaptotagmin interaction. As shown in Fig. 2 (right panel), Ba 2ϩ and Sr 2ϩ ions promoted the association of syntaxin 1 and synaptotagmin, albeit less potently than Ca 2ϩ . In contrast, Mg 2ϩ was virtually inactive and, in addition, failed to antagonize the effects of Ca 2ϩ , even at high concentrations (5 FIG. 1. Coprecipitation of synaptotagmin with syntaxin 1 from rat brain detergent extracts is stimulated by Ca 2؉ . Syntaxin 1 (syx) was immunoprecipitated from rat brain detergent extracts as described under "Experimental Procedures" using monoclonal antibodies HPC-1 and 78.2 in the presence of EGTA (2 mM, ϪCa 2ϩ ) or Ca 2ϩ (0.5 mM, ϩCa 2ϩ ). As a control, synaptophysin (syp) was immunoprecipitated in parallel with monoclonal antibody Cl 7.3. The immunoprecipitates were separated by SDS-PAGE and analyzed by immunoblotting for synaptotagmin, syntaxin 1, synaptophysin, and synaptobrevin. Immunoreactive bands were visualized using 125 I-protein A and autoradiography. Note that equal levels of syntaxin 1 were precipitated by the antibodies under all conditions. As expected, synaptophysin antibodies did not precipitate syntaxin 1 or synaptotagmin but efficiently coprecipitated synaptobrevin with synaptophysin, in agreement with our earlier observations (Edelmann et al., 1995).
FIG. 2. Phospholipid and divalent cation dependence for the association of synaptotagmin with syntaxin 1. Left panel, Ca 2ϩdependent binding of synaptotagmin to syntaxin 1 does not require acidic phospholipids. Recombinant synaptotagmin I (residues 97-421; designated C2AB) and syntaxin 1A (full length) were prepared as described under "Experimental Procedures." C2AB (0.8 M) was incubated in 20 mM Tris, pH 7.2, 150 mM NaCl, 0.5% Triton X-100, supplemented with EGTA (2 mM), Ca 2ϩ (0.5 mM), or Ca 2ϩ (0.5 mM) plus phospholipids (ϩPL; 3.7 mM phosphatidylcholine, 1.25 mM phosphatidylserine) in the presence (ϩ) or absence (Ϫ) of recombinant syntaxin 1 (0.8 M). Binding was monitored by immunoprecipitation using a purified anti-syntaxin 1 monoclonal antibody (HPC-1) as described in Experimental Procedures. The immunoprecipitates were separated by SDS-PAGE and analyzed by immunoblotting. Immunoreactive bands were visualized using 125 I-protein A and autoradiography. C2AB was precipitated only in the presence of syntaxin 1. Coprecipitation was strongly enhanced by Ca 2ϩ and was independent of negatively charged phospholipids. Right panel, divalent cation specificity for promoting binding of synaptotagmin to syntaxin 1. Binding of synaptotagmin to syntaxin 1 was assayed by coimmunoprecipitation as described in legend for the left panel in the presence of EGTA (2 mM) or the following divalent cation concentrations: Mg 2ϩ alone, 2.5 mM; Ca 2ϩ , Ba 2ϩ , Sr 2ϩ , 0.5 mM each; Mg 2ϩ (5 mM) in the presence of Ca 2ϩ (0.5 mM). Note that equal amounts of syntaxin 1 were immunoprecipitated under all conditions. In the absence of syntaxin 1, C2AB was not detected in the immunoprecipitates (not shown). mM; Fig. 2, right panel). These findings agree with physiological studies demonstrating that Ba 2ϩ and Sr 2ϩ , but not Mg 2ϩ , ions can partially substitute for Ca 2ϩ in regulated exocytosis (Augustine et al., 1987).
The Ca 2ϩ dependence of the interaction was determined by incubating rat brain detergent extracts, containing native synaptotagmin, with immobilized syntaxin under varying Ca 2ϩ concentrations. Full-length syntaxin 1A was expressed as a GST fusion protein and immobilized using glutathione-Sepharose. After extensive washing, binding of synaptotagmin was determined by immunoblot analysis. As shown in Fig. 3, the Ca 2ϩ dependence of the syntaxin-synaptotagmin interaction appears to be composed of two distinct components. By curve fitting these individual components (see legend, Fig. 3) we estimate that the EC 50 values for the high and low affinity binding sites are 0.7 and 180 M Ca 2ϩ , respectively. There is evidence that at least four calcium ions bind to the calcium sensor in a cooperative manner to activate synaptic vesicle fusion (for example, see Heidelberger et al. (1994)). Determining whether there is a cooperative relationship between the Ca 2ϩ concentration and the interaction between syntaxin and synaptotagmin will require the dissection of the high and low Ca 2ϩ affinity components.
The affinity of the synaptotagmin-syntaxin interaction was determined by titrating syntaxin 1 with increasing concentrations of C2AB in the presence of Ca 2ϩ or EGTA. Again, binding was assayed by coprecipitation and immunoblot analysis (Fig.   4, left panels). In the presence of Ca 2ϩ , syntaxin 1 was saturated at low micromolar C2AB concentrations (EC 50 ϳ0.5 M). At saturation, the binding stoichiometry approached 1:1 (Fig.  4, right panel). In the presence of EGTA, only weak binding was observed, even at the highest concentration of synaptotagmin tested (Fig. 4). Assuming that the binding stoichiometry at saturation is not affected by Ca 2ϩ , we estimate that Ca 2ϩ shifts the affinity of the interaction by approximately two orders of magnitude (Fig. 4, lower panel).
To understand in more detail the interaction between syntaxin 1 and synaptotagmin, we investigated the structural determinants which mediate Ca 2ϩ -dependent binding. As in the Ca 2ϩ titration experiments, full length and truncated forms of syntaxin 1A were expressed as GST fusion proteins (Fig. 5A). These fusion proteins were immobilized using glutathione-Sepharose and incubated with synaptosomal detergent extracts in the presence of Ca 2ϩ or EGTA. After washing, binding of synaptotagmin, and, as controls, synaptobrevin and Rab 3A, were determined by immunoblot analysis (Fig. 5B). As expected, immobilized full-length syntaxin 1 bound to synaptotagmin in a Ca 2ϩ -dependent manner. This effect was specific for synaptotagmin, since Ca 2ϩ did not modulate the amount of synaptobrevin bound to immobilized syntaxin 1 and did not cause other synaptic proteins (e.g. Rab 3A, Fig. 5B) to become associated with the immobilized syntaxin. Synaptotagmin did not bind to residues 1-193 of syntaxin but bound efficiently, in a Ca 2ϩ -dependent manner, to residues 194 -288 (Fig. 5B). Interestingly, the same region of syntaxin 1 also mediates its interaction with SNAP-25 (Chapman et al., 1994), synaptobrevin (Calakos et al., 1994; Fig. 5B), ␣-SNAP Kee et al., 1995) and N-type Ca 2ϩ -channels (Sheng et al., 1994). This region of syntaxin is composed of heptad repeats with a high probability of forming coiled-coils (Chapman et al., 1994;FIG. 3. Ca 2؉ dependence of synaptotagmin-syntaxin binding. Full-length syntaxin 1 was prepared as a GST-fusion protein and immobilized using glutathione-Sepharose. The immobilized protein (0.5 nmol) was incubated with 1 mg of the Triton X-100 rat brain synaptosomal extract (1 mg/ml) in the presence of 2 mM EGTA or 1 mM EGTA and sufficient Ca 2ϩ to yield the indicated free Ca 2ϩ concentration, for 4 h at 4°C. In parallel, GST alone was incubated with the detergent extract as a control. Beads were washed three times and the association of native synaptotagmin determined by SDS-PAGE and immunoblot analysis using anti-synaptotagmin 1 monoclonal antibody Cl 41.1. Immunoreactive bands were visualized with 125 I-protein A (upper panel). These data were quantified by phosphorimaging and are plotted in the lower panel. Error bars represent the standard deviation from the mean of three independent experiments. Two distinct inflection points in the binding curve corresponding to high and low affinity Ca 2ϩ binding sites were observed. Curves were fit to each of these two components using a logistic equation (De Lean et al., 1978). From this analysis, EC 50 values of 0.7 and 180 M Ca 2ϩ were calculated and are denoted by arrows (lower panel). E corresponds to assays carried out in 2 mM EGTA.
FIG. 4. Concentration dependence of synaptotagmin binding to syntaxin 1 in the presence and absence of Ca 2؉ . Left panel, a fixed amount of syntaxin 1 (0.5 M) was incubated with increasing amounts of C2AB in 2 mM EGTA or 0.5 mM Ca 2ϩ . Binding was assayed by immunoprecipitation and immunoblot analysis as described under "Experimental Procedures." The bands immunoreactive for synaptotagmin were quantified by phosphorimaging. These data are plotted in the lower panel (closed circles, binding measured in Ca 2ϩ ; open circles, binding measured in EGTA). For the estimation of the EC 50 , (0.5 M in Ca 2ϩ ), curves were fit to the data using a logistic equation (De Lean et al., 1978). Right panel, Coomassie Blue staining of the anti-syntaxin 1 immunoprecipitates obtained after incubating 10 M C2AB with or without 0.5 M syntaxin 1 in 2 mM EGTA or 0.5 mM Ca 2ϩ . Note that the staining of C2AB and syntaxin 1 is of approximately equal intensity in the presence of Ca 2ϩ . H and L denote the heavy and light chains, respectively, of the HPC-1 IgG used for immunoprecipitation. Kee et al., 1995), suggesting that at least some of these interactions may be mediated by intermolecular coiled coils.
It is notable that the removal of the transmembrane domain of syntaxin (residues 266 -288) resulted in decreased synaptotagmin binding activity and also diminished the ability of syntaxin to bind synaptobrevin (Fig. 5B) and ␣-SNAP . In addition, insertion of the transmembrane region into membranes is required for cleavage of syntaxin by botulinum neurotoxin C1 (Blasi et al., 1993b). While it is unlikely that the transmembrane domain participates in direct contacts with other proteins, these data suggest it may be essential for oligomerization and/or folding of syntaxin into its correct conformation.
The data described above demonstrate that Ca 2ϩ regulates the interaction between syntaxin 1 and synaptotagmin in the absence of phospholipids. As mentioned above, this contrasts with the finding that purified synaptotagmin binds Ca 2ϩ only in the presence of phospholipids (Brose et al., 1992), a property conferred by the first C2-domain (Davletov and Sü dhof, 1993;Chapman and Jahn, 1994a;Fukuda et al., 1994). Within this domain, a short sequence of highly conserved residues (SD-PYVK--L) has been identified that is crucial for Ca 2ϩ binding. Deletion of, or point mutations within, this motif abolish Ca 2ϩdependent phospholipid binding to the isolated first C2-domain (Davletov and Sü dhof, 1993;Chapman and Jahn, 1994a). In contrast, the isolated second C2-domain does not bind to phospholipids in a Ca 2ϩ -dependent manner, even though this domain contains the SDPYVK--L motif (Fukuda et al., 1994). To determine the role of these motifs in Ca 2ϩ -dependent syntaxin 1 binding, we prepared mutant synaptotagmins (Fig. 6A) which contained this deletion in either the first (designated C2A⌬B) or second C2-domain (designated C2AB⌬). The Ca 2ϩ -dependent phospholipid and syntaxin 1 binding properties of these mutants were then compared.
Deletion of the conserved motif within the first C2-domain abolished Ca 2ϩ -dependent phospholipid binding, whereas deletion of the corresponding motif within the second C2-domain had no effect (Fig. 6B). In contrast, binding of both deletion mutants to syntaxin 1 was stimulated by Ca 2ϩ to the same extent as that of the wild type cytoplasmic domain of synaptotagmin (Fig. 6C). These results clearly demonstrate that Ca 2ϩdependent binding of synaptotagmin to syntaxin 1 involves structural features distinct from those required for the Ca 2ϩdependent interaction of synaptotagmin with phospholipids. Therefore, Ca 2ϩ regulates the synaptotagmin-syntaxin interaction via a novel Ca 2ϩ -binding site within the complex.

DISCUSSION
Recent studies have provided insights into the sequence of events that may lead to bilayer fusion (reviewed by Rothman and Warren (1994) and Jahn and Ferro-Novick (1994)). An early step in this pathway includes assembly of the SNARE proteins synaptobrevin, syntaxin 1, and SNAP-25, linking the target membrane to the incoming carrier vesicle. The assembled SNARE complex then recruits ␣-SNAP, enabling NSF to bind. NSF is an ATPase and upon hydrolysis of ATP dissociates the SNARE complex, an event proposed to result in membrane fusion (Söllner et al., 1993b).
How can the Ca 2ϩ -stimulated interaction between synaptotagmin and syntaxin 1 be integrated into this model? Söllner et al. (1993b) observed that in neuronal detergent extracts, a small (substoichiometric) amount of synaptotagmin is associated with syntaxin 1 which could be displaced by the addition of exogenous ␣-SNAP (Söllner et al., 1993b). It was therefore suggested that synaptotagmin interacts with the SNARE complex before ␣-SNAP and NSF bind and dissociate the complex. This view, however, is difficult to reconcile with the dramatic increase in affinity of synaptotagmin for syntaxin upon Ca 2ϩ influx. Rather, we believe that in the nerve terminal the association of synaptotagmin with syntaxin 1 occurs after dissociation of the complex by NSF (O'Conner et al., 1994). In this scenario, the NSF-dependent dissociation of the complex can be viewed as an ATP-dependent priming step that is necessary but not sufficient for membrane fusion. For exocytosis to proceed, dissociation needs to be succeeded by the Ca 2ϩ -dependent association of synaptotagmin with syntaxin 1. Such a model would provide an explanation for the apparent lack of ATP dependence in the final step of exocytosis and puts the Ca 2ϩ sensor, synaptotagmin, closer to the fusion event (Hay and Martin, 1992;Bittner and Holz, 1992;Thomas et al., 1993;Neher and Zucker, 1993).
The effect of Ca 2ϩ on the affinity of the synaptotagminsyntaxin interaction is likely to reflect a conformational change in either one or both of these proteins. The significance of the high and low affinity Ca 2ϩ -dependent components is currently under investigation. However, the component which displays a low affinity for Ca 2ϩ (EC 50 ϭ 180 M) is the first calcium-dependent interaction which corresponds to the calcium depend- FIG. 5. Residues 194 -288 of syntaxin 1 mediate Ca 2؉ -dependent synaptotagmin binding. A, Coomassie Blue-stained gel of full-length and truncated forms of syntaxin 1 fused to GST. GST, full-length and truncated forms of syntaxin 1A fused to GST, were purified using glutathione-Sepharose and subjected to SDS-PAGE on 15% gels. Recombinant proteins were visualized by staining with Coomassie Blue. B, identification of the synaptotagmin binding domain of syntaxin. The GST-syntaxin 1 fusion proteins (0.5 nmol) were immobilized using glutathione-Sepharose and incubated with Triton X-100 extract of synaptosomes (0.5 mg of protein) in the presence of 2 mM EGTA (Ϫ) or 0.5 mM Ca 2ϩ (ϩ) for 4 h at 4°C. Beads were extensively washed and subjected to SDS-PAGE and immunoblot analysis to detect synaptotagmin and, as controls, synaptobrevin and Rab 3A. Immunoblots were visualized with 125 I-protein A. Total corresponds to 10 g of the synaptosomal extract. ence of neurotransmitter release reported in neurons (EC 50 ϭ 194 M; Heidelberger et al., 1994). We therefore propose that the putative conformational changes associated with this component functions in the late acting triggering step in exocytosis.