Phosphorylation of syntaphilin by cAMP-dependent protein kinase modulates its interaction with syntaxin-1 and annuls its inhibitory effect on vesicle exocytosis.

cAMP-dependent protein kinase (PKA) can modulate synaptic transmission by acting directly on the neurotransmitter secretory machinery. Here, we identify one possible target: syntaphilin, which was identified as a molecular clamp that controls free syntaxin-1 and dynamin-1 availability and thereby regulates synaptic vesicle exocytosis and endocytosis. Deletion mutation and site-directed mutagenesis experiments pinpoint dominant PKA phosphorylation sites to serines 43 and 56. PKA phosphorylation of syntaphilin significantly decreases its binding to syntaxin-1A in vitro. A syntaphilin mutation of serine 43 to aspartic acid (S43D) shows similar effects on binding. To characterize in vivo phosphorylation events, we generated antisera against a peptide of syntaphilin containing a phosphorylated serine 43. Treatment of rat brain synaptosomes or syntaphilin-transfected HEK 293 cells with the cAMP analogue BIMPS induces in vivo phosphorylation of syntaphilin and inhibits its interaction with syntaxin-1 in neurons. To determine whether PKA phosphorylation of syntaphilin is involved in the regulation of Ca(2+)-dependent exocytosis, we investigated the effect of overexpression of syntaphilin and its S43D mutant on the regulated secretion of human growth hormone from PC12 cells. Although expression of wild type syntaphilin in PC12 cells exhibits significant reduction in high K(+)-induced human growth hormone release, the S43D mutant fails to inhibit exocytosis. Our data predict that syntaphilin could be a highly regulated molecule and that PKA phosphorylation could act as an "off" switch for syntaphilin, thus blocking its inhibitory function via the cAMP-dependent signal transduction pathway.

Neurotransmitter release involves a series of protein interactions between the membranes of synaptic vesicles and presynaptic terminals, culminating in the calcium-dependent fusion of the two membranes (for reviews see Refs. 1-4). The synaptic vesicle-associated protein synaptobrevin interacts with two membrane-associated proteins, SNAP 1 -25 and syn-taxin, to form a stable SNARE complex (5)(6)(7)(8)(9)(10)(11)(12). Formation of the SNARE complex has been proposed to bring the synaptic vesicle and plasma membranes into close apposition and provide the energy that drives the mixing of the two lipid bilayers (6,7,10). Calcium influx into the presynaptic terminal triggers the calcium sensor of the fusion machinery, upon which the SNARE core complex binding matures from a trans-state to a cis-state, resulting in the complete fusion of the two membranes and the release of neurotransmitter. The very stable cis-SNARE core complex is then dissociated by the action of ␣-SNAP and the ATPase N-ethylmaleimide-sensitive factor (10). The assembly and disassembly of the SNARE complex must be highly regulated to gain plasticity in neurotransmitter release. Recent studies have made significant progress in understanding this regulatory mechanism by the isolation of the SNARE complex interacting proteins, including Munc18/n-Sec1/rbSec1, complexins, Munc-13, tomosyn, cysteine string protein, snapin, and syntaphilin (13)(14)(15)(16)(17)(18)(19), which bind to individual SNARE proteins and regulate their availability to form functional SNARE complexes at release sites.
Second messenger regulation of the protein interactions underlying neurotransmission is one mechanism by which cellular events modulate synaptic transmission (4). Although the time course between action potential arrival at the nerve terminal and synaptic vesicle fusion is too short for protein phosphorylation/dephosphorylation to exact a direct and acute role in a single round of vesicle exocytosis, protein kinases and phosphatases may have significant effects on subsequent neurotransmitter release events. It is reasonable to speculate that the phosphorylation/dephosphorylation states of synaptic proteins that mediate vesicle exocytosis could regulate the biochemical pathways leading from synaptic vesicle docking to fusion. Activation of cAMP-dependent protein kinase (PKA) has been shown to facilitate synaptic transmission at many synapses (20 -24). Application of the adenylate cyclase activator forskolin on adult rat and mouse hippocampal slices or hippocampal neuron cultures increased the frequency of spontaneous miniature excitatory postsynaptic currents without affecting their amplitude (21,25,26). In hippocampal neurons, activation of PKA increases neurotransmitter release by directly acting on the exocytotic apparatus (27). The number of functional presynaptic boutons of hippocampal neuron cultures increased after long term exposure to the cAMP analogue (28,29). However, there was no change in the number of functioning terminals and nor in the number of docked vesicles within individual synapses after 3 min of treatment of cultured hippocampal neurons with forskolin (27), suggesting that the short term effects of cAMP may affect the secretory machinery di-rectly. Thus, identifying the PKA target(s) that regulates assembly/disassembly of the fusion machinery and the priming of docked vesicles is critical to the elucidation of the molecular mechanisms underlying synaptic transmission and presynaptic plasticity.
Syntaphilin is a brain-enriched protein that we first characterized as a binding partner of syntaxin-1 (18). Binding of syntaphilin to syntaxin inhibits the binding of syntaxin to SNAP-25 and thus prevents the formation of the SNARE core complex. Functionally, overexpression of syntaphilin in cultured hippocampal neurons inhibits neurotransmitter release; furthermore, injection of the syntaphilin syntaxin-binding peptide into the presynaptic cell body of superior cervical ganglion neurons in culture results in the inhibition of neurotransmission, suggesting that syntaphilin may function as a molecular clamp that controls free syntaxin-1 availability for the assembly of the SNARE complex and thereby regulates synaptic vesicle exocytosis. Syntaphilin also binds to dynamin-1 and inhibits its interaction with amphiphysin, suggesting an inhibitory role for this protein in dynamin-mediated endocytosis (30). The expression of syntaphilin is developmentally regulated, and it is more prominently expressed in the mature rat brain in areas undergoing synaptic plastic changes (31). Our finding that syntaphilin expression is limited to a subset of synapses led us to ask whether it might be an expressionlimited modulator of synaptic activity. Syntaphilin is serinerich (12% of total amino acid content) and contains numerous consensus sites for protein phosphorylation, suggesting that its functional roles on synaptic vesicle recycling are further regulated through phosphorylation. In this study we show that syntaphilin can be phosphorylated by PKA both in vitro and in vivo and that this phosphorylation inhibits its binding to syntaxin-1A. Furthermore, mutation of the phosphorylation site serine 43 to aspartic acid (mimicking a constitutive phosphorylation) annuls the inhibitory effect of syntaphilin on Ca 2ϩdependent exocytosis in PC12 cells. Thus, PKA activation could act as an "off" switch for syntaphilin by blocking its inhibitory functions at the nerve terminal.
In Vitro Phosphorylation Assays-Recombinant His-tagged fulllength syntaphilin was immunoprecipitated from HEK 293 cell lysate using anti-T7-His monoclonal antibody (Novagen) and protein A-Sepharose CL-4B resin (Amersham Biosciences). Equal amounts (ϳ7 pmol) of resin-bound proteins were incubated either with or without the catalytic subunit of PKA to a final concentration of 0.5 unit/l or 100 g of PKA inhibitor peptide PKI (Promega) in phosphorylation buffer (Buffer A: 40 mM Tris-HCl, pH 7.4, 20 mM MgCl 2, 0.1 mM ATP) containing 20 M [␥-32 P]ATP for 1 h at 30°C. The 25-l reactions were terminated by boiling with 12.5 l of Tricine sample buffer (Bio-Rad) containing 4.5% 2-mercaptoethanol and 45 mM dithiothreitol (Sigma). Phosphorylation products were separated by 10 -20% Tricine-SDS-PAGE; the gels were stained with Coomassie Blue, dried, and exposed to x-ray film. Bacte-rially expressed His-tagged syntaphilin 1-130, 130 -205, and 203-469 (50 pmol of each) was bound to nickel-nitrilotriacetic acid beads and phosphorylated as described for the full-length protein. 1 g of syntaphilin mutants of affinity-purified GST-syntaphilin 1-469 were phosphorylated by the same method. For back-phosphorylation assays, glutathione-Sepharose bound in vitro phosphorylated GST-syntaphilin 1-469 and His-syntaphilin immunoprecipitated from HEK 293 cells was subjected to back-phosphorylation by incubating with PKA catalytic subunit (to a final concentration of 0.5 unit/l) and 0.125 mM [␥-32 P]ATP in Buffer A for 1 h at 30°C. The products were separated by 10 -20% Tricine-SDS-PAGE, and then the gels were stained with Coomassie Blue, dried, and exposed to x-ray film.
Stoichiometric Measurements-For phosphorylation time course experiments, 60 pmol of purified recombinant GST-syntaphilin 1-469 was phosphorylated by 80 units of PKA catalytic subunit in Buffer A at 30°C. 6 l of each sample (containing 7.5 pmol of GST-syntaphilin) was removed from the reaction mixture between 5 and 240 min, and these reactions were terminated by the addition of SDS sample buffer and boiling. The products were separated by SDS-PAGE, stained with Coomassie Blue, and exposed to x-ray film. The gel slices containing syntaphilin were excised and scintillation-counted, and the molar ratio of inorganic phosphate incorporated per mol of syntaphilin was calculated and plotted as a function of time.
In Vitro Binding Experiments-Equal amounts (5 g) of GST-syntaphilin 1-469 or GST alone were immobilized on glutathione-Sepharose beads (Amersham Biosciences). The resin-bound proteins were incubated either with or without the catalytic subunit of PKA (247.5 units/reaction) in Buffer A at 30°C for 3 h under continuous agitation. After washing with TBST/PI buffer (50 mM TBS, pH 7.4, 0.1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 1 mM leupeptin, and 1 mM aprotinin), one-fifth of each phosphorylation reaction was subjected to back-phosphorylation to verify the initial phosphorylation. One-third of each reaction was incubated with equal volumes of bacterial lysates containing syntaxin-1A in TBST/PI on a microtube rotator at 4°C for 3 h. After washing with TBST/PI, the bound complexes were separated by SDS-PAGE, transferred onto nitrocellulose membranes, and immunoblotted with anti-syntaxin-1 (10H5) antibody. The membranes were stripped in stripping buffer (62.5 mM Tris-HCl, pH 7.5, 20 mM dithiothreitol, and 1% SDS) at 60°C for 20 min and blotted with anti-GST antibody (Pierce). Horseradish peroxidase-conjugated secondary antibodies and ECL chemiluminescence (Amersham Biosciences) were used to visualize the bands. For the binding dose-response curve, GST-syntaphilin 1-469 immobilized on glutathione-Sepharose beads was divided into two parts and incubated either with or without the catalytic subunit of PKA (410 units/reaction) in Buffer A at 30°C for 3 h under continuous agitation. After washing with TBST/PI, each phosphorylation reaction was divided into seven equal aliquots, and each part was incubated with increasing amounts of purified His-syntaxin-1A (0, 15, 30, 50, 100, 150, and 450 pmol, respectively) and 200 g of bovine serum albumin for blocking nonspecific binding in TBST/PI on a microtube rotator at 4°C for 3 h. After washing with TBST/PI, the bound complexes were separated by SDS-PAGE, transferred onto nitrocellulose membranes, and immunoblotted with anti-syntaxin-1 (10H5) antibody, followed by stripping, and blotting with anti-GST antibody. For quantitative analysis, the binding intensities were measured with National Institutes of Health Image 1.61. Care was taken during exposure of the ECL film to ensure that all readings were in the linear range of a standard. The percentage of binding relative to the nonphosphorylated or wild type syntaphilin was calculated based on standard curves. Student's t tests were performed, and the results are presented as the means Ϯ S.E. of three or four independent experiments.
Preparation of Synaptosomes, Stimulation of PKA in Vivo, and Immunoprecipitation-Rat brain synaptosomes were prepared by differential and discontinuous Percoll gradient centrifugation (32). Briefly, whole rat brains were homogenized in ice-cold 21.9% sucrose buffer, pH 7.4. The homogenates were centrifuged in a Beckman GSR-6RHT centrifuge at 3,000 rpm for 10 min. The supernatant was placed on top of Percoll gradients (23,15,10, and 3% in sucrose buffer) and spun in an SS34 rotor at 17,250 rpm for 5 min. The synaptosome bands between the 15 and 23% gradients were collected, mixed with Wash Buffer (122.8 mM NaCl, 5 mM KCl, 1.15 mM NaH 2 PO 4 , 20 mM PIPES, 0.1% D(ϩ)-glucose, pH 6.8), and then spun in an SS34 rotor at 11,000 rpm for 15 min. The synaptosomes were resuspended in Neurobasal medium supplemented with B-27 (both from Invitrogen), then divided into two equal volumes, and incubated in the presence of either 50 M of the cAMP analogue Sp-5,6-DCl-cBIMPS (BIMPS) or Me 2 SO alone (vehicle control) at 37°C for 1 h. The samples were then lysed in lysis buffer supplemented with phosphatase inhibitors (50 mM TBS, pH 7.4, 1% Triton X-100, 0.5% deoxycholic acid, 1 mM phenylmethylsulfonyl fluoride, 1 mM leupeptin, 1 mM aprotinin, 1 mM EDTA, 1 mM EGTA, 1 mM ␤-glycerophosphate, 1 mM sodium vanadate, 100 nM cyclosporin A, 10 nM okadaic acid; all from Sigma) at 4°C for 30 min and centrifuged by 13,000 rpm for 20 min. The protein concentrations were determined by the BCA protein assay (Pierce). Solubilized proteins from both BIMPS-and Me 2 SO-treated synaptosomes (300 g each) were incubated with anti-syntaxin-1 (10H5) antibody or normal mouse IgG as control in 500 l of TBST/PI supplemented with phosphatase inhibitors overnight at 4°C. Immune complexes were resolved by SDS-PAGE, and co-precipitated syntaphilin was detected by immunoblotting with syntaphilin polyclonal antibody. Phosphorylation of syntaphilin in Me 2 SOand BIMPS-treated synaptosomes was detected by immunoblotting with the phospho-specific syntaphilin antibody anti-pS43. For semi-quantitative analysis, the intensity of the syntaphilin signals was measured with National Institutes of Health Image 1.61, and the relative amounts were calculated using a linear standard curve of syntaphilin from the synaptosomal lysates using a paired Student's t test for statistical analysis.
High K ϩ -stimulated Exocytosis in PC12 Cells-Exocytosis assays were generally conducted as previously described, with a few modifications (33). Briefly, PC12 cells (2 ϫ 10 6 cells grown in 35-mm collagencoated dishes (Corning Inc.)) were co-transfected the next day with a plasmid expressing human GH (pXGH5) along with 3.6 g of empty pcDNA vector or pcDNA vector encoding either wild type syntaphilin or its mutant by using LipofectAMINE 2000 in OPTI-MEM (both from Invitrogen). After 48 h of incubation in RPMI culture medium supplemented with 10% horse serum, and 5% fetal bovine serum (all from ATCC) and 100 unit/ml penicillin G and 100 g/ml streptomycin (both from Invitrogen) at 37°C in 10% CO 2 , the cells were lifted and replated (split from one 35-mm dish into two 22-mm dishes). 24 h after replating, the cells were subjected to exocytosis assays by incubating in low K ϩ (basal) and high K ϩ (stimulated) release conditions. Release buffers consisted of low (5.6 mM) or high (56 mM) KCl solutions in 145 mM NaCl, 2.2 mM CaCl 2 , 0.5 mM MgCl 2 , 5.6 mM glucose, and 15 mM HEPES, pH 7.4, at room temperature for 30 min. The medium was then removed and centrifuged by 13,000 rpm at 4°C for 5 min. Human growth hormone (hGH) in the supernatant was taken as secreted hGH. The cells from the dishes were harvested in Tris-buffered saline, pH 7.4, containing protease inhibitors, 1 mM EDTA, 0.5% Triton X-100, and added to the centrifuged pellet of the secretion medium. Following a 20-min incubation at 4°C on a microtube rotator, the cell lysates were centrifuged at 13,000 rpm for 20 min. The supernatant of this step was taken as the intracellular hGH that was not secreted after K ϩ stimulation. hGH levels in the various samples were measured by the hGH enzyme-linked immunosorbent assay kit (Roche Applied Science) according to the manufacturer's instructions. The percentage of hGH released was calculated as the hGH in supernatant divided by the total hGH obtained from solubilized cells plus hGH in the supernatant. All of the experiments were carried out in duplicate or triplicate each time. The statistical analyses were performed with the Student's t test, and the results are presented as the means Ϯ S.E. of five independent experiments.

RESULTS
In Vitro Phosphorylation of Syntaphilin by PKA-Our previous studies indicate that syntaphilin is a developmentally regulated and expression-limited protein capable of inhibiting both synaptic vesicle exocytosis and endocytosis (18,30,31). Syntaphilin is serine-rich (12% of total amino acid content) and contains numerous sites for protein phosphorylation, suggesting that its role is further regulated through phosphorylation. Because PKA activation induces vesicle cycling at both previously silent cerebellar granule cell synapses (34) and in synapses of hippocampal cultures (35), we wondered whether PKA phosphorylation of syntaphilin could influence its inhibitory functions at the nerve terminal via the cAMP signaling pathway. To address this question, we examined the ability of recombinant syntaphilin to serve as a substrate for PKA. 7 pmol of immunoprecipitated His-tagged syntaphilin expressed in HEK 293 T cells was incubated with PKA catalytic subunit in the presence of [␥-32 P]ATP. The reactions were terminated by adding SDS sample buffer and heating, and the products were separated by SDS-PAGE and stained with Coomassie Blue to verify that equal amounts of protein were loaded in each lane, and the gel was dried and exposed to x-ray film to detect 32 P incorporation. As shown in Fig. 1A, syntaphilin functions as a substrate for PKA, and its phosphorylation is blocked by the pseudosubstrate PKA inhibitor peptide PKI.
To map phosphorylation sites in syntaphilin, we examined the capacity of syntaphilin truncated mutants to serve as PKA substrates. We incubated 50 pmol of purified His-tagged syntaphilin truncated mutants including 1-130, 130 -205, and 203-469 with PKA and found that syntaphilin amino-terminal domain (1-130) was efficiently phosphorylated in vitro by PKA (Fig. 1B, upper panel). In contrast, both the coiled-coil domain (130 -205) and the carboxyl half (203-469) could not serve as a PKA substrate under our experimental conditions. Immunoblotting of the reaction mixtures with anti-T7-His antibody demonstrated that approximately equal amounts of proteins were used for in vitro phosphorylation (Fig. 1B, lower panel).
We then conducted a sequence search against the Scansite FIG. 1. Phosphorylation of syntaphilin by PKA in vitro. A, PKA phosphorylation of syntaphilin. 7 pmol of T7-His-tagged syntaphilin (His-philin) was immunoprecipitated (IP) by immobilized anti-T7-His antibody and then incubated with [␥-32 P]ATP in a 35-l reaction in the absence (Ϫ) or presence (ϩ) of the catalytic subunit of PKA or PKA inhibitor peptide (PKI), as indicated. The phosphorylation products were separated by SDS-PAGE, stained with Coomassie Blue (lower panel), dried, and exposed to x-ray film (upper panel). B, PKA phosphorylation of syntaphilin truncated mutants. Approximately 50 pmol of purified Histagged syntaphilin truncated mutant proteins was phosphorylated by PKA as described under "Experimental Procedures." The reactions were separated by SDS-PAGE, dried, and then exposed to x-ray film for autoradiography (upper panel). Equal amounts of the mutant syntaphilin proteins were further confirmed by immunoblotting with anti-T7-His tag antibody (lower panel). The numbers indicate the amino acid boundaries of each truncated mutant.
Motif Scanner Program (36), and several consensus PKA phosphorylation residues were predicted between amino acids 1 and 130. We focused on three serine residues at positions 43, 56, and 64 that show high potential as PKA targets. To identify the amino acid residue(s) phosphorylated by PKA, we used sitedirected mutagenesis to generate syntaphilin mutants where the serine at positions 43, 56, or 64 was substituted with alanine. Approximately 1 g of the GST-syntaphilin mutants were incubated with PKA catalytic subunit at 30°C for 1 h, separated by SDS-PAGE, stained with Coomassie Blue ( Fig.  2A, lower panel), and then exposed to x-ray film ( Fig. 2A, upper  panel). To quantitate 32 P incorporation, the gel slices corresponding to phosphorylated syntaphilin bands were excised and scintillation counted. Although the S64A mutant was as efficiently phosphorylated as the wild type syntaphilin, the 32 P incorporation into S43A and S56A mutants was significantly decreased. Double mutation of S43A and S56A resulted in further reduction of the phosphorylation signal, suggesting that both the Ser 43 and Ser 56 residues serve as primary PKA phosphorylation sites in syntaphilin. Furthermore, we confirmed the site-directed mutagenesis results by mass spectrometry analysis using the phosphorylated syntaphilin and found that serine 43 is indeed a primary target for PKA phosphoryl-ation in vitro. The surrounding sequence of syntaphilin serine 43 (RTHS 43 ) and serine 56 (RRTS 56 ) match one of the consensus sequences for PKA (RRX(S/T), RX(S/T), and RX 2 (S/T)).
To further confirm that PKA phosphorylates only two residues in syntaphilin, we performed stoichiometric analysis of phosphorylation for purified recombinant syntaphilin protein.
The reactions included 60 pmol of GST-syntaphilin and [␥-32 P]ATP. 6 l of samples containing 7.5 pmol of GST-syntaphilin were removed from the reaction mixture at various time points between 5 and 240 min, and these reactions were terminated by the addition of SDS sample buffer and boiling. The products were separated by SDS-PAGE, and inorganic P incorporation was quantitated by scintillation counting of excised gel slices corresponding to phosphorylated syntaphilin bands. The stoichiometry values are expressed as the ratios of moles of inorganic phosphate (P i ) incorporated per mole of syntaphilin protein and plotted against reaction time (Fig. 2B). Under conditions optimized for maximal phosphorylation, we found that PKA phosphorylation of syntaphilin reached a plateau level after 120 min of incubation at 30°C, and the maximal stoichiometry is nearly 2, consistent with the two primary PKA sites identified by the mutagenetic studies.
In Vivo Phosphorylation of Syntaphilin by PKA-Although our biochemical experiments showed that PKA incorporates 32 P into syntaphilin, the conditions used for in vitro phosphorylation may not reflect conditions found in the native cellular environment. In addition, treatment of proteins with detergent for solubilization may expose sites that are normally not available for phosphorylation in vivo. Therefore, to investigate in vivo phosphorylation, we performed back-phosphorylation assays. We transfected a T7-His-tagged syntaphilin construct into HEK 293 T cells where no endogenous syntaphilin could be detected. After 2 days the transfected cells were incubated with the nonhydrolyzable and cell-permeable cAMP analogue BIMPS (50 M) or Me 2 SO as vehicle control at 37°C for 30 min. BIMPS is better than its parent compound, Sp-cAMPS, in several respects including higher metabolic stability, greater lipophilicity, and excellent membrane permeability (37). In this procedure, endogenous phosphate is incorporated in vivo into exogenously expressed syntaphilin after stimulation with the cAMP analogue BIMPS. The cells from stimulated and nonstimulated HEK 293 T cell cultures were solubilized in 1% Triton X-100 containing TBS supplemented with protease and phosphatase inhibitors. Syntaphilin was then immunoprecipitated with an anti-T7-His antibody and then processed for in vitro phosphorylation (back-phosphorylation) with purified PKA to incorporate 32 P into syntaphilin that was left unphosphorylated in vivo after stimulation. In this protocol, a decrease in back-phosphorylation reflects in vivo PKA phosphorylation of syntaphilin in the transfected HEK 293 T cells in response to stimulation with the cAMP analogue. As expected for a direct PKA phosphorylation event on syntaphilin in vivo, the amount of 32 P incorporated during back-phosphorylation was significantly reduced when transfected cells were stimulated in vivo with BIMPS (Fig. 3A, upper panel). Coomassie Blue staining (Fig. 3A, lower panel) corroborated the identity of the phosphorylated band (68 kDa) as syntaphilin and was used to normalize back-phosphorylation to the amount of syntaphilin immunoprecipitated by the anti-T7-His antibody. Quantitative analysis of the scintillation counting data of 32 P incorporated syntaphilin protein bands isolated from gels of three independent experiments showed that stimulation with the cAMP analogue BIMPS significantly reduced back-phosphorylation of syntaphilin to 58 Ϯ 5% of the Me 2 SO-treated control value (Fig. 3B).
To elucidate the possible in vivo occurrence and physiological GST-syntaphilin or mutants was incubated with PKA phosphorylation reagents at 30°C for 1 h, separated by SDS-PAGE, stained with Coomassie Blue (lower panel), dried, and exposed to x-ray film for autoradiography (upper panel). B, stoichiometry of PKA phosphorylation for syntaphilin. Equal amounts (7.5 pmol) of syntaphilin were phosphorylated with PKA for the indicated time periods. The reactions were terminated by the addition of SDS sample buffer and boiling, the products were separated by SDS-PAGE, and the gel was stained with Coomassie Blue and then exposed to x-ray film for 1 h. Gel slices containing labeled syntaphilin were excised and scintillation counted, and then the moles of inorganic phosphate (P i ) incorporated per mole of syntaphilin protein were calculated and plotted against the reaction time.
relevance of cAMP-dependent phosphorylation of syntaphilin on serine 43, we raised an antibody that selectively recognizes the phosphorylated form of the protein. A peptide corresponding to LTRTH(pS)LMAC (residues 38 -45) of syntaphilin with a phosphoserine at position 43 was coupled to keyhole limpet hemocyanin and used to generate an immune response in rabbit. The resulting polyclonal antiserum was passed over columns carrying an unrelated peptide and the related non-phosphopeptide to remove nonspecific antibodies. Finally, the antiserum was affinity-purified by binding to the phosphorylated peptide. The affinity-purified anti-Ser(P) 43 polyclonal antibody (anti-pS43) proved to be highly specific for the PKAphosphorylated form of syntaphilin. In immunoblot experiments, PKA-phosphorylated and unphosphorylated GST-syntaphilin were resolved side by side by SDS-PAGE and probed with anti-pS43 or anti-syntaphilin (anti-philin), a polyclonal antibody directed against syntaphilin (18). Anti-pS43 antibody was able to recognize 100 ng of the GST-syntaphilin after in vitro PKA phosphorylation and did not cross-react with same amount of unphosphorylated syntaphilin on the same blot (Fig.  4A, left panel). Although anti-philin detected equally well the phosphorylated and unphosphorylated GST-syntaphilin, anti-pS43 showed selective recognition of the phosphorylated protein (Fig. 4A, right panel). The specificity of anti-pS43 antibody was further demonstrated by the fact that it recognized phosphorylated enhanced green fluorescent protein-syntaphilin expressed in COS cells followed by stimulation with the cAMP analogue BIMPS but did not cross-react with the S43A and S43D mutants of eGFP-syntaphilin (data not shown).
Having established the specificity of anti-pS43 antibody in recognizing only the phosphorylated form of syntaphilin, we sought to investigate PKA phosphorylation of syntaphilin in nerve terminal-enriched synaptosomes and to confirm whether serine 43 is one of the primary phosphorylation sites in vivo. Freshly prepared synaptosomes were stimulated with either 50 M BIMPS or Me 2 SO alone as a control at 37°C for 1 h and then solubilized in 1% Triton X-100, 0.5% deoxycholic acid in TBS with protease and phosphatase inhibitors. Equal amounts of total proteins from stimulated or nonstimulated synaptosomes were separated with SDS-PAGE, and syntaphilin was detected by immunoblotting with both anti-philin and anti-pS43 antibodies. In control (Me 2 SO) synaptosomes, a low basal level of phosphorylated syntaphilin was observed that was significantly increased (133.5 Ϯ 3.8% relative to control, n ϭ 4, p Ͻ 0.05) after PKA activation with BIMPS (Fig. 4B, left panel). Immunoblotting with anti-philin antibody showed equal loading of syntaphilin from Me 2 SO-and BIMPS-treated synapto- as vehicle control at 37°C for 30 min and then solubilized with Triton X-100. Syntaphilin (His-philin) was then immunoprecipitated (IP) with an anti-T7-His antibody and then processed for in vitro phosphorylation (back-phosphorylation) with purified PKA to incorporate 32 P into syntaphilin that was left unphosphorylated after in vivo stimulation. Note that the amount of 32 P incorporated during back-phosphorylation was significantly reduced when transfected cells were stimulated in vivo with BIMPS (upper panel). Coomassie Blue staining (lower panel) was used to normalize back-phosphorylation to the amount of syntaphilin immunoprecipitated. B, normalized percentages of syntaphilin back-phosphorylation from unstimulated versus stimulated (BIMPS) HEK 293 cells after transfection. Bar, mean Ϯ S.E., n ϭ 3. Back-phosphorylation was decreased to 58 Ϯ 5% (p Ͻ 0.05) of the Me 2 SO treated control value, indicating that syntaphilin is phosphorylated by endogenous PKA in vivo in its native conformation.

FIG. 4. cAMP-dependent phosphorylation of syntaphilin in synaptosomes.
A, the affinity-purified anti-pS43 antibody is specific for the serine 43-phosphorylated form of syntaphilin. Bacterially expressed GST-syntaphilin (philin) was incubated under phosphorylation conditions in the presence (ϩ) or absence (Ϫ) of the catalytic subunit of PKA at 30°C for 3 h. 100 ng of either nonphosphorylated or phosphorylated GST-philin was resolved by SDS-PAGE and blotted with either anti-pS43 (left panel) or anti-philin antibody (right panel). Although anti-philin recognized both phosphorylated and nonphosphorylated syntaphilin, anti-pS43 detected only the phosphorylated protein. B, the affinity-purified anti-pS43 antibody recognizes phosphorylated syntaphilin in rat brain synaptosomes following PKA activation. Freshly prepared synaptosomes were incubated in Me 2 SO (Ϫ) or 50 M BIMPS (ϩ) at 37°C for 1 h. After solubilization with 1% Triton X-100, 0.5% deoxycholic acid, synaptosomal lysates (100 g of total proteins for anti-pS43 blotting and 25 g of total proteins for anti-philin blotting) were separated by SDS-PAGE and then immunoblotted with either anti-pS43 (left panel) or anti-philin (right panel) antibodies. Note that detection of syntaphilin by the anti-philin antibody confirmed equal protein loading. somes (Fig. 4B, right panel), suggesting that syntaphilin in nerve terminals could be phosphorylated at serine 43 by endogenous PKA in response to cAMP analogue stimulation. All together, the results of our back-phosphorylation studies from the transfected HEK 293 cells and the immunoblot with the anti-pS43 antibody from synaptosomal preparations indicate that syntaphilin is a direct target of cAMP-dependent phosphorylation in vivo.
Biochemical Effect of PKA Phosphorylation of Syntaphilin-Our previous studies have shown that syntaphilin competes with SNAP-25 for binding to syntaxin-1A and thereby prevents formation of the SNARE complex. To determine whether PKA phosphorylation of syntaphilin could modulate its interaction with syntaxin-1A, we performed in vitro binding studies with phosphorylated syntaphilin. Glutathione-Sepharose beads with immobilized GST-syntaphilin or GST alone were incubated with phosphorylation buffer in the presence or absence of PKA for 3 h. After extensive washing, one-fifth aliquots of each reaction were back-phosphorylated in the presence of [␥-32 P]ATP to verify the initial phosphorylation (Fig. 5A, right  panel). The rest of the beads with immobilized PKA-phosphorylated or nonphosphorylated syntaphilin or GST control were FIG. 5. Biochemical effect of syntaphilin phosphorylation by PKA. A, PKA phosphorylated syntaphilin exhibited decreased binding to syntaxin-1A. Glutathione-Sepharose beads with immobilized GST-syntaphilin or GST alone were incubated with phosphorylation buffer in the presence or absence of PKA for 3 h. After extensive washing, one-fifth aliquots of each reaction were back-phosphorylated in the presence of [␥-32 P]ATP and PKA to verify the initial phosphorylation (right panel). The beads with immobilized PKA-phosphorylated (GST-philin-P) or nonphosphorylated syntaphilin (GST-philin) or GST control were then incubated with recombinant syntaxin-1A. The binding complexes were separated by SDS-PAGE and processed for immunoblotting with anti-syntaxin antibody (upper left panel), followed by stripping and secondary blotting with anti-GST antibody (lower leftpanel). B, relative binding of phosphorylated syntaphilin to syntaxin-1A. The binding intensities were semi-quantified using NIH Image. The relative binding was calculated based on a linear standard curve of syntaxin-1A aliquots and expressed as a percentage of syntaxin-1A binding to nonphosphorylated GST-philin. PKA phosphorylation of syntaphilin significantly reduces its binding to syntaxin-1A (43 Ϯ 7%, mean Ϯ S.E., n ϭ 4, p Ͻ 0.05) relative to nonphosphorylated syntaphilin. C, dose-response curves of syntaxin binding activity of both PKA phosphorylated and unphosphorylated syntaphilin. 10 pmol of GST-philin with or without PKA phosphorylation was incubated with increasing amounts of purified His-syntaxin-1A for 3 h as indicated. After washing, the bound complexes were separated by SDS-PAGE and immunoblotted with anti-syntaxin-1 (10H5) antibody, followed by stripping and blotting with anti-GST antibody. Dose-response curves were generated using semi-quantitative analysis of the blots, and the results indicate that phosphorylation of syntaphilin by PKA decreases its maximal binding capacity to syntaxin-1A by 37% Ϯ 3 (mean Ϯ S.E.). The curve shown here is a representative of three experiments. D, replacing serine 43 of syntaphilin with aspartic acid (S43D) mimics the effect of the PKA phosphorylation on its binding to syntaxin-1A. Approximately 0.2 g of GST-philin wild type, S43A, or S43D mutants immobilized on glutathione-Sepharose beads were incubated with equal amounts of cell lysates containing syntaxin-1A for 3 h. After extensive washing the bound proteins were separated by SDS-PAGE and blotted with anti-syntaxin antibody. The binding intensities were semi-quantified using NIH Image. The relative levels of syntaxin-1A bound to syntaphilin mutants S43D and S43A were calculated based on a linear standard curve of syntaxin-1A aliquots. Note that although the S43A mutation had no significant effect on the binding properties, the S43D mutation decreased syntaphilin binding to syntaxin-1A by 32 Ϯ 5% (mean Ϯ S.E., n ϭ 3, p Ͻ 0,05) relative to normalized wild type syntaphilin. incubated with recombinant syntaxin-1A. The binding complexes were then separated by SDS-PAGE and processed for immunoblotting with anti-syntaxin antibody (10H5), followed by stripping and secondary blotting with anti-GST antibody. As shown in Fig. 5A, PKA phosphorylation of syntaphilin exhibited decreased binding to syntaxin-1A compared with the nonphosphorylated syntaphilin. Semi-quantitative analysis revealed that PKA phosphorylation of syntaphilin significantly reduces its binding to syntaxin-1A (43 Ϯ 7%, n ϭ 4, p Ͻ 0.05) relative to nonphosphorylated syntaphilin (Fig. 5B).
To further characterize the binding capability of phosphorylated syntaphilin, we performed dose-dependent syntaxin-1A binding experiments with either phosphorylated or unphosphorylated syntaphilin. Equal amounts of GST-syntaphilin were incubated either with or without the catalytic subunit of PKA for 3 h. After extensive washing, each reaction was divided into seven aliquots, and then each aliquot containing approximately ϳ10 pmol of GST-syntaphilin was incubated with increasing amounts of purified His-syntaxin-1A for 3 h as indicated in Fig.  5C. After washing, the bound complexes were separated by SDS-PAGE and immunoblotted with anti-syntaxin-1 antibody, followed by stripping and blotting with anti-GST antibody. Dose-response curves generated using semi-quantitative analysis of the blots showed that phosphorylation of syntaphilin by PKA decreases its maximal binding capacity to syntaxin-1A by 37% Ϯ 3 (Fig. 5C).
Our mutagenetic studies and mass spectrometry analysis demonstrated that serine 43 of syntaphilin is one of the primary sites for PKA phosphorylation, which is located within the amino-terminal proline-rich region and adjacent to the binding site (coiled-coil domain) for syntaxin. To test whether the introduction of a negatively charged residue at the site has any effect on the binding of syntaphilin to syntaxin-1A, we attempted to mimic complete phosphorylation of serine 43 by mutating it to a negatively charged aspartic acid residue (S43D) or to a noncharged alanine residue (S43A) as a control and then analyzed the resultant binding properties of the mutated proteins. 0.2 g of GST-syntaphilin wild type, S43A, or S43D mutant proteins immobilized on glutathione-Sepharose beads were incubated with equal amounts of cell lysates containing syntaxin-1A for 3 h. After extensive washing, the bound proteins were separated by SDS-PAGE and blotted with antisyntaxin antibody. Our binding studies with quantitative analysis showed that although the S43A mutation had no significant effect on the binding properties, the S43D mutation decreased syntaphilin binding to syntaxin-1A by 32 Ϯ 5% (n ϭ 3, p Ͻ 0,01; Fig. 5D).
Because our in vitro binding results indicated that the cAMPdependent phosphorylation of syntaphilin decreases its binding capacity to syntaxin-1A, we next sought to determine whether this phosphorylation event affects the association of syntaphilin with syntaxin-1A in vivo by co-immunoprecipitation from synaptosomal lysates with an anti-syntaxin (10H5) antibody. The activation of PKA in synaptosomes with BIMPS significantly decreased immunoprecipitation of syntaphilin with syntaxin-1 (Fig. 6A, left panel). BIMPS-stimulated phosphorylation of syntaphilin at serine 43 was confirmed by immunoblotting with anti-pS43 antibody of both Me 2 SO-and BIMPS-treated synaptosomal lysates (Fig. 6A, right panel). Quantitative analysis of three independent experiments showed that the BIMPS stimulation of synaptosomes significantly reduced syntaphilin co-immunoprecipitated with syntaxin-1 to 29 Ϯ 18% (n ϭ 3, p Ͻ 0.05) relative to Me 2 SO treated controls (Fig. 6B). Therefore, both in vitro binding assays using the recombinant proteins and co-immunoprecipitation studies from purified synaptosomes indicate that the cAMP-dependent phosphorylation of syntaphilin at serine 43 significantly decreased its binding capacity to syntaxin-1.
Physiological Effect of Syntaphilin and Its S43D Mutant on Ca 2ϩ -dependent Exocytosis in PC12 Cells-In our earlier study, we found that syntaphilin overexpression in cultured hippocampal neurons was associated with a decrease in excitatory postsynaptic current amplitude and the frequency of spontaneous events (18). Both were attributed to an inhibition of SNARE complex formation by syntaphilin and thus of synaptic vesicle exocytosis. Given our current results, it is possible that PKA phosphorylation of syntaphilin at serine 43 has a significant effect on its binding to syntaxin-1A, thus turning off the inhibitory function of syntaphilin through cAMP-dependent signal cascades. To determine whether this phosphorylation event is implicated in the regulation of Ca 2ϩ -dependent exocytosis, the effect of expression of syntaphilin or its mutant on SNARE-mediated vesicle fusion was evaluated using a hGH release assay in transfected PC12 cells (38,39). In transfected PC12 cells hGH is concentrated in large dense core vesicles and undergoes Ca 2ϩ -dependent exocytosis in response to depolarization by high K ϩ . hGH secreted into the culture medium following high K ϩ -induced depolarization can then be used as a measure of large dense core vesicle release, allowing for a functional evaluation of SNAREs and other presynaptic proteins. Syntaphilin pcDNA constructs including wild type and S43D mutant or pcDNA vector alone as a control were cotransfected with the vector pXGH5 encoding hGH. Stimulated FIG. 6. In vivo PKA phosphorylation of syntaphilin inhibits its interaction with syntaxin-1 in synaptosomes. A, synaptosomes were incubated without (Ϫ) or with (ϩ) 50 M BIMPS at 37°C for 1 h and then lysed with 1% Triton X-100, 0.5% deoxycholic acid. Solubilized synaptosomal lysates (300 g of total protein) were incubated with anti-syntaxin antibody (␣-stx-1) or normal mouse IgG (IgG) as control. The immunoprecipitates were separated by SDS-PAGE. Co-immunoprecipitation of syntaphilin was detected by anti-philin antibody (left panel). In vivo PKA phosphorylation of syntaphilin at serine 43 in control (Ϫ) and BIMPS (ϩ) treated synaptosomal lysates was confirmed by immunoblotting with anti-pS43 antibody (right panel). Approximately equal amounts of syntaxin-1 immunoprecipitated were confirmed by immunoblot with anti-syntaxin antibody. B, normalized percentages of syntaphilin immunoprecipitated by anti-syntaxin antibody from BIMPS untreated (Ϫ) versus treated (ϩ) synaptosomes. The data are expressed as the means Ϯ S.E. BIMPS stimulation reduced the co-precipitation of syntaphilin with syntaxin-1 to 29 Ϯ 18% (n ϭ 3, p Ͻ 0.05) relative to syntaphilin precipitated from untreated synaptosomes.
hGH secretion was determined after 30 min of incubation with a high K ϩ buffer, whereas basic hGH secretion was defined in a low K ϩ buffer under the same conditions. Released hGH in the medium was expressed as a percentage of total (released and intracellular) hGH measured by enzyme-linked immunosorbent assay. In cells transfected with vector alone, high K ϩ -induced depolarization was associated with a release of 13.6 Ϯ 1.4% of the total cellular hGH (Fig. 7A). Consistent with its biochemical properties, transfection of PC12 cells with syntaphilin caused a decrease in hGH release (10.5 Ϯ 1.2%, p Ͻ 0.05), whereas no such inhibition was seen in cells transfected with a syntaphilin mutant (⌬CC) lacking the syntaxin-1Abinding domain (data not shown). These studies demonstrate congruence between the biochemical and functional properties of syntaphilin binding to syntaxin-1A and raise the possibility that syntaphilin could attenuate synaptic activity by inhibiting SNARE-mediated fusion.
To normalize results from repeated experiments, high K ϩstimulated hGH release from PC 12 cells transfected with control vector alone was set at 100%. Overexpression of wild type syntaphilin caused a significant decrease (by more than 25%) in stimulated hGH release without affecting basal hGH secretion, which is consistent with our previous findings that overexpression of syntaphilin in cultured hippocampal neurons inhibits neurotransmitter release and injection of syntaphilin into the presynaptic superior cervical ganglion neurons in culture results in inhibition of neurotransmission. Because our biochemical studies showed that both PKA phosphorylation of syntaphilin as well as S43D mutation of this protein reduce the interaction of syntaphilin with syntaxin-1A, we asked whether the S43D mutant could reverse the effect of wild type syntaphilin on release. Our studies demonstrated that overexpression of S43D mutant, which mimics 100% phosphorylation of this residue, did not show significant inhibitory effect on high K ϩinduced hGH secretion (Fig. 7B). Approximately equal expression levels of both syntaphilin wild type and its S43D mutant in transfected PC12 cells were confirmed by immunoblotting the transfected lysates with anti-syntaphilin antibody (Fig. 7B). No significant inhibitory effect of the syntaphilin S43D mutation on high K ϩ -stimulated secretion further supports the notion that syntaphilin is a highly regulated protein capable of inhibiting synaptic vesicle exocytosis and its inhibitory activity could be "turned off" via the cAMP-PKA signaling pathway.

DISCUSSION
Our previous results provided both biochemical and physiological evidence that syntaphilin competes with SNAP-25 for binding to syntaxin-1 and inhibits SNARE complex formation. Transient overexpression of syntaphilin in cultured hippocampal neurons significantly reduced neurotransmitter release. Introduction of syntaphilin fragments containing the syntaxinbinding sequence into presynaptic superior cervical ganglion neurons in culture effectively inhibited synaptic transmission. We interpreted this inhibitory effect as syntaphilin acting as a syntaxin-1 clamp controlling the formation of the functional SNARE fusion complex and consequently inhibiting the process of synaptic vesicle exocytosis. The results of the current study demonstrate that syntaphilin can be phosphorylated by PKA both in vitro and in vivo and that this phosphorylation inhibits the binding of this protein to syntaxin-1A. Furthermore, mutation of the dominant phosphorylation site serine 43 to aspartic acid (mimicking 100% of phosphorylation) annuls the inhibitory effect of syntaphilin on Ca 2ϩ -dependent exocytosis in PC12 cells. Our studies suggest that syntaphilin can serve as a target for PKA at the synapse, and the phosphorylation of syntaphilin by PKA appears to modulate neurotransmitter release through a cAMP-dependent signal cascade. Thus, PKA phosphorylation might act as an off switch for syntaphilin, therefore facilitating the input of free syntaxin-1 into the SNARE complexes.
At many synapses, presynaptic activation of PKA can enhance the release of neurotransmitter per action potential (20 -22, 24). The cAMP/PKA cascade augments vesicular release at cerebellar granule cell synapses in primary culture (34). Long term potentiation in the mossy fiber pathway in the hippocampal CA3 region appears to be a presynaptic mechanism that is mediated by cAMP-PKA signaling (23,40). In agreement with these functional observations, several SNAREs and their binding proteins have been implicated as PKA targets in the nerve terminal. For example, PKA phosphorylates SNAP-25 both in vivo and in vitro (41,42), and the phosphorylation site threonine 138 lies in the core of the SNARE complex. However, the functional effect implicated in this phosphorylation event on neurotransmitter release still remains to be elucidated. Forskolin induces the phosphorylation of rabphilin-3A, a synaptic vesicle-associated protein, in CA3 hippocampal synaptosomes, FIG. 7. Physiological effect of syntaphilin and its S43D mutant on Ca 2؉ -dependent exocytosis in PC12 cells. A, syntaphilin inhibits stimulated hGH release in PC12 cells. hGH expression vector (pXGH5) was co-transfected with syntaphilin cDNA or vector alone into PC12 cells. hGH release was calculated by dividing hGH measured in the culture medium by the total hGH (intracellular and extracellular). Both basal release (at low K ϩ ) and high K ϩ -induced release are expressed as the averages Ϯ standard error (n ϭ 5). B, the S43D mutant effectively annuls the inhibitory effect of syntaphilin on high K ϩ -induced secretion in PC12 cells. High K ϩ -induced release of hGH from PC12 cells transfected with empty pcDNA vector (V) is used as a control (100%) to normalize relative release from cells transfected with the His-tagged syntaphilin wild type (W) or S43D mutant (S43D). Immunoblot from transfected PC12 cell extracts with antibody for the T7-His tag confirmed equal expression level of His-tagged syntaphilin wild type and the S43D mutant.
suggesting that it may be a PKA target for inducing mossy fiber long term potentiation in the hippocampus (43)(44)(45). Cysteine string protein, a presynaptic chaperone, is phosphorylated by PKA in vivo (46,47). PKA phosphorylation of snapin increases its binding to SNAP-25 and consequently enhances the association of synaptotagmin with the SNARE complex and increases the release probability of dense core vesicles in chromaffin cells (17,48). Phosphorylation of synapsin Ia by PKA and calciumcalmodulin-dependent kinase I/IV controls its dissociation from synaptic vesicles, which in turn regulates the kinetics of vesicle pool turnover (49,50). PKA phosphorylation of RIM1␣, a presynaptic scaffolding protein and Rab effector, has also been implicated in presynaptic long term potentiation in the cerebellum (51). Thus, the existence of multiple PKA targets in the presynaptic terminal might underlie the robust effects of the cAMP/PKA signal cascade on synaptic vesicle release. Alternatively, phosphorylation of the multiple SNARE regulatory proteins provides a molecular basis for the heterogeneity of PKAinduced synaptic facilitation.
Our physiological results showing that PKA phosphorylation of syntaphilin is involved in the regulation of Ca 2ϩ -dependent exocytosis of hGH from PC12 cells are in agreement with our biochemical findings. Although the expression of wild type syntaphilin in PC12 cells exhibits a significant reduction in high K ϩ -induced hGH release, the S43D mutant reverses its inhibitory effect on exocytosis. By using the S43D mutant to mimic a constitutive phosphorylation of serine 43 instead of direct activation of the cAMP/PKA cascade in PC12 cells, our data demonstrate for the first time that phosphorylation of a single protein at a single residue is able to influence vesicle exocytosis in PC 12 cells. This assay with hGH as a reporter for regulated exocytosis has been widely used for functional studies of presynaptic proteins. Thus, the approach of our current studies excludes the possibility that our observation in PC 12 cells is due to PKA phosphorylation of other synaptic regulatory proteins or activation of downstream protein kinases. Thus, our results suggest that syntaphilin might be a highly regulated PKA target that modulates neurotransmitter release at synapses. Conceivably, PKA activation, which has been shown to induce vesicle release at both previously silent cerebellar granule cell synapses (34) and in synapses of hippocampal slices (29), could act as an off switch for syntaphilin, thus blocking its inhibitory functions at the nerve terminals or activating previously silent synapses in which syntaphilin is present. The studies reported here highlight the potential importance of PKA phosphorylation of syntaphilin in the regulation of synaptic vesicle exocytosis. They provide the foundation for further studies aimed at better understanding the structural and functional aspects of these events. Using phosphopeptide-specific antisera to examine spatial and temporal signal activation of this regulatory pathway in vivo will allow us to better understand the molecular mechanisms underlying PKA-dependent presynaptic modulation.