The Stimulus-induced Tyrosine Phosphorylation of Munc18c Facilitates Vesicle Exocytosis*

Stimulus-induced tyrosine phosphorylation of Munc18c was investigated as a potential regulatory mechanism by which the Munc18c-Syntaxin 4 complex can be dissociated in response to divergent stimuli in multiple cell types. Use of [32P]orthophosphate incorporation, pervanadate treatment, and phosphotyrosine-specific antibodies demonstrated that Munc18c underwent tyrosine phosphorylation. Phosphorylation was apparent under basal conditions, but levels were significantly increased within 5 min of glucose stimulation in MIN6 beta cells. Tyrosine phosphorylation of Munc18c was also detected in 3T3L1 adipocytes and increased with insulin stimulation, suggesting that this may be a conserved mechanism. Syntaxin 4 binding to Munc18c decreased as Munc18c phosphorylation levels increased in pervanadate-treated cells, suggesting that phosphorylation dissociates the Munc18c-Syntaxin 4 complex. Munc18c phosphorylation was localized to the N-terminal 255 residues. Mutagenesis of one residue in this region, Y219F, significantly increased the affinity of Munc18c for Syntaxin 4, whereas mutation of three other candidate sites was without effect. Moreover, Munc18c-Y219F expression in MIN6 cells functionally inhibited glucose-stimulated SNARE complex formation and insulin granule exocytosis. These data support a novel and conserved mechanism for the dissociation of Munc18c-Syntaxin 4 complexes in a stimulus-dependent manner to facilitate the increase in Syntaxin 4-VAMP2 association and to promote vesicle/granule fusion.

Stimulus-induced tyrosine phosphorylation of Munc18c was investigated as a potential regulatory mechanism by which the Munc18c-Syntaxin 4 complex can be dissociated in response to divergent stimuli in multiple cell types. Use of [ 32 P]orthophosphate incorporation, pervanadate treatment, and phosphotyrosine-specific antibodies demonstrated that Munc18c underwent tyrosine phosphorylation. Phosphorylation was apparent under basal conditions, but levels were significantly increased within 5 min of glucose stimulation in MIN6 beta cells. Tyrosine phosphorylation of Munc18c was also detected in 3T3L1 adipocytes and increased with insulin stimulation, suggesting that this may be a conserved mechanism. Syntaxin 4 binding to Munc18c decreased as Munc18c phosphorylation levels increased in pervanadate-treated cells, suggesting that phosphorylation dissociates the Munc18c-Syntaxin 4 complex. Munc18c phosphorylation was localized to the N-terminal 255 residues. Mutagenesis of one residue in this region, Y219F, significantly increased the affinity of Munc18c for Syntaxin 4, whereas mutation of three other candidate sites was without effect. Moreover, Munc18c-Y219F expression in MIN6 cells functionally inhibited glucose-stimulated SNARE complex formation and insulin granule exocytosis. These data support a novel and conserved mechanism for the dissociation of Munc18c-Syntaxin 4 complexes in a stimulus-dependent manner to facilitate the increase in Syntaxin 4-VAMP2 association and to promote vesicle/granule fusion.
In the early 1990s, the discovery of soluble N-ethylmaleimidesensitive factor attachment protein receptor (SNARE) 2 proteins elucidated the molecular basis for synaptic vesicle exocytosis (1,2). Vesicle exocytosis entails the pairing of a vesicle-associated membrane protein (VAMP) SNARE with a binary cognate receptor complex at the target membrane composed of SNAP-25/23 and syntaxin proteins (target SNAREs) to form the SNARE core complex (3)(4)(5)(6)(7). Shortly thereafter, SNARE protein complexes were identified and found to regulate secretory processes in many diverse cell types ranging from yeast and Caenorhabditis elegans to humans, which has led to the general concept that most types of regulated vesicle fusion occur by a common mechanism.
Concurrent with the discovery of SNARE proteins was the discovery of the yeast Sec1 secretory protein, which was found to interact directly with the target SNARE syntaxin (8). Homologs in C. elegans (UNC-18), Drosophila melanogaster (ROP), and mammalian cells (Munc18a-c) were also identified (9 -13). Collectively, the Sec1 and Munc18 protein families are now referred to as "SM" proteins for Sec1 and Munc18. Munc18 proteins are ϳ66 -68 kDa in size and are soluble factors with no transmembrane domain (14). They are found localized to the cytosol and also to the plasma membrane through high affinity binding to their cognate syntaxin (15,16). Munc18a (also known as Munc18-1/neural Sec1/rat brain Sec1) was demonstrated to interact with Syntaxin 1 in a manner mutually exclusive of the other SNARE core complex proteins and is expressed in neurons and islet cells (14,17), whereas Munc18b and Munc18c were found to be ubiquitously expressed (18). Munc18a and Munc18b share binding with Syntaxins 1-3, whereas only Munc18c binds Syntaxin 4 with high affinity (9,13,18). Null mutations in the genes encoding SM proteins are lethal, indicating their essentiality. However, the mechanistic/functional role(s) of SM proteins in vesicle exocytosis has remained unclear.
Crystallographic and NMR structural analyses of the Munc18a-Syntaxin 1A complex suggest that the Munc18 protein serves to hold the four helical domains of the syntaxin protein in a "closed" conformational state in a 1:1 Munc18/syntaxin molar ratio. It has been proposed that, upon stimulation, Munc18 releases syntaxin and assists in the transition of syntaxin into its "open" conformation for subsequent interaction with SNAP-25 and VAMP2 in SNARE core complexes (19 -21). Structural studies of the SM protein family have revealed a similar overall structure in which a small folded N-terminal domain mediates their interaction with plasma membrane syntaxins (22,23). The C-terminal domain of Munc18c carries out the effector function that appears to be essential for fusion, and a particular loop 2/3 domain within this region may be critical for this effector function (22,23). Consistent with this, we have shown previously that an inhibitory peptide directed at this region or a single point mutation within it disrupts the interaction of Munc18c with Syntaxin 4 in 3T3L1 adipocytes (24,25). However, the residues of Munc18c required for its dissociation from Syntaxin 4 in the stimulus-induced "opening" of Syntaxin 4 have yet to be defined. Syntaxin 4 and Munc18c are ubiquitously expressed, with evidence emerging from multiple fields of research demonstrating their importance in numerous tissues that coordinately regulate whole body euglycemia. For example, Syntaxin 4 and Munc18c are functionally essential in glucose uptake via insulin-stimulated GLUT4 vesicle translocation in skeletal muscle and adipose tissues (16, 26 -30). In addition, recent evidence supports a role for the Syntaxin 4 and Munc18c proteins in glucose-stimulated insulin granule exocytosis/insulin secretion from pancreatic islet ␤-cells (30 -33). Because insulin granule exocytosis and insulin-stimulated GLUT4 vesicle translocation rapidly and coordinately maintain whole body euglycemia, we propose that commonalities in post-translational modifications may provide the basis for a conserved mechanism by which the Munc18c-Syntaxin 4 complex regulates exocytosis.
The post-translational modification of Munc18 proteins by serine/ threonine phosphorylation has been suggested as a mechanism to regulate exocytosis (34). For example, the phosphorylation of Munc18a by protein kinase C and Cdk5 (cyclin-dependent kinase 5) reduces the amount of Munc18 that binds to syntaxin and promotes secretory granule exocytosis (35)(36)(37)(38)(39)(40). Munc18c was also recently shown to be phosphorylated by protein kinase C in rat parotid acinar cells and endothelial cells (41,42). Furthermore, the platelet Munc18c homolog PSP (platelet Sec1 protein) becomes phosphorylated in vitro by protein kinase C, and phosphorylated PSP fails to bind to Syntaxin 4 in vitro (43) as well as in thrombin-stimulated platelets (44). However, this serine/threonine phosphorylation of Munc18 proteins has failed to fully account for the stimulus-induced alterations in the Munc18-syntaxin complex.
In this study, we present evidence demonstrating that Munc18c is regulated by tyrosine phosphorylation and that increased tyrosine phosphorylation of Munc18c can dissociate it from Syntaxin 4. Munc18c phosphorylation was mapped to multiple tyrosine residues present in the N-terminal 255 amino acids, and mutation of Tyr 219 increased its binding to Syntaxin 4. Conversely, enhancement of Munc18c phosphorylation using the protein-tyrosine phosphatase inhibitor pervanadate decreased its association with Syntaxin 4. Glucose stimulation rapidly enhanced the tyrosine phosphorylation of endogenous Munc18c in MIN6 pancreatic beta cells, and Tyr 219 was found to confer this effect. Moreover, mutation of Munc18c Tyr 219 resulted in inhibited glucosestimulated SNARE complex formation and insulin secretion. Because we also found that Munc18c became rapidly tyrosine-phosphorylated in response to insulin in 3T3L1 adipocytes, this stimulus-induced Munc18c phosphorylation may be part of a conserved mechanism in vesicle/granule fusion.

EXPERIMENTAL PROCEDURES
Materials-The rabbit anti-Munc18c antibodies were generated as described (16). The rabbit anti-Syntaxin 4 polyclonal and mouse anti-VAMP2 monoclonal antibodies were obtained from Chemicon International, Inc. (Temecula, CA), and Synaptic Systems (Göttingen, Germany), respectively. The rabbit anti-Myc polyclonal and mouse anti-Myc monoclonal (9E10) antibodies were purchased from Upstate Biotechnology (Lake Placid, NY) and Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), respectively. The anti-phosphotyrosine monoclonal antibodies were obtained from Upstate Biotechnology (4G10) and BD Transduction Laboratories (PY20). The MIN6 cells were a gift from Dr. John C. Hutton (University of Colorado Health Sciences Center). TransFectin lipid reagent was purchased from Bio-Rad. Radioimmunoassay grade bovine serum albumin and D-glucose were purchased from Sigma. Enhanced chemiluminescence reagent and Hyperfilm-MP were obtained from Amersham Biosciences. [ 32 P]Orthophosphate (1 mCi/ml orthophosphoric acid in water) and X-Omat Blue XB-1 film were purchased from PerkinElmer Life Sciences.
Cell Culture-MIN6 beta cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 25 mM glucose and supplemented with 15% fetal bovine serum, 100 units/ml penicillin, 100 g/ml streptomycin, 292 g/ml L-glutamine, and 50 M ␤-mercaptoethanol as described previously (45,46). 3T3L1 preadipocytes were purchased from the American Type Culture Collection and cultured in DMEM containing 25 mM glucose and 10% calf serum at 37°C and 8% CO 2 . At confluence, cells were differentiated by incubation in medium containing 25 mM glucose, 10% fetal bovine serum, 1 g/ml insulin, 1 mM dexamethasone, and 0.5 mM isobutyl-1-methylxanthine. After 4 days the medium was changed to DMEM, 25 mM glucose, and 10% fetal bovine serum. Differentiated adipocytes were used for the experiments 12 days after the initiation of differentiation and were placed in serumdepleted medium for 2-3 h prior to insulin stimulation.
[ 32 P]Orthophosphate Labeling of MIN6 Beta Cells-Metabolic labeling using [ 32 P]orthophosphate was performed as described previously (47) with minor modification. Briefly, MIN6 beta cells (70% confluent on 10-cm 2 tissue culture dishes) were incubated in DMEM containing 2.8 mM glucose at 37°C for 16 h. The cells were washed twice with phosphate-free DMEM and incubated for 4 h in 5 ml of the same medium containing 100 Ci/ml [ 32 P]orthophosphate. The medium was then removed, and the cells were washed twice with ice-cold buffer containing 20 mM HEPES (pH 7.4) and 150 mM NaCl. One milliliter of ice-cold Nonidet P-40 lysis buffer was added to the cells, and cleared cell lysates were prepared and used for immunoprecipitation with anti-Munc18c antibody. The Munc18c immunoprecipitates were subjected to 10% SDS-PAGE, and orthophosphate incorporation was detected by autoradiography.
Subcellular Fractionation-Subcellular fractions of beta cells were isolated as described previously (48,49). Briefly, MIN6 beta cells at 80 -90% confluence were harvested in 1 ml of homogenization buffer (20 mM Tris-HCl (pH 7.4), 0.5 mM EDTA, 0.5 mM EGTA, 250 mM sucrose, 1 mM dithiothreitol, and 1 mM sodium orthovanadate containing the protease inhibitors leupeptin (10 g/ml), aprotinin (4 g/ml), pepstatin (2 g/ml), and phenylmethylsulfonyl fluoride (100 M)). Plasma membrane fractions were isolated as described previously (50). Cells were disrupted by 10 strokes through a 27-gauge needle, and homogenates were centrifuged at 900 ϫ g for 10 min. The resulting pellets were mixed with 1 ml of Buffer A (0.25 M sucrose, 1 mM MgCl 2 , and 10 mM Tris-HCl (pH 7.4)) and 2 volumes of Buffer B (2 m sucrose, 1 mM MgCl 2 , and 10 mM Tris-HCl (pH 7.4)). The mixture was overlaid with Buffer A and centrifuged at 113,000 ϫ g for 1 h to obtain an interface containing the plasma membrane fraction. The interface was collected and diluted to 2 ml with homogenization buffer for centrifugation at 3000 ϫ g for 10 min, and the resulting pellet was collected as the plasma membrane fraction. Fractions were assayed for soluble protein content as described previously (51).
Plasma membrane fractions of 3T3L1 adipocytes were obtained using the differential centrifugation method as described previously (16,52). Briefly, 3T3L1 adipocytes were washed with and resuspended in HES buffer (20 mM HEPES (pH 7.4), 1 mM EDTA, and 255 mM sucrose containing 1 mM phenylmethylsulfonyl fluoride, 10 g/ml pepstatin, 10 g/ml aprotinin, and 5 g/ml leupeptin). Lysates were sheared 10 times through a 22-gauge needle and centrifuged at 19,000 ϫ g for 20 min at 4°C. The resulting pellet was resuspended in HES buffer and layered onto a 1.12 M sucrose cushion for centrifugation at 100,000 ϫ g for 60 min. The plasma membrane layer was removed from the cushion and centrifuged at 40,000 ϫ g for 20 min. The plasma membrane pellet was resuspended in HES buffer and assayed for protein content as described above.
Co-immunoprecipitation and Immunoblotting-MIN6 beta cells were incubated with MKRBB for 2 h and preincubated with or without pervanadate for 5 min prior to glucose stimulation. Pervanadate was made immediately prior to use by combining 1 mM sodium orthovanadate with 3 mM hydrogen peroxide for 15 min (53). Cells were subsequently lysed in Nonidet P-40 lysis buffer. MIN6 beta cell cleared detergent homogenates (2-3 mg) were combined with rabbit anti-Munc18c antibody, rabbit anti-Syntaxin 4 antibody, or mouse anti-phosphotyrosine antibody for 2 h at 4°C, followed by a second incubation with protein G Plus-agarose for 2 h. The resultant immunoprecipitates were subjected to 10% SDS-PAGE, followed by transfer to polyvinylidene difluoride (PVDF) membranes for immunoblotting. Anti-Munc18c, anti-Syntaxin 4, and anti-phosphotyrosine (PY20) antibodies were used at 1:5000, 1:500, and 1:1000 dilutions, respectively, and horseradish peroxidase-conjugated secondary antibodies were used at a dilution of 1:5000 for visualization by chemiluminescence. Immunoprecipitations using adipocyte cleared detergent cell lysates were performed similarly to those using MIN6 cell lysates.

Munc18c Undergoes Tyrosine Phosphorylation in MIN6 Beta Cells-
To first determine whether Munc18c can be phosphorylated in MIN6 beta cells, metabolic labeling with [ 32 P]orthophosphate and immunoprecipitation with anti-Munc18c antibody were performed (Fig. 1A). MIN6 cell lysates contained a high abundance of [ 32 P]phosphate-labeled proteins. However, immunoprecipitation with anti-Munc18c antibody revealed incorporation into a single band at ϳ68 kDa, corresponding to a band detected by anti-Munc18c antibody in cell lysates. Immunoprecipitation of labeled lysates using anti-Syntaxin 4 antibody failed to show any incorporation of phosphate, indicating that Syntaxin 4 is not phosphorylated in beta cells (data not shown) and that phosphorylation is specific to Munc18c in this complex.
To determine whether Munc18c phosphorylation occurs in a tyrosine-specific manner, we treated MIN6 cells with the protein-tyrosine phosphatase inhibitor pervanadate. Treatment with pervanadate was predicted to increase the abundance of tyrosine-phosphorylated Munc18c, but not serine/threonine phosphorylation, if in fact Munc18c The radiolabeled cell lysates were used for immunoprecipitation (IP) with rabbit anti-Munc18c antibody, and protein was resolved by 10% SDS-PAGE. The gel was dried and subjected to autoradiography. B, MIN6 beta cells were incubated in MKRBB for 2 h prior to incubation with either vehicle or pervanadate (pV; 1 mM sodium orthovanadate plus 3 mM hydrogen peroxide) for 5 min. Cell lysates were prepared for immunoprecipitation reactions with anti-phosphotyrosine antibody (p-Tyr). Proteins were separated by 10% SDS-PAGE, transferred to PVDF membrane, and immunoblotted (IB) with anti-Munc18c antibody. C, lysates from cells treated with vehicle or pervanadate for 5 min were immunoprecipitated with anti-Munc18c antibody. Proteins were subjected to 10% SDS-PAGE and immunoblotted with anti-Munc18c and anti-phosphotyrosine antibodies, which recognized an identically sized band at 68 kDa. Data are representative of at least three independent experiments for each panel.
undergoes tyrosine phosphorylation. Pervanadate was added to MIN6 cells for 5 min prior to preparation of lysates for immunoprecipitation. Pervanadate treatment resulted in 2.6 Ϯ 0.5-fold (p Ͻ 0.02) increased levels of phosphotyrosine-modified Munc18c as detected by immunoprecipitation with anti-phosphotyrosine antibodies (Fig. 1B). Munc18c abundance in cell lysates was unaffected by pervanadate treatment. In a reciprocal immunoprecipitation using anti-Munc18c antibody, pervanadate treatment induced a similar increase in phosphorylated Munc18c (Fig. 1C). Subsequent immunoblotting of the same membrane confirmed the migration of the phosphotyrosine band as identical to that of the Munc18c band at 68 kDa. This indicated that Munc18c and not some other tyrosine-phosphorylated protein with a molecular mass of ϳ68 kDa was indeed tyrosine-phosphorylated.
Tyrosine Phosphorylation of Munc18c Leads to Its Dissociation from Syntaxin 4: A Conserved Mechanism-To determine whether the tyrosine phosphorylation of Munc18c alters its ability to interact with Syntaxin 4, we treated MIN6 cells with increasing concentrations of pervanadate and prepared lysates for immunoprecipitation with anti-Munc18c antibody. Similar levels of Munc18c were immunoprecipitated under all treatment conditions (Fig. 2). Munc18c became increasingly tyrosine-phosphorylated as the pervanadate concentration was increased. At 1 mM pervanadate (Fig. 2, lane 4), Munc18c tyrosine phosphorylation exceeded that of the basal level (lane 1) by 4 Ϯ 1-fold (p Ͻ 0.05). By contrast, Syntaxin 4 coprecipitating with Munc18c decreased as pervanadate levels were increased. Syntaxin 4 protein expression levels were unaffected by pervanadate treatment. These data showed that pervanadate treatment specifically increased Munc18c phosphorylation at tyrosine residues, which correlated with its dissociation from Syntaxin 4.
We (16) and others (28) have proposed that stimulation of vesicle fusion must somehow trigger the dissociation of the Munc18c-Syntaxin 4 complex, enabling Syntaxin 4 to then form the heterotrimeric SNARE core complex with SNAP-25/23 and VAMP2. Moreover, this dissociation is predicted to occur at the plasma membrane, the location of the Syntaxin 4 and SNAP-25/23 proteins. Although Munc18c is a soluble protein, it localizes to the plasma membrane through its interaction with Syntaxin 4 (16). To investigate whether this particular pool of Munc18c can be tyrosine-phosphorylated in a stimulus-dependent manner, MIN6 cells were subfractionated according to previously described methods (48), and plasma membrane fractions were used in immunoprecipitation experiments. Munc18c immunoprecipitated from plasma membrane fractions prepared from unstimulated cells showed some phosphotyrosine immunoreactivity (Fig. 3A). However, stimulation for 5 min with glucose increased the amount of tyrosinephosphorylated Munc18c by 2.8 Ϯ 0.6-fold (p Ͻ 0.05). From the same fraction preparations, Syntaxin 4 co-immunoprecipitated phosphorylated Munc18c under unstimulated conditions (Fig. 3B); however, within 5 min of glucose stimulation, the amount of phosphorylated Munc18c immunoprecipitated decreased by ϳ50 Ϯ 16% (p Ͻ 0.02). The phosphotyrosine band was identified as Munc18c based upon Munc18c immunoblotting and migration of the band at ϳ68 kDa. In addition, the identity of this band was confirmed not to be the tyrosine-phosphorylated forms of other Syntaxin 4-binding proteins such as N-ethylmaleimide-sensitive factor (NSF) or SNAP-25 because N-ethylmaleimidesensitive factor migrated at ϳ75-80 kDa and SNAP-25 migrated at 25-30 kDa (data not shown) and were distinctly different from the 68-kDa band seen here. Moreover, the Syntaxin 4-interacting protein Synip (70 kDa) has also been suggested to be phosphorylated, but at serine residues and not tyrosine. These data suggested that, as Munc18c became increasingly tyrosine-phosphorylated in response to glucose stimulation, it dissociated from Syntaxin 4.
To determine whether Munc18c-Syntaxin 4 complex dissociation by Munc18c tyrosine phosphorylation in response to stimuli is a conserved mechanism, parallel experiments were performed in 3T3L1 adipocytes. In 3T3L1 adipocytes and skeletal muscle, Munc18c-Syntaxin 4 interactions are critical for mediating insulin-stimulated GLUT4 vesicle fusion (16,54). Fully differentiated 3T3L1 adipocytes left unstimulated or stimulated with 100 nM insulin for 5 min were harvested, and plasma membrane fractions were prepared as described previously (25). Munc18c was equivalently immunoprecipitated from each fraction, and tyrosinephosphorylated Munc18c was increased by ϳ1.5-fold in fractions prepared from cells stimulated with insulin (Fig. 3C). Immunoprecipitation of Syntaxin 4 resulted in the co-immunoprecipitation of 80% less  tyrosine-phosphorylated Munc18c from insulin-stimulated lysates compared with unstimulated lysates (Fig. 3D). With 15 min of insulin stimulation, the tyrosine phosphorylation of Munc18c increased by Ͼ2.5-fold, which corresponded to a 90% decrease in binding to Syntaxin 4 (data not shown). These results show that there are mechanistic parallels in the stimulus-induced dissociation of tyrosine-phosphorylated Munc18c from Syntaxin 4 between insulin-responsive adipocytes and insulin-secreting beta cells.
The N Terminus of Munc18c Is Required for Tyrosine Phosphorylation and Interaction with Syntaxin 4-To delineate the region of Munc18c required for modification by tyrosine phosphorylation, N-terminal deletion constructs were generated. Because truncated forms of Munc18c protein are generally poorly expressed or unstable, we generated deletions at internal Met residues that had reasonably good Kozak sequences. These deletions resulted in the truncation of the first 172, 255, or 332 amino acids, which were then tagged at the C terminus with Myc and His epitope tags to allow for discrimination of recombinant from endogenous Munc18c (Fig. 4A). MIN6 cells were electroporated with each construct, and cell lysates were used for immunoprecipitation with anti-Myc antibody to assess the expression levels of the various truncation products. Myc immunoblotting showed the Munc18c-WT, Munc18c-⌬172, Munc18c-⌬255, and Munc18c-⌬332 proteins to be 67, 43, 39, and 33 kDa in size, respectively (Fig. 4B, upper panel). Confocal immunofluorescence microscopy confirmed that the localization of Munc18c-WT and the truncation mutants was similar to that of endogenous Munc18c (data not shown). Expression of the Munc18c-⌬255 and Munc18c-⌬332 proteins was reduced compared with that of the Munc18c-WT and Munc18c-⌬172 proteins (Fig. 4B, lanes 3 and 4 versus lanes 1 and 2). Phosphotyrosine immunoblotting of this membrane showed that Munc18c-WT was phosphorylated (Fig. 4B, middle panel), which recapitulated the phosphorylation of endogenous Munc18c. However, phosphorylation was reduced in the Munc18c-⌬172 mutant and altogether absent in the Munc18c-⌬255 and Munc18c-⌬332 mutants (Fig. 4B, boxes). Moreover, Syntaxin 4 bound only to Munc18c-WT (Fig. 4B, lower panel). In a second approach, these truncation mutants were linked at the C terminus to EGFP and showed an identical pattern of tyrosine phosphorylation, indicating that this phosphorylation was independent of the epitope tags (data not shown). The results were similar using lysates prepared from glucose-stimulated cells, suggesting that residues 1-255 contain all tyrosine residues essential for phosphorylation and that the N terminus of Munc18c is essential for its interaction with Syntaxin 4.
To identify candidate tyrosine residues within the N terminus of Munc18c, we used ClustalW (available at ca.expasy.org). Although no classic NPXY motifs were found, four tyrosine residues were selected as likely candidates for phosphorylation using a computer prediction program (available at www.cbs.dtu.dk) that searches for alternative motifs (59). Alternative motifs have at least one acidic residue in the four residues immediately upstream of the candidate tyrosine (Tyr ϭ 0, P4 to P1). Among the sites with at least two acidic residues, EE and ED are the most frequently occurring dipeptides at positions P4 and P3. Candidate sites were detected between amino acids 65 and 220. Alignment of mouse Munc18a and Munc18b, C. elegans UNC-18, D. melanogaster ROP, and S. cerevisiae Sec1 against residues 60 -262 of mouse Munc18c is shown in Fig. 5. The four tyrosine residues meeting these criteria are located at positions 66, 103, 218, and 219 in the mouse Munc18c sequence (Fig. 5, shaded). Of these four sites, only Tyr 103 is conserved among all SM proteins in the alignment.
Site-directed mutations of tyrosine to phenylalanine were made at residues 66, 103, 218, and 219 in the context of Munc18c-WT-Myc/His (Fig. 6A). Each construct was electroporated into MIN6 cells, and lysates were prepared for immunoprecipitation with anti-Myc antibody (Fig. 6B). Each mutant was expressed at a level below that of Munc18c-WT in the lysates as detected by Myc immunoblotting. Quantification of the Tyr/Myc ratio in each immunoprecipitation showed that there were no significant changes in modification (data not shown). Moreover, the phosphotyrosine level of each mutant correlated with its abundance in the lysate, suggesting that no single site confers the entire phosphorylation signal. Although each mutant retained the ability to co-immunoprecipitate Syntaxin 4, optical density scanning quantitation revealed that the Munc18c-Y219F mutant immunoprecipitated 2.5-fold more Syntaxin 4 than did Munc18c-WT (Fig. 6C). The amount of Syntaxin 4 immunoprecipitated by Munc18c-Y103F also increased, but failed to reach statistical significance. Of the four mutants, expression of Munc18c-Y103F was the most severely impaired relative to Munc18c-WT, and Syntaxin 4 levels, as well as total protein levels, tended to be reduced in lysates expressing Munc18c-Y103F. This suggests that Munc18c-Y103F is somehow toxic to cell viability. Altogether, these data suggest that Tyr 219 is important for Syntaxin 4 binding. FIGURE 4. The N terminus of Munc18c is tyrosine-phosphorylated. A, three N-terminally truncated forms of Munc18c were obtained by PCR amplification using different 5Ј-primers that initiated at internal Met sites with good Kozak sequences. These deletions resulted in the removal of the first 172, 255, or 332 residues. PCR products as well as Munc18c-WT were subcloned into pcDNA3.1/myc-His for expression in mammalian cells. B, DNA constructs were electroporated into MIN6 beta cells, and 48 h later, detergent lysates were prepared for immunoprecipitation (IP) with anti-Myc antibody. Myc-tagged proteins are denoted by asterisks. Proteins were subjected to 10% SDS-PAGE and initially immunoblotted (IB) for Myc to assess the size and level of expression of each truncation product (upper panel). Subsequently, the membrane was stripped and immunoblotted with anti-phosphotyrosine (p-Tyr; middle panel) and anti-Syntaxin 4 (Syn4; lower panel) antibodies. Boxes in the phosphotyrosine immunoblot delineate the locations of the truncated protein bands detected by anti-Myc antibody.
To confirm that the increased binding of Syntaxin 4 to Munc18c-Y219F was independent of effects due to cell electroporation, MIN6 cells were transiently transfected by lipofection, and lysates were used for immunoprecipitation (Fig. 6D). Myc-tagged Munc18c-Y219F co-immunoprecipitated ϳ3-fold more Syntaxin 4 compared with Munc18c-WT, consistent with results obtained using electroporation to transfect cells. In addition, confocal immunofluorescence microscopy analysis demonstrated that Myc/His-tagged Munc18c-WT and Munc18c-Y219F were similarly localized to the cytosolic and membrane compartments, indicating that the increased affinity of Munc18c-Y219F for Syntaxin 4 was not due to altered localization of the Munc18c-Y219F protein (data not shown). These data suggest that the specific phosphorylation of Tyr 219 may be pivotal in the dissociation of the Munc18c-Syntaxin 4 complex in response to stimuli such that mutation of the site results in reduced dissociation and enhanced binding of the complex.
To determine whether Tyr 219 is indeed a site of phosphorylation, we first attempted to use two-dimensional phosphopeptide mapping. However, the tryptic cleavage sites flanking the peptide containing Tyr 219 were too closely spaced and generated a fragment too small for analysis. As an alternative strategy, we isolated Tyr 219 in a small fragment of Munc18c encoding residues 173-255 and performed site-di- FIGURE 5. Sequence alignment of the SM protein family and identification of four candidate tyrosine residues for phosphorylation. ClustalW was used to align sequences of SM proteins relative to amino acids 60 -262 of mouse Munc18c (available at ca.expasy.org/tools/#align). Candidate residues for tyrosine phosphorylation (shaded) were identified based upon prediction (79) of non-NPXY motifs using the prediction program from the Center for Biological Sequence Analysis at the Technical University of Denmark (available at www.cbs.dtu.dk). Amino acid positions are listed to the right of each sequence. rected mutagenesis within this fragment to assess its requirement for tyrosine phosphorylation. Residues 173-255 were linked upstream of EGFP such that EGFP could provide an epitope for detection and also be used as a gauge of transfection efficiency (Fig. 7A). This fragment of Munc18c also contained Tyr 218 , but eliminated sites 66 and 103. Within the context of this construct (Munc18c-(173-255)-WT-EGFP), sites 218 and 219 were individually mutagenized to phenylalanine to determine their requirement for phosphotyrosine immunoprecipitation. Both sites were mutagenized in the double mutant to verify their requirement for tyrosine phosphorylation in this fragment. Munc18c-(173-255)-WT-EGFP and mutant DNAs were transiently transfected into MIN6 beta cells, and unstimulated lysates were prepared for immunoprecipitation with anti-phosphotyrosine antibody (Fig. 7B). Immunoblotting for EGFP expression in cell lysates demonstrated equivalent expression levels among the WT and mutant proteins (Fig. 7B, lanes  1-4). Munc18c-(173-255)-WT-EGFP was immunoprecipitated by anti-phosphotyrosine antibody, confirming that this fragment of Munc18c contained phosphorylated tyrosine residues (Fig. 7B, lane 5). Munc18c-(173-255)-Y218F-EGFP and Munc18c-(173-255)-Y219F-EGFP were also immunoprecipitated by anti-phosphotyrosine antibody, demonstrating that each site underwent phosphorylation (Fig. 7B,  lanes 6 and 7). Although we had predicted that Tyr 219 might be phosphorylated only in response to glucose, the abundance of Munc18c-(173-255)-Y218F-EGFP immunoprecipitated by anti-phosphotyrosine consistently exceeded that of Munc18c-(173-255)-WT-EGFP or Munc18c-(173-255)-Y219F-EGFP. One possible explanation for this might be that the Y218F mutation induced a conformational change that exaggerated the exposure of the Tyr 219 active site, increasing the number of molecules carrying only the Tyr 219 site capable of being phosphorylated under basal conditions. The double mutant, containing Y218F and Y219F, was not immunoprecipitated by anti-phosphotyrosine antibody at all, indicating that Tyr 218 and Tyr 219 were required to confer the phosphotyrosine immunoreactivity in this fragment of Munc18c (Fig. 7B, lane 8). Interestingly, a third tyrosine residue present in this fragment that failed to meet the criteria of the prediction program as a potential site of phosphorylation (Tyr 196 ) did not sustain phosphorylation in the double mutant. Tyr 196 was not initially tested because it has no Glu or Asp residues at positions Ϫ4 to Ϫ2 in the sequence. Ponceau S staining of the PVDF membrane was used to verify equivalent loading of immunoprecipitation reactions. Thus, these data demonstrated that Munc18c residues Tyr 218 and Tyr 219 were each phosphorylated.
In parallel experiments, we tested whether mutation of Tyr 219 disrupts the ability of glucose to increase the level of phosphorylation of   (Fig. 8,  lanes 5 and 7), similar to the data shown in Fig. 6. However, no increase in Munc18c-(173-255)-Y219F-EGFP immunoprecipitation by antiphosphotyrosine antibody was observed in glucose-stimulated lysates compared with Munc18c-(173-255)-WT-EGFP (Fig. 8, lane 8 versus  lane 6). These data showed that Tyr 218 phosphorylation did not increase in response to glucose, suggesting that Tyr 219 is required for glucoseinduced phosphorylation.
The utility of this as an assay system to test the importance of Tyr 219 for function in SNARE complex assembly lies in the ability of Syntaxin 4 overexpression to relieve the inhibition caused by Munc18c-WT expression. We have demonstrated previously that inhibition of GLUT4 translocation in 3T3L1 adipocytes caused by overexpression of Munc18c can be reversed if Syntaxin 4 levels are coordinately increased (24). In this, Syntaxin 4 is used as a "molecular sponge" to bind excess Munc18c. In the present study, Syntaxin 4 overexpression in Munc18c-WT-expressing cells restored the level of VAMP2 coprecipitated by anti-Syntaxin 4 antibody in unstimulated lysates to ϳ90% of the control level (pcDNA3-transfected cells) and to ϳ80% of the control level in glucose-stimulated lysates (Fig. 9B, lanes 1 and 2 and lanes 5 and 6,  respectively). However, Syntaxin 4 overexpression in Munc18c-Y219Fexpressing cells failed to restore the glucose-induced increase in Syn-taxin 4-VAMP2 association (Fig. 9B, lane 4 versus lane 2). Overexpression of the Munc18c-WT and Munc18c-Y219F proteins was confirmed by immunoblotting for the Myc epitope, although no decrease in Munc18c-WT binding was observed in glucose-stimulated lysates. This may be a reflection of the transient nature of the dissociation between Munc18c and Syntaxin 4 or of transient phosphorylation caused by alterations in the balance of kinase and phosphatase activities. Overall, these results showed that the increased affinity of the Munc18c-Y219F mutant for Syntaxin 4, combined with the inability of the mutant to undergo glucose-induced phosphorylation, resulted in the failure of Munc18c-Y219F to release Syntaxin 4 upon glucose stimulation and impaired the glucosestimulated increase in Syntaxin 4-VAMP2 association.
Phosphorylation of Tyr 219 Is Functionally Important for Glucose-stimulated Secretion-To determine whether the effects of Munc18c-Y219F expression on Syntaxin 4-VAMP2 association correlate with functional inhibition of glucose-stimulated insulin secretion, we used the human C-peptide reporter assay. Human C-peptide (synthesized from a human proinsulin cDNA) is packaged in an identical fashion to mouse C-peptide and insulin in granules, but is immunologically distinct from mouse endogenous C-peptide and serves as a reporter from transfected cells. MIN6 cells cotransfected with pcDNA3 control DNA plus human pro-  insulin DNA exhibited 160 Ϯ 5.2% (p Ͻ 0.005) of basal human C-peptide secretion in response to glucose, an appropriate increase in human C-peptide secretion in response to glucose in this assay system (55,56). In contrast, overexpression of Munc18c-WT or Munc18c-Y219F reduced glucose-stimulated secretion to 115 and 121% of the basal level, respectively (Fig. 10, bars 1 and 3), consistent with their inhibition of endogenous Syntaxin 4 association with VAMP2. The inhibitory effect of Munc18c overexpression is also consistent with inhibition of glucosestimulated insulin secretion in islets from Munc18c-overexpressing transgenic mice (31). Overexpression of Syntaxin 4 with Munc18c-WT partially and significantly increased human C-peptide secretion to 132% of the basal level (Fig. 10, bar 2). This partial restoration of human C-peptide secretion correlates with the partial restoration of Syntaxin 4-VAMP2 binding induced by overexpression of Syntaxin 4 with Munc18c-WT in Fig. 9B. However, Syntaxin 4 overexpression with Munc18c-Y219F failed to increase glucose-stimulated insulin secretion at all. These data suggest that phosphorylation of Tyr 219 is essential for glucose-regulated dissociation, which is important for the function of the Munc18c-Syntaxin 4 complex in regulating glucose-stimulated exocytosis and secretion.

DISCUSSION
All types of SNARE-mediated vesicle/granule exocytosis require an SM protein (57). Using insulin-stimulated GLUT4 vesicle translocation as our model system, we have shown previously that Munc18c-Syntaxin 4 complexes are important for the fusion step of vesicle exocytosis (25), although the mechanism for this has remained unknown. Here, we have presented evidence of a novel and conserved mechanism for Munc18c-Syntaxin 4 dissociation, a key step that is required for vesicle fusion to occur. This model mechanism entails stimulus-induced tyrosine phosphorylation of Munc18c, which dissociates it from Syntaxin 4, enabling Syntaxin 4 to bind with VAMP2.
The conserved nature of this model is supported by evidence showing that Munc18c becomes increasingly tyrosine-phosphorylated in response to stimuli that initiate the exocytosis of vesicles/granules in two distinct cell types, pancreatic beta cells and adipocytes. Increased tyrosine phosphorylation of Munc18c correlated with its diminished interaction with Syntaxin 4. Tyrosine phosphorylation was mapped within the N-terminal 255 residues of Munc18c, and this region was found to be essential for its interaction with Syntaxin 4. Analysis of this region revealed the presence of four candidate tyrosine residues for phosphorylation; however, only mutation of Tyr 219 was found to alter Munc18c binding affinity for Syntaxin 4. Tyr 219 was found to be required to confer the glucoseinduced phosphorylation to a fragment of Munc18c containing only this and a second candidate tyrosine residue (Tyr 218 ). Our studies using the Y219F mutant of Munc18c indicated that phosphorylation of Tyr 219 was functionally important for Syntaxin 4 binding to VAMP2 and also for stimulus-induced secretion, altogether suggesting that phosphorylation of Tyr 219 is crucial for the subsequent function of Syntaxin 4 in granule fusion and insulin release.
Munc18c was found to be phosphorylated in the basal state at multiple tyrosine residues because the individual mutation of Tyr 66 , Tyr 103 , Tyr 218 , or Tyr 219 in full-length Munc18c failed to eliminate phosphotyrosine immunoreactivity. Munc18c-(173-255)-WT, which contained only the candidate sites Tyr 218 and Tyr 219 , was able to recapitulate the glucose-stimulated increase in tyrosine phosphorylation of endogenous Munc18c, suggesting that one or both of the two included candidate tyrosines was phosphorylated preferentially in response to glucose stimulation. Our data further showed that mutation of Tyr 219 , but not Tyr 218 , resulted in loss of this glucose-induced phosphorylation, sug-gesting Tyr 219 as the primary site for stimulus-induced phosphorylation. Although Tyr 66 and Tyr 103 were not similarly tested in the context of a smaller fragment of Munc18c for the ability to confer stimulusenhanced phosphorylation, neither the Y66F nor Y103F mutation exerted a significant alteration in affinity for Syntaxin 4, suggesting against either site as crucial for Syntaxin 4 function with VAMP2 in stimulus-induced exocytosis. However, it is unclear as to why any molecules of the Munc18c-(173-255)-Y218F-EGFP mutant, which had only Tyr 219 intact, were immunoprecipitated by anti-phosphotyrosine antibody from unstimulated lysates. In fact, Munc18c-(173-255)-Y218F-EGFP was immunoprecipitated in slightly greater abundance than Munc18c-(173-255)-WT-EGFP. One possible explanation could be that the mutation in the context of this small fragment of Munc18c induced a conformational change that enhanced the accessibility of the site for modification by phosphorylation. In addition, it is possible that the non-candidate site, Tyr 196 , was somehow made active as a site of phosphorylation and contributed to the tyrosine phosphorylation. Other proteins such as Sic1 and NFAT1 (nuclear factor of activated T cells 1) are phosphorylated at multiple sites, at which a threshold level of phosphorylation serves as a binary switch (58 -60). Alternatively, the transcription activator Ets-1 is phosphorylated at multiple sites, which act additively to produce graded DNA binding affinity, shifting Ets-1 from a dynamic conformation poised to bind DNA to a well folded inhibited state (61). However, it is clear from the double mutant that elimination of Tyr 218 and Tyr 219 together eliminated all phosphotyrosine immunoreactivity, indicating that they were indeed phosphorylated. Future studies will be required to determine whether Munc18c phosphorylation proceeds via a binary switch or a "graded rheostat" mechanism.
Tyrosine phosphorylation results from the net effects of protein-tyrosine kinases balanced by the effects of protein-tyrosine phosphatases (PTPs) (62). The data presented here suggest that Munc18c can be tyrosine-phosphorylated under basal conditions. One candidate protein-tyrosine kinase may be the Fes/Fer non-receptor protein-tyrosine kinase, which has been shown to participate in SNARE-mediated vesicle fusion (63). Moreover, glucose stimulation increased the level of Munc18c tyrosine phosphorylation, and this effect was mimicked by inhibition of PTPs using pervanadate, suggesting that glucose may enhance phosphorylation via inactivation of PTPs. If analogous in adipocytes, inactivation of PTPs would be triggered by insulin stimulation. Although it was clearly shown that increased phosphorylation of Munc18c correlated with its dissociation from Syntaxin 4, significant loss of Syntaxin 4 binding was observed only when cells had been treated with the PTP inhibitor pervanadate. This suggests that the concurrent activities of protein-tyrosine kinases and PTPs at the multiple sites of phosphorylation may contribute to our diminished ability to detect these transient changes in Munc18c-Syntaxin 4 binding. In support of a role for glucose in the inactivation of PTPs, it has been shown that glucose stimulation can produce H 2 O 2 , which causes oxidative stress in beta cells (64), and that oxidative stress can inactivate PTPs (65). Thus, glucose stimulation may induce the inactivation of PTPs, which causes a net increase in Munc18c tyrosine phosphorylation. Future studies will be focused on identification of the protein-tyrosine kinases and PTPs involved in the tyrosine phosphorylation of Munc18c.
One very important finding from this study is that the inhibition of exocytosis caused by overexpression of Munc18c correlated with a marked decrease in endogenous Syntaxin 4-VAMP2 interaction. Although we (16,54) and others (28) have shown previously that overexpression of Munc18c in islet cells, adipocytes, and skeletal muscle tissue results in significantly impaired regulated exocytosis, the under-lying cause of this inhibition was unknown. The data presented here suggest that the underlying mechanism for the inhibitory effect of Munc18c overexpression involves Munc18c sequestration of endogenous Syntaxin 4, which reduces Syntaxin 4-VAMP2 association. This was further demonstrated by the partial rescue of both Syntaxin 4-VAMP2 binding and functional exocytosis upon addition of exogenous Syntaxin 4. Because the restoration of binding and exocytosis was indeed only partial, it remains possible that overexpression of Munc18c impairs/sequesters molecules other than just Syntaxin 4.
In fact, dissociation of the Munc18-syntaxin complex is also thought to be catalyzed by additional SNARE-interacting proteins. For example, the dissociation of yeast homologs Sly1p (Munc18 homolog) and Sed5p (syntaxin homolog) has been shown to require the small GTPase protein Ypt1p (66). Similarly, in adipocytes, the Munc18c-Syntaxin 4 complex has been proposed to be dissociated upon activation of the Rab4 GTPase (67) or the phosphatidylinositol 3-kinase/Akt-dependent phosphorylation the of Rab GTPase-activating protein AS160 (68,69). In support of this, the crystal structure of Munc18-1 revealed a region that is very similar to the Ypt1p-binding site of Sly1p (70). Moreover, we (71) and others (72) have targeted this region of Munc18-1 and Munc18c with small interfering peptides to demonstrate the importance of this loop region to vesicle fusion. However, other than Ypt1p in yeast, no direct interactions between mammalian Munc18 and Rab proteins have been demonstrated to date, suggesting that the mechanisms for dissociation of the mammalian Munc18-syntaxin complexes may differ from those of their yeast counterparts.
Although our demonstration of the regulation of the Munc18-syntaxin complex by tyrosine phosphorylation of Munc18c is a novel concept, modification of Munc18 and syntaxin proteins by serine and threonine phosphorylation has been investigated previously. For example, serine phosphorylation of Munc18-1 by protein kinase C disrupts interaction with Syntaxin 1A (35,38), and threonine phosphorylation of Munc18-1 by Cdk5 results in the disassembly of the Munc18a-Syntaxin 1A complex (37). Furthermore, serine phosphorylation of Syntaxin 1A by the death-associated protein kinase DAPK significantly decreases interaction with Munc18-1 (73). In addition, Syntaxins 4 and 1A can be phosphorylated by multiple kinases; however, this has not been shown to occur in adipocytes or islet beta cells (73)(74)(75)(76). Moreover, in our beta cell metabolic labeling studies, immunoprecipitation of Munc18c did not co-immunoprecipitate phosphate-incorporated Syntaxin 4 (data not shown). These data suggest two possibilities: 1) Syntaxin 4 is not phosphorylated in beta cells, or 2) Munc18c cannot bind to Syntaxin 4 that is phosphorylated. Thus, the sites of serine/threonine phosphorylation in the Munc18c and Syntaxin 4 proteins remain unknown and require future investigation. In all, although the serine phosphorylation of other Munc18 isoforms has been demonstrated, it has yet to be demonstrated that this can fully account for the mechanism underlying alterations in the Munc18-syntaxin complex in response to stimuli. It may be that the phosphorylation of Munc18 proteins at serine, threonine, and tyrosine residues coordinately regulates this interaction.
Munc18c and Syntaxin 4 are ubiquitously expressed proteins that are emerging as key regulators of additional exocytotic events that impact human health other than the diabetes field. For example, Syntaxin 4based complexes regulate hemostasis through platelet exocytosis, implicating their importance in cardiovascular disease (43,77). In addition, Syntaxin 4 and Munc18c have been shown to be regulated during macrophage activation to function in membrane traffic and tumor necrosis factor-␣ cytokine secretion (78), suggesting a role for the Munc18c-Syntaxin 4 complex in macrophage-mediated host defense and inflammatory disease. Finally, the exocytosis of Weibel-Palade bod-ies containing von Willebrand factor, P-selectin, and interleukin-8 within minutes after stimulation has been shown recently to be regulated by Syntaxin 4 in endothelial cells (42), showing the potential importance of Munc18c-Syntaxin 4 complexes in vascular homeostasis.
Significant progress has been made in defining signaling cascades that lead to the secretion of insulin from islet beta cells and the uptake of glucose into skeletal muscle and adipose tissues via GLUT4 vesicle translocation; however, each cascade is rate-limited by the distal steps of vesicle exocytosis. The functional regulation of Syntaxin 4 by Munc18c is thought to be crucial in the fusion event, although the details have remained unclear. Given the evidence showing that the same Munc18c-Syntaxin 4 complex regulates insulin secretion and insulin action, therapies that target or even affect secondarily these SNARE and SNAREassociated proteins will have consequences in regulation of glucose homeostasis. This could be beneficial if the regulation occurs in parallel, but hazardous if the complex regulation is opposing (i.e. up-regulation of Syntaxin 4 results in decreased glucose uptake but increased insulin secretion). Thus, the mechanistic data presented here begin to fill an important gap in our understanding of the etiology of insulin resistance and diabetes.