Doc2β Is a Novel Munc18c-interacting Partner and Positive Effector of Syntaxin 4-mediated Exocytosis*

The widely expressed Sec/Munc18 (SM) protein Munc18c is required for SNARE-mediated insulin granule exocytosis from islet beta cells and GLUT4 vesicle exocytosis in skeletal muscle and adipocytes. Although Munc18c function is known to involve binding to the t-SNARE Syntaxin 4, a paucity of Munc18c-binding proteins has restricted elucidation of the mechanism by which it facilitates these exocytosis events. Toward this end, we have identified the double C2 domain protein Doc2β as a new binding partner for Munc18c. Unlike its granule/vesicle localization in neuronal cells, Doc2β was found principally in the plasma membrane compartment in islet beta cells and adipocytes. Moreover, co-immunoprecipitation and GST interaction assays showed Doc2β-Munc18c binding to be direct and complexes to be devoid of Syntaxin 4. Supporting the notion of Munc18c binding with Syntaxin 4 and Doc2β in mutually exclusive complexes, in vitro competition with Syntaxin 4 effectively displaced Munc18c from binding to Doc2β. The second C2 domain (C2B) of Doc2β and an N-terminal region of Munc18c were sufficient to confer complex formation. Disruption of endogenous Munc18c-Doc2β complexes by addition of the Doc2β binding domain of Munc18c (residues 173–255) was found to selectively inhibit glucose-stimulated insulin release. Moreover, increased expression of Doc2β enhanced glucose-stimulated insulin secretion by ∼40%, whereas siRNA-mediated depletion of Doc2β attenuated insulin release. All changes in secretion correlated with parallel alterations in VAMP2 granule docking with Syntaxin 4. Taken together, these data support a model wherein Munc18c transiently switches from association with Syntaxin 4 to association with Doc2β at the plasma membrane to facilitate exocytosis.

included (24). These data suggest that a cellular factor may be required to mediate the transient displacement of Munc18c.
There is precedence for the existence of proteins that bind and impact SM-Syntaxin complexes. For example, the yeast protein Ypt1p, a Rab GTPase, binds the Sly1p-Sed5p complex, yeast SM-Syntaxin homologues (25). In neurons, the SM protein Munc18-1 has several known binding proteins other than its cognate Syntaxin, several of which are C2 domain-containing proteins (26 -28). One of these is Doc2␤, which was shown to bind directly to Munc18-1 (27). Doc2␤ binds to a range of ligands including calcium, phospholipids, and intracellular proteins. Whereas a second isoform, Doc2␣, is found primarily in brain/neuronal tissue, Doc2␤ is more widely expressed (27, 29 -32).
In this report we demonstrate that Doc2␤ is expressed in insulin-secreting beta cells as well as insulin-responsive adipocytes, in which it associates with Munc18c in a manner mutually exclusive of Munc18c's other known binding partner Syntaxin 4. Our results delineate binding domains of each protein sufficient to confer the interaction and further show that endogenous Munc18c-Doc2␤ association is functionally important for glucose-stimulated insulin secretion. Loss of association was coupled to decreased SNARE docking (Syntaxin 4 association with VAMP2 granules), whereas increased association or Doc2␤ expression was correlated with increased SNARE docking. In vitro studies revealed that Syntaxin 4 could displace Munc18c from preformed Munc18c-Doc2␤ complexes in a dose-dependent fashion, altogether supportive of a model in which Munc18c switches binding partners from Syntaxin 4 to Doc2␤ at the plasma membrane to promote exocytosis.

EXPERIMENTAL PROCEDURES
Materials-Rabbit anti-Munc18c and anti-GLUT4 antibodies were generated as previously described (9). The rabbit polyclonal anti-Syntaxin 4 and mouse monoclonal anti-VAMP2 antibodies were obtained from Chemicon (Temecula, CA) and Synaptic System (Gottingen, Germany), respectively. The rabbit polyclonal anti-Doc2 antibody was a kind gift from Dr. Matthijs Verhage (Vrije Universiteit, Netherlands). FLAG and clathrin antibodies were obtained from Sigma and BD Transduction Labs (Franklin Lakes, NJ), respectively. Rabbit and mouse anti-GFP antibodies were acquired from Abcam (Cambridge, MA) and Clontech Laboratories (Mountain View, CA), respectively. Rabbit polyclonal GFP and Syntaxin 1A antibodies were purchased from Upstate Biotechnology (Lake Placid, NY). The monoclonal anti-Myc (9E10) antibody and protein G plus agarose were obtained from Santa Cruz Biotechnology (Santa Cruz, CA), respectively. MIN6 cells were a gift from Dr. John Hutton (University of Colorado Health Sciences Center). Goat anti-mouse and anti-rabbit horseradish peroxidase secondary antibodies and transfectin lipid reagent were acquired from Bio-Rad. Lipofectamine 2000 was purchased from Invitrogen (Carlsbad, CA). Radioimmunoassay (RIA) grade bovine serum albumin and D-glucose were purchased from Sigma. Enhanced chemiluminescence reagent and Hyperfilm-MP were obtained from Amersham Biosciences. The human C-peptide and rat insulin RIA kits were purchased from Linco Research Inc (St. Charles, MO). Doc2␤ siRNA oligonucleotides were purchased from Ambion (Austin, TX): siDoc2 number 2-GGCAAAUAAGCUCAGAAC-Att; siDoc2 number 1-UCAUCACACACGGAGAUCCtc.
Cell Culture, Transient Transfection, and Secretion Assays-MIN6 beta cells were cultured in Dulbecco's modified Eagle's medium (DMEM with 25 mM glucose) 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 (15). MIN6 beta cells at 50 -60% confluence were transfected with 40 g of plasmid DNA per 10 cm 2 dish using transfectin (Bio-Rad) to obtain ϳ30 -50% transfection efficiency. After 48 h of incubation, cells were washed twice with and incubated for 2 h in freshly prepared modified Krebs-Ringer bicarbonate buffer (MKRBB: 5 mM KCl, 120 mM NaCl, 15 mM Hepes pH 7.4, 24 mM NaHCO 3 , 1 mM MgCl 2 , 2 mM CaCl 2 , and 1 mg/ml BSA). Cells were stimulated with 20 mM glucose or 50 mM KCl for the times indicated in the figures. Cells were subsequently lysed in Nonidet P-40 lysis buffer (25 mM Tris, pH 7.4, 1% Nonidet P-40, 10% glycerol, 50 M sodium fluoride, 10 mM sodium pyrophosphate, 137 mM sodium chloride, 1 mM sodium vanadate, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, 1 g/ml pepstatin, and 5 g/ml leupeptin), and lysates were cleared by microcentrifugation for 10 JULY 27, 2007 • VOLUME 282 • NUMBER 30 min at 4°C for subsequent use in co-immunoprecipitation experiments. For measurement of human C-peptide release, MIN6 beta cells were transiently co-transfected with each plasmid plus human proinsulin cDNA (kind gift from Dr. Chris Newgard, Duke University), using transfectin with 2 g of each DNA per 35-mm dish of cells at 50% confluence. 48 h following transfection, cells were preincubated for 2 h in MKRBB buffer and stimulated with 20 mM glucose for 1 h. MKRBB was collected for quantitation of human C-peptide released.

Munc18c-Doc2␤ Complexes Function in Exocytosis
Transfection of siRNA oligonucleotides into MIN6 cells was achieved using Lipofectamine 2000 (Invitrogen) with 100 nM oligonucleotides to obtain ϳ70 -80% transfection efficiency. A non-targeting RNA (scrambled siRNA, obtained from Ambion) was included as a control in parallel experiments. Transfected cells were maintained in supplemented DMEM for 48 h, starved in MKRBB and stimulated as described above, and insulin-secreted into the MKRBB quantitated by RIA. Cells were harvested in 1% Nonidet P-40 lysis buffer for detecting Doc2␤ depletion.
CHO-K1 cells were purchased from the American Type Culture collection (Manassas, VA) and cultured in Ham's F-12 medium supplemented with 10% fetal bovine serum, 100 units/ml penicillin, 100 g/ml streptomycin, and 292 g/ml L-glutamine. At 80 -90% confluence, cells were electroporated with 40 g of DNA as previously described (9). After 48 h of incubation, cells were harvested in 1% Nonidet P-40 lysis buffer and lysates cleared by centrifugation at 14,000 ϫ g for 10 min at 4°C for subsequent use in coimmunoprecipitation experiments.
Subcellular Fractionation-Subcellular fractions of beta cells were isolated as described previously (34). Briefly, MIN6 beta cells at 80 -90% confluence were harvested into 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). Cells were disrupted by 10 strokes through a 27-gauge needle, and homogenates were centrifuged at 900 ϫ g for 10 min. Postnuclear supernatants were centrifuged at 5,500 ϫ g for 15 min, and the subsequent supernatant centrifuged at 25,000 ϫ g for 20 min to obtain the secretory granule fraction in the pellet. The supernatant was further centrifuged at 100,000 ϫ g for 1 h to obtain the cytosolic fraction. Plasma membrane fractions (PM) were obtained by mixing the postnuclear pellet 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 6,000 ϫ g for 10 min, and the resulting pellet was collected as the plasma membrane fraction. All pellets were resuspended in 1% Nonidet P-40 lysis buffer to solubilize membrane proteins.
Subcellular fractions of 3T3L1 adipocytes were obtained using the differential centrifugation method as described pre-viously (9). 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 low speed (HDM) fraction was obtained by centrifugation of the resulting supernatant at 41,000 ϫ g for 20 min at 4°C. The supernatant was removed and centrifuged at 180,000 ϫ g for 75 min at 4°C to generate the high speed (LDM) fraction. The plasma membrane fraction (PM) was obtained by resuspending the pellet from the initial 19,000 ϫ g centrifugation in HES buffer followed by layering onto a 1.12 M sucrose cushion for centrifugation at 100,000 ϫ g for 60 min. The plasma membrane layer was then removed from the cushion and centrifuged at 40,000 ϫ g for 20 min, and that pellet then resuspended in HES buffer.
Co-immunoprecipitation and Immunoblotting-MIN6 beta cells were preincubated in MKRBB for 2 h followed by glucose stimulation. 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-Syntaxin4 antibody, or rabbit anti-Doc2␤ antibody for 2 h at 4°C followed by a second incubation with protein G Plusagarose for 2 h. The resultant immunoprecipitates were subjected to 10% SDS-PAGE followed by transfer to PVDF membranes for immunoblotting. Munc18c, Syntaxin 4, and Doc2␤ antibodies were used at 1:5000, 1:500, and 1:1000 dilutions, respectively, and secondary antibodies conjugated to horseradish peroxidase were diluted at 1:5000 for visualization by chemiluminescence. Immunoprecipitations using CHO-K1 detergent-cleared cell lysates were performed similar to that of the MIN6 cell lysates.
Adenoviral Transduction of MIN6 Cells-MIN6 cells at 60% confluence were transduced with pAd5CMV-Munc18c CsClpurified particles (generated by the University of Iowa Gene Targeting Vector Core, Iowa City, IA) for 2 h at 37°C (MOI ϭ 100). Transduced cells were then washed twice with phosphate-buffered saline and incubated for 48 h in complete medium at 37°C, 5% CO 2 . Transduced cells were subsequently preincubated in MKRBB for 2 h and stimulated with 20 mM D-glucose for 5 min. Cleared detergent lysates were prepared as described above for the GST pull-down assay.
Statistical Analysis-All data are expressed as mean Ϯ S.E. Data were evaluated for statistical significance using the Student's t test.

Doc2␤ and Munc18c
Associate in MIN6 Beta Cells-Doc2␤ is considered to be a ubiquitously expressed protein, enriched in brain, heart, and lung tissues (27,31), but expression in adipocytes and islet cells has not yet been established. To address this, tissue and cultured cell lysates were used for immunoblotting to detect the presence of Doc2␤ protein using a previously validated antibody provided by Dr. Matthias Verhage. The Doc2␤ antibody recognized a single 45-kDa band in both human and mouse islet lysates, MIN6 beta cell lysate and recombinant purified full-length Doc2␤ protein (Fig. 1A). Moreover, of the three subcellular fractions of MIN6 beta cells examined, Doc2␤ was almost exclusively localized to the PM fraction (Fig. 1A, lane 4), and was nearly undetectable in granule and cytosolic fractions (data not shown). The purity of the fractions was validated by the presence of the PM protein Syntaxin 1A exclusively in the plasma membrane fraction (Fig. 1A) and by insulin content as we have documented previously (15,34).
In 3T3L1 adipocytes, Doc2␤ was also primarily localized to the PM fraction (Fig. 1B), with very little present in intracellular vesicle fractions (LDM and HDM). Subcellular fractions prepared from insulin-stimulated 3T3L1 adipocytes were validated by detection of GLUT4 protein in PM, LDM, and HDM fractions and detection of Syntaxin 4 protein principally in the PM fraction. RT-PCR was also used to confirm the presence of Doc2␤ mRNA in the MIN6 cells (data not shown). Thus, Doc2␤ and Munc18c both localize to the PM fraction in cell types where Munc18c-Syntaxin 4 complexes are known to be important for regulated exocytotic events.
To next determine whether Doc2␤ could bind to Munc18c, Doc2␤ was expressed as a GST fusion protein in E. coli and attached to beads as bait for precipitating interacting proteins. To increase the abundance of Munc18c and enhance detection of a binding event, MIN6 lysates prepared from cells transduced to express recombinant FLAG-tagged Munc18c were initially used. Both anti-FLAG and anti-Munc18c antibodies detected Munc18c in precipitates with GST-Doc2␤ but not GST control ( Fig. 2A). Coimmunoprecipitation was used as an independent approach to confirm this interaction. MIN6 cell lysates prepared from cells transfected to express recombinant Myc-tagged Doc2␤ were immunoprecipitated with anti-Myc antibody or with the vector control (pcDNA3.1-myc-his). As shown in Fig. 2B, Doc2␤-myc was able to co-precipitate endogenous Munc18c, whereas the vector control had no effect. Finally, to determine if endogenous Doc2␤-Munc18c complexes formed, anti-Munc18c antibody was used to co-precipitate Doc2␤ (Fig. 2C). Immunoprecipitation with an IgG control verified the specificity of the interaction. Parallel studies conducted in 3T3L1 adipocyte lysates gave identical results, indicating that the endogenous complex is not specific for islet beta cells (data not shown). Moreover, the interaction between recombinant tagged forms of Munc18c and Doc2␤ occurred   JULY 27, 2007 • VOLUME 282 • NUMBER 30 irrespective of whether epitope tags were placed at the N or C termini of either protein, suggesting that the interaction may occur through more internal regions of each.

Munc18c-Doc2␤ Complexes Function in Exocytosis
The C2B Domain of Doc2␤ Is Sufficient to Mediate Direct Binding to Munc18c-To determine whether the association of Doc2␤ and Munc18c is direct or indirect, we combined bacterially expressed and purified recombinant proteins in an in vitro binding assay. A truncated form of GST-Doc2␤ (residues 123-375) was found to confer binding to His-tagged Munc18c (Fig.  3A), indicating that the Doc2␤ N-terminal Munc13-interacting domain (MID) was dispensable for interaction with Munc18c, similar to an earlier finding with Doc2␤-Munc18-1 interaction (27). To further delineate the minimal binding domain of Doc2␤ sufficient to confer direct binding to Munc18c, each individual C2 domain was tested (Fig. 3B). Whereas the GST-C2A protein was incapable of interaction with Munc18c, the GST-C2B protein showed nearly equal affinity for Munc18c as did the full GST-C2AB protein. This binding specificity of Munc18c for the C2B domain of Doc2␤ is distinctly different from that of Munc18-1, which was shown to associate with the C2A domain (27). This suggests that unlike the Munc18-isoform specific binding to cognate Syntaxins, Doc2␤ is not discretely paired with one particular Munc18 protein, but that its interaction with each Munc18 protein is partitioned by differential affinity of each to different C2 domains.
The N Terminus of Munc18c Is Required for Binding to Doc2␤-To determine the minimal binding domain of Munc18c required for its interaction with Doc2␤, a deletion series of Munc18c-GFP was used as previously described (23). Munc18c-GFP mutants were co-electroporated with Doc2␤myc into CHOK1 cells, because CHOK1 cells transfect very efficiently (Fig. 4A). Lysates prepared from transfected cells were used in anti-Myc (Doc2␤) immunoprecipitation reactions to determine which region of Munc18c conferred the highest level of interaction (Fig. 4B, lanes 1-6). Although both N-terminal regions containing residues 1-172 or 173-255 of Munc18c could be co-precipitated by Doc2␤-myc, the 173-255 region showed consistently higher binding affinity to recombinant Myc-tagged Doc2␤ (Fig. 4B, lanes 8 and 9); in three of five experiments no binding of the 1-172 fragment to GST-Doc2␤ was observed, whereas in the remaining two experiments a low level of binding was observed. This was not caused by discrepancies in protein expression or batch of cells or DNA used for electroporation. Thus it appears there could be two binding sites, but the affinity of the 1-172 fragment for GST-Doc2␤ is reproducibly lesser than that of the 173-255 region. The specificity of this binding was validated by lack of binding of GFP alone to Doc2␤-myc (Fig. 4B, lane 12). In addition, full-length Munc18c-GFP bound less well than did the 173-255 region (Fig. 4B, lane 7), suggesting that the smaller isolated region may adopt a more accessible conformation. Interestingly, this 173-255 region is known to contain a pivotal tyrosine at position 219 that we have recently demonstrated to be required to mediate Munc18c dissociation from Syntaxin 4 (23). Thus the region required to displace Munc18c from Syntaxin 4 is also the region of Munc18c sufficient to accept binding to Doc2␤, consistent with a switch mechanism of Munc18c binding to either in a mutually exclusive manner.
Syntaxin 4 Is Excluded from the Munc18c-Doc2␤ Complex-To test the notion that the Munc18c-Doc2␤ complex is mutually exclusive of Munc18c-Syntaxin 4 complex, GST-Doc2␤ linked to beads was incubated with MIN6 cell lysates and eluted proteins were immunoblotted with anti-Munc18c and anti-Syntaxin 4 antibodies. As shown in Fig. 5A, endogenous Munc18c but not Syntaxin 4 was precipitated from MIN6 cell lysates.
The inability of Syntaxin 4 to coprecipitate with GST-Doc2␤ and Munc18c may have been caused by issues of protein abundance or additional cellular factors, because the assay was conducted using MIN6 cell lysates. Therefore, to test the notion that Munc18c forms mutually exclusive complexes with Doc2␤ and Syntaxin 4, we performed in vitro competition assays (Fig.  5B). GST-Doc2␤ coupled to beads was initially preincubated with His-tagged Munc18c protein for 2 h at 4°C. GST-Doc2␤-Munc18c complexes were then pelleted by centrifugation and unbound excess Munc18c washed away. The presence of unbound Munc18c demonstrated that Munc18c protein was not limiting for the formation of complex (Fig. 5B, lanes 1-4). Increasing amounts of soluble recombinant Syntaxin 4 protein were subsequently added for an additional 2 h. A 3-fold molar excess of Syntaxin 4 resulted in dissociation of more than 60% of the His-Munc18c from GST-Doc2␤ (Fig. 5B, lane 4). Consistent with data from the GST-Doc2␤ interaction studies using cell lysate above, a negligible amount of recombinant Syntaxin 4 protein was seen to precipitate with either GST-Doc2␤ or GST alone, with greater than 99% of protein left unbound. Control experiments conducted using Syntaxin 1A, a non-Munc18cbinding protein, in place of Syntaxin 4 as the competitor failed to show disruption of the interaction between GST-Doc2␤ and His-Munc18c (Fig. 5B, lane 7), indicating that the competition by Syntaxin 4 is not simply the result of increased and nonspecific protein interference in the assay. Furthermore, the above experiments were repeated using the truncated GST-Doc2␤-C2AB protein and gave identical results. The reciprocal experiment gave similar results: GST-Syntaxin 4 failed to precipitate recombinant His-Doc2␤ protein (Fig. 5C). Thus, under in vitro conditions, we have established a switching binding mode of Munc18c to Syntaxin 4 and to Doc2␤. Though the molar ratio threshold for Syntaxin 4 to completely compete off Munc18c is high in vitro, it is possible that there might be cellular signals required to trigger the switch of binding partner in vivo.
Munc18c-Doc2␤ Complexes Exist Only at the Plasma Membrane-We have shown previously that the dissociation of Munc18c from Syntaxin 4 is only observed from the PM compartment of pancreatic beta cells or 3T3L1 adipocytes although displaced Munc18c was never seen to move into either cytosolic or granule compartments (23). Subcellular fractions prepared from MIN6 cells contained Doc2␤, Munc18c, and Syntaxin 4 in large abundance in the PM fraction (Fig. 6A, lane 1). Trace amounts of Doc2␤ were also localized to the cytosolic fraction, which happens to be where ϳ50% of total cellular Munc18c is found (Fig.  6A, lane 3). Co-immunoprecipitations were performed from these subcellular fractions using anti-Munc18c or anti-Doc2␤ antibodies. Munc18c co-immunoprecipitated both Syntaxin 4 and Doc2␤ exclusively from the PM fraction, even though Doc2␤ and Munc18c were both also present in the cytosolic fraction (Fig. 6B, lanes 1 and 3). Reciprocal immunoprecipitation of Doc2␤ similarly resulted in co-precipitation of Munc18c from the PM fraction but failed to precipitate Syntaxin 4 (Fig. 6B, lane 4). No net changes in abundance of complexes formed between Munc18c, Syntaxin 4, or Doc2␤ in response to glucose stimulation were observed however (data not shown). These data might suggest that complexes are highly transient and/or that there is no net switch of Munc18c to Doc2␤ at any one particular point in time during exocytosis.
Endogenous Munc18c-Doc2␤ Complexes Are Essential for Insulin Exocytosis-To evaluate the requirement for the endogenous Doc2␤-Munc18c complexes in exocytosis we used the minimal Munc18c region (173-255) as a competitive inhibitor in a human C-peptide reporter assay. This small region of Munc18c fails to bind Syntaxin 4 (data not shown). MIN6 cells were co-transfected with Munc18c-(173-255) or vector control together with human proinsulin cDNA. Human C-peptide (derived from human proinsulin) is synthesized and packaged in an identical fashion to mouse C-peptide and insulin in granules, but is immunologically distinct from mouse C-peptide and serves as a reporter of secretion from transfectable cells (36,37). MIN6 cells transfected with vector control exhibited 130% of basal human  JULY 27, 2007 • VOLUME 282 • NUMBER 30

JOURNAL OF BIOLOGICAL CHEMISTRY 21791
C-peptide secretion in response to glucose (Fig. 7A, bars 1  and 2). In contrast, cells transfected to express the 173-255 fragment showed abolished glucose-stimulated secretion (Fig.  7A, bars 2 and 3), and was without effect upon basal secretion (1.1 Ϯ 0.1-fold of control, n ϭ 3). However, KCl-stimulated human C-peptide release was not affected by the presence of the 173-255 region (Fig. 7A, bars 4 and 5), suggesting that the Doc2␤-Munc18c is required only for glucose-stimulated exocytosis, and that expression of this region did not have a global dampening effect upon exocytosis events in general.
To gain mechanistic data underlying this alteration in function we assessed SNARE complex assembly (Syntaxin-VAMP association) in MIN6 lysates prepared from cells transfected with the 173-255 region or vector control. As we have shown previously (23), glucose stimulation for 5 min increased the co-immunoprecipitation of VAMP2-granules with Syntaxin 4 in lysates prepared from vector-treated cells (Fig. 7B). However, MIN6 cells overexpressing Munc18c-(173-255) abolished glucose-stimulated Syntaxin 4-VAMP2 association (Fig. 7C), but was without significant effect upon unstimulated levels of Syntaxin 4-VAMP2 association. SNARE complex assembly in unstimulated cells has been reported to represent predocked granules used for the initial burst of insulin release that can be stimulated by either glucose or KCl (38). Taken together, these data suggest that endogenous Munc18c-Doc2␤ interaction is functionally important for regulating glucose-stimulated VAMP2-granule docking with Syntaxin 4, but not necessary for predocking granules.
In a second approach to determine the requirement for Doc2␤ we utilized siRNA-mediated depletion. Two different commercially available siRNA oligonucleotides were transiently transfected into MIN6 cells (designated as siDoc2␤) using the reagent Lipofectamine 2000 to obtain greater than 80% of cells transfected, as determined using fluorescently labeled oligonucleotides. Depletion by one of the two oligonucleotides, designated siDoc number 1, resulted in ϳ40% reduction in Doc2␤ protein (Fig. 8A). This depletion corresponded to ϳ40% reduction of glucose-induced insulin release relative to secretion in siControl-transfected cells (Fig. 8B). Transfection with the second siRNA, siDoc number 2, resulted in no signif-  icant depletion of endogenous Doc2␤ protein or upon insulin release, and thus all subsequent studies used siDoc number 1.
The effect of Doc2␤ depletion on Syntaxin 4-Munc18c association and Syntaxin 4-VAMP2 association was also examined. Consistent with the in vitro data showing competition between Syntaxin 4 and Doc2␤ for binding to Munc18c, Fig. 8C shows that Syntaxin 4-Munc18c association was significantly increased in glucose-stimulated siDoc2␤-treated cells (53 Ϯ 12%, p Ͻ 0.05, compared with siCon-treated cells). In addition, siDoc2␤-treated cells failed to show any glucose-stimulated increase in SNARE complex assembly (quantified in Fig. 8D), compared with appropriate 2-fold increase in glucose-stimulated VAMP2 granule association with Syntaxin 4 in siControltreated cells. Similar to results obtained using the 173-255 region in this assay, siDoc2␤ treatment had no significant effect upon unstimulated SNARE complex assembly nor Syntaxin 4-Munc18c association. Specific depletion of Doc2␤ was confirmed by immunoblotting of lysates using anti-Doc2␤ and -clathrin antibodies. Taken together these data indicate that reduction of Doc2␤ levels directly impact the ability of Munc18c to associate with Syntaxin 4, and that Doc2␤ is required for Syntaxin 4-mediated exocytosis of glucose-stimulated granules.
Doc2␤ Functionally Enhances Insulin Exocytosis and SNARE-mediated Vesicle Docking-To further investigate the relationship of the Doc2␤ role in exocytosis with its role as a Munc18c-binding protein, we tested the ability of Doc2␤ overexpression to relieve the inhibition upon insulin secretion caused by increased Munc18c expression. We have shown previously that overexpression of Syntaxin 4 can act as a molecular sponge to bind excess Munc18c and resume normal exocytosis, and that non-Munc18c-binding proteins, or a non-PM localized form of Syntaxin 4 fail to rescue exocytosis (23,39). As shown in Fig. 9A, Munc18c overexpression reduced glucosestimulated human C-peptide release by ϳ50% (119% of basal versus 141% attained in vector-expressing cells, p Ͻ 0.05). Coexpression of Doc2␤ with Munc18c fully counteracted this inhibition (Fig. 9A, bars 6 versus 4), consistent with the Doc2␤ role as a PM-localized Munc18c-binding factor. Coordinately, expression of Munc18c-myc dramatically decreased VAMP2 association with Syntaxin 4, under both unstimulated and stimulated conditions (Fig. 9B, lanes 3 and 4). Of note, the recombinant Munc18c-myc protein is not recognized by the Munc18c antibody (antibody raised against the far C terminus of Munc18c protein), such that the Munc18c immunoblot in Fig. 9B represents only endogenous Munc18c binding with Syn4. Expression of Myc-tagged forms of Munc18c and Doc2␤ in MIN6 lysates was confirmed by anti-Myc immunoblotting. Consistent with the secretion results, the addition of Doc2␤myc protein in Munc18c-myc overexpressing cells restored the level of VAMP2 co-precipitated by anti-Syntaxin 4 antibody in unstimulated lysates to the control levels (Fig. 9B, compare  lanes 1, 3, and 5). However, there was no difference between unstimulated and glucose-stimulated VAMP2 association with Syntaxin 4 in cells overexpressing both Doc2␤-myc and Munc18c-myc (Fig. 9B, lanes 5 and 6).
Remarkably, Doc2␤ overexpression alone caused release of ϳ40% more human C-peptide compared with vector-transfected cells (Fig. 9A, bars 2 versus 8), and without significantly increasing basal secretion. Consistent with this, Doc2␤ overexpression in MIN6 cells reduced association of Syntaxin 4 with endogenous Munc18c (Fig. 9B, lanes 7 and 8) under both unstimulated as well as glucose-stimulated conditions (by 23 Ϯ 8% and 41 Ϯ 4% under unstimulated and stimulated conditions, respectively, compared with vector-expressing cells). By contrast, Doc2␤ overexpression resulted in significantly increased Syntaxin 4-VAMP2 co-precipitation from both unstimulated and glucose-stimulated lysates (Fig. 9B, quantified in Fig. 9C). These data suggest that the stimulatory effect of Doc2␤ overexpression in glucose-stimulated insulin secretion correlated with its ability to increase the docking of granules at Syntaxin 4 sites.

DISCUSSION
In this report we have identified Doc2␤ as a new binding partner for the exocytotic protein Munc18c and demonstrated an essential functional role for this complex in insulin granule exocytosis. Doc2␤ was found localized to the plasma membrane compartment in both cell types known to utilize Munc18c for exocytosis, islet beta cells and 3T3L1 adipocytes, and bound to Munc18c exclusively in this cellular compartment. The N-terminal residues 173-255 of Munc18c bound directly to the Doc2␤ second C2 domain, and expression of the 173-255 region of Munc18c or siRNA-mediated depletion of Doc2␤ significantly impaired glucose-stimulated insulin exocytosis. Syntaxin 4 failed to bind Munc18c-Doc2␤ complexes, and in vitro binding studies revealed that Syntaxin 4 could competitively inhibit Munc18c binding to Doc2␤. Furthermore, increased expression of Doc2␤ was found to enhance glucose-stimulated insulin exocytosis, resultant from increased VAMP2granule docking at Syntaxin 4 sites, and decreased Munc18c occupancy of Syntaxin 4 sites. Taken together, these data support a model whereby Munc18c switches between binding partners Syntaxin 4 and Doc2␤ at the plasma membrane, such that Munc18c remains in close proximity to the SNARE fusion apparatus to exert its function in regulated exocytosis.
SM proteins are proposed to function in catalysis of vesicle fusion, fusion being the final and irreversible step of vesicle trafficking. However SM proteins are also thought to add another layer of specificity of vesicle docking in tandem with Rab-like proteins. In yeast the Sly1p-Sed5p complex, analogous to SM-Munc18 complexes, affinity is reduced by the addition of the Rab-like protein Ypt1p to promote SNARE pairing and fusion (40,41). Similarly, the Munc18-1-Syntaxin 1A complex crystal structure shows a putative Rab-interacting site in the loop connecting domain 2 (composed of residues 135-245 plus 480 -592) with domain 3b of Munc18-1 that would serve as a hinge mechanism for releasing the constrictive hold on Syntaxin 1A (18). Consistent with this, we have previously shown that mutation within domain 2 reduces Munc18c affinity for Syntaxin 4 in 3T3L1 adipocytes (33). However, neither candidate Rab proteins Rab4 found on GLUT4 vesicles nor Rab3A in beta cells have been confirmed to carry out this disruptive function.
Alternatively, the calcium-phospholipid-binding Doc2␤ protein has also been localized to vesicles in neuronal cell types and shown to bind to Munc18-1 via its first C2 domain (C2A) (27,42), making it a suitable candidate as a Munc18c-Syntaxin 4 displacement factor. However, two key findings presented here suggest a novel role for Doc2␤ function in endocrine cell exocytosis: 1) Munc18c does not bind the C2A domain of Doc2␤, but rather binds to the C2B domain instead; 2) Doc2␤ localized principally to the plasma membrane fractions in both islet beta cells and 3T3L1 adipocytes. Moreover, in PC12 and chromaffin cells Doc2␤ on vesicles translocates to the plasma membrane upon Ca 2ϩ stimulation to facilitate vesicle priming (32,(43)(44)(45). In contrast, we did not observe this translocation event in MIN6 beta cells, a cell type well known to elicit insulin exocytosis in response to Ca 2ϩ stimulation. Neither did we observe Doc2␤ translocation in stimulated 3T3L1 adipocytes (data not shown). Given the persistent plasma membrane localization of Munc18c in both MIN6 beta cells and 3T3L1 adipocytes, it seemed likely that any factor binding to displaced Munc18c would also be localized to the plasma membrane. The Doc2␤ plasma membrane localization in both beta cells and adipocytes fits such a prediction well. Thus, the mechanisms underlying alterations of Munc18c-Syntaxin 4 complex conformation likely differ significantly from those of yeast and neuronal cells.
Although there is a high degree of sequence homology between family members within the syntaxin and SM protein families, there are notable differences that have been proposed to exert conformational changes that significantly alter how the complexes form and function. For example, comparison of Syntaxin 4 with Syntaxin 1A suggests there are differences in Syntaxin 4 contacts between the Habc and H3 domains that will alter its presentation to the variable hairpin region in domain 3a of its SM partner (Munc18c) (18). In addition, very recent reports have shown that the Munc18-1 isoform binds to the far N terminus of Syntaxin 1A to allow Syntaxin 1A to adopt a transitional conformation (20,21). A KDR motif present in the far N terminus of Syntaxin 1A mediates its interaction with Munc18-1 (46). However, we cannot necessarily expect Syntaxin 4 to bind and function as does Syntaxin 1A, since Syntaxin 4 lacks the KDR motif.
Our in vitro competition studies demonstrated that Syntaxin 4 displaced Munc18c from binding to Doc2␤ (IC 50 ϳ 16 nM) suggesting that in the cell Doc2␤ might function either as an acceptor of displaced Munc18c, or play a more active role by inducing the disruption of Munc18c-Syntaxin complexes. However, no significant increase in abundance of Munc18c-Doc2␤ complexes following glucose stimulation in MIN6 beta cells was detected, analogous to Munc18c-Syntaxin 4 complexes (23). There are several possibilities for this: 1) Munc18c is not fully displaced from Syntaxin 4 in vivo; 2) the switching is highly transient and significant net changes in complex abundances are not reached at any given time point. In the case of the first possibility, it has been shown in vitro that Munc18c may remain attached to Syntaxin 4 during SNARE complex assembly (24); however, this has yet to be demonstrated in vivo in these endocrine cell types. Alternatively, SNARE complex interactions in vitro are known to occur more promiscuously than in vivo (3). This may be related to the absence of posttranslational modifications, such as stimulus-induced phosphorylation of Munc18c, occurring only in vivo, that displace it from Syntaxin 4 (23). Even in cells the phosphorylation event was highly transient, requiring phosphatase inhibitor to trap phospho-Munc18c to detect its displacement from Syntaxin 4 (23). Thus, this second possibility is particularly intriguing because the region of Munc18c found sufficient to confer Doc2␤ binding contains the regulatory Tyr 219 residue.
Another possible trigger of Munc18c-Doc2␤ association could be elevated [Ca 2ϩ ] i . Granule priming is known to require elevated [Ca 2ϩ ] i in islet beta cells, and if Doc2␤ is involved in glucose-induced priming of granules then our data showing loss of insulin release from Doc2␤-depleted cells would support FIGURE 9. Overexpression of Doc2␤ enhances insulin exocytosis and VAMP2 granule docking with Syntaxin 4. A, MIN6 cells were transfected with pcDNA3 vector control, pcDNA3-Munc18c/myc-His, pcDNA3-Doc2␤/ myc-His, or both together plus human proinsulin DNA. After 48 h of incubation, cells were preincubated in MKRBB for 2 h followed by stimulation with 20 mM glucose. Human C-peptide secreted into the media was measured by RIA and normalized for total cellular protein content. Data in each of four independent experiments were normalized to basal ϭ 1 and are shown as the average Ϯ S.E.; *, p Ͻ 0.05, versus Munc18c with glucose; **, p Ͻ 0.05 versus vector with glucose. B, overexpression of Doc2␤ impairs Syntaxin 4-Munc18c binding to enhance VAMP2 association with Syntaxin 4. Detergent cell lysates prepared from unstimulated (0) or glucose-stimulated (5 min) MIN6 cells transiently transfected with vector, Munc18c, Doc2␤, or both together were used for immunoprecipitation with anti-Syntaxin 4 antibody. Immunoprecipitated proteins were resolved on 12% SDS-PAGE for immunodetection of Munc18c and VAMP2. Anti-Syntaxin 4 immunoblot shows equal Syntaxin 4 precipitation in reactions. Expression of Doc2␤-myc and Munc18c-myc proteins in lysates was confirmed by anti-Myc immunoblotting. Data are representative of three independent experiments. C, optical density scanning quantitation of VAMP2 associated with Syntaxin 4, normalized to the level of association under unstimulated conditions in each of three independent experiments. *, p Ͻ 0.02, versus vector-transfected cells under basal conditions; **, p Ͻ 0.05, versus vector-transfected cells under stimulated conditions.

Munc18c-Doc2␤ Complexes Function in Exocytosis
this. Upon binding to calcium, the affinity of Doc2␤ for binding to phospholipids is significantly increased (47), and the C2A but not C2B domain of Doc2␤ mediates this in vitro (29). Interestingly, Doc2␤ also associated with the Munc18-1 isoform through this C2A domain in neurons, in a calciumindependent fashion (27). However our data clearly show that Doc2␤ binds to Munc18c through the C2B domain, unlike its interaction with Munc18-1. Given this distinction it may be that Munc18c-Doc2␤ binding in beta cells is calcium-dependent.
Our observation that both disrupting endogenous Munc18c-Doc2␤ complex and depletion of endogenous Doc2␤ led to reduced Syntaxin 4-VAMP2 association specific to glucose stimuli suggests that Munc18c-Doc2␤ interaction is dispensable for pre-docking of granules but is critically important for docking of mobilized insulin granules. This interpretation stems from several studies showing a correlation between the co-immunoprecipitation of VAMP2 with Syntaxin 4 from unstimulated beta cell lysates with the abundance of predocked granules, those granules released in response to [Ca 2ϩ ] i , elevated by either glucose or KCl in the first phase of insulin release (23,38,49). In contrast, overexpression of Doc2␤ enhanced both basal and glucose-stimulated VAMP2 association with Syntaxin 4. The mechanism behind this could lie in the ability of Doc2␤ to compete Munc18c away from Syntaxin 4 in the absence of stimulus, but could also be mediated by its interaction with other molecules such as Munc13-1, which has been demonstrated to influence both first phase and second phase insulin secretion in studies using islet cells isolated from Munc13-1 heterozygous knockout mice (50).
Doc2␤ was found to function as a Munc18c-binding protein in a rescue assay, whereby the inhibition of exocytosis caused by overexpression of Munc18c can be relieved and normal exocytosis restored by co-expression of the Munc18c-binding protein. We have previously shown that co-expression of Syntaxin 4 rescues insulin secretion as well as VAMP2 granule docking (23). However, while Doc2␤ overexpression also rescued secretion and predocked VAMP2 granules at Syntaxin 4 sites, it failed to fully restore increased docking of glucose-mobilized granules. One possible explanation could be that protein stoichiometry of Munc18c: Doc2␤ in islet cells is crucial for insulin granule docking and exocytosis, and alterations of this stoichiometry by overexpression of both proteins did not necessarily reconstitute the proper Munc18c:Doc2␤ complex molar ratio. This was not too surprising because protein stoichiometry of SNARE accessory proteins has been shown to be important for maintaining normal granule exocytosis events (48,51). Future islet perifusion studies will be required to discern whether Munc18c, Doc2␤, and the Munc18c-Doc2␤ complex are required for first phase or second phase insulin secretion.
In summary, we have identified Doc2␤ as a new binding partner for Munc18c in endocrine cells and have demonstrated a functional requirement for the Doc2␤-Munc18c complex exocytosis and SNARE core complex formation. We have further revealed that Doc2␤ can compete with the Munc18c cognate SNARE protein, Syntaxin 4. Doc2␤ has the unique ability to compete with Syntaxin 4 for Munc18c binding, and may represent the missing factor in the mechanism by which Munc18c facilitates the transition of Syntaxin 4 from a monomeric closed conformation into its fusion competent form. Because Munc18c, Doc2␤ and Syntaxin 4 are ubiquitously expressed, the mechanism identified here may be conserved throughout numerous secretory events, particularly in endocrine cells.