Introduction

Cysteine string proteins (CSPs) are highly conserved proteins associated with synaptic vesicles (Zinsmaier et al., 1990; see Braun and Scheller, 1995; Buchner and Gundersen, 1997). The proteins were originally discovered as synapse-associated antigens in the nervous system of Drosophila and, independently, in Torpedo as candidate functional components of presynaptic voltage-sensitive N-type Ca²⁺-channels, expressed ectopically in frog oocytes (Gundersen and Umbach, 1992). CSPs are characterized by a central multiple palmitoylated string of cysteine residues and by an N-terminal J-domain, the latter being a 70 amino acid region of homology shared by bacterial DnaJ, a signature domain present in all members of the DnaJ molecular chaperone/heat shock protein (Hsp) family (Caplan et al., 1993; Braun and Scheller, 1995; see Buchner and Gundersen, 1997). CSPs exist in at least two isoforms, CSP1 and CSP2, generated by alternate RNA splicing (Zinsmaier et al., 1990; Chamberlain and Burgoyne, 1996). CSP2 represents a truncated version of CSP1 lacking the C-terminal 31 amino acid sequence. Several lines of evidence show that CSPs are associated with rat brain synaptic vesicle membranes (Mastrogiacomo and Gundersen, 1995). A general role for CSPs in trafficking events in both neuronal and non-neuronal cells has been implicated since CSP has been identified in membranes of pancreatic zymogen granules (Braun and Scheller, 1995), secretory granules of neurosecretory neurons in the neurohypophysis (Pupier et al., 1997) and in a chromaffin granule-enriched fraction (Kohan et al., 1995; Chamberlain et al., 1996). Detection of CSP mRNA in all human tissues examined suggests a widespread expression of the csp gene (Coppola and Gundersen, 1996).

Functionally, CSPs have been demonstrated to be regulatory components of synaptic exocytosis, since deletion of the csp gene in Drosophila generates a phenotype characterized by premature death and temperaturesensitive inhibition of synaptic transmission, resulting in reversible paralysis at elevated temperatures (Zinsmaier et al., 1994). However, the precise role of CSPs in synaptic vesicle exocytosis remains largely unknown. Apart from a suggested functional interaction of CSPs with presynaptic voltage-sensitive N-type Ca²⁺-channels (Gundersen and Umbach, 1992; Mastrogiacomo et al., 1994a), it has been proposed that CSPs operate as molecular chaperones, e.g. assisting in folding or the conformational change of proteins participating in membrane trafficking (Caplan et al., 1993; Zinsmaier et al., 1994; Bohen et al., 1995; see Buchner and Gundersen, 1997). The cysteine-rich string domain in CSPs has led to another hypothetical model for the functional role of CSPs. The cysteine residues of CSPs have been shown to possess a high degree of post-translational fatty acid acylation. This compact acylation conveys a hydrophobic domain to the protein, which also contains highly polar N- and C-termini, and as a consequence CSP has been proposed to be uniquely suited to catalyze membrane fusion (Gundersen et al., 1995).

Stimulus–secretion coupling in the pancreatic -cell involves complex interaction between different signal transduction pathways (see Efendic et al., 1991; Berggren and Larsson, 1994). Glucose metabolism results in an increase in the cellular ATP:ADP ratio, closure of K_ATP-channels, depolarization of the plasma membrane and opening of voltage-gated L-type Ca²⁺-channels. The subsequent Ca²⁺-influx is a major signal for initiating insulin secretion in the pancreatic -cell (Berggren and Larsson, 1994). Several proteins that regulate vesicular docking and fusion events, originally characterized in the nervous system, have recently been identified in the endocrine pancreas (Jacobsson et al., 1994), and roles for several of these proteins in the secretion of insulin have been revealed (Jacobsson et al., 1994; Martin et al., 1995, 1996; Regazzi et al., 1995, 1996; Sadoul et al., 1995; Kirali-Bourri et al., 1996; Nagamatsu et al., 1996, 1997; Wheeler et al., 1996; Lang et al., 1997; Mizuta et al., 1997). With the suggestion that CSP may be a modulator of Ca²⁺-channel activity and Ca²⁺-influx being the key event in the initiation of insulin secretion, we have attempted to identify the presence of CSP in pancreatic -cells and to determine its role in insulin secretion. We have investigated the cellular localization of CSP in the pancreas at the mRNA and protein levels by using in situ hybridization, RT–PCR, RNase protection assay, immunofluorescence histochemistry combined with confocal laser microscopy, immunoelectron microscopy, immunoblotting and subcellular fractionation. The possible functional role of CSP in the -cell has been evaluated by studies of voltage-gated Ca²⁺-channel activity and insulin secretion in individual cells using patch–clamp and amperometric techniques.

Results

CSP mRNA in insulin-producing cells

In situ hybridization of sections of cultured RINm5F cells as well as rat brain, each with two probes complementary to two different regions of CSP mRNA, showed a strong hybridization signal (data not shown). Hybridization of sections from RINm5F cells with radiolabeled CSP probe in the presence of an excess (100) of cold probe did not result in any hybridization signal (data not shown). Slides that were emulsion-dipped and hematoxylin-eosin counter-stained, showed silver grains overlying the cytoplasm of individual RINm5F cells (data not shown).

Two isoforms of CSP have so far been described, CSP1 and CSP2. The mRNA of the latter isoform has a 72 bp insertion, which contains a stop codon and leads to a shorter 3.3 kDa protein (Chamberlain and Burgoyne, 1996). In order to study which of the isoforms is expressed in the pancreatic islets and in the insulin-producing cell line INS-1, we performed RT–PCR analysis employing primers that flank the insertion region. As shown in Figure 1A, INS-1 cells express both isoforms (lane 2) of CSP, whereas pancreatic islets contain only CSP1 (lanes 3 and 4). Therefore, we selected the CSP1 isoform for overexpression studies.

CSP-immunoreactivity in rat pancreatic islets, INS-1 and mouse -cells

CSP protein was demonstrated by Western blotting in homogenates from rat pancreatic islets, ob/ob mouse -cells and INS-1 cells (Figure 1B).

Incubation of cryostat sections with an antiserum generated to CSP peptide or an antiserum generated to recombinant CSP protein gave identical labeling patterns in both pancreas and brain. In the pancreas, there was strong CSP-like immunoreactivity (CSP-LI) in many cells of the islets of Langerhans and also in nerve fibers and terminals present around blood vessels and extending into the islets (Figure 2A and B). Confocal laser microscopy revealed that CSP-LI was punctate within the cytoplasm of endocrine cells (Figure 2B). The exocrine pancreas demonstrated slightly weaker, punctate CSP-LI in the apical region of acinar cells (Figure 2C).

Incubation with non-immune serum gave no fluorescence in the rat pancreas (data not shown). Preabsorption of the CSP peptide antiserum with the corresponding immunogen (10^-6 M) gave no immunostaining in exocrine or endocrine cells.

Direct double-labeling immunofluorescence histochemistry of rat pancreatic islets demonstrated CSP-LI in insulin- (cf. Figure 3A with B), glucagon- (cf. Figure 3C with D) and somatostatin- (cf. Figure 3E with F) containing cells.

Punctate CSP-LI was also demonstrated in cultured RINm5F cells (data not shown), INS-1 cells (Figure 4A) and in cells from ob/ob mouse -cells (Figure 4B). Double-labeling of such cells showed extensive, but not complete, co-localization with insulin (Figure 4C).

Ultrastructural localization of CSP

Immunoelectron microscopy, using an antiserum to recombinant CSP, revealed gold particles localized to the membrane area of secretory vesicles in virtually all endocrine cells of the islets of Langerhans (Figure 5A and B). Insulin was detected in several secretory vesicles. Acinar cells of the exocrine pancreas showed gold particles localized to the membrane area of the zymogen granules (Figure 5C). Localization of gold particles to other organelles of the exocrine pancreas was not apparent.

CSP proteins in subcellular fractions and tissue/cell homogenates

The presence and subcellular localization of CSP protein were also examined in pancreas by subcellular fractionation and immunoblotting using antiserum to recombinant CSP. Subcellular fractions of insulinoma cells showed an enrichment of 34 and 36 kDa proteins in the granule fraction with a weak signal at 27 kDa (Figure 6). The membrane fraction showed an enrichment of a 27 kDa protein with a very weak signal at 36 and 34 kDa (Figure 6). The cytosol fraction demonstrated a 27 kDa protein (Figure 6). Homogenates of rat pancreas and liver showed a 36 kDa protein together with a very weakly stained 27 kDa protein (data not shown). Rat brain contained a strong protein band at 34 kDa and a weak band at 27 kDa (Figure 6). Higher molecular weight proteins that reacted with our antiserum to recombinant CSP were also detected. Proteins of 72/70 kDa were identified in all preparations except in brain where the molecular weight was 68 kDa (Figure 6).

Effect of CSP overexpression and CSP antibodies on voltage-dependent Ca²⁺-channels

The possible effects of CSP on voltage-dependent Ca²⁺-channel activity was tested using the whole-cell configuration of the patch–clamp technique. Overexpression of rat CSP1 in INS-1 cells led to an 400-fold increase in CSP mRNA amounts (Figure 7A) and a 2.6-fold increase in CSP protein levels (Figure 7B). Immunohistochemical analysis of CSP1 overexpressing cells compared with mock-transfected cells did not reveal any obvious differences in either CSP distribution or co-localization with insulin (Figure 7C).

To analyze the effect of CSP1 overexpression on the function of voltage-dependent Ca²⁺-channels at the single-cell level, we co-expressed CSP1 and green fluorescent protein (GFP) in INS-1 cells. Following identification of the positively transfected cells via GFP fluorescence (Figure 8A), the whole-cell configuration was established and membrane currents were recorded during voltage steps to membrane potentials between -60 and +50 mV, from a holding potential of -70 mV. Figure 8B shows the current–voltage (IV) relationship in INS-1 cells following overexpression of CSP1 (CSP1 and GFP, filled circles) or mock-transfection (GFP, open circles), using the lipofectamine transfection technique. No difference in Ca²⁺-channel activity between CSP1 and mock-transfected cells was obtained (Figure 8B). Example current traces from the descending phase of the IV-curve are shown in Figure 8B.

In a second approach, we studied the effects of a CSP antiserum on voltage-dependent Ca²⁺-channel currents. Cy3-conjugated secondary antibodies were allowed to perfuse into individual cells via the patch pipette in order to verify accumulation of antibody in the cells (Figure 8C). Mouse -cells were perfused with CSP antibodies (diluted 1:100) via the recording patch pipette. The cells were then depolarized for 100 ms, from -70 to 0 mV every 20 s. No effect of the CSP antibodies (Figure 8D, bottom traces) on depolarization-induced Ca²⁺-channel currents was obtained during a time period of 15 min, as compared with control cells perfused with mouse IgG (Figure 8D, top traces).

Effect of CSP overexpression on single-cell insulin secretion

To investigate the effects of CSP on -cell secretory capacity, we applied an amperometric technique to study release of serotonin from INS-1 cells preloaded with 5-hydroxy-DL-tryptophan 5-HTP. It is well-established that 5-HTP is converted to serotonin in the -cell and that serotonin is loaded into the secretory vesicles and co-secreted with insulin by exocytosis (Smith et al., 1995; Aspinwall et al., 1997). Again, CSP1-overexpressing cells were identified by co-expression of GFP. The secretory response to K⁺-stimulation was significantly decreased in cells overexpressing CSP1 (Figure 9A) as compared with mock-transfected cells (Figure 9B). The number of spikes decreased from 24.8 3.6 (n = 6) in mock-transfected cells to 8.0 1.1 (n = 8) in cells overexpressing CSP1 (P <0.001; Figure 9C). The area of current spikes (measured in coulombs) can be used to quantitate the moles of detected hormone or transmitter released per vesicle using Faraday's law (Wightman et al., 1991). No difference in mean spike area between CSP1 and mock-transfected cells could be demonstrated (Figure 9D), indicating that the quality of released serotonin was unaffected by CSP1 overexpression.

Discussion

The present study demonstrates that CSP mRNA is expressed in insulin-producing cells and that CSP-LI in these cells is associated with vesicular membranes. Since we could demonstrate the presence of both CSP1 and CSP2 mRNA in insulin-secreting cell lines, we suggest that the 34/36 kDa and the 27 kDa bands represent CSP1 and CSP2, respectively. Furthermore, the granule fraction contains mainly CSP1, whereas CSP2 dominates in the membrane and cytosol fractions of insulinoma cells. Predominance of CSP1 in the granule fraction and the absence of CSP2 in normal islet cells suggests that CSP1 may be involved in catalyzing membrane fusion. There is evidence that CSP is post-translationally palmitoylated in vivo, since treatment with deacylating reagents results in a 7 kDa shift, converting Torpedo CSP from 34 to 27 kDa (Gundersen et al., 1994). Data show that 12 of the 13 cysteine residues of Torpedo CSP are fatty-acylated (Gundersen et al., 1994) and that the cysteine residues are present in a restricted domain, spanning Cys113 to Cys136 (Gundersen et al., 1994). Extensive lipidation has been described for a number of other proteins, including the proteolipid protein of myelin, GAP-43, SNAP-25, rhodopsin and transferrin receptors (see Gundersen et al., 1994). It has been suggested that this unprecedented degree of acylation in CSP would be well-suited to mediate interactions at membrane interfaces such as those occurring during vesicular membrane fusion events (see Gundersen et al., 1994). It has been shown that the palmitoylated form of CSP2 appears following electrophoretic separation as a 27 kDa protein (Chamberlain and Burgoyne, 1996).

Our results show that CSP also occurs in a higher molecular weight (72 kDa) form, which has previously been suggested to represent dimerization of CSP (Braun and Scheller, 1995; Mastrogiacomo and Gundersen, 1995; Chamberlain and Burgoyne, 1996). In this context, it is of interest that two models, the 'acyl-flip' and the 'bilayer collapse' models, have been proposed to explain how CSP operates as a candidate fusion-promoting agent (Gundersen et al., 1994). Both models suggest that pairs of CSP monomers would be required for fusion to occur (Mastrogiacomo et al., 1994a,b; Gundersen et al., 1995).

As shown in the present study, CSP1 overexpression in individual INS-1 cells resulted in a significant decrease in insulin release as compared with mock-transfected cells. The fact that CSP1 overexpression inhibits insulin release may suggest that CSP indeed has a regulatory role on insulin exocytosis under physiological conditions.

CSP has earlier been implicated as a regulator of presynaptic Ca²⁺-channels (Gundersen and Umbach, 1992). Using a suppression cloning strategy, whereby N-type Ca²⁺-channels were expressed in Xenopus oocytes, a putative Ca²⁺-channel subunit cDNA was identified (Gundersen and Umbach, 1992). After the injection of CSP antisense RNA, electrophysiological studies demonstrated a complete suppression of Ca²⁺-channel activity (Gundersen and Umbach, 1992). Unexpectedly, sequence analysis revealed that this candidate Ca²⁺-channel subunit was CSP, which has been cloned in Drosophila by Zinsmaier et al. (1990). The inhibitory effect of CSP1 overexpression on insulin secretion, but the absence of effect of CSP1 overexpression or intracellular administration of CSP antibodies on voltage-gated Ca²⁺-currents is interesting and suggests that CSP is a regulator of -cell exocytosis via a mechanism not involving voltage-gated Ca²⁺-channels. Recent results obtained from Drosophila csp mutants which have a temperature-dependent block of synaptic transmission (Umbach et al., 1994) show that CSPs appear to regulate an early step in the Ca²⁺-dependent neurotransmitter secretion at the Drosophila neuromuscular junction (Umbach and Gundersen, 1997). The dominant presence of N-type Ca²⁺-channels in neurons and L-type Ca²⁺-channels in -cells (Berggren and Larsson, 1994) may explain the modulatory effect of CSP on Ca²⁺ fluxes in neurons and absence of effect on Ca²⁺-channel activity in endocrine -cells.

It is of interest that synaptotagmin, but not CSP, synaptophysin or Rab 3A, have been detected in P/Q-type Ca²⁺-channel-associated complexes (Martin-Moutout et al., 1996). Trimeric SNARE complexes are implicated in Ca²⁺-dependent exocytosis, which is believed to be partly regulated by synaptotagmin. These observations suggest that CSP is not a part of the SNARE complex.

The occurrence of a J-domain indicates that CSP belongs to the DnaJ chaperone/Hsp family. DnaJ proteins have been associated with chaperone-like functions, e.g. to aggregate and disaggregate proteins, facilitate protein transport, facilitate folding or conformational changes in protein, or to regulate protein function (Caplan et al., 1993; see Braun and Scheller, 1995). The common mechanism underlying such diverse functions appears to be the ability of Hsp70 to bind transiently and sequester unfolded regions of substrate proteins, thereby preventing unproductive aggregation (Pelham, 1986). The Hsp70 protein binding and release kinetics are governed by an intrinsic ATPase activity. Both the ATPase activity and the protein-binding activity can be modulated by protein cofactors, including DnaJ and its homologs. It is known that DnaJ interacts with Hsp70, stimulating its ATPase activity (see Caplan et al., 1993). It has been suggested that DnaJ homologues are regulators of Hsp70 activity (Caplan et al., 1993). It was shown recently that CSP activates the constitutive Hsp70 (Hsc70) in a dose-dependent manner (Braun et al., 1996; Chamberlain and Burgoyne, 1997). Hsc70 is an abundant neural protein with coupled protein binding and ATPase activity, which has been shown to uncoat clathrin-coated vesicles in vitro, an obligate step in the fusion of transport vesicles to their target membranes (Schlossman et al., 1984). Hsc70 may function within the biochemical pathways of exo- or endocytosis to promote formation or dissociation of multimeric complexes or to regulate conformational changes (see Braun et al., 1996). Taken together, our results showing reduced insulin secretion after CSP1 overexpression, but absence of regulatory action on voltage-gated Ca²⁺ currents, may be compatible with a chaperone function of CSP in pancreatic -cells.

Materials and methods

In situ hybridization

For in situ hybridization, RINm5F cells were used (Gazdar et al., 1980). RINm5F cells were centrifuged into pellets that were fixed in 4% paraformaldehyde for 1 h, washed with phosphate-buffered saline (PBS) and frozen on dry ice. The pellets were cut at 14 m section thickness in a cryostat and mounted onto ProbeOn™ microcope Slides (Fisher Scientific, Pittsburgh, PA). Rat brain and newborn mice were frozen and sectioned and used for control purposes. Two oligonucleotide probes (CAGTCCAAGAACATGGTATAACGATTCCCCGGAAGTAGAGAGTGAGCG, probe 1; and GAGTACGAAGTAGGTGTTGACATTCTCCTCCCCAAATTGCTCAGCCAC, probe 2), with an optimal G-C content (50–65%) and homology <80% with DDBJ/EMBL/GenBank-entered nucleotide sequences, were synthesized (Scandinavian Gene Synthesis, Köping, Sweden), reversed and complementary to rat CSP mRNA (Braun and Scheller, 1995; Mastrogiacomo and Gundersen, 1995). In situ hybridization was performed essentially as previously described (Young, 1990; Dagerlind et al., 1992). Probes were 3'-end labeled with [-³⁵S]dATP (NEN, Boston, MA), using terminal deoxynucleotidyl transferase (Amersham Ltd, Amersham, UK) and purified using Nensorb 20 columns (NEN). Sections were air-dried and incubated for 16 h at 42°C with 10⁶ c.p.m. of the labeled probe in a hybridization solution containing 50% deionized formamide, 4 standard saline citrate (SSC) (1 SSC = 0.15 M NaCl, 0.015 M sodium citrate), 1 Denhardt's solution [0.02% bovine serum albumin (BSA), 0.02% Ficoll, 0.02% polyvinylpyrolidone], 1% N-lauroylsarcosine, 0.02 M NaPO₄ (pH 7.0), 10% dextran sulfate, 500 g/ml denatured salmon sperm DNA and 200 mM dithiothreitol. Following hybridization, the sections were rinsed in 1 SSC at 55°C for 60 min, rinsed in distilled water, dehydrated in 60 and 95% ethanol and apposed to -max autoradiography film (Amersham) at -20°C. After two weeks of exposure (probe 1) and one week of exposure (probe 2), the films were developed with Kodak LX 24 and fixed with Kodak AL 4. In addition to film autoradiography, the sections were dipped in Kodak NTB₂ autoradiography emulsion in distilled water, exposed for 6 and 3 weeks (probe 1 and probe 2, respectively) at 4°C, developed in Kodak D19 and fixed in Kodak 3000. Sections were rinsed in distilled water and counter-stained with hematoxylin-eosin. All sections were examined and photographed under bright- or dark-field illumination, using a Nikon Microphot-SA microscope and Kodak Tmax 100 ASA film.

RT–PCR and RNase protection analysis

Detection of CSP1 and CSP2 mRNA was performed by RT–PCR using as the upstream primer 5'-TTCGTCGTCTGTGGCCTCCT-3' and as the downstream primer 5'-TGGTCTCTGTGGCGGATGCT-3'. Five micrograms of total RNA obtained from INS-1 cells, rat pancreatic islets, ob/ob mouse islets and rat cerebellum were reverse-transcribed using MMLV revertase. Aliquots of the generated cDNA were used for PCR-mediated amplification using the RT–PCR Kit (Stratagene) and [-³²P]dCTP. PCR was performed in an AutogeneII-thermocycler (Grant, UK) using a linked program (1 cycle of 5 min at 94°C, 5 min at 54°C, 2 min at 72°C; and 30 cycles of 1 min at 94°C, 2 min at 54°C, 2 min at 72°C). PCR products were separated on a 2% agarose gel in TBE.

Levels of CSP mRNA were analyzed by RNase-protection analysis. Therefore, the rat CSP cDNA fragment from nucleotides 22 to 467 (according to DDBJ/EMBL/GenBank accession number S81917) was subcloned into pBluescriptII SK(-) (Stratagene, La Jolla, CA). Radiolabeled cRNA was generated on the linearized plasmids by employing the SP6/T7 in vitro-transcription kit (Boehringer Mannheim) and [-³²P]dCTP. Following purification by polyacrylamide gel electrophoresis (6% acrylamide, 7 M urea in 1 TBE), equal c.p.m. of the labeled cRNA probes (510⁴ c.p.m. final activity) were mixed with the total RNA in hybridization solution, incubated for 5 min at 90°C and hybridized at 45°C overnight. RNase protection was performed using the RPA II kit (Ambion, Austin, TX). Quantification of protected complexes was performed by phosphoimaging. Values obtained for CSP mRNA were normalized by -actin-mRNA values.

CSP antisera

Rabbit polyclonal anti-CSP serum was generated against a CSP/ glutathione S-transferase (GST) fusion protein (anti-CSP). A cDNA encoding the entire rat CSP open reading frame was obtained by PCR from the 2 kb rat CSP clone. The sequence was verified, constructed as a GST-fusion protein and expressed in Escherichia coli. The fusion protein was purified on an agarose–glutathione column before immunization. The characterization of the CSP antipeptide antiserum has been described previously (Braun and Scheller, 1995).

Immunofluorescence histochemistry and confocal microscopy

Male Sprague-Dawley rats (B & K Universal; 150–200 g) were anesthetized with sodium pentobarbital (Mebumal®; 40 mg/kg i.p.) and perfused via the ascending aorta with Ca²⁺-free Tyrode's solution (37°C), followed by an ice-cold mixture of formalin–picric acid (4% paraformaldehyde and 0.4% picric acid in 0.16 M phosphate buffer, pH 6.9). The pancreas was rapidly removed and immediately fixed by immersion in the same fixative for 90 min, and rinsed for at least 24 h in a 0.1 M phosphate buffer (pH 7.4) containing 10% sucrose, 0.02% bacitracin and 0.01% sodium azide. Sections were cut at 10 m thickness in a cryostat and processed for indirect immunofluorescence. INS-1 cells and ob/ob mouse -cells were prepared and cultured on coverslips. The cells were fixed by immersion in fixative (as above) for 10 min and thereafter washed several times in PBS. The sections and cells were incubated in rabbit antisera to CSP peptide or recombinant CSP protein (diluted 1:500 and 1:800, respectively) for 18–22 h at 4°C. The sections were then washed in PBS for 30 min and incubated with fluoroscein isothiocyanate (FITC)-conjugated donkey anti-rabbit (diluted 1:40; Jackson Immuno Research Laboratories Inc., West Grove, PA) secondary antibody. After a further wash in PBS, the sections were mounted in 0.1% p-phenylendiamine dissolved in PBS and glycerol (1:3). Sections were double-stained by combining rabbit antiserum to CSP peptide with guinea-pig antiserum to insulin (diluted 1:2000; UCB-Bioproducts S.A., Braine-l'Alleud, Belgium) or mouse monoclonal antibodies to glucagon (diluted 1:1000; Novo Nordisk AS, Bagsvaerd, Denmark) or mouse monoclonal antibodies to somatostatin (diluted 1:200; Buchan et al., 1985). The combinations were visualized using a mixture of FITC-conjugated donkey anti-rabbit and lissamine-rhodamine (LRSC)conjugated goat anti-guinea-pig (1:40) or goat anti-mouse (1:40; both from Jackson Immuno Research Laboratories Inc.) secondary antibodies. Control sections were stained using non-immune serum (diluted 1:500) as primary antibody and by preabsorption of the antiserum to CSP-peptide with cognate CSP-peptide at 10^-6 M. Sections were also examined in a Bio-Rad MRC-600 laser scanning confocal imaging system equipped with a krypton/argon mixed gas laser and a Nikon Optiphot II microscope. The standard K1/K2 dual channel filter sets combined with an excitation filter (488 DF 10 for FITC-induced fluorescence and a 568 DF 10 for LRSC-induced fluorescence) was used to examine the immunoreactivity. The images were produced using a Tektronix Phaser IIsd printer.

Electron microscopy

Pancreatic tissue was collected from male Sprague-Dawley rats (B & K Universal, Stockholm, Sweden; 150–200 g) and 1 mm³ tissue blocks were fixed at 4°C in a mixture of 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) for 4 h. After fixation, the tissue was partially dehydrated in 70% ethanol and embedded in LR White resin (Agar Scientific Ltd, Cambridge, UK). Thin sections were collected on formvar coated nickel grids. The grids were washed in distilled water and incubated in 5% normal goat serum in PBS containing 0.1% BSA and 0.13% sodium azide (buffer A) for 30 min at room temperature. The sections were placed in primary rabbit anti-recombinant CSP (diluted 1:500) in PBS containing 1% BSA and 0.13% sodium azide for 18–22 h at 4°C. Control sections were incubated in non-immune serum. The grids were washed thoroughly in buffer A and then transferred to 10 nm gold-conjugated goat anti-rabbit secondary antibodies (diluted 1:50; Amersham) for 30 min at room temperature, followed by a wash in buffer A. The sections were subsequently post-fixed in 2% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) for 15 min and finally stained with uranyl acetate (15 min) and lead citrate (2 min) and examined with a JEOL-1200EX electron microscope. Kodak 4489 film was used for photography.

-cell granule preparation and tissue/cell homogenization

Transplantable rat insulinoma was propagated in New England Deaconess Hospital (NEDH) rats, which were obtained from an in-house breeding colony at the Joslin Diabetes Center, Harvard Medical School, Boston, MA. The subcellular fractionation of the harvested insulinoma by differential and density gradient centrifugation was performed as described previously. Insulinoma tissue was homogenized at 4°C in 0.27 M sucrose, 10 mM HEPES, 1 mM EGTA, pH 6.5 using 8–10 strokes of a Potter homogenizer. The homogenate was then centrifuged at 1700 g for 7 min at 4°C to remove cell debris and nuclei. The supernatant was applied to an iso-osmotic discontinuous density gradient composed of 19.2% (w/v) Nycodenz (Nyegaard Diagnostica AS, Oslo, Norway) mixed with 0.27 M sucrose in the proportions 1:0, 1:1 and 1:3 (5 ml of each) and then centrifuged at 4°C in a Beckman SW 28 rotor at 100 000 g for 60 min. Three fractions consisting of insulin secretory granules, plasma membrane and cytosol were obtained. The band consisting of enriched -cell granules was removed from the interface between the high and medium density solutions and then washed in the homogenization buffer. The granule fraction was mixed with eight volumes of 27% (v/v) Percoll (Pharmacia, Uppsala, Sweden) in 0.27 M sucrose/10 mM MES, pH 6.5 and centrifuged at 35 000 g at 4°C for 45 min using a Sorvall SS34 rotor. The purified -cell granule fraction was recovered and washed five or six times in 0.25 M sucrose/10 mM MES, pH 6.5, to remove the gradient material. The fraction was stored in liquid nitrogen.

Tissue homogenates were prepared from male Sprague-Dawley rats by homogenization in 50 mM Tris–HCl, 100 mM NaCl and 1 mM EGTA (pH 7.5) containing the following protease inhibitors: 0.2 mM PMSF, 2 M leupeptin, 10 M pepstatin, 50 M benzamidine, 0.5 M soya trypsin inhibitor and 0.005% DNase (Sigma Chemical Co., St Louis, MO) using a teflon/glass homogenizer. HIT-T15 cells were homogenized in 4 mM HEPES (pH 7.4) and the same protease inhibitors as above. The homogenates were centrifuged at 5000 g for 10 min at 4°C in a Beckman TLA100.2 rotor. RINm5F cell homogenates were prepared by homogenization in 1% SDS using a Branson ultrasonic homogenizer. Protein estimation was performed by protein microassay (Bio-Rad, Hercules, CA).

SDS–PAGE and Western blotting

Samples were denatured for 5 min at 100°C in SDS–PAGE sample buffer (Hames, 1990). Analysis was performed on a 12% polyacrylamide SDS–PAGE gel (Laemmli, 1970). The proteins were transferred to 0.2 m nitrocellulose (Schleicher and Schuell, GmbH, Dassel, Germany) in 20% methanol, 20 mM Tris, 150 mM glycine and 0.05% SDS for 12–14 h at 0.35 mA (Towbin et al., 1979). Non-specific binding of the primary antibody was blocked by incubation in 5% milk powder in buffer A (10 mM Tris, 150 mM NaCl, pH 7.4, containing 0.05% Tween-20). The blot was subsequently probed overnight at 4°C with polyclonal rabbit anti-recombinant CSP (diluted 1:10 000; Braun and Scheller, 1995) with buffer A containing 5% milk powder. The blots were washed with buffer A and then incubated for 1 h at room temperature with peroxidase-conjugated goat-anti-rabbit IgG (diluted 1:2000; Organon Teknika Corp., Cappel Research Products, Durham, NC) in buffer A containing 5% milk powder and finally washed with buffer A. Detection was performed using an ECL detection system (Amersham) and the immunoreactive bands visualized with Hyperfilm (Amersham).

CSP1 overexpression

The rat CSP1 cDNA was subcloned into a pRc/CMV (Invitrogen, Carlsbad, CA) backbone vector, thus generating pCMVrCSP. Plasmid pCMVGFP contains the 'humanized' form of the S65T-mutant of the GFP. All expression constructs were purified twice by CsCl-density ultracentrifugation.

INS-1 cells were grown in RPMI 1640 medium supplemented with 5.5 mM glucose, 10 mM HEPES (pH 7.4), 1 mM sodium pyruvate, 50 M -mercaptoethanol, 100 units/ml penicillin, 100 g/ml streptomycin, 2 mM glutamine and 10% fetal calf serum at 5% CO₂ and 37°C. Transfection was performed overnight by the lipofectamine technique in supplemented RPMI 1640 medium without serum and antibiotics. For co-transfection 1 g pCMVrCSP or 1 g pRc/CMV (mock) and 1 g pCMVGFP was mixed with 6 l lipofectamine (Gibco-BRL, Paisley, UK) and 1.0 ml serum-free RPMI 1640 medium supplemented as above. Cells were cultured for a further 36 h in Petri dishes for electrophysiological and amperometric studies. For analysis of CSP overexpression at RNA and protein levels, INS-1 cells were co-transfected with pHOOK1 (Invitrogen, Carlsbad, CA) and either pCMVrCSP or pRc/CMV (mock) and transfected cells were collected by the HOOK technique according to the manufacturer's instruction.

One percent SDS extracts of CSP-transfected and mock-transfected INS-1 cells (20 g) were loaded on 12.5% polyacrylamide gels, and immunoblotted using rabbit CSP antiserum (diluted 1:10 000). Blots were quantified by densitometric analysis using ImageQuant software (Molecular Dynamics Inc., Sunnyvale, CA).

Electrophysiology

The activity of the voltage-dependent Ca²⁺-channels was studied in INS-1 cells transfected with CSP as described above and in mouse pancreatic -cells. For the preparation of single -cells, mice were starved for 24 h and then killed by decapitation. Pancreatic islets from ob/ob mice were isolated by a collagenase technique (Lacy and Kostianovsky, 1967) and a -cell suspension was prepared and washed, essentially as previously described (Lernmark, 1974). The cells were resuspended in RPMI 1640 culture medium (Flow Laboratories, Scotland, UK), containing 11 mM glucose, supplemented with 10% fetal bovine serum, 100 IU/ml penicillin, 100 g/ml streptomycin and 60 g/ml gentamycin. The cell suspension was seeded into Petri dishes and incubated at 37°C in 5% CO₂ for 1–3 days.

The culture medium was removed and the cells were washed with a solution composed as follows: 138 mM NaCl; 5.6 mM KCl; 1.2 mM MgCl; 10 mM CaCl₂; 10 mM tetraethylammoniumchloride; 5 mM HEPES pH 7.4. Cells were covered with 350 l of the described solution. The pipette solution contained: 150 mM N-methyl-D-glucamine (NMDG); 110 mM HCl; 1 MgCl₂; 2 mM CaCl₂; 10 EGTA; 3 mM Mg-ATP; 5 mM HEPES pH 7.15. All experiments were performed at room temperature (22–24°C). The free concentration of Ca²⁺ was calculated to be 60 nM from the binding constants of Martell and Smith (1974). NMDG was substituted for K⁺ in the pipette solution in order to block outward-directed K⁺-currents. INS-1 cells co-expressing CSP1 and GFP were detected by GFP fluorescence using a Zeiss Axiovert 35 M fluorescence microscope with the following settings: excitation light from a 75 W xenon lamp with a 475–495 nm excitation filter, dichroic mirror and a 515–565 nm emission filter. Images for documentation were made using fluorescence digital imaging. CSP antibodies were added to the pipette solution at a dilution of 1:100. In order to prove the physiological presence of antibodies applied by the patch pipette inside the mouse -cells, we used a Cy3-conjugated rabbit IgG secondary antibody and detected it using a confocal microscope (Leica CLSM; Leica Lasertechnik GmbH, Heidelberg, Germany) with the following settings: 40/1.30 oil Leitz fluotar objective lens, excitation wavelength 568 nm (argon/krypton laser) and a long-pass 580 nm emission filter. The whole-cell configuration of the patch–clamp technique (Hamill et al., 1981) was used, utilizing an Axopatch 200 patch–clamp amplifier (Axon Instruments, Inc., Foster City, CA). Voltage-steps were generated, digitized and stored in a computer (IBM AT-clone), using the program pClamp (Axon Instruments) and Labmaster ADC (Scientific Solutions, Inc., Solon, OH). The current responses were filtered at 1 kHz, Bessel filter (-3 dB-point; Frequency Devices, Haverhill, MA).

Amperometry

Glass-encased carbon fiber microelectrodes were pulled from borosilicate glass capillaries (Hilgenberg, Malsfeld, Germany), containing a single carbon fiber of 9 m diameter (P-55S; Amoco Corp., Greenville, SC). The carbon fibers were sealed in the tip by dipping them in Sylgard (Dow Corning, Kanagawa, Japan) and then cut at a 35–45° angle. Six to 12 h prior to experiments, cells were pre-incubated in (5-HTP (Sigma Co., St Louis, MO). 5-HTP was prepared as stock solution at a concentration of 25 mM and added to the culture medium giving a final concentration of 1 mM. During experiments, Petri dishes were placed in a temperature-regulated perfusion chamber (32–36°C), perfused with a standard extracellular (EC) solution consisting of: 138 mM NaCl, 5.6 mM KCl, 1.2 mM MgCl₂, 2.6 mM CaCl₂, 5 mM HEPES–NaOH, 3 mM glucose at pH 7.40. Electrodes were placed in close contact with the cell surface with a micromanipulator (Narishige PE-2). Cells were stimulated by a transient (10–30 s) pulse of EC with the K⁺ concentration elevated to 25 mM, using a flow-injector (Transjector, Eppendorf, Hamburg, Germany). Microelectrodes were operated in amperometric mode using an EI-400 potentiostat (Ensman Instrumentation, Bloomington, IN). Records were filtered at 100 Hz (-3 dB value), digitized at 200 Hz (Axon Instrument ADC TL-1) and stored in a computer (Axotape software, Axon Instrument). A holding potential (V_C) of 600 mV versus a sodium-saturated calomel electrode was applied to the carbon fiber electrode, sufficient to oxidize 5-HTP (Zhou and Misler, 1996). Amperometric currents were analyzed using in-house software and compared using Student's t-test. For trace figures, digitized recordings were exported into CorelDraw (Corel Corp., Ottawa, Ontario, Canada) for final layout.

Acknowledgements

This research was supported by grants from the Swedish Medical Research Council (04X-10358, 03X-09890, 03X-09891, 03XS-12708, 12P-10169, 19X-00034, 03X-12549), Åke Wibergs Stiftelse, Magnus Bergvalls Stiftelse, United States Public Health Service Grant DK-35914, The Swedish Diabetes Association, The Nordic Insulin Foundation Committee, The Juvenile Diabetes Foundation, Berth von Kantzows Foundation and Funds from the Karolinska Institute.

References

Aspinwall CA, Brooks SA, Kennedy RT and Lakey JRT (1997) Effects of intravesicular H⁺ and extracellular H⁺ and Zn²⁺ on insulin secretion in pancreatic beta cells. J Biol Chem, 272, 31308–31314. | Article | PubMed | ISI | ChemPort |

Berggren P-O and Larsson O (1994) Ca²⁺ and pancreatic -cell function. Biochem Soc Trans, 22, 12–18. | PubMed | ISI | ChemPort |

Bohen SP, Kralli A and Yamamoto KR (1995) Hold 'em and fold 'em: chaperones and signal transduction. Nature, 268, 1303–1304. | ChemPort |

Braun JEA and Scheller RH (1995) Cysteine string protein, a DnaJ family member, is present on diverse secretory vesicles. Neuropharmacology, 34, 1361–1369. | Article | PubMed | ISI | ChemPort |

Braun JEA, Wilbanks SM and Scheller RH (1996) The cysteine string protein secretory vesicle protein activates Hsc70 ATPase. J Biol Chem, 271, 25989–25993. | Article | PubMed | ISI | ChemPort |

Buchan AMJ, Sikora LKJ, Levy JG, McIntosh CHS, Dyck J and Brown JC (1985) An immunocytochemical investigation with monoclonal antibodies to somatostatin. Histochemistry, 83, 175–180. | Article | PubMed | ISI | ChemPort |

Buchner E and Gundersen CB (1997) The DnaJ-like cysteine string protein and exocytotic neurotransmitter release. Trends Neurosci, 20, 223–227. | Article | PubMed | ISI | ChemPort |

Caplan AJ, Cyr DM and Douglas MG (1993) Eucaryotic homologues of Escherichia coli dnaJ: a diverse protein family that functions with Hsp70 stress proteins. Mol Biol Cell, 4, 555–563. | PubMed | ISI | ChemPort |

Chamberlain LH and Burgoyne RD (1996) Identification of a novel cysteine string protein variant and expression of cysteine string proteins in non-neuronal cells. J Biol Chem, 271, 7320–7323. | Article | PubMed | ISI | ChemPort |

Chamberlain LH and Burgoyne RD (1997) Activation of the ATPase activity of heat-shock proteins Hsc70 and Hsp70 by cysteine-string protein. Biochem J, 322, 853–858. | PubMed | ISI | ChemPort |

Chamberlain LH, Henry J and Burgoyne RD (1996) Cysteine string proteins are associated with chromaffin granules. J Biol Chem, 271, 19514–19517. | Article | PubMed | ISI | ChemPort |

Coppola T and Gundersen CB (1996) Widespread expression of human cysteine string proteins. FEBS Lett, 391, 269–272. | Article | PubMed | ISI | ChemPort |

Dagerlind Å, Friberg K, Bean AJ and Hökfelt T (1992) Sensitive mRNA detection using unfixed tissue: combined radioactive and non-radioactive in situ hybridization histochemistry. Histochemistry, 98, 39–49. | Article | PubMed | ISI | ChemPort |

Efendic S, Kindmark H and Berggren P-O (1991) Mechanisms involved in the regulation of the insulin secretory process. J Intern Med Suppl 2, 229, 9–22.

Gazdar AF, Chick WL, Oie HK, Sims HL, King DL, Weir GC and Lauris V (1980) Continuous, clonal, insulin—and somatostatin—secreting cell lines established from a transplantable rat islet cell tumour. Proc Natl Acad Sci USA, 77, 3519–3523. | Article | PubMed | ChemPort |

Gundersen CB and Umbach JA (1992) Suppression cloning of the cDNA for a candidate subunit of a presynaptic calcium channel. Neuron, 9, 527–537. | Article | PubMed | ISI | ChemPort |

Gundersen CB, Mastrogiacomo A, Faull K and Umbach JA (1994) Extensive lipidation of a Torpedo cysteine string protein. J Biol Chem, 269, 19197–19199. | PubMed | ISI | ChemPort |

Gundersen CB, Mastrogiacomo A and Umbach JA (1995) Cysteine-string proteins as templates for membrane fusion: models of synaptic vesicle exocytosis. J Theor Biol, 172, 269–277. | Article | PubMed | ISI | ChemPort |

Hames BD (1990) One-dimensional polyacrylamide gel electrophoresis. In Hames,B.D. and Rickwood,D. (eds), Gel Electrophoresis of Proteins. A Practical Approach. 2nd edn. IRL Press, Oxford, UK, pp. 1–147.

Hamill OP, Marty A, Neher E, Sakman B and Sigworth FJ (1981) Improved patch–clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Arch, 391, 85–100. | Article | PubMed | ISI | ChemPort |

Jacobsson G, Bean AJ, Scheller RH, Juntti-Berggren L, Deeney JT, Berggren P-O and Meister B (1994) Identification of synaptic proteins and their isoform mRNAs in compartments of pancreatic endocrine cells. Proc Natl Acad Sci USA, 91, 12487–12491. | Article | PubMed | ChemPort |

Kirali-Bourri C, Morgan A, Burgoyne R, Weller U, Wollheim CB and Lang J (1996) Soluble N-ethylamide sensitive factor attachment protein and N-ethylamide insensitive factors are required for Ca²⁺-stimulated exocytosis of insulin. Biochem J, 314, 199–203. | PubMed | ChemPort |

Kohan SA, Pescatori M, Brecha NC, Mastrogiacomo A, Umbach JA and Gundersen CB (1995) Cysteine string protein immunoreactivity in the nervous system and adrenal gland of rat. J Neurosci, 15, 6230–6238. | PubMed | ISI | ChemPort |

Lacy PE and Kostianovsky M (1967) Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes, 16, 35–39. | PubMed | ISI | ChemPort |

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685. | Article | PubMed | ISI | ChemPort |

Lang J, Fukuda M, Zhang H, Mikoshiba K and Wollheim CB (1997) The first C2 domain of synaptotagmin is required for exocytosis of insulin from pancreatic -cells: action of synaptotagmin at low micromolar calcium. EMBO J, 16, 5837–5846. | Article | PubMed | ISI | ChemPort |

Lernmark Å (1974) The preparation of and studies on, free cell suspensions from mouse pancreatic islets. Diabetologia, 10, 431–438.

Martell AE and Smith RM (1974) Critical stability constants. Vol I. Amino Acids. Vol II. Amines. Plenum Press, New York, NY.

Martin F, Moya F, Gutierrez LM, Reig JA and Soria B (1995) Role of syntaxin in mouse pancreatic beta cells. Diabetologia, 38, 860–863. | Article | PubMed | ISI | ChemPort |

Martin F, Salinas E, Vazquez J, Soria B and Reig JA (1996) Inhibition of insulin release by synthetic peptides shows that the H3 region at the C-terminal domain of syntaxin-1 is crucial for Ca²⁺ but not guanosine-5'-[-thio]triphosphate-induced secretion. Biochem J, 320, 201–205. | PubMed | ISI | ChemPort |

Martin-Moutout N, Charvin N, Leveque C, Sato K, Nishiki T, Kozaki S, Takahashi M and Seager M (1996) Interaction of SNARE complexes with P/Q-type calcium channels in rat cerebellar synaptosomes. J Biol Chem, 271, 6567–6570. | Article | PubMed | ChemPort |

Mastrogiacomo A and Gundersen CB (1995) The nucleotide and deduced amino acid sequence of a rat cysteine string protein. Mol Brain Res, 28, 12–18. | Article | PubMed | ISI | ChemPort |

Mastrogiacomo A, Parsons SM, Zampighi GA, Jenden DJ, Umbach JA and Gundersen CB (1994a) Cysteine string proteins: a potential link between synaptic vesicles and presynaptic calcium channels. Science, 263, 981–982. | Article | PubMed | ISI | ChemPort |

Mastrogiacomo A, Evans CJ and Gundersen CB (1994b) Antipeptide antibodies against a Torpedo cysteine-string protein. J Neurochem, 62, 873–880. | ISI | ChemPort |

Mizuta M, Kurose T, Miki T, Shoji-Kasai Y, Takahashi M, Seino S and Matsukura S (1997) Localization and functional role of synaptotagmin III in insulin secretory vesicles in pancreatic -cells. Diabetes, 46, 2002–2006. | Article | PubMed | ISI | ChemPort |

Nagamatsu S, Fujiwara T, Nakamichi Y, Watanabe T, Katahira H and Seino S (1996) Expression and functional role of syntaxin1/HPC-1 in pancreatic beta cells. Syntaxin 1A, but not 1B, plays a negative role in regulatory insulin release pathway. J Biol Chem, 2271, 1160–1165.

Nagamatsu S, Nakamichi Y, Yamaguchi K, Sawa H and Akagawa K (1997) Overexpressed syntaxin1A/HPC-1 inhibits insulin secretion via a regulated pathway, but does not influence glucose metabolsim and intracelllular Ca²⁺ in insulinoma cell line TC3. Biochem Biophys Res Comm, 231, 89–93. | Article | PubMed | ISI | ChemPort |

Pelham HR (1986) Speculations on the functions of the major heat shock and glucose-regulated proteins. Cell, 46, 959–961. | Article | PubMed | ISI | ChemPort |

Pupier S, Leveque C, Marqueze B, Kataoka M, Takahashi M and Seagar MJ (1997) Cysteine string proteins associated with secretory granules of the rat neurohypophysis. J Neurosci, 17, 2722–2727. | PubMed | ISI | ChemPort |

Regazzi CB et al. (1995) VAMP-2 and cellubrevin are expressed in pancreatic -cells and are essential for Ca²⁺ but not for GTPS-induced insulin secretion. EMBO J, 14, 2723–2730. | PubMed | ISI | ChemPort |

Regazzi CB, Sadoul K, Meda P, Kelly RB, Halban PA and Wollheim CB (1996) Mutational analysis of VAMP domains implicated in Ca²⁺-induced insulin exocytosis. EMBO J, 15, 6951–6959. | PubMed | ISI | ChemPort |

Sadoul K, Lang J, Montecucco C, Weller U, Regazzi R, Catsicas S, Wollheim CB and Halban PA (1995) SNAP-25 is expressed in islets of Langerhans and is involved in insulin release. J Cell Biol, 128, 1019–1028. | Article | PubMed | ISI | ChemPort |

Schlossman DM, Schmid SL, Braell WA and Rothman J (1984) An enzyme that removes clathrin coats: purification of an uncoating ATPase. J Cell Biol, 99, 723–733. | Article | PubMed | ISI | ChemPort |

Smith PA, Duchen MR and Ashcroft FM (1995) A fluorimetric and amperometric study of calcium and secretion in isolated mouse pancreatic beta-cells. Pflügers Arch, 430, 808–818. | Article | ChemPort |

Towbin H, Staehelin T and Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA, 76, 4350–4354. | Article | PubMed | ChemPort |

Umbach JA and Gundersen CB (1997) Evidence that cysteine string proteins regulate an early step in the Ca²⁺-dependent secretion of neurotransmitter at Drosophila neuromuscular junction. J Neurosci, 17, 7203–7209. | PubMed | ISI | ChemPort |

Umbach JA, Zinsmaier KE, Eberle KK, Buchner E, Benzer S and Gundersen CB (1994) Presynaptic dysfunction in Drosophila csp mutants. Neuron, 13, 899–907. | Article | PubMed | ISI | ChemPort |

Wheeler MB et al. (1996) Characterization of the SNARE protein expression in beta-cell lines and pancreatic islets. Endocrinology, 137, 1340–1348. | Article | PubMed | ISI | ChemPort |

Wightman RM, Jankowski JA, Kennedy RT, Kawagoe KT, Schroeder TJ, Leszczyszyn DJ, Near JA, Diliberto EJ,Jr and Viveros OH (1991) Temporally resolved catecholamine spikes correspond to single vesicle release from individual chromaffin cells. Proc Natl Acad Sci USA, 88, 10754–10758. | Article | PubMed | ChemPort |

Young WS,III (1990) In situ-hybridization histochemistry. In Björklund,A., Hökfelt,T., Wouterlood,F.G. and van den Pol,A.N. (eds), Handbook of Chemical Neuroanatomy: Analysis of Neuronal Microcircuits and Synaptic Interactions. Elsevier, Amsterdam, The Netherlands, pp. 481–512.

Zhou Z and Misler S (1996) Amperometric detection of quantal secretion from patch-clamped rat pancreatic -cells. J Biol Chem, 270, 270–277.

Zinsmaier KE, Hofbauer A, Heimbeck G, Pflugfelder GO, Buchner S and Buchner E (1990) A cysteine-string protein is expressed in retina and brain of Drosophila. J Neurogenet, 7, 15–29. | PubMed | ISI | ChemPort |

Zinsmaier KE, Eberle KK, Buchner E, Walter N and Benzer S (1994) Paralysis and early death in cysteine string protein mutants of Drosophila. Science, 263, 977–980. | Article | PubMed | ISI | ChemPort |