To delete syntaxin in (E18) chromaffin cells we expressed BoNT/C light chain from a bicistronic message containing enhanced green fluorescent protein (egfp) using the Semliki forest virus (SFV) expression system [24]. In control chromaffin cells, endogenous syntaxin 1 localized in defined clusters at the plasma membrane, in agreement with previous studies [12], [27], and many amperometric spikes were induced by high potassium stimulation (Figure 1B). Syntaxin1 staining revealed a major reduction at the plasma membrane after 6 hours of SFV BoNT/C infection (Figure 1A), and no secretion events were induced by stimulation (Figure 1B), as shown before [28]. A low cytoplasmic syntaxin staining remained after BoNT/C proteolysis (Figure 1A, see also Figure 5A) probably because the HPC1 antibody still recognizes the proteolysed protein. The cytoplasmic levels of this (soluble) protein may be low due to non-specific degradation and/or may be lost during permeabilization in preparation for staining.

We examined the effect of syntaxin deletion on secretory vesicle docking using morphometric analyses of electron micrographs. In control chromaffin cells, large numbers of secretory vesicles were morphologically docked to the plasma membrane in ultrathin sections (Figure 2A and B) [24]. After syntaxin deletion, a severe reduction of these morphologically docked secretory vesicles was observed (Figure 2A–C). This docking defect was not due to a decreased biogenesis of secretory vesicles, because the total number of vesicles (Figure 2E) was unaffected. The average vesicle diameter was also normal (control: 85.4±11.0 nm, control+BoNT/C: 87.3±9.6 nm; ANOVA p>0.01 for respectively n = 561 and n = 548 vesicles in N = 4 animals and n = 20 cells). A comparison between vesicle distribution in BoNT/C expressing chromaffin cells with munc18-1 null chromaffin cells [24] (grey line in Figure 2B) indicates that syntaxin deletion produces an exact phenocopy of munc18-1 deletion (Figure 2B). A small but significant difference between syntaxin and munc18-1 deletion was observed in the amount of secretory vesicles within 0–100 nm from the plasma membrane (inset of Figure 2B; ANOVA p<0.05). The subset of secretory vesicles in this 100 nm region of the plasma membrane is thought to represent the morphological correlates of unprimed vesicles [14]. Syntaxin deletion resulted also in a significant decrease of secretory vesicles<30 nm distance from the plasma membrane (Figure 2D; ANOVA, p<0.001), that might represent a separate vesicle pool [14]. These data indicate that syntaxin is involved in secretory vesicle docking in chromaffin cells.

Earlier studies demonstrated that in synapses syntaxin cleavage does not affect synaptic vesicle docking [4], [29]. To elucidate the apparent contrast with chromaffin cells, we expressed BoNT/C in cultured autaptic hippocampus neurons at DIV16. BoNT/C infected and control neurons maintained a dense network of synapses on the timescale of this experiment No other morphological changes were observed in the BoNT/C infected cultures (neurite length, branching, not shown). To confirm that synaptic transmission was silenced in BoNT/C expressing neurons, we examined synaptic function by measuring spontaneous miniature (‘minis’) and evoked excitatory currents. The spontaneous release frequency was strongly reduced but not completely blocked in autaptic hippocampus neurons infected with BoNT/C (Figure 3A and B; control n = 5, control+BoNT/C n = 4) as observed before in rat neurons [25], [30]. Consistent with previous studies [4], [29], action potential triggered release is completely abolished in the BoNT/C infected cells (Figure 3C; excitatory postsynaptic current (EPSC) amplitude in control cells was 2294±514 pA, n = 5, in BoNT/C transfected cells EPSC amplitude could not be detected, n = 4).

Next, we examined the synaptic ultrastructure in BoNT/C expressing autaptic neurons. The overall synapse morphology was not altered (Figure 4A). At low magnification the general appearance of synapses infected with SFV BoNT/C was similar to controls. The ultra structure of asymmetrical synapses was also unchanged, showing docked vesicles in immediate contact with the active zone membrane, many synaptic vesicles in the periphery of the active zone, and a post-synaptic density (Figure 4A). Instead of a docking defect, we observed an increase of docked vesicles at the active zone (Figure 4B), as previously observed in the Drosophila neuromuscular junction [5] and giant synapses of squid [29]. The length of the active zone was also increased, and the number of docked vesicles per active zone length was unaltered (Figure 4B). Thus, in contrast to chromaffin cells, synapses do not show a synaptic vesicle docking defect after syntaxin proteolysis. We also quantified the number of vesicles within 150 nm (approximately 3 times the synaptic vesicle size) from the active zone, and the total number of vesicles present in the synapse, and observed for both vesicle populations a significant increase (Figure 4B), consistent with previous observations [3], [29]. No significant difference in the cumulative vesicle distribution was observed (Figure 4C).The synaptic vesicle cluster perimeter did not significantly differ between control and BoNT/C expressing synapses (Figure 4B), while the mean synaptic vesicle diameter was also unchanged (control: 30.7±0.7 nm, control+BoNT/C: 33.9±0.9 nm; respectively n = 1222 and n = 1099 vesicles quantified in n = 62 and n = 46 synapses from N = 4 animals). Within the synaptic vesicle pool we observed an increased number of larger (>45 nm) synaptic vesicles (control: 5.5% and control+BoNT/C: 14.7%; ANOVA p<0.001 for n = 1099 vesicles in n = 62 and n = 46 synapses from N = 4 animals) in syntaxin-deleted synapses (Figure 4D). The larger synaptic vesicles were observed throughout the synapse also in direct contact with the active zone membrane (Figure 4A). A similar phenotype was detected in Drosophila strains lacking syntaxin [5].

Expression of BoNT/C shows the same docking defect as we previously found with Munc18-1 deletion [20], [24]. An impaired targeting or local accumulation of Munc18-1 at docking sites may therefore explain the docking defect in syntaxin-deleted chromaffin cells. Here, we performed a global analysis of the cellular distribution of Munc18-1 within the entire cell. We assumed that the cellular distribution of Munc18-1 along the entire cell diameter is similar to specialized cell areas like cell-cell contacts or areas of substrate contact (‘footprints’), and therefore Munc18-1 localization in these specialized cell areas was not analyzed. Examination of the subcellular localization of Munc18-1 using immunofluorescence staining revealed a punctate labeling in the cytoplasm and discrete puncta at the plasma membrane. This distribution was similar in the presence or absence of syntaxin (Figure 5A). Quantification of Munc18-1 staining revealed indeed a significant, but small reduction of the total average intensity (14.8%, ANOVA p<0.05 for n = 22 cells from N = 2 animals) and plasma membrane staining in BoNT/C expressing cells (14.9%, ANOVA p<0.01; Figure 5B and C). These small reductions in Munc18-1 levels appear to be insufficient to explain the drastic docking defect.

Changes in the sub-membrane actin cortex of secretory cells influence the number of docked secretory vesicles [31]. Previously we demonstrated that Munc18-1 thinners and fenestrate the F-actin cortex and hereby regulates the vesicle ‘hit-rate’ at the target [24]. Since syntaxin is a major binding partner of Munc18-1, we evaluated whether syntaxin proteolysis has similar effects. We visualized the actin cytoskeleton using rhodamin-phaloidin. Munc18-1 under- or overexpression increased or reduced respectively the amount of actin in the F-actin cortex as published before [24], but the intensity/intactness of the sub-membranous actin was similar in the presence or absence of syntaxin (Figure 6). We conclude that syntaxin, unlike its binding partner Munc18-1, does not influence docking by changing the organization of the actin cortex.

Embryonic (E18) mouse chromaffin cells were prepared as described [8] and experiments performed on the 2^nd–4^th day after isolation. Microisland hippocampus cultures were prepared from wild-type mouse embryos at E18 according to [48] and experiments performed at DIV16. Acute expression of heterologous genes was induced using Semliki Forest Virus (SFV). Genes of interest were expressed from a bi-cistronic message containing egfp [20], [24]. Experiments were performed after 6 hr of infection. BoNT/C light chain (kind gift from T Galli, INSERM, Paris, France) and Munc18-1 have been described before [24]. Constructs were verified by sequencing.

Chromaffin cells from wild type or munc18-1 deficient (E18) mice were plated on rat tail type 1 collagen-coated (32 µg/ml; Beckton Dickinson labware, USA) Bellco gridded glass coverslips (Bellco Glass Inc., USA) and infected (DIV2) with BoNT/C-ires-egfp, munc18-1-ires-egfp, or SFV-egfp as a control. Cells were observed under a fluorescence microscope 6h after infection and the location of infected/control cells were mapped. At the time secretion was blocked in BoNT/C expressing cells, cells were fixed for 45 min at room temperature with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4). After fixation cells were washed three times for 5 min with 0.1 M cacodylate buffer (pH 7.4), post-fixed for 2 hr at room temperature with 1% OsO₄ in bidest, washed and stained with 1% uranyl acetate for 40 min in the dark. Following dehydration through a series of increasing ethanol concentrations, cells were embedded in Epon and polymerized for 24 h at 60°C. The coverslip was removed by alternately dipping in liquid nitrogen and hot water. Cells of interest were selected by observing the flat Epon embedded cell monolayer (containing the gridded Bellco print) under the light microscope, and mounted on pre-polymerized Epon blocks for thin sectioning. Ultra thin sections (∼90 nm) were cut parallel to the cell monolayer and collected on single-slot, formvar-coated copper grids, and stained with uranyl acetate and lead citrate. For each condition the relative frequency of docked vesicles, and vesicles within 30 or 100 nm from the plasma membrane were calculated in three different grids per animal in a JEOL 1010 electron microscope. Docked vesicles were without any measurable distance between granule and plasma membrane. Distances from the granule membrane to the plasma membrane were measured on digital images taken at 20.000× magnification using analySIS software (Soft Imaging System, Gmbh, Germany). Secretory vesicles were recognized by their round, dense core and had a diameter of approximately 90 nm. The observer was blinded for the genotype.

Hippocampus islands cultures of wild-type mice (E18) were grown on Bellco gridded coverslips that contain micro islands of glia cells. Wild-type hippocampus neurons were infected (DIV16) with BoNT/C-ires-egfp or SFV-egfp as control and observed under a fluorescence microscope 6h after infection to map the location of infected cells. Only glia islands containing a single neuron were used for analysis. Fixation was performed at the time when evoked-release was blocked in BoNT/C expressing neurons. Fixation, embedding and sectioning were the same as for chromaffin cells (see above). Autaptic synapses were selected at low magnification using a JEOL 1010 electron microscope. The distribution of synaptic vesicles, total vesicle number, size of the vesicle cluster, post synaptic density and active zone length were measured on digital images taken at 100.000× magnification using analySIS software (Soft Imaging System, Gmbh, Germany). The observer was blinded for the genotype. No difference was observed in any of the parameters measured between wild-type synapses expressing SFV-egfp and non-infected wild-type synapses, these synapses were therefore pooled.

Mouse chromaffin cells were infected (DIV2) with SFV-egfp, BoNT/C-ires-egfp or munc18-1-ires-egfp and fixed after 6 hr in 4% paraformaldehyde, permeabilized in 0.1% Triton X-100 and blocked with 2% goat serum. Cells were incubated with primary antibody (anti-Munc18-1, polyclonal #2701 produced in our laboratory; anti-syntaxin1, monoclonal #HPC1, Sigma), washed and stained with the secondary antibody (goat anti-rabbit Alexa 594 or goat anti-mouse Alexa 647 or 594). Filamentous actin was stained by incubation with 0.25 U/ml rhodamin-phalloidin (Molecular Probes) in PBS for 40 min. As a control filamentous actin was stained in munc18-1 null chromaffin cells infected with SFV-egfp. In each experiment, coverslips were viewed with a 63× objective Zeiss LSM510 fluorescence microscope and confocal images were acquired using identical photomultiplier settings and corrected for background fluorescence using unlabelled specimens. The Munc18-1 expression level on the plasma membrane was determined along line scans of the entire cell diameter using the Metamorph software (Universal Imaging Corporation, West Chester, USA) and normalized to the total intensity.

Carbon-fiber amperometry on mouse chromaffin cells were performed 6h after infection at 30–32°C. Single-stranded insulated carbon fibers (diameter 6 µm, model CC-18, van den Hul, Oene, The Netherlands) were mounted in glass micro capillaries (GC150-10, Harvard Apparatus Ltd, Kent, UK). GigaOhm resistance (2–5 GW) to ground was achieved by insulating the microelectrode and carbon fiber with Sylgard. The tip of the carbon fiber was cut just before the experiment to ensure cleanliness and sensitivity of the exposed tip surface. Microelectrodes were filled with 1 M KCl and placed in close apposition to the cell surface. Amperometric currents were recorded with an EPC8 amplifier (HEKA Electronics, Lambrecht, Germany; electrode voltage set to +650 mV), sampled at 10 kHz and filtered at 3 kHz. Release was evoked using high 30 nM K⁺ solution.

Neurons were infected 6 hours before electrophysiological recordings with Semliki Forest virus. Whole cell voltage-clamp recordings were performed on cultured hippocampus autaptic neurons between DIV 15 and 16. The patch pipette contained the following solution (in mM): 125 K⁺-gluconic acid, 10 NaCl, 4.6 MgCl₂, 4 K₂-ATP, 15 creatine phosphate, 1 EGTA and 20 U/ml phospocreatine kinase (pH 7.30). External medium contained (in mM): 140 NaCl, 2.4 KCl, 4 CaCl₂, 4 MgCl₂, 10 HEPES, 10 glucose (pH 7.30). Axopatch 200A (Axon Instruments, Union City, USA) was used for whole-cell recordings and signals were acquired using Digidata 1322A and Clampex 8.1 (Axon Instruments). Clampfit 8.0 (Axon Instruments) and Mini Analysis (Synaptosoft) was used for offline analysis. All experiments were conducted at 31°C.

Data are shown as mean values±SEM Statistical significance was determined by comparison between experiments (means of all chromaffin cells or synapses within an experiment) using paired student's t-test or analysis of variance.