![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 19, Issue 9, 3769-3781, September 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,
*Molecular Mechanism of Exocytosis, Max Planck Institute for Biophysical Chemistry, 37077 Göttingen, Germany; and ||Department of Neurobiology, Max Planck Institute for Biophysical Chemistry, 37077 Göttingen, Germany
Submitted December 6, 2007;
Revised May 23, 2008;
Accepted June 18, 2008
Monitoring Editor: Thomas F. J. Martin
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Jahn and Scheller, 2006; Rizo et al., 2006) and characterized by a stretch of 60–70 amino acids, known as the SNARE motif (Terrian and White, 1997; Weimbs et al., 1997). The assembly of four SNARE domains into a coiled-coil bundle is a general mechanism in fusion of intracellular membrane compartments. The bundle is held together by layers of hydrophobic interaction, with the exception of the middle layer (layer "0"), which is formed by an arginine and three glutamines (Sutton et al., 1998). The SNARE domains can be subdivided into four homologous groups, named after their contribution to the zero layer: R, Qa, Qb, and Qc (Fasshauer et al., 1998b). SNARE assembly requires the participation of one domain from each group, resulting in a certain degree of specificity (McNew et al., 2000; Scales et al., 2000a).
In most cases, the four SNARE-domains are encoded by separate membrane-targeted proteins, but the SNAREs driving the fusion of vesicles with the plasma membrane (exocytosis) are special in that three proteins provide the four domains (Weimbs et al., 1998; Fukuda et al., 2000). One of the SNAREs in this pathway, exemplified by the best-known isoform synaptosomal-associated protein of 25 kDa (SNAP-25), seems to have been created by fusion of the Qb and Qc SNAREs. This arrangement necessitates a flexible linker, which runs back along the complex from the C-terminal end of the first (Qb) SNARE-domain and connects to the N-terminal end of the second SNARE-domain (Qc). This antiparallel linker is a special feature of exocytotic SNARE complexes in eukaryotic organisms from yeast to human.
The only function so far ascribed to the linker domain is the membrane-targeting of SNAP-25 and SNAP-23 through the palmitoylation of four to five cysteine residues in the N-terminal end of the linker (Gonzalo et al., 1999; Loranger and Linder, 2002). However, other members of the family (SNAP-29, SNAP-46, Sec9, and SPO20) lack linker cysteines. It remains controversial whether palmitoylation is needed for the function of SNAP-25 in exocytosis, or only for membrane targeting, and it is unknown whether the linker plays any other, more active role in determining the special features that distinguish exocytosis from other membrane fusion reactions: calcium-triggering and a high rate of execution of the fusion process.
Ca2+-triggered exocytosis from Snap-25 null mouse chromaffin cells is nearly abolished, but it can be rescued by viral expression of SNAP-25 isoforms (Sorensen et al., 2003b). Here, we used this approach and fast electrophysiological techniques to address the role of the SNAP-25 linker domain in exocytosis. By mutating linker-cysteines in SNAP-25 and substituting the SNAP-23 linker for its SNAP-25 counterpart, we verify the function of cysteines in membrane targeting, and we show that the cysteine-containing part of the SNAP-25 and SNAP-23 linker is interchangeable. However, a short 10-amino acid stretch at the C-terminal end of the SNAP-25 linker is necessary for fast calcium triggering of exocytosis. These data establish the SNAP-25 linker as an integral part of the membrane fusion machinery, and they suggest that the arrangement of the Qa and Qb motifs in one protein together with a linker is an adaptation toward fast exocytosis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). SNAP-25a cysteine mutants and SNAP-25a/SNAP-23 chimeras were produced by polymerase chain reaction (PCR), and all constructs were verified by DNA sequencing. Semliki Forest virus particles expressing mutant/chimera together with enhanced green fluorescent protein from a bicistronic message were generated as described (Ashery et al., 1999). For bacterial expression of the SNAP-25a and SNAP-23 linker domain constructs, corresponding DNAs were cloned into a pET28a vector via the NdeI/EcoRI cleavage sites.
Immunofluorescence on Plasma Membrane Sheets
Plasma membrane sheets were generated 6 h after viral infection of mouse embryonic chromaffin cells and subsequently fixed, washed, and blocked as described previously (Nagy et al., 2005). The membrane sheets were incubated with the primary antibodies (mouse anti-SNAP-25, dilution 1:100; rabbit anti-syntaxin 1, dilution 1:100; Synaptic Systems, Göttingen, Germany) for 3 h and subsequently with cyanine (Cy)3- and Cy5-coupled secondary antibodies for 60 min (dilutions 1:200; Jackson ImmunoResearch Laboratories, West Grove, PA). All antibodies were diluted in phosphate-buffered saline (PBS) containing 1% bovine serum albumin. 1-(4-Trimethyl-aminiumphenyl)-6-phenyl-1,3,5-hexatriene (Invitrogen, Carlsbad, CA) was used for visualizing the plasma membrane. Samples were examined with an Axiovert 100TV fluorescence microscope (Carl Zeiss, Oberkochen, Germany) with a 100 x 1.4 numerical aperture plan achromate objective by using appropriate fluorescence filters (Carl Zeiss). Images were taken with a back-illuminated frame transfer charge-coupled device camera (512 x 512-NTE Chip, 24- x 24-µm pixel size; Scientific Instruments, Monmouth Junction, NJ) with a magnifying lens (2.5x Optovar). Digital image analysis was performed using MetaMorph software (Molecular Devices, Sunnyvale, CA). To quantify fluorescence intensity, a region-of-interest was defined on the randomly selected membrane and transferred to the other channels. The fluorescence intensity was calculated by measuring the average intensity of the area and subtracting the local background. At least 10 membrane sheets from each animal were analyzed, and the mean value for each animal was used to calculate population mean and SEM (5–13 animals per condition). Correlative features of fluorescent spots were analyzed as described in Nagy et al. (2005).
Immunoblotting
Bovine chromaffin cell preparation was performed essentially as described previously (Nagy et al., 2002). The primary antibodies used were rabbit anti-SNAP-25 (1:3000; catalog no. 111 002, Synaptic Systems, Göttingen, Germany) and anti-valosin–containing protein (VCP, 1:3000; catalog no, ab11433, Abcam, Cambridge, United Kingdom), which was used as a loading control. Equal amounts of proteins were separated on a 4–20% SDS-polyacrylamide gel (Ready Gel; Bio-Rad Laboratories, Hercules, CA), and they were blotted onto nitrocellulose membranes (Amersham Hybond-ECL; GE Healthcare Bio-Sciences, Uppsala, Sweden). After incubation with secondary antibodies (goat anti-rabbit/anti-mouse horseradish peroxidase-conjugated immunoglobulin G, 1:2000; Jackson ImmunoResearch Laboratories), the membranes were washed three times and incubated in ECL Western blotting detection reagent (SuperSignal, West Pico; Pierce Chemical, Rockford, IL). Chemiluminescence was detected by a digital gel documentation system; quantification was performed by densitometry using ImageJ software (National Institutes of Health, Bethesda, Maryland). The expression level was corrected for the infection efficiency, which was estimated as the ratio of green fluorescent protein (GFP)-expressing cells to the total cell number.
Ca2+ Uncaging and Measurements, Electrophysiology, and Electrochemistry
Mouse chromaffin cells were infected and 5–8 h were allowed for expression of wild-type and mutant constructs in the same preparations. Whole-cell patch clamp, ratiometric intracellular calcium ([Ca2+]i) measurements, flash photolysis of caged Ca2+, and amperometry and membrane capacitance measurements were performed as described previously (Nagy et al., 2002). Data were analyzed using Igor Prosoftware (Wavemetrics, Lake Oswego, OR). Pool sizes and fusion time constants were obtained by fitting a sum of exponential functions to individual capacitance traces; data are presented as mean ± SEM. The significance was tested using nonparametric Mann–Whitney test.
Protein Expression and Purification
The expression constructs for cysteine-free SNAP-25A (C84S, C85S, C90S, C92S; amino acids [aa] 1–206), for the syntaxin 1a SNARE motif (SyxH3; aa 180–262), and for synaptobrevin 2 (Syb; aa 1–96) have been described previously (Fasshauer et al., 1998a).
For expression of SNAP-25a linker peptide (NKLKSSDAYKKAWGNNQDGVVASQPARVVDEREQMAISGGFIRRVTNDARE) and SNAP-23 linker peptide (NRTKNFESGKNYKATWGDGGDNSPSNVVSKQPSRITNGQPQQTTGAASGGYIKRITNDARE) the fragments were introduced into the pET28a plasmid between the NheI and the EcoRI sites. Protein expression was induced in Escherichia coli BL21(DE3) by adding 0.8 mM isopropyl-β-D-thiogalactopyranoside to a bacterial culture grown in Luria-Bertani medium to OD600 
0.8. The bacteria were pelleted and resuspended in extraction buffer (500 mM NaCl, 50 mM Tris, pH 7.4, and 8 mM imidazole). Cells were sonified in the presence of 6 M urea and subsequently incubated with nickel-nitrilotriacetic acid. After 1 h at 8°C, the Ni2+ beads were washed with extraction buffer, and the recombinant protein was eluted with 400 mM imidazole (plus half of the extraction buffer). The His-tag was cleaved off by thrombin during overnight dialysis in 20 mM Tris, 50 mM NaCl, and 1 mM dithiothreitol, pH 7.4. After purification by ion exchange chromatography on an Ákta system (GE Healthcare Bio-Sciences) using Mono-S column (GE Healthcare Bio-Sciences), all peptides were pure as judged by SDS- polyacrylamide gel electrophoresis (PAGE) analysis. Protein concentration was determined by absorption at 280 nm, and peptides were stored at –20°C.
SDS-PAGE was carried out as described previously (Schagger et al., 1988). For testing SNARE-complex assembly, equimolar amounts of the SNARE proteins were incubated for one hour at room temperature. When testing for SDS resistance, samples were solubilized in SDS sample buffer (not boiled) before analysis on a 12% polyacrylamide gel.
Circular Dichroism (CD) Spectroscopy Measurements
CD measurements were performed using a Jasco model J-720 instrument (Applied Photophysics, Leatherhead, United Kingdom) by using quartz cuvettes with 1-mm pathlength (Helma, Mülheim, Germany). All experiments were carried out in 20 mM sodium phosphate buffer, pH 7.4, in the presence of 100 mM NaCl at 25°C. Liposomes (65% brain phosphatidylcholine, 30% brain phosphatidylserine, 5% phosphatidylinositol-4,5-bisphosphate were mixed in chloroform and dried; Avanti Polar Lipids, Alabaster, AL) were prepared in 20 mM Tris buffer with 100 mM NaCl, pH 7.4, by the SMART system (GE Healthcare Bio-Sciences).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). In the quadruple C/S mutation, both the overall expression level was reduced (Figure 1C; 0.6- to 1-fold expression, 3 experiments) and the amount of mutated protein on the plasma membrane, which reached only 4% of the level measured after WT overexpression (Figure 1B). Costaining of membrane sheets against syntaxin-1 showed that the level of syntaxin-1 staining was unchanged between sheets overexpressing WT SNAP-25a (308 ± 18 arbitrary units [a.u.], 4 experiments with 31–58 sheets each) and sheets overexpressing the quadrupel C/S mutant (303 ± 20 a.u., 4 experiments with 43–52 sheets each).
|
). This analysis showed that the fast secretory burst (representing the fusion of the RRP) and the slow burst (representing the SRP) as well as the time constants of fusion from the two pools were similar to the WT situation (Figure 2, A and B). The observed increase in RRP size in the presence of C/S-1 or C/S-4 was not statistically significant. Also the sustained rate of release was unperturbed. Therefore, the ablation of single cysteines in the SNAP-25a linker region does not impair the ability of SNAP-25a to mediate fast Ca2+-triggered exocytosis.
|
50%), in which no secretory defect was identified (Sorensen et al., 2003b) (Supplemental Figure 1). This presented an opportunity for testing whether the palmitoylation plays a role for exocytosis per se, in addition to membrane targeting. Even this mutant resulted in significant rescue of exocytosis (Figure 2, C and D). However, compared with WT protein the overall level of exocytosis was somewhat reduced, due to a reduced size of both releasable pools. In addition, the RRP fusion time constant and the time delay between the UV-flash and the start of the capacitance increase were both significantly increased, indicating that fusion triggering was negatively affected.
We next infused expressing chromaffin cells with a solution containing 13.5 µM calcium and recorded single resolved amperometric spikes (Table 1). Because each vesicle fusion event gives rise to one amperometric spike, it is possible to estimate parameters of individual fusion events. Especially, it is possible to distinguish between the formation of a narrow fusion pore, which gives rise to a prespike "foot signal" in the amperometric recording, and the full fusion event, which results in the amperometric spike (Jackson and Chapman, 2006). Interestingly, in 4xC/S-expressing chromaffin cells the duration of the pre-spike foot signal was prolonged (Table 1). Notably, spikes in the Snap-25 knockout cells were found to have on average shorter prespike feet (Sorensen et al., 2003b). This finding demonstrates that the presence of the 4xC/S mutation does not result in an intermediate phenotype between the Snap-25 knockout and the wild-type situation caused by the lower expression level, but that the 4xC/S mutant causes slower secretion even on the level of individual fusion events. In addition, the 4xC/S mutation resulted in a mild and less significant increase in the charge of each spike, i.e., the quantal size.
|
The SNAP-25 Linker Domain Speeds Up Exocytosis Triggering
We reported previously that the expression of the ubiquitous isoform SNAP-23 in Snap-25 null chromaffin cells supports some rescue of calcium-dependent exocytosis, but without a burst component (Sorensen et al., 2003b). Here, we first repeated this experiment with similar results, and then applied stronger stimulation into the >100 µM range to investigate whether exocytosis in the presence of SNAP-23 might be shifted to higher [Ca2+]i and therefore have gone undetected in previous studies. The high-Ca2+ stimulation elicited much larger capacitance increases that were not correlated with an amperometric signal (Supplemental Figure 2, A and B). This observation has been attributed to the SNARE-independent fusion of a population of vesicles not containing catecholamines at [Ca2+] >100 µM (Xu et al., 1998). Therefore, only the amperometric measurements were used for assaying catecholamine release under these conditions. The high-calcium stimulations only elicited slightly more release of catecholamines than stimulation to 20–30 µM, and the secretion followed a similar time course as during flashes to 20–30 µM calcium (Supplemental Figure 2, A and B). Thus, Snap-25 null cells do not support fast LDCV release even when expressing SNAP-23 and stimulated to [Ca2+]i > 100 µM.
Next, we asked whether there are any functional differences between the SNAP-23 and SNAP-25 linkers, which both act to anchor the protein in the plasma membrane. We therefore constructed chimeric proteins where the two SNARE domains from SNAP-25 were joined by the SNAP-23 linker. The crossover points of the chimera were the lysine-83(SNAP-25)/lysine-78(SNAP-23), and the glutamate-148(SNAP-25)/glutamate-153(SNAP-23) (also see Figures 4 and 5). The chimera (denoted SN25aL23) was overexpressed at similar levels as SNAP-25 WT protein (Figure 1C). When overexpressed in Snap-25 null chromaffin cells, SN25aL23 did not fully restore secretion (Figure 3), but it led to a slowdown and a reduction in exocytosis within the first second of stimulation (Figure 3A). This partial rescue already indicates that the linker plays another role than just membrane targeting (see also below). We investigated whether stronger stimulation might recover exocytosis driven by the chimera. We increased [Ca2+]i to 60–80 µM to speed up exocytosis without eliciting capacitance increases uncorrelated to amperometric charge, which typically happens at >100 µM in an all-or-none manner (Xu et al., 1998). Indeed, we found that at 60–80 µM calcium, the burst of secretion within the first 0.5 s was largely restored, whether evaluated by capacitance, or by amperometric measurements (Figure 3B). The agreement between the two types of measurements shows that these recordings were not severely contaminated with fusing vesicles not containing catecholamines.
|
20 (Figure 3E, bottom left), whereas the secretory delay was threefold longer in the SN25aL23-expressing cells (Figure 3E, bottom right). At higher postflash [Ca2+]i (Figure 3, B, D, and F), both WT and mutant cells displayed faster kinetics, as indicated by faster time constants and shorter secretory delays (Figure 3, D and F). In SN25aL23-expressing cells, two exocytotic phases were now distinguishable (Figure 3D). Strikingly, when considering cells with a calcium-increase to between 35 and 100 µM, the amplitudes of the two burst components were normal (Figure 3F, top left and middle), indicating that the releasable vesicle pool sizes were unaffected by the SNAP-23 linker. However, both the fast and slow time constant of release and the secretory delay were longer in the chimera, indicating slower exocytosis triggering with the SNAP-23 linker (Figure 3, D and F). To investigate the calcium-dependence of release, we plotted the rate constant of fast release (= 1/
) against the postflash [Ca2+]i (Figure 3G). Likewise, we plotted the secretory delay against post-flash [Ca2+]i (Figure 3H). In Figure 3, G and H, we included data up to 250 µM calcium, even though at >100 µM the population of noncatecholamine containing vesicles referred to earlier became dominating. However, because these vesicles fuse with a slow time constant of 500 ms (refer to Supplemental Figure 2), it was still possible to identify the kinetics of fast burst release from these measurements, even though properties of the slow burst could not be identified. The graphical representation in Figure 3, G and H, shows that with increasing [Ca2+]i the rate constant of fast release was speeded up, whereas the delays became shorter, as described previously (Voets, 2000). This calcium-dependence of secretion was shifted toward much higher concentrations in the chimeric construct containing the SNAP-23 linker. In addition to the effect on exocytosis triggering, at high [Ca2+]i it became clear that the sustained rate of secretion was reduced in the chimeric construct (Figure 3, B and F, top right). This parameter has often been associated with the rate of refilling the releasable vesicle pools.
Recording of single resolved amperometric spikes did not reveal significant differences between SNAP-25a and SNAP25aL23 (Table 1), in line with the lack of significant differences following SNAP-25a and SNAP-23 overexpression (Sorensen et al., 2003b). This shows that a slow-down of exocytosis triggering (found both in 4xC/S mutant and in the SNAP25aL23 chimera) in some cases might correlate with a change in fusion pore duration; in other cases not. Thus, exocytosis triggering and fusion pore expansion is driven by partly different—but probably overlapping—processes.
Overall, the replacement of the SNAP-23 for the SNAP-25 linker causes a displacement of the calcium dependence of exocytosis toward much higher concentrations without changing the releasable vesicle pool sizes, and it also seems to affect slower phases of exocytosis that assay vesicle priming (see Discussion).
The Membrane-anchoring Parts of SNAP-25 and SNAP-23 Are Interchangeable
Even though the role of both the SNAP-25 and the SNAP-23 linker in anchoring the protein to the plasma membrane is well-established, it could not be ruled out that the chimeric SN25aL23 protein might somehow be defective in membrane-anchoring due to a mismatch between the linker and other parts of the protein. We therefore again compared the membrane anchoring, by using an antibody recognizing both SNAP-25 WT and SN25aL23. Both proteins were found in the plasma membrane after overexpression in mouse cells at 12- to 13-fold the endogeneous concentration (Figure 4, A and B). Costaining of membrane sheets against syntaxin-1 showed that the level of syntaxin-1 staining was unchanged between sheets overexpressing WT SNAP-25a (938 ± 13 a.u., 3 experiments with 34–44 sheets each) and sheets overexpressing the SN25aL23 (941 ± 20 a.u., 3 experiments with 34–48 sheets each).
|
), we also investigated the colocalization between our chimeric construct and syntaxin by using double labeling and cross-correlation analysis (Nagy et al., 2005). This analysis showed no major difference in the correlation coefficient, with syntaxin between SNAP-25a and SN25aL23 (the point at 0 radial distance in Figure 4C). The r decays to zero when one of the two pictures are displaced with respect to the other (radial distance >0), indicating the finite dimension of correlating domains. This analysis shows that the punctate staining pattern for syntaxin-1 and SNAP-25a overlap to the same degree as syntaxin-1 and SN25aL23.
Next, we investigated whether the SN25aL23 chimera can form SNARE-complexes, by mixing equimolar amounts of bacterially expressed and purified SNARE-proteins for 1 h at room temperature, following by SDS-PAGE (Figure 4D). Both full-length SNAP-25a, and the SN25aL23 and SN25aL23
3C-2 chimeras (see below), were able to form ternary SNARE-complexes with syntaxin-1 and synaptobrevin-2 (bands at arrow 50 kDa in lanes 6–8, Figure 4D).
To identify the part of the linker responsible for the difference in exocytosis triggering, we constructed a number of chimeras (Figure 5) and tested them after overexpression in Snap-25 null cells. The expression level of these chimeras was unchanged as compared with SNAP-25 WT protein (Figure 1C). We first investigated whether the fact that the SNAP-23 linker is 10 amino acid residues longer than the SNAP-25 linker is functionally relevant. However, a SNAP-25aL23 chimera with the extra 10 amino acids deleted (SN25AL2310) still displayed what we will refer to as the "linker phenotype," i.e., incomplete rescue and a lower rate of exocytosis triggering at intermediate [Ca2+]i, which was overcome at higher concentrations. This is shown in Figure 6A, where we flashed to normal [Ca2+]i (20 µM, light blue traces) and very high [Ca2+]i (150–200 µM, blue traces). The capacitance trace at very high [Ca2+]i is contaminated by the population of noncatecholamine containing vesicles. Even though this procedure therefore does not allow detailed analysis like in Figure 3, the amperometric measurements performed in parallel (Figure 6A, bottom traces) clearly show rescue of catecholamine release at high [Ca2+]i, and therefore allows the fast distinction between the normal and the linker phenotype. Next, we constructed two chimeras where either the N- or the C-terminal half of the linker was from SNAP-25 (Figure 5). The former, SN25aL2310-1, where the domain containing the linker cysteines was from SNAP-25, nevertheless displayed the linker phenotype (Supplemental Figure 3). Conversely, the chimera where the N-terminal part—and therefore the linker cysteines—was from SNAP-23, but the rest of the linker from SNAP-25 (SN23aL2310-2) exhibited full rescue and fast secretion indistinguishable from SNAP-25a (Figure 6B).
|
|
A Short Amino Acid Stretch in the C-Terminal Half of the SNAP-25 Linker Speeds Up Exocytosis
To further narrow down the decisive part of the linker, we continued testing the chimeras displayed in Figure 5. In brief, we found that all mutants depicted in red displayed the linker phenotype, i.e., secretion like SN25aL23, whereas mutants in green had a SNAP-25a–like secretory phenotype and mutants in blue showed an intermediate phenotype. The minimum region that gave an unperturbed SNAP-25-like phenotype was a 10-amino acid stretch encompassing the positions 120–129 in SNAP-25. Chimeras including this stretch were expressed and targeted to the membrane at wild-type levels (Figures 1C and 4B) and restored secretion to normal amplitude and kinetics (Figure 7 and Supplemental Figure 4A), regardless of whether the construct included the extra seven amino acids present in SNAP-23 N-terminal of this domain. However, a chimera containing the extra three amino acids (GAA) present in SNAP-23 immediately C-terminal of the 10-amino acid stretch from SNAP-25 displayed an intermediate phenotype (Supplemental Figure 4B), showing that the domain needs to be fused directly to the remaining part of the linker, which is largely conserved between SNAP-25 and SNAP-23.
|
A chimeric protein where the two initial valines were substituted for the SNAP-23 residues isoleucine and threonine (SN25aL2310-6) resulted in a construct with an intermediate phenotype between the linker phenotype and SNAP-25a (Figure 8A). With this construct, the normal amplitude of the exocytotic burst was recognizable, but the triggering rate was still somewhat decreased and also the sustained component of release was depressed. Therefore, the N-terminal hydrophobicity of the SNAP-25 domain is necessary for full rescue. Another chimera (SN25aL2310-4) was created by substituting the last three amino acids of the stretch (MAI) with the uncharged SNAP-23 residues (QTT). This construct displayed the linker phenotype (Figure 8B); thus, the C-terminal part of the domain is absolutely necessary for fast exocytosis triggering.
|
10-8) sufficed to speed the secretion up compared with the linker phenotype, even though not quite as much as when all three C-terminal amino acids MAI were from SNAP-25 (SN25aL2310-7, Figure 9, A and B). The latter construct restored the size of the burst, but still left time constants of triggering and the delay slowed down by a factor 2–4. Also the sustained component of release was depressed. This is interesting, because in this construct the proline-127 from SNAP-23 was present, and the central charges (DERE) in SNAP-25 were missing. Furthermore, the two chimeras where the last three amino acids (MAI) were from SNAP-25 (SN25aL2310-6 and SN25aL2310-7) displayed indistinguishable phenotypes (compare Figure 8A and Figure 9A), even though the former construct contained the middle charged stretch (DEREQ) from SNAP-25, whereas the latter had the SNAP-23 sequence (NGQPQ). Thus, both the C-terminal and N-terminal hydrophobicity of this stretch and especially methionine-127 is important for fast exocytosis triggering, whereas the middle stretch seems less important, because even a proline in this area does not impair secretion.
|
Altogether, we conclude that the hydrophobicity of the N- and C-terminal ends of this linker domain is of critical importance for fast triggering, and the domain cannot be further subdivided without loosing functionality. Therefore, the 10 amino acids in the SNAP-25 linker from position 120–129 display the features of a single protein domain acting in fast exocytosis triggering.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We analyzed the role of the palmitoylated cysteines in a null background, by expression in Snap-25 knockout chromaffin cells. Our data confirm the function of the cysteines in targeting SNAP-25 to the plasma membrane (Hess et al., 1992; Veit et al., 1996; Lane and Liu, 1997; Gonzalo et al., 1999; Vogel and Roche, 1999; Gonelle-Gispert et al., 2000; Koticha et al., 2002; Loranger and Linder, 2002; Kammer et al., 2003), in which it acts in exocytosis by binding to the Q-SNARE partner syntaxin-1. Removing single cysteines reduced the amount of SNAP-25 on the plasma membrane to <50%, but it did not impair secretion, as previously shown in insulin-secreting cells (Gonelle-Gispert et al., 2000). A SNAP-25 without cysteines was expressed at lower levels than WT protein, but still induced significant rescue. The resulting secretion was slowed down and the duration of the fusion pore increased. Thus, the cysteines are important for the speed of exocytosis. The observed rescue agree with the finding by several groups that nonpalmitoylated SNAP-25 can support membrane fusion in in vitro assays (Scales et al., 2000b; Schuette et al., 2004) or after infusion into BoNT/E-treated PC12 cells (Scales et al., 2000b). The slow-down of secretion found here using high time-resolution methods would not have been noticeable in those intrinsically slow assays.
It has been suggested that SNAP-25 and SNAP-23 are differentially targeted to membrane rafts, due to the presence of an extra cysteine in SNAP-23 (Salaun et al., 2005a,b). Others have failed to find SNAREs in lipid rafts (Lang et al., 2001), but they identified cholesterol-dependent SNARE clusters in plasma membrane sheets. We showed that the secretory phenotype of a chimeric SNAP-25a construct where the N-terminal half of the linker—including all cysteines—was from SNAP-23 did not deviate from SNAP-25a in our assay. Thus the difference in secretory phenotype between SNAP-25 and SNAP-23, which was found after overexpression (Sorensen et al., 2003b), cannot be due to a different arrangement of cysteines. Likewise, we found that the cysteine clusters of the two SNAP-25 splice variants are functionally equivalent (Nagy et al., 2005). Nevertheless, it remains possible that at lower expression levels differential targeting of SNAP-25 and SNAP-23 could be functionally important.
Using chimeric proteins we identified a 10-amino acid stretch in the C-terminal half of the SNAP-25 linker, which is necessary for fast exocytosis triggering. This stretch begins immediately C-terminal of the minimal domain required for palmitoylation of SNAP-25 in vivo (Gonzalo et al., 1999) and did not affect expression level or membrane targeting. It consists of hydrophobic and charged amino acids, whereas the stretch in SNAP-23 is hydrophilic, but uncharged. Testing of chimera underlined the importance of the hydrophobicity for functionality. This domain is conserved in SNAP-25 from zebrafish to human (Figure 10A), whereas in the fly and worm one or two of the initial hydrophobic amino acids are present together with the important methionine and one-three charges in the middle (Figure 10A). An interesting exception is formed by the sea urchin, where the critical methionine is replaced by a cysteine. Cysteines are quite hydrophobic, even when not palmitoylated. Limited structural studies performed using CD spectroscopy showed that both are unstructured in solution (Supplemental Figure 5), in agreement with previous data (Margittai et al., 2001). Also, our electrophysiological data indicate the lack of secondary structure, since replacing the middle part of the 10-amino acid stretch with residues from SNAP-23, which involved the insertion of a proline, did not further exacerbate the intermediate phenotype found upon replacement of the first two valines. Finally, the stretch is flanked to both sides by helix-breakers (prolines, glycines), making it unlikely that this domain would change the structure of the SNARE-domains through a cis-action.
|
). However, Chieregatti et al. (2004) also showed that synaptotagmin-7 suppresses SNAP-23–induced basal release. Synaptotagmin-7 is expressed in and modulates secretion from chromaffin cells (Schonn et al., 2008). Thus, the ability of SNAP-23 to stimulate exocytosis under resting conditions depends on the expression of auxiliary proteins and/or the cell type. In addition, we found that the rate of sustained release at high [Ca2+]i was slowed down by the SNAP-23 linker (Figure 3, B and F). In a sequential model of exocytosis (Figure 10B), the priming rate can be measured from the sustained release component, but only if the fusion rate is kept much higher than the priming rate during the experiment. In the SN25L23 chimera, this assumption breaks down because of the dramatic slowdown of fusion triggering, which might have caused newly recruited vesicles to become "trapped" in the SRP. Simulations of the model in Figure 10B by using triggering rates estimated from Figure 3 showed that about half of the decrease in sustained component might be explained by the decrease in fusion triggering (data not shown), whereas the other half might represent a real depression of the vesicle priming reaction.
The displacement of the intracellular calcium dependence of exocytosis triggering is a very specific phenotype. Indeed, it is striking that the only molecular manipulations that are known—or are likely based on published literature—to change the intracellular calcium-dependence of exocytosis triggering are mutations in synaptotagmins and SNAP-25 (Sorensen et al., 2003a; Wang et al., 2003; Chieregatti et al., 2004; Rhee et al., 2005; Wang et al., 2005; Nagy et al., 2006; Pang et al., 2006; Sorensen et al., 2006). In contrast, manipulations of proteins involved in exocytosis more often lead to changes in pool sizes and/or recruitment in the absence of a change in triggering: examples include tomosyn (Yizhar et al., 2004), Munc13 (Ashery et al., 2000), Munc18 (Voets et al., 2001; Gulyas-Kovacs et al., 2007), CAPS1 (Speidel et al., 2005), SV2 (Xu and Bajjalieh, 2001), Snapin (Tian et al., 2005), synaptobrevin (Borisovska et al., 2005), and 
-SNAP/NSF (Xu et al., 1999). Our present work extends the limited number of molecular manipulations that change the intracellular calcium dependence of exocytosis triggering to include the SNAP-25 linker.
There is ample evidence that synaptotagmin-1 binds to the SNARE complex or to individual SNAREs through interactions with one or both C2-domains (Bennett et al., 1992; Sollner et al., 1993; Chapman et al., 1995; Schiavo et al., 1997; Zhang et al., 2002; Rickman and Davletov, 2003; Shin et al., 2003; Bai et al., 2004; Bhalla et al., 2006; Pang et al., 2006; Lynch et al., 2007). It is possible that the SNAP-25 linker participates in synaptotagmin binding in situ. Another possibility, with would agree with the importance of the hydrophobic residues, is that the SNAP-25 linker affects interactions with the membrane that take place simultaneously with Ca2+ binding to synaptotagmin-1. Finally, the linker might affect stages of SNARE complex assembly, which could change both vesicle priming and fusion reactions (Sorensen et al., 2006), possibly by stabilizing an intermediate conformation of the SNARE complex (An and Almers, 2004). The picture emerging is that SNAREs, synaptotagmins, and lipids form an integrated fusion machine (Bhalla et al., 2006; Pang et al., 2006; Dai et al., 2007), whose calcium dependence is set by the properties of the entire assembly. The SNAP-25 linker must be seen as an integral part of this machine and it is tempting to speculate that the fusion of the Qb and Qc-SNARE motifs might have evolved as an adaptation toward calcium triggering of exocytosis.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
These authors contributed equally to this work. 
Present addresses: 
National Centre for Stereotactic Radiosurgery, Royal Hallamshire Hospital, Sheffield S10 2JF, United Kingdom;
Department of Cell Biology, Yale University, School of Medicine, New Haven, 06510 CT.
Address correspondence to: Jakob B. Sørensen (jsoeren{at}gwdg.de)
Abbreviations used: LDCV, large dense-core vesicle; RRP, readily-releasable pool; SNAP-25, synaptosome-associated protein of 25 kDa; SNARE, soluble N-ethylmaleimide-sensitive factor attachment protein receptors; SRP, slowly releasable pool; WT, wild type.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ashery, U., Betz, A., Xu, T., Brose, N., and Rettig, J. (1999). An efficient method for infection of adrenal chromaffin cells using the Semliki Forest virus gene expression system. Eur. J. Cell Biol 78, 525–532.[Medline]
Ashery, U., Varoqueaux, F., Voets, T., Betz, A., Thakur, P., Koch, H., Neher, E., Brose, N., and Rettig, J. (2000). Munc13-1 acts as a priming factor for large dense-core vesicles in bovine chromaffin cells. EMBO J 19, 3586–3596.[CrossRef][Medline]
Bai, J., Wang, C. T., Richards, D. A., Jackson, M. B., and Chapman, E. R. (2004). Fusion pore dynamics are regulated by synaptotagmin*t-SNARE interactions. Neuron 41, 929–942.[CrossRef][Medline]
Bennett, M. K., Calakos, N., and Scheller, R. H. (1992). Syntaxin: a synaptic protein implicated in docking of synaptic vesicles at presynaptic active zones. Science 257, 255–259.
Bhalla, A., Chicka, M. C., Tucker, W. C., and Chapman, E. R. (2006). Ca(2+)-synaptotagmin directly regulates t-SNARE function during reconstituted membrane fusion. Nat. Struct. Mol. Biol 13, 323–330.[CrossRef][Medline]
Borisovska, M., Zhao, Y., Tsytsyura, Y., Glyvuk, N., Takamori, S., Matti, U., Rettig, J., Sudhof, T., and Bruns, D. (2005). v-SNAREs control exocytosis of vesicles from priming to fusion. EMBO J 24, 2114–2126.[CrossRef][Medline]
Chapman, E. R., Hanson, P. I., An, S., and Jahn, R. (1995). Ca2+ regulates the interaction between synaptotagmin and syntaxin 1. J. Biol. Chem 270, 23667–23671.
Chieregatti, E., Chicka, M. C., Chapman, E. R., and Baldini, G. (2004). SNAP-23 functions in docking/fusion of granules at low Ca2+. Mol. Biol. Cell 15, 1918–1930.
Dai, H., Shen, N., Arac, D., and Rizo, J. (2007). A quaternary SNARE-synaptotagmin-Ca(2+)-phospholipid complex in neurotransmitter release. J. Mol. Biol 367, 848–863.[CrossRef][Medline]
Fasshauer, D. (2003). Structural insights into the SNARE mechanism. Biochim. Biophys. Acta 1641, 87–97.[Medline]
Fasshauer, D., Eliason, W. K., Brunger, A. T., and Jahn, R. (1998a). Identification of a minimal core of the synaptic SNARE complex sufficient for reversible assembly and disassembly. Biochemistry 37, 10354–10362.[CrossRef][Medline]
Fasshauer, D., Sutton, R. B., Brunger, A. T., and Jahn, R. (1998b). Conserved structural features of the synaptic fusion complex: SNARE proteins reclassified as Q- and R-SNAREs. Proc. Natl. Acad. Sci. USA 95, 15781–15786.
Fukuda, R., McNew, J. A., Weber, T., Parlati, F., Engel, T., Nickel, W., Rothman, J. E., and Sollner, T. H. (2000). Functional architecture of an intracellular membrane t-SNARE. Nature 407, 198–202.[CrossRef][Medline]
Gonelle-Gispert, C., Molinete, M., Halban, P. A., and Sadoul, K. (2000). Membrane localization and biological activity of SNAP-25 cysteine mutants in insulin-secreting cells. J. Cell Sci 113, 3197–3205.[Abstract]
Gonzalo, S., Greentree, W. K., and Linder, M. E. (1999). SNAP-25 is targeted to the plasma membrane through a novel membrane-binding domain. J. Biol. Chem 274, 21313–21318.
Gulyas-Kovacs, A., de Wit, H., Milosevic, I., Kochubey, O., Toonen, R., Klingauf, J., Verhage, M., and Sorensen, J. B. (2007). Munc18-1, sequential interactions with the fusion machinery stimulate vesicle docking and priming. J. Neurosci 27, 8676–8686.
Hess, D. T., Slater, T. M., Wilson, M. C., and Skene, J. H. (1992). The 25 kDa synaptosomal-associated protein SNAP-25 is the major methionine-rich polypeptide in rapid axonal transport and a major substrate for palmitoylation in adult CNS. J. Neurosci 12, 4634–4641.[Abstract]
Jackson, M. B., and Chapman, E. R. (2006). Fusion pores and fusion machines in Ca2+-triggered exocytosis. Annu. Rev. Biophys. Biomol. Struct 35, 135–160.[CrossRef][Medline]
Jahn, R., and Scheller, R. H. (2006). SNAREs–engines for membrane fusion. Nat. Rev. Mol. Cell Biol 7, 631–643.[CrossRef][Medline]
Kammer, B., Schmidt, M. F., and Veit, M. (2003). Functional characterization of palmitoylated and nonacylated SNAP-25 purified from insect cells infected with recombinant baculovirus. Mol. Cell Neurosci 23, 333–340.[CrossRef][Medline]
Kloepper, T. H., Kienle, C. N., and Fasshauer, D. (2007). An elaborate classification of SNARE proteins sheds light on the conservation of the eukaryotic endomembrane system. Mol. Biol. Cell 18, 3463–3471.
Koticha, D. K., McCarthy, E. E., and Baldini, G. (2002). Plasma membrane targeting of SNAP-25 increases its local concentration and is necessary for SNARE complex formation and regulated exocytosis. J. Cell Sci 115, 3341–3351.
Lane, S. R., and Liu, Y. (1997). Characterization of the palmitoylation domain of SNAP-25. J. Neurochem 69, 1864–1869.[Medline]
Lang, T., Bruns, D., Wenzel, D., Riedel, D., Holroyd, P., Thiele, C., and Jahn, R. (2001). SNAREs are con
