| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Department of Neurobiology, The Hebrew University, Jerusalem, Israel; and 2Department of Neurochemistry, Max-Planck Institute for Brain Research, Frankfurt am Main, Germany
Submitted 14 December 2005; accepted in final form 20 May 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Südhof 2004). Among these, the synaptosomal soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs) syntaxin, SNAP-25 and synaptobrevin and their cytoplasmic binding partners N-ethylmaleimide-sensitive fusion protein (NSF) and soluble NSF attachment protein (
,
,
-SNAP) have received particular attention because these proteins are central to an evolutionarily conserved protein cascade of protein interactions that is thought to catalyze constitutive and regulated membrane fusion in all eukaryotic cells (Rothman 1994; Ungar and Hughson 2003). Accordingly, SNARE complex formation is essential to drive the exocytotic fusion reaction, whereas NSF catalyzed dissociation of the postfusion SNARE complex allows the SNAREs to re-acquire fusion competence.
Although consensus exists on the postfusion role of NSF (and SNAPs) in SNARE recycling, its importance for membrane fusion per se is less clear. Gene ablation studies in Drosophila (Kawasaki et al. 1998; Tolar and Pallanck 1998) and biochemical experiments with different eukaryotic preparations (reviewed in Haas 1998; Whiteheart et al. 2001) are consistent with NSF functioning solely as a SNARE chaperone that is required to maintain the readily releasable pool of synaptic vesicles. Membrane fusion of docked secretory vesicles in sea urchin eggs occurs efficiently in the absence of NSF (Whalley et al. 2004), and the effects of N-ethyl-maleimide on chromaffin granule exocytosis monitored amperometrically indicate a requirement for NSF at an early priming step of release that precedes the exocytotic fusion reaction and is uncoupled from the terminal fusion event (Xu et al. 1999a). On the other hand, microinjection experiments at the squid giant synapse have suggested that NSF may contribute to mechanisms that control the time course of evoked transmitter release. This conclusion was based on the observation that both the rise and decay times of the synaptic current were slowed on injection of two inhibitory NSF-derived peptides, NSF-2 and NSF-3 (Schweizer et al. 1998). In an attempt to resolve these discrepancies, we now performed NSF peptide microinjection studies using a nerve terminal preparation that is suited to electrophysiologically dissect different aspects of the release process.
The crayfish opener neuromuscular junction (NMJ) is a glutamatergic synapse, in which single quanta can be readily detected (Dudel 1981). Its first daughter branch (15–20 µm) is sufficiently large to impale with a microelectrode and to inject different compounds including proteins (Ravin et al. 1997). Moreover, the time course of transmitter release can be accurately followed in this preparation by establishing synaptic delay histograms. We report that microinjection of the NSF peptides, NSF-2 and NSF-3, into presynaptic terminals of the crayfish neuromuscular junction reduced the quantal content, as found in the squid (Schweizer et al. 1998) but did not alter the synaptic delay histogram. Our results corroborate an essential function of NSF in neurotransmitter exocytosis but argue against NSF having a general role in controlling the time course of neurotransmitter release.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Recombinant hamster His6-NSF-myc prepared as described previously (Pellegrini et al. 1995) was used to immunize two rabbits (Eurogentec, Seraing, Belgium). Both sera reacted with His6-NSF-myc, but not with His6--SNAP, immobilized on nitrocellulose, and recognized a single band at 80 kDa, i.e., the expected size for NSF, in rat brain homogenate.
Frozen crayfish ganglia or rat brains were homogenized in 10 volumes of 20 mM HEPES, pH 7.4, 100 mM KCl, containing the protease inhibitor cocktail "complete" [Roche Molecular Biochemicals, Mannheim, Germany]. Approximately 30 µg of protein dissolved in sample buffer were separated by 8% SDS polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were transferred to nitrocellulose, incubated with NSF antiserum, and bound immunoreactivity was detected using an Enhanced Chemiluminescence substrate (KMF, St. Augustin, Germany) as described (Pellegrini et al. 1995).
Specificity was assessed by comparing the immunoreactivities of untreated sera with those of antisera preincubated with recombinant His6-NSF-myc. To this end, 200 µg of His6-NSF-myc were either immobilized on 500 µl of Ni-agarose (Qiagen, Hilden, Germany) or resolved by 8% SDS-PAGE and transferred to nitrocellulose. The region of the nitrocellulose containing His6-NSF-myc was cut into 1-cm strips to provide an immobilized source of denatured NSF. Diluted (1:1000) sera were then incubated with both forms of immobilized His6-NSF-myc for 16 h at 4°C before being used for Western analysis.
Physiological experiments
The opener muscle of the first walking leg of the crayfish Procambarus clarkii was used as described previously (Dudel 1981; Ravin et al. 1997). Crayfish were imported from Louisiana (Atchafalaya, Raceland, LA) and kept in tanks with filtered running fresh water. The preparation was mounted on the stage of a compound microscope (Zeiss Axioscop FS) equipped with a long-distance objective (Acroplan x40) (Ravin et al. 1997). Unless otherwise stated, the preparation was constantly superfused (Gilson pump Minipulse 3, Villiers, France) at 11–12°C with a modified van Harreveld solution composed of the following salts (in mM) 220 NaCl, 5.4 KCl, 13.5 CaCl2, 2.5 MgCl2, and 10 Tris, with the pH being adjusted to 7.4 by adding NaOH.
Intra-preterminal injections
The first daughter branch, after the main bifurcation of the excitatory axon, was impaled, about midway from the main bifurcation and the end of the branch with a microelectrode filled with injection solution. This intra-axonal electrode was also used to record the axon action potential. Because of the small size of the axon and release boutons, two brief pressure puffs, of 10 s each, were sufficient to fill the boutons (using a pico-injector, PLI-100, Medical System, Greenvale, NY) as assessed by accumulation of fluorescently labeled rhodamine. Peptides were dissolved at a final concentration of 1 mM together with 150 mM KCl and 10 mM Tris buffer. The pH was adjusted to 7.2 with KOH. The solution which also contained 0.4% dextran-Rhodamine B (MW 10,000, molecular probes) was filtered twice through 0.2-µm filters. A part of this solution was used immediately for injections. The rest was divided into aliquots of 0.2 ml and kept at –70°C. No differences were seen between fresh or frozen solutions. The NSF-2 or NSF-3 concentration in a bouton was estimated by fluorescence of the co-injected Rhodamin dextran (Schweizer et al. 1998).
Stimulation and recording
Two modes of electrical stimulation were employed. For axon stimulation, a small suction electrode was placed over the parent branch of the single excitatory axon that innervates the opener muscle. Pulses of 0.2 ms, with twice the threshold, were used. With this method of stimulation, the quantal content depends on the amplitude and shape of the action potential arriving at the presynaptic terminal, and these depend on ionic conductances that may be affected by the injected peptide.
The other mode of stimulation, a direct depolarization of a single release bouton by brief negative current pulses using a macropatch electrode, is insensitive to changes as large as, and exceeding, 50% in the presynaptic membrane resistance (Dudel 1981, 1983). Thus unless otherwise stated, we employed mainly direct depolarizing pulses (Dudel 1981; Ravin et al. 1997). Pulses of –0.7 µA, 0.7 ms, which produced a quantal content of 0.15–0.5, were used. These experiments were done in the presence of 50 nM TTX (Dudel 1981; Ravin et al. 1997).
For both modes of stimulation, quanta were recorded with a macropatch electrode (Dudel 1981) as detailed in Ravin et al. (1997). The macropatch electrode, which had an opening of 
8 mm, was placed under visual control over a single release bouton at a distance of 50 µm away from the intraaxonal injection site. Quantal content was established for consecutive sets of 360 pulses at 3 Hz. Between successive measurements, the quantal content could vary by 10–12%. Therefore changes in this range were not considered to be produced by the injected peptide. When the quantal content had stabilized under control conditions, the peptide was injected, and quantal content was monitored continuously. Values of quantal content are given as mean ± SD. Statistical significance was evaluated with the paired or unpaired t-test. To obtain an average size of single quanta, we sampled 100 single quanta during each experiment without any selection. The beginning of each quantum was shifted to a common zero point for averaging.
Establishing synaptic delay histograms
Traces were digitized (50 kHz) using the Labview interface (AT-MIO-16F-5,NI-DAQ 4.9.0 driver software, National Instruments, Austin, TX) and a Pentium computer (Pentium 3,500 MHz). For action-potential-evoked release, and to cancel the effect of distance between the stimulating and recording electrodes, the delay between the negative peak of the excitatory nerve terminal current (ENTC) and the beginning of each quantum was measured (Fig. 5). For direct depolarizing pulses, the delay was measured from the beginning of the depolarizing pulse (Ravin et al. 1999). Delay histograms were constructed by grouping the delay values into bins of 0.25 ms and presenting the data as a continuous delay curve by connecting the midpoint of each bin. To allow for quantitative comparison between histograms, the number of pulses applied under each of the experimental conditions tested (control, injected, recovery) was kept constant (Wojtowicz et al. 1987) in each experiment, being 
2,000. The delay curves presented correspond to the number of quanta during a given time bin as percent of the number of quanta at the peak of the histogram (normalized histograms). The delay curves were smoothed by weighted average of nine nearest neighbors (Graphpad prism 3.0). The decay phase of the delay histogram was fitted by a single exponent to establish the time constant of decay (
D) (Slutsky et al. 2001).
|
Another sensitive way to analyze synaptic delay histograms is to produce a curve of the cumulative delays (Bukcharaeva et al. 1999). The interval between the minimal synaptic delay and the time at which 90% of all measured quantal delays is defined as 90. To obtain the average cumulative curves, the minimal delays of each curve were adjusted to a common zero point. Statistical significance of the difference between the 90 of the average curves was assessed by unpaired t-test.
Reagents
Tetrodotoxin was purchased from Sigma (Rehovot, Israel). Peptides were synthesized, and their purity confirmed by HPLC, at Research Genetics (Huntsville, AL). Rhodamin-Dextran B, M.W. 10,000, was purchased from Molecular Probes (Eugene, OR).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
We used antisera raised against recombinant hamster His6-NSF-myc to identify endogenous NSF in crayfish nervous tissue. As shown in Fig. 1A (and data not shown), the sera recognized a single protein of 80 kDa in rat brain membranes. In crayfish, two bands of 75 and 85 kDa, respectively, were stained by these antisera (Fig. 1A). The specificity of this immunoreactivity was confirmed by preadsorbing the sera with recombinant His6-NSF-myc. Immunoreactivities of both the rat and crayfish NSF bands were strongly reduced on preincubation with the original antigen (Fig. 1B). The presence of two bands in crayfish may reflect proteolysis of the endogenous protein or indicate the presence of two related, but distinct, NSF genes, as described for other invertebrate species (Boulianne and Trimble 1995). The ability to detect crayfish NSF with antisera raised against a mammalian NSF protein further supports the previously noted high conservation of NSF sequences between vertebrates and invertebrates (Schweizer et al. 1998).
|
). Here we used these peptides to block NSF activity in the presynaptic terminal of the crayfish NMJ. NSF and NSF-3 reduce the quantal content when inducing release by direct focal depolarization
To unravel whether NSF affects neurotransmitter release at the crayfish NMJ, we microinjected NSF-2 or NSF-3 peptides into the presynaptic terminal. This resulted in rapid filling (within 2–3 min) of the terminal and the release boutons as evident from measuring fluorescence intensity of the co-injected dye. The intracellular concentration of NSF-2 (or NSF-3), as estimated from dye fluoresfscence, reached values of 0.5–0.6 mM, and this concentration remained constant for >40 min. Figure 2A shows samples of recordings under control conditions (preinjection) (a), after NSF-2 injection (b), and after recovery (c). The quantal content (m) under control conditions stabilized at an average of m = 0.19 ± 0.03. Five minutes after NSF-2 (estimated intraterminal concentration of 0.6 mM) injection the quantal content declined, on average, to 0.04 ± 0.01 (corresponding to an 80% block, significant at P < 0.0001), and after 30 min, the quantal content recovered to 0.17 ± 0.02 (90% of control, insignificant P > 0.324). In 14 similar experiments injection of NSF-2 (0.5–0.6 mM intraterminally) reduced the quantal content on average by 63 ± 12.1% (P < 0.0001) and recovered to 98.9 ± 4.4% of the control.
|
30 min. This result implies that the dye, and probably the peptide, did not diffuse backwards into the axon. Several explanations may account for the recovery of the quantal content. First, the peptide might be hydrolyzed by proteolytic enzymes. Alternatively the peptide might be bound to other proteins or be sequestered to intracellular compartments. Another possibility could be some sort of desensitization to peptide action. This, however, can be excluded because after recovery of the quantal content, a second injection (1-min interval) of the peptide reduced release to a similar extent. This is illustrated in the experiment shown in Fig. 2C. Here, the control quantal content was 0.22 ± 0.03. After the first injection, the quantal content dropped by 64% to 0.08 ± 0.02. Release recovered after 9 min to a value of 0.2 ± 0.02, and a second peptide injection reduced release again to a value of 0.08 ± 0.02, which after 12 min, recovered to 0.22 ± 0.02. In both cases, the reduction was highly significant (P < 0.0001). Similar results were obtained in five independent experiments. After the first injection, the quantal content dropped on average by 46 ± 3.8% and after the second injection (1–2 min interval) by 53.4 ± 6.5%. The duration of peptide inhibition appeared to be correlated to the reduction in quantal content. The smaller the reduction in quantal content the shorter was the period of reduced quantal content (Fig. 2D). These experiments also suggest that there was no leakage of the peptide from the injection pipette into the terminal because recovery of quantal content was always close to the preinjection values, and always high pressure pulses were required to obtain peptide effect.
Control experiments showed that the average size of the single quantum event (sensitive to seal resistance and distance from the release bouton) did not change during the entire experiment (Fig. 2B, 
).
Also, injection of a peptide with the same amino acid composition as NSF-2, but a scrambled amino acid sequence had no effect on the quantal content, even when two injections were given and similar concentrations (0.6 mM) were reached (Fig. 2E). The small fluctuations in quantal content seen after injection were in the same range as observed during the preinjection period. In four experiments with scrambled NSF-2, the quantal content did not decline after injection (average quantal content after injection remained 101 ± 8.6% of the control). In these experiments, we also established synaptic delay histograms (see following text) and found no change in synaptic delay histograms after the injections (D in the control was 0.38 ± 0.1 ms and after injection D remained 0.38 ± 0.01 ms). This shows that the effect of NSF-2 is sequence-specific.
Microinjection of NSF-3 reduced the quantal content to a similar extent as NSF-2. For the experiment shown (Fig. 2F), control quantal content was 0.22 ± 0.01. After NSF-3 injection (0.5 mM), the quantal content declined to 0.09 ± 0.02 (60% block) and recovered after 40 min to 0.18 ± 0.01 (82% of the control P = 0.078). In 18 experiments, the average reduction in quantal content after NSF-3 injection (estimated intraterminal concentration of 0.5–0.6 mM) was 69 ± 9.02%. Again, the amplitude of the single quanta events did not change throughout the experiment (Fig. 2F, ).
Effects of NSF-2 and NSF-3 on synaptic delay histograms
The results presented in the preceding text corroborate the notion that NSF is involved in processes that determine the amount of release or quantal content (Schweizer et al. 1998). To examine whether NSF is also involved in the control of the time course of evoked release at the crayfish NMJ, we constructed synaptic delay histograms under control conditions and after intraterminal injection of NSF-2, or NSF-3 at the time of maximal reduction of quantal content. Figure 3A shows that NSF-2 injection (1st 
) reduced the quantal content (similar concentration and time course as in Fig. 2). During the period of suppressed release, we administered two additional injections to ensure a longer duration of peptide inhibition.
|
D of the control was 0.37 ms, and it remained 0.37 ms after injection. Furthermore, the cumulative delay histograms were actually the same, and in both cases the 90 was 2.6 ms (Fig. 3C). Figure 3D shows that NSF-3 injection similarly did not change the time course of release. The quantal content dropped from m = 0.23 ± 0.01 in control to m = 0.08 ± 0.01 after injection of NSF-3 (0.6 mM), and again, the two delay histograms were indistinguishable. In 32 experiments with NSF-2 or NSF-3 injections, the average D in the control was 0.37 ± 0.01 ms, and it remained 0.37 ± 0.01 ms after injection. We hence conclude that, in the crayfish NMJ, both NSF-2 and NSF-3 are effective in reducing the quantal content but have no effect on the time course of glutamate release. Experiments at 22°C
Our results resemble those of Schweizer et al. (1998) in that both NSF peptides reduced the quantal content. However, in contrast to these authors, we did not observe any effect of the peptides on the time course of release. One possible reason for this difference could be that our experiments were performed at 11–12°C, whereas Schweizer et al. (1998) used room temperature. To examine this point, we repeated the experiments shown in Figs. 2 and 3 at 22°C.
Figure 4A demonstrates that also at 22°C NSF-2 (0.5 mM, 2 injections, in Fig. 4A) reduced the average quantal content (control: m = 0.26 ± 0.02 after injection of NSF-2: m = 0.07 ± 0.01). Constructing synaptic delay histograms we found that at 22°C (Fig. 4B) release started sooner (0.4 ms), and the delay histogram was narrower, than at 11–12°C (Fig. 3B). Irrespective of these differences, also at 22°C injection of NSF-2 had no effect on the time course of release (Fig. 4, B and C).
|
Another possible reason for the differences between the results obtained here and by Schweizer et al. (1998) could be the stimulus employed. We depolarized the release bouton directly, whereas Schweizer et al. (1998) induced release by action potentials. Hence, the motor axon was stimulated by a suction electrode, whereas release was recorded from a single bouton with the macropatch electrode. Figure 5A shows three samples of recordings. In all three traces, the ENTC is visible (). The first trace shows failure of release, whereas the other two show single quanta events appearing at different delays (*). Figure 5C shows that also for action potential evoked release, injection of NSF-2 reduced release. The quantal content dropped from a control value of m = 0.18 ± 0.02 to m = 0.08 ± 0.01 after injection and the effect lasted for 25 min. The reduction in the quantal content was not caused by NSF-2 effects on the action potential. Figure 5B shows that both the intracellularly recorded action potential and the ENTC were the same under control conditions and during the period of suppressed release. Also, the time course of action potential induced release was unchanged after injection of NSF-2 (Fig. 5, D and E). In this experiment, the delay histograms before and after injection superimpose. In five experiments, D of the control was 0.37 ± 0.02 and it remained 0.37 ± 0.01 after injection. The average drop in quantal content was 70 ± 20.2%.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Schweizer et al. 1998) and forms the basis for our microinjection experiments with NSF-derived peptides. The main finding of the latter is that presynaptic injection of NSF-2 or NSF-3 reduced the quantal content without altering the time course of transmitter release, showing again that different mechanisms control the quantal content and time course of evoked release (Parnas and Parnas 1999). The peptides had no effect on the amplitude or shape of single quantal release events, nor did NSF-2 affect the amplitudes of the presynaptic action potential and the ENTCs. Also a peptide having the same amino acid composition as NSF-2 but a scrambled sequence did not affect transmitter release. We therefore conclude that NSF-2 and NSF-3 peptides specifically interfere with the mechanism that regulates the amount of neurotransmitter released.
The decline in quantal content caused by NSF-2 and NSF-3 (on average 66 ± 12%, n = 32) confirms the essential role of NSF in synaptic vesicle exocytosis. On microinjecting the NSF-2 or NSF-3 peptides, derived from the highly conserved D1 domain of NSF, into the preterminal region of the crayfish opener neuromuscular junction, we observed a comparatively rapid decline in quantal content. This rapid decline can be attributed to the small size of the release bouton (2–5 µm as compared with the large presynaptic terminal of the squid giant synapse) and its proximity to the injection electrode, parameters that both allow for rapid increases in peptide concentration within the bouton. Our findings are consistent with the data of Schweizer et al. (1998), who demonstrated reduced release at the squid giant synapse and inhibition of -SNAP stimulation of the ATPase activity of recombinant mammalian NSF by both NSF-2 and NSF-3.
The rise time of the current generated by a single quantum event is mainly determined by properties of the postsynaptic ionotropic receptors but also by the rate of discharge of the transmitter from the vesicle (Khanin et al. 1996, 1997; Van der Kloot 1995). The decay phase of the current depends not only on postsynaptic receptor channel properties but also on the time period during which the transmitter remains in the synaptic cleft (Hille 1992). Because the amplitude and shape of single quantum events were not changed on NSF-2 or NSF-3 injection, we conclude that the transmitter content of the synaptic vesicles was unaltered and that the time required for formation and dilation of the fusion-pore (Almers et al. 1991; Breckenridge et al. 1987) and for discharge of the transmitter into the synaptic cleft (Khanin et al. 1996, 1997) was not impaired by the peptides. Thus NSF seems not to be involved in these stages of the release process. This information could not be obtained from measurement of synaptic currents composed of many quanta as in the case of the squid giant synapse.
The reduced quantal content found after NSF-2 or NSF-3 injection is consistent with peptide inhibition of NSF-dependent SNARE complexes at a reaction important for the acquisition of fusion competence (Banerjee et al. 1996; Xu et al. 1999a). At the squid giant synapse, inhibition of transmitter release by NSF-2 or NSF-3 was paralleled by some slowing in the rise and decay times of the postsynaptic current (Schweizer et al. 1998). This was interpreted to indicate a change in the synchrony or kinetics of transmitter release. In case of the squid giant synapse, assaying release by monitoring postsynaptic currents did not discern whether the quantal and kinetic effects of NSF-2 or NSF-3 inhibition arose from the same or distinct molecular disruptions.
Two potential presynaptic reactions that might give rise to a kinetic effect of NSF-2 or NSF-3 in fast synapses include desynchronized stimulus-secretion coupling and a slowing in the fusion kinetics of individual vesicles (Schweizer et al. 1998). Despite the reduction in quantal content, the delay histogram analysis of evoked quanta performed here did not unravel a change in the time course of evoked release. Thus in the fast synapse of the crayfish, the function of NSF in regulating the amount of neurotransmitter released seems to be distinct from the reactions that control release kinetics. Our results confirm and extend a previous study in which, based on butulinum toxin E cleavage experiments, the SNARE complex was concluded to determine the probability of synaptic vesicle fusion but not release kinetics (Finley et al. 2002).
Both the quantal release analyzed in crayfish and the multiquantal release found at the squid giant synapse depend on NSF as an essential determinant of the amount of release. However, only in the squid did microinjection of NSF-2 or NSF-3 produce some slowing in the rise time of the postsynaptic current. This discrepancy may be explained by a distinct organization of the vesicle release machinery in these two synaptic preparations (Markram et al. 1998). Furthermore, the vesicle pools that contribute to the evoked response under conditions of low (single quanta) and high (multi-quantal) release probabilities may differ in harboring distinct conformations of NSF-dependent SNARE complexes (Chen et al. 1999; Xu et al. 1999a). Thus unraveling the reasons for the differential effects of NSF-2 or NSF-3 noted in crayfish and squid may provide insight into how the vesicle release machinery contributes to the control of transmitter release kinetics.
From the crayfish study, it seems that NSF is not involved in determining the probability of a single quantum to be released in time from the immediately releasable vesicle pool. This will also probably be correct for the release of a few quanta from a single or a few release sites in the crayfish. Ultrastructural studies have shown that single release boutons in the crayfish opener NMJ have 1–5 release sites (Atwood and Morin 1970) and that this synapse does not fatigue even after a long period of stimulation at high frequency. In the squid, the number of release sites in a presynaptic terminal is not known, and the postsynaptic current is composed of hundreds, if not more, quanta. Furthermore, in the squid the synapse depletes rapidly even after a single impulse. Is it possible that such a special system demands the recruitment of vesicles to the presynaptic membrane even during the release produced by a single action potential? Our findings indicate that NSF does not play a role in the control of the release kinetics of those vesicles which are already available, i.e., docked and "primed" (Südhof 2004) for release. However, they do not exclude a role of NSF in the rapid recruitment of nondocked vesicles even during release induced by a single action potential.
In conclusion, the data presented here extend and confirm the essential role of NSF in synaptic vesicle exocytosis. In addition, our study highlights the value of using single-quantum analysis to more precisely dissect the consequences of disrupting the function of presynaptic protein complexes. Our results contribute to a unified picture of NSF action that assigns a prefusion role to this cytoplasmic fusion protein (Banerjee et al. 1996; Hay and Scheller 1997; Xu et al. 1999a,b). In addition, they show that NSF is not a generic determinant of release kinetics under conditions of low quantal content. Here, presynaptic inhibitory autoreceptors appear to have a decisive role in determining the precise kinetics of exocytotic membrane fusion in nerve terminals (Parnas et al. 2000, 2005; Slutsky et al. 2001, 2003).
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present addresses: V. O'Connor, University of Southampton, Cell Sciences, Biomedical Sciences Bldg., Bassett Crescent East, Southampton SO16 7PX, UK; and O. El-Far, INSERM Unité 464, Faculté de Médicine Secteur Nord, Blvd. Pierre Dramard, 13916 Marseille Cedex 20, France.
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: I. Parnas, Dept. of Neurobiology, The Hebrew University, Jerusalem 91904, Israel (E-mail: parnas{at}huji.ac.il)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Atwood HL and Morin WA. Neuromuscular and axoaxonal synapses of the crayfish opener muscle. J Ultrastruct Res 32: 351–369, 1970.[CrossRef][ISI][Medline]
Banerjee A, Barry A, DasGupta BR, and Martin TFJ. N-ethylmaleimide-sensitive factor acts at a perfusion ATP-dependent step in Ca2+-activated exocytosis. J Biol Chem 271: 20223–20226, 1996.
Boulianne GL and Trimble WS. Identification of a second homolog of N-ethylmaleimide-sensitive fusion protein that is expressed in the nervous system and secretory tissues of Drosophila. Proc Natl Acad Sci USA 92: 7095–7099, 1995.
Breckenridge LJ and Almers W. Currents through the fusion pore that forms during exocytosis of a secretory vesicle. Nature 328: 814–817, 1987.[CrossRef][Medline]
Bukcharaeva EA, Kim KC, Moravec J, Nikolsky EE, and Vyskocil F. Noradrenaline synchronizes evoked quantal release at frog neuromuscular junctions. J Physiol 517: 789–888, 1999.
Chen YA, Scales SJ, Patel SM, Doung YC, and Scheller RH. SNARE complex formation is triggered by Ca2+ and drives membrane fusion. Cell 97: 165–174, 1999.[CrossRef][ISI][Medline]
Dudel J. The effect of reduced calcium on quantal unit current and release at the crayfish neuromuscular junction. Pflügers Arch Eur J Physiol 391: 35–40, 1981.[CrossRef][ISI][Medline]
Dudel J. Graded or all-or-nothing release of transmitter quanta by local depolarizations of nerve terminals on crayfish muscle? Pflügers Arch Eur J Physiol 398: 155–164, 1983.[CrossRef][ISI][Medline]
Finley MFA, Pate SM, Madison DV, and Scheller RH. The core membrane fusion complex governs the probability of synaptic vesicle fusion but not transmitter release kinetics. J Neurosci 22: 1266–1272, 2002.
Haas A. NSF-fusion and beyond. Trends Cell Biol 8: 471–473, 1998.[CrossRef][ISI][Medline]
Hay JC and Scheller RH. SNAREs and NSF in targeted membrane fusion. Curr Opin Cell Biol 9: 505–512, 1997.[CrossRef][ISI][Medline]
Hille B. Ionic Channels of Excitable Membranes (2nd ed.). Sunderland, MA: Sinauer, 1992.
Khanin R, Parnas H, and Segel L. A mechanism for discharge of charged excitatory neurotransmitter. Biophys J 72: 507–521, 1997.
Khanin R, Segel L, Parnas H, and Ratner E. Neurotransmitter discharge and postsynaptic rise time. Biophys J 70: 2030–2032, 1996.
Kawasaki F, Mattiuz AM, and Ordway RW. Synaptic physiology and ultrastructure in comatose mutants define an in vivo role for NSF in neurotransmitter release. J Neurosci 18: 10241–10249, 1998.
Lin RC and Scheller RH. Mechanisms of synaptic vesicle exocytosis. Annu Rev Cell Dev Biol 16: 19–49, 2000.[CrossRef][ISI][Medline]
Markram H, Pikus D, Gupta A, and Tsodyks M. Potential for multiple mechanisms, phenomena and algorithms for synaptic plasticity at single synapses. Neuropharmacology 37: 489–500, 1998.[CrossRef][ISI][Medline]
Parnas I and Parnas H. Different mechanisms control the amount and time course of neurotransmitter release. J Physiol 517.3: 629, 1999.
Parnas H, Segel L, Dudel J, and Parnas I. Autoreceptors, membrane potential and regulating of transmitter release. Trends Neurosci 23: 60–68, 2000.[CrossRef][ISI][Medline]
Parnas H, Slutsky I Rashkovan G, Silman I, Wess J, and Parnas I. Depolarization initiates phasic acetylcholine release by relief of a tonic block imposed by presynaptic M2 muscarinic receptors. J Neurophysiol 93: 3257–3269, 2005.
Pellegrini LL, O'Connor V, Lottspeich F, and Betz H. Clostridial neurotoxins compromise the stability of a low energy SNARE complex mediating NSF activation of synaptic vesicle fusion. EMBO J 14: 4705–4713, 1995.[ISI][Medline]
Ravin R, Parnas H, Spira ME, and Parnas I. Partial uncoupling of neurotransmitter release from [Ca2+]i by membrane hyperpolarization. J Neurophysiol 81: 3044–3053, 1999.
Ravin R, Spira ME, Parnas H, and Parnas I. Simultaneous measurement of intracellular Ca2+ and asynchronous transmitter release from the same crayfish bouton. J Physiol 501: 251–262, 1997.
Rothman JE. Mechanisms of intracellular protein transport. Nature 372: 55–63, 1994.[CrossRef][Medline]
Schweizer FE, Dresbach T, DeBello WM, O'Connor V, Augustine GJ, and Betz H. Regulation of neurotransmitter release kinetics by NSF. Science 279: 1203–1206, 1998.
Slutsky I, Silman I, Parnas H, and Parnas I. Presynaptic M2 muscarinic receptors are involved in controlling the kinetics of ACh release at the frog neuromuscular junction. J Neurophysiol 536: 717–725, 2001.
Slutsky I, Wess J, Gomeza AJ, Dudel J, Parnas I, and Parnas H. Use of knockout mice reveals involvement of M2-muscarinic receptors in control of the kinetics of acetylcholine release. J Neurophysiol 89: 1954–1967, 2003.
Söllner T, Bennett MK, Whiteheart SW, Scheller RH, and Rothman JE. A protein assembly disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle docking, activation, and fusion. Cell 75: 409–418, 1993.[CrossRef][ISI][Medline]
Sudhof TC. The synaptic vesicle cycle. Annu Rev Neurosci 27: 509–547, 2004.[CrossRef][ISI][Medline]
Sutton RB, Fasshauer D, Jahn R, and Brunger AT. Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 A° resolution. Nature 395: 347–353, 1998.[CrossRef][Medline]
Tolar LA and Pallanck L. NSF function in neurotransmitter release involves rearrangement of the SNARE complex downstream of synaptic vesicle docking. J Neurosci 18: 10250–10256, 1998.
Ungar D and Hughson FM. SNARE protein structure and function. Annu Rev Cell Dev Biol 19: 493–517, 2003.[CrossRef][ISI][Medline]
Van der Kloot W. The rise times of miniature endplate currents suggest that acetylcholine may be released over a period of time. Biophys J 69: 148–154, 1995.
Whallev T, Timmers K, Coorssen J, Bezrukov L, Kingsley DH, and Zimmerberg J. Membrane fusion of secretory vesicles of the sea urchin egg in the absence of NSF. J Cell Sci 117: 2345–2356, 2004.
Whiteheart SW, Schraw T, and Matveeva EA. N-ethylmaleimide sensitive factor (NSF) structure and function. Int Rev Cytol 207: 71–118, 2001.[ISI][Medline]
Wojtowicz JM, Parnas I, Parnas H, and Atwood HL. Latency of transmitter release at crayfish motor nerve endings examined by intracellular depolarization. Can J Physiol Pharmacol 65: 105–108, 1987.[ISI][Medline]
Xu T, Ashery U, Burgoyne RD, and Neher E. Early requirement for alpha-SNAP and NSF in the secretory cascade in chromaffin cells. EMBO J 18: 3293–3304, 1999a.[CrossRef][ISI][Medline]
Xu T, Rammner B, Margittai M, Artalejo AR, Neher E, and Jahn R. Inhibition of SNARE complex assembly differentially affects kinetics components of exocytosis. Cell 99: 713–722, 1999b.[CrossRef][ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  Y. M. Kupchik, G. Rashkovan, L. Ohana, T. Keren-Raifman, N. Dascal, H. Parnas, and I. Parnas Molecular mechanisms that control initiation and termination of physiological depolarization-evoked transmitter release PNAS, March 18, 2008; 105(11): 4435 - 4440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kuner, Y. Li, K. R. Gee, L. F. Bonewald, and G. J. Augustine Photolysis of a caged peptide reveals rapid action of N-ethylmaleimide sensitive factor before neurotransmitter release PNAS, January 8, 2008; 105(1): 347 - 352. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, quantal content. 
). The immunoreactivity of the 40-kDa band (* in A) recognized in the crayfish homogenate was not reduced on preadsorption, indicating a nonspecific interaction with the antiserum.
