HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 551/1/103    most recent jphysiol.2003.041152v1 Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Teng, H. Articles by Wilkinson, R. S. Search for Related Content PubMed PubMed Citation Articles by Teng, H. Articles by Wilkinson, R. S.

J Physiol (2003), 551.1, pp. 103-114
© Copyright 2003 The Physiological Society
DOI: 10.1113/jphysiol.2003.041152

'Delayed' endocytosis is regulated by extracellular Ca²⁺ in snake motor boutons

Haibing Teng and Robert S. Wilkinson

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

When cooled below ~7 °C, recently endocytosed vesicles in the motor terminals of the garter snake fail to shed their clathrin coats. Perhaps as a result, the terminals complete only about one-half of the compensatory endocytosis expected after a given period of stimulation. Upon return to room temperature (RT), endocytosis resumes immediately and is complete within minutes. This 'delayed' endocytosis following release from cold block provides an opportunity to study clathrin-dependent endocytotic mechanisms in temporal isolation from those events, such as Ca²⁺ entry and consequent exocytosis, that are normally associated with the activation of nerve terminals. We have taken advantage of clathrin decoating blockade to examine the rate, temperature dependence and extracellular Ca²⁺ dependence of endocytosis at the snake nerve-muscle synapse. Endocytosis was fast at RT (complete in < 1 min) and markedly faster still at 35 °C. Moreover, the rate of endocytosis varied significantly with change in [Ca²⁺]_o; the rate at 7.2 mM (single exponential time constant, ~3 s) was approximately double that at 0 mM (single exponential time constant, ~7 s). Thus, membrane retrieval via clathrin is rapid and, due to its dependence on [Ca²⁺]_o, potentially regulated by changes in the milieu of the synaptic cleft during neural activity.

(Received 7 February 2003; accepted after revision 12 May 2003; first published online 17 June 2003)
Corresponding author R. S. Wilkinson: Department of Cell Biology and Physiology, Washington University School of Medicine, 660 South Euclid Avenue, Box 8228, St Louis, MO 63110, USA. Email: wilk{at}cellbio.wustl.edu

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

Fast chemical synapses necessarily recycle synaptic vesicles rather than rely on the synthesis of new ones. Thus, activation of a nerve terminal not only triggers the exocytosis of transmitter, but also initiates the concomitant endocytosis of spent vesicular components for reuse. That the supply of competent vesicles (as well as the surface area of the bouton's plasma membrane) remains constant over time attests to the precision with which these exo- and endocytotic processes are matched (Betz et al. 1992; Wu & Betz, 1996; Murthy & Stevens, 1998; Sun et al. 2002). Yet, relatively little is known about what actually triggers endocytosis or how the matched complementary events of membrane fusion and fission are co-regulated. A problem in addressing these points experimentally is that exo- and endocytosis ordinarily occur simultaneously. Thus, exocytosis itself might initiate or regulate endocytosis, for example via stresses in the plasma membrane or by presentation of luminal vesicle membrane proteins to the synaptic cleft. Alternatively, factors such as [Ca²⁺]_o and [Ca²⁺]_i or other signalling mechanisms may combine in order to co-regulate exocytosis and endocytosis concomitantly.

Recently, we observed a curious feature of snake (and perhaps other) motor terminals, namely that decoating of clathrin-coated vesicles is blocked at temperatures below ~7 °C. Thus, after brief stimulation in a cold bath containing the endocytotic probe FM1-43, all FM1-43-labelled internalized vesicles exhibited clathrin coats (Teng & Wilkinson, 2000). Moreover, the coated vesicles remained near the plasma membrane, suggesting that downstream events, such as movement towards the vesicle pool, required completion of the decoating step. Here we demonstrate that a key event upstream of decoating is also impeded: about one-half of the compensatory endocytosis expected after a brief stimulation is blocked when the preparation is kept cold. This endocytotic 'debt' can remain for hours, but is relieved by a temperature step to room temperature (RT) or above. By manipulating the cooled preparation prior to or during the temperature step, delayed endocytosis can be made to resume under conditions different from those that existed during the period of transmitter release. Using this new technique, we have studied the dependence of clathrin-mediated endocytosis on temperature and the level of extracellular Ca²⁺. In a normal bath and at RT, endocytosis was rapid, being complete < 1 min after the temperature step to RT. As expected, the rate of endocytosis exhibited a positive temperature coefficient. The rate of endocytosis also increased markedly with increasing [Ca²⁺]_o over the range 0-7.2 mM. Part of this work had been described in abstract form (Teng & Wilkinson, 2001).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Garter snakes (Thamnophis sirtalis) were anaesthetized with pentobarbital sodium (80 mg kg^-1, I.P.) and killed by rapid decapitation. Several contiguous segments of the single-fibre-thick transversus abdominis muscle were dissected from the animal, placed in reptilian saline solution, and divided as needed to provide two to eight individual (and nearly identical) three-segment, nerve-muscle preparations. Details of the muscle's anatomy, dissection procedure and composition of the physiological saline (Ringer solution) are described elsewhere (Wilkinson & Lichtman, 1985). All procedures followed the Washington University Guidelines for Animal Studies.

Electrical stimulation, intracellular recording and activity-dependent staining

Nerve-muscle preparations were mounted in a temperature-controlled chamber on the stage of an inverted microscope. The cut end of the muscle nerve was placed in a hook-in-oil electrode for stimulation with 200 µs negative-going rectangular pulses. The supramaximal stimulus amplitude (2-7 V) was determined by visual observation of muscle contraction. The centre muscle in each preparation was stimulated, with the two adjacent muscles serving as unstimulated controls for activity-dependent staining, as described below. Intracellular recording of endplate potentials (EPPs) in curarized (30 µM) preparations and subsequent analyses were performed using methods described elsewhere (Wilkinson et al. 1992); endplate sites were visualized using differential interference contrast optics. Sulphorhodamine 101 (SR101, 160 µg ml^-1; Lichtman et al. 1985; Teng et al. 1999) was the endocytotic probe. The dye is amphiphilic due to its charged sulphonic acid moieties and, particularly in reptiles, resembles FM1-43 in its activity-dependent staining properties (Betz & Bewick, 1992). However, unlike FM1-43, it rinses easily from fixed or partially fixed plasma membranes. It was therefore possible to terminate each dye uptake period at a precise time by rapid exchange of the SR101-containing solution with fixative (1 % formaldehyde). SR101-containing bath solutions were also exchanged with SR101-free solutions at various times during experiments, as described in Results (exchange time < 1 s).

Measurement of endocytosis and the rate of exocytosis

The fluorescence of endocytosed SR101 was quantified in fixed, slide-mounted preparations. Two confocal microscopes were used, a Zeiss LSM-510 and a Bio-Rad MRC 1024 (details of fixation and microscopy are given in Teng et al. 1999). Images of nerve terminals were reconstructed as projections of 9-35 separate optical sections obtained at 400-700 nm steps in the z direction. Individual boutons in each reconstructed image were outlined with 'region of interest' software (Scion Image for Windows, Scion Corp., Frederick, MD, USA; www.scioncorp.com) and the average fluorescence of pixels within the bouton noted. Boutons from four to eight terminals (21-443 boutons in total; each twitch terminal comprises on average 58 boutons) were assayed in each preparation. Background staining, the average staining intensity of the endplate regions located between boutons at a neuromuscular junction (NMJ), was subtracted from the average measurement of bouton staining intensity at that NMJ. All preparations also included an unstimulated control muscle to assess possible SR101 uptake due to constitutive and spontaneous endocytosis, including the endocytosis associated with spontaneous transmitter release. The staining intensity of boutons in the unstimulated muscle was either undetectable or only barely detectable, averaging about 17 % of the background fluorescence described above. This second type of background was ignored in all but those experiments involving short incubation periods (5-15 s; see Fig. 4 and below). Here, stimulated boutons were themselves weakly stained, and both types of background were subtracted. Total endocytosis over a time of incubation with SR101 was taken as the mean SR101 staining intensity determined in this manner (expressed as eight-bit arbitrary pixel brightness units, ABU, 0-255). The rate of endocytosis at a particular time was measured by adding the probe for a precisely timed brief period (e.g. 5 s; starting 2.5 s before and ending 2.5 s after the desired time point) and dividing ABU by that time period. The rate is expressed as ABU s^-1. Because fixed tissue was used, each data point in experiments aimed at measuring the rate of endocytosis over time required a separate nerve-muscle preparation. Care was taken to use adjacent muscles from the same snake and to hold all adjustments of the confocal microscope constant throughout such experiments.

Bath temperature (monitored with a calibrated thermistor) was changed between 7 °C, RT (21-24 °C) and 35 °C by rapid exchange with a solution of the desired temperature. Bath composition was changed in the same manner; Ca²⁺ concentrations of ~0 (no added Ca²⁺), ~0 with 10 mM EGTA, 3.6 mM (normal [Ca²⁺]) and 7.2 mM were accompanied by equimolar changes in Na⁺ concentration so that osmolarity was held nearly constant. All baths contained 1.8 mM Mg²⁺. Separate experiments were performed to ensure that the inherent brightness of SR101 and the dye's concentration within internalized vesicles were not themselves significantly influenced by temperature or [Ca²⁺]_o over the range of these variables studied (see Results). Any change in brightness or concentration of the dye would be indistinguishable from changes in the rate of endocytosis.

Measurement of [Ca²⁺]_i

Intracellular Ca²⁺ levels were measured to assess whether changes in [Ca²⁺]_o were accompanied by detectable changes in [Ca²⁺]_i. With the preparation pinned out in a dish, fura-2 (pentapotassium salt, 50 mM; Molecular Probes, Eugene OR, USA) was added to a small well (0.6 µl) formed from silicone sealant and located adjacent to the muscle bath. The freshly cut end of the muscle nerve was placed into the well, which was then covered with petroleum jelly to prevent drying. The preparation was incubated for 4 h (RT), then rinsed with saline and refrigerated (5 °C) overnight to permit diffusion and equilibration of the indicator into the nerve terminals (see Wu & Betz, 1996). Ratiometric imaging was by standard methods using a Zeiss inverted microscope equipped with an Intelligent Imaging Innovations (Denver, CO, USA) digital camera system and software (Wurth & Zweifach, 2002). Image pairs (340 nm and 380 nm excitation wavelengths) were acquired with a SensiCam digital camera (Cooke, Tonawanda, NY, USA). Pixels were binned (4 4). Image pairs were obtained at intervals of 4-6 s. Ratio images were displayed on a computer monitor and regions of individual boutons were outlined manually. The fura-2 dissociation constant at a particular temperature was taken as that given by Larsson et al. (1999). Background fluorescence was averaged from three regions of the muscle fibre surrounding the terminals, and subtracted. The corrected image pair ratios were then converted to [Ca²⁺]_i using equations given by Grynkiewiez et al. (1985).

Unless indicated otherwise, data are presented as means ± S.D.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Delayed endocytosis

Our initial evidence for low-temperature partial blockade of endocytosis was that with brief stimulation in the presence of SR101 or FM1-43 at 7 °C, nerve terminals took up less dye than those stimulated at 7 °C but then warmed to RT with the dye still present. Compensatory endocytosis was therefore incomplete at 7 °C, and there remained endocytotic debt. Subsequent studies of this phenomenon are described in Fig. 1 and Fig. 2. Figure 1 shows the time course of endocytosis in snake motor terminals at three temperatures: 7 °C (Fig. 1a), 23 °C (RT; Fig. 1b) and 35 °C (Fig. 1c). In all experiments, the preparation was briefly stimulated (5 Hz for 30 s) at the indicated temperature and then held at that temperature until fixation. In the left panel of each sequence, SR101 was added to the bath during stimulation (30 s) and the preparation was immediately fixed and prepared for imaging. As can be seen by comparing staining intensities, endocytosis during the stimulation period became far more robust as the temperature increased. We confirmed that this marked increase in the rate of endocytosis was not simply due to increased exocytosis. Transmitter release, estimated electrophysiologically in separate experiments under the same conditions (5 Hz, 30 s stimulation with partial curare block), was not strongly influenced by temperature. With the mean endplate potential (EPP) amplitude at 7 °C (8.3 ± 3.7 mV, n = 8 fibres from two snakes) taken as 100 %, the mean EPP amplitude recorded from the same fibres was 129 ± 62 % at RT and 97 ± 26 % at 35 °C. Neither change was significant (P > 0.26; see also Adams, 1987, 1989). When SR101 was added at a time subsequent to stimulation (middle and right panels in Fig. 1a-c), a different picture emerged. At 7 °C (Fig. 1a) the rate of endocytosis was quite slow initially, and continued with only a slight decrement for the duration of the experiment. Thus, dye uptake sampled 60-80 s after stimulation (right) was about as robust as during (left) or immediately after stimulation (centre). At higher temperatures, however (Fig. 1b-c), dye uptake diminished with time. The dye intensity measurements are plotted in Fig. 1d. Endocytosis was nearly complete in ~20 s at 35 °C, and in ~50 s at 23 °C. In all experiments carried out using the same protocol as for Fig. 1, dye uptake was no longer measurable (using a 5-30 s incubation period) after 19 ± 4 s (n = 5) at 35 °C and after 38 ± 12 s (n = 6) at RT (21-24 °C). At 7 °C, dye uptake was weak, but there was no indication of completion of endocytosis. Thus, while the rate of endocytosis approximately doubled as temperature was increased to ~14 °C above RT, a similar decrease in temperature slowed endocytosis by a much larger factor (> 100, see below).

		View larger version [in this window] [in a new window]
Figure 1. The rate of endocytosis in motor terminals at three different temperatures Each micrograph here and in Figs 2 and 3 shows a small region (a few boutons) of one terminal; punctate SR101 hot spots are clusters of internalized vesicles near active zones. Times on micrographs (and plot below) correspond to the middle of the dye uptake period, with 0 s indicating the end of stimulation. a, dye uptake at 7 °C was weak during stimulation, but continued only slightly diminished throughout the experiment. Shown are uptake during the 30 s, 5 Hz stimulation (0 s), during a 20 s period immediately after stimulation (10 s) and during a 20 s period beginning 60 s after stimulation (70 s). b, uptake at 23 °C was moderate during stimulation, but diminished monotonically with time. c, uptake at 35 °C was intense during stimulation, but diminished rapidly and was nearly complete by 25 s. d, plot of SR101 uptake rates at three different temperatures; each data point is averaged from the staining intensity of six to eight terminals (two snakes). The most brightly stained boutons here and in Fig. 2 are shown slightly saturated so that relative staining intensity among all panels can be appreciated. Error bars are S.D.

		View larger version [in this window] [in a new window]
Figure 2. Delayed endocytosis invoked by a temperature pulse to RT a-e, SR101 uptake in terminals from five muscles cold stimulated for 60 s at 5 Hz. a, uptake at 6-7 °C during stimulation and for 90 s after it (Stim.). b-e, uptake during a temperature pulse to RT beginning at 90 s (b), 30 min (c), 60 min (d) and 120 min (e) after stimulation. The uptake period (60 s) at RT was long enough to ensure completion of endocytosis at all delay times. f, mean staining intensity of 73-134 boutons (five to eight terminals) in each of the preparations of a-e. The endocytotic 'debt' recovered by the RT pulse declined slowly with time due to continued slow endocytosis at 6-7 °C. g, averaged results from four experiments using the same protocols as above. Shown also is dye uptake representing the total compensatory endocytosis that occurred after the 5 Hz, 60 s stimulation (measured in separate experiments and normalized to 100 %; see text). Uptake during the cold stimulus averaged about one-half of the total, with the other half recoverable as delayed endocytosis via a pulse to RT 60-90 s after stimulation (left-most circle). Recoverable endocytosis declined slowly over time, presumably due to slow endocytosis in the cold. After 120 min, the endocytotic debt recoverable by an RT pulse was about one-fifth of the total. Error bars are S.D.

Indeed, we found that endocytosis at 7 °C could continue for > 2 h, slowly diminishing but not erasing the endocytotic debt incurred by stimulation. However, the debt remaining at any time could be erased via a brief (~1 min) elevation of temperature to RT or above. These observations are shown in Fig. 2. The micrographs in Fig. 2a-e illustrate results from an experiment in which five preparations from one snake were identically cold stimulated (5 Hz, 60 s, 6-7 °C). Figure 2a shows a few terminal boutons from a preparation that was stimulated in the presence of SR101. The probe remained in the cold bath for an additional 90 s after stimulation, at which time the preparation was fixed and prepared for imaging. In Fig. 2b, SR101 was absent during stimulation and for 90 s thereafter, but was added at 90 s when the bath was exchanged for one at RT. The probe remained for 60 s before fixation, a time sufficient for virtually all compensatory endocytosis to be completed at RT (Fig. 1). Figure 2a and b shows roughly the same staining intensity, suggesting that about one-half of total endocytosis was completed during 150 s at cold temperature, with the other half delayed until the preparation was warmed. Figure 2c-e shows the staining intensity observed using the same protocol as in Fig. 2b, except that successively longer times (up to 2 h) elapsed before the preparation was warmed. The mean SR101 staining intensities (in ABU; see Methods) from the experiment of Fig. 2a-e are plotted in Fig. 2f. Consistent with the experiments represented in Fig. 1, endocytosis continued slowly at 7 °C (see Fig. 1d), so that as time passed, the amount of delayed endocytosis evoked by warming the preparation diminished. Even after 2 h, however, a significant fraction of the total endocytosis needed to compensate for transmitter release remained incomplete and was recovered by the temperature pulse to RT. A summary of results from several experiments using protocols similar to those described above is given in Fig. 2g. Dye uptake is plotted as a percentage of the maximum possible. The mean brightness corresponding to maximal uptake is expressed as 100 % on the ordinate and was measured in separate experiments by adding SR101 both during stimulation (30 s at 7 °C) and for a sufficiently long period (60 s) at RT immediately after to assure that endocytosis was complete (square symbol). The remaining symbols in Fig. 2g correspond to data from the single experiment of Fig. 2f. Average dye uptake in preparations held at 7 °C during and just after stimulation was about 50 % of the total possible, leaving an endocytotic debt of about 50 %. This could be recovered by warming the preparation soon after stimulation (leftmost circle). With the preparation held at 7 °C for longer times, endocytosis continued slowly, so that the dye uptake initiated by a temperature pulse diminished progressively with time. However, about 20 % of the initial endocytotic debt remained at 120 min, the longest delay time examined.

Dependence on [Ca²⁺]_o

We developed a technique utilizing delayed endocytosis to test whether the rate of endocytosis is sensitive to the concentration of bath Ca²⁺. Preparations were stimulated (5 Hz, 60 s) in a normal Ca²⁺ bath (3.6 mM) at 6-7 °C. To decrease the rate of spontaneous transmitter release (and consequent endocytosis) following stimulation, the bath was exchanged for one containing no added Ca²⁺ and 10 mM EGTA, still at 6-7 °C. After 90 min, delayed endocytosis was initiated by replacing the cold bath with one containing SR101 and a particular [Ca²⁺]_o at RT. Incubation was for 1 min, sufficient time for endocytosis to be completed in normal bath Ca²⁺ (Fig. 1d and additional experiments; data not shown). The preparation was then fixed. Results are shown in Fig. 3. There was no qualitative difference in the punctate distribution of SR101 uptake at various levels of [Ca²⁺]_o. In contrast, the quantity of SR101 taken up in 1 min depended systematically on [Ca²⁺]_o (filled bars in Fig. 3; statistics in Fig. 3 legend). In a bath containing no added Ca²⁺, the staining intensity was only 49 ± 6 % of that in normal (3.6 mM) Ca²⁺ (100 %). The staining intensity was the same (51 ± 2 %) in a Ca²⁺-free bath containing 10 mM EGTA (with normal 1.8 mM Mg²⁺). At 1.8 mM bath Ca²⁺, the staining intensity was 70 ± 10 % of that in normal [Ca²⁺]_o. We then increased [Ca²⁺]_o to 7.2 mM (twice normal) and, as expected, saw no significant increase (104 ± 14 %) in SR101 staining intensity compared to that observed with 3.6 mM. Since endocytosis at 3.6 mM was complete within the time allotted, the protocol could not be used to measure whether or not 7.2 mM Ca²⁺ produced a still-higher rate of endocytosis. We therefore decreased the incubation time from 60 to 25 s in order to assess dye uptake at 7.2 mM Ca²⁺ (unfilled bars in Fig. 3). With uptake at 3.6 mM once again normalized to 100 %, uptake at 7.2 mM was 135 ± 18 %. Thus, the rate of endocytosis more than doubled (from 49 % of normal to 135 % of normal) over the range 0-7.2 mM bath Ca²⁺.

		View larger version [in this window] [in a new window]
Figure 3. Delayed endocytosis depends on bath Ca2+ a-d, micrographs showing dye uptake in four preparations that were stimulated identically (5 Hz, 60 s) and kept cold. After 90 min, preparations were incubated in SR101 at RT for 1 min, a time just sufficient for the completion of delayed endocytosis in 3.6 mM [Ca2+]o. Dye uptake increased with increasing [Ca2+]o up to 3.6 mM (P < 0.02); further increase to 7.2 mM [Ca2+]o did not significantly increase dye uptake (P > 0.34). e, bar graph (filled bars) showing dependence of uptake on [Ca2+]o; uptake at 3.6 mM [Ca2+]o is normalized to 100 %. Addition of EGTA to the zero-Ca2+ bath had no significant effect (leftmost bar; P > 0.47). In other experiments, the incubation time at RT was reduced (25 s) to test the [Ca2+]o dependence at higher concentrations (open bars; P < 0.04); uptake at 3.6 mM is again shown normalized to 100 %. The rate of endocytosis more than doubled over the [Ca2+]o range 0-7.2 mM. Error bars are S.D. Results and significance levels are from 241-389 boutons (three to four snakes) at each Ca2+ concentration.

The experiments shown in Fig. 3 utilized relatively long incubation times for dye uptake. The endocytotic signal was integrated over time, providing well-stained preparations with relatively little variability. We next attempted to complement this approach by examining the time course of endocytosis at different values of [Ca²⁺]_o in more detail. Several nerve-muscle preparations were stimulated identically (5 Hz, 60 s, 6-7 °C) and kept cold. After 15 min the bath was exchanged for one at RT and containing 0 (no added Ca²⁺), 3.6 or 7.2 mM Ca²⁺. To estimate the rate of endocytosis immediately after the temperature jump, SR101 was included in the RT bath. To estimate the instantaneous rate at later time points (e.g. after 5, 10, 20 or 40 s), the RT bath was once again exchanged for one that contained SR101 with the same Ca²⁺ level and at the same (room) temperature. Incubation in SR101 was as short as possible (determined in separate experiments; usually 5 s), but was longer (up to 15 s) at some 20 and 40 s time points, when endocytosis was found to be nearly complete and the rate of endocytosis was extremely low. Each preparation was then immediately fixed. The dye uptake during a particular ~5 s sampling interval was taken as representing, approximately, the instantaneous rate of endocytosis at the time midway in that interval (e.g. 2.5, 7.5, 12.5, 22.5 or 42.5 s after the temperature jump to RT). Figure 4a-c shows results from three individual experiments with [Ca²⁺]_o values of 0, 3.6 and 7.2 mM, respectively. Each data point represents the staining intensity averaged from 21-221 terminal boutons in one muscle preparation. Five to seven muscle preparations from one snake were used in each of the experiments. The rate of endocytosis was highest in the first test period for all three levels of [Ca²⁺]_o (0-5 s after the jump to RT, plotted as 2.5 s). This rate (calculated as ABU s^-1; see Methods) is shown normalized to 1.0 to allow comparison of all examples in Fig. 4. The rate of endocytosis declined along an approximate single-exponential time course (best fit shown by continuous lines) for all levels of [Ca²⁺]_o. The rate of decline varied systematically, being about twice as rapid (time constant, 6.5 s) at 7.2 as at 0 mM Ca²⁺ (time constant, 11.5 s; curves from Fig. 4a-c shown superimposed in Fig. 4d). Thus, consistent with experiments carried out using longer incubation times (Fig. 3), compensatory endocytosis was completed more rapidly as [Ca²⁺]_o increased. The averaged results from all experiments with protocols similar to that used to obtain the data presented in Fig. 4a-c are shown in Fig. 4e. The single-exponential time constants (from best-fit curves as in Fig. 4a-c) decreased by a factor of approximately two as [Ca²⁺]_o increased over the three levels studied (Fig. 4f and Fig. 4 legend).

		View larger version [in this window] [in a new window]
Figure 4. Rapid time course of delayed endocytosis at three bath Ca2+ concentrations a-c, SR101 uptake rates in preparations that were stimulated (5 Hz, 60 s, 6-7 °C) and kept cold in normal Ringer solution (3.6 mM Ca2+) for 15 min before a change (0 s on plots) to RT and the desired [Ca2+]o. For each data point, SR101 was added to the RT bath for a 5-15 s period, whose midpoint is indicated by the data point's value on the abscissa (each point is one preparation; all preparations in one plot are from the same snake). The fastest rates of endocytosis were from the earliest incubations after the temperature change (initial data point, 2.5 s) and are shown normalized to 1.0. Rates declined exponentially at all [Ca2+]o (best fits indicated by continuous lines; the time constants are: a, 11.5 s; b, 7.7 s; c 6.5 s). d, superposition of best-fit exponentials from a-c. e, best-fit exponential curves from all experiments (curve a, four experiments at 0 mM Ca2+; curve b, three experiments at 3.6 mM Ca2+; curve c, three experiments at 7.2 mM Ca2+). In each experiment, five to seven preparations were made from one snake. f, range of time constants seen in all experiments (same preparations as in e); data are means ± S.E.M. Time constants were scattered, but on average decreased (rates increased) with increasing bath Ca2+ (P < 0.08). The examples in a-c are of the slowest rate of endocytosis observed at each Ca2+ level.

In the experiments represented in Fig. 3 and Fig. 4 we assumed that the staining intensity seen at the light level was proportional to the number of vesicles endocytosed. However, Ca²⁺ might affect staining directly. In particular, the activity, or effective concentration, of SR101 near the presynaptic membrane could depend on [Ca²⁺]_o because Ca²⁺ is a counterion that is capable of modifying the interaction between SR101 and the membrane surface charge. If true, staining might reflect not only the number of endocytosed vesicles, but also the amount of SR101 loaded into each vesicle. Two observations from control experiments indicate that this was not the case. First, the faint staining of unstimulated nerve terminals did not depend significantly on [Ca²⁺]_o. The brightness was 22.6 ± 1.2 ABU in 0 mM [Ca²⁺]_o and 21.2 ± 5.2 ABU in 7.2 mM [Ca²⁺]_o (P > 0.78, n = 66 boutons from two snakes; 3 min incubation). Second, while the rate of endocytosis varied in a Ca²⁺-dependent manner (Fig. 4e), the maximum dye uptake obtainable was similar in all preparations, regardless of [Ca²⁺]_o (53.0 ± 6.3 ABU in 0 mM [Ca²⁺]_o; 54.6 ± 16.4 ABU in 7.2 mM [Ca²⁺]_o; P > 0.72, n = 370 boutons from three snakes; 3 min dye incubation). Since preparations were stimulated identically (and therefore in need of identical compensatory endocytosis), we expected this result only if the staining intensity depended exclusively upon the number of vesicles internalized. These same observations may also rule out a second artefact, namely that [Ca²⁺]_o during incubation influenced the quantum yield of SR101 when it was subsequently imaged in fixed tissue.

Invariance of [Ca²⁺]_i

Because [Ca²⁺]_i is a putative regulator of the rate of endocytosis (see Discussion), the apparent dependence upon [Ca²⁺]_o might instead reflect the influence of [Ca²⁺]_i. We measured the [Ca²⁺]_i increase in boutons during and following 5 Hz, 30 s stimulation to determine whether [Ca²⁺]_i remained elevated for long periods after cold stimulation (Fig. 5). This would not influence the dependence of delayed endocytosis on [Ca²⁺]_o, but might influence the rate of delayed endocytosis generally. In experiments at RT (n = 9) and at 7 °C (n = 8), baseline [Ca²⁺]_i (~50 nM) rose rapidly to 200-250 nM, as expected during the stimulus train, then decayed to baseline (Fig. 5e). Decay times were adequately fitted with double exponentials (data not shown). Cooling had no measurable effect on the faster time constant (~1 s) but, as shown in Fig. 5f, increased the slower one significantly (82 ± 46 s; n = 8 at 7 °C versus 14.8 ± 6.1 s, n = 9 at RT; P < 0.005). Despite this slowing, however, the return to baseline at 7 °C was complete in 173 ± 91 s, well before the 15 or 90 min time points we employed to assay delayed endocytosis (the record with the slowest return to baseline at 7 °C is that shown in Fig. 5e). Thus, [Ca²⁺]_i was not measurably elevated due to previous stimulation at the time that delayed endocytosis was assayed.

		View larger version [in this window] [in a new window]
Figure 5. Stimulus-induced intraterminal Ca2+ transients returned to baseline before measurements of the rate of endocytosis a-d, fura-2 images of typical terminals at 7 °C (left) and RT (right); a table relating Ca2+ concentrations to colours in the ratio images is given at the far right. a, images of a resting terminal prior to stimulation. The 380 nm image was displayed on the screen of a computer and boutons were outlined manually with a computer mouse (mask) for subsequent analysis. The 380 and 340 nm raw images are shown for 7 °C data only. b, fura-2 images during stimulation. c-d, fura-2 images after stimulation. e, time course of [Ca2+]i transients; each data point is from one ratio image at the time indicated. The RT record is typical; the record at 7 °C exhibits the slowest decay observed among eight preparations. f, average time constants of exponential decay curves fitted to the decay phase of Ca2+ transients (see text); error bars are S.D. (RT, n = 9; 7 °C, n = 8).

[Ca²⁺]_i could also be influenced indirectly by changes in [Ca²⁺]_o. To test this possibility we measured [Ca²⁺]_i in terminals while they were subjected to conditions similar to those used to obtain the data shown in Fig. 3 and Fig. 4. Preparations with fura-2-loaded terminals were incubated in Ringer solution containing 1.8, 3.6, 7.2 or 21.6 mM Ca²⁺ for 30 min. The bath was then changed to a different Ca²⁺ concentration for 30 min, then back to the first for 30 min. We measured [Ca²⁺]_i by fura-2 ratio imaging immediately before and immediately after solution changes, plus every 10 min in a particular solution. Each measurement was the average of indicated [Ca²⁺]_i from ratios of 10 image pairs obtained over ~1 min. While measurements of [Ca²⁺]_i varied among preparations (range, 25-150 nM), we detected no systematic dependence upon [Ca²⁺]_o. These results are summarized in Fig. 6. In a single terminal (example, Fig. 6a-d) there was some scatter among repeated measurements, but there was no discernable change in [Ca²⁺]_i after changing [Ca²⁺]_o (Fig. 6e). Figure 6f summarizes the results from six experiments (70 solution changes) similar to that of Fig. 1e. Increasing or decreasing [Ca²⁺]_o by factors ranging from two to more than 10 did not change [Ca²⁺]_i. Results were similar at RT (left, four experiments) and in preparations cooled to 7 °C (right, two experiments). One can appreciate the stability of [Ca²⁺]_i by noting the insignificant trend of the regression lines in Fig. 1f as compared to the overall scatter in fura-2 measurements.

		View larger version [in this window] [in a new window]
Figure 6. Intraterminal Ca2+ was unaffected by changes in bath Ca2+ a-d, fura-2 images of a typical terminal. a, 380 nm excitation; mask on boutons as in Fig. 5. b, image at 340 nm excitation. c, ratio image with 21.6 mM bath Ca2+ for 30 min. d, ratio image 1 min after change to 1.8 mM bath Ca2+; see Fig. 5 for colour table. e, typical measurements of indicated [Ca2+]i values before and after a step in [Ca2+]o from 21.6 mM (30 min; last 1 min shown) to 1.8 mM (30 min). Each data point is an average of the indicated [Ca2+] obtained from a masked ratio image of a terminal as in c-d above. f, summary of results from various [Ca2+]o steps as in e. Each data point compares indicated [Ca2+]i averaged from measurements of 10 terminals 1 min before and 1 min after a change in [Ca2+]o. The logarithmic abscissa is a ratio of both decreasing (< 1) and increasing (> 1) steps in bath Ca2+ of concentrations ranging from 1.8 to 21.6 mM. Corresponding changes in intraterminal Ca2+ (expressed as a percentage) varied among experiments but showed no pattern at either RT or 7 °C (P > 0.6). The scatter in data, measured before and after the step in [Ca2+]o, was similar to the scatter in repeated measurements at the same [Ca2+]o (example in e). The slope of the regression lines is near to zero.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Temporal isolation of endocytosis by low temperature

The mechanism underlying delayed endocytosis is low-temperature blockade of clathrin decoating (Illinger et al. 1991; Teng et al. 1999; Teng & Wilkinson, 2000; see also Morgan et al. 2001). In previous work we stimulated snake nerve-muscle preparations in the presence of FM1-43 and kept them cold until fixation. With the aid of electron microscopy (EM) we observed clusters of internalized vesicles located just within (< 300 nm) the presynaptic membrane near active zones. All such labelled vesicles had clathrin coats (Teng & Wilkinson, 2000). In contrast, stimulated snake motor terminals held at RT until fixation contained very few, if any, coated vesicles, in agreement with previous EM studies of endocytosis in RT preparations (Heuser, 1989). Thus, cooling appears to block the decoating of internalized vesicles, thereby impeding endocytosis. Our preparation therefore resembles the Drosophila shibire mutant, which contains a temperature-sensitive version of the clathrin-associated protein dynamin (Ramaswami et al. 1994; Koenig & Ikeda, 1996; Kuromi & Kidokoro, 1998). As is true for the shibire model, where endocytosis is delayed until a transition to the permissive temperature, delayed endocytosis in snake should serve as a useful tool for the study of various putative regulators of vesicle processing. Examples include neurotransmitters and neuromodulators (acetylcholine and adenosine compounds) that have been implicated in the putative feedback regulation of neural activity, and intraterminal Ca²⁺ (Rochon et al. 2001; see below).

The relatively fast rates of endocytosis measured by our method argue that rapid retrieval is possible via a clathrin-dependent mechanism. In experiments where endocytosis resumed at 3.6 mM [Ca²⁺]_o and RT, virtually all dye uptake was complete in ~1 min (Figs. 3 and 4). Elevation of the temperature to 35 °C approximately doubled the rate of endocytosis compared to RT; we did not study the temperature dependence of delayed endocytosis in detail. The time course of delayed endocytosis at RT was approximated by a single-exponential decay, with time constants ranging from 2.1 to 7.7 s at 3.6 mM [Ca²⁺]_o. This scatter probably reflects both biological variability and measurement error. To increase temporal resolution, we chose short periods for dye uptake (5-15 s; so that measurements at later time points were just above background) and therefore probably failed to detect some of the uptake. Thus (particularly at 0 mM bath Ca²⁺; see below), we observed dye uptake ~1 min after the step to RT when SR101 was allowed to remain continuously in the bath (experiments represented in Fig. 3), whereas our time-constant measurements (Fig. 4) suggest that endocytosis should have been complete in < 1 min. We saw no evidence in delayed endocytosis of the two-exponential-component process that has been reported in other preparations (von Gersdorff & Matthews, 1994a; Wu & Betz, 1996; Neves & Lagnado, 1999; Heidelberger et al. 2002), although the temporal resolution might have been insufficient for us to do so. It is also possible that the substantial endocytosis not delayed by temperature block represents one kinetic component, and delayed endocytosis the other (see below).

We note also that any putative steps in the vesicle retrieval process that occur before actual membrane fission could be initiated or completed at cold temperature before the pulse to RT. Endocytotic times measured with delayed endocytosis might therefore underestimate the total time required. However, our results agree with the recent findings of Sun et al. (2002), who observed directly endocytotic time constants of 115 ms to tens of seconds, depending upon the stimulus rate, in the rat calyx of Held. Richards et al. (2000) also provide evidence for rapid recycling (< 2 min) at frog NMJs. While the endocytotic mechanisms in these studies were not identified, we argue, based on the present work, that the retrieval rates observed are consistent with a clathrin-dependent process and that putative faster mechanisms, such as kiss-and-run transmitter release, need not be invoked.

Only about one-half of the total compensatory endocytosis was blocked by low temperature and therefore 'delayed' (Fig. 2g). The rest occurred during and immediately after stimulation. Heidelberger et al. (2002) recently described a technique whereby about one-half of endocytosis was similarly delayed for several minutes. Their method was to transiently increase hydrostatic pressure in patch-clamped retinal bipolar terminals. Upon return to permissive pressure, endocytosis resumed and exhibited a single exponential time constant (~40 s). Interestingly, the undelayed portion of endocytosis that was insensitive to pressure exhibited a significantly faster (~3 s) time constant. Thus, the two known kinetically distinct components of endocytosis in bipolar terminals (von Gersdorff & Matthews, 1994a; Rouze & Schwartz, 1998; Neves & Lagnado, 1999) were separated in time. In the present work, snake motor terminals also exhibited delayed endocytosis with single-exponential kinetics similar to those seen by Heidelberger et al. (2002), suggesting that the earlier round of endocytosis that was insensitive to cooling in snake was also kinetically distinct. Although membrane capacitance measurements have not been performed on snake terminals, we have observed two functionally distinct components of endocytosis using optical probes: macropinocytosis and clathrin-mediated endocytosis (Teng et al. 1999; Teng & Wilkinson, 2000; see also Koenig & Ikeda, 1996; Richards et al. 2000; Holt et al. 2003). Since delayed endocytosis in snake is clathrin-mediated, we suggest that the mode of endocytosis that is observed to be insensitive to cooling in this animal (and perhaps to pressure in bipolar terminals) is macropinocytosis.

Negative regulation of the rate of endocytosis by [Ca²⁺]_o

The total endocytotic debt brought about by cold stimulation was erased by a temperature increase to RT at all levels of bath Ca²⁺. However, the rate of this delayed endocytosis, assessed either as total dye uptake in a fixed period or as the approximate time constant of uptake, increased markedly with increasing [Ca²⁺]_o. Physiologically, systemic levels of Ca²⁺ are tightly regulated in vertebrates. However, Ca²⁺ within the narrow synaptic cleft is isolated by a diffusion barrier and probably decreases with increasing rates of transmitter release, due to the opening of Ca²⁺-permeable channels in both the pre- and postsynaptic membranes (Nicholson, 1980; Vassilev et al. 1997; Stanley, 2000). Such a decrease in cleft Ca²⁺ should, according to our observations, decrease the rate of endocytosis. Since the rate of transmitter release is ultimately dependent upon the number of recycled vesicles available in various pools, the effect of this putative mechanism would be a type of negative feedback. In particular, the maximal rate of release would be self-limiting, thereby providing a mechanism of synaptic depression (see Borst & Sakmann, 1999). A similar negative feedback mechanism has been reported for other synapses. In vertebrate retinal bipolar terminals, the increase in [Ca²⁺]_i associated with activity was found to inhibit endocytosis (Von Gersdorff & Matthews, 1994b). In the rat calyx of Held, exocytosis itself was inferred to inhibit endocytosis because endocytosis was most rapid after a single stimulus and slowest after a tetanus (Sun et al. 2002). However, because protocols in these studies did not examine endocytosis in isolation, it is possible that activity-dependent diminution of cleft Ca²⁺ was responsible, at least in part, for the diminution of the rate of endocytosis seen.

The regulation of endocytosis by [Ca²⁺]_o could be direct or, alternatively, mediated by some mechanism that is itself influenced by [Ca²⁺]_o. In either case, the location and nature of the [Ca²⁺]_o sensor are unknown. During neurotransmitter release, compensatory endocytosis presents the luminal domains of vesicular membrane proteins to the cleft. At least one vesicle membrane protein is known to undergo interactions with cleft proteins (Son et al. 2000). A hypothesis consistent with these observations is that some presynaptic membrane [Ca²⁺]_o sensor, perhaps a membrane protein of exocytosed vesicles, influences the rate of subsequent endocytosis.

Neither [Ca²⁺]_o nor Ca²⁺ influx is required for delayed endocytosis

Delayed endocytosis was slowed but not abolished in a Ca²⁺-free bath, with or without 10 mM EGTA. Bath Ca²⁺ is also not required for endocytosis at the larval Drosophila NMJ (measured by exploiting delayed endocytosis in shibire; see above; Ramaswami et al. 1994). In rat hippocampal terminals, endocytosis not only persists in a zero-Ca²⁺ bath, but its rate is reported to be independent of [Ca²⁺]_o over a wide range (0-10 mM; Ryan et al. 1993, 1996). However, Ryan et al. (1996) measured FM1-43 uptake during a 1 min period after the delivery of only 20 stimuli. This time might have sufficed for a terminal to complete endocytosis at any Ca²⁺ concentration (as was almost the case in snake), and dependence on [Ca²⁺]_o, if present, would not have been observed. In contrast, bath Ca²⁺ seems to be required for endocytosis when transmitter release is induced by drugs such as -latrotoxin (frog NMJ; Ceccarelli & Hurlbut, 1980; Henkel & Betz, 1995) or Ba²⁺ (rat synaptosomes; Cousin & Robinson, 1998). One possible explanation for this disparity is that the mode of endocytosis occurring during stimulation requires [Ca²⁺]_o, while the mode blocked by low temperature (or high temperature or pressure in shibire, and bipolar terminals, respectively) does not (see also Ramaswami et al. 1994). The dependence of endocytosis on [Ca²⁺]_o might also be species dependent or synapse specific.

Intraterminal Ca²⁺ has been studied as a possible regulator of compensatory endocytosis because of its key role in exocytosis. Our experiments were performed under conditions where [Ca²⁺]_i remained constant. There was no neural activity for at least 15 min before RT endocytosis began, and we confirmed that [Ca²⁺]_i had returned to baseline levels in this time (Fig. 5). Moreover, in separate experiments we found that changing [Ca²⁺]_o over a wide range had no measurable effect on bulk [Ca²⁺]_i (Fig. 6; see Hodgkin & Keynes, 1957). Our imaging system would not have detected 'hot spots' of Ca²⁺ just within the terminal membrane that could result from transient inward flux through channels. However, we cannot envision how changes in [Ca²⁺]_o over a few millimolar could create such a flux. Thus, we conclude that the endocytotic process can be triggered in the absence of stimulus-mediated (or other) Ca²⁺ entry, an observation that is also supported by an EM study (Gad et al. 1998). Nevertheless, regulation of endocytosis by [Ca²⁺]_i, which we did not study, remains an important putative mechanism. Depending on conditions, [Ca²⁺]_i might upregulate endocytosis (Klingauf et al. 1998; Sankkaranarayanan & Ryan, 2000), downregulate it (Von Gersdorff & Matthews, 1994b), have no measurable effect, (Wu & Betz, 1996; Sun et al. 2002) or affect the slow and fast components differentially (Neves et al. 2001). Both [Ca²⁺]_i and [Ca²⁺]_o might contribute to the activity-dependent regulation of endocytosis, either in opposition or cooperatively.

	REFERENCES

Top Abstract Introduction Methods Results Discussion References

Adams BA, (1987). Thermal dependence of passive electrical properties of lizard muscle fibres. J Exp Biol 133, 169-182	[Abstract]
Adams BA, (1989). Temperature and synaptic efficacy in frog skeletal muscle. J Physiol 408, 443-455	[Abstract]
Betz WJ , & Bewick GS (1992). Optical analysis of synaptic vesicle recycling at the frog neuromuscular junction. Science 255, 200-203
Betz WJ, Bewick GS & Ridge RM (1992). Intracellular movements of fluorescently labeled synaptic vesicles in frog motor nerve terminals during nerve stimulation. Neuron 9, 805-813	[Medline]
Borst JG , & Sakmann B (1999). Depletion of calcium in the synaptic cleft of a calyx-type synapse in the rat brainstem. J Physiol 521, 123-133	[Abstract/Full Text]
Ceccarelli B , & Hurlbut WP (1980). Ca2+-dependent recycling of synaptic vesicles at the frog neuromuscular junction. J Cell Biol 87, 297-303	[Abstract]
Cousin MA , & Robinson PJ (1998). Ba2+ does not support synaptic vesicle retrieval in rat cerebrocortical synaptosomes. Neurosci Lett 253, 1-4	[Medline]
Gad H, Lö WP, Zotova E, Brodin L & Shupliakov O (1998). Dissociation between Ca2+-triggered synaptic vesicle exocytosis and clathrin-mediated endocytosis at a central synapse. Neuron 21, 607-616	[Medline]
Grynkiewicz G, Poenie M & Tsien RY (1985). A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem 260, 3440-3450	[Abstract]
Heidelberger R, Zhou Z-Y & Matthews G (2002). Multiple components of membrane retrieval in synaptic terminals revealed by changes in hydrostatic pressure. J Neurophysiol 88, 2509-2517	[Abstract/Full Text]
Henkel AW , & Betz WJ (1995). Monitoring of black widow spider venom (BWSV) induced exo- and endocytosis in living frog motor nerve terminals with FM1-43. Neuropharmacology 34, 1397-1406	[Medline]
Heuser J, (1989). The role of coated vesicles in recycling of synaptic vesicle membrane. Cell Biol Int Rep 13, 1063-1076	[Medline]
Hodgkin AL , & Keynes RD (1957). Movements of labelled calcium in squid giant axons. J Physiol 138, 253-281
Holt M, Cooke A, Wu MM & Lagnado L (2003). Bulk membrane retrieval in synaptic terminals of retinal bipolar cells. J Neurosci 23, 1329-1339	[Abstract/Full Text]
Illinger D, Poindron P & Kuhry J-G (1991). Fluid phase endocytosis investigated by fluorescence with trimethylamino-diphenylhexatriene in L929 cells: the influence of temperature and of cytoskeleton depolymerizing drugs. Biol Cell 73, 131-138	[Medline]
Klingauf J, Kavalali ET & Tsien RW (1988). Kinetics and regulation of fast endocytosis at hippocampal synapses. Nature 394, 581-585
Koenig JH , & Ikeda K (1996). Synaptic vesicles have two distinct recycling pathways. J Cell Biol 135, 797-808	[Abstract]
Kuromi H , & Kidokoro Y (1998). Two distinct pools of synaptic vesicles in single presynaptic boutons in a temperature-sensitive Drosophila mutant, shibire. Neuron 20, 917-925	[Medline]
Larsson D, Larsson B, Lundgren T & Sundell K (1999). The effect of pH and temperature on the dissociation constant for fura-2 and their effects on [Ca2+]i in enterocytes from a poikilothermic animal, Atlantic Cod (Gadus Morhua). Anal Biochem 273, 60-65	[Medline]
Lichtman JW, Wilkinson RS & Rich MM (1985). Multiple innervation of tonic endplates revealed by activity-dependent uptake of fluorescent probes. Nature 314, 357-359	[Medline]
Morgan JR, Prasad K, Jin S, Augustine GJ & Lafer EM (2001). Uncoating of clathrin-coated vesicles in presynaptic terminals: roles for Hsc70 and auxilin. Neuron 32, 289-300	[Medline]
Murthy VN , & Stevens CF (1998). Synaptic vesicles retain their identity through the endocytic cycle. Nature 392, 497-501	[Medline]
Neves G , & Lagnado L (1999). The kinetics of exocytosis and endocytosis in the synaptic terminal of goldfish retinal bipolar cells. J Physiol 515, 181-202	[Abstract/Full Text]
Neves G, Gomis A & Lagnado L (2001). Calcium influx selects the fast mode of endocytosis in the synaptic terminal of retinal bipolar cells. Proc Natl Acad Sci U S A 98, 15282-15287	[Abstract/Full Text]
Nicholson C, (1980). Modulation of extracellular calcium and its functional implications. Fed Proc 39, 1519-1523	[Medline]
Ramaswami M, Krishnan KS & Kelly RB (1994). Intermediates in synaptic vesicle recycling revealed by optical imaging of Drosophila neuromuscular junctions. Neuron 13, 363-375	[Medline]
Richards DA, Guatimosim C & Betz WJ (2000). Two endocytic recycling routes selectively fill two vesicle pools in frog motor nerve terminals. Neuron 27, 551-559	[Medline]
Rochon D, Rousse I & Robitaille R (2001). Synapse-glia interactions at the mammalian neuromuscular junction. J Neurosci 21, 3819-3829	[Abstract/Full Text]
Rouze NC , & Schwartz EA (1998). Continuous and transient vesicle cycling at a ribbon synapse. J Neurosci 18, 8614-8624	[Abstract/Full Text]
Ryan TA, Reuter H, Wendland B, Schweizer FE, Tsien RW & Smith SJ (1993). The kinetics of synaptic vesicle recycling measured at single presynaptic boutons. Neuron 11, 713-724	[Medline]
Ryan TA, Smith SJ & Reuter H (1996). The timing of synaptic vesicle endocytosis. Proc Natl Acad Sci U S A 93, 5567-5571	[Abstract]
Sankaranarayanan S , & Ryan TA (2000). Real-time measurements of vesicle-SNARE recycling in synapses of the central nervous system. Nat Cell Biol 2, 197-204	[Medline]
Son Y-J, Scranton TW, Sunderland WJ, Baek SJ, Miner JH, Sanes JR & Carlson SS (2000). The synaptic vesicle protein SV2 is complexed with an 5-containing laminin on the nerve terminal surface. J Biol Chem 275, 451-460	[Abstract/Full Text]
Stanley EF, (2000). Presynaptic calcium channels and the depletion of synaptic cleft calcium ions. J Neurophysiol 83, 477-482	[Abstract/Full Text]
Sun JY, Wu XS & Wu LG (2002). Single and multiple vesicle fusion induce different rates of endocytosis at a central synapse. Nature 417, 555-559	[Medline]
Teng H, Cole JC, Roberts RL & Wilkinson RS (1999). Endocytic active zones: hot spots for endocytosis in vertebrate neuromuscular terminals. J Neurosci 19, 4855-4866	[Abstract/Full Text]
Teng H , & Wilkinson RS (2000). Clathrin-mediated endocytosis near active zones in snake motor boutons. J Neurosci 20, 7986-7993	[Abstract/Full Text]
Teng H , & Wilkinson RS (2001). Temperature and extracellular calcium dependence of endocytosis in snake motor terminals. Soc Neurosci Abstr 27, 383.15
Vassilev PM, Mitchel J, Vassilev M, Kanazirska M & Brown EM (1997). Assessment of frequency-dependent alterations in the level of extracellular Ca2+ in the synaptic cleft. Biophys J 72, 2103-2116	[Abstract]
Von Gersdorff H , & Matthews G (1994a). Dynamics of synaptic vesicle fusion and membrane retrieval in synaptic terminals. Nature 367, 682-683	[Medline]
Von Gersdorff H , & Matthews G (1994b). Inhibition of endocytosis by elevated internal calcium in a synaptic terminal. Nature 370, 652-655	[Medline]
Wilkinson RS , & Lichtman JW (1985). Regular alternation of fiber types in the transversus abdominis muscle of the garter snake. J Neurosci 5, 2979-2988	[Abstract]
Wilkinson RS, Lunin SD & Stevermer JJ (1992). Regulation of single quantal efficacy at the snake neuromuscular junction. J Physiol 448, 413-436	[Abstract]
Wu LG , & Betz WJ (1996). Nerve activity but not intracellular calcium determines the time course of endocytosis at the frog neuromuscular junction. Neuron 17, 769-779	[Medline]
Wurth GA , & Zweifach A (2002). Evidence that cytosolic calcium increases are not sufficient to stimulate phospholipid scrambling in human T-lymphocytes. Biochem J 362, 701-708	[Medline]

Acknowledgements

We thank J. Cole and L.-G. Wu for helpful discussions and for critical reading of the manuscript. This work was supported by United States Public Health Service grant NS-24752.

This article has been cited by other articles:

				W. L. Coleman, C. A. Bill, F. Simsek-Duran, G. Lonart, D. Samigullin, and M. Bykhovskaia Synapsin II and calcium regulate vesicle docking and the cross-talk between vesicle pools at the mouse motor terminals J. Physiol., October 1, 2008; 586(19): 4649 - 4673. [Abstract] [Full Text] [PDF]

				M. A. Gaffield and W. J. Betz Synaptic Vesicle Mobility in Mouse Motor Nerve Terminals with and without Synapsin J. Neurosci., December 12, 2007; 27(50): 13691 - 13700. [Abstract] [Full Text] [PDF]

				R. Renden and H. von Gersdorff Synaptic Vesicle Endocytosis at a CNS Nerve Terminal: Faster Kinetics at Physiological Temperatures and Increased Endocytotic Capacity During Maturation J Neurophysiol, December 1, 2007; 98(6): 3349 - 3359. [Abstract] [Full Text] [PDF]

				M. Y Lin, H. Teng, and R. S Wilkinson Vesicles in snake motor terminals comprise one functional pool and utilize a single recycling strategy at all stimulus frequencies J. Physiol., October 15, 2005; 568(2): 413 - 421. [Abstract] [Full Text] [PDF]

				H. Teng and R. S Wilkinson Clathrin-mediated endocytosis in snake motor terminals is directly facilitated by intracellular Ca2+ J. Physiol., June 15, 2005; 565(3): 743 - 750. [Abstract] [Full Text] [PDF]

				A. M. Hofer Another dimension to calcium signaling: a look at extracellular calcium J. Cell Sci., March 1, 2005; 118(5): 855 - 862. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 551/1/103    most recent jphysiol.2003.041152v1 Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Teng, H. Articles by Wilkinson, R. S. Search for Related Content PubMed PubMed Citation Articles by Teng, H. Articles by Wilkinson, R. S.