| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Journal of Physiology (2001), 532.3, pp. 637-647
© Copyright 2001 The Physiological Society
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
MEPCs), measured at the same junction immediately after stimulation, by 16 %. Three thousand stimuli reduced size by 23 %. When the solution contained 10 µM neostigmine (NEO) 3000 stimuli reduced MEPCs by 60 %, because with acetylcholinesterase (AChE) inhibited, MEPC size is more sensitive to changes in acetylcholine (ACh) content. Similar decreases in miniature endplate potential size (MEPP) followed repetitive stimulation of contracting preparations.
MEPCs increased back to pre-stimulus values with a half-time of 8-10 min. Recovery was blocked by (-)-vesamicol (VES), by hemicholinium-3 (HC3) and by nicotinic cholinergic agonists - all of which inhibit ACh loading into synaptic vesicles.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Most chemical synapses release their neurotransmitter in packets, or quanta. Quantal release produces a high local concentration of the transmitter in the synaptic cleft, which enables the synapse to operate rapidly (reviewed by Van der Kloot & Molgó, 1994). The quanta are packaged within synaptic vesicles. After the quanta are released the membrane of the synaptic vesicles is retrieved and formed into new vesicles, which are then reloaded with transmitter. These recycled vesicles then become available for release. Within this broad picture, many details of the retrieval and reloading processes remain fuzzy (Betz & Angleson, 1998).
A large body of previous work concluded that repetitive stimulation does not alter the size of the quanta at the neuromuscular junction (reviewed by Van der Kloot & Molgó, 1994). Contrary reports that repetitive nerve stimulation reduces quantal size did not establish whether the size was smaller because less ACh was released or because the endplate was less sensitive to ACh (Doherty et al. 1984; Glavinovi'c, 1988). We set out to clarify these points.
The present paper follows on recent work in which the rate of quantal release was accelerated with elevated K+ (Van der Kloot et al. 2000). Because elevated K+ solutions depress choline (Ch) uptake into nerve terminals, they cannot be used to see whether release itself decreases quantal size (Naves et al. 1996). Two methods used in the elevated K+ work are used again: inhibiting ACh loading into synaptic vesicles and measuring the store of releasable quanta. Three inhibitors of ACh loading were used. (-)-Vesamicol (VES) blocks active ACh transport into isolated synaptic vesicles (reviewed by Prior et al. 1992; Parsons et al. 1993; Van der Kloot & Molgó, 1994). NH4+ inhibits ACh uptake by diminishing the proton gradient across the vesicular membrane that is required for transmitter accumulation (Van der Kloot, 1987). Hemicholinium-3 (HC3) blocks the Na+-driven Ch uptake mechanism in the membrane of the motor nerve terminal, which may be directly coupled to ACh synthesis and loading (Yamamura & Snyder, 1973; Gylys & Jenden, 1996).
The number of vesicles in the releasable store was estimated by counting how many quanta were released by the proton ionophore carbonyl cyanide m-chlorophenylhydrazone (CCCP) (Molgó & Pécot-Dechavassine, 1988). The CCCP experiments showed that the number of quanta released from different preparations varied substantially. However, the numbers released by CCCP from paired preparations from the same frog are correlated, which means that the muscle from one side of the body can be used as a control for the other (Van der Kloot et al. 2000). Therefore CCCP was used in the present paper to see whether there are fewer quanta in the releasable store after prolonged repetitive stimulation, as reported by Ceccarelli & Hurlbut (1980).
The results are discussed in light of the recent discovery that there are two distinct endocytic pathways for reforming synaptic vesicles at the frog neuromuscular junction (Richards et al. 2000).
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Preparations
Sartorius or cutaneous pectoris nerve-muscle preparations were dissected from Rana pipiens, which were killed by double pithing in accordance with University guidelines (Van der Kloot et al. 1998). The Ringer solution contained (mM): NaCl, 120; KCl, 2.0; CaCl2, 2.5; Tes-NaOH, 4.0 (pH 7.4). Many of the recordings were made in Ringer solution containing 0.03 µM tetrodotoxin, to suppress spontaneous muscle contractions.
Use of neostigmine
NEO (10 µM) was used in many experiments because it was found that inhibiting much of the AChE accentuates experimentally induced changes in MEPC sizes (Van der Kloot et al. 1994). This is predicted by mathematical modelling of the generation of MEPCs in the presence and absence of AChE. We used the model of Wathey et al. (1979), which ignores the anatomical complexities of the endplate included in more sophisticated models (Stiles et al. 1996), which for present purposes should make little difference. The equations, the methods used for solving them and the parameters used are in Van der Kloot et al. (1994). Eliminating AChE has little effect on the peak amplitudes of the model MEPCs (Fig. 1A), but has a substantial effect on their time integrals (Fig. 1B). Without AChE the plot of MEPC size as a function of quantal ACh content has a steeper slope. Therefore with AChE inhibited, the changes in the MEPCs reflect changes in quantal ACh content better than changes in amplitude (see also Van der Kloot, 1997).
Electrophysiology
When indicated in the text, nerve-muscle preparations were pre-treated by soaking in Ringer solution containing 2 M formamide for 17-20 min to largely eliminate excitation-contraction coupling (del Castillo & Escalona de Motta, 1978). The preparations were pinned on a layer of silicone rubber in an acrylic chamber with a volume of about 5 ml. Recordings were at room temperature (18-22 °C).
The motor nerve was drawn into a suction electrode for stimulation. The pulses were at least twice supra-maximal, 80 µs square waves, which in most experiments were produced by an FHC pulsar 6B stimulator (Brunswick, ME, USA) driving a WPI A360 stimulus isolator (Sarasota, FL, USA).
MEPCs were recorded using the two-electrode voltage clamp; the recording methods and criteria for accepting data were described in detail by Van der Kloot et al. (1994). Briefly, the glass microelectrodes were bevelled, with DC resistances between 2 and 4.5 M
. Clamping was done with an Axon Instruments Axoclamp-2A voltage clamp amplifier (Foster City, CA, USA). The signals were passed on to an Axon Instruments CyberAmp 300 signal conditioner, which amplified over a bandwidth from 0.1 to 1000 Hz. The signal emerging from the CyberAmp was split, one branch going to an Axon Instruments AI 2020 event detector, the other to a ComputerBoards DAS-16 330 A/D converter (Mansfield, MA, USA). When a MEPC passed a threshold set just above the noise level, the event detector delivered a TTL pulse to the A/D board, which digitized at 10 kHz. The 200 points (2 ms) before the TTL pulse and the 800 (8 ms) after the pulse were output to a computer. An operator observed the signal. If the MEPC was not contaminated by an overlapping MEPC or by electrical noise, its voltage-time integral, MEPC, was calculated as the measure of the quantal size (Van der Kloot, 1997). If the holding current began to markedly increase during an experiment the clamp was discontinued and the data discarded.
MEPPs and MEPPs were recorded with the same apparatus, except that only a single intracellular electrode was used. The amplitudes of the MEPPs were corrected to a standard resting potential of -90 mV by the method of Katz & Thesleff (1957). Examples of the MEPPs and MEPCs were shown previously (Van der Kloot, 1991; Naves & Van der Kloot, 1996).
Estimating the releasable store of quanta with CCCP
Briefly, the number of quanta released following exposure to CCCP was measured as follows (details in Van der Kloot et al. 2000). An endplate was penetrated. CCCP was added to the bath to give a final concentration of 10 µM. Every 100 s the computer recorded a series of A/D points at 2 kHz extending for 0.64 s. If the membrane potential of the fibre fell to a level where seeing the MEPPs became problematical the microelectrode was inserted into a fresh endplate. In some sets all recording was from a single fibre, in others as many as three junctions were penetrated. Previous work showed that the number of quanta released from paired muscles was correlated, regardless of how many penetrations were required (Van der Kloot et al. 2000). Recording was terminated when the miniature endplate potential frequency (FMEPP) reached a low value, usually < 1 s-1. After the recording was completed, the operator reviewed the records in the form of 64 ms segments, counting the number of MEPPs in each segment. The total number of MEPPs released was estimated assuming that the measured rate persisted until the next set of measurements was taken in.
Measuring evoked quantal output
To test for possible effects of a drug on evoked quantal output nerve-muscle preparations were placed in a solution containing (mM): NaCl, 120; KCl, 2.5; MgCl2, 2.5; CaCl2, 0.2; and TES, 4.0. This solution substantially decreased quantal output. Mean quantal outputs (m0) were estimated by the method of failures, counting the number of stimuli at 0.5 Hz that were not followed by an endplate potential (n0) and the total number of stimuli (N). Then:
m0 = ln(N/n0).
The standard error of the estimate (S.E.) is given by:
(Martin, 1966). Differences between mean quantal outputs were tested by Student's t test.
Detecting subpopulations of quantal size
The methods described in this section were used and tested for reproducibility previously (Van der Kloot et al. 2000). To perform the test adequately, at least 300 MEPPs must be recorded in each set of data. The MEPPs from each set were sorted in ascending order and plotted as a cumulative curve. Each of these curves was fitted with one and with two cumulative lognormal distributions, as illustrated in Fig. 8, using the Levenberg-Marquardt method (Press et al. 1989). If the fit to two distributions appeared to be better, the next step was to determine whether improvement was because the two distributions actually fitted the data better or merely because more fitting parameters were used (Horn, 1987; Van der Kloot et al. 2000). The MEPPs, arranged in ascending order, were sorted into bins, each of which contained a minimum of five observations. The predicted values from the one lognormal and two lognormal curves were sorted into the same bins. The likelihood functions, ll = Fobs 
ln(fobs/fpred) were calculated for each bin. These likelihood functions were then summed, 
llone and lltwo. Then:
G = 2(llone - lltwo).
G is distributed as 
2 with 3 degrees of freedom, because the curve that is the sum of two lognormal distributions is calculated with three more parameters than the single one. The probability that the data are best described by one lognormal curve rather than two is obtained from the 2 and the degrees of freedom
Statistics
All of the means reported in the text and in the figures are given with the ±95 % confidence intervals, which were determined by resampling (Van der Kloot, 1996). In the experiments on the recovery of quantal size after stimulation, resampling was also used to determine if there was a statistically significant difference between the first 30-50 MEPCs after the end of the repetitive stimulation and the last 30-50 MEPCs recorded at the junction (Van der Kloot, 1996). If P < 0.05 the differences are termed significant.
Chemicals
The drugs were from Sigma (St Louis, MO, USA), except for (-)-vesamicol (2-(4-phenylpiperidino)cyclohexanol), HC3 and nicotine bitartrate, which were obtained from RBI (Natick, MA, USA).
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Quantal size decreases with stimulation
Preparations pretreated with formamide to disrupt excitation-contraction coupling were placed in Ringer solution without NEO. Junctions were voltage clamped and MEPCs were measured. Then the clamp was discontinued, the motor nerve was stimulated 200 or 3000 times at 10 Hz, the junction was re-clamped and 50-100 MEPCs were measured. Since FMEPP was elevated following the stimulation, all of these MEPCs were recorded in less than 5 min; the short recording time is important because, as will be shown shortly, size recovers as time passes. There was a significant decrease in MEPC size following stimulation (Fig. 2A). The decrease in MEPC size following 200 stimuli was almost identical to the decrease in MEPP amplitude measured by Doherty et al. (1984), who withdrew the recording electrode during stimulation. When we repeated our measurements in the presence of 10 µM NEO the decreases in size following 3000 stimuli were greater (Fig. 2A). This is accounted for by a mathematical model of MEPC generation, which shows that the slope of the plot relating the time integral of the number of channels opened as a function of the ACh released is steeper when AChE is inhibited (Fig. 1). No significant decreases in MEPC sizes were detected following 50 (n = 4) or 100 stimuli (n = 4) in 10 µM NEO. Doherty et al. (1984) showed that size decreased because there were fewer MEPPs with large amplitudes, and an increase in the proportion with smaller amplitudes. They argued that this shows that the quanta are smaller because less ACh is released in each packet. We will take up this question shortly.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 1. Calculations of model MEPCs with and without active AChE
| ||
Because of the natural variation in quantal size at junctions and between junctions, for accurate measurement of size changes it is necessary to measure from the same junction before and after treatment, as in the experiments just reported, or to measure from five or more fibres in each preparation before and after treatment (reviewed by Van der Kloot, 1991). Measuring from a number of fibres after stimulation was not feasible in the present work, because quantal size recovers over time (see below). The best we could do to see whether quantal size also decreased following stimulation when excitation-contraction coupling was intact was to stimulate at 10 Hz and then rapidly measure MEPPs (Fig. 2B). Size surely decreased following stimulation, so the decrease is not an artifact somehow produced by the uncoupling in formamide.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 2. Tetanic stimulation at 10 Hz reduces miniature sizes A, the reduction in
| ||
Stimulation does not substantially decrease sensitivity to ACh
Quantal sizes might be smaller after stimulation because less ACh is released per quantum, because the endplate is less sensitive to ACh, or both. Repetitive stimulation decreases quantal size at the Drosophila neuromuscular junction by desensitizing receptors (Adelsberger et al. 1997) and there is evidence for desensitization with repetitive stimulation at the frog neuromuscular junction as well (Giniatullin et al. 1989). One experimental test used to help decide between these alternatives was to apply iontophoretic pulses of ACh to endplates in Ringer solution containing 10 µM NEO, the solution in which stimulation produced the largest decreases in quantal size. The preparations were pretreated with formamide to uncouple contraction. The depolarization of the endplate in response to each ACh pulse was measured as the time integral of the endplate depolarization. After control responses had been collected, the iontophoretic pulses were stopped while the motor nerve was stimulated 3000 times at 10 Hz. Then the iontophoretic pulses were resumed and the responses measured once again. Stimulation decreased the response to ACh only slightly (Fig. 3), even though MEPCs were reduced by roughly 60 % (Fig. 2A). Similar results were obtained in five additional experiments. This suggests that quantal size decreases mainly because less ACh is released per quantum. Additional evidence supporting this conclusion will be presented shortly.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 3. A period of repetitive stimulation does not reduce the endplate response to an iontophoretic pulse of ACh A, the larger signal, peaking at 60 ms, is an example of the depolarization produced by the ACh pulse. The smaller signal, at 20 ms, is a MEPP. The solution contained 10 µM NEO. B, the time integrals of the responses to ACh pulses over time. During the interruption in the recording the motor nerve was stimulated 3000 times at 10 Hz.
| ||
Recovery of MEPC size following stimulation
Formamide-treated preparations were placed in Ringer solution containing 10 µM NEO. MEPCs were measured. The clamp was discontinued and the motor nerve stimulated 3000 times at 10 Hz. As soon as the stimulation was terminated, the clamp was reestablished and MEPCs measured once again. Immediately after stimulation the MEPC size was substantially decreased compared to before stimulation; the mean decrease was to 50 ± 12 % (n = 24). Following stimulation quantal size rose back toward the pre-stimulation levels over a period of minutes (Fig. 4A). Routinely there is a considerable spread in quantal sizes, so the means of relatively small samples of MEPCs have substantial variance. Because of this scatter the half-time for recovery can only be approximated; 8-10 min seems a reasonable estimate from the 24 experiments. Doherty et al. (1984), who recorded MEPPs in contracting preparations by withdrawing the recording electrode during stimulation and then reinserting in the same fibre, observed similar time courses of recovery.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 4. All solutions contained 10 µM NEO. The filled bars show the mean
| ||
Effects of ACh loading inhibitors on the recovery of MEPC size
The inhibitors were added to the solution just before the beginning of recording, because in 10 µM NEO in unstimulated preparations they gradually decrease MEPC size by as much as 30 % (Van der Kloot et al. 1994; Naves & Van der Kloot, 1996).
(-)-Vesamicol. The results so far suggest that quantal size recovers because additional ACh is added to the quanta. To test this hypothesis further, we measured MEPCs in the presence of VES. Quantal size did not recover significantly when the solution contained 2 µM VES, but did recover in 0.2 µM (Fig. 4B and Fig. 5). The inhibition of recovery by VES supports the explanation that quantal size decreases after stimulation mainly because less ACh is released per quantum, and that recovery occurs when additional ACh is packed into the vesicles.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 5. The effects of drugs on the recovery of Each column shows the ratio of the size in the final time bin following recovery to the size in the initial time bin, immediately after the stimulation. The error bars on the right of the columns show the +95 % confidence limits. The vertical dashed line shows the ratio expected if there was no recovery in quantal size. The numbers to the right of each bar show (number of experiments in which the sizes in the final time bins were significantly larger at P < 0.05)*/(total number of experiments).
| ||
Hemicholinium-3. HC3 at 1 µM blocked the recovery of quantal size following the stimulation, while 0.1 µM HC3 was ineffectual (Fig. 4C and Fig. 5). (In some of the experiments with 1 µM HC3 the quantal size decreased significantly during the recovery period following the stimulation.) Much of the previous work on the effects of HC3 on quantal size was done with concentrations of 10 µM or even 30 µM. Such high concentrations should be avoided because they might have additional effects on the preparation.
If the uptake of Ch is important for the recovery of quantal size following stimulation, we were concerned that the extent of the recovery might be diminished by the anticholinesterase in the extracellular solution, which by blocking ACh hydrolysis might materially diminish the quantity of Ch available for recycling. Therefore we repeated the control experiments in solutions containing 10 µM Ch. The presence of Ch did not significantly alter the extent of the recovery (Fig. 5).
HC3 might act on another target. In snail neurons, HC3 is an agonist for a neuronal-type ACh receptor (Poulain et al. 1987). We can see whether HC3 is also an agonist for the neuronal ACh receptor on the frog motor nerve terminal because if so it will decrease evoked quantal output (Van der Kloot, 1993). The mean quantal output, m0, was estimated by the method of failures. In control preparations 2 µM carbachol, an ACh agonist, decreased evoked quantal output, m0, to 56 ± 11 % (n = 10; 95 % confidence interval) of the pre-drug control. In 1 µM HC3 evoked quantal output was 128 ± 22 % (n = 10) of the control. Clearly HC3 does not act like a nicotinic agonist by decreasing evoked output. HC3 probably halts the recovery of MEPC size by blocking the Ch transporter.
ACh agonists. Nicotinic ACh agonists block the addition of ACh to quanta in preparations that have been treated to produce an increase in quantal size (Van der Kloot, 1993). In the next experiments 1 µM carbachol was added to the extracellular solution before the repetitive stimulation. In five experiments quantal size did not recover significantly after the stimulation; in two of the five experiments quantal size decreased significantly during the recovery period (Fig. 5). Nicotine and 1,1-dimethyl-4-phenyl-piperazinium (DMPP, an agonist which targets nicotinic ACh receptors) also blocked the recovery of quantal size (Fig. 4D and Fig. 5).
MEPPs after stimulation in the presence of ACh-loading inhibitors
Preparations were exposed to 1 µM HC3, 2 µM VES, 20 mM NH4+, or all three simultaneously. MEPPs were measured at several junctions, and then the motor nerves were stimulated at 10 Hz for varying times. After the stimulation ended, MEPPS were measured once again. In all the inhibitor-containing solutions, the size of the MEPPs decreased with stimulation (Fig. 6). The fewest stimuli used was 3000; more stimuli did not appear to decrease quantal size further. The declines were very similar in the presence of the individual inhibitors or when all three inhibitors together were present. Even after the longest stimulation periods, MEPPs were readily seen and their frequency was in the normal range.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 6. The error bars show the ±95 % confidence limits.
| ||
The number of quanta released by CCCP following repetitive stimulation
To see whether the store of releasable quanta falls following repetitive stimulation and whether quanta containing ACh can be formed when ACh-loading inhibitors are present, we used CCCP to estimate the store of releasable quanta (Molgó & Pécot-Dechavassine, 1988; Van der Kloot et al. 2000). To answer the first question the nerve from one preparation of the pair was stimulated 3000 times at 10 Hz. Then it was placed in 10 µM CCCP and the number of quanta subsequently released measured. The second preparation was exposed to 10 µM CCCP without any stimulation (Fig. 7A). Stimulated preparations released fewer quanta than the unstimulated controls. To see whether rest increases the releasable store once again, one preparation was stimulated 3000 times at 10 Hz and then kept in Ringer solution for 4-5 h before exposure to 10 µM CCCP. The other preparation was exposed to CCCP immediately after stimulation (Fig. 7B). The rested preparations released more quanta than those treated with CCCP immediately after stimulation.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 7. Changes in the number of quanta released by 10 µm CCCP as a result of stimulation, after recovery from simulation, and after recovery in the presence of ACh-loading inhibitors A, one muscle was unstimulated (x-axis), while the second was stimulated 3000 times at 10 Hz (y-axis) before exposure to the CCCP. The continuous line shows the expectation if the number of releasable quanta was unchanged. The dotted line is the regression coefficient for the data, R 2 = 0.47. B, one muscle was exposed to the CCCP immediately after being stimulated 3000 times (x-axis), while the other was allowed to rest for 3-5 h before exposure to CCCP (y-axis). The continuous line shows the expectation if rest had no effect. The dotted line is the regression, R 2 = 0.70. C, preparations stimulated 3000 times at 10 Hz either without drugs or in the presence of ACh-loading inhibitors. The number of quanta released by CCCP after 3-5 h rest was measured. The continuous line is the expectation if the drugs had no effect on releasable number. The dotted lines are the ±95 % confidence limits for the regression. The slope of the regression, which is not shown, was 0.82, R 2 = 0.82.
| ||
We used paired preparations once again to see whether ACh-loading inhibitors would decrease the number of quanta added to the releasable store during a recovery period following repetitive stimulation. One of the pair was stimulated 3000 times at 10 Hz in Ringer solution, and then allowed to rest for 3-5 h before being exposed to 10 µM CCCP and the number of quanta released estimated. The second preparation was treated the same, except that during stimulation and recovery the solution contained either 2 µM VES, or 1 µM HC3 + 2 µM VES. The inhibitors did not decrease the size of the releasable store (Fig. 7C). Since releasable quanta are formed during the rest period, it seems that they can be made even in the presence of ACh-loading inhibitors.
Subpopulations of quantal sizes following stimulation in VES
The approach in these experiments was used previously when we enhanced the rate of quantal release with elevated K+ in the presence of an ACh-uptake inhibitor and then examined the MEPPs to see whether there were two size classes (Van der Kloot et al. 2000). Since there were no obvious differences in the effects of the inhibitors tested, all the experiments looking for two subpopulations of quantal sizes were done in 2 µM VES. (We stimulated at 30 Hz because we hoped that we could use others' data to estimate how many quanta were released by the stimulation, but this required too many assumptions to be useful.) Van der Kloot et al. (2000) showed that at junctions most MEPP distributions are fitted by a single lognormal probability distribution function. However, after 1000 or more stimuli, the cumulative distributions of the MEPPs from most of the junctions were fitted best by two lognormal curves in 46 out of 61 cases (Fig. 8). The figure shows the data plotted on probability scales; the classical approach for detecting two subpopulations was by examining such plots for breaks between two lines fitted to the plot. The figures show how difficult this approach is in practice. In earlier experiments in elevated K+ the fraction of the MEPPs in the larger subpopulation decreased as more quanta were released (Van der Kloot et al. 2000, Fig. 5B). A decrease in the fraction in the larger subpopulation was by no means as sure in nerve stimulation experiments, perhaps because the rate at which the number of quanta released falls off with nerve stimulation varies from one preparation to the next (data not shown). A decrease in the fraction in the large subpopulation suggests that in ACh-loading inhibitors the newly formed quanta are smaller. The ratio of the mean size of the MEPPs in the larger subpopulation to that in the smaller was 2.6 ± 0.26 (n = 46), close to the ratio found in the elevated K+ experiments. The advantage of the K+ data is that we can estimate the total number of quanta released by the treatment: 60 % of the MEPPs were in the larger subpopulation when 100 000 quanta had been released.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 8. Examples of the fitting of the distribution of The points are shown as black circles. Red curve, one lognormal; blue curve, two lognormals. The
| ||
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Katz (1969), who introduced the term 'quantum', later pointed out that the name misleads some readers. By a misconceived analogy with the physical quantum, they acquire the false idea that miniatures have almost uniform size. In reality quantal size at a junction varies over a fivefold range; in untreated preparations the CV is about 0.3. With treatment, the range is much greater.
A variety of treatments substantially increase quantal size (references in Van der Kloot et al. 1998; reviewed by Sulzer & Pothos, 2000). In the most dramatic instances at the neuromuscular junction, mean size increases fourfold. At both the frog and the mouse neuromuscular junctions size increases because more ACh is loaded into the readily releasable pool of quanta (Yu & Van der Kloot, 1991; Van der Kloot, 1993).
Surely it is less surprising that quanta can also be made smaller. What is unexpected is that relatively modest stimulation, releasing fewer than 20 % of the quanta in the terminal, significantly decreases size. The best data come from experiments in which MEPC size was measured in the same fibre before and after the stimulation, which bypasses problems with the variations in size from junction to junction and from preparation to preparation or the uncertainty of reinserting electrodes into the same fibre before and after a bout of contraction (Fig. 2A). Size is significantly decreased by 200 stimuli. Three thousand stimuli reduced the MEPCs recorded in 10 µM NEO by roughly 60 %. Experiments in which MEPPs from randomly chosen endplates are sampled before and after tetanic stimulation are less reliable for measuring the magnitude of the size changes, but they clearly show that quantal size also decreases following stimulation of preparations in which excitation- contraction coupling was intact (Fig. 2B).
There are two reasons to think that after stimulation the quanta are smaller largely because they contain less ACh. First, the response of the endplate to ACh applied by iontophoresis was little changed after repetitive nerve stimulation that would decrease MEPC size by 60 % (Fig. 3). Second, after stimulation was over quantal size gradually rose back toward the pre-stimulus value as the preparation rested. Recovery was blocked by the inhibitors of ACh loading into synaptic vesicles: VES, HC3, NH4+, and nicotinic agonists (Fig. 5).
When do vesicles receive their stockpile of ACh? The answer seems to be that loading can occur in several steps. Two hundred stimuli at 10 Hz releases somewhere between 10 000 and 20 000 quanta (Doherty et al. 1984; Van der Kloot, 1993). Doherty et al. (1984) pointed out that this is about the number of vesicles in the readily releasable pool, probably those attached to the active zones at the motor nerve terminals. So it seems that once the vesicles join the readily releasable pool they receive an appreciable supplementary loading of ACh. We called this second-stage loading (Van der Kloot, 1991; Naves et al. 1996; Naves & Van der Kloot, 1996).
This name should be replaced. It was based in the idea that the readily releasable pool is replenished by vesicles coming from the reserve pool in the axoplasm (Betz & Angleson, 1998). Loading ACh into vesicles in the reserve pool was the first stage and further loading in the readily releasable pool was the second. This picture has been redrawn by Richards et al. (2000), who show that when stimulation first begins the vesicles in the readily releasable pool are replenished via a discrete, fast endocytic pathway. It requires more prolonged stimulation to move vesicles from the reserve pool into the readily releasable pool. A second, slower endocytic pathway that involves membrane infoldings and cisternae replenishes the reserve pool. Since there is no single first stage, the term second-stage loading is misleading. 'Final loading' seems a better name for the pumping of additional ACh into the vesicles once they are added to the readily releasable pool. Previous work showed that the enlargement of quantal size occurs largely by increased final loading, so the amount of ACh incorporated in final loading can be regulated.
There is substantial evidence for final loading (Doherty et al. 1984; Van der Kloot, 1991; Naves et al. 1996; Naves & Van der Kloot, 1996). Final loading can be blocked by all the inhibitors of vesicular ACh uptake that we have used. The half-time for final loading is 8-10 min (Doherty et al. 1984; Van der Kloot, 1993; Fig. 4). Vesicles reaching the readily releasable pool either via the rapid endocytic pathway or from the reserve pool seem to receive final loading. Quantal size is transitorily decreased following 200 stimuli, in which most of the vesicles released from the readily releasable pool presumably are replenished by the rapid endocytic pathway. It is also transitorily decreased by 3000 stimuli, when many of the replacement vesicles must come from the reserve pool, judging from the substantial decrease in the number of releasable quanta in the terminals (Fig. 7).
Our data give no more information about the initial loading of ACh into the vesicles formed by the fast endocytic pathway. They do suggest that the filling of the vesicles in the reserve pool involves more than one step. After prolonged stimulation in the presence of VES two subpopulations of quantal size were often detected (Fig. 8). Note that final loading is not operating in these experiments because it has been blocked by the VES. Apparently during the stimulation vesicles released from the reserve pool are replaced by recycled vesicles containing less ACh. This was more clearly demonstrated in earlier work when the release rate was increased with 30 mM K+ solution. The fraction in the subpopulation of larger size quanta decreased as more quanta were released (Van der Kloot et al. 2000).
This leaves us with the question of how ACh first gets into reformed, recycling vesicles in the reserve pool. Even after thousands of stimuli in the presence of ACh uptake inhibitors, quanta were still readily seen and measured (Fig. 6). One possibility was that these quanta were all loaded before the inhibitor was present, and the recycled vesicles contain no transmitter. At the snake neuromuscular junction empty vesicles are formed and fuse with the terminal membrane when the preparation is stimulated (Parsons et al. 1999). We do not know whether this occurs in the frog. Previous work shows that repetitive stimulation reduces the number of releasable quanta and the number of vesicles in the terminal (Ceccarelli & Hurlbut, 1980). After 9000 stimuli frog nerve terminals are maximally stained with FM2-10 after resting for 20 min (Richards et al. 2000). We estimated the number of releasable ACh quanta by counting the MEPPs released by CCCP treatment. The CCCP-releasable store decreased when the drug was added immediately following 3000 stimuli (Fig. 7A). On the other hand, the number of releasable quanta increased again following a period of rest, and this increase also occurred when uptake inhibitors were present (Fig. 7B and C). Apparently some ACh is loaded in recycled vesicles even when an ACh-loading inhibitor is present in a concentration adequate to block the recovery of quantal size in final loading (Fig. 5). The initial loading of ACh into recycled vesicles does not seem to require the VES-inhibited ACh transporter or the proton gradient diminished by NH4+.
The most difficult aspect of these results to account for is the formation of reserve vesicles containing some ACh in the presence of HC3, which might be expected to block all ACh production in the terminal by halting Ch recycling. We speculate that there is a pathway for ACh synthesis that obtains Ch from another source, and that HC3 acts because Ch reuptake, ACh synthesis and transport into vesicles are linked into a chain (Gylys & Jenden, 1996). More work is needed on how ACh is loaded when inhibitors are present.
Another unresolved question is the pathway by which nicotinic agonists depress ACh loading into recycling quanta or into quanta that are increasing in size (Van der Kloot, 1993), because we are accustomed to think of these as directly operating membrane ions channels. Neurons from the rat dorsal septal nucleus hyperpolarize in response to nicotinic agonists (Sorenson & Gallagher, 1996). The hyperpolarization is eliminated when GTP
S is present in the patch pipette, strongly suggesting a G-protein link between this neuronal nicotinic receptor and the K+ channel that is activated. Perhaps a second messenger is also involved in the frog motor nerve terminal as a link between the nicotinic neuronal receptor on the motor nerve terminal and the mechanism for loading synaptic vesicles with ACh. Motor nerve terminals have nicotinic receptors (Tsuneki et al. 1995), which are implicated in the control of evoked quantal output (Van der Kloot, 1993; Tian et al. 1997). The possibility that feedback mechanisms alter quantal loading as well as quantal release deserves attention at other synapses where autoinhibition occurs (reviewed by Wu & Saggau, 1997; MacDermott et al. 1999).
The combination of final loading and a fast endocytic pathway explains results that seriously challenged the concept that quantal release occurs from vesicles prepackaged with transmitter. Biochemical data show that newly synthesized ACh is released more rapidly than expected if it mixed with all the ACh within the terminal. Newly synthesized false transmitters also soon appear in the quanta (reviewed by Van der Kloot & Molgó, 1994; Naves et al. 1996). Since most of the terminal's total stock of ACh is in the reserve vesicles, newly synthesized ACh or false transmitter should be put into quanta during final loading. Vesicles formed by the fast endocytic pathway should be filled mostly with newly synthesized ACh. Therefore the prompt release of newly synthesized transmitter seems to be well accounted for.
When the link between quanta and vesicles was first proposed, it seemed reasonable to think that vesicles were formed, filled and then stand by until released. Now we know there are two endocytic pathways for recycling and several steps in transmitter loading. At least the final step in transmitter loading can be regulated to vary the ACh content of the quanta. The picture now lacks the elegance of simplicity, but does account for the data. Naturally enough, as we have pointed out, there are still open questions to be answered.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| ADELSBERGER H., HECKMANN, M. & DUDEL, J. (1997). The amplitude of quantal currents is reduced during short-term depression at neuromuscular synapses in Drosophila. Neuroscience Letters 225, 5-8 | [Medline] |
| BETZ W. J. & ANGLESON, J. K. (1998). The synaptic vesicle cycle. Annual Review of Physiology 60, 347-363. | [Abstract/Full Text] |
| CECCARELLI B. & HURLBUT, W. P. (1980). Vesicle hypothesis of the release of quanta of acetylcholine. Physiological Reviews 60, 396-441 | [Medline] |
| DEL CASTILLO J. & ESCALONA DE MOTTA, G. (1978). A new method for excitation-contraction uncoupling in frog skeletal muscle. Journal of Cell Biology 78, 782-784 | [Abstract] |
| DOHERTY P., HAWGOOD, B. J. & SMITH, I. C. H. (1984). Changes in miniature end-plate potentials after brief nervous stimulation at the frog neuromuscular junction. Journal of Physiology 356, 349-358 | [Abstract] |
| GINIATULLIN R. A., KHAMITOV, G., KHAZIPOV, R., MAGAZANIK, L. G., NIKOLSKY, E. E., SNETKOV, V. A. & VYSKOCIL, F. (1989). Development of desensitization during repetitive end-plate activity and single end-plate currents in frog muscle. Journal of Physiology 412, 113-122 | [Abstract] |
| GLAVINOVIC M. I. (1988). Changes in miniature end-plate currents due to high potassium and calcium at the frog neuromuscular junction. Synapse 2, 636-643 | [Medline] |
| GYLYS K. H. & JENDEN, D. J. (1996). Regulation of acetylcholine synthesis in the presence of hemicholinium mustard. Life Sciences 58, 1937-1946 | [Medline] |
| HORN R. (1987). Statistical methods for model descrimination. Biophysical Journal 51, 255-263 | [Abstract] |
| KATZ B. (1969). The Release of Neural Transmitter Substances. Liverpool University Press, Liverpool | |
| KATZ B. & THESLEFF, S. (1957). On the factors which determine the amplitude of the 'miniature endplate potential'. Journal of Physiology, 137, 267-278 | |
| MACDERMOTT A. B., ROLE, L. W. & SIEGELBAUM, S. A. (1999). Presynaptic ionotropic receptors and the control of transmitter release. Annual Review of Neuroscience 22, 443-485 | [Abstract/Full Text] |
| MARTIN A. R. (1966). Quantal nature of synaptic transmission. Physiological Reviews 46, 41-66 | |
| MOLGÓ J. & PÉCOT-DECHAVASSINE, M. (1988). Effects of carbonyl cyanide m-chlorophenylhydrazone (CCCP) on quantal transmitter release and ultrastructure of frog motor nerve terminals. Neuroscience 24, 695-708 | [Medline] |
| NAVES L. A., BALEZINA, O. P. & VAN DER KLOOT, W. (1996). Monoethylcholine as a false transmitter precursor at the frog and mouse neuromuscular junctions. Brain Research 730, 58-66 | [Medline] |
| NAVES L. A. & VAN DER KLOOT, W. (1996). Transmitter packaging at frog neuromuscular junctions exposed to anticholinesterases; the role of second stage acetylcholine loading. Journal of Neurophysiology 76, 2614-2625 | [Medline] |
| PARSONS R. L., CALUPCA, M. A., MERRIAM, L. A. & PRIOR, C. (1999). Empty synaptic vesicles recycle and undergo exocytosis at vesamicol-treated motor nerve terminals. Journal of Neurophysiology 81, 2696-2700 | [Abstract/Full Text] |
| PARSONS S. M., PRIOR, C. & MARSHALL, I. G. (1993). Acetylcholine transport, storage, and release. International Review of Neurobiology 35, 279-390 | [Medline] |
| POULAIN B., FOSSIER, P., BAUX, G. & TAUC, L. (1987). Hemicholinium-3 facilitates the release of acetylcholine by acting on presynaptic nicotinic receptors at a central synapse in Aplysia. Brain Research 435, 63-70 | [Medline] |
| PRESS W. H., FLANNERY, B. P., TEUKOLSKY, S. A. & VETTERLING, W. T. (1989). Numerical Recipes in Pascal. Cambridge University Press, Cambridge | |
| PRIOR C., MARSHALL, I. G. & PARSONS, S. M. (1992). The pharmacology of vesamicol: an inhibitor of the vesicular acetylcholine transporter. General Pharmacology 23, 1017-1022 | [Medline] |
| RICHARDS D. A., GUATIMOSIM, C. & BETZ, W. J. (2000). Two endocytic recycling routes selectively fill two vesicle pools in frog motor nerve terminals. Neuron 27, 551-559. | [Medline] |
| SORENSON E. M. & GALLAGHER, J. P. (1996). The membrane hyperpolarization of rat dorsolateral septal nucleus neurons is mediated by a novel nicotinic receptor. Journal of Pharmacology and Experimental Therapeutics 277, 1733-1743 | [Abstract] |
| STILES J. R., VANHELDEN, D., BARTOL, T. M., SALPETER, E. E. & SALPETER, M. M. (1996). Miniature endplate current rise times < 100 µs from improved dual recordings can be modeled with passive acetylcholine diffusion from a synaptic vesicle. Proceeding of the National Academy of Sciences of the USA 93, 5747-5752 | |
| SULZER D. & POTHOS, E. N. (2000). Regulation of quantal size by presynaptic mechanisms. Reviews in the Neurosciences 11, 159-212 | [Medline] |
| TIAN L., PRIOR, C., DEMPSTER, J. & MARSHALL, I. G. (1997). Hexamethonium- and methyllycacoconitine-induced changes in acetylcholine release from rat motor nerve terminals. British Journal of Pharmacology 122, 1025-1034 | [Medline] |
| TSUNEKI H., KIMURA, I., DEZAKI, K., KIMURA, M., SALA, C. & FUMAGALLI, G. (1995). Immunohistochemical localization of neuronal nicotinic receptor subtypes at the pre- and postjunctional sites in mouse diaphragm muscle. Neuroscience Letters 196, 13-16 | [Medline] |
| VAN DER KLOOT W. (1987). Inhibition of packing of acetylcholine into quanta by ammonium. FASEB Journal 1, 298-302 | [Abstract] |
| VAN DER KLOOT W. (1991). The regulation of quantal size. Progress in Neurobiology 36, 93-130 | [Medline] |
| VAN DER KLOOT W. (1993). Nicotinic agonists antagonize quantal size increases and evoked release at frog neuromuscular junction. Journal of Physiology 468, 567-589 | [Abstract] |
| VAN DER KLOOT W. (1996). Statistics for studying quanta at synapses: resampling and confidence limits on histograms. Journal of Neuroscience Methods 65, 151-155 | [Medline] |
| VAN DER KLOOT W. (1997). Making quantal analysis easier and more accurate. Journal of Neuroscience Methods 77, 129-133 | [Medline] |
| VAN DER KLOOT W., BALEZINA, O. P., MOLGÓ, J. & NAVES, L. A. (1994). The timing of channel opening during miniature endplate currents at the frog and mouse neuromuscular junctions: effects of fasciculin-2, other anti-cholinesterases and vesamicol. Pflügers Archiv 428, 114-126 | [Medline] |
| VAN DER KLOOT W., BENJAMIN, W. & BALEZINA, O. P. (1998). Calcitonin gene-related peptide acts presynaptically to increase quantal size and output at frog neuromuscular junctions. Journal of Physiology 507, 689-695 | [Abstract/Full Text] |
| VAN DER KLOOT W., COLASANTE, C., CAMERON, R. & MOLGÓ, J. (2000). Recycling and refilling of transmitter quanta at the frog neuromuscular junction. Journal of Physiology 523, 247-258 | [Abstract/Full Text] |
| VAN DER KLOOT W. & MOLGÓ, J. (1994). Quantal acetylcholine release at the vertebrate neuromuscular junction. Physiological Reviews 74, 899-991 | [Medline] |
| VAN DER KLOOT W. & NAVES, L. A. (1996). Accounting for the shapes and size distributions of miniature endplate currents. Biophysical Journal 70, 2175-2184 | [Abstract] |
| WATHEY J. C., NASS, M. M. & LESTER, H. A. (1979). Numerical reconstruction of the quantal event at nicotinic synapses. Biophysical Journal 27, 145-164 | [Abstract] |
| WU L.-G. & SAGGAU, P. (1997). Presynaptic inhibition of elicited neurotransmitter release. Trends in Neurosciences 20, 204-212 | [Medline] |
| YAMAMURA H. I. & SNYDER, S. (1973). High affinity transport of choline into synaptosomes of rat brain. Journal of Neurochemistry 21, 1355-1374 | [Medline] |
| YU S. P. & VAN DER KLOOT, W. (1991). Increasing quantal size at the mouse neuromuscular junction and the role of choline. Journal of Physiology 433, 677-704 | [Abstract] |
Acknowledgements
This work was supported by Grant 10320 from the National Institute of Neurological Diseases and Stroke. We thank Judy Samarel for assistance.
Corresponding author
W. Van der Kloot: The Boat House, 1 Fort Gate, Newhaven, East Sussex, BR9 9DR, UK.
Email: wvanderkloot{at}post.harvard.edu
Author's present address
L. A. Naves: Department of Physiology and Biophysics, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil.
Email: lnaves{at}mono.icb.ufmg.br
This article has been cited by other articles:
|
|
 |
|
  X. Wang, Y. Li, K. L. Engisch, S. T. Nakanishi, S. E. Dodson, G. W. Miller, T. C. Cope, M. J. Pinter, and M. M. Rich Activity-Dependent Presynaptic Regulation of Quantal Size at the Mammalian Neuromuscular Junction In Vivo J. Neurosci., January 12, 2005; 25(2): 343 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Axmacher, M. Stemmler, D. Engel, A. Draguhn, and R. Ritz Transmitter Metabolism as a Mechanism of Synaptic Plasticity: A Modeling Study J Neurophysiol, January 1, 2004; 91(1): 25 - 39. [Abstract] [Full Text]
|
|
|
|
|
|
J. D. Talbot, G. David, and E. F. Barrett Inhibition of Mitochondrial Ca2+ Uptake Affects Phasic Release From Motor Terminals Differently Depending on External [Ca2+] J Neurophysiol, July 1, 2003; 90(1): 491 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. O. Rizzoli and W. J. Betz Effects of 2-(4-Morpholinyl)-8-Phenyl-4H-1-Benzopyran-4-One on Synaptic Vesicle Cycling at the Frog Neuromuscular Junction J. Neurosci., December 15, 2002; 22(24): 10680 - 10689. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, with the enzyme active; 
, with the enzyme inhibited. A, the peak amplitudes; B, the time integrals. With the integrals the slope of the relation between the number of molecules in the quantum and the magnitude of the response is steeper when there is no enzyme.
, in Ringer solution; 
, in 10 µM NEO. In parentheses above each bar is the number of experiments. The error bars show the 95 % confidence intervals. B, the decrease in 
, in 30 mm NH4+; 
, in 2 µM VES;