Journal of Neuroscience Research 31502-506 (1992)
Focal, Extracellular Recording of Slow Miniature Junctional Potentials at the Mouse Neuromuscular Junction J. Vautrin and M.E. Kriebel Universite Paris XII, Creteil, France (J.V.) and Physiology Department, SUNY Health Science Center at Syracuse, Syracuse, New York (M.E.K.) time courses are observed and cannot be incorporated into the quantal theory (Kriebel et al., 1990). These MEPPs have neither a defined amplitude nor a defined rise time (Vautrin and Kriebel, 1991). The amplitude distributions of these MEPPs are skewed (see Kriebel, 1988 for review) and show a mode 10 times smaller than that of the bell shaped distribution (bell-MEPPs, see Fatt and Katz, 1952). Samples containing high incidence of skew-MEPPs showed also skewed rise time distributions with little or no change in the mode (Vautrin and Kriebel, 1991). Long rise time MEPPs (also called slow-MEPPs) were attributed to a different type of postsynaptic receptor or were suspected to have a remote origin. Electrotonic (passive) transient membrane potentials are distorted during their conduction (see Jack et al., 1975). Their amplitude decreases and their rise time increases with distance from the generation site. In contrast, Vautrin and Kriebel (1991) proposed that bell-, skew-, slow-, and giant MEPPs have the same origin. They attributed long rise time events to desynchronization of subunits that compose MEPPs (Kriebel et al., 1988). Bursts of releases generate postsynaptic responses that integrate into a single peak signal with the interval between individual presynaptic events is shorter than the rise time of individual postsynaptic responses (Vautrin and Kriebel, 1991). Extracellular electrodes detect local Key words: extracellular junctional potential, giant miniature junctional potentials (MEJPs) corresponding to miniature endplate potential, slow miniature end- an activity of the synapse which is restricted within a few plate potential, quantal subunit micrometers around the electrode (Del Castillo and Katz, 1956). INTRODUCTION
Miniature endplate potentials (MEPPs) with slow rising phase can be attributed either to burst of transmitter releases or to distortion of conduction from remote releasing sites. The spontaneous activity of neuromuscular junctions recorded extracellularly at mouse diaphragms using sharp electrodes was analyzed to test these two hypotheses. The miniature junctional potentials (MEJPs) frequencies observed intracellularly as compared to MEPP frequency measured intracellularly in controls indicate that most events recorded extracellularly are induced by the presence of the electrode. All types of MEPPs (bellMEPPs, skew-MEPPs, slow-, and giant MEPPs) previously described with intracellular recording methods (Vautrin and Kriebel, Neuroscience 41:71-88, 1991) were observed extracellularly and showed similar characteristics. This means that the presynaptic and postsynaptic zones that generate these synaptic events are restricted within areas of a few micrometers squared of synaptic contact. Long rise times of extracellularly recorded synaptic spontaneous events may be explained by multiple transmitter releases at intervals shorter than the rise time of individual events, which postsynaptic responses fuse into a single peak.
Quanta1 theory proposes that transmitter is released from motor nerve terminal by packets of standard size which generate postsynaptic events with identical amplitudes and time courses, the miniature endplate potentials (MEPPs) (Fatt and Katz, 1952; Del Castillo and Katz, 1954). Small variations in amplitude were attributed to measurement error or to random origin (Fatt and Katz, 1952). Yet, in many circumstances, such as development, increased release frequency or nerve terminal intoxication, MEPPs with readily different amplitudes and 0 1992 Wiley-Liss, Inc.
METHODS Spontaneous junctional activity was recorded extracellularly with sharp (3-5MCl) electrodes filled with KCl (3M) or CsCl (8M) at the mouse diaphragm using conventional voltage recording electrophysiological Address reprint requests to J. Vautrin, LNP, NINDS NIH, 36/2C02, 9000 Rockville Pike, Bethesda, MD 20892. Received March 22, 1991; revised June 10, 1991; accepted June 11, 1991.
Extracellular Slow Miniature .Junctional Potentials TABLE I. Mean Peak Amplitude and Rise Time of MEJP Series Successively Sampled Successive samples From
To
1 501
500 1000 1500
1001
Amplitude ?
2.28 2.33 2.05
SD ? ? &
.07 .07 .08
0.78 2 .02 0.79 _t .02 0.76 ? .02
"10-90% of peak amplitude.
technic. Both saline contained (in Mm): NaCl(l15), KC1 (2-3), CaC1, (2) and Tris or HEPES buffer adjusted to pH 7.4. Temperature was maintained at 24-26°C. No cholinesterase was used. Only recordings with 1% noiseto-signal ratio or less were considered for analysis. After recording on a magnetic tape recorder the synaptic signal was filtered at 3KHz and digitized at 30-50khz. MEJPs were detected and stored on a microcomputer using the SCAN program from Strathclyde University (Glasgow, UK). SCAN sampled continuously the synaptic signal and stored it in a temporary buffer. Each incoming sample was compared with a zero level and when the difference exceeded a threshold level a complete signal record (including a number of pre-trigger samples) was collected and stored for analysis. The zero level used for comparison was updated so as to constantly track slow drifts in the baseline. The threshold was set to 2-3% of the full scale which corresponded to 1% or less of the mean amplitude of the synaptic events analyzed. All events detected were checked individually. A transient negative slope on the rising phase was considered an objective criterion for a multiple releases event and these were rejected so that only MEJPs showing a rising phase with a constantly positive slope to the peak were included in the statistics.
RESULTS MEJPs were recorded extracellularly using sharp electrodes. The frequency observed was initially 5 to 20hz, which is higher than the frequency observed intracellularly at rest (0.1 to lhz) and indicates that most of the synaptic events detected were induced by the electrode. This frequency depended highly on the positioning of the electrode. Usually MEJP frequency declined with time but moderate pressure of the recording electrode made it possible to obtain stationary frequency for a few minutes (1-5) (Table I). There was generally a high incidence of small amplitude MEJPs and the distribution was then skewed (Fig. 1). A second mode or a hump was generally observed (at about 0.3mV-0.4mV on Fig. l), suggesting the underlying presence of a population of MEJPs which have a bell-shaped amplitude distribution.
I
80
Rise Timea ? SD
503
60 I
Q)
E 3 Z
40
20
0 0
1
2
Peak amplitude Fig. 1. Amplitude distribution of 2000 miniature junctional potentials (MEJPs) recorded with an extracellular electrode at a mouse diaphragm neuromuscular junction. The general shape of the distribution is skewed. There is a hump at 0.3-0.5mV.
Rise times measured between 10% and 90% of the peak formed a distribution with a mode near 0.6 ms (Fig. 2) The distribution was highly skewed and some rise times lasted several milliseconds. There was no discontinuity in the distribution between the shortest rise times (SRTs) and the longer rise times (LRTs). This is a relative classification since there was a continuum from SRTs and LRTs with no apparent limit. A typical scatter diagram of MEJP peak amplitude versus MEJP rise time is shown on Figure 3. SRTs increased slightly with MEJP amplitude. The minimum amplitude increased with the rise time. In some cases the positive correlation between amplitude and rise time was striking (not shown, also reported by Dunant, 1986). Such plots show a high density of points near the minimum rise time with amplitude from 2 to 6 times the minimum amplitude. Superpositions of following and unselected MEJPs are displayed in Figure 4. Selected LRT MEJPs are displayed in Figure 5B. There were often breaks on the rising phase of the LRT MEJPs. Breaks were never seen on SRT MEJP rise times. When SRT MEJPs were selected and superposed (Fig. 5A) all time courses were alike. The SRT MEPP amplitude is known to vary by steps (Kriebel and Gross, 1974; Vautrin, 1986) and SRT MEJPs clustered into fascicles corresponding to the preferred amplitudes and formed a banding pattern as observed for intracellular recording of MEPPs (Kriebel, 1988) or miniature endplate currents (Erxleben and Kriebel, 1988a, 1988b). The 80 LRT MEJPs of a series of 2000 were selected on Figure 5B to be compared with the SRT MEJPs of the same series (Fig. 5A). The time courses of the LRT MEJPs were more scattered. Despite the large
Vautrin and Kriebel
504
300
200 L
a,
-Q
E
3
=
100
0 0
1
I
I
1
2
3
4
Rise time (ms) Fig. 2. Distribution of 2000 MEJP rise times measured between 10 and 90% of the peak amplitude. Same events as in Figure 1. Note some very long rise times.
-. 2
5 Fig. 4. Two sets (A,B) of superposed unselected MEJPs. The vertical line shows the fastest rise times. The dots show longer rise times. Note the breaks on the time course of long rise time MEJPs. The early phase of the time courses is duplicated on the left to show that early phases are independent of the rise times.
E
Y
a, -0 3 c. .a
5
Y
a
l
0,
between the onset of the MEJPs and 0.68 ms after this onset. Figure 6 shows that early slope of the LRT MEJPs had a similar distribution to that of the slope of the whole population of MEJPs. LRT MEJPs could not be differentiated from SRT MEJPs during the earlier phase of the rise (about 500 ms).
n
DISCUSSION
0 0
3
6
Rise time (ms) Fig. 3. peak amplitude VS rise time of mouse diaphragm MEJPs. Same data as in Figures 1 and 2.
range of rise times and time courses some organization may be detected in MEJP time courses. Some LRT MEJPs were clustered into groups of similar rise time and eventually groups of similar amplitude. The inserts on the left of Figure 4 show that LRT MEJP group into the same fascicles as SRT MEJPs for most of the rising phase. The early slope of rising phases were measured
When a transient transmembrane potential propagates, its amplitude decreases and its rise time increases with distance (Jack et al., 1975). Therefore, transmitter releases occurring at some distance from the recording site generate synaptic signals with increased rise times and reduced amplitudes. Remote releases show amplitude and time course inversely correlated. LRT synaptic events, intracellularly (Vautrin and Kriebel, 1991; Vautrin, 1992) or extracellularly (as reported here) recorded, show no alteration of the early phase of rise as compared to faster events. They show no or a rather positive correlation between rise time and amplitude, which suggests that LRT events are not remote (see also Dunant, 1986).
Extracellular Slow Miniature Junctional Potentials
505
A
c, -
G
P
B
b
0
1.2
2.4
Rising slope (mV.s-’)
Time Fig. 5. A: Selection of 40 short rise time MEJPs. The vertical line helps to compare the rise times. The black dot shows the smallest MEJP (probably a single subunit, see text). The star shows a secondary event on a falling phase; it is the size of the subunit. B: Selection of the 80 MEJPs with the longest rise times of a parent series of 2000. The parent series is the same as in A . Although there is a wide range of amplitudes and time courses, time courses cluster into groups of similar rise time or/and similar amplitude.
Del Castillo and Katz (1956) estimated that transmitter releases occurring 20pm apart from the extracellular electrode cannot be resolved. SRT and LRT events reported here were recorded with an extracellular electrode, which means that these currents were probably generated by ACh sensitive membrane patches that cannot be more distant than a few micrometers from the electrode and that the sites that released the corresponding ACh packets are located within a similar distance (Wathey et al., 1979). Wernig (1976) demonstrated that the amplitude of a synaptic signal generated 3-4pm from the extracellular electrode is greatly reduced. As MEPPs (Kriebel and Gross, 1974) and miniature endplate currents (MEPCs) (Erxleben and Kriebel, 1988b), SRT MEJPs show preferred amplitudes. This cannot be explained by attenuation of currents generated at random distance from the
Fig. 6 . Distribution of slope of MEJPs measured between 0 and 0.68ms of the rising phase. Dark histobars: all 2000 MEJPs of the series. White histobars (2 different scales): 80 MEJPs with the longest rise times. SRT MEJPs and LRT MEJPs show the same early slope distribution.
electrode. It would make it impossible to recognize the preferred amplitudes observed intracellularly and suggests that most of these events happened close to the tip of the electrode. When superposed, preferred amplitudes of SRT MEJPs group MEJP time courses into fascicles. Early phases of SRT and LRT MEJPs organize into the same fascicles. Attenuation of remote events would disturb this organization. It is known that up to 50% of the spontaneous activity of the terminal may occur near an extracellular electrode (Del Castillo and Katz, 1956; Wernig, 1976; Bennett and Lavidis, 1982; see also Robitaille and Tremblay, 1987). Most of extracellularly detected events are probably induced by the mechanical pressure of the electrode tip itself (see Boroff et al., 1974) and/or from the local increase of osmolarity induced by leakage of the highly concentrated saline from the electrode tip. Extracellularly recorded MEJPs show the same minimum rise time and mode of rise time (0.6 ms) as for MEPC recorded intracellularly (Vautrin and Kriebel, 1991). MEJPs rise time distribution is skewed as for MEPPs or MEPCs that form a skewed amplitude distribution. There is a continuum of rise times from SRTs and LRTs and it is difficult to attribute the very long rise times (which seem to have no maximum limit) to differ-
506
Vautrin and Kriebel
ent properties of the postsynaptic apparatus. Therefore, a presynaptic origin of the LRTs appears more likely. Since SRT MEJPs and LRT MEJPs show the same distribution of early slope and organize into the same fascicles for 200-500 microseconds, there is no evidence that the process generating the initial part of the LRT MEJPs is different from the process that generates SRT MEJPs. After that delay, additional current is necessary to explain the time course of LRT MEJPs as compared to SRT MEJP time course. The same observations were made on MEPCs recorded intracellularly (Vautrin and Kriebel, 1991). There is no MEJP, MEPP, or MEPC that cannot be explained by transmitter subunits released in different numbers and at different times (Matteson et al., 1981; Kriebel et al., 1988). Although LRT MEJPs show a wide range of time courses, many of them cluster into groups of similar time course (see fascicles of LRT MEJP in Fig. 5B). This reveals the deterministic nature of the process that releases bursts of transmitter packets (see also Vautrin, 1988; Kriebel et al., 1990; Vautrin and Kriebel, 1991). The volume of the synaptic vesicles is thought to determine the size of the transmitter packets. Vesicle volume distribution is bell shaped and fails to explain changes in the size of the transmitter packet (Kriebel and Gross, 1974; Pecot-Dechavassine and Molgo, 1982; Kriebel et al., 1990; Van der Kloot, 1991). Vesicles responsible for LRT MEJPs and MEPPs may be more labile and thus not resolved by standard fixation protocols. Vesicles may contain only a sub-quantum of transmitter and synchronous exocytoses, which could explain the variations in MEJP amplitude and time course (Kriebel and Gross, 1974; Wernig and Stirner, 1977). The amount of transmitter in a packet may also not be determined by the volume of synaptic vesicles (Kriebel et al., 1990; Vautrin et al., 1992).
ACKNOWLEDGMENTS John Dempster from Strathclyde University provided the computer program SCAN. This work was supported by ‘ ‘Association Frangaise contre les Myopathies”, NS 25683-03 and NINDS, NIH.
REFERENCES Bennett MR, Lavidis NA (1982): Variation in quantal secretion at different release sites along developing and mature motor terminal branches. Dev Brain Res S:1-9. Boroff DA, Del Castillo J, Evoy WH, Steinhardt RA (1974): Observations on the action of type A botulinum toxin on frog neuromuscular junctions. J Physiol 240:227-253. Del Castillo J, Katz B (1954): Quanta1 components of the end-plate potentials. J Physiol 124:560-573.
Del Castillo J, Katz B (1956): Localization of active spots within the neuromuscular junction of the frog. J Physiol 1 3 2 6 3 0 4 4 9 , Dunant Y (1986): On the mechanism of acetylcholine release. Prog Neurobiol 2655-92. Erxleben C, Kriebel ME (1988a): Characteristics of spontaneous miniature endplate currents at the mouse neuromuscular junction. J Physiol 400:645-658. Erxleben C, Kriebel ME (1988b): Subunit composition of the spontaneous miniature endplate currents at the mouse neuromuscular junction. J Physiol 400:659-676. Fatt P, Katz B (1952): Spontaneous subthreshold activity at motor nerve endings. J Physiol 117:109-128. Jack JBB, Noble D, Tsien RW (1975): “The Spread of Current in Excitable Cells. Oxford: Clarendon Press. Kriebel ME (1988): The neuromuscular junction. In Whittaker, (Ed) “Handbook of Experimental Pharmacology. ” Springer-Verlag, Berlin-Heidelberg: 86537-566. Kriebel ME, Gross CE (1974): Multimodal distribution of frog miniature end-plate potentials in adult, denervated, and tadpole leg muscle. J Gene Physiol 64:85-103. Kriebel ME, Hanna R, Muniak C (1986): Synaptic vesicle diameter and synaptic cleft width at the mouse diaphragm in neonates and adults. Dev Brain Res 27:19-29. Kriebel ME, Vautrin J, Llados F (1988): A unifying theory for the basis of the different classes of quanta found in the neuromuscular junction based on sub-units. Am Neurosci Ann Meet, Toronto. Kriebel ME, Vautrin J, Holsapple J (1990): Transmitter release: Prepackaging and random mechanism or dynamic and deterministic process. Brain Res Review 15:167-178. Matteson DR, Kriebel ME, Llados F (1981): A statistical model indicates that miniature end-plate potentials and unitary evoked end-plate potentials are composed of subunits. J Theor Biol 90~337-363. Pecot-Dechavassine M, Molgo J (1982): Attempt to detect morphological correlates for the “giant” miniature end-plate potentials induced by 4-aminoquinoline. Biol Cell 46:93-96. Robitaille R, Tremblay JP (1987): Non-uniform release at the frog neuromuscular junction: evidence of morphological and physiological plasticity. Brain Res Rev 1295-1 16. Van der Kloot W (1991): The regulation of quantal size. Prog Neurobiol 36:93-130. Vautrin J (1986): Subunits in quantal transmission at the mouse neuromuscular junction: test of peak intervals in amplitude distributions. J Theor Biol 120363-370. Vautrin J (1988): Classes of unitary evoked responses at the vertebrate neuromuscular junction. Brain Res 438:304-306. Vautrin J (1992): Miniature endplate potentials induced by ammonium chloride, hypertonic shock, and botulinum toxin. J Neurosci Res 31:318-326. Vautrin J, Kriebel ME (1991): Characteristics of slow miniature endplate currents show a subunit composition. Neuroscience 41: 71-88. Vautrin J, Kriebel ME, Holsapple J (1992): Further evidence for the dynamic formation of transmitter quanta at the neuromuscular junction. J Neurosci Res, in press. Wathey JC, Nass WM, Lester HA (1979): Numerical reconstruction of the quantal event at nicotinic synapses. Biophys 27:145. Wernig A (1976): Localization of active sites in the neuromuscular junction of the frog. Brain Res 118:63-72. Wernig A, Stirner H (1977): Quantum amplitude distributions point to junctional unity of the synaptic “active zone”. Nature 269: 820-822.
Focal, Extracellular Recording of Slow Miniature Junctional Potentials at the Mouse Neuromuscular Junction J. Vautrin and M.E. Kriebel Universite Paris XII, Creteil, France (J.V.) and Physiology Department, SUNY Health Science Center at Syracuse, Syracuse, New York (M.E.K.) time courses are observed and cannot be incorporated into the quantal theory (Kriebel et al., 1990). These MEPPs have neither a defined amplitude nor a defined rise time (Vautrin and Kriebel, 1991). The amplitude distributions of these MEPPs are skewed (see Kriebel, 1988 for review) and show a mode 10 times smaller than that of the bell shaped distribution (bell-MEPPs, see Fatt and Katz, 1952). Samples containing high incidence of skew-MEPPs showed also skewed rise time distributions with little or no change in the mode (Vautrin and Kriebel, 1991). Long rise time MEPPs (also called slow-MEPPs) were attributed to a different type of postsynaptic receptor or were suspected to have a remote origin. Electrotonic (passive) transient membrane potentials are distorted during their conduction (see Jack et al., 1975). Their amplitude decreases and their rise time increases with distance from the generation site. In contrast, Vautrin and Kriebel (1991) proposed that bell-, skew-, slow-, and giant MEPPs have the same origin. They attributed long rise time events to desynchronization of subunits that compose MEPPs (Kriebel et al., 1988). Bursts of releases generate postsynaptic responses that integrate into a single peak signal with the interval between individual presynaptic events is shorter than the rise time of individual postsynaptic responses (Vautrin and Kriebel, 1991). Extracellular electrodes detect local Key words: extracellular junctional potential, giant miniature junctional potentials (MEJPs) corresponding to miniature endplate potential, slow miniature end- an activity of the synapse which is restricted within a few plate potential, quantal subunit micrometers around the electrode (Del Castillo and Katz, 1956). INTRODUCTION
Miniature endplate potentials (MEPPs) with slow rising phase can be attributed either to burst of transmitter releases or to distortion of conduction from remote releasing sites. The spontaneous activity of neuromuscular junctions recorded extracellularly at mouse diaphragms using sharp electrodes was analyzed to test these two hypotheses. The miniature junctional potentials (MEJPs) frequencies observed intracellularly as compared to MEPP frequency measured intracellularly in controls indicate that most events recorded extracellularly are induced by the presence of the electrode. All types of MEPPs (bellMEPPs, skew-MEPPs, slow-, and giant MEPPs) previously described with intracellular recording methods (Vautrin and Kriebel, Neuroscience 41:71-88, 1991) were observed extracellularly and showed similar characteristics. This means that the presynaptic and postsynaptic zones that generate these synaptic events are restricted within areas of a few micrometers squared of synaptic contact. Long rise times of extracellularly recorded synaptic spontaneous events may be explained by multiple transmitter releases at intervals shorter than the rise time of individual events, which postsynaptic responses fuse into a single peak.
Quanta1 theory proposes that transmitter is released from motor nerve terminal by packets of standard size which generate postsynaptic events with identical amplitudes and time courses, the miniature endplate potentials (MEPPs) (Fatt and Katz, 1952; Del Castillo and Katz, 1954). Small variations in amplitude were attributed to measurement error or to random origin (Fatt and Katz, 1952). Yet, in many circumstances, such as development, increased release frequency or nerve terminal intoxication, MEPPs with readily different amplitudes and 0 1992 Wiley-Liss, Inc.
METHODS Spontaneous junctional activity was recorded extracellularly with sharp (3-5MCl) electrodes filled with KCl (3M) or CsCl (8M) at the mouse diaphragm using conventional voltage recording electrophysiological Address reprint requests to J. Vautrin, LNP, NINDS NIH, 36/2C02, 9000 Rockville Pike, Bethesda, MD 20892. Received March 22, 1991; revised June 10, 1991; accepted June 11, 1991.
Extracellular Slow Miniature .Junctional Potentials TABLE I. Mean Peak Amplitude and Rise Time of MEJP Series Successively Sampled Successive samples From
To
1 501
500 1000 1500
1001
Amplitude ?
2.28 2.33 2.05
SD ? ? &
.07 .07 .08
0.78 2 .02 0.79 _t .02 0.76 ? .02
"10-90% of peak amplitude.
technic. Both saline contained (in Mm): NaCl(l15), KC1 (2-3), CaC1, (2) and Tris or HEPES buffer adjusted to pH 7.4. Temperature was maintained at 24-26°C. No cholinesterase was used. Only recordings with 1% noiseto-signal ratio or less were considered for analysis. After recording on a magnetic tape recorder the synaptic signal was filtered at 3KHz and digitized at 30-50khz. MEJPs were detected and stored on a microcomputer using the SCAN program from Strathclyde University (Glasgow, UK). SCAN sampled continuously the synaptic signal and stored it in a temporary buffer. Each incoming sample was compared with a zero level and when the difference exceeded a threshold level a complete signal record (including a number of pre-trigger samples) was collected and stored for analysis. The zero level used for comparison was updated so as to constantly track slow drifts in the baseline. The threshold was set to 2-3% of the full scale which corresponded to 1% or less of the mean amplitude of the synaptic events analyzed. All events detected were checked individually. A transient negative slope on the rising phase was considered an objective criterion for a multiple releases event and these were rejected so that only MEJPs showing a rising phase with a constantly positive slope to the peak were included in the statistics.
RESULTS MEJPs were recorded extracellularly using sharp electrodes. The frequency observed was initially 5 to 20hz, which is higher than the frequency observed intracellularly at rest (0.1 to lhz) and indicates that most of the synaptic events detected were induced by the electrode. This frequency depended highly on the positioning of the electrode. Usually MEJP frequency declined with time but moderate pressure of the recording electrode made it possible to obtain stationary frequency for a few minutes (1-5) (Table I). There was generally a high incidence of small amplitude MEJPs and the distribution was then skewed (Fig. 1). A second mode or a hump was generally observed (at about 0.3mV-0.4mV on Fig. l), suggesting the underlying presence of a population of MEJPs which have a bell-shaped amplitude distribution.
I
80
Rise Timea ? SD
503
60 I
Q)
E 3 Z
40
20
0 0
1
2
Peak amplitude Fig. 1. Amplitude distribution of 2000 miniature junctional potentials (MEJPs) recorded with an extracellular electrode at a mouse diaphragm neuromuscular junction. The general shape of the distribution is skewed. There is a hump at 0.3-0.5mV.
Rise times measured between 10% and 90% of the peak formed a distribution with a mode near 0.6 ms (Fig. 2) The distribution was highly skewed and some rise times lasted several milliseconds. There was no discontinuity in the distribution between the shortest rise times (SRTs) and the longer rise times (LRTs). This is a relative classification since there was a continuum from SRTs and LRTs with no apparent limit. A typical scatter diagram of MEJP peak amplitude versus MEJP rise time is shown on Figure 3. SRTs increased slightly with MEJP amplitude. The minimum amplitude increased with the rise time. In some cases the positive correlation between amplitude and rise time was striking (not shown, also reported by Dunant, 1986). Such plots show a high density of points near the minimum rise time with amplitude from 2 to 6 times the minimum amplitude. Superpositions of following and unselected MEJPs are displayed in Figure 4. Selected LRT MEJPs are displayed in Figure 5B. There were often breaks on the rising phase of the LRT MEJPs. Breaks were never seen on SRT MEJP rise times. When SRT MEJPs were selected and superposed (Fig. 5A) all time courses were alike. The SRT MEPP amplitude is known to vary by steps (Kriebel and Gross, 1974; Vautrin, 1986) and SRT MEJPs clustered into fascicles corresponding to the preferred amplitudes and formed a banding pattern as observed for intracellular recording of MEPPs (Kriebel, 1988) or miniature endplate currents (Erxleben and Kriebel, 1988a, 1988b). The 80 LRT MEJPs of a series of 2000 were selected on Figure 5B to be compared with the SRT MEJPs of the same series (Fig. 5A). The time courses of the LRT MEJPs were more scattered. Despite the large
Vautrin and Kriebel
504
300
200 L
a,
-Q
E
3
=
100
0 0
1
I
I
1
2
3
4
Rise time (ms) Fig. 2. Distribution of 2000 MEJP rise times measured between 10 and 90% of the peak amplitude. Same events as in Figure 1. Note some very long rise times.
-. 2
5 Fig. 4. Two sets (A,B) of superposed unselected MEJPs. The vertical line shows the fastest rise times. The dots show longer rise times. Note the breaks on the time course of long rise time MEJPs. The early phase of the time courses is duplicated on the left to show that early phases are independent of the rise times.
E
Y
a, -0 3 c. .a
5
Y
a
l
0,
between the onset of the MEJPs and 0.68 ms after this onset. Figure 6 shows that early slope of the LRT MEJPs had a similar distribution to that of the slope of the whole population of MEJPs. LRT MEJPs could not be differentiated from SRT MEJPs during the earlier phase of the rise (about 500 ms).
n
DISCUSSION
0 0
3
6
Rise time (ms) Fig. 3. peak amplitude VS rise time of mouse diaphragm MEJPs. Same data as in Figures 1 and 2.
range of rise times and time courses some organization may be detected in MEJP time courses. Some LRT MEJPs were clustered into groups of similar rise time and eventually groups of similar amplitude. The inserts on the left of Figure 4 show that LRT MEJP group into the same fascicles as SRT MEJPs for most of the rising phase. The early slope of rising phases were measured
When a transient transmembrane potential propagates, its amplitude decreases and its rise time increases with distance (Jack et al., 1975). Therefore, transmitter releases occurring at some distance from the recording site generate synaptic signals with increased rise times and reduced amplitudes. Remote releases show amplitude and time course inversely correlated. LRT synaptic events, intracellularly (Vautrin and Kriebel, 1991; Vautrin, 1992) or extracellularly (as reported here) recorded, show no alteration of the early phase of rise as compared to faster events. They show no or a rather positive correlation between rise time and amplitude, which suggests that LRT events are not remote (see also Dunant, 1986).
Extracellular Slow Miniature Junctional Potentials
505
A
c, -
G
P
B
b
0
1.2
2.4
Rising slope (mV.s-’)
Time Fig. 5. A: Selection of 40 short rise time MEJPs. The vertical line helps to compare the rise times. The black dot shows the smallest MEJP (probably a single subunit, see text). The star shows a secondary event on a falling phase; it is the size of the subunit. B: Selection of the 80 MEJPs with the longest rise times of a parent series of 2000. The parent series is the same as in A . Although there is a wide range of amplitudes and time courses, time courses cluster into groups of similar rise time or/and similar amplitude.
Del Castillo and Katz (1956) estimated that transmitter releases occurring 20pm apart from the extracellular electrode cannot be resolved. SRT and LRT events reported here were recorded with an extracellular electrode, which means that these currents were probably generated by ACh sensitive membrane patches that cannot be more distant than a few micrometers from the electrode and that the sites that released the corresponding ACh packets are located within a similar distance (Wathey et al., 1979). Wernig (1976) demonstrated that the amplitude of a synaptic signal generated 3-4pm from the extracellular electrode is greatly reduced. As MEPPs (Kriebel and Gross, 1974) and miniature endplate currents (MEPCs) (Erxleben and Kriebel, 1988b), SRT MEJPs show preferred amplitudes. This cannot be explained by attenuation of currents generated at random distance from the
Fig. 6 . Distribution of slope of MEJPs measured between 0 and 0.68ms of the rising phase. Dark histobars: all 2000 MEJPs of the series. White histobars (2 different scales): 80 MEJPs with the longest rise times. SRT MEJPs and LRT MEJPs show the same early slope distribution.
electrode. It would make it impossible to recognize the preferred amplitudes observed intracellularly and suggests that most of these events happened close to the tip of the electrode. When superposed, preferred amplitudes of SRT MEJPs group MEJP time courses into fascicles. Early phases of SRT and LRT MEJPs organize into the same fascicles. Attenuation of remote events would disturb this organization. It is known that up to 50% of the spontaneous activity of the terminal may occur near an extracellular electrode (Del Castillo and Katz, 1956; Wernig, 1976; Bennett and Lavidis, 1982; see also Robitaille and Tremblay, 1987). Most of extracellularly detected events are probably induced by the mechanical pressure of the electrode tip itself (see Boroff et al., 1974) and/or from the local increase of osmolarity induced by leakage of the highly concentrated saline from the electrode tip. Extracellularly recorded MEJPs show the same minimum rise time and mode of rise time (0.6 ms) as for MEPC recorded intracellularly (Vautrin and Kriebel, 1991). MEJPs rise time distribution is skewed as for MEPPs or MEPCs that form a skewed amplitude distribution. There is a continuum of rise times from SRTs and LRTs and it is difficult to attribute the very long rise times (which seem to have no maximum limit) to differ-
506
Vautrin and Kriebel
ent properties of the postsynaptic apparatus. Therefore, a presynaptic origin of the LRTs appears more likely. Since SRT MEJPs and LRT MEJPs show the same distribution of early slope and organize into the same fascicles for 200-500 microseconds, there is no evidence that the process generating the initial part of the LRT MEJPs is different from the process that generates SRT MEJPs. After that delay, additional current is necessary to explain the time course of LRT MEJPs as compared to SRT MEJP time course. The same observations were made on MEPCs recorded intracellularly (Vautrin and Kriebel, 1991). There is no MEJP, MEPP, or MEPC that cannot be explained by transmitter subunits released in different numbers and at different times (Matteson et al., 1981; Kriebel et al., 1988). Although LRT MEJPs show a wide range of time courses, many of them cluster into groups of similar time course (see fascicles of LRT MEJP in Fig. 5B). This reveals the deterministic nature of the process that releases bursts of transmitter packets (see also Vautrin, 1988; Kriebel et al., 1990; Vautrin and Kriebel, 1991). The volume of the synaptic vesicles is thought to determine the size of the transmitter packets. Vesicle volume distribution is bell shaped and fails to explain changes in the size of the transmitter packet (Kriebel and Gross, 1974; Pecot-Dechavassine and Molgo, 1982; Kriebel et al., 1990; Van der Kloot, 1991). Vesicles responsible for LRT MEJPs and MEPPs may be more labile and thus not resolved by standard fixation protocols. Vesicles may contain only a sub-quantum of transmitter and synchronous exocytoses, which could explain the variations in MEJP amplitude and time course (Kriebel and Gross, 1974; Wernig and Stirner, 1977). The amount of transmitter in a packet may also not be determined by the volume of synaptic vesicles (Kriebel et al., 1990; Vautrin et al., 1992).
ACKNOWLEDGMENTS John Dempster from Strathclyde University provided the computer program SCAN. This work was supported by ‘ ‘Association Frangaise contre les Myopathies”, NS 25683-03 and NINDS, NIH.
REFERENCES Bennett MR, Lavidis NA (1982): Variation in quantal secretion at different release sites along developing and mature motor terminal branches. Dev Brain Res S:1-9. Boroff DA, Del Castillo J, Evoy WH, Steinhardt RA (1974): Observations on the action of type A botulinum toxin on frog neuromuscular junctions. J Physiol 240:227-253. Del Castillo J, Katz B (1954): Quanta1 components of the end-plate potentials. J Physiol 124:560-573.
Del Castillo J, Katz B (1956): Localization of active spots within the neuromuscular junction of the frog. J Physiol 1 3 2 6 3 0 4 4 9 , Dunant Y (1986): On the mechanism of acetylcholine release. Prog Neurobiol 2655-92. Erxleben C, Kriebel ME (1988a): Characteristics of spontaneous miniature endplate currents at the mouse neuromuscular junction. J Physiol 400:645-658. Erxleben C, Kriebel ME (1988b): Subunit composition of the spontaneous miniature endplate currents at the mouse neuromuscular junction. J Physiol 400:659-676. Fatt P, Katz B (1952): Spontaneous subthreshold activity at motor nerve endings. J Physiol 117:109-128. Jack JBB, Noble D, Tsien RW (1975): “The Spread of Current in Excitable Cells. Oxford: Clarendon Press. Kriebel ME (1988): The neuromuscular junction. In Whittaker, (Ed) “Handbook of Experimental Pharmacology. ” Springer-Verlag, Berlin-Heidelberg: 86537-566. Kriebel ME, Gross CE (1974): Multimodal distribution of frog miniature end-plate potentials in adult, denervated, and tadpole leg muscle. J Gene Physiol 64:85-103. Kriebel ME, Hanna R, Muniak C (1986): Synaptic vesicle diameter and synaptic cleft width at the mouse diaphragm in neonates and adults. Dev Brain Res 27:19-29. Kriebel ME, Vautrin J, Llados F (1988): A unifying theory for the basis of the different classes of quanta found in the neuromuscular junction based on sub-units. Am Neurosci Ann Meet, Toronto. Kriebel ME, Vautrin J, Holsapple J (1990): Transmitter release: Prepackaging and random mechanism or dynamic and deterministic process. Brain Res Review 15:167-178. Matteson DR, Kriebel ME, Llados F (1981): A statistical model indicates that miniature end-plate potentials and unitary evoked end-plate potentials are composed of subunits. J Theor Biol 90~337-363. Pecot-Dechavassine M, Molgo J (1982): Attempt to detect morphological correlates for the “giant” miniature end-plate potentials induced by 4-aminoquinoline. Biol Cell 46:93-96. Robitaille R, Tremblay JP (1987): Non-uniform release at the frog neuromuscular junction: evidence of morphological and physiological plasticity. Brain Res Rev 1295-1 16. Van der Kloot W (1991): The regulation of quantal size. Prog Neurobiol 36:93-130. Vautrin J (1986): Subunits in quantal transmission at the mouse neuromuscular junction: test of peak intervals in amplitude distributions. J Theor Biol 120363-370. Vautrin J (1988): Classes of unitary evoked responses at the vertebrate neuromuscular junction. Brain Res 438:304-306. Vautrin J (1992): Miniature endplate potentials induced by ammonium chloride, hypertonic shock, and botulinum toxin. J Neurosci Res 31:318-326. Vautrin J, Kriebel ME (1991): Characteristics of slow miniature endplate currents show a subunit composition. Neuroscience 41: 71-88. Vautrin J, Kriebel ME, Holsapple J (1992): Further evidence for the dynamic formation of transmitter quanta at the neuromuscular junction. J Neurosci Res, in press. Wathey JC, Nass WM, Lester HA (1979): Numerical reconstruction of the quantal event at nicotinic synapses. Biophys 27:145. Wernig A (1976): Localization of active sites in the neuromuscular junction of the frog. Brain Res 118:63-72. Wernig A, Stirner H (1977): Quantum amplitude distributions point to junctional unity of the synaptic “active zone”. Nature 269: 820-822.
