Brain Research, 125 (1977) 123-140
123
© Elsevier/North-Holland Biomedical Press, Amsterdam- Printed in The Netherlands
O S C I L L A T I O N OF A C E T Y L C H O L I N E D U R I N G THE TORPEDO ELECTRIC ORGAN
NERVE
A C T I V I T Y IN
Y. DUNANT, M. ISRAEL, B. LESBATS and R. MANARANCHE Laboratoire de Neurobiologie Cellulaire, Neurochimie, C.N.R.S., 91190 Gif sar-Yvette (France) and (KD.) Ddpartement de Pharmacologie, Ecole de Mddecine, 20, 1211 GenOve 4 (Switzerland)
(Accepted July 27th, 1976)
SUMMARY The amount of transmitter in the electric organ of Torpedo was measured with a time resolution of 1 sec in the course of stimulation. In parallel, the modifications of the electrophysiological response were analysed by determining the conductance increase (AG) apd the electromotive force of electroplaques. Large changes in the level of total acetylcholine (ACh) were seen during stimulation. These changes were twofold: a slow wave and, superimposed on it, a rapid oscillation. The slow wave raised total ACh to the initial level, or evet~ higher. It was probably related to modifications in the amount of ACh released since it corresponded to characteristic inflections in the evolution of the AG curve. The slow wave and this physiological parameter were similarly affected when the experiments were performed at a reduced temperature. The rapid oscillation had an amplitude of about 20 4 0 ~ of the total ACh. It was undamped and its period was 4-5 sec. In contrast to the slow wave, no clear physiological change associated with the rapid oscillation has been observed. The slow wave and rapid oscillation occurred in the 'free pool' of ACh, whereas bound ACh, the fraction associated with synaptic vesicles, was not affected by these changes. A dynamic description of synaptic activity is proposed. The content of 'free' ACh is used and renewed completely after a few tens of impulses, so that transmission seems to imply the continual recycling of the same pool of transmitter rather than utilization of a large preloaded store. The release processes must then be integrated in rapid metabolic loops.
INTRODUCTION A joint electrophysiological and biochemical description is presented in this work, which is a dynamic analysis of transmission in the electric organ of the fish,
124 Torpedo marmorata. This very homogeneous tissue consists of a great number of prisms, arranged side by side, each being built up of about 500 superposed electroplaques, which are covered on their ventral surface by a rich network of nerve terminals. The cholinergic nature of transmission has been demonstrated in this organlS, ~6, When the nerve is active, the terminals release acetylcholine (ACh), which acts on receptors situated at the ventral membrane of the plates. This results in a large number of synchronized postjunctional potentials which sum to generate a strong discharge. positive towards the dorsal face of the fish. The released transmitter is lhen rapidly hydrolyzed, and choline and acetate ~5 taken up into the nerve terminals where ACh is resynthesized. Since hydrolysis of the released transmitter occurs in a matter of msec, all the ACh found in the tissue when it is quenched for extraction must be regarded as intracellular and most probably intraterminal ')1. Advantage has been taken of this to measure nerve terminal ACh as a function of time during stimulation. In preliminary experiments, the level of transmitter was found to change quickly; it was therefore necessary to improve the rapidity of the biochemical methods, which finally reached a time resolution of 1 sec. The changes of tissue ACh have been correlated, when possible, with modifications of the electrical parameters of the response. The peculiar arrangement of the electric organ allowed us to record not only the amplitude of the discharge, which is an overall measurement of synaptic efficiency, but also the inversion potential and the conductance change of the active tissue. The inversion potential provides useful information on the ionic electromotive force acting on the postsynaptic membrane; the conductance change is expected to be a function of the number of receptors which are activated in the membrane at a given time. Most of the present work has been carried out by analyzing the total ACh of the tissue. Experiments were also performed to determinethe compartment of ACh modified during stimulation. Bound ACh is that which remains after the tissue has been homogenized. Most of it is present in synaptic vesicles which can be isolated and purified by fractionation methods2e,~':3,":L 'Free ACh' is the part, about half of the total, which is hydrolyzed by esterases when the tissue is disrupted. The term 'free' must be considered as operational since the subcellular localization of this compartment has not yet been elucidated (see review in ref. 28). Previous work has shown that stimulation increases the turnover of 'free ACh" but not that of bound ACh s,9. Partial descriptions of these electrophysiological and biochemical lindings have been published as short communicationsT,W,t:3,'-'L METHODS Animals Torpedo marmorata were supplied by the 'Station de Biologie Marine', Arcachon, France. Most of the fish were females of 30-60 cm length. They were kept in sea water at about 15 C . Incubation medium When excised, the tissue was kept in an elasmobranch saline solution containing
125 (in m M ) : NaC1, 280; KC1, 3.0; CaClo., 3.4; MgClz, 1.8; NaHCOa, 5; NaH~PO4, 1.2; urea, 300; saccharose, 100; and glucose 5.5. This medium was oxygenated and its pH kept between 7.1 and 7.4. A good conservation of the tissue during several hours was shown by morphological, biochemical and physiological controls. When required, drugs or labelled precursors were added to this solution. Unless mentioned, experiments were carried out at room temperature.
Electrophysiological methods Stimuhttion by the nerve. The Torpedo was anaesthetized by tricaine methane sulphonate (MS 222, Sandoz) at a concentration of I g/3 1 sea water. Controls showed that this short anaesthesia has no effect on the efficiency of nerve-electroplaque transmission. Fragments of electric organ were carefully excised with their nerve intact. They usually contained a few prisms of electroplaques. After dissection the tissue was allowed to recover in the saline medium for at least I h. Stimulation was then applied to the nerve and the response potential recorded between the ventral and the dc, rsal ends of the prisms. Field stimulation of the electrogenic tissue. When 2 or 3 dissected prisms of electric tissue are submitted to a brief (1 msec) electric shock, the response discharge recorded is separated from the stimulus artefact by an irreducible latency of 2-3 msec. Nerve degeneration or treatment with curare completely abolishes this response 3. These observations and microelectrode investigations 4 demonstrated that Torpedo electroplaques are not able to generate and propagate action potentials. Field stimulation only excites nerves and nerve terminals in the tissue directly. The electrical discharge is an indirect response of electroplaques to ACh released by nerve terminals; more precisely, it is the sum of individual postjunctional potentials. Field stimulation was found to be very convenient in experiments where the tissue had to be incubated in solutions containing drugs or radioactive precursors, or when a high resolution in time was wanted. Fragments of electric organ were excised and incubated in the saline medium. Care was taken not to damage the electroplaques by cutting through the prisms longitudinally. The pieces were weighed and set on a piece of nylon cloth between the two stimulating electrodes which were parallel to the prisms. Stimulation is more efficient ill this way since the field can easily reach the nerves at the edges of the electroplaques and between them iv. The set-up was either immersed or superfused by a continuous flow of the saline solution. Supramaximal shocks were applied and the responses recorded between the two extremities of the prism. When the prisms were kept at room temperature and their response tested at large time intervals, a slight increase in the voltage amplitude was observed during the 2 h following dissection; the response then remained at a very stable level for more than 24 h.
Determination of inversion potential and conductance change during stimulation. These parameters were measured using a method recently described v, based on an observation performed on the electric organ of Torpedo nobiliana 4. A piece of electric organ was dissected with its nerve. The ventral and dorsal ends of the fragment were placed between two Ag-AgCI.~ plates. The tissue was stimulated through the
126 nerve and produced the usual response of amplitude, V. This discharge was modified by applying brief polarizing currents to the whole piece of tissue through the silver plates. The electromotive force of the fragment was measured under these conditions by giving an equal but opposite back-electromotive force. This was seen by the suppression of the voltage change of the response. This force has been called 'inversion potential' (E~nv) since the polarity of the discharge was reversed when stronger external potentials were applied. The conductance change (AG) was the difference between the conductance of the tissue measured at the peak of the discharge and that measured in the resting state. These parameters are relatively easy to determine in the electrogenic organ of Torpedo since the voltage changes are linearly related to the applied currents not only at rest but also during activity. Isolated prisms could also be submitted to this technique; they were stimulated through the silver plates by a brief field shock preceding the application of the polarizing pulse.
Biochemical methods' Sample~. Tissue samples were incubated for at [east l h in the physiological solution. A group of samples were then stimulated and, at different times, one of them was dropped into the extraction solution. In the best conditions, the latency of the transfer was reduced to about 0.25 sec. Then new groups were taken and stimulated in such a manner that a time resolution of I sec was achieved. When, after the first experiments, it appeared that ACh changes were periodical, the sampling was randomized to avoid any systematic periodicity in the protocol. Two different rapid quenching procedures were used. For the first, the samples were dropped into liquid nitrogen or isopentane cooled by liquid nitrogen. After freezing, the sample was powdered and the powder extracted in 5 ml of 5 o~ trichloracetic acid (TCA). The TCA was then removed by washing with ether. In the second method, the samples were dropped into 5 ml of hot HC1 (87 '~C, pH 2.7) and left for 7 rain. At the end of this time, the pH had risen to about 5. The second procedure was more convenient and gave similar results to the first. These procedures extract the total ACh content of the tissue. To determine the amount of bound or vesicular ACh, the samples were dropped into the physiological solution at about 5 '~C and immediately homogenized. TCA was then added to the tube without delay (5 ~/~ifinal concentration). Alternatively, an equal volume of N/50 HCI at 87 °C was added, and left for 7 rain. Estimation of ACh. The ACh of the extracted samples was, for most of the experiments reported here, estimated with the conventional frog rectus method. In order to confirm the results, and to facilitate the analysis of a large number of samples, a radiochemical procedure to measure ACh was developed 1°, and used in some experiments. Labelling of the acetyl moiety of ACh. [l-14C]acetate was added to the incubation medium when necessary, and the radioactive acetylcholine synthesized by the tissue was extracted and separated frorn the acetate as previously described2L Other estimations. Choline acetyltransferase was assayed by the method described by Fonnum 18 and proteins were measured according to Lowry et ai. zv,
127 RESULTS
Conductance change and inversion potential in the course of repetitive stimulation It has been known for a long time that repeated excitation of the electric fish results in a 'fatigue' of the discharge. Fig. 1 shows a detailed analysis of this phenomenon. The tissue was stimulated through its nerve at a frequency of 5/sec. The electrical response discharge (V) was initially 20 V, and its amplitude fell to half after about 30 sec stimulation. There then occurred a 'plateau', in which the level of V was maintained or showed only ~ slight decrease. Subsequently the discharge amplitude decreased again to reach a very low level. Often, though not seen in this experiment, further plateau regions were found. Short pulses of external current were briefly applied before and at different times during the stimulation to measure the inversion potential (Einv)and conductance change (AG). Sample records are shown in Fig. 1, and the modifications of Einv and
0
1.5min
Rec. 2rain
5min
r
--
--'-S--.-
~
~
" ~-----
~ r
,/"--..x.
100
i
Stimulation
5 / sec
>
Einv
~d
.....
5o
0
I
0
I
o--
l
2
1
-~
I
4
6
t
min
Fig. ]. Electrical changes in the Torpedo e]ectrogenJc tissue during 'fatigue' and recovery of transmission. A piece of electric organ (3~- cm length ; 0. ] 3 sq.cm section ; 5 prisms) was stimulated through its nerve at 5~see.Short pulses of external current were briefly applied at different moments of'activity. The voltage records are seen corresponding to 0, 1.5 and 5 rain stimulation, and 2 rain recovery. Arrows indicate the level of the inversion potential (Einv) which gives the measurement of the electromotive force of the tissue. Calibration shows I0 V and 1 msec. The conductance change (AG) is the change in tissue c o n d u c t a n c e between the resting state a n d the peak of activity. In the graph, the a m p l i t u d e o f the discharge (V) as well as A G have been plotted as per cent of their initial values, which were V 0 :- 22 V; E°inv -- 42 V; AG O 0.62 × 10.3 ~ ~. A complete recovery of all these parameters was seen 30-60 m i n after the end o f activity.
128 A G with time are plotted in the graph. Einv decreased e x p o n e n t i a l l y d u r i n g the first 2 nnin o f s t i m u l a t i o n to r e m a i n stable at a b o u t h a l f o f its initial level. The recovery o l E~nv was f o u n d to be a r a t h e r slow process since the force r e t u r n e d to the initial value only l h after the end o f s t i m u l a t i o n . The changes o f A G were quite different f r o m those o f Einv. A l t h o u g h there was again a large initial decrease, this was followed by clear plateau, followed by a decrease to a very low value. A G partially recovered after a few minutes rest. In o t h e r experiments, the polarizing pulses were a p p l i e d at s h o r t e r time intervals and further plateaus o f A G were found, c o r r e s p o n d i n g to those o f V (see ref. 7). T h e full recovery o f A G was achieved only after 30-60 rain. These changes in Em~ a n d A G d u r i n g repetitive activity were similar when small prisms were stimulated by field shocks a n d when the tissue was s u b m i t t e d to a b u n d a n t superfusion with the saline m e d i u m . As expected in the last c o n d i t i o n , the a m p l i t u d e , o f V was r e d u c e d by the short-circuiting action o f the saline, b u t the values o f E~,,~ and A G were not modified. A G is expected to be related to the n u m b e r o f effective A C h - r e c e p t o r interactions. M o d i f i c a t i o n s o f this p a r a m e t e r can c o n s e q u e n t l y be indicative o f changes in the a m o u n t o f t r a n s m i t t e r released in a given pulse. F o r this reason, a t t e n t i o n was
S
B"-.'-.
i
___
'
'
'
ii ,,o
-
""'1
ii
.
II
ii
~
iJ II. •
•
o
c •
B
ii JJ
BO
sec onds
'~. . . . . . ; ~ ' " " ; a
..... ;e"
seconds
,'~" . . . . ¢o . . . . . ~, . . . . .
~ ...... ~""'~,'
s~¢ceds
di"~"~'~
...... ~ '
Fig. 2. S l o w w a v e a n d fast o s c i l l a t i o n o f t o t a l A C h in the course o f s t i m u l a t i o n at 5~see. Inset shows a curve describing the slow ACh changes of the tissue which appeared when the sampling was performed with a low time resolution. This curve with its successive peaks has been called 'slow wave'. Regions A, B and C of the slow wave have been analyzed with a high time resolution of sampling (every second) in the lower graphs. The rapid oscillations became apparent. They occurred at different mean levels corresponding to three phases of the slow wave. The rapid oscillations were undamped and involved about 35 % of total ACh. Their period was approximately 5 sec. Lower graphs A, B and C are from separate experiments. The extraction procedure was 'hot HCI' as described in 'Methods'. The results are expressed as per cent of unstimulated samples.
129 paid to any possible correlation between the time course of AG and that of tissue ACh.
ACh changes as a function of time during stimulation The changes in total ACh during repetitive activity were two-fold : a slow wave and, superimposed on it, a rapid oscillation. They are illustrated in Fig. 2 by experiments in which the tissue samples were taken at two different frequencies during the course of stimulation at 5/sec. When the tissue was quenched at relatively large time intervals (inset in Fig. 2), the total amount of ACh was seen to diminish during the first decline of AG, to a level usually about 60 of of controls. It then increased from this low level to a peak which was equal to or even higher than control values. The time of occurrence of the peak corresponded to the plateau phase in the changes of AG. The total ACh then diminished from this peak value, coinciding with the decrease of AG from the first plateau. Often additional peaks were found which appeared to correspond with the subsequent plateaus of AG. These relatively slow changes in ACh involved about 40 % of the total. There was, however, in certain experiments a rather large dispersion of the ACh values, especially during the first minute of stimulation. Consequently the frequency of sampling was progressively increased to finally reach a time resolution of I sec. It was then found that the apparent dispersion was in fact due to the presence of a rapid oscillation of ACh, which was superimposed on the slow wave. The rapid oscillation showed no sign of attenuation; its amplitude was about 20-40 o~ of the total ACh and its period about 5 sec. Very important was the fact that, when the rapid oscillation was analyzed at different times during the course of stimulation, it occurred at different mean levels. This is illustrated in Fig. 2, where the rapid oscillation was measured at times corresponding with the initial fall (A), the first peak (B) and a later phase (C) of the slow wave previously described. It can be seen that the mean ACh levels were lower than controls in A, above in B, and below in C. This explains well the shape of the curve obtained with a low frequency of sampling. In contrast to the slow wave, no clear changes of the electrophysiological parameters were found corresponding to the rapid oscillation. There were in any case no proportionate variations in the amplitude or duration of the discharge. Statistical analysis of the rapid oscillation How far is the observed rapid fluctuation a real periodic function? This was tested in experiments where the oscillation was followed for at least 60 sec, i.e. 12 periods. During the second minute of stimulation at 5/sec, the mean level of the rapid oscillation was stable enough to allow a power spectrum analysis. The autocorrelation function (Fig. 3a) demonstrated that the pattern was not random, having a period of about 5 sec. In other experiments, although the rapid oscillation was clearly apparent, the level changes due to the slow wave made it difficult to calculate this function. Another way to test the reliability of the rapid oscillation has been to stimulate a number of samples for the same times, so that mean values, with standard deviations,
0
.
=~
0
-~\
.~I
i
0
J
i
,
controls
,
C~
,
o ~
=
~
i
i
I
0
~
I
,
0
i
controls
i
i
o
,
i
i
~---e
i
i
,
=
0
0
°
I
I
o
i
, i , controls
,
o
i
i
,
,
,
i
i
,
i
=
\
i
¢="
i
.
~------ =---- • je
controls
i
I
I
i
0
o
--
I
o
0
o
1
-4
O"
3 o_
~-
o
¢~
Q
!
P I
01-
0
131 lOO
Total ACh nmole/g
Voltage in % 1400
B
A
8O
120 60
\
1000
\
40
....
2
',,
19°C
~.
~g
80 ¸
\O.~o~ °
•
" . %
5oc
o---e~ 60
¸
minutes ][" 0
I 1
I 2
I a
4
minutes
o
Fig. 4. A: effect of temperature on the evolution of the electrophysiological response to stimulation. (V is expressed as per cent o f initial discharge.) Field stimulation at 5/sec was applied at 5 ~'C or 19 ~C. T h e plateau was depressed at 5 °C. B: simultaneous variations of total A C h in the s a m e experiments. The rise of the slow wave was markedly reduced at 5 "C.
could be compared for several points in time during a single period of the oscillation. It was found that, in an interval of time lower or equal to the presumed period of oscillation (5 sec), there were always at least two points having highly significantly different means (Fig. 3b). Finally, systematic errors were excluded by experiments where unstimulated samples were quenched as a function of time using the same protocol as in other experiments. These controls were indicative of the normal range of experimental error which did not exceed ± 10~ nor show any sign of periodicity (Fig. 5, lower graph). Effeet of the temperature on the slow wave and the rapid oseillation Fig. 4A shows that the first plateau of the curve describing the evolution of the electrophysiological response was depressed when the stimulation was delivered at 5 °C rather than at room temperature. As previously shown 9, the rise of total ACh
Fig. 3. a: autocorrelation analysis of the rapid A C h oscillation. Experiment of the 2nd m i n of stimulation at 5/sec (a part of it is seen in Fig. 2B). N u m b e r of samples 60; n u m b e r of periods = 12. T h e autocorrelation function clearly shows that the fluctuation is not r a n d o m a n d that the period is slightly larger t h a n 5 sec. b: Statistical evidence for the rapid A C h oscillation occurring during stimulation. In 4 different experiments, a n u m b e r (n) of samples were stimulated at 5/sec for the s a m e length of time, quenched a n d analyzed for total A C h . Ordinate, m e a n (:!- S.E.M.) of totaJ A C h in nmoles/g o f wet tissue. There is at least one point in each graph having a m e a n lower t h a n a previous or subsequent point. The following m e a n s are statistically different. G r a p h A, 7 sec point c o m p a r e d to 9 sec, P. 0.01;7sec I 8secpointscomparedto9sec ~ 10 sec points, P 0.001. G r a p h B , 1 3 s e c p o i n t c o m p a r e d to 14 sec, P :~ 0.001 ; 13 sec c o m p a r e d to 15 sec, P < 0.01. G r a p h C, 38 sec point c o m p a r e d to 39 sec, P 0.02. G r a p h D, 42 sec point c o m p a r e d to 45 sec, P -< 0.01. In exp. D, the presence of tubocurarine (10 4 M ) in the incubation m e d i u m abolished the electrophysiological response to stimulation but did not alter the oscillation. In B, the radiochemical m e t h o d was used to estimate A C h (see Methods). n was 6, 10, 5 a n d 5 in experiments A, B, C a n d D respectively.
132 16o
25°C
14.
/~
i ~
/'
',j f
i t i t
~l,)
~Zl
11o .~ 1oo
'
I
20oC _
310 '
I
'
I
l
I
' 318 '
_,_,_~=._.-..~.._.__~
I
',
'
• ,
°
I
'
. ....
I
' 4 i6 '
I
seconds II '
I
'
I
54
_ ~_~._.
;,
"~
~'i"i',
I
) ' ) , i , t , ) , ] , 5
9
13
17
, i ,
seconds 21~ '
~,L
i , , , ) , ) 25
29
Fig. 5. Effect of temperature on the rapid oscillation of ACh. Upper graph: oscillation of total ACh when the stimulation was delivered at 25 ~C (frequency of stimulation 5/see). The period of the oscillation was a b o u t 4 sec and its amplitude fluctuated at a mean level above 100 %. Middle graph: similar experiment performed at 6 °C. There was a reduction in the amplitude of the oscillation, and the fluctuation was irregular with a mean level below 100 %. These fluctuations were still higher than the statistical variation of unstimulated samples seen in lower graph. Lower graph: unstimutated controis at 20 ~C. In this experiment the samples were quenched as a function of time using the same protocol as in the other experiments, but stimulation was not applied. In all three cases, the samples were frozen in isopentane cooled by liquid nitrogen, powdered and extracted with T C A as indicated in Methods.
occurring during this plateau was markedly reduced at the low temperature (Fig. 4B). In order to analyze the effect of cooling on the fast ACh oscillation, we studied an interval of stimulation of 25 sec with samples taken every second. This interval of time corresponded to the first plateau of 2xG. The results are shown in Fig. 5 which
133 compares the samples stimulated at 25 c'C (upper graph) to those stimulated at 6 ~'C (middle graph). In the lower graph, the samples were left unstimulated at 20 °C, but quenched as a function of time using the same protocol as in the other experiments. When the stimulation was delivered at 6 °C, the dispersion of ACh values was larger than in controls but the periodicity did not clearly appear. At a temperature of 25 °C, a clear periodicity of about 4 sec was noticed and the amplitude of the fast oscillation was markedly greater. When expressed as per cent of unstimulated samples, most of the ACh values obtained at 6 °C were below 100 %; in contrast, at 25 °C the mean level of the curve was higher than 100~o. The difference in level between the two curves reflects the temperature dependence of the slow wave, as seen in Fig. 4B. Two similar experiments were performed at different stages of the slow wave. Their results were the same as those shown in Fig. 5. When the tissue was stimulated at a reduced temperature, the peaks of the slow wave were markedly depressed and the amplitude of the rapid oscillation was smaller. The period of the rapid oscillation seemed not much altered, but this last point was difficult to establish because of the lower amplitude. 110
10(
90_
/\ /\ ._...
BO
seconds , 10
-6 0
,
,
,
, 14
j
p
.
.
.
.
18
.
.
.
. 22
.
.
.
.
.
. 26
2O/s
~'~ 14C
cO
~ 12( 0
u~
10C
--e
8c
/
eJ
e
second/
Fig. 6. Rapid ACh oscillation at different frequencies of stimulation. The figure shows that, in a 20 sec interval, 4-5 peaks of total ACh can be seen at the frequencies of stimulation of I or 20/see, corresponding to periods of oscillation of 4-5 sec. Extraction procedure, 'liquid nitrogen-TCA' in the upper graph, 'hot HCI' in the lower graph.
134
of the frequency of stimulation on total A Ch changes The rise of ACh, occurring simultaneously with the first plateau of the electrophysiological response starts after 20-30 sec, independent of the frequency of stimulation between 1 and 10/see 9. The length of this plateau was somewhat variable and further plateaus could be distinguished, especially when AG was measured 7, or at lower frequencies. They corresponded to additional peaks of total ACh (as seen in Fig. 2, inset). The periodicity of these secondary peaks seemed to depend on the frequency of stimulation 13. However, because of the discovery of the rapid oscillation, more experiments will be needed to establish the precise relationship between the periodicity of the slow wave and the frequency of stimulation On the other hand, the frequency of stimulation appeared to have little or no effect on the period of the rapid oscillation. For a frequency of 5/see, the period of the rapid oscillation was close to 4 sec in the experiment of Fig. 5, and about 5 sec in the experiments of Fig. 2. Similar experiments were performed at frequencies of stimulation of 1/sec and 20/see (Fig. 6). The rapid oscillation was found for these two frequencies of stimulation but its period remained the same. It is possible that the amplitude of the oscillation was less at l/see, but a large number of further measurements will be needed to certify this point. Effect
Oscillations of endogenous and ' 14C /A Ch In previous reports we have shown that, when stimulation is applied in the presence of a labelled precursor, there is an increased incorporation into ACh, which only occurs after several minutes. This implies that the same pool of transmitter is released and taken up continuously, at least for the first minutes of stimulation s,9. An experiment was carried out to confirm this and measure the turnover with a higher time resolution (Fig. 7). Samples were incubated in [l-14C]acetate for 2 h, to allow incorporation of the
7I /17,, ,,
"
. . . . .
,,, /.-'(I
,,'d~
;11
.
- I
,i
I
,
\ /
" pot=
,,
;it k'
+_,
• i
\
,
,.,
-
.
I
,,-', ,, ",,,
',//
I
,;
•
.
;%/ ;11,,,
/
=-= : ,
123
© Elsevier/North-Holland Biomedical Press, Amsterdam- Printed in The Netherlands
O S C I L L A T I O N OF A C E T Y L C H O L I N E D U R I N G THE TORPEDO ELECTRIC ORGAN
NERVE
A C T I V I T Y IN
Y. DUNANT, M. ISRAEL, B. LESBATS and R. MANARANCHE Laboratoire de Neurobiologie Cellulaire, Neurochimie, C.N.R.S., 91190 Gif sar-Yvette (France) and (KD.) Ddpartement de Pharmacologie, Ecole de Mddecine, 20, 1211 GenOve 4 (Switzerland)
(Accepted July 27th, 1976)
SUMMARY The amount of transmitter in the electric organ of Torpedo was measured with a time resolution of 1 sec in the course of stimulation. In parallel, the modifications of the electrophysiological response were analysed by determining the conductance increase (AG) apd the electromotive force of electroplaques. Large changes in the level of total acetylcholine (ACh) were seen during stimulation. These changes were twofold: a slow wave and, superimposed on it, a rapid oscillation. The slow wave raised total ACh to the initial level, or evet~ higher. It was probably related to modifications in the amount of ACh released since it corresponded to characteristic inflections in the evolution of the AG curve. The slow wave and this physiological parameter were similarly affected when the experiments were performed at a reduced temperature. The rapid oscillation had an amplitude of about 20 4 0 ~ of the total ACh. It was undamped and its period was 4-5 sec. In contrast to the slow wave, no clear physiological change associated with the rapid oscillation has been observed. The slow wave and rapid oscillation occurred in the 'free pool' of ACh, whereas bound ACh, the fraction associated with synaptic vesicles, was not affected by these changes. A dynamic description of synaptic activity is proposed. The content of 'free' ACh is used and renewed completely after a few tens of impulses, so that transmission seems to imply the continual recycling of the same pool of transmitter rather than utilization of a large preloaded store. The release processes must then be integrated in rapid metabolic loops.
INTRODUCTION A joint electrophysiological and biochemical description is presented in this work, which is a dynamic analysis of transmission in the electric organ of the fish,
124 Torpedo marmorata. This very homogeneous tissue consists of a great number of prisms, arranged side by side, each being built up of about 500 superposed electroplaques, which are covered on their ventral surface by a rich network of nerve terminals. The cholinergic nature of transmission has been demonstrated in this organlS, ~6, When the nerve is active, the terminals release acetylcholine (ACh), which acts on receptors situated at the ventral membrane of the plates. This results in a large number of synchronized postjunctional potentials which sum to generate a strong discharge. positive towards the dorsal face of the fish. The released transmitter is lhen rapidly hydrolyzed, and choline and acetate ~5 taken up into the nerve terminals where ACh is resynthesized. Since hydrolysis of the released transmitter occurs in a matter of msec, all the ACh found in the tissue when it is quenched for extraction must be regarded as intracellular and most probably intraterminal ')1. Advantage has been taken of this to measure nerve terminal ACh as a function of time during stimulation. In preliminary experiments, the level of transmitter was found to change quickly; it was therefore necessary to improve the rapidity of the biochemical methods, which finally reached a time resolution of 1 sec. The changes of tissue ACh have been correlated, when possible, with modifications of the electrical parameters of the response. The peculiar arrangement of the electric organ allowed us to record not only the amplitude of the discharge, which is an overall measurement of synaptic efficiency, but also the inversion potential and the conductance change of the active tissue. The inversion potential provides useful information on the ionic electromotive force acting on the postsynaptic membrane; the conductance change is expected to be a function of the number of receptors which are activated in the membrane at a given time. Most of the present work has been carried out by analyzing the total ACh of the tissue. Experiments were also performed to determinethe compartment of ACh modified during stimulation. Bound ACh is that which remains after the tissue has been homogenized. Most of it is present in synaptic vesicles which can be isolated and purified by fractionation methods2e,~':3,":L 'Free ACh' is the part, about half of the total, which is hydrolyzed by esterases when the tissue is disrupted. The term 'free' must be considered as operational since the subcellular localization of this compartment has not yet been elucidated (see review in ref. 28). Previous work has shown that stimulation increases the turnover of 'free ACh" but not that of bound ACh s,9. Partial descriptions of these electrophysiological and biochemical lindings have been published as short communicationsT,W,t:3,'-'L METHODS Animals Torpedo marmorata were supplied by the 'Station de Biologie Marine', Arcachon, France. Most of the fish were females of 30-60 cm length. They were kept in sea water at about 15 C . Incubation medium When excised, the tissue was kept in an elasmobranch saline solution containing
125 (in m M ) : NaC1, 280; KC1, 3.0; CaClo., 3.4; MgClz, 1.8; NaHCOa, 5; NaH~PO4, 1.2; urea, 300; saccharose, 100; and glucose 5.5. This medium was oxygenated and its pH kept between 7.1 and 7.4. A good conservation of the tissue during several hours was shown by morphological, biochemical and physiological controls. When required, drugs or labelled precursors were added to this solution. Unless mentioned, experiments were carried out at room temperature.
Electrophysiological methods Stimuhttion by the nerve. The Torpedo was anaesthetized by tricaine methane sulphonate (MS 222, Sandoz) at a concentration of I g/3 1 sea water. Controls showed that this short anaesthesia has no effect on the efficiency of nerve-electroplaque transmission. Fragments of electric organ were carefully excised with their nerve intact. They usually contained a few prisms of electroplaques. After dissection the tissue was allowed to recover in the saline medium for at least I h. Stimulation was then applied to the nerve and the response potential recorded between the ventral and the dc, rsal ends of the prisms. Field stimulation of the electrogenic tissue. When 2 or 3 dissected prisms of electric tissue are submitted to a brief (1 msec) electric shock, the response discharge recorded is separated from the stimulus artefact by an irreducible latency of 2-3 msec. Nerve degeneration or treatment with curare completely abolishes this response 3. These observations and microelectrode investigations 4 demonstrated that Torpedo electroplaques are not able to generate and propagate action potentials. Field stimulation only excites nerves and nerve terminals in the tissue directly. The electrical discharge is an indirect response of electroplaques to ACh released by nerve terminals; more precisely, it is the sum of individual postjunctional potentials. Field stimulation was found to be very convenient in experiments where the tissue had to be incubated in solutions containing drugs or radioactive precursors, or when a high resolution in time was wanted. Fragments of electric organ were excised and incubated in the saline medium. Care was taken not to damage the electroplaques by cutting through the prisms longitudinally. The pieces were weighed and set on a piece of nylon cloth between the two stimulating electrodes which were parallel to the prisms. Stimulation is more efficient ill this way since the field can easily reach the nerves at the edges of the electroplaques and between them iv. The set-up was either immersed or superfused by a continuous flow of the saline solution. Supramaximal shocks were applied and the responses recorded between the two extremities of the prism. When the prisms were kept at room temperature and their response tested at large time intervals, a slight increase in the voltage amplitude was observed during the 2 h following dissection; the response then remained at a very stable level for more than 24 h.
Determination of inversion potential and conductance change during stimulation. These parameters were measured using a method recently described v, based on an observation performed on the electric organ of Torpedo nobiliana 4. A piece of electric organ was dissected with its nerve. The ventral and dorsal ends of the fragment were placed between two Ag-AgCI.~ plates. The tissue was stimulated through the
126 nerve and produced the usual response of amplitude, V. This discharge was modified by applying brief polarizing currents to the whole piece of tissue through the silver plates. The electromotive force of the fragment was measured under these conditions by giving an equal but opposite back-electromotive force. This was seen by the suppression of the voltage change of the response. This force has been called 'inversion potential' (E~nv) since the polarity of the discharge was reversed when stronger external potentials were applied. The conductance change (AG) was the difference between the conductance of the tissue measured at the peak of the discharge and that measured in the resting state. These parameters are relatively easy to determine in the electrogenic organ of Torpedo since the voltage changes are linearly related to the applied currents not only at rest but also during activity. Isolated prisms could also be submitted to this technique; they were stimulated through the silver plates by a brief field shock preceding the application of the polarizing pulse.
Biochemical methods' Sample~. Tissue samples were incubated for at [east l h in the physiological solution. A group of samples were then stimulated and, at different times, one of them was dropped into the extraction solution. In the best conditions, the latency of the transfer was reduced to about 0.25 sec. Then new groups were taken and stimulated in such a manner that a time resolution of I sec was achieved. When, after the first experiments, it appeared that ACh changes were periodical, the sampling was randomized to avoid any systematic periodicity in the protocol. Two different rapid quenching procedures were used. For the first, the samples were dropped into liquid nitrogen or isopentane cooled by liquid nitrogen. After freezing, the sample was powdered and the powder extracted in 5 ml of 5 o~ trichloracetic acid (TCA). The TCA was then removed by washing with ether. In the second method, the samples were dropped into 5 ml of hot HC1 (87 '~C, pH 2.7) and left for 7 rain. At the end of this time, the pH had risen to about 5. The second procedure was more convenient and gave similar results to the first. These procedures extract the total ACh content of the tissue. To determine the amount of bound or vesicular ACh, the samples were dropped into the physiological solution at about 5 '~C and immediately homogenized. TCA was then added to the tube without delay (5 ~/~ifinal concentration). Alternatively, an equal volume of N/50 HCI at 87 °C was added, and left for 7 rain. Estimation of ACh. The ACh of the extracted samples was, for most of the experiments reported here, estimated with the conventional frog rectus method. In order to confirm the results, and to facilitate the analysis of a large number of samples, a radiochemical procedure to measure ACh was developed 1°, and used in some experiments. Labelling of the acetyl moiety of ACh. [l-14C]acetate was added to the incubation medium when necessary, and the radioactive acetylcholine synthesized by the tissue was extracted and separated frorn the acetate as previously described2L Other estimations. Choline acetyltransferase was assayed by the method described by Fonnum 18 and proteins were measured according to Lowry et ai. zv,
127 RESULTS
Conductance change and inversion potential in the course of repetitive stimulation It has been known for a long time that repeated excitation of the electric fish results in a 'fatigue' of the discharge. Fig. 1 shows a detailed analysis of this phenomenon. The tissue was stimulated through its nerve at a frequency of 5/sec. The electrical response discharge (V) was initially 20 V, and its amplitude fell to half after about 30 sec stimulation. There then occurred a 'plateau', in which the level of V was maintained or showed only ~ slight decrease. Subsequently the discharge amplitude decreased again to reach a very low level. Often, though not seen in this experiment, further plateau regions were found. Short pulses of external current were briefly applied before and at different times during the stimulation to measure the inversion potential (Einv)and conductance change (AG). Sample records are shown in Fig. 1, and the modifications of Einv and
0
1.5min
Rec. 2rain
5min
r
--
--'-S--.-
~
~
" ~-----
~ r
,/"--..x.
100
i
Stimulation
5 / sec
>
Einv
~d
.....
5o
0
I
0
I
o--
l
2
1
-~
I
4
6
t
min
Fig. ]. Electrical changes in the Torpedo e]ectrogenJc tissue during 'fatigue' and recovery of transmission. A piece of electric organ (3~- cm length ; 0. ] 3 sq.cm section ; 5 prisms) was stimulated through its nerve at 5~see.Short pulses of external current were briefly applied at different moments of'activity. The voltage records are seen corresponding to 0, 1.5 and 5 rain stimulation, and 2 rain recovery. Arrows indicate the level of the inversion potential (Einv) which gives the measurement of the electromotive force of the tissue. Calibration shows I0 V and 1 msec. The conductance change (AG) is the change in tissue c o n d u c t a n c e between the resting state a n d the peak of activity. In the graph, the a m p l i t u d e o f the discharge (V) as well as A G have been plotted as per cent of their initial values, which were V 0 :- 22 V; E°inv -- 42 V; AG O 0.62 × 10.3 ~ ~. A complete recovery of all these parameters was seen 30-60 m i n after the end o f activity.
128 A G with time are plotted in the graph. Einv decreased e x p o n e n t i a l l y d u r i n g the first 2 nnin o f s t i m u l a t i o n to r e m a i n stable at a b o u t h a l f o f its initial level. The recovery o l E~nv was f o u n d to be a r a t h e r slow process since the force r e t u r n e d to the initial value only l h after the end o f s t i m u l a t i o n . The changes o f A G were quite different f r o m those o f Einv. A l t h o u g h there was again a large initial decrease, this was followed by clear plateau, followed by a decrease to a very low value. A G partially recovered after a few minutes rest. In o t h e r experiments, the polarizing pulses were a p p l i e d at s h o r t e r time intervals and further plateaus o f A G were found, c o r r e s p o n d i n g to those o f V (see ref. 7). T h e full recovery o f A G was achieved only after 30-60 rain. These changes in Em~ a n d A G d u r i n g repetitive activity were similar when small prisms were stimulated by field shocks a n d when the tissue was s u b m i t t e d to a b u n d a n t superfusion with the saline m e d i u m . As expected in the last c o n d i t i o n , the a m p l i t u d e , o f V was r e d u c e d by the short-circuiting action o f the saline, b u t the values o f E~,,~ and A G were not modified. A G is expected to be related to the n u m b e r o f effective A C h - r e c e p t o r interactions. M o d i f i c a t i o n s o f this p a r a m e t e r can c o n s e q u e n t l y be indicative o f changes in the a m o u n t o f t r a n s m i t t e r released in a given pulse. F o r this reason, a t t e n t i o n was
S
B"-.'-.
i
___
'
'
'
ii ,,o
-
""'1
ii
.
II
ii
~
iJ II. •
•
o
c •
B
ii JJ
BO
sec onds
'~. . . . . . ; ~ ' " " ; a
..... ;e"
seconds
,'~" . . . . ¢o . . . . . ~, . . . . .
~ ...... ~""'~,'
s~¢ceds
di"~"~'~
...... ~ '
Fig. 2. S l o w w a v e a n d fast o s c i l l a t i o n o f t o t a l A C h in the course o f s t i m u l a t i o n at 5~see. Inset shows a curve describing the slow ACh changes of the tissue which appeared when the sampling was performed with a low time resolution. This curve with its successive peaks has been called 'slow wave'. Regions A, B and C of the slow wave have been analyzed with a high time resolution of sampling (every second) in the lower graphs. The rapid oscillations became apparent. They occurred at different mean levels corresponding to three phases of the slow wave. The rapid oscillations were undamped and involved about 35 % of total ACh. Their period was approximately 5 sec. Lower graphs A, B and C are from separate experiments. The extraction procedure was 'hot HCI' as described in 'Methods'. The results are expressed as per cent of unstimulated samples.
129 paid to any possible correlation between the time course of AG and that of tissue ACh.
ACh changes as a function of time during stimulation The changes in total ACh during repetitive activity were two-fold : a slow wave and, superimposed on it, a rapid oscillation. They are illustrated in Fig. 2 by experiments in which the tissue samples were taken at two different frequencies during the course of stimulation at 5/sec. When the tissue was quenched at relatively large time intervals (inset in Fig. 2), the total amount of ACh was seen to diminish during the first decline of AG, to a level usually about 60 of of controls. It then increased from this low level to a peak which was equal to or even higher than control values. The time of occurrence of the peak corresponded to the plateau phase in the changes of AG. The total ACh then diminished from this peak value, coinciding with the decrease of AG from the first plateau. Often additional peaks were found which appeared to correspond with the subsequent plateaus of AG. These relatively slow changes in ACh involved about 40 % of the total. There was, however, in certain experiments a rather large dispersion of the ACh values, especially during the first minute of stimulation. Consequently the frequency of sampling was progressively increased to finally reach a time resolution of I sec. It was then found that the apparent dispersion was in fact due to the presence of a rapid oscillation of ACh, which was superimposed on the slow wave. The rapid oscillation showed no sign of attenuation; its amplitude was about 20-40 o~ of the total ACh and its period about 5 sec. Very important was the fact that, when the rapid oscillation was analyzed at different times during the course of stimulation, it occurred at different mean levels. This is illustrated in Fig. 2, where the rapid oscillation was measured at times corresponding with the initial fall (A), the first peak (B) and a later phase (C) of the slow wave previously described. It can be seen that the mean ACh levels were lower than controls in A, above in B, and below in C. This explains well the shape of the curve obtained with a low frequency of sampling. In contrast to the slow wave, no clear changes of the electrophysiological parameters were found corresponding to the rapid oscillation. There were in any case no proportionate variations in the amplitude or duration of the discharge. Statistical analysis of the rapid oscillation How far is the observed rapid fluctuation a real periodic function? This was tested in experiments where the oscillation was followed for at least 60 sec, i.e. 12 periods. During the second minute of stimulation at 5/sec, the mean level of the rapid oscillation was stable enough to allow a power spectrum analysis. The autocorrelation function (Fig. 3a) demonstrated that the pattern was not random, having a period of about 5 sec. In other experiments, although the rapid oscillation was clearly apparent, the level changes due to the slow wave made it difficult to calculate this function. Another way to test the reliability of the rapid oscillation has been to stimulate a number of samples for the same times, so that mean values, with standard deviations,
0
.
=~
0
-~\
.~I
i
0
J
i
,
controls
,
C~
,
o ~
=
~
i
i
I
0
~
I
,
0
i
controls
i
i
o
,
i
i
~---e
i
i
,
=
0
0
°
I
I
o
i
, i , controls
,
o
i
i
,
,
,
i
i
,
i
=
\
i
¢="
i
.
~------ =---- • je
controls
i
I
I
i
0
o
--
I
o
0
o
1
-4
O"
3 o_
~-
o
¢~
Q
!
P I
01-
0
131 lOO
Total ACh nmole/g
Voltage in % 1400
B
A
8O
120 60
\
1000
\
40
....
2
',,
19°C
~.
~g
80 ¸
\O.~o~ °
•
" . %
5oc
o---e~ 60
¸
minutes ][" 0
I 1
I 2
I a
4
minutes
o
Fig. 4. A: effect of temperature on the evolution of the electrophysiological response to stimulation. (V is expressed as per cent o f initial discharge.) Field stimulation at 5/sec was applied at 5 ~'C or 19 ~C. T h e plateau was depressed at 5 °C. B: simultaneous variations of total A C h in the s a m e experiments. The rise of the slow wave was markedly reduced at 5 "C.
could be compared for several points in time during a single period of the oscillation. It was found that, in an interval of time lower or equal to the presumed period of oscillation (5 sec), there were always at least two points having highly significantly different means (Fig. 3b). Finally, systematic errors were excluded by experiments where unstimulated samples were quenched as a function of time using the same protocol as in other experiments. These controls were indicative of the normal range of experimental error which did not exceed ± 10~ nor show any sign of periodicity (Fig. 5, lower graph). Effeet of the temperature on the slow wave and the rapid oseillation Fig. 4A shows that the first plateau of the curve describing the evolution of the electrophysiological response was depressed when the stimulation was delivered at 5 °C rather than at room temperature. As previously shown 9, the rise of total ACh
Fig. 3. a: autocorrelation analysis of the rapid A C h oscillation. Experiment of the 2nd m i n of stimulation at 5/sec (a part of it is seen in Fig. 2B). N u m b e r of samples 60; n u m b e r of periods = 12. T h e autocorrelation function clearly shows that the fluctuation is not r a n d o m a n d that the period is slightly larger t h a n 5 sec. b: Statistical evidence for the rapid A C h oscillation occurring during stimulation. In 4 different experiments, a n u m b e r (n) of samples were stimulated at 5/sec for the s a m e length of time, quenched a n d analyzed for total A C h . Ordinate, m e a n (:!- S.E.M.) of totaJ A C h in nmoles/g o f wet tissue. There is at least one point in each graph having a m e a n lower t h a n a previous or subsequent point. The following m e a n s are statistically different. G r a p h A, 7 sec point c o m p a r e d to 9 sec, P. 0.01;7sec I 8secpointscomparedto9sec ~ 10 sec points, P 0.001. G r a p h B , 1 3 s e c p o i n t c o m p a r e d to 14 sec, P :~ 0.001 ; 13 sec c o m p a r e d to 15 sec, P < 0.01. G r a p h C, 38 sec point c o m p a r e d to 39 sec, P 0.02. G r a p h D, 42 sec point c o m p a r e d to 45 sec, P -< 0.01. In exp. D, the presence of tubocurarine (10 4 M ) in the incubation m e d i u m abolished the electrophysiological response to stimulation but did not alter the oscillation. In B, the radiochemical m e t h o d was used to estimate A C h (see Methods). n was 6, 10, 5 a n d 5 in experiments A, B, C a n d D respectively.
132 16o
25°C
14.
/~
i ~
/'
',j f
i t i t
~l,)
~Zl
11o .~ 1oo
'
I
20oC _
310 '
I
'
I
l
I
' 318 '
_,_,_~=._.-..~.._.__~
I
',
'
• ,
°
I
'
. ....
I
' 4 i6 '
I
seconds II '
I
'
I
54
_ ~_~._.
;,
"~
~'i"i',
I
) ' ) , i , t , ) , ] , 5
9
13
17
, i ,
seconds 21~ '
~,L
i , , , ) , ) 25
29
Fig. 5. Effect of temperature on the rapid oscillation of ACh. Upper graph: oscillation of total ACh when the stimulation was delivered at 25 ~C (frequency of stimulation 5/see). The period of the oscillation was a b o u t 4 sec and its amplitude fluctuated at a mean level above 100 %. Middle graph: similar experiment performed at 6 °C. There was a reduction in the amplitude of the oscillation, and the fluctuation was irregular with a mean level below 100 %. These fluctuations were still higher than the statistical variation of unstimulated samples seen in lower graph. Lower graph: unstimutated controis at 20 ~C. In this experiment the samples were quenched as a function of time using the same protocol as in the other experiments, but stimulation was not applied. In all three cases, the samples were frozen in isopentane cooled by liquid nitrogen, powdered and extracted with T C A as indicated in Methods.
occurring during this plateau was markedly reduced at the low temperature (Fig. 4B). In order to analyze the effect of cooling on the fast ACh oscillation, we studied an interval of stimulation of 25 sec with samples taken every second. This interval of time corresponded to the first plateau of 2xG. The results are shown in Fig. 5 which
133 compares the samples stimulated at 25 c'C (upper graph) to those stimulated at 6 ~'C (middle graph). In the lower graph, the samples were left unstimulated at 20 °C, but quenched as a function of time using the same protocol as in the other experiments. When the stimulation was delivered at 6 °C, the dispersion of ACh values was larger than in controls but the periodicity did not clearly appear. At a temperature of 25 °C, a clear periodicity of about 4 sec was noticed and the amplitude of the fast oscillation was markedly greater. When expressed as per cent of unstimulated samples, most of the ACh values obtained at 6 °C were below 100 %; in contrast, at 25 °C the mean level of the curve was higher than 100~o. The difference in level between the two curves reflects the temperature dependence of the slow wave, as seen in Fig. 4B. Two similar experiments were performed at different stages of the slow wave. Their results were the same as those shown in Fig. 5. When the tissue was stimulated at a reduced temperature, the peaks of the slow wave were markedly depressed and the amplitude of the rapid oscillation was smaller. The period of the rapid oscillation seemed not much altered, but this last point was difficult to establish because of the lower amplitude. 110
10(
90_
/\ /\ ._...
BO
seconds , 10
-6 0
,
,
,
, 14
j
p
.
.
.
.
18
.
.
.
. 22
.
.
.
.
.
. 26
2O/s
~'~ 14C
cO
~ 12( 0
u~
10C
--e
8c
/
eJ
e
second/
Fig. 6. Rapid ACh oscillation at different frequencies of stimulation. The figure shows that, in a 20 sec interval, 4-5 peaks of total ACh can be seen at the frequencies of stimulation of I or 20/see, corresponding to periods of oscillation of 4-5 sec. Extraction procedure, 'liquid nitrogen-TCA' in the upper graph, 'hot HCI' in the lower graph.
134
of the frequency of stimulation on total A Ch changes The rise of ACh, occurring simultaneously with the first plateau of the electrophysiological response starts after 20-30 sec, independent of the frequency of stimulation between 1 and 10/see 9. The length of this plateau was somewhat variable and further plateaus could be distinguished, especially when AG was measured 7, or at lower frequencies. They corresponded to additional peaks of total ACh (as seen in Fig. 2, inset). The periodicity of these secondary peaks seemed to depend on the frequency of stimulation 13. However, because of the discovery of the rapid oscillation, more experiments will be needed to establish the precise relationship between the periodicity of the slow wave and the frequency of stimulation On the other hand, the frequency of stimulation appeared to have little or no effect on the period of the rapid oscillation. For a frequency of 5/see, the period of the rapid oscillation was close to 4 sec in the experiment of Fig. 5, and about 5 sec in the experiments of Fig. 2. Similar experiments were performed at frequencies of stimulation of 1/sec and 20/see (Fig. 6). The rapid oscillation was found for these two frequencies of stimulation but its period remained the same. It is possible that the amplitude of the oscillation was less at l/see, but a large number of further measurements will be needed to certify this point. Effect
Oscillations of endogenous and ' 14C /A Ch In previous reports we have shown that, when stimulation is applied in the presence of a labelled precursor, there is an increased incorporation into ACh, which only occurs after several minutes. This implies that the same pool of transmitter is released and taken up continuously, at least for the first minutes of stimulation s,9. An experiment was carried out to confirm this and measure the turnover with a higher time resolution (Fig. 7). Samples were incubated in [l-14C]acetate for 2 h, to allow incorporation of the
7I /17,, ,,
"
. . . . .
,,, /.-'(I
,,'d~
;11
.
- I
,i
I
,
\ /
" pot=
,,
;it k'
+_,
• i
\
,
,.,
-
.
I
,,-', ,, ",,,
',//
I
,;
•
.
;%/ ;11,,,
/
=-= : ,
