- Email: [email protected]
Action potentials produce a long-term enhancement of M-current in frog sympathetic ganglion
Brain Research, 580 (1992) 281-287 ~) 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/92/$05.00
281
BRES 17735
Action potentials produce a long-term enhancement of M-current in frog sympathetic ganglion Alfredo Kirkwood* and John E. Lisman Program in Biophysics, Department of Biology and the Center for Complex Systems, Brandeis University, Waltham MA 02154 (USA) (Accepted 31 December 1991)
Key words: Long-term potentiation; Excitability; Feedback; Potassium conductance
M-current is a voltage-gated K+ current that can be turned off by the muscarinic action of acetylcholine. We examined the effects of postsynaptic action potential firing on the level of M-current in B-cells of the bullfrog sympathetic ganglion. High frequency stimulation of action potentials induced an approximately two-fold increase in the level of the M-current that could last up to 35 min. The 'enhanced' M-current was similar to the 'resting' one in its time-dependence, voltage-dependence and sensitivity to neurotransmitters. Experiments were undertaken to examine the functional consequences of the enhanced M-current. Following high frequency stimulation the number of spikes evoked by depolarizing current was reduced. In addition, the excitatory postsynaptic potential (EPSP) evoked by maximal input became subthreshold, thereby blocking information flow through the ganglion cell. These results indicate that the enhancement of M-current by spikes provides a negative feedback mechanism for the control of excitability. It has been reported that postsynaptic stimulation of ganglion cells also produces a long-term increase in the nicotinic EPSP, but we were unable to confirm this observation. INTRODUCTION K ÷ conductances are important determinants of neuronal excitability, but the role of different K ÷ conductances and the mechanisms controlling these conductances remain unclear. O n e K ÷ conductance k n o w n to affect excitability is the Ca2+-activated K + conductance 11'16'17'21. The intracellular Ca 2÷ elevation caused by action potentials turns on Ca2+-activated K ÷ channels which then reduce subsequent firing. These K ÷ channels turn off as Ca 2÷ levels fall, a process that takes about 10 s in frog sympathetic ganglion cells2s. These findings demonstrate that there is a negative feedback mechanism that can regulate excitability. Here we report that frog sympathetic ganglion cells have a second feedback mechanism that regulates excitability over much longer time periods ( - 3 0 min). This mechanism involves M-current, a non-inactivating, voltage-dependent K + current first identified in frog sympathetic ganglion cells 2, but subsequently found in m a n y central and peripheral n e u r o n s 6. M-current is the major K ÷ current near spike threshold and therefore has marked effects on n e u r o n a l excitability2'21'23. Previous work on frog sympathetic ganglion cells has shown that the magnitude of M-current can be modulated by extrinsic factors including acetylcholine2, peptide neurotransmitters 9 and nucleotides2. O u r results show M-current
can also be modulated by action potentials intrinsic to ganglion cells and that this modulation has long-term effects on cell excitability. MATERIALS AND METHODS
Dissection and mounting Experiments were done on B cells in the 9th and 10th sympathetic ganglia of the bullfrog (Rana catesbeiana). The ganglia were excised, mounted in a small plexiglas chamber and treated with 3% collagenase (type IIs Sigma) for 5-10 rain to facilitate desheathing. Ringer composition was (in mM) NaCI 115.5, KCI 2, CaC12 3.6, MgCI2 5, TRIS 5 (pH, 7.3). All drugs except BAPTA (from Boehringer) and chicken LHRH (from Peninsula) were purchased from Sigma. Recording procedures Cells were impaled with conventional microelectrodes filled with 3 M KCI (60-90 Mff2). B-cells were identified by their short latencies and their responsiveness to presynaptic stimulation above the 7th ganglion2°. The strength of presynaptic electrical stimulation was always adjusted to give a maximal postsynaptic response. Muscarinic synaptic currents and agonist-induced nicotinic currents were recorded using discontinuous single electrode voltage-clamp (Axoclamp 2A; sampling frequency = 3-4 KHz). These current records were filtered at 200 Hz. To measure the M-current, the cells were voltage-clamped with two electrodes and the data filtered at 2 KHz. Extracellular fast synaptic potentials were recorded in the 10th ganglion using an AC differential amplifier. In these experiments, the whole sympathetic chain (from the 10th to the 5th ganglion) was freed of connective tissue. The chain was placed in a plexiglas chamber which was divided in a linear array of compartments interconnected by small slits. Each ganglia was placed in a different compartment with the interconnecting nerve running through the
* Present address: Center for Neural Sciences, Brown University, Providence, RI 02912, USA. Correspondence: J.E. Lisman, Department of Biology, Brandeis University, Waltham, MA 02154, USA.
282 3.0
c
10th ganglion; the other lead was connected to the compartment containing the postganglionic nerve.
2.0
2 •
c
k
2
• 00
~
1.o
RESULTS • 1
O0
•
1 0.0
000
•
•
•
A
1'0
Enhancement of the M-current by high frequency stimulation (HFS)
A
2'0
310
,~
i sec
E x p e r i m e n t s w e r e d o n e in bullfrog B type g a n g l i o n cells 7'2°. Cells in the intact ganglion w e r e i m p a l e d with
Time (minutes) Fig. 1. Time course (left) of enhancement of M-current by high frequency stimulation (HFS) (arrowheads). Right: high time resolution view of current relaxations due to M-current before (1) and after (2) HFS. Holding voltage -35 mV; currents evoked by 1 s hyperpolarizing pulse to -65 mV.
two m i c r o e l e c t r o d e s .
M-current
was m e a s u r e d
under
v o l t a g e - c l a m p by s t a n d a r d p r o c e d u r e s 3. T h e m e m b r a n e p o t e n t i a l was h e l d at - 3 5 m V while r e p e t i t i v e l y applying 1 s c o m m a n d pulses to - 6 5 mV. H y p e r p o l a r i z a t i o n to - 6 5 m V turns off the M - c u r r e n t . R e p o l a r i z a t i o n to - 3 5
slits, which were sealed with vaseline• To record synaptic responses from the 10th ganglion, one AgCI lead from the amplifier was connected through an agar bridge to the compartment containing the
m V r e a c t i v a t e s t h e M - c u r r e n t , p r o d u c i n g the slow outw a r d relaxations s h o w n in Fig. 1 (right). T h e size of t h e s e r e l a x a t i o n s was u s e d to q u a n t i f y M - c u r r e n t 3. T h e s e
B
A
5.0 ¸ ,,-%
.<
4.0
t'-
2 nA
3.0
.•
•.'
, z(
2.0
~'.~ t,0
L__
A" "~'
0.0-
A.. ,L...~,''',"'~h:~:'~
-1.0 -80
I
I --60
I -40
-20
Voltage (mV)
C
D 1.0-
<( C
0.0. £)
1.0 tO (J
/
2
'4-.
(J
I
-1.0
-lOO
!
I
!
!
-80
-60
-40
-20
Voltage (mV)
0.2
o
I
I
I
w
i
loo
200
~oo
400
500
Time (msec)
Fig. 2. HFS increases M-current without affecting the time and voltage-dependence of M-current. A: currents evoked by a series of pulses (from -85 mV to -20 mV) from a holding potential of -72 mV before (left) and 3 min after (right) HFS. B: the I - V relationship for the M-current. C: the I-V relationship for the instantaneous current at the beginning of the depolarizing command pulses; open symbols, before HFS; closed symbols after HFS. Resistance = 80 Mff~ before, 76 MQ after HFS. D: time-constant of M-current activation evoked by a depolarizing command pulse to -32 mV before (C)) and after (O) HFS. Graph is a semi-logarithmic plot of (1-Mr/Mr) vs. time. M t and M e are the amount of M-current measured at the time t and at the final time of the depolarization, respectively.
283 measurements were done repetitively over a 5-15 min period to determine the baseline level of M-current. Cells were then stimulated to fire action potentials using brief (1 ms) suprathreshold current pulses (4-9 nA) passed through one of the microelectrodes. In our high frequency stimulation protocol, these current pulses were delivered in bursts (30 pulses at 30 Hz) at the rate of 1 burst every 2 s over a 1 min period. Fig. 1 (left) shows that HFS produced a 120% increase in M-current. This enhanced M-current slowly declined to control levels in about 10 min. The decay was not due to cell damage because a second period of HFS again enhanced the M-current. The enhancement of the M-current induced by HFS was found to be a robust phenomenon, observed in 27 of 34 cells. The average increase in M-current was 101 + 8% (mean + S.D.). The duration of the M-current enhancement was variable from preparation to preparation, ranging from 3 to 35 min (10 + 6 min, mean + S.D., n = 20). As illustrated in Fig. 2, the main effect of HFS on M-current was to increase its conductance. In this experiment M-current was measured at several different voltages (Fig. 2A). The M-current was enhanced by the same percentage ( - 3 0 % ) at all the voltages tested (Fig. 2B), indicating that there was no substantial change in the voltage-dependence of the M-current (n = 6). The leak resistance, which was calculated from the instantaneous current at the beginning of the depolarizing command pulses (Fig. 2C) was not affected (n = 10). HFS
also did not affect (Fig. 2D) the time-constant ( - 2 0 0 ms) for the activation of M-current (n = 10). Taken together, these results indicate that HFS increases the M-current conductance, without strongly affecting the gating process. The effect of HFS on M-current might be due to a direct effect of voltage on the channels that generate M-current. Alternatively, the effect might be indirect, involving cellular biochemistry. To explore this question we tested the effect of removing extracellular Ca 2+. Under these conditions the enhancement of M-current was very small (3.7 + 5.6%; n = 6). This finding indicates that depolarization enhances M-current by an indirect process involving Ca 2÷.
A control
15 rain
45 rain
10
sec
0.5 nA
B
A
1.5" 1
-
1
< t2
v
¢~ "1o
I'
nA
20 sec
1 sec
1.0.
~cL o.5! A
B 1
1 0.0
2 w
.
.
.
.
.
.
f" f .
.
0
I
I
I
I
30
60
90
~ 20
---~
~ so
Time (minutes) Fig. 4. HFS increases synaptically evoked slow excitatory postsynaptic current (EPSC). A: slow EPSC before (top), 12 min after and 45 min after HFS. B: EPSC amplitude as function of time before
Fig. 3. Effects of neurotransmitters on M-current enhanced by HFS. A: response (left) to 1 s application (horizontal bar) of muscarinic agonist (0.1 mM Bethanechol) ejected by pressure from a pipette close to the impaled cell. B" same as (A) but 6 min after HFS. Traces at right show M-current relaxations (induced as in Fig. 1) before (t) and during (2) agonist suppression. Dotted line in (B) marks the holding current (0.47 nA) before HFS.
and after HFS (arrowheads). Same cell as in (A). The cell was voltage clamped with one electrode (sampling frequency = 3 KHz) at -40 inV. Dotted lines in A indicate the holding current before HFS (1.2 nA). The slow EPSC was evoked at rate of 1 every 3 min by applying a brief tetanus (20 Hz, 1 s) to the preganglionic chain above the 7th ganglion. 0.5 mM curare was added to prevent the nicotinic responses.
284
Effects of neurotransmitters on the enhanced levels of M-current M-current in B sympathetic cells can be synaptically suppressed by a variety of neurotransmitters; among them, acetylcholine acting through muscarinic receptors and neuropeptides like luteinizing hormone-releasing hormone ( L H R H ) 3'1°. It was of interest to know whether the enhanced levels of M-current induced by HFS could also be suppressed by neurotransmitters, Fig. 3 shows the suppression of the M-current by muscarinic agonist before and after HFS. Stimulation in this case produced a 102% increase in M-current. The size of the current suppressed by the agonist was now much larger than before stimulation, but the percentage of the M-current suppressed was not significantly affected (49% and 46% before and after HFS respectively). Similar results were obtained in other cells (-4.9 + 1.3% change in the fraction of M-current suppressed; 147 + 13% increase in the M-current size induced by HFS; n = 7). In one cell tested with chicken L H R H , HFS increased M-current 98%, but there was only a -2% change in the fractional
A
12
rain ~
I 40
mV
15111irl I~
msec
B 0-8 rain
~
lll I
il
~
III II fNItlII II I NIl
I12°
III
mY
I rain
12-20 rain
suppression of M-current by L H R H . These results indicate that the 'enhanced' and 'resting' M-current are similar in their susceptibility to suppression by neurotransmitters. To test whether the 'enhanced' M-current is also sensitive to transmitter released by the preganglionic nerve, we measured the slow excitatory postsynaptic current (EPSC) before and after HFS. This slow EPSC is due to the muscarinic action of released acetylcholine on M-current I and so should be larger after HFS if the 'enhanced' M-current is sensitive to synaptically released transmitter. Fig. 4 shows that this is the case. The enhancement of the slow EPSC was observed in 16 out of 22 cells (82 + 9%, n = 16) and lasted between 4 and 45 min (17 +__ 11, n = 15).
Effects of high frequency stimulation on excitability Previous work has demonstrated that M-current decreases the spike rate during steady depolarizing current (spike accommodation) 2'8'1°. Studies in bullfrog sympathetic cells also show that voltage-dependent activation of M-current can occur during the fast EPSP 27 and that suppression of the M-current enhances the fast EPSP 23. It would therefore be expected that enhancement of M-current by HFS should decrease the fast EPSP and decrease excitability, To test this possibility, we studied the effect of HFS on spike accommodation and on the ability of the fast EPSP to evoke a spike. Fig. 5A shows an experiment in which HFS increased spike accommodation for several minutes. Before stimulation, a 3 n A current pulse (0.5 s) evoked 6 action potentials on 8 successive trials; after HFS the same current pulse evoked fewer spikes. This decrease in excitability lasted for - 1 5 min. The effect of HFS on the spike evoked by the fast EPSP is shown in Fig. 5B. Before HFS the fast EPSP was large enough to evoke an action potential (large deflections in Fig. 5B). After HFS the fast EPSP became subthreshold (small deflections in Fig. 5B). The probability of the EPSP evoking a spike was reduced for about 10 min, but eventually recovered. Thus the effects produced by HFS are powerful enough to produce functionally important changes in excitability.
HFS
Fig. 5. HFS reduces excitability. A: HFS increases spike accommodation. The cell was stimulated once a minute with a depolarizing current pulse (3 nA, 500 ms; indicated by the bar). Prior to HFS the current pulse consistently evoked 6 spikes (top trace). After HFS the number of spikes evoked by the current was reduced for -15 min. On the left of each trace is the time after HFS. Dotted line indicates resting potential (-43 mV) prior to HFS. B: HFS makes EPSP subthreshold. Large deflections are action potential induced by fast EPSP (0.1 Hz); small deflections in,~;cate that the EPSP was subthreshold. The two traces were recorded consecutively at the indicated times. Action potentials are truncated because of the low frequency response of the recording system.
HFS does not induce postsynaptic potentiation of the fast EPSP It has been reported that high frequency tetanic firing of frog ganglion cells induces a long-term enhancement of the fast nicotinic EPSP 14'15. Since such potentiation would counteract the effect of enhanced M-current on excitability, we were interested in studying potentiation of the fast EPSP. In these experiments we used the same Ringer composition (1.8 mM Ca 2+ and no Mg 2+ added)
285
A 4 E v 3
"1o
2
2 1
40
I
I
20
0
msec
I
4O
60
B 0.10,
o.o8 .
I
0.08.
40 .=,=,=
^
0.04 0.02
0
!
I
I
I
I
20
40
60
80
I00
C
'10 "1
Q. E <:
, ~
2
' II nA
eeeeeeeeeee ~
|
i ,i
eeee
eeee
-"~ ,.=e,:
^
0 0
0 10
I
I
I
I
20
30
40
50
Time (minutes) Fig. 6. Lack of effect of antidromic tetanic stimulation on the fast synaptic responses. The graphs are plots of time vs. the amplitude of the nicotinic response vs. time. Arrow shows HFS of the postganglionic sciatic nerve (20 Hz, 30 s). A: amplitude of the fast EPSP recorded intracellularly in the presence of sufficient curare (50/~M) to make the EPSP subthreshold. B: amplitude of the fast EPSP in intact 10th ganglion recorded extracellularly by a vaseline gap method (50/zM curare). C: amplitude of the inward current evoked by applied acetylcholine as measured by a single-electrode voltage-clamp (sampling rate = 3.5 KHz). Acetylcholine was delivered by iontophoresis (80 nA, 100 ms) from a microelectrode (1-2 MC~) filled with 1 M acetylcholine (a backcurrent of -3 nA was passed throughout the experiment). All data points are averages of 6 consecutive responses evoked at 0.1 Hz. Superimposed in the inset of each plot are two of these averaged responses recorded before (1) and after (2) HFS. A, B and C are from different experiments.
and the same postsynaptic stimulation protocol (shocks were applied at a rate of 20 Hz for 5 s to the postganglionic nerve) as described by Kumamoto and Kuba 15.
We found, however, that the likelihood of potentiating the fast nicotinic response was very low. Antidromic stimulation failed to potentiate the fast EPSP recorded
286 intracellularly in 16 out of 17 cells (Fig. 6A). Three additional cells were somatically stimulated with the same protocol as used to enhance the M-current, but no potentiation of the fast EPSP was observed. In contrast, orthodromic tetanus (20 Hz, 5 s), which has been shown to induce robust presynaptic potentiation of transmitter release 5'13, reliably enhanced the fast EPSP. It was possible that damage produced by impalement prevented potentiation of the fast EPSP. To check this possibility we measured the responses of whole intact ganglia with an extracellular vaseline gap method. The experiments were done in 50/~M curare to reduce the size of the fast EPSP and make them subthreshold. Antidromic tetanic stimulation of the postganglionic nerve failed to potentiate the fast EPSP in 8 out of 9 ganglia (Fig. 6B). We furthermore considered the possibility that the increase in M-current caused by HFS might obscure the potentiation of the fast EPSP. To address this possibility we measured the current induced by applied acetylcholine while the cell was voltage-clamped to -70 mV, a voltage sufficiently negative to inactivate M-current 1. Under these conditions the nicotinic response was not changed by HFS in 7 of 8 cells tested (Fig. 6C). Taken together, these results show that postsynaptic potentiation of the fast EPSP is hard to induce ( < 9% of the trials). Kuba and Kumamoto TM report that the probability of inducing potentiation is 66%. This difference might conceivably be due to the fact that we treated the ganglia enzymatically prior to impalement whereas Kuba and Kumamoto did not. However, it is also possible, that the high success rate reported by Kuba and Kumamoto TM is an overestimate. Their only published graph of the time course of the potentiation has a pretetanus baseline showing a positive drift that could account for the subsequent apparent potentiation. DISCUSSION
Enhancement of the M-current by high frequency stimulation (HFS) The principal finding reported in this paper is a longlasting enhancement of the M-current by HFS of action potentials in frog sympathetic ganglion cells (Fig. 1). We found that the enhancement of the M-current is a robust phenomenon, observed in 80% of the trials. The duration of enhancement varied from cell to cell, but could be as long as 35 min. The main effect of HFS was to increase the conductance of the M-current; the voltage and time-dependence of the M-current were not substantially affected (Fig. 2). The enhanced M-current was just as sensitive to exogenous or endogenous neurotransmitter as the M-current of unstimulated cells (Figs. 3 and 4).
The mechanism by which the M-current is enhanced is unclear. Recently, it has been reported that Ca 2+ entry during a short train of action potentials (10 Hz, 5 s) can produce a short-term ( - 2 0 s) enhancement of the M-current TM. Our finding that the long-term enhancement of M-current requires external Ca 2+ suggests that Ca 2÷ entry during HFS may also trigger this phenomenon. Proceeding on this assumption, one possibility is that Ca 2÷ influx releases a neurotransmitter which in turn acts back on the cell to enhance the M-current. There is a precedent both for neurotransmitter release from sympathetic neurons 22'26, and, in other systems, of enhancement of M-current by neurotransmitters ~9'24. An altogether different mechanism is suggested by the recent observation that M-current is enhanced by raising internal Ca 2+ and then inhibited at higher Ca 2+18'29. Thus one possibility is that Ca 2÷ entry through voltagedependent Ca 2÷ channels produces a modest elevation of Ca 2+ and enhances M-current. Perhaps then, muscarinic agonists produce larger, local elevations and suppress M-current, as suggested by our previous results 12.
Functional consequence of the M-current enhancement Two lines of evidence indicate that the enhancement of M-current produced by HFS affects excitability. First, HFS produced a long-lasting decrease in the number of spikes evoked by depolarizing current passed into the ganglion cells (Fig. 5A). This can be described as an increase in accommodation. Second, the fast EPSP can become subthreshold (Fig. 5B). The fact that the EPSP can become subthreshold carries special significance in frog sympathetic ganglion because these cells are typically innervated by only a single axon 28. We can therefore conclude that the effect of HFS is powerful enough to make a maximal EPSP subthreshold, thereby blocking information flow through the ganglion cell. While it is clear that the enhancement of M-current by HFS contributes to these large changes in excitability, we cannot exclude the possibility that other voltage-dependent conductances are affected by HFS and also affect excitability. Because M-current is enhanced by cell firing and, in turn, serves to decrease excitability, the M-current appears to be part of a negative feedback mechanism that regulates excitability. Previous work on the CaZ+-acti vated K + conductance indicates that it is part of such a mechanism regulating excitability 11'17, but the period involved is only several seconds. The regulation of M-current reported here acts over much longer periods. Theoretical work on cortical synaptic plasticity has pointed to the need for activity-dependent, long-term regulation of postsynaptic excitability 4.
287 REFERENCES 1 Adams, P.R. and Brown, D.A., Synaptic inhibition of the M-current: slow excitatory post-synaptic potential mechanism in bullfrog sympathetic neurones, J. Physiol., 332 (1982) 263-272. 2 Adams, ER., Brown, D.A. and Constanti, A., M-currents and other potassium currents in bullfrog sympathetic neurones, J. Physiol., 330 (1982) 537-572. 3 Adams, ER., Brown, D.A. and Constanti, A., Pharmacological inhibition of the M-current, J. Physiol., 332 (1982) 223-262. 4 Bear, M.E, Cooper, L.N. and Ebner, E E , A physiological basis for a theory of synaptic modification, Science, 237 (1987) 42-48. 5 Briggs, C.A., Brown, T.H. and McAfee, D.A., Neurophysiology and pharmacology of long-term potentiation in the rat sympathetic ganglion, J. Physiol., 359 (1985) 503-521. 6 Brown, D.A., M-currents: an update, Trends Neurosci., 11 (1988) 294-299. 7 Dodd, J. and Horn, J.P., A reclassification of B and C neurones in the 9th and 10th paravertebral sympathetic ganglia of bullfrog, J. Physiol., 334 (1983) 225-269. 8 Goh, J.W. and Pennefather, P.S., Pharmacological and physiological properties of the after-hyperpolarization current of bullfrog ganglion neurones, J. Physiol., 394 (1987) 315-330. 9 Jan, Y.N., Jan, L.Y. and Kuffler, S.W., A peptide as a possible transmitter in sympathetic ganglia of the frog, Proc. Natl. Acad. Sci. USA, 76 (1979) 1501-1505. 10 Jones, S.W., Muscarinic and peptidergic excitation of bullfrog sympathetic neurones, J. Physiol., 366 (1985) 63-87. 11 Jones, S.W. and Adams, P.R., The M-current and other potassium currents of vertebrate neurons. In L. Kaczmarec and I.B. Levitan (Eds.), Neuromodulation, Oxford University Press, NY, 1987. 12 Kirkwood, A. and Lisman, J.E., Muscarinic suppression of the M-current is mediated by a rise in internal Ca 2+ concentration, Neuron, 6 (1991) 1009-1014. 13 Koyano, K., Kuba, K. and Minota, S., Long-term potentiation of transmitter release induced by repetitive presynaptic activities in bull-frog sympathetic ganglia, J. Physiol., 359 (1991) 219-233. 14 Kuba, K. and Kumamoto, E., Long-term potentiation in vertebrate synapse: a variety of cascades with common subprocesses, Prog. Neurobiol., 34 (1990) 197-269. 15 Kumamoto, E. and Kuba, K., Sustained raise in ACH sensitivity of a sympathetic ganglion cell induced by postsynaptic elec-
trical activities, Nature, 305 (1983) 145-146. 16 Madison, D.V. and Nicoll, R.A., Noradrenaline blocks accommodation of pyramidal cell discharge in the hippocampus, Nature, 299 (1982) 636-638. 17 Madison, D.V. and Nicoll, R.A., Control of repetitive discharge of rat CA1 pyramidal neurones in vitro, J. Physiol., 354 (1984) 319-331. 18 Marrion, N.V., Zucker, R.S., Marsh, S.J. and Adams, ER., Modulation of M-current by intracellular Ca 2÷, Neuron, 6 (1991) 533-545. 19 Moore, S.D., Madamba, S.G., Joels, M. and Siggins, G.R., Somatostatin augments the M-current in hippocampal neurons, Science, 239 (1988) 278-280. 20 Nishi, S., Soeda, H. and Koketsu, K., Studies on sympathetic B and C neurons and patterns of preganglionic innervation, J. Cell Comp. Physiol., 66 (1965) 19-32. 21 Pennefather, P., Lancaster, B., Adams, P.R. and Nicholl, R., Two distinct Ca-dependent K currents in bullfrog sympathetic ganglion cells, Proc. Natl. Acad. Science USA, 82 (1985) 30403044. 22 Rubio, R., Bencherif, M. and Berne, R.M., Release of purines from postsynaptic structures of amphibian ganglia, J. Neurochem., 51 (1988) 1717-1723. 23 Schulman, J.A. and Weight, E E , Synaptic transmission: longlasting potentiation by a post synaptic mechanism, Science, 194 (1976) 1437-1439. 24 Sims, S.M., Singer, J.J. and Walsh, J.V., Antagonistic adrenergic-muscarinic regulation of M current in smooth muscle cells, Science, 239 (1988) 190-193. 25 Smith, S.J., MacDermott, A.B. and Weight, E E , Detection of intracellular Ca 2+ transients in sympathetic neurones using arsenazo III, Nature, 304 (1983) 350-352. 26 Suetake, K., Kojima, H., Inanaga, K. and Koketsu, K., Cathecholamines are released from non-synaptic cell-soma membrane: histochemical evidence in bullfrog sympathic ganglion cells, Brain Res., 205 (1981) 436-440. 27 Tosaka, T., Tasaka, J., Miyazaki, T. and Libet, B., Hyperpolarization following activation of K + channels by excitatory postsynaptic potentials, Nature, 305 (1983) 148-150. 28 Weitsen, H.A. and Weight, E E , Synaptic innervation of sympathetic ganglion cells in the bullfrog, Brain Res., 128 (1977) 197-211. 29 Yu, EP., Marrion, N.V., Villaroel, A. and Adams, E, Effects of intracellular calcium on M-current revealed by intracellular perfusion, Biophys. J., 59 (1991) 78A.
281
BRES 17735
Action potentials produce a long-term enhancement of M-current in frog sympathetic ganglion Alfredo Kirkwood* and John E. Lisman Program in Biophysics, Department of Biology and the Center for Complex Systems, Brandeis University, Waltham MA 02154 (USA) (Accepted 31 December 1991)
Key words: Long-term potentiation; Excitability; Feedback; Potassium conductance
M-current is a voltage-gated K+ current that can be turned off by the muscarinic action of acetylcholine. We examined the effects of postsynaptic action potential firing on the level of M-current in B-cells of the bullfrog sympathetic ganglion. High frequency stimulation of action potentials induced an approximately two-fold increase in the level of the M-current that could last up to 35 min. The 'enhanced' M-current was similar to the 'resting' one in its time-dependence, voltage-dependence and sensitivity to neurotransmitters. Experiments were undertaken to examine the functional consequences of the enhanced M-current. Following high frequency stimulation the number of spikes evoked by depolarizing current was reduced. In addition, the excitatory postsynaptic potential (EPSP) evoked by maximal input became subthreshold, thereby blocking information flow through the ganglion cell. These results indicate that the enhancement of M-current by spikes provides a negative feedback mechanism for the control of excitability. It has been reported that postsynaptic stimulation of ganglion cells also produces a long-term increase in the nicotinic EPSP, but we were unable to confirm this observation. INTRODUCTION K ÷ conductances are important determinants of neuronal excitability, but the role of different K ÷ conductances and the mechanisms controlling these conductances remain unclear. O n e K ÷ conductance k n o w n to affect excitability is the Ca2+-activated K + conductance 11'16'17'21. The intracellular Ca 2÷ elevation caused by action potentials turns on Ca2+-activated K ÷ channels which then reduce subsequent firing. These K ÷ channels turn off as Ca 2÷ levels fall, a process that takes about 10 s in frog sympathetic ganglion cells2s. These findings demonstrate that there is a negative feedback mechanism that can regulate excitability. Here we report that frog sympathetic ganglion cells have a second feedback mechanism that regulates excitability over much longer time periods ( - 3 0 min). This mechanism involves M-current, a non-inactivating, voltage-dependent K + current first identified in frog sympathetic ganglion cells 2, but subsequently found in m a n y central and peripheral n e u r o n s 6. M-current is the major K ÷ current near spike threshold and therefore has marked effects on n e u r o n a l excitability2'21'23. Previous work on frog sympathetic ganglion cells has shown that the magnitude of M-current can be modulated by extrinsic factors including acetylcholine2, peptide neurotransmitters 9 and nucleotides2. O u r results show M-current
can also be modulated by action potentials intrinsic to ganglion cells and that this modulation has long-term effects on cell excitability. MATERIALS AND METHODS
Dissection and mounting Experiments were done on B cells in the 9th and 10th sympathetic ganglia of the bullfrog (Rana catesbeiana). The ganglia were excised, mounted in a small plexiglas chamber and treated with 3% collagenase (type IIs Sigma) for 5-10 rain to facilitate desheathing. Ringer composition was (in mM) NaCI 115.5, KCI 2, CaC12 3.6, MgCI2 5, TRIS 5 (pH, 7.3). All drugs except BAPTA (from Boehringer) and chicken LHRH (from Peninsula) were purchased from Sigma. Recording procedures Cells were impaled with conventional microelectrodes filled with 3 M KCI (60-90 Mff2). B-cells were identified by their short latencies and their responsiveness to presynaptic stimulation above the 7th ganglion2°. The strength of presynaptic electrical stimulation was always adjusted to give a maximal postsynaptic response. Muscarinic synaptic currents and agonist-induced nicotinic currents were recorded using discontinuous single electrode voltage-clamp (Axoclamp 2A; sampling frequency = 3-4 KHz). These current records were filtered at 200 Hz. To measure the M-current, the cells were voltage-clamped with two electrodes and the data filtered at 2 KHz. Extracellular fast synaptic potentials were recorded in the 10th ganglion using an AC differential amplifier. In these experiments, the whole sympathetic chain (from the 10th to the 5th ganglion) was freed of connective tissue. The chain was placed in a plexiglas chamber which was divided in a linear array of compartments interconnected by small slits. Each ganglia was placed in a different compartment with the interconnecting nerve running through the
* Present address: Center for Neural Sciences, Brown University, Providence, RI 02912, USA. Correspondence: J.E. Lisman, Department of Biology, Brandeis University, Waltham, MA 02154, USA.
282 3.0
c
10th ganglion; the other lead was connected to the compartment containing the postganglionic nerve.
2.0
2 •
c
k
2
• 00
~
1.o
RESULTS • 1
O0
•
1 0.0
000
•
•
•
A
1'0
Enhancement of the M-current by high frequency stimulation (HFS)
A
2'0
310
,~
i sec
E x p e r i m e n t s w e r e d o n e in bullfrog B type g a n g l i o n cells 7'2°. Cells in the intact ganglion w e r e i m p a l e d with
Time (minutes) Fig. 1. Time course (left) of enhancement of M-current by high frequency stimulation (HFS) (arrowheads). Right: high time resolution view of current relaxations due to M-current before (1) and after (2) HFS. Holding voltage -35 mV; currents evoked by 1 s hyperpolarizing pulse to -65 mV.
two m i c r o e l e c t r o d e s .
M-current
was m e a s u r e d
under
v o l t a g e - c l a m p by s t a n d a r d p r o c e d u r e s 3. T h e m e m b r a n e p o t e n t i a l was h e l d at - 3 5 m V while r e p e t i t i v e l y applying 1 s c o m m a n d pulses to - 6 5 mV. H y p e r p o l a r i z a t i o n to - 6 5 m V turns off the M - c u r r e n t . R e p o l a r i z a t i o n to - 3 5
slits, which were sealed with vaseline• To record synaptic responses from the 10th ganglion, one AgCI lead from the amplifier was connected through an agar bridge to the compartment containing the
m V r e a c t i v a t e s t h e M - c u r r e n t , p r o d u c i n g the slow outw a r d relaxations s h o w n in Fig. 1 (right). T h e size of t h e s e r e l a x a t i o n s was u s e d to q u a n t i f y M - c u r r e n t 3. T h e s e
B
A
5.0 ¸ ,,-%
.<
4.0
t'-
2 nA
3.0
.•
•.'
, z(
2.0
~'.~ t,0
L__
A" "~'
0.0-
A.. ,L...~,''',"'~h:~:'~
-1.0 -80
I
I --60
I -40
-20
Voltage (mV)
C
D 1.0-
<( C
0.0. £)
1.0 tO (J
/
2
'4-.
(J
I
-1.0
-lOO
!
I
!
!
-80
-60
-40
-20
Voltage (mV)
0.2
o
I
I
I
w
i
loo
200
~oo
400
500
Time (msec)
Fig. 2. HFS increases M-current without affecting the time and voltage-dependence of M-current. A: currents evoked by a series of pulses (from -85 mV to -20 mV) from a holding potential of -72 mV before (left) and 3 min after (right) HFS. B: the I - V relationship for the M-current. C: the I-V relationship for the instantaneous current at the beginning of the depolarizing command pulses; open symbols, before HFS; closed symbols after HFS. Resistance = 80 Mff~ before, 76 MQ after HFS. D: time-constant of M-current activation evoked by a depolarizing command pulse to -32 mV before (C)) and after (O) HFS. Graph is a semi-logarithmic plot of (1-Mr/Mr) vs. time. M t and M e are the amount of M-current measured at the time t and at the final time of the depolarization, respectively.
283 measurements were done repetitively over a 5-15 min period to determine the baseline level of M-current. Cells were then stimulated to fire action potentials using brief (1 ms) suprathreshold current pulses (4-9 nA) passed through one of the microelectrodes. In our high frequency stimulation protocol, these current pulses were delivered in bursts (30 pulses at 30 Hz) at the rate of 1 burst every 2 s over a 1 min period. Fig. 1 (left) shows that HFS produced a 120% increase in M-current. This enhanced M-current slowly declined to control levels in about 10 min. The decay was not due to cell damage because a second period of HFS again enhanced the M-current. The enhancement of the M-current induced by HFS was found to be a robust phenomenon, observed in 27 of 34 cells. The average increase in M-current was 101 + 8% (mean + S.D.). The duration of the M-current enhancement was variable from preparation to preparation, ranging from 3 to 35 min (10 + 6 min, mean + S.D., n = 20). As illustrated in Fig. 2, the main effect of HFS on M-current was to increase its conductance. In this experiment M-current was measured at several different voltages (Fig. 2A). The M-current was enhanced by the same percentage ( - 3 0 % ) at all the voltages tested (Fig. 2B), indicating that there was no substantial change in the voltage-dependence of the M-current (n = 6). The leak resistance, which was calculated from the instantaneous current at the beginning of the depolarizing command pulses (Fig. 2C) was not affected (n = 10). HFS
also did not affect (Fig. 2D) the time-constant ( - 2 0 0 ms) for the activation of M-current (n = 10). Taken together, these results indicate that HFS increases the M-current conductance, without strongly affecting the gating process. The effect of HFS on M-current might be due to a direct effect of voltage on the channels that generate M-current. Alternatively, the effect might be indirect, involving cellular biochemistry. To explore this question we tested the effect of removing extracellular Ca 2+. Under these conditions the enhancement of M-current was very small (3.7 + 5.6%; n = 6). This finding indicates that depolarization enhances M-current by an indirect process involving Ca 2÷.
A control
15 rain
45 rain
10
sec
0.5 nA
B
A
1.5" 1
-
1
< t2
v
¢~ "1o
I'
nA
20 sec
1 sec
1.0.
~cL o.5! A
B 1
1 0.0
2 w
.
.
.
.
.
.
f" f .
.
0
I
I
I
I
30
60
90
~ 20
---~
~ so
Time (minutes) Fig. 4. HFS increases synaptically evoked slow excitatory postsynaptic current (EPSC). A: slow EPSC before (top), 12 min after and 45 min after HFS. B: EPSC amplitude as function of time before
Fig. 3. Effects of neurotransmitters on M-current enhanced by HFS. A: response (left) to 1 s application (horizontal bar) of muscarinic agonist (0.1 mM Bethanechol) ejected by pressure from a pipette close to the impaled cell. B" same as (A) but 6 min after HFS. Traces at right show M-current relaxations (induced as in Fig. 1) before (t) and during (2) agonist suppression. Dotted line in (B) marks the holding current (0.47 nA) before HFS.
and after HFS (arrowheads). Same cell as in (A). The cell was voltage clamped with one electrode (sampling frequency = 3 KHz) at -40 inV. Dotted lines in A indicate the holding current before HFS (1.2 nA). The slow EPSC was evoked at rate of 1 every 3 min by applying a brief tetanus (20 Hz, 1 s) to the preganglionic chain above the 7th ganglion. 0.5 mM curare was added to prevent the nicotinic responses.
284
Effects of neurotransmitters on the enhanced levels of M-current M-current in B sympathetic cells can be synaptically suppressed by a variety of neurotransmitters; among them, acetylcholine acting through muscarinic receptors and neuropeptides like luteinizing hormone-releasing hormone ( L H R H ) 3'1°. It was of interest to know whether the enhanced levels of M-current induced by HFS could also be suppressed by neurotransmitters, Fig. 3 shows the suppression of the M-current by muscarinic agonist before and after HFS. Stimulation in this case produced a 102% increase in M-current. The size of the current suppressed by the agonist was now much larger than before stimulation, but the percentage of the M-current suppressed was not significantly affected (49% and 46% before and after HFS respectively). Similar results were obtained in other cells (-4.9 + 1.3% change in the fraction of M-current suppressed; 147 + 13% increase in the M-current size induced by HFS; n = 7). In one cell tested with chicken L H R H , HFS increased M-current 98%, but there was only a -2% change in the fractional
A
12
rain ~
I 40
mV
15111irl I~
msec
B 0-8 rain
~
lll I
il
~
III II fNItlII II I NIl
I12°
III
mY
I rain
12-20 rain
suppression of M-current by L H R H . These results indicate that the 'enhanced' and 'resting' M-current are similar in their susceptibility to suppression by neurotransmitters. To test whether the 'enhanced' M-current is also sensitive to transmitter released by the preganglionic nerve, we measured the slow excitatory postsynaptic current (EPSC) before and after HFS. This slow EPSC is due to the muscarinic action of released acetylcholine on M-current I and so should be larger after HFS if the 'enhanced' M-current is sensitive to synaptically released transmitter. Fig. 4 shows that this is the case. The enhancement of the slow EPSC was observed in 16 out of 22 cells (82 + 9%, n = 16) and lasted between 4 and 45 min (17 +__ 11, n = 15).
Effects of high frequency stimulation on excitability Previous work has demonstrated that M-current decreases the spike rate during steady depolarizing current (spike accommodation) 2'8'1°. Studies in bullfrog sympathetic cells also show that voltage-dependent activation of M-current can occur during the fast EPSP 27 and that suppression of the M-current enhances the fast EPSP 23. It would therefore be expected that enhancement of M-current by HFS should decrease the fast EPSP and decrease excitability, To test this possibility, we studied the effect of HFS on spike accommodation and on the ability of the fast EPSP to evoke a spike. Fig. 5A shows an experiment in which HFS increased spike accommodation for several minutes. Before stimulation, a 3 n A current pulse (0.5 s) evoked 6 action potentials on 8 successive trials; after HFS the same current pulse evoked fewer spikes. This decrease in excitability lasted for - 1 5 min. The effect of HFS on the spike evoked by the fast EPSP is shown in Fig. 5B. Before HFS the fast EPSP was large enough to evoke an action potential (large deflections in Fig. 5B). After HFS the fast EPSP became subthreshold (small deflections in Fig. 5B). The probability of the EPSP evoking a spike was reduced for about 10 min, but eventually recovered. Thus the effects produced by HFS are powerful enough to produce functionally important changes in excitability.
HFS
Fig. 5. HFS reduces excitability. A: HFS increases spike accommodation. The cell was stimulated once a minute with a depolarizing current pulse (3 nA, 500 ms; indicated by the bar). Prior to HFS the current pulse consistently evoked 6 spikes (top trace). After HFS the number of spikes evoked by the current was reduced for -15 min. On the left of each trace is the time after HFS. Dotted line indicates resting potential (-43 mV) prior to HFS. B: HFS makes EPSP subthreshold. Large deflections are action potential induced by fast EPSP (0.1 Hz); small deflections in,~;cate that the EPSP was subthreshold. The two traces were recorded consecutively at the indicated times. Action potentials are truncated because of the low frequency response of the recording system.
HFS does not induce postsynaptic potentiation of the fast EPSP It has been reported that high frequency tetanic firing of frog ganglion cells induces a long-term enhancement of the fast nicotinic EPSP 14'15. Since such potentiation would counteract the effect of enhanced M-current on excitability, we were interested in studying potentiation of the fast EPSP. In these experiments we used the same Ringer composition (1.8 mM Ca 2+ and no Mg 2+ added)
285
A 4 E v 3
"1o
2
2 1
40
I
I
20
0
msec
I
4O
60
B 0.10,
o.o8 .
I
0.08.
40 .=,=,=
^
0.04 0.02
0
!
I
I
I
I
20
40
60
80
I00
C
'10 "1
Q. E <:
, ~
2
' II nA
eeeeeeeeeee ~
|
i ,i
eeee
eeee
-"~ ,.=e,:
^
0 0
0 10
I
I
I
I
20
30
40
50
Time (minutes) Fig. 6. Lack of effect of antidromic tetanic stimulation on the fast synaptic responses. The graphs are plots of time vs. the amplitude of the nicotinic response vs. time. Arrow shows HFS of the postganglionic sciatic nerve (20 Hz, 30 s). A: amplitude of the fast EPSP recorded intracellularly in the presence of sufficient curare (50/~M) to make the EPSP subthreshold. B: amplitude of the fast EPSP in intact 10th ganglion recorded extracellularly by a vaseline gap method (50/zM curare). C: amplitude of the inward current evoked by applied acetylcholine as measured by a single-electrode voltage-clamp (sampling rate = 3.5 KHz). Acetylcholine was delivered by iontophoresis (80 nA, 100 ms) from a microelectrode (1-2 MC~) filled with 1 M acetylcholine (a backcurrent of -3 nA was passed throughout the experiment). All data points are averages of 6 consecutive responses evoked at 0.1 Hz. Superimposed in the inset of each plot are two of these averaged responses recorded before (1) and after (2) HFS. A, B and C are from different experiments.
and the same postsynaptic stimulation protocol (shocks were applied at a rate of 20 Hz for 5 s to the postganglionic nerve) as described by Kumamoto and Kuba 15.
We found, however, that the likelihood of potentiating the fast nicotinic response was very low. Antidromic stimulation failed to potentiate the fast EPSP recorded
286 intracellularly in 16 out of 17 cells (Fig. 6A). Three additional cells were somatically stimulated with the same protocol as used to enhance the M-current, but no potentiation of the fast EPSP was observed. In contrast, orthodromic tetanus (20 Hz, 5 s), which has been shown to induce robust presynaptic potentiation of transmitter release 5'13, reliably enhanced the fast EPSP. It was possible that damage produced by impalement prevented potentiation of the fast EPSP. To check this possibility we measured the responses of whole intact ganglia with an extracellular vaseline gap method. The experiments were done in 50/~M curare to reduce the size of the fast EPSP and make them subthreshold. Antidromic tetanic stimulation of the postganglionic nerve failed to potentiate the fast EPSP in 8 out of 9 ganglia (Fig. 6B). We furthermore considered the possibility that the increase in M-current caused by HFS might obscure the potentiation of the fast EPSP. To address this possibility we measured the current induced by applied acetylcholine while the cell was voltage-clamped to -70 mV, a voltage sufficiently negative to inactivate M-current 1. Under these conditions the nicotinic response was not changed by HFS in 7 of 8 cells tested (Fig. 6C). Taken together, these results show that postsynaptic potentiation of the fast EPSP is hard to induce ( < 9% of the trials). Kuba and Kumamoto TM report that the probability of inducing potentiation is 66%. This difference might conceivably be due to the fact that we treated the ganglia enzymatically prior to impalement whereas Kuba and Kumamoto did not. However, it is also possible, that the high success rate reported by Kuba and Kumamoto TM is an overestimate. Their only published graph of the time course of the potentiation has a pretetanus baseline showing a positive drift that could account for the subsequent apparent potentiation. DISCUSSION
Enhancement of the M-current by high frequency stimulation (HFS) The principal finding reported in this paper is a longlasting enhancement of the M-current by HFS of action potentials in frog sympathetic ganglion cells (Fig. 1). We found that the enhancement of the M-current is a robust phenomenon, observed in 80% of the trials. The duration of enhancement varied from cell to cell, but could be as long as 35 min. The main effect of HFS was to increase the conductance of the M-current; the voltage and time-dependence of the M-current were not substantially affected (Fig. 2). The enhanced M-current was just as sensitive to exogenous or endogenous neurotransmitter as the M-current of unstimulated cells (Figs. 3 and 4).
The mechanism by which the M-current is enhanced is unclear. Recently, it has been reported that Ca 2+ entry during a short train of action potentials (10 Hz, 5 s) can produce a short-term ( - 2 0 s) enhancement of the M-current TM. Our finding that the long-term enhancement of M-current requires external Ca 2+ suggests that Ca 2÷ entry during HFS may also trigger this phenomenon. Proceeding on this assumption, one possibility is that Ca 2÷ influx releases a neurotransmitter which in turn acts back on the cell to enhance the M-current. There is a precedent both for neurotransmitter release from sympathetic neurons 22'26, and, in other systems, of enhancement of M-current by neurotransmitters ~9'24. An altogether different mechanism is suggested by the recent observation that M-current is enhanced by raising internal Ca 2+ and then inhibited at higher Ca 2+18'29. Thus one possibility is that Ca 2÷ entry through voltagedependent Ca 2÷ channels produces a modest elevation of Ca 2+ and enhances M-current. Perhaps then, muscarinic agonists produce larger, local elevations and suppress M-current, as suggested by our previous results 12.
Functional consequence of the M-current enhancement Two lines of evidence indicate that the enhancement of M-current produced by HFS affects excitability. First, HFS produced a long-lasting decrease in the number of spikes evoked by depolarizing current passed into the ganglion cells (Fig. 5A). This can be described as an increase in accommodation. Second, the fast EPSP can become subthreshold (Fig. 5B). The fact that the EPSP can become subthreshold carries special significance in frog sympathetic ganglion because these cells are typically innervated by only a single axon 28. We can therefore conclude that the effect of HFS is powerful enough to make a maximal EPSP subthreshold, thereby blocking information flow through the ganglion cell. While it is clear that the enhancement of M-current by HFS contributes to these large changes in excitability, we cannot exclude the possibility that other voltage-dependent conductances are affected by HFS and also affect excitability. Because M-current is enhanced by cell firing and, in turn, serves to decrease excitability, the M-current appears to be part of a negative feedback mechanism that regulates excitability. Previous work on the CaZ+-acti vated K + conductance indicates that it is part of such a mechanism regulating excitability 11'17, but the period involved is only several seconds. The regulation of M-current reported here acts over much longer periods. Theoretical work on cortical synaptic plasticity has pointed to the need for activity-dependent, long-term regulation of postsynaptic excitability 4.
287 REFERENCES 1 Adams, P.R. and Brown, D.A., Synaptic inhibition of the M-current: slow excitatory post-synaptic potential mechanism in bullfrog sympathetic neurones, J. Physiol., 332 (1982) 263-272. 2 Adams, ER., Brown, D.A. and Constanti, A., M-currents and other potassium currents in bullfrog sympathetic neurones, J. Physiol., 330 (1982) 537-572. 3 Adams, ER., Brown, D.A. and Constanti, A., Pharmacological inhibition of the M-current, J. Physiol., 332 (1982) 223-262. 4 Bear, M.E, Cooper, L.N. and Ebner, E E , A physiological basis for a theory of synaptic modification, Science, 237 (1987) 42-48. 5 Briggs, C.A., Brown, T.H. and McAfee, D.A., Neurophysiology and pharmacology of long-term potentiation in the rat sympathetic ganglion, J. Physiol., 359 (1985) 503-521. 6 Brown, D.A., M-currents: an update, Trends Neurosci., 11 (1988) 294-299. 7 Dodd, J. and Horn, J.P., A reclassification of B and C neurones in the 9th and 10th paravertebral sympathetic ganglia of bullfrog, J. Physiol., 334 (1983) 225-269. 8 Goh, J.W. and Pennefather, P.S., Pharmacological and physiological properties of the after-hyperpolarization current of bullfrog ganglion neurones, J. Physiol., 394 (1987) 315-330. 9 Jan, Y.N., Jan, L.Y. and Kuffler, S.W., A peptide as a possible transmitter in sympathetic ganglia of the frog, Proc. Natl. Acad. Sci. USA, 76 (1979) 1501-1505. 10 Jones, S.W., Muscarinic and peptidergic excitation of bullfrog sympathetic neurones, J. Physiol., 366 (1985) 63-87. 11 Jones, S.W. and Adams, P.R., The M-current and other potassium currents of vertebrate neurons. In L. Kaczmarec and I.B. Levitan (Eds.), Neuromodulation, Oxford University Press, NY, 1987. 12 Kirkwood, A. and Lisman, J.E., Muscarinic suppression of the M-current is mediated by a rise in internal Ca 2+ concentration, Neuron, 6 (1991) 1009-1014. 13 Koyano, K., Kuba, K. and Minota, S., Long-term potentiation of transmitter release induced by repetitive presynaptic activities in bull-frog sympathetic ganglia, J. Physiol., 359 (1991) 219-233. 14 Kuba, K. and Kumamoto, E., Long-term potentiation in vertebrate synapse: a variety of cascades with common subprocesses, Prog. Neurobiol., 34 (1990) 197-269. 15 Kumamoto, E. and Kuba, K., Sustained raise in ACH sensitivity of a sympathetic ganglion cell induced by postsynaptic elec-
trical activities, Nature, 305 (1983) 145-146. 16 Madison, D.V. and Nicoll, R.A., Noradrenaline blocks accommodation of pyramidal cell discharge in the hippocampus, Nature, 299 (1982) 636-638. 17 Madison, D.V. and Nicoll, R.A., Control of repetitive discharge of rat CA1 pyramidal neurones in vitro, J. Physiol., 354 (1984) 319-331. 18 Marrion, N.V., Zucker, R.S., Marsh, S.J. and Adams, ER., Modulation of M-current by intracellular Ca 2÷, Neuron, 6 (1991) 533-545. 19 Moore, S.D., Madamba, S.G., Joels, M. and Siggins, G.R., Somatostatin augments the M-current in hippocampal neurons, Science, 239 (1988) 278-280. 20 Nishi, S., Soeda, H. and Koketsu, K., Studies on sympathetic B and C neurons and patterns of preganglionic innervation, J. Cell Comp. Physiol., 66 (1965) 19-32. 21 Pennefather, P., Lancaster, B., Adams, P.R. and Nicholl, R., Two distinct Ca-dependent K currents in bullfrog sympathetic ganglion cells, Proc. Natl. Acad. Science USA, 82 (1985) 30403044. 22 Rubio, R., Bencherif, M. and Berne, R.M., Release of purines from postsynaptic structures of amphibian ganglia, J. Neurochem., 51 (1988) 1717-1723. 23 Schulman, J.A. and Weight, E E , Synaptic transmission: longlasting potentiation by a post synaptic mechanism, Science, 194 (1976) 1437-1439. 24 Sims, S.M., Singer, J.J. and Walsh, J.V., Antagonistic adrenergic-muscarinic regulation of M current in smooth muscle cells, Science, 239 (1988) 190-193. 25 Smith, S.J., MacDermott, A.B. and Weight, E E , Detection of intracellular Ca 2+ transients in sympathetic neurones using arsenazo III, Nature, 304 (1983) 350-352. 26 Suetake, K., Kojima, H., Inanaga, K. and Koketsu, K., Cathecholamines are released from non-synaptic cell-soma membrane: histochemical evidence in bullfrog sympathic ganglion cells, Brain Res., 205 (1981) 436-440. 27 Tosaka, T., Tasaka, J., Miyazaki, T. and Libet, B., Hyperpolarization following activation of K + channels by excitatory postsynaptic potentials, Nature, 305 (1983) 148-150. 28 Weitsen, H.A. and Weight, E E , Synaptic innervation of sympathetic ganglion cells in the bullfrog, Brain Res., 128 (1977) 197-211. 29 Yu, EP., Marrion, N.V., Villaroel, A. and Adams, E, Effects of intracellular calcium on M-current revealed by intracellular perfusion, Biophys. J., 59 (1991) 78A.