- Email: [email protected]
Glutamate-induced inhibition of paired pulse facilitation of monosynaptic excitatory post-synaptic potentials in frog spinal motoneurons
Brahz Research, 597 (1992) 124-130 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
124
BRES 18295
Glutamate-induced inhibition of paired pulse facilitation of monosynaptic excitatory post-synaptic potentials in frog spinal motoneurons F u s a o N a k a m u r a , Miyuki K u n o a n d S h i u s h i M a t s u u r a Department of Physiology., Osaka City Unit'ersity Medical School, Abeno-ku, Osaka (Japan) (Accepted 30 June 1992)
Key words: Excitatory amino acid; Autoregulation; Short-term potentiation; Modulation of transmitter release
To evaluate actions of glutamate on excitatory synaptic transmission in the central nervous system, we examined glutamate-induced changes in the paired pulse facilitation of monosynaptic excitatory post-synaptic potentials evoked by stimulation of the lateral column fibers (LC-EPSPs) on lumbar motoneurons in the frog spinal cord. Glutamate (1 raM) depolarized motoneurons both in the presence and absence of Mg 2+. In most cells perfused with Mg2+-free or high Ca2+-Mg 2+ solutions, the glutamate potential was accompanied by a reduction in peak amplitude of EPSPs, although the degree of change varied with the cells. Glutamate enhanced the EPSP amplitude in a few cells with Mg2+-free and high Ca2+-Mg 2+ solutions, and in most cells with high Mg 2+ medium. In 3/5 cells tested, the paired pulse facilitation of EPSPs was reduced by glutamate when the EPSP amplitude either increased or decreased. NMDA (50 /~M), kainate (50-100 /zM), quisqualate (5-50 /~M) and L-2-amino-4-phosphonobutyrate (L-AP4, 1 raM) also decreased the facilitation in about hull of the cells tested. The glutamate-induced decrease in the facilitation was observed in both the presence and absence of Mg 2+ and was not affected by the concomitant application of glutamate and antagonists for non-NMDA or NMDA receptors, such as 6-cyano-7-nitro-quinoxalinediones(CNQX, 60 ~M) or 2-amino-5-phosphonovalerate (APV, 250/~M). Glutamate reduced the facilitation of excitatory post-synaptic currents (EPSCs) recorded at a constant membrane potential under voltage clamp, when the EPSC amplitude either increased or decreased and when the input conductance either increased or decreased. Thus changes in membrane potential or input conductance of the post-synaptic membrane were not essential to the glutamate-induced inhibition of the facilitation. These results suggest that glutamate modulates release of excitatory transmitters via mechanisms insensitive to Mg 2+, CNQX and APV.
INTRODUCTION Excitatory amino acids are distributed throughout the central nervous system (CNS) and are putative excitatory neurotransmitters 28,s5. The exogenous application of L-glutamate mimics the situation when the level of extracellular glutamate is elevated throughout the CNS, and allows the study of the modulation of neuronal activity by the action of glutamate. Biochemical studies have shown that excitatory amino acids either inhibit 3,2°,21.24 or enhance 3,7,13,1~ the release of excitatory amino acids in the CNS. A paired pulse • facilitation of EPSPs has been considered to result from a variation in transmitter release 38 and may be valid to monitor actions of glutamate on the transmitter release. However, there are relatively few reports
on the action of glutamate on the facilitation at the excitatory amino acid-mediated synaptic transmission in the CNS. In the frog spinal cord, the excitatory synapses between the descending lateral column fiber and motoneurons may utilize excitatory amino acids as neurotransmitters 3~. Intraarterial perfusion z2 enables rapid application of glutamate throughout the spinal cord. We analyzed glutamate-induced changes in the paired pulse facilitation of monosynaptic EPSPs in Mge+-free and Mg2+-containing solutions. Evidence was obtained that glutamate reversibly reduced the paired pulse facilitation of EPSPs via mechanisms insensitive to Mg 2+ and antagonists for both NMDA and non-NMDA receptors (APV and CNQX), and that changes in membrane potential or input conductance of the post-synaptic membrane were not essential to
Correspondencc: M. Kuno, Department of Physiology, Osaka City University Medical School, Abeno-ku, Osaka 545, Japan. Fax: (81) (06) 632-7114.
125 the r e d u c t i o n of the facilitation. T h u s it seems t h a t g l u t a m a t e m o d u l a t e s the t r a n s m i t t e r release at the excitatory a m i n o a c i d - m e d i a t e d excitatory synapses. A p r e l i m i n a r y account has b e e n m a d e 27.
decay phase of EPSPs s2. The location of synapses generating monosynaptic EPSPs was roughly estimated from the shape indices 32, using parameters describing a nerve cell model following the study by Jack et al. on la-fiber EPSPs 12'16. The input conductance was determined from the steady-state current responses to 100-500 ms, 4-10 mV hyperpolarizing command voltage steps. Input resistance of the cells ranged from 1.3 to 10 M~.
M A T E R I A L S AND M E T H O D S
Data analysis Adult bullfrogs (Rana catesbiana) weighing 150-350 g were anesthetized with pentobarbital sodium (Nembutal; Abbott) (30 mg/kg of body weight) and the entire spinal cord with attached spinal roots was isolated. The isolated spinal cord was placed in the recording chamber and perfused with ox'ygenated Ringers' solution through a fine polyethylene cannula inserted into the ventral spinal artery. Details of this procedure have been described elsewhere 15'22. All experiments were carried out at room temperature (15-24°O.
Sohaions The Mg2+-free Ringers' solution contained (in mM): 111 NaCI, 2.0 KCI, 1.8 CaCI 2, 1.5 NaHCO 3 (pH 7.3-7.4) and was supplemented with 0.1% glucose. In later experiments, the pH was adjusted using 10 mM N-[2-hydroxyethyl] piperazine-N'-[2-ethanesulfonic acid] (HEPES)-NaOH. The high Mg 2+ medium was made by adding 4 mM MgCI2 to the above medium. In some experiments, the high Ca2+-Mg 2+ (4.5 mM Ca2+ plus 8 mM Mg2+ ) medium was used to eliminate poly-synaptic inputs 14'37. The osmolarity of the high Ca 2+Mg 2+ and the high Mg2+ solutions was compensated for by changes in the concentration of NaCI.
Electrical signals were digitized at 10-20 kHz using an analoguedigital converter (MacLab/4, Analogue Digital Instrument) and were stored on a personal computer (Macintosh SE/30, Apple). Averages of 4-20 records and measurements of peak amplitude of EPSP or EPSC were conducted using a computer program (Scope ver. 3.1, Analogue Digital Instrument) or an averaging computer (ATAC 350, Nihon Kohden). When a notch was observed on the rising phase of the EPSPs, the amplitude of the monosynaptic EPSP was measured at this notch. The peak amplitude of the EPSPs was within 5 mV in most cells and was not corrected for non-linear summation.
Chemicals L-Glutamate (Sigma), kainate (Sigma), quisqualate (Sigma), Nmethyl-D-aspartate (NMDA, Sigma), L-2-amino-4-phoshphonobutyrate (L-AP4, Sigma), 2-amino-5-phosphonovalerate (APV, Sigma), and 6-cyano-7-nitro-quinoxalinediones (CNQX, Tocris);:ere dissolved in Ringers' solutions and applied by switching the valve connected to the cannula. A stock solution of CNQX (21 mM) was made in distilled water supplemented with 0.05 N NaOH: the pH of the perfusate containing the final concentration of CNQX (60 ~M) was 7.3-7.4.
Stimulation The 9th or 10th ventral roots were placed on Ag/AgCI wire electrodes for antidromic stimulation of motoneurons. A bipolar tungsten electrode insulated except for the tip was inserted into the lateral column (LC) at the third spinal segment to stimulate the descending fibers. Excitatory post-synaptic potentials were evoked by stimulation of the ipsilateral lateral column fibers (LC-EPSPs). Electrical stimuli (duration, 30-50 /zs) were applied at 0.2-1.0 Hz. A paired pulse was applied at an interval of 50-200 ms.
hztracellular recordings Lumbar motoneurons in the 9th and 10th spinal segments were identified by antidror~dc stimulation of the ventral root. The glass microelectrodes were filled with 4 M potassium acetate solution. The resistance of microelectrodes ranged from 10 to 20 M/]. In some experiments, intracellular and extracellular potentials were simultaneously recorded, and the extracellular potentials were subtracted from the intracellular ones. The distance between the tips of the two microelectrodes was 10-30 p,m. Similar results were obtained with or without the subtraction. The electrical signals were recorded via a preamplifier (MEZ-7101 and MEZ-8201, Nihon Kohden.), low-pass filtered at 3 kHz, and displayed on an oscilloscope and a chart recorder. The data were stored on a VHS video tape recorder (VC FI7, Sharp) via a digital audio processor (PCM-501ES, Shoshin EM) with a sampling rate of 44.1 kHz for later analysis. The resting membrane potential of cells used in this study ranged from - 8 4 to 50 inV. -
Voltage clamp Voltage clamp recordings were obtained from some cells impaled with two separate microelectrodes, as described elsewhere 16'37. The distance between the tips of the two microelectrodes was 10-30 p,m. We discarded cases in which the remaining EPSP during voltage clamp was beyond 3% of the EPSP before the clamp. We analyzed EPSCs only when the synaptic location was assessed to be generated at 0.2-0.8 space constant from the soma, using normalized shape indices: the shape indices of monosynaptic EPSPs were obtained from the 10-90% rise time and the half width of averaged EPSPs normalized by the membrane time constant. The membrane time constant was estimated from an optimal exponential fit for the later
RESULTS
Glutamate potentials and associated changes in EPSP amplitude L - G l u t a m a t e (1 m M ) d e p o l a r i z e d l u m b a r spinal mot o n e u r o n s ( g l u t a m a t e potential), both in the presence and absence of Mg 2+. The m a g n i t u d e of tne quasi steady-state g l u t a m a t e potential induced by perfusion of 1 m M g l u t a m a t e for 3 - 5 min was normalized to that at - 5 0 m V by assuming that the reversal potential was 0 m V a n d t h a t the c u r r e n t - v o l t a g e relationship was linear. T h e n o r m a l i z e d g l u t a m a t e potential was 13.2 + 5.1 m V (n = 11) in the Mg2+-free m e d i u m , 12.8 5:4.4 m V (n = 5) with 4.5 m M Ca 2+ plus 8 m M Mg 2+ (high C a 2 + - M g 2+) a n d 8.4 + 6.0 m V (n = 6) with 1.8 m M Ca 2+ plus 4 m M M g 2+ (high Mg2+). In some cells, perfusion o f 1 m M g l u t a m a t e for a longer period resulted in a slight reduction of the g l u t a m a t e potential following t h e s t e a d y state, but f u r t h e r analyses were c o n d u c t e d on E P S P s r e c o r d e d at the quasi steady state n e a r to the m a x i m u m g l u t a m a t e potential. T h e p e a k a m p l i t u d e of m o n o s y n a p t i c E P S P evoked by stimulation of the lateral c o l u m n fibers ( L C - E P S P ) is c o n s i d e r e d to be mainly m e d i a t e d through nonN M D A r e c e p t o r s 37, although a late slow c o m p o n e n t of E P S P sensitive to a N M D A antagonist, 2-amino-5p h o s p h o n o v a l e r a t e ( A P V ) , was occasionally observed in the MLZ+-free m e d i u m ( d a t a not shown). In most cells tested in the Mg2+-free m e d i u m , the E P S P ampli-
126 A
^EPSP2
I\
C llmv
I \
li~.u o l ~
3
o
~°°~) .:':.. °°
%
O
• EPSP1 o EPSP2
0o
-.
3
o
~
4
S
a
Time (min)
1.2to
:=-.- 1.0
i
2
ooo o:f
o
o
o.6
U.
I_____.
l
~mv 2ores
0"60
2
4 6 Time (min)
8
Fig. !. Inhibition of the paired pulse facilitation by glutamate. A: LC-EPSPs evoked by a single and a paired pulse stimulations were superimposed. B: EPSPs evoked by a paired pulse stimulation during the glutamate-induced depolarization. C: a time-course of glutamate-induced changes in amplitude of the first (e) and second (o) EPSPs. Each point represents amplitude of averaged EPSPs of four successive trials evoked at every 2.5 s. D: the facilitation ratio (EPSP2/EPSPI) before, during and after perfusion of I mM glutamate. (B, C and D were obtained from the same ceil.)
tude was reversibly reduced by glutamate, although a transient enhancement of the peak amplitude was occasionally observed during the early phase of the glutamate potential. A sustained enhancement of the EPSP amplitude during the quasi steady state of the glutamate potential was seen in one cell in the absence of Mg 2+ and more often in the high Mg 2+ solution. The glutamate-induced depolarization was accompanied by firing from EPSP in some cells and by a negative component of evoked potential in a few cells, but, in this study, we analyzed EPSPs without contamination of firings or negative potentials.
A
B
EPSP increase
A paired pulse stimulation often causes an increase or a decrease in the second EPSP amplitude. These phenomena are known as synaptic facilitation or depression, and have been considered to result from a variation in transmitter release from the presynaptic terminal as. When the strength and interval of stimulus was adjusted to evoke facilitation in LC-EPSPs, the facilitation ratio was calculated as the ratio of the amplitude of the second EPSP (EPSP2) to that of the first EPSP (EPSP1) (Fig. 1A). In Fig. 1B, EPSP2 was
EPSP decrease
P=0.009, n=7
2.0
Effects of glutamate on paired pulse facilitation of EPSPs
P=0.048, n=10
.o_ m 1.5
1.5
C
EPSP decrease P=0.029, n=6
1.5
J
C
.g '--- 1.0 o~
u
i
u_
1.0
1.0
0.5
control
glutamate
control
glutamate
control
glutamate +CNQX Fig. 2. Glutamate-induced reduction in the facilitation ratio. Each point represents the mean facilitation ratio calculated from 5-20 trials in each cells. A and B: the facilitation ratio for EPSPs with an increase (A) or a decrease (B) in peak amplitude on the action of glutamate. The data were obtained from 7 (A) and 10 (B) cells in the Mg2÷-free (o), the high Mg2+ (e) and the high Ca2÷-Mg 2+ (11) solutions. C: the facilitation ratio with co-application of glutamate and CNOX (60 ~M). Points represent values for 6 measurements using 5 cells. Glutamate, 1 mM.
127 greater than EPSP1 in the control (bottom), but EPSP2 was almost equal or was slightly smaller than EPSP1 at the depolarization induced by 1 mM glutamate (top). Fig. 1C shows the peak amplitude of the averaged EPSP1 (closed circles) and EPSP2 (open circles) of four successive records, before, during the peffusion and after the washout of glutamate. The amount of reduction by glutamate was greater with EPSP2, the result being a reversible reduction in the facilitation ratio (Fig. 1D). Fig. 2 shows the mean facilitation ratio in controls and in the presence of 1 mM glutamate for cells in which the EPSP amplitude was increased (Fig. 2A) or decreased (Fig. 2B) by glutamate in the MgE+-free (open circles), the high Mg 2+ (closed circles) and the high C a 2 + - M g 2+ (closed squares) solutions. The amount of change varied among cells and was negligible in some, but reduction of the ratio by glutamate was significant according to paired t-tests (P < 0.01 for Fig. 2A, P < 0.05 for Fig. 2B). Table I summarizes the results from data with both increased and decreased EPSPs. Similar reversible reduction in the paired pulse facilitation was seen with depolarization induced by kainate (50-100 ~M) in 3 of 6 cells, by quisqualate (5-50/zM) in 4 of 6 cells, and by NMDA (50/~M) in 4 of 7 cells (Table I). NMDA was applied in the absence of Mg 2+. L-2-Amino-4-phosphonobutyrate (L-AP4), a glutamate agonist, reduced the peak amplitude of the first EPSP to 58 + 15% (n = 9) of the controls without appreciable changes in the membrane potential. A decrease in the facilitation ratio was seen in 4 of 9 cells tested (Table I). The glutamate potential was decreased by concomitant application of 6-cyano-7-nitro-quinoxalinediones (CNQX), a non-NMDA receptor antagonist. When the
TABLE 1
Effects of excitatory amino acids on the paired pulse facilitation. Facilitation ratio
Glutamate in = 17) Kainate in -- 6) Quisqualate (n = 6) NMDA (n = 7) L-AP4 in = 9)
control
test
number of cells depressed
1.36+0.22 1.27+0.18 1.28 + 0.21 1.38+0.26 1.33+0.23
1.11 +0.24 1.03+0.22 1.01 + 0.20 1.15+0.26 1.19+0.13
10 3 4 4 4
EPSP amplitude was either increased or decreased by glutamate, kainate, quisqualate and NMDA, and was decreased by L-AP4. Data are mean + SD of the facilitation ratio in the absence icontrol) and presence itest) of the excitatory amino acids, n; number of cells tested. The number of cells with a significant reduction (P < 0.05) in the facilitation ratio by the excitatory amino acids is shown in the right column. Glutamate, 1 mM; kainate, 50-100 ~,M; quisqualate, 5-50/zM; NMDA, 50 p,M; L-AP4, 1 mM.
1.8
LI. 1.0 0.8
APV +
+
GLU
+
"
"
+
÷ +
Order of experiments
Fig. 3. Change in the facilitation ratio of a cell by co-application of 1 mM glutamate and 250 ~M APV in the Mg2+-free solution. The perfusate was repeatedly changed according to the number on the abscissa. Each point and bar represent the mean and S.D. of the facilitation ratio during the quasi steady state with each solution.
normalized glutamate potential was decreased by coapplication of 60 p,M CNQX to less than 5 mV (3.1 + 1.8 mV, n = 6), the peak EPSP amplitude at the quasi steady state ranged from 39 to 89% (71 + 18%, n = 6) of controls and the facilitation ratio was decreased (Fig. 2C, P <0.05). Application of 60 /.tM CNQX alone for 3-5 min, a period similar to that used for the co-application, decreased the peak EPSP amplitude to 62 + 25% (44-91%, n = 3) of control without any apparent reduction in the facilitation ratio. When the Mg2+-free perfusion medium with and without APV was repeatedly changed, the glutamate-induced reduction in the facilitation ratio was observed in the presence of 250 g M APV, although APV alone did not seem to change the ratio (Fig. 3).
Effects of glutamate on paired pulse facilitation of EPSCs The amplitude of EPSP was influenced both by the membrane potential and the input conductance of motoneurons ~6'37. We recorded excitatory post-synaptic currents (EPSCs) at a constant membrane potential under voltage clamp and examined the effects of glutamate on the paired pulse facilitation of EPSCs. As shown in Fig. 4A, there was a good agreement between the facilitation ratio in LC-EPSPs (trace a) and that in LC-EPSCs (trace d) recorded at the resting potential. Fig. 4B summarizes the effects of 1 mM glutamate on the facilitation ratio of EPSCs obtained from 5 measurements in 4 cells in the Mg2+-free (circles) or the high C a 2 + - M g 2+ (squares) solutions. The membrane potential was voltage clamped near to the resting potential measured just before application of glutamate. Glutamate increased the input conductance in two cells to 153 and 125% of control, and decreased conductance to 90 and 81% in two cells. Glutamate (1 mM) either decreased (open symbols) or increased
128
A
B
1.6
0
.~--(.1 .':U)
_o
d ~ ~ ~
'¢¢m
1•0
o.a control
glu
Fig. 4. Glutamate-induced inhibition in the paired pulse facilitation of EPSCs. A: LC-EPSPs (trace a) and the current monitor (trace b) before voltage clamp, and the voltage monitor (trace c) and LCEPSCs (trace d) during the voltage clamp. In each record, 5 successive sweeps were superimposed. The membrane potential was held at the resting potential during the voltage clamp. B: the paired pulse facilitation ratio of LC-EPSCs in controls and in the presence of 1 mM glutamate. The values were obtained from 5 measurements in 4 cells. Glutamate decreased (open symbols) or increased (closed symbols) peak amplitude of EPSCs in response to the first stimulus. Cells were perfused with the Mg2*-free (circles) or the high Ca 2+Mg 2 + (square) solutions.
(closed symbols) the amplitude of peak EPSCs, but the inhibitory effects of glutamate on the facilitation ratio were seen with both decreased and increased EPSCs. DISCUSSION We obtained evidence that glutamate inhibited the paired pulse facilitation of monosynaptic excitatory synaptic transmission between the spinal motoneurons and the descending lateral column fibers. The exogenously applied glutamate either decreased or increased the peak amplitude of the EPSP, but the glutamate-induced inhibition of the facilitation was seen with both decreased and increased EPSPs. The amplitude of EPSPs is influenced by factors of the driving force for ions involved in EPSCs, the input conductance of the post-synaptic cell, and the amplitude of EPSCs 37. The driving force decreases with glutamate-induced depolarization, since the current-voltage relation of LCEPSC in the Mg2+-free medium was linear and the reversal potential was close to 0 mV ~6. Glutamate-induced depolarization in spinal motoneurons was reported to be associated with an increase 4'6,17.37, a decrease 6 or no change 6 in the input conductance. In the present study, glutamate reduced the facilitation of monosynaptic EPSCs recorded at a constant membrane potential under voltage clamp and the reduction was seen with both increased and decreased conductances, thereby suggesting that the membrane potential or the input conductance of the post-synaptic membrane was not essential for the inhibition in the facilitation.
The reversible inhibition of the paired pulse facilitation of EPSP or EPSC suggests that glutamate modulates release of transmitters 3s, although a quantal analysis might be required to demonstrate the presynaptic action directly. In hippocampal cultured neurons, a reversible reduction of EPSCs by a low concentration of glutamate via presynaptic action was noted in about 3/5 of cells using a statistical analysis based on the variance method 8. Glutamate has been considered to act as a mixed agonist at both NMDA and non-NMDA receptors on spinal cord neurons 23. In the present study, both NMDA and non-NMDA agonists (NMDA, kainate and quisqualate) reduced the facilitation. Biochemical studies revealed the inhibitory effects of NMDA or non-NMDA antagonists on the glutamateinduced inhibition of the release of excitatory amino acids in the C N S 3'20'24, but the glutamate action on the facilitation presented herein was insensitive to APV and CNQX. Mechanisms of the glutamate-induced inhibition of the paired pulse facilitation remains to be defined. At the lobster neuromuscular synapse, glutamate hyperpolarizes the presynaptic membrane by activation of the K + channel via pertussis toxin-sensitive GTP binding protein 26. In the CNS, glutamate and L-AP4 were found to enhance the paired pulse facilitation associated with reduction in the first EPSP amplitude. The enhancement of the facilitation was suggested to be due to a decrease in the release of neurotransmitters in response to the first stimulus ~'5''~'~'~'s6,possibly via mechanisms other than activation of the terminal K + channel ~. Reduction in the facilitation ratio with a decrease in the first EPSP amplitude presented herein could not be explained solely by the "depletion model ''as. Glutamate and its agonists can elevate the concentration of extracellular K + either due to repetitive firings or K + efflux through the glutamate receptor channel 3~,a3. In the frog spinal cord, accumulation of extracellular K + depolarizes motoneurons associated with reduction in EPSP amplitude and an increase in spontaneous activity 34. In the present study, the glutamate-induced inhibition of the facilitation was observed with the small glutamate potential without repetitive firings in the presence of CNQX, and L-AP4 reduced the facilitation ratio without depolarization. Thus, an elevation of the extracellular K + level does not seem to be essential to generate the glutamate-induced inhibition of the facilitation. We did not, however, measure the extracellular K + level, and could not exclude slight and indirect effects of the K + level on the inhibition. Extracellular M g 2+ is noted to modify activities of frog spinal motoneurons: in the high M g 2+ or the high
129
Ca2+-Mg 2+ solutions, poly-synaptic inputs were eliminated and a stronger stimulus was required to evoke the same amplitude of LC-EPSPs as that in the absence of Mg 2+37, possibly due to antagonistic action of Mg 2+ to Ca 2+ on synaptic transmission. Glutamate-induced inward currents were reduced by Mg 2+ due to a block of NMDA receptors 17. In addition, an increase in the EPSP amplitude by glutamate was more often seen when the high Mg 2+ medium was used. An increase in frequency of spontaneous EPSPs by excitatory amino acids has been noted in the vestibular organ 3° in the presence of Mg 2+. Thus, the cellular responses to glutamate might differ in the presence of Mg 2+. However, the glutamate-induced inhibition of the facilitation was seen in both the presence and absence of Mg 2+, thereby indicating that the glutamate action on the transmitter release was insensitive to Mg 2+ Autoreceptors on the presynaptic membrane may be involved in the excitatory amino acid-mediated transmission in the CNS 35. In the present study, involvement of indirect actions of glutamate via activation of other cells in the transmitter release would need to be considered, since exogenous glutamate can activate receptors throughout the spinal cord. For an example, exogenously applied excitatory amino acids were seen to increase taurine in the extracellular space 2'18'25, and taurine reduced the release of the neurotransmitter at the excitatory synapse 37. The increase in taurine due to excitatory amino acids was, however, inhibited by NMDA or non-NMDA antagonists 25, thereby differing from the present observation. The presynaptic action via uptake of glutamate is another possibility Hum'29. Ubiquitous distribution of glutamate in the CNS suggests that the level of extracellular glutamate might be elevated at various regions in the CNS, under physiological or pathological conditions. Multiple actions of glutamate may function as negative or positive feedback systems and are likely to produce variability in modulation of neuronal activities in the CNS. Acknowledgements. We thank Ms. J. Kawawaki for technical assistance.
REFERENCES 1 Anson, J. "and Collins, G.G.S., Possible presynaptic actions of 2-amino-4-phosphonobutyrate in rat olfactory cortex, Br. J. Pharmacol., 91 (1987) 753-761. 2 Butcher, S.P., Lazarewicz, J.W. and Hamberger, A., In vivo microdialysis studies on the effects of decortication and excitotoxic lesions on kainic acid-induced calcium fluxes, and endogenous amino acid release in the rat striatum, J. Neurochem., 49 (1987) 1355-1360. 3 Collins, G.G.S., Anson, J. and Surtees, L., Presynaptic kainate and N-methyl-a-aspartate receptors regulate excitatory amino acid release in the olfactory cortex, Brain Res., 265 (1983) 157159.
4 Corradetti, R., King, A.E., Nistri,. A., Rovira, C. and Sivilotti, L., Pharmacological characterization of D-aminophosphonovaleric acid antagonism of amino acid and synaptically evoked excitations on frog motoneurones in vitro: an intracelluiar study, Br. J. Pharmacol., 86 (1985) 19-25. 5 Cotman, C.W., Flatman, J.A., Ganong, A.H. and Perkins, M.N., Effects of excitatory amino acid antagonists on evoked and spontaneous excitatory potentials in guinea-pig hippocampus, J. Phys. iol. 378 (1986) 403-415. 6 Engberg, I., Flatman, J~A. and Lambert, J.D.C., The action of excitatory amino acids on motoneurones in the feline spinal cord, Z Physiol., 288 (1979) 227-261. 7 Ferkany, J.W., Zaczek, R. and Coyle, J.T., Kainic acid stimulates excitatory amino acid neurotransmitter release at presynaptic receptors, Nature, 298 (1982) 757-759. 8 Forsythe, I.D. and Clements, J.D., Presynaptic glutamate receptors depress excitatory monosynaptic transmission between mouse hippocampal neurones, J. Physiol., 429 (1990) 1-16. 9 Harris, E.W. and Cotman, C.W., Effects of acidic amino acid antagonists on paired-pulse potentiation at the lateral perforant path, Exp. Brain Res., 52 (1983) 455-460. 10 Harris, K.M. and Miller, R.J., Excitatory amino acid-evoked release of [3H]GABA from hippocampai neurons in primary culture, Brah~ Res., 482 (1989) 23-33. 11 Hoehn, K. and White, T.D., Glutamate-evoked release of endogenous adenosine from rat cortical synaptosomes is mediated by glutamate uptake and not by receptors, J. Neurochem., 54 (1990) 1716-1724. 12 Jack, J.J.B., Miller, S., Porter, R. and Redman, S.J., The timecourse of minimal excitatory post-synaptie potentials evoked in spinal motoneurones by group la afferent fibres, Z Physiol., 215 (1971) 353-380. 13 Jacobson, I. and Hamberger, A., Kainic acid-induced changes of extracellular amino acid levels, evoked potentials and EEG activity in the rabbit olfactory bulb, Brab~ Res., 348 (1985) 289-296. 14 Jahr, C.E. and Yoshioka, K., la afferent excitation of motoneurunes in the in vitro new-born rat spinal cord is selectively antagonized by kynurenate, J. PhysioL, 370 (1986) 515-530. 15 Kuno, M. and Matsuura, S., Sites and mechanisms of action of lidocaine upon the isolated spinal cord of the frog, Brain Res., 249 (1982) 87-93. 16 Kuno, M., Yasunami, T. and Matsuura, S., Excitatory post-synaptic currents in response to different synaptic inputs of frog spinal motoneurons, Neu~'osci. Res., 5 (1988)240-252. 17 Lacey, G. and Nistri, A., Changes in glutamate-evoked currents of frog motoneurones by extracellular Mg a+, Neurosci. Lett., 90 (1988) 213-218. 18 Lehmann, A., Isacsson, H. and Hamberger, A., Effects of in vivo administration of kainic acid on the extracellular amino acid pool in the rabbit hippocampus, J. Neurochem., 40 (1983) 1314-1320. 19 Liu, G., Feldman, J.L. and Smith, J.C., Excitatory amino acidmediated transmission of inspiratory drive to phrenic motoneuruns, J. Neurophysiol., 64 (1990) 423-436. 20 L6pez-Colom6, A.M. and Roberts, P.J., Effect of excitatory amino acid analogues on the release of o-[3H]aspartate from chick retina, Eur. J. Pharmacol., 142 (1987) 409-417. 21 Martin, D., Bustos, G.A., Bowe, M.A., Bray, S.D. and Nadler, J.V., Autoreceptor regulation of glutamate and aspartate release from slices of the hippocampal CAI area, J. Neurochem., 56 (1991) 1647-1655. 22 Matsuura, S., Kawaguchi, S., Ichiki, M., Sorimachi, M., Kataoka, K. and Inouye, A., Perfusion of frog's spinal cord as convenient method for neuropharmacological studies, Eur. J. Pharmacol., ~ (1969) 13-16. 23 Mayer, M.L. and Westbrook, G.L., Mixed agonist action of excitatory amino acids on mouse spinal cord neurones under voltage clamp, J. Physiol., 354 (1984)29-53. 24 McBean, G.J. and Roberts, P.J., Glutamate-preferring receptors regulate the release of D-[3H]aspartate from rat hippocampai slices, Nature, 291 (1981) 593-594. 25 Men6ndez, N., Herreras, O., Solis, J.M., Herranz, A.S. and del Rio, R.M., Extracellular taurine increase in rat hippocampus
130
26 27 28 29
30 31
evoked by specific glutamate receptor activation is related to the excitatory potency of glutamate agonists, Neurosci. Lett., 102 (1989) 64-69. Miwa, A., Ui, M. and Kawai, N., G protein is coupled to p~synaptic glutamate and GABA receptors in lobster neuromuscular synapse, J. Neurophysiol., 63 (1990) 173-180. Nakamura, F., Kuno, M. and Matsuura, S., Effects of giutamate on monosynaptic excitatory transmission at spinal motoneurons, Jpn. J. Physiol., 41 (1991) S166. Nicoll, R.A., Malenka, R.C. and Kauer, J.A., Functional comparison of neurotransmitter receptor subtypes in mammalian central nervous system, Physiol. Rer., 70 (1990) 513-565. Pin, J.-P. and Bockaert, J., Two distinct mechan!sms, differentially affected by excitatory amino acids, trigger GAB~ release from fetal mouse striatal neurons in primary culture, J. Neurosci., 9 (1989) 648-656. Prigioni, I., Russo, G., Valli, P. and Masetto, S., Pre- and postsynaptic excitatory action of glutamate agonists on frog vestibular receptors, Hearing Res., 46 (1990) 253-260. Pumain, R., Kurcewicz, I. and Louvel, J., Ionic changes induced by excitatory amino acids in the rat cerebral cortex, Can. J. Physiol. PharmacoL, 65 (1987) 1067-1077.
32 Rail, W., Burke, R.E., Smith, T.G., Nelson, P.G. and Frank, K., Dendritic location of synapses and possible mechanisms for the monosynaptic EPSP in motoneurons, J. Neurophysiol. 30 (1967) 1169-1193. 33 Rice, M.E. and Nicholson, C., Glutamate- and aspartate-induced extracellular potassium and calcium shifts and their relation to those of kainate, quisqualate and N-methyl-D-aspartate in the isolated turtle cerebellum, Neuroscience, 38 (1990) 295-310. 34 Sykov;i, E. and Orkand, R.K., Extracellular potassium accumulation and transmission in frog spinal cord, Neuroscience, 5 (1980) 1421-1428. 35 Takeuchi, A., The transmitter role of glutamate in nervous systems, Jpn. J. Physiol., 37 (1987) 559-572. 36 Trombley, P.Q. and Westbrook, G.L., Excitatory synaptic transmission in cultures of rat olfactory bulb, J. Neurophysiol., 64 (1990) 598-606. 37 Yasunami, T., Kuno, M. and Matsuura, S., Voltage-clamp analysis of taurine-induced suppression of excitatory post-synaptic potentials in frog spinal motoneurons, J. Neurophysiol., 60 (1988) 1405-1418. 38 Zucker, R.S., Short-term synaptic plasticity, Annu. Rev. Neurosci., 12 (1989) 13-31.
124
BRES 18295
Glutamate-induced inhibition of paired pulse facilitation of monosynaptic excitatory post-synaptic potentials in frog spinal motoneurons F u s a o N a k a m u r a , Miyuki K u n o a n d S h i u s h i M a t s u u r a Department of Physiology., Osaka City Unit'ersity Medical School, Abeno-ku, Osaka (Japan) (Accepted 30 June 1992)
Key words: Excitatory amino acid; Autoregulation; Short-term potentiation; Modulation of transmitter release
To evaluate actions of glutamate on excitatory synaptic transmission in the central nervous system, we examined glutamate-induced changes in the paired pulse facilitation of monosynaptic excitatory post-synaptic potentials evoked by stimulation of the lateral column fibers (LC-EPSPs) on lumbar motoneurons in the frog spinal cord. Glutamate (1 raM) depolarized motoneurons both in the presence and absence of Mg 2+. In most cells perfused with Mg2+-free or high Ca2+-Mg 2+ solutions, the glutamate potential was accompanied by a reduction in peak amplitude of EPSPs, although the degree of change varied with the cells. Glutamate enhanced the EPSP amplitude in a few cells with Mg2+-free and high Ca2+-Mg 2+ solutions, and in most cells with high Mg 2+ medium. In 3/5 cells tested, the paired pulse facilitation of EPSPs was reduced by glutamate when the EPSP amplitude either increased or decreased. NMDA (50 /~M), kainate (50-100 /zM), quisqualate (5-50 /~M) and L-2-amino-4-phosphonobutyrate (L-AP4, 1 raM) also decreased the facilitation in about hull of the cells tested. The glutamate-induced decrease in the facilitation was observed in both the presence and absence of Mg 2+ and was not affected by the concomitant application of glutamate and antagonists for non-NMDA or NMDA receptors, such as 6-cyano-7-nitro-quinoxalinediones(CNQX, 60 ~M) or 2-amino-5-phosphonovalerate (APV, 250/~M). Glutamate reduced the facilitation of excitatory post-synaptic currents (EPSCs) recorded at a constant membrane potential under voltage clamp, when the EPSC amplitude either increased or decreased and when the input conductance either increased or decreased. Thus changes in membrane potential or input conductance of the post-synaptic membrane were not essential to the glutamate-induced inhibition of the facilitation. These results suggest that glutamate modulates release of excitatory transmitters via mechanisms insensitive to Mg 2+, CNQX and APV.
INTRODUCTION Excitatory amino acids are distributed throughout the central nervous system (CNS) and are putative excitatory neurotransmitters 28,s5. The exogenous application of L-glutamate mimics the situation when the level of extracellular glutamate is elevated throughout the CNS, and allows the study of the modulation of neuronal activity by the action of glutamate. Biochemical studies have shown that excitatory amino acids either inhibit 3,2°,21.24 or enhance 3,7,13,1~ the release of excitatory amino acids in the CNS. A paired pulse • facilitation of EPSPs has been considered to result from a variation in transmitter release 38 and may be valid to monitor actions of glutamate on the transmitter release. However, there are relatively few reports
on the action of glutamate on the facilitation at the excitatory amino acid-mediated synaptic transmission in the CNS. In the frog spinal cord, the excitatory synapses between the descending lateral column fiber and motoneurons may utilize excitatory amino acids as neurotransmitters 3~. Intraarterial perfusion z2 enables rapid application of glutamate throughout the spinal cord. We analyzed glutamate-induced changes in the paired pulse facilitation of monosynaptic EPSPs in Mge+-free and Mg2+-containing solutions. Evidence was obtained that glutamate reversibly reduced the paired pulse facilitation of EPSPs via mechanisms insensitive to Mg 2+ and antagonists for both NMDA and non-NMDA receptors (APV and CNQX), and that changes in membrane potential or input conductance of the post-synaptic membrane were not essential to
Correspondencc: M. Kuno, Department of Physiology, Osaka City University Medical School, Abeno-ku, Osaka 545, Japan. Fax: (81) (06) 632-7114.
125 the r e d u c t i o n of the facilitation. T h u s it seems t h a t g l u t a m a t e m o d u l a t e s the t r a n s m i t t e r release at the excitatory a m i n o a c i d - m e d i a t e d excitatory synapses. A p r e l i m i n a r y account has b e e n m a d e 27.
decay phase of EPSPs s2. The location of synapses generating monosynaptic EPSPs was roughly estimated from the shape indices 32, using parameters describing a nerve cell model following the study by Jack et al. on la-fiber EPSPs 12'16. The input conductance was determined from the steady-state current responses to 100-500 ms, 4-10 mV hyperpolarizing command voltage steps. Input resistance of the cells ranged from 1.3 to 10 M~.
M A T E R I A L S AND M E T H O D S
Data analysis Adult bullfrogs (Rana catesbiana) weighing 150-350 g were anesthetized with pentobarbital sodium (Nembutal; Abbott) (30 mg/kg of body weight) and the entire spinal cord with attached spinal roots was isolated. The isolated spinal cord was placed in the recording chamber and perfused with ox'ygenated Ringers' solution through a fine polyethylene cannula inserted into the ventral spinal artery. Details of this procedure have been described elsewhere 15'22. All experiments were carried out at room temperature (15-24°O.
Sohaions The Mg2+-free Ringers' solution contained (in mM): 111 NaCI, 2.0 KCI, 1.8 CaCI 2, 1.5 NaHCO 3 (pH 7.3-7.4) and was supplemented with 0.1% glucose. In later experiments, the pH was adjusted using 10 mM N-[2-hydroxyethyl] piperazine-N'-[2-ethanesulfonic acid] (HEPES)-NaOH. The high Mg 2+ medium was made by adding 4 mM MgCI2 to the above medium. In some experiments, the high Ca2+-Mg 2+ (4.5 mM Ca2+ plus 8 mM Mg2+ ) medium was used to eliminate poly-synaptic inputs 14'37. The osmolarity of the high Ca 2+Mg 2+ and the high Mg2+ solutions was compensated for by changes in the concentration of NaCI.
Electrical signals were digitized at 10-20 kHz using an analoguedigital converter (MacLab/4, Analogue Digital Instrument) and were stored on a personal computer (Macintosh SE/30, Apple). Averages of 4-20 records and measurements of peak amplitude of EPSP or EPSC were conducted using a computer program (Scope ver. 3.1, Analogue Digital Instrument) or an averaging computer (ATAC 350, Nihon Kohden). When a notch was observed on the rising phase of the EPSPs, the amplitude of the monosynaptic EPSP was measured at this notch. The peak amplitude of the EPSPs was within 5 mV in most cells and was not corrected for non-linear summation.
Chemicals L-Glutamate (Sigma), kainate (Sigma), quisqualate (Sigma), Nmethyl-D-aspartate (NMDA, Sigma), L-2-amino-4-phoshphonobutyrate (L-AP4, Sigma), 2-amino-5-phosphonovalerate (APV, Sigma), and 6-cyano-7-nitro-quinoxalinediones (CNQX, Tocris);:ere dissolved in Ringers' solutions and applied by switching the valve connected to the cannula. A stock solution of CNQX (21 mM) was made in distilled water supplemented with 0.05 N NaOH: the pH of the perfusate containing the final concentration of CNQX (60 ~M) was 7.3-7.4.
Stimulation The 9th or 10th ventral roots were placed on Ag/AgCI wire electrodes for antidromic stimulation of motoneurons. A bipolar tungsten electrode insulated except for the tip was inserted into the lateral column (LC) at the third spinal segment to stimulate the descending fibers. Excitatory post-synaptic potentials were evoked by stimulation of the ipsilateral lateral column fibers (LC-EPSPs). Electrical stimuli (duration, 30-50 /zs) were applied at 0.2-1.0 Hz. A paired pulse was applied at an interval of 50-200 ms.
hztracellular recordings Lumbar motoneurons in the 9th and 10th spinal segments were identified by antidror~dc stimulation of the ventral root. The glass microelectrodes were filled with 4 M potassium acetate solution. The resistance of microelectrodes ranged from 10 to 20 M/]. In some experiments, intracellular and extracellular potentials were simultaneously recorded, and the extracellular potentials were subtracted from the intracellular ones. The distance between the tips of the two microelectrodes was 10-30 p,m. Similar results were obtained with or without the subtraction. The electrical signals were recorded via a preamplifier (MEZ-7101 and MEZ-8201, Nihon Kohden.), low-pass filtered at 3 kHz, and displayed on an oscilloscope and a chart recorder. The data were stored on a VHS video tape recorder (VC FI7, Sharp) via a digital audio processor (PCM-501ES, Shoshin EM) with a sampling rate of 44.1 kHz for later analysis. The resting membrane potential of cells used in this study ranged from - 8 4 to 50 inV. -
Voltage clamp Voltage clamp recordings were obtained from some cells impaled with two separate microelectrodes, as described elsewhere 16'37. The distance between the tips of the two microelectrodes was 10-30 p,m. We discarded cases in which the remaining EPSP during voltage clamp was beyond 3% of the EPSP before the clamp. We analyzed EPSCs only when the synaptic location was assessed to be generated at 0.2-0.8 space constant from the soma, using normalized shape indices: the shape indices of monosynaptic EPSPs were obtained from the 10-90% rise time and the half width of averaged EPSPs normalized by the membrane time constant. The membrane time constant was estimated from an optimal exponential fit for the later
RESULTS
Glutamate potentials and associated changes in EPSP amplitude L - G l u t a m a t e (1 m M ) d e p o l a r i z e d l u m b a r spinal mot o n e u r o n s ( g l u t a m a t e potential), both in the presence and absence of Mg 2+. The m a g n i t u d e of tne quasi steady-state g l u t a m a t e potential induced by perfusion of 1 m M g l u t a m a t e for 3 - 5 min was normalized to that at - 5 0 m V by assuming that the reversal potential was 0 m V a n d t h a t the c u r r e n t - v o l t a g e relationship was linear. T h e n o r m a l i z e d g l u t a m a t e potential was 13.2 + 5.1 m V (n = 11) in the Mg2+-free m e d i u m , 12.8 5:4.4 m V (n = 5) with 4.5 m M Ca 2+ plus 8 m M Mg 2+ (high C a 2 + - M g 2+) a n d 8.4 + 6.0 m V (n = 6) with 1.8 m M Ca 2+ plus 4 m M M g 2+ (high Mg2+). In some cells, perfusion o f 1 m M g l u t a m a t e for a longer period resulted in a slight reduction of the g l u t a m a t e potential following t h e s t e a d y state, but f u r t h e r analyses were c o n d u c t e d on E P S P s r e c o r d e d at the quasi steady state n e a r to the m a x i m u m g l u t a m a t e potential. T h e p e a k a m p l i t u d e of m o n o s y n a p t i c E P S P evoked by stimulation of the lateral c o l u m n fibers ( L C - E P S P ) is c o n s i d e r e d to be mainly m e d i a t e d through nonN M D A r e c e p t o r s 37, although a late slow c o m p o n e n t of E P S P sensitive to a N M D A antagonist, 2-amino-5p h o s p h o n o v a l e r a t e ( A P V ) , was occasionally observed in the MLZ+-free m e d i u m ( d a t a not shown). In most cells tested in the Mg2+-free m e d i u m , the E P S P ampli-
126 A
^EPSP2
I\
C llmv
I \
li~.u o l ~
3
o
~°°~) .:':.. °°
%
O
• EPSP1 o EPSP2
0o
-.
3
o
~
4
S
a
Time (min)
1.2to
:=-.- 1.0
i
2
ooo o:f
o
o
o.6
U.
I_____.
l
~mv 2ores
0"60
2
4 6 Time (min)
8
Fig. !. Inhibition of the paired pulse facilitation by glutamate. A: LC-EPSPs evoked by a single and a paired pulse stimulations were superimposed. B: EPSPs evoked by a paired pulse stimulation during the glutamate-induced depolarization. C: a time-course of glutamate-induced changes in amplitude of the first (e) and second (o) EPSPs. Each point represents amplitude of averaged EPSPs of four successive trials evoked at every 2.5 s. D: the facilitation ratio (EPSP2/EPSPI) before, during and after perfusion of I mM glutamate. (B, C and D were obtained from the same ceil.)
tude was reversibly reduced by glutamate, although a transient enhancement of the peak amplitude was occasionally observed during the early phase of the glutamate potential. A sustained enhancement of the EPSP amplitude during the quasi steady state of the glutamate potential was seen in one cell in the absence of Mg 2+ and more often in the high Mg 2+ solution. The glutamate-induced depolarization was accompanied by firing from EPSP in some cells and by a negative component of evoked potential in a few cells, but, in this study, we analyzed EPSPs without contamination of firings or negative potentials.
A
B
EPSP increase
A paired pulse stimulation often causes an increase or a decrease in the second EPSP amplitude. These phenomena are known as synaptic facilitation or depression, and have been considered to result from a variation in transmitter release from the presynaptic terminal as. When the strength and interval of stimulus was adjusted to evoke facilitation in LC-EPSPs, the facilitation ratio was calculated as the ratio of the amplitude of the second EPSP (EPSP2) to that of the first EPSP (EPSP1) (Fig. 1A). In Fig. 1B, EPSP2 was
EPSP decrease
P=0.009, n=7
2.0
Effects of glutamate on paired pulse facilitation of EPSPs
P=0.048, n=10
.o_ m 1.5
1.5
C
EPSP decrease P=0.029, n=6
1.5
J
C
.g '--- 1.0 o~
u
i
u_
1.0
1.0
0.5
control
glutamate
control
glutamate
control
glutamate +CNQX Fig. 2. Glutamate-induced reduction in the facilitation ratio. Each point represents the mean facilitation ratio calculated from 5-20 trials in each cells. A and B: the facilitation ratio for EPSPs with an increase (A) or a decrease (B) in peak amplitude on the action of glutamate. The data were obtained from 7 (A) and 10 (B) cells in the Mg2÷-free (o), the high Mg2+ (e) and the high Ca2÷-Mg 2+ (11) solutions. C: the facilitation ratio with co-application of glutamate and CNOX (60 ~M). Points represent values for 6 measurements using 5 cells. Glutamate, 1 mM.
127 greater than EPSP1 in the control (bottom), but EPSP2 was almost equal or was slightly smaller than EPSP1 at the depolarization induced by 1 mM glutamate (top). Fig. 1C shows the peak amplitude of the averaged EPSP1 (closed circles) and EPSP2 (open circles) of four successive records, before, during the peffusion and after the washout of glutamate. The amount of reduction by glutamate was greater with EPSP2, the result being a reversible reduction in the facilitation ratio (Fig. 1D). Fig. 2 shows the mean facilitation ratio in controls and in the presence of 1 mM glutamate for cells in which the EPSP amplitude was increased (Fig. 2A) or decreased (Fig. 2B) by glutamate in the MgE+-free (open circles), the high Mg 2+ (closed circles) and the high C a 2 + - M g 2+ (closed squares) solutions. The amount of change varied among cells and was negligible in some, but reduction of the ratio by glutamate was significant according to paired t-tests (P < 0.01 for Fig. 2A, P < 0.05 for Fig. 2B). Table I summarizes the results from data with both increased and decreased EPSPs. Similar reversible reduction in the paired pulse facilitation was seen with depolarization induced by kainate (50-100 ~M) in 3 of 6 cells, by quisqualate (5-50/zM) in 4 of 6 cells, and by NMDA (50/~M) in 4 of 7 cells (Table I). NMDA was applied in the absence of Mg 2+. L-2-Amino-4-phosphonobutyrate (L-AP4), a glutamate agonist, reduced the peak amplitude of the first EPSP to 58 + 15% (n = 9) of the controls without appreciable changes in the membrane potential. A decrease in the facilitation ratio was seen in 4 of 9 cells tested (Table I). The glutamate potential was decreased by concomitant application of 6-cyano-7-nitro-quinoxalinediones (CNQX), a non-NMDA receptor antagonist. When the
TABLE 1
Effects of excitatory amino acids on the paired pulse facilitation. Facilitation ratio
Glutamate in = 17) Kainate in -- 6) Quisqualate (n = 6) NMDA (n = 7) L-AP4 in = 9)
control
test
number of cells depressed
1.36+0.22 1.27+0.18 1.28 + 0.21 1.38+0.26 1.33+0.23
1.11 +0.24 1.03+0.22 1.01 + 0.20 1.15+0.26 1.19+0.13
10 3 4 4 4
EPSP amplitude was either increased or decreased by glutamate, kainate, quisqualate and NMDA, and was decreased by L-AP4. Data are mean + SD of the facilitation ratio in the absence icontrol) and presence itest) of the excitatory amino acids, n; number of cells tested. The number of cells with a significant reduction (P < 0.05) in the facilitation ratio by the excitatory amino acids is shown in the right column. Glutamate, 1 mM; kainate, 50-100 ~,M; quisqualate, 5-50/zM; NMDA, 50 p,M; L-AP4, 1 mM.
1.8
LI. 1.0 0.8
APV +
+
GLU
+
"
"
+
÷ +
Order of experiments
Fig. 3. Change in the facilitation ratio of a cell by co-application of 1 mM glutamate and 250 ~M APV in the Mg2+-free solution. The perfusate was repeatedly changed according to the number on the abscissa. Each point and bar represent the mean and S.D. of the facilitation ratio during the quasi steady state with each solution.
normalized glutamate potential was decreased by coapplication of 60 p,M CNQX to less than 5 mV (3.1 + 1.8 mV, n = 6), the peak EPSP amplitude at the quasi steady state ranged from 39 to 89% (71 + 18%, n = 6) of controls and the facilitation ratio was decreased (Fig. 2C, P <0.05). Application of 60 /.tM CNQX alone for 3-5 min, a period similar to that used for the co-application, decreased the peak EPSP amplitude to 62 + 25% (44-91%, n = 3) of control without any apparent reduction in the facilitation ratio. When the Mg2+-free perfusion medium with and without APV was repeatedly changed, the glutamate-induced reduction in the facilitation ratio was observed in the presence of 250 g M APV, although APV alone did not seem to change the ratio (Fig. 3).
Effects of glutamate on paired pulse facilitation of EPSCs The amplitude of EPSP was influenced both by the membrane potential and the input conductance of motoneurons ~6'37. We recorded excitatory post-synaptic currents (EPSCs) at a constant membrane potential under voltage clamp and examined the effects of glutamate on the paired pulse facilitation of EPSCs. As shown in Fig. 4A, there was a good agreement between the facilitation ratio in LC-EPSPs (trace a) and that in LC-EPSCs (trace d) recorded at the resting potential. Fig. 4B summarizes the effects of 1 mM glutamate on the facilitation ratio of EPSCs obtained from 5 measurements in 4 cells in the Mg2+-free (circles) or the high C a 2 + - M g 2+ (squares) solutions. The membrane potential was voltage clamped near to the resting potential measured just before application of glutamate. Glutamate increased the input conductance in two cells to 153 and 125% of control, and decreased conductance to 90 and 81% in two cells. Glutamate (1 mM) either decreased (open symbols) or increased
128
A
B
1.6
0
.~--(.1 .':U)
_o
d ~ ~ ~
'¢¢m
1•0
o.a control
glu
Fig. 4. Glutamate-induced inhibition in the paired pulse facilitation of EPSCs. A: LC-EPSPs (trace a) and the current monitor (trace b) before voltage clamp, and the voltage monitor (trace c) and LCEPSCs (trace d) during the voltage clamp. In each record, 5 successive sweeps were superimposed. The membrane potential was held at the resting potential during the voltage clamp. B: the paired pulse facilitation ratio of LC-EPSCs in controls and in the presence of 1 mM glutamate. The values were obtained from 5 measurements in 4 cells. Glutamate decreased (open symbols) or increased (closed symbols) peak amplitude of EPSCs in response to the first stimulus. Cells were perfused with the Mg2*-free (circles) or the high Ca 2+Mg 2 + (square) solutions.
(closed symbols) the amplitude of peak EPSCs, but the inhibitory effects of glutamate on the facilitation ratio were seen with both decreased and increased EPSCs. DISCUSSION We obtained evidence that glutamate inhibited the paired pulse facilitation of monosynaptic excitatory synaptic transmission between the spinal motoneurons and the descending lateral column fibers. The exogenously applied glutamate either decreased or increased the peak amplitude of the EPSP, but the glutamate-induced inhibition of the facilitation was seen with both decreased and increased EPSPs. The amplitude of EPSPs is influenced by factors of the driving force for ions involved in EPSCs, the input conductance of the post-synaptic cell, and the amplitude of EPSCs 37. The driving force decreases with glutamate-induced depolarization, since the current-voltage relation of LCEPSC in the Mg2+-free medium was linear and the reversal potential was close to 0 mV ~6. Glutamate-induced depolarization in spinal motoneurons was reported to be associated with an increase 4'6,17.37, a decrease 6 or no change 6 in the input conductance. In the present study, glutamate reduced the facilitation of monosynaptic EPSCs recorded at a constant membrane potential under voltage clamp and the reduction was seen with both increased and decreased conductances, thereby suggesting that the membrane potential or the input conductance of the post-synaptic membrane was not essential for the inhibition in the facilitation.
The reversible inhibition of the paired pulse facilitation of EPSP or EPSC suggests that glutamate modulates release of transmitters 3s, although a quantal analysis might be required to demonstrate the presynaptic action directly. In hippocampal cultured neurons, a reversible reduction of EPSCs by a low concentration of glutamate via presynaptic action was noted in about 3/5 of cells using a statistical analysis based on the variance method 8. Glutamate has been considered to act as a mixed agonist at both NMDA and non-NMDA receptors on spinal cord neurons 23. In the present study, both NMDA and non-NMDA agonists (NMDA, kainate and quisqualate) reduced the facilitation. Biochemical studies revealed the inhibitory effects of NMDA or non-NMDA antagonists on the glutamateinduced inhibition of the release of excitatory amino acids in the C N S 3'20'24, but the glutamate action on the facilitation presented herein was insensitive to APV and CNQX. Mechanisms of the glutamate-induced inhibition of the paired pulse facilitation remains to be defined. At the lobster neuromuscular synapse, glutamate hyperpolarizes the presynaptic membrane by activation of the K + channel via pertussis toxin-sensitive GTP binding protein 26. In the CNS, glutamate and L-AP4 were found to enhance the paired pulse facilitation associated with reduction in the first EPSP amplitude. The enhancement of the facilitation was suggested to be due to a decrease in the release of neurotransmitters in response to the first stimulus ~'5''~'~'~'s6,possibly via mechanisms other than activation of the terminal K + channel ~. Reduction in the facilitation ratio with a decrease in the first EPSP amplitude presented herein could not be explained solely by the "depletion model ''as. Glutamate and its agonists can elevate the concentration of extracellular K + either due to repetitive firings or K + efflux through the glutamate receptor channel 3~,a3. In the frog spinal cord, accumulation of extracellular K + depolarizes motoneurons associated with reduction in EPSP amplitude and an increase in spontaneous activity 34. In the present study, the glutamate-induced inhibition of the facilitation was observed with the small glutamate potential without repetitive firings in the presence of CNQX, and L-AP4 reduced the facilitation ratio without depolarization. Thus, an elevation of the extracellular K + level does not seem to be essential to generate the glutamate-induced inhibition of the facilitation. We did not, however, measure the extracellular K + level, and could not exclude slight and indirect effects of the K + level on the inhibition. Extracellular M g 2+ is noted to modify activities of frog spinal motoneurons: in the high M g 2+ or the high
129
Ca2+-Mg 2+ solutions, poly-synaptic inputs were eliminated and a stronger stimulus was required to evoke the same amplitude of LC-EPSPs as that in the absence of Mg 2+37, possibly due to antagonistic action of Mg 2+ to Ca 2+ on synaptic transmission. Glutamate-induced inward currents were reduced by Mg 2+ due to a block of NMDA receptors 17. In addition, an increase in the EPSP amplitude by glutamate was more often seen when the high Mg 2+ medium was used. An increase in frequency of spontaneous EPSPs by excitatory amino acids has been noted in the vestibular organ 3° in the presence of Mg 2+. Thus, the cellular responses to glutamate might differ in the presence of Mg 2+. However, the glutamate-induced inhibition of the facilitation was seen in both the presence and absence of Mg 2+, thereby indicating that the glutamate action on the transmitter release was insensitive to Mg 2+ Autoreceptors on the presynaptic membrane may be involved in the excitatory amino acid-mediated transmission in the CNS 35. In the present study, involvement of indirect actions of glutamate via activation of other cells in the transmitter release would need to be considered, since exogenous glutamate can activate receptors throughout the spinal cord. For an example, exogenously applied excitatory amino acids were seen to increase taurine in the extracellular space 2'18'25, and taurine reduced the release of the neurotransmitter at the excitatory synapse 37. The increase in taurine due to excitatory amino acids was, however, inhibited by NMDA or non-NMDA antagonists 25, thereby differing from the present observation. The presynaptic action via uptake of glutamate is another possibility Hum'29. Ubiquitous distribution of glutamate in the CNS suggests that the level of extracellular glutamate might be elevated at various regions in the CNS, under physiological or pathological conditions. Multiple actions of glutamate may function as negative or positive feedback systems and are likely to produce variability in modulation of neuronal activities in the CNS. Acknowledgements. We thank Ms. J. Kawawaki for technical assistance.
REFERENCES 1 Anson, J. "and Collins, G.G.S., Possible presynaptic actions of 2-amino-4-phosphonobutyrate in rat olfactory cortex, Br. J. Pharmacol., 91 (1987) 753-761. 2 Butcher, S.P., Lazarewicz, J.W. and Hamberger, A., In vivo microdialysis studies on the effects of decortication and excitotoxic lesions on kainic acid-induced calcium fluxes, and endogenous amino acid release in the rat striatum, J. Neurochem., 49 (1987) 1355-1360. 3 Collins, G.G.S., Anson, J. and Surtees, L., Presynaptic kainate and N-methyl-a-aspartate receptors regulate excitatory amino acid release in the olfactory cortex, Brain Res., 265 (1983) 157159.
4 Corradetti, R., King, A.E., Nistri,. A., Rovira, C. and Sivilotti, L., Pharmacological characterization of D-aminophosphonovaleric acid antagonism of amino acid and synaptically evoked excitations on frog motoneurones in vitro: an intracelluiar study, Br. J. Pharmacol., 86 (1985) 19-25. 5 Cotman, C.W., Flatman, J.A., Ganong, A.H. and Perkins, M.N., Effects of excitatory amino acid antagonists on evoked and spontaneous excitatory potentials in guinea-pig hippocampus, J. Phys. iol. 378 (1986) 403-415. 6 Engberg, I., Flatman, J~A. and Lambert, J.D.C., The action of excitatory amino acids on motoneurones in the feline spinal cord, Z Physiol., 288 (1979) 227-261. 7 Ferkany, J.W., Zaczek, R. and Coyle, J.T., Kainic acid stimulates excitatory amino acid neurotransmitter release at presynaptic receptors, Nature, 298 (1982) 757-759. 8 Forsythe, I.D. and Clements, J.D., Presynaptic glutamate receptors depress excitatory monosynaptic transmission between mouse hippocampal neurones, J. Physiol., 429 (1990) 1-16. 9 Harris, E.W. and Cotman, C.W., Effects of acidic amino acid antagonists on paired-pulse potentiation at the lateral perforant path, Exp. Brain Res., 52 (1983) 455-460. 10 Harris, K.M. and Miller, R.J., Excitatory amino acid-evoked release of [3H]GABA from hippocampai neurons in primary culture, Brah~ Res., 482 (1989) 23-33. 11 Hoehn, K. and White, T.D., Glutamate-evoked release of endogenous adenosine from rat cortical synaptosomes is mediated by glutamate uptake and not by receptors, J. Neurochem., 54 (1990) 1716-1724. 12 Jack, J.J.B., Miller, S., Porter, R. and Redman, S.J., The timecourse of minimal excitatory post-synaptie potentials evoked in spinal motoneurones by group la afferent fibres, Z Physiol., 215 (1971) 353-380. 13 Jacobson, I. and Hamberger, A., Kainic acid-induced changes of extracellular amino acid levels, evoked potentials and EEG activity in the rabbit olfactory bulb, Brab~ Res., 348 (1985) 289-296. 14 Jahr, C.E. and Yoshioka, K., la afferent excitation of motoneurunes in the in vitro new-born rat spinal cord is selectively antagonized by kynurenate, J. PhysioL, 370 (1986) 515-530. 15 Kuno, M. and Matsuura, S., Sites and mechanisms of action of lidocaine upon the isolated spinal cord of the frog, Brain Res., 249 (1982) 87-93. 16 Kuno, M., Yasunami, T. and Matsuura, S., Excitatory post-synaptic currents in response to different synaptic inputs of frog spinal motoneurons, Neu~'osci. Res., 5 (1988)240-252. 17 Lacey, G. and Nistri, A., Changes in glutamate-evoked currents of frog motoneurones by extracellular Mg a+, Neurosci. Lett., 90 (1988) 213-218. 18 Lehmann, A., Isacsson, H. and Hamberger, A., Effects of in vivo administration of kainic acid on the extracellular amino acid pool in the rabbit hippocampus, J. Neurochem., 40 (1983) 1314-1320. 19 Liu, G., Feldman, J.L. and Smith, J.C., Excitatory amino acidmediated transmission of inspiratory drive to phrenic motoneuruns, J. Neurophysiol., 64 (1990) 423-436. 20 L6pez-Colom6, A.M. and Roberts, P.J., Effect of excitatory amino acid analogues on the release of o-[3H]aspartate from chick retina, Eur. J. Pharmacol., 142 (1987) 409-417. 21 Martin, D., Bustos, G.A., Bowe, M.A., Bray, S.D. and Nadler, J.V., Autoreceptor regulation of glutamate and aspartate release from slices of the hippocampal CAI area, J. Neurochem., 56 (1991) 1647-1655. 22 Matsuura, S., Kawaguchi, S., Ichiki, M., Sorimachi, M., Kataoka, K. and Inouye, A., Perfusion of frog's spinal cord as convenient method for neuropharmacological studies, Eur. J. Pharmacol., ~ (1969) 13-16. 23 Mayer, M.L. and Westbrook, G.L., Mixed agonist action of excitatory amino acids on mouse spinal cord neurones under voltage clamp, J. Physiol., 354 (1984)29-53. 24 McBean, G.J. and Roberts, P.J., Glutamate-preferring receptors regulate the release of D-[3H]aspartate from rat hippocampai slices, Nature, 291 (1981) 593-594. 25 Men6ndez, N., Herreras, O., Solis, J.M., Herranz, A.S. and del Rio, R.M., Extracellular taurine increase in rat hippocampus
130
26 27 28 29
30 31
evoked by specific glutamate receptor activation is related to the excitatory potency of glutamate agonists, Neurosci. Lett., 102 (1989) 64-69. Miwa, A., Ui, M. and Kawai, N., G protein is coupled to p~synaptic glutamate and GABA receptors in lobster neuromuscular synapse, J. Neurophysiol., 63 (1990) 173-180. Nakamura, F., Kuno, M. and Matsuura, S., Effects of giutamate on monosynaptic excitatory transmission at spinal motoneurons, Jpn. J. Physiol., 41 (1991) S166. Nicoll, R.A., Malenka, R.C. and Kauer, J.A., Functional comparison of neurotransmitter receptor subtypes in mammalian central nervous system, Physiol. Rer., 70 (1990) 513-565. Pin, J.-P. and Bockaert, J., Two distinct mechan!sms, differentially affected by excitatory amino acids, trigger GAB~ release from fetal mouse striatal neurons in primary culture, J. Neurosci., 9 (1989) 648-656. Prigioni, I., Russo, G., Valli, P. and Masetto, S., Pre- and postsynaptic excitatory action of glutamate agonists on frog vestibular receptors, Hearing Res., 46 (1990) 253-260. Pumain, R., Kurcewicz, I. and Louvel, J., Ionic changes induced by excitatory amino acids in the rat cerebral cortex, Can. J. Physiol. PharmacoL, 65 (1987) 1067-1077.
32 Rail, W., Burke, R.E., Smith, T.G., Nelson, P.G. and Frank, K., Dendritic location of synapses and possible mechanisms for the monosynaptic EPSP in motoneurons, J. Neurophysiol. 30 (1967) 1169-1193. 33 Rice, M.E. and Nicholson, C., Glutamate- and aspartate-induced extracellular potassium and calcium shifts and their relation to those of kainate, quisqualate and N-methyl-D-aspartate in the isolated turtle cerebellum, Neuroscience, 38 (1990) 295-310. 34 Sykov;i, E. and Orkand, R.K., Extracellular potassium accumulation and transmission in frog spinal cord, Neuroscience, 5 (1980) 1421-1428. 35 Takeuchi, A., The transmitter role of glutamate in nervous systems, Jpn. J. Physiol., 37 (1987) 559-572. 36 Trombley, P.Q. and Westbrook, G.L., Excitatory synaptic transmission in cultures of rat olfactory bulb, J. Neurophysiol., 64 (1990) 598-606. 37 Yasunami, T., Kuno, M. and Matsuura, S., Voltage-clamp analysis of taurine-induced suppression of excitatory post-synaptic potentials in frog spinal motoneurons, J. Neurophysiol., 60 (1988) 1405-1418. 38 Zucker, R.S., Short-term synaptic plasticity, Annu. Rev. Neurosci., 12 (1989) 13-31.