Neuroseience Letters, 143 (1992) 131-134 ~) 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
131
NSL 08874
Excitatory amino acid receptors involved in primary afferent-evoked polysynaptic EPSPs of substantia gelatinosa neurons in the adult rat spinal cord slice M e g u m u Yoshimura and Syogoro Nishi Department of Physiology, Kurume University School of Medicine, Kurume (Japan) (Received 29 April 1992; Revised version received 19 May 1992: Accepted 22 May 1992)
Key words: Monosynaptic EPSP; Nociception; NMDA receptor; CNQX; APV; Pain; Dorsal horn Intracellular recordings were made from substantia gelatinosa (SG) neurons in spinal cord slices to determine a subclass of excitatory amino acid receptors involved in polysynaptic excitatory postsynaptic potentials (EPSPs). In the majority of neurons, polysynaptic EPSPs evoked by A6 fiber were not affected by 2-amino-5-phosphonovaleric acid (APV), while all EPSPs including monosynaptic EPSPs were depressed by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). All spontaneous EPSPs were blocked by CNQX, while spontaneous EPSPs in a few SG neurons were attenuated by APV. These observations suggest that polysynaptic EPSPs evoked through A~ fibers are predominantly mediated by activation of the non-N-methylD- aspartate (non-NMDA) receptor subclass.
It has been proposed that polysynaptic excitatory postsynaptic potentials (EPSPs) of spinal cord neurons evoked by primary afferent stimulation are mediated by N-methyl-D-aspartic acid (NMDA) receptors whereas monosynaptic EPSPs are mediated by non-NMDA (AMPA/kainate) receptors. This notion is derived from observations that NMDA receptor antagonists depressed the slow component of the dorsal root-ventral root potential (DR-VRP) that was believed to be polysynaptic, and often left the fast component unaffected in the isolated spinal cord [6, 8]. Experiments with new NMDA and non-NMDA receptor antagonists and other preparations support this notion [3-5, 9]. Recently, it has been confirmed that monosynaptic EPSPs evoked by primary afferent stimulation of substantia gelatinosa neurons in spinal dorsal horn are mediated by the nonNMDA receptor subclass [11]. On the other hand, not all studies concur with the view of the involvement of NMDA receptors in the polysynaptic excitation of spinal neurons. An intracellular study of frog motoneurons, for example, showed that the polysynaptic EPSPs evoked by primary afferent stimulation were less sensitive to D-c~amino adipate (DAA) and 2-amino-5-phosphonovaleric
Correspondence: M. Yoshimura, Department of Physiology, Kurume University School of Medicine, 67 Asahi-machi, Kurume, 830 Japan.
acid (APV) than the monosynaptic EPSPs [1, 2]. Thus, the involvement of NMDA receptors in polysynaptic pathway in the spinal cord is still controversial. In the present study, we recorded intracellularly from substantia gelatinosa (SG, lamina I1 of Rexed) [10] neurons of the dorsal horn of adult rat spinal cord slices which retained an attached dorsal root and tested the effect of NMDA and non-NMDA receptor antagonists on primary afferent-evoked polysynaptic and spontaneous EPSPs. The methods for obtaining slices of the adult rat spinal cord have been described elsewhere [11]. Briefly, lumbosacral spinal cords were removed from adult rats (8 16 weeks old) under anesthesia with urethane (1.5 g/kg, i.p.). Alter removal of the dura mater, all ventral and dorsal roots were cut, with the exception of the L5 or L6 dorsal root on one side. The pia-arachnoid membrane was removed and then the spinal cord was mounted on a plexy-glass stage with cyanoacrylate glue. About 500/Ira thickness of the transverse slice was cut on a vibratome. The slice of spinal cord with the dorsal root attached was placed on a nylon mesh in the recording chamber and was perfused with Krebs solution (16-18 ml/min) equilibrated with 95% 02 and 5% COn at 36 +_ I°C. The composition of the Krebs solution was (in mM): 117 NaCI, 3.6 KC1, 2.5 CaC12, 1.2 MgC12, 1.2 NaH2PO4, 25 NaHCO3 and 11 glucose. The minimum stimulus inten-
132
2.5 mV
B1
B2
Fig. 1. Synapticresponsesin substantiagelatinosaneuronsevokedby primaryafferentAd fibers.AI: monosynapticEPSPsfollowedby polysynaptic EPSPs(arrows). A2: stimulationof A~fiberevokedonlypolysynapticEPSPs.B1: stimulationof A~fiberevokedboth monosynapticand polysynaptic EPSPs. B2: the polysynapticEPSPswere not observedin solutioncontaining5 mM calciumand 5 mM magnesium.Note that the polysynaptic EPSPs had long-and variable-latency.Three superimposedtraces are shown.
sity required to activate A~ fibers was determined by monitoring the compound action potentials in an isolated sciatic nerve-dorsal root preparation using a silverwire hook electrode. The minimum stimulus intensity and duration required to activate A6 fibers was approximately 2.0 V and 0.1 ms, respectively. The same suction electrode was then used to activate dorsal root afferents which had a length of 6-10 mm in the in vitro slice preparation. Intracellular recordings were obtained from 72 SG neurons that exhibited either monosynaptic-polysynaptic EPSPs or polysynaptic EPSPs alone responded to the A~ afferent fiber stimulation. Recorded SG neurons had the resting membrane potential of -65_+7 mV (mean+S.D., n=34) and the apparent input resistance of 260+70 Mr9 (n=34). Single low intensity stimuli (2.0-5.0 V, 0.1 ms) sufficient to activate A~ fibers applied to dorsal root evoked fast EPSPs with short- and constant-latency in 79% of the SG neurons. The EPSPs were followed by long- and variable-latency EPSPs (Fig. 1A 1). In the remaining 21% of SG neurons, stimulation of A8 fibers evoked longand variable-latency EPSPs alone (Fig. 1A2). Stimulations of Aot/fl fibers (less than 2.0 V, 0.1 ms) did not evoke short- and constant-latency EPSPs. The short- and constant-latency EPSPs have been shown to be monosynaptic based on the evidence that the latency of EPSPs was short (0.7-2.2 ms) and was constant in response to a high-frequency repetitive (20 Hz) dorsal root stimulation. In addition, the shape of EPSPs was similar in each trial, although the amplitude of EPSPs was dependent on the stimulus intensity (Fig. 1A1 ,B). In contrast, the long-
latency EPSPs were variable in latency in responses to a repetitive (10 Hz) stimulation and the latency of EPSPs (ranging from 3.0 to 20 ms) was much longer than that of the monosynaptic EPSPs. In addition, the amplitude and shape of long-latency EPSPs were variable in response to the stimulation of dorsal root even with a constant stimulus intensity (Fig. IA,B and 2A-C). Moreover, the longlatency EPSPs disappeared in the solution containing high Ca 2+ and high Mg2+ (Fig. 1B2) [11]. Blockade of the long-latency EPSPs by high divalent cations was not due to a voltage-dependent block of NMDA channels by the high concentration of Mg 2+, since the long-latency EPSPs were not sensitive to APV in the majority of neurons tested. These observations suggest that the longand variable-latency EPSPs evoked by primary afferent Ad~fibers are polysynaptic in nature. In 75% of the SG neurons which exhibited either monosynaptic-polysynaptic EPSPs or polysynaptic EPSPs alone, an NMDA receptor antagonist APV (50-200/aM) had little effect on the amplitude and frequency of the polysynaptic EPSPs (Fig. 2A). In the remaining 25% of SG neurons, APV effectively depressed the polysynaptic EPSPs (Fig. 2B). Both APV-insensitive and APV-sensitive polysynaptic EPSPs as well as monosynaptic EPSPs were blocked by a non-NMDA receptor antagonist 6-cyano-7-nitroquinoxatine-2,3-dione (CNQX; 5-10 /aM) (not shown). The neuron shown in Fig. 2C exhibited only polysynaptic EPSPs in response to the primary afferent A6 fiber stimulation. These polysynaptic EPSPs were not affected by APV (Fig. 2C2), but attenuated reversibly by CNQX (Fig. 2C3). This observation suggests that the NMDA receptor is not significantly involved in
133
A1
c1
B1
,
Control
Control
Control
C2 F-.
A2
APV
B2 APV
*
APV
C3
I '"v"
CNQX
J 10 mV i
80 ms
10
ms
20 ms
Fig. 2. Effects of APV and CNQX on the polysynaptic EPSPs evoked by Aa fibers. A 1: stimulation of the A6 fiber evoked monosynaptic t'PSPs and marked polysynaptic EPSPs. A2: APV had little effect on the amplitude and frequency of the polysynaplic EPSPs. Asterisks indicate truncated spikes, B: polysynaptic EPSPs (arrow) evoked by Aa fiber stimulation were blocked by APV. CI: stimulation of Aa fiber evoked only polys~naptic EPSPs. C2: APV had little effect on the polysynaptic EPSPs. CNQX blocked the polysynaptic EPSPs (C3). Three or 4 superimposcd traces arc shox~ n.
this polysynaptic pathway, although it cannot be ruled out that a small fraction of the polysynaptic EPSPs is mediated by the N M D A receptor. To elucidate whether interneurons involved in the polysynaptic pathway release glutamate or a related amino acid as a transmitter, we tested the effect of CNQX and APV on spontaneous EPSPs and EPSPs evoked by focal stimulation applied to SG neurons near the impaled cell. All SG neurons exhibited spontaneous EPSPs with frequency of I 50 Hz and amplitude of 0.5 20 mV. In 5 out of 7 neurons, APV (50/aM) had no detectable effect on the amplitude and frequency of spontaneous EPSPs: the amplitude and frequency of spontaneous EPSPs in the neuron shown in Fig. 3 before application of APV were 3.1+2.1 mV and 37 Hz, respectively, while the mean amplitude and frequency were 3.2+2.1 mV and 35 Hz, respectively, in the presence of APV (50/aM). In the other 2 out of 7 SG neurons, spontaneous EPSPs were attenuated by APV (not shown). All spontaneous EPSPs were almost completely blocked by CNQX (n=12) (Fig. 3C). Blockade of all spontaneous EPSPs by CNQX was also observed in Mg 2+ free solutions. A single low intensity focal stimulus with the monopolar electrode, which was placed on the SG ventral region close to the recorded neurons, evoked EPSPs in all SG neurons tested. All evoked EPSPs had a short-latency ranging from 0.5 to 0.6 ms suggesting that the EPSPs were evoked by the activation of interneurons and/or axons of interneurons which established monosynaptic contact with the recorded neurons. CNQX (5 10/aM)
depressed these EPSPs in amplitude by more than 75%, while APV depressed EPSPs by less than 30%. APV, however, shortened the half decay time of the EPSPs in all SG neurons. The evoked EPSPs were almost completely blocked by solution containing both CNQX (10 # M ) and APV (50 MM) (not shown). These findings are consistent with other studies showing that N M D A receptors contribute primarily to later components of EPSPs in spinal neurons [7, l 1]. These observations suggest that not only primary afferent Aa fiber but also interneurons release glutamate or a related amino acid as a transmitter onto the SG neurons. Our results suggest that although the N M D A receptor is likely to be involved in the polysynaptic pathway in a small fraction of the SG neurons, the polysynaptic EPSPs evoked by primary afferent Aa fiber stimulation are mediated predominantly by the activation of nonN M D A receptor subclass. This is inconsistent with the view reported previously [3 6, 8, 9]. There are several possible explanations for the discrepancy. Our results were obtained exclusively from SG neurons and by stimulation of A6 afferent fibers, while previous studies used unidentified primary afferents to activate the polysynaptic pathway and recorded from various kinds of spinal neurons. It has been established that the different diameter of primary afferent fibers terminate at the peculiar laminae and neurons in the spinal cord. It is, theretk)re, possible that the polysynaptic EPSPs in the other laminae are mediated predominantly by the N M D A receptor. Other factors need to be considered are species dif-
134
A
N M DA receptor, whole cell patch clamp analysis of miniature EPSPs in slices will be required.
Control
We are grateful to Drs. H. Higashi and A.B. MacDermott for valuable comments on the manuscript. ]his work was supported by The Ishibashi Research Fund and a Grand-in-Aid for Scientific Research by the Ministry of Education, Science and Culture of Japan.
B
C
APV
CNQX
10 rrlV
D
Wmeh
I
][
]]]
100 ms Fig. 3. Spontaneous EPSPs in substantia gelatinosa neurons were not sensitive to APV but blocked by CNQX. Spontaneous EPSPs were observed at membrane potential of -70 mV. APV had little effect on the amplitude and frequency of EPSPs (B). CNQX almost completely blocked the spontaneous EPSPs (C). The effect of CNQX was reversible (D). Three superimposed traces are shown.
ferences, maturity of the neurons and different ionic concentrations, particularly Mg 2+, known to be a N M D A receptor antagonist. To address whether spinal neurons exhibit the synaptic response which is mediated solely by activation of the
1 Arenson, M.S., Berti, C., King, A.E. and Nistri, A., The effects of D-x-amino adipate on excitatory amino acid responses recorded intracellularly from motoneurones of the frog spinal cord, Neurosci. Lett., 49 (1984) 99-104. 2 Corradetti, R., King, A.E., Nistri, A., Rovera, C. and Sivilotti, L., Pharmacological characterization of DAPV antagonism of amino acid and synaptically evoked excitations on frog motoneurones in vitro: an intracellular study, Br. J. Pharmacol., 86 {1985) 19 25. 3 Davies, J. and Watkins, J.C., Role of excitatory amino acid receptors in mono- and polysynaptic excitation in the cat spinal cord, Exp. Brain Res., 49 (1983) 280 290. 4 Davies, J. and Watkins, J.C., Depressant actions of ),-t>glutamylaminomethyl sulphonate (GAMS) on amino acid induced and synaptic excitation in the cat spinal cord, Brain Res., 327 (1985) 113- 120. 5 Elmslie, K.S. and Yoshikami, D., Effects of kynurenate on root potentials evoked by synaptic activity and amino acids in the frog spinal cord, Brain Res., 330 (1985) 265-272. 6 Evans, R.H., Francis, A.A., Jones, A.W., Smith, D.A.S. and Watkins, J.C., The effects of a series of ~-phosphonic ~-carboxylic amino acids on electrically evoked and excitant amino acid induced responses in isolated spinal cord preparations; Br. J. Pharmacol., 75 (1982) 65-75. 7 Forsythe, I.D. and Westbrook, G.L., Slow excitatory postsynaptic currents mediated by N-methyl-D-aspartate receptor on cultured mouse central neurones, J. Physiol., 396 (1988) 515-533. 8 Padjen, A.L. and Smith, RA., Specific effects of c~-D,L-aminoadipjc acid on synaptic transmission in frog spinal cord, Can. J. Physiol. Pharmacol., 58 (1980) 692-698. 9 Polc, E. 2APH depresses gamma motoneurons and polysynaptic reflexes in the cat spinal cord, Eur. J. Pharmacol., 117 (1985) 3 8 7 389. 10 Rexed, B., The cytoarchitectonic organization of the spinal cord in the cat, J. Comp. Neurol., 96 (1952) 4154t95. 11 Yoshimura, M. and Jessell, T.M., Amino acid-mediated EPSPs at primary afferent synapses with substantia gelatinosa neurones in the rat spinal cord, J. Physiol., 430 (1990) 315-335.
131
NSL 08874
Excitatory amino acid receptors involved in primary afferent-evoked polysynaptic EPSPs of substantia gelatinosa neurons in the adult rat spinal cord slice M e g u m u Yoshimura and Syogoro Nishi Department of Physiology, Kurume University School of Medicine, Kurume (Japan) (Received 29 April 1992; Revised version received 19 May 1992: Accepted 22 May 1992)
Key words: Monosynaptic EPSP; Nociception; NMDA receptor; CNQX; APV; Pain; Dorsal horn Intracellular recordings were made from substantia gelatinosa (SG) neurons in spinal cord slices to determine a subclass of excitatory amino acid receptors involved in polysynaptic excitatory postsynaptic potentials (EPSPs). In the majority of neurons, polysynaptic EPSPs evoked by A6 fiber were not affected by 2-amino-5-phosphonovaleric acid (APV), while all EPSPs including monosynaptic EPSPs were depressed by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). All spontaneous EPSPs were blocked by CNQX, while spontaneous EPSPs in a few SG neurons were attenuated by APV. These observations suggest that polysynaptic EPSPs evoked through A~ fibers are predominantly mediated by activation of the non-N-methylD- aspartate (non-NMDA) receptor subclass.
It has been proposed that polysynaptic excitatory postsynaptic potentials (EPSPs) of spinal cord neurons evoked by primary afferent stimulation are mediated by N-methyl-D-aspartic acid (NMDA) receptors whereas monosynaptic EPSPs are mediated by non-NMDA (AMPA/kainate) receptors. This notion is derived from observations that NMDA receptor antagonists depressed the slow component of the dorsal root-ventral root potential (DR-VRP) that was believed to be polysynaptic, and often left the fast component unaffected in the isolated spinal cord [6, 8]. Experiments with new NMDA and non-NMDA receptor antagonists and other preparations support this notion [3-5, 9]. Recently, it has been confirmed that monosynaptic EPSPs evoked by primary afferent stimulation of substantia gelatinosa neurons in spinal dorsal horn are mediated by the nonNMDA receptor subclass [11]. On the other hand, not all studies concur with the view of the involvement of NMDA receptors in the polysynaptic excitation of spinal neurons. An intracellular study of frog motoneurons, for example, showed that the polysynaptic EPSPs evoked by primary afferent stimulation were less sensitive to D-c~amino adipate (DAA) and 2-amino-5-phosphonovaleric
Correspondence: M. Yoshimura, Department of Physiology, Kurume University School of Medicine, 67 Asahi-machi, Kurume, 830 Japan.
acid (APV) than the monosynaptic EPSPs [1, 2]. Thus, the involvement of NMDA receptors in polysynaptic pathway in the spinal cord is still controversial. In the present study, we recorded intracellularly from substantia gelatinosa (SG, lamina I1 of Rexed) [10] neurons of the dorsal horn of adult rat spinal cord slices which retained an attached dorsal root and tested the effect of NMDA and non-NMDA receptor antagonists on primary afferent-evoked polysynaptic and spontaneous EPSPs. The methods for obtaining slices of the adult rat spinal cord have been described elsewhere [11]. Briefly, lumbosacral spinal cords were removed from adult rats (8 16 weeks old) under anesthesia with urethane (1.5 g/kg, i.p.). Alter removal of the dura mater, all ventral and dorsal roots were cut, with the exception of the L5 or L6 dorsal root on one side. The pia-arachnoid membrane was removed and then the spinal cord was mounted on a plexy-glass stage with cyanoacrylate glue. About 500/Ira thickness of the transverse slice was cut on a vibratome. The slice of spinal cord with the dorsal root attached was placed on a nylon mesh in the recording chamber and was perfused with Krebs solution (16-18 ml/min) equilibrated with 95% 02 and 5% COn at 36 +_ I°C. The composition of the Krebs solution was (in mM): 117 NaCI, 3.6 KC1, 2.5 CaC12, 1.2 MgC12, 1.2 NaH2PO4, 25 NaHCO3 and 11 glucose. The minimum stimulus inten-
132
2.5 mV
B1
B2
Fig. 1. Synapticresponsesin substantiagelatinosaneuronsevokedby primaryafferentAd fibers.AI: monosynapticEPSPsfollowedby polysynaptic EPSPs(arrows). A2: stimulationof A~fiberevokedonlypolysynapticEPSPs.B1: stimulationof A~fiberevokedboth monosynapticand polysynaptic EPSPs. B2: the polysynapticEPSPswere not observedin solutioncontaining5 mM calciumand 5 mM magnesium.Note that the polysynaptic EPSPs had long-and variable-latency.Three superimposedtraces are shown.
sity required to activate A~ fibers was determined by monitoring the compound action potentials in an isolated sciatic nerve-dorsal root preparation using a silverwire hook electrode. The minimum stimulus intensity and duration required to activate A6 fibers was approximately 2.0 V and 0.1 ms, respectively. The same suction electrode was then used to activate dorsal root afferents which had a length of 6-10 mm in the in vitro slice preparation. Intracellular recordings were obtained from 72 SG neurons that exhibited either monosynaptic-polysynaptic EPSPs or polysynaptic EPSPs alone responded to the A~ afferent fiber stimulation. Recorded SG neurons had the resting membrane potential of -65_+7 mV (mean+S.D., n=34) and the apparent input resistance of 260+70 Mr9 (n=34). Single low intensity stimuli (2.0-5.0 V, 0.1 ms) sufficient to activate A~ fibers applied to dorsal root evoked fast EPSPs with short- and constant-latency in 79% of the SG neurons. The EPSPs were followed by long- and variable-latency EPSPs (Fig. 1A 1). In the remaining 21% of SG neurons, stimulation of A8 fibers evoked longand variable-latency EPSPs alone (Fig. 1A2). Stimulations of Aot/fl fibers (less than 2.0 V, 0.1 ms) did not evoke short- and constant-latency EPSPs. The short- and constant-latency EPSPs have been shown to be monosynaptic based on the evidence that the latency of EPSPs was short (0.7-2.2 ms) and was constant in response to a high-frequency repetitive (20 Hz) dorsal root stimulation. In addition, the shape of EPSPs was similar in each trial, although the amplitude of EPSPs was dependent on the stimulus intensity (Fig. 1A1 ,B). In contrast, the long-
latency EPSPs were variable in latency in responses to a repetitive (10 Hz) stimulation and the latency of EPSPs (ranging from 3.0 to 20 ms) was much longer than that of the monosynaptic EPSPs. In addition, the amplitude and shape of long-latency EPSPs were variable in response to the stimulation of dorsal root even with a constant stimulus intensity (Fig. IA,B and 2A-C). Moreover, the longlatency EPSPs disappeared in the solution containing high Ca 2+ and high Mg2+ (Fig. 1B2) [11]. Blockade of the long-latency EPSPs by high divalent cations was not due to a voltage-dependent block of NMDA channels by the high concentration of Mg 2+, since the long-latency EPSPs were not sensitive to APV in the majority of neurons tested. These observations suggest that the longand variable-latency EPSPs evoked by primary afferent Ad~fibers are polysynaptic in nature. In 75% of the SG neurons which exhibited either monosynaptic-polysynaptic EPSPs or polysynaptic EPSPs alone, an NMDA receptor antagonist APV (50-200/aM) had little effect on the amplitude and frequency of the polysynaptic EPSPs (Fig. 2A). In the remaining 25% of SG neurons, APV effectively depressed the polysynaptic EPSPs (Fig. 2B). Both APV-insensitive and APV-sensitive polysynaptic EPSPs as well as monosynaptic EPSPs were blocked by a non-NMDA receptor antagonist 6-cyano-7-nitroquinoxatine-2,3-dione (CNQX; 5-10 /aM) (not shown). The neuron shown in Fig. 2C exhibited only polysynaptic EPSPs in response to the primary afferent A6 fiber stimulation. These polysynaptic EPSPs were not affected by APV (Fig. 2C2), but attenuated reversibly by CNQX (Fig. 2C3). This observation suggests that the NMDA receptor is not significantly involved in
133
A1
c1
B1
,
Control
Control
Control
C2 F-.
A2
APV
B2 APV
*
APV
C3
I '"v"
CNQX
J 10 mV i
80 ms
10
ms
20 ms
Fig. 2. Effects of APV and CNQX on the polysynaptic EPSPs evoked by Aa fibers. A 1: stimulation of the A6 fiber evoked monosynaptic t'PSPs and marked polysynaptic EPSPs. A2: APV had little effect on the amplitude and frequency of the polysynaplic EPSPs. Asterisks indicate truncated spikes, B: polysynaptic EPSPs (arrow) evoked by Aa fiber stimulation were blocked by APV. CI: stimulation of Aa fiber evoked only polys~naptic EPSPs. C2: APV had little effect on the polysynaptic EPSPs. CNQX blocked the polysynaptic EPSPs (C3). Three or 4 superimposcd traces arc shox~ n.
this polysynaptic pathway, although it cannot be ruled out that a small fraction of the polysynaptic EPSPs is mediated by the N M D A receptor. To elucidate whether interneurons involved in the polysynaptic pathway release glutamate or a related amino acid as a transmitter, we tested the effect of CNQX and APV on spontaneous EPSPs and EPSPs evoked by focal stimulation applied to SG neurons near the impaled cell. All SG neurons exhibited spontaneous EPSPs with frequency of I 50 Hz and amplitude of 0.5 20 mV. In 5 out of 7 neurons, APV (50/aM) had no detectable effect on the amplitude and frequency of spontaneous EPSPs: the amplitude and frequency of spontaneous EPSPs in the neuron shown in Fig. 3 before application of APV were 3.1+2.1 mV and 37 Hz, respectively, while the mean amplitude and frequency were 3.2+2.1 mV and 35 Hz, respectively, in the presence of APV (50/aM). In the other 2 out of 7 SG neurons, spontaneous EPSPs were attenuated by APV (not shown). All spontaneous EPSPs were almost completely blocked by CNQX (n=12) (Fig. 3C). Blockade of all spontaneous EPSPs by CNQX was also observed in Mg 2+ free solutions. A single low intensity focal stimulus with the monopolar electrode, which was placed on the SG ventral region close to the recorded neurons, evoked EPSPs in all SG neurons tested. All evoked EPSPs had a short-latency ranging from 0.5 to 0.6 ms suggesting that the EPSPs were evoked by the activation of interneurons and/or axons of interneurons which established monosynaptic contact with the recorded neurons. CNQX (5 10/aM)
depressed these EPSPs in amplitude by more than 75%, while APV depressed EPSPs by less than 30%. APV, however, shortened the half decay time of the EPSPs in all SG neurons. The evoked EPSPs were almost completely blocked by solution containing both CNQX (10 # M ) and APV (50 MM) (not shown). These findings are consistent with other studies showing that N M D A receptors contribute primarily to later components of EPSPs in spinal neurons [7, l 1]. These observations suggest that not only primary afferent Aa fiber but also interneurons release glutamate or a related amino acid as a transmitter onto the SG neurons. Our results suggest that although the N M D A receptor is likely to be involved in the polysynaptic pathway in a small fraction of the SG neurons, the polysynaptic EPSPs evoked by primary afferent Aa fiber stimulation are mediated predominantly by the activation of nonN M D A receptor subclass. This is inconsistent with the view reported previously [3 6, 8, 9]. There are several possible explanations for the discrepancy. Our results were obtained exclusively from SG neurons and by stimulation of A6 afferent fibers, while previous studies used unidentified primary afferents to activate the polysynaptic pathway and recorded from various kinds of spinal neurons. It has been established that the different diameter of primary afferent fibers terminate at the peculiar laminae and neurons in the spinal cord. It is, theretk)re, possible that the polysynaptic EPSPs in the other laminae are mediated predominantly by the N M D A receptor. Other factors need to be considered are species dif-
134
A
N M DA receptor, whole cell patch clamp analysis of miniature EPSPs in slices will be required.
Control
We are grateful to Drs. H. Higashi and A.B. MacDermott for valuable comments on the manuscript. ]his work was supported by The Ishibashi Research Fund and a Grand-in-Aid for Scientific Research by the Ministry of Education, Science and Culture of Japan.
B
C
APV
CNQX
10 rrlV
D
Wmeh
I
][
]]]
100 ms Fig. 3. Spontaneous EPSPs in substantia gelatinosa neurons were not sensitive to APV but blocked by CNQX. Spontaneous EPSPs were observed at membrane potential of -70 mV. APV had little effect on the amplitude and frequency of EPSPs (B). CNQX almost completely blocked the spontaneous EPSPs (C). The effect of CNQX was reversible (D). Three superimposed traces are shown.
ferences, maturity of the neurons and different ionic concentrations, particularly Mg 2+, known to be a N M D A receptor antagonist. To address whether spinal neurons exhibit the synaptic response which is mediated solely by activation of the
1 Arenson, M.S., Berti, C., King, A.E. and Nistri, A., The effects of D-x-amino adipate on excitatory amino acid responses recorded intracellularly from motoneurones of the frog spinal cord, Neurosci. Lett., 49 (1984) 99-104. 2 Corradetti, R., King, A.E., Nistri, A., Rovera, C. and Sivilotti, L., Pharmacological characterization of DAPV antagonism of amino acid and synaptically evoked excitations on frog motoneurones in vitro: an intracellular study, Br. J. Pharmacol., 86 {1985) 19 25. 3 Davies, J. and Watkins, J.C., Role of excitatory amino acid receptors in mono- and polysynaptic excitation in the cat spinal cord, Exp. Brain Res., 49 (1983) 280 290. 4 Davies, J. and Watkins, J.C., Depressant actions of ),-t>glutamylaminomethyl sulphonate (GAMS) on amino acid induced and synaptic excitation in the cat spinal cord, Brain Res., 327 (1985) 113- 120. 5 Elmslie, K.S. and Yoshikami, D., Effects of kynurenate on root potentials evoked by synaptic activity and amino acids in the frog spinal cord, Brain Res., 330 (1985) 265-272. 6 Evans, R.H., Francis, A.A., Jones, A.W., Smith, D.A.S. and Watkins, J.C., The effects of a series of ~-phosphonic ~-carboxylic amino acids on electrically evoked and excitant amino acid induced responses in isolated spinal cord preparations; Br. J. Pharmacol., 75 (1982) 65-75. 7 Forsythe, I.D. and Westbrook, G.L., Slow excitatory postsynaptic currents mediated by N-methyl-D-aspartate receptor on cultured mouse central neurones, J. Physiol., 396 (1988) 515-533. 8 Padjen, A.L. and Smith, RA., Specific effects of c~-D,L-aminoadipjc acid on synaptic transmission in frog spinal cord, Can. J. Physiol. Pharmacol., 58 (1980) 692-698. 9 Polc, E. 2APH depresses gamma motoneurons and polysynaptic reflexes in the cat spinal cord, Eur. J. Pharmacol., 117 (1985) 3 8 7 389. 10 Rexed, B., The cytoarchitectonic organization of the spinal cord in the cat, J. Comp. Neurol., 96 (1952) 4154t95. 11 Yoshimura, M. and Jessell, T.M., Amino acid-mediated EPSPs at primary afferent synapses with substantia gelatinosa neurones in the rat spinal cord, J. Physiol., 430 (1990) 315-335.
