JOURNALOF NEUROPHYSIOLOGY Vol. 65, No. 1, January 1991. Printed
in U. X,4.
Initiation of Epileptiform Activity by Excitatory Amino Acid Receptors in the Disinhibited Rat Neocortex WAI-LING LEE AND JOHN J. HABLITZ Neurobiology Research Center, Department of Physiology at Birmingham, Birmingham, Alabama 35294 SUMMARY
AND
CONCLUSIONS
1. Intracellular recordings were obtained from neurons in layer II-III of rat frontal cortex maintained in vitro. The role of excitatory amino acid receptors in generation of picrotoxin (PTX)-induced epileptiform activity was investigated with the use of D-2amino+phosphonovaleric acid (D-APV) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) as selective antagonists of N-methyl-D-aspartate (NMDA) and non-NMDA receptors, respectively. 2. Bath application of PTX resulted in a decrease in evoked inhibitory postsynaptic potentials (IPSPs) in neocortical neurons and a concomitant increase in a polysynaptic late excitatory postsynaptic potential (1EPSP). Epileptiform burst responses, termed paroxysmal depolarizing shifts (PDSs), subsequently developed. Based on response duration, two types of PDSs were identified. Long PDSs were > 100 ms in duration, whereas short PDSs lasted ~50 ms. An early depolarizing potential preceded both types of epileptiform burst response. 3. The NMDA receptor antagonist D-APV reduced the peak amplitude and duration of the PDS. D-APV-insensitive portions of the PDS were greatly attenuated or abolished by CNQX. The non-NMDA antagonist also increased the latency to PDS onset and reduced its duration without affecting peak amplitude. CNQX-insensitive components of the PDS, when present, were abolished by D-APV. 4. Short-duration PDSs could be blocked by CNQX. In these neurons, increasing the stimulation strength produced epileptiform responses of reduced amplitude. 5. Under control conditions, PDS amplitude was a linear function of membrane potential, increasing with hyperpolarization and diminishing on depolarization. The early depolarizing response preceding the PDS showed a variable dependence on voltage. Both decreases and increases in amplitude were observed, in different neurons, on membrane depolarization. 6. In the presence of CNQX, PDS amplitude remained linearly related to membrane potential. Early depolarizations preceding the PDS displayed an unconventional voltage sensitivity, increasing with depolarization over the range - 100 to -60 mV. 7. We conclude that, in the PTX disinhibition model of epilepsy, both NMDA and non-NMDA receptors contribute to epileptiform burst responses. An NMDA-receptor-mediated early depolarization may serve as a trigger for burst responses. NMDA receptors are also involved in maintenance of the peak of the paroxysmal depolarizations, whereas non-NMDA receptors are necessary for full expression of the PDS.
INTRODUCTION
Intracellular recordings from neocortical neurons in vivo (Matsumoto et al. 1969) and in vitro (Gutnick et al. 1982)
and Biophysics,
University of Alabama
have shown that interictal spike discharges in acute epileptogenic foci are accompanied by action-potential discharges riding on a large membrane depolarization [originally termed a paroxysmal depolarizing shift (PDS) by Matsumoto and Ajmone-Marsan ( 1964)]. Synaptic events have been suggested to play an important role in elaboration of neocortical epileptiform activity (Gutnick et al. 1982; Matsumoto et al. 1969) with a subpopulation of neurons in layer IV and upper layer V possibly responsible for initiation of paroxysmal events (Chagnac-Amitai and Connors 1989a,b; Connors 1984). The changes in normal synaptic activity responsible for generation of synchronous epileptiform discharges have not been described in detail in the neocortex. Moreover, although excitatory amino acid neurotransmission has been extensively investigated in the neocortex (Sutor and Hablitz 1989a,b; Thomson 1986), the role of specific receptor subtypes in initiation of the PDS and maintenance of the peak depolarization has not been determined. In the rat neocortex, the excitatory amino acids glutamate and aspartate are prominent neurotransmitter candidates (Streit 1984), acting on both N-methyl-D-aspartate (NMDA) and non-NMDA receptors. Application of the NMDA receptor antagonist D-2-amino-5-phosphonovaleric acid (D-APV) blocks a late excitatory postsynaptic potential (IEPSP) produced by activation of layer IV (Sutor and Hablitz 1989a,b) and late synaptic responses evoked by white matter stimulation (Thomson 1986). Early EPSPs (eEPSPs) are selectively blocked by the non-NMDA receptor antagonist 6 - cyano - 7 - nitroquinoxaline - 2,3 - dione (CNQX) (Hablitz and Sutor 1990). Studies of picrotoxininduced epileptiform activity in the hippocampus have shown bath application of 5 PM CNQX to significantly reduce or abolish evoked PDSs in both CA1 and CA3 neurons. In cells where a CNQX-insensitive component in the PDS was manifest, this remaining activity was abolished by the NMDA receptor antagonist D-APV (20 PM), suggesting the existence of an underlying NMDA-mediated synaptic potential (Lee and Hablitz 1989a). The role of the two receptor types in epileptiform activity in the neocortex is not well established. The present study sought to determine the contribution each receptor type made to the triggering or initiation of PDSs as opposed to the full expression or maintenance of the PDS waveform. In addition, we tested whether development of synchronized burst responses after application of the GABA,-receptor antagonist picrotoxin (PTX) resulted from a loss of inhibitory control of polysynaptic EPSPs in the neocortex.
0022-3077/9 1 $1 SO Copyright 0 199 1 The American Physiological Society Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
87
88
W.-L.
LEE
AND
A preliminary account of some of these findings has appeared (Lee and Hablitz 1989b). METHODS
Neocortical brain slices were obtained from adult SpragueDawley rats of both sexes. The preparation and maintenance of these slices employed methods described previously (Sutor and Hablitz 1989a). Briefly, rats were decapitated under ether anesthesia; the brains were quickly dissected out and placed in ice cold saline for 30-60 s. Coronal slices of frontal cortex (500 ,ccrnthick) were cut with a McIlwain tissue chopper and transferred to an interface type recording chamber. After 30 min at room temperature, slices were slowly warmed to the recording temperature of 34 t 1 “C and allowed to equilibrate at least 1.5 h before recording. The slices were continuously perfused with an oxygenated (95% O,-5% COJ saline consisting of the following (in mM): 124 NaCl, 5 MCI, 1.25 NaH,PO,, 2.0 CaCI,, 2.0 MgCl,, 26 NaHCO,, and 10 dextrose. Epileptiform activity was induced by addition of PTX (50 PM) to the perfusate. Intracellular recording electrodes were pulled from filamentcontaining borosilicate glass tubings (1.5 mm OD) and had resistances of 50- 100 ML! when filled with 4 M potassium acetate. Recordings were made with the use of a single-electrode timeshare circuit as described previously (Sutor and Hablitz 1989a). Bipolar stimulating electrodes were placed in layer IV - 1 mm lateral to the recording electrode. The strength of the orthodromic stimulation was set at 2 times threshold intensity needed for evoking a PDS. After impalement, the resting membrane potential was allowed to stabilize for a minimum of 30 min. Orthodromic stimulation was then applied once per minute for 10 min. After this baseline period, excitatory amino acid antagonists were bath applied while orthodromic stimulation continued to be applied. Antagonist effects were assessed after 1 h of drug perfusion with each neuron serving as its own control. Because of the long duration of the studies, only one cell was examined in each preparation. Neuronal input resistance (RN) and resting membrane potential were continuously monitored before and after antagonist application. This was accomplished by passing hyperpolarizing current pulses of -0.5 nA through the recording electrode and observing the resulting voltage deflection. To assess the effect of changes in membrane potential on the PDS amplitude, long-duration (3 s) current pulses were given with orthodromic stimulation applied 500 ms after pulse onset. Data were recorded on a Neurocorder DR-384 and analyzed with a PDP 1 l/23+ computer. Data are presented as mean t SE. The drugs used in this study were PTX (Sigma), D-APV (Cambridge Research Biochemicals), and CNQX (Tocris Neuramin).
J. J. HABLITZ
A ---iA-
IlOmV 20 ms
FIG. 1. Development of picrotoxin-induced epileptiform responses in a neocortical neuron. A: under control conditions, weak stimulation evokes an eEPSP followed by multiple, small-amplitude, late polysynaptic EPSPs. B: stronger stimulation produces an EPSP-IPSP complex. The IPSP is depolarizing because of the high RMP (-84 mV). C and D: responses evoked by same stimuli 5 min after starting bath application of 50 PM picrotoxin. ,Y and F: paroxysmal bursts are evoked after 10 min PTX exposure.
tial (IPSP) followed by a GABA,-receptor-mediated potassium-dependent hyperpolarization. Five minutes after starting bath application of 50 PM PTX, the eEPSP evoked by weak stimulation was virtually unaffected, whereas the 1EPSPwas considerably enhanced (Fig. 1C). Strong stimulation now evoked an action potential and a smaller afterdepolarization (Fig. 10). These changes are consistent with a partial block of GABA,-mediated inhibition. After further exposure to PTX, 1EPSPSwere further enhanced and, after a long latent period, a PDS was triggered (Fig. 1E). Increasing the stimulus strength evoked a PDS at a short latency (Fig. 1F). After PTX application, under the experimental conditions of this study, spontaneous PDS were rare (see also Gutnick et al. 1982). The waveform of the orthodromically evoked PDSs, although characteristic for a given neuron, varied between cells. With the use of duration asa criterion, RESULTS two categories of PDSs were identified. Long PDSs had durations (measured at half amplitude) > 100 ms (165 t 16 Induction of‘epileptiform activity by picvotoxin ms, n = 16), whereas short PDSs persisted for (50 ms (45 t Responsesof a layer II neuron to low- and high-intensity 5 ms, n = 7). Neurons in the two groups did not differ in R, stimulation of layer IV are shown in Fig. 1, A and B, respec- or resting membrane potential (RMP). Examples of long tively. Weak stimulation evoked an eEPSP followed by a and short PDSs are shown in Fig. 2, A and B, respectively. late depolarization composed of multiple polysynaptic Although the PDS is generally viewed as being “all or 1EPSPs.High intensity stimulation evoked an initial depo- none” in nature, the observed differences in PDS duration larization that was rapidly curtailed and a long-lasting depo- could be due to variability in the strength of the stimulus larization followed by a hyperpolarization. Previous studies used for activation. If this were the case, increasing the infrom this laboratory (Sutor and Hablitz 1989a; Weiss and tensity of stimulation would be expected to differentially Hablitz 1984) and others (Connors et al. 1988; Howe et al. affect the two types of responses. However, in agreement 1987) have shown that this potential consists of an EPSP with previous studies on cortical neurons (Gutnick et al. overlapped in time by a depolarizing GABA,-receptor-me1982) increasing the stimulation strength had little effect diated chloride-dependent inhibitory postsynaptic poten- on the duration of long- or short-duration PDSs (Fig. 2, A
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA
A
L
T
AND
NON-NMDA
RECEPTORS
AND
EPILEPTIFORM
89
ACTIVITY
r-:‘T
k2T
25mV
I I
1
1 OOms
B I
XIT
x2T
x3T
x4T
FIG. 2. Paroxysmal depolarizing shifts (PDSs) recorded in PTX-treated rat neocortical neurons. Epileptiform activity was evoked by threshold [ 1 times threshold (T)] and suprathreshold (2-4 times threshold) stimuli. A: example of a neuron displaying a long-duration PDS. Increasing stimulus strength produces a decrease in latency to onset of the PDS from 29 to 19 ms. Peak PDS amplitudes and durations were (from left to right) 63 mV, 120 ms; 6 1 mV, 124 ms; 6 1 mV, 120 ms; and 60 mV, 120 ms. RMP, -88 mV; R,, 16 MQ. B: short-duration PDSs recorded in another neuron. The most prominent effect of increasing stimulation strength was to decrease the time to onset of the PDS. Latencies were 13,4, 3, and 3 ms for stimuli of l-4 times threshold, respectively. The corresponding peak amplitudes and durations were 72 mV, 54 ms; 7 1 mV, 54 ms; 7 1 mV, 55 ms; and 71 mV, 56 ms. RMP, -85 mV; R,, 22 MQ.
and B, respectively). A decrease in the latency to onset was the only observable effect of increasing stimulus strength. Extracellular recordings obtained after loss of intracellular impalements indicated that the duration of the field potential corresponded closely to that of the intracellularly recorded potential. Efect of NMDA and non-NMDA neocortical PDSs
antagonists on
The effect of bath application of D-APV on evoked epileptiform activity was studied in nine neurons with RMPs and R,s of -80.3 t 1.9 mV and 20.6 t 1.1 M& respectively. D-APV (20 ,uM) had no effect on basic membrane properties but reduced both PDS duration and amplitude (Fig. 3A). Peak amplitude and duration were diminished by 29.8 t 7.7 and 15 t 3.5%, respectively. The time to onset and rate of rise of the PDS were not altered by D-APV (Fig. 3, A3 and B3). Short-duration PDSs were observed in only one of the neurons tested with D-APV; this neuron was similarly affected (Fig. 3B). CNQX is a non-NMDA receptor antagonist (Fletcher et al. 1988; Honore et al. 1988) that selectively and reversibly blocks EPSPs in neocortical neurons (Hablitz and Sutor 1990). CNQX had no apparent effect on RMP or R, (-85 t 1.4 mV and 20.5 t 1.5 M& respectively; n = 12). The effect of CNQX on evoked epileptiform discharges (5 PM), which took 30-50 min to fully develop, is shown in Fig. 4. The antagonist either reduced or abolished PDSs, depending on their initial duration. Long-duration PDSs (189.1 t- 1.8 ms, n = 7) were shortened by 56.9 t 8.6% (n =
7) with no appreciable change in peak amplitude (averaged reduction 1.3 t 4.7%, n = 7). The latency to onset of the PDS was increased and the rate of rise reduced. Short-duration PDSs (42.3 t 7.0 ms, n = 5) were virtually abolished by CNQX as shown in Fig. 4B2. These effects were reversible on washing (Fig. 4B4). Increasing orthodromic stimulation in the presence of CNQX could partially restore epileptiform burst responses. Figure 5B shows a neuron where, after 1 h, CNQX had blocked PDSs evoked by 2 times threshold stimulation leaving a small postsynaptic depolarization. Stimulation at 3 and 4 times threshold evoked progressively larger, multicomponent depolarizing responses, whereas at 5 times threshold a PDS-like response was observed. These results indicate that a CNQX-resistant component of synaptic transmission is present in PTX-treated slices. Figure 5C shows that PDS-like responses were obtained with lower stimulation 30 min after starting washout of CNQX, indicating partial reversal. Neurons were not studied long enough to demonstrate complete recovery from CNQX exposure. Sensitivity of CNQX-resistant antagonists
responses to NMDA
As described above, epileptiform discharges in PTXtreated slices were reduced, but not abolished, by the NMDA antagonist D-APV. We therefore tested its effect on the CNQX-insensitive component of the PDS (termed the residual PDS). As shown in Fig. 6, control responses (Fig. 6A) were reduced >.in the presence of CNQX. Addition of 20
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
W.-L. LEE AND J. J. HABLITZ
90
3
30ms
25mV
B
2
1
L 150ms
3
Control 25mV
L
40ms
FIG. 3. Effect of D-APV on long- and short-duration PDSs. A: examples of long-duration (186 ms) PDS before (1) and after (2) bath application of 20 PM D-APV. Records are shown superimposed in 3. D-APV reduced PDS duration by 26% and peak amplitude by 6%. RMP, -88 mV; R,, 22 MSt. B: similar experiment in a neuron showing a short-duration (5 1 ms) PDS. BI: control response before D-APV. B2: after 1 h in D-APV. Duration and peak amplitude were reduced by 20 and 24%, respectively. B3: superimposition of traces from 1 and 2. RMP, -8 1 mV; R,, 20 MS2.
D-APV rapidly abolished the residual PDS (Fig. 6C). This was true in all neurons tested (n = 4). Figure 6D shows partial recovery 11 min after removal of CNQX and DAPV. In a similar manner, CNQX readily abolished resid-
pM
ual PDSs in neurons initially exposed to D-APV (n = 3; not shown). Thus epileptiform burst responses were not observable when both NMDA and non-NMDA receptors were blocked.
A 1
2
3
CNQX 1 hr
L
1
20mvl100ms
3
CNQX
I
4
CNQX 1 hr
Wash
FIG. 4. Alterations of long- and short-duration PDSs by the non-NMDA receptor antagonist CNQX. A: long-duration ( 169 ms) PDS before (1) and after (2) bath application of 5 PM CNQX. Duration was reduced by 30% with no change in peak amplitude. After 1 h in CNQX (3) the duration was reduced by 76%, and the peak amplitude was increased by 6%. Note the increased latency to onset to PDS and slower initial rate of rise. RMP, -8 1 mV; R,, 16 Mfi. B: example of a short-duration (66 ms) PDS in another neuron. Bl: control response before CNQX. B2: after CNQX, duration was reduced by 30% and peak amplitude by 10%. Latency to PDS onset also increased. B3: after 1 h in CNQX, the PDS was abolished. B4: partial recovery of PDS after washing for 14 min. RMP, -87 mV; RN, 18 M&
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA
A
AND
NON-NMDA
RECEPTORS
AND
EPILEPTIFORM
ACTIVITY
91
x2T
x2T
CNQX 18min
Control
25mV c
L50ms
-7
x3T
x2T
x4T
1 hr CNQX
--
I
C 30
x2T
-7----x,3T
1
x4T
min
wash._
---
h
m
FIG. 5. Action of CNQX on short-duration PDSs is stimulus dependent. A: specimen records showing marked reduction in short-duration (42 ms) PDS by 5 PM CNQX. Stimulation intensity was 2 times threshold intensity needed to evoke a PDS under control conditions (x2T). B: after 1 h exposure to CNQX, stimulation at levels used during control evokes little postsynaptic response. Increases (3-5 times threshold) in stimulation elicit progressively larger responses until a PDS-like (i.e., spikes riding on an underlying depolarization) event appears. C: partial recovery from the effects of CNQX was observed after 30 min of wash. RMP, -83 mV; R,, 16 Mfi.
Voltage dependence of neocortical PDSs The sensitivity of residual PDSs to D-APV suggested the involvement of NMDA receptors. An additional test for
A
Control
B
CNQX
lhr
NMDA receptor involvement is the voltage-dependence of the response. Conventional synaptic responses (Sutor and Hablitz 1989a; Thomson 1986) and PDSs (Gutnick et al. 1982) in the neocortex increase in amplitude with hyperpo-
C
D
APV+CNQX
Wash
i-
25mV
i,
I 1 OOms
FIG. 6. Effect of the NMDA receptor antagonist D-APV on the CNQX-resistant component of the long-duration PDS. A: recording of long-duration (105 ms) epileptiform responses under control conditions. B: residual PDS evoked 1 h after starting CNQX. Note decrease in duration ( 12%) and increase in delay to onset of the epileptiform response. Peak amplitude was unaffected. C: 6 min exposure to 20 PM D-APV, in the continued presence of CNQX, completely suppressed postsynaptic activity. D: effectsof CNQX and D-APV were partially reversed on 11 min washing with antagonist-free saline. RMP, -93 mV; R,, 20 MQ.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
W.-L. LEE AND J. J. HABLITZ
92
larization and decrease with depolarization. Near the RMP, NMDA responses decrease in amplitude with both hyperpolarization and depolarization. Figure 7,A and B, shows examples of PDSs at a variety of membrane potentials before and after CNQX, respectively. A plot of peak PDS amplitude versus membrane potential is shown in Fig. 70. PDS amplitude was linearly related to membrane potential under both conditions. The extrapolated reversal potential for PDSs and residual PDSs (i.e., those recorded in presence of CNQX) were -33.8 t 7.3 (n = 6) and -32.2 t 5.5 mV (n = 6), respectively. The relatively negative values obtained for reversal potentials result from an inability to depolarize the membrane potential beyond the reversal potential and the necessity to rely on extrapolated versus interpolated measurements of the reversal point. Careful examination of records obtained during these experiments indicated the presence of an early depolarization preceding the full PDS (Fig. 7C). Because this potential seemed to trigger or initiate the PDS, we determined whether this response had a unique voltage dependence. In the absence of CNQX, complex voltage dependencies were observed such that depolarization produced decreases in amplitude in two neurons and increases in two others. This
A
indicates varying contributions from NMDA and nonNMDA receptors. In the presence of CNQX, early response amplitude increased with depolarization over the range - 110 to -56 mV. This was a consistent finding in all neurons tested (n = 6). A typical example is shown in Fig. 7,C2 and E. This increase in amplitude with depolarization can be attributed in part to activation of voltage-dependent currents, particularly a persistent sodium conductance (Flatman et al. 1986; Sutor and Hablitz 1989b) because R, did increase with depolarization in all the neurons tested (Fig. 7F). However, in the neuron shown, despite the increase in resistance, the small depolarization preceding the PDS in the absence of CNQX decreased rather than increased with depolarization (Fig. 7, CI and E). To more directly examine the voltage dependence of the early depolarization without contaminating influences from active membrane conductances, the lidocaine derivative QX-3 14 was used to block sodium currents (Connors and Prince 1982; Sutor and Hablitz 1989b). After intracellular injection of QX-3 14, neurons had relatively constant R,s over the range -90 to -60 mV (Fig. SE), and fast action potentials were not evoked. Examples of PDSs recorded in QX-3 14-injected neurons are shown in Fig. 8
Control -66mV
-1 OOmV
-76mV
-89mV
4 nl
’
-100
25mVL 1 OOms
@i
E
CNQX -98mV
-76mV
-65mV
s 2s N3 .A b al --” 8” 4 -U-xe- 0 c
L
.
.
_
’
.
-80’
.
.
-50 Vm (mV)
1
Cl
10
s-
:-‘::;- wo-
’ -100
--.
.
.
-60
Vm (mV>
C
1
2
I -7OmV
-69mV
F
501
-1OOmV -99mV
2-
25-
’
5
-80 Vm hV> FIG. 7. Voltage dependence of PDSs and early depolarizations in the presence and absence of CNQX. A: control PDSs recorded at 4 different membrane potentials. Note the increase in PDS amplitude at more hyperpolarized membrane potentials. B: residual PDSs after CNQX application. PDS amplitude still increased with hyperpolarization. C: expanded time base records of early depolarization that preceded the PDS. Superimposed records of early depolarizations obtained at 2 membrane potentials are shown. Cl: early depolarizations under control conditions. C2: in the presence of CNQX, the early depolarization was bigger at -69 than -99 mV. D: plot of peak PDS amplitude as a function of membrane potential (mV). Control PDS amplitude (0) decreased linearly with membrane depolarizations. Similar results were obtained in the presence of CNQX (0). Each point is the average of 4 records. E: plot of early depolarization amplitude vs. membrane potential. Under control conditions (o), early depolarizations were measured 8.3 ms after stimulus artifact. Early depolarizations in the presence of CNQX (0) were measured 34.2 ms after the stimulus artifact. Note the gradual increase in amplitude as the membrane potential was depolarized from - 100 to -62 mV. F: R, as function of membrane potential. RMP, -89 mV.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA
AND
NON-NMDA
RECEPTORS
AND
EPILEPTIFORM
ACTIVITY
OL--1
-120
c
25mV
L
.
-80
1
1
-40
0
Vm (mV>
1 OOms
-8OmV
2mV(oms
B /
/
-4gmV
-68mV
FIG. 8. Voltage dependence of CNQX-resistant PDSs and early depolarizations in a QX-3 14-injected rat neocortical neuron. A: specimen records of epileptiform responses obtained at different membrane potentials. Fast Na-dependent action potentials were abolished. Remaining regenerative responses are attributable to calcium electrogenesis or sodium conductances in remote regions of the neuron. B: superimposed records of early depolarizations. Bl: selected responses obtained at membrane potentials where R, was constant. The early depolarization was largest at -68 mV and became progressively smaller at -80 and -92 mV. B2: early depolarizations obtained at the extremes of membrane polarization. Response was largest at -49 mV despite a decrease in R,. C: plot of PDS amplitude as a function of membrane potential. PDS amplitude decreased linearly between -86 and -49 mV, and the extrapolated reversal potential was 7.5 mV. D: early depolarization amplitude plotted vs. membrane potential. Note the gradual increase in its amplitude with membrane depolarization. E: plot of R, vs. membrane potential. Note the relatively constant R, at potentials between -92 and -68 mV. RMP, -80 mV.
and resemble responses previously described by Gutnick et al. ( 1982). In QX-3 14-injected cells, the amplitude of residual PDSs recorded in the presence of CNQX decreased with depolarization (Fig. 8, A and C). The extrapolated reversal potential was 9.8 t 5.4 mV (n = 4). Simultaneously measured early depolarizations increased in amplitude with depolarization. Over a region where R, remained constant, early response amplitude increased as the membrane potential was depolarized. The amplitude of the early depolarization continue to increase with depolarization beyond -60 mV despite a progressive decrease in RN (Fig. 8, B2, D, and E), suggesting that the increase in early depolarization amplitude was not simply due to changes in passive properties. DISCUSSION
The results of the present series of experiments suggest that reductions of inhibition in the neocortex produce synchronized epileptiform discharges by enhancing late polysynaptic EPSPs. Fully developed paroxysmal discharges were grouped into two classes on the basis of duration. In each case, an early depolarizing event was seen preceding
the PDS and appeared responsible for initiation of paroxysmal activity. Epileptiform discharges were reduced by antagonists of both the NMDA and non-NMDA types of excitatory amino acid receptors, suggesting that, as in the hippocampus (Lee and Hablitz 1989a), both receptor types contribute to generation of burst discharges. Generation
of PDSs
PDSs recorded here were similar to those described previously in studies of disinhibited neocortex in vivo (Matsumoto and Ajmone-Marsan 1964; Matsumoto et al. 1969) and in vitro (Gutnick et al. 1982; Thomson and West 1986). However, in the present study, based on duration, there appeared to be two types of PDSs. During our studies of excitatory amino acid antagonists, generally only one neuron was studied in a given preparation. Short- and longduration PDSs were therefore seen in different slices, usually from different preparations. This raises the possibility that variations in the orientation of the brain during slice preparation, with variable preservation of intracortical connections, contributed to the observed differences. Shortduration PDSs were not a sign of poor slice viability be-
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
94
W.-L.
LEE
AND
J. J. HABLITZ
cause such responses were stable over time, evoked at cqmparable stimulus intensities and observed in neurons whose properties were representative of the whole population. Slice orientation has been shown to affect variables such as dye coupling (Gutnick et al. 1985), and further investigation of this factor may yield information concerning the intracortical connections regulating the degree of excitability seen during epileptiform discharges. The greater sensitivity of short-duration PDSs to blockade by CNQX suggests that a different population of cells or fibers were activated in these preparations. Considerable evidence has accumulated indicating that a common mechanism of action of many convulsant drugs is to decrease GABA-mediated, chloride-dependent inhibition (Macdonald 1984). PDSs generated in this disinhibition model of epilepsy have increased our understanding of the mechanisms underlying paroxysmal bursts (Hablitz 1984; Wong et al. 1984). Previous studies have shown that the neocortex is particularly sensitive to reductions in inhibition, demonstrating seizure activity when GABA-mediated responses in the brain stem and vestibular nuclei were barely affected (Davenport et al. 1979; Tribble et al. 1983). Under normal conditions, inhibition in the neocortex controls the appearance of a late polysynaptic EPSP in -layer II/III neurons (Sutor & Hablitz 1989a) and can mask the expression of long-term potentiation (Sutor and Hablitz 1989~). Similarly, it has been shown that partial suppression of GABA,-receptor-mediated inhibition expands the horizontal or tangential spread of activity in the cortex (Chagnac-Amitai and Connors 1989a). Thus the synaptic mechanism whereby decreases in GABAergic inhibition results in the generation of synchronous epileptiform discharges appears to be a loss of inhibitory control over local excitatory circuits within the neocortex. A small EPSP-like depolarization has been previously observed to precede PDS onset in the neocortex (see also Gutnick et al. 1982; Thomson and West 1986). However, no attempt was made to examine the properties of this event. We have shown that this initial depolarization is synaptic in origin. Under control conditions, the effects of alterations in membrane potential were variable with both increases and decreases in amplitude seen on depolarization. When CNQX was used to block non-NMDA receptors, the early depolarization demonstrated a non-conventional NMDA receptor type voltage dependence. Similar nonlinear behavior of EPSPs has been described in the hippocampus (Dingledine et al. 1986) and suggests that, under disinhibited conditions, an NMDA-mediated synaptic potential can be evoked. This event may then serve as a trigger for initiation of the actual PDS, which has both NMDA and non-NMDA components. Directly evoked NMDA-mediated synaptic potentials have been seen in disinhibited hippocampal slices after blockade of non-NMDA receptors (Lee and Hablitz 1990), and similar potentials appear to underlie seizures originating from midbrain structures (Pierson et al. 1989). Such NMDA-mediated depolarizations may thus serve as triggers for epileptiform activity in many brain structures.
acid receptors of both the NMDA and non-NMDA type were blocked. Selective blockade of individual receptor types indicated that NMDA receptor blockade affected both the amplitude and duration of the PDS, whereas antagonism of non-NMDA receptors influenced the time of onset and duration of PDSs but not their amplitude. Previous studies of the effect of NMDA antagonists on epileptiform activity in the neocortex (Thomson and West 1986) and hippocampus (Dingledine et al. 1986) have shown PDSs to be reduced in amplitude and duration. In the present study, measured from the RMP, PDSs had amplitudes of -60 mV, with the peak of the PDS corresponding to a membrane potential near -20 mV. This is a region of membrane potential where the voltage-dependent block by Mg2+ is minimal (Nowak et al. 1984), and a contribution from NMDA receptors would be prominent and a decrease in amplitude expected after antagonist exposure. Previous studies on in vitro slices of hippocampus (Collingridge et al. 1988), neocortex (Jones and Baughman 1988), and hippocampal cells in culture (Forsythe and Westbrook 1988) have suggested that NMDA-mediated synaptic potentials are long in duration. Blockade of such a long-duration component would account for the decrease in PDS duration produced by D-APV. At 5 PM, CNQX is a quite specific and effective antagonist of non-NMDA receptors (Fletcher et al. 1988; Hablitz and Sutor 1990); therefore responses observed in the presence of agent are likely to be governed predominantly by NMDA receptor activation. The slower rising phase of the PDS in the presence of CNQX and the concomitant increase in latency to onset may reflect less synchronous activation of NMDA receptors and poorer recruitment of local excitatory pathways. Despite the presence of a D-APV-sensitive, i.e., NMDAmediated, component to PDSs and residual PDSs recorded in the presence of CNQX, plots of response amplitude versus membrane potential were always linear, indicating a reversal potential near 0 mV. This linear behavior is consistent with previous studies of the voltage dependence of epileptiform activity (Gutnick et al. 1982; Johnston and Brown 198 1). The lack of an unconventional NMDA-like voltage dependence suggests that other conductances, both synaptic and voltage dependent, are of a sufficient magnitude to mask the voltage-dependent behavior of the NMDA channels. A similar high conductance state has been seen in response to iontophoretically applied NMDA in both hippocampus (Hablitz 1982) and neocortex (Flatman et al. 1983; Sutor and Hablitz 1989b). Thus the lack ofa NMDAlike voltage dependence is not a sufficient criteria for ruling out contributions from NMDA receptors.
Efects of excitatory amino acid antagonists Another principal finding of this study was that epileptiform discharges were not observed when excitatory amino
CHAGNAC-AMITAI, Y. AND CONNORS, B. W. Horizontal spread of synchronized activity in neocortex and its control by GABA-mediated inhibition. J. Neurophysiol. 61: 747-758, 1989a. CHAGNAC-AMITAI, Y. AND CONNORS, B. W. Synchronized excitation and
This work was supported by National Institute of Neurological Disorders and Stroke Grants NS- 18 145 and NS-22373. Address for reprint requests: J. J. Hablitz, Neurobiology Research Center, University of Alabama at Birmingham, Birmingham, AL 35294. Received
13 April
1990; accepted
in final
form
6 September
1990.
REFERENCES
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA inhibition
driven
by intrinsically
bursting
AND
NON-NMDA
neurons
in neocortex.
RECEPTORS J. Neu-
rophysiol. 62: 1149- 1162, 1989b. COLLINGRIDGE, G. L., HERRON, C. E., AND LESTER, R. J. Synaptic activation of N-methyl-D-aspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus. J. Physiol. Lond. 399: 283-300, 1988. CONNORS, B. W. Initiation of synchronized bursting in neocortex. Nature Lond. 310: 685-687, 1984. CONNORS, B. W., MALENKA, R. C., AND SILVA, L. R. Two inhibitory postsynaptic potentials, and GABA, receptor-mediated responses in neocortex of rat and cat. J. Physiol. Lond. 406: 443-468, 1988. CONNORS, B. W. AND PRINCE, D. A. Effects of local anesthetic QX-3 14 on the membrane properties of hippocampal pyramidal neurons. J. Pharmacol. Exp. Ther. 220: 476-48 1, 1982. DAVENPORT, J., SCHWINDT, P. C., AND CRILL, W. E. Epileptogenic doses of penicillin do not reduce a monosynaptic GABA-mediated postsynaptic inhibition in the intact anesthetized cat. Exp. Neural. 65: 552-572, 1979. DINGLEDINE, R., HYNES, M. A., AND KING, G. L. Involvement of Nmethyl-D-aspartate receptor in epileptiform bursting in the rat hippocampal slice. J. Physiol. Land. 380: 175-l 89, 1986. FLATMAN, J. A., SCHWINDT, P. C., AND CRILL, W. E. Multiple actions of N-methyl-D-aspartate on cat neocortical neurons in vitro. Brain Res. 266: 169-173,1983. FLATMAN, J. A., SCHWINDT, P. C., AND CRILL, W. E. The induction and modification of voltage-sensitive responses in cat neocortical neurons by iV-methyl-D-aspartate. Brain Res. 363: 62-77, 1986. FLETCHER, E. J., MARTIN, D., ARAM, J. A., AND HONORE, T. Quinoxalinediones selectively block quisqualate and kainate receptors and synaptic events in rat neocortex and hippocampus and frog spinal cord in vitro. Br. J. Pharmacol. 95: 585-597, 1988. FORSYTHE, I. D. AND WESTBROOK, G. L. Slow excitatory postsynaptic currents mediated by N-methyl-D-aspartate receptors on cultured mouse central neurons. J. Physiol. Lond. 396: 5 15-533, 1988. GUTNICK, M. J., CONNORS, B. W., AND PRINCE, D. A. Mechanisms of neocortical epileptogenesis in vitro. J. Neurophysiol. 48: 132 I- 1335, 1982. GUTNICK, M. J., LOBEL-YAAKOV, R., AND RIMON, G. Incidence of neuronal dye-coupling in neocortical slices depends on the plane of section. Neuroscience 15: 659-666, 1985. HABLITZ, J. J. Conductance changes induced by DL-homocysteic acid and N-methyl-DL-aspartic acid in hippocampal neurons. Brain Res. 247: 149-153,1982. HABLITZ, J. J. Picrotoxin-induced epileptiform activity in hippocampus: role of endogenous versus synaptic factors. J. Neurophysiol. 5 1: 10 1 l1027, 1984. HABLITZ, J. J. AND SUTOR, B. Excitatory postsynaptic potentials in rat neocortical neurons in vitro. III. Effects of a quinoxalinedione nonNMDA receptor antagonist. J. Neurophysiol. 64: 1282- 1290, 1990. HONORE, T., DAVIES, S. N., DREJER, J., FLETCHER, E. J., JACOBSEN, P., LODGE, D., AND NIELSEN, F. E. Quinoxalinedione: potent competitive non-NMDA glutamate receptor antagonists. Science Wash. DC 24 1: 701-703, 1988. HOWE, J. R., SUTOR, B., AND ZIEGLGANSBERGER, W. Characteristics of long-duration inhibitory postsynaptic potentials in rat neocortical neurons in vitro. Cell Mol. Neurobiol. 7: I- 18, 1987. JOHNSTON, D. AND BROWN, T. H. Giant synaptic potential hypothesis for epileptiform activity. Science Wash. DC 2 11: 294-297, 198 1.
AND
EPILEPTIFORM
ACTIVITY
95
JONES, K. A. AND BAUGHMAN, R. W. NMDAand non-NMDA-receptor components of excitatory synaptic potentials recorded from cells in layer V of rat visual cortex. J. Neurosci. 8: 3522-3534, 1988. LEE, W. L. AND HABLITZ, J. J. Involvement of non-NMDA receptors in picrotoxin-induced epileptiform activity in the hippocampus. Neurosci. Lett. 107: 129-134, 1989a. LEE, W. L. AND HABLITZ, J. J. Role of NMDA and non-NMDA receptors in picrotoxin-induced epileptiform activity in rat neocortex. Sot. Neurosci. Abstr. 15: 1213, 1989b. MACDONALD, R. L. Anticonvulsant and convulsant drug actions on vertebrate neurones in primary dissociated cell culture. In: Electrophysiology OfEpilepsy, edited by P. A. Schwartzkroin and H. Wheal. London: Academic, 1984, p. 353-388. MATSUMOTO, M. AND AJMONE-MARSAN, C. Cortical cellular phenomena in experimental epilepsy: interictal manifestations. Exp. Neural. 9: 286326, 1964. MATSUMOTO, M., AYALA, G. F., AND GUMNIT, R. J. Neuronal behavior and triggering mechanism in cortical epileptic focus. J. Neurophysiol. 32:688-703, 1969. MAYER, M. L., WESTBROOK, G. L., AND GUTHRIE, P. B. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature Lond. 309: 26 l-263, 1984. NOWAK, L., BREGESTOVSKI, P., ASCHER, P., HERBET, A., AND PROCHIANTZ, A. Magnesium gates glutamate-activated channels in mouse central neurones. Nature Lond. 307: 462-465, 1984. PIERSON, M. G., SMITH, K. L., AND SWANN, J. S. A slow NMDA-mediated synaptic potential underlies seizures originating from midbrain. Brain Res.486: 381-386, 1989. STREIT, P. Glutamate and aspartate as transmitter candidates for system of the cerebral cortex. In: Cerebral Cortex. Functional Properties ofcortical Cells, edited by E. J. Jones and A. Peters. New York: Plenum, 1984, vol. 2, p. 119-143. SUTOR, B. AND HABLITZ, J. J. EPSPs in rat neocortical neurons in vitro. I. Electrophysiological evidence for two distinct EPSPs. J. Neurophysiol. 61.607-620, 1989a. SUTOR, B. AND HABLITZ, J. J. EPSPs in rat neocortical neurons in vitro. II. Involvement of N-methyl-r>-aspartate receptors in generation of EPSPs. J. Neurophysiol. 6 1: 62 l-634, 1989b. SUTOR, B. AND HABLITZ, J. J. Long-term potentiation in frontal cortex: role of NMDA-modulated polysynaptic excitatory pathways. Neurosci. Lett. 97: 11 l-l 17, 1989~. THOMSON, A. M. A magnesium-sensitive post-synaptic potential in rat cerebral cortex resembles neuronal responses to N-methylaspartate. J. Physiol. Lond. 370: 53 l-549, 1986. THOMSON, A. M. AND WEST, D. C. N-methylaspartate receptors mediate epileptiform activity evoked in some, but not all, conditions in rat neocortical slices. Neuroscience 19: 116 1- 1177, 1986. TRIBBLE, G. L., SCHWINDT, P. C., AND CRILL, W. E. Reduction of postsynaptic inhibition tolerated before seizure initiation: brain stem. Exp. Neural. 80: 304-320, 1983. WEISS, D. S. AND HABLITZ, J. J. Interaction of penicillin and pentobarbital with inhibitory synaptic mechanisms in neocortex. Cell. Idol. Neuro-
biol. 4: 301-317, 1984. WONG, R. K. S., TRAUB, R. D., AND MILES, R. Epileptogenic mechanisms as revealed by studies of the hippocampal slice. In: Electrophysiology of Epilepsy, edited by P. A. Schwartzkroin and H. Wheal. London: Academic, 1984, p. 253-276.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
in U. X,4.
Initiation of Epileptiform Activity by Excitatory Amino Acid Receptors in the Disinhibited Rat Neocortex WAI-LING LEE AND JOHN J. HABLITZ Neurobiology Research Center, Department of Physiology at Birmingham, Birmingham, Alabama 35294 SUMMARY
AND
CONCLUSIONS
1. Intracellular recordings were obtained from neurons in layer II-III of rat frontal cortex maintained in vitro. The role of excitatory amino acid receptors in generation of picrotoxin (PTX)-induced epileptiform activity was investigated with the use of D-2amino+phosphonovaleric acid (D-APV) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) as selective antagonists of N-methyl-D-aspartate (NMDA) and non-NMDA receptors, respectively. 2. Bath application of PTX resulted in a decrease in evoked inhibitory postsynaptic potentials (IPSPs) in neocortical neurons and a concomitant increase in a polysynaptic late excitatory postsynaptic potential (1EPSP). Epileptiform burst responses, termed paroxysmal depolarizing shifts (PDSs), subsequently developed. Based on response duration, two types of PDSs were identified. Long PDSs were > 100 ms in duration, whereas short PDSs lasted ~50 ms. An early depolarizing potential preceded both types of epileptiform burst response. 3. The NMDA receptor antagonist D-APV reduced the peak amplitude and duration of the PDS. D-APV-insensitive portions of the PDS were greatly attenuated or abolished by CNQX. The non-NMDA antagonist also increased the latency to PDS onset and reduced its duration without affecting peak amplitude. CNQX-insensitive components of the PDS, when present, were abolished by D-APV. 4. Short-duration PDSs could be blocked by CNQX. In these neurons, increasing the stimulation strength produced epileptiform responses of reduced amplitude. 5. Under control conditions, PDS amplitude was a linear function of membrane potential, increasing with hyperpolarization and diminishing on depolarization. The early depolarizing response preceding the PDS showed a variable dependence on voltage. Both decreases and increases in amplitude were observed, in different neurons, on membrane depolarization. 6. In the presence of CNQX, PDS amplitude remained linearly related to membrane potential. Early depolarizations preceding the PDS displayed an unconventional voltage sensitivity, increasing with depolarization over the range - 100 to -60 mV. 7. We conclude that, in the PTX disinhibition model of epilepsy, both NMDA and non-NMDA receptors contribute to epileptiform burst responses. An NMDA-receptor-mediated early depolarization may serve as a trigger for burst responses. NMDA receptors are also involved in maintenance of the peak of the paroxysmal depolarizations, whereas non-NMDA receptors are necessary for full expression of the PDS.
INTRODUCTION
Intracellular recordings from neocortical neurons in vivo (Matsumoto et al. 1969) and in vitro (Gutnick et al. 1982)
and Biophysics,
University of Alabama
have shown that interictal spike discharges in acute epileptogenic foci are accompanied by action-potential discharges riding on a large membrane depolarization [originally termed a paroxysmal depolarizing shift (PDS) by Matsumoto and Ajmone-Marsan ( 1964)]. Synaptic events have been suggested to play an important role in elaboration of neocortical epileptiform activity (Gutnick et al. 1982; Matsumoto et al. 1969) with a subpopulation of neurons in layer IV and upper layer V possibly responsible for initiation of paroxysmal events (Chagnac-Amitai and Connors 1989a,b; Connors 1984). The changes in normal synaptic activity responsible for generation of synchronous epileptiform discharges have not been described in detail in the neocortex. Moreover, although excitatory amino acid neurotransmission has been extensively investigated in the neocortex (Sutor and Hablitz 1989a,b; Thomson 1986), the role of specific receptor subtypes in initiation of the PDS and maintenance of the peak depolarization has not been determined. In the rat neocortex, the excitatory amino acids glutamate and aspartate are prominent neurotransmitter candidates (Streit 1984), acting on both N-methyl-D-aspartate (NMDA) and non-NMDA receptors. Application of the NMDA receptor antagonist D-2-amino-5-phosphonovaleric acid (D-APV) blocks a late excitatory postsynaptic potential (IEPSP) produced by activation of layer IV (Sutor and Hablitz 1989a,b) and late synaptic responses evoked by white matter stimulation (Thomson 1986). Early EPSPs (eEPSPs) are selectively blocked by the non-NMDA receptor antagonist 6 - cyano - 7 - nitroquinoxaline - 2,3 - dione (CNQX) (Hablitz and Sutor 1990). Studies of picrotoxininduced epileptiform activity in the hippocampus have shown bath application of 5 PM CNQX to significantly reduce or abolish evoked PDSs in both CA1 and CA3 neurons. In cells where a CNQX-insensitive component in the PDS was manifest, this remaining activity was abolished by the NMDA receptor antagonist D-APV (20 PM), suggesting the existence of an underlying NMDA-mediated synaptic potential (Lee and Hablitz 1989a). The role of the two receptor types in epileptiform activity in the neocortex is not well established. The present study sought to determine the contribution each receptor type made to the triggering or initiation of PDSs as opposed to the full expression or maintenance of the PDS waveform. In addition, we tested whether development of synchronized burst responses after application of the GABA,-receptor antagonist picrotoxin (PTX) resulted from a loss of inhibitory control of polysynaptic EPSPs in the neocortex.
0022-3077/9 1 $1 SO Copyright 0 199 1 The American Physiological Society Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
87
88
W.-L.
LEE
AND
A preliminary account of some of these findings has appeared (Lee and Hablitz 1989b). METHODS
Neocortical brain slices were obtained from adult SpragueDawley rats of both sexes. The preparation and maintenance of these slices employed methods described previously (Sutor and Hablitz 1989a). Briefly, rats were decapitated under ether anesthesia; the brains were quickly dissected out and placed in ice cold saline for 30-60 s. Coronal slices of frontal cortex (500 ,ccrnthick) were cut with a McIlwain tissue chopper and transferred to an interface type recording chamber. After 30 min at room temperature, slices were slowly warmed to the recording temperature of 34 t 1 “C and allowed to equilibrate at least 1.5 h before recording. The slices were continuously perfused with an oxygenated (95% O,-5% COJ saline consisting of the following (in mM): 124 NaCl, 5 MCI, 1.25 NaH,PO,, 2.0 CaCI,, 2.0 MgCl,, 26 NaHCO,, and 10 dextrose. Epileptiform activity was induced by addition of PTX (50 PM) to the perfusate. Intracellular recording electrodes were pulled from filamentcontaining borosilicate glass tubings (1.5 mm OD) and had resistances of 50- 100 ML! when filled with 4 M potassium acetate. Recordings were made with the use of a single-electrode timeshare circuit as described previously (Sutor and Hablitz 1989a). Bipolar stimulating electrodes were placed in layer IV - 1 mm lateral to the recording electrode. The strength of the orthodromic stimulation was set at 2 times threshold intensity needed for evoking a PDS. After impalement, the resting membrane potential was allowed to stabilize for a minimum of 30 min. Orthodromic stimulation was then applied once per minute for 10 min. After this baseline period, excitatory amino acid antagonists were bath applied while orthodromic stimulation continued to be applied. Antagonist effects were assessed after 1 h of drug perfusion with each neuron serving as its own control. Because of the long duration of the studies, only one cell was examined in each preparation. Neuronal input resistance (RN) and resting membrane potential were continuously monitored before and after antagonist application. This was accomplished by passing hyperpolarizing current pulses of -0.5 nA through the recording electrode and observing the resulting voltage deflection. To assess the effect of changes in membrane potential on the PDS amplitude, long-duration (3 s) current pulses were given with orthodromic stimulation applied 500 ms after pulse onset. Data were recorded on a Neurocorder DR-384 and analyzed with a PDP 1 l/23+ computer. Data are presented as mean t SE. The drugs used in this study were PTX (Sigma), D-APV (Cambridge Research Biochemicals), and CNQX (Tocris Neuramin).
J. J. HABLITZ
A ---iA-
IlOmV 20 ms
FIG. 1. Development of picrotoxin-induced epileptiform responses in a neocortical neuron. A: under control conditions, weak stimulation evokes an eEPSP followed by multiple, small-amplitude, late polysynaptic EPSPs. B: stronger stimulation produces an EPSP-IPSP complex. The IPSP is depolarizing because of the high RMP (-84 mV). C and D: responses evoked by same stimuli 5 min after starting bath application of 50 PM picrotoxin. ,Y and F: paroxysmal bursts are evoked after 10 min PTX exposure.
tial (IPSP) followed by a GABA,-receptor-mediated potassium-dependent hyperpolarization. Five minutes after starting bath application of 50 PM PTX, the eEPSP evoked by weak stimulation was virtually unaffected, whereas the 1EPSPwas considerably enhanced (Fig. 1C). Strong stimulation now evoked an action potential and a smaller afterdepolarization (Fig. 10). These changes are consistent with a partial block of GABA,-mediated inhibition. After further exposure to PTX, 1EPSPSwere further enhanced and, after a long latent period, a PDS was triggered (Fig. 1E). Increasing the stimulus strength evoked a PDS at a short latency (Fig. 1F). After PTX application, under the experimental conditions of this study, spontaneous PDS were rare (see also Gutnick et al. 1982). The waveform of the orthodromically evoked PDSs, although characteristic for a given neuron, varied between cells. With the use of duration asa criterion, RESULTS two categories of PDSs were identified. Long PDSs had durations (measured at half amplitude) > 100 ms (165 t 16 Induction of‘epileptiform activity by picvotoxin ms, n = 16), whereas short PDSs persisted for (50 ms (45 t Responsesof a layer II neuron to low- and high-intensity 5 ms, n = 7). Neurons in the two groups did not differ in R, stimulation of layer IV are shown in Fig. 1, A and B, respec- or resting membrane potential (RMP). Examples of long tively. Weak stimulation evoked an eEPSP followed by a and short PDSs are shown in Fig. 2, A and B, respectively. late depolarization composed of multiple polysynaptic Although the PDS is generally viewed as being “all or 1EPSPs.High intensity stimulation evoked an initial depo- none” in nature, the observed differences in PDS duration larization that was rapidly curtailed and a long-lasting depo- could be due to variability in the strength of the stimulus larization followed by a hyperpolarization. Previous studies used for activation. If this were the case, increasing the infrom this laboratory (Sutor and Hablitz 1989a; Weiss and tensity of stimulation would be expected to differentially Hablitz 1984) and others (Connors et al. 1988; Howe et al. affect the two types of responses. However, in agreement 1987) have shown that this potential consists of an EPSP with previous studies on cortical neurons (Gutnick et al. overlapped in time by a depolarizing GABA,-receptor-me1982) increasing the stimulation strength had little effect diated chloride-dependent inhibitory postsynaptic poten- on the duration of long- or short-duration PDSs (Fig. 2, A
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA
A
L
T
AND
NON-NMDA
RECEPTORS
AND
EPILEPTIFORM
89
ACTIVITY
r-:‘T
k2T
25mV
I I
1
1 OOms
B I
XIT
x2T
x3T
x4T
FIG. 2. Paroxysmal depolarizing shifts (PDSs) recorded in PTX-treated rat neocortical neurons. Epileptiform activity was evoked by threshold [ 1 times threshold (T)] and suprathreshold (2-4 times threshold) stimuli. A: example of a neuron displaying a long-duration PDS. Increasing stimulus strength produces a decrease in latency to onset of the PDS from 29 to 19 ms. Peak PDS amplitudes and durations were (from left to right) 63 mV, 120 ms; 6 1 mV, 124 ms; 6 1 mV, 120 ms; and 60 mV, 120 ms. RMP, -88 mV; R,, 16 MQ. B: short-duration PDSs recorded in another neuron. The most prominent effect of increasing stimulation strength was to decrease the time to onset of the PDS. Latencies were 13,4, 3, and 3 ms for stimuli of l-4 times threshold, respectively. The corresponding peak amplitudes and durations were 72 mV, 54 ms; 7 1 mV, 54 ms; 7 1 mV, 55 ms; and 71 mV, 56 ms. RMP, -85 mV; R,, 22 MQ.
and B, respectively). A decrease in the latency to onset was the only observable effect of increasing stimulus strength. Extracellular recordings obtained after loss of intracellular impalements indicated that the duration of the field potential corresponded closely to that of the intracellularly recorded potential. Efect of NMDA and non-NMDA neocortical PDSs
antagonists on
The effect of bath application of D-APV on evoked epileptiform activity was studied in nine neurons with RMPs and R,s of -80.3 t 1.9 mV and 20.6 t 1.1 M& respectively. D-APV (20 ,uM) had no effect on basic membrane properties but reduced both PDS duration and amplitude (Fig. 3A). Peak amplitude and duration were diminished by 29.8 t 7.7 and 15 t 3.5%, respectively. The time to onset and rate of rise of the PDS were not altered by D-APV (Fig. 3, A3 and B3). Short-duration PDSs were observed in only one of the neurons tested with D-APV; this neuron was similarly affected (Fig. 3B). CNQX is a non-NMDA receptor antagonist (Fletcher et al. 1988; Honore et al. 1988) that selectively and reversibly blocks EPSPs in neocortical neurons (Hablitz and Sutor 1990). CNQX had no apparent effect on RMP or R, (-85 t 1.4 mV and 20.5 t 1.5 M& respectively; n = 12). The effect of CNQX on evoked epileptiform discharges (5 PM), which took 30-50 min to fully develop, is shown in Fig. 4. The antagonist either reduced or abolished PDSs, depending on their initial duration. Long-duration PDSs (189.1 t- 1.8 ms, n = 7) were shortened by 56.9 t 8.6% (n =
7) with no appreciable change in peak amplitude (averaged reduction 1.3 t 4.7%, n = 7). The latency to onset of the PDS was increased and the rate of rise reduced. Short-duration PDSs (42.3 t 7.0 ms, n = 5) were virtually abolished by CNQX as shown in Fig. 4B2. These effects were reversible on washing (Fig. 4B4). Increasing orthodromic stimulation in the presence of CNQX could partially restore epileptiform burst responses. Figure 5B shows a neuron where, after 1 h, CNQX had blocked PDSs evoked by 2 times threshold stimulation leaving a small postsynaptic depolarization. Stimulation at 3 and 4 times threshold evoked progressively larger, multicomponent depolarizing responses, whereas at 5 times threshold a PDS-like response was observed. These results indicate that a CNQX-resistant component of synaptic transmission is present in PTX-treated slices. Figure 5C shows that PDS-like responses were obtained with lower stimulation 30 min after starting washout of CNQX, indicating partial reversal. Neurons were not studied long enough to demonstrate complete recovery from CNQX exposure. Sensitivity of CNQX-resistant antagonists
responses to NMDA
As described above, epileptiform discharges in PTXtreated slices were reduced, but not abolished, by the NMDA antagonist D-APV. We therefore tested its effect on the CNQX-insensitive component of the PDS (termed the residual PDS). As shown in Fig. 6, control responses (Fig. 6A) were reduced >.in the presence of CNQX. Addition of 20
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
W.-L. LEE AND J. J. HABLITZ
90
3
30ms
25mV
B
2
1
L 150ms
3
Control 25mV
L
40ms
FIG. 3. Effect of D-APV on long- and short-duration PDSs. A: examples of long-duration (186 ms) PDS before (1) and after (2) bath application of 20 PM D-APV. Records are shown superimposed in 3. D-APV reduced PDS duration by 26% and peak amplitude by 6%. RMP, -88 mV; R,, 22 MSt. B: similar experiment in a neuron showing a short-duration (5 1 ms) PDS. BI: control response before D-APV. B2: after 1 h in D-APV. Duration and peak amplitude were reduced by 20 and 24%, respectively. B3: superimposition of traces from 1 and 2. RMP, -8 1 mV; R,, 20 MS2.
D-APV rapidly abolished the residual PDS (Fig. 6C). This was true in all neurons tested (n = 4). Figure 6D shows partial recovery 11 min after removal of CNQX and DAPV. In a similar manner, CNQX readily abolished resid-
pM
ual PDSs in neurons initially exposed to D-APV (n = 3; not shown). Thus epileptiform burst responses were not observable when both NMDA and non-NMDA receptors were blocked.
A 1
2
3
CNQX 1 hr
L
1
20mvl100ms
3
CNQX
I
4
CNQX 1 hr
Wash
FIG. 4. Alterations of long- and short-duration PDSs by the non-NMDA receptor antagonist CNQX. A: long-duration ( 169 ms) PDS before (1) and after (2) bath application of 5 PM CNQX. Duration was reduced by 30% with no change in peak amplitude. After 1 h in CNQX (3) the duration was reduced by 76%, and the peak amplitude was increased by 6%. Note the increased latency to onset to PDS and slower initial rate of rise. RMP, -8 1 mV; R,, 16 Mfi. B: example of a short-duration (66 ms) PDS in another neuron. Bl: control response before CNQX. B2: after CNQX, duration was reduced by 30% and peak amplitude by 10%. Latency to PDS onset also increased. B3: after 1 h in CNQX, the PDS was abolished. B4: partial recovery of PDS after washing for 14 min. RMP, -87 mV; RN, 18 M&
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA
A
AND
NON-NMDA
RECEPTORS
AND
EPILEPTIFORM
ACTIVITY
91
x2T
x2T
CNQX 18min
Control
25mV c
L50ms
-7
x3T
x2T
x4T
1 hr CNQX
--
I
C 30
x2T
-7----x,3T
1
x4T
min
wash._
---
h
m
FIG. 5. Action of CNQX on short-duration PDSs is stimulus dependent. A: specimen records showing marked reduction in short-duration (42 ms) PDS by 5 PM CNQX. Stimulation intensity was 2 times threshold intensity needed to evoke a PDS under control conditions (x2T). B: after 1 h exposure to CNQX, stimulation at levels used during control evokes little postsynaptic response. Increases (3-5 times threshold) in stimulation elicit progressively larger responses until a PDS-like (i.e., spikes riding on an underlying depolarization) event appears. C: partial recovery from the effects of CNQX was observed after 30 min of wash. RMP, -83 mV; R,, 16 Mfi.
Voltage dependence of neocortical PDSs The sensitivity of residual PDSs to D-APV suggested the involvement of NMDA receptors. An additional test for
A
Control
B
CNQX
lhr
NMDA receptor involvement is the voltage-dependence of the response. Conventional synaptic responses (Sutor and Hablitz 1989a; Thomson 1986) and PDSs (Gutnick et al. 1982) in the neocortex increase in amplitude with hyperpo-
C
D
APV+CNQX
Wash
i-
25mV
i,
I 1 OOms
FIG. 6. Effect of the NMDA receptor antagonist D-APV on the CNQX-resistant component of the long-duration PDS. A: recording of long-duration (105 ms) epileptiform responses under control conditions. B: residual PDS evoked 1 h after starting CNQX. Note decrease in duration ( 12%) and increase in delay to onset of the epileptiform response. Peak amplitude was unaffected. C: 6 min exposure to 20 PM D-APV, in the continued presence of CNQX, completely suppressed postsynaptic activity. D: effectsof CNQX and D-APV were partially reversed on 11 min washing with antagonist-free saline. RMP, -93 mV; R,, 20 MQ.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
W.-L. LEE AND J. J. HABLITZ
92
larization and decrease with depolarization. Near the RMP, NMDA responses decrease in amplitude with both hyperpolarization and depolarization. Figure 7,A and B, shows examples of PDSs at a variety of membrane potentials before and after CNQX, respectively. A plot of peak PDS amplitude versus membrane potential is shown in Fig. 70. PDS amplitude was linearly related to membrane potential under both conditions. The extrapolated reversal potential for PDSs and residual PDSs (i.e., those recorded in presence of CNQX) were -33.8 t 7.3 (n = 6) and -32.2 t 5.5 mV (n = 6), respectively. The relatively negative values obtained for reversal potentials result from an inability to depolarize the membrane potential beyond the reversal potential and the necessity to rely on extrapolated versus interpolated measurements of the reversal point. Careful examination of records obtained during these experiments indicated the presence of an early depolarization preceding the full PDS (Fig. 7C). Because this potential seemed to trigger or initiate the PDS, we determined whether this response had a unique voltage dependence. In the absence of CNQX, complex voltage dependencies were observed such that depolarization produced decreases in amplitude in two neurons and increases in two others. This
A
indicates varying contributions from NMDA and nonNMDA receptors. In the presence of CNQX, early response amplitude increased with depolarization over the range - 110 to -56 mV. This was a consistent finding in all neurons tested (n = 6). A typical example is shown in Fig. 7,C2 and E. This increase in amplitude with depolarization can be attributed in part to activation of voltage-dependent currents, particularly a persistent sodium conductance (Flatman et al. 1986; Sutor and Hablitz 1989b) because R, did increase with depolarization in all the neurons tested (Fig. 7F). However, in the neuron shown, despite the increase in resistance, the small depolarization preceding the PDS in the absence of CNQX decreased rather than increased with depolarization (Fig. 7, CI and E). To more directly examine the voltage dependence of the early depolarization without contaminating influences from active membrane conductances, the lidocaine derivative QX-3 14 was used to block sodium currents (Connors and Prince 1982; Sutor and Hablitz 1989b). After intracellular injection of QX-3 14, neurons had relatively constant R,s over the range -90 to -60 mV (Fig. SE), and fast action potentials were not evoked. Examples of PDSs recorded in QX-3 14-injected neurons are shown in Fig. 8
Control -66mV
-1 OOmV
-76mV
-89mV
4 nl
’
-100
25mVL 1 OOms
@i
E
CNQX -98mV
-76mV
-65mV
s 2s N3 .A b al --” 8” 4 -U-xe- 0 c
L
.
.
_
’
.
-80’
.
.
-50 Vm (mV)
1
Cl
10
s-
:-‘::;- wo-
’ -100
--.
.
.
-60
Vm (mV>
C
1
2
I -7OmV
-69mV
F
501
-1OOmV -99mV
2-
25-
’
5
-80 Vm hV> FIG. 7. Voltage dependence of PDSs and early depolarizations in the presence and absence of CNQX. A: control PDSs recorded at 4 different membrane potentials. Note the increase in PDS amplitude at more hyperpolarized membrane potentials. B: residual PDSs after CNQX application. PDS amplitude still increased with hyperpolarization. C: expanded time base records of early depolarization that preceded the PDS. Superimposed records of early depolarizations obtained at 2 membrane potentials are shown. Cl: early depolarizations under control conditions. C2: in the presence of CNQX, the early depolarization was bigger at -69 than -99 mV. D: plot of peak PDS amplitude as a function of membrane potential (mV). Control PDS amplitude (0) decreased linearly with membrane depolarizations. Similar results were obtained in the presence of CNQX (0). Each point is the average of 4 records. E: plot of early depolarization amplitude vs. membrane potential. Under control conditions (o), early depolarizations were measured 8.3 ms after stimulus artifact. Early depolarizations in the presence of CNQX (0) were measured 34.2 ms after the stimulus artifact. Note the gradual increase in amplitude as the membrane potential was depolarized from - 100 to -62 mV. F: R, as function of membrane potential. RMP, -89 mV.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA
AND
NON-NMDA
RECEPTORS
AND
EPILEPTIFORM
ACTIVITY
OL--1
-120
c
25mV
L
.
-80
1
1
-40
0
Vm (mV>
1 OOms
-8OmV
2mV(oms
B /
/
-4gmV
-68mV
FIG. 8. Voltage dependence of CNQX-resistant PDSs and early depolarizations in a QX-3 14-injected rat neocortical neuron. A: specimen records of epileptiform responses obtained at different membrane potentials. Fast Na-dependent action potentials were abolished. Remaining regenerative responses are attributable to calcium electrogenesis or sodium conductances in remote regions of the neuron. B: superimposed records of early depolarizations. Bl: selected responses obtained at membrane potentials where R, was constant. The early depolarization was largest at -68 mV and became progressively smaller at -80 and -92 mV. B2: early depolarizations obtained at the extremes of membrane polarization. Response was largest at -49 mV despite a decrease in R,. C: plot of PDS amplitude as a function of membrane potential. PDS amplitude decreased linearly between -86 and -49 mV, and the extrapolated reversal potential was 7.5 mV. D: early depolarization amplitude plotted vs. membrane potential. Note the gradual increase in its amplitude with membrane depolarization. E: plot of R, vs. membrane potential. Note the relatively constant R, at potentials between -92 and -68 mV. RMP, -80 mV.
and resemble responses previously described by Gutnick et al. ( 1982). In QX-3 14-injected cells, the amplitude of residual PDSs recorded in the presence of CNQX decreased with depolarization (Fig. 8, A and C). The extrapolated reversal potential was 9.8 t 5.4 mV (n = 4). Simultaneously measured early depolarizations increased in amplitude with depolarization. Over a region where R, remained constant, early response amplitude increased as the membrane potential was depolarized. The amplitude of the early depolarization continue to increase with depolarization beyond -60 mV despite a progressive decrease in RN (Fig. 8, B2, D, and E), suggesting that the increase in early depolarization amplitude was not simply due to changes in passive properties. DISCUSSION
The results of the present series of experiments suggest that reductions of inhibition in the neocortex produce synchronized epileptiform discharges by enhancing late polysynaptic EPSPs. Fully developed paroxysmal discharges were grouped into two classes on the basis of duration. In each case, an early depolarizing event was seen preceding
the PDS and appeared responsible for initiation of paroxysmal activity. Epileptiform discharges were reduced by antagonists of both the NMDA and non-NMDA types of excitatory amino acid receptors, suggesting that, as in the hippocampus (Lee and Hablitz 1989a), both receptor types contribute to generation of burst discharges. Generation
of PDSs
PDSs recorded here were similar to those described previously in studies of disinhibited neocortex in vivo (Matsumoto and Ajmone-Marsan 1964; Matsumoto et al. 1969) and in vitro (Gutnick et al. 1982; Thomson and West 1986). However, in the present study, based on duration, there appeared to be two types of PDSs. During our studies of excitatory amino acid antagonists, generally only one neuron was studied in a given preparation. Short- and longduration PDSs were therefore seen in different slices, usually from different preparations. This raises the possibility that variations in the orientation of the brain during slice preparation, with variable preservation of intracortical connections, contributed to the observed differences. Shortduration PDSs were not a sign of poor slice viability be-
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
94
W.-L.
LEE
AND
J. J. HABLITZ
cause such responses were stable over time, evoked at cqmparable stimulus intensities and observed in neurons whose properties were representative of the whole population. Slice orientation has been shown to affect variables such as dye coupling (Gutnick et al. 1985), and further investigation of this factor may yield information concerning the intracortical connections regulating the degree of excitability seen during epileptiform discharges. The greater sensitivity of short-duration PDSs to blockade by CNQX suggests that a different population of cells or fibers were activated in these preparations. Considerable evidence has accumulated indicating that a common mechanism of action of many convulsant drugs is to decrease GABA-mediated, chloride-dependent inhibition (Macdonald 1984). PDSs generated in this disinhibition model of epilepsy have increased our understanding of the mechanisms underlying paroxysmal bursts (Hablitz 1984; Wong et al. 1984). Previous studies have shown that the neocortex is particularly sensitive to reductions in inhibition, demonstrating seizure activity when GABA-mediated responses in the brain stem and vestibular nuclei were barely affected (Davenport et al. 1979; Tribble et al. 1983). Under normal conditions, inhibition in the neocortex controls the appearance of a late polysynaptic EPSP in -layer II/III neurons (Sutor & Hablitz 1989a) and can mask the expression of long-term potentiation (Sutor and Hablitz 1989~). Similarly, it has been shown that partial suppression of GABA,-receptor-mediated inhibition expands the horizontal or tangential spread of activity in the cortex (Chagnac-Amitai and Connors 1989a). Thus the synaptic mechanism whereby decreases in GABAergic inhibition results in the generation of synchronous epileptiform discharges appears to be a loss of inhibitory control over local excitatory circuits within the neocortex. A small EPSP-like depolarization has been previously observed to precede PDS onset in the neocortex (see also Gutnick et al. 1982; Thomson and West 1986). However, no attempt was made to examine the properties of this event. We have shown that this initial depolarization is synaptic in origin. Under control conditions, the effects of alterations in membrane potential were variable with both increases and decreases in amplitude seen on depolarization. When CNQX was used to block non-NMDA receptors, the early depolarization demonstrated a non-conventional NMDA receptor type voltage dependence. Similar nonlinear behavior of EPSPs has been described in the hippocampus (Dingledine et al. 1986) and suggests that, under disinhibited conditions, an NMDA-mediated synaptic potential can be evoked. This event may then serve as a trigger for initiation of the actual PDS, which has both NMDA and non-NMDA components. Directly evoked NMDA-mediated synaptic potentials have been seen in disinhibited hippocampal slices after blockade of non-NMDA receptors (Lee and Hablitz 1990), and similar potentials appear to underlie seizures originating from midbrain structures (Pierson et al. 1989). Such NMDA-mediated depolarizations may thus serve as triggers for epileptiform activity in many brain structures.
acid receptors of both the NMDA and non-NMDA type were blocked. Selective blockade of individual receptor types indicated that NMDA receptor blockade affected both the amplitude and duration of the PDS, whereas antagonism of non-NMDA receptors influenced the time of onset and duration of PDSs but not their amplitude. Previous studies of the effect of NMDA antagonists on epileptiform activity in the neocortex (Thomson and West 1986) and hippocampus (Dingledine et al. 1986) have shown PDSs to be reduced in amplitude and duration. In the present study, measured from the RMP, PDSs had amplitudes of -60 mV, with the peak of the PDS corresponding to a membrane potential near -20 mV. This is a region of membrane potential where the voltage-dependent block by Mg2+ is minimal (Nowak et al. 1984), and a contribution from NMDA receptors would be prominent and a decrease in amplitude expected after antagonist exposure. Previous studies on in vitro slices of hippocampus (Collingridge et al. 1988), neocortex (Jones and Baughman 1988), and hippocampal cells in culture (Forsythe and Westbrook 1988) have suggested that NMDA-mediated synaptic potentials are long in duration. Blockade of such a long-duration component would account for the decrease in PDS duration produced by D-APV. At 5 PM, CNQX is a quite specific and effective antagonist of non-NMDA receptors (Fletcher et al. 1988; Hablitz and Sutor 1990); therefore responses observed in the presence of agent are likely to be governed predominantly by NMDA receptor activation. The slower rising phase of the PDS in the presence of CNQX and the concomitant increase in latency to onset may reflect less synchronous activation of NMDA receptors and poorer recruitment of local excitatory pathways. Despite the presence of a D-APV-sensitive, i.e., NMDAmediated, component to PDSs and residual PDSs recorded in the presence of CNQX, plots of response amplitude versus membrane potential were always linear, indicating a reversal potential near 0 mV. This linear behavior is consistent with previous studies of the voltage dependence of epileptiform activity (Gutnick et al. 1982; Johnston and Brown 198 1). The lack of an unconventional NMDA-like voltage dependence suggests that other conductances, both synaptic and voltage dependent, are of a sufficient magnitude to mask the voltage-dependent behavior of the NMDA channels. A similar high conductance state has been seen in response to iontophoretically applied NMDA in both hippocampus (Hablitz 1982) and neocortex (Flatman et al. 1983; Sutor and Hablitz 1989b). Thus the lack ofa NMDAlike voltage dependence is not a sufficient criteria for ruling out contributions from NMDA receptors.
Efects of excitatory amino acid antagonists Another principal finding of this study was that epileptiform discharges were not observed when excitatory amino
CHAGNAC-AMITAI, Y. AND CONNORS, B. W. Horizontal spread of synchronized activity in neocortex and its control by GABA-mediated inhibition. J. Neurophysiol. 61: 747-758, 1989a. CHAGNAC-AMITAI, Y. AND CONNORS, B. W. Synchronized excitation and
This work was supported by National Institute of Neurological Disorders and Stroke Grants NS- 18 145 and NS-22373. Address for reprint requests: J. J. Hablitz, Neurobiology Research Center, University of Alabama at Birmingham, Birmingham, AL 35294. Received
13 April
1990; accepted
in final
form
6 September
1990.
REFERENCES
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
NMDA inhibition
driven
by intrinsically
bursting
AND
NON-NMDA
neurons
in neocortex.
RECEPTORS J. Neu-
rophysiol. 62: 1149- 1162, 1989b. COLLINGRIDGE, G. L., HERRON, C. E., AND LESTER, R. J. Synaptic activation of N-methyl-D-aspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus. J. Physiol. Lond. 399: 283-300, 1988. CONNORS, B. W. Initiation of synchronized bursting in neocortex. Nature Lond. 310: 685-687, 1984. CONNORS, B. W., MALENKA, R. C., AND SILVA, L. R. Two inhibitory postsynaptic potentials, and GABA, receptor-mediated responses in neocortex of rat and cat. J. Physiol. Lond. 406: 443-468, 1988. CONNORS, B. W. AND PRINCE, D. A. Effects of local anesthetic QX-3 14 on the membrane properties of hippocampal pyramidal neurons. J. Pharmacol. Exp. Ther. 220: 476-48 1, 1982. DAVENPORT, J., SCHWINDT, P. C., AND CRILL, W. E. Epileptogenic doses of penicillin do not reduce a monosynaptic GABA-mediated postsynaptic inhibition in the intact anesthetized cat. Exp. Neural. 65: 552-572, 1979. DINGLEDINE, R., HYNES, M. A., AND KING, G. L. Involvement of Nmethyl-D-aspartate receptor in epileptiform bursting in the rat hippocampal slice. J. Physiol. Land. 380: 175-l 89, 1986. FLATMAN, J. A., SCHWINDT, P. C., AND CRILL, W. E. Multiple actions of N-methyl-D-aspartate on cat neocortical neurons in vitro. Brain Res. 266: 169-173,1983. FLATMAN, J. A., SCHWINDT, P. C., AND CRILL, W. E. The induction and modification of voltage-sensitive responses in cat neocortical neurons by iV-methyl-D-aspartate. Brain Res. 363: 62-77, 1986. FLETCHER, E. J., MARTIN, D., ARAM, J. A., AND HONORE, T. Quinoxalinediones selectively block quisqualate and kainate receptors and synaptic events in rat neocortex and hippocampus and frog spinal cord in vitro. Br. J. Pharmacol. 95: 585-597, 1988. FORSYTHE, I. D. AND WESTBROOK, G. L. Slow excitatory postsynaptic currents mediated by N-methyl-D-aspartate receptors on cultured mouse central neurons. J. Physiol. Lond. 396: 5 15-533, 1988. GUTNICK, M. J., CONNORS, B. W., AND PRINCE, D. A. Mechanisms of neocortical epileptogenesis in vitro. J. Neurophysiol. 48: 132 I- 1335, 1982. GUTNICK, M. J., LOBEL-YAAKOV, R., AND RIMON, G. Incidence of neuronal dye-coupling in neocortical slices depends on the plane of section. Neuroscience 15: 659-666, 1985. HABLITZ, J. J. Conductance changes induced by DL-homocysteic acid and N-methyl-DL-aspartic acid in hippocampal neurons. Brain Res. 247: 149-153,1982. HABLITZ, J. J. Picrotoxin-induced epileptiform activity in hippocampus: role of endogenous versus synaptic factors. J. Neurophysiol. 5 1: 10 1 l1027, 1984. HABLITZ, J. J. AND SUTOR, B. Excitatory postsynaptic potentials in rat neocortical neurons in vitro. III. Effects of a quinoxalinedione nonNMDA receptor antagonist. J. Neurophysiol. 64: 1282- 1290, 1990. HONORE, T., DAVIES, S. N., DREJER, J., FLETCHER, E. J., JACOBSEN, P., LODGE, D., AND NIELSEN, F. E. Quinoxalinedione: potent competitive non-NMDA glutamate receptor antagonists. Science Wash. DC 24 1: 701-703, 1988. HOWE, J. R., SUTOR, B., AND ZIEGLGANSBERGER, W. Characteristics of long-duration inhibitory postsynaptic potentials in rat neocortical neurons in vitro. Cell Mol. Neurobiol. 7: I- 18, 1987. JOHNSTON, D. AND BROWN, T. H. Giant synaptic potential hypothesis for epileptiform activity. Science Wash. DC 2 11: 294-297, 198 1.
AND
EPILEPTIFORM
ACTIVITY
95
JONES, K. A. AND BAUGHMAN, R. W. NMDAand non-NMDA-receptor components of excitatory synaptic potentials recorded from cells in layer V of rat visual cortex. J. Neurosci. 8: 3522-3534, 1988. LEE, W. L. AND HABLITZ, J. J. Involvement of non-NMDA receptors in picrotoxin-induced epileptiform activity in the hippocampus. Neurosci. Lett. 107: 129-134, 1989a. LEE, W. L. AND HABLITZ, J. J. Role of NMDA and non-NMDA receptors in picrotoxin-induced epileptiform activity in rat neocortex. Sot. Neurosci. Abstr. 15: 1213, 1989b. MACDONALD, R. L. Anticonvulsant and convulsant drug actions on vertebrate neurones in primary dissociated cell culture. In: Electrophysiology OfEpilepsy, edited by P. A. Schwartzkroin and H. Wheal. London: Academic, 1984, p. 353-388. MATSUMOTO, M. AND AJMONE-MARSAN, C. Cortical cellular phenomena in experimental epilepsy: interictal manifestations. Exp. Neural. 9: 286326, 1964. MATSUMOTO, M., AYALA, G. F., AND GUMNIT, R. J. Neuronal behavior and triggering mechanism in cortical epileptic focus. J. Neurophysiol. 32:688-703, 1969. MAYER, M. L., WESTBROOK, G. L., AND GUTHRIE, P. B. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature Lond. 309: 26 l-263, 1984. NOWAK, L., BREGESTOVSKI, P., ASCHER, P., HERBET, A., AND PROCHIANTZ, A. Magnesium gates glutamate-activated channels in mouse central neurones. Nature Lond. 307: 462-465, 1984. PIERSON, M. G., SMITH, K. L., AND SWANN, J. S. A slow NMDA-mediated synaptic potential underlies seizures originating from midbrain. Brain Res.486: 381-386, 1989. STREIT, P. Glutamate and aspartate as transmitter candidates for system of the cerebral cortex. In: Cerebral Cortex. Functional Properties ofcortical Cells, edited by E. J. Jones and A. Peters. New York: Plenum, 1984, vol. 2, p. 119-143. SUTOR, B. AND HABLITZ, J. J. EPSPs in rat neocortical neurons in vitro. I. Electrophysiological evidence for two distinct EPSPs. J. Neurophysiol. 61.607-620, 1989a. SUTOR, B. AND HABLITZ, J. J. EPSPs in rat neocortical neurons in vitro. II. Involvement of N-methyl-r>-aspartate receptors in generation of EPSPs. J. Neurophysiol. 6 1: 62 l-634, 1989b. SUTOR, B. AND HABLITZ, J. J. Long-term potentiation in frontal cortex: role of NMDA-modulated polysynaptic excitatory pathways. Neurosci. Lett. 97: 11 l-l 17, 1989~. THOMSON, A. M. A magnesium-sensitive post-synaptic potential in rat cerebral cortex resembles neuronal responses to N-methylaspartate. J. Physiol. Lond. 370: 53 l-549, 1986. THOMSON, A. M. AND WEST, D. C. N-methylaspartate receptors mediate epileptiform activity evoked in some, but not all, conditions in rat neocortical slices. Neuroscience 19: 116 1- 1177, 1986. TRIBBLE, G. L., SCHWINDT, P. C., AND CRILL, W. E. Reduction of postsynaptic inhibition tolerated before seizure initiation: brain stem. Exp. Neural. 80: 304-320, 1983. WEISS, D. S. AND HABLITZ, J. J. Interaction of penicillin and pentobarbital with inhibitory synaptic mechanisms in neocortex. Cell. Idol. Neuro-
biol. 4: 301-317, 1984. WONG, R. K. S., TRAUB, R. D., AND MILES, R. Epileptogenic mechanisms as revealed by studies of the hippocampal slice. In: Electrophysiology of Epilepsy, edited by P. A. Schwartzkroin and H. Wheal. London: Academic, 1984, p. 253-276.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (132.174.254.155) on December 5, 2018. Copyright © 1991 the American Physiological Society. All rights reserved.
