The Journal of Neurophysiology Vol. 79 No. 2 February 1998, pp. 659-669
Copyright ©1998 by the American Physiological Society

Transient Suppression of GABA_A-Receptor-Mediated IPSPs After Epileptiform Burst Discharges in CA1 Pyramidal Cells

F.E.N. Le Beau and B. E. Alger

Department of Physiology, University of Maryland School of Medicine, Baltimore, Maryland 21201

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Le Beau, F.E.N. and B. E. Alger. Transient suppression ofGABA_A-receptor-mediated IPSPs after epileptiform burst discharges in CA1 pyramidal cells. J. Neurophysiol. 79: 659-669, 1998. Epileptiform burst discharges were elicited in CA1 hippocampal pyramidal cells in the slice preparation by perfusion with Mg²⁺-free saline. Intracellular recordings revealed paroxysmal depolarization shifts (PDSs) that either occurred spontaneously or were evoked by stimulation of Schaffer collaterals. These bursts involved activation of N-methyl-D-aspartate receptors because burst discharges were reduced or abolished by DL-2-amino-5-phosphonovaleric acid. Bath application of carbachol caused an increase in spontaneous activity that was predominantly due to -aminobutyric acid-A-receptor-mediated spontaneous inhibitory postsynaptic potentials (sIPSPs). A marked reduction in sIPSPs (31%) was observed after each epileptiform burst discharge, which subsequently recovered to preburst levels after ~4-20 s. This sIPSP suppression was not associated with any change in postsynaptic membrane conductance. A suppression of sIPSPs also was seen after burst discharges evoked by brief (100-200 ms) depolarizing current pulses. N-ethylmaleimide, which blocks pertussis-toxin-sensitive G proteins, significantly reduced the suppression of sIPSPs seen after a burst response. When increases in intracellular Ca²⁺ were buffered by intracellular injection of ethylene glycol bis(-aminoethyl)ether-N,N,N,N-tetraacetic acid, the sIPSP suppression seen after a single spontaneous or evoked burst discharge was abolished. Although we cannot exclude other Ca²⁺-dependent mechanisms, this suppression of sIPSPs shared many of the characteristics of depolarization-induced suppression of inhibition (DSI) in that it involved activation of G proteins and was dependent on increases in intracellular calcium. These findings suggest that a DSI-like process may be activated by the endogenous burst firing of CA1 pyramidal neurons.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

A number of different mechanisms that modify -aminobutyric acid (GABA) release, either transiently or in the long term, now have been identified. Depolarization of pyramidal cells in the hippocampus (Alger et al. 1996; Pitler and Alger 1992a, 1994) and cerebellar Purkinje cells (Llano et al. 1991; Vincent and Marty 1993; Vincent et al. 1992) is followed by a reduction (lasting from 30 s to 1 min) inGABAergic inhibitory postsynaptic currents (IPSCs). This decrease in inhibition, observed with both spontaneous and evoked GABA responses, has been termed depolarization-induced suppression of inhibition (DSI). The decreased inhibition occurring as a consequence of DSI should have a significant effect on neuronal integration in the hippocampus. Pyramidal cell excitability is increased during DSI in the hippocampus (Wagner and Alger 1996), whereas decreases in GABAergic inhibition facilitate the induction of long-term potentiation (LTP) (Stelzer et al. 1994; Wigstrom and Gustafsson 1983) and the onset of certain types of epileptic activity (Stelzer et al. 1987). Considerable evidence suggests that DSI is dependent on an influx of Ca²⁺ into the pyramidal cell after the depolarizing stimulus (Pitler and Alger 1992a). This Ca²⁺ influx then triggers the induction of a retrograde signal, which causes a transient decrease in GABA release from the presynaptic terminal (Alger and Pitler 1995; Pitler and Alger 1992a, 1994). A recent study proposed that glutamate, or a glutamate-like substance, is the retrograde messenger in cerebellum (Glitsch et al. 1996). However, the identity of the retrograde messenger in the hippocampus remains to be determined.

In most studies of DSI, the depolarizing stimulus was either brief trains of current pulses producing a series of action potentials or a 1- or 2-s depolarizing voltage step delivered to the postsynaptic cell. The question remains whether DSI occurs as a result of cell firing after synaptic stimulation. A single Ca²⁺ spike is sufficient to produce DSI (Pitler and Alger 1994), suggesting that DSI could occur under conditions in which there is sufficient influx of Ca²⁺ such as might occur, for example, during an epileptiform burst discharge. Hippocampal pyramidal cells can fire burst responses under both normal (Kandel et al. 1961) and epileptic conditions. Burst responses are composed, at least in part, of a voltage-dependent Ca²⁺ component (Schwartzkroin and Prince 1978; Wong and Prince 1979). The aim of this study was, therefore, to determine whether a suppression of sIPSPs in CA1 pyramidal cells could be observed after an epileptiform burst discharge.

The most commonly studied acute models of epilepsy induction, including penicillin, bicuculline, and picrotoxin, all involve a depression of GABAergic inhibition (for review see Schwartzkroin 1993) and were, therefore, unsuitable for use in this study. However, lowering extracellular Mg²⁺ removes the voltage-dependent block of the N-methyl-D-aspartate (NMDA) receptor by Mg²⁺ (Mayer et al. 1984) and enhances NMDA transmission. Perfusion of hippocampal slices in a solution nominally free of Mg²⁺ results in epileptiform burst discharges (Anderson et al. 1986; Avoli et al. 1987; Mody et al. 1987; Neuman et al. 1989; Schneiderman and MacDonald 1987; Tancredi et al. 1990). GABAergic inhibition remains intact under these conditions (Benardo 1993; Tancredi et al. 1990), and the Mg²⁺-free saline model thus provides a useful tool with which to study changes in inhibition during epileptiform activity.

Our results show that a transient reduction in sIPSPs is observed after an epileptiform burst response induced in Mg²⁺-free saline. This suppression of sIPSPs shares many of the characteristics of DSI as previously described and suggests that DSI can occur under more physiological conditions and therefore could play an important role in information coding in the hippocampus.

	METHODS

Abstract Introduction Methods Results Discussion References

Preparation of slices

Adult male Sprague-Dawley rats (30-40 days) were anesthetized deeply with halothane and decapitated. The brain was removed quickly, and the hippocampi dissected free. Hippocampal slices (400-µm thick) were cut in an agar block with a Vibratome (Technical Products, International). The slices then were allowed to recover for 1 h in a holding chamber at the interface of a physiological saline and humidified 95% O₂-5% CO₂ atmosphere at room temperature. Recordings were carried out in a submerged, perfusion-type chamber (Nicoll and Alger 1981) where the slice was perfused at 0.5-1 ml/min at 29-31°C.

Solutions and drugs

Standard saline contained (in mM) 120 NaCl, 25 NaHCO₃, 3 KCl, 2.5 CaCl₂, 2 MgSO₄, 1 NaH₂PO₄, and 10 glucose. To induce epileptiform bursting, MgSO₄ was omitted from the perfusate, and in most experiments, the concentration of CaCl₂ was increased to 3.5 mM. 6-Cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 µM) and 2-amino-5-phosphonovaleric acid (APV, 50-100 µM) were purchased from Research Biochemicals (Natick, MA). Bicuculline methiodide (10-20 µM), tetraethylammonium chloride (TEA; 10 mM), 4-aminopyridine (4-AP; 10-50 µM), N-ethylmaleimide (NEM; 250 µM), and ethylene glycol bis(-aminoethyl)ether-N,N,N,N-tetraacetic acid (EGTA; 100 mM) were purchased from Sigma (St. Louis, MO).

Recordings and data acquisition

Intracellular recordings were made from CA1 pyramidal cells using 3 M KCl filled glass microelectrodes pulled to resistances of 50-90 M. In most cases, a moderate holding current (less than 0.5 nA) was used to maintain a slightly hyperpolarized membrane potential (70 to 80 mV) and enhance the amplitude of the sIPSPs. TEA (10 mM) was added to the electrode solution to reduce any residual afterhyperpolarizations (AHPs) not blocked by bath application of carbachol. In cells recorded with EGTA in the electrode, EGTA either diffused into the cell passively or was injected using 0.1- to 0.5-nA hyperpolarizing current steps (150- to 200-ms duration).

The signals were recorded with an Axoclamp-2A amplifier, amplified, digitized, and stored on a personal computer using pCLAMP 6.0 (Axon Instruments, Foster City, CA). In addition, data were stored on a VCR-based tape recorder system (Neurocorder, Neuro Data Instruments, New York, NY) and could be played back for off-line analysis either on a rectilinear chart recorder (Gould, Cleveland, OH) or into the appropriate pCLAMP program. A bipolar tungsten stimulating electrode (David Kopf Instruments, Tujunga, CA) was positioned in stratum radiatum of CA2/CA3 region to activate the Schaffer collateral system orthodromically. Evoked burst discharges in response to electrical stimuli (duration 50 µs) were adjusted to produce a response just suprathreshold for action potential generation.

Data analysis

Because both the frequency and amplitude of sIPSPs were altered after a burst, the response record was integrated to obtain a measure that would reflect both these changes (Pitler and Alger 1992a). Responses were integrated in 1- or 2-s time bins during a 5-s period before the onset of the burst. This was compared with a 5-s postburst period commencing 1 s after the onset of the burst or, for multiple burst responses, 1 s after the onset of the last burst. Control synaptic activity therefore is plotted as the integral value (mV·ms), whereas changes in the integral values after a burst response are expressed as a percentage of the control value. For simplicity, we refer to the integrated value as the "sIPSP magnitude." Integration was performed using the Fetchan program of pCLAMP 6.0. The magnitude of the epileptiform burst response elicited in Mg²⁺-free saline also was measured as an integral value (mV·ms). A burst was measured from the point of inflection of the first action potential until the membrane potential returned to baseline. This measure of burst size was chosen because the integral value incorporates both the paroxysmal depolarization shifts (PDS) and the action potentials elicited by each burst. Changes in synaptic activity after application of bicuculline were determined by comparing the level of synaptic activity (integral) during a 7- to 8-s window in control and bicuculline. Input resistance was assessed using 0.2- to 0.3-nA hyperpolarizing current pulses (100-ms duration) at 1-s intervals. To compare conductances in the presence and absence of a drug at the same membrane potential, DC was injected through the electrode to compensate for any drug-induced change in membrane potential.

Unless otherwise stated (Fig. 6), the results reported were obtained by averaging the data from three burst responses in each neuron. Results are expressed as means ± SE. Statistical analysis was carried out on Sigmaplot using a Student's t-test or a Pearson product moment correlation. The significance level chosen wasP < 0.05. 

View larger version (23K): [in this window] [in a new window]   FIG. 6. Magnitude of sIPSP suppression for single vs. multiple burst responses. In cells recorded in the absence of EGTA (A), the degree of sIPSP suppression increases with increasing number of either spontaneous or evoked burst responses. When EGTA is included in the electrode solution (B), there is no significant sIPSP suppression after a single burst and a reduced suppression after a series of multiple bursts. Pairs of numbers above each bar represent first, the total number of cells recorded and second, the total number of burst responses measured. * Significant difference as determined by Student's t-test.

	RESULTS

Abstract Introduction Methods Results Discussion References

Recordings were made from CA1 pyramidal neurons with resting membrane potentials greater than 50 mV and action potential amplitudes >70 mV. Control input resistances ranged from 39.2 to 65.2 M (mean 48.5 ± 8.0 M,n = 20).

Postsynaptic pyramidal cell firing decreases spontaneous IPSPs

Figure 1A shows an example of one cell in which DSI of sIPSPs was induced (in normal saline) in response to a depolarizing voltage step (0.7 nA, 1-s duration) as previously described (Pitler and Alger 1992a, 1994). APV (50 µM), CNQX (10 µM), and carbachol (1 µM) were added to the bathing medium. Carbachol greatly increases the spontaneous excitability of interneurons, via activation of muscarinic receptors (Martin and Alger 1996), leading to an increase in the number of large spontaneous GABA_A IPSPs recorded in pyramidal cells (Pitler and Alger 1992b). Although DSI can occur in the absence of carbachol, application of carbachol greatly facilitates the study of DSI of spontaneous IPSP activity. In this study, epileptiform burst discharges were evoked in two cells in Mg²⁺-free saline in the absence of carbachol. In both cases, the overall level of synaptic activity was too low to identify any changes in sIPSPs, therefore, for all the data reported here, 1-5 µM carbachol was included in both the normal saline and Mg²⁺-free saline perfusates.

View larger version (31K): [in this window] [in a new window]   FIG. 1. Postsynaptic depolarization and epileptiform burst discharges cause suppression of spontaneous inhibitory postsynaptic potentials (sIPSPs). A: 6 consecutive traces (4-s duration) showing sIPSPs recorded in normal saline with DL-2-amino-phosphonovaleric acid (APV, 50 µM), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 µM), and carbachol (1 µM). A 1-s depolarizing pulse (0.7 nA) given in trace 2 (*) caused a dramatic (66%) reduction in the magnitude of integrated sIPSPs. Bottom trace: recovery of sIPSPs ~50 s after the depolarizing step. Response to the depolarizing step is truncated for clarity but is illustrated in full in B and illustrates carbachol's ability to block spike accommodation. Resting potential 57 mV. (As detailed in METHODS moderate holding currents were used in all cells to hyperpolarize the membrane between 70 and 80 mV). C: intracellular recordings of spontaneous epileptiform burst discharges for 1 cell recorded 15, 40, and 60 min after commencing perfusion with Mg2+-free saline. Burst shape and magnitude remain constant over time. D: in a 2nd cell perfused in Mg2+-free saline, evoked epileptiform burst discharges represented by the vertical lines (truncated for clarity) were followed by a period of decreased synaptic activity. Input conductance during the experiment was assessed using 0.3 nA, 100-ms duration, hyperpolarizing pulses every second. There was no change in conductance after the burst that could account for the suppression of sIPSPs. Resting potential 68 mV. [In this and all other cells shown, Mg2+-free saline contained 3.5 mM [Ca2+]o and 1-3 µM carbachol; electrode contained 10 mM tetraethylammonium (TEA)].

In Fig. 1A, the 1-s depolarizing current pulse (*) produced a dramatic (66%) reduction in sIPSPs from control levels that gradually recovered, in this case after 20-30 s. The response to the depolarizing step is shown in Fig. 1B. A small proportion of cells do not appear to exhibit DSI, so we first tested for the presence of DSI in normal divalent cation conditions in the majority of cells recorded in this study (except those with EGTA, see further) before switching to Mg²⁺-free saline. Only cells that exhibited DSI were studied further.

Application of bicuculline (10-20 µM) decreased the integrated level of synaptic activity by 72 ± 5.2% (n = 5). Thus the majority of the observed postsynaptic potentials were due to GABA_A-receptor-mediated events. In view of this, changes in postsynaptic activity will be referred to as changes in sIPSPs.

Magnesium-free saline induced burst responses

Replacement of the normal saline with one essentially free of Mg²⁺ leads to the emergence of epileptiform burst responses in in vitro brain slices (Anderson et al. 1986; Avoli et al. 1987; Mody et al. 1987; Neuman et al. 1989; Schneiderman and MacDonald 1987; Tancredi et al. 1990). Magnesium-free solutions enhance NMDA responses by relieving the Mg²⁺-dependent voltage-sensitive blockade of NMDA channels. Generally, burst responses either occurring spontaneously or evoked by stimulation in s. radiatum were seen within 15-20 min of switching to Mg²⁺-free saline. Examples of single spontaneous bursts for one cell, recorded at different times in Mg²⁺-free saline, are shown in Fig. 1C. The burst consisted of a large depolarization (15-25 mV) with a mean total duration of 333 ± 109 ms (n = 9), and several fast action potentials superimposed on the depolarizing response. The depolarizing burst response, termed a PDS, was qualitatively similar to that seen previously in Mg²⁺-free saline (Benardo 1993; Tancredi et al. 1990). However, because of the inclusion of 10 mM TEA in the recording solution (see METHODS) and 1-5 µM carbachol in the bath solution, the large Ca²⁺-dependent afterhyperpolarizations that usually follow the PDS (Alger and Nicoll 1980; Schwartzkroin and Stafstrom 1980) were reduced significantly or absent in this study. Although the magnitude and form of the burst responses varied slightly among cells, the responses were relatively constant within each cell and remained stable for long recording periods. This is illustrated in Fig. 1C, which shows spontaneous burst responses for one neuron, recorded 15, 40, and 60 min after switching to Mg²⁺-free saline. The burst responses induced in Mg²⁺-free saline were abolished gradually on perfusion with normal saline, and the cells regained their original firing characteristics (data not shown).

In experiments where no other cation was added to compensate for the removal of Mg²⁺, the frequency of bursting rapidly increased such that, after 20-30 min in Mg²⁺-free saline, burst responses occurred every few seconds. The frequency of spontaneous bursting was not affected by hyperpolarizing the membrane potential. The uncontrollable nature of these spontaneous events and the frequency with which they occurred, rendered it impossible to study changes in sIPSPs for more than a few minutes in Mg²⁺-free saline. Attempts to control the burst frequency by adding back small amounts of Mg²⁺ (0.5 and 0.25 mM Mg²⁺) were unsuccessful as burst responses were seen only in saline nominally Mg²⁺ free, as has been observed previously (Schneiderman and MacDonald 1987). However, other studies (Benardo 1993; Mody et al. 1987) showed that the frequency of bursting in Mg²⁺-free saline could be controlled by increasing the calcium concentration present in the bathing medium. In this study, increasing extracellular Ca²⁺ in the Mg²⁺-free saline from 2.5 to 4.5 mM almost totally abolished spontaneous bursts, although evoked burst responses still could be obtained. With 3.5 mM extracellular Ca²⁺, spontaneous bursts still occurred but at a sufficiently low rate to permit study. All subsequent experiments therefore were carried out with 3.5 mM extracellular Ca²⁺ in the Mg²⁺-free saline. In three cells in which no further experimental treatments were employed, the effects of perfusion in Mg²⁺-free saline (with carbachol) on sIPSP activity were assessed. We measured the integrated sIPSP value between 10 and 60 min into perfusion with Mg²⁺-free saline and found no significant change (<10%) with increasing time in Mg²⁺-free saline.

Suppression of sIPSPs after burst responses

In nearly all cells (n = 23) in which DSI was observed in normal saline with APV, CNQX and carbachol before perfusion with Mg²⁺-free saline, some suppression of sIPSPs after a burst response was seen in the Mg²⁺-free condition. An example of this reduction in sIPSPs, which usually lasted between 4 and 20 s, is illustrated in Fig. 1D, where there is a clear decrease in sIPSPs after each evoked burst discharge lasting ~20 s. This figure also demonstrates that the decrease in sIPSPs was not due to a change in pyramidal cell conductance after the epileptiform burst as the responses to brief hyperpolarizing current injections were unchanged after the burst.

This suppression of sIPSPs is further illustrated in Fig. 2A, which shows consecutive traces of synaptic activity before and after a spontaneous epileptiform discharge (*). In this cell, large sIPSPs were observed before the burst and were suppressed dramatically immediately after the burst response but gradually recovered over the next 8-10 s. The mean reduction in sIPSP magnitude (see METHODS) after a single spontaneous burst was 31 ± 4.1% (n = 7). There was no significant change in input conductance after a single spontaneous or evoked burst (<5%, n = 4).

View larger version (20K): [in this window] [in a new window]   FIG. 2. Single burst discharges transiently suppress sIPSPs. A: 4 consecutive traces (4-s duration) show a single spontaneous epileptiform burst response (*, trace 2) induced by perfusion with Mg2+-free saline. Burst discharge is shown in full below. After the burst, there is a marked suppression of sIPSPs for 4-8 s. Magnitude of integrated sIPSPs progressively recovers to preburst levels. In B, the scatter plot shows there is no correlation (P > 0.05) between the magnitude of sIPSP suppression after an epileptiform burst discharge and the size of the burst response. Changes in sIPSPs and burst responses are both measured as integral values (mV·ms). (Resting potential 52 mV).

Although there was some variation in the amount of sIPSP suppression observed after a single burst, the magnitude of suppression was not correlated with the small fluctuations in burst size. These results are shown in Fig. 2B, which shows a scatter plot of the burst integral (see METHODS) for three single spontaneous bursts from seven cells (n = 21) versus the percentage change in sIPSPs occurring after each burst. There was no significant correlation between the size of a single burst discharge and the percentage suppression of postburst sIPSPs. However, multiple burst discharges, occurring in quick succession, did result in significant increases in the degree of sIPSP suppression observed (see below, Fig. 6).

Suppression of synaptic activity does not require synaptic input

As illustrated in Fig. 1A, DSI in normal saline usually is induced with a 1-s depolarizing current pulse and occurs in the absence of fast excitatory synaptic input (APV, CNQX in perfusion medium). We, therefore, wished to determine whether the postburst suppression in sIPSPs observed in Mg²⁺-free saline was dependent on synaptic activation or also could be induced after depolarizing current pulses, i.e., whether it, like DSI, required only activation of the postsynaptic cell. Short-duration (100-200 ms) current pulses were employed to mimic more closely the epileptiform burst discharges occurring spontaneously. Injection of 0.5-1.0 nA, 100- to 200-ms current pulses resulted in small depolarizing responses on which action potentials were superimposed. A clear suppression in sIPSPs was observed after each response with a mean reduction in sIPSPs of 54 ± 14% (n = 2).

View larger version (28K): [in this window] [in a new window]   FIG. 3. N-ethylmaleimide (NEM) decreases postburst suppression of sIPSPs. Consecutive traces (4-s duration) in control (A) and NEM (B, 250 µM). A burst discharge (*) in control (Mg2+-free saline) results in a clear suppression of sIPSPs for 4-8 s. With NEM (B), there is a small increase in the overall level of synaptic activity in this cell so that large sIPSPs give rise to action potentials. NEM almost completely blocks the postburst suppression of sIPSPs, which, in this example, occurs without any change in the size of the burst discharge. (Resting potential 60 mV). C: mean (%) reduction in the magnitude of integrated sIPSP in control (Con) and NEM (n = 4). Neither the level of synaptic activity before an epileptiform burst discharge (D) nor the size of the mean burst in NEM (E) were significantly different from control (P = 0.79 and 0.24, respectively). * Significant difference from control responses as determined by Student's t-test.

These results show that both synaptic stimulation and short depolarizing current pulses result in a brief suppression of sIPSPs. This raised the question whether this suppression involved the same mechanisms as have been described previously for DSI (Pitler and Alger 1992a, 1994).

At a concentration of 50 µM, 4-AP, which blocks a transient voltage-dependent K current (I_D) in hippocampus (Storm 1988) and increases neurotransmitter release (Tong and Jahr 1994), blocks DSI (Alger et al. 1996). Attempts to block the postburst suppression of sIPSPs with 10-50 µM 4-AP were, however, unsuccessful. By itself 4-AP induces epileptiform burst discharges in CA3 (Perreault and Avoli 1991; Rutecki et al. 1987), and application of 10-50 µM 4-AP to Mg²⁺-free saline dramatically increased burst frequency such that cells fired with a burst frequency of 1-3 s, making it impossible to study changes in sIPSPs.

Postburst suppression of sipsps is reduced by NEM

DSI is not observed in slices from hippocampi of pertussis-toxin-treated animals (Pitler and Alger 1994), suggesting that a pertussis-toxin-sensitive G protein may be involved in the DSI signal pathway. In addition NEM, a sulfhydryl alkylating agent that blocks pertussis-toxin-sensitive G protein actions (Shapiro et al. 1994), recently has been shown to block DSI of both spontaneous and evoked IPSCs in the hippocampus via an apparent presynaptic mechanism (Morishita et al. 1997) (see DISCUSSION).

The effects of NEM on the postburst suppression of sIPSPs was investigated, and Fig. 3 (A and B) illustrates the effects of NEM (250 µM) on one cell. Consecutive traces in control Mg²⁺-free saline (A) show a marked suppression of sIPSPs lasting 4-8 s after a spontaneous burst discharge (*). After bath application of NEM (B), this suppression is abolished almost completely. In nearly all cells NEM caused an initial decrease, followed by an increase, in the integrated sIPSP magnitude. For the cell in B, a few large sIPSPs gave rise to action potentials in NEM. In this cell, the suppression of sIPSPs occurred without any change in the size of the burst response. The changes in activity seen in NEM were associated with only very small changes in input resistance (less than ±12%, n = 4). In two additional cells, the effects of NEM on input resistance were assessed in normal saline with addition of 10 µM bicuculline to block any change in IPSPs. Under these conditions NEM produced a 10 ± 8.6% change in input resistance. The reduction in sIPSP suppression after a burst discharge in the presence of NEM therefore could not be accounted for by changes in the cells' input resistance. The effects of NEM were irreversible with 45 min of wash.

Figure 3C shows that the mean postburst suppression in sIPSPs before perfusion with NEM was 34.8 ± 6.4%. NEM significantly decreased this suppression to 13.7 ± 4.6(n = 4) and (P = 0.007). The reduction in postburst suppression of sIPSPs in NEM could not be explained by any significant change in the baseline level of synaptic activity. The overall level of synaptic activity, measured during a5-s window before burst onset, was not significantly different (P = 0.79) for the four cells in NEM (Fig. 3D). In addition, the reduction in postburst suppression was not a consequence of changes in the size of the burst responses. Two cells showed no change in burst size in NEM, whereas in the other two, the burst increased, and the mean (n = 4) burst integrals for control and NEM were 3.5 ± 0.6 and 6.9 ± 1.8 mV·ms (P = 0.24), respectively (Fig. 3E). The reduction in sIPSP suppression was not, therefore, attributable to a decrease in burst magnitude. The results show that NEM blocked the postburst suppression of sIPSPs in a manner consistent with the abolition of DSI shown previously (Morishita et al. 1997). This suggests that the reduction in sIPSPs seen after a burst response also may involve a pertussis-toxin-sensitive G protein.

Postburst suppression of sIPSPs requires an increase in postsynaptic calcium

DSI in normal saline is blocked when 100 mM EGTA is included in the recording solution (Pitler and Alger 1992a), suggesting that, although the expression of DSI represents a presynaptic mechanism, its induction is dependent on an increase in postsynaptic Ca²⁺. To determine whether the reduction in synaptic activity after a burst was also Ca²⁺-dependent, recordings were made with 100 mM EGTA in the electrode solution. The ability of EGTA to leak into the cell was confirmed by a marked decrease in spike accommodation seen in pyramidal cells (in the absence of carbachol) in response to 1-s depolarizing pulses (Madison and Nicoll 1982).

It was hoped initially that it would be possible to observe a postburst suppression in sIPSPs immediately after cell impalement and subsequently to see that suppression become blocked progressively as EGTA leaked into the cell. This was not, however, possible because, by the time the cell had stabilized adequately to allow data recording, sufficient EGTA appeared to have leaked already into the cell. Group comparisons therefore have been made between neurons recorded with and without EGTA in the recording electrode. These results are illustrated in Fig. 4A, which shows that the mean suppression of sIPSPs after a single spontaneous burst in neurons recorded without EGTA was 31 ± 4.1% (n = 7) compared with 7 ± 5.7% with EGTA (n = 6).The difference between these two groups was significant(P = 0.02).

View larger version (14K): [in this window] [in a new window]   FIG. 4. Ethylene glycol-bis(-aminoethyl ether)-N,N,N,N-tetraacetic acid (EGTA) blocks sIPSP suppression after a single epileptiform burst discharge. A: mean suppression of sIPSPs for cells recorded in control (Con) in the absence of EGTA, and those with 100 mM EGTA in the recording electrode. The presence of EGTA significantly decreased the reduction in sIPSPs seen after a single spontaneous burst from 31 ± 4.1% (n = 7) to 7 ± 5.7% (n = 6). Neither burst magnitude nor the overall level of synaptic activity was significantly different in EGTA (P = 0.2 and 0.3, respectively). * Significant difference from control responses as determined by Student's t-test.

As with NEM, it was important to exclude the possibility that these results may simply reflect a decrease in the overall level of synaptic activity or size of the burst responses between these two groups. A comparison of the burst integral for cells with and without EGTA included in the recording electrode is illustrated in Fig. 4B. The mean burst integrals were 4.5 ± 0.7 and 6.1 ± 0.9 mV·ms for neurons recorded in the absence of EGTA and those with EGTA, respectively. This difference was not significant (P = 0.2). In Fig. 4C, the mean levels of synaptic activity before the bursts in the two groups were compared and revealed no significant difference between those cells recorded with, and those without, EGTA in the electrode (P = 0.35).

For those neurons recorded with EGTA in the electrode, the presence of DSI was not initially demonstrated in normal saline with APV, CNQX, and carbachol as in Fig. 1. It is, therefore, conceivable that the EGTA group might have included some of the small proportion of units in which no DSI is evident. However, for all of the cells recorded in this study with EGTA, although a single spontaneous burst produced little or no suppression of sIPSPs, multiple evoked or spontaneous bursts occurring in quick succession could elicit a reduction in sIPSPs. The reduction in synaptic activity observed after either a single or multiple burst discharge without EGTA is shown in Fig. 5, A and B. In this cell (A), a single spontaneous burst produced a decrease of 49% in sIPSPs after the burst discharge. An even more pronounced suppression of sIPSPs (63%) occurred after a series of three spontaneous bursts (Fig. 5B). In contrast, an example of a cell recorded with 100 mM EGTA in the electrode shows no significant suppression of sIPSPs (<10%) after a single spontaneous burst (Fig. 5C) although a 54% reduction after a series of four spontaneous bursts (Fig. 5D) was observed. As with the sIPSP reduction seen after a single burst, the decreases in sIPSPs occurring after multiple burst responses were also not attributable to changes in input resistance. Changes in conductances after multiple burst responses were measured in three cells, and the average change in input resistance was only 2.8 ± 5.5% with no cell showing a >10% change.

View larger version (27K): [in this window] [in a new window]   FIG. 5. Comparison of single vs. multiple burst discharges on sIPSP suppression. A: 4 consecutive traces (4-s duration) for 2 cells: 1 recordedwithout EGTA in the electrode (resting potential 55 mV; A and B) and 1 with 100 mM EGTA in the electrode (resting potential 52 mV; C and D). A: for the cell recorded without EGTA, a single burst produced a clear suppression of sIPSPs, which was enhanced after 3 spontaneous burst discharges. B: with 100 mM EGTA in the electrode, there is no suppression of sIPSPs after a single spontaneous burst, but the same cell was capable of showing a reduction in sIPSPs after a series of 4 spontaneous epileptiform bursts.

These results demonstrate that, although EGTA significantly reduced the sIPSP suppression seen after a single burst, these cells were still capable of exhibiting some reduction of sIPSPs. The possible reasons EGTA failed to block all sIPSP suppression completely are discussed later (see DISCUSSION).

These results show that the reduction in sIPSPs seen after a single burst response was reduced as a result of chelating postsynaptic calcium influxes with EGTA. Like DSI, therefore, the suppression of sIPSPs seen after an epileptiform burst is dependent on (an increase in) postsynaptic calcium.

Suppression of sIPSPs is related to the number of burst discharges

The data in Fig. 5 suggest that there may be a relationship between the number of bursts occurring in quick succession and the degree of sIPSP suppression. To assess this, we have compared the degree of sIPSP suppression seen after a single burst discharge with that after a series of multiple burst discharges, in cells recorded with and without EGTA in the electrode. The results of this analysis are shown in Fig. 6. The numbers above the bars on the histogram indicate, first, the number of cells recorded and, second, the total number of bursts measured. Where more than one burst response was obtained for a cell, the mean of these responses was taken. For cells recorded without EGTA (n = 7) (Fig. 6A), there is a progressive increase in sIPSP suppression from 31 to 49% as the number of epileptiform burst discharges in the response increased from one to three or four. Although the difference in the magnitude of suppression between one and two burst responses was not significantly different(P = 0.07), there was significantly greater sIPSP suppression after the multiple, three- to four-burst responses than that seen after a single burst (P = 0.01). With 100 mM EGTA in the electrode there was, as also shown earlier (Figs. 4 and 5), no significant sIPSP suppression after a single burst response (7%). However, all cells exhibited some form of multiple burst discharges. Under these circumstances either two or three to four burst responses resulted in a sIPSP suppression of 32 ± 9.3% and 28 ± 4.7%, respectively. However, the degree of sIPSP suppression seen with three- to four-burst responses with EGTA in the electrode was significantly less than that seen after similar multiple (3-4) bursts recorded without EGTA (P = 0.01). These results suggest that, despite its efficacy in preventing sIPSP suppression after a single burst, EGTA can only reduce, but not prevent, sIPSP suppression after multiple bursts.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The results show that a transient depression of GABA_A-receptor-mediated sIPSPs can occur after a single epileptiform burst discharge induced in Mg²⁺-free saline. This suppression of sIPSPs shares many of the characteristics of DSI that have previously been described (Alger et al. 1996; Morishita et al. 1997; Pitler and Alger 1992a, 1994) and demonstrates that a DSI-like process may be induced by epileptiform burst activity.

Reducing extracellular Mg²⁺ enhances neuronal excitation in a number of preparations, including the hippocampus, and causes epileptiform burst discharges (Anderson et al. 1986; Avoli et al. 1987; Mody et al. 1987; Neuman et al. 1989; Schneiderman and MacDonald 1987; Tancredi et al. 1990). In this study, removal of extracellular Mg²⁺ led to the generation of epileptiform burst discharges, the magnitudes of which were highly stable over time (Fig. 1C). Whittington et al. (1995) have described a progressive rundown of IPSC amplitude in Mg²⁺-free saline that was greatest after 3-4 h. Such long recording times were not carried out in the present study in which the maximum time in Mg²⁺-free saline for most cells was ~1 h. A comparison of the level of synaptic activity between 10 and 60 min after commencing perfusion with Mg²⁺-free saline showed no significant difference.

Although many mechanisms could contribute to the generation of burst discharges induced by perfusion with Mg²⁺-free saline (Frankenhaeuser and Hodgkin 1957), the major effect of Mg²⁺-free saline was probably to relieve the voltage-dependent Mg²⁺ block of the NMDA receptor (Mayer et al. 1984). Burst discharges recorded in Mg²⁺-free saline were reduced or abolished after application of 50-100 µM APV (data not shown) as has been shown in previous studies in CA1 (Tancredi et al. 1990; Westerhoff et al. 1995). Adding back 2 mM Mg²⁺ to the bathing solution also rapidly and completely abolished all epileptiform activity. In our experiments, extracellular Ca²⁺ was increased from 2.5 to 3.5 mM, which helped to control the frequency of bursting, probably by partially compensating for the decreased Mg²⁺ concentration.

The induction of DSI is dependent on a Ca²⁺ influx into the postsynaptic cell (Alger and Pitler 1995; Pitler and Alger 1992a). Similarly the suppression of sIPSPs after a burst discharge was dependent on intracellular Ca²⁺. After intracellular injection of 100 mM EGTA, a single spontaneous burst discharge failed to reduce sIPSPs significantly, with a mean reduction of 7 ± 5.7% compared with 31 ± 4.1% in those neurons recorded in the absence of EGTA (Fig. 4). All neurons with EGTA (n = 6) in which a single burst failed to elicit any sIPSP suppression, however, were capable of exhibiting a reduction after multiple burst discharges (Figs. 5 and 6).

Although this and other studies (Llano et al. 1991; Pitler and Alger 1992a) have shown clearly that the induction of DSI depends on Ca²⁺ influx, the precise relationship between DSI and Ca²⁺ is currently unclear. The small variations in the size of a single burst occurring within a cell did not correlate with the observed fluctuations in the magnitude of postburst suppression (Fig. 2B). However, increasing the number of bursts occurring in quick succession did result in a greater degree of sIPSP suppression (Figs. 5 and 6). This suggests that the magnitude of sIPSP suppression may be related directly to the amount of Ca²⁺ influx. Pitler and Alger (1992a) made a similar observation when DSI was elicited by a train of action potentials. They found that increasing the number of action potentials in the train increased both the magnitude and duration of the IPSP suppression.

The magnitude of the reduction in sIPSPs seen after a burst response (31%) was similar to that observed after depolarizing voltage steps. In contrast, the duration of the sIPSP suppression was consistently shorter (4-20 s) compared with that seen after a depolarizing voltage pulse(30-60 s). The duration of postburst suppression of sIPSPs was, however, consistent with that seen after a train of action potentials (Pitler and Alger 1992a). This may be attributable to differences in the magnitude or distribution of the Ca²⁺ influx occurring under the different experimental protocols (see further). The shorter duration of postburst suppression of sIPSPs was unlikely to be due to any impairment in Ca²⁺ regulation in Mg²⁺-free saline. Connor et al. (1988) found that resting intracellular Ca²⁺ levels recorded using fura-2 in isolated CA1 neurons in Mg²⁺-free saline were not significantly different from those obtained in normal saline, although this does not exclude the possibility that changes in dynamic Ca²⁺ regulation still could have occurred.

There are a number of possible reasons EGTA in the current study failed to block the postburst suppression of sIPSPs after multiple bursts in contrast to the complete prevention of DSI by EGTA observed after a train of action potentials produced by depolarizing current pulses (Pitler and Alger 1994). This may simply reflect the slow calcium chelation by EGTA (Tsien 1980) so that sufficient Ca²⁺ remains unbound and able to trigger the mechanism responsible for the postburst suppression of sIPSPs. Alternatively there may be differences in the spatial and temporal patterns of calcium channel activation and Ca²⁺ influx after a synaptic input compared with that after intrasomatic current injection. Miyakawa et al. (1992) found that increases in intracellular Ca²⁺ generated by synaptic stimulation were distributed in more distal regions of the dendritic tree than the elevation produced by current injection. It is, therefore, possible that in the current study EGTA was unable to gain access to the region in which the Ca²⁺ signal is initiated at least at a sufficient concentration to block the presumably larger influx of Ca²⁺ after a multiple burst response.

The observed suppression of sIPSPs after a burst could be due to a postsynaptic down regulation of GABA_A responses after Ca²⁺ influx via the NMDA receptor (Chen and Wong 1995; Chen et al. 1990; Inoue et al. 1986; Stelzer and Shi 1994; Stelzer et al. 1988). Increased intracellular Ca²⁺ suppresses postsynaptic GABA_A responses via a dephosphorylation-dependent process (Chen et al. 1990) in acutely isolated hippocampal neurons. However, these mechanisms are not involved in DSI as previously described (Alger et al. 1996; Morishita et al. 1997; Pitler and Alger 1992a, 1994) because they involve a decrease in postsynaptic responsiveness to GABA after Ca²⁺ influx via the NMDA receptor that is blocked by APV. APV does not block DSI, and in either the hippocampus or the cerebellum there is no decrease in postsynaptic GABA_A receptor sensitivity as determined by either iontophoretic application of GABA (Llano et al. 1991; Pitler and Alger 1992a) or by analysis of spontaneous TTX-resistant miniature IPCSs (Alger et al. 1996; Llano et al. 1991). In the present study, however, at least part of the burst response reflected activation of NMDA receptors so it is possible the reduction in sIPSPs is due to increased intracellular Ca²⁺ resulting in a modification of the postsynaptic GABA_A receptors (Chen and Wong 1995; Stelzer and Shi 1994). We currently cannot exclude the possibility that this mechanism also may account for the suppression of sIPSPs seen after an epileptiform burst discharge.

Alternatively sIPSP suppression could involve a presynaptic modulation of GABA release as proposed for DSI (Pitler and Alger 1992a, 1994). Although the induction of DSI and the post-burst suppression of sIPSPs are both dependent on an influx of Ca²⁺ into the postsynaptic cell, the expression of DSI is presynaptic (Alger and Pitler 1995; Alger et al. 1996; Morishita et al. 1997). One agent that blocks DSI, probably via presynaptic mechanisms, is NEM (Morishita et al. 1997). Morishita et al. (1987) showed that NEM could be a useful tool for investigating G-protein-mediated events in the CNS. They found that, in addition to blocking DSI, bath application of NEM (250 µM) increased sIPSP activity and increased the frequency of TTX-insensitive miniature IPSCs without affecting iontophoretic GABA_A responses. In addition, block of postsynaptic G proteins by omitting GTP from the recording pipette does not abolish DSI. These observations are consistent with a presynaptic site of action and suggest that NEM blocks DSI by interfering with presynaptic G-protein mediated events.

A block of postburst sIPSP suppression by NEM would be evidence for a presynaptic modification and would suggest a DSI-like process was involved in the suppression of sIPSPs. Our results show that NEM caused a significant reduction in the suppression of sIPSPs after both spontaneous and evoked epileptiform discharges (Fig. 3, B and C) that was not caused by changes in the overall level of synaptic activity or the burst magnitude (Fig. 3, D and E). Application of NEM initially decreased baseline activity before increasing it in all cells tested. This initial decrease was not seen in experiments in which EPSCs were blocked by APV and CNQX. It is, therefore, possible that a complex interplay of excitatory and inhibitory interactions could account for this observation.

The function of this transient reduction in sIPSPs remains to be determined, but it could, for example, play an important role in determining pyramidal cell excitability. DSI can enhance pyramidal cell excitability as indicated by an increase in the excitatory postsynaptic currents (EPSCs) recorded during DSI (Wagner and Alger 1996). These changes in EPSCs were clearly attributable to changes in the IPSCs rather than to a direct effect on the EPSC. In addition, a reduction in inhibition after a single burst discharge could facilitate the induction of LTP. Huerta and Lisman (1995) have found that a single burst elicited at the peak of a theta oscillation was sufficient to evoke a long-lasting potentiation of extracellularly recorded postsynaptic potentials. This potentiation only occurred when theta oscillations were induced by 50 µM carbachol, as in the absence of carbachol there was no change in the EPSP. Muscarinic agonists, by blocking several voltage-activated K currents, cause a depolarized membrane potential and an increase in neuronal excitability (e.g., Benardo and Prince 1982; Cole and Nicoll 1983; Gahwiler and Brown 1985; Halliwell and Adams 1982; Storm 1989). In addition, carbachol also causes an increase in spontaneous IPSCs (Alger et al. 1996; Martin and Alger 1996; Pitler and Alger 1992b) that could serve to prevent the uncontrolled excitation of pyramidal cells. It is well established that the induction of LTP is facilitated by a reduction in inhibition (Wigstrom and Gustafsson 1983). Therefore, the suppression of sIPSPs observed after a burst discharge could provide a means of producing a brief reduction in inhibition that would allow the temporal integration of excitatory inputs. An investigation into whether this suppression of sIPSPs occurs after normal burst discharges in CA3 will be important to elucidate its physiological significance. In addition, further work is required to establish whether the suppression of sIPSPs after an epileptiform burst discharge reflects activation of the same mechanisms as those involved in DSI or the postsynaptic Ca²⁺-dependent dephosphorylation of the GABA_A receptor (Chen and Wong 1995).

	ACKNOWLEDGEMENTS

  We thank L. A. Martin, S. E. Mason, R. Lenz, and W. Morishita for helpful comments on this manuscript and E. Elizabeth for editorial assistance.

  This work was supported by National Institute of Neurological Disorders and Stroke Grants NS-22010 and NS-30219 to B. E. Alger.

	FOOTNOTES

  Address reprint requests to B. E. Alger.

  Received 2 July 1997; accepted in final form 6 October 1997 .

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ALGER, B. E., NICOLL, R. A. Epileptiform burst afterhyperpolarization: calcium-dependent potassium potential in hippocampal CA1 pyramidal cells. Science 210: 1122-1124, 1980.[Abstract/Free Full Text]
• ALGER, B. E., PITLER, T. A. Retrograde signaling at GABA_A-receptor synapses in the mammalian CNS. Trends Neurosci. 18: 333-340, 1995.[Medline]
• ALGER, B. E., PITLER, T. A., WAGNER, J. J., MARTIN, L. A., MORISHITA, W., KIROV, S. A., LENZ, R. A. Retrograde signaling in depolarization-induced suppression of inhibition in rat hippocampal CA1 cells. J. Physiol. (Lond.) 496: 197-209, 1996.[Abstract/Free Full Text]
• ANDERSON, W. W., LEWIS, D. V., SWARTZWELDER, H. S., WILSON, W. A. Magnesium-free medium activates seizure-like events in the rat hippocampal slice. Brain Res. 398: 215-219, 1986.[Medline]
• AVOLI, M., LOUVEL, J., PUMAIN, R., OLIVIER, A. Seizure-like discharges induced by lowering [Mg²⁺]_o in the human epileptogenic neocortex maintained in vitro. Brain Res. 417: 199-203, 1987.[Medline]
• BENARDO, L. S. Recruitment of inhibition by enhanced activation of synaptic NMDA responses in the rat cerebral cortex. Brain Res. 627: 314-324, 1993.[Medline]
• BENARDO, L. S., PRINCE, D. A. Ionic mechanisms of cholinergic excitation in mammalian hippocampal pyramidal cells. Brain Res. 249: 333-344, 1982.[Medline]
• BUCKLE, P. J., HAAS, H. L. Enhancement of synaptic transmission by 4-aminopyridine in hippocampal slices of the rat. J. Physiol. (Lond.) 326: 109-122, 1982.[Abstract/Free Full Text]
• CHEN, Q. X., STELZER, A., KAY, A. R., WONG, R.K.S. GABA_A receptor function is regulated by phosphorylation in acutely dissociated guinea-pig hippocampal neurones. J. Physiol. (Lond.) 420: 207-221, 1990.[Abstract/Free Full Text]
• CHEN, Q. X., WONG, R.K.S. Suppression of GABA_A receptor responses by NMDA application in hippocampal neurons acutely isolated from the adult guinea-pig. J. Physiol. (Lond.) 482: 353-362, 1995.[Abstract/Free Full Text]
• COLE, A. E., NICOLL, R. A. Acetylcholine mediates a slow synaptic potential in hippocampal pyramidal cells. Science 221: 1299-1301, 1983.[Abstract/Free Full Text]
• CONNOR, J. A., WADMAN, W. J., HOCKBERGER, P. E., WONG, R.K.S. Sustained dendritic gradients of Ca²⁺ induced by excitatory amino acids in CA1 hippocampal neurons. Science 240: 649-653, 1988.[Abstract/Free Full Text]
• FRANKENHAEUSER, B., HODGKIN, A. L. The action of calcium on the electrical properties of squid axons. J. Physiol. (Lond.) 137: 218-244, 1957.
• GAHWILER, B. H., BROWN, D. A. Functional innervation of cultured hippocampal neurones by cholinergic afferents from co-cultured septal explants. Nature 313: 577-579, 1985.[Medline]
• GLITSCH, M., LLANO, I., MARTY, A. Glutamate as a candidate retrograde messenger at interneurone-Purkinje cell synapses of rat cerebellum. J. Physiol. (Lond.) 497: 531-537, 1996.[Abstract/Free Full Text]
• HALLIWELL, J. V., ADAMS, P. R. Voltage-clamp analysis of muscarinic excitation in hippocampal neurons. Brain Res. 250: 71-92, 1982.[Medline]
• HUERTA, P. T., LISMAN, J. E. Bidirectional synaptic plasticity induced by a single burst during cholinergic theta oscillation in CA1 in vitro. Neuron 15: 1053-1063, 1995.[Medline]
• INOUE, M., OOMURA, Y., YAKUSHIJI, T., AKAIKE, N. Intracellular calcium ions decrease the affinity of the GABA receptor. Nature 324: 156-158, 1986.[Medline]
• KANDEL, E. R., SPENCER, W. A., BRINLEY, F. J. JR Electrophysiology of hippocampal neurons. I. Sequential invasion and synaptic organization. J. Neurophysiol. 24: 225-242, 1961.[Free Full Text]
• LLANO, I., LERESCHE, N., MARTY, A. Calcium entry increases the sensitivity of cerebellar Purkinje cells to applied GABA and decreases inhibitory synaptic currents. Neuron 6: 565-574, 1991.[Medline]
• MADISON, D. V., NICOLL, R. A. Noradrenaline blocks accommodation of pyramidal cell discharge in the hippocampus. Nature Lond. 299: 636-638, 1982.[Medline]
• MARTIN, L. A., ALGER, B. E. Muscarinic receptor activation of interneurons susceptible to depolarization-induced suppression of inhibition (DSI). Soc. Neurosci. Abstr. 22: 1989, 1996.
• MAYER, M. L., WESTBROOK, G. L., GUTHRIE, P. B. Voltage-dependent block by Mg²⁺ of NMDA responses in spinal cord neurones. Nature 309: 261-263, 1984.[Medline]
• MIYAKAWA, H., ROSS, W. N., JAFFE, D., CALLAWAY, J. C., LASSER-ROSS, N., LISMAN, J. E., JOHNSTON, D. Synaptically activated increases in Ca²⁺ concentration in hippocampal CA1 pyramidal cells are primarily due to voltage-gated Ca²⁺ channels. Neuron 9: 1163-1173, 1992.[Medline]
• MODY, I., LAMBERT, J.D.C., HEINEMANN, U. Low extracellular magnesium induces epileptiform activity and spreading depression in rat hippocampal slices. J. Neurophysiol. 57: 869-888, 1987.[Abstract/Free Full Text]
• MORISHITA, W., KIROV, S. A., PITLER, T. A., MARTIN, L. A., LENZ, R. A., ALGER, B. E. N-Ethylmaleimide blocks depolarization-induced suppression of inhibition and enhances GABA release in the rat hippocampal slice in vitro. J. Neurosci. 17: 941-950, 1997.[Abstract/Free Full Text]
• NEUMAN, R. S., CHERUBINI, E., BEN-ARI, Y. Endogenous and network bursts induced by N-methyl-D-aspartate and magnesium free medium in the CA3 region of the hippocampal slice. Neuroscience 28: 393-399, 1989.[Medline]
• NICOLL, R. A., ALGER, B. E. A simple chamber for recording from submerged brain slices. J. Neurosci. Methods 4: 153-156, 1981.[Medline]
• PERREAULT, P., AVOLI, M. Physiology and pharmacology of epileptiform activity induced by 4-aminopyridine in rat hippocampal slices. J. Neurophysiol. 65: 771-785, 1991.[Abstract/Free Full Text]
• PITLER, T. A., ALGER, B. E. Postsynaptic spike firing reduces synaptic GABA_A responses in hippocampal pyramidal cells. J. Neurosci. 12: 4122-4132, 1992a.[Abstract]
• PITLER, T. A., ALGER, B. E. Cholinergic excitation of GABA interneurones in the rat hippocampal slice. J. Physiol. (Lond.) 450: 127-142, 1992b.[Abstract/Free Full Text]
• PITLER, T. A., ALGER, B. E. Depolarization-induced suppression of GABAergic inhibition in rat hippocampal pyramidal cells: G protein involvement in a presynaptic mechanism. Neuron 13: 1447-1455, 1994.[Medline]
• RUTECKI, P. A., LEBEDA, F. J., JOHNSTON, D. 4-Aminopyridine produces epileptiform activity in hippocampus and enhances synaptic excitation and inhibition. J. Neurophysiol. 57: 1911-1924, 1987.[Abstract/Free Full Text]
• SCHNEIDERMAN, J. H., MACDONALD, J. F. Effects of reduced magnesium on hippocampal synchrony. Brain Res. 410: 174-178, 1987.[Medline]
• SCHWARTZKROIN, P. A. In: Epilepsy: Models, Mechanisms, and Concepts., . Cambridge, UK: Cambridge Univ. Press, 1993, p. 1
• SCHWARTZKROIN, P. A., PRINCE, D. A. Cellular and field potential properties of epileptogenic hippocampal slices. Brain Res. 147: 117-130, 1978.[Medline]
• SCHWARTZKROIN, P. A., STAFSTROM, C. E. Effects of EGTA on the calcium-activated afterhyperpolarization in hippocampal CA3 pyramidal cells. Science 210: 1125-1126, 1980.[Abstract/Free Full Text]
• SHAPIRO, M. S., WOLLMUTH, L. P., HILLE, B. Modulation of Ca²⁺ channels by PTX-sensitive G-proteins is blocked by N-ethylmaleimide in rat sympathetic neurons. J. Neurosci. 14: 7109-7116, 1994.[Abstract]
• STELZER, A., KAY, A. R., WONG, R.K.S. GABA_A-receptor function in hippocampal cells is maintained by phosphorylation factors. Science 241: 339-341, 1988.[Abstract/Free Full Text]
• STELZER, A., SHI, H. Impairment of GABA_A receptor function by N-methyl-D-aspartate-mediated calcium influx in isolated CA1 pyramidal cells. Neuroscience 62: 813-828, 1994.[Medline]
• STELZER, A., SIMON, G., KOVACS, G., RAI, R. Synaptic disinhibition during maintenance of long-term potentiation in the CA1 hippocampal subfield. Proc. Natl. Acad. Sci. USA 91: 3058-3062, 1994.[Abstract/Free Full Text]
• STELZER, A., SLATER, N. T., TEN BRUGGENCATE, G. Activation of NMDA receptors blocks GABAergic inhibition in an in vitro model of epilepsy. Nature 326: 698-701, 1987.[Medline]
• STORM, J. F. Temporal integration by a slowly inactivating K⁺ current in hippocampal neurons. Nature 336: 379-381, 1988.[Medline]
• STORM, J. F. An afterhyperpolarization of medium duration in rat hippocampal pyramidal cells. J. Physiol. (Lond.) 409: 171-190, 1989.[Abstract/Free Full Text]
• TANCREDI, V., HWA, G.G.C., ZONA, C., BRANCATI, A., AVOLI, M. Low magnesium epileptogenesis in the rat hippocampal slice: electrophysiological and pharmacological features. Brain Res. 511: 280-290, 1990.[Medline]
• TONG, G., JAHR, C. E. Multivesicular release from excitatory synapses of cultured hippocampal neurons. Neuron 12: 51-59, 1994.[Medline]
• TSIEN, R. Y. New calcium indicators and buffers with high selectivity against magnesium and protons: design, synthesis, and properties of prototype structures. Biochemistry 19: 2396-2404, 1980.[Medline]
• VINCENT, P., ARMSTRONG, C. M., MARTY, A. Inhibitory synaptic currents in rat cerebellar Purkinje cells: modulation by postsynaptic depolarization. J. Physiol. (Lond.) 456: 453-471, 1992.[Abstract/Free Full Text]
• VINCENT, P., MARTY, A. Neighboring cerebellar Purkinje cells communicate via retrograde inhibition of common presynaptic interneurons. Neuron 11: 885-893, 1993.[Medline]
• WAGNER, J. J., ALGER, B. E. Increased neuronal excitability during depolarization-induced suppression of inhibition in rat hippocampus. J. Physiol. (Lond.) 495: 107-112, 1996.[Abstract/Free Full Text]
• WESTERHOFF, C.H.A., DOMANN, R., WITTE, O. W. Inhibitory mechanisms in epileptiform activity induced by low magnesium. Pflügers Arch. 430: 238-245, 1995.[Medline]
• WHITTINGTON, M. A., TRAUB, R. D., JEFFERYS, J.G.R. Erosion of inhibition contributes to the progression of low magnesium bursts in rat hippocampal slices. J. Physiol. (Lond.) 486: 723-734, 1995.[Abstract/Free Full Text]
• WIGSTROM, H., GUSTAFSSON, B. Facilitated induction of hippocampal long-lasting potentiation during blockade of inhibition. Nature 301: 603-604, 1983.[Medline]
• WONG, R.K.S., PRINCE, D. A. Dendritic mechanisms underlying penicillin-induced epileptiform activity. Science 204: 1228-1231, 1979.[Abstract/Free Full Text]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society