- Email: [email protected]
calmodulin-dependent protein kinase II
Neuropharmacology 40 (2001) 203–211 www.elsevier.com/locate/neuropharm
Epileptiform activity and EPSP-spike potentiation induced in rat hippocampal CA1 slices by repeated high-K +: involvement of ionotropic glutamate receptors and Ca2+/calmodulin-dependent protein kinase II Alexey Semyanov 1, Oleg Godukhin
*
Institute of Theoretical and Experimental Biophysics, Russian Academy of Sciences, Pushchino, Moscow Region, 142292, Russia Received 24 January 2000; received in revised form 20 June 2000; accepted 2 August 2000
Abstract We have previously demonstrated that repeated brief increases in extracellular K + (K +o) induce a hyperexcitability in CA1 pyramidal cells that persists for a long time after the final application of K + [Neurosci. Lett. 223 (1997) 177; Epilepsy Research (2000) 75]. This epileptiform activity, which was associated with a lasting excitatory postsynaptic potential (EPSP)-spike potentiation, presented some of the characteristic features of traditional in vivo kindling. We have also found that Ca2+ influx through Ltype voltage-sensitive Ca2+ channels is essential for the development of both in vitro kindling and EPSP-spike potentiation. The aims of this study were to investigate the involvement of ionotropic glutamate receptors, especially those of the NMDA subtype, and the requirement for Ca2+/calmodulin-dependent protein kinase II (CaMKII) in these phenomena. Field EPSPs with presynaptic fibre volleys from the stratum radiatum, and population spikes from the stratum pyramidale, were recorded in the CA1 area of rat hippocampal slices in response to electrical stimulation of the Schaffer collateral/commissural fibres. Repeated (three episodes) brief (30 s) increases in extracellular K + induced a sustained decrease in the threshold for development of evoked epileptiform discharges (i.e. an in vitro kindling-like state) and a lasting potentiation of the EPSP-spike transfer in CA1 pyramidal neurons (EPSP-spike potentiation). The selective antagonist of NMDA receptors, APV (50 µM), blocked the EPSP-spike potentiation, depressed the induction phase of the in vitro kindling-like state, and blocked the maintenance phase of this state. In contrast to APV, the blockade of AMPA/kainate receptors by CNQX (10 µM) had no effect. Like APV, KN62 (3 µM), a selective membrane permeable inhibitor of CaMKII, blocked the EPSP-spike potentiation and the maintenance phase of the in vitro kindling-like state. Our previous and present results therefore demonstrate that Ca2+ influxes through L-type voltage-dependent—and NMDA receptor-dependent—Ca2+ channels contribute differentially to the development of an in vitro kindling-like state, and both induce EPSP-spike potentiation in CA1 hippocampal pyramidal cells in response to repeated brief increases in K +o. It is suggested that these effects of intracellular Ca2+ on the maintenance phase of the in vitro kindling-like state and EPSP-spike potentiation are mediated by CaMKII-dependent mechanisms. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Hippocampal CA1 slices; Field potentials; Epileptiform activity; EPSP-spike potentiation; NMDA glutamate receptor; Ca2+/calmodulindependent protein kinase II
1. Introduction Kindling (i.e. progressive reduction of threshold for induction of seizure activity) is an important model of * Corresponding author. Fax: +7-967-790-553. E-mail addresses: [email protected] (A. Semyanov), [email protected] (O. Godukhin). 1 Present address: Department of Clinical Neurology, Institute of Neurology, UCL, Queen Square, London WC1N 3BG, UK. Fax: +4420-7278-5616.
epileptogenesis, but in vivo investigations have been unable so far to clarify the processes leading to hyperexcitability (Goddard et al., 1969; Lothmann and Williamson, 1994; Della Paschoa et al., 1997). Hippocampal slices are widely used to study the cellular mechanisms of epileptogenesis (Anderson et al., 1986; Mody et al., 1987; Wilson et al., 1992; Fleck et al., 1992; Bernard and Wheal, 1995), and we have previously demonstrated that repeated (3–6 episodes) brief (30 s) increases in extracellular K + (K +o) induce a hyperexcitability in CA1
0028-3908/01/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 0 8 ( 0 0 ) 0 0 1 4 7 - 7
204
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
pyramidal cells that persists for a long time after the final application of K + (Semyanov and Godukhin, 1997; Semyanov et al., 2000). This epileptiform activity, which is associated with a lasting excitatory postsynaptic potential (EPSP)-spike potentiation, presents some of the characteristic features of traditional in vivo kindling (i.e. repeated pattern of stimulation needed to kindle; progressive reduction of the threshold of evoked epileptiform discharges dependent on the number of K +o episodes applied, and persistence). Furthermore, as demonstrated previously, electrical and audiogenic kindlings in vivo occluded the development of the in vitro kindling-like state induced by repeated increases in K +o CA1 slices (Semyanov et al., 2000). We have also found that Ca2+ influx through L-type voltage-sensitive Ca2+ channels is essential for the development of both in vitro kindling and EPSP-spike potentiation in CA1 hippocampal slices in response to the repeated increases in K +o (Semyanov and Godukhin, 1997). The first aim of this study was to investigate the involvement of ionotropic glutamate receptors, especially those of the NMDA subtype, in these phenomena because NMDA receptors are involved in the induction of both synaptic long-term potentiation (LTP) (Bliss and Lynch, 1988; Kuba and Kumamoto, 1990) and epileptiform activity (Jefferys, 1990; Wilson et al., 1992; Bernard and Wheal, 1995) in CA1 pyramidal neurons. Intracellular Ca2+, through activation of Ca2+/calmodulin-dependent kinase II (CaMKII), is known to play a key role in the long-lasting, activitydependent synaptic modifications that underlie various forms of neuronal plasticity such as LTP and traditional kindling (Wasterlain and Farber, 1984; Malenka et al., 1989; Malinov et al., 1989; Ito et al., 1991; Butler et al., 1993; Lisman, 1994; Zhou et al., 1994; Fukunaga et al., 1995; Huber et al., 1995; Murray et al., 1995; Otmakhov et al., 1997). Therefore, the second aim of this study was to examine the requirement for CaMKII in the development of EPSP-spike potentiation and the in vitro kindling-like state induced by repeated high K + episodes. We have used a pharmacological approach using antagonists of glutamate NMDA and AMPA/kainate receptors, and KN62 as a specific, membrane permeable inhibitor of CaMKII (Tokumitsu et al., 1990). Because synchronization of the cell population is the principal feature of epileptiform discharges, we used an extracellular recording protocol to provide a better insight into the collective behaviour of the cell population. 2. Methods 2.1. Preparation and maintenance of CA1 hippocampal slices All experiments were carried out with male Wistar rats (Animal laboratory of ITEB RAN, Pushchino,
Russia) weighing 140–150 g (n=30). CA1 hippocampal slice preparation and superfusion, and methods for the recording of field excitatory postsynaptic potentials (fEPSPs) with presynaptic fibre volleys (PrVs) from the stratum radiatum and population spikes (PSs) from the stratum pyramidale in response to electric stimulation of the Schaffer collateral/commissural fibres (SC/CF) were as described previously (Semyanov and Godukhin, 1997; Semyanov et al. 1997, 2000). To functionally isolate the CA1 area from CA2 and CA3 areas, an orthogonal cut of the stratum pyramidale extending to the distal edge of the mossy fibre layer was made (Fig. 1a). Slices were placed into a submersion recording chamber and superfused at 2.5 ml/min (32°C). The composition of the standard medium for electrophysiological recording was (in mM): 124 NaCl, 3 KCl, 1.25 KH2PO4, 2 MgSO4, 2 CaCl2, 26 NaHCO3, 10 d-glucose; pH 7.4 was adjusted with 95% O2/5% CO2. 2.2. Extracellular electrophysiology Slices were allowed to equilibrate in the chamber for at least 3 h before data collection. Field responses were recorded from the stratum radiatum and stratum pyramidale of the CA1 area with two glass microelectrodes filled with standard medium (Fig. 1a). 2.3. Synaptic LTP and EPSP-spike potentiation PrV amplitudes, fEPSP slopes and PS amplitudes were obtained from measurements of individual responses to a series of 25 separate stimuli from minimum to maximum stimulus intensities (for PS generation) applied at 10-s intervals. Each series of 25 stimuli was separated from the next by 20 min. The three components of the overall input–output relationships, PrV amplitude versus current intensity, fEPSP slope versus PrV amplitude and PS amplitude versus fEPSP slope were taken as indices of the excitability of the glutamatergic SC/CF fibres, the glutamatergic synaptic transmission efficiency and the EPSP-spike transfer efficiency, respectively (Semyanov and Godukhin, 1997; Semyanov et al., 1997, 2000). The time course of the fEPSP slope for a given PrV amplitude and the time course of the amplitude of the first PS in PS responses for a given fEPSP slope were constructed. Typically, time courses of fEPSP slope were determined within an intermediate range of PrV amplitudes (around 0.5 mV), and time courses of PS amplitude within an intermediate range of fEPSP slopes (around 0.4 mV/ms). However, these parameters were investigated over the full dynamic range of stimulus intensities to avoid errors that could arise if a single arbitrary stimulus intensity was used (McEachern and Shaw, 1996).
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
205
Fig. 1. Representative field potentials recorded in CA1 hippocampal slices in response to a single electrical stimulus applied to the stratum radiatum. (a, left panel) Schematic drawing of a CA1 hippocampal slice showing the position of the stimulation electrode (Stim), and the two sites of extracellular recordings: 1, fEPSP with PrV were recorded in the stratum radiatum (rad); 2, single PS or multiple PS discharges were recorded in the stratum pyramidale (pyr). (a, right panel) Extracellulary recorded field potentials from electrode sites 1 and 2, with changes in the pattern of PS response induced by repeated K +o applications (dashed line, additional PS). sch, Schaffer collateral/commissural fibres. (b) A typical set of PS responses recorded in CA1 hippocampal slices without (Control) and 1 h after 1, 3, or 6 K +o applications. All recordings were made 4 h after the start of slice superfusions. Vertical and horizontal calibration bars are 1 mV and 10 ms, respectively.
2.4. Epileptiform activity
2.5. Pharmacological manipulations
Epileptiform activity was induced by three brief (30 s) episodes of exposure to 20 mM K +o (with equimolar decreases in Na+) applied at 10-min intervals. The appearance of multiple PSs in response to a single electrical stimulus was taken as indicator of the development of epileptiform activity in the CA1 pyramidal neurons (Wheal et al., 1998). Two parameters of such activity were measured: (1) the stimulus intensity (µA) of the appearance of the second (additional) PS (marked by an arrow on the dashed line in Fig. 1a) was characterized as the threshold of generation of an additional PS in PS discharges (TASG); and (2) the number of PSs in the PS discharge (NPS); this parameter was measured for the current intensity of TASG taken prior to the application of K +o episodes (or at the same time points for control experiments: ⫺20 min–0 min) (Semyanov et al., 2000). Fig. 1b shows typical PS responses recorded from CA1 hippocampal slices without (baseline control) and 1 h after 1, 3 and 6 applications of K +. All recordings were made 4 h after the start of slice superfusions.
d,l-(+)-2-Amino-5-phosphonovaleric acid (APV) and 6-cyano-7-nitroquinoxaline-2,3-(1H,4H)dione (CNQX), used as selective antagonists of NMDA and AMPA/kainate receptors, respectively, were purchased from Sigma. CNQX and APV were dissolved in DMSO and distilled water, respectively, prior to dilution in standard medium. APV (50 µM) and CNQX (10 µM, 0.1% DMSO) were applied for 2 min before, during and 2 min after the repeated K +o episodes. At these concentrations, APV and CNQX completely, but reversibly, block currents in CA1 pyramidal neurons induced by NMDA and kainate, respectively. 1-[N,O-Bis(5-isoquinolinesulphonyl)N-methyl-l-tyrosyl]-4-phenylpiperazine (KN62) was purchased from Sigma. Stock solutions of KN62 (3 mM) were prepared in DMSO and stored at ⫺20°C prior to dilution (1:1000) in standard medium. Slices were preincubated in KN62 (3 µM, 0.1% DMSO) for 60 min before being placed in the recording chamber. At this concentration, KN62 inhibits CaMKII (Ki=0.9 µM) but has no direct effect on Ca2+ channels in CA1 neurons of hippocampal slices (Tokumitsu et al., 1990; Wyllie and Nicoll,
206
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
1994; Sihra and Pearson, 1995; Otmakhov et al., 1997). Preliminary experiments showed (data not shown) that treatment of CA1 slices with DMSO (0.1%), KN62 (3 µM) or APV (50 µM) alone without K +o increases (the baseline control conditions) had no effect on the time courses of the parameters of neuronal activity under study. In the baseline control conditions, CNQX (10 µM) depressed fEPSP slope and PS amplitude during a 60-min “washout” period. 2.6. Data analyses and statistics All electrophysiological data were digitized at 20 kHz and analyzed using a computer with software developed in-house for the measurement of PrV amplitude, PS amplitude, fEPSP slope, TASG and NPS. All values in Section 3 are expressed as mean±S.E.M. Because there were no significant differences between the parameters of field responses as well as the parameters of epileptiform activity for different time points (+20 min–+120 min) in baseline conditions (without K +o increases) compared to pre-repeated K +o increases (⫺20 min–0 min), these parameters were compared before and after three repeated K +o increases. To determine whether there were any significant differences (P⬍0.05) between these parameters, one-way analysis of variance (ANOVA) or onetailed Student’s t-tests were employed. In some cases, to determine whether there were significant differences between the magnitudes of changes in electrophysiological parameters produced by different stimulation protocols, two-tailed independent t-tests were employed.
3. Results Fig. 1b shows the effects of repeated applications of K + on the development of the PS discharges in CA1 hippocampal slices. Three and six, but not one, K +o episodes induced the development of electrical stimulusevoked PS discharges, which persisted for more than 2 h after the final K +o increase. Three K +o episodes evoked long-lasting decreases in TASG and increases in NPS (Fig. 2). When slices were treated with APV (50 µM), the decreases in TASG and increases in NPS induced by repeated K + applications were only short lasting (40–60 min) (Fig. 2). APV (50 µM) abolished the development of the PS discharges induced by three K +o episodes (compare Fig. 2c with Fig. 1b). Analysis of the effects of APV on the time courses of TASG decrease and NPS increase showed that this drug depresses (by around 33%) the early induction phase (i.e. 20 min after the last K +o increase) and shortens the late maintenance phase (i.e. epileptiform activity only lasted around 40 min after the last K +o increase) of the development of an in vitro kindling-like state (Fig. 2). Three K +o episodes had no effect on the time course of the
Fig. 2. Effects of APV on the time courses of TASG (a) and NPS (b) produced by repeated brief exposure of CA1 hippocampal slices to high K +o. TASG values were normalised relative to the pre-K +o episodes (⫺20 min–0 min) TASG values. The magnitudes of NPS are expressed in absolute values. The dashed lines are the time courses of the normalised TASG values (a) and NPS values (b) in baseline control conditions (after APV application without K +o increase). Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) during APV (50 µM) application; *P⬍0.05, n=6. (c) Exemplary PS responses obtained before (Pre 3K(+)+APV) and 120 min after 3K(+)+APV episodes. Calibration bars: 1 mV and 3 ms.
fEPSP slope but induced an increase in the peak amplitude of the PS for the given fEPSP slope (i.e. EPSPspike potentiation) as shown in Fig. 3. K +o episodes during bath application of APV (50 µM) induced a delayed depression of the fEPSP slope, and completely blocked EPSP-spike potentiation (Fig. 3). Three brief applications of APV (50 µM) alone (in baseline control condition) had no effect on the time courses of TASG, NPS, fEPSP slope and PS amplitude in post-treated CA1 hippocampal slices (data not shown). In contrast to APV, the AMPA/kainate-receptor antagonist, CNQX (10 µM), had no significant effects on the decreases in TASG (Fig. 4a) and increases in NPS (data not shown) induced by the repeated K +o increases. Furthermore, CNQX delayed by around 60 min the EPSPspike potentiation induced by K +o episodes (Fig. 4b).
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
207
Fig. 3. Effects of APV on the time courses of fEPSP slope (a) and PS amplitude (b) produced by repeated brief exposures to high K +o. Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) during APV (50 µM) application; *P⬍0.05, n=6.
However, this delayed potentiating effect of K +o episodes during bath application of 10 µM CNQX can be explained by the slow washout kinetics of CNQX because this glutamate antagonist must be “washed out” for about 60 min (in baseline control condition—without K +o episodes) in order to allow the fEPSP slope and first PS amplitude to return to control values (for ⫺20 min– 0 min time points in the time scale) (data not shown). In contrast to APV, CNQX (10 µM) did not abolish the development of the PS discharges induced by three K +o applications (Fig. 4c comparatively to Fig. 2c). When the slices were preincubated with 3 µM KN62, the K +o episodes induced only short-lasting (40–60 min) decreases in TASG and increases in NPS (Fig. 5), and failed to induce the EPSP-spike potentiation (Fig. 6b). Like APV, KN62 (3 µM) also abolished the development of the PS discharges induced by three K +o episodes (compare Fig. 5c with Fig. 2c and Fig. 1b). These effects of KN62 were similar to those of APV but, in contrast to the latter, KN62 did not alter the early induction phase (i.e. 20 min after the final K +o increase) of the in vitro kindling-like state (Fig. 5 and Fig. 2) and did not induce a delayed depression of the EPSP slope (Fig. 6a and Fig. 3a). KN62 (3 µM) alone (in baseline control condition) had no effects on the time courses of TASG, NPS, fEPSP slope and PS amplitude in post-treated CA1 hippocampal slices (data not shown).
Fig. 4. Effects of CNQX on the time courses of TASG (a) and PS amplitude (b) produced by repeated brief exposures to high K +. TASG values were normalised relative to the pre-K +o episodes (⫺20 min–0 min) TASG values. The dashed line (a) is the time course of the normalised TASG values in baseline control conditions (after CNQX application without K +o increase). Time (min) is the timing in relation to the last K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) during CNQX (10 µM) application; *P⬍0.05, n=6. (c) Exemplary PS responses obtained before (Pre 3K(+)+CNQX) and 100 min after 3K(+)+CNQX episodes. Calibration bars: 1 mV and 5 ms.
4. Discussion The characteristic feature of the kindling model of epileptogenesis is the sustained predisposition of a certain brain region to seizure susceptibility. Therefore, one of the more important tasks in the elucidation of kindling mechanisms involves understanding the cellular–molecular mechanisms of plastic alterations in neurons associated with the start of kindling-induced epileptogenesis. Although hippocampal slices are used widely to study the electrophysiological and biochemical processes ongoing during a seizure or an interictal spike, it has
208
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
Fig. 6. Effects of KN62 on the time courses of fEPSP slope (a) and PS amplitude (b) produced by repeated brief exposures to high K +o. Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) after preincubation with KN62 (3 µM); * P⬍0.05, n=6.
Fig. 5. Effects of KN62 on the time courses of TASG (a) and NPS (b) produced by repeated brief exposures to high K +o. TASG values were normalised relative to the pre-K +o episodes (⫺20 min–0 min) TASG values. The magnitudes of NPS are expressed in absolute values. The dashed lines are the time courses of the normalised TASG values (a) and NPS values (b) in baseline control conditions (after KN62 application without K +o increase). Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) after preincubation with KN62 (3 µM); *P⬍0.05, n=6. (c) Exemplary PS responses obtained before (Pre 3K(+)+KN62) and 120 min after (3K(+)+KN62) preincubation with KN62. Calibration bars: 1 mV and 5 ms.
been difficult to find an in vitro model that might be useful for understanding how epileptogenesis arises. Therefore, the study of the development mechanisms of the kindling phenomenon by using in vitro models of epileptogenesis is problematic. In our previous experiments we demonstrated that repeated brief exposures of the CA1 hippocampal slices to high K +o induces the expression of two events (Semyanov and Godukhin, 1997; Semyanov et al. 1997, 2000): (i) a long-lasting EPSP-spike potentiation and (ii) a progressive sustained decrease in the threshold for development of stimulus-evoked epileptiform discharges in CA1 pyramidal neurons (an in vitro kindling-like state). EPSP-spike potentiation results in a larger popu-
lation spike amplitude for a given fEPSP size and a higher probability of single spike discharge. Several possible mechanisms for EPSP-spike potentiation have been proposed: (1) reduced inhibition mediated by GABAergic interneurons, (2) increased membrane excitability at the initial segment of the axon, possibly resulting from increased Na+ channel density, and (3) amplification or reduced attenuation of distal dendritic inputs via the clustering of low-voltage activated Ca2+ channels (Wheal et al., 1998). In our previous and present experiments we found that L-type high-voltage activated and NMDA-dependent Ca2+ channels are also essential for this component of potentiation. In all likelihood, EPSP-spike potentiation involves more than one mechanism. Thus, this potentiation could act as a strong factor for the induction of epileptiform discharge. Indeed, our previous and present findings demonstrate that the K +o episode-induced in vitro kindling-like state is associated with this component of long-term potentiation. Our results also demonstrate that induction of both events depends strongly on the activity of L-type voltage-sensitive Ca2+ channels (Semyanov and Godukhin, 1997), and, as we found in the present study, Ca2+ influx through NMDA-dependent Ca2+ channels may also be essential for the development of both the in vitro kindling-like state and EPSP-spike potentiation. As we demonstrated, a selective antagonist of NMDAreceptors, APV (50 µM), blocks the EPSP-spike potenti-
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
ation, depresses the induction phase of the in vitro kindling-like state, and blocks the maintenance phase of this state. In contrast to APV, the blockade of AMPA/kainate receptors by CNQX (10 µM) had no effect on the development of the in vitro kindling-like state, and delayed (by around 60 min) EPSP-spike potentiation induced by the repeated K +o increases (Fig. 4a and b). However, this delayed potentiating effect of K +o episodes during the CNQX treatment can be explained by the slow washout kinetics of CNQX (Section 3). In other words, the K +o episodes are probably capable of inducing EPSP-spike potentiation during the first 60 min of the “washout” period but their effects are masked by the blocking action of the remaining CNQX on synaptic responses. One of the more important questions addressed by our findings is how the relatively short-term increases in intracellular Ca2+ induced by the repeated brief K +o increases can maintain EPSP-spike potentiation and the in vitro kindling-like state in the CA1 area for a long time. Like APV, KN62 (3 µM), a selective membrane permeable inhibitor of CaMKII, blocked the EPSP-spike potentiation and the maintenance phase of the in vitro kindling-like state. Lisman (1994) proposed that CaMKII molecules became phosphorylated during a synaptic event that was to be “remembered”, even after the event was over. It has been proposed that Ca2+dependent autophosphorylation of this kinase can switch the molecule into an “on” state which retains its activity even after Ca2+ is removed (Dosemeci and Albers, 1996). These findings indicate that Ca2+ could participate in the induction of EPSP- spike potentiation and the maintenance mechanisms of the in vitro kindling-like state via autophosphorylation of CaMKII. Repeated brief increases in K +o can activate CaMKII via both the voltage-sensitive Ca2+ channels and the NMDA-receptor operated Ca2+ ionophores localized on the soma/proximal dendrites and distal dendrites in CA1 pyramidal neurons, respectively (Westenbrock et al., 1990; Muller and Connor, 1991; Christie et al., 1995; Huber et al., 1995; Rao and Craig, 1997). It is known that Ca2+ influx via NMDA-operated ionophores activates this kinase concentrated in the postsynaptic density of glutamate synapses to induce synaptic long-term potentiation (Malenka et al., 1989; Fleck et al., 1992; Onyang et al., 1997). Our findings indicate that EPSPspike potentiation induced by the repeated increases in K +o depends on the activation of NMDA receptors. Such NMDA-dependent EPSP-spike potentiation can essentially facilitate the development of epileptiform events in CA1 pyramidal neurons in response to the repeated increases in K +o induced in the CA1 hippocampal area under certain pathological conditions (for example, under repeated episodes of hypoxia or ischaemia). On the basis of our previous (Semyanov and Godukhin, 1997; Semyanov et al., 2000) and present data, and
209
from the literature findings, we should like to propose the following hypothetical scheme for the involvement of L-type voltage-sensitive Ca2+ channels, NMDAdependent Ca2+ channels and CaMKII in the development of epileptiform bursts in CA1 pyramidal neurons in response to repeated increases in K +o (Fig. 7). The repetition of these K +o increases induces long-lasting changes in the properties of L-type Ca2+ channels in CA1 pyramidal neurons. The intrinsic properties of the cells are modified such that the pyramidal neurones them-
Fig. 7. Hypothetical scheme for the involvement of L-type voltagesensitive Ca2+ channels, NMDA-dependent Ca2+ channels and CaMKII in the development of epileptiform bursts in CA1 pyramidal neurons in response to repeated brief K +o increases. It is suggested that synchronous epileptiform bursts in many CA1 pyramidal neurones are associated with the prolonged membrane depolarisation shifts which are responsible for driving the rapid bursts of Na+-dependent action potentials. The prolonged membrane depolarisation shifts in CA1 pyramidal neurons are induced by influx of Ca2+ via L-type voltage-sensitive Ca2+ channels (VSCCs). Influxes of Ca2+ via L-type VSCCs and NMDA receptor-dependent Ca2+ channels also induce EPSP-spike potentiation (ESP) which can facilitate the epileptiform burst development (possibly via activation of L-type VSCCs). These effects of Ca2+ on EPSP-spike potentiation are mediated by the activation of CaMKII. The repeated K +o increases switch an inactive form of CaMKII to an active one which in turn induces long-lasting changes in L-type VSCC properties (possibly via an increase in a population of the channels which are active at resting membrane potentials). The properties of these channels are modified such that the pyramidal neurons become hyperresponsive to excitatory inputs. In other words, the repeated increases in K +o induce a sustained decrease in the threshold of evoked epileptiform discharge development. The broken arrows in the scheme show the pathways terminating the depolarising shift in membrane potential induced by the Ca2+ influx via L-type VSCC. SC/CF, Schaffer collaterals/commissural fibres; IN, CA1 inhibitory interneurons.
210
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
selves become hyperresponsive to excitatory inputs. Epileptiform bursts are observed as multiple Na+-dependent action potentials superimposed on synchronized, L-type Ca2+-channel-mediated, depolarising shifts of membrane potential in many CA1 pyramidal neurons. This hypothesis is confirmed by the following data. Under normal conditions (without repeated K +o-episode stimulation), the most conspicuous electrophysiological property of CA1 pyramidal neurons is their regular—spiking but not burst firing—behaviour (Mason, 1993). The L-type Ca2+ channel has high conductivity and slow kinetics of inactivation, so that once activated by an excitatory input, it will sustain the depolarisation for tens of milliseconds, driving rapid bursts of fast-action potentials mediated by voltage-dependent sodium channels. It is possible that the repeated K +o episodes can increase a population of dihydropyridine-sensitive Ca2+ channels (possibly Ltype) which are active at resting membrane potentials in hippocampal CA1 pyramidal neurons (Magel et al., 1996). EPSP-spike potentiation could increase the burst firing probability of the cells through the activation of L-type voltage-sensitive Ca2+ channels. The contribution of this component of long-term potentiation to L-type Ca2+channel-mediated depolarising shift expression can be essential for a sustained decrease in the threshold for development of evoked epileptiform discharges (in vitro kindling-like state) in CA1 pyramidal neurons. The facilitatory effect of NMDA-dependent Ca2+ influx on the development of epileptiform bursts is mediated by CaMKII which participates in the maintenance mechanisms of the in vitro kindling-like state in the CA1 hippocampal area in response to repeated increases in K +o. This scheme also includes Ca2+-dependent K + channels and GABA-mediated recurrent inhibition which could be involved in a termination of the depolarising shift and epileptiform burst in CA1 pyramidal neurons (Jefferys, 1990; Traub and Dingledine, 1990; Bradford, 1995; Le Beau and Alger, 1998). We propose that this hypothetical scheme can help to explain some of the cellular–molecular mechanisms of the plastic alterations in CA1 pyramidal neurons associated with the start of kindling-induced epileptogenesis. Acknowledgements This work was supported by grants from INTAS Open Call (# 97-0382) and the Russian Foundation of Basic Research (# 99-04-48444). References Anderson, W.W., Lewis, D.V., Swartzwelder, H.S., Wilson, W.A., 1986. Magnesium-free medium activates seizure-like events in the rat hippocampal slice. Brain Research 398, 215–219.
Bernard, C., Wheal, H.V., 1995. Plasticity of AMPA and NMDA receptor-mediated epileptiform activity in a chronic model of temporal lobe epilepsy. Epilepsy Research 21, 95–107. Bliss, T.V.P., Lynch, M.A., 1988. Long-term potentiation of synaptic transmission in the hippocampus: properties and mechanisms. In: Landfield, P.W., Deadwyler, S.A. (Eds), Long-term potentiation: from biophysics to behavior. Neurology and Neurobiology. Alan R. Liss, Inc. 35 (4), 3–72. Bradford, H.F., 1995. Glutamate, GABA and epilepsy. Progress in Neurobiology 47, 477–511. Butler, L., Silva, A., Tonegawa, S., McNamara, J.O., 1993. Enhanced seizure susceptibility yet impairment of kindling development in α-calcium-calmodulin kinase II mutant mice. Society for Neuroscience Abstracts 19, 1030. Christie, B.R., Eliot, L.S., Ito, K.I., Miyakawa, H., Johnston, D., 1995. Different Ca2+ channels in soma and dendrites of hippocampal pyramidal neurons mediate spike-induced Ca2+ influx. Journal of Neurophysiology 73, 2553–2557. Della Paschoa, O.E., Kruk, M.R., Hamstra, R., Voskuyl, R.A., Danhov, M., 1997. Seizure patterns in kindling and cortical stimulation models of experimental epilepsy. Brain Research 770, 221–227. Dosemeci, A., Albers, R.W., 1996. A mechanism for synaptic frequency detection through autophosphorylation of CaM kinase II. Biophysical Journal 70, 2493–2501. Fleck, M.W., Palmer, A.M., Barrionuevo, G., 1992. Potassium-induced long-term potentiation in rat hippocampal slices. Brain Research 580, 100–105. Fukunaga, K., Muller, D., Miyamoto, E., 1995. Increased phosphorylation of Ca2+/calmodulin-dependent protein kinase II and its endogenous substrates in the induction of long-term potentiation. Journal of Biological Chemistry 270 (11), 6119–6124. Goddard, G.V., McIntyre, D.C., Leech, C.K., 1969. A permanent change in brain function resulting from daily electrical stimulation. Experimental Neurology 15, 295–330. Huber, K.M., Mank, M.D., Kelly, P.T., 1995. LTP induced by activation of voltage-dependent Ca2+ channels requires protein kinase activity. NeuroReport 6, 1281–1284. Ito, I., Hidaka, H., Sugiyama, H., 1991. Effects of KN62, a specific inhibitor of calcium/calmodulin-dependent protein kinase II, on long-term potentiation in the rat hippocampus. Neuroscience Letters 121, 119–121. Jefferys, J.G.R., 1990. Basic mechanisms of focal epilepsies. Experimental Physiology 75, 127–162. Kuba, K., Kumamoto, E., 1990. Long-term potentiations in vertebrate synapses: a variety of cascades with common subprocesses. Progress in Neurobiology 34, 197–269. Le Beau, F.E.N., Alger, B.E., 1998. Transient suppression of GABAAreceptor-mediated IPSPs after epileptiform burst discharges in CA1 pyramidal cells. Journal of Neurophysiology 79, 659–669. Lisman, J., 1994. The CaM kinase II hypothesis for the storage of synaptic memory. Trends in Neurological Science 17 (10), 406– 412. Lothmann, E.W., Williamson, J.M., 1994. Closely spaced recurrent hippocampal seizures elicit two types of heightened epileptogenesis: a rapidly developing, transient kindling and slowly developing, enduring kindling. Brain Research 649, 71–84. McEachern, J.C., Shaw, C.A., 1996. An alternative to the LTP orthodoxy: a plasticity–pathology continuum model. Brain Research Reviews 22, 51–92. Magel, J.C., Avery, R.B., Christie, B.R., Johnston, D., 1996. Dihydropyridine-sensitive, voltage-gated Ca2+ channels contribute to the resting intracellular Ca2+ concentration of hippocampal CA1 pyramidal neurons. Journal of Neurophysiology 76 (5), 3460–3470. Malenka, R.C., Kauer, J.A., Perkel, J.D., Mauk, M.D., Kelly, P.T., Nicoll, R.A. et al., 1989. An essential role for post-synaptic calmodulin and protein kinase activity in long-term potentiation. Nature 340, 554–556.
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
Malinov, R., Schulmann, H., Tsien, R.W., 1989. Inhibition of postsynaptic PKC or CaMKII blocks induction but not expression of LTP. Science 245, 826–830. Mason, A., 1993. Electrophysiology and burst-firing of rat subicular pyramidal neurons in vitro: a comparison with area CA1. Brain Research 600, 174–178. Mody, I., Lambert, J.D.C., Heinemann, U., 1987. Low extracellular magnesium induces epileptiform activity and spreading depression in rat hippocampal slices. Journal of Neurophysiology 57, 869–888. Muller, W., Connor, J.A., 1991. Dendritic spines as individual neuronal compartments for synaptic Ca2+ responses. Nature 354, 73– 76. Murray, K.D., Gall, C.M., Benson, D.L., Jones, E.G., Isackson, P.J., 1995. Decreased expression of the alpha subunit of Ca2+/calmodulin-dependent protein kinase type II mRNA in the adult rat CNS following recurrent limbic seizures. Brain Research 32 (2), 221–232. Onyang, Y., Kantor, D., Harris, K.M., Schuman, E.M., Kennedy, M.B., 1997. Visualization of the distribution of autophosphorylated calcium/calmodulin-dependent protein kinase II after tetanic stimulation in the CA1 area of the hippocampus. The Journal of Neuroscience 17, 5416–5427. Otmakhov, N., Griffith, L.C., Lisman, J.E., 1997. Postsynaptic inhibitors of calcium/calmodulin-dependent protein kinase type II block induction but not maintenance of pairing-induced long-term potentiation. The Journal of Neuroscience 17 (14), 5357–5365. Rao, A., Craig, A.M., 1997. Activity regulates the synaptic localization of the NMDA receptor in hippocampal neurons. Neuron 19 (4), 801–812. Semyanov, A., Godukhin, O., 1997. Kindling-like state in rat hippocampal CA1 slices induced by the repeated short-term extracellular K + increases: the role of L-type Ca2+ channels. Neuroscience Letters 223, 177–180. Semyanov, A., Morenkov, E., Godukhin, O., 1997. The decreased susceptibility to the development of in vitro kindling-like state in hippocampal CA1 slices of rats sensitive to audiogenic seizures. Neuroscience Letters 230, 187–190. Semyanov, A., Savin, A., Morenkov, E., Godukhin, O., 2000. In vivo
211
hippocampal kindling occludes the development of in vitro kindling-like state in CA1 rat hippocampal slices. Epilepsy Research 38 (1), 75–85. Sihra, T.S., Pearson, H.A., 1995. Ca/calmodulin-dependent kinase II inhibitor KN62 attenuates glutamate release by inhibiting voltagedependent Ca2+ channels. Neuropharmacology 34 (7), 731–741. Tokumitsu, H., Chijiwa, T., Hagiwara, M., Mizutani, A., Terasawa, M., Hidaka, H., 1990. KN62, 1-[N,O-bis(isoquinolinesulfonyl)-Nmethyl-l-tyrosyl]-4-phenylpiperazine, a specific inhibitor of Ca2+/calmodulin-dependent protein kinase II. Journal of Biological Chemistry 265, 4315–4320. Traub, R.D., Dingledine, R., 1990. Model of synchronized epileptiform bursts induced by high potassium in CA3 region of rat hippocampal slice. Role of spontaneous EPSPs in initiation. Journal of Neurophysiology 64, 1009–1018. Wasterlain, C.G., Farber, D.B., 1984. Kindling alters the calcium/calmodulin dependent phosphorylation of synaptic plasma membrane proteins in rat hippocampus. Proceedings of the National Academy of Sciences of the United States of America 81, 1253–1257. Westenbrock, R.E., Ahlijanian, M.K., Catterall, W.A., 1990. Clustering of L-type Ca2+ channels at the base of major dendrites in hippocampal pyramidal neurons. Nature 347, 281–284. Wheal, H.V., Bernard, C., Chad, J.E., Cannon, R.C., 1998. Pro-epileptic changes in synaptic function can be accompanied by pro-epileptic changes in neuronal excitability. Trends in Neurological Science 21 (4), 167–174. Wilson, W.A., Stasheff, S., Swartzwelder, S., Clark, S., Anderson, W.W., 1992. The role of NMDA receptors in in vitro epileptogenesis. In: Avanzini, G., Engel, J. Jr., Faricllo, R., Heinemann, U. (Eds.), Neurotransmitters in Epilepsy (Epilepsy Research Supplement). Elsevier, Amsterdam, pp. 157–166 Suppl. 8. Wyllie, D.J.A., Nicoll, R.A., 1994. A role for protein kinases and phosphatases in the Ca2+-induced enhancement of hippocampal AMPA receptor-mediated synaptic responses. Neuron 13, 635–643. Zhou, X.R., Suzuki, T., Shimizu, H., Nishino, H., 1994. Amygdala kindling activates the phosphorylation of Ca2+/calmodulin-dependent protein kinase II in rat hippocampus. Neuroscience Letters 12, 45–48.
Epileptiform activity and EPSP-spike potentiation induced in rat hippocampal CA1 slices by repeated high-K +: involvement of ionotropic glutamate receptors and Ca2+/calmodulin-dependent protein kinase II Alexey Semyanov 1, Oleg Godukhin
*
Institute of Theoretical and Experimental Biophysics, Russian Academy of Sciences, Pushchino, Moscow Region, 142292, Russia Received 24 January 2000; received in revised form 20 June 2000; accepted 2 August 2000
Abstract We have previously demonstrated that repeated brief increases in extracellular K + (K +o) induce a hyperexcitability in CA1 pyramidal cells that persists for a long time after the final application of K + [Neurosci. Lett. 223 (1997) 177; Epilepsy Research (2000) 75]. This epileptiform activity, which was associated with a lasting excitatory postsynaptic potential (EPSP)-spike potentiation, presented some of the characteristic features of traditional in vivo kindling. We have also found that Ca2+ influx through Ltype voltage-sensitive Ca2+ channels is essential for the development of both in vitro kindling and EPSP-spike potentiation. The aims of this study were to investigate the involvement of ionotropic glutamate receptors, especially those of the NMDA subtype, and the requirement for Ca2+/calmodulin-dependent protein kinase II (CaMKII) in these phenomena. Field EPSPs with presynaptic fibre volleys from the stratum radiatum, and population spikes from the stratum pyramidale, were recorded in the CA1 area of rat hippocampal slices in response to electrical stimulation of the Schaffer collateral/commissural fibres. Repeated (three episodes) brief (30 s) increases in extracellular K + induced a sustained decrease in the threshold for development of evoked epileptiform discharges (i.e. an in vitro kindling-like state) and a lasting potentiation of the EPSP-spike transfer in CA1 pyramidal neurons (EPSP-spike potentiation). The selective antagonist of NMDA receptors, APV (50 µM), blocked the EPSP-spike potentiation, depressed the induction phase of the in vitro kindling-like state, and blocked the maintenance phase of this state. In contrast to APV, the blockade of AMPA/kainate receptors by CNQX (10 µM) had no effect. Like APV, KN62 (3 µM), a selective membrane permeable inhibitor of CaMKII, blocked the EPSP-spike potentiation and the maintenance phase of the in vitro kindling-like state. Our previous and present results therefore demonstrate that Ca2+ influxes through L-type voltage-dependent—and NMDA receptor-dependent—Ca2+ channels contribute differentially to the development of an in vitro kindling-like state, and both induce EPSP-spike potentiation in CA1 hippocampal pyramidal cells in response to repeated brief increases in K +o. It is suggested that these effects of intracellular Ca2+ on the maintenance phase of the in vitro kindling-like state and EPSP-spike potentiation are mediated by CaMKII-dependent mechanisms. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Hippocampal CA1 slices; Field potentials; Epileptiform activity; EPSP-spike potentiation; NMDA glutamate receptor; Ca2+/calmodulindependent protein kinase II
1. Introduction Kindling (i.e. progressive reduction of threshold for induction of seizure activity) is an important model of * Corresponding author. Fax: +7-967-790-553. E-mail addresses: [email protected] (A. Semyanov), [email protected] (O. Godukhin). 1 Present address: Department of Clinical Neurology, Institute of Neurology, UCL, Queen Square, London WC1N 3BG, UK. Fax: +4420-7278-5616.
epileptogenesis, but in vivo investigations have been unable so far to clarify the processes leading to hyperexcitability (Goddard et al., 1969; Lothmann and Williamson, 1994; Della Paschoa et al., 1997). Hippocampal slices are widely used to study the cellular mechanisms of epileptogenesis (Anderson et al., 1986; Mody et al., 1987; Wilson et al., 1992; Fleck et al., 1992; Bernard and Wheal, 1995), and we have previously demonstrated that repeated (3–6 episodes) brief (30 s) increases in extracellular K + (K +o) induce a hyperexcitability in CA1
0028-3908/01/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 0 8 ( 0 0 ) 0 0 1 4 7 - 7
204
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
pyramidal cells that persists for a long time after the final application of K + (Semyanov and Godukhin, 1997; Semyanov et al., 2000). This epileptiform activity, which is associated with a lasting excitatory postsynaptic potential (EPSP)-spike potentiation, presents some of the characteristic features of traditional in vivo kindling (i.e. repeated pattern of stimulation needed to kindle; progressive reduction of the threshold of evoked epileptiform discharges dependent on the number of K +o episodes applied, and persistence). Furthermore, as demonstrated previously, electrical and audiogenic kindlings in vivo occluded the development of the in vitro kindling-like state induced by repeated increases in K +o CA1 slices (Semyanov et al., 2000). We have also found that Ca2+ influx through L-type voltage-sensitive Ca2+ channels is essential for the development of both in vitro kindling and EPSP-spike potentiation in CA1 hippocampal slices in response to the repeated increases in K +o (Semyanov and Godukhin, 1997). The first aim of this study was to investigate the involvement of ionotropic glutamate receptors, especially those of the NMDA subtype, in these phenomena because NMDA receptors are involved in the induction of both synaptic long-term potentiation (LTP) (Bliss and Lynch, 1988; Kuba and Kumamoto, 1990) and epileptiform activity (Jefferys, 1990; Wilson et al., 1992; Bernard and Wheal, 1995) in CA1 pyramidal neurons. Intracellular Ca2+, through activation of Ca2+/calmodulin-dependent kinase II (CaMKII), is known to play a key role in the long-lasting, activitydependent synaptic modifications that underlie various forms of neuronal plasticity such as LTP and traditional kindling (Wasterlain and Farber, 1984; Malenka et al., 1989; Malinov et al., 1989; Ito et al., 1991; Butler et al., 1993; Lisman, 1994; Zhou et al., 1994; Fukunaga et al., 1995; Huber et al., 1995; Murray et al., 1995; Otmakhov et al., 1997). Therefore, the second aim of this study was to examine the requirement for CaMKII in the development of EPSP-spike potentiation and the in vitro kindling-like state induced by repeated high K + episodes. We have used a pharmacological approach using antagonists of glutamate NMDA and AMPA/kainate receptors, and KN62 as a specific, membrane permeable inhibitor of CaMKII (Tokumitsu et al., 1990). Because synchronization of the cell population is the principal feature of epileptiform discharges, we used an extracellular recording protocol to provide a better insight into the collective behaviour of the cell population. 2. Methods 2.1. Preparation and maintenance of CA1 hippocampal slices All experiments were carried out with male Wistar rats (Animal laboratory of ITEB RAN, Pushchino,
Russia) weighing 140–150 g (n=30). CA1 hippocampal slice preparation and superfusion, and methods for the recording of field excitatory postsynaptic potentials (fEPSPs) with presynaptic fibre volleys (PrVs) from the stratum radiatum and population spikes (PSs) from the stratum pyramidale in response to electric stimulation of the Schaffer collateral/commissural fibres (SC/CF) were as described previously (Semyanov and Godukhin, 1997; Semyanov et al. 1997, 2000). To functionally isolate the CA1 area from CA2 and CA3 areas, an orthogonal cut of the stratum pyramidale extending to the distal edge of the mossy fibre layer was made (Fig. 1a). Slices were placed into a submersion recording chamber and superfused at 2.5 ml/min (32°C). The composition of the standard medium for electrophysiological recording was (in mM): 124 NaCl, 3 KCl, 1.25 KH2PO4, 2 MgSO4, 2 CaCl2, 26 NaHCO3, 10 d-glucose; pH 7.4 was adjusted with 95% O2/5% CO2. 2.2. Extracellular electrophysiology Slices were allowed to equilibrate in the chamber for at least 3 h before data collection. Field responses were recorded from the stratum radiatum and stratum pyramidale of the CA1 area with two glass microelectrodes filled with standard medium (Fig. 1a). 2.3. Synaptic LTP and EPSP-spike potentiation PrV amplitudes, fEPSP slopes and PS amplitudes were obtained from measurements of individual responses to a series of 25 separate stimuli from minimum to maximum stimulus intensities (for PS generation) applied at 10-s intervals. Each series of 25 stimuli was separated from the next by 20 min. The three components of the overall input–output relationships, PrV amplitude versus current intensity, fEPSP slope versus PrV amplitude and PS amplitude versus fEPSP slope were taken as indices of the excitability of the glutamatergic SC/CF fibres, the glutamatergic synaptic transmission efficiency and the EPSP-spike transfer efficiency, respectively (Semyanov and Godukhin, 1997; Semyanov et al., 1997, 2000). The time course of the fEPSP slope for a given PrV amplitude and the time course of the amplitude of the first PS in PS responses for a given fEPSP slope were constructed. Typically, time courses of fEPSP slope were determined within an intermediate range of PrV amplitudes (around 0.5 mV), and time courses of PS amplitude within an intermediate range of fEPSP slopes (around 0.4 mV/ms). However, these parameters were investigated over the full dynamic range of stimulus intensities to avoid errors that could arise if a single arbitrary stimulus intensity was used (McEachern and Shaw, 1996).
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
205
Fig. 1. Representative field potentials recorded in CA1 hippocampal slices in response to a single electrical stimulus applied to the stratum radiatum. (a, left panel) Schematic drawing of a CA1 hippocampal slice showing the position of the stimulation electrode (Stim), and the two sites of extracellular recordings: 1, fEPSP with PrV were recorded in the stratum radiatum (rad); 2, single PS or multiple PS discharges were recorded in the stratum pyramidale (pyr). (a, right panel) Extracellulary recorded field potentials from electrode sites 1 and 2, with changes in the pattern of PS response induced by repeated K +o applications (dashed line, additional PS). sch, Schaffer collateral/commissural fibres. (b) A typical set of PS responses recorded in CA1 hippocampal slices without (Control) and 1 h after 1, 3, or 6 K +o applications. All recordings were made 4 h after the start of slice superfusions. Vertical and horizontal calibration bars are 1 mV and 10 ms, respectively.
2.4. Epileptiform activity
2.5. Pharmacological manipulations
Epileptiform activity was induced by three brief (30 s) episodes of exposure to 20 mM K +o (with equimolar decreases in Na+) applied at 10-min intervals. The appearance of multiple PSs in response to a single electrical stimulus was taken as indicator of the development of epileptiform activity in the CA1 pyramidal neurons (Wheal et al., 1998). Two parameters of such activity were measured: (1) the stimulus intensity (µA) of the appearance of the second (additional) PS (marked by an arrow on the dashed line in Fig. 1a) was characterized as the threshold of generation of an additional PS in PS discharges (TASG); and (2) the number of PSs in the PS discharge (NPS); this parameter was measured for the current intensity of TASG taken prior to the application of K +o episodes (or at the same time points for control experiments: ⫺20 min–0 min) (Semyanov et al., 2000). Fig. 1b shows typical PS responses recorded from CA1 hippocampal slices without (baseline control) and 1 h after 1, 3 and 6 applications of K +. All recordings were made 4 h after the start of slice superfusions.
d,l-(+)-2-Amino-5-phosphonovaleric acid (APV) and 6-cyano-7-nitroquinoxaline-2,3-(1H,4H)dione (CNQX), used as selective antagonists of NMDA and AMPA/kainate receptors, respectively, were purchased from Sigma. CNQX and APV were dissolved in DMSO and distilled water, respectively, prior to dilution in standard medium. APV (50 µM) and CNQX (10 µM, 0.1% DMSO) were applied for 2 min before, during and 2 min after the repeated K +o episodes. At these concentrations, APV and CNQX completely, but reversibly, block currents in CA1 pyramidal neurons induced by NMDA and kainate, respectively. 1-[N,O-Bis(5-isoquinolinesulphonyl)N-methyl-l-tyrosyl]-4-phenylpiperazine (KN62) was purchased from Sigma. Stock solutions of KN62 (3 mM) were prepared in DMSO and stored at ⫺20°C prior to dilution (1:1000) in standard medium. Slices were preincubated in KN62 (3 µM, 0.1% DMSO) for 60 min before being placed in the recording chamber. At this concentration, KN62 inhibits CaMKII (Ki=0.9 µM) but has no direct effect on Ca2+ channels in CA1 neurons of hippocampal slices (Tokumitsu et al., 1990; Wyllie and Nicoll,
206
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
1994; Sihra and Pearson, 1995; Otmakhov et al., 1997). Preliminary experiments showed (data not shown) that treatment of CA1 slices with DMSO (0.1%), KN62 (3 µM) or APV (50 µM) alone without K +o increases (the baseline control conditions) had no effect on the time courses of the parameters of neuronal activity under study. In the baseline control conditions, CNQX (10 µM) depressed fEPSP slope and PS amplitude during a 60-min “washout” period. 2.6. Data analyses and statistics All electrophysiological data were digitized at 20 kHz and analyzed using a computer with software developed in-house for the measurement of PrV amplitude, PS amplitude, fEPSP slope, TASG and NPS. All values in Section 3 are expressed as mean±S.E.M. Because there were no significant differences between the parameters of field responses as well as the parameters of epileptiform activity for different time points (+20 min–+120 min) in baseline conditions (without K +o increases) compared to pre-repeated K +o increases (⫺20 min–0 min), these parameters were compared before and after three repeated K +o increases. To determine whether there were any significant differences (P⬍0.05) between these parameters, one-way analysis of variance (ANOVA) or onetailed Student’s t-tests were employed. In some cases, to determine whether there were significant differences between the magnitudes of changes in electrophysiological parameters produced by different stimulation protocols, two-tailed independent t-tests were employed.
3. Results Fig. 1b shows the effects of repeated applications of K + on the development of the PS discharges in CA1 hippocampal slices. Three and six, but not one, K +o episodes induced the development of electrical stimulusevoked PS discharges, which persisted for more than 2 h after the final K +o increase. Three K +o episodes evoked long-lasting decreases in TASG and increases in NPS (Fig. 2). When slices were treated with APV (50 µM), the decreases in TASG and increases in NPS induced by repeated K + applications were only short lasting (40–60 min) (Fig. 2). APV (50 µM) abolished the development of the PS discharges induced by three K +o episodes (compare Fig. 2c with Fig. 1b). Analysis of the effects of APV on the time courses of TASG decrease and NPS increase showed that this drug depresses (by around 33%) the early induction phase (i.e. 20 min after the last K +o increase) and shortens the late maintenance phase (i.e. epileptiform activity only lasted around 40 min after the last K +o increase) of the development of an in vitro kindling-like state (Fig. 2). Three K +o episodes had no effect on the time course of the
Fig. 2. Effects of APV on the time courses of TASG (a) and NPS (b) produced by repeated brief exposure of CA1 hippocampal slices to high K +o. TASG values were normalised relative to the pre-K +o episodes (⫺20 min–0 min) TASG values. The magnitudes of NPS are expressed in absolute values. The dashed lines are the time courses of the normalised TASG values (a) and NPS values (b) in baseline control conditions (after APV application without K +o increase). Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) during APV (50 µM) application; *P⬍0.05, n=6. (c) Exemplary PS responses obtained before (Pre 3K(+)+APV) and 120 min after 3K(+)+APV episodes. Calibration bars: 1 mV and 3 ms.
fEPSP slope but induced an increase in the peak amplitude of the PS for the given fEPSP slope (i.e. EPSPspike potentiation) as shown in Fig. 3. K +o episodes during bath application of APV (50 µM) induced a delayed depression of the fEPSP slope, and completely blocked EPSP-spike potentiation (Fig. 3). Three brief applications of APV (50 µM) alone (in baseline control condition) had no effect on the time courses of TASG, NPS, fEPSP slope and PS amplitude in post-treated CA1 hippocampal slices (data not shown). In contrast to APV, the AMPA/kainate-receptor antagonist, CNQX (10 µM), had no significant effects on the decreases in TASG (Fig. 4a) and increases in NPS (data not shown) induced by the repeated K +o increases. Furthermore, CNQX delayed by around 60 min the EPSPspike potentiation induced by K +o episodes (Fig. 4b).
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
207
Fig. 3. Effects of APV on the time courses of fEPSP slope (a) and PS amplitude (b) produced by repeated brief exposures to high K +o. Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) during APV (50 µM) application; *P⬍0.05, n=6.
However, this delayed potentiating effect of K +o episodes during bath application of 10 µM CNQX can be explained by the slow washout kinetics of CNQX because this glutamate antagonist must be “washed out” for about 60 min (in baseline control condition—without K +o episodes) in order to allow the fEPSP slope and first PS amplitude to return to control values (for ⫺20 min– 0 min time points in the time scale) (data not shown). In contrast to APV, CNQX (10 µM) did not abolish the development of the PS discharges induced by three K +o applications (Fig. 4c comparatively to Fig. 2c). When the slices were preincubated with 3 µM KN62, the K +o episodes induced only short-lasting (40–60 min) decreases in TASG and increases in NPS (Fig. 5), and failed to induce the EPSP-spike potentiation (Fig. 6b). Like APV, KN62 (3 µM) also abolished the development of the PS discharges induced by three K +o episodes (compare Fig. 5c with Fig. 2c and Fig. 1b). These effects of KN62 were similar to those of APV but, in contrast to the latter, KN62 did not alter the early induction phase (i.e. 20 min after the final K +o increase) of the in vitro kindling-like state (Fig. 5 and Fig. 2) and did not induce a delayed depression of the EPSP slope (Fig. 6a and Fig. 3a). KN62 (3 µM) alone (in baseline control condition) had no effects on the time courses of TASG, NPS, fEPSP slope and PS amplitude in post-treated CA1 hippocampal slices (data not shown).
Fig. 4. Effects of CNQX on the time courses of TASG (a) and PS amplitude (b) produced by repeated brief exposures to high K +. TASG values were normalised relative to the pre-K +o episodes (⫺20 min–0 min) TASG values. The dashed line (a) is the time course of the normalised TASG values in baseline control conditions (after CNQX application without K +o increase). Time (min) is the timing in relation to the last K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) during CNQX (10 µM) application; *P⬍0.05, n=6. (c) Exemplary PS responses obtained before (Pre 3K(+)+CNQX) and 100 min after 3K(+)+CNQX episodes. Calibration bars: 1 mV and 5 ms.
4. Discussion The characteristic feature of the kindling model of epileptogenesis is the sustained predisposition of a certain brain region to seizure susceptibility. Therefore, one of the more important tasks in the elucidation of kindling mechanisms involves understanding the cellular–molecular mechanisms of plastic alterations in neurons associated with the start of kindling-induced epileptogenesis. Although hippocampal slices are used widely to study the electrophysiological and biochemical processes ongoing during a seizure or an interictal spike, it has
208
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
Fig. 6. Effects of KN62 on the time courses of fEPSP slope (a) and PS amplitude (b) produced by repeated brief exposures to high K +o. Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) after preincubation with KN62 (3 µM); * P⬍0.05, n=6.
Fig. 5. Effects of KN62 on the time courses of TASG (a) and NPS (b) produced by repeated brief exposures to high K +o. TASG values were normalised relative to the pre-K +o episodes (⫺20 min–0 min) TASG values. The magnitudes of NPS are expressed in absolute values. The dashed lines are the time courses of the normalised TASG values (a) and NPS values (b) in baseline control conditions (after KN62 application without K +o increase). Time (min) is the timing in relation to the final K +o episode (3 K +o↓↓↓). (䊊) without and (䉬) after preincubation with KN62 (3 µM); *P⬍0.05, n=6. (c) Exemplary PS responses obtained before (Pre 3K(+)+KN62) and 120 min after (3K(+)+KN62) preincubation with KN62. Calibration bars: 1 mV and 5 ms.
been difficult to find an in vitro model that might be useful for understanding how epileptogenesis arises. Therefore, the study of the development mechanisms of the kindling phenomenon by using in vitro models of epileptogenesis is problematic. In our previous experiments we demonstrated that repeated brief exposures of the CA1 hippocampal slices to high K +o induces the expression of two events (Semyanov and Godukhin, 1997; Semyanov et al. 1997, 2000): (i) a long-lasting EPSP-spike potentiation and (ii) a progressive sustained decrease in the threshold for development of stimulus-evoked epileptiform discharges in CA1 pyramidal neurons (an in vitro kindling-like state). EPSP-spike potentiation results in a larger popu-
lation spike amplitude for a given fEPSP size and a higher probability of single spike discharge. Several possible mechanisms for EPSP-spike potentiation have been proposed: (1) reduced inhibition mediated by GABAergic interneurons, (2) increased membrane excitability at the initial segment of the axon, possibly resulting from increased Na+ channel density, and (3) amplification or reduced attenuation of distal dendritic inputs via the clustering of low-voltage activated Ca2+ channels (Wheal et al., 1998). In our previous and present experiments we found that L-type high-voltage activated and NMDA-dependent Ca2+ channels are also essential for this component of potentiation. In all likelihood, EPSP-spike potentiation involves more than one mechanism. Thus, this potentiation could act as a strong factor for the induction of epileptiform discharge. Indeed, our previous and present findings demonstrate that the K +o episode-induced in vitro kindling-like state is associated with this component of long-term potentiation. Our results also demonstrate that induction of both events depends strongly on the activity of L-type voltage-sensitive Ca2+ channels (Semyanov and Godukhin, 1997), and, as we found in the present study, Ca2+ influx through NMDA-dependent Ca2+ channels may also be essential for the development of both the in vitro kindling-like state and EPSP-spike potentiation. As we demonstrated, a selective antagonist of NMDAreceptors, APV (50 µM), blocks the EPSP-spike potenti-
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
ation, depresses the induction phase of the in vitro kindling-like state, and blocks the maintenance phase of this state. In contrast to APV, the blockade of AMPA/kainate receptors by CNQX (10 µM) had no effect on the development of the in vitro kindling-like state, and delayed (by around 60 min) EPSP-spike potentiation induced by the repeated K +o increases (Fig. 4a and b). However, this delayed potentiating effect of K +o episodes during the CNQX treatment can be explained by the slow washout kinetics of CNQX (Section 3). In other words, the K +o episodes are probably capable of inducing EPSP-spike potentiation during the first 60 min of the “washout” period but their effects are masked by the blocking action of the remaining CNQX on synaptic responses. One of the more important questions addressed by our findings is how the relatively short-term increases in intracellular Ca2+ induced by the repeated brief K +o increases can maintain EPSP-spike potentiation and the in vitro kindling-like state in the CA1 area for a long time. Like APV, KN62 (3 µM), a selective membrane permeable inhibitor of CaMKII, blocked the EPSP-spike potentiation and the maintenance phase of the in vitro kindling-like state. Lisman (1994) proposed that CaMKII molecules became phosphorylated during a synaptic event that was to be “remembered”, even after the event was over. It has been proposed that Ca2+dependent autophosphorylation of this kinase can switch the molecule into an “on” state which retains its activity even after Ca2+ is removed (Dosemeci and Albers, 1996). These findings indicate that Ca2+ could participate in the induction of EPSP- spike potentiation and the maintenance mechanisms of the in vitro kindling-like state via autophosphorylation of CaMKII. Repeated brief increases in K +o can activate CaMKII via both the voltage-sensitive Ca2+ channels and the NMDA-receptor operated Ca2+ ionophores localized on the soma/proximal dendrites and distal dendrites in CA1 pyramidal neurons, respectively (Westenbrock et al., 1990; Muller and Connor, 1991; Christie et al., 1995; Huber et al., 1995; Rao and Craig, 1997). It is known that Ca2+ influx via NMDA-operated ionophores activates this kinase concentrated in the postsynaptic density of glutamate synapses to induce synaptic long-term potentiation (Malenka et al., 1989; Fleck et al., 1992; Onyang et al., 1997). Our findings indicate that EPSPspike potentiation induced by the repeated increases in K +o depends on the activation of NMDA receptors. Such NMDA-dependent EPSP-spike potentiation can essentially facilitate the development of epileptiform events in CA1 pyramidal neurons in response to the repeated increases in K +o induced in the CA1 hippocampal area under certain pathological conditions (for example, under repeated episodes of hypoxia or ischaemia). On the basis of our previous (Semyanov and Godukhin, 1997; Semyanov et al., 2000) and present data, and
209
from the literature findings, we should like to propose the following hypothetical scheme for the involvement of L-type voltage-sensitive Ca2+ channels, NMDAdependent Ca2+ channels and CaMKII in the development of epileptiform bursts in CA1 pyramidal neurons in response to repeated increases in K +o (Fig. 7). The repetition of these K +o increases induces long-lasting changes in the properties of L-type Ca2+ channels in CA1 pyramidal neurons. The intrinsic properties of the cells are modified such that the pyramidal neurones them-
Fig. 7. Hypothetical scheme for the involvement of L-type voltagesensitive Ca2+ channels, NMDA-dependent Ca2+ channels and CaMKII in the development of epileptiform bursts in CA1 pyramidal neurons in response to repeated brief K +o increases. It is suggested that synchronous epileptiform bursts in many CA1 pyramidal neurones are associated with the prolonged membrane depolarisation shifts which are responsible for driving the rapid bursts of Na+-dependent action potentials. The prolonged membrane depolarisation shifts in CA1 pyramidal neurons are induced by influx of Ca2+ via L-type voltage-sensitive Ca2+ channels (VSCCs). Influxes of Ca2+ via L-type VSCCs and NMDA receptor-dependent Ca2+ channels also induce EPSP-spike potentiation (ESP) which can facilitate the epileptiform burst development (possibly via activation of L-type VSCCs). These effects of Ca2+ on EPSP-spike potentiation are mediated by the activation of CaMKII. The repeated K +o increases switch an inactive form of CaMKII to an active one which in turn induces long-lasting changes in L-type VSCC properties (possibly via an increase in a population of the channels which are active at resting membrane potentials). The properties of these channels are modified such that the pyramidal neurons become hyperresponsive to excitatory inputs. In other words, the repeated increases in K +o induce a sustained decrease in the threshold of evoked epileptiform discharge development. The broken arrows in the scheme show the pathways terminating the depolarising shift in membrane potential induced by the Ca2+ influx via L-type VSCC. SC/CF, Schaffer collaterals/commissural fibres; IN, CA1 inhibitory interneurons.
210
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
selves become hyperresponsive to excitatory inputs. Epileptiform bursts are observed as multiple Na+-dependent action potentials superimposed on synchronized, L-type Ca2+-channel-mediated, depolarising shifts of membrane potential in many CA1 pyramidal neurons. This hypothesis is confirmed by the following data. Under normal conditions (without repeated K +o-episode stimulation), the most conspicuous electrophysiological property of CA1 pyramidal neurons is their regular—spiking but not burst firing—behaviour (Mason, 1993). The L-type Ca2+ channel has high conductivity and slow kinetics of inactivation, so that once activated by an excitatory input, it will sustain the depolarisation for tens of milliseconds, driving rapid bursts of fast-action potentials mediated by voltage-dependent sodium channels. It is possible that the repeated K +o episodes can increase a population of dihydropyridine-sensitive Ca2+ channels (possibly Ltype) which are active at resting membrane potentials in hippocampal CA1 pyramidal neurons (Magel et al., 1996). EPSP-spike potentiation could increase the burst firing probability of the cells through the activation of L-type voltage-sensitive Ca2+ channels. The contribution of this component of long-term potentiation to L-type Ca2+channel-mediated depolarising shift expression can be essential for a sustained decrease in the threshold for development of evoked epileptiform discharges (in vitro kindling-like state) in CA1 pyramidal neurons. The facilitatory effect of NMDA-dependent Ca2+ influx on the development of epileptiform bursts is mediated by CaMKII which participates in the maintenance mechanisms of the in vitro kindling-like state in the CA1 hippocampal area in response to repeated increases in K +o. This scheme also includes Ca2+-dependent K + channels and GABA-mediated recurrent inhibition which could be involved in a termination of the depolarising shift and epileptiform burst in CA1 pyramidal neurons (Jefferys, 1990; Traub and Dingledine, 1990; Bradford, 1995; Le Beau and Alger, 1998). We propose that this hypothetical scheme can help to explain some of the cellular–molecular mechanisms of the plastic alterations in CA1 pyramidal neurons associated with the start of kindling-induced epileptogenesis. Acknowledgements This work was supported by grants from INTAS Open Call (# 97-0382) and the Russian Foundation of Basic Research (# 99-04-48444). References Anderson, W.W., Lewis, D.V., Swartzwelder, H.S., Wilson, W.A., 1986. Magnesium-free medium activates seizure-like events in the rat hippocampal slice. Brain Research 398, 215–219.
Bernard, C., Wheal, H.V., 1995. Plasticity of AMPA and NMDA receptor-mediated epileptiform activity in a chronic model of temporal lobe epilepsy. Epilepsy Research 21, 95–107. Bliss, T.V.P., Lynch, M.A., 1988. Long-term potentiation of synaptic transmission in the hippocampus: properties and mechanisms. In: Landfield, P.W., Deadwyler, S.A. (Eds), Long-term potentiation: from biophysics to behavior. Neurology and Neurobiology. Alan R. Liss, Inc. 35 (4), 3–72. Bradford, H.F., 1995. Glutamate, GABA and epilepsy. Progress in Neurobiology 47, 477–511. Butler, L., Silva, A., Tonegawa, S., McNamara, J.O., 1993. Enhanced seizure susceptibility yet impairment of kindling development in α-calcium-calmodulin kinase II mutant mice. Society for Neuroscience Abstracts 19, 1030. Christie, B.R., Eliot, L.S., Ito, K.I., Miyakawa, H., Johnston, D., 1995. Different Ca2+ channels in soma and dendrites of hippocampal pyramidal neurons mediate spike-induced Ca2+ influx. Journal of Neurophysiology 73, 2553–2557. Della Paschoa, O.E., Kruk, M.R., Hamstra, R., Voskuyl, R.A., Danhov, M., 1997. Seizure patterns in kindling and cortical stimulation models of experimental epilepsy. Brain Research 770, 221–227. Dosemeci, A., Albers, R.W., 1996. A mechanism for synaptic frequency detection through autophosphorylation of CaM kinase II. Biophysical Journal 70, 2493–2501. Fleck, M.W., Palmer, A.M., Barrionuevo, G., 1992. Potassium-induced long-term potentiation in rat hippocampal slices. Brain Research 580, 100–105. Fukunaga, K., Muller, D., Miyamoto, E., 1995. Increased phosphorylation of Ca2+/calmodulin-dependent protein kinase II and its endogenous substrates in the induction of long-term potentiation. Journal of Biological Chemistry 270 (11), 6119–6124. Goddard, G.V., McIntyre, D.C., Leech, C.K., 1969. A permanent change in brain function resulting from daily electrical stimulation. Experimental Neurology 15, 295–330. Huber, K.M., Mank, M.D., Kelly, P.T., 1995. LTP induced by activation of voltage-dependent Ca2+ channels requires protein kinase activity. NeuroReport 6, 1281–1284. Ito, I., Hidaka, H., Sugiyama, H., 1991. Effects of KN62, a specific inhibitor of calcium/calmodulin-dependent protein kinase II, on long-term potentiation in the rat hippocampus. Neuroscience Letters 121, 119–121. Jefferys, J.G.R., 1990. Basic mechanisms of focal epilepsies. Experimental Physiology 75, 127–162. Kuba, K., Kumamoto, E., 1990. Long-term potentiations in vertebrate synapses: a variety of cascades with common subprocesses. Progress in Neurobiology 34, 197–269. Le Beau, F.E.N., Alger, B.E., 1998. Transient suppression of GABAAreceptor-mediated IPSPs after epileptiform burst discharges in CA1 pyramidal cells. Journal of Neurophysiology 79, 659–669. Lisman, J., 1994. The CaM kinase II hypothesis for the storage of synaptic memory. Trends in Neurological Science 17 (10), 406– 412. Lothmann, E.W., Williamson, J.M., 1994. Closely spaced recurrent hippocampal seizures elicit two types of heightened epileptogenesis: a rapidly developing, transient kindling and slowly developing, enduring kindling. Brain Research 649, 71–84. McEachern, J.C., Shaw, C.A., 1996. An alternative to the LTP orthodoxy: a plasticity–pathology continuum model. Brain Research Reviews 22, 51–92. Magel, J.C., Avery, R.B., Christie, B.R., Johnston, D., 1996. Dihydropyridine-sensitive, voltage-gated Ca2+ channels contribute to the resting intracellular Ca2+ concentration of hippocampal CA1 pyramidal neurons. Journal of Neurophysiology 76 (5), 3460–3470. Malenka, R.C., Kauer, J.A., Perkel, J.D., Mauk, M.D., Kelly, P.T., Nicoll, R.A. et al., 1989. An essential role for post-synaptic calmodulin and protein kinase activity in long-term potentiation. Nature 340, 554–556.
A. Semyanov, O. Godukhin / Neuropharmacology 40 (2001) 203–211
Malinov, R., Schulmann, H., Tsien, R.W., 1989. Inhibition of postsynaptic PKC or CaMKII blocks induction but not expression of LTP. Science 245, 826–830. Mason, A., 1993. Electrophysiology and burst-firing of rat subicular pyramidal neurons in vitro: a comparison with area CA1. Brain Research 600, 174–178. Mody, I., Lambert, J.D.C., Heinemann, U., 1987. Low extracellular magnesium induces epileptiform activity and spreading depression in rat hippocampal slices. Journal of Neurophysiology 57, 869–888. Muller, W., Connor, J.A., 1991. Dendritic spines as individual neuronal compartments for synaptic Ca2+ responses. Nature 354, 73– 76. Murray, K.D., Gall, C.M., Benson, D.L., Jones, E.G., Isackson, P.J., 1995. Decreased expression of the alpha subunit of Ca2+/calmodulin-dependent protein kinase type II mRNA in the adult rat CNS following recurrent limbic seizures. Brain Research 32 (2), 221–232. Onyang, Y., Kantor, D., Harris, K.M., Schuman, E.M., Kennedy, M.B., 1997. Visualization of the distribution of autophosphorylated calcium/calmodulin-dependent protein kinase II after tetanic stimulation in the CA1 area of the hippocampus. The Journal of Neuroscience 17, 5416–5427. Otmakhov, N., Griffith, L.C., Lisman, J.E., 1997. Postsynaptic inhibitors of calcium/calmodulin-dependent protein kinase type II block induction but not maintenance of pairing-induced long-term potentiation. The Journal of Neuroscience 17 (14), 5357–5365. Rao, A., Craig, A.M., 1997. Activity regulates the synaptic localization of the NMDA receptor in hippocampal neurons. Neuron 19 (4), 801–812. Semyanov, A., Godukhin, O., 1997. Kindling-like state in rat hippocampal CA1 slices induced by the repeated short-term extracellular K + increases: the role of L-type Ca2+ channels. Neuroscience Letters 223, 177–180. Semyanov, A., Morenkov, E., Godukhin, O., 1997. The decreased susceptibility to the development of in vitro kindling-like state in hippocampal CA1 slices of rats sensitive to audiogenic seizures. Neuroscience Letters 230, 187–190. Semyanov, A., Savin, A., Morenkov, E., Godukhin, O., 2000. In vivo
211
hippocampal kindling occludes the development of in vitro kindling-like state in CA1 rat hippocampal slices. Epilepsy Research 38 (1), 75–85. Sihra, T.S., Pearson, H.A., 1995. Ca/calmodulin-dependent kinase II inhibitor KN62 attenuates glutamate release by inhibiting voltagedependent Ca2+ channels. Neuropharmacology 34 (7), 731–741. Tokumitsu, H., Chijiwa, T., Hagiwara, M., Mizutani, A., Terasawa, M., Hidaka, H., 1990. KN62, 1-[N,O-bis(isoquinolinesulfonyl)-Nmethyl-l-tyrosyl]-4-phenylpiperazine, a specific inhibitor of Ca2+/calmodulin-dependent protein kinase II. Journal of Biological Chemistry 265, 4315–4320. Traub, R.D., Dingledine, R., 1990. Model of synchronized epileptiform bursts induced by high potassium in CA3 region of rat hippocampal slice. Role of spontaneous EPSPs in initiation. Journal of Neurophysiology 64, 1009–1018. Wasterlain, C.G., Farber, D.B., 1984. Kindling alters the calcium/calmodulin dependent phosphorylation of synaptic plasma membrane proteins in rat hippocampus. Proceedings of the National Academy of Sciences of the United States of America 81, 1253–1257. Westenbrock, R.E., Ahlijanian, M.K., Catterall, W.A., 1990. Clustering of L-type Ca2+ channels at the base of major dendrites in hippocampal pyramidal neurons. Nature 347, 281–284. Wheal, H.V., Bernard, C., Chad, J.E., Cannon, R.C., 1998. Pro-epileptic changes in synaptic function can be accompanied by pro-epileptic changes in neuronal excitability. Trends in Neurological Science 21 (4), 167–174. Wilson, W.A., Stasheff, S., Swartzwelder, S., Clark, S., Anderson, W.W., 1992. The role of NMDA receptors in in vitro epileptogenesis. In: Avanzini, G., Engel, J. Jr., Faricllo, R., Heinemann, U. (Eds.), Neurotransmitters in Epilepsy (Epilepsy Research Supplement). Elsevier, Amsterdam, pp. 157–166 Suppl. 8. Wyllie, D.J.A., Nicoll, R.A., 1994. A role for protein kinases and phosphatases in the Ca2+-induced enhancement of hippocampal AMPA receptor-mediated synaptic responses. Neuron 13, 635–643. Zhou, X.R., Suzuki, T., Shimizu, H., Nishino, H., 1994. Amygdala kindling activates the phosphorylation of Ca2+/calmodulin-dependent protein kinase II in rat hippocampus. Neuroscience Letters 12, 45–48.