N-methylaspartate receptors mediate epileptiform activity evoked in some, but not all, conditions in rat neocortical slices

Bioluminescence calcium imaging of network dynamics and their cholinergic modulation in slices of cerebral cortex from male rats

The activity of neuronal ensembles was monitored in neocortical slices from male rats using wide-field bioluminescence imaging of a calcium sensor formed with the fusion of green fluorescent protein and aequorin (GA) and expressed through viral transfer. GA expression was restricted to pyramidal neurons and did not conspicuously alter neuronal morphology or neocortical cytoarchitecture. Removal of extracellular magnesium or addition of GABA receptor antagonists triggered epileptiform flashes of variable amplitude and spatial extent, indicating that the excitatory and inhibitory networks were functionally preserved in GA-expressing slices. We found that agonists of muscarinic acetylcholine receptors largely increased the peak bioluminescence response to local electrical stimulation in layer I or white matter, and gave rise to a slowly decaying response persisting for tens of seconds. The peak increase involved layers II/III and V and did not result in marked alteration of response spatial properties. The persistent response involved essentially layer V and followed the time course of the muscarinic afterdischarge depolarizing plateau in layer V pyramidal cells. This plateau potential triggered spike firing in layer V, but not layer II/III pyramidal cells, and was accompanied by recurrent synaptic excitation in layer V. Our results indicate that wide-field imaging of GA bioluminescence is well suited to monitor local and global network activity patterns, involving different mechanisms of intracellular calcium increase, and occurring on various timescales.

Brief activation of GABAergic interneurons initiates the transition to ictal events through post-inhibitory rebound excitation

Activation of γ-aminobutyric acid (GABA_A) receptors have been associated with the onset of epileptiform events. To investigate if a causal relationship exists between GABA_A receptor activation and ictal event onset, we activated inhibitory GABAergic networks in the superficial layer (2/3) of the somatosensory cortex during hyperexcitable conditions using optogenetic techniques in mice expressing channelrhodopsin-2 in all GABAergic interneurons. We found that a brief 30ms light pulse reliably triggered either an interictal-like event (IIE) or ictal-like ("ictal") event in the in vitro cortical 4-Aminopyridine (4-AP) slice model. The link between light pulse and epileptiform event onset was lost following blockade of GABA_A receptors with bicuculline methiodide. Additionally, recording the chronological sequence of events following a light pulse in a variety of configurations (whole-cell, gramicidin-perforated patch, and multi-electrode array) demonstrated an initial hyperpolarization followed by post-inhibitory rebound spiking and a subsequent slow depolarization at the transition to epileptiform activity. Furthermore, the light-triggered ictal events were independent of the duration or intensity of the initiating light pulse, suggesting an underlying regenerative mechanism. Moreover, we demonstrated that brief GABA_A receptor activation can initiate ictal events in the in vivo 4-AP mouse model, in another common in vitro model of epileptiform activity, and in neocortical tissue resected from epilepsy patients. Our findings reveal that the synchronous activation of GABAergic interneurons is a robust trigger for ictal event onset in hyperexcitable cortical networks.

Regulation of ERK1/2 mitogen-activated protein kinase by NMDA-receptor-induced seizure activity in cortical slices

Extracellular signal-regulated kinase 1/2 (ERK1/2) that belongs to a subfamily of mitogen-activated protein kinases (MAPKs) plays diverse roles in the central nervous system. Activation of ERK1/2 has been observed in various types of neuronal excitation, including seizure activity in vivo and in vitro, as well as in NMDA-receptor (NMDA-R)-dependent long-term potentiation in the hippocampus. On the other hand, recent studies in cultured neurons have shown that NMDA-R stimulation could result in either ERK1/2 activation or non-activation, depending on the pharmacological manipulations. To assess NMDA-R-dependent regulation of ERK1/2 activity in vivo, here we examined the effect of NMDA-R-induced seizure activity on ERK1/2 activation by using rat cortical slice preparations. NMDA-R-dependent seizure activity introduced by Mg2+ -free condition did not cause ERK1/2 activation. On the other hand, when picrotoxin was added to concurrently suppress GABAA-receptor-mediated inhibition, profound ERK1/2 activation occurred, which was accompanied by strong phospho-ERK1/2-staining in the superficial and deep cortical layer neurons. In this case, prolonged membrane depolarization and enhanced burst action potential firings, both of which were much greater than those in Mg2+ -free condition alone, were observed. Differential ERK1/2 activation was supported by the concurrent selective increase in phosphorylation of a substrate protein, phospho-site 4/5 of synapsin I. These results indicate that NMDA-R activation through a release from Mg2+ -blockade, which accompanies enhancement of both excitatory and inhibitory synaptic transmission, was not enough, but concurrent suppression of GABAergic inhibition, which leads to a selective increase in excitatory synaptic transmission, was necessary for robust ERK1/2 activation to occur within the cortical network.

The role of glutamatergic inputs onto parvalbumin-positive interneurons: relevance for schizophrenia

Cognitive impairment, a core feature of schizophrenia, has been suggested to arise from a disturbance of gamma oscillations that is due to decreased neurotransmission from the parvalbumin (PV) subtype of interneurons. Indeed, PV interneurons have uniquely fast membrane and synaptic properties that are crucially important for network functions such as feedforward inhibition or gamma oscillations. The causes leading to impairment of PV neurotransmission in schizophrenia are still under investigation. Interestingly, NMDA receptors (NMDARs) antagonism results in schizophrenia-like symptoms in healthy adults. Additionally, systemic NMDAR antagonist administration increases prefrontal cortex pyramidal cell firing, apparently by producing disinhibition, and repeated exposure to NMDA antagonists leads to changes in the GABAergic markers that mimic the impairments found in schizophrenia. Based on these findings, PV neuron deficits in schizophrenia have been proposed to be secondary to (NMDAR) hypofunction at glutamatergic synapses onto these cells. However, NMDARs generate long-lasting postsynaptic currents that result in prolonged depolarization of the postsynaptic cells, a property inconsistent with the role of PV cells in network dynamics. Here, we review evidence leading to the conclusion that cortical disinhibition and GABAergic impairment produced by NMDAR antagonists are unlikely to be mediated via NMDARs at glutamatergic synapses onto mature cortical PV neurons.

Homeostatic increase in excitability in area CA1 after Schaffer collateral transection in vivo

PURPOSE

Epilepsy is a significant long-term consequence of traumatic brain injury (TBI) and is likely to result from multiple mechanisms. One feature that is common to many forms of TBI is denervation. We asked whether chronic partial denervation in vivo would lead to a homeostatic increase in the excitability of a denervated cell population.

METHODS

To answer this question, we took advantage of the unique anatomy of the hippocampus where the input to the CA1 neurons, the Schaffer collaterals, could be transected in vivo with preservation of their outputs and only minor cell death.

KEY FINDINGS

We observed a delayed increase in neuronal excitability, as apparent in extracellular recordings from hippocampal brain slices prepared 14 days (but not 3 days) post lesion. Although population spikes in slices from control and lesioned animals were comparable under resting conditions, application of solutions that were mildly proconvulsive (high K(+) , low Mg(2+) , low concentrations of bicuculline) produced increases in the number of population spikes in slices from lesioned rats, but not in slices from unlesioned sham controls. Denervation did not produce changes in several markers of γ-aminobutyric acid (GABA)ergic synaptic inhibition, including the number of GABAergic neurons, α1 GABA(A) receptor subunits, the vesicular GABA transporter, or miniature inhibitory postsynaptic currents.

SIGNIFICANCE

We conclude that chronic partial denervation does lead to a delayed homeostatic increase in neuronal excitability, and may, therefore, contribute to the long-term neurologic consequences of TBI.

Glutamate receptor subtypes mediating synaptic activation of prefrontal cortex neurons: relevance for schizophrenia

Schizophrenia may involve hypofunction of NMDA receptor (NMDAR)-mediated signaling, and alterations in parvalbumin-positive fast-spiking (FS) GABA neurons that may cause abnormal gamma oscillations. It was recently hypothesized that prefrontal cortex (PFC) FS neuron activity is highly dependent on NMDAR activation and that, consequently, FS neuron dysfunction in schizophrenia is secondary to NMDAR hypofunction. However, NMDARs are abundant in synapses onto PFC pyramidal neurons; thus, a key question is whether FS neuron or pyramidal cell activation is more dependent on NMDARs. We examined the AMPAR and NMDAR contribution to synaptic activation of FS neurons and pyramidal cells in the PFC of adult mice. In FS neurons, EPSCs had fast decay and weak NMDAR contribution, whereas in pyramidal cells, EPSCs were significantly prolonged by NMDAR-mediated currents. Moreover, the AMPAR/NMDAR EPSC ratio was higher in FS cells. NMDAR antagonists decreased EPSPs and EPSP-spike coupling more strongly in pyramidal cells than in FS neurons, showing that FS neuron activation is less NMDAR dependent than pyramidal cell excitation. The precise EPSP-spike coupling produced by fast-decaying EPSCs in FS cells may be important for network mechanisms of gamma oscillations based on feedback inhibition. To test this possibility, we used simulations in a computational network of reciprocally connected FS neurons and pyramidal cells and found that brief AMPAR-mediated FS neuron activation is crucial to synchronize, via feedback inhibition, pyramidal cells in the gamma frequency band. Our results raise interesting questions about the mechanisms that might link NMDAR hypofunction to alterations of FS neurons in schizophrenia.

Dopamine D1 receptor activation regulates sodium channel-dependent EPSP amplification in rat prefrontal cortex pyramidal neurons

Dopamine (DA) effects on prefrontal cortex (PFC) neurons are essential for the cognitive functions mediated by this cortical area. However, the cellular mechanisms of DA neuromodulation in neocortex are not well understood. We characterized the effects of D1-type DA receptor (D1R) activation on the amplification (increase in duration and area) of excitatory postsynaptic potentials (EPSPs) at depolarized potentials, in layer 5 pyramidal neurons from rat PFC. Simulated EPSPs (sEPSPs) were elicited by current injection, to determine the effects of D1R activation independent of modulation of transmitter release or glutamate receptor currents. Application of the D1R agonist SKF81297 attenuated sEPSP amplification at depolarized potentials in a concentration-dependent manner. The SKF81297 effects were inhibited by the D1R antagonist SCH23390. The voltage-gated Na+ channel blocker tetrodotoxin (TTX) abolished the effects of SKF81297 on sEPSP amplification, suggesting that Na+ currents are necessary for the D1R effect. Furthermore, blockade of 4-AP- and TEA-sensitive K+ channels in the presence of TTX significantly increased EPSP amplification, arguing against the possibility that SKF81297 up-regulates currents that attenuate sEPSP amplification. SKF81297 application attenuated the subthreshold response to injection of depolarizing current ramps, in a manner consistent with a decrease in the persistent Na+ current. In addition, D1R activation decreased the effectiveness of temporal EPSP summation during 20 Hz sEPSP trains, selectively at depolarized membrane potentials. Therefore, the effects of D1R activation on Na+ channel-dependent EPSP amplification may regulate the impact of coincidence detection versus temporal integration mechanisms in PFC pyramidal neurons.

Emergence of a spermine-sensitive, non-inactivating conductance in mature hippocampal CA1 pyramidal neurons upon reduction of extracellular Ca2+: dependence on intracellular Mg2+ and ATP

Large and protracted elevations of intracellular [Ca(2+)] and [Na(+)] play a crucial role in neuronal injury in ischemic conditions. In addition to excessive glutamate receptor activation, other ion channels may contribute to disruption of intracellular ionic homeostasis. During episodes of ischemia, extracellular [Ca(2+)] falls significantly. Here we report the emergence of an inward current in hippocampal CA1 pyramidal neurons in acute brain slices from adult mice upon reduction/removal of [Ca(2+)](e). The magnitude of the current was 100-300pA at -65mV holding potential, depending on intracellular constituents. The current was accompanied by intense neuronal discharge, observed in both whole-cell and cell-attached patch configurations. Sustained currents and increased neuronal firing rates were both reversed by restoration of physiological levels of [Ca(2+)](e), or by application of spermine (1mM). The amplitudes of the sustained currents were strongly reduced by raising intracellular [Mg(2+)], but not by extracellular [Mg(2+)] increases. Elevated intracellular ATP also reduced the current. This conductance is similar in several respects to the "calcium-sensing, non-selective cation current" (csNSC), previously described in cultured mouse hippocampal neurons of embryonic origin. The dependence on intracellular [ATP] and [Mg(2+)] shown here, suggests a possible role for this current in disruption of ionic homeostasis during metabolic stress that accompanies excessive neuronal stimulation.

Synaptic hyperexcitability of deep layer neocortical cells in a genetic model of absence seizures

We used sharp-electrode, intracellular recordings in an in vitro brain slice preparation to study the excitability of neocortical neurons located in the deep layers (>900 microm from the pia) of epileptic (180-210-days old) Wistar Albino Glaxo/Rijswijk (WAG/Rij) and age-matched, non-epileptic control (NEC) rats. Wistar Albino Glaxo/Rijswijk rats represent a genetic model of absence seizures associated with generalized spike and wave (SW) discharges in vivo. When filled with neurobiotin, these neurons had a typical pyramidal shape with extensive apical and basal dendritic trees; moreover, WAG/Rij and NEC cells had similar fundamental electrophysiological and repetitive firing properties. Sequences of excitatory postsynaptic potentials (EPSPs) and hyperpolarizing inhibitory postsynaptic potentials (IPSPs) were induced in both the strains by electrical stimuli delivered to the underlying white matter or within the neocortex; however, in 24 of 55 regularly firing WAG/Rij cells but only in 2 of 25 NEC neurons, we identified a late EPSP that (1) led to action potential discharge and (2) was abolished by the N-methyl-D-aspartate (NMDA) receptor antagonist 3,3-(2-carboxypiperazine-4-yl)-propyl-1-phosphonate (20 microM; n = 8/8 WAG/Rij cells). Finally, we found that the fast and slow components of the stimulus-induced IPSPs recorded during the application of glutamatergic receptor antagonists had similar reversal potentials in the two strains, while the peak conductance of the fast IPSP was significantly reduced in WAG/Rij cells. These findings document an increase in synaptic excitability that is mediated by NMDA receptors, in epileptic WAG/Rij rat neurons located in neocortical deep layers. We propose that this mechanism may be instrumental for initiating and maintaining generalized SW discharges in vivo.

Glutamate transporters regulate excitability in local networks in rat neocortex

Excitatory postsynaptic currents (EPSCs) in the neocortex are principally mediated by glutamate receptors. Termination of excitation requires rapid removal of glutamate from the synaptic cleft following release. Glutamate transporters are involved in EPSC termination but the effect of uptake inhibition on excitatory neurotransmission varies by brain region. Epileptiform activity is largely mediated by a synchronous synaptic activation of cells in local cortical circuits, presumably associated with a large release of glutamate. The role of glutamate transporters in regulating epileptiform activity has not been addressed. Here we examine the effect of glutamate transport inhibition on EPSCs and epileptiform events in layer II/III pyramidal cells in rat neocortex. Inhibiting glutamate transporters with DL-threo-beta-benzyloxyaspartic acid (TBOA; 30 microM) had no effect on the amplitude or decay time of evoked, presumably alpha-amino-3-hydroxyl-5-methyl-isoxazolepropionic acid-mediated, EPSCs. In contrast, the amplitude and duration of epileptiform discharges were significantly enhanced. TBOA resulted also in a decreased threshold for evoking epileptiform activity and an increased probability of occurrence of spontaneous epileptiform discharges. TBOA's effects were not inhibited by the group I and II metabotropic glutamate receptors antagonist (S)-alpha-methyl-4-carboxyphenylglycine or the kainate receptor antagonist [(3S,4aR, 6S, 8aR)-6-((4-carboxyphenyl)methyl-1,2,3,4,4a,5,6,7,8,8a-decahydroisoquinoline-3-carboxylic acid]. D-(-)-2-amino-5-phosphonovaleric acid could both prevent excitability changes by TBOA and block already induced changes. Dihydrokainate (300 microM) had effects similar to TBOA suggesting involvement of the glial transporter GLT-1. Inhibiting glutamate transport increases local network excitability under conditions where there is an enhanced release of glutamate. Our results indicate that uptake inhibition produces an elevation of extracellular glutamate levels and activation of N-methyl-D-aspartate receptors.

alpha-[11C]-Methyl-L-tryptophan PET identifies the epileptogenic tuber and correlates with interictal spike frequency

Epilepsy surgery has been successfully performed in patients with tuberous sclerosis complex (TSC) and seizures arising from a restricted epileptogenic area. The outcome of cortical excision depends on accurate pre-surgical identification of the epileptogenic tuber. [11C] alpha-methyl-L-tryptophan (alpha-MTrp) was originally developed to measure serotonin synthesis in vivo with positron emission tomography (PET). However in pathologic conditions its uptake may also depend on the synthesis of quinolinic or kynurenic acid via the kynurenine pathway. Increased levels of serotonin and quinolinic acid have been observed in epileptogenic lesions, raising the possibility that alpha-MTrp PET may localize the epileptogenic area. The aim of this study was to correlate alpha-MTrp PET uptake with the localization of the epileptogenic area and with interictal spike frequency in patients with TSC. alpha-MTrp uptake was measured in 8 patients (2 males, mean age 29.6+/-14.9 years, range 3-50 years) with intractable partial epilepsy due to TSC. All patients underwent scalp EEG monitoring during the PET scan. In four (50%), increased uptake of alpha-MTrp occurred in the epileptogenic area alone. Two (25%) patients showed multifocal abnormalities and the remaining two (25%) did not show focal changes. PET localization was mostly seen in patients with frequent interictal abnormalities on the EEG. Furthermore, there was a significant correlation between alpha-MTrp uptake and the frequency of interictal spikes (r=0.6; P<0.05). alpha-MTrp PET is a promising diagnostic tool in the localization of the epileptogenic area in patients with TSC.

Pre- rather than co-application of vigabatrin increases the efficacy of tiagabine in hippocampal slices

PURPOSE

The antiepileptic drug vigabatrin (VGB) increases intracellular availability of the inhibitory transmitter gamma-aminobutyric acid (GABA) by inhibition of GABA-transaminase. A blockade of the GABA uptake is the main mechanism of action of tiagabine (TGB). Based on this, the two antiepileptic drugs (AEDs) can be speculated to act synergistically so that their combined antiepileptic efficacy is supraadditive.

METHODS

To test this, experiments were performed on hippocampal slices of guinea-pigs. As an epilepsy model, epileptiform field potentials (EFPs) were induced by omission of Mg2+ from the bath solution and recorded in stratum pyramidale of the CA3 region. VGB (7.5 microM) and TGB (0.75 microM) were added to the superfusate.

RESULTS

VGB, given alone, failed to decrease the repetition rate of EFPs. Similarly, TGB applied alone only transiently led to a nonsignificant reduction of the EFP frequency. Combining VGB and TGB, their suppressive efficacy increased, yielding a significant reduction of EFP frequency, which, however, again did not persist. Pretreatment of the preparations with VGB for 2 h, followed by additional application of TGB, or TGB alone, drastically and persistently potentiated the effects.

CONCLUSIONS

These results demonstrate that VGB and TGB show favorable pharmacodynamic interactions, provided VGB is allowed to block intracellular GABA degradation before GABA uptake block by TGB.

Coactivation of Local Circuit NMDA Receptor Mediated epsps Induces Lasting Enhancement of Minimal Schaffer Collateral epsps in Slices of Rat Hippocampus

Lasting enhancement of minimal Schaffer collateral epsps in CA1 pyramidal neurons was induced when these epsps were coactivated with putative local circuit N-methyl-d-aspartate (NMDA) receptor-mediated epsps. In transverse slices of hippocampus, minimal epsps were evoked by stimulating the Schaffer collateral/commissural pathway within CA3 stratum radiatum and the local circuit axons were stimulated within CA1 stratum oriens or alveus using fine, two barrel glass micropipettes. Minimal Schaffer collateral epsps were insensitive to d-2-amino-5-phosphonovalerate (AP-5) and displayed conventional voltage relations. Local epsps were sensitive to AP-5 and increased in amplitude and duration with membrane depolarization. High frequency tetanic stimulation of the minimal Schaffer collateral epsps induced no lasting enhancement. However, pairing of the Schaffer epsp with a frequency potentiated local epsp resulted in a delayed, but lasting increase in the Schaffer epsp. When the local epsp did not exhibit frequency potentiation during pairing, no enhancement resulted. When either epsp was contaminated with an ipsp, no lasting increase resulted. When transmission through the alveus was blocked by focal application of local anaesthetic, traditional protocols for long-term potentiation (LTP) induction were relatively ineffective. These results indicate that a significant source of the NMDA receptor activation required for the induction of LTP in the Schaffer collateral input to CA1 derives from local circuit connections between CA1 pyramidal neurons and, moreover, that lasting enhancement can be induced with low frequency activation, at postsynaptic resting potential, in the presence of Mg2+ and with local inhibition intact.

Distinct firing patterns of neuronal subtypes in cortical synchronized activities

Cortical neurons, especially GABAergic interneurons, are composed of very diverse subtypes. It remains to be investigated whether each subtype shows a unique firing pattern during the synchronized activities generated by the intracortical circuit. By lowering extracellular Mg(2+) in vitro, we induced NMDA receptor-dependent spontaneous activities in the rat frontal cortex at 30 degrees C. After a series of spontaneous depolarization shifts, the long bursts occurred. The long bursts were composed of initial discharges and fast run-like potentials (FRLP) (4-10 Hz). Large inhibitory currents were induced at the initial discharge. After the strong inhibition decreased, the FRLP started. However, the periodical inhibition survived during the FRLP. At each phase of the synchronized activities, cortical neuron types exhibited distinct firing patterns. Pyramidal cells increased firing frequency periodically up to approximately 25-55 Hz during the FRLP cycles. Fast-spiking (FS) cells fired at the highest frequency in the initial discharge, up to 400 Hz, and could continue firing faster than 200 Hz for several seconds. In the FRLP, the firing frequency of FS cells rhythmically increased up to 150 Hz. In contrast, large cholecystokinin basket cells fired, very similarly to pyramidal cells, at each phase. Somatostatin and vasoactive intestinal polypeptide cells fired faster than pyramidal cells at the initial discharge, but showed the similar firings to pyramidal cells during the FRLP. The firing patterns of cortical neurons are not only determined by the strength and temporal pattern of synchronized inputs but also strongly dependent on the neuronal subtype with specific physiological, chemical, and morphological characteristics.

Epileptiform propagation patterns mediated by NMDA and non-NMDA receptors in rat neocortex

PURPOSE

The neocortex can generate various forms of epileptiform activity, including one that depends on N-methyl-D-aspartate (NMDA)-type glutamate receptors (NMDARs), and another dependent on non-NMDA-type (AMPA) glutamate receptors (AMPARs). Previous work in vitro suggests that both forms of activity are initiated by neurons of layer 5, but the spatial patterns of horizontal propagation have been studied only for the AMPAR form. We have tested the hypothesis that both types of epileptiform activity spread via common pathways in one cortical layer, suggesting that lamina-specific intervention might selectively interrupt both.

METHODS

Slices of rat somatosensory cortex were maintained in vitro and treated with the gamma-aminobutyric acid type A (GABA(A))-receptor antagonist picrotoxin. Single all-or-none epileptiform discharges were evoked with an electrical stimulus, and extracellular microelectrodes were used to track the vertical and lateral spread of the discharges.

RESULTS

In both high and low concentrations of picrotoxin, the non-NMDAR antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) completely blocked propagation, whereas the NMDAR antagonist D-2-amino-5-phosphonovaleric acid (DAPV) only shortened the duration of discharges. When extracellular [Mg2+] was reduced in the presence of picrotoxin and CNQX, NMDAR-dependent epileptiform discharges could be initiated. NMDAR-dependent discharges spread at about one fifth the conduction velocity of AMPAR-dependent events. Analysis of spatiotemporal field-potential patterns suggested that both NMDAR- and AMPAR-mediated propagation involved early activity in layers 5 and 6, followed by larger-amplitude activity in upper cortical layers along the path of propagation.

CONCLUSIONS

Our results imply that a common pathway mediates the propagation of these two forms of epileptiform activity, and suggests that lamina-specific surgical intervention might maximize anticonvulsant effect while minimally disrupting cortical function.

Effect of a glutamate receptor antagonist (GYKI 52466) on 4-aminopyridine-induced seizure activity developed in rat cortical slices

In the present experiments we have tested the effect of the noncompetitive AMPA antagonist GYKI 52466 (20-80 microM) on spontaneous epileptic discharges developed as the consequence of 4-aminopyridine application in neocortex slices of adult rats. Parallel to the changes of spontaneous activity, the field potentials, evoked by electrical stimulation of the corpus callosum, were also analyzed. Glass microcapillary extracellular recording electrode was positioned in the third layer of the somatosensory cortex slice, while the stimulating electrode was placed at the border of the white and gray matter. 4-aminopyridine and GYKI 52466 were bath-applied. The application of 40 microM GYKI 52466 caused about 40% decrease in the frequency and the amplitude of spontaneous seizures as well as the duration of each discharges developed in 4-amino-pyridine. Pre-incubation with the AMPA antagonist effectively inhibited both the development of seizure activity and the maintenance of the discharges. GYKI 52466 also decreased the duration and amplitude of field responses evoked by stimulation of the corpus callosum. This inhibitory effect was dose-dependent. Our data in the in vitro cortex slice epilepsy model suggest that the non-competitive AMPA antagonist GYKI 52466 is a potent anticonvulsant and neuroprotective compound because it reduced the fully developed epileptic discharges or prevented their development.

Reduced Mg2+ blockade of synaptically activated N-methyl-D-aspartate receptor-channels in CA1 pyramidal neurons in kainic acid-lesioned rat hippocampus

Unilateral kainic acid lesion in the hippocampus caused a long-term change in the balance between excitatory and inhibitory drive onto CA1 pyramidal cells, making these cells hyperexcitable several weeks post-lesion. In this study, we have shown an enhanced N-methyl-D-aspartate receptor-mediated component in the excitatory synaptic transmission together with a reduced GABA(A) receptor-mediated inhibition in CA1 pyramidal cells one-week post kainic acid lesion. In these cells, pharmacologically isolated N-methyl-D-aspartate receptor-mediated whole-cell excitatory postsynaptic currents were significantly larger at negative holding potentials, and the voltage-dependence of N-methyl-D-aspartate receptor channels was shifted in the hyperpolarizing direction. The plot of relative conductance (g/gMax) shifted significantly (P<0.01) to more negative holding potentials by 19 mV (-28+/-4 mV in control slices and -47+/-4 mV in kainic acid slices) at the half maximal conductance point (g/gMax =0.5). This shift gives a larger N-methyl-D-aspartate receptor-mediated component in the excitatory synaptic transmission at resting membrane potentials (around -60 mV). The shifted voltage dependence is highly sensitive to extracellular Mg2+ ions. Moderate increases in [Mg2+]o from 1 mM to 2.6 mM more than compensated for the negative shift and effectively suppressed the population epileptiform bursting activity. Fitting the voltage dependence to an ionic block model revealed a higher dissociation constant of N-methyl-D-aspartate receptor channels for Mg2+ in kainic acid-lesioned slices (52 mM at 0 mV; 330 microM at -60 mV) than in control slices (7.7 mM at 0 mV; 93 microM at -60 mV). While a simple single site model adequately fitted the control data for [Mg2+]o at 1 mM and 2.6 mM, no consistent model of this form was found for the kainic acid-lesioned slices. These results revealed changed properties of N-methyl-D-aspartate receptor channels in the kainic acid-lesioned model of epilepsy. The reduced Mg2+ blockade of N-methyl-D-aspartate receptor channels contributed significantly to the epileptiform bursting activity.

NMDA receptor-mediated changes of spontaneous activity patterns in thalamocortical slice cultures

Spontaneous activity is a hallmark of the thalamocortical system in vivo. Up until now, in vitro preparations of this system have been shown to be spontaneously active only when inhibition was reduced or N-methyl-D-aspartate (NMDA) receptor-mediated currents were facilitated via low extracellular magnesium levels. This study investigated the dependence of spontaneous thalamocortical activity patterns on NMDA receptor function via variation of extracellular magnesium levels (0-1 mM) and by the application of the specific NMDA receptor-antagonist D-2-amino-5-phosphonovalerate (AP5) in the absence of magnesium. We used cocultures of rat neocortical and thalamic slices which have been shown to develop reciprocal synaptic connections similar to those in vivo. Multi-site extracellular recordings revealed that the cultures were spontaneously active at all concentrations of magnesium and AP5, albeit with a high variability among cultures. Activity consisted of burst-like events which were largely synchronized within as well as among the neural tissues, and thalamic background activity during periods of neocortical quiescence. Each tissue was capable of triggering activity in the other, indicating that both thalamocortical and corticothalamic synaptic connections were functional. With increasing magnesium concentration, activity rates declined in both tissues and the site of origin of the synchronous, burst-like events shifted from neocortex to thalamus. AP5 in magnesium-free perfusion solution had qualitatively similar effects. We conclude that thalamic activity is not as dependent on the facilitation of NMDA receptor-mediated currents as neocortical activity and consequently, that the thalamus is the pacemaker of thalamocortical synchronized activity in physiological in vitro conditions.

'Dormant' inhibitory neurons: do they exist and what is their functional impact?

The concept of dormant interneurons is proving to be hard to define precisely. We argue here that the term is best used as an operational description of interneurons which are not lost from the epileptic brain, but which fail to perform adequately. We present evidence for the existence of functionally dormant interneurons in the tetanus toxin model of chronic epilepsy, and we explore the roles of a partial dormancy (and also of charge-screening) in the acute low magnesium model of epilepsy.

Molecular mechanisms that underlie structural and functional changes at the postsynaptic membrane during synaptic plasticity

The synaptic plasticity that is addressed in this review follows neurodegeneration in the brain and thus has both structural as well as functional components. The model of neurodegeneration that has been selected is the kainic acid lesioned hippocampus. Degeneration of the CA3 pyramidal cells results in a loss of the Schaffer collateral afferents innervating the CA1 pyramidal cells. This is followed by a period of structural plasticity where new synapses are formed. These are associated with changes in the numbers and shapes of spines as well as changes in the morphometry of the dendrites. It is suggested that this synaptogenesis is responsible for an increase in the ratio of NMDA to AMPA receptors mediating excitatory synaptic transmission at these synapses. Changes in the temporal and spatial properties of these synapses resulted in an altered balance between LTP and LTD. These properties together with a reduction in the inhibitory drive increased the excitability of the surviving CA1 pyramidal cells which in turn triggered epileptiform bursting activity. In this review we discuss the insights that may be gained from studies of the underlying molecular machinery. Developments in one of the collections of the cogs in this machinery has been summarized through recent studies characterizing the roles of neural recognition molecules in synaptic plasticity in the adult nervous systems of vertebrates and invertebrates. Such investigations of neural cell adhesion molecules, cadherins and amyloid precursor protein have shown the involvement of these molecules on the morphogenetic level of synaptic changes, on the one hand, and signal transduction effects, on the other. Further complex cogs are found in the forms of the low-density lipoprotein receptor (LDL-R) family of genes and their ligands play pivotal roles in the brain development and in regulating the growth and remodelling of neurones. Evidence is discussed for their role in the maintenance of cognitive function as well as Alzheimer's. The molecular mechanisms responsible for the clustering and maintenance of transmitter receptors at postsynaptic sites are the final cogs in the machinery that we have reviewed. Postsynaptic densities (PSD) from excitatory synapses have yielded many cytoskeletal proteins including actin, spectrin, tubulin, microtubule-associated proteins and calcium/calmodulin-dependent protein kinase II. Isolated PSDs have also been shown to be enriched in AMPA, kainate and NMDA receptors. However, recently, a new family of proteins, the MAGUKs (for membrane-associated guanylate kinase) has emerged. The role of these proteins in clustering different NMDA receptor subunits is discussed. The MAGUK proteins are also thought to play a role in synaptic plasticity mediated by nitric oxide (NO). Both NMDA and non-NMDA receptors are highly clustered at excitatory postsynaptic sites in cortical and hippocampal neurones but have revealed differences in their choice of molecular components. Both GABAA and glycine (Gly) receptors mediate synaptic inhibition in the brain and spinal cord. Whilst little is known about how GABAA receptors are localized in the postsynaptic membrane, considerable progress has been made towards the elucidation of the molecular mechanisms underlying the formation of Gly receptors. It has been shown that the peripheral membrane protein gephyrin plays a pivotal role in the formation of Gly receptor clusters most likely by anchoring the receptor to the subsynaptic cytoskeleton. Evidence for the distribution as well as function of gephyrin and Gly receptors is discussed. Postsynaptic membrane specializations are complex molecular machinery subserving a multitude of functions in the proper communication between neurones. Despite the fact that only a few key players have been identified it will be a fascinating to watch the story as to how they contribute to structural and functional plasticity unfold.

Gabapentin potentiation of the antiepileptic efficacy of vigabatrin in an in vitro model of epilepsy

1. An enhancement of promoted release of gamma-aminobutyric acid (GABA) and a change in GABA-metabolism have been suggested as mechanisms of action of gabapentin. Vigabatrin is supposed to act mainly via inhibition of GABA-transaminase but it also interferes with GABA-release and GABA-uptake. On the basis of these mechanisms of action, a pharmacodynamic interaction of the two antiepileptic drugs could be supposed which might be of relevance in the sense of a rational polypharmacy. 2. To address the aforementioned hypothesis, experiments were carried out on hippocampal slices (n=107) of guinea-pigs (n=70). Epileptiform field potentials (e.f.p.) were induced by omission of magnesium from the bath solution and recorded in the stratum pyramidale of the CA3 region. Gabapentin (30-600 microM; 5.1-102.72 microg ml(-1)), vigabatrin (50-200 microM, 6.45-25.8 microg ml(-1)) and the GABA(A)-receptor antagonist bicuculline (100 microM) were added to the bath solution for 3 h. 3. Gabapentin, in concentrations up to 600 microM, failed to decrease the repetition rate or duration of e.f.p. (n=19). However, vigabatrin, evoked a dose-dependent reduction of the repetition rate of e.f.p. For a concentration of 100 microM (12.9 microg ml(-1)) there was a reduction down to 48+/-5% (mean+/-s.e.mean) of the initial value within 3 h (n=11). With simultaneous administration of vigabatrin (100 microM) and gabapentin (60 microM) for 3 h (n=15), the repetition rate of e.f.p. decreased down to 8+/-3%, which is significantly different from the values obtained after administration of 100 microM vigabatrin alone (P<0.0001). Both, the antiepileptic effect of vigabatrin alone and the enhancement by gabapentin were blocked by the GABA(A)-receptor antagonist bicuculline (100 microM, n=16). 4. These results demonstrate that gabapentin is able to augment the antiepileptic effects of vigabatrin significantly. It is possible that a change in the GABA-release machinery is induced by vigabatrin which then can be augmented by gabapentin.

Characterization of the in vitro antiepileptic activity of new and old anticonvulsant drugs

1. The in vitro antiepileptiform effects of some old and new anticonvulsants in the experimental model of the "epileptiform" hippocampal slice have been reviewed. 2. On the basis of their influence on in vitro epileptogenesis and basal neuronal excitability, anticonvulsants can be classified into three main categories: (1) anticonvulsants (prototypical drug phenytoin) affecting basal neuronal excitability but not epileptogenesis; (2) anticonvulsants (prototypical drugs barbiturates) affecting basal neuronal excitability and epileptogenesis; (3) anticonvulsants (prototypical drug felbamate) affecting epileptogenesis but not basal neuronal excitability. 3. It is concluded that the model of the "epileptiform" hippocampal slices can be considered a previsional test for the study and the screening of new anticonvulsant drugs.

Epileptiform discharge induced by 4-aminopyridine in magnesium-free medium in neocortical neurons: physiological and pharmacological characterization

An in vitro model of epileptiform activity was developed to study the role of excitatory and inhibitory neurotransmitters in the epileptogenesis. Intracellular recordings were obtained from rat neocortical slices exposed to 4-aminopyridine in a magnesium-free solution. Spontaneous epileptiform activity consisting of paroxysmal depolarization shifts with associated spontaneous depolarizing postsynaptic potentials were observed. The paroxysmal depolarization shifts were blocked either by D,L-2-amino-5-phosphonovalerate (50 microM), an N-methyl-D-aspartate receptor antagonist, or by 6-cyano-7-nitroquinoxaline-2.3-dione (10 microM), a non-N-methyl-D-aspartate receptor antagonist. These glutamate receptor antagonists also reduced the occurrence of spontaneous depolarizing postsynaptic potentials. Bicuculline methiodide, an antagonist of GABAA receptors, suppressed spontaneous depolarizing postsynaptic potentials, while it reduced the frequency of paroxysmal depolarization shifts and increased their duration. Hyperpolarization of the membrane potential by continuous current injection increased the frequency of paroxysmal depolarization shifts and reduced their duration, but it reduced the occurrence of spontaneous postsynaptic potentials. Paroxysmal depolarization shifts were blocked by tetrodotoxin (1 microM). The duration and the frequency of paroxysmal depolarization shift were reduced by dopamine (30-300 microM) in a dose-dependent manner. Our model suggests a different involvement of excitatory and inhibitory processes in the generation of epileptiform activity.

Serotonin blocks different patterns of low Mg2+-induced epileptiform activity in rat entorhinal cortex, but not hippocampus

Low Mg2+-induced epileptiform activity in the entorhinal cortex is characterized by an initial expression of seizure-like events followed by late recurrent discharges. Both these forms of activity as well as the transition between them were blocked by serotonin. In contrast, serotonin had little effect upon the epileptiform activity in areas CA3 and CA1 of the hippocampus. Both forms of epileptiform activity in the entorhinal cortex are sensitive to N-methyl-D-aspartate receptor antagonists and it is shown here that serotonin blocked both types of epileptiform activity through an effective concentration-dependent reduction of N-methyl-D-aspartate receptor-mediated excitatory postsynaptic potentials in deep layer entorhinal cortex cells. Serotonin also prolonged or even prevented the transition between the two types of epileptiform activity and we suggest that this may be through activation of the Na+/K+-ATPase. The resistance of epileptiform activity in CA1 and CA3 to serotonin was most likely related to the inability of serotonin to reduce Schaffer collateral-evoked excitatory postsynaptic potentials. Given the strong serotonergic inputs to both the hippocampus and entorhinal cortex, the differential sensitivity of the two regions to serotonin suggests functional differences. In addition since the late recurrent discharges in the entorhinal cortex are resistant to all clinically used anticonvulsants, serotonin may open new avenues for the development of novel anticonvulsant compounds.

Review of the role of inhibitory neurons in chronic epileptic foci induced by intracerebral tetanus toxin

Blocking inhibition provides one of the most common experimental means of triggering epileptic activity in hippocampus and neocortex. However, it has proved much more difficult to show that chronic models of epilepsies are due to disinhibition. One problem is knowing how much inhibition needs to be blocked to provide a sufficient mechanism for epileptic activity. We have found that inhibitory (GABAA) transmission, estimated from evoked monosynaptic IPSCs, must be reduced to 17% of their control amplitude (by 4-7 microM bicuculline) before hippocampal slices generate all-or-none epileptic discharges. Similar estimates of inhibition in chronic epileptic foci induced by intrahippocampal injection of tetanus toxin showed that monosynaptic IPSCs dropped to 10% of control in the injected hippocampus during the first 2 weeks after injection. At all other stages of the active epileptic foci in the two hippocampi the reduction in IPSCs was not alone sufficient for epileptic activity; at 4-6 weeks IPSCs were normal despite continued epileptic activity. One likely mechanism for the late epileptic activity is a reduction of either the intrinsic excitability, or the synaptic excitation, of inhibitory interneurons so they fail to be recruited normally. Alternative mechanisms include the formation of new excitatory connections, as found at modest levels in the dentate gyrus. Several mechanisms may play a part in chronic foci such as those induced by tetanus toxin, either acting together, or sequentially during the progression of the epileptic focus.

Hypoxia-induced hyperexcitability in vivo and in vitro in the immature hippocampus

Hypoxia is the most common cause of neonatal seizures and encephalopathy. We have previously developed an in vivo experimental model of perinatal hypoxia which exhibits age-dependent acute and chronic epileptogenic effects. Between postnatal day (P) 10-12, the rat exhibits acute seizure activity during global hypoxia, while no seizures are induced at earlier (P5) or older (P60) ages. Rats exposed to hypoxia between P10-12 have reduced seizure thresholds to chemical convulsants in adulthood. The nonNMDA antagonists NBQX appears to suppress both the acute and long term epileptogenic effects of hypoxia. The age-dependency of the hyperexcitable response to hypoxia in vivo can be reproduced in vitro using hippocampal slices. In Mg(2+)-free media, hypoxia induced ictal discharges within 60 s of onset in 79% of slices from normal P10 rat pups compared to 11% of adult slices (p < 0.001). Model systems such as that described here allow for correlation of in vitro and in vivo electrophysiology and should provide data regarding the pharmacological and physiological characteristics of hypoxia-induced seizure activity in the immature brain which could ultimately be applied to therapeutic strategies.

Extracts from rhizomes of Cyperus articulatus (Cyperaceae) displace [3H]CGP39653 and [3H]glycine binding from cortical membranes and selectively inhibit NMDA receptor-mediated neurotransmission

The marshland plant Cyperus articulatus (Cyperaceae) is commonly used in traditional medicine in Africa and Latin America to treat a wide variety of human diseases ranging from headache to epilepsy. We tested the hypothesis that the purported anti-epileptic effect of this plant might be due to a functional inhibition of excitatory amino acid receptors. One or several component(s) contained in the extracts inhibited the binding of [3H]CGP39653 to the NMDA recognition site and of [3H]glycine to the strychnine-insensitive glycine site of the NMDA receptor complex from rat neocortex. Water extracts from rhizomes of Cyperus articulatus dose-dependently reduced spontaneous epileptiform discharges and NMDA-induced depolarizations in the rat cortical wedge preparation at concentrations at which AMPA-induced depolarizations were not affected. We conclude that the purported beneficial effects of Cyperus articulatus might at least partially be due to inhibition of NMDA-mediated neurotransmission.

Effects of glutamate application on the rhythm of low magnesium-induced epileptiform activity in hippocampal slices of guinea-pigs

The extracellular concentration of glutamate has previously been reported to increase to more than 10-fold the basal level during seizure activity. In the present study, we tested whether localized increases in extracellular glutamate concentration influence the rhythm of epileptiform discharges in the low-magnesium epilepsy model. In hippocampal slices of guinea-pigs, epileptiform activity was induced by omission of magnesium from the bath fluid. Glutamate and its subreceptor agonists N-methyl-D-aspartate (NMDA) and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) were ejected into different strata of the CA3 and CA1 regions using microiontophoretic and micropressure application. Glutamate, NMDA and AMPA applied to the CA3 region, but not to the CA1 region, induced a short-lasting increase in epileptiform discharge frequency, often followed by a transient reduction. The effect was most pronounced with application into the stratum lacunosum-moleculare of the CA3 region and could only be evoked in slices exceeding 400 microns in thickness. The effects on the rhythm of epileptiform discharges induced by NMDA and AMPA were blocked by their specific receptor antagonists. They were not influenced by application of GABAA and GABAB receptor antagonists. Changes in somatic membrane potential of CA3 pyramidal neurons did not correlate with changes in the rhythm of epileptiform discharges elicited in this region. The transient suppression of epileptiform discharges that followed the increase in discharge frequency was abolished by an adenosine A1 receptor antagonist. We propose that localized increases in extracellular glutamate concentration modify the rhythm of epileptiform discharges due to changes in neuronal network activity.

A role for synaptic and network plasticity in controlling epileptiform activity in CA1 in the kainic acid-lesioned rat hippocampus in vitro

1. Stimulation of the surviving afferents in the stratum radiatum of the CA1 area in kainic acid-lesioned hippocampal slices produced graded epileptiform activity, part of which (> 20%) involved the activation of N-methyl-D-aspartate (NMDA) receptors. There was also a failure of synaptic inhibition in this region. In this preparation, we have tested the effects of low-frequency stimulation (LFS; 1 Hz for 15 min) on synaptic responses and epileptiform activity. 2. LFS resulted in long-term depression (LTD) of excitatory synaptic potentials (EPSPs), long-term decrease of population spike amplitudes (PSAs) and EPSP-spike (E-S) potentiation. Evoked epileptiform activity was reduced but neurons had a higher probability of discharge. LTD could be reversed by subsequent tetanic stimulation whereas E-S dissociation remained unchanged. Synaptic and network responses could be saturated towards either potentiation or depression. However, E-S potentiation was maximal following the first conditioning stimulus. 3. NMDA receptor-mediated responses were pharmacologically isolated. LFS resulted in LTD of synaptic responses, long-term decrease of PSAs and E-S depression. These depressions could not be reversed by subsequent tetanic stimulation. alpha-Amino-3-hydroxy-5-methylisoxazolepropionic acid (AMPA) and NMDA receptor-mediated responses were then measured in isolation before and following conditioning stimuli. LFS was shown to simultaneously produce LTD of AMPA and NMDA receptor-mediated responses. E-S potentiation of the AMPA component and E-S depression of the NMDA component occurred coincidentally. 4. LTD of AMPA and NMDA receptor-mediated responses were shown to be NMDA dependent. In contrast, E-S potentiation and depression occurred even when NMDA receptors were pharmacologically blocked. 5. These findings indicate that synaptic responses could be modified bidirectionally in the CA1 area of kainic acid-lesioned rat hippocampus in an NMDA receptor-dependent manner. However, E-S dissociations were independent of the activation of NMDA receptors, hinting at mechanisms different from those of synaptic LTD. We suggest that changes in E-S coupling were caused by a modification of the firing threshold of the CA1 pyramidal neurons. Furthermore, the firing mechanisms controlling NMDA and AMPA receptor-mediated network activity appeared to be different. The possible use of LFS applied to the hippocampus as a clinical intervention to suppress epileptiform activity is discussed.

Anticonvulsant drug effects on spontaneous thalamocortical rhythms in vitro: phenytoin, carbamazepine, and phenobarbital

When perfused with a medium containing no added Mg2+, rodent thalamocortical brain slices generate spontaneous generalized thalamocortical discharges of several types. Two of these discharges, termed simple and complex thalamocortical burst complexes (sTBCs and cTBCs), are physiologically and pharmacologically similar to the spike-wave discharges of generalized absence epilepsy and to the discharges underlying generalized tonic-clonic seizures, respectively. In a further characterization of the pharmacology of generalized thalamocortical discharges recorded in rodent thalamocortical slices, the actions of anticonvulsants effective in control of partial and generalized tonic-clonic seizures, but not generalized absence seizures, were studied on these rhythms. The effects of phenytoin, carbamazepine, and phenobarbital were tested against sTBCs and cTBCs recorded in vitro in rodent thalamocortical slices. When applied in clinically relevant concentrations, phenytoin and carbamazepine were very effective in reducing or blocking cTBCs. These drugs were much less effective in controlling sTBCs. Phenobarbital was effective in controlling both sTBCs and cTBCs, but the level of block was greater for cTBCs. Therefore, it appears that sTBCs and cTBCs are quite distinct in their relative sensitivity to anticonvulsant drugs, and this differential sensitivity parallels the relative effectiveness of these drugs in controlling generalized absence and generalized tonic-clonic seizures.

Erosion of inhibition contributes to the progression of low magnesium bursts in rat hippocampal slices

1. Bathing slices of rat hippocampus in media containing no magnesium ions results in epileptic discharges that originate in hippocampal area CA3. These discharges increase in severity gradually over periods of hours. 2. The progression of epileptic activity was much slower than the equilibration of extracellular magnesium activity and the resulting increase in strength of monosynaptic NMDA receptor-mediated excitation. Its time course matched that of a progressive decrease in pharmacologically isolated, evoked GABAA receptor-mediated inhibitory postsynaptic current (IPSC) in the CA3 pyramidal cells. Conductance decreased to 37 +/- 6% of control values after 4 h. Responses to exogenous GABA application decreased to 52 +/- 12%. 3. Maximal IPSC conductance in 0 mM extracellular Mg2+ ([Mg2+]o) also decreased gradually when epileptic activity was abolished by bath application of 20 microM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 50 microM D-2-amino-5-phosphonovaleric acid (D-APV) throughout the 4 h incubation period. It reached 61 +/- 8% of control values, a significantly smaller decline than that seen after 4 h of epileptic activity. 4. The decrease in mean IPSC conductance only partially reversed when the recording electrode contained 100 mM Mg2+. Complete recovery of IPSC strength occurred when electrodes also contained either 50 mM MgATP or 20 mM BAPTA. Reintroduction of 1 mM [Mg2+]o rapidly abolished epileptic activity and caused a slow, partial increase in IPSC conductance. 5. In the presence of 1 mM [Mg2+]o, GABAA receptor-mediated inhibition had to decrease to 17 +/- 11% of control values, in the presence of 4-7 microM bicuculline, to reach threshold for epileptic activity. 6. These data demonstrate a postsynaptic decrease in GABAA receptor-mediated inhibition in the in vitro low magnesium model of epilepsy. We propose that the apparent leaching of intracellular Mg2+ ([Mg2+]i) from cells leads to loss of ATP and consequent partial dephosphorylation of the GABAA receptor and that this is exacerbated by epileptic activity.

An in vitro model of persistent epileptiform activity in neocortex

An in vitro model of persistent epileptiform activity was developed to study the mechanisms involved in epileptogenesis. Extracellular recordings were obtained from rat neocortical slices exposed to magnesium-free solution for 2 h. During exposure to magnesium-free solution spontaneous epileptiform activity consisting of interictal bursting and ictal-like discharges were observed. Interestingly, this activity persisted for hours after the slices were returned to magnesium-containing control solution. The N-methyl-D-aspartate (NMDA) receptor antagonist CPP prevented the development of the epileptiform activity, while the non-NMDA receptor antagonist CNQX abolished the epileptiform discharge that persisted after slices were returned to control solution. These findings suggest there are two distinct phases in the development of epileptic activity in this model, namely, induction (mediated by NMDA receptor activity) and maintenance (supported largely by non-NMDA receptor activity). The similarities and possible parallels between the mechanisms underlying this epileptogenesis and other forms of use-dependent modification of synaptic excitation, such as long-term potentiation, are discussed. This in vitro model of neocortical epileptogenesis may provide insights into the events underlying the development of clinical partial epilepsy.

NMDA antagonists: antiepileptic-neuroprotective drugs with diversified neuropharmacological profiles

In conclusion, NMDA antagonists as anticonvulsants are especially active in preventing the generalization of the behavioural and electrical seizures and display a typical spectrum of in vitro antiepileptiform activities. In addition, based on in vitro and in vivo limbic kindled studies, the drugs should be regarded more as an antiepileptiform than as an anticonvulsant drugs. As neuroprotective drugs, NMDA antagonists are effective against many types of neuronal injury and show a window of activity which does not exceed 1-2 h, thus suggesting an influence of NMDA receptors in the 'early' or 'acute' mechanisms of brain damage. Among NMDA antagonists, glycine antagonists or the morphinans dextromethorphan and dextrorphan showed a spectrum of antiepileptiform and neuroprotective activities broader than other NMDA antagonists. The primary pharmacological activities of NMDA antagonists are accompanied by some effects including perturbation of many sensory, psychological or motor processes. Typical behavioural and EEG changes were also induced by the drugs. In spite of the side-effects elicited by the drugs, differential effects detected among the various classes of NMDA antagonists (i.e. lack of induction of typical EEG-behavioural effects and of typical cortical neurotoxicity) might render some of these suitable for full clinical application as anticonvulsant-neuroprotective drugs.

Animal models of perinatal asphyxia: contributions, contradictions, clinical relevance

Animal models have contributed immensely to our understanding of hypoxic ischemic encephalopathy in the newborn. A number of animal models have been used, including both primate and subprimate species. Although the Rhesus monkey model has a dramatically similar pathological distribution of brain injury when compared with the human, other pathologic processes secondary to asphyxia may be more appropriately assessed in other species. The maxim that because primates are closer on the phylogenetic tree to humans than are subprimates all observations in the primate are applicable to the human is simply not true. Understanding of the neurochemical consequences of asphyxia in the past decade have arisen from experiments primarily in the neonatal rat. We have come to understand that not only is the hypoxic event of major significance, but that, once reperfused, reoxygenation causes further injury. Free-radical generation following reperfusion may be massive and may further contribute to cell membrane injury. These observations have lead to rational theoretic approaches to the treatment of hypoxic ischemic brain injury. On the other hand, previously used treatments such as osmotic agents and glucocorticoids would appear to be not only inefficacious but hazardous in the treatment of hypoxic ischemic brain injury. The role of nitric oxide (NO) in the pathogenesis of brain injury is yet uncertain, but there is little doubt that it plays a significant role. Although survival of the immature animal subjected to hypoxic environment is longer than in the mature animal, the central nervous system of the immature animal is more sensitive to glutamate and N-Methyl-D-aspartate (NMDA) receptor-mediated injury.

An animal model of hypoxia-induced perinatal seizures

Clinically, neonatal hypoxic encephalopathy is commonly associated with seizure activity. Here we describe a rodent model of cerebral hypoxia in which there is are age dependent effects of hypoxia, with hypoxia inducing seizure activity in the immature rat, but not in the adult. Global hypoxia (3-4% O2) induced acute seizure activity during a window of development between postnatal day (P5-17), peaking at P10-12. Animals which had been rendered hypoxic between P10-12 had long term decreases in seizure threshold, while animals exposed at younger (P5) or older (P60) ages did not. Antagonists of excitatory amino acid (EAA) transmission appear to be superior to benzodiazepines in suppressing the acute and long term effects of perinatal hypoxia, suggesting involvement of the EAA system in these phenomena. No significant histologic damage occurs in this model, suggesting that functional alterations take place in neurons when exposed to an hypoxic insult at a critical developmental stage. Future work is directed at evaluating molecular and cellular events underlying the permanent increase in seizure susceptibility produced by this model.

The role of excitatory amino acids in the generation of spontaneous hippocampal oscillations

We examined the role of N-methyl-D-aspartate (NMDA) and non-NMDA glutamate receptors in generating spontaneous CA3 field potential oscillations in the hippocampal slice. Non-NMDA EPSPs are responsible for a portion of the spontaneous activity recorded in standard perfusion medium. NMDA receptors are not activated when inhibition is intact, however, recurrent excitation via NMDA receptors accounts for an increasing proportion of the low frequency (2-4 Hz) rhythms produced as inhibition is progressively blocked by low concentrations of the GABAA antagonist, penicillin. Production of these rhythms involves complex interactions among NMDA, non-NMDA and GABAA receptors. NMDA EPSPs can drive the rhythm in the absence of non-NMDA receptors only when inhibition is suppressed by more than 50%. Otherwise non-NMDA EPSPs appear to be necessary to depolarize neurons before NMDA EPSPs can be activated.

Thalamocortical rhythm generation in vitro: extra- and intracellular recordings in mouse thalamocortical slices perfused with low Mg2+ medium

Mouse thalamocortical slices maintain synaptic connections between thalamus and cortex. When perfused with low Mg2+ medium, thalamocortical slices exhibited spontaneous thalamocortical 2-6 s long bursts of rhythmic activity every 15-30 s, which was potentiated by pentylenetetrazol and abolished by severing thalamocortical connections. These oscillations were similar in frequency and duration to normal and pathological thalamocortical rhythms in vivo. In vitro studies of these thalamocortical rhythms may prove valuable in understanding cellular factors important in rhythm generation.

Fluctuations in pyramid-pyramid excitatory postsynaptic potentials modified by presynaptic firing pattern and postsynaptic membrane potential using paired intracellular recordings in rat neocortex

Single axon excitatory connections between pairs of neocortical pyramidal neurons were studied using paired intracellular recordings in layers II/III and IV of coronal slices of adult rat somatosensory/motor cortex. Excitatory postsynaptic potentials evoked with different presynaptic firing patterns and at different postsynaptic membrane potentials were compared. Two methods of statistical analysis were used in attempts to determine whether changes in mean excitatory postsynaptic potential amplitude were due to presynaptic or postsynaptic modifications. Analysis of the decrease in mean excitatory postsynaptic potential amplitude associated with increases in presynaptic firing rate were consistent with a change in probability of transmitter release. Paired pulse depression appeared to exhibit both presynaptic and postsynaptic components when the interspike interval was < 10 ms, but could be explained simply by a decrease in probability of release with interspike intervals between 10 and 80 ms. Previous studies had demonstrated that these excitatory postsynaptic potentials are partially mediated by N-methyl-D-aspartate receptors. In contrast to the apparently presynaptic effects of firing pattern, postsynaptic membrane depolarization appeared to produce an increase in quantal amplitude. In addition to this increase at low frequencies, a form of frequency-dependent, self-potentiation involving the recruitment of an additional, longer-latency postsynaptic component occurred at higher presynaptic firing rates. The possibility is discussed that two different mechanisms are involved in the replacement of vesicles at release sites. Over a few tens of milliseconds (paired-pulse depression) availability of releasable transmitter may be determined by the rate of replacement of discharged vesicles from a readily releasable pool of vesicles. Over longer periods of firing at 0.33-2 Hz, the readily releasable pool may become exhausted and require replenishment. Postsynaptic depolarization increases the duration of these excitatory postsynaptic potentials, facilitating summation and enables two components of excitatory postsynaptic potential enhancement at N-methyl-D-aspartate receptor-mediated synapses; one that is present at all firing rates and relates simply to voltage dependent events and one that occurs at higher firing rates and involves a gradual, time dependent event. These data also indicate that the optimal pyramidal firing pattern if another pyramid is to be activated is a tonic, or brief burst pattern at relatively low repetition rates. Long bursts of many presynaptic spikes recruit little that is not activated by pairs of spikes. This situation is in stark contrast to the results obtained in the following paper in which excitatory inputs from pyramids to non-pyramids are described.

The effects of selective glutamate receptor antagonists on synchronized firing bursts in layer III of rat visual cortex

In the rat visual cortex in vitro, single-shock stimulations applied to the border between layer VI and the white matter evoke synchronized burst-firing by units in layer III. We have examined the effects of glutamate receptor antagonists on this activity, with antagonists applied via the bath to allow correlation of effects with concentrations. All synaptically driven components (recorded extracellularly as field potential 'S2' spikes, dipoles 'W1' and 'W2', and coinciding single-unit spikes) were inhibited by greater than 90% in 1.0 mM kynurenic acid and in 3 or 10 microM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, which selectively blocks AMPA/kainate receptors). S2 spike amplitudes were reduced by half in 0.7 microM CNQX. 2-Amino-5-phosphonovalerate (APV), a specific blocker of NMDA receptors, did not prevent S2 spike burst or horizontal spread of bursting within layer III. However, APV reduced the duration of synchronized bursts and the slower potentials which followed. In Mg(2+)-free medium, new components appeared which were APV-sensitive: (1) low amplitude spikes, distributed spatially like S2 spike, but recurring more slowly, and (2) slow potentials, distributed spatially like W1 and W2 potentials, but lasting for hundreds of milliseconds. The amplitudes of these spikes were reduced by half in 3 microM D-APV. Our data imply that: (1) glutamate receptors play a major role in mediating local, excitatory neurotransmission in the supragranular layers of neocortex, with NMDA and AMPA/kainate subtypes each subserving somewhat different functions; (2) AMPA/kainate receptors mediate rapid excitatory transmission between layer III neurons, responsible for driving the first 15 ms of synchronized bursts; (3) currents gated by NMDA receptors determine the duration of coherent firing bursts, and drive asynchronous neuronal firing following bursts; and (4) under conditions which circumvent block by extracellular Mg2+, activation of NMDA receptors greatly enhances and prolongs the response to single-shock stimulations. In vivo, activation of layer III neurons is likely to depend significantly upon currents gated by NMDA receptors whenever repetitively firing excitatory inputs summed over several tens of milliseconds provide enough depolarization to lift block by extracellular Mg2+.

Excitatory amino acid receptors appear to mediate paroxysmal depolarizing shifts in rat neocortical neurons in vitro

This study was designed to assess some of the contributions of excitatory amino acids to locally evoked responses in neurons in slices from frontal motor cortex in Sprague-Dawley rats. Intracellular recordings were obtained from 54 cortical neurons. Paroxysmal depolarization shifts (PDS) were evoked by local single pulse stimulation in cortex or in a small number of cases (n = 2) occurred spontaneously. These potentials could be abolished by application of kynurenic acid, a broad spectrum excitatory amino acid receptor antagonist. They were enhanced in Mg(2+)-free medium and could then be antagonized by application of D,L-2-amino-5-phosphonovalerate (AP5), a selective blocker of the N-methyl-D-aspartate (NMDA) receptors.

Assessment of epileptogenic potential: experimental, clinical and epidemiological approaches

There are experimental, clinical and epidemiological methods of assessing the epileptogenic potential of psychotropic drugs. In the laboratory it has been shown that there is a range of cellular and synaptic processes in the cerebral cortex and hippocampus that give rise to epileptiform neuronal activity. In addition to the classical suppression of GABA-mediated inhibitory synaptic mechanisms, in vitro studies in animal models of epilepsy and on human tissue suggest a prominent role for the N-methyl D-aspartate (NMDA) subtype of excitatory amino acid receptors. Any mechanism that leads to the depolarization of the neurones is likely to result in a facilitation of the NMDA-receptor involvement in excitatory neurotransmission. This is particularly true in the cortex and hippocampus where the densities of the NMDA-receptor are highest. Data are presented in this paper on how this epileptogenic mechanism can be studied in vitro. In humans, the importance of an accurate diagnosis is stressed and the advantages and disadvantages of routine EEG recordings and ambulatory monitoring discussed. Descriptions of large-scale systems of drug safety monitoring and their application to the assessment of the epileptogenic properties of psychotropic drugs are given.

Cesium potentiates epileptiform activities induced by bicuculline methiodide in rat neocortex maintained in vitro

We report that extracellular application of cesium (Cs+, 3 mM) potentiated the epileptiform discharge evoked by GABAA-receptor antagonist bicuculline methiodide (BMI 50 microM) in rat neocortical slices maintained in vitro. Cs+ changed BMI-induced epileptiform burst of a few hundred milliseconds evoked by extracellular focal stimuli into epileptiform discharge only a few seconds long (1.8-7 s). Moreover, Cs+ induced the appearance of spontaneously occurring epileptiform activities (0.038-0.15 Hz). Simultaneous intracellular/extracellular recordings indicated that each intracellular epileptiform burst was correlated with a field discharge. Variation of the membrane potential modified only the amplitude of the epileptiform burst and did not alter its frequency of occurrence, indicating that each discharge was a synchronous population event. The epileptiform discharges were not blocked by the N-methyl-D-aspartate (NMDA) receptor antagonist 3-((+-)-2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid (CPP 5-10 microM). In contrast, the non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX 0.5-5 microM) greatly reduced the duration of each epileptiform discharge by abolishing its afterdischarges in a concentration-dependent manner. This reduction in duration was accompanied by an increase in frequency of occurrence, however. After blockade of non-NMDA receptors with CNQX, a CPP-sensitive spontaneous discharge could be observed. These findings indicate that the inorganic cation Cs+ applied extracellularly can induce spontaneously occurring epileptiform activities in BMI-treated neocortical slices. In addition, receptors of excitatory amino acids play a major role in synchronizing this type of Cs+/BMI-induced spontaneous epileptiform activities.

Epileptogenic effect of hypoxia in the immature rodent brain

The response to cerebral hypoxia/ischemia may be different in the neonate compared to other age groups. An in vivo model was developed in the rat to determine whether there are age-dependent differences in the effects of hypoxia on electroencephalographic (EEG) activity. EEG recordings were obtained from Long Evans hooded rats deprived of oxygen at five ages: postnatal days 5 to 7, 10 to 12, 15 to 17, 25 to 27, and 50 to 60. Oxygen concentration was varied from 0, 2, 3, and 4% between animals. EEGs were recorded in all animals before, during, and at 1 hour after exposure to the hypoxic condition and at 1 to 7 days afterward in a subset of animals. All animals were deprived of oxygen until the onset of apnea and bradycardia to 20 to 40% of baseline heart rate values. Hypoxia resulted in isoelectric EEG significantly more frequently in the animals deprived of oxygen at postnatal days 25 to 27 and 50 to 60 than in the younger age groups. A highly significant effect was that the animals deprived at postnatal days 5 to 17 revealed a high incidence of epileptiform EEG activity during hypoxia. In contrast, the older animals exhibited only rare isolated EEG spikes before reaching an isoelectric EEG. The severity of hypoxia-induced epileptiform EEG changes was highest in the animals subjected to moderately hypoxic conditions (3% and 4% oxygen) at postnatal days 10 to 12. Furthermore, epileptiform changes persisted for hours to days following prolonged episodes of hypoxia in the younger animals. This study demonstrates a unique response of the immature brain to exhibit epileptiform activity during hypoxia.

Horizontal long-term potentiation of responses in rat somatosensory cortex

The search for mechanisms in neocortex that change synaptic efficacy and produce associative learning through activity-dependent processes has focused on the role of glutamate receptors of the N-methyl-D-aspartate (NMDA) type. NMDA receptor activation is necessary for the induction of long-term potentiation (LTP) in hippocampus and in neocortex. The effect of NMDA receptor activation is modulated in several ways, including Mg2+ block of the NMDA-dependent channel which prevents Ca2+ entry until neurons become partially depolarized. We report that when NMDA receptor activation is facilitated by lowering the extracellular [Mg2+] in the bathing medium, a low-frequency train presented in layer VI induces potentiated responses throughout a wide horizontal extent of layer II/III in neocortical slices. The response amplitudes potentiated by 34-200% over baseline values depending on the intensity of the repetitive conditioning stimulus and the distance of the recording site from the stimulus. At the same time that pre-existing evoked responses were potentiated, horizontal spread of activity in layer II/III was facilitated resulting in responses appearing at sites more than 1 mm from the stimulus. This enhanced transmission of responses persisted for greater than 2 h, and its induction was prevented by selective NMDA receptor antagonists. The results show that the horizontal spread of activity can be increased by altering the conditions of the stimulus presentation. We conclude that the mechanisms supporting LTP could determine the area of neocortex that is activated by a sensory input.

NMDA receptor-independent epileptiform activity induced by magnesium-free solution in rat amygdala neurons is blocked by CNQX

The effect of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), a specific non-N-methyl-D-aspartate (non-NMDA) receptor antagonist, on NMDA-independent epileptiform activity induced by Mg2(+)-free medium was studied in rat basolateral amygdala (BLA) neurons using intracellular recording techniques. Twenty to 30 min after switching to Mg2(+)-free medium, spontaneous and evoked epileptiform activity were observed in 16 out of 18 amygdala slices. Superfusion of D-2-amino-5-phosphonovalerate (D-APV), a selective NMDA receptor antagonist, reduced the duration of epileptiform activity by an average of 83%. However, there was a residual depolarizing component which remained in the presence of D-APV. This D-APV-resistant component could be completely blocked by CNQX suggesting that it is mediated by non-NMDA receptors.

The lateral spread of ictal discharges in neocortical brain slices

The in vitro brain slice technique was used to examine the lateral propagation of spontaneous electrographic ictal episodes across adjacent areas of guinea pig neocortex. Epileptiform activity was induced by perfusing slices with Mg-free artificial CSF. Simultaneous field potential recordings of ictal episodes were obtained from 4 micropipettes placed 1-3 mm apart across coronal slices in middle-cortical layers. Two types of lateral spread were characterized. Ictal episodes often developed focally and then spread as a slowly moving wavefront traveling at less than 0.3 mm/sec into adjacent, uninvolved cortex. By contrast, other episodes began nearly synchronously at all cortical sites. The individual afterdischarges that composed each ictal episode propagated rapidly across the cortex at greater than 30 mm/sec and were triggered by multiple pacemakers. Ictal episodes always terminated abruptly across the entire slice. The NMDA-receptor antagonist, 2-amino-phosphono-valerate, applied focally between recording sites, blocked rapid propagation across treated areas and resulted in the emergence of spatially separate, independent pacemakers. Pacemaker failure is the proposed mechanism for simultaneous and generalized termination of ictal episodes in this in vitro model of epileptogenesis.

Magnesium removal induces paroxysmal neuronal firing and NMDA receptor-mediated neuronal degeneration in cortical cultures

Removal of extracellular Mg2+ triggered the onset of repetitive excitatory discharges in cultured murine cortical neurons, detected by recording with patch electrodes in the whole cell configuration. The discharges were suppressed by 100 microM D-2-amino-5-phosphonovalerate. Over the next 24-72 h substantial numbers of neurons, but not glia, degenerated, releasing lactate dehydrogenase to the bathing medium. The neuronal death induced by removal of extracellular Mg2+ could be attenuated by either 3 microM tetrodotoxin or 50 microM dextrorphan, and thus likely reflects excessive activation of N-methyl-D-aspartate receptors triggered by excitatory discharges. This Mg2+ removal model may be a useful model in which to study certain aspects of epileptic neocortical injury.

The NMDA antagonist, MK-801, suppresses long-term potentiation, kindling, and kindling-induced potentiation in the perforant path of the unanesthetized rat

Antagonism of NMDA-mediated transmission by MK-801 has been shown to block long-term potentiation (LTP) in vitro and delay electrical kindling of the amygdala. The present experiment sought to examine the relationship between synaptic potentiation of the perforant path-granule cell synapse and development of perforant path kindling. MK-801 (0.1 and 1.0 mg/kg) blocked induction of LTP of the perforant path in the unanesthetized animal measured 24 h after train delivery. The 1.0 mg/kg dosage also increased afterdischarge (AD) thresholds, delayed kindling development from daily stimulation of the perforant path (means = 8.82 +/- 1.19 and 22.9 +/- 3.66 sessions to the first stage 5 seizure), and increased AD durations. Kindling produced a significant potentiation of the EPSP (47%) and population spike (49%) after the first evoked AD in control animals. No significant enhancement of either component of the field potential was observed in MK-801-treated animals. Animals treated with this dosage of MK-801, did, however, kindle in the absence of potentiation at this synapse. It was concluded that although NMDA-mediated potentiation may facilitate kindling, synaptic potentiation does not appear to be a critical requirement for kindling to develop. These findings support the notion that development of the burst response and not synaptic enhancement may be the critical physiological alteration that underlies the kindling phenomenon.