Long-lasting potentiation of epileptiform bursts by group I mGluRs is NMDA receptor independent.

In CA3 pyramidal cells of guinea pig hippocampal slices, picrotoxin (50 microM) elicited spontaneous, rhythmically recurring epileptiform bursts 285-435 ms in duration. The addition of (S)-3, 5-dihydroxyphenylglycine (DHPG, 50 microM, 90 min application), a selective group I metabotropic glutamate receptor (mGluR) agonist, resulted in a rapid-onset transient increase in burst frequency. This was followed by a slowly progressive increase in burst duration, with bursts reaching 1.5-3.8 s in duration at 90 min of DHPG application. The potentiation of epileptiform burst duration persisted at least 2 h after agonist removal. To determine whether N-methyl-D-aspartate (NMDA) receptor activation participates in the mGluR-induced potentiation of epileptiform bursts, experiments were carried out in the presence of D-2-amino-5-phosphonovaleric acid (APV, 50-100 microM), an NMDA receptor antagonist. Application of DHPG in the presence of APV resulted in a significantly enhanced transient increase in burst frequency. Nevertheless, when compared with the control described above, there was no significant alteration in the rate of development of the burst prolongation nor its persistence after washout. In other experiments, application of APV in the presence of fully developed mGluR-induced potentiated bursts (after 90 min washout of DHPG) resulted in no significant change in either burst frequency or duration. The data reveal that both induction and maintenance of group I mGluR-mediated potentiation of epileptiform discharges are NMDA receptor-independent processes, suggesting that epileptogenesis, when induced by group I mGluR activation, may occur and be sustained in the absence of NMDA receptor activation.

Role of metabotropic glutamate receptors in epilepsy.

Considerable information is available regarding the role of ionotropic glutamate receptors in the generation of interictal spikes. Progress in the study of metabotropic glutamate receptors (mGluRs) makes clear that activation of these receptors can contribute greatly to seizure discharges and epileptogenesis. The effects of activation of the different mGluR subgroups on neuronal hypersynchrony and the initiation and propagation of seizure discharges in hippocampal slices are discussed herein. To help one understand the mechanisms that underlie these effects, information regarding the action of mGluRs on cellular and synaptic properties is summarized. The data bring to the forefront the critical role of mGluRs in epilepsy and emphasize the anticonvulsant effects of group II and III mGluR activation as opposed to the convulsant action of group 1, which elicits seizure discharges and epileptogenesis in experimental models.

Group I mGluR-mediated silent induction of long-lasting epileptiform discharges.

Picrotoxin, an antagonist of GABA(A) receptor-mediated activity, elicited 320- to 475-ms synchronized bursts from the CA3 region of the guinea pig hippocampal slice. The addition of the selective group I metabotropic glutamate receptor (mGluR) agonist (S)-3, 5-dihydroxyphenylglycine (DHPG, 50 microM; 20- to 45-min application) gradually increased the burst duration to 1-4 s; this effect persisted 2-3 h after agonist removal. To determine whether the induction of this long-lasting effect required ongoing synchronized activity during mGluR activation, DHPG application in a second set of experiments took place in the presence of CNQX and (R, S)-CPP, antagonists of AMPA/kainate and NMDA receptors, respectively. In these experiments, synchronized bursting was silenced during the mGluR agonist application, yet after wash out of the DHPG and the ionotropic glutamate receptor (iGluR) blockers, epileptiform discharges 1-10 s in duration appeared and persisted at least 2 h after wash out of the mGluR agonist. The potentiated bursts were reversibly shortened by application of 500-1,000 microM (+)-alpha-methyl-4-carboxyphenylglycine (MCPG) or (S)-4-carboxyphenylglycine (4CPG), agents with group I mGluR antagonist activity. These data suggest that transient activation of group I mGluRs, even during silencing of synchronized epileptiform activity, may have an epileptogenic effect, converting brief interictal-length discharges into persistent seizure-length events. The induction process is iGluR independent, and the maintenance is largely mediated by the action of endogenous glutamate on group I mGluRs, suggesting that autopotentiation of the group I mGluR-mediated response may underlie the epileptogenesis seen here.

Requirement of protein synthesis for group I mGluR-mediated induction of epileptiform discharges.

Picrotoxin (50 microM) elicited rhythmic synchronized bursting in CA3 pyramidal cells in guinea pig hippocampal slices. Addition of the selective group I metabotropic glutamate receptor (mGluR) agonist (S)-3,5-dihydroxyphenylglycine (25 microM) elicited an increase in burst frequency. This was soon followed by a slowly progressive increase in burst duration (BD), converting the brief 250-520 ms picrotoxin-induced synchronized bursts into prolonged discharges of 1-5 s in duration. BD was significantly increased within 60 min and reached a maximum after 2-2.5 h of agonist exposure. The protein synthesis inhibitors anisomycin (15 microM) or cycloheximide (25 microM) significantly impeded the mGluR-mediated development of the prolonged bursts; 90-120 min of agonist application failed to elicit the expected burst prolongation. By contrast, the mGluR-mediated enhancement of burst frequency progressed unimpeded. Furthermore, protein synthesis inhibitors had no significant effect on the frequency or duration of fully developed mGluR-induced prolonged discharges. These results suggest that the group I mGluR-mediated prolongation of synchronized bursts has a protein synthesis-dependent mechanism.

Role of group I metabotropic glutamate receptors in the patterning of epileptiform activities in vitro.

In guinea pig hippocampal slices, picrotoxin elicited spontaneous epileptiform bursts 300-550 ms in duration. Additional application of (R,S)-3,5-dihydroxyphenylglycine or (S)-3-hydroxyphenylglycine, agonists specific for group I metabotropic glutamate receptors (mGluRs), or (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid, a broad-spectrum mGluR agonist, converted picrotoxin-induced interictal bursts into prolonged discharges measured on the order of seconds. The prolonged discharges induced by selective group I mGluR agonist continued to be produced for hours after agonist removal. The antagonists (S)-4-carboxyphenylglycine and (+)-alpha-methyl-4-carboxyphenylglycine had no effect on the duration of picrotoxin-induced interictal bursts. However, after agonist exposure, the persistent prolonged discharges occurring in the absence of agonist were reversibly suppressed by the antagonists, suggesting that the activity is maintained via endogenous activation of group I mGluRs by synaptically released glutamate. Our results suggest that, under some conditions, activation of group I mGluRs produces long-lasting enhancement of synaptic responses, mediated at least in part by autopotentiation of the group I mGluR response itself, which may result in the production of seizure discharges and contribute to epileptogenesis.

Synchronized oscillations in hippocampal CA3 neurons induced by metabotropic glutamate receptor activation.

The metabotropic glutamate receptor agonist (1S,3R)-1-aminocyclopentane-1,3 dicarboxylic acid (ACPD) at concentrations above 60 microM produced stereotypic oscillatory activity in CA3 pyramidal cells of rat hippocampal slices. This oscillatory activity consisted of trains of depolarizations with overriding action potentials. On average, individual trains lasted 7 sec and recurred at intervals of 24 sec. During each train, the constituent depolarizations achieved a maximum frequency of 27 Hz, then slowed to 8 Hz toward the end of the train. Extracellular and dual intracellular recordings suggested that this ACPD-induced oscillatory activity occurred synchronously in the CA3 population. The oscillations persisted in the presence of GABAA, GABAB, and NMDA receptor antagonists. In contrast, the oscillations were blocked by the AMPA/kainate receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10-30 microM). Likewise, the oscillations were blocked by the metabotropic glutamate receptor antagonists (+)-alpha-methyl-4-carboxyphenylglycine [(+)-MCPG; 1 mM], (S)-4-carboxy-3-hydroxyphenylglycine [(S)-4C3HPG; 1mM] and (S)-4-carboxyphenylglycine [(S)-4CPG; 1 mM]. The results suggest that activation of metabotropic glutamate receptors can result in a permissive state that allows AMPA/kainate receptor-mediated conductances to mediate synchronized activity among hippocampal CA3 neurons.

Role of metabotropic glutamate receptor subtypes in the patterning of epileptiform activities in vitro.

1. Epileptiform activities were elicited from the in vitro guinea pig hippocampus by the addition of picrotoxin. Modification of the picrotoxin-induced activities by agents active at metabotropic glutamate receptors (mGluRs) was examined using intracellular and extracellular recordings. 2. Picrotoxin typically elicited synchronized discharges (epileptiform bursts) in CA3 neurons. These spontaneously occurred at regular intervals. In the presence of (+)-alpha-methyl-4-carboxyphenylglycine (MCPG; 700-1,000 microM), an antagonist at multiple mGluR subtypes, the frequency of spontaneous epileptiform bursts decreased. In contrast, when the mGluR agonists (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (ACPD; 5 microM) or (2S,3S,4S)-alpha-(carboxycyclopropyl)-glycine (L-CCG-I; 10 microM) were added to the incubating medium, the frequency of epileptiform bursts increased. No consistent change in membrane potential, burst duration, nor burst afterhyperpolarization was associated with the changes in burst frequency. 3. When spontaneous burst frequency was reduced in MCPG, stimulation at a higher frequency entrained bursts without failure. Bursts evoked in MCPG were similar in waveform and amplitude to those evoked in the control state. 4. (S)-4-carboxyphenylglycine (S-4CPG) and (R,S)-4-carboxy-3-hydroxyphenylglycine (RS-4C3HPG) are antagonists at mGluR subtypes 1 and 5 but agonists at mGluRs 2 and 3. Addition of either of these agents increased the frequency of epileptiform bursts. 5. These results suggest that sufficient glutamate is released during epileptiform activities to activate mGluRs. The overall effect is to increase the frequency of synchronized discharges. This modulatory action on burst frequency is probably mediated via the mGluR 2 and 3 receptor subclass.

Synaptic modifications accompanying epileptogenesis in vitro: long-term depression of GABA-mediated inhibition.

We used an in vitro model similar to kindling to examine the processes underlying epileptogenesis. A 60 Hz train was applied every 5-10 min to the Schaffer collateral pathways in guinea pig hippocampal slices until epileptiform bursting was elicited in the CA3 region. The resultant alterations in both spontaneous and evoked activities were studied using intracellular recordings from CA3 pyramidal cells. An attempt was made to elucidate the synaptic modifications responsible for the conversion to this state of enhanced excitability. Analyses revealed that the emergence of epileptiform discharge was accompanied by a long-term depression of evoked inhibitory conductances. This tetanus-induced reduction of inhibition involved both the early and late phases of the evoked hyperpolarization, suggesting modification of both the GABAA and GABAB receptor-mediated events. Previous studies have suggested that NMDA receptor activation plays an important role in the induction of epileptiform activity in this model. Our data, showing that depression of inhibition can be induced in the presence of CNQX, is consistent with this hypothesis. The parallel development of long-term depression of inhibition and epileptiform bursting following tetanic stimulation suggests that plasticity of the inhibitory transmission process is a potential source of vulnerability contributing to epileptogenesis.