
The level of the mRNAs encoding the AMPA-selective glutamate receptors-A and -B, alternatively splice variants, Flip and Flop, was studied by in situ hybridization in the brains of rats kindled by Schaffer collateral/commissural-fiber stimulation. In comparison to control animals, the expression level of the Flip variant of both GluR-A and GluR-B mRNAs was bilaterally enhanced in the dentate granule neurons of kindled animals 24 h after last-generalized seizure, whereas no obvious alterations were observed in the GluR-A Flop and GluR-B Flop mRNA variants. In kindled animals, studied 1 month after the last seizure, GluR-A Flip and GluR-B Flip mRNA had returned to control levels. We suggest that these changes may result in an enhanced glutamate receptor sensitivity in the fascia dentata during kindling.
While GABAergic inhibition plays a major role in the regulation of neuronal excitability, a role for altered GABAergic inhibition in the pathogenesis of epilepsy remains to be proven. The demonstration that GABAA receptors are composed of multiple subunits and that the properties and pharmacology of GABAA receptors are different for different subunit combinations, suggests that GABAA receptor heterogeneity may be of importance in determining the properties of GABAergic inhibition in different regions of the nervous system. While it is clear that GABAA receptor heterogeneity is present in the nervous system, a role for receptor heterogeneity in the pathogenesis of epilepsy remains uncertain. GABAA receptor heterogeneity may have implications for the treatment of epilepsy. It is quite possible that drugs which regulate GABAergic function may have variable efficacy in different regions of the nervous system due to expression of receptors with subunits that have different sensitivity to allosteric regulators. In situ hybridization studies indicate the colocalization of alpha 1 beta 1 gamma 2L and delta subunit mRNAs in hippocampal dentate gyrus granule cells while only the alpha 1, beta 1 and gamma 2L and not the delta subunit mRNAs colocalize in the pyramidal cells of the hippocampus. The reduced rate of acute desensitization and the slow recovery of GABA-evoked currents typical of delta-containing subunit combinations could generate tonic inhibition via long-lasting IPSPs in the dentate gyrus and thus play a role in preventing seizures. By the same rationale, a reduction in the level of expression of the delta subunit mRNA in the dentate gyrus or its absence as in the hippocampal pyramidal cells could be associated with a reduced seizure threshold. Furthermore, it is likely that there are developmental changes in the stoichiometry or subunit composition of GABAA receptors rendering the developing nervous system more or less sensitive to the effects of GABAergic anticonvulsant drugs. In addition to the heterogeneous expression of GABAA receptors, other issues concerning the regulation of GABAergic function are of potential importance. The regulatory events that control the expression of specific receptor subtypes and levels of GABA receptors are unknown. To understand the role of GABAA receptor heterogeneity in the pathogenesis of epilepsy will require the combination of biophysical and molecular biological techniques. It will be important to determine not only whether the properties of GABAA receptors have been altered in a specific form of epilepsy but also whether gene expression has been altered.
To investigate possible changes in the GABA, receptor agonist site in the CAI area and fascia dentata of rats kindled by stimulation of Schaffer collaterals, a quantitative autoradiographic study of the [H-3]muscimol binding was carried out. Two kindled groups were studied, at 24 h (fully kindled stage) and at 28 days (long-term stage) after the last class V seizure. Several concentrations of [3H]muscimol were tested in the range of the high/intermediate (5-40 nM) and low-affinity (60-100 nM) binding sites. In the fully kindled group, the binding over the complete range of tested [H-3]muscimol concentrations was significantly increased by 30-50% in the fascia dentata, while the binding was significantly decreased by 10-25% in the CA1 area. The high/intermediate-affinity binding Was still significantly increased by 20-30% in the fascia dentata 28 days after the last seizure. In this long-term group there was still a significant decrease of 10-18% of the low-affinity binding in the CA1 area. These results show that kindling epileptogenesis induces long-lasting changes in the GABA(A) receptor agonist binding sites that are region specific. We hypothesize that the changes encountered at the fully kindled stage, i.e. increased binding in the fascia dentata and decreased binding in the CA1 area, may underly the electrophysiologically observed increased paired-pulse depression of field potentials in the former and the decreased paired-pulse depression in the latter area [Kamphuis et al. (1992) Neurosci. Lett. 141, 101-105; Kamphuis et al. (1988) Blain Res, 440, 205-215; Zhao and Leung (1991) Brain Res. 564, 220-229; Zhao and Leung (1992) Brain Res. 582, 163-167].We conclude that the observed changes may not only contribute to the induction of kindling epileptogenesis but may also play a role in the maintenance of the kindled State.