Search

The function of the gamma-aminobutyric acid type A receptor (GABA(A)R) is maintained by endogenous phosphorylation. We have shown that the corresponding kinase is the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH), using the locally produced glycolytic ATP. In addition, using cerebral tissue obtained during curative surgery for epilepsy, we showed that both the endogenous phosphorylation and the GABA(A)R function are significantly reduced in the "epileptogenic" cerebral cortex when compared to "control" tissue. This dysfunction likely contributes to seizure generation and/or transition from the interictal to the ictal state. Glucose utilization is decreased in the epileptogenic cortex of patients with partial epilepsy in the interictal state, but the relationship to the disorder remains unclear. We propose that this hypometabolism is related to the deficiency in the endogenous phosphorylation of GABA(A)R and the resulting greater lability of GABAergic inhibition. Several lines of evidences indeed suggest that GABAergic inhibition is costly in terms of metabolic consumption. The deficiency of this glycolysis-dependent mechanism may thus link epileptogenicity to glucose hypometabolism. The antiepileptic effect of ketogenic diets may be mediated by the subsequent rise in the NADH/NAD(+) index, which favors GABA(A)R endogenous phosphorylation and should contribute to restoration of GABAergic inhibition in the epileptogenic zone.

We have previously described a new endogenous phosphorylation mechanism that maintains ionotropic gamma-aminobutyric acid receptor (GABAAR) function and have shown that the kinase involved is the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH). This enzyme is closely associated with the receptor and phosphorylates the alpha1 subunit of the receptor. In a wealth of studies, a reduction in GABAergic neurotransmission has been suggested as a pathophysiological mechanism for human epilepsy. In this paper, we present evidence showing both reduced efficacy of this glycolysis-dependent GABAAR phosphorylation mechanism and of GABAergic inhibition in epileptogenic cortical tissue samples obtained during curative surgery of patients with partial seizures, as compared to non-epileptogenic human cortical tissue. This feature is not due to a reduction in the density of GABAAR alpha1 subunits in the epileptogenic tissue as evidenced by photoaffinity labeling. Maintaining the receptor in a phosphorylated state either by favoring the endogenous phosphorylation or by inhibiting a membrane-bound phosphatase sustains the GABAAR responses in the human epileptogenic cortex. The deficiency in endogenous phosphorylation and the associated decreased GABAAR function can account for transient failures of GABAergic inhibition and may favor seizure initiation and propagation. These findings suggest a functional link between epileptic pathology and the regional cerebral glucose hypometabolism observed in patients with partial epilepsies, since the dysfunction of the GABAergic mechanism is dependent on locally produced glycolytic ATP. They also point to new targets for developing molecules active in drug-resistant epilepsies.

GABAA receptors were characterized in cellular fractions isolated from adult bovine brain. The fraction enriched in cortical astrocytes is very rich in high-affinity binding sites for [3H]flunitrazepam and other "central-type" benzodiazepine ligands. The amount of specific [3H]flunitrazepam binding was more than five times higher in the glial fraction than in synaptosomal and perikaryal fractions. [3H]Flunitrazepam was displaced by low concentrations of clonazepam and other specific ligands for central GABAA receptors. Specific binding sites for GABA, flunitrazepam, barbiturates, and picrotoxin-like convulsants were characterized. Allosteric interactions between the different sites were typical of central-type GABAA receptors. The presence of alpha-subunit(s), as revealed by [3H]flunitrazepam photoaffinity labeling, was demonstrated in all brain fractions at molecular mas 51-53 kDa. Photoaffinity labeling was highest in the glial fraction. However, in primary cultured astrocytes from neonate rat cortex, no photoaffinity labeling was detected. Information obtained from astrocytes in culture should thus be taken with caution when extrapolated to differentiated astroglial cells. Our results actually show that, in mature brain, most of the fully pharmacologically active GABAA receptors are extrasynaptic and expressed in astroglia.

Partially purified (Na+,K+)-ATPase (E.C. 3.6.1.3.) was investigated in the epileptic cortex of audiogenic DBA/2 mice and in the primary and secondary foci of cats with acute or chronic freeze lesions. No differences in specific activities measured at 3 mM K+ were observed between epileptic and control cortex, except an increase of enzymic activities in the primary focus of acutely lesioned cats. The (Na+,K+)-ATPase catalytic subunits were resolved by SDS-gel electrophoresis and their phosphorylation levels were measured in presence of K+ ions and phenytoin. K+ was more effective in inducing maximal dephosphorylation of (Na+,K+)-ATPase in C57/BL, with identical affinity in the two strains. Phenytoin decreased the net phosphorylation level of (Na+,K+)-ATPase by about 50% in C57/BL mice, but only by 20% in DBA/2 mice. Both K+ and phenytoin dephosphorylating influences were decreased in primary and secondary foci of acutely lesioned cats. Those changes were limited to the alpha(-) subunit. In chronic cats, the dephosphorylating step of the (Na+,K+)-ATPase catalytic subunit recovered a normal affinity to K+, but its sensitivity to phenytoin remained decreased. Those differences in K+ and phenytoin influences on brain (Na+,K+)-ATPases between control and epileptic cortex might be responsible for the ictal transformation and seizure spread. In cats, the alteration of the alpha(-) isoform could mainly affect the glial cells.

We examined the activity and phosphorylation level of (Na+,K+)-ATPase (E.C. 3.6.1.3) partially purified from normal and epileptic human cortices. Control patients (n = 11) were operated on for a non-epileptogenic deep brain lesion, while epileptic patients (n = 10) were operated on for temporal or frontal originating partial seizures, resistant to medications or secondary to evolutive brain tumors. No differences in the specific activity of microsomal (Na+,K+)-ATPase were observed between the two groups of patients. After partial purification of the enzyme followed by SDS-polyacrylamide gel electrophoresis, (Na+,K+)-ATPase catalytic subunit had a decreased affinity for K+ in human epileptic cortex and lost its sensitivity to phenytoin dephosphorylation. Indirect evidence suggests that those abnormalities of (Na+,K+)-ATPase in human epileptic cortex hold preferentially true for the alpha(-) enzymatic subunit. Those results indicate that, in human epileptic cortex, (Na+,K+)-ATPase and most probably its glial subtype is altered in its K+ regulation and phenytoin sensitivity and could be responsible for ictal transformation and seizure spread.