Current views of the pathobiochemistry of epilepsy

Review: cholinergic mechanisms and epileptogenesis. The seizures induced by pilocarpine: a novel experimental model of intractable epilepsy

High-dose treatment with pilocarpine hydrochloride, a cholinergic muscarinic agonist, induces seizures in rodents following systemic or intracerebral administration. Pilocarpine seizures are characterized by a sequential development of behavioral patterns and electrographic activity. Hypoactivity, tremor, scratching, head bobbing, and myoclonic movements of the limbs progress to recurrent myoclonic convulsions with rearing, salivation, and falling, and status epilepticus. The sustained convulsions induced by pilocarpine are followed by widespread damage to the forebrain. The amygdala, thalamus, olfactory cortex, hippocampus, neocortex, and substantia nigra are the most sensitive regions to epilepsy-related damage following convulsions produced by pilocarpine. Spontaneous seizures are observed in the long-term period following the administration of convulsant doses of pilocarpine. Developmental studies show age-dependent differences in the response of rats to pilocarpine. Seizures are first noted in 7-12 day-old rats, and the adult pattern of behavioral and electroencephalographic sequelae of pilocarpine is seen in 15-21-day-old rats. During the third week of life the rats show an increased susceptibility to the convulsant action of pilocarpine relative to older and younger animals. The developmental progress of the convulsive response to pilocarpine does not correlate with evolution of the brain damage. The adult pattern of the damage is seen after a delay of 1-2 weeks in comparison with the evolution of seizures and status epilepticus. The susceptibility to seizures induced by pilocarpine increases in rats aged over 4 months. The basal ganglia curtail the generation and spread of seizures induced by pilocarpine. The caudate putamen, the substantia nigra, and the entopeduncular nucleus govern the propagation of pilocarpine-induced seizures. The antiepileptic drugs diazepam, clonazepam, phenobarbital, valproate, and trimethadione protect against pilocarpine-induced convulsions, while diphenylhydantoin and carbamazepine are ineffective. Ethosuximide and acetazolamide increase the susceptibility to convulsant action of pilocarpine. Lithium, morphine, and aminophylline also increase the susceptibility of rats to pilocarpine seizures. The pilocarpine seizure model may be of value in designing new therapeutic approaches to epilepsy.

Do glia contribute to behaviour? A neuromodulatory review

1. The links between behavioural state, gross electrophysiology and the activity of neurons and astrocytes are reviewed to stimulate interest in the contributions that glia make to behaviour. 2. Behavioural arousal in which neuronal responsivity ("sensitivity") is elevated is also associated with a sustained (0.5-10 sec) potential shift (SPS). 3. There is powerful and accumulating evidence that the SPS is primarily of glial origin. 4. In epilepsy neurons are hyperactive and there is a massive SPS during seizures. In seizure free periods, epileptic animals frequently have elevated arousal responses and increased neuronal sensitivity, indicating that seizures may be due to elevation of the activity of a normally adaptive sensitizing mechanism. 5. The common finding of an astrocytic pathology in epilepsy and the links between arousal, neuronal sensitization, SPSs and seizures implicates a modulatory role for astrocytes in both health and disease. 6. Glia, especially astrocytes, may modulate neuronal responsiveness by regulation of the microenvironment. 7. At the current state of knowledge, regulation of extracellular ionic K+, Ca2+ and neurotransmitter glutamate and GABA seem to be the most important candidates for modulating neuronal sensitivity in arousal and abnormally for seizure genesis. 8. Both in phylogeny and in ontogeny, glia and neurons have intimate associations. 9. The functional astrocytic syncitium is in a prime position to control the ecology of neuronal populations and thereby their activity. 10. The physiology and biochemistry of glia-neuronal interactions offers exciting new prospects for developments in behavioural neuroscience.

Brain neurotransmitters in glycine encephalopathy

We measured neurotransmitter markers in autopsied brain of infants with glycine encephalopathy (GE). Because patients with GE develop intractable seizures, special attention was devoted to those neurotransmitter systems implicated in human epilepsy. Mean levels of glycine in the frontal cortex of GE patients were three times higher than control values. No abnormalities were observed for concentrations of gamma-aminobutyric acid (and related receptors), other major neurotransmitter amino compounds, or activities of cholineacetyltransferase and aspartate aminotransferase. Mean acetylcholinesterase activity was significantly elevated by 46%. As experimental data suggest, glycine markedly potentiates the action of the excitatory neurotransmitter glutamic acid. To the extent that the brain seizures in patients with GE can be explained by this mechanism, pharmacotherapy with excitatory amino acid antagonists may represent a new approach to the treatment of GE.

Inhibition of the development of electrical kindling of the prepyriform cortex by daily focal injections of excitatory amino acid antagonists

The effects of daily focal injections of excitatory amino acid antagonists into the prepyriform cortex on the development of electrically kindled seizures at this site were studied. The selective 'NMDA receptor' antagonists 2-amino-7-phosphonoheptanoic acid (AP7) and 3-[+/-)-2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid (CPP) significantly inhibited the development of the electrically evoked afterdischarge over a 10 day period and prevented the development of the motor seizure responses. The 'kainate and quisqualate receptor' antagonist gamma-D-glutamylaminomethyl sulphonic acid (GAMS) showed less potent but still significant inhibitory actions on these responses. When drug treatment ceased, kindling progressed in all animals at a rate similar to that of the control (buffer-treated) animals. These results suggest a critical role for NMDA receptors in the primary neuronal events initiating the epileptiform activity in this animal model of epilepsy.

Increased activity of choline acetyltransferase and acetylcholinesterase in actively epileptic human cerebral cortex

We measured the activities of the cholinergic marker synthetic and catabolic enzymes choline acetyltransferase (ChAT) and acetylcholinesterase (AChE) in surgical specimens obtained from 38 patients immediately following anterior temporal lobectomy for intractable epilepsy. Samples from patients with actively spiking lateral temporal cortex were compared to non-spiking lateral temporal cortex obtained from patients in whom the epileptic discharges were confined to the hippocampus. Mean activities of ChAT and AChE were increased by 25% (P less than 0.01) and 30% (P less than 0.025) respectively in the spiking vs. non-spiking cortex. We suggest that the above-normal activity of these cholinergic marker enzymes may reflect sprouting of cholinergic nerve terminals in spontaneously spiking cortex of some patients and/or increased acetylcholine metabolism secondary to the stimulatory effect of the ongoing epileptic discharge.

Evidence that activation of GABA receptors in the substantia nigra suppresses spontaneous spike-and-wave discharges in the rat

The involvement of intranigral gamma-aminobutyric acid (GABA) receptors in the control of generalized non-convulsive epilepsy was investigated in a genetically determined model of petit mal epilepsy in the rat. Bilateral intranigral injection of muscimol (2 ng/0.2 microliter/side), a GABA agonist, significantly suppressed EEG-recorded spike-and-wave discharges for about 80 min, both at the cortical and thalamic levels. This suppressive effect was shown to be dose-dependent, reproduced by bilateral intranigral injections of 4,5,6,7-tetrahydro-isoxazolo [5,4-c]pyridin 3-ol (THIP), another GABA agonist, and reversed by a subsequent intranigral injection of bicuculline methiodide, a GABA antagonist. In addition, the anti-absence effects of bilateral intranigral injections of muscimol were well localized to the substantia nigra. Bilateral intranigral injections of GABA antagonists per se had no effects on the discharges. These results demonstrate that activation of GABA receptors in the substantia nigra suppresses the occurrence of spike-and-wave discharges in an animal model of generalized non-convulsive epilepsy. The results are compared to data obtained in models of generalized convulsive epilepsy and the hypothesis of nigral GABAergic mechanisms as a control system for generalized epilepsies is discussed.

Potentiation by kainate of excitatory amino acid release in striatum: complementary in vivo and in vitro experiments

The effect of kainate on extracellular levels of amino acids in corpus striatum was investigated in vitro and in vivo, to elucidate the mechanism underlying its neurotoxicity. Kainate increased extracellular glutamate and aspartate in both striatal slices in vitro and intact striatum in vivo, as previously reported. Both in vitro and in vivo, DL-threo-3-hydroxyaspartate increased extracellular glutamate and aspartate levels (to between 150 and 200% of basal), and also enhanced their kainate-evoked release. The action of kainate in vivo was reduced by prior frontal decortication, whereas in vitro the kainate-evoked responses were only slightly reduced by tetrodotoxin, and remained above control values. These results confirm that kainate increases extracellular glutamate and aspartate, and provide evidence that this is due to synaptic release evoked by an action on receptors on glutamatergic neurone terminals. These findings may be relevant to the understanding of epilepsy.

Continuous glutamate leakage from brain cells is balanced by compensatory high-affinity reuptake transport

The glutamate (and aspartate) uptake blocker threo-3-hydroxyaspartate (20 microM) was added to superfusion fluids employed for in vivo microdialysis of corpus striatum, and to incubation medium for striatal slices (5 microM). In vivo it caused an increase in glutamate and aspartate concentrations in the superfusion fluid. In vitro it caused increases in the levels of glutamate, aspartate, GABA, taurine and glutamine in the incubation fluid. Tetrodotoxin (1 microM) did not influence the rises in glutamate or aspartate. It is concluded from these results that there is a continuous outward leakage of glutamate, and aspartate, from neural cells which is normally balanced by an inward flux due to reuptake processes. This leakage is distinct from synaptic release of these substances due to spike activity, since tetrodotoxin added to striatal slices did not diminish the action of threo-3-hydroxyaspartate. The significance of the findings for mechanisms leading to ischaemic or hypoxic brain damage, and basic mechanisms in epilepsy is discussed.

Common anticonvulsants inhibit Ca2+ uptake and amino acid neurotransmitter release in vitro

Phenytoin (PHT), phenobarbital (PB), and carbamazepine (CBZ) dose-dependently inhibited veratrine-stimulated calcium influx and evoked amino acid neurotransmitter release in rat cortical slices at relatively low concentrations. The action on Ca2+ influx was in the clinical effective dose range for these anticonvulsant drugs, whereas the action on amino acid release was mostly well above this range. Neither ethosuximide (ESM) nor sodium valproate (VPA) had any effect on the veratrine-stimulated Ca2+ influx. Stimulated amino acid release was not affected by ESM, whereas VPA specifically inhibited the release of aspartate in preference to glutamate and GABA at concentrations well within the clinically effective dose range. The actions of VPA and ESM on other parameters measured were detectable only at very high concentrations, which are likely to be irrelevant in defining their clinical mode of action.