Acta neurol. belg., 2000, 100, 201-213 Review article Pathophysiology of epilepsy S. ENGELBORGHS, R. D’HOOGE, P. P. DE DEYN Department of Neurology, A.Z. Middelheim, Antwerp, and Department of Neurology, Laboratory of Neurochemistry and Behavior, Born-Bunge Foundation, University of Antwerp, Antwerp, Belgium ———— Abstract This work reviews the current knowledge on epileptogenesis and pathophysiology of epilepsy. Recently, gene defects underlying four monogenic epilepsies (generalized epilepsy with febrile seizures, autosomal dominant nocturnal frontal lobe epilepsy, benign familial neonatal convulsions and episodic ataxia type 1 with partial seizures) have been identified, shedding new light on the pathophysiology of epilepsy as these diseases are caused by ion channel mutations. Although epileptic syndromes differ pathophysiologically, common ictogenesis-related characteristics as increased neuronal excitability and synchronicity are shared as well as mechanisms involved in interictal-ictal transition. Emerging insights point to alterations of synaptic functions and intrinsic properties of neurons as common mechanisms underlying hyperexcitability. This work also reviews the neurochemical mechanisms of epilepsy. An imbalance between glutamate and g-aminobutyric acid neurotransmitter systems can lead to hyperexcitability but catecholaminergic neurotransmitter systems and opioid peptides were shown to play a role in epileptogenesis as well. An overview of currently available anti-epileptic drugs and their presumed mechanisms of action is given as an illustration of the neurochemistry of epileptogenesis. Most anti-epileptic drugs exert their anti-epileptic properties through only a few neurochemical mechanisms that are meanwhile basic pathophysiological mechanisms thought to cause seizures. Key words : Epilepsy ; pathophysiology ; epileptogenesis ; ictogenesis ; neurochemistry ; anti-epileptic drugs. 1. Introduction Several decades have been devoted to the study of the pathophysiology of epilepsy. Increasing knowledge in the field only contributed to a partial understanding of the underlying mechanisms. Nevertheless, insight in the pathophysiology of epilepsy and its underlying histological and neurochemical alterations has contributed to rational development strategies of new anti-epileptic drugs (AEDs). Although various epileptic syndromes were shown to differ pathophysiologically, they appar- ently share common ictogenesis-related characteristics such as increased neuronal excitability and synchronicity. Emerging insights point to alterations of synaptic functions and intrinsic properties of neurons as common mechanisms underlying hyperexcitability. Progress in the field of molecular genetics revealed arguments in favor of this hypothesis as mutations of genes encoding ion channels were recently discovered in some forms of human epilepsy. This work reviews the current knowledge on the pathophysiology of epilepsy with special emphasis on ictogenesis, mechanisms of interictal-ictal transition and neurochemical mechanisms underlying epilepsy. Where possible, examples concerning pathophysiological mechanisms underlying distinct epileptic syndromes will be given. 2. Pathophysiology of epilepsy Epileptic seizures arise from an excessively synchronous and sustained discharge of a group of neurons. The single feature of all epileptic syndromes is a persistent increase of neuronal excitability. Abnormal cellular discharges may be associated with a variety of causative factors such as trauma, oxygen deprivation, tumors, infection, and metabolic derangements. However, no specific causative factors are found in about half of the patients suffering from epilepsy. Underlying causes and pathophysiological mechanisms are (partially) understood for some forms of epilepsy, e.g. epilepsies caused by disorders of neuronal migration and monogenic epilepsies. For several other types of epilepsy, current knowledge is only fragmentary. 2.1. DISORDERS OF NEURONAL MIGRATION The major developmental disorders giving rise to epilepsy are disorders of neuronal migration that may have genetic or intrauterine causes (Meldrum, 1994). Abnormal patterns of neuronal migration lead to various forms of agyria or pachygyria whereas lesser degrees of failure of neuronal 202 S. ENGELBORGHS ET AL. migration induce neuronal heterotopia in the subcortical white matter. Recent experimental data suggest that cortical malformations can both form epileptogenic foci and alter brain development in a manner that diffuse hyperexcitability of the cortical network occurs (Chevassus au Louis et al., 1999). Other studies revealed increases in postsynaptic glutamate receptors and decreases in g-aminobutyric acid (GABA) (A) receptors in microgyric cortex which could promote epileptogenesis (Jacobs et al., 1999). Tuberous sclerosis is a developmental disorder with autosomal dominant inheritance in which disordered neuronal migration and epilepsy are commonly found. Periventricular heterotopia is an Xlinked dominant disorder of cerebral cortical development. Fox et al. (1998) showed that mutations in the filamin 1 gene prevent migration of cerebral cortical neurons causing periventricular heterotopia. Affected females present with epilepsy whereas affected males die embryonically. Recently, however, a male patient with bilateral periventricular and subcortical heterotopia was described which raises the possibility of a novel gene involved in brain formation (Sisodiya et al., 2000). X-linked lissencephaly and double cortex syndrome is another disorder of neuronal migration. Double cortex or subcortical band heterotopia often occurs in females whereas more severe lissencephaly is found in affected males. A causal mutation in a gene called doublecortin was recently identified (Gleeson et al., 1998). It was suggested that doublecortin acts as an intracellular signaling molecule critical for the migration of developing neurons (Allen and Walsh, 1999). Although these disorders are relatively rare, studying the underlying pathophysiological mechanisms may shed light on the pathophysiology of more common epileptic syndromes. 2.2. GENETICS OF HUMAN EPILEPSY Epilepsies with complex inheritance About 40% of patients suffering from epilepsy have a genetic background that contributes to the aetiology of epilepsy (Gardiner, 2000). Most familial epilepsies like juvenile myoclonic epilepsy, childhood absence epilepsy, and benign childhood epilepsy with centrotemporal spikes have a complex mode of inheritance resulting from the interaction of several loci together with environmental factors (McNamara, 1999). In patients with absence seizures (and their first degree relatives), biochemical changes (e.g. increased plasma glutamate levels) have been identified which can be related to a generalized increase in cortical excitability (Van Gelder et al., 1980). Probably, the genetic predisposition of absence epilepsy is based on a gene-dependent biochemical derangement leading to increased cortical excitability. Genetic data generated by studies on animal models of absence epilepsy show a relative simple inheritance factor of one gene that determines being epileptic or not while other genes determine number and duration of epileptic fits (Renier and Coenen, 2000). Monogenic epilepsies Monogenic epileptic disorders are rare, accounting for no more than 1% of patients. Recent advances in the genetics and molecular biology of these diseases unravelled the underlying pathophysiology of some of these epileptic syndromes. In 1996, Berkovic et al. described a new epileptic syndrome : familial temporal lobe epilepsy. Simple partial seizures with psychic or autonomic symptoms are frequently occurring seizure types whereas complex partial fits are infrequent. Pedigree analysis suggested autosomal dominant inheritance with age-dependent penetrance (Berkovic et al., 1996). Linkage to chromosome 10q has been reported in one family but the genetic defect remains to be elucidated (Berkovic and Scheffer, 1997a). Autosomal dominant partial epilepsy with auditory features is characterized by auditory hallucinations, although other sensory symptoms have been reported as well (Winamer et al., 2000). Clinical semiology points to a lateral temporal localization which is supported by electroencephalogram (EEG)-data that revealed inconstant focal abnormalities over both temporal regions (Michelucci et al., 2000). In a single case, brain Magnetic Resonance Imaging (MRI) showed atrophy with an increased T2 signal in the lateral portion of the right temporal lobe (Michelucci et al., 2000). This epileptic syndrome was found to be linked to chromosome 10q22-24 (Winamer et al., 2000). Gene defects underlying four other monogenic epilepsies (generalized epilepsy with febrile seizures, autosomal dominant nocturnal frontal lobe epilepsy, benign familial neonatal convulsions and episodic ataxia type 1 with partial seizures) have recently been identified, shedding new light on the pathophysiology of epilepsy as these diseases are caused by ion channel mutations (Steinlein, 1998 ; Zuberi et al., 1999). Generalized epilepsy with febrile seizures type I is an autosomal dominant epileptic syndrome that is caused by a point mutation in the b1-subunit of a voltage-gated Na+ channel (Wallace et al., 1998) whereas type II is caused by a point mutation in the a1-subunit of a voltage-gated Na+ channel (Escayg et al., 2000). These mutations cause distinct types of epilepsy in different members of the same family, which may result from inheritance of the mutant gene in the context of other susceptibility genes or environmental factors (McNamara, 1999). PATHOPHYSIOLOGY OF EPILEPSY 203 Benign familial neonatal convulsions is a syndrome that is inherited in an autosomal dominant pattern. Mutations of two distinct but related voltage-gated K+ channel genes have been identified (Biervert et al., 1998). Although both genes (KCNQ2 and KCNQ3) are located on different chromosomes (20q and 8q respectively), their coexpression explains how these 2 different mutations cause an identical disease phenotype. In some families, autosomal dominant nocturnal frontal lobe epilepsy is caused by a point mutation in a gene on chromosome 20q (CHRNA4), encoding the a4 subunit of the neuronal nicotinic acetylcholine (ACh) receptor (Steinlein et al., 1995). At least some ACh receptors are located presynaptically, thus promoting the release of neurotransmitters as GABA. The mutant receptor causes a reduction of ACh-mediated Ca2+ flux, which results in a decrease of GABA released from presynaptic terminals leading to synaptic disinhibition (McNamara, 1999). However, the majority of the families with autosomal dominant nocturnal frontal-lobe epilepsy are not linked to CHRNA4, indicating the presence of genetic heterogeneity (Gardiner, 2000). Episodic ataxia type 1 is a rare autosomal dominant disorder, characterized by brief episodes of ataxia associated with myokymia (Zuberi et al., 1999). The patients suffering from this syndrome also show partial epileptic fits. The syndrome is associated with point mutations in the human voltage-gated potassium channel gene on chromosome 12p13 (Zuberi et al., 1999). As potassium channels determine the excitability of neurons, it is suggested that this mutation is pathogenic (Zuberi et al., 1999). These recent discoveries illustrate that ion channel dysfunctions can play a crucial role in the pathophysiology of epilepsy. As several AEDs act on ion channels, these findings are relevant to other epileptic syndromes in man. metabolism at the medial thalamic nucleus (Juhász et al., 1999). These findings are common and have strong lateralization value for the seizure focus in human temporal lobe epilepsy. The decreased benzodiazepine receptor binding possibly reflects neuronal loss but may also indicate decreased benzodiazepine receptor density in the medial thalamic nucleus which remains to be elucidated. This structure may indeed play an important role in temporal lobe epilepsy as the nucleus medialis thamali has strong reciprocal connections with other parts of the limbic system (Engelborghs et al., 1998b ; Juhász et al., 1999). Interictal PET studies revealed increased glucose metabolism and FMZ binding in the lateral thalamus of patients with temporal lobe epilepsy, possibly reflecting an upregulation of GABA-mediated inhibitory circuits (Juhász et al., 1999). A recent study investigated expression and distribution of GABA(A)-receptors in the hippocampus of pilocarpine-treated rats (Fritschy et al., 1999). A loss of a critical number of interneurons in the gyrus dentatus was noticed, which might play a role in seizure initiation (Fritschy et al., 1999). Meanwhile, long-lasting upregulation of GABA(A)-receptors in granule cells was found, which might represent a compensatory response to seizure activity (Fritschy et al., 1999). Central benzodiazepine receptor density in the CA1 region was shown to be significantly reduced by means of autoradiography in post-mortem samples of patients with hippocampal sclerosis (Hand et al., 1997). On the other hand, affinity for FM2 was increased in subiculum and gyrus dentatus (Hand et al., 1997). Other publications suggest that enhanced sensitivity to glutamate may be an important element in the pathophysiology of temporal lobe epilepsy as a quantitative autoradiographic analysis of ionotropic glutamate receptor subtypes revealed an upregulation in the reorganized human epileptogenic hippocampus (Brines et al., 1997). 2.3. PATHOPHYSIOLOGY OF DISTINCT TYPES OF EPILEPSY Gelastic epilepsy Mesial temporal lobe epilepsy Mesial temporal lobe epilepsy is characterized by recurrent complex partial seizures and hippocampal sclerosis. Ipsilateral to the epileptogenic focus, hippocampal neuronal loss results in significantly reduced hippocampal volumes as measured by means of MRI (Jokeit et al., 1999). Besides hippocampal volumetry, MR proton spectroscopy was shown to be a valuable tool to correctly lateralize patients with mesial temporal lobe epilepsy (Kuzniecky et al., 1998). In patients with intractable temporal lobe epilepsy, interictal Positron Emission Tomography (PET) studies found decreased [11C]flumazenil (FMZ) binding (benzodiazepine receptor binding) and glucose Gelastic seizures are frequently caused by hypothalamic hamartomas (Engelborghs et al., 2000). In a series of 9 patients with gelastic seizures, 4 had hypothalamic hamartoma (Striano et al., 1999). Hypothalamic hamartomas are rare congenital malformations and often present as a clinical syndrome, characterized by pubertas praecox, mental retardation, and gelastic seizures. Later, refractory epilepsy with multiple seizure types develops. Patients with hypothalamic hamartoma may as well have focal epileptiform discharges in the anterior temporal or frontal lobe on ictal electrocorticographic recordings but focal cortical resection was shown not to affect seizure frequency (Cascino et al., 1993). According to depth electrode data showing ictal onset from the lesion, seizures seem to 204 S. ENGELBORGHS ET AL. begin in the intrinsically epileptogenic hamartoma (Munari et al., 1995 ; Kuzniecky et al., 1997). Ictal single-photon emission computed tomography performed during typical gelastic seizures demonstrated hyperperfusion in the hamartoma and the hypothalamic region (Kuzniecky et al., 1997). Moreover, complete surgical extirpation or a gamma knife radiosurgical treatment of the hypothalamic hamartoma results in seizure remission (Kuzniecky et al., 1997 ; Georgakoulias et al., 1998 ; Unger et al., 2000). It therefore seems probable that epileptic discharges arise in the hypothalamus and spread via hypothalamic-amygdala connections to produce focal temporal lobe ictal discharges (Saper, 1990 ; Berkovic et al., 1997b). Rasmussen encephalitis Rasmussen encephalitis is a rare, progressive, neurodegenerative illness of unknown cause that typically affects children in the first decade of life (Rasmussen et al., 1958). Severe seizures that are refractory to anti-epileptic medication, hemispheric atrophy, and dementia are cardinal features of this disease. Rabbits immunized with glutamate receptor subunit 3 (GluR3) protein developed epilepsy and cerebral histopathological changes characteristic of Rasmussen encephalitis (Rogers et al., 1994). This led to the discovery of anti-GluR3 antibodies in serum of patients with Rasmussen encephalitis (Rogers et al., 1994). Effectiveness of plasma exchanges or intravenous immunoglobin therapy as treatment of Rasmussen encephalitis was demonstrated in a series of patients (Andrews et al., 1996 ; Topcu et al., 1999), which further proves the auto-immune pathogenesis of Rasmussen encephalitis. Insights gained from the study of Rasmussen encephalitis may help to increase our knowledge about more common forms of epilepsy. Some patients undergoing temporal lobectomy for refractory epilepsy show localized inflammatory histopathological changes and increased auto-antibodies in serum (Dambinova et al., 1997 ; Peltola et al., 2000) ; whether or not this is caused by an autoimmune pathogenesis remains to be elucidated (McNamara, 1999). 3. Kindling and epileptogenesis Goddard (1967) was the first to describe that periodic stimulation of neural pathways progressively leads to recurrent behavioral and electrographic seizures. Kindling procedures have provided a substrate for the study of the role of enhanced synaptic efficacy in seizure disorders. It is now considered to be a first choice experimental procedure in the study of the potential mechanisms of epileptogenesis. The phenomenon can be evoked in various brain regions, but amygdala kindling is most frequently used in epilepsy research as a model for complex partial seizures (Fisher, 1989). Although kindling has been shown to be phenomenologically different from other types of plastic changes in the central nervous system (Sutula, 1991), there are many points of similarity between kindling and the process of long-term potentiation. Kindling has been shown to depend upon functional as well as structural changes in glutamatergic synapses. The anticonvulsant effects of glutamate receptor blocking agents like NMDA antagonists seem to be at least partly due to their inhibitory effects on in vitro kindling. 4. Ictogenesis Excitability is a key feature of ictogenesis that may originate from individual neurons, neuronal environment or a population of neurons (Traub et al., 1996). Excitability arising from single neurons may be caused by alterations in membrane or metabolic properties of individual neurons. When regulation of environmental, extracellular concentrations of ions or neurotransmitters is suboptimal, the resulting imbalance might enhance neuronal excitation. Collective anatomic or physiologic neuronal alterations may convert neurons into a hyperexcitable neuronal population. In reality, these three theoretical mechanisms are thought to interact during specific ictal episodes. Each epileptic focus is unique as the differential contribution of these three concepts leading to ictal events is thought to differ from focus to focus. 4.1. EXCITABILITY ARISING FROM INDIVIDUAL NEURONS Functional and perhaps structural changes occur in the postsynaptic membrane, thus altering the character of receptor protein - conductance channels, thereby favoring development of paroxysmal depolarizing shift (PDS) and enhanced excitability. Epileptic neurons appear to have increased Ca2+ conductance. It may be that latent Ca2+ channels are used, that the efficacy of Ca2+ channels is increased or that the number of Ca2+ channels is chronically elevated. However, development of burst activity depends on the net inward current and not on the absolute magnitude of the inward current. When extracellular K+ concentrations are increased (as during seizure activity), the K+ equilibrium across the neuronal membrane is reduced, resulting in reduced outward K+ currents. The net current will become inward, depolarizing the neuron to the extent that Ca2+ currents will be triggered. This results in a PDS and a burst of spikes (Fig. 1) (Dichter, 1997). 4.2. EXCITABILITY ARISING FROM NEURONAL MICRO- ENVIRONMENT Both functional and structural alterations occur in epileptic foci. The functional changes involve PATHOPHYSIOLOGY OF EPILEPSY 205 MFS was demonstrated in patients with refractory temporal lobe epilepsy with hippocampal sclerosis on neuroimaging as well as in numerous animal models of temporal lobe epilepsy (Sutula et al., 1988 ; Sutula et al., 1989). In normal conditions, the dentate granule cells limit seizure propagation through the hippocampal network. However, the formation of recurrent excitatory synapses between dentate granule cells, as is thought to occur after MFS, may transform the dentate granule cells into an epileptogenic population of neurons (Figure 2) (McNamara et al., 1999). Possibly, a vicious circle develops : seizures cause neuronal death which results in MFS which in turn increases seizure frequency. FIG. 1. — The hallmark of a discharge is the paroxysmal depolarization shift. The ability to record such events at the scalp suggests that numerous neurons are firing in synchrony (From : Browne T.R., Feldman R.G., eds. Epilepsy : Diagnosis and Management. Boston : Little, Brown and Company, 1983, p.12). concentrations of cations and anions, metabolic alterations, and changes in neurotransmitter levels. The structural changes involve both neurons and glia. Excessive extracellular K+ depolarizes neurons and leads to spike discharge. During seizures, changes in extracellular Ca2+ (a decrease of 85%) precede those of K+ by milliseconds and Ca2+ levels return to normal more quickly than K+. Glia are able to clear neurotransmitters from the extracellular space and to buffer K+ thus correcting the increased extracellular K+ concentrations that occur during seizures. Epileptic foci may show proliferation of glia that differ however in morphological and physiological properties (Bordey and Sontheimer, 1998). Gliosis will affect glial K+ buffering capacity and hence may contribute to seizure generation (Grisar et al., 1999). 4.3. THE EPILEPTIC CELL POPULATION Collective anatomic or physiologic neuronal alterations might produce progressive, networkdependent facilitation of excitability, perhaps coupled with a decrease of inhibitory influences, e.g. due to selective loss of inhibitory neurons. Mossy fiber sprouting (MFS) is an example of neuronal alterations leading to increased excitability (Cavazos et al., 1991). FIG. 2. — Many central neurons burst, but rarely in a prolonged manner (panel A). In an epileptic focus, neurons burst spontaneously and for prolonged periods (note the paroxysmal depolarization shift) (panel B). 5. Mechanisms of interictal-ictal transition Mechanisms producing signal amplification, synchronicity, and spread of activity are likely to be involved in interictal-ictal transitions. Different theoretical mechanisms of interictal-ictal transition are summarized in table 1 and discussed in the following section. In vivo, interictal-ictal transition can seldom be attributed to one theoretical mechanism, but often results from the interaction of different mechanisms. 5.1. NONSYNAPTIC MECHANISMS Alterations in ionic microenvironment Repetitive ictal and interictal activity causes increases in extracellular K+ leading to increased neuronal excitability (Moody et al., 1974). Some neurons are very sensitive to changes in membrane K+ currents, e.g. pyramidal cells in the CA1 region of the hippocampus (Rutecki et al., 1985). 206 S. ENGELBORGHS ET AL. Table 1 Summary of mechanisms of interictal-ictal transition Nonsynaptic mechanisms 1. Alterations in ionic microenvironment ; e.g. increased extracellular K+, decreased extracellular Ca++ 2. Decreases in size of extracellular space 3. Failure of ion transport : Na+-K+ pump or Cl--K+ co-transport 4. Presynaptic terminal bursting 5. Ephaptic interactions Synaptic mechanisms 1. 2. 3. 4. Depression of GABA-ergic inhibition NMDA receptor activation ; voltage-dependent EPSPs Frequency potentiation of EPSPs Actions of modulators Active ion transport Activation of the Na+-K+ pump is important for regulation of neuronal excitability during excessive neuronal discharges (Ayala et al., 1970). Substances like ouabain that block the Na+-K+ pump can induce epileptogenesis in animal models. Hypoxia or ischemia can result in Na+-K+ pump failure thus promoting interictal-ictal transition. A Cl--K+ co-transport mechanism controls the intracellular Cl- concentration and the Cl- gradient across the cell membrane which regulates effectiveness of GABA-activated inhibitory Cl- currents. Interference with this process could cause a progressive decrease in the effectiveness of GABAergic inhibition leading to increased excitability (Prince, 1988). Presynaptic terminal bursting The amount of transmitter released is related to depolarization of presynaptic terminals. Changes in axon terminal excitability will have effects on synaptic excitation (Prince, 1988). Abnormal bursts of action potentials occur in the axonal arborizations of thalamocortical relay cells during epileptogenesis. Since one thalamocortical relay cell ends on a large number of cortical neurons, synchronization can occur which might play an important role in interictal-ictal transition (Engelborghs et al., 1998b). Ephaptic interaction Ephaptic interactions are produced when currents from activated neurons excite adjacent neurons in the absence of synaptic connections. Ephaptic effects are strongly dependent on the size of the extracellular space. When extracellular space is small, ephaptic interactions are promoted (Traub et al., 1985). 5.2. SYNAPTIC MECHANISMS Two theoretical mechanisms can occur : decreased effectiveness of inhibitory synaptic mechanisms or facilitation of excitatory synaptic events. Both mechanisms will be discussed in the following section. 6. Neurochemical mechanisms underlying epilepsy 6.1. GABA The GABA hypothesis of epilepsy implies that a reduction of GABA-ergic inhibition results in epilepsy whereas an enhancement of GABA-ergic inhibition results in an anti-epileptic effect (De Deyn et al., 1990). Inhibitory postsynaptic potentials (IPSPs) gradually decrease in amplitude during repetitive activation of cortical circuits. This phenomenon might be caused by decreases in GABA release from terminals, desensitization of GABA receptors that are coupled to increases in Clconductance or alterations in the ionic gradient because of intracellular accumulation of Cl- (Wong and Watkins, 1982). In case of intracellular accumulation of Cl-, passive redistribution is ineffective. Moreover, Cl--K+ co-transport becomes less effective during seizures as it depends on the K+ gradient. As Cl--K+ co-transport depends on metabolic processes, its effectiveness may be affected by hypoxia or ischemia as well. These mechanisms may play a critical role in ictogenesis and interictal-ictal transition. Several studies have shown that GABA is involved in pathophysiology of epilepsy in both animal models and patients suffering from epilepsy. GABA levels and glutamic acid decarboxylase (GAD) activity were shown to be reduced in epileptic foci surgically excised from patients with intractable epilepsy and in CSF of patients with certain types of epilepsy (De Deyn et al., 1990). In stiff-man syndrome, a disease associated with epilepsy and diabetes mellitus, auto-antibodies to GAD were demonstrated (Solimena et al., 1988). A reduction of 3H-GABA binding has been reported in human brain tissue from epileptic patients whereas PET studies demonstrated reduced benzodiazepine receptor binding in human epileptic foci (Savic et al., 1996). The degree of benzodiazepine receptor reduction showed a positive correlation with seizure frequency. The GABA receptor complex is involved in various animal models of epilepsy as well. Low CSF levels of GABA were revealed in dogs with epilepsy (Löscher and Swartz-Porsche, 1986). Reduced GAD levels were revealed in the substantia nigra of amygdala-kindled rats (Löscher and Schwark, 1985). Significant alterations in GABA and benzodiazepine binding have been shown in 207 PATHOPHYSIOLOGY OF EPILEPSY the substantia nigra of genetically seizure-prone gerbils (Olsen et al., 1985). Mice with a genetic susceptibility to audiogenic seizures have a lower number of GABA receptors than animals of the same strain that are not seizure prone (Horton et al., 1982). Several endogenous (guanidino compounds) and exogenous (e.g. bicuculline, picrotoxin, penicillin, pilocarpine, pentylenetetrazol) convulsants inhibit GABAergic transmission through inhibition of GABA synthesis or through interaction with distinct sites at the postsynaptic GABA(A) receptor (De Deyn and Macdonald, 1990 ; De Deyn et al., 1992 ; D’Hooge et al. 1999). Convulsant agents that block synaptic GABA-mediated inhibition, amplify the dendritic spike-generating mechanism that involves Ca2+ (Dichter and Ayala, 1987 ; Fisher, 1989). Synaptic inputs are thought to trigger and synchronize this process throughout a population of cells which then might result in an epileptic fit. Several AEDs are GABA analogues, block GABA metabolism (e.g. vigabatrin, tiagabine, valproate) or facilitate postsynaptic effects of GABA. However, a study evaluating dose-dependent behavioral effects of single doses of vigabatrin in audiogenic sensitive rats, suggests that the antiepileptic properties of vigabatrin not only depend on GABA-ergic neurotransmission but might also be explained by decreased central nervous system levels of excitatory amino acids or increased glycine concentrations (Engelborghs et al., 1998a). 6.2. GLUTAMATE Glutamatergic synapses play a critical role in all epileptic phenomena. Activation of both ionotropic and metabotropic postsynaptic glutamate receptors is proconvulsant. Antagonists of N-methyl-Daspartate (NMDA) receptors are powerful anti-convulsants in many animal models of epilepsy. Several genetic alterations have been shown to be epileptogenic in animal models but no specific mutation relating to glutamatergic function has yet been linked to a human epilepsy syndrome. Nevertheless, there is evidence for altered NMDA receptor function in acquired epilepsy in animal models and in men. An increased sensitivity to the action of glutamate at NMDA receptors is seen in hippocampal slices from kindled rats and in cortical slices from cortical foci in human epilepsy (Hwa and Avoli, 1992). This results in an enhanced entry of Ca2+ into neurons during synaptic activity (Louvel and Pumain, 1992). Changes in metabotropic glutamate receptor function may also play a key role in epileptogenesis (Chapman, 1998). Epileptic seizures and epilepsy form frequent complications of uremia. As a possible underlying mechanism, we have demonstrated the accumulation of a series of uremic guanidino compounds which were shown to inhibit GABA-ergic neurotransmission (De Deyn and Macdonald, 1990). One of these endogenous agents was in addition shown to be an agonist at the excitatory NMDA receptor (D’Hooge et al., 1996 ; De Deyn et al., in press (a)). In patients with absence seizures, plasma glutamate levels were found to be significantly increased (Van Gelder et al., 1980). Neuronal membranes are exposed to increased amounts of extracellular glutamate thus increasing neuronal excitability. A recent study on a genetic rat model of epilepsy (WAG/RIJ rats ; spontaneous spikewave (SW) discharges accompanied by behavioral abnormalities) provides evidence for an interaction of glutamatergic and serotonergic mechanisms in the triggering and maintenance of epilepsy (Filakovszky et al., 1999). Intracerebroventricular injection of 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT), a 5HT1A receptor agonist, caused marked increase of the cumulative duration and number of SW discharges whereas dizocilpine (MK-801), a NMDA receptor antagonist, decreases SW discharges (Gerber et al., 1998). Both substances opposed each other’s effects in this rat model of epilepsy. 6.3. CATECHOLAMINES Abnormalities of CNS catecholamines have been reported in several genetic models of epilepsy. In spontaneous epileptic rat, dopamine was decreased in the nucleus caudatus whereas noradrenaline was increased in midbrain and brainstem (Hara et al., 1993). Decreased levels of dopamine have been found in epileptic foci of epilepsy patients (Mori et al., 1987). In animal models of absence epilepsy, seizures are exacerbated by dopamine antagonists while fits are alleviated by dopamine agonists (Snead, 1995). These results suggest that decreased dopamine facilitates appearance of seizures by lowering the threshold triggering such seizures. Tottering mice have an absence-like syndrome that is characterized by episodes of behavioral arrest associated with 6 to 7 Hz cortical SW EEG discharges. Selective destruction of the ascending noradrenergic system at birth prevents the onset of the syndrome. Therefore, it has been suggested that the syndrome is caused by a noradrenergic hyperinnervation of the forebrain (Meldrum, 1994). Recent data indicate that the serotonergic system regulates epileptiform activity in a genetic rat model of absence epilepsy as intraperitoneal or intracerebroventricular administration of 8-OHDPAT caused marked and dose-dependent increases in number and duration of SW discharges (Gerber et al., 1998). 208 S. ENGELBORGHS ET AL. 6.4. OPIOID PEPTIDES Established AEDs decrease neuronal membrane excitability by interacting with ion channels or neurotransmitter receptor complexes. AEDs that decrease membrane excitability through interaction with ion channels act on sodium and calcium channels. Most AEDs that interact with neurotransmitter complexes promote inhibitory GABA-ergic neurotransmission although some more recently developed drugs act through inhibition of excitatory neurotransmission as well. The most frequently used AEDs and their (presumed) mechanisms of action are summarized in table 3. Numerous AEDs interfere with ion channel functioning. Both ethosuximide and valproate block voltage-dependent Ca++ channels of the Ttype which explains their anti-epileptic efficacy in generalized absence epilepsy (Stefani et al., 1997). Carbamazepine, felbamate, lamotrigine, oxcarbazepine, phenytoin, topiramate, valproate and zonisamide are known to reduce voltage-dependent Na+ currents. Several established AEDs promote inhibitory neurotransmission. Vigabatrin is an enzymeactivated irreversible inhibitor of GABA-transaminase and has a weaker GABA uptake inhibitory effect whereas tiagabine is a pure GABA uptake inhibitor. Both AEDs thus increase the functional pool of GABA. Benzodiazepines enhance GABAergic inhibition through interaction with the GABA (A) receptor. Till present, progabide is the only AED exerting GABA agonistic effects at both type A and B sites. Several other AEDs have (weaker) GABA-ergic properties but act on other mechanisms as well. The main mechanism underlying phenobarbital’s anti-epileptic effects is attributed to In experimental studies, opioids and opioid peptides had both convulsant and anticonvulsant properties (Meldrum, 1994). Kappa agonists suppress SW discharges in an animal model of abscence epilepsy (Przewlocka et al., 1995). Peptides with a m agonist action induce hippocampal or limbic seizures when administered intraventricularly possibly due to inhibition of inhibiting interneurons. In patients with complex partial seizures, PET studies pointed out that m receptor density is increased in the temporal cortex (Mayberg et al., 1991). 7. Pathophysiology of epilepsy and mechanisms of action of AEDs (Levy et al., 1995 ; Macdonald and Kelly, 1995 ; Thomas, 1999) Most of the presently used AEDs were discovered by screening without a rationale as to the mechanism of action. As knowledge on the pathophysiology of epilepsy increases and the mechanisms of action of most AEDs are at least partially unravelled, it becomes clear that most AEDs exert their anti-epileptic properties through only a few neurochemical mechanisms that are meanwhile basic pathophysiological mechanisms thought to cause epileptic fits (table 2). Thanks to increased use of animal models of epilepsy, a better insight in the pathophysiology of epilepsy and improved knowledge on mechanisms of action of AEDs, several new more rationally designed AEDs were developed and marketed during the past decade (De Deyn et al., in press (b)). Table 2 Correlation between mechanisms of epileptogenesis and mechanisms of action of AEDs Mechanisms of epileptogenesis GABA ● ● ● ● ● Glu ● ● ● Reduced GABA in microgyric cortex Reduced benzodiazepine receptor binding in medial thalamic nucleus (mesial temporal lobe epilepsy) Reduced benzodiazepine receptor density in CA1 region (hippocampal sclerosis) Reduced GABA levels and GAD activity (epileptic foci) Auto-antibodies to GAD (Stiff-man syndrome) Upregulation of hippocampal ionotropic glutamate receptors (temporal lobe epilepsy) Anti-gluR3 antibodies (Rasmussen encephalitis) Increased plasma glutamate levels (absence seizures) Na+ ● Mutation voltage-gated Na+ channel (generalized epilepsy with febrile seizures) K+ ● Mutation voltage-gated K+ channel (benign familial neonatal convulsions) Ca++ ● Reduced ACh-mediated Ca flux (nocturnal frontal lobe epilepsy) → Increased membrane excitability Mechanisms of actions of AEDs ● ● ● ● ● ● Increased functional pool of GABA (vigabatrin, tiagabine) Enhanced GABA-ergic inhibition (benzodiazepines) GABA agonistic effects (progabide) (Weaker) GABA-ergic properties (phenobarbital, gabapentin, topiramate, valproate, zonisamide) Inhibition of glutamate release (lamotrigine) Block of glycine site at NMDA receptor (felbamate) ● Reduction of voltage-gated Na+ currents (carbamazepine, felbamate, lamotrigine, oxcarbazepine, phenytoin, topiramate, valproate, zonisamide) ● Reduction of T-type Ca++ currents (ethosuximide, valproate) → Decreased membrane excitability PATHOPHYSIOLOGY OF EPILEPSY 209 Table 3 Mechanisms of actions of anti-epileptic agents Anti-epileptic agent Mechanism(s) of action Benzodiazepines Enhances GABA action Reduces sustained repetitive firing Blocks voltage-dependent Na+ channels Limitation of sustained repetitive firing Reducing T-type Ca++ currents Blocking synchronized thalamic firing Inhibition of glutamatergic neurotransmission (reduces NMDA action, blocks glycine-site on NMDA receptor) GABA potentiation Blocks voltage-dependent Na+ channels Blocks L-type Ca++ channels GABA analog but does not bind to GABA receptors Increases synaptic GABA : activation of Glutamic Acid Decarboxylase ? May block amino acid transporter Binds to voltage-dependent Ca++channels Æ reduced intraneuronal concentration of Ca++ Possibly : inactivation of Na+ channels Reduces glutamate release Inhibits voltage-activated Ca++ currents, blocks voltage-dependent Na+ channels Unknown mechanism of action Increases seizure threshold and inhibits seizure spread in kindled rats Inhibition of voltage-dependent Na+ channels Inhibition of voltage-activated Ca++ currents Enhances GABA action Reduces sustained repetitive firing Reduces voltage-dependent Ca++ currents Blocks voltage-gated Na+ channels Reduces Ca++ currents Reduces sustained repetitive firing - blocks voltage-dependent Na+ currents GABA agonist at A and B sites NMDA receptor antagonist Inactivation of Na+ channels Neuronal and glial GABA-uptake inhibitor Na+ channel block Reduction of L-type Ca++ currents Potentiation of GABA at the GABA(A) receptor : enhancement of Cl- flux Inhibition of glutamatergic neurotransmission : weak block of AMPA/kainate receptors Inhibition of carbonic anhydrase Increases CNS GABA levels by increased synthesis and reduced catabolism Blocks T-type Ca++ currents Enhances Na+ channel inactivation GABA-Transaminase inhibitor Inhibits GABA uptake Blocks Na+ channels Blocks T-type Ca++ channels Enhances GABA action Inhibition of carbonic anhydrase Carbamazepine Ethosuximide Felbamate Gabapentin Lamotrigine Levetiracetam Oxcarbazepine Phenobarbital Phenytoin Primidone Progabide Remacemide Tiagabine Topiramate Valproate Vigabatrin Zonisamide the reduction of voltage-dependent Ca++ currents although this drug enhances GABA-ergic neurotransmission as well. Gabapentin, topiramate, valproate and zonisamide are other examples of drugs that have amongst others GABA-ergic properties. Lamotrigine is currently the best example of a drug acting through excitatory neurotransmission as it inhibits the release excitatory amino acids, especially glutamate. Although lamotrigine also blocks ion channels, its effect on glutamate release is thought to be primarily responsible for antiepileptic properties. Felbamate blocks the glycine site at the NMDA receptor which is at least partially related to its anti-epileptic effects. Besides ion channel inhibition and potentiation of GABA-ergic neurotransmission, topiramate interferes with excitatory neurotransmission as it weakly blocks amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptors. A link between from one side the preclinical profile and mechanisms of action of AEDs and from the other side clinical profile can be made. In general, AEDs exerting anti-epileptic properties through interaction with one single mechanism of action, have a narrow clinical profile (e.g. ethosuximide). However, most AEDs interfere with a combination of basic mechanisms of anti-epileptic action such as inhibitory neurotransmission and Na 210 S. ENGELBORGHS ET AL. and/or Ca channel functioning (e.g. valproate). The relative importance of basic mechanisms of anti-epileptic action involved (partially) determines the clinical profile of AEDs. Drugs effective against myoclonic seizures generally enhance GABA-ergic inhibition. Drugs effective against generalized tonic-clonic and partial seizures appear to reduce sustained high-frequency repetitive firing by delaying recovery of Na+ channels from activation (e.g. carbamazepine, valproate). It has been suggested that T-type Ca++ channels of thalamic relay neurons play a critical role in the generation of 3 Hz spike-and-wave discharges that characterize generalized absence epilepsy. Indeed, drugs that block T-type Ca++ currents are effective against generalized absence seizures (ethosuximide, valproate). Conclusions Although different epileptic syndromes differ pathophysiologically, ictogenesis-related mechanisms are often shared. The study of rare epileptic syndromes as monogenic epilepsies shed new light on ictogenesis and consequently substantially increased our knowledge and understanding of the pathophysiology of epilepsy. It now is generally accepted that ictogenesis mainly results from neuronal membrane hyperexcitability. Both neurotransmitter systems and ion channels play a crucial role in neuronal excitability. Most of the presently used AEDs were discovered by screening without a rationale as to the mechanism of action. As our knowledge on the pathophysiology of epilepsy increases and the mechanisms of action of most AEDs are at least partially unravelled, it becomes clear that most AEDs exert their anti-epileptic properties through only a few neurochemical mechanisms leading to decreased neuronal membrane excitability. At this point, pathophysiological mechanisms of ictogenesis meet with mechanisms of action of AEDs. This knowledge allowed linking preclinical mechanisms of action with clinical profiles of AEDs. A more detailed understanding of the pathophysiology of epilepsy and epileptogenesis will thus further contribute to still more specific and efficacious pharmacological interactions in this field of clinical neurology. Acknowledgements This work was supported by Born-Bunge Foundation, University of Antwerp, Neurosearch Antwerp and Fund for Scientific Research - Flanders (F.W.O. - Vlaanderen, grant G.0027.97). S. E. is a Research Assistant and R. D. is a post-doctoral fellow of the Fund for Scientific Research - Flanders (F.W.O. - Vlaanderen). REFERENCES ALLEN K. M., WALSH C. A. Genes that regulate neuronal migration in the cerebral cortex. Epilepsy Res, 1999, 36 : 143-154. ANDREWS P. I. DICHTER M. A., BERKEVIC S. F., NEWTON M. R., MCNAMARA J. O. Plasmapheresis in Rasmussen’s encephalitis. Neurology, 1996, 46 : 242-246. AYALA G., MATSUMOTO H., GUMNIT R. Excitability changes and inhibitory mechanisms in neocortical neurons during seizures. J. Neurophysiol., 1970, 33 : 73-85. BERKOVIC S. F., MCINTOSH A., HOWELL R. A., MITCHELL A., SHEFFIELD L. J. et al. Familial temporal lobe epilepsy : a common disorder identified in twins. Ann. Neurol., 1996, 40 : 227-235. BERKOVIC S. F., SCHEFFER I. E. Epilepsies with single gene inheritance. Brain Dev., 1997a, 19 : 13-18. BERKOVIC S. F., KUZNIECKY R. I., ANDERMANN F. Human epileptogenesis and hypothalamic hamartomas : new lesions from an experiment of nature. Epilepsia, 1997b, 38 : 1-3. BIERVERT C., SCHROEDER B. C., KUBISCH C., BERKOVIC S. F., PROPPING P. et al. A potassium channel mutation in neonatal human epilepsy. Science, 1998, 279 : 403-406. BORDEY A., SONTHEIMER H. Properties of human glial cells associated with epileptic foci. Epilepsy Res., 1998, 32 : 286-303. BRINES M. L., SUNDARESAN S., SPENCER D. D., DE LANEROLLE N. C. Quantitative autoradiographic analysis of ionotropic glutamate receptor subtypes in human temporal lobe epilepsy : up-regulation in reorganized epileptogenic hippocampus. Eur. J. Neurosci., 1997, 9 : 2035-2044. CASCINO G. D., ANDERMANN F., BERKOVIC S. F., KUZNIECKY R. I., SHARBROUGH F. W. et al. Gelastic seizures and hypothalamic hamartomas : evaluation of patients undergoing chronic intracranial EEG monitoring and outcome of surgical treatment. Neurology, 1993, 43 : 747-750. CAVAZOS J. E., GOLARAI G., SUTULA T. P. Mossy fiber synaptic reorganization induced by kindling : time course of development, progression, and permanence. J. Neurosci., 1991, 11 : 2795-2803. CHAPMAN A. G. Glutamate receptors in epilepsy. Prog. Brain Res., 1998, 116 : 371-383. CHEVASSUS AU LOUIS N., BARABAN S. C., GAIARSA J. L., BEN ARI Y. Cortical malformations and epilepsy : new insights from animal models. Epilepsia, 1999, 40 : 811-821. DAMBINOVA S. A., IZYKENOVA G. A., BUROV S. V., GRIGORENKO E. V., GROMOV S. A. The presence of autoantibodies to N-terminus domain of GluR1 subunit of AMPA receptor in the blood serum of patients with epilepsy. J. Neurol. Sci., 1997, 152 : 93-97. DE DEYN P. P., MACDONALD R. L. Guanidino compounds that are increased in uremia inhibit GABA- and glycine-responses on mouse neurons in cell culture. Ann. Neurol., 1990, 28 : 627-633. DE DEYN P. P., MARESCAU B., MACDONALD R. L. Epilepsy and the GABA-hypothesis. A brief review and PATHOPHYSIOLOGY OF EPILEPSY some examples. Acta Neurol. Belg., 1990, 90 : 6581. DE DEYN P. P., D’HOOGE R., MARESCAU B., PEI Y. Q. Chemical models of epilepsy and their applicability in the development of anticonvulsants. Epilepsy Res., 1992, 12 : 87-110. DE DEYN P. P., D’HOOGE R., VAN BOGAERT P. P., MARESCAU B. Endogenous guanidino compounds as uremic neurotoxins. Kidney Int., In Press (a). DE DEYN P.P., ENGELBORGHS S., D’HOOGE R. Animal models of focal epilepsy and their contribution to the development of new antiepileptic agents. In : Neuropharmacology of Epilepsy. GROSS R. (ed.). Humana Press, In Press (b). D’HOOGE R., RAES A., LEBRUN P., DILTOER M., VAN BOGAERT P. P. et al. N-methyl-D-aspartate receptor activation by guanidinosuccinate but not by methylguanidine : behavioral and electrophysiological evidence. Neuropharmacology, 1996, 35 : 433-440. D’HOOGE R., DE DEYN P. P., VAN DE VIJVER G., ANTOONS G., RAES A. et al. Uraemic guanidino compounds inhibit g-aminobutyric acid-evoked whole cell currents in mouse spinal cord neurones. Neurosci. Lett., 1999, 265 : 83-86. DICHTER M., AYALA G. Cellular mechanisms of epilepsy : a status report. Science, 1987, 237 : 157-164. DICHTER M. A. Basic mechanisms of epilepsy : targets for therapeutic intervention. Epilepsia, 1997, 38 (Suppl. 9) : S2-S6. ENGELBORGHS S., PICKUT B. A., D’HOOGE R., WIECHERT P., HAEGELE K. et al. Behavioral effects of vigabatrin correlated with whole brain gaminobutyric acid metabolism in audiogenic sensitive rats. Arzneim.-Forsch. / Drug Res., 1998a, 7 : 713-716. ENGELBORGHS S., MARIËN P., MARTIN J. J., DE DEYN P. P. Functional anatomy, vascularisation and pathology of the human thalamus. Acta Neurol. Belg., 1998b, 98 : 252-265. ENGELBORGHS S., MARIËN P., DE DEYN P. P. Gelastic epilepsy caused by hypothalamic hamartoma. Lancet, 2000, 356 : 940. ESCAYG A., MACDONALD B. T., MEISLER M. H., BAULAC S., HUBERFELD G. et al. Mutations of SCN1A, encoding a neuronal sodium channel, in two families with GEFS+2. Nat. Genet., 2000, 24 : 343-345. FILAKOVSZKY J., GERBER K., BAGDY G. A serotonin-1A receptor agonist and an N-methyl-D-aspartate receptor antagonist oppose each others effects in a genetic rat epilepsy model. Neurosci. Lett., 1999, 261 : 89-92. FISHER R. S. Animal models of the epilepsies. Brain Res. Rev., 1989, 14 : 245-278. FOX J. W., LAMPERTI E. D., EKSIOGLU Y. Z., HONG S. E., FENG Y. et al. Mutations in filamin 1 prevent migration of cerebral cortical neurons in human periventricular heterotopia. Neuron, 1998, 21 : 1315-1325. FRITSCHY J. M., KIENER T., BOUILLERET V., LOUP F. GABAergic neurons and GABA(A)-receptors in temporal lobe epilepsy. Neurochem. Int., 1999, 34 : 435-445. 211 GARDINER R. M. Impact of our understanding of the genetic aetiology of epilepsy. J. Neurol., 2000, 247 : 327-334. GEORGAKOULIAS N., VIZE C., JENKINS A., SINGOUNAS E. Hypothalamic hamartomas causing gelastic epilepsy : two cases and a review of the literature. Seizure, 1998, 7 : 167-171. GERBER K., FILAKOVSKY J., HALASZ P., BAGDY G. The 5HT1A agonist 8-OH-DPAT increases the number of spike-wave discharges in a genetic rat model of absence epilepsy. Brain Res., 1998, 807 : 243245. GLEESON J. G., ALLEN K. M., FOX J. W., LAMPERTI E. D., BERKOVIC S. et al. Doublecortin, a brain-specific gene mutated in human X-linked lissencephaly and double cortex syndrome, encodes a putative signaling protein. Cell, 92 : 63-72. GODDARD G. V. Development of epileptic seizures through brain stimulation at low intensity. Nature, 1967, 214 : 1020-1021. GRISAR T., LAKAYE B., THOMAS E., BETTENDORF L., MINET A. The molecular neuron-glia couple and epileptogenesis. In : Jasper’s basic mechanisms of the epilepsies. Third edition. Advances in Neurology, Vol. 79. DELGADO-ESCUETA A. V., WILSON W. A., OLSEN R. W., PORTER R. J. (eds.). Philadelphia, Lippincott Williams & Wilkins, 1999, 591-602. HAND K. S., BAIRD V. H., VAN PAESSCHEN W., KOEPP M. J., REVESZ T. et al. Central benzodiazepine receptor autoradiography in hippocampal sclerosis. Br. J. Pharmacol., 1997, 122 : 358364. HARA M., SASA M., KAWABATA A., SERIKAWA T., YAMADA T. et al. Decreased dopamine and increased norepinephrine levels in the spontaneously epileptic rat, a double mutant rat. Epilepsia, 1993, 34 : 433-440. HORTON R. W., PRESTWICH S. A., MELDRUM B. S. g-Aminobutyric acid and benzodiazepine binding sites in audiogenic seizure-susceptible mice. J. Neurochem., 1982, 39 : 864-870. HWA G. G. C., AVOLI M. Excitatory synaptic transmission by NMDA and non-NMDA receptors in the superficial/middle layers of the epileptogenic human neocortex maintained in vitro. Neurosci. Lett., 1992, 143 : 83-86. JACOBS K. M., KHARAZIA V. N., PRINCE D. A. Mechanisms underlying epileptogenesis in cortical malformations. Epilepsy Res., 1999, 36 : 165-188. JOKEIT H., EBNER A., ARNOLD S., SCHULLER M., ANTKE C. et al. Bilateral reductions of hippocampal volume, glucose metabolism, and Wada hemispheric memory performance are related to the duration of mesial temporal lobe epilepsy. J. Neurol., 1999, 246 : 926-933. JUHÁSZ C., NAGY F., WATSON C., DA SILVA E. A., MUZIK O. et al. Glucose and [11C]flumazenil positron emission tomography abnormalities of thalamic nuclei in temporal lobe epilepsy. Neurology, 1999, 53 : 2037-2045. KUZNIECKY R., GUTHRIE B., MOUNTZ J., BEBIN M., FAUGHT E. et al. Intrinsic epileptogenesis of hypothalamic hamartomas in gelastic epilepsy. Ann. Neurol., 1997, 42 : 60-67. 212 S. ENGELBORGHS ET AL. KUZNIECKY R., HUGG J. W., HETHERINGTON H., BUTTERWORTH E., BILIR E. et al. Relative utility of 1H spectroscopic imaging and hippocampal volumetry in the lateralization of mesial temporal lobe epilepsy. Neurology, 1998, 51 : 6671. LEVY R. H., MATTSON R. H., MELDRUM B. S. Antiepileptic drugs. New York, Raven Press, 1995. LÖSCHER W., SCHWARK W. S. Evidence for impaired GABAergic activity in the substantia nigra of amygdaloid kindled rats. Brain Res., 1985, 339 : 146-150. LÖSCHER W., SWARTZ-PORSCHE D. Low levels of gaminobutyric acid in cerebrospinal fluid of dogs with epilepsy. J. Neurochem., 1986, 46 : 13221325. LOUVEL J., PUMAIN R. N-Methyl-D-aspartate-mediated responses in epileptic cortex in man : An in vitro study. In : Neurotransmitters in epilepsy. AVANZINI S., ENGEL J., FORELLO R., HEINEMANN U. (eds.). Amsterdam, Elsevier, 1992, 361-367. MACDONALD R. L., KELLY K. M. Antiepileptic drug mechanisms of action. Epilepsis, 1995, 36 Suppl. 2 : S2-12. MAYBERG H. S., SADZOT B., MELTZER C. C., FISHER R. S., LESSER R. P. et al. Quantification of mu and nonmu opiate receptors in temporal lobe epilepsy using positron emission tomography. Ann. Neurol., 1991, 30 : 3-11. MCNAMARA J. O. Emerging insights into the genesis of epilepsy. Nature, 1999, 399 (suppl) : A15-A22. MELDRUM B. Epileptic seizures. In : Basic Neurochemistry (5th edition). SIEGEL G.J., AGRANOFF B. W., ALBERS R. W., MOLINOFF P. B. (eds.). New York, Raven Press, 1994, 885-898. MICHELUCCI R., PASSARELLI D., PITZALIS S., DAL CORSO G., TASSINARI C. A. et al. Autosomal dominant partial epilepsy with auditory features : description of a new family. Epilepsia, 2000, 41 : 967-970. MOODY W., FUTAMACHI K., PRINCE D. Extracellular potassium activity during epileptogenesis. Exp. Neurol., 1974, 42 : 248-263. MORI A., HIRAMATSU M., NAMBA S., NISHIMOTO A., OHMOTO T. et al. Decreased dopamine level in the epileptic focus. Res. Commun. Chem. Path. Pharmacol., 1987, 56 : 157-164 MUNARI C., KAHANE P., FRANCIONE S., HOFFMANN D., TASSI L. et al. Role of the hypothalamic hamartoma in the genesis of gelastic fits (a video-stereoEEG study). Electroencephalogr. Clin. Neurophysiol., 1995, 95 : 154-160. OLSEN R. W., WAMSLEY J. K., MCCABE R. T., LEE R. J., LOMAX P. Benzodiazepine / gamma-aminobutyric acid receptor deficit in the midbrain of the seizure-susceptible gerbil. Proc. Nat. Acad. Sci. USA, 1985, 82 : 6701-6705. PELTOLA J., KULMALA P., ISOJARVI J., SAIZ A., LATVALA K. et al. Autoantibodies to glutamic acid decarboxylase in patients with therapy resistent epilepsy. Neurology, 2000, 55 : 46-50. PRINCE D. Cellular mechanisms of interictal-ictal transitions. In : Mechanisms of epileptogenesis : the transition to seizure. DICHTER M. (ed.). New York, Plenum Press, 1988, 57-71. PRZEWLOCKA B., LASON W., MACHELSKA H., VAN LUIJTELAAR G., COENEN A. et al. Kappa opioid receptor agonists suppress absence seizures in WAG/Rij rats. Neurosci. Lett., 1995, 186 : 131134. RASMUSSEN T., OLSZEWSKI J., LLOYD-SMITH D. K. Focal seizures due to chronic localized encephalitis. Neurology, 1958, 8 : 435-455. RENIER W. O., COENEN A. M. L. Human absence epilepsy : the WAG/Rij rat as a model. Neurosci. Res. Comm., 2000, 26 : 181-191. ROGERS S. W., ANDREWS P. I., GAHRING L. C., WHISENAND T., CAULEY K. et al. Autoantibodies to glutamate receptor GluR3 in Rasmussen’s encephalitis. Science, 1994, 265 : 648-651. RUTECKI P., LEBEDA F., JOHNSTON D. Epileptiform activity induced by changes in extracellular potassium in hippocampus. J. Neurophysiol., 1985, 54 : 1363-1374. SAPER C. B. Hypothalamus. In : The human nervous system. PAXINOS G. (ed.). San Diego, Academic Press, 1990, 389-413. SAVIC I., SVANBORG E., THORELL J. O. Cortical benzodiazepine receptor changes are related to frequency of partial seizures : a positron emission tomography study. Epilepsia, 1996, 37 : 236-244. SISODIYA S. M., FREE S. L., DUNCAN J. S., STEVENS J. M. Bilateral periventricular and subcortical heterotopia in a man with refractory epilepsy. Epilepsia, 2000, 41 : 352-354. SNEAD O. C. III. Basic mechanisms of generalized absence seizures. Ann. Neurol., 1985, 37 : 146157. SOLIMENA M., FOLLI F., DENIS-DONINI S., COMI G. C., POZZA G. et al. Autoantibodies to glutamic acid decarboxylase in a patient with stiff-man syndrome. N. Engl. J. Med., 1988, 318 : 1012-1020. STEFANI A., SPADONI F., BERNARDI G. Voltage-activated calcium channels : targets of antiepileptic drug therapy ? Epilepsia, 1997, 38 : 959-965. STEINLEIN O. K., MULLEY J. C., PROPPING P., WALLACE R. H., PHILLIPS H. A. et al. A missense mutation in the neuronal nicotinic acetylcholine receptor a4 subunit is associated with autosomal dominant nocturnal frontal lobe epilepsy. Nature Genet.,1995, 11 : 201-204. STEINLEIN O. K. New insights into the molecular and genetic mechanisms underlying idiopathic epilepsies. Clin. Genet., 1998, 54 : 169-175. STRIANO S., MEO R., BILO L., CIRILLO S., NOCERINO C. et al. Gelastic epilepsy : symptomatic and cryptogenic cases. Epilepsia, 1999, 40 : 294-302. SUTULA T., HE X. X., CAVAZOS J., SCOTT G. Synaptic reorganization in the hippocampus induced by abnormal functional activity. Science, 1988, 239 : 11471150. SUTULA T., CASCINO G., CAVAZOS J., PARADA I., RAMIREZ L. Mossy fiber synaptic reorganization in the epileptic human temporal lobe. Ann. Neurol., 1989, 26 : 321-330. SUTULA T. Kindling and synaptic potentiation : a role for enhanced synaptic efficacy in the development of epilepsy. In : Kindling and Synaptic Plasticity. MORRELL F. (ed.). Boston, Birkhäuser, 1991, 5477. PATHOPHYSIOLOGY OF EPILEPSY THOMAS P. Traitement médical des épilepsies. Encycl. Méd. Chir., Neurologie, 1999, 17-045-A-50. TOPCU M., TURANLI G., AYNACI F. M., YALNIZOGLU D., SAATCI I. et al. Rasmussen encephalitis in childhood. Child’s Nerv. Syst., 1999, 15 : 395-402. TRAUB R., DUDEK F., TAYLOR C., KNOWLES W. Simulation of hippocampal afterdischarges synchronized by electrical interactions. Neuroscience, 1985, 14 : 1033-1038. TRAUB R. D., BORCK C., COLLING S. B., JEFFERYS J. G. R. On the structure of ictal events in vitro. Epilepsia, 1996, 37 : 879-891. UNGER F., SCHROTTNER O., HASELSBERGER K., KORNER E., PLOIER R. et al. Gamma knife radiosurgery for hypothalamic hamartomas in patients with medically intractable epilepsy and precocious puberty. J. Neurosurg., 2000, 92 : 726-731. VAN GELDER N., ASLAM JANGUA N., METRAKOS K., MACGIBBOB B., METRAKOS J. Plasma aminoacids in 3/sec spike-wave epilepsy. Neurochem. Res., 1980, 5 : 661-673. WALLACE R. H., WANG D. W., SINGH R., SCHEFFER I. E., GEORGE A. L. Jr. et al. Febrile seizures and gener- 213 alized epilepsy associated with a mutation in the Na+-channel b1 subunit gene SCN1B. Nature Genet., 1998, 19 : 366-370. WINAMER M. R., OTTMAN R., HAUSER W. A., PEDLEY T. A. Autosomal dominant partial epilepsy with auditory features : defining the phenotype. Neurology, 2000, 54 : 2173-2176. WONG R., WATKINS D. Cellular factors influencing GABA response in hippocampal pyramidal cells. J. Neurophysiol., 1982, 48 : 938-951. ZUBERI S. M., EUNSON L. H., SPAUSCHUS A., DE SILVA R., TOLMIE J. et al. A novel mutation in the human voltage-gated potassium channel gene (Kv1.1) associates with episodic ataxia type 1 and sometimes with partial epilepsy. Brain, 1999, 122 : 817-825. P. P. DE DEYN, Laboratory of Neurochemistry and Behaviour, University of Antwerp, Universiteitsplein 1, 2610 Antwerp (Belgium).