El Mice Research Articles

Molecular mechanism underlying epileptic seizure: forwards development of novel drugs for untreatable epilepsy

For the development of new drugs for hitherto untreatable epilepsy, it is necessary to clarify the basic pathophysiology involved in such epileptic seizures and find the target site. This review focused on molecular events related to the expression and expansion of the epileptic focus which are the target of novel antiepileptics. Immediate early genes such as c-fos followed by expression of nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) have been evidenced as initial important phenomena in the cascade of molecular systems that develop and complement the transient neuronal excitation to long-term neuronal plasticity. Non-receptor type tyrosine kinase Fyn in the Src family has been suggested to promote kindling development via tyrosine phosphorylation of the NMDA-receptor subunit, NR2B. The cause of abnormality in the inhibitory system is induced by lowering of glutamate-dependent GABA release in the epileptic focus within the hippocampus in human temporal epilepsy. This is probably attributed to a decrease in GABA transporters. Regarding abnormality of the excitatory system, there is an increase in glutamate release prior to convulsive seizures, an enhancement of NMDA receptor responsiveness and high levels of AMPA receptors related to convulsion after completion of kindling. In gene analysis of human familiar epilepsy, abnormalities and point mutations have recently been found in the following genes: KCNQ 2 and KCNQ3, coding for K+ channels; CHRNA4 of the nicotinic receptor subunit alpha 4; and the cystatin B gene. In epilepsy model mice, EL mice with several gene mutations known to be involved in the seizures, the El-1 gene contains an abnormality of the ceruloplasmin gene. SER (spontaneously epileptic rat: zi/zi, tm/tm), a double mutant, manifests a deletion of the region containing the aspartoacylase gene related to the tm gene. Since an increase in N-acetyl-L-aspartate (NAA) is observed in the SER brain, NAA may serve to evoke seizures.

S10–3 - Genetical basis of abnormal plasticity in EL mice.

S10–3 - Genetical basis of abnormal plasticity in EL mice.

Antiepileptic effects of allopurinol on EL mice are associated with changes in SOD isoenzyme activities.

We have investigated the potential antiepileptic action of superoxide dismutase (SOD) activities in the brain of the epileptic mutant EL mouse. EL mice which experienced frequent seizures (EL[s]) had abnormally low levels of SOD isoenzyme activity in the hippocampal area. Once epileptogenicity was established in these animals, activity of cyanide-sensitive Cu,Zn-SOD was maintained at significantly lower levels than in control mice. However, cyanide-insensitive Mn-SOD activity was not different from non-epileptic controls. In EL mice which had not experienced seizure provoking stimulations and exhibited no seizures (EL[ns]) there was moderately lower levels of SOD isoenzyme activities compared to controls. In spite of the low level of Cu,Zn-SOD activity in EL[s] mice, the Cu,Zn-SOD protein content was high in the hippocampus of these animals, suggesting that inactive Cu,Zn-SOD might be induced during development. After allopurinol (ALP) was given orally to EL[s] mice, Cu,Zn-SOD activities increased dramatically in the hippocampus and seizure activity was decreased. Even after 48 h, when antiepileptic action of ALP was lost, the SOD activity was maintained at the high level associated with initial ALP administration. EL[s] mice also showed DNA fragmentation in the hippocampal CA1 region and the parietal cortex, detected with in situ terminal transferase-mediated dUTP nick labeling with the aid of alkaliphosphatase or peroxidase. The degree of DNA fragmentation was less severe in EL[ns] mice. We propose that abnormalities in region specific Cu,Zn-SOD isoenzyme activity might produce free radicals, leading to DNA fragmentations and cell loss. This might contribute to hippocampal epileptogenesis in EL mice.

Increased expression of 3-hydroxyanthranilate 3,4-dioxygenase gene in brain of epilepsy-prone El mice

The El mouse is an established animal model for human epilepsy. We previously reported that the level of quinolinic acid (QUIN), an excitotoxin, was high in the brain of epilepsy-prone El mice and that the increased production of QUIN was secondary to an increased activity of 3-hydroxyanthranilate 3,4-dioxygenase (3-HAO, EC 1.13.11.6) in the brains of these mice. In this study, we cloned and sequenced the cDNA for 3-HAO and showed that its expression in the brain of El mice was higher than that of control ddY mice. These results suggest that a genetic defect leading to derepression of the 3-HAO gene expression in the brain may be involved in the pathogenesis for the epileptic diseases of El mice.

Role of nitric oxide in the hippocampus of epileptic mutant el mice

Exposure to ionizing radiation causes damage to living tissues; however, only a small number of agents have been approved for use in radiation injuries. Radioprotector is the primary countermeasure to radiation injury and none radioprotector has indeed reached the drug development stage. Repurposing the long list of approved, non-radioprotective drugs is an attractive strategy to find new radioprotective agents. Here, we applied a computational approach to discover new radioprotectors in silico by comparing publicly available gene expression data of ionizing radiation-treated samples from the Gene Expression Omnibus (GEO) database with gene expression signatures of more than 1309 small-molecule compounds from the Connectivity Map (cmap) dataset. Among the best compounds predicted to be therapeutic for ionizing radiation damage by this approach were some previously reported radioprotectors and baclofen (P < 0.01), a chemical that was not previously used as radioprotector. Validation using a cell-based model and a rodent in vivo model demonstrated that treatment with baclofen reduced radiation-induced cytotoxicity in vitro (P < 0.01), attenuated bone marrow damage and increased survival in vivo (P < 0.05). These findings suggest that baclofen might serve as a radioprotector. The drug repurposing strategy by connecting the GEO data and cmap can be used to identify known drugs as potential radioprotective agents.

Abnormalities in SOD Isoenzyme Patterns and DNA Fragmentation During Development of the Epileptic Mutant EL Mouse

Purpose: Several lines of evidence suggest that free radicals and their scavenger enzyme superoxide dismutase (SOD) are involved in long‐term potentiation (LTP) and epileptogenesis in the CA1 field of the hippocampus. In rats with kainate‐induced epilepsy and tottering mice (a model for petit ma1 seizures), “apoptosis” has been observed in neurons. Free radicals in the brain have been suggested to facilitate the apoptosis. The EL mouse, a mutant model of epilepsy, shows secondarily generalized seizures. In the mutant mice, the susceptibility to seizures gradually increases during development and through the experience of repetitive seizures. The parietal cortex in EL mice plays an important role in the seizure initiation, and the hippocampus shows a high electrical excitability during generalization of the seizures. In this study, we examined SOD isoenzyme activities and protein contents in the parietal cortex and hippocampus of EL mice during development. To investigate the region‐specific targets of free radicals in EL mice, the spatial specificity of DNA fragmentation also was studied. Methods: SOD activities were determined by using the method of Fridovich (1991). Total SOD activity included cyanide‐insensitive Mn‐SOD and cyanide‐sensitive Cu, Zn‐SOD. After coincubation with 1 mM cyanide, the activities of Mn‐SOD were assayed. Cu, Zn‐SOD activities were obtained from total SOD activities by subtracting Mn‐SOD activities. Mn‐SOD and Cu, Zn‐SOD contents were assayed with a commercially available enzyme‐linked immunosorbent assay (ELISA) kit (Boehringer). DNA fragmentation was detected in EL mice by using in situ terminal transferase‐mediated deoxyuridine triphosphate (dUTP) nick‐labeling detected with alkaline phosphatase and peroxidase. Results: EL mice which had experienced frequent seizures and established epileptogenicity, EL[s], revealed abnormal SOD isoenzyme activity patterns characterized by a remarkable decrease in the hippocampus. Once epileptogenicity was established, SOD activities, especially cyanide‐sensitive Cu, Zn‐SOD activities, were maintained at low levels. Despite the conspicuous decrease in Cu, Zn‐SOD activities, the protein contents of this isoenzyme were increased in the hippocampus. Cyanide‐insensitive Mn‐SOD activities were not changed. On the other hand, EL mice that had not experienced seizure‐provoking stimulations and exhibited no seizures, EL[ns], showed a moderate decrease in their SOD isoenzyme activities. In EL[s] mice, there was a fragmentation of DNA in the CAI field of the hippocampus and the parietal cortex, whereas in EL[ns], the DNA fragmentation was restricted to the parietal cortex and was much less obvious. Conclusions: The results revealed a region‐specific decrease in the Cu, Zn‐SOD activity and an increase in the regional free radical concentration in the EL brain during epileptogenesis. The discrepancy between enzyme activity and protein content might indicate the induction of an inactive Cu, Zn‐SOD isoenzyme during development. We speculated that region‐specific abnormalities in the activity of Cu, Zn‐SOD isoenzyme, presumably due to a disorder of Cu metabolism, produce spatially specific free radicals, which in turn could cause DNA fragmentation. This process might contribute to the common pathway for epileptogenesis.

Large frequency potentiation induced by 2 Hz stimulation in the hippocampus of epileptic El mice

El mouse has been found to be characteristics with hippocampal disinhibition, and has been suggested decrease in GABAergic synaptic transmission [Ono et al., Brain Res. 745 (1997) 165–172; Fueta et al., Brain Res. 779 (1998) 324–328]. The efficacy of GABAergic synapses can be modulated in response to trains of low frequency stimulation. The frequency potentiation of a population spike (PS) and the field excitatory postsynaptic potential (fEPSP) induced by a low frequency stimulation (2 Hz for 15 s) were recorded for the CA3 subfield, and PS alone for the CA1 subfield and dentate gyrus. PS frequency potentiation was greater in El mice than in non-epileptic control ddY mice. Especially the CA3 subfield exhibited a high PS frequency potentiation (300±73%) compared to age-matched ddY mice (64±24%). However, EPSP frequency potentiation was similar in El and ddY mice. The degree of PS frequency potentiation in CA3 was decreased by the reduction of extracellular Ca 2+ from 2 to 1 mM in both strains, suggesting presynaptic involvement. The potentiation in El mice was suppressed by AMPA/kainate type receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dion (CNQX), but more than half of the control value remained at 5 μM, whereas the potentiation in ddY mice was abolished at this concentration. N-methyl- d-aspartate (NMDA) type receptor antagonist 3-3 (2-carboxypiperazine-4-yl) propyl-1-phosphonate (10 μM; CPP) did not affect the potentiation. Bicuculline (5 μM), GABA A receptor antagonist, did not increase the amplitude of PS during stimulation but induced epileptic (multiple PSs) potentials. High PS frequency potentiation of El mice was mimicked to the degree of that in ddY mice by a low dose of GABA B receptor agonist baclofen (3 μM). The suppression by baclofen was partially reversed by the antagonist saclofen (500 μM). The large frequency potentiation in young El mice, which do not have seizure-susceptibility, indicates an intrinsic property in El mice. It is suggested that the high synchronization of CA3 neurons in El mice is due to a little activation of GABA B receptor activation and also to enhancement of non-NMDA type synaptic transmission.

Influence of dietary zinc on convulsive seizures and hippocampal NADPH diaphorase-positive neurons in seizure susceptible EL mouse

Adequate, high and deficient dietary levels of zinc (Zn) were compared in seizure-susceptible EL mice with respect to convulsions and to nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase-positive hippocampal neurons. Diaphorase positivity is associated with nitric oxide (NO) production. Convulsive seizures in the EL mice given the various diets did not differ over 1–4 weeks, but convulsions in EL mice given the Zn-deficient diet for 4 weeks were more effectively suppressed by injection of zonisamide (ZNS) (75 mg/kg intraperitoneally) than in mice receiving high- or adequate-Zn diet for the same period. Numbers of NADPH diaphorase-positive neurons in the CA1/CA2 region of the hippocampal formation were significantly higher in mice given the Zn-deficient diet for 4 weeks than in mice fed adequate Zn. Mice receiving the high-Zn diet for the same period had significantly fewer NADPH diaphorase-positive neurons in the subiculum than mice with adequate Zn. These results suggest that Zn deficiency inhibits convulsive seizures of EL mice, and that dietary Zn influences numbers of NO producing neurons in the hippocampal formation.

Regional differences in hippocampal excitability manifested by paired-pulse stimulation of genetically epileptic El mice

Hippocampal excitability in El mice was studied by analyzing paired-pulse responses of population excitatory postsynaptic potentials (EPSPs) and population spikes (PSs). In vitro slice preparations from seizure-susceptible adult (15 weeks old) and non-seizure susceptible young (5 weeks old) El mice were compared with age-matched mother strain ddY mice. In CA1 area, paired-pulse inhibition of PSs was reduced by about 50% at 10 ms interpulse interval (IPI) in both 5 and 15 weeks old El mice when compared to ddY mice. Phenobarbital (200 μM) decreased paired-pulse ratio (PPR) by 30% in El mice, and bicuculline (1 μM) increased PPR by 80% in ddY mice at 10 ms IPI. These results suggest an intrinsic existence of decreased GABA A receptor-mediated inhibition in CA1 of El mice. In dentate gyrus (DG), an increase in paired-pulse facilitation of PSs was observed at intermediate IPIs (50–200 ms) in El mice at both ages, especially at 15 weeks of age, when 52%-increased PPR was recorded. The facilitation was not due to GABA A receptor-mediated inhibition and was not age-dependent. In CA3 area, increased paired-pulse facilitation of PSs and EPSPs over the range of 10–1000 ms IPIs was observed only in the 15-week-old El mice. The age-dependent appearance of seizure susceptibility was associated with the increase in excitatory synaptic transmission in CA3. Our results show that El mice possess excitatory/inhibitory synaptic transmission abnormalities in the hippocampus that could contribute to seizure predisposition.

Developmental and regional alteration of kappa-opioid receptors in seizure-susceptible EL mouse brain.

The binding of [3H]ethylketocyclazocine ([3H]EKC) under the suppression of mu and delta sites in the brain of EL mice (seizure-susceptible) was examined to determine the relationship between seizures and the dynorphinergic system. The density of kappa-opioid receptors in the cerebrum of adult EL mice during interictal periods significantly increased, without changes in apparent affinities, compared with that of adult ddY mice (seizure-nonsusceptible; the mother strain of EL). Subsequently, the binding of 0.8 nM [3H]EKC in 8 brain regions was examined in the 2 strains. The [3H]EKC binding in 25-day-old EL mice that had no seizures significantly increased in the hippocampus and amygdala. At the age of 50 days, EL mice displayed abortive seizures, and the number of kappa sites in EL mice was significantly greater in the hippocampus, amygdala and cerebral cortex. It was further shown that the binding of [3H]EKC in 150-day-old adult EL mice during interictal periods was markedly increased in the hippocampus, amygdala, cerebral cortex and striatum, compared with the corresponding regions in ddY mice. The up-regulation of kappa receptors in the EL mouse brain may result from deficits in endogenous dynorphins, which could be involved in the pathogenesis of seizure diathesis and seizure manifestations in the EL mouse.

Effects of Saiko-ka-ryukotsu-borei-to on irritable characteristics in El mice.

In the present study, we attempted to characterize the circadian rhythm of locomotor activity of El mice and to examine the duration of sodium pentobarbital-induced sleep during the light and dark periods. The effect of Saiko-ka-ryukotsu-borei-to, a representative Kampo medicine used for treating insomnia, was studied on the locomotor activity and on the duration of sodium pentobarbital-induced sleep in El mice. The spontaneous locomotor activity of El mice during the dark period was not so different from that of ddY mice, whereas the activity during the light period was significantly higher in the El mice, and the duration of sodium pentobarbital-induced sleep of El mice was very short during the light period, nearly equal to that during the dark period. The administration of Saiko-ka-ryukotsu-borei-to caused marked reduction in the locomotor activity during the light period, and dose-dependent prolongation in the duration of sodium pentobarbital-induced sleep time during the light period, but that during the dark period was unaffected. These findings suggest that the El mouse is a model with a tendency to be easily excitable during the light period compared to ddY mice. Furthermore, it was confirmed that Saiko-ka-ryukotsu-borei-to reduced the excitation in El mice and might be useful against sleep disorder due to excitation.

Environmental Influences on Epilepsy Gene Mapping in El Mice

The epileptic EL mouse has been studied extensively as a genetic model for idiopathic complex partial seizures in humans. The seizures in EL mice occur during routine handling at approximately 90 days of age, but can be induced at younger ages (50 days) by repeated rhythmic vestibular stimulation, e.g., tossing. Six seizure frequency quantitative trait loci (QTLs), EU-E16, were previously mapped in crosses between EL and non-epileptic strains using mechanical tossing procedures beginning at 30 days of age. The presence of these seizure frequency QTLs depended upon genetic background and the type of cross. Here we confirm Chromosome 2 and 9 QTLs in a backcross to the seizure-resistant ABP/LeJ strain with mice tested beginning at 200 days of age. However, the mapping of epilepsy genes was influenced by the seizure testing procedure, i.e., repeated tossing. The maximum Z-score for Ell (Chromosome 9) was 3.7 after 6 tests, but decreased to 2.4 after 15 tests. In contrast, the maximum Z-score for Ell (Chromosome 2) was 2.0 after 6 tests, but increased to 3.9 after 15 tests. In addition to nonallelic interactions (epistasis), our findings indicate that the genetic complexity of tossing-induced seizure susceptibility in EL mice also arises from genotype-environmental interactions involving the seizure test, seizure history, and age.

Selective inhibition by tiagabine of age-dependent hyperexcitability responsible for generating seizures in epileptic EL mice

Selective inhibition by tiagabine of age-dependent hyperexcitability responsible for generating seizures in epileptic EL mice

Nuclear polyadenylate polymerase activity in the brain of seizure-prone EL mice.

Nuclear polyadenylate polymerase from I activity in the brains of seizure-prone EL mice was significantly higher than in seizure-non-susceptible progenitor ddY mice. This finding may be essential in acquiring susceptibility to seizures, since there was no significant difference between EL(S) mice and those that did not receive stimulation, EL(NS) mice. Lower form II enzymatic activity was observed in both groups of EL mice but not in ddY mice. Moreover, significantly lower activities of form II 7 days after seizures were found in EL(S) mice compared with EL(NS) mice, suggesting that this is a consequence of repeated seizures. The activity of form I enzyme decreased immediately and at 30 and 60 min after seizures, then returned to control levels at 100 min. Form II enzymatic activity was significantly decreased only at 30 min after seizures, implying that seizures exerted a later effect on form II enzyme. These changes may cause a decrease in the rate of polyadenylation in the brain; thus, alteration of post-transcriptional events, including messenger RNA processing and transport, may occur during epileptic seizures.

Granule cell disinhibition in dentate gyrus of genetically seizure susceptible El mice

Paired-pulse inhibition was investigated electrophysiologically in the dentate gyrus using hippocampal slices from epileptic El mice. At short interpulse intervals (IPIs), the inhibition was 30% in the El, and 90% in the control ddY mice at the ages of 10 and 15 weeks. No difference in inhibition was observed at the age of 5 weeks. Bicuculline, a GABA A receptor antagonist, attenuated the inhibition during short IPIs in the ddY mice, while in the El mice, phenobarbital and flunitrazepam, which enhance GABA A receptor function, restored the inhibitory activity comparable to that of the ddY. The disinhibition progressed with growth, closely correlating with seizure development in El mice. These results suggest that decrease in the GABAergic inhibition occurs in the dentate gyrus of the El mice with growth. GABA concentration in the hippocampus was also quantified using HPLC. In El mice, GABA level was significantly lower than that in ddY mice at the ages of 5 and 15 weeks. Thus, the disinhibition observed in the El dentate gyrus at 15 weeks of age does not appear to be directly related to the content of GABA. GABAergic disinhibition suggests possible loss of unknown inhibition control factor(s) in the El dentate gyrus as growth progresses. The growth-dependent disinhibition in the granule cells may be a prerequisite for epileptogenesis in El mice.

238 Potentiation and inhibition of GABA receptors expressed in xenopus oocyte by lipid hydroperoxide and food additives

Anticonvulsant actions of the nootropic drug nefiracetam were studied using EL mice, an animal model of epilepsy, in which peripheral-type benzodiazepine receptors (PBRs) might be involved in their epileptogenesis. Nefiracetam, when administered orally to EL mice, inhibited convulsions induced by the PBR agonist, Ro 5-4864, with an ED50 of 17.2 mg/kg, whereas it did not inhibit the drug-induced convulsions in control DDY mice. When administered intravenously (i.v.) to DDY mice, nefiracetam and other piracetam-like nootropics inhibited the Ro 5-4864-induced convulsions in the sequence of nefiracetam>aniracetam≫oxiracetam, piracetam. Spontaneous EL mouse seizures were also inhibited by these nootropics with a similar rank order of potencies. Binding studies for PBRs, performed on crude membranes of brain tissues of these mice, revealed that [3H]Ro 5-4864 and [3H]PK 11195 bindings were both inhibited by micromolar concentrations of nootropic agents in the sequence of nefiracetam>aniracetam≫oxiracetam, piracetam. The results suggest that nefiracetam may exert an anticonvulsant action through interacting with a low-affinity type of PBR in the brain, and could be developed as a promising therapeutic drug for neurological disorders including epilepsies.

Alcohol intake decreases brain Zonisamide concentration in inbred EL mice.

This study was undertaken to evaluate the effect of ethanol intake on the anticonvulsive effects of Zonisamide (ZNS), a sulfonamide derivative. EL mice, which are highly susceptible to seizures, were given a 10% ethanol solution ad lib for 1-4 weeks. In mice given ethanol for 4 weeks, seizures were not suppressed by ZNS at a dose of 75 mg kg-1, i.p. Serum ZNS concentrations following ethanol consumption for 1-4 weeks were higher than in untreated mice; however, brain ZNS concentrations following ethanol consumption were lower than those in untreated mice. These results suggest that alcohol intake decreases the brain concentration of ZNS, but not the serum concentration.

Developmental and seizure-related regional differences in immediate early gene expression and GABAergic abnormalities in the brain of EL mice.

To examine the hypothesized role of the immediate early gene (IEG) response in synaptic plasticity and in epileptogenesis, we studied the spatial specificity of the expression of IEG in EL mice, a well known mutant model of epilepsy. Also to examine the 'GABA hypothesis' in epilepsy, GABA concentration and GAD activity was determined in micro brain regions (10-300 ng) of EL mice related to the focus in the parietal cortex and the hippocampus. We found that the IEG expression after seizures is not related to the seizure pattern, but to the seizure history, seizure threshold and development of EL[s]. Even in the interictal period, EL mice with long seizure histories and very low seizure thresholds demonstrate IEG expression continuously. This is probably strengthened by repeated seizures. The IEG expression site is however located in the hippocampal CA1, which is the final terminal of various inputs from other areas of the limbic system. It is proposed that the continuous expression of IEGs might play a different role from that of transiently expressed IEGs. Developmentally, the site of IEG expression shifted from one site to another in a very similar manner as in the IEG expression with propagation of paroxysmal discharges in each seizure, and the three-dimensional expression area was gradually expanded, suggesting a change in the regional active site during epileptogenesis. These lines of evidence suggest that during development as well as repetitive seizures, frequent expressions of IEGs and syntheses of Fos and Zif proteins might facilitate synaptic conductivity involved in epileptogenesis. The sites of abnormal GABA concentrations and GAD activities were almost the same in the parietal cortex, around Sidman atlas coronal section No. 300 and in the hippocampal CA1 pyramidal cells as the spatio-temporal specific IEG expression sites. These findings strongly suggest that IEG expression and abnormal GABAergic functions are involved in epileptogenesis in EL mice.

Analysis of the kinetics of synaptic inhibition points to a reduction in GABA release in area CA1 of the genetically epileptic mouse, El.

In order to determine whether changes in synaptic inhibition are involved in chronic models of epilepsy, it is necessary to understand the factors which determine the kinetics of fast gamma-aminobutyric acid (GABA)ergic inhibition. For this purpose, we analyzed the decaying phase of isolated inhibitory postsynaptic currents (IPSC) in rats CA1 pyramidal cells. Reduction of GABA release (by reducing [Ca2+]o or paired-pulse stimulation) or blockade of GABA uptake (with tiagabine) led to the conclusion that small changes in the amount of GABA available for postsynaptic binding have little effect on the peak amplitude, but have marked effect on the duration of the IPSC. We then studied isolated GABAA receptor-mediated inhibition in area CA1 of the El mouse strain, which is genetically predisposed to epilepsy. Results were compared with the non-epileptogenic mother strain, ddY. Inhibitory postsynaptic potentials (IPSPs) in El mice (IPSPEl) were not significantly different in amplitude of those from ddY mice (IPSPddY). However, the rise-time and duration of IPSPEl were respectively about 25% and 50% shorter than those of IPSPddY. With appropriate pharmacological manipulation of GABA release or uptake, IPSPEl could be made to resemble the IPSPddY and vice versa. It is concluded that the synaptic release of GABA in area CA1 of the El mouse is decreased compared to that of the ddY mouse.

Down-regulation of S-adenosylhomocysteine hydrolase in the active methyl transfer system in the brain of genetically epileptic El mice.

To study the role of cerebral methylation in epileptogenesis, we investigated the active methyl transfer pathway in the brain of genetically epileptic El mice. We examined S-adenosylhomocysteine (AdoHcy) hydrolase activity (by high performance liquid chromatography), the corresponding mRNA levels (by competitive reverse transcription-polymerase chain reaction), as well as S-adenosylmethionine (AdoMet) and AdoHcy levels in epileptic El and non-epileptic control ddY mice. The level of AdoHcy, a potent feedback inhibitor in the methyl transfer system, was relatively high throughout growth, particularly in 15 week old El mice where it was increased approximately 20% compared to control ddY mice. AdoHcy hydrolase activity in the El mice brain did not increase during growth, which was confirmed by the finding that mRNA synthesis from the hydrolase gene behaved likewise. In contrast, ddY mice exhibited a gradual increase in the mRNA synthesis up to three fold and 20% increase in the enzyme activity at 15 weeks of age compared to those of El mice in which the seizure frequency was 100%. The levels of AdoMet, a versatile methyl donor, did not change throughout growth. We concluded that the down-regulation of AdoHcy hydrolase results in the accumulation of AdoHcy, possibly inducing an unstable state including seizures in the El mouse brain. El mice predisposed to epilepsy may be characterized by disordered feedback regulation of the AdoMet-dependent methyl transfer pathway.