Cerebroprotective effect of lamotrigine after focal ischemia in rats

BACKGROUND AND PURPOSE

Glutamate receptor antagonists are protective in animal models of focal cerebral ischemia. Lamotrigine (3,5-diamino-6-[2,3-dichlorophenyl]-1,2,4-triazine) is an anticonvulsant drug that blocks voltage-gated sodium channels and inhibits the ischemia-induced release of glutamate. We describe the cerebroprotective effect of lamotrigine (as the isethionate salt) after middle cerebral artery occlusion in rats.

METHODS

Neurological deficit and infarct volume (visualized by the lack of reduction of 2,3,5-triphenyltetrazolium chloride) 24 hours after permanent left middle cerebral artery occlusion were studied in Fischer rats (n = 8 per group per dose).

RESULTS

Lamotrigine at 20 mg/kg i.v. over 10 minutes administered immediately after middle cerebral artery occlusion reduced total infarct volume by 31% and cortical infarct volume by 52%. Lamotrigine at 8 mg/kg i.v. over 10 minutes reduced cortical infarct volume by 38%. Lamotrigine at 50 mg/kg i.v. for 10 minutes was not cerebroprotective and induced a decrease of 29 +/- 15 mm Hg in mean arterial blood pressure (P < .05, n = 8). The optimum dose of lamotrigine (20 mg/kg i.v. over 10 minutes) when administered with a 1-hour delay after middle cerebral artery occlusion reduced cortical infarct volume by 41%. Lamotrigine (20 mg/kg i.v. over 10 minutes) with a 2-hour delay after middle cerebral artery occlusion was ineffective. Neurological deficits after 24 hours were improved after immediate treatment with lamotrigine at 20 mg/kg i.v. over 10 minutes.

CONCLUSIONS

The cerebroprotective effect of lamotrigine in rats is limited to a narrow dose range between 8 and 20 mg/kg. Lamotrigine or analogous compounds may be useful when given shortly after the onset of stroke.

Reduced postischemic expression of a glial glutamate transporter, GLT1, in the rat hippocampus

Perturbations of the synaptic handling of glutamate have been implicated in the pathogenesis of brain damage after transient ischemia. Notably, the ischemic episode is associated with an increased extracellular level of glutamate and an impaired metabolism of this amino acid in glial cells. Glutamate uptake is reduced during ischemia due to breakdown of the electrochemical ion gradients across neuronal and glial membranes. We have investigated, in the rat hippocampus, whether an ischemic event additionally causes a reduced expression of the glial glutamate transporter GLT1 (Pines et al. 1992) in the postischemic phase. Quantitative immunoblotting, using antibodies recognizing GLT1, revealed a 20% decrease in the hippocampal contents of the transporter protein, 6 h after an ischemic period lasting 20 min induced by four vessel occlusion. In situ hybridization histochemistry with 35S labelled oligonucleotide probes or digoxigenin labelled riboprobes directed to GLT1 mRNA showed a decreased signal in the hippocampus, particularly in CA1. This reduction was more pronounced at 3 h than at 24 h after the ischemic event. We conclude that the levels of GLT1 mRNA and protein show a modest decrease in the postischemic phase. This could contribute to the delayed neuronal death typically seen in the hippocampal formation after transient ischemia.

Neuroprotective effect of free radical scavengers on beta-N-oxalylamino-L-alanine (BOAA)-induced neuronal damage in rat hippocampus

The neurotoxin beta-N-oxalylamino-L-alanine (BOAA), found in Lathyrus sativus seeds, is thought to be the causative agent of neurolathyrism. We have investigated the neuroprotective effects of free radical scavengers on BOAA-induced toxicity following focal injection (1 microliter) of BOAA and comparing the pathological outcome with the effects of injections of alpha-amino-3-hydroxy-5-methyl-isoxazole-4-propionate (AMPA), kainate (KA) or N-methyl-D-aspartate (NMDA) into the dorsal hippocampus of male Wistar rats. Cellular damage was assessed histologically. BOAA (50 nmol) induced a highly selective pattern of hippocampal damage identical with that seen with AMPA (1 nmol). BOAA-induced neurotoxicity, but not AMPA, KA (0.5 nmol) or NMDA (25 nmol)-induced neurotoxicity, was prevented in a dose-dependent manner by focal co-injection of four potential free radical scavengers; dimethyl sulphoxide (DMSO) (1750-7000 nmol), dimethylthiourea (DMTU) (8000 nmol), dimethylformamide (DMF) (7000 nmol) and mannitol (1000 nmol). These findings suggest that hippocampal damage induced by BOAA involves an interaction between AMPA receptors and free radicals.

Lamotrigine--a novel approach

Lamotrigine is an anticonvulsant chemically related to the antifolate compound pyrimidine. In the maximal electroshock test in rodents it is more potent and has a longer duration of action than other anticonvulsants, and it exhibits little acute neurotoxicity. Lamotrigine suppresses sustained repetitive firing by prolonging inactivation of the sodium channel in the neuronal membrane. It blocks the pathological, but not the normal, release of glutamate and aspartate. Lamotrigine is also cerebroprotective in rodent models of stroke or focal ischaemia. This effect correlates with the suppression of glutamate release and is probably due principally to the action on sodium channels. Lamotrigine has a favourable pharmacokinetic profile. It is rapidly and completely absorbed and has linear pharmacokinetics. Protein binding is moderate and, as lamotrigine does not induce liver enzymes, it does not alter the pharmacokinetics of other anticonvulsants. The half-life and clearance of lamotrigine are altered by concomitant antiepileptic therapy. In the presence of the enzyme inhibitor sodium valproate the half-life is doubled and this interaction is clinically significant.

The role of glutamate in epilepsy and other CNS disorders

Glutamate is the principal excitatory neurotransmitter in the brain and, as such, it inevitably plays a role in the initiation and spread of seizure activity. It also plays a critical role in epileptogenesis. The process of "kindling" limbic seizures in rodents by repeated electrical stimulation is dependent on activation of N-methyl-D-aspartate (NMDA) receptors. The function of these receptors is enhanced in the hippocampus of kindled rats and in the cerebral cortex of patients with focal epilepsy. Microdialysis studies show an increase in the extracellular concentration of glutamate and aspartate before or during seizure onset, suggesting that either enhanced amino acid release or impaired uptake contributes to seizure initiation. Glutamate antagonists selective for NMDA or non-NMDA receptors are potent anticonvulsants when given systemically in a wide variety of animal models of epilepsy. They are of limited efficacy against kindled seizures in rats and (on the basis of preliminary evidence) in patients with drug-refractory complex partial seizures. Cognitive side effects appear to be a significant problem with competitive, as well as noncompetitive, NMDA antagonists. Glutamate receptor antagonists provide significant protection against brain damage following global or focal cerebral ischemia or acute traumatic injury in rodent models. Anticonvulsant compounds of the lamotrigine type, which act on sodium channels and reduce ischemia-induced glutamate release, are cerebroprotective in rodent ischemia models and are free from the cognitive side effects of NMDA-receptor antagonists.

The pyrimidine derivative BW 1003C87 protects against excitotoxic lesions induced by kainate in the rat striatum

We have investigated the neuroprotective effect of the pyrimidine derivative BW 1003C87 (5-[2,3,5-trichlorophenyl] pyrimidine-2,4-diamine ethane sulphonate) against striatal and hippocampal lesions induced by kainic acid (KA), N-methyl-D-aspartate (NMDA) and (S)-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ((S)-AMPA) in the rat. BW 1003C87 20 mg/kg i.p. administered pre- and post-treatment (20 min prior to excitotoxic injection and again 4 h later) protects against the lesions induced by KA (1.1 nmol) in the hippocampus (CA2 pyramidal cells only; 40% protection, P < 0.05). In the striatum, the same dose of BW 1003C87 significantly reduces KA toxicity (80% protection, P < 0.001). BW 1003C87 has no significant effect on the lesions induced by NMDA (30 nmol) or S-AMPA (6 nmol) in either brain region. These results are consistent with previous studies showing that the neurotoxicity of KA occurs via an indirect mechanism involving glutamate release.

Global ischaemia: hippocampal pathology and spatial deficits in the water maze

Spatial deficits were assessed in male Wistar rats which had undergone 4 vessel occlusion for 5, 10, 15 or 30 min. Relationships between the extent of brain damage, the duration of 4-vessel occlusion, and the behavioural impairment consequent upon ischaemia were investigated. Starting 13-18 days after occlusion, rats were trained to find a hidden platform in a Morris water maze. All ischaemic groups were impaired on some performance indices relative to controls, in both acquisition and retention of the platform location. Increasing the duration of ischaemia increased behavioural deficits on some measures, but there was no clear-cut evidence that longer durations of ischaemia resulted in increased behavioural impairments. Histological assessment, at two coronal levels in hippocampus and four coronal levels in cortex and striatum, revealed CA1 cell loss in all ischaemic groups, which varied between 10-100% across the range of durations employed. CA1 cell loss increased as both a linear and quadratic function of increasing the duration of ischaemia. In rats subjected to 5-15 min ischaemia, cell loss was almost exclusively confined to the CA1 area. In rats subjected to 30 min ischaemia there was additional, variable damage in hippocampal areas CA2, 3 and 4, substantial cell loss in the striatum (50-70%) and some neuronal damage in the cortex (largely in layer III). However correlations between CA1 cell loss in ischaemic rats and indices of spatial ability were non-significant, despite avoiding bias in the analysis by ensuring that only those rats with submaximal CA1 cell loss estimates and behavioural impairments were included. Given the lack of correlation between damage to the CA1 region and behaviour, it is suggested that CA1 cell loss may not be the only determinant of the water maze deficits displayed by 4-vessel occlusion ischaemic rats.

Lack of effect of lamotrigine against HPNS in rodent and primate models

The neurophysiological effects of the novel anticonvulsant lamotrigine on the high pressure neurological syndrome, HPNS, were investigated in the rat and nonhuman primate Papio anubis. Rats were exposed to pressure at a rate of 3 ATA per min in a helium/oxygen environment. They were pretreated with either lamotrigine isethionate 15, 30, or 60 mg/kg IP or control vehicle. After 15 and 30 mg/kg there were no changes in onset pressures for any of the grades of tremor or myoclonus. After 60 mg/kg, tremor was much slower, at 7-9 Hz, than the 15-20 Hz seen in controls. Four baboons were exposed to pressure at 0.33 ATA per min in the same environment and treated with lamotrigine isethionate at 7.5 mg/kg/h i.v. Each animal underwent a control and a drug-treated exposure. No changes in the onset or severity of HPNS behavioural signs were observed. However, an increase in alpha wave amplitude of the EEG was almost prevented. In both species sustained myoclonic jerking occurred at pressures similar to those at which seizure activity was observed in control exposures. It is concluded that although lamotrigine is protective in several models of neuronal excitation, it is ineffective in protecting against behavioural signs associated with high atmospheric pressure.

Protection from high pressure induced hyperexcitability by the AMPA/kainate receptor antagonists GYKI 52466 and LY 293558

The neurophysiological effects of 2 novel AMPA/kainate receptor antagonists, GYKI 52466 and LY 293558, on the high pressure neurological syndrome have been investigated in the rat and baboon (GYKI 52466) and rat (LY 293558). Rats were exposed to increasing ambient pressures of helium and oxygen at 3 ATA/min, on one occasion each. GYKI 52466 at 20 mumol/kg i.v. immediately before, followed by 70 mumol/kg/hr i.v. during compression delayed tremor by 85% and myoclonus by 30%, compared with control vehicle, and no side effects were observed. Seizure activity was not affected by any of the doses used. LY 293558 at 36 mumol/kg i.p. delayed tremor and myoclonus (44% and 12%), LY 293558 72 mumol/kg additionally delayed seizure activity (21%). Side effects, principally tranquilization at the higher dose, were also noted. Six baboons were exposed to a maximum pressure of 91 ATA at 0.3 ATA/min, in the same environment, on two occasions. One exposure was treated with an i.v. infusion of GYKI 52466 15.2 mumol/kg/hr, the other with the same volume of control vehicle. Limb and face tremor and myoclonus were delayed and the severity of signs reduced. No seizures were observed in the drug treated group before 91 ATA. EEG changes associated with exposure to pressure were not affected. It is concluded that antagonism at the AMPA/kainate receptor by GYKI 52466 and LY 293558 beneficially alters HPNS signs but in a manner which is dependent on both the drug and species being studied.

Effect of the non-NMDA receptor antagonist GYKI 52466 on the microdialysate and tissue concentrations of amino acids following transient forebrain ischaemia

The effect of the non-N-methyl-D-aspartate (non-NMDA) receptor antagonist 1-(4-aminophenyl)-4-methyl-7,8-methylenedioxy-5H-2,3-benzodiazepine hydrochloride (GYKI 52466) on ischaemia-induced changes in the microdialysate and tissue concentrations of glutamate, aspartate, and gamma-aminobutyric acid (GABA) was studied in rats. Twenty minutes of four-vessel occlusion resulted in a transient increase in microdialysate levels of glutamate, aspartate, and GABA in striatum, cortex, and hippocampus. Administration of GYKI 52466 (10 mg/kg bolus + 10 mg/kg/60 min intravenously starting 20 min before onset of ischaemia) inhibited ischaemia-induced increases in microdialysate glutamate and GABA in striatum without affecting the increases in hippocampus or cortex. Twenty minutes of four-vessel occlusion resulted in immediate small decreases and larger delayed (72 h) decreases in tissue levels of glutamate and aspartate. Transient increases in tissue levels of GABA were shown in all three structures at the end of the ischaemic period. At 72 h, after the ischaemic period, significantly reduced GABA levels were observed in striatum and hippocampus. GYKI 52466, given under identical conditions as above, augmented the ischaemia-induced decrease in striatal tissue levels of glutamate and aspartate, without significantly affecting the decreases in hippocampus and cortex. Twenty minutes of ischaemia resulted in a large increase in microdialysate dopamine in striatum. GYKI 52466 failed to inhibit this increase. Kainic acid (500 microM infused through the probe for 20 min) caused increases in microdialysate glutamate and aspartate in the striatum. GYKI 52466 (10 mg/kg bolus + 10 mg/kg/60 min) completely inhibited the kainic acid-induced glutamate release. In conclusion, the action of the non-NMDA antagonist, GYKI 52466, in the striatum is different from that in the cortex and hippocampus. The inhibition by GYKI 52466 of ischaemia-induced and kainate-induced increases in microdialysate glutamate concentration in the striatum may be related to the neuroprotection provided by GYKI 52466 in this region.

The effect of the non-NMDA receptor antagonist GYKI 52466 and NBQX and the competitive NMDA receptor antagonist D-CPPene on the development of amygdala kindling and on amygdala-kindled seizures

A competitive (NBQX) and a non-competitive (GYKI 52466) AMPA antagonist, and a competitive NMDA antagonist (D-CPPene) were tested against the development of kindling and against fully kindled seizures in amygdala-kindled rats. GYKI 52466, 10 mg/kg given i.p. 5 min prior to electrical stimulation in fully kindled animals, reduces both the cortical after-discharge duration and the behavioural seizure score. GYKI 52466, 20 mg/kg, reduces seizure score and after-discharge duration significantly (after 5-30 min) but the animals show severe motor side effects and an irregular cortical and hippocampal EEG. Administration of GYKI 52466, 10 mg/kg, prior to kindling stimulation on days 3-8, does not slow the development of kindling. NBQX, 20 mg/kg or 40 mg/kg i.p., 30 min prior to stimulation, significantly reduces the seizure score in fully kindled animals. NBQX 20 mg/kg i.p. has no effect on the development of kindling. D-CPPene, 8 mg/kg or 12 mg/kg, 120 min prior to stimulation reduces the behavioural seizure score in fully kindled animals. D-CPPene, 8 mg/kg on days 3-8, delays the development of kindling. NMDA receptors play a key role in the kindling process. Expression of kindled seizures involves non-NMDA and NMDA receptors.

Contrasting effects of D- and L-(E)-4-(3-phosphono-2-propenyl)piperazine-2-carboxylic acid as anticonvulsants and as inhibitors of potassium-evoked increases in hippocampal extracellular glutamate and aspartate levels in freely moving rats

Microdialysis experiments performed in the dorsal hippocampus of freely moving rats showed that L-(E)-4-(3-phosphono-2-propenyl)piperazine-2-carboxylic acid (L-CPPene) is 10 times as potent as D-CPPene in inhibiting potassium-induced increases in extracellular levels of aspartate and glutamate. In control experiments, two 100 mM KCl stimuli (S1 and S2) applied for 10 min each (separated by a 40-min recovery period) produced substantial (300-500%) increases in the extracellular levels of aspartate, glutamate, taurine, and GABA and a 50% decrease in the glutamine level. S2/S1 ratios in the control groups were 0.67 (aspartate), 0.78 (glutamate), 0.83 (GABA), and 0.85 (taurine). In the experimental groups, D- or L-CPPene was applied via the probe during the second potassium stimulus (S2). L-CPPene (25 or 250 microM) produced selective suppression of potassium-induced increases of extracellular glutamate (S2/S1 ratio: 0.25) and aspartate (S2/S1 ratio: 0.20) levels, whereas 250 microM D-CPPene was required to inhibit the extracellular aspartate and glutamate increases. Neither enantiomer of CPPene affected the potassium-induced increases of GABA and taurine or the decrease in extracellular glutamine concentration. An additional study comparing the anticonvulsant potencies of D- and L-CPPene was performed using audiogenic DBA/2 mice. The anticonvulsant potency of D-CPPene, as assessed against sound-induced seizures in DBA/2 mice, was an order of magnitude higher than that of L-CPPene [ED50 clonic phase (intraperitoneal, 45 min): 1.64 mumol/kg and 16.8 mumol/kg, respectively]. We attribute the anticonvulsant action of D-CPPene to its antagonist action at the NMDA receptor.(ABSTRACT TRUNCATED AT 250 WORDS)

Foetal grafts from hippocampal regio superior alleviate ischaemic-induced behavioural deficits

Transitory global cerebral ischaemia produced in rats by four vessel occlusion for 15 min produced substantial loss of CA1 cells in dorsal hippocampus and minimal other intra- and extra-hippocampal damage. Ischaemic rats showed a long-lasting impairment in spatial navigation in the water maze, and such impairment was sensitive to task difficulty. Groups of ischaemic animals were implanted with foetal tissue dissected from hippocampal regio superior (SUP--containing CA1 field), regio inferior (INF--containing dentate gyrus), and basal forebrain, with grafts sited in the alveus above the damaged CA1 region. Behavioral testing in the water maze (acquisition, retention and a working memory task) was conducted over a period of 4 to 12 weeks after grafting. Only rats receiving the SUP graft showed consistent improvement in water maze performance, relative to ischaemic controls, when tested in retention and working memory. Although the selective effect of CA1-containing grafts suggests repairing of the damaged host circuit, functional recovery may have been related to the greater ability of SUP grafts to survive and grow in the host ischaemic hippocampus.

Neuronal damage induced by beta-N-oxalylamino-L-alanine, in the rat hippocampus, can be prevented by a non-NMDA antagonist, 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline

The neurotoxin beta-N-oxalylamino-L-alanine (BOAA), found in Lathyrus sativus seeds, is thought to be the causative agent of neurolathyrism. We have investigated the in vivo mechanism of action of BOAA by focal injection (1 microliter) in the dorsal hippocampus of male Wistar rats and comparing the pathological outcome with the effects of injections (1 microliter) of alpha-amino-3-hydroxy-5-methyl-isoxazole-4-propionate (AMPA), kainate (KA) or N-methyl-D-aspartate (NMDA). Cellular damage induced by the excitatory amino acids in the pyramidal (CA1-CA4) and dentate granule neurones (DG) was assessed histologically 24 h after the injection. The study shows that BOAA (50 nmol) induces hippocampal toxicity with a highly selective pattern of regional cellular damage. The CA1, CA4 and DG subfields show 70-90% neuronal injury whereas CA2 and CA3 show only minimal damage. This pattern of cellular damage is similar to that induced by AMPA (1 nmol) and NMDA (25 nmol) but not KA (0.5 nmol). BOAA-induced neurotoxicity is prevented in a dose-dependent manner by focal co-injection of the non-NMDA receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX) (1-25 nmol) but not by a dose of MK-801 (3 mg/kg i.p.) which is neuroprotective against an injection of NMDA. Delayed focal injections of NBQX (25 nmol) up to 2 h after the BOAA injection result in a significant protection of all pyramidal and granular cell regions. These results indicate that the in vivo hippocampal toxicity of BOAA is mediated by AMPA receptors rather than by KA or NMDA receptors.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of ischaemia and reperfusion on the extra- and intracellular distribution of glutamate, glutamine, aspartate and GABA in the rat hippocampus, with a note on the effect of the sodium channel blocker BW1003C87

The redistribution of neurotransmitter amino acids resulting from 20 min of ischaemia was studied in the rat hippocampus by quantitative, electron microscopic immunocytochemistry and by in vivo microdialysis. Changes in the distribution of glutamate, glutamine, aspartate and GABA in various cell compartments of CA1 were analysed immediately after ischaemia or after 60 min of reperfusion, by incubating ultrathin sections with antisera raised against protein glutaraldehyde conjugates of the respective amino acids and subsequently with a secondary antibody coupled to colloidal gold particles. Transverse microdialysis probes coupled with HPLC and implanted in the same animals were used to determine the extracellular concentration of amino acids in the left hippocampus and to apply a drug (BW1003C87) believed to modify the extracellular release of amino acids induced by ischaemia. Forebrain ischaemia was induced by temporary occlusion of the common carotid arteries in rats with permanently occluded vertebral arteries. The extracellular concentrations of glutamate, aspartate and GABA increased markedly during ischaemia, but returned rapidly to normal during reperfusion. BW1003C87 (250 microM, in the dialysis fluid) did not modify the increase in extracellular concentration of amino acids during ischaemia. Glutamate-like immunoreactivity was reduced in pyramidal cell somata both immediately after ischaemia and after 60 min of reperfusion. This reduction appeared to be somewhat less pronounced for cells in the left hemisphere (perfused with BW1003C87) than in the contralateral hemisphere. Ischaemia caused no consistent changes in terminals. The ratio between the intracellular levels of glutamate and glutamine was assessed by double-labelling immunocytochemistry, using two different gold particle sizes. The glutamate-glutamine ratio in glial cells was greatly increased after ischaemia, but recovered to a normal level within 1 h of reperfusion. Aspartate-like immunoreactivity was substantially reduced in pyramidal cell somata both immediately and 60 min after ischaemia, while profiles that were immunopositive for GABA in control brains showed increased GABA immunolabelling. These results suggest that postsynaptic neuronal elements as well as glial cells contribute to the extracellular overflow of excitatory amino acids during an ischaemic event: post-synaptic elements by leaking or releasing glutamate and aspartate, and glial cells by losing their ability to convert glutamate to glutamine effectively.(ABSTRACT TRUNCATED AT 400 WORDS)

Excitotoxicity and selective neuronal loss in epilepsy

The early stages of selective neuronal loss occurring in the hippocampus and other brain regions after prolonged epileptic activity have fine structural characteristics matching those induced by excitotoxic agents. NMDA receptor antagonists provide protection against such damage. The extracellular concentration of glutamate or aspartate may be transiently raised prior to or early in seizure activity but tends not to match the levels associated with hypothalamic damage in the original paradigm of excitotoxicity. Various aspects of the excitotoxic process are examined to see if they can account for particular details of the pattern of selective neuronal loss. A full explanation of selective vulnerability will take into account not only a range of characteristics of the vulnerable neuron but also its functional network during sustained activity.

Selective excitotoxic pathology in the rat hippocampus

The pattern of cell loss and neuronal degeneration resulting from multiple microinjections of N-methyl-D-aspartate (NMDA), ibotenate (IBO), quisqualate (QUIS), and kainate (KA) into hippocampus was studied, together with the protection provided by the NMDA antagonist 3-(+/-)-2-carboxypiperazin-4-yl-propyl-1-phosphonate (CPP). Histological evaluation was carried out after 7 days of survival. NMDA and IBO resulted in an extensive loss of all cells in the hippocampus including dentate gyrus, hilar cells, and CA3-CA1 pyramidal cells, but there was an absence of damage to areas and structures outside hippocampus. After QUIS and KA injections the hippocampal damage was limited to hilar cells in the dentate gyrus, CA3 pyramidal cells, and partial loss of CA1 cells; there was extensive extrahippocampal damage including entorhinal cortex, amygdala, layers III, V, and VI of ventral neocortex, olfactory areas, and various thalamic nuclei. CPP provided almost complete protection from the effects of intrahippocampal injections of NMDA and IBO, but did not affect the hippocampal cell loss found after QUIS and KA (with the exception of minor protection of some CA1 cells). CPP protected most extrahippocampal sites from the damage resulting from QUIS and KA, indicating that such excitotoxic cell death is indirect and involves NMDA receptor activation by an endogenous agent. The use of multiple microinjections as opposed to single injections allows a clearer interpretation of selective excitotoxic vulnerability.

Cerebroprotective effect of BW619C89 after focal or global cerebral ischaemia in the rat

BW619C89 (4-amino-2-(4-methylpiperazin-l-yl)-5-(2,3,5-trichlorophenyl) pyrimidine) was evaluated for cerebroprotection after focal or global ischaemia. BW619C89, as the mesylate dihydrate salt, 20 mg kg-1, i.v. for 10 min immediately, or with a 1 h delay after permanent middle cerebral artery occlusion in Fischer rats reduces cortical infarct volume (visualized with (2,3,5-triphenyltetrazolium) by 49% (p < 0.05) or by 61% (p < 0.001) and improves neurological deficit. Administration of BW619C89 with a 2 h delay is ineffective. BW619C89, given i.p. 0 and 4 h after 20 min of transient bilateral common carotid artery occlusion in vertebral artery-occluded Wistar rats reduces glutamate release and neuronal cell loss in the hippocampal CA1 sector (p < 0.01) and striatum (p < 0.05). BW619C89 resembles BW1003C87 (5-(2,3,5-trichlorophenyl)-2,4-diaminopyrimidine) in inhibiting veratrine-induced glutamate release and protecting against ischaemic brain damage.

The orally active NMDA receptor antagonist CGP 39551 ameliorates the high pressure neurological syndrome in Papio anubis

The neurophysiological effects of a novel, orally active, competitive N-methyl-D-aspartate (NMDA) receptor antagonist (DL-(E)-2-amino-4-methyl-5-phosphono-3-pentenoic acid ethyl ester), CGP 39551, on the high pressure neurological syndrome (HPNS) were investigated in the non-human primate Papio anubis. Six animals were exposed to maximum pressures of 81 ATA in a helium and oxygen environment, on two occasions. One exposure was pretreated orally with CGP 39551 100 mg/kg 24 h before compression, the other pretreated with an equivalent volume of vehicle, in this case water. CGP 39551 significantly ameliorated the signs of HPNS, compared with controls, at pressures above 31 ATA and prevented the severe signs from occurring at the higher pressures. Onset pressures of the mild signs at low pressures were, however, unaffected. Among EEG changes, the pressure induced reduction in delta wave amplitude was prevented by CGP 39551, but the increase in the amplitude of the 7-9 Hz band was not. It is concluded that CGP 39551 may play an important role in the prophylactic treatment of HPNS.

The pyrimidine-derivative, BW1003C87, protects CA1 and striatal neurons following transient severe forebrain ischaemia in rats. A microdialysis and histological study

The four vessel occlusion model of severe transient global ischaemia in Wistar rats has been used to study changes in the extracellular concentration of amino acids in hippocampus and striatum during and after ischaemia. We have investigated the effect of a pyrimidine derivative. BW1003C87 [5-(2,3,5-trichlorophenyl) pyrimidine-2,4-diamine 1.1 ethanesulphonate], 10 or 20 mg/kg, administered before or after ischaemia, on the amino acid accumulation and the pathological outcome. BW1003C87, 10 or 20 mg/kg, given intraperitoneally 20 min prior to ischaemia significantly reduces the extracellular accumulation of the amino acid in hippocampus and in striatum. BW1003C87, 10 or 20 mg/kg, injected 20 min prior to and 4 h after ischaemia protects against the hippocampal (CA1) and the striatal lesions. Administration of BW1003C87, 20 mg/kg, at 0 and 4 h or at 2 and 6 h post-ischaemia, significantly reduces CA1 lesions whereas no significant protection is seen in the striatum. These data indicate that BW1003C87 is able to inhibit the extracellular accumulation of amino acids during severe forebrain ischaemia. The compound is also a potent neuroprotector in this model (in which N-methyl-D-aspartate receptor antagonists fail to protect CA1 neurons). The effect of BW1003C87 on ischaemic glutamate release may contribute to protection in the striatum; it cannot explain the neuroprotection in CA1 since delayed administration is still highly protective. An action on voltage-sensitive sodium channels may contribute to the effects on amino acid release and to the cerebroprotective effect.

Proconvulsant and anticonvulsant effects of Evans blue dye in rodents

The effect of i.c.v. administration of Evans blue to sound sensitive DBA/2 mice and to genetically epilepsy-prone rats was studied. In mice, Evans blue (3.3-52 nmol) induced: hyperlocomotion, wild running, scratching, clonic muscle spasms, tonic seizure (latency 10-45 min), followed by death or recovery. The CD50 value for clonic seizures for Evans blue was 35(23-53) nmol. Pretreatment (45 min) with Evans blue (13-52 nmol, i.c.v.) dose-dependently reduced the incidence of sound-induced seizures in DBA/2 mice (ED50 value against clonic seizures = 30 [15-58] nmol, i.c.v). In rats, Evans blue (104 nmol, i.c.v.) induced electroencephalographic seizures in the hippocampus and cortex and behavioural limbic seizures with a latency of 15-20 min. A reduction in the mean score (from 5 to 2-3) for behavioural seizures was observed which lasted for 4-5 days in rats electrically-kindled daily in the hippocampal CA3 subsector. Sound-induced clonic seizures in kindled and non-kindled rats were reduced for 3-4 days after administration of Evans blue (104 nmol, i.c.v.).

Excitatory amino acid antagonists, lamotrigine and BW 1003C87 as anticonvulsants in the genetically epilepsy-prone rat

Sound-induced seizures in genetically epilepsy-prone rats were used to compare the anticonvulsant effect of phenytoin and diazepam with compounds which decrease glutamatergic neurotransmission including excitatory amino acid antagonists acting at N-methyl-D-aspartate (NMDA) receptors: D(-)CPPene, CGP 37849 and MK 801 or at the glycine/NMDA site: ACPC (1-aminocyclopropane-dicarboxylic acid) (partial agonist) or non-NMDA receptors: NBQX (2,3-dihydroxy-6-nitro-7-sulfamoylbenzo[f]-quinoxaline.Li) and GYKI 52466 (1-(aminophenyl)-4-methyl-7,8-methylene-dioxy-5H-2,3-benzodiazepin e.HCl) or acting at sodium channels to decrease glutamate release: lamotrigine and BW 1003C87 (5(2,3,5-trichlorophenyl)-2,4-diaminopyrimidine ethane sulphonate). ED50 values against clonic seizures (in mumol/kg at the time of peak anticonvulsant effect) were: phenytoin 30.5 (2 h), diazepam 0.5 (0.5 h), MK 801 0.01 (4 h), D(-)CPPene 1.9 (4 h), CGP 37849 2 (1 h), GYKI 52466 24 (0.25 h), NBQX 40 (0.5 h), ACPC 1053 (0.5 h), BW 1003C87 2.2 (1 h), lamotrigine 4.8 (4 h). BW 1003C87, lamotrigine, MK 801, phenytoin, diazepam and CGP 37849 had the most favourable therapeutic indices (rotarod locomotor deficit ED50/anticonvulsant ED50).

Effects of fetal hippocampal field grafts on ischaemic-induced deficits in spatial navigation in the water maze

Transitory global cerebral ischaemia induced in rats by four vessel occlusion for 15 min produced substantial loss of CA1 cells in dorsal hippocampus, and minimal damage in other intra- and extrahippocampal forebrain regions examined. Ischaemic rats showed long-lasting deficits in spatial navigation in the water-maze, consisting of impaired learning to locate a hidden platform in a novel pool, a substantial increase in time spent searching close to the platform without finding it, and moderate deficits in matching to position in a working memory task. Groups of ischaemic rats were implanted with fetal tissue dissected from hippocampal CA1 field, containing glutamatergic CA1 pyramidal cells, from dentate gyrus, containing glutamatergic dentate granule cells, and from basal forebrain, containing cholinergic cells, with grafts sited in the alveus above the damaged CA1 region, for comparison with non-grafted ischaemic and non-ischaemic control groups, over a series of tests from four to 20 weeks after grafting. All ischaemic groups showed comparable acquisition deficits prior to transplantation, and similar loss of CA1 cells on post mortem examination. When tested in a familiar pool in retention and reversal learning of the original platform position, and a working memory task, all ischaemic rats performed better than in initial acquisition. However, rats receiving CA1 grafts showed the most consistent improvement relative to ischaemic controls. When tested in a second (i.e. novel) pool, ischaemic rats again showed marked impairment, whereas rats with CA1 grafts were significantly superior, and learned as rapidly as non-ischaemic controls. The performance of groups with dentate granule and basal forebrain grafts was similar to that of the non-grafted ischaemic control group throughout testing. These results suggest that ischaemic rats are impaired in the adaptive use of spatial information, as shown by acquisition and working memory deficits, but not in long- or short-term memory storage processes, and are also impaired in precise spatial localization. The effects of CA1 grafts in restoring spatial abilities, shown most clearly when rats were tested in a novel environment, suggest that these grafts may have assisted with repair to the damaged host circuit, rather than acted through the release of an appropriate neurotransmitter, since the glutamatergic dentate granule grafts were ineffective. However, CA1 grafts showed better survival and growth than the other types of transplant, so that functional recovery may have been related to graft viability rather than to the specific type of graft.

Aniracetam reverses the anticonvulsant action of NBQX and GYKI 52466 in DBA/2 mice

Aniracetam (1-p-anisoyl-2-pyrrolidinone) selectively reverses the anticonvulsant activities of the non-NMDA receptor antagonists, GYKI 52466 (1-(4-aminophenyl)-4-methyl-7,8-methylenedioxy-5H-2,3- benzodiazepine.HCl) and, to a lesser extent, NBQX (2,3-dihydroxy-6-nitro-7-sulfamoylbenzo(F)quinoxaline), without affecting the anticonvulsant activity of the competitive NMDA receptor antagonist, D(-)-CPPene, in DBA/2 mice. Pretreatment with aniracetam (50 nmol i.c.v., 15 min before drugs) increases the ED50 values (mumol/kg i.p., 15 min) for GYKI 52466-induced protection against sound-induced clonic seizures in DBA/2 mice 7 fold, from 20.1 (11.9-33.9) to 142 (91.7-219), and for NBQX-induced protection 2 fold, from 39.7 (33.8-46.7) to 85.6 (63.9-115), respectively. Aniracetam on its own (12.5-100 nmol i.c.v.) has no convulsant activity, but reverses the anticonvulsant effect of GYKI 52466 (60 mumol/kg i.p., 15 min) in a dose-dependent manner.