Influence of Ivabradine on the Anticonvulsant Action of Four Classical Antiepileptic Drugs Against Maximal Electroshock-Induced Seizures in Mice

Although the role of hyperpolarization-activated cyclic nucleotide-gated (HCN) channels in neuronal excitability and synaptic transmission is still unclear, it is postulated that the HCN channels may be involved in seizure activity. The aim of this study was to assess the effects of ivabradine (an HCN channel inhibitor) on the protective action of four classical antiepileptic drugs (carbamazepine, phenobarbital, phenytoin and valproate) against maximal electroshock-induced seizures in mice. Tonic seizures (maximal electroconvulsions) were evoked in adult male albino Swiss mice by an electric current (sine-wave, 25 mA, 0.2 s stimulus duration) delivered via auricular electrodes. Acute adverse-effect profiles of the combinations of ivabradine with classical antiepileptic drugs were measured in mice along with total brain antiepileptic drug concentrations. Results indicate that ivabradine (10 mg/kg, i.p.) significantly enhanced the anticonvulsant activity of valproate and considerably reduced that of phenytoin in the mouse maximal electroshock-induced seizure model. Ivabradine (10 mg/kg) had no impact on the anticonvulsant potency of carbamazepine and phenobarbital in the maximal electroshock-induced seizure test in mice. Ivabradine (10 mg/kg) significantly diminished total brain concentration of phenytoin and had no effect on total brain valproate concentration in mice. In conclusion, the enhanced anticonvulsant action of valproate by ivabradine in the mouse maximal electroshock-induced seizure model was pharmacodynamic in nature. A special attention is required when combining ivabradine with phenytoin due to a pharmacokinetic interaction and reduction of the anticonvulsant action of phenytoin in mice. The combinations of ivabradine with carbamazepine and phenobarbital were neutral from a preclinical viewpoint.     

Anticonvulsant profile of nimodipine and nitrendipine against pentylenetetrazole induced seizures in rats

The activity of nimodipine and nitrendipine against pentylenetetrazole (PTZ) induced seizures in Albino rats was studied alone and in combination with valproate. The median effective dose [ED50] of valproate, nimodipine and nitrendipine were initially determined. All the 3 drugs were injected i.p. 30 min before the induction of seizures. Seizures were induced by PTZ 85 mg/kg i.p., and subsequently the effect of combining ED50 doses of nimodipine and nitrendipine with ED50 dose of valproate was evaluated. ED50 of valproate and nitrendipine were 129 and 2.5 mg/kg respectively. ED50 of nimodipine could not be established since a dose-response relationship was not obtained. Hence, for the purpose of combination studies, 4 mg/kg of nimodipine was used. Both nimodipine (4 mg/kg) and nitrendipine (2.5 mg/kg) decreased the ED50 of valproate from 129 to 40 mg/kg. Both nimodipine and nitrendipine potentiate the activity of valproate against PTZ induced seizures and can be considered as potential adjuvant anticonvulsants which merit further study.     

Dose-finding study with nicotine as a proconvulsant agent in PTZ-induced seizure model in mice

The present study was conducted to investigate the possible interaction between low doses of nicotine and pentylenetetrazole (PTZ) in vivo and also to evaluate the influence of nicotine on the antiseizure efficacy of topiramate and sodium valproate in the PTZ-induced seizure model in mice. Graded dose–response study with nicotine showed the CD50 value for nicotine at 6.76 mg/kg. i.p. Subtheshold dose of nicotine (4 mg/kg, i.p.) pretreatment significantly decreased the CD50 value for PTZ from 47.86 mg/kg, i.p. (of PTZ per se) to 31.62 mg/kg, i.p. Sodium valproate but not topiramate, significantly inhibited PTZ-induced seizures in mice with an ED50 value of 177.83 mg/kg, i.p. Nonconvulsive dose of nicotine (1 mg/kg, i.p.) significantly antagonized the protective efficacy of sodium valproate against PTZ-induced seizures and increased the ED50 value to 338.84 mg/kg, i.p. PTZ-induced seizures significantly increased the mouse brain levels of MDA and reduced the level of GSH while sodium valproate reversed such changes. Nicotine pretreatment reversed the anti-lipid peroxidative action of sodium valproate in the PTZ-induced seizure model in mice. The study highlighted the convulsant as well as proconvulsant role of nicotine and established dose discrimination for nicotine as a proconvulsant agent and an anti-antiseizure agent. The study bears significant clinical relevance particularly amongst epileptic smokers who may show failure of efficacy of antiepileptic agents and present with breakthrough seizure attacks on exposure to nicotine.     

Influence of valproate and carbamazepine on symmetrical cortical penicillin foci in the rat

Two epileptogenic foci were formed in homotopic areas of sensorimotor cortices of both hemispheres by application of penicillin. Both valproate (200 and/or 400 mg/kg i.p.) and carbamazepine (50 and/or 100 mg/kg i.p.) did not significantly alter the synchronization of interictal focal discharges. The incidence of ictal phases was suppressed by both drugs in a dose-dependent manner--higher doses (valproate--400 mg/kg; carbamazepine--100 mg/kg i.p.) fully blocked the seizures. Both drugs did not suppress the projection of focal discharges into the opposite hemisphere but were efficacious against secondary generalization.     

Phencyclidine-induced stereotyped behavior and serotonergic syndrome in rat

Phencyclidine (50 mg/kg, i.p.) induced in rats a biphasic response consisting of serotonergic syndrome followed by stereotyped behavior. The initial serotonergic syndrome was significantly reduced by cinanserin (20 mg/kg, i.p.) and cyproheptadine (2.0 mg/kg, i.p.) but not influenced by haloperidol (0.5 mg/kg, i.p.). The later stereotyped behavior was significantly reduced by haloperidol (0.5 mg/kg, i.p.) but not influenced by cinanserin (20 mg/kg, i.p.) or cyproheptadine (2.0 mg/kg, i.p.). A lower dose of phencyclidine (5.0 mg/kg, i.p.) elicited only haloperidol-sensitive stereotyped behavior. These results indicate that serotonergic and dopaminergic mechanisms may mediate the behavioral effects of phencyclidine in rats.     

Role of tricyclic antidepressants in the central regulation of hyperalgesia and stress analgesia

The effect of hypophysectomy (HE) on pain thresholds was studied in female noninbred rats. Hyperalgesia was observed after HE since the first till the sixth day of the observation period. Droperidol (1 mg/kg i.p.) and amitryptyline (5 mg/kg i.p.) produced hyperalgesia in sham-operated rats, which was potentiated in hypophysectomized animals. In rats taken into the experiment 3 days after operation, no increase in the pain threshold was recordable during the 30-minute painful stress, and poststress autoanalgesia did not develop subsequently. The opposing data were obtained in sham-operated animals. On intraperitoneal administration of phentanyl (25 micrograms/kg) after the 30-minute painful stress hypophysectomized rats did not manifest any potentiation of its analgesic effect in contradistinction to sham-operated animals. Simultaneous administration of phentanyl at the same dose and melipramine (5 mg/kg i.p.) produced considerable potentiation of analgesia if administered after stress. In hypophysectomized rats, that effect was somewhat reduced.     

Benzodiazepine antagonists abolish electrophysiological effects of sodium valproate in the rat

A hypothesis was considered that anti-epileptic potency of sodium valproate (VPA) may be associated with its action via the benzodiazepine system. The ability of anti-petit mal drugs to suppress the slow secondary negative wave (SNW) of the visually evoked potential was used as a sensitive electrophysiological “tag” for comparison of VPA (200 mg/kg, i.p.) and Diazepam (5 mg/kg, i.p.) effects. Both drugs induced a profound inhibition of the SNW. Benzodiazepine antagonists Ro 5-3663 (2 mg/kg, i.p.) and Ro 15-1788 (5 mg/kg, i.p.) caused recovery of the SNW amplitude within several minutes of injection. Both antagonists abolished immobility and sedation produced by VPA and Diazepam. The possibility should be considered that therapeutic effects of VPA are mediated through the benzodiazepine receptor coupled to GABA.     

5,5-Diphenylhydantoin Antagonizes Neurochemical and Behavioral Effects of p, p'-DDT but Not of Chlordecone

Abstract: Rats were given 75 mg/kg of 5,5-diphenylhydantoin (phenytoin) or vehicle 30 min prior to 75 mg/kg of 1, 1, 1-trichloro-bis( p -chlorophenyl)ethane ( p, p' -DDT) (p.o.) or chlordecone (i.p.) and tremor was measured 12 h later. Rats were then killed, and regional brain levels of biogenie amines and their acid metabolites and amino acids were determined. Pretreatment with phenytoin significantly attenuated the tremor produced by p, p' -DDT but enhanced that produced by chlordecone. p, p' -DDT had significant effects on the levels of asparate, glutamate, 5-hydroxyindoleacetic acid (5-HIAA), and 3-methoxy-4-hydroxyphenylglycol (MHPG), whereas chlordecone increased glycine, 5-HIAA, and MHPG levels. Pretreatment with phenytoin blocked p.p' -DDT-induced increases of aspartate in the brainstem and spinal cord, 5-HIAA in the hippocampus, and MHPG in the brainstem and hypothalamus. Phenytoin significantly enhanced chlordecone-induced increases of MHPG in the brainstem. These data indicate that organo-chlorine-induced increases in noradrenergic activity in the brainstem and spinal cord may be directly related to the tremorigenic effects of these chemicals.     

Decreased Benzodiazepine Receptor Density in Rat Cerebellum Following Neurotoxic Doses of Phenytoin

Abstract: The effect of acute and chronic administration of phenytoin on [³H]-flunitrazepam binding was examined in the rat cerebellum. There was no significant effect of phenytoin on [³H]flunitrazepam binding in the rat cerebellum 1 and 6 h after a single i.p. injection of 200 mg/kg of phenytoin. However, after 14 days and 28 days of chronic phenytoin administration, significant de-creases in [³H]flunitrazepam receptor density were observed, with no changes in apparent affinity constants in the rat cerebellum. This effect of phenytoin was dose-dependent, as lower doses of phenytoin (100 mg/kg/day) for 14 or 28 days produced no alterations in [³H]flunitrazepam binding in the rat cerebellum. Light-microscopic examination of the rat cerebellum treated with 200 mg/kg/day of phenytoin for 14 days showed degeneration of the Purkinje cells, with edematous Bergmann astrocytes. These data provide evidence for the neuronal localization of benzodiazepine receptors on cerebellar Purkinje cells.     

Valproic acid and bicuculline affect the formation of glycerolipid in rat brain

To ascertain the effects of bicuculline and of sodium valproate on the incorporation of glycerol into rat brain lipid, rats were divided into 5 groups: (a) controls; (b) treated with sodium valproate (400 mg/kg body wt); (c) treated with bicuculline (12.5 μmol/kg body wt); (d) treated with sodium valproate as in (b) + bicuculline as in (c); and (e) treated with bicuculline (25 μmol/kg body wt). Only rats of group (c) had seizures, which lasted until the end of the experiment. Each animal received 20 μCi of [2-³H]glycerol by intraventricular route and was sacrificed 12 min afterwards. Hippocampi and cerebella were taken and lipid extracted and separated by chromatography.

The type of treatment influenced very much the fate of injected, labeled glycerol. Indeed, total recovered radioactivity increased following either convulsions or the administration of valproate, whereas both treatments decreased the amount of radioactivity incorporated into lipid. These effects were more evident in cerebella than in hippocampi.

The distribution of radioactivity among lipid classes (diglyceride, triglyceride and total phospholipid) was also affected by seizures, which decreased the labeling ratio phospholipid/neutral lipid. The distribution of radioactivity among phospholipid classes was influenced by bicuculline (both at convulsant and non-convulsant doses) and these effects were sometimes antagonized by valproate. We conclude that some effects of bicuculline are exerted through the systemic modifications due to seizures and that other effects are probably connected to neuronal hyperfiring. The data reported in this paper are consistent with both mechanisms of action proposed for valproate, i.e. increased membrane permeability and modifications of GABAergic systems.     

Phenytoin-valproate interaction: importance of saliva monitoring in epilepsy.

Sodium valproate is often used with phenytoin when epilepsy cannot be controlled by a single drug. Sodium valproate depresses phenytoin protein binding and so invalidates plasma phenytoin monitoring as a means of determining precise phenytoin dosage requirements. Plasma and saliva phenytoin and plasma valproate concentrations were measured in 42 patients with epilepsy receiving both drugs. Phenytoin protein binding was also measured by ultrafiltration in 19 of these patients and 19 patients taking phenytoin alone. Saliva phenytoin concentration bore the same close correlation to unbound (therapeutically active) phenytoin in patients receiving both drugs as it did in patients receiving phenytoin alone, whereas plasma total phenytoin did not. The same therapeutic range for saliva phenytoin (4-9 mumol/1; 1-2 microgram/ml) was therefore valid in both groups. The depression of phenytoin binding was directly related to the plasma concentration of valproate both in random samples taken from the 42 patients and in samples taken throughout the day in two of these patients. This was confirmed in vitro. Even when the concentration of valproate is known the degree of binding cannot be predicted. Saliva rather than plasma monitoring of phenytoin treatment is therefore valuable in the presence of valproate and with reduced phenytoin binding from any cause.     

Which drug for the adult epileptic patient: phenytoin or valproate?

A series of 140 previously untreated patients with tonic-clonic or partial seizures were randomised to receive either phenytoin or sodium valproate. There was no difference between the treatment groups in pretreatment variables that might influence outcome. Sodium valproate and phenytoin in the treatment of tonic-clonic or partial seizures showed no difference in efficacy as regards time to two year remission or time to first seizure. When the possible prognostic factors were studied, including history and results of clinical examination and investigations before treatment; the only factor which influenced the proportion of patients achieving two year remission was type of seizure. Patients with a clinical history of partial seizures did significantly less well than those with a history of tonic-clonic seizures only. This study showed no major difference in efficacy between sodium valproate and phenytoin in adults with recent onset of epilepsy, irrespective of the type of seizures that the patient suffered.     

Potentiation of bradykinin-induced vascular permeability increase by prostaglandin E2 and arachidonic acid in rabbit skin.

The activity of prostaglandins (PG) in producing vascular permeability was quantitated by dye extraction method in skin of anaesthetized rabbits. PGE1 and PGE2 (0.01-10 mug) produced increase in vascular permeability. Activity was approximately equal to that of histamine (Hist) and 1/20 of that of bradykinin (BK) on a weight basis. The activity of PGF1alpha and PGF2alpha was only 1/20 of that of PGE1 or PGE2. In spite of the relatively low potency of PGE1 and PGE2 in the rabbit, near threshold doses (0.1 or 1 mug) of PGE2 could potentiate permeability responses to bradykinin (0.1 mug) by 10 or 100-fold, respectively. Equivalent doses (0.1 or 1 mug) of histamine could not potentiate the bradykinin responses. Arachidonic acid (AA) at 1 mug, produced a 10-fold potentiation in the permeability response to bradykinin (0.1 mug). Pretreatment of the rabbits with indomethacin (20 mg/kg, i.p.) reduced the responses of BK (0.1 mug) + AA (1 mug) down to a similar magnitude of those seen with bradykinin alone. However, indomethacin did not block responses to either, BK alone, BK + PGE2, or BK + Hist. Various doses (1, 10, 100 and 300 mug) of arachidonic acid alone also produced increase in cutaneous vascular permeability, although its potency was only 1/3-1/8 of that of PGE2. This activity of arachidonic acid was attributed in part to its bioconversion to PGE2, since its activity was significantly reduced by the prostaglandin antagonist, diphloretin phosphate (DPP) (60 mg/kg, i.v.) and by indomethacin (20 mg/kg, i.p.), which blocks conversion of arachidonic acid to prostaglandins. Arachidonic acid may owe some of its permeability increasing effects to histamine release, since its effects were also reduced by the anti-histamine, pyrilamine (2.5 mg/kg, i.v.).     

Identical phalangeal defects induced by phenytoin and nifedipine suggest fetal hypoxia and vascular disruption behind phenytoin teratogenicity.

In previous experimental studies in rabbits, we have shown that vasodilating drugs (including nifedipine) cause distal digital defects. These defects were preceded by edema, hemorrhage, and finally necrosis of the developed cartilage in the phalanges. The underlying mechanism is most likely a fetal hypoxic response, secondary to maternal hypotension and decreased uteroplacental blood flow. Since phenytoin is known to cause distal digital defects both in man and rabbits, we decided to compare the defects provoked by oral administration of phenytoin (100 mg/kg) versus nifedipine (8.3 mg/kg) to New Zealand White rabbits on days 6-18 of gestation. In order to investigate phase-specificity, phenytoin (150 mg/kg) was given on days 14-17. The result of single dose administration on day 16 of phenytoin (300 mg/kg) versus nifedipine (33.2 mg/kg) was also studied. In this latter experiment maternal heart rate was measured up to 21 hours after phenytoin administration. Phenytoin induced digital defects identical with those produced by nifedipine and caused marked maternal cardiodepression. The defects consisted of a reduction, absence, or abnormal structure of the distal phalanges. The distal phalanx of the fourth digit on the hindpaw was the first to be affected, with inclusion of other phalanges, both on the hind- and forepaws, with increasing dose. The sensitive period for induction and histological appearance of these defects was identical for phenytoin and nifedipine. These results suggest that vascular disruption due to a fetal hypoxic response lies behind phenytoin teratogenicity, as has been shown for vasodilators. A cardiodepressive action on the maternal and fetal hearts, possibly in combination with decreased uteroplacental blood flow, is discussed as a probable factor behind phenytoin teratogenicity.     

Antinociceptive effect induced by intraperitoneal administration of vitamin K2 (menatetrenone) in ICR mice

The antinociceptive effect of vitamin K2 (menatetrenone) in mice was examined using tail-flick and formalin test. Menatetrenone at doses of 10, 50 and 100 mg/kg, i.p. produced a dose-dependent and significant inhibition of the tail-flick response in mice. Menatetrenone (50 and 100 mg/kg, i.p.) had no significant effect on the duration of the first phase of the formalin-induced flinching. However, menatetrenone (100 mg/kg, i.p.) significantly inhibited the second phase of the formalin-induced flinching. I.p. administration of menatetrenone (100 mg/kg) significantly reduced the duration of nociceptive responses induced by i.t. injection of bradykinin, but not of substance P, prostaglandin E2 or N-methyl-D-aspartate (NMDA). These present data suggest that i.p. pretreatment with menatetrenone produced dose-dependent antinociceptive effect in mice. This effect may be, at least in part, mediated by the inhibition of bradykinin dependent nociceptive transmission in the spinal cord.     

Influence of protective drugs on the elevation of extracellular potassium ion concentration in the brain during ischaemia

Influence of drugs on the changes of extracellular potassium ion concentration in the brain during total cerebral ischaemia was investigated. The aorta of the dogs was clamped twice with an intermittent reperfusion period of 60 min. In control experiments no significant difference was found in the elevation of extracellular potassium ion concentration of the brain during the first and second clampings. In the present study drugs were administered 10 min prior the second aorta occlusion. Verapamil in a dose of 0.125 mg/kg proved to be ineffective. Piridoxilate in a dose of 10 mg/kg and piracetam in a dose of 100 mg/kg delayed to a small extent the potassium outflow. The following drugs enhanced significantly the duration before the steep increase of potassium ion outflow: phenytoin in a dose of 10 mg/kg by 31.8 sec (p less than 0.01), ethyl-butyl-thiobarbital in a dose of 15 mg/kg by 30.2 sec (p less than 0.05), and lidocaine in a dose of 100 mg/kg by 115.8 sec (p less than 0.01). Comparing present results to our earlier data (obtained after 50 sec ischaemia) it can be concluded, that these protective influences become more effective during longer ischaemic period (2-5 min), when lidocaine, phenytoin and ethyl-butyl-thiobarbital were used. Moreover, in spite of the observation seen during shorter ischaemia, even piridoxilate and piracetam exerted some degree of protective effect. No such effect of verapamil could be detected in the present experimental model.     

Effect of Withania somnifera glycowithanolides on a rat model of tardive dyskinesia.

Withania somnifera glycowithanolides (WSG) were investigated for their preventive effect on the animal model of tardive dyskinesia (TD), induced by once daily administration of the neuroleptic, haloperidol (1.5 mg/kg, i.p.), for 28 days. Involuntary orofacial movements (chewing movements, tongue protusion and buccal tremors) were assessed as TD parameters. WSG (100 and 200 mg, p.o.), administered concomitantly with haloperidol for 28 days, inhibited the induction of the neuroleptic TD. Haloperidol-induced TD was also attenuated by the antioxidant, vitamin E (400 and 800 mg/kg, p.o.), but remained unaffected by the GABA-mimetic antiepileptic agent, sodium valproate (200 and 400 mg/kg, p.o.), both agents being administered for 28 days like WSG. The results indicate that the reported antioxidant effect of WSG, rather than its GABA-mimetic action, may be responsible for the prevention of haloperidol-induced TD.     

Studies on anticonvulsant actions of L-deprenyl

L-Deprenyl (Selegeline) introduced for use in parkinson's disease, is implicated to show beneficial effects in epilepsy, alzheimer's disease, cognition, depression and other age related neurological diseases. In this study, we investigated the CNS effects of L-deprenyl with special reference to epilepsy, anxiety and cognition and memory in mice. L-deprenyl (10, 20 and 40 mg/kg) showed a significant anticonvulsant activity against pentylenetetrazole (PTZ)-induced convulsions. Combination of L-deprenyl (10 mg/kg) with the sub-protective dose of diazepam (1 mg/kg) showed potentiation of the anticonvulsant effect. In the maximal-electroshock (MES)-induced convulsions, L-deprenyl (10 mg/kg) significantly delayed the onset and decreased the duration of extensor phase. Its combination with the lower dose of phenytoin (10 mg/kg) showed potentiation in response compared to the per se effect of both the drugs. However, L-deprenyl did not show any protective effect in lithium-pilocarpine induced status epilepticus. Acute treatment with L-deprenyl had no effect on learning and memory. In chronic treatment, L-deprenyl per se showed no effect on learning and memory but did improve the condition in mice with scopolamine induced memory deficit. L-Deprenyl per se was anxiogenic though in combination with diazepam (1 mg/kg) it potentiated the antianxiety effect of the latter. The above observations suggest that in epilepsy, L-deprenyl might be acting partially by influencing the GABAA/benzodiazepine mechanism in the brain (similar to diazepam and phenytoin), and in cognition enhancing effect, the cholinergic system might be playing a role. Thus, L-deprenyl could prove to be an adjuvant in the antiepileptic therapy and beneficial in dementia associated with epilepsy.     

Inhibition of GABA system involved in cyclosporine-induced convulsions.

In this study, we attempted to clarify the mechanisms mediating cyclosporine-evoked convulsions. Cyclosporine (50 mg/kg, i.p.) significantly enhanced the intensity of convulsions induced by bicuculline (GABA receptor antagonist), but not those induced by strychnine (glycine receptor antagonist), N-methyl-D-aspartic acid, quisqualic acid or kainic acid (glutamate receptor agonists). Bicuculline plus cyclosporine-induced convulsions were significantly suppressed by an activation of GABAergic transmission with diazepam, phenobarbital and valproate. The GABA turnover estimated by measuring aminooxyacetic acid-induced GABA accumulation in the mouse brain was significantly inhibited by cyclosporine (50 mg/kg, i.p.). When cultured rat cerebellar granule cells were exposed to 1 microM cyclosporine for 24 hr, the specific [3H]muscimol (10 nM) binding to intact granule cells decreased to 53% of vehicle controls. The present study provides the first evidence suggesting that cyclosporine inhibits GABAergic neural activity and binding properties of the GABAA receptor. These events are closely related to the occurrence of adverse central effects including tremors, convulsions, coma and encephalopathy under cyclosporine therapy.     

Sodium valproate exerts anti-conflict activity in rats without any concomitant rise in forebrain GABA level

The role of GABA in the mediation of anti-conflict activity by drugs remains controversial. Amino-oxyacetic acid (AOAA), 30 mg/kg, i.p., 2 h, and γ-vinyl GABA (GVG) 900 mg/kg, i.p., 4 h, elevated rat forebrain GABA, but failed to exert any anti-conflict activity in a waterlick paradigm in rats. The GABA analogue, THIP, 0.1–10.0 mg/kg, i.p., 30 min, was also ineffective. Sodium valproate (VPA), 400 mg/kg, i.p., showed no increase in forebrain GABA at 5 min and 4 h, and a very weak elevation, to 106% of control, at 30 min. However, VPA elicited anti-conflict activity at 5 as well as at 30 min. The VPA mediated anti-conflict behavior at 5 min unrelated to increased forebrain GABA level and the lack of anti-conflict activity of AOAA and GVG in spite of significantly elevated GABA suggest an anti-conflict mechanism independent of increased brain GABA concentration. A GABA receptor involvement in the anti-conflict mechanism of VPA was nevertheless indicated by the ability of bicuculline, 0.3 mg/kg, s.c., 15 min, to completely suppress VPA elicited anti-conflic response at 5 min.