Search

This study examined the effects of carbamazepine (CBZ) or levetiracetam (LEV) and sub-therapeutic doses of the combination of the two conventional antiepileptics on some of the markers of motor coordination. Twenty-four male Wistar rats (140 ± 5 g) were randomized into 4 groups (n = 6). Group I rats received physiological saline (0.2 ml), group II were administered CBZ (25.0 mg/kg), group III received LEV (50 mg/kg), while group IV rats were given sub-therapeutic doses of CBZ (12.5 mg/kg) and LEV (25 mg/kg) intraperitoneally for 28 days. Thereafter the animals were subjected to behavioral and biochemical investigations, while the frontal lobe and cerebellar tissue were preserved for histological investigation. Data were subjected to descriptive and inferential statistics, and the results presented as mean ± SEM, analyzed using one-way Analysis of variance (ANOVA) and Student- Newman Keuls post hoc analysis where appropriate. p < 0.05 was considered statistically significant. There was significant alteration in fine and skilled movement after the CBZ, and CBZ + LEV chronic treatment compared with the control. The CBZ, and CBZ + LEV combination treatment increased the frontal lobe and cerebellar activities of acetylcholinesterase, malondialdehyde concentration, tissue necrotic factor alpha and decreased the activities of super oxide dismutase relative to the control. Disorganization of the histoarchitecture of the frontal lobe and cerebellum was characterized by cellular atrophy, chromatolysis and hyalinization. Chronic CBZ, and CBZ + LEV combination treatment produced psychomotor dysfunction and neurotoxicity in this order CBZ + LEV > CBZ > LEV in the rats., (Copyright © 2021 Elsevier B.V. All rights reserved.)

The scorpion-envenoming syndrome has an incidence of approximately 8000 accidents/year in Brazil; with most severe cases occurring during childhood and elderly. Previous results from our laboratory suggest that the effects of scorpion toxins on the central nervous system play a major role on the lethality induced by scorpion envenoming. Our group has shown that the pre-treatment with carbamazepine (CBZ) is able to increase the latency-to-death in developing animals inoculated with tityustoxin, a toxic fraction of the Tityus serrulatus crude venom. Nevertheless, in order to perceive CBZ as potentially useful in clinical practice, the efficiency of CBZ against crude venom inoculation and the pharmacological treatment introduced after envenomation must be addressed. Thus, the objective of this work was to evaluate CBZ therapeutic efficiency against scorpion envenomation in developing rats. Animals were treated with i.p. injections of either vehicle or CBZ (50 mg/kg or 100 mg/kg) 10 min after injected with a s.c. fixed volume of either saline or crude T. serrulatus venom extract (48 mg/kg). The dose chosen for venom inoculation was 16 times its DL50 for 21-day-old Wistar rats, invariably inducing death within 2 h. Although CBZ did not significantly reduce the pulmonary edema, it was effective in increasing survival rate by approximately 75% in treated rats. In conclusion, CBZ was effective in the treatment of T. serrulatus envenomation even though not blocking the pulmonary edema.

Effects of carbamazepine on behavioral sensitization to morphine in mice has been investigated. Mice treated daily for 7 days with morphine (10 mg/kg) induced behavioral sensitization. Carbamazepine (10, 20, 40 mg/kg, i.p.) itself dose-dependently inhibited the locomotor activity of mice, but did not affect the acute morphine induced hyperactivity. Chronic treatment with carbamazepine had no effect on the development of morphine behavioral sensitization. Co-administration of carbamazepine 30 min prior to morphine had no significant effect on the development of behavioral sensitization. After the behavioral sensitization has been established, carbamazepine (10, 20, 40mg/kg, i.p.) did not affect the expression of morphine sensitization. However, carbamazepine (10, 20, 40mg/kg, i.p.) dose-dependently potentiated the transfer of morphine sensitization. The data of the present study implies that carbamazepine may influence the maintenance process of behavioral sensitization, which results in the enhancement of the transfer of behavioral sensitization. In clinic, the present results suggest that chronic use of carbamazepine might have abuse potential in opioid abusers.

We have recently reported that ictal-like afterdischarges (ADs) analogous to those in in vivo kindling models are induced by high-frequency stimulation (100 Hz, 1s) to the stratum radiatum of the CA1 region of rat hippocampal slices. To test whether this model can also serve as an in vitro seizure model for temporal lobe epilepsy, we examined the effects of antiepileptic drugs on this model and compared them with their effects on seizures in patients with temporal lobe epilepsy. ADs were progressively enhanced following repetitive high-frequency stimulations to slices treated with 4-aminopyridine, a proconvulsive A-type potassium channel blocker. Bath application of phenytoin (1-100 microM) and carbamazepine (1-100 microM) suppressed AD generation in a concentration-dependent manner. At a clinically relevant concentration of 10 microM, phenytoin reduced the number of spikes in an AD to 50.6% and carbamazepine to 39.7% of the control values. On the other hand, ethosuximide had no effect on AD generation at a concentration of 1 mM, which is clinically effective against absence seizures, but enhanced it at a toxic concentration of 10 mM. These findings indicate that the pharmacological profiles of antiepileptic drugs applied to our model correspond to those applied to seizures in patients with temporal lobe epilepsy. We therefore conclude that this model can be a useful in vitro model for the ictal manifestation of temporal lobe epilepsy.

Fyn tyrosine kinase is thought to be involved in the control of neuronal intracellular signal transduction elicited by neurotransmitter stimulation. Emotional disorders, such as fearfulness in Fyn-deficient mice, prompted us to investigate the neural mechanisms that lead to defective emotional expression by using functional neuroanatomical methods. In order to examine the reactivity of a specific neural network to excitatory neurotransmitter administration, we mapped the distribution of c-Fos-immunoreactive neurons after administering N-methyl-D-aspartate (NMDA) to control and mutant mice at the subthreshold dose for seizure induction. The induction of neuronal c-Fos-immunoreactivity by NMDA was enhanced in the Fyn-deficient mice, and there was a much greater increase in immunopositive neurons in certain well-defined areas, such as the amygdaloid medial nuclear subdivisions, hypothalamic paraventricular nucleus, and midbrain periaqueductal gray, of the mutant. NMDA-induced c-Fos expression was attenuated by pretreatment with D-(-)-2-amino-5-phosphonovaleric acid, a competitive NMDA antagonist, both in the control and the mutant mice. These findings suggest that the excitability of the projection system from the amygdala to the hypothalamus and midbrain, the main pathways of emotional expression, is enhanced in Fyn-deficient mice.

The veratridine epileptiform model was utilized to assess the antiepileptic effect of Carbamazepine (CBZ) in rat hippocampal CA1 pyramidal neurons using conventional intracellular recording techniques. In the veratridine model, where brain slices are treated with veratridine (0.3 microM), a single intracellular stimulus evokes epileptiform bursting. Additionally, spontaneous epileptiform activity commonly appears on prolonged exposure to veratridine in this model. In this model, therapeutic (7-15 microM) and high (50 microM) concentrations of CBZ inhibited the evoked and spontaneous epileptiform bursting in a concentration- and voltage-dependent manner. At all concentrations tested, CBZ produced inhibition of epileptiform activity without affecting the membrane resting potential or input resistance. However, at 50 microM, the drug increased the firing threshold of neurons. These results confirm the suitability of this model for testing sodium channel-dependent antiepileptic agents.

Application of 4-aminopyridine (4-AP) has previously been reported to produce different patterns of epileptiform discharges in entorhinal cortex (EC)-hippocampal slices: recurrent short discharges (RSDs) in hippocampal area CA1, seizure-like events (SLEs) and negative-going potentials (NGPs) in the medial entorhinal cortex (mEC). Using recordings of field potentials, we investigated the pharmacological effects of the clinically employed standard anticonvulsant drugs phenytoin (PHT), carbamazepine (CBZ), valproic acid (VPA) and phenobarbital (PHB) and those of pentobarbital (PB) on 4-AP-induced epileptiform activity. The anticonvulsant drugs showed different effects: SLEs were completely blocked by all tested drugs. Valproic acid, which suppressed all epileptiform activities, seemed to have the most fundamental effect of all drugs on 4-AP induced activity, because under phenytoin and carbamazepine, some epileptiform activity was still observable. The RSDs in hippocampal area CA1 of the hippocampus did not respond to the different anticonvulsants. In contrast, PB decreased the frequency of the RSDs in CA1 and enhanced the frequency of the NGPs in the EC. We propose that the activities induced by 4-AP in the combined entorhinal cortex-hippocampal slices may provide an in vitro model for the development of new drugs against difficult-to-treat focal epilepsy.

Inositol uptake was measured at concentrations of 25, 40 and 50 microM in human astrocytoma cell cultures treated for 1-3 weeks with pharmacologically relevant concentrations of LiCl, valproic acid or carbamazepine as well as in control cultures that had not been treated with any drug. After at least 2 weeks of treatment, each of these 3 conventional anti-bipolar drugs increased the uptake significantly at 25 microM inositol, had no effect at 40 microM, and decreased it at 50 microM inositol. Reduction of the drug concentrations by 50% abolished the stimulation of uptake at 25 microM inositol by lithium and valproic acid and reduced that by carbamazepine. These findings may contribute to an understanding of the mechanisms of action for anti-bipolar medication, and explain the controversy in the literature whether or not brain inositol is reduced after chronic administration of lithium.

The efficacy of lithium and carbamazepine in treatment of bipolar affective disorder is well established. Although a number of biochemical effects have been found the exact molecular mechanisms underlying their therapeutic actions have not been elucidated. Nor have the target regions in the brain been located. The objectives of the present investigation were to identify the selective effects and target regions of long-term treatment, with either lithium or carbamazepine, on G-protein subunit expression in rat brain. Effects were measured in hippocampus, hypothalamus, amygdala, frontal cortex, neostriatum, thalamus, raphe nuclei and cerebellum. At the protein level amounts of Galphao decreased significantly (P < 0.01) in neostriatum and Gbeta increased in frontal cortex in response to both drug treatments. At the mRNA level amounts of Galphai1 increased significantly (P < 0.01) in neostriatum. Galphas messenger amounts decreased in frontal cortex and increased in thalamus. These effects were common for both drugs, however, in addition also some differential effects, specific for either of the two drugs, were observed. We conclude frontal cortex and neostriatum are important target regions of long-term treatment with either lithium or carbamazepine and suggest Galphao, Galphas, Galphai1 and Gbeta to be selective target molecules.

We examined the alteration of endogenous mono ADP-ribosylation in the hippocampus of amygdaloid kindled rats to clarify the neurochemical basis of epilepsy. A significant increase of the ADP-ribosylation on the 38 kDa protein was observed in the hippocampal membrane of the kindled rat. Several antiepileptics (phenytoin, phenobarbital, carbamazepine, sodium valproate) significantly decreased the ADP-ribosylation on the 38 kDa protein and effaced the increase in the kindled group. The ADP-ribosylation was largely increased by sodium nitroprusside, a nitric oxide generating compound, in both the kindled and control groups. Carbamazepine could not affect the ADP-ribosylation in the presence of sodium nitroprusside. Twenty amino acids from the N-terminus of the ADP-ribosylated 38 kDa protein were determined by sequential analysis. The sequence was completely identical to that of glyceraldehyde-3-phosphate dehydrogenase. These results indicate that the endogenous mono-ADP-ribosylation which increased in the kindled group and decreased by the antiepileptics might be a specific reaction associated with the mechanisms of epileptogenesis.

Adenosine depresses the excitability of pyramidal neurones in the hippocampus. This effect is lost in calcium-free media and we have now investigated the mechanism of this. Extracellular recordings were made of antidromically and orthodromically evoked population potentials from CA1 region of rat hippocampal slices. It was observed that the activity of adenosine can be restored in the presence of procaine or carbamazepine, known inhibitors of sodium channels. The GABAB agonist baclofen was able to depress potential size but did not restore sensitivity to adenosine. It is concluded that the loss of postsynaptic sensitivity to adenosine in calcium-free solution results from the increased sodium conductances.

In calcium-free media, neurones in the rat hippocampal slice develop bursts of population potentials and lose their sensitivity to adenosine. The present paper reports the unexpected and paradoxical finding that the xanthines theophylline and cyclopentyltheophylline, the latter of which is selective for A1 purine receptors, depressed the excitability of hippocampal pyramidal neurones in calcium-free media. Chelating residual calcium with EGTA reduced excitability which was additive with the xanthine effect, while 100 microM calcium depressed the response to theophylline. The inhibition by xanthines was prevented by adenosine, which had no effect by itself, but was not reproduced or modified by adenosine deaminase. The xanthine effects were also prevented by baclofen and carbamazepine. A common feature of adenosine, baclofen and carbamazepine which may account for their antagonism of the xanthines is the blockade of calcium fluxes. It is proposed that in the presence of low external concentrations of calcium xanthines can reduce excitability by promoting the mobilisation and trans-membrane movement of residual calcium in the medium or neuronal membranes.

Tolerance to carbamazepine's anticonvulsant effects on amygdala kindled seizures occurs contingently, that is, only when carbamazepine is given prior to, but not after the seizure occurs. Biological correlates of contingent tolerance were examined using in situ hybridization and receptor binding techniques for thyrotropin-releasing hormone (TRH) mRNA and TRH receptor binding. Rats were fully kindled and given daily injections of carbamazepine (15 mg/kg, i.p.) either 15 min before (CBZ-before) or after (CBZ-after) amygdala stimulation until the CBZ-before rats became tolerant. Kindled rats were matched so that the two groups had an equal number of seizures and doses of CBZ. Three other groups were also used for comparison: kindled rats that received vehicle injections, and sham-kindled animals treated with either vehicle or CBZ. Rats were sacrificed 4 h after the last seizure or sham stimulation. Both sham-kindled rat groups had barely detectable levels of TRH mRNA. In the CBZ-after (non-tolerant) and vehicle-kindled rats, TRH mRNA levels were increased in the dentate gyrus, pyriform, entorhinal, and perirhinal cortices. In contrast to the other kindled animals, the CBZ-before rats (tolerant) had dramatically diminished TRH mRNA levels bilaterally in the dentate gyrus and pyriform cortex, and ipsilateral to the stimulation in the entorhinal cortex. Decreases in TRH receptor binding were demonstrated autoradiographically in the dentate gyrus and perirhinal cortex in all of the kindled groups with no differences between tolerant and non-tolerant rats.(ABSTRACT TRUNCATED AT 250 WORDS)

Traumatic nerve injury in man can often result in the development of neuropathy. An animal model of neuropathic hyperalgesia is produced by loose ligation of the sciatic nerve in the rat. We have examined the effect of pre-emptive treatment with a number of drugs on the development of hyperalgesia in this model. Animals received clonidine (1 mg.kg-1, s.c.), morphine (5 mg.kg-1, s.c.), (+/-)-baclofen (40 mg.kg-1, s.c.), carbamazepine (50 mg.kg-1, s.c.) or vehicle (4 ml.kg-1, s.c.) 30 min prior to loose ligation or sham-operation. Morphine- and clonidine-treated animals were administered a second dose of drug 6h following surgery. Twenty-six or twenty-nine days following surgery, the ipsilateral (ipsi.) and contralateral (contra.) paw withdrawal thresholds were determined using an Algesymeter. In vehicle-treated animals the ipsilateral paw withdrawal threshold (118 +/- 7 g) was significantly lower (P < 0.0001, paired t-test) than the contralateral (195 +/- 7 g). In contrast, in animals pre-treated with clonidine no significant difference between ipsilateral (200 +/- 9 g) and contralateral (191 +/- 7 g) paw withdrawal thresholds was detected. Morphine pre-treatment was less effective in preventing development of hyperalgesia; however, whilst the ipsilateral (146 +/- 18 g) paw withdrawal threshold tended to be lower than the contralateral (183 +/- 8 g), this was not significant.(ABSTRACT TRUNCATED AT 250 WORDS)

Repeated administration of a subconvulsant dose of a local anesthetic will eventually induce seizures, a phenomenon similar to electrical kindling. We have investigated the effect of repeated lidocaine and cocaine administration on the phosphoinositide (PI) hydrolysis induced by batrachotoxin (BTX), a specific Na channel activator. Rats were injected with cocaine or saline daily for 6 days and PI hydrolysis was assayed in sliced frontal cortex. Cocaine treatment had no effect on BTX-induced PI hydrolysis while in vitro cocaine blocked the BTX effect. In a second experiment, rats received daily injections of lidocaine or saline. After a rat developed at least two seizures, it was sacrificed together with a rat receiving lidocaine injections which had never seized and a rat receiving saline injections. Basal, BTX and ibotenic acid (IBO; a glutamate receptor agonist)-stimulated PI hydrolysis did not differ among the three groups in slices of either hippocampus (HC) or piriform cortex (PC) though IBO-stimulated PI hydrolysis was much greater in the HC than in the PC. Neither in vitro nor in vivo carbamazepine altered the effect of cocaine on BTX-induced PI hydrolysis. These results demonstrate that local anesthetic kindling does not alter PI hydrolysis coupled to Na channel or IBO activation.

Initial studies of the effect of chronic carbamazepine (CBZ) against cocaine-induced seizures indicated that there were genetic differences in both the time course for the development of the anticonvulsant effects of CBZ against cocaine-induced seizures and the persistence of these effects. The present studies were initiated to investigate the time course for the development and persistence of the anticonvulsant effects of chronic CBZ against cocaine seizures in BALB/cByJ, C57Bl/6J and SJL/J mice. The anticonvulsant actions of CBZ were dependent on the duration of CBZ administration, requiring 4-7 days to achieve maximal efficacy. However, once the anticonvulsant effects of CBZ were manifest, the effect persisted for up to 5 days after stopping CBZ treatment depending on the genotype. The levels of CBZ and its active epoxide metabolite were determined in plasma and brain at various time points during and after chronic CBZ treatment. The levels of CBZ and CBZ-10,11-epoxide were substantially reduced over the course of treatment in all three strains, such that the levels of the two compounds in plasma and brain could not account for the decreased susceptibility to cocaine seizures observed following chronic CBZ. These results suggest that the effects of CBZ on cocaine seizures are mediated by relatively long-term changes in one or more biological systems associated with cocaine's convulsant effects.

The effects of chronic carbamazepine (CBZ) on the development and expression of cocaine-kindled seizures and seizures produced by an acute injection of cocaine were evaluated in BALB/cByJ, C57Bl/6J and SJL/J mice. The repeated administration of a subconvulsant dose of cocaine initially resulted in the development of an increased sensitivity to the convulsant effects of cocaine in the three strains. Chronic, dietary carbamazepine attenuated this initial sensitization to cocaine-induced seizures. While the continued administration of cocaine resulted in a relatively permanent sensitization to cocaine-induced seizures among SJL mice, tolerance to cocaine-induced seizures ultimately developed among C57 mice and to a lesser degree among BALB mice. Genetic factors were found to mediate the effects of chronic CBZ on the development of sensitization and/or tolerance to the convulsant effects of cocaine. Among BALB mice, chronic CBZ appears to have eliminated the development of tolerance to cocaine-induced seizures and allowed an underlying sensitization to be manifest. Among SJL mice, however, the sensitization observed following repeated cocaine injections was reduced, but not eliminated. Genetic factors were also found to be associated with the effects of CBZ on seizures induced by the acute administration of cocaine. BALB and C57 mice, but not SJL mice, chronically treated with dietary CBZ were less susceptible to a consulvant dose of cocaine than their corresponding dietary controls for at least 72 h after stopping CBZ administration. In addition, there were genotype-specific lethal effects associated with the concurrent administration of CBZ and cocaine.

Somatostatin (SRIF), a peptide widely distributed in the central nervous system, has been implicated in the genesis of seizure activity in a number of animal models of epilepsy. We examined the effects of the anticonvulsants, phenytoin, carbamazepine and diazepam, on the release of SRIF from dispersed adult rat neuronal cells in short-term culture. Each of these agents caused dose-dependent inhibition of ouabain-stimulated SRIF release in a well-characterized hypothalamic dispersed cell system. We also examined the effects of phenytoin on SRIF release from dispersed rat cortical cells and inhibition of stimulated SRIF secretion was again observed. These findings support the hypothesis that the inhibition of neuronal SRIF release may represent a pharmacological mechanism of action of anticonvulsants.

The influence of diazepam, an agonist, and flumazenil (Ro 15-1788), an antagonist of the benzodiazepine receptor, on repetitive firing of action potentials in cultured spinal neurons and on voltage-dependent Na+ currents in cultured N2A neuroblastoma cells was examined. The effects were compared to those of the antiepileptics phenytoin and carbamazepine and the local anesthetic lidocaine. The whole-cell configuration of the patch-clamp technique was used for potential and current recording. Diazepam (10 microM) or phenytoin (10 microM) reduced the duration of repetitive action potential discharges in 50 or 67% of the spinal neurons, respectively. At a concentration of 100 microM repetitive firing was completely blocked. Flumazenil (100 microM) had no effect. In N2A neuroblastoma cells diazepam, phenytoin, carbamazepine and lidocaine, but not flumazenil, at a concentration of 100 microM reduced the Na+ current to 60-67% of control. At 10 microM no or only a weak depression was seen with any drug. In the presence of diazepam (100 microM) the Na+ channel inactivation curve was shifted in the hyperpolarizing direction by -4.8 +/- 0.5 mV. Phenytoin, carbamazepine and lidocaine (all 100 microM) caused stronger shifts of -17.4 +/- 2.1, -10.6 +/- 0.9 and -17.0 +/- 2.1 mV, respectively. Inhibition of the Na+ current by diazepam increased use-dependently over 9 depolarizing pulses repeated at high frequency (200 Hz), whereas use-dependent effects of the other compounds developed less rapidly. At a low stimulation rate (7 Hz) use-dependent block was pronounced with lidocaine, but weak or absent with diazepam and carbamazepine.(ABSTRACT TRUNCATED AT 250 WORDS)

The piriform cortex, especially its deep anterior part, has been recently suggested to be a crucial epileptogenic site in the rat brain. We investigated the susceptibility of different parts of the piriform cortex to the development of electrical kindling as compared to that of the basolateral amygdala. A locus in the deep cell layer (layer III) of the rostral portion of the posterior piriform cortex (PPC) is described, which is considerably more sensitive to electrical stimulation than adjacent areas of the PPC, including the deep prepiriform cortex or the amygdala. The sensitive locus in the PPC can be readily kindled, and focal seizure thresholds in fully kindled rats are 60-90% lower than respective thresholds in rats kindled from other loci. Treatment of fully kindled rats with antiepileptic drugs diazepam, carbamazepine, phenobarbital, and valproate showed that anticonvulsant effects of these drugs in animals kindled from stimulation of the PPC were comparable to respective effects in animals kindled from stimulation of the basolateral amygdala, although the locus in the PPC tended to be more resistant. The data support the idea that the piriform area may contain the most sensitive neuronal tissue responsible for the generation of seizures during kindling. It remains to be determined if the described locus in the PPC is critical to the kindling process when kindling is induced from other structures within the olfactory-limbic system.

Acute effects of 4 anticonvulsants on hippocampal kindled seizures induced with about 3 Hz electrical stimulations were assessed in cats. The number of stimulating pulses required for the triggering of epileptic afterdischarge (pulse-number threshold, PNT) was used as the indicator for the seizure threshold. Duration of afterdischarge (ADD), ictal and interictal behaviors and serum drug levels were also recorded. PB produced a PNT-increase more prominently than an ADD-decrease with seizure stage regression. PHT produced only a proconvulsive effect by decreasing PNT. CBZ also produced a proconvulsive effect by decreasing PNT at a low dose, and decreased PNT and ADD simultaneously at a high dose. Conversely VPA increased PNT and ADD simultaneously. These results were discussed comparing mainly with a previous study of amygdala-generating seizures.

The effects of carbamazepine (CBZ) treatment on local anesthetic-kindled seizures and lethality were evaluated in different stages of the kindling process and under different methods of CBZ administration. Chronic oral CBZ inhibited the development of both lidocaine- and cocaine-induced seizures, but had little effect on the fully developed local anesthetic seizures. Chronic CBZ also decreased the incidence of seizure-related mortality in the cocaine-injected rats. Acute CBZ over a range of doses (15-50 mg/kg) had no effect on completed lidocaine-kindled or acute cocaine-induced seizures. Repeated i.p. injection of CBZ (15 mg/kg) also was without effect on the development of lidocaine- or cocaine-kindled seizures. The differential effects of CBZ depending upon stage of seizure development suggest that distinct mechanisms underlie the development versus maintenance of local anesthetic-kindled seizures. The effectiveness of chronic but not repeated, intermittent injections of CBZ suggests that different biochemical consequences result from the different treatment regimens. The possible utility of chronic CBZ in preventing the development of toxic side effects in human cocaine users is suggested by these data, but remains to be directly evaluated.

Evoked potential (EP) analysis of the somatomotor pathway in cats revealed that the temporal lobe anticonvulsant carbamazepine suppresses the thalamic relay, whereas the petit mal antiepileptic ethosuximide acts on the cortex. Moreover, a history of temporal lobe epilepsy (amygdala kindling) maximized thalamic response to carbamazepine, especially during sleep states vulnerable to generalized kindled seizures. Ethosuximide accentuated cortical response during sleep and awakening states vulnerable to generalized spike-wave complexes, regardless of a history of petit mal seizures (systemic penicillin epilepsy). The findings provide a neural basis for differential drug effects on generalized temporal lobe and petit mal epilepsy and further suggest that a chronic seizure disorder can create a predisposition for antiepileptic drug sensitivity.

Bath application of therapeutic concentrations of the anticonvulsant carbamazepine suppressed penicillin-induced synchronized afterdischarging in immature rat CA3 hippocampal pyramidal cells. Afterdischarging was completely abolished in all preparations at a concentration of 30 microM (IC50 = 8.5 +/- 1.4 microM; mean +/- S.E.M.). The duration of the preceding epileptiform burst was not altered at this concentration and was diminished by only 24.4 +/- 1.2% at a supratherapeutic concentration of 100 microM. These results suggest that a carbamazepine-sensitive neurophysiological mechanism distinct from those responsible for epileptiform burst generation plays a key role in the generation of afterdischarges in developing hippocampus.

Spontaneous epileptiform activity (SEA) develops in area CA1 of hippocampal slices, when the Ca2+ concentration in the perfusate is lowered to 0.2 mM, at which level evoked chemical synaptic transmission is blocked. We investigated the effects of different anticonvulsants on this autonomous activity, in order to determine whether the antiepileptic effect can be ascribed to an influence on neuronal excitability. Carbamazepine was the most effective to block SEA at concentrations of 1-15 microM. Phenobarbital and phenytoin depressed SEA at concentrations of 25 microM. Valproate was effective at concentrations of 2-5 mM. Midazolam, a water-soluble benzo-diazepine agonist and the N-methyl-D-aspartate antagonists, DL-alpha-aminoadipic acid and 2-amino-7-phosphonoheptanoic acid were ineffective in blocking SEA suggesting that they exert their antiepileptic action by interference with synaptic mechanisms.

The possibility that excitatory and inhibitory inputs to neurons can affect the generation and degeneration of neuroarchitecture was examined in hippocampal pyramidal neurons in isolated cell culture. Dendritic outgrowth and cell survival were directly monitored in neurons exposed to: the excitatory neurotransmitter glutamate, the inhibitory transmitter GABA, anticonvulsants or combinations of these agents. Glutamate caused a graded series of changes in pyramidal neuron cytoarchitecture: a selective inhibition in dendritic outgrowth and dendritic pruning was observed with subtoxic levels of glutamate while cell death was induced by higher levels. Low levels of GABA alone or in combination with diazepam, carbamazepine, phenobarbital or phenytoin were without effect on dendrite outgrowth while higher levels caused moderate reductions in outgrowth. Neither GABA nor the anticonvulsants affected cell survival. GABA plus diazepam, phenobarbital, carbamazepine and phenytoin each significantly reduced the dendritic regression and cell death normally caused by glutamate. Elevation of extracellular K+ to 50 mM caused dendritic regression and 100 mM K+ caused cell death; these effects were greatly reduced by GABA and anticonvulsants. The calcium channel blocker Co2+ prevented the dendritic regression and cell death caused by both glutamate and K+ indicating that calcium influx was required for the neuroarchitectural responses. Taken together, these results demonstrate that neurotransmitters and neuromodulatory drugs can have direct and interactive effects on both neurite outgrowth and cell survival. Such neurotransmitter actions may play roles in both the formation and degeneration of the neuronal circuits in which they participate in information coding.