Eric W. Lothman

Kainic acid induced limbic seizures: metabolic, behavioral, electroencephalographic and neuropathological correlates

Increasing amounts (0.3–12 mg/kg) of kainic acid (KA) were given intravenously to rats and behavioral, electrographic, 2-deoxyglucose autoradiographic and neuropathologic responses were studied. A dose of 12 mg/kg kainic acid caused a stereo-typed sequence of staring spells, wet dog shakes, automatisms — mild limbic convulsions and severe limbic convulsions (rearing, bilateral upper extremity clonus, falling and salivation) that developed over 1–2 h. Smaller doses showed different thresholds for these behavioral phenomena but a similar time course of development. Electrographic seizures in limbic areas were similar to afterdischarges produced with electrical stimulation and their threshold was 4 mg/kg. The earliest electrical changes occurred in the hippocampus and were accompanied by staring spells. Subsequently seizures were synchronized in limbic centers during which mild convulsive activity was seen. Finally electrical seizures appeared in both limbic and surface leads, coincident with severe convulsions. Quantitative deoxyglucose autoradiography showed that the hippocampul CA3 region was most sensitive to metabolic changes. Mild limbic convulsions were associated with increased glucose utilization in the hippocampus, subiculum, pyriform and entorhinal cortices, septum and amygdala. Severe limbic convulsions led to even larger changes in these areas as well as the substantia nigra and part of the thalamus. Neuronal damage was seen without systemic metabolic derangements with a threshold of 7 mg/kg. The regions of neuropathology overlapped with areas of intense seizure activity. n nThese data show that systemic kainic acid preferentially activates seizures in the limbic system, particularly the hippocampus, and that different behavioral concomitants of limbic seizures depend on specific patterns of activation of limbic and extra limbic circuits. A scheme for the anatomic spread of seizures in limbic and non-limbic structures is proposed. A state of ‘limbic status’ underlies the neurotoxicity of kainic acid.

Functional anatomy of limbic seizures: focal discharges from medial entorhinal cortex in rat.

Focal seizure discharges were induced in the ventral aspect of the medial entorhinal cortex of awake, freely moving rats, either with cannula injections of penicillin or picrotoxin (0.02 microliters every 10-15 min) or by repetitive tetanic electrical stimulation. [14C]Deoxyglucose autoradiography (DG) was performed when animals were in a steady-state with respect to electrographic discharges and/or behavioral changes. During simple interictal spikes behavior remained normal and DG labeling was increased only in the entorhinal focus and stratum moleculare of the ventral dentate gyrus. With complex spikes and short seizures animals exhibited staring, decreased responsiveness, and occasional wet dog shakes. DG labeling was increased in all layers of the dentate gyrus, Ammons horn (ipsilateral greater than contralateral) and, to a lesser degree, in ipsilateral amygdala, and the accumbens-ventral pallidum area. During strong seizures, rearing and forelimb clonus occurred and metabolism was strongly activated bilaterally in the hippocampal formation, amygdala, accumbens, substantia nigra, and the anterior and periventricular thalamic nuclei. These studies indicate that the dentate gyrus initially restricts the entry of seizures from entorhinal cortex into the rest of hippocampus. As this is overcome there is rapid bilateral spread through the hippocampal formation with passive interruption of normal behavior. With prolonged seizure discharges there is further capture of amygdala and subcortical extrapyramidal and thalamic nuclei associated with behavioral convulsions.

Cellular and synaptic basis of kainic acid-induced hippocampal epileptiform activity

The effects of kainic acid (KA) were studied using extracellular and intracellular recordings in the hippocampal slice preparation. In sufficient concentrations, KA led to a loss of all evoked responses. However, the amount of drug needed for this varied according to anatomic region. CA3 was more sensitive (1 microM) than CA1 or the dentate gyrus (10 microM). These results can be understood in terms of a profound and long-lasting depolarization of neurons. Lower concentrations of KA (0.05-0.1 microM) did not change the resting membrane potential or input resistance of hippocampal pyramidal cells but produced spontaneous epileptiform activity which originated in CA3 and propagated to CA1. Epileptiform discharges were not present in the dentate gyrus. Coincident with the induction of paroxysms, the following changes were observed: (1) an increase in the excitability of CA3 and CA1 pyramidal cells as measured by a left shift in the input-output curves of evoked responses and a lowered threshold stimulus intensity necessary for activation of action potentials in single neurons; (2) augmentation and synchronization of bursting in pyramidal cells; and (3) prolonged EPSPs without an increase in their amplitude. These findings indicate that multiple changes, involving both the properties of single neurons and synaptic connections, are involved in the development of hippocampal paroxysms and that CA3 and CA1 have different roles in the generation of these discharges.

Kainic acid‐induced limbic seizures Electrophysiologic studies

surface and depth electroencephalograms (EEGs) were studied after intravenous injections of kainic acid (KA). High frequency oscillations and spikes appeared in the hippocampus at a dose (1 mg per kilogram) that did not affect other structures. Higher doses ≤ 4 mg per kilogram) led to electrical seizures in limbic structures, similar to those in temporal lobe epilepsy. In hippocampal slices maintained in vitro, 0.1 to 1.0 μM KA produced spontaneous epileptiform spikes, originating in CA1, and increased evoked potentials. Systemic KA is a potent means of inducing limbic seizures with a primary action in the hippocampus. We propose that this selective activation arises when KA augments excitatory glutamatergic synapses in critical epileptogenic areas, such as the CA, region of the hippocampus.

Blood-brain barrier changes with kainic acid-induced limbic seizures

Rats were treated with kainic acid (KA) i.v. to produce increasingly severe limbic seizures that were monitored with a behavioral rating scale. At various times after the induction of seizures, the animals; blood-brain barriers (B-BB) were studied with alpha-[14C]aminoisobutyric acid ([14C]AIBA) autoradiography. Using optical density ratios, a coefficient was devised to assess the functional integrity of the B-BB in discrete anatomic regions and to quantitatively compare these measurements among different groups of experimental animals. In animals that exhibited only mild seizures, the B-BB was not different from controls. Animals with severe limbic seizures, however, showed alterations. For as long as 2 h after delivery of KA, the B-BB appeared normal; from 2 to 24 h, the permeability to [14C]AIBA was markedly increased throughout the brain, especially in limbic regions; from 24 h to 7 days the B-BB returned to normal except for a small residual change in limbic structures. These findings were confirmed with Evans blue dye studies of the B-BB. A correlation between focal accentuation of B-BB alterations and neuropathologic changes was found. These experiments indicted that recurrent limbic seizures may lead to a breakdown in the B-BB independent of systemic metabolic derangements. Marked focal metabolic and electrical changes, however, occurred in several limbic structures several hours before the blood-brain barrier was altered.

Self‐sustaining limbic status epilepticus.: I. Acute and chronic cerebral metabolic studies: Limbic hypermetabolism and neocortical hypometabolism

Regional cerebral glucose utilization (RCGU) increases during seizures whereas hypometabolism occurs in postictal and interictal states. Recently, we developed a model of nonconvulsive, self-sustaining limbic status epilepticus (SSLSE) in which electrographic seizures persist 12 to 24 hours after 90 minutes of continuous hippocampal stimulation. The present studies define the functional anatomy of SSLSE and the states thereafter. RCGU was studied by 2-deoxyglucose autoradiography in (1) a group of rats acutely (1 hour after induction) during SSLSE, and (2) two groups of rats chronically (1 week or 1 month) after SSLSE. RCGU measurements in these groups were compared with those obtained in naive and electrode-implanted control rats. In the acute group, there were bilateral increases in RCGU in the hippocampus, retrohippocampal structures, and associated limbic and subcortical nonlimbic regions; hypometabolism was found in several neocortical structures. Chronically, RCGU was elevated in certain limbic areas at 7 days but returned to control values at 30 days. On the basis of our findings, we postulate a feedback network involving the hippocampus and neighboring parahippocampal structures (the hippocampal-parahippocampal loop) as a critical substrate for establishing limbic system status epilepticus. In addition, the results indicate that metabolic responses can persist long after the cessation of status epilepticus and that both increases and decreases in RCGU can be seen in acute limbic status epilepticus.Regional cerebral glucose utilization (RCGU) increases during seizures whereas hypometabolism occurs in postictal and interictal states. Recently, we developed a model of nonconvulsive, self-sustaining limbic status epilepticus (SSLSE) in which electrographic seizures persist 12 to 24 hours after 90 minutes of continuous hippocampal stimulation. The present studies define the functional anatomy of SSLSE and the states thereafter. RCGU was studied by 2-deoxyglucose autoradiography in (1) a group of rats acutely (1 hour after induction) during SSLSE, and (2) two groups of rats chronically (1 week or 1 month) after SSLSE. RCGU measurements in these groups were compared with those obtained in naive and electrode- implanted control rats. In the acute group, there were bilateral increases in RCGU in the hippocampus, retrohippocampal structures, and associated limbic and subcortical nonlimbic regions; hypometabolism was found in several neocortical structures. Chronically, RCGU was elevated in certain limbic areas at 7 days but returned to control values at 30 days. On the basis of our findings, we postulate a feedback network involving the hippocampus and neighboring parahippocampal structures (the hippocampal-parahippocampal “loop”) as a critical substrate for establishing limbic system status epilepticus. In addition, the results indicate that metabolic responses can persist long after the cessation of status epilepticus and that both increases and decreases in RCGU can be seen in acute limbic status epilepticus.

Effect of anticonvulsant drugs on kainic acid-induced epileptiform activity

Abstract Five antiepileptic drugs were tested for their ability to block limbic seizures induced by systemic injection of kainic acid and to suppress kainic acid-induced epileptiform discharges in incubated hippocampal slices. Phenytoin, phenobarbital, ethosuximide, and valproic acid inhibited epileptiform discharges in hippocampal slices at concentrations approximating their respective clinically effective anticon-vulsant blood concentration in humans, and diazepam had a similar action at significantly higher concentrations. At these concentrations none of the drugs blocked evoked orthodromic responses of monosynaptic excitatory connections in the hippocampal slices. In contrast, none of the drugs, at therapeutic doses, prevented kainic acid-induced seizure discharges in the hippocampus, in situ. Phenobarbital and diazepam were effective at higher concentrations. These data demonstrate that antiepileptic drugs do not have identical effects on seizure discharges in one type of brain tissue in situ and in vitro even when both are elicited by the same convulsant agent. The results also indicate that limbic seizures induced by kainic acid in vivo, like many cases of complex partial seizures in humans, are highly resistant to conventional anticonvulsant drug therapy.

Loss of GABAA receptors during partial status epilepticus.

There is a critical role for loss of GABA-mediated inhibition in the CA1 region of the hippocampus in the emergence of partial status epilepticus (SE) in experimental animals. We demonstrated loss of GABA-mediated inhibition in the CA1 region of the hippocampus by the paired-pulse method in several different experimental models of partial SE. The cellular mechanism underlying loss of GABA-mediated inhibition during SE remains unclear. This study investigates the effect of experimental SE on rat forebrain GABA receptor. We studied the effect of a combination of lithium and pilocarpine and lithium alone on GABA-mediated inhibition in the CA1 region of the hippocampus by paired-pulse inhibition technique.2 Methods. SE was induced by a method described by Honchar et aL3 Briefly, adult male Sprague-Dawley rats weighing 175 to 250 grams were each given 3 mEqlkg lithium chloride as a single intraperitoneal injection followed 20 hours later by 50 mglkg pilocarpine. Animals started having seizures 13 to 46 minutes after pilocarpine injection, and these seizures continued for 30 to 47 minutes before termination. The seizures were terminated by ether anesthesia and the forebrains were rapidly removed from the skulls, placed in a chilled sucrose buffer, and homogenized. The sucrose buffer contained 0.32 M sucrose, 100 mM PIPES, 10 mM EGTA, and 20 mM EDTA. The homogenate was spun at 5,000 g for 10 minutes in a Beckman 52-21 centrifuge. The supernatant was then spun at 18,000 g for 20 minutes. The resultant pellet (P2 fraction) was osmotically shocked in a lox volume of distilled water. The synaptic plasma membrane fractions derived by this procedure were suspended in 50 mM Tris citrate buffer (pH 7.4) and washed three times in Tris buffer. After three washes, the membranes were stored at -70 C. Fifty p1 of 3H-muscimol was added to 100 pl membranes and 850 p1 50 mM Tris citrate to reach final muscimol concentrations varying from 3 nM to 100 nM. Based on preliminary experiments, nonspecific binding was obtained in the presence of 100 mM GABA. The membranes were incubated with radioligand for 20 minutes and then reaction was terminated by rapid filtration under vacuum on Whatman GF/B filters. Filters were washed three times, then digested in 3 ml of Insta-Gel Packard scintillation fluid. Radioactivity was measured by conventional scintillation counting. Binding data is presented as specific binding, calculated by subtracting binding in presence of GABA from total binding. Paired-pulse inhibition technique was utilized to measure GABA-mediated inhibition in the CA1 region of the hippocampus of urethane (1.3 to 1.5 gkgbanesthetized animals. We have previously described* this procedure for studying paired-pulse inhibition in detail. In one set of experiments, naive animals were studied; in the second set of experiments, rats were given 3 mEqlkg lithium chloride intraperitoneally 20 hours prior to urethane anesthesia. In the lithium-pretreated animals, inhibition was measured, 50 m g k g pilocarpine was administered, and 15 minutes later paired-pulse inhibition was measured again. Results. All animals injected with lithium and pilocarpine experienced seizures for at least 30 minutes. No seizures were observed in animals treated with lithium alone or in naive animals. A specific high-affinity saturable binding site for 3H-muscimol was defined in rat forebrain synaptic plasma membranes. In three separate experiments, a saturation curve for specific 3H-muscimol binding was generated by measuring total and nonspecific binding for free 3H-muscimol concentration ranging from 6 nM to 100 nM. Specific binding was approximately 50% of total binding. For each experiment, synaptic plasma membranes from five naive rats and from five animals undergoing SE were pooled; for each 3H-muscimol concentration, three separate measurements of total and nonspecific binding were made. Each measurement was repeated once. Synaptic plasma membranes from naive rats and from those undergoing SE were 0 8 -

Learning deficits after lesions of dentate gyrus granule cells

The effect of destroying granule cells in the dentate gyrus on learning was examined with a behavioral testing protocol. These neurons were destroyed by microinjections of the selective neurotoxin colchicine in the hippocampal formation of rats. After a 30-day recovery period, the animals were trained in an operant chamber with an appetitive conditioning paradigm. The learning abilities of the animals with lesions were compared with two control groups--naive, unoperated rats and those with control injections of saline. The basic task required the animal to discriminate between two spatially separate visual stimuli which represented positive and negative cues. Testing and training was separated into four progressively more difficult phases with various stimulus schedules, contingencies of reinforcement, and stimulus positions. Colchicine-treated animals demonstrated significantly poorer performance than naive animals and those receiving saline control injections. None of the colchicine-treated animals achieved criterion performance in the stimulus position reversal paradigm, and half had difficulty with variable ratio schedules of reinforcement. Our experiments suggested that granule cells in the dentate gyrus played a pivotal role in certain learning tasks.

Self‐sustaining limbic status epilepticus.: II. Role of hippocampal commissures in metabolic responses

In prior work, we developed a model of self-sustaining limbic status epilepticus (SSLSE) induced by continuous hippocampal stimulation (CHS). Previous electrographic studies showed that SSLSE was synchronized between the cerebral hemispheres. On the basis of this initial work, we postulated that hippocampal commissures were critical for the initiation and maintenance of SSLSE. In the current experiments, we tested this hypothesis by applying CHS in animals with (CMX) or without (— CMX) hippocampal commissurotomies. In the — CMX group, electrographic SSLSE was synchronized between the stimulated and contralateral sides. In the CMX group, SSLSE developed only on the stimulated sides. Regional cerebral glucose utilization (RCGU) was also studied acutely (1 hour) after CHS using 2-deoxyglucose autoradiography. In the — CMX group, there was symmetrically increased RCGU in the hippocampus, retrohippocampal structures, and associated limbic and subcortical nonlimbic regions. In the CMX group, a similar pattern was found, but confined to the side of stimulation. CMX alone did not change RCGU values from those in control (— CMX, nonstimulated) brain in any of the regions studied. Areas of bilateral neocortical hypometabolism were found in both (CMX and —CMX) SSLSE groups. These results lead to rejection of the hypothesis that hippocampal commissures play an essential role in the initiation and maintenance of SSLSE. Instead, a feedback circuit involving the hippocampus and its adjacent structures seems to be the critical anatomic substrate for SSLSE. The presence of neocortical hypometabolism after CMX indicates that the structures other than the hippocampal commissure (eg, the thalamus or other forebrain commissures) mediate this effect.