Electroencephalographic and pathophysiologic features of rat parenteral penicillin epilepsy

Acute and chronic pharmacological models of generalized absence seizures

This article reviews the contribution of pharmacologically induced acute and chronic animal models to our understanding of epilepsies featuring non-convulsive generalized seizures, the typical and atypical absence seizures. Typical absences comprise about 5% of all epilepsies regardless of age and the atypical ones are even more common. Although absence epilepsy was thought to be relatively benign, children with childhood epilepsy (CAE) turn out to have a high rate of pretreatment attention deficits that persist despite seizure freedom. The phenomenon of the absence seizure has long attracted research interest because of the clear temporal relationship of the conspicuous EEG rhythm of 3 Hz generalized spike and wave discharges (GSWD) and the parallel transient "loss of consciousness" characterizing these seizures which is time-locked with the GSWD. Indeed, clinical epileptologists, basic scientists and neurophysiologists have long recognized in GSWD a unique electrographic and behavioral marker of the genetic predisposition to most types of epilepsy. Interestingly, the subject is still controversial since it has recently been proposed that both classification terms of CAE currently in use: idiopathic and primary generalized, be abandoned - a point of debate. Both issues - underlying mechanisms and focal origin of absence seizures - may be further enlightened by observations in valid animal models.

An in vivo technique for investigating electrophysiological effects of centrally administered drugs on single neurons and network behaviour

Single neuronal juxtacellular recording with simultaneous cortical electroencephalogram (EEG) in whole-animal preparations in vivo has allowed the study of the behaviour of individual neurons in relation to whole brain activity. Data on single neuron firing, neural synchrony, network behaviour and their responses to pharmacological agents can be obtained with dual recordings. However, pharmacological effects on cellular and network activity during paired single-unit recordings have not been possible due to the difficulties in maintaining recordings of two cells for a prolonged period. Here, we describe a method of maintaining stable dual cell juxtacellular recordings from distinct brain regions, allowing the assessment of single unit activity before, during and after the intracerebroventricular (ICV) injection of drugs. Data collection using this technique allows correlation both between the two cells and with whole-brain EEG, and their responses to pharmacological interventions. This is particularly useful for the investigation of the effects of anti-epileptic drugs on animal models of epilepsy, where single unit activity of two cells from distinct regions can be correlated with each other and with whole-brain activity during pre-ictal, ictal and interictal states. We also describe standardised analytical methods of quantifying cell firing patterns, the rhythmicity of individual neurons and the synchronicity of firing between two neurons in ictal and interictal periods and their responses to drug exposure.

Animal models

Epilepsy accounts for a significant portion of the dis-ease burden worldwide. Research in this field is fundamental and mandatory. Animal models have played, and still play, a substantial role in understanding the patho-physiology and treatment of human epilepsies. A large number and variety of approaches are available, and they have been applied to many animals. In this chapter the in vitro and in vivo animal models are discussed,with major emphasis on the in vivo studies. Models have used phylogenetically different animals - from worms to monkeys. Our attention has been dedicated mainly to rodents.In clinical practice, developmental aspects of epilepsy often differ from those in adults. Animal models have often helped to clarify these differences. In this chapter, developmental aspects have been emphasized.Electrical stimulation and chemical-induced models of seizures have been described first, as they represent the oldest and most common models. Among these models, kindling raised great interest, especially for the study of the epileptogenesis. Acquired focal models mimic seizures and occasionally epilepsies secondary to abnormal cortical development, hypoxia, trauma, and hemorrhage.Better knowledge of epileptic syndromes will help to create new animal models. To date, absence epilepsy is one of the most common and (often) benign forms of epilepsy. There are several models, including acute pharmacological models (PTZ, penicillin, THIP, GBL) and chronic models (GAERS, WAG/Rij). Although atypical absence seizures are less benign, thus needing more investigation, only two models are so far available (AY-9944,MAM-AY). Infantile spasms are an early childhood encephalopathy that is usually associated with a poor out-come. The investigation of this syndrome in animal models is recent and fascinating. Different approaches have been used including genetic (Down syndrome,ARX mutation) and acquired (multiple hit, TTX, CRH,betamethasone-NMDA) models.An entire section has been dedicated to genetic models, from the older models obtained with spontaneous mutations (GEPRs) to the new engineered knockout, knocking, and transgenic models. Some of these models have been created based on recently recognized patho-genesis such as benign familial neonatal epilepsy, early infantile encephalopathy with suppression bursts, severe myoclonic epilepsy of infancy, the tuberous sclerosis model, and the progressive myoclonic epilepsy. The contribution of animal models to epilepsy re-search is unquestionable. The development of further strategies is necessary to find novel strategies to cure epileptic patients, and optimistically to allow scientists first and clinicians subsequently to prevent epilepsy and its consequences.

7-12 Hz high-voltage rhythmic spike discharges in rats evaluated by antiepileptic drugs and flicker stimulation

Paroxysmal 7- to 12-Hz high-voltage rhythmic spike (HVRS) or spike-wave discharges often appear in several particular strains of rats. However, functional hypotheses of these 7-12 Hz high-voltage cortical oscillations (absence seizure vs. idling mu rhythm) are inconclusive. The mu rhythm can be provoked by flicker stimulation (FS) in most people, but FS is less effective at eliciting absence epileptic activity. Therefore FS and antiepileptic drugs were used to verify the role of HVRS activity in Long-Evans rats with spontaneous HVRS discharges and Wistar rats without spontaneous HVRS discharges. The occurrence of HVRS discharges was significantly reduced by antiabsence drugs (ethosuximide, valproic acid, and diazepam) in dose-dependent manners, but high-dose carbamazepine displayed little effect. On the other hand, oscillation frequencies and durations of spontaneous HVRS discharges were not altered by FS. Under asynchronous brain activity, many FSs (>60%) elicited small-amplitude mu-rhythm-like activity in the barrel cortex concomitant with FS-related rhythms in the occipital cortex and resulted in significant augmentation of 7-12 Hz power in the parietal region. Furthermore, a large portion of FSs (>60%) revealed increase of 7-12 Hz power of the parietal cortex after ethosuximide administration (100 mg/kg ip) in Long-Evans rats. Similar FS-elicited phenomena also appeared in Wistar rats. Characteristics of FS-elicited mu-rhythm-like activities were consistent with those observed in humans, and they remarkably differed from those of spontaneous HVRS discharges. These results support the hypothesis that HVRS activity in Long-Evans rats may be an absence-like seizure activity rather than the mu rhythm.

Is spontaneous high-voltage rhythmic spike discharge in Long Evans rats an absence-like seizure activity?

A distinct high-voltage rhythmic spike (HVRS) discharge characterized by a barrage of negative spikes oscillating at 5-12 Hz was observed in chronically implanted Long Evans rats. Spontaneous HVRS discharges were exhibited in 90% of 40 Long Evans rats and occurred during sudden arrest of ongoing behavior (immobility) with occasional facial/whisker twitching. However, the function of HVRS discharges in Long Evans rats remains inconclusive to date and has been associated with alpha tremor/mu rhythm, attentive mu wave, and absence seizure. To elucidate the function of HVRS discharges in Long Evans rats, several experiments were performed. In a 6-h recording session (12:00-18:00), HVRS activities primarily occurred in several specific vigilance states, being particularly abundant in a short-lasting period before vigilance changes. Several characteristics, such as durations, oscillatory frequencies, and interspike intervals (ISIs) of HVRS discharges, were altered during wake-sleep states. Oscillatory frequencies were negatively correlated with durations of HVRS segments. In addition, ISIs of a HVRS episode exhibited a crescendo-decrescendo pattern. These variable ISIs could explain why a negative correlation was found between oscillatory frequencies and durations of HVRS episodes. Moreover, HVRS discharges were demonstrated to have widespread and near-synchronous distribution to bilateral cortical areas. In addition, innocuous electrical stimuli were unable to stop ongoing HVRS discharges. By contrast, noxious stimuli elicited behavioral arousal and immediately terminated most HVRS discharges. Cortical-evoked potentials in response to mild electrical stimulation under HVRS discharges were different from those under waking state but resemble those under slow-wave sleep with a smaller magnitude. Moreover, the temporal and spectral characteristics of spontaneous HVRS activities were analogous to those of seizure activities induced by penicillin and pentylenetetrazol. The incidence of spontaneous HVRS discharges was significantly decreased by ethosuximide administration. Based on these results, HVRS discharge might not be associated with a voluntary mu-rhythm behavior, instead it behaves as an absence-like seizure activity. These results were also collaborated using other genetic absence-seizure rats, such as WAG/Rij and GAERS rats. Possible mechanisms for the generation and termination of paroxysmal HVRS discharges are also discussed.

Effect of neurotoxin DSP4 on EEG power spectra in the rat acute model of epilepsy

The effect of the adrenergic neurotoxin N-(chloroethyl)-N-ethyl-2-bromobenzylamine (DSP4) on electroencephalographic (EEG) activity was studied in the model of epilepsy induced by systemic application of penicillin (1,000,000 IU/kg, i.p). DSP4 (50 mg/kg, i.p.) was administrated to male Wistar rats, while the control animals were rats from the same litters. EEG activity was recorded in acute and chronic experiments 3 or 4 weeks after DSP4 treatment, before and after penicillin administration. Occasional locus coeruleus (LC) stimulation served as an electrophysiological test of DSP4 toxic effect. EEG power spectra in DSP4 treated animals showed a tendency to be greater in lower frequency bands than in controls before penicillin administration; there was almost no effect of electrical LC stimulation, regardless on penicillin treatment. In the model of epilepsy, the mean total EEG power spectra were greater in the period of 135-330 min after penicillin administration, as well as during 345-540 min, in DSP4 treated animals as compared to the controls. It seems that neurotoxin DSP4 is an optimal tool for studying the removal of LC influence in the acute model of epilepsy. It is also suggested that norepinephrine (NE) may have a modulatory role in the systemic penicillin epilepsy.

The effect of cortical lesion on systemic penicillin epilepsy in rats

There is a certain recovery of function following brain damage, due to neuronal plasticity. The experiments were performed in order to investigate the effects of cortical lesion on seizural activity in rats induced by systemic application of penicillin. The sensorimotor cortex was unilaterally removed in the lesioned animals, while the control animals were only sham operated or non-operated (before implantation of the electrodes). Seizural activity was recorded by means of electroencephalograms before and after penicillin treatment (1,000,000 I.U./kg, i.p). Testing of penicillin started at least 30 days after cortical lesion. Seizural activity was characterized by spike and wave complexes accompanied by vigilance reduction and sometimes by mild myoclonic jerks in both control and lesioned animals. The early period (about 2 h after penicillin administration) with appearance of the spike-wave discharges with relative increase of the mean total electroencephalogram powers as well as the succeeding period 2.5-5.5 h after penicillin administration) with maximum number of spike-wave discharges did not differ in the electroencephalogram of the control and lesioned animals. The late period of penicillin effect (from 6-11 h after penicillin administration) with frequent spike-wave discharges and still large mean total electroencephalogram powers was observed only in lesioned animals. It is concluded that a cortical lesion destabilizes the brain function in the rat model of epilepsy induced by parenteral administration of penicillin.

Pharmacological models of generalized absence seizures in rodents

A number of animal models of generalized absence seizures in rodents are described. These include absence seizures induced by gamma-hydroxybutyrate (GHB), low dose pentylenetetrazole, penicillin, THIP, and AY-9944. All of these models share behavioral and EEG similarity to human absence seizures and show pharmacologic specificity for antiabsence drugs such as ethosuximide and trimethadione. Moreover, the absence seizures induced by these agents are exacerbated by GABAergic agonists, a property unique to experimental absence seizures. These models are predictable, reproducible, and easy to standardize. They are useful both in studying mechanisms of pathogenesis of absence seizures as well as in screening for antiabsence activity of potential antiepileptic drugs.

Aggravation of penicillin-induced epilepsy in rats with locus ceruleus lesions

The rate and pattern of development of seizures induced by penicillin injected intraperitoneally were determined in rats that had been depleted of brain norepinephrine (NE) by bilateral injections of the neurotoxin 6-OH dopamine into the locus ceruleus. Behavioral observations and scalp electrographic recordings were made after injection and the efficacy of NE depletion was determined by high performance liquid chromatography measurement of cortical levels of NE and its metabolites. We found that in comparison to sham-operated control rats, NE-depleted rats had a significantly shorter latency to first observable myoclonic jerk, the first epileptic discharge, the first convulsion with sustained epileptic discharges, and a longer duration of convulsions. We observed a similar electrographic pattern of multifocal spikes with bilateral synchrony in both groups. However, more of the control rats (six of 12) had convulsions as compared to the lesioned rats (four of 12). These findings are consistent with previous evidence that depletion of neocortical NE facilitates the development of epileptiform activity in the CNS; however, a convulsive state was not induced by NE depletion.

Increased gamma-hydroxybutyric acid receptors in thalamus of a genetic animal model of petit mal epilepsy

The distribution and kinetics of specific binding sites for gamma-hydroxybutyrate (GHB), a naturally occurring compound known to produce absence-like seizures, was studied in the brains of Wistar rats with spontaneous, bilaterally synchronous spike wave discharges (SWDs), a model of petit mal epilepsy, and non-epileptic controls using [3H]GHB autoradiography. [3H]GHB receptor binding was increased 40-60% in lateral thalamic nuclei of the epileptic animals. Kinetic analysis showed that the increase in the binding was due to an increase in density of low affinity GHB binding sites in the epileptic animals. Given the ability of GHB to produce petit mal-like seizures when administered to animals, and the fact that the SWDs in the Wistar rat model seem to emanate from lateral thalamus, these data raise the possibility that GHB-mediated mechanisms may play a role in the pathogenesis of petit mal seizures.

gamma-Hydroxybutyrate model of generalized absence seizures: further characterization and comparison with other absence models

gamma-Hydroxybutyrate (GHB) produces a constellation of EEG and behavioral events that respond selectively to antiabsence antiepileptic drugs. The GHB-induced seizure was quantitated in the presence of three other absence seizure models: pentylenetetrazole, systemic penicillin, and the flash evoked afterdischarge (FEAD). Penicillin and pentylenetetrazole produced a significant prolongation of GHB-induced seizure in a dose-dependent fashion. This potentiation of GHB seizure was observed when these compounds were given either before administration of gamma-butyrolactone (GBL), the prodrug of GHB, or at the onset of GHB-induced seizure. Photic stimulation given in a manner to produce FEAD also resulted in a significant prolongation of GHB-induced seizure. All of these maneuvers lowered the threshold to GHB seizure, but none interfered with the brain kinetics of GHB in the animals treated with GBL. Ethosuximide pretreatment significantly shortened the GHB seizure and overcame the potentiating effect of penicillin and pentylenetetrazole in this model. These data confirm the GHB-treated animal as a model of generalized absence seizure. The GHB model meets appropriate criteria for an absence seizure model and compares favorably with other models of absence currently in use.

Attenuation of penicillin models of epilepsy by nonsteroidal anti-inflammatory drugs

Prostaglandin and thromboxane levels increase in mammalian brain during seizures, but whether a metabolite of arachidonic acid has a role in induction, or is merely a by-product, of seizures is still not clear. It has been shown that nonsteroidal anti-inflammatory drugs block arachidonic acid metabolism through inhibition of the enzyme, cyclooxygenase. In the present study, the antiepileptic action of nonsteroidal antiinflammatory drugs was investigated utilizing penicillin-induced focal (250 IU) and primary generalized (2.5 million IU kg-1) models of epilepsy. The effects of the drugs on the electrocortical and motor manifestations of both models were examined. As a group, the nonsteroidal anti-inflammatory drugs significantly affected the central nervous system excitation produced by penicillin. However, individual drugs affected different parameters of the excitation. Paracetamol, the most effective, delayed and/or blocked spikes and seizures in both models. Ibuprofen was more effective against the primary generalized model. Mefenamic acid decreased the number of seizures and indomethacin decreased the voltage output in both models. The present study, together with earlier papers, suggests that an arachidonic acid metabolite plays a role in induction of several animal models of epilepsy.

Interictal EEG pattern in rabbit penicillin epilepsy

The effects of Na-penicillin G (1,000,000 I.U./kg i.v.) were studied in 28 rabbits implanted with surface and deep electrodes in the medial (mT) and lateral (lT) nuclei of the thalamus and in the cornu Ammonis dorsalis (CAd). Attention was focused on interictal spike activity, cortical spindle activity and spikes-spindles relationship. A multifocal interictal EEG pattern, represented by cortical and thalamic spikes, was observed in 20 animals. A clear association between cortical spikes and spindles occurred almost constantly. Moreover a statistically significant increase in the average duration of spindles without changes in the average frequency was noticed. Seizure discharges had always a focal start and secondary generalization. The EEG features of parenteral penicillin epilepsy in the rabbit appeared to be more similar to those of the rat than to those of the cat. The spindle activity changes and the paroxysmal spike activity can be interpreted as two independent penicillin induced features appearing during raised cortical excitability periods.

A model of chronic spontaneous petit mal-like seizures in the rat: comparison with pentylenetetrazol-induced seizures

Of 100 randomly chosen, adult male Wistar rats in the breeding colony at the Centre de Neurochimie , Strasbourg, 31 presented spontaneous, nonconvulsive epileptic seizures: wave-and-spike discharges, 7-11 cycles/s, 200-600 microV, accompanied by behavioral arrest and myoclony of the vibrissae and of the facial and cervical muscles. Pentylenetetrazol (PTZ) 10 and 20 mg/kg increased the duration and number of seizures by 100-150% in these spontaneously epileptic animals, and caused identical seizures in apparently normal rats. Sodium valproate, diazepam, trimethadione, and ethosuximide suppressed the spontaneous seizures and protected against PTZ-induced seizures in a dose-dependent fashion. Carbamazepine and diphenylhydantoin were inefficacious or aggravative in the two cases. The clinical, EEG, and pharmacological observations suggest that the Wistar rats displaying spontaneous seizures constitute a valid physiological and pharmacological model of petit mal absences, presenting advantages compared to the usual models in which seizures are induced by injected epileptogenic drugs.