Hierarchy of seizure states in the electrogenic limbic status epilepticus model: behavioral and electrographic observations of initial states and temporal progression

Increased Susceptibility to Pilocarpine-Induced Status Epilepticus and Reduced Latency in TRPC1/4 Double Knockout Mice

Canonical transient receptor potential channels (TRPCs) are a family of calcium-permeable cation channels. Previous studies have shown that heteromeric channels comprising TRPC1 and TRPC4 mediate epileptiform bursting in lateral septal neurons and hippocampal CA1 pyramidal neurons, suggesting that TRPC1/4 channels play a pro-seizure role. In this study, we utilized electroencephalography (EEG) recording and spectral analysis to assess the role of TRPC1/4 channels in the pilocarpine model of status epilepticus (SE). We found that, surprisingly, TRPC1/4 double knockout (DKO) mice exhibited an increased susceptibility to pilocarpine-induced SE. Furthermore, SE latency was also significantly reduced in TRPC1/4 DKO mice. Further studies are needed to reveal the underlying mechanisms of our unexpected results.

Pilocarpine-induced status epilepticus in mice: A comparison of spectral analysis of electroencephalogram and behavioral grading using the Racine scale

Pilocarpine-induced status epilepticus (SE) is a widely used seizure model in mice, and the Racine scale has been used to index seizure intensity. The goal of this study was to analyze electroencephalogram (EEG) quantitatively using fast Fourier transformation (FFT) and statistically evaluate the correlation of electrographic seizures with convulsive behaviors. Simultaneous EEG and video recordings in male mice in a mixed genetic background were conducted and pilocarpine was administered intraperitoneally to induce seizures. The videos were graded using the Racine scale and the root-mean-square (RMS) power analysis of EEG was performed with Sirenia Seizure Pro software. We found that the RMS power was very weakly correlated with convulsive behavior induced by pilocarpine. Convulsive behaviors appeared long before electrographic seizures and showed a strong negative correlation with theta frequency activity and a moderate positive correlation with gamma frequency activity. Racine scores showed moderate correlations with RMS power across multiple frequency bands during the transition from first electrographic seizure to SE. However, there was no correlation between Racine scores and RMS power during the SE phase except a weak correlation with RMS power in the theta frequency. Our analysis reveals limitations of the Racine scale as a primary index of seizure intensity in status epilepticus, and demonstrates a need for quantitative analysis of EEG for an accurate assessment of seizure onset and severity.

Treatment of late lithium-pilocarpine-induced status epilepticus with diazepam

PURPOSE

It is well known that status epilepticus (SE) becomes increasingly difficult to control over time. Previous studies have indicated that the electroencephalographic pattern at the time of intervention is predictive of the probability of successful treatment. However, these findings are based on studies limited to the first 2h of SE onset. Little data is available on the efficacy of treating SE at later time points.

METHODS

The aim of the present study was to investigate the efficacy of diazepam (DZP) treatment given at two different phases of SE in a lithium-pilocarpine rat model: during continuous ictal discharges (CIDs, phase 3), and during late periodic epileptiform discharges (late PEDs, phase 5). Changes in cortical and hippocampal electroencephalographs (EEGs) were observed continuously during the phases of SE, as well as at 24, 36, 48, and 72h after SE onset. The effects of DZP treatment during CIDs or during late PEDs were compared to control DZP-untreated rats.

RESULTS

In all three groups, hippocampal and cortical EEGs displayed five distinct phases of SE. There were no statistical differences in the duration of phases 1 and 2 among the three groups. Although DZP administration during CIDs did not terminate CIDs in most rats, it did significantly shorten the duration of phases 3 and 4 of SE. Importantly, DZP given during phase 5 successfully ended behavioral and electrographic seizures in most rats.

CONCLUSIONS

Hippocampal and cortical EEGs displayed five distinct phases of SE that were similarly responsive to DZP treatment. Termination of electrographic seizures with DZP treatment was more effective in the last phase of SE (late PEDs) than in phase 3 (CIDs). These findings suggest that previous reports of DZPs decrease in efficacy over time may not be applicable to DZP treatment at 4h-post onset.

Status epilepticus severity influences the long-term outcome of neurogenesis in the adult dentate gyrus

Status epilepticus (SE) is characterized by continual seizure activity that can vary widely in the intensity of convulsions. We induced seizures by applying continuous electrical stimulation to the hippocampus in adult rats to explore the effects of three different SE states on neurogenesis and neuronal death in the hippocampus. Rats exhibiting the most severe SE state (fully convulsive) demonstrated profound increases in cell proliferation in the dentate gyrus (DG) at 1 week post-insult, but the majority of the new neurons had died at 4 weeks. In contrast, rats exhibiting less severe SE states (ambulatory or masticatory, partial convulsive) had the same degree of cell proliferation at 1 week, but most new neurons survived at 4 weeks. As compared to partially convulsive SE rats, fully convulsive SE rats had significantly greater DG pathology. Our data indicate that SE of varying severity triggers similar short-term proliferation of neural progenitors, but that the long-term outcome of neurogenesis is influenced by the degree of insult-induced degeneration in the DG tissue environment.

Self-sustaining status epilepticus: a condition maintained by potentiation of glutamate receptors and by plastic changes in substance P and other peptide neuromodulators

We describe a model of self-sustaining status epilepticus (SSSE) induced by stimulation of the perforant path in free-running rats. In this model, seizures can be transiently suppressed by intrahippocampal injection of a blocker of alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid/ kainate synapses but return in the absence of further stimulation when the drug ceases to act. However, seizures are irreversibly abolished by blockers of N-methyl-D-aspartate receptors given locally or systemically. SSSE is enhanced by substance P and its agonists and blocked by its antagonists. SSSE induces novel expression of substance P-like immunoreactivity in hippocampal principal cells. These changes and those in other limbic peptides may contribute to the maintenance of SSSE and to the modulation of hippocampal excitability during epileptic seizures. NMDA

Electrocorticographic changes during generalized convulsive status epilepticus in soman intoxicated rats

Generalized convulsive status epilepticus (GCSE) is the most common and potentially most damaging form of status epilepticus (SE). It has been previously reported, in both human GCSE and animal models of GCSE, that the electroencephalographs (EEGs) and electrocorticographs (ECoGs) recorded during GCSE contain an ordered sequence of five identifiable patterns: discrete seizures (phase 1), waxing and waning ictal discharges (phase 2), continuous ictal discharges (phase 3), continuous activity with flat periods (phase 4), and periodic epileptiform discharge on a flat background (phase 5). In this paper, we report the same pattern of ECoG changes in 15 rats exposed to soman, an acetylcholinesterase (AChE) inhibitor. Phase 1 was observed in 12 of 15 animals, but phases 2-5 were recorded in all the animals. Taken together, these findings suggest that the sequence of EEG changes is independent of the initiating cause, represent a common electrical response to GCSE, and reflect a common underlying neurochemical mechanism.

Electroencephalographic and histological characteristics of a model of limbic status epilepticus permitting direct control over seizure duration

Status epilepticus is a neurological emergency associated with substantial morbidity and mortality. Experimental and clinical investigations suggest that prolonged seizure activity is associated with injury to vulnerable neurons. Compounds with neuroprotective properties may minimize such injury. Existing methods of inducing experimental status epilepticus result in seizure activity of variable duration and neuronal injury of variable degree. To minimize such variability, status epilepticus may be stopped with anticonvulsants, but this limits the ability to screen for independent neuroprotective properties. We have developed a simple and reliable non-pharmacological model of limbic status epilepticus in which the duration of status epilepticus is under direct experimental control. Status epilepticus is induced by continuous, unilateral hippocampal stimulation. Using this model, the degree of hippocampal pyramidal cell injury varies in direct proportion to status epilepticus duration across a range of 15-140 min. A progressive sequence of EEG changes unfolds with increasing status epilepticus duration, resembling that seen in other models. This model may serve as a reference against which the effects of potential neuroprotective compounds can be studied.

Neuropharmacological mechanisms of nerve agent-induced seizure and neuropathology

This paper proposes a three phase "model" of the neuropharmacological processes responsible for the seizures and neuropathology produced by nerve agent intoxication. Initiation and early expression of the seizures are cholinergic phenomenon; anticholinergics readily terminate seizures at this stage and no neuropathology is evident. However, if not checked, a transition phase occurs during which the neuronal excitation of the seizure per se perturbs other neurotransmitter systems: excitatory amino acid (EAA) levels increase reinforcing the seizure activity; control with anticholinergics becomes less effective; mild neuropathology is occasionally observed. With prolonged epileptiform activity the seizure enters a predominantly non-cholinergic phase: it becomes refractory to some anticholinergics; benzodiazepines and N-methyl-D-aspartate (NMDA) antagonists remain effective as anticonvulsants, but require anticholinergic co-administration; mild neuropathology is evident in multiple brain regions. Excessive influx of calcium due to repeated seizure-induced depolarization and prolonged stimulation of NMDA receptors is proposed as the ultimate cause of neuropathology. The model and data indicate that rapid and aggressive management of seizures is essential to prevent neuropathology from nerve agent exposure.

Mapping of limbic seizure progressions utilizing the electrogenic status epilepticus model and the 14C-2-deoxyglucose method

We have previously described a model of limbic status epilepticus in which chronic prolonged seizure states of immobile, exploratory, minor convulsive or clonic convulsive behavior are induced by intracerebral electrical stimulation; these states appear to belong to the same behavioral progression as kindled seizures. We postulated that the underlying seizure substrates, as mapped by the 14C-2-deoxyglucose method, should reflect a corresponding anatomic progression of discharge spread. Status epilepticus was induced in rat by pulsed-train current delivered for up to 90 min to one of several subcortical areas. Autoradiographs revealed that most of the observed patterns of seizure-induced metabolic activation comprised a hierarchical sequence, such that progressively more extensive patterns subsumed anatomic territories activated in less extensive patterns, thus allowing inferences as to the progression of discharge spread. In this sequence, the basolateral amygdala ipsilateral to the induction electrode was among the first structures to be activated. In successively larger activation patterns a small unilateral network related to basolateral amygdala was involved; this evolved through a transitional state to a unilateral extensive limbic pattern; which in turn was succeeded by bilateral extensive limbic activation. This hierarchical sequence culminated in a neocortical activation pattern, in which most of the forebrain was involved in intense seizure-induced activation. Seizure behaviors increased in severity in correspondence with the underlying seizure-activated anatomic substrate. In contrast, patterns of seizure activation were observed which did not fit within the early stages of the above sequence, although analysis indicates that the later stages of spread may be shared. The study of these patterns and those reported in the literature indicates that although limbic seizure networks may be anatomically distinct at their origination, further expansion is characterized by overlap; upon assumption of extensive patterns of activation the number of nuclei participating is so vast that the identity of the limbic originator is lost and common convulsive manifestations occur.

A new, non-pharmacologic model of convulsive status epilepticus induced by electrical stimulation: behavioral/electroencephalographic observations and response to phenytoin and phenobarbital

Much remains to be learned about mechanisms underlying entry into, and temporal progression of, status epilepticus (SE). This report describes a non-pharmacologic model of generalized convulsive SE in rat. Pulsed trains of suprathreshold electric current, were administered bilaterally to either of four rostral forebrain sites: orbital cortex, medial precentral cortex, deep prepiriform cortex, or rostral caudate-putamen (n = 8 per site). This induction method resulted in 30/32 animals attaining limb-clonic convulsive SE within a mean of 30-35 min for each forebrain site, with no differences between sites. Subsequent SE proceeded without further interventions, permitting observation of the natural course of progression. A stereotyped behavioral/electrographic sequence occurred, characterized by devolution. Behaviorally, animals progressed from predominantly limb clonus to head clonus, then to subtle twitching, and finally to electrical SE before cessation of spikes. The corresponding electrographic progression was from fast and slow spiking to periodic epileptiform discharges (PEDs). In 20 animals surviving to 48 h, pathologic damage affected mainly limbic sites; damage was related to total convulsive time rather than to clonic activity. High-dose phenobarbital but not phenytoin suppressed SE when given during orbital cortex-induced limb-clonic SE. These findings are compatible with human observations and indicate that this model will enable investigations of generalized SE mechanisms and evaluation of new therapeutic agents for refractory SE.

Anticonvulsant effects of the experimental induction of hippocampal theta activity

Epileptic seizures occur less during wakefulness or paradoxical sleep, conditions during which hippocampal theta rhythm is seen. This leads to the hypothesis that this rhythm indicates a physiological state of the hippocampal formation which opposes its recruitment into seizures. This was tested by determining the effects of experimental induction or suppression of hippocampal theta activity on seizures. Hippocampal theta activity can be induced by chemical or electrical stimulation of the medial septal nucleus and adjacent nucleus of the diagonal band of Broca. Microinjections of the muscarinic agonist carbachol in the medial septum during pentylenetetrazol (PTZ) induced facial-forelimb seizures stopped behavioral seizures and EEG spiking within five seconds, and caused hippocampal theta activity. Medial septal electrical stimulation at 4-8 Hz had similar effects. Electrolytic medial septal lesions abolished hippocampal theta activity and lowered myoclonic and facial-forelimb PTZ seizure thresholds. Medial septal carbachol injections were also made during electrically kindled limbic status epilepticus. Within ten seconds, ictal behavior stopped and the EEG spike rate decreased by half with a gradual return to the baseline rate over three minutes. These results demonstrate that the hippocampal theta rhythm corresponds to a seizure-resistant condition, providing a possible explanation for the seizure promoting properties of slow wave sleep.

Effect of an adenosine antagonist and an adenosine agonist on status entry and severity in a model of limbic status epilepticus

Adenosine is an endogenous neuromodulator that suppresses excitatory neurotransmission. We postulated that adenosine-mediated mechanisms resist status epilepticus (SE) entry and limit SE severity. In the first experiment rats were given an adenosine agonist (2-chloroadenosine), an adenosine antagonist (aminophylline), or saline vehicle, prior to SE induction with pulsed-train current delivered to amygdala in successive 5-min current-on sessions. Saline-treated animals entered limbic SE, with predominantly exploratory behavior, after 6.0 +/- 0.9 current-on sessions. Aminophylline increased major convulsive activity during stimulation and resulted in entry into convulsive SE after only 2.1 +/- 0.1 sessions. 2-Chloroadenosine, in contrast, suppressed major convulsive activity during stimulation, and blocked (in 3/7) or delayed (4/7) SE entry, with successes requiring 12.8 +/- 0.9 stimulation sessions. In a second experiment, animals already in exploratory SE were administered a single injection of saline vehicle, aminophylline, or 2-chloroadenosine. Aminophylline converted exploratory SE into lethally severe convulsive SE. 2-Chloroadenosine suppressed SE behaviorally and electrographically, and protected recipients from the seizure-associated cerebral damage seen in saline-administered SE controls. These results support the hypothesis that endogenous adenosine mechanisms resist SE entry, modulate the severity of ongoing SE, and limit the anatomic spread of seizure activity.

Electrogenic status epilepticus induced from numerous limbic sites

We have previously shown that pulsed-train electrical stimulation of the amygdala results, within 60-90 min, in status epilepticus in which one of four behavioral states is predominant: immobile, exploratory, minor convulsive and clonic convulsive. These four states form a hierarchy which appears, on electrographic and behavioral grounds, to belong to the same order of severity as limbic-kindled seizures, but not representing steady-state versions. We tested the hypothesis that it should be possible to induce prolonged seizure states from numerous limbic sites, as is the case with kindled seizures. It was found that electrogenic status epilepticus could be generated at any of 11 extra-amygdala sites, including parts of hippocampal formation, olfactory/limbic cortical areas, and caudate putamen. These four status categories could each be engendered from numerous sites without any site-specific behavioral features. Midline cortical and caudate stimulation was more prone to elicit clonic convulsive status. Such findings provide further evidence that kindled seizures and electrogenic status states belong to the same progression, and share the same underlying anatomic substrates.