Animal Models of Epilepsy


This chapter includes an accompanying lecture presentation that has been prepared by the authors: .

This chapter includes an accompanying lecture presentation that has been prepared by the authors: .

Key Concepts

  • There are numerous animal models of epilepsies modeling focal onset or generalized epilepsies of genetic origin or induced with a variety of chemical, physical, or electrical stimulation methods.

  • An increasing number of models of early-onset epilepsies have been generated and extensively used in advancing knowledge regarding mechanisms or treatments.

  • Drug-resistant models have been generated in adult models of temporal lobe or neocortical epilepsy, traumatic brain injury (TBI), early-onset epilepsies such as epileptic encephalopathies, and atypical absence epilepsy.

  • Surgical approaches have been used in certain models including corpus callosotomy (CCx), resections, and novel approaches such as viral gene delivery, deep brain stimulation, or neuronal transplantation.

Animal models have provided a valuable tool to study pathomechanisms of diseases and help develop treatments to improve the quality of life of individuals affected by them. In epilepsy there has been a dramatic increase in the numbers of animal models of seizures and epilepsies, developed to simulate different types of seizures and epilepsies. This chapter discusses some of the models that are relevant to common types of epilepsies, with special focus on those that may require epilepsy surgery investigations, such as focal onset epilepsies due to structural lesions or pathologies, drug-resistant epilepsies (DREs), special pediatric epileptic syndromes, or epilepsies with a genetic-structural etiology, such as tuberous sclerosis. Models in which surgical interventions have been implemented to ameliorate epilepsies are also discussed. In certain cases, pathologies that are considered strongly associated with surgically remediable epilepsies in humans may not necessarily suffice to generate epilepsy in animals, and therefore are discussed in light of this limitation.

Developmental Equivalency Across Species

Rodent models are the most commonly used models. Rats and mice have substantial biologic differences from humans, including a briefer life span of approximately 2 years and brains that lack the typical gyrations seen in humans. Newborn pups are considered as equivalent to full-term newborns between postnatal days 8 and 13 (PN8–13) based on crude estimates of brain growth. Eye opening occurs at around PN13–15, puberty starts at around PN32–36 in females and around PN35–45 in males, and adulthood begins at around PN60. However, the maturation of different developmental processes, whether behavioral or motor milestones, trajectories of signaling pathways’ maturation, neuronal migration, or differentiation and integration occur asynchronously and with significant differences across species. For example, whereas humans start to ambulate after the first year of life, rodent pups can already ambulate in their home cage in the third week of life. Comparisons therefore across species for various developmental milestone trajectories should be done separately for each studied process.

Electrical Stimulation Models of Temporal Lobe Epilepsy or Limbic Epileptogenesis

Electrical Kindling

The concept of seizures predisposing to more severe seizures was proposed by Gowers in 1881. Experimentally, this concept was shown in the kindling model, which was described in 1969 by Goddard as an example of “permanent change in brain function resulting from daily electrical stimulation.” Kindling consists of repeated electrical stimulation of hippocampal or limbic regions that results in progressively more severe seizure activity until stage 5 (rearing and falling) seizures are seen. In adults, stimulations are usually twice daily. The progression of the kindled seizure severity includes initial appearance of focal seizures (mouth and facial automatisms, head nodding, or focal clonic seizures) followed by secondarily generalized seizures (bilateral clonic, rearing, falling). The sequence of behaviors associated with the various electrographic kindling-induced events is widely known as the Racine scale and includes mouth and facial automatisms (class 1), head nodding (class 2), forelimb clonus (class 3), rearing (class 4), and rearing and falling (class 5). Pinel and Rovner expanded this scale to include repetitive class 5 seizures (renamed as class 6); circling, jumping, and/or rolling over with vocalization (class 7); and class 7 seizures followed by tonus (class 8).

Kindling in immature animals was first described in 1981 by Moshé. , The age-dependent differences in the methods of inducing kindling were striking in that more frequent stimulations were needed to establish the kindled state in young pups and postictal refractoriness was shorter than in adult rats, and different numbers of stimulations were required in different age groups. Behaviorally, the kindling stages in pups include a progression from behavioral arrest (stage 0), mouth clonus (stage 1), head bobbing (stage 2), unilateral or alternating forelimb clonus (stage 3 or 3.5), bilateral forelimb clonus without (stage 4) or with progression to rearing and falling (stage 5), wild running and jumping with vocalizations (stage 6), and tonus (stage 7).

Kindling has been used in various species, including rodents, cats, primates, and amphibians, and in other brain regions, such as the frontal cortex. The involvement of genetic factors or inherent biologic factors that modify susceptibility to kindling has been demonstrated by McIntyre in the FAST and SLOW amygdala kindling rats. Region-specific differences in kindling were also demonstrated, with faster kindling rates at the perirhinal and piriform cortices and slower rates at stimulation of the dentate gyrus stimulation. Beyond seizures, FAST kindlers also demonstrate more behavioral traits, such as attention deficit and hyperactivity, easy distraction, and impulsive behavior, compared with SLOW kindlers.

Kindling has been extensively used to model the process of limbic epileptogenesis and for testing drugs for their ability to either retard kindling epileptogenesis (i.e., increase the number of stimuli required to reach the kindled state) or suppress fully kindled seizures. However, spontaneous seizures have been described only rarely, after hundreds of stimulations in adults, and accordingly it is more common that kindling is used as model of epileptogenesis than of pure epilepsy. Whether and, if so, when kindling may occur in humans to explain progression or secondary epileptogenesis has been open to discussion and debate.

Post–Status Epilepticus Electrical Stimulation Models of Temporal Lobe Epilepsy

Unilateral or bilateral repetitive electrical stimulation of amygdala, hippocampus, or perforant pathway has been used to induce status epilepticus (SE) and mesial temporal lobe epilepsy (MTLE). The advantage of these methods is that they strictly use the effect of electrical stimulation, partly simulating the effect of repetitive epileptic activities or seizures originating from limbic regions. The acute seizures include SE with stage 4 (rearing) or stage 5 (rearing and falling) seizures. Mortality is generally lower than in the pharmacologic models of SE. Epilepsy ensues in most of the animals, and seizures in response to sensory stimuli (auditory, handling) have also been reported. Pathology includes neuronal loss, and gliosis in hippocampal regions. In the chronic phase, cognitive deficits may occur.

Chemoconvulsant Models of Temporal Lobe Epilepsy

TLE is one of the most common types of epilepsy in adults; MTLE is the most common DRE type, and hippocampal sclerosis (HS) is the most common pathology in resected temporal lobe specimens from patients with DRE. TLE may have genetic or acquired etiologies or predisposing factors, although in many cases the etiology remains unknown. In rodent models, several induction procedures that have been used to reproduce aspects of the phenotype include pharmacologic or electrical stimulation of limbic structures of naïve animals or animals harboring certain predisposing conditions. The outcomes and phenotypes may differ according to the age at which these induction methods have been applied, as is discussed in the following sections. The most commonly used induction process is SE by administration of glutamatergic or cholinergic receptor agonists.

Post–Status Epilepticus Models of Temporal Lobe Epilepsy: Kainic Acid Models

Excitatory amino acids that have been used to induce SE include kainic acid (KA), quisqualic acid, N -methyl- d -aspartate (NMDA), homocysteine, and homocysteic acid. These compounds have been administered either systemically or focally in the brain or intracerebroventricularly, resulting in different types of injury, seizures, and long-term phenotypes. Because of the specific relevance to TLE, we will focus here on KA.

KA, also known as kainate, is an excitatory amino acid, an analogue of l -glutamic acid, that occurs naturally in seaweed and activates kainate ionotropic glutamatergic receptors. KA has neurotoxic potential and has been used to induce local lesions in the brain or for its anthelmintic properties. , However, KA has garnered a lot of scientific interest because it can also induce seizures lasting for longer than 30 minutes and SE that may eventually develop into epilepsy and hippocampal pathology reminiscent of HS, the telltale pathology of MTLE. However, age- and sex-specific differences have been reported in seizure susceptibility, ensuing pathology, or molecular and neurophysiologic changes and subsequent epilepsy. , ,

In rodents, acute intraperitoneal injection(s) result in hypoactivity followed by automatisms, “wet-dog shakes,” scratching, clonic seizures, rearing and falling, or hyperactivity; tonic-clonic seizures are rarely seen except with high doses. , Induction can be done either via single injection or with repetitive low-dose intraperitoneal injections tailored according to the severity of seizures that appear, so as to minimize mortality. A modification of the KA model includes focal injection of KA in limbic regions or intracerebroventricularly. Dose adjustment is needed in younger age groups, and with different strains and species, to minimize the mortality often seen with KA-SE. In general, immature pups develop SE with much lower doses and lower mortality than in adults. , , Electroencephalographic (EEG) recordings show focal epileptic discharges at limbic areas, which subsequently spread to cortical regions. These may last for hours and may persist for days without overt convulsive behaviors.

In adults, systemic or intraventricular KA-induced SE causes neuronal loss and gliosis in the hippocampus, subiculum, and entorhinal cortex; dentate granule cell dispersion; and mossy fiber sprouting. Pathology can be ameliorated if acute seizures are controlled with antiseizure drugs. , However, if KA-SE is induced in the first week of life, there is no overt injury and no long-term epilepsy, whereas incremental age-specific injury and epileptogenesis are seen if KA-induced SE is induced in older pups. , Sex-specific effects of early-life KA-SE on GABA A receptor signaling maturation or cognitive impairment have been described. ,

Following the acute period, there is a latent period that follows before the chronic spontaneous epileptic seizures appear. The latent period can be variable and typically requires EEG recordings to capture behaviorally silent electrographic seizures that typically precede the appearance of electroclinical seizures seen in the chronic phase. The frequency of seizures and their association with behavioral or motor manifestations, as well as the neuropathologic changes, increase with time, suggesting a progressive nature. , In adults, almost all animals expressing acute SE after systemic KA injection develop spontaneous seizures in long-term recordings. In the focal KA-SE model, the rate of epilepsy development is higher if induction is done in awake animals compared with those under anesthesia. In immature animals, KA-SE, however, does not result in epilepsy if induction is in the first week of life, and the rate of epilepsy increases with age thereafter. , , Comorbidities associated with the KA-SE model include deficits in visuospatial learning and/or memory, hyperexcitability, and aggressive behaviors during handling.

Post–Status Epilepticus Models of Temporal Lobe Epilepsy: Pilocarpine Models

Pilocarpine is a muscarinic receptor agonist, derived from the plant Pilocarpus microphyllus, that was introduced as a method to induce SE in rodents in 1983. Pilocarpine can induce SE at high doses; however, mortality can be significant. Pretreatment with scopolamine and, in certain cases, early termination with benzodiazepines minimizes the peripheral side effects and high mortality that may ensue. Subcutaneous rather than intraperitoneal injection has also been used to reduce mortality. More recently, pretreatment with lithium chloride, given the day before, has allowed reduction in the pilocarpine doses, minimizing side effects while still resulting in SE. A reduced-intensity status epilepticus (RISE) lithium-pilocarpine SE model has been proposed as a low-mortality, high-morbidity variation of the model that still produces the chronic epilepsy phenotype. The RISE protocol includes lithium chloride subcutaneous injection a day prior and scopolamine 30 minutes before the first low pilocarpine subcutaneous dose; pilocarpine can be repeated if resultant seizures are not severe enough (i.e., bilateral forelimb clonus with rearing). If such seizures appear, xylazine is administered followed by a combination of MK-801 (NMDA receptor antagonist), diazepam, and MPEP (metabotropic glutamate receptor 5 antagonist) given subcutaneously an hour later. Intracranial infusion of pilocarpine in the ventricle or hippocampus has also been used. Intrahippocampal, unilateral pilocarpine application causes SE but with minimal mortality, and pathology that is more contained in the hippocampal regions, resembling to human TLE, while preserving the epilepsy phenotype.

In general, the pilocarpine SE models have been reported to have a similar spectrum of manifestations as the KA-SE models, including the acute SE stage, latent and chronic epilepsy stages, pathology that exhibits features of HS, or learning disabilities, anxiety, hyperactivity, irritability, and aggression. As with KA-SE, age- and sex-specific factors greatly influence the manifestations of the pilocarpine models and need to be considered in the study design. These include susceptibility to pilocarpine seizures, subsequent development of epilepsy, and the induced neuropathology. , ,

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here