Abstracts

Rationale: Stroke has long been recognized as the most common cause of epileptic seizures in adults age 60 and over, accounting for over 35% of new-onset symptomatic cases of epilepsy in the elderly population. Though this association between stroke and epilepsy is well recognized, it is not well understood. Acquired epilepsy (AE) is one of the most common neurological disorders with significant consequences on quality of life. Neuronal insults such as stroke can induce neuronal plasticity changes that lead to epilepsy during the latent period after injury known as epileptogenesis. Various changes in synapses and receptor organization can be seen in epileptic tissue, but very little is known about the molecular mechanisms underlying this condition. To more efficiently study the plasticity changes that occur during epileptogenesis, we have developed a model of spontaneous recurrent epileptiform discharges (seizures) after a stroke-like injury in organotypic hippocampal slice cultures (OHSCs). OHSCs represent an advantageous model to study stroke-induced AE because they maintain neuronal morphology, cellular and anatomical relations, and network connections. Methods: Our model employs a glutamate injury paradigm that is representative of a stroke-like injury. Treatment protocol consisted of exposure to 3.5 mM glutamate for 35 minutes at DIV 21 followed by removal of glutamate and return to maintenance medium. Cell death was measured using propidium iodide (PI). Electrophysiological techniques consisted of field potential recordings and whole-cell current clamp recordings. Results: Glutamate injury revealed neuronal loss similar to that seen in other stroke models: compared to controls at 24-h cell death was 60% greater, and at 72-h it was 23% greater in glutamate treated slices. Beyond this time point cell death was not different from control. These data indicate that the initial injury phase is responsible for most of the cell death in OHSCs. The surviving, but injured cells are an ideal substrate for epileptogenesis. Indeed, field potential recordings at DIV 28-31 revealed seizure-like activity in approximately 46% of OHSCs, compared to 7.14% of controls. In addition, intracellular recordings from single CA3 pyramidal cells displayed spontaneous recurrent epileptiform discharges in about half of the glutamate-injured OHSCs, but not in the controls. Further, in field potential recordings, seizure activity was inhibited by the anticonvulsants phenobarbital and phenytoin, but not by ethosuximide. Conclusions: The morphological changes, electrophysiological characteristics and response to standard anticonvulsant treatment observed in our model is similar to stroke-induced AE in the human condition and in vivo models of epilepsy, making this a simple and powerful model to explore the underlying molecular mechanisms and changes that occur during epileptogenesis.