Abstract
Epilepsy is a chronic neurological disorder characterized by recurrent seizures resulting from abnormal electrical activity in the brain. Experimentally, it is studied using animal models relying on artificial external induction protocols. However, this does not accurately reflect what may occur in vivo, and the processes underlying how and under which condition an otherwise normal neuronal activity can lead to a seizure are poorly understood. To better understand these processes in a hippocampal CA1 pyramidal neuron, we have implemented and validated a realistic computational model to investigate how and to what extent a local dendritic accumulation of extracellular K(+), during a sustained synaptic activation can give rise to epileptic activity similar to what observed experimentally. The model makes the experimentally testable prediction that an increase in the time constant for K(+) reuptake by the neuron or astrocytes, can bring the neuron close to a state that, while normally responding to a train of synaptic activations, may occasionally generate seizure-like activity. Furthermore, the model provides an experimentally testable prediction on how control conditions can be rescued through pharmacological intervention.