Abstracts

Electrical stimulation of the brain has been considered as an alternative treatment to reduce seizure frequency or duration. The mechanism of seizure suppression by electrical stimulations is unclear. In a variety of experimental models, both low and high frequency stimulation have been reported to suppress seizure-like activity, which suggests involvement of various mechanisms. Here we model the effect of different frequencies of electrical stimulation on simulated epileptiform activity in neural networks. Electrical stimulation is modeled as a train of action potentials (APs) in axons of presynaptic neurons. The frequency of APs varies between 1 Hz and 150 Hz. The duration of stimulation varies from 10 ms to 300 ms for high frequency and from 1 s to 60 s for low frequency stimulation. Neurons are simulated using a single compartment conductance-based model. Excitatory and inhibitory neurons (total 1440) are synaptically connected and this network model simulates the neurons of an epileptic focus. Synaptic plasticity is simulated using a phenomenologic model (Shouval et al., 2002) in which the synaptic strength is regulated by the calcium concentration and the temporal pattern of calcium transients in synaptic spines. This model of synaptic plasticity produces either synaptic potentiation or depression depending on the temporal pattern of the synaptic activation. We included in the synaptic model several mechanisms regulating calcium concentration in spines. Low frequency stimulations (1 - 3 Hz) lead to long-term depression of excitatory synapses in a network. Several seconds of low frequency stimulation of neurons before the onset of the simulated epileptiform activity results in diminished synchronization of neuronal activity and delay of seizure onset. The brief high-frequency stimulations ([gt] 30 Hz) after onset resulted in prompt suppression of ongoing epileptiform activity. High frequency stimulations lead to transient synaptic depression and momentarily suppress ongoing epileptiform activity in a network. Low frequency stimulation applied before the onset followed by brief high frequency stimulation after onset is the most effective in suppressing epileptiform activity in the network. Suppression of simulated epileptiform activity requires stimulation of multiple neurons of the seizure focus. Synaptic plasticity may explain modulating effect of electrical stimulation. (Supported by NIH grant NS38958.)