Abstracts

RATIONALE: Recent experimental data suggest that mossy fiber sprouting and other forms of synaptic reorganization contribute to recurrent excitatory networks in evolving epileptic foci. Using a nonspecific neuronal network model, incorporating both excitatory and inhibitory connections, we assessed the changes in network burst generation after increasing the number of synaptic connections. METHODS: Networks of synaptically connected neurons were modeled using the modified Av-Rinzel reduced model of a single neuron. Each neuron has inputs from randomly chosen excitatory and inhibitory neurons. The network is constantly activated by random excitatory input. The number of synaptic inputs and synaptic weights can be modified for individual neurons or the network in general. The initial parameters of the model and number of synaptic connections were set to values not producing synchronized bursting activity. In subsequent simulations the number of connections were increased. For a given set of parameters, multiple networks (10) were modeled. RESULTS: Increasing the number of excitatory connections per neuron, a simulation of a sprouting paradigm, increases bursting activity in the network. Increased bursting activity is also frequently observed, however, when both excitatory and inhibitory connectivity is proportionately, but still randomly, increased throughout the network. CONCLUSIONS: An increase in random synaptic connections in a model excitatory-inhibitory neural network increases bursting activity of the network not only when excitatory connections are added but also when both excitatory and inhibitory connections are increased. Thus evolving epileptogenesis may occur in neural networks where sprouting is not restricted to excitatory neurons. Supported by the NIH grant NS 38958