The goal of this proposal is to understand the reciprocal interactions between the activity of a neural network and the properties of the synapses that connect the individual neurons. The rationale for these experiments is based on the recent findings that in area CA3 of the hippocampus many aspects of synchronous network activity are determined by the properties of the glutamatergic recurrent collateral synapses that connect CA3 pyramidal cells. For example, synchronous CA3 network activity is terminated not by feedback inhibition but by depression of the recurrent collateral synapses; further, the strength of the recurrent collateral synapses determines the probability of initiating synchronous network activity, but does not ,determine the duration of that activity. We propose to develop these initial results by testing 2 related hypotheses. First, synchronous activation of the CA3 network is regulated by quantifiable properties of the recurrent collateral synapses between CA3 cells: the supply of releasable glutamate, the probability of glutamate release, and postsynaptic determinants of synaptic strength. Second, under appropriate conditions, network activity is sufficient to produce both short and long-tern changes of these synaptic properties. We will test these hypotheses using single and dual whole-cell recordings in the hippocampal slice preparation. These studies will elucidate principles by which real neural networks operate, such as how synaptic plasticity alters network operation. In addition, understanding how normal synchronous network activity is terminated will delineate new targets for epilepsy therapy. For example, the minor role of postsynaptic inhibition in CA3 burst termination indicates that GABA-mediated inhibition is not the optimal anticonvulsant target in at least some neural networks, and preliminary studies on the link between synaptic strength and network activity suggest a novel epilepsy therapy: exploiting interictal network activity to produce a specific, long-term weakening of the synapses that underlie the propensity for seizures.
Showing the most recent 10 out of 16 publications
