No abstract

Kainate receptors (KARs) on CA1 pyramidal cells make no detectable contribution to EPSCs. We report that these receptors have a metabotropic function, as shown previously for CA1 interneurons. Brief kainate exposure caused long-lasting inhibition of a postspike potassium current (I(sAHP)) in CA1 pyramidal cells. The pharmacological profile was independent of AMPA receptors or the GluR5 subunit, indicating a possible role for the GluR6 subunit. KAR inhibition of I(sAHP) did not require ionotropic action or network activity, but was blocked by the inhibitor of pertussis toxin-sensitive G proteins, N-ethylmaleimide (NEM), or the PKC inhibitor calphostin C. These data suggest how KARs, putatively containing GluR6, directly increase excitability of CA1 pyramidal cells and help explain the propensity for seizure activity following KAR activation.

Prolonged modification of intrinsic neuronal excitability is gaining prominence as an activity-dependent form of plasticity. Here we describe a potential synaptic initiation mechanism for these changes in which release of the transmitter glutamate acts on kainate receptors to regulate the postspike slow afterhyperpolarization (sAHP). This action of synaptically released glutamate was occluded by previous kainate application. Furthermore, inhibition of glutamate uptake enhanced the effects of synaptic activation. Glutamatemediated kainate receptor inhibition of sAHP current (I sAHP ) was blocked by the PKC inhibitor calphostin C, confirming the requirement for a metabotropic signaling cascade. These data describe a new physiological function for glutamate release: activation of metabotropic kainate receptors, which control directly the excitability of pyramidal cells and probably contribute to prolonged excitability changes.

In this study we describe a model of connectivity linking the different neurons in the CA3 and CA1 areas of the young male rat hippocampus. The anatomical and electrophysiological values of the parameters used in the model were selected from the available literature. Each type of synapse was characterized by its spatial location on the dendritic tree, its weight, its probability of activation, and the ionotropic receptors involved. We have shown that the degree of convergence and divergence of inputs is highly dependent upon the type of neuron and its spatial location. The different gradients of connectivity we describe support the lamellar hypothesis from a functional point of view, even if the anatomical patterns seem diffuse. The analysis of the proportion of common afferents to a class of neurons further confirmed this point. It is suggested that the circuitry creates local coherence in terms of processing of information by establishing restricted areas where information is preferentially treated. The functional consequences and limitations of these findings are also discussed. This model is the first step in the development of a network model of the hippocampus with realistic architecture.

No abstract