We constructed a computer model of the in vitro CA3 region of the rat hippocampal slice bathed in a high-potassium medium. Our aim was to understand better the mechanisms of initiation of synchronized bursts and the processes that regulate the interburst interval in the experimental system. 2. Our model began with a previously published model of the longitudinal CA3 hippocampal slice. The model contains three interconnected cell populations: 9,000 (excitatory) pyramidal cells; 450 inhibitory cells whose postsynaptic action is somatic and decays quickly, corresponding to chloride-dependent inhibition mediated by gamma-aminobutyric acid (GABA)A channels, and 450 inhibitory cells whose postsynaptic action is dendritic, of delayed onset and long lasting, that corresponds to K-dependent inhibition mediated by GABAB channels. 3. The model was then modified to account for specific features of the high-K experimental system: 1) the pyramidal cells do not generate intrinsic bursts; 2) EIPSP(CI) and EK are both shifted in a depolarizing direction; 3) spontaneous (i.e., not caused by presynaptic firing) excitatory postsynaptic potentials (EPSP)s were included; and 4) a steady current was injected into the pyramidal cells to depolarize them. 4. This model generates synchronized population bursts with interburst intervals of approximately 1.0-1.5 s. Bursts in individual pyramidal cells are preceded by barrages of EPSPs. These results agree with experiment. 5. Our model agrees with the following additional experiments: 1) synchronized bursts are abolished by partial blockade of excitatory synapses; 2) burst frequency is increased by partial blockade of a slow-intrinsic-K conductance; and 3) blockade of chloride-dependent inhibition leads to bursts of longer duration with longer interburst intervals. 6. The basic structural features of this model are similar to, but not identical to, the model of the disinhibited hippocampal slice. Spontaneous EPSPs appear to be critical in the high-K system for initiating, but not for synchronizing, population bursts. The experimental data and simulation results raise interesting questions about the role of spontaneous EPSPs in initiating synchronized discharges in other epileptic systems and on the possible role of spontaneous EPSPs in the normal brain.
