Abstracts

Rationale: There is great interest in the role of coherent oscillations in the brain. In some cases, high frequency oscillations (HFOs) are integral to normal brain function, while at other times they are implicated as markers of epileptic tissue. Ripples (100-200 Hz) and fast ripples (250-500 Hz oscillations) are promising biomarkers of epileptic tissue, but the mechanism of their generation is unknown. We investigate random synaptic activity (noise) as a potential initiator of HFOs, and explore the network parameters necessary to produce these events. Methods: We implemented a physiological computer model of hippocampus containing 80 pyramidal cells and 20 basket cells. Our model included synaptic noise sources, gap junctions, and recurrent axons, all of varying strengths. Noise sources were AMPA synapses in the apical dendrites of pyramidal cells and cell bodies of basket cells. The output of the network was the summed somatic voltages of the pyramidal cells. Results: These simulations demonstrate that, under normal coupling conditions, random synaptic noise can drive the network to oscillate coherently at gamma (30-100 Hz) frequencies. Abnormal levels of gap junctions and/or recurrent axons allowed the network to oscillate faster, at ripple frequencies, with high levels of noise. The frequency of oscillation was highly dependent on basket cell activity. The model predicts that, in order to reach fast ripple (> 250 Hz) frequencies, the feedback inhibition from basket cells cannot be organized as a single syncytium. Fast ripples were formed with high levels of synaptic noise only when the basket cells were divided into multiple independent populations. These fast ripples were different in character from those previously reported, but are very similar to those found in human recordings currently being collected in our laboratory. Conclusions: The model predicts that increased gap junctions, recurrent action potentials, and disruption of normal basket cell activity are able to produce HFOs when random synaptic activity is high. These network phenomena are common experimental findings in epileptic tissue. These results suggest that fast ripples and ripples have a physiological basis as potential biomarkers of epilepsy: they can be formed by the same pathological changes seen in epilepsy. Furthermore, they suggest a possible mechanistic difference between fast ripples and some forms of ripples.