Detailed Models of Network Bursts and Oscillations

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Epilepsy, a condition of recurrent spontaneous seizures, affects 0.5-1% of the population, and remains an intractable condition in many cases. Additionally, the problem of anticipating incipient seizures, based on brain electrical signals, remains difficult. Very fast brain oscillations (VFOs) at >100 Hz are a locational marker of epileptogenic tissue; they can occur during interictal ("between-seizure") potentials, and also can occur - as a harbinger - just prior to frank seizures. [Some authors, such as A. Bragin and J. Engel, Jr., distinguish "ripples" at <~250 Hz from "fast ripples" at >~250 Hz, but for the present, let us lump them together.] VFO, however, also occurs during normal brain events, such as physiological sharp-wave ripple (SPW-R) complexes, events that appear important in the consolidation of spatial memories: thus, VFO may be critical for normal cognitive processes. The similar appearance of pathological and physiological VFO events (although not fast ripples) suggests shared mechanisms. We are therefore faced with a dilemma: on the one hand, it is essential to distinguish the mechanisms underlying pathological vs. normal VFO, because suppressing pathological VFO might prevent seizures, while suppressing normal VFO might interfere with memory; but on the other hand, the likely sharing of mechanisms implies that this distinction may not be straightforward. VFO can occur experimentally in conditions when chemical synapses are blocked. A large body of evidence suggests that VFO originates from gap junctional coupling of pyramidal neurons, via axons. Models based on electrical coupling account for multiple features of VFO, including the frequency range, spatial properties (in epileptic neocortex), intracellular potentials and synaptic currents, pH sensitivity, and pharmacology. (Other models of VFO, based on chemical synapses and/or field effects have run afoul of experiments showing persistence of VFO with synapses blocked, and of the very small amplitudes of the extracellular potentials.) Other aspects of VFO are not understood: for example, the limited somatic firing during physiological sharp- wave ripples; or the ability of pathological tissue to generate either "ripples" or "fast ripples", one after the other, as apparently discrete, separable events. Nor is it known how, during a physiological SPW-R, VFO occurs superimposed upon summated synaptic potentials; while, prior to a seizure, VFO can occur by itself. This proposal seeks to investigate these and other aspects of VFO, using both highly detailed and also highly simplified network models; and using data from rodent experiments as well as human tissue recorded in situ and in vitro. The computational models will be predictive and motivate further experimental and eventually clinical investigations. A long-term goal is to use refined understanding of VFO mechanisms to be able to prevent or suppress pathological fast brain oscillations, with minimal perturbation of the normal ones.