Cellular contributions to ictal population signals

Abstract

AbstractThe amplitude of ictal activity is a defining feature of epileptic seizures, but the determinants of this amplitude have not been studied. Pathological synchronization of neuronal activity is assumed to be the main driver of increased ictal amplitudes. However, cell intrinsic changes that drive pathological neuronal discharges may also contribute to ictal amplitude. Clinically, ictal amplitudes are measured electrographically (using e.g. EEG, ECoG, and depth electrodes), but these methods do not enable the assessment of the contributions of the amplitudes of individual neurons as well as synchronization to the amplitude of the ictal signal. We therefore measured ictal population activity by recording both electrographic local field potentials (LFP) and neuronal calcium levels, the latter measured using optical imaging of neuronal transgenic calcium-sensitive fluorophores. Spontaneous seizure activity was first assessed in an awake, behaving mouse model of focal cortical injury, using paired GCaMP6-based calcium imaging and LFP electrical recording. Spontaneous recurrent seizures were then measured in organotypic hippocampal slice cultures (OHSC), with a combination of GCaMP7 calcium imaging and LFP recordings. OHSC is an in vitro preparation in which all electrical and calcium signals could be unambiguously ascribed to the epileptic network. Population-averaged calcium activity was highly correlated with the time integral of electrographic field recordings, both in vivo and in vitro. We found that cell intrinsic changes were the largest contributor to the population signal during seizure onset. In other words, the network signal was not merely the summation of highly synchronous physiological activity, but that individual neurons were generating pathological discharges. During frank seizure, individual neurons continued to generate pathological activity, which also became highly synchronized. Interestingly, recruitment of newly active neurons accounted for only a small fraction of synchronous activity, particularly in vivo, revealing that network synchrony was largely due to reactivation of the same population of neurons.