Neuronal and Network Mechanisms of Ictogenesis

摘要

Epilepsy is a chronic disorder marked by recurrent, unpredictable seizures. Classic seizure prediction methods use signal processing algorithms on EEG signals to identify the pre-seizure period. However, this approach has produced disappointing results. Here, our goal is to elucidate the mechanisms that underlie ictogenesis, the process by which normal neural activity synchronizes during the pre-seizure period. We focus on neuronal and network-level mechanisms and how they interact to create ictogenic conditions.;At the neuronal level, we use computational modeling to suggest that intrinsic properties of individual neurons can determine a brain region's propensity for ictogenesis. We then compared the intrinsic properties of neurons in piriform cortex (PC), a region implicated in temporal lobe epilepsy, to neurons in somatosensory cortex (S1), a clinically stable region. Consistent with our model, we found that PC neurons are more excitable than those in S1.;At the network level, we use computational and experimental methods to explore how network parameters contribute to ictogenesis. In our simulations, we found a specific relationship between the network's propensity for ictogenesis and network parameters such as size and connectivity. In our experiments, we use fluorescent imaging of cortical slices from S1 and PC to compare activity during ictogenesis. Our results suggest that the spatially distributed circuitry of PC is reflected in the nonlocal pattern of activation, whereas S1 activity patterns reflect the recruitment of local circuitry in a smooth, sequential manner.;Finally, we examine the interaction of underlying mechanisms by analyzing an abstract model of neuronal population activity. Using dynamic systems analysis, we formalize the concept of network threshold, and show that it depends on a complex interaction of neuronal and network mechanisms, as well as the variability in the network. Our analysis suggests that network threshold is not a static parameter, but a dynamic and potentially modifiable characteristic of an active network.;Taken together, our data elucidates how both neuronal and network mechanisms contribute to ictogenesis, and how the process is further complicated by their interactions. By focusing on the mechanisms underlying ictogenesis, we hope to provide direction for clinical interventions that can stop an epileptic episode before it begins.