Abstracts

Rationale: Acquired epilepsies are characterized by spontaneous, recurrent seizures that emerge following injury. Understanding seizure initiation, or ictogenesis, is critical for developing treatments for medically refractory epilepsy. However, ictogenesis is difficult to study, due to the spontaneous nature of seizures and poor spatial sampling available for capturing seizure onset in vivo. Here, we developed novel techniques for chronic cellular resolution imaging of neuronal activity in the hippocampal organotypic slice culture model of post-traumatic epilepsy to analyze seizure onset at the single neuron level.

Methods: Hippocampal slice cultures generate spontaneous seizure-like events following slicing injury. With this preparation, it is feasible to image the entire network, ensuring that the earliest pathological activity is captured (as no outside inputs exist). Pan-neuronal GCaMP7 expression was paired with DLX-cre-targeted tdtomato labeling of interneurons to separate the calcium activity of principal cells and interneurons. We used a custom imaging system to observe undisturbed organotypic hippocampal slice cultures over several weeks to capture the calcium activity across dozens of seizures. With this approach, the activity of hundreds of individual neurons could be tracked over time. We used this preparation to test whether a subpopulation of neurons consistently initiates seizure activity.

Results: Electrophysiological recordings and calcium imaging demonstrated 3 unique types of seizure onset in this preparation: low amplitude fast, sentinel spike and spike burst + low amplitude fast activity onset. These patterns recapitulate common features of human seizure onset, including low voltage fast activity and spike discharges. Weeks-long imaging of seizure activity showed a characteristic evolution in onset type and a refinement of the seizure onset zone. Our cellular level tracking revealed that neuronal onset sequences were highly variable seizure-to-seizure and argues against seizure initiation by a small number of neuronal drivers, such as hub cells.

Conclusions: Using novel network-wide, cellular resolution activity imaging, we reveal that seizure onset is stochastic at the single neuron level. In other words, the neuronal activation order in sequential seizures is highly variable, which suggests that no stereotyped onset sequence exists at the cellular level. Transitions to seizures may therefore be driven by state changes that allow for re-entrant seizure activity in non-stereotyped sequences.

Funding: Please list any funding that was received in support of this abstract.: NIH-NINDS.