Abstracts

This abstract has been invited to present during the Broadening Representation Inclusion and Diversity by Growing Equity (BRIDGE) poster session

Rationale: One of our long-term goals is to develop AAV-based therapies for temporal lobe epilepsy (TLE). Since intraparenchymal injection of AAV only leads to regional infection of cells, a detailed understanding of the epileptic circuits is required to inform injection strategies.  Studies by the Kapur lab have shown that TRAP2 reporter mice can be used to map evoked seizures at cellular resolution.  Therefore, we applied this method to map spontaneous seizures in epileptic TRAP2 mice.

Methods: Mice were engineered to express a tamoxifen-regulated Cre recombinase (Cre-ER) under the control of the Fos promoter. Crossing with a Cre-dependent reporter line, such as the tdTomato line Ai9 (henceforth TRAP2 reporter mice), allows mapping of the neural circuits activated by seizures. TRAP2 are on a mixed C57Bl/6 background, a strain that does not develop epilepsy after electrical kindling. We avoided TLE models that begin with status epilepticus as this leads to extensive neuronal death. Therefore, we developed a hybrid kindling (HK) approach that combines subconvulsant dosing of kainic acid with electrical stimulation of the hippocampus. We analyzed tdTomato expression in the following groups: controls (naive, vehicle), evoked seizures (electrical, kainic acid), and HK mice. Mice were implanted with an EEG headset plus stimulating electrode. Mice were IP injected with 10 mg/kg kainate and then electrically kindled. Video/EEG was recorded 24/7. 4-Hydroxy-tamoxifen (4OHT) was injected 90 min after the event to be TRAPed. Neurons were counted in 10x images of horizontal brain slices using Cellsens software (every 8th 30-um slice).

Results: TRAP2 mice have a very low background of tdTomato-labeled cells. Low background was also observed in epileptic mice that received the vehicle control. Mice TRAPed after 2-electrically evoked seizures showed robust tdTomato expression throughout the hippocampal trisynaptic circuit. Hybrid kindled mice TRAPed after the last evoked seizure and epileptic HK mice TRAPed after a spontaneous seizure, also showed robust expression throughout the trisynaptic circuit; however, there was also strong activation throughout the extended limbic system. Profound activation was observed in prefrontal cortex, lateral septum, midline thalamus, amygdala, and piriform cortex. Too many neurons expressed tdTomato to allow for axon tracing. Therefore, we analyzed the Allen Mouse Brain Connectivity Atlas to determine the main axonal projections from identified brain regions. Combining these two data sets led to a map of the epileptic circuit.

Conclusions: All stimuli that evoked electrographic seizures led to very strong activation of dentate granule cells, particularly in the septal hippocampus. Similarly, seizures led to strong activation of the subiculum, but in this case, predominantly in the temporal hippocampus. While discrete seizures activated mainly the hippocampus, fully kindled and epileptic mice showed extended activation of the limbic circuit. These studies suggest that targeting gene therapies to the output neurons of the entorhinal cortex may prevent a focal seizure from being convulsive. 

Funding: Supported by NS097726