Abstracts

Rationale: Many individuals with temporal lobe epilepsy (TLE) show sparing of area CA2 relative to area CA3 and CA1, and past studies suggest that area CA2 may promote seizure activity in TLE and animal models of TLE. Therefore, we hypothesized that silencing CA2 neurons selectively would reduce seizures in vivo. To test this hypothesis, we used a common mouse model of TLE where pilocarpine is first injected to produce status epilepticus (SE). To selectively examine the role of CA2, SE was induced in Amigo2-Cre+/- mice, where Cre is expressed relatively selectively in CA2 pyramidal cells. Cre-dependent AAV (AAV2/5-hSYN-DIO-hM4D(Gi)-mCherry) was injected bilaterally in dorsal CA2 to express hM4D(Gi) (inhibitory designer drugs activating designer receptors; iDREADDs), tagged with mCherry. During video-EEG of chronic seizures, iDREADDs were activated by clozapine-N-oxide (CNO) in the drinking water. Comparisons were made to water without CNO, Amigo2-Cre-/- mice, and other controls.

Methods: Adult male and female Amigo2-Cre+/- and -/- mice were injected with pilocarpine to induce SE. Four weeks later, after mice exhibited spontaneous seizures, they were injected stereotaxically in dorsal CA2 bilaterally with AAV2/5-hSYN-DIO-hM4D(Gi)-mCherry or AAV2/5-hSYN-DIO-mCherry. Immediately afterwards, mice were implanted with 4 subdural electrodes over the dorsal hippocampus bilaterally, left frontal cortex and right occipital cortex. After 4 weeks, mice were recorded for 6 weeks using continuous video-EEG with 10 mg/kg CNO during one 3 week-long period and without CNO during the other 3 week-long period. Order was randomized. Afterwards mice were perfusion-fixed to confirm expression of virus in CA2. In addition, 18 mice were perfusion fixed >3 weeks after viral injection to analyze expression. An additional cohort of mice were tested with both 5 mg/kg and 10 mg/kg CNO and effects on seizures could not be distinguished.

Results: In this model, most seizures were convulsive (stages 4-5 on the Racine scale) and were evident at all 4 recording sites. Seizures were frequent, typically occurring up to 5x/day. Chronic convulsive seizure frequency was significantly reduced by CNO in iDREADD-expressing Amigo2-Cre+/- mice relative to the same mice that drank water without CNO, and this was true in males and females (two-way ANOVA, treatment and sex as main factors, F(1,16)14.2, p=0.002), with no interaction of factors. There were no statistically significant effects of CNO on convulsive seizure duration, the latency from seizure onset to the convulsion, or the number of seizure-free days.

Conclusions: The results support a role of area CA2 in the regulation of seizures in a mouse model of TLE. Inhibiting CA2 in the chronic period reduced seizure frequency but when a seizure occurred, its duration and severity appeared to be unchanged. These data suggest CA2 silencing limits seizure propagation through hippocampus. The results suggest that ultimately CA2 might be targeted as a therapeutic strategy in TLE.

Funding: Please list any funding that was received in support of this abstract.: Supported by NIH R01 NS106983.