Abstracts

Rationale: The stargazer mouse model of absence epilepsy has an AMPA receptor trafficking defect in parvalbumin-expressing interneurons in the neocortex. These fast-spiking interneurons are implicated in the generation of gamma oscillations between 30-100 Hz. We have previously shown that the peak change in interictal relative gamma power has an inverse relationship with the total duration of seizures after administering anti-epileptic drugs in stargazer mice. Here we examine the temporal dynamics of interictal gamma oscillations with respect to spike-wave seizure onset and offset. Methods: Stargazer and wild type mice were implanted epidurally with silver wire electrodes over bilateral somatosensory cortex, and analysis was performed on 51 frequencies between 2-100 Hz in the electroencephalogram during a 30-60 minute period after the start of the recording. Seizures were digitally removed and only interictal data points were analyzed. Relative power (RP) at any given frequency was defined as the absolute power (AP) at that frequency divided by the total power (TP; RP=AP/TP). Power and coherence analyses were performed in EEGLab (Matlab). All statistical analyses were performed using a 1- or 2-way repeated measures ANOVA with Bonferroni correction for multiple comparisons, and significance was set at p < 0.05. Results: Six stargazer mutants aged P55 and above (mean number of seizures per mouse = 34) were analyzed for temporal changes in relative gamma power with respect to seizure onset and offset. During a one second interval prior to seizure onset (S-1), compared to the fifth second prior to seizure onset (S-5), the peak change in relative gamma power was at 70 Hz, reduced by 32.1% (p < 0.001). During the one second interval immediately after seizure offset (S+1), compared to (S-1), the peak change in relative gamma power was again at 70 Hz, but increased by 33.9%, returning to baseline gamma power (p < 0.001). Absolute power at 70 Hz was also significantly reduced in a similar pattern with a return to baseline after seizure offset. Interhemispheric phase coherence was also evaluated at frequencies between 2-300 Hz with no significant difference in interictal coherence at any frequency between WT and stargazer mutants, before and after administering ethosuximide, and in five second windows before and after seizure onset and offset, respectively (p>0.05). Conclusions: Here we show a distinct reduction in relative gamma power with no change in interhemispheric phase coherence just prior to seizure onset in a mouse model of absence epilepsy. These findings are in contrast to the dynamics of gamma oscillations reported prior to the onset of focal seizures, which have been shown to increase in power and phase coherence prior to seizure onset, providing a framework for understanding the differences in cellular and network processes that underlie focal- versus generalized-onset seizures. Funding: NINDS K08 NS096029-01