Abstracts

Rationale: Focal neocortical epilepsy forms a network of synaptically connected nodes and exhibits drug resistance in about 25% of patients, of which only about 50% achieve seizure-freedom with surgery. To better understand how seizures initiate and spread in cortical networks, we developed an experimental paradigm that permits simultaneous electrographic and mesoscopic calcium imaging of Thy1-positive excitatory or parvalbumin-positive (PV) inhibitory neurons across a defined bilateral neocortical network.


Methods: We injected 4-Aminopyridine (4-AP, 2 mM), a chemoconvulsant, into primary somatosensory cortex (S1), a site with known connections to ipsilateral secondary motor cortex (iM2) and contralateral S1 (cS1). This was performed in two strains of awake mice, one expressing activity dependent fluorescence in Thy1 excitatory neurons (n=6 mice, 19 seizures) and the other in PV inhibitory neurons (n=6, 24 seizures). Following injection of 4-AP, we observed robust focal seizures at S1 that exhibited variable but non-stochastic recruitment of monosynaptically connected nodes (n=12 mice). To test a mechanism of this propagation pattern, we used electrical microstimulation to probe the difference in excitatory (n=5 mice) and inhibitory activity (n=5 mice): stimulus trains of 0.5 mA of current were delivered to S1 for either 100-ms, 1-sec, or 10-sec (50 Hz) during widefield imaging of Thy1 and PV neural activity across the network. Finally, the application of the sodium-channel blocker tetrodotoxin (5 uM) in iM2 was used to determine the importance of this secondary network node in bilateral seizure propagation.


Results: Despite strong anatomical connection between S1 and cS1, seizures never (0/43) propagated to cS1 without first recruiting iM2, and rarely (2/43) propagated to cS1 without first recruiting the contralateral M2 (cM2), a site with little connection to S1. A critical role of iM2 in contralateral propagation was confirmed by in vivo tetrodotoxin ablation of this region, which significantly reduced calcium responses in the contralateral hemisphere. Based on differences in Thy1 and PV cell imaging, we find that cross-callosal excitation/inhibition balance varies between the nodes of the S1-M2 network. This effect that was also reflected in our microstimulation experiments, where PV activity was significantly higher in cS1 compared to cM2 (based on change in fluorescence change during stimulation, ttest p< .01 for 1- and 10-sec stimulation, p< .05 for 100 ms), and Thy1 activity was significantly higher in cM2 compared to cS1 (p< .001 for 1- and 10-sec stimulation, p< .05 for 100 ms).


Conclusions: Our findings demonstrate that propagation patterns of focal to bilateral neocortical seizures are guided by excitatory/inhibitory balance, which varies across the corpus callosum. Our data also highlight that specific network nodes outside of the seizure onset zone, in this case iM2, may serve as targets to control seizure propagation. These findings will guide subsequent studies examining how different cell types and brain regions outside of the ictal focus can be manipulated to constrain or prevent focal neocortical seizures.


Funding: None