Rationale:
Epilepsy affects about 1% of the US population and nearly a third of epilepsy patients fail to respond adequately to anti-epileptic drugs. An understanding of how seizures start and spread through the brain is likely critical to developing new therapeutic options. Epilepsy is increasingly considered a network disorder in which multiple connected brain regions interact in a way that permits seizures to propagate out from an initial focus, sometimes generalizing to the whole brain. This propagation may also be underpinned by differences in excitatory and inhibitory drive, a relationship that may also differ between brain regions. To address this question, an experimental paradigm is needed that permits simultaneous cell-type-specific imaging of individual seizures as they propagate through brain regions with known connectivity.
Methods:
Widefield calcium imaging was performed in transgenic mice expressing GCaMP6f in either Thy1+ cells (putative excitatory pyramidal neurons) or PV+ inhibitory cells. Following cranial window implantation and recovery, awake mice are injected with 4-Aminopyridine (4AP, 2.5 mM) into a focal neocortical site in primary somatosensory cortex (S1). This site was chosen based on data from the Allen Institute showing strong connectivity with ipsilateral secondary motor cortex (M2) and contralateral S1, providing a clearly defined anatomical network to observe how excitation and inhibition vary spatiotemporally throughout seizures. Large cortical windows permit clear imaging and electrophysiology of these network sites on both hemispheres of the brain.
Results:
Prior to seizure onset in the electrographic data, neural activity of both excitatory and inhibitory cells increases at the seizure focus, though there was no significant difference between these cell types. After ictal onset, seizures propagated in one of two ways: some seizures propagated contiguously, spreading outward from the focus, and some seizures spread first to the well-connected ipsilateral M2 site. Next, some seizures spread bilaterally. Interestingly, when this occurred, our imaging data revealed that the cross-hemisphere propagation was nearly always first through the frontal cortex and never through posterior connections. Contralateral S1, despite receiving direct callosal projections from our ictal focus, was recruited significantly later than cM2, a di-synaptic node, as measured in both PV and Thy1 calcium activity.
Conclusions:
The observed difference in seizure propagation between contralaterally connected brain nodes suggests that callosal projections may vary in function depending on the initial source (such as S1 vs M2). This may depend on the relative amounts of excitation vs inhibition at the propagation sites or the gross number of cross-callosal connections, or both. Regardless of the mechanisms, this finding highlights how specific choke-points, in our case ipsilateral M2, may exist in different seizure networks and provide potential targets for surgical or cell-type-specific intervention to constrain seizure spread.
Funding: None