Abstracts

Rationale: Impaired consciousness during limbic seizures has a large impact on patient quality of life. However, the neural circuit responsible for this alteration of brain function is still poorly understood. During temporal lobe seizures, brain imaging shows cortical deactivation, coupled with sleep-like slow wave electrical activity and reduced cholinergic neurotransmission. We previously demonstrated that the lateral septum (LS), a putative inhibitory region strongly connected to hippocampus, could induce both cortical slow waves and decrease of cholinergic neurotransmission during electrostimulation. Because most cholinergic input to cortex is provided by the nucleus basalis (NB), we propose that focal seizures originating from the hippocampus depress cholinergic neuronal activity in NB via subcortical pathways (eg. LS), resulting in cortical depression. Methods: We first examined the functional implication of cholinergic neurons from NB during induced partial seizures in hippocampus in a lightly anesthetized rat model by applying an optogenetic approach to restore cholinergic arousal. Then, we used a combination of retrograde and anterograde tracing to identify the anatomic network between hippocampus, subcortical regions and NB. Finally, to determine the neuronal activity profile during focal seizures from subcortical regions of interest, we measured multiunit activity (MUA) in vivo. Results: We found that cortical slow waves dramatically convert to fast waves in response to optogenetic cholinergic stimulation during partial seizure (n=4), which supports the hypothesis that cholinergic output from NB plays a critical role in maintaining cortical arousal. We showed no direct anatomical connection between LS and NB (n=4). However, strong output from LS and input to NB was observed in the paratenial thalamic nucleus (PT) (n=2), located among the midline thalamic nuclei known to participate in arousal. MUA recordings in PT revealed a decrease of neuronal firing during seizures (n=7), which support an inhibitory effect from LS output and suggests a reduced excitatory effect on NB. Conclusions: These results show that cortical depression induced by limbic partial seizure can be facilitated by disruption of a polysynaptic subcortical network including LS, PT and NB. Further investigation of this network may provide new leads for therapeutic interventions to improve cortical function and consciousness during and following seizures. Funding: NIH R01 NS066974 and R01 NS096088