ABSTRACT
Sudden unexpected death in epilepsy (SUDEP) is the leading cause of death in patients with refractory
epilepsy. SUDEP is second only to stroke in years of potential life lost to neurological disease and is a
major public health problem. Several etiologies have been proposed for SUDEP including cardiac and
respiratory dysregulation. Another that is postulated is impaired arousal. Seizures impair arousal. Among
arousal stimuli, one that may be particularly relevant to SUDEP is CO2. CO2 rises following seizures and
is part of the seizure cessation mechanism. Seizures are frequently associated with ictal and post-ictal
central and obstructive apneas. Apnea further exacerbates the accumulation of CO2. Impairment of CO2-
arousal is proposed as an etiological factor in another sudden death entity, sudden infant death
syndrome, which has many parallels with SUDEP. We discovered that seizures impaired CO2-arousal in
seizure naïve mice and those that do not have a particularly profound death rate. Whether this is true in
mouse models with strong SUDEP phenotypes is unknown. Thus, our goal in this proposal is to determine
how seizures in different sleep states impair CO2-arousal in mouse models of temporal lobe epilepsy
(TLE) and the genetic epileptic encephalopathy, Dravet Syndrome (DS). In Aim 1 we will determine the
extent to which CO2-arousal is impaired in epilepsy models. We will focus on the pilocarpine-TLE model
as many patients that die of SUDEP have TLE, and the DS model, as patients with DS have a
disproportionately high SUDEP risk. We will also determine whether simply having epilepsy in these
models impairs CO2-arousal as a potential easily measurable biomarker for SUDEP risk. In Aim 2 we will
determine whether neuronal function, assessed via fiber photometry, of arousal system components in
the dorsal raphe nucleus and parabrachial nucleus, two important contributors to sleep-wake regulation
and key nodes in CO2-arousal, is impaired by seizures and epilepsy. In Aim 3 we will determine whether
optogenetically stimulating a DRN-PBN circuit in DS mice, using a novel mouse model, prior to seizures
prevents seizure-induced death lending direct insights into possible therapeutic measures. Since the
models employed have known death rates, we will be able to compare findings between mice that die
and those that survive making these studies more relevant to SUDEP. Combining these findings with our
previous work, we will have a powerful, rigorous, translatable approach to identify convergent and
divergent mechanisms across models for how impaired CO2-arousal in epilepsy contributes to SUDEP
risk. We expect to be able to leverage these mechanisms to identify at-risk individuals and reduce death
from this devastating disease.