Abstracts

Rationale: Epilepsy is the 4^th most prevalent neurological condition with 50 million cases worldwide. Despite this, there is a poor understanding of the cause of epilepsy-associated cognitive decline. As a result, many anti-epileptic medications effective for seizure suppression inadvertently exacerbate comorbidities in patients with co-existing epilepsy-related cognitive decline. The mechanism driving the development and persistence of epilepsies remains poorly understood, and no disease modifying therapeutics – those that would reduce, or reverse seizure burden and cognitive decline in epilepsy patients - exist. Previous data identified STAT3 as a driver of upregulated genes across rodent models of epilepsy and human temporal lobe epilepsy. Recent research pointing to the role of inflammation in compromising cognition in diseases such as Alzheimer’s, Huntington’s disease, and age-associated cognitive decline. We aim to test whether we can target STAT3 via its upstream JAK kinases using the FDA approved drug CP690550 to enduringly rescue cognitive decline.


Methods: Bioinformatic analysis of microarray data from human patients with temporal lobe epilepsy used MAGIC to identify clusters of genes and their associated master regulators. Ontological analysis was used to obtain descriptions of the functions of gene products for given gene clusters. We utilized molecular and biochemical techniques to validate the levels of STAT3 post status epilepticus (SE) at acute and chronic time points to validate our therapeutic windows for CP690550 treatment. We utilized spontaneous and forced alternation tests to monitor deficits in working and short-term memory after seizures, as well as the ability of CP690550 to rescue decline. We applied Motion Sequencing (MoSeq), a machine learning platform combined with 3D video analysis, to holistically find ‘hidden’ behavioral differences at a frame-by-frame rate.


Results: Using the forced alternation test and behavioral seizure recording, we found that attempts to target the transient activation of STAT3 immediately post-SE fail to suppress spontaneous seizures and prevent declines in short-term memory. However, we discovered that STAT3 reignites during spontaneous recurring seizures in our mouse model of epilepsy, opening a second, potent therapeutic window for treatment. We show that two weeks of treatment with CP690550 during chronic epilepsy enduringly rescues working and short-term memory for at least two weeks post drug washout. Furthermore, using MoSeq, we show a general, holistic rescue of behavioral alterations after two weeks of treatment with CP690550.


Conclusions: We used data-driven hypothesis generation to identify STAT3 as a driver of disease in both acute and chronic epilepsy. We then showed that the reignition of the JAK/STAT signaling pathway in chronic epilepsy opens a second, potent therapeutic window targetable using the FDA approved drug, CP690550, to enduringly rescue epilepsy-associated comorbidities.


Funding: Supported by CURE (AR), Lily’s Fund (AR), NIH grant T32 GM14103 (JK), 1R01NS108756 (AR, RD) and R01NS105628, R01NS102937, and R21NS120868 (JM).