Development of a Genetic Rabbit Model of Kcnh2-Mediated Epilepsy, SUDEP, & Long QT Syndrome Type 2

Sudden Unexpected Death in Epilepsy (SUDEP) is the leading cause of death in epilepsy. Many SUDEP cases
had genetic variants linked to cardiac arrhythmias, particularly Long QT Syndrome (LQTS). Our lab and others
showed that LQT2 patients are at a >2-fold higher risk of epilepsy, compared to genotype negative family
members. LQT2 is caused by KCNH2 variants that encode a K+ channel protein (Kv11.1), which produces K+
current (IKr). It is critical for cardiac repolarization, stabilizes the neuronal resting membrane potential, and
suppresses repetitive firing. People with KCNH2-mediated epilepsy and LQT2 need safe and effective anti-
seizure medications (ASMs). A critical barrier is the lack of a clinically relevant animal model of LQT2 with
epilepsy. We will develop the first translational model of Kcnh2-mediated epilepsy, SUDEP, and LQT2. As LQT2
patients with KCNH2 pore-domain variants are at the highest risk of seizures, we used CRISPR-Cas9 to generate
rabbits with a heterozygous frameshift mutation in the Kcnh2 pore-domain (Kcnh2(+/mut)). This model has superior
construct validity. Due to rodent vs. human differences in cardiac electrical function, and unphysiological Kcnh2
expression patterns, present models are not appropriate for translational studies of Kcnh2-mediated neuro-
cardiac pathologies. Many rodent models fail to reproduce the natural progression of clinical epilepsy, include
non-seizure related neuronal damage, require triggers that are not physiologically relevant, and have low
predictive validity for ASM screening. Rabbits are established models for drug testing, and studying seizures,
arrhythmias, and sudden death. In contrast to rodents, neuronal cell-types are similar in humans and rabbits.
R61: External Face Validation: Using quantifiable and clinically relevant endpoints, we will test if Kcnh2(+/mut)
rabbits reproduce the neuro-cardiac pathologies seen in LQT2 patients with epilepsy. Preliminary data indicates
that we generated a clinically-relevant rabbit model of Kcnh2-mediated epilepsy, SUDEP, and LQT2. There is
reduced Kv11.1 expression in the brain and heart, QTc prolongation, spontaneous noncardiogenic epileptic
seizures, and spontaneous seizure-mediated sudden death. R33: Despite LQT2 patients being at a high risk of
epilepsy and SUDEP, there are no established ASMs for LQT2 patients with epilepsy. We demonstrated that
LQT2 patients are at an increased risk of arrhythmias when on vs. off ASMs, particularly Na+ channel blocking
ASMs (e.g., phenytoin). External Face Validation: (1) We will demonstrate that myocytes and cortical neurons
from Kcnh2(+/mut) rabbits have reduced IKr and are hyperexcitable. (2) Similar to LQT2 patients, cellular and in
vivo assays will test if phenytoin has adverse cardiac effects in Kcnh2(+/mut) rabbits. Preliminary data indicates
phenytoin blocks IKr and causes a larger increase in QTc in Kcnh2(+/mut) vs. WT rabbits, which suggests predictive
validity of our model. Impact: We will develop a model of Kcnh2-mediated epilepsy, SUDEP, and LQT2 that
reproduces human LQT2 neuro-cardiac pathologies. It will provide a platform for identifying effective and safe
ASMs to reduce seizures and SUDEP in LQT2, and complements our research using the LQTS patient registry.

Grant Number: 5R33NS133273-04
NIH Institute/Center: NIH
Principal Investigator: David Auerbach