Development of a Transgenic Rabbit Model of Dravet Syndrome

Abstract

Rationale: Sudden Unexpected Death in Epilepsy (SUDEP) is a leading cause of death in patients with epilepsy. While SUDEP mechanisms are not understood, there is evidence to implicate apnea, autonomic dysfunction, and cardiac arrhythmias. Loss-of-function variants in SCN1A are linked to Dravet syndrome (DS). DS patients have the highest SUDEP risk, up to 20%. There are few effective therapies for any of the genetic epilepsies and no reliable biomarkers for SUDEP risk. Importantly, SCN1A is expressed in both heart and brain in humans and rodents. Because of this, we proposed that cardiac arrhythmias contribute to the mechanism of SUDEP in DS. Scn1a+/- DS mice have increased cardiac myocyte sodium current density, action potential (AP) prolongation, increased incidence of delayed afterdepolarizations, and cardiac arrhythmias. Induced pluripotent stem cell (iPSC)-derived cardiac myocytes from DS patients have substrates for arrhythmia, including increased sodium current density, rates of spontaneous contraction, and incidence of early- and late-afterdepolarizations. However, no mouse or iPSC model can completely replicate the human DS phenotype.

Methods: Rabbits more closely replicate the human cardiac AP than mice and, unlike human iPSCs, provide a complete organism to translate to the clinical setting. Thus, we worked with the University of Michigan Center for Advanced Models for Translational Sciences and Therapeutics to generate a New Zealand White (NZW) rabbit Scn1a deletion model using CRISPR-Cas9 gene editing. Rabbits were tested using video EEG/ECG telemetry as well as surface ECG under anesthesia. Optical mapping studies were performed in isolated hearts and patch clamp electrophysiology was performed in acutely isolated cardiac myocytes.

Results: We generated transgenic NZW rabbits with a germline 14 base pair deletion in exon 1 of Scn1a, resulting in a frameshift/truncation mutation. NZW Scn1a-/- kits died by postnatal day (P) 7-9 and showed evidence for differences in cardiac rhythm, including bradycardia, long QT intervals, and AP prolongation. Cardiac myocytes isolated from NZW Scn1a+/- kits showed increased late sodium current compared to WT. We did not observe behavioral seizures or arrhythmias in NZW Scn1a+/- rabbits, which live normal life spans and breed normally. These data suggested, like mouse models, that the DS rabbit phenotype may be background strain dependent. To test this hypothesis, we crossed Scn1a+/- NZW rabbits to the Dutch Belted (DB) strain. We found that F1:NZW x DB Scn1a+/- rabbits have spontaneous seizures as well as pre- and post-ictal cardiac arrhythmia as assessed by video/EEG/ECG telemetry. Approximately 22% of the kits undergo premature death. Severe seizures are observed in adult animals, but to date no adults have died during seizures. Optical mapping experiments showed increased propensity to ventricular arrhythmia which were characterized by spiral wave reentry in isolated F1:NZW x DB Scn1a+/- hearts.

Conclusions: F1:NZW x DB Scn1a+/- rabbits model DS in terms of seizures and SUDEP and are an optimal model to investigate neuro-cardiac mechanisms of SUDEP as well as to develop novel therapeutics for DS. To our knowledge, this is the first transgenic large animal model of a developmental and epileptic encephalopathy.