Abstracts

Rationale: Up to 30% of epilepsy associated ion channel genes demonstrate altered splicing, including AMPA receptor GluR subunits, which exist as mutually exclusive alternatively spliced flip or flop variants. GluR1 flip has higher glutamate sensitivity than GluR1 flop, thus increasing GluR1 flip would lead to network hyperexcitability and increased seizure susceptibility. Post-seizure increases in GluR1 flip are implicated in the pathogenesis of seizures and epilepsy, and reducing GluR1 flip levels should have potent antiepileptic and antiepileptogenic actions. This is particularly important in the neonatal period where GluR expression peaks developmentally. Splice modulating oligonucleotides (SMOs) are a class of compounds that bind to and block pre-mRNA regulatory sites to direct alternative splicing. Thus, re-direction of GluR1 pre-mRNA splicing away from flip production may provide a potent therapeutic option. We have developed a novel SMO that selectively and potently decreases GluR1 flip levels. Methods: For all experiments and treatment groups SMO or vehicle was delivered by free hand ICV injection into both lateral ventricles of FVB mouse pups on postnatal (P) days P1, P3, and P5, with evaluation at P10. We determined SMO dose-response for GluR flip and flop levels by real-time PCR in hippocampus and cortex. For all electrophysiology experiments, whole-cell patch clamp recordings of CA1 pyramidal neurons were performed on P10 hippocampal slices. Additionally, single dose (3mg/kg) and incremental kainate IP dosing paradigms were used to evaluate SMO effect on seizure severity, latency, and threshold. Results: ICV delivery of this SMO resulted in a dose-dependent down-regulation of GluR1 flip at P10 in cortex and hippocampus. SMO directed GluR1 flip reduction,decreased AMPA receptor-mediated excitatory post-synaptic currents (EPSCs) generated at Schaeffer collateral/CA1 synapses by about 40%, while NMDA-mediated EPSCs were unaffected. Increases in AMPA-EPSC amplitude and duration by cyclothiazide (a GluR flip specific modulator) were robustly depressed after SMO mediated knockdown of GluR1 flip, indicating GluR1 flip contributes most of the cyclothiazide-sensitive EPSC in neonatal mice. Decreasing GluR1 flip levels with our SMO provided strong protection against kainate-induced seizures in neonatal mice. Specifically, we demonstrate decreased seizure severity with SMO treatment after single kainate dose. Further, with incremental kainate dosing, SMO-treated mice required 76% more kainate to induce status epilepticus. Importantly, CA1 long term potentiation, a cellular model of hippocampal-dependent learning, was unaffected by decreasing the GluR1 flip levels. Conclusions: We developed a highly potent and specific modulator of GluR1 alternative splicing which provided robust neonatal antiseizure activity without affecting synaptic plasticity, suggesting it may not have anti-cognitive properties. Our results show that modulation of AMPA channel alternative splicing is a novel therapeutic strategy for preventing seizures and epileptogenesis.