Authors :
Presenting Author: Divyalakshmi Soundararajan, MS – Columbia Vagelos College of Physicians & Surgeons
Gabriel Jimenez, BS – Columbia University Vagelos College of Physicians & Surgeons
Scott Harper, PhD – Nationwide Children's Hospital
Wayne Frankel, PhD – Columbia University Vagelos College of Physicians & Surgeons
Tristan Sands, MD, PhD – Columbia University Vagelos College of Physicians & Surgeons
Rationale:
Pathogenic gain-of-function GRIN2A variants cause developmental and epileptic encephalopathy (DEE) with seizures and cognitive impairment. In Grin2a S644G mice, homozygotes develop lethal seizures before weaning, while heterozygotes display hyperactivity, anxiety-like behavior, and altered sensory responses. Available pharmacologic interventions offer little benefit, highlighting the need for targeted therapies. RNA interference (RNAi) provides a mutation-agnostic approach to reduce pathogenic GRIN2A expression, potentially mitigating toxic gain-of-function effects without variant-specific customization. We tested whether neonatal intracerebroventricular delivery of AAV9 carrying a miRNA against mouse and human GRIN2A could rescue survival and improve neurological outcomes in this Grin2a S664G preclinical model.
Methods:
A miRNA (miG2a) targeting conserved regions of mouse and human GRIN2A was cloned into a self-complementary AAV9 vector under a U6 promoter. Neonatal Grin2a S644G and wild-type littermates received ICV injections at different doses at postnatal day 1 (P1). Treatment effects were assessed using survival analysis, 6 Hz seizure threshold testing, open-field locomotion, acoustic startle response, repetitive behavior quantification, and gait analysis. Molecular effects were evaluated by qPCR measurement of Grin2a and Grin2b mRNA from brain tissue.
Results:
In homozygous Grin2a S644G mice, miG2a treatment at all tested doses extended survival beyond the typical lethal seizure window. One treatment dose restored 6 Hz electroconvulsive threshold (ECT) values to wild-type levels. Open-field testing revealed sex-dependent effects on locomotor activity; however, miG2a treatment did not reverse the hyperactivity phenotype in Grin2a S644G mice. In the acoustic startle test, miG2a attenuated startle responses regardless of genotype, indicating a treatment-specific effect independent of the underlying mutation.
Conclusions:
Postnatal AAV9-mediated delivery of miG2a extended survival and showed promising restoration of seizure thresholds in Grin2a S644G mice, but had limited effects on behavioral phenotypes. While these findings indicate that RNAi can modulate some disease-relevant outcomes, the partial efficacy highlights the need for further optimization of dose, timing, and targeting strategies before considering broader translational applications.
Funding:
CureGRIN Foundation