Abstracts

Dravet syndrome (DS) is a severe developmental and epileptic encephalopathy characterized by early onset, with febrile and afebrile generalized tonic-clonic seizures and high mortality risk by sudden unexpected death in epilepsy (SUDEP). While epilepsy becomes less severe with disease progression, cognitive and behavioral symptoms persist.

DS is caused by haploinsufficiency of SCN1A gene, encoding Nav1.1 channel, essential for action potential generation and propagation in GABAergic interneurons (IN). IN hypoexcitability is considered the initial trigger of the disease, leading to circuit hyperexcitability. Because of this crucial role in DS pathogenesis, IN are the main target of current gene therapy strategies aimed at increasing SCN1A expression. However, evidence from animal models suggests that IN firing defect normalizes in DS chronic phase, dominated by secondary circuit modifications, raising the question of DS symptom reversibility.

We previously generated a DS mouse model carrying a STOP cassette in the Scn1a locus (Scn1a^Stop/+) that could be conditionally removed by Cre recombination. We showed that ubiquitous restoration of Nav1.1 physiological levels, both perinatal and postsymptomatic, ameliorated epileptic and behavioral symptoms. However, whether specific cell types should be preferentially targeted at different disease stages remains unclear.

Here, we took advantage of the Scn1a^Stop/+ model to  determine whether a gene therapy targeting only IN could be sufficient to ameliorate DS phenotype. To this aim we crossed Scn1a^Stop/+ mice with either constitutively active or tamoxifen-inducible GABAergic Cre drivers (Gad2Cre and Gad2CreER) to control the timing of Scn1a gene re-expression in INs.

Scn1a^Stop/+;Gad2Cre mice with constitutive Nav1.1 restoration in IN show a striking phenotypic amelioration with reduced seizure burden, nearly abolished susceptibility to thermally induced seizures and rescued survival up to postnatal day (P) 60. However, mild epileptic events persist during the active phase of circadian rhythm, with alterations in power spectral density and incomplete recovery of Θ-γ coupling. Intriguingly, IN firing properties and inhibitory synaptic transmission in the hippocampus are not fully rescued either, suggesting the presence of  subtle defects that cannot be corrected by an IN specific therapeutic intervention.

To mimic the effect of a post-symptomatic IN specific gene therapy, we restored Nav1.1 levels in Scn1a^Stop/+; Gad2CreER mice at P60. Preliminary data show no improvement in spontaneous seizures up to one month post-correction and slightly increased susceptibility to thermally induced seizures with respect to control, suggesting that compensatory mechanisms established during disease progression might hinder therapeutic efficacy.

Our findings underscore the importance of timing in gene therapy and highlight the limitations of targeting only IN population in a complex network, pointing to the need for broader approaches to rebalance neural circuits in DS.