Abstracts

Rationale: Dravet syndrome (DS) is a severe, intractable, epileptic encephalopathy with early onset in the first 6-18 months of life. Since the first description of DS 40 years ago, there has been considerable progress in determining the etiopathology of this devastating disorder. Remarkably however, there are few studies that describe the earliest perinatal mechanisms underlying DS pathogenesis. To this end, we characterized a new DS model carrying a missense mutation in scn1lab, the zebrafish orthologue of SCN1A, the primary gene mutated in DS that encodes the voltage-gated sodium channel alpha subunit 1.  Methods: We used CRISPR/Cas9 mutagenesis to introduce a mutation that perturbs ion transport function in all channel isoforms. We then performed automated behavioral tracking and local field potential recordings to measure convulsions and epileptiform discharges in larvae as previously described (1), followed by single-cell RNASeq, morphometric analysis of transgenic reporter labeled GABAergic neurons and pharmacological profiling of scn1lab mutant larvae.(1) Afrikanova T et al. Validation of the zebrafish pentylenetetrazol seizure model: locomotor versus electrographic responses to antiepileptic drugs. PLoS One2013; 8(1): e54166.  Results: Electrographic recordings showed that homozygous mutant (scn1labmut/mut) larvae display spontaneous seizures with clear interictal, pre-ictal and ictal discharges from 4 days post-fertilization (dpf) (Mean no. of seizures = 7 per 20-min recording; ****P<0.0001; significance calculated using one-way ANOVA with Tukey's post-hoc test). DropSeq analysis revealed a 2:1 shift in the ratio of glutamatergic (excitatory) to GABAergic (inhibitory) neurons in scn1labmut/mut larval brains compared to wild type (WT) that resulted from a progressive loss of GABAergic neurons over the course of development. Additionally, dynamic changes in the proportion of neuronal, glial and progenitor cell populations were observed. To explore disease pathophysiology further in scn1labmut/mut larvae, we quantified dendritic arborization in GABAergic neurons and identified a 40% reduction in arbor number compared to WT 40% (P<0.001; n=15 mutant, n=16 WT). We hypothesize that the significant reduction in inhibitory arbors causes an inhibitory to excitatory neurotransmitter imbalance that can account for seizure onset and an enhanced electrical brain activity level measured in scn1labmut/mut larval brains (in particular, in the range of high frequencies), with eventual loss of GABAergic neurons and astrogliosis due to excitotoxicity. Early chronic exposure to fenfluramine (FEN), a serotonin-releasing agent and 5-HT2 receptor agonist, completely restored the number of dendritic arbors to normal in scn1labmut/mut larvae, whereas similar treatment with the benzodiazepine diazepam reduced seizure frequency, but had no effect on the observed structural deficits. Furthermore, BrdU labeling revealed increased cell proliferation in the brains of scn1labmut/mut larvae that were also rescued by chronic FEN treatment.  Conclusions: In summary, our findings provide insight into the very early mechanisms of DS pathogenicity, describe new observations on neuronal and glial cell population changes in the scn1labmut/mut brain, present novel evidence for disease modifying activity of FEN and validate a new in vivo model for elucidating mechanisms underlying epileptogenesis. Funding: CVE, NCMM startup; KG, Marie Curie 798703-GEMZ-H2020-MSCA-IF-2017; AS, National Research Fund Luxembourg C14/BM/7975668/CaSCAD; INTER/DFG/17/11583046 MechEPI.