Abstracts

Rationale: Homozygous recessive loss-of-function variants in SCN1B, encoding the voltage-gated sodium channel b1/b1B subunits, are linked to Dravet syndrome (DS). The Scn1b null mouse model of DS recapitulates key DS patient phenotypes including severe seizures, developmental delay, ataxia, and early lethality.  Electrophysiological studies showed abnormal neuronal excitability in brain slices from Scn1b null mice.  Here, we compared the differences in neuronal excitability in human induced pluripotent stem cell (iPSC) neurons derived directly from SCN1B p.R89C DS or control patients. Methods: Whole exome sequencing was performed for the proband, healthy parents, and healthy sibling under IRB. Mapping of the reads to the reference genome was done using Burrows-Wheeler Aligner. De novo variants were called using DeNovoGear (DNG) and the generated list of variants was filtered using the criteria: read depth in all individuals = 8; allele balance in the proband between 0.25 and 0.75 and in the parents = 0.95; exclusion of variants in tandem repeats and segmental duplications; posterior probability of de novo calling of DNG = 0.5; and exclusion of variants seen in > 1 individual. No de novo variants were identified in the proband. The dataset was further filtered using the criteria: read depth in all individuals = 8; allele balance in the proband = 0.95 and in the parents between 0.25 and 0.75 for filtering under a homozygous model and between 0.25 and 0.75 in the proband and parents for the compound heterozygous model; exclusion of variants in tandem repeats and segmental duplications; and a frequency of = 1% in control databases. These criteria identified a homozygous variant in SCN1B, c.265C>T, p.R89C (NM_199037), in the proband, with parents and sibling as heterozygous carriers. The variant was seen twice in the Exac database (allele frequency 1.647x10^-05) and was validated with Sanger sequencing. Dermal fibroblasts obtained from the proband under IRB were reprogrammed into iPSCs and differentiated to excitatory cortical-like neurons. The electrophysiological properties of these neurons were analyzed and compared to iPSC neurons differentiated in parallel from a healthy control. Results: Peak transient sodium current (INa) density evoked by a test pulse to -30 mV from a pre-pulse of -120 mV was increased in SCN1B p.R89C vs. control (-193±25 pA/pF control, -287±25 pA/pF SCN1B p.R89C (p=0.003)). There were no significant differences in INa voltage-dependent properties. Current-clamp recordings showed no significant differences in resting membrane potential, capacitance, input resistance, or maximum firing frequency between genotypes of APs evoked by current injections from -60 pA to 180 pA.  However, SCN1B p.R89C neurons had a lower threshold for initiating APs compared to controls. The input/output relationships showed differences in AP firing frequency between genotypes in a stimulation intensity-dependent manner. SCN1B p.R89C neurons showed increased repetitive firing frequency evoked by depolarizing current injections from 50 pA to -90 pA (P < 0.05, control, n = 41; SCN1B p.R89C, n= 47).  A subset of SCN1B p.R89C neurons had abnormal AP waveforms with wider duration and slowed decay kinetics and fired spontaneously. Conclusions: These data suggest that increased transient INa may contribute to pathogenic mechanisms in SCN1B-linked DS. Funding: Supported by R37 NS076752 (LLI).