Self-limited Neonatal Epilepsy associated with global developmental delay and high-threshold sensory profile: A novel KCNQ3 de novo variant unusually located in the voltage sensor S4 segment of the Kv7.3 channel subunit

Background: Pathogenic variants in KCNQ2 and KCNQ3 are established causes of self-limited neonatal epilepsy (SeLNE). While over 300 KCNQ2 variants have been reported, the number of KCNQ3 variants is increasing, with an expanding phenotypic spectrum including Developmental and Epileptic Encephalopathy (DEE), Intellectual Disability (ID), and Autism Spectrum Disorder (ASD). In a single case study, we report the clinical, neuropsychological, and molecular characterization of a novel KCNQ3 variant to refine genotype-phenotype correlations. Methods: Deep phenotyping included standardized neurodevelopmental assessment (BSID-III) and sensory profiling (Toddler Sensory Profile-2). Genetic analysis was performed using a targeted Next-Generation Sequencing panel with Sanger confirmation of candidate variants. Variant interpretation followed ACMG criteria. Structural impact was assessed using AlphaFold-based modelling and FoldX in silico mutagenesis; pathogenicity was further evaluated with PolyPhen-2, SIFT, and MutationTaster. Results: BSID-III showed global developmental delay (cognitive 75, language 63, motor 67, socio-emotional 75). Sensory assessment identified a high-threshold profile with increased Seeking and Registration patterns across modalities. NGS identified a de novo heterozygous KCNQ3 variant (NM_004519.4:c.730G>A; p.Gly244Ser), absent from population databases. The variant affects a highly conserved residue (PhyloP100=7.9) within the S4 voltage-sensing domain. Structural modelling predicted destabilization (ΔΔG=+1.56 kcal/mol). In silico tools consistently supported a deleterious effect; ACMG classification was likely pathogenic (PS2, PM2, PP3). Seizures were effectively controlled with levetiracetam. Conclusion: This study expands the KCNQ3 mutational spectrum and supports the contribution of S4 domain variants to combined epileptic and neurodevelopmental phenotypes, highlighting the value of integrated deep phenotyping and molecular characterization.