Characteristic spatial and frequency distribution of mutations in SCN1A.

BACKGROUND

SCN1A is the most well-recognized and commonly mutated gene related to epilepsy. This study analyzed the characteristic spatial and frequency distributions of SCN1A mutations, aiming to provide important insight into the mutagenesis etiopathology of SCN1A-associated epilepsy.

METHODS

Epilepsy-associated SCN1A variants were retrieved from the SCN1A mutation database, the HGMD database, and literature reviews. The base substitutions, mutation frequencies in CpG dinucleotides, and spatial distributions of mutations in terms of exons and structural domains were analyzed.

RESULTS

A total of 2621 SCN1A variants were identified in 5106 unrelated cases. The most common type was missense mutation, followed by frameshift mutations and splice site mutations. Among the missense mutations, transitions within CpG dinucleotides were much more recurrently identified than transitions within non-CpG dinucleotides, and the most common type was the G > A transition. Among the nonsense mutations, the most predominant type of single-base substitution was the C > T transition, among which 75.3% (235/312) were within CpG sites. The most common "hotspot" codons for missense mutations were codons 101, 946, and 1783; while for nonsense mutations it was codon 712. One-base deletion or insertion was the most common type of frameshift mutation, causing protein truncation. The three most common frameshift mutations were c.5536_5539delAAAC, c.4554dupA, and c.5010_5013delGTTT. Splice mutations were the most frequently identified in exon 4 with a hotspot site c.602 + 1G > A. The spatial distribution of missense mutations showed that exons 22 and 4 had the highest mutation density (111 and 84 mutations per 100 bp, respectively), and exon 12 had the lowest mutation density, with 4 mutations per 100 bp. Further distribution analysis of the protein domains revealed that missense mutations were more common in the pore region and voltage sensor (231 mutations per 100 amino acids, respectively), and the protein truncation mutations were distributed evenly among the domains.

CONCLUSIONS

SCN1A mutations tend to cluster at distinct sites, depending on the characteristic CpG dinucleotides, exons, and functional domains. Higher mutation density in particular regions, such as exon 22 and exon 4, offers promising targets for therapeutic genetic interventions.