An integrative coding and non-coding SNPs analysis of the CLDN-3 gene in humans to identify high-priority variants...

By creating intramembrane tight-junction barriers, claudin-3 (CLDN-3), a vital member of the claudin protein family, helps to preserve cell polarity. Through aberrant signaling pathways, the dysregulation of CLDN-3 has been linked to gynecological and other epithelial cancers. While previous studies have explored its role in cancer progression, the functional impact of both coding and non-coding single-nucleotide polymorphisms (SNPs) in CLDN-3 remains largely unexplored.

In this study, in-silico tools were employed to identify high-priority variants for gynecological and other epithelial cancers.

Using SIFT, PolyPhen-2, and SNPs&GO, we first assessed 140 non-synonymous SNPs (nsSNPs). Out of these, 29 were consistently predicted as high-priority pathogenic by all three tools. Stability analysis based on ΔΔG using I-Mutant 2.0 and MUpro identified six corensSNPs that were stabilized after mutation and were predicted to impair protein function by MutPred2. Among them, N140I and Y147C lie near the CLDN-3/clostridium perfringens enterotoxin (CPE) binding domain, suggesting a novel mechanism that may interfere with CPE-based therapeutics. These predictions were further supported by molecular dynamics (MD) simulation, which shows that these mutations may compromise CLDN-3 structural integrity and function. For non-coding SNPs analysis, we prioritized 28 high-impact untranslated region (UTR) variants out of 256 based on RegulomeDB scores (≤ 2b). Among these, four SNPs in the 3' UTR were found to influence miRNA binding, indicating dual targetgain/loss effects that could affect post-transcriptional regulation. GTEx-based eQTL analysis identified three 5' UTR SNPs that significantly alter CLDN-3 expression levels, indicating high-value effects on transcriptional regulation. We used the TRANSFAC database to perform transcription factor binding site (TFBS) analysis in order to further evaluate the regulatory landscape; no noteworthy binding motifs were found. Further, minimum free energy (MFE) analysis revealed five SNPs that significantly alter mRNA secondary structure and stability, providing additional insights into the post-transcriptional regulation of CLDN-3.

These comprehensively identified in silico coding and non-coding SNPs may serve as highpriority biomarkers requiring further experimental validation for future clinical applications.