Alport syndrome (AS) is a hereditary progressive kidney disease caused by pathogenic variants in the COL4A3, COL4A4, and COL4A5 genes. Aberrant pre-mRNA splicing represents a major disease mechanism in AS, and both synonymous and intronic variants can disrupt splice-site recognition or regulatory elements, leading to exon skipping or intron retention. However, the clinical interpretation of putative spliceogenic variants remains challenging, particularly when subtle splicing changes coexist with background alternative splicing. Fourteen synonymous and intronic variants in COL4A3 and COL4A4 were screened using in silico splicing prediction tools and subsequently functionally evaluated by minigene assays in HEK293T and HeLa cells. Thirteen variants exhibited aberrant splicing, most commonly as exon skipping, with variable effects on transcript structure and abundance. Several variants predominantly generated frameshift transcripts with premature termination codons, whereas others resulted in in-frame exon skipping or produced low-level aberrant transcripts comparable to wild-type splicing. By integrating quantitative splicing outcomes with population frequency data and predicted protein consequences, all variants were reclassified according to ACMG/AMP guidelines, resulting in four pathogenic, nine likely pathogenic, and one likely benign variant. This study demonstrates that synonymous and intronic variants in COL4A3 and COL4A4 can exert pathogenic effects primarily through disruption of pre-mRNA splicing rather than through direct alteration of the coding sequence. Incorporation of quantitative functional splicing data into ACMG/AMP variant interpretation provides a robust framework for resolving variants of uncertain significance and improves the molecular diagnosis of AS.
Wang et al. (Wed,) studied this question.