In vitro pathogenicity evaluation of deep intronic variants for recessive genetic retinal diseases

Deep intronic variants activating cryptic exons and altering mRNA splicing, measured by a high-throughput minigene splicing assay (HTSA) in HEK293 cells

Non-coding variants that alter mRNA splicing are increasingly recognised as a cause of Mendelian disorders; deep intronic variants (DIV) that activate cryptic exons (CE) are thought to underlie about 20% of inherited retinal disease (IRD) cases, yet their pathogenicity remains difficult to predict accurately. The authors therefore measured the effect of 640 very rare DIV, chosen in trans with a known mutation in patients with recessive IRD, using a high-throughput splicing assay (HTSA) based on a split-GFP minigene into which the 270-bp sequences flanking each variant are cloned; the plasmid library is transfected into HEK293 cells, and the transcripts are amplified by RT-PCR, sequenced and quantified. The assay shows that 98 variants activate a cryptic exon more than 100 times more than the reference allele and are classified as pathogenic in this experimental setting, validated in a longer context, while 26 are classified as variants of uncertain significance; only 6 of 90 variants (6.6%) were predicted pathogenic by in silico algorithms (SpliceAI), and the diagnosis is confirmed in 50 of 78 patients (64%). Overall, 19.4% of the deep intronic variants tested activate a cryptic exon, which the authors read as a sign that a significant portion of undiagnosed patients with recessive IRD carries a pathogenic intronic mutation.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The contrast between 98 functionally pathogenic variants and only 6 of 90 predicted by SpliceAI is the result to retain: for deep intronic variants, a predictive score cannot substitute for a splicing assay. The yield of 50 patients out of 78 is nonetheless obtained in an artificial system — a minigene in HEK293 cells with 270-bp flanking sequences — and the authors themselves speak of pathogenic classification in this experimental setting; confirmation on patient RNA remains desirable before translating this result into clinical classification. Since these are very rare variants selected in trans with a known mutation, the 19.4% rate cannot be transposed to all intronic variants, and this is a preprint not yet peer reviewed.

Analysis by Dr Thibaut Benquey