Variants in RNU4-2 , encoding the small nuclear RNA (snRNA) U4, were recently identified as a major cause of neurodevelopmental disorders (ReNU syndrome). Here, we investigated de novo variants in 50 snRNAs in a French cohort of 23,649 individuals with rare disorders and collected data of additional patients through an international collaboration. Altogether, we identified 133 probands with pathogenic or likely pathogenic variants in RNU4-2 and 15 individuals with de novo and/or recurrent variants in constrained regions of RNU5B-1 , one of five genes encoding U5. These variants cluster in evolutionarily conserved regions of U4 and U5 essential for splicing. RNU4-2 variants affecting stem III are associated with milder phenotypes than those in the T-loop (quasi-pseudoknot). Phaseable variants associated with severe phenotypes occurred on the maternal allele. Individuals with RNU4-2 variants show specific defects in alternative 5’ splice site usage, correlating with variant location and clinical severity. Additionally, we report an episignature associated with severe ReNU syndrome. This study further highlights the importance of de novo variants in snRNAs and establishes RNU5B-1 as a new neurodevelopmental disorder gene.
Skeletal dysplasia, also called osteochondrodysplasia, is a category of disorders affecting bone development and children’s growth. Up to 552 genes, including fibroblast growth factor receptor 3 (FGFR3), have been implicated by pathogenic variations in its genesis. Frequently identified causal mutations in osteochondrodysplasia arise in the coding sequences of the FGFR3 gene: c.1138G>A and c.1138G>C in achondroplasia and c.1620C>A and c.1620C>G in hypochondroplasia. However, in some cases, the diagnostic investigations undertaken thus far have failed to identify the causal anomaly, which strengthens the relevance of the diagnostic strategies being further refined. We observed a Caucasian adult with clinical and radiographic features of achondroplasia, with no common pathogenic variant. Exome sequencing detected an FGFR3(NM_000142.4):c.1075+95C>G heterozygous intronic variation. In vitro studies showed that this variant results in the aberrant exonization of a 90-nucleotide 5′ segment of intron 8, resulting in the substitution of the alanine (Ala359) for a glycine (Gly) and the in-frame insertion of 30 amino acids. This change may alter FGFR3’s function. Our report provides the first clinical description of an adult carrying this variant, which completes the phenotype description previously provided in children and confirms the recurrence, the autosomal-dominant pathogenicity, and the diagnostic relevance of this FGFR3 intronic variant. We support its inclusion in routinely used diagnostic tests for osteochondrodysplasia. This may increase the detection rate of causal variants and therefore could have a positive impact on patient management. Finally, FGFR3 alteration via non-coding sequence exonization should be considered a recurrent disease mechanism to be taken into account for new drug design and clinical trial strategies.