A novel splice-altering frameshift variant in the COL1A1 gene underlies osteogenesis imperfecta type I: molecular characterization of a four-generation Chinese pedigree and literature review

Abstract

Backgroud: Osteogenesis imperfecta (OI) is a phenotypically and genetically heterogeneous group of inherited connective tissue disorder. This investigation aims to elucidate the molecular etiology underlying a four-generation Chinese family affected by OI.

Methods: Whole-exome sequencing was employed to identify pathogenic variants in the proband, with subsequent Sanger sequencing performed for familial co-segregation analysis. A minigene assay was conducted to investigate the effect of variant on splicing patterns. The pathogenic potential of variant was evaluated through protein structural modeling and HEK293 cell-based functional studies. COL1A1 splicing variants were further collated to analyze its occurrence frequency across geographically diverse OI cohorts, intronic distribution patterns and potential hotpots for mild versus severe subtypes.

Results: Multiple affected members within an non-consanguineous Chinese pedigree exhibited clinical manifestations fitting OI-associated phenotypic spectrum. A novel heterozygous COL1A1 splicing variant (c.370-2A > C) inherited from the mother was identified in the proband. The splicing variant altered the canonical acceptor site (AG) at the intron 4-exon 5 junction, activating a adjacent cryptic splicing site in exon 5. This abberrant splicing event introduced a frameshift variant (c.370_379delGGACCCGCAG), and generated a premature termination codon that truncates the COL1A1 protein (p.Gly124Alafs*138). AlphaFold3-based protein structural modeling revealed the loss of the triple-helical domain in this truncated protein. In vitro functional assays showed that mRNA and protein expression levels of mutant COL1A1 were significantly increased than wild-type COL1A1 (p < 0.05). Comprehensive literature analysis indicated that COL1A1 splicing variants account for < 10% of variants in OI cohorts from the vast majority of regions. The acceptor site of intron 9 and the donor sites of intron 35 are hotspots for COL1A1 splicing variant occurrence. Moreover, the majority of COL1A1 splicing variants, expecially those proximal to the 5' and 3' terminal regions, result in mild manifestations of OI type I, whereas variants at donor sites of introns 14, 20, and 46, may be candidate hotspots for lethal OI type II.

Conclusions: Our study revealed the pathogenic mechanism of a novel COL1A1 splicing variant in a four-generation Chinese family with OI, and provided updated data on COL1A1 splicing variants and its potential hotpots for mild versus severe OI subtypes.

Keywords: COL1A1; Minigene assay; Osteogenesis imperfecta; Splicing variant; Whole-exome sequencing.

Fig. 1

Clinical characteristics of the proband and affected family members with OI. (A) Pedigree chart of the proband and affected family members. Males are represented by squares and females by circles, with filled symbols indicating affected individuals and open symbols representing unaffected family members. Black arrow indicates the proband. (B) Ophthalmic examination revealed characteristic blue sclerae in affected individuals III-2, III-3, IV-1, IV-2, and IV-3. (C) Radiographic images of the proband’s hand and arm revealed characteristic skeletal abnormalities, with yellow arrows highlighting upper arm fracture. (D) Anterior and lateral radiographic views of the spine in the proband demonstrated mild scoliosis, as indicated by the yellow arrows

Fig. 2

Molecular genetic detection. (A) A single base-pair change (chr17: 48276690, T to G or A to C) in the COL1A1 gene was visualized using the IGV. (B) Familial segregation analysis of the COL1A1 variant was performed through Sanger sequencing in the proband and affected family members. +/- indicates the presence of a heterozygous variant, and -/- indicates the absence of a heterozygous variant

Fig. 3

Mini-gene splicing assays. (A) Two recombinant vectors were constructed, containing either wild-type or mutant COL1A1 sequences spanning exons 3 to 6. (B) RT-PCR analysis was performed using the primer pairs for amplifying the COL1A1 gene. C: control. E: empty vector. W: wild type. M: mutant. (C) RT-PCR analysis was performed using the specific primer pairs for amplifying the EGFP-COL1A1 gene. (D) The RT-PCR products were analyzed by Sanger sequencing. The red solid line indicates the deleted nucleotide in the mutant COL1A1 sequence. (E) Western blot analysis of wild-type and mutant protein molecular weights. Uncropped blots were available in Supplementary file 1. WT: wild type. MT: mutant

Fig. 4

Pathogenicity analysis using structural modeling and protein expression. (A) 3D structure prediction of both WT and mutant COL1A1 protein using AlphaFold3. (B) Representative fluorescence images showing transfection efficiency of WT and mutant COL1A1 constructs. (C) qRT-PCR analysis of relative mRNA expression for WT and mutant COL1A1. (D) Expression levels of wild type and mutant COL1A1 protein by Western blot and quantitative analysis. Representative blots from three independent biological replicates are shown. Uncropped blots were available in Supplementary file 1

Fig. 5

Systematic analysis of COL1A1 splicing variants. (A) The distribution and prevalence of COL1A1 splicing variants was systematically analyzed across multinational cohorts of OI patients. (B) The distribution of alternative splicing variants within all COL1A1 introns. (C) The number of splicing variants associated with donor and acceptor sites in each intron of the COL1A1 gene

Fig. 6

The composition of OI subtypes caused by different splicing variants at donor and acceptor sites within the COL1A1 intron