Conclusion of diagnostic odysseys due to inversions disrupting GLI3 and FBN1

Abstract

Many genetic testing methodologies are biased towards picking up structural variants (SVs) that alter copy number. Copy-neutral rearrangements such as inversions are therefore likely to suffer from underascertainment. In this study, manual review prompted by a virtual multidisciplinary team meeting and subsequent bioinformatic prioritisation of data from the 100K Genomes Project was performed across 43 genes linked to well-characterised skeletal disorders. Ten individuals from three independent families were found to harbour diagnostic inversions. In two families, inverted segments of 1.2/14.8 Mb unequivocally disrupted GLI3 and segregated with skeletal features consistent with Greig cephalopolysyndactyly syndrome. For one family, phenotypic blending was due to the opposing breakpoint lying ~45 kb from HOXA13 In the third family, long suspected to have Marfan syndrome, a 2.0 Mb inversion disrupting FBN1 was identified. These findings resolved lengthy diagnostic odysseys of 9-20 years and highlight the importance of direct interaction between clinicians and data-analysts. These exemplars of a rare mutational class inform future SV prioritisation strategies within the NHS Genomic Medicine Service and similar genome sequencing initiatives. In over 30 years since these two disease-gene associations were identified, large inversions have yet to be described and so our results extend the mutational spectra linked to these conditions.

Keywords: Gene Rearrangement; Genomics; Molecular Diagnostic Techniques; Musculoskeletal Diseases; Sequence Inversion.

Figure 1

Pedigrees and characteristic read-alignment signatures for rare diagnostic inversions in three Families from 100KGP. (A) Pedigrees and genetic segregation. Shading in Family 1 indicates polysyndactyly of hands/feet, relative macrocephaly and suspected Greig syndrome. Shading in Family 2 indicates radial dysplasia, toe syndactyly and variable urogenital features, as detailed in online supplemental figure S13. Shading in Family 3 indicates thoracic aortic aneurysm and suspected Marfan syndrome. Clinical status of the proband’s son is unknown. *WGS data available from 100KGP. NA, genetic testing not performed. (B) Read-alignments viewed with IGV showing inversions of chr7:42 051 297–43 254 780 (Family 1) and chr7:27 245 456–42 072 394 (Family 2). Both GLI3-disrupting inversions have breakpoints in intron 4, confirming that truncation of the gene at this point is a bona fide disease mechanism. (C)Distal breakpoint of inversion (chr15:46 635 052–48 604 302) disrupting FBN1 shared by proband (upper track) and mother (middle). †Control (lower) is unrelated individual from 100KGP analysed using similar methods. GRCh38 read-alignments are coloured by pair orientation such that read-pairs where both reads map to the +ve genomic strand are highlighted in green. Read-pairs where both reads map to the –ve strand (blue) are seen on the other side of the breakpoint.

Figure 2

Diagnostic odyssey timelines for Families 1–3. For Family 1, precise dates were unavailable for karyotyping and array testing. *Sequence data initially analysed in 2016 using GRCh37 as a reference. The same data were remapped and reanalysed on GRCh38 in February 2020. †Variant identified on Rare Disease Day 2021. WGS, whole-genome sequencing.