EXT1 and EXT2 Variants in 22 Chinese Families With Multiple Osteochondromas: Seven New Variants and Potentiation of Preimplantation Genetic Testing and Prenatal Diagnosis

Abstract

Multiple osteochondromas (MO), the most common type of benign bone tumor, is an autosomal dominant skeletal disorder characterized by multiple cartilage-capped bony protuberances. In most cases, EXT1 and EXT2, which encode glycosyltransferases involved in the biosynthesis of heparan sulfate, are the genes responsible. Here we describe the clinical, phenotypic and genetic characterization of MO in 22 unrelated Chinese families involving a total of 60 patients. Variant detection was performed by means of a battery of different techniques including Sanger sequencing and whole-exome sequencing (WES). The pathogenicity of the missense and splicing variants was explored by means of in silico prediction algorithms. Sixteen unique pathogenic variants, including 10 in the EXT1 gene and 6 in the EXT2 gene, were identified in 18 (82%) of the 22 families. Fourteen (88%) of the 16 variants were predicted to give rise to truncated proteins whereas the remaining two were missense. Seven variants were newly described here, further expanding the spectrum of MO-causing variants in the EXT1 and EXT2 genes. More importantly, the identification of causative variants allowed us to provide genetic counseling to 8 MO patients in terms either of preimplantation genetic testing (PGT) or prenatal diagnosis, thereby preventing the reoccurrence of MO in the corresponding families. This study is the first to report the successful implementation of PGT in MO families and describes the largest number of subjects undergoing prenatal diagnosis to date.

Keywords: EXT1 gene; EXT2 gene; multiple osteochondromas; pathogenic variant; preimplantation genetic testing; prenatal diagnosis.

FIGURE 1

Pedigrees of the 22 Chinese families with multiple osteochondromas. Filled squares or circles denote patients. Open symbols denote clinically unaffected family members. Circles with question marks denote members with questionable clinical features and no imaging tests. Arrows indicate probands. + identifies those subjects subjected to genetic analysis. In Families 1–18, all patients subjected to genetic analysis harbored the variant of interest whereas the clinically healthy subjects did not. In Families 19–22, no pathogenic variants were found in any of the individuals subjected to genetic analysis. Those individuals who were born healthy through the preimplantation genetic testing are denoted by # whereas those who directly received prenatal diagnosis as a fetus are denoted by *. Subjects analyzed by WES were indicated by squares.

FIGURE 2

Special exostosis lesions and complications in some affected MO individuals. (A,B) Radiographs of Family 13 member III-5 as anteroposterior and lateral views, revealing the scoliosis, thoracolumbar kyphosis and vertebral fusion (arrow). (C) Right forearm of Family 11 member III-1, showing exostosis in the ulna and dislocation of the elbow joint (arrow). (D) Right forearm of Family 4 member II-2, showing exostosis in the ulna and bowed forearm conferring dislocation of radioulnar joint (arrow). (E) Radiograph of Family 7 member III-1, revealing the exostoses in the scapula (arrow). (F) Radiograph of Family 4 member II-2, showing the exostosis at the right side of the first rib (arrow). (G) Pelvic radiograph of Family 4 member II-2, displaying exostosis on the left side of the ilium (arrow). (H) Ultrasound images of Family 7 member III-1, showing the absence of a uterus (arrow).

FIGURE 3

In silico analyses with respect to three novel variants. (A) Prediction of the potential effect of EXT1 c.1284 + 1G > C on splicing by Alamut. (B) Prediction of the potential effect of EXT2 c.1173 + 2dupT on splicing by Alamut. (C) Structure model of EXT1. The N-terminal region (amino acid residues 29–461) is shown here in cartoon form. Helix, sheet and loop are colored in red, yellow and green, respectively. The side chains of Cys355 and Cys334 are shown as spheres colored in cyan.

FIGURE 4

Distribution of the described EXT1 (A) and EXT2 (B) variants in accordance with their respective gene/protein structures. The green boxes denote the numbered exons. Novel mutations are shown in red. The horizontal line indicates the approximate demarcations of the genomic deletions in EXT2 (red) and duplication in EXT1 (blue). The protein domains of EXT1 and EXT2 are presented at the bottom of each gene depiction, and were derived from the Protein Families (Pfam) database.