Atypical Sotos syndrome caused by a novel splice site variant

Sotos syndrome is usually caused by haploinsufficiency of NSD1; it is characterized by overgrowth, craniofacial features, and learning disabilities. We describe a boy with Sotos syndrome caused by a splicing variant (c.4378+5G>A). The clinical manifestations included severe connective tissue involvement, including joint hypermobility, progressive scoliosis, pectus deformity, and skin hyperextensibility; no overgrowth was observed.

section. At birth, his weight was 4072 g (+2.9 SD), and his length was 56 cm (+3.9 SD); occipitofrontal circumference (OFC) was not available. He showed prominent hypotonia and neonatal jaundice requiring phototherapy. At the age of 5 months, he developed pneumonia; diagnostic chest radiographs incidentally revealed scoliosis. He began walking unsupported at the age of 18 months; he began to speak single meaningful words at the age of 24 months. No abnormal karyotype was found. Although Sotos syndrome was suspected based on overgrowth at birth, global developmental delay, and facial features (e.g., broad forehead, downslanting palpebral fissures, small nose, short philtrum, and pointed chin) ( Fig. 1A-D), fluorescence in situ hybridization revealed no 5q35 microdeletion encompassing NSD1. The proband first experienced otitis media at the age of 2 years. He developed epilepsy at the age of 3 years, which was controlled with carbamazepine. Scoliosis gradually progressed, such that he began wearing a brace at the age of 2 years (Fig. 1L); he underwent implantation of a growth-promoting vertical expandable prosthetic titanium rib at the age of 5 years (Fig. 1M, N). Subsequently, he underwent growing rod surgery; the rods were lengthened at 6-month intervals, followed by final fixation surgery at the age of 12 years ( Fig. 1O, P).
When the proband was 7 years of age, he underwent examination at our clinic. His main medical concerns were connective tissue-related complications, including progressive scoliosis and severe pes planus. He also showed soft and mildly hyperextensible skin; joint hypermobility; and a slender build with decreased subcutaneous fat, long and slender limbs, large hands, and large and flat feet (Fig. 1K). Consequently, a hereditary connective tissue disorder was suspected. The proband's weight was 19.9 kg (-1.2 SD), height was 120.9 cm (-0.8 SD), OFC was 51.2 cm (-0.5 SD), and arm span was 128 cm. He showed characteristic craniofacial features (e.g., broad forehead, long and downslanting palpebral fissures, small nose, short philtrum, everted upper lip, prominent ears, high arched palate, and numerous dental caries) (Fig. 1E, F). When he was 9 years of age, array comparative genomic hybridization and gene panel analysis of hereditary connective tissue disorders (e.g., Marfan syndrome and Ehlers-Danlos syndrome) revealed no pathogenic variants.
The proband attended a special class for physically handicapped children at an elementary school because of his severe scoliosis. Evaluation by the Wechsler Intelligence Scale for Children, 4th edition, showed that he had an IQ of 61 at the age of 11 years; thus, he was admitted to a junior high school with a special class for intellectual disabilities. He had nocturia since childhood; when he was 15 years of age, urological examinations revealed bilateral vesicoureteral reflux and a reduced bladder capacity of 180 mL. Therefore, he began oral fesoterodine therapy.
We last examined the proband when he was 16 years of age; his weight was 39.1 kg (-1.6 SD), height was 158.0 cm (-1.1 SD), and OFC was 54.0 cm (-1.4 SD). He showed craniofacial characteristics (e.g., frontal bossing, downslanting palpebral fissures, and a pointed chin) (Fig. 1G, H), skeletal features (e.g., slender build, joint hypermobility in the thumbs [Beighton score 2/9], severe scoliosis, large hands with slender fingers [Fig. 1I, J], and pes planus), and skin with softness and mild hyperextensibility but without fragility or bruisability. Skin hyperextensibility was measured using the method by Remvig et al. 5 (length of skin stretching ≥cutoff values in 4/5 areas). The systemic score in the revised Ghent criteria for Marfan syndrome 6 was 7/20 (≧7 indicating systemic involvement), although he did not have aortic dilatation or ectopia lentis. He had neither overgrowth nor macrocephaly, suggesting an atypical form of Sotos syndrome.
An unknown disorder with connective tissue involvement was suspected; thus, whole-exome sequencing (WES) was performed. This study was approved by the Ethics Committee at Shinshu University School of Medicine. After written informed consent was obtained from the proband and his parents, genomic DNA was extracted from peripheral blood leukocytes and subjected to WES. Libraries were prepared with SureSelect Human All Exon kit V6 (Agilent Technologies, Santa Clara, CA, USA). Libraries were sequenced on NovaSeq (Illumina, San Diego, CA, USA) with 151bp paired-end reads. After read quality was assessed using FastQC v.0.11.8 (https://github.com/s-andrews/FastQC/), the reads were aligned to the human reference genome (UCSC hg19, NCBI build  (Fig. 2D).
The detected variant was an ultrarare novel splice site variant that was not registered in databases such as gnomAD, ExAC, GEM-J WGA, and HGVD or in our in-house databases. In silico prediction of splice site variants (http://www.fruitfly.org/seq_tools/ splice.htmL) showed that the variant had a prediction score of 0.96, indicating that the variant is likely to cause splicing abnormalities and produce aberrant transcripts. Another single nucleotide substitution at the same position, c.4378+5G>C, was previously identified as a causative variant of Sotos syndrome 8 . The variant c.4378+5G>A was classified as a likely pathogenic variant according to the American College of Medical Genetics and Genomics and the Association for Molecular Pathology guidelines (PS2: de novo, PM2: absent from controls) 9 . Considering these clinical and molecular findings, the proband was diagnosed with Sotos syndrome.
Although Sotos syndrome had been suspected when the proband was an infant because of intrauterine overgrowth and craniofacial features, connective tissue involvement (e.g., skin softness and mild hyperextensibility, joint hypermobility in the thumbs, and severe scoliosis) were the main clinical manifestations thereafter. We initially presumed that he had an unknown connective tissue disorder based on his substantial connective tissue-related features along with normal OFC and nonovergrowth, which are atypical for Sotos syndrome. These atypical features might be attributed to the characteristics of the detected variant, c.4378+5G>A. However, there were no data concerning detailed clinical features in a previously described patient with Sotos syndrome who had a variant at the same position (c.4378+5G>C) 8 . The molecular consequences of this probable splicing variant might be identified through an mRNA-based analysis, but mRNA could not be obtained from the proband. Furthermore, other gene(s) might be related to connective tissue involvement in the proband, although WES excluded variants in genes relevant to known heritable connective tissue disorders.
In conclusion, we identified a novel splice site mutation of NSD1 in a patient with atypical Sotos syndrome that included substantial connective tissue involvement without overgrowth. These observations might extend the phenotypic spectrum of Sotos syndrome.

DATA AVAILABILITY
The relevant data from this Data Report are hosted at the Human Genome Variation Database at https://doi.org/10.6084/m9.figshare.hgv.3252.