A report of 2 cases of Cornelia de Lange syndrome (CdLS) and an analysis of clinical and genetic characteristics in a Chinese CdLS cohort

BACKGROUND Cornelia de Lange syndrome (CdLS) is a rare dominantly inherited developmental disorder with an estimated prevalence of 0.5-10:100,000 and no racial disparity in prevalence. The aim of this study was to present two unrelated Chinese CdLS individuals with mutations in NIPBL and to perform a comprehensive analysis of a Chinese cohort with CdLS. SUBJECTS AND METHODS Two unrelated Chinese patients complaining of short stature were referred to the outpatient department of Peking Union Medical College Hospital (PUMCH). Their clinical data at birth and at the most recent assessment were collected. Mutation analysis was carried out by whole exome sequencing. Twenty-four Chinese cases with CdLS were identified through a systematic review of the literature published between 1987 and 2017. RESULTS Two patients presented with typical phenotypes, characteristic complications of CdLS and mutations in the NIPBL gene. The average age at diagnosis of the 26 Chinese cases was higher than that of other cohorts. The frequencies of characteristic manifestations of CdLS were similar with those of other populations. CONCLUSIONS By investigating 26 Chinese cases of CdLS, we observed that the clinical data and gene variants in the Chinese cohort of CdLS patients were generally in accordance with those of other populations.

Herein, we present two unrelated Chinese CdLS cases. Furthermore, the clinical characteristics of Chinese CdLS cases published previously were summarized and compared with those of other populations.

| Ethical compliance
Examination protocols were approved by the ethics committees of Chinese Academy of Medical Science and Peking Union Medical College with the following reference number: JS-1663. Written informed consents were obtained from the subjects for the publication of the case report and any accompanying images.
Two patients with CdLS-like phenotypes from two unrelated Chinese families were referred to the outpatient department of Peking Union Medical College Hospital (PUMCH). Demographic parameters were requested at birth and at the most recent assessment. No parental consanguinity or positive family history of any deformity was found in any of the cases. All subjects underwent complete evaluation, focusing on psychomotor development, malformations or medical complications.

| Exome sequencing protocol
Two-milliliter peripheral blood samples were taken from two patients. Genomic DNA was isolated using the QIAamp DNA Blood Mini Kit. Two micrograms of DNA were fragmented randomly to an average size of 250 bp using a Covaris acoustic system. An adapter-ligated library was pooled and hybridized with a BGI in-house 59M whole exome capture kit to capture the target regions. Quantitative PCR was performed to estimate the magnitude of enrichment. Then, the qualified captured library was sequenced on a HiSeq 2,500 analyzer (Illumina, Inc.) for a 100-bp paired-end run.

| Read alignment, variant calling, and annotation
Illumina Pipeline (version 1.3.4) was used for image analysis, error estimation, base calling and generating the primary sequence data. The clean data containing paired-end reads were mapped to the human genome (NCBI37/hg19) using BWA software (Burrows Wheeler Aligner, http://sourc eforge.net/proje cts/bio-bwa/). SNVs (single nucleotide variants) and small InDels (insertion and deletions) were identified by SOAP Snp software (http://soap.genom ics.org.cn/) and SAMtools Pileup software (http://sourc eforge.net/proje cts/samto ols/), respectively. The variants were annotated using Gaea, a BGI in-house-developed annotation pipeline. The public databases used to calculate the frequency in the normal population included the 1K genome database (http:// www.1000g enomes.org/), ESP6500 database, dbSNP database and BGI in-house database. The software used for nonsynonymous functional predictions included PolyPhen-2, SIFT and Ens Condel.

| Literature review
We systematically searched the online literature databases Pubmed, Embase, Medline, Wanfang, China National Knowledge Infrastructure (CNKI) and the Cochrane Collaboration Library for articles associated with a Chinese cohort of CdLS patients. Finally, 24 previously reported cases of CdLS in the Chinese population were identified (Table 1). All subjects in the included papers complied with the genetic sequencing diagnosis or clinical diagnostic criteria for CdLS suggested by the CdLS Foundation's Medical Director, Kline et al. (2018).

| Applying the consensus criteria of clinical phenotype in Chinese patients with CdLS
Each individual was evaluated by the consensus criteria of clinical phenotype of CdLS, which integrates facial dysmorphisms, growth retardation, psychomotor development, hand oligodactyly and/or adactyly, hirsutism and other major malformations (Kline et al., 2018). The patients are classified into four phenotypes. Individuals having at least three cardinal features and an overall score of ≥11 are considered as "classic CdLS" phenotype. An overall score of 9-10 with at least two cardinal features indicates "non-classic CdLS". A score of ≥4 and at least one cardinal feature is considered as being sufficient to indicate molecular testing for CdLS. A score of <4 is insufficient to conduct such testing. No other cohorts of Chinese CdLS patients have been investigated with this consensus criteria of clinical phenotypes.

| Clinical data of the two patients in the present study
Both of the two patients in the present study were children of clinically asymptomatic nonconsanguineous parents with normal height and no genetic defects. Patient 1 (Figure 1a) was born at a gestational age of 34 weeks by cesarean section. At birth, her weight was 2.15 kg. At the age of 5 years, her height was 100 cm (−2.43 SDS). Her bone age was unmeasured. She manifested with a low hair line, synophrys, hypertrichosis of the brow and long curly eyelashes. She had bone anomalies that included 5th clinodactyly in the fingers of both hands and 2nd and 3rd syndactyly in the toes of both feet. She also presented microcephaly, patent ductus arteriosus and dental anomalies. Her serum IGF-1 level was 73.8 ng/ml (−1.72 SDS) and GH level was 16.3 ng/ml. Patient 2 (Figure 1b) was born full-term by a normal delivery, with a birth weight of 2.5 kg. She was 17 years old, and her height was 134.3 cm (−4.81 SDS). Her bone age was 16-18 years. She had synophrys and long eyelashes. She also exhibited bone anomalies, including 5th clinodactyly in the fingers of both hands and 3rd and 4th syndactyly in the toes of both feet ( Figure 2). In addition, she had impaired expressive communication and learning disabilities. Her serum IGF-1 was 351 ng/ml (−1.52 SDS). No abnormalities were found in ophthalmologic examination, echocardiography or other evaluations.
All of the patients had the typical craniofacial characteristics of CdLS, including synophrys (  gene (NM_133433.3, NG_006987.2 RefSeqGene) in patient 1. In patient 2, a frameshift mutation c.8274_8275 ins CT was detected in exon 47 of the NIPBL gene, which has been predicted to cause the replacement of a valine with a tryptophan at residue 2,760 (p. Val2760Trpfs). Furthermore, c.8274_8275 ins CT was suggested to be pathogenic by the Clinvar database. These two mutations have been previously reported (Oliveira et al., 2010;Richards et al., 2015;Teresa-Rodrigo et al., 2014). None of these subjects had chromosomal abnormalities.

| Neonatal features
The male/female ratio in the Chinese cohort of 26 individuals with CdLS was 1:1.6 (10:16). Selicorni et al. (2007) found that this ratio was 1:0.63 (38:24) in their study of 62 Italian patients with CdLS. 50% of the 26 Chinese individuals were diagnosed with CdLS when they were less than one year old. This is consistent with the study conducted by Ramos et al. in which 52% of 101 Italian CdLS patients were diagnosed within the first year of life (Oliver et al., 2010). The proportion of prenatal growth retardation in the group of Chinese patients with CdLS (88.5%) was higher than that in the 62 Italian CdLS patients (56.4%) observed by Selicorni et al. (2007). The specific reason for the disparity between these findings warrants further investigation.

| Craniofacial characteristics
Facial abnormalities are pathognomonic for CdLS and constitute the clinical hallmark of the syndrome, which can lead to the initial diagnosis (Boyle, Jespersgaard, Brøndum-Nielsen, Bisgaard, & Tümer, 2015). Kline et al. (2018) collected auxological parameters of over 500 individuals with CdLS and observed that the most typical facial dysmorphias were long and thick eyelashes (99%), synophrys and hypertrichosis of the brows (98%), thin lips with down-turned corners (94%), a depressed nasal bridge with anteverted nares (85%), widely spaced teeth and micrognathia (84%), hirsutism (78%) and cutis marmorata (60%). The frequency of the craniofacial abnormalities observed in the study cohort of 26 Chinese individuals with CdLS was consistent with the above findings (Table 2).

| Limb anomalies
Among the 26 Chinese individuals diagnosed with CdLS, 25 patients of them had some extremity malformations including minor differences in morphologies involving the limbs (e.g., proximally placed thumb, small hand, clinodactyly, syndactyly) ( Table 2). This result was consistent with previously described results in the literature, in which 86% of the 310 CdLS individuals observed by Jackson, Kline, Barr, and Koch (1993) and 100% of the 378 CdLS subjects in the large cohort studied by Mehta et al. (2016)  CdLS patients are more common than those of the lower extremities (Mehta et al., 2016). And this is also true for the 26 Chinese CdLS cases: 8 cases had both upper and lower extremities anomalies, 16 cases had only upper extremities anomalies and no cases had only lower extremities manifestations (Table 2).

| Intellectual disability
Cognitive impairment with behavioral and neurological problems is a hallmark of CdLS. We found that 23 patients among the 26 Chinese individuals diagnosed with CdLS had impaired psychomotor development and intellectual disabilities, which is consistent with previous studies (Grados, Alvi, & Srivastava, 2017). Srivastava et al. (2014) studied 41 children with CdLS from 5 to 18 years old and found that 17% did not have autism, 41% had mild autism, and another 41% had severe autism based on the Childhood Autism Rating Scale (CARS). Only 2 of the 26 Chinese cases with CdLS were reported to have obvious autism manifestations and this percentage (7.7%) is much lower than the data found by Srivastava et al. This might be attributable to the fact that only 5 cases of the 26 Chinese CdLS patients were older than 5 years old when they were diagnosed as CdLS and the autism features of the rest cases hadn't manifested yet. However, it doesn't mean the behavioral abnormalities in Chinese CdLS cohorts are less common. On the contrary, it might indicate that many elder children with CdLS haven't been diagnosed in China yet which should arouse further attention of clinicians.

| Genetic analysis
CdLS is a genetically and clinically heterogeneous disorder. DNA sequence variations in NIPBL, SMC3 and RAD21 are closely correlated with the autosomal dominant form of CdLS, while SMC1A and HDAC8 abnormalities result in the X-linked form of CdLS (Kaiser et al., 2014). The mutations in these five genes leading to CdLS are all involved in the cohesin pathway. Cohesin and regulatory cohesin genes are responsible for chromosome segregation, DNA repair, and gene transcription regulation (Liu & Krantz, 2009). NIPBL gene mutations have been found to be the most common causes of CdLS worldwide and contribute to approximately 50% of all CdLS cases, while SMC1A and SMC3 alterations were estimated to account for 5% and < 1%, respectively (Oliveira et al., 2010). Reports of the occurrence of mutations in the RAD21 and HDAC8 genes are rather anecdotal; thus, further studies in larger cohorts are needed for assessment .

| CONCLUSIONS
Herein, we present 2 unrelated Chinese CdLS patients. Furthermore, we are the first to apply the consensus criteria of clinical phenotype in CdLS patients proposed by Kline et al. in 2018 to Chinese patients with CdLS. We observed that the clinical data and gene variants in all 26 Chinese cases with CdLS collected by literature review did not significantly vary from those of other populations. This is the first time that a summarization of clinical and genetic characteristics of Chinese patients with CdLS has been reported. However, further studies in larger cohorts are needed to elucidate the genotype-phenotype correlations and the prevalence of variable NIPBL mutations in Chinese populations with CdLS.