Background

Syndactyly (SD) refers to the deformity of partial or complete fusion of adjacent fingers or toes caused by abnormal influencing factors during limb development. SD has obvious phenotypic heterogeneity and genetic heterogeneity, and SD is characterized by being symmetric or asymmetric and unilateral or bilateral. The complex clinical phenotype of SD leads to different degrees of deformity, which can manifest as abnormal skin texture or involve phalanges, metacarpals or metatarsals and more seriously extend to carpal or tarsal bones [1]. At present, 9 kinds of nonsyndromic SD (type I-IX) have been reported. Among these 9 types, the five main types (I-V type) are autosomal dominant inheritance, and the other types and subtypes are inherited in an autosomal recessive or X‑linked hereditary pattern [2]. Syndactyly type I is the most common type, and it can manifest as centre axis webbing of 3/4 syndactyly and/or 2/3 toe syndactyly. According to the phenotype and genotype, this type can be divided into four subtypes (type I-a, type I-b, type I-c, and type I-d). Syndactyly I-c (Montagu) is associated with skin/bone fusion of only the third and fourth fingers and occasionally the third to fifth fingers [3]. In the process of limb development, the signalling and transcriptional regulation of genes are involved in this highly complex growth process. It is known that the genes HOXD13, SHH, GLI3, GJA1, LRP4, WNT7A and APC play an important role in limb development. Different mutations of these genes may lead to limb deformities. It has been reported that deletions, nonsense mutations, missense mutations, frameshift mutations or polyalanine expansion mutations of the HOXD13 gene could lead to SD [4, 5]. In this study, we examined a four-generation Mongolian family with congenital syndactyly from Eastern Inner Mongolia, China. To identify the pathogenic genetic variants of this family, we used exon sequencing analysis and related analysis methods to research this Mongolian family.

Methods

Clinical data collection

A Mongolian family with congenital syndactyly in eastern Inner Mongolia was selected as the research object. The proband (II4) in this family was a 65-year-old female. Genetic counselling was conducted at the Affiliated Hospital of Inner Mongolia Minzu University for the first time in March 2014. This condition has been passed down for 4 generations in this family. The medical history of 38 members of this family can be traced back by investigation (Fig. 1), and 9 patients with this condition were identified (4 males and 5 females). III16 underwent finger separation surgery with the traditional folk method in infancy and had a poor prognosis. Two patients, IV6 and IV7, underwent operations in the Department of Hand and Foot Surgery of the Affiliated Hospital of Inner Mongolia Minzu University between the ages of 5 and 12, and the prognosis was good (Table 1). The hand pictures and X-ray results of II4, III10, III12 and IV7 are shown in Fig. 2.

All study protocols were approved by the Ethics Committee of the Affiliated Hospital of Inner Mongolia Minzu University. The ethical approval ID is NM-LL-2018-03-03-01. All experimental work was performed in accordance with the Declaration of Helsinki. Informed consent forms were obtained from all the participants for genetic analysis of the DNA samples and publication of the genetic data. We collected basic information, medical histories and X-ray results. To date, the clinical data of 15 members have been collected. The amount of blood collected was 10ml, and it was stored in two EDTA anticoagulant tubes for further study.

Fig. 1
figure 1

A pedigree with syndactyly

■and●represent male patients and female patients respectively;□and○represent normal men and normal women respectively;↗represents the proband (II4);+/−: heterozygous for HOXD13, +/+: wild type for HOXD13.

Full size image
Fig. 2
figure 2

Pictures and X-rays of syndactyly of pedigree patients

II4 (proband)the third and fourth fingers of the left hand were combined, and the toes were normal; III10 (patient)the third and fourth fingers of the right hand were combined, and the toes were normal; III12 Normal phenotype but carrier of pathogenic gene; IV7 third and fourth fingers of both hands were combined, and two repair operations have been performed.

Full size image
Table 1 Clinical characteristics of the family
Full size table

Extraction of genomic DNA

We collected 5ml peripheral blood from family volunteers with EDTA anticoagulant. Blood genomic DNA was extracted by protease K-SDS-phenol/chloroform extraction. Extracted DNA was quantified using a Qubit 3.0 Fluorometer. Agarose gel electrophoresis was used to analyse the extent of DNA degradation and to assess whether there was contamination of RNA and protein.

Exon sequencing and data analysis

In this experiment, six members of the family (three patients, II4, III10 and IV7, and two unaffected members, III2, III13 and III12) were selected for whole exon sequencing to capture the candidate genes. Exon sequencing was completed by Novo gene.

The operation process is reported here. After extracting genomic DNA, the library was constructed, and the exons of the library products were captured. In the capture process, an Agilent liquid-phase probe capture system was used to efficiently enrich human DNA in the whole exon region. After PCR amplification, library quality inspection was carried out, and Illumina NovaSeq 6000 PE150 sequencing (pe150, pair end 150 bp, which means it is high-throughput two terminal sequencing with 150 bp measured at each end) was carried out according to the effective concentration of the qualified library and the data output requirements (Fig. 3). The sequencing depth was 100x, and the original sequence (raw reads) was obtained by sequencing. To guarantee the quality of information analysis, raw reads were finely filtered to obtain clean reads, and the subsequent analysis is based on the clean reads. The specific process involved the removal of reads paired with the adapter and the removal of reads paired with a proportion of N (N indicates that the base information cannot be determined) greater than 10% in the single-end sequencing read. When the base number of low-quality (less than 5) bases contained in the single-end sequencing read exceeds 50% of the length ratio of the read, this pair of reads was removed. This ensures that the quality of the sequencing data is mainly distributed over Q30 (≥ 80%), and the remaining valid data are clean reads.

Clean reads were compared to the reference genome (GRCh37/hg19) through BWA software. The initial comparison results in BAM format were obtained and then sorted with SAMtools. Sambamba marks repeated reads (marked duplicate reads) and identifies SNP sites through Sentieon. The conversion/transversion ratio (Ts:Tv) can reflect the accuracy of the SNP datasets, and this ratio in the whole genome was approximately 2.2. ANNOVAR software was used to annotate SNPs, including the annotation information of the dbSNP database, thousand human genome project database and other existing databases. CNVkit software was used to detect CNVs. CNVs were detected by determining the depth distribution of reads on a reference genome. ANNOVAR software was used to annotate the copy number variations. Finally, candidate genes were screened through annotation.

Fig. 3
figure 3

Library construction and exon capture

The genomic DNA was randomly broken into segments with a growth of 180-280 bp.After end repair and adding A tail, connectors were connected at both ends of the fragment to prepare DNA library. The library with specific index was pooled and hybridized with biotin labeled probe in liquid phase, and then the exons on the gene were captured.

Full size image

PCR amplification and Sanger sequencing validation

Primer design, PCR amplification and Sanger sequencing were carried out for the six candidate genes obtained by exon sequencing. The primers for PCR amplification of the discovered pathogenic variant were as follows: HOXD13 forward primer, 5’-CTGCACCCCTGCAAACGCAC-3’; and HOXD13 reverse primer, 5’-GCTGTCTGTGGCCAACCTGG-3’. The PCR amplification reaction conditions were as follows: 3 min predenaturation at 95 °C; 30 s denaturation at 94 °C, 30 s annealing at 55 °C, and 45 s extension at 72 °C for 30 cycles; and 72 °C for 5 min. The primer sequences and amplification conditions of the FBN2, FMN1, FREM2, LRP4 and MYO10 genes are indicated in Additional file 1: Table S1.

The PCR products were detected by 2% agarose gel electrophoresis, recovered from the gel (TIAN gel Midi Purification Kit), and sent to Beijing Liuhe Huada Gene Technology Co Ltd. for Sanger sequencing.

Results

Phenotypic characteristics

This Mongolian family included 38 members that spanned 4 generations. There were 9 patients involved, including 4 males and 5 females. Patients appear in each generation. The incidence probability for men and women is equal, which reflects an autosomal dominant inheritance model. I2, II4, III5, III7 and III10 were unilateral 3/4 finger fusions, and all of these patients had normal toes. III16, IV3, IV,6 and IV,7 had bilateral 3/4 finger fusion, and all of these patients had normal toes. No other diseases were identified for the 9 patients and other major family members who underwent a systematic physical examination. The mother (III12) of patient IV7 had no clinical phenotype, but the test results showed that she was a carrier of pathogenic genes.

Exon sequencing results

Exon sequencing for performed for six family members, including the proband (3 patients, 1 suspected pathogenic gene carrier and 2 unaffected members), and a total of 6 candidate genes related to finger (toe) development were screened, namely, the FBN2, FMN1, FREM2, HOXD13, LRP4 and MYO10 genes (Table 2).

Table 2 Genetic information
Full size table

Sanger sequencing results

The HOXD13, FBN2, FMN1, FREM2, LRP4 and MYO10 genes were amplified by PCR. All samples contained the target DNA fragments (target fragment lengths were 388 bp for HOXD13, 445 bp for FBN2, 382 bp for FMN1, 688 bp for FREM2, 764 bp for LRP4, and 685 bp for MYO10). Further Sanger sequencing validation showed that in this family, the patient and the carrier without clinical phenotype (III12) had a G > A mutation in the second exon at position 917 of the HOXD13 gene (HOXD13, c.917G > A) (Fig. 4) that resulted in the change of arginine at position 306 to glutamine. This mutation was not seen in the unaffected family members, indicating that this site has a cosegregation phenomenon between patients and unaffected members of this family. In addition, the mutation sites of the FBN2, MYO10, LRP4, FREM2 and FMN1 genes were not cosegregated in the patients and unaffected members of this family (Additional file 2: Figure S1). Moreover, referring to the genomes of 1000 people and 201 Mongolian genome databases, this HOXD13 gene mutation (c.917G > A) was not found, indicating that the HOXD13 gene c.917G > A, p.R306Q mutation is the main cause of disease in this Mongolian family.

Fig. 4
figure 4

Verification results (HOXD13)

A.2%agarosegel electrophoresis was used to detect PCR products: subjects; H2O.Negative control; M. DL 2000 DNA marker (Note: the gel figure was cropped; The original gel image is in Additional file3: Figure S2). B. Sanger sequencing validation results: II4 proband is GA heterozygous; The carriers with incomplete penetrance of III12 has normal phenotype, but the genotype is GA heterozygous; III8 normal member, genotype is GG homozygous; III2 normal member, the genotype is GG homozygous.

Full size image

HOXD13 function prediction analysis

The HOXD13 gene is a member of the Hox gene family and is located at 2q31.1. It contains two exons and is 1365 bp in length. It encodes a 343 amino acid polypeptide sequence. The HOXD13 gene c.917G > A mutation is located in exon 2, which encodes the homeodomain in the C-terminal region (a highly conserved DNA binding motif). A protein secondary structure prediction website (http://bioinf.cs.ucl.ac.uk/psipred/) was used to predict the structure of the mutated sequence. The results show that amino acids at the 30th ~ 40th, 110th ~ 120th and 240th ~ 250th positions composing the alpha helix form an irregular curl. The homologous protein sequence for homebox protein Hox-D13 [Homo sapiens] was obtained from NCBI (National Center for Biotechnology Information). The protein three-dimensional structure prediction website SWISS-MODEL (www.swissmodel.expasy.org/) was used to predict the protein structure. The mutation of the HOXD13 gene c.917G > A, p.R306Q changes the main chemical bonds that maintain the three-dimensional structure of the protein (Fig. 5).

Fig. 5
figure 5

Function prediction of HOXD13 gene mutation

(A) HOXD13 is located on chromosome 2 q31.1 (red bar). (B) The gene structure of the HOXD13. (C) Sequence Plot. (D) Protein structure of HOXD13 wild type and mutant type.

Full size image

Discussion

According to the data of Online Mendelian Inheritance in Man (OMIM), 287 gene mutations have been associated with syndactyly. In this study, six candidate genes related to finger (toe) development were screened by exon sequencing: the FBN2, FMN1, FREM2, HOXD13, LRP4 and MYO10 genes. The variation sites of the FBN2 gene are mainly missense variations and nonsense variations, and these variations are an important cause of congenital contractile arachnodactyly (CCA). The main manifestation of this condition is that patients usually have permanent joint curvature (contracture), such as finger and toe curvature (toe flexion) [6, 7]. The Cenani-Lenz syndrome phenotype is related to various mutations of the LRP4 gene. Its potential mechanism of action may be involved in the impaired membrane transport of LRP4 protein [8, 9]. The rearrangement of the GREM1-FMN1 gene and FMN1 mutations and abnormal limb malformations, such as heart malformation, cartilage hypoplasia and polydactyly, are associated with the loss of FMN1 protein [10]. A study of Fraser syndrome in a foetus found that mutations of FREM2 may lead to symptoms of double fingers and hidden eyeballs in foetuses with congenital high airway obstruction syndrome and renal hypoplasia after birth. However, abnormalities in the fused eyelids and fingers, larynx and vagina are caused by failed programmed cell death [11, 12]. The MYO10 gene plays a certain role in limb development and promotes cell migration in vivo and in vitro [13, 14]. The six candidate genes obtained by second-generation sequencing in this study were analysed by Sanger sequencing and other methods. The results showed that there was no cosegregation of FBN2, FMN1, FREM2, LRP4, MYO10 or other gene mutations in patients and unaffected members of this family, so these genes were excluded as the main pathogenic genetic variants of this family.

Clinically, HOXD13 gene mutations have the strongest correlation with familial syndactyly. According to the mutation location and type, mutations can be divided into functional mutations, such as alanine chain length variations, homologous box amino acid variations, and splice site mutations [15]. The HOXD13 gene has two exons. Exon 1 encodes a 15 residue polyalanine region of the HOXD13 N-terminal region. Mutations of the homologous box amino acid encoded by exon 2 can reduce the DNA binding activity of HOXD13 or change the recognition of its target gene. This results in a multi finger (toe) or short finger (toe) phenotype [16].

Through genetic research involving several SD families, it was found that syndactyly type I-c, SPD1 (synpolydactyly type 1) and syndactyly type V (SDTY5; MIM# 186,300) were caused by mutations in homeobox D13 (HOXD13) and the c. 917G > A (p.R306Q) mutant affects the ability of HOXD13 to activate transcription. Mutations in the HOXD13 homologous domain (p.R306Q) can cause SD1-c, and there is a similarity between SD1-c and polydactyly syndrome [17, 18]. Another study reported that an SPD deformity caused by a missense mutation of the HOXD13 gene homeodomain was found in a Chinese family [19]. There are also studies on the mutation of the HOXD13 gene in an SPD family, which further confirms that PAE mutations in the HOXD13 gene in the Chinese population can lead to typical SPD [20]. Studies have reported that the HOXD13 gene has different genetic defects in mouse models simulating human multi finger and toe phenotypes [21]. These results further support the importance of the HOXD13 gene in limb development.

SPD1 is characterized by syndactyly of the middle axis (3/4 fingers) and polydactyly of the posterior axis (4/5 toes). Type V syndactyly showed 4/5 syndactyly with metacarpal fusion, 4/5 metacarpal dysplasia and axial webbing. The clinical features of type I-c syndactyly are only 3/4 fingers, skin/bone fusion and normal toes [1, 2].

In the Mongolian family of this study, patients I2, II4, III5, III,7 and III10 had unilateral 3/4 finger fusion, III16, IV3, IV6 and IV7 had bilateral 3/4 finger fusion, and no toes were affected. This indicates type I-c syndactyly. The mother (III12) of patient IV7 had no clinical phenotype, but the test results showed that she was a carrier of a pathogenic gene. II2 in the pedigree died. According to the relatives’ description, this individual was asymptomatic, and their children (III5 and III7) were affected. This indicates that the family has the characteristics of incomplete penetrance. In addition, the patients in this family had inconsistent presentations. Some had unilateral hand involvement, and some had bilateral hand involvement. Furthermore, the phenomenon was gradually aggravated with transmission from generation to generation. These complex clinical phenotypes may be related to gene interactions and genetic backgrounds, and these considerations need to be further studied.

Conclusion

Exon sequencing was used to further verify that there was a G > A mutation (HOXD13, c.917G > A, p.R306Q) in the second exon at position 917 of the HOXD13 gene in patients and carriers of this family. This mutation resulted in the change of arginine at position 306 to glutamine. This mutation could lead to the occurrence of syndactyly, and it is the main cause syndactyly in this Mongolian family. Our discovery also enriched the expression profile of the HOXD13 gene.