A Heterozygous Mutation in the Triple Helical Region of the Alpha 1 (II) Chain of the COL2A1 Protein Causes Non-Lethal Spondyloepiphyseal Dysplasia Congenita

Abstract

Objective:

Heterozygous pathogenic variants in the COL2A1 gene result in several clinical features including impaired skeletal growth, ocular and otolaryngological abnormalities. Missense mutations in the triple helical region of the COL2A1 protein have been associated with lethal spondyloepiphyseal dysplasia (SED). In this study, we aimed to identify the underlying cause of a case of SED congenita (SEDC) in a 27-month-old child.

Materials and Methods:

A patient who was diagnosed initially with osteochondrodysplasia underwent a detailed clinical and radiological examination to obtain a conclusive diagnosis. The patient did not show any clinical features of hypochondrogenesis. Whole exome sequencing of the COL2A1 gene was carried out to identify the underlying genetic cause of the disorder.

Results:

Variant annotation and filtration detected a heterozygous missense mutation c.1357G>A (p.G453S) in the exon 21 of the COL2A1 gene of the proband which was confirmed by Sanger sequencing. Neither parent carried the mvariant suggesting this was a new mutation.

Conclusion:

The COL2A1 mutation (c.1357G>A), identified in this case, results in more mild phenotype than other missense mutations in exon 21 which are known to cause lethal hypochondrogenesis. We showed, for the first time, that a missense mutation (p.G453S) in the triple helical region of the alpha 1 (II) chain of the COL2A1 protein underlies SEDC and is not always lethal.

Introduction

Mutations in COL2A1 (collagen type 2 alpha 1) gene is known to cause variety of clinical conditions including spondyloepiphyseal dysplasia congenita (SEDC) (Liu et al., 2016; Sangsin et al., 2016; Xiong et al., 2018), Kniest dysplasia (Al-Hashmi et al., 2013; Barat-Houari et al., 2016b), Stickler syndrome (Higuchi et al., 2017; Huang et al., 2018), osteoarthritis with chondrodysplasia (Jakkula et al., 2005), Legg-Calve-Perthes disease (Miyamoto et al., 2007; Kannu et al., 2011), and other skeletal conditions (Reinartz et al., 2017; Güneş et al., 2018; Liu et al., 2018). Conditions that arise as a result of COL2A1 gene mutations are collectively termed type II collagenopathies (Barat-Houari et al., 2016a). Both autosomal recessive and autosomal dominant phenotype have been reported due to COL2A1 mutations (Tham et al., 2015; Barat-Houari et al., 2016c; Xiong et al., 2018). Clinical characteristics of COL2A1 mutations include skeletal deformities (epiphyseal dysplasia, metaphyseal deformities, disproportionate short stature, midface hypoplasia, spinal deformity, and hypoplastic pelvis), ocular abnormalities, deafness, and cleft palate. However, intrafamilial and interfamilial phenotypic variabilities are common (Nagendran et al., 2012).

SEDC exists, in most cases, as an autosomal dominant condition with skeletal deformities at birth. SEDC is characterized by short-trunk dwarfism, odontoid hypoplasia, cervical spine subluxation, scoliosis, kyphosis, lumbar lordosis, coxa vara, genu valgum, clubfoot, metaphyseal changes, abnormal epiphyses, and flattened vertebral bodies. Extra-skeletal features include midface hypoplasia, sensorineural hearing loss, myopia and/or retinal degeneration with retinal detachment, micrognathia, thoracic hyperkyphosis, and cleft palate (Xia et al., 2007; Mark et al., 2011; Zhang et al., 2011; Veeravagu et al., 2013; Xu et al., 2014; Liu et al., 2016; Higuchi et al., 2017; Xiong et al., 2018).

Here, we studied a patient who was initially diagnosed as having osteochondrodysplasia. Whole exome sequencing identified a heterozygous missense mutation in the COL2A1 gene. Detailed clinical examination followed by extensive genetic analysis confirm the case as SEDC.

Materials and Methods

Patient

A 6-month-old female was referred to the Centre for Genetics and Inherited Diseases (CGID). The affected individual was delivered by a caesarean section. The subject underwent detailed clinical and radiological examination at the department of genetics, Madinah maternity and children hospital (MMCH), Saudi Arabia. The study was approved by Ethical Review Board of MMCH, Almadinah Almunawarah, Saudi Arabia. Signed informed consents were obtained from parents of the affected individual for genetic analysis and publication of photographs. Parents were called to the CGID and interviewed to record history of incidence of the disease and the disease status. Profile photographs of patients were taken for presentation/analysis of facial dysmorphism. Head circumferences were measured to exclude the microcephalic feature in the patient. Plain chest radiographs and an echocardiogram were taken to rule out any heart diseases. A computed tomography (CT) scan of the brain of the affected individual was also performed to rule out any brain abnormalities. Blood and urine biochemical tests were also performed to exclude any metabolic disorders.

Extraction of genomic DNA and targeted sequencing

Peripheral blood samples were collected from both parents and an affected child in an EDTA coated tube. Nucleic acid was extracted using QIAquick DNA extraction kit. DNA was quantified by Nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA) and Qubit fluorometer (Thermo Fisher Scientific).

Initially, 25 candidate genes (based on functional and expression data) were sequenced by Sanger approach. Ensembl genome browser^* was used to download the genomic sequences of the candidate genes. Primers were designed using primer3 software^† to amplify the exonic regions and flanking sites. The patient's genomic sequences were aligned to the reference sequence using BIOEDIT sequence alignment editor version 6.0.7 (Ibis Biosciences, Inc., Carlsbad, CA).

Sequencing of all the protein-coding genes

Genomic DNA from the affected individual was subjected to exome sequencing. Library preparation and exome enrichment was done by using SureSelect Target Enrichment kit (Agilent Technologies, Santa Clara, CA). DNA sequencing and cluster generation was carried out on NextSeq500 machine (Illumina, San Diego, CA). Exome sequencing protocol is described elsewhere (Hashmi et al., 2016). For analysis purposes, a BaseSpace (Illumina) was used to align fastq files to the reference genome. Variants were called by using genome analysis tool kit (GATK). Genomic variants were annotated and filtered by VariantStudio (Illumina).

Results

Clinical and radiological analysis of the patient

The affected individual, who is now 27 months old, was born to phenotypically normal parents and was recruited for this study (Fig. 1). Heights of both parents were in normal range. Father is 36 years old with a height of 165.4 cm and mother is 29 years old and her height is 153 cm.

FIG. 1.

Radiographs of an affected child showing typical features of spondyloepiphyseal dysplasia congenital. (a) Anterior posterior view of the skull. Large calavera and enlarged orbital regions are evident. (b) Lateral view of the spine showing platyspondyly (flat vertebrae) and posterior wedging of lumbar vertebrae. Anterior-posterior diameter is more than the height of the vertebrae. (c) Anterior posterior view of the chest, abdomen, pelvis, and bilateral femur. Bell-shaped chest, absent pubic bones, epiphysial dysplasia, and deformed ends of the long bones can be seen.

The age of the affected individual was 6 months at the time of recruitment. We were unable to record the growth parameters at the time of birth. However, growth parameters at the age of 6 months were recorded as a height of 39 cm less than 3rd percentile, head circumference (occipitofrontal diameter) of 42 cm on the 75th percentile, and a weight of 5.5 kg corresponding to the 25th percentile. Cognition and mentality were found preserved. A facial dysmorphism was found in terms of relative macrocephaly, flat midface with frontal bossing, hypertelorism, and depressed nasal bridge with a high arched palate. A patient was presented with a cleft of the soft palate. A short neck with severe trunk shortening and a barrel chest (pectus carinatum) deformity was also observed. Patient exhibits features of disproportionate short stature. Involvement of the spine was also observed with kyphoscoliosis and mild lumbar lordosis. A rhizomelic limb shortening with coxa vara; bilateral clubfoot deformity; and a mild degree of restricted movements of large joints including elbow, hip, and knee joints was also observed. Laboratory investigations including measurement of amino acid levels, creatine, and creatinine level were normal, making a metabolic disorder unlikely. Ophthalmic examination excluded issues related to eye/vision. Abdominal ultrasonography and echocardiogram values were in normal range. However, the abdomen was found mildly protuberant without any organomegaly. CT scanning of brain showed periventricular white matter hypo-density prominent in the frontal lobes suggestive of early hypoxic ischemic changes.

Skeletal survey showed relatively large calvarium, bell-shaped chest, platyspondyly, posterior wedging of the vertebrae, long bones metaphysical flaring, and absent pubic bones suggestive of SEDC (Fig. 1).

Screening candidate genes did not reveal any pathogenic mutation

Patient was referred to CGID with multiple clinical features including asphyxiating thoracic dystrophy, hypophosphatasia (based on low alkaline phosphatase level), atelogenesis defect, mild facial hypoplasia, short thoracic cage with short extremities, osteochondrodyspasia, short stature, and cleft palate. For this reason, we first selected and screened 25 previously known candidates genes that cause abovementioned phenotypes. This includes ALPL, CBFA1, CEP120, DYNC2H1, DYNC2L11, FLNA, FLNB, IFT10, IFT140, IFT172, IFT52, IFT80, KIAA0586, NEK1, RUNX2, RUNX3, SLC26A2, SOX9, TCTEX1D2, TTC21B, WDR19, WDR19, WDR34, WDR35, and WDR60. Sequencing data analysis identified an 18-bps heterozygous in-frame deletion variant (c.243-260delGGCGGCTGCGGCGGCGGC) in the RUNX2 gene. Initially, we considered this variant as a causal mutation for the patient's phenotype based on previous report (Zeng et al., 2018). However, further analysis of variant revealed that it is present in high frequency in variety of genome variation databases. Moreover, segregation analysis discovered the presence of variant in mother as well. Therefore, this variant was considered as a population polymorphism (Hashmi et al., 2018).

Exome data analysis identified a novel mutation in COL2A1 gene

WES libraries were generated using 51 Mb SureSelect V6 kits (Agilent Technologies). A 103 bps paired end reads were obtained with an average read depth of 130. Reads were aligned to the human genome reference sequence hg19 using Burrows-Wheeler Aligner (BWA). In total, more than 97,000 variants were obtained. Different filters were applied to obtain candidate variants in the coding regions and exon-intron splice junctions. The details of filtering steps are described elsewhere (Alkhiary et al., 2017). An autosomal recessive model did not reveal any variant of interest. A de novo model was used to select heterozygous alterations with minor allele frequency of <0.0l, and predicted protein effect of “damaging” identified a novel heterozygous missense variant in COL2A1 gene (ENST380518: c.1357G>A; p.Gly453Ser). The detected variant causes the substitution of a glycine into a serine (hydrophobic to hydrophilic substitution) residue at position 453. It is absent in gnomAD and dbSNP databases. Multiple in silico tools predict a deleterious effect of this missense change.

Sanger sequencing validate the presence of variant

Reference sequence of COL2A1 (ENST00000380518.7) was obtained from the ensembl genome browser.^‡ A primer pair (forward: 5′-GACACCCCTCCCTTCTTCTC-3′ and reverse: 5′-ACACCAGATTCTCTCCAGGG-3′) for polymerase chain reaction amplification of the candidate variant was designed using Primer3Plus software.^** The exonic segment was amplified in the DNA of patient and both parents followed by their sequencing using BigDye Terminator v3.1 Cycle Sequencing Kit and ABI 3500 Genetic Analyzer (Applied Biosystems, Foster City, CA). Sequence variant was identified through BioEdit sequence alignment editor version 6.0.7.^‡‡ Patient DNA sequence showed missense variant (c.1357G>A) in heterozygous state, however, parental DNA analysis and DNA analysis of unaffected population controls did not have the variant (Fig. 2).

FIG. 2.

Sequencing chromatogram showing partial sequence of exon 21 of the COL2A1 gene. (a) DNA sequence of the affected child showing G > A heterozygous mutation. (b) Sequence chromatogram of the control individual. Both parents also showed wild-type sequence. Arrow indicates the point of mutation. Color images are available online.

Discussion

Type II collagen is an extracellular protein of cartilaginous tissue and is required for normal endochondral development and cartilage homeostasis. NAD-dependent protein deacetylase Sirtuin 1 (SirT1) binds to the chromatin regions of COL2A1 promoter and enhancer and thus enhances COL2A1 gene expression (Dvir‐Ginzberg et al., 2008). Moreover, it has been shown that Set7/9 forms a protein complex with SirT1 on the promoter region of COL2A1 and enhances COL2A1 expression (Oppenheimer et al., 2014).

In this study, whole exome sequencing was performed on a DNA sample from a patient presented with osteochondrodysplasia. Data analysis identified an 18-bps heterozygous in-frame deletion variant (c.243-260delGGCGGCTGCGGCGGCGGC) in the RUNX2 gene. Initially, we considered this variant as a causation mutation for the patient's phenotype based on previous report (Zeng et al., 2018). However, further analysis of variant revealed that it is present in high frequency in different databases, apparently healthy individuals and in the DNA of healthy mother (Hashmi et al., 2018). Three-dimensional structures of the wild-type and mutant RUNX2 protein (p.Ala82_Ala87del) were analyzed and it was found that both wild-type and mutant proteins show similar secondary structure patterns. This lead to the argument that the 18-bps heterozygous in-frame deletion variant in RUNX2 gene is a population polymorphism and not a pathogenic mutation (Hashmi et al., 2018). Further filtration of exome variants detected a heterozygous missense mutation (c.1357G>A) in the COL2A1 gene. This variant substitutes a bulky serine residue for a glycine amino acid (p.G453S) in a triple helical region of alpha 1 (II) chain of the COL2A1 protein. Flanking region of this variant in exon 21 was amplified in DNA samples from proband and both parents to validate the presence of variant. Sanger sequencing validate the presence of variant in proband, however, it is absent in the DNA samples of both parents.

Heterozygous missense mutations in the triple helical region of alpha 1 (II) chain of the COL2A1 protein cause dominant negative effects. These mutations normally substitute bulky amino acid for glycine residues and thus may result in the assembly of abnormal alpha 1 (II) chain into trimeric collagen molecules. Abnormal collagen molecules cause spectrum of SED (Nishimura et al., 2005). Specifically, mutations in the exon 21 and 22 lead to the clinical phenotype of hypochondrogenesis, a lethal SED. Most carriers of heterozygous missense mutations in exon 21 and 22 died in the first year of life (Nishimura et al., 2005). However, in this case, the patient showed typical features of SEDC and the patient is alive at the age of 2.5 years. Characteristics features of hypochondrogenesis including mesomelia, square-shaped short iliac bone, short and broad neck of the femur, elongated distal fibula, short pedicle of lumbar spine, dorsal concavity of lumbar vertebral body, and shortening of distal ulna have not been observed in this case. Phenotypic variability can be attributed to incomplete penetrance or variable expressivity. Multiple factors have been identified as an underlying cause of reduced penetrance and variable expressivity including genetic modifiers, allelic variations, and complex genetic and environmental interaction (Cooper et al., 2013). We hypothesize that mutation(s) in regulatory regions are responsible for phenotypic variability and reduced penetrance. Mutations in regulatory regions cannot be detected using whole exome sequencing approach and therefore, the whole genome sequencing may reveal the deep intronic or regulatory regions variants as an underlying cause of the phenotypic variability and reduced penetrance in this patient.

In conclusion, the missense mutation (c.1357G>A; p.G453S) in the triple helical region of alpha 1 (II) chain of the COL2A1 protein underlies non-lethal SEDC.

Ethics Approval and Consent to Participate

All laboratory experimentations were approved by the Ethical Review Committee of Taibah University. All participants signed informed written consent.

Footnotes

Authors' Contributions

A.A. performed Sanger validation; F.A. recruited the case and performed clinical investigations; Y.N.K. performed skeletal survey and drafted X-ray report; S.A., J.A.H., and A.M.A. performed exome sequencing; A.U. performed exome data analysis; S.B. designed the study and prioritized variants; J.A.H. wrote the initial draft of the article. All authors contributed to and have approved the final article.

Acknowledgments

We are grateful to all the participants for their willingness to participate in this study. We are also thankful to Dr. Amir Ali Khan who performed thorough English editing and corrected the grammatical mistakes in the article.

Author Disclosure Statement

No competing financial interests exist.

*

†

‡

**

‡‡

References

Al-Hashmi

, Imtiaz

, Ramzan

, et al. (2013) Novel splice (IVS18 + 1G>C) mutation in COL2A1 causing Kniest dysplasia. Clin Dysmorphol, 22:39-41.

Alkhiary

, Ramzan

, Ilyas

, et al. (2017) Whole exome sequencing analysis identifies a missense variant in COL1A2 gene which causes osteogenesis imperfecta Type IV in a family from Saudi Arabia. J Musculoskelet Surg Res, 1:33-38.

Barat-Houari

, Baujat

, Tran Mau Them

, et al. (2016a) Confirmation of autosomal recessive inheritance of COL2A1 mutations in spondyloepiphyseal dysplasia congenita: lessons for genetic counseling. Am J Med Genet A, 170A:263-265.

Barat-Houari

, Dumont

, Fabre

, et al. (2016b) The expanding spectrum of COL2A1 gene variants in 136 patients with a skeletal dysplasia phenotype. Eur J Hum Genet, 24:992-1000.

Barat-Houari

, Sarrabay

, Gatinois

, et al. (2016c) Mutation update for COL2A1 gene variants associated with type II collagenopathies. Hum Mutat, 37:7-15.

Cooper

, Krawczak

, Polychronakos

, et al. (2013) Where genotype is not predictive of phenotype: towards an understanding of the molecular basis of reduced penetrance in human inherited disease. Hum Genet, 132:1077-1130.

Dvir-Ginzberg

, Gagarina

, Lee

, et al. (2008) Regulation of cartilage specific gene expression in human chondrocytes by SirT1 and nicotinamide phosphoribosyltransferase. J Biol Chem, 52:36300-36310.

Güneş

, Yeşil

, Beng

, et al. (2018) Longitudinal follow-up of two patients with dysspondyloenchondromatosis due to novel heterozygous mutations in COL2A1. Mol Syndromol, 9:134-140.

Hashmi

, Al-Harbi

, Ramzan

, et al. (2016) A novel splice-site mutation in the ASPM gene underlies autosomal recessive primary microcephaly. Ann Saudi Med, 36:391-396.

10.

Hashmi

, Almatrafi

, Latif

, et al. (2018) An 18 bps in-frame deletion mutation in RUNX2 gene is a population polymorphism rather than a pathogenic variant. Eur J Med Genet, 62:124-128.

11.

Higuchi

, Hasegawa

, Yamashita

, et al. (2017) A novel mutation in the COL2A1 gene in a patient with Stickler syndrome type 1: a case report and review of the literature. J Med Case Rep, 11:237.

12.

Huang

, Lin

, Chen

, et al. (2018) Targeted next-generation sequencing identifies two novel COL2A1 gene mutations in Stickler syndrome with bilateral retinal detachment. Int J Mol Med, 42:1819-1826.

13.

Jakkula

, Melkoniemi

, Kiviranta

, et al. (2005) The role of sequence variations within the genes encoding collagen II, IX and XI in non-syndromic, early-onset osteoarthritis. Osteoarthritis Cartilage, 13:497-507.

14.

Kannu

, Irving

, Aftimos

, Savarirayan

(2011) Two novel COL2A1 mutations associated with a Legg-Calvé-Perthes disease-like presentation. Clin Orthop Relat Res, 469:1785-1790.

15.

Liu

, Xiong

, Liu

, et al. (2018) COL2A1 mutation (c.3508G>A) leads to avascular necrosis of the femoral head in a Chinese family: a case report. Mol Med Rep, 18:254-260.

16.

Liu

, Pang

, Jiang

, et al. (2016) Novel COL2A1 mutations causing spondyloepiphyseal dysplasia congenita in three unrelated Chinese families. Eur Spine J, 25:2967-2974.

17.

Mark

, Torres-Martinez

, Lachman

, et al. (2011) Association of a p.Pro786Leu variant in COL2A1 with mild spondyloepiphyseal dysplasia congenita in a three-generation family. Am J Med Genet A, 155A:174-179.

18.

Miyamoto

, Matsuda

, Kitoh

, et al. (2007) A recurrent mutation in type II collagen gene causes Legg-Calvé-Perthes disease in a Japanese family. Hum Genet, 121:625-629.

19.

Nagendran

, Richards

, McNinch

, et al. (2012) Somatic mosaicism and the phenotypic expression of COL2A1 mutations. Am J Med Genet A, 158A:1204-1207.

20.

Nishimura

, Haga

, Kitoh

, et al. (2005) The phenotypic spectrum of COL2A1 mutations. Hum Mutat, 26:36-43.

21.

Oppenheimer

, Kumar

, Meir

, et al. (2014) Set7/9 impacts COL2A1 expression through binding and repression of SirT1 histone deacetylation. J Bone Miner Res, 29:348-360.

22.

Reinartz

, Mohwinkel

, Sürie

, et al. (2017) Germline mutation within COL2A1 associated with lethal chondrodysplasia in a polled Holstein family. BMC Genomics, 18:762.

23.

Sangsin

, Srichomthong

, Pongpanich

, et al. (2016) Whole-exome sequencing reveals a novel COL2A1 mutation in a patient with spondylo-epiphyseal dysplasia congenita. Genet Mol Res, 15:15017624.

24.

Tham

, Nishimura

, Geiberger

, et al. (2015) Autosomal recessive mutations in the COL2A1 gene cause severe spondyloepiphyseal dysplasia. Clin Genet, 87:496-498.

25.

Veeravagu

, Lad

, Camara-Quintana

, et al. (2013) Neurosurgical interventions for spondyloepiphyseal dysplasia congenita: clinical presentation and assessment of the literature. World Neurosurg, 80:437 .e1-8.

26.

Xia

, Cui

, Huang

, et al. (2007) Molecular prenatal diagnosis in 2 pregnancies at risk for spondyloepiphyseal dysplasia congenita. Clin Chim Acta, 387:153-157.

27.

Xiong

, Liu

, Xue

et al. (2018) A novel de novo mutation in COL2A1 leading to spondyloepiphyseal dysplasia congenita in a Chinese family. Hum Genome Var, 5:17059.

28.

, Qiu

, Zhu

, et al. (2014) A novel mutation in COL2A1 leading to spondyloepiphyseal dysplasia congenita in a three-generation family. Eur Spine J Suppl, 2:271-277.

29.

Zeng

, Wei

, Zhao

, et al. (2018) A novel 18-bp in-frame deletion mutation in RUNX2 causes cleidocranial dysplasia. Arch Oral Biol. 96:243-248.

30.

Zhang

, He

, Fu

, et al. (2011) Identification of three novel mutations in the COL2A1 gene in four unrelated Chinese families with spondyloepiphyseal dysplasia congenita. Biochem Biophys Res Commun, 413:504-508.