Original Article Received: June 24, 2014 Accepted after revision: November 27, 2014 Published online: January 27, 2015
Gynecol Obstet Invest DOI: 10.1159/000370196
Mutation Analysis of COL1A1 and COL1A2 in Fetuses with Osteogenesis Imperfecta Type II/III Wenbo Wang a Qichang Wu a Lin Cao b Li Sun a Yasong Xu a Qiwei Guo a
b
Prenatal Diagnosis Center of Xiamen’s Maternal and Child Health Care Hospital, Xiamen City, Fujian Province, and Beijing Genomics Institute at Shenzhen, Shenzhen, PR China
Key Words COL1A1 gene · COL1A2 gene · Gene sequencing · Mutation · Fetal osteogenesis imperfecta
Abstract Aim: To analyze COL1A1/2 mutations in prenatal-onset OI for determine the proportion of mutations in type I collagen genes among prenatal onset OI and to provide additional data for genotype-phenotype analyses. Material and Methods: Ten cases of severe fetal short-limb dwarfism detected by antenatal ultrasonography were referred to our center. Before the termination of pregnancy, cordocentesis was performed for fetal karyotype and COL1A1/2 gene sequencing analysis. Postmortem radiographic examination was performed at all instances for definitive diagnosis. Results: COL1A1 and COL1A2 SNP and mutations were identified in all the cases. Among these, one synonymous SNP and four synonymous SNPs were recognized in COL1A1/2, respectively, seven cases have distinct heterozygous mutations and six new COL1A1/2 gene mutations were identified. Conclusion: There has been substantial progress in the identification of the molecular defects responsible for skeletal dysplasias. With the constant increase in the number of identified mutations in COL1A1 and COL1A2, genotype-phenotype correlation is becoming increasingly pertinent. © 2015 S. Karger AG, Basel
© 2015 S. Karger AG, Basel 0378–7346/15/0000–0000$39.50/0 E-Mail [email protected] www.karger.com/goi
Introduction
Osteogenesis imperfecta (OI) is a heritable skeletal disorder characterized by bone fragility and, often, short stature [1]. The disease varies in severity from mild to perinatal lethal and exhibits both autosomal dominant and recessive inheritance patterns [2]. About 90% of individuals with OI are heterozygous for causative variants in the COL1A1 and COL1A2 genes [3]. However, these findings have mostly been restricted to surviving individuals, including early-childhood mortality cases. Some perinatally lethal OI (type II) and progressively deforming OI (type III) are often not considered in research data because of prenatal lethality. The proportion of mutations and the correlations between genotype and phenotype in prenatal-onset fetuses with OI are completely unknown. Recent developments in prenatal ultrasonography have made possible to identify fetuses with OI type II/III. In the case of prenatal-onset OI fetuses with clinically unaffected parents and no skeletal malformations in their family history, the causative variant mutations are almost de novo and are related with COL1A1 and COL1A2. Therefore, we analyzed COL1A1/2 mutations in five cases of OI type II and five cases of OI type III that were diagnosed by postmortem radiographs; we found that the COL1A1/2 mutation causes OI type II/III in fetuses. The aim of this study was Qichang Wu Prenatal Diagnosis Center of Xiamen’s Maternal and Child Health Care Hospital Xiamen City, Fujian Province (PR China) E-Mail qichang_wu @ 163.com
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
a
Materials and Methods Subject Population The Prenatal Diagnosis Center of Xiamen Maternal and Child Health Care Hospital is a regional referral center for fetuses with suspected anomalies and/or genetic syndromes. It provides prenatal services for a considerable percentage of suspected anomalous pregnancies in the southwest area of Fujian of mainland China. In China, pregnancy can be terminated in any trimester if the fetus has severe malformations. The study was approved by the Ethics Committee of Xiamen Maternal and Child Health Care Hospital. We diagnosed 10 cases of fetal OI between November 2010 and December 2012. The detected gestational ages of these cases ranged from 19 to 27 weeks, in women ranging from 20 to 35 years of age. Among these women, 7 were primigravida and 3 were multigravida. These women were in non-consanguineous marriages and had a normal course of pregnancy; their family histories were devoid of other reports of skeletal malformations. These cases of severe fetal short-limb dwarfism detected by antenatal ultrasonography were referred to our center. Fetal shortlimb dwarfism were defined according to sonographic measurements that demonstrated a fetal femur and/or humerus length that was >4 standard deviations below the mean femoral length (FL) and/or humeral length (HL) for the gestational age, as predicted by Hadlock et al. [4]. Since prenatal ultrasonography findings of these cases such as severe micromelia, narrow thorax, protuberant abdomen, bent bones, bone mineralization, multiple bone fractures, and polyhydramnios were predicted to have potential lethal outcomes, the families decided to terminate pregnancy. Before termination of pregnancy, cordocentesis was performed for fetal karyotype and COL1A1/2 molecular analysis. Postmortem radiographic examination was performed at all instances for definitive diagnosis, based on the description by Sillence et al. [2]. The diagnosis of OI type II was consistent with the radiographs, which showed notable undermineralization of the calvarium, multiple fractures of the ribs in a discontinuous pattern with normal rib parts in between callus formation (discontinuous beading), and broad long bones of the lower extremities that were deformed and shortened because of multiple fractures. The diagnosis of OI type III was also consistent with the radiographs, which showed diminished but visible mineralization of the calvarium, slender ribs without fractures, and multiple fractures of the femora. Collagen Gene Sequencing and Variant Detection First, qualified genomic DNA samples were randomly fragmented by using a Covaris shearing system into fragments of 100– 200 bp, and then adapters were ligated to both ends of the resulting fragments. The extracted DNA was then amplified by ligation-mediated polymerase chain reaction (LM-PCR), purified, and hybridized to the Nimblegen Human custom array for enrichment, and the non-hybridized fragments were washed out. Both non-captured and captured LM-PCR products were subjected to quantitative PCR to estimate the magnitude of enrichment. Each captured library was then loaded on a HiSeq2000 platform to perform high-
2
Gynecol Obstet Invest DOI: 10.1159/000370196
throughput sequencing independently to ensure that each sample met the desired average fold-coverage. Raw image files were processed using the CASAVA Software 1.7 (Illumina) for base calling adopting the default parameters, and the sequence of each individual was generated as 90-bp paired-end reads. The bioinformatics analysis began from the sequencing data (raw data) generated by the Illumina pipeline. First, the adapter sequence in the raw data was removed, and low-quality reads, which have too many Ns and low-quality bases, were discarded. Second, the sequencing reads were aligned to the reference genome sequence using Burrows-Wheeler Aligner (BWA), which provides flexible parameter setup and generates the alignment output in sequence alignment/ map (SAM) format. More information about SAM format is available at http://samtools.sourceforge.net/SAM1.pdf. For mapping, we used the human genome (hg19) as the reference genome. Moreover, we used SOAPsnp to detect single-nucleotide polymorphisms (SNPs). After identification of the SNPs, ANNOVAR was used to perform annotation and classification. To test the mutation sites further, PolyPhen (http://genetics.bwh.harvard.edu/pph/) and SIFT (http://sift.jcvi.org/www/sift chr coords submit. html) software were utilized to predict the pathogenic effect of the function of a protein.
Results
COL1A1 and COL1A2 SNPs One synonymous SNP was recognized in COL1A1 (p.Thr766Thr) in all the cases. Four synonymous SNPs were recognized in COL1A2 (p.Thr29Thr, p.Asp82Asp, p.Pro482Pro, and p.Val626Val). Only one non-synonymous SNP was recognized in each gene: p.Ala1075Thr in COL1A1 and p.Pro549Ala in COL1A2. COL1A1 and COL1A2 Mutations We identified mutations in type I/II collagen genes in 10 samples, among which seven cases were identified to have distinct heterozygous mutations. No deletions or duplications were found. The details are presented in table 1.
Discussion
In our study, the samples were selected according to the criterion of post-termination delivery radiographs that showed findings consistent with OI type II and III phenotypes as per the description provided by Sillence. In seven of the ten cases, distinct heterozygous type I collagen mutations were detected. Of the seven samples in which collagen mutations were identified, five had causative mutations in COL1A1 and two in COL1A2. No deletions or duplications were found. Among the five distinct COL1A1 causative mutations, four had not been previously detected (p.Gly689Asp, p.Gly695Ser, p.Gly419Glu, Wang/Wu/Cao/Sun/Xu/Guo
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
to suggest that sequence analysis of COL1A1/2 genes variants should be added to the diagnostic process of prenatal-onset fetuses.
Table 1. Mutations in COL1A1 and COL1A2
F3 F4 F5 F6 F7 F10 F11
Type
II II II II III III III
Gene
COL1A1 COL1A1 COL1A1 COL1A1 COL1A2 COL1A2 COL1A1
cDNA
c.2066G>A c.2083G>A c.1256G>A c.3856G>C c.2717G>A c.3124G>A c.3599G>C
Protein
p.Gly689Asp p.Gly695Ser p.Gly419Glu p.Val1289Leu p.Arg906His p.Gly1042Ser p.Gly1187Arg
Triple helix
511 517 241 1,111 816 952 1,009
Exon
31 31 19 49 42 47 48
Software prediction PolyPhen
SIFT
damaging damaging damaging damaging damaging damaging damaging
damaging damaging damaging damaging damaging damaging damaging
and p.Val1289Leu) and were diagnosed as type II OI. The two distinct COL1A2 causative mutations that were identified in our study have not been previously reported (p.Arg906His and p.Gly1042Ser); the diagnosis was type III OI in both cases. We identified mutations in samples from four subjects with type II OI and three subjects with type III OI. These results provide an estimate of the relative frequencies of the spectrum of mutations associated with prenatal-onset OI. With the current relatively low coverage of these mutations, most of the newly studied fetuses with OI will have previously unrecognized mutations. The COL1A1 and COL1A2 genes code for α1 and α2 chains of collagen type I. Type I collagen is a heterotrimer consisting of two α1 chains and one α2 chain [5]. It is initially synthesized as a proα chain with a propeptide at each end (N-propeptide and C-propeptide). The propeptides are necessary for proα chain association and triple helix formation. The triple helical domains are composed of uninterrupted repeats of Gly-X-Y tripeptides [6]. Numerous studies have shown that over 80% of lethal to moderately severe cases of OI are heterozygous for mutations of the COL1A1 or COL1A2 genes, which alter the structure of the triple helical domains or the carboxylterminal propeptides of the proα1 or proα2 chains of type I procollagen [7]. In our study, among the seven distinct COL1A1 and COL1A2 causative mutations, five resulted in substitution for triple-helix glycine residues and two resulted in non-glycine substitution within the Gly-X-Y triplet domain of the triple helix. To our knowledge, the mutation c.3856G>C (p.Val1289Leu) in COL1A1 and the mutation c.2717G>A (p.Arg906His) in COL1A2 have not been reported earlier. The contribution of the non-glycine substitution to the phenotype is not known. However, the mutation c.3856G>C (p.Val1289Leu) was identified in the C-propeptide region of COL1A1, which was reported to carry more severe than variants in the triple
helix region. Mutations of the C-terminus of the α chain can also cause dramatic changes in type I collagen metabolism [8]. Two exclusively lethal regions of glycine substitutions in the α1 chain (helix positions 691–823 and 910–964) and eight clusters of lethality in the α2 chain (helix positions 319–364, 451–502, 547–580, 622–637, 694–06, 757–811, 859–907, and 937–994) have been identified [9]. In our study, the four mutations in COL1A1 resulting in the substitution of glycines in the triple helix were all exceptions to the lethal regions, but among the four cases, three had been diagnosed as perinatally lethal OI (type II) by radiographic analysis. One mutation in COL1A2 (p.Gly1042Ser, helix position 952) was in the lethal region, but the case was diagnosed as progressively deforming OI (type III) by radiographic analysis. Although the identification of some lethal regions provides a useful rule of thumb, there are certainly many exceptions. Very little is known about how a specific type I collagen mutation leads to a particular phenotype. In general, the phenotypic severity depends on the affected α chain, position of the mutation, the substituting amino acid, or on the combination of these variables [10]. Our data suggest that multiple pathways lead to lethality. Models relating genotype to phenotype must capture the diversity of these mechanisms. There has been substantial progress in the identification of the molecular defects responsible for skeletal dysplasias. Many of these discoveries have led to the availability of DNA diagnostics for both molecular confirmation of ultrasonography and postmortem findings, as well as invasive prenatal diagnosis for at-risk families. With the discovery of COL1A1/2 genes variants as a cause of OI, sequence analysis of these genes should be added to the diagnostic process of prenatal-onset fetuses. Even in the absence of well-defined pathways and mechanisms, the accumulation of associations between mutations and clinical data is an important part of genetic counseling
COL1A1/2 Mutation in Prenatal OI
Gynecol Obstet Invest DOI: 10.1159/000370196
3
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
Subject
and contributes to the understanding of the etiological factors of OI for the evaluation and refinement of the current models relating genotype to phenotype. To date, more than 1,000 distinct variants in the COL1A1 and COL1A2 genes have been determined to cause OI [11]. With the constant increase in the number of identified mutations in COL1A1 and COL1A2, genotype-phenotype correlation is becoming increasingly pertinent.
Acknowledgment This study was supported by the Science and Technology Committee of Fujian Province (2011D008).
Disclosure Statement There are no conflicts of interest to declare.
References
4
Gynecol Obstet Invest DOI: 10.1159/000370196
6 Pepin M, Atkinson M, Starman BJ, Byers PH: Strategies and outcomes of prenatal diagnosis for osteogenesis imperfecta: a review of biochemical and molecular studies completed in 129 pregnancies. Prenat Diagn 1997; 17: 559– 570. 7 Byers PH, Tsipouras P, Bonadio JF, Starman BJ, Schwartz RC: Perinatal lethal osteogenesis imperfecta (OI type II): a biochemically heterogeneous disorder usually due to new mutations in the genes for type I collagen. Am J Hum Genet 1988;42:237–248. 8 Cole WG, Dalgleish R: Perinatal lethal osteogenesis imperfecta. J Med Genet 1995; 32: 284–289.
9 Marini JC, Forlino A, Cabral WA, et al: Consortium for osteogenesis imperfecta mutations in the helical domain of type I collagen: regions rich in lethal mutations align with collagen binding sites for integrins and proteoglycans. Hum Mutat 2007;28:209–221. 10 Bodian DL, Chan TF, Poon A, Schwarze U, Yang K, Byers PH, Kwok PY, Klein TE: Mutation and polymorphism spectrum in osteogenesis imperfecta type II: implications for genotype-phenotype relationships. Hum Mol Genet 2009;18:463–471. 11 van Dijk FS, Byers PH, Dalgleish R, Malfait F, Maugeri A, Rohrbach M, Symoens S, Sistermans EA, Pals G: EMQN best practice guidelines for the laboratory diagnosis of osteogenesis imperfecta. Eur J Hum Genet 2012; 20: 11–19.
Wang/Wu/Cao/Sun/Xu/Guo
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
1 Rauch F, Glorieux FH: Osteogenesis imperfecta. Lancet 2004;363:1377–1385. 2 Sillence DO, Senn A, Danks DM: Genetic heterogeneity in osteogenesis imperfecta. J Med Genet 1979;16:101–116. 3 Steiner RD, Adsit J, Basel D: COL1A1/2-Related Osteogenesis Imperfecta. Gene ReviewsTM. Seattle (WA): University of Washington, 1993–2013. 4 Hadlock FP, Deter RL, Harrist RB, Park SK: Estimating fetal age: computer-assisted analysis of multiple fetal growth parameters. Radiology 1984;152:497–501. 5 Ben Amor IM, Glorieux FH, Rauch F: Genotype-phenotype correlations in autosomal dominant osteogenesis imperfecta. J Osteoporos 2011;2011:540178.
Gynecol Obstet Invest DOI: 10.1159/000370196
Mutation Analysis of COL1A1 and COL1A2 in Fetuses with Osteogenesis Imperfecta Type II/III Wenbo Wang a Qichang Wu a Lin Cao b Li Sun a Yasong Xu a Qiwei Guo a
b
Prenatal Diagnosis Center of Xiamen’s Maternal and Child Health Care Hospital, Xiamen City, Fujian Province, and Beijing Genomics Institute at Shenzhen, Shenzhen, PR China
Key Words COL1A1 gene · COL1A2 gene · Gene sequencing · Mutation · Fetal osteogenesis imperfecta
Abstract Aim: To analyze COL1A1/2 mutations in prenatal-onset OI for determine the proportion of mutations in type I collagen genes among prenatal onset OI and to provide additional data for genotype-phenotype analyses. Material and Methods: Ten cases of severe fetal short-limb dwarfism detected by antenatal ultrasonography were referred to our center. Before the termination of pregnancy, cordocentesis was performed for fetal karyotype and COL1A1/2 gene sequencing analysis. Postmortem radiographic examination was performed at all instances for definitive diagnosis. Results: COL1A1 and COL1A2 SNP and mutations were identified in all the cases. Among these, one synonymous SNP and four synonymous SNPs were recognized in COL1A1/2, respectively, seven cases have distinct heterozygous mutations and six new COL1A1/2 gene mutations were identified. Conclusion: There has been substantial progress in the identification of the molecular defects responsible for skeletal dysplasias. With the constant increase in the number of identified mutations in COL1A1 and COL1A2, genotype-phenotype correlation is becoming increasingly pertinent. © 2015 S. Karger AG, Basel
© 2015 S. Karger AG, Basel 0378–7346/15/0000–0000$39.50/0 E-Mail [email protected] www.karger.com/goi
Introduction
Osteogenesis imperfecta (OI) is a heritable skeletal disorder characterized by bone fragility and, often, short stature [1]. The disease varies in severity from mild to perinatal lethal and exhibits both autosomal dominant and recessive inheritance patterns [2]. About 90% of individuals with OI are heterozygous for causative variants in the COL1A1 and COL1A2 genes [3]. However, these findings have mostly been restricted to surviving individuals, including early-childhood mortality cases. Some perinatally lethal OI (type II) and progressively deforming OI (type III) are often not considered in research data because of prenatal lethality. The proportion of mutations and the correlations between genotype and phenotype in prenatal-onset fetuses with OI are completely unknown. Recent developments in prenatal ultrasonography have made possible to identify fetuses with OI type II/III. In the case of prenatal-onset OI fetuses with clinically unaffected parents and no skeletal malformations in their family history, the causative variant mutations are almost de novo and are related with COL1A1 and COL1A2. Therefore, we analyzed COL1A1/2 mutations in five cases of OI type II and five cases of OI type III that were diagnosed by postmortem radiographs; we found that the COL1A1/2 mutation causes OI type II/III in fetuses. The aim of this study was Qichang Wu Prenatal Diagnosis Center of Xiamen’s Maternal and Child Health Care Hospital Xiamen City, Fujian Province (PR China) E-Mail qichang_wu @ 163.com
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
a
Materials and Methods Subject Population The Prenatal Diagnosis Center of Xiamen Maternal and Child Health Care Hospital is a regional referral center for fetuses with suspected anomalies and/or genetic syndromes. It provides prenatal services for a considerable percentage of suspected anomalous pregnancies in the southwest area of Fujian of mainland China. In China, pregnancy can be terminated in any trimester if the fetus has severe malformations. The study was approved by the Ethics Committee of Xiamen Maternal and Child Health Care Hospital. We diagnosed 10 cases of fetal OI between November 2010 and December 2012. The detected gestational ages of these cases ranged from 19 to 27 weeks, in women ranging from 20 to 35 years of age. Among these women, 7 were primigravida and 3 were multigravida. These women were in non-consanguineous marriages and had a normal course of pregnancy; their family histories were devoid of other reports of skeletal malformations. These cases of severe fetal short-limb dwarfism detected by antenatal ultrasonography were referred to our center. Fetal shortlimb dwarfism were defined according to sonographic measurements that demonstrated a fetal femur and/or humerus length that was >4 standard deviations below the mean femoral length (FL) and/or humeral length (HL) for the gestational age, as predicted by Hadlock et al. [4]. Since prenatal ultrasonography findings of these cases such as severe micromelia, narrow thorax, protuberant abdomen, bent bones, bone mineralization, multiple bone fractures, and polyhydramnios were predicted to have potential lethal outcomes, the families decided to terminate pregnancy. Before termination of pregnancy, cordocentesis was performed for fetal karyotype and COL1A1/2 molecular analysis. Postmortem radiographic examination was performed at all instances for definitive diagnosis, based on the description by Sillence et al. [2]. The diagnosis of OI type II was consistent with the radiographs, which showed notable undermineralization of the calvarium, multiple fractures of the ribs in a discontinuous pattern with normal rib parts in between callus formation (discontinuous beading), and broad long bones of the lower extremities that were deformed and shortened because of multiple fractures. The diagnosis of OI type III was also consistent with the radiographs, which showed diminished but visible mineralization of the calvarium, slender ribs without fractures, and multiple fractures of the femora. Collagen Gene Sequencing and Variant Detection First, qualified genomic DNA samples were randomly fragmented by using a Covaris shearing system into fragments of 100– 200 bp, and then adapters were ligated to both ends of the resulting fragments. The extracted DNA was then amplified by ligation-mediated polymerase chain reaction (LM-PCR), purified, and hybridized to the Nimblegen Human custom array for enrichment, and the non-hybridized fragments were washed out. Both non-captured and captured LM-PCR products were subjected to quantitative PCR to estimate the magnitude of enrichment. Each captured library was then loaded on a HiSeq2000 platform to perform high-
2
Gynecol Obstet Invest DOI: 10.1159/000370196
throughput sequencing independently to ensure that each sample met the desired average fold-coverage. Raw image files were processed using the CASAVA Software 1.7 (Illumina) for base calling adopting the default parameters, and the sequence of each individual was generated as 90-bp paired-end reads. The bioinformatics analysis began from the sequencing data (raw data) generated by the Illumina pipeline. First, the adapter sequence in the raw data was removed, and low-quality reads, which have too many Ns and low-quality bases, were discarded. Second, the sequencing reads were aligned to the reference genome sequence using Burrows-Wheeler Aligner (BWA), which provides flexible parameter setup and generates the alignment output in sequence alignment/ map (SAM) format. More information about SAM format is available at http://samtools.sourceforge.net/SAM1.pdf. For mapping, we used the human genome (hg19) as the reference genome. Moreover, we used SOAPsnp to detect single-nucleotide polymorphisms (SNPs). After identification of the SNPs, ANNOVAR was used to perform annotation and classification. To test the mutation sites further, PolyPhen (http://genetics.bwh.harvard.edu/pph/) and SIFT (http://sift.jcvi.org/www/sift chr coords submit. html) software were utilized to predict the pathogenic effect of the function of a protein.
Results
COL1A1 and COL1A2 SNPs One synonymous SNP was recognized in COL1A1 (p.Thr766Thr) in all the cases. Four synonymous SNPs were recognized in COL1A2 (p.Thr29Thr, p.Asp82Asp, p.Pro482Pro, and p.Val626Val). Only one non-synonymous SNP was recognized in each gene: p.Ala1075Thr in COL1A1 and p.Pro549Ala in COL1A2. COL1A1 and COL1A2 Mutations We identified mutations in type I/II collagen genes in 10 samples, among which seven cases were identified to have distinct heterozygous mutations. No deletions or duplications were found. The details are presented in table 1.
Discussion
In our study, the samples were selected according to the criterion of post-termination delivery radiographs that showed findings consistent with OI type II and III phenotypes as per the description provided by Sillence. In seven of the ten cases, distinct heterozygous type I collagen mutations were detected. Of the seven samples in which collagen mutations were identified, five had causative mutations in COL1A1 and two in COL1A2. No deletions or duplications were found. Among the five distinct COL1A1 causative mutations, four had not been previously detected (p.Gly689Asp, p.Gly695Ser, p.Gly419Glu, Wang/Wu/Cao/Sun/Xu/Guo
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
to suggest that sequence analysis of COL1A1/2 genes variants should be added to the diagnostic process of prenatal-onset fetuses.
Table 1. Mutations in COL1A1 and COL1A2
F3 F4 F5 F6 F7 F10 F11
Type
II II II II III III III
Gene
COL1A1 COL1A1 COL1A1 COL1A1 COL1A2 COL1A2 COL1A1
cDNA
c.2066G>A c.2083G>A c.1256G>A c.3856G>C c.2717G>A c.3124G>A c.3599G>C
Protein
p.Gly689Asp p.Gly695Ser p.Gly419Glu p.Val1289Leu p.Arg906His p.Gly1042Ser p.Gly1187Arg
Triple helix
511 517 241 1,111 816 952 1,009
Exon
31 31 19 49 42 47 48
Software prediction PolyPhen
SIFT
damaging damaging damaging damaging damaging damaging damaging
damaging damaging damaging damaging damaging damaging damaging
and p.Val1289Leu) and were diagnosed as type II OI. The two distinct COL1A2 causative mutations that were identified in our study have not been previously reported (p.Arg906His and p.Gly1042Ser); the diagnosis was type III OI in both cases. We identified mutations in samples from four subjects with type II OI and three subjects with type III OI. These results provide an estimate of the relative frequencies of the spectrum of mutations associated with prenatal-onset OI. With the current relatively low coverage of these mutations, most of the newly studied fetuses with OI will have previously unrecognized mutations. The COL1A1 and COL1A2 genes code for α1 and α2 chains of collagen type I. Type I collagen is a heterotrimer consisting of two α1 chains and one α2 chain [5]. It is initially synthesized as a proα chain with a propeptide at each end (N-propeptide and C-propeptide). The propeptides are necessary for proα chain association and triple helix formation. The triple helical domains are composed of uninterrupted repeats of Gly-X-Y tripeptides [6]. Numerous studies have shown that over 80% of lethal to moderately severe cases of OI are heterozygous for mutations of the COL1A1 or COL1A2 genes, which alter the structure of the triple helical domains or the carboxylterminal propeptides of the proα1 or proα2 chains of type I procollagen [7]. In our study, among the seven distinct COL1A1 and COL1A2 causative mutations, five resulted in substitution for triple-helix glycine residues and two resulted in non-glycine substitution within the Gly-X-Y triplet domain of the triple helix. To our knowledge, the mutation c.3856G>C (p.Val1289Leu) in COL1A1 and the mutation c.2717G>A (p.Arg906His) in COL1A2 have not been reported earlier. The contribution of the non-glycine substitution to the phenotype is not known. However, the mutation c.3856G>C (p.Val1289Leu) was identified in the C-propeptide region of COL1A1, which was reported to carry more severe than variants in the triple
helix region. Mutations of the C-terminus of the α chain can also cause dramatic changes in type I collagen metabolism [8]. Two exclusively lethal regions of glycine substitutions in the α1 chain (helix positions 691–823 and 910–964) and eight clusters of lethality in the α2 chain (helix positions 319–364, 451–502, 547–580, 622–637, 694–06, 757–811, 859–907, and 937–994) have been identified [9]. In our study, the four mutations in COL1A1 resulting in the substitution of glycines in the triple helix were all exceptions to the lethal regions, but among the four cases, three had been diagnosed as perinatally lethal OI (type II) by radiographic analysis. One mutation in COL1A2 (p.Gly1042Ser, helix position 952) was in the lethal region, but the case was diagnosed as progressively deforming OI (type III) by radiographic analysis. Although the identification of some lethal regions provides a useful rule of thumb, there are certainly many exceptions. Very little is known about how a specific type I collagen mutation leads to a particular phenotype. In general, the phenotypic severity depends on the affected α chain, position of the mutation, the substituting amino acid, or on the combination of these variables [10]. Our data suggest that multiple pathways lead to lethality. Models relating genotype to phenotype must capture the diversity of these mechanisms. There has been substantial progress in the identification of the molecular defects responsible for skeletal dysplasias. Many of these discoveries have led to the availability of DNA diagnostics for both molecular confirmation of ultrasonography and postmortem findings, as well as invasive prenatal diagnosis for at-risk families. With the discovery of COL1A1/2 genes variants as a cause of OI, sequence analysis of these genes should be added to the diagnostic process of prenatal-onset fetuses. Even in the absence of well-defined pathways and mechanisms, the accumulation of associations between mutations and clinical data is an important part of genetic counseling
COL1A1/2 Mutation in Prenatal OI
Gynecol Obstet Invest DOI: 10.1159/000370196
3
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
Subject
and contributes to the understanding of the etiological factors of OI for the evaluation and refinement of the current models relating genotype to phenotype. To date, more than 1,000 distinct variants in the COL1A1 and COL1A2 genes have been determined to cause OI [11]. With the constant increase in the number of identified mutations in COL1A1 and COL1A2, genotype-phenotype correlation is becoming increasingly pertinent.
Acknowledgment This study was supported by the Science and Technology Committee of Fujian Province (2011D008).
Disclosure Statement There are no conflicts of interest to declare.
References
4
Gynecol Obstet Invest DOI: 10.1159/000370196
6 Pepin M, Atkinson M, Starman BJ, Byers PH: Strategies and outcomes of prenatal diagnosis for osteogenesis imperfecta: a review of biochemical and molecular studies completed in 129 pregnancies. Prenat Diagn 1997; 17: 559– 570. 7 Byers PH, Tsipouras P, Bonadio JF, Starman BJ, Schwartz RC: Perinatal lethal osteogenesis imperfecta (OI type II): a biochemically heterogeneous disorder usually due to new mutations in the genes for type I collagen. Am J Hum Genet 1988;42:237–248. 8 Cole WG, Dalgleish R: Perinatal lethal osteogenesis imperfecta. J Med Genet 1995; 32: 284–289.
9 Marini JC, Forlino A, Cabral WA, et al: Consortium for osteogenesis imperfecta mutations in the helical domain of type I collagen: regions rich in lethal mutations align with collagen binding sites for integrins and proteoglycans. Hum Mutat 2007;28:209–221. 10 Bodian DL, Chan TF, Poon A, Schwarze U, Yang K, Byers PH, Kwok PY, Klein TE: Mutation and polymorphism spectrum in osteogenesis imperfecta type II: implications for genotype-phenotype relationships. Hum Mol Genet 2009;18:463–471. 11 van Dijk FS, Byers PH, Dalgleish R, Malfait F, Maugeri A, Rohrbach M, Symoens S, Sistermans EA, Pals G: EMQN best practice guidelines for the laboratory diagnosis of osteogenesis imperfecta. Eur J Hum Genet 2012; 20: 11–19.
Wang/Wu/Cao/Sun/Xu/Guo
Downloaded by: UCSF Library & CKM 128.218.166.245 - 5/14/2015 11:16:48 PM
1 Rauch F, Glorieux FH: Osteogenesis imperfecta. Lancet 2004;363:1377–1385. 2 Sillence DO, Senn A, Danks DM: Genetic heterogeneity in osteogenesis imperfecta. J Med Genet 1979;16:101–116. 3 Steiner RD, Adsit J, Basel D: COL1A1/2-Related Osteogenesis Imperfecta. Gene ReviewsTM. Seattle (WA): University of Washington, 1993–2013. 4 Hadlock FP, Deter RL, Harrist RB, Park SK: Estimating fetal age: computer-assisted analysis of multiple fetal growth parameters. Radiology 1984;152:497–501. 5 Ben Amor IM, Glorieux FH, Rauch F: Genotype-phenotype correlations in autosomal dominant osteogenesis imperfecta. J Osteoporos 2011;2011:540178.
