- Email: [email protected]
Identification and molecular characterization of two novel mutations in COL1A2 in two Chinese families with osteogenesis imperfecta
Available online at www.sciencedirect.com
Journal of Genetics and Genomics 38 (2011) 149e156 www.jgenetgenomics.org
Identification and molecular characterization of two novel mutations in COL1A2 in two Chinese families with osteogenesis imperfecta Zhenping Xu a,b,1, Yulei Li a,1, Xiangyang Zhang a, Fanming Zeng a, Mingxiong Yuan a, Mugen Liu a, Qing Kenneth Wang a,c,*, Jing Yu Liu a,* a
Key Laboratory of Molecular Biophysics of the Ministry of Education, Center for Human Genome Research, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China b Key Open Laboratory for Tissue Regeneration of Henan Universities, Department of Life Science and Technique, Xinxiang Medical University, Xinxiang 453003, China c Department of Molecular Cardiology, Cleveland Clinic, Cleveland, OH 44195, USA Received 10 August 2010; revised 21 January 2011; accepted 6 March 2011
Abstract Osteogenesis imperfecta (OI, also known as brittle bone disease) is caused mostly by mutations in two type I collagen genes, COL1A1 and COL1A2 encoding the pro-a1 (I) and pro-a2 (I) chains of type I collagen, respectively. Two Chinese families with autosomal dominant OI were identified and characterized. Linkage analysis revealed linkage of both families to COL1A2 on chromosome 7q21.3-q22.1. Mutational analysis was carried out using direct DNA sequence analysis. Two novel missense mutations, c.3350A>G and c.3305G>C, were identified in exon 49 of COL1A2 in the two families, respectively. The c.3305G>C mutation resulted in substitution of a glycine residue (G) by an alanine residue (A) at codon 1102 (p.G1102A), which was found to be mutated into serine (S), argine (R), aspartic acid (D), or valine (V) in other families. The c.3350A>G variant may be a de novo mutation resulting in p.Y1117C. Both mutations co-segregated with OI in respective families, and were not found in 100 normal controls. The G1102 and Y1117 residues were evolutionarily highly conserved from zebrafish to humans. Mutational analysis did not identify any mutation in the COX-2 gene (a modifier gene of OI). This study identifies two novel mutations p.G1102A and p.Y1117C that cause OI, significantly expands the spectrum of COL1A2 mutations causing OI, and has a significant implication in prenatal diagnosis of OI. Keywords: Osteogenesis imperfecta; Mutation; COL1A2; COX-2
1. Introduction Osteogenesis imperfecta (OI) is a clinically and genetically heterogeneous disorder characterized by bone fractures with minimal trauma. The reported incidence of OI varies from approximately 1/100000 to 1/25000 (Martin and Shapiro, 2007; Phillipi et al., 2008). The severity of OI varies from mild (OI type I, IV and V), severe (OI type III, IV, VI, VII and VIII), to
* Corresponding authors. Tel: þ86 27 8779 2649, fax: þ86 27 8779 4549. E-mail addresses: [email protected] (Q.K. Wang), liujy@mail. hust.edu.cn (J.Y. Liu). 1 These authors contributed equally to this study.
perinatally lethal (OI type II, VII and VIII). The inheritance mode of OI includes autosomal dominant (OI type IeV) and recessive (OI type VII and VIII) (Basel and Steiner, 2009). Severe and mild forms of OI all share the major clinical feature of bone fragility. Other clinical features have also been observed in differential OI types, including hypermobile joints, altered sclera, dentinogenesis imperfecta (DI), progressive conductive and sensorineural hearing loss, short stature, and bone deformity (Basel and Steiner, 2009). Type I, II, III and IV OI are caused by mutations in either COL1A1 or COL1A2 (Rauch and Glorieux, 2004; Basel and Steiner, 2009). Type VII and VIII OI are associated with mutations in CRTAP (Morello et al., 2006; Baldridge et al., 2008) and LEPRE1 (Cabral et al., 2007; Baldridge et al., 2008), respectively. The molecular pathogenic determinants for type V and
1673-8527/$ - see front matter Copyright Ó 2011, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, and Genetics Society of China. Published by Elsevier Limited and Science Press. All rights reserved. doi:10.1016/j.jgg.2011.03.002
150
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
VI have not been identified (Basel and Steiner, 2009). Recently, mutations in PPIP, FKBP10 and SERPINH1 were found in patients with severe OI (van Dijk et al., 2009; Alanay et al., 2010; Christiansen et al., 2010). Type I collagen is the major structural protein of the bone, skin, and other tissues, and consisted of two alpha-1 chains and one alpha-2 chain (Witecka et al., 2008). Each chain contains 338 units of uninterrupted repeats of the Gly-X-Y triplet. The conserved glycine residue in every third position of triple-helical domains is essential for the alpha chains to intertwine correctly and for the integrity of the protein, whereas the X and Y may be any amino acid residue (Bodian et al., 2008; Faqeih et al., 2009). The most common mutations in types IeIV OI result in the substitution of the glycine residue by another amino acid in the triple-helical domain of either alpha chain, which affects the stability of collagens (Marini et al., 2007; Rauch et al., 2010). The COL1A2 gene is located on chromosome 7q22.1, contains 52 exons and 51 introns, and encodes the alpha-2 chain of type I collagen. Over 400 OI-associated mutations in the COL1A2 gene have been deposited in the Osteogenesis Imperfecta Variant Database (https://oi.gene.le.ac.uk/variants. php?action=search_unique&select_db=COL1A2). Among the mutations, 14 were located in exon 49 of the COL1A2 gene, including 12 missense mutations and 2 microdeletions (summarized in Table 1). It is interesting that there is an apparent intrafamilial variability among individuals with the same mutation and interfamilial variability among individuals with the same type of OI (Basel and Steiner, 2009). The mutations at the C-terminus of alpha-2 chain appear to be more severe than those near the N-terminus, but mutations at N-terminal glycine residues of the alpha-1 chain caused more lethality (Bodian et al., 2008; Witecka et al., 2008).However, it remains difficult to establish correlation between a specific genotype and the resulting phenotype (Marini et al., 2007; Basel and Steiner,
2009; Faqeih et al., 2009). Therefore, identification of new mutations will significantly facilitate further clinically genotypeephenotype correlation studies. Recently, Brooks et al. (2009) found a genetic locus on chromosome 1q that influenced the clinical expression of OI. The COX-2 gene located within the 1q locus had a high expression level in osteoblasts and may regulate bone formation. Brooks et al. (2009) suggested the COX-2 gene might be a modifier gene of OI and affect the clinical phenotype associated with Type I collagen mutations. OI is a worldwide, hereditary, metabolic bone disorder, and currently there are no effective therapies for the affected OI individuals (Monti et al., 2010). Thus, genetic diagnostic technique may be important for providing families with accurate information and appropriate counseling regarding the health of the fetus. In this study, we studied two Chinese OI families with variable phenotypes of type I, III and IV OI and identified two novel heterozygous mutations, c.3350A>G and c.3305G>C in exon 49 of the COL1A2 gene. Furthermore, a prenatal genetic diagnosis using amniotic fluid cells was performed to determine whether two fetuses from a pregnant patient carried the mutation in family 1. 2. Materials and methods 2.1. Study subjects and isolation of genomic DNA Two Chinese families with OI were identified and characterized. Informed consent was obtained from patients or their legal guardians in accordance with the study protocols approved by the Ethics Committee of Huazhong University of Science and Technology, China. The participants were clinically examined at the Third Affiliated Hospital of Xinxiang Medical University and People’s Hospital of Xinye, China. Genomic DNA was extracted from peripheral whole blood
Table 1 Summary of mutations identified in exon 49 of COL1A2. Mutation
Type of OI
Reference
p.Gly1090Cys p.Gly1090Asp p.Gly1096Ala p.Gly1099_Pro1101del p.Gly1099Arg p.Pro1100_Gly1102del p.Gly1102Ser
OI III/IV OI III OI III OI IV OI III OI IV OI I OI I OI IV OI IV OI III/IV OI IV OI I/III/IV OI I/III/IV Unknown OI III OI IV OI I
Marini et al., 2007 Faqeih et al., 2009 Lu et al., 1995 Pace et al., 2001 Marini et al., 2007; Faqeih et al., 2009 Lund et al., 1996; Lee et al., 2006 Hartikka et al., 2004 Roschger et al., 2008 Wenstrup et al., 1988 Faqeih et al., 2009 Marini et al., 2007
p.Gly1102Arg p.Gly1102Asp p.Gly1102Val p.Gly1102Ala* p.Tyr1117Cys* p.Asp1120Ala p.Thr1148pro p.Cys1163Arg p.Asp1165Glu
Asterisk (*) indicates the novel mutation identified in this study.
This study This study Li, 2009 (http://oi.gene.le.ac.uk) Oliver et al., 1996 Pace et al., 2008 Ponnapakkam et al., 2006 (http://www.ashg.org/genetics/ashg06s/f30600.htm)
Note
Reported 2 times Reported 2 times
Reported 3 times
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
151
samples or amniotic fluid cells by using Wizard Genomic DNA Purification Kit (Promega, USA). 2.2. Linkage and mutation analyses There are four microsatellite markers (D17S1868, D17S798, D7S657 and D7S515) from the ABI PRISM Linkage Mapping Set MD10 panel that are closely linked to COL1A1 and COL1A2. These four markers were selected for linkage analysis in the two Chinese families with OI. Genotypes were analyzed using the GeneMapper 2 Software program (Applied Biosystems, USA). Two-point linkage analysis was performed using MLINK as previously described (Dai et al., 2010), assuming a gene frequency of 0.001, a penetrance rate of 100%, a phenocopy rate of 0%, and an allele frequency of 1/n, where n equals the number of alleles observed. The haplotypes were constructed manually. Mutational analysis of all exons and exon-intron boundaries of COL1A2 and a potential OI modifier gene COX-2 was carried out using direct DNA sequencing analysis with DNA samples from the probands in both families. 2.3. Restriction fragment length polymorphism (RFLP) analysis RFLP was used to determine whether the identified mutations co-segregated with the disease in the families and whether the mutations were absent in 100 normal Chinese individuals of Han nationality. The 49th exon/intron boundary of the COL1A2 gene was amplified using PCR primers 50 ttaagtccatccctgcaagtgt-30 (forward primer) and 50 ggtgctggttgggagaagatg-30 (reverse primer). The wild-type COL1A2 allele does not contain Pst I and Pvu II restriction sites. Mutation c.3350A>G creates the Pst I site, whereas mutation c.3305G>C creates the Pvu II site. The 594 bp PCR products were digested with Pvu II (NEB, USA) or Pst I (Fermentas, Canada) at 37 C overnight. The digested products were separated on 2% agarose gels to distinguish the wild type and mutant alleles. 2.4. Genetic analysis of amniotic fluid cells (AFC) Amniotic fluid samples were obtained after informed consent from the carrier patient (the eldest sister of the proband) in family 1 at the 18th week of gestation. Isolation and culture of AFCs was carried out as described by Bossolasco et al. (2006). DNA was extracted from AFC samples and used for analysis of the identified mutation in family 1. 3. Results 3.1. Clinical characteristics of the two Chinese families with OI The two probands in the two Chinese OI families were clinically characterized in detail. Their radiographs of the left femoral bone and tibiofibula are shown in Fig. 1.
Fig. 1. Radiographic images of the left femoral bone and tibiofibula of the probands in two Chinese families with OI. A: family 1 proband; B: tibiofibula, family 2 proband; C: femoral bone, family 2 proband. The arch curvature is clearly visible in both probands.
The proband in family 1 (Fig. 2A, individual III-3) was a 21 year old male patient with a height of 125 cm. He appeared to be healthy before the age of two years. The first fracture occurred at the age of two years. After the age of three years, multiple fractures of long bones were noted. His femoral bone and tibiofibula had an arch curvature (Fig. 1A). He was only mobile by wheelchair at the time of this study. There was a family history of bone fractures in family 1 (Fig. 2A). The proband’s elder sister (Fig. 2A, III-2) was 23 years of age and had a height of 150 cm. Multiple fractures occurred at the age of four years, and she appeared to be healthy before that. Her femoral bone and tibiofibula were similar to the proband. She could not walk at the present time. Individual III-1 had a less severe clinical symptom, a normal height of 158 cm, and only one fracture occurred at the age of seven. Family member II-3 had a normal height of 168 cm, multiple fractures and caput femoris necrosis after 50 years of age. After transplanting articulation of hip, he could walk by using a walking stick. All OI patients in the family had normal mental development. Other family members, including individual I-1 and I-2, were not affected with OI. The proband in family 2 (Fig. 2B, individual III-1) was a 23-year-old female patient with a height of 132 cm. She was born with blue sclera, which disappeared after she was ten years old. She displayed lower limb weakness when she imitated walking, and could not skip or jump. The first fracture occurred at the age of twenty months. Afterward, multiple fractures in two femoral bones occurred, but the number of fractures of the right long bone was less than that of the left
152
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
Fig. 2. Pedigree structure for Chinese OI family 1 (A) and family 2 (B). The affected individuals are shown with filled circles (females) or squares (males), and the unaffected individuals are shown with empty symbols. The probands are indicated by arrows. The deceased individuals are slashed. The haplotypes predicted from genotyping data of two markers D7S657 and D7S515 are shown below each symbol. The COL1A2 gene is located between D7S657 and D7S515. The 3-3 haplotype cosegregates with the disease in family 1. Individuals II:1 and II:3 carry the same haplotype, but II:1 is unaffected. Their parents are also normal. This suggests that a de novo mutation causing OI occurred in individual II:3 and was transmitted to III-1, III-2, and III-3 (A). The 1-2 haplotype cosegregates with the disease in family 2 (B).
one. She was only mobile with wheelchair at the time of this study. Radiography showed that her femoral bone and tibiofibula had an arch curvature (Figs. 1B and 2C), which was less severe than the proband in family 1 (Fig. 1A). There were multiple patients with OI in family 2 (Fig. 2B). The OI phenotypes varied in the family. Patient I-1 had a normal height of 168 cm, and a mild phenotype with only
two fractures. Patient II-4 had a similar phenotype as the proband and died in an accident. The proband’s mother (Fig. 2B, II-2) was 46 years old and her height was 152 cm. The first fracture occurred at the age of two years, and there were five fractures before the school age. Bone fractures occurred rarely after she was twelve years old. Her bone fractures occurred usually in the left long bone. The length of
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
153
the left leg was shorter than the right one. Patient III-2 had light blue sclera. Patient III-4 was an 11-year-old male with the first fracture at the birth and blue sclera, which disappeared later. The second fracture occurred when he was one year old. He had four fractures during the age of 3e4 years. The fractures occurred in his upper limbs.
from the pregnant patient was performed at the 18th week of gestation. The results showed that the first fetus conceived by the pregnant patient carried the c.3350A>G mutation, but the second fetus did not carry the mutation (Fig. 4B).
3.2. Linkage of two Chinese families with OI to the COL1A2 gene
In this study, we identified two Chinese families with autosomal dominant OI and variable phenotypes of type I, III and IV OI according to the original Sillence classification criteria (Sillence et al., 1979). Both families were likely linked to a chromosome 7 region where the COL1A2 gene is located because positive LOD scores were obtained for both families for two markers spanning the COL1A2 gene, D7S657 and D7S515. The LOD scores in family 1 were only 0.6 for both D7S657 and D7S515 due to the small size of the family, but haplotype 3-3 co-segregated with OI in the family. The LOD scores in family 2 were 2.11 and 1.96 for the two markers, respectively, and haplotype analysis showed that the two markers co-segregated with the disease in the family. Two novel mutations were subsequently identified in the two families in exon 49 of COL1A2, including c.3350A>G (p.Y1117C) in family 1 and c.3305G>C (p.G1102A) in family 2. The two mutations co-segregated with OI in corresponding families, and were not present in 100 normal controls. These results suggest that the two novel mutations are pathogenic to OI in the two families. To date, 14 OI mutations were previously reported in exon 49 of COL1A2, including 12 missense mutations and 2 microdeletions (Table 1). The p.G1102A mutation identified in this study occurs at an amino acid residue where four other mutations were previously reported, including p.G1102S, p.G1102R, p.G1102D, and p.G1102V (Table 1). The results in this study further demonstrate that the p.G1102 residue of COL1A2 is a hot spot for mutations. This suggests that p.G1102 is critical to the function of COL1A2. The p.Y1117C mutation identified in family 1 is a novel mutation which occurs at the C-terminal telopeptide (Fig. 4C). The p.Y1117 residue was highly conserved from zebrafish to humans during evolution (Fig. 4A). The mutation was identified in the proband’s father, but not in the proband’s uncle. As the proband’s father and uncle shared the identical haplotype (Fig. 2A), the proband’s grandparents are not patients, we suggest that p.Y1117C may be a de novo mutation. It is well-known that the severity of OI varies even among patients in the same family (Marini et al., 2007; Witecka et al., 2008; Basel and Steiner, 2009; Brooks et al., 2009; Dimasi et al., 2010). In this study, phenotypic variability was also observed between family 1 and family 2 as well as among patients in each family. The phenotypic variability may be related to variants in a modifier gene (Brooks et al., 2009). Brooks et al. (2009) suggested that variants in the COX-2 (PTGS2) gene, which is expressed in osteoblasts and may regulate bone formation, can affect the clinical phenotype associated with collagen mutations. Thus, we hypothesized that a COX-2 variant could mediate phenotypic variability of OI in the two Chinese families in this study. We carried out
Both Chinese OI families showed an autosomal dominant pattern of inheritance. Because about 90% of families with autosomal dominant OI were associated with mutations in the COL1A1 and COL1A2 genes (Bodian et al., 2009), we performed linkage analysis for the two families with four microsatellite markers (D17S1868, D17S798, D7S657 and D7S515) that span the COL1A1 and COL1A2 genes. Linkage results excluded the COL1A1 gene in both families because multiple recombinants were identified with markers D17S1868 and D17S798 (data not shown). However, positive linkage was found with the COL1A2 gene. Haplotype analysis for markers D7S657 and D7S515 was performed in the two families (Fig. 2A and B). Haplotype 3-3 cosegregated with OI in family 1, and haplotype 1-2 co-segregated with the disease in family 2. Maximum LOD scores for markers D7S657 and D7S515 were 0.6 and 0.6 in family 1, respectively, and 2.11 and 1.96 in family 2, respectively. 3.3. Identification of mutations in the COL1A2 gene All exons and exon-intron boundaries of COL1A2 from the two probands in two Chinese families were sequenced to identify mutations. We identified two novel missense mutations, c.3350A>G, p.Y1117C (Fig. 3A and B) in family 1 and c.3305G>C, p.G1102A (Fig. 3D and E) in family 2, both of which are located in exon 49. RFLP analysis was performed to confirm that the mutations were associated with the disease in two families. All four patients in family 1 were heterozygous for both mutant (333 bp and 261 bp) and wild-type allele (594 bp) (Fig. 3C), whereas the unaffected family members carried only the wildtype allele. Similar results were obtained in family 2 for mutation c.3305G>C (mutant allele, 381 bp and 213 bp; wildtype allele, 594 bp) (Fig. 3F). Both mutations were not found in 100 normal controls (data not shown). The results of RFLP analysis showed that the two mutations co-segregated with OI in corresponding families (Figs. 2A, 2B, 3C and 3F). In family 1, individual II-1 carried disease haplotype 3-3, but had the normal phenotype and did not carry the mutation. The proband’s grandparents were phenotypically normal. Thus, it is likely that the disease-causing mutation in the family may be arose de novo in individual II-3, which was then transmitted to three offsprings with OI. 3.4. Prenatal diagnosis for a pregnant patient in family 1 In family 1, the proband’s eldest sister had two times of gestation. A prenatal diagnosis using amniotic fluid samples
4. Discussion
154
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
Fig. 3. Identification of two novel heterozygote mutations in exon 49 of the COL1A2 gene. A>G (c.A3350G, p.Y1117C) in family 1 and G>C (c.G3305C, p.G1102A) in family 2. A and D: sequences for wild-type alleles from normal family members; B and E: sequences for patients showing heterozygote mutant alleles (A>G, mutation 1 in family 1) and (G>C, mutation 2 in family 2); C and F: results from RFLP analysis for family 1 and family 2, respectively. The mutant allele in family 1 showed two bands of 333 bp and 261 bp, whereas wild-type allele showed only one band of 594 bp (C). The mutant allele in family 2 showed two bands of 381 bp and 213 bp, and wild-type allele was 594 bp (F).
mutational analysis (direct DNA sequencing analysis) of all exons and exon-intron boundaries of the COX-2 gene, but failed to identify any COX-2 mutation in the two Chinese families. Future studies are needed to further explore the mechanisms for the phenotypic variability between or within OI families. There are more than 350 genetic skeletal disorders (Hall, 2002), and OI is a severe skeletal dysplasia that may manifest intrauterine fractures and even lethality in OI type II and type III. OI can be diagnosed by MRI, prenatal ultrasonography and conventional X-ray fetography (Hsieh et al., 2008; Solopova et al., 2008). However, type I and type IV OI may not manifest or show clinical features that are detectable by MRI,
ultrasonography and X-ray fetography before birth (Morgan and Marcus, 2010). Thus, a molecular genetic diagnosis may be valuable to identify a potentially affected fetus. In this study, we showed that the prenatal molecular diagnosis was an effective approach to determine whether or not a fetus carries the identified mutation in family 1. In conclusion, this study identified two novel mutations in exon 49 of the COL1A2 gene, including p.Y1117C and p.G1102A that cause OI. Interestingly, the p.G1102A mutation occurs at a residue that can be mutated into several other amino acid residues, including G1102S, G1102R, G1102D, and G1102V, demonstrating that pG1102 is a mutation hot spot. Our results expand the spectrum of COL1A2 mutations
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
155
Fig. 4. Prenatal diagnosis of COL1A2 mutation p.Y1117C and the structure of type I collagen alpha-2 chain. A: the alignment of amino acid residues around Y1117 revealed evolutionary conservation of this residue from zebrafish (Danio rerio) to humans (Homo sapiens); B: prenatal diagnosis of two fetuses from a pregnant patient in the family 1. The first fetus showed a mutant allele (two bands of 333 bp and 261 bp) and a wild allele (594 bp), but the second fetus showed only wild allele (594 bp). F, the father of the fetus; M, the mother of the fetus; No.1, the first fetus; No.2, the second fetus; C: structure of type I collagen alpha-2 chain and the location and types of COL1A2 mutations in exon 49 (asterisk, novel mutations identified in this study).
causing OI, and suggest that prenatal molecular diagnosis may be an effective technique to provide a family with accurate information about the fetus, allowing appropriate counseling. Acknowledgements This work was supported by the China Natural Science Foundation grants (Nos. 30670736 and 30972655) and the National Basic Research Program of China (973 Program) (No. 2007CB512002). We are grateful to the members of two families for their enthusiastic participation in this study. References Alanay, Y., Avaygan, H., Camacho, N., Utine, G.E., Boduroglu, K., Aktas, D., Alikasifoglu, M., Tuncbilek, E., Orhan, D., Bakar, F.T., Zabel, B., SupertiFurga, A., Bruckner-Tuderman, L., Curry, C.J., Pyott, S., Byers, P.H., Eyre, D.R., Baldridge, D., Lee, B., Merrill, A.E., Davis, E.C., Cohn, D.H., Akarsu, N., Krakow, D., 2010. Mutations in the gene encoding the RER protein FKBP65 cause autosomal-recessive osteogenesis imperfecta. Am. J. Hum. Genet. 86, 551e559. Baldridge, D., Schwarze, U., Morello, R., Lennington, J., Bertin, T.K., Pace, J. M., Pepin, M.G., Weis, M., Eyre, D.R., Walsh, J., Lambert, D., Green, A., Robinson, H., Michelson, M., Houge, G., Lindman, C., Martin, J., Ward, J., Lemyre, E., Mitchell, J.J., Krakow, D., Rimoin, D.L., Cohn, D.H., Byers, P. H., Lee, B., 2008. CRTAP and LEPRE1 mutations in recessive osteogenesis imperfecta. Hum. Mutat. 29, 1435e1442.
Basel, D., Steiner, R.D., 2009. Osteogenesis imperfecta: recent findings shed new light on this once well-understood condition. Genet. Med. 11, 375e385. Bodian, D.L., Madhan, B., Brodsky, B., Klein, T.E., 2008. Predicting the clinical lethality of osteogenesis imperfecta from collagen glycine mutations. Biochemistry 47, 5424e5432. Bodian, D.L., Chan, T.F., Poon, A., Schwarze, U., Yang, K., Byers, P.H., Kwok, P.Y., Klein, T.E., 2009. Mutation and polymorphism spectrum in osteogenesis imperfecta type II: implications for genotype-phenotype relationships. Hum. Mol. Genet. 18, 463e471. Bossolasco, P., Montemurro, T., Cova, L., Zangrossi, S., Calzarossa, C., Buiatiotis, S., Soligo, D., Bosari, S., Silani, V., Deliliers, G.L., Rebulla, P., Lazzari, L., 2006. Molecular and phenotypic characterization of human amniotic fluid cells and their differentiation potential. Cell Res. 16, 329e336. Brooks, P., Marcaillou, C., Vanpeene, M., Saraiva, J.P., Stockholm, D., Francke, S., Favis, V., Cohen, N., Rousseau, F., Tores, F., Lindenbaum, P., Hager, J., Philippi, A., 2009. Robust physical methods that enrich genomic regions identical by descent for linkage studies: confirmation of a locus for osteogenesis imperfecta. BMC Genet. 10, 16. Cabral, W.A., Chang, W., Barnes, A.M., Weis, M., Scott, M.A., Leikin, S., Makareeva, E., Kuznetsova, N.V., Rosenbaum, K.N., Tifft, C.J., Bulas, D.I., Kozma, C., Smith, P.A., Eyre, D.R., Marini, J.C., 2007. Prolyl 3-hydroxylase 1 deficiency causes a recessive metabolic bone disorder resembling lethal/severe osteogenesis imperfecta. Nat. Genet. 39, 359e365. Christiansen, H.E., Schwarze, U., Pyott, S.M., AlSwaid, A., Al Balwi, M., Alrasheed, S., Pepin, M.G., Weis, M.A., Eyre, D.R., Byers, P.H., 2010. Homozygosity for a missense mutation in SERPINH1, which encodes the collagen chaperone protein HSP47, results in severe recessive osteogenesis imperfecta. Am. J. Hum. Genet. 86, 389e398.
156
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
Dai, X., Gao, Y., Xu, Z., Cui, X., Liu, J., Li, Y., Xu, H., Liu, M., Wang, Q.K., Liu, J.Y., 2010. Identification of a novel genetic locus on chromosome 8p21.1-q11.23 for idiopathic basal ganglia calcification. Am. J. Med. Genet. B. Neuropsychiatr. Genet. 153B, 1305e1310. Dimasi, D.P., Chen, J.Y., Hewitt, A.W., Klebe, S., Davey, R., Stirling, J., Thompson, E., Forbes, R., Tan, T.Y., Savarirayan, R., Mackey, D.A., Healey, P.R., Mitchell, P., Burdon, K.P., Craig, J.E., 2010. Novel quantitative trait loci for central corneal thickness identified by candidate gene analysis of osteogenesis imperfecta genes. Hum. Genet. 127, 33e44. Faqeih, E., Roughley, P., Glorieux, F.H., Rauch, F., 2009. Osteogenesis imperfecta type III with intracranial hemorrhage and brachydactyly associated with mutations in exon 49 of COL1A2. Am. J. Med. Genet. A 149A, 461e465. Hall, C.M., 2002. International nosology and classification of constitutional disorders of bone (2001). Am. J. Med. Genet. 113, 65e77. Hartikka, H., Kuurila, K., Ko¨rkko¨, J., Kaitila, I., Gre´nman, R., Pynno¨nen, S., Hyland, J.C., Ala-Kokko, L., 2004. Lack of correlation between the type of COL1A1 or COL1A2 mutation and hearing loss in osteogenesis imperfecta patients. Hum. Mutat. 24, 147e154. Hsieh, C.T., Yeh, G.P., Wu, H.H., Wu, J.L., Chou, P.H., Lin, Y.M., 2008. Fetus with osteogenesis imperfecta presenting as increased nuchal translucency thickness in the first trimester. J. Clin. Ultrasound 36, 119e122. Lee, K.S., Song, H.R., Cho, T.J., Kim, H.J., Lee, T.M., Jin, H.S., Park, H.Y., Kang, S., Jung, S.C., Koo, S.K., 2006. Mutational spectrum of type I collagen genes in Korean patients with osteogenesis imperfecta. Hum. Mutat. 27, 599. Li, C., 2009. Increased Bone Mineral Density and Body Mass Index in Families of Osteogenesis Imperfecta. ASHG Meeting 2009 Program Number: 2630 (https://oi.gene.le.ac.uk). Lu, J., Costa, T., Cole, W.G., 1995. A novel G1006A substitution in the alpha 2(I) chain of type I collagen produces osteogenesis imperfecta type III. Hum. Mutat. 5, 175e178. Lund, A.M., Skovby, F., Schwartz, M., 1996. Deletion of a Gly-Pro-Pro repeat in the pro alpha2(I) chain of procollagen I in a family with dominant osteogenesis imperfecta type IV. Hum. Genet. 97, 287e290. Marini, J.C., Forlino, A., Cabral, W.A., Barnes, A.M., San Antonio, J.D., Milgrom, S., Hyland, J.C., Ko¨rkko¨, J., Prockop, D.J., De Paepe, A., Coucke, P., Symoens, S., Glorieux, F.H., Roughley, P.J., Lund, A.M., Kuurila-Svahn, K., Hartikka, H., Cohn, D.H., Krakow, D., Mottes, M., Schwarze, U., Chen, D., Yang, K., Kuslich, C., Troendle, J., Dalgleish, R., Byers, P.H., 2007. 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. 28, 209e221. Martin, E., Shapiro, J.R., 2007. Osteogenesis imperfecta: epidemiology and pathophysiology. Curr. Osteoporos. Rep. 5, 91e97. Monti, E., Mottes, M., Fraschini, P., Brunelli, P., Forlino, A., Venturi, G., Doro, F., Perlini, S., Cavarzere, P., Antoniazzi, F., 2010. Current and emerging treatments for the management of osteogenesis imperfecta. Ther. Clin. Risk Manag. 6, 367e381. Morello, R., Bertin, T.K., Chen, Y., Hicks, J., Tonachini, L., Monticone, M., Castagnola, P., Rauch, F., Glorieux, F.H., Vranka, J., Ba¨chinger, H.P., Pace, J.
M., Schwarze, U., Byers, P.H., Weis, M., Fernandes, R.J., Eyre, D.R., Yao, Z., Boyce, B.F., Lee, B., 2006. CRTAP is required for prolyl 3-hydroxylation and mutations cause recessive osteogenesis imperfecta. Cell 127, 291e304. Morgan, J.A., Marcus, P.S., 2010. Prenatal diagnosis and management of intrauterine fracture. Obstet. Gynecol. Surv. 65, 249e259. Oliver, J.E., Thompson, E.M., Pope, F.M., Nicholls, A.C., 1996. Mutation in the carboxy-terminal propeptide of the Pro alpha 1(I) chain of type I collagen in a child with severe osteogenesis imperfecta (OI type III): possible implications for protein folding. Hum. Mutat. 7, 318e326. Pace, J.M., Atkinson, M., Willing, M.C., Wallis, G., Byers, P.H., 2001. Deletions and duplications of Gly-Xaa-Yaa triplet repeats in the triple helical domains of type I collagen chains disrupt helix formation and result in several types of osteogenesis imperfecta. Hum. Mutat. 18, 319e326. Pace, J.M., Wiese, M., Drenguis, A.S., Kuznetsova, N., Leikin, S., Schwarze, U., Chen, D., Mooney, S.H., Unger, S., Byers, P.H., 2008. Defective C-propeptides of the proalpha2(I) chain of type I procollagen impede molecular assembly and result in osteogenesis imperfecta. J. Biol. Chem. 283, 16061e16067. Phillipi, C.A., Remmington, T., Steiner, R.D., 2008. Bisphosphonate therapy for osteogenesis imperfecta. Cochrane Database Syst. Rev. 4, CD005088. Ponnapakkam, T.P., Sledge, D., Gensure, R., 2006. A novel COL1A2 mutation leading to osteogenesis imperfecta type 1 in identical twins. Pediatrics Ochsner Clinic Foundation, New Orleans, LA. (http://www.ashg.org/ genetics/ashg06s/f30600.htm). Rauch, F., Glorieux, F.H., 2004. Osteogenesis imperfecta. Lancet 363, 1377e1385. Rauch, F., Lalic, L., Roughley, P., Glorieux, F.H., 2010. Genotype-phenotype correlations in nonlethal osteogenesis imperfecta caused by mutations in the helical domain of collagen type I. Eur. J. Hum. Genet. 18, 642e647. Roschger, P., Fratzl-Zelman, N., Misof, B.M., Glorieux, F.H., Klaushofer, K., Rauch, F., 2008. Evidence that abnormal high bone mineralization in growing children with osteogenesis imperfecta is not associated with specific collagen mutations. Calcif. Tissue Int. 82, 263e270. Sillence, D.O., Senn, A., Danks, D.M., 1979. Genetic heterogeneity in osteogenesis imperfecta. J. Med. Genet. 16, 101e116. Solopova, A., Wisser, J., Huisman, T.A., 2008. Osteogenesis imperfecta type II: fetal magnetic resonance imaging findings. Fetal Diagn. Ther. 24, 361e367. van Dijk, F.S., Nesbitt, I.M., Zwikstra, E.H., Nikkels, P.G., Piersma, S.R., Fratantoni, S.A., Jimenez, C.R., Huizer, M., Morsman, A.C., Cobben, J.M., van Roij, M.H., Elting, M.W., Verbeke, J.I., Wijnaendts, L.C., Shaw, N.J., Ho¨gler, W., McKeown, C., Sistermans, E.A., Dalton, A., MeijersHeijboer, H., Pals, G., 2009. PPIB mutations cause severe osteogenesis imperfecta. Am. J. Hum. Genet. 85, 521e527. Wenstrup, R.J., Cohn, D.H., Cohen, T., Byers, P.H., 1988. Arginine for glycine substitution in the triple-helical domain of the products of one alpha 2(I) collagen allele (COL1A2) produces the osteogenesis imperfecta type IV phenotype. J. Biol. Chem. 263, 7734e7740. Witecka, J., Augusciak-Duma, A.M., Kruczek, A., Szyd1o, A., Lesiak, M., Krzak, M., Pietrzyk, J.J., Ma¨nnikko¨, M., Sieron, A.L., 2008. Two novel COL1A1 mutations in patients with osteogenesis imperfecta (OI) affect the stability of the collagen type I triple-helix. J. Appl. Genet. 49, 283e295.
Journal of Genetics and Genomics 38 (2011) 149e156 www.jgenetgenomics.org
Identification and molecular characterization of two novel mutations in COL1A2 in two Chinese families with osteogenesis imperfecta Zhenping Xu a,b,1, Yulei Li a,1, Xiangyang Zhang a, Fanming Zeng a, Mingxiong Yuan a, Mugen Liu a, Qing Kenneth Wang a,c,*, Jing Yu Liu a,* a
Key Laboratory of Molecular Biophysics of the Ministry of Education, Center for Human Genome Research, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China b Key Open Laboratory for Tissue Regeneration of Henan Universities, Department of Life Science and Technique, Xinxiang Medical University, Xinxiang 453003, China c Department of Molecular Cardiology, Cleveland Clinic, Cleveland, OH 44195, USA Received 10 August 2010; revised 21 January 2011; accepted 6 March 2011
Abstract Osteogenesis imperfecta (OI, also known as brittle bone disease) is caused mostly by mutations in two type I collagen genes, COL1A1 and COL1A2 encoding the pro-a1 (I) and pro-a2 (I) chains of type I collagen, respectively. Two Chinese families with autosomal dominant OI were identified and characterized. Linkage analysis revealed linkage of both families to COL1A2 on chromosome 7q21.3-q22.1. Mutational analysis was carried out using direct DNA sequence analysis. Two novel missense mutations, c.3350A>G and c.3305G>C, were identified in exon 49 of COL1A2 in the two families, respectively. The c.3305G>C mutation resulted in substitution of a glycine residue (G) by an alanine residue (A) at codon 1102 (p.G1102A), which was found to be mutated into serine (S), argine (R), aspartic acid (D), or valine (V) in other families. The c.3350A>G variant may be a de novo mutation resulting in p.Y1117C. Both mutations co-segregated with OI in respective families, and were not found in 100 normal controls. The G1102 and Y1117 residues were evolutionarily highly conserved from zebrafish to humans. Mutational analysis did not identify any mutation in the COX-2 gene (a modifier gene of OI). This study identifies two novel mutations p.G1102A and p.Y1117C that cause OI, significantly expands the spectrum of COL1A2 mutations causing OI, and has a significant implication in prenatal diagnosis of OI. Keywords: Osteogenesis imperfecta; Mutation; COL1A2; COX-2
1. Introduction Osteogenesis imperfecta (OI) is a clinically and genetically heterogeneous disorder characterized by bone fractures with minimal trauma. The reported incidence of OI varies from approximately 1/100000 to 1/25000 (Martin and Shapiro, 2007; Phillipi et al., 2008). The severity of OI varies from mild (OI type I, IV and V), severe (OI type III, IV, VI, VII and VIII), to
* Corresponding authors. Tel: þ86 27 8779 2649, fax: þ86 27 8779 4549. E-mail addresses: [email protected] (Q.K. Wang), liujy@mail. hust.edu.cn (J.Y. Liu). 1 These authors contributed equally to this study.
perinatally lethal (OI type II, VII and VIII). The inheritance mode of OI includes autosomal dominant (OI type IeV) and recessive (OI type VII and VIII) (Basel and Steiner, 2009). Severe and mild forms of OI all share the major clinical feature of bone fragility. Other clinical features have also been observed in differential OI types, including hypermobile joints, altered sclera, dentinogenesis imperfecta (DI), progressive conductive and sensorineural hearing loss, short stature, and bone deformity (Basel and Steiner, 2009). Type I, II, III and IV OI are caused by mutations in either COL1A1 or COL1A2 (Rauch and Glorieux, 2004; Basel and Steiner, 2009). Type VII and VIII OI are associated with mutations in CRTAP (Morello et al., 2006; Baldridge et al., 2008) and LEPRE1 (Cabral et al., 2007; Baldridge et al., 2008), respectively. The molecular pathogenic determinants for type V and
1673-8527/$ - see front matter Copyright Ó 2011, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, and Genetics Society of China. Published by Elsevier Limited and Science Press. All rights reserved. doi:10.1016/j.jgg.2011.03.002
150
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
VI have not been identified (Basel and Steiner, 2009). Recently, mutations in PPIP, FKBP10 and SERPINH1 were found in patients with severe OI (van Dijk et al., 2009; Alanay et al., 2010; Christiansen et al., 2010). Type I collagen is the major structural protein of the bone, skin, and other tissues, and consisted of two alpha-1 chains and one alpha-2 chain (Witecka et al., 2008). Each chain contains 338 units of uninterrupted repeats of the Gly-X-Y triplet. The conserved glycine residue in every third position of triple-helical domains is essential for the alpha chains to intertwine correctly and for the integrity of the protein, whereas the X and Y may be any amino acid residue (Bodian et al., 2008; Faqeih et al., 2009). The most common mutations in types IeIV OI result in the substitution of the glycine residue by another amino acid in the triple-helical domain of either alpha chain, which affects the stability of collagens (Marini et al., 2007; Rauch et al., 2010). The COL1A2 gene is located on chromosome 7q22.1, contains 52 exons and 51 introns, and encodes the alpha-2 chain of type I collagen. Over 400 OI-associated mutations in the COL1A2 gene have been deposited in the Osteogenesis Imperfecta Variant Database (https://oi.gene.le.ac.uk/variants. php?action=search_unique&select_db=COL1A2). Among the mutations, 14 were located in exon 49 of the COL1A2 gene, including 12 missense mutations and 2 microdeletions (summarized in Table 1). It is interesting that there is an apparent intrafamilial variability among individuals with the same mutation and interfamilial variability among individuals with the same type of OI (Basel and Steiner, 2009). The mutations at the C-terminus of alpha-2 chain appear to be more severe than those near the N-terminus, but mutations at N-terminal glycine residues of the alpha-1 chain caused more lethality (Bodian et al., 2008; Witecka et al., 2008).However, it remains difficult to establish correlation between a specific genotype and the resulting phenotype (Marini et al., 2007; Basel and Steiner,
2009; Faqeih et al., 2009). Therefore, identification of new mutations will significantly facilitate further clinically genotypeephenotype correlation studies. Recently, Brooks et al. (2009) found a genetic locus on chromosome 1q that influenced the clinical expression of OI. The COX-2 gene located within the 1q locus had a high expression level in osteoblasts and may regulate bone formation. Brooks et al. (2009) suggested the COX-2 gene might be a modifier gene of OI and affect the clinical phenotype associated with Type I collagen mutations. OI is a worldwide, hereditary, metabolic bone disorder, and currently there are no effective therapies for the affected OI individuals (Monti et al., 2010). Thus, genetic diagnostic technique may be important for providing families with accurate information and appropriate counseling regarding the health of the fetus. In this study, we studied two Chinese OI families with variable phenotypes of type I, III and IV OI and identified two novel heterozygous mutations, c.3350A>G and c.3305G>C in exon 49 of the COL1A2 gene. Furthermore, a prenatal genetic diagnosis using amniotic fluid cells was performed to determine whether two fetuses from a pregnant patient carried the mutation in family 1. 2. Materials and methods 2.1. Study subjects and isolation of genomic DNA Two Chinese families with OI were identified and characterized. Informed consent was obtained from patients or their legal guardians in accordance with the study protocols approved by the Ethics Committee of Huazhong University of Science and Technology, China. The participants were clinically examined at the Third Affiliated Hospital of Xinxiang Medical University and People’s Hospital of Xinye, China. Genomic DNA was extracted from peripheral whole blood
Table 1 Summary of mutations identified in exon 49 of COL1A2. Mutation
Type of OI
Reference
p.Gly1090Cys p.Gly1090Asp p.Gly1096Ala p.Gly1099_Pro1101del p.Gly1099Arg p.Pro1100_Gly1102del p.Gly1102Ser
OI III/IV OI III OI III OI IV OI III OI IV OI I OI I OI IV OI IV OI III/IV OI IV OI I/III/IV OI I/III/IV Unknown OI III OI IV OI I
Marini et al., 2007 Faqeih et al., 2009 Lu et al., 1995 Pace et al., 2001 Marini et al., 2007; Faqeih et al., 2009 Lund et al., 1996; Lee et al., 2006 Hartikka et al., 2004 Roschger et al., 2008 Wenstrup et al., 1988 Faqeih et al., 2009 Marini et al., 2007
p.Gly1102Arg p.Gly1102Asp p.Gly1102Val p.Gly1102Ala* p.Tyr1117Cys* p.Asp1120Ala p.Thr1148pro p.Cys1163Arg p.Asp1165Glu
Asterisk (*) indicates the novel mutation identified in this study.
This study This study Li, 2009 (http://oi.gene.le.ac.uk) Oliver et al., 1996 Pace et al., 2008 Ponnapakkam et al., 2006 (http://www.ashg.org/genetics/ashg06s/f30600.htm)
Note
Reported 2 times Reported 2 times
Reported 3 times
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
151
samples or amniotic fluid cells by using Wizard Genomic DNA Purification Kit (Promega, USA). 2.2. Linkage and mutation analyses There are four microsatellite markers (D17S1868, D17S798, D7S657 and D7S515) from the ABI PRISM Linkage Mapping Set MD10 panel that are closely linked to COL1A1 and COL1A2. These four markers were selected for linkage analysis in the two Chinese families with OI. Genotypes were analyzed using the GeneMapper 2 Software program (Applied Biosystems, USA). Two-point linkage analysis was performed using MLINK as previously described (Dai et al., 2010), assuming a gene frequency of 0.001, a penetrance rate of 100%, a phenocopy rate of 0%, and an allele frequency of 1/n, where n equals the number of alleles observed. The haplotypes were constructed manually. Mutational analysis of all exons and exon-intron boundaries of COL1A2 and a potential OI modifier gene COX-2 was carried out using direct DNA sequencing analysis with DNA samples from the probands in both families. 2.3. Restriction fragment length polymorphism (RFLP) analysis RFLP was used to determine whether the identified mutations co-segregated with the disease in the families and whether the mutations were absent in 100 normal Chinese individuals of Han nationality. The 49th exon/intron boundary of the COL1A2 gene was amplified using PCR primers 50 ttaagtccatccctgcaagtgt-30 (forward primer) and 50 ggtgctggttgggagaagatg-30 (reverse primer). The wild-type COL1A2 allele does not contain Pst I and Pvu II restriction sites. Mutation c.3350A>G creates the Pst I site, whereas mutation c.3305G>C creates the Pvu II site. The 594 bp PCR products were digested with Pvu II (NEB, USA) or Pst I (Fermentas, Canada) at 37 C overnight. The digested products were separated on 2% agarose gels to distinguish the wild type and mutant alleles. 2.4. Genetic analysis of amniotic fluid cells (AFC) Amniotic fluid samples were obtained after informed consent from the carrier patient (the eldest sister of the proband) in family 1 at the 18th week of gestation. Isolation and culture of AFCs was carried out as described by Bossolasco et al. (2006). DNA was extracted from AFC samples and used for analysis of the identified mutation in family 1. 3. Results 3.1. Clinical characteristics of the two Chinese families with OI The two probands in the two Chinese OI families were clinically characterized in detail. Their radiographs of the left femoral bone and tibiofibula are shown in Fig. 1.
Fig. 1. Radiographic images of the left femoral bone and tibiofibula of the probands in two Chinese families with OI. A: family 1 proband; B: tibiofibula, family 2 proband; C: femoral bone, family 2 proband. The arch curvature is clearly visible in both probands.
The proband in family 1 (Fig. 2A, individual III-3) was a 21 year old male patient with a height of 125 cm. He appeared to be healthy before the age of two years. The first fracture occurred at the age of two years. After the age of three years, multiple fractures of long bones were noted. His femoral bone and tibiofibula had an arch curvature (Fig. 1A). He was only mobile by wheelchair at the time of this study. There was a family history of bone fractures in family 1 (Fig. 2A). The proband’s elder sister (Fig. 2A, III-2) was 23 years of age and had a height of 150 cm. Multiple fractures occurred at the age of four years, and she appeared to be healthy before that. Her femoral bone and tibiofibula were similar to the proband. She could not walk at the present time. Individual III-1 had a less severe clinical symptom, a normal height of 158 cm, and only one fracture occurred at the age of seven. Family member II-3 had a normal height of 168 cm, multiple fractures and caput femoris necrosis after 50 years of age. After transplanting articulation of hip, he could walk by using a walking stick. All OI patients in the family had normal mental development. Other family members, including individual I-1 and I-2, were not affected with OI. The proband in family 2 (Fig. 2B, individual III-1) was a 23-year-old female patient with a height of 132 cm. She was born with blue sclera, which disappeared after she was ten years old. She displayed lower limb weakness when she imitated walking, and could not skip or jump. The first fracture occurred at the age of twenty months. Afterward, multiple fractures in two femoral bones occurred, but the number of fractures of the right long bone was less than that of the left
152
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
Fig. 2. Pedigree structure for Chinese OI family 1 (A) and family 2 (B). The affected individuals are shown with filled circles (females) or squares (males), and the unaffected individuals are shown with empty symbols. The probands are indicated by arrows. The deceased individuals are slashed. The haplotypes predicted from genotyping data of two markers D7S657 and D7S515 are shown below each symbol. The COL1A2 gene is located between D7S657 and D7S515. The 3-3 haplotype cosegregates with the disease in family 1. Individuals II:1 and II:3 carry the same haplotype, but II:1 is unaffected. Their parents are also normal. This suggests that a de novo mutation causing OI occurred in individual II:3 and was transmitted to III-1, III-2, and III-3 (A). The 1-2 haplotype cosegregates with the disease in family 2 (B).
one. She was only mobile with wheelchair at the time of this study. Radiography showed that her femoral bone and tibiofibula had an arch curvature (Figs. 1B and 2C), which was less severe than the proband in family 1 (Fig. 1A). There were multiple patients with OI in family 2 (Fig. 2B). The OI phenotypes varied in the family. Patient I-1 had a normal height of 168 cm, and a mild phenotype with only
two fractures. Patient II-4 had a similar phenotype as the proband and died in an accident. The proband’s mother (Fig. 2B, II-2) was 46 years old and her height was 152 cm. The first fracture occurred at the age of two years, and there were five fractures before the school age. Bone fractures occurred rarely after she was twelve years old. Her bone fractures occurred usually in the left long bone. The length of
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
153
the left leg was shorter than the right one. Patient III-2 had light blue sclera. Patient III-4 was an 11-year-old male with the first fracture at the birth and blue sclera, which disappeared later. The second fracture occurred when he was one year old. He had four fractures during the age of 3e4 years. The fractures occurred in his upper limbs.
from the pregnant patient was performed at the 18th week of gestation. The results showed that the first fetus conceived by the pregnant patient carried the c.3350A>G mutation, but the second fetus did not carry the mutation (Fig. 4B).
3.2. Linkage of two Chinese families with OI to the COL1A2 gene
In this study, we identified two Chinese families with autosomal dominant OI and variable phenotypes of type I, III and IV OI according to the original Sillence classification criteria (Sillence et al., 1979). Both families were likely linked to a chromosome 7 region where the COL1A2 gene is located because positive LOD scores were obtained for both families for two markers spanning the COL1A2 gene, D7S657 and D7S515. The LOD scores in family 1 were only 0.6 for both D7S657 and D7S515 due to the small size of the family, but haplotype 3-3 co-segregated with OI in the family. The LOD scores in family 2 were 2.11 and 1.96 for the two markers, respectively, and haplotype analysis showed that the two markers co-segregated with the disease in the family. Two novel mutations were subsequently identified in the two families in exon 49 of COL1A2, including c.3350A>G (p.Y1117C) in family 1 and c.3305G>C (p.G1102A) in family 2. The two mutations co-segregated with OI in corresponding families, and were not present in 100 normal controls. These results suggest that the two novel mutations are pathogenic to OI in the two families. To date, 14 OI mutations were previously reported in exon 49 of COL1A2, including 12 missense mutations and 2 microdeletions (Table 1). The p.G1102A mutation identified in this study occurs at an amino acid residue where four other mutations were previously reported, including p.G1102S, p.G1102R, p.G1102D, and p.G1102V (Table 1). The results in this study further demonstrate that the p.G1102 residue of COL1A2 is a hot spot for mutations. This suggests that p.G1102 is critical to the function of COL1A2. The p.Y1117C mutation identified in family 1 is a novel mutation which occurs at the C-terminal telopeptide (Fig. 4C). The p.Y1117 residue was highly conserved from zebrafish to humans during evolution (Fig. 4A). The mutation was identified in the proband’s father, but not in the proband’s uncle. As the proband’s father and uncle shared the identical haplotype (Fig. 2A), the proband’s grandparents are not patients, we suggest that p.Y1117C may be a de novo mutation. It is well-known that the severity of OI varies even among patients in the same family (Marini et al., 2007; Witecka et al., 2008; Basel and Steiner, 2009; Brooks et al., 2009; Dimasi et al., 2010). In this study, phenotypic variability was also observed between family 1 and family 2 as well as among patients in each family. The phenotypic variability may be related to variants in a modifier gene (Brooks et al., 2009). Brooks et al. (2009) suggested that variants in the COX-2 (PTGS2) gene, which is expressed in osteoblasts and may regulate bone formation, can affect the clinical phenotype associated with collagen mutations. Thus, we hypothesized that a COX-2 variant could mediate phenotypic variability of OI in the two Chinese families in this study. We carried out
Both Chinese OI families showed an autosomal dominant pattern of inheritance. Because about 90% of families with autosomal dominant OI were associated with mutations in the COL1A1 and COL1A2 genes (Bodian et al., 2009), we performed linkage analysis for the two families with four microsatellite markers (D17S1868, D17S798, D7S657 and D7S515) that span the COL1A1 and COL1A2 genes. Linkage results excluded the COL1A1 gene in both families because multiple recombinants were identified with markers D17S1868 and D17S798 (data not shown). However, positive linkage was found with the COL1A2 gene. Haplotype analysis for markers D7S657 and D7S515 was performed in the two families (Fig. 2A and B). Haplotype 3-3 cosegregated with OI in family 1, and haplotype 1-2 co-segregated with the disease in family 2. Maximum LOD scores for markers D7S657 and D7S515 were 0.6 and 0.6 in family 1, respectively, and 2.11 and 1.96 in family 2, respectively. 3.3. Identification of mutations in the COL1A2 gene All exons and exon-intron boundaries of COL1A2 from the two probands in two Chinese families were sequenced to identify mutations. We identified two novel missense mutations, c.3350A>G, p.Y1117C (Fig. 3A and B) in family 1 and c.3305G>C, p.G1102A (Fig. 3D and E) in family 2, both of which are located in exon 49. RFLP analysis was performed to confirm that the mutations were associated with the disease in two families. All four patients in family 1 were heterozygous for both mutant (333 bp and 261 bp) and wild-type allele (594 bp) (Fig. 3C), whereas the unaffected family members carried only the wildtype allele. Similar results were obtained in family 2 for mutation c.3305G>C (mutant allele, 381 bp and 213 bp; wildtype allele, 594 bp) (Fig. 3F). Both mutations were not found in 100 normal controls (data not shown). The results of RFLP analysis showed that the two mutations co-segregated with OI in corresponding families (Figs. 2A, 2B, 3C and 3F). In family 1, individual II-1 carried disease haplotype 3-3, but had the normal phenotype and did not carry the mutation. The proband’s grandparents were phenotypically normal. Thus, it is likely that the disease-causing mutation in the family may be arose de novo in individual II-3, which was then transmitted to three offsprings with OI. 3.4. Prenatal diagnosis for a pregnant patient in family 1 In family 1, the proband’s eldest sister had two times of gestation. A prenatal diagnosis using amniotic fluid samples
4. Discussion
154
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
Fig. 3. Identification of two novel heterozygote mutations in exon 49 of the COL1A2 gene. A>G (c.A3350G, p.Y1117C) in family 1 and G>C (c.G3305C, p.G1102A) in family 2. A and D: sequences for wild-type alleles from normal family members; B and E: sequences for patients showing heterozygote mutant alleles (A>G, mutation 1 in family 1) and (G>C, mutation 2 in family 2); C and F: results from RFLP analysis for family 1 and family 2, respectively. The mutant allele in family 1 showed two bands of 333 bp and 261 bp, whereas wild-type allele showed only one band of 594 bp (C). The mutant allele in family 2 showed two bands of 381 bp and 213 bp, and wild-type allele was 594 bp (F).
mutational analysis (direct DNA sequencing analysis) of all exons and exon-intron boundaries of the COX-2 gene, but failed to identify any COX-2 mutation in the two Chinese families. Future studies are needed to further explore the mechanisms for the phenotypic variability between or within OI families. There are more than 350 genetic skeletal disorders (Hall, 2002), and OI is a severe skeletal dysplasia that may manifest intrauterine fractures and even lethality in OI type II and type III. OI can be diagnosed by MRI, prenatal ultrasonography and conventional X-ray fetography (Hsieh et al., 2008; Solopova et al., 2008). However, type I and type IV OI may not manifest or show clinical features that are detectable by MRI,
ultrasonography and X-ray fetography before birth (Morgan and Marcus, 2010). Thus, a molecular genetic diagnosis may be valuable to identify a potentially affected fetus. In this study, we showed that the prenatal molecular diagnosis was an effective approach to determine whether or not a fetus carries the identified mutation in family 1. In conclusion, this study identified two novel mutations in exon 49 of the COL1A2 gene, including p.Y1117C and p.G1102A that cause OI. Interestingly, the p.G1102A mutation occurs at a residue that can be mutated into several other amino acid residues, including G1102S, G1102R, G1102D, and G1102V, demonstrating that pG1102 is a mutation hot spot. Our results expand the spectrum of COL1A2 mutations
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
155
Fig. 4. Prenatal diagnosis of COL1A2 mutation p.Y1117C and the structure of type I collagen alpha-2 chain. A: the alignment of amino acid residues around Y1117 revealed evolutionary conservation of this residue from zebrafish (Danio rerio) to humans (Homo sapiens); B: prenatal diagnosis of two fetuses from a pregnant patient in the family 1. The first fetus showed a mutant allele (two bands of 333 bp and 261 bp) and a wild allele (594 bp), but the second fetus showed only wild allele (594 bp). F, the father of the fetus; M, the mother of the fetus; No.1, the first fetus; No.2, the second fetus; C: structure of type I collagen alpha-2 chain and the location and types of COL1A2 mutations in exon 49 (asterisk, novel mutations identified in this study).
causing OI, and suggest that prenatal molecular diagnosis may be an effective technique to provide a family with accurate information about the fetus, allowing appropriate counseling. Acknowledgements This work was supported by the China Natural Science Foundation grants (Nos. 30670736 and 30972655) and the National Basic Research Program of China (973 Program) (No. 2007CB512002). We are grateful to the members of two families for their enthusiastic participation in this study. References Alanay, Y., Avaygan, H., Camacho, N., Utine, G.E., Boduroglu, K., Aktas, D., Alikasifoglu, M., Tuncbilek, E., Orhan, D., Bakar, F.T., Zabel, B., SupertiFurga, A., Bruckner-Tuderman, L., Curry, C.J., Pyott, S., Byers, P.H., Eyre, D.R., Baldridge, D., Lee, B., Merrill, A.E., Davis, E.C., Cohn, D.H., Akarsu, N., Krakow, D., 2010. Mutations in the gene encoding the RER protein FKBP65 cause autosomal-recessive osteogenesis imperfecta. Am. J. Hum. Genet. 86, 551e559. Baldridge, D., Schwarze, U., Morello, R., Lennington, J., Bertin, T.K., Pace, J. M., Pepin, M.G., Weis, M., Eyre, D.R., Walsh, J., Lambert, D., Green, A., Robinson, H., Michelson, M., Houge, G., Lindman, C., Martin, J., Ward, J., Lemyre, E., Mitchell, J.J., Krakow, D., Rimoin, D.L., Cohn, D.H., Byers, P. H., Lee, B., 2008. CRTAP and LEPRE1 mutations in recessive osteogenesis imperfecta. Hum. Mutat. 29, 1435e1442.
Basel, D., Steiner, R.D., 2009. Osteogenesis imperfecta: recent findings shed new light on this once well-understood condition. Genet. Med. 11, 375e385. Bodian, D.L., Madhan, B., Brodsky, B., Klein, T.E., 2008. Predicting the clinical lethality of osteogenesis imperfecta from collagen glycine mutations. Biochemistry 47, 5424e5432. Bodian, D.L., Chan, T.F., Poon, A., Schwarze, U., Yang, K., Byers, P.H., Kwok, P.Y., Klein, T.E., 2009. Mutation and polymorphism spectrum in osteogenesis imperfecta type II: implications for genotype-phenotype relationships. Hum. Mol. Genet. 18, 463e471. Bossolasco, P., Montemurro, T., Cova, L., Zangrossi, S., Calzarossa, C., Buiatiotis, S., Soligo, D., Bosari, S., Silani, V., Deliliers, G.L., Rebulla, P., Lazzari, L., 2006. Molecular and phenotypic characterization of human amniotic fluid cells and their differentiation potential. Cell Res. 16, 329e336. Brooks, P., Marcaillou, C., Vanpeene, M., Saraiva, J.P., Stockholm, D., Francke, S., Favis, V., Cohen, N., Rousseau, F., Tores, F., Lindenbaum, P., Hager, J., Philippi, A., 2009. Robust physical methods that enrich genomic regions identical by descent for linkage studies: confirmation of a locus for osteogenesis imperfecta. BMC Genet. 10, 16. Cabral, W.A., Chang, W., Barnes, A.M., Weis, M., Scott, M.A., Leikin, S., Makareeva, E., Kuznetsova, N.V., Rosenbaum, K.N., Tifft, C.J., Bulas, D.I., Kozma, C., Smith, P.A., Eyre, D.R., Marini, J.C., 2007. Prolyl 3-hydroxylase 1 deficiency causes a recessive metabolic bone disorder resembling lethal/severe osteogenesis imperfecta. Nat. Genet. 39, 359e365. Christiansen, H.E., Schwarze, U., Pyott, S.M., AlSwaid, A., Al Balwi, M., Alrasheed, S., Pepin, M.G., Weis, M.A., Eyre, D.R., Byers, P.H., 2010. Homozygosity for a missense mutation in SERPINH1, which encodes the collagen chaperone protein HSP47, results in severe recessive osteogenesis imperfecta. Am. J. Hum. Genet. 86, 389e398.
156
Z. Xu et al. / Journal of Genetics and Genomics 38 (2011) 149e156
Dai, X., Gao, Y., Xu, Z., Cui, X., Liu, J., Li, Y., Xu, H., Liu, M., Wang, Q.K., Liu, J.Y., 2010. Identification of a novel genetic locus on chromosome 8p21.1-q11.23 for idiopathic basal ganglia calcification. Am. J. Med. Genet. B. Neuropsychiatr. Genet. 153B, 1305e1310. Dimasi, D.P., Chen, J.Y., Hewitt, A.W., Klebe, S., Davey, R., Stirling, J., Thompson, E., Forbes, R., Tan, T.Y., Savarirayan, R., Mackey, D.A., Healey, P.R., Mitchell, P., Burdon, K.P., Craig, J.E., 2010. Novel quantitative trait loci for central corneal thickness identified by candidate gene analysis of osteogenesis imperfecta genes. Hum. Genet. 127, 33e44. Faqeih, E., Roughley, P., Glorieux, F.H., Rauch, F., 2009. Osteogenesis imperfecta type III with intracranial hemorrhage and brachydactyly associated with mutations in exon 49 of COL1A2. Am. J. Med. Genet. A 149A, 461e465. Hall, C.M., 2002. International nosology and classification of constitutional disorders of bone (2001). Am. J. Med. Genet. 113, 65e77. Hartikka, H., Kuurila, K., Ko¨rkko¨, J., Kaitila, I., Gre´nman, R., Pynno¨nen, S., Hyland, J.C., Ala-Kokko, L., 2004. Lack of correlation between the type of COL1A1 or COL1A2 mutation and hearing loss in osteogenesis imperfecta patients. Hum. Mutat. 24, 147e154. Hsieh, C.T., Yeh, G.P., Wu, H.H., Wu, J.L., Chou, P.H., Lin, Y.M., 2008. Fetus with osteogenesis imperfecta presenting as increased nuchal translucency thickness in the first trimester. J. Clin. Ultrasound 36, 119e122. Lee, K.S., Song, H.R., Cho, T.J., Kim, H.J., Lee, T.M., Jin, H.S., Park, H.Y., Kang, S., Jung, S.C., Koo, S.K., 2006. Mutational spectrum of type I collagen genes in Korean patients with osteogenesis imperfecta. Hum. Mutat. 27, 599. Li, C., 2009. Increased Bone Mineral Density and Body Mass Index in Families of Osteogenesis Imperfecta. ASHG Meeting 2009 Program Number: 2630 (https://oi.gene.le.ac.uk). Lu, J., Costa, T., Cole, W.G., 1995. A novel G1006A substitution in the alpha 2(I) chain of type I collagen produces osteogenesis imperfecta type III. Hum. Mutat. 5, 175e178. Lund, A.M., Skovby, F., Schwartz, M., 1996. Deletion of a Gly-Pro-Pro repeat in the pro alpha2(I) chain of procollagen I in a family with dominant osteogenesis imperfecta type IV. Hum. Genet. 97, 287e290. Marini, J.C., Forlino, A., Cabral, W.A., Barnes, A.M., San Antonio, J.D., Milgrom, S., Hyland, J.C., Ko¨rkko¨, J., Prockop, D.J., De Paepe, A., Coucke, P., Symoens, S., Glorieux, F.H., Roughley, P.J., Lund, A.M., Kuurila-Svahn, K., Hartikka, H., Cohn, D.H., Krakow, D., Mottes, M., Schwarze, U., Chen, D., Yang, K., Kuslich, C., Troendle, J., Dalgleish, R., Byers, P.H., 2007. 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. 28, 209e221. Martin, E., Shapiro, J.R., 2007. Osteogenesis imperfecta: epidemiology and pathophysiology. Curr. Osteoporos. Rep. 5, 91e97. Monti, E., Mottes, M., Fraschini, P., Brunelli, P., Forlino, A., Venturi, G., Doro, F., Perlini, S., Cavarzere, P., Antoniazzi, F., 2010. Current and emerging treatments for the management of osteogenesis imperfecta. Ther. Clin. Risk Manag. 6, 367e381. Morello, R., Bertin, T.K., Chen, Y., Hicks, J., Tonachini, L., Monticone, M., Castagnola, P., Rauch, F., Glorieux, F.H., Vranka, J., Ba¨chinger, H.P., Pace, J.
M., Schwarze, U., Byers, P.H., Weis, M., Fernandes, R.J., Eyre, D.R., Yao, Z., Boyce, B.F., Lee, B., 2006. CRTAP is required for prolyl 3-hydroxylation and mutations cause recessive osteogenesis imperfecta. Cell 127, 291e304. Morgan, J.A., Marcus, P.S., 2010. Prenatal diagnosis and management of intrauterine fracture. Obstet. Gynecol. Surv. 65, 249e259. Oliver, J.E., Thompson, E.M., Pope, F.M., Nicholls, A.C., 1996. Mutation in the carboxy-terminal propeptide of the Pro alpha 1(I) chain of type I collagen in a child with severe osteogenesis imperfecta (OI type III): possible implications for protein folding. Hum. Mutat. 7, 318e326. Pace, J.M., Atkinson, M., Willing, M.C., Wallis, G., Byers, P.H., 2001. Deletions and duplications of Gly-Xaa-Yaa triplet repeats in the triple helical domains of type I collagen chains disrupt helix formation and result in several types of osteogenesis imperfecta. Hum. Mutat. 18, 319e326. Pace, J.M., Wiese, M., Drenguis, A.S., Kuznetsova, N., Leikin, S., Schwarze, U., Chen, D., Mooney, S.H., Unger, S., Byers, P.H., 2008. Defective C-propeptides of the proalpha2(I) chain of type I procollagen impede molecular assembly and result in osteogenesis imperfecta. J. Biol. Chem. 283, 16061e16067. Phillipi, C.A., Remmington, T., Steiner, R.D., 2008. Bisphosphonate therapy for osteogenesis imperfecta. Cochrane Database Syst. Rev. 4, CD005088. Ponnapakkam, T.P., Sledge, D., Gensure, R., 2006. A novel COL1A2 mutation leading to osteogenesis imperfecta type 1 in identical twins. Pediatrics Ochsner Clinic Foundation, New Orleans, LA. (http://www.ashg.org/ genetics/ashg06s/f30600.htm). Rauch, F., Glorieux, F.H., 2004. Osteogenesis imperfecta. Lancet 363, 1377e1385. Rauch, F., Lalic, L., Roughley, P., Glorieux, F.H., 2010. Genotype-phenotype correlations in nonlethal osteogenesis imperfecta caused by mutations in the helical domain of collagen type I. Eur. J. Hum. Genet. 18, 642e647. Roschger, P., Fratzl-Zelman, N., Misof, B.M., Glorieux, F.H., Klaushofer, K., Rauch, F., 2008. Evidence that abnormal high bone mineralization in growing children with osteogenesis imperfecta is not associated with specific collagen mutations. Calcif. Tissue Int. 82, 263e270. Sillence, D.O., Senn, A., Danks, D.M., 1979. Genetic heterogeneity in osteogenesis imperfecta. J. Med. Genet. 16, 101e116. Solopova, A., Wisser, J., Huisman, T.A., 2008. Osteogenesis imperfecta type II: fetal magnetic resonance imaging findings. Fetal Diagn. Ther. 24, 361e367. van Dijk, F.S., Nesbitt, I.M., Zwikstra, E.H., Nikkels, P.G., Piersma, S.R., Fratantoni, S.A., Jimenez, C.R., Huizer, M., Morsman, A.C., Cobben, J.M., van Roij, M.H., Elting, M.W., Verbeke, J.I., Wijnaendts, L.C., Shaw, N.J., Ho¨gler, W., McKeown, C., Sistermans, E.A., Dalton, A., MeijersHeijboer, H., Pals, G., 2009. PPIB mutations cause severe osteogenesis imperfecta. Am. J. Hum. Genet. 85, 521e527. Wenstrup, R.J., Cohn, D.H., Cohen, T., Byers, P.H., 1988. Arginine for glycine substitution in the triple-helical domain of the products of one alpha 2(I) collagen allele (COL1A2) produces the osteogenesis imperfecta type IV phenotype. J. Biol. Chem. 263, 7734e7740. Witecka, J., Augusciak-Duma, A.M., Kruczek, A., Szyd1o, A., Lesiak, M., Krzak, M., Pietrzyk, J.J., Ma¨nnikko¨, M., Sieron, A.L., 2008. Two novel COL1A1 mutations in patients with osteogenesis imperfecta (OI) affect the stability of the collagen type I triple-helix. J. Appl. Genet. 49, 283e295.