Association of JAG1 with Bone Mineral Density and Osteoporotic Fractures: A Genome-wide Association Study and Follow-up Replication Studies
2010; Elsevier BV; Volume: 86; Issue: 2 Linguagem: Inglês
10.1016/j.ajhg.2009.12.014
ISSN1537-6605
AutoresAnnie W.C. Kung, Su-Mei Xiao, Stacey S. Cherny, Gloria Hoi‐Yee Li, Yi Gao, Gloria Hoi Wan Tso, Kam Shing Lau, K.D.K. Luk, Jianmin Liu, Bin Cui, Min-Jia Zhang, Zhen-lin Zhang, Jin-Wei He, Hua Yue, Wia-bo Xia, Lian-mei Luo, Shu-li He, Douglas P. Kiel, David Karasik, Yi-Hsiang Hsu, L. Adrienne Cupples, Serkalem Demissie, Unnur Styrkársdóttir, Bjarni V. Halldórsson, Gunnar Sigurðsson, Unnur Thorsteinsdottir, Kári Stéfansson, J. Brent Richards, Guangju Zhai, Nicole Soranzo, Ana M. Valdes, Tim D. Spector, Pak C. Sham,
Tópico(s)Genetic Associations and Epidemiology
ResumoBone mineral density (BMD), a diagnostic parameter for osteoporosis and a clinical predictor of fracture, is a polygenic trait with high heritability. To identify genetic variants that influence BMD in different ethnic groups, we performed a genome-wide association study (GWAS) on 800 unrelated Southern Chinese women with extreme BMD and carried out follow-up replication studies in six independent study populations of European descent and Asian populations including 18,098 subjects. In the meta-analysis, rs2273061 of the Jagged1 (JAG1) gene was associated with high BMD (p = 5.27 × 10−8 for lumbar spine [LS] and p = 4.15 × 10−5 for femoral neck [FN], n = 18,898). This SNP was further found to be associated with the low risk of osteoporotic fracture (p = 0.009, OR = 0.7, 95% CI 0.57–0.93, n = 1881). Region-wide and haplotype analysis showed that the strongest association evidence was from the linkage disequilibrium block 5, which included rs2273061 of the JAG1 gene (p = 8.52 × 10−9 for LS and 3.47 × 10−5 at FN). To assess the function of identified variants, an electrophoretic mobility shift assay demonstrated the binding of c-Myc to the "G" but not "A" allele of rs2273061. A mRNA expression study in both human bone-derived cells and peripheral blood mononuclear cells confirmed association of the high BMD-related allele G of rs2273061 with higher JAG1 expression. Our results identify the JAG1 gene as a candidate for BMD regulation in different ethnic groups, and it is a potential key factor for fracture pathogenesis. Bone mineral density (BMD), a diagnostic parameter for osteoporosis and a clinical predictor of fracture, is a polygenic trait with high heritability. To identify genetic variants that influence BMD in different ethnic groups, we performed a genome-wide association study (GWAS) on 800 unrelated Southern Chinese women with extreme BMD and carried out follow-up replication studies in six independent study populations of European descent and Asian populations including 18,098 subjects. In the meta-analysis, rs2273061 of the Jagged1 (JAG1) gene was associated with high BMD (p = 5.27 × 10−8 for lumbar spine [LS] and p = 4.15 × 10−5 for femoral neck [FN], n = 18,898). This SNP was further found to be associated with the low risk of osteoporotic fracture (p = 0.009, OR = 0.7, 95% CI 0.57–0.93, n = 1881). Region-wide and haplotype analysis showed that the strongest association evidence was from the linkage disequilibrium block 5, which included rs2273061 of the JAG1 gene (p = 8.52 × 10−9 for LS and 3.47 × 10−5 at FN). To assess the function of identified variants, an electrophoretic mobility shift assay demonstrated the binding of c-Myc to the "G" but not "A" allele of rs2273061. A mRNA expression study in both human bone-derived cells and peripheral blood mononuclear cells confirmed association of the high BMD-related allele G of rs2273061 with higher JAG1 expression. Our results identify the JAG1 gene as a candidate for BMD regulation in different ethnic groups, and it is a potential key factor for fracture pathogenesis. Osteoporosis is a major public health problem, a disease characterized by low bone mass, poor bone quality, and an increased predisposition to fracture. BMD, an important clinical predictor of fracture, is a complex trait of strong genetic determination and is associated with a heritability estimate of 0.6 to 0.8.1Dequeker J. Nijs J. Verstraeten A. Geusens P. Gevers G. Genetic determinants of bone mineral content at the spine and radius: a twin study.Bone. 1987; 8: 207-209Abstract Full Text PDF PubMed Scopus (274) Google Scholar, 2Arden N.K. Baker J. Hogg C. Baan K. Spector T.D. The heritability of bone mineral density, ultrasound of the calcaneus and hip axis length: A study of postmenopausal twins.J. Bone Miner. Res. 1996; 11: 530-534Crossref PubMed Scopus (336) Google Scholar, 3Ng M.Y. Sham P.C. Paterson A.D. Chan V. Kung A.W. Effect of environmental factors and gender on the heritability of bone mineral density and bone size.Ann. Hum. Genet. 2006; 70: 428-438Crossref PubMed Scopus (70) Google Scholar, 4Ioannidis J.P. Ng M.Y. Sham P.C. Zintzaras E. Lewis C.M. Deng H.W. Econs M.J. Karasik D. Devoto M. Kammerer C.M. et al.Meta-analysis of genome-wide scans provides evidence for sex- and site-specific regulation of bone mass.J. Bone Miner. Res. 2007; 22: 173-183Crossref PubMed Scopus (139) Google Scholar Some large cohort studies and meta-analyses have started to clearly delineate the association of certain polymorphisms.5Uitterlinden A.G. Ralston S.H. Brandi M.L. Carey A.H. Grinberg D. Langdahl B.L. Lips P. Lorenc R. Obermayer-Pietsch B. Reeve J. et al.APOSS InvestigatorsEPOS InvestigatorsEPOLOS InvestigatorsFAMOS InvestigatorsLASA InvestigatorsRotterdam Study InvestigatorsGENOMOS StudyThe association between common vitamin D receptor gene variations and osteoporosis: A participant-level meta-analysis.Ann. Intern. Med. 2006; 145: 255-264Crossref PubMed Scopus (228) Google Scholar, 6Ralston S.H. Uitterlinden A.G. Brandi M.L. Balcells S. Langdahl B.L. Lips P. Lorenc R. Obermayer-Pietsch B. Scollen S. Bustamante M. et al.GENOMOS InvestigatorsLarge-scale evidence for the effect of the COLIA1 Sp1 polymorphism on osteoporosis outcomes: The GENOMOS study.PLoS Med. 2006; 3: e90Crossref PubMed Scopus (169) Google Scholar, 7Ioannidis J.P. Ralston S.H. Bennett S.T. Brandi M.L. Grinberg D. Karassa F.B. Langdahl B. van Meurs J.B. Mosekilde L. Scollen S. et al.GENOMOS StudyDifferential genetic effects of ESR1 gene polymorphisms on osteoporosis outcomes.JAMA. 2004; 292: 2105-2114Crossref PubMed Scopus (270) Google Scholar, 8Tran B.N. Nguyen N.D. Eisman J.A. Nguyen T.V. Association between LRP5 polymorphism and bone mineral density: A Bayesian meta-analysis.BMC Med. Genet. 2008; 9: 55Crossref PubMed Scopus (42) Google Scholar Several genome-wide association studies (GWAS) have recently confirmed some associated loci, although they demonstrate only a small effect size on BMD variation in the general population.9Kiel D.P. Demissie S. Dupuis J. Lunetta K.L. Murabito J.M. Karasik D. Genome-wide association with bone mass and geometry in the Framingham Heart Study.BMC Med. Genet. 2007; 8: S14Crossref PubMed Scopus (207) Google Scholar, 10Richards J.B. Rivadeneira F. Inouye M. Pastinen T.M. Soranzo N. Wilson S.G. Andrew T. Falchi M. Gwilliam R. Ahmadi K.R. et al.Bone mineral density, osteoporosis, and osteoporotic fractures: a genome-wide association study.Lancet. 2008; 371: 1505-1512Abstract Full Text Full Text PDF PubMed Scopus (548) Google Scholar, 11Styrkarsdottir U. Halldorsson B.V. Gretarsdottir S. Gudbjartsson D.F. Walters G.B. Ingvarsson T. Jonsdottir T. Saemundsdottir J. Center J.R. Nguyen T.V. et al.Multiple genetic loci for bone mineral density and fractures.N. Engl. J. Med. 2008; 358: 2355-2365Crossref PubMed Scopus (512) Google Scholar, 12Styrkarsdottir U. Halldorsson B.V. Gretarsdottir S. Gudbjartsson D.F. Walters G.B. Ingvarsson T. Jonsdottir T. Saemundsdottir J. Snorradóttir S. Center J.R. et al.New sequence variants associated with bone mineral density.Nat. Genet. 2009; 41: 15-17Crossref PubMed Scopus (278) Google Scholar, 13Xiong D.H. Liu X.G. Guo Y.F. Tan L.J. Wang L. Sha B.Y. Tang Z.H. Pan F. Yang T.L. Chen X.D. et al.Genome-wide association and follow-up replication studies identified ADAMTS18 and TGFBR3 as bone mass candidate genes in different ethnic groups.Am. J. Hum. Genet. 2009; 84: 388-398Abstract Full Text Full Text PDF PubMed Scopus (163) Google Scholar It has also been suggested that genes predisposing to the risk of osteoporosis vary between different ethnic groups.14Dvornyk V. Liu X.H. Shen H. Lei S.F. Zhao L.J. Huang Q.R. Qin Y.J. Jiang D.K. Long J.R. Zhang Y.Y. et al.Differentiation of Caucasians and Chinese at bone mass candidate genes: Implication for ethnic difference of bone mass.Ann. Hum. Genet. 2003; 67: 216-227Crossref PubMed Scopus (66) Google Scholar, 15Lambrinoudaki I. Kung A.W. Absence of high-risk "s" allele associated with osteoporosis at the intronic SP1 binding-site of collagen Ialpha1 gene in Southern Chinese.J. Endocrinol. Invest. 2001; 24: 499-502Crossref PubMed Scopus (24) Google Scholar Because BMD is a polygenic trait, it is probable that many additional associated variants or ethnic-specific loci remain to be identified. The identification of additional genes that determine BMD variation would therefore provide insight into the complex genetic architecture and pathogenetic mechanisms of osteoporosis and offer strategies for therapeutic development. In this GWAS, a multistep genome-wide association strategy (Figure S1 available online) was adopted to identify additional and/or ethnic-specific genes and their variants underlying BMD variation. The association of genome-wide data was first studied in 800 unrelated Hong Kong Southern Chinese (HKSC) females with extreme BMD (discovery sample). The most promising single-nucleotide polymorphisms (SNPs) were then tested for replication in another 720 HKSC subjects with extreme BMD (first-step replication). An association between replicated SNPs and BMD variation was confirmed in five other independent cohorts including 17,378 subjects of European descent or Asian descent (second-step replication). Finally, a region-wide association study was conducted to detect the causal variant of identified gene in the pooled extreme sample of discovery cohort and first-step replication cohort (n = 1520), and the potential biological function of the gene variant was further validated. The characteristics of population are shown in Table 1. In the discovery stage, 800 unrelated women with extreme high or low BMD were selected from a HKSC cohort with extreme BMD (n = 1520). These subjects were selected from a database (>7000 Southern Han Chinese volunteers) of the Osteoporosis Centre of the University of Hong Kong. The low-BMD subjects are defined as an individual having BMD Z-score ≤−1.28 at either lumbar spine (LS) or femoral neck (FN) (the lowest 10% of the total cohort), while high-BMD subjects are individuals with BMD Z-score ≥+1.0 at either site. Subjects who reported that they had diseases or environmental factors that may affect BMD and bone metabolism were excluded. The recruitment procedure and exclusion criteria have been detailed elsewhere.16Kung A.W. Lai B.M. Ng M.Y. Chan V. Sham P.C. T-1213C polymorphism of estrogen receptor beta is associated with low bone mineral density and osteoporotic fractures.Bone. 2006; 39: 1097-1106Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar Subjects were excluded for having a history of chronic medical illness, premature menopause below age 40 years, malabsorption, major gastrointestinal operation, metabolic bone diseases, endocrine disorders including hyper- and hypothyroidism, or medications that may affect bone and calcium metabolism, as well as for taking hormonal replacement therapy, antiosteoporosis medications, and active vitamin D3 metabolites. BMD (g/cm2) at LS and FN was measured by dual-energy X-ray absorptiometry (DXA; Hologic QDR 4500 plus, Hologic Waltham, MA, USA) with standard protocol. The age-corrected and standardized BMD (mean 0, SD 1), termed BMD Z-scores, was generated for each gender. The in vivo precision of the machine was 1.2% and 1.5% for LS and FN BMD, respectively.17Kung A.W. Yeung S.S. Lau K.S. Vitamin D receptor gene polymorphisms and peak bone mass in southern Chinese women.Bone. 1998; 22: 389-393Abstract Full Text Full Text PDF PubMed Scopus (49) Google ScholarTable 1Description of Studied PopulationsDiscovery CohortReplication CohortsFirst StepSecond StepHong Kong Southern Chinese Discovery CohortHong Kong Southern Chinese First-Step Replication CohortHong Kong Osteoporosis Study Prospective CohortNorthern Chinese PopulationGroupHigh BMDLow BMDHigh BMDLow BMDBMDNonfractureFractureBMDNumber376424264 (168)456 (124)188116372441584Age (years)46.6 ± 14.851.1 ± 15.948.3 ± 16.349.0 ± 16.062.3 ± 8.960.7 ± 8.267.4 ± 9.956.2 ± 11.3Height (m)1.58 ± 0.061.53 ± 0.071.65 ± 0.091.57 ± 0.081.52 ± 0.061.53 ± 0.061.50 ± 0.061.60 ± 0.06Weight (kg)61.1 ± 9.949.1 ± 6.767.1 ± 11.452.1 ± 9.555.4 ± 9.555.7 ± 9.554.1 ± 9.359.5 ± 9.4BMD Z-ScoreLumbar Spine1.05 ± 0.78−1.59 ± 0.531.36 ± 0.89−1.34 ± 0.72−0.43 ± 1.18−0.39 ± 1.18−0.72 ± 1.17−0.05 ± 1.24Femoral Neck1.15 ± 0.79−1.36 ± 0.601.01 ± 0.84−1.16 ± 0.69−0.35 ± 0.97−0.29 ± 0.97−0.78 ± 0.930.71 ± 0.92Data are expressed as mean ± standard deviation. The subject number is total number (the number of male subjects, if applicable, is given in parentheses) in each cohort. A total of 1520 subjects with extreme BMD phenotype were selected from a growing database of Hong Kong Southern Chinese (more than 7000 volunteers). The low BMD subjects are defined as an individual having BMD Z-score ≤ −1.28 at either lumbar spine (LS) or femoral neck (FN), which is equivalent to the lowest 10% of the total cohort, while high BMD subjects are individuals with BMD Z-score ≥ +1.0 at either of the two skeletal sites. First, a subset of 800 women was selected from the Hong Kong Southern Chinese (n = 1520) as the discovery cohort, and then the remainder of individuals (n = 720, 59.4% women) of the extreme population were used for the first-step replication. Open table in a new tab Data are expressed as mean ± standard deviation. The subject number is total number (the number of male subjects, if applicable, is given in parentheses) in each cohort. A total of 1520 subjects with extreme BMD phenotype were selected from a growing database of Hong Kong Southern Chinese (more than 7000 volunteers). The low BMD subjects are defined as an individual having BMD Z-score ≤ −1.28 at either lumbar spine (LS) or femoral neck (FN), which is equivalent to the lowest 10% of the total cohort, while high BMD subjects are individuals with BMD Z-score ≥ +1.0 at either of the two skeletal sites. First, a subset of 800 women was selected from the Hong Kong Southern Chinese (n = 1520) as the discovery cohort, and then the remainder of individuals (n = 720, 59.4% women) of the extreme population were used for the first-step replication. The discovery sample was genotyped via the Infinium assay (Illumina, San Diego, CA) with Human610-quad chip, including 564,214 SNPs. PLINK (version 1.04) was used for the data management and quality control statistics.18Purcell S. Neale B. Todd-Brown K. Thomas L. Ferreira M.A. Bender D. Maller J. Sklar P. de Bakker P.I. Daly M.J. Sham P.C. PLINK: A tool set for whole-genome association and population-based linkage analyses.Am. J. Hum. Genet. 2007; 81: 559-575Abstract Full Text Full Text PDF PubMed Scopus (19746) Google Scholar 785 individuals and 488,853 SNPs were retained for analysis after the exclusion based on strict quality-control criteria. Subjects were excluded according to the following criteria: (1) genotyping call rate less than 95% (n = 5); (2) autosomal heterozygosity less than 27% or more than 31% (the same five subjects with low genotyping call rate); (3) being related or identical to other individuals in the sample (n = 7); and (4) discordance of observed gender and estimated sex (n = 3). SNPs were excluded if (1) genotyping call rate was 95% or less (1158 SNPs); (2) Hardy-Weinberg equilibrium (HWE) p value was less than 1.0 × 10−4 (904 SNPs); (3) minor allele frequency (MAF) was less than 0.01 (73,589 SNPs). Finally, the average genotyping call rate of those retained SNPs was 99.91%. In the genome-wide association (GWA) scan, the association tests of SNPs with standardized BMD were implemented via PLINK (version 1.04). For each SNP, the asymptotic p value for the relationship between the number of minor alleles and BMD was derived from a two-sided t-statistic assuming the minor allele having an additive effect. Each SNP was tested for its association with BMD at both LS and FN. We used the genomic control19Devlin B. Roeder K. Genomic control for association studies.Biometrics. 1999; 55: 997-1004Crossref PubMed Scopus (2297) Google Scholar and EIGENSTRAT20Price A.L. Patterson N.J. Plenge R.M. Weinblatt M.E. Shadick N.A. Reich D. Principal components analysis corrects for stratification in genome-wide association studies.Nat. Genet. 2006; 38: 904-909Crossref PubMed Scopus (6863) Google Scholar methods for detecting possible bias resulting from population substructure and stratification. Quantile-quantile plots were constructed via R-2.5.2 for Windows. The entire genome association data and the LD block structure of genomic region were plotted via Haploview v4.1.21Barrett J.C. Fry B. Maller J. Daly M.J. Haploview: Analysis and visualization of LD and haplotype maps.Bioinformatics. 2005; 21: 263-265Crossref PubMed Scopus (12049) Google Scholar A two-step replication method was used in this study to validate the top SNPs identified from the GWA scan. SNPs from the GWA scan that had a false-positive report probability of less than 0.5 (corresponding to a p value threshold of 4.75 × 10−5 or less) as suggested by Wacholder et al. (2004)22Wacholder S. Chanock S. Garcia-Closas M. El Ghormli L. Rothman N. Assessing the probability that a positive report is false: an approach for molecular epidemiology studies.J. Natl. Cancer Inst. 2004; 96: 434-442Crossref PubMed Scopus (1416) Google Scholar and the fact that one in 5000 SNPs is hypothesized as a true positive (according to results from previous GWAS10Richards J.B. Rivadeneira F. Inouye M. Pastinen T.M. Soranzo N. Wilson S.G. Andrew T. Falchi M. Gwilliam R. Ahmadi K.R. et al.Bone mineral density, osteoporosis, and osteoporotic fractures: a genome-wide association study.Lancet. 2008; 371: 1505-1512Abstract Full Text Full Text PDF PubMed Scopus (548) Google Scholar, 23Samani N.J. Erdmann J. Hall A.S. Hengstenberg C. Mangino M. Mayer B. Dixon R.J. Meitinger T. Braund P. Wichmann H.E. et al.WTCCC and the Cardiogenics ConsortiumGenomewide association analysis of coronary artery disease.N. Engl. J. Med. 2007; 357: 443-453Crossref PubMed Scopus (1655) Google Scholar) were used for the first-step replication. Furthermore, SNPs with p value of less than 0.05 in the first-step replication and also those reaching genome-wide significance in the pooled analysis conducted in the 1520 HKSC subjects with extreme BMD were further validated in five independent cohorts, including the Hong Kong Osteoporosis Study prospective cohort (n = 1881 women), a Northern Chinese population (n = 1584 women), and three in silico replication cohorts: the Framingham Osteoporosis Study (FOS, n = 3569), the deCODE Genetics Study (dCG, n = 7610, ∼85.6% women), and the TwinsUK Study (TUK, n = 2734 women). In this GWA, the p value threshold for GWA significance was 1.02 × 10−7 or less (Bonferroni correction, 0.05 divided by 488,853 SNPs). The first-step replication sample included the remainder subjects with extreme BMD (428 females and 292 males) from the same HKSC extreme cohort, who we did not included in the discovery phase. SNPs were genotyped with the high-throughput Sequenom genotyping platform (Sequenom, San Diego, CA). Genotyping was performed blind to the phenotype status of the samples and was repeated in 5% of the samples for verification and quality control. The association analyses were the same as those used for the discovery cohort. Then the joint analysis were conducted in the pooled subjects with extreme BMD including the HKSC discovery cohort and first-step replication cohort (n = 1520). The three in silico replication cohorts have been well described previously. The FOS is a community-based, longitudinal, prospective cohort comprising three generations of individuals in multigenerational pedigrees and additional unrelated individuals.9Kiel D.P. Demissie S. Dupuis J. Lunetta K.L. Murabito J.M. Karasik D. Genome-wide association with bone mass and geometry in the Framingham Heart Study.BMC Med. Genet. 2007; 8: S14Crossref PubMed Scopus (207) Google Scholar The dCG is a population-based sample for genetics study of complex diseases.11Styrkarsdottir U. Halldorsson B.V. Gretarsdottir S. Gudbjartsson D.F. Walters G.B. Ingvarsson T. Jonsdottir T. Saemundsdottir J. Center J.R. Nguyen T.V. et al.Multiple genetic loci for bone mineral density and fractures.N. Engl. J. Med. 2008; 358: 2355-2365Crossref PubMed Scopus (512) Google Scholar, 12Styrkarsdottir U. Halldorsson B.V. Gretarsdottir S. Gudbjartsson D.F. Walters G.B. Ingvarsson T. Jonsdottir T. Saemundsdottir J. Snorradóttir S. Center J.R. et al.New sequence variants associated with bone mineral density.Nat. Genet. 2009; 41: 15-17Crossref PubMed Scopus (278) Google Scholar The TUK is a population-based sample of twins from the UK studying the hereditary basis of a variety of age-related traits and diseases.10Richards J.B. Rivadeneira F. Inouye M. Pastinen T.M. Soranzo N. Wilson S.G. Andrew T. Falchi M. Gwilliam R. Ahmadi K.R. et al.Bone mineral density, osteoporosis, and osteoporotic fractures: a genome-wide association study.Lancet. 2008; 371: 1505-1512Abstract Full Text Full Text PDF PubMed Scopus (548) Google Scholar BMD was measured in the three cohorts at LS and FN via DXA with standard protocols (Lunar DPX-L for FOS, Hologic QDR4500A for dCG, and Hologic QDR2000W for TUK). The three cohorts were genotyped with the Illumina Infinium HumanHap300 or HumanHap550 or HumanCNV370 Beadchip (TUK and dCG) or the Affymetrix 500K mapping array plus 50K supplemental array (FOS). For the FOS cohort, sex- and cohort (Original and Offspring)-specific standardized residuals (mean 0, SD 1) of phenotypes were calculated. GWAS was performed with linear mixed effects regression models, with fixed SNP genotype effects, and random individual effects that correlate within pedigree according to kinship relationship.24Therneau T. Kinship: Mixed Effects Cox Models, Sparse Matrices, and Modelling Data from Large Pedigrees. R Package Version 1.1.0.Nineteenth Edition. R Foundation for Statistical Computing, Vienna, Austria2008Google Scholar To account for the effect of population substructure, age- and weight-corrected BMD were also adjusted for the first four principal components from the EIGENSTRAT analysis. For the dCG and TUK cohorts, age- and weight-corrected BMD was computed for each sex separately to have a mean 0 and standard deviation 1. Single SNP association tests were performed with an additive genetic effect model that estimated the effect of one copy increment of the reference allele for all three cohorts. The p values were also corrected for genomic control in the dCG cohort. Hong Kong Osteoporosis Study prospective cohort: This population is also a part of the HKSC database with both BMD and fracture data. A total of 1881 unrelated postmenopausal women were prospectively followed for incident fractures.25Kung A.W. Luk K.D. Chu L.W. Tang G.W. Quantitative ultrasound and symptomatic vertebral fracture risk in Chinese women.Osteoporos. Int. 1999; 10: 456-461Crossref PubMed Scopus (47) Google Scholar, 26Kung A.W. Lee K.K. Ho A.Y. Tang G. Luk K.D. Ten-year risk of osteoporotic fractures in postmenopausal Chinese women according to clinical risk factors and BMD T-scores: A prospective study.J. Bone Miner. Res. 2007; 22: 1080-1087Crossref PubMed Scopus (72) Google Scholar The duration of follow-up corresponded to the time from baseline to the occurrence of fracture, death, or the last follow-up visit data. A low-traumatic fracture was defined as a fracture sustained after a fall from standing height or less. All nonosteoporotic site fractures (nose, toe, head, jaw, skull, and hands) and traumatic fractures were excluded. Radiographic morphometric vertebral fractures were defined by a reduction in vertebral height of at least 3 standard deviations in anterior, mid, or posterior ratios compared with normative means.27Black D.M. Cummings S.R. Stone K. Hudes E. Palermo L. Steiger P. A new approach to defining normal vertebral dimensions.J. Bone Miner. Res. 1991; 6: 883-892Crossref PubMed Scopus (233) Google Scholar Fractures were verified by retrieval of the X-ray report and hospital discharge summaries from the computerized patient record system of the Hospital Authority of the Hong Kong Government. The Northern Chinese population: 1584 female Northern Chinese aged 20 years or above were recruited from three osteoporosis centers (two in Shanghai and one in Beijing) by posters. All the study subjects are residents of Shanghai or Beijing and are Chinese of Han origin. Patients with chronic diseases and conditions that might potentially affect bone mass and bone metabolism were excluded with the same exclusion criteria as the HKSC population.16Kung A.W. Lai B.M. Ng M.Y. Chan V. Sham P.C. T-1213C polymorphism of estrogen receptor beta is associated with low bone mineral density and osteoporotic fractures.Bone. 2006; 39: 1097-1106Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar, 28Liu J.M. Zhao H.Y. Ning G. Zhao Y.J. Zhang L.Z. Sun L.H. Xu M.Y. Chen J.L. Relationship between body composition and bone mineral density in healthy young and premenopausal Chinese women.Osteoporos. Int. 2004; 15: 238-242Crossref PubMed Scopus (52) Google Scholar, 29Zhao H.Y. Liu J.M. Ning G. Zhao Y.J. Chen Y. Sun L.H. Zhang L.Z. Xu M.Y. Chen J.L. Relationships between insulin-like growth factor-I (IGF-I) and OPG, RANKL, bone mineral density in healthy Chinese women.Osteoporos. Int. 2008; 19: 221-226Crossref PubMed Scopus (25) Google Scholar The DXA machine (GE-Lunar Corp. or Hologic Inc.) was used to measure BMD of the LS and FN via standard protocols. These two Chinese second-step replication cohorts were genotyped with the TaqMan system (Applied Biosystems, Foster City, CA). The statistical procedures of associations between SNP and BMD variation were the same as those used for the HKSC discovery cohort, with the addition of an "osteoporosis center" variable also included as a covariate in the association analysis of Northern Chinese population to eliminate the potential effect of subjects being recruited from different centers. The logistic regression model was used to examine the association of SNP with fracture risk by adjusting for covariates such as age and follow-up duration. Skeletal site-specific analyses were also conducted for spine and hip fractures. These statistical analyses were performed with SPSS v.16 (SPSS, Chicago, IL). In all studies, all subjects gave informed consent and the study protocols were reviewed and approved by local institutional review boards. The meta-analysis of SNP with BMD variation in (1) all of the Chinese populations used for the second-step replication (n = 3,465); (2) all of the second-step replication subjects of European descent (n = 13,913); and (3) all of the studied subjects from seven independent populations (n = 18,898) were performed via the inverse variance meta-analysis approach. Furthermore, to study the association of the identified gene with BMD in more detail and to detect the functional variant, a region-wide and tagSNP-based method followed by imputation and sequencing was conducted. Selected SNPs were genotyped with the high-throughput Sequenom genotyping platform (Sequenom, San Diego, CA) in all of the HKSC subjects with extreme BMD (n = 1520). SNPs with genotyping call rate < 90% and HWE p < 0.001 were excluded. Both single marker- and LD block-based haplotype association analyses with BMD were performed with the PLINK (version 1.04). The single SNP analysis was the same as that used for the HKSC discovery cohort. The conditional haplotype testing model was also used to assess the correlation structure and identify the most promising candidate causal variant(s).30Han S. Kim-Howard X. Deshmukh H. Kamatani Y. Viswanathan P. Guthridge J.M. Thomas K. Kaufman K.M. Ojwang J. Rojas-Villarraga A. et al.Evaluation of imputation-based association in and around the integrin-alpha-M (ITGAM) gene and replication of robust association between a non-synonymous functional variant within ITGAM and systemic lupus erythematosus (SLE).Hum. Mol. Genet. 2009; 18: 1171-1180Crossref PubMed Scopus (95) Google Scholar The imputation of genotypes for all untyped SNPs from HapMap in the studied genomic region was conducted with the IMPUTE program.31Marchini J. Howie B. Myers S. McVean G. Donnelly P. A new multipoint method for genome-wide association studies by imputation of genotypes.Nat. Genet. 2007; 39: 906-913Crossref PubMed Scopus (1970) Google Scholar We used the release 27 HapMap Chinese (CHB [Chinese Han Beijing] and CHD [Chinese in Denver]) data as the reference panel. The IGG program32Li M.X. Jiang L. Ho S.L. Song Y.Q. Sham P.C. IGG: A tool to integrate GeneChips for genetic studies.Bioinformatics. 2007; 23: 3105-3107Crossref PubMed Scopus (8) Google Scholar was used for the integration of HapMap data and the exportation of IMPUTE format. Sequencing was conducted in the coding region of identified gene and its ∼500 bp flanking genomic sequence in 25 high-BMD subjects who carried the protective allele of the identified SNP and 25 low-BMD subjects who carried the risk allele. In addition, to test the best-fitted genetic model for the identified SNP, the likelihood ratio test of the additive or dominant or recessive regression model (1 df) against the full 2 df model was performed. The interaction with sex, age ( 50 years), and menopause was tested with regression models. Finally, the potential biological function of the identified gene variant was studied. The FASTSNP program33Yuan H.Y. Chiou J.J. Tseng W.H. Liu C.H. Liu C.K. Lin Y.J. Wang H.H. Yao A. Chen Y.T. Hsu C.N. FASTSNP: An always up-to-date and extendable service for SNP function analysis and prioritization.Nucleic Acids Res. 2006; 34: W635-W641Crossref PubMed Scopus (449) Google Scholar was used for predicting the function of SNPs of interest. An electrophoretic mobility shift assay (EMSA) w
Referência(s)