Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Osteoporos Int. Author manuscript; available in PMC 2010 August 9.
Published in final edited form as:
PMCID: PMC2917903

Replication study of candidate genes/loci associated with osteoporosis based on genome-wide screening



Osteoporosis is a major public health problem characterized by low bone mineral density (BMD). This replication study confirmed 38 single-nucleotide polymorphisms (SNPs) out of 139 SNPs previously reported in three recent genome-wide association studies (GWASs) in an independent US white sample. Ten SNPs achieved combined p<3.6×10−4.


BMD is under strong genetic control. This study aims to verify the potential associations between BMD and candidate genes/loci reported by GWAS of FHS100K, Icelandic deCODE, and UK-NL.


Eight promising (at the genome-wide significant level after Bonferroni correction) and 131 available sub-promising (at the most stringent p value, p<5.5×10−5 in the three GWASs reports) SNPs were selected. By using genotypic information from Affymetrix 500 K SNP arrays, we tested their associations with BMD in 1,000 unrelated US whites. Fisher’s combined probability method was used to quantify the overall evidence of association. BMD was measured by dual energy X-ray absorptiometry.


Two promising SNPs, rs3762397 and rs3736228, were replicated in the current study with p<0.05. Besides, 36 sub-promising SNPs were replicated at the same significant level. Ten SNPs achieved significant combined p<3.6×10−4 (0.05/139 SNPs, corrected for multiple testing).


Osteoporosis susceptibility of 38 SNPs was replicated in 1,000 unrelated US whites. This study showed promise for replication of some initial genome-wide association signals.

Keywords: Bone mineral density, Genome-wide association, Osteoporosis, Replication


Osteoporosis is a common skeletal disease characterized by low bone mass and micro-architectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to osteoporotic fractures [1]. Osteoporosis is defined clinically through the measurement of bone mineral density (BMD, g/cm2), which is one of the most important predictors of primary osteoporotic fractures [2, 3]. Genetic factors play important roles in determining population variation of BMD [4]. More than 20 genome-wide linkage studies and hundreds of candidate gene association studies have revealed multiple genetic loci related to BMD. However, discrepant and conflicting results were reported across studies, necessitating replication studies in independent samples [5]. So far, specific genetic factors of osteoporosis are largely unknown. Identifying the genes/loci for osteoporosis is still challenging.

With the rapid advancement in high throughput single-nucleotide polymorphism (SNP) genotyping technique and determination of haplotype tagging SNPs attributed to the International HapMap Project (, genome-wide association study (GWAS) is now widely applied to dissect the genetic determination of common complex disorders, including osteoporosis. Considering conflicting results from previous association studies and probability of false positive association due to large number of tests involved in GWAS, replication studies in an independent sample are requested.

As a follow-up replication study, the present work was attempted to test in 1,000 unrelated US whites the associations between 139 candidate SNPs and BMD variation, which were originally disclosed in three recently published GWASs: FHS100K [6], Icelandic deCODE [7], and UK-NL [8].

Materials and methods

Study populations

Current GWAS replication population

The study of current 500 K GWAS was approved by the involved Institutional Review Board. Signed informed consent documents were obtained from all study participants. We studied a total of 1,000 unrelated subjects, including 501 women and 499 men. Our study subjects of US whites were identified from an established cohort containing ~6,000 subjects recruited from the Midwestern US. All subjects were normal healthy subjects defined by a comprehensive suite of exclusion criteria detailed previously [4]. Briefly, subjects with chronic diseases and conditions involving vital organs (heart, lung, liver, kidney, and brain) and severe endocrinological, metabolic, and nutritional diseases that might affect bone metabolism were excluded from this study. By following the above exclusion criteria, we expected to exclude potential confounders which may interfere with association test and increase the power of detecting modest genetic effect on BMD variation in our study population.

Previous three GWASs populations

The FHS100K GWAS in osteoporosis [6] consisted of 1,141 subjects, including 495 men and 646 women with BMD data.

The Icelandic deCODE GWAS [7] consisted of 5,861 Icelandic subjects, including 776 men and 5,085 women.

The UK-NL GWAS [8] consisted of 8,557 participants from four population-based cohorts. Among them, 2,094 women came from the TwinsUK discovery cohort, which was a population-based sample of Britons previously shown to be representative of singleton populations, and the general UK population [9]. The other three cohorts included Rotterdam cohort (4,081 subjects, 784 men, and 3,297 women), TwinsUK replication cohort (1,692 women), and Chingford cohort (690 women). All the participants were of white European ancestry. The general relevant characteristics of these four studies are summarized in Table 1.

Table 1
Basic characteristics of the replication sample and initial GWAS samples

BMD measurement

In current 500K GWAS, we measured BMD (g/cm2) at the following skeletal sites with daily calibrated Hologic 4500 dual energy radiograph absorptiometry (DXA; Hologic, Bedford, MA, USA): BMD at the lumbar spine L1–L4 (SPNBMD), combined BMD of femoral neck, trochanter, and intertrochanter areas (HIPBMD), and BMD at the femoral neck (FNBMD). The coefficient of variation (CV) values of the DXA measurements were 1.98%, 1.87%, and 1.87% for SPNBMD, HIPBMD and FNBMD, respectively.

In the FHS100K GWAS, BMD was measured by Lunar DPX-L (Lunar Corp., Madison, WI, USA). The CVs for SPNBMD, trochanter BMD, and FNBMD were 0.9%, 2.5%, and 1.7%, respectively [10]. In the Icelandic deCODE GWAS, BMD was measured at the lumbar spine and hip by DXA [7]. In the UK-NL GWAS, BMD at the lumbar spine and femoral neck was measured by DXA [8].


Current 500K GWAS

Genomic DNA was extracted from peripheral blood sample using a commercial isolation kit (Gentra systems, Minneapolis, MN, USA). Genotyping with the Affymetrix Mapping 250K Nsp and Affymetrix Mapping 250K Sty arrays was performed by the Vanderbilt Microarray Shared Resources ( using the standard protocol of the Affymetrix. Genotyping calls were determined from the fluorescent intensities using the dynamic model (DM) algorithm with a 0.33 p value setting [11] as well as the B-RLMM algorithm, an extension of the RLMM [12] developed for the Mapping 500K product. DM calls were used for quality control while the B-RLMM calls were used for all subsequent data analyses. B-RLMM clustering was performed with 94 samples per cluster.

The final average B-RLMM call rate across the entire sample was 99.14%. However, out of the initial full set of 500,568 SNPs, we discarded 32,961 SNPs with call rates <95% in the sample, an addition of 36,965 SNPs with allele frequencies deviating from Hardy–Weinberg equilibrium (HWE; p<0.001), and 51,323 SNPs with minor allele frequencies (MAF)<1%. Therefore, the final SNP set maintained in the subsequent analyses contained 379,319 SNPs, yielding an average SNP spacing of ~7.9 kb throughout the human genome.


The Affymetrix 100K SNP GeneChip was used for genotyping. The study tested 70,987 SNPs with genotypic call rates ≥80%, HWE p≥0.001, MAF≥10% for association with BMD [6].

The Icelandic deCODE GWAS

The Infinium HumanHap300 or the HumanCNV370 SNP chip from the Illumina (San Diego, CA) was used for genotyping. In total, 301,019 SNPs with genotypic call rates ≥98%, HWE p>10−7, and MAF>5% were used for association analyses with BMD [7].


The Infinium assays (Illumina, San Diego, CA), Hap300 Duo, Hap300, and Hap550, were used for genotyping TwinsUK discovery samples (2,094 women). Following the inclusion criteria of the genotypic call rates >90%, HWE p≥0.0001, MAF≥1%, 314,075 SNPs were used for association analyses with BMD. These 314,075 SNPs were then assayed in the Rotterdam samples with the HumanHap 550 v3.0 assays (Illumina, San Diego, CA), applying the same quality-control criteria. In the study, significant SNPs were further replicated in the Chingford cohort and/or TwinsUK replication cohort, which were genotyped with Taqman system (Applied Biosystems, Foster City, CA, USA) [8].

Selection of SNPs for current replication study

For replication analyses in our samples, a total of 139 SNPs were selected from three previous GWASs (FHS100K, Icelandic deCODE and UK-NL). Firstly, we selected eight so-called promising SNPs (three from the FHS100K GWAS, three from the Icelandic deCODE GWAS, and two from the UK-NL GWAS), which all reached a genome-wide significant level after Bonferroni correction in the respective studies, i.e., 7.04×10−7 (0.05/70,987 SNPs) in the FHS100K GWAS, 1.66×10−7 (0.05/301,019 SNPs) in the Icelandic deCODE GWAS, and 1.59×10−7 (0.05/314,075 SNPs) in the UK-NL GWAS. Considering that the Bonferroni correction is overly strict, secondly, we selected an addition of 131 available SNPs (15 from FHS100K, 82 from Icelandic deCODE, and 34 from UK-NL) with p value less than 5.5×10−5 (i.e., a most stringent cutoff p value for data report in the three previous GWASs) i.e., so-called sub-promising SNPs.

Statistical analyses

Statistical analysis in current replication sample

Prior to the association analyses, we adjusted the raw phenotypic values with the same covariates applied in the GWAS of FHS100K, Icelandic deCODE, and UK-NL, respectively. For those promising and sub-promising SNPs, which were missing in our 500K Affymetrix assays, we imputed the genotypes with the IMPUTE program ( Assuming the same genetic model in this replication study and the previous GWASs, we then used the SNPTEST program ( to test the association in our sample. To quantify the overall evidence of association between SNPs and BMD, Fisher’s combined probability method [13] was used to calculate a combined p value in both previous GWASs and the current replication study, stratified by sex and skeletal site.

Statistical analysis in three previous GWASs

In FHS100K GWAS [6], BMD values were adjusted by age, age2, height, BMI, smoking, physical activity, and estrogen therapy. Multivariate regression analysis was performed in each sex (men and women) and cohort (original and offspring). Association analyses were performed by using both family-based association tests and additive generalized estimating equation models.

In Icelandic deCODE GWAS [7], BMD values were adjusted for age, sex, and weight. For each SNP, a linear regression analysis, with the genotype as an additive covariate and standardized BMD as the response variable, was fitted to test for association. Each SNP was tested separately for its association with HIPBMD and SPNBMD.

In UK-NL GWAS [8], BMD values were adjusted for age. Association analyses were performed using the PLINK software package (version 1.01) (, with family structure in the sample taken into account [14]. Some of the subjects are monozygotic twins, and for these sib-pairs, genotypic information for only one individual per pair was included in the analyses, since monozygotic twins share identical genetic information. Where a single dizygotic twin had missing data, or was excluded, the remaining sibling was treated as a singleton in the statistical analysis.

Quality control of the replication sample

The Structure 2.2 ( was used to detect potential population stratification in our GWAS sample, which uses a Markov chain Monte Carlo algorithm to cluster individuals into different cryptic subpopulations on the basis of multilocus genotype data [15]. Specifically, 200 randomly selected unlinked SNPs were used for the clustering. For the reliability of our results, we performed independent analyses under three assumed numbers for population strata (k=2, 3, and 4). Existence of substructure is suggested if the subjects were clustered into two or more groups. We further tested our sample for population stratification using the genomic control method [16]. Based on genome-wide SNP information, we estimated the inflation factor (λ), a measure for population stratification. Ideally, for a homogeneous population with no stratification, the value of λ should be equal or near to 1.0.


Characteristics of the current replication sample

Basic characteristics of our current replication sample are presented in Table 1. The STRUCTURE program revealed that all subjects in this US whites sample were clustered together and could not be assigned into any subgroups, indicating that there was no significant population stratification within the sample. Further, the genomic control method estimated the λ value to be 1.007, confirming the results achieved through the Structure 2.2 software, indicating that there was essentially no population stratification in this sample. The relative homogeneity of this study sample eliminates potential spurious associations due to population stratification.

Replicated SNPs

Thirty-eight SNPs among the 139 selected SNPs achieved p values less than 0.05 in this replication study. Sixteen out of the 38 SNPs attained p value less than 0.01. Specifically, (1) for the eight selected promising SNPs, two SNPs (rs3762397 and rs3736228) were replicated (see Table 2). SNP rs3762397 is located in the intron of the nuclear receptor subfamily 5, group A, member 2 (NR5A2) gene, also known as liver receptor homolog (LRH-1) gene; SNP rs3736228 is located in the coding exon 18 of the lipoprotein-receptor-related protein (LRP5) gene (see Table 2). (2) For the remaining 131 selected sub-promising SNPs, 36 SNPs were replicated (see Table 3). Among them, 17 SNPs were associated with HIPBMD, six SNPs with SPNBMD, and 13 SNPs with both HIPBMD and SPNBMD.

Table 2
Replication outcome of the promising SNPs selected from initial GWASs
Table 3
Replication outcome of the sub-promising SNPs (p<5.5×10−5) selected from initial GWASs

Fisher’s combined probability test

Among the 38 replicated SNPs, 10 SNPs achieved a significant combined p value, i.e., less than 3.6×10−4 (0.05/139 SNPs, with multiple-testing of SNPs taken into account). The combined p values of the 10 SNPs are presented in Table 4.

Table 4
Replicated SNPs with significant p values of Fisher’s combined probability test

Notably, two SNPs located within the LRP5 gene, rs3736228 and rs2306862, achieved combined p values of 5.3×10−12 and 6.0×10−6 for SPNBMD, respectively, in a combined test for UK-NL sample and this replication sample. These two SNPs were also found to be associated with SPNBMD in the entire population of Icelandic deCODE study with p values of 6.5×10−4 and 4.9×10−4, respectively. The other two SNPs, rs3020331 in the estrogen receptor 1 (ESR1) gene and rs4870044, ~45 kb upstream of the ESR1 gene and in the intron of the chromosome 6 open reading frame 97 (C6orf97) gene, were associated with SPNBMD, with combined p values of 1.2×10−6 and 4.9×10−8 in the whole populations of Icelandic deCODE GWAS and this replication sample. Besides the above four SNPs in the well-known osteoporosis candidate genes, the other six SNPs were also found to play important roles in the BMD variation at different skeletal site. Three SNPs, rs11898505 in SPTBN1 gene, rs4276378 in ADCY2 gene, and rs11239762 in BMS1L gene, were found to be associated with SPNBMD. Two SNPs, rs12437971 in ADAMTS17 gene and rs6696981 in an anonymous gene, were associated with HIPBMD. The SNP, rs1823926 in LOC51334 gene, was associated with FNBMD.


Following up three recently published GWASs on osteoporosis conducted in different cohorts (FHS100K, Icelandic deCODE, and UK-NL) [68], the present study explored the association of the top significant SNPs with BMD variation in an independent sample of 1,000 unrelated US whites. We selected a total of 139 SNPs for replication test, including eight promising SNPs and 131 sub-promising SNPs available from the three previous GWASs reports. Among them, two promising and 36 sub-promising SNPs were replicated when the significant threshold of p value was set at 0.05. The average replication rate was 25% for the promising SNPs, compared to 27% for the sub-promising SNPs. After combining outcomes from previous GWAS and the current replication study, 10 SNPs attained significant level even after stringent Bonferroni correction. In general, our data showed that significant and suggestive GWAS findings are likely to be replicated by independent study. Suggestive SNPs, though not attaining genome-wide significance threshold in initial GWAS sample, should also be valued for evaluation in independent samples. Therefore, besides the 20–50 top significant SNPs generally listed in GWAS reports, we suggest more comprehensive data be released for further exploration and replication in genetic research community.

One of the significant SNP replicated in this study, rs3762397, is located in a novel candidate gene NR5A2 at 1q32. This gene is expressed in all major steroidogenic tissues and tissues such as skeletal muscle, bone marrow [17, 18]. NR5A2 is the major NR5A subfamily member expressed in the preovulatory follicle and the corpus luteum. It may play a key role in the regulation of gonadal steroidogenic gene expression [17], accordingly affecting bone metabolism via hormonal regulation.

Another significant SNP replicated in the current study, rs3736228, is located in LRP5 gene, which encodes a transmembrane protein from the low density lipoprotein receptor family, the low-density LRP5. LRP5 is expressed in osteoblasts [19]. As a well-known Wnt coreceptor, the protein transduces Wnt signaling and affects bone mass [20, 21]. Relationship between LRP5 polymorphisms and BMD has been widely investigated [22, 23]. Some polymorphisms in the LRP5 gene (e.g., rs4988321, rs312009, rs2508836, rs729635, rs643892) were associated with reduced bone mass and/or increased fracture risk [24, 25], while some other LRP5 mutations (e.g. LRP5171V) were associated with high bone mass [26, 27]. The UK-NL GWAS [8] disclosed the essential effect of rs3736228 on decreased BMD. In a recent meta-analysis comprised of 16,705 individuals, a significant association between rs3736228 and decreased SPNBMD was revealed in both Asian and Caucasian population, although the effect of this LRP5 gene polymorphism on BMD variation was modest [28]. Our current replication study found consistent association of rs3736228 with SPNBMD in the whole population (p=0.028) and in female subgroup (p=0.019). These findings were strengthened by the following combined analyses, in which rs3736228 ranked as the top significant SNP, indicating a strong association with SPNBMD in the overall sample of 9,557 subjects (8,557 from UK-NL and 1,000 individuals from current study). And the SNP was significantly associated with SPNBMD in 2,595 women (2,094 from Twins UK discovery and 501 from our current replication study). The other SNP in LRP5 gene, rs2306862, was also significantly associated with SPNBMD in this 2,595 female subgroup. Interestingly, the two replicated SNPs in LRP5 gene, rs2306862 and rs3736228, were found interacted with physical activity on SPNBMD in a recent study [29].

Interestingly, the two replicated genes, LRP5 and NR5A2 (LRH-1), are involved in lipid and lipoprotein metabolism [3034]. LRP5 deficiency mice had increased plasma cholesterol levels [31]. LRH-1 directly regulates apolipoprotein M (APOM) transcription by binding to an LRH-1 response element located at the APOM promoter region [32]. In addition, studies reported associations between serum lipid and BMD [35, 36], suggesting a close relationship between lipids and bone metabolism.

GWAS with a larger sample size has greater power to detect genetic variants that confer modest disease risks without relying on prior knowledge of any specific genes/genomic regions [37, 38]. Population stratification is an important source of spurious association in genetic association studies [39, 40]. The study cohort, used for replication test, came from an apparently homogenous US midwest white population, living in Omaha, Nebraska and its surrounding areas. This population is largely dominant of Caucasians as the major ethnic group in this area for many generations. Consistently, no significant population stratification was detected, which strongly warranted the reliability of the replication results.

We noted that a majority of the promising SNPs previously found associated with BMD were not replicated in this study. Potential reasons might be: (1) study methods differ among GWAS studies. For example, phenotype data is not completely the same across the GWASs: SPNBMD was a combination of L2–L4 in FHS100K and Icelandic deCODE but a combination of L1–L4 in UK-NL and the present replication study. Different genotyping platforms (i.e., Affymetrix in current GWAS, Illumina in UK-NL) and different SNP genotype quality control criteria was used among studies. (2) The result probably also partially reflect the complexity of osteoporosis pathogenesis, which is determined not only by genetic factors but also by environmental factors and their interactions. Populations from the same ethnic origin but different geographic regions or with different cultures have different exposures to environmental factors. Factors such as smoking, alcohol drinking, nutritional status, and exercises may have significant influence on BMD determination in humans. These factors, however, are sometimes difficult to assess, quantify, and controlled accurately, and their influence on BMD variation could not be judiciously accounted for thus brings interfering noise for statistical analyses.

In summary, following up three previous GWASs on osteoporosis, this study tested association with BMD for 139 significant or suggestive SNPs in an independent US whites sample, with 38 SNPs replicated. This study showed promise for replication of initial genome-wide association signals in an independent study.


Investigators of this work were partially supported by grants from NIH (R01 AR050496-01, R21 AG027110, R01 AG026564, R21 AA015973, and P50 AR055081). Yin-Ping Zhang was supported by China Scholarship Council, Ministry of Education. The study also benefited from grants from National Science Foundation of China, Huo Ying Dong Education Foundation, Hunan Province, Xi’an Jiaotong University, and the Ministry of Education of China. We thank all the study subjects for volunteering to participate in the study.


Conflicts of interest None.

Contributor Information

Y.-P. Zhang, Key Laboratory of Environment and Genes Related to Diseases, Ministry of Education, College of Medicine, Xi’an Jiaotong University, Xi’an 710061, People’s Republic of China.

F.-Y. Deng, Departments of Orthopedic Surgery and Basic Medical Sciences, University of Missouri, Kansas City, MO 64108, USA.

Y. Chen, Key Laboratory of Biomedical Information Engineering, Ministry of Education and Institute of Molecular Genetics, School of Life Science and Technology, Xi’an Jiaotong University, No. 28, Xianning West Road, Xi’an, Shaanxi 710049, People’s Republic of China.

Y.-F. Pei, Key Laboratory of Biomedical Information Engineering, Ministry of Education and Institute of Molecular Genetics, School of Life Science and Technology, Xi’an Jiaotong University, No. 28, Xianning West Road, Xi’an, Shaanxi 710049, People’s Republic of China.

Y. Fang, Departments of Orthopedic Surgery and Basic Medical Sciences, University of Missouri, Kansas City, MO 64108, USA.

Y.-F. Guo, Departments of Orthopedic Surgery and Basic Medical Sciences, University of Missouri, Kansas City, MO 64108, USA. Key Laboratory of Biomedical Information Engineering, Ministry of Education and Institute of Molecular Genetics, School of Life Science and Technology, Xi’an Jiaotong University, No. 28, Xianning West Road, Xi’an, Shaanxi 710049, People’s Republic of China.

X. Guo, Key Laboratory of Environment and Genes Related to Diseases, Ministry of Education, College of Medicine, Xi’an Jiaotong University, Xi’an 710061, People’s Republic of China.

X.-G. Liu, Key Laboratory of Biomedical Information Engineering, Ministry of Education and Institute of Molecular Genetics, School of Life Science and Technology, Xi’an Jiaotong University, No. 28, Xianning West Road, Xi’an, Shaanxi 710049, People’s Republic of China.

Q. Zhou, Key Laboratory of Biomedical Information Engineering, Ministry of Education and Institute of Molecular Genetics, School of Life Science and Technology, Xi’an Jiaotong University, No. 28, Xianning West Road, Xi’an, Shaanxi 710049, People’s Republic of China.

Y.-J. Liu, Departments of Orthopedic Surgery and Basic Medical Sciences, University of Missouri, Kansas City, MO 64108, USA.

H.-W. Deng, Departments of Orthopedic Surgery and Basic Medical Sciences, University of Missouri, Kansas City, MO 64108, USA. Key Laboratory of Biomedical Information Engineering, Ministry of Education and Institute of Molecular Genetics, School of Life Science and Technology, Xi’an Jiaotong University, No. 28, Xianning West Road, Xi’an, Shaanxi 710049, People’s Republic of China. College of Life Sciences and Engineering, Beijing Jiao Tong University, Beijing 100044, People’s Republic of China.


1. Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med. 1993;94:646–650. [No authors listed] [PubMed]
2. Cummings SR, Melton LJ. Epidemiology and outcomes of osteoporotic fractures. Lancet. 2002;359:1761–1767. [PubMed]
3. Kanis JA, Borgstrom F, De Laet C, Johansson H, Johnell O, Jonsson B, Oden A, Zethraeus N, Pfleger B, Khaltaev N. Assessment of fracture risk. Osteoporos Int. 2005;16:581–589. [PubMed]
4. Deng HW, Chen WM, Conway T, Zhou Y, Davies KM, Stegman MR, Deng H, Recker RR. Determination of bone mineral density of the hip and spine in human pedigrees by genetic and lifestyle factors. Genet Epidemiol. 2000;19:160–177. [PubMed]
5. Liu YJ, Shen H, Xiao P, Xiong DH, Li LH, Recker RR, Deng HW. Molecular genetic studies of gene identification for osteoporosis: a 2004 update. J Bone Miner Res. 2006;21:1511–1535. [PMC free article] [PubMed]
6. Kiel DP, Demissie S, Dupuis J, Lunetta KL, Murabito JM, Karasik D. Genome-wide association with bone mass and geometry in the Framingham Heart Study. BMC Med Genet. 2007;8:S14. [PMC free article] [PubMed]
7. Styrkarsdottir U, Halldorsson BV, Gretarsdottir S, Gudbjartsson DF, Walters GB, Ingvarsson T, Jonsdottir T, Saemundsdottir J, Center JR, Nguyen TV, Bagger Y, Gulcher JR, Eisman JA, Christiansen C, Sigurdsson G, Kong A, Thorsteinsdottir U, Stefansson K. Multiple genetic loci for bone mineral density and fractures. N Engl J Med. 2008;358:2355–2365. [PubMed]
8. Richards JB, Rivadeneira F, Inouye M, Pastinen TM, Soranzo N, Wilson SG, Andrew T, Falchi M, Gwilliam R, Ahmadi KR, Valdes AM, Arp P, Whittaker P, Verlaan DJ, Jhamai M, Kumanduri V, Moorhouse M, van Meurs JB, Hofman A, Pols HA, Hart D, Zhai G, Kato BS, Mullin BH, Zhang F, Deloukas P, Uitterlinden AG, Spector TD. Bone mineral density, osteoporosis, and osteoporotic fractures: a genome-wide association study. Lancet. 2008;371:1505–1512. [PMC free article] [PubMed]
9. Arden NK, Baker J, Hogg C, Baan K, Spector TD. The heritability of bone mineral density, ultrasound of the calcaneus and hip axis length: a study of postmenopausal twins. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research. 1996;11:530–534. [PubMed]
10. Hannan MT, Felson DT, Dawson-Hughes B, Tucker KL, Cupples LA, Wilson PW, Kiel DP. Risk factors for longitudinal bone loss in elderly men and women: the Framingham Osteoporosis Study. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research. 2000;15:710–720. [PubMed]
11. Di X, Matsuzaki H, Webster TA, Hubbell E, Liu G, Dong S, Bartell D, Huang J, Chiles R, Yang G, Shen MM, Kulp D, Kennedy GC, Mei R, Jones KW, Cawley S. Dynamic model based algorithms for screening and genotyping over 100 K SNPs on oligonucleotide microarrays. Bioinformatics (Oxford, England) 2005;21:1958–1963. [PubMed]
12. Rabbee N, Speed TP. A genotype calling algorithm for affymetrix SNP arrays. Bioinformatics (Oxford, England) 2006;22:7–12. [PubMed]
13. Fisher RA. The resemblance between twins, a statistical examination of Lauterbach’s Measurements. Genetics. 1925;10:569–579. [PubMed]
14. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Maller J, Sklar P, de Bakker PI, Daly MJ, Sham PC. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81:559–575. [PubMed]
15. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155:945–959. [PubMed]
16. Devlin B, Roeder K. Genomic control for association studies. Biometrics. 1999;55:997–1004. [PubMed]
17. Boerboom D, Pilon N, Behdjani R, Silversides DW, Sirois J. Expression and regulation of transcripts encoding two members of the NR5A nuclear receptor subfamily of orphan nuclear receptors, steroidogenic factor-1 and NR5A2, in equine ovarian cells during the ovulatory process. Endocrinology. 2000;141:4647–4656. [PubMed]
18. Nishimura M, Naito S, Yokoi T. Tissue-specific mRNA expression profiles of human nuclear receptor subfamilies. Drug Metab Pharmacokinet. 2004;19:135–149. [PubMed]
19. Kato M, Patel MS, Levasseur R, Lobov I, Chang BH, Glass DA, 2nd, Hartmann C, Li L, Hwang TH, Brayton CF, Lang RA, Karsenty G, Chan L. Cbfa1-independent decrease in osteoblast proliferation, osteopenia, and persistent embryonic eye vascularization in mice deficient in Lrp5, a Wnt coreceptor. J Cell Biol. 2002;157:303–314. [PMC free article] [PubMed]
20. Krishnan V, Bryant HU, Macdougald OA. Regulation of bone mass by Wnt signaling. J Clin Invest. 2006;116:1202–1209. [PMC free article] [PubMed]
21. Gong Y, Slee RB, Fukai N, Rawadi G, Roman-Roman S, Reginato AM, Wang H, Cundy T, Glorieux FH, Lev D, Zacharin M, Oexle K, Marcelino J, Suwairi W, Heeger S, Sabatakos G, Apte S, Adkins WN, Allgrove J, Arslan-Kirchner M, Batch JA, Beighton P, Black GC, Boles RG, Boon LM, Borrone C, Brunner HG, Carle GF, Dallapiccola B, De Paepe A, Floege B, Halfhide ML, Hall B, Hennekam RC, Hirose T, Jans A, Jèuppner H, Kim CA, Keppler-Noreuil K, Kohlschuetter A, LaCombe D, Lambert M, Lemyre E, Letteboer T, Peltonen L, Ramesar RS, Romanengo M, Somer H, Steichen-Gersdorf E, Steinmann B, Sullivan B, Superti-Furga A, Swoboda W, van den Boogaard MJ, Van Hul W, Vikkula M, Votruba M, Zabel B, Garcia T, Baron R, Olsen BR, Warman ML. LDL receptor-related protein 5 (LRP5) affects bone accrual and eye development. Cell. 2001;107:513–523. [PubMed]
22. Xiong DH, Lei SF, Yang F, Wang L, Peng YM, Wang W, Recker RR, Deng HW. Low-density lipoprotein receptor-related protein 5 (LRP5) gene polymorphisms are associated with bone mass in both Chinese and whites. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research. 2007;22:385–393. [PubMed]
23. Koay MA, Woon PY, Zhang Y, Miles LJ, Duncan EL, Ralston SH, Compston JE, Cooper C, Keen R, Langdahl BL, MacLelland A, O’Riordan J, Pols HA, Reid DM, Uitterlinden AG, Wass JA, Brown MA. Influence of LRP5 polymorphisms on normal variation in BMD. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research. 2004;19:1619–1627. [PubMed]
24. Grundberg E, Lau EM, Lorentzson M, Karlsson M, Holmberg A, Groop L, Mellstrèom D, Orwoll E, Mallmin H, Ohlsson C, Ljunggren O, Akesson K. Large-scale association study between two coding LRP5 gene polymorphisms and bone phenotypes and fractures in men. Osteoporos Int. 2008;19:829–837. [PubMed]
25. Agueda L, Bustamante M, Jurado S, Garcia-Giralt N, Ciria M, Salâo G, Carreras R, Noguâes X, Mellibovsky L, Dâiez-Pâerez A, Grinberg D, Balcells S. A haplotype-based analysis of the LRP5 gene in relation to osteoporosis phenotypes in Spanish postmenopausal women. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research. 2008;23:1954–1963. [PubMed]
26. Boyden LM, Mao J, Belsky J, Mitzner L, Farhi A, Mitnick MA, Wu D, Insogna K, Lifton RP. High bone density due to a mutation in LDL-receptor-related protein 5. N Engl J Med. 2002;346:1513–1521. [PubMed]
27. Johnson ML. The high bone mass family—the role of Wnt/Lrp5 signaling in the regulation of bone mass. J Musculoskelet Neuronal Interact. 2004;4:135–138. [PubMed]
28. Tran BN, Nguyen ND, Eisman JA, Nguyen TV. Association between LRP5 polymorphism and bone mineral density: a Bayesian meta-analysis. BMC medical genetics. 2008;9:55. [PMC free article] [PubMed]
29. Kiel DP, Ferrari SL, Cupples LA, Karasik D, Manen D, Imamovic A, Herbert AG, Dupuis J. Genetic variation at the low-density lipoprotein receptor-related protein 5 (LRP5) locus modulates Wnt signaling and the relationship of physical activity with bone mineral density in men. Bone. 2007;40:587–596. [PMC free article] [PubMed]
30. Guo YF, Xiong DH, Shen H, Zhao LJ, Xiao P, Guo Y, Wang W, Yang TL, Recker RR, Deng HW. Polymorphisms of the low-density lipoprotein receptor-related protein 5 (LRP5) gene are associated with obesity phenotypes in a large family-based association study. J Med Genet. 2006;43:798–803. [PMC free article] [PubMed]
31. Fujino T, Asaba H, Kang MJ, Ikeda Y, Sone H, Takada S, Kim DH, Ioka RX, Ono M, Tomoyori H, Okubo M, Murase T, Kamataki A, Yamamoto J, Magoori K, Takahashi S, Miyamoto Y, Oishi H, Nose M, Okazaki M, Usui S, Imaizumi K, Yanagisawa M, Sakai J, Yamamoto TT. Low-density lipoprotein receptor-related protein 5 (LRP5) is essential for normal cholesterol metabolism and glucose-induced insulin secretion. Proc Natl Acad Sci U S A. 2003;100:229–234. [PubMed]
32. Venteclef N, Haroniti A, Tousaint JJ, Talianidis I, Delerive P. Regulation of anti-atherogenic apolipoprotein M gene expression by the orphan nuclear receptor LRH-1. J Biol Chem. 2008;283:3694–3701. [PubMed]
33. Francis GA, Fayard E, Picard F, Auwerx J. Nuclear receptors and the control of metabolism. Annu Rev Physiol. 2003;65:261–311. [PubMed]
34. Ory DS. Nuclear receptor signaling in the control of cholesterol homeostasis: have the orphans found a home? Circ Res. 2004;95:660–670. [PubMed]
35. Adami S, Braga V, Zamboni M, Gatti D, Rossini M, Bakri J, Battaglia E. Relationship between lipids and bone mass in 2 cohorts of healthy women and men. Calcif Tissue Int. 2004;74:136–142. [PubMed]
36. Dennison EM, Syddall HE, Aihie Sayer A, Martin HJ, Cooper C. Lipid profile, obesity and bone mineral density: the Hertfordshire Cohort Study. QJM. 2007;100:297–303. [PMC free article] [PubMed]
37. Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science (New York, NY) 1996;273:1516–1517. [PubMed]
38. Freimer NB, Sabatti C. Human genetics: variants in common diseases. Nature. 2007;445:828–830. [PubMed]
39. Chanock SJ, Manolio T, Boehnke M, Boerwinkle E, Hunter DJ, Thomas G, Hirschhorn JN, Abecasis G, Altshuler D, Bailey-Wilson JE, Brooks LD, Cardon LR, Daly M, Donnelly P, Fraumeni JF, Jr, Freimer NB, Gerhard DS, Gunter C, Guttmacher AE, Guyer MS, Harris EL, Hoh J, Hoover R, Kong CA, Merikangas KR, Morton CC, Palmer LJ, Phimister EG, Rice JP, Roberts J, Rotimi C, Tucker MA, Vogan KJ, Wacholder S, Wijsman EM, Winn DM, Collins FS. Replicating genotype–phenotype associations. Nature. 2007;447:655–660. [PubMed]
40. Ioannidis JP. Non-replication and inconsistency in the genome-wide association setting. Hum Hered. 2007;64:203–213. [PubMed]