|
nif:isString
|
-
The Avon Longitudinal Study of Parents and their Children (ALSPAC) is a longitudinal population-based birth cohort that initially included over 13,000 women and their children in Avon, UK, in the early 1990s. This cohort is described in detail on the website (http://www.alspac.bris.ac.uk) and elsewhere [14]. Both mothers and children have been extensively followed from the 8th gestational week onwards using a combination of self-reported questionnaires, medical records and physical examinations. Ethical approval was obtained from the ALSPAC Law and Ethics committee and relevant local ethics committees, and written informed consent provided by all parents. Blood samples were taken and DNA extracted as previously described [15].
pQCT scans were performed on approximately 4500 children when they attended the age 15 research clinic at which time total body DXA scans were also performed. Results for hip DXA scans, collected at the age 13 research clinic, were also analysed. At both time points, height was measured using a Harpenden stadiometer (Holtain Limited, Wales) and weight using a Tanita Body Fat Analyser. Cortical bone mineral content (BMCC), cortical bone mineral density (BMDC) and cortical bone area (BAC), were measured on a single slice at the mid tibia using the Stratec XCT2000L (Germany). Periosteal circumference (PC), endosteal circumference (EC) and cortical thickness (CT) were derived using a circular ring model. A threshold routine was used for defining cortical bone, which specified a voxel with a density >650 mg/cm3 as cortical bone. Of the 4500 pQCT scans obtained in ALSPAC 89 were rejected as being of insufficient quality. The coefficients of variation (CVs) based on 139 ALSPAC subjects scanned a mean of 31 days apart, were 2.7%, 1.3% and 2.9% for BMCC, BMDC, BAC, respectively. Total body BMD and BMD of the left femoral neck were measured using a Lunar Prodigy scanner, for which CVs were 1.0% (146 subjects) and 1.7% (166 subjects) respectively.
1760 ALSPAC individuals were genotyped using the Illumina HumanHap317K SNP chip. Markers with <1% minor allele frequency, >5% missing genotypes or which failed an exact test of Hardy-Weinberg equilibrium (HWE) (p<1×10−7) were excluded from further analysis. We also excluded any individuals who did not cluster with the CEU individuals in multidimensional scaling analysis, who had >5% missing data, heterozygosity of >36.4% or <34.3% and a male who scored heterozygous at many X chromosome loci. After data cleaning we were left with 1518 individuals (999 with pQCT data) and 315,807 SNPs. We carried out imputation to HapMap release 22 using Mach 1.0, Markov Chain Haplotyping [16].
Genotyping (of the SNPs with p<1×10−5 in the GWAS meta-analysis) was carried out on the entire ALSPAC cohort for whom DNA was available (10121 individuals) by KBioscience (http://www.kbioscience.co.uk), who employ a novel form of competitive allele specific PCR (KASPar) for genotyping. Those individuals who were included in the ALSPAC discovery analysis as well as those of non-white reported ethnicity, with >10% missing genotypes and with siblings in the cohort were excluded from the analysis. For each SNP there were between 2753 and 2789 individuals with both pQCT and genotype data. All but one SNP (rs9541712) were successfully genotyped. All were in HWE.
Measurement of free RANKL levels: Soluble RANKL levels (sRANKL) were measured on fasting blood samples (which were obtained according to standard protocols, collected in heparin tubes, and stored at minus 80 degrees until further use), in 37 male individuals attending the age 15 research clinic, randomly selected after stratification by RANKL rs1021188 genotype. Measurements, which were performed blind, were carried out using the ampli-sRANKL enzyme-linked immunosorbent assay (ELISA) from Biomedica (Vienna, Austria) according to manufacturer's instructions, with the exception that 90µl of plasma was used per well. Measurement ranges, intra- and inter-assay coefficients of variation (CVs) were 0.02–2pmol/l, <9% and <6% respectively. Duplicate samples with a coefficient of variation of <20% were considered for further statistical analysis. In sample(s) where the recorded concentration was below the detection limit of 0.02 pmol/l, the latter value was entered in subsequent analyses.
The Gothenburg Osteoporosis and Obesity Determinants (GOOD) study was initiated to determine both environmental and genetic factors involved in the regulation of bone and fat mass [17], [18]. Male study subjects were randomly identified in the greater Gothenburg area in Sweden using national population registers, contacted by telephone, and invited to participate. To be enrolled in the GOOD study, subjects had to be between 18 and 20 years of age. There were no other exclusion criteria, and 49% of the study candidates agreed to participate (n = 1068). The subjects of the GOOD cohort were also analysed after five years of follow-up, between 23 and 25 years of age. The GOOD study was approved by the local ethics committee at Gothenburg University. Written and oral informed consent was obtained from all study participants. Height was measured using a wall-mounted stadiometer, and weight was measured to the nearest 0.1 kg.
BMDC, BMDC and BAC were measured on a single tibial diaphyseal slice (at 25% of the bone length in the proximal direction of the distal end of the bone) using the Stratec XCT2000 (Germany). PC, EC and CT were derived using a circular ring model. A threshold routine was used for defining cortical bone, which specified a voxel with a density >710 mg/cm3 as cortical bone. Trabecular vBMD was measured using a scan through the distal metaphysis (at 4% of the bone length in the proximal direction of the distal end of the bone) of tibia. Tibia length was measured from the medial malleolus to the medial condyle of the tibia. The CVs were <1% for all pQCT measurements. aBMD (g/cm2) of the whole body, femoral neck (of the left leg), and lumbar spine were assessed using the Lunar Prodigy DXA (GE Lunar, Madison, WI, USA). The CVs for the aBMD measurements ranged from 0.5% to 3%, depending on application.
Genotyping was performed with Illumina HumanHap610 arrays at the Genetic Laboratory, Department of Internal Medicine, Erasmus Medical Center, Rotterdam, the Netherlands. Genotypes were called using the BeadStudio calling algorithm. Genotypes from 935 individuals passed the sample quality control criteria [exclusion criteria: sample call rate <97.5%, gender discrepancy with genetic data from X-linked markers, excess autosomal heterozygosity >0.33 (∼FDR<0.1%), duplicates and/or first degree relatives identified using IBS probabilities (>97%), ethnic outliers (3 SD away from the population mean) using multi-dimensional scaling analysis with four principal components]. Across 22 duplicate samples, genotype concordance exceeded 99.9%. We carried out imputation to HapMap release 22 (after excluding SNPs with MAF<1%, SNP call rate <98% and HWE p value <1×10−6) using Mach 1.0, Markov Chain Haplotyping [16].
The Osteoporotic Fractures in Men (MrOS) study is a multicenter, prospective study including older men in Sweden (3014), Hong Kong (≅2000), and the United States (≅6000). In the present study, associations between candidate polymorphism and skeletal parameters were investigated in the Swedish cohort (Table 1), which consists of three sub-cohorts from three different Swedish cities (n = 1005 in Malmö, n = 1010 in Göteborg, and n = 999 in Uppsala). Study subjects (men aged 69–81) were randomly identified using national population registers, contacted and asked to participate. To be eligible for the study, the subjects had to be able to walk without assistance, provide self-reported data, and sign an informed consent; there were no other exclusion criteria [19]. The study was approved by the ethics committees at the Universities of Gothenburg, Lund, and Uppsala. Informed consent was obtained from all study participants.
Validated pQCT analyses were available for the Gothenburg and Malmö cohorts. BMCC, BMDC and BAC, were measured on a single tibial diaphyseal slice slice (at 38% of the bone length in the proximal direction of the distal end of the bone) using the Stratec XCT2000 (Germany). PC, EC and CT were derived using a circular ring model. A threshold routine was used for defining cortical bone, which specified a voxel with a density >710 mg/cm3 as cortical bone. Trabecular vBMD was measured using a scan through the distal metaphysis (at 4% of the bone length in the proximal direction of the distal end of the bone) of tibia. The CVs were <1% for all pQCT measurements. Adjustments for study centre were performed.
Total body areal BMD (aBMD, g/cm2), as well as aBMD of the femoral neck and lumbar spine (L1-L4) were assessed at baseline using the Lunar Prodigy dual energy X-ray absorptiometry (DXA) (n = 2004 from the Uppsala and Malmö cohorts; GE Lunar Corp., Madison, WI, USA) or Hologic QDR 4500/A-Delphi (n = 1010 from the Göteborg cohort; Hologic, Waltham, MA, USA). The CVs for the aBMD measurements ranged from 0.5% to 3%, depending on the application. To be able to use DXA measurements performed with equipment from two different manufacturers, a standardized BMD (sBMD) was calculated, as previously described [19]. Adjustments for study centre were performed.
Genotyping (of the SNPs with p<1×10−5 in the GWAS meta-analysis) was carried out using matrix-assisted laser desorption ionization-time of flight mass spectrometry on the Sequenom MassARRAY platform (San Diego, CA, USA). The genotyping call rate was >97% and the SNP-s were in Hardy-Weinberg equilibrium.
The ALSPAC discovery set (n = 999) and GOOD (n = 935) contributed to the genome-wide meta-analysis. We analysed only those imputed SNPs which had a minor allele frequency of >0.01 and an r2 imputation quality score of >0.3 in both sets (n = 2,417,199). We carried out genome-wide association analyses for BMDC using additive linear regression in Mach2QTL for both ALSPAC and GOOD (using GRIMP [20] for the GOOD analyses). We included age, sex, height and weight(ln) as covariates in ALSPAC, and age, height and weight(ln) as covariates in the male only GOOD cohort. We carried out meta-analyses of the results from the two cohorts using two methods in METAL (www.sph.umich.edu/csg/abecasis/metal). In the p-value meta-analysis study-specific Z-statistics are calculated (which summarise the p-values and direction of effect) for each SNP's association. The Z-statistics are then summed across studies, using weights proportional to the square-root of each study's sample size, to provide a summary p-value for each association. In the inverse variance method standardized betas and standard errors from each study are combined using a fixed effect model which weights the studies using the inverse variance. We also carried out the meta-analyses with and without genomic control. The results using each method were very similar and the selection of SNPs was based on the p-value meta-analysis not adjusted for genomic control. Genome-wide significance was taken to be p<5×10−8. We selected one SNP from each independent region that had a p<1×10−5 for replication in the ALSPAC replication cohort and MrOS Sweden. Additive linear regression analyses were carried out for the associations between these SNPs and BMDC in PLINK [21] (ALSPAC) or in SPSS Statistics 17.0 (MrOS Sweden) using age, sex, height and weight(ln) as covariates. We calculated the combined p-value (for all four cohorts) using METAL (method described above).
Association between the RANKL SNP and other traits: For the RANKL SNP (rs1021188) we also tested for associations with other bone traits; BAC, BMCC, PC, EC and CT using the pQCT data and TB BMD, FN BMD and LS BMD in all cohorts (where the appropriate measures were available). The ALSPAC discovery and replication cohorts were combined and we also show the association results from GOOD from two time-points (the original GWAS time-point (19 years) and data from the five year follow-up visit (24 years). We carried out association analyses using additive linear regression in PLINK for ALSPAC and in SPSS Statistics 17.0 for GOOD and MrOS Sweden. We included age, sex, height and weight(ln) as covariates in the model in ALSPAC, and age, height and weight(ln) as covariates in the male only GOOD and MrOS Sweden cohorts. Pubertal stage was also included as a covariate in the FN BMD analyses of ALSPAC. An interaction between sex and rs1021188 in association with BMDC in the ALSPAC cohort was tested in PLINK, which tests the significance of the interaction term in the linear model. We carried out meta-analyses using standardized betas and standard errors from each of the studies. This was carried out in METAL using the inverse-variance method described above. sRANKL levels were transformed using the Rank-Based Inverse Normal Transformation. Linear regression was used to determine the relationship between sRANKL level and the addition of the minor rs1021188 allele.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}