ICCBH2013 Oral Communications Biology (6 abstracts)
1Department of Internal Medicine, Erasmus MC, Rotterdam, The Netherlands; 2The Generation R Study Group, Erasmus MC, Rotterdam, The Netherlands; 3Department of Epidemiology, Erasmus MC, Rotterdam, The Netherlands; 4Hebrew SeniorLife and Harvard Medical School, Boston, Massachusetts, USA. *Winner of New Investigator Award
Aim: Lean and bone mass have considerably high phenotypic and genetic correlations with a shared heritability estimate ranging between 30 and 40% in adults. A genome-wide association study (GWAS) on total body lean mass and a bivariate GWAS on lean mass and BMD were ran in a cohort of children to identify genes with pleiotropic effects on muscle mass and peak bone mass attainment.
Methods: Subjects are part of the Generation R study, a prospective multiethnic birth cohort in Rotterdam, The Netherlands; we included 4096 children (mean age=6.2, S.D.=0.50 years) with total body DXA measurements (GE-Lunar iDXA) and genomewide genotyping (Illumina 660K). The univariate and bivariate GWAS were adjusted for age, sex, height, fat percent and 20 genomic principal components using bivariate PLINK. A P<5×10−8 was considered genome-wide significant (GWS).
Results: Genomic inflation factors were close to unity indicating adequate correction for stratification. In the univariate analysis we identified a GWS association with lean mass (β=0.13, P=2.9×10−8) for a SNP mapping to 11q13.2, in the PPP6R3/LRP5 locus. The SNP explained 0.8% of the variation in lean mass and was nominally significantly associated with BMD (β=0.10, P=7.6×10−5) explaining 0.4% of BMD variation. The association with lean mass was reduced after additional correction for bone mineral content (β=0.08, P=0.001), explaining 0.2% of the phenotypic variance. In the bivariate GWAS this SNP was also associated at GWS level (P=4.8×10−8) showing positive correlations of the bivariate trait with both lean mass (0.96) and BMD (0.68).
Conclusion: While LRP5 (an ubiquitously expressed gene member of the Wnt signaling pathway) is known to play a key role in bone mechanosensing with GWAS showing association with BMD and fracture in elderly adults; we showed that genetic variation in the PPP6R3/LRP5 locus exerts pleitropic effects on muscle mass and peak bone mass acquisition of children. Given the high regional LD it is difficult to establish from which gene the GWAS signal is arising. PPP6R3 is a gene of unknown function ubiquitously expressed across tissues among others in bone and muscle. Replication in additional children cohorts is underway while the exact same SNP has been found associated at genome significant level in a bivariate GWAS of bone strength and lean mass in two large consortia of adult individuals. Such pleiotropic effects on muscle mass and BMD observed in children are still evident later in life.