Searchable abstracts of presentations at key conferences on calcified tissues
Bone Abstracts (2014) 3 LB1 | DOI: 10.1530/boneabs.3.LB1

ECTS2014 Late Breaking Abstracts (1) (5 abstracts)

Endochrondral ossification, mesenchymal stem cell and Wnt pathway specific loci predict differential skeletal effects in high bone mass

Celia Gregson 1 , John Kemp 2, , Mhairi Marshall 2 , George Davey Smith 4 , Matthew Brown 2 , Emma Duncan 2, & Jon Tobias 1


1Musculoskeletal Research Unit, University of Bristol, Bristol, UK; 2University of Queensland Diamantina Institute, Brisbane, Queensland, Australia; 3Royal Brisbane and Women’s Hospital, Brisbane, Queensland, Australia; 4MRC Integrative Epidemiology Unit, University of Bristol, Bristol, UK.


Extreme high bone mass (HBM) may be monogenic (e.g. LRP5 mutations) or polygenic, due to variants in the same genes determining bone mineral density (BMD) as found in the general population. We aimed to determine how variation in established BMD loci, in different functional pathways, explains the HBM phenotype.

241 unexplained HBM cases (lumbar spine(LS)1+total hip(TH) Z-scores≥+4.4) were recruited from 15 UK centres, by screening 335 115 DXA scans. Established LRP5 mutations were excluded by Sanger sequencing (n=6). Infinium OmniExpress-12v1.0 genotypes were imputed to UK10K; 57 SNPs had certainty>0.8. Using Estrada’s Femoral Neck(FN) or LS Betas, we calculated weighted genetic risk scores (GRSFN/GRSLS) for total (n=57), Wnt (n=12), OPG-RANK-RANKL (n=3), endochrondral ossification (n=6) and mesenchymal stem cell (MSC) differentiation (n=4) annotated loci.

A one SD increase in total GRSFN was associated with a 0.13 S.D. (−0.01, 0.27), P=0.06, increase in TH BMDTH and 0.13 (0.01, 0.25), P=0.04, increase in total body BMDTB, explaining 1.6 and 3.2% phenotypic variance respectively however, GRSLS was independent of LS BMDLS (0.12 (−0.05, 0.29), P=0.17, r2=0.8%).

Wnt GRS was associated with a 0.15 S.D. (0.01, 0.29), P=0.04, increase in BMDTH (r2=2%), but was independent of BMDTB (0.03 (−0.08, 0.15), P=0.57, r2=0.3%) and BMDLS (−0.02 (−0.20, 0.15), P=0.79, r2=0%). Conversely, the endochrondral ossification GRS was independent of BMDTH (0.01 (−0.13, 0.15), P=0.85, r2=0%) and BMDLS (0.14 (−0.03, 0.31), P=0.11, r2=1.2%), but explained 2.3% of variance in BMDTB (0.11 (−0.01, 0.23), P=0.08). Whereas, MSC GRS was associated with a 0.19 S.D. (0.02, 0.37), P=0.03, r2=2.2%) increase in BMDLS, but was independent of both BMDTB and BMDTH (P>0.8, r2=0%). All BMD sites were independent of OPG-RANK-RANKL GRS.

In conclusion, BMD in HBM appears driven by osteoblast rather than osteoclast pathways. A greater proportion of phenotypic variance in i) hip BMD is explained by wnt pathway loci, ii) total body BMD by endochrondral ossification loci and iii) lumbar spine BMD by mesenchymal stem cell differentiation loci, suggesting differential genetic regulation of individual skeletal sites.

Volume 3

European Calcified Tissue Society Congress 2014

Prague, Czech Republic
17 May 2014 - 20 May 2014

European Calcified Tissue Society 

Browse other volumes

Article tools

My recent searches

No recent searches.