PT - JOURNAL ARTICLE AU - Daniel S Evans AU - Frederic Cailotto AU - Neeta Parimi AU - Ana M Valdes AU - Martha C Castaño-Betancourt AU - Youfang Liu AU - Robert C Kaplan AU - Martin Bidlingmaier AU - Ramachandran S Vasan AU - Alexander Teumer AU - Gregory J Tranah AU - Michael C Nevitt AU - Steven R Cummings AU - Eric S Orwoll AU - Elizabeth Barrett-Connor AU - Jordan B Renner AU - Joanne M Jordan AU - Michael Doherty AU - Sally A Doherty AU - Andre G Uitterlinden AU - Joyce B J van Meurs AU - Tim D Spector AU - Rik J Lories AU - Nancy E Lane TI - Genome-wide association and functional studies identify a role for <em>IGFBP3</em> in hip osteoarthritis AID - 10.1136/annrheumdis-2013-205020 DP - 2015 Oct 01 TA - Annals of the Rheumatic Diseases PG - 1861--1867 VI - 74 IP - 10 4099 - http://ard.bmj.com/content/74/10/1861.short 4100 - http://ard.bmj.com/content/74/10/1861.full SO - Ann Rheum Dis2015 Oct 01; 74 AB - Objectives To identify genetic associations with hip osteoarthritis (HOA), we performed a meta-analysis of genome-wide association studies (GWAS) of HOA.Methods The GWAS meta-analysis included approximately 2.5 million imputed HapMap single nucleotide polymorphisms (SNPs). HOA cases and controls defined radiographically and by total hip replacement were selected from the Osteoporotic Fractures in Men (MrOS) Study and the Study of Osteoporotic Fractures (SOF) (654 cases and 4697 controls, combined). Replication of genome-wide significant SNP associations (p ≤5×10−8) was examined in five studies (3243 cases and 6891 controls, combined). Functional studies were performed using in vitro models of chondrogenesis and osteogenesis.Results The A allele of rs788748, located 65 kb upstream of the IGFBP3 gene, was associated with lower HOA odds at the genome-wide significance level in the discovery stage (OR 0.71, p=2×10−8). The association replicated in five studies (OR 0.92, p=0.020), but the joint analysis of discovery and replication results was not genome-wide significant (p=1×10−6). In separate study populations, the rs788748 A allele was also associated with lower circulating IGFBP3 protein levels (p=4×10−13), suggesting that this SNP or a variant in linkage disequilibrium could be an IGFBP3 regulatory variant. Results from functional studies were consistent with association results. Chondrocyte hypertrophy, a deleterious event in OA pathogenesis, was largely prevented upon IGFBP3 knockdown in chondrocytes. Furthermore, IGFBP3 overexpression induced cartilage catabolism and osteogenic differentiation.Conclusions Results from GWAS and functional studies provided suggestive links between IGFBP3 and HOA.