PLoS ONE (Jan 2011)

Large scale replication study of the association between HLA class II/BTNL2 variants and osteoarthritis of the knee in European-descent populations.

  • Ana M Valdes,
  • Unnur Styrkarsdottir,
  • Michael Doherty,
  • David L Morris,
  • Massimo Mangino,
  • Agu Tamm,
  • Sally A Doherty,
  • Kalle Kisand,
  • Irina Kerna,
  • Ann Tamm,
  • Margaret Wheeler,
  • Rose A Maciewicz,
  • Weiya Zhang,
  • Kenneth R Muir,
  • Elaine M Dennison,
  • Deborah J Hart,
  • Sarah Metrustry,
  • Ingileif Jonsdottir,
  • Gudbjorn F Jonsson,
  • Helgi Jonsson,
  • Thorvaldur Ingvarsson,
  • Cyrus Cooper,
  • Timothy J Vyse,
  • Tim D Spector,
  • Kari Stefansson,
  • Nigel K Arden

DOI
https://doi.org/10.1371/journal.pone.0023371
Journal volume & issue
Vol. 6, no. 8
p. e23371

Abstract

Read online

Osteoarthritis (OA) is the most common form of arthritis and a major cause of disability. This study evaluates the association in Caucasian populations of two single nucleotide polymorphisms (SNPs) mapping to the Human Leukocyte Antigen (HLA) region and deriving from a genome wide association scan (GWAS) of knee OA in Japanese populations. The frequencies for rs10947262 were compared in 36,408 controls and 5,749 knee OA cases from European-descent populations. rs7775228 was tested in 32,823 controls and 1,837 knee OA cases of European descent. The risk (major) allele at rs10947262 in Caucasian samples was not significantly associated with an odds ratio (OR) = 1.07 (95%CI 0.94 -1.21; p = 0.28). For rs7775228 the meta-analysis resulted in OR = 0.94 (95%CI 0.81-1.09; p = 0.42) for the allele associated with risk in the Japanese GWAS. In Japanese individuals these two SNPs are in strong linkage disequilibrium (LD) (r(2) = 0.86) with the HLA class II haplotype DRB1*1502 DQA1*0103 DQB1*0601 (frequency 8%). In Caucasian and Chinese samples, using imputed data, these SNPs appear not to be in LD with that haplotype (r(2)<0.07). The rs10947262 and rs7775228 variants are not associated with risk of knee OA in European descent populations and they do not appear tag the same HLA class II haplotype as they do in Japanese individuals.