Spector Tim D, Reneland Richard H, Mah Steven, Valdes Ana M, Hart Deborah J, Kammerer Stefan, Langdown Maria, Hoyal Carolyn R, Atienza Josephine, Doherty Michael, Rahman Proton, Nelson Matthew R, Braun Andreas
Twin Research and Genetic Epidemiology Unit, St. Thomas' Hospital, London, UK.
Arthritis Rheum. 2006 Feb;54(2):524-32. doi: 10.1002/art.21624.
To perform a large-scale association analysis of single-nucleotide polymorphisms (SNPs) in patients with radiographically defined osteoarthritis (OA) of the knee.
We examined >25,000 SNPs located within approximately 14,000 genes for associations with radiographically defined knee OA, using polymerase chain reaction and MassExtend amplification techniques. Allele frequencies were estimated initially in DNA pools from 335 female patients with knee OA and 335 asymptomatic and radiographically negative female control subjects. All were of northern European ancestry. Significant allele frequency differences were validated by genotyping of individual DNA samples. Confirmed significant findings were verified in 2 additional case-control samples from the UK (443 cases and 303 controls) and Newfoundland (346 cases and 264 controls). Chondrosarcoma cell lines were used to test for potential differences in gene expression.
The marker most strongly associated with the risk of knee OA was rs912428, a C/T polymorphism in intron 1 of LRCH1, a gene on chromosome 13q14 that encodes a novel protein of as-yet-unknown function. The frequency of the T allele compared with controls was consistently increased by 40% across all 3 case-control groups. Additional subanalyses in case-control samples with hip OA and hand OA suggested similar trends, but did not reach statistical significance. Association fine-mapping using 10 additional SNPs in LRCH1 confirmed intron 1 as the region of highest association but failed to reveal variations with significance stronger than the marker SNP, as did the haplotype analysis. LRCH1 was not up-regulated or overexpressed in chondrosarcoma cell lines exposed to inflammatory stimuli, suggesting a possible structural role.
A genetic variant in LRCH1 was consistently associated with knee OA in 3 samples from 2 populations. Our results also suggest that the same association with OA may exist at other sites. Additional genetic and experimental work is needed to elucidate the precise mechanism by which the LRCH1 gene influences OA risk.
对影像学确诊的膝关节骨关节炎（OA）患者进行单核苷酸多态性（SNP）的大规模关联分析。
我们使用聚合酶链反应和MassExtend扩增技术，检测了约14000个基因中的25000多个SNP与影像学确诊的膝关节OA的关联。最初在335名患膝关节OA的女性患者和335名无症状且影像学检查阴性的女性对照者的DNA池中估计等位基因频率。所有受试者均为北欧血统。通过对个体DNA样本进行基因分型来验证显著的等位基因频率差异。在另外2个来自英国（443例病例和303例对照）和纽芬兰（346例病例和264例对照）的病例对照样本中验证了得到确认的显著结果。使用软骨肉瘤细胞系检测基因表达的潜在差异。
与膝关节OA风险关联最密切的标记是rs912428，它是位于13q14染色体上LRCH1基因内含子1中的C/T多态性，该基因编码一种功能未知的新型蛋白质。在所有3个病例对照组中，与对照组相比，T等位基因的频率始终增加40%。在患髋OA和手OA的病例对照样本中进行的额外亚组分析显示出类似趋势，但未达到统计学显著性。使用LRCH1中的另外10个SNP进行的关联精细定位证实内含子1是关联度最高的区域，但未能揭示比标记SNP显著性更强的变异，单倍型分析也是如此。在暴露于炎症刺激的软骨肉瘤细胞系中，LRCH1未上调或过表达，提示其可能具有结构作用。
LRCH1中的一个基因变异在来自两个群体（人群）的3个样本中均与膝关节OA始终相关。我们的结果还提示在其他部位可能也存在与OA的相同关联。需要开展更多的遗传学和实验研究工作来阐明LRCH1基因影响OA风险的确切机制。
