Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Fine Mapping and Expression Studies Point to KIF5A as the Gene Responsible for Association of the 12q13 Locus with Rheumatoid Arthritis.

McClure,  Annie, Eyre,  Steve, Thomson,  Wendy, Worthington,  Jane, Barton,  Anne


A region on chromosome 12q13 was shown to have modest association with rheumatoid arthritis (RA) in the Wellcome Trust genome-wide association study (GWAS) of seven common diseases. Two parallel studies have since confirmed this as a RA susceptibility locus in UK and US populations (rs1678542 P=1.3×10-7 and P=5.4×10-6, respectively) and the same SNP has also been associated with type 1 diabetes (T1D). Variants mapping to nearby genes have also been associated with T1D and multiple sclerosis. The associated variant, rs1678542, is located in intron 15 of KIF5A. KIF5A is a microtubule motor protein involved in the transport of neurofilaments and expression was reported to be confined to neurons. However other ubiquitous kinesin family members have been shown to be involved in cell motility and intracellular transport of HLA-class II molecules.


The aim of the current study was to investigate the 12q13 region further to identify the most likely RA susceptibility gene in the region.


A region of ~400kb surrounding the associated variant on 12q13 was selected for fine mapping. Tag SNPs were chosen to capture all the known variation across the region (r2 >0.8). Any SNPs showing larger effect sizes than rs1678542 were then genotyped in a larger, independent validation cohort of 3633 RA cases and 2908 controls.

To investigate the expression of KIF5A, RNA was extracted from the blood of 52 RA patients and cDNA was synthesised. Whole blood RNA samples were selected to represent the range of possible genotypes at rs1678542 (major allele homozygous n=20, heterozygous n=19, minor allele homozygous n=13). KIF5A is known to be expressed in the brain so RNA was extracted from mouse brain to act as a positive control. Primers were designed to amplify KIF5A in human and mouse transcriptomes. The expression of two housekeeping genes (GAPDH and ACTIN) was also measured in order to normalise KIF5A expression. QRT-PCR was carried out using the ABI 7500.


In a fine-mapping cohort of 1,000 RA cases and 1,000 controls, 5 tag SNPs were identified that exhibited greater effect sizes than rs1678542. These 5 SNPs and rs1678542 were then genotyped in the larger validation cohort. Only rs1678542 remained significantly associated (p=0.004, OR 0.90 95%, CI 0.84–0.97). QRT-PCR of KIF5A confirmed its expression in whole blood. Correlation of expression with genotype at rs1678542 by linear regression showed a significant trend in expression by genotype (1×10-3). Individuals homozygote for the protective major allele had reduced KIF5A expression when compared to individuals who were heterozygote or homozygote for the minor allele at rs1678542.


After fine mapping of the region, association remains strongest with the original SNP identified in the WTCCC GWAS mapping to the KIF5A gene. The demonstration of expression of KIF5A in whole blood improves its potential for involvement in RA pathogenesis. Furthermore we present the first evidence of a correlation between KIF5A expression and genotype at a SNP associated with RA. The next step will be to validate this correlation using allele specific expression.

To cite this abstract, please use the following information:
McClure, Annie, Eyre, Steve, Thomson, Wendy, Worthington, Jane, Barton, Anne; Fine Mapping and Expression Studies Point to KIF5A as the Gene Responsible for Association of the 12q13 Locus with Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1593
DOI: 10.1002/art.29359

Abstract Supplement

Meeting Menu