Genome-Wide Association Study On the Severity of Joint Destruction in Autoantibody Positive Rheumatoid Arthritis Identifies a Role for Sperm Associated Antigen 16

Abstract

Background/Purpose: Recent genome-wide association studies (GWAs) have identified > 30 SNPs predisposing to Rheumatoid Arthritis(RA). These variants are helpful in unraveling the pathogenesis of RA. However, most therapeutic strategies target pathways of disease progression. Genetic factors account for a considerable proportion of variance in joint damage, but thus far only a few replicated severity factors are known and no GWAS has been performed. We aimed to increase the understanding of the processes underlying the inter-individual differences in joint damage in anti-citrullinated peptide anti-bodies (ACPA)-positive RA by performing a 3-staged GWAS on joint damage progression using high-quality radiology data, followed by in vitro and ex vivo studies. Methods: Stage 1 was performed on 385 ACPA-positive RA-patients Methods: Stage 1 was performed on 385 ACPA-positive RA-patients from the NARAC using Illumina HumanHap 550k BeadChips. Stage 2 concerned of 1,567 X-rays of 301 ACPA-positive RA-patients included in a Dutch cohort with 7 years follow-up. In stage 3, 861 X-rays of 742 North-American ACPA-positive RA-patients included in the NDB and Wichita-cohorts were studied. All X-rays were scored using the Sharp-van der Heijde method (ICCs all 0.9). The expression of SPAG16 variants was studied by RT-qPCR using a RA synovium cDNA library and cDNA derived from other RA tissues and fibroblast-like synoviocytes (FLS). Expression levels of MMP1 and MMP3 of FLS before and after stimulation with TNF- (10 ng/ml) and IL1 (1ng/ml) were evaluated by RT-qPCR and ELISA (cell culture supernatants). Finally serum MMP3 levels were measured in RA patients of stage 2 using ELISA. Results: In stage 1, the strongest association was observed for a cluster of SNPs at 2q34, the region of Sperm associated AntiGen16 (SPAG16, P 4.55 10 7, 0.77 fold progression rate per year per minor allele). Independent replication was obtained in stage 2 and 3, again observing a protective effect on damage progression (P 2.16 10 2 and 2.29 10 2 resp.). Apart from its role in spermatozoa, the function of SPAG16 is incompletely known. We detected SPAG16 isoforms in RA tissues and FLS. No relation between expression levels of SPAG16 transcripts in FLS and SPAG16 genotypes could be detected. However, the matrix degrading capacity of FLS may be affected. FLS of patients with the minor allele tended to express less MMP3 mRNA and secreted lower levels of MMP3 (P 2.28 10 2). Also after cytokine stimulation the minor allele was associated with less production of MMP3. Furthermore, RA-patients carrying the minor allele had lower serological levels of MMP3 (P 4.59 10 2) and lower MMP3 levels were associated with less progression of damage(P 3.09 10 3). Conclusion: A genetic variant in SPAG16 is associated with less production of MMP3 by FLS and protection against joint damage progression. These findings indicate a new pathway involved in joint damage in ACPA-positive RA.