Background Segmental duplication affecting the FCGR locus on chromosome 1 has led to ambiguous mapping of underlying genetic variants. Consequently the region is misrepresented in SNP databases, has been untested by genome wide association studies and dismissed as a potential contributor to genetic predisposition. Several large-scale studies identified genetic markers close to FCGR2A with divergent, yet reproducible associations with a number of autoimmune and inflammatory diseases. Our objective was to define the causal variants associated with FcγRIIa in disease susceptibility.

Methods Gene specific resequencing of FCGR2A was performed in a panel of 32 controls of mixed ethnicity. Fine scale linkage disequilibrium was used to infer trans-segmental duplication marker relationships, allowing reanalysis of Immunochip data from several autoimmune and inflammatory disease studies. Immunochip data for the FCGR2A locus from UK and Spanish RA, Spanish systemic sclerosis, Caucasian ankylosing spondylitis, UK and Irish psoriatic arthritis, European systemic lupus erythematosus, UK myositis, UK and Spanish giant cell arteritis and US/Turkish Takayasu arteritis cohorts were reanalysed. FcγRIIa ectodomain proteins of each allotype were synthesised and their IgG subclass binding preferences and biophysical properties were interrogated using surface plasmon resonance and SEC-MALLS.

Results Re-analysis of Immunochip data identified two groups of associated SNPs stratified by minor allele frequency, tagging Q27W and H131R substitutions. Haplotypes of these two risk markers showed different susceptibility associations with inflammatory diseases compared to the common 27Q-131R haplotype. Both UK and Spanish RA (OR (95% CI) 1.17 (1.07–1.27) p = 0.0003 and 1.24 (1.05–1.47) p = 0.011 respectively) and US Takayasu arteritis (OR 1.78 (1.21–2.63) p = 0.0037 were associated with the 27W-131H haplotype. The 27Q-131H haplotype predisposed to ankylosing spondylitis in Europeans (OR 1.15 (1.10–1.20) p = 2.04 × 10^–11), but was protective from SLE (OR 0.79 (0.74–0.84) p = 1.3 × 10^–13) and myositis (OR 0.83 (0.72–0.95) p = 0.0083). Using biophysical techniques these nonsynonymous coding SNPs altered  the affinity and binding preferences of the receptor for IgG subclasses and influenced receptor oligomerisation.

Conclusion Segmental duplication and structural variation confounded iChip SNP analyses, however FCGR2A specific resequencing allowed analysis of trans-segmental duplication markers. We confirmed two independent genetic risk factors at this locus. The functional impact and potential pathogenic role for FCGR2A in autoimmune/inflammatory disease is discussed.