We genotyped ten SNPs in PTPN22, including two missense (rs2476601 (R620W), rs33996649 (R263Q)) and 5′-UTR (rs2488457) SNPs, in four large multi-ethnic populations. Using these genotypes and HapMap data we imputed 107 SNPs in diverse ethnic groups to elucidate the effect size and potential relation of these variants with SLE risk across populations. Our results were consistent with previous reports: we reinforced that SLE-association with PTPN22 is largely accounted for by rs2476601 in individuals of European ancestry [1], [48], [49], [50]. While there is a strong North-South gradient of the risk allele frequency across Europe [26], we did not have sufficient AIMs determine North vs. South European ancestry we could not assess the effect of this gradient in our study. We were unable to detect association with rs2476601 in AA or AS, or any other SNP within PTPN22. This result is consistent with earlier reports that this SNP is rare and not associated with SLE in AA [3], [51] or in Han Chinese [52].
