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.
A Comprehensive Replication Study and Functional Analysis of SLE Associated Variants in BLK, ATG5, PXK and SCUBE1.
Delgado-Vega6, Angelica M., Kozyrev6, Sergey, Frostegard4, Johan, Truedsson5, Lennart, Pons-Estel8, Bernardo A., D'Alfonso11, Sandra, Witte1, Torsten
Hannover Medical School, Germany
U do Porto, Porto, Portugal
U Eastern Piedmont, Novara, Italy
U of Szeged, Hungary
Hosp Santo Antonio and ICBAS, Porto, Portugal
Inst d Biomedicina y Parasitología López-Neyra, Granada, Spain
Karolinska University Hosp at Huddinge, Stockholm, Sweden
Lund U, Sweden
Rudbeck Laboratory, Uppsala U, Sweden
Rudbeck Laboratory, Uppsala U, Sweden. OMRF, OK. Andalucian Center f Genomics and Oncological Research Pfizer-University of Granada-Junta de Andalucia, Spain
Sanatorio Parque, Rosario, Argentina
U Catholique d Louvain, Bruxells, Belgium
The HapMap International project has provided an extensive catalogue of the common human genetic variation. It made possible the design of genome-wide association studies, which have radically contributed to the dissection of the genetic basis of SLE and multiple susceptibility variants have been identified. The 1000 genomes project, although not completed yet, is providing a deep source of newly identified variants that are ready to be used. In the present work, we have fine mapped four SLE-associated genes and imputed a handful of SNPs using these data, revealing novel potentially functional variants.
Fine mapping of BLK, ATG5, PXK and SCUBE1 was performed in the European multicenter collection "BIOLUPUS". Individuals with <90% of European ancestry identified by using PCA and STRUCTURE were removed as well as duplicated or related samples. Publicly available reference haplotypes and fine-scale recombination maps were downloaded from the 1000 genomes project (release August 2009) and HapMap phase 3 (CEU+TSI) for chromosomes 3, 6, 8 and 22. We combine these two reference panels and imputed the missing SNPs in our study data set by IMPUTEv2. Only SNPs for which the probabilities of each imputed genotype was >90% and the missing rate was <5% were included. A total of 299 SNPs in PXK, 60 SNPs in ATG5, 66 SNPs in BLK and 115 SNPs in SCUBE1 were tested for association in 1256 cases and 1576 controls, adjusting by the country of origin using GENABELv1.44 and PLINKv7. Multiple testing was corrected by genomic control and false discovery rate methods. The minimum set of SNPs explaining the association was calculated by multivariate logistic regression. Any SNP belonging to this set was typed, if it was an imputed SNP. A set of selected regions for each gene was re-sequenced in a number of individuals. Functional studies including expression and splicing analysis were performed.
The only gene with genome-wide significance of association was BLK (P= 6.5×10-7 for the strongest SNP). The previously associated variants rs13277113 (P= 1.58×10-6) and rs2736340 (P= 1.82×10-6) ranked 2nd and 4th respectively. Conditional analysis suggested three independent groups of SNPs: one located in the promoter region (including rs13277113), another one in intron 1 and a third group of low frequency variants (MAF < 0.01) in the 3'UTR region.
In ATG5, the association was explained by two independent SNPs in the surroundings of exon 5. In PXK, a long haplotype of highly correlated SNPs in LD with the previously associated variant rs6445975 was delimited to the end of the gene. Association analysis with the levels of expression of these three genes is presented.
None of the SNPs in SCUBE1 reached significance after correction for multiple testing. Extremely high heterogeneity in the distribution of haplotypes and patterns of LD were observed among different countries.
We not only replicated the association of BLK, ATG5 and PXK but also present an extensive mapping of these genes. We narrowed down the associated regions and performed functional studies in order to understand the role of potentially causal variants on gene function.
To cite this abstract, please use the following information:
Delgado-Vega, Angelica M., Kozyrev, Sergey, Frostegard, Johan, Truedsson, Lennart, Pons-Estel, Bernardo A., D'Alfonso, Sandra, et al; A Comprehensive Replication Study and Functional Analysis of SLE Associated Variants in BLK, ATG5, PXK and SCUBE1. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1578