Abstract
Background Systemic lupus erythematosus (SLE) is a complex autoimmune disorder with significant genetic underpinnings. Multiple genome-wide association studies (GWAS) have been conducted with identification of several SLE genes and loci. However, these only explain about 15% of SLE genetic risk indicating that additional loci of modest effect remain to be discovered. Association clustering methods such as OASIS are more apt than single-variant analysis for identifying modest genetic effects.
Methods Previously, OASIS analysis of two dbGAP GWAS datasets (6077 subjects; 0.75 million single-nucleotide polymorphisms (SNPs)) identified 30 SLE associated loci. In the present study, 410 genes were mapped to these loci and investigated for gene expression in 3 SLE GEO datasets. A fourth expression dataset, GSE50395, from Cordoba, Spain was used for validation. Expression quantitative trait loci (eQTL) analysis was performed using Blood eQTL browser for significant SNPs in loci harboring differentially expressed genes. Protein network analysis was performed using STRING to determine functional pathways involved. Confirmatory qPCR on monocytes of 12 healthy donors and 12 SLE patients was performed for the significant gene.
Results This genetic and functional analysis identified 187 genes in GEO datasets that were differentially expressed in SLE. Of these 55 genes were found to be differentially expressed in at least 2 GEO datasets with all probes directionally aligned. In the Cordoba dataset, 10 genes were found to have altered expression in SLE. DDX11 was significantly down regulated in both the GEO (P=3.60E-02) and the Cordoba (P=8.02E-03) datasets. The most significant SNP, rs3741869 (GWAS P=3.2E-05) in OASIS locus 19 (chromosome 12p11.21), containing the gene DDX11, was found to be a cis-eQTL regulating the expression of DDX11 (P=8.62E-05). DDX11 expression in SLE was confirmed to be down regulated using qPCR (P=0.001), ratifying DDX11 as a novel gene associated with SLE. DDX11 interacted with multiple genes identified using genomic convergence of OASIS loci and expression analysis, including STAT1/STAT4. This study identified several novel SLE candidate genes as well, such as NRF1, TTC9 and BARHL1 which need additional data for verification.
Conclusion Genomic convergence approach with a locus-based test, OASIS, and expression studies on multiple datasets can help identify novel SLE genes. This study showed that DDX11, a RNA helicase thought to be involved in genome stability, is repressed in SLE and its expression is modulated by rs3741869.