Identification of a Sjögrens syndrome-associated variant that influences OAS1 isoform switching. H. Li1,2, J. A. Ice1, J. A. Kelly1, I. Adrianto1, S. B. Glenn1, K. S. Hefner3, E. Vista4, D. U. Stone2, R. Gopalakrishnan5, G. D. Houston2, D. M. Lewis2, M. Rohrer5, P. Hughes5, J. B. Harley6, C. G. Montgomery1, J. Chodosh7, J. A. Lessard8, J. Anaya9, B. M. Segal10, N. L. Rhodus5, L. Radfar2, R. H. Scofield1, C. J. Lessard1,2, K. L. Sivils1 1) Oklahoma Medical Research Foundation, Oklahoma City, OK; 2) University of Oklahoma Health Sciences Center, Oklahoma City, OK; 3) Hefner Eye Care and Optical Center, Oklahoma City, OK; 4) University of Santo Tomas Hospital, Manila, Philippines; 5) University of Minnesota, Minneapolis, MN; 6) Cincinnati Childrens Hospital Medical Center and the Department of Veterans Affairs Medical Center, Cincinnati, OH; 7) Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, MA; 8) Valley Bone and Joint Clinic, Grand Forks, ND; 9) Universidad del Rosario, Bogotá, Colombia; 10) Hennepin County Medical Center, Minneapolis, MN.
Sjögrens syndrome (SS) is a common, progressive autoimmune exocrinopathy characterized by symptoms of dry eyes and mouth. Our previous gene expression profiling (GEP) study using peripheral blood in 180 SS cases and 73 controls identified dysregulation of interferon (IFN) pathways in SS. Notably, OAS1, an IFN-inducible gene involved in the inhibition of virus replication, was overexpressed in SS patients. Through integration of GEP and genome-wide association study (GWAS) data, we identified multiple cis-expression quantitative trait loci (eQTL) in OAS1. The most significant eQTL in OAS1 is a splice site variant, rs10774671, located at the intersection between intron 5 and exon 6. Support for association of rs10774671 with SS was observed in our GWAS dataset with P=6×10-3. Our objective was to further evaluate the association of this eQTL with SS and characterize its functional mechanism. We tested association of rs10774671 with SS in an independent set of samples consisting of 648 cases and 2927 controls of European ancestry. Meta-analysis was then performed using a weighted Z score that generated Pmeta=9×10-6, with the A allele conferring risk. To determine the functional impact of rs10774671, we evaluated the alternative splicing events in OAS1 using both GEP and RNA-sequencing (RNA-seq) data. In GEP, one probe specifically recognizes a truncated form of OAS1 (p42). RNA-seq experiments were performed in 57 SS cases and 27 healthy controls on the Illumina platform. After quality control, OAS1 transcripts were reconstructed using Cufflinks and the relative abundance of each isoform was compared across samples with different rs10774671 genotypes. Both GEP and RNA-seq showed that the risk allele A, which demolishes the splicing consensus sequence, was correlated with higher expression of p42 (Pmicro=2×10-16 and Pseq=1×10-15). RNA-seq results also showed correlation of the A allele with higher proportions of p48 and p44 isoforms, but lower expression of the functionally normal isoform, p46. By using multiple genetic and genomic tools, we identified a SS-associated variant in OAS1 that switches the primary isoform from p46 to p42. Interestingly, p42 has been reported to result in a decreased OAS1 enzyme activity. This IFN-inducible gene is involved in viral RNA degradation and the inhibition of virus replication. These results indicate the risk allele in rs10774671 may cause vulnerability to virus infection that contributes to SS susceptibility.
You may contact the first author (during and after the meeting) at