P685. Ulcerative colitis risk loci in Lithuanian-Latvian case control sample – a replication study
J. Skieceviciene1, L. Kupcinskas1, L. Jonaitis1, A. Zvirbliene1, G. Denapiene2, J. Valantinas2, M. Leja3, A. Derovs4, G. Kiudelis1, S. Schreiber5, A. Franke5, 1Lithuanian University of Health Sciences, Institute for Digestive Research, Kaunas, Lithuania, 2Vilnius University, Center of Hepatology, Gastroenterology and Dietetics, Vilnius, Lithuania, 3University of Latvia, Institute of Oncology, Riga, Latvia, 4Riga Stradins University, 5Internal Disease Department, Riga, Latvia, 5Christian-Albrechts-University, Institute of Clinical Molecular Biology, Kiel, Germany
Ulcerative colitis (UC) is a chronic, relapsing inflammatory condition of the colon. Although the precise etiology is unknown, current data support the hypothesis of a complex interplay between genetic and environmental factors in disease pathogenesis. However, despite the recent investigation of considerable part of the genetic variance in ulcerative colitis patients, the possible epistatic interactions between these genetic variants are still unaccounted for. The aim of the current study was to investigate the role of known IBD-associated single nucleotide polymorphisms (SNPs) in a subset of UC patients from Lithuania and Latvia and evaluated possible epistatic interactions between these genetic variants.
We investigated 77 SNPs with moderate to strong association in five previously published genome-wide association studies of CD or UC, respectively. Our replication study comprised 447 Lithuanian and Latvian UC patients and 1,154 healthy controls. Single-marker case-control, genotype-phenotype association, SNP-SNP epistasis analyses and meta-analysis were performed.
After correcting for multiple testing, we confirmed associations at 21q21.1 (rs1736135, P = 8.01×10−6), 6q21 (rs7746082, P = 6.41×10−5), JAK2 (rs10758669, P = 8.08×10−6), RNF186 (rs3806308, P = 2.40×10−6), and ORMDL3 (rs2872507, P = 1.24×10−6). Pooling our data with the original UC data substantiated four of the associations and revealed additional six loci to be associated with UC at genome-wide level (P < 5×10−8). No association with any disease subphenotype was found. SNP-SNP interaction analysis showed significant epistasis between SNPs in the PTPN22 (rs2476601) and C13orf31 (rs3764147) genes and increased risk for UC (P = 1.64×10−6, OR = 2.44). In silico prediction of the interactive network of these genes further validated a possible interaction.
We confirmed the association of five loci (21q21.1, 6q21, JAK2, RNF186, and ORMDL3) with UC in the Lithuanian-Latvian population. SNP-SNP interaction analysis showed that the combination of SNPs in the PTPN22 (rs2476601) and C13orf31 (rs3764147) genes increase the risk for UC.