Włodzimierz Zych, Agnieszka Kowalik, Agnieszka Paziewska, Alina Kanikowska, Tomasz Mach, Bożena Walewska-Zielecka, Jakub Karczmarski, Marek Krawczyk, Agnieszka Rogowska, Tomasz Bobiński, Jerzy Ostrowski, Grzegorz Boryczka, Maria Janiak, Halina Cichoż-Lach, Andrzej Habior, Joanna Musialik, Ewa Wunsch, Piotr Milkiewicz, Marek Hartleb, Michal Mikula, Irena Ciecko-Michalska, Małgorzata Ferenc, Joanna Raczyńska, Krzysztof Mucha, Joanna Raszeja-Wyszomirska, Michalina Dabrowska, Michał Wasilewicz, Rafał Stankiewicz, Marian Grzymisławski, Krzysztof Goryca, and Filip Ambrozkiewicz
Background Primary biliary cholangitis (PBC) and primary sclerosing cholangitis (PSC) are forms of hepatic autoimmunity, and risk for both diseases has a strong genetic component. This study aimed to define the genetic architecture of PBC and PSC within the Polish population. Methods Subjects were 443 women with PBC, 120 patients with PSC, and 934 healthy controls recruited from Gastroenterology Departments in various Polish hospitals. Allelotyping employed a pooled-DNA sample-based genome-wide association study (GWAS) approach, using Illumina Human Omni2.5-Exome BeadChips and the following novel selection criteria for risk loci: blocks of at least 10 single nucleotide polymorphisms (SNPs) in strong linkage disequilibrium, where the distance between each adjacent SNP pair in the block was less than 30 kb, and each SNP was associated with disease at a significance level of P < 0.005. A selected index SNP from each block was validated using TaqMan SNP genotyping assays. Results Nineteen and twenty-one SNPs were verified as associated with PBC and PSC, respectively, by individual genotyping; 19 (10/9, PBC/PSC) SNPs reached a stringent (corrected) significance threshold and a further 21 (9/12, PBC/PSC) reached a nominal level of significance (P < 0.05 with odds ratio (OR) > 1.2 or < 0.83), providing suggestive evidence of association. The SNPs mapped to seven (1p31.3, 3q13, 6p21, 7q32.1, 11q23.3, 17q12, 19q13.33) and one (6p21) chromosome region previously associated with PBC and PSC, respectively. The SNP, rs35730843, mapping to the POLR2G gene promoter (P = 1.2 × 10-5, OR = 0.39) demonstrated the highest effect size, and was protective for PBC, whereas for PSC respective SNPs were: rs13191240 in the intron of ADGRB3 gene (P = 0.0095, OR = 0.2) and rs3822659 (P = 0.0051, OR = 0.236) along with rs9686714 (P = 0.00077, OR = 0.2), both located in the WWC1 gene. Conclusions Our cost-effective GWAS approach followed by individual genotyping confirmed several previously identified associations and discovered new susceptibility loci associated with PBC and/or PSC in Polish patients. However, further functional studies are warranted to understand the roles of these newly identified variants in the development of the two disorders. Electronic supplementary material The online version of this article (doi:10.1186/s12920-016-0239-9) contains supplementary material, which is available to authorized users.