Genome-wide association of serum bilirubin levels in Korean population

Abstract

A large-scale, genome-wide association study was performed to identify genetic variations influencing serum bilirubin levels using 8841 Korean individuals. Significant associations were observed at UGT1A1 (rs11891311, P = 4.78 × 10 ⁻¹⁴⁸ ) and SLCO1B3 (rs2417940, P = 1.03 × 10 ⁻¹⁷ ), which are two previously identified loci. The two single-nucleotide polymorphisms (SNPs) were replicated (rs11891311, P = 3.18 × 10 ⁻¹⁵ ) or marginally significant (rs2417940, P = 8.56 × 10 ⁻⁴ ) in an independent cohort of 1096 individuals. In a conditional analysis adjusted for the top UGT1A1 variant (rs11891311), another variant in UGT1A1 (rs4148323, P = 1.22 × 10 ⁻¹²¹ ) remained significant; this suggests that in UGT1A1 at least two independent genetic variations influence the bilirubin levels in the Korean population. The protein coding variant rs4148323, which is monomorphic in European-derived populations, may be specifically associated with serum bilirubin levels in Asians ( P = 2.56 × 10 ⁻⁷⁰ ). The SLCO1B3 variant (rs2417940, P = 1.67 × 10 ⁻¹⁸ ) remained significant in a conditional analysis for the top UGT1A1 variant. Interestingly, there were significant differences in the associated variations of SLCO1B3 between Koreans and European-derived populations. While the variant rs2417940 at intron 7 of SLCO1B3 was more significantly associated in Koreans, variants rs17680137 ( P = 0.584) and rs2117032 ( P = 2.76 × 10 ⁻⁵ ), two of the top-ranked SNPs in European-derived populations, did not reach the genome-wide significance level. Also, variants in SLCO1B1 did not reach genome-wide significance in Koreans. Our result supports the idea that there are considerable ethnic differences in genetic association of bilirubin levels between Koreans and European-derived populations.