A polymorphism in the Irisin-encoding gene (FNDC5) associates with hepatic steatosis by differential miRNA binding to the 3′UTR

Background & Aims: Irisin, the cleaved extra-cellular fragment of the Fibronectin type III domain-containing protein 5 (FNDC5) is a myokine that is proposed to have favorable metabolic activity. We aimed to elucidate the currently undefined role of variants in the FNDC5 gene in non-alcoholic fatty liver disease (NAFLD). Methods: We prioritized single nucleotide polymorphisms in FNDC5 on the basis of their putative biological function and identified rs3480 in the 3′ untranslated region (3′UTR). We studied the association of rs3480 with liver disease severity and the metabolic profile of 987 Caucasian patients with NAFLD. Functional investigations were undertaken using luciferase reporter assays of the 3′UTR of human FNDC5, pyrosequencing for allele-specific expression of FNDC5 in liver, measurement of serum irisin, and bioinformatics analysis. Results: The rs3480 (G) allele was associated with advanced steatosis (OR 1.29; 95% CI 1.08–1.55; p = 0.004), but not with other histological features. This effect was independent but additive to PNPLA3 and TM6SF2. The rs3480 polymorphism influenced FNDC5 mRNA stability and the binding of miR-135a-5P. Compared with controls, hepatic expression of this microRNA was upregulated while FNDC5 expression was downregulated. Elevated serum irisin was associated with reduced steatosis, and an improved metabolic profile. Conclusions: Carriage of the FNDC5 rs3480 minor (G) allele is associated with more severe steatosis in NAFLD through a microRNA-mediated mechanism controlling FNDC5 mRNA stability. Irisin is likely to have a favorable metabolic impact on NAFLD. Lay summary: Irisin is a novel protein produced mainly by muscle, which is known to be released into the circulation, with an unclear role in liver fat deposition. This study demonstrates that genetic variants in the gene encoding the irisin protein modulate the risk of liver fat in patients with fatty liver disease. Interestingly, these effects are independent of, but additive to those of other recently described genetic variants that contribute to liver fat. In functional studies, we have deciphered the detailed molecular mechanisms by which this genetic variant mediates its effects.