The relationship between genetically-driven liver fat and coronary heart disease (CHD) remains unclear. ApoB-containing lipoproteins are known causal factors for CHD and may explain this relationship.

We conducted a genome-wide association study (GWAS) in the UK Biobank to identify genetic variants associated with liver fat. We then investigated the effects that these genetic variants had on both apoB-containing lipoproteins and CHD. Using Mendelian Randomization (MR) analyses, we examined if the relationship between genetically-driven liver fat and CHD could be attributed to its effect on apoB-containing lipoproteins. We found 25 independent liver-fat associated single-nucleotide polymorphisms (SNPs) with differing effects on lipoprotein metabolism. The SNPs were classified into three groups/clusters. The first cluster (N = 3 SNPs) displayed lipoprotein-raising effects. The second cluster (N = 12 SNPs) displayed neutral effects on lipoproteins and the third cluster (N = 10 SNPs) displayed lipoprotein-lowering effects. For every 1% higher liver fat, the first cluster showed an increased risk of CHD (OR = 1.157 [95% CI: 1.108–1.208]). The second cluster showed a non-significant effect on CHD (OR = 0.988 [95% CI: 0.965–1.012], whereas the third cluster showed a protective effect of increased liver fat on CHD (OR = 0.942 [95% CI: 0.897–0.989]). When adjusting for apoB, the risk for CHD became null.

Here, we identify 25 liver-fat associated SNPs. We find that SNPs that increase, decrease or have neutral effects on apoB-containing lipoproteins show increased, decreased or neutral effects on CHD respectively. Therefore, the relationship between genetically-driven liver fat and CHD is mediated by the causal effect of apoB.