NEW YORK (GenomeWeb) – A pair of studies has linked additional loci and ancestry-specific variants to plasma lipid levels.
Genome-wide association studies have linked 175 genetic loci to plasma lipid levels in people, and two studies appearing today in Nature Genetics have now built upon that work. In one, an international team of researchers conducted an exome-wide association study of more than 300,000 individuals — they identified 250 loci linked to lipid levels and examined the relationship between these variants and coronary artery disease, age-related macular degeneration, and type 2 diabetes. The other study examined lipid level-linked variants in more than 47,000 East Asians and found novel as well as population-specific variants.
In the first paper, the Broad Institute's Sekar Kathiresan and his colleagues tested the association of genotypes from the Human Exome BeadChip array with lipid levels in 73 studies, encompassing more than 300,000 participants.
From this, they identified 444 variants that were associated with total cholesterol, high-density-lipoprotein cholesterol, low-density-lipoprotein cholesterol, or triglyceride levels. They traced these variants to 75 novel loci and 175 loci that had previously been linked to lipid traits.
They selected two coding variants at JAK2 and A1CF for functional analysis. In mice, they found that the JAK2 p.Val617Phe variant led to lower plasma total cholesterol levels, as it does in people, while the A1CF p.Gly398Ser variant led to increased triglyceride and total cholesterol levels as well as to a higher risk of coronary artery disease.
Kathiresan and his colleagues also examined links between the 444 variants they uncovered and other conditions.
For instance, they noticed that a rare null mutation in HBB, which is linked to beta-thalassemia, was strongly associated with lower total cholesterol. An additional analysis of 31,156 coronary artery disease cases and 65,787 controls found that people carrying loss-of-function HBB variants are protected against coronary artery disease.
Variants within the CETP gene, meanwhile, were linked to higher LDL-C levels and were correlated with increased risk of age-related macular degeneration.
The researchers also looked into whether lipid-modifying medications that lower LDL-C levels would always increase type 2 diabetes risk, as has been reported. They confirmed that the PCSK9 p.Arg46Leu variant, which is linked to lower LDL-C levels, is, at the same time, associated with an increased risk of type 2 diabetes. Additionally, among 113 variants associated with variations in LDL-C levels, the researchers noted an inverse effect between their influence on LDL-C levels and diabetes risk.
In the second paper, the University of Michigan's Cristen Willer and her colleagues used a custom-content exome array to study protein-coding variants associated with blood lipid levels among 47,532 individuals of East Asian ancestry. The researchers found 255 variants at 41 loci that reached significance, including three novel loci and 14 variants specific to East Asian populations. The three novel loci specific to East Asians were in the ACVR1C, MCU, and CD163 genes.
For 38 of the loci previously linked to lipid levels, the researchers identified 31 protein-altering variants linked to 25 functional lipid genes. Coding variants, the researchers noted, were more likely to have consistent effect sizes across ancestries than non-coding variants.
To examine whether the novel and potentially functional variants affected coronary artery disease risk, the researchers conducted an association study in 28,889 Chinese individuals with and without the condition as well as about 185,000 cases and controls from the CADIoGRAMplusC4D dataset.
They found that the variant near MCU that was associated with lower LDL-C levels was also linked to reduced CAD risk among Chinese individuals. The other variants exhibited a consistent direction of effect between lipid traits and disease risk, they noted.
By folding in data from the Global Lipids Genetics Consortium on more than 300,000 people, mostly of European ancestry, Willer and her colleagues identified an additional nine variants associated with lipid traits. All these variants were common and they included four coding variants, in FAM114A2, MGAT1, ASCC3, and PLCE1.
Both the East Asian and GLGC datasets homed in on the same nine functional genes, the researchers reported. However, different variants were associated in each ancestry group. For instance, the variants uncovered at PCSK9, CD36, ABCA1, CETP, PMFBP1, LIPG, LDLR, and PPARA in the GLGC dataset exhibited lower minor allele frequencies in the East Asian samples. At the same time, PCSK9, APOB, CD36, CETP, LDLR, and PPARA variants identified in East Asian samples had lower MAFs in the GLGC samples.
"Overall, our findings demonstrate that rare and low-frequency coding variants are more likely to be population specific, which underscores the value of discovering ancestry-specific rare variants in diverse populations, particularly for low-frequency variations," the researchers wrote.