Elevated blood cholesterol levels are a major risk factor for heart attack, the leading cause of death in developed countries. Although drugs exist that target low-density lipoprotein cholesterol (LDL-C), the so-called “bad cholesterol,” they do not reduce LDL-C and heart disease risk in all patients. The search is on for additional treatments.
An international team of researchers reported in the Aug. 4 Nature the results of a systematic investigation into genetic contributions to cholesterol levels and heart disease. In two papers, the researchers describe 95 variations across the genome that affect blood cholesterol in multiple ethnic groups and zero in on a single variant to identify the gene responsible and the mechanism by which it influences cholesterol levels.
The link between blood cholesterol levels, heart disease and genetics is well known, but exactly which genes are involved and how they affect cholesterol levels has been difficult to uncover. These two studies expose numerous genes implicated in cholesterol metabolism, each one a potential target for therapeutic intervention.
The researchers ran a genomewide association screen of more than 100,000 individuals to look for genetic regions, or “signposts,” that were significantly associated with coronary artery disease and four key factors: LDL-C; high-density lipoprotein cholesterol (HDL-C), or “good cholesterol”; total cholesterol; and triglyceride levels. They identified 95 genetic markers significantly associated with blood lipid levels.
“These signposts were validated across different ethnic groups, showing that cholesterol is regulated in largely the same ways no matter what race you are,” said Kiran Musunuru, a postdoctoral research associate at the Center for Human Genetics Research at Massachusetts General Hospital and a co–lead author on both papers. Musunuru and colleagues also went the extra mile by validating several of the genes directly in mouse models.
A Role for Noncoding DNASurprisingly, most of the 95 signposts were located in noncoding regions of the chromosome. Though noncoding variants do not carry instructions for protein synthesis themselves, they apparently have the potential to influence observable characteristics, or phenotype, an effect few scientists have demonstrated experimentally.
In the second study, the researchers used an array of powerful genetic tools and technologies to explore whether one such noncoding variant truly caused the low LDL-C phenotype or if it simply marked a coding variant in a nearby region that was responsible for the difference.
The researchers chose a site on chromosome 1 that was strongly associated with both low LDL-C levels and decreased coronary artery disease risk. They studied it exhaustively, looking to see if it pointed to a precise genetic alteration.
The research team was excited to find that the noncoding variant itself was responsible for the observed difference in cholesterol levels. The noncoding DNA variation—a single base change—creates a transcription factor–binding site expressed in the liver, where cholesterol metabolism is centered. When the transcription factor binds to the site, it ramps up production of a gene called SORT1. The protein encoded by SORT1 directly altered LDL-C levels by modulating the secretion of very low-density lipoprotein from the liver, ultimately controlling bloodstream lipid levels and the risk of coronary artery disease.
New Targets for TherapyOne novel aspect of these studies is that they establish a roadmap for translating human genetic associations into new biologic insights. The researchers started with human population studies, in which they scanned the genome to find signposts related to cholesterol. Then, by focusing on one signpost, they found the single DNA base change that affected cholesterol levels and identified the relevant gene, SORT1. Next, they determined the mechanism by which that DNA change influenced cholesterol. Finally, the action of the SORT1 gene was validated in mouse models.
“We connect all the dots and provide evidence that this noncoding variant is causal for the low LDL-C level phenotype,” said Sekar Kathiresan, director of preventive cardiology at MGH and co–senior author on both papers. “Importantly, this study also provides concrete evidence that a noncoding variant can have a clinical effect.”
Together, these two papers show that there are a lot of spots in the genome that affect cholesterol, many of them previously unappreciated. Now that Kathiresan, Musunuru and colleagues have thoroughly examined one of the signposts associated with cholesterol, they say it is time to move on to the other 94. “Deeper investigation, along the model shown in our second paper, at the other 94 loci will yield more information on how lipid levels in the blood are regulated and, consequently, expose new targets for therapies,” said Kathiresan.
The therapeutic implications of this research include searching for ways to increase SORT1 activity in the liver to decrease levels of LDL-C in the bloodstream, Kathiresan said. He also pointed out that there are no proven therapies that raise HDL-C and thereby lower risk for coronary artery disease. “This work should expose additional genes that can be manipulated to lower bad cholesterol or raise good cholesterol and, in turn, reduce heart attacks,” he said.
For more information, students may contact Sekar Kathiresan at skathiresan@partners.org.
Conflict Disclosure: The authors declare no conflicts of interest.
Funding Source: The National Institutes of Health
Disclaimer: The researchers are unable to provide treatment recommendations for individual cases.
