NEW YORK (GenomeWeb) – In an illustration of the benefits of large-scale population sequencing, researchers involved with the National Heart, Lung, and Blood Institute's Exome Sequencing Project, along with researches from the UK10K project to sequence 10,000 individuals, have identified a rare variant associated with lower blood levels of triglycerides.
The NHLBI group reported its findings in June in the New England Journal of Medicine, while the UK10K group reported its results this week in Nature Communications.
Nicole Soranza, a group leader at the Wellcome Trust Sanger Institute, which led the sequencing for the UK10K project, said that the Nature Communications publication is the first report of a disease-related variant from the UK10K Project. "It's an early report of a much bigger project," she said. "We have now analyzed 64 cardio-metabolic traits, including lipid subfractions, obesity measures, blood pressure, blood function, liver function, and renal function." The members of the UK10K Project are in the process of preparing the results of the project for publication, she said.
The UK10K Project kicked off in 2010 and involved whole-genome sequencing of 4,000 individuals from two different cohorts — a group known as the ALSPAC cohort, which includes families with a vast amount of clinical, genetic, and environmental information; and the TwinsUK cohort, which includes individuals entered into a twin registry. In addition, researchers have sequenced 6,000 exomes from individuals who have a neurodevelopmental disorder, are obese, or have a rare disease.
In the recent studies, the UK10K and NHLBI groups both sequenced thousands of individuals, but used different strategies. The UK10K group used a low-pass whole-genome sequencing approach on 3,202 individuals for whom triglyceride measurements were also available, while the NHLBI group used deep exome sequencing, about 89x coverage, on 3,734 individuals. The two groups found variants in the APOC3 gene that were associated with lower levels of blood triglycerides. The UK10K group confirmed the finding in an additional 12,831 individuals from Northern and Southern European populations, while the NHLBI group went one step further and demonstrated in 110,970 individuals that the variants not only impacted blood triglyceride levels, but that they are also associated with a lower risk of coronary heart disease.
Sekar Kathiresan from the Cardiovascular Research Center and Center for Human Genetic Research at Massachusetts General Hospital and corresponding author of the NEJM study, told Clinical Sequencing News that the studies demonstrate that "the sample sizes needed to show relationship to disease are considerably larger than what people originally suspected." He noted a third study published in the same NEJM issue that sequenced just the APOC3 gene, but in over 10,000 individuals, and similarly found that loss-of-function variants in the gene were associated with lower risk of heart disease.
In the NHLBI's NEJM study, the researchers found four APOC3 mutations, including three loss-of-function mutations and one missense mutation, and found that the variant was present in one out of 150 people. They also found that carriers had a 40 percent lower risk of coronary heart disease among non-carriers.
The finding that individuals with loss-of-function variants in APOC3 had lower levels of triglycerides was not surprising, since the association of that gene with lipids has been known for some time, Kathiresan said. However, the association between long-term lower levels of blood lipids caused by APOC3 loss of function and coronary artery disease had not previously been demonstrated, he said, and could have a significant clinical impact.
Already, he said, pharmaceutical companies are working on compounds that would inhibit the APOC3 gene. For example, Isis Pharmaceuticals is developing a compound to inhibit apoCIII protein production that it plans to test in a phase III trial this year in patients with severely high triglyceride levels.
Soranzo told CSN that the finding would also help give better risk predictions. "It's an example of how having a better evaluation and understanding of low-frequency variants … can begin to help us in the transition to really understanding the impact on a personalized level, understanding what is an individual's risk."
The UK10K group identified the same splice site loss-of-function variant as the NHLBI group, finding that it had a minor allele frequency of 0.25 percent.
Soranzo said that one key to identifying the APOC3 variant was the large number of genomes evaluated. "That's a step change from previous studies," she said. "We now have a full repertoire of sequence variation, down to variants at very low minor allele frequencies." Sequencing the whole genomes of 4,000 individuals "allows us to look at variants below 1 percent frequency, which are unrepresented in other studies."
Whether deep exome sequencing or low-pass whole-genome sequencing is a better method for such population studies is a current debate in the community, Kathiresan said. One problem with low-pass whole-genome sequencing, he said, is that it will have trouble identifying really rare variants present in only a few individuals. The APOC3 mutation "is rare, but it's not super rare," he said.
Soranzo noted that the benefit of whole-genome sequencing is that it will pick up variants outside of the protein coding regions and also pick up variants in the exome that are poorly covered by exome capture.
However, Kathiresan noted that the debate may soon be moot as population sequencing studies will increasingly move to deeper whole-genome sequencing as costs continue to come down.