Genetic predisposition linked to coronary calcification

Issue: October 2012

ByRobert Roberts, MD, FRSC, FRCPC, MACC, LLD (Hon.), FAHA

ByBrandon A. Roberts, MSc, MD

ByCalvin W. Roberts, MD

ByWilliam L. Roberts, MHA, CPA

In an international consortium of 14 genome-wide association studies referred to as the Coronary Artery Disease Genome-wide Replication and Meta-Analysis, or CARDIoGRAM, 23 genetic risk variants were validated as having an association with coronary artery disease.

These results and those from many other studies have confirmed that CAD has a significant genetic predisposition. It is also well established that a relationship exists between the presence of coronary artery calcium and atherosclerosis. CT has become a reliable means of detecting coronary artery calcium. Data from family and epidemiological studies suggest that coronary calcification, similar to CAD, has genetic predisposition.

Risk variants for coronary calcification

Robert Roberts

In the CARDIoGRAM study by O’Donnell and colleagues, a genome-wide association study (GWAS) was performed to determine whether coronary calcification has a genetic predisposition. In GWAS, one simply genotypes the DNA sample from each individual case and control with 1 million single-nucleotide polymorphisms (SNPs) selected to span the human genome. Any SNP occurring more frequently in the cases than controls is considered to be in a DNA region that increases risk for the disease. However, since one is using 1 million endpoints, rather than a single endpoint, there is the potential for many false-positives, so the conventional P value of .05 must be corrected; by convention, this has been a Bonferroni correction, which gives a P value of 5 x 10^-8, commonly known as genome-wide significance. For the study, coronary artery calcium was detected by CT scanner and the Agatston scoring method was used to determine the extent of calcification.

The discovery population consisted of five cohorts of cases and controls with a total sample size of 9,961 evenly distributed between men and women (mean ages ranged from 52 to 76 years). Mean calcification scores ranged from 131 to 386. Maximum calcification scores ranged from 4,867 to 12,611. The genotyping was performed with either 500,000 or 1 million SNPs, using a variety of platforms, including Illumina and Affymetrix. The statistical threshold for an association between a SNP and that of coronary artery calcium was selected at genome-wide significance of 5 x 10^-8.

Results of the initial discovery population showed that the well-known 9p21 locus and that of 6p24 were associated with coronary calcification. These findings were confirmed in an independent population of 6,032 people. Five other known risk variants for CAD showed significant association with coronary artery calcium; however, the association was less than genome-wide significant. To determine whether the association also occurred in those with MI, researchers analyzed 1,150 SNPs showing moderate association with coronary calcification (P=5 x 10^-3) for replication in a population of 34,508 cases with coronary artery calcium (6,811 had MI). Again, 9p21 and 6p24 showed a genome-wide significant association with coronary calcification. Several other genetic risk variants also showed an association with coronary calcification but were not of genome-wide significance. Of note, all 23 genetic risk variants validated to be associated with CAD were genotyped in the population with coronary calcification and only 9p21 and 6p24 showed genome-wide significance.

Investigating genetic predisposition

The function of the 9p21 region, despite being the first CAD risk variant to be identified, remains unknown. It is in a region in which there is no gene encoding for protein; instead, the 9p21 risk region encodes for a long noncoding RNA of unknown function. Nevertheless, in all of the studies, 9p21 is highly associated with CAD and occurs in 75% of the world’s population, except for Africans. We, and other researchers, have shown that 9p21 is associated with atherosclerosis, but not with MI. This finding has been consistently confirmed across many populations and countries, indicating that the 9p21 defect is associated with atherosclerosis.

There is conclusive evidence to show that 9p21 is not associated with thrombosis or plaque rupture, the accepted underlying mechanisms precipitating MI. It is also confirmed that the site of action of 9p21 is at the vessel wall. The locus at 6p24 is in a region associated with a known protein coding gene; namely that of Phactr1, which encodes for the protein phosphatase-1. Phosphatase-1 is known to regulate many cellular processes through dephosphorylation of several different substrates. The exact mechanism whereby phosphatase-1 involves calcification remains unknown.

These findings indicate that calcium is part of the process leading to atherosclerosis and has a genetic predisposition. Although only two risk variants showed genome-wide significance (9p21 and 6p24), others with less significance (CXCL12, SORT1, MRAS, COL4AI/COL4A2 and ADAMTS7) may provide genetic predisposition and may be genome-wide significant in larger samples.

Since only certain genetic risk factors contribute to coronary calcification, it may ultimately provide a clue to the molecular mechanism contributing to pathogenesis of atherosclerosis. The formation of calcium does not appear to be pivotal to the development of atherosclerosis because one can have atherosclerosis and MI without calcification. Nevertheless, the mechanism whereby calcium contributes to the pathogenesis and whether it could be a target for drug therapy remains to be determined. Since calcium formation is age-dependent and occurs in several other organs, it is likely to be a byproduct rather than fundamental to the pathogenesis.

Robert Roberts, MD, is president and CEO of the University of Ottawa Heart Institute and director of the Ruddy Canadian Cardiovascular Genetics Centre at the University of Ottawa Heart Institute. He is also a member of the Cardiology Today Editorial Board. Brandon A. Roberts, MD, MSc, is a resident in medicine and the critical care unit at Louisiana State University in New Orleans.

For more information:

O’Donnell CJ. Circulation. 2011;124:2855-2864.
Roberts R. Clin Chem. 2012;58:104-112.
Schunkert H. Nat Genet. 2011;43:333-338.

Disclosures:

Roberts and Roberts report no relevant financial disclosures.

Robert Roberts, MD, can be reached at the Ruddy Canadian Cardiovascular Genetics Center, 40 Ruskin St., Suite H-2404, Ottawa, Ontario, Canada K1Y 4W7; his email is rroberts@ottawaheart.ca.