Genetic Epidemiology of Coronary Heart Disease

BACKGROUND:

The Rochester Family Heart Study was initiated in 1983 to study the genetic basis for familial aggregation of apolipoprotein A-I and hypertension in Rochester, Minnesota. Recruitment began in 1985 and resulted in data on 2,259 individuals in 300 three generation pedigrees. The study was supported by R01HL24489.

Lipid levels play a central role in determining the predisposition to coronary heart disease. In Western industrialized societies diseases of the heart rank as the number one cause of morbidity and mortality in the adult population. Thirty percent of all deaths in the United States in 1982 were attributable to coronary heart disease. Epidemiological studies of total communities including Framingham, Massachusetts and Tecumseh, Michigan have established a long list of factors that predict increased risk to coronary heart disease. Those that have been implicated include male gender, aging, positive family history, sedentary life style, diabetes, high blood pressure, cigarette smoking, overweight and abnormal plasma levels of lipids and lipoproteins. Although there is great variation among individuals in the risk factors associated with the development of coronary heart disease, the disease typically involves the infiltration of cholesterol into the intimal wall of the arteries beginning early in life. It has been hypothesized that each individual inherits or experiences an array of risk factors that determine susceptibility to the disease process that progresses through the phases of injury to the vascular wall, infiltration of lipids, accumulation of cholesterol esters, a proliferation of cells, stenosis and finally occlusion that leads to the coronary heart disease outcome. Lipid metabolism plays a central role in determining the progression of this process. A role for elevated lipids in the etiology of coronary heart disease is supported by the Lipid Research Clinics trial of cholestyramine which established that a 20 percent reduction of plasma low density lipoprotein (LDL) in particular, results in a 19 percent reduction of coronary heart disease after seven years in high risk males. Most studies to date have focused on environmental factors associated with variability in risk factors. Biometrical studies have estimated the role of genes in determining interindividual variability in lipid and lipoprotein phenotypes by comparing and contrasting the levels in related and unrelated individuals living together and apart. With the advent of new techniques to measure variability in gene products and at the DNA level, it is now possible to expand our studies of the role of genes in both lipid metabolism and coronary heart disease. Knowledge about the genetic epidemiology of coronary heart disease will provide valuable tools for clinical screening and diagnosis and will make it possible to develop a rational strategy to intervene. Not all individuals respond in the same way to lipid-lowering strategies. By knowing the genes involved in regulating the metabolic factors that are altered in coronary heart disease we will be able to sort out those individuals who will respond to specific treatments.

DESIGN NARRATIVE:

Beginning in 1987 and using blood samples and data from the RFHS, variation at the DNA level was determined using restriction fragment length polymorphisms and at the protein level by isoelectric focusing. The relationships were defined in the population among the apolipoproteins, cholesterol, triglycerides, high density lipoprotein cholesterol phenotypes and estimates were done on the effects of age, sex, smoking, drug use, and obesity on these relationships. Estimates were determined for the fraction of interindividual variation in the apos attributable to gene loci coding for these molecules. Estimates were also determined for the fraction of interindividual variation in cholesterol, triglycerides, and high density lipoprotein cholesterol attributable to gene loci coding for the apos and the LDL receptor. A determination was also made as to whether genotypic variation for the gene loci coding for the apo and LDL-receptor contributed to the prediction of coronary heart disease beyond that provided by the effects of these gene loci on the interindividual variation in the apos, cholesterol, triglycerides, and high density lipoprotein cholesterol. Radioimmunoassay and medical information collection took place at the Mayo Clinic. DNA probes, DNA studies of special families, and definition of new restriction fragment length polymorphism took place at Charing Cross Hospital in London, England. RFLP typing of genes coding for the apolipoproteins and the LDL receptor and electrophoretic typing of the known apo E and AIV protein polymorphisms took place at the University of Pittsburgh. The genetic architecture responsible for phenotypic variability in the population at large was defined at the University of Michigan and Washington University, St. Louis.

The study was renewed in 1992 to carry out six aims that 1) provided new information about four polymorphic genes involved in reverse cholesterol transport; 2) completed the investigators' studies to evaluate the utility of genetic variation in 12 candidate genes for prediction of coronary artery disease; 3) identified individuals who carry mutations that have an impact on coronary artery disease; and 4) began molecular studies of these individuals to define the DNA changes that are responsible. New noninvasive measures of coronary artery disease in asymptomatic individuals using ultrafast cardiac computed tomography complemented efforts to evaluate the role of measured genetic variation in predicting coronary artery disease.

The study was renewed again in 1997 with five specific aims. Aims 1 and 2 identified genomic regions containing new coronary artery disease susceptibility genes, using 234 tandem repeat marker loci at a 10 centimorgan (cM) density and the variance component method of linkage in 274 multigeneration pedigrees, both globally and in particular contexts defined by gender, age, body mass index and smoking. For those regions that showed linkage, confirmatory linkage analysis was conducted in a second independent sample of 301 multigeneration pedigrees. In regions where linkage was replicated, fine structure linkage analyses was conducted, using additional marker loci, located approximately every 1 cM. Specific aims 3 and 4 identified functional DNA variations in the new candidate coronary artery disease susceptibility genes. Direct DNA sequencing and cladistic analyses were used to estimate the average, context dependent, dominant, and epistatic effects associated with these functional DNA variations. Specific aim 5 estimated the association between the risk of coronary artery disease and functional genotypes identified by the cladistic analyses carried out in specific aims 3 and 4 before and after considering the intermediate biological risk factors that link the effects of genetic and environmental variation to variation in risk of disease.

The study was renewed in 2001 through July 2006 to continue to investigate innovative analytic and molecular aspects of risk of coronary artery disease, in particular, the contribution of polymorphic variation at the ApoE locus and two other candidate regions of the genome on variation in plasma ApoE level.