Throughout the course of human evolution, wide fluctuations in food supply promoted the selection of DNA sequences that optimize the extraction, storage, and utilization of dietary nutrients. With the advent of the industrial age, agricultural practices changed rapidly from subsistence farming to massive-scale food production. Our genome has had insufficient time to adapt to the dietary abundance and physical inactivity of the modern era. Consequently, the increased intake of dietary cholesterol and saturated fats has had a profound effect on human health. Diseases of dietary excess (e.g., diabetes and atherosclerosis) rather than insufficiency (malnutrition and infections) are the major cause of death and disability in the Western world. My research program is focused on the identification of genetic factors that contribute to (and protect from) diseases of dietary excess, especially heart and liver disease.
Multiple Factors Contribute to Heart Disease
Atherosclerosis is a complex and heterogeneous disorder resulting from the interplay of genetic susceptibilities and environmental challenges. To better define the factors that contribute to coronary heart disease (CHD), we established a population-based cohort, the Dallas Heart Study, in which more than 3,500 individuals (50 percent African American) were characterized with respect to behavioral, environmental, metabolic, and genetic risk factors for CHD. Imaging studies were performed to quantify atherosclerotic burden, heart size, heart function, and body fat distribution. Genomic DNA was obtained from each subject, and a comprehensive panel of blood and urinary analytes was measured. The extensive phenotypic database generated from this study is being used to identify new factors that will enhance our ability to predict who will get heart disease. To identify factors that promote the progression of metabolic and cardiovascular disease, we invited the study participants to return for a follow-up clinical assessment, which was completed in 2010.
Cholesterol and Heart Attacks
The most important risk factor for CHD is the plasma level of cholesterol. Cholesterol is transported in the blood in lipid-protein complexes called lipoproteins. Low-density lipoprotein (LDL) is the major cholesterol-carrying lipoprotein, and the incidence of heart disease is directly related to the levels of LDL-cholesterol in the blood. A major focus of our laboratory has been to identify genes that contribute to differences in plasma levels of LDL-cholesterol among individuals, with a goal of identifying new therapeutic approaches to prevent CHD.
Initially, we focused on rare diseases that cause very high plasma levels of LDL-cholesterol. We showed that mutations in genes required for the clearance of circulating LDL (ARH or LDLRAP) and for the excretion of cholesterol from the body (ABCG5 and ABCG8) both cause hypercholesterolemia and premature CHD. These studies confirmed that a high plasma level of LDL-cholesterol, irrespective of the underlying cause, is sufficient for the development of CHD.
A Genetic Cause of Low Plasma Levels of LDL-Cholesterol
A high plasma level of cholesterol is sufficient for the development of CHD but is it necessary? Can we protect people from heart disease by lowering their cholesterol even if they have other risk factors? To address these questions, we set out to identify sequence variations that confer low plasma levels of LDL-cholesterol. DNA sequencing of Dallas Heart Study participants revealed that mutations that disrupt the function of a circulating protein, proprotein convertase subtilisin/kexin type 9 (PCSK9), result in lower plasma levels of LDL. Surprisingly, 1 of every 50 African Americans and 1 of every 30 Caucasians has an inactivating mutation in PCSK9.
Next, we showed that the rate of CHD was more markedly reduced over a 15-year period in those individuals with inactivating mutations in PCSK9 than was seen in clinical trials that used statins to lower cholesterol. PCSK9 carriers have a lower blood cholesterol level for a lifetime, whereas most statin users do not have reduced LDL-C levels until they are middle-aged, after the disease is already established. Our findings, taken together with other studies, suggest that initiating lipid-lowering treatment earlier in life, either by changing dietary composition or by taking low doses of cholesterol-reducing medications, is the optimal strategy to prevent development of heart disease.
By screening the families of participants heterozygous for a loss-of-function mutation in PCSK9, we identified a 34-year-old woman with no functional PCSK9 and an LDL-C of only 14 mg/dL. These findings confirmed the central role that PCSK9 plays in LDL metabolism in humans and suggested that PCSK9 would be a safe target for cholesterol-lowering therapy. Jay Horton (University of Texas Southwestern Medical Center at Dallas) showed that PCSK9 binds to the extracellular domain of the LDL receptor on the surface of liver cells. We mapped the site of binding of PCSK9 to the LDL receptor and several pharmaceutical companies have now developed antibodies that interfere with this binding. Recent testing of these antibodies in humans has shown a dramatic reduction of plasma LDL in the blood. This finding is important because approximately half of the individuals currently taking cholesterol-lowering drugs do not achieve the target LDL-C level. Inactivation of PCSK9 provides a new therapeutic approach for successful cholesterol-lowering therapy.
A Common Genetic Variation Increases Heart Attack Risk
We compared the genomes of individuals with early symptomatic CHD to those of older asymptomatic subjects and identified a highly reproducible association between a 58-kilobase interval on chromosome 9 and CHD. Individuals with two copies of the "risk" allele have a 40 percent increase in CHD that is not explained by any of the known risk factors. Other investigators have shown that genetic variation in the same genomic interval also predisposes to aortic and cerebral aneurysms. The implicated sequences are in a region of the genome that does not code for any proteins but does encode a long noncoding RNA. We are probing the mechanistic link between the risk allele and CHD.
Rare Genetic Defects Cumulatively Contribute to Complex Disorders, Such as Metabolic Syndrome
The obesity epidemic has resulted in a dramatic increase in the prevalence of metabolic risk factors for diabetes and heart disease, including insulin resistance, high plasma levels of triglycerides, low plasma levels of high-density lipoprotein cholesterol (HDL-C), fatty liver disease, and hypertension. We have taken two different approaches to identify genetic variants that contribute to metabolic risk for heart disease. First, we compared the number of mutations in selected genes in individuals at the extremes of the distribution of various metabolic traits, including plasma levels of LDL-cholesterol and HDL-cholesterol. For several genes, we found an excess of rare and low-frequency loss-of-function mutations that contribute to these traits, and we have shown that a significant proportion of the general population has mutations that result in a complete loss of gene function. Thus, mutations with large effects on protein function are not rare in the general population, and these mutations collectively contribute significantly to complex traits. Characterization of individuals with such mutations provides the opportunity to assess directly the role of a gene in humans. Second, we sequenced candidate genes in all the participants of the Dallas Heart Study and tested for association with various metabolic traits. We found a plethora of mutations in genes encoding three proteins of the angiopoeitin family that are associated with reduced plasma levels of triglycerides (ANGPTL3, ANGPTL4, and ANGPTL5). Antibodies to these proteins are now being developed as potential therapeutics for the treatment of severe hypertriglyceridemia.
Gene Defect in Fatty Liver Disease
Previously, we showed that one-third of the subjects in the Dallas Heart Study had an excess of triglyceride stored in lipid droplets in the cytoplasm of hepatocytes (hepatic steatosis). This condition was most prevalent in Hispanics and least prevalent in African Americans. To identify genes that contribute to hepatic steatosis, we performed an unbiased screen of the genome for variants that change the amino acid sequences of proteins and tested for association between these variants and hepatic fat content. This analysis revealed a variant in a member of the patatin-like phospholipase domaincontaining family of proteins (PNPLA), PNPLA3, that was associated with hepatic fat content. The variant is most common in Hispanics and is not associated with any of the factors known to promote hepatic steatosis (e.g., body weight, insulin sensitivity, and plasma lipid levels). The mutation associated with elevated liver fat content is also associated with liver disease (hepatitis and cirrhosis).
PNPLA3 has a domain at its N terminus that resembles patatin, an enzyme with acylhydrolase activity that was originally discovered in the potato tuber. Structural modeling of PNPLA3 suggests that the mutation associated with hepatic steatosis masks the active site of the enzyme. We have purified the recombinant protein and shown that the variant associated with hepatic steatosis impairs the ability of the enzyme to break down triglycerides. We are now focused on determining how the variant promotes the development of hepatitis and cirrhosis in individuals with hepatic steatosis.
Determining the DNA sequence variations that confer susceptibility to metabolic and cardiovascular disease will enhance our understanding of the underlying processes that contribute to these diseases in the population, providing the opportunity to identify new treatment targets, diagnostic tests, and therapeutic interventions for patients with metabolic risk factors that contribute to heart disease.
Grants from the National Heart, Lung, and Blood Institute provide partial support for these projects.
As of May 30, 2012