|Home | About | Journals | Submit | Contact Us | Français|
Epigenomic mechanisms, such as DNA methylation and histone modifications, represent a critical link between genomic coding and phenotype expression that is influenced by both underlying genetic and environmental factors. Animal models indicate that epigenomic mechanisms have fundamental roles in the development of inflammation, obesity, diabetes, atherosclerotic lesions, and cardiovascular disease. Aberrant DNA methylation and histone modifications have been found in pathology investigations of atherosclerotic lesions. In the present article, we propose an overview of the major milestones in epigenetics, as well as of recent development and technologies for epigenomic research. We review recent findings demonstrating the links of individual characteristics, risk factors, and biomarkers with epigenetic signatures. We present available evidence from human studies that have investigated the relation of epigenetic mechanisms with subclinical and clinical cardiovascular disease. Also, we provide a list of online resources that can help retrieve additional information and facilitate further work in epigenomics. Finally, we discuss open questions for future research in cardiovascular epigenomics, including issues related to the analysis of surrogate and cardiovascular tissues, conduction and analysis of genome-wide epigenetic studies, integration of genomic and transcriptomic data in epigenomic investigations, and perspectives for epigenomics in preventive cardiovascular medicine and pharmacology.
Epigenetics is the study of heritable alterations in phenotypes and gene expression that occur without changes in the DNA sequence. Epigenetic mechanisms are flexible genomic factors that can change genome function under exogenous influence, but also provide a molecular substrate that allows for the stable propagation of gene expression states from one generation of cells to the next.
The best understood of the epigenetic mechanisms is DNA methylation,1 a covalent modification that, in mammals, occurs predominantly at cytosines followed by guanines (i.e., CpG dinucleotides) to form 5-methylcytosines. CpG methylation generally is associated with gene silencing. DNA-methylation states may vary over an individual’s lifetime and have been shown to regulate biological processes underlying cardiovascular disease, such as atherosclerosis, inflammation, hypertension, and diabetes. Methylation within gene promoters and CpG-dense sequences (CpG Islands, usually unmethylated) has long been held to have the highest functional relevance to gene expression control (Figure 1). However, recent work has identified regions up to 2 kb from the islands (i.e., on the island edges), termed CpG island shores, that have variable methylation and are the most enriched with functional CpG sites.2 DNA methylation also is believed to play important roles in the maintenance of genome integrity by transcriptional silencing of retrotransposons, i.e. retrovirus-like DNA sequences that can duplicate and transpose themselves across the genome. In particular, some retrotransposons, such as Alu and Long-Interspersed Nucleotide Elements-1 (LINE-1) have up to 1-million copies in the human genome, are heavily methylated, and have been extensively investigated in human epigenetics studies. DNA methylation is mitotically stable and can be transmitted through cell division and, in some loci identified in animal experiments, transgenerationally. Histones are globular proteins that can be post-translationally modified at specific amino-acid residues. Histone 3 (H3) and Histone 4 (H4) can be covalently modified by acetylation, methylation, ubiquitination, phosphorylation, sumoylation, citrullination, and ADP-ribosylation. These modifications alter histone interactions with DNA and other nuclear proteins. Consequently, histone modifications induce either repression or activation of transcription, depending on the type of modification and on the position of the modified amino-acid residue.
The conceptual foundations of epigenetics were laid out in the early 1940s by Sir Conrad Waddington, who first defined epigenetics as the “the branch of biology which studies the causal interactions between genes and their products, which bring the phenotype into being”. Just a few years later DNA methylation was first described in bacterial genomes (Figure 2), but only in 1975 two groups independently hypothesized that 5-methylcytosines was duplicated through cell division and regulated mRNA expression. At that time, Allfrey et al. had already demonstrated that histone acetylation and methylation contributed to gene expression control.
The last 20 years have seen the development and diffusion of easily accessible techniques for DNA methylation and histone modification analysis. Qualitative methods such as bisulfite sequencing and methylation-specific PCR have been developed and eventually adapted to produce semi-quantitative calls. Quantitative methods, such as pyrosequencing and MALDI-TOF (Matrix assisted laser desorption/ionization-time of flight), recently have allowed for precise measurements of the copies of methylated genes/loci in a DNA sample.
Techniques for genome-scale analysis of both DNA methylation and histone modifications have become available that provide increasing degrees of genomic coverage. Currently, relatively inexpensive chip arrays are available that allow quantitative analysis of methylation at approximately 486,000 individual CpG sites. At a higher cost, whole genome tiling arrays featuring 2.1 million probes can be used to provide qualitative calls on the methylation of genes or loci, but without achieving individual methylation measures on single CpGs.
Next-generation sequencing is increasingly used for histone modification analysis, coupled with chromatin immunoprecipitation (CHIP-Seq), and is being further developed for DNA methylation analysis. Current genome-wide scale methods for epigenomic analyses are limited in that they are prone to analytical artifacts and measurement errors. Technical validation through a more accurate technology, such as pyrosequencing or MALDI-TOF, is widely regarded a necessary step to validate genome-scale findings.
Starting from the mid-1990s’, there has been a steady growth in the number of laboratories and investigators involved in epigenetic research. In the words of Andrew Feinberg,1 epigenetics has been moving at the very “epicenter of modern medicine because it can help to explain the relationship between an individual’s genetic background, the environment, aging, and disease”. All those questions that can be addressed in epigenetic studies are critical to cardiovascular research. Cardiovascular epigenetic research has been growing quasi-exponentially in recent years, albeit the increase in the number of cardiovascular publications has lagged a few years behind the general production in epigenetics (Figure 3). Only 12 epigenetic publications that involved cardiovascular disease or related issues appeared in PubMed in 1995 and that number was just 35 in 2002. The projected number for 2010 is 525, amounting to a 43-fold increase since 1995. As a benchmark, the total number of PubMed publications on cardiovascular disease or related issues increased from 50,320 to 95,953 from 1995 to 2010 (91% fold increase). In 2010, the proportion of epigenetic publications on cardiovascular disease or related issues will be 5.4%, compared to 1.6% in 1995.
Epigenomics represents a critical link between genomic coding and phenotype expression that is influenced by both underlying genetic and environmental factors (Figure 4). Available data indicate that cardiovascular risk factors might influence and remodel epigenomic patterns, and that cardiovascular biomarkers are associated with epigenetic modifications. Epigenetic modifications may contribute to subclinical and clinical cardiovascular disease. Epigenomics is inherently interconnected with genetics, because epigenetic modifications can alter the expression of genetic variations, as well as genetic variation is one of the determinants of DNA methylation and histone modifications. In the online supplement we report an extended list of epigenetic publications on cardiovascular risk factors, biomarkers, subclinical or clinical cardiovascular disease (Table S1).
Experimental animal models have demonstrated that DNA methylation has critical roles in the development of atherosclerosis and cardiovascular disease. Mice deficient in genes coding for methylation enzymes, such as DNMTs (that establish or replicate DNA methylation) or MTHFR (related with methyl donor generation), show hypomethylation of their DNA. Increased expression of inflammatory mediators is found in leukocytes from DNMT-deficient mice.3 In MTHFR-deficient mice, DNA hypomethylation has been shown to precede the formation of aortic fatty streaks.4 Also, atherosclerosis-prone Apoe−/ − mice develop specific changes in DNA methylation of transcribed gene sequences, as well as in inter-dispersed elements, both in peripheral blood leukocytes and in the aorta prior to developing vascular lesions.5 In Apoe−/ −mice, changes in DNA methylation in peripheral blood leukocytes have been shown to contribute to the dysregulation of inflammation and promotion of atherosclerosis.5
Animal experiments also show how environmental influences during critical periods of pregnancy and early development may cause epigenetic modifications in the fetus, resulting in permanent offspring health effects. In Agouti viable yellow (Avy) mice, the Avy allele results from insertion of a retrotransposon (metastable epiallele) into the 5’ end of the wild type A allele. Unlike A/A wild type (brown) and a/a (black), Avy/a mice have variable coat color, ranging from yellow to mottled to brown. Providing female mice around the time of conception with dietary supplements such as folic acid, vitamin B12, choline and betaine increases the availability of methyl-groups and has been shown to cause higher CpG methylation at the Avy locus in the offspring.4 Avy methylation results in a phenotype characterized by brown coat color, reduced susceptibility to obesity, insulin resistance, and cancers, and lengthened life span.
Hypomethylation is characteristic of areas of smooth-muscle cell proliferation that has been found in advanced atherosclerotic plaques in human pathology specimens, as well as in atheromas of ApoE knockout mice and in neointimal thickenings of New Zealand White rabbit aortas. ApoE knock out mouse aortas exhibit a decrease in DNA methylation that can been detected as early as at four weeks of age, thus anticipating any histological changes associated with atherosclerosis.6 In human atherosclerotic tissues, the Estrogen Receptor (ER) α and β promoters show increased methylation. ER promoter methylation has been well demonstrated to increase with age even in normal tissues and reach near complete methylation level in the elderly.
Animal models have linked alterations in histone modifications with the development of atherosclerosis and cardiovascular disease. Trichostatin A, a specific histone deacetylase inhibitor (HDACi), accelerates macrophage infiltration and development of fatty streak lesions and in LDLR−/− mice.6 In mouse models of type 2 diabetes, smooth-muscle cells show increased expression of proatherogenic inflammatory genes. Turunen et al. have suggested that aberrant changes in histone modifications may represent a primary substrate for metabolic memory and proinflammatory phenotypes in smooth-muscle cells in diabetic vascular disease.6
The epigenome, unlike the genome, undergoes dynamic changes throughout the life course that can contribute to, and maintain adaptive and/or deviant gene expression states. DNA methylation is established during embryonic stages (de novo DNA methylation) by DNA methyltransferases 3a and 3b (DNMT3a & DNMT3b). In somatic cells, DNMT1 is responsible for mitotic replication (maintenance) of DNA methylation during mitosis. In mammalian cells, the fidelity of maintenance of methylation is 97–99.9% per mitosis. In addition, de novo methylation occurs in adult somatic cells in as much as 3–5% of mitoses, thus generating additional epigenetic changes.
Age, sex, and cardiovascular risk factors have been associated with specific patterns of DNA methylation and histone modifications. Loss of genomic DNA methylation has been found cross-sectionally in a variety of common age-related diseases. In work with the Normative Aging Study of men who receive care in Veterans Association hospitals, Bollati et al. showed a longitudinal decline in the average blood genomic DNA methylation of repetitive sequences such as Alu and LINE-1 over 8-years of follow-up.7 Genome-wide profiling of DNA methylation in blood DNA samples taken 11–16 years apart in recent studies from two cohorts from Iceland and Utah demonstrated both losses and gains in methylation over time, depending on the loci.8 The dynamic changes in DNA methylation appear to be influenced by additional factors related with cardiovascular risk. Three independent studies have consistently demonstrated that exposure to air pollution, an established risk factor for ischemic heart disease and stroke, was associated with reduced blood methylation of LINE-1 (for a review of epigenetic effects of environmental factors, see Baccarelli and Bollati9). Using a candidate-gene approach, hyper- and hypomethylation of specific genes was related to air pollutant exposures, including increased CDKN2B methylation, and decreased NOS2 and MAGEA1 methylation.9 Recently, Breton et al. have shown that second hand smoke induced lower Alu and LINE-1 DNA methylation, in child buccal cell DNA, as well as changes in methylation of specific genes identified through methylation profiling.10 In peripheral blood leukocytes of patients with hypertension, recent studies have shown a loss of global genomic methylation content,11 as well as hypermethylation of the HSD11B2 gene, linking epigenetics to blood pressure control.12
Several genes that are critical to glucose and lipid metabolism have been shown to be under epigenetic control, as reviewed by Ling and Groop.13 Participants in the The Dutch Hunger Winter Families Study, who were exposed in-utero to the 1944–1945 famine – a condition that has been associated with overweight, impaired glucose homeostasis and increased cardiovascular risk in adulthood, exhibited hypomethylation of the imprinted IGF2 and INSIGF genes, and hypermethylation of the GNASAS, MEG3, IL10, ABCA1, and LEP genes compared to unexposed siblings.14
The epigenome, due to its dynamic nature, may show signatures associated with cardiovascular risk biomarkers. Also, the individual epigenomic background may determine the levels of these biomarkers or their responses to acquired risk factors. In particular, DNA methylation has been linked to several cardiovascular-related biomarkers, including homocysteine,15 and C-reactive protein.16 DNA methylation is emerging as a primary regulator of inflammation. Methylation has been shown to control leukocytes functions related to cardiovascular risk, including the expression of soluble mediators, as well as of surface molecules that direct margination, adhesion, and migration of blood leukocytes in vascular tissues. Recent work in the Normative Aging Study has shown that mean serum vascular cell adhesion molecule-1 (VCAM-1) was associated with decreasing blood methylation levels measured in LINE-1 repetitive elements adjusting for potential confounders.17
Cardiovascular disease often develops following a prolonged asymptomatic phase. Identifying epigenomic biomarkers that parallel the development of subclinical disease might open new paths for risk stratification and prevention. Based on results from animal models showing atherosclerosis-related DNA methylation alterations in peripheral blood leukocytes, DNA methylation has been suggested to reflect hyperproliferation and altered functions of cell types involved in immune or inflammatory responses during atherosclerosis.18 Peripheral blood leukocytes can be easily obtained in the community and, have high potentials for the development of novel epigenomic biomarkers because of their roles inflammation, atherosclerosis, and cardiovascular disease etiology. A study of 93 patients with chronic kidney disease failed to reveal significant associations between global DNA methylation content in peripheral blood DNA and intima media thickness, taken as a measure of subclinical atherosclerosis.19 Larger studies including other measures of subclinical disease are needed.
Conversely, recent data have associated DNA methylation profiles, as measured in peripheral blood leukocytes, with clinical cardiovascular disease. Castro et al. found lower DNA methylation content in peripheral blood leukocytes from patients with atherosclerotic cardiovascular disease.20 Recent findings from the Normative Aging Study have shown that lower LINE-1 methylation in peripheral blood leukocytes is a predictor of incidence and mortality from ischemic heart disease and stroke.21 Elevated, not decreased, Alu methylation in peripheral blood leukocytes recently was related to prevalence of cardiovascular disease and obesity in Chinese individuals.22 Global methylation measures provide average estimates of methylation across the entire genome, or in wide portions of the DNA such as repetitive elements, and as such do not have the resolution necessary to pinpoint individual genes or sequences responsible for CVD risks. Decreased LINE-1 and Alu methylation may be accompanied by reactivation of different sets of silenced genes, which may be responsible for the opposite associations with cardiovascular risk.
DNA methylation and histone modifications represent attractive disease mechanisms, as in principle they might help explain how environmental and lifestyle factors can impose aberrant gene expression patterns in an individual's lifetime that can result in increased cardiovascular risk. At the current stage, several questions are open in cardiovascular epigenomics.
This review is only a brief and necessarily partial introduction to cardiovascular epigenomics that we hope will help motivate future research. A list of selected online resources that may help retrieve additional information and facilitate further work in epigenetics is presented in Table 1. We look forward to future epigenetic investigations that will produce fundamental knowledge about the pathophysiology of cardiovascular disease and lead to improved prevention, risk stratification and disease management.
Dr. Rienstra is supported by a grant from the Netherlands Organization for Scientific Research (Rubicon grant 825.09.020). This work was supported by grants from the NIH to Dr. Benjamin (1R01HL092577, 1RC1HL101056, 1R01HL102214, R01AG028321) and Dr. Baccarelli (ES000002, 1R21ES019773).
Conflict of Interest Disclosures