Epigenetics: The New Frontier for Vascular Diseases
By Thomas G. Brock, Ph.D.
In the world of personal healthcare, there aren’t many things that seem as personally adjustable as cardiovascular health. There are ‘heart smart’ foods, suggesting that simple changes in diet can reap benefits for your heart. A regular schedule of aerobic exercise will strengthen your cardiovascular system. Adjust your sleep patterns, get a pet, find a spouse, eat less, or listen to classical music and you might enjoy a long life of vascular bliss. The ways you can manipulate the system seem almost endless.
Naturally, there is the genetic side of vascular diseases. Genetic defects in such diverse parameters as lipid metabolism, arterial structure, endothelial and smooth muscle function, platelet adhesiveness, and fibrinolysis can clearly contribute to the development of atherosclerosis, for example. The phenotypic consequences of such defects may range from subtle to pronounced. Superficially, genetic problems seem so much more refractory to treatment than the personal, everyday lifestyle options.
Now, everything is about to get a lot more complicated as well as very interesting. The ways that all those adjustable, personal factors impact gene expression in prolonged, meaningful ways is part of the field of epigenetics. Technically, ‘epigenetics’ refers to heritable changes in gene expression that do not involve changes in DNA sequence or copy number. The cell biologist uses epigenetic events to help explain how pluripotent stem cells differentiate into diverse cell types, while developmental biologists see epigenetic processes at work in the morphogenesis of an embryo into discrete organs. Similarly, epigenetic signaling that is put in place in response to prenatal or early life events can contribute to disease development much later in life.
Epigenetics: The Basics
In molecular terms, epigenetic changes commonly involve adding or removing various “marks” on chromatin. The basic structural unit of chromatin is the nucleosome, which consists of a short stretch of DNA wrapped twice around a protein octamer composed of two sets of the core histones, H2A, H2B, H3, and H4 (Figure 1). The C-termini of the histones are highly conserved and form the protein core of each nucleosome, while the N-terminal tails project from the nucleosome. These tails can interact with DNA or with other proteins. For example, each tail is rich in regularly spaced, positively-charged residues (lysine (K), arginine (R)), which can associate with the anionic backbone of DNA. Acetylation or methylation of these residues alters their charge and size, affecting DNA binding. Phosphorylation, often on sites adjacent to basic residues, puts a large negatively-charged moiety on the histone. These types of chemical marks also lead to structural changes in the histone tail, providing binding sites for proteins or protein complexes, which in turn modulate further protein or histone changes. Histone marks can be reversed enzymatically, or they can persist for longer periods, extending through multiple cell divisions to maintain cell phenotype. A much more stable modification involves methylation of cytosine residues on DNA, most notably on ‘CpG islands’, genomic regions that are rich in CG pairs. Methylation of DNA is particularly stable and often results in the suppression of transcription of specific genes by physically interfering with the transcriptional machinery. Methylated DNA, like marked histones, provides binding sites for proteins and the establishment of protein complexes that can modify histones and alter gene expression. These are some of the basic ways that phenotype can be stably altered without changing DNA sequence or copy number.
Food For Thought
You are what you eat, but does your diet affect your children’s health? About ten years ago, Bygren and colleagues studied a cohort of individuals born in 1905 in the Överkalix parish in northernmost Sweden, where annual harvests are heavily impacted by the weather.1 County records provided birth and death dates for the 1905 cohort, their parents and their grandparents. Additional records indicated years of poor, moderate, or superior availability of crop food during the preceding century. Was the lifespan of the individuals born in 1905 affected by the nutritional experience of their predecessors? To focus this question, the authors hypothesized that, in order for famine or food surplus to have persistent effects, the dietary impact must occur in a sexually formative period. Interestingly, they found a significant correlation between food availability for paternal grandfathers when they were 9-12 years old and the survival of their grandchildren. Perhaps more remarkably, grandchild lifespan shortened if there was an excess of food for the paternal grandfather and increased if the grandfather experienced famine during this critical developmental stage. The effect was not trivial: the difference in survival for grandchildren averaged 32 years!
A follow-up study by the same group expanded the study to include cohorts from 1890, 1905, and 1920 and examined cause of death data.2 They reported that a father experiencing famine during the critical 9-12 years of age passed protection against cardiovascular disease to his children. To restate: famine for the father meant less cardiovascular disease for his children. No significant correlations were found for mothers, grandmothers, or grandparents and offspring cardiovascular disease. These correlative studies anticipated a decade of studies into the now popular concept that the diet of the mother might reprogram the genes of the fetus, without mutation, leading to persistent effects in health and development of the child.3
Case in Point: Epigenetics in Hypertension
Aldosterone is a corticosteroid hormone that, through mineralocorticoid receptors, alters the uptake and secretion of salts by collecting ducts of the kidneys, thus altering water uptake, blood volume, and blood pressure. Excess aldosterone contributes to arterial hypertension, congestive heart failure, chronic kidney disease, coronary artery disease, and stroke.4 Clearly, the mechanism of action of aldosterone at the kidney collecting duct must be important.
One enzyme serves to hypermethylate the lysine residue 79 on histone 3 (H3K79). In yeast, ablation of this methyltransferase causes disruption of telomere silencing, leading to the name DOT1. A mouse DOT1 homolog, DOT1L, also hypermethylates H3K79 in mice, but this serves to repress the expression of a subset of genes. DOT1L binds to AF9, a DNA-binding protein which positions DOT1L only on those nucleosomes whose associated DNA has a sequence that is recognized by AF9. This confers selectivity for which nucleosomes are targeted for hypermethylation of H3K79 and which genes are consequently repressed. One such gene encodes a sodium channel subunit, epithelial Na+ channel subunit α (ENaCα).
Aldosterone alters the expression of ENaCα by modulating the DOT1L/AF9 repressive complex (Figure 2). Aldosterone induces the expression of serum- and glucocorticoid-induced kinase 1 (Sgk1), which phosphorylates AF9 and disrupts the complex.5 Aldosterone also inhibits the expression of both DOT1L and AF9.6,7 These steps are necessary and sufficient to increase ENaCα expression in the presence of the hormone-receptor complex.
It is worth noting that this pathway involves a short term and reversible mechanism involving histone modification. Such transient changes are the hallmark of the ‘signaling model’ of epigenetic modifications. These contrast with other pathways where histone modifications are inherited through cell division, the so-called ‘histone code hypothesis’ of epigenetic signaling.
Classical Epigenetic Studies
The Dutch Famine Birth Cohort Studies relate to babies born around a 5-month period of extreme food shortage during the winter of 1944-1945. Numerous characteristics of mothers and offspring were obtained around birth and the offspring have been followed since. In one study, exposure to famine during gestation led to a higher cumulative incidence of coronary artery disease, as well as an earlier age at onset, than was found for those not exposed to famine during gestation.8 Moreover, these effects were found only if exposure to famine occurred early in gestation. Such correlative studies leave us to wonder: what could have happened during those first few months of development that became evident only years later?
A series of recent studies have come closer to demonstrating epigenetics in action. While obviously the diet of a pregnant mother can affect fetal development, these studies indicate that dietary supplements, taken by the mother, may mark the genome of the fetus and affect adult health. In the viable yellow agouti (Avy) mouse, expression of the agouti gene leads to a switch from brown to yellow coat color (Figure 3). Moreover, adult agouti mice suffer from obesity, hyperlipidemia, and hypertension. The expression of the agouti gene is initiated from a cryptic promoter in a retrotransposon inserted in agouti pseudoexon 1A (PS1A).9 It’s known that cytosine methylation on the transposable element prevents agouti gene expression, producing a brown-coated mouse that is referred to as ‘pseudoagouti’.10 Waterland and Jirtle demonstrated that supplementing normal chow given to pregnant mice with methyl donors (folic acid, betaine, vitamin B12, choline) produced an increase in methylation of CpG sites within PS1A in the offspring and shifted the coat color distribution toward the brown (pseudoagouti) phenotype.9 These methylation patterns were found in cells from diverse tissues, showing that Avy methylation is determined in the early embryo and maintained with high fidelity throughout development.
In another study, genistein, an isoflavonoid naturally found in soy, increased both PS1A methylation and frequency of pseudoagouti expression in offspring when added to the mother’s diet.11 Importantly, ectopic agouti expression is associated with adult-onset obesity, diabetes, and cancer.12 Hypermethylation of PS1A, in mice from genistein-fed mothers, persisted into adulthood, decreasing agouti expression, and, remarkably, protecting mice from obesity.11 Conversely, hypomethylation of PS1A, following exposure to the common chemical bisphenol A (BPA), increased agouti expression;13 BPA is known to promote obesity, hyperlipidemia, and hypertension in mice. Taken together, these results demonstrate that maternal diet has in utero effects on the epigenome of the early embryo that can alter susceptibility to disease into adulthood.
1. Bygren, L.O., Kaati, G., and Edvinsson, S. Acta Biotheor. 49, 53-59 (2001).
2. Kaati, G., Bygren, L.O., and Edvinsson, S. Eur. J. Hum. Genet. 10, 682-688 (2002).
3. Heerwagen, M.J., Miller, M.R., Barbour, L.A., et al. Am. J. Physiol. Regul. Integr. Comp. Physiol. 299(3), R711-R722 (2010).
4. Calhoun, D.A. Circulation 114, 2572-2574 (2006).
5. Zhang, W., Xia, X., Reisenauer, M.R., et al. J. Clin. Invest. 117(3), 773-783 (2007).
6. Zhang, W., Xia, X., Reisenauer, M.R., et al. J. Biol. Chem. 281(26), 18059-18068 (2006).
7. Zhang, D., Yu, Z.-Y., Cruz, P., et al. Kidney Int. 75(3), 260-267 (2009).
8. Painter, R.C., de Rooij, S.R., Bossuyt, P.M., et al. Am. J. Clin. Nutr. 84, 322-327 (2006).
9. Waterland, R.A. and Jirtle, R.L. Mol. Cell. Biol. 23(15), 5293-5300 (2003).
10. Morgan, H.D., Sutherland, H.G.E., Martin, D.I.K., et al. Nat. Genet. 23, 314-318 (1999).
11. Dolinoy, D.C., Weidman, J.R., Waterland, R.A., et al. Environ. Health Perspect. 114(4), 567-572 (2006).
12. Yen, T.T., Gill, A.M., Frigeri, L.G., et al. FASEB J. 8, 479-488 (1994).
13. Dolinoy, D.C., Huang, D., and Jirtle, R.L. Proc. Natl. Acad. Sci. USA 104(32), 13056-13061 (2007).
Figure 1. Histone tails
A.) Positioning of the histone tail relative to the C-terminal folded region. B.) Amino acid sequences of core histone N-terminal tails, indicating sites of phosphorylation (p), acetylation (a), ADP ribosylation (rib), and methylation (m).
Figure 2. Aldosterone promotes the phosphorylation of Af9 by Sgk1, dissociating DOT1L from Af9, allowing demethylation of H3K79 and the expression of ENaCα
Figure 3. Genetically identical week 15 littermates representing coat colors ranging from agouti (left) to pseudoagouti (right); Note differences in size
Most figures are available in high-resolution formats for use in other works at a nominal fee.