Transgenerational Epigenetics: The Role of Maternal Effects in Cardiovascular Development

Abstract

Transgenerational epigenetics, the study of non-genetic transfer of information from one generation to the next, has gained much attention in the past few decades due to the fact that, in many instances, epigenetic processes outweigh direct genetic processes in the manifestation of aberrant phenotypes across several generations. Maternal effects, or the influences of maternal environment, phenotype, and/or genotype on offsprings’ phenotypes, independently of the offsprings’ genotypes, are a subcategory of transgenerational epigenetics. Due to the intimate role of the mother during early development in animals, there is much interest in investigating the means by which maternal effects can shape the individual. Maternal effects are responsible for cellular organization, determination of the body axis, initiation and maturation of organ systems, and physiological performance of a wide variety of species and biological systems. The cardiovascular system is the first to become functional and can significantly influence the development of other organ systems. Thus, it is important to elucidate the role of maternal effects in cardiovascular development, and to understand its impact on adult cardiovascular health. Topics to be addressed include: (1) how and when do maternal effects change the developmental trajectory of the cardiovascular system to permanently alter the adult’s cardiovascular phenotype, (2) what molecular mechanisms have been associated with maternally induced cardiovascular phenotypes, and (3) what are the evolutionary implications of maternally mediated changes in cardiovascular phenotype?

Introduction

Epigenetics can be studied within an individual’s lifetime (cell-to-cell transmission of non-genetic information), and/or as effects on subsequent generations (body-to-body transmission). The non-genetic transfer of information across generations can be further divided into two general non-exclusive categories, transgenerational epigenetic “effects/transfer” and transgenerational epigenetic “inheritance.” The former term is used to describe the transmission of information from parent to offspring that is non-genomic in nature, but whose exact mechanism of action often is not yet known (Youngson and Whitelaw 2008; Ho and Burggren 2010). Transgenerational epigenetic inheritance, on the other hand, refers to the transmission of specific epigenetic marks/processes across generations, often via the germline (Jablonka and Lamb 1989). The most heavily studied epigenetic marks are methylation and acetylation of DNA or histones. However, other epigenetic inheritance systems such as self-perpetuating loops, non-coding RNAs, and structural inheritance have been implicated in transgenerational epigenetic inheritance (Jablonka and Raz 2009).

Maternal effects, or “the causal influence of maternal phenotype or genotype on offspring phenotype” (Wolf and Wade 2009), fall under the umbrella of transgenerational epigenetics. Due to the intimate role of the mother during embryonic and juvenile development in most animals, there is much interest in investigating the means by which maternal effects can shape the individual. The first reported study of maternal effects was in the pulmonate snail, Limnaea peregra, in which directionality of shell-coiling was determined to be inherited in a non-Mendelian fashion via a yet-to-be discovered factor in the yolk (Boycott and Diver 1923; Sturtevant 1923). Since then, maternal effects in early development has been studied in a wide variety of species and biological systems. Prior to the start of zygotic transcription, maternal effects are responsible for cellular organization and determination of the axis of the body, while in later stages of development, maternal effects are influential in the initiation and maturation of organ systems and the general performance of the organism (Nusslein-Volhard et al. 1987; Bernardo 1996; Eising et al. 2006; Youngson and Whitelaw 2008; Ho and Burggren 2010; Champagne 2012). In particular, the cardiovascular system is largely affected by early environmental perturbations, especially during perinatal development. Thus, it is important to highlight the role of maternal effects in cardiovascular development and in the subsequent cardiovascular phenotype of the adult, and to discuss the work that remains to be done.

In this review, I discuss: (1) how and when maternal effects, through non-genomic processes, can change developmental trajectory of the cardiovascular system to permanently alter the adult’s cardiovascular phenotype, (2) what molecular mechanisms have been associated with maternally induced cardiovascular phenotypes and how persistent these effects are across generations, and (3) the evolutionary implications of maternally mediated changes in cardiovascular phenotype.

Evidence of maternal effects in cardiovascular development

Critical windows of cardiovascular development

Transmission of non-genomic information from mother to offspring most often occurs early in the life of the offspring. In oviparous and viviparous animals, early life, or the perinatal period, can be divided into two sequential stages, (1) pre-hatch or prenatal (in ovo or in utero, respectively) stage, and (2) post-hatching or post-natal stage. The environment in which the offspring’s development occurs defines each period of life. The pre-hatch or prenatal period is largely defined by the nature of the egg or the uterine environment, whereas the postnatal environment is largely defined by maternal behavior. During these early stages of life, small perturbations can induce large changes in the developmental trajectory of the cardiovascular system, leading to a permanently altered phenotype in adulthood (Burggren 1999; Burggren and Reyna 2011).

The change in developmental trajectory is especially profound and stable when environmental perturbations occur during sensitive periods or “critical windows” of development. Often, a critical window coincides with a period of drastic transition in a biological system. The cardiovascular system undergoes extensive structural and functional changes both during prenatal and postnatal periods of life; thus, one can imagine the great potential for maternally induced shifts in developmental trajectory. This is especially pertinent when the cardiovascular system transitions from functioning to support its own development as well as the growth and development of many other organ systems such as the renal, nervous, and lymphatic systems (Burggren et al. 2013), to functioning as a convective transport system for the body (Pelster and Burggren 1996; Jacob et al. 2002; Rombough 2002). There is evidence that the transition from hyperplastic to hypertrophic growth in the heart is a critical period of development (Anatskaya et al. 2010). During these critical windows, the interplay of the environment and genetics define the final cardiovascular phenotype (Mone et al. 2004). Each period of development is characterized by unique maternal contributions and mother–fetus/offspring interactions that may contribute to the offspring’s early cardiovascular phenotype, thereby leading to an altered phenotype in the adult.

Prenatal/pre-hatch maternal effects

As formation and growth of vessels (vasculogenesis and angiogenesis), and morphogenesis of the heart (cardiomyocyte proliferation and maturation) ensues during the in utero/in ovo period of life, the embryo or fetus relies heavily, and at times solely, on maternal contribution to the embryonic and fetal environment. In oviparous species, egg quality and size, which are dependent upon life-history traits (i.e., fecundity, gestational time, age to sexual maturity, age at reproduction, and size at maturity) of the mother, have been shown to play a direct role in cardiovascular phenotype of the of the F1 generation. Egg size, which is determined by maternal phenotype, has been shown to allometrically determine embryonic heart rate in birds (Tazawa et al. 2001) and some species of snake (Aubret 2013). Also, breed-specific differences in composition of the yolk in chickens determine embryonic heart rate as well as growth rate (Ho et al. 2011). Maternal nutrition has been shown to alter the lipid content in the eggs of chickens, and this has been implicated in inducing metabolic and cardiovascular disease in the hatchlings (Cherian 2007). In zebrafish, an oviparous species, maternal nutrition affects hematological parameters as well as the heart’s performance (heart rate, stroke volume, and cardiac output) of the offspring at various stages of larval and juvenile development (Schwerte et al. 2005).

Great public interest in the origins of human health and disease (i.e., fetal programming) has prompted large efforts to shed light on the potential influence of maternal environment in the establishment of the offspring’s cardiovascular health. The effect of maternal phenotype and environment, such as nutrition, toxins, stress, and exercise on the offspring’s cardiovascular-related health and disease has been studied in humans, and these data are strongly supported by studies in laboratory animals (Hoet and Hanson 1999; Bertram and Hanson 2002; Brawley et al. 2003; Melzer et al. 2010; Siebel et al. 2012; Mottola 2013). In general, these studies show that suboptimal birth weight due to adverse maternal conditions leads to increased predisposition to cardiovascular-related pathologies in adulthood. Interestingly, studies suggest that there is a robust correlation between maternal hypertension and low birth weight of offspring in humans and rodents (Drake and Walker 2004). As low birth weight is highly correlated with high systolic blood pressure and increased risk for cardiovascular disease later in life, it comes to reason that this maternal effect may persist far beyond the F1 and F2 generations (Drake and Walker 2004).

Epidemiological studies based on individuals who were born during the Dutch famine of 1944–1945 reveal that adults subjected to maternal under-nutrition during prenatal life had higher incidences of obesity, metabolic disease, and cardiovascular disease, and this was unrelated to the child’s size at birth (Roseboom et al. 2000). More specifically, maternal malnourishment in the third trimester led to higher blood pressure in their children upon reaching adulthood (Roseboom et al. 2001). Similarly, the children of women that suffered the Chinese famine during 1959–1961 had significantly higher mean systolic blood pressure and higher prevalence of hypertension as adults than did their unexposed counterparts (Chen et al. 2013). Maternal over-nutrition resulting in maternal obesity during gestation also predisposes the offspring to increased risk of cardiovascular-related diseases; however, the underlying mechanisms are different from those of maternal under-nutrition (Dong et al. 2013). Moderate maternal exercise during pregnancy has been suggested to counter the ill-effects of maternal obesity on fetal cardiovascular health (Mottola 2013). Exercise during pregnancy improves maternal health and reduces the incidence of gestational diabetes, which may have far-reaching effects on the cardiovascular functions of the child (Melzer et al. 2010). Additionally, May et al. (2012) reported that fetal heart rate was negatively correlated with maternal activity level, whereas variability in heart rate was positively correlated with maternal activity level during the third trimester (May et al. 2010). The authors suggest that maternal exercise induces direct cardiovascular effects on the fetus that is similar to aerobic training in adults.

Postnatal maternal effects

Cardiovascular growth and development continues beyond prenatal stages of life. In vertebrates, the heart rapidly undergoes hyperplastic growth soon after birth, followed by a switch towards hypertrophic growth (Li et al. 1996). This switch occurs at distinct points in time; in humans this occurs 3–6 months after birth and in rodents this occurs during the first 2 weeks of postnatal life (Chen et al. 2004). In mammals, capillary density and microvasculature maturation of pulmonary, retinal, renal, muscle, and cerebral vessels continue after birth (Schnitzer 1988; Wang et al. 1992; Stingl and Rhodin 1994). During this distinct period of cardiovascular development, maternal effects reflect the quality of maternal behavior/care.

Historically, examination of the influence of maternal care during postnatal life has focused largely on the offspring’s behavioral phenotype, with emphasis on the programming of the sympathoadrenal–medullary (SAM) and the hypothalamic–pituitary–adrenal (HPA) systems (Essex et al. 2011; Juruena 2013). Convergent data from a wide variety of animals indicate that suboptimal maternal care sensitizes the offspring’s HPA axis to subsequent stressors by altering release of glucocorticoid and catecholamine, and the expression of glucocorticoid receptor (reviewed by Sanchez 2006; Meaney et al. 2007; Loria et al. 2014). It has been well-established that prolonged activation of the SAM and HPA systems leads to the development and progression of cardiovascular disease; however, only recently has stress in early life in the form of maternal behavior/care been linked to acute cardiovascular disturbances as well as to chronic disease in adulthood. In humans, studies have strongly suggested that adverse early-life events or chronic behavioral stressors that induce long-term anxiety during childhood are independent risk factors for the development of cardiovascular disease in adulthood (Alastalo et al. 2009, 2013). These early-life stressors include maternal separation, neglect, and abuse. In rodents, altered maternal care in the form of maternal separation during the first few weeks of life results in changes in adults’ cardiovascular function (Loria et al. 2014). Maternal separation in rodents involves the temporary daily separation of offspring from the mother during the early period of life when the offspring is most reliant on maternal care for development and survival. Adult rats that have been subjected to maternal separation not only have higher blood pressure in response to acute stressors (Sanders and Anticevic 2007), but they also have increased sensitivity to vasoactive and hypertension-promoting factors such as angiotensin II and norepinephrine, suggesting a dysfunction in adults’ cardiovascular and renal function (Loria et al. 2010, 2011, 2013). Also, in mice, maternal separation induces vascular endothelial dysfunction in adulthood (Ho et al. 2012).

Although interesting, many studies that focus on the cardiovascular aspect of perinatal maternal effects report cardiovascular outcomes in the adult, especially risk of disease in humans (Hanson et al. 2011). This provides little insight into the direct impact of maternal phenotype during cardiovascular development. Only a few studies have focused on early embryonic, fetal, and early postnatal cardiovascular parameters as endpoints (Brenner et al. 1999; Schwerte et al. 2005; Momoi et al. 2008; May et al. 2010, 2012; Ho et al. 2011). Even so, these few studies lack the mechanistic investigation of the long-term effects of these maternally induced perinatal cardiovascular changes in the offspring.

Potential mechanisms of maternally induced transgenerational transfer of cardiovascular phenotype

Observations of human disease, the use of transgenic animals, and the study of comparative physiology have provided great insight into highly conserved factors and mechanisms that regulate cardiac and vessel development. Mechanical forces such as shear stress, stretch, and intercardiac fluid forces, and the oxygen environment of the embryo have been shown to guide early cardiovascular development by inducing the transcription of vasculogenic, angiogenic, and cardiogenic factors such as vascular endothelial growth factor and hypoxia inducible factor-1 (Hove et al. 2003; le Noble et al. 2004; Burggren 2004; Bagatto and Burggren 2006; Simon and Keith 2008; Han et al. 2010). Additionally, hormones such as thyroid hormones, testosterone, and the newly discovered peptide hormone, elabela, have been implicated in early cardiovascular development (Moussavi et al. 1985; Schjeide et al. 1989; Er et al. 2007; Goldman-Johnson et al. 2008; Chng et al. 2013). More recently, genome-wide expression profiles provide a glimpse of the complexity underlying cardiovascular development. Changes in gene-expression profile over the course of development have uncovered novel factors involved in the unfolding of the cardiovascular system (Chen et al. 2004; Miquerol and Kelly 2013; Park et al. 2013). Along the same lines, mechanistic study of the epigenome have identified a multitude of epigenetic processes and marks that are crucial in regulating the expression of genes during cardiovascular development (for a thorough review on the topic, see (Skinner 2011; Chang and Bruneau 2012; Vallaster et al. 2012). With the study of models of cardiovascular disease, transgenic laboratory animals (knock-out of epigenetic markers), and studies of monozygotic and dizygotic twins in humans, it has become more evident that epigenetic processes are important in understanding how cardiovascular development can be molecularly regulated to give rise to permanent phenotypes of adults (Chang and Bruneau 2012; Sun et al. 2013). Only recently have mechanistic studies broadened our understanding of how maternal effects can regulate many of the above-mentioned factors/mechanisms known to control early cardiovascular development, with still many questions left unanswered.

In oviparous animals such birds and reptiles, maternal effects that occur in ovo are the result of the quality of the protective barrier of the egg, and/or maternally produced nutrient stores, lipids, mRNAs, transcription factors, immune factors, antioxidants, and hormones deposited into the egg during oogenesis (Smith and Ecker 1965; Craig and Piatigorsky 1971; Rose and Orlans 1981; Schwabl 1993; Kudo 2000; Dzialowski et al. 2009; Peluc et al. 2012). The importance of maternally derived yolk factors in cardiovascular development is highlighted by studies in which manipulation/alteration of maternally derived egg factors result in morphological and functional cardiovascular changes in the embryo. Ho et al. (2011) used a xenobiotic approach to examine the effects of maternally derived environment of the yolk in the determination of cardiovascular phenotype in two breeds of chicken, the layer and the broiler chicken (Ho et al. 2011). During embryonic, juvenile, and adult stages of life, the layer chicken possesses significantly lower growth rate, mass, metabolic rate, and heart rate compared with the broiler chicken (Latimer and Brisbin 1987; Martinez-Lemus et al. 1998, 1999; Boerjan 2004; MacRae et al. 2006; Sato et al. 2006; Everaert et al. 2008). The authors determined that the difference in yolk environment between the two breeds was partially responsible for the breed-specific difference in embryonic heart rate (Ho et al. 2011). In the same study, it was determined that the yolk of broilers and layers contain significantly different concentrations of thyroid hormone and testosterone, both of which have cardiogenic and vasculogenic effects in embryonic tissues. Additional studies are necessary to elucidate the specific factors responsible for these maternal effects.

The molecular mechanisms underlying maternal effects on offspring’s cardiovascular phenotype have been heavily pursued in the area of epidemiology, resulting in the emerging field of epigenetic epidemiology (Mill and Heijmans 2013). Individuals who were exposed to famine during prenatal periods of life during the Dutch famine not only are at higher risk of cardiovascular disease, but they have lower DNA methylation of the IGF2 (insulin-like growth factor 2) differentially methylated region, and the INSIGF [alternative splice-variant of insulin (INS) transcript and IGF2 transcript] loci. Additionally, these affected individuals have higher DNA methylation of the IL10 (interleukin 10), LEP (leptin), ABCA1 (an ATP-binding cassette transporter), GNASAS [long non-coding RNA antisense to the stimulatory G-protein alpha subunit (GNAS) gene], and MEG3 (maternally expressed 3) loci compared with their unexposed siblings (Heijmans et al. 2008; Tobi et al. 2009). These differentially methylated transcripts are involved in the regulation of metabolic pathways as well as in cardiovascular, neurological, and immunological function. Taken together with evidence that there is increased adiposity and potential for increased risk of cardiovascular disease in the affected F2 generation in the Dutch famine cohort, this reinforces the longevity of the effects of maternal condition on the offspring (Veenendaal et al. 2013).

Aberrant maternal care leads to detectable epigenetic marks that can persist over multiple generations (Curley et al. 2009). Increased maternal licking and grooming in mice caused changes in epigenetic regulation of hippocampal glucocorticoid receptors in the offspring (Weaver 2007), whereas maternal separation resulted in an altered state of DNA methylation in the brain (Murgatroyd et al. 2009; Murgatroyd 2013). Maternal separation can also dynamically change expression and activity of histone deacetylase (HDAC) in the murine brain over the course of development (Levine et al. 2011), and more recently, maternal separation in mice has been shown to increase HDAC expression in the adults’ vasculature (Ho and Pollock 2012). Much is known of the transmission and induction of epigenetic marks in the brain due to perturbations in maternal care; however, there is a need for further exploration of the mechanism by which maternal separation may alter epigenetic marks in the cardiovascular system.

Evolutionary potential of maternal effects in cardiovascular phenotype

Although the impact of maternal effects in pre-zygotic organization of tissue and post-zygotic development has gone in and out of fashion throughout the past century, it is an undeniable factor in evolution (Mousseau and Fox 1998; McAdam et al. 2002; Reinhold 2002; Garamszegi et al. 2007; Badyaev 2008). Of evolutionary importance is the impact of maternal effects on life-history traits of the offspring that have a clear and direct bearing on fitness (Galloway et al. 2009). These effects have been observed in a wide range of plants and animal species (Sakwinska 2004; Wilson et al. 2005; Garamszegi et al. 2007; Solberg et al. 2007; Allen et al. 2008; Biard et al. 2009). In vertebrate animals, the variation in life-history traits such as birth weight, stamina, performance, and juvenile mass have been directly attributed to maternal provision and care (Sinervo and Huey 1990; Wilson and McNabb 1997; Finkler et al. 1998; Dzialowski and Sotherland 2004; Eising et al. 2006; Biard et al. 2009; Dzialowski et al. 2009; Ho et al. 2011; Ho and Burggren 2012; Braun et al. 2013; Burton et al. 2013; King et al. 2013).

Currently there is no direct evidence that maternal effects on cardiovascular development can alter maternally induced evolutionary trends. However, traits that are under the influence of maternally driven evolutionary pressure can be largely influenced by cardiovascular performance and function. It will be exciting to see if, and how, maternal effects on cardiovascular development may contribute to the alteration of life-history traits and thus alter evolutionary trends.

Conclusion

To date, studies of the influence of maternal provision and care on offspring’s cardiovascular phenotype focus largely on the adult endpoints, with very few studies examining the influence of maternally derived factors on cardiovascular development per se. This poses the question: are maternally-induced changes in cardiovascular phenotype at the prenatal and early postnatal stages stable, and if so, can adults’ cardiovascular phenotype be truly attributed to maternal influences in early development? Furthermore, there is a paucity of studies that examine maternal effects on cardiovascular phenotype beyond the first and second generations. In some instances of environmentally induced maternal effects, this raises the question of whether there is an actual transmission of information from mother to offspring or if effects on the offspring’s cardiovascular phenotype is a result of direct environmental influences. Some argue that a true transgenerational epigenetic occurrence requires the effect to be seen in the F3 generation, independent of the original environmental context of the parental (F0) generation to avoid direct environmental effects on the germline (Skinner et al. 2010). In other words, to be a true transgenerational epigenetic effect, the offspring’s phenotype must not be a result of the offspring experiencing the same environment as the mother. This is an important consideration for future studies.

We are a long way from fully understanding the intricacies and complexity of transgenerational epigenetics and cardiovascular development, especially in the context of maternal influence; however, from the existing literature, we gain an appreciation of the immense, multiplicative nature of maternal effects in the development of the cardiovascular system.

Funding

Funding was provided by an F32 postdoctoral grant from the National Institutes of Health (HL116145) to D.H.H.