Epigenetic mechanisms linking early-life exposures to long-term health outcomes

The mechanisms underlying developmental programming remain still unclear. However, increasing evidence has been accumulated indicating the key role of epigenetic regulation in the developmental programming phenomenon.Reference Barouki, Gluckman and Grandjean^6, Reference Hochberg, Feil and Constancia⁷ Epigenetic regulation of gene expression involves mitotically and meiotically heritable changes in gene expression which are not because of changes in the primary DNA sequence. The most studied epigenetic mechanisms affecting gene expression are: (1) DNA methylation, the enzymatic addition of a methyl group (-CH₃) to the 5-position of the cytosine ring,Reference Auclair and Weber⁸ (2) enzymatic modifications of histones, including acetylation, methylation, phosphorylation and ubiquitinationReference Verdone, Agricola, Caserta and Di Mauro⁹ and (3) synthesis of small (∼20–30 nucleotides) non-coding RNAs (ncRNAs), including microRNAs (miRNAs), Piwi-interacting RNAs (piRNAs) and long non-coding RNAs (long ncRNAs), that can target messenger RNA to interfere with transcription or translation.Reference Portnoy, Huang, Place and Li¹⁰ The epigenetic mechanisms act in a coordinated manner to regulate gene expression. In general, DNA methylation represses transcription, whereas histone acetylation is linked to transcriptional activation. Acetylated histones are associated with unmethylated DNA and are almost absent in methylated DNA regions. Schematic representation of main epigenetic mechanisms is presented in Fig. 1.

Fig. 1

Schematic representation of major epigenetic mechanisms. The major epigenetic mechanisms affecting gene expression are: (1) DNA methylation, (2) enzymatic modifications of histones and (3) synthesis of small non-coding RNAs (ncRNAs).

The traditional view of epigenetic modifications as static on/off switches in the control of gene expression is currently being altered by the idea that these marks are dynamic, and different environmental factors can act on the epigenetic machinery to bring about either short-term or long-term outcomes.Reference Hochberg, Feil and Constancia⁷ In human beings, epigenetic mechanisms have been shown to be very sensitive to environmental stimuli other than the nutrients and physiological factors that are normative, in evolutionary terms, to the human environment.Reference Barouki, Gluckman and Grandjean⁶ The different xenobiotic environmental exposures, including drugs, industrial chemicals, tobacco smoke and others, can lead to adverse consequences such as increased disease risk.

There are specific stages of development during which the epigenome is more labile than it is during adulthood, therefore the environment with the potential for the most profound influence on epigenetic states is that encountered in prenatal life.Reference Thompson and Einstein¹¹ During early development, the genome-wide changes in epigenetic marks orchestrate chromatin in a way destined to form different organs and tissues in the body. Once established, the epigenetic marks are stably maintained through somatic cell divisions and create unique, lineage-specific patterns of gene expression. In mammalian development, there are two main periods of enhanced sensitivity of the epigenome to environmental stimuli: gametogenesis and early embryogenesis.Reference Vickaryous and Whitelaw¹² During germ cell development, the genome-wide DNA demethylation occurs followed by the remethylation before fertilization. Early embryogenesis is then characterized by a second genome-wide demethylation wave, and patterns of methylation are re-established after implantation. The postfertilization demethylation and remethylation phases are likely to play a role in the removal of acquired epigenetic modifications, which can be influenced by individual genetic and environmental factors.Reference Reik, Dean and Walter¹³ The epigenome is therefore likely to be particularly vulnerable to the adverse influences during these stages of development. In human beings, the window of epigenetic developmental plasticity extends from preconception to early childhood and involves epigenetic responses to environmental changes, which exert their effects during life-history phase transitions.Reference Hochberg, Feil and Constancia⁷ Currently, the environmentally induced alteration in epigenetic markers is considered to be a key mechanism connecting the early-life events with age-associated diseases including cancer, neurodegenerative diseases and type 2 diabetes.Reference Gluckman, Hanson and Low^14, Reference Vaiserman¹⁵

Alcohol exposure in early development

Ethanol is a classic teratogen capable of inducing a wide range of developmental abnormalities. It is well known that heavy prenatal alcohol exposure during early life can severely impair human physical and neurobehavioral development. Epidemiological data offer numerous evidence that paternal alcohol consumption can affect birth weight, congenital heart defects and mild cognitive impairments.Reference Abel^16–Reference Ramsay¹⁸ Prenatal exposure to alcohol is generally known to have adverse effects on offspring neuroendocrine and behavioral functions. Alcohol readily crosses the placenta, thus directly affecting developing fetal endocrine organs. In addition, alcohol-induced changes in maternal endocrine function can disrupt the normal hormonal interactions between the pregnant female and fetal systems, altering the normal hormone balance and, indirectly, affecting the development of fetal metabolic, physiological and endocrine functions.Reference Weinberg, Sliwowska, Lan and Hellemans¹⁹

Some animal studies have explored long-term adverse effects of prenatal alcohol exposure. In utero alcohol exposure was found to cause long-term reduction in plasma insulin-like growth factor 1 (IGF1) concentrations,Reference Breese, D'Costa and Ingram²⁰ and permanent alterations in the thyroid functionReference Wilcoxon and Redei²¹ of the adult offspring. Altered hypothalamic–pituitary–adrenal (HPA) axis functionReference Weinberg, Sliwowska, Lan and Hellemans¹⁹, behavior,Reference Abel and Berman^22, Reference Janicke and Coper²³ enhanced age-related neural degeneration,Reference Church, Hotra and Holmes²⁴ memory lossReference Dumas and Rabe²⁵ and diabetesReference Pennington, Shuvaeva and Pennington^26, Reference Yao, Chen and Nyomba²⁷ have been found as well. In the study by Abel et al.,Reference Abel, Church and Dintcheff²⁸ in utero alcohol-exposed animals died at a significantly younger age and never attained the same maximum body weights as control animals. For females prenatally exposed to alcohol, life span was shortened by about 20 weeks; in male cohorts, life span was shortened by about 2.5–7 weeks.

The detrimental effects following in utero alcohol exposure [fetal alcohol spectrum disorders (FASD)] have been recognized for centuries, but only recently new data were obtained providing a molecular insight into the mechanisms involved in FASD. Three developmental periods are particularly vulnerable: preconception, preimplantation and gastrulation. These periods of alcohol teratogenesis correlate with peak periods of epigenetic reprogramming. Along with the evidence that ethanol interferes with one-carbon metabolism, DNA methylation, histone modifications and non-coding RNA, it suggests an important role of epigenetic mechanisms in the etiology of FASDs.Reference Haycock²⁹ Moreover, alcohol is known to lead to a variety of nutritional disturbances including nutrient intake, absorption, utilization and excretion. Whereas some nutrients can essentially affect gene expression, the alcohol-induced nutrient disbalance may be a major contributor to impaired gene expression in FASD. Several metabolites, namely, acetate, S-adenosylmethionine, nicotinamide adenine dinucleotide and zinc are supposed to be especially relevant to alcohol metabolism and alcohol-induced liver disease.Reference Moghe, Joshi–Barve and Ghare³⁰

Prenatal alcohol exposure has been shown to reduce global DNA methylation in the developing mouse fetus. For example, in the study by Garro et al.,Reference Garro, McBeth, Lima and Lieber³¹ the acute administration of ethanol to pregnant mice during mid gestation caused reduced methylase activity and genome-wide hypomethylation of fetal DNA. Among the epigenetically regulated loci shown to be associated with FASD, one of the best characterized is differentially methylated region of the imprinted insulin-like growth factor 2 (Igf2) gene. IGF2 is thought to be a major fetal growth factor in contrast to IGF1, which is more involved in postnatal growth. IGFs are believed to play a key role in regulating energy metabolism, body size, longevity and various organ-specific functions, as well as in tumor development.Reference Novosyadlyy and Leroith³² The study from the cohort prenatally exposed to the Dutch famine 1944–1945 found that epigenetic changes, including lower Igf2 methylation which have occurred during gestation, can be persistent, potentially contributing to age-associated impairments.Reference Heijmans, Tobi and Stein³³ The regulation of Igf2 gene imprinting occurs at CpG enriched imprinting control regions, a differentially methylated region (DMR) that is variably methylated depending upon parent of origin during gonadal development and is involved in gene silencing.Reference Giannoukakis, Deal and Paquette³⁴ Recently, Haycock and Ramsay hypothesized that disruption of imprinting control in the H19/Igf2 domain may be a mechanism of ethanol-induced growth retardation – a key clinical feature of the FASD.Reference Haycock and Ramsay³⁵ In their study, paternal alleles were significantly less methylated in ethanol-treated mouse placentae in comparison with saline-treated controls. In embryos from C57BL/6J mice that are susceptible to in utero growth retardation, Downing et al.Reference Downing, Johnson and Larson³⁶ found small decreases in DNA methylation at four CpG sites in one of the DMRs of the Igf2 locus following a prenatal alcohol exposure. Only one of the four sites showed a statistically significant decrease. All Igf2 transcripts showed ∼1.5-fold decreases following the intrauterine alcohol exposure. In a more recent study by Downing et al.,Reference Downing, Flink and Florez-McClure³⁷ the prenatal alcohol exposure altered the expression of a subset of genes, including genes involved in methylation, chromatin remodeling, protein synthesis and mRNA splicing. In a mouse model for FASD, the alcohol exposure during an early neurulation induced aberrant changes in DNA methylation patterns with associated changes in gene expression.Reference Liu, Balaraman and Wang³⁸ Specifically, increased methylation of genes known to play a role in metabolism (Cyp4f13) and decreased methylation of genes associated with development (Nlgn3, Elavl2, Sox21 and Sim1), imprinting (Igf2r) and chromatin (Hist1h3d) was confirmed.

The potential role of histone acetylation in an alcohol-induced developmental programming remains largely undefined. It has been assumed that its role in linking early-life alcohol consumption to later-life outcomes is significant, since it is shown that alcohol administration causes hyperacetylation of histones in both cell culture and animal studies.Reference Moghe, Joshi–Barve and Ghare³⁰ Perhaps this is because ethanol consumption results in the increased production of acetate. Acetate is the end-product of ethanol metabolism in the liver, and it has been shown to be the mediator of alcohol-induced histone acetylation.Reference Moghe, Joshi–Barve and Ghare³⁰ The first experimental evidence that prenatal alcohol exposure can influence histone acetylation in liver of adult rat offspring was provided by the study of Yao and Nyomba.Reference Yao and Nyomba³⁹

To date, only few studies have provided the information about the role of miRNAs and their target genes in the pathogenesis of fetal alcohol syndrome. Wang et al.Reference Wang, Zhang and Li⁴⁰ determined the expression levels of 509 miRNAs in fetal mouse brains with or without a prenatal ethanol exposure. Ethanol treatment caused teratogenesis in developmentally exposed mice and mental retardation in their offspring. Among the analyzed miRNAs, miR-10a, miR-10b, miR-9, miR-145, miR-30a-3p and miR-152 were up-regulated, whereas miR-200a, miR-496, miR-296, miR-30e-5p, miR-362, miR-339, miR-29c and miR-154 were down-regulated. Sathyan et al. Reference Sathyan, Golden and Miranda⁴¹ also showed that miRNAs can mediate teratogenic effects of ethanol by using a fetal mouse cerebral cortex-derived neurosphere culture model. Among 218 miRNAs screened, the expression of four miRNAs (miR-21, miR-335, miR-9 and miR-153) was significantly suppressed at an ethanol level similar to those attained by alcoholics, whereas a lower ethanol concentration, attainable during social drinking, induced miR-335 expression. MiR-335 is a pro-apoptotic, antimitogenic factor, which decreases neural stem cells proliferation and whose actions are antagonistic to miR-21.Reference Sathyan, Golden and Miranda⁴¹ Conversely, miR-335 suppression, which is equivalent to the effect of ethanol exposure, resulted in increased neural stem cells proliferation and resistance to apoptosis.

Prenatal exposure to ethanol via maternal preference drinking induced subtle but consistent changes to the global gene expression in the adult brain of C57BL/6J mice.Reference Kleiber, Laufer and Wright⁴² A number of genes identified have previously been implicated in networks involved in cellular and tissue development, free radical scavenging and small molecule metabolism, as well as in FASD-relevant neurobehavioural phenotypes such as cognitive function (Ache, Bcl2, Cul4b, Dkc1, Ebp, Lcat, Nsdh1, Sstr3), anxiety (Bcl2), attention deficit hyperactivity disorder (Nsdh1) and mood disorders (Bcl2, Otx2, Sstr3).

A wide range of fetal abnormalities and birth defects have also been repeatedly reported in animals and humans after preconceptional alcohol exposure. The mechanisms are likely to involve alcohol-induced epigenetic changes in the gametes, resulting in the ‘FASD-like’ phenotypes in unexposed generations.Reference Abel^16, Reference Ramsay¹⁸ A link between chronic alcohol use in men and hypomethylation of paternally imprinted loci in sperm DNA in genomic regions critical for embryonic development has been demonstrated, thus providing a mechanism for paternal effects in the etiology of FASDs.Reference Ouko, Shantikumar and Knezovich⁴³ In this study, the methylation patterns of two DMRs, H19 and IG-DMR, in the sperm DNA from male volunteers, were analyzed. There was a pattern of increased demethylation with alcohol consumption at these loci along with a significant difference observed at the IG-DMR between the non-drinking and the heavy alcohol consuming groups. In the mouse study, the effects of alcohol administration in pregnant females on the methylation pattern of five imprinted genes (H19, Gtl2, Peg1, Snrpn and Peg3) in somatic and sperm cell DNAs of the male offspring were evaluated.Reference Stouder, Somm and Paoloni-Giacobino⁴⁴ A significant decrease in the number of methylated CpGs of H19 in the first generation (F1) offspring sperm and in the second generation (F2) offspring brain was observed in the alcohol-exposed group, along with a decrease in the mean sperm concentration. CpGs 1 and 2 of the H19 CTCF-binding site 2 that are important for the regulation of Igf2 gene expression were also demethylated. One other study, however, found no difference in DNA methylation in the sperm of alcohol-exposed males, indicating that the transmission of the epigenetic signal at conception may be not because of altered methylation, but may be the result of an RNA-mediated mechanism or an altered chromatin remodeling.Reference Knezovich and Ramsay⁴⁵

Maternal smoking during pregnancy

Maternal tobacco and cannabis smoking during pregnancy also has a number of well-known short-term adverse effects on offspring, such as shortened gestation and fetal growth restriction. Prenatal maternal smoking can also result in adult health problems.Reference Cupul-Uicab, Skjaerven and Haug⁴⁶ Data from rodent experiments support an association between a fetal nicotine exposure and future development of some components of the metabolic syndrome in the offspring, such as obesity, increased blood pressure and altered glucose homeostasis. In the rat model, the maternal nicotine exposure during gestation and lactation caused defective metabolic, reproductive, cardiovascular, pulmonary and hypoxia-sensing outcomes in the offspring.Reference Bruin, Gerstein and Holloway⁴⁷ Remarkably, such adverse effects of maternal nicotine exposure were not limited to the F1 offspring but also affected the F2. Specifically, F2 offspring whose mothers were exposed to nicotine during their development had elevated blood pressure, increased fasting serum insulin and an enhanced insulin response to an oral glucose challenge.Reference Holloway, Cuu and Morrison⁴⁸ The nicotine exposure only during lactation programme rodents for adipocyte hypertrophy and the further development of leptin and insulin resistance, and thyroid and adrenal dysfunction in adult life.Reference Freitas, Abreu-Villaça, Passos and Lisboa^49, Reference Lisboa, de Oliveira and de Moura⁵⁰

Human studies have also shown that children born to mothers who smoke during pregnancy are at an increased risk of obesity, hypertension and diabetes.Reference Von, Toschke, Koletzko and Slikker^51, Reference Wideroe, Vik, Jacobsen and Bakketeig⁵² Among adults, such causal link has been established in respect of several adverse outcomes such as obesity, hypertension and diabetes mellitus,Reference Cupul-Uicab, Skjaerven and Haug^46, Reference Somm, Schwitzgebel and Vauthay^53–Reference Thomas, Hypponen and Power⁵⁶ vascular damageReference Geerts, Bots, Grobbee and Uiterwaal⁵⁷ and raised cholesterol.Reference Jaddoe, de Ridder and van den Elzen⁵⁸ In utero exposure to maternal cigarette smoking can be associated with an increased risk of cancer later in life.Reference Doherty, Grabowski and Hoffman⁵⁹ Recent studies have also shown that tobacco smoking during pregnancy can permanently impair the lung structure and function in offspring; these changes can increase the risk of respiratory illness and accelerate the decline in lung function with age.Reference Maritz and Harding⁶⁰

Maternal smoking has been shown to be involved in fetal programming, and in utero tobacco exposure has been found to be associated with epigenetic changes in the offspring.Reference Rogers^61–Reference Knopik, Maccani, Francazio and McGeary⁶⁴ An altered DNA methylation and dysregulated expression of miRNA may be important modes by which maternal cigarette smoking during pregnancy influence health outcomes throughout the life course. In the Breton et al.Reference Breton, Byun and Wenten⁶⁵ study, prenatal tobacco smoke exposure influenced global and gene-specific DNA methylation in offspring. The authors concluded that life-long effects of in utero exposures may be mediated through alterations in DNA methylation, and variants in detoxification genes may modulate the effects of in utero exposure through epigenetic mechanisms. In a more recent study, it has been shown that in utero tobacco smoke exposure negatively impacts respiratory health in later life and some of the detrimental effects of this exposure are likely because of epigenetic modifications.Reference Breton, Siegmund and Kong⁶⁶ Infants born to smokers had higher methylation at the IGF2 DMR than those born to never smokers or those who quit during pregnancy.Reference Murphy, Adigun and Huang⁶⁷ The smoking-related increase in methylation was most pronounced in male offspring, for whom ∼20% of smoking-related low birth weight was mediated by DNA methylation at the IGF2 DMR. In a genome-wide survey of site-specific CpG methylation in asthmatics, the authors demonstrated evidence of wide spread epigenetic consequences of in utero smoke exposure, with differential methylation of more than 25% of all genes tested, including numerous cancer and asthma-related genes. A persistent effect of prenatal tobacco smoke on DNA methylation levels, as measured by repetitive element satellite 2 (Sat2) in adulthood, has been also found.Reference Flom, Ferris and Liao⁶⁸ Suter et al.Reference Suter, Ma and Harris⁶⁹ observed that the expression of 623 genes and the methylation of 1024 CpG dinucleotides were significantly altered among smokers, with only 38 CpGs showing significant differential methylation. They identified a significant correlation between placental transcriptional regulation and differential CpG methylation in only 25 genes among non-smokers but in 438 genes among smokers, with a dominant effect among oxidative stress pathways.

It is known that prenatal exposure to maternal cigarette smoking is associated with long-lasting variations in brain and behavior. Toledo-Rodriguez et al. Reference Toledo-Rodriguez, Lotfipour and Leonard⁷⁰ showed that epigenetic mechanisms may mediate some of the consequences of maternal smoking through methylation of DNA in genes important for brain development and associated with cognitive and affective disorders, such as the brain-derived neurotrophic factor (BDNF). They also assessed DNA methylation of the BDNF promoter in the blood of adolescents whose mothers smoked during pregnancy. Prenatal nicotine exposure was associated with higher rates of DNA methylation in the BDNF-6 exon. These results suggest that maternal smoking may lead to long-term down-regulation of BDNF expression via the increase of DNA methylation in its promoter region. Such mechanisms could, in turn, lead to modifications in both development and plasticity of the brain exposed in utero to maternal cigarette smoking.

In a more recent study, Toledo-Rodriguez et al. Reference Toledo-Rodriguez, Loyse and Bourdon⁷¹ investigated in the rat model, whether prenatal exposure to nicotine can alter kidney morphology and gene expression. The results showed that, in spontaneously hypertensive rats, the prenatal exposure to nicotine decreases kidney glomerular mass and increases renal expression of the angiotensin II type 1b receptor gene; the latter is not mediated through changes in DNA methylation of the proximal promoter of this gene. In the rats, the fetal nicotine exposure increased the sympathetic neurotransmitter release in the fetal heart and caused the programming of oncogenic protein kinase C epsilon (PKCε) gene repression through promoter methylation, linking maternal smoking to pathophysiological consequences in the offspring heart.Reference Lawrence, Chen and Xiong⁷² Perinatal nicotine exposure also altered AT 1 and AT 2 receptor expression pattern in the brain of fetal and offspring rats.Reference Mao, Zhang and Xiao⁷³ Maternal nicotine exposure had no influence on rat cytochrome P450 CYP1A1 mRNA, but resulted in a marked increase in the expression of CYP2A3 mRNA and CYP2B1 mRNA.Reference Gamieldien and Maritz⁷⁴

Several recent studies have investigated the influence of perinatal smoke exposure on miRNA expression.Reference Maccani and Knopik⁶³ The effect of the maternal cigarette smoking during pregnancy on aberrant expression of placental miRNA has been obtained in the Maccani et al.Reference Maccani, Avissar-Whiting and Banister⁷⁵ study, when maternal smoking during pregnancy was associated with the down-regulation of miR-16, miR-21 and miR-146a in the placenta. The most obvious consequence of the perinatal nicotine exposure is impaired lung structure and function;Reference Maritz and Harding⁶⁰ therefore, changes in miRNA expression levels were studied mainly in lung tissues. The down-regulation of miRNA expression in the lung of rats neonatally exposed to environmental cigarette smoke has been found in the Izzotti et al. Reference Izzotti, Calin and Arrigo⁷⁶ study. In this research, of the 484 miRNAs analyzed, 126 miRNAs (26.0%) were down-regulated twofold and 24 miRNAs were down-regulated more than threefold; only seven miRNAs (1.4%) were upregulated twofold. The most remarkably down-regulated miRNAs were miRNAs that regulate stress response, apoptosis, proliferation, angiogenesis and gene expression. In the other study by Izzotti et al.,Reference Izzotti, Calin and Steele⁷⁷ newborn mice were exposed to environmental cigarette smoke for 5 weeks, corresponding to the weaning period. Such exposure produced dramatic changes in miRNA expression, mostly in the sense of down-regulation, reflecting both adaptive mechanisms and activation of pathways involved in the pathogenesis of pulmonary diseases. In the more recent study by Izzotti et al.,Reference Izzotti, Larghero and Longobardi⁷⁸ the alterations in expression of several miRNAs (namely, miR-34b, miR-345, miR-421, miR-450b, miR-466 and miR-469) induced in neonatal mice by the exposure to mainstream smoke remained down-regulated after 1 week of cessation. Remarkably, most of these miRNAs are thought to be implicated in carcinogenesis. MiR-34b is a main effector of the p53 tumor suppressor pathway. MiR-345, miR-421 and miR-450b are involved in the inhibition of the RAS oncogene pathway. In summarizing these data, Russ and Slack suggested in their recent analytical survey,Reference Russ and Slack⁷⁹ that neonatal exposure to cigarette smoke can lead to dysregulation of miRNAs, and particularly their down-regulation, associated with the subsequent development of lung cancer. Specifically, it can cause dysregulation by modulating the p53 tumor suppression pathway. The down-regulation of miRNAs such as let-7 family members can lead to an overactivation of their oncogene targets, whereas upregulation in a few miRNAs can lead to inhibition of important tumor suppressors. Furthermore, such dysregulation may be associated with duration of smoke exposure. A short-term exposure leads to reversible changes. A longer exposure is needed for irreversibility of changes in gene expression. However, the important question is whether nicotine alone or some of the over 4000 xenobiotic compounds identified in tobacco smoke are most responsible for the induction of the epigenetic modifications in lung tissue.

Importantly, the epigenetic alterations induced by perinatal exposure to nicotine may not be limited to the F1 generation but they also may be transmitted through the germline into subsequent generations. For example, in the study by Rehan et al.,Reference Rehan, Liu and Naeem⁸⁰ the nicotine-induced impairment of lung function (including the increase in total respiratory system resistance and the decrease in dynamic compliance following methacholine challenge) was accompanied with decreased peroxisome proliferator-activated receptor gamma (PPARγ) expression, along with significant changes in the global lung and gonadal DNA methylation and histone 3 and 4 acetylation in both male and female F1 and F2 rat offspring.

Evidence has also shown that the effects of maternal marijuana smoking during pregnancy on offspring health status can be about the same as those of tobacco smoking. Ingestion of cannabis derivatives during pregnancy might result in negative consequences for the offspring in later life, since cannabinoids are able to cross the placental barrier during gestation and can also be transferred through maternal milk during lactation.Reference Dinieri and Hurd⁸¹ A number of human longitudinal studies have linked prenatal cannabis exposure with increased rates of fetal distress, growth retardation, as well as long-term behavioral abnormalities and increased vulnerability to neuropsychiatric disorders in adulthood. Studies using rodent models were able to detect a multitude of lasting neurobiological alterations in the offspring of cannabinoid-treated mothers, such as changes in different neuroendocrine and neurotransmitter systems, with the most pronounced effects reported in the dopaminergic system.Reference Dinieri and Hurd^81, Reference Navarro, Rubio and de Fonseca⁸² Continuing these studies, Campolongo et al. Reference Campolongo, Trezza and Ratano⁸³ found in their recent rodent study, that maternal exposure to even low doses of cannabinoid compounds resulted in atypical locomotor activity, cognitive impairments, altered emotional behavior and enhanced sensitivity to drugs of abuse in the adult offspring. Some of the observed behavioral abnormalities might be related to alterations in stress hormone levels induced by maternal cannabis exposure.

The long-term consequences of prenatal cannabis exposure are also assumed to be associated with alterations in epigenetic pathways such as DNA methylation and histone modification that can potentially maintain abnormal gene regulation and related behavioral disturbances.Reference Morris, DiNieri, Szutorisz and Hurd⁸⁴ Prenatal cannabis exposure decreased dopamine receptor D2 (DRD2) mRNA expression in the human ventral striatum [nucleus accumbens (Nac)], a key brain reward region.Reference DiNieri, Wang and Szutorisz⁸⁵ No significant alterations were observed for the other genes in cannabis-exposed subjects. Maternal cigarette use was associated with reduced NAc prodynorphin messenger RNA expression. Cannabis-related gene disturbances observed in the human fetus were subsequently characterized with an animal model of prenatal Δ-9-tetrahydrocannabinol exposure. The data obtained suggest that maternal cannabis use alters developmental regulation of mesolimbic D2 receptor in offspring through epigenetic mechanisms that regulate histone lysine methylation, and the ensuing reduction of D2 receptor might contribute to addiction vulnerability later in life. Wang et al. Reference Wang, Dow-Edwards and Anderson⁸⁶ found a specific reduction, particularly in male fetuses, of the D(2) mRNA expression levels in the amygdala basal nucleus in association with maternal marijuana use. This reduction was positively correlated with the amount of maternal marijuana intake during pregnancy. Moreover, the prenatal cannabis exposure was significantly associated with increased μ receptor expression in the amygdala, the reduced κ receptor mRNA in mediodorsal thalamic nucleus and reduced preproenkephalin expression in the caudal putamen.Reference Wang, Dow-Edwards and Anderson⁸⁷ These results indicate that maternal cannabis exposure during pregnancy impairs opioid-related genes in distinct brain circuits that may have long-term effects on cognitive and emotional behaviors.

Prenatal psychoactive drug exposure

Laboratory animals prenatally exposed to psychoactive drugs, including cocaine, heroin and amphetamine, exhibit shortened life span and other signs of enhanced age-related degeneration.Reference Church, Holmes, Tilak and Hotra⁸⁸ In the study by Lu et al.,Reference Lu, Liu, Long and Ma⁸⁹ a maternal cocaine or heroin administration during pregnancy caused morphological alterations in pyramidal neurons in the somatosensory cortex suggesting that prenatal administration of addictive substances may impair short-term spatial memory in adult offspring. Prenatal cocaine exposure can also alter long-term cognitive function under basal conditions in the Morris water maze and drug challenge in the radial-arm maze, birth weight and genital development.Reference Cutler, Wilkerson, Gingras and Levin⁹⁰ Cocaine-induced cognitive deficits were predominately in working memory rather than reference memory or long-term retention. Prenatal cocaine exposure also resulted in altered behavioral reactivity in adult rats in terms of the immediate and delayed response to stressful experiences.Reference Molina, Wagner and Spear⁹¹ Substantially more behavioral adaptation to chronic stress was evident in adult offspring of non-treated dams than in offspring of dams exposed to cocaine during pregnancy.Reference Campbell, Bliven and Silveri⁹² Prenatal cocaine exposure increased the heart susceptibility to ischemia-reperfusion injury in adult male but not female rats,Reference Bae, Gilbert, Ducsay and Zhang⁹³ and also induced abnormal apoptosis and myocyte hypertrophy in postnatal heart, leading to an increased heart susceptibility to ischemic insults in postnatal life.Reference Bae and Zhang⁹⁴ The data from the study by Xiao et al. Reference Xiao, Yang and Zhang⁹⁵ suggest that prenatal cocaine exposure causes reprogramming of coronary myogenic tone via changes in the Ca²⁺ sensitivity in a sex-dependent manner, leading to an increased risk of dysfunction of coronary autoregulation in adult offspring. In the other study by Xiao et al.,Reference Xiao, Huang and Xu⁹⁶ it has been found that prenatal cocaine exposure can programme vascular contractility via changes in endothelial NO synthase-regulated Ca²⁺ sensitivity of myofilaments in the sex- and tissue-dependent manners in resistance arteries leading to an increased risk of hypertension in male offspring. The results of the study by Church et al. Reference Church, Holmes, Tilak and Hotra⁸⁸ provided converging evidence that prenatal cocaine exposure caused enhanced age-related degeneration. The experimental rats prenatally exposed to cocaine showed reduced body weights in old age and 7–12% shorter life spans than the untreated control group.

The pathophysiology of the effects of cocaine on fetal development has been described along two major pathways: neurochemical effects and vasoconstrictive effects. In addition, a third pathway has been recognized concerning the effect on fetal programming.Reference Lester and Padbury^97, Reference Biliński, Wojtyła and Kapka-Skrzypczak⁹⁸ Cocaine as a stressor can likely alter the expression of key candidate genes, increasing exposure to catecholamines and fetal cortisol-altering neuroendocrine (HPA axis) activity, leading to infant behavioral dysregulation, poor behavioral control and emotion regulation in later life.Reference Lester and Padbury⁹⁷ A direct effect of cocaine in epigenetic modification of DNA methylation and programming of cardiac PKCε gene repression linking prenatal cocaine exposure and pathophysiological consequences in the heart of adult offspring was demonstrated.Reference Meyer, Zhang and Zhang⁹⁹ The results by Zhang et al. Reference Zhang, Darwanto and Linkhart¹⁰⁰ also indicate a role for DNA methylation in cocaine-mediated PKCε gene repression in the developing heart and suggest an epigenetic mechanism affecting this gene linked with vulnerability of ischemic injury in the heart of adult offspring. Fetal exposure to cocaine also caused programming of PKCε gene repression in the left ventricle of adult rat offspring.Reference Zhang, Meyer and Zhang¹⁰¹ The data from the Meyer et al. Reference Meyer, Zhang and Zhang¹⁰² study suggest that prenatal cocaine exposure causes a sex-specific loss of cardioprotection by the ischemic preconditioning in adult offspring, which is most likely because of fetal programming of PKCε gene repression, resulting in a down-regulation of PKCε function in the heart of adult male offspring.

Several studies in which animals were exposed to lower doses of either amphetamine or methamphetamine throughout pregnancy indicate that this class of drugs can also have long-term neurochemical and behavioral consequences for rodent offspring.Reference Middaugh¹⁰³ The exposure to these substances can act as a stressor that affects fetal programming, disrupts fetal placental monoamine transporter expression and alters neuroendocrine and neurotransmitter system development.Reference Middaugh¹⁰³ The behavioral effects include abnormal responding on aversively motivated tasks, and heightened motor activity which might be associated with slower habituation. Moreover, it has been found that amphetamine can have long-term effects of monoaminergic systems in the brains of adults, including neural degeneration. Like cocaine, methamphetamine could also alter the developmental trajectory of the fetus by acting on the HPA axis, disrupting fetal placental monoamine transporter expression and altering gene expression through epigenetic mechanisms, such as chromatin remodeling and DNA methylation.Reference Salisbury, Ponder, Padbury and Lester¹⁰⁴

Concluding remarks

Exposures to maternal smoking and alcohol/substance abuse during intrauterine and/or lactation periods are known to be associated with a number of short- and long-term adverse health consequences and are an increasing public health problem worldwide. A growing body of literature highlighted the key role of epigenetic mechanisms in mediating the association between exposure to harmful substances in early life and lifelong health outcomes. Several investigations have examined specific mechanisms of epigenetic regulation involved in the developmental programming of adult disease by early-life exposure to addictive substances, including tobacco, alcohol, marijuana or cocaine. To date, DNA methylation is the most studied epigenetic mechanism underlying these exposure-disease associations. In recent years, increasing attention is drawn to the role of non-coding RNAs, including miRNAs, in developing such associations.

Currently, the DOHaD research field is mainly focused on the adverse effects of maternal malnutrition. The undernutrition in early life is known to predispose to intrauterine growth restriction, low birth weight, and different aspects of the metabolic syndrome and cardiovascular disease in adulthood. Less attention was paid until recently to other potentially important programming factors including prenatal substance abuse, even though the long-term impact of these factors on health may be comparable to that of nutritional inadequacy in early life. In Fig. 2, a schematic diagram of relationships between maternal smoking, alcohol consumption and drug use during intrauterine development or infancy, and disease in adulthood, as well as the hypothetical epigenetic modulations linking early-life substance abuse to later-life health outcomes is presented. As we can see from this figure, prenatal exposure to substances of abuse may, similarly to dietary insufficiency, result in impaired intrauterine growth and subsequently lead to metabolic disorders such as type 2 diabetes and obesity, as well as to cardiovascular disease. However, there exist some discrepancy between the programming effects of malnutrition and substance abuse. The characteristic feature of developmental programming by prenatal exposure to substances of abuse is that they cause impaired brain development, as well as serious cognitive, behavioral and emotional problems in adulthood. Moreover, the early-life exposure to all these factors can lead to a cancer. In this case, the risk of cancer is associated with intrauterine growth restriction and low birth weight, while in the case of nutritional programming the increased risk of cancer development has been shown to be rather associated with fetal macrosomia (arbitrarily defined as a birth weight of more than 4000 g).Reference Vaiserman¹⁰⁵

Fig. 2

Scheme of hypothetical mechanisms linking prenatal substance abuse to later-life health outcomes. These mechanisms include epigenetic modulation of genes involved in the development of the hypothalamic–pituitary–adrenal (HPA), insulin-like growth factor (IGF), protein kinase C epsilon (PKCε), p53 tumor suppressor and brain-derived neurotrophic factor (BDNF) pathways.

Data providing conclusive evidence on the long-term consequences of maternal substance abuse for the child exposed in utero have significant clinical implications and should be considered by health associations when making policy decisions about the health risks associated with prenatal and/or neonatal exposure to alcohol, nicotine and other potentially harmful substances. A deeper understanding of the mechanisms underlying the life-long effects of prenatal drug exposure can help professionals to develop better diagnostic tools and devise appropriate therapeutic interventions to improve the condition of children perinatally exposed to drugs of abuse.