Genetic and Epigenetic Influences Associated with Intrauterine Growth Restriction Due to In Utero Tobacco Exposure

Abstract

While many fetuses are exposed to tobacco in utero, not all experience adverse outcomes as a result of this exposure. Mechanisms leading to the attenuation of fetal birth weight and adverse pregnancy outcomes are complex. Therefore many studies have begun to focus, not only on the contribution of maternal and fetal genes to phenotypic outcome, but also on epigenetic changes associated with exposure to maternal tobacco smoke. In this review, we detail the epidemiologic evidence associating an adverse pregnancy outcome to maternal tobacco use. We provide a brief summary of studies demonstrating an association between maternal and fetal gene polymorphisms with low birth weight in response to maternal tobacco exposure. We also review the literature showing epigenetic changes in the offspring associated with in utero tobacco exposure. The complex interplay of genomic and epigenomic factors may contribute to specific phenotypic outcomes and can help begin to elucidate the differential susceptibilities to tobacco smoke in utero.

While many fetuses are exposed to common known harmful substances such as alcohol and tobacco in utero, not all experience adverse outcomes as a result of this exposure. This discrepancy cannot be accounted for by dose effect alone. Thus, current efforts aimed at understanding the potential genetic, epigenetic, and metabolic basis for susceptibility to the more common exposures such as tobacco are of paramount importance in perinatal medicine. Indeed, multiple population-based studies have identified maternal tobacco use as one of the strongest modifiable risk factors for intrauterine growth restriction (IUGR). It is also implicated with other adverse pregnancy outcomes such as preterm birth, placental abruption, placenta previa, amniotic fluid emboli and maternal deep vein thrombi (1-6). In spite of increased public awareness of poor fetal outcomes associated with tobacco use during pregnancy, the prevalence of smoking during pregnancy is reported to be as high as 12-20% (7).

We now have over 40 years of clinical evidence suggesting a causal relationship between tobacco use and delivery of SGA infants. Studies dating back to 1957 by Simpson and Linda showed a significant 200 gram difference in birth weight among infants whom were exposed to as little as 10 cigarettes per day. (8) Yet the mechanisms leading to the attenuation of fetal birth weight and adverse pregnancy outcomes are complex, involving multiple epidemiologic, genetic or epigenetic, and socio-demographic factors.

In this review, we will first detail the epidemiologic evidence associating an adverse pregnancy outcome to maternal tobacco use or exposure (Summarized in Table 1). We will then examine the biologic evidence linking the clinical observation to our understanding of genomic and epigenomic regulation of gene expression. In this manner, we will seek to understand the complex interplay of genomic and epigenomic factors which relate a given clinical condition or phenotype, to susceptibility by virtue of genotype and environmental exposure.

Table 1. Summary of smoking related risks and pregnancy complications.

Pregnancy Complication	Relative Risk	Dose Response	Smoking Cessation influences risk?	Consistency between studies	References
SGA/IUGR	1.5-2.9	Yes	Yes	Yes	(8, 11, 71-73)
Preterm Birth	1.2-2.6	Yes	Yes	Yes	(2, 17, 18, 74)
Pre-eclampsia	0.5-0.7	No	?	Yes	(74, 75)
Placenta Previa	1.5-3.0	?	?	Yes	(74)
SIDS	2.0-3.0	Yes	?	Yes	(74, 76-78)
Stillbirth	1.3-1.8	Yes	Yes	Yes	(74, 79, 80)

Fetal Growth Restriction

Maternal tobacco intake during pregnancy has been causally linked to in utero growth restriction and decreased fetal birth weight in a number of population-based studies. An evaluation of more than 37,000 self identified smokers from a population-based perinatal database collected at the University of Utah (2008, (9)) supported the observation made five decades ago by Simpson and Linda: maternal cigarette use bears deleterious affect on fetal growth. The study observed a persistent association between maternal tobacco use and decreased BW ranging from 170-260 g less than that of infants of nonsmokers. These effects may be dose-related, as there is good evidence that mean birth weight decrements are greater with increased numbers of cigarettes smoked during pregnancy. This is especially pronounced in woman who smoke >10 cigarettes/day (10). The cessation of smoking during pregnancy can impact current and future pregnancies. Women who quit smoking during pregnancy have infants who weigh more than those who continue to smoke, and also have an increased likelihood of having infants who weigh the same when compared to women who never smoked (11).

So what is the harm in being growth restricted in utero? In summary of over three decades of human and animal studies, according to the fetal or developmental origins of adult disease hypothesis, perturbations in the gestational milieu influence the development of diseases later in life through the static reprogramming of gene expression via alterations in chromatin infrastructure. As discussed further below, others have previously shown that uteroplacental insufficiency induced through bilateral uterine artery ligation of the pregnant rat dam results in asymmetrical IUGR and, similar to the human, causes demonstrable postnatal disease including postnatal growth lag, neurodevelopmental delay, and later in life risk of metabolic disease, obesity, and cardiovascular disease; these alterations are associated with modifications of the fetal epigenome. It is these observations which comprise the essential elements of the developmental origins of disease hypothesis.

In addition to fetal growth restriction, other adverse fetal and infant outcomes have been associated with in utero tobacco exposure. For example, sudden infant death syndrome (SIDS) has been linked to maternal smoking and low birth weights. A British study investigated the association between smoking, growth retardation and SIDS. Evaluation of 104 SIDS victims and 206 controls showed that SIDS to be significantly related to gestation and smoking but not independently to birth weights. The study was able to conclude that most of the risk associated with SIDS was associated with growth restriction which may be accounted for maternal smoking (12).

Mechanisms leading to growth restriction following in utero tobacco exposure are poorly understood but often attributed to chronic fetal hypoxia. One predominant hypothesis is that nicotine, a principle alkaloid of tobacco smoke and a potentially harmful DNA adduct, is a key mediator of placental blood vessel constriction leading to increased apoptosis of placental syncytiotrophoblasts and resulting in fetal growth restriction. (10) This state of chronic hypoxia could potentially explain growth restriction after in utero tobacco exposure; however, it does not completely address the observation that not all infants exposed to tobacco are small for gestational age. Therefore a shift in research has focused on the potential role gene polymorphisms and epigenetic inheritance has on the development of IUGR among fetuses exposed to tobacco in utero.

Similarly, studies looking at endothelial cells of the umbilical vein from newborns of smokers showed that smoking reduced nitric oxide production in the fetal vascular bed contributing to the reduction of vasodilatory capacity and hypoxic state (13). Aagaard-Tillery et al (2010) investigated whether functional or fetal genotypes along well characterized metabolic pathways may account for smoking related risk of reduced birth weight. The study showed that fetuses with GSTT1(del) had a mean birth weight reduction of 262g among smokers suggesting an interaction between metabolic genes and in utero tobacco exposure in regard to fetal growth (14). These studies are expounded upon further below.

Preterm Birth/PROM

Preterm birth is defined as birth occurring prior to 37 weeks completed gestational age. The rate of premature birth has increased by 36% since the 1980s accounting for 12% of all live births in the United States (15). Conditions such as IUGR/ Pre-eclampsia account for more than one third of all preterm births in developed countries. Unfortunately tocolytic medicines have not been as successful in preventing preterm births and a focus on reduction of harmful behaviors is a promising intervention (16). A study performed in Sweden evaluated 311,977 live births and was able to show that smoking was most heavily associated with increased risks of very preterm birth and spontaneous preterm birth. They also validated prior studies by showing that the highest impact of smoking-related risks of preterm birth was seen on women who smoked at least 10 cigarettes/ day (2).

The mechanisms associated with smoking and preterm births are not fully understood but studies have linked impaired maternal immunity and acute inflammation in the umbilical cord and placenta to maternal tobacco use (13). In addition smoking has been linked to increased production of prostaglandins in fetal membranes which had had strong link to preterm labor and birth (17-19).

In addition to preterm birth, smoking has been linked to preterm premature rupture of membranes (preterm PROM). This process is defined as rupture of membranes prior to 37 weeks of gestation and is the second most common cause of preterm delivery. In a multicenter case control study assessing the association of preterm premature rupture of membranes and 41 potential risk factors maternal cigarette smoking increased the risk by 2.1 fold (20). This clinical finding may be explained by smoking related increased susceptibility to infection, reduction in elastic properties of fetal membranes, or even reduction in copper blood plasma which plays a important part for collagen synthesis and maintenance (21).

Pre-eclampsia

Preeclampsia is a hypertensive disorder of pregnancy associated with adverse pregnancy outcomes. However studies have shown that smoking decreases the risk of pre-eclampsia up to 33% (22) This paradoxical association is not completely understood but may also be dependent on a dose response. A study preformed at George Washington Hospital was able to show that women with salivary cotinine levels of greater than 200ng/ml had a significantly lower incidence of preeclampsia, infants with lower birth weights and a higher incidence of small for gestational age infants than women with cotinine levels of 200ng/ml or less (23). In a study looking at the accuracy of self reported cigarette smoking among pregnant women in the 1990s evaluating subjects from the CPEP trial (NICHD trial of Calcium for Preeclampsia Prevention) cotinine confirmed the report of 84.6% of women who reported smoking and 94.5% of women who denied smoking. In addition to validating accuracy of self reported cigarette smoking via cotinine levels the study was also able to show that the accuracy of self reporting did not change when compared to studies done thirty years prior in spite of increased social pressures to hid smoking habits.

The exact mechanisms through which maternal tobacco use during pregnancy reduces the risk of pre-eclampsia is not yet fully understood. Some have hypothesized the risk reduction is directly related to tobacco-mediated increased hypoxia-induced apoptosis of the syncitiotrophoblast layer and up regulation of antioxidant systems within the placenta, thereby reducing the risk of preeclampsia (22). Others have hypothesized that both smoking and preeclampsia are associated with alterations in circulating angiogenic factors. A study performed at Magee-Womens Research Institute showed that cigarette smoking is associated with lower maternal sFlt-1, a tyrosine kinase soluble angiogenic factor, during pregnancy and pre-eclampsia (24).

The relationship between cigarette smoking and preeclampsia is still not completely understood, however an interesting study looking at the risk of adverse pregnancy outcomes in preeclamptic women who smoked when compared to nonpreeclamptic women who do not smoke showed that smoking decreases the risk of preeclampsia, but smokers with preeclampsia have an increased risk of adverse pregnancy outcomes, i.e.: preterm birth (OR 5.77), abruption (OR 6.16), stillbirth (OR 3.39) (25).

Although maternal tobacco use is reportedly decreased, it still remains as one of the leading causes for preventable causes of infant morbidity and adverse pregnancy outcomes. Despite some level of knowledge of pregnant women regarding adverse effects of smoking, there is a strong need for education on smoking cessation during pregnancy. A Turkish study looking at the level of knowledge about the effects of cigarette smoking and maternal smoking status before and during pregnancy was able to show that awareness of intrauterine fetal death as a harmful effect was the single most important factor associated with quitting smoking (26). The role of practitioners educating women of childbearing age is imperative in the reduction of smoking related adverse pregnancy outcomes.

Maternal Smoking, Gene Polymorphisms and Infant Birth Weight

The carcinogenic properties of more than 4000 chemical compounds in tobacco smoke have been studied for decades (reviewed in (27)). Yet an individual's susceptibility to cancer due to smoking varies not only by tobacco dosage (years × packs per day) but by a combination of exposure and genetics. Polymorphisms of a myriad of genes involved in the detoxification of harmful chemicals and their metabolic intermediates from tobacco smoke tweak an individual's ability to metabolize different classes of potential carcinogens (Summarized in (Table 2).Similarly, not all individuals exposed to tobacco smoke in utero are growth restricted. The results of many recent studies indicate that susceptibility to IUGR from maternal tobacco exposure is likely due to the interactions of maternal and fetal genes with the environment, as well as distinct epigenetic changes to the fetus.

Table 2. Polymorphisms associated with lower birth weight with maternal tobacco exposure.

Gene	Polymorphism	Reference
CYP1A1	MspI	(36, 38, 39, 41)
GSTT1	null	(14, 36, 38, 39, 41)
NQO1	Pro187Ser; rs1800566	(43, 44)
CYP2E1	CYP2E1*5; rs3813864	(44)
GSTM1	null	(41)
AhR	Arg554Lys	(45)

While differing susceptibilities to in utero tobacco exposure are poorly understood, epidemiologic studies have recently begun to focus on polymorphisms in genes involved in processing polycyclic aromatic hydrocarbons (PAH), or addictive (nicotine) compounds into excretable, water soluble compounds. PAH can be found in the environment and the carcinogenic properties are well characterized. (28) The use of PAH as a biomarker for environmental exposure to carcinogens is well studied. (29, 30) Catalysis of PAH usually requires the coordination of two phases of enzymatic activity. The first step in processing xenobiotic compounds is metabolism by the Phase I enzymes, which include members of the Cytochrome P450 (CYP) family and Epoxide Hydrolase (31).

Phase I enzymes catalyze the conversion of PAH into reactive unstable intermediates. These intermediates, which act as reactive oxygen species (ROS) can damage DNA through the formation of harmful DNA adducts or oxidation of the base pairs. If left unrepaired, damaged DNA can readily lead to mutations during DNA synthesis. CYP2A6, a Phase I enzyme, is the major Phase I CYP family member enzyme responsible for breaking down nicotine (32) (Figure 1C). CYP1A1, another Phase I enzyme, is a major player involved in metabolism of the PAH found in tobacco smoke. (9) Many Phase I enzymes, such as CYP1A1, contain a “Xenobiotic Response Element” or XRE in the promoter. The aryl hydrocarbon receptor (AhR) is a ligand-activated transcription factor which binds xenobiotic compounds in the cell, couples with ARNT, and binds to XRE-containing genes to initiate transcription in response to toxins (Figure 1A) (33).

Figure 1. Cellular mechanisms of processing xenobiotics.

A. AhR is a receptor which, when bound to xenobiotics, heterodimerizes with ARNT and binds XRE-containing promoters to drive expression of genes involved in xenobiotic metabolism. B Phase I enzymes process PAH into unstable intermediates which can damage DNA. Phase II enzymes convert the unstable intermediates to excretable compounds. C. CYP2A6 is the major CYP family enzyme involved in processing nicotine into an excretable compound.

During Phase II, these reactive metabolites are converted to polar products. The GST family of gene products conjugates these intermediates into excreted polar electrophiles. (34) The NAD(P)H:quinone acceptor oxidoreductase 1 (NQO1) gene is a cytosolic flavoenzyme that catalyzes the reduction of quinones to hydroquinones. (35) Because the products from Phase I metabolism are potentially toxic to the cell due to their high reactivity, coordination of both phases of enzymatic activity is essential for cell survival. (Figure 1B)

The most extensively studied gene polymorphisms associated with maternal tobacco smoke exposure and fetal growth are CYP1A1 and GSTT1. A seminal study by Wang et al. in 2002 characterized the association of maternal polymorphisms in these two genes with reduced fetal birth weight and maternal smoking behavior (36). Maternal CYP1A1 genotype was determined as either AA, Aa or aa, where “a” represents the well-characterized MspI polymorphism (37). This study reports that smoking mothers with the AA genotype had offspring with an overall reduction in birth weight of approximately 252g while mothers with either the Aa or aa genotype had offspring with a reduction in birth weight of 520g. Offspring of mothers who smoked and had the GSTT1 gene showed a mean reduction in birth weight of 285g, where mothers with the GSTT1 (del) genotype had infants with a mean birth weight reduction of 642g. In mothers who smoked and were either Aa or aa for CYP1A1 and GSTT1 (del), the offspring showed a mean reduction in birth weight of 1285g (36). When comparing non smoking mothers with the AA genotypes of CYP1A1 and the GSTT1 present allele to smoking mothers with either Aa or aa CYP1A1 and GSTT1 (null) there is also a significant association with pre term delivery and lower gestational age (38, 39).

One limitation of the Wang et al. study is that it focused solely on the maternal genotype. In a study where both maternal and conceptus genotypes were analyzed, only offspring GSTT1 genotype was associated with reduced birth weight (14). In mothers who smoked, offspring null for GSTT1 showed a reduced mean birth weight of 262g compared with fetuses that had the GSTT1 allele. No associations were found between CYP2A6 or CYP1A1 genotype of the conceptus and birth weight when exposed to maternal tobacco smoke.

GSTM1 and GSTP1 are GST family members which are highly polymorphic. These polymorphisms affect the way an individual can process toxins and xenobiotics (40). The association between maternal smoking, maternal GSTM1 and GSTP1 polymorphisms and birth weight has also been analyzed. In a study of IUGR, the GSTM1 (null) polymorphism, when combined with maternal smoking, showed an increased prevalence for IUGR (41). However, in a separate study of 646 women, there was no significant association found between GSTM1 (null) genotype, maternal smoking and IUGR (42).

A study of the association between fetal genotype, fetal growth and maternal smoking behavior looked at SNPs within 5 xenobiotic metabolic enzymes (CYP1B1, CYP1A2, EPHX1, GSTP1 and NQO1) as well as 6 genes involved in folate metabolism. They report an association between maternal smoking and fetal growth only with the fetal genotype of NQO1 (43). Maternal NQO1 genotype, when combined with maternal smoking, also influences birth weight (44). Infants born to smoking mothers with two copies of the NQO1 allele had reduced birth weight. Similar effects were seen if the mother had two wild type copies of CYP2E1 (44) or AhR (45).

Maternal Smoking, Epigenetics and IUGR

During mitosis, chromatin is in its most compact form. In order to achieve this high level of organization, the cell utilizes specialized proteins to package the DNA. Chromosomes are organized into repeating units along the length of DNA called nucleosomes. Nucleosomes are made of 2 copies of each of the four histone proteins (H2A, H2B, H3 and H4) which form a core histone octamer. The DNA double helix wraps around these histones proteins to form the nucleosome (46).

The field of epigenetics builds upon the classical view of genetics by postulating that the heritable information is not only due to changes in the DNA sequence. Heritable alterations in gene expression can be attributed to various epigenetic mechanisms. Studies characterizing epigenetic changes associated with IUGR in animal models have been steadily increasing over the past 5 years. Most have focused specifically on changes in DNA methylation and histone modifications.

DNA can be covalently modified by the addition of a methyl group to cytosines within CpG dinucleotides to regulate transcription. Most CpGs within the mammalian genome are methylated. The exception is GC-rich regions of the genome with high levels of unmethylated CpGs called “CpG islands” (47). These islands usually coincide with promoters of constitutively active genes (47). DNA methylation is an essential process in imprinting and X-inactivation in mammals and is required for gene silencing (48,49).

Non-coding RNAs (ncRNAs) also epigenetically regulate gene expression. There are two classes of ncRNAs: long and small. (50) Long ncRNAs are involved in genomic imprinting, such as Xist, a long ncRNA which regulates X-inactivation (51). Small ncRNAs, such as micro RNAs, control gene expression through translational repression, mRNA degradation or chromatin modifications (50).

The addition of post-translational modifications to the histone proteins is another means of modifying gene expression without altering the underlying genetic code. These modifications, such as methylation, acetylation and phosphorylation, make up the histone code. (52,53). Patterns of different combinations of histone modifications can be read by downstream “machinery” to promote DNA damage repair, gene expression, or the recruitment of silencing factors (53). Generally, these modifications are associated with specific chromatin states. Acetylation of Lysine 14 of histone H3 is associated with maintaining a euchromatic, or open, chromatin state (54). This allows the transcriptional machinery to access the DNA. Trimethylation of lysine 9 of histone H3, however, is associated with a heterochromatic state, inhibiting access to the transcriptional machinery and promoting gene silencing (54).

The correlation between IUGR and epigenetic changes in the offspring is a field that has recently been gaining attention. IUGR associated with maternal tobacco smoke is usually attributed to uteroplacental insufficiency. A surgical method of introducing IUGR in the offspring of rats has been well established. Bilateral uterine artery ligation (BUAL) induces uteroplacental insufficiency and retards fetal growth. (55) This model system has been used to study the correlation between IUGR and DNA methylation and histone modification patterns in the offspring.

IUGR is associated with adult onset of diabetes. Using the BUAL method in rat, the methylation status of DNA within pancreatic islets of IUGR offspring was determined. Growth restricted offspring had differential methylation in approximately 1400 CpG sites compared to sham surgery controls at 7 weeks of age, before the onset of adult diabetes. (56) Studies using BUAL induction of IUGR in rat hippocampus and periventricular white matter show an overall reduction in CpG island methylation and increased H3K9 and H3K14 acetylation. (57) Expression of chromatin regulators such as DNA methyltransferase I (DNMT1) and methyl-CpG binding protein 2 (MECP2) and histone deacetylase I (HDAC1) were also altered in IUGR animals. Acetylation of histone H3 was also altered in liver from IUGR pups (58, 59).

Gene specific epigenetic alterations were also seen within the hepatic IGF1 gene in IUGR rats.(60) Fu et al. reported that the histone code was altered along the length of the IGF1 gene. The epigenetic changes within the Pdx1 gene in the pancreas in rat IUGR offspring from birth to adulthood were characterized (61). Pdx1 expression was permanently reduced in IUGR offspring and this reduction was characterized by changing histone modification and DNA methylation patterns within the promoter region.

Within the past few years the study of epigenetic changes die to maternal smoking exposure has begun to gain momentum. The fetal origins of adult disease hypothesis postulates that adverse exposures in utero make an individual susceptible to the adult onset of metabolic disease, such as cardiovascular and diabetes (62-64). The possibility remains that exposure to maternal smoking in utero may alter the fetal epigenome. The memory of in utero exposure may be maintained by epigenetic mechanisms.

A study in 2008 determined global methylation in 73 fathers, 69 mothers and 156 offspring (65). Although smoking per se did not affect global methylation, a significant correlation between paternal DNA methylation and that of the offspring was lost if either the father or offspring ever smoked. Smoking may therefore affect global DNA methylation. In a study of global DNA methylation in 85 blood samples from women as part of the National Collaborative Perinatal Project in N Y, prenatal smoke exposure showed a statistically significant association with higher levels of DNA methylation in adult women (66).

Global DNA methylation changes due to in utero tobacco exposure were studied as part of the Children's Health Study (67). Global methylation analysis of samples from 272 children in kindergarten and first grade was performed. Children exposed in utero to maternal smoking showed significantly lower methylation of AluYb8. Children with the GSTT1 null genotype who were exposed in utero also showed differences in LINE1 methylation. Differential methylation was also found within 8 genes due to prenatal smoking exposure.

Global DNA methylation status in newborn cord blood is associated with cord blood serum cotinine levels (68). In a study of 30 newborns with high, low or very low levels of cotinine in the serum, global DNA methylation showed an inverse relationship with cord blood cotinine levels. Global DNA methylation levels were lowest in newborns with the highest serum cotinine levels.

In a study of gene specific changes of DNA methylation in placenta from smoking and non-smoking mothers, differential methylation of the proximal promoter of the Phase I enzyme, CYP1A1 was found (69). CYP1A1 expression is upregulated in placenta from smoking mothers (69, 70). Bisulfite sequencing analysis was performed for 59 CpG sites in the CYP1A1 proximal promoter. A region surrounding the XRE within the CYP1A1 promoter was found to be hypomethylated in placenta from smoking mothers. This hypomethylation was significantly correlated with the decrease in gene expression.

These studies have all focused on DNA methylation. Already it appears that in utero exposure to maternal tobacco smoke causes changes in DNA methylation patterns in the offspring. More detailed analysis are in order to determine if other epigenetic changes such as histone modifications or non-coding RNAs are affected as well. The significance of the changes in DNA methylation patterns needs to be determined.

Summary

Although our discussion herein is unique to maternal tobacco exposure and sequeale in the fetus, it is not without broader implications to understanding biologic plausibility and merit of studies on exposure and genotype. First, while mechanisms leading to fetal growth restriction following in utero tobacco exposure have generally often been attributed to chronic fetal hypoxia, nicotine, cotinine, and potentially harmful DNA adducts are known to cross or collect in the placenta of smokers. As discussed, phase I gene-products, such as CYP1A1, are integral in metabolic activation of PAH compounds into oxidized derivatives, resulting in reactive oxygen intermediates capable of covalently binding DNA to form adducts; as a balance to such intermediary forming reactions, conjugation with endogenous species to form hydrophilic glutathione conjugates which are then readily excretable occurs. Thus while it is possible that chronic hypoxia is a primary mediator of fetal growth restriction in response to in utero tobacco exposure, our data supports the notion that it is equally likely that the discrepant variation in fetal susceptibility to smoking-related growth restriction results from the collective effects along the phase I and phase II PAH metabolic pathways. We demonstrate in our analyses that that an imbalance between maternal/placental-mediated increases in reactive intermediates (e.g., phase I CYP1A1 promoter CpG hypomethylation in the XRE-primed region) alongside diminished ability of the fetus to in turn excrete these reactive intermediates (fetal phase II GSTT1(del)) may mediate fetal growth potential in response to in utero tobacco exposure.

There is an increasing recognition of the role genetic and epigenetic mechanisms play in modifying individual susceptibility to perinatal health and disease. Our discussion recognizes such mechanisms. The implications of our discussion are two-fold. First, it illustrates that a fetal metabolic gene (GSTT1) which is integral in the excretion of reactive intermediates of aromatic hydrocarbons modifies fetal growth specifically in response to in utero tobacco exposure. These findings imply that tobacco metabolites may reach the fetus and thus modify fetal growth if not excreted. Second, future studies aimed at illuminating the complex interplay of genomic-epigenomic-environmental interactions may help dissect multifactorial etiologies and identify at-risk populations for the common adverse health and disease outcomes in women and children, and lead to a deeper understanding of later in life disease.