The Plausibility of Maternal Nutritional Status Being a Contributing Factor to the Risk for Fetal Alcohol Spectrum Disorders: The Potential Influence of Zinc Status as an Example

Abstract

There is increasing evidence that human pregnancy outcome can be significantly compromised by suboptimal maternal nutritional status. Poor diet results in a maternal-fetal environment in which the teratogenicity of other insults such as alcohol might be amplified. As an example, there is evidence that zinc (Zn) can interact with maternal alcohol exposure to influence the risk for fetal alcohol spectrum disorders (FASD). Studies with experimental animals have shown that the teratogenicity of alcohol is increased under conditions of Zn deficiency, while its teratogenicity is lessened when animals are given Zn supplemented diets or Zn injections prior to the alcohol exposure. Alcohol can precipitate an acute phase response resulting in a subsequent increase in maternal liver metallothionein, which can sequester Zn and lead to decreased Zn transfer to the fetus. Importantly, the teratogenicity of acute alcohol exposure is reduced in metallothionein knockout mice, which can have improved Zn transfer to the conceptus relative to wild-type mice. Consistent with the above, Zn status has been reported to be low in alcoholic women at delivery. Preliminary data from two basic science and clinical nutritional studies that are ongoing as part of the international Collaborative Initiative on Fetal Alcohol Spectrum Disorders (CIFASD) support the potential role of Zn, among other nutritional factors, relative to risk for FASD. Importantly, the nutrient levels being examined in these studies are relevant to general clinical populations and represent suboptimal levels rather than severe deficiencies. These data suggest that moderate deficiencies in single nutrients can act as permissive factors for FASD, and that adequate nutritional status or intervention through supplementation may provide protection for some of the effects of prenatal alcohol exposure.

1. Introduction

It is estimated that, conservatively, 50% of human concepti are lost before or during implantation, and of those that successfully implant, an additional 15–20% are lost prior to delivery [1,2]. With respect to completed pregnancies, approximately 3% result in a child with one or more severe malformations. These numbers, as discouraging as they are, do not take into account the increased risk many children can have for chronic diseases later in life, including obesity, diabetes, hypertension and vascular disease, as a consequence of select prenatal or early postnatal insults [3,4]. While a diversity of factors can contribute to the occurrence of developmental abnormalities, there is increasing evidence that human pregnancy outcome can be significantly compromised by suboptimal maternal nutritional status, and that poor nutrition may be a leading cause of preventable birth defects. A number of investigators since the early 1940’s have reported that women who consume diets that can be classified as “poor” have an increased risk for pregnancy complications compared to women who consume diets that can be classified as “good” (Table 1). It is worth noting that this association has been reported over a period of time where many would argue as to what constitutes a good versus a poor diet; however, a consistent theme over the past seven decades has been that “good” diets are characterized by high micronutrient content. Importantly, data from a variety of nutrition intervention trials, ranging from the provision of whole foods, to multivitamin-mineral supplements, to single nutrients such as folate and iodine, provide evidence that improvements in a mother’s diet can result in marked reductions in her risk for a complicated pregnancy [5–11]. When viewed in its totality, the data supporting the concept that maternal nutritional status is a key predictor of human pregnancy outcome seem overwhelming; however, a complication in the story is that while the diets of women who consume “poor’ diets may not be ideal, pronounced essential nutrient deficiencies are rare. This observation has led to the idea that the increased risk for pregnancy complications observed in this group of women may, in many cases, be more due to the fact that the poor diet results in a maternal-fetal environment in which the teratogenicity of other insults is amplified, than due to a simple deficit of one or more essential nutrients in the diet.

Table 1.

Maternal Diet and Pregnancy Outcome: “Good” vs. “Poor” Diet

Author	Study
Ebbs et al., 1941	Intervention
Burke et al., 1943	Observational
Jeans et al., 1955	Observational
Primrose and Higgins, 1971	Intervention
Laurence et al., 1983	Intervention
Friel et al., 1995	Observational
Wright, 1995	Observational
Torfs et al., 1998	Observational
Velie et al., 1999	Observational
Di Cintio et al., 2001	Observational
Krapels et al., 2004	Observational
Rees et al., 2005	Observational
Lam and Torfs, 2006	Observational
Knudsen et al., 2008	Observational
Yang et al., 2008	Observational
Baker et al., 2009	Observational
Haggarty et al., 2009	Observational

In the current paper, we explore this concept using alcohol as an example of a developmental insult whose teratogenicity can be modulated by maternal diet. It is now widely accepted that excessive maternal alcohol intake during pregnancy can result in a number of developmental abnormalities commonly referred to as the Fetal Alcohol Syndrome (FAS), or more recently, the Fetal Alcohol Spectrum Disorder (FASD) [12]. FASD is now thought to be one of the most common causes of developmental mental retardation in the human population. Due to space constraints we will focus our discussion on the ability of zinc (Zn) to modulate the teratogenic expression of alcohol. In the last section of this paper, preliminary data from an ongoing multi-nutrient supplementation trial in Russia and the Ukraine with women who are at high risk for having a child with FASD will be discussed. We would like to emphasize that while our comments are focused on Zn, there is good evidence that numerous other nutrients, including copper (Cu), iron (Fe), magnesium (Mg), selenium (Se), methionine, choline, vitamin B₁₂ and folate, can modulate alcohol’s developmental toxicity. Owing to space constraints, review articles are cited in several instances and the reader is directed to them for additional references.

2. Alcohol-Zn Interactions

Although the concomitant effects of alcohol and specific nutrient deficiencies on the developing brain are not well understood, animal studies have shown that compromised nutrition can exacerbate ethanol’s teratogenic effects. Many adverse effects of prenatal alcohol exposure including low birth weight [13], physical anomalies [14], brain damage [15] and reduced IGF levels [16] have been reported to be more severe when the alcohol is consumed along with poor diets; however, a complication of this finding is that blood alcohol levels are often higher among malnourished subjects than in subjects thought to be characterized by similar alcohol intakes, but better diets. On the positive side, results from early animal studies suggest that select nutritional supplements can attenuate some of the effects of prenatal alcohol exposure, although the effectiveness depends on many factors, including the level of alcohol exposure and the outcome measure [17]. The contribution of select nutritional factors to the risk for FASD in humans has not been well characterized, however, the overall nutritional status of heavy drinkers is generally recognized to be poor [18]. It is critically important to define the role that malnutrition or under-nutrition plays in the risk for FASD. Of particular interest in the pregnant heavy drinker is assessing the status for those micronutrients that are critical for normal neurulation and development of the central nervous system (CNS). From work with experimental animals, it is well documented that deficiencies of certain nutrients, including folate, vitamin B₁₂, Zn, Fe and Cu, during pregnancy can result in abnormal CNS development, and deficiencies of these nutrients are commonly noted in alcoholics [18–23].

With respect to the above nutrients, the hypothesis that maternal Zn status is an important predictor of the risk for FASD has received particular attention. Over 25 years ago, Flynn et al reported that maternal plasma Zn and fetal cord plasma Zn were lower in pregnant women who consumed alcohol versus non-alcohol drinking women [24]. Importantly, these investigators reported that there were negative correlations between maternal plasma Zn concentrations and the severity and frequency of birth defects in the infants, suggesting an etiologic role for Zn deficiency in human FASD. Consistent with the idea that maternal Zn status might be a predictor of the risk for FASD are the early data of Miller et al [25]; these investigators reported that the reproductive toxicity of alcohol in rats was elevated in dams fed marginal Zn diets (10 µg Zn/g diet) compared to that in dams fed diets with control amounts of Zn (45 µg/g) [25]. It is important to note that in the above study by Miller et al, the amount of Zn that was provided in the diet of the marginal Zn group would not be viewed as “deficient”, as the consumption of this diet throughout pregnancy would not typically result in marked developmental anomalies. Consistent with the data of Miller et al, the teratogenicity of alcohol has been reported to be amplified in pregnant rats and mice fed Zn deficient diets (1 µg Zn/g) [26–28].

The relevance of the above work by Miller et al is underscored by the observation that in certain human populations, dietary Zn intake during pregnancy can be well below current recommended dietary intakes [29–32], suggesting that suboptimal Zn status is common in the human population. The concept that a deficit of Zn during early development presents a risk to the human conceptus is reasonable, given that in experimental animals, the teratogenicity of severe Zn deficiency is well documented, resulting in malformations that affect multiple organ systems including the CNS [33]. Even moderate deficiencies of this nutrient during development can result in persistent adverse effects on the immune system and neurobehavioral abnormalities [2,34,35]. Mechanistically, Zn deficiency is thought to influence embryonic and fetal development through multiple mechanisms including abnormal nucleic acid metabolism, reduced protein synthesis, impaired cell migration and cell signaling due to alterations in the cytoskeleton secondary to impaired tubulin polymerization, excessive cellular oxidative stress, reduced binding of transcription factors and hormones that are dependent on Zn-finger regions, and reductions in insulin and IGF-signaling [2,23,34,36–40] (Table 2; Figure 1).

Table 2.

Potential mechanisms of Zn deficiency teratogenicity

Indirect

Zinc deficiency-induced changes in maternal metabolism

Direct

Impaired DNA replication and transcription

Chromosomal damage

Decreased binding of growth factors and hormones to their receptors

Excessive oxidative damage

Altered patterns of cell death

Altered cell signaling

Impaired cell migration

Low activities of numerous zinc enzymes

Figure 1.

Alcohol-induced acute phase response (APR) and precipitation of conceptal Zn deficiency. An APR can also lead to changes in the metabolism of other essential nutrients. Multiple mechanisms underlie the teratogenicity of alcohol.

It is important to note that mechanistically, one would predict multiple synergistic interactions between an alcohol insult and a condition of marginal Zn deficiency. As an example of the above, it is well documented that Zn contributes to the oxidant defense system through multiple means, including through its ability to: (1) regulate Cu-Zn superoxide dismutase (CuZn SOD) activity; (2) regulate metallothionein levels; (3) protect sulfhydryl groups from oxidation; (4) modulate intracellular thiol groups; and (5) inhibit the binding of redox active metals, such as Fe and Cu, to intracellular sites where they can generate free radical reactions (e.g., Fenton-type reactions). Given the above, it is evident that one functional consequence of Zn deficiency is an increased susceptibility to exogenous oxidative stressors, such as smoking, endotoxin challenge, and, particularly germane to this paper, alcohol [27,39,41]. The consequences of excessive tissue oxidative stress in the embryo can include lipid, protein and DNA oxidative damage, and an increase in apoptosis, all of which can trigger abnormal development. It is important to note that all of the above are common findings in animal models for FASD [23].

Another potential point of interaction between Zn deficiency and an alcohol challenge involves sonic hedgehog (Shh) signaling. Shh signaling is critical for polarizing activity, and Shh null fetuses are characterized by a postaxial forelimb ectrodactyly in mouse models [42]. Shh is a Zn-dependent developmental trigger [43], and reduced Shh expression has been implicated in ethanol-induced postaxial forelimb ectrodactyly in the mouse [44]. Schreiner and co-workers [45] have suggested that a state of embryonic Zn deficiency secondary to an alcohol-induced acute-phase response (see below; [46–48]) in the mother results in reduced Shh signaling with subsequent dysmorphology [45]. This is an interesting hypothesis that merits further investigation. Importantly, if it is shown to be correct, there are numerous other Zn-dependent developmental proteins, many in the hedgehog signaling pathway (e.g., Gli Zn finger transcription regulators), that might also be affected through the mechanism described above.

A third finding that is common to experimental Zn deficiency, and alcohol challenges during pregnancy, is a reduction in IGF levels and action [49,50]. Zn deficiency can alter IGF-1 and IGF-binding protein metabolism in maternal and fetal blood [49] and may contribute to growth retardation of the offspring. Growth deficit is also commonly noted in offspring of alcohol-exposed women. Administration of ethanol by intubation (5.25 g/kg/day) on postnatal days 4–9 results in motor coordination impairments that are significantly improved with intranasal administration of IGF on postnatal days 10–13 [50].

Finally, a common finding in animal models for developmental Zn deficiency and FASD is an elevated occurrence of apoptosis in the embryo and fetus. Multiple mechanisms have been implicated in the inducing-effects of alcohol and Zn deficiency on apoptosis including disruptions in growth factor signal transduction pathways mediated by receptor tyrosine kinases, an increase in the expression of caspase-3, a down-regulation of NF-κB-dependent anti-apoptotic genes, and an increase in cellular oxidative stress. [36,38,51].

Evidence that severe Zn deficiency presents a reproductive risk in humans is provided by the observation that women with a congenital disorder in Zn absorption (acrodermatitis enteropathica; AE) have a very high risk for pregnancy complications unless they are given dietary Zn supplements [2]. It is now recognized that the AE is due to a defective Zn transporter. The analogue of this transporter in mice is Zip4, and consistent with the human literature, homozygous Zip4-knockout mouse embryos die during early embryogenesis and are characterized by multiple defects [52]. As is depicted in Table 3, a number of studies have reported that low plasma Zn concentrations in the first or third trimester is associated with an increased risk for several pregnancy complications, including birth defects and growth retardation. In further support of the concept that maternal Zn status can be a predictor of pregnancy complications, numerous investigators have reported that even in non-AE populations, the provision of dietary Zn supplements during pregnancy results in a reduced risk for pregnancy complications (Table 3). However, as is also provided in Table 3, there are several studies in which the provision of Zn supplements during pregnancy had no measurable positive effects. Indeed, in a recent Cochrane Review on the effects of Zn supplementation on pregnancy outcome [53], the authors concluded that with the exception of the risk for prematurity, there was no consistent positive effect of Zn supplements during pregnancy. In our opinion, the somewhat negative finding by Mahomed et al [53] is due to the fact that in many cases, the Zn supplementation trials have been done with relatively healthy, non-stressed populations. This may be important since, in addition to low dietary Zn intake, stressor-induced changes in the metabolism of Zn and other nutrients are often secondary to an acute phase response (APR), which can be triggered by a diverse set of cytokines that are released following tissue injury.

Table 3.

Evidence for an influence of Zn on human pregnancy outcome

Poor pregnancy outcome in women with acrodermatitis enteropathica

Hambidge et al., 1975

Low Zn status (low blood and hair Zn concentrations) is associated with an increased risk for complications

Jameson

Meadows et al.

Cavdar et al.

Neggers et al.

Srinivas et al.

Scheplyagina

Golalipour et al.

Zn supplements are associated with improved pregnancy outcome

Kynast and Saling, 1986

Cherry et al., 1989

Cavdar et al., 1991

Goldenberg et al., 1995

Osendarp et al., 2001

Saaka et al., 2009

Hozyasz et al., 2009

Danesh et al., 2009

Zn supplements do not reduce pregnancy complications

Hunt et al., 1985

Mahomed et al., 1989

Jonsson et al., 1996

Caulfield et al., 1999

Hafeez et al., 2005

2.1. Acute Phase Response-Induced Fetal Zn Deficiency

It has been hypothesized that the developmental toxicity of a wide variety of toxicants and environmental insults is mediated in part through the induction of the APR, which increases maternal hepatic metallothionein synthesis and Zn sequestration, leading to reduced Zn transfer to the conceptus. If severe enough, this can result in abnormal development [2,46,54–56] (Figure 1). The potential human relevance of the above work with experimental animals is illustrated by the finding that pregnant women who are infected with cytomegalovirus, a common etiologic agent of intrauterine infection, are characterized by high cytokine levels including TNF-α and IL-6, and lower than normal plasma Zn concentrations [57]. Underscoring the potential significance of the above finding is the recent report by Collier et al [58] that maternal infections during pregnancy are common. In rats, disease and environmental factors that reduce maternal serum Zn concentrations can disrupt Zn-dependent processes in the embryo and produce developmental defects, even when the mother consumes a Zn “adequate” diet [54,59]. Significantly, the teratogenic effects of the APR-induced hypozincemia can be amplified by marginal Zn diets, and reduced with Zn supplementation [46,54]. Critical to the current paper, in rodent models, acute alcohol exposure is associated with APR-induced reductions in fetal Zn uptake [48,54–56,60,61]. That the above reductions in fetal Zn uptake are functionally significant is suggested by the observation that the teratogenicity of alcohol is lessened when animals are given Zn prior to the alcohol exposure [60–63]. Moreover, the teratogenicity of acute alcohol exposure is reduced in metallothionein knockout mice compared to metallothionein wild-type mice [55].

It is important to note that in the above discussion, the focus has been on the potential positive effects of Zn supplementation on the expression of FASD in the offspring. Zn supplementation has also been reported to be of value in attenuating ethanol-induced liver damage in the adult [64–66] and mitigating lung epithelial and macrophage dysfunction induced by chronic alcohol intake [67]. Thus, it can be speculated that Zn supplementation of the high risk patient could directly benefit the mother, as well as her child.

In addition to pregnant women who are exposed to alcohol, infants with FASD have been reported to have low plasma Zn concentrations and higher twenty-four hour urinary Zn excretions compared to normal controls, indicating altered Zn homeostasis [68]. Interestingly Murillo-Fuentes and coworkers [69] reported that rat pups whose mothers were given alcohol during the period of lactation were also characterized by higher than normal levels of urinary Zn excretion. It is important to stress that the above reports are preliminary in nature. However, if the finding can be replicated, it represents a critical observation as it could provide one explanation for the persistent immunological abnormalities that have been reported for some children with FASD [70,71]. If these persistent immunological abnormalities are secondary in part to persistent abnormalities in Zn metabolism, they could be responsive to Zn supplements. In support of this concept, Zn given by injection [72], as well as the feeding of diets high in Zn [73], have been reported to reduce neurobehavioral abnormalities in mouse and rat offspring of ethanol-exposed dams. While these data are compelling, others, using a rat model, did not observe a protective effect of Zn supplementation on alcohol-induced developmental brain abnormalities [74]. In addition, it is important to note that there can be adverse interactions among nutrients. For example, supplementation of an alcohol-containing liquid diet with 300 µg/ml of Zn resulted in severe fetal Cu deficiencies relative to controls [75]. Cu deficiency during pregnancy is also teratogenic, affecting cardiac, vascular, neurological, pulmonary, skeletal, and immune systems [21,76,77].

Ethanol exposure has also been shown to alter Fe regulation and homeostasis. Chronic ethanol consumption increases body stores of Fe and is associated with a significant risk of Fe overload [78–81]. Levels of non-transferrin-bound Fe are higher in alcohol abusers [78], which could contribute to reactive oxygen species formation via its involvement in Fe-catalyzed Fenton reactions [82]. When antioxidants, Fe chelators or sulfhydryl compounds that increase cellular glutathione are administered, ethanol toxicity is significantly reduced [83]. Miller and coworkers [84] reported that ethanol consumption by rat dams perturbs the temporal patterns between Fe concentrations and Fe-regulatory proteins in brain regions of offspring. That fetal alcohol exposure can result in low Fe stores in the human infant has been reported by Carter et al [85].

It is critical to note that the metabolism of Fe, Zn and Cu are interrelated, and it has been demonstrated in experimental animal models that a maternal deficit of any one of the above elements can result in alterations in the metabolism of the other elements in the mother as well as the fetus [86]. The above observation underscores the potential risk of focusing on a signal nutrient when one is studying alcohol-nutrient interactions with respect to the risk for FASD. In this regard, it is important to note that while hypozincemia is a hallmark of an APR, as is depicted in Figure 1, an APR can also result in marked disturbances in the metabolism of Fe, Cu, folate and vitamins, as well as other essential nutrients. The extent to which APR-induced changes in theses nutrients might affect fetal development has been largely unexplored.

3. Testing of the Zn/Alcohol Hypothesis: The Need for Large-Scale Prospective Trials

There is a need for multiple prospective studies investigating whether alterations in maternal Zn status can affect fetal outcome in alcohol-exposed women. These studies should be designed in a way that they could later be analyzed via meta-analysis as is done in Cochrane reviews to determine whether suboptimal Zn status is a predictor of adverse fetal outcomes. Moreover, maternal nutrient supplementation could be instituted as this is easily modifiable. Towards this goal, a longitudinal prospective study that aims to examine maternal nutritional status and its contribution to risk for FASD, as well as to test a multi-micronutrient intervention with or without choline supplementation, is currently underway at sites in Ukraine, as part of the Collaborative Initiative on Fetal Alcohol Spectrum Disorders (CIFASD). The study protocol was approved by institutional review boards in Russia and Ukraine and institutional review boards at the University of California, San Diego and the University of California, Davis; all study participants provided informed consent.

In the above study, alcohol-exposed and comparison group subjects are selected from prenatal patients at each of the two sites who were screened by prenatal care providers at first prenatal visit, using a short standardized screening tool. Subjects are considered eligible for the alcohol-exposed group based on quantity and frequency of alcohol consumption during pregnancy. A positive screen for quantity and frequency of alcohol consumption is defined as at least four episodes of five or more standard drinks, at least five episodes of 3–4 standard drinks, or at least 10 episodes of 1–2 standard drinks either in the month around the time of conception, or the most recent month of pregnancy. Subjects are considered eligible for the comparison group based on minimal to no alcohol consumption during either time period, as reported at the time of the initial screening.

All eligible alcohol-exposed women are invited to enroll, and comparison group women are recruited in an approximate 1:1 ratio. This is accomplished following enrollment of each alcohol-exposed woman by approaching the next pregnant woman presenting for prenatal care who reports low to no alcohol exposure in response to the screening questionnaire and who agrees to participate in the study.

As part of the pregnancy follow-up procedure, enrolled women participate in comprehensive standardized interviews, newborn physical examinations, and standard neurobehavioral testing of infants. For the nutrition component of this study, all subjects are randomized to receive a nutritional intervention involving three arms: multi-micronutrient supplement provided from time of enrollment to delivery, multi-micronutrient supplement with choline supplement provided from the time of enrollment to delivery, or no treatment (current standard of care in Russia and Ukraine). Maternal blood samples are drawn at the time of enrollment to establish baseline nutritional status, and again in the 3^rd trimester to evaluate change in status, and to validate the impact of treatment group on change in nutritional status.

The samples collected to date in Russia have been analyzed in the laboratory of one of the authors (AS) at the Institute for Biotech Medicine in Moscow. The samples collected to date from the Ukraine have been analyzed by three of the authors in the U.S. at the University of California, Davis (CLK, JYUA, KG). Due to potential variability in the laboratory procedures, and to differences in the characteristics of subjects, preliminary data from these two sites are presented separately. Baseline maternal blood samples were collected in heparinized tubes and analyzed for Zn, Cu, Mg, Fe and Ca by inductively coupled plasma optical emission spectrometry (ICP) (Trace scan; Thermo Elemental, Franklin, MA) [87]. Certified reference solutions (QC 21, Spec Centri Prep, Metuchen, NJ) were used to generate standard curves for each element. National Bureau of Standards reference samples were included with each run to ensure accuracy and reproducibility.

Preliminary data on analysis of mineral status at the time of enrollment is available for a total of 69 subjects from the two sites. Women in the alcohol-exposed groups at the two sites were similar in age, primigravidity, and years of education to their respective no or low-exposed groups, but more likely to be single mothers and to be current smokers relative to their comparison groups (Table 4).

Table 4.

Characteristics of pregnant women enrolled in nutrition and FASD study in Russia and Ukraine

	Russian Sample		Ukraine Sample
Maternal Characteristic	Alcohol Exposed N = 10	Alcohol Unexposed N = 10	Alcohol Exposed N = 28	Alcohol Unexposed N = 21
Maternal Age – years, mean (SD)	27.7 (8.0)	27.5 (6.0)	23.9 (4.6)	24.2 (4.9)
Primigravid n (%)	5 (50.0)	5 (50.0)	11 (52.4)	19 (67.9)
Education – years, mean (SD)	14.7 (2.2)	16.4 (2.0)	13.0 (2.4)	14.1 (1.6)
Marital Status n (%)
Single	4 (40.0)	1 (10.0)	7 (25.0)	2 (9.5)
Married	2 (20.0)	8 (80.0)	21 (75.0)	19 (90.5)
Separated	3 (40.0)	1 (10.0)	0	0
Smokers n (%)	8 (80.0)	2 (20.0)	7 (25.0)	1 (4.8)

Table 5 describes the characteristics of maternal alcohol use as reported at the time of enrollment regarding the month around the time of conception. Consistent with the group selection criteria, women in the alcohol-exposed groups were predominately binge drinkers, with almost no women reporting daily drinking, even in small amounts.

Table 5.

Reported alcohol consumption in month around time of conception among pregnant women enrolled in nutrition and FASD study in Russia and Ukraine

	Russian Sample		Ukraine Sample
# Drinks per Occasion in Month	Alcohol Exposed N = 10	Alcohol Unexposed N = 10	Alcohol Exposed N = 28	Alcohol Unexposed N = 21
5 or more n (%) # of occasions	4 (40.0) 0–5	0	12 (42.9) 0–3	0
3 or 4 n (%) # of occasions	10 (100) 2–5	0	24 (85.7) 0–10	0
1 or 2 n (%) # of occasions	9 (90.0) 0–7	7 (70.0) 0–3	28 (100) 1–28	7 (33.3) 0–2
Daily n (%)	0	0	2 (7.1)	0

As shown in Table 6, although numbers are small in each sample, for most of the minerals, consistently lower mean values were observed in the alcohol-exposed groups than in the controls; these differences were statistically significant for Zn at both sites, and for Cu at the Ukraine site.

Table 6.

Maternal plasma mineral concentrations at enrollment among pregnant women participating in nutrition and FASD study in Russia and Ukraine

Mineral Mean (SD)	Russia Sample			Ukraine Sample
	Alcohol Exposed N = 10	Alcohol Unexposed N = 10	p-value*	Alcohol Exposed N = 28	Alcohol Unexposed N = 21	p-value*
Ca (µg/mL)	88.5 ± 3.3	93.8 ± 4.5	0.348	83.4 ± 1.2	83.5 ± 0.7	0.915
Cu (µg/mL)	2.2 ± 0.2	2.4 ± 0.2	0.409	1.7 ±0.1	1.9 ± 0.1	0.043
Fe (µg/mL)	1.4 ± 0.3	1.1 ±0.1	0.342	0.78 ± 0.06	0.82 ± 0.05	0.637
Mg (µg/mL)	18.1 ± 0.7	19.0 ±1.0	0.489	15.9 ± 0.02	16.2 ± 0.2	0.187
Zn (µg/mL)	0.59 ± 0.04	0.73 ± 0.06	0.073	0.57 ± 0.02	0.64 ± 0.05	0.022

^*t-test appropriate for equal or unequal variances

4. Summary and Concluding Comments

The above comments have been aimed at the overarching hypothesis that the occurrence of numerous features of FASD, including growth deficiency, structural features, persistent immunological defects and neurobehavioral impairment, is influenced by the nutritional status of the mother, as well as the conceptus. Current data support the concept that select micronutrient deficiencies increase the risk for the occurrence of FASD in high-risk populations. In theory, these nutritional deficiencies can arise as a consequence of poor diets, as well as a consequence of tissue injury-induced alterations in the metabolism of select nutrients. If the above concepts are correct, it is reasonable to predict that the use of select micronutrient supplements could reduce the frequency and severity of FASD in these populations.

Consistent with this hypothesis, preliminary data from the ongoing studies in Ukraine and Russia show that plasma Zn and Cu concentrations are low in pregnant women who report high alcohol intakes. In addition to the mounting evidence that certain micronutrients can affect normal structural development, it is evident that prenatal nutrition is an important factor both in prenatal growth and in postnatal cognitive performance [88,89]. As described above, Zn, as well as Cu, is involved in multiple biochemical pathways that are critical for brain growth and function.

As the toxicity of alcohol is thought to be due, in part, to free radical-induced oxidative damage [83], deficits of Zn and Cu would both be predicted to increase the sensitivity of the developing conceptus to alcohol, given that these nutrients contribute to the oxidative defense system [90]. The reported blunting of alcohol’s adverse effects by folate, vitamin B₁₂, choline and Zn supplementation indicates that there may be some commonalities among these nutrients [22,91–93]. For example, all of these nutrients can affect redox stress, as well as gene methylation patterns, underscoring the fact that none of these nutrients should be looked at in isolation. Important future research directions include determination of the mechanisms underlying the developmental effects of the “suboptimal” nutritional status that can occur with alcohol exposure as well as delineation of how specific, persistent, and important these effects are.

A major aim of the ongoing CIFASD trial is to evaluate nutritional risk modifiers for FASD and to determine if maternal micronutrient supplementation during pregnancy in drinking women has substantial promise as an easily accomplished environmental manipulation that should substantially improve pregnancy outcome for both the mother and infant. Large prospective FASD studies that utilize well-“validated” biomarkers for maternal nutritional status (acute and chronic) prior to and during early pregnancy are a critical research need. In these studies, maternal nutritional status should be evaluated at regular intervals to capture environment-induced changes. These large prospective nutrient supplementation trials should be implemented with populations of pregnant women who are at a high risk for FASD. Moreover, nutrient supplementation trials should also be done with FASD children to determine whether nutritional intervention can alleviate long-term adverse outcomes in the offspring.

In closing, it should be noted that the concept that women who abuse alcohol during pregnancy should be encouraged to use multivitamin, multimineral supplements is not new. Indeed, in 1990, the American Institute of Medicine identified this group of women who should be particularly encouraged to use these supplements during pregnancy [94]. To date, our experience in Russia, as well as in Ukraine, is that women at high risk for having a FASD child are very receptive to using supplements. Obviously, the most appropriate message to these women is to stop drinking. However, on a practical level, combining this message with one that is aimed at improving their overall nutritional status would seem to be an appropriate public health strategy.