Fetal Alcohol Spectrum Disorder (FASD) refers to the range of adverse effects that can occur in the children of women who consume alcohol during pregnancy (Riley and McGee, 2005; Manning and Eugene Hoyme, 2007). At the most severe end of the spectrum is Fetal Alcohol Syndrome (FAS), which may occur with prolonged consumption of relatively high amounts of alcohol (Jones and Smith, 1973; Jones et al., 1973). Prenatal exposure to alcohol may also produce either partial FAS (some of the diagnostic features occur) or may result in Alcohol-Related Birth Defects, or Alcohol-Related Neurodevelopmental Disorders (Stratton et al., 1996; Johnson et al., 2018) Interestingly, whilst the level of alcohol exposure may determine the severity of the observed effects, other cognitive, neuropsychological, and behavioral deficits have consistently been observed (Streissguth, 1986; Astley et al., 2009; Guerri et al., 2009; Kable et al., 2016; Temple et al., 2019).

Although not recognized as a symptom of FASD, there is evidence for an association between prenatal alcohol exposure and alcohol use disorders in offspring. Despite the considerable range of variables that might influence alcohol consumption by individuals, the few existing prospective longitudinal studies analyzing the relation between prenatal alcohol exposure and alcohol use by offspring all agree on one chief outcome: maternal alcohol intake during gestation is one of the best predictors of later alcohol use from adolescence to young-adulthood (Baer et al., 1998, 2003; Griesler and Kandel, 1998; Alati et al., 2006; Streissguth, 2007; Cornelius et al., 2016; Goldschmidt et al., 2019). Concurrently, numerous experimental studies with animals have shown increased alcohol intake in the offspring of dams that had consumed alcohol during gestation. These studies have been extensively reviewed in three previous publications, in which the effects of prenatal alcohol exposure on infantile, adolescent and adult alcohol acceptance and intake were described, analyzing the possible mechanism involved in those effects (Spear and Molina, 2005; Chotro et al., 2007; Abate et al., 2008). In the present review, we continue to explore this issue by analyzing specific research on the subject from the last 15 years, particularly those studies that have used animal models, focusing on the importance of mechanisms related to fetal learning about alcohol.

Most of the clinical evidence on the association between prenatal alcohol exposure and alcohol use disorders is derived mainly from a few longitudinal studies. The first of these is the “Seattle Prospective Longitudinal Study on Alcohol and Pregnancy,” in which the effects of prenatal alcohol exposure were examined on a cohort of children born in 1974 to mothers selected as representative of the Seattle/King County population in the USA (Streissguth et al., 1981). In this study, among many other variables, alcohol use problems were measured in the offspring at three different ages: 14, 21 and 25 years. When subjects were 14 years old, three sets of data from 439 families were collected: adolescent alcohol use, family history of alcohol problems, and prenatal alcohol exposure history; and using these data they conducted correlational analyses. Prenatal alcohol exposure was found to be a better predictor of adolescent alcohol use than a family history of alcohol problems (Baer et al., 1998). When these same subjects were 21 years old, the families were re-evaluated, and it was reported that prenatal alcohol exposure was still strongly associated with a higher number of symptoms of alcohol dependence in early adulthood. This relationship persisted independently of the effects of a family history of alcohol problems, nicotine exposure, other prenatal exposures, and postnatal environmental factors including parental use of other drugs (Baer et al., 2003). At the age of 25, the association between prenatal alcohol exposure and adult alcohol use problems was still present (Streissguth, 2007). In all these cases the results were obtained after adjusting for maternal demographic characteristics, maternal use of tobacco and other drugs during pregnancy, and maternal and familial alcohol problems after birth. In line with those outcomes, alcohol exposure during gestation was also found to be a key factor for predicting alcohol use disorders in adults who were adopted at birth (Yates et al., 1998). In one study, 197 adoptees were evaluated for the use of alcohol, tobacco, and other drugs. Of those, 21 had received prenatal alcohol exposure and their outcomes were compared with those of 102 control adoptees who had received no alcohol exposure. The results showed that even when controlling for biological parental alcohol abuse or dependence, the prenatal alcohol exposure factor was still the best predictor of alcohol use disorders. This study highlights the relevance of alcohol exposure during prenatal development for alcohol abuse in adulthood, a relationship that appears to exist independently of confounding postnatal environmental variables (Yates et al., 1998).

Another follow-up study with participants from the “Mater–University of Queensland Study of Pregnancy and Its Outcomes” was designed specifically to analyze the association between maternal alcohol exposure and the onset of alcohol disorders. This study was conducted with 2,138 participants from a population-based birth cohort, born in Brisbane, Australia in 1981 (Alati et al., 2006). Mothers and their sons/daughters were followed from pregnancy to the offspring’s early adulthood, and the onset of alcohol disorders was registered from adolescence to 21 years old. The results revealed that in utero exposure to three or more drinks containing alcohol was related to alcohol use disorders at the age of 21, increasing its risk by almost three times compared with subjects exposed to either smaller amounts of alcohol or those that had received no exposure. In addition, they reported that the sons and daughters of mothers who had consumed three or more glasses of alcohol during early pregnancy were almost four times more likely to show an early onset of alcohol disorders at the age of 21 than those whose mothers had consumed less than two drinks at any time. This association was robust, even after adjusting for a number of biological and environmental factors (Alati et al., 2006). In another publication, similar results were reported when subjects were 14 years old (Alati et al., 2008).

Another longitudinal prospective study aimed to analyze the relative contribution of familial risk and prenatal exposure to substance use in offspring. This study was carried out with a sample of 209 third-generation offspring of families from the area of Pittsburgh, USA, specifically selected for being at either high or low risk of developing alcohol dependence (O’Brien and Hill, 2014). High-risk families were selected based on the presence of two alcohol-dependent sisters and low-risk families were selected on the basis of having a minimal number of first and second-degree relatives with alcohol dependence. The results of this study showed that prenatal alcohol exposure increased the risk of alcohol use disorders in both high and low-risk participants, although high-risk mothers were more likely to use alcohol and cigarettes during each trimester of pregnancy. In addition, it was reported that among the high-risk offspring, the effects of prenatal exposure were more specific to the particular substance exposed, i.e., prenatal alcohol exposure was associated with alcohol problems in offspring, while cigarette exposure was associated with cigarette use.

Another study analyzed the link between maternal self-reported alcohol consumption during pregnancy and adolescent self-reported drinking in a sample of 185 mothers and their first-borns recruited from the New York State Cohort (Griesler and Kandel, 1998). Maternal drinking was assessed retrospectively, at an average of 3.3 years after delivery, with reports covering a period of 18 months (including pregnancy). Adolescent (age 9–17) life-time and current alcohol drinking data were obtained from self-reports. The results indicated that maternal alcohol drinking during pregnancy—particularly moderate to heavy consumption—was associated with the current drinking of their female offspring. No association was found, however, between maternal drinking, either during or after pregnancy, and alcohol drinking in sons.

Finally, a fifth longitudinal study was conducted with participants from cohorts of the “Maternal Health Practices and Child Development Project.” These participants were recruited between 1983 and 1986, also from Pittsburg, USA, and they have been followed since the fourth gestational month. The data from this study demonstrated that the level of adolescent drinking (at the age of 16) was directly predicted by prenatal alcohol exposure, as well as lower levels of parental strictness and exposure to maltreatment and violence during childhood (Cornelius et al., 2016). These authors also reported that heavier drinking during adolescence is directly predicted by maternal alcohol consumption during pregnancy. A similar relation between these variables was observed when analyzing data collected during young adulthood, at the age of 22 (Goldschmidt et al., 2019). The results of these studies, together with the one described previously, provide clear evidence of the direct connection between maternal and adolescent drinking, which not only includes alcohol dependence and alcohol-related problems.

In sum, the outcomes of all these studies support the existence of an association between prenatal alcohol exposure and either early onset of alcohol drinking and/or with the development of alcohol use disorders in adolescence and young adulthood. In addition, most of these studies highlight a critical role for prenatal alcohol exposure in the idea of a causal pathway that leads to alcohol use disorders. Several mechanisms have been proposed to underlie the link between prenatal alcohol exposure and these consequences, although research studies at this level have primarily been conducted with laboratory animals.

Experimental studies with animals provide abundant evidence confirming the results found in humans i.e., prenatal exposure to alcohol—in addition to producing numerous harmful effects—will, in most cases, be followed by an increased acceptance (i.e., attraction and consumption) of alcohol. These effects have been reviewed comprehensively in previous publications in which the outcomes of numerous studies with rodents exposed prenatally and perinatally to alcohol are described and analyzed, focusing on the behavioral effects of prenatal exposure, and in particular, on the factors that play a role in the postnatal response to alcohol (Spear and Molina, 2005; Chotro et al., 2007; Abate et al., 2008). Table 1 includes the studies in rodents described in those reviews, as well as more recent publications, in which the effect of prenatal alcohol exposure on postnatal alcohol intake was assessed. As explained in those early reviews, increased alcohol intake has been observed after prenatal exposure to different doses of alcohol, ranging from relatively low to high alcohol concentrations. Those effects have been found in studies in which pregnant dams were given alcohol in a liquid diet made available for 24 h, as well as when alcohol was administered intragastrically in controlled amounts, modeling the so-called “binge drinking” of alcohol. In terms of the period of exposure, most of those studies show that exposure to alcohol during the entire gestation period of the rat (22 days), or only during the last 2 weeks, induces high alcohol consumption in the offspring. But even short binge-like exposures to relatively moderate alcohol doses, restricted to the final gestation days (GD 17–20), have systematically resulted in heightened alcohol intake. These results have been observed at different postnatal stages, infancy, adolescence, and adulthood. In general, there are more studies reporting this increased alcohol intake effect when tested early in ontogeny than in adulthood. However, there are some showing that this effect can be directly detected in late adolescence and even in adulthood, although in some cases postnatal re-exposure to alcohol seems necessary. The importance of other factors such as sex, genetic differences, and stress conditions at testing, on the detection of an effect of increased alcohol intake after prenatal alcohol exposure, has also been thoroughly discussed in those reviews (Spear and Molina, 2005; Chotro et al., 2007; Abate et al., 2008).

During the last decade, new studies showing evidence of an increased intake of alcohol following prenatal alcohol exposure in rodents have been added to those reviewed previously. Among all of this literature, those studies proposing and testing possible mechanisms by which fetal exposure to alcohol may increase the avidity for this drug are of particular interest for this review.

On the basis of the results found in animal studies, a number of different mechanisms have been proposed to explain the increased consumption of alcohol that is observed following prenatal alcohol exposure. Some propose an indirect link between high alcohol consumption and other alterations induced by prenatal alcohol exposure, while others suggest more direct pathways between prenatal alcohol consumption and an increased acceptance of the drug.

After reviewing all of the evidence from controlled experimental studies with animals, the connection between prenatal alcohol exposure and augmented alcohol intake during infancy, adolescence, and even adulthood appears to be clear. This link suggests that prenatal alcohol exposure increases the probability of early onset of alcohol use, which in turn has been described as a strong predictor of alcohol dependence (Grant, 1998). Although the causal relationship between adolescent-onset and adult alcohol use is still under debate (Prescott and Kendler, 1999), it is clear that adolescence is a vulnerable period for the neurobehavioral effects of alcohol (Spear, 2000, 2015). Unlike what occurs in adult subjects, initiation of alcohol use in adolescence has been found to accelerate the course of alcohol dependence, without the need for a long history of alcohol consumption (Clark et al., 1998; Spear, 2002). Therefore, any situation that favors the initiation of alcohol use during this period of development, such as stress or prenatal alcohol exposure, can be considered a risk factor for later alcohol use and misuse.

Thus, understanding the mechanisms by which prenatal alcohol exposure favors the early initiation of alcohol consumption may allow us to find strategies to mitigate the behavioral effects of this fetal experience. Moreover, it would be interesting to consider prenatal alcohol exposure as a case of involuntary early onset of alcohol use when designing prevention policies. This is particularly important if we assume that (as indicated by the longitudinal studies reviewed here), the sooner that alcohol intake begins, the greater the possibility of a long history of alcohol consumption and hence, the higher the likelihood of developing an alcohol use disorder.

Given these considerations, it is clear that more clinical and preclinical research is needed to explore prenatal alcohol exposure as a causal pathway leading to alcohol disorders in adolescence and adulthood. As discussed in this review, it has been well established that prenatal learning about the sensory and pharmacological properties of alcohol is a mechanism that plays an important role in facilitating the early onset of alcohol consumption. The opioid system has been found to mediate this prenatal learning, in which acetaldehyde acts as the main appetitive reinforcer. Future research should aim to find the link between acetaldehyde and the activation of the opioid system, with the dopaminergic system and/or salsolinol being the main candidates for this role. In addition to fetal alcohol learning, other identified coexistent and interacting mechanisms undoubtedly need to be investigated in more depth. For instance, stress is a factor that interacts with prenatal exposure to alcohol, facilitating in many cases the observed increase in alcohol intake in the exposed offspring. Further research is also needed to identify the causal processes by which prenatal alcohol-induced alterations in the activity of the HPA-axis drive the subject to consume more alcohol. Furthermore, it would be interesting to continue elucidating the role of the anxiolytic effects of alcohol on the increased intake response observed in subjects prenatally exposed to alcohol. Finally, the mechanistic connection between epigenetic modifications induced by prenatal alcohol exposure and the resulting changes in alcohol intake remains an underexplored but promising field of research, which needs to be addressed through rigorous research. In this regard, the first steps have already been taken, as shown, for example, in the results of a study already mentioned in this review (Wille-Bille et al., 2020).

MG, AA-A, and MC contributed equally to manuscript writing, revision, read and approved of the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.