A Narrative Review on the Effect of Valproic Acid on the Placenta

ABSTRACT

Background

Valproic acid (VPA) is an antiepileptic and mood‐stabilizing drug with well‐established teratogenic risks when taken during pregnancy. While its harmful effects on fetal development are well known, less attention has been given to its impact on placental development and function, despite the placenta's critical role in pregnancy.

Aim

This narrative review examines how VPA exposure affects placental growth, morphology, nutrient transport, and epigenetic modifications. It also considers whether placental dysfunction may contribute VPA's teratogenic effects.

Results

Evidence suggests that VPA disrupts placental structure and growth, alters the expression of nutrient transporters, such as those for folate, glucose, and amino acids, and modifies the placental epigenome, including globally decreased DNA methylation and increased histone acetylation.

Discussion

It is hypothesized that these epigenetic changes may influence chromatin remodelling and trophoblast gene expression, though this connection has not been fully established. Such epigenetic dysregulation may result in aberrant gene expression that underlies the structural and functional impairments observed in the placenta, potentially compromising its ability to support fetal development and contributing to VPA's teratogenic effects. Findings across studies, however, are inconsistent, varying with dose, timing of exposure, and model system. Furthermore, there is a lack of research examining sex‐specific differences in placental responses to VPA, despite evidence that male and female placentas exhibit distinct growth patterns, gene expression profiles, and susceptibilities to environmental insults.

Conclusion

Addressing these knowledge gaps through targeted research will improve our understanding of how VPA affects the placenta and its role in teratogenesis.

Abbreviations

ABC

ATP‐binding cassette

BBB

blood–brain barrier

BCRP

breast cancer resistance protein

CTL

choline transporter‐like protein

FATP

fatty acid transporting protein

FCCP

carbonyl cyanide‐p‐trifluoromethoxy phenylhydrazone

FRα

folate receptor alpha

GCM1

glial cell missing factor 1

GD

gestational day

GLUT

glucose transporter

HDAC

histone deacetylase

HIF

hypoxia‐inducible factor

LAT

L amino acid transporter

MCT

monocarboxylate transporter

MRP

multidrug resistance protein

NTD

neural tube defect

OATP

organic anion transporting polypeptide

OCTN

organic cation transporter novel

PCFT

proton‐coupled folate transporter

RFC

reduced folate carrier

VPA

valproic acid

1. Introduction

1.1. Teratogenic Risk and Clinical Challenges of Valproic Acid Use During Pregnancy

Since the 1980s, researchers and clinicians have reported on the teratogenic effects of valproic acid (VPA), a highly effective broad‐spectrum antiepileptic and mood‐stabilizing drug. VPA is prescribed to manage various neurological and psychiatric conditions, including epilepsy, bipolar disorder, and migraines. However, when taken during pregnancy, VPA can cause severe adverse outcomes. The U.S. Food and Drug Administration previously classified VPA as a Category D teratogen (Peckman 2010), warning users there is positive evidence of human fetal risk, but its use may be justified in life‐threatening emergencies or when safer drugs are ineffective (Friedman 1993; Schwarz et al. 2005). This classification system has since been retired. Under the U.S. Food and Drug Administration's Pregnancy and Lactation Labelling Rule, implemented in 2015, VPA's risks are now described in a detailed risk summary that promotes individualized risk–benefit assessments to guide clinical decision‐making (Pernia and DeMaagd 2016).

Despite government and clinical recommendations to avoid VPA as an initial treatment for females of reproductive age (Tomson et al. 2015), recent findings show that it is still being frequently prescribed, particularly for mood disorders and migraine prophylaxis (Adedinsewo et al. 2013; Smolinski et al. 2024). This may be partly due to the fact that discussion about VPA's teratogenic risks has predominately occurred within neurology‐related fields rather than in psychiatric literature (Rakitin 2020). In addition, guidelines for psychiatrists on VPA usage in females of reproductive age can be inconsistent, contributing to its continued prescription despite known risks (Rakitin 2020; Wartman and VandenBerg 2022). Considering approximately 50% of pregnancies are unplanned (Bearak et al. 2018, 2020; Finer and Henshaw 2006), pregnant individuals may inadvertently expose their unborn children to VPA during critical development windows. Such exposure significantly increases the risk of birth defects and impaired neurodevelopment in the fetus. Specifically, in utero VPA exposure in the first trimester is associated with a 9%–10% incidence of major congenital malformations (Battino et al. 2024; Hernández‐Díaz et al. 2012; Pack et al. 2024). These malformations include neural tube defects (NTDs), genitourinary and musculoskeletal anomalies, cleft lip and/or palate, and congenital heart defects (reviewed in Ornoy et al. 2023). NTDs are the most common birth defect observed with VPA exposure as they occur in 1%–2% of exposed pregnancies, which is 10–20 times greater than in the general population (Ornoy et al. 2023).

Moreover, taking VPA during pregnancy, particularly in the second and third trimester, raises the risk of fetal neurodevelopmental disorders by 30%–40% (Bromley et al. 2013; Coste et al. 2020). This could lead to delayed motor skills, reduced cognitive function, language impairment, autism spectrum disorder, attention deficit hyperactivity disorder, and other behavioral defects, all of which have been described in VPA‐exposed children (reviewed in Ornoy et al. 2023). These neurodevelopmental challenges are now recognized as part of fetal valproate spectrum disorder (Bluett‐Duncan et al. 2023). While this condition was initially identified through distinct facial features and congenital malformations (Moore et al. 2000), neurodevelopmental impairments are now considered central to its presentation (Clayton‐Smith et al. 2019). The consequences of in utero exposure extend beyond birth, with many individuals growing up to face significant physical, mental, and developmental challenges that profoundly affect their quality of life and those of their families (Khanom et al. 2024).

Discontinuing VPA treatment during pregnancy is discouraged due to the risks of uncontrolled seizures and psychiatric episodes, which jeopardize both maternal and fetal health (Hixson 2010; Schmidt 2011; Shih and Ochoa 2009; Viguera et al. 2007). Patients must be gradually tapered off VPA while a replacement drug is introduced simultaneously. However, stopping VPA treatment will not reverse any teratogenic damage already inflicted. While other antiepileptic and mood‐stabilizing drugs are available, they also carry teratogenic risks and may not be as effective for all patients (Meador and Loring 2016). These challenges contribute to the continued use of VPA in certain cases, despite the known risks. Therefore, it is critical to elucidate the mechanisms underlying VPA teratogenesis so that safer and equally effective therapeutic alternatives may be developed.

1.2. Placental Toxicity as a Potential Mediator of VPA Teratogenesis

Most research to date investigating VPA's teratogenic mechanisms has focused on fetal toxicity, indicating that VPA elicits its toxic effects directly on the fetus. However, there is also evidence to suggest that the placenta—a transient organ essential for pregnancy—may also be a target of VPA‐induced toxicity.

Historically, the placenta has been overlooked in birth defects research, as investigators have underestimated its role in fetal development and postnatal health. The placenta is not merely a waste product to be discarded after birth but an organ that holds valuable information about the pregnancy and the health of the offspring. For example, the size and shape of the placenta in relation to the fetus at birth has been shown to predict vulnerability for cardiovascular disease later in life (reviewed in Thornburg et al. 2010). In addition, placental abnormalities are thought to be responsible for most euploid miscarriages (miscarriages involving fetuses with a normal set of chromosomes) and are directly associated with several pregnancy complications, such as preeclampsia, intrauterine growth restriction, preterm birth, and stillbirth (Brosens et al. 2011; Burton and Jauniaux 2018; Del Gobbo et al. 2020; Freedman et al. 2019; Hemberger and Dean 2023; Jauniaux et al. 2006; Kaufmann et al. 2003; Pijnenborg et al. 2006). Many of these issues stem from insufficiencies in the placenta's primary function. Serving as the interface between mother and fetus, the placenta mediates the exchange of nutrients, gases, and waste products between the maternal and fetal circulations. However, the placenta is not just a passive conduit; it actively adapts its capacity for delivering nutrients in response to maternal availability and fetal demand to support fetal growth and development (Kramer et al. 2023). Beyond this, the placenta synthesizes and secretes various hormones, proteins, lipids, and even neurotransmitters that are essential for sustaining pregnancy (Kramer et al. 2023; Rosenfeld 2021).

The influence of placental dysfunction on congenital birth defects is far less recognized, but evidence suggests that the placenta plays a role beyond just providing nutrients (Hemberger and Dean 2023). Fetal organs, such as the brain and heart, are particularly vulnerable to placental abnormalities, which have been linked to adverse neurodevelopmental and cardiovascular outcomes in the offspring (reviewed in Mahadevan et al. 2023; Rosenfeld 2021). The terms “placenta‐brain axis” and “placenta‐heart axis” describe the intricate relationship between the development of these vital organs. These axes suggest that placental changes might provide a mechanistic understanding of neurodevelopmental disorders and heart defects in the offspring (Mahadevan et al. 2023; Rosenfeld 2021).

Given that in utero VPA exposure is associated with both neurodevelopmental disorders and congenital heart defects, it is plausible that these effects may result from VPA‐induced disruption of placental development. Currently, there is evidence suggesting that in utero VPA exposure impairs placental structure and function; however, it is limited and conflicting. Since placental function plays a crucial role in both fetal development and postnatal health, understanding how VPA alters its molecular processes is essential for uncovering its broader teratogenic potential. These insights could improve the early detection and prevention of complications in VPA‐exposed pregnancies and help identify individuals at greater risk (Walker et al. 2017).

To better understand how VPA‐induced placental toxicity may contribute to its overall teratogenesis, this review summarizes the current evidence of VPA's effect on placental development. A comprehensive literature search was conducted collaboratively by two authors using Embase, Medline, and Web of Science to identify primary research articles investigating the effects of VPA on the placenta. The search utilized the keywords valproic acid and placenta, yielding 189 articles that met the search criteria. After screening for relevance, 25 articles were deemed pertinent and included in this review.

2. Effect of VPA on Placental Growth

2.1. Variability in Its Impact on Placental Size

Fetal and placental size are representative of the placenta's functional capacity to support fetal growth and reduced fetal and/or placental growth can be indicative of placental toxicity. While in utero VPA exposure has consistently been shown to negatively impact fetal growth, studies investigating the effect of VPA on placental growth report conflicting results. For example, we previously exposed pregnant female CD‐1 mice to a single dose of either saline or 400 mg/kg VPA on gestational day 9 (GD9) and examined its impact on fetal and placental growth in fetuses with and without exencephaly (a NTD observed with VPA exposure in mice) on GD13, GD15, GD17, and GD19. We found that non‐exencephalic fetuses exposed to VPA had significantly reduced fetal weight and crown‐rump length compared to saline controls at all time points except GD19, while exencephalic fetuses consistently showed significantly lower fetal weights. However, VPA's impact on placental growth was inconsistent: placental weights and diameters varied across time points and between non‐exencephalic and exencephalic groups (Shafique and Winn 2021a). Therefore, while VPA consistently affected fetal growth, its impact on placental growth was less clear.

A study by Hrubec et al. (2006) further supports evidence of placental growth disruption following VPA exposure. Female CD‐1 mice were administered 500 mg/kg VPA on the morning of GD8, with placental and fetal samples collected on GD17. The researchers found that both fetal and placental weights were significantly reduced in the VPA‐treated group compared to controls. These findings provide support for the potential of VPA to negatively influence placental growth.

Another study by Ruyani and Sumarsono (2023) assessed pregnant BALB/c mice divided into four groups who were either untreated (control), or 400 or 600 mg/kg VPA. Each group was further split into subgroups that were consecutively dosed from GD10 to either GD15 or GD17, with placentas harvested on GD15 and GD17. They found no significant differences in placental weight between exposure groups at each time point (Ruyani and Sumarsono 2023). However, it cannot be assumed from the placental weight results alone that VPA did not cause placental toxicity, as the authors did not report any fetal growth data, which could provide insight into how the placenta functioned following VPA exposure.

Contrastingly, in another study conducted in our lab, we found that pregnant CD‐1 mice exposed to 600 mg/kg VPA on GD13 had significantly lower placental weights on GD18 compared to saline controls (Jackson et al. 2024). However, there was no significant difference in the average fetal weight or placental efficiency per litter between the treated and untreated groups (Jackson et al. 2024). Placental efficiency is the ratio of fetal to placental weight, which is used as a proxy measurement for placental function as it evaluates if a placenta is properly transporting nutrients to the fetus. Small placentas connected to large fetuses are considered highly efficient, whereas large placentas connected to small fetuses are deemed insufficient. Placental growth should be assessed relative to fetal growth to provide insight into placental functioning. For example, if fetal weight decreases with VPA exposure but placental weight does not, it can be indicative of reduced efficiency, which may be caused by drug‐induced placental toxicity.

Overall, VPA's effect on placental growth remains a complex issue. While most studies show consistent reductions in fetal growth following VPA exposure, its impact on placental growth is more variable, with some studies reporting significant changes and others finding no effect. One important factor that may contribute to this variability is measurement error. Measurement variability in fetal and placental weight arises from instrument precision, human handling, biological differences, and timing of measurement. Due to its smaller size, placental weight is particularly susceptible to measurement error, as even minor absolute discrepancies represent a larger proportion of its total weight compared to fetal weight. This can artificially inflate variability in placental weight data, potentially masking true biological effects of VPA or creating inconsistencies across studies. In addition, differences in measurement protocols—such as whether placentas are blotted dry before weighing or inconsistencies in placental extraction—can further contribute to variability. To improve the reliability of placental weight assessments, it is essential to implement standardized measurement protocols, ensure regular scale calibration, and consider complementary metrics like placental diameter and placental efficiency, which evaluate placental size and function relative to fetal weight, respectively, rather than relying solely on absolute placental weight.

2.2. Structural Alterations in the Placenta

Although the impact of VPA on placental growth remains uncertain, changes in placental size are often accompanied by changes to placental structure. Most investigations into placental morphology following VPA exposure have utilized rodent models, particularly mice and rats, due to the extensive limitations associated with using human placentas. Both humans and rodents have discoid, hemochorial placentas where maternal and fetal blood are separated by multiple layers of specialized trophoblast cells, forming the interhaemal barrier, which facilitates nutrient, gas, and waste exchange. While human and rodent placentas have structural differences, the molecular regulation behind their placentation is similar (Hemberger et al. 2020); therefore, mechanistic studies evaluating the structural alterations to the placenta in rodents can inform the potential mechanisms in humans.

In brief, the rodent placenta comprises three primary layers: the maternal decidua and the two fetal‐derived layers, the junctional zone and the labyrinth (Figure 1). Each layer has distinct functions that are crucial for supporting fetal growth and development. The maternal decidua develops from the blastocyst implantation site and anchors the growing embryo to the uterine wall, acting as structural support (Panja and Paria 2021). The decidua is also the site of maternal vascular remodeling to form spiral arteries that supply the fetus with adequate blood flow throughout pregnancy (Panja and Paria 2021). Deeper within the placenta, situated between the decidua and the labyrinth, is the junctional zone, which is comprised of two major cell types: spongiotrophoblast cells and glycogen trophoblast cells. Spongiotrophoblast cells make up most of the junctional zone and play a critical endocrine role (Elmore et al. 2022), whereas glycogen trophoblast cells store and supply glycogen to support both fetal and placental growth. The innermost and perhaps most critical layer of the placenta is the labyrinth. The labyrinth is the primary site of nutrient, gas, and waste exchange between the mother and fetus, housing an intricate network of maternal and fetal blood vessels. The proper development and function of each placental layer is essential for the successful progression of pregnancy, and defects in their structural development have been associated with developmental failure and growth deficits (reviewed in Woods et al. 2018).

FIGURE 1.

Schematic representation of rodent placenta layers. Representative images from a GD18 mouse placenta visualized with hematoxylin and eosin staining to highlight the cellular structure of each placental layer. (A) High magnification of the maternal decidua. (B) High magnification of the junctional zone, highlighting its two major cell types: Spongiotrophoblast cells (SpTs) and glycogen trophoblast cells (GTCs). (C) High magnification of the labyrinth, highlighting the blood spaces containing anucleated red blood cells (RBCs) and sinusoidal trophoblast giant cells (S‐TGCs). Figure created using BioRender.com.

Although few studies have assessed placental structure following in utero VPA exposure, their findings do suggest that VPA alters placental morphology; however, the results differ depending on the experimental model. Khera (1992a, 1992b) observed significant progressive changes to the placental decidua and labyrinth in Sprague–Dawley rats for up to 48 h following a single exposure to 600, 800, and 1000 mg/kg VPA on GD13. These changes included thrombosis and angiectasis (abnormal widening) of the maternal blood vessels, necrosis of trophoblast cells responsible for nutrient transfer between the mother and fetus, suppressed proliferation of the fetal capillaries, and reduced diameters of the umbilical vessels. Together, these degenerative changes were hypothesized to compromise the placenta's ability to facilitate proper maternal–fetal exchange, ultimately posing serious developmental risks for the embryo (Khera 1992a, 1992b).

Recent studies further corroborate these findings. The aforementioned study by Ruyani and Sumarsono (2023) reported that consecutive exposure to 400 and 600 mg/kg VPA from GD10 to GD15 and GD17 significantly reduced the area and vascular density of the placental labyrinth in BALB/c mice. They also found that VPA notably decreased the relative mRNA levels of key proangiogenic genes, including Nrp‐1 (neuropilin‐1), Vegfa (vascular endothelial growth factor A), and Kdr (vascular endothelial growth factor receptor 2) at both GD15 and GD17. Also, the relative mRNA levels of the antiangiogenic gene sFlt‐1 (soluble fms‐like tyrosine kinase‐1) were significantly higher on GD17 in all VPA‐exposed placentas (Ruyani and Sumarsono 2023). Perhaps unsurprisingly, Kühnel et al. (2017) found that placental overexpression of human sFlt‐1 in mice restricted fetal intrauterine growth and reduced placental weights. Placental‐specific overexpression of human sFlt‐1 altered placental morphology and trophoblast differentiation, characterized by a reduced placental area, a diminished labyrinth layer, and a decreased number of glycogen cells in the junctional zone (Kühnel et al. 2017). The findings of Kühnel et al. (2017) indicate that the increased sFlt‐1 expression in VPA‐exposed placentas, as seen by Ruyani and Sumarsono (2023), could significantly impair placental structural development. Overall, Ruyani and Sumarsono (2023) hypothesized that the altered expression of angiogenesis‐related genes contributed to the reduced vascular density observed in VPA‐exposed placentas, indicating that VPA may disrupt placental angiogenesis. This hypothesis is supported by VPA's ability to inhibit tumor angiogenesis in various cancer models (Berendsen et al. 2012; Gao et al. 2007; Sawai et al. 2023; Wang et al. 2013; Zhang et al. 2014; Zhao et al. 2016).

In another study, Aljuboury and Al‐Hayani (2025) investigated the effect of repeated VPA exposure, administering 0, 250, or 500 mg/kg VPA once daily for 18 days to pregnant mice. They found that placentas exposed to VPA exhibited severe vacuolated glycogen cells and necrosed spongiotrophoblasts within the junctional zone, and hypoplasia of villi (Aljuboury and Al‐Hayani 2025). The presence of edema in the labyrinth at the highest dose further indicates that VPA may disrupt placental homeostasis and compromise the placental barrier, potentially contributing to the teratogenic fetal outcomes. However, this study's conclusions should be interpreted cautiously due to the lack of quantitative analysis of placental changes, which could undermine the generalizability and robustness of these findings.

Contrastingly, in the aforementioned study by our lab, no significant histological differences were observed in the area of the fetal‐derived layers (the junctional zone and the labyrinth) of GD18 placentas (treated on GD13) between the control and the 600 mg/kg VPA‐treated groups in CD‐1 mice (Jackson et al. 2024). Similarly, Emmanouil‐Nikoloussi et al. (2004) histologically examined GD18 placentas exposed to 500 or 600 mg/kg VPA from GD8 to GD10 and found no significant differences among controls and drug‐treated placentas. In addition, Hrubec et al. (2006) reported that placentas from VPA‐exposed mice were histologically indistinguishable from those of unexposed controls.

Collectively, these studies underscore the complexity and variability in VPA's impact on placental morphology. The current evidence indicates that both single and repeated VPA doses can compromise the structural integrity of the placental layers, potentially impairing fetal development (Aljuboury and Al‐Hayani 2025; Khera 1992a, 1992b; Ruyani and Sumarsono 2023). However, the extent of these effects is influenced by both the dose and timing of exposure. VPA's teratogenicity is well‐established as dose‐dependent, with higher doses correlating with more severe developmental outcomes (Ornoy et al. 2023). This dose–response relationship may help explain why significant morphological alterations were reported in studies using doses of 400 mg/kg VPA or higher (Aljuboury and Al‐Hayani 2025; Khera 1992a, 1992b; Ruyani and Sumarsono 2023).

The timing of exposure is equally crucial. Early in pregnancy, placental trophoblast cells are actively growing and differentiating to form the mature placenta. The structural consequences associated with VPA exposure may depend on the stage of placental development at that time. Exposures earlier in development could have more structural consequences compared to later in development when the placenta is fully established. The placenta also possesses remarkable plasticity, adjusting its phenotype in response to intrauterine environmental cues. Interestingly, studies that observed no significant structural differences between VPA‐treated and control placentas (Emmanouil‐Nikoloussi et al. 2004; Hrubec et al. 2006; Jackson et al. 2024) administered VPA early in gestation but examined the placentas on GD18, late in pregnancy. In contrast, studies that identified structural abnormalities assessed placentas within 48 h of the last VPA exposure, suggesting that placental adaptation may occur over time. This raises the intriguing possibility that the placenta can recover from VPA‐induced insults by modulating its structure and function throughout gestation. Further research is needed to validate this hypothesis and to explore how placental adaptation influences fetal development, emphasizing the need for temporal analyses across multiple gestational stages to fully capture the dynamic responses of the placenta. Results from these studies could offer insight into the development and postnatal health of offspring from mothers who were gradually tapered off VPA during their pregnancy. In addition, investigations examining the effect of VPA administered throughout gestation (e.g., daily doses) are crucial, as this dosing regimen more accurately mirrors clinical scenarios where pregnant individuals may continue VPA treatment. Such studies could help inform safer treatment strategies and improve outcomes for both mothers and their children.

3. Placental Nutrient Transport: VPA Transfer and Its Effect on Transporter Function

As the interface between the mother and fetus, the placenta shields the fetus from direct exposure to maternal circulation while enabling the transfer of essential nutrients and molecules. The maternal and fetal bloodstreams are separated by the interhaemal barrier, a specialized structure composed of differentiated trophoblast cells, called syncytiotrophoblasts, that facilitate the transport of substances necessary for fetal growth. The composition of this barrier differs between species. In humans, a single layer of syncytiotrophoblast cells separates the maternal blood and the fetal capillary endothelium, whereas the rodent placenta has two syncytiotrophoblast layers as well as a sinusoidal trophoblast giant cell layer that separates the maternal and fetal vascular spaces. Despite these morphological differences, the mechanisms by which molecules cross the placental barrier are similar in both species. Substances must traverse the maternal‐facing membrane, pass through the cell layer(s), and cross the fetal‐facing membrane to reach fetal circulation. Depending on their properties, molecules can cross the placental barrier via passive or facilitated diffusion, or active transport. Nutrient transporters embedded in the trophoblast cell membranes play a critical role in moving compounds across these layers, and the placenta can dynamically adjust transporter expression based on maternal nutrient availability and fetal demand—ensuring proper fetal growth and development. While the placenta offers some protection against maternally administered drugs, certain compounds, like VPA, can cross the placental barrier and enter the fetal compartment, where they may elicit teratogenic effects.

3.1. Transplacental VPA Transfer

Uncovering the mechanism behind VPA's transplacental transfer is crucial for understanding fetal exposure and its associated risks. VPA transfer has been studied using in vitro models of the placental barrier, particularly human placental choriocarcinoma lines such as BeWo and JEG‐3 cells. These studies reveal that VPA transport across the placenta is not governed by passive diffusion alone but involves carrier‐mediated mechanisms. VPA is a branched, eight‐carbon carboxylic acid with a molecular weight of 114 Da and a pK _a of approximately 4.9. At physiological pH (~7.4), the majority of VPA exists in its ionized form, which restricts passive diffusion across lipid membranes. While non‐ionized VPA can cross membranes more easily under acidic conditions, the predominance of the ionized form at physiological pH necessitates the involvement of specialized transport systems to facilitate its movement across the placental barrier.

3.1.1. Carrier‐Mediated VPA Transport

One of the first studies to show that VPA involved carrier‐mediated transport across the placenta was conducted by Utoguchi and Audus (2000) who demonstrated that VPA uptake in BeWo cells is a saturable process. Specifically, they found that VPA uptake was blocked by metabolic inhibitors and the proton disruptor carbonyl cyanide‐p‐trifluoromethoxy phenylhydrazone (FCCP), indicating that its transport requires cellular energy expenditure and is driven by a proton gradient. These findings suggest that VPA, in its ionized state, relies on proton‐dependent transporters to cross the trophoblast barrier (Utoguchi and Audus 2000). Building on these results, VPA was observed to have a higher permeability coefficient in the apical‐to‐basolateral direction, indicating preferential transport toward the fetus (Ushigome et al. 2001). Furthermore, dose‐dependent saturation of VPA transport in the same direction was demonstrated, reinforcing the role of carrier‐mediated mechanisms in VPA's fetal‐directed transfer (Ikeda et al. 2015).

3.1.2. Involvement of Monocarboxylate Transporters

The proton‐dependent nature of VPA transport led researchers to investigate the potential involvement of monocarboxylate transporters (MCTs), which are active, proton‐gradient‐dependent transporters expressed in the human placenta and facilitate the transport of various monocarboxylic acids. Utoguchi and Audus (2000) found that VPA uptake in BeWo cells was inhibited by certain monocarboxylic acids but not dicarboxylic acids, suggesting MCT involvement. In addition, benzoic acid, a known MCT substrate, competitively inhibited VPA uptake (Utoguchi and Audus 2000), further supporting this conclusion.

Nakamura et al. (2002) reported similar findings in human placental brush‐border membrane vesicles. They observed that the uptake of both VPA and lactic acid, a well‐characterized MCT substrate, was decreased by the MCT inhibitor α‐cyano‐4‐hydroxycinnamate (Nakamura et al. 2002). Lactic acid also competitively inhibited VPA uptake, reinforcing the hypothesis that VPA uses the same transport pathways as other monocarboxylates. Several MCT isoforms, including MCT1, MCT3, MCT4, MCT5, MCT6, and MCT7, are expressed in the placenta (Price et al. 1998). VPA may be actively transported into the fetal compartment by any one of these isoforms, potentially contributing to elevated fetal VPA concentrations observed clinically. However, one study found that knocking down MCT1 and MCT4 in JEG‐3 cells did not significantly affect VPA transport (Ishiguro et al. 2018), suggesting that other proton‐dependent transporters or MCT isoforms may play a role in VPA's placental transfer (Ishiguro et al. 2018).

3.1.3. Key Insights and Future Directions for Transplacental VPA Transfer

The available in vitro evidence strongly suggests that VPA's transplacental transfer is facilitated by carrier‐mediated mechanisms, not solely passive diffusion. These processes are likely proton‐dependent and possibly involve MCTs, which have been implicated in the movement of VPA and other fatty acids. However, more research must be conducted to fully elucidate VPA's transport mechanism across the placenta.

Given the similarities between the blood–brain barrier (BBB) and the placental barrier in controlling the passage of substances, insights into VPA transport across the BBB may provide clues about the specific transporters involved in its placental transfer. For example, Naora and Shen (1995) found that VPA's uptake across the BBB was not affected by classical MCT substrates such as pyruvate and lactate, suggesting that MCTs may not play a primary role in VPA transport. Instead, their data indicated that VPA likely uses medium‐chain fatty acid carriers and other organic anion systems. In parallel, Guo and Jiang (2016) identified organic anion transporting polypeptide 2 (OATP2) as a key transporter for VPA in the BBB. They found that increased OATP2 expression enhanced VPA uptake. OATP2 is also expressed in the placenta and therefore may also contribute to its distribution from maternal to fetal compartments. Since VPA transport across the BBB exhibits directionality, favoring the blood‐to‐brain direction, it is plausible that similar transport systems could facilitate the maternal‐to‐fetal transfer seen across the placenta. While dicarboxylate and organic anion carriers have been identified as important in VPA transport toward the brain (Guo and Jiang 2016; Naora and Shen 1995), their roles in the placenta remain unexplored. Further research is essential to elucidate whether OATP2, MCTs, or other proton‐linked transporters contribute to the elevated fetal VPA levels observed clinically, and how these systems influence fetal drug exposure and outcomes.

One unexplored pathway of potential VPA transport across the placenta is endocytosis. Endocytosis enables internalization of a region of the plasma membrane, facilitating the uptake of extracellular cargo—including substances bound to plasma proteins—across the trophoblast layers to reach fetal circulation (Cooke et al. 2021). The placenta has been shown to internalize albumin‐bound cargo, as demonstrated by the uptake of vitamin D via endocytosis (Ashley et al. 2022). Given that VPA is highly protein‐bound in maternal blood, primarily to albumin (Nau and Krauer 1986; Patel and Levy 1979), it is plausible that a VPA–albumin complex could be transported across the placenta via this mechanism, enabling maternal‐to‐fetal transfer. Investigating this pathway could offer valuable insights into how VPA crosses the placental barrier, contributing to fetal exposure and drug effects.

Overall, the evidence reviewed highlights the complex mechanisms underlying VPA's transplacental transfer, emphasizing the potential roles of carrier‐mediated transport. While similarities between the placental barrier and the BBB suggest that transporters such as OATP2 and other organic anion systems may facilitate VPA movement, the precise contribution of these transporters in the placenta remains unclear. In addition, the possibility of VPA–albumin complexes crossing the placenta via endocytosis offers a novel avenue for investigation. Uncovering these mechanisms is critical for understanding fetal VPA exposure and may reveal targeted therapy options to prevent its transplacental transport; therefore, more research in this area is needed.

3.2. Effect on Nutrient Transporter Expression

The fetus is reliant on its mother to supply the nutrients essential for its growth and development. Nutrient transporters, located on the apical and basolateral membranes of syncytiotrophoblasts, facilitate the uptake of essential nutrients to ensure fetal growth and development. The relative expression of these transporters changes across pregnancy depending on maternal availability and fetal demand. Dysregulated nutrient transporter expression impacts nutrient availability, consequently affecting fetal growth and development. VPA has been shown to alter the expression of various placental nutrient transporters; however, the direction of expression changes depends on the experimental model, making it difficult to determine the effect of VPA on fetal nutrient availability.

3.2.1. Folate Transporters

Folate is an essential B vitamin required for cellular metabolic reactions, one‐carbon metabolism, and gene regulation. During pregnancy, maternal folate demand increases significantly due to its essential role in fetal development, particularly neurodevelopment (Li and Meador 2022). The placenta transports folate through three main carriers: folate receptor alpha (FRα; encoded by FOLR1), reduced folate carrier (RFC; encoded by SLC19A1), and proton‐coupled folate transporter (PCFT; encoded by SLC46A1). Disruptions in the folate pathway have been implicated in VPA teratogenesis, yet studies on the effects of VPA on placental folate transport show conflicting results.

The expression of FRα has been extensively studied, with conflicting results depending on experimental conditions. In vitro, VPA exposure has been shown to increase FOLR1 mRNA levels in trophoblast cell lines. Rubinchik‐Stern et al. (2015) reported a dose‐dependent increase in FOLR1 mRNA levels in BeWo cells after 5 days of exposure to 83 and 166 μg/mL VPA. Kurosawa et al. (2018) similarly found that 24‐h exposure to 500 μM VPA significantly increased FOLR1 mRNA expression in BeWo and JEG‐3 cells, reaching 171% and 225% of control levels, respectively. However, ex vivo and in vivo studies suggest the opposite trend. Rubinchik‐Stern et al. (2018) found that FOLR1 mRNA levels decreased by 57% and 72% in full‐term human placentas perfused with 83 and 166 μg/mL VPA, respectively. Similarly, Furugen et al. (2021) observed that repeated VPA administration at 400 mg/kg/day from GD16 to GD19 significantly reduced Folr1 mRNA levels in GD20 rat placentas to 64% of control levels, with a concurrent 35% reduction in FRα protein. In contrast, Meir et al. (2016) saw no significant changes in Folr1 mRNA expression in GD13.5 and GD18.5 mouse placentas after repeated doses of 200 mg/kg VPA from GD8.5–GD12.5 and GD13.5–GD17.5. Similarly, in early human placentas perfused ex vivo with 42 μg/mL VPA for 5 days, Tetro et al. (2019) found no significant changes in FOLR1 expression. These discrepancies suggest that VPA's effects on FOLR1 may depend on the model system, exposure duration, dosage, and gestational timing.

Compared to FOLR1, studies on SLC19A1 more consistently indicate that VPA suppresses its expression. Rubinchik‐Stern et al. (2015) reported significant reductions in SLC19A1 mRNA levels in BeWo cells after 2‐ and 5‐day incubations with 166 μg/mL VPA. RFC protein expression also decreased in a dose‐dependent manner after 5 days, reaching 57%, 53%, and 46% of control levels at 42, 83, and 166 μg/mL VPA, respectively. These findings are supported by in vivo studies, where Meir et al. (2016) reported a decrease in Slc19a1 mRNA levels in GD13.5 mouse placentas following repeated 200 mg/kg VPA injections, though no changes were observed at GD18.5. Similarly, Tetro et al. (2019) demonstrated a reduction in SLC19A1 mRNA and protein expression in early human placentas perfused ex vivo with 42 μg/mL VPA; however, this decrease was not significant. Interestingly, Rubinchik‐Stern et al. (2018) reported an increase in SLC19A1 gene expression in full‐term human placentas perfused ex vivo with 42, 83, and 166 μg/mL VPA, further highlighting how gestational timing may influence the placental response to VPA exposure.

The effects of VPA on SLC46A1 gene expression are less clear. Kurosawa et al. (2018) observed significant increases in SLC46A1 mRNA levels in BeWo and JEG‐3 cell lines following a 24‐h exposure to 500 μM VPA. However, other studies investigating SLC46A1 gene expression found no significant changes following VPA exposure, whether in full‐term human placentas (Rubinchik‐Stern et al. 2018; Tetro et al. 2019). Given the limited number of studies investigating PCFT regulation, its role in placental folate transport following VPA exposure remains uncertain.

3.2.2. Choline Transporters

Choline is an essential nutrient required for multiple physiological processes, including phospholipid synthesis, bile production, and lipoprotein formation (Radziejewska and Chmurzynska 2019). It is crucial for sphingomyelin synthesis, axon nerve myelination (Michel et al. 2006), and plays a key role in neurotransmission via acetylcholine synthesis (Sarter and Parikh 2005). In addition, choline serves as a secondary methyl donor for S‐adenosyl methionine synthesis, which regulates cellular methylation reactions (Zeisel and Blusztajn 1994). Deficiencies during the perinatal period have been linked to an increased risk of NTDs and orofacial clefts, and impaired cognitive outcomes (Boeke et al. 2012; Shaw et al. 2006, 2004).

Because the fetus depends entirely on maternal choline, placental transport is necessary to meet fetal demands (Garner et al. 1995). This transport is facilitated by choline transporter‐like proteins 1 (CTL1; encoded by SLC44A1) and 2 (CTL2; encoded by SLC44A2) (Radziejewska and Chmurzynska 2019). While research on VPA's effects on placental choline transport is limited, existing evidence suggests a tendency for VPA to decrease CTL expression. Rubinchik‐Stern et al. (2018) found significant differences in SLC44A1 mRNA expression across full‐term human placentas perfused ex vivo with 0, 42, 83, and 166 μg/mL VPA. Similarly, Tetro, Moushaev, et al. (2021) demonstrated that BeWo cells incubated with 42, 83, or 166 μg/mL VPA for 2 or 5 days exhibited significant reductions in SLC44A1 mRNA expression after 5 days at 166 μg/mL and SLC44A2 mRNA expression after 5 days at 83 μg/mL. Protein expression of both CTL1 and CTL2 significantly decreased after 5 days of incubation at 83 and 166 μg/mL (Tetro, Moushaev, et al. 2021).

3.2.3. Glucose Transporters

Glucose is the primary energy source for fetal metabolism, making its placental transport essential for fetal growth and development (Sibiak et al. 2022). Since the placenta produces minimal glucose until late gestation, it relies on maternal glucose uptake to sustain fetal energy needs and support placental glycogen synthesis (Hahn et al. 1999). Placental glycogen acts as an energy reserve, ensuring a continuous glucose supply during fluctuations in maternal levels, which is critical for maintaining fetal growth (Armistead et al. 2020; Hahn et al. 2001).

Placental glucose transport is mediated by glucose transporter proteins (GLUTs), which facilitate sodium‐independent diffusion across the placental barrier (Illsley 2000; Ingermann 1987; James‐Allan et al. 2019). Among them, GLUT‐1 (encoded by SLC2A1) is the primary placental isoform (Brown et al. 2011; Sakata et al. 1995). In a healthy pregnancy, GLUT‐1 expression increases with gestation to meet fetal nutritional demands (Stanirowski et al. 2021). Disruptions to GLUT‐1 function can impair fetal glucose availability, potentially leading to fetal growth restrictions and developmental abnormalities (Stanirowski et al. 2021).

Research on the effects of VPA exposure on placental glucose transport remains limited, but existing evidence suggests a negative impact on GLUT‐1. Rubinchik‐Stern et al. (2018) reported that VPA exposure at 83 and 166 μg/mL reduced SLC2A1 mRNA expression in perfused human full‐term placentas. Similarly, Kitahara et al. (2024) found that 0.3–1.2 mM VPA decreased glucose uptake and GLUT‐1 protein expression in BeWo cells. VPA‐induced downregulation may impair placental glycogen storage and glucose buffering, potentially disrupting fetal growth and highlighting the need for further investigation across models.

3.2.4. Organic Anion Transporting Polypeptides

Organic transporting polypeptides (OATPs) are membrane transporters responsible for moving large (> 300 Da) organic anions into cells (Kovacsics et al. 2017; Walker et al. 2017). Of the 11 identified human OATPs, only a few are present in the placenta (Berveiller et al. 2015; Kis et al. 2010; Staud et al. 2012; Wang et al. 2012). Their broad substrate specificity enables the transport of bile acids, bilirubin, eicosanoids, prostaglandins, thyroid hormones, and their sulfate or glucuronate conjugates (Ali et al. 2020; Kovacsics et al. 2017). OATPs operate as electroneutral transporters by coupling substrate uptake to the efflux of counterions like glutathione, conjugated glutathione, bicarbonate, or glutamate (Kovacsics et al. 2017; Roth et al. 2012).

VPA exposure alters several OATP subtypes, including OATP1A2, OATP2A1, OATP2B1, and OATP4A1, encoded by SLCO1A2, SLCO2A1, SLCO2B1, and SLCO4A1, respectively. Rubinchik‐Stern et al. (2015) found that 166 μg/mL VPA significantly reduced SLCO1A2 mRNA levels after 2 days, though this effect was not sustained after 5 days, and OATP1A2 protein levels remained unchanged. Jinno et al. (2020) was significantly increased in the GD13 rat placenta following both single and repetitive administrations of 400 mg/kg VPA, and that Slco2b1 mRNA expression was significantly reduced following a single administration of 400 mg/kg VPA in the GD20 rat placenta.

The most consistent findings involve SLCO4A1 (OATP4A1). Rubinchik‐Stern et al. (2015) found that 83 and 166 μg/mL VPA reduced SLCO4A1 mRNA levels after 2 and 5 days, while protein expression decreased only at 42 μg/mL after 5 days. Meir et al. (2016) also observed reduced Slco4a1 mRNA expression in GD13.5 mouse placentas following VPA exposure, and Jinno et al. (2020) found similar reductions in GD20 rat placentas after both single and repeated VPA treatments. Given OATP4A1's critical role in thyroid hormone transport, its downregulation is concerning, as thyroid hormones are essential for fetal neurodevelopment (Moog et al. 2017). Disruptions in their placental transfer could contribute to the developmental abnormalities linked to VPA exposure during pregnancy.

3.2.5. Carnitine Transporters

Carnitine is essential for fetal development, transporting long‐chain fatty acids across the mitochondrial inner membrane for oxidation, providing an alternative energy source alongside glucose metabolism (Houten and Wanders 2010; Shekhawat et al. 2004). It also protects against free radical damage, preventing lipid peroxidation, cell death, and apoptosis (Shekhawat et al. 2004). Carnitine reserves support metabolism and promote tissue development, and deficiencies have been linked to apnea, cardiac arrest, cardiac hypertrophy, cardiomyopathy, decrease in interventricular septum thickness, Reye‐like syndrome, lethargy, hepatotoxicity, encephalopathy, and sudden infant death syndrome (Wu et al. 2004). Notably, several of these symptoms overlap with fetal valproate spectrum disorder, hinting at a potential link between carnitine deficiency and VPA‐induced developmental abnormalities (Wu et al. 2004).

Placental carnitine transport relies on organic cation transporters, particularly OCTN1 (encoded by SLC22A4) and OCTN2 (encoded by SLC22A5) (Tamai et al. 1998; Wu et al. 2000). VPA exposure alters their expression in a gestation‐dependent manner. Jinno et al. (2020) found that 400 mg/kg VPA significantly reduced Slc22a4 mRNA expression in GD20 rat placentas following both single and repetitive dosing, while Slc22a5 gene expression significantly increased in GD13 placentas after repeated exposure. These opposing effects suggest that VPA's impact on carnitine transport varies with gestational timing.

3.2.6. ATP‐Binding Cassette Efflux Transporters

ATP‐binding cassette (ABC) transporters, including multidrug resistance proteins (MRPs) and breast cancer resistance protein (BCRP), are essential for placental function. They actively transport nutrients like inorganic ions, glucose, amino acids, metal ions, cholesterol, and phospholipids while protecting the fetus by effluxing toxicants back into maternal circulation (Walker et al. 2017). Among ABC transporters, MRPs in the ABCC gene subfamily are well studied, with MRP1 (encoded by ABCC1), MRP2 (encoded by ABCC2), MRP3 (encoded by ABCC3), MRP4 (encoded by ABCC4), and MRP5 (encoded by ABCC5) playing key roles in placental drug transport (Borst et al. 2000).

VPA exposure alters MRP expression. Jinno et al. (2020) reported that repetitive 400 mg/kg VPA reduced Abcc1 mRNA expression in GD20 rat placenta. Similarly, Rubinchik‐Stern et al. (2015) found that ABCC2 mRNA was reduced in BeWo cells after 2 days with 166 μg/mL VPA and 5 days with 83 and 166 μg/mL. Jinno et al. (2020) also observed Abcc3 mRNA expression decrease in GD13 placentas following repeated VPA exposure and Abcc4 mRNA reductions in GD20 placentas after both single and repeated doses. Conversely, Abcc5 gene expression increased in GD13 placentas after single and repeated exposures (Jinno et al. 2020).

BCRP, encoded by ABCG2, is another key efflux transporter, with expression increasing across gestation to regulate placental transport (Lye et al. 2013). Unlike MRPs, BCRP expression appears to increase with VPA exposure—Rubinchik‐Stern et al. (2015) reported elevated ABCG2 mRNA and BCRP protein levels at 83 and 166 μg/mL VPA after 5 days of incubation in BeWo cells. These findings suggest that VPA differentially regulates ABC transporters, potentially altering placental barrier function and the balance between fetal protection and nutrient transfer.

3.2.7. System L Transporters

Amino acids are vital for fetal growth and are actively transported across the placenta by over 10 families of amino acid carriers (Larqué et al. 2013). System L, a sodium‐independent bidirectional transporter, plays a key role in fetal nutrition by exchanging large, neutral amino acids (Cleal and Lewis 2008; Gaccioli et al. 2015). L‐type amino acid transporter 1 (LAT1; encoded by SLC7A5) and 2 (LAT2; encoded by SLC7A8) are key components of system L, as they regulate the amino acid availability between maternal and fetal circulation (Larqué et al. 2013).

VPA exposure has opposing effects on LAT1 and LAT2 expression. LAT1 expression decreased in BeWo cells after 2 and 5 days with 83 and 166 μg/mL VPA, though protein levels remained unchanged (Rubinchik‐Stern et al. 2015). Similar reductions in Slc7a5 mRNA expression were reported in GD13.5 mouse placentas (Meir et al. 2016) and GD20 rat placentas following single and repeated 400 mg/kg VPA doses (Jinno et al. 2020). In contrast, Slc7a8 mRNA expression increased in GD13 rat placentas after both single and repetitive 400 mg/kg VPA treatments (Jinno et al. 2020) and in BeWo cells in a concentration‐dependent manner after 2 and 5 days (Tetro, Moushaev, et al. 2021). VPA's differential effect on different system L transporters may disrupt the balance of the tightly regulated exchange of essential amino acids, potentially affecting fetal growth and development by altering the composition and availability of key nutrients.

3.2.8. Fatty Acid Transporting Protein

The fatty acid transporting protein (FATP; encoded by SLC27A4) family facilitates the transfer of essential fatty acids, such as linoleic acid, α‐linoleic acid, and long‐chain polyunsaturated fats. These fatty acids are critical during periods of rapid fetal growth, especially during the brain growth spurt, as they support the development of the nervous system and other tissues (Larqué et al. 2013). SLC27A4 expression was reduced following exposure to 42 μg/mL VPA for 5 days in an ex vivo human placenta model (Tetro et al. 2019). The reduction in SLC27A4 expression following VPA exposure suggests a potential disruption in the placental transfer of essential fatty acids, which are vital for fetal brain and nervous system development.

3.2.9. Key Insights and Future Directions for VPA's Effect on Nutrient Transporters

VPA exposure alters the expression of multiple placental nutrient transporters, which may have significant implications for fetal growth and development. Dysregulated transporter expression may limit nutrient availability, contributing to growth impairments, birth defects, and neurodevelopmental disorders seen with VPA exposure. For example, VPA‐induced downregulation of folate, choline, and/or thyroid hormone transporters may underlie its neurodevelopmental effects, while carnitine deficiency could contribute to congenital heart malformations. However, inconsistencies across studies complicate the interpretation of these findings. Figure 2 provides a visual representation of the differences in mRNA and protein nutrient transporter expression across in vitro, in vivo, and ex vivo models. It is hypothesized that differences in model type and timing of exposure likely underlie these discrepancies. For example, FOLR1 mRNA expression increased with VPA exposure in vitro (Kurosawa et al. 2018; Rubinchik‐Stern et al. 2015), but decreased in both GD20 rat and full‐term human placentas (Furugen et al. 2021; Rubinchik‐Stern et al. 2018), highlighting the influence of experimental model on transporter expression. Similarly, FOLR1 expression was significantly decreased in ex vivo full‐term human placentas but not early human placentas (Tetro et al. 2019), emphasizing the importance of exposure timing. Yet not all genes were inconsistently altered by VPA exposure. For example, decreases in expression across in vitro and in vivo models were observed for SLCO4A1, which further complicates the analysis of VPA's placental alteration.

FIGURE 2.

Schematic representation of changes in nutrient transporter gene and protein expression following VPA exposure across in vitro, in vivo, and ex vivo experimental models. The schematic illustrates the directionality of expression changes (increase or decrease) for key nutrient transporters, including folate [FRɑ (FOLR1), RFC (SLC19A1), and PCFT (SLC46A1)], choline [CTL1 (SLC44A1) and CTL2 (SLC44A2)], organic anions [OATP1A2 (SLCO1A2), OATP2B1 (SLCO2B1), OATP4A1 (SLCO4A1)], carnitine [OCTN1 (SLC22A4) and OCTN2 (SLC22A5)], drug efflux transporters [MRP1 (ABCC1) MRP (ABCC2), MRP3 (ABCC3), MRP4 (ABCC4), MRP5 (ABCC5), and BCRP (ABCG2)], amino acids [LAT1 (SLC7A5) and LAT2 (SLC7A8)], and fatty acids [FATP4 (SLC27A4)] in response to VPA exposure. (A) The in vitro panel highlights findings from studies using placental cell lines such as BeWo and JEG‐3 cells. (B) The in vivo panel highlights changes observed in animal models, particularly rat and mouse placentas which have two syncytiotrophoblasts layers denoted as SynT‐I and SynT‐II. (C) The ex vivo panel represents alterations reported in human placental perfusion studies. This figure is not intended to depict the precise location of these transporters within tissues or cells but rather provide a comprehensive overview of VPA's impact on transporter expression across different experimental models. This figure is inspired by Kozlosky et al. (2022) and created using BioRender.com.

Moreover, most studies conducted to date focus on mRNA expression, which does not always translate to altered protein expression or functional transporter activity, leaving uncertainty about whether VPA exposure truly limits nutrient availability. There is evidence that VPA can decrease RFC, CTL1, CTL2, and GLUT‐1 expression in BeWo cells (Kitahara et al. 2024; Rubinchik‐Stern et al. 2015; Tetro et al. 2019); however, these results require corroboration in vivo to account for real‐time pathophysiological changes (Liu et al. 2023). Future research should prioritize assessing changes in protein expression and nutrient concentrations across the placental barrier to establish a direct link between disrupted nutrient transport and VPA teratogenesis. This integrative approach will provide definitive insights into how VPA exposure compromises placental function and fetal nutrient supply, ultimately improving our understanding of its teratogenic mechanisms.

4. VPA‐Induced Epigenetic Changes in the Placenta

Several mechanisms have been proposed to explain VPA‐induced developmental toxicity, including increasing oxidative stress (Tung and Winn 2011), inhibiting the folate pathway (Lloyd 2013), suppressing nitric oxide signaling (Tiboni et al. 2021), and perturbing gene expression (Shafique and Winn 2020)—all of which could be linked by VPA's capacity to alter the epigenome. The epigenome encompasses the collection of chemical modifications to DNA and histone proteins that, without altering the underlying DNA sequence, regulate gene expression and cellular function. Such modifications determine the transcriptional status of genes, influencing their activation or repression, and are pivotal in processes such as cell differentiation, development, and response to environmental stimuli.

VPA modifies the epigenome by inhibiting histone deacetylase (HDAC) enzymes (Gottlicher et al. 2001), which remove acetyl groups (─COCH₃) from lysine residues on histone proteins. The positively charged lysine residues can then bind to the negatively charged DNA, thereby influencing chromatin structure and gene expression. VPA inhibits HDACs by binding to the zinc‐containing catalytic site within the HDAC pocket, disrupting the active site required for deacetylation (Abouzeid et al. 2007). HDACs are divided into four classes based on their structure and function, and VPA inhibits HDACs in Classes I and II (Gurvich et al. 2004). Class I HDACs (HDAC1, HDAC2, HDAC3, HDAC8) are found primarily in the nucleus, whereas Class II HDACs (HDAC4, HDAC5, HDAC6, HDAC7, HDAC9, HDAC10) shuttle between the nucleus and the cytoplasm to perform tissue‐specific functions (reviewed in de Ruijter et al. 2003). When HDACs are inhibited, the acetyl groups on histone tails accumulate, modifying chromatin structure and influencing the DNA's accessibility to transcription factors. These changes influence gene expression in diverse ways depending on the specific genomic and cellular context. The biological consequences of HDAC inhibition are highly dependent on which HDAC isoforms are affected, the cell type, and the genes involved.

Another way in which VPA alters the epigenome is by altering histone and DNA methylation. Histone acetylation and DNA methylation are interdependent processes that collectively influence chromatin structure, so perturbing one mechanism often cascades into alterations in the other. Therefore, following VPA induction of histone hyperacetylation, the tightly connected interplay between histone acetylation and DNA methylation is disrupted. This disruption in methylation status affects chromatin stability and gene regulation (Mello 2021). Ultimately, epigenetic modifications could plausibly lead to differential gene expression and disrupt pathways critical for placental development. In the placenta, the implications of VPA‐induced HDAC inhibition are not yet fully understood, but elucidation of how VPA influences the epigenetic landscape, particularly in the context of placental development and function, as well as what potential role epigenetics plays in mediating adverse pregnancy outcomes, is an important avenue of investigation.

4.1. Changes to the Placental Epigenome

VPA has been shown to alter the placental epigenome. For example, Downing et al. (2010) hypothesized that differences in VPA teratogenesis observed in B6 and D2 mice were partly due to strain differences in VPA‐induced HDAC inhibition (Downing et al. 2010). Therefore, to test this hypothesis, they reciprocally mated B6 and D2 mice and examined GD18 fetuses and placentas from F1 litters for changes in global histone protein acetylation following a single dose of 600 mg/kg VPA on GD9. They found that both strains showed a similar increase in histone 3 (H3) acetylation in fetuses following in utero VPA exposure, but fetuses from D2 dams exhibited a greater increase in histone 4 (H4) acetylation compared to fetuses from B6 dams. In contrast, the placenta showed smaller increases in acetylation, with B6 placentas displaying a higher increase in H3 acetylation than D2 placentas. While differences in VPA‐induced epigenetic modifications were observed between strains, no direct correlations were made between increased or decreased acetylation and the corresponding fetal and placental outcomes.

Furugen et al. (2021) conducted an in vivo study on pregnant female Wistar rats to further investigate the impact of VPA on global placental histone acetylation and HDAC expression. The results showed that both single and repeated administrations of 400 mg/kg VPA during mid (GD12 and GD9–GD12) and late gestation (GD19 and GD16–GD19) increased H3 acetylation, with repeated doses showing a significant 2.5‐fold increase. Similarly, in a previous study conducted in our lab, we examined VPA‐induced epigenetic changes 1‐, 3‐, and 6‐h post‐VPA exposure (400 mg/kg) on GD9 in the decidua of the CD‐1 mouse placenta. Our study revealed that VPA induced a significant increase in global H3 and H4 acetylation for up to 3‐ and 6‐h post‐dosing, respectively (Shafique and Winn 2021b). In addition, a more recent study from our lab found that in utero exposure to 600 mg/kg VPA in CD‐1 mice on GD13 resulted in several epigenetic changes (Jackson et al. 2024). The findings showed a significant increase in the staining intensity of acetylated H4 for up to 3‐h post‐dosing, a significant increase in the methylation of H3 lysine 4 for up to 24‐h post‐dosing, but a significant decrease in global DNA methylation at 1‐ and 24‐h post‐dosing. However, there was no change in global DNA methylation at the 3‐h time point (Jackson et al. 2024). While these studies provide valuable insights into the placental epigenetic changes induced by VPA, the results focus on global rather than gene‐specific changes, making it difficult to draw conclusions regarding the biological consequences of VPA‐induced epigenetic modulation in the placenta. Focusing on determining the precise role of these epigenetic modifications in mediating the teratogenic effects of VPA would provide more insights into this relationship.

4.2. Epigenetic Modifications as an Adverse Outcome Pathway

VPA‐induced HDAC inhibition may initiate an adverse outcome pathway by changing the epigenome and modulating gene expression, leading to placental toxicity. In placental tissue, VPA has been shown to decrease global DNA methylation and increase histone acetylation, which may drive the structural and functional changes observed with VPA exposure. In human embryonic stem cells, exposure to VPA led to extensive gene deregulation characterized by significant alterations in both transcriptional and microRNA expression profiles (Meganathan et al. 2015). These changes were directly linked to HDAC inhibition and histone hyperacetylation, resulting in disrupted stem cell differentiation. VPA repressed critical transcription factors involved in dorsal forebrain and neural tube development (e.g., OTX2, EMX2, and SOX10) while upregulating genes associated with ventral forebrain differentiation and axonogenesis, such as SLIT1, SEMA3A, and DLX2, which are involved in neuronal migration and axon guidance. This dual regulatory effect destabilized ectodermal lineage specification, highlighting the role of epigenetic modifications in mediating VPA's teratogenic effects. Further analysis revealed that VPA‐induced overexpression of miR‐378 was a key mediator of these transcriptional changes. miR‐378 directly suppressed dorsal forebrain markers, including OTX2 and EMX2, disrupting normal neuronal differentiation. Functional experiments demonstrated that knockdown of miR‐378 restored the expression of these transcription factors, confirming the role of miR‐378 in mediating VPA's effects. This study provides evidence that epigenetic perturbations, such as those initiated by VPA, can cascade into widespread transcriptional changes, ultimately compromising development.

The placenta originates from trophoblast stem cells within the trophectoderm: the outer layer of the blastocyst. Initially, these cells exhibit high developmental plasticity, but as gestation progresses, they differentiate into specialized cell types essential for placental structure and function (Hemberger and Dean 2023). When epigenetic regulation is disrupted, chromatin remodeling can compromise this carefully orchestrated process, jeopardizing both placental formation and function. Thus, it is plausible that VPA, which is known to disrupt chromatin structure, causes such placental changes.

While DNA methylation and histone acetylation patterns are shown to be altered in VPA‐exposed placentas compared to controls, the studies that reported these findings did not investigate the corresponding gene expression changes or physiological consequences of these epigenetic alterations. However, there is evidence that VPA exposure can alter the expression of genes similarly dysregulated in other placenta models, where such changes were linked to decreased DNA methylation and histone deacetylation. For example, Ohyama et al. (2023) demonstrated that VPA disrupts trophoblast conversion to syncytiotrophoblasts. In their study, 24‐h VPA treatment decreased the mRNA expression of genes ERVW‐1 and ERVFRD‐1, which encode for the proteins syncytin‐1 and syncytin‐2, respectively, and suppressed the cell fusion required for a syncytiotrophoblast‐like phenotype in both BeWo and human trophoblast stem cell models (Ohyama et al. 2023). While the authors did not investigate VPA‐induced epigenetic changes in this model, there is evidence to suggest that HDAC inhibition is responsible for this dysregulated differentiation. In a separate study, it was found that both the inhibition and the knockdown of HDAC1 and HDAC2 together prevented syncytiotrophoblast differentiation in BeWo cells (Jaju Bhattad et al. 2020). The link between HDAC inhibition and failure of syncytiotrophoblast formation could be in the differential expression of glial cell missing factor 1 (GCM1). Transcription factor GCM1 is pivotal in guiding trophoblast stem cell differentiation to syncytiotrophoblasts by modulating the expression of cell fusion proteins like syncytin‐1 and syncytin‐2 (Baczyk et al. 2009; Liang et al. 2010). Indeed, the same study that saw decreases in ERVW‐1 and ERVFRD‐1 mRNA expression also found that VPA exposure reduced GCM1 expression in differentiated BeWo cells (Ohyama et al. 2023). Decreased Gcm1 was previously linked to epigenetic modifications by Arima et al. (2006), who observed that placentas from DNA methyltransferase knockout mice failed to express Gcm1, resulting in impaired syncytiotrophoblast formation and consequently labyrinth structure. Because epigenetic alterations—specifically decreased global DNA methylation and histone deacetylation—and VPA exposure have both been linked to decreases in ERVW‐1, ERVFRD‐1, and GCM1 gene expression, it is possible that VPA impairs syncytiotrophoblast formation via HDAC inhibition.

Another way by which VPA's HDAC inhibition may cause placental toxicity is by impairing placental vascular development. The blood vessels of the placental vascular network develop via two key processes: vasculogenesis (formation of new blood vessels from precursor cells) and angiogenesis (growth of new blood vessels from existing ones). These tightly coordinated events ensure the expanding placental surface is well‐vascularized, enabling efficient blood flow and nutrient delivery to the fetus. VPA has been shown to disrupt angiogenesis in multiple models, including cancer cell studies and in mouse placentas. In cancer models, VPA‐induced HDAC inhibition is proposed to repress angiogenesis by destabilizing the oxygen‐sensitive alpha regulatory subunit of hypoxia‐inducible factor 1 (HIF‐1), leading to reduced HIF expression (Kim et al. 2017; Zhao et al. 2016). This destabilization occurs through hyperacetylation‐induced disruption of Hsp90 chaperone activity, leading to proteasomal degradation of HIF‐1α (reviewed in Liang et al. 2006). Although this destabilization of HIF‐1α is not mediated directly through chromatin remodeling and is instead an example of the direct consequences of failed deacetylation, this process can still significantly modulate angiogenic gene expression, consequently impairing placental development. Cowden Dahl et al. (2005) found that placentas from Hif‐1α deficient mouse embryos exhibit defective placental vascularization. Independently, VPA was shown to decrease HIF‐1α expression in BeWo cells (Kitahara et al. 2024) and decrease placental vascular density and pro‐angiogenic gene expression in mouse placentas (Ruyani and Sumarsono 2023), which suggests that VPA may modulate placental vascular development through the HIF‐1 signaling pathway. Moreover, the role of HIF‐1 extends beyond angiogenesis. HIF‐1 regulates genes involved in glucose transport, cellular proliferation, and invasion, all of which are essential for placental development. For instance, HIF‐1 activates the expression of glucose transporters GLUT‐1 and GLUT‐3, increasing the efficiency of glucose uptake (Chang et al. 2022; Mimura et al. 2012; Wicks and Semenza 2022). Consistent with this, Kitahara et al. (2024) reported simultaneous decreases in HIF‐1α and GLUT‐1 expression in BeWo cells exposed to VPA. When analyzed together, these independent studies suggest that VPA‐induced HDAC inhibition has broad implications for placental development, potentially due to the consequences of HIF‐1 pathway disruption.

VPA‐induced HDAC inhibition may also impact the expression of other nutrient transporters. Section 3.2 of this review describes VPA's effect on nutrient transporter expression, highlighting that VPA can modulate the expression of various transporters, which has implications for nutrient availability. However, to date, only one study has directly linked nutrient transporter expression to epigenetic changes. It was found that VPA‐induced HDAC inhibition in BeWo cells increased histone acetylation in the FOLR1 promoter region, leading to its elevated expression (Miyazawa et al. 2024). More studies connecting the dots between epigenetic changes and nutrient transporter expression need to be conducted to confirm this potential adverse outcome pathway.

Currently, it is well established that VPA can modify the placental epigenome, yet the direct impact of these epigenetic modifications on gene expression, protein synthesis, and placental physiology remains poorly understood. The potential implications of dysregulated epigenetics in the placenta are highlighted in Figure 3. Future research should focus on elucidating the connections between epigenetic changes in the placenta and their downstream effects on gene and protein expression, as well as the resulting physiological consequences. For instance, studies could investigate whether decreased DNA methylation or increased histone acetylation at specific loci leads to reduced expression of key proteins involved in trophoblast differentiation, like GCM1 and HIF‐1. These protein expression changes could subsequently be linked to physiological changes such as impaired syncytiotrophoblast formation and placental vascularization. To address these knowledge gaps, researchers could employ advanced techniques such as bisulfite sequencing to analyze DNA methylation patterns, chromatin immunoprecipitation for assessing histone modifications and transcription factor binding, RNA sequencing for transcriptomic profiling, and proteomics for protein expression analysis. These molecular approaches should be complemented by detailed structural and functional analyses of the placenta to uncover the mechanisms underlying VPA‐induced toxicity and its potential contributions to adverse pregnancy outcomes.

FIGURE 3.

Cartoon diagram of VPA's hypothesized mechanism of placental toxicity. (1) In placental tissue, VPA inhibits histone deacetylase (HDAC) enzymes, inducing global changes to the epigenome, including increased histone acetylation and decreased DNA methylation. (2) These epigenetic changes induce chromatin remodeling and, consequently, differential gene expression in placental trophoblast cells which can impair cell differentiation to mature trophoblast subtypes, like syncytiotrophoblasts (2a), and can alter the expression of nutrient transporters key to fetal growth and development (2b). (3) Overall, these changes in gene expression can alter placental structure and function, which can detrimentally impact fetal growth and development. This figure was created using BioRender.com.

5. Conclusion

Overall, the placenta warrants greater attention in understanding the mechanisms underlying VPA teratogenesis. VPA's ability to alter the epigenome and modulate gene expression carries profound implications for placental development, potentially contributing to the fetal toxicity observed with in utero VPA exposure. However, significant gaps remain in our understanding of how VPA impacts the placenta, emphasizing the need for further investigation into its specific mechanisms of action.

One of the most critical yet understudied areas in VPA teratogenesis is the role of placental sex. Placental sex plays a significant role in modulating responses to environmental exposures, including toxicants like VPA. Male and female placentas exhibit inherent differences in structure, gene expression, and response to maternal and environmental factors, even under healthy conditions (Goodman et al. 2023). For example, male placentas tend to be heavier and prioritize growth, reflecting differing resource allocation strategies (Ishikawa et al. 2003; Roland et al. 2014), whereas female placentas often exhibit greater adaptability to adverse maternal environments (Goodman et al. 2023). Studies in other toxicant models provide critical insights into the importance of placental sex in modulating exposure outcomes. Punshon et al. (2019) found that higher cadmium concentrations in the placenta were associated with reduced placental weight, but female placentas were more significantly affected, indicating their greater susceptibility to cadmium‐induced dysfunction. This susceptibility may stem from sex‐specific differences in gene expression and adaptive responses to toxicants. Similarly, Cuffe et al. (2011) found that short‐term maternal exposure to dexamethasone resulted in a significant reduction in placental weights in female fetuses but not in males, demonstrating the sex‐specific nature of placental responses to environmental factors. These differences suggest that placental sex could play a critical role in determining susceptibility to toxicant‐induced placental dysfunctions (Cuffe et al. 2011; Punshon et al. 2019). Despite this growing evidence, many studies on VPA fail to account for placental sex in their analyses, potentially obscuring critical insights into sex‐specific vulnerabilities. Incorporating placental sex as a variable in future research would improve the accuracy of toxicological assessments and lead to a clearer understanding of VPA's teratogenic risks.

Overall, future research should prioritize conducting comprehensive analyses of placental structure and function with the aim of directly linking these findings to changes in gene expression and epigenetic regulation. In studying placental structure, particular focus should be placed on characterizing the effects of VPA on trophoblast differentiation and evaluating changes across specific trophoblast subpopulations, such as syncytiotrophoblasts. In addition, a more thorough evaluation of VPA's impact on nutrient transporter expression is essential to determine whether VPA reduces protein expression and nutrient availability. These insights will help to clarify whether placental toxicity contributes to VPA teratogenesis. Key variables such as fetal sex, fetal birth defect status, dose, timing, and frequency of exposure, and experimental model should also be accounted for in future studies to understand their influence in VPA teratogenesis and their relevance to human pregnancy. By integrating these approaches, researchers might discover how VPA compromises placental function, laying the foundation for targeted interventions. These efforts would not only provide crucial insights into VPA‐induced placental toxicity but also advance our understanding of placental biology as a key mediator of fetal health. Moreover, they will inform the evaluation of how other therapeutic drugs affect the placenta, ultimately driving progress in maternal–fetal medicine and improving drug safety during pregnancy.

Author Contributions

Louise M. Winn: writing – review and editing, supervision, project administration, funding acquisition. Lauren T. L. Brown: writing – original draft, writing – review and editing, methodology, conceptualization. Delaine Pereira: writing – original draft, methodology.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The authors have nothing to report.