Translational and Regulatory Sciences
Online ISSN : 2434-4974
REVIEW
Alternatives to animal testing for assessing transplacental transfer
Takeshi HORIHirokazu KAJI
Author information
JOURNAL OPEN ACCESS FULL-TEXT HTML

2025 Volume 7 Issue 1 Pages 1-7

Details
Abstract

In drug development, reproductive and developmental toxicity tests play a crucial role in safeguarding the health of future generations. However, effective methods for predicting toxicity have not yet been established. Differences among species make toxicity testing challenging because it is difficult to extrapolate animal testing results to humans. Recently, in vitro testing using human cells has gained attention as a promising alternative, offering the potential to complement or partially replace traditional animal reproductive and developmental toxicity tests. In this review, we introduce several ‘placental barrier models’ for assessing chemical transfer from the mother to the fetus, including both conventional approaches and our latest in vitro methods.

Highlights

Assessment of the transplacental transfer of substances is essential for the development of safer drugs for fetuses. This review introduces various models including ex vivo perfusion systems, insert column models, and microfluidic chips. A notable advancement has been the development of trophoblast stem cell-based placental barrier models that offer enhanced physiological relevance. These innovations align with the principle of the 3Rs, paving the way for reliable reproductive and developmental toxicity tests, while reducing reliance on animal testing.

Introduction

Reproductive and developmental toxicity tests assess substances related to reproductive health, such as those affecting conception and fetal development, and are essential for safeguarding the health of parents and future generations. One notable example that highlights the importance of such assessments is the thalidomide tragedy. In the 1950s and 1960s, thalidomide was widely prescribed as a sedative and antinauseant, particularly for morning sickness during pregnancy. Tragically, its use has led to the death of approximately 2,000 unborn babies and has caused limb malformations in over 10,000 children worldwide, particularly in Germany and Japan [1]. Notably, animal studies using rodents have not revealed thalidomide-induced malformations, underscoring the complexities of reproductive and developmental toxicity, and the challenges of accurate prediction. Even without apparent congenital effects such as those caused by thalidomide, the Developmental Origins of Health and Disease (DOHaD) hypothesis suggests that prenatal chemical exposure may significantly increase the risk of chronic diseases in adulthood, affecting millions globally [2].

The safety of the substances used is often evaluated through extensive animal testing. Reproductive and developmental toxicity tests require a large number of animals, which is costly and raises significant concerns for animal welfare [3, 4]. Reproductive toxicity testing mainly focuses on parental reproductive capability (the establishment and maintenance of pregnancy), whereas developmental toxicity testing primarily examines the toxicity that disrupts normal fetal and embryonic development [5]. Reproductive and developmental toxicity tests are generally conducted in accordance with the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) guidelines for pharmaceuticals and international guidelines (mainly from the Organization for Economic Cooperation and Development [OECD]) for chemicals. In addition, the European Union Registration, Evaluation, Authorization, and Restriction of Chemicals (REACH) regulations mandate further safety testing of chemicals, requiring companies that produce or import over 1 ton of chemicals per year to register these substances [5]. REACH encompasses an estimated 70,000 to 100,000 substances [5,6,7,8]. Although REACH was implemented in 2008, by December 2022 approximately 2.7 million animals have been used for reproductive and developmental toxicity testing, and this number continues to increase [9, 10].

The placenta not only produces hormones to sustain pregnancy and facilitates gas and nutrient exchange between the mother and fetus, it also serves as a barrier, limiting fetal exposure to various xenobiotics, including certain drugs present in the maternal blood [11]. Despite this, certain xenobiotics can cross the placental barrier. Therefore, the accurate quantification of drug transfer through the placenta is vital for predicting potential fetal toxicity at the drug development stage. However, assessing placental permeability poses challenges, as the placental structure and cellular composition differ significantly between humans and experimental animals [12, 13]. Following recent advances in human-specific therapeutics, such as antibodies, nucleic acids, and peptide drugs, the reliance on animal testing for reproductive and developmental toxicity testing has been increasingly questioned. As the limitations of animal testing have become increasingly evident, new approach methodologies (NAMs) have been developed, including in vitro methods that utilize human tissues and cells as complements or partial replacements for animal testing. NAMs align with the 3Rs principle (replacement, reduction, and refinement) [14], a guiding framework for animal testing that has gained substantial support worldwide and are recognized as alternative to animal testing. In this review, we highlight the latest alternatives to animal testing for reproductive and developmental toxicity evaluation, focusing on placental barrier permeability assessment methods from conventional to cutting-edge approaches currently under development.

Structure and Properties of the Placental Barrier

When evaluating the placental permeability, it is essential to consider the structure and properties of the placental barrier. Significant differences exist among species in terms of the placental structure and cellular composition.

In early human pregnancy, the primary placental barrier comprises a bilayer of outer syncytiotrophoblast (ST) and inner cytotrophoblast (CT) cells (Fig. 1A). CT cells align laterally to form a nearly continuous monolayer [15], and ST cells emerge from the fusion of CT cells, creating a large single-cell layer that covers the surface of the villi. This unique ST layer consists of a vast fused cytoplasm containing tens of billions of nuclei with a surface area of over 12 square meters [16]. By term, CT cells largely disappear, implying that CT cells, as precursors of ST cells, are important for early villous development and morphological changes (Fig. 1B). Additionally, transporters and receptors on ST and CT cells facilitate substance transport, with endocytosis occurring in ST cells and probably in CT cells [17]. The release of syncytial nuclear aggregates (syncytial knots) is another dynamic biological event occurring at the maternal-fetal interface [18]. In addition to trophoblasts, other cellular structures also influence substance transfer. Fibroblasts within the CT layer form the villous core, and vascular fetal blood vessels, consisting of endothelial cells within the villi, also contribute to this barrier. Thus, these cells should also be considered in permeability testing to evaluate the transfer of substances from maternal to fetal blood.

Fig. 1.

Cross-sectional structures of human placental villi. Cross-sections of the human placental villus at first trimester (A) and at term (B). ST: syncytiotrophoblast; CT: cytotrophoblast; EC: endothelial cell.

To evaluate the placental permeability of substances, it is necessary to select a model that considers their similarity to the actual human placenta and their utility as a testing system. Below, we introduce four model types: animal, ex vivo perfusion, insert column and chip.

Methods for Assessing the Placental Permeability of Substances

Animal models

Traditionally, animal models (particularly mice and rats) have been used to investigate the effects of substances on fetal development. These models are valuable for studying complex biological processes involving multiple organ systems, as they allow for the assessment of both placental permeability and the impact of the substance on embryonic and fetal development. Although mice, rats, and humans all have a hemochorial placenta, where fetal trophoblasts (chorial) are in direct contact with maternal blood (hemo) [19, 20], there are notable structural differences.

Human placentas are hemomonochorial, with a main single-layered trophoblast layer (as in the late-stage placenta), whereas mice and rats’ placentas are hemotrichorial, with three trophoblastic layers (Fig. 2A and 2B) [21, 22]. Additionally, mouse and rat placentas have a labyrinthine structure that differs significantly from that of the human placenta. In addition to these structural differences, protein expression profiles in mouse and rat placentas differ from those in humans, resulting in variations in substance transport capabilities [23, 24]. Therefore, to accurately assess placental permeability in humans, it is crucial to use data derived from human tissues and cells.

Fig. 2.

Comparison of cross-sections of hemomonochorial and hemotrichorial placental barriers. (A) hemomonochorial, a single-layer trophoblast barrier, is observed in the human placenta at term. (B) hemotrichorial, a three-layer trophoblast barrier, is observed in the placentas of mice and rats. ST: syncytiotrophoblast; CT: cytotrophoblast; EC: endothelial cell; S-TGC: sinusoidal trophoblast giant cell.

Ex vivo human placental perfusion model

The ex vivo perfusion model allows the assessment of the placental permeability of substances using the entire placenta [25,26,27]. In this system, tubes are often inserted into the chorionic lumen on the maternal side, and into the vein and artery on the fetal side of the placenta, with solutions being perfused artificially using a pump. By circulating a drug into the maternal side, researchers can evaluate its transfer to the fetal circulation by measuring the drug concentrations in the maternal and fetal solutions. This system provides valuable data using human tissues and offers insights closely aligned with human physiology [28,29,30,31,32,33,34,35]. However, it is technically difficult to construct a reliable perfusion setup. Properly inserting and securing the tubes within the placental tissue is challenging and often leads to issues such as leakage. The method also has limitations owing to its reliance on delivered placentas, which limits sample availability and consistency as the results may vary between individuals. Additionally, as the system requires a term placenta, it cannot directly evaluate drug permeability during the organogenesis stage, when sensitivity to substances is the highest [36]. Another limitation is that the exact surface area of the placental villi is difficult to measure accurately, preventing the calculation of the apparent permeability coefficient (Papp), which is possible with other model types such as insert column systems. This restriction hinders direct comparison across placental samples with varying surface areas.

Insert column models

Insert columns placed in standard culture plates are commonly used to evaluate the permeability of substances across cell barriers in vitro [37, 38]. These insert columns have a permeable membrane, such as a porous or collagen-based sheet, at the bottom. By culturing cells on a membrane, a cell sheet that completely covers the membrane can be constructed. The permeability of the substances was tested by applying them to one side of the cell sheet and measuring their amount on the other side over time.

In the early stages of pregnancy, the human placental barrier is composed of a ST, a CT, a stromal cell, and a fetal vascular endothelial cell layer, arranged in that order from the maternal blood side [39]. The ST layer constitutes the main barrier; therefore, trophoblasts have been used to develop simple placental barrier models.

When constructing an in vitro placental barrier, it is essential to choose appropriate trophoblast cells for the model. Human choriocarcinoma-derived trophoblast cell lines such as BeWo and JEG-3 offer high proliferation rates and ease of maintenance. BeWo cells are widely used because they differentiate into ST-like cells in response to forskolin, which increases intracellular cyclic adenosine monophosphate (cAMP) and leads to the partial fusion of cells [24, 40]. JEG-3 and Jar cells (a cell line from the human hydatid mole) are less suitable barrier models, as they do not fuse under standard culture conditions, despite their ability to produce human chorionic gonadotropin (hCG) [41, 42]. Some models incorporate both BeWo and endothelial cell layers on either side of a permeable membrane, creating a more faithful placental barrier [39, 43]. However, even in BeWo models, the fusion rate of cells is significantly lower than that observed in vivo [44]. Additionally, choriocarcinoma-derived trophoblast cell lines accumulate genetic mutations over multiple passages, leading to genetic and phenotypic divergence from in vivo trophoblast cells [19]. These cells show different expression patterns of genes regarding transporters and drug-metabolizing enzymes, which may lead to results that differ from those in human placental cells [24, 45].

Primary trophoblasts isolated directly from human placentas retain in vivo functions and avoid the limitations of trophoblastic cell lines derived from choriocarcinoma [46,47,48,49]. These cells rapidly differentiate into ST cells when cultured in standard media, making their long-term maintenance difficult. Consequently, these cells must be freshly isolated for each experiment, which is time-consuming and can introduce variability between samples. Although some studies have reported models using primary cells [50, 51], an effective ST barrier model has not yet been sufficiently validated. Difficulties in maintaining primary trophoblasts also hinder the ability to perform repeated studies to identify suitable cell culture conditions for creating in vitro placental barrier models.

Chip models

To culture cells in environments that closely mimic in vivo conditions, microfluidic chip technology, also known as organ-on-a-chip or microphysiological systems, has been developed using microfabrication techniques [52,53,54]. These small chips (typically 2–3 cm) feature channels of varying designs, with some channels measuring several hundred micrometers in width and more than one centimeter in length. Specific organ-derived cells can be introduced into the chip, allowing for a more physiologically relevant culture environment by precisely controlling cell positioning, medium flow, extracellular matrix composition, and other factors [55].

In the human placenta, trophoblastic cells of the placental villi are immersed in flowing maternal blood within the intervillous space. To mimic this circumstance, BeWo cells are introduced into a chip channel and cultured under medium perfusion. Their differentiation and microvilli formation are promoted, providing information on the differentiation of ST cells [56]. By exposing trophoblasts cultured on chips to drugs or nanoparticles under perfusion, in vitro results that closely reflect in vivo interactions can be expected [57,58,59].

Models Using Human Placental Stem Cells

In vitro cell culture models are promising alternatives to animal and ex vivo perfusion models. However, a major limitation is that they often rely on cancer cell lines, which differ significantly from in vivo trophoblastic cells. To address this limitation, trophoblast stem (TS) cells with properties similar to those of in vivo cytotrophoblasts that retained capacity for proliferation were established.

In 1998, Tanaka et al. first established mouse TS cells [60], and later developed a placental barrier model using these cells [61]. However, the derivation of human TS cells has not been achieved for a long time. Initial attempts to generate human TS-like cells from primed pluripotent stem cells were not fully successful because these cells did not meet the criteria proposed by Lee et al. for human primary first-trimester trophoblast cells. These criteria included specific protein marker expression (GATA3, KRT7, and TFAP2C), HLA class I profile, hypomethylation of the ELF5 promoter, and expression of microRNAs (miRNAs) from the chromosome 19 miRNA cluster (C19MC) [62]. Later, Okae et al. reported conditions for establishing human TS cells from early placental trophoblasts or blastocysts that met Lee et al.’s criteria, and resembled cytotrophoblast cells shortly after implantation (8–10 days post-implantation) [63, 64].

Human TS cells offer advantages over primary trophoblast cells owing to their high proliferative ability in 2D cultures, without changing their stemness for over 5 months or more than 80 passages. This capability makes these cells highly suitable for the development of human placental barrier models with ST layers. Following the establishment of human TS cells, these were cultured in chips, although early models had limited ST-layer coverage of approximately 50% [65]. Recently, Hori et al. reported an insert column-type barrier model with nearly 100% ST coverage [66] (Table 1). This model displayed microvilli on the ST layer surface, similar to in vivo villi, and its barrier permeability to substances such as caffeine, antipyrine, and glyphosate showed a tendency similar to that of ex vivo perfusion systems. In the future, it will be important to validate the biological similarity of the model to in vivo placental villi using a more comprehensive range of substances. Furthermore, this model has the potential to allow the study of the effects of pathogens (e.g., viruses and bacteria), nanoparticles, and other substances on the placental barrier, thereby expanding our understanding of placental interactions with various substances [67,68,69,70].

Table 1.Comparison of placental barrier permeability assessment methods

Model Advantages Disadvantages
Animal models - Useful for studying complex biological processes. - Differences between animal and human placentas.
- Evaluation of both placental permeability and fetal effects. - Ethical concerns and high costs.
Ex vivo human placental perfusion model - Use of the human placental organ to obtain data reflecting human physiology. - Cumbersome setup with risk of leakage.
- Measurement of drug transfer rates to the fetus. - Limited availability of placentas.
- Not suitable for early pregnancy studies.
Insert column models - Relatively simple and reproducible. - Use of cancer cell lines that differ from normal trophoblasts.
- Useful for the calculation of the apparent permeability coefficient (Papp). - Lack of established syncytiotrophoblast (ST) barrier models with primary cells.
Chip models - Mimics in vivo conditions with controlled medium flow and microfabrication techniques. - Limited ST layer coverage in early models.
- Analysis of cellular responses under dynamic conditions. - Requires specialized technology and skills.
Recent models combining human placental stem cells and insert columns - Nearly 100% ST coverage. - Need for further characterization.
- High physiological relevance, including both structure and function. - Need for strict protocols to achieve international standardization of permeability testing methods.
- Capability for repeated model construction using human trophoblast stem (TS) cells.

Discussion

The establishment of human TS cells and their use in placental barrier models is a significant step towards a more physiologically relevant screening of substance permeability [66]. It is anticipated that permeability testing using this model will become standardized and widely adopted as an alternative to animal testing for reproductive and developmental toxicity assessments. To realize this potential, further studies are required to thoroughly characterize the placental barrier model and improve its usability and reliability as a test system.

The drug development industry is showing a heightened interest in developing alternatives to animal testing. For instance, in the United States, the Food and Drug Administration (FDA) Modernization Act 2.0, passed in December 2022, eliminated the requirement for animal testing in non-clinical studies, fueling interest in alternative approaches [71]. This movement is global and translational research aimed at leveraging laboratory-developed cell culture systems for drug development is expected to grow in importance.

Funding Sources

This work received partial financial support from JSPS KAKENHI (Grant Numbers 23H01821, 22K18936 and 21K04852) (to T.H. and H.K.), the Mandom International Research Grants on Alternatives to Animal Experiments (to T.H.), AMED-CREST (Grant Number JP21gm1310001) (to T.A. and H.K.), the JST Adaptable and Seamless Technology Transfer Program through Target-driven R&D (Grant Numbers JPMJTM22BD) (to H.K.), the Research Center for Biomedical Engineering (to T.H. and H.K.), and TMDU priority research areas grant (to T.H.).

Conflict of Interest

The Institute of Science Tokyo filed a patent application covering the protocol and methods for the generation of human trophoblast organoids. T.H. and H.K. own stock as members of a recently established company, HPS Inc., and could potentially receive compensation from the company.

Acknowledgments

The authors thank Ms. Inês M. Gonçalves for her valuable comments on this manuscript.

References
 
© 2025 Catalyst Unit

This article is licensed under a Creative Commons [Attribution-NonCommercial-NoDerivatives 4.0 International] license.
https://creativecommons.org/licenses/by-nc-nd/4.0/
feedback
Top