🧬 biology

Reconstituting Placenta–Embryo Crosstalk in a Microfluidic Chip for Human‑Relevant Developmental Toxicity Studies

This study presents a human-relevant microfluidic co-culture platform integrating a placental barrier model with hiPSC-derived embryoid bodies to dynamically recapitulate placenta-embryo crosstalk, thereby enabling more accurate mechanistic assessment of developmental toxicity by demonstrating how placental transport and metabolism modulate the effects of teratogens.

Original authors: Tina Buerki-Thurnherr, Manon Murdeu, Isabel Wegner, Xialein Lin, Jacob Folz, Paula Navascués, Andreas Hierlemann, Julia A. Boos, Vera Kissling

Published 2026-08-04

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Tina Buerki-Thurnherr, Manon Murdeu, Isabel Wegner, Xialein Lin, Jacob Folz, Paula Navascués, Andreas Hierlemann, Julia A. Boos, Vera Kissling

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine trying to understand how a new medicine affects a baby growing inside a mother's womb. For decades, scientists have had to rely on animal tests, which are like trying to guess how a human heart works by studying a frog's heart. It gives you a hint, but it's not the whole story. The real challenge is that the mother and the baby are connected by a super-specialized "security gate" called the placenta. This isn't just a wall; it's a busy, intelligent border control center. It decides what gets through, what gets filtered out, and even changes some chemicals before they reach the baby. If we want to know if a drug or a pollutant is safe for a developing human, we can't just test it on the baby cells alone; we have to test it with the placenta gatekeeper in the room.

This is where the story gets tricky. We can't just peek inside a pregnant woman's body to see these tiny interactions in real-time, and growing a whole human embryo in a lab dish is impossible and unethical. So, scientists have been trying to build tiny, artificial versions of this system in the lab. They use "microfluidic chips," which are basically tiny plastic devices with microscopic rivers and channels that can hold living cells. The goal is to recreate the conversation between the mother's side and the baby's side to see how they talk to each other and how they react to dangerous chemicals.

The Paper's Big Idea: A Tiny City with a Gatekeeper

In this study, a team of researchers built a high-tech, miniature city on a glass slide to solve this puzzle. They created a "human-relevant" model that mimics the critical connection between a pregnant mother and her developing baby. Think of their device as a two-story apartment building with a very special, semi-permeable floor between the top and bottom floors.

On the top floor (the "maternal" side), they grew a layer of cells taken from a real human placenta. These cells naturally fused together to form a thick, protective skin called the syncytiotrophoblast. This layer acts like the placenta's security guard, complete with tiny hair-like projections (microvilli) that increase its surface area, just like in a real human body. On the bottom floor (the "embryonic" side), they placed a tiny, floating ball of human stem cells called an "embryoid body." This ball is a mini-model of a very early human embryo, capable of turning into the three main building blocks of the body: the brain/nerves, the muscles/bones, and the gut.

The magic happens because these two floors are connected by a tiny, gravity-driven river. The researchers tilted the chip back and forth, creating a gentle flow that allowed nutrients and signals to drift from the top floor to the bottom, just like blood flowing between a mother and her baby. This setup allowed them to watch how the "security guard" (the placenta) interacted with the "growing baby" (the embryoid body) in real-time.

The Great Test: Two Villains, Two Different Stories

To see if their tiny city worked, the team introduced two famous "villains" known to cause birth defects: Valproic Acid (VPA) and All-Trans Retinoic Acid (RA). They wanted to see what happened when these villains tried to cross the security gate.

First, they tested Valproic Acid (VPA). When they exposed the baby cells without the placenta guard, the cells were mostly fine. But when they added the placenta guard to the mix, something surprising happened. The researchers observed that the concentration of VPA on the baby's side was actually higher than what was seen in a simple barrier without cells. This enrichment suggests that the placenta isn't just a passive wall; it appears to use active transport mechanisms to move the chemical across, resulting in a higher dose reaching the developing brain cells. Consequently, when the baby cells were exposed to the placenta, they received this enriched dose of the chemical, leading to a disaster for the developing brain cells. The baby cells lost their ability to form proper neural structures, showing that the placenta's active transport mechanisms significantly increased the exposure of the embryo to the toxin. This suggests that for VPA, the placenta acts more like an active transporter than a passive shield.

Then, they tested All-Trans Retinoic Acid (RA). This time, the story flipped completely. When the baby cells were exposed to RA without the placenta, they were damaged, and their brain structures failed to form. However, when the placenta guard was present, the baby cells were safe! The placenta acted like a chemical factory, breaking down the dangerous RA into a harmless version before it could reach the baby. The guard successfully filtered out the threat, protecting the developing embryo.

What This Means

The most important discovery here is that the placenta isn't a passive wall. It's an active participant that can change the outcome of a pregnancy. Depending on the chemical, the placenta might actively transport a toxin to the baby, block it, or even transform it into something else.

The researchers found that their tiny chip could mimic these complex behaviors with high accuracy. They showed that the placental barrier they built could transport drugs and nanoparticles in a way that matched real human data. They also proved that you can't just test drugs on baby cells alone; you have to test them with the placenta, because the placenta decides the final dose the baby actually receives.

By recreating this delicate dance between the mother's side and the baby's side, this study offers a new, human-based way to test for developmental toxicity. It suggests that in the future, we might be able to predict how drugs or pollutants affect a pregnancy without needing animal tests, simply by watching how our tiny, chip-based security guards handle the incoming traffic. The paper doesn't claim to have solved all the mysteries of pregnancy, but it has built a powerful new tool that suggests we are finally getting a clearer, more human picture of how the placenta protects—or sometimes fails to protect—our future generations.

Technical Summary: Reconstituting Placenta–Embryo Crosstalk in a Microfluidic Chip for Human‑Relevant Developmental Toxicity Studies

Problem Statement
Current developmental toxicity assessment relies heavily on in vivo animal studies, which offer limited mechanistic insight into human-relevant effects of drugs, nanoparticles (NPs), and environmental pollutants during pregnancy. While recent advances include human induced pluripotent stem cell (hiPSC)-derived embryoid bodies (EBs) for in vitro teratogenicity assessment, existing models typically expose embryonic systems directly to test compounds. This approach fails to capture the critical role of the placenta in modulating embryo-foetal exposure, transport, metabolism, and signalling. Consequently, there is a lack of predictive, human-relevant models that can reproduce maternal-foetal transport and elucidate how maternal exposures translate into adverse developmental outcomes.

Methodology
The authors developed a fully human-based co-culture platform on a microfluidic chip that integrates a functional placental barrier with a hiPSC-derived embryoid body (EB) to model early embryonic development.

Device Architecture: The microfluidic chip consists of two poly(dimethylsiloxane) (PDMS) layers separated by a semipermeable polyethylene terephthalate (PET) membrane (1 µm pore size). The device features six independent fluidic units. The "maternal" compartment houses the placental barrier, while the "embryonic" compartment contains the EB.
Placental Barrier Model: Primary human cytotrophoblast (CTB) cells isolated from term placentas are seeded on the maternal side of the membrane, while human placental vascular endothelial cells (HPVECs) are seeded on the embryonic side. The CTBs spontaneously fuse to form a syncytiotrophoblast (STB) layer. To ensure confluence, a double-seeding strategy was employed. The PDMS surface was coated with a plasma-deposited organosilicon thin film (SiO:CH) to minimize the adsorption of small molecules and NPs.
Embryonic Model: hiPSCs were aggregated in hanging drops to form undifferentiated EBs, which were then transferred to the chip's embryonic compartment on Day 4 in vitro (DIV4).
Flow Dynamics: Bidirectional tilting of the device (±5°) induces gravity-driven flow, enabling the exchange of nutrients, signalling molecules, and secretory factors between the placenta and the EB without mechanical pumps.
Experimental Workflow: The placental barrier was established over DIV0–4. EBs were formed in parallel and transferred to the chip on DIV4 for co-culture until DIV9. Compounds were introduced to the maternal compartment to assess transport and developmental toxicity.
Analytical Readouts: The study utilized immunofluorescence, transmission/scanning electron microscopy (TEM/SEM), hCG ELISA, and high-performance liquid chromatography coupled with tandem mass spectrometry (HPLC-ESI-MS/MS) to assess barrier integrity, morphology, endocrine function, and compound translocation.

Key Results

Barrier Formation and Function: The model successfully formed a polarized, syncytialized trophoblast layer with preserved endocrine function (hCG secretion) and barrier integrity. The barrier exhibited size-selective permeability: small molecules (sodium fluorescein) and small NPs (25 nm PMMA) crossed the barrier, while larger molecules (40 kDa FITC-dextran) and larger NPs (150 nm PMMA) were retained.
Physiological Relevance: Transport data for six reference compounds (e.g., antipyrine, caffeine, digoxin) across the on-chip barrier strongly correlated with human ex vivo placenta perfusion data (Pearson $r^2$ = 0.7374), validating the model's physiological relevance.
Placental Modulation of Toxicity (Valproic Acid - VPA): Exposure to the known teratogen VPA (1000 µM) resulted in disrupted neuroectodermal features (elongated lumens, altered rosette morphology) in EBs only when co-cultured with the placental barrier. In monoculture (direct exposure), VPA effects were limited. This discrepancy was attributed to the placenta actively transporting VPA, leading to higher concentrations in the embryonic compartment compared to the acellular control.
Placental Protection (All-Trans Retinoic Acid - RA): Conversely, exposure to RA (11 nM) caused a loss of neural rosette formation in EB monocultures. However, in the presence of the placental barrier, EBs maintained normal neural rosette morphology. LC-MS analysis detected the metabolite 4-hydroxy retinoic acid (4-OH-RA) in the embryonic compartment only when the placenta was present, indicating that the placenta metabolizes RA, thereby preventing its accumulation and toxicity in the embryonic compartment.
Viability: EBs remained viable throughout the 9-day co-culture period, with preserved metabolic activity (ATP levels), confirming the system's stability for toxicity screening.

Significance and Claims
The paper claims that this platform successfully reproduces key features of the dynamic placenta-embryo interplay, providing a human-relevant in vitro system for mechanistic studies of placental contributions to developmental toxicity.

The authors emphasize that their findings demonstrate the placental barrier's active role in regulating embryonic/foetal exposure through transport and metabolism. Specifically, the study shows that:

Placental transport can exacerbate toxicity: In the case of VPA, the placenta facilitated higher embryonic exposure, leading to toxicity that was not observed in direct exposure models.
Placental metabolism can protect against toxicity: In the case of RA, the placenta metabolized the compound, shielding the embryo from teratogenic effects that were evident in direct exposure models.

The authors conclude that incorporating placental function is essential for accurate developmental toxicity assessment. They argue that their integrated system allows for the investigation of exposure-response relationships and placenta-mediated effects that cannot be obtained using isolated tissue models, representing a critical step toward mechanism-informed, predictive investigation of human developmental toxicity. The paper does not claim to replace in vivo studies entirely but positions the model as a tool to bridge the gap between current in vitro assays and human physiological reality.

More like this