Building Efficient Stem Cell-Derived Mouse Blastocyst Models for Environmental Studies

Summary

This protocol describes the generation and analysis of iGATA4-blastoids, a scalable stem cell–based model that recapitulates key features of the mouse blastocyst, including lineage organization and cavitation, and enables reproducible, high-throughput assessment of environmental and molecular perturbations affecting early embryonic development.

Abstract

Stem cell–based embryo models offer a powerful platform for investigating early mammalian development and its sensitivity to environmental perturbations. Here, a robust and accessible protocol is presented for generating high-fidelity mouse blastoids using embryonic stem cells (ESCs), trophoblast stem cells (TSCs), and embryonic stem cells engineered for inducible GATA4 expression (iGATA4-ESCs). This system produces iGATA4-blastoids that accurately recapitulate natural E4.5 blastocyst morphology, lineage proportions, and transcriptional signatures, providing a reliable in vitro model of the primitive endoderm (PE), epiblast (EPI), and trophectoderm (TE) compartments. This protocol first demonstrates how to maintain and prepare the three stem cell populations required for blastoid formation. The step-by-step assembly of iGATA4-blastoids under pre-implantation culture conditions is then outlined, highlighting critical experimental parameters. To enable rigorous characterization, a tested workflow for blastoid fixation, staining, and imaging is provided. Finally, an example is presented demonstrating how iGATA4-blastoids can be used for high-throughput assays. By exposing large cohorts of blastoids to defined environmental factors, changes in morphology, lineage allocation, and developmental progression can be quantified rapidly and reproducibly. Together, this protocol establishes a scalable platform for modeling pre-implantation development, dissecting early lineage decisions, and performing multiparametric screens to assess environmental or molecular influences on blastocyst formation.

Introduction

Understanding how early mammalian embryos self-assemble and how common environmental factors may negatively affect their formation requires model systems that are reproducible, experimentally accessible, and scalable. Integrated embryo-like models generated by combining embryonic and extraembryonic stem cells provide such a platform^1,2. This protocol describes how embryonic stem cells (ESCs), trophoblast stem cells (TSCs), and ESCs engineered for inducible GATA4 expression (iGATA4-ESCs) can be assembled under pre-implantation conditions to produce blastoids with robust and properly proportioned primitive endoderm (PE), epiblast (EPI), and trophectoderm (TE) lineages. Recent work has demonstrated that these structures, termed iGATA4-blastoids³, resemble the E4.5 blastocyst in lineage proportions, transcriptional signatures, and functional marker expression. When cultured in vitro, 12.5% of these structures transition to post-implantation morphology. Importantly, they are amenable to toxicology and environmental screens.

The development of this technique was motivated by limitations in pre-existing blastoid systems^2,4,5. Although many protocols^{6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21} generate cavitated structures resembling early blastocysts, most do so with relatively low efficiency or produce structures with incomplete PE formation, undefined cell identities, or incorrect lineage proportions. These inconsistencies reduce their suitability for mechanistic studies and high-throughput perturbation screens. Because PE specification in vivo follows the activation of GATA6, SOX17, GATA4, and SOX7²², and defects in either the PE or its derivative, the visceral endoderm (VE), impair peri-implantation development^{23,24,25,26,27}, enforcing this transcriptional program in vitro improves blastoid fidelity. ESCs carrying a conditional GATA4 transgene robustly transdifferentiate into VE and, when combined with ESCs and TSCs under post-implantation conditions, generate integrated embryo models that gastrulate and initiate organogenesis^28,29,30,31. Here, this principle is adapted to pre-implantation conditions, demonstrating that iGATA4-ESCs reliably generate PE and yield blastoids that closely resemble natural blastocysts³.

This method overcomes previous limitations in blastocyst modeling using stem cells and offers high blastoid formation efficiency (80% of structures formed are blastoids, 75% of which demonstrate proper lineage organization and proportions) and high embryo-likeness. Moreover, the modular assembly of iGATA4-blastoids, in which each blastocyst lineage is derived from an independently cultured stem cell population, provides direct experimental access to each embryo compartment. These properties make the system particularly well suited for multiparametric screening applications. Because it yields large cohorts of morphologically consistent structures with accurately proportioned TE, EPI, and PE lineages, it enables parallel perturbations and quantification of lineage-specific responses with statistical power unattainable using natural embryos.

As a proof of principle, the effects of common environmental factors presented at physiological ranges^{32,33,34,35,36,37,38} on blastoid development were measured. The impact of caffeine, nicotine, ethanol, and amino acid modifications on cavitation rate, morphology, fitness, and lineage proportions was assessed across a range of concentrations and temporal treatment windows. Most conditions (caffeine, ethanol, nicotine, and those mimicking low-protein diets) resulted in aberrant blastoid formation and impaired structural quality. In contrast, increasing the concentration of branched-chain amino acids (similar to high-protein diets) improved blastoid formation efficiency and quality. These results mirrored effects observed in in vitro–cultured natural mouse blastocysts³, demonstrating that this system captures subtle, dose-dependent phenotypes and supports scalable, multidimensional screening pipelines not feasible with natural embryos.

Within the broader context of embryo modeling, this technique demonstrates that the same cells used to model integrated post-implantation development can also be deployed as pre-implantation models. This method is particularly suitable for applications requiring scalable modeling of blastocyst-like structures with accurate PE formation and robust lineage integration. However, as with other stem cell–based models of the mouse blastocyst, iGATA4-blastoids fail to implant and develop further after transfer to a host mouse and are therefore not suitable for studies of embryo–maternal interactions. Overall, this protocol provides a reliable and accessible approach for generating high-fidelity blastoids that model early cell-type decisions in the mammalian embryo.

Protocol

All procedures described in this protocol involve established cell lines and were performed in accordance with institutional biosafety guidelines at the California Institute of Technology. No live animal or human subject work was conducted.

Multiple ESC and TSC lines have been tested and found to support iGATA4-blastoid formation. ESCs, iGATA4-ESCs, and TSCs from previous studies²⁸ have been used successfully to complete all the steps described in this protocol. The same authors describe the generation of iGATA4-ESCs from a previously established CD1 (ICR) mouse ESC line.

TSCs are cultured on top of commercially available mouse embryonic fibroblasts (MEFs). All chemicals, reagents, kits, equipment, and software used in this protocol are listed in the Table of Materials.

1. Culture of mouse stem cells for blastoid formation

1. Preparation of MEF feeder plates
   1. Coat a 6-well cell culture plate with 1.5 mL of 0.2% (w/v) gelatin in deionized (DI) water per well. Incubate at 37 °C in normoxia for at least 45 min.
   2. During incubation, prepare a 15 mL centrifuge tube with 4 mL of FC media (Table 1) pre-warmed to 37 °C.
   3. Thaw the MEF vial in a water bath using a figure-eight motion until only a small ball of ice remains. Transfer the contents into the prepared tube, then pipette up and down to thoroughly mix. Centrifuge at 500 × g for 4 min.
   4. Aspirate the supernatant and resuspend the pellet in 1 mL of FC medium. Adjust to a sufficient volume to plate ~160,000 cells per well in 1.5 mL per well.
   5. Aspirate the gelatin and plate 1.5 mL of MEFs per well. Incubate in normoxia overnight without disturbance.
   6. Replace the medium with fresh 1.5 mL per well FC medium the following day. Change the medium every 48 h. Use MEFs 2–7 days after plating. MEFs may remain healthy for longer, but their ability to support TSC growth may be compromised.

CAUTION: Perform all mouse cell culture in biosafety level (BSL) 1 laminar flow cabinets using the BSL-2 technique.

2. Thawing and passaging of mouse embryonic stem cells (WT ESCs and iGATA4-ESCs)
   1. Thaw WT ESCs or iGATA4-ESCs following steps 1.1.1–1.1.3. After centrifugation, resuspend the cells in 1 mL of N2B27/2i/LIF medium (Table 1) and seed ~50,000 cells per well in a gelatin-coated 6-well plate with ~1.5 mL per well. Incubate at 37 °C in normoxia and change medium every 48 h.
   2. Ensure that colonies are tightly packed and round with minimal protrusions before passaging. Passage every 2–3 days.
      NOTE: WT ESCs and iGATA4-ESCs are indistinguishable morphologically until induction with doxycycline (DOX).
   3. Coat a 6-well plate with 1.5 mL of 0.2% gelatin in DI water per well. Incubate at 37 °C for at least 45 min.
   4. Add 500 µL of cell dissociation reagent per well and incubate for 3 min. Add 2 mL of FC medium while gently pipetting around the edges. Confirm >90% detachment under a microscope. If cells remain attached, gently pipette to aid resuspension.
   5. Transfer the cell suspension into a 15 mL centrifuge tube and centrifuge at 500 × g for 4 min.
   6. Aspirate the supernatant and resuspend the pellet in 1 mL of N2B27/2i/LIF medium. Determine cell concentration using 10 µL on a hemocytometer. Adjust volume to seed ~50,000 cells per well.
   7. Aspirate the gelatin and seed the cells. Incubate in normoxia. Change medium the following day and every 48 h thereafter.
3. Thawing and passaging of mouse trophoblast stem cells (TSCs)
   1. Thaw TSCs following steps 1.1.1–1.1.3. After centrifugation, resuspend the cells in 1 mL of TSF4H medium (Table 1) and seed ~20,000 cells per well in a MEF-coated 6-well plate. Incubate at 37 °C, 5% CO₂, 5% O₂ (hypoxia) and change medium every 24 h.
   2. Ensure colonies are large (>75 µm diameter) and flattened before passaging. Add 500 µL of cell dissociation reagent per well and incubate for 5 min. Add 2 mL of FC medium while gently pipetting. Confirm >90% detachment.
   3. Transfer cells to a centrifuge tube and spin at 500 × g for 4 min.
   4. Resuspend in 1 mL TSF4H medium. Determine concentration and adjust to ~20,000 cells per well.
   5. Plate onto MEFs and incubate in hypoxia. Change medium every 24 h and passage every 2–3 days before colony merging.

2. Formation of iGATA4-blastoids

1. Preparation of Microwell aggregation plate for cell aggregation
   1. Add 500 µL of anti-adherence rinsing solution to each well of a Microwell aggregation plate. Centrifuge at 2,000 × g for 5 min with proper balancing to remove trapped air from the microwells. Incubate at 37 °C in normoxia for at least 1 h or overnight.
2. Blastoid preparation, day 0 — Preparation and plating of WT and iGATA4-ESCs
   1. Pre-warm N2B27/2i/LIF, FC, and FCYL media (FC + 10 ng/mL mouse recombinant LIF (mLIF) + 2 µM ROCK inhibitor Y27632) to 37 °C. Prepare 0.2% gelatin in DI water and have the cell dissociation reagent ready.
   2. Aspirate anti-adherence solution without touching the well bottom. Wash twice with 1 mL PBS and once with 1 mL FCYL medium. Pipette along the wall to avoid damaging the microwells.
   3. Prepare a gelatin-coated 6-well plate (≥45 min incubation) for passaging cells.
   4. Harvest WT and iGATA4-ESCs by aspirating medium and adding 500 µL cell dissociation reagent per well. Incubate for 3 min at 37 °C. Add 2 mL FC medium and gently pipette. Confirm complete detachment under a microscope. Transfer to a 15 mL tube and centrifuge at 500 × g for 4 min.
   5. Resuspend in 1 mL FCYL medium. Determine cell concentration using 10 µL on a hemocytometer. Adjust to 500,000–700,000 cells/mL. Calculate the volume required for 5 WT ESCs and 5 iGATA4-ESCs per microwell (6,000 cells per well).
      NOTE: Optimize for other ESC lines if needed. A starting range is 4–7 ESCs per microwell with a 1:1 ratio of ESCs to iGATA4-ESCs.
   6. Prepare 500 µL of cell suspension per well. Add dropwise to the Microwell aggregation plate using a 1000 µL pipette. Centrifuge at 100 × g for 3 min and verify even distribution. Incubate for 6 h until aggregates form.
   7. Passage remaining ESCs as described in Section 1.2.
3. Blastoid preparation, day 0 — Preparation of TX medium
   1. Prepare ~20 mL TX basal medium using DMEM/F12 (Table 2). Prepare fresh for each experiment.
      1. Add 200 µL 200 mM L-alanyl-L-glutamine supplement (final 2 mM). Store stock at 4 °C.
      2. Add 200 µL 10,000 U/mL penicillin/streptomycin (final 100 U/mL each). Store at −20 °C.
      3. Add 40 µL 32 mg/mL L-ascorbic acid (final 64 µg/mL). Prepare in DI water, filter sterilize, and store at −20 °C.
      4. Add 20 µL sodium selenite (final 14 ng/mL). Prepare and store at −20 °C.
      5. Add 240 µL NaHCO₃ (final 543 µg/mL). Mix thoroughly after thawing.
      6. Add 20 µL holo-transferrin (final 10.7 µg/mL). Mix after thawing.
      7. Add ~37.5 µL insulin (final 19.4 µg/mL; batch dependent). Store at 4 °C.
   2. Immediately before use, add Y27632 to a final concentration of 20 µM, CHIR99021 to 3 µM, IL-11 to 30 ng/mL, FGF4 to 25 ng/mL, heparin to 1 µg/mL, mLIF to 10 ng/mL, TGFβ to 15 ng/mL, and 8Br-cAMP to 25 µM.
4. Blastoid preparation, day 0 — Preparation and plating of TSCs
   1. Pre-warm TSF4H, FC (Table 1), and TX media (Section 2.3). Prepare gelatin-coated plates (≥45 min) for MEF depletion.
   2. Harvest TSCs by adding 500 µL cell dissociation reagent per well and incubating for 5 min in hypoxia. Add 2 mL FC medium and gently pipette. Confirm detachment. Transfer to a tube and centrifuge at 500 × g for 4 min.
      NOTE: TSCs cultured >4 days may require up to 10 min cell dissociation reagent treatment.
   3. Resuspend in 1 mL TSF4H medium and adjust to appropriate volume for plating at a 1:2 split.
   4. Plate onto gelatin-coated wells for MEF depletion. Incubate ~45 min in hypoxia. Ensure MEFs attach while TSCs remain mostly unattached.
   5. Replace 900 µL medium from Microwell aggregation plate wells with 500 µL TX medium. Return to hypoxia.
   6. Transfer TSC-containing supernatant to a tube without disturbing attached MEFs. Determine concentration (target 500,000–700,000 cells/mL).
   7. Calculate volume for 12 TSCs per microwell (14,400 cells per well). Centrifuge if further concentration is required.
      NOTE: Optimize for other TSC lines (10–15 cells per microwell recommended).
   8. Add 500 µL cell suspension per well dropwise. Centrifuge at 100 × g for 3 min. Confirm distribution and incubate in hypoxia for 24 h.
   9. Passage remaining TSCs as needed (Section 1.3).
5. Blastoid preparation, day 1 — Induction
   1. Around 24 h after TSC addition, add 20 µL of 5 mM 8Br-cAMP and 1 µL of 1 µg/mL DOX per well. Incubate for 48–72 h in hypoxia.
6. Blastoid preparation, day 3 — Fixation
   1. Resuspend aggregates using a cut 1000 µL pipette tip. Transfer to 2 mL tubes and allow sedimentation for 5 min.
   2. Remove supernatant and add 250 µL of 4% paraformaldehyde (PFA). Resuspend gently.
      CAUTION: Handle PFA in a fume hood with appropriate PPE.
   3. Incubate 15 min, flicking every 5 min.
   4. Remove PFA and add 300 µL PBST (PBS + 0.1% Tween-20). Mix gently. Store at 4 °C for <1 week before staining.
7. Immunofluorescence of blastoids
   1. Transfer 150 µL per sample to a 96-well U-bottom plate.
   2. Add 100 µL permeabilization buffer and incubate 20 min. Replace with primary antibody solution and incubate overnight at 4 °C.
   3. Wash 3× with PBST (5 min each). Add secondary antibodies.
   4. Incubate 2 h at room temperature in the dark.
   5. Wash 3× and transfer to imaging dish.
   6. Image using confocal microscopy (≥25× objective, NA ≥0.8).

3. High-throughput screens using the iGATA4-blastoid model

1. iGATA4-blastoid preparation and testing of environmental factors
   1. Prepare Microwell aggregation plate as described in Section 2.1 and generate iGATA4-blastoids following Sections 2.2–2.4. Prepare sufficient wells, including appropriate controls. Include a blank control (blastoids prepared using the standard protocol), a carrier control, and one or more experimental conditions.
   2. Incubate blastoids overnight in hypoxia. Approximately 24 h after TSC seeding, prepare TX medium supplemented with 8Br-cAMP, doxycycline (DOX), and the compound of interest (or carrier control). The following conditions have been tested:
      1. Caffeine: 3.5 mg/L, diluted from ultrapure water stocks.
      2. Nicotine: 330 ng/L, diluted from 100% ethanol (EtOH) stocks.
      3. Ethanol: 17 mM, diluted from absolute EtOH.
      4. High branched-chain amino acids (BCAA): +5%, +25%, +50% relative to baseline concentrations (0.45 mM L-leucine, 0.45 mM L-valine, 0.42 mM L-isoleucine), prepared from concentrated stocks in ultrapure water.
      5. Low methionine and leucine: −5%, −25%, −50%, −100% relative to baseline concentrations (0.12 mM L-methionine, 0.45 mM L-leucine), prepared using Met-/Leu-free DMEM/F12. Met-/Leu-free DMEM (LM-DMEM) is supplemented with 1x Ham's F-12 Nutrient Mix before use.
         CAUTION: Prepare and handle concentrated caffeine and nicotine stocks in a fume hood using appropriate personal protective equipment (PPE; gloves, lab coat, eye protection). Ethanol is flammable and must be kept away from heat sources.
   3. Aspirate the medium from each well and add 1 mL of compound-containing TX medium. Process one well at a time to avoid drying. Return the plate to the hypoxia incubator and culture until day 3.
   4. Capture representative brightfield images on day 3 using a transmitted light microscope. Fix the structures and perform immunofluorescence as described in Sections 2.6 and 2.7.
      NOTE: Adjust treatment windows depending on the experimental objective. Treatments can be applied during early aggregation (day 1) or later stages (e.g., cavity expansion). Example windows include day 1–3, day 1–2 (followed by media replacement), and day 2–3.
2. Multiparametric iGATA4-blastoid analysis
   1. Use brightfield images to quantify the percentage of wells containing cavitated structures (i.e., structures with a clearly defined central cavity) and measure blastoid diameter.
      1. Open images in the image analysis software.
      2. Use the straight-line tool to draw a line across each structure.
      3. Measure diameter using Analyze > Measure.
   2. Select representative immunostained samples (20–40 structures per biological replicate, with at least three biological replicates per condition) and image as described in Section 2.7.
   3. Quantify total cell number and lineage composition using immunofluorescence markers: TE (CDX2-positive), EPI (SOX2-positive and SOX17-negative), and PE (SOX17-positive).
   4. Assess structural organization and morphology by determining whether PE cells are spatially segregated from EPI cells within the inner cell mass (ICM) (binary classification: yes/no).
   5. Classify morphology as good (clear blastocyst-like structure with a defined cavity and properly organized TE, EPI, and PE compartments), mediocre (minor abnormalities such as irregular shape or partially formed cavity), or bad (major abnormalities such as absence of cavity or disorganized/unsorted lineages).
   6. Quantify additional parameters, including cavity size or volume, where applicable. Use bioinformatic tools or other tools for plotting and statistical analysis, and 3D image analysis software for cell counting and/or image preparation for display.

Results

Maintaining ESCs and iGATA4-ESCs (normoxia) and TSCs (hypoxia) under appropriate culture conditions is essential for successful blastoid generation. When cultured as described, ESCs and iGATA4-ESCs should be largely indistinguishable, forming mid-sized, round, and well-defined colonies after 3 days. The appearance of flattened colonies or protrusive edges indicates incipient differentiation, and such cultures should not be used for blastoid formation. TSCs should form flat epithelial colonies with slightly raised edges and flatter centers. Overgrown or merging TSC colonies (typically after >3 days in culture) often result in poor blastoid outcomes (Figure 1).

During iGATA4-blastoid assembly, several early hallmarks indicate protocol success (Figure 2A, B). Approximately 6 h after ESC and iGATA4-ESC aggregation, tight cell aggregates should be visible at the bottom of the microwells. Loose aggregates suggest excessive cell dissociation reagent exposure, which reduces cell–cell adhesion. Progression to the next step (TSC addition; Section 2.4) should only occur once tight aggregates are observed (Figure 2B, D0.5 timepoint). Immunostaining at this stage should show homogeneous SOX2⁺ and GATA4⁻ aggregates (Figure 2B, D0.5 timepoint).

Following TSC addition and overnight incubation, a single cohesive aggregate surrounded by CDX2⁺ cells should be evident (Figure 2B). The presence of large, unaggregated cells at the surface of aggregates often indicates residual MEFs. MEF contamination can reduce blastoid formation efficiency; therefore, adequate MEF depletion is critical to obtain a predominantly TSC population prior to aggregation. A 40 µm cell strainer may be used to remove residual MEFs, although this may reduce TSC yield. Incomplete or excessive CDX2⁺ coverage can be corrected by adjusting the number of TSCs added per microwell (recommended range: 10–15).

Following the addition of DOX and 8Br-cAMP, cavities are typically observed by day 2 and may persist until day 3. Under brightfield microscopy, cavities appear as clear, empty spaces within blastoids (Figure 2B, C; white dashed line). GATA4⁺ cells appear shortly after DOX induction and progressively sort to the surface of the ICM-like compartment (Figure 2B). Successful blastoid formation is characterized by high cavitation efficiency (typically ~80% of structures per Microwell aggregation plate; Figure 2D), a single-layered CDX2⁺ TE-like outer epithelium, a blastocoel-like cavity, and a rounded ICM-like compartment containing SOX2⁺ EPI-like cells overlaid by GATA4⁺ PE-like cells (Figure 2C–E).

Common suboptimal outcomes include non-cavitated structures, irregularly shaped blastoids with multiple cavities or ICM-like compartments, and wells containing few structures or cellular debris. While approximately 20% abnormal structures are expected (Figure 2C), higher rates may indicate issues such as incorrect TX medium preparation, poor starting cell quality, failure to add DOX or 8Br-cAMP on day 1, or displacement of structures during medium changes.

For high-throughput screening applications, untreated and vehicle control wells should yield blastoids comparable to positive-control conditions (Figure 3A,B). Experimental treatments may affect multiple quality metrics, including blastoid size (Figure 3C), cavitation efficiency, total and lineage-specific cell numbers (Figure 3D), PE sorting, and overall morphology (Figure 3E). For example, treatment with caffeine, nicotine, or ethanol during blastoid formation resulted in smaller structures with reduced cavitation efficiency, lower total or ICM cell numbers, impaired PE sorting, and poorer morphology. In contrast, increasing concentrations of branched-chain amino acids produced larger, more highly cavitated blastoids with increased cell numbers and improved morphology (Figure 3C–E). Biological variability between experiments is expected, highlighting the importance of including both blank and carrier controls in each experiment.

Figure 1: Examples of good- and poor-quality WT ESCs, iGATA4-ESCs, and TSCs. Brightfield micrographs of WT ESCs, iGATA4-ESCs, and TSCs showing expected colony morphologies after 2–3 days of culture in N2B27/2i/LIF (ESCs) or TSF4H medium (TSCs). The panel titled “WT ESCs – bad quality” shows cells with suboptimal morphology, including flattened colonies with protruding edges (red arrowheads). The panel “TSC overgrown” shows merged and enlarged TSC colonies after 4 days of culture. Scale bar: 100 µm for all panels. Please click here to view a larger version of this figure.

Figure 2: Schematic and expected morphology of iGATA4-blastoids. Reprinted from Developmental Cell, Vol. 61, Jorgensen V, Bao M, Junyent S et al., Efficient stem cell-derived mouse embryo models for environmental studies, pp. 193–207.e6, Copyright (2025)³, with permission from Elsevier. (A) Schematic representation of natural mouse blastocyst development (top) and stem cell–based blastocyst modeling using WT ESCs, iGATA4-ESCs, and TSCs.
(B) Micrographs showing expected morphology and marker expression at different stages of the protocol. Maximum intensity projection (MIP); dashed lines delineate the cavity. Scale bar: 20 µm. (C) Brightfield image of iGATA4-blastoids at day 3. Scale bar: 100 µm.
(D) Cavitation rates of iGATA4-blastoids generated using TSCs pre-cultured in normoxia or hypoxia; aggregates were cultured under hypoxic conditions. n ≥ 4; error bars represent the standard error of the mean. (E) Immunostaining of E4.5 blastocysts and day 3 iGATA4-blastoids for lineage markers (CDX2, cyan; OCT4, red; GATA4, green). Maximum intensity projection (MIP); single Z-plane shown where indicated. Scale bar: 25 µm. Please click here to view a larger version of this figure.

Figure 3. Effects of environmental factors on blastoid formation and quality. (A) Representative immunofluorescence 3D images (top) and IMARIS surface reconstructions (bottom) of iGATA4-blastoids treated with environmental factors (caffeine, nicotine, ethanol, and conditions mimicking high- and low-protein diets). Scale bars: 25 µm for all images. (B) Schematic of the experimental workflow. (C) Blastoid diameter under different treatment conditions. The range represents the 10th–90th percentiles; boxes indicate the interquartile range; the line indicates the median. Statistical analysis was performed using one-way ANOVA with multiple comparisons. (D) Quantification of total (gray), trophectoderm (cyan), epiblast (red), and primitive endoderm (green) cell numbers in control and treated blastoids. Bars represent the mean; error bars represent the standard deviation. Statistical analysis was performed using one-way ANOVA with multiple comparisons. (E) Quantification of the proportion of blastoids classified as “good” (proper lineage sorting with minimal morphological abnormalities), “mediocre” (largely correct morphology with minor abnormalities, including incomplete EPI–PE sorting), or “bad” (severely abnormal morphology with disrupted lineage organization) under control and treatment conditions. Statistical analysis was performed using two-way ANOVA relative to control conditions. For all panels, n ≥ 3. Statistical significance is indicated as *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Please click here to view a larger version of this figure.

Media	Reagent	Final concentration
FC
	DMEM	n.a.
	Heat-inactivated FBS (iFBS)	15%
	GlutaMax	2 mM
	2-Mercaptoethanol (2-ME)	0.1 mM
	Non-essential amino acids (NEAA)	1x
	Na Pyruvate	1x
	Penicillin–Streptomycin	1x
N2B27/2i/LIF
	DMEM/F12	50%
	NeurobasalA	50%
	B27 supplement	1x
	N2 supplement	0.5x
	2-ME	100 µM
	Penicillin–Streptomycin	1%
	GlutaMax	2 mM
	PD0325901	1 µM
	CHIR99021	3 µM
	Mouse recombinant LIF	10 ng/mL
TSF4H
	Megacell RPMI-1640 Medium	n.a.
	iFBS	20%
	GlutaMax	2 mM
	Na Pyruvate	1x
	Penicillin–Streptomycin	1x
	FGF4	25 ng/mL
	Heparin	1 μg/mL

Table 1: Media recipes for FC, N2B27/2i/LIF, and TSF4H. Recipes to prepare FC media, N2B27/2i/LIF media, and TSF4H media, with final concentrations.

Component	Final concentration	Stock concentration	Solvent and storage	Volume to ~20 mL
8Br-cAMP	25 µM	5 mM	PBS, store at -20 °C	100 µL
CHIR99021	3 µM	10 mM	DMSO, store at -20 °C	6 µL
DMEM/F12	N/A	N/A	Store at 4 °C	20 mL
FGF4	25 ng/mL	100 µg/mL	0.1% BSA in PBS, store at -20 °C	5 µL
L-alanyl-L-glutamine Supplement	2 mM	200 mM	Store at 4 °C	200 µL
Heparin	1 µg/mL	1 mg/mL	PBS, store at -20 °C	20 µL
Holo-transferrin	10.7 µg/mL	10.7 mg/mL	DI water, store at -20 °C	20 µL
Human TGF-β1	15 ng/mL	20 µg/mL	4 mM HCl + 0.1% BSA, store at -20 °C	15 µL
IL-11	30 ng/mL	25 µg/mL	0.1% BSA in PBS, store at -20 °C	24 µL
Insulin	19.4 µg/mL	9.5 - 11.5 mg/mL	Concentration is batch dependent, store at 4 °C	33.7–40.8 µL, adjust volume according to stock concentration.
L-ascorbic acid-2-phosphate magnesium	64 µg/mL	32 mg/mL	DI water, store at -20 °C	40 µL
Mouse recombinant LIF	10 ng/mL	10 µg/mL	PBS + 0.1% BSA, store at -20 °C	20 µL
NaHCO3	543 µg/mL	45.25 mg/mL	DI water, store at -20 °C	240 µL
Penicillin/Streptomycin	1x	100x	Store at -20 °C	200 µL
Sodium selenite	14 ng/mL	14 µg/mL	DI water, store at -20 °C	20 µL
Y27632	20 µM	20 mM	DMSO, store at -20 °C	20 µL

Table 2: Media recipe for TX. Recipe to prepare TX media, with details on how to prepare and store small molecules and other reagents, and final concentrations.

Discussion

iGATA4-blastoids provide a highly efficient, embryo-like, stem cell–based model of the mouse blastocyst by self-organizing three in vitro–cultured cell types: ESCs, inducible GATA4-ESCs (iGATA4-ESCs), and hypoxia-cultured TSCs. These structures recapitulate key features of the E4.5 blastocyst, including appropriate lineage proportions and transcriptional identity, and can be used for multiparametric environmental screening, reliably reflecting blastocyst-stage developmental defects³. A recent study further demonstrates that aggregation of ESCs, iGATA4-ESCs, and TSCs in basal blastoid media also yields high-efficiency blastoid formation, highlighting the robustness and adaptability of this approach across laboratories and culture conditions³⁹. iGATA4-ESCs have previously been shown to recapitulate the functions of VE cells in post-implantation stem cell–based embryo models^28,29,30,31. Together, these results illustrate how iGATA4-ESCs can serve as flexible and reliable stand-ins for PE/VE cells in stem cell–based embryo modeling across developmental stages.

Several steps in the protocol are critical for achieving high-quality blastoids. These include the quality of the starting cell populations, particularly hypoxia-cultured TSCs (Figure 1); the preparation of TX medium, which should be freshly prepared on day 0 (Table 2; Section 2.3); and careful handling of plates following aggregation. Key readouts of protocol success include cavitation efficiency (typically ~80%), appropriate morphology (~100 µm diameter structures with a blastocoel-like cavity, a CDX2⁺ TE-like outer layer, and a sorted SOX2⁺ and GATA4⁺ inner cell mass with an approximate 3 TE:1 EPI:1 PE ratio), and the presence of a properly sorted GATA4⁺ PE-like layer. These are metrics commonly used in the field^1,2,4,5. Among cavitated blastoids, approximately 75% of structures display all of these features. These metrics can be used as initial indicators for troubleshooting. Suboptimal cell quality is typically reflected in abnormal colony morphology, resulting in poorly aggregated or non-cavitated structures. Improper medium preparation may lead to incomplete aggregation or lineage defects, while rough handling can result in multiple aggregates per microwell or empty wells.

Further optimization may include adjusting the number of seeded cells (recommended starting ranges: 4–7 ESCs and iGATA4-ESCs, and 10–15 TSCs per microwell), refining MEF depletion methods (e.g., filtration through a 40 µm cell strainer), and modifying the TX medium composition. For example, removal of FGF4 and heparin can still support efficient blastoid formation while promoting a more mural TE-like identity. iGATA4-blastoids can be maintained in culture for up to 6 days; however, structures from days 4–6 often increase in size while retaining blastocyst-like proportions of the lineage. The protocol yields consistent results across biological replicates, but appropriate controls should be included in each experiment.

Compared with other mouse blastoid systems, which vary in starting cell populations and culture conditions^{6,7,8,9,10,11,12,13,14,15,16,17,18,19,20}, iGATA4-blastoids achieve both high formation efficiency and accurate lineage representation. While many existing protocols generate cavitated structures and are valuable for specific biological applications, they often exhibit incomplete PE formation, heterogeneous or undefined cell identities, or incorrect lineage proportions. By combining three lineage-directed stem cell populations in a stepwise manner, iGATA4-blastoids overcome these limitations and provide a reproducible, modular, and experimentally accessible system.

Despite these advantages, several limitations remain. Extended in vitro culture in TX medium beyond 4–6 days results in overgrowth, even if lineage proportions remain blastocyst-like. Approximately 12.5% of iGATA4-blastoids transition to peri-implantation–like morphologies in vitro, characterized by a cup-shaped, pseudostratified epiblast surrounding a pro-amniotic cavity, an expanded extraembryonic ectoderm-like compartment adjacent to the proximal epiblast, and an outer visceral endoderm–like layer. However, iGATA4-blastoids fail to implant and develop further following transfer into a host mouse. Therefore, iGATA4-blastoids, and blastoids more broadly, are not suitable for studying implantation or embryo–maternal interactions.

The molecular basis of blastoids' failure to implant remains unclear, although current evidence suggests that incomplete trophectoderm modeling may be a contributing factor. TE-like cells in iGATA4-blastoids exhibit an overly polar molecular identity. Future efforts to engineer TSCs that more closely resemble mural TE cells, or to introduce alternative cell lines that better recapitulate implantation-competent TE, represent promising directions to overcome these limitations. The modular nature of iGATA4-blastoids, in which TE-specific modifications can be implemented independently of EPI and PE formation, makes this system particularly well-suited for such developments.

Taken together, the iGATA4-blastoid protocol provides a robust and scalable platform for studying lineage specification and early developmental processes in the mammalian embryo. Its reproducibility and adaptability make it well-suited for high-throughput environmental, pharmacological, and genetic screening applications. As such, iGATA4-blastoids represent a powerful system for investigating early developmental mechanisms and assessing how diverse perturbations influence pre-implantation development.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
2-Mercaptoethanol	Thermo Fisher Scientific	31350010	Used for general media preparation
8-Bromo-cAMP	STEMCELL Technologies	73602	Used for TX media preparation and blastoid cavitation
AggreWell400	STEMCELL Technologies	34415	Microwell aggregation plate; Used for blastoid preparation
Anti-Adherence Rinsing Solution	STEMCELL Technologies	7010	Used for blastoid preparation
ART Wide Bore Filtered Pipette Tips	Thermo Fisher Scientific	2079GPK	Wide bore pipettes for blastoid movement during immunostaining.
B27 Supplement	Thermo Fisher Scientific	17504044	Used for media preparation
Caffeine	Sigma	58-08-2	Used as environemental factor to test in iGATA4-blastoids
CDX2 antibody, mouse	Biogenex	MU392A5UC	For immunofluorescence, used at 1:200
CellXpert (normoxia incubator)	Eppendorf	6731010015	Other brands/models may work too.
CHIR99021	STEMCELL Technologies	72054	Used for media preparation
Countess Cell Counting Chamber slides	Invitrogen	C10228	Used for general tissue culture purposes
DAPI	Thermo Fisher Scientific	D1306	Used for immunofluorescence, used at 1:500-1:1,000
DMEM	Gibco	11995040	Used for FC media preparation
DMEM/F12	Thermo Fisher Scientific	21331-020	Used for N2B27 and TX media preparation
DMEM-LM	Thermo Fisher Scientific	30030	Supplement DMEM-LM with Ham's F-12 Nutrient Mix powder to a final 1x concentration to obtain Met- and Leu-free DMEM/F12
Donkey anti-Goat IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 488	Invitrogen	A-11055	Used for immunofluorescence, used at 1:500-1:1,000
Donkey anti-Mouse IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 647	Invitrogen	A-31571	Used for immunofluorescence, used at 1:500-1:1,000
Donkey anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 568	Invitrogen	A10042	Used for immunofluorescence, used at 1:500-1:1,000
Donkey anti-Rat IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 647	Invitrogen	A78947	Used for immunofluorescence, used at 1:500-1:1,000
Donkey Serum	Jackson ImmunoResearch	017-000-121	Used for imunofluorescence. Thaw at 4°C;, and store aliquoted at -20°C;.
Doxycycline	Sigma	D9891-5G	For iGATA4 induction during blastoid preparation
Eppendorf Centrifuge 5702	Eppendorf	5702	used in our lab, others will work. Requires tube and plate adaptors.
Falcon 96-well Clear Round Bottom Not Treated Microplate, with Lid, Individually Wrapped	Corning	351177	Used for immunofluorescence
FGF4	R&D Systems	5846-F4	Used for media preparation
Fiji	Schindelin et al., 2012	https://imagej.net/Fiji	Used for image analysis
GATA4 antibody, rabbit	Cell Signalling	36966S	Used for immunofluorescence, used at 1:500
Gelatin solution, 2%	Sigma	G1393	Used for general tissue culture purposes
Glass Bottom Dish 35 mm	Ibidi	81218-200	For immunofluorescence and imaging
GlutaMAX	Thermo Fisher Scientific	35050-061	L-alanyl-L-glutamine supplement; Used for media preparation
Ham's F-12 Nutrient Mix powder	Thermo Fisher Scientific	21700075	Used as supplement for DMEM-LM, for low-protein diet conditions.
Heparin	Sigma	H3149	Used for media preparation
Heracell VIOS 160i (hypoxia incubator)	Thermo Fisher Scientific	51030400	Other brands/models may work too.
Holo-transferrin	Sigma	T4132	Used for TX media preparation.
Human recombinant TGF-ß1	PeproTech	100-21	Used for TX media preparation.
IL-11	PeproTech	200-11	Used for TX media preparation.
Imaris	Oxford Instruments	https://imaris.oxinst.com/	Used for image analysis and image preparation.
Inactivated Fetal Bovine Serum	Corning	35-010-CV	Used for media preparation
Inducible GATA4 ESCs (tetO-GATA4-ESCs)	Amadei et al., 2021	NA	Cell line available from the Zernicka-Goetz lab upon request
Insulin	Sigma	I9278	Used for TX media preparation.
Irradiated C57bl/6 Mouse Embryonic Fibroblasts	Thermo Fisher Scientific	A34960	Used as feeder layer for TSC culture. Thaw as indicated in the protocol and use within a week.
L-ascorbic acid-2-phosphate magnesium	Sigma	A8960	Used for TX media preparation.
L-isoleucine	Thermo Fisher Scientific	73-32-5	Used to supplement TX media.
L-leucine	Thermo Fisher Scientific	61-90-5	Used to supplement TX media.
L-valine	Thermo Fisher Scientific	72-18-4	Used to supplement TX media.
Megacell RPMI-1640 Medium	Sigma	M3817	Used for media preparation
Miscellaneous pipette tips	Multiple providers	NA	For general liquit handling
Mouse recombinant LIF	STEMCELL Technologies	78056	Used for media preparation
N2 Supplement	Thermo Fisher Scientific	17502048	Used for media preparation
NaHCO3	Sigma	S5761	Used for TX media preparation.
Neurobasal A	Thermo Fisher Scientific	10888-022	Used for media preparation
Nicotine	Sigma	54-11-5	Used as environemental factor to test in iGATA4-blastoids
Non-essential amino acids	Thermo Fisher Scientific	11140-050	Used for general media preparation
Parafilm	Thermo Fisher Scientific	13-374-10	For immunofluorescence
Paraformaldehyde, 16%	Electron Microscopy Sciences	15710	Used for immunofluorescence, prepare 4% working concentration in PBS before use.
PBS	Thermo Fisher Scientific	10010031	Used for general tissue culture purposes and immunofluorescence
PD0325901	STEMCELL Technologies	72184	Used for media preparation
Penicillin–Streptomycin	Thermo Fisher Scientific	15140122	Used for media preparation
Pluriselect pluriStrainer (Cell Strainer) 40 um mesh-size PET	Thermo Fisher Scientific	NC1401489	Optional, for MEF depletion
Prism10	GraphPad	https://www.graphpad.com/scientific-software/prism/	Used for data analysis and plotting.
ROCK inhibitor Y27632	STEMCELL Technologies	72304	Used for media preparation and blastoid formation
Sodium Pyruvate	Thermo Fisher Scientific	11360070	Used for media preparation
Sodium selenite	Sigma	S5261	Used for TX media preparation.
SOX17, goat	R&D Systems	AF1924	Used for immunofluorescence, used at 1:500-1:1,000
SOX2, rat	Invitrogen	14981180	Used for immunofluorescence, used at 1:500-1:1,000
Triton X-100	Sigma	T8787	Used for immunofluorescence
TrypLE Select enzyme	Thermo Fisher Scientific	12563011	Cell dissociation reagent; Used for general tissue culture purposes
TSC-eGFP	Tanaka et al., 1998	NA	Cell line available from the Zernicka-Goetz lab upon request
Tween-20	Sigma	P1379	Used for immunofluorescence
Ultrapure water for tissue culture	Thermo Fisher Scientific	10977023	Used for general tissue culture purposes
WT CD1 ESCs	Amadei et al., 2021	NA	Cell line available from the Zernicka-Goetz lab upon request
WT TSCs	Sozen et al., 2019	NA	Cell line available from the Zernicka-Goetz lab upon request
Zeiss 980 confocal microscope	ZEISS	NA	Microscope, other makes and models may work too.