Robust self-organization of livestock pluripotent stem cells into post-gastrulation embryo models with advanced neuronal and mesodermal structures

This study establishes robust stem cell-based embryo models for sheep and pigs that successfully recapitulate post-gastrulation development, including the formation of complex neuronal and mesodermal structures like somites and neural tubes, thereby filling a critical gap in ungulate developmental biology and enabling comparative research across mammalian species.

The Gist

Imagine you are trying to understand how a complex city is built. You can't just look at the finished skyscrapers; you need to see the blueprints, the construction crews, and how the roads, power plants, and houses organize themselves from a single pile of raw materials.

For a long time, scientists could only study this "construction phase" of life (embryonic development) using mice and humans. But mice and humans are like two very different types of architects. They build their cities in unique ways. What about the "livestock" architects—sheep and pigs? Their construction plans are different, but until now, we couldn't watch them build in a lab because their embryos develop inside the mother in a way that's very hard to study.

This paper is like a breakthrough in construction simulation software. The researchers figured out how to build "miniature, self-assembling embryo models" using sheep and pig stem cells. Here is how they did it, explained simply:

1. The Raw Materials: The "Lego Bricks"

The scientists started with pluripotent stem cells. Think of these as magical Lego bricks that haven't decided what they want to be yet. They could become a brain cell, a muscle cell, or a kidney cell. The researchers took these bricks from sheep and pigs and put them in a special bowl.

2. The Spark: Waking Up the Builders

Just throwing the bricks in a bowl isn't enough; they need instructions. The researchers gave the cells a chemical "wake-up call."

• The Gastruloid Phase (The Rough Draft): First, they let the cells clump together. Within 24 hours, the clump stopped being a round ball and started stretching out, breaking its symmetry. It was like a blob of clay suddenly deciding, "Okay, I have a head and a tail!"
• The Result: They created "Gastruloids." These are tiny, 3D models that mimic the very early stages of an embryo, showing how the body plan starts to form. Interestingly, the sheep and pig models acted slightly differently, showing that even at this early stage, different species have their own unique "personality."

3. The Big Leap: Building the "Trunk"

The real magic happened when they added a secret ingredient: ECM (Extracellular Matrix).

• The Analogy: Imagine the cells are trying to build a house on a slippery, flat table. They keep sliding apart. The ECM is like pouring concrete or mud around the foundation. It gives the structure something to hold onto.
• The Result: With this "concrete" added, the sheep cells didn't just stop at a rough draft. They built Ovine Trunk-Like Structures (oTLSs). These are sophisticated mini-models that look like the middle section (the trunk) of a developing sheep embryo.

4. What Did They Build?

These mini-embryos were incredibly impressive. They didn't just grow randomly; they organized themselves with military precision:

• The Spine: They built a central tube that looks exactly like a spinal cord (the neural tube).
• The Muscles: Flanking the spine, they built segmented blocks that look like somites (the precursors to ribs and back muscles), stacking up like a row of bricks.
• The Brain & Nerves: They didn't just make the back; they made the front too! They generated neural crest cells (which become skin, nerves, and facial features) and even early brain regions.
• The Kidneys: Perhaps most surprisingly, they built kidney precursors. This is a huge deal because previous models mostly only built the back of the body. These models built the "middle" organs too.

5. Why Is This a Big Deal?

Think of it like this: Before this, we only had a manual for building a house in New York (Human) and one for a house in Tokyo (Mouse). We didn't have the manual for a house in a rural farm setting (Sheep/Pig).

• Comparative Biology: Now we can compare how different mammals build their bodies. Why does a pig's early development differ from a human's? This model lets us watch the differences happen in real-time.
• Safety Testing: If a new medicine or pesticide is toxic to a developing sheep embryo, we can test it on these mini-models in a petri dish instead of hurting real animals.
• Agriculture: This could help us understand how to improve livestock health or even help with breeding programs.

The Bottom Line

The researchers took sheep and pig stem cells, gave them a little chemical nudge and a bit of "mud" (ECM), and watched them self-organize into complex, living structures that look like the trunk of a developing animal.

It's like watching a pile of sand spontaneously turn into a working sandcastle with a moat, a tower, and a drawbridge, all without a human hand touching it after the initial setup. This proves that the "instructions" to build a complex body are deeply embedded in the cells themselves, and we can finally watch livestock follow those instructions in a lab.

Technical Summary

1. Problem Statement

• Limitations of Current Models: Stem cell-based embryo models (SEMs), such as gastruloids and trunk-like structures (TLS), have been successfully developed for mice and humans. However, they are currently restricted to rodent and primate species.
• Biological Gaps: Ungulate species (e.g., sheep and pigs) exhibit distinct developmental trajectories compared to rodents and primates, including prolonged peri-implantation stages and the onset of somitogenesis prior to implantation.
• Technical Challenges: Direct study of ungulate embryos is difficult due to their in utero development. Furthermore, existing SEMs largely recapitulate posterior embryonic structures and lack robust generation of anterior neural tissues, renal primordia, and complex dorso-ventral patterning found in post-gastrulation development.
• Goal: To establish the first in vitro models for post-gastrulation development in ungulates to enable comparative mammalian developmental studies, toxicology, and agricultural biotechnology.

2. Methodology

The researchers utilized recently established embryonic disc-like pluripotent stem cells (EDSCs) from sheep (ovine) and pigs (porcine).

• Cell Culture & Differentiation:
  • Cells were maintained in a "primed" pluripotent state using AFX medium (Activin A, FGF2, and the WNT inhibitor XAV939).
  • To induce gastrulation, XAV939 was withdrawn, followed by transient activation of the WNT pathway using the agonist CHIR99021.
  • Cells were aggregated in ultra-low-attachment U-bottom plates to form spheroids.
• Gastruloid Generation:
  • Protocols were adapted from mouse/human systems, adjusting for species-specific developmental tempos.
  • Aggregates were cultured for up to 72 hours to observe symmetry breaking and axial elongation.
• Trunk-Like Structure (TLS) Optimization (Ovine only):
  • Building on ovine gastruloids, the team introduced Extracellular Matrix (ECM) (Geltrex) at 24 hours post-aggregation (immediately after symmetry breaking) to support higher-order tissue organization.
  • Parameters optimized included CHIR concentration (6 µM pulse), timing of ECM addition, and aggregate size (4,000 cells per aggregate was optimal).
  • Retinoic acid was excluded as it caused aberrant morphologies.
• Characterization Techniques:
  • Immunofluorescence: Whole-mount staining for markers including Brachyury (Bry), Sox2, Sox1, Pax3, Foxa2, Cdx2, Gata6, and renal/neural markers (WT1, Pax2, NCAM1, TUJ1).
  • Single-Cell RNA Sequencing (scRNA-seq): Time-resolved sequencing (Days 2–8) to map lineage trajectories, cell type diversity, and HOX gene dynamics.
  • Cross-Species Annotation: Data was mapped against human and mouse embryonic atlases using SingleR to validate cell identities.
  • Morphometric Analysis: Quantification of elongation, somite counts, and structural integrity.

3. Key Contributions

• First Ungulate SEMs: Successfully generated the first in vitro embryo models for sheep and pigs, extending SEM technology beyond rodents and primates.
• Robust Ovine TLS (oTLS): Developed a highly reproducible protocol for ovine trunk-like structures that sustain development for 8–10 days, significantly longer than mouse or human TLS models.
• Expanded Lineage Repertoire: Demonstrated that ungulate PSCs can self-organize into complex structures containing dorsal neural derivatives (neural crest, roof plate), anterior neuronal populations (midbrain-hindbrain boundary), and renal primordia (kidney progenitors), which are often missing in existing trunk models.
• Species-Specific Divergence: Revealed fundamental differences in the self-organization mechanisms between ovine and porcine cells, specifically regarding Neuromesodermal Progenitor (NMP) formation and axial mesoderm patterning.

4. Key Results

A. Gastruloid Formation (Sheep vs. Pig)

• Morphology: Both species exhibited symmetry breaking within 24 hours and axial elongation over 72 hours.
• Molecular Differences:
  • Ovine: Formed a clear Neuromesodermal Progenitor (NMP) population (Bry+/Sox2+), leading to a neural tube-like axis and paraxial mesoderm. Pax3 was expressed in both neural and mesodermal lineages (dorsal bias).
  • Porcine: Lacked a canonical NMP population. Instead, they formed an axial mesoderm-like cluster (Bry+/Foxa2+) with internal distribution. Pax3 was confined to mesoderm, and Foxa2 was prominent, suggesting a ventral-skewed profile.
• Outcome: While both formed gastruloids, only the ovine system successfully progressed to robust TLSs under the tested conditions.

B. Ovine Trunk-Like Structures (oTLS)

• Structural Integrity: oTLSs developed a central neural tube-like structure flanked by segmented somite-like domains. They maintained structural integrity for up to 8 days without collapse.
• Robustness: The system tolerated a wide range of initial aggregate sizes (400–8,000 cells), with 4,000 cells being optimal. Over 90% of aggregates elongated along a single axis, and >80% formed clear neural tubes and somites.
• Lineage Diversification (scRNA-seq):
  • Identified 14 distinct cell types.
  • Neural: Generated roof plate cells, neural crest, glial precursors, midbrain-hindbrain boundary (MHB) progenitors, and postmitotic interneurons.
  • Mesodermal: Progressed from presomitic mesoderm (Msgn1, Tbx6) to somitic and myogenic progenitors (Pax7, Myf5).
  • Renal: Generated intermediate mesoderm derivatives, including nephron progenitors, podocytes, and collecting duct-like epithelium. Spatially organized as bilateral WT1+ domains.
  • Endothelial: Presence of endothelial cell populations.
• Dynamics:
  • HOX Patterning: Showed temporal collinearity of HOX genes, consistent with anteroposterior patterning.
  • Metabolism: Transitioned from a bivalent metabolic state (glycolysis + OxPhos) to a glycolytic-dominant state, mimicking in vivo post-implantation shifts.
  • Segmentation Clock: Exhibited rhythmic somite formation with a periodicity of ~6.6 hours per somite pair.

C. Comparison with In Vivo and Other Models

• The oTLSs recapitulated the mediolateral positioning of the kidney field and the epithelialization of nephric structures seen in human fetal kidneys.
• Unlike mouse TLSs, oTLSs did not require extraembryonic tissues to initiate complex trunk organization, highlighting the intrinsic self-organizing capacity of ungulate PSCs.

5. Significance

• Comparative Developmental Biology: Provides a tractable platform to study the unique developmental tempos and geometries of large mammals (ungulates), filling a critical gap between rodent/primate models and farm animals.
• Biomedical Applications: The ability to model renal primordia and anterior neural structures in a livestock context offers new avenues for veterinary toxicology and understanding congenital defects in agriculture.
• Fundamental Insights: The study reveals that the extracellular matrix (ECM) acts as an active regulatory cue for morphogenesis in ungulates, enabling the transition from simple aggregates to complex, multi-lineage trunk architectures.
• Future Directions: Establishes a foundation for refining livestock models to include endodermal derivatives and primordial germ cells, potentially revolutionizing agricultural biotechnology and reproductive medicine.

Limitations Noted: The study used a single cell line per species, relied on cross-species annotation (mouse/human atlases) due to the lack of a sheep reference atlas, and was conducted under hypoxic conditions which may influence specific lineage outcomes.