Caltech researchers have developed a powerful new method for creating embryo-like structures from stem cells that could transform how we study fertility.
Using stem cells rather than a traditional fertilized egg, the team has built mouse embryo models called iG4-blastoids that closely mimic natural blastocysts, the stage of development when an embryo implants into the uterus. This implantation stage is when the majority of pregnancies fail, including those resulting from in vitro fertilization, and little is known about why. The new models will enable researchers to study the effects of environmental factors, such as exposure to caffeine, nicotine, alcohol, and diet.
The research was led by the laboratory of Magdalena Zernicka-Goetz , Bren Professor of Biology and Biological Engineering at Caltech, and is described in a paper appearing in the journal Developmental Cell.
"This model could revolutionize fertility research-helping us understand why some pregnancies fail and how to support the ones that can succeed," Zernicka-Goetz says. "These live models allow us to watch, in a dish, how early embryos organize themselves, and to test how common environmental exposures-like caffeine, alcohol, nicotine, and even high or low protein diets-affect that process."
The team found that iG4-blastoids responded to toxins and nutrient changes in the same way as natural embryos. For example, caffeine and nicotine in the earliest stages of pregnancy reduced cell numbers and impaired development, while altering amino acid availability mimicked the impact of high- or low-protein diets on embryo growth.
The team built the iG4-blastoids from three types of stem cells, including one engineered to express a key developmental gene (GATA4). Expressing GATA4 was a key breakthrough where other attempts at creating stem-cell-derived blastoids had failed. Importantly, the method results in properly developed blastoids 80 percent of the time. This high efficiency enables the researchers to produce thousands of the models and conduct robust experiments; for example, screening for impacts of caffeine on specific days of development, or the effects of different concentrations at different times.
Although this work was performed in mice, the models will profoundly enhance studies of the basic biological processes underlying reproduction, such as the factors that prevent otherwise genetically healthy embryos from implanting-a critical milestone of development-and the mechanisms underlying why certain environmental factors may inhibit progression.
"This method is a game-changer," says Sergi Junyent, a Caltech postdoctoral scholar and co-first author of the new paper. "When applied to humans, it could help us understand why some embryos thrive while others fail, and how to optimize conditions for conception-whether natural or assisted."
While iG4-blastoids cannot yet progress far beyond implantation stages, the researchers emphasize that their high fidelity to natural embryos makes them an unprecedented model system. Ongoing work aims to refine the models further and explore their translation to human fertility research in a safe, ethical, and regulated framework.
The paper is titled "Efficient stem cell-derived mouse embryo models for environmental studies." Former Caltech graduate student Victoria Jorgensen (PhD '23), former Caltech postdoctoral scholar Min Bao (now of First Affiliated Hospital of Wenzhou Medical University in Wenzhou, China), and Junyent are co-first authors. In addition to Zernicka-Goetz, additional Caltech co-authors are former SURF student Christoph M. Häfelfinger; postdoctoral scholars Laura Amaya, Zhaodi Liao, and Dong-Yuan Chen; Brian A. Williams, director of the Transcriptome Function and Technology Program; graduate student Amanda Wu; and Matt Thomson , professor of computational biology and Heritage Medical Research Institute Investigator. Funding was provided by the National Institutes of Health, Open Philanthropy Award, the Human Frontiers Science Program, the National Natural Science Foundation of China, the Ministry of Science and Technology of China, the Swiss Study Foundation, the Werenfels Fund of the Free Academic Society Basel, and the Claudine and Hans-Heiner Zaeslin-Bustany Foundation. Zernicka-Goetz and Thomson are affiliated faculty members with the Tianqiao and Chrissy Chen Institute for Neuroscience at Caltech .
