Lab-grown structures mimic human embryo's earliest stage yet

Immunofluorescence co-staining images of ZO1 (Zonula Occludent-1, green) and phalloidin (red) in a human blastoid.

Multiple teams of researchers have created artificial blastocysts like this one from human stem cells.Credit: UT Southwestern

Scientists have used human stem cells to mimic the earliest stage yet of embryo growth.

Multiple research groups independently report that they have grown balls of cells that look like human blastocysts, which form about 4 days after an egg is fertilized by sperm. Two teams published their results in Nature on 17 March1,2; last week, two other groups reported similar results on the bioRxiv preprint server3,4 that have not been peer reviewed. These experiments offer a window into a crucial time in human development, and an opportunity to better understand pregnancy loss and infertility without experimenting on human embryos.

“This is an important milestone,” says Jianping Fu, a bioengineer at the University of Michigan in Ann Arbor.

Some of what scientists currently understand about this early stage of development has come from studies on human embryos. But access to these embryos is limited and tightly regulated in light of ethical considerations. Because blastocysts grown in the laboratory from human stem cells differ from human embryos, they might avoid some of the ethical limits on human embryo research and could increase access to this type of work, scientists say. They do not expect the new blastocyst-like structures to have the ability to develop into a complete embryo.

Researchers have previously grown blastocysts in the lab from mouse stem cells5, but mice have different developmental pathways from humans, so the resulting structures weren’t a perfect model of human development.

The latest studies are “really putting all the pieces together in a way that you can potentially model how the embryo actually develops in its earliest stages”, says Janet Rossant, a developmental biologist at the Hospital for Sick Children in Toronto, Canada. “This is a period we don’t understand very much about.”


During a pregnancy, a blastocyst would implant in the wall of the uterus at around 7 or 8 days. At this point, it has an outer layer of cells that would give rise to the placenta and a clump of cells within that have the potential to develop into a fetus.

Scientists have previously used human embryonic stem cells to look at later stages of embryo development, around 18–20 days, after the blastocyst has implanted6. But the new experiments are the earliest stage of development to be modelled in a lab.

Human blastocyst in uterus, light micrograph.

A blastocyst will implant in the wall of the uterus around day 7 or 8 of human development.Credit: Lennart Nilsson, TT/SPL

Some knowledge of this phase comes from research groups growing human embryos in the lab for up to 13 days. Laws in about a dozen countries, as well as guidelines issued by the International Society for Stem Cell Research (ISSCR), limit embryo development in the lab to 14 days after fertilization. By this time, implantation is complete, and what’s known as a primitive streak appears in the embryo, marking a point at which the cells within are becoming more differentiated and complex.

In one of the two Nature studies1, a team of scientists at the University of Texas Southwestern Medical Center in Dallas and Kunming Medical University in China treated human stem cells with a succession of growth factors to form artificial blastocysts, called ‘human blastoids’. The researchers showed they could do this using stem cells derived from human embryos, or using adult skin cells that had been reprogrammed into stem cells.

In the other peer-reviewed study2, Jose Polo at Monash University in Clayton, Australia, and his colleagues reprogrammed adult skin cells to generate a mixture of cells, some of which grew into human blastoids.

“Is [a human blastoid] exactly equivalent to a human embryo?” asks Aryeh Warmflash, a stem-cell biologist at Rice University in Houston, Texas. “Almost certainly not. Is it a pretty good model for the blastocyst-stage embryo? I think it probably is.”

Both teams showed that their artificial structures were built like blastocysts, with a cavity in the centre and a mass of cells — what could continue developing into fetal tissues in real blastocysts — in one corner. They also showed that the structures contained three signature cell types that make up a blastocyst. And they coaxed their human blastoids to ‘implant’ onto plastic sheets and mature into a state similar to a human blastocyst after it implants into the uterine wall.

The protocols are familiar to researchers such as Fu, who has seen similar methods used to develop mouse blastoids. “Nonetheless, this is a very important next step,” he says.

The two teams that posted preprints showed similar results while working with extended pluripotent stem cells.

“Looking to the future, we want to use this model to gain more insight into early human development, and to understand different gene functions, as well as their mutations,” says Jun Wu, a molecular biologist at Southwestern, who led one of the Nature studies1.

A workaround

Still, the teams acknowledge that their methods can be improved. Both Nature studies reported that only about 10% of the reprogrammed or transformed cells developed into human blastoids. Also, both teams acknowledged that there were some cells in the structures that are not typically found in human blastocysts.

“It’s a good start,” Rossant says of the studies. But on the basis of these factors, “you would predict that it’s not going to be incredibly reproducible”.

Neither of the teams that published in Nature allowed their structures to grow beyond about the equivalent of a 2-week-old embryo, mindful of the 14-day rule’s limit on growing human embryos in the laboratory.

Still, some developmental biologists think that these artificial structures differ from human embryos in a key way. Scientists do not expect the structures to be viable beyond this stage of development, on the basis of evidence7 showing that mouse blastoids do not develop into embryos when implanted in a mouse uterus.

But their similarity to human blastocysts still raises ethical questions. The ISSCR is already aware of this, and is due to release revised guidelines for work with embryo-like structures in May.

The sophistication of these model structures and the uncertainty about their developmental potential and whether they should be treated as embryos have made it hard to get funding for the work. In the United States, the National Institutes of Health (NIH) has been reluctant to fund such work, citing a section of federal law known as the Dickey–Wicker Amendment, which bars the government from funding research that creates or destroys human embryos. Researchers argue that the structures are different from natural human embryos and have called for clarity from the agency on the criteria that guide its funding decisions.

The agency’s policy office last year convened a meeting of leading researchers at the US National Academies of Sciences, Engineering, and Medicine in Washington DC to discuss milestones in the field. This month, NIH director of science policy, Carrie Wolinetz, wrote that the agency would consider funding stem-cell-based model structures that mimic embryo development on a case-by-case basis.

Nature 591, 510-511 (2021)


  1. 1.

    Yu, L. et al. Nature https://doi.org/10.1038/s41586-021-03356-y (2021).

    Article  Google Scholar 

  2. 2.

    Liu, X. et al. Nature https://doi.org/10.1038/s41586-021-03372-y (2021).

    Article  Google Scholar 

  3. 3.

    Sozen, B., Jorgensen, V., Zhu, M., Cui, T. & Zernicka-Goetz, M. Preprint at bioRxiv https://doi.org/10.1101/2021.03.12.435175 (2021).

  4. 4.

    Fan, Y. et al. Preprint at bioRxiv https://doi.org/10.1101/2021.03.09.434313 (2021).

  5. 5.

    Rivron, N. C. et al. Nature 557, 106–111 (2018).

    PubMed  Article  Google Scholar 

  6. 6.

    Moris, N. et al. Nature 582, 410–415 (2020).

    PubMed  Article  Google Scholar 

  7. 7.

    Sozen, B. et al. Dev. Cell 51, 698–712 (2019).

    Article  Google Scholar 

Download references

Nature Briefing

An essential round-up of science news, opinion and analysis, delivered to your inbox every weekday.

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing