Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Embryo-like structures created from human stem cells

Coloured scanning electron micrograph (SEM) of a human embryo at the eight cell stage

An early human embryo. Structures with similar features could be used to study human development.Credit: Dr Yorgos Nikas/SPL

Biologists have developed a way to use human stem cells to make structures that mimic early embryos. The embryo-like structures are the first to produce rudimentary reproductive cells; and also go through stages that resemble several other landmarks in early human development.

Research groups are seeking to make ever more sophisticated artificial embryo-like structures that can be used, not for reproduction, but to study early-stage embryonic development. The latest method for making these structures, published in Nature1 today, has a much higher success rate than previous attempts, and can reliably produce them on demand.

Using these structures for research should be less controversial than working on embryos left over from in vitro fertilization (IVF) procedures, says Jianping Fu, a bioengineer at the University of Michigan in Ann Arbor who led the latest study.

Hear more about this research on the Nature Podcast.

Pharmaceutical companies might also one day use the structures to test whether drugs are safe for pregnant women. And physicians could use them to investigate why some woman have multiple miscarriages.

“This study could help to understand and help prevent early pregnancy loss,” says Amander Clark, a stem-cell biologist at the University of California, Los Angeles, who wrote a News & Views article2 to accompany the study. “Women who have repeat early pregnancy failure should now have hope that scientists are working on approaches to help understand why this occurs,” she adds.

But the research is likely to raise its own ethical issues. Although these structures could not grow into a person, they develop features in the lab that some people consider the point when an embryo becomes an individual.

Timing is everything

Two years ago, Fu reported3 making his first embryo-like structures using colonies of human pluripotent stem cells (PSCs), some from embryos and some made from skin cells reprogrammed to an embryo-like state. Using the right mix of biochemical signals at the right time, Fu was able to coax the colonies of PSCs — which can differentiate into other cell types — to mimic the first step by which an early embryo’s mostly homogenous cells become various tissue types.

The structures also showed early signs of developing a feature called the primitive streak, which establishes the head-to-tail axis.

But Fu says the result was frustrating because the method worked only about 5% of the time — not enough to be a reliable research tool, much less an aid in pharmaceutical clinical research.

To gain greater control over the process, Fu’s team replaced the conventional culture plates used to grow colonies of PSCs with a small device that contains different materials channels. The middle one is filled with a gel and lined with support posts that anchor the PSC colonies. The colonies are loaded through another channel, and the third channel is used to deliver a couple dozen of biochemical signals at precise times.

Fu says that the embryo-like structures produced using this process were more similar to natural embryos than the structures the team previously made. The primitive streak was better defined, and precursors of the cells that go on to form eggs and sperm emerged. The embryo-like sacs formed 95% of the time, with structures developing and cellular changes occurring at almost the same times, says Fu. “To see 10 or 15 structures develop in such a controlled and synchronized way, it was amazing,” he says.

Eric Siggia, a physical biologist at the Rockefeller University in New York City, says the reliability is the most impressive aspect of the study. “It’s tremendously powerful,” says Siggia, who also makes synthetic embryo-like structures.

Clark anticipates that synthetic embryo-like structures will be used to improve understanding of how the primitive streak forms. “This is one of the most important and least understood events in human life,” she says.

Fu’s synthetic structures showed signs of a primitive streak in just four days. He had to stop the experiments after that because the structures outgrew the channels, but he plans to improve his devices so the sacs can develop further.

Ethical line

But the presence of the primitive streak in these synthetic structures could also prove controversial, because some people consider this to be when an embryo becomes an individual human being. Some countries, such as the United States, have guidelines that forbid research on human embryos past 14 days after fertilization, which is around the time the primitive streak forms. Others, including the United Kingdom, have explicit laws against it.

The US National Institutes of Health (NIH) does not have a clear policy on research using synthetic embryo structures that have developed a primitive streak. But Fu and Siggia say none of their projects have been funded. A 2019 NIH call for proposals on brain research specifies that work on synthetic embryos cannot be funded. “This causes a lot of confusion and uncertainty, discouraging researchers working in this field from applying for NIH grants,” says Fu, whose research was mostly funded by his university.

An NIH spokesperson told Nature that research proposals using synthetic embryo structures are judged on a case-by-case basis.

Some countries will probably introduce regulations for this kind of research, says Magdalena Zernicka-Goetz, a developmental biologist at the California Institute of Technology in Pasadena. “We will have to confront ourselves with the question of what is a human embryo, and whether these models really have the potential to develop into one,” she says.

Fu, Siggia and Clark think that because synthetic embryos are not the same as intact human embryos, they should not be subject to the same rules as those donated from IVF clinics. The embryo-like structures lack a placenta and other cells crucial for development, and could not possibly develop into a person, says Siggia. “It’s like putting four wheels on a frame and saying it's a car, even though there is no engine,” he says.


Read the related News & Views: ‘Human embryo implantation modelled in microfluidic channels’.

Updates & Corrections

  • Correction 16 September 2019: An earlier version of the story stated that Jianping Fu was stem-cell biologist.


  1. 1.

    Zheng, Y. et al. Nature (2019).

    Article  Google Scholar 

  2. 2.

    Clark, A. T. Nature (2019).

    Article  Google Scholar 

  3. 3.

    Shao, Y. et al. Nature Commun. 8, 208 (2017).

    PubMed  Article  Google Scholar 

Download references


Nature Careers


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


Quick links