Lit from below by the microscope’s soft glow, the translucent mouse embryos looked exactly as they should. On day 3 they began to elongate, from spheres into cylinders. On one end, the neural tube started to fold around day 6, on the other a tail began to bud. By day 8, a beating heart began to circulate blood through vessels forming around the embryo’s yolk sac.
But these embryos weren’t the product of an egg and a sperm. They weren’t even growing in the uterus of a female mouse. They were developed inside a bioreactor, and made up entirely of stem cells cultured in a Petri dish.
The experiments, performed at the Weizmann Institute of Science in Israel and published Monday in Cell, mark the first time researchers have grown fully synthetic mouse embryos — that is without the use of sperm or eggs — outside the womb.
The advance opens up new avenues for studying how stem cells form various organs in the developing embryo and better understanding how certain mutations drive various developmental diseases. It also raises profound questions about whether other animals, including humans, might one day be cultured from stem cells in a lab.
“As soon as the science starts to move into a place where it’s feasible to go from a stem cell population in a Petri dish all the way through to organ development — which suggests one day it will be possible to go all the way to creating a living organism — it’s a pretty wild and remarkable time,” said Paul Tesar, a developmental biologist at Case Western Reserve University School of Medicine who was not involved in the study.
Since the 1980s, developmental biologists have been taking apart embryos cell by cell to try to understand how they eventually become all the specialized tissues that allow fish to swim and mice to scurry and humans to walk and talk. In the last decade or so, researchers have learned enough about the signals that send stem cells down these differentiation paths to be able to put them back together into things that resemble organs (organoids), recently fertilized eggs (blastoids), and even embryos (embryoids).
But these balls of mouse and human cells could only be grown in dishes and test tubes for a short amount of time before they’d poop out. They needed a living womb to support their further development, or even better, an artificial approximation of one.
Jacob Hanna, an embryonic stem cell biologist at the Weizmann, spent seven years engineering a tubed system of spinning glass vials housed in an incubator to do just that. Last year, his team reported in Nature that their mechanical uterus could keep natural mouse embryos alive for up to 11 days.
“That really showed that mammalian embryos can grow outside the uterus — it’s not really patterning or sending signals to the embryo so much as providing nutritional support,” Hanna told STAT in an interview. The next step was to see what would happen if they put stem cells — rather than natural embryos — into their contraption. “Can these cells make an entire embryo? That was a big unanswered question for the field.”
In this latest work, the team combined that system with a novel cocktail of stem cells, some of which had been chemically coaxed to over-express genes that switched on development of the placenta and yolk sac — tissues that are vital for supporting the healthy growth of embryos.
The synthetic embryos were able to grow to day 8.5, developing the beginnings of a well-shaped brain, a neural tube, and an intestinal tract, as well as a beating heart. Analyses of the synthetic embryos’ gene expression patterns across different tissues showed that they were 95% similar to a natural mouse embryo of the same age.
“We found that these cells do have this incredible self-organizing capability that can be unleashed if given the right artificial settings,” said Hanna.
However, the work has some important limitations. Day 8.5 is still relatively early; the full gestational time for a mouse is 20 days. And the embryos that survived that long were a rarity. Only about 50 of 10,000 cellular clumps self-organized into embryos. The rest failed to develop properly.
“This is just one step, but a very important step for us to be able to study early development,” said Tesar. “We’re crossing into the realm of being able to generate an embryo from scratch, and potentially a living organism. It’s been a really notable switch for the field.”
While scientists have gotten very good at rewinding mature cells to the more primitive stem cell state, figuring out exactly which chemical signals will cause a stem cell to produce the precursors of a liver or kidney has been much more challenging. Experiments trying to nudge stem cells to form specialized tissues have tended to produce jumbled mixes of cells instead, lacking organization and with the wrong compositions of cell types.
Researchers say the new work from Hanna’s team should provide a way forward for getting those recipes right, in part because the transparent bioreactor allows scientists to observe organs developing in front of their eyes, but still in the context of surrounding support tissues. And because by starting with stem cells instead of fertilized eggs, they can produce these embryonic structures in a much more scalable and controlled way.
“This is going to tremendously refine the roadmap to tissue and organ formation,” said Nicolas Rivron, a stem cell biologist at the Institute of Molecular Biotechnology of the Academy of Sciences in Vienna. “It’s going to teach us the minimal structures, the minimal elements that will be necessary to eventually form full-fledged organs. That alone is absolutely priceless.”
Beyond basic research though, the bigger impact of this work is its potential to one day be applied to other species, including humans. Synthetic embryos derived from stem cells offer scientists the opportunity to probe in unprecedented detail the early days of human development while providing a less controversial and ethically fraught alternative to human embryos — the study of which has historically been limited by funding bans and the willingness of IVF donors.
The synthetic embryology revolution isn’t going to happen tomorrow. There are numerous technical hurdles to translation — humans have much longer gestation periods and they grow much larger than a mouse, as well as being a more complicated organism. But that kind of work always starts somewhere, and it usually starts with mice. That means it’s not too soon to start thinking about where this could all be headed.
“The more and more we show the capacity for pushing stem cell-derived embryos further and further in development, the more synthetic embryos and natural embryos begin to merge,” said Tesar. “There will always be a gray area, but as scientists and as a society we need to come together to decide where the line is and define what is ethically acceptable.”
Hanna, for his part, isn’t interested in synthetic embryos for reproductive purposes. The ultimate goal he’s working toward is making organs and tissues for transplantation and to treat human diseases. He sees synthetic organoids not as potential lifeforms so much as as biology’s best 3D printer.
“You can view this as a universal differentiation protocol,” he said. Rather than needing different complicated chemical recipes to make a stem cell become a liver or a lung, embryoids, even very early-stage ones can give a stem cell all the signals it needs to produce potentially life-saving therapies.
Imagine a patient with untreatable leukemia — they need a bone marrow transplant to survive. In Hanna’s future, scientists can take a biopsy of skin cells from that patient, wind them back into stem cells, grow them in naive conditions, and put them in this specialized bioreactor. The end result? An army of bone marrow stem cells that can be given to that patient, without them having to wait for a donor match that might never come. “It’s early days but we’re really opening up the field to explore these possibilities more seriously,” said Hanna. “We’re moving from science fiction to science.”