The human body and all its complexity arise from just a small collection of cells that divide and morph into different types of tissues. But exactly how this occurs is hard to study because embryos are hidden inside their mothers. Some embryos are donated to science by individuals who have undergone in vitro fertilization, but these embryos are limited in availability and by strict ethical and legal regulations.
Consequently, scientists have turned to laboratory models to study the process: embryo-like models made from both mice and human stem cells, rather than eggs and sperm. Now, reporting in the journal Nature, Magdalena Zernicka-Goetz, Bren Professor of Biology and Biological Engineering at Caltech, and her colleagues have generated a human embryo–like model that mimics aspects of the second week of human development, a time after embryos become implanted in the womb.
These models are not living entities that can grow into actual embryos, but they may offer insights into human defects and diseases, how embryos develop, why some pregnancies fail, and may even lead to new ways to develop synthetic organs for transplants.
"Our human embryo–like model, created entirely from pluripotent human stem cells, gives us access to the developing structure at a stage that is normally hidden from us due to the implantation of the tiny embryo into the mother's womb," said Zernicka-Goetz, who is in the process of moving her lab to Caltech from the University of Cambridge's Department of Physiology, Development and Neuroscience, where the work was carried out.
"This exciting development allows us to manipulate genes to understand their developmental roles in a model system. This will let us test the function of specific factors, which is difficult to do in the natural embryo," says Zernicka-Goetz, who is also an affiliated faculty member with Caltech's Tianqiao and Chrissy Chen Institute for Neuroscience.
Today it was announced that Zernicka-Goetz has been selected as the recipient of the 2023 Ogawa-Yamanaka Stem Cell Prize, awarded by Gladstone Institutes and supported by Cell Press, for her pioneering work to develop the first integrated stem cell–derived embryo models, research she began 10 years ago. These integrated models combine both embryonic and extraembryonic stem cells, which are those cells that develop into structures that support a growing embryo, such as the placenta and the yolk sac.
The models were developed more fully in a series of papers published between 2017 and 2021, during which time her team created mouse embryo–like models from the embryonic and extraembryonic stem cells. In 2022 they, and other scientists, showed that these mouse models could develop further to form progenitors of all brain regions, spinal cord, gut tube, and primordial germ cells. The mouse models also had beating hearts.
The new human embryo–like models are not as advanced as their mouse counterparts—they do not contain beating heart-like structures, for instance—but they do contain both embryonic and extraembryonic tissues, which would normally develop to form the placenta, the yolk sac, and the amnionic sac.
The new models also develop primordial germ cells, the cells that go on to become sperm and egg in living humans. In fact, this is the first time the primordial germ cells have been created in a model containing both embryonic and extraembryonic tissues, as is the case in natural development.
"In the lab, we are very interested in how embryonic and extraembryonic tissues interact and pattern each other to allow for cells to differentiate and for the embryo to develop properly. Importantly, there are several cell types in the human that are not present or are specified differently in the mouse, and this new model allows us to study these," says Bailey Weatherbee of the University of Cambridge, who co-led the study with Carlos Gantner of the University of Cambridge.
The researchers say a key to their new model is the lack of what are called exogenous factors—chemical signals that are often added to an embryo-like model in a dish to trigger developmental changes. In the new embryo-like model, the growing cells produced chemicals that trigger these changes on their own. In other words, the system behaved closer to what happens in a human womb.
"Not having these exogeneous factors really opens the ability to study how these tissues—the embryonic and extraembryonic tissues—naturally talk to each other," says Gantner. "This tissue interaction is incredibly important, and we think it fails in a proportion of pregnancies leading to their loss."
There are clear regulations governing stem cell–based models of human embryos, and all researchers doing embryo-modeling work must first be approved by ethics committees. Journals require proof of this ethics review before they accept scientific papers for publication. Zernicka-Goetz's laboratory holds these approvals.
The new paper, titled "A post-implantation human embryo model derived from pluripotent stem cells," was funded by several grants to Zernicka-Goetz such as those from the Wellcome Trust, the European Research Council, OpenAtlas, the Gates Cambridge Trust, and the Leverhulme Trust Early Career Fellowship. Other study authors include Lisa Iwamoto-Stohl of the University of Cambridge; and Riza Daza, Nobuhiko Hamazaki and Jay Shendure of the University of Washington. Shendure is also affiliated with the Brotman Baty Institute for Precision Medicine, the Howard Hughes Medical Institute, and the Allen Discovery Center for Cell Lineage Tracing. Zernicka-Goetz is a NOMIS Distinguished Scientist and Scholar Awardee.