Origin Stories

Developmental biologist Lewis Wolpert famously declared that “it is not birth, marriage or death, but gastrulation which is truly the most important time in your life.”

He had a point. The earliest events in development—in which a cluster of identical stem cells develops a distinct axis and then starts to form its gut—defines an animal’s entire body plan. Doing it one way builds a cat; doing it just a little differently, a crocodile.

“What’s really interesting is there are processes at the beginning of development that are conserved across animal life, yet even within mammals, they can be very different,” says Mijo Simunovic, PhD, assistant professor in the Departments of Genetics & Development and Chemical Engineering and member of the Columbia Stem Cell Initiative.

A better understanding of early embryogenesis would not only reveal how an embryo takes on its unique shape but could eventually have immense medical impact.

Most pregnancy failures occur at implantation, and placental defects are among the biggest causes of pregnancy-related mortality. But despite decades of effort, early embryogenesis has been very hard to study.

In humans and other placental mammals, this early stage of development occurs while the embryo implants into the wall of the uterus and forms the placenta. Mouse embryos can help answer many questions about embryogenesis. But it is becoming increasingly clear that mice and humans develop very differently at certain stages of early development, particularly implantation.

“Understanding early human embryogenesis demands human models,” says Simunovic, who has been working for nearly a decade to create such models from human stem cells, first as a Simons Junior Fellow at Rockefeller University and now in their own lab which opened at Columbia in 2019.

“We simply didn’t have any access to studying implantation in a lab, and now we’re hoping with our stem cell model that this is going to be possible, at least in some ways.”

Mimicking human development

To build a human model of implantation, Simunovic’s lab looks for a way to simplify the problem. “We’ve asked: ‘Can we build up these embryonic phenomena from their most basic components?’” Simunovic says. A paper the team published in Cell Stem Cell last year revealed just the beginning of how far the team has come toward that goal.

Beginning with human pluripotent stem cells, the researchers used advanced tissue engineering techniques to produce a model of the epiblast, a spherical ball of cells that forms at the beginning of development and, in a real embryo, represents tissues that creates the organs. From the same pluripotent stem cells, the team innovated a method of producing extra-embryonic tissues, mimicking placenta formation in a dish. Co-culturing the epiblast models with the extra-embryonic tissues produced an assembloid, in which the epiblast self-organizes with the extra-embryonic layers to form structures that mimic post-implantation development.

“The process that we’re mimicking is the one that happens right after the embryo implants into the uterus,” says Simunovic, “and this is now giving us access to study the molecular mechanisms of this process.”

Since publication of Simunovic’s paper, several groups around the world reported their own stem cell models that mimic human post-implantation development, casting a spotlight on this new field of embryo modeling and raising considerable interest from the press.

The assembloids provide one of the best models to date of early human development, but Simunovic is quick to emphasize an important point: These are not embryos, and the team has no interest in producing embryos. “Our field is not only driving new biology, but also new bioethics, which is why it’s very important that we present an accurate picture, beyond any sensationalism,” says Simunovic.

The importance and rigor of Simunovic’s assembloid work is increasingly being recognized and supported. In 2023, Simunovic was named an Allen Distinguished Investigator, to lead a project to understand how sex hormones influence early organ development; a NYSCF Robertson Investigator by the New York Stem Cell Foundation; a Pew Scholar in the Biomedical Sciences; and a Schaefer Research Scholar. Their assembloid work also receives support from the Burroughs-Wellcome Fund Next Gen Pregnancy Initiative.

A zoo of assembloids

Now that they have a model of human embryonic implantation, Simunovic’s team is working on duplicating the work in other species hoping to catalyze research into nonconventional biological models that hold secrets to our own development.

“We have actually accumulated a whole zoo of various kinds of stem cells in the lab, beyond human, that includes various primates, small and some large mammals like rhinoceros and horse and cow,” says Simunovic.

Within the next few years, the researchers hope to investigate the evolutionary conserved mechanisms that shape mammals at the molecular level, unlocking the secrets of the most important time of their lives.

In the Beginning, There Was Physics

For someone who studies development, Simunovic followed an unusual developmental path, obtaining a PhD in chemistry from the University of Chicago, followed by a PhD in physics from the Sorbonne Université in Paris. “I have an affinity for the quantitative and physical world, but I am much more interested in the biological questions,” says Simunovic. Postdoctoral training in developmental biology and embryology followed the two doctorates, and Simunovic has since focused on melding the two branches of science: “I think that modern biology really kind of requires, in addition to classically trained biologists, also physicists and engineers to really make advancements.” Unsurprisingly, that attitude has led to the creation of a broad-based research effort. Simunovic’s lab includes a mix of engineers and biologists, and their collaborators range from physicists to clinicians. The team’s projects cover a similarly diverse range. “Some people in the lab are studying the fundamental developmental biology of early human development, some are studying processes of evolutionary convergence in developmental signaling, and there are also people who are employing engineering approaches to build organoid models of the human uterus,” says Simunovic.