This lab-grown ball of human cells shares many similarities with 5-day-old human embryos.


A human embryo at the blastocyst stage is smaller than the tip of a ballpoint pen and may contain fewer than 100 cells, but this developmental waypoint has long puzzled and vexed biologists and physicians. Many miscarriages occur during this stage, for example, and a blastocyst can also split to create twins. Now, multiple research groups have found ways to mimic blastocysts, coaxing lab-grown human cells to form clusters that closely resemble the true thing.

The feat, described in two Nature papers this week and two recent preprints, could enable researchers to tackle important questions about human fertility, such as why in vitro fertilization (IVF) often fails. Moreover, the ersatz blastocysts “will be windows into this stage of human development,” says stem cell biologist Aryeh Warmflash of Rice University, who wasn’t connected to the work. “They will allow us to study it in ways we could not do before.”

We were all blastocysts once. This phase, which in humans starts about 5 days after fertilization and lasts only a couple of days, is a watershed. “The blastocyst is the first stage in which we have specialized cell types developing,” says developmental biologist Janet Rossant of the Hospital for Sick Children and the University of Toronto. The stage also initiates another momentous event: implantation, in which the blastocyst nestles into the uterine lining and begins to interact with the mother’s cells to build the placenta.

But answering questions such as which genes orchestrate blastocyst development and why implantation is so often unsuccessful has been difficult. The only source for human blastocysts is donated embryos originally generated for IVF treatments, which are scarce and carry hefty ethical baggage. In the United States, for instance, researchers cannot use funding from the National Institutes of Health to study these blastocysts. Seeking an alternative, several groups of scientists have induced mouse stem cells to form blastocystlike clumps dubbed blastoids, but they don’t perfectly re-enact what happens in a human embryo.

To create a human blastoid, cell biologist Jun Wu of the University of Texas Southwestern Medical Center and colleagues initially harnessed embryonic stem (ES) cells, which can be isolated from human blastocysts and give rise to all the cell types in our bodies. Under certain culture conditions, the cells can form each of the three cell types in the blastocyst, researchers previously found. Wu and his team took that discovery a step further and showed that when they stimulated cultured human ES cells with two molecular mixtures, the cells congregated into dead ringers for blastocysts.

Because ES cells come from human blastocysts, they share many of the same ethical and practical limitations. But with the right molecular prodding, researchers can convert mature cells, such as fibroblasts from the skin, into induced pluripotent stem (iPS) cells, which have the same tissue-generating capabilities as ES cells but don’t require the destruction of embryos. Nudging human iPS cells with the same two molecular mixtures also yields blastocystlike cell clusters, Wu’s team now reports in Nature.

The second group publishing in Nature, led by stem cell biologist Jose Polo of Monash University in Australia, chanced on a different recipe for making human blastoids while studying how skin cells morph into iPS cells. The group noticed that intermediate cells, which hadn’t fully converted into iPS cells, could spawn all three types of blastocyst cells. On standard culture plates the cells couldn’t display their full potential. But in roomier quarters, they converged into spheres that closely resembled blastocysts. In preprints posted last week two independent groups, led by developmental biologists Magdalena Zernicka-Goetz of the California Institute of Technology and Yang Yu of Peking University Third Hospital, reported also making blastocystlike clusters from “extended” human stem cells.

Polo’s and Wu’s groups demonstrated that their blastoids recapitulated many characteristics of human blastocysts. They contained about the same number of cells, for example, and switched on many of the same genes. And at least in the culture dish, blastoids re-create some early steps of implantation.

Making the clusters was inefficient, and those that did form showed several key differences from IVF-derived blastocysts. “There’s a lot going on that we don’t understand,” says reproductive and developmental biologist Susan Fisher of the University of California, San Francisco. Still, she stresses, “As a first step it’s extremely exciting, and a huge amount can be learned.”

Although the new techniques are inefficient, Polo notes they can still produce blastoids in large numbers. That could enable researchers to use blastoids to test whether certain chemicals disrupt embryonic development, trace how mutations lead to birth defects, and refine IVF.

The blastoids are not embryos, Wu cautions, but are “a collection of cells that undergoes the early stages of embryogenesis.” A human blastoid could not develop into a fetus, he adds. A widely accepted research guideline, codified into law in some countries, forbids growing blastocysts for more than 14 days—and all four groups abided by that limit with their blastoids. New recommendations from the International Society for Stem Cell Research, due for release in May, could provide further guidance about working with embryolike structures such as blastoids.

But the public’s reaction to these new creations is uncertain, Fisher says. “It’s a test case for how scientists and lay people feel about a collection of cells.”