Read our COVID-19 research and news.

A human embryo is injected with gene-editing molecules.

Dr. Kathy Niakan/Nature

Scientists use gene-editing technology to knock out genes in human embryos for first time

For the first time, scientists have used gene-editing techniques on human embryos to probe how they develop. The work suggests that a protein called OCT4 has additional roles in human embryos compared with those of mice. But biologists say it is more important as a proof of principle; previous human embryo–editing research has focused instead on correcting faulty genes. 

The new experiments, published today in Nature, are also a first test of the United Kingdom’s carefully crafted embryo-editing research regulations, which require researchers to undergo a review by the national Human Fertilisation and Embryology Authority and receive a license before going forward. Kathy Niakan, a developmental biologist at the Francis Crick Institute in London, applied in 2015 to use the CRISPR editing technique on human embryos to learn more about the genes active in early development. The researchers planned to focus first on OCT4, known as a marker for pluripotent stem cells—cells that can become all tissues in the body.

Niakin and colleagues used CRISPR to deactivate the gene that codes for OCT4 in 37 single-cell human embryos left over after in vitro fertilization treatments and donated by couples. Mouse embryos lacking the protein form mostly placental cells; the cells destined to form the fetus don’t appear. But in the human embryo knockouts, placental cells and yolk sac cells also failed to form. That suggests that in humans OCT4 is plays a role in the development of all three of these fundamental cell types. The researchers plan further work with CRISPR to find out exactly what genes OCT4 controls in the different cell types.

The work shows that that “you can do [CRISPR] effectively enough and efficiently enough” to study development, says Janet Rossant, a developmental biologist at The Hospital for Sick Children and the University of Toronto in Canada. Researchers have relied on mouse models to understand early mammalian development, she adds, but to understand human development and how it can go wrong, the real thing may be best.