Human embryos, newly fertilized (left) and at the eight-cell stage (right), that had DNA edited by CRISPR.

Shoukhrat Mitalipov

First U.S.-based group to edit human embryos brings practice closer to clinic

The ethical and practical debates over using the DNA-editing method CRISPR to alter human embryos just got less hypothetical. A week after the news leaked out, a U.S.-based team has published the first rigorous demonstration that CRISPR can efficiently repair a gene defect in human embryos—one that would cause a potentially deadly heart condition—without introducing new mutations elsewhere. Although none of the labmade embryos were transferred into women, the research team, led by embryologist Shoukhrat Mitalipov of Oregon Health and Science University (OHSU) in Portland, says the success paves the way for using the technique in the clinic to prevent the transmission of genetic disease.

Because their approach appears to avoid the problems of patchy and imprecise editing seen in previous CRISPR tests on human embryos, the researchers claim it’s a viable strategy for rescuing mutated embryos that would otherwise be screened out of in vitro fertilization (IVF) procedures. But evidence of the technique’s long-term safety is still lacking, and many researchers and ethicists have argued that germline editing—making permanent, heritable changes to the genome that could correct genetic disease, but also theoretically introduce other designer traits—should for now be limited to research exploring basic biology.

“I’m uncomfortable, honestly, with the sort of stated purpose of this study,” says Jennifer Doudna, a molecular biologist at the University of California, Berkeley, who is among the pioneers of the CRISPR method. “It’s not about research, I don’t think. It’s about how we get to a clinical application of this technology.”

Mitalipov’s lab has navigated ethically complicated embryo research before. He advanced a technique to prevent the transmission of disease-causing mutations in a woman’s mitochondria, organelles with their own genes, by transferring her nuclear DNA into a donor egg, which is controversial because any offspring would inherit DNA from three individuals.

But the CRISPR editing project was especially hard to sell to his university, Mitalipov says. He first made the proposal to OHSU’s institutional review board about 3 years ago. His plan was to use CRISPR—a DNA-cutting enzyme and an RNA that guides it to a target sequence—to slice into the gene MYBPC3 at the site of a mutation that leads to an enlarged heart and can cause sudden cardiac arrest, even in young, seemingly healthy athletes. The researchers would also insert short DNA strands bearing the healthy gene sequence. Then they would rely on a human embryo’s natural ability to repair cuts in its DNA, hoping it would use the healthy sequence as a template.

The university set up two committees to judge the proposal, one evaluating its ethics and the other its scientific merits. Some of their members—kept anonymous even from Mitalipov—were hesitant to sign off, he says. The three other published human embryo–editing experiments, all from Chinese research teams using small numbers of embryos, have suggested that CRISPR’s enzyme sometimes cuts unintended targets in the DNA. They also produced embryos that were mosaic: A portion of their cells contained the healthy gene, whereas others kept the mutated one. Committee members thus worried the technique was too inefficient and risky to improve on current IVF procedures, Mitalipov recalls.

Others questioned whether CRISPR technology was needed at all. A person carrying a mutated copy of the MYBPC3 gene still has a 50% chance of passing on the other, healthy copy, and doctors can already screen out mutated embryos during IVF. (People with two mutated copies of the gene are more rare, and their condition is more severe.) Mitalipov pushed back against reluctant committee members with his own take. “Discarding 50% of embryos [in IVF], knowing that you could actually correct the mutation, is morally wrong,” he told them.

Because the work required creating and destroying human embryos, it was barred from receiving U.S. government funding. The OHSU lab used institutional funds; collaborators at the Salk Institute for Biological Studies in San Diego, California, relied on funding from three charitable foundations; Korean and Chinese collaborators also got federal or regional funding to help with the project.

The OHSU group collected eggs from healthy women recruited and compensated for the research, and sperm from a man whom the OHSU’s cardiovascular institute had identified as having one mutated copy of MYBPC3. Instead of injecting the CRISPR system hours after the egg is fertilized, as in previous published experiments, they added it right alongside the sperm, hoping to prevent mosaicism by catching the new embryo before it had a chance to divide and make copies of the mutated gene. And to reduce the chance of cuts at unintended parts of the genome, they relied on a short-lived version of CRISPR, whose enzyme and guide RNA wouldn’t stick around in the cell after making their initial edit.

An earlier edit

Introducing CRISPR machinery at the point of fertilization appeared to eliminate patchy genome repair known as mosaicism.

Healthyegg Egg’sDNA SpermDNA Variationsin repair CRISPR added as DNA replicates Mosaicembryo Sperm withmutation Nucleus Postfertilization editing Healthyegg Uniformhealthy embryo Simultaneous injection CRISPR Mutated sperm and CRISPR introduced Egg’sDNA Corrected sperm DNA
G. Grullón/Science


Of 58 IVF embryos that developed after the CRISPR injection, nearly three-quarters managed to repair the paternal MYBPC3. None of these successfully edited embryos harbored cells with the mutated gene, the researchers reported this week in Nature. And they found no evidence that CRISPR had cut outside the intended site.

Unexpectedly, all but one of the embryos repaired the sliced MYBPC3 using the existing healthy copy of the gene (inherited from the egg donor), instead of the added template. Compared with more developed cells, maybe the early embryo “evolved to more efficiently repair errors in the DNA,” says Jun Wu, a stem cell biologist at Salk and a collaborator on the project. (Mitalipov cautions, however, that the repair process at work in these embryos is unlikely to be efficient if an embryo inherits mutated versions of a disease-causing gene from both parents. In such cases, researchers would need a more effective DNA template, Mitalipov says.)

“This paper seems to allay many of the concerns about risk,” says George Daley, a stem cell researcher at Boston Children’s Hospital and dean of Harvard Medical School in Boston, “but it’s really, really important not to overinterpret and generalize.” Efficiency and the risk of off-target editing can vary based on the targeted gene, he notes.

To some clinicians, even a slight increase in the healthy IVF embryo pool for certain couples seems justification enough for turning to CRISPR. Even if half of the embryos don’t inherit a mutated gene, those with the healthy gene may bear other abnormalities in older parents, and finding a viable one through screening can be difficult, says James Grifo, a reproductive endocrinologist at New York University’s Langone Medical Center in New York City. “I don’t think we should be alarmist about these possibilities of treating and avoiding disease.”

But earlier this year, a committee convened by the U.S. National Academy of Sciences and the National Academy of Medicine in Washington, D.C., took a different position: that clinical use of germline editing could be allowed, but only in situations where a couple otherwise has no chance of a healthy biological child. In the new study, “we already have a case that challenges [those] criteria,” says Jeffrey Kahn, a bioethicist at Johns Hopkins University in Baltimore, Maryland, and a member of the panel. “As these kinds of research findings accumulate … at what point will somebody say it’s time to use this in a clinical context?”

Mitalipov shares the concern that embryo editing will be used in the clinic before it’s fully understood. Congress prohibits the U.S. Food and Drug Administration from approving clinical trials involving embryo editing. But despite a similar restriction on the mitochondrial method Mitalipov pioneered, a U.S. fertility specialist used it last fall to produce an apparently healthy baby in Mexico. As Mitalipov’s team continues to optimize the gene-editing technique for a possible clinical trial, “we’re not going to transfer [embryos to a uterus] without oversight,” he says. But, he adds, “the private clinics will be using it one way or another.”