All products featured on WIRED are independently selected by our editors. However, we may receive compensation from retailers and/or from purchases of products through these links.
When Chinese scientists announced in April they had edited human embryos using a new genetic tool called Crispr, the headlines blared “designer babies,” and the world woke up to Crispr’s power. But for those looking closely at the data, the paper was a relief.
Crispr did not work very well to edit the human embryos—nonviable ones, to reduce the ethical problems with the research. In fact, it worked terribly. Out of 54 embryos the team altered with Crispr and tested, only four ended up with the right genetic tweaks. And even these four embryos had off-target mutations, places where Crispr edited the DNA even though it shouldn’t have. Off-target mutations in the wrong gene can be dangerous, and the paper was proof it was too unsafe to “fix” viable human embryos with Crispr. At least for now.
Today, on the first day of an international summit on human gene editing in Washington, DC, MIT researchers announced that they have tweaked a Crispr protein to reduce those off-target effects. It’s the latest in a series of improvements to the Crispr system that, together, are inching the error rate down toward practically zero.
As long as Crispr creates too many off-target mutations, it’s easy to assert that the technology is not predictable, and that editing human embryos is therefore not ethical. Of the five steps Crispr co-inventor Jennifer Doudna recently proposed in Nature for oversight of human gene editing, the first is adopting standards for measuring efficiency and off-target effects. But if gene editing with Crispr is safe, do the ethics change? That’s a big question for the DC meeting—one that Crispr scientists have often tiptoed around.
“Safety is something we all agree is important. There’s a language for that that’s acceptable to all sides,” says Josephine Johnston, director of research at the Hastings Center. “But the other kinds of concerns that people have about this work are much more difficult to have conversations about.” Those concerns are not scientific but social, including questions about consent of unborn babies and the roles of humans in making permanent genetic changes. As the Crispr technology improves, the safety concerns are peeling away to reveal the ethical concerns at the core.
On face of it, the MIT paper is yet another molecular nuts and bolts study with a barely scrutable title (“Rationally engineered Cas9 nucleases with improved specificity”). Crispr is essentially a pair of DNA scissors, and Cas9 is the protein in the Crispr system that unzips DNA and runs along looking for its target by matching the DNA sequence against a snippet of its guide RNA. When Cas9 finds its target, snip snip. The problem is that Cas9 will sometimes think it’s found a target even when up to five of the guide RNA’s approximately 20 letters do not match the DNA—hence the off-target mutations.
Feng Zhang, another scientist with a claim to CRISPR's invention as a gene-editing technique, and his colleagues at MIT changed one part of Cas9 slightly so its guide RNA binds fewer mismatched DNA sequences. This improved Cas9’s specificity by about 25-fold at the sites they tested. Zhang’s lab, along with that of other labs of Harvard, have come up with several other ways to make Cas9 better behaved, such as pairing up Cas9s so they only work in twos and both guide RNAs match.
With these improvements, Harvard geneticist George Church estimates that Crispr’s error rate, best-case scenario, could be just 1 in 300 trillion letters of DNA. (The rate can vary quite widely in different types of cells and with different guide RNA designs.) Even at the higher end, that’s comparable to the spontaneous mutation rate in humans, says Church.
He asserts that off-target mutations are already a solved problem. With typical Churchian flair, he says, “I’m beginning to feel like this is hitting a fly with sledgehammer. I think the thing is already dead.” Even if not all scientists agree about off-target mutations yet, solving the problem of Crispr’s non-specificity is looking more like a question of when rather than if.
A conversation about the ethics of human modification has been kicking around since the 1980s, when scientists started using cruder forms of gene editing. “For decades, we’ve been able to say it’s not there yet, so we’re not going to do [human gene editing]. It was an easy way to stop the conversation,” says Debra Mathews, a bioethicist at Johns Hopkins University. “We’re now at a point where it is precise enough that we do actually just have to have the conversation.” Over the next three days in DC, scientists and policymakers will have to do just that.
You may also like:
