The Chinese scientist He Jiankui, who revealed last fall that he used CRISPR gene editing to try to make twin newborn girls immune to HIV infection, might have also given them a higher risk of death. That’s according to a new study from University of California, Berkeley data scientists who analyzed the records of more than 400,000 people in the United Kingdom.
Looking at that large population sample, the Berkeley researchers studied the overall effects of the mutated gene that He Jiankui inserted into the embryos of the two girls with a single goal in mind—to protect them from the virus that causes AIDS.
The Berkeley study suggested that the gene mutation also comes with some significant downsides. The researchers compared death records and genomic profiles in the UK Biobank database. People who had both copies of the mutation that provides protection against HIV, known as CCR5-Δ32 (pronounced “delta-32”), were 20 percent more likely to die before age 76 than people with one copy of Δ32—or none at all.
“The cost of resistance to HIV may be increased susceptibility to other, and perhaps more common, diseases,” write Xinzhu “April” Wei and Rasmus Nielsen in Nature Medicine. Wei is a postdoctoral researcher in Nielsen’s Berkeley lab, which focuses on genetics and computational biology.
They found that the death rate of people with one copy of Δ32 was the same as in people with no copies, which could have implications in the life spans of the two CRISPR babies. He Jiankui has said that one of the twins in his experiment was given two copies of CCR5-Δ32 —a homozygous mutation—and the other twin, just one copy of the mutated gene.
While other scientists have conducted genetic editing experiments in nonviable human embryos—that is, never intended for birth—no one until He, at least publicly, had dared produce a live birth for a range of ethical and medical reasons, not least of which are the unintended harms that the Berkeley study suggests. The twin newborns are the first babies known to be the products of such an experiment.
He (pronounced “Huh”) announced last fall that he and his colleagues had edited twin embryos, then reimplanted them in their mother’s womb. He refers to the girls as Lulu and Nana. He did the experiment at the Southern University of Science and Technology in Shenzhen, China. Officials there condemned his activities once they came to light.
The ethical breaches of He’s work have spurred many, but not all, of the field’s leaders to call for a worldwide moratorium on the gene editing of human embryos. Experts stopped short of making that call at an international summit in 2015.
He’s work has not been published in a scientific journal. What’s known about it comes from slides he presented at a gene-editing summit in Hong Kong (pictured above) days after making his stunning announcement.
Soon after that, scientists around the work began to dissect what kind of changes He might have made. Sean Ryder of the University of Massachusetts Medical School concluded that the changes He claims to have made does not match any of the “wild-type” Δ32 mutations found naturally in humans, most often in Northern Europeans. In other words, he had created edits with no known precedent. Ryder called it “unconscionable.”
Stanford professor Matthew Porteus, one of the two gene-editing experts who interviewed He onstage at the Hong Kong summit, told Xconomy one week later that “to call [He’s work] a clinical trial or an experiment gives it an imprimatur that it doesn’t deserve. It was neither of these things. It was his… activity.” He Jiankui was in touch with Porteus and other scientists and ethicists in the US as he planned the procedure.
CCR5 plays a wide role in humans, and tinkering with it gets complicated fast, Wei and Nielsen write: “A mutation can be advantageous or disadvantageous depending on environmental conditions and developmental stages. In fact, despite the protection that Δ32 provides against HIV, and possibly other pathogens such as smallpox and flavivirus, and although it facilitates recovery after stroke, it also appears to reduce protection against certain other infectious diseases such as influenza.”
There are limitations to the study. The UK Biobank’s average age of enrollment is 56 years old; the sample of people large enough to do analysis fell between 41 and 76, so the data did not include people who died at an older age. As for people who died younger than the studied range, Nielsen noted that the Biobank held “a deficit of individuals” who carried both copies of Δ32, which suggests to him that the different death rates could also be at play at earlier ages.
The study might not translate to other regions, cautioned Nielsen. “Our results cannot easily be extrapolated to other populations with different environmental conditions,” he said via email. “The mortality rates we estimated could be different, for example, in China.”
He said that he and Wei hope to repeat the study for other large populations in the future.