In 2018, the Chinese scientist He Jiankui, Ph.D., altered the DNA of two human embryos, which were born as two healthy twin girls named Lulu and Nana. Against the violent ethical uproar in the medical community, He defended his experiment by claiming that the genetic changes he made were benign — they were meant to promote HIV resistance — but research published Monday in Nature Medicine has revealed that he may have spoken too soon.
Using the gene-editing tool CRISPR, He created a mutation in the CCR5 gene called the ∆32 mutation in the embryos. People with two copies of this mutation seem resistant to the HIV virus, and gene therapy involving this gene is a promising avenue for HIV treatment in adults. But He’s approach created a change in the embryos that would in theory make the twins HIV-resistant before birth — and be passed on to all future generations — but it didn’t go according to plan. He only succeeded in creating this mutation in one of the girls, and as the new paper shows, doing so may have unforeseen consequences.
According to the study, this mutation seems to sharply increase an individual’s susceptibility to other major diseases, like the flu, points out lead study author Rasmus Nielsen, Ph.D., an integrative biologist at the University of California, Berkeley. Nielsen estimates that people with two copies of the mutation have a 21 percent greater chance of dying before age 76 than those who don’t.
“It’s very hard to predict what the effect of a mutation is,” he tells Inverse. ”You might think that is’t good to have a mutation because it does something beneficial to a child, but it may come with all kinds of other risks that are too hard to predict.”
An Unexpected Downside
Nielsen’s study of the CCR5 ∆32 mutation used genetic and lifestyle data from 409,693 people of British ancestry who make up the population of the UK Biobank. He and his team investigated the survival rate of people who naturally had one copy of the mutation, two copies, or no copies, and doing so revealed a much lower survival rate in people with two copies — like the genetically modified twin. Nielsen posits this pattern is due to the mutation’s effect on the impact of common — but sometimes fatal — diseases.
The CCR5 ∆32 mutation has shown great promise for creating resistance to HIV because it changes immune cells in a way that makes it difficult for the HIV virus to latch onto them. But that change seems to also allow the influenza virus to become more dangerous. For example, an analysis of genetic markers from 171 patients who contracted flu during the 2009 pandemic found that people with two copies of the mutation had a 17.4 percent mortality rate. By comparison, people without the mutation only had a 4.7 percent mortality rate.
Do these staggering statistics mean that the CCR5 mutation is harmful? Not necessarily, says Nielsen. If you live somewhere where flu is a threat, having this mutation is not ideal, but CCR5 is still helpful in some contexts, like in the Northern European populations in which it confers resistance to HIV and occurs most often.
What Does This Mean for Lulu and Nana?
Nielsen’s study takes a big-picture look at a small but powerful mutation in an attempt to predict whether the mutation might help increase survival or do the opposite.
“To me, like with any other medication, you have to weigh the benefits against the costs,” he says.
At least in the British population he studied, his analysis showed that the downsides outweigh the upsides. Whether those results will hold true for the twins in He’s experiment remains to be seen. Because this study used data on people of British ancestry, we can’t be too hasty to predict how the CCR5 ∆32 mutation might affect someone with a different genetic background, like the Chinese twins.
“Genetic background matters a lot,” explains Dana Carroll, Ph.D., a biochemist at the University of Utah who was not involved with this study. “∆32 and CCR5 is extremely rare in Chinese populations, so I think it’s harder to predict what the consequences might be in someone of a different genetic background. I think that’s something that people have kind of lost sight of, and that obviously Dr. He didn’t take into consideration at all.”
It remains unclear how the future will play out for Lulu and Nana with regard to CCR5, which, Carroll adds, underscores an important point. “I think that one has to be very careful about taking out an allele that’s prevalent in one population and moving into other populations where there’s been no adaptation to what the consequences of that allele might be,” he says.
In other words, we have no idea what it means not only to tweak the genome but to mix and match genes have persisted in totally different environmental circumstances.
“It certainly does add another perspective to CCR5 knockouts,” Carroll says.
For that reason, and many others, Nielsen maintains hat there was no excuse for He’s work. “I think there’s no justification for doing these experiments in embryos that are implanted,” he says.
That’s not to say that there are no good reasons to figure out a safe way to edit the human genome. Those ideas are being considered right now. Carroll is a member of newly created Commission on Human Germline Editing, an expert commission convened by the United States National Academy of Sciences and the UK’s Royal Society that was created partially to address issues just like the questions that Nielsen’s study raises. Is there an ethical way to edit the human genome? And if there is, can we truly predict what that outcome might be?
The need to predict how one gene may cause ripple effects will be crucial, and that day is probably coming very soon.
We use the genotyping and death register information of 409,693 individuals of British ancestry to investigate fitness effects of the CCR5-∆32 mutation. We estimate a 21% increase in the all-cause mortality rate in individuals who are homozygous for the ∆32 allele. A deleterious effect of the ∆32/∆32 mutation is also independently supported by a significant deviation from the Hardy–Weinberg equilibrium (HWE) due to a deficiency of ∆32/∆32 individuals at the time of recruitment.