Skip to Content
Biotechnology and health

Years before CRISPR babies, this man was the first to edit human embryos

In 2015, an unknown Chinese scientist edited the DNA of human embryos. It was a step on an inexorable path to designer babies.
December 11, 2018
Photograph of Junjiu Huang
Photograph of Junjiu Huang

It could have been anyone. It was so easy. But it was him. Junjiu Huang.

In 2015, Huang, a stem-cell researcher at Sun Yat-Sen University in Guangzhou, first reported using the gene-editing tool CRISPR on human embryos. His paper was rejected by top Western journals on the grounds that it didn’t follow ethics rules and presented scant science, but that April it found its way into print in an obscure English-language publication in Beijing.

The result was, in Chinese, xuān rán dà bō (轩然大波), or “towering waves”—a sensational controversy.

Huang had only carried out a lab experiment, in which he’d tried to fix a gene error that causes a blood disease. His test subjects were abnormal IVF embryos, smaller than the period at the end of this sentence, and they were soon destroyed. No attempt was made to create a child.

Still, Huang had broken a taboo: altering the DNA of so-called germline cells, those that affect heredity. The implication was clear. Genetically edited people could one day be born. And those changes would be passed on to future generations.

The reaction to Huang’s work was instant, visceral, and global. Humankind could drive its own evolution, but the person holding the wheel was a youthful-looking biologist from southern China whom no one had ever heard of. His scientific effort was called “totally premature” and a “dreaded” experiment. The dean of Harvard Medical School ascribed to Huang potentially “deranged motivations.”

I wrote about Huang’s research in 2015, so it was with a sense of déjà vu that I watched the international reaction this November when He Jiankui, a scientist at the Southern University of Science and Technology in Shenzhen, announced he’d changed the DNA of human embryos to make them HIV-resistant and implanted them into women—one of whom, He claimed, gave birth to twin girls. Once again, an ambitious Chinese scientist had crossed into unknown territory to score a controversial first. Once again, his papers were rejected and attacked by furious Western scientists.

Photo of Jiankui He
Jiankui He
AP Photo/Mark Schiefelbein

These events have made it apparent that the scientific community is deeply uncertain and conflicted about how to roll out a technology that will affect humanity’s shared gene pool. He’s shocking baby announcement came just before a major international summit in Hong Kong whose purpose had been to discuss gene-editing technology and its implications for reproduction—that is, to determine whether there should be CRISPR babies at all. Despite the wave of anger at He’s news, the summit leaders did not conclude that humankind isn’t responsible enough to engineer its own heredity or should enact a moratorium while we learn a bit more. Rather, it ended with the clearest call yet by science leaders to move the technology toward medical use in IVF clinics.

The dream is that future generations will benefit from longer and healthier lives because their bodies possess genetic vaccines against heart disease, Alzheimer’s, and more. The nightmare, though, may be pretty close to what He was allowed to present on stage in Hong Kong: children whose genomes were clumsily and needlessly mutated as part of an international science race.

The dream is that future generations will benefit from longer and healthier lives. The nightmare, though, is children whose genomes are clumsily and needlessly mutated as part of an international science race.

It could still be that China’s government cracks down on the gene-­editing efforts: at the time of writing, He was under investigation by everyone from the local health board to China’s ministry of science and technology, and he had vanished from view. In the US, starting pregnancies with gene-edited embryos was blocked by Congress in 2015. But American researchers have been prowling labs in China, looking to set up shop there and push CRISPR babies forward in ways they can’t at home.

Once gene editing was developed, the pace of technological exploration made it inevitable that someone would create a birth, and someone will again. Even if the technology doesn’t move forward in China, it will just happen somewhere else.

The spark
I had traveled to China in October, a full month before the announcement of the CRISPR babies, to understand the country’s intentions for embryo engineering, an area in which its scientists had taken a notable lead: of the 10 papers I could locate describing lab-edited embryos, eight were from China, and one each from the US and the UK. Whatever was happening, China would be the place to learn about it. If there was a secret project to make a baby, perhaps I could unearth it. I began with Huang, who had dropped off the media radar entirely in 2015.

Huang told me that our interview, in a tea house in Guangzhou, was the first he’d given since his paper was published three and a half years ago. Even now, he did not want to recall the international censure his first embryo-editing report had generated, nor the towering pile of inquiries that flowed into his in-box. “I don’t remember,” he said.

Huang was ready to speak because passions had cooled and embryo editing had become an accepted, if limited, line of research. (When US scientists edited embryos in 2017, it was heralded as a “breakthrough.”) Many had begun to see the procedure as a potential new way to prune risks of genetic disease from tomorrow’s children. Huang slowly relaxed his guard. He told me he was married and played volleyball. The means to edit embryos and alter heredity? “A necessity of history,” he said.

Yet Huang appeared to have no inkling of the news He would soon reveal. When I asked him what advice he would have for anyone in a rush to start a clinical trial in an IVF clinic, he said he thought it was unlikely anyone was trying. “We are far from the proper timing for this,” he said.

Huang’s life story would be familiar to many Chinese scientists—“just normal,” he said. He grew up on a farm, but his family moved to a town so his parents could work in a factory building boat parts and send him to school, where he was singled out for his high grades.

Even as a boy, he says, he found the embryo a source of fascination. He mated purple, white, and green corn to make hybrids, and during the 1990s he followed popular reports of a Chinese effort to clone a panda. “I thought the embryo was a very mysterious type of cell,” he says. “It has all the information needed to form something, but how does the process work?”

So Huang noticed when, in 2012, US, European, and South Korean scientists developed a versatile new way to alter the DNA information inside living cells. Called by its acronym, CRISPR allows scientists to easily cut open the double helix at any location so they can add or remove genetic instructions. A few hundred dollars’ worth of supplies and chemicals is all that’s needed.

By May 2013, a team at MIT, led by Rudolf Jaenisch, had injected CRISPR into mouse embryos, leading to the birth of the first CRISPR-modified mammals. But China, with its loose rules on animal research and its ambitions to become a worldwide leader in the technology, rapidly claimed the rest of the animal zoo. “We lost the mouse. But we won the sheep, the goat, and the monkey,” says Huang Xingxu (no relation), a professor of biology at Shanghai Tech University.

The birth of two edited monkeys in Yunnan province, announced in January 2014, led some observers to realize that edited humans might be next. But who would take such a step? What kind of social understanding or global agreement might be needed? There was none.

Junjiu Huang says he first introduced CRISPR into human embryos just three months after the monkey report. It was, as some feared, very simple to do. “We spent about half a year to finish the project, because it’s not a very complex experiment,” he says.

Huang was well positioned to do it. Guangzhou has large, well-established IVF clinics interested in research. And Huang also perceived the need for a new form of treatment. About 10% of the population in the growing city of 13 million carries a genetic error that creates a risk for beta thalassemia, a blood disease. What if CRISPR could be used to replace the broken gene with a working copy in embryos? That, he imagined, would be a “new technique” to eradicate the disease in newborn children. In his tests, Huang worked with abnormal embryos rejected by the IVF lab.

“I was doing basic research … to test the feasibility,” he says. He admits he had “no idea what the reaction would be.”

“Irresponsible” for now
It wouldn’t take long to find out. Huang’s effort to install a normal beta thalassemia gene in embryos did work sometimes, but there were serious problems. CRISPR is error prone and can make unwanted edits, called “off targets.” It meant the chance of introducing new and potentially harmful mutations. Also, the process wasn’t efficient. Often the embryo ended up with a mixture of corrected and uncorrected cells, a so-called ­mosaic—a problem that was to end up afflicting most of He’s embryos as well.

To many scientists, the risk of unwanted and undetected errors is what makes it so unwise to create a CRISPR baby. “It’s clearly not ready,” says Zheng-Yi Chen, a Harvard University scientist who works with CRISPR in pigs in China. “You don’t know the consequences to develop a whole human being. Any subtle difference could be magnified by a ­billion- or trillion-fold. It could change the landscape.”

Huang’s lab results alarmed top Western biologists who reviewed them starting in late 2014. They claimed the work was sloppy and made sure his submission was rejected by both Science and Nature, the world’s premier science journals. But in truth, the experts were shocked by how advanced the Chinese work was. Before Huang could publish his report elsewhere, American biotech executives who had seen his text called for an immediate moratorium on all embryo editing. Their editorial, published by Nature, was titled “Don’t edit the human germline.”

It would be the first, and last, organized call from gene-editing experts to shut down the lab research. Days later, a broader group of specialists, writing in Science and including Jennifer Doudna, a co-discoverer of CRISPR, took the position that lab studies should be encouraged but called for an urgent international meeting to “explore responsible uses of this technology.” That meeting, eventually held at the National Academy of Sciences in Washington, DC, in December 2015, drew top biologists and ethicists from around the world, including China.

Biologists have struggled to understand their own power and to contain the risk that governments could step in with regulatory restrictions on CRISPR. Seeing how quickly the technology was moving, David Baltimore, the former president of Caltech, speaking for the conference organizers, hit the pause button: he stated that making a baby would be “irresponsible” for now. Such an undertaking needed to wait until the technology was better studied and until there was a “broad societal consensus” about why we’d want to change the gene pool at all.

That summit statement after the meeting in Washington was agreed to by the scientific academies of the US, the United Kingdom, and China—the last of these an arm of the central government. Huang, the junior researcher who started it all, wasn’t at the historic forum. “I was not invited,” he says.

When “no” means “maybe”
A year later, in February 2017, the US National Academy of Sciences published a detailed set of recommendations written by a cadre of mostly American senior scientists. It found that no country was yet in a position to safely create a human whose genes were altered with CRISPR. But the technology was not in itself impermissible, the scientists said. So long as such a project was aimed at preventing serious disease, was preceded by safety studies, and met other—somewhat undefined—conditions, it might be acceptable to try for a live birth.

In Shenzhen, He Jiankui was listening. A biophysicist and expert in DNA sequencing, he had studied at Rice and Stanford but had returned to China. He had a professorship, funding, and, it appears, the ambition of being the first in the world to produce a child genetically engineered with CRISPR. That March, according to documents, he came to an ethics committee at a South China hospital with a proposal for a clinical trial of a treatment intended to make children immune to HIV. He believed he could win the Nobel Prize. In those documents, He cited the US academy’s report, telling others the Americans had “approved” the idea of germline editing—which, in some sense, they had.

In Shenzhen, He quickly began assembling the data that would let him meet the academy’s standard—or something resembling it. He focused on two genes whose deletion from a person’s genome can have a health benefit. One was CCR5, without which people can’t usually get HIV. Deleting the other, PCSK9, leads to extraordinarily low levels of “bad” cholesterol and a much reduced chance of heart disease.

The ideas were ambitious—closer to an enhancement than a cure, since they’d prevent diseases in the future rather than correcting a DNA defect in the embryo. But they also met one of the American report’s criteria—that CRISPR not be used to modify children if there were “reasonable” alternatives.

In most cases where parents carry a risk gene for an inherited disease, like cystic fibrosis, they will pass it on to only half their children—so the genetic error can be weeded out during IVF procedures by testing embryos and picking those that didn’t inherit the faulty gene. Many scientists believe gene-editing embryos will never be necessary, for that reason. Only CRISPR, though, can endow a child with a trait—such as HIV immunity—that the parents didn’t have. That’s why He considered his approach to be fair game.

He’s students began working on what turned into tests on more than 300 human embryos, plus countless mouse and monkey cells. The work was enabled by the Western CRISPR industry that had grown up to distribute the technology’s components and, often, collect profits. For instance, in 2016 He had e-mailed the Broad Institute, in Cambridge, Massachusetts, seeking a license to its important CRISPR patent and rights to use the gene-editing tool in human beings. Broad declined, since it had already sold rights for human therapeutics to its own spinout company, Editas Medicine. Broad did, however, sell He a license to market CRISPR’s key molecular ingredients. (An official with Broad noted that all its legal agreements prohibit “any human germline modification” and that making babies is “a clear and flagrant violation.” )

The He team also reached out to the scientific community on issues of science and ethics. Kiran Musunuru, a gene-editing scientist at the Perelman School of Medicine in Pennsylvania, recalls getting peppered with questions from a graduate student in He’s lab: “Do you think these are reasonable and feasible?” wondered Feifei Cheng about some mouse experiments. The Chinese scientists didn’t exactly hide their purpose, either. In one e-mail, Cheng said, “I think our research will illustrate whether … genome editing in embryos, not in adult, is efficient and safe for the first time.”

The focus on safety was the giveaway clue. Safe or not safe doesn’t matter for a research embryo; they are all destroyed after a few days of growth in a laboratory. Safety would only matter if you intended to create a pregnancy.

He Jiankui even took part in meetings and ethics symposia meant to determine whether babies should be made. “He spoke, but he didn’t seem to listen,” says Stuart Newman, a biologist at New York Medical College, who attended a January 2017 workshop at Berkeley organized by Doudna, where He was among those present. Doudna had always worried about how CRISPR could be misused and told a gathering of journalists that August that the sudden announcement of a CRISPR baby could be a “worst-case scenario.” Now, He was headed back to China to do exactly that.

He was not dissuaded even by his own data. After he gene-zapped test embryos, his detailed DNA-sequencing studies of the outcome convinced him that the problem of unwanted, off-target edits was minimal and he would be able to see any that did occur. But he was not so easily able to control mosaicism—a result of the fact that CRISPR didn’t always edit every cell in the embryo, or ended up editing cells in different ways. His data on test embryos found that the majority were mosaics, according to a 2017 presentation.

Photograph of He
He: CRISPR babies are here.
ASSOCIATED PRESS

Gene superpower
Chinese scientists are working in an atmosphere of loose regulations and great ambitions. China’s government wants to lead the world in biotechnology. At an institute in Shanghai, I heard a speech by Zhou Qi, a prominent stem-cell scientist who was among a group that met last March in Beijing to map out a new government strategy for achieving that goal. “China,” he said in the speech, “will put biotechnology as a very high, very important priority.”

To achieve that, China has paid special attention to both gene editing and stem-cell research—the corner of biology dealing with the Promethean capacity of certain cells (including the fertilized egg) to form hearts, lungs, and any other body part. In October, Zhou led a team that—through a complex series of steps involving stem cells, gene editing, and cloning—had shown that two male mice could have offspring together. And I heard him tick off a list of unpublished results that sounded scientifically important and, I thought, were also all likely to cause sensational headlines.

In the Chinese-language literature, I found scattered boasting and calls to move forward quickly. “In the area of human germline editing technology, we have already stepped to the forefront,” researcher Liu Jian-Qiao wrote a few months ago. Liu said that research should be “bounded” by international norms established in 2015, but that China should also influence those standards: “We should … strive for more of a right to have our voices heard, and for greater authority to take initiative in the area of clinical applications research.”

In China, there wasn’t the same reluctance as in the US to think about the technology’s benefits. The embryo-editing teams I interviewed were all clearly preparing the technology for eventual use in humans. Some of them, for instance, were attempting to locate cases where—in line with the US National Academy of Sciences’ criteria for using CRISPR—germline gene editing might be the only answer. One such scientist was Fan Yong, who like Huang is based in Guangzhou. “We are currently mainly selecting diseases that can only be cured by using embryonic gene editing treatments,” he wrote in an e-mail. One candidate group Fan was exploring is deaf men and women who marry, a common occurrence in a country as large as China. If both parents’ hearing loss has the same genetic cause, it can mean they can’t have a hearing child. Fan told me he thinks correcting deafness in embryos “is a natural choice for public health in China.”

Still, none of the Chinese scientists I spoke to said they thought a baby was coming anytime soon. They noted that under a 2003 government guideline, no IVF clinic was supposed to take a genetically modified embryo and start a pregnancy. “From the perspective of technology as well as how society would accept it, I don’t think we’re at that point yet,” Huang told me when I visited with him. Maybe one day, he said, when the technology was more advanced.

Too late
In Shenzhen, He wasn’t waiting.

It appears that by about February 2018, He’s team had transferred edited embryos into the uterus of a woman (who remains unidentified). They continued to monitor the pregnancy, taking blood draws and peering at the twins’ genomes that way. It remains unclear what authorities—if any—in China signed off on the trial. He’s own university now claims it knew nothing about the study, which was carried out quietly if not secretly.

He may have planned to make his big reveal of the CRISPR babies in November during the Second International Summit on Human Genome Editing in Hong Kong, a meeting whose purpose was to debate the prospect of making such babies, and where he was among some 70 scheduled speakers. “I suspect he was planning to pull a Steve Jobs–style ‘one last thing’ during his talk,” says Musunuru, referring to the Apple founder’s trick of saving the biggest news for last. It would be the ultimate fait accompli—a “bombshell” set off before the world. Instead, reports of He’s undertaking leaked just before the meeting, and the scientist rushed to post a series of recorded statements to YouTube. The twins were named Lulu and Nana. “We hope you have mercy for them,” He said. “I believe families need this technology.”

Oversight? It’s unclear if there was much. A daring cure? Not really. HIV can be treated or prevented with other, cheaper methods and still affects less than 0.1% of the population in China. As David Liu, a Harvard biologist, asked He in Hong Kong: “What is the unmet medical need?” The question cut to the quick—in medicine, risky interventions for healthy people are unacceptable. The embryos He chose to edit were normal.

“When I realized what had happened, I screamed—literally,” says Musunuru, who has seen the genetic readouts on the embryos that had turned into the twin girls. There had been “a mess of different edits”—that is, cells were changed in different ways. The girls may prove to be made up of a mosaic of differently edited cells as well. “The first gene­-edited baby was going to be a disturbing event, no matter what,” he says. “The fact that it happened this way, with flawed embryos, by researchers who were either clueless about the problem or simply couldn’t care less, made it 100 times worse.”

The criticisms after He announced the births were unrelenting. “Beyond belief,” said Feng Zhang, of the Broad Institute, another of CRISPR’s co-inventors. Huang weighed in with a statement that what his counterpart had done was “against the law, regulation, and medical ethics of China.”

“The first gene­-edited baby was going to be a disturbing event, no matter what. The fact that it happened this way, with flawed embryos, by researchers who were either clueless about the problem or simply couldn’t care less, made it 100 times worse.”

Some scientists, it turns out, were in the know about He’s work. One can only guess why they remained silent. But we know how the He affair ended in Hong Kong. George Daley, the dean of Harvard Medical School and the same person who implied a few years back that Huang wasn’t right in the head, took the stage and, though he called He’s work a “misstep,” did not condemn it. Instead, Daley spoke in favor of using CRISPR in IVF clinics in the future, saying it was time to move past the question of “ethical permissibility” and on to the question of how to do it correctly.

In their final statement, summit leaders gave no mention to a criterion they had once agreed was paramount before moving into clinical applications: “broad societal consensus.” There is no consensus yet about whether we should engineer human beings—and there may never be. But He’s claim helped make the question moot. We already have.

It’s unclear what will happen to He. His clinical trial was halted, and the government seems about equally likely to block any further babies as to endorse them. He left the Hong Kong conference early, saying he would “remain in China, my home country, and cooperate fully with all inquiries about my work.”

Huang, meanwhile, went back to Guangzhou and is continuing his research. In our interview, which occurred before the announcement of the CRISPR babies, I had asked him whether he would do anything differently if he could go back in time. Would he put CRISPR in a human embryo? “I would do it again,” he said. “The development of science and technology is unavoidable.”

Deep Dive

Biotechnology and health

Scientists are finding signals of long covid in blood. They could lead to new treatments.

Faults in a certain part of the immune system might be at the root of some long covid cases, new research suggests.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

The next generation of mRNA vaccines is on its way

Adding a photocopier gene to mRNA vaccines could make them last longer and curb side effects.

Ready, set, grow: These are the biotech plants you can buy now

For $73, I bought genetically modified tomato seeds and a glowing petunia.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.