Editing Humanity: The CRISPR Revolution and the New Era of Genome Editing

CHAPTER 16 BREAKING THE GLASS

In March 2016, Michael Deem visited Shenzhen to speak at a small symposium on “Biodynamical Systems.” In the official group photo, Deem stands tall in the center of the group. His smiling protegé, He Jiankui, is easily recognizable. Deem’s lecture was on CRISPR, the same topic on which JK and Deem had once published together. How far CRISPR had progressed since then: it was now a universal genome editing tool, spawning several biotech companies hoping to cure cancer and genetic disorders.

Six months later, SUSTech announced that Deem, listed as a member of the university’s Department of Biology, and He Jiankui had won a $5.7 million grant as part of Shenzhen’s Peacock Plan for further development of their sequencing system. The thrill of building a company like Direct Genomics with the prize of cracking China’s clinical market would have satisfied the ambitions of most biotech entrepreneurs. But although he did not have any hands-on experience using CRISPR, JK was contemplating an even more daring venture.

JK first discussed his idea to edit human embryos privately with Quake, his former Stanford supervisor, in 2016. The response wasn’t what he was expecting. “That’s a terrible idea. Why would you do that?” Quake told him.¹ JK reluctantly agreed with Quake’s advice that, at a minimum, he should obtain the appropriate ethical approval and patients’ informed consent. JK promised to do so, emailing later: “I will take your suggestion that we will get a local ethic approve [sic] before we move on to the first genetic edited human baby. Please keep it in [sic] confidential.” Quake thus became the first of a small but not insignificant “circle of trust”—scientists and ethicists in whom JK confided and sought advice. Their reactions varied upon learning of JK’s intentions, but no one broke a confidence or blew the whistle.

That summer, JK traveled to New York to attend the biggest conference on CRISPR and genome editing, held at Cold Spring Harbor Laboratory on Long Island. During one of the breaks, JK introduced himself to Doudna, one of the meeting co-organizers, and posed for a selfie with her sitting in the front row of the Grace auditorium, which he posted on WeChat.² Doudna took such requests in stride: she is about as popular as Taylor Swift, graciously allowing photographs from admiring students and scientists at every event.

In late 2016, JK reached out to Mark DeWitt, a lesser known scientist at Doudna’s Innovative Genomics Institute (IGI), who had recently published a paper on genetically editing cells for sickle-cell disease.³ After meeting for lunch, DeWitt accepted JK’s surprise invitation to visit China to give a lecture at SUSTech, but not before reiterating that Berkeley had no interest in supporting experiments on human embryos. “I’m a little reluctant to speak about it again,” DeWitt told me by phone. “It’s important but it’s not what I want to be known for.”⁴

In January 2017, JK was honored to be invited to attend a small workshop at IGI organized by Doudna and Stanford bioethicist William Hurlbut on the topic of public education and engagement around human genome editing. JK was the youngest scientist at the meeting, and other than delivering a short talk presenting preliminary data on editing in human embryos, focusing on the safety of CRISPR technology, kept a low profile. On his blog, he posted a summary of his talk, which highlighted five safety issues, including off-target effects and mosaicism, that would require further study. He signed off his report by stating that the behavior of anyone performing germline editing would be “extremely irresponsible,” unless and until these important safety issues were resolved. He also noted something else of interest: the Valentine’s Day publication of the National Academy of Sciences report on germline editing, which (as we will see later) in JK’s view gave a “yellow light” for human germline editing.⁵

The meeting began with a bang: a public lecture by George Church, provocatively titled “Future, Human, Nature: Reading, Writing, Revolution.”⁶ Church spoke openly about germline therapy, speculating that mitochondrial germline therapy would move into treatments for infertility. Compared to our ancestors, “we are already augmented,” Church said, mostly through physics and chemistry rather than genetics. He noted humanity’s ability to see, to hear, to fly, to reach altitudes, to plumb the ocean’s depths, further than our ancestors could have dreamed. But geneticists would get their chance: near the end of his talk, Church displayed a list of “protective alleles”—variants in certain genes that offer beneficial traits with no known or serious downsides. Among those genes was CCR5. “Wow, that was wonderful,” said Doudna, Church’s host for the evening, probably echoing JK’s sentiments in the audience.

During the meeting, JK also met William Hurlbut and his son Ben, a science historian at Arizona State University. Looking back, Ben Hurlbut concluded that JK’s motivations were familiar and mundane,

driven by the high-octane milieu of contemporary biotechnology, both in the United States and in China. He internalized ideas that led him to believe that his experiment would elevate his status in the international scientific community, advance his country in the race for scientific and technological dominance, and drive scientific progress forward against the headwinds of ethical conservatism and public fear.⁷

Two years later, JK shared with Hurlbut that one particular statement at the IGI conference really affected him. “Many major breakthroughs are driven by one or a couple of scientists… cowboy science,” JK said. “You need a person to break the glass.”⁸

JK was following progress in the arena of CRISPR gene editing. China was racing ahead in offering CRISPR-based somatic therapies to dozens of cancer patients “unhampered by rules,” although it was unclear how successful these early efforts were.⁹

But JK was interested in going much further, to “break the glass”: engineering a permanent, heritable change to human DNA as a means to treat or prevent a disease not only in a patient but also in their children forevermore. In the three years following the groundbreaking human embryo experiments performed in Guangzhou in 2015,¹⁰ nine more human embryo editing studies were published. All but two came from research groups in China.^{11, 12, 13, 14, 15, 16, 17} In March 2017, a group at Guangzhou Medical School reported editing a pair of disease genes in human embryos. Although none of the studies had been designed to actually implant an edited embryo, there was a feeling that it was only a matter of time. When Nature Biotechnology asked leaders in the genomics field whether germline editing was inevitable, including Doudna, Dana Carroll, Hank Greely, Robin Lovell-Badge, and Craig Venter, they concurred it was.¹⁸

In contrast to the Chinese reports on gene editing in human embryos, two other studies—one in Europe, the other in the United States—were given top billing in Nature. One was led by Kathy Niakan at the Francis Crick Institute in central London, the $1 billion crown jewel of British molecular biology research. The institute was officially opened by the Queen in 2016, who unveiled a portrait of Crick that had been commissioned by Jim Watson. Niakan is a widely respected developmental biologist who studies the fundamental mechanisms by which embryos develop in the critical first few days and weeks after fertilization. Understanding how specific genes orchestrate cell growth and development could help identify risk factors for miscarriages and other complications of pregnancy, and improve the success rate of IVF.

In 2016, the UK HFEA granted Niakan limited permission to use CRISPR-Cas9 to engineer DNA alterations in human embryos.¹⁹ She used spare embryos created during IVF and donated for research, growing them for no more than fourteen days. “I promise you she has no intention of the embryos ever being put back into a woman for development,” insisted her colleague, Lovell-Badge. In September 2017, using CRISPR, Niakan’s team reported in Nature a key role for a gene called OCT4 in human embryonic development.²⁰

A bigger splash came just a month earlier. ONE GIANT STEP FOR DESIGNER BABIES was the exclusive front-page headline in a British tabloid. The scoop, from science journalist Steve Connor, also ran in MIT Technology Review under a more sober title: “First human embryos edited in U.S.”²¹ Connor had heard the news several weeks before Nature was due to publish a report from a leading American IVF center. The journal editors were “incandescent that their thunder had been stolen,” which is music to any science reporter’s ears.^I

The study’s lead author was Shoukhrat Mitalipov of the privately funded Oregon Health & Science University in Portland, so was not subject to the federal funding prohibition on human embryo research. Originally from Kazakhstan, Mitalipov trained in Moscow before emigrating to the United States in the mid-1990s. An expert in stem cell and embryo research, his group reported the first cloned monkeys and the first individualized stem cells (by genetically reprogramming a patient’s skin cells). His work in mitochondrial replacement therapy earned him recognition as one of Nature’s top ten personalities of 2013.

So why did Nature, despite the many Chinese human embryo precedents, elect to publish Mitalipov’s paper? It surely wasn’t just because the controversial field of human embryo editing had been carried out on American soil.²² Mitalipov’s work stood out for two reasons: first, the team’s results were technically impressive. Mitalipov’s group had created dozens of embryos by fertilizing donated eggs with sperm from a man with hypertrophic cardiomyopathy. (This disease can cause sudden, fatal heart attacks, notably in young athletes such as former Boston Celtics star Reggie Lewis.) The Oregon team used CRISPR to repair the faulty copy of the MYBPC3 gene, which had a four-base deletion. Of the fifty-eight treated embryos, forty-two were apparently repaired with two healthy copies of MYBPC3, without any serious off-target effects.

The other novel finding was the manner in which the mutant gene had been corrected. Mitalipov had expected that if gene editing occurred, the CRISPR machinery would cut the MYBPC3 gene and paste the guide sequence that was co-injected into the embryos. But instead, the repaired sequence appeared to emanate from the corresponding gene on the partner chromosome. This suggested a novel DNA repair mechanism—a gene replacement rather than an edit. Consequently, Mitalipov didn’t use the word “editing” in the title of his report but “correction.” “Everyone always talks about gene editing,” Mitalipov said. “I don’t like the word editing. We didn’t edit or modify anything. All we did was un-modify a mutant gene using the existing wild-type maternal gene.”²³ Others weren’t so sure. Embryologist Tony Perry argued that the genomes of egg and sperm were too far apart for this mechanism to work. “The genomes are separated by what are, in cellular terms, intergalactic distances,” he said.²⁴

But most observers judged the work to be an important step forward. “It feels a bit like a ‘one small step for (hu)mans, one giant leap for (hu)mankind’ moment,” said Doudna.²⁵ And she predicted correctly that Mitalipov’s results would be “encouraging to those who hope to use human embryo editing for either research or eventual clinical purposes.”^II Asked what was the biggest challenge in his research, Mitalipov said it was nothing technical but rather the regulatory approval required from three different committees vetting the research, which could be a forerunner to clinical trials in the future. He matter-of-factly said that dominantly inherited disorders would be prime targets. This would include cancer predisposition genes such as BRCA1 and BRCA2, which when mutated increase the odds of breast and/or ovarian cancer.

Mitalipov did not revel in the spotlight, appearing uncomfortable during a live interview on the Charlie Rose television show. But in his “giant leap for (hu)mankind,” Mitalipov felt his research offered hope for millions of people suffering from or carrying serious genetic diseases who worried about transmitting those genes to their children. “We now have a chance to prevent disease at the earliest possible stage of life,” he said.²⁶ But given the backlash that had greeted the first human embryo tinkering in 2015, who would dare take the next step?

In Shenzhen, JK read the headlines and hype around Mitalipov’s study with incredulity bordering on disgust. Two years earlier, American scientists had rushed to condemn China’s initial exploration of human embryo editing, which had used nonviable embryos, but here they were basking in the media spotlight with an article in Nature, no less, using viable embryos, without anything like the same criticism. JK charged this was a double standard, nothing less than “scientific racism against the Chinese.”²⁷ Feeling a sense of patriotic duty, JK was emboldened to push on.

In mid-2017, JK started talking to Lombardi about the idea of launching a gene-editing company. He wanted to hire a CEO for Direct Genomics so he could free up time to focus on the science of gene editing. He asked Lombardi for help to attract funding so that his “NewCo” would be an international company. “What do you want to do?” Lombardi asked. “We’re going to gene edit humans,” JK said. Lombardi was almost speechless. “You realize there’s a moratorium, right?”²⁸

JK didn’t share Lombardi’s concern. The 2017 National Academies report on germline editing had surprised many observers in suggesting that germline editing could be permitted under certain conditions. JK sent Lombardi a rudimentary business plan, focusing on addressing the HIV health crisis in Southeast Asia. According to the plan, China was setting up a district that would do medical tourism around gene editing. “My stuff will be part of this,” JK told Lombardi. For the next twelve months, Lombardi scheduled exploratory meetings, with the intent of letting JK discuss his commercial aspirations with potential investors. Each time, JK canceled his trip at short notice.

One reason was that JK was starting to enroll couples in his clinical trial. Working with a Beijing-based organization that helps Chinese people with HIV, JK recruited eight couples where the father was HIV-positive. One couple subsequently dropped out. On a Saturday in June 2017, JK met two suitable couples interested in volunteering for his trial. For just under an hour, JK walked the couples through the informed consent process, presenting a document that described the risks of the procedures. The study was billed as an “AIDS vaccine development project.” JK was the project leader, while funding was provided by SUSTech. The project would knock out the CCR5 gene “to help these CCR5 gene-editing [sic] babies to obtain the genotype of the Northern European to naturally immunize against HIV-1 virus.” Each couple would receive about $40,000 for medical costs for the IVF procedure, genome editing, and hospital visits.²⁹

While the document outlined the risks of genome editing, JK explained his team was not responsible for any off-target effects, which were “beyond the consequence of the existing medical science and technology.” And only JK’s team had the right to make any public announcement about the trial. Sitting at the end of the boardroom table watching JK invite these couples to make history were two notable observers.³⁰ One was Michael Deem. The other, casually dressed in a blue T-shirt as if he’d just dropped in after a 5K run, was Yu Jun, a very prominent Chinese scientist, one of the four cofounders of BGI. Yu had sold his share of the organization, and was based in Beijing where he is on the faculty of the Chinese Academy of Sciences (CAS)—China’s equivalent of the NIH.³¹ Moreover, he is the chief editor of the official genomics journal of the Genetics Society of China. It was a glaring illustration of the high-profile friendships that JK had forged during his short time back in China.

Two months later, JK returned to the Cold Spring Harbor Laboratory, this time as an invited speaker. He stood at the podium where countless legends of molecular biology have presented, under the gaze of a large portrait of Jim Watson.^III JK talked about the improving efficiency of gene editing, measuring off-target mutations by genome sequencing in monkey and human embryos. There was no indication that JK was about to rush into clinical germline editing, at least not until his penultimate slide—a New York Times magazine headline marking the death of Jesse Gelsinger. Human germline editing could occur “in the near future,” JK said, but “we should do this slow and with caution. A single case of failure may kill the entire field, just like this case—‘the biotech death of Jesse Gelsinger’ for gene therapy.”³²

JK’s final slide—the traditional acknowledgements to collaborators and funding agencies—thanked Deem, DeWitt, and William Hurlbut. He answered a few technical questions from the audience, but no one packed into the auditorium seriously thought the young Chinese scientist was about to break from the international consensus on germline editing. Many (like me) were just hoping the evening session would wrap up on time so they could head down to the bar for a well-earned beer.

Three months later, during an advisory board meeting for Direct Genomics, JK welcomed his latest prize recruit, Nobel laureate Craig Mello, to Shenzhen. JK asked Mello about the possibility of using CRISPR to prevent HIV transmission in human embryos, but Mello didn’t think JK was really serious. Meanwhile, one of JK’s PhD students, Feifei Cheng, sent a series of emails to Penn cardiologist Kiran Musunuru, for advice about using CRISPR to target another gene, PCSK9: individuals with rare mutations in PCSK9 exhibit dramatically lower cholesterol levels. Musunuru replied to her first email but ignored the others. He forgot about the exchange completely until twelve months later, when Marchione emailed him JK’s draft paper.³³

Back in China, JK’s profile and reputation were growing fast. The Central Government selected JK for its top multidisciplinary science program, Qianren Jihua (the Thousand Talents Plan), self-described as one of the world’s most prestigious science awards. CCTV-13, the biggest news channel on Chinese state television, aired a fawning four-minute video showcasing JK based on a Chinese idiom, “The waves behind drive on those before.” JK is shown joking with colleagues, and proudly displaying an encyclopedia of books he had published containing a complete human genome sequence, dozens of volumes stacked in a double helix tower configuration. There were also scenes of him enthusiastically playing five-a-side soccer as his wife and baby look on.³⁴

Early in 2018, JK was back in San Francisco. He told DeWitt over dinner that he had received approval from his institutional review board (IRB). He also met Stanford’s Matthew Porteus, revealing in confidence that he was starting work on implanting genetically edited human embryos. A stunned Porteus berated JK, telling him it was a terrible idea. “I told him he was putting the entire field at risk through his reckless actions,” Porteus recalled, and he needed to talk to more experts.³⁵ JK sat poker-faced; he had expected a more supportive reaction. In retrospect, Porteus says he wishes he’d raised the alarm more publicly. Between a trio of Stanford faculty, Deem, Mello, DeWitt, Ben Hurlbut, and presumably others, JK took a significant number of American scientists into what Ferrell called his “circle of trust.”³⁶ He was popping up everywhere like Zelig, Regalado said.

JK visited William Hurlbut several times throughout 2017 and 2018, enjoying some lengthy and intense ethics discussions. JK wanted to understand the objections to heritable genome editing in the West. On one occasion, JK brought his thumb and forefinger together and asked Hurlbut if something that small could be as important as a baby? Hurlbut thought JK was “humble and well-meaning,” trying to advance science and help others. But Hurlbut didn’t suspect that JK was seriously contemplating clinical work. “He’s an idealist,” Hurlbut said. “He’s an inexperienced, perhaps naïve, optimist. I kind of knew [He] was involved in something of significance. But it’s unfortunate that it had to happen this way. Sad, really, because he seems like a guy with good intentions.”³⁷

Hurlbut’s sympathetic assessment was not widely shared.

In February 2018, JK took unpaid leave from his university. He was a man on a mission. He wanted to make his mark on the world stage, and to help his fellow citizens, especially those at risk of HIV. And he wanted to deliver a victory for Chinese science. He later told Ben Hurlbut that he had no interest in waiting for an international societal consensus that in his view was never going to happen. “But once one or a couple of scientists make [the] first kid, [it’s] safe, healthy, then the entire society including science, ethics, law, will be accelerated… So, I break the glass.”

Following IVF, JK’s team implanted thirteen embryos into five of the seven women enrolled in the trial. Two women became pregnant. In April 2018, JK had some exciting news to share with a few of his American confidants. He sent virtually identical emails to Mello, Quake, and DeWitt, with an emphatic subject line: “Success!”

Good News! The women [sic] is pregnant, the genome editing success! The embryo with CCR5 gene edited was transplanted to the women [sic] 10 days ago, and today the pregnancy is confirmed.³⁸

Quake decided to share JK’s email with a colleague but didn’t receive any concrete advice. Mello was also put in an awkward position. “I’m glad for you,” he replied, “but I’d rather not be kept in the loop on this. You are risking the health of the child you are editing… I just don’t see why you are doing this. I wish your patient the best of luck for a healthy pregnancy.”³⁹ Mello’s disapproval was tempered by his respect for JK. “I know you mean well,” he wrote. Meanwhile, DeWitt was in shock and didn’t know what he should do.

In August 2018, JK made another quick trip to the United States. In Boston, JK finally got to meet Feng Zhang, seeking advice on methods to reduce off-target effects in mouse and human embryos. “He was having the same challenges as other researchers around lack of efficiency and lack of precision,” Zhang told a reporter in Hong Kong.⁴⁰ “I told him that the technology is neither efficient nor precise enough for real-world application in embryos, including in human IVF applications.” From there, JK traveled to New York, where he visited Chengzu Long, a Chinese CRISPR scientist at New York University. He made quite the impression. “It was hilarious,” the fast-talking Long recalled. “He had a male personal assistant open the car door for him and carry his briefcase. Very Chinese!”⁴¹

JK also traveled to Pennsylvania to meet executives of Geisinger, a leading healthcare company. The two-hour meeting on August 16 was hosted by two distinguished geneticists, Hunt Willard and David Ledbetter. JK’s team was interested in starting a project called “Gene Achieve” in China, based on Geisinger’s MyCode precision medicine biobank, in which volunteers opt in to receive genetic and other medical results. Willard said JK discussed his CRISPR research without mentioning his embryo work. “It wasn’t obvious to me how [or] why Geisinger would want to get involved in any collaborations with either Direct Genomics or the hospital system,” Willard told me.⁴²

Despite his reservations, Mello traveled to China to attend a Direct Genomics scientific advisory board meeting. On November 19, 2018, a smiling Mello, sitting next to JK, posed for a photograph with company staff in a boardroom. As usual, JK posted the photograph on WeChat. It was just ten days before his scheduled appearance at the Hong Kong summit. But JK was keeping a massive secret.

A few weeks earlier, in late October, a visibly nervous JK had taken a flight north from Shenzhen to an undisclosed destination. Ferrell noticed there was a distinct look of relief upon his return. He carried the news that, by emergency caesarean section, not one but two CRISPR babies had been born.

I. Sadly it was Connor’s last major scoop; he died of prostate cancer a few months later, aged sixty-two.

II. If Mitalipov’s observations on the DNA repair mechanism hold up, then investigators will have more work in cases where both copies of the gene are mutated. They will have to figure out a way to coax the embryo to accept the externally provided repair patch.

III. The Watson portrait was removed 18 months later after Watson uttered racist comments in a PBS documentary. Cold Spring Harbor Laboratory subsequently cut all ties with its former president and chancellor, removing his name from the graduate school.