In late November 2018, a Chinese biophysicist named He Jiankui announced that twin girls had been born from embryos he had edited using CRISPR, a tool that allows scientists to cut and alter DNA at precise locations. The girls, referred to by the pseudonyms Lulu and Nana, were the first known humans to be born with deliberately modified genetic material that could be passed to their own children. The announcement stunned the scientific world, not because genome editing was new, but because He had crossed a line that the global research community had broadly agreed should not yet be crossed, and had done so with troubling technical and ethical shortcomings that put the children at risk.
What He Jiankui Actually Did
He Jiankui’s stated goal was to make the twin girls resistant to HIV. He targeted a gene called CCR5, which produces a protein on the surface of immune cells that HIV uses as a doorway to infect them. A naturally occurring mutation of this gene, known as CCR5-delta32, is found in a small percentage of people of European descent. People who carry two copies of this mutation lack functional CCR5 receptors, making them highly resistant to the most common strains of HIV.
He used CRISPR to try to disable CCR5 in embryos created through in vitro fertilization. The fathers in the study were HIV-positive, and the mothers were HIV-negative. He implanted the edited embryos, and the twins were born sometime in October 2018. A third baby, from a separate couple, was reportedly born later.
The revelation came on November 25, 2018, just before a scheduled presentation at the Second International Summit on Human Genome Editing in Hong Kong. The news broke through media reports and a YouTube video He had posted, rather than through peer-reviewed publication, which itself was a departure from scientific norms.1PubMed Central. CRISPR’d babies: human germline genome editing in the ‘He Jiankui affair’
Why the Experiment Was Medically Unnecessary
One of the most damning criticisms of He’s work is that the procedure solved a problem that already had safe, effective solutions. The couples involved were HIV-discordant, meaning the father was HIV-positive and the mother was not. In such cases, the risk of transmitting HIV to the child during conception can be essentially eliminated through sperm washing, a well-established technique used in assisted reproduction.
A systematic review of the evidence found that no HIV transmission occurred across more than 11,500 cycles of assisted reproduction using washed semen, involving nearly 4,000 women. Even among men who did not have their viral load fully suppressed at the time of the procedure, no seroconversions occurred in over a thousand women. And no cases of mother-to-child transmission were reported either. Roughly 56% of couples in those studies achieved a clinical pregnancy, making sperm washing not only safe but reasonably effective as a fertility treatment.2PubMed Central. Effectiveness of semen washing to prevent human immunodeficiency virus (HIV) transmission and assist pregnancy in HIV-discordant couples: a systematic review and meta-analysis
Beyond sperm washing, the fathers in He’s study were already on antiretroviral therapy, which suppresses viral load to undetectable levels and dramatically reduces transmission risk on its own. There was no clinical scenario in which editing embryonic DNA was the only or even the best option for protecting the children from HIV. The experiment exposed the babies to unknown risks in exchange for a benefit they could have received through routine medical care.
The Editing Did Not Work as Intended
Even setting aside the question of whether the experiment should have happened at all, the technical execution was deeply flawed. He Jiankui did not actually recreate the naturally protective CCR5-delta32 mutation. Instead, CRISPR introduced different, novel mutations at the CCR5 gene, mutations whose effects on the protein had never been studied. Whether these mutations confer any HIV resistance at all remains unclear.
Analysis of the embryos revealed that the editing was incomplete. One of the twins, Lulu, had only one of her two CCR5 gene copies altered, with a 15-base-pair deletion. The other copy remained completely normal. Since HIV resistance from natural CCR5-delta32 requires both copies to be nonfunctional, Lulu likely gained no meaningful protection. The other twin, Nana, appeared to have both copies disrupted, but with two different types of mutations rather than the same one on each copy.3PLoS Biology. Gene-edited babies: What went wrong and what could go wrong
Additionally, when He tested embryos for off-target edits, meaning unintended cuts at other locations in the genome, he established only a single embryonic stem cell line. Twelve out of nineteen embryos still contained entirely unedited CCR5 copies, suggesting the editing process was far from reliable. Because off-target effects can cause unpredictable damage, from disrupting tumor-suppressor genes to triggering other health problems, the incomplete screening was a serious failure of due diligence.
Ethical Violations That Went Beyond the Science
The ethical problems ran deeper than the science itself. He Jiankui operated largely outside established oversight structures. He did not publish his work through normal channels before proceeding to implantation, and the informed consent process was grossly inadequate. The consent documents given to the parents failed to disclose He’s financial conflicts of interest and did not adequately explain the risks of germline editing, including the fact that no one had ever done this in humans and that long-term consequences were entirely unknown.4Nature Biotechnology. Ten ways in which He Jiankui violated ethics
Recruiting HIV-positive men for this kind of study raised its own ethical red flags. People living with HIV in China face substantial social stigma, which can create a power imbalance between researchers and participants. Couples desperate to have children without the stigma of HIV might agree to experimental procedures they would not otherwise accept, particularly if the full scope of the risks is obscured. Ethical guidelines on clinical trial recruitment exist specifically to guard against this kind of coercion, and He’s procedures violated those guidelines significantly.
There is also the question of the children themselves, who could not consent to having their genomes permanently altered. Unlike somatic gene therapy, which modifies only the patient’s own cells and does not pass to the next generation, germline editing changes DNA in every cell of the body, including eggs and sperm. Any mutations introduced, intended or accidental, will be inherited by the children’s own future offspring. The children were enrolled in an experiment whose consequences will unfold over generations.
The Hidden Risks of Disabling CCR5
Even if the editing had been technically flawless and had perfectly replicated the CCR5-delta32 mutation, removing CCR5 function is not a free trade. CCR5 is not just an HIV entry point; it plays active roles in the immune system, particularly in directing immune cells to sites of infection in the brain and elsewhere.
Research on people who naturally carry two copies of CCR5-delta32 has found a significantly increased risk of severe West Nile virus infection. In one study of confirmed symptomatic West Nile virus cases, people homozygous for the mutation were dramatically overrepresented compared to the general population, with an odds ratio of about 4.4 in an Arizona cohort and 9.1 in a Colorado cohort. Having no functional CCR5 was also linked to a much higher chance of dying from the infection.5PubMed Central. CCR5 deficiency increases risk of symptomatic West Nile virus infection Mouse studies confirmed the mechanism: without CCR5, immune cells fail to traffic properly to the brain during West Nile virus infection, leading to uncontrolled viral replication and higher mortality.6Virus Research. Beyond HIV infection: Neglected and varied impacts of CCR5 and CCR5Δ32 on viral diseases
West Nile virus is not the only concern. CCR5 has been implicated in the body’s response to other infections, including tick-borne encephalitis and influenza. Removing a protein that the immune system actively uses to fight infections in exchange for resistance to one specific virus, one that can already be prevented with existing drugs, is a trade-off that no ethical review board had approved or even fully evaluated.
A Tangled Connection to Memory and Cognition
One of the more unexpected dimensions of the CCR5 story involves the brain. Research in mice has shown that reducing or eliminating CCR5 function enhances learning and memory. Mice lacking the gene performed better on standard memory tests, including spatial navigation and fear conditioning, and showed increased synaptic plasticity, the ability of connections between neurons to strengthen with use.7PubMed Central. CCR5 is a suppressor for cortical plasticity and hippocampal learning and memory
More recent work has elaborated on the mechanism. CCR5 appears to act as a brake on neural plasticity, dampening the molecular signaling pathways that consolidate memories and remodel synapses. When that brake is removed, mice show increased dendritic spine turnover, more learning-related structural changes in the cortex, and faster experience-dependent adaptation.8Experimental Neurology. CCR5 as a key modulator in neurocognitive disorders
These findings are in mice, and it would be a mistake to leap from rodent studies to conclusions about human intelligence. But they do raise an uncomfortable question: did He Jiankui’s editing, even accidentally, alter the cognitive development of the children? Nobody knows the answer, and the fact that nobody can know the answer is itself part of the problem. The experiment created human subjects who will carry genetic changes with unpredictable downstream effects for the rest of their lives. Researchers studying CCR5’s role in cognition have been quick to point out that more plasticity is not always better. Enhanced synaptic remodeling could, under the wrong circumstances, contribute to problems like seizure susceptibility or anxiety, outcomes that have not been ruled out.
The Retracted Mortality Study
Shortly after He’s announcement, a widely covered study published in Nature Medicine claimed that people homozygous for CCR5-delta32 had a higher overall mortality rate, suggesting the mutation came with a life-expectancy cost. The study drew enormous attention because it seemed to provide direct evidence that what He had done was not only reckless but actively harmful to the children’s long-term survival.
The paper was later retracted. Researchers identified significant errors in the dataset, including problems with the UK Biobank data used to draw conclusions about mortality. The retraction did not prove that CCR5-delta32 homozygosity is harmless; it simply meant that the specific claim about higher overall mortality was not supported by the data presented.9PubMed Central. Retraction Note: CCR5-∆32 is deleterious in the homozygous state in humans The episode illustrated how charged the scientific atmosphere had become. Researchers and journalists alike were eager for a clean narrative about the consequences of He’s experiment, and the retraction was a reminder that the actual picture is more complicated and less resolved than either side wanted it to be.
China’s Regulatory Response
He Jiankui was sentenced in December 2019 to three years in prison by a Chinese court for “illegal medical practice.” He was also fined and banned from working in reproductive technology. Two collaborators received shorter sentences. The punishment was widely seen as swift by Chinese standards, but it also raised questions about how He had been allowed to proceed in the first place.
In the aftermath, China undertook substantial regulatory reform. The National People’s Congress, China’s highest legislative body, along with the Ministry of Science and Technology and the National Health Committee, moved to tighten oversight. The most prominent step was incorporating regulation of genetic research and human embryo research into the Chinese Civil Code, giving these rules the weight of foundational law rather than department-level guidelines that could be more easily sidestepped.10Medical Law International. After He Jianku: China’s biotechnology regulation reforms
Further reforms have focused on strengthening ethics governance. A national ethics committee was established to coordinate oversight, and new requirements for ethics review and external monitoring of research were introduced. Legal reforms also addressed the protection of human dignity, biosafety risk prevention, and the regulation of technological crimes, making it harder for a lone researcher to bypass institutional review.11PubMed. How should China set ethical guardrails for medical research? Whether these reforms are sufficient in practice remains an open question. The original regulations technically should have prevented He’s work, but enforcement was lax and institutional oversight was weak enough that he was able to recruit couples, edit embryos, implant them, and bring babies to term without anyone stopping him.
The Global Fallout and International Guidelines
The CRISPR babies case reverberated far beyond China. Within two years, three major international bodies issued reports on the governance of human genome editing: the National Academies of Sciences and the Royal Society published a joint report through an international commission, the World Health Organization released its own governance framework, and the European Group on Ethics weighed in as well.12PubMed Central. The impact of the three major human genome editing reports on the governance landscape All three reports grappled with the same tension: the technology is advancing faster than the rules to govern it.
The broad consensus that emerged from these reports, and from the wider scientific community, was that heritable human genome editing should not proceed to clinical use at this time. The reasoning combined scientific caution (off-target effects remain poorly understood, mosaicism is common in edited embryos, and long-term effects cannot be predicted from preclinical data) with ethical concern (the inability of future generations to consent, the risk of exacerbating social inequality, and the absence of medical conditions that would justify the risk when alternatives exist).
A strict prohibitive stance was deemed necessary not just as punishment for He’s actions but as a signal about the seriousness of the boundary he had crossed.13Medical Research Archives. The Precautionary Principle: A Public Policy Tool to Support the Application of Heritable Human Genome Editing? At the same time, there was general agreement that basic research on human embryos, using CRISPR in laboratory settings without implantation, should continue under strict oversight. The line, in other words, is not between editing and not editing. It is between editing embryos in a lab to understand biology and implanting those embryos to create people.
Germline Editing Versus Somatic Gene Therapy
The controversy around He’s experiment can obscure the fact that gene editing is being used successfully in medicine right now, just not in the way He did it. Somatic gene therapy modifies cells in a living patient, typically targeting a specific tissue like blood or the eye. Because the edits happen only in those tissues and do not affect eggs or sperm, the changes cannot be inherited. If something goes wrong, the consequences are limited to the patient, who has given informed consent as an adult.
The U.S. National Institutes of Health has invested heavily in somatic cell editing, creating a dedicated consortium and allocating roughly $190 million over six years to accelerate its development. The consortium’s mandate is strictly limited to somatic applications; germline editing is explicitly excluded, not just as an area of disinterest but as an unacceptable outcome.14Fertility and Sterility. The history, use, and challenges of therapeutic somatic cell and germline gene editing
Somatic therapies have already reached patients. Treatments using CRISPR to edit blood stem cells for sickle cell disease and beta-thalassemia have been approved in the U.S. and Europe. These are serious genetic diseases with limited treatment options, and the benefit clearly outweighs the risk in ways that He’s experiment never could claim. The distinction matters: the technology itself is not the problem. The problem is applying it to the germline, where the effects are permanent, heritable, and unpredictable over generational timescales.
What Happened to the Children
The twins, Lulu and Nana, and the third child born from He’s experiments, are now young children living in China. Almost nothing is publicly known about their health or development. Chinese authorities have stated that the children will be monitored, but details about what that monitoring entails, who is conducting it, or what if any findings have emerged have not been made public.
This silence is itself part of the ethical wreckage. The scientific community broadly agrees that the children should receive ongoing medical follow-up, particularly given the incomplete editing, the novel mutations introduced, and the unknown downstream effects of those mutations on immune function and potentially cognition. But there is no international framework for monitoring them, and their privacy is understandably a concern. They did not choose to become the subjects of the most controversial experiment in modern genetics, and they deserve to grow up without being treated as specimens. Balancing their right to medical oversight with their right to a normal life is a problem no one has fully figured out how to solve.
The case also opened a question that has no satisfying answer: if the children develop health problems that appear linked to the editing, what is the remedy? You cannot undo a germline edit. The mutations are in every cell. If CCR5 disruption increases their vulnerability to West Nile virus or another pathogen, that vulnerability is permanent. If the novel mutations He introduced turn out to have unforeseen effects on protein function, there is no way to restore the original sequence. The irreversibility is the core reason the scientific community drew the line where it did, and He stepped over it anyway.