On November 26, 2018, at a gene editing conference in Hong Kong, Dr. Jiankui He announced the delivery of twins Lulu and Nina, conceived with embryos that had been genetically modified using Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR-Cas9). Dr. He reported that he altered the gene CCR5 in those embryos. This gene produces a protein that most strains of human immunodeficiency virus (HIV) require to enter cells.1 This genetic modification was done on the embryos to protect the resulting children from contracting HIV. The children's father had HIV.
Although this experiment has not yet been presented in a peer-reviewed scientific journal, it has been extensively discussed in both lay and science-oriented media. It seems that genetic engineering of human embryos is “here” despite its limited public support. Ed Yong reported in The Atlantic2 that modifying germ cells before human implantation is illegal in 15 of 22 countries in Western Europe. In the United States, the U.S. Food and Drug Administration is prohibited from even reviewing any drug or biological product involving the modification of a human embryo.3 The scientific community has recently expressed hesitation about the long-term safety of modifying germ cells that will become babies and has recommended proceeding cautiously.4–6 The discovery of CRISPR-Cas9 has made genetic modifications easier to engineer and less expensive to perform than previously used methods such as zinc finger nucleases and transcription activator-like effector nucleases. These older methods had more complex requirements for programming them to bind to DNA at the targeted site.4,5 Further development of this molecular tool is likely to make genetic modifications of cells common for preventing and treating human diseases,3,4 for manufacturing commercial products that use live organisms,4,5,7 for exploring the intricacies of the human genome,8 and even for novel applications such as the rescue of extinct species.9 We will explain what CRISPR-Cas9 is and how it was used by Dr. He to prenatally change the genome of a baby. We will then look more deeply into the details of this experiment.
The human genome is contained in 23 chromosome pairs, primarily consisting of DNA. Each DNA molecule is a dimer containing about six billion nucleotides connected by a phosphate backbone. Only four different nucleosides occur in this very long sequence: adenine (A), guanine (G), thymine (T) and, cytosine (C). Each of these four nucleotides can form hydrogen bonds with only one other nucleotide (A with T and G with C). This is the basis for the genetic continuity of cells during mitosis or meiosis. Each strand of the DNA molecule can exactly dictate the opposite subunit by acting as a template in which each nucleotide in the sequence establishes hydrogen bonds with its unique mate and also forms a covalent bond to the proceeding nucleotide to create an elongated phosphate spine reconstituting the double-stranded molecule.
Most life contains DNA, including bacteria and many viruses. Viruses that attack bacteria, or bacteriophages, are very numerous.10 Bacteria have developed defensive weapons that attach to and destroy invading bacteriophages by cutting apart their DNA. In their study of bacteria, scientists have identified some of these weapons (referred to as restriction enzymes) and have used them routinely as tools in the study of DNA. From different species of bacteria, more than 3,000 of these bacterial weapons have been identified. Different restriction enzymes bind to different short sequences of DNA (usually 4–8 nucleotides long) and then make double-stranded cuts in the DNA near those oligonucleotides.
A more recent discovery in bacteria is a defensive tool called CRISPR-Cas9 (CRISPR stands for Clustered Regularly Interspaced Short Palindromic Repeats.) As with restriction enzymes, CRISPR-Cas9 variants can be found in many bacteria.11 CRISPR-Cas9 acts like an immune system for its bacterium.12 If a bacterial cell survives an attack by a bacteriophage, CRISPR-Cas9 retains a small piece of that virus's DNA. This retained memory of prior exposure enables that bacterium to rapidly defend itself, if it is attacked again by the same species of bacteriophage.
The CRISPR-Cas9 system (Fig. 1), adapted for genetic engineering differs from restriction enzymes in that its binding–recognition site is defined by a longer sequence of about 20 nucleotides. Because restriction enzyme binding sequences are short, they bind at many sites and make many cuts in a long DNA molecule. The binding site of CRISPR-Cas9 molecules is long enough to be used for precise binding if the CRISPR-Cas9 tool is well designed. Zinc finger nucleases and transcription activator-like effector nucleases have more complex attachments, which make them harder to engineer. For example, it takes many months to create zinc finger nuclease and transcription activator-like effector nuclease editing tools, whereas it takes days to create a CRISPR-Cas9 tool (with a cost reduction of similar magnitude).4
Once CRISPR-Cas9 has attached to the DNA molecule, the CRISPR-Cas9–associated nuclease Cas9 makes a double-stranded cut in a defined location on the DNA molecule. To make this precise cut, CRISPR-Cas9 must also simultaneously bind to a short DNA sequence (called a PAM sequence). The cut is made two or three nucleotides away from the PAM sequence.5,13 PAM sequences differ and depend on the bacteria from which the CRISPR-Cas9 tool was isolated.
CRISPR-Cas9 is guided to its binding site by an engineered piece of RNA created to attach to the target location on the DNA on the sense strand.14 The double-stranded break created by CRISPR-Cas9 needs to be repaired rapidly by the cell or the cell is likely to die. Prevalent exonucleases will delete nucleotides from bare cut ends of DNA. Random nucleotides may attach to the bare ends. These deletion and insertion mutations may make the gene's protein product nonfunctional. The investigator may also supply small fragments of DNA that enable the cell to rebuild a targeted cutting site by using these pieces as a template in its repair.
CRISPR-Cas9 may be used to cut out a long sequence of DNA or to remove an entire gene from a cell's DNA by using two CRISPR-Cas9 molecules designed to cut the DNA on both sides of the undesired sequence. However, cutting or disabling an entire gene from a cell's functional DNA may have a negative effect on that organism, because many genes code for more than one protein; genes and their products may have unexpected interactions to other genes and proteins, and genes may play different roles when expressed in different parts of the body.
DR. JIANKUI HE
Dr. He received his PhD from Rice University in 2010, which was followed by postgraduate work at Stanford University. After training in the United States, Dr. He was recruited back to China as part of China's “Thousand Talents Plan” (www.1000plan.org/en/young.html). Genetics is one of many areas of interest to the Chinese government.15 It is rumored that China will spend $9.2 billion on precision (genetic) medicine over the next 15 years.16 In the next year, Sichuan University's West China Hospital alone plans to sequence the genome of more than one million people.16 Meghna Kataria, writing in BioNews,17 reported that China has already treated 86 adult patients with advanced cancer using CRISPR-Cas9 technologies. A description of at least 11 ongoing clinical trials in China, involving treatment of cancer and HIV, using CRISPR-Cas9, can be found in the U.S. clinical trial registry (www.clinicaltrials.gov). These trials go back as far as 2015 and have a projected enrollment of more than 200 patients. Several Chinese teams have previously edited human embryos in experiments to eventually treat important diseases, but without the intent of using them for human implantation.18,19
Through conferences and travel, Dr. He maintained communication with many U.S. scientists involved in genetics and genetic engineering. He discussed his interest in modifying the genome of embryos with several U.S. scientists, although, according to news reports,20 these scientists did not believe that he would proceed with this project at that time.
Dr. He intended to create embryos that would be resistant to most strains of HIV. The gene CCR5 codes for a protein that most HIV strains use to enter cells.1 His objective was to eliminate production of this protein in individuals born from the modified embryos. He initially recruited eight couples in which the father had HIV and the couple wished to intervene so that their child would be prevented from ever contracting HIV.
Dr. He's approach was to delete a specific 32 base pair segment of the CCR5 gene. This specifically targeted deletion is called the CCR5 gene's delta 32 mutation. It occurs naturally in about 10% of Europeans.1 Creating a delta 32 mutation as opposed to some other mutation, would be reassuring in that the planned genetic alteration occurs naturally and therefore would be unlikely to cause unexpected adverse health effects in the resulting children. However, it is known that the delta 32 mutation is not entirely benign. Having a nonfunctional CCR5 gene increases a person's susceptibility to West Nile virus and Japanese encephalitis.21 It also may increase the likelihood of an individual dying from influenza.22,23
One may question why Dr. He chose to modify the CCR5 gene. Human immunodeficiency virus is treatable, and the knowledge of how to decrease the risk of contracting it is widespread. Perhaps the choice to modify CCR5 gene was made because this modification had been used successfully in somatic cells to make CD4+ T cells resistant to HIV infection (using zinc finger nuclease technology).24 Finding a worthwhile candidate gene to modify in germline cells may be difficult. Established technology, namely preimplantation genetic testing, can be used to avoid most identified mutations that cause recessive, X-linked, or dominantly inherited diseases and still enable carriers of these abnormal genes to have genetically related children.25 Preimplantation genetic testing screens embryos created with in vitro fertilization for the gene of interest so that only nonaffected embryos are transferred back to the patient. At this early stage of development, genetic editing of human embryos needs to prevent diseases that cannot be prevented with existing technology.
In 2017, at a conference held at Cold Springs Harbor Laboratory titled “Cancer and Stem Cells” (available on YouTube), Dr. He presented research on mouse, monkey, and human embryos. In this research he looked at the problems of avoiding mosaicism and off-target DNA changes using CRISPR-Cas9 on the CCR5 gene.
Reporters from The Wall Street Journal were able to question Dr. He's spokesperson after his talk announcing this birth at the Hong Kong conference.26 Five women underwent in vitro fertilization, and a total of 22 oocytes were retrieved. Each father's sperm was prepared in a manner to limit or eliminate HIV in the sperm specimen. Reportedly, a sperm and the CRISPR-Cas9/Cas9 enzymes were injected into 18 of the unfertilized oocytes. Thirteen edited embryos were transferred back into the five women. Two of the four embryos from one couple contained modifications of the CCR5 gene. According to Yong reporting for The Atlantic,2 one of those transferred embryos was known to still contain one copy of a normal CCR5 gene. This embryo transfer led to the birth of Lulu and Nina. Lucas Laursen reported in Fortune that there is at least one other pregnancy with a modified genome that is still ongoing.27
How to best perform gene editing of embryos efficiently and avoid creating mosaic embryos while modifying the genes inherited from both the mother and father is a subject of active research. The first Chinese team using CRISPR-Cas9 in 2015, and editing genes of nonviable embryos, was able to change the genome as desired in only 4 out of 86 embryos.19 More recently, a team at Oregon Health and Science University edited genes of donated embryos (without the intention of implanting them) with markedly better results.28 The CRISPR-Cas9 enzymes were likely injected into zygotes as close to the time of fertilization as possible.
Dr. He was successful in modifying the genes of embryos that subsequently became babies. However, it is troubling that he did not achieve what he set out to do. At the Hong Kong conference where the births were announced (and in the YouTube video that Dr. He produced, dated November 25, 2018), he reported that the genome of each child had been sequenced and showed no changes other than in the CCR5 genes. This suggests that this use of CRISPR-Cas9 made no unintended changes in the embryo's genome. However, because Lulu has one normal CCR5 gene that will still produce the normal protein, she will not be resistant to HIV infection. Of greater concern, according to an (unnamed) expert who attended the conference where the births were announced, the slides presented did not show a delta 32 mutation in any of the CCR5 genes.2 The mutations that were created in the CCR5 genes of Nina may or may not produce functional proteins. If this report is accurate, the gene modifications actually performed may not prevent Nina from contracting HIV and may or may not have other health consequences. In addition, even monitoring the health of Lulu and Nina over their lifetimes will not necessarily demonstrate that the objective of preventing HIV infection was met, because they may never be exposed to HIV.
Most countries regulate or ban research on human embryos.29 China has guidelines for such research. Investigators must present their research proposal to the institutional review board for approval at the institution where the research is to take place. Dr. He was on the faculty of Shenzhen-based Southern University of Science and Technology. Yong reported the University had turned down Dr. He's research proposal.2 The University subsequently issued a statement condemning Dr. He's experiment and denying any knowledge of it (various news sources report that he was subsequently fired.) Dr. He had taken a 3-year unpaid leave from the University starting in February of 2018. He also worked at HarMoniCare Women and Children's Hospital in Shenzhen, which is independent of Southern University of Science and Technology.26 Yong also reported that Dr. He obtained institutional review board approval from that hospital.2 The approval was published in the China Clinical Trial Registry, but the hospital claims that the signature on the approval form was forged.
As reported in the Atlantic, the Associated Press, and The Wall Street Journal,2,20,26 much of the scientific community was shocked at the risks intrinsic to this undertaking. Although Dr. He achieved a “first”; there are tangible risks involved in modifying embryos at this time. Several studies have suggested that off-target mutations after the use of CRISPR-Cas9 remain a significant problem and active area of research.14,30–33 Gene editing in animals using both CRISPR-Cas9 and prior techniques has demonstrated occasional unintended consequences. For example, there are a number of animal experiments of commercial value in which an MSTN-like gene was deleted (using CRISPR-Cas9 and prior tools). This gene codes for myostatin, which limits how large muscles can grow in humans.34 Animal gene editing experiments to delete this gene had the intended effect of doubling or tripling muscle mass, but genetically modified cattle were created that were too big to exit though the birth canal,35 some pigs were born with extra thoracic vertebrae,7 and rabbits and pigs were born with enlarged tongues.36 Gene function is difficult to fully deduce and some applications of CRISPR-Cas9 result in unintended genetic code changes with uncertain effect.14,30–33
Dr. Jiankui He was the first scientist involved in the birth of a baby with edited genes. He chose to edit a gene related to a disease, HIV, which could both be avoided and treated with established therapies. If the recipient of the gene modification does not contract HIV, it would not demonstrate the efficacy of this gene modification because the individual may never be exposed to HIV. He also chose to modify a gene to eliminate a gene product that did not completely protect the resulting child from the disease of concern. The child could still be infected by strains of HIV that used a different binding protein. Successfully eliminating this gene product by creating a delta 32 mutation, as planned, was known to create alternative health issues for the recipient of the mutation. He aided in the birth of a child who was heterozygotic for a mutated CCR5 gene and who therefore would not be immune to HIV infection. The other child born had mutations of her CCR5 genes that might still produce functional proteins, or as seen in animal experiments, might present with unintended health outcomes.
More concisely, although the thought of gene editing of embryos is an exciting prospect, our present experience using gene editing for the treatment of adults with severe disease or for beneficial genome modification of animal populations is limited. Many aspects of the experiment undertaken by Dr. He were troubling. Even with the discovery of CRISPR-Cas9, suboptimal control of molecular tools for gene editing and a review of the history of gene editing suggest the need for more caution and more collaboration before undertaking additional attempts to modify germline cells to create babies.
1. Martinson JJ, Chapman NH, Rees DC, Luis Y-T, Clegg JB. Global distribution of the CCR5 gene 32-basepair deletion. Nat Genet 1997;16:100–3.
2. Yong E. The CRISPR baby scandal gets worse by the day. The Atlantic December 3, 2018.
3. Cohen IG, Adashi EY. The FDA is prohibited from going germline. Science 2016;353:545–6.
4. Doudna J, Sternberg SH. A crack in creation: gene editing and the unthinkable power to control evolution. New York (NY): Houghton Mifflin; 2017.
5. National Academies of Sciences, Engineering, and Medicine. Human genome editing: science, ethics, and governance. Washington, DC: The National Academies Press; 2017.
6. Baltimore D, Berg P, Botchan M, Carroll D, Charo RA, Church G, et al. Biotechnology. A prudent path forward for genomic engineering and germline gene modification. Science 2015;348:36–8.
7. Qian L, Tang M, Yang J, Wang Q, Cai C, et al. Targeted mutations in myostatin by zinc-finger nucleases result in double-muscled phenotype in Meishar pigs. Scientific Rep 2015;5:14435.
8. Qi LS, Larson MH, Gilbert LA, Doudna JA, Weissman JS, Arkin AP, et al. Repurposing CRISPR as an RNA-guided platform for sequence specific control of gene expression. Cell 2013;152:1173–83.
9. Shapiro B. Mammoth 2.0: will genome engineering resurrect extinct species? Genome Biol 2015;16:228.
10. Ackermann H-W. Bacteriophage taxonomy. Microbiol Aust 2011;32:90–4.
11. Markarva KS, Haft DH, Barrangou R, Broins SJ, Charpentier E, Horvath P, et al. Evolution and classification of CRISPR-Cas systems. Nat Rev Microbiol 2011;9:467–77.
12. Horvath P, Barrangou R. CRISPR/Cas, the immune system of bacteria and archaea. Science 2010;327:167–70.
13. Anders C, Niewoehner O, Duerst A, Jinek M. Structural basis of PAM-dependent target DNA recognition by the Cas9 endonuclease. Nature 2014;513:569–73.
14. Doench JG, Fusi N, Sullender M, Hegde M, Vaimberg EW, Donovan KF, et al. Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas 9. Nat Biotechnol 2016;34:184–91.
15. Zweig D, Wang H. Can China bring back the best? The Communist Party Organizes China's Search for Talent. China Q 2013;215:590–615.
16. Cyranoski D. China embraces precision medicine on a massive scale. Nature 2016;529:9–10.
17. Kataria M. China has treated 86 people with CRISPR genome editing. BioNews January 29, 2018.
18. Kang X, He W, Huang Y, Yu Q, Chen Y, Gao X, et al. Introducing precise genetic modifications into human 3PN embryos by CRISPR/Cas-mediated genome editing. J Assisted Reprod Genet 2016;33:581–8.
19. Liang P, Xu Y, Zhang X, Ding C, Huang R. CRISPR/Cas9-mediated gene editing in human tripronuclear zygotes. Protein Cell 2015;6:363–72.
20. Marchione M. Chinese researcher claims first gene-edited babies. AP November 26, 2018.
21. Glass WG, McDermott DH, Lim JK, Lekhong S, You SF, Frank WA, et al. CCR5 deficiency increases risk of symptomatic West Nile virus infection. J Exp Med 2006;203:35–40.
22. Tyner JW, Uchida O, Kajiwara N, Kim EY, Patel AC, O'Sullivan MP, et al. CCL5-CCR5 interaction provides antiapoptotic signals for macrophage survival during viral infection. Nat Med 2008;11:1180–7.
23. Dawson TC, Beck MA, Kaziel WA, Henderson F, Maeda N. Contrasting effects of CCR5 and CCR2 deficiency in the pulmonary inflammation response to influenza A virus. Am J Pathol 2000;156:1951–9.
24. Didigu CA, Wilen CB, Wang J, Duong J, Secreto AJ, Danet-Desnoyers GA, et al. Simultaneous zinc-finger nuclease editing of the HIV coreceptors ccr5 and cxcr4 protects CD4+ T cells from HIV-1 infection. Blood 2014;123:61–9.
25. Brezina PR, Kutteh WH. Clinical applications of preimplantation genetic testing. BMJ 2015;350:7611–20.
26. Rana P, Fan W. Chinese scientist claims world’s first genetically modified babies. Wall Street Journal November 26, 2018.
27. Laursen L. Chinese scientist defends his gene-edited babies, claims a third is on the way. Fortune November 28, 2018.
28. Ma H, Marti-Gutierrez N, Park S-W, Wu J, Lee Y, Suzuki K, et al. Correction of a pathological gene mutation in human embryos. Nature 2017;548:413–19.
29. Araki M, Ishii T. International regulatory landscape and integration of corrective genome editing into in vitro fertilization. Reprod Biol Endocrinol 2014;12:108–19.
30. Mon H, Smith JL, Peng L, Yin H, Moore J, Zhang XO, et al. CRISPR/Cas 9 mediated genome editing induces exon skipping by alternate splicing on exon deletion. Genome Biol 2017;18:108–15.
31. Kosicki M, Tomberg K, Bradley A. Repair of double-strand breaks induced by CRISPR-Cas 9 leads to large deletions and complex rearrangements. Nat Biotechnol 2018;36:765–71.
32. Fu Y, Foder JA, Khayter C, Maeder ML, Reyon D. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat Biotechnol 2013;32:822–6.
33. Cradick JJ, Fine EJ, Antico CJ, Bao G. CRISPR/Cas9 systems targeting beta-globin and CCR5 genes have substantial off target activity. Nucleic Acid Res 2013;41:9584–92.
34. Lee SJ, Reed LA, Davies MV, Girgenrath S, Goad MEP, Tomkinson KN, et al. Regulation of muscle growth by multiple ligand signaling through activin type II receptors. Proc Natl Acad Sci U S A 2005;102:18117–22.
35. Bellinge RHS, Liberies DA, Laschi SPA, O'Brian PA, Jay GK. Myostatin and its implications on animal breeding: a review. Anim Genet 2005;36:1–6.
36. Qingyan L, Yuan L, Deng J, Chen M, Wang Y, Zeng J, et al. Efficient generation of myostatin gene mutated rabbit by CRISPR/Cas9. Scientific Rep 2016; 6:25029.