The Gene
The technique is still cumbersome and inefficient. Obviously, due to stringent restrictions in the creation of artificial human embryos, whether these spermlike and egglike cells can give rise to human embryos capable of normal development is yet unknown. But the basic derivation of cells capable of transmitting heredity has been achieved. In principle, if the parent ES cells can be modified using any genetic technique—including gene editing, genetic surgery, or the insertion of a gene using a virus—any genetic change can permanently and heritably be etched in the human genome.
It is one thing to manipulate genes. It is quite another thing to manipulate genomes. In the 1980s and 1990s, DNA-sequencing and gene-cloning technology allowed scientists to understand and manipulate genes and thereby control the biology of cells with extraordinary dexterity. But the manipulation of genomes in their native context, particularly in embryonic cells or germ cells, opens the door to a vastly more powerful technology. What is at stake is no longer a cell, but an organism—ourselves.
In the spring of 1939, Albert Einstein, mulling over recent advances in nuclear physics in his study at Princeton University, realized that every step required to achieve the creation of an unfathomably powerful weapon had been individually completed. The isolation of uranium, nuclear fission, the chain reaction, the buffering of the reaction, and its controlled release in a chamber had all fallen into place. All that was required was sequence: if you strung these reactions together in order, you obtained an atomic bomb. In 1972, at Stanford, Paul Berg stared at bands of DNA on a gel and found himself at a similar juncture. The cutting and pasting of genes, the creation of chimeras, and the introduction of these gene chimeras into bacterial and mammalian cells allowed scientists to engineer genetic hybrids between humans and viruses. All that was needed was the threading of these reactions into a sequence.
We are at a similar moment—a quickening—for human genome engineering. Consider the following steps in sequence: (a) the derivation of a true human embryonic stem cell (capable of forming sperm and eggs); (b) a method to create reliable, intentional genetic modifications in that cell line; (c) the directed conversion of that gene-modified stem cell into human sperm and eggs; (d) the production of human embryos from these modified sperm and eggs by IVF . . . and you arrive, rather effortlessly, at genetically modified humans.
There is no sleight of hand here; each of the steps lies within the reach of current technology. Of course, much remains unexplored: Can every gene be efficiently altered? What are the collateral effects of such alterations? Will the sperm and egg cells formed from ES cells truly generate functional human embryos? Many, many minor technical hurdles remain. But the pivotal pieces of the jigsaw puzzle have fallen into place.
Predictably, each of these steps is currently barricaded by strict regulations and bans. In 2009, after a prolonged ban on federally funded research on ES cells, the Obama administration lifted the injunction on the derivation of new ES cells in the United States. But even with the new regulations, the NIH categorically prohibits two kinds of research on human ES cells. First, scientists are not permitted to introduce these cells into humans or animals to enable their development into live embryos. And second, genome modifications on ES cells cannot be performed in circumstances that “might be transmitted into the germline”—i.e., into sperm or egg cells.
In the spring of 2015, as I completed this book, a group of scientists, including Jennifer Doudna and David Baltimore, issued a joint statement seeking a moratorium on the use of gene-editing and gene-altering technologies in the clinical setting, and particularly in human ES cells. “The possibility of human germline engineering has long been a source of excitement and unease among the general public, especially in light of concerns about initiating a ‘slippery slope’ from disease-curing applications toward uses with less compelling or even troubling implications,” the moratorium reads. “A key point of discussion is whether the treatment or cure of severe diseases in humans would be a responsible use of genome engineering, and if so, under what circumstances. For example, would it be appropriate to use the technology to change a disease-causing genetic mutation to a sequence more typical among healthy people? Even this seemingly straightforward scenario raises serious concerns . . . because there are limits to our knowledge of human genetics, gene-environment interactions, and the pathways of disease.”
Many scientists find the call for a moratorium understandable, even necessary. “Gene editing,” the stem cell biologist George Daley noted, “raises the most fundamental of issues about how we are going to view our humanity in the future and whether we are going to take the dramatic step of modifying our own germ line and in a sense take control of our genetic destiny, which raises enormous peril for humanity.”
In many ways, the proposed scheme of restrictions is reminiscent of the Asilomar moratorium. It seeks to limit the use of technology until the ethical, political, social, and legal implications of the technology can be ascertained. It calls for a public appraisal of the science and its future. It is also a frank acknowledgment of how tantalizingly close we are to making embryos with permanently altered human genomes. “It is very clear that people will try to do gene editing in humans,” Rudolf Jaenisch, the MIT biologist who created the first mouse embryos from ES cells, said. “We need some principled agreement that we want to enhance humans in this way or we don’t.”
The word to watch in that last sentence is enhance, for it signals a radical departure from the conventional limits of genomic engineering. Prior to the invention of genome-editing technologies, techniques such as embryo selection allowed us to cull information away from the human genome: by selecting embryos via preimplantation genetic diagnosis (PGD), the Huntington’s disease mutation, or the cystic fibrosis mutation, could be eliminated from a particular family’s lineage.
CRISPR/Cas9-based genomic engineering, in contrast, allows us to add information to the genome: a gene can be changed in an intentional manner, and new genetic code can be written into the human genome. “This reality means that germline manipulation would largely be justified by attempts to ‘improve ourselves,’ ” Francis Collins wrote to me. “That means that someone is empowered to decide what an ‘improvement’ is. Anyone contemplating such action should be aware of their hubris.”
The crux, then, is not genetic emancipation (freedom from the bounds of hereditary illnesses), but genetic enhancement (freedom from the current boundaries of form and fate encoded by the human genome). The distinction between the two is the fragile pivot on which the future of genome editing whirls. If one man’s illness is another man’s normalcy, as this history teaches us, then one person’s understanding of enhancement may be another’s conception of emancipation (“why not make ourselves a little better?” as Watson asks).
But can humans responsibly “enhance” our own genomes? What are the consequences of augmenting the natural information encoded by our genes? Can we make our genomes a “little better” without risking the possibility of making ourselves substantially worse?
In the spring of 2015, a laboratory in China announced that it had casually crossed the barricade. At the Sun Yat-sen University in Guangzhou, a team led by Junjiu Huang obtained eighty-six human embryos from an IVF clinic and tried to use the CRISPR/Cas9 system to correct a gene responsible for a common blood disorder (only embryos that were nonviable in the long term were chosen). Seventy-one embryos survived. Of the fifty-four embryos tested, only four were found to have the corrected gene inserted. More portentously, the system was found to have inaccuracies: in one-third of all the embryos tested, unintentional mutations in other genes were also introduced, including mutations in genes essential for normal development and survival. The experiment was stopped.
It was a daring, if slapdash, experiment, meant to provoke a response—and it did. Around the world, scientists reacted to the attempted modification of a human embryo with extreme anguish and concern. The highest-ranking scientific journals, including Nature, Cell, a
nd Science, refused to publish the results, citing broad violations of safety and ethical concerns (the results were eventually published in a scarcely read online journal, Protein + Cell). Yet, even as they read the study with apprehension and horror, biologists already knew that this was just the first step past the breach point. The Chinese researchers had taken the shortest route to permanent human genome engineering, and predictably, the embryos had been littered with unforeseen mutations. But the technique could be modified with multiple variations to make it potentially more efficient and accurate. If embryonic stem cells, and stem-cell-derived sperm and eggs, had been used, for instance, these cells could have been screened up front to cull away any deleterious mutations, and the efficiency of gene targeting might have been greatly increased.
Junjiu Huang told a journalist that he was “planning to decrease the number of off-target mutations [using] different strategies—tweaking the enzymes to guide them more precisely to the desired spot, introducing the enzymes in a different format that could help to regulate their lifespans and thus allow them to be shut down before mutations accumulate.” In a few months, he hoped to attempt another variation of the experiment—this time, he expected, with much higher efficiency and fidelity. He was not exaggerating: the technology to modify the genome of a human embryo may be complex, inefficient, and inaccurate—but it does not lie out of scientific reach.
While scientists in the West continue to watch Junjiu Huang’s experiments on human embryos with justified apprehension, Chinese scientists are far more sanguine about such experiments. “I don’t think China wants to take a moratorium,” one scientist reported in the New York Times in late June 2015. A Chinese bioethicist clarified, “Confucian thinking says someone becomes a person after they are born. That is different from the United States and other countries with a Christian influence, where because of religion they may feel research on embryos is not okay. Our ‘red line’ here is that you can only experiment on embryos that are younger than fourteen days old.”
Another scientist wrote of the Chinese approach, “Do first, think later.” Several public commentators seemed to agree with this strategy; in the comments section of the New York Times, readers advocated lifting the bans on human genomic engineering and urged a ramp-up in experimentation in the West, in part to remain competitive with the efforts in Asia. The Chinese experiments had evidently raised the stakes throughout the world. As one writer put it, “If we don’t do this work, China will.” The drive to change the genome of a human embryo has turned into an intercontinental arms race.
As of this writing, four other groups in China are reportedly working on introducing permanent mutations in human embryos. By the time this book is published, I would not be surprised if the first successful targeted genome modification of a human embryo had been achieved in a laboratory. The first “post-genomic” human might be on his or her way to being born.
We need a manifesto—or at least a hitchhiker’s guide—for a post-genomic world. The historian Tony Judt once told me that Albert Camus’s novel The Plague was about the plague in the same sense that King Lear is about a king named Lear. In The Plague, a biological cataclysm becomes the testing ground for our fallibilities, desires, and ambitions. You cannot read The Plague except as a thinly disguised allegory of human nature. The genome is also a testing ground for our fallibilities and desires, although reading it does not require understanding allegories or metaphors. What we read and write into our genome is our fallibilities, desires, and ambitions. It is human nature.
The task of writing that complete manifesto belongs to another generation, but perhaps we can scribe its opening salvos by recalling the scientific, philosophical, and moral lessons of this history:
1. A gene is the basic unit of hereditary information. It carries the information needed to build, maintain, and repair organisms. Genes collaborate with other genes, with inputs from the environment, with triggers, and with random chance to produce the ultimate form and function of an organism.
2. The genetic code is universal. A gene from a blue whale can be inserted into a microscopic bacterium and it will be deciphered accurately and with nearly perfect fidelity. A corollary: there is nothing particularly special about human genes.
3. Genes influence form, function, and fate, but these influences typically do not occur in a one-to-one manner. Most human attributes are the consequence of more than one gene; many are the result of collaborations between genes, environments, and chance. Most of these interactions are nonsystematic—i.e., they occur through the intersection between a genome and fundamentally unpredictable events. And some genes tend to influence only propensities and tendencies. We can thus reliably predict the ultimate effect of a mutation or variation on an organism for only a minor subset of genes.
4. Variations in genes contribute to variations in features, forms, and behaviors. When we use the colloquial terms gene for blue eyes or gene for height, we are really referring to a variation (or allele) that specifies an eye color or height. These variations constitute an extremely minor portion of the genome. They are magnified in our imagination because of cultural, and possibly biological, tendencies that tend to amplify differences. A six-foot man from Denmark and a four-foot man from Demba share the same anatomy, physiology, and biochemistry. Even the two most extreme human variants—male and female—share 99.688 percent of their genes.
5. When we claim to find “genes for” certain human features or functions, it is by virtue of defining that feature narrowly. It makes sense to define “genes for” blood type or “genes for” height since these biological attributes have intrinsically narrow definitions. But it is an old sin of biology to confuse the definition of a feature with the feature itself. If we define “beauty” as having blue eyes (and only blue eyes), then we will, indeed, find a “gene for beauty.” If we define “intelligence” as the performance on only one kind of problem in only one kind of test, then we will, indeed, find a “gene for intelligence.” The genome is only a mirror for the breadth or narrowness of human imagination. It is Narcissus, reflected.
6. It is nonsense to speak about “nature” or “nurture” in absolutes or abstracts. Whether nature—i.e., the gene—or nurture—i.e., the environment—dominates in the development of a feature or function depends, acutely, on the individual feature and the context. The SRY gene determines sexual anatomy and physiology in a strikingly autonomous manner; it is all nature. Gender identity, sexual preference, and the choice of sexual roles are determined by intersections of genes and environments—i.e., nature plus nurture. The manner in which “masculinity” versus “femininity” is enacted or perceived in a society, in contrast, is largely determined by an environment, social memory, history, and culture; this is all nurture.
7. Every generation of humans will produce variants and mutants; it is an inextricable part of our biology. A mutation is only “abnormal” in a statistical sense: it is the less common variant. The desire to homogenize and “normalize” humans must be counterbalanced against biological imperatives to maintain diversity and abnormalcy. Normalcy is the antithesis of evolution.
8. Many human diseases—including several illnesses previously thought to be related to diet, exposure, environment, and chance—are powerfully influenced or caused by genes. Most of these diseases are polygenic—i.e., influenced by multiple genes. These illnesses are heritable—i.e., caused by the intersection of a particular permutation of genes—but not easily inheritable—i.e., likely to be transmitted intact to the next generation, since the permutations of genes are remixed in each generation. Instances of each single-gene—monogenic—disease are rare, but, in sum, they turn out to be surprisingly common. More than ten thousand such diseases have been defined thus far. Between one in a hundred and one in two hundred children will be born with a monogenic disease.
9. Every genetic “illness” is a mismatch between an organism’s genome and its environment. In some cases, the appropriate medical intervention to miti
gate a disease might be to alter the environment to make it “fit” an organismal form (building alternative architectural realms for those with dwarfism; imagining alternative educational landscapes for children with autism). In other cases, conversely, it might mean changing genes to fit environments. In yet other cases, the match may be impossible to achieve: the severest forms of genetic illnesses, such as those caused by nonfunction of essential genes, are incompatible with all environments. It is a peculiar modern fallacy to imagine that the definitive solution to illness is to change nature—i.e., genes—when the environment is often more malleable.
10. In exceptional cases, the genetic incompatibility may be so deep that only extraordinary measures, such as genetic selection, or directed genetic interventions, are justified. Until we understand the many unintended consequences of selecting genes and modifying genomes, it is safer to categorize such cases as exceptions rather than rules.
11. There is nothing about genes or genomes that makes them inherently resistant to chemical and biological manipulation. The standard notion that “most human features are the result of complex gene-environment interactions and most are the result of multiple genes” is absolutely true. But while these complexities constrain the ability to manipulate genes, they leave plenty of opportunity for potent forms of gene modification. Master regulators that affect dozens of genes are common in human biology. An epigenetic modifier may be designed to change the state of hundreds of genes with a single switch. The genome is replete with such nodes of intervention.