Scientists are engineering a new living thing: a radically modified version of the lowly bacterium E. coli. In an article in Science from August, researchers at Harvard University described an ongoing project to build the genetic code of E. coli from scratch, but with major revisions to create a new strain unlike any in existence.
The modified E. coli is meant to be so foreign to viruses that they will not be able to infect it. George Church, the principal investigator on the study, told me the finished product “will be completely, unassailably resistant to all viruses, even viruses never seen before.” This offers important advantages in industrial processes such as pharmaceutical and biofuel manufacturing, said Matthieu Landon, one of the lead authors on the paper. (Genzyme, a biotechnology company, suffered a crippling viral contamination in 2009 at its Allston, MA, plant that made two essential drugs unavailable for the patients who depended upon them. A virus-resistant cell would not be vulnerable to such contamination.)
The E. coli project is a step toward the synthesis of a human genome, a goal that holds scientific promise but has also raised ethical concerns. In June, Church and a consortium of researchers published information about a planned project to synthesize the human genome. The news was met with criticism by some scientists not involved with the project, who expressed concern that details had been discussed behind closed doors and without adequate consideration for ethical concerns.
Critics worry that a synthetic human genome could be used in unethical ways, such as for the creation of “designer babies.” Laurie Zoloth, a bioethicist at Northwestern University, has advocated for the formation of ethical oversight committees to supervise this type of research. In a recent interview, she asked, “Who’s watching to say if this is a good or bad thing?” Unlike for clinical trials, there is no regulatory body for basic science research.
Church’s consortium did not endorse the synthesis of a human genome within a reproductive cell—that is, within an egg or sperm that could be used to create a living human. Instead, the researchers intend to engineer blood or tissue cells that could only be grown in a laboratory as individual cells.
Critics raise two reasonable concerns. The first is that even though “legitimate” scientists like Church declare no intention to create a synthetic human being, if the technology exists it would be hard to stop less mainstream scientists from giving it a shot. If induced pluripotent stem cells, a type of cell often used in research, were created with a synthetic genome, could a scientist then steer the cells to develop into reproductive cells? “This enters into the world of reproduction fairly easily. That’s why someone has to make sure that there are careful rules to govern what is done,” Zoloth told me.
An oversight committee sounds reasonable at first, but also carries risks. Creating a rigid bureaucratic structure could slow the pace of innovation by forcing researchers into fixed paths. Research cannot be planned years in advance; it must pivot constantly as researchers learn from the experiments they conduct. A better approach would be to hold researchers to a set of ethical principles and discipline scientists who deviate from those principles. Punishment could take several forms—loss of funding, a ban on publishing in peer-reviewed journals, and revocation of university affiliation would all be potent deterrents. This would not regulate specific scientific activities but the overall approach to science, similar to the function of the Hippocratic oath in medicine.
In a project running in parallel with Church’s E. coli effort, a consortium led by Jef Boeke at NYU aiming to synthesize a yeast genome has adopted a set of 11 guidelines across 4 domains (societal benefit, intellectual property, safety, and governance) intended to place guardrails on the actions of the groups’ members. Such a document could be used as a model to limit unethical endeavors on a broader scale without impeding valid research. Boeke has partnered with Church on the human genome synthesis project.
A second major complaint about the proposed project is that there is no clear application for synthetic human cells. Church argues that synthesized human genomes could be used in stem cell or organ transplantation. By creating virus-resistant stem cells, for example, patients could gain some protection from dangerous infections. “When the patient is given a choice between two cell sources, one of which is pathogen-resistant and another which is not, the patient is going to choose the pathogen-resistant one,” Church said. Yet this approach would only confer virus resistance to the cells in the transplant (blood cells or solid organ) itself, not the transplant recipient’s native cells, so the benefit would be modest.
In an interview with JAMA last week, Church also suggested that if a pig’s genome were synthesized, it could be used to produce organs that would be easier to transplant into a human. This, however, would require a separate pig genome synthesis, which would be just as resource-intensive as a human genome synthesis, with no guarantee of successfully creating human-friendly pig organs.
Even though the medical applications are not totally clear in this case, science should not be restricted to areas where there are obvious applications. A classic example: when Einstein developed the general theory of relativity, he did not expect that people would use it to help find the nearest Taco Bell, but today’s GPS systems do depend somewhat on Einstein’s theory. Church’s consortium should not have to justify its work by defining exactly what will be done with it. Applications of basic science research are usually hard to predict (see also transistors, lasers, and the atomic bomb).
While the ethical debate continues, Church and company say they are already raising funds for the human genome synthesis, which they have called “HGP-write” in reference to the original Human Genome Project. The human genome is one thousand times larger than that of E. coli, but the vast majority of DNA in a human is “non-coding.” The coding regions of human DNA are only five times larger than E. coli’s and the pace of synthesis is increasing exponentially. While the E. coli project initially required intensive troubleshooting with each DNA revision, with more experience the editing process is becoming streamlined, saving many hours of work. Landon, a boyish Frenchman who captains the Harvard Business School rugby team and who Church described as “one of the world’s best” at this troubleshooting, told me, “As design rules are better known, troubleshooting lethal elements will probably be almost 100 percent eliminated.”
A synthetic human genome is “years, not decades” away, according to Church. There are risks. But the bigger risk would be to stifle the scientific process and push innovation into darker corners of the scientific community where there would be less oversight and more nefarious motives.