Last year, researchers working to synthesize the genome of a strain of yeast began to eye a much bigger prize: assembling from scratch the 3 billion base pairs of DNA that drive a human cell. The idea caught the attention of other prominent scientists, and inspired a proposal published online in Science today. The so-called Human Genome Project–Write (HGP-write) aims to synthesize entire genomes—of humans and other species—from their chemical components, and get them to function in living cells.
The initiative generated buzz last month after an invitation-only meeting to discuss the prospect at Harvard Medical School in Boston. Its organizers aimed to keep the details under wraps until this more formal proposal was published—a move that rankled those hoping for a broader public discussion about its ethical, legal, and social implications. Some speculated that scientists would use these engineered cells to create designer humans with no genetic parents.
The new proposal makes clear that HGP-write has no such aim, the authors say. The main goal instead is to drive down the cost of engineering large stretches of DNA and testing their activity in cells. “HGP-write would push current conceptual and technical limits by orders of magnitude,” the authors write.
That “learning by building” approach would put scientists’ understanding of the genome to the test, says chromosome biologist Torsten Waldminghaus of Philipp University of Marburg in Germany, who is not involved in the project. “You know all the parts needed [to make a chromosome], so you take these parts and rebuilt it,” he says. “If it’s functional, you see that you were right.”
Researchers have already constructed functioning viral and bacterial genomes, and the yeast genome project, known as Sc2.0, aims to have all 16 chromosomes—roughly 10 million base pairs—assembled by the end of next year. But a mammalian genome is a different prospect, says synthetic biologist Tom Ellis of Imperial College London, an Sc2.0 collaborator who attended the Harvard meeting. Researchers would need to choose an appropriate cell line to act as a host and then gradually swap out large chunks of its genome with the synthetic DNA. It’s not yet clear how best to physically shuttle this DNA into a mammalian cell, or how to design synthetic sequences that keep their host cell functioning normally. “How do you debug it if you’re throwing in a million bases at a time?” Ellis says. “That’s a lot of hunting.”
Those challenges were enough to give pause to a veteran of the field. “I’m kind of a conservative guy, and at first my reaction [to HGP-write] was not very enthusiastic,” says geneticist Jef Boeke of New York University’s Langone Medical Center in New York City, another Sc2.0 collaborator who eventually helped organize the Harvard meeting and is the lead author on the new paper. When Boeke met with Sc2.0 colleagues last summer to discuss their next target organism, San Francisco, California–based futurist Andrew Hessel of the software company Autodesk “sort of made this impassioned speech” for a new human genome project that would capture the public’s imagination and inspire the field around a single goal, Boeke says. As more potential collaborators got onboard, including Harvard geneticist George Church, Boeke says he “became more and more convinced that this really was a good focus.”
Compared with gene editing with tools such as CRISPR/Cas9, constructing a full genome could allow more widespread manipulation, Church explains. CRISPR is already being refined to make large numbers of genetic changes to cells with increasing precision, but to recode various points throughout the genome, “it might actually be lower-cost to just synthesize the whole thing,” he says.
The new publication describes pilot projects that might harness large-scale synthesis for human health applications. One set of examples involves “ultrasafe” human cell lines: Cells used to secrete proteins used in drug treatments could be engineered to be resistant to viruses that might contaminate the product. Or stem cells injected into a patient as therapy might be designed so that their tumor suppressor genes are less likely to mutate and cause cancer. One possible pilot project would create human cells that can tolerate a simpler and less costly culture medium by surviving without the typical amino acids and vitamins. Another would engineer a pig genome—eliminating embedded viral genes and genes encoding molecules that are immunogenic, for example—so that researchers could grow pig organs suitable for human transplant.
The paper’s 25 authors hope to launch their project this year with $100 million in funding—a figure Boeke calls “a somewhat arbitrary milestone.” At least for now, the collaborators aren’t expecting a big infusion of new federal funding. The initial sum could be pooled from existing grant funding for labs already working in genome synthesis, he says.
But the mere principle of a fully synthetic human cell is problematic to some. The list of potential benefits is “not an adequate reason to take such an enormous moral step,” says Laurie Zoloth, a bioethicist at Northwestern University, Chicago, in Illinois. “I think developing the tools to make large genetic sequences is an important human goal. Creating an entirely new [human] genomes—that’s a different kind of a project.” Zoloth and synthetic biologist Drew Endy of Stanford University in Palo Alto, California, first drew attention to the May meeting when they published a critical response, arguing that these initial discussions should have been open to the broader public.
The authors had hoped that this paper would coincide with last month’s event, Church explains, but “we didn’t really have a completed message,” when the meeting date arrived. The new proposal promises an “open and ongoing dialogue” about the ethical, legal, and social implications of the work, and suggests that a percentage of the project’s funding should go to supporting such discussions. How genetic engineering might be applied to humans is “one of these things where people keep talking about it,” Church says. “That’s good. It’s not going to happen accidentally.”