Adam Bluestein

Mycoplasma capricolum

The breathless media coverage of the J. Craig Venter Institute’s latest synthetic biology breakthrough, published in Science, is somewhat puzzling, but given Venter’s Barnumesque self-promotion gene, not totally surprising. What’s a bit disappointing, though, is the uncritical repetition of Venter’s claim that his group’s latest accomplishment — transferring the genome of one type of bacteria into a yeast cell, modifying it, and then transplanting it into another bacterium — inevitably puts us on the brink of creating a “fully synthetic organism,” as BBC News reports.

Certainly, Venter sounds confident — but he always does, to the amusement and occasional consternation of other leading minds in the synbio arena. “Assuming we don’t make any errors, I think it should work and we should have the first synthetic species by the end of the year,” he told the press last week.

So, what actually happened, and what does the claim about creating “synthetic species” really mean? Some background: In 2007, researchers at the Venter Institute in Rockville, Maryland, managed to transfer the genome (the complete genetic code, embedded in DNA) of one type of bacteria into another closely related species from which the original genetic material had been removed. The transfer was successful: the recipient cell  “booted up” using the genetic code from the donor, taking on its characteristics — the (very) rough equivalent of a Freaky Friday-style personality transfer. Then, in 2008, the same group synthesized the genome of the bacteria Mycoplasma genitalium — that is, created it entirely from pieces of DNA derived from chemicals in a lab. However, when they tried to install the synthetic genome into a “hollowed out” bacterial cell, it failed to come alive.

The Venter group now posits that the problem was akin to that of organ rejection in transplant patients — that the synthetic genome, which was assembled in a yeast cell, was perceived by the new host bacterium as a foreign invader and was therefore destroyed in self-defense. The new paper, authored by Sanjay Vashee and colleagues, describes a technique for overcoming the genome-rejection problem. The researchers first transplanted the entire genome of a bacteria, Mycoplasma mycoides, into yeast — a first. Then, using existing tools for genetic engineering in yeast, they modified the genome so that it would carry the molecular markings of a bacteria rather than a yeast, an entirely different organism. The trick worked — when the modified genome was taken from the yeast and inserted into an “empty” recipient cell, Mycoplasma capricolum (a species closely related to the donor), it rebooted as a viable mycoides cell that went on to undertake multiple rounds of cell division.

So far, so cool. Here’s why. Up until now, most synthetic biology work has been done in E. coli and strains of yeast, which are quite amenable to tinkering. However, there are many other microorganisms that are much less hackable but whose mechanical apparatus — say, the ability to perform photosynthesis or survive in extremely harsh environments — makes them attractive potential tools for industrial applications like biofuel production or environmental remediation. Having a technique that enables researchers to move genomes around between different hosts could prove extremely useful for engineering a broader catalog of life forms.

But does the new technique described by the Venter team also lead us, inevitably, to synthetic life forms, as Venter would have us believe?  Some aren’t so sure. The Scientist‘s report on the paper notes parenthetically that the current study was conducted in a natural Mycoplasma genome — not the synthetic genome the group assembled last year. This is a crucial difference. Venter’s team assures us that the problem with rebooting using the synthetic genome was what I’m calling the rejection issue. But it could just as well be some other problem with the synthesized genome itself, or some other element of the interaction between the synthetic genome and the host.

“When you write a computer program and it doesn’t work, do you blame the computer or do you blame the programmer?” says a synthetic biology researcher at Harvard who spoke with me off-record. ” When you generate a synthetic genome, you don’t really know what it’s going to do in any of these hosts. The booting up process here is a chicken-and-egg problem. All these proteins are in there to do transcription and translation, and now what you’re saying is I’m going to throw a genome in and say, ‘Just use this as your substrate rather than this other stuff,’ and there’s no reason to believe that that’s going to work. What’s the matter with it — why it’s not working — could be for some reason we just haven’t discovered yet.

“However, this ability to translocate between one thing and another could be useful in that way. If I have a computer program and it doesn’t work on this computer, I can go try it on a bunch of other computers to see where it might boot. If I had a synthetic genome and I couldn’t get it to work in this [organism], maybe I could get it to work in something else. If you look at the steps of synthetic life, you need all of these engineering principles to make things work … the tools to generate a genome from scratch — we’ve done that essentially — and we also need a way to move that genome around into different hosts, and this is what they’ve shown.”

To be sure, the idea of building an entirely synthetic genome that will operate a single-celled organism is becoming more and more plausible — something I’m convinced we’ll see in the next few years, if not the few months that Venter predicts. But there are just as certainly a few more unforeseen obstacles standing in the way. And even then, will we really have created “life from scratch”? After all, the resulting life forms now being discussed will still rely on a biological “chassis,”  the “hollowed out” cell. As the Harvard researcher points out, “That shell is not such an empty husk — if that’s all it was, it wouldn’t work. If the empty husk is like a computer and has the basics of microprocessor and so forth … the unique thing about biology is that not only does it read the program, it also modifies the program, turning genes on and off.”

“Life” as most of us understand it, is the combination of the software and hardware — the genome and the not-really-empty cell. A working synthetic genome, while a significant breakthrough with potentially enormous practical applications, would still be a far cry from something we could honestly call “life from scratch.”

It’s also worth pointing out that the whole “life from scratch” idea is one that holds a lot more fascination for laypeople (and journalists) than for many of the scientists actually practicing synthetic biology. “It’s kind of the ultimate powerful thing — you know, ‘I made life from scratch,'” Pam Silver, of Harvard Medical School, told me in a recent interview. “But it’s not something I personally want to be doing very much. I think we have good systems that develop and divide, which is the key to all of this this — that these are engineered machines that develop and divide, that self-perpetuate as opposed to static machines. So I’m pretty satisfied with what we’ve got, though I find what’s going on in that area interesting.” In an interview I conducted with Harvard’s George Church this summer, he told me: “I’d be delighted if we did it [made life from scratch] en route to something else, but we’re trying to keep our eye on things that are useful and intellectually interesting. I suppose that making life from scratch is somewhat intellectually interesting, but it’s not a well-formulated difficult paradox unless you believe in vitalism. So I think we’re going to be focused on basic enabling technologies and then applying them to societal problems.”