This story was updated at 5:00 p.m.

The first microbe to live entirely by genetic code synthesized by humans has started proliferating at a lab in the J. Craig Venter Institute (JCVI). Venter and his colleagues used a synthetic genome—the genetic instruction set for life—to build and operate a new, synthetic strain of Mycoplasma mycoides bacteria, according to an online report published May 20 by Science.

"This is the first self-replicating cell on the planet to have a computer for a parent," said J. Craig Venter during a press briefing on May 20. "It's also the first species to have a Web site in its genetic code."

For the past 15 years, the genomes of thousands of organisms have been sequenced and deposited in databases. "We call this digitizing biology," JCVI molecular biologist Daniel Gibson told Scientific American. "We now show that it is possible to reverse this and synthesize cells starting from this digitized information….We refer to the cell we have created as being a synthetic cell because it is a cell controlled by a genome assembled from chemically synthesized pieces of DNA."

In other words, a chemical synthesizer stitched together various short iterations of man-made adenine, cytosine, guanine and thymine that were then assembled into a working genome that can successfully produce the proteins that enable life. Using stretches of DNA, known as cassettes, roughly 1,000 base-pairs in length, the researchers assembled a simplified version of M. mycoides genome from scratch in a succession of E. coli and yeast cells. The final synthetic genome—more than a million base-pairs long—was then inserted into an existing Mycoplasma capricolum cell. The synthetic cell then went on to behave as a M. mycoides, producing proteins from the instructions encoded by the synthetic genome and even dividing and growing.

"It is a big deal," geneticist and technology developer George Church of Harvard Medical School says of the achievement. "It's not incremental, but it's not final either," noting that other groups are already delivering useful products from partially reengineered genomes, such as biofuels from engineered E. coli.

Biological engineer Drew Endy of Stanford University clarified how to think of this creation. "It's not genesis, it's not as if mice are coming from a pile of dirty rags in a corner," he says. "The correct word is poesis, human construction. We can now go from information and get a reproducing organism. It lays down the gauntlet for us to learn how to engineer genomes."

Getting to this point was not without its challenges, including requiring at least $40 million in investment into relevant experiments over the past 15 years, primarily funded by Venter's private company Synthetic Genomics and the U.S. Department of Energy, among others. The researchers started with the intention of synthesizing the genome of Mycoplasma genitalium, which has the smallest known natural genetic instruction set. But that organism's slow growth and other properties led them to ditch it in favor of genetically more complex cousins such as M. mycoides and M. capricolum. To simplify things, they deleted 14 genes from M. mycoides natural genome, leaving behind hundreds.

Then the researchers could not find a way to transfer genomes from one bacterial species to another, eventually enlisting the yeast as an assembly waystation, permitting easier manipulation of genetic material and overcoming natural resistance in the microbes to tinkering with their DNA. The yeast also copies the synthetic genome numerous times with its own to allow spares for experiments, while adding its own genetic twists, such as eight single nucleotide polymorphisms now found in the synthetic genome. In fact, there are 19 total nucleotide sequence differences between the synthetic genome and its natural analog. And, thus far, genomes can only be swapped between closely related species. "Right now, we don't know how far phylogenetically speaking the donor and recipient can be," said JCVI microbiologist Carole Lartigue at the May 20 briefing.

But once this synthetic genome was inserted—the would-be host cell failed "and we did not know why," Gibson says. By cross-checking the entire genome gene by gene, they found the fatal flaw after three months of work: a single missing base in the dnaA gene, which is required for life. "Accuracy is essential," Venter said. "There are parts of the genome where it cannot tolerate even a single error."

Of course, the rest of the original cell remains "naturally" made, from the cytoplasm on down, but the billions of daughter cells are assembled entirely from proteins encoded by the synthetic genome. Once the perfected synthetic M. mycoides genome was inserted into M. capricolum, on March 26, it booted up the natural cell's machinery and busily set to work living, making proteins and, ultimately, dividing and thriving. By March 29, the researchers found a thriving blue colony of M. capricolum living as synthetically driven M. mycoides. "The cells with only the synthetic genome are self-replicating and capable of logarithmic growth," the researchers wrote, and grow "slightly faster" than their natural peers.

Venter and his colleagues also included four "watermarks" in the code to distinguish the synthetic microbe—dubbed Mycoplasma mycoides JCVI-syn1.0—from natural organisms, including 46 names of scientific contributors to the synthetic genome, an email address and a web site based on a code derived from the four letters of the bases and 64 combinations of the four letters, or triplets, possible in the genetic code. "When you put English text into [the code], it generates very frequent stop codons in the genetic code and won't produce big proteins," said JCVI microbiologist Hamilton Smith, a Nobel Laureate in medicine. "It's designed to be biologically neutral."

Gibson adds: "If one is able to translate the watermark sequences, they will be able to send us an email and prove that they decoded the sequences."

The man-made genetic code also includes three quotes: "To live, to err, to fall, to triumph, and to recreate life out of life" from James Joyce; "see things not as they are but as they might be" from Robert Oppenheimer via the Ethical Culture School in New York City; and "what I cannot build, I cannot understand" from physicist Richard Feynmann.

As for the first synthetic cells, they now lie dormant in a JCVI freezer. "If there's a cell museum, we may donate it," Venter said. "If we need it, we can thaw it out and it will start replicating again."

What could go wrong
The mere fact of human-directed life in the lab raises its own concerns, including the potential for synthetic life to escape the lab and exterminate its natural cousins, or infect them with synthetic DNA through horizontal gene transfer. Various methods to control this have been suggested, including building genetic sequences that cannot exist in nature, engineering in weaknesses to man-made cells, or even inserting suicide genes that kill the organism if it is removed from its lab environment. "We depend on algae for a fair amount of the oxygen we breathe, it would be bad if we messed that up," Venter noted.  

Man-made creations are likely to be fragile compared to their robust natural counterparts that have been engineered by billions of years of evolution and competition, Church notes, but he also calls for strict oversight to be built into the process of working with or creating such synthetic organisms. "The first safeguard turns out to be to have other people review the work you're going to do so it's not one person coming up with an idea at the bench," Endy adds. "It's a buddy system if you will."

After all, the JCVI scientists "are now ready to build different organisms," Gibson says. "We would like to use available sequencing information and create cells that can produce energy, pharmaceuticals, industrial compounds and sequester carbon dioxide."

In fact, Venter hopes to use the techniques to begin synthesizing antiviral vaccines in days rather than weeks or months. "We have ongoing funding from [the National Institutes of Health] in a program with Novartis to use these new synthetic DNA tools to perhaps make the flu vaccine you might get next year," Venter said, as well as to develop vaccines for viruses that had previously eluded treatment because of their ability to rapidly mutate, such as rhinovirus (the common cold) and HIV (AIDS). And the researchers hope to tinker with the at least 2 million base pairs of an algae genome to help it more efficiently turn sunlight and CO2 into hydrocarbons.

Tackling even more complex genomes remains a daunting task, so many of the researchers involved will now focus on an attempt to create the simplest genome possible that can still permit life. "We can whittle away at the synthetic genome and repeat transplantation experiments until no more genes can be disrupted and the genome is as small as possible," Gibson says, estimating that this could be less than half of the more than a million base pairs required by this first synthetic genome. "This will help us to understand the function of every gene in a cell and what DNA is required to sustain life in its simplest form." As well as what DNA might be desired for a future synthetic biology.