Artificial Life

In the mid-1990s, Craig Venter rose to fame by claiming that he and his colleagues would decipher the human genome long before a huge team of government scientists would. He at least managed a tie: Both groups have provided increasingly accurate versions of the genome since 2000, and Venter has just published the first genome sequence from one person (himself) that includes all the chromosomes inherited from his parents. As important as sequencing the human genome has been, however, Venter is overseeing another experiment that could someday eclipse it. Scientists at the J. Craig Venter Institute and Venter’s biotech firm, Synthetic Genomics, are trying to make a genome from scratch. “I plan to show that we understand the software of life by creating artificial life,” Venter declares in his new memoir, A Life Decoded.

Venter first announced this plan in 2002, and he has been doggedly pursuing it ever since. Step one of the plan: Identify the fewest number of genes a microbe needs to survive in a lab. The scientists would then synthesize that minimal genome from raw ingredients and insert it into a host cell. The genome would make its own proteins and gradually transform the cell into a new creature.

To build the minimal genome, Venter turned to a microbe he and his colleagues had already been studying for several years, a pathogen known as Mycoplasma genitalium that causes urinary tract infections. He and his colleagues had determined (pdf) that the parasitic microbe has just 482 genes (18,000 reside in a human). They then began testing each of those genes to see which were essential to the microbe’s survival. Last year, they reported that M. genitalium can survive without 100 of its genes. “We know which genes we can eliminate one at a time, but we don’t know which we can eliminate together,” Venter says. To see if the remaining 382 genes meet the minimum requirement for life, Venter’s team will have to build a genome with them and drop it into a cell.

Venter knew that no one had ever successfully transplanted a bacterial genome, and there were a lot of reasons to suspect it might not work. “Cells in general don’t like another cell’s DNA injected into them,” he says. But this June, he and his colleagues delicately teased out the entire genome of Mycoplasma mycoides (which infects goats) and slipped it into Mycoplasma capricolum, a related but distinctly separate species. Tests revealed that the recipient bacteria lost their old genomes, while the donor genomes began to take over. “It’s the key breakthrough in this field,” Venter says.
Now Venter’s team is creating a minimal genome themselves and transplanting it into a cell. more...