Bacterial genomes are tiny—just a few megabases in size—but they’re not without feature-bloat. Most genomes encode proteins that bacteria require only occasionally, if at all. The question is, how many? To put it another way, what’s the bare minimum genome required to sustain life?
In a new report in Science, Venter and his team describe Mycoplasma mycoides JCVI-syn3.0, a downsized version of the “synthetic cell” they first described 6 years earlier (1). That first genome, M. mycoides JCVI-syn1.0, was a fully synthetic rewrite of the 1.08-Mbp, 901-gene wild-type M. mycoides genome, albeit with a few watermarks and other genetic tweaks built in—the functional equivalent of retyping a manuscript from the original text (2). syn3.0, measuring 573-kb and encoding just 438 proteins and 35 RNAs, is half that size—a stripped down, streamlined rewrite. Incredibly, nearly a third of the remaining genes (149) have no known function.
Although it’s possible that a few additional genes still could be pared away under the right growth conditions or if, say, the team were willing to accept a slower doubling time, the key point, said Rich Roberts, from New England Biolabs, is that the team have come up with a working definition of a minimal genome—a platform for understanding the basic requirements of life.
“If you can make a minimal genome and really define some minimal set of genes that are necessary for life, then you’ve got a good starting point for trying to understand what is the basic stuff, if you like, that needs to go on in a cell in order for it to be alive,” he explained. syn3.0, he added, meets that definition: “It’s a pretty small number of genes, and it is alive. There’s just no two ways about it; this thing is alive, it lives.”
Farren Isaacs from Yale University called the study a “a tour-de-force” that advances both genomics and synthetic biology. “From a technological perspective, I think it advances capabilities for design, building, and testing of synthetic genomes, which is a significant improvement from the JCVI’s prior work between 2007 and 2010, where they can much more effectively iterate between different genome designs and experimental testing and validation.”
Building a Better Genome
syn3.0 is the culmination of nearly two decades of work for Venter and his team—work that began with the 1995 sequencing of the genome of a related microbe, Mycoplasma genitalium. (M. genitalium held the previous record for the smallest known genome, with 525 genes.) Since then, his team has used bioinformatics and transposon mutagenesis to calculate a presumptive minimal set of 375 genes; worked out how to synthesize and assemble a genome from scratch; and transplanted a chromosome into a cell to “reboot” it with a new genetic operating system. The culmination of that work was syn1.0.
But according to Venter, syn1.0 was “basically the control experiment—it proved that we could actually chemically synthesize a 1.1-Mb genome and boot that up and get a whole new cell from that process.”
The team next turned its attention to minimizing the syn1.0 genome. Using their predictions about what constituted core genetic functionality, Venter and his colleagues first attempted to write a new genome from scratch, but they failed to produce a viable cell. “I was actually relatively certain that one or more of them would work based on the results of data from the last 20-some-odd years,” Venter said. “But our study shows that, clearly, those estimates were wrong, because they were only based on the known world” —that is, on assumptions of what were and were not essential genes.
So, the researchers reversed course, starting with an intact genome and pruning away what wasn’t absolutely required. They divided the genome up into eighths, deleting genes previously identified as nonessential and testing each fragment in the context of a seven-eighths complete syn1.0 genome. Then, through the use of an iterative design-build-test strategy that involved comprehensive transposon mutagenesis, the judicious deletion and restoration of genes, and mixing and matching of genome segments, the team arrived at an intermediate design they called syn2.0 (576-kbp, 478 genes) and, ultimately, at syn3.0. That final organism has a doubling time of 3 hours, one-third that of syn1.0.
According to Venter, the process was hampered by synthetic lethals, pairs of genes that are nonessential on their own but lethal if both are missing. “It ended up being largely a trial-and-error kind of process,” he concluded.
A Versatile Platform
The Venter team called syn3.0 “a versatile platform for investigating the core functions of life and for exploring whole-genome design.” They showcased some of those possibilities by first reorganizing the genes in one eighth-genome segments to group them by function—a process Venter called “defragging”—and then altering the codon optimization and usage in a separate 5-kb segment.
Venter said his team is now investigating the functions of the 149 unknown genes they identified, as well as working to make syn3.0 available for the community. He also is considering a contest “to see who can add the most interesting steps in evolution to this basic chassis of a cell and a genome”—the ability to photosynthesize, for instance.
The process of genome synthesis is easier and cheaper than ever, Venter said, and he fully expects other researchers to follow his team’s footsteps to build other minimal genomes in other organisms. His company, Synthetic Genomics Inc., even sells a “DNA printer” to facilitate the process.
But George Church from Harvard Medical School noted that gene editing techniques such as CRISPR and MAGE may be far more accessible for most researchers. In this study, Venter’s team built full-scale genomes from scratch in about 3 weeks. But using genome editing technologies—altering the genome “with the motor running,” as Church put it—researchers can explore billions of designs per day, a strategy Church and Isaacs have used to recode regulatory elements and DNA triplets in the genetic code. “Admittedly, there are only combinatorial changes among the billion genomes, but that’s the whole point: Those subtle differences are what makes or breaks a genome.”
As for Roberts, the study’s most significant aspect is not technological, but genetic: “These 149 genes are essential to life, [and] we haven’t got a clue what they do.” That, he said, represents “a challenge to the biological community,” and “absolutely the most stunning discovery that came out of this.”
1. Hutchison, C.A., III, et al., “Design and synthesis of a minimal bacterial genome,” Science, 351:aad6253, March 25, 2016. [http://science.sciencemag.org/content/351/6280/aad6253]
2. Gibson, D.G., et al., “Creation of a bacterial cell controlled by a chemically synthesized genome,” Science 329:52–6, 2010. [http://science.sciencemag.org/content/329/5987/52]