For the last few decades, scientists have been intently decoding the genes of dozens of organisms, from bacteria to humans. The effort, which culminated in 2000 with the deciphering of the human genome’s roughly thirty thousand genes, reflects researchers’ increasing adeptness at “reading” the language of DNA. It’s a biological literacy that has meant dramatic advances in understanding the genetic basis of health and disease, bringing with them the promise of safer and more effective drugs.
But now a small group of researchers are looking to a far more ambitious goal than simply reading the sequence of genetic material: they are attempting to write entirely new genomes from scratch. In essence, they hope to create new synthetic forms of life, the likes of which have never before existed, by painstakingly spelling out exact sequences of DNA that hold all the instructions for the new organisms.
It is biotech’s most brazen attempt, so far, to play God. So the fact that Craig Venter, the legendary self-aggrandizing visionary of genomics, is leading the charge should come as no surprise. After all, it was Venter, then the president of Celera Genomics, who headed the controversial private effort to sequence the human genome-and to do so ahead of the public Human Genome Project. Now through the Rockville, MD-based Institute for Biological Energy Alternatives, a nonprofit organization Venter launched last April, he is gearing up to build a synthetic bacterium, by first writing out its genome. It’s a project that would not only help meet the center’s goal of creating high-utility microorganisms specifically designed to mop up carbon dioxide, say, or produce hydrogen fuel with the utmost efficiency; it’s a project that could also upend genetic engineering itself.
Venter’s objective is not merely to tweak existing life forms by inserting genes that confer specific traits-the main tactic in conventional genetic engineering. Instead he wants to assemble an entire genome, DNA letter by DNA letter, putting together only the genes he wants: those necessary for an organism’s survival and those that will allow it to carry out a desired task. “The long-term advantage of creating an organism from a chemically synthesized genome is that it allows complete flexibility of design,” says University of North Carolina biologist Clyde Hutchison. No longer limited to nature’s repertoire, researchers could create a wide variety of synthetic organisms, each made to perform a specific chore, such as sopping up oil slicks or producing a plastic. And because such a bacterium would devote most of its energy to its assigned job, it could, in theory, be much more efficient than a counterpart made via conventional genetic engineering.
Building such an organism “would be a momentous achievement,” says Eugene Koonin, an evolutionary biologist at the National Center for Biotechnology Information, a part of the National Institutes of Health. But, Koonin points out, creating life from scratch presents a few daunting challenges. Even a simple single-celled organism such as a bacterium can have hundreds of thousands or millions of DNA letters. And even if scientists do figure out the exact sequence of DNA that would create the organism they want, they would still have to figure out how to “write” those letters in long stretches of DNA. Though researchers routinely synthesize short DNA strands, today’s DNA-making machines can’t handle anything longer than a hundred letters or so. And piecing together an entire genome from such tiny fragments is a monumentally time-consuming, error-prone task.
Still, despite the audacity of the endeavor, Venter is not alone in his ambition to rewrite the language of life. A small cadre of researchers in academia and industry are working out the details of technologies that could make genome writing routine; one such tool is a machine capable of synthesizing, automatically and with high accuracy, genome length stretches of DNA. And while those scientists are honing their DNA-writing skills, others are looking to transform the genetic language itself by adding entirely new letters to the DNA alphabet-thus creating the potential to give organisms abilities not seen in nature. “The program got written four billion years ago,” says MIT computer scientist Tom Knight, who studies the interface between biology and computing. “It’s time to rewrite the program.”
The roots of Venter’s ambitions lie in an earlier effort called the Minimal Genome Project. In the mid-1990s, researchers at Venter’s Institute for Genomic Research sequenced the first two complete genomes-both bacterial. Armed with new information on the genes that make each species unique, the researchers were curious to see which of those genes were necessary to sustain life. It wasn’t just curiosity that drove the project though. If researchers could identify those genes, they reasoned, they would be able to model the most basic of cell activities. “That would be helpful in understanding more complex cells and in designing new ones,” says Hutchison.
The team started with a lowly bacterium called Mycoplasma genitalium, whose tiny genome consists of just 517 genes made up of about 580,000 DNA letters. “I raised the question: Does the bacterium need all those genes?” says Hutchison, who had taken a sabbatical from the University of North Carolina to work on the project. By selectively disabling different genes, the researchers discovered that only 265 to 350 were essential: few enough to make it conceivable that researchers could assemble the entire genome from scratch, though the endeavor might take a decade or more.
But the group put the Minimal Genome Project on hold in 1999, while Venter focused on sequencing the human genome. Now he has revived the project-this time with a specific application in mind: creating artificial bacteria that could help provide cleaner sources of energy. Under the guidance of Nobel Laureate Hamilton Smith, who left Celera last fall to become the scientific director of the Institute for Biological Energy Alternatives, researchers will try first to build a minimal genome and place it inside a bacterial cell whose own genome has been destroyed. The researchers’ hope is that the synthetic genome will take over, and a new life form will be born. If they succeed with this preliminary experiment, the researchers will then begin to create organisms whose minimal genomes will be supplemented with additional genes that provide instructions for metabolizing carbon dioxide, say, or producing hydrogen fuel.
Not only does designing genomes from scratch allow researchers to engineer new organisms with extreme precision, Venter says, it also allows them to strip the cells of a host of natural functions needed to survive and reproduce in the wild. As a result, synthetic organisms would function only under tightly controlled or rarified conditions such as those inside a biological pollution filter on the smokestack of a fossil-fuel-burning power plant.
“By our back-of-the-envelope calculations,” Venter says, “it wouldn’t be that complicated to have a carbon dioxide scrubber, directly associated with those power plants, that captures all the carbon dioxide and converts it into something economically useful”-like a plastic. Or researchers could engineer bacteria that use methane from waste sites, for instance, to produce hydrogen fuel. “As far as I know, there is no existing organism that can either capture carbon dioxide or produce hydrogen efficiently enough with its existing metabolism to make it economically feasible,” says Venter.
For now, synthetic bacteria with custom-made metabolisms exist only on the blackboard, and it could take a decade before the chalk dust gives way to living creatures. But there is growing evidence that, just maybe, it could happen.
Last summer researchers from the State University of New York at Stony Brook proved that it’s possible to synthesize an organism-albeit one found in nature-from nothing more than genome sequence information and chemicals in a test tube. Molecular biologist Eckard Wimmer and his colleagues used a combination of DNA synthesizers and brute force to reconstruct in DNA the 7,500-letter-long polio genome sequence. Next, they converted that sequence into RNA, a biochemical cousin of DNA that makes up many viral genomes, combined that RNA with enzymes and other molecules in a test tube, and watched as whole polio viruses assembled spontaneously. It was the first time scientists had ever synthesized a virus-or any other organism for that matter-from scratch.
When the work was announced, many scientists called it a publicity stunt, arguing that the Stony Brook team had chosen to build a dangerous microbe in a bid for headlines. But the work did highlight a very real technological problem: assembling long strands of DNA by conventional means is an almost prohibitively time-consuming task. Researchers like the ones at Stony Brook first synthesize short fragments using conventional DNA synthesizers. Such machines use a complicated series of chemical reactions to attach each DNA letter to the next. Because errors can occur at each step, the longer the fragment, the more errors it contains; so researchers typically limit fragments to fewer than 80 letters. To assemble longer DNA strands, they toss the fragments sequentially into test tubes, together with enzymes that link the fragments end to end. This process introduces a multitude of tiny, single-letter errors though. Detecting and correcting those errors adds more work and more time to the job. Had the Stony Brook team chosen an organism with a genome longer than 7,500 letters, it might still be working on the project.
But that might be about to change. Glen Evans, CEO of San Diego, CA-based Egea Biosciences, thinks his company has a solution-one that could help propel the idea of engineering wholly new organisms into reality. “It took the polio researchers two years to synthesize the virus,” says Evans. “We could have done that in less than a week.”
The source of Evans’s bravado is his company’s newly developed machine, which can rapidly synthesize long strands of DNA with relatively good accuracy: the device makes a mistake only once for every 10,000 DNA letters, or bases, Evans says, whereas conventional techniques typically have an error rate of one in 100. Right now Evans’s DNA-writing machine is accurate enough to make several genes at once, but he hopes to get the error rate down low enough to make the larger DNA strands that are required for building entire genomes. Evans first conceived of the technology at the University of Texas Southwestern Medical Center, where he was director of the Human Genome Sequencing Center. When researchers completed a draft of the human genome, he says, “we realized we had read the genetic code, but we didn’t have the ability to write the genetic code.”
To fill that gap, he built a gene-writing system that combines hardware and software. Using the software, a researcher can design a protein-a new drug, for instance-on a computer, which in turn determines the sequence of DNA bases needed to encode the protein. “It’s kind of like word processing for DNA,” says Evans. The hardware, essentially a robotic chemistry lab, assembles long stretches of DNA automatically, circumventing what would otherwise require endless hours of tedium for humans. The machine first synthesizes fragments of the gene, each measuring 50 to 100 letters long, and arrays the fragments into tiny wells. It then grabs each fragment in sequence, attaching one to the next with a customized cocktail of enzymes and ultimately producing the whole gene, which can amount to a few thousand bases. Evans says Egea has developed a prototype machine that, thanks to automation, can synthesize 10,000 bases in just two days. He says the technology could be extended to yield in a matter of weeks highly accurate strands 100,000 bases in length-long enough to make a very simple bacterial genome. Automation and robotics also allow for careful control of each chemical reaction in the process. That control, combined with the enzyme cocktail, helps keep long DNA strands largely free of the errors that plague conventionally made strands.
Though Venter and other researchers bent on creating synthetic organisms will still have a lot of scientific heavy lifting to do before they’re able to design new genomes readily, technology like Egea’s, says MIT’s Knight, could lighten the burden of genome construction. And a handful of biotech companies are now also getting into the business of souped-up DNA synthesis. “Pretty soon, we won’t have to store DNA in large refrigerators,” says Knight. “We’ll just write it when we need it.”
Playing with Blocks
While Evans and others are working on machines that could expand researchers’ ability to write genes, chemists at the Scripps Research Institute in La Jolla, CA, are expanding the genetic alphabet itself. “Our repertoire of bases is naturally limited,” to the familiar DNA letters A, T, C, and G, says Scripps chemist Floyd Romesberg. Because these letters tell an organism which proteins to make, the types of proteins that can be specified by the genome are limited as well. Getting, say, a bacterium to make novel types of proteins would require adding new DNA letters.
That’s exactly what Romesberg’s lab has done. Building on the pioneering work of biologist Steven Benner at the University of Florida, Romesberg and his colleagues have created a letter in the form of the chemical fluorobenzene. This artificial DNA letter looks nothing like a natural one, he says, so the challenge is to trick the cell’s DNA replication and translation machinery into recognizing it. So far, the Scripps researchers have synthesized short fragments of DNA that incorporate the new letter and have successfully created an enzyme that can replicate the modified code. The next step is to design a system for translating the code into a completely unnatural protein-a novel drug, for instance.
To do this, Romesberg is collaborating with another Scripps chemist, Peter Schultz. While Romesberg’s team was rejiggering the DNA alphabet, Schultz’s lab was tinkering with another set of biological building blocks: the amino acids that form proteins. Living things use 20 amino acids, which are strung together as proteins, following instructions encoded in the DNA. Schultz’s group created a bacterium that has 21, the 21st being a chemically modified version of a natural amino acid. Such synthetic amino acids offer the chance to build new functions into proteins. “There’s a huge range of chemical groups that we could put into proteins to make them do interesting things,” says Schultz. He is, for instance, working on creating photosensitive amino acids which, in response to light, could trigger specific reactions in a cell.
What’s more, the two Scripps teams are working to combine their techniques. The goal: to create cells with all the enzymes and other molecules necessary to translate DNA code that bears Romesberg’s artificial letter into proteins that incorporate Schultz’s artificial amino acids. The technology could have a huge impact on the development of new protein therapeutics, says Schultz. Protein drug researchers typically modify natural proteins-adding a specific sugar that binds to a cancer cell, for example-to increase their effectiveness. “But what these people are doing is kind of dirty chemistry,” says Schultz. Treating these extra chemical groups as artificial amino acids and directly encoding them in a synthetic gene would enable researchers to modify proteins with incredible selectivity and simultaneously create living factories that churn out the new proteins.
“We’ve removed a billion-year-old constraint on what we can do with proteins,” says Schultz. “And so we’re taking the point of view that if God had worked on Sunday, and he had more amino acids to work with, what would have been the outcome?” Would an organism with an expanded genetic code and amino acid inventory have an evolutionary advantage? Perhaps there is a reason why all known organisms share those 20 building blocks. “Is it just a chance of history that early life took this route?” asks Stephen Freeland, an evolutionary geneticist at the University of Maryland, Baltimore County. “Or is there more to it?”
If scientists could answer such big theoretical questions, Freeland says, it might be possible one day to discover on other planets life that might not otherwise be recognizable. And if the synthetic-genome technologies in the works at Scripps, Egea, Venter’s institute, and elsewhere pan out, life right here on Earth could soon look a little less familiar-and a lot more diverse.
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
These materials were meant to revolutionize the solar industry. Why hasn’t it happened?
Perovskites are promising, but real-world conditions have held them back.
Why China is still obsessed with disinfecting everything
Most public health bodies dealing with covid have long since moved on from the idea of surface transmission. China’s didn’t—and that helps it control the narrative about the disease’s origins and danger.
A quick guide to the most important AI law you’ve never heard of
The European Union is planning new legislation aimed at curbing the worst harms associated with artificial intelligence.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.