Earlier this year, out in the remote test farms of North Dakota, researchers sprayed weedkiller on canola, a delicate golden-flowered plant used for cooking oil. A touch of the herbicide would have killed most plants, but the canola refused to wilt. It survived to late summer harvest.
It was not an uncommon sight in modern farming. Like many U.S. crops, the canola had been engineered to survive a heavy dose of weedkiller. But this stubborn plant was different from most genetically modified crops, which carry copies of bacterial genes in their DNA. The canola survived by tapping only the natural gifts of its genetic code -- or, rather, a lightly "edited" version of its genetic code.
Created by the biotech firm Cibus LLC, this herbicide-tolerant canola is likely to be the first in a wave of crops created by targeted mutation, a long-sought technique that allows tailored changes in plant genes, down to single pairs of DNA. The technology is poised to upend the debate on modified crops, forcing regulators and the public to face a simple question: What does "genetically engineered" mean?
Approaches vary on targeted mutation, with Cibus using an older and controversial technique compared to its peers, which draw on bleeding-edge approaches to human gene therapy. But while the technologies diverge, at their most fundamental they all resemble altering one letter in one word of the newspaper, said Peter Beetham, Cibus' scientific head.
Flipping that one letter, he said, "potentially changes the meaning of the whole paragraph."
Though far from public notice, invoking random crop mutations is a time-tested technology. Throughout the past century, breeders have used harsh chemicals and radiation to mutate the cells of food staples like wheat; the reaped improvements were an essential part of the Green Revolution. Despite these gains, though, mutation has remained a random, inefficient process, mostly producing crops riven with lesions and genetic flaws, drowning out the rare improvement.
Targeted mutation, also known as genome editing, changes this dynamic. Over the past several years, a clutch of small biotech firms has developed tools that allow scientists to induce errors in DNA repair -- such mistakes are the source of mutation -- with great specificity. These tools, which use complex protein structures called zinc fingers or meganucleases, can also selectively insert or silence genes in crop species like corn. Combined, they will knock years off development time, scientists say.
"These tools are going to be widely used to study plant gene function, to modify multiple [genetic] pathways and to understand how plant genes work," said Dan Voytas, a biologist at the University of Minnesota and one of the leading developers of genome editing. "We're at a true inflection point in how we do plant biotechnology."
In perhaps the best measure of genome editing's spreading influence, the advance has attracted the attention and capital of the world's largest biotech companies, including Monsanto Co., Dow Chemical Co., Dupont Co. and BASF SE. Over the past few years, these chemical and seed companies have conducted a quiet arms race, partnering with startups like Sangamo Biosciences and Precision Biosciences to snatch up licenses to the companies' technology.
In the industry, expectations are high that genome editing, building off the reams of gene function data pouring into private and public databases, will reorder their business. As Vipula Shukla, one of Dow's leading scientists, wrote last year in the journal Nature, these technologies make "precise modification of native genomes in plants practical and feasible for the first time. This approach ... establishes an efficient and precise strategy for plant genome engineering."
The technology, licensed from Sangamo, is well beyond proof of concept, Shukla added in an interview.
"It's something that's very much embedded in our efforts to develop all kinds of traits," she said. Dow is using it on cash crops -- the Nature paper applied genome editing in corn -- and has signed deals to apply it in potatoes, tomatoes and cassava, among other plants. "We are using it wherever and whenever it makes sense from product development," Shukla said.
Many scientists have expressed hope that these new genetic tools could spur the public to embrace, or at least not fear, the bioengineering of plants, since the crops would not carry the "yuck" factor of foreign genes. (A frost-tolerant tomato carrying a fish gene, proposed in the early 1990s, is still widely cited, though it was never sold.)
Such hopes misread public sentiment, said Janet Cotter, a Greenpeace scientist based at the University of Exeter. The public's objections are simple, she said: "They don't like people meddling with DNA."
The distinction between targeted mutation and traditional biotech crops is too fine, added Jeff Wolt, an agronomist at Iowa State University. If it looks like biotechnology and it acts like biotechnology, it will be mixed up into the current debate, he said.
"I don't think the public understands the subtleties," he said.
As targeted mutation plows forward, crops developed with it are bound to run up against a U.S. regulatory system that has been based on one technology: the somewhat random insertion of largely bacterial DNA into plants.
Genome editing could instead see crops that have had multiple native genes slightly altered, targeting complex qualities like their water use. Without carrying foreign genes, such plants could face little to no regulatory oversight.
For instance, the U.S. Department of Agriculture, which controls growth of genetically modified crops, concluded six years ago in a letter, seen by Greenwire, that it had no authority to regulate crops generated with "mutagenesis techniques" like those employed by Cibus. The firm has faced no limits on its field trials for its herbicide-tolerant canola and will likely be able to sell the crop without facing the USDA controls that have regulated bioengineered crops for more than a decade.
Indeed, regulators in the United States and Europe are bracing themselves for the explosion of genetic approaches set to unfold in plants. Beyond genome editing, plants could have expression of their genes silenced with RNA -- a practice already used in soy and papaya -- or scientists may use genetic tools to insert DNA from plants in the same species, avoiding the randomness of breeding (Greenwire, Dec. 21, 2009).
For several years, USDA has been reworking its rules for bioengineered crops. It remains a mystery whether or how it will choose to regulate genome-edited crops, but the department is "considering where zinc-finger technology falls within our regulatory authority," said Andre Bell, a spokesman.
The possibility of a light regulatory load -- and access to a European market still conflicted about traditional genetically modified crops -- is "one of the real attractions of our technologies to our partners, especially those in Europe," said Matthew Kane, CEO of Precision Biosciences, which has licensed its editing technology to Dupont and, earlier this month, to BASF, the German chemical giant.
Looking simply at the technology of genome editing, less regulation would seem appropriate, Iowa State's Wolt said. "Using biotechnology that doesn't insert a foreign gene means that we should be simplifying, minimizing or eliminating a lot of these regulatory requirements," he said. Given public opinion, however, USDA officials "have their hands somewhat tied in what they can do," he added.
Given the regulatory uncertainty on both sides of the Atlantic, the large crop firms have not committed to the notion that edited crops should skirt regulators. Indeed, it is likely Dow and BASF will use targeted mutation to improve crops that also carry bacterial genes, falling squarely under existing rules.
Still, there is an optimism that genome editing could roam free, said Voytas, who is also the chief scientist at Cellectis Plant Sciences, a firm that recently began collaborating with Monsanto.
"Conservatively, people are saying if we deliver a nuclease protein ... that's not so different than a [mutation-causing agent]," he said. "We've not genetically altered the cell that's receiving it."
There will be difficult distinctions for regulators to untangle, though, since multiple, single-letter changes can be introduced with genome editing, Voytas said. Perhaps one mutation is fine, but scientists will be able to cause three, five or 10 letter changes in genes.
"There's a threshold," he said. "When do you have a new gene?"
'New phase of what's possible'
Engineering plants has long been a messy affair. While researchers can have their way with flexible microbes, inserting DNA into plants relies on technology pioneered decades ago. Using bacteria or gene guns -- the latter blast plant cells with DNA-encrusted metals -- scientists can only insert DNA at random. With luck, the genes do not interfere with existing DNA, and the crop is cultivated.
To Cibus' Beetham, splicing foreign DNA into plants seems inelegant, a "transition technology."
Plant genomes are like grandfather clocks, he said, and throwing a foreign gene into the genome amounts to adding a fast-spinning cog "that throws off everything." Genome editing, in Beetham's analogy, would instead tweak the bite in one cog's tooth.
Cibus' approach to mutation is controversial. The technique exploded onto the scientific scene in the late 1990s with high -- and failed -- promises for gene therapy; it has been freighted with skepticism ever since.
During those heady days, the technology, developed in the lab of Eric Kmiec at Thomas Jefferson University in Philadelphia, looked miraculously simple: Introducing a hybrid DNA-RNA template would cause a small number of cells, during division, to copy genetic code off the template, causing mutation.
More than 10 years later, Cibus' approach remains similar. The firm injects single strands of synthesized DNA into a cell. The strands match the code of one genome section, except for one intentional mistake. When the cell divides, on rare occasions the template lines up with its targeted DNA. Repair enzymes sense a mismatch and swoop in to "fix" the DNA, copying off Cibus' template and creating a permanent mutation. Cibus' single-strand DNA, its job done, is then broken apart by the cell.
"The only change is a single mutation," Beetham said. "It's a much clearer and safer product."
The tools used by Precision, Cellectis and Sangamo are more proactive, relying on customized protein molecules that recognize genetic strands 20 or so letters in length -- a signature long enough to imply a unique position -- and there break apart the DNA, rather than waiting for cell replication. Then, a DNA template similar to Cibus' can be used, or scientists can wait for the repair enzymes to make a mistaken mutation themselves. Either way, an inheritable change is made at high rates, and no foreign genetic material is left behind, according to Precision's Kane.
"Once the [protein] has made its cut, it gets out of the way," he said.
There are limits to what can be done with genome editing, since it is restricted to variations on genes already in a plant. For example, one pillar of modern agricultural biotechnology is the Cry bacterial gene, which produces a protein toxic to select insects and harmless to humans. Used widely in cotton and corn, the gene is wholly foreign to plants. It could not be created by mutation.
The true potential for genome editing, beyond more efficient hypothesis testing, involves attributes that cannot be tied to a single gene: traits like height, drought tolerance and photosynthesis. As more genes are linked to these traits, targeted mutation can be used to increase or limit their expression, the type of wholesale bioengineering that is currently more common in bacteria.
"Those more complex changes can be considered now and can be accomplished in a relatively quick period of time," Kane said. "We are entering a new phase of what's possible."
Several environmental groups that have long been opposed to genetically modified crops, like Greenpeace and the Center for Food Safety, are conflicted when it comes to this new phase. They welcome the notion that more precise tools will lower the roulette-type insertion of genes in modified crops. However, they are uncertain if their campaigns against edited crops will mirror their past opposition.
Since bioengineered crops began to be sold more than a decade ago, the debate has pivoted on whether or not a plant carries foreign genes, said Bill Freese, science policy analyst at the Center for Food Safety. "Frankly, I'm getting a little tired of that [debate]," he said.
The center's concern with Cibus' canola, and future edited plants, is that herbicide resistance will remain a prime goal of biotech companies, particularly if such resistance can be imparted without USDA rules. Already, a number of weeds have developed tolerance to glyphosate, the popular herbicide that is routinely overused on transgenic crops, the National Academy of Sciences warned earlier this year.
These crops "demonstrably lead us exactly the wrong way," Freese said.
Greenpeace, meanwhile, is unequivocal in considering edited crops genetically modified. Inserting any heritable material into the cell, even a single-strand piece of DNA unable to integrate into the genome, should qualify the crop as engineered, Greenpeace's Cotter said.
"It's tinkering around with an approach that's still working against nature," she said.
Matter of time for edited crops?
Whether it is working with or against nature, Cibus has grown herbicide-tolerant canola in North Dakota, where most of the crop is already modified to resist Monsanto's Roundup. The company has bred its mutated canola, designed to resist a class of sulfonylurea weedkillers, into high-yielding varieties. It plans to sell the crop in little more than a year, partnering with the seed distributor BrettYoung.
All of Cibus' successful mutations have come in developing weedkiller-resistant crops. It has deals to develop similar traits in potatoes and flax, and last year it received $37 million from Makhteshim-Agan, an Israeli herbicide company, to develop five "major" crops for the European market.
The company has also seen plants resist its mutation efforts. Its plans for sorghum have faltered, for example. Not all crops will necessarily work with the Cibus' technology, Beetham said. And some outside scientists are skeptical it can target any traits beyond herbicide resistance.
One of Beetham's past collaborators, Chongmei Dong, a plant geneticist at the University of Sydney, believes the company's mutations are not a result of their DNA templates at all, but rather the natural rate of mutation seen in cultured plant cells. By exposing these cells to herbicide, it is simple to find resistant plants -- just look for the surviving cells. When Dong tried using Cibus' techniques in embryonic wheat cells five years ago, the mutation would not take.
"I started thinking, '[Does] this technology really work?'" Dong said.
Beetham, who worked at Kimeragen, the failed company that rose out of Kmiec's lab, published one of the earliest studies using DNA-RNA templates in 1999, inducing weedkiller resistance in tobacco. Since then, his team has plugged away for 10 years, improving the template and the screening used to find the minute number of successfully induced mutations. The mechanism is clear to Beetham, and Cibus understands the timing of DNA repair well enough to target complex traits, he added.
"You can circumvent selection," Beetham said. "That's a challenge. I'm not saying that it's a slam dunk. But we have the tools now that can go after that challenge."
It may be scientifically impossible to be certain, given published data, whether Cibus is creating its mutations or simply picking out accidental mutations it can use. KeyGene, a decades-old Dutch biotech company, has begun using a process similar to Cibus, targeting weedkiller resistance in vegetables. Beetham said the company's data are clear that they are generating mutations above the background levels.
Still, Dong believes that it will be zinc fingers and similar molecules that will lead the field in the future, a sentiment the major biotech firms seem to share, given their recent deals.
Researchers are racing to publish high-profile papers indicating high efficiencies -- mutations that are successful 10 percent of the time, a previously unheard success --- and the cutting mechanisms are well-understood. It is only a matter of time before the edited crops arrive, scientists say.
Given the demands that will be placed on agriculture in the coming decades -- slowly supplementing and replacing fossil fuels in the economy, and feeding a rising population -- these tools could not have come soon enough, Minnesota's Voytas added.
"These new technologies allow you to do things with much more precision, with much more accuracy and with much more certainty," Voytas said. "You know what you created."