Skip to Content

The Next Great GMO Debate

Deep inside its labs, Monsanto is learning how to modify crops by spraying them with RNA rather than tinkering with their genes.

The Colorado potato beetle is a voracious eater. The insect can chew through 10 square centimeters of leaf a day, and left unchecked it will strip a plant bare. But the beetles I was looking at were doomed. The plant they were feeding on—bright green and carefully netted in Monsanto’s labs outside St. Louis—had been doused with a spray of RNA.

The experiment took advantage of a mechanism called RNA interference. It’s a way to temporarily turn off the activity of any gene. In this case, the gene being shut down was one vital to the insect’s survival. “I am pretty sure 99 percent of them will be dead soon,” said Jodi Beattie, a Monsanto scientist who showed me her experiment.

The discovery of RNA interference earned two academics a Nobel Prize in 2006 and set off a scramble to create drugs that block disease-causing genes. Using this same technology, Monsanto now thinks it has hit on an alternative to conventional genetically modified organisms, or GMOs. It can already kill bugs by getting them to eat leaves coated with specially designed RNA. And if the company succeeds in developing sprays that penetrate plant cells, as it’s attempting to, it could block certain plant genes, too. Imagine a spray that causes tomatoes to taste better or helps plants survive a drought.

Monsanto isn’t the only one working on genetic sprays. Other large agricultural biotech companies, including Bayer and Syngenta, are also investigating the technology. The appeal is that it offers control over genes without modifying a plant’s genome—that is, without creating a GMO.

That means sprays might sidestep much of the controversy around agricultural biotechnology. Or so companies hope. What’s certain is that a way to accomplish the goals of genetic engineering without having to develop a GMO could bring commercial rewards. Sprays might be quickly tailored to do battle with an insect infestation or a new type of virus. Not only could this be faster than creating new GM crops, but the gene-silencing effects of RNA interference last only a few days or weeks. That means you might spray on traits such as drought resistance in times of water shortage without affecting the plant’s performance in times of normal rainfall.

Beattie showed me a large glass jar in which dried, purified RNA glistened like crumbled packing peanuts. A few years ago, this much RNA might have cost $1 million, one reason few would have thought to spray it from tractors rumbling through rows of corn. But the cost of making RNA has plummeted. Monsanto estimates that it now costs $50 a gram. A tenth that amount, the company says, is potent enough to kill 100 percent of beetles on an acre of plants.

Monsanto has spent millions learning how to control plant traits using genetic sprays. Opponents see a new risk.

At Monsanto I met Robb Fraley, the company’s chief technology officer, who oversees a research staff of 5,000. Three years ago Fraley designated the RNA sprays as one of Monsanto’s new areas for product development. He thinks that within a few years they will “open up a whole new way to use biotechnology” that “doesn’t have the same stigma, the same intensive regulatory studies and cost that we would normally associate with GMOs.” He’s told people he thinks the tools are “incredible” and “breathtaking” and that “of all the platforms we are working on, this is the one that reminds me the most of the early days of biotech.”

It was Fraley who made Monsanto’s first GM plants in the 1980s—petunias resistant to a plant poison. Today, Monsanto has revenues of about $9 billion a year from GM seeds for crops that produce the insect toxin Bt or resist the weed killer Roundup. GM corn, soy, and cotton plants now spread across 180 million hectares. And it has generated a public controversy just as vast. To its strongest critics, the company is simply “Monsatan.”

But with the RNA spray technology, which Monsanto calls BioDirect, the company may have found something that will bedevil opponents. The sprays are made from a ubiquitous molecule that degrades quickly in soil. They can be genetically precise enough to kill potato bugs but spare their ladybug cousins. And so far, consuming RNA molecules appears no more toxic to people than drinking a glass of orange juice. As Monsanto put it in a letter to U.S. regulators, “humans have been eating RNA as long as we have been eating.”

Public opposition, regulations, and the slow pace of plant breeding mean that on average, bringing a new GM crop to market costs more than $100 million and takes around 13 years. But imagine you wanted to fight a plant virus, says James Carrington, head of a Missouri nonprofit called the Danforth Plant Science Center and an advisor to Monsanto. “If you can gain control with a spray, you can envision a product that can change very rapidly, that you can test faster, experiment with faster, and bring to market faster,” he says. “You could respond to issues as they arise.”

Not everyone is convinced, though, that applying RNA will be commercially feasible or any less controversial than genetic modification. “The public is not accepting GMOs, and this could be more alarming. People are going to say you are taking the RNA and spraying this in the open,” says Kassim Al-Khatib, a plant physiologist at the University of California, Davis. “The acceptance of biotech has to be there before you can deliver another approach. This isn’t a technology for tomorrow. It’s for the day after tomorrow.”

Sprays might be quickly tailored to do battle with an insect infestation or a new type of virus.

When I met Fraley, he didn’t deny that there are obstacles—in fact, that’s what reminds him so much of biotech’s early days. He says no one yet understands exactly how to get RNA inside a plant’s cells using a field sprayer—at least not with the sort of inexpensive, works-every-time efficiency farmers would be looking for. Many insects are also not easily affected. Monsanto has been spending millions to crack these problems, collaborating with biotech companies specializing in drug delivery. “We’re still a few breakthroughs away,” he says.

Weed control

The cells of plants and animals carry their instructions in the form of DNA. To make a protein, the sequence of genetic letters in each gene gets copied into matching strands of RNA, which then float out of the nucleus to guide the protein-making machinery of the cell. RNA interference, or gene silencing, is a way to destroy specific RNA messages so that a particular protein is not made.

Top: A Colorado
potato beetle.

Middle: Oranges affected by citrus greening disease.

Bottom: A jar of purified RNA on display at Monsanto.

The mechanism is a natural one: it appears to have evolved as a defense system against viruses. It is triggered when a cell encounters double-stranded RNA, or two strands zipped together—the kind viruses create as they try to copy their genetic material. To defend itself, the cell chops the double-stranded RNA molecule into bits and uses the pieces to seek out and destroy any matching RNA messages. What scientists learned was that if they designed a double-stranded RNA corresponding to an animal or plant cell’s own genes, they could get the cells to silence those genes, not only those of a virus.

Some GM plants already use RNA interference to disable unwanted enzymes, or to kill viruses or pests. The Flavr Savr tomato—the first genetically modified crop to be approved in the United States, back in 1994—harnessed the mechanism to block an enzyme that makes tomatoes soft, so they could ripen longer on the vine. Like Monsanto’s Roundup Ready cotton and corn, the Flavr Savr was a GMO. Its seeds have an extra gene that manufactures a specific RNA molecule. Since then, companies have engineered a few other plants to take advantage of RNA interference. This year a Granny Smith apple genetically modified to silence a gene that turns apple slices brown won clearance from regulators. Before that, the Hawaiian papaya industry was saved by plants engineered to produce RNA that defends against the ringspot virus. And Monsanto is awaiting approval to sell corn plants that use RNA interference to kill the western corn rootworm. That plant is the first GMO to incorporate an insecticidal RNA into its genetic makeup.

But what if you could just spray the RNA on instead of tinkering with a plant’s genome? A chemist named Doug Sammons was the first person inside Monsanto to have the idea. He studies weeds that have become resistant to glyphosate, the herbicide that Monsanto markets as Roundup. These weeds have become a huge problem for farmers and for Monsanto. Sammons determined that some resistant weeds have as many as 160 extra copies of a gene called EPSPS. That’s the very enzyme glyphosate interferes with, blocking plant growth. The super-weeds had found a trick to overwhelm the herbicide.

Sammons thought the weed’s extra genes could be knocked back into line with RNA interference. The problem was that since weeds are wild, Monsanto didn’t have any way to control their genetic makeup, as it could with a corn plant. “So he came to us and said, Why don’t we just spray it on a plant? We were like, ‘Really?’” says Gregory Heck, a research manager at Monsanto. “We’d only thought of [GMOs] until that time.”

It seemed unlikely to work—but it did, according to Monsanto. In lab tests and at a roadside plot in Illinois that’s been overrun by weeds, a mixture of Roundup and double-stranded RNA coded to match the EPSPS gene made resistant weeds wilt. According to Monsanto’s patents, the technique also involved spraying a silicone surfactant that let the RNA molecules slip into air-exchange holes in the plant’s surface. Somehow, soaking the leaves with RNA caused the silencing effect to spread through the entire plant, affecting it long enough to let the herbicide take hold.

The technology could give Monsanto a new, exclusive formulation of Roundup (which lost its original patent several years ago) and help deal with the troublesome weeds, which have spread across U.S. farmland. “It’s definitely a prize if you can reënable glyphosate,” says Heck. But the company’s scientists saw that it could do much more: they could theoretically reach in and temporarily block any gene in any crop. “It could be a weed or a corn plant,” says Lyle Crossland, a senior program manager at Monsanto. “You could just dial in the sequence information. You could turn off the gene that makes fruits brown; you could do something with drought tolerance, photosynthesis. We have a lot of probing going on.”

“It’s a way of elegantly targeting particular genes and turning those genes off. And there are undesirable trait genes in everything.”

Some plant experts aren’t convinced it’s practical yet. Stephen Powles, director of the Australian Herbicide Resistance Initiative and a professor at the University of Western Australia, told me he’d had a “bit of a go” at repeating Monsanto’s weed experiment but hadn’t been able to make it work. “Getting double-stranded RNA sprayed onto plants and getting it into plants, and killing a plant, is not easy, and in fact it’s very, very difficult,” he says. “There’s the formulation technology, the shelf life, and can it bounce around in the back of a pickup for a week at 110 °F.”

Richard Jorgensen, a plant biologist who was the first to observe RNA interference, thinks modifying traits with a spray “might be really patchy” compared with a true GMO. Say you wanted to turn flowers a specific color. “Would you spray it every week and hope it gets into every cell in the plant bud? I think there are lots of limitations compared to [GMOs],” he says. To Powles, however, the idea of spray-on traits has strong appeal. “It’s a way of elegantly targeting particular genes and turning those genes off. And there are undesirable trait genes in everything,” he says.

Skunk works

After the weed discovery, which occurred in 2010, Monsanto began spending heavily to build a position in RNA technology. It took over a company called Beeologics, which had found a way to introduce RNA into sugar water that bees feed on in order to kill a parasitic mite that infests hives. That company had also come up with a much cheaper way to make RNA.

Monsanto also began trying to crack the problem of getting RNA into plants more efficiently. It paid $30 million for access to the RNA interference know-how and patents held by the biotech company Alnylam, and it did a similar deal with Tekmira, an RNA delivery specialist based in Burnaby, British Columbia. Monsanto is also the financial backer of a 15-person company called Preceres, a kind of skunk works it established just off the campus of MIT, where robotic mixers are busy stirring RNA together with coatings of specialized nanoparticles.

The startup was created by drug delivery specialists, including MIT professors Daniel Anderson and Robert Langer, who have spent a decade learning how to get RNA drugs into human cells—a problem so difficult it almost derailed the idea of such medicines. Anderson told me the crop project faces substantial difficulties, too. “It’s easier to envision if you are injecting a person in their veins, but if you are spraying out of a plane, that would be a whole different set of challenges,” he said. “We don’t have to worry about wind currents with drugs.”

The basic task at Preceres is how to get a large, electrically charged molecule like RNA to move through a plant’s waxy cuticle and into its cells. To do it, researchers there are working to encapsulate the RNA in synthetic nanoparticles called lipidoids—greasy blobs with specialized chemical tails. The idea is to slip them into a plant, where the coating will dissolve, releasing the RNA. Formulations get shipped out to St. Louis for testing in greenhouses.

Roger Wiegand, the company’s CEO, says the company is also trying to kill insects that aren’t as easily affected by RNA as the potato beetle. “There are insects that just laugh at naked double-stranded RNA,” he says. Those include a caterpillar now infesting Brazil’s soybean crops. He says some of the formulations get tested for endurance in caterpillar spit that Monsanto sends to Cambridge.

If they are able to sort out the delivery problems, Wiegand thinks, RNA sprays will be “a big frickin’ deal” and “a breakthrough at the same level GMO plants were.” Yet so far, only a few scientific publications even mention the idea of RNA sprays. That makes it hard to judge companies’ claims. And many aren’t talking at all. Bayer declined to comment on its research program. So did Syngenta, which in 2012 paid $523 million to acquire Devgen, a European biotech with which it had worked on RNA insecticides.

One project I did learn about is led by Nitzan Paldi, an Israeli entrepreneur who’d been a cofounder of Beeologics. His current startup, called Forrest Innovations, is investigating a solution to citrus greening disease, a blight that’s destroying Florida’s citrus industry and is also present in Brazil. Caused by bacteria spread by an invasive insect called the Asian citrus psyllid, it leaves oranges hard and discolored, with juice the flavor of jet fuel. Last year, 22 percent of oranges in Florida suddenly fell off the trees.

Paldi isn’t willing to disclose exactly how he’s applying the RNA, but he did say he’s hoping to block genes involved in the trees’ reaction to the bacteria. It’s their immune response to the infection that causes the greening symptoms. If the treatment works, Paldi believes, an RNA intervention could sail past regulators. With growers desperate, and the prospect of no more Florida orange juice, the public may be open-minded too. “We are potentially riding in on the horse and saving the day,” he says.

Killer match

At Monsanto, the effort to develop an RNA spray to kill potato beetles has overtaken the weed idea. It could reach the market by 2020, says Jeremy Williams, a Monsanto geneticist who directs the insect program. The company has settled on a gene target and has begun efforts to make the spray rainproof so it grips the plant leaf and doesn’t wash away for at least a week.

One reason the potato beetle is an interesting target for RNA sprays is that it’s famous for becoming resistant to conventional insecticides. Since 1952, it’s developed resistance to more than 60 of them, starting with DDT. But RNA interference is a means of attack that Williams doesn’t think will be easy to overcome. If the beetle does evolve to resist an RNA molecule, he says, geneticists could easily launch a new assault: just “slide the sequence over” by a few letters or target several genes at once.

“Humans have been eating RNA as long as we have been eating.”

Monsanto has also been interested in the problem facing orange growers. It collaborates with Wayne Hunter, a spiky-haired entomologist at the U.S. Department of Agriculture’s research laboratory in Fort Pierce, on Florida’s Atlantic coast, where grapefruit and orange orchards are affected by citrus greening disease. With assistance from Monsanto, Hunter has been trying to kill the psyllid insect with RNA. He toured me through a plot of 100 orange trees, explaining that he’d drenched their roots with RNA or injected it into their trunks. Hunter’s most interesting result is that orange trees seem to soak up double-stranded RNA and hold onto it. He applies a relatively huge dose to each tree, about 200 milligrams, and finds traces of the molecules still in their canopies three months later.

In Hunter’s lab, psyllids were feeding on cuttings from trees resting in cups of liquid spiked with double-stranded RNA. Hunter was testing specific sequences that match crucial genes in the insect. One, which codes for arginine kinase, interferes with its ability to make energy.

Before picking a target, scientists can sift through online archives of DNA data to avoid matches with the genes of friendly insects, like honeybees. It takes an exact match of about 20 consecutive genetic letters for RNA interference to work. The resulting double-stranded RNA molecules, usually about 200 letters long, are then fed to other species, including bees, aphids, and whiteflies, as a practical test for “off target” effects. Monsanto has found that its sequences—which it calls triggers—usually don’t affect any but the most closely related species, bugs in the same genus. “The differences are genetic,” says Hunter. “The genes of insects are not identical. If it does not match, it does not kill.”

In contrast, conventional insecticides wipe out helpful insects along with the bad ones. To stave off the greening disease, growers in Florida have been applying such chemicals as often as every two weeks. One, imidacloprid, is restricted in Europe for its suspected link to bee colony collapse. “We’ve just got to get away from hard-core pesticide use,” says David Hall, leader of the subtropical-insect research unit that Hunter works in.

So far, it looks as though RNA treatments would be at best an adjunct in the orange groves, not a silver bullet. RNA doesn’t knock bugs out instantly, as a chemical neurotoxin does. In Hunter’s lab, insects only start dying after four days, and some live two weeks. “It’s a biopesticide—it takes longer,” he says. Perhaps partly for that reason, the field study of 100 trees supported by Monsanto yielded ambiguous results. The trees remained covered with psyllids, but they might have flown in from elsewhere. Hunter is planning to try again in a large enclosed greenhouse where he can apply RNA to every tree, mimicking what would happen if growers used an “area-wide” application.

Meanwhile, growers are trying anything. Some grind up infected trees. There is also a GM tree that’s resistant to the blight, thanks to an added gene from a spinach plant. But even if consumers accepted GM orange juice, those trees couldn’t be planted fast enough to replace the millions of sick ones in Florida’s groves. Hunter’s RNA molecules probably won’t arrive soon enough either. “We are still 10 years away,” he says. “That is a problem with this technology. Around here, there is an enormous amount of pressure to come up with a solution.”

Big questions

People on Monsanto’s public relations staff told me they hoped to communicate better on RNA sprays than they had on GMOs. (Visitors to the company’s offices can pick up a handout titled “12 Myths about Monsanto”; number 1 is the rumor that it bars GMOs from its own cafeteria.) Until now, the sprays have been too deep in the R&D pipeline to attract the attention of GMO opponents. But plants genetically engineered to use RNA silencing have drawn attacks. In 2012, the Safe Food Foundation in Australia alleged that experimental wheat developed by the Australian government could kill people. They said the RNA trigger designed to change the plant’s starch content might match the gene for a human liver enzyme and interfere with it, too. The charge was fanciful, mostly because RNA does not appear to make it past a person’s saliva or stomach acids. Even so, says Wiegand, “the big question any skeptic will raise is: ‘If you are killing insects, what will this do to me?’”

Monsanto has been laying groundwork for the inevitable safety debate. It sent staffers to grocery stores and farm stands to collect fruits and vegetables that appeared to be suffering from viral infections. Analyzing these, they found thousands of fragments of viral RNA, many of which closely matched human genes. Yet it’s not known that anyone has been harmed by RNA in produce. Given this “history of safe consumption,” the company concluded, mere matches between RNA triggers and human genes have “little biological relevance.”

RNA may be natural. But introducing large amounts of targeted RNA molecules into the environment is not.

Last year the U.S. Environmental Protection Agency asked a panel of experts to help it decide how to regulate RNA insecticides, including sprays as well as those incorporated into a plant’s genes. In an 81-page letter to the agency, Monsanto lobbied against any special rules. It said RNA products should actually be spared safety tests it called irrelevant, including those designed to assess whether they were toxic to rodents and whether they could cause allergies, as well as in-depth studies of what happens to the molecules in the environment. Only proteins cause allergies, Monsanto said. And when the company doused dirt with RNA, it degraded and was undetectable after 48 hours.

Company research probably won’t ever satisfy critics. The National Honey Bee Advisory Board told the EPA that using RNA interference at this point would put natural systems at “the epitome of risk” and could be as regrettable as our earlier embrace of DDT. “We are decades away from enough scientific understanding to allow sustainable and predictable use of this technology under field conditions,” they said. The beekeepers worry that pollinators could be hurt by unintended effects. They made the point that the genomes of many insects aren’t yet known, so scientists can’t predict whether their genes will match an RNA target.

The EPA’s advisors, in their report last year, agreed that there was little evidence of a risk to people from eating RNA. But is there some kind of ecological risk? This question they found harder to answer. Monsanto paints RNA as safe and quick to disappear, yet the aim is to make it lethal to insects and weeds, and the company wants to develop longer-lasting formulations. How long? In Hunter’s trees the molecules persisted for months. What’s more, Monsanto’s own discoveries have underscored the surprising ways in which double-stranded RNA can move between species.

These unfolding discoveries suggest that complex biology is at work, leading the EPA’s advisors to say that the “potential scale” of RNA used in agriculture “warrants exploration of the potential for unintended ecological effects.” RNA may be natural. But introducing large amounts of targeted RNA molecules into the environment is not. The advisory panel concluded that “knowledge gaps make it difficult to predict” exactly what problems might arise.

Yet the biggest challenge to RNA sprays, Nitzan Paldi told me, isn’t going to come from regulators. The real problem can be summarized in a single word: Monsanto. “For half the world, that is enough to know it’s evil,” he says. “Monsanto is introducing a new technology, full stop. But Monsanto is also the best way to make this real. For the scientifically literate, this is the dream molecule.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.