When farmers began seeing increased damage from the western corn rootworm a few years ago, they knew that the beetle was developing resistance to a widely used toxin produced by the bacterium Bacillus thuringiensis (Bt). The tried-and-true approach of genetically engineering corn to produce Bt toxin was losing its effectiveness against the insect, and crops were once again being destroyed.
Monsanto, which commercialized a strain of Bt corn in 2003 that kills the corn rootworm, raced to find an alternative weapon to combat the devastating pest. The company is now close to seeking regulatory approval for a new product—a strain of corn engineered to produce double-stranded RNA (dsRNA). When a rootworm larva eats the roots of the corn, it ingests dsRNA, which silences an essential gene in the pest, thereby killing it.
The approach relies on a biological process called RNA interference (RNAi), which is used by cells to suppress gene expression. The process is triggered by dsRNA, which cells break into pieces that interfere with the transcription of messenger RNA and the translation of that mRNA into protein.
Monsanto and other firms are excited about RNAi-based technology because it offers many potential benefits for growers, including protection from insects and roundworms without the use of traditional pesticides. It also offers the possibility of reversing herbicide resistance in weeds, and it could even help save honeybees when used as a weapon against mites.
But some scientists are raising concerns about the potential risks of the technology to nontarget organisms, including humans.
Anticipating that companies will soon seek regulatory approval for RNAi-based pesticides, the Environmental Protection Agency held a meeting last month to get advice from experts on whether its current framework is adequate for assessing the risks of the new technology. These experts and others say that regulators must better understand the toxicity issues unique to RNAi-based pesticides, as well as the persistence of dsRNA in the environment.
EPA has some experience examining the risks associated with RNAi-based pesticides, but most of the products were reviewed before scientists had a good understanding of the RNAi process, says Chris Wozniak, a biologist with EPA’s Biopesticides & Pollution Prevention Division. The first RNAi-based pesticides were approved by EPA in 1997 for targeting infectious plant viruses, but the mechanism behind RNAi was not widely understood until 1998 (Nature, DOI: 10.1038/35888).
1992 The Food & Drug Administration declares nucleic acids “generally recognized as safe.”
1994 FDA approves the genetically modified Flavr Savr tomato, which uses antisense RNA—a similar biological process to RNA interference (RNAi)—to regulate expression of an enzyme responsible for ripening fruit. The product is discontinued because it isn’t profitable.
1997 EPA approves virus-resistant papaya, squash, and potato, all of which use antisense RNA technology.
1998 RNAi phenomenon is first described in literature.
2010 EPA approves virus-resistant plum that uses posttranscriptional gene silencing.
2010 & 2011 USDA approves two high-oleic soybean traits for cultivation in the U.S. The technology uses RNAi to block expression of certain soybean enzymes, leading to increased monounsaturated fat and reduced saturated fat soybean oil.
2013 EPA grants Monsanto a permit to field-test an RNAi-based pesticide that targets western corn rootworm.
These products incorporated virus resistance traits into the genomes of plants, “triggering an innate defense mechanism in plants,” says John Kough, a senior scientist in EPA’s Microbial Pesticides Branch. EPA is more concerned about the newer generation of RNAi-based pesticides because they are “designed to kill free-living organisms,” he notes.
Thus far, all of the RNAi-based pesticides reviewed by EPA have used a similar strategy: They require incorporation of the pesticide into the plant genome to be effective. But companies are also working on RNAi-based products that are intended to be directly sprayed on plants, insects, or other pests, or applied in granular form. Such products could be mixed with surfactants, stabilizers, and other pesticide active ingredients.
For example, Monsanto is currently developing an RNAi-based spray designed to make weeds less resistant to its herbicide glyphosate, also known as Roundup. The company and others are also investigating an RNAi-based product to protect honeybees from the varroa mite. Meanwhile, Syngenta acquired the Belgian biotech company Devgen in late 2012 to work on RNAi-based pesticide sprays.
EPA has more than 20 years of experience assessing the risks of biobased pesticides that are incorporated into plant genomes, primarily with plants that express Bt proteins. “The agency does not, however, have any experience with hazard and risk assessments for dsRNA products that are applied directly to the environment,” says Russell Jones, a senior biologist in EPA’s Biopesticides & Pollution Prevention Division.
The current screening-level tests for pesticides may be “a starting point for assessing the ecological hazards and risks of directly applied dsRNA products, but they may not be completely applicable due to the unique modes of action of these active ingredients,” Jones says.
Some of the hazards that are unique to dsRNA active ingredients include off-target gene silencing, silencing of the target gene in unintended organisms, immune system stimulation, and the saturation and disruption of the RNAi machinery in nontarget organisms, Jones notes. These hazards were first pointed out by Department of Agriculture entomologists Jonathan G. Lundgren and Jian J. Duan in a BioScience paper last year (DOI: 10.1525/bio.2013.63.8.8).
The USDA scientists also highlighted areas where information is lacking, including the persistence of dsRNA in the environment. Without such information, it is difficult to assess exposure.
EPA acknowledges that there are insufficient data in the published literature on the stability and activity of dsRNA in soil and aquatic systems to determine how long a dsRNA molecule will last in the environment. Its lifetime could depend on the size of the molecule, whether it is in a clay soil or a loamy soil, and whether it binds to organic matter or other soil particles, Wozniak says.
Monsanto has conducted studies looking at the degradation of dsRNA in soil. Results of those studies have been submitted for publication, says David Carson, a lead scientist in environmental fate and microbiology at Monsanto. For its RNAi-based corn rootworm product, the rate of degradation was independent of dose, taking less than 30 hours to degrade by half in soil. The dsRNA degrades completely within two days in soil, he notes.
Another Monsanto study of the corn rootworm pesticide suggests the technology has potential for high taxonomic specificity. The study examined the lethal and sublethal effects of dietary exposure to the dsRNA in 18 different insect species, representing 10 families and four orders(Transgenic Res. 2013, DOI: 10.1007/s11248-013-9716-5). The corn rootworm was the only species affected. That study helps rule out the potential for effects of dsRNA pesticides on nontarget organisms that eat the corn or the plant’s roots, says Steven L. Levine, a senior science fellow at Monsanto.
Much of the concern related to human health effects of dsRNA was raised after a study led by Chen-Yu Zhang of Nanjing University, in China, that reports the presence of exogenous plant micro dsRNA in human blood plasma after the consumption of rice (Cell Res. 2012, DOI: 10.1038/cr.2011.158). The researchers suggested that once inside humans, such microRNA can regulate genes, such as those associated with cholesterol.
The paper required a correction (CellRes. 2012, DOI: 10.1038/cr.2011.174), and several researchers have since disputed the findings, saying there is no evidence for the uptake of plant microRNA into humans or mice fed a rice diet. Nonetheless, the paper stoked fears and concerns about the safety of RNA-based technology in food.
The reaction prompted regulators at Food Standards Australia New Zealand to assess the safety of genetically modified crops that incorporate gene-silencing technology. The agency concluded last May that “the weight of scientific evidence published to date does not support the view that small dsRNAs in foods are likely to have adverse consequences for humans.”
Also, medical research on RNAi may calm concerns about dsRNA in foods. The pharmaceutical industry has wanted to develop drugs that could silence essential genes associated with various cellular functions, cancerous tumors, or pathogens.
But most dsRNA-based drugs have not been effective unless they are encapsulated or protected from stomach acids and various enzymes. This suggests that exposure to dsRNA by oral ingestion is likely to be minimal.
EPA believes that dsRNA expressed by plant cells “may not be taken up as an effective agent of RNAi by the mammalian gastrointestinal tract,” EPA’s Kough says. However, more data are needed about the environmental fate of dsRNA with complicated structures, such as hairpins or supercoiling, or dsRNA formulated to have increased stability, he says.
For companies developing RNAi-based pesticides, they will have to balance dsRNA stability to ensure the pesticide is around long enough to be effective but degrades fast enough to not be a safety concern, Kough says.
Beekeepers are also excited about RNAi-based technology. “Existing products used in the beekeeping industry to combat bee parasites and diseases have limited effectiveness, and the industry is looking with hope at new technologies such as RNAi to improve honeybee health,” says Christi Heintz, executive director of Project Apis m., a nonprofit organization that funds research to improve honeybee health.
The technology could benefit beekeepers because of its high specificity and low chance to cause resistance. However, the group is urging EPA to thoroughly evaluate RNAi designed to protect crops to look for unintended effects on nontarget organisms such as bees, she says.