Monsanto has petitioned EPA to eliminate the structured non-Bt refuge requirement for farmers in Texas, the Mid-South and the Southeast when they grow the company's Bollgard II cotton technology.
Currently, growers have two refuge options when they plant transgenic Bt (Bollgard or Bollgard II) cotton: Option 1 requires a 20 percent or greater refuge of non-Bt cotton and Option 2 calls for a 5 percent non-Bt refuge that cannot be treated with insecticides for bollworm or tobacco budworm. (If the 5 percent refuge is embedded, it can only be treated if the associated Bt cotton is treated at the same time.)
Both options were designed to provide a source of susceptible bollworm/tobacco moths that can mate with moths that might become resistant to the single Bt gene in Bollgard cotton.
The requirements were suggested by a scientific advisory panel convened by EPA in the mid-1990s to develop strategies to prevent the development of resistance in bollworm and tobacco budworm to the technology.
But Monsanto researchers and some university scientists are saying that the non-Bt cotton refuge acres required to prevent target pests from developing resistance over time are not necessary in the two-gene system of the Bollgard II technology.
The issue, researchers say, is not that they don't need susceptible insects; it's that susceptible insects may come from a large number of alternate crop and native host plants. Scientists have known bollworms can complete larval development on a number of plants, including corn, soybeans and grain sorghum. But they thought the number of hosts for tobacco budworm was limited — until now.
“We have good data from North Carolina and Georgia that show tobacco budworm can be found living on a number of crops and other plants,” says J.R. Bradley, professor of entomology at North Carolina State University. “I've even found them on my wife's petunias in my backyard here in Raleigh.”
North Carolina and Georgia are two of the six cotton-producing states where Monsanto and cooperating university scientists conducted studies over the last four years to determine the host range for bollworm and tobacco budworm. The tests, which used carbon isotope (C-3/C-4 assays) and gossypol analyses to determine the insects' larval food sources, were expanded to include southeast Texas in 2005.
“We've done two years of studies with bollworm, and we pretty well understand that bollworms (a) have a lot of hosts throughout the season and (b) they're flying significant distances all season long so there is significant mixing from these alternate hosts,” said Walt Mullins, Monsanto's technical manager for Bollgard and Bollgard II.
“We have solid reasons to believe there will be contributions on the bollworm side from hosts such as soybeans, peanuts, weedy hosts, corn, sorghum. All are contributing to the overall population, and we've been able to measure that, primarily with C-3/C-4 analyses.”
The missing piece of the puzzle has been the tobacco budworm. Most scientists have assumed tobacco budworms would be mostly found in cotton and tobacco, especially during July, August and September when susceptible insects are needed to mate with resistant individuals to prevent resistance development.
But the tests in North Carolina and Georgia showed that minimally 80 to 90 percent of the tobacco budworms found in those states were “non-cotton” moths; that is, they had fed as larvae on plants other than cotton.
“We found significant numbers of tobacco budworm moths in the study, but they were largely non-cotton moths all season long,” said Mullins. “We anticipated a certain degree of that based on tobacco as a host. But peanuts are also a fairly good host for tobacco budworms along with soybeans and the diversity of other, weedy hosts for that area.
“But what we were really amazed at was how much the cotton's contribution to moth production was getting drowned out by everything else that was going on. We knew there would be some non-cotton produced tobacco budworms, but we were really surprised at the magnitude of that number.”
In the Mid-South and southeast Texas, where growers tend to plant fewer crops besides cotton, the percentages of non-cotton tobacco budworm moths weren't as high. Monsanto conducted the tests in Arkansas, Louisiana, Mississippi and Tennessee in 2004 and added southeast Texas in 2005.
“Southeast Texas is where we saw some of the first pyrethroid resistance in tobacco budworm, and we decided we should include that region in the sampling as well,” said Mullins.
In what he called the “worst-case scenario,” university and Monsanto researchers saw no less than 10 percent of the moths that were non-cotton moths in any location all season long. The lowest numbers were found in southeast Texas and the lower Delta states.
“That means if you have 100 moths in the comparison at least 10 of those moths came from some other host than cotton,” said Mullins, “and, more typically, in Mississippi and northeast Arkansas it would be closer to 40 to 50 percent.”
Applying statistical analysis to those numbers means that you could be dealing with numbers of non-cotton moths as high as 10 percent or as low as 1 percent — if you want to be conservative, says Mullins.
From the time Bt cotton was introduced in 1996, entomologists have been using computer models to try to predict how soon and under what conditions tobacco budworms or bollworms could theoretically develop resistance to the single Bacillus thuringiensis or Bt gene contained in Bollgard cotton.
The primary motivation for developing Bollgard II, which contains two different Bt genes, was to enhance the resistance management value of the Bt product, according to Mullins. “It was broadly believed that if you put two genes in the plant that worked differently on the insect, you would reduce the probability of resistant insects surviving both genes and significantly decreasing the effective life of the product.
“If we're talking only 1 percent of the tobacco budworm moths in the lower Delta being non-cotton moths, the model would suggest that the percent of the effective natural refuge that would be required based on the strength of the two-gene system would be less 0.1 of a percent,” he said.
“We are proposing that when you look at the amount of refuge necessary for a two-gene product, even in the worst case scenario in our data, that we still have significantly more refuge than what would be required to maintain efficacy of Bollgard II for an extended period of time.”
Bradley says entomologists on the East Coast have long believed that cotton producers in the region had ample natural refuge or alternate hosts to provide susceptible tobacco budworm moths for mating with resistant months.
“But we told Monsanto we would not get into the public arena on this until they developed a data base that provided evidence of what was actually happening in the field,” he said. “Now we have the data base.”
Long considered one of the leading experts on insect pests in cotton, Bradley says the refuge requirement is more than just a theoretical issue for producers. “The fact is that the refuge costs us money — us being the grower,” he said. “And if there is no biological need for it, why burden farmers when they need to be doing everything they can to stay in business?”
Roger Leonard, research entomologist with the LSU AgCenter who conducted much of the bollworm/tobacco budworm host study in Louisiana, says he's not ready to weigh into the natural refuge debate just yet.
“I'm on the fence at this point, and I'm willing to listen and review all the data,” he said. “I recognize the considerable differences in the ecology of these insects among the cotton production regions, and I want to look at all the data more closely before I start expressing opinions.
“I believe the good news is that there are populations of these insects coming from somewhere other than cotton, and, even if we disagree with some of the findings, some level of these pests are being produced on plants other than cotton.
“I also support recommendations that would increase cotton acreage planted to Bollgard II cultivars. The two genes in Bollgard II provide a broader spectrum of pest control, offer better insect resistance management or IRM qualities and can further reduce oversprays with conventional insecticides.”
Mullins said the package Monsanto has submitted to EPA petitions the agency for a natural refuge option for all Bollgard II cotton from Texas east.
“You get west of Texas and you get into pink bollworm area,” he said. “They're basically in an eradication mode with the pink bollworm, and these data were generated specifically for bollworm and tobacco budworm.”
In February, Monsanto will conduct a series of meetings with university researchers and Extension entomologists, consultants and growers to make them more aware of what the company is proposing.
“We want to provide them with the opportunity to make comments either in support of or opposition to the proposal. We expect EPA to appoint a scientific advisory panel to review the proposal in the spring or early summer, and we want everyone to be fully informed so we can have as broad a discussion as possible.”
Both Bollgard and Bollgard II technologies are scheduled to be re-registered in September, and EPA could announce a decision on Monsanto's proposal that coincides with the re-registration decision.
Regardless of when the decision is made, growers must continue to follow EPA's guidelines on Bollgard and Bollgard II. “This in no way means that we are backing away from our commitment to be in compliance with the current refuge requirement for Bollgard and Bollgard II,” says Mullins.
Monsanto anticipates that the refuge requirements for Bollgard will remain unchanged. “We will have to work through how we handle that interim period when we still have Bollgard and Bollgard II in the marketplace and appropriately address refuge requirements for both products.”
Leonard says he welcomes the meetings, which will be held on a state-by-state basis in the Delta region. “We need to have some frank discussions,” he said. “Some of our consultants are skeptical about the information, and we need to thoroughly review the data to be certain we don't do anything that would eliminate this technology as an IPM tool.
“This is the most effective technology that has ever been developed for our producers to manage caterpillar pests in cotton. In the absence of Bt cotton, cotton producers in the Mid-South would suffer dramatic increases in production costs for insect pest management. This is a significant issue that must be considered carefully by the entire U.S. cotton industry. At this point, I'm not sure the public realizes how important this decision could be.”