NEW YORK (GenomeWeb) – Although transgenic crops engineered to express multiple insecticidal proteins based on the same toxin have been used for years to combat pest resistance, a new study published this week indicate that this approach is not always as effective as indicated by many mathematical models.
The findings, which also point to the potential of cross-resistance between unrelated toxins, suggest that current efforts to anticipate the effects of pyramided insecticidal toxins too often assume ideal conditions and point to the need for more realistic values in prediction simulations.
Since their introduction to the market in 1995, crops modified to express pesticidal Bt proteins derived from the bacterium Bacillus thuringiensis have been widely used against a variety of pests, particularly those affecting corn and cotton. As originally developed, these plants expressed a single Bt protein, but their narrow range of action and concerns over pest resistance led to the development of plants that express more than one type of Bt toxin against the same pest — a process known as stacking or pyramiding.
The beneficial effects of Bt plants have been substantial, increasing yields for growers and reducing the need for topical insecticides. However, growers have become increasingly reliant on these transgenic crops and, in recent years, have been reducing refuges of non-Bt plants, which produce susceptible pests that can mate with resistant insects feeding on Bt crops.
As a result, there have been an increasing number of reports of pests that are resistant to Bt traits, including some US populations of Helicoverpa zea that can survive on pyramided Bt cotton.
Last year researchers led by the University of Arizona's Yves Carriere reported data indicating that pyramiding can be ineffective when conditions differ from the ideal. He and his colleagues have now published a new report analyzing how the predicted effects of Bt stacking compares with real-world outcomes.
According to the latest study, which appeared in Nature Biotechnology, among the conditions favorable to preventing resistance to stacked Bt crops are toxins that kill all or nearly all susceptible pests; the absence of cross-resistance between pyramided toxins; and a high rate of redundant killing, wherein each toxin alone can kill nearly all susceptible pests.
But in reviewing data from various studies conducted by other groups, the U of A researchers found that the key factors affecting the durability of Bt pyramids often "deviate from ideal conditions." Most notably, in many cases the survival of insects susceptible to Bt traits is greater than the threshold value of .25 percent commonly used in prediction models.
"The pyramids should kill all or most susceptible insects," Carriere told GenomeWeb. "If that does not happen, one of the key assumptions of redundant killing is unlikely to be met, so it will be easier for those [susceptible] pests that are not killed … to evolve resistance."
Additionally, the models currently used to develop resistance-management strategies don't always incorporate the most relevant values for key parameters, the scientists wrote in their paper.
For instance, the U of A group identified significant relationships between Bt toxin amino acid sequence similarity in domains II or III and the effects of toxin combinations on cross-resistance and mortality of susceptible insects, but noted this is a factor rarely considered in simulations.
"When you put toxins in the pyramid, you want them to be different in domain II to minimize cross-resistance and different in domain III to have greater mortality of susceptible insects," Carriere said. Knowing this could help ag-bio firms better select which Bt traits to stack in their products and help in designing chimeric toxins.
"It's also important for modelers," he added. "If you model the evolution of resistance to a specific pyramid and a specific pest, you know the homology of the toxins in that pyramid so you can guess what would be the expected mortality of susceptible insects, you can guess … the expected cross-resistance, [and] you can build models that are more realistic."
Further, while amino acid similarity between Bt toxins is known to be a frequent source of cross-resistance stemming from mutations that reduce binding to midgut receptors, Carriere and his team found that cross-resistance can occur between toxins with no structural homology. This, they wrote, suggests that cross-resistance can occur through other mechanisms than ones that affect toxin binding.
Indeed, when genetic mutations are associated with toxin binding to receptors, existing simulations are fairly accurate because such changes are expected, Jonathan Lundgren, a US Department of Agriculture entomologist who was not involved in the Nature Biotechnology study, explained. However, "if the allele that's involved is a behavioral change or … is not in the target organism — maybe it's in [a pest's] symbiont community — then our models are suddenly a little less useful."
This, the U of A scientists wrote, highlights the need for additional research into the mechanisms behind this kind of atypical cross-resistance.
In the end, Carriere said that the findings reported in Nature Biotechnology should assist in the optimization of Bt pyramids for sustainable pest management. "There is a lot of assessment of [Bt toxins'] impact on the environment and we know that, in general, they are environmentally friendly ways of controlling pests. So we want those Bt toxins to stay … as long as we can keep them around."
But to Lundgren, a key step toward maintaining the efficacy of Bt toxins will be their use in conjunction with other kinds of technologies.
"Whenever you use one technology to combat pests, resistance will occur," he said. "History and science tells us this consistently; no matter what we throw against them, the pests are eventually going to become resistant."
And at least one industry player agrees. Monsanto, which markets a number of stacked Bt crop plants, is currently in the final stages of developing a corn rootworm resistance strain of corn that expresses two different Bt traits, as well as dsRNA that silence a gene essential to the insects.
But RNAi is not a cure-all, and resistance to the gene-silencing technology will eventually develop if it is relied upon too heavily, Lundgren cautioned. Therefore, increasing the size of refuges is also a necessity.
"The science is pretty strong on this point," he said. "That is really what needs to happen."