For US cotton farmers the world changed dramatically in 1996. During the prior year, the nibblings of tobacco budworms, cotton bollworms, and pink bollworms cost the industry upwards of a quarter billion dollars in lost yields. But in 1996 a new kind of cotton arrived on the market, one genetically engineered to kill insects without the additional application of pesticides. Bt-cotton became widely popular, accounting for half of all cotton grown in the United States by 2003.
By 2003 however, the crop was experiencing problems, formerly susceptible insects were developing resistance. So, a new kind of Bt cotton was introduced, one with “stacked” or “pyramided” toxins. These varieties have since taken over, now 90% of cotton grown in the United States produces two toxins derived from the bacterium Bacillus thuringiensis.
A key assumption made by regulators, farmers, and most scientists is that two-toxin plants will kill insects that are resistant to just one toxin. In a recent PNAS Early Edition paper, a team of agricultural researchers test this assumption and finds it is false.
Instead, insect strains selected for resistance to one poison, Cry1Ac, survived better on “pyramided” cotton producing two insecticidal proteins (Cry1Ac and Cry2Ab) than their parent strain. For the researchers, this was a surprise.
“Also, we did not expect to find pervasive cross-resistance between Cry1Ac and Cry2Ab,” says Yves Carriere, an entomologist at the University of Arizona. “These findings have important implications for management of resistance in pests with low susceptibility to Bt toxins.”
In 19 out of 21 experiments analyzed, they found that selection with one toxin increased resistance to another, even if the insect had not been exposed to it before. This, the authors suggest, means it is significantly easier for populations of some cotton pests, like Helicoverpa zea (the cotton bollworm) and H. armigera (the African bollworm), to evolve resistance.
“The evolution of resistance to Bt crops by pest is the most serious threat to the benefits they provide,” Carriere says. To slow the evolution of resistance, the US Environmental Protection Agency mandates refuges–areas of non-Bt cotton where insects susceptible to Bt can survive and mate with resistant insects to keep the proportion of recessive resistance genes low in the population. However, the authors write, many of the simulation
models used to direct the size and location of these refuges rely on incorrect assumptions and produce over-optimistic estimates.
The team is following up on this work by examining the effects of seasonal declines in Bt toxin concentration and taking a broader look at cross-resistance.