What is in this article?:
- How genes jump from crop to crop
- Pollinating insects had relatively small impact on gene flow
- A new data-driven statistical model that incorporates the surrounding landscape in unprecedented detail describes the transfer of an inserted bacterial gene via pollen and seed dispersal in cotton plants more accurately than previously available methods.
- The transfer of genes from genetically modified crop plants is a hotly debated issue. Many consumers are concerned about the possibility of genetic material from transgenic plants mixing with non-transgenic plants on nearby fields. Producers, on the other side, have a strong interest in knowing whether the varieties they are growing are free from unwanted genetic traits.
Pollinating insects had relatively small impact on gene flow
Surprisingly, the team found that pollinating insects, widely believed to be the key factor in moving transgenic pollen into neighboring crop fields, had a small impact on gene flow compared to human farming activity, with less than one percent of seeds collected around the edges of non-Bt cotton fields resulting from bee pollination between Bt and non-Bt cotton.
Most previous studies focused on the distance between the non-transgenic crop field and the nearest source of transgenic plants.
"Although this approach is simple, it is potentially less useful for understanding gene flow in commercial agriculture where there can be many sources of transgenic plants," Heuberger said.
Heuberger and her co-workers broadened the scope to include flower-pollinating bees, humans moving seeds around and the area of all cotton fields in a three-kilometer (1.9 mile) radius. This approach turned out to be more powerful in understanding the effect of surrounding fields than using the customary model based solely on distance.
For the study, the scientists chose 15 fields across the state of Arizona planted with cotton that did not have the transgenic protein encoded by a gene from the bacterium Bacillus thuringiensis, or Bt. They assessed the number of pollinators visiting cotton flowers through field observations and determined the transfer of genes by collecting samples of cotton bolls and determining their genetic identity.
"We saw a need for a spatially explicit model that would account for the whole surrounding landscape," Heuberger said. "Our model takes into account the distance and area of all relevant neighboring fields, the effect of pollinators like bees and human factors that can result in the mixing of seed types."
Heuberger's findings have implications not just for genetically engineered traits but also more generally for seed production.
"When you grow a crop and want the variety to be pure, just being able to know how far gene flow will occur and how it is affected by pollinators and human farming activity in the area is very valuable."
The research was funded by Western Region Sustainable Agriculture Research and Education and an Environmental Protection Agency STAR Fellowship.