Skip Navigation

Transgenic crop may have bred with wild weed

July 25, 2005 By Michael Hopkin This article courtesy of Nature News.

Oilseed rape hybrid unlikely to become 'superweed', say researchers.

British researchers have found evidence that transgenic oilseed rape in test plots is interbreeding with related wild species, raising fears that herbicide-tolerance could spread among weeds.

The government-funded research, carried out at the Centre for Ecology and Hydrology (CEH) in Dorset, UK, suggests that oilseed rape, Brassica napus, may have hybridized with charlock, Sinapis arvensis, a related weed species.

Surprisingly, one of the suspected crosses appears to be a healthy, fertile plant. But, researchers add, concerns that 'superweeds' will take over fields are unfounded.

Herbicide-tolerant weeds would mostly be a problem for farmers trying to rid their fields of unwanted plants, comments Les Firbank, a crop researcher at the CEH research station in Lancaster, who led Britain's previous farm-scale evaluations of the effects of transgenic crops on biodiversity. "It's a management problem for farmers, not an environmental problem," he says.

Mix and match

It's a management problem for farmers, not an environmental problem.
Les Firbank
Centre for Ecology and Hydrology, Lancaster
In the three-year study, researchers analysed weed species growing in 28 fields sown half with transgenic oilseed rape and half with non-transgenic crop. They identified two plants that seem to possess characteristics of both oilseed rape and charlock.

One fertile plant resembled charlock, but was not killed by the Liberty herbicide that the oilseed rape had been engineered to be resistant against. When the researchers extracted and analysed its DNA, they identified the genetic sequence that confers this tolerance1.

Another plant, found in the middle of a non-transgenic plot, seemed to have physical characteristics halfway between those of oilseed rape and charlock, showing that the two species can hybridize. This plant was infertile.

Some environmental groups are claiming this is evidence that transgenes can escape into the plant community at large. Emily Diamand, a spokesperson for campaign group Friends of the Earth, comments that "we're seeing the real possibility of superweeds being created".

Natural selection

But the plants are not necessarily cause for worry, comments Brian Johnson, an ecological geneticist with English Nature, which advises the government on wildlife issues. "Quite frankly, this does not demonstrate the creation of anything resembling a superweed," he told

Johnson points out that the herbicide-resistant wild plant identified in the study may not be a true hybrid. The technique used to identify the gene sequence is very sensitive, and could simply have picked up contamination from oilseed-rape pollen, and the herbicide resistance seen in the plant itself could have arisen naturally, he argues.

Even if it were a true hybrid, the resistance to Liberty herbicide would be unlikely to confer a benefit outside that field, so the plant would not be expected to spread widely.

As for the hybrid plant found in the middle of the plot, Johnson says "it's one thing to be a robust plant, but that doesn't mean anything if you're firing blanks."

One at a time

The finds are not entirely unexpected: oilseed rape has previously been found to hybridize with wild turnip. However, in a comment attached to the report, the study's confidential reviewer says, "this unusual occurrence merits further study."

Johnson remains confident that, with careful management, 'superweeds' resistant to several herbicides will not arise. One tactic is not to license different crops with engineered resistance to different weedkillers. They should be made resistant to one weedkiller at a time, he suggests.

The discovery may be of interest in the United States, where herbicide-tolerant crops are widely grown. "Farmers will have to pay attention to how they manage their crops with herbicide," Firbank says.


  1. Daniels R., Boffey C., Mogg R., Bond J. & Clarke R. Report to DEFRA, (2005).


Need Assistance?

If you need help or have a question please use the links below to help resolve your problem.