WEEDS have been shown for the first time to become stronger and fitter by cross-breeding with genetically engineered crops, in this case sunflowers. And at the same time, a team in France has demonstrated how easily weeds might be able to swap genes with the GM strains of sugar beet already in field trials.
The findings emphasise the need for developers of GM crops to be cautious about which traits they introduce into plants, in case they spread irreversibly to weeds. They also strengthen the case for using technologies that would prevent gene spread altogether, argues Jeremy Sweet of the National Institute for Agricultural Botany in Cambridge. 鈥淚f you鈥檙e worried about a gene which alters the fitness of wild populations, then stopping the GM plant breeding has got to be a good thing,鈥� he says.
Allison Snow鈥檚 team at Ohio State University has shown in controlled tests that wild sunflowers, considered a weed by many farmers in the US, become hardier and produce 50 per cent more seeds if they are crossed with a GM sunflower resistant to seed-nibbling moth larvae. 鈥淲e were shocked,鈥� says Snow. However, Pioneer Hi-Bred of Iowa, which developed the GM sunflower, says it has no plans to sell the strain commercially.
Advertisement
Snow, whose results were presented to a conference last week, cautions against overstating the significance of the results. 鈥淚t doesn鈥檛 prove all GM crops are dangerous,鈥� she says. 鈥淚 just think we need to be careful because genes can be very valuable for a weed and persist for ever once they鈥檙e out there.鈥�
Pioneer Hi-Bred spokesman Doyle Karr adds that existing GM crops such as soybeans and maize don鈥檛 have any wild relatives in the US. And although GM canola, or oilseed rape, is related to wild mustard, the only spread of genes so far has been to commercial non-GM rape, especially in Canada. 鈥淚t鈥檚 all gene- and crop-specific,鈥� he says. 鈥淵ou ask beforehand what the implications are if there鈥檚 crossover, and that鈥檚 been true all along.鈥�
However, various companies are developing GM sugar beets. Studies of normal beet fields by Henk van Dijk and his colleagues at the University of Lille in France suggest that they have underestimated the likelihood of GM beets swapping genes with the beet weeds that grow among them. 鈥淲e found gene flow to be possible between all forms,鈥� they say in the Journal of Applied Ecology (vol 39, p 561).
The situation with beet is particularly complicated because there is a two-way flow, with weed genes often polluting farm strains and reducing yields. The beet weeds could become even more of a nuisance to farmers if they pick up herbicide-resistance genes.
Van Dijk says that while tricks such as doubling the number of chromosomes in GM strains could reduce the chance of gene spread, they would not eliminate it. 鈥淚t鈥檚 almost inevitable,鈥� he says. But despite the risk, he still believes GM strains could help farmers.