Cotton plants native to Mexico’s Yucatan Peninsula may all look the same — unkempt and untamed bushes with flowers that shift from pale yellow to violet as pollinators visit them. But genes that have escaped from genetically modified cotton crops have made some of these native plants fundamentally different, changing their biology and the way they interact with insects.
One type of escaped gene makes wild cotton exude less nectar. With no means to attract defensive ants that protect it from plant eaters, the cotton is devoured. Another escaped gene makes the wild cotton produce excess nectar, enticing a lot of ants that might keep other insects, including pollinators, at bay. It’s the first case that really suggests that a whole ecosystem can be disrupted after transgenes enter a wild population.
The results challenge one long-held view that when genes from genetically modified crops escape into the wild, they have only a neutral effect on wild plants or pass on their benefits to weeds. The findings confirm that unexpected outcomes of this genetic transfer, some of which were never imagined, or at least were not assumed as possible, do happen sometimes.
Scientists have previously tried to explain what happens after DNA from genetically modified crops ends up in their wild relatives. But the majority of studies have been done under carefully controlled conditions, and very few have tested the consequences, if any, of these gene transfers on natural ecosystems.
The scarce evidence motivated scientists to find out. The country was their natural lab. The cotton we know (Gossypium hirsutum) first appeared and diversified between 2 million and 1.5 million years ago in Mexico, and native variants still sprout across the land. In the last 25 years, vast fields of fluffy genetically engineered cotton have also cropped up across the northern part of the country. During that time, the team explored Mexico searching for wild cotton, only to find it at the edge of cliffs, municipal dumps or the middle of a highway. Wild cotton likes to grow in the most inhospitable locations, where it does not have to compete with other species, she says. In 2018, the team traveled to the Ría Lagartos biosphere reserve, an isolated coastal area in the Yucatan Peninsula. With the whitest beaches just a few feet away, the researchers spent long days observing and sampling cotton plants under the scorching sun as swarms of mosquitoes bit them nonstop.
In the lab, the team extracted DNA from the 61 plants it had collected and found that 24 of the plants did not have any transgenes. Twenty-one plants had a transgene that conferred resistance to the herbicide glyphosate; seven could now produce a lethal toxin that kills destructive insects; and the remaining nine had incorporated both escaped genes into their genetic code.
When slathered in a stress-inducing chemical, the plants with glyphosate resistance produced a lot less nectar than wild plants. The nectar is a sugary snack that wild cotton secretes whenever it’s eaten in exchange for the bodyguard services of particularly aggressive ant species. These plants were also the ones that looked the most ragged before the samples were taken. With no tasty reward to offer, and no ants to protect the cotton from hungry herbivores, these plants suffered the most damage compared with native plants that didn’t have the transgene.
Treated with the same chemical, the plants with the insecticide gene exuded nectar all the time, secreting more than the wild plants with no escaped genes and becoming an irresistible beacon to protective ants. But in the researchers’ sample of plants, there weren’t as many with the insecticide gene, suggesting that either the ants or the transgene itself were scaring off other insects. That may have interfered with the pollination of the cotton’s flowers, preventing the plant from reproducing.
Sources:
V. Vázquez-Barrios et al. Ongoing ecological and evolutionary consequences by the presence of transgenes in a wild cotton population. Scientific Reports. Published online January 21, 2021. doi:10.1038/s41598-021-81567-z.
Comments