GM crops detoxify glyphosate

DNA shuffling used to introduce bacterial enzyme, while some raise outcrossing concerns

May 21, 2004
Charles Choi(

Scientists report in the May 21 Science that they have engineered plants capable of detoxifying lethal doses of the heavily used herbicide glyphosate, while showing no setback in growth. The new genes could help corn, soybeans, and other crops benefit from the herbicide later than before in the growing season without affecting flower and seed development, says the team that published their findings, and field trials are underway.

“Commercialization of this trait may provide competition to Roundup Ready crops,” researcher Linda Castle of Verdia in Redwood City, Calif., told The Scientist. “Verdia is working with Pioneer Hi-Bred International Inc. to evaluate the commercial potential in corn and with Delta and Pine Land Company to evaluate the commercial utility of the trait in cotton.”

However, the potential for such enhanced herbicide resistance genes spreading to wild cousins to create superweeds “is a serious concern,” said Henry Daniell, professor of molecular biology at the University of Central Florida in Orlando. “Imagine a gene that is 10,000 times more potent, but outcrossing with weeds. This should be introduced only in combination with additional transgene containment strategies, tightly linked,” said Daniell, who was not involved in the study.

Globally, Roundup Ready glyphosate tolerance is the number one transgenic trait, occupying most of the 122 million acres of transgenic herbicide-tolerant crops grown in 2003. Glyphosate inhibits the enzyme enolpyruvylshikimate-3-phosphate synthase (EPSPS) in the plant chloroplast-localized pathway that leads to the biosynthesis of aromatic amino acids.

“Because the target pathway is not present in animals or insects, glyphosate is nontoxic to them. In addition, glyphosate is rapidly broken down in the soil and does not contribute to groundwater contamination,” Castle said. “The use of glyphosate tolerant crops has resulted in a dramatic increase in the use of no-till cropping systems, which improves nutrient retention in the soil, increases soil organic matter content, and reduces soil erosion.”

Currently, Roundup Ready plants do not detoxify glyphosate, but rather overproduce EPSPS to resist lethal glyphosate levels, Daniell said. In these plants, the herbicide still accumulates in meristems, where it may interfere with reproductive development and lower crop yield.

The researchers looked for a way to detoxify glyphosate instead. They searched within their microbial collection and in 2000 found glyphosate N-acetyltransferase (GAT) enzymes from Bacillus licheniformis that had weak glyphosate-detoxifying capabilities. The gene variants encoding the GAT enzymes were isolated for use in DNA shuffling, and the researchers selected the gene variants that encoded the most active GAT enzymes. After 11 generations of recombination of “directed evolution,” they created enzymes that are nearly 10,000 times more active than the original parents when expressed in Escherichia coli. “The work on various iterations of multigene shuffling to recombine diversity of genes is quite solid,” Daniell said.

The transgene proved less effective in plants than in bacteria, Daniell said, but transgenic corn expressing the best GAT enzymes were tolerant to six times the usual lethal field rate of glyphosate, while showing no visual damage or growth setback.

“There is some reason why tolerance is low in crop plants even though the GAT is hyperactive. Maybe because the site of action of the glyphosate is EPSPS, which is compartmentalized within plastids, but the authors did not target their enzyme to the chloroplast,” Daniell said. “Also, these genes originated from B. licheniformis, a prokaryote which is not efficiently translated in a eukaryotic environment, like BT genes that needed codon optimization. Thus, this technology has a lot of room for improvement.”

While Castle said they could further improve the enzyme, she noted they believe they have genes more than adequate for crop development. Currently, they have corn in small-scale trials looking at different gene variants and testing the efficacy of glyphosate sprays at different concentrations. “It takes about 5 years of field trials to get regulatory approval and to get breeding done in elite varieties. We're just at the beginning,” Castle said.

“What is impressive is they started out with an enzyme not of value commercially that could only carry out limited glyphosate acetylation reaction, and were able to use this technology to evolve it,” said Nam-Hai Chua, professor of plant molecular biology at Rockefeller University in New York, who did not participate in the study. That suggested that DNA shuffling could prove successful with other traits of interest, he said.