Transgenes migrate into old races of maize.

On multiple occasions, genes from new-fangled corn produced via bioengineering have slipped into the traditional races of maize grown in southern Mexico.

That's one of the conclusions of the first scientific paper documenting transgenic corn leaking genes into old maize varieties. Such leaks seem "relatively common," report David Quist and Ignacio H. Chapela of the University of California, Berkeley in the Nov. 29 NATURE.

Concerned by early results of this study, the Mexican government investigated. In mid-September, it announced finding bits of a distinctive transgene in a few percent of corn kernels in samples from remote Mexican fields and up to 60 percent near population centers. The findings have ignited concern that such genetic incursions might diminish the diversity of Mexico's traditional maize.

To check for straying transgenes, Quist and Chapela sampled four remote maize fields for genetic material commonly used in engineering corn. One such tool, the bit of cauliflower mosaic virus that was also detected in the government study, showed up in five of seven samples. Also, samples revealed two bits of bacterial DNA, including part of the gene that enables corn to make insecticide.

To estimate the frequency of transgene invasions, Quist and Chapela analyzed the maize genes surrounding transgenes. These neighbors varied so much, Quist contends, that transgenes probably entered the maize genome at multiple sites along the plant's DNA. There's been "a high level of gene flow," the researchers say.

No transgenic corn has been licensed for planting in Mexico since 1998, although some transgenic corn is imported for food. It's not clear whether the stray transgenes result from corn planted before 1998 or from recent, unlicensed plantings.

Ten years ago, people tended to dismiss the possibility of modern genes wandering into traditional varieties, comments ecologist David Andow of the University of Minnesota in St. Paul. With improved tracing of gene flow, that view has been changing, and this paper pushes the trend, he says.

Corn breeders in the tropics still collect old Mexican varieties, although traditional U.S. corn breeders rarely go to Mexico for material, notes geneticist John Doebley of the University of Wisconsin-Madison. He'd advise keeping an eye on the wandering transgenes but says, "I'm not an alarmist about this."