American chestnuts were once a dominant tree, and a major source of food, in the forests of eastern North America.

ANDREW NEWHOUSE

To save iconic American chestnut, researchers plan introduction of genetically engineered tree into the wild

SYRACUSE, NEW YORK—Two deer-fenced plots here contain some of the world's most highly regulated trees. Each summer researchers double-bag every flower the trees produce. One bag, made of breathable plastic, keeps them from spreading pollen. The second, an aluminum mesh screen added a few weeks later, prevents squirrels from stealing the spiky green fruits that emerge from pollinated flowers. The researchers report their every move to regulators with the U.S. Department of Agriculture (USDA). "We tell them when we plant and where we plant and how many we plant," says Andrew Newhouse, a biologist at the nearby State University of New York College of Environmental Science and Forestry (SUNY ESF).

These American chestnut trees (Castanea dentata) are under such tight security because they are genetically modified (GM) organisms, engineered to resist a deadly blight that has all but erased the once widespread species from North American forests. Now, Newhouse and his colleagues hope to use the GM chestnuts to restore the tree to its former home. In the coming weeks, they plan to formally ask U.S. regulators for approval to breed their trees with nonengineered relatives and plant them in forests.

If the regulators approve the request, it would be "precedent setting"—the first use of a GM tree to try to restore a native species in North America, says Doria Gordon, lead senior scientist at the Environmental Defense Fund (EDF) in Washington, D.C. But deciding whether to unleash a GM tree into the wild could take years.

American chestnuts, towering 30 meters or more, once dominated forests throughout the Appalachian Mountains. But in the early 1900s, a fungal infection appeared on trees at the Bronx Zoo in New York City, and then spread rapidly. The so-called chestnut blight—an accidental import from Asia—releases a toxin that girdles trees and kills everything above the infection site, though still-living roots sometimes send up new shoots. By midcentury, large American chestnuts had all but disappeared.

In 1990, SUNY ESF tree geneticists William Powell and Charles Maynard (now retired) decided to try to create resistant chestnuts with the then-new technology of genetic engineering. Eventually, they inserted into the tree's genome a wheat gene that codes for an enzyme called oxalate oxidase, or OxO. It breaks down the oxalic acid the pathogen releases, which is what kills the trees. "We're basically taking the weapon away from the fungus," Powell says.

Researchers seal off the flowers of a chestnut carrying a wheat gene that neutralizes a fungal toxin.

ANDREW NEWHOUSE

It didn't work at first. Then, the scientists changed the wheat gene's promoter sequence to cause OxO to be expressed at high levels. In 2014, they reported that a GM tree named Darling 58 both resisted blight infection and transmitted resistance to its offspring. Subsequent tests showed that it produces nuts indistinguishable from those of native trees, Newhouse says. And its pollen, flowers, and decaying leaves don't harm bees, beneficial soil fungi, or tadpoles that hatch in pools on the forest floor.

But the request to release it is likely to face a lengthy regulatory road. The United States, China, and Brazil have approved some transgenic trees for use in fruit orchards, biofuel plantations, and afforestation projects. But like GM crops and animals, GM trees are controversial, and ethical and ecological concerns are heightened because the chestnut trees would grow wild. Regulators from three federal agencies are likely to take a close look at those concerns. USDA officials, for instance, will seek to determine whether the tree could become a weed or otherwise threaten existing plants. The Food and Drug Administration will study whether the tree's fruit is safe to eat, and the Environmental Protection Agency will consider whether the trees' blight-blocking enzyme should be regulated as a fungicide.

Regulators also "need a really clear process for transparently incorporating … cultural and spiritual values into the decision-making," says Gordon, who serves on a committee convened by the National Academies of Sciences, Engineering, and Medicine to examine issues surrounding GM trees. The American chestnut was a culturally important tree and important food source for many Native Americans, and some are wary of genetically altering a species with which they have a long relationship, says Neil Patterson, a member of the Tuscarora Nation and assistant director of the Center for Native Peoples and the Environment at SUNY ESF.

If the tree survives the regulatory gauntlet, researchers not directly involved in its development are cautiously optimistic that it could help with restoration. It seems to fend off blight better than hybrids produced so far through traditional breeding methods, says Jared Westbrook, chief scientist of the Asheville, North Carolina–based American Chestnut Foundation, which has spent 35 years attempting to breed a blight-resistant chestnut and helped fund the GM tree research. But to maximize survival, the GM trees—which are all descended from clones of one "founder" tree—will need to be crossed with trees adapted to local climate and diseases, Westbrook says. "We're not going to restore a species with a clone."

All that work could be undone if the fungus evolves a way around the defense, says Richard Sniezko, a tree geneticist with the U.S. Forest Service in Cottage Grove, Oregon. Powell and Newhouse doubt that kind of natural selection will occur, because their tree does not actually kill the fungus. Still, "None of us wants something to be put out there … and it fails after 10 years," Sniezko says.

The continuing influx of insects and pathogens from abroad also could present a new chestnut killer, says Gary Lovett, an ecologist at the Cary Institute of Ecosystem Studies in Millbrook, New York. Creating resistant varieties "is a good thing," he says. "But it doesn't do any good if we keep introducing new pests."