Strategy Worries Crop Up in Biotechnology's War on Pests
Some Scientists Fear Genetically Modified Plants Could Pass Traits to
Enemy Weeds or Kill Helpful Insects
By Rick Weiss
Washington Post Staff Writer
Monday, September 21, 1998; Page A03
Lush fields of corn, soybeans and cotton look peaceful from a distance,
but peek between the leaves and there's a war going on. Ravenous insects
and aggressively encroaching weeds are constantly on the attack, and the
crops are responding with invisible salvos of naturally occurring
insect- and weed-repelling compounds.
American farmers tilted the balance of power this year with large-scale
plantings of new, genetically engineered crops armed with secret
weapons. Some of these plants have been endowed with a gene that
suffuses them with a potent insecticide called Bt. Others have special
genes that make them resistant to commercial herbicides, allowing
farmers to spray withering doses of weedkillers that previously would
have wiped out the crops as well.
But recent research suggests that the war is far from won. In one new
experiment, engineered plants spread their new herbicide tolerance genes
much more quickly than expected to surrounding weeds -- the molecular
equivalent of passing secrets to the enemy. Other work suggests that
Bt-producing plants may be killing not only targeted insects but also
beneficial insects that kill plant pests.
Findings such as these have escalated a long-standing war of words
between opponents and advocates of agricultural biotechnology.
"This says we need stronger oversight and we need more testing before
some of these crops are released," said Jane Rissler, senior staff
scientist at the Union of Concerned Scientists. U.S. regulatory agencies
do not require companies to document the rate at which engineered plants
pass their new genes to weeds, she notes -- a lapse she predicts will
spur the growth of "superweeds" and other ecological problems.
Not so, countered Thomas Nickson, an ecological technology coordinator
for Monsanto Co., the St. Louis-based plant biotechnology giant. "The
risk of creating a superweed," Nickson said, "is truly an insignificant
Behind these opposing opinions is a unified recognition that
agricultural pests represent a major threat to the global food supply.
Insects consume an estimated 13 percent of the world's food production.
To fight them, and to keep weeds at bay, costs billions of dollars every
Farmers have traditionally relied upon old-fashioned plant breeding to
make new varieties of plants with natural resistance to pests. But in
recent years, scientists have speeded up that process by placing
specific, useful genes directly into crops. Companies commercializing
this technology have promised environmental and economic savings,
especially through reduced dependence on chemical pesticides.
The total acreage of engineered crops has skyrocketed since the first
varieties went on sale three years ago. Fully 45 percent of this year's
U.S. cotton crop was engineered by Monsanto, either to protect it
against the company's popular Roundup herbicide or to make Bt to protect
its bolls against insects. Tens of millions of acres of engineered
soybeans, corn, canola and potatoes have also been planted, and
scientists are working on gene-altered sunflowers, squash and other
In some respects the effort has been a success. In the Southeast, for
example, insecticide doses on cotton have been reduced by as much as 45
percent and weed control has been made easier. But the latest studies
have cast a shadow on this promising landscape.
In the Sept. 3 issue of Nature, Joy Bergelson and colleagues at the
University of Chicago described their shocking discovery that mustard
plants engineered to be tolerant of a DuPont Co. weedkiller passed their
fancy genes to nearby wild mustard plants at a rate 20 times higher than
seen in ordinary mustard plants or plants that had naturally occurring
resistance to the herbicide.
It's not clear why the engineered plants became, as the journal put it,
so promiscuous. "No one ever expected the process of making a plant
transgenic to affect its outcrossing rate," said Allison Snow, an Ohio
State University ecologist.
Perhaps, Bergelson and others said, new genes can disrupt existing genes
-- including some controlling pollination and fertility -- when they are
blasted into plants, creating what Rissler calls a Viagra.
If true, that would undercut biotech companies' claims that
engineered plants pose no greater ecological risk than naturally
occurring mutant plants.
Making matters worse, Snow recently discovered that, contrary to what
experts had hoped, weeds that are pollinated by engineered plants can
pass those new genes to their offspring for generations. With additional
pollinations from other engineered varieties, that could lead to the
creation of weeds resistant to multiple herbicides.
Some engineered crops, such as cotton and corn, present little risk
because they have few weedy relatives to pass their herbicide resistance
to. But others -- including squash, rapeseed (canola), sunflowers and
sorghum -- have close wild relatives. If these weeds were to catch their
cousins' high-tech genes, they could quickly become bigger pests than
Scientists said they are especially concerned about sorghum, whose weedy
relative, Johnson grass, has become an enormous ecological pest since
its U.S. introduction from Africa in the mid-1800s. Monsanto has decided
not to pursue development of a Roundup-resistant sorghum because of that
risk, a spokesman said. But at least one other company has conducted
laboratory studies with engineered sorghum, and there is economic
incentive to improve that crop in Africa, where it is an important
In another potential problem, beneficial insects such as ladybug beetles
and lacewings may be inadvertently harmed by Bt when they feed on
insects that have fed on Bt plants, according to several studies
published in the past two years. In some cases, mortality rates for
these insect predators were about double those seen in predators raised
on Bt-free diets.
Scientists stressed that most of these worrisome results are not
definitive. In contrast to the work with ladybugs and lacewings, for
example, it appears that parasitic wasps -- which also kill insect pests
-- are not harmed by Bt-laden prey. And even if some insect predators
are killed, some argue, hardier beneficial species may move into those
Moreover, even Bergelson notes that transfer of a herbicide resistance
gene to a weed is just the first of many steps needed to make a
superweed. "It has to jump, it has to [function], it has to place those
individual weeds at a competitive advantage or free them from attack,
and the effect also has to be big enough to change the ecological
interaction," she said. "Our study looks at the first step only, and
In April, scientists reported that they had engineered crops in a clever
way that keeps new genes in the plants' photosynthetic organs and away
from the pollen, perhaps eliminating the risk of cross-pollination.
Other scientists quickly questioned, though, whether these plants will
prove less vigorous because of the added genetic burden in the
The agricultural arms race, it appears, has a long way to go. How will
"We're really flying by the seat of our pants right now," said John
Ruberson, an entomologist at the University of Georgia in Tifton. "Ask
the question in four or five years and we'll all be experts."
© Copyright 1998 The Washington Post Company