What is StarLink corn?
StarLink corn is one of several kinds of Bt corn that have
been genetically engineered to produce insecticide within
the plant itself so that external applications of pesticides
to fields can be reduced or eliminated. While other Bt corns
on the market in 1999 and 2000 produced a Bt toxin called
Cry1A(b), StarLink had Cry9C, a slightly different version
of the protein.
In tests required for government approval to grow the
crop, the Cry9C protein had been slower to break down under
artificial digestibility tests than Cry1A(b) and had raised
the suspicions of EPA reviewers by exhibiting several other
characteristics of allergens. Because the issue of Cry9C
allergenicity was unresolved, the EPA granted permission
to grow the crop as long as it was not used for human food.
Since the majority of the corn harvest in the United States
is used for animal feed and the production of fuel alcohol,
this restriction did not spell the end of the line for StarLink
corn as a crop. Farmers would simply have to ensure that
their StarLink harvest was directed into channels that didn't
lead to the human food supply.
|This decision to approve a so-called "split"
registration has been the target of much criticism.
Although Aventis, the agbiotech company that owned StarLink,
promised to have farmers sign agreements that their
corn would not end up as human food, in practice it
would have been difficult to guarantee. Some field corn
varieties, such as white corn destined for food use,
are kept separate during the handling process, but most
of the country's field corn--the kind that ends up in
tacos and animal feed and automobile tanks--is stored
and distributed in bulk. Grain dealers don't maintain
separate silos for the human food supply and the animal
In addition, corn pollen often is blown into nearby fields
where it can pollinate other varieties of corn. A low level
of cross-pollination is acceptable among varieties that
are approved for human consumption, but cross-pollination
would be unacceptable under the zero tolerance for StarLink
in human food.
The split registration proved to be an immediate failure.
A survey of 230 farmers, commissioned by Aventis in December
1999, indicated that two farmers had sold their StarLink
harvest for human food or for export and another 29 did
not know to what use their crop had been directed after
The results of the survey were reported to the government.
The failure to act on this early notice that the split approval
was not being honored has been another point of criticism.
Because Aventis was still asking for full approval instead
of split approval, the EPA prepared in the spring of 2000
to test StarLink for allergenicity. A Scientific Advisory
Panel was established to suggest methods for testing and
criteria for acceptability of StarLink for human food. But
it was too late.
StarLink found in grocery stores
In September 2000 an independent laboratory confirmed that
samples of taco shells submitted for testing by an anti-transgenic
activist group had tested positive for the presence of StarLink
Within days, supermarkets were pulling corn products from
their shelves. Aventis halted sales of StarLink seeds for
the spring 2001 planting and announced it would purchase
the entire 2000 harvest of Starlink corn to prevent any
further use of the corn in food products.
|EPA administrators began
to reconsider the split approval. The EPA has since
said that it will never again approve a biotech crop
for split use. Any crop that has both human food and
feed or industrial uses will have to meet the standards
for human food use in order to gain government approval.
Tracking the transgene
The EPA usually does not require a company to develop analysis
methods that will detect the presence of an approved transgenic
ingredient, and none was required for the Bt in StarLink
corn. The government and Aventis scrambled to develop a
test. So did several private companies. Soon grain elevators
and seed companies began testing their supplies for the
presence of StarLink DNA or protein. Recommended testing
procedures were published on the USDA's website at http://www.usda.gov/gipsa/biotech/starlink/cry9cdetection.htm.
This example, using a test for Bt
Cry1Ac, shows one kind of test available for Bt protein.
The top result is for conventional corn. There is
only one pink line across the stick. The bottom result
is for Bt corn. There are two pink lines across the
stick, indicating that Bt Cry1Ac protein is present
in the corn sample that was tested.
Although the StarLink acreage was not large, StarLink
DNA or protein was detected in hundreds of products. Some
corn varieties that were not supposed to have the StarLink
gene tested positive for StarLink. Eventually StarLink DNA
was detected in products made from white corn (http://washingtonpost.com/wp-dyn/articles/A16045-2001Jul3.html),
although no white corn StarLink varieties had been produced.
Foreign buyers quickly became nervous. Some countries already
had in place prohibitions against importing transgenic food
stuffs. Other countries made moves to institute such prohibitions.
The StarLink episode highlighted a problem with the entire
U.S. corn supply: the U.S. didn't keep conventional and
transgenic loads separate. Buyers turned to South Africa,
China, Argentina and Brazil for corn that was free of StarLink
In the spring of 2001, the president, general counsel,
and vice president for market development of the U.S. crop
sciences division of Aventis CropScience were fired. A spokesperson
for Aventis said it was fair to link the firings to the
StarLink fiasco. (A story on the dismissal of the three
executives is available at http://www.guardianunlimited.co.uk/gmdebate/Story/0,2763,437347,00.html.)
|Meanwhile, Aventis was attempting to win temporary
approval for small amounts of StarLink in human food
until the existing corn supplies were used up. Aventis
submitted several volumes of test results and analysis
to bolster the request for temporary approval. The Cry9C
protein was degraded by moisture and heat, Aventis argued,
so cooking would destroy most of the protein. People
who ate StarLink corn products would be exposed to extremely
small amounts of the protein, not enough to cause any
harm. Aventis commissioned the production and testing
of tortillas, chips, and corn bread made from 100% StarLink
corn to demonstrate the breakdown of the protein during
the cooking process. The Aventis request and the accompanying
data are available at http://www.epa.gov/pesticides/biopesticides/
Woman shelling corn in Guatemala
But the EPA's scientific advisory panel found the evidence
insufficient. Panel members suggested that the testing procedures
developed by Aventis might not accurately measure the amount
of protein in food. They decided that no safe limit for
StarLink could be determined from the data submitted. The
government should be more aggressive about collecting information
on possible allergic reactions, they suggested. Relying
of the opinion of the scientific advisory committee, the
EPA refused to allow StarLink in food. The scientific panel's
report is available at http://www.epa.gov/scipoly/sap/2001/july/julyfinal.pdf.
The Centers for Disease Control investigated claims that
51 people had suffered allergic reactions shortly after
eating corn products, but concluded that none of the reported
symptoms could be attributed to the StarLink protein. The
CDC's report is available at http://www.cdc.gov/nceh/ehhe/Cry9cReport/.
As the controversy continued, seed companies gradually
screened their stocks for StarLink. By the summer of 2001
the amount of StarLink in corn supplies was estimated to
be less than 0.125%.
In September 2003, a coalition of groups from several regions
of Mexico reported that the GM gene from StarLink corn had
entered the native corn populations of Mexico. The commercial
cultivation of any kind of GM corn is prohibited in Mexico
because of concerns about gene flow to Mexico's indigenous
corn varieties, but GM corn kernels can be imported for
use as food. A press release on this discovery is available
In November 2003, scientists reported that additional tests
had failed to demonstrate the presence of an allergy in
a person who had complained of allergic reactions to StarLink
corn. The person was fed StarLink corn, non-StarLink corn,
and a placebo on different days under conditions in which
neither he nor the technicians who provided the food knew
which kind he was receiving. The person did not experience
allergic reactions on any of the days of testing. A report
on the tests is available at http://www.nytimes.com/2003/11/10/business/10cornxx.html.
Colored corn varieties from Mexico
|The StarLink episode is a case study in
the issues related to transgenic technology: the adequacy
of governmental regulation, the ability to control gene
flow, consumer acceptance of transgenic products at
home and abroad, and questions about the safety of the
products themselves. The U.S. experience with StarLink
corn may help to improve future policies as nations
sort out the risks and benefits of transgenic technology.
More information on StarLink corn is available from the
The StarLink Case: Issues for the Future (http://www.pewagbiotech.org/research/starlink/starlink.pdf)
Michael Taylor and Jody Tick of Resources for the Future
describe the StarLink controversy and pose questions about
future handling of biotech foods. Resources for the Future
is an economic think tank supported by government grants
and donations from foundations and corporations.
Environmental Protection Agency web site
EPA's StarLink news archive
Aventis Crop Science web site, list of press releases
for 2000 and 2001
StarLink News page on this web site
More information on Bt corn is available from the following
Bt corn questions and answers html
Dr. Frank Peairs in the department of Bioagricultural
Sciences and Pest Management at Colorado State University
developed this set of questions and answers about Bt corn.
Because StarLink is one kind of Bt corn, this information
may be of interest to some viewers.