Taco dispute underscores need for standardized tests

URL: http://www.nature.com/cgi-taf/DynaPage.taf?file=/nbt/journal/v18/n11/full/nbt1100_1136b.html

Nature Biotech, November 2000 Volume 18 Number 11 pp 1136 - 1137

Emma Dorey

                On October 12, Safeway (Pleasanton, CA) removed its

                  home-brand taco shells from supermarket shelves after Genetic

                  Engineering Food Alert (GEFA; Washington, DC), a coalition of

                  seven anti-biotechnology groups, said traces of a GM corn not

                  yet approved for human consumption had been found in the

                  shells. The move follows that by Kraft Foods (Northfield, IL),

                  which recalled all Taco Bell Home Originals taco shell products

                  on September 22 after tests commissioned by GEFA

                  suggested the presence of the same GM corn. It was the first

                  time a product containing GM ingredients had been withdrawn

                  in the US, and the episode sparked a flurry of demands for

                  tightening of biotech regulations including discussions between

                  members of congress and the FDA commissioner Jane

                  Henney. However, there has been no risk to human health, and

                  the incident only illustrates the need for standardized tests and

                  the importance that EPA set thresholds for the accidental

                  mixing of ingredients.

                  The corn in question, StarLink, has been genetically modified

                  by Aventis CropScience (Lyon, France) to express the

                  insecticidal protein Cry9C from Bacillus thuringiensis. It is

                  currently approved for animal consumption, but approval for

                  human food use is pending.

                  Genetic ID (Fairfield, IA), which was commissioned by Friends

                  of the Earth to test 23 products after the group "heard that

                  genetically engineered corn is deadly to monarch butterflies",

                  says it found 1% of corn DNA in the taco shells was StarLink.

                  Kraft Foods subsequently withdrew the products from the

                  market, and Aventis halted sales of StarLink seed, agreeing to

                  pay the USDA to buy all StarLink corn grown this year (at least

                  $90 million) to prevent it from "tainting" the US food supply.

                  Rep. Dennis Kucinich (D-OH) and 15 members of congress

                  wrote to Henney reiterating calls for "stringent pre-approvals for

                  ag and biotech products" and anti-biotech activists had a field

                  day promoting the recall as a public health measure. The FDA

                  subsequently announced it would test a wide range of food

                  products for the presence of StarLink corn, and it is now likely

                  food manufacturers, hoping to avoid a recall of their own, will

                  begin specifying the use of conventional grain.

                  Lost in all this was that human health was not in jeopardy.

                  StarLink is only pending approval for food use because it did

                  not pass all of EPA's screens for allergenicity—Cry9C was not

                  immediately broken down in digestion tests, which can be a

                  sign of food allergens. "I do not believe there has been any risk

                  to the public," says Steve Taylor of the department of Food

                  Science and Technology at the University of Nebraska. He

                  points out that Cry9C actually does not resemble known

                  allergens, "so in fact may not be an allergen." In addition he

                  adds that most allergenic proteins are present at levels of

                  1–40%, but that Aventis indicates that the Cry9C protein is

                  present in corn kernels at 0.3% and that taco shells would

                  contain far less.

                  In any event, it was not Cry9C but the gene encoding for its

                  expression that was allegedly detected in the taco shells. "My

                  understanding is that the protein has been looked for

                  extensively and noone, but noone, has found any trace of the

                  protein," says Val Giddings, vice president of food and

                  agriculture at BIO. The gene is not suspected of having any

                  health effects whatsoever.

                  Moreover, there is some debate that StarLink DNA has actually

                  been detected. Jeffrey Smith, Genetic ID's vice president of

                  marketing communications, says the test was run three times,

                  each time in duplicate with a separate DNA extraction, and all

                  PCR reactions were consistent. In addition, he says the result of

                  the PCR reaction with the Starlink specific primer set was sent

                  to another laboratory for sequencing to verify the sequence was

                  StarLink. Genetic ID, he says, is "extremely confident in our

                  procedures and in the result." In addition, Kucinich says that

                  Kraft and the FDA have confirmed the presence of StarLink

                  corn in the taco bell shells.

                  However, Aventis spokesperson Rick Rountree says Aventis

                  has "not been able to re-confirm [Genetic ID's] results," and that

                  "There is still a strong likelihood that another genetically

                  modified ingredient in a corn that is approved for food use may

                  have been there and could be triggering a false positive."

                  Richard Birkmeyer, president and CEO of Strategic

                  Diagnostics (SDI; Newark, DE), which produces protein tests

                  based on immunassay technology, explains: "The more

                  processed the food," he says, "the less accurate and reliable

                  the results, no matter what method, because you're destroying

                  protein and DNA and you can get very erroneous results."

                  DNA tests based on PCR (like the one used by Genetic ID) are

                  very reliable if done properly, says Giddings, but there are lots

                  of ways to generate false positives, particularly if used on

                  processed foods. "None of the tests [out there] are validated for

                  use on processed foods," he adds. Indeed, Genetic ID, which

                  was founded by anti-biotech proponent John Fagan, dean of

                  Maharishi University of Management, has been involved in a

                  false-alarm dispute in the past.

                  Although not necessary from a human-health standpoint, the

                  controversy highlights the growing need for a reliable way to

                  detect the presence of GMOs, and several groups, including

                  Kraft, Kucinich, and BIO, advocate validated, standardized

                  tests.

                  Addressing this, the Grain Inspection, Packers & Stockyards

                  Administration (GIPSA; Washington, DC) is setting up the

                  Biotechnology Reference Laboratory (BRL; Kansas City, MO).

                  GIPSA deputy administrator David Shipman says the aim will

                  be to facilitate the marketing of grains by providing standard

                  testing references so that buyer and seller can have confidence

                  that the results they're getting are accurate and consistent. The

                  voluntary program will accredit test kits and analytical

                  laboratories on the basis of all sorts of criteria, including

                  detection of GM ingredients. In addition, European authorities

                  are in discussion with GIPSA about what form standardized

                  tests should take as part of the new GMO traceability

                  requirement of the European Council (Nat. Biotechnol. 18,

                  705;MEDLINE). Although not the aim of BRL, credible

                  independent verification of the reliability of the tests being

                  marketed may go some way to allaying irrational consumer

                  fears, and Birkmeyer, Kucinich, and Giddings think the BRL will

                  be good for the industry.

                  Meanwhile, much of the taco-shell palaver occurred because

                  the EPA has not yet established the tolerance level of Cry9C for

                  human food use. While the major risk to human health comes

                  from living microbes, most food chains or processing systems

                  have "aesthetic" limits for the accidental mixing of foreign

                  material. For example, the FDA will allow 25 insect fragments

                  and 2 rodent hairs or 1 "rodent excreta fragment" in 50 grams

                  of cornmeal, and 5 fly eggs or 2 maggots in every 100 grams of

                  tomato juice. The EPA must set a similar limit for Cry9C. "That

                  absolutely has to take place," says Kucinich.