March 2001

.pdf version

Engineering Multiple Genes in a Single Transformation Event
Cross-Pollination Leads to Triple Herbicide Resistance
Wheat Streak Mosaic Virus As A Gene Expression Vector
Results of Six-Month Review of Federal Biotechnology Policy Released for Comment
Outcry Over Cry9C
The StarLink Incident: Changing the Face of the US Grain Industry

Another Milestone In Plant Biotechnology

In the post-genomic era, with research emphasis increasingly placed on gene function, it has become clear that many of the desired agronomic traits under development will require the simultaneous engineering of multiple genes or pathways. In addition, new genetic engineering tools will be needed to fully exploit the knowledge gained from functional genomic research. Though the ability to provide the coordinated expression of multiple genes to produce valuable agronomic traits is considered the Holy Grail of plant biotechnology, this area remains a challenging one for those involved in nuclear genetic engineering.

Ye et al. successfully introduced three genes into rice nuclear genome to form a short biosynthetic pathway that expressed ß-carotene, resulting in "golden rice."1 However, this effort took seven years to accomplish. Despite the ability to transform a number of plant species with increased efficiency, position effect and gene silencing continue to hamper efficient expression of foreign genes via the nuclear genome, making it challenging to engineer a single plant to contain multiple genes. The introduction of multiple genes via the nuclear genome requires the generation of individual transgenic plants and subsequent backcrosses to reconstitute the entire pathway or multi-subunit proteins.

Routine expression of polycistrons via the chloroplast genome provides an extraordinary opportunity to express foreign pathways in a single transformation event, and this process is not subject to gene silencing or position effect. A study featured in Nature Biotechnology in January 2001 demonstrates the expression of a three-gene Bacillus thuringiensis Cry2Aa2 operon and a selectable marker, resulting in multiple-gene expression in a single step.2

In this study, the Bacillus thuringiensis (Bt) cry2Aa2 operon is used as a model system to demonstrate operon expression and crystal formation via the chloroplast genome. Cry2Aa2 is the distal gene of a three-gene operon. The orf 2 gene immediately upstream of cry2Aa2 codes for a putative chaperonin that facilitates the folding of the CRY protein (and others) to form proteolytically stable cuboidal crystals. CRY protein levels decrease in plant tissues late in the growing season or under physiological stress.3 A stabler protein expressed at high levels throughout the growing season is necessary to increase toxicity of Bt transgenic plants to target insects and help eliminate the development of Bt resistance.

Therefore, the cry2Aa2 bacterial operon was expressed in tobacco chloroplasts to test the resultant transgenic plants for increased expression and improved persistence of the accumulated insecticidal protein.2 Stable foreign gene integration was confirmed by PCR and Southern blot analyses in T0 and T1 transgenic plants. Cry2Aa2 operon-derived protein accumulated at 45.3% of the total soluble protein in mature leaves and remained stable even in old bleached leaves (46.1%), thereby increasing the efficacy of transgenic plants throughout the growing season. This is the highest level of foreign gene expression ever reported in transgenic plants. Insects known to be exceedingly difficult to control (10-day old cotton bollworm, beet armyworm) were 100% killed after consuming transgenic leaves.

One common concern arising from the use of nuclear transgenic crops expressing Bt toxins in suboptimal levels is the development of Bt-resistant pests. Plant-specific recommendations to reduce the development of Bt resistance include increasing Bt expression levels (high dose strategy), expressing the protein only in tissues highly sensitive to damage (tissue specific expression), or expressing multiple toxins (gene pyramiding). All three approaches are attainable through chloroplast genetic engineering. For example, we have recently shown that hyper-expression of several thousand copies of a Bt gene via chloroplast genetic engineering results in 100% mortality of insects that were up to 40,000-fold resistant to other Bt proteins.4

Electron micrographs showed the presence of the insecticidal protein folded into cuboidal crystals similar in shape to Cry2Aa2 crystals observed in Bacillus thuringiensis.2 In contrast to currently marketed transgenic plants with soluble CRY proteins, folded protoxin crystals will be processed only by target insects that have high alkaline gut environment; this approach should improve the safety of Bt transgenic plants.

Electron micrograph of Cry2Aa2 crystals (immunogold labeled) in a transgenic chloroplast expressing the cry2A operon

For example, it has been shown that there is a more than 30-fold concentration difference in the activity of soluble and crystalline CRY protein, and this difference is due to the host mid-gut alkaline environment.2 Plants transformed with the cry2Aa2 operon show a large accumulation and improved persistence of the expressed insecticidal protein(s) throughout the life of the plant. This is most likely because of the folding of the insecticidal protein into cuboidal crystals, thereby protecting it from cellular proteases.

Another hotly debated and controversial environmental concern is the potential toxicity of transgenic pollen to non-target insects such as the Monarch butterfly. This study demonstrates that even though chloroplasts in leaves contained as much as 47% CRY protein of the total soluble protein, the insecticidal protein was absent in transgenic pollen.2 Absence of insecticidal proteins in transgenic pollen eliminates potential toxicity to non-target insects via pollen.

Chloroplast genetic engineering is emerging as an alternative new technology that overcomes many of the environmental concerns of nuclear genetic engineering. One common environmental concern is the escape of foreign genes through pollen or seed dispersal from transgenic crops to their weedy relatives, creating super weeds, or, among other crops, causing genetic pollution. Although pollen from a few plants contains metabolically active plastids, the plastid DNA itself is lost during the process of pollen maturation and hence is not transmitted to the next generation. Maternal inheritance of foreign genes through chloroplast genetic engineering is highly desirable when there is potential for outcrossing among crops or between crops and weeds.5

Introducing blocks of foreign genes in a single transformation event would avoid complications inherent in putting one gene at a time into random locations in the nuclear genome, a process that results in position effects or gene silencing. This is especially true for multi-subunit biopharmaceutical proteins (such as monoclonals) in which subunits should be synthesized in stoichiometric amounts for complete assembly. Crystallization of foreign proteins should also serve as a model system for large-scale production of foreign proteins within chloroplasts in a folded configuration, which enhances their stability and facilitates single step purification via centrifugation. This is the first demonstration of the expression of a bacterial operon in transgenic plants and opens the door to engineering novel pathways in plants in a single transformation event.

Chloroplast genetic engineering has been accomplished so far only in tobacco and potato. However, genes for herbicide, insect, and pathogen resistance and drought tolerance have been expressed to very high levels, conferring the desired traits. Chloroplast genome has been shown to express biopharmaceuticals, including human proteins, at levels 300-fold higher than nuclear expression, resulting in a properly folded and fully functional configuration. Efforts are underway to extend chloroplast genetic engineering to other economically important crops both in academic and industrial laboratories. All of these findings augur well for environmentally friendly genetic engineering approaches in the next generation of transgenic crops.


1. Ye X, Al-Babili S, Klöti A, Zhang J, Lucca P, Beyer P, and Potrykus I. 2000. Engineering the provitamin A (b-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science 287: 303-305.

2. De Cosa B, Moar W, Lee SB, Miller M, and Daniell H. 2001. Overexpression of the Bt cry2Aa2 operon in chloroplasts leads to formation of insecticidal crystals. Nature Biotechnology 19: 71-74.

3. Greenplate J. 1999. Quantification of Bacillus thuringiensis insect control protein Cry1Ac over time in bollgard cotton fruit and terminals. Journal of Economic Entomology 92: 1377-1383.

4. Kota M, Daniell H, Varma S, Garczynski SF, Gould F, and Moar WJ. 1999. Overexpression of the Bacillus thuringiensis (Bt) Cry2A protein in chloroplasts confers resistance to plants against susceptible and Bt-resistant insects. Proceedings of the National Academy Science USA 96: 1840-1845.

5. Daniell H, Datta R, Varma S, Gray S, and Lee SB. 1998. Containment of herbicide resistance through genetic engineering of the chloroplast genome. Nature Biotechnology 16: 345-348.

Henry Daniell
Department of Molecular Biology & Microbiology
University of Central Florida


One of the risks frequently cited in association with transgenic crops is the escape of a foreign gene via sexual reproduction. The recipient plant in such cases may be a non-transgenic variety of the same crop or a sexually compatible relative. Depending on the gene and trait considered, adverse environmental or agricultural impacts may result from such transfers, ranging from issues of genetic purity of neighboring crops to the generation of "super weeds." While this issue is receiving increasing attention by researchers, a recent report by Hall et al.1 describes a truly remarkable example of herbicide resistance transfer via pollen among Brassica napus varieties. What is unusual here is not so much that it happened at all, but that it occurred rapidly and multiple times, such that, through completely random crossing, certain plants were found to be resistant to three different herbicides.

The problem

The events that precipitated this study occurred in Alberta, Canada where in 1997 a producer planted three varieties of B. napus in the same or adjacent fields. One variety was resistant to glyphosate, another was resistant to glufosinate, and a third resistant to herbicides of the imidazolinone type. In field 1, the producer started planting glufosinate-resistant B. napus, but, after 15 hectares had been sowed, switched to imidazolinone-resistant B. napus for the remainder of the field. In field 2, which was located across a road (22 m from the edge of field 1 containing the glufosinate-resistant plants), the producer planted glyphosate-resistant B. napus. The next year, the producer fallowed field 1 and planted part of field 2 with imidazolinone-resistant B. napus. Weeds in the fallowed field 1 were sprayed with glyphosate, but the producer noticed that B. napus volunteers in this field were not being controlled by the herbicide.

At this point, Hall and company stepped in and surveyed field 1 to determine whether herbicide resistance had developed. They collected B. napus volunteers that had survived the glyphosate treatments and grew them in the greenhouse in order to obtain seeds for further study. The resulting generation was then tested for resistance to glyphosate by spraying seedlings with the herbicide. They found that 20 of 34 individuals tested had a high percentage of offspring showing resistance to glyphosate. This result could be accounted for by either of two mechanisms: 1) direct movement of glyphosate-resistant seeds from field 2 across the road to field 1 (most likely transported by farm equipment), or 2) movement of pollen such that cross-hybridizations occurred between the glyphosate-resistant variety and either the glufosinate- or imidazolinone-resistant varieties. Although mechanical operations (planting and harvesting) in the two different fields were not performed at the same time, the direct movement of seed from field 2 to field 1 remained a possibility.

Evaluation of glyphosate-resistant volunteers

To confirm whether pollen flow was responsible for the glyphosate resistance, Hall et al. set forth three tests. First, if the volunteer plants had gained glyphosate resistance through pollen flow, this would have been conferred by the male parent, while glufosinate or imidazolinone resistance would have been inherited from the maternal plant. Thus, they would be resistant to two herbicides rather than just one. Second, if the resistance trait had been transferred via pollen, the progeny of the resistant plants would show a pattern of segregation for resistance consistent with Mendelian ratios for a single-locus dominant trait. Third, RFLP analysis of the progeny using markers specific for each of the parental lines would indicate whether the volunteer plants were products of a hybridization of two lines. Eight volunteer B. napus plants were put through these tests and seven passed them all, demonstrating that gene flow via pollen was the primary mechanism of resistance transfer.

Progeny of additional plants were characterized solely on the basis of resistance to herbicides. In the section of field 1 originally planted with glufosinate-resistant B. napus, all nine plants tested had progeny resistant to both glyphosate and glufosinate. In the section planted with imidazolinone-resistant plants, the progeny of 10 of 11 plants were resistant to both glyphosate and the imidazolinone herbicide, imazethapyr (the lone exception to this being a glufosinate-resistant individual that was likely moved as a seed from the nearby glufosinate resistant section of the field).

Triple resistance

In addition to the cases of double resistance, two plants from field 1 gave rise to progeny resistant to all three herbicides. This was attributed to sequential hybridization among the plants. In one such case it is probable that a glufosinate-resistant plant was pollinated by a glyphosate-resistant plant in 1997. The following year, a progeny of this plant was selected by application of glyphosate to kill competing vegetation and was subsequently cross-pollinated by imidazolinone-resistant B. napus planted in field 2 in 1998.

The triple resistance in the second plant is proposed to have arisen by a different sequence of events, with the first cross occurring between glufosinate- and imidazolinone-resistant plants. A progeny of this cross is thought to have escaped glyphosate treatment and crossed with one of the glyphosate-resistant volunteers. Although the end result is the same, this illustrates how all possible combinations and sequences of events are possible.

Lessons from this situation

This report demonstrates that pollen flow among individuals of an outcrossing species can be a very effective method for transmitting genes. B. napus is capable of both selfing and outcrossing, having an outcrossing frequency of 20-30%. In the field, the actual cross-hybridization rate is a function of distance, with percent outcrossing diminishing the farther the recipient is from the pollen source. It is therefore interesting that one of the triple-resistant plants was found over 550 m from the pollen sources, greatly exceeding the 100-m buffer mandated for seed producers. Also, in cases such as this field situation where large numbers of plants are involved, even a low percentage of outcrossing can result in significant transfer of genes via pollen. It is important to note that this research is dealing with intraspecific hybridization. Hybridization between Brassica crops and several wild relatives has been reported, but may occur at lower frequencies and hybrid plants may suffer from lack of vigor or fertility.

The authors point out that the circumstances that gave rise to the triple-resistant B. napus are highly unusual in that three varieties harboring different herbicide-resistance genes were planted in close proximity in the same year. Then the next year the only weed control method used in the fallowed field 1 was the herbicide glyphosate. B. napus is susceptible to many weed control measures, so in the second year the glyphosate-resistant volunteers should properly have been controlled by herbicides with different modes of action, tillage operations, or cultural practices. Fortunately, seeds of B. napus have a relatively short persistence (duration of dormancy of four to five years in the soil) so proper management practices will eliminate multiple-resistant weed seeds in a relatively short period of time. Nevertheless, this report points out just how rapidly genes can move within an outcrossing crop and why planting distance and crop rotation precautions and herbicide/weed control techniques need to be varied regularly to avoid developing problematic volunteer weeds. (For more information, see "Outcrossing Between Canola Varieties - A Volunteer Canola Control Issue" at < http://www.agric. >.) This example should serve as a warning to producers to use their new herbicide-resistant crops wisely according to guidelines.


Hall L, Topinka K, Huffman J, Davis L, and Good A. 2000. Pollen flow between herbicide-resistant Brassica napus is the cause of multiple-resistant B. napus volunteers. Weed Science 48: 688-694.

Jim Westwood
Department of Plant Pathology, Physiology, and Weed Science
Virginia Tech


One wheat disease agent may soon be used as a tool for expressing foreign genes in plants. Agricultural Research Service (USDA) plant pathologists Drake C. Stenger and Roy C. French have developed a strategy for using wheat streak mosaic virus (WSMV) as a vehicle for transient gene expression in cereal species without plant transformation.1 Studies on WSMV show its capability as an accurate and feasible gene expression vector.2

WSMV is a single-stranded plus-sense RNA virus in the family Potyviridae, genus Tritimovirus. Its genome is comprised of 9,384 bases encoding a polyprotein of 3,035 amino acids. The polyprotein is cleaved into several mature proteins through the action of virus-encoded proteinases.3 A complete genomic map of WSMV is accessible at

In nature, WSMV is transmitted from plant to plant through the feeding activity of the wheat curl mite (Aceria tosichella). In the laboratory, the virus is transmissible by mechanical inoculation of either virions or as the naked RNA alone. WSMV naturally infects wheat (Triticum aestivum), causing severe leaf mosaic and crop loss, and causes a mild mosaic disease in some corn (Zea mays) cultivars. WSMV also infects some wild grasses, including Japanese millet, switch grass, and weedy foxtail.

To assess the suitability of WSMV as a transient gene delivery system, Stenger and French initially introduced neomycin phosphotransferase II (NPT II) and ß-glucuronidase (GUS) genes into the WSMV genome. The foreign genes were inserted between the nuclear inclusion b (NIb) and coat protein (CP) domains of the WSMV polyprotein. NIb is a component of viral RNA polymerase whereas CP is the major structural component of the virion. The NPT II insert was stable for 18-30 days after inoculation.

Expression of NPT II showed little effect on viral CP expression, meaning there was no interference with viral replication. GUS protein was expressed in functional form in wheat and barley without interfering with viral activity. However, the GUS insert did reduce viral virulence in corn and oat. Expression of both NPT II and GUS was similar in wheat leaf and root trials.2 Unfortunately, the team discovered, using RT-PCR, that the GUS inserts were subject to deletion from the viral genome, particularly when fused to NIb.

Current strategies for wheat transformation used by Shirley Sato at the University of Nebraska include particle bombardment and Agrobacterium tumefaciens. Both techniques produce stable transgenics; however, additional improvements are needed before transgenic wheat is ready for commercial production. Stenger and French anticipate that WSMV will complement wheat transformation efforts by permitting a preliminary evaluation of foreign gene effects prior to the production of transgenic plants. WSMV is related to numerous other cereal-infecting potyviruses that also may be suitable for use as expression vectors.

More information is available by contacting Stenger at or French at rfrench@


1. Suszkiw J. 2000. Wheat streak mosaic virus: From harmful to helpful. Agricultural Research 48(12): 18.

2. Choi I-R Stenger DC, Morris TJ, and French R. 2000. Technical advance: A plant virus for systemic expression of foreign genes in cereals. Plant Journal 23(4): 547-555.

3. Choi I-R, Stenger DC, and French R. 2000. Multiple interactions among proteins encoded by the mite-transmitted wheat streak mosaic tritimovirus. Virology 267(2): 185-198.

Brian R. Shmaefsky
Department of Biology and Environmental Sciences
Kingwood College


In May 2000, President Clinton directed the Council on Environmental Quality (CEQ) and the Office of Science and Technology Policy (OSTP) to conduct a six-month interagency assessment of federal environmental regulations regarding agricultural biotechnology and to make recommendations to improve them.1,2 To conduct the assessment, CEQ and OSTP established an Interagency Working Group composed of individuals from agencies in the Departments of Agriculture, Commerce, Health and Human Services, Interior, Justice, and State, and from the Environmental Protection Agency (EPA) and Office of Management and Budget. The Working Group approached the assessment in the context of six case studies, describing in detail how specific products are regulated or how they might potentially be regulated. The case studies considered current agency practices, identifying strengths and potential areas of improvement. Due to time limitations, the Interagency Working Group did not develop conclusions or recommendations. The results of the six-month review were released to the public on January 22,3 opening a public comment period through May 1, 2001.

The case studies covered a range of biotechnology products, including some that have been approved for commercial production, others that are currently under regulatory consideration, and others that have not been presented for regulatory review. The case studies were considered within the context of the Coordinated Framework for Regulation of Biotechnology,4 which applies existing statutes and their regulations and guidelines for implementation to products of biotechnology introduced into the environment. The six-month review showed how many of these existing instruments apply within the contexts of the case studies. However, it was clear that not all aspects of biotechnology regulation are adequately addressed within the Coordinated Framework. In such cases, regulatory policy is evolving through formal and informal understandings between agencies regarding how a particular organism will be regulated. In some cases, the review showed that statutes not currently used for biotechnology oversight might be applied. In other cases, however, it was unclear whether any statute applies, and it will be necessary for agencies to determine an appropriate oversight strategy.

The six case studies and four associated sidebars were chosen to represent a range of organisms, relevant statutes, and public interest. A brief description of each case study and key findings are presented below.

Salmon expressing a growth hormone gene

Production of Atlantic salmon in floating marine net-pens poses a near-term regulatory issue.5 There is a high probability of escape from such confinements and a possibility of impacts on conspecific populations or other components of the receiving ecosystem. The lead agency drafting the case study was the Food and Drug Administration (FDA), with representatives from the National Marine Fisheries Service and the Department of Interior (DOI). Relevant statutes applied by the Coordinated Framework include the Federal Food, Drug, and Cosmetics Act (FFDCA) and the National Environmental Policy Act (NEPA). The case study identified other statutes that could be invoked to authorize more effective federal oversight of transgenic salmon, including the Endangered Species Act (ESA), the Lacey Act (concerning injurious wildlife species), the Non-Indigenous Aquatic Nuisance Prevention and Control Act, and a portion of the Rivers and Harbors Act. A sidebar on goldfish expressing an antifreeze polypeptide gene was developed under the lead of the Department of the Interior to explore issues posed by regulation of transgenic ornamental fishes. In addition to many of the statutes mentioned for the salmon, the Toxic Substances Control Act (TSCA) might also be applied.

Bt maize

Maize expressing a toxin from Bacillus thuringiensis is widely grown in the United States, and possible impacts on non-target organisms have been the subject of public and scientific debate. In addition to the Federal Insecticide, Fungicide, and Rodenticide Act, FFDCA, Federal Plant Pest Act (FPPA), Plant Quarantine Act (PQA), and Plant Protection Act (PPA), the case study discussed the possible application of the Migratory Bird Treaty Act, and the ESA. A sidebar briefly discussed how NEPA and many of the statutes mentioned for Bt maize might be applied to authorize oversight of viral DNA sequences expressed as microbial pesticides. The EPA led the drafting of this case study, with participation by USDA-Animal and Plant Health Inspection Service (APHIS) and DOI.

Herbicide-tolerant soybean

Herbicide-tolerant soybeans are widely grown and have the potential to significantly affect how herbicides are used to control agriculturally important weeds. APHIS, with help from EPA and DOI, identified the Federal Insecticide, Fungicide, and Rodenticide Act, FFDCA, FPPA, PQA, and PPA as the principle authorizing statutes and discussed the possible application of the Migratory Bird Treaty Act and ESA. A hypothetical pharmaceutical-producing plant was discussed in a sidebar to bring out environmental exposure issues posed by production under some conditions. FDA and APHIS identified the Virus-Serum-Toxin Act, Public Health Service Act, FFDCA, PPA, and NEPA as possible authorizing legislation.

Animals producing human drugs

Depending on confinement conditions, animals expressing pharmaceutical proteins potentially present environmental exposure issues. In this hypothetical case study, FDA, with help from APHIS and the Food Safety Inspection Service, identified the Public Health Service Act, FFDCA, and NEPA as relevant statutes. A sidebar was developed by APHIS, FDA, and the Food Safety Inspection Service to explore the somewhat different issues posed by production of animal biologics. Statutes discussed included Virus-Serum-Toxin Act, the Animal Quarantine Laws, TSCA, and the Animal Welfare Act.

Bioremediation using poplar trees

Perennial plants present regulators with a different set of environmental exposure issues than annual plants such as corn or soybean. Although transgenic trees are not commercially produced, the case study focusing on poplars was developed by the USDA-Forest Service, APHIS, EPA, and DOI to consider oversight of a perennial plant. The principal statutes that were identified as applicable were FPPA, PQA, PPA, and TSCA.

Bioremediation and biosensing using bacteria

EPA, DOI, and APHIS developed this case study to describe oversight of genetically modified bacteria that are neither plant pests nor pesticides. The principal statute discussed in this case study was TSCA.

In the announcement accompanying the release of the six-month review, OSTP and CEQ asked for public comment on the case studies regarding the comprehensiveness and rigor of environmental assessment, the comprehensiveness and strength of regulatory authority, transparency of the environmental assessment and decision-making process, public involvement, interagency coordination, and handling of confidential business information.

Public comments are requested by May 1, 2001. After the comment period, OSTP and CEQ will continue the assessment process and recommend appropriate steps to strengthen the science-based regulatory system. To access a copy of the assessment, go to <>, click on "What's New," and on "CEQ/OSTP Assessment of Environmental Regulation for Biotechnology." Comments should be submitted to OSTP and CEQ at addresses provided in these documents.


1. Office of the Press Secretary, The White House. 2000. Clinton Administration agencies announce food and agricultural biotechnology initiatives: Strengthening science-based regulation and consumer access to information. May 3, 2000.
< >

2. Hallerman EM. 2000. Administration announces sweeping assessment of federal agricultural and food biotech regulations. ISB News Report, June. <>

3. CEQ/OSTP. 2001. CEQ/OSTP Assessment: Case studies of environmental regulation for biotechnology. <>

4. OSTP. 1986. Coordinated framework for the regulation of biotechnology. 51 Federal Register 23301-23350 (June 26).

5. Hallerman EM. 2000. Commercialization of path-breaking transgenic salmon faces stumbling blocks. ISB News Report, April.

Eric M. Hallerman
Department of Fisheries and Wildlife Sciences
Virginia Tech

The StarLink incident continues to have broad-reaching impacts on the agbiotech industry, affecting federal regulations, trade restrictions, crop segregation methods, producers, and farmers, and even spawning a handful of lawsuits. The following two articles provide a review of some of ramifications of the StarLink event for agbiotech and what is being done to close the loopholes that allowed the unapproved corn to find its way into food products. (See also: " Unapproved Corn Slips Through Regulatory Net," ISB News Report, October 2000.) - Ruth Irwin, Editor


On September 22, 2000, Kraft Foods (Northfield, IL) announced its voluntary recall of all Taco Bell Home Originals taco shells and taco dinners sold nationwide in retail grocery outlets. The company warned consumers who had purchased the products to return them uneaten, because the food contained remnants of unapproved genetically modified corn.

The corn at the kernel of the controversy was StarLink, which expresses the insecticidal protein Cry9C from Bacillus thuringiensis (Bt) subspecies tolworthi. Commercialized by Aventis CropScience USA LP (Research Triangle Park, NC), StarLink was grown on less than one percent of US corn fields, but was inadvertently commingled with a large amount of conventional corn. Aventis SA (Strasbourg, France) estimates the company's total cost associated with the StarLink matter could run between US $100 million and $1 billion. Moreover, the controversy may set back commercialization of agricultural biotechnology and lead to a reexamination of America's regulatory system.

A ticking bomb

In 1998, the Environmental Protection Agency issued a registration to Plant Genetic Systems, Inc. (Des Moines, IA) for StarLink corn. Later that year, the registration was conveyed to AgrEvo USA (Wilmington, DE), and, subsequently, AgrEvo and Rhone-Poulenc Ag Company formed Aventis CropScience.

The EPA regulates Cry endotoxins in Bt corn as plant pesticides. The Agency can grant an exemption from a food tolerance requirement if it finds a reasonable certainty that aggregate exposures to the endotoxin will not cause harm. In the case of Cry9C, the EPA granted a limited exemption that restricted utilization to animal feed and industrial non-food uses. The agency did not extend the exemption to human food because there was concern within the EPA that Cry9C, which is not readily digested, might be a food allergen.

According to the EPA-approved plan, when farmers bought StarLink seed, they would be told that corn grown from it could not be sold for human consumption, and they would sign a "Grower Agreement" to that effect. Moreover, the seeds would come with instructions explaining the need for a 660-foot buffer strip between the planting of StarLink and other varieties. Any corn grown within the buffer zone would have to go to the feedlot, or it would be slated for a non-food use.

For the plan to work, information about StarLink would have to pass from Aventis to seed companies, then to the seed dealers, and finally to the farmers who bought the seed. However, the message did not always get through to the farmers. Furthermore, it appears that confusion was caused by the fact that corn varieties typically are not approved only for animal consumption. In short, a two-track marketing system was set into motion that required segregation of StarLink as it traveled through a complex world of grain production, transportation, storage, and processing. And there would be zero tolerance for errors.

The explosion

On September 18, 2000, Larry Bohlen held a press conference to announce that taco shells purchased from a local grocery store contained trace amounts of GM corn DNA associated with StarLink. The event was convened by Genetically Modified Food Alert (a consortium of seven consumer organizations based in Washington, D.C.), which sent a letter to Phillip Morris Companies, Inc., the parent company of Kraft Foods, about the test results.

In early October the FDA entered the picture. The FDA is charged with preventing food from reaching the market if the food contains unapproved additives or pesticides. After confirming the presence of ingredients from unapproved StarLink corn, the FDA announced that it would officially recall the taco shells. By then, however, Kraft Foods had recalled 2.5 million boxes of Taco Bell Home Originals taco shell products.

Following Kraft's action, a number of other food manufacturers issued recalls for products made from corn. Eventually, nearly 300 products were recalled. In addition, food processors like Archer Daniels Midland Co. (Decatur, IL) and Cargill, Inc. (Minneapolis, MN) began testing incoming shipments for StarLink and rejecting rail cars of corn.

The "contamination" problem was not confined to the US. In Japan, StarLink has not been approved for consumption by either humans or animals. Yet the Consumers Union of Japan reported that StarLink protein had been detected in cornmeal. Exports to Japan, the largest foreign market for US corn, dropped by about two-thirds. South Korea, the second largest consumer of US corn, banned them outright.

The fallout: Aventis

Following Bohlen's announcement, Aventis stopped sales of StarLink seed and reached an agreement with the Department of Agriculture to purchase StarLink corn with the objective of isolating the corn from the US food supply. Aventis would pay growers for StarLink corn (and any corn grown within 660 feet of StarLink corn) if the growers did not intend to use the corn as feed on the farm. In late January, Aventis formalized this agreement with the attorneys general of 17 states and included compensation to grain elevator operators for losses related to StarLink corn. Aventis also agreed to pay farmers whose non-StarLink corn had been inadvertently commingled either with StarLink corn or with corn grown within the buffer zone.

Back in October, at the urging of the EPA, Aventis announced that it was canceling the registration of StarLink corn, which would mean that StarLink could no longer be planted for any agricultural purpose. However, there was still the problem of StarLink remnants in commercially available processed foods. To avoid more product recalls, Aventis petitioned the EPA to expand the Cry9C tolerance requirement exemption to a four-year exemption for foods made from StarLink corn. Aventis contended that this time frame would allow all processed foods conceivably made from StarLink corn to pass through the channels of trade. Aventis supported the petition with new data on the potential allergenicity of the Cry9C protein.

The EPA sought an independent review of the Aventis data. The agency's scientific advisory panel found that there is a "medium likelihood" that Cry9C protein is a potential allergen, but that given the low level of Cry9C protein entering the human diet, there is a "low likelihood" that StarLink corn has resulted in sensitization of some individuals to the Cry9C protein.

The EPA is continuing its evaluation of the scientific information. In the meantime, Aventis faces three class action lawsuits alleging damages caused by a loss in domestic and foreign markets for US corn. In addition, two people have sued Aventis claiming that they had various allergy symptoms after eating Kraft taco shells. By the end of November, Aventis SA announced plans to sell off Aventis CropScience. Aventis said the decision to divest the unit was independent of the criticism over the StarLink issue.

The fallout: AgBiotech

Proposals for reform have surfaced in the wake of the StarLink recalls. In October, Senator Dick Durbin from Illinois introduced legislation that would make the FDA's voluntary review of GM foods mandatory (S 3184: The Genetically Engineered Foods Act of 2000). The FDA recently published its own proposed rule, which would require food developers to notify the FDA at least 120 days prior to the commercial distribution of GE food or animal feed and to provide information to demonstrate that the product is as safe as its conventional counterpart.

The Coordinated Framework for the Regulation of Biotechnology, a creation of the Reagan Administration, granted oversight of GM plants and animals to the EPA, FDA, and Department of Agriculture. Critics of this regulatory scheme have pointed to the StarLink case as justification for an overhaul. Several years ago, Senator Durbin introduced legislation (S1281; Safe Food Act of 1999) to consolidate the responsibilities for food safety, labeling, and inspection into a single independent federal agency. The Senator recently promised to reintroduce a bill to establish one agency with primary responsibility for food safety.

The issue of food labeling was also raised after the StarLink controversy. In December, a joint US-European Union Biotechnology Consultative Forum recommended tighter controls on GE foods, including mandatory labeling of products that contain biotech ingredients. In its recently proposed guidelines on voluntary labeling, however, the FDA reaffirmed its decision not to require special labeling of all bioengineered foods.

The StarLink case has further highlighted the question of who bears the responsibility for an accidental commingling of traditional crops with those containing a genetically engineered trait. The Senate Agriculture Committee of the Iowa legislature, for example, recently approved legislation that would protect farmers from being held liable for a bioengineered crop that contaminates another's field. The bill would cover GM crops that are either prohibited from sale or use for human consumption by the FDA or that are prohibited for import by nations that import at least ten percent of the same species of grain crop in its natural form.

The Iowa legislation acknowledges two situations requiring segregation of a genetically engineered trait. With regard to a StarLink-type two-track marketing system, one EPA official has indicated that it is not very likely that the Agency would ever grant a split registration again. The second scenario calling for segregation invokes the acceptance of GM crops in the world market. This is something that cannot be changed by federal agency policy or by legislation.


1. Barboza D. 2000. Gene-altered corn changes dynamics of grain industry. New York Times A1 (December 11).

2. Crenson M. 2000. How StarLink got loose. Pittsburgh Post-Gazette A-8 (December 4).

3. FIFRA Advisory Panel. 2000. Assessment of scientific information concerning StarLink corn.
Available: < >

4. Harl NE, Ginder RG, Hurburgh CR, and Moline S. 2000. The StarLink situation.
Available: <>

Phillip B. C. Jones, PhD., J.D.
Seattle, Washington


The StarLink corn issue may not result in significant changes in transgenic crop research, but it will indeed place more grain industry attention on systematic differentiation, i.e., the way transgenic crops are handled from farmers' fields to the marketplace. Just recently, the National Corn Growers Association (NCGA) urged growers who planted StarLink hybrids last year to make an extra effort to control possible volunteer StarLink corn in 2001. That may mean rotating to another crop or growing a herbicide-tolerant hybrid that allows farmers to control volunteer StarLink.

"The danger is [from] volunteer StarLink corn pollinating surrounding non-StarLink corn plants, further compounding the problems of keeping StarLink out of the supply of US corn," says Ohio producer Fred Yoder, chairman of the NCGA Biotech Working Group. "Rotation is the best choice," he points out. "In an ideal situation for 2001, you'd rotate ground planted to StarLink last year into soybeans, oats, or some other crop that will allow you to find and destroy volunteer corn. "But if you're locked into growing corn-on-corn, you need to plant herbicide-tolerant hybrids that let you eliminate StarLink volunteers," Yoder stresses. However, the NCGA is warning farmers not to use Roundup Ready hybrids to control StarLink volunteers since "Roundup Ready corn is not yet approved for export to the European Union and is restricted from some domestic wet-milling markets."

The recommendation on controlling StarLink volunteers is made in addition to a separate statement issued by the NCGA encouraging growers to plant seed that has been tested for Cry9C, the StarLink transgenic protein.

Some overseas buyers want proof

The StarLink corn issue has increased international and domestic purchaser concerns over transgenic foods and food safety—no food company wants to be involved in a product recall and suffer adverse press. Consequently, buyers are asking for proof of non-transgenic crop origin. In the last year, the National Sunflower Association has begun providing its members with a letter stating that US sunflower is transgene free. The US Department of Agriculture is providing similar documentation upon request.

Saudi Arabia, the largest single importer of US corn oil, has banned imports of GM grains and processed foods containing GM ingredients. "All imported foodstuffs must be accompanied by authenticated health certificates indicating that they are free of any genetically modified elements and are fit for human consumption," according to the Saudi edict. This includes prepared foods that have one or more of their ingredients from GM plant material.

Other grains establish GMO protocol

Wheat organization advisors are using the StarLink example as impetus to establish the means for handling grains such as wheat, if and when a transgenic variety becomes commercially available. For example, at their recently held Wheat Industry Conference, the National Association of Wheat Growers (NAWG), Wheat Export Trade Education Committee (WETEC), and US Wheat Associates (ASW) joint committee on biotechnology proposed the establishment of an advisory committee to work with Monsanto on the development of a closed-loop system to prevent commingling of GM wheat with conventional wheat. The advisory committee recommended involving other sectors of the wheat industry, including farmers, grain handlers, millers, bakers, and exporters in the segregation process. Establishment of a reasonable tolerance for accidental commingling of GM and non-GM grain was also adopted.

StarLink controversy portends future of the industry

The StarLink controversy may shed some light on where the future of agriculture may be headed, suggests Zach Fore, cropping systems specialist with the University of Minnesota Extension Service. He predicts that the number of grain products possessing specific traits will greatly expand in the coming years, some of which will be products of biotechnology and others will result from conventional breeding and selection. In almost all cases, grain products possessing specific traits will need to be segregated from other grains and will need to meet other criteria for handling and purity, he says.

In the simplest cases, farmers will need to plant, harvest, and store grains separately, then have them tested to meet certain purity standards. In the most complex cases, every step in the process from seed selection to final delivery will need to be documented and monitored. The product will be certified, tested, and have a paper trail that allows traceability back to its origin.

The so called "StarLink Incident" need not be looked upon as a negative issue, Fore says; rather, it will bring improvements. "It will result in agricultural products with special attributes to be managed, handled, distributed, and marketed better—to the benefit of all in the food chain, including farmers. It is critical that farmers view these developments as opportunities, not hassles. Farmers and food companies willing to respond by customizing what they produce and how they produce it for their customers will benefit," Fore says.

Post-StarLink: what some northern plains grain leaders are saying

"The StarLink problem should make the grain handlers' industry more aware of their roles in the movement of grain and the possible affects of commingling products. I don't think GMO research will be impacted very much. However, the results of the research may be treated more carefully and thoughtfully. The process/logistics of handling product will require more attention through businesses like FarmConnect."

—Art Brandli
Chairman of the Minnesota Wheat Research and Promotion Council, and Chairman of FarmConnect
< >

"North Dakota has been pretty much spared the StarLink problem. But it has served as a warning to wheat growers and handlers. I think the entire grain industry will want to know more about what's being dumped in its pits. Getting reliable info on that could be a challenge. The spring wheat and durum growing areas might be a little ahead on being able to segregate things because of the various quality specs already demanded by buyers. Up until now, corn has been corn, same with soybeans, generally speaking. We are seeing some differentiation in those crops too."

—Steve Strege
Executive Director
North Dakota Grain Dealers Association

"Here are some of my thoughts on what may happen:
• No company will release a crop that is only partially approved.
• Companies will think very hard before introducing a crop that is not approved in major export markets.
• Discussion on the need for tolerances will be continued.
• Biotech releases will come more slowly and the enthusiasm will be tempered.
• Biotech research will go on. Many companies cannot turn back and they should not. Perhaps they'll get more aggressive in looking for output traits that consumers want.
• We no longer are ignorant about what can happen if GMO grain isn't managed properly. Everyone now knows. Just say `StarLink,' and everyone knows what happened."

—David Torgerson
Executive Director
Minnesota Association of Wheat Growers

Kraft Foods Exec applauds strategic biotech approach

Kraft, which is among the Philip Morris family of companies, has four specific improvements that the company is recommending for enhancing the safe entry of biotechnology into the marketplace. Kraft is encouraging the appropriate regulatory authorities to consider the following:

• Discontinue partial approvals of advances in plant biotechnology and do not allow crops approved for animal use to enter the market unless they have also been approved for use in food;
• Require as a pre-condition to approval that a fully validated testing procedure be in place for identifying the relevant DNA in crops and in finished products;
• Require mandatory review of all plant biotechnology advances by the appropriate government agencies before those advances enter the market; and
• Strengthen the requirements for environmental stewardship of plant biotechnology to enhance the integrity of the food supply chain from farm to finished product.

Tracy Sayler
Fargo, ND

ISB News Report
120 Engel Hall
Virginia Tech
Blacksburg, VA 24061

ISB welcomes your comments and encourages article submissions. If you have a suitable article relevant to our coverage of the agricultural and environmental applications of genetic engineering, please email it to the Editor for consideration.

The material in this News Report is compiled by NBIAP's Information Systems for Biotechnology, a joint project of USDA/CSREES and the Virginia Polytechnic Institute and State University. It does not necessarily reflect the views of the U.S. Department of Agriculture or of Virginia Tech. The News Report may be freely photocopied or otherwise distributed without charge.

Ruth Irwin, Editor (

To have the News Report automatically emailed to you, send an email message to and type subscribe newsreport [your name] in the message section. Do not include a signature file or additional text. To unsubscribe, send email to and type unsubscribe newsreport [your name] in the message section, or email with your request.
Connect to for internet access to ISB News Reports, textfiles, and databases.

Information Systems for Biotechnology, 120 Engel Hall, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, tel: 540-231-3747, fax: 540-231-9070, email: