Return to inventory: Completed Consultations on Foods from Genetically Engineered Plant Varieties
Date: May 5, 2004
Subject: Bacillus thuringiensis Cry1F insect resistant cotton event 281-24-236
Keywords: Cotton, Gossypium hirsutum L., insect-protected cotton, Cry1F protein, phosphinothricin acetyltransferase (PAT), Agrobacterium tumefaciens, Bacillus thuringiensis subspecies aizawai, Lepidopteran insects, cotton bollworm, tobacco budworm.
In a submission dated March 17, 2003, Mycogen Seeds/Dow AgroSciences LLC (DAS) provided to FDA a summary of the safety and nutritional assessment they conducted on the new bioengineered insect-protected cotton line, Cry1F cotton event 281-24-236. The company provided additional information on May 30, September 5, and September 25, 2003. DAS concluded that the new insect resistant cotton is as safe and nutritious as conventional cotton varieties currently being marketed.
The intended effect of this genetic modification of cotton is to confer resistance to Lepidopteran insect pests such as the cotton bollworm, tobacco budworm, fall armyworm and soybean loopers. To accomplish this objective, DAS introduced the cry1F gene isolated from Bacillus thuringiensis (B.t.) subspecies aizawai strain PS811 and the pat gene isolated from Streptomyces viridochromogenes into the proprietary cotton Acala germplasm line known as GC510. The Cry1F cotton event 281-24-236 is toxic to Lepidopteran insects. The pat gene confers tolerance to chemically synthesized phosphinothricin products such as glufosinate-ammonium and is used as a selectable marker. The Cry1F cotton event 281-24-236 is intended to be marketed and sold as a stacked event product along with Cry1Ac cotton event 3006-210-23, submitted in a separate filing (BNF 92) with FDA.
The Environmental Protection Agency (EPA) regulates plant-incorporated protectants under the Federal Food, Drug, and Cosmetic Act and the Federal Insecticide, Fungicide, and Rodenticide Act. Under EPA regulations, the Cry1F protein in cotton event 281-24-236 is considered a pesticidal substance and the phosphinothricin acetyltransferase (PAT) protein is considered an inert ingredient. Therefore, the safety assessment of these proteins falls under the regulatory purview of EPA.
Genetic Modifications and Characterization
DAS used the Acala cotton variety GC510 in the transformation of cotton event 281-24-236. DAS used a disarmed Agrobacterium tumefaciens methodology for their transformation (Ooms et al., 1982; Zambryski, 1988). The T-DNA region of the plasmid vector pAGM281 contains the elements listed in Table 1.
|(4OCS)Δmas2'||Mannopine synthase promoter from A. tumefaciens strain LBA 4404 pTi15955 for cry1F expression.|
|cry1F||Synthetic, plant optimized, full length version of cry1F from B. thuringiensis subspecies aizawai strain PS811. Confers resistance to Lepidopteran insects.|
|ORF25 polyA||Bi-directional terminator from A. tumefaciens strain LBA 4404 pTi15955 (Barker et al., 1983) for cry1F and pat gene.|
|pat||Synthetic plant-optimized glufosinate-ammonium resistance gene, based on a phosphinothricin acetyltransferase gene sequence from Streptomyces viridochromogenes. Used as a selectable marker.|
|Ubi Zm1||Zea mays ubiquitin promoter system for pat expression.|
The pAGM281 transformation vector is a binary T-DNA vector carrying the transgenes for insertion into the plant genome between consensus T-DNA border sequences (from A. tumefaciens pTi15955 (Barker et al., 1983)), and a bacterial antibiotic resistance marker in the plasmid backbone, to facilitate cloning and maintenance of the plasmid in bacterial hosts. The plasmid backbone was derived from plasmid RK2 (Schmidhauser and Helsinki, 1985), from which the tetracycline resistance gene was deleted and replaced with a DNA fragment containing the erythromycin resistance gene for bacterial expression. DAS states that a lack of binding of the erythromycin probe to the expected DNA fragments indicates that the gene encoding for erythromycin resistance on plasmid pAGM281 has not been integrated into the cotton genome for event 281-24-236.
DAS characterized the integration and number of insertions in event 281-24-236, using restriction enzyme digestion and Southern blot analysis. The results suggest integration of a single, intact copy of the (4OCS)Δmas2' promoter, cry1F, ORF25 polyA signal, and two integrations of the Ubi Zm1 promoter and pat. The additional pat fragment contains 231 contiguous base pairs (bp) of pat plus 24 bp from an adjacent region, constituting a 255 bp open reading frame (ORF). The putative amino acid sequence of the ORF consists of 77 amino acids from the amino terminus of the PAT protein and an additional eight amino acid carboxyl terminal tail. DAS reports that the transcription of the pat fragment is approximately 16 fold less than the full length pat gene and that no protein expression was detected from the second pat fragment. DAS concludes that the presence of the additional fragment is not considered of consequence to safety of cotton products derived from cotton event 281-24-236.
DAS assessed the genetic stability of the cry1F and pat genes within and across generations using Southern Blot and segregation analysis. DAS states that the integration is stable and that the segregation of transgenes in event 281-24-236 shows a Mendelian inheritance pattern.
Cotton, Gossypium hirsutum L., is grown worldwide primarily as a source of fiber for textile manufacturing. Cottonseed, which is a by-product of the fiber production, is used in human food and animal feed production. Cottonseed contains the natural toxicants, gossypol and cyclopropenoid fatty acids (CPFAs). Whole cottonseed is used as feed, supplying energy and fiber for dairy cows.
Alternatively, cottonseed can be processed into linters, hulls, oil and meal. As a first step in cotton processing the linters are removed. Linters are short fibers that are further processed to obtain pure cellulose for use in food, for example, in casings for bologna, sausages, and frankfurters, and in products such as ice cream and salad dressings. Next the hull, a protective seed coating, is removed. Hulls are used as a fiber component of livestock feeds due to their high cellulose and lignin content. Seeds without hulls are referred to as kernels or meat. The kernels are used for oil extraction. Cottonseed oil is routinely used in human food and has a long history of safe use. Cottonseed oil intended for human consumption is highly purified. The purification process substantially reduces the content of CPFAs. The refined cottonseed oil is used as frying oil, salad and cooking oil, and in various foods including mayonnaise, salad dressing, shortening, and margarine.
Cottonseed meal is the product obtained by finely grinding the flakes or cake which remain after removal of most of the oil from cottonseed by either a solvent or mechanical extraction process. Cottonseed meal is principally used as protein containing feed for livestock.
DAS analyzed the composition of cotton event 281-24-236 to assess whether any unexpected modifications occurred. DAS used null plants for their comparison of compositional analysis. DAS describes these null plants as the F1 generation of plants lacking all transgenes after crossing the Cry1Ac and Cry1F BC3F1 lines to generate the stacked product. The null plants and the Cry1F event 281-24-236 BC3F1 line were self-pollinated to the F4 generation to provide the seed source for the nutritional composition study. DAS states that the background of the null control is very closely related to the Cry1F event because it was derived from a late (BC3F1) generation of the transgenic event. DAS further states that the null control is also very closely related to the commercial variety, PSC355, because it has been backcrossed to the PSC355 variety four times, yielding a genetic background comprised of approximately 93.75% of the cotton variety PSC355. DAS concludes that the null plants represent the genetic background of both the Cry1F event and a commercial cotton variety.
The following groups of analytes were measured:
- amino acids
- fatty acids
A list of specific analytes contained in each group is shown in Table 2.
Analyses were performed on cottonseed, terminal leaf, square, and processed cottonseed including kernel, hull, toasted meal, and refined oil. DAS tested cotton grown at six different locations. DAS analyzed cottonseed samples individually and averaged the results across all six production sites. DAS compared mean values of the transgenic cotton to the non transgenic (null) control cotton line using a mixed model analysis of variance with p values being determined by a paired 't' test and Dunnett's procedure to maintain the experiment-wide error at 0.05. This procedure was utilized because in addition to the null control and event 281-24-236, the experiment also included event 3006-210-23. Mean values were also compared to literature values. DAS analyzed processed cottonseed products by combining samples from all six sites into a single sample per treatment. Terminal leaf and square samples were reported individually from two locations.
|Proximates||Minerals||Amino Acids||Fatty Acids||Anti-Nutrients||Anti-Oxidants|
acid detergent fiber (ADF)
neutral detergent fiber (NDF)
cyclopropenoid fatty acids (malvalic, sterculic, dihydrosterculic acids)
aflatoxins B1, B2, G1, G2
1.a Raw commodity (cotton seed)
DAS determined the levels of the following components of cotton seed:
- amino acids
- fatty acids
- anti-nutrients (CPFAs, gossypol-total)
DAS reports that the mean levels of proximates and amino acids were comparable between the control and the transgenic line and were within or near literature ranges. DAS reports that the only significant (p < .05) difference found was for the mineral manganese and the fatty acids stearic and oleic acid, where event 281-24-236 had higher levels than the control. However, the level for manganese was within the stated literature range, while the levels for stearic and oleic acids were less than the lowest literature value reported. These differences are not considered biologically meaningful. DAS reported that fatty acid values for both control and transgenic cotton were slightly lower than existing literature values, and notes that it is difficult to accurately compare these values to single literature values.
DAS determined the level of CPFAs. Cyclopropenoid fatty acids, which occur naturally in cottonseed, are considered to be anti-nutritional compounds and are undesirable in human food and animal feed. They are known to inhibit the desaturation of stearic acid to oleic acid in the human or animal body and result in the alteration of membrane permeability or an increase in the melting point of fats. In cottonseed oil intended for human consumption, the CPFAs are reduced to negligible levels.(1) DAS reports no statistically significant differences between control line and transgenic line for CPFAs, and all values were within literature values.
DAS measured gossypol in cottonseed. Gossypol is a terpenoid aldehyde that naturally occurs in cottonseed and is toxic to humans and animals. DAS reports no statistically significant differences between control line and transgenic line for gossypol, and all values were within literature values.
DAS measured four major aflatoxins, B1, B2, G1, and G2, in cottonseed. Aflatoxins are mycotoxins produced by certain species of the fungus Aspergillus that may infect cotton, mainly A. flavus and A. parasiticus. Aflatoxins are highly-substituted coumarins containing a fused dihydrofuran and are endemic to cotton grown in the southwestern U.S. Aflatoxins are potent animal toxins and carcinogens and have been epidemiologically implicated as environmental carcinogens in humans. DAS reports that these aflatoxins were not detected in either the control or transgenic cotton line.
1.b Raw Commodity (terminal leaf and square)
DAS determined the levels of the following components of leaf and square tissue:
- anti-nutrients (gossypol - total, polyphenols)
DAS measured total polyphenols and gossypol in terminal leaf and square from two locations and found comparable results between the transgenic cotton and the controls.
2. Processed commodity (cottonseed oil, kernels, toasted meal, hulls)
DAS analyzed kernels, hulls, toasted meal, and refined oil from single samples from bulk processing across all six production sites.
2.a Cottonseed oil
DAS determined the levels of the following components of refined cottonseed oil:
- proximates (fat, protein, moisture)
- fatty acids
- anti-nutrients (gossypol, free and total, CPFAs)
DAS states that analysis results for proximates and fatty acids were comparable between the transgenic line and the control line and that all results were within published literature ranges. DAS reports similar results for control and transgenic lines for both gossypol and CPFAs. No free gossypol was detected and low total gossypol levels were found in the samples analyzed. The levels of CPFAs were also similar for control and transgenic samples and were close to literature values.
Cottonseed oil was also analyzed for tocopherol isomers. Tocopherols, which may function as naturally occurring antioxidants, showed similar results for control and transgenic lines and fell within Codex Alimentarius standards for occurrence of alpha, beta, gamma, and delta tocopherols in crude cotton seed oil. Oil processing steps subsequently remove tocopherols from the oil.
DAS determined the levels of the following components of kernels:
- proximates (moisture)
- anti-nutrients (gossypol, free and total)
Proximate analysis was comparable for the Cry1F event 281-24-236 line and control line and fell within literature ranges. Anti-nutrient analysis of the kernel was similar for the control line and the transgenic line. Literature values for anti-nutrients in kernels were not available.
2.c Toasted meal
DAS determined the levels of the following components of toasted meal:
- amino acids
- anti-nutrients (gossypol, free and total)
There were only inconsequential differences noted in the data analysis results for all proximates, minerals, and gossypol between the control and the transgenic line and these were within literature values. The levels of the amino acids aspartic acid, proline, alanine, valine, phenylalanine, lysine, arginine and leucine were slightly higher in the transgenic line than the control line, and also slightly higher than the literature. However, these increases are not judged to be biologically meaningful.
DAS determined the levels of the following components of hulls:
- proximates (ash, fat, moisture, protein, and carbohydrates)
Analysis results were comparable for the Cry1F event 281-24-236 and the control line, and values were near or within literature ranges.
DAS has concluded that the Cry1F cotton event 281-24-236 is not materially different in composition, safety, wholesomeness, or any relevant parameter from cotton now grown, marketed, and consumed. At this time, based on DAS's data and information, the agency considers DAS's consultation on Cry1F cotton event 281-24-236 to be complete.
Karin Ricker, Ph.D.
Barker, R., K. Idler, D. Thompson and J. Kemp. 1983. Nucleotide sequence of the T-DNA region from the Agrobacterium tumefaciens octopine Ti plasmid pTil5955. Plant Mol. Biol. 2:335-350.
Ooms, G., P. Hooykaas, R. V. Veen, P. V. Beelen, T. Regensburg-Tuink and R. Schilperoort. 1982. Octopine Ti-plasmid deletion mutants of Agrobacterium tumefaciens with emphasis on the right side of the T-region. Plasmid. 7: 15-29.
Schmidhauser, T.J. and Helinski, D.R. 1985. Regions of broad-host range plasmid RK2 involved in replication and stable maintenance in nine species of gram-negative bacteria. J. Bacteriol. 164(1), 446-455.
Zambryski, P. 1988. Basic processes underlying Agrobacterium-mediated DNA transfer to plant cells. Ann. Rev. Genet. 22: 1-30.
(1)"Cottonseed Oil." In "Bailey's Industrial Oil & Fat products." 1996. John Wiley & Sons, Inc. Vol. 2, Chapter 4, pp. 159-240.