• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services


  • Print
  • Share
  • E-mail

Biotechnology Consultation Note to the File BNF No. 000109

Return to inventory: Completed Consultations on Foods from Genetically Engineered Plant Varieties

See also Biotechnology: Genetically Engineered Plants for Food and Feed and about Submissions on Bioengineered New Plant Varieties

Date: August 29, 2008

Subject: Biotechnology Notification File (BNF) BNF 000109; Glyphosate-Tolerant Cotton event GHB614

Keywords: Cotton, Gossypium hirsutum L., transformation event GHB614, GlyTolTM, Coker 312, glyphosate, herbicide tolerant, 2mepsps gene, double mutant 2mEPSPS, 5-enolpyruvylshikimate 3-phosphate synthase, Black Mexican sweet maize, Zea mays L.

1. Background

In a submission dated December 12, 2006, Bayer CropScience USA LP (Bayer) provided to the Food and Drug Administration (FDA) a summary of the safety and nutritional assessment of bioengineered glyphosate-tolerant cotton event GHB614, hereafter referred to as cotton event GHB614. Bayer provided additional information to the FDA on June 6, 2007, January 18, 2008, February 29, 2008, and July 31, 2008. Bayer concluded that food and feed derived from cotton event GHB614 are as safe and nutritious as food and feed derived from conventional cotton varieties currently being marketed.

2. Intended Effect

The intended effect of the modification in cotton event GHB614 is to confer tolerance to the herbicide glyphosate. To accomplish this objective, Bayer introduced a gene, 2mepsps, which encodes a double-mutant 5-enolpyruvylshikimate 3-phosphate synthase (2mEPSPS), into a conventional cotton cultivar. The 2mEPSPS protein confers tolerance to the herbicide glyphosate by inhibiting the enzyme 5-enolpyruvylshikimate 3- phosphate synthase (EPSPS). EPSPS is involved in the shikimic acid pathway for aromatic amino acid biosynthesis in plants and microorganisms.

3. Genetic Modifications and Characterizations

3.1 Parental Variety

The parental variety used in this transformation was Coker 312. Coker 312 is a cross between a Coker line and a D&PL® variety. Coker 312 is a variety of upland cotton (Gossypium hirsutum). Bayer states that they chose Coker 312 because it produces embryogenic structures suitable for regeneration of the cotton plant, which makes it ideal for the original transformant.

3.2 Transformation Plasmid and Method

To generate cotton event GHB614, Bayer used an Agrobacterium tumefaciens-mediated transformation system. Bayer transformed tissue from variety Coker 312 with the vector pTEM2. Plasmid pTEM2 was derived from pGSC1700 by inserting the gene of interest, 2mepsps, between the T-DNA border repeats of pTEM2. The insert contains one open reading frame, 2mepsps. Plasmid pTEM2 was constructed in Escherichia coli, and then transferred to a suitable A. tumefaciens strain, which was used in plant transformation. The genetic elements of the T-DNA region of plasmid vector pTEM2 are listed in Table 1.

Table 1. Genetic elements contained in the T-DNA region of the plasmid vector pTEM2
Genetic Element Description
Left Border Left border sequence from the T-DNA of Agrobacterium tumefaciens
Ph4a748At Promoter region of the histone H4 gene from Arabidopsis thaliana
Intron 1 h3At First intron sequence of gene II of the histone H3.III variant from Arabidopsis thaliana
TPotp C Transit peptide derived from corn and sunflower RuBisCO. Targets the mature protein to plastids.
2mepsps 2mepsps gene derived from Zea mays L.
3'histonAt Untranslated region of the histone H4 gene of Arabidopsis thaliana functioning as a polyadenylation signal.
Right Border Right border sequence from the T-DNA of Agrobacterium tumefaciens.

Plasmid pTEM2 contains several genes on its plasmid backbone necessary for maintenance and selection of the plasmid. Bayer CropScience stated that these sequences are outside the T-DNA borders. The plasmid backbone of pTEM2 contains origins of replication that allow replication of the plasmid in both A. tumefaciens and E. coli. The plasmid also contains the adenyltransferase gene aadA conferring resistance to streptomycin and spectinomycin.

3.3 Characterization, Stability and Inheritance of the Introduced DNA

Bayer conducted Southern blot analyses to characterize the genetic material introduced into cotton event GHB614. Bayer demonstrated that a single intact copy of the gene cassette was integrated into the plant genome and no other plasmid sequences were detected in the plant genome. The integration of the gene cassette is stable over multiple generations and in different genetic backgrounds, confirming the stability of the insert and showing it segregates as one dominant Mendelian locus. PCR analysis was used to demonstrated that the 5-prime and 3-prime flanking sequences of cotton event GHB614 are of cotton plant origin.

4. Introduced Protein – 2mepsps Enzyme

The 2mepsps gene encodes a 47 kilodalton protein consisting of 445 amino acids. The mutant 2mEPSPS protein differs from the wild type maize enzyme EPSPS by two amino acid substitutions, resulting in a protein with a high tolerance to glyphosate. The epsps gene was originally isolated from a cell suspension of Black Mexican Sweet maize (Zea mays L.).

4.1 Expression Levels

Bayer determined the levels of 2mEPSPS protein in whole ginned cottonseed, also known as fuzzy seed, and fractionated agricultural products by a validated enzyme-linked immunosorbent assay (ELISA). For whole cottonseed, Bayer determined 99.5% of the 2mEPSPS protein to be present in the kernel fraction and less than 0.5% in the lint coat (kernel + lint coat = cottonseed fraction). Concentration of 2mEPSPS protein ranged from 16.2 micrograms/gram fresh weight (μg/g fw) to 30.5 μg/g fw for cottonseed from cotton event GHB614 treated with glyphosate and from 15.8 μg/g fw to 25.5 μg/g fw for cottonseed from cotton event GHB614 not treated with glyphosate.

5 Safety Assessment of the Introduced Protein

5.1 Assessment of Potential Allergenicity

Bayer provides data and information on the allergenic potential of the 2mEPSPS protein. Bayer indicates that the potential allergenicity of the 2mEPSPS protein was assessed by searching for amino acid sequence homology between the 2mEPSPS protein and known allergens. These searches were conducted using databases comprised of all known protein sequences, including identified allergenic sequences, from publicly available databases (Uniprot_Swissprot, Uniprot_TrEMBL, PIR, NRL-3D, DAD and GenPept). Bayer reports that these searches did not result in any sequence homology or shared identity between the 2mEPSPS protein and known allergens.

The 2mEPSPS protein contains potential glycosylation sites (amino acid positions 118 and 394), which Bayer notes is similar to the wild type EPSPS. However, Bayer states that 2mEPSPS is an unlikely candidate for a glycoprotein based on its cellular targeting to plastids, where glycosylation would not be expected to occur.

Bayer reports that the 2mEPSPS protein is acid and heat labile. To further address the allergenicity potential of 2mEPSPS protein, Bayer cites proteolytic stability test studies. Bayer notes the rapid degradation (< 30 seconds) of 2mEPSPS during in vitro digestibility studies using simulated gastric fluid and simulated intestinal fluid. Bayer concludes that the 2mEPSPS protein is not likely to be allergenic.

5.2 Assessment of Potential Toxicity

Bayer describes data and information on the potential toxicity of the 2mEPSPS protein. Bayer compared the complete amino acid sequence of 2mEPSPS to known protein toxins in public databases (Uniprot_Swissprot, Uniprot_TrEMBL, PIR, NRL-3D, DAD and GenPept). These searches did not result in any sequence homologies between the 2mEPSPS protein and known toxic substances.

Bayer also assessed the potential toxicity of 2mEPSPS using an acute oral toxicity study in rodents. During the study, animals received an acute oral dose of 2,000 milligrams protein/kilogram body weight. No adverse effects were reported.

Bayer also cites the results of its digestibility studies in simulated gastric and intestinal fluids as reported above as further evidence of human and animal safety.

6. Food and Feed Use

Cotton is a source of lint and vegetable oil. Cottonseed oil ranks third behind soybean and corn oil for human consumption. Cottonseed contains the natural toxicants gossypol and cyclopropenoid fatty acids (CPFA). Cottonseed oil for human consumption is therefore highly purified to substantially reduce the content of gossypol and CPFA. Refined cottonseed oil is used as a frying oil, salad and cooking oil. Cottonseed lint, or "linters," are the short fibers that remain on the cottonseed after ginning. Linters are processed into pure cellulose and are used in a wide range of different products such as casings for sausages, ice cream and salad dressings.

Whole cottonseed is an important dairy, beef and sheep feed. Cottonseed hulls are a byproduct of the oil extraction process and are used as a source of fiber for cattle to aid in digestion. Fuzzy cottonseed hulls are preferred over delinted hulls, especially in starter rations for newly weaned calves. Cottonseed meal and hulls are a major source of protein for ruminant animals.

7. Compositional Analysis

Bayer tested the composition of cottonseed and processed fractions from cotton event GHB614 and the non-transgenic variety Coker 312.

Cottonseed was collected from nine field trials conducted under current production practices in typical cotton-producing areas of the Southern United States. Bayer measured each component from six transgenic plots and three non-transgenic plots at each test site. Twenty-seven samples from each of three groups: Coker 312 cotton, cotton event GHB614 that was not sprayed with glyphosate herbicide, and cotton event GHB614 that was sprayed three times with glyphosate herbicide, were tested in total. A sample of ginned cottonseed, also known as fuzzy seed, was taken from each of the 81 field plots and shipped in frozen state to Bayer laboratories at Research Triangle, NC. The seeds were sub-sampled and shipped in a frozen state to its contracted analytical facility in Des Moines, IA.

An analysis of variance (ANOVA) was performed for all analytes for whole cottonseed except fatty acids. In addition, T-test comparisons on the analytes subjected to the ANOVA were also conducted for Coker 312 controls versus corn event GHB614 sprayed with the herbicide glyphosate and Coker 312 controls versus corn event GHB614 not sprayed with the herbicide glyphosate. A significance level of 0.01 (α = 1%) was chosen for both analyses. Bayer stated that it chose the 0.01 significance level, rather than the typical 0.05 significance level, to reduce the likelihood of "false positive" detections of significance.

Bayer analyzed the composition of processed fractions of cottonseed from a single field trial to the non-transgenic control. The seeds used in this study were derived from one field trial site and the parent plants were treated with glyphosate. Samples were processed under conditions typical of production practices.

7.1 Cottonseed

Bayer tested the levels of 47 different components from the 81 samples of cottonseed. Bayer analyzed cottonseed for proximates, amino acids, fatty acids, minerals, and other analytes of interest. A list of specific analytes is shown in Table 2.

Bayer reported no statistically significant differences comparing cotton event GHB614 with non-transgenic variety Coker 312 in any analyte measured, with the exception of the amino acids cystine and methionine (p < .01) in their ANOVA. However, Bayer also reported no significant differences by T-tests for those two amino acids or for any of the other analytes measured when comparing the analyte values for non-transgenic samples and either unsprayed or sprayed transgenic samples. Further, Bayer reported that the mean values for all analytes measured were within reported literature ranges. Bayer states that the comparison of their analytical results to reference ranges from the literature establishes that cottonseed from cotton event GHB614 provide the same nutritional value as cottonseed currently being consumed.

Table 2: Components measured in event GHB614 cottonseed
Proximates Minerals Amino Acids Fatty Acids Anti-nutrients and potential toxicants Vitamins
acid detergent fiber (ADF)
neutral detergent fiber (NDF)
aspartic acid
glutamic acid
myristic (14:0)
palmitic (16:0)
stearic (18:0)
arachidic (20:0)
behenic (22:0)
palmitoleic (16:1)
oleic (18:1)
linoleic (18:2)
linolenic (18:3)
phytic acid
total gossypol
free gossypol
sterculic acid
malvalic acid
dihydrosterculic acid
total tocopherol
*measured in refined cottonseed oil only

7.2 Processed Cottonseed Fractions

Samples of cottonseed from Coker 312 and cotton event GHB614 from one of the field trials were processed into linters, delinted seed, meal, toasted meal, hulls, crude oil and deodorized oil at the GLP Program of the Food Protein R & D Center at Texas A & M University. A matrix of the samples and analyses are provided in Table 3.

Table 3. Analyses performed on processed agricultural commodities of GHB614 and Coker 312 control cotton. An "X" indicates the analysis was performed.
Analysis Linters Delinted seed Meal Toasted meal Hulls Crude oil Deodorized oil
Moisture X X X X X    
Protein X X X X X    
Fat X X X X X    
Ash X X X X X    
Carbohydrates X X X X X    
Crude Fiber X X X X X    
ADF X X X X X    
NDF X X X X X    
Amino Acids   X X X      
Fatty Acids   X       X X
Calcium   X X X      
Magnesium   X X X      
Phosphorus   X X X      
Potassium   X X X      
Zinc   X X X      
Alpha-tocopherol   X X X     X
Beta-tocopherol             X
Delta-tocopherol             X
Gamma-tocopherol             X
Total tocopherol   X X X     X
Gossypol - free   X X X      
Gossypol - total   X X X     X
Cycloprenoid fatty acids   X X X     X
Phytic acid   X X X      

Bayer reported means for the analytes measured in processed cottonseed fractions for comparison to processed fractions derived from the non-transgenic control. Bayer also provided literature values for the analytes measured in processed fractions for comparison. Bayer concluded that the values for all the analytes measured in the processed products derived from cotton event GHB614 were comparable with those measured in the processed products derived from Coker 312 and the values for the respective products reported in the literature.

8. Conclusions

Bayer has concluded that cottonseed and cottonseed-derived products from cotton event GHB614 are not materially different in composition, safety, or any other relevant parameter from cottonseed and cottonseed-derived products now grown, marketed and consumed. At this time, based on Bayer's data and information, the agency considers Bayer's consultation on cotton event GHB614 to be complete.


Jannavi Srinivasan

Richard Bonnette