Biotechnology Consultation Note to the File BNF No. 000106
Return to inventory: Completed Consultations on Foods from Genetically Engineered Plant Varieties
Biotechnology Consultation - Note to the File
Biotechnology Notification File BNF No. 000106
July 31, 2012
HCEM485 maize, herbicide tolerant corn
Maize; corn; Zea mays; herbicide tolerance; HCEM485 corn; 2mepsps gene; modified 5-enolpyruvylshikimate-3-phosphate synthase (2mEPSPS) protein; glyphosate; Stine Seed Farm, Inc.
This document summarizes our evaluation of Biotechnology Notification File (BNF) No. 000106. In a submission dated December 9, 2010, Robert Potter Consulting submitted, on behalf of Stine Seed Farm, Inc. (Stine), a safety and nutritional assessment of bioengineered herbicide tolerant corn, transformation event HCEM485 maize (hereafter referred to as HCEM485 corn). Robert Potter Consulting provided additional information, on behalf of Stine, on October 6 and 17, 2011, and January 10, 2012. FDA evaluated the information in Stine's submissions to ensure that regulatory and safety issues regarding human food and animal feed derived from the new plant variety have been resolved prior to commercial distribution.
In our evaluation of BNF No. 000106, we considered all information provided by Stine as well as publicly available information and information in the agency's files. Here we discuss the outcome of the consultation, but do not intend to restate the information provided in the final consultation in its entirety.
The intended technical effect of the modification in HCEM485 corn is to confer tolerance to the herbicide glyphosate. To accomplish this objective, Stine introduced the 2mepsps gene that encodes a modified 5-enolpyruvylshikimate-3-phosphate synthase (2mEPSPS) protein; 2mEPSPS protein, unlike native corn EPSPS protein, is not inhibited by glyphosate.
The purpose of this evaluation is to assess whether Stine has introduced a substance that requires premarket approval as a food additive or that raises other regulatory or safety issues under the Federal Food, Drug, and Cosmetic Act (FD&C Act)
The Environmental Protection Agency (EPA) regulates the use of herbicides under the FD&C Act and the Federal Insecticide, Fungicide, and Rodenticide Act. Under EPA regulations, the herbicide residues in HCEM485 corn are considered pesticide residues.
Genetic Modification and Characterization
Stine transformed the recipient elite inbred line Stine 963 to generate HCEM485 corn.
Introduced DNA and Transformation Method
The transformation vector pHCEM was generated by cloning a 6.0 kilobase (kb) genomic fragment from a bacterial artificial chromosome library of inbred corn line B73 into the ClaI and EcoRV sites of pBluescript. Sequence analysis showed that the 6.0 kb genomic fragment contained the corn EPSPS coding sequence; its flanking sequences, including the 5' regulatory sequence; and an N-terminal chloroplast transit peptide sequence. The cloned 6.0 kb genomic fragment was then subjected to site-directed mutagenesis to introduce two nucleotide substitutions in the EPSPS coding sequence. The nucleotide substitutions were designed to change two amino acids in the corn EPSPS protein sequence: a threonine to isoleucine substitution at position 102 and a proline to serine substitution at position 106. Stine states that it sequenced the mutated 6.0 kb genomic fragment (hereafter referred to as the 2mepsps fragment) in pHCEM and confirmed that no other alterations in the sequence were introduced.
To generate HCEM485 corn, Stine excised the 2mepsps fragment from the cloning vector, isolated it by gel purification, and used aerosol beam injection to introduce it into immature corn embryos. Treated embryos were cultured on non-selective medium for five days before transfer onto media containing successively higher concentrations of the herbicide glyphosate. Transgenic plants were regenerated from glyphosate-tolerant transformants.
Characteristics, Inheritance, and Stability of the Introduced DNA
Stine characterized the introduced DNA in HCEM485 corn through Southern blot analysis of genomic DNA. Based on the results, the putative organization of the introduced DNA in HCEM485 includes four complete copies of the 2mepsps fragment at a single site of insertion in the corn genome. The results of the Southern blot analysis are consistent with the results of Western blot analysis of leaf and seed tissue extracts using polyclonal antisera as no truncated expression products were detected. The results of the Southern blot analysis also confirmed that no sequences outside the 2mepsps fragment (i.e., vector sequences) were present in HCEM485 corn.
Stine studied the inheritance of the introduced DNA through segregation analysis of the glyphosate-tolerance phenotype in two generations of HCEM485 corn. The observed ratios of glyphosate tolerant (positive) to glyphosate sensitive (negative) plants from F1 hybrid and F2 segregating generations were analyzed using Chi square analysis.(1) The observed phenotypic segregation ratios in both generations were consistent with the expected ratios. Based on the results of the phenotypic segregation analysis, Stine concludes that the 2mepsps genotype in HCEM485 corn segregates as a single locus in a Mendelian fashion.
In its submission, Stine does not discuss the stability of the introduced DNA. According to Stine, since the introduced DNA is from corn, the consequences of genetic rearrangements are not expected to be materially different from those arising by natural genetic recombination events or mutations during sexual reproduction.
Identity and Function of the Introduced Protein
HCEM485 corn was genetically engineered to express an altered 5-enolpyruvylshikimate-3-phosphate synthase (2mEPSPS) protein. Whereas native corn EPSPS protein is inhibited by glyphosate, interrupting the biosynthesis of aromatic amino acids essential for plant growth, 2mEPSPS protein is not inhibited by glyphosate. Consequently, plants expressing 2mEPSPS protein remain viable when exposed to glyphosate.
Stine compared 2mEPSPS protein in HCEM485 corn with modified EPSPS protein in GA21 corn(2) and with native EPSPS protein in corn.(3) The 2mepsps gene in HCEM485 corn was sequenced and its expression product was deduced using sequence analysis software. The deduced expression product is identical to modified EPSPS protein in GA21 corn and, with the exception of two intentionally substituted amino acids, to native corn EPSPS protein. Further, Western blot analyses of corn leaf and grain samples was conducted with a monoclonal antibody specific to 2mEPSPS to confirm the molecular weight and immunoreactivity of 2mEPSPS protein from HCEM485 corn and modified EPSPS protein from GA21 corn. Based on the results of these analyses, Stine concludes that 2mEPSPS protein expressed in HCEM485 corn is equivalent to modified EPSPS protein expressed in GA21 corn with respect to protein sequence, molecular weight, and immunochemical cross-reactivity.
Stine performed Western blot analysis of seed and leaf extracts using polyclonal antisera to compare EPSPS protein expression in HCEM485 corn, GA21 corn, and non-transgenic Stine 963 corn. The results of the analysis indicate that 2mEPSPS protein expression in HCEM485 corn is lower than modified EPSPS protein expression in GA21 corn but three- to five-fold greater than native EPSPS protein expression in Stine 963 corn; these results are consistent with the identity and activity of the respective promoters and with the total number of EPSPS gene copies in each corn variety. Stine concludes that further characterization of 2mEPSPS protein expression in HCEM485 corn is not warranted because transgenic EPSPS proteins in HCEM485 corn and GA21 corn have identical amino acid sequences and similar expression levels and evidence supports that modified EPSPS protein in GA 21 corn is not toxic (discussed below).
Safety Assessment of Potential for Toxicity and Allergenicity of the Introduced Protein
Stine based its safety assessment of 2mEPSPS protein largely on the similarity of 2mEPSPS protein to (1) EPSPS proteins that occur naturally in foods and feeds with histories of safe use; and (2) modified EPSPS protein expressed in GA21 corn. Native corn EPSPS protein, like the native EPSPS proteins from other plant and microbial food sources, has long been a part of human and animal diets and is not associated with any known health concerns. The deduced amino acid sequence of 2mEPSPS protein differs from native corn EPSPS protein(4) by only two amino acids; Stine notes that these two amino acids involve substitutions with standard amino acids common to all proteins of biological origin and that these substitutions are not known to alter the enzymatic function of the EPSPS protein.(5) As it has determined the immunochemical equivalence of 2mEPSPS protein to modified EPSPS protein from GA21 corn (discussed above), Stine refers to publicly available assessments of potential toxicity and allergenicity of modified EPSPS protein from GA21 corn in support of the safety assessment of 2mEPSPS protein from HCEM485. In these publications, the toxicity of modified EPSPS protein was assessed through bioinformatic analysis comparing its amino acid sequence to the sequences of known toxins and through an acute oral toxicity study in mice.(6) Likewise, the allergenicity of modified EPSPS protein was assessed in these publications through bioinformatic analysis comparing its amino acid sequence to the sequences of known allergens and through analysis of protein stability in simulated gastric and intestinal fluids. Stine concludes that 2mEPSPS protein from HCEM485 corn is expected to be as safe as modified EPSPS protein from GA21 corn.
Stine conducted a literature search for information relevant to the safety assessments of native EPSPS and of modified EPSPS proteins and for information published subsequent to the completion of FDA's review of GA21 corn in 1997 (BNF 000051). A number of oral toxicity studies were identified and described by Stine. No differences were reported in the growth or health of animals fed grain or forage from GA21 corn. Similarly, no signs of toxicity were observed in mice in an acute oral toxicity study of E. coli-expressed, double-mutant EPSPS protein (see FDA consultation BNF 000109; GHB614 cotton). Stine further notes that no reports of adverse effects from consumption of GA21 corn by humans or livestock were identified during its literature search.
Potential Novel Polypeptides
In its characterization of HCEM485 corn, Stine confirmed that corn is the sole source of the introduced DNA. On this basis, Stine asserts that the consequences of genetic rearrangements resulting in the production of chimeric open reading frames would not be materially different from those arising from natural genetic recombination events or mutations during sexual reproduction.
Food & Feed Use
Stine states that HCEM485 corn will be grown for the same uses as currently commercialized corn.
Corn (Zea mays L.; maize) originated in Mexico and was grown as a food crop as early as 2700 B.C. Today, corn is grown worldwide for food, feed, and industrial uses. Corn grain is used in food primarily in the form of processed products, such as high fructose corn syrup, cereals, oil, meal, flour, starch, and grits. Corn is a significant source of nutritionally important amino acids methionine and cystine, carotenoids, and vitamin E. Corn oil is rich in polyunsaturated fatty acids and is used mainly as a salad and cooking oil and in margarine production.
Corn is also used in animal feed. Corn grain is fed to cattle, poultry, and swine either as intact or processed grain or as dry or wet milling byproducts. Corn silage is primarily fed to ruminants. Corn is a good source of polyunsaturated fatty acids and essential amino acids, except for lysine and tryptophan.
Scope of Analyses
Stine analyzed the composition of forage and grain from an HCEM485 corn hybrid (the HCEM485 hybrid) and compared it to three non-transgenic control hybrids (controls). The controls were produced by crossing select inbred lines, each of which was used as a parental line in the breeding of the HCEM485 hybrid.
Study Design - Compositional Analyses
In a study conducted in 2007, Stine analyzed forage and grain from the HCEM485 hybrid and controls grown using a randomized complete block design with three replicates at each of several locations in North America. The HCEM485 hybrid was treated with glyphosate at the normal application rate. Forage samples were collected at two test sites and grain samples at four test sites; samples from the three individual controls were combined into one composite control (composite control) sample for each replicated plot. The composition data included proximates (moisture, crude protein, crude fat, ash, starch, and carbohydrates (by difference)), fiber (acid detergent fiber (ADF), neutral detergent fiber (NDF), and total dietary fiber (TDF)), fatty acids, amino acids, vitamins, minerals, secondary metabolites and anti-nutrients, and phytosterols.
Stine conducted a mixed model analysis of variance, using genotype and location as independent factors, to analyze the compositional data across all locations. The composition data analyses included mean values, 95 percent confidence intervals, and percent difference between the two treatment means. Statistical significance was set at the 5 percent level (p < 0.05). When a statistically significant difference in a component was observed between HCEM485 corn and composite control, the biological relevance of the difference from food and feed safety and nutritional perspectives was determined. Stine took into consideration (1) the reproducibility of the differences across individual sites, (2) the magnitude of the differences, and (3) the comparability/consistency of the values for the HCEM485 hybrid and the composite control with values reported in published literature.(7)
Results of Analyses - Compositional analysis of corn forage
Proximates and Fiber
Stine reports the results of compositional analyses for proximates and fiber (ADF and NDF only). Statistically significant differences were observed between the HCEM485 hybrid and the composite control for crude protein and crude fat. However, the mean concentrations for all the analyzed proximates and fiber were consistent with the ranges reported in the published literature.
Stine also reports the levels of two minerals, calcium and phosphorus, in forage. Although a statistically significant difference was observed between the HCEM485 hybrid and the composite control for calcium, the mean concentrations for both calcium and phosphorus were consistent with the ranges reported in the published literature.
Results of Analyses - Compositional analysis of corn grain
Proximates and Fiber
Stine reports the compositional analyses for proximates and fiber (ADF, NDF, and TDF). A statistically significant difference was observed between the HCEM485 hybrid and the composite control for TDF. However, the mean concentrations for all the analyzed proximates and fiber were consistent with the ranges reported in the published literature.
Stine reports the compositional analyses for eight fatty acids.(8) Statistically significant differences between the HCEM485 hybrid and the composite control were observed for palmitic, stearic, oleic, linolenic, and eicosenoic acids. However, the mean levels for all the reported fatty acids were consistent with the ranges reported in the literature.
Stine reports the compositional analyses for eighteen amino acids. A statistically significant difference was observed between the HCEM485 hybrid and the composite control for methionine. However, the mean concentrations for all the analyzed amino acids were consistent with the ranges reported in the literature.
Stine reports the compositional analyses for six vitamins, four isoforms of tocopherol, and total tocopherol.(9) Statistically significant differences were observed between the HCEM485 hybrid and the composite control for Vitamin B1, Vitamin B6, Vitamin B9, the tocopherol isoforms, and total tocopherol. However, the differences between the treatment means for these components were small in magnitude and were consistent with the range(s) reported in literature, where available.
Stine reports the compositional analysis for ten minerals.(10) Statistically significant differences were observed between the HCEM485 hybrid and the composite control for calcium, copper, and potassium. However, the mean concentrations for all the analyzed minerals were consistent with the range(s) reported in the literature.
Secondary Metabolites and Anti-nutrients
Stine reports the compositional analysis for ferulic acid, ρ-coumaric acid, inositol, phytic acid, raffinose, and trypsin inhibitor.(11) Statistically significant differences between the HCEM485 hybrid and the composite control were observed for inositol and trypsin inhibitor. However, the mean concentrations for all the analyzed secondary metabolites and anti-nutrients were consistent with the range(s) reported in the literature.
Stine reports that the compositional analysis for cholesterol, campesterol, stigmasterol, beta-sitosterol, stigmastanol and total sterols revealed no statistically significant differences between the HCEM485 hybrid and the composite control.
In summary, the magnitude of differences between the HCEM485 hybrid and the composite control were small and the mean values for the components were consistent with the respective published literature range(s). Based on the results of the compositional analysis, Stine concludes that grain and forage from HCEM485 corn is substantially equivalent in composition to the control hybrids and to other commercialized corn hybrids.
FDA evaluated Stine's submission to determine whether HCEM485 corn raises any safety or regulatory issues with respect to the intended modification or with respect to the food and feed itself. Based on the information provided by Stine and other information available to the agency, FDA did not identify any regulatory or safety issues under the FD&C Act that would require further evaluation at this time.
Stine concludes that its herbicide tolerant corn variety, HCEM485 corn, and the food and feed derived from it are as safe as conventional corn varieties and, with the exception of the herbicide tolerance trait, are not materially different in composition or other relevant parameters from other corn varieties now grown, marketed, and consumed in the United States. At this time, based on Stine’s data and information, the agency considers Stine’s consultation on HCEM485 corn to be complete.
Carrie McMahon, Ph.D.
(1) Significance for the Chi square analysis was set at the 95 percent confidence level (p < 0.05).
(2) GA21 corn expressing the modified EPSPS protein was the subject of BNF 000051. The modified EPSPS protein (1) is encoded by an epsps gene that was cloned from corn and (2) contains the same two amino acid substitutions, introduced into the cloned corn epsps gene using in vitro techniques, as in the description above for Stine's 2mEPSPS protein.
(3) Native EPSPS protein sequences from corn used in the comparison included GenBank Accession Numbers AY106729 and X63374.
(4) GenBank Accession Number AY106729.
(5) The basis for Stine's statement regarding function is discussed in ANZFA (2000).
(6) Oral toxicity study was conducted using microbially expressed protein.
(7) Stine reports values published by the International Life Sciences Institute, the Organization for Economic Cooperation and Development, and the Australia New Zealand Food Authority (now known as Food Standards Australia New Zealand).
(8) Stine notes that the values for several fatty acids were below the limit of quantification (LOQ) for one or more samples. The following fatty acids were therefore neither reported nor statistically analyzed: caprylic, capric, lauric, myristic, myristoleic, pentadecanoic, pentadecenoic, palmitoleic, heptadecanoic, hetpadecenoic, gamma-linolenic, eicosadienoic, eicosatrienoic, arachidonic, behenic, and erucic acids.
(9) Stine notes that the values for Vitamin A were below the LOQ for some samples and that the values for Vitamin B2 and beta-cryptoxanthine were below the LOQ for all samples. The Vitamin A values were reported but not statistically analyzed.
(10) Stine notes that the values for selenium were below the LOQ for some samples and that the values for chromium were below the LOQ for all samples. The selenium values were reported but not statistically analyzed.
(11) Stine notes that all values for furfural were below the LOQ and were not reported.