|
In a submission dated August 31, 2005, Syngenta Seeds, Inc. (Syngenta) provided to the Food and Drug Administration (FDA) a safety and nutritional assessment of genetically engineered AMY797E alpha-amylase corn designated as corn event 3272. Syngenta provided additional information to the FDA on May 1, 2006, August 2, 2006, November 8, 2006, January 19, 2007, March 2, 2007, March 6, 2007, March 7, 2007, May 31, 2007, July 11, 2007 and July 12, 2007. Syngenta concluded that food and feed derived from corn event 3272 are as safe and nutritious as food and feed derived from conventional corn varieties.
Corn event 3272 was genetically engineered to contain a chimeric, thermostable, alpha-amylase gene derived from alpha-amylase genes from three hyperthermophilic microorganisms within the order Thermococcales. The resulting transformed corn is intended to be mixed with other corn for use in dry grind fuel ethanol production as a source of alpha-amylase, replacing the need for the addition of microbially-produced amylase during processing. The enzyme catalyzes the hydrolysis of the alpha-1,4-glucosidic bonds of amylase and amylopectin polymers into dextrins, maltose, and glucose. The recombinant enzyme is stable and active at high temperatures of dry grind ethanol production.
Syngenta states that, based on its safety assessment, corn event 3272 is as safe as other corn lines, and as such may be used as food, feed or in the production of food products or ingredients. However, Syngenta states that grain from corn event 3272 is primarily targeted for the dry grind fuel ethanol production industry in the United States.
Syngenta constructed the plasmid vector pNOV7013. The plasmid contains two expression cassettes within its T-DNA region. The T-DNA region contains a single copy of the amy797E alpha-amylase expression cassette and a single copy of the phosphomannose isomerase (pmi) expression cassette as a selectable marker. The T-DNA region of pNOV7013 was incorporated into immature corn embryo cells from a proprietary corn line (NP2499/NP2500) using Agrobacterium-mediated transformation. The T-DNA region of pNOV7013 contains the following genetic elements.
| Cassette Grouping | Genetic Elements | Description |
|---|---|---|
| Right Border | Right border sequence from nopaline Ti plasmid. Delineates the T-DNA region for Agrobacterium-mediated transformation. | |
| amy797E | GZein | Promoter region from the Zea mays storage protein gene (zein). |
| amy797E (amy797GL3, GZein ss, ER rs) | Chimeric alpha-amylase gene (amy797GL3) derived from order Thermococcales microbes. Also contains N-terminal maize gamma-zein signal sequence (GZein ss) and a C-terminal SEKDEL endoplasmic reticulum retention sequence (ER rs). | |
| PEPC9 | Intron #9 from the phosphoenolpyruvate carboxylase gene from Zea mays. | |
| 35S terminator | Terminator sequence from the 35S RNA from the cauliflower mosaic virus genome. | |
| pmi | ZmUbiInt | Promoter region from Zea mays polyubiquitin gene. Contains the first intron. |
| pmi | E. coli manA gene encoding phosphomannose isomerase. | |
| NOS | Terminator sequence from the nopaline synthase gene of Agrobacterium tumefaciens. | |
| Left Border | Left border sequence from the nopaline Ti plasmid. Delineates the T-DNA region for Agrobacterium-mediated transformation. |
Plasmid pNOV7013 contains several genes on its plasmid backbone necessary for maintenance and selection of the plasmid. These sequences are not intended for transfer into the plant genome. The plasmid backbone of pNOV7013 contains origins of replication that allow replication of the plasmid in both Agrobacterium tumefaciens and Escherichia coli. The plasmid contains the spec gene encoding the Tn7 adenylyltransferase conferring resistance to erythromycin, streptomycin and spectinomycin. pNOV7013 also contains the virG gene, which is a gene involved in regulation of virulence in A. tumefaciens, and the repA gene, which is a gene responsible for plasmid replication in Gram-negative, plant-associated bacteria.
Syngenta performed polymerase chain reaction (PCR), restriction digests, and Southern blot analysis to support its conclusion that corn event 3272 contains one intact copy of its amy797E and pmi expression cassette. Syngenta determined the nucleotide sequence of the T-DNA region present in corn event 3272 to demonstrate the integrity of the insert. Syngenta reports that corn event 3272 contains no detectable genetic material from the pNOV7013 backbone, including the spec, virG, and repA genes.
Syngenta examined the stability and inheritance pattern of the introduced traits through conventional breeding over five generations consisting of four rounds of backcrossing to conventional inbred lines. Syngenta reported no significant differences in the observed-to-expected segregation ratios for the amy797E gene over five generations, as demonstrated by the chi-square (χ2) values. Syngenta reported that these segregation data indicate a single-locus, Mendelian inheritance pattern for the insert in corn event 3272.
Syngenta provided information on the identity, function, and characterization of the genes as well the expression levels of the gene products. They also provided information on the potential allergenicity and toxicity of the expressed proteins.
Syngenta compared the amino acid sequence of the AMY797E alpha-amylase protein to other alpha-amylases, stating that AMY797E shared a 93% homology to a microbially derived alpha-amylase BD5088 (Innovase LLC) produced through a similar recombination technique as AMY797E. The BD5088 alpha-amylase was the subject of a generally recognized as safe (GRAS) notice (GRN 126) for use in food. Syngenta also provided the complete amino acid sequence of AMY797E. Additionally, Syngenta provided information about the functional activity of AMY797E to demonstrate its functionality as an alpha-amylase.
The pmi gene, introduced as a selectable marker into corn event 3272, encodes a phosphomannose isomerase (PMI) enzyme that catalyzes the inter-conversion of mannose-6-phosphate and fructose-6-phosphate. The expression of the pmi gene in the plant allows the plant to survive and grow on media containing only mannose as the only or primary carbon source, which facilitates selection of transformed plants.
Syngenta reports that PMI enzymes exist widely in nature among both prokaryotes and eukaryotes. Syngenta states that PMI enzymes have been found in plants such as tobacco, walnut, and soybean and other legumes, although Syngenta notes that sequence homology for the pmi gene introduced into event 3272 is highest for enteric Gram-negative bacteria (70-100% homology).
Syngenta reports that AMY797E alpha-amylase expression levels were measured by enzyme-linked immunosorbent assay (ELISA) in several tissues at various stages of development. Syngenta notes that expression of AMY797E is directed primarily to the kernel by the maize gamma-zein promoter, and as expected, the highest concentration of AMY797E alpha-amylase is in the kernel. Mean concentration measured in mature or senescent kernels ranged from 838 ± 268 μg/g fresh weight (fw) (1004 ± 322 μg/g dry weight (dw)) to 955 ± 225 μg/g fw (1335 ± 358 μg/g dw). Mean concentration in younger "dough" stage (R4) kernels ranged from 874 ± 160 μg/g fw (1994 ± 228 μg/g dw) to 1627 ± 338 μg/g fw (3365 ± 780 μg/g dw). Concentration of AMY797E alpha-amylase varied in whole plant samples, at various growth stages, from < 12 μg/g fw (< 37 μg/g dw) to 281 ± 108 μg/g fw (668 ± 248 μg/g dw).
PMI protein concentrations were also measured in various tissues and developmental stages. PMI was detected in most tissues tested and concentrations were similar regardless of the developmental stage tested. The highest levels were detected in pollen, with concentrations ranging from 8.0 to 8.5 μg/g fw (17.0 - 18.2 μg/g dw). Mean concentrations of PMI in kernels over all growth stages ranged from < 0.4 μg/g fw (< 0.5 μg/g dw) to 0.8 ± 0.1 μg/g fw (1.8 ± 0.4 μg/g dw). Concentrations of PMI in whole plant samples from all developmental stages ranged from < 0.3 μg/g fw (< 0.6 μg/dw) to 1.5 ± 0.3 μg/g fw (3.6 ± 0.9 μg/g dw).
Syngenta states that the potential allergenicity of the AMY797E alpha-amylase and PMI proteins were assessed by searching for amino acid homology between these proteins and known allergen protein sequences. These searches were conducted using a database comprised of identified or putative allergen sequences from publicly available databases (GenPept, PIR, SWISS-PROT, FAARP and IUIS) and additional putative allergen sequences from the scientific literature. Syngenta also assessed the stability of the AMY797E and PMI proteins using in vitro digestibility assays.
Syngenta reports that the donor organisms (Thermococcus/Pyrococcus) used to develop AMY797E alpha-amylase protein are not known to be allergenic.
Syngenta reports that for AMY797E, there were no amino acid sequence identities of greater than 35% in segments of 80 amino acids with any entries in the database. Syngenta does note that there was a single segment of 8 contiguous amino acids in AMY797E that matched a known allergenic sequence derived from an insect. However, Syngenta notes that the allergenic epitopes are known for this allergen (American cockroach, Per a 3) and there is no overlap between these binding epitopes and the eight amino acid region of sequence identity with AMY797E alpha-amylase. Therefore, Syngenta maintains that this sequence identity is not biologically relevant and has no implication for the allergenic potential of the AMY797E alpha-amylase.
Syngenta provides data on the in vitro digestibility of AMY797E. Syngenta reports that AMY797E was susceptible to proteolytic degradation in simulated gastric fluid (SGF) containing pepsin, indicating that AMY797E is degraded within 5 minutes. Syngenta states that AMY797E is not stable to digestion and is therefore unlikely to become allergenic.
Syngenta notes that AMY797E is a thermostable protein. In addition, Syngenta reports that analysis of AMY797E as expressed in corn event 3272 does not reveal evidence of post-translational glycosylation.
Syngenta reports that the donor organism (E. coli) used to develop the PMI protein is not known to be allergenic.
Syngenta reports that for PMI, there were no amino acid sequence identities of greater than 35% in segments of 80 amino acids with any entries in the database. Syngenta does note that there was a single segment of 8 contiguous amino acids in the PMI protein that matched a known allergen, α-parvalbumin from Rana species CH2001, an edible frog. Syngenta reports that further investigation of PMI using serum from one known α-parvalbumin-sensitive individual demonstrated a lack of reactivity with PMI. Syngenta therefore concluded that the sequence identity between PMI and α-parvalbumin is not biologically relevant.
Syngenta provides data on the in vitro digestibility of PMI. Syngenta reports that the PMI protein was found to be degraded when sampled immediately after time zero. Syngenta states that PMI is not stable to digestion and is therefore unlikely to become allergenic.
Syngenta notes that the PMI protein is labile to heat. In addition, Syngenta reports that PMI is unlikely to be glycosylated, given that the PMI protein does not contain consensus amino acid sequences required for N-glycosylation and the protein is not targeted to a cellular glycosylation pathway.
Syngenta reported results from an acute oral toxicity study in mice where 1511 milligrams/kilogram body weight (mg/kg bw) AMY797E protein, or a control, was given by gavage. Syngenta indicates that the AMY797E test substance was prepared from event-3272 grain and was determined to be 42% AMY797E protein. Syngenta states that the animals were monitored for 14 days and were sacrificed. Syngenta reports that there were no effects of treatment on any observation, including body weight, food consumption, organ weight, or histopathology.
Syngenta reports that the potential toxicity of AMY797E was also assessed by comparing its amino acid sequence against all publicly available protein sequences identified as toxins in the National Center for Biotechnology Information Entrez Protein Database. Syngenta reports that no significant sequence homology to any known toxins was identified.
Syngenta reports results from an acute oral toxicity study where PMI protein was given to mice by gavage at a dose of 3030 mg/kg bw. Syngenta notes that the PMI protein used in this study was obtained by over-expressing the protein in E. coli. Syngenta states that the animals were monitored for 14 days and were sacrificed. Syngenta reports that there were no effects of treatment on any observation, including body weight, food consumption, organ weight, or histopathology.
Syngenta reports that the potential toxicity of PMI was also assessed comparing its amino acid sequence against all publicly available protein sequences identified as toxins in the National Center for Biotechnology Information Entrez Protein Database. Syngenta reports that no significant sequence homology to any known toxins was identified.
Syngenta notes that corn grown in the U.S. is primarily the yellow dent type, a commodity crop used primarily to feed domestic animals, either as grain or forage. The remainder of the crop being exported or processed by wet or dry milling to yield human food products such as high fructose corn syrup, starch or oil, grits and flour. The by-products of wet and dry milling are commonly used in animal feed.
Syngenta evaluated the composition of forage and grain from event 3272-derived corn hybrids relative to negative segregant (near-isogenic) control corn hybrids of similar genetic background.
Compositional analyses were performed on forage and grain. Syngenta analyzed corn grown at 10 different locations over a two year period, with three replicate plots of each genotype1 planted at each location in randomized complete blocks. Grain was harvested from 6 locations in both years and forage was harvested from 6 locations in 2003 and 7 locations in 2004. Three of the locations used in 2003 were again used in 2004 for growing corn grain, and four of the locations used in 2003 were again used for growing corn forage in 2004. Two hybrid pairs 2 were grown in 2003 and one hybrid pair was grown in 2004. The data were combined for statistical analysis across locations, hybrid pairs, and growing seasons. Compositional data were statistically analyzed using a mixed model analysis of variance with locations serving as blocks. Statistical significance was assigned at p < 0.05 indicating that the difference between the treatments was statistically different at the 5% customary level. Syngenta compared the compositional data with published literature values for each analyte to assess whether statistically significant differences in the composition of the event 3272-derived hybrids and the corresponding near-isogenic control maize were biologically meaningful.
Syngenta determined the levels of the following analytes in forage from corn event 3272-derived and the near-isogenic control hybrids collected at the R4 development stage. The following groups of analytes were measured:
A list of specific analytes contained in each group is shown in Table 2. Syngenta reported mean values for these analytes fell within published literature ranges.3,4 Syngenta conducted a combined statistical analysis of forage data from all three field trials for analytes5 marked with an asterisk (*) in Table 2, and found statistically significant differences between the event 3272-derived hybrids and their near-isogenic control lines for protein, carbohydrate and acid detergent fiber (ADF). Protein levels were statistically significantly higher, and carbohydrates and ADF were significantly lower in event 3272-derived hybrids than their respective near-isogenic control lines. Syngenta concluded that these differences were not biologically meaningful.
| Proximates | Minerals | Amino Acids | Fatty Acids | Anti-Nutrients | Secondary Metabolites | Vitamins |
|---|---|---|---|---|---|---|
|
ash* fat moisture* protein* carbohydrate* acid detergent fiber (ADF)* neutral detergent fiber (NDF)* total dietary fiber (TDF) starch (grain only) |
calcium* copper iron magnesium manganese phosphorus* potassium sodium zinc selenium |
methionine cysteine lysine tryptophan threonine isoleucine histidine valine alanine leucine arginine phenylalanine glycine aspartic acid glutamic acid proline serine tyrosine |
palmitic (16:0) stearic (18:0) oleic (18:1) linoleic (18:2) linolenic (18:3) |
phytic acid trypsin inhibitor raffinose |
furfural p-coumaric acid ferulic acid inositol |
beta-carotene beta-cryptoxanthin folic acid vitamin B1 vitamin B2 vitamin B3 vitamin B6 vitamin C tocopherols(α-,β-,γ-,δ-) |
Syngenta determined the levels of the following analytes in grain from corn event 3272-derived and the near-isogenic control hybrids collected at the R6 growth stage (maturity). The following groups of analytes were measured:
A list of specific components contained in each group is shown in Table 2. Syngenta reported mean values were within published literature ranges, with minor exceptions noted.6,7 Syngenta conducted a combined statistical analysis of grain data from all field trials for all analytes listed in Table 2, excluding γ-tocopherol and β-cryptoxanthin8 and the nine analytes below the limit of quantitation. Exclusion of these analytes did not, however, affect Syngenta's conclusions. Low levels of these nutrients and anti-nutrients are consistent with values reported in the published literature and variable levels of certain nutrients (e.g. selenium) found in soil throughout the U.S.
Syngenta reported statistically significant differences in mean values between event 3272-derived hybrids and controls for protein and manganese, which were higher in event 3272-derived hybrids than near-isogenic control hybrids. Syngenta also reported statistically significant differences in the mean values of carbohydrates, total dietary fiber, neutral detergent fiber, vitamin B6, β-carotene, inositol, and ferulic acid, which were lower in event 3272-derived hybrids. Event 3272-derived hybrids had significantly higher levels of almost all amino acids compared to the near-isogenic control lines, with the exception of arginine, cysteine, and lysine which were not different. Syngenta has compared all mean values to published literature ranges and concludes differences between event 3272-derived and near-isogenic control hybrids are not biologically meaningful.
Syngenta reports results from a 42-day broiler chicken feeding study comparing birds fed event-3272 grain versus the near-isogenic control and commercially-available corn. Syngenta indicates that all of the diets derived from treatment and control diets supported rapid growth with low mortality rates and excellent feed conversion ratios.9 Syngenta also reports no evidence of biologically-significant differences in growth or feed conversion in chickens fed event-3272 grain compared to near-isogenic control or commercially available corn. Furthermore, Syngenta notes the absence of any adverse nutritional or toxic effects in chickens fed these diets.
Syngenta has concluded that AMY797E alpha-amylase corn event 3272 is not materially different in composition, safety, or any other relevant parameter from corn now grown, marketed, and consumed. At this time, based on Syngenta's data and information, the agency considers Syngenta's consultation on AMY797E alpha-amylase corn event 3272 to be complete.
Richard E. Bonnette
(2)Syngenta defines hybrid pairs as the result of an initial cross-breeding of event 3272 and a non-transgenic line followed by self-crossing or cross-breeding of the resulting progeny one or more times and the selection of homozygous trait positive and negative segregants.
(3)Syngenta reported fat levels below the 0.1% limit of quantification in forage from both event 3272-derived hybrids and their respective near-isogenic control lines at a single field site. Excluding data from this site, reported mean values for fat were within published literature ranges.
(4)Values were within the ranges found in the International Life Sciences Institute Crop Composition Database (version 3.0, released April 10, 2006, available at: www.cropcomposition.org). Excluding phosphorous, means for all analytes also fell within values cited by the Organisation for Economic Co-operation and Development (OECD) "Consensus Document on Compositional Considerations for New Varieties of Maize (Zea Mays): Key Food and Feed Nutrients, Anti-Nutrients and Secondary Plant Metabolites." (Series on the Safety of Novel Foods and Feeds, No. 6, 2002).
(5)Syngenta limited forage analytes in the combined statistical analysis to those suggested by the OECD (2002). Fat was excluded from the combined statistical analysis due to levels below the limit of detection.
(6)The following 9 grain analytes had one or more observations below the limit of quantitation of the assay: sodium, selenium, raffinose, phytic acid, furfural, tocopherols (α-, β- and γ-), and vitamin C. An observation represents the mean of 3 replicate plots per location, with grain from 15 plants pooled from each plot.
(7)Values were within the ranges found in the International Life Sciences Institute Crop Composition Database (version 3.0, released April 10, 2006, available at: www.cropcomposition.org). Fatty acids mean values were within ranges cited by OECD (2002).
(8)For several of the analytes, Syngenta reported that the treatment (genotype) effect differed by location. However, based on the statistical model used, the interpretation of the data was not affected.
(9)Syngenta notes that the non-transgenic control grain appeared to contain low levels (approximately 1.6-2.6%) of event-3272 corn.
