Novel Food Information - Soybean Modified to Increase Yield MON 87712

Health Canada has notified Monsanto Canada Inc. that it has no objection to the food use of soybean modified to increase yield MON 87712. The Department conducted a comprehensive assessment of this soybean according to its Guidelines for the Safety Assessment of Novel Foods. These Guidelines are based upon internationally accepted principles for establishing the safety of foods with novel traits.


The following provides a summary of the notification from Monsanto Canada Inc. and the evaluation by Heath Canada and contains no confidential business information.

1. Introduction

Monsanto has developed a soybean line with an increased yield phenotype compared to conventional soybean varieties. This line was developed through Agrobacterium-mediated transformation of meristem tissue from A3525 (an elite commercial soybean variety) to introduce the BBX32 coding sequence derived from Arabidopsis thaliana into the Glycine max (L.). Merr. genome to produce the BBX32 protein. BBX32 is a B-box type zinc-finger regulatory accessory protein. In A. thaliana, the BBX32 protein forms a protein complex with endogenous transcription factors to modulate light signal transduction. In MON 87712, the BBX32 protein functions within a similar signalling pathway to control the plant’s response to dark-to-light transition. This modulation of diurnal metabolism in the soybean plant results in the increased yield potential of MON 87712 compared to its conventional counterpart.

The safety assessment performed by Food Directorate evaluators was conducted according to Health Canada’s Guidelines for the Safety Assessment of Novel Foods. These Guidelines are based on harmonization efforts with other regulatory authorities and reflect international guidance documents in this area (e.g., Codex Alimentarius). The assessment considered: how soybean modified to increase yield MON 87712 was developed; how the composition and nutritional quality of this soybean line compared to conventional varieties; and the potential for this line to be toxic or cause allergic reactions. Monsanto Canada Inc. has provided data that demonstrate that soybean modified to increase yield MON 87712 is as safe and of the same nutritional quality as conventional soybean varieties used as food in Canada.

The Food Directorate has a legislated responsibility for pre-market assessment of novel foods and novel food ingredients as detailed in the Food and Drug Regulations (Division 28). Soybean modified to increase yield MON 87712 is considered a novel food under the following part of the definition of a novel food:

  • “c) a food that is derived from a plant, animal or microorganism that has been genetically modified such that
    1. the plant, animal or microorganism exhibits characteristics that were not previously observed in that plant, animal or microorganism.”

2. Development of the Modified Plant

The petitioner has provided information describing the methods used to develop soybean modified to increase yield MON 87712, in addition to the molecular biology data that characterize the genetic change, which results in an increased yield phenotype.

Soybean modified to increase yield MON 87712 was developed using Agrobacterium-mediated transformation of the elite commercial soybean variety A3525 with the binary plasmid PV-GMAP5779. The binary plasmid PV-GMAP5779 contains two separate transfer DNA sequences (i.e., T-DNAs). The first T-DNA (i.e., T-DNA I) contains the gene expression cassette bearing the gene of interest (i.e., the BBX32 coding sequence and its required regulatory elements), and a second T-DNA (i.e., T-DNA II) contains the gene expression cassette of the selectable marker gene cp4 epsps. The CP4 EPSPS protein (5-enolpyruvyl shikimate 3-phosphate synthase from Agrobacterium sp. strain CP4) confers tolerance to the action of glyphosate, the active ingredient in RoundupÒ herbicide. In the transformation process, both T-DNA are integrated into the soybean genome at independent, unlinked loci, and the remainder of the binary plasmid PV-GMAP5779 backbone was not inserted into the plant genome. Traditional breeding methods were subsequently used to isolate plants that only contain the T-DNA I, and not the T-DNA II. This process resulted in the production and selection of the marker-free, MON 87712.

3. Characterization of the Modified Plant

Southern blot analysis of soybean modified to increase yield MON 87712 demonstrated the insertion of a single copy of the T-DNA I insert (i.e., the BBX32 coding sequence and its required regulatory elements) at a single locus within the plant genome. Analysis also verified the absence of the T-DNA II and any extraneous sequence related to the binary plasmid PV-GMAP5779. Comparison of the genomic DNA sequences flanking the T-DNA I insert and the identical sequence in conventional A3525 soybean indicates that a 42-base pair (bp) deletion from the conventional genome sequence occurred upon insertion of the T-DNA I. This deletion is presumably resulted from double-stranded break repair mechanisms in the plant during Agrobacterium-mediated transformation. With this exception, no major unexpected rearrangements occurred during the development of MON 87712.

Bioinformatics analyses were performed on the T-DNA I insert and flanking genomic DNA sequences. Based on current databases and tools, no known open reading frames (ORFs) were interrupted due to the insertion of the transgenic DNA in the pre-insertion locus of the A3525 genome. Analysis of all six potential reading frames of the sequences consisting of the T-DNA I insert and flanking genomic DNA sequences was conducted to search for potential newly created ORFs (conservatively defined as any sequence between two stop codons [TGA, TAG, TAA] and at least 8 amino acids in length). A total of 9 putative polypeptides that met these criteria were identified, however none of them showed any similarity to any known toxin, allergen, or other biologically active protein that could affect human or animal health.

Generational stability of the single insert was determined across multiple generations of MON 87712. Genomic DNA from five consecutive generations were analysed by Southern blot analysis and the presence of the single insert was confirmed in each generation. Chi-squared (c2) analysis of segregation data supports the conclusion that the BBX32 coding sequence in MON 87712 resides at a single locus within the soybean genome and is inherited according to Mendelian inheritance principles.

4. Product Information

Soybean modified to increase yield MON 87712 differs from its unmodified counterpart by the addition of the BBX32 coding sequence and its required regulatory elements. The insertion of the BBX32 gene results in the expression of the BBX32 protein: a B-box type zinc-finger regulatory accessory protein that interacts with one or more endogenous transcription factors to regulate the plant’s response to dark-to-light transition. This modulation of diurnal metabolism in the soybean plant results in the increased yield potential of MON 87712 compared to its unmodified counterpart.

5. Dietary Exposure

The consumption pattern of soybean-based food products is not expected to change with the addition of MON 87712 to the food supply as it will be used in the same way as currently commercialized soybeans.

The petitioner did not provide information on dietary intake of soybean; however, BBX32 protein consumption was estimated using the Dietary Exposure Evaluation Model (DEEM-FCID). DEEM food consumption data were obtained from the 1994-1996 and 1998 United States Department of Agriculture (USDA) Continuing Survey of Food Intake by Individuals (CSFII), assuming that all consumed soybean products were derived from MON 87712. As no BBX32 protein was detected in the seed tissue (i.e., the tissue consumed by humans), the petitioner used a methodology approved by the United States (US) Environmental Protection Agency (EPA) to derive a margin of exposure (MOE) between intake of BBX32 and the no observed adverse effect level (NOAEL) in an acute toxicity study in mice (described under Chemistry/Toxicology).

Using information pertaining to the consumption of soy products from a Health Canada 2008 document Intakes of food that are potential sources of melamine, and applying a calculated expression level of 5 ng BBX32 protein/g fresh weight (fwt) soybean (using the EPA-approved methodology described under Chemistry/Toxicology) leads to a BBX32 protein intake of 76.25 ng/kg body weight (bw) for 4-year olds and 29.8 ng/kg bw for 51 plus-year old females (i.e., the two highest soybean-based food product consuming groups).

According to the Organization for Economic Co-operation and Development (OECD) (2001) consensus document on soybean, the worldwide human consumption of soybean is not well documented. However, it states that the average US consumption of soybeans is low. Soybean oil makes up 94% of the soybean-based food ingredients consumed by humans. The oil would contain negligible amounts of protein (and even less BBX32 protein). According to OECD (2001), the daily consumption of soybean-based food products in Japan is 69.9 g/day. For a 70-kg person, this would be about 1 g/kg bw/day. Since Japan, unlike Canada, is one of the higher soybean-consuming countries, our own estimates of soybean-based product consumption of 15.25 and 5.96 g/kg bw for 4-year olds and 51 plus-year old females, respectively, is almost certainly high. Consequently our own estimates of respective exposures to the BBX32 protein would also be too high.

Soy-based formulas have about 1.65 to 2.1 g of protein per dL (American Academy of Pediatrics, 1998). While not all the protein is from soybean, the assumption is made for the following calculations: An infant of 0 to 3 months has a mean body weight of 4.5 kg and consumes an average of 750 mL of breast milk or formula per day (Chemical Health Hazard Assessment Division Reference Library, Health Canada). At 2 g protein/dL, 750 mL (7.5 dL) of soy-based formula provides 15 g protein/day or 3.33 g/kg/ bw/day for a 4.5-kg infant. Assuming a BBX32 protein content of 5 ng/g protein, these infants would consume 16.65 ng/kg bw calculated by the petitioner for non-nursing infants.

6. Nutrition

The petitioner provided data to compare the composition of MON 87712 to its conventional parental control variety A3525 as well as comparison with 17 other commercially available reference soybean varieties. Harvested seed and forage tissues were evaluated in the study. All samples were grown in 8 geographically diverse US field trials during the 2009 growing season. Each test site included MON 87712 (test), A3525 (control), and 3 conventional commercial soybean varieties.

The sites were planted in a randomized complete block design with 4 blocks per site using test, control, and reference substances. Seed samples were analyzed for nutrients: proximates (ash, fat, moisture, protein, and carbohydrate by calculation), acid detergent fibre (ADF), neutral detergent fibre (NDF), amino acids, fatty acids (C8-C22), and vitamin E. The anti-nutrients assessed in harvested soybean seed included raffinose, stachyose, lectin, phytic acid, trypsin inhibitors, and isoflavones. Data for vitamin K were not provided and the petitioner indicated that this was not a requirement by the previous OECD consensus document for soybean (the document being current at the time in which the notification was made).

All data for each test site and a combined-site analysis of harvested soybean seed were provided. There were 34 nutrient comparisons. Of these, 22 of the comparisons were not significantly different between MON 87712 and the conventional control A3525. The 12 analytes that differed were within the range of literature values and that of the reference varieties. In the combined-site analysis of the 8 anti-nutrients tested in soybean seed there were no significant differences observed for any of the analytes between MON 87712 and the control A3525.

7. Chemistry/Toxicology

Expression of the BBX32 protein in MON 87712 is low and thus could not be purified in sufficient quantity for use in subsequent safety studies. Therefore, recombinant BBX32 protein was produced in an Escherichia coli expression system, the sequence of which was engineered to match that of the BBX32 protein produced in MON 87712. Equivalence of the physiochemical characteristics and functional activity of plant- and microbial-derived BBX32 proteins was confirmed by a series of analytical techniques including: western blot analysis, N-terminal sequence analysis, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF), a protein-protein binding assay, and glycosylation analysis.

Due to the low level of BBX32 protein expression in MON 87712, the plant-derived BBX32 protein was not isolated to a high level of purity. Consequently, the physiochemical characterization of plant-derived BBX32 protein was limited to western blot analysis. This technique indicated that plant- and microbial-derived BBX32 proteins have equivalent apparent migrational mobility (an approximate molecular weight of 25 kDa), and both proteins are immunoreactive to anti-BBX32 monoclonal antibodies. N-terminal sequence analysis of the microbial-derived BBX32 protein revealed that the protein’s N-terminus amino acid sequence matches the N-terminus sequence predicted by the BBX32 coding sequence in MON 87712. The microbial-derived BBX32 protein also exhibited protein-binding activity to an appropriate target protein. The petitioner provided the rationale that the BBX32 protein expressed in MON 87712 also exhibits positive protein-binding activity as evident by the increased yield phenotype of the transgenic plant. Lastly, while the BBX32 protein sequence contains a single potential N-glycosylation site (i.e., NNT at position 172-174), it lacks the N-terminal signal sequence required for transport to the endoplasmic reticulum (required for both N- and O-glycosylation). The petitioner stated that this is comparable to the Agrobacterium sp. CP4 EPSPS (the sequence of which contains potential N-glycosylation sites, but lacks the signal sequence required for transportation and has been demonstrated to not be glycosylated). While equivalence between plant- and microbial-derived BBX32 proteins could only be directly assessed by western blot analysis, the rationales provided by the petitioner to support equivalence of N-terminal sequence, functional activity, and glycosylation status sufficiently demonstrate that these two proteins are equivalent by weight of evidence.

The microbial-derived BBX32 protein was used in a 14-day acute oral toxicity study, where mice (10 animals/sex/group) were gavaged with BBX32 protein (29 mg/kg bw) or bovine serum albumin (BSA) (35 mg/kg bw) as a control. Mice were observed for 14 days and then sacrificed. Necropsies were performed. The NOAEL for acute oral toxicity in mice was set at 29 mg/kg bw, based on the absence of death or any observable adverse effects.

BBX32 protein was not detected in soybean by immunoblot methodology (Limit of detection, LOD = 2.5 pg). In the absence of this information, the petitioner estimated the amount of BBX32 protein in MON 87712 soybean (expressed as ng/g fresh weight), using the following formula: ng/g fwt = lowest spiked BBX32 protein standard with a visible band (pg) ÷ 2 [volume of extract loaded per lane (μl) ÷ (buffer to tissue ratio)]. This estimate was essentially the LOD and on this basis, the amount of BBX32 protein was calculated to be 10 ng/g fwt.

The petitioner estimated exposure to the BBX32 protein based on a value equal to one-half of the LOD of the assay used to quantify the expression level (i.e., 10 ng/g fwt ÷ 2 = 5 ng/g fwt) and estimates of acute dietary exposure to soybeans for the 95th percentile “eaters only” which was determined using the Dietary Exposure Evaluation Model (DEEM-FCID), version 2.16). DEEM food consumption data are obtained from the 1994-1996 and 1998 USDA Continuing Survey of Food Intake by Individuals (CSFII) and assumed that 100% of soybean-based products consumed (excluding oils) are derived from MON 87712. This methodology is used by the US EPA for non-detectable pesticide residues when a substance is not detected in blended commodities, and was considered an acceptable methodology in this case.

The petitioner used a MOE approach to compare a single dietary intake of BBX32 protein for the general population and non-nursing infants to the NOAEL of 29 mg/kg bw in mice. The MOEs obtained were 29,000,000 for the general population and 539,000 for non-nursing infants (the subpopulation with the highest estimated exposure).

Digestibility of the BBX32 protein was assessed by incubation with simulated gastric fluid (SGF), incubation in SGF followed by incubation in simulated intestinal fluid (SIF), and incubation in SIF alone. Digestion of the protein was analyzed at specific time intervals. More than 98.75% of the full-length BBX32 protein was digested in SGF within 30 seconds. Transiently-stable protein fragments were observed at times less than 20 minutes when measured by SDS-PAGE with protein staining and at times less than 10 minutes when assessed by western blot analysis with BBX32-specific antibodies. When a 2-minute digestion in SGF was followed by digestion in SIF, the transiently-stable protein fragments were fully digested within 30 seconds of being added to SIF. Following incubation in SIF alone, the BBX32 protein was fully digested within 5 minutes. The rapidity of the digestion of the BBX32 protein in vitro indicates that it would be unlikely to survive the human digestive system and therefore not likely to pose a systemic risk.

Bioinformatics analyses were performed to assess the potential of the BBX32 protein for toxicity, allergenicity, and biological activity. The FASTA algorithm was used to compare the amino acid sequence of the BBX32 protein to sequences in toxicity (GenBank, release 181), allergenicity (FARRP, 2011), and public domain (GenBank, release 181) databases. No amino acid sequence similarities between the BBX32 protein and known toxins, allergens, or biologically active proteins of concern were found.

Based on the currently available data, potential exposure to the BBX32 protein from MON 87712 would not be a toxicological concern.

8. Allergenicity

Many known protein allergens are glycosylated and an absence of protein-glycosylation is supportive of non-allergenicity. As previously mentioned, the BBX32 protein sequence contains a potential N-glycosylation consensus sequence (i.e., NNT at position 172-174), but lacks the N-terminal signal sequence required for transport to the endoplasmic reticulum. This suggests that the BBX32 protein expressed in MON 87712 is not glycosylated and would not likely be an allergenic protein.

Additionally, the serum of soy-allergenic patients contains IgE antibodies specific to a number of known soybean allergens. The sera of 14 soybean-allergenic subjects were individually assessed for binding to extracts of MON 87712, the control variety A3525, and 17 reference soybean varieties using an enzyme-linked immunosorbent assay (ELISA). The serum IgE binding values, expressed as ng IgE/ml serum, obtained with extracts of the 17 reference soybean varieties were used to calculate a 99% tolerance interval. The IgE binding values obtained for extracts from MON 87712 and A3525 were compared to the tolerance interval derived for each serum. For all soy-allergenic subjects, the IgE binding values obtained from MON 87712 and A3525 extracts were within the reference soybean tolerance limits, indicating that the allergenicity of MON 87712 and its parental control variety A3525 are comparable to that of other conventional commercial soybean varieties.

Based on the currently available data, potential exposure to the BBX32 protein (and the MON 87712 soybean plant as a whole) would not be a greater allergenic concern than conventional soybean.


Health Canada’s review of the information presented in support of the food use of soybean modified to increase yield MON 87712 does not raise concerns related to food safety. Health Canada is of the opinion that food derived from MON 87712 is as safe and nutritious as food from current commercial soybean varieties.

Health Canada's opinion deals only with the food use of soybean modified to increase yield MON 87712. Issues related to its use as animal feed have been addressed separately through existing regulatory processes in the Canadian Food Inspection Agency (CFIA). The CFIA evaluated information provided on the environmental, animal, and human health safety of MON 87712 with the intended use in animal feed. From their assessment, the CFIA concluded that there are no concerns from an environmental and feed safety perspective. This perspective is applicable to the food and feed products derived from MON 87712 destined for commercial sale.

It is the continuing responsibility of the food manufacturer or importer to ensure that their products are in compliance with all applicable statutory and regulatory requirements. Any new information obtained in relation to these products which have potential health and safety implications should be forwarded to Health Canada for our consideration in order to ensure the continued safety and integrity of all foods available in the Canadian marketplace. The sale of a food which poses a hazard to the health of consumers would contravene the provisions of the Food and Drugs Act.

This Novel Food Information document has been prepared to summarize the opinion regarding the subject product provided by the Food Directorate, Health Products and Food Branch, Health Canada. This opinion is based upon the comprehensive review of information submitted by the petitioner according to the Guidelines for the Safety Assessment of Novel Foods.

(Également disponible en français)

For further information, please contact:

Novel Foods Section
Food Directorate
Health Products and Food Branch
Health Canada, PL2204A1
251 Frederick Banting Driveway
Ottawa, Ontario K1A 0K9

Page details

Date modified: