EPA+DHA Herbicide Tolerant Canola Event LBFLFK

Novel Food Information

On this page

• Background
  • 1. Introduction
  • 2. Development of the Modified Plant
  • 3. Characterization of the Modified Plant
  • 4. Product Information
  • 5. Dietary Exposure
  • 6. Nutrition
  • 7. Toxicology
  • 8. Allergenicity
• Conclusion

Background:

Health Canada has notified BASF Canada that it has no objection to the food use of canola event LBFLFK. The Department conducted a comprehensive assessment of this variety 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 BASF Canada and the evaluation by Heath Canada and contains no confidential business information.

1. Introduction

BASF Canada has developed LBFLFK canola to produce the long-chain polyunsaturated fatty acids docosahexaenoic acid (DHA, C22:6n-3) and eicosapentaenoic acid (EPA, C20:5n-3) in the seed and to be tolerant to the imidazolinone herbicides. Recombinant DNA techniques were used in the development of LBFLFK canola to introduce ten fatty acid desaturase and elongase proteins and the acetohydroxy acid synthase (AHAS) protein which confers tolerance to imidazolinone herbicides.

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 LBFLFK canola was developed; how the composition and nutritional quality of LBFLFK canola compared to non-modified varieties; and the potential for LBFLFK canola to be toxic or cause allergic reactions. BASF Canada has provided data that demonstrates that LBFLFK canola is as safe and of the same nutritional quality as traditional canola 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). LBFLFK canola is considered a novel food under the following part of the definition of novel foods:

"c) a food that is derived from a plant, animal or microorganism that has been genetically modified such that:

• 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 LBFLFK canola and the molecular biology data that characterize the genetic change, which results in the production of the fatty acids EPA and DHA in the plant's seeds through the expression of ten fatty acid desaturase and elongase proteins and tolerance to the imidazolinone herbicides through introduction of a modified AHAS protein.

LBFLFK canola was developed by Agrobacterium-mediated transformation of hypocotyl segments of commercial canola variety Kumily utilizing the vector LTM593. The vector contains 13 gene expression cassettes: Physcomitrella patens (P. patens) delta-6 elongase, Thraustochytrium sp. delta-5 desaturase, Ostreococcus tauri (O. tauri) delta-6 desaturase, Thalassiosira pseudonana (T. pseudonana) delta-6 elongase, Phytophthora sojae (P. sojae) dela-12 desaturase, Pythium irregulare (P. irregular) omega-3 desaturase, Phytophthora infestans (P. infestans) omega-3 desaturase, Thraustochytrium sp. delta-5 desaturase, Thraustochytrium sp. delta-4 desaturase, P. irregular omega-3 desaturase, Pavlova lutheri (P. lutheri) delta-4 desaturase, and O. tauri delta-5 elongase in the EPA and DHA biosynthetic pathway, and a modified Arabidopsis thaliana (A. thaliana) acetohydroxy acid synthase which conveys herbicide tolerance. Two gene expression cassettes were introduced twice (O3D(Pir) and D5D(Tc) but have different seed-specific promotors to increase expression during seed maturation. Expression of the ten EPA and DHA biosynthesis genes results in the conversion of oleic acid (OA) to EPA and DHA in LBFLFK canola seeds.

The genes used for EPA and DHA biosynthesis in LBFLFK canola were chemically synthesized based on the sequences originally identified and characterized from the source organisms. The DNA sequence of each gene was modified to optimize the translation rate. The amino acid sequence of each novel protein is unchanged with two exceptions: (1) the c-D6E(Tp) sequence has the amino acid substitution P196S, and (2) the c-AHAS(At) sequence contains two mutations resulting in the desired amino acid substitutions A122T and S653N that confer herbicide tolerance. The safety of the AHAS protein with these two mutations has been previously assessed by Health Canada. No component from the source organisms was directly used in the development of LBFLFK canola.

The Petitioner provided information to support the safety and historical use (where applicable) of each source organism P. patens, Thraustochytrium sp., O. tauri, T. pseudonana, P. sojae, P. irregular, P. infestans, P. lutheri in the EPA and DHA biosynthetic pathway, and a modified A. thaliana, AHAS).

3. Characterization of the Modified Plant

The number of integration sites of the T-DNA insert in LBFLFK canola was characterized through Illumina-based sequencing of a T3 line and the parent variety. Based on the results of

this whole genome sequencing (WGS) analysis, two T-DNA integration sites were identified: the first in the Cnn random chromosome and the second in the C03 chromosome of the LBFLFK canola genome.

A vector-targeted sequencing approach confirmed the absence of vector backbone sequence in LBFLFK canola.

The integrity and adjacent canola genomic sequences of the Cnn random and C03 T-DNA inserts were further characterized by locus-specific PCR followed by Sanger sequencing and analysis of Bacterial Artificial Chromosome (BAC) clones containing LBFLFK Insert1 and LBFLFK Insert2, generated from T3 leaf material. A comparison of the sequences obtained from Kumily with those from the 3' and 5' flanking regions of the two T-DNA inserts in LBFLFK revealed an 8-bp deletion of the canola genome at the integration site of the Cnn random T-DNA insert and a 31-bp deletion of the canola genome at the integration site of the C03 T-DNA insert.

PCR and Sanger sequencing confirmed that both T-DNA inserts (Cnn random and C03) contained all 13 intended gene expression cassettes. The Cnn random T-DNA insert had a 184-bp truncation of the 5΄ end of the right border and a 72-bp truncation of the 3΄ end of the left border. In addition, the first 64 bp in the right border of the Cnn random T-DNA insert was determined to be a rearrangement of short T-DNA right border-derived repeats. The C03 T-DNA insert had a 184-bp truncation of the 5΄ end of the right border and a 53-bp truncation of the 3΄ end of the left border.

The sequences of the T-DNA insert were determined to be identical to the vector reference sequence of LTM593 except for two single nucleotide changes in the Cnn random T-DNA insert and one nucleotide change in the C03 T-DNA insert. In the Cnn random insert, one cytosine to adenine nucleotide change was in the coding sequence of the delta-12 desaturase gene, c-D12D(Ps), which resulted in a phenylalanine to leucine amino acid substitution (F83L) in the D12D(Ps) protein. In addition, a cytosine to adenine nucleotide change was found in the promoter sequence p-PXR(Lu), which is part of an expression cassette containing the c-O3D(Pir) coding sequence. This nucleotide change does not result in an amino acid substitution. In the C03 insert, there was a guanine to thymine nucleotide change in the coding sequence of the delta-4 desaturase gene, c-D4D(Pl). This change resulted in an alanine to serine amino acid substitution (A102S) in the D4D(Pl) protein. These amino acid changes have no impact on the function or activity of the respective proteins.

Bioinformatics analyses based on the DNA sequence obtained for both T-DNA inserts via Sanger sequencing were performed to investigate the potential toxicity and allergenicity of potentially expressed proteins identified as open reading frames (ORFs) at the individual DNA junctions of the Cnn random and C03 T-DNA inserts. Potential ORFs were defined as any contiguous nucleic acid sequence that contains a string of 30 translated codons between two in-frame termination codons (i.e. TAA, TAG, or TGA) from any of the six potential reading frames

(three forward and three reverse reading frames). Potential ORFs were translated and evaluated for sequence identity matches to known toxins and allergens to identify any need for additional safety tests. None of the bioinformatics analyses revealed any basis to suspect that any of the putative proteins are toxigenic or allergenic.

The stability of the inserted DNA sequence in LBFLFK canola was characterized through Illumina-based sequencing. The junction sequences and sequencing read distributions over three LBFLFK canola generations (T3, T4, T5) were compared. Read depth distribution patterns across the entire T-DNA were uniform with similar peaks in read depth across the entire T-DNA sequence, and the same classes of unique junction (split-read cluster) sequences were observed in all three generations of LBFLFK canola tested. Based on the results of this whole genome sequencing (WGS) analysis, the stability of the inserted DNA sequence in LBFLFK canola was concluded.

The inheritance of the two LBFLFK canola T-DNA insertion loci was assessed in F2 and F3 generations using segregating F2 and F3 seed material derived from hemizygous parental plants. The zygosity of plants in the T3, F1, and F2 generations was assessed via real-time TaqMan^®end-point PCR assays. A Pearson's Chi-square (c2) analysis was used to statistically compare the observed segregation ratios of the LBFLFK canola inserts to the expected Mendelian ratios. The analysis confirmed that the two inserts were stably inherited according to Mendelian principles.

4. Product Information

LBFLFK canola differs from conventional canola through the introduction of the coding sequences for 10 genes (12 expression cassettes) that make up the biosynthesis pathway that converts OA into EPA and DHA. In addition, a gene encoding an acetohydroxy acid synthase (AHAS) protein from A thaliana with two mutations was introduced that conveys tolerance to the imidazolinone herbicides. In total the following 13 genes were introduced, which would be expected to result in the expression of 11 new proteins: c-D12D(Ps) encoding a delta-12 desaturase (Ps); c-D6D(Ot) encoding a delta-6 desaturase (Ot); c-D6E(Tp) encoding a delta-6 elongase (Tp); c-D6E(Pp) encoding a delta-6 elongase (Pp); c-D5D(Tc), delta-5 desaturase (Tc); c-O3D(Pi) encoding a omega-3 desaturase (Pi); c-O3D(Pir) encoding a omega-3 desaturase (Pir); c-D5E(Ot) encoding a delta-5 elongase (Ot); c-D4D(Tc) encoding a delta-4 desaturase (Tc); c-D4D(Pl) encoding a delta-4 desaturase (Pl); and c-AHAS(At) encoding an acetohydroxy acid synthase.

Field trials were initiated during the 2015 planting season to generate protein expression data for LBFLFK canola (T4) at the following canola growing locations in the United States of America: Geneva, MN; Sun River, MT; American Falls, ID; and Ephrata, WA. There was a single replication of each plot (LBFLFK unsprayed, LBFLFK sprayed with imidazolinone herbicide, and the control variety Kumily) at each site. Field sites were representative of canola producing regions suitable for commercial production. The sprayed LBFLFK plots were sprayed, at the 3-4 leaf stage with an herbicide containing the active ingredient imazamox, an imidazolinone. Immature seed samples were taken when seeds were green. For collection of mature seed, plants were swathed at approximately BBCH 85 and collected once moisture was estimated visually to be <10 percent. Protein quantification was achieved using either an enzyme-linked immunosorbent assay (ELISA) or a capillary-based quantitative western blot method. The choice of quantification method depended on the suitability based on the characteristics of each protein. Expression of AHAS(At) was quantified in plant tissues including seeds, whole plant (at rosette and flowering stages), leaf, root, and pollen using a quantitative western blot method.

ELISAs were used to determine the amounts of D12D(Ps), D6E(Pp), D5D(Tc), and D5E(Ot) present in the samples. Quantitative western blot methods were used to determine the amounts of D6D(Ot), D6E(Tp), O3D(Pir), O3D(Pi), D4D(Pl), D4D(Tc), and AHAS(At) [A122TS653N] present in the samples. The tissues analyzed were whole plants at different maturity stages, leaf tissue, root tissue, immature seed, mature seed, and pollen.

Eight of the ten proteins in the EPA and DHA biosynthesis pathway were detected in immature and/or mature seeds of LBFLFK canola. Expression of each protein is expected to be highest in the seeds because expression is driven by a seed-specific promoter. BASF stated that other tissue types were tested, and none of the proteins involved in the EPA/DHA synthesis pathway were detected. Two proteins, D6E(Pp) and O3D(Pi) could not be detected in either immature or mature seeds. The D6E(Pp), D5D(Tc), O3D(Pi), and D5E(Ot) proteins could not be quantified in immature seeds. The D6E(Pp) and O3D(Pi) proteins could not be quantified in mature seeds. The µg of protein per g of LBFLFK canola seed was calculated on a fresh weight and dry weight (DW) basis. The lowest protein level quantified in dried mature seed was 0.93 µg/g of the D12D(Ps), and the highest protein level quantified was 936.43 µg/g of the D6E(Tp) protein. AHAS(At) was quantifiable in all tissue types tested except mature seed.

The level of expression for the integral membrane proteins was below the limit of quantitation (< LOQ) in all tissues analyzed except for seed. The newly expressed AHAS(At) [A122TS653N] protein, driven by a constitutive promoter, was quantifiable in every tissue except mature seed. Canola seeds are the most likely tissue to enter the food supply, usually after processing into oil and defatted meal fractions. Levels of protein in the mature seed were as follows: average D12D(Ps) 0.93 µg/g DW, average D6E(Pp) below the limit of detection (LOD) for the assay (1.04 µg/g DW), average D5D(Tc) 1.33 µg/g DW, average D5E(Ot) 15.48 µg/g DW, average D6D(Ot) 40.22 µg/g DW, average D6E(Tp) 936.43 µg/g DW, average O3D(Pir) 504.38 µg/g DW, average O3D(Pi) below the LOD for the assay (27.08 µg/g DW), average D4D(Pl) 4.16 µg/g DW, average D4D(Tc) 11.81 µg/g DW, and average AHAS(At) [A122TS653N] below the limit of quantitation (LOQ) for the assay (3.05 µg/g DW).

The P. patens delta-6 elongase gene encodes a 46 kDa protein, the Thraustochytrium sp. delta-5 desaturase gene encodes a 50 kDa protein, the O. tauri delta-6 desaturase gene encodes a 34 kDa

protein, the T. pseudonana delta-6 elongase gene encodes a 32 kDa protein, the P. sojaedelta-12 desaturase gene encodes a 46 kDa protein, the P. infestans omega-3 desaturase gene encodes a 41 kDa protein, the Thraustochytrium sp. delta-4 desaturase gene encodes a 59 kDa protein, the P. irregular omega-3 desaturase gene encodes a 40 kDa protein, the P. lutheri delta-4 desaturase gene encodes a 49 kDa protein, the O. tauri delta-5 elongase gene encodes a 34 kDa protein, and the A. thaliana acetohydroxy acid synthase gene encodes a 73 kDa polypeptide.

Enzyme functionality and specificity of the three elongases and seven desaturases were demonstrated using yeast strains individually expressing each of the proteins. In vivo feeding studies were performed with 14 potential fatty acid intermediates in the fatty acid pathway to allow an assessment of specificity of each expressed enzyme. Membranes isolated from these yeast expression strains were used to conduct in vitro assays to assess the backbone specificity of each of the elongases and desaturases. Mass spectrometry was used to perform tryptic peptide mapping analysis against the deduced amino acid sequence of the proteins to confirm the identity of the proteins isolated from LBFLFK canola.

Western blot analysis showed that the apparent molecular weights of D12D(Ps), D12D(Ps) [F83L], D6D(Ot), D6E(Tp), D5D(Tc), O3D(Pir), D5E(Ot), D4D(Tc), D4D(Pl), D4D(Pl) [A102S], and AHAS(At) [A122TS653N] proteins produced in LBFLFK canola were consistent with the predicted molecular weights of 45.6 kDa, 45.5 kDa, 51.7 kDa, 31.8 kDa, 49.8 kDa, 40.4 kDa, 34.2 kDa, 59.0 kDa, 49.1 kDa, 49.1 kDa, and 66.1 kDa, respectively. An immunoreactive band near the calculated mass of the D6E(Pp) and the O3D(Pi) proteins was not observed in membrane-enriched samples derived from immature seeds of LBFLFK. A glycosylation analysis demonstrated the absence of post-translational glycosylation of all novel proteins produced in LBFLFK canola with two exceptions. The O3D(Pi)-specific and D6E(Pp)-specific antibodies could not detect a protein in agreement with the calculated molecular mass of the O3D(Pi) protein; therefore, the glycosylation status of this protein could not be determined.

5. Dietary Exposure

LBFLFK canola will provide an alternate source of EPA and DHA for existing markets. According to the Petitioner, EPA and DHA from LBFLFK canola will be offered in the form of refined oil for use as a food ingredient. The dietary exposure to EPA and DHA through the consumption of this oil was considered as part of the Nutritional Assessment.

6. Nutrition

Compositional data for LBFLFK canola, its parental variety Kumily, and six commercial canola reference varieties were collected from twelve field trials conducted across two growing seasons

(Winter 2014/2015 and Spring 2015) in the United States. In each trial, four replicates of each entry were planted in a randomized complete block design. Plots considered for the compositional comparisons were treated with imidazolinone and standard non-selective herbicides.

Seed samples were harvested and analyzed for proximates and fibres, fatty acids, amino acids, minerals, vitamins, phytosterols, and anti-nutrients (glucosinolates, phytic acid, phenolics). These compositional components are in line with recommendations listed in the Organisation for Economic Co-operation and Development (OECD) consensus document on the compositional considerations for new varieties of low erucic acid rapeseed (canola). The analyses for each component were conducted using internationally approved and validated analytical methods and following consistent and appropriate sample storage and preparation procedures.

The data collected from the field trials were analyzed using acceptable, statistical methodology. The data were summarized as the mean, minimum, maximum, and standard error. Where a statistically significant difference (p-value < 0.05) was identified versus the parental variety, further context for interpreting the possible nutritional relevance of the difference was gathered through comparisons with the expected range for conventional canola as defined by means reported by the Petitioner for the reference varieties, the International Life Sciences Institute Crop Composition Database, and the OECD consensus document.

In LBFLFK canola, a number of fatty acid desaturases and elongases were introduced to facilitate conversion of OA to EPA and DHA. As expected, the fatty acid profile was impacted as described in paragraphs 31 to 36.

Statistically significant differences compared to the parental variety were consistently observed across both seasons for the following fatty acids (Winter / Spring control means vs. Winter / Spring LBFLFK means expressed as % total fatty acids): C16:1n-7 (0.34 / 0.38 vs. 0.22 / 0.23), C18:0 (1.99 / 2.33 vs. 2.62 / 3.06), C18:3n-3 (8.00 / 8.99 vs. 5.11 / 5.92), C20:0 (0.66 / 0.70 vs. 0.52 / 0.66), C22:0 (0.37 / 0.41 vs. 0.26 / 0.30), C24:0 (0.20 / 0.24 vs. 0.093 / 0.15), and C24:1n-9 (0.15 / 0.19 vs. 0.09 / 0.11). Since the modified cultivar was never consistently outside the expected range for conventional canola, these differences were not considered to be a nutritional safety concern.

A statistically significant difference such that the modified cultivar was consistently outside the expected range across both seasons was observed for total trans fatty acids (Winter / Spring control means vs. Winter / Spring LBFLFK means expressed as % total fatty acids): 0.097 / 0.062 vs. 0.35 / 0.27. However, supplemental data indicated that the total trans fatty acid content of refined oil derived from the modified cultivar (0.594%) was comparable to oil derived from the parental variety (0.626%), and was within range of marine oils that are sources of EPA and DHA (at least 0.12-2.76%). Furthermore, the total trans fatty acid content of LBFLFK oil was well below that found in commonly consumed canola (2.314%) and soybean (1.66%) oils according to the Canadian Nutrient File. Therefore, this difference was not considered to be a nutritional safety concern as the primary exposure to trans fatty acids in the modified cultivar was expected to be through the refined oil.

A statistically significant difference such that the modified cultivar was consistently outside the expected range across both seasons, was observed for linoleic acid (LA; C18:2n-6) (Winter / Spring control means vs. Winter / Spring LBFLFK means expressed as % total fatty acids): 19.27 / 18.73 vs. 29.52 / 29.61. Linoleic acid (LA) is an essential nutrient for which an Adequate Intake level has been set. Since the LA content of the modified cultivar was consistently higher than the control, and an Upper Tolerable Intake Level for LA has not been established, this difference was not considered to be a nutritional concern.

Statistically significant differences were observed such that LBFLFK canola was consistently outside the expected range across both seasons for the following cis-monounsaturated fatty acids (Winter / Spring control means vs. Winter / Spring LBFLFK means expressed as % total fatty acids): C18:1n-9 (54.86 / 56.66 vs. 23.52 / 26.18), and C20:1n-9 (0.98 / 1.07 vs. 0.59 / 0.68). Monounsaturated fatty acids can be biosynthesized from other fuel sources and are, therefore, not essential in the diet. Therefore, these differences were not considered to be a nutritional safety concern.

A number of cis-polyunsaturated fatty acids, apart from EPA and DHA, were detected only in the modified cultivar (C18:2n-9, C18:3n-6, C18:4n-3, C20:2n-9, C20:3n-3, C20:3n-9, C20:3n-6, C20:4n-3, C20:4n-6, C22:4n-3, C22:4n-6, C22:5n-3, and C22:5n-6). The Petitioner provided evidence that these non-essential fatty acids are present in other organisms and foods that are safely and routinely consumed. Therefore, their presence in LBFLFK canola was not considered to be a nutritional safety concern.

EPA (C20:5n-3) and DHA (C22:6n-3) were detected in LBFLFK canola, but were not detected in the parental control or reference varieties. The average EPA content (% total fatty acids) of LBFLFK seeds from trials conducted during the Winter and Spring growing seasons was 7.21% and 6.27%, respectively. The average DHA content (% total fatty acids) of LBFLFK seeds from trials conducted during the Winter and Spring growing seasons was 1.02% and 0.77%, respectively. Supplemental data provided by the Petitioner, indicated that the combined EPA and DHA content of refined oil derived from the modified cultivar was 4.5% of total fatty acids. The Food Directorate (FD) draft document entitled "Re-evaluation of the FD position on reference intake for addition of oils rich in EPA/DHA to foods as part of the Novel Foods notification process (Division 28)" proposes reference intake levels for use in exposure modelling scenarios for oils containing EPA or DHA. These reference intakes represent upper limits above which adverse health effects associated with usual, combined intake of DHA or EPA are potentially possible. For oils containing both EPA and DHA, the reference intake is 3 g/day. It was determined that the nutritional safety of EPA and DHA found in the modified cultivar should be evaluated in relation to this reference intake based on exposure to the refined oil.

The Petitioner estimated the impact of replacing conventional canola oil in foods with LBFLFK oil using comprehensive food consumption data collected as part of the National Health and Nutrition Examination Survey in the United States. Overall, the combined, estimated intake of EPA and DHA following the replacement of canola oil with the LBFLFK oil was ≤ 0.18 g/day at the mean and ≤ 0.43 at the 90th percentile, respectively, across all population groups and genders. In addition, the Petitioner considered another scenario in which LBFLFK oil replaced fish (menhaden) oil in foods. Since the combined EPA and DHA content of menhaden oil is higher than LBFLFK oil, direct substitution of menhaden oil with LBFLFK oil can only result in less exposure to EPA and DHA. Based on these scenarios, food use of LBFLFK canola is not expected to result in usual, combined intake of EPA and DHA above the FD reference level of 3 g/d. Therefore, the presence of EPA and DHA in LBFLFK canola was not considered to be a nutritional safety concern.

The Petitioner indicated that the stereochemical attachment of EPA and DHA to triacylglycerol in LBFLFK canola may differ from that of some conventional sources, but is likely to fall within the expected range for conventional sources. In addition, there is currently insufficient scientific evidence on which to conclude that stereochemical attachment to triacylglycerol has an effect on the bioavailability of EPA and DHA. Therefore, this potential difference was not considered to be a nutritional safety concern.

Statistically significant differences were consistently observed across both seasons for the following non-lipid components (Winter / Spring control means vs. Winter / Spring LBFLFK means, units): neutral detergent fibre (18.40 / 15.26 vs. 16.65 / 14.55, % DW), vitamin K1 (0.088 / 0.11 vs. 0.097 / 0.12, mg/100 g DW), calcium (0.33 / 0.32 vs. 0.30 / 0.29, % DW), magnesium (0.33 / 0.34 vs. 0.31 / 0.32, % DW), glucobrassicin (0.34 / 0.31 vs. 0.96 / 0.67, μmol/g DW), gluconapin (1.40 / 2.30 vs. 1.72 / 2.45, μmol/g DW), sinapine (1.00 / 1.02 vs. 0.89 / 0.95, % DW), total phytosterols (0.93 / 0.89 vs. 0.88 / 0.75, % DW). Since the modified cultivar was never consistently outside the expected range for conventional canola, these differences were not considered to be a nutritional safety concern.

Based on the information provided, the Nutrition Pre-market Assessment Division (NPAD) has not identified any safety concerns regarding the food use of oil derived from LBFLFK canola from a nutritional perspective.

7. Toxicology

The Pre-Market Toxicology Assessment Section (PTAS) evaluated the safety of the LBFLFK canola by assessing the potential toxicity of the finished product, which is the highly refined canola seed oil.

No history of safe food exposure rationale could be supplied for the source organisms P. sojae, O. tauri, T. pseudonana, P. patens, Thraustochytrium sp., P. irregular, P. infestans and P. lutheri.

The level of expression for each protein was determined in the oil seed of LBFLFK canola using enzyme-linked immunosorbent assay (ELISA) or quantitative western blot. Levels of protein expression for D12D(Ps), D6D(Ot), D6E(Tp), D5D(Tc), O3D(Pir), D5E(Ot), D4D(Tc), D4D(Pl) and AHAS ranged from 1 - 936 μg/g DW in the developing and mature seeds. Two of the transgenic proteins, D6E(Pp) and O3D(Pi), were reported to be below LOD and LOQ in the LBFLFK canola seed (LOQ = 6 - 71 μg/g DW).

Novel expression of AHAS protein (containing the A122T and S653N mutations) derived from A. thaliana has been previously approved by Health Canada in various crop varieties intended for human consumption. The safety of the AHAS(At) protein is well established and does not pose a toxicological concern.

The predicted amino acid sequences of all transgenes were compared to sequences of known toxins retrieved from the National Center for Biotechnology Information (NCBI) GenBank non-redundant peptide sequence database (search date May 3, 2017; 121,686,747 sequences). An E-score cut off of ≤ 1 was considered positive against known toxins and antinutrient sequences.

The O3D(Pir) and AHAS proteins showed positive alignments with non-toxic proteins, omega-3 fatty acid desaturase and acetolactate synthase 3, respectively. These proteins are expressed in Ricinus communis (castor bean), which also produces the toxin ricin. However, these proteins are not involved in ricin biosynthesis, but are well characterized and are involved in fatty acid and branched chain amino acid synthesis, respectively. The Petitioner and Evaluator concluded that the newly expressed proteins in LBFLFK canola did not share sequence similarity with known oral toxins.

The Petitioner stated that, for food use, the LBFLFK canola will only be used for the production of highly refined canola oil. The Petitioner provided data showing that the EPA+DHA canola oil, much like other highly processed vegetable oils, is devoid of all proteins, including the 11 transgenic proteins assessed in this petition.

PTAS concluded that the 11 transgenic proteins expressed in LBFLFK canola would not be expected to pose any toxicological concerns to consumers, because they are absent from the final food product, a highly processed refined oil. Should the Petitioner request additional food uses for LBFLFK canola that contains its proteins, further assessment would be required to address the data gaps that exist in the toxicological database for these proteins.

Based on the information provided, PTAS has not identified any safety concerns regarding the food use of oil derived from LBFLFK canola from a toxicological perspective.

8. Allergenicity

The Pre-Market Toxicology Assessment Section (PTAS) evaluated the safety of the LBFLFK canola by assessing the potential allergenicity of the finished product, which is the highly refined canola seed oil.

Novel expression of AHAS (containing the A122T and S653N mutations) protein derived from A. thalianahas been previously approved by Health Canada in various crop varieties intended for human consumption. The allergenic safety of A. thaliana AHAS protein is well established and does not pose an allergenic risk.

The predicted amino acid sequences of all transgenes were compared to sequences of known allergens retrieved from the Food Allergy Research and Resource Program (FARRP) Allergen Protein database (2017; 2,035 sequences). The proteins did not share ≥ 35% amino acid identity with any known allergen or contain potential allergen epitopes (an exact sequence match over 8 amino acids). Based on the results of the bioinformatics analysis, the Petitioner and Evaluator concluded that the newly expressed proteins in LBFLFK canola did not share sequence similarity with known oral allergens.

The Petitioner demonstrated that the D12D(Ps), D6D(Ot), D6E(Tp), D5D(Tc), O3D(Pir), D5E(Ot), D4D(Tc) and D4D(Pl) proteins present in LBFLFK crude seed protein extracts were denatured when incubated at temperatures of 50 to 90 ºC for 20 minutes (as determined by enzyme activity assay and western blot). The assay demonstrated that these proteins were structurally unstable at high temperatures. The processing and refining of canola seed into oil involves similarly high temperatures (between 80 to 105 °C for 15 to 20 minutes). It is expected that the denatured proteins will be more susceptible to digestion in the gastrointestinal tract.

The Petitioner demonstrated that the D12D(Ps), D6D(Ot), D5D(Tc), O3D(Pir), D5E(Ot), D4D(Tc), D4D(Pl) proteins present in LBFLFK crude seed protein extracts were digested to completion following incubation with simulated gastric fluid (10 U pepsin/μg protein; pH 1.2) or SIF (~10 mg/ml pancreatin; pH 7.5) for 2 minutes, as determined by western blot. D6E(Tp) was shown to be susceptible to sequential digestion with SGF (30 minutes) followed by SIF (30 seconds). As such, the D12D(Ps), D6D(Ot), D6E(Tp), D5D(Tc), O3D(Pir), D5E(Ot), D4D(Tc) and D4D(Pl) proteins are expected to be readily digested under the conditions normally found in the human gastrointestinal tract.

Two of the transgenic proteins, D6E(Pp) and O3D(Pi), were reported to be below levels of detection and quantitation in the LBFLFK canola seed. Further, the Petitioner was unable to isolate sufficient quantities from microbial expression systems due to the intractable nature of the integral membrane proteins. At these extremely low levels of expression, it is unlikely that toxicologically relevant quantities of D6E(Pp) and O3D(Pi) would be present in the highly refined canola oil.

The Petitioner stated that the LBFLFK canola will only be used for the production of highly refined canola oil for food use. LBFLFK canola oil was shown, like other highly processed vegetable oils, to be devoid of all proteins, including the eleven transgenic proteins assessed in this petition.

PTAS concluded that the 11 transgenic proteins expressed in LBFLFK canola would not pose an allergenic safety concerns to consumers as they will be absent from the final highly refined oil.

Based on the information provided, PTAS has not identified any safety concerns regarding the food use of oil derived from LBFLFK canola from an allergenicity perspective.

Conclusion:

Health Canada's review of the information presented in support of the food use of highly refined oil derived from LBFLFK canola does not raise concerns related to food safety. Health Canada is of the opinion that highly refined oil derived from this canola variety is as safe and nutritious as highly refined oil derived from current commercial canola varieties.

As there is a significant change to the nutritional composition of the highly refined oil derived from LBFLFK canola, special labelling is required for this product. The Petitioner was advised to contact the Canadian Food Inspection Agency (CFIA) for the Agency to review the Petitioner's proposed common name and labelling for this product for purposes of sale.

Health Canada's opinion deals only with the food use of canola event LBFLFK. Issues related to its use as animal feed have been addressed separately through existing regulatory processes in the Canadian Food Inspection Agency (CFIA). 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 canola event LBFLFK destined for commercial sale.

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.

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
hc.bmh-bdm.sc@canada.ca