Decision Document DD2019-125
Determination of the Safety of BASF Canada Inc.'s Canola (Brassica napus L.) Event LBFLFK

This Decision Document has been prepared to explain the regulatory decisions reached under Directive 94-08 (Dir94-08) – Assessment Criteria for Determining Environmental Safety of Plants with Novel Traits, its companion document BIO2017-03 – The Biology of Brassica napus L. (Canola/Rapeseed) and Section 2.6 – Guidelines for the Assessment of Novel Feeds: Plant Sources, of Chapter 2 of the RG-1 Regulatory Guidance: Feed Registration Procedures and Labelling Standards.

The Canadian Food Inspection Agency (CFIA) – specifically the Plant Biosafety Office of the Plant Health and Biosecurity Directorate, the Plant and Biotechnology Risk Assessment Unit of the Plant Health Science Directorate and the Animal Feed Program (AFP) of the Animal Health Directorate – has evaluated information submitted by BASF Canada Inc. This information concerns canola event LBFLFK that synthesizes omega-3 long-chain polyunsaturated fatty acids (LC-PUFAs) and has tolerance to imidazolinone herbicides.

The CFIA has determined that canola event LBFLFK does not present altered risk when compared to canola varieties currently grown and permitted to be used as livestock feed in Canada, subject to the conditions described in Section 7 – Regulatory Decision.

Regarding the feed authorization, the CFIA has authorized the use of canola event LBFLFK oil as a source of omega-3 LC-PUFAs for fish feeds only. Furthermore, only pre-pressed solvent extracted defatted canola meal has been authorized for use as livestock feed. Conditions on the feed authorization are described in Section 7 – Regulatory Decision of this decision document. Given that the canola oil for this event was approved for fish feed only, a new definition for omega-3 long chain polyunsaturated fatty acids canola oil was added to Schedule IV of the Feeds Regulations.

Taking into account the CFIA's environment and livestock feed evaluations, unconfined release into the environment of canola event LBFLFK is therefore authorized by the Plant Biosafety Office of the Plant Health and Biosecurity Directorate as of December 9, 2019. Likewise, livestock feed use is authorized, with the specific use restrictions stated above, by the Animal Feed Program of the Animal Health Directorate as of December 9, 2019.

Any canola lines derived from canola event LBFLFK may also be released into the environment and used as livestock feed, provided that:

• no inter-specific crosses are performed
• the intended uses are similar and meet the conditions of the authorization
• it is known based on characterization that these plants are substantially equivalent to the authorized line, and do not display any additional novel traits, and
• the novel genes are expressed at levels similar to those of the authorized line

With respect to its unconfined release into the environment, an appropriate herbicide tolerance management plan should be implemented. If canola event LBFLFK is cultivated in Canada as an individual event or in combination with other canola events in stacked/pyramided products, BASF Canada Inc. must submit a herbicide tolerance management plan to the CFIA.

Canola event LBFLFK is subject to the same phytosanitary import requirements as unmodified canola varieties. Canola event LBFLFK is required to meet the requirements of other jurisdictions, including but not limited to, the Food & Drugs Act and the Pest Control Products Act.

Please note that the livestock feed and environmental assessments of novel feeds and plants with novel traits are critical steps in the potential commercialization of these plant types. Other requirements, such as the assessment of novel foods by Health Canada, have been addressed separately from this review.

December 9, 2019

This bulletin was created by the Canadian Food Inspection Agency. For further information, please contact the Plant Biosafety Office or the Animal Feed Program by visiting the contact page.

On this page

1. 1. Brief identification of the modified plant
2. 2. Background information
3. 3. Description of the novel traits
   1. 3.1 Development method
   2. 3.2 Synthesis of omega-3 long-chain polyunsaturated fatty acids
   3. 3.3 Tolerance to imidazolinone herbicide
   4. 3.4 Stable integration into the plant genome
4. 4. Criteria for the environmental assessment
   1. 4.1 Potential for canola event LBFLFK to become a weed of agriculture or be invasive of natural habitats
   2. 4.2 Potential for gene flow from canola event LBFLFK to sexually compatible plants whose hybrid offspring may become more weedy or more invasive
   3. 4.3 Potential for canola event LBFLFK to become a plant pest
   4. 4.4 Potential impact of canola event LBFLFK or its gene products on non-target organisms, including humans
   5. 4.5 Potential impact of canola event LBFLFK on biodiversity
5. 5. Criteria for the livestock feed assessment
   1. 5.1 Potential impact of canola event LBFLFK on livestock nutrition
   2. 5.2 Potential impact of canola event LBFLFK on animal health and human safety as it relates to the potential transfer of residues into foods of animal origin and worker/bystander exposure to the feed
6. 6. New information requirements
7. 7. Regulatory decision

1. Brief identification of the modified plant

Designation of the modified plant: Canola Event LBFLFK (OECD unique identifier BPS-BFLFK-2)

Applicant: BASF Canada Inc.

Plant species: Canola (Brassica napus L.)

Novel traits: Synthesis of omega-3 long-chain polyunsaturated fatty acids; tolerance to imidazolinone herbicides.

Trait introduction method: Agrobacterium-mediated transformation

Intended use of the modified plant: Canola event LBFLFK is intended for cultivation within the United States and processing either in the United States or Canada. BASF intends to implement an Identity Preservation (IdP) System at every step of production and handling. Canola event LBFLFK oil is intended for use in human food and fish feeds only, as a source of omega-3 long-chain polyunsaturated fatty acids. Only pre-pressed solvent-extracted defatted canola meal is intended for use as livestock feed. No authorization has been requested or granted for use of whole grain and forage derived from the canola event LBFLFK as livestock feed at this time.

2. Background information

BASF Canada Inc. has developed a canola event (LBFLFK) that produces omega-3 long-chain polyunsaturated fatty acids (LC-PUFAs), including eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) from oleic acid, and that is tolerant of imidazolinone herbicides.

Canola event LBFLFK was developed by BASF Canada Inc. using recombinant deoxyribonucleic acid (DNA) technology. The inserted DNA includes 13 gene expression cassettes that encode 10 fatty acid desaturase and elongase proteins (2 of the 10 desaturases are each encoded by 2 expression cassettes), as well as 1 modified acetohydroxyacid synthase (AHAS) enzyme.

The 10 desaturase and elongase proteins are as follows:

• delta-12 desaturase from Phytophthora sojae (D12D(Ps))
• delta-6 desaturase from Ostreococcus tauri (D6D(Ot))
• delta-6 elongase from Thalassiosira pseudonana (D6E(Tp))
• delta-6 elongase from Physcomitrella patens (D6E(Pp))
• delta-5 desaturase from Thraustochytrium sp. (D5D(Tc))
• omega-3 desaturase from Pythium irregulare (O3D(Pir))
• omega-3 desaturase from Phytophthora infestans (O3D(Pi))
• delta-5 elongase from Ostreococcus tauri (D5E(Ot))
• delta-4 desaturase from Thraustochytrium sp. (D4D(Tc))
• delta-4 desaturase from Pavlova lutheri (D4D(Pl))

The herbicide tolerance trait in canola event LBFLFK is conferred through the introduction of 1 ahas gene from Arabidopsis thaliana. The mutated ahas(At) gene contains 2 single nucleotide substitutions and encodes a modified AHAS(At) protein with 2 amino acid substitutions, an alanine to threonine at position 122 and serine to asparagine at position 653. The 2 amino acid substitutions prevent binding of imidazolinone herbicides to the modified AHAS protein, and therefore confer tolerance to these herbicides.

BASF Canada Inc. provided information on the identity of canola event LBFLFK, and a detailed description of the introduced genetic elements and new proteins encoded by these genetic elements. BASF Canada Inc. also provided information about how canola event LBFLFK compares to other canola in terms of its agronomic characteristics and environmental safety, and its nutrition and safety as an animal feed.

The Plant and Biotechnology Risk Assessment (PBRA) Unit of the Plant Health Science Directorate, CFIA, has reviewed the above information, in light of the assessment criteria for determining environmental safety of plants with novel traits, as described Directive 94-08 (Dir94-08) – Assessment Criteria for Determining Environmental Safety of Plants with Novel Traits. The PBRA Unit has considered:

• the potential for canola event LBFLFK to become a weed of agriculture or to be invasive of natural habitats;
• the potential for gene flow from canola event LBFLFK to sexually compatible plants whose hybrid offspring may become more weedy or more invasive;
• the potential for canola event LBFLFK to become a plant pest;
• the potential impact of canola event LBFLFK and their gene products on non-target organisms, including humans; and
• the potential impact of canola event LBFLFK on biodiversity.

The Animal Feed Program (AFP) of the CFIA has also reviewed the above information with respect to the assessment criteria for determining the safety and nutrition of livestock feed, as described in Section 2.6 – Guidelines for the Assessment of Novel Feeds: Plant Sources, of Chapter 2 of the RG-1 Regulatory Guidance: Feed Registration Procedures and Labelling Standards.

The AFP has considered both intended and unintended effects and similarities and differences between canola event LBFLFK, and unmodified parental canola variety relative to the safety and efficacy of feed ingredients derived from canola event LBFLFK, for its intended purpose, including:

• the potential impact of canola event LBFLFK on livestock nutrition; and
• the potential impact of canola event LBFLFK on animal health and human safety, as it relates to the potential transfer of residues into foods of animal origin and worker/bystander exposure to the feed.

The AFP has also considered whether feeds derived from canola event LBFLFK meet the definitions and requirements of feeds as listed in Schedule IV of the Feeds Regulations.

3. Description of the novel traits

3.1 Development method

Canola event LBFLFK was developed through Agrobacterium rhizogenes-mediated transformation of Brassica napus cv. Kumily hypocotyl segments with a plasmid vector. The vector included a T-DNA region containing 13 expression cassettes encoding fatty acid desaturases and elongases, to introduce a new omega-3 LC-PUFA synthesis pathway in the canola seed. As well, tolerance to imidazolinone herbicides was conferred by 1 expression cassette encoding a modified Arabidopsis thaliana acetohydroxy acid synthase (AHAS) protein. Transformed explants were selected on the basis of tolerance to imazethapyr (an imidazolinone herbicide) and regenerated to produce plants. Canola event LBFLFK was identified as a successful transformant based on molecular analyses, fatty acid profiles, herbicide tolerance efficacy and agronomic evaluations.

3.2 Synthesis of omega-3 long-chain polyunsaturated fatty acids

Canola event LBFLFK was developed to synthesize omega-3 LC-PUFAs, including eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). There were 10 integral membrane proteins (7 desaturases and 3 elongases) introduced into canola event LBFLFK. Unmodified canola produces primarily oleic and linolenic fatty acids in seeds through the combined efforts of enzymes involved in de novo fatty acid synthesis, elongation, and desaturation. Expression of the omega-3 LC-PUFA synthesis genes in canola event LBFLFK allows production of DHA and its biosynthetic intermediate EPA from these endogenous fatty acids.

The genes used for omega-3 LC-PUFA synthesis in canola event LBFLFK were chemically synthesized based on the sequences 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 was unchanged from the source organism, with the exception of the delta-6 elongase, D6E(Tp), sequence that has a proline to serine amino acid substitution at position 196 (P196S). This substitution does not occur in any known conserved domains responsible for the functionality of the D6E(Tp) protein and is therefore not anticipated to alter the enzymatic activity of the protein.

Two T-DNA inserts are present in canola event LBFLFK (see 3.4 Stable integration in the plant genome). The sequences of the T-DNA inserts in canola event LBFLFK were found to contain 2 single nucleotide changes compared to the expected sequences, resulting in 2 amino acid substitutions. In one copy of the gene encoding the D12D(Ps) protein, a single nucleotide change led to a phenylalanine-to-leucine amino acid substitution at position 83 (F83L). In one copy of the gene encoding the D4D(Pl) protein, a single nucleotide change led to an alanine-to-serine amino acid substitution at position 102 (A102S). These amino acid changes do not impact the enzymatic function or activity of these respective proteins.

Expression of omega-3 LC-PUFA synthesis proteins in canola event LBFLFK is driven by seed-specific promoters. Samples of canola tissues were collected from plants in 4 field trials in the US. Tissues were collected from unsprayed plants and plants sprayed with a herbicide containing the active ingredient imazamox. These treatments represent the growing conditions likely to be seen in crop production. The tissues analyzed were whole plants at the rosette (BBCH 16-51) and flowering (BBCH 64-65) stages, leaf tissue, root tissue, immature seed (BBCH 75-79), mature seed (BBCH 99), and pollen (BBCH 61-63). ELISAs were used to determine the amounts of D12D(Ps), D6E(Pp), D5D(Tc), and D5E(Ot) proteins present in the tissue samples, whereas quantitative western blot methods were used to determine the amounts of D6D(Ot), D6E(Tp), O3D(Pir), O3D(Pi), D4D(Pl), D4D(Tc) proteins present in the tissue samples.

Eight of the 10 proteins in the introduced omega-3 LC-PUFA synthesis pathway were detected in immature and/or mature seeds of LBFLFK canola. In all other tissues analyzed, the level of expression for all introduced omega-3 LC-PUFA synthesis proteins was below the limit of quantification (< LOQ). 2 proteins, D6E(Pp) and O3D(Pi) could not be detected in either immature or mature seeds.

The average expression of the omega-3 LC-PUFA synthesis proteins, in micrograms per gram dry weight (μg/g dw) in mature and immature seeds from unsprayed plants, were as follows:

The expression levels of the above proteins in the tissues of herbicide-treated plants were similar to those from unsprayed plants.

The potential allergenicity and toxicity of the introduced omega-3 LC-PUFA synthesis proteins was evaluated. The weight of evidence indicates that these proteins are unlikely to be allergenic, based on the following information:

• the sources of the introduced omega-3 LC-PUFA synthesis genes (Phytophthora sojae, Ostreococcus tauri, Thalassiosira pseudonana, Physcomitrella patens, Thraustochytrium sp., Pythium irregulare, Phytophthora infestans, and Pavlova lutheri) are not commonly associated with allergenicity;
• bioinformatic evaluations of the amino acid sequences of the introduced omega-3 LC-PUFA synthesis proteins confirmed the lack of relevant similarities to known allergens;
• unlike many allergenic proteins, which tend to resist digestion, 8 of the 10 proteins in the canola event LBFLFK EPA and DHA synthesis pathway (D12D(Ps), D6D(Ot), D6E(Tp), D5D(Tc), O3D(Pir), D5E(Ot), D4D(Tc) and D4D(Pl)) were shown experimentally to be rapidly degraded in simulated gastric fluid and not to be heat stable; and
• unlike many allergens, the same 8 proteins in canola event LBFLFK were shown experimentally to be unglycosylated.

Taken together, this information implies that these 8 proteins are unlikely to be allergenic. The 2 remaining introduced proteins, D6E(Pp) and O3D(Pi), could not be detected in canola event LBFLFK with O3D(Pi)-specific and D6E(Pp)-specific antibodies and therefore the exposure of livestock or non-target organisms to these proteins would be very low.

It was also concluded that the omega-3 LC-PUFA synthesis proteins introduced in canola event LBFLFK are unlikely to be toxic to livestock and non-target organisms because:

• they lack a mode of action to suggest that they are intrinsically toxic to livestock or non-target organisms
• the amino acid sequences of the omega-3 LC-PUFA synthesis proteins lack relevant similarities to known toxins.

For a more detailed discussion of the potential allergenicity and toxicity of the introduced proteins involved in the synthesis of omega-3 LC-PUFAs in canola event LBFLFK, see Section 5.2: Potential impact of canola event LBFLFK on animal health and human safety as it relates to the potential transfer of residues into foods of animal origin and worker/bystander exposure to the feed.

3.3 Tolerance to imidazolinone herbicide

AHAS is an enzyme found in bacteria, certain other micro-organisms and plants. This enzyme catalyses the first step in the biosynthesis of the branched chain amino acids valine, leucine and isoleucine. Imidazolinone herbicides block the normal function of AHAS proteins, resulting in a lethal decrease in protein synthesis.

Canola event LBFLFK was developed to be tolerant to imidazolinone herbicides by introduction of a modified AHAS protein from Arabidopsis thaliana. The modified AHAS(At) enzyme contains 2 amino acid substitutions, an alanine to threonine substitution at position 122 and a serine to asparagine substitution at position 653. These mutations render the enzyme insensitive to imidazolinone herbicides.

The protein expression of the modified AHAS(At) in canola event LBFLFK is driven by a constitutive promoter. Samples of canola tissues were collected from plants in 4 field trials in the US. Tissues were collected from unsprayed plants and plants sprayed with a herbicide containing the active ingredient imazamox. The average protein expression of the modified AHAS(At) in micrograms per gram fresh weight (μg/g fw) from unsprayed plants, as evaluated by a quantitative western blot method, ranged from below the limit of detection in mature seed to 30.63 μg/g fw in pollen.

The potential allergenicity and toxicity of the modified AHAS(At) protein to livestock and non-target organisms were evaluated. The weight of evidence indicates that the modified AHAS(At) protein is unlikely to be allergenic, based on the following information:

• the source of the ahas gene, Arabidopsis thaliana, is not commonly associated with allergenicity;
• the modified AHAS protein amino acid sequence lacks relevant similarities to known allergens;
• unlike many allergens, the modified AHAS(At) protein in canola event LBFLFK was shown experimentally to be rapidly degraded in simulated gastric fluid and not to be heat stable; and
• unlike many allergens, the modified AHAS(At) protein in canola event LBFLFK was shown experimentally to be unglycosylated.

It was also concluded that the modified AHAS(At) protein is unlikely to be toxic to livestock and non-target organisms because:

• it lacks a mode of action to suggest that it is intrinsically toxic to livestock or non-target organisms
• the AHAS(At) protein amino acid sequence lacks relevant similarities to known toxins.

For a more detailed discussion of the potential allergenicity and toxicity of the modified AHAS(At) protein, see Section 5.2: Potential impact of canola event LBFLFK on animal health and human safety as it relates to the potential transfer of residues into foods of animal origin and worker/bystander exposure to the feed.

3.4 Stable integration into the plant genome

Molecular characterization by Illumina-based sequencing, including analyses of junction sequences and sequencing read distributions, demonstrated that canola event LBFLFK contains 2 T-DNA integration sites containing all 13 intended gene expression cassettes, one in the Cnn random chromosome and the second in the C03 chromosome of the canola genome. No additional elements, including intact or partial DNA fragments of the gene cassettes or backbone sequences from the plasmid vector, linked or unlinked to the intact inserts, were detected in canola event LBFLFK .

The stability of the inserts within canola event LBFLFK was verified by Illumina-based sequencing over 3 generations. The inheritance pattern of both T-DNA inserts across 3 segregating generations of canola event LBFLFK showed that the 2 inserts segregated independently according to Mendelian rules of inheritance for 2 unlinked genetic loci.

4. Criteria for the environmental assessment

4.1 Potential for canola event LBFLFK to become a weed of agriculture or be invasive of natural habitats

Canola (B. napus) possesses some of the characteristics that are common to weeds and invasive plants. It is an annual crop that may persist in unmanaged ecosystems without human intervention. There have been reports of B. napus becoming a weed of agriculture in North America and other parts of the world; however, it has not become an abundant or problematic weed in Canada, despite being cultivated in Canada for many years. B. napus plants can grow as volunteers in cultivated fields in the seasons following a B. napus crop, but they are usually eliminated by soil cultivation or the use of herbicides. According to the information provided by BASF Canada Inc., canola event LBFLFK was determined not to be significantly different from unmodified canola varieties in this respect.

The CFIA evaluated data submitted by BASF Canada Inc. on the reproductive biology and life history traits of canola event LBFLFK. Canola event LBFLFK was field-tested at 6 fall-seeded locations in the southern United States (US) in 2014 and 2015, and at 8 spring-seeded locations in the northern US in 2015. The locations of the trials in the northern US share similar environmental and agronomic conditions to the canola production areas of Manitoba, Saskatchewan and Alberta, and were considered representative of the major Canadian canola growing regions.

During the field trials, canola event LBFLFK was compared to the unmodified control canola variety with a similar genetic background. Reference canola varieties were also included in these trials to establish ranges of comparative values that are representative of currently grown canola varieties in Canada. Phenotypic and agronomic traits were evaluated, covering a broad range of characteristics that encompass the entire life cycle of the canola plant. The traits evaluated in the northern US trials included field emergence, early plant stand, final plant stand, pod count, plant height, seed moisture, 1000-seed weight, and yield. Although instances of statistically significant differences were observed between canola event LBFLFK and the unmodified control canola variety for some traits in the individual-site analyses, there were no consistent trends in the data across locations that would indicate the differences were due to the genetic modification. The 6 locations in the southern US were not representative of major Canadian growing regions, but the results from these locations were consistent with the results from the Canadian-equivalent locations. The statistical analysis of these observations showed no biologically meaningful differences between canola event LBFLFK and the unmodified control canola variety, and therefore supports a conclusion of phenotypic and agronomic equivalence to currently grown canola varieties.

BASF Canada Inc. evaluated the seed germination of canola event LBFLFK under a standard regime (25°C for 8 hours in light, 15°C for 16 hours in dark), a warm temperature regime (25°C with 8/16 hours light/dark cycle) and a cold temperature regime (10°C in the dark for 10 days then 25°C 8/16 hours light/dark cycle for 3 days). Seed of canola event LBFLFK harvested from the 6 northern US field trials was compared to the unmodified control canola variety and reference varieties harvested from the same field trials. The germination rates of canola event LBFLFK seeds were statistically lower than those of the unmodified control canola variety seeds at all 3 temperature regimes, but within the range observed for the reference varieties for the standard and warm temperature regimes. The germination rate of canola event LBFLFK seeds was below the range of the reference varieties for the cold temperature regime. The trend toward reduced germination is likely a consequence of the modified fatty acid profile of the seed of canola event LBFLFK. A secondary dormancy test was performed to confirm that reduced germination of canola event LBFLFK is not associated with increased seed secondary dormancy. The results showed that the secondary dormancy of canola event LBFLFK seed was not increased, and that the lower seed germination rates observed for canola event LBFLFK were due to reduced seed viability. Therefore, the introduction of the novel traits tends to reduce the viability of seed of canola event LBFLFK. As reduction of seed viability does not confer a competitive advantage to the plant, it is not associated with increased weediness potential of canola event LBFLFK.

The response of canola event LBFLFK to abiotic stressors was observed at the agronomic trial sites. The abiotic stressors included excessive rainfall, wind, drought, heat, and cold/wet weather. Observations were collected at 4 crop development stages. No differences were observed between canola event LBFLFK and the unmodified control canola variety for their responses to the abiotic stressors at all measured growth stages.

The susceptibility of canola event LBFLFK to a range of canola pests and pathogens was evaluated in the field at 3 sites in the northern US (Idaho, South Dakota, and Minnesota) (see Section 4.3: Potential for canola event LBFLFK to become a plant pest). No consistent trend in decreased susceptibility to pests or pathogens was observed in canola event LBFLFK compared to the unmodified control canola variety.

No competitive advantage was conferred to plants of canola event LBFLFK, other than that conferred by tolerance to imidazolinone herbicides, as the reproductive characteristics, growth characteristics and tolerance to abiotic and biotic stressors of canola event LBFLFK were comparable to those of the unmodified control canola variety, and seed viability tended to be lower. Tolerance to imidazolinone herbicides provides a competitive advantage only when these herbicides are used and will not, in itself, make an imidazolinone tolerant plant weedier or more invasive of natural habitats. Canola event LBFLFK plants growing as volunteers will not be controlled if imidazolinone herbicides are used as the only weed control tool. However, control of canola event LBFLFK as a volunteer weed in subsequent crops or in fallow ground can be achieved by the use of other classes of herbicides or by mechanical means.

The novel traits have no intended or observed effects on weediness or invasiveness. The CFIA has therefore concluded that canola event LBFLFK has no altered weediness or invasiveness potential in Canada compared to currently grown canola varieties.

The CFIA considers the changes in usual agronomic practices that may arise from volunteer plants with novel herbicide tolerances. Similarly, the CFIA considers the potential that continued application of the same herbicide in subsequent rotations may lead to increased selection pressure for herbicide tolerant weed populations. In order to address these issues, a herbicide tolerance management plan that includes integrated weed management strategies should be implemented. These plans may include a recommendation to rotate or combine weed control products with alternate modes of action and to employ other weed control practices.

With respect to its unconfined release into the environment, cultivation of canola event LBFLFK is subject to herbicide tolerance management requirements. According to BASF Canada Inc., canola event LBFLFK is not intended to be cultivated in Canada. Therefore, a herbicide tolerance management plan specific to this product is not required at this time. However, if canola event LBFLFK is cultivated in Canada, a herbicide tolerance management plan must be implemented in order to manage canola event LBFLFK volunteers and delay the development of tolerance to imidazolinone herbicides in local weed populations within agro-ecosystems.

4.2 Potential for gene flow from canola event LBFLFK to sexually compatible plants whose hybrid offspring may become more weedy or more invasive

Successful interspecific and intergeneric crosses between B. napus and some sexually compatible species have been reported in the scientific literature. However, many of these crosses have required extensive human intervention, and the rates of natural hybridization between B. napus and weedy relatives resulting in fertile offspring appear to be very low. Sinapis arvensis is considered the most weedy of B. napus relatives in Western Canada. Hybrids between both species can be produced under field conditions, however at very low frequency. Additionally, backcrossing of the hybrids to S. arvensis failed to produce viable progeny. Therefore, the likelihood of introgression of traits from B. napus to S. arvensis appears to be very low. In crosses with other wild related species (that is, Raphanus raphanistrum and Erucastrum gallicum), no viable hybrid seed was produced. Stable gene transfer from B. napus is most likely with Brassica crops such as B. juncea and B. rapa (see biology document BIO2017-03: The Biology of Brassica napus L. (Canola/Rapeseed) for more information).

The modified fatty acid composition in seeds of canola event LBFLFK is unrelated to weediness characteristics. Field trials confirmed that the reproductive characteristics, growth characteristics and tolerance to abiotic and biotic stresses of canola event LBFLFK are comparable to those of the unmodified control canola variety (see Section 4.1: Potential for the canola event LBFLFK to become a weed of agriculture or be invasive of natural habitats).

Similarly, it is anticipated that with regards to weediness characteristics, no competitive advantage would be conferred to hybrids arising through interspecific or intergeneric hybridization with canola event LBFLFK, other than that conferred by tolerance to imidazolinone herbicides. The imidazolinone tolerance trait would confer no competitive advantage to these plants unless challenged by imidazolinone herbicides. This would only occur in managed ecosystems where imidazolinone herbicides are used for weed control. As with imidazolinone tolerant canola event LBFLFK volunteers, these herbicide tolerant individuals, should they arise, could be controlled using mechanical means or herbicides other than imidazolinone herbicides. Hybrids, if they developed, could potentially result in the loss of imidazolinone herbicides as a tool to control these species. This, however, can be avoided by the use of sound crop management practices.

With respect to its unconfined release into the environment, cultivation of canola event LBFLFK is subject to herbicide tolerance management requirements. According to BASF Canada Inc., canola event LBFLFK is not intended to be cultivated in Canada. Therefore, a herbicide tolerance management plan specific to this product is not required at this time. However, if canola event LBFLFK is cultivated in Canada, a herbicide tolerance management plan must be implemented in order to minimize transfer of the herbicide tolerance trait into weedy relatives within agro-ecosystems.

This information led the CFIA to conclude that gene flow from canola event LBFLFK to related species in Canada is possible, but would not result in increased weediness or invasiveness of the resulting progeny.

4.3 Potential for canola event LBFLFK to become a plant pest

Canola is not considered to be a plant pest in Canada, and the imidazolinone herbicide tolerance trait introduced into canola event LBFLFK is unrelated to plant pest potential (that is, the potential for the plant to harbour new or increased populations of pathogens or pests). However, considering the known roles of fatty acids in plant-pathogen and plant-pest interactions, BASF Canada Inc. submitted studies to examine the response of canola event LBFLFK to plant pathogens and pests.

The susceptibility of canola event LBFLFK to canola pest insects and pathogens was evaluated in the field at 3 sites in northern US (Idaho, South Dakota, and Minnesota). The pest insects observed included:

• flea beetles
• blister beetles
• diamond moths
• cabbage looper
• beet webworms
• stink bugs
• lygus
• aphids
• thrips
• earwigs
• crane flies
• grasshoppers
• leafhoppers

The pathogens observed included:

• Alternaria
• aster yellow
• bacterial leaf spot
• black leg
• Cercospora leaf spot
• clubroot
• downy mildew
• Fusarium wilt
• Phytophthora
• powdery mildew
• root rot
• Sclerotinia

Insects were sampled at 4 crop developmental stages using visual observations, sticky traps, and pitfall traps. Sticky traps target flying and foliage-dwelling arthropods, while pitfall traps target ground surface-dwelling arthropods.

With the exception of flea beetles, ecological evaluations of canola event LBFLFK did not show any increase or decrease in susceptibility to pathogens and pest insects compared to the unmodified control canola variety and reference canola varieties grown at the same locations. Higher numbers of flea beetles were observed on canola event LBFLFK compared to the unmodified control variety in all 3 locations and regardless of the sampling method. In some instances the difference was statistically significant and the flea beetles observed on canola event LBFLFK outnumbered those observed on reference canola varieties. However, separate field damage observations conducted across 14 sites in the US in spring 2015 did not record higher numbers of flea beetles on canola event LBFLFK compared to the unmodified control canola variety. It was noted that potential increased attractiveness of canola event LBFLFK on flea beetles, if any, would be managed using standard flea beetle control practices currently used on canola, which were not deployed during the field trial testing conditions.

The CFIA has therefore concluded that, under standard pest control practices deployed in commercial agriculture, canola event LBFLFK does not display altered plant pest potential compared to currently grown canola varieties.

4.4 Potential impact of canola event LBFLFK and its gene products on non-target organisms, including humans

The CFIA evaluated the potential impacts of the novel traits expressed by canola event LBFLFK (that is, modified fatty acid profile and tolerance to the imidazolinone herbicide) and the proteins that confer the novel traits on organisms interacting with canola.

The imidazolinone tolerance trait introduced into canola event LBFLFK is unrelated to a potential impact on non-target organisms.

Detailed characterization of the introduced novel proteins (D12D(Ps), D6D(Ot), D6E(Tp), D6E(Pp), D5D(Tc), O3D(Pir), O3D(Pi), D5E(Ot), D4D(Tc), D4D(Pl), and modified AHAS(At)) led to the conclusion that none of these proteins displayed any characteristics of a potential toxin or allergen (see Section 5.2: Potential impact of canola event LBFLFK on animal health and human safety as it relates to the potential transfer of residues into foods of animal origin and worker/bystander exposure to the feed).

Omega-3 long chain-polyunsaturated fatty acids (LC-PUFAs), like DHA and EPA, have many functions in both aquatic and terrestrial organisms. Cultivating canola event LBFLFK will expose some organisms to levels of omega-3 LC-PUFAs that they do not normally encounter. The main potential route of exposure is via seed consumption because the novel long-chain polyunsaturated fatty acids are only produced in the seed. Indirect exposure via the consumption of prey that has fed on canola event LBFLFK seed has been shown to be negligible. Hixson et al. (2016)^Footnote 1 studied the effect of DHA and EPA in the diet of a canola pest, cabbage white butterfly (Pieris rapae). Authors concluded that the presence of EPA and DHA in diets of larval P. rapae may alter adult mass and wing morphology. No effect was seen on other parameters measured, including developmental phenology, larval or pupal weight, food consumption and larval mortality. Therefore, the novel proteins produced in canola event LBFLFK are unlikely to have a direct or indirect effect on arthropods in Canada, with the possible exception of limited impacts on pest species feeding on canola seed.

The abundance of beneficial arthropods was evaluated in the field at the same locations in the US and using the same protocol as for evaluating the abundance of canola pests (see Section 4.3: Potential for canola event LBFLFK to become a plant pest). The beneficial arthropods observed included lady beetles, bees, parasitic wasps, spiders, lacewings, hoverflies, mayflies and dragonflies. These observations showed similar abundance of beneficial arthropods in plots of canola event LBFLFK compared to plots of the unmodified control canola variety and reference canola varieties grown at the same locations.

Collectively, these information elements indicate that the interactions between canola event LBFLFK and the populations of animals and microorganisms interacting with canola crops will be similar compared to currently grown canola varieties.

The CFIA has therefore determined that the unconfined release of canola event LBFLFK in Canada will not result in altered impacts on non-target organisms, including humans, compared to currently grown canola varieties.

4.5 Potential impact of canola event LBFLFK on biodiversity

Canola event LBFLFK displays no novel phenotypic characteristics that would extend its range beyond the current geographic range of canola production in Canada. The novel traits expressed by canola event LBFLFK have been determined to be unlikely to cause adverse effects on non-target organisms and canola event LBFLFK does not display increased weediness, invasiveness or plant pest potential. Canola (B. napus) can outcross to B. rapa and B. juncea, and potentially to wild relatives, under natural conditions in Canada. However, the consequences of the transfer of the novel traits are minimal. The novel herbicide tolerance trait does not confer any selective advantage in the absence of the herbicide, and imidazolinone-tolerant hybrids can be controlled by herbicides with other modes of action, or through mechanical means. The transfer of omega-3 LC-PUFA synthesis proteins and their corresponding products is not expected to confer a selective advantage or modify the interactions between the hybrid progeny and organisms interacting with these plants.

Canola event LBFLFK has tolerance to the imidazolinone herbicide. The use of this herbicide in cropping systems has the intended effect of reducing local weed populations within agro-ecosystems. This may result in a reduction in local weed species biodiversity, and may have effects on other trophic levels which utilize these weed species. It must be noted however that the goal of reduction in weed biodiversity in agricultural fields is not unique to the use of plants with novel traits, canola event LBFLFK, or the cultivation of canola. According to BASF Canada Inc., canola event LBFLFK is not intended to be cultivated in Canada. Even if it was grown in Canada, it is unlikely that canola event LBFLFK would have any indirect effects on biodiversity, in comparison to the effects that would be expected from cultivation of currently grown canola varieties.

The CFIA has concluded that the introduced genes and their corresponding novel traits do not confer to canola event LBFLFK any characteristic that would result in unintended environmental effects following unconfined release. The CFIA has therefore concluded that the potential impact on biodiversity of canola event LBFLFK is unlikely to be different from that of the canola varieties that are currently grown in Canada.

5. Criteria for the Livestock Feed Assessment

The Animal Feed Program (AFP) considered the safety and efficacy of feed ingredients derived from canola event LBFLFK, including nutrient and anti-nutrient profiles; the presence of gene products, residues, and metabolites, in terms of animal health and human safety as it relates to the potential transfer of residues into foods of animal origin and worker/bystander exposure to the feed; and whether feeds derived from canola event LBFLFK meet the definitions and requirements of feeds as listed in Schedule IV of the Feeds Regulations.

5.1 Potential impact of canola event LBFLFK on livestock nutrition

5.1.1 Nutrient and anti-nutrient composition

The nutritional equivalence of canola event LBFLFK (non-sprayed and sprayed with imazamox or imidazolinone herbicide) to its unmodified control canola variety (Kumily) and 6 reference canola varieties was assessed from replicated field sites in the US during the winter (2014/15) and spring (2015) growing seasons. Canola grain samples harvested from all field plots were used for composition data analysis. For analysis of the meal and oil, grain samples were harvested from replicated plots of canola event LBFLFK (sprayed with Beyond® herbicide), the unmodified control canola variety and 3 reference varieties grown during the 2016 growing season. Nutrients analyzed in the grain, meal and oil included the following: proximate (protein, crude fat, moisture and ash), crude fibre (CF), acid detergent fibre (ADF), neutral detergent fibre (NDF), total dietary fibre (TDF), amino acids, fatty acids, minerals, vitamins, phytosterols, glucosinolates and anti-nutrients (phytic acid, tannins, sinapine, coumaric and ferulic acids) as per the guidelines of the Organization for Economic Co-operation and Development (OECD, 2011). Analysis of variance was used for statistical examination of the compositional data and comparison of means was conducted at a 95% confidence level (P<0.05). Statistically significant differences observed between canola event LBFLFK and the unmodified control canola variety were assessed within the range of the reference canola varieties included in the trials. Mean values were also compared to the range of means generated from the Agriculture & Food Systems Institute (AFSI) Crop Composition Database (formerly known as ILSI), and peer-reviewed scientific literature and OECD consensus documents (2011) to provide context for the comparative analyses and assess the broader biological relevance of the results.

5.1.1.1 Seed

No statistically significant differences were observed between canola event LBFLFK and its unmodified control canola variety for all proximate analysis, fiber components and amino acids in the 2014/15 season. Statistically significant differences were observed between canola event LBFLFK and its unmodified control canola variety for ADF, CF, NDF, alanine, aspartic acid, leucine, methionine, tyrosine, and valine. However, the means of these nutrients in canola event LBFLFK were within the ranges of the reference canola varieties across all sites, and were also within the range of values in peer-reviewed literature and AFSI Crop Composition database. Therefore, the observed differences were not considered biologically relevant. For vitamins analyzed in the 2014/15 season; delta-tocopherol was statistically significantly lower and vitamin K₁ were statistically significantly higher in canola event LBFLFK (sprayed only) when compared to the unmodified control variety. In the spring 2015 season, vitamin K₁ was statistically significantly higher in canola event LBFLFK compared to the unmodified control variety. Calcium and magnesium were statistically significantly lower in canola event LBFLFK compared to the unmodified control canola variety, for both seasons, while phosphorus was statistically significantly higher in canola event LBFLFK (non-sprayed) compared to the unmodified control canola variety in the 2015 season. All mean values were within the natural variation of the reference canola varieties, published scientific literature and AFSI Crop Composition database. Therefore, the observed differences were not considered biologically relevant.

No statistically significant differences were observed between canola event LBFLFK and the unmodified control canola variety for phytic acid and ferulic acid. In both seasons, tannins were consistently below the limits of quantification for all measurements. Sinapine was statistically significantly lower for canola event LBFLFK compared to the unmodified control variety in both seasons, while coumaric acid was statistically significant different in canola event LBFLFK (non-sprayed) compared to the control, in the 2015 season. All means were within the range of the reference varieties and AFSI Crop Composition database values. Total glucosinolates in canola event LBFLFK (12.78 μmol/g dry weight) were statistically significantly higher than in the unmodified control variety (11.62 μmol/g dry weight) in the 2015 season, and a similar difference was seen in the comparison of canola event LBFLFK (non-sprayed) to the unmodified control variety in the winter 2014/15 season. However, the mean values for canola event LBFLFK were within the range of the reference varieties. Additionally, the measured total glucosinolate values for canola event LBFLFK met the quality standards (4-26.8 μmol/g dry weight) for canola (OECD, 2011). Therefore, these differences were not considered biologically relevant in terms of canola quality. In the 2014/15 season, delta-5 avenasterol and stigmasterol in canola event LBFLFK were statistically significantly lower and delta-7 stigmastenol was statistically significantly higher than the unmodified canola control variety. In contrast, beta-sitosterol and campesterol were statistically significantly lower in canola event LBFLFK compared to the control in spring 2015. Brassicasterol and total phytosterols were statistically significantly lower for canola event LBFLFK compared to unmodified control variety across both seasons. However, the means were within the range of the reference varieties and or published literature values.

5.1.1.2 Defatted meal

Defatted meal (pre-pressed, solvent extracted) from canola meal event LBFLFK did not show any significant differences in proximate (moisture, crude fat, protein, ash, and carbohydrates) and fiber (crude fiber, acid detergent fiber, and neutral detergent fiber) composition when compared to the unmodified control canola variety. The crude fat content between both defatted meal fractions was comparable, with a mean of 1.12% for canola event LBFLFK versus 0.94% in the unmodified control variety. Except for histidine and cysteine, no statistically significant differences were observed across sites between canola meal event LBFLFK and the unmodified control meal. The mean values for histidine and cystine in canola event LBFLFK were within the range of values reported in OECD 2011. No statistically significant differences were observed in the vitamin content of the meal from canola event LBFLFK compared to the unmodified control variety. Calcium was statistically significantly lower and potassium was statistically significantly higher in defatted meal from canola event LBFLFK compared to the unmodified control variety but were within the range of the reference control varieties and/or literature ranges. Values for phytic acid and tannins were not statistically significantly different between canola event LBFLFK and the unmodified control meal. Statistically significant differences were observed for progoitrin, 4-hydroxyglucobrassicin, glucobrassicin, ferulic acid, p-coumaric acid, and sinapine, with values lower in canola event LBFLFK compared to the unmodified control meal. However, mean values were either within the range of reference varieties or the AFSI Crop Composition database. Both canola event LBFLFK (17.0 μmol/g total glucosinolates) and unmodified control variety (20.8 μmol/g total glucosinolates) contained less than 30 μmol/g total glucosinolates as expected in the composition of canola meal (OECD, 2011) and the Canadian Feeds Regulations (Schedule IV, Part 1).

5.1.1.3 Oil

Endogenous fatty acids not impacted by novel trait

The introduction of the omega-3 LC-PUFA trait and the associated enzymatic pathway in canola event LBFLFK did not impact other fatty acids already present in canola crop. No comparative statistical analysis could be performed for the fatty acids: hexadecatrienoic (C16:3, n-3); erucic (C22:1, n-9) and docosadienoic (C22:2, n-6), which were below the level of quantification across all samples. Erucic acid levels for both canola event LBFLFK and unmodified control variety were below the standard maximum of 2% specified for the definition of canola quality in the oil component (OECD 2011, and the Canadian Feeds Regulations (Schedule IV, Part 1)). The fatty acid profiles of canola event LBFLFK (grain and oil) showed that myristic (C14:0), palmitic (C16:0), cis-7-hexadecenoic (C16:1, n-9); margaric (C17:0), magaroleic (C17:1), cis-vaccenic (C18:1, n-7) and eicosadienoic (C20:2, n-6) fatty acids were not impacted by the omega-3 LC-PUFA trait as there were either no statistically significant differences between the levels in canola event LBFLFK compared to the unmodified control variety, or they were within the range of values of the reference varieties, AFSI Crop Composition database, Codex Standard or published literature values.

Endogenous fatty acids impacted by novel trait

The introduction of the omega-3 LC-PUFA trait and the associated enzymatic pathway in canola event LBFLFK resulted in the modification of some fatty acids already present in canola. Statistically significant differences between canola event LBFLFK and the unmodified control variety were consistently observed for the following fatty acids: palmitoleic (C16:1, n-7), stearic (C18:0), oleic (C18:1, n-9), linoleic (C18:2, n-6), cis,cis-6,9-octadecadienoic (C18:2, n-9), alpha-linolenic (C18:3, n-3), arachidic (C20:0) (grain), gondoic (C20:1, n-9), 6,11-eicosadienoic (C20:2, n-9), behenic (C22:0), lignoceric (C24:0), nervonic (C24:1, n-9) and total trans fatty acids (grain). The mean values in canola event LBFLFK oil for C16:1, n-7, C18:0, C18:2, n-9 C18:3, n-3, C20:0, C20:1, n-9, C20:2, n-9; C22:0, C24:0, C24:1, n-9 were either within or outside the range of the reference canola varieties. Those outside the range of the reference canola varieties were however either within AFSI Crop Composition database, Codex Standard or scientific literature values and therefore the observed differences were not considered biologically relevant.

In canola event LBFLFK oil, oleic acid, the starting substrate fatty acid for the newly introduced omega-3 LC-PUFA synthesis pathway, was significantly decreased compared to the unmodified control canola variety (29.5% vs. 59% of the total oil profile), while linoleic acid was significantly increased compared to the unmodified control variety (30% vs. 19.6% of the total oil profile). The mean values were consistently outside of the range of the reference canola varieties and outside the range of natural variation based on the peer-reviewed literature, Codex standards and the AFSI Crop Composition database. These observed differences in canola event LBFLFK were considered biologically relevant and therefore were not equivalent to the oleic and linoleic fatty acids of conventional canola oil. BASF Canada Inc. explained that this observation is to be expected, as oleic and linoleic acids are the primary precursors for the synthesis of EPA and DHA and therefore the conversion of oleic acid into longer chain and more highly unsaturated fatty acids was expected to have an impact on the levels on oleic and linoleic acids in canola event LBFLFK. An additional secondary effect was observed on the levels of stearic acid, which was significantly higher in canola event LBFLFK compared to the unmodified control variety and outside the range of the reference canola varieties. Additionally, the higher relative linoleic acid content in canola event LBFLFK was also attributable to the newly expressed delta-12 desaturase from Phytophthora sojae that produces this fatty acid from oleic acid.

Fatty acids introduced by novel trait

The introduction of the omega-3 LC-PUFA trait and the associated enzymatic pathway in canola event LBFLFK resulted in the introduction of 13 new fatty acids not present in the oil of the unmodified control canola variety and reference varieties. Thirteen LC-PUFAs: gamma-linolenic (C18:3, n-6), stearidonic (C18:4, n-3), eicosatrienoic (C20:3, n-3), dihomo-gamma linolenic acid (C20:3, n-6), mead (C20:3, n-9), bishomostearidonic (C20:4, n-3), arachidonic (C20:4, n-6), eicosapentaenoic (C20:5, n-3 (EPA)), docosatetraenoic (C22:4, n-3), adrenic (C22:4, n-6), clupanodonic (C22:5, n-3), osbond (C22:5, n-6), and docosahexaenoic (C22:6, n-3 (DHA)) observed in canola event LBFLFK grain and oil were consistently below the level of quantification in the unmodified control variety and the reference canola varieties and therefore no statistical comparisons were computed. The presence of these fatty acids only in canola event LBFLFK was attributed to the introduction of the omega-3 LC-PUFA trait. The levels of EPA in canola event LBFLFK grain ranged from 5-8% for the winter and 5.5-7% spring seasons, respectively, while the levels in canola event LBFLFK crude oil and refined oil were 5.4% and 4.1%, respectively. The DHA levels in canola event LBFLFK grain ranged from 0.6-0.73%, while it was 0.4-0.5% in the crude and refined oil. In total, the level of omega-3 LC-PUFAs increased in canola event LBFLFK oil (10-15%) compared to the unmodified control variety and reference varieties (2-7%). A comparative analysis of the fatty acid composition of various edible oils and fat-containing foods showed that all 13 new fatty acids introduced by the omega-3 LC-PUFA trait in canola event LBFLFK oil are already present at varying levels in other food and feed products such as fish, algae, cod, menhaden oils, salmon, eggs and dairy products that have a history of safe use for animal and human consumption.

5.1.1.4 Feeding study with canola event LBFLFK oil

Cargill, in partnership with BASF Canada Inc., conducted a fish study in Chile to test the nutritional efficacy and safety of canola event LBFLFK oil compared to fish oil and conventional canola oil which are commonly used in aquaculture. Three diets were formulated with fish oil, LBFLFK oil and conventional canola oil. Canola event LBFLFK oil was included at 16% of the diet. Dietary levels of EPA and DHA content were standardized among the 3 feed types with other feed ingredients. Diets were pelleted and fed to 100 Atlantic salmon per 350L tank for each diet. Each feed type was replicated 4 times. Feed was provided 4 times daily and non-consumed feed were collected twice daily. Fish were weighed at beginning and end of trial. Fatty acid composition and liver histology were examined. At the end of trial, whole fish from each treatment were ground and solvent-extracted, then fatty acid composition was determined for the extracted oil. No statistically significant effects were observed among fish on the 3 diets for final weight, weight increase, feed intake, feed conversion and survival. The liver histology was similar and normal for all 3 treatments. Bioavailability of canola event LBFLFK oil was similar to fish oil. Fish fed with canola event LBFLFK derived oil had similar amounts of EPA (3.37g/100g fat) and DHA (7.53 g/100g fat) as fish fed with a fish oil diet (2.48 g/100g fat and 7.54 g/100g fat). Fish fed with conventional canola oil showed significantly lower EPA (1.62g/100g fat) and DHA (5.98 g/100g fat) values as expected.

5.1.2 Conclusions

It was concluded based on the evidence provided by BASF Canada Inc. that the nutritional composition of canola event LBFLFK is similar to conventional canola for all nutrients, except for the modified levels of oleic (C18:2, n-9) and linoleic (C18:2, n-6) acids and the presence of 13 new polyunsaturated fatty acids. LBFLFK canola grain and oil, has introduced levels of omega-3 LC-PUFAs including EPA (eicosapentaenoic acid, 20:5, n-3) and DHA (docosahexaenoic acid, 22:6, n-3) which are absent in the conventional canola oil. Canola event LBFLFK oil would be used in human food and aquaculture feeds, as a source of omega-3 LC-PUFAs. As a result of the changed fatty acid profile of canola event LBFLFK and its new usage in aquaculture, a new ingredient describing the oil from canola event LBFLFK was created in Part 1, Schedule IV Feeds Regulations. In contrast, the pre-pressed, solvent-extracted defatted meal (1% oil) from canola event LBFLFK was not nutritionally different from conventional canola meal and would therefore be used similarly as the conventional canola meal in livestock feeds. As such, no ingredient definition was created for the meal, however the definition for conventional canola meal in Part 1, Schedule IV Feeds Regulations has been modified to include pre-pressed, solvent extracted meal from canola event LBFLFK.

5.2 Potential impact of canola event LBFLFK on animal health and human safety as it relates to the potential transfer of residues into foods of animal origin and worker/bystander exposure to the feed

Canola event LBFLFK synthesizes omega-3 LC-PUFAs including EPA and DHA due to the insertion of genes encoding fatty acid desaturase and elongase proteins and is tolerant to imidazolinone herbicides due to production of modified acetohydroxy acid synthase protein, AHAS(At). A weight-of-evidence approach was used to evaluate the risk to livestock consuming feed ingredients from canola event LBFLFK, humans consuming foods of animal origin derived from those livestock, and workers/bystanders exposed to the feed ingredients from this event due to the following significant changes:

• modified protein acetohydroxy acid synthase from Arabidopsis thalianoa (AHAS(At))
• novel protein delta-12 desaturase from Phytophthora sojae (D12D(Ps))
• novel protein delta-6 desaturase from Ostreococcus tauri (D6D(Ot))
• novel protein delta-6 elongase from Thalassiosira pseudonana (D6E(Tp))
• novel protein delta-6 elongase from Physcomitrella patens (D6E(Pp))
• novel protein delta-5 desaturase from Thraustochytrium sp. (D5D(Tc))
• novel protein omega-3 desaturase from Phytophthora infestans (O3D(Pi))
• novel protein omega-3 desaturase from Phytophthora infestans (O3D(Pi))
• novel protein delta-5 elongase from Ostreococcus tauri (D5E(Ot))
• novel protein delta-4 desaturase from Thraustochytrium sp. (D4D(Tc))
• novel protein delta-4 desaturase from Pavlova lutheri (D4D(Pl))
• EPA/DHA levels in foods of animal origin

5.2.1 Acetohydroxy acid synthase (At) protein – AHAS(At)

The potential allergenicity and toxicity of AHAS(At) protein were evaluated. With respect to its potential allergenicity, the source of the ahas(At) gene, Arabidopsis thaliana, is not known to produce allergens and there is a history of safe use of the host organism and a history of safe exposure to the AHAS(At). A bioinformatics analysis of the AHAS(At) protein amino acid sequence confirmed the lack of relevant similarities between the AHAS(At) protein and known allergens. Unlike many allergens, canola event LBFLFK produced AHAS(At) protein was shown experimentally to be degraded in simulated gastric fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the AHAS(At) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the AHAS(At) protein lacks a mode of action to suggest that it is intrinsically toxic. There is a history of safe use for donor organism and safe exposure to the AHAS(At) in canola. A bioinformatics analysis of the AHAS(At) protein amino acid sequences confirmed the lack of relevant similarities between the AHAS(At) protein and known toxins. The exposure to the AHAS(At) protein is expected to be extremely low as the AHAS(At) protein is expressed at very low level in canola event LBFLFK and is degraded under conditions which simulate the mammalian digestive tract. The weight of evidence thus indicates that the AHAS(At) protein is unlikely to cause a toxic concern.

Therefore, the modified AHAS(At) in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.2 Delta-12 desaturase (Ps) – D12D(Ps) protein

The potential allergenicity and toxicity of the D12D(Ps) protein were evaluated. With respect to its potential allergenicity, the source of the D12D(Ps) gene, Phytophthora sojae, is not known to produce allergens and a bioinformatics evaluation of the D12D(Ps) protein amino acid sequences confirmed the lack of relevant similarities between the D12D(Ps) protein and known allergens. Unlike many allergens, canola event LBFLFK produced D12D(Ps) protein was shown experimentally to be rapidly degraded in simulated gastric and intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the D12D(Ps) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D12D(Ps) protein lacks a mode of action to suggest that it is intrinsically toxic and a bioinformatics evaluation of the D12D(Ps) protein amino acid sequences confirmed the lack of relevant similarities between the D12D(Ps) protein and known toxins. The exposure to the D12D(Ps) protein is expected to be extremely low as the D12D(Ps) protein is expressed at very low level in canola event LBFLFK and is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the D12D(Ps) protein is unlikely to cause a toxic concern.

Therefore, the D12D(Ps) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans and workers/bystanders.

5.2.3 Delta-6 desaturase (Ot) – D6D(Ot) protein

The potential allergenicity and toxicity of the D6D(Ot)protein were evaluated. With respect to its potential allergenicity, the source of the D6D(Ot) gene, Ostreococcus tauri, is not known to produce allergens and a bioinformatics analysis of the D6D(Ot) protein amino acid sequences confirmed the lack of relevant similarities between the D6D(Ot) protein and known allergens. Unlike many allergens, canola event LBFLFK D6D(Ot) protein was shown experimentally to be degraded in simulated gastric and intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the D6D(Ot) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D6D(Ot) protein lacks a mode of action to suggest that it is intrinsically toxic and a bioinformatics analysis of the D6D(Ot) protein amino acid sequences confirmed the lack of relevant similarities between the D6D(Ot) protein and known toxins. The exposure to the D6D(Ot) protein is expected to be extremely low as the D6D(Ot) protein is expressed at very low level in canola event LBFLFK and is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the D6D(Ot) protein is unlikely to cause a toxic concern.

Therefore, the D6D(Ot) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans and workers/bystanders.

5.2.4 Delta-6 elongase (Tp) – D6E(Tp) protein

The potential allergenicity and toxicity of the D6E(Tp) protein were evaluated. With respect to its potential allergenicity, the source of the D6E(Tp) gene, Thalassiosira pseudonana, is not known to produce allergens and there is a history of safe exposure to the delta-6 elongase (D6E) protein. A bioinformatics analysis of the D6E(Tp) protein amino acid sequences confirmed the lack of relevant similarities between the D6E(Tp) protein and known allergens. Unlike many allergens, canola event LBFLFK D6E(Tp) protein was shown experimentally to be degraded in simulated intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the D6E(Tp) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D6E(Tp) protein lacks a mode of action to suggest that it is intrinsically toxic and a bioinformatics analysis of the D6E(Tp) protein amino acid sequences confirmed the lack of relevant similarities between the D6E(Tp) protein and known toxins. The exposure to the D6E(Tp) protein is expected to be low as the D6E(Tp) protein is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the modified D6E(Tp) protein is unlikely to cause a toxic concern.

Therefore, the D6E(Tp) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.5 Delta-6 elongase (Pp) – D6E(Pp) protein

The potential allergenicity and toxicity of the D6E(Pp) protein were evaluated. With respect to its potential allergenicity, the source of the D6E(Pp) gene, Physcomitrella patens, is not known to produce allergens and there is a history of safe exposure to the Delta-6-elongases (D6E). A bioinformatics analysis of the D6E(Pp) protein amino acid sequences confirmed the lack of relevant similarities between the D6E(Pp) protein and known allergens. The exposure to the D6E(Pp) protein is expected to be negligible as this protein was undetectable in tissues of canola event LBFLFK, using ELISA with a limit of detection 2.78 ppm, and the protein was potentially degraded under feed processing and digestion. The weight of evidence thus indicates that the D6E(Pp) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D6E(Pp) protein lacks a mode of action to suggest that it is intrinsically toxic. There is a history of safe use for donor organism and safe exposure to the delta-6-elongases (D6E). A bioinformatics evaluation of the D6E(Pp) protein amino acid sequences confirmed the lack of relevant similarities between the D6E(Pp) protein and known toxins. The exposure to the D6E(Pp) protein is expected to be negligible as the protein was undetectable in tissues of canola event LBFLFK, using ELISA with a limit of detection 2.78 ppm, and the protein was potentially degraded under feed processing and digestion. The weight of evidence thus indicates that the D6E(Pp) protein is unlikely to cause a toxic concern.

Therefore, the D6E(Pp) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans and and workers/bystanders.

5.2.6 Delta-5 desaturase (Tc) – D5D(Tc) protein

The potential allergenicity and toxicity of the D5D(Tc) protein were evaluated. With respect to its potential allergenicity, the source of the D5D(Tc) gene, Thraustochytrium sp., is not known to produce allergens and a bioinformatics evaluation of the D5D(Tc) protein amino acid sequences confirmed the lack of relevant similarities between the D5D(Tc) protein and known allergens. Unlike many allergens, canola event LBFLFK D5D(Tc) protein was shown experimentally to be degraded in simulated gastric and intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the D5D(Tc) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D5D(Tc) protein lacks a mode of action to suggest that it is intrinsically toxic and a bioinformatics evaluation of the D5D(Tc) protein amino acid sequences confirmed the lack of relevant similarities between the D5D(Tc) protein and known toxins. The exposure to the D5D(Tc) protein is expected to be negligible as the D5D(Tc) protein is expressed at very low levels in canola event LBFLFK and is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the modified D5D(Tc) protein is unlikely to cause a toxic concern.

Therefore, the D5D(Tc) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.7 Omega-3 desaturase (Pir) – O3D(Pir) protein

The potential allergenicity and toxicity of the O3D(Pir) protein were evaluated. With respect to its potential allergenicity, the source of the O3D(Pir) gene, Pythium irregulare, is not known to produce allergens and a bioinformatics evaluation of the O3D(Pir) protein amino acid sequences confirmed the lack of relevant similarities between the O3D(Pir)and known allergens. Unlike many allergens, canola event LBFLFK O3D(Pir) protein was shown experimentally to be degraded in simulated gastric and intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the O3D(Pir) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the O3D(Pir) protein lacks a mode of action to suggest that it is intrinsically toxic and a bioinformatics evaluation of the O3D(Pir) protein amino acid sequences confirmed the lack of relevant similarities between the O3D(Pir) protein and known toxins. The exposure to the O3D(Pir) protein is expected to be low as the O3D(Pir) protein is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the modified O3D(Pir) protein is unlikely to cause a toxic concern.

Therefore, the O3D(Pir) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.8 Omega-3 desaturase (Pi) – O3D(Pi) protein

The potential allergenicity and toxicity of the O3D(Pi) protein were evaluated. With respect to its potential allergenicity, the source of the D3D(Pi) gene, Phytophthora infestans, is not known to produce allergens and there is a history of safe exposure to the Omega-3 desaturase (O3D) protein. A bioinformatics evaluation of the O3D(Pi) protein amino acid sequences confirmed the lack of relevant similarities between the O3D(Pi) protein and known allergens. The exposure to the O3D(Pi) protein is expected to be negligible as the protein was undetectable in tissues of canola event LBFLFK, using western blot with a limit of detection 27.08 ppm, and the protein was potentially degraded under feed processing and digestion. The weight of evidence thus indicates that the O3D(Pi) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the O3D(Pi) protein lacks a mode of action to suggest that it is intrinsically toxic. There is a history of safe use for donor organism and safe exposure to the Omega-3 desaturase (O3D) protein. A bioinformatics evaluation of the O3D(Pi) protein amino acid sequences confirmed the lack of relevant similarities between the O3D(Pi) protein and known toxins. The exposure to the O3D(Pi) protein is expected to be negligible as the protein was undetectable in tissues of canola event LBFLFK, using western blot with a limit of detection 27.08 ppm, and the protein was potentially degraded under feed processing and digestion. The weight of evidence thus indicates that the O3D(Pi) protein is unlikely to cause a toxic concern.

Therefore, the O3D(Pi) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.9 Delta-5 elongase (Ot) – D5E(Ot) protein

The potential allergenicity and toxicity of the D5E(Ot) protein were evaluated. With respect to its potential allergenicity, the source of the D5E(Ot) gene, Ostreococcus tauri, is not known to produce allergens and a bioinformatics evaluation of the D5E(Ot) protein amino acid sequences confirmed the lack of relevant similarities between the D5E(Ot) protein and known allergens. Unlike many allergens, canola event LBFLFK D5E(Ot) protein was shown experimentally to be degraded in simulated gastric and intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the D5E(Ot) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D5E(Ot) protein lacks a mode of action to suggest that it is intrinsically toxic to livestock and a bioinformatics evaluation of the D5E(Ot) protein amino acid sequences confirmed the lack of relevant similarities between the D5E(Ot) protein and known toxins. The exposure to the D5E(Ot) protein is expected to be extremely low as the D5E(Ot) protein is expressed at very low levels in canola event LBFLFK and is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the D5E(Ot) protein is unlikely to cause a toxic concern.

Therefore, the D5E(Ot) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.10 Delta-4 desaturase (Tc) – D4D(Tc) protein

The potential allergenicity and toxicity of the D4D(Tc) protein were evaluated. With respect to its potential allergenicity, the source of the D4D(Tc) gene, Thraustochytrium sp., is not known to produce allergens and a bioinformatics evaluation of the D4D(Tc) protein amino acid sequences confirmed the lack of relevant similarities between the D4D(Tc) protein and known allergens. Unlike many allergens, canola event LBFLFK D4D(Tc) protein was shown experimentally to be degraded in simulated gastric and intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the D4D(Tc) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D4D(Tc) protein lacks a mode of action to suggest that it is intrinsically toxic to livestock and a bioinformatics evaluation of the D4D(Tc) protein amino acid sequences confirmed the lack of relevant similarities between the D4D(Tc) protein and known toxins. The exposure to the D4D(Tc) protein is expected to be extremely low as the D4D(Tc) protein is expressed at low levels in canola event LBFLFK and is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the D4D(Tc) protein is unlikely to cause a toxic concern.

Therefore, the D4D(Tc) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.11 Delta-4 desaturase (Pl) – D4D(Pl) protein

The potential allergenicity and toxicity of the D4D(Pl) protein were evaluated. With respect to its potential allergenicity, the source of the D4D(Pl) gene, Pavlova lutheri, is not known to produce allergens and a bioinformatics evaluation of the D4D(Pl) protein amino acid sequences confirmed the lack of relevant similarities between the D4D(Pl) protein and known allergens. Unlike many allergens, canola event LBFLFK D4D(Pl) protein was shown experimentally to be degraded in simulated gastric and intestinal fluid, not heat stable, and to be unglycosylated. The weight of evidence thus indicates that the D4D(Pl) protein is unlikely to cause an allergenic concern.

In terms of the potential toxicity, the D4D(Pl) protein lacks a mode of action to suggest that it is intrinsically toxic to livestock and a bioinformatics evaluation of the D4D(Pl) protein amino acid sequences confirmed the lack of relevant similarities between the D4D(Pl) protein and known toxins. The exposure to the D4D(Pl) protein is expected to be extremely low as the D4D(Pl) protein is expressed at low levels in canola event LBFLFK and is degraded under conditions which simulate the mammalian digestive tract and is unstable under heating conditions expected to be encountered during processing of some canola products. The weight of evidence thus indicates that the D4D(Pl) protein is unlikely to cause a toxic concern.

Therefore, the D4D(Pl) protein in canola event LBFLFK is unlikely to pose a risk to livestock, humans, and workers/bystanders.

5.2.12 EPA/DHA levels in foods of animal origin

The safety of EPA/DHA levels in foods of animal origin, following application of canola event LBFLFK oil in fish feeds, was also evaluated as part of the feed safety assessment.

It was determined that canola event LBFLFK oil, when used in fish feeds, would not present a safety concern to humans via the potential transfer of EPA/DHA into foods of animal origin, when comparing the estimated exposure to EPA/DHA derived from the farmed and wild Atlantic salmonid fish.

5.2.13 Conclusion

It was concluded, based on the evidence provided by BASF Canada Inc., that the production of the 7 desaturases, 3 elongases and modified acetohydroxy acid synthase protein, AHAS(At) in canola event LBFLFK are unlikely to pose a risk to livestock, humans and workers/by-standers. Therefore, pre-press solvent extracted defatted meal (1% oil) from canola event LBFLFK seed is considered as safe as meal from conventional canola species currently available in the Canadian market. Additionally, the use of canola event LBFLFK oil in fish feeds would not present a safety concern to humans via the potential transfer of EPA/DHA into foods of animal origin.

The safety and efficacy of the levels of omega-3 LC PUFAs in foods of animal origin such as milk, eggs and meat by feeding canola event LBFLFK oil to livestock species other than fish were not assessed at this time. As such, BASF Canada Inc. is required to use the identity preservation (IdP) system to ensure that the oil derived from canola event LBFLFK seeds are used in fish feeds only and not fed to other livestock species.

Furthermore, the recycling or repurposing of the oil derived from canola event LBFLFK intended for human consumption (for example, spent omega-3 canola oil obtained from human usages) is not authorized for use in livestock species other than fish.

Additionally, no authorization has been granted for use of whole seeds and forage derived from the canola event LBFLFK as livestock feed at this time since no data was provided to support the use of these ingredients as livestock feed. As such, whole seed and parts of omega-3 canola crop (for example, forage) shall not be grazed by livestock or fed as chop feed or forage for livestock.

6. New Information Requirements

If at any time, BASF Canada Inc. becomes aware of any new information regarding risk to the environment, livestock or human health, which could result from the unconfined environmental release or livestock feed use of canola event LBFLFK or lines derived from it, BASF Canada Inc. is required to immediately provide such information to the CFIA. On the basis of such new information, the CFIA will re-evaluate the potential impact of canola event LBFLFK on the environment, livestock and human health, and may re-evaluate its decision with respect to the livestock feed use and unconfined environmental release authorizations of canola event LBFLFK.

7. Regulatory Decision

Based on the review of the data and information submitted by BASF Canada Inc. and input from other relevant scientific sources, the Plant and Biotechnology Risk Assessment Unit of the Plant Health Science Directorate, CFIA, has concluded that the unconfined environmental release of canola event LBFLFK does not present altered environmental risk when compared to canola varieties that are currently grown in Canada.

Based on the review of the data and information submitted by BASF Canada Inc. and input from other relevant scientific sources, the Animal Feed Program of the Animal Health Directorate, CFIA, has authorized the use of canola event LBFLFK oil as a source of omega-3 long-chain polyunsaturated fatty acids for fish feeds only. Furthermore, only pre-pressed solvent extracted defatted canola meal has been authorized for use as livestock feed.

Unconfined release into the environment and use as livestock feed of canola event LBFLFK is therefore authorized by the Plant Biosafety Office of the Plant Health and Biosecurity Directorate and the Animal Feed Program of the Animal Health Directorate, respectively, as of December 9, 2019. Any canola lines derived from canola event LBFLFK may also be released into the environment and used as livestock feed with the specified conditions stated below, provided that:

• no inter-specific crosses are performed
• the intended uses are similar, and meet the conditions of the authorization
• it is known based on characterization that these plants do not display any additional novel traits, and
• the novel genes are expressed at levels similar to that of the authorized line

With respect to its unconfined release into the environment, an appropriate herbicide tolerance management plan should be implemented. If canola event LBFLFK is cultivated in Canada as an individual event or in combination with other canola events in stacked/pyramided products, BASF Canada Inc. must submit a herbicide tolerance management plan to the CFIA.

Regarding the feed authorization, canola event LBFLFK was approved as follows:

• As a result of the modified and introduced fatty acids within canola event LBFLFK and a different usage than conventional canola oil (that is, as a source of omega-3 LC PUFAs), a new ingredient definition for omega-3 long chain polyunsaturated fatty acids canola oil has been created for class 8, Part I, Schedule IV of the Feeds Regulations. In addition, BASF Canada Inc. provided information in the submission for the use of the oil derived from canola event LBFLFK seed as a source of n-3 LC-PUFAs in fish feeds. Considering this information, the new ingredient definition captures that this ingredient has only been approved for fish feeds.
• The information presented to the AFP indicated that the whole seed from canola event LBFLFK will be subjected to prepress solvent extraction resulting in a meal that contains a small fraction of oil (1%). The canola meal definition 5.3.3 in Part I, Schedule IV of the Feeds Regulations has been modified to indicate that the canola meal can be derived from whole omega-3 canola seeds of the species Brassica napus developed to synthesize omega-3 long chain polyunsaturated fatty acids (n-3 LC-PUFAs) including eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). The meal derived from seeds of the canola event LBFLFK can be combined with and used in the same way as conventional canola meal in livestock feeds.
• The safety and efficacy of the levels of omega-3 LC PUFAs in foods of animal origin such as milk, eggs and meat by feeding long chain omega-3 canola oil or whole seeds from canola event LBFLFK to livestock species other than fish were not assessed at this time. As such BASF Canada Inc. is required to use an identity preservation system to ensure that the oil derived from omega-3 canola seeds are used in fish feeds only and not fed to other livestock species.
• The disposal of the oil derived from canola event LBFLFK intended for human consumption (for example, spent omega-3 canola oil obtained from human usages) is not authorized for use in livestock species other than fish.
• No approval has been granted for use of whole seeds and forage derived from the canola event LBFLFK as livestock feed at this time.

Canola event LBFLFK is subject to the same phytosanitary import requirements as unmodified canola varieties. Canola event LBFLFK is required to meet the requirements of other jurisdictions, including but not limited to, the Food & Drugs Act and the Pest Control Products Act.

Please refer to Health Canada's Decisions on Novel Foods for a description of the food safety assessment of canola event LBFLFK.