Novel Food Information: Canola Protein Isolate and Cruciferin-rich Canola Protein Isolate

On this page


Health Canada has notified Merit Functional Foods Corporation (Winnipeg, MB) that it has no objection to the food use of a canola protein isolate (CPI) and a cruciferin-rich canola protein isolate (CRCPI) as food ingredients and alternative protein sources. The Department conducted a comprehensive safety assessment 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 Merit Functional Foods Corporation (Merit) and the evaluation by Health Canada. This document contains no confidential business information.

The term "canola" refers to species of rapeseed (Brassica napus L. and Brassica rapa L.)and mustard (Brassica juncea L.) which produce oil that is low in erucic acid and glucosinolates. Canola species belong to the mustard or cabbage family of plants.

Cruciferin and napin refer to the two major classes of storage proteins found in canola seed.

1. Introduction

Merit has developed a CPI and a CRCPI for use as alternative protein sources in food.

CPI and CRCPI are high protein (≥ 90% dry basis) powders derived from cold-pressed canola meal. The manufacturing process allows for the separation of cruciferin and napin proteins. CRCPI is enriched in cruciferin proteins such that they are the predominant proteins present (≥ 80% of total protein), whereas CPI contains significant amounts of both cruciferin and napin proteins. Health Canada has already authorized a napin-rich canola protein isolate (NRCPI) manufactured by Merit.

Merit indicated that the notified CRCPI is substantially equivalent to a CRCPI that is generally recognized as safe (GRAS) in the United States as per GRAS Notice No. 327 (GRN No. 327), with respect to which the Food and Drug Administration (FDA) issued a Letter of No Questions. The manufacturing process for both cruciferin-rich isolates is very similar except that the starting material in GRN No. 327 is heat- and solvent-treated canola meal versus cold-pressed canola meal for the notified isolate. As such, Merit indicated that the notified CRCPI will largely conform to the specifications of the CRCPI presented in GRN No. 327. Merit also indicated that the NRCPI presented in GRN No. 327 is substantially equivalent to the previously notified NRCPI which has already been authorized by Health Canada. GRN No. 327 also proposes that the napin-rich and cruciferin-rich canola protein isolates may be used alone or together as ingredients in the same food. As such, Merit indicated that information presented in GRN No. 327 can be used to support the safety of the notified CRCPI and the notified CPI.

A safety assessment was conducted by Food Directorate evaluators 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 the development of CPI and CRCPI, the specifications and manufacturing methods, their intended use, nutritional composition, microbiological and toxicological information, as well as the presence of potential allergens. Merit Functional Foods Corporation provided data to support that its CPI and CRCPI are safe for use as food ingredients and alternative protein sources in Canada.

The Food Directorate has a legislated responsibility for the pre-market assessment of novel foods and novel food ingredients as detailed in Division 28 (Novel Foods) of Part B of the Food and Drug Regulations. Food use of protein fractions from canola has not been reported to any great extent. Therefore, CPI and CRCPI are considered novel under the following part of the definition for novel foods: "(a) a substance, including a microorganism that does not have a history of safe use as a food".

2. Development of the Product

The raw material for the notified CPI and CRCPI is meal derived from the seeds of conventional (i.e., not genetically modified) canola (i.e., low erucic acid—low glucosinolate rapeseed) species that have undergone mechanical extrusion of their oil without the use of heat or organic solvents (i.e., cold-pressed).

The notified CPI and CRCPI are manufactured by mixing cold-pressed canola meal with salt water at a specific temperature and pH. Insoluble materials, including residual oil, are then removed by decantation. The resulting solution is concentrated by ultrafiltration, which reduces contaminant and anti-nutrient levels. The concentrated protein solution can be heat-treated, spray dried, and sieved to form the final CPI. Alternatively, the concentrated protein solution can be heat-treated, cooled and diluted causing the cruciferin fraction to precipitate as micelles, which are pelleted out by centrifugation. The pellets are then re-solubilized, spray dried and sieved to form the final CRCPI.

3. Dietary Exposure

Merit estimated mean and 90th percentile intake of CPI and CRCPI for various age- and sex-groups based on data collected in the United States as part of the National Health and Nutrition Examination Survey (NHANES) 2013-2014. The dietary exposure estimates were prepared with the understanding that current protein sources in foods will be replaced with either CPI or CRCPI, and therefore, are reflective of collective exposure to both of the notified protein isolates. The exposure assessment assumed ingredient use of CPI and CRCPI in a wide variety of foods at levels ranging from 2-35%. Adolescent males (12-19 y) were estimated to have the greatest per capita intakes with mean and 90th percentile values of 33.43 and 58.85 g/day, respectively. Children (3-11 y) were estimated to have the greatest intake expressed on a per kilogram of body weight basis with mean and 90th percentile values of 0.84 and 1.44 g/kg bw/day, respectively.

Merit indicated that there is a high degree of similarity in the reference amounts for common foods between Canada and the United States, and indicated that market research found that Americans and Canadians reported very similar preferences for plant-based proteins. It is noted that the estimated intake of CPI and CRCPI exceeds those determined for soy protein from all food sources using data from the 2004 Canadian Community Health Survey—Nutrition (Cycle 2.2). Footnote 1 Furthermore, the mean estimate for adults exceeded mean intake of protein by adults from all plant-based foods as determined using data from the 2015 Canadian Community Health Survey—Nutrition. Footnote 2 Thus, the exposure assessment is unlikely to underestimate actual exposure to CPI and CRCPI.

4. Chemistry

Merit analyzed multiple batches of the notified CPI and CRCPI for trace elements (arsenic, cadmium, lead, and mercury), 11 mycotoxins, two polycyclic aromatic hydrocarbons, dioxins, and 17 pesticides deregistered in Canada and therefore considered to be environmental contaminants. The majority of these contaminants were present at either very low concentrations or were "not detected". The analytical limits of detection (LOD) employed for all contaminants evaluated were deemed to be suitably low. Using either the quantified analytical results or assuming the LOD values, it was determined that addition of CPI or CRCPI to the specified foods would have a negligible impact on total dietary exposure to these contaminants.

It was noted that CPI and CRCPI are discreet ingredients that exist independent of any processing aids, additives, or other substances that may be used to manufacture them. Use of substances to manufacture the notified protein isolates must comply with relevant legal provisions where they exist, and must not violate Section 4 of the Food and Drugs Act.

5. Microbiology

The temperature and filtration steps of the manufacturing process help to ensure the microbiological safety of CPI and CRCPI. To demonstrate compliance with microbiological specifications, Merit provided data on three non-consecutive production lots of the notified protein isolates. The various microbiological tests, the tolerance limits, and methodologies were reviewed as they pertain to microbiological safety. The microbial analysis was conducted using validated methods from Health Canada's Compendium of Analytical Methods, and the specifications were considered acceptable. The results of the batch testing demonstrated that the microbiological specifications for the finished canola protein isolates were met consistently in a production run.

6. Nutrition

The reported proximate and amino acid composition of the notified CRCPI is considered substantially similar to the CRCPI presented in GRN No. 327.

Merit provided the Protein Digestibility-Corrected Amino Acid Score (PDCAAS) and the adjusted Protein Efficiency Ratio (PER) of the notified isolates as measures of protein quality. Both the PDCAAS and PER were less than values for casein. However, the protein quality of both the notified CPI and CRCPI is within the range of common food proteins, and better than some other plant-based proteins.

As CPI and CRCPI are incomplete proteins of lower quality than casein, use as the only sources of dietary protein would be inappropriate, similar to many other plant-based proteins. Merit proposed that CPI and CRCPI will be used as alternative protein sources as part of a mixed diet consumed ad libitum. In this context, individuals can easily meet metabolic demands by consuming a variety of food proteins, which complement each other in terms of their amino acid profile. Only in an extreme scenario, in which consumers exclude nearly all other sources of protein from their diet, is potential inadequacy possible. However, such a worst case scenario is unrealistic and implies the consumption of an unbalanced diet, which is generally not recommended.

Merit measured the total glucosinolate and erucic acid content of the notified CPI and CRCPI, which were both present at levels lower than those found in canola meal and commonly eaten foods. The erucic acid content was negligible. Merit indicated that the anti-nutrient and erucic acid composition of the notified CRCPI derived from cold-pressed canola meal would be entirely comparable to the CRCPI obtained from heat-solvent treated meal as presented in GRN No. 327. Based on the close similarities in nutritional composition and manufacturing method, this rationale was accepted for erucic acid and all measured anti-nutrients (glucosinolates, total phenolic acids, sinapic acid, phytic acid) except for the glucosinolate hydrolysis products (isothiocyanates, nitriles), whose levels are difficult to estimate and were not directly measured in the notified protein isolates.

Glucosinolate hydrolysis products are formed from glucosinolates by the action of the enzyme myrosinase when they are brought into contact with each other following the rupture of plant cell walls. Exposure to these hydrolysis products is common following consumption of foods containing glucosinolates and myrosinase (e.g., cruciferous vegetables). Seed cooking during production of canola meal normally eliminates most myrosinase activity, but is not completely effective in eliminating exposure to glucosinolate hydrolysis products due to the action of myrosinases produced by the intestinal microflora. As heat treatment of seeds prior to pressing is not part of the production process for either of the notified isolates, it is possible that glucosinolate hydrolysis products may be present in higher amounts than in isolates prepared from the heat treated canola seed, such as those presented in GRN No. 327. However, these levels are expected to be reduced by filtration steps during manufacturing. In addition, exposure to endogenously produced hydrolysis products is limited by the total amount of glucosinolates, which are present in amounts well below those found in broccoli.

Importantly, unacceptably high levels of glucosinolate hydrolysis products would be indicated by the adjusted PER. As noted previously, the adjusted PERs of the notified protein isolates were within the range of common food proteins, and were greater than that of some other plant-based proteins. Thus, the potential nutritional hazards posed by glucosinolate hydrolysis products in the notified protein isolates are not a concern, as they are ultimately indistinguishable from the effects of eating any one of many non-novel food proteins of similar or worse quality.

7. Toxicology

To support safety of the notified CPI and CRCPI, Merit primarily referenced studies presented in GRN No. 327. The test articles for these studies were cruciferin-rich and napin-rich canola proteins isolates that Merit indicated were substantially similar to the notified CRCPI and the previously authorized NRCPI. The previously authorized NRCPI is mentioned here as it pertains to the safety of the notified CPI.

The cold-pressed method used by Merit for removing oil from the canola meal for the notified isolates omits the heat treatment and hexane extraction steps that are used for the protein isolates presented in GRN No. 327. Therefore, in order to bridge the safety data from GRN No. 327, Merit demonstrated substantial similarity between the canola protein isolates obtained from both production methods. Similarity was based on the constituents present (i.e., protein, moisture, ash); the proportions of those constituents (e.g., both the GRAS and notified formulations contained ≥ 90% protein); and the ratio of the storage protein (i.e., both the GRAS and notified formulations contained ≥ 80% napin for NRCPI, and ≥ 80% cruciferin for CRCPI). Therefore, the data from GRN No. 327 may be used to support the safety of the notified CRCPI and of the previously notified NRCPI.

No genotoxicity was observed for either CRCPI or NRCPI in an Ames assay, an in vivo bone marrow micronucleus assay, or an in vitro mammalian cell gene mutation test (mouse lymphoma assay).

In a well-conducted 90-day oral (dietary) rat study of CRCPI, statistically lower body weight gains were observed in the mid- and high-dose groups for males, and in the high-dose group for females; however, body weights (in all dose groups) were stated to be within historical control ranges. These reduced body weight gains were associated with reduced food consumption, due to the poor palatability of the dietary test material. Animals were otherwise determined to be in good health. The highest dose tested was considered to be the no-observed-adverse-effect level (NOAEL), and is 11.24 g/kg bw/day (males, most sensitive sex).

Similar effects were observed for an identical study of NRCPI. The NOAEL was also the highest dose tested (12.46 g/kg bw per day; males, most sensitive sex); however, to be conservative, the aforementioned NOAEL for CRCPI (i.e., 11.24 g/kg bw/day) was used as the point of departure for determining margins of exposure (discussed below).

The results of these toxicological studies also apply to CPI given that it is a combination of the napin and cruciferin canola proteins that are individually enriched in NRCPI and CRCPI following a similar manufacturing process.

Merit also noted that canola species are bred to be low in erucic acid and glucosinolates. Merit submitted data for the notified CPI and CRCPI demonstrating that levels of glucosinolates and erucic acid are lower than levels found in common foods in the diet, and were comparable to levels present in the test articles from the aforementioned 90-day oral toxicity studies.

Merit estimated a maximum dietary exposure of 1.44 g/kg bw/day, observed for the 90th percentile in children (3-11 years old). This is approximately 10-fold lower than the NOAEL obtained for the 90-day oral toxicity study in rats (i.e., 11.24 g/kg bw/day). This margin of exposure was determined to be sufficient, given the lack of toxicity in the safety studies, as well as the lack of genotoxicity. It was noted that other jurisdictions (Food Standards Australia New Zealand and the European Food Safety Authority) have approved canola protein isolates based on higher estimated maximum dietary exposures, ranging from 2.28 to 4.73 g/kg bw/day.

8. Allergenicity

Merit provided data that showed substantial homology between canola cruciferin protein and mustard protein (up to 90% over the full protein length). This degree of similarity between these two proteins suggests that some individuals who have an allergic reaction to mustard are likely to also react to canola cruciferin protein. That is, there could be allergenic cross-reactivity between the two proteins. To mitigate the risk for mustard allergic consumers, Merit will label its CPI and CRCPI with a cautionary statement indicating that the notified protein isolates may be unsuitable for individuals with mustard allergy.

Merit also provided data that showed some homology between canola cruciferin protein and proteins from different tree nuts (almond, hazelnut, cashew, pecan, pistachio and walnut). The upper values for homology ranged from 61-66% over a sliding 80 amino acid (AA) window and 46-48% over the entire protein sequence for the various tree nuts.

Historically, Health Canada has relied on criteria first developed by a joint Food and Agriculture Organization of the United Nations (FAO) / World Health Organization (WHO) expert consultation to identify potential allergic cross-reactions Footnote 3 . This guidance states that any homology with a known allergen equal or greater than 35% over a sliding 80-AA window is considered to indicate significant homology and that further investigation, such as serum screening, should be completed.

It has been suggested that cross-reactivity is rare below 50% homology over the entire protein sequence Footnote 4, and that the 35% cutoff over 80 AA is conservative, resulting in an overestimation of cross reactivity Footnote 5 .

Merit provided an opinion from subject matter experts indicating that the homology between cruciferin and tree nuts was not likely to represent a risk to tree nut allergic consumers. The opinion was based on the use of a threshold for cross reactivity of 50% over the entire protein sequence and the lack of any good evidence in the allergy literature that mustard allergic consumers experience allergic cross reactivity to tree nut allergens. The subject matter experts assumed that if people with tree nut allergy would cross-react to canola then that would also mean that consumers with mustard allergy would cross-react to tree nuts, something that has not been observed in consumers with mustard allergy. The subject matter experts concluded that there is no need to alert tree nut allergic consumers about possible cross-reactivity with canola protein isolates.

Merit also provided market data from the United States regarding current use of their canola protein isolates as ingredients in several products. These data indicated that Americans have been exposed to a significant number of individual servings of cruciferin protein from Merit's NRCPI and CPI. None of these currently marketed products include tree nuts as ingredients or are otherwise labelled in a way that would discourage tree nut allergic consumers from eating them. On the contrary, some specifically says "no nut flours" on the label, and are described on the manufacturer's website as being "free of most common allergens…" including specifically, "no nuts."

Given the volume of products that have been sold with some being advertised as "nut free", or as having "no nut flours", there is reason to believe that, in the United States, persons with tree nut allergy have been exposed to canola cruciferin protein by consuming food containing Merit's CPI and NRCPI. Merit has not received any complaints or reports of adverse reactions in consumers with tree nut allergy since the products have been on the market.

In summary, the information provided by Merit supports that there is low risk of cross-reaction for persons with tree nut allergy. Consequently, labelling of CPI and CRCPI with a cautionary statement for tree nut allergic consumers is not warranted.

Due to concerns regarding potential cross-reactivity between canola and mustard allergens, Health Canada intends to develop educational material for persons with mustard allergy about the possibility of reacting to canola proteins, which will be published online and circulated to food allergy associations.


Health Canada's review of the information presented in support of use of CPI and CRCPI as food ingredients and alternative protein sources has concluded that there are no food safety concerns for the general population. To address the allergenic risk to mustard allergic individuals, it is expected that CPI and CRCPI will carry a label statement about this risk. Furthermore, Health Canada will update its webpages to inform consumers that canola protein should not be consumed by those with mustard allergies, and will proactively inform food allergy associations of this opinion.

It is the continuing responsibility of the manufacturers, sellers and distributors of CPI and CRCPI to ensure that marketed products are in compliance with all statutory and regulatory requirements, including labelling and advertising.

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


Footnote 1

Mudryj et al. (2015). Br J Nutr 113:299-309.

Return to footnote 1 referrer

Footnote 2

Auclair & Burgos. (2021). Appl Physiol Nutr Metab 46:501-510.

Return to footnote 2 referrer

Footnote 3

FAO/WHO. (2001). Evaluation of Allergenicity of Genetically Modified Foods. Rome: FAO.

Return to footnote 3 referrer

Footnote 4

Aalberse RC. (2000). J Allergy Clin Immunol 106(2):228-238.

Return to footnote 4 referrer

Footnote 5

Abdelmoteleb et al. (2021). Food Chem Tox 147:111888.

Return to footnote 5 referrer

Page details

Date modified: