Transport engineering as a strategy to realize rapeseed's potential as a protein‐rich food

Abstract

Rapeseed is the world's second‐largest oilseed crop, and the low‐value press cake that remains after oil extraction contains 30–40% protein with an excellent amino acid composition; however, it is unexploited as food owing to its content of anti‐nutritional glucosinolates. The potential of rapeseed press cake as source of plant‐based protein for human consumption faces a food and a farming challenge. We suggest using molecular breeding based on transport engineering of glucosinolate transporters to prevent the accumulation of glucosinolates, specifically in the seed, as a strategy that can enable the raw, unprocessed rapeseed press cake to be used as food (and improved feed), and potentially also improve the crop's inherent pest and disease resistance. © 2025 The Author(s). Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

The heavily pesticide‐dependent rapeseed is an under‐utilized protein source for food owing to anti‐nutritional glucosinolates. Transport engineering technology may reduce glucosinolates in seeds while increasing defense in vegetative parts.

1. INTRODUCTION

The green transition demands a shift in the global diet from consumption of animal‐based protein to plant‐based protein. ¹ This shift requires multiple sources of plant‐based proteins, of which rapeseed is an existing, yet underutilized protein source. Today, rapeseed is grown as an oil crop and the rapeseed press cake that remains after the oil is pressed out of the seeds, is a low‐value sidestream despite its content of 30–40% protein with good amino acid composition. The rapeseed press cake is used as protein feed primarily for ruminants and to a lesser degree for non‐ruminants, such as chicken and pigs. The latter is a consequence of the goitrogenic (and therefore growth inhibitory) activity of specific glucosinolate hydrolysis products in the press cake.

Realization of the potential of rapeseed press cake as a source of plant‐based protein for human consumption has two major challenges. One is the food challenge related to the press cake's content of anti‐nutritional glucosinolates. The other is the farming challenge related to the serious difficulties farmers face in mitigating damages from pathogens, birds, slugs and the many insect pests of brassicaceous plants, ² which has made rapeseed farming heavily pesticide‐dependent. ³ Here, we suggest applying a transport engineering strategy in breeding to address the food and possibly also the farming challenge.

2. HISTORY OF RAPESEED

Rapeseed (Brassica napus L.) originated as a hybridization event between (domesticated) Asian cabbage or turnip (Brassica rapa L.) and Mediterranean cabbage (Brassica oleracea L.) a few thousand years ago. ⁴ It was not, however, until the second half of the 20th century, when breeders developed new rapeseed varieties with higher seed quality that large‐scale farming of rapeseed picked up. First, breeding resulted in reduced levels of the unhealthy eruca acid in lines referred to as single‐low. In the late 1970s, the development of the Bronowski cultivar with reduced levels of anti‐nutritional glucosinolates in the seeds resulted in the generation of the double‐low lines with ‘canola quality’ seeds low in both eruca acid and glucosinolates. ⁵ Although rapeseed is primarily grown for its oil that is used for consumption and biofuel, the reduction of glucosinolates in seeds enabled the press cake to be used as protein‐rich feed, thus increasing the commercial value and contributing to an increase in rapeseed farming worldwide. ⁶

2.1. Glucosinolates and their hydrolysis products

Glucosinolates are amino acid‐derived defense compounds containing a thioglucose and an organic sulfate moiety and they are uniquely characteristic of the Brassicales order. More than 130 different glucosinolates derived from few precursor amino acids (mainly methionine, tryptophan and phenylalanine) have been identified in nature. ⁷ The high chemical diversity of glucosinolates arises from extensive modifications of the amino acid‐derived side chain. ⁷ Glucosinolates are part of a two‐component defense system and upon tissue disruption, such as that resulting from chewing insect activity, glucosinolates are hydrolyzed by otherwise spatially‐separated myrosinases (i.e. thioglucoside glucohydrolases). The resulting aglycone is unstable and degrades rapidly to biologically active hydrolysis products, of which the major ones are isothiocyanates and nitriles. ⁸ Many hydrolysis products can be further metabolized to inorganic thiocyanate ions in the animal body. ⁹ , ¹⁰

The Brassicaceae family includes many well‐known vegetables (e.g. cabbages) and oilseed‐producing plants (e.g. rapeseed). They differ in their glucosinolate profiles, but are typically dominated by aliphatic (methionine‐derived) glucosinolates and to a lesser extent indolic glucosinolates. ¹¹ The diversity in glucosinolate profiles gives rise to different profiles of hydrolysis products and thus different biological activities. ⁸ The effect of individual glucosinolates and their hydrolysis product on insects and pathogens is species‐dependent. For example, the isothiocyanate hydrolysis products from many different glucosinolates have been reported to be toxic to a range of generalist pests, ¹³ , ¹⁴ but the same compounds are readily detoxified by specialist pests adapted to feed on glucosinolate‐containing plants. ¹² For humans, some glucosinolates are well‐known as condiments due to the pungent flavor of their hydrolysis products (e.g. sinigrin in wasabi), some are associated with health promotion (e.g. glucoraphanin in broccoli), whereas others are anti‐nutritional owing to goitrogenic properties (see Section 3 below). ¹³ , ¹⁴

3. THE FOOD CHALLENGE ‐ GOITROGENICITY OF BRASSICACEOUS CROPS

In animals, goitrogenic compounds interfere with the function of the thyroid gland and can cause goiter (enlarged thyroid gland). In rapeseed, the seeds are enriched with progoitrin [2‐(R)‐OH‐3‐butenyl glucosinolate] ¹⁵ that upon hydrolysis gives rise to an isothiocyanate, which is further converted via intramolecular cyclization to goitrin (L‐5‐vinyl‐2‐thiooxazolidone). As the name implies, goitrin has goitrogenic activity ¹³ , ¹⁴ and affects the metabolic rate of the whole body.

The thyroid hormones T3 and T4 produced by the thyroid gland have, respectively, three and four iodine atoms covalently bound in their structure. ¹⁶ Goitrin inhibits thyroid peroxidase, a key enzyme in the incorporation of the iodine atoms in the biosynthesis of T3 and T4 hormones. ¹⁶ Inorganic thiocyanate ions (derived from isothiocyanate and nitrile hydrolysis products of many different glucosinolates; see Section 2.1 above) are competitive inhibitors of iodide uptake in the thyroid gland. ¹⁷ The goitrogenic effect of inorganic thiocyanate ions can be alleviated by adequate supply of iodine in the diet, whereas this is not the case for goitrin. ¹⁷ , ¹⁸ Consequently, to alleviate the food challenge related to the goitrogenic effect of glucosinolate hydrolysis products in the rapeseed press cake, it is key that the level of progoitrin (giving rise to goitrin) is reduced in the seeds either through breeding or post‐harvest processing.

3.1. Novel food status of rapeseed products in the EU

Novel foods in general include raw plant parts as well as isolates and products produced from plant parts that have not been used for human consumption to a significant degree within the EU before May 1997. ¹⁹ Currently, three EFSA novel food approvals exist for food products based on rapeseed press cake: (i) an isolate with ≥90% protein, and (ii) a partially defatted powder of rapeseed press cake with 28–43% protein and (iii) a rapeseed protein‐fibre concentrate with 28–45% protein. ²⁰ , ²¹ , ²² These three products have a total glucosinolate content of <1, <0.3 and 0.1 mmol kg⁻¹ for, respectively, protein isolate, press cake powder and protein‐fibre concentrate, ²⁰ , ²¹ , ²² compared to ~10–25 mmol kg⁻¹ in raw, unprocessed press cake of canola quality seeds. ²³ Recently, Ferm Food ApS (ferm-food.com) obtained an amendment of the existing EFSA novel food approval on partially defatted powder of rapeseed press cake, ²¹ which now also permits fermentation of the rapeseed press cake. ²⁴ Interestingly, the specifications for glucosinolates and phytic acid content are unchanged in the amended novel food approval of fermented press cake, but have a lower limit for protein and lipid content, and an upper limit for carbohydrate, fibres, water content and ash. These changes are expected to allow for a wider range of suitable raw materials.

As prescribed by the EFSA rules, production process, compositional data, type of food products and daily dose have to be specified along with an estimation of the expected intake of the novel food in the different population groups. ¹⁹ Thus, an EFSA approval authorizes use of a novel food only in the food products listed in the application and within a maximum dose. Thereby EFSA‐approved novel foods differ from other food products in that the amount of a novel food in a food product is regulated. For example, the EFSA‐approved partially defatted powder of rapeseed press cake can be used in meat imitates, but it must not exceed 10% of the total weight of the product. The maximum allowable content of rapeseed press cake in any of the food products is 20% (i.e. 20 g per 100 g), and this is in cereal and muesli products. ²¹

A notable example of an EFSA rapeseed novel food application that was not granted is the use of 4% whole seeds of rapeseed in gourmet breads. ²⁵ The seeds were of a double‐low rapeseed variety with ≤15 mmol glucosinolates/kg seeds. The applicant was requested to provide data from a human intervention study to determine the impact of consumption of goitrogenic glucosinolates on iodine status resulting from consumption of the bread with whole rapeseed seeds at the proposed levels, in addition to intakes of glucosinolates from background diets. ²⁵ This study was not provided and thus the food product not approved.

3.2. The potential of rapeseed press cake

In the EU, consumption of protein in food amounted to ~9.8 million tonnes (Mt) animal‐based protein and ~7.2 Mt plant‐based protein in 2021. ²⁶ Interestingly, in the same year EU production of rapeseed protein used for feeding EU livestock amounted to ~2.8 Mt, ²⁷ showing that protein from rapeseed theoretically can replace over a quarter of animal‐based protein in the EU. Moreover, compared to other sources of plant‐based protein, rapeseed protein contains especially high amounts of sulfur‐containing amino acids, cysteine and methionine, of which the latter has been reported to be reduced in the serum of people with a plant‐based diet. ²⁸ , ²⁹ , ³⁰

The example with the rejection of use of 4% seeds from rapeseed in bread (see above) shows that for EFSA the total content of glucosinolates in the seeds is the primary factor blocking the use of seeds and, consequently, also the raw press cake as food ingredient. This suggests that if rapeseed varieties with none or very few glucosinolates in the seeds could be developed through breeding, then the seeds as well as the raw rapeseed press cake could obtain EFSA novel food approval.

Why not just go with protein isolate and defatted press cake powder already approved by EFSA today? It is great that these products have already made it to the emerging market of rapeseed food products as a source of plant‐based protein. However, the raw, unprocessed rapeseed press cake offers unique opportunities as a commodity protein source to be used as a whole food ingredient for everyday cooking by the general population. From a health perspective, unprocessed food is preferable to processed food. ³¹ Additionally, greenhouse gas emissions from the preparation of processed food ingredients such as the three approved press cake‐based novel foods may mitigate any emission benefits compared to animal‐based protein sources. ³² Moreover, by making downstream processing unnecessary, the raw press cake could become a low‐cost commodity protein source for export to other less affluent markets.

4. THE FARMING CHALLENGE

A global survey by Zheng et al. ² amongst rapeseed experts showed that insect pests are currently the major biotic stress affecting rapeseed farming. In contrast to disease management, where farmers employ different management options such as crop rotations, chemical and biological control, and resistant cultivars to mitigate disease occurrence, established management options for insect pests are currently limited to pesticides. ² The use of pesticides has been linked to declines in insect diversity and, more specifically, pollinator diversity. ³ With the ban on neonicotinoid pesticides in the EU in 2013, specialist pests adapted to feeding on Brassicas (such as Psylliodes chrysocephala) have become an increasingly serious problem as resistance to the remaining permitted pesticides such as pyrethroids is building up ‐ with regional variations. ³³ , ³⁴ , ³⁵ , ³⁶ In the UK, the area of rapeseed fields is reduced by >50% since 2013 (Fig. 1). ³ , ³⁷ Thus, establishing new management options, such as breeding more pest‐resistant cultivars, is a key challenge that must be addressed to ensure sustainability of rapeseed farming in Europe and elsewhere.

Figure 1.

The farming challenge of rapeseed. (A) Following the EU‐ and UK‐wide ban of the seed‐coated neonicotinoid pesticides in 2013 (designated by the dotted line), the total area used for growing rapeseed in the UK has reduced by ~50%. The cause has been attributed to a major challenge for farmers to manage pests in their fields. ² , ³⁷ (B) Difference in field performance between neonicotinoid‐treated and untreated seeds, highlighting the critical pest pressure on rapeseed farming in EU and UK, here exemplified by a UK field. Photo courtesy of Samantha Cook, Rothamsted Research.

As introduced above, the two‐component (glucosinolate‐myrosinase) defense system deters predators through the biological activity of glucosinolate hydrolysis products. Rapeseed is plagued by both generalist pests (e.g. insects, birds, slugs, mammals) and specialist insect pests evolved to detoxify the glucosinolate‐mediated defense. ² It has been reported that plants with higher levels of leaf glucosinolates are less susceptible to some generalist pests, ³⁸ , ³⁹ but more susceptible to specialist pests. ³⁹ , ⁴⁰ The relationship between glucosinolate defense and contrasting selection pressure by generalist and specialist pests dictates the need for a balanced level of defense compounds for a given environment as predicted in mathematical models by van der Meijden and Chakraborty et al. ⁴¹ , ⁴² Besides total glucosinolate levels, evidence suggests that also the glucosinolate profile is a key trait for defense against both specialist and generalist pests. ⁴³ , ⁴⁴ Furthermore, the specialists respond species‐specifically to individual glucosinolates (for review, see Obermeier et al. ⁴⁴ ). As an example, Züst et al. ⁴³ observed that the dislike of specific glucosinolates by different specialist aphids resulted in domination of Arabidopsis ecotypes with a specific glucosinolate chemotype, underlining that glucosinolate profiles influence the presence of specialist insects.

As suggested by Mithen, ³⁸ breeding for rapeseed cultivars with glucosinolate levels and profiles fitted to provide optimal defense against the most critical herbivores and diseases could be a way to improve the plant's pest and pathogen resistance. Recent analyses of extensive and diverse cultivar collections ⁴⁵ , ⁴⁶ , ⁴⁷ demonstrate a positive correlation between leaf and seed glucosinolate concentrations, revising earlier conclusions ‐ based on limited cultivar samples ‐ that no such correlation exists. ³⁸ , ⁴⁸ , ⁴⁹ Zhang et al. attributed this correlation to breeding for low seed glucosinolate levels. These findings highlight an urgent need to: (i) conduct further research on the relationship between specialist herbivory in relation to glucosinolate profile and concentration; and (ii) develop alternative strategies for controlling seed glucosinolate levels, rather than relying solely on the genetic background of existing double‐low cultivars. Here, blocking transport of glucosinolates to the seeds could be such an alternative means of reducing seed glucosinolate levels (see Section 5 below) that should allow breeders to develop rapeseed lines with more variation in glucosinolate profile and concentration without having to compromise on the seed glucosinolate levels.

5. STRATEGY TO ADDRESS FOOD AND FARMING CHALLENGES: TRANSPORT ENGINEERING TECHNOLOGY

Transport engineering uses altered expression of transporter genes to change metabolite distribution patterns; it has emerged as a powerful technology in agriculture as transporters have proven important for controlling key traits. ⁵⁰ , ⁵¹ , ⁵² In the model plant Arabidopsis thaliana (closely related to rapeseed), we have identified glucosinolate importers (GTRs‐Glucosinolate TRansporters) ⁵³ and exporters (UMAMITs‐Usually Multiple Acids Move In and Out Transporter), ⁵⁴ and shown that they are essential for seed loading. ⁵⁵ In the model plant, it was reported that loss‐of‐function of individual umamit glucosinolate transporters did not result in smaller seeds. A reduction in seed size was, however, observed both when mutating multiple umamit exporter genes ⁵⁴ and in a gtr1 gtr2 double mutant. ⁵³ Several successful applications of glucosinolate transport engineering in brassicaceous oilseed crops such as rapeseed and Indian mustard (Brassica juncea L. Czern) have been reported, focusing on blocking transport of glucosinolates to the seeds (that has no de novo synthesis), while maintaining them in non‐seed tissues (Fig. 2). ⁴⁶ , ⁵¹ , ⁵⁶ , ⁵⁷ The polyploidy of rapeseed is a key challenge in targeted breeding. Due to the polyploid genome structure of rapeseed, each gene in Arabidopsis is expected to have six copies in rapeseed, which can exhibit functional redundancy. ⁴ , ⁵⁸ A noticeable example, however, is that loss‐of‐function of a single bnGTR2 homolog in the double‐high cultivar ZY821 resulted in 76% reduction of the seed glucosinolate levels without a change in 1000‐seed weight. ⁵⁶ This demonstrates that transport engineering via GTRs can be implemented without yield penalties when targeting a single GTR gene.

Figure 2.

Illustration of strategy for glucosinolate transport engineering. Translocation of seed‐bound glucosinolates from site of synthesis in the source tissue (i.e. funiculus and/or silique walls) to the seed requires the combined action of UMAMIT glucosinolate exporters and GTR glucosinolate importers. ⁵⁵ With the transport engineering strategy, loss‐of‐function of these transporters will specifically disrupt the transport of seed‐bound glucosinolates, thus preventing their accumulation in seeds. Implementation of transport engineering would allow for both breeding of rapeseed cultivars with very low levels of glucosinolates in the seeds and altered levels and profile of glucosinolate defense compounds in the non‐seed tissues. Rapeseed plant illustration from biorender. ⁶²

With the recent identification of the UMAMIT glucosinolate exporters, it was shown that loss‐of‐function of three glucosinolate exporters caused a ~95% reduction in seed glucosinolate levels in A. thaliana. ⁵⁴ No translation of this finding to rapeseed has yet been reported. Noticeably, Xu et al. ⁵⁴ showed that rosette leaves do not contribute to the seed glucosinolate level, and that the inflorescence was the source tissue. Moreover, the funiculus ‐ hitherto thought to just be a highway connecting the seed to the replum in the silique ‐ was shown to be a hotspot for biosynthesis of seed‐bound glucosinolates. This suggests that funiculus‐expressed transporter genes are likely to be key players in the translocation of seed‐bound glucosinolates and that they should be considered as targets in molecular breeding programs. ⁵⁵

Implementation of glucosinolate transport engineering offers a promising approach for breeding rapeseed lines with altered levels and profile of defensive glucosinolates in the non‐seed tissues and low levels of glucosinolates in the seeds, starting from either double‐low or single‐low genetic background. This strategy can readily be implemented by generating loss‐of‐function mutations in key bnUMAMIT and bnGTR homologs with modern gene editing technologies such as CRISPR‐Cas, either delivered via Agrobacterium‐mediated transformation or directly into protoplasts as ribonucleoprotein complexes (both techniques suitable for multiplex editing). ⁵⁹ , ⁶⁰ , ⁶¹ Incoming legislation in the EU on ‘new genomic techniques’ is expected to allow cultivation of transgene‐free CRISPR‐Cas‐edited plants on similar terms as conventianally bred crops.

6. PERSPECTIVES

Glucosinolate transport engineering technology is a promising, unique opportunity to address the food and possibly also the farming challenges of rapeseed. With an EFSA novel food‐approved raw press cake produced from a more pest‐ and disease‐resistant rapeseed cultivar, this crop has the potential to be not only an oilseed crop, but also a major source of high‐quality protein food in a more sustainable agriculture.

CONFLICT OF INTEREST

Some of the subject matters are covered in European patent no. EP2590996 and international patent application no. PCT/EP2023/082018. D.X. is listed as inventor of the latter whereas B.A.H. is listed as inventor of both. University of Copenhagen is listed as assignee on both documents. The other author, J.S.T., declares no competing interests.

DATA AVAILABILITY STATEMENT

Data sharing: not applicable to this article as no datasets were generated or analysed during the current study.