Navigating the old and the new of anti-nutritional factors: A comprehensive and critical exploration of sustainable protein-rich ingredients and innovative processing techniques

Navigating the old and the new of anti-nutritional factors: A comprehensive and critical exploration of sustainable protein-rich ingredients and innovative processing techniques Giovanni D’Auria a,1 , Matilde Milana b,1 , Ever Hernandez Olivas c , Sara Arranz d , Mercedes Caro d , Pasquale Ferranti a,h , Andr´ e Brodkorb c , Itziar Tueros d , Clare Mills e,f , Nesli Sozer g,* , Chiara Nitride a,** a Department of Agricultural Sciences of the University of Naples, Federico II, via Universit` a 100, Portici, (Na), 80055, Italy b Wageningen Food Safety Research-WFSR, Wageningen University & Research, the Netherlands c Teagasc, Food Research Centre, Moorepark, Fermoy, Cork, P61 C996, Ireland d AZTI, Food Research, Basque Research and Technology Alliance (BRTA). Parque Tecnol´ ogico de Bizkaia, Astondo Bidea, Edificio 609, Derio, Bizkaia, 48160 Spain e School of Biosciences, University of Surrey, Guildford, United Kingdom f Division of Immunology, Immunity to Infection and Respiratory Medicine, School of Biological Sciences, Manchester Institute of Biotechnology, University of Manchester, Manchester, United Kingdom g VTT Technical Research Centre of Finland, Espoo, Finland h Institute of Food Science & Technology, National Research Council, Via Roma 52, Avellino, 83100, Italy ARTICLE INFO Keywords: Anti-nutrients Food processing Microalgae Single-cell proteins Pulses Rapeseed Insect ABSTRACT Background: Anti-nutritional factors (ANFs) are a key consideration in the development of novel, sustainable, protein-rich ingredients, as their levels are influenced by both ingredient selection and food processing techniques. Scope and approach: This review, part of the Giant Leaps Horizon Europe-co-funded project, examines the chemical characterization, biological effects, and mechanisms of action of ANFs in a diverse range of alternative protein sources, including legumes, insects, algae, and microbial biomass. This study assesses how traditional and innovative food processing methods, such as fermentation, germination, enzymes, extrusion, affect ANF activity and the nutritional quality of alternative ingredients. Key findings and conclusions: Innovative processing can mitigate the adverse effects of ANFs while preserving or even enhancing the health-promoting properties of foods. However, limitations and inconsistencies in current analytical methods for quantifying ANFs can lead to a misrepresentation of their levels, activity, processing stability, and bioactivity, thereby impacting the nutritional quality of ingredients. Furthermore, the interactions between ANFs and the gut microbiota are considered, particularly on the production of bioactive compounds like short-chain fatty acids. In conclusion, this review underscores the critical need for further research into ANF dynamics and the development of improved analytical methods. Accurate data on ANF levels are crucial for effective safety assessments and for ensuring that alternative protein ingredients are not nutritionally inferior. Ultimately, consumer trust in the safety and nutrition of novel foods is essential for their market acceptance and for advancing the transition towards sustainable food systems that address global environmental issues. * Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (G. D’Auria), [email protected] (M. Milana), [email protected] (E. Hernandez Olivas), [email protected] (S. Arranz), [email protected] (M. Caro), [email protected] (P. Ferranti), [email protected] (A. Brodkorb), [email protected] (I. Tueros), [email protected], [email protected] (C. Mills), [email protected] (N. Sozer), [email protected] (C. Nitride). 1 these authors contributed equally to the work. Contents lists available at ScienceDirect Trends in Food Science & Technology journal homepage: www.elsevier.com/locate/tifs https://doi.org/10.1016/j.tifs.2025.105365 Received 25 March 2025; Received in revised form 2 October 2025; Accepted 4 October 2025 Trends in Food Science & Technology 165 (2025) 105365 Available online 8 October 2025 0924-2244/© 2025 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). 1. Introduction The European Green Deal promotes sustainable food systems by aiming to reduce the environmental impact of traditional animal agriculture impact. A key component, the Farm to Fork Strategy, targets “net zero” emissions, partly by encouraging a shift towards alternative proteins. This transition is supported by growing consumer interest in plantbased proteins, such as those from lentils, chickpeas, and peas, driven by environmental, health, and ethical concerns (Etter et al., 2024). Adopting a so called “flexitarian diet”, defined as reducing animal protein intake by 25 % in favour of plant-based alternatives, could lower greenhouse gas emissions by 40 % and water use by 10 % (Mariotti & Gardner, 2019). The adoption of flexitarian diets in Europe varies geographically and demographically, with rates ranging from 5 % to 26 % of the population. For instance, Greece shows the highest proportion of individuals following this dietary pattern, whereas Eastern Europe lags. Young adults are the demographic most likely to adopt this lifestyle (Kamin et al., 2024; Portugal-Nunes et al., 2023; Statista Search Department, 2024a, 2024b). Shifting from highly digestible animal proteins to alternatives necessitates that the latter possess at least an equivalent nutritional quality. Globular seed storage proteins from plants (e.g., 7S and 11S globulins) are inherently less digestible than animal proteins (e.g., casein), primarily because their rigid structures conceal cleavage sites from digestive enzymes. Furthermore, the presence of ANFs in both plant and non-plant alternative protein sources can further decrease their digestibility. ANFs, which vary in structure and molecular weight, are naturally produced by organism as a defence mechanism against biotic and abiotic stresses and pest. They are classified as non-proteinaceous (np-ANFs) and proteinaceous (p-ANFs) types and can negatively affect the health and growth of organisms that ingest them. They may act as toxins, interfering with nutrient digestion, absorption and metabolism in the gastrointestinal tract, or exert adverse metabolic effects in other tissues. Their concentration in ingredients and food products depends on environmental conditions, downstream treatments, and processing methods, like milling and fractionation (Salim et al., 2023). Notably, ANF levels may increase with rising global temperatures due to climate change (Kashyap et al., 2024). Their presence in alternative protein ingredients can also exacerbate nutritional deficiencies among the vulnerable populations, such as infants, children, and older adults, who have undeveloped or impaired gastrointestinal systems (Moughan et al., 2024). The activity of ANFs, such as trypsin inhibitors, phytates, tannins, oxalates, and glucosinolates, can be modulated throughout traditional methods like soaking or bio-processes, such as fermentation and germination. These processed are commonly used in home cooking and industrial food preparation, for example in canning legumes. Fermented and germinated ingredients have traditionally been used to produce foods like sauerkraut or malt for beer. Similarly, the preparation of legumes sprouts and fermented vegetables like kimchi, practice long established in Asia, are gaining popularity in Europe. Technological processes, including ingredients fractionation for protein isolate and concentrate production, heat treatment, debittering, and extrusion, can also influence the abundance and activity of ANFs in commercial ingredients and food products (Patterson et al., 2017). In many parts of the world, including Canada, UK and the European Union, numerous alternative protein ingredients are considered novel. Consequently, a general safety assessment is required before these foods can be placed on the market. This assessment includes an evaluation of their nutritional quality. Information on the levels and activity of ANFs in a novel ingredient forms an integral part of the data package required for this nutritional assessment. Although ANFs in plants are welldocumented, these compounds can also be present in novel ingredients produced from other types of organism and may become concentrated in fractionated ingredients. Therefore, it is crucial to understand how existing knowledge of plant-based ANFs can be applied to novel foods entering the market, such as algae, insects, and microbial biomass, as substitutes for animal proteins. New types of ANFs, unique to these sources, may also require evaluation. As part of the co-funded Giant Leaps Horizon Europe project, this review compares the role of ANFs in affecting the nutritional quality of conventional ingredients, such as alternative pulses, cereals and protein fractions, and innovative sources like unicellular organisms (e.g., microalgae, microbial biomasses), rapeseed meal, and insects is compared. The ingredients were strategically selected to represent a broad spectrum of alternative protein sources, ranging from wellestablished options like pulses to emerging ones such as microbial biomasses, microalgae, and insects This selection allows the review to address varying levels of scientific understanding, technological maturity, and regulatory readiness. Furthermore, in silico bioinformatic analyses are being conducted to examine the expression of p-ANFs in emerging ingredients and the metabolic pathways leading to np-ANF formation. The review also includes a critical evaluation of existing analytical methods for determining ANFs. The goal is to identifying gaps that could compromise their accurate assessment and to guide future research toward more reliable techniques. Finally, the review explores the interactions between ANFs and gut microbiota in shedding light on potential health implications. 2. Global perspectives on novel food regulation Many countries and regions worldwide, including the European Union (EU) member states, the UK, Canada, Australia and New Zealand, have regulations requiring that a safety assessment be conducted before a novel food can be placed on the market. This risk assessment is conducted by the European Food Safety Authority (EFSA) in the EU and the Food Standards Agency (FSA) in the UK. In Canada, the assessment is undertaken by Health Canada, while Food Standards Australia New Zealand (FSANZ) performs this role for both Australia and New Zealand. All jurisdictions provide guidance and adhere to the principals of risk assessment laid out by the CODEX Alimentarius Commission with modifications relevant to the populations whose health they are designed to protect (Brooke-Taylor & Grinter, 2023; Crevel, 2023; Taylor & Godefroy, 2023). When a whole genome sequence is available for microorganisms, a bioinformatic analysis for ANFs is also performed. Because these risk assessments focus on specific ingredients, they consider the impact of the manufacturing process on the resulting levels and activities of ANFs. Other jurisdictions may take different approaches, notably the USA, which follows a self-regulatory model. In this system, manufacturers must conduct their own safety assessment to establish a consensus among qualified experts that the substance is safe under its condition of use. Manufacturers have the option to notify the Food and Drug Administration (FDA) that an ingredient has been given “Generally Recognized as Safe” (GRAS) status. The FDA then responds either with no comments regarding the notifier’s GRAS status determination or by concluding that the notice provides an insufficient basis for such a determination. 3. Chemical characterization, mechanism of action and analysis of ANFs Integrating bioinformatics with genetic, metabolomic, and proteomic analyses serves as an initial screening method to guide in vitro assessments, and enhance the understanding of ANFs, particularly p-ANFs, in food ingredients. However, risk assessment frameworks require data on ANFs in ingredients be determined using validated detection and quantification methods, preferably performed by accredited laboratories. Existing analytical techniques are often unsuitable for this purpose, especially for novel foods. While previous work has addressed the detection of active molecules in raw ingredients (Purohit et al., 2023), the quantification of active ANFs post-production remains challenging due to insufficient extraction under physiological conditions. The G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 2 extraction of p-ANFs is especially complex, as they must retain their native structure to remain active. Protein solubility depends on factors such as primary structure, interactions within the food, and location within the matrix. Consequently, low extraction yields may underestimate the presence or potency of p-ANF. Structural barriers, such as exoskeletons of insect and the cell walls of microalgae, fungi, and plants, further hinder protein extraction. Although complex buffer systems aid exploratory proteomics analyses, they may also inactivate p-ANFs, rendering them unsuitable for physiological assessment. In contrast, npANFs do not require physiological buffers, which simplifies their quantification. In this review, we examined current methods for quantifying both pand np-ANFs as summarised in Supplementary Tables 1 and 3 We evaluate existing techniques for identifying active ANFs in ingredients and foods, focusing on those relevant to dietary transitions, including lectins, enzyme inhibitors, and phytic acids. This review also highlights the limitations of these methods when applied to food processing, which can lead to significant data misinterpretation. We complement this analysis with data mining of protein and metabolic pathway databases. Table 1 summarises the key findings for a set of analysed alternative protein ingredients from plants and exemplar insect, microalgal, and microbial species. While this table focuses on the whole source material, the main body of the review incorporates available information about its derived fractions. This distinction is crucial, as an ANF present in the whole source may be significantly enriched or lost during processing. The presence of these compounds in the whole sources and their fractions is discussed in Section 3.1 and 3.2, while the effects of processing are reviewed in Section 4. 3.1. Non-proteinaceous anti-nutritional factors (np-ANFs) 3.1.1. Saponins Saponins are plant secondary metabolites found in cereals, legumes, green leaves, and some marine organisms. These amphiphilic molecules comprise a hydrophobic aglycone, typically a 30-carbon triterpenoid or steroid skeleton, and a hydrophilic glycone with one or more sugars (Timilsena et al., 2023). The aglycone structure determines the classification of saponins as triterpenoid or steroid saponins. The steroidal skeleton has five rings and 27 carbon atoms, forming either furostanol or spirostanol structure, while triterpenoid saponins have six rings and 30 carbon atoms (Fig. 1). Both types contain various functional groups (−OH, −COOH, −CH3), which generate structural diversity influenced by both aglycone content and the sugar chain composition. The analysis of saponins from plant tissue has been comprehensively reviewed (Supplementary Table 1) (Cheok et al., 2014). Due to their amphipathic nature, saponins can disrupt cell membranes, leading to haemolysis in vitro. However, large doses are required to cause local gut mucosal inflammation. They can also act as ANFs by forming insoluble complexes with proteins, lipids, minerals like iron, zinc and calcium, influencing negatively nutrient absorption (Schreiner et al., 2022). The concentration of total saponins is notable in lentils (3.50 mg/g), faba beans (4.00 mg/g), oats (2.17 mg/g), and quinoa (24.20 mg/g) (Ahuja et al., 2015; Bhinder et al., 2021; S. Kaur et al., 2019; Sharma, 2021). Interestingly, insect meals, from species such as cricket, contain approximately 53 mg/g of saponins. These compounds contribute to the insects’ defence mechanisms, with levels varying based on rearing conditions and feed types (Siddiqui et al., 2024). Yeast and other fungi, species like Pleurotus (0.0016 mg/g) can also produce saponin-like compounds as secondary metabolites (Effiong et al., 2024), with the levels that can change depending on the fermentation condition. Saponins are also secondary metabolites of algae, where they contribute to ecological interactions and can be present up to 57.5 mg/g in Chlorella spp. (Ridlo et al., 2023). 3.1.2. Phytates Phytic acid, also known as myoinositol (6)-hexakis (dihydrogen phosphate) or InsP6, is a compound that occurs naturally in plants, animals, and soil. It typically exists as phytate, forming salts with monoand divalent cations such as potassium (K+), magnesium (Mg 2+ ), and calcium (Ca 2+ ) (Salim et al., 2023). With its six phosphate groups, myo-inositol hexaphosphate (InsP6) is the most potent form. Lower phosphorylated forms, such as InsP5 and InsP4, also contribute to chelation, although their effect diminishes with fewer phosphate groups. Consequently, the ion-chelating ability of these compounds is directly proportional to their degree of phosphorylation. This chelation of divalent cations leads reduces the bioavailability of essential minerals. Phytates can also affect the protein digestion by binding to proteins and digestive enzymes, hindering their break down into amino acids (Nissar et al., 2017) (Fig. 2). Phytate-protein interactions are pH-dependent; specifically, when the pH is below a protein’s isoelectric point, the anionic phosphate groups of phytate bind strongly to the cationic groups of that protein. Phytic acid is present in most legumes, with concentrations of approximately 13 mg/g in lentils and an average concentration of 32 mg/g in faba bean flour (Adamidou et al., 2011). The levels can vary depending on the cultivar, the environmental factors and ingredient processing. It is also found in oats (11 mg/g), and quinoa (4 mg/g) (Bhinder et al., 2021; Gołębiewska et al., 2022; S. Kaur et al., 2019). Low levels are also detected in microalgae and insects; for example, Chlorella Table 1 ANFs in alternative protein sources. The table results from a literature review, implemented with p-ANFs with a review of the UNIPROT database to gather information about protein sequences. Vicia faba Lens culinaria Avena sativa Chenopodium quinoa Brassica napus Chlorella sp Xanthobacter sp Acheta sp PROTEIC ANTI-NUTRIENTS α -amylase inhibitors – – x– – – – – Lectins $ $ –x x x x,+– Serine protease inhibitors $ $ x x x x – – Thiaminase – – – + + + – – NON-DIGESTIBLE CARBOHYDRATES Chitin – – – – – – – ■ Oligosaccharides ■ ■ – – – – – – Polysaccharides – – ■ – – – – – LOW MOLECULAR WEIGHT ANTI-NUTRIENTS Phytates ■ ■ ■ ■ ■ ■ – ■ Phenolic compounds ■ ■ ■ ■ ■ ■ – ■ Oxalates – – – ■ – – – ■ Saponins ■ ■ ■ ■ – ■ – ■ Toxic ANFs ■ – – – – ■ – ■ Legend: $ evidence at protein level, * evidence at transcript level, x inferred from homology, +protein predicted, ■ scientific literature available, - no knowledge available. G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 3 Fig. 1. Chemical structures of representative non-proteinaceous anti-nutritional factors: (a) Phytic acid, (b) Gallic acid, (c) Vicine, (d) Linamarin, (e) N-acetyl-Dglucosamine – chitin monomer, (f) Progoitrin, (g) 3-Butenyl isothiocyanate, and (h) Solanine. Fig. 2. Graphical representation of np-ANFs inhibiting a generic protein, with polyphenols (right) and phytate (left) interacting through key amino acids. The left panel of phytate shows the chelating action of bivalent cations. G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 4 spp. contains approximately 0.6 mg/g, while cricket meal contains between 0.001 up to 2.4 mg/g of phytates (Asaniyan & Ogundele, 2020; Siddiqui et al., 2024). As with other ANFs, the level of phytate in insects depends on their diet. Several methods for phytic acid quantification exist, but they have notable limitations, as comprehensively reviewed by Kahrıman et al. (2020). Although quantifying phytic acid with the Wade reagent (a ferric chloride and sulfosalicylic acid reactive) is rapid, this approach can be expensive and yields more variable results than other spectrophotometric methods, such as the Chen or Haug-Lantzsch methods. Additionally, the Wade method can be affected by the ingredient’s protein content, as iron can form complexes with proteins that subsequently precipitate and interfere with the determination (Kahrıman et al., 2020). The AOAC Official Method SM 986.11, based on Anion Exchange Chromatography (AEC), acid hydrolysis, and colorimetric phosphorus determination, is the widely accepted standard. This method separates phytic acid (InsP6), the most abundant myo-inositol phosphate in seeds (>90 %). However, minor forms of myo-inositol phosphates (i.e., InsP3, InsP4, InsP5), which can be present at high concentrations in processed samples, or those exposed to phytase, an enzyme found in animals, plants, and microorganisms, can co-elute with the InsP6-chelating form. This leads to an overestimation of the “active phytic acid”, the forms with the highest capacity to chelate divalent ions (Baruah et al., 2017; McKie & McCleary, 2016). Consequently, the simple quantification of total myo-inositol phosphates can obscure the true chelating potential of an ingredient. An alternative method involves the use of an InsP n -specific phytase following acidic extraction, which recognises phytic acid (InsP6) and the lower myo-inositol phosphate forms (i.e., InsP2, InsP3, InsP4, InsP5) (McKie and McCleary (2016). This method should improve the detection specificity of processed ingredients and foods. However, it shares a key assumption with the official AOAC method 986.11: It presumes that all released phosphorus originates exclusively from the InsP6 form. This assumption ignores potential contributions from other phosphate esters, such as the lesser-phosphorylated myo-inositol phosphates. Consequently, a critical assessment reveals that currently available methods for phytic acid determination are inadequate for validating deactivation processes, as they cannot reliably confirm the reduction of the most active, ion-chelating forms of phytate. Therefore, evaluating the true effectiveness of a deactivation method might require bioaccessibility assays. In these assays, the digestibility of the treated ingredient is compared to its untreated counterpart, and the key metric is the quantity of freely absorbable divalent ions (e.g., iron, zinc). This approach provides a direct measure of nutrient availability, rather than relying on the concentration of phytic acid itself. Regarding naturally occurring phytase, its presence could potentially serve as an indirect indicator of phytic acid in an ingredient. Although the UniProt and NCBI protein databases do not annotate phytase in Chlorella or Acheta, microbial biomass like Xanthobacter does contain phytase-like proteins. However, this does not imply or roll out the presence of an InsP6-active form in the ingredient, which still needs to be assessed analytically. 3.1.3. Phenolic compounds Although polyphenols and dietary fibres are essential components of the diet due to their antioxidant properties and positive effects on the gut microbiota, this review classifies them as ANFs. This classification is based on their potential to interact with proteins and digestive enzymes, thereby reducing protein digestibility and absorption, particularly in plant-based meat and dairy substitutes compared to their animal counterparts (Moughan, 2021). Phenolic compounds, defined by one or more hydroxyl groups attached to an aromatic ring (Fig. 1), are broadly categorised into four groups: (i) phenolic acids, (ii) flavonoids, (iii) polyphenolic amides, and (iv) other polyphenols, such as resveratrol, ellagic acid derivatives, curcumin, and rosmarinic acid (Tsao, 2010). Tannins, a water-soluble and astringent subgroup of polyphenols, are classified as either as hydrolysable or condensed. They interfere with the digestion by forming complexes with carbohydrates, proteins, and mineral ions which impedes efficient nutrient utilisation (Ojo, 2022). This interaction is driven by their ability to bind with specific amino acids residues. For instance, the astringent sensation from high-tannins foods like tea and coffee is caused by tannins binding to proline-Rich Proteins (PRPs) in saliva (Fig. 2)(Ozdal et al., 2013). Because these insoluble complexes precipitate in the mouth, their subsequent impact on nutrient absorption in the gastrointestinal track may be reduced or eliminated. Other binding mechanisms also contribute to the anti-nutritional effects of phenolics. Quinones, which are highly reactive, can form covalent bond with nucleophilic side chains of amino acids (e.g. lysine, cysteine, methionine), affecting both dietary proteins and digestive enzymes (Seczyk et al., 2019). Additionally, weaker hydrophobic interactions can occur between phenols and hydrophobic amino acids (Fig. 2). Despite these effects, tannins can also offer prebiotic benefits. Their ability to associate with dietary fibre makes them resistant to digestion, allowing them to reach the large intestine where they can serve as a substrate for gut microbes (Rana et al., 2022). Phenolic compounds, including phenolic acids and flavonoids and tannins and condensed tannins, are present in most emerging alternative proteins ingredients. They are found in legumes, cereals and rapeseeds as well as in microalgae, crickets and edible fungi (S. Kaur et al., 2019; Lin et al., 2022; J. Lu et al., 2024; Siddiqui et al., 2024). In faba beans and lentils, the predominant forms of phenolic acids are hydroxycinnamic compounds, while procyanidins (a type of condensed tannin) are the main flavonoids (Y. Lu et al., 2018; Singh et al., 2017). Catechins are present in both, though epicatechin and quercetin are especially abundant in faba beans (Y. Lu et al., 2018). Lentils also contain flavan-3-ols, dihydrochalcones, stilbenes, and gallates (Singh et al., 2017). Quinoa contains high levels of quercetin and kaempferol, which are ubiquitous in many fruits and vegetables (Pereira et al., 2020). Avenanthramides are a group of phenolic alkaloids found mainly in oats (Soycan et al., 2019). Rapeseed primarily contains sinapine, sinapic acid, and sinapoyl-glucosides (Zhang et al., 2022). Tannins concentrations vary significantly across these sources. Reported levels include faba bean (13 mg/g), lentils (9 mg/g), oats (3 mg/ g), quinoa (0.3 mg/g), the microalgae Chlorella spp. (3 mg/g), and rapeseed hulls (up to 0.02 mg/g) (Adamidou et al., 2011; Asaniyan & Ogundele, 2020; F. Chen et al., 2022; S. Kaur et al., 2019; Y. Li et al., 2024; Siddiqui et al., 2024). Insects, including crickets, also contain tannins, with cricket meals containing 0.06 mg/g (Asaniyan & Ogundele, 2020; Siddiqui et al., 2024). 3.1.4. Oxalates Like phytates, oxalates are chelating agents that inhibit the absorption of minerals such as calcium, iron, and zinc by forming insoluble complexes. Oxalic acid is a dicarboxylic organic acid characterised by its low molecular weight, high acidity, and strong chelating and reducing abilities (Karamad et al., 2019). Oxalates are particularly abundant in quinoa (2.3 mg/g) and can be found in insects, with concentrations in cricket meals ranging from 0.1 up to 9 mg/g (Asaniyan & Ogundele, 2020; Siddiqui et al., 2024). 3.1.5. Glucosinolates and isothiocyanates Glucosinolates (formerly known as thioglucosides) are sulphur-rich, anionic secondary metabolites. They are hydrolysed by the enzyme myrosinase into bioactive compounds, including isothiocyanates, nitriles, goitrin, and thiocyanates (Fig. 1) (Hanschen & Schreiner, 2017). Glucosinolates are classified into three main groups based on their amino acid precursors: aliphatic (derived from amino acids like alanine, leucine, isoleucine, valine, and methionine), indole (from tryptophan), and aromatic (from phenylalanine or tyrosine) (Bell et al., 2018; Ishida et al., 2014). The taste and flavour of Brassicaceae vegetables are significantly G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 5 shaped by glucosinolates and their hydrolysis products, which can impact consumer palatability. Many intact glucosinolates, including sinigrin, gluconapin, and progoitrin, are associated with a bitter taste. The hydrolysis of progoitrin produces goitrin, which is noted for its extreme bitterness. In contrast, the characteristic pungent or "hot" sensation is caused by isothiocyanates. For instance, allyl isothiocyanate (derived from sinigrin) is responsible for the sharp taste of mustard and horseradish, while raphasatin (from dehydroerucin) contributes to the pungency of radish. However, not all these compounds impart strong flavours; glucoraphanin, for example, is considered having little perceptible taste (Bell et al., 2018). Among the hydrolysis products, thiocyanates are major goitrogens that compete with iodine for thyroid uptake, potentially causing goitre and hypothyroidism, especially with low iodine intake. Isothiocyanates, containing a N=C=S group, exhibit cytotoxic effects that can also disrupt thyroid function. Additionally, glucosinolates may reduce the availability of essential mineral, including copper and selenium (Akram et al., 2021). Glucosinolates are widely present in Brassica species, such as cabbage, broccoli, rapeseed, mustard, Brussels sprouts, and horseradish. Their concentration in de-oiled rapeseed seeds can be as high as 150 mg/ g, compared to 4.2 mg/g in commercial canola and rapeseed meals (Clarke, 2010; Gołębiewska et al., 2022; Prieto et al., 2019). The most abundant glucosinolates in rapeseed meal including progoitrin, gluconapin, glucobrassicanapin, glucoraphanin, sinigrin and gluconasturtiin (Xie et al., 2022). While a predicted pathway for glucosinolate formation in Chlorella is available in the Kyoto Encyclopaedia of Genes and Genomes (KEGG) database (pathway: cvr00966), key enzymes for this pathway are absent or unannotated in currently available genomic sequences. Despite their potential negative effects, glucosinolates also offer health benefits, including anti-cancer, anti-inflammatory, and antioxidant properties. They may help reduce the risk of cancers, such as prostate cancer, by inhibiting cancer cell growth and reducing inflammation (Akram et al., 2021). 3.1.6. Non digestible carbohydrates Non digestible carbohydrates (Fig. 1) can hinder the digestive enzymes by increasing the viscosity of the chyme and binding proteins through weak interaction (Moughan, 2021). The concentration of these non-digestible carbohydrates varies significantly across protein ingredients: flours typically contain significant amounts of fibre, protein concentrates have low fibre content, and isolates contain virtually none. In alternative novel ingredients, the composition of non-digestible carbohydrates is also diverse. For instance, crickets contain chitin (Kipkoech, 2023), while the alga Chlorella contains a blend of cellulose, hemicellulose and pectin (Zanchetta et al., 2021). For ingredients derived from microbial biomass, these carbohydrates, encompassing both soluble and insoluble fibres, can constitute a substantial proportion of the final product. They range from plant cell walls and insect exoskeletons. Both plant-derived fibres and chitin are known to interact with the mucous layer covering the enterocytes (Collado-Gonz´ alez et al., 2019; Meldrum et al., 2017). Polysaccharides like β-glucans are found in oats at an average of 16 mg/g (S. Kaur et al., 2019), while legume cell walls are rich in pectic polysaccharides and xyloglucans (Jarvis et al., 2003). Chitin and its deacetylated form, chitosan, are present in insects; crickets, for example, contain up to 49.8 mg/kg (fresh weight) and 137.2 mg/kg (dry weight) (Magara et al., 2021). These are complemented by α -galactosides, which are found in legumes such as lentils and include raffinose (0.42 mg/g), stachyose (1.87 mg/g) and verbascose (0.49 mg/g). Their levels can be almost tenfold higher in faba beans, with a concentration of 3.90 mg/g for raffinose, 13.70 mg/g of stachyose and 15.00 mg/g of verbascose (Mayer Labba et al., 2021; Wang & Daun, 2006). Since human enzymes cannot digest these compounds, they are fermented by gut bacteria in the colon, producing gases like methane and hydrogen that cause gastrointestinal discomfort. This is especially problematic for short-chain oligosaccharides known as FODMAPs (fermentable oligosaccharides, disaccharides, monosaccharides, and polyols). When consumed in sufficient quantities by susceptible individuals, FODMAPs can cause gastrointestinal distress, leading some to adopt a low FODMAP diet (Algera et al., 2022). 3.2. Proteinaceous anti-nutritional factors (p-ANFs) Understanding the mechanism of action of ANFs is key to assessing their impact on nutrient absorption. This section details the classification of ANFs based on their structure and function. Supplementary Table 2 outlines their structural features and inhibition activity, while Supplementary Table 3 summarises quantitative and accredited reference methods for their analysis. 3.2.1. Serine-protease inhibitors Protease inhibitors (PIs) are proteins that prevent harmful protease activity during seed dormancy. While their role in plant stress tolerance is well-established, their involvement in other physiological processes remains less understood. In their active form, PIs can interfere with human digestive enzymes and impair macronutrient absorption (Kårlund et al., 2021). Plant seeds and tubers contain inhibitors of trypsin, chymotrypsin, carboxypeptidases, elastase, and other serine proteases. This inhibition can be reversible or irreversible; for instance, suicide inhibitors function by being cleaving the active site peptide (Fig. 3). However, most serine PIs operate through a tight-binding, reversible mechanism (Vorster et al., 2023). For example, trypsin and chymotrypsin inhibitors have been reported of 4.47 and 3.56 IU/mg, respectively, in raw faba beans (Alonso et al., 2001). The extraction and analysis of various PI isoforms slight differences in protein length, amino acid sequence, specificity, and structural stability following heat treatment. Protease inhibitors are classified by molecular weight into two major categories: Kunitz-type and BowmanBirk inhibitors (BBIs). Beyond inhibiting trypsin and chymotrypsin, BBIs also target elastase, cathepsin G, chymase, protein kinase B, and PI3K. Clinical trials and in vitro studies have demonstrated the effectiveness of BBIs against tumour cells, highlighting their potential as a therapeutic agent in cancer treatment. Research indicates that BBIs exhibit antiproliferative and pro-apoptotic effects on various cancer cell lines, including breast, prostate, and colorectal cancers (Gitlin-Domagalska et al., 2020). The protease inhibitory properties contribute to the modulation of cellular signalling pathways associated with tumour progression and metastasis, supporting the ongoing investigation of BBI as a promising bioactive compound in oncology. BBIs are notably stable to heat and gastric juice and are resistant to proteolytic enzymes, including pepsin and pronase (Clemente, 2014). In a pioneering 1973 study, Odani and Ikenaka showed that a soybean BBI treated with cyanogen bromide and pepsin yielded two active fragments, one inhibiting trypsin and the other inhibiting chymotrypsin, thus demonstrating the dual and distinct action of this protease inhibitor (Odani & Ikenaka, 1973). Similar BBIs exist in lima beans, garden beans, azuki beans, mung beans, groundnuts, and chickpeas, as well as in non-legume sources, such as wheat germ and rice. Interestingly, although chickpea BBI treated with cyanogen bromide and pepsin exhibits features comparable to soybean BBI, the proximity of its double-headed trypsin-chymotrypsin inhibitory sites prevents simultaneous inhibition activity in its active form (Clemente, 2014). In dicotyledonous plants, BBI proteins typically have a conserved ’double-headed’ structure. In contrast, most monocotyledonous (monocot) plant BBIs lack the conserved cysteine residues required for chymotrypsin inhibition in the second inhibitory loop, suggesting that monocots possess only BBI-trypsin inhibitory activity (Xie et al., 2021). These examples highlight the high complexity of specificity and stability within the same family of inhibitors. Serpins (SERin Protease INhibitor) are a well-conserved superfamily of proteins widespread throughout all kingdoms of life. They have the demonstrated ability to inhibit both serine and cysteine proteases. The inhibitory function of serpins has been associated with the regulation of endogenous proteinases (e.g., controlling cell death), and defence G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 6 against exogenous proteases (i.e., pest control) (Jamal et al., 2013). LR-Serpins inhibit both serine and cysteine proteases through a unique “suicide-like” substrate inhibitory mechanism, which has been described in the Supplementary Material 1. Enzyme inhibition tests are used to directly assess active enzyme inhibitors, making protein extraction under physiological conditions essential for accurate evaluation. Techniques such as high-pressure homogenisation or bead milling can enhance extractability from recalcitrant tissues. Furthermore, testing inhibitory activity before and after thermal denaturation (e.g., by autoclaving) helps confirm that the observed inhibition is specifically related to the enzyme inhibitor. 3.2.2. Lectins Lectins are a family of carbohydrate-binding proteins found ubiquitously in plants, animals, and microorganisms. Their wide-ranging biological functions have been comprehensively reviewed by others (Elumalai & Lakshmi, 2021). In plants, lectins are present in the edible parts like seeds, bark, and leaves, where they serve as defensive tools against insects and pathogens. They also facilitate protein storage, maintain seed dormancy, and contribute to carbohydrate metabolism. (Dias et al., 2015). Some lectins interact with carbohydrates in plant cells (endogenous targets), while others recognise carbohydrates on exogenous targets like pathogen surface glycans (De Coninck & Van Damme, 2021). The concentration of lectins varies significantly among edible seed. For instance, kidney beans have the highest concentration at up to 10 mg/g, whereas soybeans, faba beans, and lentils contain a lower amount of approximately 1 mg/g. Cereals generally have the lowest levels, around 0.5 mg/g (Pryme & Aarra, 2021). Depending on environmental factors and the plant’s developmental stage, lectins in seeds can constitute up to 40 % of the total seed storage protein (De Coninck & Van Damme, 2021; Levchuk et al., 2013). In animals and bacteria, these proteins take part in various metabolic pathways, including defence mechanisms. Lectins take part in endocytosis, intracellular transportation, apoptosis induction in tumour cells, Human Immunodeficiency Virus (HIV) infection blocking, bacterial adhesion and migration regulation, blood protein level control, and immune system function by recognising pathogen-specific carbohydrates (Dias et al., 2015). Structurally, lectins are defined as proteins with at least one domain that specifically binds to the sugar portions of glycoconjugates. Their binding specificity varies widely; for instance, some recognise mannose (e.g., favin from faba bean, lentil lectin, concanavalin A from the jackbean), others bind to galactose (e.g., ricin from castor seed, jacalin Fig. 3. Graphical representation of Bowman-Birk inhibitors of trypsin (left) and chymotrypsin (right). Panel A) shows the free enzymes with key amino acids involved in inhibition, while Panel B) depicts the enzyme-inhibitor complex. The figure was generated using AlphaFold2 protein structure database. G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 7 from jackfruit seed, peanut agglutinin) and still others target N-acetylglucosamine moieties (e.g., wheat germ agglutinin) (De Coninck & Van Damme, 2021). Based on their domains, lectins are categorised in merolectins (a single, monovalent carbohydrate-binding domain), hololectins (at least two domains binding the same sugar), chimerolectins (a binding domain paired with an enzymatic domain), and superlectins (binding structurally unrelated sugars) (Elumalai & Lakshmi, 2021). Based on their carbohydrate-binding specificity and structural characteristics, lectins can be further classified in: L-type (leguminous), C-type (calcium-dependent), P-type (plant), and F-type (fucose-specific), G-type (galactose-binding lectins), among others, each exhibiting unique biological functions and roles in various organisms (Elumalai & Lakshmi, 2021). Due to their biological activity, the presence of active lectins is a critical consideration in food safety. Injection of active lectins can cause gastrointestinal distress, interfere with nutrient absorption, and potentially damage the gut lining. The risk has led to food product recalls in Europe, such as for semi-processed frozen red kidney beans, as documented in the EU’s Rapid Alert System for Food and Feed (RASFF) online database (https://webgate.ec.europa.eu/rasff-window/screen/se arch). Consequently, the quantification of active lectins is a crucial step in the safety assessment of alternative food ingredients like legume protein fractions, especially as novel lectin-containing foods are introduced. Lectins are found in many alternative protein sources, from lentils and faba beans (0.1–1 mg/g) to the microalgae Chlorella spp., which exhibits low hemagglutination activity (Chen et al., 2022). To predict the presence of lectins across different protein sources, we utilised computational tools. The UniLectin database (https://unilectin. unige.ch/), a repository of curated and predicted lectins, classifies them into domain folds, and families based on sequence similarity (Bonnardel et al., 2019, 2021; Schnider et al., 2024). Using this resource, we generated a heatmap of the lectin-like proteins in legumes and cereals of interest, with Phaseolus vulgaris (common bean) as a reference due to its association with gastroenteritis. This analysis identified rapeseed and chickpea as having the highest number of expressed lectin-like proteins, followed by lentil and fava, with the L-type lectins being the most represented (Supplementary Fig. 1). Furthermore, we used LectomeXplore to generate a lectin class heatmap for the phyla Arthropoda (focusing on Acheta domesticus), Chlorophyta (focusing on Chlorella vulgaris), and Pseudomonadota (focusing on microbial biomass like Xanthobacter) (Supplementary Fig. 2a). As data for Acheta domesticus (October 2024) were unavailable at the time of the analysis, we used the related species Gryllus bimaculatus as a proxy (Fig. 4 and Supplementary Fig. 2b). This analysis revealed that Gryllus bimaculatus (Fig. 4) has the most complex lectinlike profile, including chitin-binding, rhamnose-binding, and the ERGIC-VIP (Endoplasmic Reticulum-Golgi Intermediate Compartment Vesicular Integral Membrane Protein) L-type lectins, a secretory glycoprotein that serves as a transport receptor. In Chlorella vulgaris, we observed Fucose (F)- and P-type lectin families, along with calnexin-calreticulin-like lectins. For microbial biomass like Xanthobacter, only the cyanovirin-like lectin, an 11-kDa protein of cyanobacterial origin with potent virucidal activity, was annotate (Balzarini, 2006; Mitchell et al., 2017; Sharon & Ofek, 2007). These in silico findings suggest that the actual protein expression levels in these ingredients and their potential health implications warrant a dedicated risk assessment. However, accurately quantifying active lectins is challenging. Currently, there are no universal reference methods, except for the hemagglutination assay food pulses (Supplementary Table 3). This assay uses red blood cells and protein extracts in physiological buffers to detect agglutination. A key limit is the need to maintain osmolarity to prevent cell lysis or lectin deactivation. While effective for easily solubilised seed proteins, applying this method to recalcitrant tissues (e.g., leaves) and walled organisms (e.g., microalgae) is problematic. Fig. 4. Heatmap visualization of lectin expression in the novel food ingredients (Gryllus bimaculatus, Chlorella vulgaris and Xanthobacter autotrophicus), with a minimum similarity score of 0.35 applied to the reference (Bonnardel et al., 2021). G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 8 Similarly, fractionation processes leading that result in poor protein solubility can lead to misinterpretation of results. Extraction methods must also preserve protein structure, precluding the use of ultrasound, or certain chaotropic or reducing chemicals. The commonly used buffers may selectively extract certain proteins, leading to inconclusive or falsenegative results in the assay. A multi-platform approach could address these limitations and highlight gaps in current knowledge. Finally, a lack of standardised reporting units, with activity expressed as Hemagglutination Units per mg of ingredient, total protein, or extracted protein, adds confusion to the risk assessment process. 3.2.3. Alpha amylase inhibitors Alpha-amylase inhibitors ( α AIs) block starch degradation by endogenous α -amylase during seed dormancy and protect the seed against pests and predators (Neeta & Discov, 2016). This digestive enzyme is also present in the hydrolytic enzyme pool of moulds, insects, and mammals. To date, researchers have established the existence of six types of α AI in the plant domain: the knottin type, γ thionin-like type, CM-proteins type, Kunitz type, legume lectin-like and thaumatin-like type (Peddio et al., 2022). A closer look at these types reveals distinct characteristics. The CM-type α AIs, also referred as “cereal type” α AIs, are small proteins (120–160 AA) soluble in chloroform/methanol. Many CM-type α AIs are abundant in barley seeds at different stages of maturation (Finnie et al., 2002). The Kunitz-type α AIs can inhibit either α -amylase or trypsin, while some variants can inhibit both. The lectin-like α AIs (namely α AI1 and α AI2) are glycoproteins isolated from a variety of kidney beans. Their name arises from their high degree of sequence homology with lectins expressed from the same legumes. These two isoforms can inhibit amylases in different organisms as; for example, α AI1 can inhibit both mammalian and insect amylases. The thaumatin-like AI is a protein with a molecular mass of approximately 22 kDa. These proteins share significant sequence similarity with pathogenesis-related group 5 (PR-5) proteins and with thaumatin, a potent sweet-tasting protein. Among this class, zeamatin, a bifunctional inhibitor found in Zea mays, is the best known. It inhibits porcine pancreatic trypsin and the intestinal amylase of various insects. However, despite the high N-terminal sequence similarity, other thaumatin-like proteins, such as those from barley seeds, show no inhibitory activity against trypsin or α -amylases (Franco et al., 2002). While the knottin-like, the Kunitz, and the thaumatin-like types appear to inhibit amylase only in insects, the other types (γ Thionin-like type, cereal type, and legume lectins-like) can inhibit amylase in both insects and mammals (Svensson et al., 2004). Investigating mammal α AIs is a critical aspect of assessing digestibility and bioavailability of macronutrients, especially in the context of alternative starchy food ingredients from pseudocereals, minor legumes and cereals. For example, the content and activity of α AIs have been reported for various legumes, with faba beans containing 18.90 IU/g (Alonso et al., 2001). 3.2.4. Thiaminases Thiaminases are enzymes that degrade thiamine (vitamin B1), a crucial nutrient for energy metabolism. These enzymes exist in two forms, Thiaminase I (Thiamine pyridinylase, EC 2.5.1.2, InterPro: IPR030901) and Thiminase II (Aminopyrimidine aminohydrolase, EC 3.5.99.2, InterPro: IPR027574, IPR004305), and are found in certain raw fish, shellfish, bacteria some plants and seeds. When consumed regularly without proper preparation, they can pose a risk of thiamine deficiency (Bitsch, 2003). Cooking or food processing typically inactivates thiaminases, thereby reducing the risk. The biological function of these enzymes has long been an enigma. Thiaminases break down thiamine by catalysing the base-exchange reaction that substitutes the thiazole moiety with a nucleophile. Thiaminase I, in particular, is associated with thiamine deficiency syndromes in animals. However, recent research has provided new insights. For instance, Sannino et al. (2018) demonstrated that Thiaminase I improved survival in Burkholderia thailandensis strains grown in thiamine-depleted media compared to isogenic strains lacking the enzyme. Their work suggests that Thiamine I aids survival by cleaving thiamine, its phosphorylated forms, and toxic analogues to release precursors for de novo thiamine synthesis. This mechanism allows thiamine auxotrophs to grow more effectively on these precursors than on thiamine itself. The KEGG db includes thiamine metabolism pathways for microbial biomass like Xanthobacter (xau00730) and Chlorella variabilis (cvr00730). However, in the available genomic sequences for both organisms, thiaminase I and II are either absent or unannotated. Conversely, the UNIPROT database lists four predicted Thiaminase II isoforms for Chlorella vulgaris, derived from EMBL/GenBank/DDBJ whole-genome shotgun (WGS) data (EMBL: KAI3427436.1). These isoforms (accession numbers A0A9D4TK12, A0A9D4THA1, A0A9D4TGC6, and A0A9D4TWS4) share 20–59 % sequence identity with the Thiaminase II domain. Similarly, for Xanthobacter, a protein inferred from homology (EMBL: TLX41961.1) has been identified as a Pyrroloquinolinequinone synthase (UNIPROT ID: A0A6C1KDR4) containing a Thiaminase II domain (residues 16–225). 3.3. Toxic anti-nutritional factors (t-ANFs) Certain components of protein sources, as well as compounds generated during processing regimes such as biotransformation, can be toxic when present above normal concentrations. Among these, nucleic acids, including deoxyribonucleic acid (DNA) and ribonucleic acid (RNA), are essential polymers of nucleotides found within all cells and viruses (Salim et al., 2023). They are particularly abundant in single-cell proteins and fermentation side-streams, such as brewers’ spent yeast (Alonso-Ria˜ no et al., 2021), their high concentration is a result of rapid microbiological growth phases. Unlike some anti-nutritional factors, elevated nucleic acid intake does not hinder nutrient absorption. Instead, it can disrupt purine metabolism, potentially leading to hyperuricemia (high uric acid in the blood), kidney stone formation, and gout (Sharif et al., 2021). Single-cell proteins and hydrolysed yeasts can contain 6 %–15 % nucleic acids, with a daily consumption of up to 33 g of yeast considered Generally Recognized As Safe (GRAS) (Podpora et al., 2016). A review by Ritala et al. (2017) identified nucleic acids as a significant safety concern in SCPs. For example, bacterial SCPs may contain 8–12 % nucleic acids (primarily RNA), while fungi-based SCPs contain slightly lower levels of 7–10 %. To address this, enzymatic strategies have been developed to reduce nucleic acid content. In the manufacturing of Quorn™ mycoproteins from Fusarium venenatum, heat-activated endogenous ribonucleases are employed to lower RNA levels, ensuring the product is safe for consumption. It is also worth noting that some mycoproteins may present other metabolic effects; for instance, in vitro evidence suggests they can sequester bile salts, potentially affecting lipid absorption (Colosimo et al., 2020). A separate class of anti-nutritional compounds includes toxic glycosides, which are secondary metabolites that serve as defence mechanisms in organisms. These can be broadly categorised: Cyanogenic glycosides found in over 2500 plant species, these compounds are glycosides of α -hydroxynitriles that release cyanide upon enzymatic hydrolysis (Ritmejeryt˙ e et al., 2022). Amygdalin, first isolated from bitter almond seeds (Amygdalus communis) is a well-known example. Some plant species, such as tobacco, contain high levels of alkaloid glycosides like nicotine. These are cyclic organic compounds containing nitrogen, found in sources ranging from microalgae and insects to common foods like lupins, potatoes, tomatoes and eggplants. They are classified by their heterocyclic ring structure and can be anti-nutritional by inhibiting digestion or causing symptoms such as vomiting and diarrhoea (Hill, 2003; Pfister et al., 2001; Wink, 2003). For instance, vicine and convicine are glycoalkaloids faba beans responsible for favism. This is a genetic condition where their consumption by individuals with a glucose-6-phosphate dehydrogenase (G6PD) enzyme deficiency leads to acute hemolytic anemia, the rapid destruction of red blood cells G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 9 Linsberger-Martin, G., Weiglhofer, K., Thi Phuong, T. P., & Berghofer, E. (2013). High hydrostatic pressure influences antinutritional factors and in vitro protein digestibility of split peas and whole white beans. LWT - Food Science and Technology, 51(1), 331–336. https://doi.org/10.1016/j.lwt.2012.11.008 Loh, J. S., Mak, W. Q., Tan, L. K. S., Ng, C. X., Chan, H. H., Yeow, S. H., Foo, J. B., Ong, Y. S., How, C. W., & Khaw, K. Y. (2024). Microbiota–gut–brain axis and its therapeutic applications in neurodegenerative diseases. Signal Transduction and Targeted Therapy, 9(1), 37. https://doi.org/10.1038/s41392-024-01743-1 Lu, Y., Tian, C., Gao, C., Wang, B., Yang, W., Kong, X., Chai, L., Chen, G., Yin, X., & He, Y. (2018). Phenolic composition, antioxidant capacity and inhibitory effects on α -glucosidase and lipase of immature faba bean seeds. International Journal of Food Properties, 21(1), 2366–2377. https://doi.org/10.1080/10942912.2018.1522331 Lu, J., Zamaratskaia, G., Langton, M., R¨ ohnisch, H. E., & Karkehabadi, S. (2024). Minimizing anti-nutritional factors in wet protein extraction from Swedish faba beans through the application of response surface methodology. Food Chemistry, 460, Article 140700. https://doi.org/10.1016/j.foodchem.2024.140700 Magara, H. J. O., Niassy, S., Ayieko, M. A., Mukundamago, M., Egonyu, J. P., Tanga, C. M., Kimathi, E. K., Ongere, J. O., Fiaboe, K. K. M., Hugel, S., Orinda, M. A., Roos, N., & Ekesi, S. (2021). Edible crickets (orthoptera) around the world: Distribution, nutritional value, and other benefits—A review. Frontiers in Nutrition, 7. https://doi.org/10.3389/fnut.2020.537915 Mariotti, F., & Gardner, C. D. (2019). Dietary protein and amino acids in vegetarian diets—A review. Nutrients, 11(11), 1–19. https://doi.org/10.3390/nu11112661 Mayer Labba, I.-C. C., Frøkiær, H., & Sandberg, A.-S. S. (2021). Nutritional and antinutritional composition of fava bean (Vicia faba L., var. minor) cultivars. Food Research International, 140(December 2020), Article 110038. https://doi.org/ 10.1016/j.foodres.2020.110038 McKie, V. A., & McCleary, B. V. (2016). A novel and rapid colorimetric method for measuring total phosphorus and phytic acid in foods and animal feeds. Journal of AOAC International, 99(3), 738–743. https://doi.org/10.5740/jaoacint.16-0029 Meldrum, O. W., Yakubov, G. E., Gartaula, G., McGuckin, M. A., & Gidley, M. J. (2017). Mucoadhesive functionality of cell wall structures from fruits and grains: Electrostatic and polymer network interactions mediated by soluble dietary polysaccharides. Scientific Reports, 7(1), Article 15794. https://doi.org/10.1038/ s41598-017-16090-1 Miao, X., Niu, H., Sun, M., Dong, X., Hua, M., Su, Y., Wang, J., & Li, D. (2024). A comparative study on the nutritional composition, protein structure and effects on gut microbiota of 5 fermented soybean products (FSPs). Food Research International, 183, Article 114199. https://doi.org/10.1016/j.foodres.2024.114199 Mitchell, C. A., Ramessar, K., & O’Keefe, B. R. (2017). Antiviral lectins: Selective inhibitors of viral entry. Antiviral Research, 142, 37–54. https://doi.org/10.1016/j. antiviral.2017.03.007 Moughan, P. J. (2021). Population protein intakes and food sustainability indices: The metrics matter. Global Food Security, 29, Article 100548. https://doi.org/10.1016/j. gfs.2021.100548 Moughan, P. J., Fulgoni, V. L., & Wolfe, R. R. (2024). The importance of dietary protein quality in Midto high-income countries. The Journal of Nutrition, 154(3), 804–814. https://doi.org/10.1016/j.tjnut.2024.01.020 Muzquiz, M., Welham, T., Altares, P., Goyoaga, C., Cuadrado, C., Romero, C., Guillamon, E., & Domoney, C. (2004). The effect of germination on seed trypsin inhibitors in Vicia faba and Cicer arietinum. Journal of the Science of Food and Agriculture, 84(6), 556–560. https://doi.org/10.1002/jsfa.1695 Nasrollahzadeh, F., Roman, L., Swaraj, V. J. S., Ragavan, K. V., Vidal, N. P., Dutcher, J. R., & Martinez, M. M. (2022). Hemp (Cannabis sativa L.) protein concentrates from wet and dry industrial fractionation: Molecular properties, nutritional composition, and anisotropic structuring. Food Hydrocolloids, 131 (November 2021), Article 107755. https://doi.org/10.1016/j.foodhyd.2022.107755 Neeta, K., & Discov, L. K.-B. (2016). In vitro screening of proteinaceous amylase inhibitors (T. castaneum α -amylase inhibitors) in different seed extracts. undefined. (2016) Jbsd.In, 7(2), 166–173. Ney, L.-M., Wipplinger, M., Grossmann, M., Engert, N., Wegner, V. D., & Mosig, A. S. (2023). Short chain fatty acids: Key regulators of the local and systemic immune response in inflammatory diseases and infections. Open Biology, 13(3). https://doi. org/10.1098/rsob.230014 Nikmaram, N., Leong, S. Y., Koubaa, M., Zhu, Z., Barba, F. J., Greiner, R., Oey, I., & Roohinejad, S. (2017). Effect of extrusion on the anti-nutritional factors of food products: An overview. Food Control, 79, 62–73. https://doi.org/10.1016/j. foodcont.2017.03.027 Nissar, J., Ahad, T., Naik, H. R., & Hussain, S. Z. (2017). A review phytic acid: As antinutrient or nutraceutical. Journal of Pharmacognosy and Phytochemistry, 6(6), 1554–1560. https://www.phytojournal.com/archives/2017.v6.i6.2263/a-review-ph ytic-acid-as-antinutrient-or-nutraceutical. Nogal, A., Valdes, A. M., & Menni, C. (2021). The role of short-chain fatty acids in the interplay between gut microbiota and diet in cardio-metabolic health. Gut Microbes, 13(1). https://doi.org/10.1080/19490976.2021.1897212 Odani, S., & Ikenaka, T. (1973). Scission two of soybean fragments having activity proteinase either trypsin inhibitor or into small chymotrypsin inhibitory department of biochemistry , Niigata university school of when anti-chymotryptic activity gel fractions were reacted with 0 . 05mg. The Journal of Biochemistry, 74(4), 857–860. Ohanenye, I. C., Ekezie, F.-G. C., Sarteshnizi, R. A., Boachie, R. T., Emenike, C. U., Sun, X., Nwachukwu, I. D., & Udenigwe, C. C. (2022). Legume seed protein digestibility as influenced by traditional and emerging physical processing technologies. Foods, 11(15), 2299. https://doi.org/10.3390/foods11152299 Ojo, M. A. (2022). Tannins in foods: Nutritional implications and processing effects of hydrothermal techniques on underutilized hard-to-cook legume Seeds-A review. Preventive Nutrition and Food Science, 27(1), 14–19. https://doi.org/10.3746/ pnf.2022.27.1.14 Ozdal, T., Capanoglu, E., & Altay, F. (2013). A review on protein–phenolic interactions and associated changes. Food Research International, 51(2), 954–970. https://doi. org/10.1016/j.foodres.2013.02.009 Ozolina, K., Sarenkova, I., & Muizniece-Brasava, S. (2023). The anti-nutritional factors of legumes and their treatment possibilities: A review (Vols. 68–76). https://doi.org/ 10.22616/RRD.29.2023.010 Patterson, C. A., Curran, J., & Der, T. (2017). Effect of processing on antinutrient compounds in pulses. Cereal Chemistry, 94(1), 2–10. https://doi.org/10.1094/ CCHEM-05-16-0144-FI Peddio, S., Padiglia, A., Cannea, F. B., Crnjar, R., Zam, W., Sharifi-Rad, J., Rescigno, A., & Zucca, P. (2022). Common bean (Phaseolus vulgaris L.) α -amylase inhibitors as safe nutraceutical strategy against diabetes and obesity: An update review. Phytotherapy Research, 36(7), 2803–2823. https://doi.org/10.1002/ptr.7480 Pereira, E., Cadavez, V., Barros, L., Encina-Zelada, C., Stojkovi´ c, D., Sokovic, M., Calhelha, R. C., Gonzales-Barron, U., & Ferreira, I. C. F. R. (2020). Chenopodium quinoa willd. (quinoa) grains: A good source of phenolic compounds. Food Research International, 137, Article 109574. https://doi.org/10.1016/j.foodres.2020.109574 Pfister, J. A., Panter, K. E., Gardner, D. R., Stegelmeier, B. L., Ralphs, M. H., Molyneux, R. J., & Lee, S. T. (2001). Alkaloids as anti-quality factors in plants on Western U.S. rangelands. Journal of Range Management, 54(4), 447. https://doi.org/ 10.2307/4003116 Podpora, B., ´ Swiderski, F., Sadowska, A., Rakowska, R., & Wasiak-Zys, G. (2016). Spent brewer’s yeast extracts as a new component of functional food. Czech Journal of Food Sciences, 34(6), 554–563. https://doi.org/10.17221/419/2015-CJFS P¨ ori, P., Aisala, H., Liu, J., Lille, M., & Sozer, N. (2023). Structure, texture, and sensory properties of plant-meat hybrids produced by high-moisture extrusion. LWT, 173, Article 114345. https://doi.org/10.1016/j.lwt.2022.114345 Portugal-Nunes, C., Nunes, F. M., Saraiva, C., & Gonçalves, C. (2023). Public interest in food sustainability: An infodemiology study of google trends data in Europe from 2010–2021. International Journal of Food Sciences & Nutrition, 74(1), 95–106. https:// doi.org/10.1080/09637486.2022.2151988 Prieto, M. A., L´ opez, C. J., & Simal-Gandara, J. (2019). Chapter six - Glucosinolates: Molecular structure, breakdown, genetic, bioavailability, properties and healthy and adverse effects. In I. C. F. R. Ferreira, L. B. T.-A. in F., & N. R. Barros (Eds.), Functional food ingredients from plants (Vol. 90, pp. 305–350). Academic Press. https://doi.org/ 10.1016/bs.afnr.2019.02.008. Pryme, I. F., & Aarra, T. M. (2021). Exhaustive overview of dietary plant lectins: Prospective importance in the mediterranean diet. American Journal of Biomedical Science & Research, 13(3), 339–358. https://doi.org/10.34297/ AJBSR.2021.13.001883 Raes, K., Knockaert, D., Struijs, K., & Van Camp, J. (2014). Role of processing on bioaccessibility of minerals: Influence of localization of minerals and anti-nutritional factors in the plant. Trends in Food Science & Technology, 37(1), 32–41. https://doi. org/10.1016/j.tifs.2014.02.002 Rana, N., & Dahiya, S. (2021). Proximate composition, in vitro digestibility and anti nutritional factors of millets and legume grains. Agricultural Science Digest - A Research Journal, Of. https://doi.org/10.18805/ag.D-5248 Rana, A., Samtiya, M., Dhewa, T., Mishra, V., & Aluko, R. E. (2022). Health benefits of polyphenols: A concise review. Journal of Food Biochemistry, 46(10). https://doi.org/ 10.1111/jfbc.14264 Ridlo, A., Pringgenies, D., Perangin-angin, R. A. B., & Ariyanto, D. (2023). Phytochemicals and antioxidant activity of microalgae Dunaliella salina and Botryococcus braunii. Jurnal Ilmiah Perikanan Dan Kelautan, 15(2), 438–447. https:// doi.org/10.20473/jipk.v15i2.36740 Ritmejeryt˙ e, E., Miller, R. E., Bayly, M. J., & Boughton, B. A. (2022). Visualization of cyanogenic glycosides in floral tissues. In Applied environmental metabolomics (pp. 29–44). Elsevier. https://doi.org/10.1016/B978-0-12-816460-0.00004-6. Rosa-Sibakov, N., Re, M., Karsma, A., Laitila, A., & Nordlund, E. (2018). Phytic acid reduction by bioprocessing as a tool to improve the in vitro digestibility of faba bean protein. Journal of Agricultural and Food Chemistry, 66(40), 10394–10399. https:// doi.org/10.1021/acs.jafc.8b02948 Roy, A., Ghosh, S., & Kundagrami, S. (2019). Food processing methods towards reduction of antinutritional factors in chickpea. International Journal of Current Microbiology and Applied Sciences, 8(1), 424–432. https://doi.org/10.20546/ijcmas.2019.801.044 Salim, R., Nehvi, I. B., Mir, R. A., Tyagi, A., Ali, S., & Bhat, O. M. (2023). A review on anti-nutritional factors: Unraveling the natural gateways to human health. Frontiers in Nutrition, 10(August), 1–15. https://doi.org/10.3389/fnut.2023.1215873 Sanaei Nasab, S., Tahmouzi, S., Feizollahi, E., & Mollakhalili-Meybodi, N. (2024). Impacts of novel non-thermal processing (NTP) on anti-nutritional compounds of food grains and seeds. Food Control, 162, Article 110469. https://doi.org/10.1016/j. foodcont.2024.110469 Sannino, D. R., Kraft, C. E., Edwards, K. A., & Angert, E. R. (2018). Thiaminase I provides a growth advantage by salvaging precursors from environmental thiamine and its analogs in Burkholderia thailandensis. Applied and Environmental Microbiology, 84 (18). https://doi.org/10.1128/AEM.01268-18 Sattar, A., Neelofar, & Akhtar, M. A. (1990). Irradiation and germination effects on phytate, protein and amino acids of soybean. Plant Foods for Human Nutrition, 40(3), 185–194. https://doi.org/10.1007/BF01104141 Schnider, B., M’Rad, Y., el Ahmadie, J., de Brevern, A. G., Imberty, A., & Lisacek, F. (2024). HumanLectome, an update of UniLectin for the annotation and prediction of human lectins. Nucleic Acids Research, 52(D1), D1683–D1693. https://doi.org/ 10.1093/nar/gkad905 G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 16 Schreiner, T. B., Dias, M. M., Barreiro, M. F., & Pinho, S. P. (2022). Saponins as natural emulsifiers for nanoemulsions. Journal of Agricultural and Food Chemistry, 70(22), 6573–6590. https://doi.org/10.1021/acs.jafc.1c07893 Seczyk, L., Swieca, M., Kapusta, I., & Gawlik-Dziki, U. (2019). Protein–phenolic interactions as a factor affecting the physicochemical properties of white bean proteins. Molecules, 24(3), 1–24. https://doi.org/10.3390/molecules24030408 Sharan, S., Zanghelini, G., Zotzel, J., Bonerz, D., Aschoff, J., Saint-Eve, A., & Maillard, M. (2021). Fava bean (Vicia faba L.) for food applications: From seed to ingredient processing and its effect on functional properties, antinutritional factors, flavor, and color. Comprehensive Reviews in Food Science and Food Safety, 20(1), 401–428. https://doi.org/10.1111/1541-4337.12687 Sharif, M., Zafar, M. H., Aqib, A. I., Saeed, M., Farag, M. R., & Alagawany, M. (2021). Single cell protein: Sources, mechanism of production, nutritional value and its uses in aquaculture nutrition. Aquaculture, 531, Article 735885. https://doi.org/10.1016/ j.aquaculture.2020.735885 Sharma, A. (2021). A review on traditional technology and safety challenges with regard to antinutrients in legume foods. Journal of Food Science and Technology, 58(8), 2863–2883. https://doi.org/10.1007/s13197-020-04883-8 Sharon, N., & Ofek, I. (2007). Microbial lectins. In Comprehensive glycoscience (pp. 623–659). Elsevier. https://doi.org/10.1016/B978-044451967-2/00068-4. Siddiqui, S. A., Zhao, T., Fitriani, A., Rahmadhia, S. N., Alirezalu, K., & Fernando, I. (2024). Acheta domesticus (house cricket) as human foods - an approval of the european commission - A systematic review. Food Frontiers, 5(2), 435–473. https:// doi.org/10.1002/fft2.358 Silventoinen, P., Kortekangas, A., Nordlund, E., & Sozer, N. (2022). Impact of phytase treatment and calcium addition on gelation of a protein-enriched rapeseed fraction. Food and Bioprocess Technology, 15(6), 1422–1435. https://doi.org/10.1007/s11947022-02810-7 Simons, C., Hall, C., & Biswas, A. (2017). Characterization of pinto bean high-starch fraction after air classification and extrusion. Journal of Food Processing and Preservation, 41(6), Article e13254. https://doi.org/10.1111/jfpp.13254 Singh, B., Singh, J. P., Shevkani, K., Singh, N., & Kaur, A. (2017). Bioactive constituents in pulses and their health benefits. Journal of Food Science and Technology, 54(4), 858–870. https://doi.org/10.1007/s13197-016-2391-9 Soycan, G., Sch¨ ar, M. Y., Kristek, A., Boberska, J., Alsharif, S. N. S., Corona, G., Shewry, P. R., & Spencer, J. P. E. (2019). Composition and content of phenolic acids and avenanthramides in commercial oat products: Are oats an important polyphenol source for consumers? Food Chemistry X, 3, Article 100047. https://doi.org/ 10.1016/j.fochx.2019.100047 Statista Search Department. (2024a). Share of adults following a flexitarian diet in the european union (EU-27) in 2023/24. Statista Search Department. (2024b). Share of respondents who follow a flexitarian diet in selected countries in Europe in 2023, by country. Svensson, B., Fukuda, K., Nielsen, P. K., & Bønsager, B. C. (2004). Proteinaceous α -amylase inhibitors. Biochimica et Biophysica Acta - Proteins and Proteomics, 1696(2), 145–156. https://doi.org/10.1016/j.bbapap.2003.07.004 Taylor, S. L., & Godefroy, S. B. (2023). Novel foods and ingredients: Laws and regulations in USA and Canada. In Reference module in food science. Elsevier. https://doi.org/ 10.1016/B978-0-12-823960-5.00068-8. Timilsena, Y. P., Phosanam, A., & Stockmann, R. (2023). Perspectives on saponins: Food functionality and applications. International Journal of Molecular Sciences, 24(17), Article 13538. https://doi.org/10.3390/ijms241713538 Tsao, R. (2010). Chemistry and biochemistry of dietary polyphenols. Nutrients, 2(12), 1231–1246. https://doi.org/10.3390/nu2121231 Vergnolle, N. (2016). Protease inhibition as new therapeutic strategy for GI diseases. Gut, 65(7), 1215–1224. https://doi.org/10.1136/gutjnl-2015-309147 Vogelsang-O’Dwyer, M., Petersen, I. L., Joehnke, M. S., Sørensen, J. C., Bez, J., Detzel, A., Busch, M., Krueger, M., O’Mahony, J. A., Arendt, E. K., & Zannini, E. (2020). Comparison of faba bean protein ingredients produced using dry fractionation and isoelectric precipitation: Techno-functional, nutritional and environmental performance. Foods, 9(3), 322. https://doi.org/10.3390/foods9030322 Vorster, J., van der Westhuizen, W., du Plessis, G., Marais, D., Sparvoli, F., Cominelli, E., … Kunert, K. (2023). In order to lower the antinutritional activity of serine protease inhibitors, we need to understand their role in seed development. Frontiers in Plant Science, 14(October), 1–6. https://doi.org/10.3389/fpls.2023.1252223 Wang, N., & Daun, J. (2006). Effects of variety and crude protein content on nutrients and anti-nutrients in lentils ( ). Food Chemistry, 95(3), 493–502. https://doi.org/ 10.1016/j.foodchem.2005.02.001 Wang, Y., Rosa-Sibakov, N., Edelmann, M., Sozer, N., Katina, K., & Coda, R. (2022). Enhancing the utilization of rapeseed protein ingredients in bread making by tailored lactic acid fermentation. Food Bioscience, 50, Article 102028. https://doi. org/10.1016/j.fbio.2022.102028 Wink, M. (2003). Alkaloids | toxicology. In Encyclopedia of food sciences and nutrition (pp. 134–143). Elsevier. https://doi.org/10.1016/B0-12-227055-X/00030-4. Xie, C., Li, W., Gao, R., Yan, L., Wang, P., Gu, Z., & Yang, R. (2022). Determination of glucosinolates in rapeseed meal and their degradation by myrosinase from rapeseed sprouts. Food Chemistry, 382, Article 132316. https://doi.org/10.1016/j. foodchem.2022.132316 Xie, Y., Ravet, K., & Pearce, S. (2021). Extensive structural variation in the bowman-birk inhibitor family in common wheat (Triticum aestivum L.). BMC Genomics, 22(1), 1–21. https://doi.org/10.1186/s12864-021-07475-8 Xue, J., Wu, J., Ji, Y., Sun, S., Gao, Y., Yang, H., & Wu, J. (2024). Effect of microbial fermentation on the quality of soybean meal. International Journal of Food Science and Technology, 59(1), 72–83. https://doi.org/10.1111/ijfs.16817 Zanchetta, E., Damergi, E., Patel, B., Borgmeyer, T., Pick, H., Pulgarin, A., & Ludwig, C. (2021). Algal cellulose, production and potential use in plastics: Challenges and opportunities. Algal Research, 56, Article 102288. https://doi.org/10.1016/j. algal.2021.102288 Zhang, Y., Xiao, H., Lv, X., Wang, D., Chen, H., & Wei, F. (2022). Comprehensive review of composition distribution and advances in profiling of phenolic compounds in oilseeds. Frontiers in Nutrition, 9. https://doi.org/10.3389/fnut.2022.1044871 Zhao, F., Wang, Z., & Huang, H. (2024). Physical cell disruption technologies for intracellular compound extraction from microorganisms. Processes, 12(10), 2059. https://doi.org/10.3390/pr12102059 Zhou, Y., & Xu, B. (2023). New insights into anti-diabetes effects and molecular mechanisms of dietary saponins. Critical Reviews in Food Science and Nutrition, 63 (33), 12372–12397. https://doi.org/10.1080/10408398.2022.2101425 G. D’Auria et al. Trends in Food Science & Technology 165 (2025) 105365 17