Abstract
Multiple sensorial, technological, and nutritional challenges must be overcome when developing plant‐based fermented dairy alternatives (PBFDA) to mimic their dairy counterparts. The elimination of plant‐derived off‐flavors (green, earthy, bitter, astringent) and the degradation of antinutrients are crucial quality factors highlighted by the industry for their effect on consumer acceptance. The adaptation of plant‐derived lactic acid bacteria (LAB) species into plant niches is relevant when developing starter cultures for PBFDA products due to their evolutionary acquired ability to degrade plant‐based undesirable compounds (off‐flavors and antinutrients). Some plant‐isolated species, such as Lactiplantibacillus plantarum and Limosilactobacillus fermentum, have been associated with the degradation of phytates, phenolic compounds, oxalates, and raffinose‐family oligosaccharides (RFOs), whereas some animal‐isolated species, such as Lactobacillus acidophilus strains, can metabolize phytates, RFOs, saponins, phenolic compounds, and oxalates. Some proteolytic LAB strains, such as Lacticaseibacillus paracasei and Lacticaseibacillus rhamnosus, have been characterized to degrade phytates, protease inhibitors, and oxalates. Other species have also been described regarding their abilities to biotransform phytic acid, RFOs, saponins, phenolic compounds, protease inhibitors, oxalates, and volatile off‐flavor compounds (hexanal, nonanal, pentanal, and benzaldehyde). In addition, we performed a blast analysis considering antinutrient metabolic genes (42 genes) to up to 5 strains of all qualified presumption of safety‐listed LAB species (55 species, 240 strains), finding out potential genotypical capabilities of LAB species that have not conventionally been used as starter cultures such as Lactiplantibacillus pentosus, Lactiplantibacillus paraplantarum, and Lactobacillus diolivorans for plant‐based fermentations. This review provides a detailed understanding of genes and enzymes from LAB that target specific compounds in plant‐based materials for plant‐based fermented food applications.
Keywords: antinutrients, fermentation, lactic acid bacteria, off‐flavors, plant‐based alternatives
1. INTRODUCTION
Milk fermentation processes have been used due to their preservation and sensorial characteristics, contributing to the safety and quality of final fermented food products such as yogurt, cheese, and kefir. In the last decade, the plant‐based dairy‐fermented market has increased due to the growth of consumers’ demand and awareness toward more sustainable, healthier, and more ethical food production processes and products (Engels et al., 2022; Harper et al., 2022; Ma et al., 2022; Tangyu et al., 2019). The plant‐based milk industry utilizes raw materials that require less energy, emit less greenhouse gas emissions, and need less water and land usage to be produced compared to the processing of bovine milk (Harper et al., 2022). The main challenges of the plant‐based fermented dairy alternatives (PBFDA) industry in replacing dairy products are mimicking dairy‐like flavors, color, and texture properties characteristic of dairy‐fermented products, and developing products without off‐flavors and antinutrients to improve their acceptance (Genet, Molina, et al., 2023; Sedó Molina et al., 2022; Sedó Molina, Shetty, et al., 2024).
The terms “antinutrients” and “off‐flavors” refer to multiple and physiochemically different undesirable compounds that affect a food product's nutritional and sensory profiles, in this case, PBFDAs. On the one hand, there are several antinutrients present in different plant‐based matrices that either decrease the digestibility of proteins, such as protease inhibitors, reduce the bioavailability of nutrients (minerals, vitamins, and proteins) such as phytates, polyphenolic compounds, and saponins, or cause diverse health effects such as plant oligosaccharides (raffinose, stachyose, and verbascose) that are attributed to result in flatulence (Gemede & Ratta, 2014; López‐Moreno et al., 2022; Popova & Mihaylova, 2019). On the other hand, off‐flavors are classified into volatile, such as specific plant aldehydes, alcohols, ketones, furans, and pyrazines, and non‐volatile, such as phenolic compounds, particular peptides, and saponins (Leonard et al., 2022). Volatile compounds provide “green,” “grassy,” “earthy,” and “mushroom” notes, whereas non‐volatile are defined as “bitter” and “astringent” (Mittermeier‐Kleßinger et al., 2021). Different plant‐based matrices are characterized to have chemically different antinutrients and off‐flavors that need to be removed to produce high‐quality PBFDAs.
The transition from dairy to plant‐based milk also implies a change in the physical and chemical composition perspective, influencing the subsequent fermentation process (Sedó Molina, Shetty, et al., 2024). Overall, bovine milk is mainly composed of protein (3.2%–3.4%), in which casein accounts for 82%, fat (3.2%–3.7%), and lactose (4.6%) as the main carbohydrate. In comparison, plant‐based milk offers a more diverse composition depending on the plant‐based substrate utilized, its processing conditions, and the formulation when producing the product. The most common plant substrates, such as soybeans, almonds, and coconut, are composed of 7.3%, 4.4%, and 6.2% sugars, without considering fibers, 19.9%, 49.9%, and 33.5% fat, and 36.5%, 21.2%, and 3.3% protein, respectively, in raw form (Gorissen et al., 2018; Vanga & Raghavan, 2018). Moreover, plant‐based fermentable sugars present in plant‐based milk differ from glucose, maltose, fructose, sucrose, raffinose, and starch, among others, depending on the plant‐based material used (Elango et al., 2022; Sedó Molina, Shetty, et al., 2024). Moreover, plant‐based proteins differ in structure, sequence, and amino acid composition but generally are characterized by their large size and multimeric and globular structures in comparison to casein (Christensen et al., 2022; Mittermeier‐Kleßinger et al., 2021; Sedó Molina, Shetty, et al., 2024). The differences between bioavailable carbon and nitrogen sources and the presence of certain minerals and vitamins in each type of milk will affect the fermentation and metabolism of microorganisms, thus influencing the overall sensory properties of the final PBFDAs (Merzlov et al., 2022; Samtiya et al., 2021; Sedó Molina, Shetty, et al., 2024).
Traditionally, some lactic acid bacteria (LAB) species, which have been inhabiting dairy milk niches, have been extensively used throughout the fermentation of dairy products. Historically, dairy fermentation procedures have moved from using spontaneous and back‐slopping fermentation techniques in small or medium batches to employing specific starter cultures in large‐scale production to generate more controllable, predictable, and safe final fermented products and processes. Milk‐isolated LAB strains have been demonstrated to be useful as starter cultures due to their evolutionarily adapted ability to consume milk substrates, such as lactose and casein, and mainly produce lactic acid, among other secondary metabolites, which fast acidifies milk and generates specific flavor, texture, and other characteristics to the desired final product (Alemayehu et al., 2014; de Vos & Vaughan, 1994; Fontana et al., 2018; Szutowska, 2020). LAB species are ubiquitous in nature and have been found in multiple sources, such as plants, fruits and vegetables, insects, and meat (Filannino et al., 2016; Konings et al., 2000; Wu et al., 2017). LABs are characterized by genetically adapting to their inhabiting niches, resulting in a diverse genotype among species and strains that have acquired or lost genes through evolutionary events in specific environments (Bachmann et al., 2017; Bačun‐Družina et al., 2009; Makarova & Koonin, 2007; Ventura et al., 2009). Thus, LAB evolution is essential when selecting LAB strains to be used as starter cultures, specifically in PBFDAs. Some LAB species and strains from plant‐based niches have acquired genes and encode enzymes that are useful for the degradation, transformation, or removal of such antinutrients and off‐flavors (Gaur & Gänzle, 2023; Sedó Molina, Shetty, et al., 2024). Thus, it may be relevant to redirect the use of conventional milk LAB starter cultures to the utilization of plant LAB isolates that phenotypically express those genes, thereby reducing the concentrations of unwanted compounds while increasing the quality of PBFDAs.
Multiple studies have shown the successful removal of undesirable compounds by physicochemical and biological approaches such as germination, dehulling, soaking, and fermentation (Das et al., 2022). This review focuses on the key metabolic machinery characterized in LAB strains involved in removing those compounds through fermentation. Thus, we explore the state‐of‐the‐art genes, enzymes, and pathways that have been described to be present in LAB species and relevant to the degradation of antinutrients and off‐flavors present in different plant‐based matrices, including a homology‐based analysis with all qualified presumption of safety (QPS)‐listed LAB species. Our focus is on several antinutrients, including phytates, plant oligosaccharides, saponins, phenolic compounds, protease inhibitors, and oxalates, as well as several volatile off‐flavors, including aldehydes such as hexanal, pentanal, and nonanal, and furans such as 2‐pentylfuran and others.
2. METHODS
2.1. Blast analysis of genes involved in the degradation of antinutrients
A homology‐based analysis has been conducted in 240 LAB strains from 55 LAB species (considered QPS) blasting 42 different genes (accession numbers found in File S1) found in LAB. These genes have been characterized to be involved in the biochemical transformation of phytate, raffinose‐family oligosaccharides (RFOs), saponins, phenolic compounds, enzyme inhibitors, and oxalates. The annotated “.faa” files (translated gene‐based protein sequences) of 240 LAB strains, generated using Prokka, were retrieved from the National Center of Biotechnology Information database according to the QPS list. These files were used as templates for the BLAST analysis of the targeted protein sequences. Positive matches were considered if there was a ≥50% amino acid homology and E‐value < 0.0005. Genes, LAB strains, and the number of positive blast matches can be found in File S1. Two graphs were generated from the blast outcome using OriginLab Pro 2021 (OriginLab Corporation), as shown in Figure 2. Figure 2a is a hierarchical clustering heatmap generated based on the average number of genes present in each LAB species, considering all the strains tested from each species. The clustering was generated based on Pearson correlation distance matrix, and the linkage was based on each species’ average number of genes. In the heatmap, only rows (LAB species) were clustered, and the color scale was from red (lowest values) to dark blue (highest values), passing through light yellow (medium values). Figure 2b is a PCA biplot graph with loadings visually representing closely related species based on their genetic similarities linked to their genetic capabilities to metabolize the different categories of antinutrients. The PCA considers 49.7% of the variation (between both components). The number of LAB strains from each species was included as follows: Bifidobacterium spp.: adolescentis (4), animalis (4), bifidum (3), breve (5), longum (5), Carnobacterium divergens (4), Companilactobacillus spp.: alimentarius (4), farciminis (4), Fructilactobacillus sanfranciscensis (5), Lacticaseibacillus spp.: casei (5), paracasei (5), rhamnosus (3), Lactiplantibacillus spp.: argentoratensis (4), paraplantarum (5), pentosus (5), plantarum (5), Lactobacillus spp.: acidophilus (4), amylolyticus (4), amylovorus (4), crispatus (5), delbrueckii (5), gallinarum (4), gasseri (3), helveticus (5), johnsonii (5), kefiranofaciens (5), Lactococcus lactis (4), Lapidilactobacillus dextrinicus (3), Latilactobacillus spp.: curvatus (5), sakei (4), Lentilactobacillus spp.: buchneri (4), diolivorans (4), hilgardii (5), kefiri (5), parafarraginis (3), Leuconostoc spp.: citreum (5), lactis (5), mesenteroides (5), pseudomesenteroides (5), Levilactobacillus brevis (4), Ligilactobacillus spp.: animalis (5), aviarius (4), salivarius (3), Limosilactobacillus spp.: fermentum (5), mucosae (5), panis (3), pontis (5), reuteri (5), Loigolactobacillus coryniformis (4), Oenococcus oeni (5), Pediococcus spp.: acidilactici (5), parvulus (5), pentosaceus (5), Secundilactobacillus collinoides (2), Streptococcus thermophilus (4).
FIGURE 2.
Homology‐based blast analysis of genes (42) involved in the degradation of antinutrients to all QPS‐listed LAB species (240 strains). Gene and species abbreviations are included in File S1. (a) Species‐based clustered heatmap representing the average number of copies of genes found in the different LAB strains from each LAB species. (b) PC‐biplot (PC2 = 21.7% and PC1 = 28.0%) of the average number of genes found in each LAB species standardized to a scale of 0–1 based on the maximum LAB species representative of each antinutrient category. The five antinutrient categories were included as loadings, whereas each species was included as a score with a colored‐based dot scale. The closer the score (dot) is to the loading, the higher the number of species’ genes as an average of the representative strains analyzed from that species. EI, enzyme inhibitors; OX, oxalates; PA, phytate; PHE, phenolic compounds; RFOs, raffinose‐family oligosaccharides; SP, saponins. Source: Plots were created using OriginPro 2021.
3. IMPORTANCE OF THE EVOLUTION OF LAB: FROM MILK TO PLANT NICHE ADAPTATION
LAB is a group of gram‐positive bacteria widely used in multiple areas, including biotechnological, medical, and food applications (Kandler, 1983). In the food industry, some LAB species have been mainly used in food fermentation processes as starter cultures to produce safe and high‐quality fermented products in robust, replicable, and controlled fermentation processes. Other LAB species can spontaneously grow during fermentation without being added (nonstarter LABs), influencing the final fermented product outcome. In the food fermentation industry, some LAB species benefit from a generally recognized as safe (GRAS) status in the United States and/or QPS in Europe, allowing them to be applied and considered safe for consumption as technological adjuvants.
LAB species are metabolically divided into obligate homofermentative (OHOM), facultative heterofermentative (FHET), and obligate heterofermentative (OHET) based on their carbohydrate metabolic pathways. OHOM genera, such as Lactococcus, Pediococcus, Enterococcus, Streptococcus, and some other Lactobacillaceae species (L. acidophilus, L. delbrueckii, and L. helveticus), utilize the Embden–Meyerhoff pathway for glucose metabolism, producing lactic acid as the main metabolic product, predominantly use phosphotransferase (PTS)‐based transport systems, and are not (with some exceptions) majorly able to ferment pentoses (Batt, 2014; Kandler, 1983; Lonvaud‐Funel, 2014). FHET genera, such as some Lactobacillaceae genera (L. casei, L. plantarum, and L. sakei, e.g.), metabolically behave as OHOM under excess of hexose concentrations but switch their metabolism to OHET under hexose‐limiting conditions, producing acetic acid, and ethanol and CO2 (apart from lactic acid) using the pentose‐phosphate pathway. OHET genera such as Leuconostoc, Oenococcus, and some other Lactobacillaceae genera such as L. brevis, L. fermentum, and L. reuteri only utilize the pentose‐phosphate pathway for hexose metabolism and usually permeases for sugar transport. They can ferment pentoses into mainly lactic and acetic acids (Batt, 2014; Kandler, 1983; Lonvaud‐Funel, 2014).
The evolution of Lactobacillales has been characterized as reflecting more gene loss rather than acquisition events, mainly regarding biosynthetic genes due to niche adaptation (Makarova & Koonin, 2007). It is suggested that 600–1200 genes have been lost, whereas less than 100 genes have been gained after the evolutionary divergence from the Bacilli ancestor (Makarova & Koonin, 2007). Differences have been observed among LAB species and strains and specific niche isolates. Milk, considered a nutrient‐rich substrate, has influenced the genetic niche‐specific adaptation of milk‐isolated LAB strains toward substantial gene loss events. For instance, S. thermophilus, which is mainly found in milk and has been used as a starter culture in yogurt and specific cheese production, is an example of an ongoing process of genome decay with an average genome size of 1.8 Mb in comparison to an average Bacilli such as Bacillus subtilis, which has a size of 4.2 Mb (Cavanagh et al., 2015; Makarova & Koonin, 2007). Moreover, specific milk‐isolated L. lactis strains have been characterized to possess plasmid complements that provide a phenotypic advantage in milk, generally involved in bacteriophage resistance, lactose utilization, bacteriocin production, and proteolytic activities. These genes could have been acquired through horizontal gene transfer (HGT) events such as conjugation, which has been the most probable HGT mechanism through LAB (Cavanagh et al., 2015). Other mobile elements, such as transposons that formed genomic islands and gene duplication events, have played an advantageous role for some L. casei strains throughout evolution and niche adaptability in milk and plant‐based niches (Cai et al., 2009). Group A L. casei strains in milk have mainly been through genome decay as they became “dairy adaptative specialists.” In contrast, Group B L. casei strains have acquired new genes through HGT, becoming “niche generalists” an environmental adaptative group of L. casei strains (Cai et al., 2009).
In the plant‐based niche, L. plantarum, which is mainly associated with living in plant‐based environments, is characterized by counterbalancing its ancestral gene loss with acquiring new genes throughout gene duplication and HGT primarily carried out by bacteriophages or conjugative pathways, having an average genome size of 3.3 Mb (Makarova & Koonin, 2007). For instance, genetic and phenotypical variation has also been found among different isolates of L. plantarum from olive brine, tomatoes, and grain fermentations. Yu et al. (2021) showed how only L. plantarum strains that were isolated from olive‐brined sources were able to degrade raffinose to some extent, demonstrating their niche adaptation to raffinose‐rich environments such as olives in comparison to the other sources. L. plantarum has been shown the ability to adapt and specialize in different plant‐based niches while at the same time sustaining intraspecific diversity as a way of promoting genetic stability in various environments (Yu et al., 2021).
Most of the intraspecific traits investigated to compare different niche isolates from the same species have focused on carbohydrate, protein, and amino acid metabolisms, which are crucial due to their high impact on the quality, flavor, and texture profiles of fermented products such as cheese (Buron‐Moles et al., 2019; Cai et al., 2009; Cavanagh et al., 2015; Okoye et al., 2022). Nevertheless, fewer studies have been made on the inter‐ and intraspecific genetic traits acquired by LAB strains related to the degradation of antinutrients and off‐flavor molecules present in plant‐based matrices. Understanding genetic variations, evolutionary events, and adaptability of LAB species in plant niches may also help select LAB strains as starter cultures for producing high‐quality PBFDAs.
4. ANTINUTRIENTS AND OFF‐FLAVORS
4.1. Antinutrients in plant‐based food matrices
Plant‐based food matrices used for the development of PBFDAs are usually divided into five main different categories: legumes (soybeans, peas, chickpeas), grains (oats), pseudograins (quinoa), drupes (cashew, almonds, coconut), and tubers (potato). These raw materials have been used as substrates to develop commercial plant‐based yogurts and cheese (Grasso et al., 2021). For example, yogurts made of soymilk and/or soybeans (Alpro) and oats (Oatly AB) (Fazer Oy), sometimes supplemented with potato protein isolate and/or starch, have been found in commercial retail (Grasso et al., 2020; Montemurro et al., 2021). On the other hand, pea, potato, soy, and cashew protein isolates have been some of the most used plant‐based matrices to produce plant‐based cheeses (Grasso et al., 2021). Based on those sources, phytic acid, saponins, phenolic compounds, lectins, protease inhibitors, RFOs, and oxalates are the antinutrients that mainly influence the nutritional (and sometimes sensorial) value of PBFDAs (Harper et al., 2022; Popova & Mihaylova, 2019; Pua et al., 2022; Samtiya et al., 2020, 2021).
Each raw material contains different macro‐ and micronutrients and antinutrients that affect the overall nutritional value of PBFDAs (Samtiya et al., 2020). Generally, the adverse effects of plant antinutrients are divided into two main antinutritional properties, which are the reduction of bioavailability of nutrients (amino acids, minerals, vitamins, and lipids) and the decrease of protein digestibility, apart from others that specifically have health adverse effects (Harper et al., 2022; Popova & Mihaylova, 2019; Pua et al., 2022; Samtiya et al., 2020, 2021). Moreover, plant metabolites that are considered antinutrients differ in their physicochemical nature, composition, and structure; for example, proteins (protease inhibitors), carbohydrates (RFOs), glycosylated compounds (saponins and phenolic compounds), and organic (oxalate and phytate) acids (Table 1). The removal, biotransformation, and/or degradation through microbial metabolism involve different pathways and specific enzymes for each compound. Furthermore, antinutrients that belong to the same physicochemical metabolic family might be structurally different from plant‐to‐plant family, leading to a more complex understanding of their degradation pathways and enzymatic hetereospecificity variability. As an example, steroidal saponins are secondary metabolites that are composed of sugar moieties (mono‐ and oligosaccharides) and aglycone structures, present in multiple plant sources such as legumes (soybean, peanuts, and chickpeas), tea, and cereals (oats). Soybean saponins, such as soyasapogenol A, have a different sugar moiety and aglycone than oat saponins, such as avenacoside A (Oleszek & Oleszek, 2020); thus, different enzymatic degradation complexes and/or different enzyme heterospecificity substrate affinity might be needed to be able to degrade them.
TABLE 1.
List of the main plant antinutrients, their composition and structure nature, health adverse effects, and the primary sources relevant when developing plant‐based fermented dairy alternatives (PBFDAs).
| Antinutrient | Types/forms | Nature of the compound | Sources | Main adverse effects |
|---|---|---|---|---|
| Phytic acid | Inositol with 1‐6P, bound to Zn2+, Fe2+, Mg2+, Ca2+ | Myoinositol is bound to 6 phosphate molecules bound to minerals. | Legumes, grains, pseudograins, drupes, tubers | Decrease plant mineral bioavailability |
| Saponins | Avenacosides, avenacins, soysaponins, α‐chaconine, α‐solanine, etc. | Triterpenes or steroidal hydrophobic aglycones attached to sugars (Glu, Ara, Gal, Rha, GlucA) | Legumes, grains, pseudograins, and tubers | Inhibition of digestive enzymes, intestinal inflammation |
| Phenolic compounds | Flavonoids, phenols, hydroxycinnamic acids, coumarins, stilbenes, xanthones, etc. | Monomeric or oligomeric phenol‐based compounds are sometimes attached to sugars (Glu, Gal, Rha, Xyl, Gluc and GlucA acids), organic acids, lipids, amines, etc. | Legumes, grains, drupes, and tubers | Some bind and precipitate proteins and bind minerals (Fe3+, Cu2+, Zn2+, Na+, Al3+) lowering the bioavailability of minerals and digestibility of proteins |
| Protease inhibitors | Trypsin, chemotrypsin, elastase, aspartic, serine protease inhibitors | Small tertiarily structured proteins with disulfide bonds | Legumes, grains | Inhibition of digestive enzymes decreases the digestibility of plant protein and other nutrients |
| Oligosaccharides (RFOs) | Raffinose, stachyose, verbascose | Tri, tetra‐, and penta‐oligosaccharides (Glu, Gal, Fru) | Legumes, tubers | Flatulence and abdominal discomfort |
| Oxalates | K+, Na+, Ca2+, Mg2+, Fe3+ bound oxalates | Organic acid (oxalic acid) bound to minerals | Legumes, grains, nuts, tubers | Decrease of mineral bioavailability |
Abbreviation: RFO, raffinose‐family oligosaccharide.
Some secondary plant metabolites are considered antinutrients for human consumption but play an essential role, among others, in plant metabolisms, influencing plant–microbiome, plant–plant, and microbiome–microbiome communications, the regulation of abiotic and biotic stresses, antibacterial and antifungal activities, and defense mechanisms (Cid‐Gallegos et al., 2022, Li et al., 2022, Pang et al., 2021). For example, phenolic compounds such as flavonoids, present in some legumes, bind to specific plant proteins to regulate the symbiosis between rhizobia species, attracting nitrogen‐fixing bacteria. In contrast, terpenoids regulate the growth of peculiar root‐associated bacteria, increasing Arenimonas proliferation and inhibiting Arthrobacter growth in Arabidopsis sp. (Pang et al., 2021). Other phenolic compounds, such as chlorogenic acid, act as herbivore‐resistant agents against aphids such as Aphis pomi and other insects such as Western thrips (X. Liu et al., 2017). Protease inhibitors are found in storage plant tissues and deplete insect digestive system functionality as a plant protection mechanism (Cid‐Gallegos et al., 2022). Oxalic acid influences plant metabolisms, such as ion homeostasis, pH regulation, metal detoxification, and plant defense (Li et al., 2022). RFOs serve as carbon storage sources for anticipating fluctuations in environmental conditions, providing energy and carbon in the seed germination period, and diminishing seed desiccation by water replacement and vitrification mechanisms (Elango et al., 2022). Phytic acid is the central phosphorus storage in seeds (López‐Moreno et al., 2022).
Although those plant compounds have been considered antinutrients in the last decades, few studies have started focusing on health‐promoting positive effects that some specific metabolites might have. For instance, flavonoids such as quercetin, which are phenolic compounds, have shown health‐promoting benefits in vivo by reducing the risk of cardiovascular diseases (Bravo, 1998). For instance, phenolic compounds have been associated with binding and precipitating proteins, thus reducing their digestibility and sometimes the action of digestive enzymes. However, this has only been shown for phenolic compounds with at least three flavonol monomers, such as polymeric tannins (Bravo, 1998). Conversely, some phenolics, such as ortho‐ and para‐diphenolics, have shown health‐promoting capabilities such as antioxidant and scavenging effects on free and peroxyl radicals, sometimes only effective in their aglycone (non‐glycoside) forms. Metabolite classification (Table 1) is based on physicochemical properties, not their nutritional health effect; hence, different compounds from the same family might have different antinutritional side effects. For instance, saponins from lentils, beans, and peas have been shown to inhibit digestive enzymes. In contrast, saponins from tea (Camellia sinensis) have been demonstrated to have anti‐inflammatory and antioxidant properties. In addition, saponins from garlic (Allium sativum) and oats (Avena sativa) were found to show antifungal and antimicrobial effects against specific microorganisms (Oleszek & Oleszek, 2020). Saponins from oats exert antioxidant activity but increase intestinal permeability when consumed (Oleszek & Oleszek, 2020). Although RFOs are mostly considered antinutrients because they cause flatulence and abdominal discomfort, they have lately been associated with some positive attributes, such as prebiotic, anti‐allergic, anti‐obesity, and anti‐diabetic properties (Elango et al., 2022).
4.2. Volatile off‐flavors in plant‐based food matrices
Plant volatile organic compounds (VOCs) consist of a wide range of secondary metabolites generally characterized by their lipophilic and low molecular weight nature with high vapor pressure at ambient temperatures (Dudareva et al., 2013; Picazo‐Aragonés et al., 2020). There are multiple classes of VOCs, differing from their biosynthetic origin in plants and physicochemical properties, such as amino acid derivates, fatty acid derivates, terpenoids, and phenylpropanoid/benzenoid compounds (Dudareva et al., 2013; Picazo‐Aragonés et al., 2020). VOC production and regulation in plants is a complex multifactorial process that involves multiple factors such as specific activities of certain biosynthetic enzymes, the concentration of substrate, precursor availability, epigenetics, and environmental factors such as plant nutrient concentration, uptake, and availability, temperature, humidity, light intensity, among others (Dudareva et al., 2013; Picazo‐Aragonés et al., 2020). Interestingly, VOCs have been considered relevant for plant evolution and environmental adaptation because their involvement in defense and pollinator attraction explains their primary production in flowers. However, VOCs are also produced in leaves, fruits, and stems. They play an essential molecular role in plant–plant, plant‐pollinator, and plant‐herbivore interactions (Dudareva et al., 2013; Picazo‐Aragonés et al., 2020).
Apart from their influence on plant metabolism and biology, plant VOCs affect the overall flavor profiles of plant‐based foods, consequently affecting consumer acceptance of certain food products. It is considered one of the significant consumer drawbacks when producing PBFDAs, along with appearance and texture (Gupta et al., 2022; Short et al., 2021; Yilmaz‐Ersan & Topcuoglu, 2022). As in milk products such as yogurts and fresh, hard, and mold‐based cheeses, plant‐based dairy alternatives have different flavor profiles due to their difference in the raw material used, ingredients used, fermentation process, starter cultures (if inoculated), and the overall production processes employed. Nevertheless, based on flavor and Western consumer acceptance studies of PBFDAs, some flavor attributes have generally been associated with off‐flavor traits, such as “beany,” “green,” and “earthy” notes (Sedó Molina, Shetty, et al., 2024). VOCs denoting those flavor attributes differ depending on the plant‐based raw material used, processing, and formulation. In addition, the presence of VOCs alone does not entirely explain the flavor profile of such food because flavor is a multifactorial and often subjective phenomenon. VOC concentration is crucial to be considered due to its direct influence on perception (Spence & Kilcawley, 2021).
Although the current analytical chemical techniques can detect the presence of hundreds of VOCs in each food and beverage product, the average human nose can detect 30–40 VOCs that are considered to significantly contribute to the overall flavor perception (Spence & Kilcawley, 2021). VOCs mainly considered off‐flavors in PBFDAs belong to aldehydes, alcohols, ketones, acids, pyrazines, furans, and sulfur compounds (Saffarionpour, 2023). They belong to different VOC classes and mainly originate from lipid oxidation through lipoxygenase pathway (LOX), autooxidation, and amino acid derivatives generated by decarboxylation, aminotransferase, deamination, and aldehyde synthesis (Dudareva et al., 2013). VOCs with the highest impact on flavor perception have lower perception odor thresholds (the compound's lowest concentration that humans can smell), such as aldehydes, furans, and pyrazines.
Table 2 lists the VOC off‐flavors typically found in plant‐based raw materials, such as peas, soybeans, oats, almonds, cashews, walnuts, and potatoes used to generate PBFDA. Aldehydes are the most common off‐flavors found in PBFDA and are usually generated from Strecker degradation, autooxidation, or LOX‐mediated oxidations. Alcohol and ketones are typically generated from aldehydes through alcohol dehydrogenase (ADH) reactions in plants. Moreover, ketones might be produced through aldol condensations of aldehydes or from decarboxylation reactions of 3‐oxo‐acids. Furans and pyrazines are typically produced through Maillard reactions from polyunsaturated fatty acids (among others) and α‐aminocarbonyl, respectively. Sulfur compounds are mainly produced from methanethiol oxidation, a compound formed from Stecker degradation of methionine (Fischer et al., 2022; Leonard et al., 2022; Saffarionpour, 2023; Wang et al., 2023; C. Zhang et al., 2020). It has been demonstrated how off‐flavor molecules, such as aldehydes, pyrazines, alcohols, ketones, and furans, can interact with plant proteins through multiple interactions such as hydrophobic, hydrogen, or covalent bonds. Degradation of plant proteins through fermentation might cause an increase in protein solubility and a decrease in hydrophobicity surface, thus releasing off‐flavor compounds (Leonard et al., 2022; Saffarionpour, 2023; C. Zhang et al., 2020).
TABLE 2.
Volatile organic compounds (VOCs) list that contribute to off‐flavor attributes in the production of plant‐based fermented dairy alternatives (PBFDAs) (CAS number, thresholds, “odor description”).
| Group of VOCs | Compound | CAS number | Odor description | Odor threshold in water (ppb) |
|---|---|---|---|---|
| Aldehyde | 2‐Heptenal | 18829‐55‐5 | Green, spicy, hazelnut, apple | 13 |
| (E)‐2‐Hexenal | 505‐57‐7 | Leafy, apple, cheesy, vegetable, fat, banana, rancid, fatty, sweet, plum, fruity, aldehydic, almond, green | 17 | |
| Nonanal | 124‐19‐6 | Citrus, lime, orange peel, rose, fat, green, fishy, waxy, fresh, fatty, aldehydic, orris, grapefruit | 1 | |
| Benzaldehyde | 100‐52‐7 | Cherry, almond, sweet, burnt sugar, sharp, strong, bitter | 350–3500 | |
| Hexanal | 66‐25‐1 | Leafy, grass, sweaty, tallow, fat, fresh, fatty, fruity, aldehydic, green | 4.5–5 | |
| Octanal | 124‐13‐0 | Lemon, citrus, soap, orange peel, fat, waxy, fatty, aldehydic, green | 0.7 | |
| Heptanal | 111‐71‐7 | Citrus, ozone, fat, herbal, fresh, wine‐lee, rancid, fatty, aldehydic, green | 3 | |
| Pentanal | 110‐62‐3 | Almond, bitter, malt, oil, pungent | 12–42 | |
| 2,4‐Decadienal (E, E) | 25152‐84‐5 | n.a. | 0.07 | |
| Alcohols | 1‐Penten‐3‐ol | 616‐25‐1 | Butter, fish, green, oxidized, wet earth | 400 |
| 1‐Octen‐3‐ol | 3391‐86‐4 | Raw, fishy, oily, earthy, fungal, chicken, mushroom, green | 1 | |
| 1‐Pentanol | 71‐41‐0 | Oil, balsamic, vanilla, fusel, sweet, balsam | 4000 | |
| 1‐Hexanol | 111‐27‐3 | Oil, alcoholic, ethereal, resin, fuel, sweet, fruity, flower, green | 2500 | |
| 1‐Heptanol | 111‐70‐6 | Leafy, coconut, herbal, peony, strawberry, chemical, musty, sweet, woody, violet, green | 3 | |
| 1‐Nonanol | 143‐08‐8 | Fat, floral, green, oil | 90 | |
| 2‐Ethyl‐1‐hexanol | 104‐76‐7 | Green, rose | 270,000 | |
| Carboxylic acids | Hexanoic acid | 142‐62‐1 | Cheese, oil, pungent, sour | 3000 |
| Esters | Ethyl acetate | 141‐78‐6 | Pineapple, ethereal, sweet, anise, fruity, balsam, weedy, green | 5–5000 |
| Furans | 2‐Pentylfuran | 3777‐69‐3 | Green, bean, vegetable, earthy, beany, fruity, metallic, green | 6 |
| 2‐Ethylfuran | 3208‐16‐0 | Earthy, sweet, burnt, malty | – | |
| Ketones | 3‐Hexanone | 589‐38‐8 | Grape, fruity | n.a. |
| 5‐Methyl‐2‐hexanone | 110‐12‐3 | n.a. | 2.1 | |
| 2‐Nonanone | 821‐55‐6 | Fragrant, fruit, green, hot milk | 5–200 | |
| 1‐Octen‐3‐one | 4312‐99‐6 | Herbal, earthy, metal, musty, mushroom, dirty | 0.005 | |
| 2‐Octanone | 111‐13‐7 | n.a. | 50 | |
| 2‐Hexanone | 591‐78‐6 | Ethereal, ether | 76–3000 | |
| 2‐Heptanone | 110‐43‐0 | Coconut, soap, herbal, sweet, woody, fruity, spicy, cinnamon | 140–3000 | |
| Pyrazines | 2‐Isobutyl‐3‐methoxypyrazine | 24683‐00‐9 | Spice, pepper, green pepper, bell, earth, galbanum, pea, green | 0.002–0.016 |
| 2‐Isopropyl‐3‐methoxypyrazine | 25773‐40‐4 | Earthy, beany, pea, earthy | 0.0002 | |
| 2‐Methoxy‐3‐isopropyl‐5‐methylpyrazine | 32021‐41‐3 | Green, pea, earthy | – | |
| Sulfur compounds | Dimethyl sulfide (DMS) | 75‐18‐3 | Sulfurous, cabbage, onion, vegetable, sweet corn, tomato, radish, wild, green | 0.3–1 |
| Dimethyl disulfide (DMDS) | 624‐92‐0 | Citrus, cabbage, sulfurous, putrid, onion, vegetable, fatty, floral, earthy, woody, fruity, herbaceous, nutty, green | 0.16–12 | |
| Dimethyl trisulfide (DMTS) | 3658‐80‐8 | Cabbage, fish, onion, sulfur | 0.005–0.01 | |
| Dibutyl disulfide (DBDS) | 211‐091‐3 | Sulfurous | 0.21 |
Some VOCs could be used as chemical signatures influencing off‐odor perception due to their high concentrations in plant bases and low odor thresholds, such as hexanal, 2‐pentylfuran, nonanal, heptanal, octanal, 1‐octen‐3‐ol, 1‐octen‐3‐one, (E, E)‐2,4‐nonadienal, 2,4‐E, E‐decadienal, pyrazines, and sulfur compounds that are associated with “green,” “beany,” and “earthy” notes. Commonly, hexanal, (E, E)‐2,4‐nonadienal, and (E, E)‐2,4‐decadienal have been considered to contribute, majorly, to the off‐flavor notes in pea and soy (C. Zhang et al., 2020) and oat milk (McGorrin, 2019), based on combining analytical and sensory analyses. Additionally, pyrazines and 1‐octen‐3‐one were also associated explicitly with off‐odors in pea and soy milk, respectively (C. Zhang et al., 2020), and (E, E, Z)‐2,4,6‐nonenal and 2‐methyl‐3‐furanthiol mainly contributed to the oat characteristic aroma in oat milk (McGorrin, 2019). In nut‐based milk, hexanal, heptanal, and nonanal were found to be present at significantly higher concentrations in walnut, peanut, and almond (Type 1) milk, respectively (Manousi & Zachariadis, 2019). Differences in VOCs prevalence in plant‐based milks are associated with different activities of LOX‐1, ‐2, ‐3, ADH, and hydroperoxide lyase (C. Zhang et al., 2020). Moreover, other parameters, such as crop varieties, processing, and the different vegetable‐based oils used for milk formulations, influence the concentration and presence of the different VOCs. The metabolic involvement of LAB in degrading off‐flavor VOCs present in PBDA milk through fermentation is discussed in Section 6.
5. METABOLIC INSIGHTS OF LAB: BIOTRANSFORMATION OF PLANT‐BASED ANTINUTRIENTS
Genes and enzymes involved in the transport and degradation of phytates, RFOs, saponins, phenolic compounds, enzyme inhibitors, and oxalates, documented in LAB species and strains, are discussed in depth and summarized in Figure 1 and Table 3.
FIGURE 1.
Lactic acid bacteria (LAB) metabolic pathways involved in raffinose‐family oligosaccharides (RFOs), phytate, saponin, phenolic acid, oxalate, and enzyme inhibitor metabolisms. All enzymes are marked in bold. LacS (sugar transport permease), rafP (raffinose permease), levS (levansucrase), ftfA (levansucrase), melA (α‐galactosidase), phy (phytase), glu (β‐glucosidase), gal (β‐galactosidase), xyl (β‐xylosidase), ram (α‐rhamnosidase), gus (β‐glucuronidase), tan (tannase), lpdC (gallate decarboxylase), est_1092 (estA, feruloyl esterase), pad (phenolic acid decarboxylase), hcrB (flavocytochrome c), hcrF (FAD‐dependent oxidoreductase), par1/par2 (FAD‐binding reductases), frc (formyl‐coA transferase), oxc (oxalyl‐coA transferase), prt (envelope‐proteinase), opp (oligopeptide transport system), dpp (di‐/tripeptide transport system). Source: Created with BioRender.
TABLE 3.
Summary of transport systems, enzymes, and genes characterized in lactic acid bacteria (LAB) species and strains, including the specific target substrates tested and the factors influencing those phenotypes, for the degradation of phytates, raffinose‐family oligosaccharides (RFOs), saponins, phenolic compounds, protease inhibitors, and oxalates.
| Antinutrients | Transport systems | Relevant enzymes and genes | LAB species and strains (phenotypically and/or genotypically characterized) | Substrates | Factors influencing phenotypes | References |
|---|---|---|---|---|---|---|
| Phytates | Unknown | phytase (phy, phyLP), tyrosine‐like phosphatase (phyLf), putative phosphatases | Lactiplantibacillus plantarum (CRL1964, JBPRS, DPC2739), Lentilactobacillus buchneri, Pediococcus pentosaceus, Levilactobacillus brevis a , Bifidobacterium spp., Limosilactobacillus reuteri, Limosilactobacillus. fermentum (DSM 20052, NKN51), Companilactobacillus alimentarius, Lactobacillus acidophilus, Lacticaseibacillus casei 40 W, Lactobacillus amylovorus (B4552 b ), Lacticaseibacillusrhamnosus, Fructilactobacillus sanfranciscensis 22E and CB1, Fructilactobacillus fructivorans DA106, Lactiplantibacillus pentosus CECT 4023 | Phytate, p‐nitrophenyl phosphate, d‐fructose‐6‐phosphate, and d‐glucose‐6‐phosphate, sodium phytate, α‐phospho‐dl‐serine/threonine/tyrosine | pH, temperature, metal ions, phosphoric ester bonds type, C‐source, substrate specificity | Andrabi et al. (2016), Damayanti et al. (2017), De Angelis et al. (2003), García‐Mantrana et al. (2015), Haros et al. (2005), Lee et al. (2013), Palacios et al. (2005), Sandez Penidez et al. (2020), N. Sharma et al. (2020), R. Sharma et al. (2018), Sreeramulu et al. (1996), Sumengen et al. (2013), Sümengen et al. (2012), Zamudio et al. (2001) |
| RFOs | PTS transporter (lacS), ABC transporter, ECF‐type transporter (M6PI), permeases (rafP) | α‐Galactosidases (galA, melA), levansucrase (levS, ftFA), sucrose phosphorylase (scrP, gtfA), fructofuranosidase (sacA/sacB, bfrA) | Leuconostoc mesenteroides PM173 and pseudomesenteroides NFICC 99 and 2004, Lacticaseibacillus paracasei ssp. paracasei, Streptococcus thermophilus S3, L. plantarum, L. fermentum LF314, L. brevis, L. reuteri ATCC 23271, L. casei, L. curvatus, and L. acidophilus ATCC 23271, B. lactis HNO19, P. pentosaceus AM173 | Raffinose, stachyose, verbascose, melibiose | Substrate specificity, plant food matrices | Carević et al. (2016), Gänzle and Follador (2012), Hati et al. (2014), Kahala et al. (2023), Kanwal et al. (2023), Oguntoyinbo (2007), Sedó Molina et al. (2022), Sedó Molina, Shetty et al. (2024), Silvestroni et al. (2002); Tsangalis and Shah (2004), Wätjen et al. (2023), Yoon and Hwang (2008), Zartl et al. (2018), Zhao and Gänzle (2018), Zhong et al. (2018) |
| Saponins | PTS systems (phospho‐β‐glucosidases and phospho‐β‐galactosidases) | β‐d‐Glucosidases (gluA, GH1,2,3, 30), β‐galactosidases (bgl, GH1,2,3,30,35,43), α‐L‐rhamnosidases (ramA), α‐L‐arabinosidases (araA), β‐D‐xylosidases (xynB), xylose isomerase (xylA) and β‐glucuronidases (gus, GH30) | L. plantarum (USC1 b and DSM 20205), L. casei ATCC 393, O. oeni ATC BAA‐1163, L. brevis SK3 and L. mesenteroides, L. pentosus 6105, Companilactobacillus paralimenterius LH4, Lactobacillus bulgaricus L3, L. acidophilus (NCFM and DSM 9126), Lactobacillus crustorum | ginsenoside Rb1 saponin, steroidal saponins, tea and herbal saponins, soysaponins | Substrate specificity, type of sugar moiety, growth C‐source | Acin‐Albiac et al. (2021), Carević et al. (2016), Coulon et al. (1998), Fu et al. (2020), Gueguen et al. (1997), He et al. (2019), Michlmayr and Kneifel (2014), Overney and Huang (2020), Pham et al. (2023), Qian et al. (2018); Sakurama et al. (2014), Sestelo et al. (2004), Theilmann et al. (2017), Ventura et al. (2009) |
| Phenolic compounds | Unknown, PTS systems (phospho‐β‐glucosidases) | Hydroxycinnamic acid esterases (lP_0796, lj0536, lj1228, hceP, and est_1092), phenolic acid decarboxylase (padC), hydrocinnamic acid reductases (hcrB. hcrF, and par1/par2), tannases (tanA, tanB), vinyl phenol reductases (vrpA), hydroxybenzoic acid decarboxylases (lpdC), β‐glucosidases (gluA), rhamnosidases (ram1), β‐glucurunosidases (lcGUS30) | L. plantarum (ATCC 14917, WCFS1, TMW1.460, DSM 1055, DSM 20205), L. fermentum FUA3589, L. johnsonii N6.2, Furfurilactobacillus millii FUA 83583, L. brevis, L. acidophilus NCFM, L. lactis, P. pentosaceus, and L. brevis FERM BP‐4693, L. crispatus, L. helveticus | ferulic acid, caffeic acid, p‐coumaric acid, oleuropein, flavonoids, tannic acid, chlorogenic acid, rosmarinic acid, hydroxycinnamic methyl and ethyl esters, gallocatechin, epigallocatechin gallate | Substrate specificity, type of sugar moiety (glycosidases), cell physiological state, outcompeting gene regulation | Gaur et al. (2020, 2023), Gaur and Gänzle (2023), Jiménez et al. (2013, 2014), Landete et al. (2021), Michlmayr and Kneifel (2014), Sakurama et al. (2014), Santamaría et al. (2018) |
| Protease inhibitors | Di‐ and oligopeptide intracellular transporters (opp, dpp) (hydrolyzed peptides from protease inhibitors) | Proteinases (CEPs—prt) and proteases | L. paracasei (Lpa4), L. pseudomesenteroides (Leps1), P. pentosaceus (Pp3), L. corynniformis (Lco4), L. plantarum (Lpl5 and NRRL B‐4496), L. salivarius H32.1, L. mucosae D5a1, and L. rhamnosus LE3, L. brevis and L. lactis | KTI and BBI inhibitors (soy, chickpeas, peas, wheat) | Substrate specificity and growth conditions (pH, redox balance) | Çabuk et al. (2018), Caminero et al. (2019), Fraberger et al. (2020), Gao et al. (2013), Sáez et al. (2022) |
| Oxalates | Putative permeases and ABC transporters. | Formyl coenzyme A transferase (frc) and oxalyl coenzyme A decarboxylase (oxc) | L. acidophilus (NCFM), L. gasseri (ATCC 33323), L. plantarum, L. fermentum, L. murinus (1222, 3133, 5333), L. animalis (5323, 5241, 5342, 5121, 1221, 6331, 223C), L. reuteri, L. paracasei, Weissella confusa, W. cibaria, L. rhamnosus, Loigolactobacillus bifermentans, L. salivarius and Bifidobacterium animalis subsp. lactis BI07, Bifidobacterium dentium Bd1 | Oxalate | pH, oxalate intracellular transport | Azcarate‐Peril et al. (2006), Cho et al. (2015), Gomathi et al. (2014), Kullin et al. (2014), Lewanika et al. (2007), Murphy et al. (2009), Tavasoli et al. (2022), Turroni et al. (2010), Ventura et al. (2009), Youssef (2024) |
aRepresents strains characterized to produce the corresponding enzyme/s extracellularly and intracellularly.
bRepresents strains that have been characterized to produce the corresponding enzyme/s extracellularly.
Abbreviation: PTS, phosphotransferase.
5.1. Phytates
Phytic acid (myoinositol hexaphosphate) is a crucial plant metabolite responsible for 50%–80% phosphorus storage in plants, particularly cereals, seeds, and legumes. This compound is often considered an antinutrient due to its biochemical ability to chelate minerals such as Zn2+, Fe2+, and Mg2+, affecting their solubility, absorption, and bioavailability in the GIT (López‐Moreno et al., 2022; N. Sharma et al., 2020; Vats & Banerjee, 2004). Its negative charge facilitates the chelation reaction with positively charged divalent and trivalent cations. Furthermore, phytic acid has been associated with binding proteins and carbohydrates, impacting their digestibility and diminishing the nutritional value of phytate‐containing foods (Popova & Mihaylova, 2019). Degradation of phytic acid by microbial fermentation has been mainly associated with a phytase (myoinositol hexakisphosphate phosphohydrolase), encoded by a gene phy, a histidine acid phosphatase that catalyzes the hydrolysis of phosphomonoester bonds of phytates (IP6) into myoinositol phosphate derivatives (IP5, IP4, IP3, IP2, IP) releasing myoinositol and six inorganic phosphates, depleting the ability to chelate metal ions (Figure 1) (Vats & Banerjee, 2004).
There are two main types of phytases: 3‐phytase (EC 3.1.3.8) and 6‐phytase (3.1.3.26), depending on the position of the ester bond hydrolyzed (Vats & Banerjee, 2004). This enzyme is also found in plants (phyt1, phyt2) such as soybean, oats, canola seeds, spelt, and maize, which might be activated through the lactic acid fermentation with optimal activities at pH between 4.5 and 5 (Vats & Banerjee, 2004). Furthermore, differences in the origin of phytase synthesis, food matrix, pH, temperature, metal ions present, and the type of phosphoric ester bonds of the target molecule might influence phytase activities. Apart from phytases, other non‐specific phytate phosphatases that were able to dephosphorylate phytates thereof having phytase activities, such as tyrosine phosphatase‐like phytase (Phylf), described in L. fermentum NKN51, have been characterized (R. Sharma et al., 2018). Phenotypical studies in LAB strains have also been performed in cereal and legume‐based fermentations, demonstrating their ability to decrease phytate content. LAB species such as L. plantarum, L. buchneri, P. pentosaceus, L. brevis, Bifidobacterium spp., L. reuteri, L. fermentum, C. alimentarius, L. acidophilus, L. casei, L. amylovorus, L. rhamnosus, F. sanfranciscensis, among others, have been characterized to express this phenotype in specific conditions (Andrabi et al., 2016; De Angelis et al., 2003; García‐Mantrana et al., 2015; Haros et al., 2005; Lee et al., 2013; Sandez Penidez et al., 2020; R. Sharma et al., 2018; Sreeramulu et al., 1996; Sumengen et al., 2013; Sümengen et al., 2012; Zamudio et al., 2001). For instance, phytase from Fructilactobacillus fructivorans DA106 showed the second‐highest phytase activity (295.3 U/mL) among 13 LAB strains tested, whereas the one from F. sanfranciscensis 22E did not show high phytase expression (De Angelis et al., 2003). Other studies have demonstrated a substrate multispecificity of the phytases within LAB strains. F. sanfranciscensis CB1 phytase (50 kDa) had a substrate multispecificity toward acetyl phosphate and adenosine‐phosphate derivatives, among others, also showing higher activities with other substrates such as p‐nitrophenyl phosphate (De Angelis et al., 2003). L. pentosus CECT 4023 acid phosphatase (69 kDa) showed multispecificity towards other phosphate‐bound molecules such as p‐nitrophenyl phosphate, d‐fructose‐6‐phosphate, and d‐glucose‐6‐phosphate, with higher activities than toward phytate (Palacios et al., 2005). Phytase from L. plantarum CRL1964 (55 kDa) showed higher substrate specificity towards sodium phytate than to other substrates (Sandez Penidez et al., 2020). Other factors can influence LAB's phytase activity, as seen with the selection of carbon source and growth phase. For instance, F. sanfranciscensis CB1, a species mainly found in traditional sourdough fermentation, intracellularly produced phytase with an activity of 420.8 U/mL when sodium phytate was present. Its expression was increased by 35% when this strain was grown with maltose and fructose in comparison to other carbohydrates (maltose with glucose or sucrose), which was correlated by the cell counts reached after fermentation (De Angelis et al., 2003). Its highest peak of phytase activity was during the stationary phase. In addition, L. pentosus CECT 4023 intracellularly produces an acid phosphatase with phytase activity at maximum rates during its stationary growth phase. This was reduced by high concentrations of glucose (2% w/v) and optimal at 0.5% glucose‐MRS (Palacios et al., 2005). However, L. amylovorus B4552 showed the highest phytase activity extracellularly (125–146 U/mL) with glucose preferably and during the exponential growth phase (Sreeramulu et al., 1996). L. plantarum CRL1964 was determined to express significantly higher specific phytase (pLhyLP) activity when growing in raffinose, maltose, and xylose in comparison to glucose, fructose, sucrose, and galactose (Sandez Penidez et al., 2020). Phytase relative activity was increased by the presence of raffinose and lactose (20 g/L) in comparison to glucose (20 g/L) in L. casei 40 W and L. fermentum DSM 20052 and the presence of fructo‐oligosaccharides in L. plantarum JBPRS (Damayanti et al., 2017). In all strains tested, phytase activity was severely reduced when inorganic phosphate was added to the growth media, although growth yields were unaffected (Damayanti et al., 2017). Finally, metal ions and other chemical compounds enhance or deplete phosphatase and phytase activities. L. pentosus CECT 4023 phosphatase activity was increased by adding 1 mM of CoCl2 (15× factor) and l‐ascorbic acid (1.25× factor) (Palacios et al., 2005), whereas in L. fermentum NKN51 was increased by 5 mM Ag+ (1.5× factor) (R. Sharma et al., 2018). Intracellular phytase relative activity was increased in L. brevis with 1 mM of HgCl2, MnCl2, MgCl2, CuCl2, and FeCl2 by 1.4, 1.4, 1.3, 1.3, and 1.2 factors, respectively, whereas extracellular phytase activity was decreased by all compounds tested (Sümengen et al., 2012). In another study, PhyLP was increased by 10 mM ascorbic acid and Co2+ (Sandez Penidez et al., 2020). Even though an acid phosphatase from L. plantarum DPC2739 had a considerable range of substrate specificity (hydrolyzing phospho‐ester bonds from α‐phospho‐dl‐serine/threonine/tyrosine, d‐glucose‐6‐phosphate, and d‐fructose‐6‐phosphate, among others), no phytate specificity was tested (Magboul & McSweeney, 1999). Moreover, based on the blast analysis (Figure 2), phyLf was found homologous in all L. fermentum and L. brevis strains, setting the positive hit threshold on 50% sequence homology and E‐value < 0.0005. No hits were found with phyLP in any strains analyzed (Section 2.1—data not shown).
5.2. Raffinose‐family oligosaccharides (RFOs)
Raffinose‐family oligosaccharides (RFOs) are soluble oligomeric carbohydrates in the plant kingdom that are considered α‐galactosyl derivatives of sucrose. They are particularly abundant in the legume family (Fabaceae), such as soybean (0.1–19.9 g/kg), lentil (4.5–5.5 mol/100 g), chickpea (1.2–48.5 g/kg), and peas (22.6–63.4 g/kg) (Elango et al., 2022). The most common antinutritional attributes associated with RFOs are indigestion and flatulence caused during its transition in the upper intestine. The lack of RFOs‐degrading enzymes in the human gut triggers their accumulation in the large intestine and finally ferments them, producing hydrogen, methane, and CO2 (Elango et al., 2022; Kanwal et al., 2023). There are five major RFO di‐ and oligosaccharides: melibiose, raffinose, stachyose, verbascose, and ajugose. Their main difference is their polymeric length, in which melibiose (Gal–Glu), raffinose (Gal–Glu–Fru), stachyose (Gal–Gal–Glu–Fru), and verbascose (Gal–Gal–Gal–Glu–Fruc) are constituted by two, three, four, and five monosaccharides, respectively. Structurally, they differ in the number of galactose monomers linked to sucrose (α‐d‐glucose and β‐d‐fructose) through glycosidic bonds to the glucose monomer.
Several microorganisms have shown the ability to degrade RFOs through the action of two main enzymes: α‐galactosidases (galA, melA) (EC 3.2.1.22), which hydrolyze the α‐1,6 bond between Gal–Glu and Gal‐Gal releasing galactose and sucrose (from raffinose) or galactose and glucose (from melibiose), and levansucrases (EC 2.4.1.10) such as evS (F. sanfransciscensis) and FtFA (L. reuteri), which are cell‐wall glycosylases that degrade RFOs cleaving β‐1,2 (Glu–Fru) and transferring the fructose monomer to a fructosyl acceptor, generating fructose and melibiose (Figure 1) (Gänzle & Follador, 2012). Other genes participate in RFO metabolism, especially in sucrose, such as sucrose phosphorylase (EC 2.4.1.7) (ScrP, GftA) and fructofuranosidase (EC 3.2.1.26) (SacA/SacB, BfrA), which break down sucrose through phosphorylation or hydrolysis, respectively (Gänzle & Follador, 2012).
In Figure 2, the homology‐based analysis considered two different protein sequences of MelA (L. fermentum and L. mesenteroides) and LevS (L. salivarius and L. mesenteroides). On average, L. plantarum, L. aviarius, and L. mucosae encoded the highest number of genes involved in RFO metabolism. In contrast, C. divergens, Companilactobacillus spp., Lacticaseibacillus spp., L. gasseri, L. amylolyticus, L. sakei, L. curvatus, L. lactis, L. dextrinicus, L. coryniformis, S. thermophilus, and P. parvulus did not encode for any gene or only one from the ones considered in RFO metabolism. B. bifidum showed the highest average gene number from Bifidobacterium spp. encoding for two copies of BfrA in each strain. Specific strains encoded multiple copies of the same gene, such as one L. aviarius, two L. plantarum, and one L. mesenteroides strains encoded 13, 7, and 4 copies of levS in their genome. Moreover, the highest copies of melA were found in L. mucosae DSM 13345, L. gallinarum DSM 10532, and L. acidophilus, with 6, 3, and 4 copies. Levansucrases (FtfA) were highly present in F. sanfranciscensis (1–2 copies), L. pentosus (0–4 copies), L. helveticus (up to 6 copies), and L. fermentum (1–2 copies).
RFOs are transported in LAB through specific transport systems such as PTS, ABC, and permeases, among others, and subsequently hydrolyzed intracellularly by the action of glycolytic enzymes mentioned (Kanwal et al., 2023; Zhong et al., 2018). Based on the proteomic and transcriptomic analysis, OHOM LAB strains, such as L. acidophilus CICC 22162, have been shown to use PTS systems and ECF‐type transporters to transport stachyose intracellularly. Moreover, higher expression of M6PI isomerase (EC 5.3.1.8) that converts β‐d‐mannose‐6‐phosphate (M6P) to d‐fructose‐6‐phosphate (F6P) was observed in comparison to the control grown in glucose, indicating fructose as one of the products of stachyose degradation (Zhong et al., 2018). In OHET LAB, such as L. reuteri, extracellular levansucrase activities are preferred due to a facilitated transport of less complex oligosaccharides (Gänzle & Follador, 2012). Nevertheless, L. reuteri has also been suggested to transport raffinose through LacS and has also shown broad carbohydrate specificity to lactose and melibiose and consecutively cleaved by GalA intracellularly, releasing galactose, melibiose, and/or sucrose oligosaccharides. Galactose might be catabolized by the expression of the gal operon (galM, galK, galT, galE, and galU) to glucose‐1‐P and enter the phosphoketolase pathway. Sucrose might be degraded into glucose and fructose by invertase and follow the same metabolic pathway (Zhao & Gänzle, 2018). In FHET LAB, such as L. plantarum, raffinose transport has been associated with rafP, located in the upstream part of the melA gene, and has high homology with lacS from S. thermophilus. However, the latter has a low affinity to raffinose (Silvestroni et al., 2002).
LAB species such as L. mesenteroides and L. pseudomesenteroides, L. paracasei ssp. paracasei, S. thermophilus, L. plantarum, L. fermentum, L. brevis, L. reuteri, L. casei, L. curvatus, and L. acidophilus have been documented to be able to degrade raffinose, stachyose, and/or verbascose during fermentation of plant‐based raw materials, such as soy milk (Garro et al., 1993; Hati et al., 2014; Oguntoyinbo, 2007; Sedó Molina, Shetty, et al., 2024; Yoon & Hwang, 2008). Bifidobacterium spp. also showed α‐galactosidase activity (Carević et al., 2016; Tsangalis & Shah, 2004). As the structural complexity of RFOs increases, fewer LAB strains can metabolize them; for example, only a limited number of strains such as L. reuteri, P. pentosaceus, L. mesenteroides, and Bifidobacterium lactis HNO19 have been able to degrade verbascose, a pentasaccharide (Kahala et al., 2023; Kanwal et al., 2023), among 14 LAB strains RFOs‐degrading tested.
α‐Galactosidase is a glycosyl hydrolase that belongs to the GH36 glycosidase family and binds to unbranched oligosaccharides such as melibiose, raffinose, and stachyose, hydrolyzing α‐1,6 bond between Gal and Glu (Kanwal et al., 2023). They are classified into two groups (Groups I and II); the first hydrolyzes RFOs, and the second hydrolyzes higher polymer galactomannans (Zartl et al., 2018). Carević et al. (2016) showed how L. reuteri ATCC 23271 was the most promising strain among those tested due to its high α‐galactosidase activity (1.27 IU/mL), followed by L. acidophilus ATCC 23271 (0.25 IU/mL) and S. thermophilus S3 (0.181 IU/mL). L. rhamnosus ATCC 7469, Lactobacillus bulgaricus ATCC 11842, and L. helveticus ATCC 15009 did not show activity. Furthermore, in a study conducted by Kahala et al., 2023, P. pentosaceus (AM173) and L. mesenteroides (PM173), both isolated from pulses, showed the highest consumption rates of raffinose, stachyose, and verbascose present in fava bean and pea after 48 h of fermentation. In this study, some LAB strains, such as L. fermentum (LF314), showed high abilities to degrade raffinose and stachyose but lower when degrading verbascose. This case is an example of substrate specificity of α‐galactosidase, even though the last galactosyl monomer in verbascose is attached with the same type of bond as the galactosyl monomer in stachyose. Moreover, Weisella cibaria (AM3) showed a high ability to degrade all RFOs in fava beans but low performance in pea fermentation, suggesting plant‐matrix dependency influenced by the ability to metabolize fermentable nutrients provided by the specific raw material (Kahala et al., 2023). Additionally, certain LAB strains, such as B. lactis HNO19, have demonstrated better growth on stachyose than on raffinose. This may be due to a preference for consuming galactose over glucose or fructose, as stachyose contains an extra galactose monomer in its structure (Zartl et al., 2018).
Several studies have considered L. pseudomesenteroides strains (NFICC 99 and 2004) as RFOs high‐degrading candidates in soy milk (Sedó Molina et al., 2022; Sedó Molina, Shetty, et al., 2024; Wätjen et al., 2023) as well as promising candidates for plant‐based fermentations due to their high proteolytic capabilities toward plant proteins (Christensen et al., 2022; Sedó Molina, Shetty, et al., 2024). Among QPS Leuconostoc species, over 95% of L. pseudomesenteroides strains can ferment raffinose, whereas only a few L. mesenteroides strains are capable; however, neither Leuconostoc lactis nor L. citreum can ferment raffinose (Lonvaud‐Funel, 2014). Generally, dairy‐derived L. lactis species cannot ferment raffinose, although the galA gene has been found in some strains (Batt, 2014; Sedó Molina, Shetty, et al., 2024), although larger genome‐sized plant‐derived lactococci might be able to degrade plant complex sugars (Khayatt et al., 2020). Raffinose consumption was also detected in a hybrid soy drink (Madsen et al., 2021).
5.3. Saponins and glycoalkaloids
Saponins are a vast and diverse group of molecules found in plants chemically characterized by glycoside compounds attached to their aglycone structure (Oleszek & Oleszek, 2020). Different mono‐ and oligosaccharides forming diverse sugar chains can be found depending on the different saponins’ type, source, and chemical structure. Some glycosidic bonds between sugars are similar to the ones found in some phenolic compounds with sugar moieties attached (Bravo, 1998).
Various plant saponins have been found in PBFDAs, influencing the fermented products’ nutritional and non‐volatile flavor attributes. In soybeans, for example, three main types of saponins are found: soyasapogenol A (A1 and A2), B (I, II, III, IV, V, V1), and E. In soyasapogenol A, there are multiple variations of sugar chains in C‐3 (aglycone) containing specific chains of Glu, Gal, Rha, Ara, and glucuronic acid (GluA) and chains of Glu, Ara, and Xyl in C‐22 (Oleszek & Oleszek, 2020). In oats, the major saponins found are triterpene saponins, such as avenacins, and steroidal saponins, such as avenacosides. Avenacosides (A, B, C) and 26‐deglucoavenacosides (A, B) are characterized by containing sugar chains of Glu and Rha disposed of in different distributions in C‐3 and a Glu monomer in C‐26 (only in avenacosides) (Oleszek & Oleszek, 2020; Yang et al., 2016). In potatoes, glycoalkaloids, such as α‐solanine and α‐chaconine, are characterized by their toxicity and bitter taste. They contain mono‐, di‐ and tri‐saccharide molecules attached in their aglycone structure, which can vary from Rha, Glu, and Gal, forming different types of glycosidic bonds and structures (Baur et al., 2021; Hennessy et al., 2018; Oleszek & Oleszek, 2020).
In general, multiple structures of aglycones attached to numerous chains of sugars can be found depending on the plant‐based raw material source used. Nevertheless, specific glycosidic bonds are found throughout the different saponins. For instance, Glu–Rha, Gal–Glu, Rha–Gal, Ara–Glu, Gal–Ara, and GluA–Glu, among others, are recurrently found in the different plant saponins (Oleszek & Oleszek, 2020). In fact, hydrolyzing their glycosidic bonds reduces their toxicity (Hennessy et al., 2020). Here, the glycosidic hydrolases present in LAB might play a role in the degradation of saponins (Fu et al., 2020; Overney & Huang, 2020; Pham et al., 2023). Specific LAB strains that have been evolutionarily adapted to have high glycosidic activities toward specific glycosidic bonds might be useful as starter cultures to biotransform saponins into less toxic compounds.
Enzymes such as β‐d‐glucosidases (EC 3.2.1.21), β‐galactosidases (EC 3.2.1.23), α‐l‐rhamnosidases (EC 3.2.1.40), α‐l‐arabinosidases (EC 3.2.1.55), β‐d‐xylosidases (EC 3.2.1.37), and β‐glucuronidases (3.2.1.31) are present in specific LAB species but only phenotypically expressed in specific strains and conditions might be useful to investigate when selecting LAB strains for the generation of PBFDAs. Indeed, the substrate specificity of those enzymes toward saponin sugar moieties is relevant to consider due to the presence of multiple similar glycosidases in whole genome LAB sequences.
In the homology‐based analysis presented in Figure 2a, different families of glycoside hydrolases (GH) characterized to have β‐glucosidases (GH1, 2, 3, and 30), β‐galactosidases (GH1, 2, 3, 30, 35, and 43), and α‐glucosidases (31) activities as well as glucuronidases (Gus), l‐α‐arabinosidases (AraA), β‐xylosidases (XynB), xylose isomerases (XylA), and l‐α‐rhamnosidases (RamA/l‐ram) were considered to play an important role in degrading sugar moieties attached to aglycones. Those GH families for β‐glucosidases and β‐galactosidases were chosen based on the substrate specificity toward glucose or galactose‐linked moieties. Overall, L. diolivorans, L. pentosus, P. acidilactici, L. brevis, and L. rhamnosus encoded for more variety and multicopy GH. On the other hand, B. bifidum, B. breve, F. sanfranciscensis, L. amylolyticus, L. delbrueckii, L. gallinarum, L. gasseri, L. aviarius, L. salivarius, and S. thermophilus encoded for 0 to 1 copies of all those genes. GH1 and GH2 families were more conserved in most of the LAB species tested compared to the other families, such as Companilactobacillus, Lactiplantibacillus, Lacticaseibacillus, and Latilactobacillus genera for GH1 and Lactiplantibacillus, Lactobacillus, Lentilactobacillus, Leuconostoc, Limosilactobacillus, Loigolactobacillus, and Latilactobacillus genera for GH2. Most of the strains of L. pentosus also encoded for GH3 and GH43, whereas some Lacticaseibacillus species blasted positive against GH35. Gus showed higher positive hits with B. longum, L. rhamnosus, L. buchneri, L. diolivorans, and L. parafarraginis. AraA was found in B. breve and B. longum, L. buchneri, L. diolivorans, and one strain of L. hilgardii; xynB was found in all strains of B. adolescentis, B. animalis, L. lactis, L. diolivorans, L. pseudomesenteroides, L. brevis, and Pediococcus acidiilactici; xylA was found in B. adolescentis, B. animalis, L. lactis, L. diolivorans, L. pseudomesenteroides, L. brevis, L. mucosae, P. acidiilactici, and S. collinoides. Finally, ramA was only found in specific strains of C. alimentarius (1), L. pentosus (4), L. acidophilus (3), L. fermentum (1), and L. coryniformis (1).
β‐d‐glucosidases are glycosidic hydrolases that have been widely characterized in LAB species. Enzymes from GH1, GH2, GH3, and GH30 have shown β‐d‐glucosidase activity in LAB. They have been associated with hydrolyzing sugar moieties (mainly glucose‐based) present in different plant metabolites, including plant saponins (Michlmayr & Kneifel, 2014). Some glucosidases have shown additional activities toward pentose‐based glycosidic bonds, such as xylose and arabinose (Michlmayr & Kneifel, 2014). They have been characterized to be either extracellularly (L. plantarum USC1) or intracellularly (L. casei ATCC 393, O. oeni ATC BAA‐1163, L. brevis SK3, and L. mesenteroides) produced (Coulon et al., 1998; Gueguen et al., 1997; Michlmayr & Kneifel, 2014; Sestelo et al., 2004). β‐glucosidase activities in plant‐based fermented products, such as kimchi, were also studied in L. pentosus 6105 and Companilactobacillus paralimentarius LH4 about ginsenoside Rb1 saponin biotransformation (He et al., 2019).
Moreover, phosphor‐β‐glucosidases and phosphor‐β‐galactosidases, which are involved in the PTS‐based sugar transport, mainly in OHOM and FHET LAB species, might be involved in the degradation of such moieties as it has been demonstrated for phenolic compounds (Acin‐Albiac et al., 2021). Interestingly, L. brevis is one of the few OHET species that harbor those putative genes in their genome (Michlmayr & Kneifel, 2014) together with L. pseudomesenteroides DSM 20193 that encodes for one cellobiose phosphor‐β‐glucosidase PTS system (Acin‐Albiac et al., 2021). The transport of plant glucoside metabolites was studied in L. acidophilus NCFM, where different PTS transport systems were upregulated depending on the compounds in which they were grown. Nevertheless, some upregulated genes were shared within different compounds. For instance, the EIIABC component of the PTS system and a phospho‐β‐glucosidase that belonged to glycoside hydrolase Family 1 were upregulated when cells were grown in amylagdin, esculin, and salicin, in comparison to when they were grown in glucose (Theilmann et al., 2017). Moreover, β‐glucosidase (gluA), β‐galactosidase (bgl), and α‐rhamnosidase (rhaA) expressed in Arthrobacter sp. S41 has been demonstrated to play an essential role in the degradation of glucose, galactose, and rhamnose moieties in α‐chaconine and α‐solanine glycoalkaloids from potatoes (Hennessy et al., 2020).
Furthermore, few studies have been conducted on β‐galactosidase activities of LAB about saponin deglycosylation. β‐galactosidase in LAB has been mainly studied for its function on the degradation of lactose into free glucose and free galactose, cleaving the β‐1,4‐glycosidic bond (Vasudha et al., 2023). In some plant saponins, this bond type (demissidine in potatoes) and similar bonds (β‐1,2‐glycosidic bond) between glucose and galactose are found in different plant saponins such as soyasaponins (Oleszek & Oleszek, 2020). High β‐galactosidase activities have been characterized in L. acidophilus, L. helveticus, and L. reuteri based on the o‐nitrophenol‐β‐d‐galactopyranoside assay (Carević et al., 2016). Lactobacillus bulgaricus L3 β‐galactosidase has been concluded to be involved in the glycosylation of steroidal saponins, used as sugar substrate acceptors using lactose as a sugar donor (J. Zhang et al., 2016).
Rhamnosidase activities have also been studied in LAB due to their relevance in the hydrolysis of rhamnose moieties attached to flavonoids, thus increasing their bioavailability and consequently performing their favorable biological properties as anti‐inflammatory and antimutagenic compounds (Mueller et al., 2018). Although flavonoids are polyphenolic compounds, their sugar‐attached bonds are si the ones found in some plant saponins, therefore suggesting that similar enzymatic activities might be important to correlate among different plant metabolites (Bravo, 1998; Oleszek & Oleszek, 2020). Strains of L. plantarum, L. brevis, P. adidilactici, and L. fermentum have shown correlations between in vitro rhamnosidase activity and the ability to hydrolyze rhamnose from specific flavonoids (hesperidin) (Mueller et al., 2018). Nevertheless, the hydrolysis of specific flavonoids has been shown to be substrate and strain‐dependent. For instance, some strains showed high rhamnosidase activities but low flavonoid hydrolysis, whereas others showed different flavonoid hydrolysis toward different saponins within the same LAB strain (Mueller et al., 2018). Other rhamnosidases have been characterized in L. plantarum DSM 20205 and L. acidophilus DSM 9126, showing evolutionary differences among them, where the former seemed to be more closely related to Aspergillus genera, whereas the latter to ramA gene of Clostridium stercorarium, suggesting HGT events (Beekwilder et al., 2009). Although their presence and activity in those strains, their application in fruit juice was limited by their pH‐dependent activity, which was close to none in pH < 4.5 (Beekwilder et al., 2009). More studies on the performance of rhamnosidases in the hydrolysis of rhamnose‐containing saponins are needed to evaluate their promising activities.
Furthermore, it has been shown how some strains of L. brevis (L. brevis subsp. coagulans) and Lactobacillus crustorum encoded and expressed β‐glucuronidase (gus) that demonstrated to hydrolyze glucuronic acid moiety attached to herbal and tea saponins attached to the C‐3 of the aglycone (Qian et al., 2018; Sakurama et al., 2014). This enzyme belongs to the glycoside hydrolase Family 30 (GH30), and it has shown substrate specificity preferences toward some flavonoids (baicalin) in comparison to others (estrone) (Sakurama et al., 2014).
Future research on factors that affect glycoside hydrolase expression and their specificity toward sugar moieties attached to aglycone is needed. LAB strains capable of producing specific GH also need to be adapted to express them under specific fermentation conditions, considering pH, temperature, other C‐sources, and coculture conditions with other LAB strains. Saponins might not necessarily up–downregulate the transcription of glycoside‐related genes, but their expression has been demonstrated to influence saponin's degradation. Therefore, finding other ways to upregulate their expression might positively affect the degradation outcome.
5.4. Polyphenolic compounds
Phenolic and polyphenolic compounds are widely present as plant metabolites, constituting an important factor in the human diet (Bravo, 1998). There are more than 8000 types of polyphenols differing in structure, chemistry, biological action, origin, and composition. However, they are all considered secondary metabolites originating through the shikimate or the acetate pathways in plants. Polyphenols are generally found in nature in conjugated forms, frequently bound to sugars (such as glucose, galactose, rhamnose, xylose, and arabinose), carboxylic and organic acids, lipids, and other phenolic compounds (Bravo, 1998). They are classified into multiple groups, including phenolic acid derivatives (acetophenones, coumarins, chromones, and hydroxycinnamic acids), flavonoids (flavonols, flavanones, isoflavonoids, etc.), stilbenes, and lignans, among others (Bravo, 1998; Rodríguez et al., 2009). A more recent classification separates phenolic compounds based on their negative or positive effect on human health. First, review articles have focused on their adverse action due to their participation in precipitating proteins, inhibiting digestive enzymes, and reducing the absorption of plant vitamins and minerals (Bravo, 1998; Rodríguez et al., 2009). Nevertheless, some other studies have postulated some beneficial health effects in specific types of phenolic compounds (flavonoids) due to their antioxidant and free‐scavenging abilities (Bravo, 1998; Rodríguez et al., 2009). Apart from those attributes, phenolics contribute to the volatile and non‐volatile flavor profiles (bitterness and astringency) and color of specific food products (Gaur & Gänzle, 2023; Rodríguez et al., 2009).
Tannins, which are the polyphenolic compounds mainly studied due to their antinutritional effects, are high molecular weight multimeric phenolic compounds that are highly hydroxylated and form insoluble complexes with carbohydrates and proteins and are responsible for the astringency of tannin‐rich food products (Bravo, 1998). Hydrolysable tannins are constituted by gallic acid monomers (gallotannins) and/or hexahydroxydiphenic acids bound to glucose (usually) (ellagitannins). In contrast, insoluble condensed tannins are multimeric flavan‐3‐ol (catechin, etc.) polymers. In cereals, hydroxycinnamic acids are the most abundant phenolic compounds, such as ferulic acid esterified to arabinoxylans (barley and rye) or ferulic, p‐coumaric, and caffeic acid esterified to glycerol (oat and sorghum).
The metabolic effect of specific LAB species and strains on the biotransformation of different phenolic compounds has been extensively studied through the last decades toward specific phenolic compounds such as hydroxycinnamic, hydroxybenzoic, and tannic acids found in specific plant‐based matrices such as fruits (mangoes, grapes, cherries, and pomegranate), vegetables (broccoli), cereals (oats, wheat), and legumes (soybeans). Specific esterases, reductases, and decarboxylases are important in degrading some phenolic compounds (Gaur & Gänzle, 2023). These authors have explained that hydroxycinnamic esters might be released by esterases such as LP_0796, Lj0536, Lj1228, HceP, and Est_1092, forming hydroxycinnamic acids, which can subsequently be decarboxylated by phenolic acid decarboxylases (Pad), producing vinyl phenol derivatives or reduced by HcrB. HcrF and Par1/Par2 reductase coding genes. Vinyl phenol derivatives can also be reduced through other reductase coding genes such as vrpA. In contrast, hydroxybenzoic esters can be transformed into hydroxybenzoic acids by other esterases, such as tannases (TanA or TanB), and subsequently decarboxylated to simple phenols throughout other decarboxylases such as LpdC. Furthermore, as mentioned in the saponin section, some LAB glycosidases, such as β‐glucosidases, phosphor‐β‐glucosidases, and rhamnosidases, have been demonstrated to be involved in the deglycosylation of flavonoids. For instance, L. plantarum, L. pentosus, L. brevis, and P. pentosaceus have been characterized to hydrolyze glucose moiety of oleuropein through β‐glucosidase activities (Michlmayr & Kneifel, 2014).
The activities of the enzymes mentioned above have been characterized in specific strains of L. plantarum (TanA, TanB, HceP, Lp_2945, Lp_366, est_1092, Lp_0796, Est_1092, HcrB, Pad, VrpA, and Ram1), L. fermentum (HcrF), L. johnsonii (Lj0536), Furfurilactobacillus millii (EstR, Par1, Par2), L. brevis (LcGUS30), and L. acidophilus (RamAL A ) (Gaur et al., 2020, 2023; Jiménez et al., 2013, 2014; Sakurama et al., 2014; Santamaría et al., 2018). For instance, in some LAB strains such as F. millii FUA 83583, the reductase activity of Par1/Par2 was out‐competing the decarboxylase activity of Pad, producing more dihydro‐derivatives than vinyl derivatives out of ferulic, caffeic, and p‐coumaric acids. However, more vinyl derivatives were produced when phenotypically testing Δpar1/par2 mutants (Gaur et al., 2023) throughout decarboxylation pathways, suggesting overexpression of pad in reductive growth conditions. All the genes mentioned above have been documented to be produced intracellularly, thus requiring an intracellular transport system to be transformed, except TanA, which is extracellularly produced (Jiménez et al., 2014).
L. plantarum has been the LAB species primarily studied in the degradation of phenolic compounds (Barthelmebs et al., 2001; Jiménez et al., 2013, 2014; Landete et al., 2021; Pulido‐Mateos et al., 2022; Santamaría et al., 2018). Some of their genes have been cloned to other LAB species to characterize their activity when heterologously expressed. Tannase genes were more expressed in L. lactis MG1363 than in L. casei BL23, L. reuteri CECT, or S. thermophilus INIA 468 when cloning the same gene from L. plantarum (Landete et al., 2021). Esterase activities toward feruroyl compounds (est_1092) have also been reported in L. fermentum, L. crispatus, L. acidophilus, and L. helveticus. Moreover, pad activity has been found in L. lactis, P. pentosaceus, and L. brevis, apart from L. plantarum species (Landete et al., 2021).
Niche adaptation of LAB into specific environments might explain the evolutionary events concerning phenolic acid degradation and HTS. For instance, fructophilic LAB (FLAB) that live in fructose‐rich niches such as pollen and insect gastrointestinal tracts have demonstrated the advantage of using their genotypical characteristics to acquire metabolic benefits from phenolic compounds present in those environments (Filannino et al., 2016). Specifically, Filannino et al. (2016) suggested that heterofermentative Lactobacillaceae such as FLAB may use p‐coumaric acid as a final external acceptor when others such as fructose, oxygen, or pyruvate are low or not present. Thus, the regeneration of NAD+ from the reduction of p‐coumaric to phloretic acid might enhance glucose intake by FLAB throughout glycolysis, increasing ATP yield in the phosphoketolase pathway (Filannino et al., 2016; Gaur & Gänzle, 2023). In contrast, decarboxylation pathways of phenolic compounds throughout the pad may be used by LAB as detoxification reactions of those compounds into less toxic derivatives. Moreover, decarboxylation reactions involve the consumption of H+, which might contribute to the internal and external pH balance and the production of CO2 that increases buffer capacity (Barthelmebs et al., 2000; Zuljan et al., 2016).
All genes and enzymes mentioned above were considered for generating the amino acid‐based blast analysis (Figure 2). Strains from C. alimentarius, L. argentoratensis, L. paraplantarum, L. pentosus, S. collinoides (one strain), and L. plantarum encoded for more than 10 genes were analyzed. Most strains of L. argentoratensis, L. paraplantarum, and L. plantarum encoded for 3–4 copies of HcrB, finding up to 7 copies in one L. plantarum strain. LpdC was found in S. collinoides to have 3–4 copies. TanA coding genes were found not only in L. plantarum but also in three strains of C. alimentarius, two strains of L. argentoratensis, four strains of L. paraplantarum and L. pentosus, and four strains of O. oeni. PadC was found in most of the strains belonging to L. lactis, Pediococcus spp., C. farciminis, L. helveticus, Latilactobacillus spp., L. brevis, L. aviarius, L. mucosae, and L. fermentum, apart from the ones mentioned previously (not O. oeni). Par1 was only found in Lactiplantibacillus spp., and only specific strains of C. farciminis (1), L. hilgardii (3), S. collinoides (1), L. coryniformis (2), and L. parafarraginis (1). Furthermore, homologs of estA gene were found in more than 60% of strains belonging to L. fermentum, P. parvulus, L. reuteri, L. aviarius, L. johnsonii, L. gallinarum, L. crispatus, L. plantarum, L. pentosus, C. farciminis, and C. alimentarius. EstA was not found in any of L. paraplantarum strains.
5.5. Protease inhibitors
Plant protease inhibitors are small proteins (5–70 kDa) found in storage and aerial plant tissues and are essential to them due to their function against insects and pathogens, interfering in their enzymatic digestive system. They are frequently classified into 12 families, including the 2 most studied families, which are Bowman–Birk inhibitors (BBI), which generally inhibit trypsin, chymotrypsin, and/or elastase, and Kunitz‐type inhibitors (KTI), which act on serine, cysteine, and aspartic proteases (Cid‐Gallegos et al., 2022). The former is typically found in plant families such as Fabaceae (soy, pea, chickpea, lentils) and Poaceae (maize, wheat, rice, barley, millet, and oat), whereas the second also includes Solanaceae (potato), Cruciferae, among others, apart from the previously mentioned. Moreover, specific α‐amylase and trypsin inhibitors from Poaceae belong to Family VI of protease inhibitors, as well as specific potato inhibitors (I and II) (Families III and IV) and carboxypeptidase inhibitors (Family X), which are mainly present in Solanaceae plant family (Cid‐Gallegos et al., 2022). They are considered antinutrients due to their ability to bind digestive proteases, inhibiting their capability to break down proteins and limiting the absorption of peptides and amino acids during digestion (Cid‐Gallegos et al., 2022). Structurally, they can form mono, di, and tetrameric protein structures, depending on their source (Boakye et al., 2022; Laatikainen et al., 2017).
Research studies on the degradation of protease inhibitors by LAB have focused on α‐amylase and trypsin inhibitors (ATIs) degradation, mainly during legume and cereal‐based fermentations (Caminero et al., 2019; Fraberger et al., 2020; Huang et al., 2020). It has generally been associated with pH‐based ATI activity inhibition or proteolytic activities of LAB toward ATI hydrolysis. However, limited research has been conducted on the latter. LAB strains belonging to L. paracasei (Lpa4), L. pseudomesenteroides (Leps1), P. pentosaceus (Pp3), L. corynniformis (Lco4), and L. plantarum (Lpl5) have shown high abilities (40%–85%) of degrading wheat‐extracted ATIs among 87 sourdough LAB isolates (Fraberger et al., 2020). Caminero et al. (2019) compared the high ATI‐degrading ability of L. salivarius H32.1, L. mucosae D5a1, and L. rhamnosus LE3 with the low ATI‐degrading ability of L. fermentum R39.3 and L. reuteri R12.22 in vivo, all isolated from human. ATI‐degrading ability was demonstrated to be strain‐specific. ATI‐degrading ability has been studied phenotypically, but transcriptomic and genetic studies of LAB proteolytic enzymes involved in their degradation need further exploration. Nevertheless, interactions between serine‐type endopeptidases (prt family) and KTI have been modeled, showing binding energy changes and high relative affinities in proteases encoded by L. plantarum and L. lactis (B7VFD1 and Q49SH0 UniProt accession numbers) toward three different KTI and BBI chickpea protease inhibitors. Peptidase from L. plantarum showed higher relative affinity to all three inhibitors in the model (Sáez et al., 2022). Cell envelope proteinases (CEPs) present in LAB are proteases that belong to the prt family and play an essential role in the primary degradation of extracellular proteins, such as milk or plant proteins, into oligopeptides that can be transported intracellularly by specific peptide transporters and serve as N‐source for growth and metabolism. Six main types of CEPs have been identified based on protein sequences, including PrtP (L. paracasei subsp. paracasei, L. lactis subsp. cremoris, and subsp. lactis, L. rhamnosus), PrtB (L. delbrueckii subsp. bulgaricus), PrtH (L. helveticus), PrtS (S. thermophilus), PrtR (L. rhamnosus), and PrtL (L. delbrueckii subsp. lactis) (Ji et al., 2021). Other Leuconostoc spp., such as L. pseudomesenteroides and L. mesenteroides, have also been characterized to encode for prtP genes (Christensen et al., 2022; Sedó Molina et al., 2024). Other species, such as L. acidophilus (BGRAA3, CH2, V74), L. casei (HN14, IFPL731), L. plantarum (LP69), and F. sanfranciscensis (CB1), have also been characterized to encode for CEPs. The specificity of CEPs has been mainly studied toward milk proteins, and different cleavage sites have been found depending on the type of milk protein targeted, such as αs1, αs2, β, and κ‐caseins (Ji et al., 2021). Moreover, different prt genes from LAB strains have been shown to cleave in specific sites in milk protein sequences (Ji et al., 2021). Although specificity, structure, and activity of CEP studies have mainly been conducted on milk proteins, more phenotypical, transcriptomic, and genetic studies might be relevant to assess their activity toward ATIs, as Sáez et al., 2022 modeled CEPs‐ATI interactions.
Furthermore, studies investigating the decrease of trypsin inhibitory activities during and after fermentation have also been reported. Trypsin and chymotrypsin inhibitory activities in pea protein concentrate were reduced by 18% and 70%, respectively, after 11 h of fermentation with L. plantarum NRRL B‐4496 (Çabuk et al., 2018). Moreover, cocultured solid‐state fermentation (SSF) using L. brevis with Aspergillus oryzae in soybean meal reduced trypsin inhibitor content by 89.2% after 72 h, majorly contributed by the action of the filamentous fungi (Gao et al., 2013). Hoffmann et al. (2003) showed how soybean trypsin inhibitor was inactivated and degraded in an in vitro rumen incubation. Three different concentrations of trypsin inhibitor (2, 4, 10 mg/mL) were subjected to three different rumen incubations, being completely inactivated after 6, 9, and 12 h, respectively. In another context, the ATI content was significantly reduced (by 41%) after 12 h of sourdough fermentation using a Type 1 sourdough starter (Boakye et al., 2022). Interestingly, Laatikainen et al. (2017) compared ATI concentration in wheat after yeast‐based and sourdough fermentation. Sourdough fermentation decreased the ATI polymeric (60 kDa) content while increasing the ATI monomeric forms (<14 kDa), significantly higher than upon yeast‐based fermentation. This could mainly be due to the reduction of disulfide bonds between ATI monomers under the reductive redox conditions of sourdough.
Prt‐based homologous genes were considered in the blast. The high presence of prt genes was found to be associated with L. rhamnosus, L. delbrueckii, L. gallinarum, L. kefiranofaciens, L. panis, and S. collinoides. prtP and prtR were predominantly found in Lacticaseibacillus spp., whereas prtP and prtB were in L. delbrueckii and L. gallinarum. PrtM was found in two strains of B. animalis and one strain of B. bifidum, two strains of C. alimentarius, L. reuteri, O. oeni, P. pentosaceus, L. mesenteroides, and L. panis, and three of L. citreum, and one strain of Companilactobacillus farciminis, F. sanfranciscensis, L. crispatus, L. sakei, L. diolivorans, L. coryniformis, and L. pentosus. Two strains of L. kefiranofaciens were found to have one copy of prtM, prtH, and prtB each, whereas only prtB was found in the other three strains. PrtH was found in the same two strains of C. farciminis, L. kefiranofaciens, L. panis, O. oeni, and P. pentosaceus, and in one strain of B. bifidum, L. pentosus, L. crispatus, L. citreum, L. diolivorans, L. sakei, L. mesenteroides, and L. reuteri.
Further research is needed to understand the transcriptomic and protease substrate specificity factors behind the ability to degrade different protease inhibitors in multiple crops. Since proteinases target specific peptide bonds, proteomic studies could elucidate the protease expression patterns during fermentation under different pH levels and using different protease inhibitors as substrates. However, highly proteolytic LAB strains isolated from crops with high inhibitor content might be a promising LAB selection target for PBDA fermentations. The presence of other plant proteins that are more accessible and found at higher concentrations in plant‐based matrices might influence the inhibitor degradation outcome.
5.6. Oxalates
Oxalic acid is a dicarboxylic acid widely present in diverse families of plants, fungi, lichens, and algae involved as a secondary metabolite in the glycolate, oxaloacetate, and ascorbic acid pathways (Li et al., 2022). Oxalic acid is relevant for its participation in calcium regulation, photosynthesis, metal detoxification, and pH homeostasis in plants. Nevertheless, its increased intake and absorption might cause health diseases such as hyperoxaluria, which involves the formation of calcium oxalate stones in humans (Li et al., 2022; Mogna et al., 2014; Tavasoli et al., 2022). There are two main types of oxalic acid in plants: soluble and insoluble oxalic acid. Soluble oxalic acid is usually bound to sodium (Na+), ammonium (NH4 +), or potassium (K+), whereas the insoluble forms complexes with calcium (Ca2+), magnesium (Mg2+), and iron (Fe2+) (Li et al., 2022). In plants, oxalic acid degradation mainly involves three enzymes: oxalyl‐CoA synthase (EC 6.2.1.8) (Aae3), oxalate oxidase (EC 1.2.3.4) (Oxo), and oxalate decarboxylase (EC 4.1.1.2) (Oxdc). Aae3 is essential in the oxalate acetylation pathway, which is ATP‐dependent. Oxalate degradation through this pathway involves four different enzymes that catabolize consecutive reactions until producing CO2 after decarboxylase, transferase, and dehydrogenase transformations. Both Oxo and Oxdc are Mn‐dependent enzymes that catalyze the oxidation of oxalate into CO2 and H2O2 and into CO2 and formate, respectively. Oxdc is also pH‐dependent (Li et al., 2022).
Some species have been genetically and phenotypically evaluated in bacteria to encode for oxalate degradation enzymatic machinery. For instance, Oxalobacter formigenes has been studied for its metabolic mechanism of using oxalate as the main C‐source in the GIT (Kullin et al., 2014; M. Liu et al., 2021; Mogna et al., 2014). Because oxalate is found in the GIT due to its uptake from oxalate‐rich foods, specific strains of L. acidophilus (NCFM), L. gasseri (ATCC 33323), L. plantarum, L. fermentum, L. murinus, L. animalis, L. reuteri, L. paracasei, Weissella confusa, Weissella cibaria, L. rhamnosus, Loigolactobacillus bifermentans, L. salivarius, and Bifidobacterium spp. have been tested for the presence of oxalate‐degrading phenotypes as well as their potential to be used as probiotic strains to avoid oxalate derived adverse health effects (Azcarate‐Peril et al., 2006; Cho et al., 2015; Gomathi et al., 2014; Kullin et al., 2014; Lewanika et al., 2007; Murphy et al., 2009; Tavasoli et al., 2022; Turroni et al., 2010; Youssef, 2024). Based on genetic and transcriptomic studies, two critical genes in LAB are involved in positive oxalate‐degrading LAB strains: formyl coenzyme A transferase (EC 2.8.3.16) (frc) and oxalyl coenzyme A decarboxylase (EC 4.1.1.8) (oxc). The former is involved in transferring CoA moiety from formyl‐CoA to oxalate, producing oxalyl‐CoA and formate. The latter decarboxylates oxalyl‐CoA into formyl‐CoA and CO2, resulting in one molecule of formate and CO2 per molecule of oxalate (Figure 1) (Azcarate‐Peril et al., 2006; Kullin et al., 2014). In some LAB strains, such as L. gasseri ATCC 33323 and L. acidophilus NCFM, frc and oxc appear to form an operon because both genes have been analyzed to have terminators (Azcarate‐Peril et al., 2006; Lewanika et al., 2007).
Transport systems encoding genes are in the same cluster as those that encode oxalate degradation in some LAB strains. L. gasseri ATCC 33323 showed a putative predicted permease, L. acidophilus NCFM had a putative predicted ABC transporter, and L. reuteri encoded for a xanthine, uracil, and vitamin C putative permease next to frc locus (Azcarate‐Peril et al., 2006; Lewanika et al., 2007). Differently, O. formigenes encodes for an antiport transport system of oxalate and formate (oxIT), which belong to the major facilitator superfamily of transporters (Iyalomhe et al., 2015).
Interestingly, Youssef (2024) stated that among five positive oxalate‐degrading LAB strains screened, all encoded for the frc gene, while only three encoded for both frc and oxc genes, which were L. bifermentans SA5, L. paracasei SA9, and L. plantarum SA37. In L. reuteri 100‐23C, both genes were present, but no phenotype was observed. This fact might be explained by the lack of an oxalate transport mechanism in this strain (Kullin et al., 2014). Bifidobacterium animalis subsp. lactis BI07, which encoded for oxc and frc genes, showed a pH dependency of the gene expression based on transcriptomics. The oxalate degradation rate was increased when this strain was exposed to pH 4.5, supporting that frc and oxc genes require mildly acidic conditions to be expressed (Turroni et al., 2010). Moreover, an ORF encoding a putative permease (ORF‐1) flanking oxc gene was found and might be involved in intracellular oxalate transport and counteracting pH changes (Turroni et al., 2010). Similarly, both strains L. gasseri ATCC 33323 and L. acidophilus NCFM showed higher oxalate‐degrading gene expression when cells were adapted to sodium oxalate and transferred to pH = 5.5 growth media (Azcarate‐Peril et al., 2006; Lewanika et al., 2007). Furthermore, Bifidobacterium dentium Bd1, a LAB strain isolated from dental plaque, upregulated both frc and oxc when the strain was exposed to acidic conditions (pH = 4), suggesting their influence under acidic‐based stress conditions (Ventura et al., 2009). Phenotypical‐based studies performed in L. acidophilus ATCC5344 and L. acidophilus (probiotic 1) have shown a decrease in oxalate concentrations by 21.7% and 41.2%, respectively. In contrast, L. plantarum (probiotic 2) increased its concentration significantly (Cho et al., 2015). Some L. animalis (5323, 5241, 5342, 5121, 1221, 6331, 223C) and L. murinus (1222, 3133, 5333) showed oxalate‐degrading phenotype in vitro, but only 223C and 5323 were active in vivo (Murphy et al., 2009).
Only the presence of homologs of frc and oxc was tested throughout the blast analysis. B. animalis was the only Bifidobacterium spp. that encoded for at least one of both genes (frc), although there was only one strain that encoded for both. L. acidophilus, L. amylovorus, L. gasseri, L. helveticus, L. johnsonii, L. kefiranofaciens, L. aviarius, L. mucosae, and L. reuteri strains were 100% positive in at least one of both genes. Only all the strains belonging to L. acidophilus and L. mucosae tested positive for both genes. However, one strain of L. acidophilus and L. mucosae had two copies each, and one strain of L. helveticus had three copies of each gene. All L. reuteri strains had both genes, but one strain, although there was another strain with two copies of frc and one copy of oxc.
However, there are still some challenges to overcome. The regulation of frc and oxc, particularly dependent on pH, could change over fermentation time because pH is decreased. Moreover, most oxalate‐degrading strains are particularly associated with being isolated from the gastrointestinal tract, perhaps limiting their application to plant‐based fermentations because their frc and oxc expression might be adapted to the gut conditions.
6. METABOLIC INSIGHTS OF LAB: BIOTRANSFORMATION OF OFF‐FLAVOR VOCS
VOCs considered off‐flavors for the generation of PBFDAs display undesired flavor attributes such as green, beany, and earthy notes that negatively influence the consumer acceptance of plant‐based fermented products that mimic their milk‐based counter products (Table 2). As discussed previously, they usually have low odor thresholds that make them perceivable at low concentrations.
However, associating a VOC with a particular note is often challenging because a high or a low concentration can affect the perceived notes. This also applies when VOCs are found in combination with other odorant and non‐odorant volatiles (Fischer et al., 2022). For instance, hexanal displays “beany” notes when it is present at low concentrations (1–10 ppm), although it does not have this attribute by itself. Similarly, 1‐octen‐3‐one displays sweet‐green notes at 100 ppm, whereas it is beany at lower concentrations (1–10 ppm) and 1‐octen‐3‐ol is perceived as earthy at low (1–10 ppm) and high (>1000 ppm), but “beany” in between those concentrations. The binary combinations of 1‐octen‐3‐one (10 ppm) and hexanal or (E, E)‐2,4‐decadienal (2 ppm) display a higher “beany” concentration range (from 1 to 1000 ppm) than others such as hexanal (10 ppm) with (E)‐2‐nonenal (2 ppm) that are perceived as “beany” at <100 ppm (Trindler et al., 2022).
Multiple studies have investigated the potential of LAB strains to remove aldehydes, ketones, pyrazines, furans, and sulfur‐based VOCs, some of those off‐flavors via fermentation. Still, the underlying mechanisms of specific enzymes involved remain to be understood. Tangyu et al. (2023) analyzed the transformation of VOCs in four different plant‐based milks (oat, sunflower seed, faba, and pea milk) using 14 single‐strain LAB different fermentations. Although other types of milk contained various VOCs, “green” and “earthy” off‐flavor compounds were consistently present across all types, including hexanal, benzaldehyde, nonanal, heptanal, 2‐pentylfuran, and 1‐octen‐3‐ol, at varying concentrations. Interestingly, certain LAB strains could transform these compounds into their respective reduced or oxidized forms. For instance, hexanal, nonanal, and heptanal may be reduced to n‐hexanol, n‐nonanol, and n‐heptanol or oxidized to hexanoic, nonanoic, and heptanoic acids (Figure 3). Less is known about the biological transformation products of furans and pyrazines in LAB. However, it has been studied in other bacteria, such as Stenotrophomonas sp. and Mycobacterium sp. (Rajini et al., 2011). Moreover, the biotransformation of sulfur compounds in LAB has been more focused on the production of dimethyl sulfide, dimethyl disulfide, and dimethyl trisulfide throughout l‐methionine and l‐cysteine metabolisms than to its degradation (Landaud et al., 2008). Furthermore, OHET LAB species, such as L. mesenteroides, showed a higher reduction of aldehyde‐based off‐flavor rates than other OHOM LAB species, such as S. thermophilus strains, in most of the plant‐based milk (sunflower, faba, and pea). Aldehydes were mainly transformed into their alcohol forms, although carboxylic acid forms were also present in pea and oat milk. Similarly, Sedó Molina et al. (2024) demonstrated how the removal of aldehyde‐based off‐flavors was higher when fermenting with L. pseudomesenteroides NFICC 2004 than with L. lactis NFICC 2005 in soy, oat, and pea milks. That phenotype was conserved when both were cocultured.
FIGURE 3.
Reductive–oxidative reaction of aldehyde volatile organic compound (VOC) off‐flavors to their carboxylic acids and primary alcohol forms through lactic acid bacteria (LAB) aldehyde (aldh) and alcohol (adh) dehydrogenases, respectively. NAD(P)H might mediate those reactions by acting as an electron donor or acceptor molecule. Source: Created with Biorender.
Fischer et al. (2022) stated that aldehyde (Aldh) and alcohol (Adh) dehydrogenases are the most important enzymes involved in the transformation of aldehyde and ketone‐based VOCs into their alcohol and carboxylic acid respective forms through redox reactions; thus, decreasing off‐flavor note's perception (Figure 3). In the central LAB carbon metabolism, the adh and aldh genes play an essential role in establishing redox balance through NAD(P)+ production, depending on the type of metabolism a LAB species has. There have been classified three types of NAD(P)+ dependent ADH in LAB: long‐chain zinc‐type ADH (Group I—around 350 amino acids), short‐chain metal‐free ADH (Group II—250 amino acids), and iron‐containing ADH (Group III—around 385 amino acids per subunit) (Chen et al., 2016). It has been suggested that Group III ADHs could have been primarily involved in reducing aldehydes (Elleuche & Antranikian, 2013). Some Group III ADHs homologues of YqhD (NADPH‐dependent ADH) in E. coli and DhaT (1,3‐propanediol dehydrogenase) in Klebsiella pneumoniae, have been phenotypically involved in reduction of aldehydes with broad substrate specificity, and have been found in genomes of specific LAB species such as O. oeni (Elleuche & Antranikian, 2013). Conversely, the bifunctional aldehyde/alcohol dehydrogenase gene (adhE), present in the genomes of most LAB species (except specific fructophilic LABs and some OHOM strains), encodes an iron‐containing ADH that is constitutively expressed in OHET species during mixed‐acid fermentation of carbohydrates (phosphoketolase pathway) (Fischer et al., 2022; Verce et al., 2020). It catabolizes the reduction of acetyl‐CoA (ALDH domain) into acetaldehyde and then to ethanol (ADH domain), present in the phosphoketolase pathway (Pony et al., 2020). Microbial strains that constitutively produce ALDH and/or ADH are suggested to display better capabilities in removing aldehyde‐based off‐flavor molecules. Interestingly, LAB species, such as Fructobacillus fructosus, Apilactobacillus kunkeei, and Apilactobacillus apinorum, which are considered fructophilic LAB, have naturally lost either the ADH or both domains from the adhE through adaptation to fructose‐rich niches (Endo et al., 2014; Maeno et al., 2019; Sedó Molina et al., 2022). Vermeulen et al. (2007) studied how F. sanfranciscensis used hexanal as a hydrogen acceptor to regenerate NADH, producing hexanol and higher levels of acetic acid when hexanal was present. Thus, OHET LAB strains might use the reduction of aldehydes such as hexanal, pentanal, and nonanal, mediated by the oxidation of NAD(P)H to NAD(P)+, to produce additional ATP through acetyl‐P dephosphorylation reaction to acetate. This fact suggests that OHET species have developed their competitiveness in nature by utilizing alternative electron acceptors to increase growth and outcompete other microbial strains in specific environments (Gänzle, 2015); therefore, they are more metabolically prone to utilize aldehydes as electron acceptors than OHOM LAB species.
It has been observed that other enzymes, such as glutamate aldehyde dehydrogenase and aspartate aldehyde dehydrogenase, can present alcohol and/or aldehyde dehydrogenase activities in LAB (Chen et al., 2016; Fischer et al., 2022; Hu et al., 2019). In L. reuteri DSM 20016, five different ADHs were identified, cloned, and characterized (Hu et al., 2019). They showed substrate heterospecifity on the different adh where the highest activities of ADH‐1, ‐2, ‐3, ‐4, and ‐5 were demonstrated on heptanal (heptaldehyde), butyraldehyde, propionaldehyde, heptanal, and 3‐methyl butanal, respectively. On the one hand, they showed that ADH‐2 and ADH‐3 could not reduce benzaldehyde, a key aldehyde‐based VOC. On the other hand, ADH‐2 and 4 had the highest relative activity toward pentanal (valeraldehyde) (82.6% and 81.1%) and hexanal (76.2% and 85.1%), whereas ADH‐1 showed the highest with benzaldehyde (49.8%) (Hu et al., 2019). Finally, all ADH activities were stimulated with the presence of Zn2+, suggesting them as zinc‐dependent dehydrogenases. A benzyl ADH was characterized in L. plantarum WCFS1 to have substrate specificity to oxidize aliphatic and aromatic alcohols such as benzyl alcohol, which could be involved in benzaldehyde metabolism (Landete et al., 2008). Benzaldehyde can also be produced from phenylalanine through non‐enzymatic reactions (Russo et al., 2021). Moreover, benzoic acid produced by L. casei has been found to have an inhibitory effect on soybean LOX, thereby reducing the oxidative action toward plant fatty acids (Sekhar Rao et al., 2002).
Moreover, higher concentrations of aldehydes such as (E, E)‐2,4‐decadienal have been reported after fermentation due to the LAB strains and conditions that could lead to creating an oxidizing environment (ex: H2O2), hence promoting lipid oxidation in the matrix toward aldehyde production (Vermeulen et al., 2007). Little is known about the biotransformation of 2‐pentylfuran by LAB, although it has been characterized in Saccharomyces cerevisiae (Xu et al., 2022). In this, 2‐pentylfuran is transformed into acetic acid hexyl ester and o‐cymene through consecutive enzymatic reactions that involve epoxide hydrolase, adh, aldo‐keto reductases, dehydrogenases, oxidases, acyltransferases, and ester synthases (Xu et al., 2022). Interestingly, o‐cymene was increased in sunflower seed fermentation by L. johnsonii and L. plantarum, although 2‐pentylfuran concentration was kept equal after fermentation (Tangyu et al., 2023). Most yeast species used in food fermentations have higher esterase activities than LAB species, which are usually characterized by low esterase expression. Therefore, alcohol and carboxylic forms of VOCs might be frequently transformed by yeast to ester derivatives by ester synthases and acyltransferases to produce ethyl ester derivatives (Xu et al., 2022). In LAB, some strains have one or more esterase‐coding genes, typically having preferences with substrates such as acetate (Katz et al., 2002).
7. DISCUSSION
The high demand for adequate nutritional, sensorial, and technological PBFDAs, such as plant‐based cheeses and yogurts, leads to investigating the microbial‐based transformation of plant compounds that might affect quality attributes in the final fermented products. Here, we described the state‐of‐the‐art genes, enzymes, and metabolic pathways involved in the biotransformation of antinutrients (phytic acid, RFOs, polyphenols, saponins, enzyme inhibitors, and oxalic acid) (Table 3) and off‐flavor VOCs (Table 2) by LAB fermentation. LAB is the most important group of bacteria that has been used for centuries for food fermentation, especially for the generation of milk‐based fermented products (Harper et al., 2022; Smit et al., 2005). Most LAB species are ubiquitous in nature and have genetically evolved differently depending on the environment, abiotic and biotic conditions, and microbial diversity, where they have been inhabiting and adapting, among others (Makarova & Koonin, 2007). Some LAB species have lost or gained relevant competence genes through genetic transfer events during their environmental adaptation, thereby increasing or reducing their microbial fitness (Bačun‐Družina et al., 2009). Most LAB strains used in dairy‐fermented products have been adapted to grow in milk, allowing them to utilize lactose, milk proteins, and other nutrients for their growth and metabolism (de Vos & Vaughan, 1994; Fontana et al., 2018). Therefore, the adapted dairy starter cultures may not be optimal in producing fermented plant‐based alternatives where different nutritional compositions and physicochemical properties have been observed.
The development of PBFDAs, derived from legumes, cereals, and nuts, may require the use of other LAB species and/or strains that will be better equipped (both genetically and enzymatically) to thrive in those new environments and cope with the undesired plant metabolites present (Gaur et al., 2020; Zartl et al., 2018). This review suggests the importance of selecting specific LAB species for detailed research when screening and characterizing strains for targeted plant‐based applications involving LAB fermentation. It emphasizes the use of particular species, such as L. plantarum, L. fermentum, L. acidophilus, L. paracasei, and L. rhamnosus, which have been characterized to encode relevant genes in multiple antinutrient categories based on research studies. On the other hand, the blast analysis genotypically validates the potential of employing those species while showing the promising capabilities of other unconventional QPS‐listed LAB species such as L. pentosus, L. diolivorans, S. collinoides, and L. paraplantarum as suitable candidate species to be used as starter cultures for plant‐based fermentations.
Genetic annotation, gene and enzyme expression, and enzymatic substrate specificity have been the most critical factors when evaluating promising LAB strains to be used as starter cultures for plant‐based fermentations. The evaluation of the presence of relevant genes in a pool of LAB strain genomes might be helpful as a primary genotype screening tool when the genes have been functionally characterized and are not highly conserved genes among LAB species. For example, tanA has been mainly described in specific strains of Lactiplantibacillus spp. (Jiménez et al., 2014), although found in other species such as O. oeni (this study), and involved in the degradation of tannic acid (Figure 1). In addition, genetic annotation and analysis of upstream and downstream regions of relevant genes might provide information on the presence of other non‐homologous genes involved in the transport of target molecules, such as an ABC transporter found in L. acidophilus NCFM, a putative permease found in L. gasseri ATCC 33323, and a putative permease found in L. reuteri 100‐23C participating as analogous oxalate LAB transport systems, in the vicinity of the frc locus (Azcarate‐Peril et al., 2006; Kullin et al., 2014; Lewanika et al., 2007). Nevertheless, genetic annotation might not be useful in highly conserved genes such as adhE, which is suggested to be involved in reducing aldehyde‐based off‐flavor compounds in OHET LAB species (Fischer et al., 2022; Sedó Molina, Shetty, et al., 2024).
The gene expression analysis under certain environmental conditions could be highly interesting when selecting LAB strains as starter cultures. However, genes might be present but not necessarily expressed under specific growth and environmental conditions, or expressed but having low activities. In fact, this might be a problem when phenotyping LAB strains using minimal media, where switching to complex media such as plant‐based drinks might influence the output expression of targeted phenotypes, resulting in an undesired product quality. Variable growth and physicochemical conditions such as pH, temperature, carbohydrate and mineral composition, growth phase, and redox balance are some of the most relevant. Experimental and fermentation variables such as temperature and salt addition also might affect targeted LAB gene expression when fermenting plant bases. Moreover, some antinutrients, such as polyphenolic compounds, can be metabolized by multiple parallel enzymatic pathways, where gene expression plays an important role in which metabolic route outcompetes the others as occurred with F. millii FUA 83583 (Gaur et al., 2023).
Third, substrate enzymatic specificity is key when characterizing LAB strains relevant to biotransform specific antinutrients and volatile VOCs. Some key enzymes have been demonstrated to have substrate multispecificity towards different compounds belonging to the same category. Alcohol dehydrogenases, phytases, RFO transporters, galactosidases, and saponin glycosidases are some examples. The combination of substrate specificity studies linked with protein structure modeling might be relevant to predict multispecificity characteristics of target enzymes toward common plant metabolites present in different plant kingdoms.
Moreover, the degradation rate of antinutrients and off‐flavors by LAB fermentation is influenced by many variables such as the plant‐based matrix, LAB stains fitness and metabolism, bacterial physiological state, fermentation conditions, and synergistic interactions within strains (Sedó Molina et al., 2024). Nevertheless, Nugroho et al. (2024) demonstrated that the removal of off‐flavor aldehydes could be performed under 1 h of biological activity by limiting the amounts of fermentable sugar conditions.
There is still a lack of knowledge of the transport systems involved in the transfer of the discussed compounds intracellularly, such as phytic acid, polyphenols, and oxalates, regulatory pathways involved in balancing the expression of the different enzymes mentioned, and paralogous enzymatic studies that prove similar activities in phylogenetically distant genes. Moreover, the considerable amount of structurally and chemically different compounds that are classified in the same category but are treated as similar, although found in various matrices, makes the applicability of the metabolic study limited. For instance, strains that might specifically hydrolyze specific sugar moieties attached to oat saponins might not apply to the same sugar moieties attached to other aglycone structures found in soybeans because of substrate specificity. Another limitation is the assumption that different plant‐based matrices are similar, leading to attempts to use the same starter cultures. This overlooks that these cultures exhibit distinct phenotypic performances depending on the specific matrix.
The increasing demand for plant‐based products might lead to the discovery of new safety challenges that the milk industry has been evaluating for a long time. New and unexplored safety hazards found in plant‐based matrices may be considered risk‐associated compounds that could also be eliminated by fermentation, such as biogenic amines, allergenic compounds, pesticide residues, mycotoxins, and bacterial heat‐resistant toxins generated from spoilage, and pathogenic agents such as Aspergillus sp. and Bacillus cereus (Bartula et al., 2023; Lin et al., 2023). Developing plant‐based products requires reestablishing food processing methods distinct from those used in the milk industry. These new approaches could reduce those associated risks. Pre‐processing and processing methods such as plant protein extraction methods or pasteurization might affect the fermentation performance of starter cultures in plant‐based matrices, affecting protein, carbohydrate, and/or nutrient bioavailability and changing their metabolic outcome. Nevertheless, some methods, such as soaking and heat treatment, might help to reduce the antinutrient and off‐flavor concentrations, such as phytates and trypsin inhibitors, in plant bases, facilitating the antinutrient degradation outcome after fermentation. A deeper understanding of how methods such as salting in cheese processing influence the metabolic activity of LAB during ripening could shed light on their role in the reduction of antinutrient compounds. This could help clarify how such processes affect the breakdown or removal of undesirable compounds during cheese ripening (Hickey et al., 2017).
Figure 2 represents the results of blasting multiple protein sequences to multiple LAB strains on the QPS list. Figure 2 is divided into a and b, in which the blast results are differently represented. Figure 2a analyzes the average number of copies of each gene in all strains from the same species. In contrast, Figure 2b shows that species have a higher number of genes from each antinutrient category, standardizing the presence of all genes from the same category on a scale from 0 to 1. Figure 2b helps to decide, which species to select when searching for a specific application based on the genes analyzed. For instance, L. pentosus strains showed high scores in the glycosidases considered relevant for degrading saponins but also a high number of genes in the phenolic metabolism, considering it a good candidate for those two phenotypes. The biplot adds extra information on the correlation between antinutrient categories and species, showing genetically similar species by the scores in relation to each loading. The PCA plot represents around 50% of the exposed data, and species such as L. plantarum, L. aviarius, and L. brevis highly contribute to the phenolic, oxalate, and saponin degrading phenotypic variables. Moreover, the PCA plot might provide a starting point when selecting LAB species for synergistic purposes, in strain mixture designs, and when selecting the best candidates from each category for combining them when developing starter cultures.
Figure 2 reveals that species from the Lactiplantibacillus (L. argentoratensis, L. paraplantarum, L. pentosus, and L. plantarum) genus encoded for multiple genes and copies of the same involved in the degradation of phenolic acid compounds where L. plantarum is the species with higher number of genes involved in this metabolism (Figure 2b). Moreover, tyrosine‐like phosphatase (phytase) found in L. fermentum was blasted against all genomes, only finding homologous genes in all strains of L. brevis and L. fermentum. Gut and animal‐derived LAB species such as L. johnsonii, L. acidophilus, L. mucosae, L. aviarius, and L. reuteri showed a higher presence of frc and oxc genes mainly involved in oxalate metabolism. For instance, one strain of L. acidophilus, L. mucosae, and L. helveticus species was determined to encode 2, 2, and 3 copies, respectively, of both genes in their genome. Furthermore, β‐glucosidase, β‐galactosidase, β‐glucuronosidase, α‐l‐arabinosidase, β‐xylosidase, and α‐l‐rhamnosidases were blasted in the form of GH families due to their heterospecific enzymatic activities showing L. diolivorans, L. brevis, L. pentosus, and P. acidilactici as the species with higher number of homologous genes involved in the degradation of glycoside bonds present in saponins (Figure 2a,b). Higher presence of envelope‐based proteinase coding genes possibly involved in the enzyme inhibitor degradation was found in L. rhamnosus, L. gallinarum, L. delbrueckii, and L. kefiranofaciens having prtP and prtB as the primary proteolytic genes present. Finally, a higher number of genes and copies of genes involved in RFOs degradation were present in L. plantarum, L. aviarius, and L. mucosae mainly due to the high presence of levS transporters, melA, and ftfA high‐copies in some the strains. Figure 2a presents the mean values of genes detected in each strain from the evaluated species without indicating variability within strains. In certain instances, some genes were highly conserved across strains within the same species, while in other cases, the presence of genes depended on the specific strain. For instance, the five L. plantarum strains evaluated showed 6, 7, 7, 0, and 0 copies of levS transporter genes, as an example of a species with high genetic variability due to its adaptability throughout gene transfer events to its living environment. Moreover, S. collinoides and L. helveticus showed three copies of hcrB and GH1 in one strain, respectively, but no presence or one copy of the gene in the other strains analyzed. Thus, LAB‐conserved genes within species might suggest the importance of those in ancestral evolutive genetic fitness and surveillance of the LAB species. In contrast, strain‐based genetic variability might be related to specific niche‐adaptation events. New knowledge on uncharacterized paralogous genes involved in identical phenotypes or similar substrate heterospecificities might be relevant for future research. Moreover, this genetic analysis can inspire the selection of unconventional LAB species that have not yet been used as starter cultures or investigated in spontaneous fermentations, such as L. diolivorans, L. pentosus, and S. collinoides.
8. CONCLUSIONS
The genetic evolution of LAB strains is crucial when selecting LAB strains as starter cultures for fermentation. With the high demand for PBFDA food products with good nutritional and sensory attributes, other, perhaps unconventional, LAB species and strains might have genetic traits that could provide desired phenotypes for that purpose. We evaluated the state‐of‐the‐art of relevant LAB genetic tools that are characterized to be associated with the degradation of antinutrients (phytates, RFOs, saponins, phenolic compounds, oxalates, and enzymatic inhibitors) and volatile off‐flavors (aldehydes, pyrazines, furans, sulfur compounds, and ketones) present in legumes, cereals, tubers, and nuts. Genetic characterization, enzymatic expression and activity, and enzymatic substrate heterospecificity are important characteristics when genotyping and phenotyping LAB strains for fermented food applications. The homology‐based gene analysis of all QPS‐listed LAB strains has provided genetic information on unconventional species with promising genetic tools for each phenotype and strain‐dependent genetic variabilities of target genes, functional for future starter culture selections and microbial characterization. Future work on linking protein sequences and structures to enzymatic expression and specificity through modeling might be required to speed up the starter culture selection process and design mixtures of LAB strains with different genetic tools to be used as starter cultures for developing PBFDAs.
AUTHOR CONTRIBUTIONS
Guillermo Eduardo Sedó Molina: Conceptualization; investigation; writing—original draft; methodology; visualization; writing—review and editing; formal analysis; data curation. Geoffrey Ras: Conceptualization; validation; visualization; writing—review and editing; supervision; formal analysis. Denise Felix da Silva: Conceptualization; validation; visualization; writing—review and editing; formal analysis; supervision. Lene Duedahl‐Olesen: Conceptualization; validation; visualization; writing—review and editing; formal analysis; supervision. Egon Bech Hansen: Conceptualization; validation; visualization; writing—review and editing; formal analysis; supervision. Claus Heiner Bang‐Berthelsen: Conceptualization; funding acquisition; methodology; validation; writing—review and editing; formal analysis; project administration; supervision; resources.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
Supporting information
Supporting Information
Supporting Information
Supporting Information
ACKNOWLEDGMENTS
Authors would like to acknowledge the National Food Institute, Technical University of Denmark (DTU), and Kongens Lyngby, Denmark, for their support and the support from Novonesis. Additionally, the authors would like to thank AgriFoodTure for providing the financial foundation for the REPLANTED project. The grant supports several of the authors listed (GESM, CHBB, LDO, and EBH). Additionally, GR and DFDS are full‐time employees of Novonesis.
Molina, G. E. S. , Ras, G. , da Silva, D. F. , Duedahl‐Olesen, L. , Hansen, E. B. , & Bang‐Berthelsen, C. H. (2025). Metabolic insights of lactic acid bacteria in reducing off‐flavors and antinutrients in plant‐based fermented dairy alternatives. Comprehensive Reviews in Food Science and Food Safety, 24, e70134. 10.1111/1541-4337.70134
REFERENCES
- Acin‐Albiac, M. , Filannino, P. , Arora, K. , Da Ros, A. , Gobbetti, M. , & Di Cagno, R. (2021). Role of lactic acid bacteria phospho‐β‐glucosidases during the fermentation of cereal by‐products. Foods, 10(1), 97. 10.3390/FOODS10010097 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Alemayehu, D. , Hannon, J. A. , McAuliffe, O. , & Ross, R. P. (2014). Characterization of plant‐derived lactococci on the basis of their volatile compounds profile when grown in milk. International Journal of Food Microbiology, 172, 57–61. 10.1016/j.ijfoodmicro.2013.11.024 [DOI] [PubMed] [Google Scholar]
- Andrabi, S. T. , Bhat, B. , Gupta, M. , & Bajaj, B. K. (2016). Phytase‐producing potential and other functional attributes of lactic acid bacteria isolates for prospective probiotic applications. Probiotics and Antimicrobial Proteins, 8(3), 121–129. 10.1007/S12602-016-9220-3 [DOI] [PubMed] [Google Scholar]
- Azcarate‐Peril, M. A. , Bruno‐Bárcena, J. M. , Hassan, H. M. , & Klaenhammer, T. R. (2006). Transcriptional and functional analysis of oxalyl‐coenzyme A (CoA) decarboxylase and formyl‐CoA transferase genes from Lactobacillus acidophilus . Applied and Environmental Microbiology, 72(3), 1891–1899. 10.1128/AEM.72.3.1891-1899.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bachmann, H. , Molenaar, D. , Branco dos Santos, F. , & Teusink, B. (2017). Experimental evolution and the adjustment of metabolic strategies in lactic acid bacteria. FEMS Microbiology Reviews, 41(Suppl. 1), S201–S219. 10.1093/FEMSRE/FUX024 [DOI] [PubMed] [Google Scholar]
- Bačun‐Družina, V. , Mrvčić, J. , Butorac, A. , & Gjuračić, K. (2009). The influence of gene transfer on the lactic acid bacteria evolution. Mljekarstvo: Časopis Za Unaprjeđenje Proizvodnje i Prerade Mlijeka, 59(3), 181–192. [Google Scholar]
- Barthelmebs, L. , Diviès, C. , & Cavin, J. F. (2001). Molecular characterization of the phenolic acid metabolism in the lactic acid bacteria Lactobacillus plantarum . Le Lait, 81(1–2), 161–171. 10.1051/LAIT:2001120 [DOI] [Google Scholar]
- Barthelmebs, L. , Divies, C. , & Cavin, J.‐F. (2000). Knockout of the p‐coumarate decarboxylase gene from Lactobacillus plantarum reveals the existence of two other inducible enzymatic activities involved in phenolic acid metabolism. Applied and Environmental Microbiology, 66(8), 3368–3375. 10.1128/AEM.66.8.3368-3375.2000 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bartula, K. , Begley, M. , Latour, N. , & Callanan, M. (2023). Growth of food‐borne pathogens Listeria and Salmonella and spore‐forming Paenibacillus and Bacillus in commercial plant‐based milk alternatives. Food Microbiology, 109, 104143. 10.1016/j.fm.2022.104143 [DOI] [PubMed] [Google Scholar]
- Batt, C. A. (2014). LACTOCOCCUS | Introduction. Encyclopedia of food microbiology (2nd ed., pp. 439–441). Elsevier. 10.1016/B978-0-12-384730-0.00181-6 [DOI] [Google Scholar]
- Baur, S. , Frank, O. , Hausladen, H. , Hückelhoven, R. , Hofmann, T. , Eisenreich, W. , & Dawid, C. (2021). Biosynthesis of α‐solanine and α‐chaconine in potato leaves (Solanum tuberosum L.)—A 13CO2 study. Food Chemistry, 365, 130461. 10.1016/j.foodchem.2021.130461 [DOI] [PubMed] [Google Scholar]
- Beekwilder, J. , Marcozzi, D. , Vecchi, S. , de Vos, R. , Janssen, P. , Francke, C. , van Hylckama Vlieg, J. , & Hall, R. D. (2009). Characterization of rhamnosidases from Lactobacillus plantarum and Lactobacillus acidophilus . Applied and Environmental Microbiology, 75(11), 3447–3454. 10.1128/AEM.02675-08 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Boakye, P. G. , Kougblenou, I. , Murai, T. , Okyere, A. Y. , Anderson, J. , Bajgain, P. , Philipp, B. , LaPlante, B. , Schlecht, S. , Vogel, C. , Carlson, M. , Occhino, L. , Stanislawski, H. , Ray, S. S. , & Annor, G. A. (2022). Impact of sourdough fermentation on FODMAPs and amylase‐trypsin inhibitor levels in wheat dough. Journal of Cereal Science, 108, 103574. 10.1016/J.JCS.2022.103574 [DOI] [Google Scholar]
- Bravo, L. (1998). Polyphenols: Chemistry, dietary sources, metabolism, and nutritional significance. Nutrition Reviews, 56(11), 317–333. 10.1111/J.1753-4887.1998.TB01670.X [DOI] [PubMed] [Google Scholar]
- Buron‐Moles, G. , Chailyan, A. , Dolejs, I. , Forster, J. , & Mikš, M. H. (2019). Uncovering carbohydrate metabolism through a genotype‐phenotype association study of 56 lactic acid bacteria genomes. Applied Microbiology and Biotechnology, 103(7), 3135–3152. 10.1007/S00253-019-09701-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Çabuk, B. , Nosworthy, M. G. , Stone, A. K. , Korber, D. R. , Tanaka, T. , House, J. D. , & Nickerson, M. T. (2018). Effect of fermentation on the protein digestibility and levels of non‐nutritive compounds of pea protein concentrate. Food Technology and Biotechnology, 56(2), 257–264. 10.17113/ftb.56.02.18.5450 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cai, H. , Thompson, R. , Budinich, M. F. , Broadbent, J. R. , & Steele, J. L. (2009). Genome sequence and comparative genome analysis of Lactobacillus casei: Insights into their niche‐associated evolution. Genome Biology and Evolution, 1, 239–257. 10.1093/gbe/evp019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caminero, A. , McCarville, J. L. , Zevallos, V. F. , Pigrau, M. , Yu, X. B. , Jury, J. , Galipeau, H. J. , Clarizio, A. V. , Casqueiro, J. , Murray, J. A. , Collins, S. M. , Alaedini, A. , Bercik, P. , Schuppan, D. , & Verdu, E. F. (2019). Lactobacilli degrade wheat amylase trypsin inhibitors to reduce intestinal dysfunction induced by immunogenic wheat proteins. Gastroenterology, 156(8), 2266–2280. 10.1053/J.GASTRO.2019.02.028 [DOI] [PubMed] [Google Scholar]
- Carević, M. , Banjanac, K. , Ćorović, M. , Jakovetić, S. , Milivojević, A. , Vukašinović, S. M. , & Bezbradica, D. (2016). Selection of lactic acid bacteria strain for simultaneous production of α‐ and β‐galactosidases. Zaštita Materijala, 57(2), 265–273. 10.5937/ZASMAT1602265C [DOI] [Google Scholar]
- Cavanagh, D. , Fitzgerald, G. F. , & McAuliffe, O. (2015). From field to fermentation: The origins of Lactococcus lactis and its domestication to the dairy environment. Food Microbiology, 47, 45–61. 10.1016/j.fm.2014.11.001 [DOI] [PubMed] [Google Scholar]
- Chen, L. , Bromberger, P. D. , Nieuwenhuiys, G. , & Hatti‐Kaul, R. (2016). Redox balance in Lactobacillus reuteri DSM20016: Roles of iron‐dependent alcohol dehydrogenases in glucose/glycerol metabolism. PLOS ONE, 11(12), e0168107. 10.1371/JOURNAL.PONE.0168107 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cho, J. G. , Gebhart, C. J. , Furrow, E. , & Lulich, J. P. (2015). Assessment of in vitro oxalate degradation by Lactobacillus species cultured from veterinary probiotics. American Journal of Veterinary Research, 76(9), 801–806. 10.2460/AJVR.76.9.801 [DOI] [PubMed] [Google Scholar]
- Christensen, L. F. , García‐Béjar, B. , Bang‐Berthelsen, C. H. , & Hansen, E. B. (2022). Extracellular microbial proteases with specificity for plant proteins in food fermentation. International Journal of Food Microbiology, 381, 109889. 10.1016/J.IJFOODMICRO.2022.109889 [DOI] [PubMed] [Google Scholar]
- Cid‐Gallegos, M. S. , Corzo‐Ríos, L. J. , Jiménez‐Martínez, C. , & Sánchez‐Chino, X. M. (2022). Protease inhibitors from plants as therapeutic agents—A review. Plant Foods for Human Nutrition, 77(1), 20–29. 10.1007/s11130-022-00949-4 [DOI] [PubMed] [Google Scholar]
- Coulon, S. , Chemardin, P. , Gueguen, Y. , Arnaud, A. , & Galzy, P. (1998). Purification and characterization of an intracellular β‐glucosidase from Lactobacillus casei ATCC 393. Applied Biochemistry and Biotechnology, 74(2), 105–114. 10.1007/BF02787177 [DOI] [Google Scholar]
- Damayanti, E. , Ratisiwi, F. N. , Istiqomah, L. , Sembiring, L. , & Febrisiantosa, A. (2017). Phytate degrading activities of lactic acid bacteria isolated from traditional fermented food. Polish Journal of Food and Nutrition Sciences, 58(1), 020053. 10.1063/1.4978126 [DOI] [Google Scholar]
- Das, G. , Sharma, A. , & Sarkar, P. K. (2022). Conventional and emerging processing techniques for the post‐harvest reduction of antinutrients in edible legumes. Applied Food Research, 2(1), 100112. 10.1016/J.AFRES.2022.100112 [DOI] [Google Scholar]
- De Angelis, M. , Gallo, G. , Corbo, M. R. , McSweeney, P. L. H. , Faccia, M. , Giovine, M. , & Gobbetti, M. (2003). Phytase activity in sourdough lactic acid bacteria: Purification and characterization of a phytase from Lactobacillus sanfranciscensis CB1. International Journal of Food Microbiology, 87(3), 259–270. 10.1016/S0168-1605(03)00072-2 [DOI] [PubMed] [Google Scholar]
- de Vos, W. M. , & Vaughan, E. E. (1994). Genetics of lactose utilization in lactic acid bacteria. FEMS Microbiology Reviews, 15(2–3), 217–237. 10.1111/J.1574-6976.1994.TB00136.X [DOI] [PubMed] [Google Scholar]
- Dudareva, N. , Klempien, A. , Muhlemann, J. K. , & Kaplan, I. (2013). Biosynthesis, function and metabolic engineering of plant volatile organic compounds. New Phytologist, 198(1), 16–32. 10.1111/NPH.12145 [DOI] [PubMed] [Google Scholar]
- Elango, D. , Rajendran, K. , Van der Laan, L. , Sebastiar, S. , Raigne, J. , Thaiparambil, N. A. , El Haddad, N. , Raja, B. , Wang, W. , Ferela, A. , Chiteri, K. O. , Thudi, M. , Varshney, R. K. , Chopra, S. , Singh, A. , & Singh, A. K. (2022). Raffinose family oligosaccharides: Friend or foe for human and plant health? Frontiers in Plant Science, 13, 829118. 10.3389/fpls.2022.829118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Elleuche, S. , & Antranikian, G. (2013). Bacterial group III alcohol dehydrogenases—Function, evolution and biotechnological applications. OA Alcohol, 1(1), 2–6. 10.13172/2053-0285-1-1-489 24432188 [DOI] [Google Scholar]
- Endo, A. , Tanaka, N. , Oikawa, Y. , Okada, S. , & Dicks, L. (2014). Fructophilic characteristics of Fructobacillus spp. may be due to the absence of an alcohol/acetaldehyde dehydrogenase gene (adhE). Current Microbiology, 68(4), 531–535. 10.1007/s00284-013-0506-3 [DOI] [PubMed] [Google Scholar]
- Engels, W. , Siu, J. , van Schalkwijk, S. , Wesselink, W. , Jacobs, S. , & Bachmann, H. (2022). Metabolic conversions by lactic acid bacteria during plant protein fermentations. Foods, 11(7), 1005. 10.3390/FOODS11071005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Filannino, P. , Di Cagno, R. , Addante, R. , Pontonio, E. , & Gobbetti, M. (2016). Metabolism of fructophilic lactic acid bacteria isolated from the Apis mellifera L. bee gut: Phenolic acids as external electron acceptors. Applied and Environmental Microbiology, 82(23), 6899–6911. 10.1128/AEM.02194-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fischer, E. , Cayot, N. , & Cachon, R. (2022). Potential of microorganisms to decrease the “beany” off‐flavor: A review. Journal of Agricultural and Food Chemistry, 70(15), 4493–4508. 10.1021/acs.jafc.1c07505 [DOI] [PubMed] [Google Scholar]
- Fontana, A. , Zacconi, C. , & Morelli, L. (2018). Genetic signatures of dairy Lactobacillus casei group. Frontiers in Microbiology, 9, 401945. 10.3389/fmicb.2018.02611 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fraberger, V. , Ladurner, M. , Nemec, A. , Grunwald‐Gruber, C. , Call, L. M. , Hochegger, R. , Domig, K. J. , & D'amico, S. (2020). Insights into the potential of sourdough‐related lactic acid bacteria to degrade proteins in wheat. Microorganisms, 8(11), 1689. 10.3390/MICROORGANISMS8111689 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fu, G. , Chen, K. , Wang, J. , Wang, M. , Li, R. , Wu, X. , Wu, C. , Zhang, P. , Liu, C. , & Wan, Y. (2020). Screening of tea saponin‐degrading strain to degrade the residual tea saponin in tea seed cake. Preparative Biochemistry & Biotechnology, 50(7), 697–707. 10.1080/10826068.2020.1731827 [DOI] [PubMed] [Google Scholar]
- Gänzle, M. G. (2015). Lactic metabolism revisited: Metabolism of lactic acid bacteria in food fermentations and food spoilage. Current Opinion in Food Science, 2, 106–117. 10.1016/J.COFS.2015.03.001 [DOI] [Google Scholar]
- Gänzle, M. G. , & Follador, R. (2012). Metabolism of oligosaccharides and starch in lactobacilli: A review. Frontiers in Microbiology, 3(SEP), 33219. 10.3389/FMICB.2012.00340/BIBTEX [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gao, Y. L. , Wang, C. S. , Zhu, Q. H. , & Qian, G. Y. (2013). Optimization of solid‐state fermentation with Lactobacillus brevis and Aspergillus oryzae for trypsin inhibitor degradation in soybean meal. Journal of Integrative Agriculture, 12(5), 869–876. 10.1016/S2095-3119(13)60305-6 [DOI] [Google Scholar]
- García‐Mantrana, I. , Monedero, V. , & Haros, M. (2015). Reduction of phytate in soy drink by fermentation with Lactobacillus casei expressing phytases from bifidobacteria. Plant Foods for Human Nutrition, 70(3), 269–274. 10.1007/S11130-015-0489-2 [DOI] [PubMed] [Google Scholar]
- Garro, M. S. , de Giori, G. S. , de Valdez, G. F. , & Oliver, G. (1993). Characterization of alpha‐galactosidase from Lactobacillus fermentum . Journal of Applied Bacteriology, 75(5), 485–488. 10.1111/J.1365-2672.1993.TB02805.X [DOI] [Google Scholar]
- Gaur, G. , Damm, S. , Passon, M. , Lo, H. K. , Schieber, A. , & Gänzle, M. G. (2023). Conversion of hydroxycinnamic acids by Furfurilactobacillus milii in sorghum fermentations: Impact on profile of phenolic compounds in sorghum and on ecological fitness of Ff. milii . Food Microbiology, 111, 104206. 10.1016/J.FM.2022.104206 [DOI] [PubMed] [Google Scholar]
- Gaur, G. , & Gänzle, M. G. (2023). Conversion of (poly)phenolic compounds in food fermentations by lactic acid bacteria: Novel insights into metabolic pathways and functional metabolites. Current Research in Food Science, 6, 100448. 10.1016/J.CRFS.2023.100448 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gaur, G. , Oh, J.‐H. , Filannino, P. , Gobbetti, M. , van Pijkeren, J.‐P. , & Gänzle, M. G. (2020). Genetic determinants of hydroxycinnamic acid metabolism in heterofermentative lactobacilli. Applied and Environmental Microbiology, 86(5), e02461–19. 10.1128/AEM.02461-19 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gemede, H. F. , & Ratta, N. (2014). Antinutritional factors in plant foods: Potential health benefits and adverse effects. International Journal of Nutrition and Food Sciences, 3(4), 284–289. 10.11648/J.IJNFS.20140304.18 [DOI] [Google Scholar]
- Genet, B. M. L. , Molina, G. E. S. , Wätjen, A. P. , Barone, G. , Albersten, K. , Ahrné, L. M. , Hansen, E. B. , & Bang‐Berthelsen, C. H. (2023). Hybrid cheeses—Supplementation of cheese with plant‐based ingredients for a tasty, nutritious and sustainable food transition. Fermentation, 9(7), 667. 10.3390/FERMENTATION9070667 [DOI] [Google Scholar]
- Gomathi, S. , Sasikumar, P. , Anbazhagan, K. , Sasikumar, S. , Kavitha, M. , Selvi, M. S. , & Selvam, G. S. (2014). Screening of indigenous oxalate degrading lactic acid bacteria from human faeces and South Indian fermented foods: Assessment of probiotic potential. The Scientific World Journal, 2014(1), 648059. 10.1155/2014/648059 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gorissen, S. H. M. , Crombag, J. J. R. , Senden, J. M. G. , Waterval, W. A. H. , Bierau, J. , Verdijk, L. B. , & van Loon, L. J. C. (2018). Protein content and amino acid composition of commercially available plant‐based protein isolates. Amino Acids, 50(12), 1685–1695. 10.1007/S00726-018-2640-5/TABLES/2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grasso, N. , Alonso‐Miravalles, L. , & O'Mahony, J. A. (2020). Composition, physicochemical and sensorial properties of commercial plant‐based yogurts. Foods, 9(3), 252. 10.3390/FOODS9030252 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grasso, N. , Roos, Y. H. , Crowley, S. V. , Arendt, E. K. , & O'Mahony, J. A. (2021). Composition and physicochemical properties of commercial plant‐based block‐style products as alternatives to cheese. Future Foods, 4, 100048. 10.1016/J.FUFO.2021.100048 [DOI] [Google Scholar]
- Gueguen, Y. , Chemardin, P. , Labrot, P. , Arnaud, A. , & Galzy, P. (1997). Purification and characterization of an intracellular β‐glucosidase from a new strain of Leuconostoc mesenteroides isolated from cassava. Journal of Applied Microbiology, 82(4), 469–476. 10.1046/J.1365-2672.1997.00136.X [DOI] [Google Scholar]
- Gupta, M. K. , Torrico, D. D. , Ong, L. , Gras, S. L. , Dunshea, F. R. , & Cottrell, J. J. (2022). Plant and dairy‐based yogurts: A comparison of consumer sensory acceptability linked to textural analysis. Foods, 11(3), 463. 10.3390/FOODS11030463/S1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haros, M. , Bielecka, M. , & Sanz, Y. (2005). Phytase activity as a novel metabolic feature in Bifidobacterium . FEMS Microbiology Letters, 247(2), 231–239. 10.1016/j.femsle.2005.05.008 [DOI] [PubMed] [Google Scholar]
- Harper, A. R. , Dobson, R. C. J. , Morris, V. K. , & Moggré, G. J. (2022). Fermentation of plant‐based dairy alternatives by lactic acid bacteria. Microbial Biotechnology, 15(5), 1404–1421. 10.1111/1751-7915.14008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hati, S. , Vij, S. , Mandal, S. , Malik, R. K. , Kumari, V. , & Khetra, Y. (2014). α‐Galactosidase activity and oligosaccharides utilization by lactobacilli during fermentation of soy milk. Journal of Food Processing and Preservation, 38(3), 1065–1071. 10.1111/JFPP.12063 [DOI] [Google Scholar]
- He, Y. , Hu, Z. , Li, A. , Zhu, Z. , Yang, N. , Ying, Z. , He, J. , Wang, C. , Yin, S. , & Cheng, S. (2019). Recent advances in biotransformation of saponins. Molecules, 24(13), 2365. 10.3390/MOLECULES24132365 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hennessy, R. C. , Jørgensen, N. O. G. , Scavenius, C. , Enghild, Jan. J. , Greve‐Poulsen, M. , Sørensen, O. B. , & Stougaard, P. (2018). A screening method for the isolation of bacteria capable of degrading toxic steroidal glycoalkaloids present in potato. Frontiers in Microbiology, 9, 2648. 10.3389/fmicb.2018.02648 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hennessy, R. C. , Nielsen, S. D. , Greve‐Poulsen, M. , Larsen, L. B. , Sørensen, O. B. , & Stougaard, P. (2020). Discovery of a bacterial gene cluster for deglycosylation of toxic potato steroidal glycoalkaloids α‐chaconine and α‐solanine. Journal of Agricultural and Food Chemistry, 68(5), 1390–1396. 10.1021/acs.jafc.9b07632 [DOI] [PubMed] [Google Scholar]
- Hickey, C. D. , Auty, M. A. E. , Wilkinson, M. G. , & Sheehan, J. J. (2017). Influence of process temperature and salting methods on starter and NSLAB growth and enzymatic activity during the ripening of cheeses produced with Streptococcus thermophilus and Lactobacillus helveticus . International Dairy Journal, 69, 9–18. 10.1016/J.IDAIRYJ.2016.12.015 [DOI] [Google Scholar]
- Hoffmann, E. M. , Muetzel, S. , & Becker, K. (2003). The fermentation of soybean meal by rumen microbes in vitro reveals different kinetic features for the inactivation and the degradation of trypsin inhibitor protein. Animal Feed Science and Technology, 106(1–4), 189–197. 10.1016/S0377-8401(02)00321-8 [DOI] [Google Scholar]
- Hu, Z. , Jia, P. , Bai, Y. , Fan, T. ping , Zheng, X. , & Cai, Y. (2019). Characterisation of five alcohol dehydrogenases from Lactobacillus reuteri DSM20016. Process Biochemistry, 86, 73–79. 10.1016/J.PROCBIO.2019.08.010 [DOI] [Google Scholar]
- Huang, X. , Schuppan, D. , Tovar, L. E. R. , Zevallos, V. F. , Loponen, J. , & Gänzle, M. (2020). Sourdough fermentation degrades wheat alpha‐amylase/trypsin inhibitor (ATI) and reduces pro‐inflammatory activity. Foods, 9(7), 943. 10.3390/FOODS9070943 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Iyalomhe, O. , Khantwal, C. M. , & Kang, D. C. (2015). The structure and function of OxlT, the oxalate transporter of Oxalobacter formigenes . Journal of Membrane Biology, 248(4), 641–650. 10.1007/s00232-014-9728-y [DOI] [PubMed] [Google Scholar]
- Ji, D. , Ma, J. , Xu, M. , & Agyei, D. (2021). Cell‐envelope proteinases from lactic acid bacteria: Biochemical features and biotechnological applications. Comprehensive Reviews in Food Science and Food Safety, 20(1), 369–400. 10.1111/1541-4337.12676 [DOI] [PubMed] [Google Scholar]
- Jiménez, N. , Esteban‐Torres, M. , Mancheño, J. M. , de las Rivas, B. , & Muñoz, R. (2014). Tannin degradation by a novel tannase enzyme present in some Lactobacillus plantarum strains. Applied and Environmental Microbiology, 80(10), 2991–2997. 10.1128/AEM.00324-14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jiménez, N. , Curiel, J. A. , Reverón, I. , de las Rivas, B. , & Muñoz, R. (2013). Uncovering the Lactobacillus plantarum WCFS1 gallate decarboxylase involved in tannin degradation. Applied and Environmental Microbiology, 79(14), 4253–4263. 10.1128/AEM.00840-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kahala, M. , Ikonen, I. , Blasco, L. , Bragge, R. , Pihlava, J. M. , Nurmi, M. , & Pihlanto, A. (2023). Effect of lactic acid bacteria on the level of antinutrients in pulses: A case study of a fermented faba bean–oat product. Foods, 12(21), 3922. 10.3390/FOODS12213922 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kandler, O. (1983). Carbohydrate metabolism in lactic acid bacteria. Antonie Van Leeuwenhoek, 49(3), 209–224. 10.1007/BF00399499 [DOI] [PubMed] [Google Scholar]
- Kanwal, F. , Ren, D. , Kanwal, W. , Ding, M. , Su, J. , & Shang, X. (2023). The potential role of nondigestible Raffinose family oligosaccharides as prebiotics. Glycobiology, 33, 274–288. 10.1093/glycob/cwad015 [DOI] [PubMed] [Google Scholar]
- Katz, M. , Medina, R. , Gonzalez, S. , & Oliver, G. (2002). Esterolytic and lipolytic activities of lactic acid bacteria isolated from ewe's milk and cheese. Journal of Food Protection, 65(12), 1997–2001. 10.4315/0362-028X-65.12.1997 [DOI] [PubMed] [Google Scholar]
- Khayatt, B. I. , van Noort, V. , & Siezen, R. J. (2020). The genome of the plant‐associated lactic acid bacterium Lactococcus lactis KF147 harbors a hybrid NRPS‐PKS system conserved in strains of the dental cariogenic Streptococcus mutans . Current Microbiology, 77(1), 136–145. 10.1007/S00284-019-01799-1/FIGURES/4 [DOI] [PubMed] [Google Scholar]
- Konings, W. N. , Kok, J. , Kuipers, O. P. , & Poolman, B. (2000). Lactic acid bacteria: The bugs of the new millennium. Current Opinion in Microbiology, 3(3), 276–282. 10.1016/S1369-5274(00)00089-8 [DOI] [PubMed] [Google Scholar]
- Kullin, B. , Tannock, G. W. , Loach, D. M. , Kimura, K. , Abratt, V. R. , & Reid, S. J. (2014). A functional analysis of the formyl‐coenzyme A (frc) gene from Lactobacillus reuteri 100‐23C. Journal of Applied Microbiology, 116(6), 1657–1667. 10.1111/JAM.12500 [DOI] [PubMed] [Google Scholar]
- Laatikainen, R. , Koskenpato, J. , Hongisto, S. M. , Loponen, J. , Poussa, T. , Huang, X. , Sontag‐Strohm, T. , Salmenkari, H. , & Korpela, R. (2017). Pilot study: Comparison of sourdough wheat bread and yeast‐fermented wheat bread in individuals with wheat sensitivity and irritable bowel syndrome. Nutrients, 9(11), 1215. 10.3390/NU9111215 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Landaud, S. , Helinck, S. , & Bonnarme, P. (2008). Formation of volatile sulfur compounds and metabolism of methionine and other sulfur compounds in fermented food. Applied Microbiology and Biotechnology, 77(6), 1191–1205. 10.1007/s00253-007-1288-y [DOI] [PubMed] [Google Scholar]
- Landete, J. M. , Plaza‐Vinuesa, L. , Montenegro, C. , Santamaría, L. , Reverón, I. , de las Rivas, B. , & Muñoz, R. (2021). The use of Lactobacillus plantarum esterase genes: A biotechnological strategy to increase the bioavailability of dietary phenolic compounds in lactic acid bacteria. International Journal of Food Sciences and Nutrition, 72(8), 1035–1045. 10.1080/09637486.2021.1900078 [DOI] [PubMed] [Google Scholar]
- Landete, J. M. , Rodríguez, H. , de las Rivas, B. , & Muñoz, R. (2008). Characterization of a benzyl alcohol dehydrogenase from Lactobacillus plantarum WCFS1. Journal of Agricultural and Food Chemistry, 56(12), 4497–4503. 10.1021/jf800500v [DOI] [PubMed] [Google Scholar]
- Lee, N. K. , Lee, E. K. , & Paik, H. D. (2013). Potential probiotic properties of phytase‐producing Lactobacillus salivarius FC113. Annals of Microbiology, 63(2), 555–560. 10.1007/S13213-012-0503-Y [DOI] [Google Scholar]
- Leonard, W. , Zhang, P. , Ying, D. , & Fang, Z. (2022). Surmounting the off‐flavor challenge in plant‐based foods. Critical Reviews in Food Science and Nutrition, 63(30), 10585–10606. 10.1080/10408398.2022.2078275 [DOI] [PubMed] [Google Scholar]
- Lewanika, T. R. , Reid, S. J. , Abratt, V. R. , Macfarlane, G. T. , & Macfarlane, S. (2007). Lactobacillus gasseri gasser AM63T degrades oxalate in a multistage continuous culture simulator of the human colonic microbiota. FEMS Microbiology Ecology, 61(1), 110–120. 10.1111/J.1574-6941.2007.00327.X [DOI] [PubMed] [Google Scholar]
- Li, P. , Liu, C. , Luo, Y. , Shi, H. , Li, Q. , Pinchu, C. , Li, X. , Yang, J. , & Fan, W. (2022). Oxalate in plants: Metabolism, function, regulation, and application. Journal of Agricultural and Food Chemistry, 70(51), 16037–16049. 10.1021/ACS.JAFC.2C04787 [DOI] [PubMed] [Google Scholar]
- Lin, X. , Duan, N. , Wu, J. , Lv, Z. , Wang, Z. , & Wu, S. (2023). Potential food safety risk factors in plant‐based foods: Source, occurrence, and detection methods. In Trends in food science and technology (Vol. 138, pp. 511–522). Elsevier Ltd. 10.1016/j.tifs.2023.06.032 [DOI] [Google Scholar]
- Liu, M. , Devlin, J. C. , Hu, J. , Volkova, A. , Battaglia, T. W. , Ho, M. , Asplin, J. R. , Byrd, A. , Loke, P. , Li, H. , Ruggles, K. V. , Tsirigos, A. , Blaser, M. J. , & Nazzal, L. (2021). Microbial genetic and transcriptional contributions to oxalate degradation by the gut microbiota in health and disease. ELife, 10, e63642. 10.7554/eLife.63642 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liu, X. , Vrieling, K. , & Klinkhamer, P. G. L. (2017). Interactions between plant metabolites affect herbivores: A study with pyrrolizidine alkaloids and chlorogenic acid. Frontiers in Plant Science, 8, 903. 10.3389/FPLS.2017.00903 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lonvaud‐Funel, A. (2014). Leuconostocaceae family. Encyclopedia of food microbiology (2nd ed., pp. 455–465). 10.1016/B978-0-12-384730-0.00185-3 [DOI] [Google Scholar]
- López‐Moreno, M. , Garcés‐Rimón, M. , & Miguel, M. (2022). Antinutrients: Lectins, goitrogens, phytates and oxalates, friends or foe? Journal of Functional Foods, 89, 104938. 10.1016/J.JFF.2022.104938 [DOI] [Google Scholar]
- Ma, K. K. , Nolden, A. A. , & Kinchla, A. J. (2022). Sensory qualities of plant protein foods. In Plant protein foods (pp. 313–342). Springer International Publishing. 10.1007/978-3-030-91206-2_11 [DOI] [Google Scholar]
- Madsen, S. K. , Priess, C. , Wätjen, A. P. , Øzmerih, S. , Mohammadifar, M. A. , & Heiner Bang‐Berthelsen, C. (2021). Development of a yoghurt alternative, based on plant‐adapted lactic acid bacteria, soy drink and the liquid fraction of brewers’ spent grain. FEMS Microbiology Letters, 368(15), fnab093. 10.1093/FEMSLE/FNAB093 [DOI] [PubMed] [Google Scholar]
- Maeno, S. , Kajikawa, A. , Dicks, L. , & Endo, A. (2019). Introduction of bifunctional alcohol/acetaldehyde dehydrogenase gene (adhE) in Fructobacillus fructosus settled its fructophilic characteristics. Research in Microbiology, 170(1), 35–42. 10.1016/J.RESMIC.2018.09.004 [DOI] [PubMed] [Google Scholar]
- Magboul, A. A. A. , & McSweeney, P. L. H. (1999). Purification and characterization of an acid phosphatase from Lactobacillus plantarum DPC2739. Food Chemistry, 65(1), 15–22. 10.1016/S0308-8146(98)00255-6 [DOI] [Google Scholar]
- Makarova, K. S. , & Koonin, E. V. (2007). Evolutionary genomics of lactic acid bacteria. Journal of Bacteriology, 189(4), 1199–1208. 10.1128/JB.01351-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Manousi, N. , & Zachariadis, G. A. (2019). Determination of volatile compounds in nut‐based milk alternative beverages by HS‐SPME prior to GC‐MS analysis. Molecules, 24(17), 3091. 10.3390/MOLECULES24173091 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McGorrin, R. J. (2019). Key aroma compounds in oats and oat cereals. Journal of Agricultural and Food Chemistry, 67(50), 13778–13789. 10.1021/acs.jafc.9b00994 [DOI] [PubMed] [Google Scholar]
- Merzlov, S. , Tsebro, A. , Rol, N. , Nadtochii, V. , & Kachan, A. (2022). Comparative analysis of the amino acid composition of beverages based on plant raw materials and cow's milk. Scientific Horizons, 25(10), 71–78. 10.48077/SCIHOR.25(10).2022.71-78 [DOI] [Google Scholar]
- Michlmayr, H. , & Kneifel, W. (2014). β‐Glucosidase activities of lactic acid bacteria: Mechanisms, impact on fermented food and human health. FEMS Microbiology Letters, 352(1), 1–10. 10.1111/1574-6968.12348 [DOI] [PubMed] [Google Scholar]
- Mittermeier‐Kleßinger, V. K. , Hofmann, T. , & Dawid, C. (2021). Mitigating off‐flavors of plant‐based proteins. Journal of Agricultural and Food Chemistry, 69(32), 9202–9207. 10.1021/acs.jafc.1c03398 [DOI] [PubMed] [Google Scholar]
- Mogna, L. , Pane, M. , Nicola, S. , & Raiteri, E. (2014). Screening of different probiotic strains for their in vitro ability to metabolise oxalates: Any prospective use in humans? Journal of Clinical Gastroenterology, 48, S91–S95. 10.1097/MCG.0000000000000228 [DOI] [PubMed] [Google Scholar]
- Montemurro, M. , Pontonio, E. , Coda, R. , & Rizzello, C. G. (2021). Plant‐based alternatives to yogurt: State‐of‐the‐art and perspectives of new biotechnological challenges. Foods, 10(2), 316. 10.3390/FOODS10020316 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mueller, M. , Zartl, B. , Schleritzko, A. , Stenzl, M. , Viernstein, H. , & Unger, F. M. (2018). Rhamnosidase activity of selected probiotics and their ability to hydrolyse flavonoid rhamnoglucosides. Bioprocess and Biosystems Engineering, 41(2), 221. 10.1007/S00449-017-1860-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Murphy, C. , Murphy, S. , O'Brien, F. , O'Donoghue, M. , Boileau, T. , Sunvold, G. , Reinhart, G. , Kiely, B. , Shanahan, F. , & O'Mahony, L. (2009). Metabolic activity of probiotics‐oxalate degradation. Veterinary Microbiology, 136(1–2), 100–107. 10.1016/j.vetmic.2008.10.005 [DOI] [PubMed] [Google Scholar]
- Nugroho, A. D. W. , van Schalkwijk, S. , Cebeci, S. , Jacobs, S. , Wesselink, W. , Staring, G. , Goerdayal, S. , Prodan, A. , Stijnman, A. , Teuling, E. , Broersen, K. , & Bachmann, H. (2024). Biopurification using non‐growing microorganisms to improve plant protein ingredients. Npj Science of Food, 8(1), 1–9. 10.1038/s41538-024-00290-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Oguntoyinbo, F. A. (2007). Identification and functional properties of dominant lactic acid bacteria isolated at different stages of solid state fermentation of cassava during traditional gari production. World Journal of Microbiology and Biotechnology, 23(10), 1425–1432. 10.1007/s11274-007-9386-7 [DOI] [Google Scholar]
- Okoye, C. O. , Dong, K. , Wang, Y. , Gao, L. , Li, X. , Wu, Y. , & Jiang, J. (2022). Comparative genomics reveals the organic acid biosynthesis metabolic pathways among five lactic acid bacterial species isolated from fermented vegetables. New Biotechnology, 70, 73–83. 10.1016/J.NBT.2022.05.001 [DOI] [PubMed] [Google Scholar]
- Oleszek, M. , & Oleszek, W. (2020). Saponins in food. In Handbook of dietary phytochemicals (pp. 1–40). Springer. 10.1007/978-981-13-1745-3_34-1 [DOI] [Google Scholar]
- Overney, C. E. , & Huang, J. J. (2020). Genome sequence of Bacillus megaterium O1, a saponin‐degrading bacterium. Microbiology Resource Announcements, 9(40), e00524–20. 10.1128/MRA.00524-20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Palacios, M. C. , Haros, M. , Rosell, C. M. , & Sanz, Y. (2005). Characterization of an acid phosphatase from Lactobacillus pentosus: Regulation and biochemical properties. Journal of Applied Microbiology, 98(1), 229–237. 10.1111/J.1365-2672.2004.02447.X [DOI] [PubMed] [Google Scholar]
- Pang, Z. , Chen, J. , Wang, T. , Gao, C. , Li, Z. , Guo, L. , Xu, J. , & Cheng, Y. (2021). Linking plant secondary metabolites and plant microbiomes: A review. Frontiers in Plant Science, 12, 621276. 10.3389/FPLS.2021.621276 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pham, H.‐H.‐T. , Kim, D.‐H. , & Nguyen, T. L. (2023). Wide‐genome selection of lactic acid bacteria harboring genes that promote the elimination of antinutritional factors. Frontiers in Plant Science, 14, 1145041. 10.3389/fpls.2023.1145041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Picazo‐Aragonés, J. , Terrab, A. , & Balao, F. (2020). Plant volatile organic compounds evolution: Transcriptional regulation, epigenetics and polyploidy. International Journal of Molecular Sciences, 21(23), 8956. 10.3390/IJMS21238956 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pony, P. , Rapisarda, C. , Terradot, L. , Marza, E. , & Fronzes, R. (2020). Filamentation of the bacterial bi‐functional alcohol/aldehyde dehydrogenase AdhE is essential for substrate channeling and enzymatic regulation. Nature Communications, 11(1), 1–11. 10.1038/s41467-020-15214-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- Popova, A. , & Mihaylova, D. (2019). Antinutrients in plant‐based foods: A review. The Open Biotechnology Journal, 13(1), 68–76. 10.2174/1874070701913010068 [DOI] [Google Scholar]
- Pua, A. , Tang, V. C. Y. , Goh, R. M. V. , Sun, J. , Lassabliere, B. , & Liu, S. Q. (2022). Ingredients, processing, and fermentation: Addressing the organoleptic boundaries of plant‐based dairy analogues. Foods, 11(6), 875. 10.3390/FOODS11060875 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pulido‐Mateos, E. C. , Lessard‐Lord, J. , Guyonnet, D. , Desjardins, Y. , & Roy, D. (2022). Comprehensive analysis of the metabolic and genomic features of tannin transforming Lactiplantibacillus plantarum strains. Scientific Reports, 12(1), 1–15. 10.1038/s41598-022-26005-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Qian, B. , Yin, L. , Yao, X. , Zhong, Y. , Gui, J. , Lu, F. , Zhang, F. , & Zhang, J. (2018). Effects of fermentation on the hemolytic activity and degradation of Camellia oleifera saponins by Lactobacillus crustorum and Bacillus subtilis . FEMS Microbiology Letters, 365(7), 14. 10.1093/FEMSLE/FNY014 [DOI] [PubMed] [Google Scholar]
- Rajini, K. S. , Aparna, P. , Sasikala, C. , & Ramana, C. V. (2011). Microbial metabolism of pyrazines. Critical Reviews in Microbiology, 37(2), 99–112. 10.3109/1040841X.2010.512267 [DOI] [PubMed] [Google Scholar]
- Rodríguez, H. , Curiel, J. A. , Landete, J. M. , de las Rivas, B. , de Felipe, F. L. , Gómez‐Cordovés, C. , Mancheño, J. M. , & Muñoz, R. (2009). Food phenolics and lactic acid bacteria. International Journal of Food Microbiology, 132(2–3), 79–90. 10.1016/J.IJFOODMICRO.2009.03.025 [DOI] [PubMed] [Google Scholar]
- Russo, M. , Dugo, P. , & Mondello, L. (2021). Aroma compounds. In Handbook of dairy foods analysis (pp. 307–320). CRC Press. 10.1201/9780429342967-14 [DOI] [Google Scholar]
- Sáez, G. D. , Sabater, C. , Fara, A. , & Zárate, G. (2022). Fermentation of chickpea flour with selected lactic acid bacteria for improving its nutritional and functional properties. Journal of Applied Microbiology, 133(1), 181–199. 10.1111/JAM.15401 [DOI] [PubMed] [Google Scholar]
- Saffarionpour, S. (2023). Off‐flavors in pulses and grain legumes and processing approaches for controlling flavor‐plant protein interaction: Application prospects in plant‐based alternative foods. Food and Bioprocess Technology, 17(5), 1141–1182. 10.1007/S11947-023-03148-4 [DOI] [Google Scholar]
- Sakurama, H. , Kishino, S. , Uchibori, Y. , Yonejima, Y. , Ashida, H. , Kita, K. , Takahashi, S. , & Ogawa, J. (2014). β‐Glucuronidase from Lactobacillus brevis useful for baicalin hydrolysis belongs to glycoside hydrolase family 30. Applied Microbiology and Biotechnology, 98(9), 4021–4032. 10.1007/s00253-013-5325-8 [DOI] [PubMed] [Google Scholar]
- Samtiya, M. , Aluko, R. E. , & Dhewa, T. (2020). Plant food anti‐nutritional factors and their reduction strategies: An overview. Food Production, Processing and Nutrition, 2(1), 1–14. 10.1186/S43014-020-0020-5 [DOI] [Google Scholar]
- Samtiya, M. , Aluko, R. E. , Puniya, A. K. , & Dhewa, T. (2021). Enhancing micronutrients bioavailability through fermentation of plant‐based foods: A concise review. Fermentation, 7(2), 63. 10.3390/FERMENTATION7020063 [DOI] [Google Scholar]
- Sandez Penidez, S. H. , Velasco Manini, M. A. , Gerez, C. L. , & Rollán, G. C. (2020). Partial characterization and purification of phytase from Lactobacillus plantarum CRL1964 isolated from pseudocereals. Journal of Basic Microbiology, 60(9), 787–798. 10.1002/JOBM.202000236 [DOI] [PubMed] [Google Scholar]
- Santamaría, L. , Reverón, I. , de Felipe, F. L. , de las Rivas, B. , & Muñoz, R. (2018). Ethylphenol formation by Lactobacillus plantarum: Identification of the enzyme involved in the reduction of vinylphenols. Applied and Environmental Microbiology, 84(17), e01064–18. 10.1128/AEM.01064-18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sedó Molina, G. E. , Ras, G. , Barone, G. , Fernández‐Varela, R. , Felix da Silva, D. , Jacobsen, C. , Duedahl‐Olesen, L. , Bech Hansen, E. , & Heiner Bang‐Berthelsen, C. (2024). Multiphasic and mixture lactic acid bacteria screening approach for the removal of antinutrients and off‐flavors present in a pea, oat and potato blend. Food Research International, 197, 115200. 10.1016/j.foodres.2024.115200 [DOI] [PubMed] [Google Scholar]
- Sedó Molina, G. E. , Shetty, R. , Jacobsen, C. , Duedahl‐Olesen, L. , Hansen, E. B. , & Bang‐Berthelsen, C. H. (2024). Synergistic effect of the coculture of Leuconostoc pseudomesenteroides and Lactococcus lactis, isolated from honeybees, on the generation of plant‐based dairy alternatives based on soy, pea, oat, and potato drinks. Food Microbiology, 118, 104427. 10.1016/J.FM.2023.104427 [DOI] [PubMed] [Google Scholar]
- Sedó Molina, G. E. , Shetty, R. , Xiao, H. , Wätjen, A. P. , Tovar, M. , & Bang‐Berthelsen, C. H. (2022). Development of a novel lactic acid bacteria starter culture approach: From insect microbiome to plant‐based fermentations. LWT—Food Science and Technology, 167, 113797. 10.1016/J.LWT.2022.113797 [DOI] [Google Scholar]
- Sekhar Rao, K. C. , Divakar, S. , Appu Rao, A. G. , Karanth, N. G. , & Sattur, A. P. (2002). Lipoxygenase inhibitor from Lactobacillus casei . Biotechnology Letters, 24(7), 511–513. 10.1023/A:1014839414395 [DOI] [Google Scholar]
- Sestelo, A. B. F. , Poza, M. , & Villa, T. G. (2004). β‐Glucosidase activity in a Lactobacillus plantarum wine strain. World Journal of Microbiology and Biotechnology, 20(6), 633–637. 10.1023/B:WIBI.0000043195.80695.17/METRICS [DOI] [Google Scholar]
- Sharma, N. , Angural, S. , Rana, M. , Puri, N. , Kondepudi, K. K. , & Gupta, N. (2020). Phytase producing lactic acid bacteria: Cell factories for enhancing micronutrient bioavailability of phytate rich foods. Trends in Food Science & Technology, 96, 1–12. 10.1016/J.TIFS.2019.12.001 [DOI] [Google Scholar]
- Sharma, R. , Kumar, P. , Kaushal, V. , Das, R. , & Kumar Navani, N. (2018). A novel protein tyrosine phosphatase like phytase from Lactobacillus fermentum NKN51: Cloning, characterization and application in mineral release for food technology applications. Bioresource Technology, 249, 1000–1008. 10.1016/J.BIORTECH.2017.10.106 [DOI] [PubMed] [Google Scholar]
- Short, E. C. , Kinchla, A. J. , & Nolden, A. A. (2021). Plant‐based cheeses: A systematic review of sensory evaluation studies and strategies to increase consumer acceptance. Foods, 10(4), 725. 10.3390/FOODS10040725 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silvestroni, A. , Connes, C. , Sesma, F. , de Giori, G. S. , & Piard, J.‐C. (2002). Characterization of the melA locus for α‐galactosidase in Lactobacillus plantarum . Applied and Environmental Microbiology, 68(11), 5464–5471. 10.1128/AEM.68.11.5464-5471.2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Smit, G. , Smit, B. A. , & Engels, W. J. M. (2005). Flavour formation by lactic acid bacteria and biochemical flavour profiling of cheese products. FEMS Microbiology Reviews, 29(3), 591–610. 10.1016/J.FMRRE.2005.04.002 [DOI] [PubMed] [Google Scholar]
- Spence, C. , & Kilcawley, N. (2021). What is the relationship between the presence of volatile organic compounds in food and drink products and multisensory flavour perception? Foods, 10(7), 1570. 10.3390/FOODS10071570 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sreeramulu, G. , Srinivasa, D. S. , Nand, K. , & Joseph, R. (1996). Lactobacillus amylovorus as a phytase producer in submerged culture. Letters in Applied Microbiology, 23(6), 385–388. 10.1111/J.1472-765X.1996.TB01342.X [DOI] [Google Scholar]
- Sümengen, M. , Dinçer, S. , & Kaya, A. (2012). Phytase production from Lactobacillus brevis . Turkish Journal of Biology, 36(5), 533–541. 10.3906/biy-1111-2 [DOI] [Google Scholar]
- Sumengen, M. , Dincer, S. , & Kaya, A. (2013). Production and characterization of phytase from Lactobacillus plantarum . Food Biotechnology, 27(2), 105–118. 10.1080/08905436.2013.781507 [DOI] [Google Scholar]
- Szutowska, J. (2020). Functional properties of lactic acid bacteria in fermented fruit and vegetable juices: A systematic literature review. European Food Research and Technology, 246(3), 357–372. 10.1007/s00217-019-03425-7 [DOI] [Google Scholar]
- Tangyu, M. , Fritz, M. , Tan, J. P. , Ye, L. , Bolten, C. J. , Bogicevic, B. , & Wittmann, C. (2023). Flavour by design: Food‐grade lactic acid bacteria improve the volatile aroma spectrum of oat milk, sunflower seed milk, pea milk, and faba milk towards improved flavour and sensory perception. Microbial Cell Factories, 22(1), 133. 10.1186/s12934-023-02147-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tangyu, M. , Muller, J. , Bolten, C. J. , & Wittmann, C. (2019). Fermentation of plant‐based milk alternatives for improved flavour and nutritional value. Applied Microbiology and Biotechnology, 103(23–24), 9263–9275. 10.1007/s00253-019-10175-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tavasoli, S. , Jalali, S. , Naji, M. , Borumandnia, N. , Majd, G. S. , Basiri, A. , Darani, K. K. , Karamad, D. , Tajabadi‐Ebrahimi, M. , & Taheri, M. (2022). Effect of a probiotic supplement containing Lactobacillus acidophilus and Bifidobacterium animalis lactis on urine oxalate in calcium stone formers with hyperoxaluria: A randomized, placebo‐controlled, double‐blind and in‐vitro trial. Urology Journal, 19(3), 179–188. 10.22037/uj.v18i.6789 [DOI] [PubMed] [Google Scholar]
- Theilmann, M. C. , Goh, Y. J. , Nielsen, K. F. , Klaenhammer, T. R. , Barrangou, R. , & Hachem, M. A. (2017). Lactobacillus acidophilus metabolizes dietary plant glucosides and externalizes their bioactive phytochemicals. MBio, 8(6), e01421–17. 10.1128/MBIO.01421-17 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Trindler, C. , Annika Kopf‐Bolanz, K. , & Denkel, C. (2022). Aroma of peas, its constituents and reduction strategies—Effects from breeding to processing. Food Chemistry, 376, 131892. 10.1016/J.FOODCHEM.2021.131892 [DOI] [PubMed] [Google Scholar]
- Tsangalis, D. , & Shah, N. P. (2004). Metabolism of oligosaccharides and aldehydes and production of organic acids in soymilk by probiotic bifidobacteria. International Journal of Food Science & Technology, 39(5), 541–554. 10.1111/J.1365-2621.2004.00814.X [DOI] [Google Scholar]
- Turroni, S. , Bendazzoli, C. , Dipalo, S. C. F. , Candela, M. , Vitali, B. , Gotti, R. , & Brigidi, P. (2010). Oxalate‐degrading activity in Bifidobacterium animalis subsp. lactis: Impact of acidic conditions on the transcriptional levels of the oxalyl coenzyme A (CoA) decarboxylase and formyl‐CoA transferase genes. Applied and Environmental Microbiology, 76(16), 5609–5620. 10.1128/AEM.00844-10 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vanga, S. K. , & Raghavan, V. (2018). How well do plant based alternatives fare nutritionally compared to cow's milk? Journal of Food Science and Technology, 55(1), 10. 10.1007/S13197-017-2915-Y [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vasudha, M. , Prashantkumar, C. S. , Bellurkar, M. , Kaveeshwar, V. , & Gayathri, D. (2023). Probiotic potential of β‐galactosidase producing lactic acid bacteria from fermented milk and their molecular characterization. Biomedical Reports, 18(3), 1–12. 10.3892/BR.2023.1605 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vats, P. , & Banerjee, U. C. (2004). Production studies and catalytic properties of phytases (myo‐inositolhexakisphosphate phosphohydrolases): An overview. Enzyme and Microbial Technology, 35(1), 3–14. 10.1016/J.ENZMICTEC.2004.03.010 [DOI] [Google Scholar]
- Ventura, M. , Turroni, F. , Zomer, A. , Foroni, E. , Giubellini, V. , Bottacini, F. , Canchaya, C. , Claesson, M. J. , He, F. , Mantzourani, M. , Mulas, L. , Ferrarini, A. , Gao, B. , Delledonne, M. , Henrissat, B. , Coutinho, P. , Oggioni, M. , Gupta, R. S. , Zhang, Z. , … Van Sinderen, D. (2009). The Bifidobacterium dentium Bd1 genome sequence reflects its genetic adaptation to the human oral cavity. PLoS Genetics, 5(12), e1000785. 10.1371/JOURNAL.PGEN.1000785 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Verce, M. , De Vuyst, L. , & Weckx, S. (2020). Comparative genomics of Lactobacillus fermentum suggests a free‐living lifestyle of this lactic acid bacterial species. Food Microbiology, 89, 103448. 10.1016/J.FM.2020.103448 [DOI] [PubMed] [Google Scholar]
- Vermeulen, N. , Czerny, M. , Gänzle, M. G. , Schieberle, P. , & Vogel, R. F. (2007). Reduction of (E)‐2‐nonenal and (E,E)‐2,4‐decadienal during sourdough fermentation. Journal of Cereal Science, 45(1), 78–87. 10.1016/J.JCS.2006.07.002 [DOI] [Google Scholar]
- Wang, X. , Kong, X. , Zhang, C. , Hua, Y. , Chen, Y. , & Li, X. (2023). Comparison of physicochemical properties and volatile flavor compounds of plant‐based yoghurt and dairy yoghurt. Food Research International, 164, 112375. 10.1016/J.FOODRES.2022.112375 [DOI] [PubMed] [Google Scholar]
- Wätjen, A. P. , De Vero, L. , Carmona, E. N. , Sberveglieri, V. , Huang, W. , Turner, M. S. , & Bang‐Berthelsen, C. H. (2023). Leuconostoc performance in soy‐based fermentations—Survival, acidification, sugar metabolism, and flavor comparisons. Food Microbiology, 115, 104337. 10.1016/J.FM.2023.104337 [DOI] [PubMed] [Google Scholar]
- Wu, C. , Huang, J. , & Zhou, R. (2017). Genomics of lactic acid bacteria: Current status and potential applications. Critical Reviews in Microbiology, 43(4), 393–404. 10.1080/1040841X.2016.1179623 [DOI] [PubMed] [Google Scholar]
- Xu, Y. X. , Jiang, Z. D. , Du, X. P. , Zheng, M. J. , Fan‐Yang, Y. , Ni, H. , & Chen, F. (2022). The identification of biotransformation pathways for removing fishy malodor from Bangia fusco‐purpurea using fermentation with Saccharomyces cerevisiae . Food Chemistry, 380, 132103. 10.1016/J.FOODCHEM.2022.132103 [DOI] [PubMed] [Google Scholar]
- Yang, J. , Wang, P. , Wu, W. , Zhao, Y. , Idehen, E. , & Sang, S. (2016). Steroidal saponins in oat bran. Journal of Agricultural and Food Chemistry, 64(7), 1549–1556. 10.1021/acs.jafc.5b06071 [DOI] [PubMed] [Google Scholar]
- Yilmaz‐Ersan, L. , & Topcuoglu, E. (2022). Evaluation of instrumental and sensory measurements using multivariate analysis in probiotic yogurt enriched with almond milk. Journal of Food Science and Technology, 59(1), 133–143. 10.1007/S13197-021-04994-W [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yoon, M. Y. , & Hwang, H. J. (2008). Reduction of soybean oligosaccharides and properties of α‐d‐galactosidase from Lactobacillus curvatus R08 and Leuconostoc mesenteroides JK55. Food Microbiology, 25(6), 815–823. 10.1016/j.fm.2008.04.008 [DOI] [PubMed] [Google Scholar]
- Youssef, H. I. A. (2024). Detection of oxalyl‐CoA decarboxylase (oxc) and formyl‐CoA transferase (frc) genes in novel probiotic isolates capable of oxalate degradation in vitro. Folia Microbiologica, 69(2), 423–432. 10.1007/s12223-024-01128-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yu, A. O. , Goldman, E. A. , Brooks, J. T. , Golomb, B. L. , Yim, I. S. , Gotcheva, V. , Angelov, A. , Kim, E. B. , & Marco, M. L. (2021). Strain diversity of plant‐associated Lactiplantibacillus plantarum . Microbial Biotechnology, 14(5), 1990. 10.1111/1751-7915.13871 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zamudio, M. , González, A. , & Medina, J. A. (2001). Lactobacillus plantarum phytase activity is due to non‐specific acid phosphatase. Letters in Applied Microbiology, 32(3), 181–184. 10.1046/J.1472-765X.2001.00890.X [DOI] [PubMed] [Google Scholar]
- Zartl, B. , Silberbauer, K. , Loeppert, R. , Viernstein, H. , Praznik, W. , & Mueller, M. (2018). Fermentation of non‐digestible raffinose family oligosaccharides and galactomannans by probiotics. Food & Function, 9(3), 1638–1646. 10.1039/C7FO01887H [DOI] [PubMed] [Google Scholar]
- Zhang, C. , Hua, Y. , Li, X. , Kong, X. , & Chen, Y. (2020). Key volatile off‐flavor compounds in peas (Pisum sativum L.) and their relations with the endogenous precursors and enzymes using soybean (Glycine max) as a reference. Food Chemistry, 333, 127469. 10.1016/J.FOODCHEM.2020.127469 [DOI] [PubMed] [Google Scholar]
- Zhang, J. , Lu, L. , Lu, L. , Zhao, Y. , Kang, L. , Pang, X. , Liu, J. , Jiang, T. , Xiao, M. , & Ma, B. (2016). Galactosylation of steroidal saponins by β‐galactosidase from Lactobacillus bulgaricus L3. Glycoconjugate Journal, 33(1), 53–62. 10.1007/s10719-015-9632-4 [DOI] [PubMed] [Google Scholar]
- Zhao, X. , & Gänzle, M. G. (2018). Genetic and phenotypic analysis of carbohydrate metabolism and transport in Lactobacillus reuteri . International Journal of Food Microbiology, 272, 12–21. 10.1016/J.IJFOODMICRO.2018.02.021 [DOI] [PubMed] [Google Scholar]
- Zhong, X. F. , Zhang, Y. B. , Huang, G. D. , Ouyang, Y. Z. , Liao, D. J. , Peng, J. W. , & Huang, W. Z. (2018). Proteomic analysis of stachyose contribution to the growth of Lactobacillus acidophilus CICC22162. Food & Function, 9(5), 2979–2988. 10.1039/C8FO00528A [DOI] [PubMed] [Google Scholar]
- Zuljan, F. A. , Mortera, P. , Alarcón, S. H. , Blancato, V. S. , Espariz, M. , & Magni, C. (2016). Lactic acid bacteria decarboxylation reactions in cheese. International Dairy Journal, 62, 53–62. 10.1016/J.IDAIRYJ.2016.07.007 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Supporting Information
Supporting Information