Effect of Solid-State Fermentation on Plant-Sourced Proteins: A Review

ABSTRACT

The growing interest in plant-based proteins is driven by sustainability concerns and potential adverse reactions to animal protein consumption. Plant proteins, while promising, are known for their inferior digestibility compared to animal counterparts. Solid-state fermentation (SSF) offers an ancient, cost-effective approach to address this limitation. SSF enhances plant-based foods by augmenting protein content through microbial hydrolysis, thereby improving overall nutritional quality. This method not only boosts protein digestibility but also reduces larger polypeptides, generates bioactive peptides, refines amino acid profiles, and eliminates anti-nutritional factors such as trypsin inhibitors and tannins. Further investigation is essential to optimize the purification of protein isolates and examine their behaviour during digestion and absorption using in vitro, cellular, and in vivo models. In conclusion, SSF represents a promising, cost-effective, and high-yield processing method to produce nutritionally enhanced plant proteins, aligning with the demands of both the industry and consumers.

KEYWORDS:

• amino acid profile
• bioactive peptides
• protein hydrolysis
• protein digestibility
• molecular weight
• protein content

Introduction

Fermentation is an ancient food processing technology used to extend the shelf-life of food and enhance its nutritional and organoleptic properties.^[1] Numerous biochemical changes occur during fermentation, leading to altered nutritive and anti-nutritive compositions and affecting product properties such as bioactivity and digestibility.^[2] Submerged (SmF) and solid-state fermentations (SSF) are two types of fermentation processes, where microorganisms are grown in liquid form for SmF but on a solid support for SSF. SSF offers numerous advantages over conventional SmF systems, such as lower water and energy requirements, higher protein productivity and less stringent requirement for sterile medium.^[3]

Recently, there has been a notable surge in the application of fermentation technology in various industries such as food, chemicals and pharmaceuticals.^[4] When considering the application of SSF to improve the nutritional properties of the plant foods such as cereals, grains, and oilseeds examination of protein content is imperative as they represent a substantial source of protein. The fermentation process may also modify the protein’s composition and nutritional profile. Prior studies have established that fermentation leads to improvements in the quality and bioavailability of protein, owing to reducing anti-nutritional factors (ANFs) and the degradation of complex macromolecular proteins to simpler smaller proteins, peptides and free amino acids.^[5]

While a significant body of literature exists on the fermentation of animal-derived proteins, mainly milk products, shifting consumer preferences and health considerations have shifted the focus towards exploring plant-based proteins.^[6] Recent trends highlight the rising demand for plant-based proteins, exemplified by a 44% increase in the consumption of lentils in the USA in 2020 compared to 2019.^[7] Previous reviews on protein fermentation focused on the bioactive peptide derived from the fermentation process and the structure-function relationship of protein.^[7,8] Specific protein sources or fermentation methods, such as plant-derived protein and SSF, were not covered. Sadh et al.^[9] discussed the general influence of SSF on proteins without referring to the biological activity of the peptides and amino acids. Up to date, no comprehensive review has yet addressed SSF of plant proteins and its fermentation products.

This review bridges this gap by summarizing the effect of SSF on various plant proteins, especially on its influence on protein content, molecular weight (MW), degree of hydrolysis, and in vitro digestibility. This paper compares the effects of different starter cultures such as bacteria, mold, yeast, and their co-culture on plant-based protein fermentation, and explores the mechanism of action. We further highlight on the reduction of antinutritional factors related to protein digestibility, changes in amino acid profile, production of peptides and modification of their bioactivity (Figure 1). This review aims to provide insights for future research on SSF of plant-derived proteins and to promote practical application of fermented proteins into the food or feed system with significantly improved nutritional value.

Figure 1. Research on the effect of solid-state fermentation on the plant protein.

Impact of starter cultures on protein content in plant foods during SSF

The protein content of SSF products varies significantly due to several factors. These factors include the type of microorganism, the availability of carbon and nutrients, and the cultivation conditions such as fermentation temperature and time.^[10] Influences of SSF by bacteria, moulds, yeast and their co-culture on the total protein content of the plant substrates are discussed in this section (Table 1).

Table 1. Effect of solid-state fermentation on protein content of selected plant-based substrates.

Substrate	Starter culture	Fermentation conditions	Observations	Reference
	Bacteria
Soybean meal	Bacillus amyloliquefaciens	Inoculum 5% ratio (w/v)37°C for 24 h	Significantly improved the crude protein content from 469.2 mg/g to 551.5 mg/g and trichloroacetic acid-soluble protein from 19.5 mg/g to 213.9 mg/g	^[Citation11]
Moringa oleifera leaf meal	Bacillus pumilus CICC 10,440	50% moisture 30°C for 24 h	Soluble protein increased from 124 mg/g to 285 mg/g	^[Citation12]
Soybean meal	Bacillus subtilis E20	50% initial moisture 40°C for 96h	The crude protein content increased from 422 mg/g to 502 mg/g	^[Citation13]
Soybean meal	Bacillus subtilis N-2	Relative humidity above 50% at 32°C for 12 days.	The water-soluble total nitrogen content in the products increased from 160 mg/g to 9.37 mg/g	^[Citation14]
Soybean meal	Bacillus subtilis SB102	10% ratio with B. subtilis SB102 and incubated for 72 h at 37°C	The crude protein increased from 344.6 mg/g to 374.3 mg/g, increased by 8.61%; the soluble protein content was increased from 202.2 mg/g to 329.8 mg/g, increased by 63.11%	^[Citation15]
Soybean meal	Bacillus subtilis natto	Initial moisture 108% 43.82°C for 62.32 h	Crude protein content increased from 507.2 mg/g to 557.1 mg/g	^[Citation16]
Cottonseed Meal	Bacillus subtilis	Moisture 60% (w/w) 30°C 48 h	Significantly increased the crude protein from 468.6 mg/g to 525.9 mg/g	^[Citation17]
Soybean meal	Bifidobacterium longum CRL 849	65% moisture 37°C for 24 h 7.50 × 10^7 CFU/g of soybean paste	Soluble protein content decreased from 55.2 mg/g to 30 mg/g	^[Citation18]
Jatropha seed cake	One bacterial strain named zxy-12	60% moisture 37°C for 5 days	The crude protein content significantly increased from about 480 mg/g to about 640 mg/g increased by 15.11%	^[Citation19]
Soybean protein meal	Lactobacillus plantarum Lp6	Inoculum size:1×10^7cfu/g 37°C for 66 h	The crude protein content increase from 574.3 mg/g to 601.5 mg/g	^[Citation20]
Maize flour	Lactobacillus plantarum	37°C for 48 h	The crude protein content increased from 90.3 mg/g to 140.6 mg/g	^[Citation21]
Cassava leaf	Lactobacillus plantarum	30°C for 48 h	Crude protein content increased from 177.9 mg/g to 240 mg/g	^[Citation22]
	Fungi
Ginkgo biloba leaves	Aspergillus niger Gyx086	60% moisture 28°C for 48 h	The crude protein content increased from 142.6 mg/g to 152.1 mg/g	^[Citation23]
A mixture of dried vegetable waste powder and oil cake mixture (soybean flour, wheat flour, groundnut oil cake and sesame oil cake at 4:3:2:1 ratio)	Aspergillus niger S14 and Aspergillus niger NCIM 616	30 ± 2°C 9 days at 35% moisture neutral pH	Fermented products using A. niger S14 increased the crude protein content from 206.2 mg/g to maximum 283.8 mg/g in day 8. Fermented products using A. niger NCIM 616 increased the crude protein content from 218.1 mg/g to maximum 283.8 mg/g in day 7.	^[Citation24]
Olive cake	Aspergillus niger, Beauveria bassiana, Fusarium flocciferum, and Rhizodiscina cf. lignyota	Inoculation size: 10^5 spores/g sample 60% moisture 25°C for 15 days	Increased the protein concentration from 42.65 mg/g to 60.03, 62.95, 82.82 and 67.22 mg/g with A. niger, B. bassiana, F. flocciferum, and Rhizodiscina cf. lignyota inoculation, respectively.	^[Citation25]
Peanut press cake	Aspergillus oryzae MTCC 3107	30°C for 6 days	The crude protein content increased from 440.5 to 496 mg/g	^[Citation26]
Flaxseed oil cake	Aspergillus oryzae DSM 1861 and Neurospora intermedia DSM 1265	Aspergillus oryzae DSM 1861: 45% moisture pH value 4–5 30°C for 2–6 days Neurospora intermedia DSM 1265: 60% moisture content, pH 4.5–5.0 30°C for 4–6 days	A. oryzae DSM 1861: increased the protein content from 285.06 ± 6.43 to 317.18 ± 4.98 and 351.28 ± 2.33 mg/g DM after 2- and 6-days fermentation, respectively. N. intermedia slightly increased the protein content from 285.06 ± 6.43 to 296.73 ± 0.50 and 309.88 ± 5.12 mg/g DM after 2- and 6-days fermentation, respectively.	^[Citation27]
Lupin flour	Aspergillus sojae, Aspergillus ficuum	30°C for 7 days	The crude protein content increased from 357 to 360 mg/g by Aspergillus sojae fermentation, decreased from 357.7 to 352.7 mg/g by Aspergillus ficuum fermentation	^[Citation28]
White, red, and black quinoa seeds	Rhizopus oligosporus	10^4 spores/g sample 30 h (2 h at 35°C to induce spore germination and 28 h at 31°C)	The total protein content increased from 142.78, 134.49, and 131.08 to 164.40, 156.18, and 157.84 mg/g in white, red, and black quinoa respectively.	^[Citation29]
Brewers’ spent grains	Rhizopus oligosporus ATCC 64,063	Inoculum size 10^6 spores/g was used and incubated at 37°C for 3 days	Increased the protein content from 252 mg/g to 341 mg/g.	^[Citation30]
Spelt wheat	Rhizopus oligosporus ATCC 64,063	Inoculated 10^4 spores/g of dry grains incubated at 31°C for 30 h	The crude protein content increased from 115 mg/g to 127 mg/g dry matter	^[Citation31]
Rapeseed presscake	Rhizopus microsporus var. oligosporus	32°C and 90–95% relative humidity for 30–48 h	Increased crude protein content from 336 to 357 mg/g	^[Citation32]
Common bean flour	Rhizopus oligosporus NRRL 2710	34.9 C for 51 h	Significantly increased the crude protein content from 230.6 to 280.6 mg/g, significantly increased the soluble protein content from 196.1 to 214 mg/g	^[Citation33]
Babassu mesocarp flour and cassava leaves	Rhizopus microsporus var. oligosporus CCT 3762	24 ± 2°C for 30 days	Increased the crude protein content from 127 mg/g to 173.1 mg/g	^[Citation34]
Palm kernel cake and palm kernel expeller	Rhizopus oryzae ME01	Initial moisture 50% for 10 days	Increased the crude protein content from by 7.64% in palm kernel cake and 3.45% in palm kernel expeller	^[Citation35]
Pineapple peels	Trichoderma viride ATCC 36,316	30°C for 4 days (NH4)2SO4 was used as nitrogen source	The crude and soluble protein contents increased considerably from 45 and 28.4 to 135.6 and 108.6 mg/g, respectively, during the first 72 h of fermentation, but the values declined to 112.1 and 93.4 mg/g respectively when fermentation reached 96 h.	^[Citation36]
Orange peel and sugar beet pulp	Trichoderma reesei and Trichoderma viride	70% moisture, T. reesei at 24°C and T. viride at 26°C for 7 days	Significantly increased the crude protein content in both substrates (P<0.01). Increase in the CP content was greater with T. viride compared with T. reesei (P<0.01).	^[Citation37]
Lentils	Pleurotus ostreatus	28°C for 14 days	Increased the crude protein content from 233 to 276 mg/g dry matter	^[Citation38]
Chia or sesame seed	Pleurotus Ostreatus	28°C for 14 days	The protein content increased from 225 to 256.9 mg/g in chia seeds, 210 to 218 mg/g in sesame seeds, respectively.	^[Citation39]
Mixture of cassava bagasse and leaves	Lentinula edodes	24 ± 2°C for 30 days	Increased the crude protein content from 127 to 173.1 mg/g	^[Citation40]
	Yeast
Wheat bran	Three commercial food grade yeast (Saccharomyces cerevisiae)	Inoculum size 0.1 g yeast/g wheat bran 32°C for 2 days	Three yeasts behaved similarly in increasing crude protein content by 11–12% compared to control.	^[Citation41]
Brown rice flour	Three brands of commercial baker’s yeast (Saccharomyces cerevisiae)	25°C for 12 h	Increased the crude protein content from 77.1 to 88.4, 86.6, and 86.4 mg/g	^[Citation42]
Soybean okara	Saccharomyces cerevisiae r. f. bayanus	28°C for 3 days	The crude protein content increased from 328 to 405 mg/g	^[Citation43]
Guar and copra meal	Saccharomyces cerevisiae	50% moisture at 40°C for 2 days	The crude protein content of guar and copra meal increased from 292.9 to 346.8 mg/g and from 216.41 to 274.02 mg/g, respectively	^[Citation44]
Maize flour	Saccharomyces cerevisiae	37°C for 2 days	The crude protein content increased from 90.3 to 134.4 mg/g	^[Citation21]
Cassava leaf	Saccharomyces cerevisiae	30°C for 2 days	Crude protein content increased from 177.9 to 224.7 mg/g	^[Citation22]
Okara	Yarrowia lipolytica	30°C for 5 days	No significant changes in protein content	^[Citation45]
The mixture of groundnut oil cake and Pistia leaves	Pichia kudriavzevii (GU939629)	35°C for 3 days	The crude protein content increased from 417.3 to 463.7 mg/g dry matter and from 118 to 153 mg/g dry matter in ground nut oil cake and pistia leaf meal, respectively	^[Citation46]
	Co-culture
Drumstick leaf flour	Aspergillus niger GIM 3.576, Candida utilis GIM 1.427 and Bacillus subtilis CICC 31,188	Inoculum proportion = 1:1:2 50% initial moisture at 32°C for 7 days	Increased the crude protein content from 284.2 to 409.8 mg/g	^[Citation10]
Milk thistle fruit	Aspergillus niger ATCC16404 and Candida tropicalis ACCC21145	Inoculum proportion= 1:1 60% initial moisture at 28°C for 3 days	The crude protein content increased from about 250 to 425 mg/g	^[Citation47]
Moringa oleifera leaf meal	Aspergillus niger, Candida utilis and Bacillus subtilis co-cultures	Inoculum proportion = 1:1:2 60% initial moisture at 32°C for 6.5 days	Increased the crude protein content from 211.2 to 301.9 mg/g, increased the soluble protein content from 54.0 to 65.5 mg/g	^[Citation48]
Lupin flour	Aspergillus sojae and Aspergillus ficuum co-cultures	Inoculum proportion= 1:1 30°C for 7 days	Co-culture fermentation increased the crude protein content from 357.7 mg/g to 364 mg/g	^[Citation28]
Soybean meal	Bacillus velezensis 157 and Lactobacillus plantarum BLCC2–0015	40% moisture First stage of fermentation: B. velezensis 157 (8.0 log CFU/g) at 37°C for 24 h Second stage: 8.0 log CFU/g of L. plantarum BLCC2–0015 incubated at 37°C for 48 h under anaerobic conditions	Increased crude protein content from 472.8 to 519.7 mg/g, and trichloroacetic acid-soluble protein from 50.7 to 117.9 mg/g	^[Citation49]
Cottonseed meal	Bacillus subtilis and Saccharomyces cerevisiae	Inoculum proportion = 1:1 60% moisture Inoculum size 2% 30°C for 72 h	The crude protein was increased from 360 mg/g to 405 mg/g	^[Citation50]
Defatted rice bran	Bacillus subtilis, Saccharomyces cerevisiae and Lactiplantibacillus plantarum	Inoculum proportion = 1:1:1 Co-culture (10^7 CFU/g) and phytase (50 U/g) were inoculated in the substrates First-stage aerobic fermentation 37°C for 36 h Second-stage anaerobic fermentation 37°C for 36 h	Increased crude protein from 122.1 to 148.1 mg/g and trichloroacetic acid soluble protein from 15.4 to 47.7 mg/g	^[Citation51]
Cassava leaf	Lactobacillus plantarum and Saccharomyces cerevisiae strains co-culture	Inoculum proportion = 1:1 30°C for 48 h	Crude protein content increased from 177.9 to 218.2 mg/g	^[Citation22]
Maize flour	Lactobacillus plantarum and Saccharomyces cerevisiae cocultures	Inoculum proportion = 1:1 37°C for 48 h	The crude protein content increased from 90.3 to 133.8 mg/g	^[Citation21]
Dehusked barley	Lactobacillus plantarum and Rhizopus oryzae	Inoculum proportion = 3:10 initial moisture 10% 32°C for 36 h	Increased soluble protein from 4.76 mg/g to 9.50 mg/g dry power	^[Citation52]

Various protein content expression formats and assay methods have been used in protein fermentation studies, where the three most prevalent formats are crude protein content, soluble protein content and trichloroacetic acid-soluble protein content. Crude protein content focus on the total nitrogen content within the sample with the application of either the Kjeldahl method or the Dumas combustion method.^[53] On the other hand, soluble protein content gauges the portion of protein which can be dissolved in different solvents (e.g. water or buffers), and can be measured via multiple methods such as Bradford protein assay,^[54] Pierce BCA assay or the Folin-Ciocalteu method (Lowry protein assay),^[55] with bovine serum albumin serving as the standard. The third way to express protein content is trichloroacetic acid-soluble protein content, which employs trichloroacetic acid for protein precipitation, followed by nitrogen content quantification in the supernatant using the Kjeldahl method. This method usually measures the small peptides with 2–20 residues and free amino acids that can be directly absorbed by the gastrointestinal tract.^[51]

Impact of bacterial start culture on protein content

As summarized in Table 1, bacteria such as Bacillus amyloliquefaciens, Bacillus subtilis, and Lactobacillus plantarum increased the crude protein content of soybean meal, cottonseed meal, maize flour, jatropha seed cake, and moringa oleifera leaf meal after fermentation.^{[11–17,19–21]} In these studies, Bacillus subtilis was the most used start culture, and soybean meal was the most common fermented substrate.^[18] However, the addition of Bifidobacterium longum CRL 849 to soybean meal decreased the protein concentration. Variations in fermentation conditions, such as inoculum size, initial moisture, fermentation temperature and time, can also influence the fermentation outcomes.

The augmentation of crude protein content within bacterial Solid-State Fermentation (SSF) processes can be ascribed to two primary factors: the concentration of nitrogen and the generation of enzymes, which inherently comprise proteins. For instance, Zhang et al.^[16] found the increase in crude protein content from 507.2 to 557.1 mg/g during SSF of soybean by Bacillus subtilis. natto is mainly due to the relative change in the loss of fiber and lipid as a result of microbial hydrolysis and the metabolism during fermentation. Likewise, Li et al.^[11] expounded that carbohydrates serve as substrates for bacterial growth and proliferation during the fermentation of soybean meal with B. amyloliquefaciens, consequently elevating the nitrogen concentration and, in turn, the crude protein content. Furthermore, the production of extracellular enzymes, inherently proteinaceous, by bacteria during SSF contributes to the escalation of crude protein content. For example, Bacillus spp. have been found to produce industrially valuable enzymes such as amylases, catalases, lipases, proteases, and xylanase.^[56] Additionally, other bacterial species, such as Lactobacillus plantarum, secretes enzymes during SSF. Terefe et al.^[21] observed a remarkable 55% increase in crude protein content in maize flour through Lactobacillus plantarum SSF as opposed to a 49% increase in Saccharomyces cerevisiae SSF. The superior increase in crude protein content can be attributed to the unique enzymes produced by Lactobacillus plantarum, such as α-amylase, esterase, lipase, α-glucosidase, β-glucosidase, enolase, phosphoketolase, lactase dehydrogenase, etc.^[57,58]

The increase in soluble protein content may be due to the enzymatic degradation of insoluble macromolecular proteins into soluble peptides and amino acids by bacterial enzymes. Su et al.^[51] reported that proteases secreted by B. subtilis BSWF hydrolysed macromolecular proteins (>35 kDa) and resulted in a dramatic increase in soluble protein in solid-state fermented rice bran. Conversely, it is noteworthy that the soluble protein content may exhibit a decrease following SSF. de Olmos et al.^[18] applied Bifidobacterium longum CRL 849 to soybean meal and observed a decrease in protein content after 24 h fermentation. This decrease could be related to the B. longum CRL 849 not being able to hydrolyze the soybean protein and instead consuming free amino acid during the fermentation progress.^[18]

Impact of mould start culture on protein content

SSF with various fungal species, including Aspergillus niger, Beauveria bassiana, Fusarium flocciferum, Rhizodiscina cf. lignyota, Rhizopus oligosporus, Trichoderma viride, Trichoderma reesei, Pleurotus ostreatus, and Lentinula edodes has been conducted with the most prevalent choices being Aspergillus niger and Rhizopus oligosporus.^{[23–34,36–40]} The increased crude protein content has been shown in various plant-based materials such as ginkgo leaves, vegetable waste powder, cake from olive, pressing seeds or grains, quinoa seeds, wheat, common bean flour, mesocarp flour, cassava leaves and bagasse, palm kernel cake, pineapple peels, orange peels, sugar beet pulp, lentils, and chia seeds. The increase in protein content depending on the type of plant material, fungal species, and fermentation conditions such as temperature, moisture, and time.

The increased crude protein content during mould dominated SSF is derived from different sources. Initially, mould may use carbon as energy sources to synthesize protein-containing mycelia during fermentation.^[36] For example, Asensio-Grau et al.^[38] fermented lentils flour with Pleurotus ostreatus and reported a net protein increment of 18.5% with a concurrent 6% decrease for carbohydrates. The growth of Pleurotus ostreatus led to the increase of unicellular protein biomass. Furthermore, the mould-secreted lignocellulases can partially degrade lignocellulose including lignin, cellulose, and hemicellulose and release small organic molecules, such as sugars. These small organic molecules can then be utilized by the fungi as a source of energy and nutrients, leading to the growth of the fungi.^[59] Wang et al.^[23] fermented ginkgo leaves with Aspergillus niger and attributed the relatively higher crude protein content of fermented sample (16.28%) compared to the unfermented one (14.26%) to the same reason. In addition, some mould strains such as Pleurotus ostreatus and Trichoderma species have specific mechanism to influence protein content. The acidophil nature of Pleurotus species and its ability to reduce pH by releasing organic acid could prevent ammoniac volatilisation driven nitrogen losses.^[60] The Trichoderma species has the ability to capture the nitrogen and urea by aerobic fermentation in a mineral-salt medium, which contributes to the significant increase of crude protein content.^[37] Moreover, similar as bacteria, mould such as Rhizopus oligosporus, Aspergillus niger can secrete extracellular enzymes, which are proteins in nature and contribute to higher crude protein content.^[30] Aspergillus niger has been found to produce proteases, pectinase, catalase during fermentation and Rhizopus oligosporus has superior protease and amylase activity.^[30] Finally, Firdaus et al.^[35] found that the size of the substrate impacted crude protein in palm kernel cake fermented by Rhizopus oryzae ME01, where smaller and uniformly sized palm kernels resulted in a higher increase in crude protein, possibly due to the larger surface area of smaller kernels that enhances protein extraction efficiency.

Reduction of crude protein content was also observed in some mould-induced SSF due to prolonged fermentation time. Aruna^[36] found that in the initial 72 hours of fermentation, the crude and soluble protein contents of pineapple peels significantly increased, but then both declined when the fermentation time reached 96 hours. The author suggested that this decline was due to either proteolysis or the consumption of protein by the mould during the later stage of SSF.^[36]

Impact of yeast start culture on protein content

Yeast fermentation has been shown to increase protein content in various food ingredients, such as wheat bran, brown rice flour, soybean okara, guar and copra meal, maize flour, cassava leaf and the mixture of groundnut oil cake and Pistia leaves, with Saccharomyces cerevisiae being the most common starter culture.^{[21,22,41–46]}

The increase in protein content can be attributed to an escalation in yeast biomass.^[41,43] Yeast is frequently applied in fermentation processes due to their facile cultivation, rapid growth rate, and inherently high protein content, making them ideal for producing single-cell proteins from agricultural by-products.^[61] Ilowefah et al.^[42] found that fermenting brown rice flour with three commercial baker’s yeasts led to a protein content increase from 77.1 to 88.4, 86.6, and 86.4 mg/g. Su et al.^[51] reported a similar increase in crude protein levels, which was due to the decrease in dry matter (mainly carbohydrates) and the elevation in proteins produced by S. cerevisiae. In contrast, Vong et al.^[45] reported an absence of significant change in protein content of okara fermented by Yarrowia lipolytica, despite an increase in total free amino acids. The lack of significant change in protein content may be due to the Kjeldahl method conducted, which measures the total nitrogen content of a sample without distinguishing between protein and free amino acids.

Impact of multiple start culture on protein content

Table 1 also summarizes the results of using multi-strain cultures in SSF with various plant-based substrates, including drumstick leaf flour, milk thistle fruit, Moringa oleifera leaf meal, lupin flour, soybean meal, cottonseed meal, defatted rice bran, cassava leaf, maize flour, and dehusked barley.^{[10,21,22,28,47–52,62]}

Previous studies have indicated that several fungal, yeast and bacterial strains can grow synergistically. Wang et al.^[52] fermented de-husked barley with a combination of Lactobacillus plantarum and Rhizopus oryzae and observed a significant increase in soluble protein content. During SSF, Lactobacillus species secrete lactic acid, lowering the pH value of the entire system. The dominant protease, aspartic protease, secreted by Rhizopus exhibited improved activity and stability in acidic environments, further contributing to the heightened soluble protein content.^[63] In a separate study fermenting Moringa oleifera leaf meal by Aspergillus niger, Candida utilis and Bacillus subtilis co-cultures, the effective fermentation is attributed in part to greater cellulase, pectinase, and amylase activities of A. niger.^[48] This facilitates the degradation of cellulose and starch to support the growth of both C. utilis and B. subtilis. Moreover, the bacteria convert non-protein nitrogen into microbial protein and secrete enzymes that accelerate cellulose and starch degradation, thereby promoting the mycelial growth of A. niger.

Recent research also suggests that several biochemical reactions may be hindered if only a single microorganism strain is used for SSF. Co-culture fermentation is required for optimal results as all available substrates can be used in a multi-strain fermentation process.^[64,65] For example, Olukomaiya et al.^[28] found that co-culture fermentation of lupin flour with Aspergillus sojae and Aspergillus ficuum co-culture resulted in the greatest increase in crude protein content than single culture fermentation. In contrast, several studies have reported a reduction in crude protein content during co-culture SSF. For instance, Terefe et al.^[22] found that instead of co-culture SSF, Lactobacillus plantarum fermentation resulted in the largest increase in crude protein content when fermented with cassava leaves. Likewise, Terefe et al.^[21] observed that the increase in protein contents of maize flour fermented solely by L. plantarum is higher than L. plantarum and S. cerevisiae coculture-fermentation. This phenomenon is not yet fully understood, but it may be linked to the increased consumption of nitrogen-containing substances like proteins, peptides, or amino acids to support the growth and metabolism of microorganisms. Further research is needed to elucidate this mechanism.

As discussed above, the selection of the starting culture is critical to the substrate protein levels. Taking soybeans as an example, Bacillus amyloliquefaciens, Bacillus subtilis, Lactobacillus plantarum, Saccharomyces cerevisiae r. f. bayanus, and Bacillus velezensis 157/Lactobacillus plantarum BLCC2–0015 co-culture have been shown to increase protein levels. However, it is noteworthy that each of these strains elicits varying increments in protein content, whereas the utilization of Bifidobacterium longum leads to a decrease in soluble protein levels. Fermentation conditions such as temperature, initial moisture, and time also play important role. Moreover, it is worth emphasizing that the majority of studies in this domain have commonly employed protein content as the primary metric to assess the nutritional value of the fermented product (Table 1). While this choice is certainly justifiable when the primary objective is to achieve an elevated protein level in the end product. However, when the focus is on the production of hydrolyzed products such as peptides and amino acids and their consequent impact on nutritional or bioactivity value, it is advisable to consider the degree of protein hydrolysis as a more informative parameter for analysis.

Overall, SSF is a complex microorganism-guided process, influenced by several factors such as timing of inoculation, fermentation time and temperature. The growth of microorganism could further be influenced by the plant substrates, making it challenging to compare different microorganisms in their capacity to alter plant protein content. Most studies only observed the effects of SSF on the protein content of substrates without follow-up experiments to explore the mechanisms. For example, the increase in soluble protein content may be due to hydrolytic enzymes produced by microorganisms, whereas the specific enzyme(s) responsible for this action was not well studied. Furthermore, the influence of co-culture fermentation on protein content and their mechanism of action is rarely studied. Future study may investigate the interaction between fungi, yeast, and bacteria during co-culture fermentation.

Degree of protein hydrolysis and molecular weight of protein after SSF

During SSF, the degree of protein hydrolysis and molecular weight (MW) of plant-based food can be influenced by various factors such as the choice of microorganisms, fermentation conditions, and the composition of the substrate. The degree of protein hydrolysis refers to the extent to which the protein molecules are broken down into smaller peptides and amino acids,^[66] while the MW of the protein molecules may undergo change as a result of the fermentation process. It is well known that both the degree of protein hydrolysis and MW have a significant impact on the bioavailability and functional properties of the fermented product.^[51] Consequently, understanding the factors that influence these parameters during SSF is crucial for optimizing the quality of plant-based food and ensuring its maximum nutritional value.

Degree of hydrolysis of plant protein after SSF

The degree of hydrolysis (DH) measures the extent of peptide bonds cleaved in the protein substrate by a proteolytic agent: the higher the value, the higher number of amino groups released. Proteolytic enzymes, primarily endogenous enzymes, hydrolyze large and intact proteins into small polypeptides and intermediate peptides.^[67] Microorganisms such as lactic acid bacteria, yeasts, and mould further catalyze the hydrolysis of intermediate peptides into small peptides and amino acids.^[67] This process significantly impacts the nutritional values and health-promoting benefits of plant food products, as most small peptides and free amino acids can be directly absorbed by the gastrointestinal tract into the body.^[51] The o-phthalaldehyde (OPA) and trinitrobenzenesulphonic acid (TNBS) amino acid determination methods are commonly employed to assess DH with results being expressed as percentage of hydrolysis.^[66]

The alteration of DH in plant-based food after SSF is shown in Table 2. Research has shown that the DH can vary greatly depending on the type of substrate employed. Lupin and quinoa, for instance, resulting in higher DH compared to wheat when fermented by Lactobacillus spp.^[68] This was attributed to the greater proteolytic activity of endogenous proteases present in lupine and quinoa, and/or the higher susceptibility of lupine and quinoa proteins to the proteolytic enzymes produced by Lactobacillus spp. Strain selection is also critical, as distinct strains of Lactobacillus spp can produce varying levels of proteolytic enzymes and exert differential effects on the DH.^[68,73] Cereals fermented by Lactobacillus reuteri K×88177 were observed to have higher DH than those by Lactobacillus plantarum DSM and Lactobacillus plantarum KX881779.^[68] Furthermore, co-culture fermentation has been shown to be beneficial for enhancing DH due to the diverse microbial enzymes present, thereby accelerating the rate of protein proteolysis.^[63,74] It is also noteworthy that the activities of proteinases and peptidases are time-dependent, with prolonged fermentation time is beneficial to the enhancement of DH.^[69,73]

Table 2. Effect of solid-state fermentation on the degree of hydrolysis and protein molecular weight profile of selected plant-based substrates.

Substrate	Starter culture	Fermentation conditions	Observations	Reference
			Degree of hydrolysis
Lupin, quinoa and wheat	Lactobacillus reuteri KX88177, Lactobacillus plantarum DSM and Lactobacillus plantarum KX881779	35°C for 72h	Fermented lupin, quinoa and wheat by three Lactobacillus spp. (Lactobacillus reuteri KX88177, Lactobacillus plantarum DSM and Lactobacillus plantarum KX881779) and reported that the DH% was increased from ∼8.5% to 35%, ∼11.5% to 30%, and ∼2.5% to 20%, respectively	^[Citation68]
Whole-grain oats	Rhizopus oryzae. or a combination of Lactobacillus plantarum B1–6 and Rhizopus oryzae.	Inoculum size: 0.5% (w/w) R. oryzae starter and 1.5% (v/w) L. plantarum 90% relative humidity 30°C for 72 h	Co-culture fermentation resulted in significantly higher DH than R. oryzae sole fermentation at all fermentation times after 24 h.	^[Citation63]
Grass pea seeds	R. microsporus var. chinensis and A. oryzae	Inoculum size: 10^4 spores/g R. microsporus var. chinensis and 10^4 spores/g A. oryzae 30°C for 32 h	The DH of Rhizopus microsporus var. chinensis and Aspergillus oryzae on grass pea seeds proteins were 24.3% and 7.6% at 32 hours fermentation, which were higher than the 15.6% and 1.5% at 24 h fermentation	^[Citation69]
Brewers’ spent grains	Rhizopus oligosporus ATCC 64,063	Inoculum size: 10^6 spores/g 37°C for 3 days	After 3 days of fermentation with R. oligosporus, proteins in brewers’ spent grains were hydrolysed to a degree of 14.6%, which is comparable to the highest DH (13%) obtained from an enzymatic hydrolysis using alcalase	^[Citation30]
			Molecular weight profile
Whole oat flour	Lactobacillus plantarum TK9 and Bifidobacterium animalis subsp. lactis V9.	L. plantarum TK9: 37°C for 28 h Bif. animalis subsp. lactis V9: 37°C 40 h	The levels of peptides with MW larger than 10 kDa in fermented oats decreased by 26.08 and 22.93% after fermentation by L. plantarum TK9 and Bif. animalis subsp. lactis V9, respectively, while the levels of peptides with MW less than 6 kDa increased by 4.44% (L. plantarum TK9) and 5.96% (Bif. animalis subsp. lactis V9)	^[Citation70]
Dehusked barley	Lactobacillus plantarum and Rhizopus oryzae co-culture	Inoculum size: 0.5% (w/w) of R. oryzae starter, 0.15% (w/w) of L. plantarum starter Initial moisture 10% 32°C for 36 h	Higher-molecular weight barley proteins, especially 18.4–25 kDa, were degraded to the low-molecular weight proteins, oligopeptides, and even amino acids; Proteins with MW >25 kDa gradually increased in fermented barley as the fermentation process progressed	^[Citation52]
Rapeseed meal	B. subtilis YY-4 and L. plantarum CICC6026	Acid protease: 0.53 mg/g initial moisture, 46% 37°C for 5 days	Most of the proteins were degraded to soluble peptides with a MW of less than 9.5 kDa, and small peptides with a MW of less than 1 kDa.	^[Citation71]
Lupin flour	Aspergillus sojae, Aspergillus ficuum and their co-cultures	30°C for 7 days	The soluble proteins in lupin flour degraded into fragments with a MW smaller than 20 kDa.	^[Citation28]
Whole soybean	Bifidobacterium animalis 937 and Bacillus subtilis natto co-culture	37°C for 48 h	Low MW peptides less than 1 kDa in the whole soybean fermented by B. animalis 937 and B. subtilis natto combination decreased by 4.6% after 24 h fermentation	^[Citation72]
Drumstick leaf flour	Aspergillus niger, Candida utilis and Bacillus subtilis	50% initial moisture 32°C for 7 days	From the third day of fermentation, proteins with molecular weight of 60–140 kDa appeared	^[Citation10]
Soybean protein meal	Lactobacillus plantarum Lp6	Inoculum size:1×10^7cfu/g 37°C for 66 h	Before and after fermentation, a similar reduced intensity of the polypeptide bands in the range of 14.4–97.4 kDa shown	^[Citation20]

One advantage of implementing SSF for protein rich foods is that it can easily achieve the same DH as commercial enzymes with being cost-effective, albeit at the expense of longer incubation times.^[30] Chin et al.^[30] reported that after 3 days of SSF with R. oligosporus, proteins in brewers’ spent grains were hydrolysed to a degree of 14.6%, which is comparable to the highest DH of 13% obtained from an enzymatic hydrolysis using alcalase.^[75] The DH of the proteins can be further demonstrated from the change of the protein MW profile, as discussed below.

Changes in protein MW profile of plant protein after SSF

The MW represents another important parameter reflecting the solubility and bioactivity of proteins. Proteases and peptidases secreted by microorganisms during fermentation breakdown proteins into smaller fragments, thus affecting the distribution of their molecular mass.^[76] This alteration in molecular mass can be determined by sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and gel filtration chromatography.^[77] Table 2 summarizes the studies that examine changes in the MW distribution of plant protein after SSF.

SSF typically induce the degradation of macromolecular proteins, yielding increased quantities of smaller proteins and peptides.^{[28,52,70,71]} Smaller MW peptide and amino acids are related to enhanced antioxidant and antihypertensive activities of fermented protein-rich materials.^[78] Additionally, reducing molecular sizes increases the digestibility of protein, improving both the nutritional value and flavor of fermented foods.^[10,52] However, the level of low MW protein can decrease during certain SSF conditions. For instance, Zhang et al.^[72] reported a 4.6% decrease in the concentration of low MW peptides (<1 kDa) in soybeans fermented by a B. animalis 937 and B. subtilis natto co-culture for 24 hours. The reduction may be due to the poor protease activity at the initial stage of the fermentation, where the soybean protein was only partly hydrolyzed and most peptides were utilized by B. animalis.^[72] In other cases, higher MW proteins may be produced in SSF. Shi et al.^[10] studied the fermentation of Moringa oleifera leaf meal by Candida utilis. In this study, proteins with MW of 60–140 kDa appeared from the third day, which could be due to the production of yeast proteins as the yeast cells proliferate. Similarly, Wang et al.^[52] observed an increase in proteins with MW > 25 kDa in fermented dehusked barley by Lactobacillus plantarum and Rhizopus oryzae, which may also be caused by the increased microbial metabolic activity. However, The effect of SSF on protein MW can sometimes be limited. Amadou et al.^[20] reported the influence of SSF by Lactobacillus plantarum Lp6 on the high MW polypeptides in soybean meal protein was restricted.

The effect of SSF on the MW changes is dependent on the substrate and fermentation time. For example, in the study of Sun et al.,^[77] Bacillus subtilis BJ-1 fermented cottonseed meal contained 56.52% low MW peptides (<1 kDa), which is comparable to a value of 46% for similar MW peptides in the fermented soybean extracts,^[79] but notably smaller than the 86.89% content observed in fermented peanut.^[80] Furthermore, Zhang et al.^[72] reported that the level of soybean low MW peptides (<1 kDa) decreased by 4.6% after 24 h of co-culture fermentation with B. animalis 937 and B. subtilis natto, but increased by 30.7% at 48 h. This indicates that fermentation time has contributed to the variation of protein molecular profile.

So far, numerous studies have explored the impact of SSF on DH and MW distribution. Future research could investigate the optimal fermentation conditions for enhancing the activity of endogenous and microbial proteases and peptidases to boost the DH of specific plant substrates. Additionally, it may be valuable to compare the DH achieved by SSF with that obtained through alternative techniques, such as enzymatic or chemical hydrolysis. This would allow developing the most cost-effective and sustainable method for industrial application. Furthermore, future studies may explore the relationship between protein MW distribution and physical properties, such as solubility, emulsification, foaming, and gelation, which could provide insight into the impact of SSF on protein functionality and offer values for practical applications.

Effect of SSF on the in vitro protein digestibility and antinutritional factors (ANF)

Plant proteins are widely recognized to possess a lower nutritional value to animal proteins due to their reduced digestibility and lower content of essential amino acids.^[81] During SSF, the enhancement of DH and the reduction of molecular weight MW of proteins have a beneficial impact on their digestibility. Digestibility is an important indicator to evaluate the nutritional value of proteins, which is depended on internal factors such as the amino acid profile and protein structure, as well as external factors such as the presence of antinutritional factors (ANF).^[82,83] Enhancing the digestibility of plant proteins reduces the levels of undigested protein that cause gastrointestinal distress and protein excretion in the stool, and eliminate gastro-intestinal immune reactions from immunogenic peptides.^[7,84]

Modification of plant protein digestibility by SSF

Several studies have reported that SSF can increase protein digestibility of plant products (Table 3). In simulating the human digestive process, in vitro pepsin-trypsin digestion is commonly employed to assess protein digestibility.^[89,92] Research has revealed that the inclusion of bacteria in SSF imparts an overall improvement in protein digestibility. For example, Bacillus amyloliquefaciens, Bacillus subtilis, Lactobacillus plantarum, Pediococcus acidilactici, Pediococcus pentosaceus, and Lactobacillus sakei applied for SSF can significantly increase the protein digestibility of soybean meal, cottonseed meal, soy seeds, lupin seeds, and rapeseed meal.^{[11,17,71,85,87,88]} The improvement in protein digestibility can be attributed to several factors. Firstly, bacterial proteolysis partially degrades complex proteins in the substrate into simple and soluble products such as peptides and amino acids, consequently bolstering digestibility.^[51] Secondly, cellulase secreted by bacteria in the substrates degrades cellulose, exposing starch granules or cell wall proteins to digestion.^[93] Thirdly, some bacteria such as lactic acid bacteria can produce organic acids, lowering the pH value of the fermented product and enhancing the activity of pepsin, leading to improved protein digestion.^[94] Finally, the rise in protein digestibility is linked to reduction of anti-nutritional factors (ANF). These ANF include inhibitors that inhibit digestive enzymes like trypsin and chymotrypsin, as well as phenolics and tannins which exacerbate the crosslinking of proteins.^[7]

Table 3. Effect of solid-state fermentation on the protein digestibility of selected plant-based substrates.

Substrate	Starter culture	Fermentation conditions	Observations	Reference
	Bacteria
Soybean meal	Bacillus amyloliquefaciens	37°C for 24 h	In vitro digestibility increased significantly from 62.91% to 72.52%	^[Citation11]
Cottonseed Meal	Bacillus subtilis	60% initial moisture 30°C for 48 h	Significantly increased in vitro protein digestibility (p<0.05)	^[Citation17]
Soy seeds	Lactobacillus plantarum B1–6	2 g/100 g sucrose 37°C for 17.61 h	In vitro protein digestibility improved significantly from 79.8% to 83.9%	^[Citation85]
Lupine wholemeal and protein isolates	Pediococcus acidilactici	CO2 atmosphere 32 ± 2°C for 48 h	in vitro protein digestibility of nonfermented and SSF samples was not significantly in lupine wholemeal and the protein isolates displayed 7.7% and 11.2% higher in vitro protein digestibility	^[Citation86]
Lupin seeds	Lactobacillus sakei KTU05–6, Pediococcus acidilactici KTU05–7 and Pediococcus pentosaceus KTU05–8	30°C for L. sakei, 32°C for P. acidilactici and 35°C for P. pentosaceus for 24 h	In comparison with the non-fermented samples, digestibility increased 17.12%, 16.73% and 18.87% in one species of lupine seed and 12.15%, 14.71% and 17.68% in the other species of lupin seed fermented with P. acidilactici, L. sakei and P. pentosaceus, respectively	^[Citation87]
Lupin and soya bean	Lactobacillus sakei KTU05–6, Pediococcus acidilactici KTU05–7 and Pediococcus pentosaceus KTU05–8	30°C for L. sakei, 32°C for P. acidilactici and 35°C for P. pentosaceus for 24 h	Protein digestibility of fermented lupin and soya bean was found higher on average by 18.3% and 15.9%, respectively, compared to untreated samples.	^[Citation88]
Rapeseed meal	B. subtilis YY-4 and L. plantarum CICC6026	Acid protease: 0.53 mg/g initial moisture, 46% 37°C for 5 days	Significantly increased protein digestibility (p = 0.012)	^[Citation71]
	Fungi
Soya bean	Aspergillus oryzae	80% relative humidity 25°C and in the dark for several days	Fermented protein showed higher in vitro digestibility due to partial protein degradation.	^[Citation89]
Flaxseed oil cake	Aspergillus oryzae (Koji Mold) and Neurospora intermedia (Oncom Mold)	Aspergillus oryzae DSM 1861 (koji strain): 45% initial moisture, pH value 4–5 30°C for 2–6 days Neurospora intermedia DSM 1265 (oncom strain): 60% initial moisture, pH 4.5–5.0 30°C for 4–6 days	In the case of N. intermedia, the standard fermentation procedure (3 days) caused a 48% increase in protein digestibility. The fermentation with A. oryzae for 4 and 6 days resulted in a significant improvement of the protein digestibility, by 24% and 42%, respectively.	^[Citation27]
De-oiled rice bran	Rhizopus oryzae NCIM 1009	50% initial moisture room temperature for 3 days	A reduction of 16.5% in protein digestibility	^[Citation90]
Common bean flour	Rhizopus oligosporus NRRL 2710	34.9°C for 51 h	Improved the in vitro protein digestibility	^[Citation33]
Lentils	Pleurotus ostreatus	28°C for 14 days	After simulating in vitro digestion, fermented flours presented a higher fraction of digested protein (17%)	^[Citation38]
Red bean flour	Cordyceps militaris (L.) Fr.	25°C for 7 days	Significantly improved the in vitro protein digestibility by 6.54%	^[Citation91]
	Co-culture
Lupin flour	Aspergillus sojae, Aspergillus ficuum and their co-cultures	30°C for 7 days	The IVPD decreased from 44.74 ± 0.32% to 34.25 ± 0.26%, 37.58 ± 0.34% and 30.22 ± 0.53% for Aspergillus sojae, Aspergillus ficuum, and co-culture fermented lupin flour, respectively.	^[Citation28]
Moringa oleifera leaf meal	Aspergillus niger, Candida utilis and Bacillus subtilis co-cultures	Inoculum proportion = 1:1:2 initial moisture 60% 32°C for 6.5 days	The in vitro protein digestibility was improved significantly from 42.86 ± 0.23% to 65.05 ± 0.89%	^[Citation48]
Maize flour	Lactobacillus plantarum, Saccharomyces cerevisiae and their cocultures	Inoculated with 7 ml of 1 × 10^6 cells/ml of L. plantarum and S. cerevisiae strains each and 3.5 ml of 1 × 10^6 cells/ml each for cocultures 37°C for 48 h	In vitro protein digestibility improved by 31%, 40%, 36%, and 34% for natural, Lactobacillus plantarum, Saccharomyces cerevisiae, and their coculture-fermented maize flour, respectively.	^[Citation21]
Cassava leaf	Lactobacillus plantarum, Saccharomyces cerevisiae strains, and their co-culture	Inoculated with 7 ml of 1 × 10^6 cells/ml of L. plantarum and S. cerevisiae strains each and 3.5 ml of 1 × 10^6 cells/ml each for cocultures 30°C for 48 h	The mean score of in vitro digestibility increased from 66.58% to 82.34%, 85.27%, 84.61%, and 83.42% for natural, Lp, and Sc and co-cultures fermented cassava leaf flours, respectively.	^[Citation22]
Grass pea seeds	Rhizopus microsporus var. chinensis and Aspergillus. oryzae	Inoculum size: 10^4 spores/g R. microsporus var. chinensis and 10^4 spores/g A. oryzae 30°C for 32 h	Tempeh made with co-culture had higher in vitro bioavailability (77.15%) than Rhizopus pure culture fermented (69.39%)	^[Citation69]
Industrial solid waste from milk thistle fruit	Aspergillus niger ATCC16404 and Candida tropicalis ACCC21145	Initial moisture 59.7% 30.8°C for 87.0 h	The digestibility was 88.92%, which was increased by 16.22%	^[Citation47]

Nevertheless, there exist instances where no discernible improvement in protein digestibility is evident. For example, there was no significant difference in the protein digestibility of lupine wholemeal fermented with Pediococcus acidilactici, contrasting with the notable increase in protein digestibility in lupine protein isolates.^[86] This divergence may due to the presence of polyphenols in the lupin wholemeal, which binds to proteins and decreases their digestibility.^[86] Polyphenols can also directly impact protein digestibility by inhibiting proteases in the digestive tract after food intake.^[95] It is important to note, however, that certain studies, the protein digestibility of lupin seeds was shown to increase after fermentation with Pediococcus acidilactici in the studies by Krunglevičiute et al.^[87] and Bartkiene et al..^[88]

Table 3 also summarizes the results of several studies applied fungi starter culture. The food substrates encompassed a diverse array of plant-based materials, including soya bean, flaxseed oil cake, de-oiled rice bran, common bean flour, lentils, and red bean flour. The fungi species applied in the studies were Aspergillus oryzae, Neurospora intermedia, Rhizopus oryzae, Rhizopus oligosporus, Pleurotus ostreatus, and Cordyceps militaris.^{[27,33,38,89–91]} The conditions for the fermentation process, such as temperature, moisture content, and pH, exhibited variations across each of these studies, reflecting the context-dependent nature of SSF and its influence on protein digestibility.

SSF with fungi has demonstrated its capacity to enhance protein digestibility, as indicated by several studies.^{[27,33,38,89,91]} Similar to bacteria, the improvement in protein in vitro digestibility by fungal SSF is due to the reduced levels of anti-nutritive compounds as well as partial pre-digestion of the substrate by fungal proteases.^[96,97] Flaxseed oil cake shown increased digestibility when fermented with both Aspergillus oryzae and Neurospora intermedia. However, with longer fermentation time, the product fermented with Aspergillus oryzae showed higher digestibility while that fermented with Neurospora intermedia did not change significantly.^[27] The discrepancy was explained by changes in the proteolytic activity of different fungal mycelium, which is strongly dependent on the culture age.

Nevertheless, certain studies have reported a reduction in protein digestibility following fungal SSF. For instance, lupin flour subjected to SSF with Aspergillus sojae, Aspergillus ficuum and their co-cultures showed lower protein digestibility.^[28] This reduction was attributed to proteins being locked within the lupin fibre matrix, thus, reducing accessibility by proteases. The authors also stated that partial protein denaturation during drying process might have also lowered the protein dispersibility and solubility, thereby resulting in a reduced protein digestibility, noted a physical change in processing prior to digestion that should be wary about. This result is aligned with the findings by Ranjan et al.^[90] that the SSF with R. oryzae has negative effects on the protein digestibility of dry defatted rice bran.

The effect of co-fermentation on protein digestibility is presented in Table 3 as well. Although co-culture SSF generally improves the protein digestibility of the substrates, it is noteworthy that the achieved increase is often less pronounced than that observed in single-strain SSF. In the study of Terefe et al.,^[21] the protein digestibility of maize flour was increased significantly after 48 h of SSF, while the highest increase was observed when employing the Lactobacillus plantarum, followed by the Saccharomyces cerevisiae, with co-culture fermentation showing the least improvement. Furthermore, in a study involving the co-fermentation of grass pea seeds with Rhizopus microsporus var. chinensis and Aspergillus oryzae, it was noted that protein digestibility exhibited improvement only when the inoculum size of A. oryzae was less than that of R. microsporus var. chinensis.^[69] This observation suggests that within co-culture fermentation, competition among microorganisms may inhibit the growth of those with a stronger capacity to enhance protein digestibility. Consequently, co-culture fermentation may not be as effective as individual fermentation in optimizing protein digestibility.

Elimination of ANF in plant food after SSF

ANF refers to chemical entities in plant products that negatively affect the digestibility of proteins or other macronutrients such as starch, hence their bioavailability.^[98] The reduction of ANF during SSF can indirectly increases the accessibility of proteins by enzymes and this in turn increases the protein digestibility.^[21] Tannins and trypsin inhibitors represent key ANFs linked to reduced protein digestibility.

Tannins represent a diverse group of phenolic compounds known for their ability to interact with proteins, thereby altering protein structure and functionality while impeding the digestibility.^[99] Several studies have shown the reduction of tannins by SSF. For instance, the tannin content was reduced significantly from 49.84% to 12.3% in maize flour fermented by co-culture of Lactobacillus plantarum and Saccharomyces cerevisiae.^[21] This reduction in tannin content can be attributed to the activities of tannases and mono- and di-oxygenases produced by the microbes during the SSF, which hydrolyze tannin-protein complexes, including both hydrolysable and condensed tannins.^[100,101] Similar result was found by Dileep et al.^[44] where the tannin content was reduced by 29.74% and 20.00% in Saccharomyces cerevisiae fermented guar and copra meal, respectively, with concurrent increase in protein digestibility.

Trypsin inhibitors are compounds that impede protein digestion by inhibiting the activity of trypsin.^[102] They form a stable complex with trypsin, leading to its inactivation and thereby reducing the availability of protein and amino acids.^[16] Multiple studies have highlighted the capacity of SSF to diminish trypsin inhibitor levels in plant-based foods. Teng et al.^[15] observed that trypsin inhibitor content in fermented soybean meal with B. subtilis SB102 and A. oryzae AO3042 was reduced by 95.33% and 81.33%, respectively, after 72 h fermentation. Similarly, Zhang et al.^[16] reported decrease in trypsin inhibitor content by 53.7% in Bacillus subtilis natto fermented soybean. Dileep et al.^[44] also reported that trypsin inhibitor activity was decreased by 56.81% and 58.02% in Saccharomyces cerevisiae fermented guar and copra meal, respectively.

The reduction of antigenic proteins through SSF is not directly linked to protein digestibility; however, it can significantly enhance digestion by mitigating gastrointestinal discomfort caused by these proteins. Antigenic proteins are known to trigger the body’s adaptive immune response, and in soybeans, glycinin (11S) and β-conglycinin (7S) are the two primary antigenic proteins, constituting 70–80% of the total seed protein.^[103] Previous studies showed that soybean meal after SSF could enhance intestinal barrier function, weaken undesirable intestinal mucosal immune response, and reduce the incidence of diarrhea of animals, which was related to the reduction of soybean meal antigenic proteins.^[104,105] For example, Li et al.^[11] reported that 92.32% of glycinin and 85.05% of β-conglycinin in soybean meal were removed by Bacillus amyloliquefaciens fermentation for 24 h. Similarly, Zhang et al.^[106] found that 84.77% of glycinin and 87.07% of β-conglycinin were degraded after 24 h fermentation with B. subtilis BS12. However, another B. subtilis strain KCCM11438P eliminated only 42% of glycinin and 70% of β-conglycinin after 24 h fermentation.^[107] Compared to B. subtilis, Bacillus amyloliquefaciens’s superior ability to decompose antigenic protein is attributed to its stronger ability to secrete several extracellular hydrolases including serine proteases, acid proteases, endoglucanases, acetylxylan and esterases. These hydrolases promote the breakdown of cell wall polymers and degradation of antigenic protein of soybean meal.^[11]

Effect of SSF on the peptide bioactivity of plant protein

Compared to the non-fermented counterparts, fermented foods show multiple benefits to human health such as reducing the risks of a number of chronic diseases including cardiovascular disease, arthritis, Type 2 diabetes, periodontitis, and bladder disorders.^[108] The bioactivities of fermented plant products by bacterial and fungi can be attributed to bioactive small peptides.^[109] This section discusses the production of bioactive peptides produced by SSF and their biological activities such as antioxidant, antihypertensive, antidiabetic, antiproliferative and immunomodulatory effects. Selected studies on the influence of SSF on the quantity and bioactivity of the peptides from plant substrates are shown in Table 4.

Table 4. Selective studies on production of bioactive peptides via solid-state fermentation of plant-based substrates.

			Observations
Substrate	Starter culture	Fermentation conditions	Amount	Bioactivity	Reference
Jatropha seed cake	One bacterial strain named zxy-12	Initial moisture 60% 37°C for 5 days	Small peptides were increased 19.61% w/w	-	^[Citation19]
Black and red quinoa seeds	Rhizopus oligosporus	Inoculum size: 10^4 spores/g of seeds 31°C for 4 days	The amount of total peptides was significantly increased 0.23 ± 0.01 g/100g to 3.48 ± 0.07 g/100g in black quinoa and 0.23 ± 0.02 to 1.93 ± 0.13 g/100 in red quinoa	Antioxidant	^[Citation110]
Hull less pumpkin seeds cake	Rhizopus oligosporus ATCC 64,063, DSM 1964 and NRRL 2710 and Rhizopus oryzae CBS 372.63	Inoculum size: 3 × 10^6 spores per 100 g of raw substrate Initial moisture 60% 50% air humidity pH 4–5 37°C the first 2 h followed by 30°C for 48, 72 and 96 h	The level of peptides increased gradually during 96 h of incubation. The most dynamic accumulation of peptides was observed in the case of the ATCC 64,063 strain. While the R. orzae is the weakest	Antioxidant	^[Citation62]
Rapeseed meal	Bacillus subtilis	85 ± 5% relative humidity 32°C for 6 days	-	Antioxidant	^[Citation111]
Rapeseed meal	B. subtilis YY-4 and L. plantarum CICC6026	Acid protease: 0.53 mg/g initial moisture, 46% 37°C for 5 days	The concentrations of soluble peptides significantly increased after fermentation (p < 0.001), from 53.33 to 231.05 ± 5.6 mg/g	Antioxidant	^[Citation71]
Drumstick leaf flour	Aspergillus niger, Candida utilis and Bacillus subtilis	50% initial moisture 32°C for 7 days	The small peptide concentration increased from 5.72 to 12.03%	Antioxidant	^[Citation10]
Cottonseed meal	Bacillus subtilis BJ-1	1.0 g/kg of papain 50% initial moisture 70% relative humidity 30°C for 48 h	After 48 h of fermentation, the peptide content reached the maximum values of 128.6 mg g^−1 from about 20 mg g^−	Antioxidant	^[Citation77]
Soya bean	Aspergillus oryzae	80% relative humidity 25°C in the dark for several days	-	Antioxidant	^[Citation89]
Okara	Yarrowia lipolytica	30°C for 5 days	-	Antioxidant	^[Citation45]
Dehusked barley	Lactobacillus plantarum and Rhizopus oryzae co-culture	Inoculum proportion = 3:10 initial moisture 10% 32°C for 36 h	The initial content of <10 kDa peptides was 34.47 ± 5.78 mg CTE/g dry power, which was reduced to 29.8 ± 1.09 mg CTE/g at 6 hr of fermentation. Then the production of peptides increased to reach a maximum value of 338.01 ± 3.98 mg CTE/g at 36 hr.	Antioxidant	^[Citation52]
Corn gluten-wheat bran mixture	Lactobacillus delbrueckii subsp.bulgaricus ATCC BAA-365 and Lactobacillus acidophilus ACCC10637	5 g/kg of acid protease Inoculum size: 5 × 10^9 CFU/g Initial moisture: 66% of LAB 36°C for 96 h	Small peptide increased from 3.14 ± 0.54% to 12.94 ± 0.89%	Antioxidant	^[Citation112]
Soybean	Bacillus subtilis	37°C for 24 h	-	Antihypertensive
Whole-grain oats	Rhizopus oryzae or a Lactobacillus plantarum B1–6 and Rhizopus oryzae.co-culture	Inoculum size: 0.5% (w/w) R. oryzae starter and 1.5% (v/w) L. plantarum 90% relative humidity 30°C for 72 h	The peptide contents of Rhizopus oryzae solely fermentation and co-culture fermentation increased by 9.72% and 15.13% compared with that of non-fermented oats (0 h).	Antihypertensive	^[Citation63]
Red bean flour	Cordyceps militaris (L.) Fr.	25°C for 7 days	An increment in the amount of small-sized proteins or peptides(< 15 kDa)	Antihypertensive	^[Citation91]
Pigeon pea	Aspergillus niger KR535626	27°C for 7 days.	-	Antihypertensive	^[Citation113]
Rapeseed (Brassica campestris L.)	Bacillus subtilis and Acrogenotheca elegans	Neutral protease: 200 U/g weight of dry rapeseed meal pH 6.5 35°C for 2 days	The peptide content were 1.63 times the original rapeseed meal	Antiproliferative	^[Citation114]
Lupin, quinoa and wheat	Lactobacillus reuteri KX88177, Lactobacillus plantarum DSM and Lactobacillus plantarum KX881779	35°C for 72h	-	Antioxidant Antihypertensive Antidiabetic Antiproliferative	^[Citation68]
Cottonseed meal	Bacillus subtilis: Saccharomyces cerevisiae co-culture	Inoculum proportion= 1:1 60% initial moisture 30°C for 72 h	Compared with the 2.78% of oligopeptide yield from the raw cottonseed meal, the oligopeptide yield increased to 24.47% in the fermented cottonseed meal	Immunomodulatory	^[Citation50]

Antioxidant peptides from plant protein produced by SSF

While relatively limited research has been conducted on the extraction and purification of antioxidant peptides from solid-state fermented products, some studies have indicated that the antioxidant properties of these products are related to the increase of peptide contents. Substrates such as quinoa seeds, rapeseed, drumstick leaf, cottonseed, soybean, okara, barley and mixture of corn gluten and wheat bran have been subjected to SSF to produce antioxidant peptides.^{[10,52,62,71,77,110,112]}

The antioxidant peptides in plant material are encrypted within their protein sequence, which are released by microbial enzymes during SSF, especially in the case of small peptides with a molecular weight of less than 3 kDa, which have been shown to have higher bioactivity than larger peptides.^[115,116] For example, Wang et al.^[71] reported that SSF of rapeseed meal by B. subtilis YY-4 and L. plantarum CICC6026 did not cause a significant change in total phenolic content, but it significantly increased DPPH radical scavenging, Fe²⁺ chelation, and reducing power with an elevated peptide content, implying the generation of antioxidant peptides during fermentation. Similarly, SSF significantly improved the antioxidant capacity of okara, which was attributed to the generation of peptides through yeast proteolysis.^[45] Furthermore, Starzyńska-Janiszewska et al.^[62] fermented pumpkin seeds cake with R. oligosporus and observed a gradual increase in peptide generation over 96 h fermentation, and the content of peptides was significantly correlated with the level of ABTS˙⁺-scavenging activity. The rapeseed peptide produced by SSF with B. subtilis 10160^[111] and B. subtilis YY-4 and L. plantarum CICC6026^[71] also showed an increase in radical scavenging activity, as well as in cottonseed fermented with B. subtilis.^[77] In addition, cotton seed fermentation produced peptides showed protective effect against H₂O₂-induced oxidant death of raw 264.7 macrophage cells in in-vitro culture at 2.5 mg/mL of protein.^[77]

However, the increased antioxidant activity of SSF products may also result from other factors. One such factor is the deglycosylation of phenolic glycosides during fermentation, leading to the release of more phenolic aglycone, thus augmenting antioxidant activity, as seen in fermented soybean with A. oryzae.^[89] Furthermore, the release of polyphenols from protein complexes can also contribute to the increased antioxidant capacity of fermented products.^[89] Additionally. Certain amino acids such as tyrosine, cysteine and methionine have been found to contribute to antioxidant activity. For instance, the increases in antioxidant capacity of the fermented rapeseed meal were correlated with increases in these amino acids. (Wang et al., 2022)

Antihypertensive peptides produced by SSF

Antihypertensive peptides is of great research interests and the majority of the reported antihypertensive peptides from food are products of enzymatic hydrolysis.^[117] Angiotensin-converting enzyme (ACE) plays an important role in blood pressure regulation in the body. It converts the inactive decapeptide angiotensin I to the active angiotensin II, which induces increased blood pressure through narrowing of blood vessels.^[118] Synthetic ACE inhibitors have been shown to lower blood pressure in in vivo clinical trials, and their development as antihypertensive drugs play an important role in promoting cardiovascular health.^[118] However, antihypertensive drugs have side effects such as fatigue, dizziness and headaches, and these side-effects are exacerbated by their requirement for lifetime use. Consequently, there has been increasing interest in natural sources of ACE-inhibiting peptides that are considered safer with less side effects.^[119] ACE inhibitory activity is measured in vitro to assess the antihypertensive properties of foods. During SSF, microbial proteases can release bioactive peptides with ACE inhibitory activity, potentially serving as antihypertensive agents.^[120]

The amino acid composition within the peptide sequence plays a crucial role in determining its activity. For a potent ACE inhibitory peptide, the most favorable N-terminus residues would be branched amino acids such as valine and isoleucine.^[121] Nawaz et al.^[113] identified a competitive ACE inhibitory octapeptide Val-Val-Ser-Leu-Ser-Ile-Pro-Arg with a molecular mass of 869.53 Da from pigeon pea fermented with A. niger (KR535626). Larger active peptides containing these octapeptide sequence were shown to be less inhibitory, which could be due to weaker binding of the peptide to the active site restricted by the extra amino acids.^[122]

Solid-state co-fermentation involving R. oryzae and L. plantarum B1–6 demonstrated a significant enhancement in ACE inhibitory activity in oat proteins. Over 24 and 72 hours of fermentation, the ACE inhibitory activity increased from 9.90% to 59.45% and 77.81%, respectively.^[63] Comparing to the co-fermentation system, extracts from R. oryzae single fermentation showed slightly lower ACE inhibitory activity (51.31% and 63.64% in 24 and 72 h, respectively). The better results obtained from co-fermentation is most likely due to the presence of multiple microorganisms that facilitate greater hydrolysis of proteins. In the study of Wu et al.,^[63] the IC₅₀ values (half maximal inhibitory concentration) of co-fermentation product was significantly reduced to 0.42 mg protein/mL compared to unfermented samples at 1.26 mg protein/mL. This value is similar to the IC₅₀ of the enzymatic hydrolysates of lentils protein (0.44 mg protein/mL)^[123] and red bean (0.63 mg protein/mL),^[91] but lower than that of fermented soybean meal (1.38 mg protein/mL).^[124] The observed differences in the ACE inhibitory activity may be attributed to a variety of causes, such as different proteins, protein extraction procedures, peptide mixture compositions, fermentation or digestion parameters (strain, fermentation method, time, temperature, pH, enzyme, substrate/enzyme ratio, etc.), or analytical methods used for the determination of ACE inhibitory activity.^[125]

In addition, Ayyash et al.^[68] found that lupin and quinoa subjected to SSF by Lactobacillus spp. showed higher ACE-inhibition and antioxidant activity than the control (without SSF). However, the authors cautioned that the bioactive fragments possessing the antioxidant activities are not necessarily the same as those that manifested the ACE-inhibition.^[68] Future studies should focus on identifying and purifying the ACE-inhibitory peptides in fermented plant, which could serve as a precursor for further research and potential clinical trials.

Antiproliferative peptides from plant protein produced by SSF

Indeed, research into anticancer peptides produced by SSF is quite limited, as reflected in Table 4. To date, only two studies have shown that SSF products possess antiproliferative activity. Xie et al.^[114] demonstrated that peptide extracts from solid-state fermented rapeseed inhibited the proliferation of human HepG2 cells, human HeLa cervical cancer cells, and human MCF-7 breast cancer cells. Additionally, they found that the rapeseed antitumor peptide participated in mitochondria-mediated apoptosis, leading to the inhibition of human HepG2 hepatoma cell proliferation.^[126] Another study by Ayyash et al.^[68] reported that lupin fermented by Lactobacillus strains exhibited greater antiproliferative activities than fermented quinoa and wheat against Caco-2 colon cancer and MCF-7 breast cancer cell lines. The author attributed the superior antiproliferative activity of fermented lupins to the higher quantity and lower MW of protein hydrolysates during fermentation. Nevertheless, none of these studies isolated nor identified the sequence of the peptides contributing to the nominated functions, which warrants further investigation.

Antidiabetic and immunomodulatory peptides from plant protein produced by SSF

Antidiabetic compounds that inhibit α-amylase and α-glucosidase activities are valuable for managing postprandial hyperglycemia and have potential applications for diabetic treatment.^[127] However, there is limited literature on the application of SSF for the generation of antidiabetic peptides from plant proteins. Ayyash et al.^[68] observed a sharp increase in α-glucosidase inhibition after 24 h fermentation of lupin, quinoa and wheat, which was attributed to the production α-glucosidase inhibitory peptides during fermentation.

For immunomodulatory capability, Liu et al.^[50] demonstrated that oligopeptides derived from SSF of cottonseed meal significantly affect the immunomodulatory ability of BALB/c mice treated with cyclophosphamide. The authors suggested that treatment with the oligopeptides protected the mice immune organs by reversing the immunosuppression caused by cyclophosphamide treatment. Also, the oligopeptide significantly increased the levels of IgG and IgM in cyclophosphamide-treated immunosuppressed mice, which indicated that the intake of oligopeptide could enhance humoral immunity effectively.^[50]

At present, the research on the production of bioactive peptides through SSF of plant-derived proteins is still limited, with a predominant focus on the antioxidant and anti-hypertensive activities. Most of the existing literature does not report on the separation, purification and identification of the bioactive peptides; instead, thy rather establish a correlation between the concentration of peptides and the increased biological activities. This highlights a big research gap in the area and further research is warranted. In addition, in vitro digestion and animal studies are needed to demonstrate the retention of these bioactive activities through processes of absorption and digestion. Such endeavours would significantly contribute to our understanding of the potential applications and benefits of SSF-derived bioactive peptides from plant sources.

The effect of SSF on amino acid profile of plant-based food

Adequate intake of dietary amino acids is essential for human health.^[128] Thus, the content of total amino acids (TAA), free amino acids (FAA), essential amino acids (EAA), and non-essential amino acids (NEAA) in a food matrix has a profound effect on its nutritional value. TAA analysis reveals the overall amino acid content, encompassing those integrated into proteins and those existing as unbound units, while FAA analysis quantifies the quantity of amino acids that are not part of a protein structure.^[129] Plant proteins, although capable of providing EAA, often present an imbalanced amino acid profile and a restricted range of amino acids when compared to animal proteins.^[87] Animal proteins, for example, often contain higher proportions of the EAA leucine, an amino acid required for efficient protein synthesis.^[130] Therefore, it is necessary to apply suitable food processing methods to improve availability of specific amino acids originated from plant proteins, and SSF has been shown to be an effective method, with selective studies summarised in Table 5.

Table 5. Effect of solid-state fermentation on the amino acid profile of plant sourced protein.

Substrate	Starter culture	Fermentation conditions	Observations		Reference
Total amino acid (TAA)
Soybean meal	Bacillus amyloliquefaciens	37°C for 24 h	Significantly increased TAAs by 6.57%		^[Citation11]
Soybean meal	Bacillus velezensis 157 and Lactobacillus plantarum BLCC2–0015	40% moisture First stage: B. velezensis 157 (8.0 log CFU/g) for 24 h at 37°C second stage: 8.0 log CFU/g of L. plantarum BLCC2–0015incubated for 48 h under anaerobic conditions at 37°C	TAA content increased by 5.05%, while the difference of AA content between the 24 h, 48 h and 72 h of fermentation was not significant (P > 0.05).		^[Citation49]
Soybean meal	Bifidobacterium longum CRL 849	65% initial moisture 37°C for 24 h	TAA decreased by 33.2% with Lys showed little difference		^[Citation18]
Soybean protein meal	Lactobacillus plantarum Lp6	Inoculum size:10^7 cfu/g 72 h at 37°C	The results showed a significant (p<0.001) decrease in TAA		^[Citation131]
Unhulled common bean seeds	Lactobacillus plantarum DSM 20,174 and Rhizopus microsporus var.chinensis co-cultivation	Inoculum proportion = 1:1 30°C for 31 h	Did not improve the amino acid profile		^[Citation132]
Cottonseed Meal	Bacillus subtilis	Initial moisture 60% 30°C 48 h	The TAA content increased (p<0.05) by 13.4%.		^[Citation17]
Red bean flour	Cordyceps militaris (L.) Fr.	25°C for 7 days	TAA significantly enhanced 13.09%		^[Citation91]
Jatropha seed cake	One bacterial strain named zxy-12	60% initial moisture 37°C for 5 days	The proportion of TAA increased from 16.516% in unfermented JSC to 29.82% in fermented JSC,		^[Citation19]
Rapeseed meal	Aspergillus niger	30°C for 48 h	TAA content was significantly increased after fermentation, the content of six essential AAs (Ile, Leu, Lys, Met, Phe, and Thr) was increased (P < 0.05), whereas that of Arg, His, and Val was not significantly changed (P < 0.05).		^[Citation133]
Drumstick leaf flour	Aspergillus niger GIM 3.576, Candida utilis GIM 1.427 and Bacillus subtilis CICC 31,188	Inoculum proportion= 1:1:2 50% initial moisture 32°C for 7 days	TAA content was 24.88%, higher than that of the unfermented substrate.		^[Citation10]
Ginkgo biloba leaves	Aspergillus niger Gyx086	60% initial moisture 28°C for 48 h.	The amount TAA was 95.07 mg/g, which was 16.49% higher than the control		^[Citation23]
Moringa oleifera leaf	Aspergillus niger	28°C for 5 days.	TAA increased from 201.0 mg/g CP to 283.9 mg/g CP		^[Citation134]
Moringa oleifera leaf meal	Aspergillus niger, Candida utilis and Bacillus subtilis co-cultures	Inoculum proportion = 1:1:2 60% initial moisture 32°C for 6.5 days	TAA were significantly increased by about 16.42% relative to unfermented one		^[Citation48]
Free amino acid
Soybean protein meal	Lactobacillus plantarum Lp6	Inoculum size:10^7 cfu/g 72 h at 37°C	FAA of fermented soybean meals increased significantly from 0.2523 to 0.7702 g/100 g sample.		^[Citation131]
Soybean meal	Bacillus subtilis E20	50% initial moisture 40°C for 96h	FAA content increased by 374.9% while hydroxyproline, asparagine, and β-alanine, were observed to decrease		^[Citation13]
Soybean meal	Bacillus subtilis N-2	Relative humidity above 50% 32°C for 12 days	The fermentation product was enriched with 16 kinds of FAAs.		^[Citation14]
Soybean meal	Bacillus subtilis natto	Initial moisture 108% 43.82°C for 62.32 h	FAA content was 12.76 times higher than that of SBM.		^[Citation16]
Soybean protein meal	Lactobacillus plantarum Lp6	Inoculum size:1×10^7cfu/g 37°C for 66 h	FAA content of fermented soybean meals increased significantly from 0.32 to 8.03 g/100 g protein.		^[Citation20]
Rapeseed meal	Bacillus subtilis YY-4 and Lactobacillus plantarum CICC6026	Acid protease: 0.53 mg/g initial moisture, 46% 37°C for 5 days	The FAAs significantly increased (p < 0.001) from 3630.80 to 7914.31 ng/g DM		^[Citation71]
White, red, and black quinoa seeds	Rhizopus oligosporus	Inoculum size: 10^4 spores/g of material 2 h at 35°C to induce spore germination and 28 h at 31°C	FAA increased form 0.29, 0.22, 0.22 g/kg DM to 2.57, 1.20, 1.24 g/kg DM in white, red, and black quinoa, respectively.		^[Citation29]
Okara	Yarrowia lipolytica	30°C for 5 days	Fermentation increased the amounts of FAA concentration by about 4-fold. Amino acid-derived volatiles, such as 3-methylbutanal and 2-phenylethanol, were also produced.		^[Citation45]
Moringa oleifera leaf	Aspergillus niger	28°C for 5 days	No significant change in FAA		^[Citation134]
Ginkgo biloba leaves	Aspergillus niger Gyx086	60% initial moisture 28°C for 48 h.	The amount of FAA was 14.77 mg/g, which was 96.41% higher than the control, respectively.		^[Citation23]
Ground nut oil cake and Pistia leaves	Pichia kudriavzevii (GU939629)	Ground nut oil cake: 60% initial moisture content, pH 6.5, 34°C, 0.4% (w/v) tannic acid supplementation, 0.5% (w/v) glucose supplementation. Pistia leaf flour.: 50% initial moisture content, pH 6.5, 32°C, 0.6% (w/v) tannic acid supplementation, 0.75% (w/v) glucose supplementation.	FAA increased from 0.85% to 1.14% in ground nut oil cake, from 0.35% to 0.68% in pistia leaf flour.		^[Citation46]
Essential (EAA) and non-essential amino acid (NEAA)
Cracked soybean seed	Rhizopus oryzae	90% relative humidity 30°C for 48 h	Among NEAA, only Ala increased significantly; among EAA only Thr increased significantly		^[Citation135]
Cracked soybean seed	Bacillus subtills	90% relative humidity 30°C for 48 h	Improved the content of most of EAA		^[Citation135]
Cracked soybean seed	Aspergillus oryzae	90% relative humidity 30°C for 48 h	Improved the content of most of EAA		^[Citation135]
Soybean meal	Bacillus subtilis natto	Initial moisture 108% 43.82°C for 62.32 h	EAA increased by 12.2%, higher than the 10.8% increase of NEAA		^[Citation16]
Soybean meal	Bacillus subtilis SB102	37°C for 72 h	The contents of lysine, threonine and arginine were increased by 6.02%, 34.55% and 50.67%, respectively.		^[Citation15]
Soybean meal	Aspergillus oryzae AO3042	37°C for 72 h	The contents of lysine, threonine and arginine were increased by 6.02%, 16.36% and 64.0%, respectively.		^[Citation15]
Soybean meal	Bacillus subtilis E20	50% initial moisture 40°C for 96h	Except for methionine and arginine, the increase of EAA and NEAA acids was above 10%.		^[Citation13]
Soybean meal	Bacillus amyloliquefaciens	37°C for 24 h	EAAs including valine, methionine, phenylalanine, and histidine increased 10–17%		^[Citation11]
Lupine wholemeal and protein isolates	Pediococcus acidilactici	32 ± 2°C for 48 h at CO2 atmosphere	Increased all the EAA content in wholemeal except lysine, decreased all the EAA content in protein isolate except Val, Thr, and Lys. Increased all the NEAA in wholemeal except Glu and Tyr, increased all the NEAA in protein isolate instead of Tyr.		^[Citation86]
Lupin wholemeal	Lactobacillus sakei KTU05–6	45% initial moisture 30°C for 48 h in anaerobic condition	The results indicate that in most cases (approx. 50% of samples) the SSF reduces the NEAA contents		^[Citation136]
Lupin seeds (Lupinus luteus L. and Lupinus albus L.)	Lactobacillus sakei KTU05–6, Pediococcus acidilactici KTU05–7 and Pediococcus pentosaceus KTU05–8	30°C for L. sakei, 32°C for P. acidilactici and 35°C forP. pentosaceus for 24 h	Leucine content increased was in all the fermented L. luteus samples, and a decrease in all the fermented L. albus samples. Lysine and histidine increased in all the fermented L. luteus samples; however, in all L. albus samples, lysine decreased.		^[Citation87]
Drumstick leaf flour	Aspergillus niger GIM 3.576, Candida utilis GIM 1.427 and Bacillus subtilis CICC 31,188	Inoculum size = 1:1:2 50% initial moisture 32°C for 7 days	The levels of most amino acids increased, whereas glutamic acid and lysine decreased. The concentration of EAA and NEAA in the fermented substrate also increased by 1.81 and 1.76%, respectively.		^[Citation10]
Rapeseed meal	Aspergillus niger	30°C for 48 h	The content of six EAAs (Ile, Leu, Lys, Met, Phe, and Thr) was increased (P < 0.05), whereas that of Arg, His, and Val was not significantly changed (P < 0.05).		^[Citation133]
Rapeseed (Brassica campestris L.)	Bacillus subtilis and Aspergillus elegans	Neutral protease: 200 U/g weight of dry rapeseed meal pH 6.5 35°C for 2 days	All amino acids increased accordingly, except Pro, the content of essential amino acids improved by 3.06% compared with the content before fermentation		^[Citation114]
Ginkgo biloba leaves	Aspergillus niger Gyx086	60% initial moisture 28°C for 48 h.	Almost all the NEAA content was improved		^[Citation23]
Jatropha seed cake	One bacterial strain named zxy-12	60% initial moisture 37°C for 5 days	The total proportion of EAA increased from 6.263% to 12.951% after SSF.		^[Citation19]
Cottonseed Meal	Bacillus subtilis	Initial moisture 60% 30°C 48 h	The content of EAAs, namely lysine, threonine, leucine, arginine, histidine, and phenylalanine, as well as all NEAAs except for proline, showed an increase.		^[Citation17]
Red bean flour	Cordyceps militaris (L.) Fr.	25°C for 7 days	All EAA significantly higher than control except for threonine. EAA significantly enhanced 13.88%		^[Citation91]
Unhulled common bean seeds	Lactobacillus plantarum DSM 20,174 and Rhizopus microsporus var. chinensis co-cultivation	Inoculum proportion = 1:1 30°C for 31 h	Decrease the most of amino acid content except of Met and Cys		^[Citation132]
Brewers’ spent grains	Rhizopus oligosporus ATCC 64,063	37°C for 3 days	Increase in some of EAA was seen (valine, methionine, lysine, histidine)		^[Citation30]
Common bean flour	Rhizopus oligosporus NRRL 2710	34.9 C for 51 h	The content of His, Ile, total sulfur (Met Cys), total aromatic (Phe Tyr), Thr, and Val increased (p<0.05) in 0.17, 0.24, 0.23, 0.84, 0.26 and 0.16 g/100 g protein, respectively. However, Leu, Lys and Trp levels decreased (p<0.05) 0.03, 0.21 and 0.09 g/100 g protein, respectively,		^[Citation33]
Cassava bagasse and leaves	Lentinula edodes	24 ± 2°C for 30 days	The profile of EAA showed high concentrations of histidine, isoleucine, valine, methionine and phenylalanine.		^[Citation40]
A mixture of dried vegetable waste powder and oil cake mixture (soybean flour, wheat flour, groundnut oil cake and sesame oil cake at 4:3:2:1 ratio)	Aspergillus niger S14 and Aspergillus niger NCIM 616	30 ± 2°C for 9 days at neutral pH	Significant increase in the concentration of EAA, except methionine and phenylalanine in the case of A. niger S14 and arginine and methionine for A. niger 616. Among NEAA, aspartic acid, serine and alanine increased in the fermented product using A. niger S14 and aspartic acid, serine, glycine and tyrosine got increased while using A. niger 616 but reduction in other non-essential amino acids		^[Citation24]
Black and red quinoa seeds	Rhizopus oligosporus	Inoculum size: 10^4 spores/g of seeds 31°C for 4 days	The black quinoa seeds significantly enriched with Thr + Ser (62 times), Ile (109 times), Tyr + Phe (95 times), Ala (72 times), His (38 times) and Lys (26 times) after fermentation		^[Citation110]
Guar and copra meal	Saccharomyces cerevisiae	50% initial moisture 40°C for 48 h	The EAA content of fermented guar meal showed a 42% increment from the raw guar meal, and the EAA of copra meal showed a 25% increment from the raw copra meal. Lysine content in guar meal and copra meal after fermentation increased 21% and 116%, respectively.		^[Citation44]
Defatted rice bran	Bacillus subtilis, Saccharomyces cerevisiae and Lactiplantibacillus plantarum co-cultivation	Inoculum proportion = 1:1:1 Co-culture (10^7 CFU/g) and phytase (50 U/g) were inoculated in the substrates. First-stage aerobic fermentation 37°C for 36 h Second-stage anaerobic fermentation 37°C for 36 h	Pyroglutamic acid, threonine, l-alanine, leucine, proline, methionine, beta-alanine and glycine, were obviously increased during second-stage anaerobic fermentation. In addition, aspartic acid, quinic acid and tryptophan were decreased.		^[Citation51]
Industrial solid waste from milk thistle fruit	Aspergillus niger ATCC16404 and Candida tropicalis ACCC21145 co-culture	Inoculum proportion= 1:1 60% initial moisture content 28°C for 3 days	The ratio of total EAA to TAA was significantly increased comparing with those of the substrates without SSF.		^[Citation47]
Wheat bran	Saccharomyces cerevisiae, Lactobacillus plantarum, and their co-culture	28°C for 18 h	Lactobacillus plantarum solely fermentation lead to higher total AAs and EAAs but lower EAA/AA. Saccharomyces cerevisiae solely fermentation showed both lower EAAs and AAs. Co-fermentation product had the highest EAA/AA ratio.		^[Citation137]
Pure teff (Eragrostis teff), pure grass-pea (Lathyrus sativus), and their mixtures, 9:1 and 8:2 (teff/grass-pea)	Aspergillus oryzae fermentation first and Rhizopus oligosporus in succession	35°C for 48h with Aspergillus oryzae fermentation, followed by 35°C for 48h with Rhizopus oligosporus fermentation	Improved the EAA profiles of teff and teff/grass-pea (9:1 and 8:2) mixtures.		^[Citation138]
Corn gluten-wheat bran mixture	Lactobacillus delbrueckii subsp.bulgaricus ATCC BAA-365 and Lactobacillus acidophilus ACCC10637	5 g/kg of acid protease Inoculum size: 5 × 10^9 CFU/g Initial moisture: 66% of LAB 36°C for 96 h	Lys content was higher (1.14% vs. 0.68%)		^[Citation112]
Sulphur-containing and aromatic amino acids
Cracked soybean seed	Lactobacillus plantarum	90% relative humidity 30°C for 48 h	Increases in sulfur amino acids (Met plus Cys), Phe, Tyr, Lys, and Thr.		^[Citation135]
Cassava bagasse and leaves	Lentinula edodes	24 ± 2°C for 30 days	The profile of EAA showed high concentrations of histidine, isoleucine, valine, methionine and phenylalanine. The sulfur containing amino acids (Met and Cys), which could reach 83.7% (in EAA mg g^−1 of protein) of the daily necessity requirements in the fermented material		^[Citation40]
Milk thistle fruit	Aspergillus niger ATCC16404 and Candida tropicalis ACCC21145	Inoculum proportion= 1:1 60% initial moisture 28°C for 3 days	Sulfur-containing amino acids (methionine and cystine), and total aromatic amino acids (phenylalanine and tyrosine) of FF were significantly increased comparing with those of the substrates without SSF.		^[Citation47]

The effect of SSF on total amino acid of plant-based food

In most cases, SSF lead to an increment in TAA content.^{[10,11,17,19,23,49,91,133,134]} The most prevalent substrate choice is soybean, however, with different choice of starter culture, the result can be different. Bacillus amyloliquefaciens, Bacillus velezensis 157 and Lactobacillus plantarum BLCC2–0015 as starter culture tend to increase TAA content, whereas Bifidobacterium longum CRL 849 and Lactobacillus plantarum Lp6 tend to decrease TAA content.^{[11,18,49,131]} In addition, with same substrate moringa oleifera leaf, using Aspergillus niger solely for SSF resulted in a greater increase in TAA content than co-culture of Aspergillus niger, Candida utilis and Bacillus subtilis, although fermentation conditions such as fermentation time may also play a significant role.^[48,134] Prolonged fermentation time may lead to the consumption of amino acids by microbes.^[45]

The observed Increase in TAA content resulting from SSF can be attributed to the biosynthesis of amino acids by microorganisms utilizing carbohydrates within the substrate via the glycerate-3P and pyruvate pathway TCA cycle (oxaloacetate and α-ketoglutarate).^[51] However, there were also instances where the TAA content was reduced or remained unchanged. de Olmos et al.^[18] fermented soybean meal with Bifidobacterium longum CRL 849 and found that total amino acid was decreased by 33.2%. The authors attributed this decrease to nutrient losses during fermentation due to leaching, destruction by light, heat or oxygen, or from microbial utilization.^[139] Starzynska-Janiszewska et al.^[132] reported that the amino acid profile remained unchanged when unhulled common bean was fermented with Rhizopus var. chinensis and Lactobacillus plantarum. However, this may be an artifact of the short fermentation time (31 h compared with 96 h in other studies) of the study and the use of whole, unhulled common beans rather than crushed or ground beans, where the nutrients are largely locked in the whole bean matrix.

The effect of SSF on free amino acid of plant-based food

In addition to TAA content, changes in free amino acid (FAA) composition have also been discussed by several researchers.^{[13,14,16,20,23,29,45,71,131,134]} Except for one study which found no change in FAA content during the fermentation of moringa oleifera leaf with Aspergillus niger, all other studies have shown that SSF increased the TAA content of final products. FAA and peptides play a crucial role in producing taste and flavour in food, and among them, FAA are more favourable for human absorption.^[23,140] In addition, improved digestibility is tie to the process of protein hydrolysis into FAA.^[45]

The increase in FAA content may result from the breakdown of large proteins into smaller amino acids through protease hydrolysis by microorganism. Zhang et al.^[16] attributed the significant increase in FAA in soybean meal to the proteolytic activity of B. subtilis natto. Wang et al.^[71] indicated that the extracellular protease hydrolytic activity of Lactobacillus plantarum can be a contributor for the increased FAA in fermented rapeseed meal, while Vong et al.^[45] stated that the increase of FAA in okara fermented by Yarrowia lipolytica was due to the superior proteolytic ability of extracellular acidic and alkaline proteases produced by the yeast.

Although amino acid degradation and deamidation may also occur during fermentation, the rate of protein hydrolysis is more likely to exceed the rate of amino acid catabolism resulting in a net increase in FAA.^[45] There are also studies discussing the decline of specific FAAs. For example, Shiu et al.^[13] applied B. subtilis to ferment soybean meal and found that most of the free amino acids in the fermented soybean meal were increased, while some of them, such as hydroxyproline, asparagine, and β-alanine, were decreased. The reduction in individual amino acid levels after fermentation can be linked to metabolism of specific amino acid by B. subtilis.^[141]

The effect of SSF of plant-based food on essential and non-essential amino acids

The quality of the protein is directly related to protein digestibility and EAA contents, which contributes to the bioavailability.^[40] Elevated consumption of EAA has health benefits such as enhancing the level of lean body mass, muscle strength, and physical function in elderly.^[142] During SSF, the conversion of amino acids by the action of transaminases could partly contribute to the increase of certain EAA or NEAA,^[33,96] and fermented foods enriched in EAA represent a valuable dietary source.^[86] The impact of SSF on EAA and NEAA has been summarized in Table 5.

The effect of SSF on EAA and NEAA profile of the same substrate varies depending on the starter culture employed. Frias et al.^[135] conducted SSF of B. subtilis, A. oryzae, R. oryzae on cracked soybean seed under the same fermentation condition. They found that R. oryzae fermentation only significantly increased the content of non-essential alanine (Ala) and essential threonine (Thr), while the other microorganisms significantly increased the levels of almost all amino acids. This discrepancies may be attributable to the proteinase profiles and proteinase secretion abilities of different microorganisms.^[143] Furthermore, even identical starter cultures may produce varying effects when applied to different cultivars of plants within the same species. Krunglevičiute et al.^[87] employed Lactobacillus sakei KTU05–6, Pediococcus acidilactici KTU05–7, and Pediococcus pentosaceus KTU05–8 for SSF on two species of lupin seed (Lupinus luteus L. and Lupinus albus L.) and observed an increase in leucine content in all fermented L. luteus samples, but a decrease in all fermented L. albus samples. The same study further identified increased lysine and histidine levels in all fermented L. luteus samples but decreased lysin in all L. albus samples. These results underscore the complexity of SSF outcomes, which are influenced by plant cultivar.

Selection of an appropriate starter culture is critical to achieve the target EAA/NEAA content. For example, level of methionine, a limiting EEAs present in the soybean, was not significantly changed by SSF using B. subtilis E20.^[13] However, an increase in methionine was reported when soybean was fermented using Bacillus subtilis natto^[16] and Bacillus amyloliquefaciens.^[11] Likewise, lysine, a limiting EAA in cereals^[144] was significantly reduced in common bean flour when fermented using Rhizopus oligosporus,^[33] and in drumstick leaf flour fermented with Aspergillus niger GIM 3.576, Candida utilis GIM 1.427 and Bacillus subtilis CICC 31,188 co-culture.^[10] However, a significant increase (1.14% vs. 0.68%) was reported by Jiang et al.^[112] using Lactobacillus delbrueckii subsp. bulgaricus to ferment corn gluten-wheat bran mixture. In contrast, lysine amount remained unaltered after fermentation of soybean meal with Bifidobacterium longum.^[18] These examples highlight how the amino acid composition of fermented plant proteins can be influenced by both the specific starter culture and the nature of the substrate. Researchers must carefully select the combination of microorganism and substrate to achieve their desired target for improving amino acid content in the final product.

It is promising to see that the EAA content in plant proteins can meet the requirements for WHO^[145] human nutrition after biological treatments like fermentation. For instance, cottonseed subjected to SSF by B. subtilis contained 296.2 mg/g protein of EAAs, which is close to the amino acid requirements for human adult nutrition (277 mg/g protein).^[77] Wang et al.^[134] also stated that the individual EAA contents in fermented moringa oleifera leaf were higher than the FAO reference protein and comparable to those in soybeans. The protein qualities of fermented teff/grass-pea mixtures have similar EAA composition as the recommended level for 2 − 5 year old children.^[138]

The content of NEAAs is also important in formulating balanced diets to support overall body growth and health.^[146] Several studies have shown SSF partially or fully increase all NEAAs.^{[10,16,17,86]} However, Rajesh et al.^[24] fermented a mixture of dried vegetable waste powder and oil cake mixture with Aspergillus niger and revealed an increase in the levels of certain NEAA including aspartic acid, serine, alanine, glycine, and tyrosine, but a decrease for other NEAA such as arginine, cysteine, glutamine, and proline. This outcome may be attributed to the utilization of these amino acids for the production of enzymes and other organic compounds by the fungal strain.^[147]

The effect of SSF on sulphur-containing and aromatic amino acids of plant-based food

Sulfur-containing amino acid, cysteine and methionine, are important for sulfur homeostasis and various cellular processes in the body.^[148] Morales et al.^[40] reported that cysteine and methionine (20.84 mg EAA/g of protein) in fermented cassava bagasse and leaves could reach 83.7% of daily requirements (25 mg EAA/g of protein) according to FAO/WHO pattern. Similarly, the total sulphur-containing amino acids of fermented milk thistle fruit with Aspergillus niger ATCC16404 and Candida tropicalis ACCC21145 co-culture were significantly increased compared to that without fermentation.^[47]

Aromatic amino acids, including tyrosine, phenylalanine, and tryptophan, also play important roles in various physiological processes. Tyrosine is essential for synthesizing important compounds like epinephrine, norepinephrine, thyroid hormones, and melanin.^[149] Phenylalanine intake is associated with neurological development and function.^[149] A previous study showed that tyrosine, phenylalanine, and tryptophan were significantly increased in SSF of milk thistle fruit.^[47] Similarly, Morales et al.^[40] observed that fermented cassava bagasse and leaves (51.44 mg EAA/g of protein) can provide approximately 81.7% of the daily requirements (63 mg EAA/g of protein) of Phe and Tyr, which is higher than that found in animal protein such as in raw bacon.

In general, the increase in amino acid content benefits the nutritional value of the fermented products. However, it is important to note that the selection of the starter culture in SSF is crucial when aiming to produce specific amino acids, such as methionine and lysine. Different microorganisms compass varying impacts on the amino acid profile of the final product, and careful consideration must be given to selecting the culture that will yield the desired amino acid composition. Additionally, some amino acids such as histidine, tyrosine, methionine, and cysteine, have been reported as antioxidants. Particularly, histidine exhibits strong radical scavenging activity due to the decomposition of its imidazole ring.^[150] Therefore, the functionalities of amino acids should also be taken into consideration when developing protein fermentation products.

Conclusion

Due to the concerns about environment, cost-effectiveness, sustainability, ethical issues, and health problems such as high cholesterol and lactose intolerance relevant to animal protein-based foods, the preference of plant protein-based foods is increasing. Fermentation is a process with a long history and SSF has the advantage of being less costly and more productive than submerged fermentation, making it a promising method for producing high-nutritional and functional foods from plant protein sources.

Generally, both soluble and total protein contents are increased after SSF. Proteins may come from substrate hydrolysis, microorganism growth, or enzyme secretion to break down macromolecules into small peptides and amino acids. The increment in protein content suggests higher nutritional quality of the final products. Plant proteins inherently have the disadvantage of being less digestible than animal proteins. SSF enhances the hydrolysis of plant-derived proteins by breaking down large molecules into smaller ones, thereby increasing protein digestibility. In addition, SSF can simultaneously eliminate anti-nutritional factors such as trypsin inhibitors and tannins, further increasing the protein digestibility. While SSF can produce bioactive peptides from plant proteins, research in this area is limited. Further research is required on isolating and purifying peptides from fermentation products and assessing their impact on in vitro digestion, cellular models, and animal models. The challenge in industrial production lies in optimizing various protein properties simultaneously. In addition, the development of SSF and its industry application are closely related to the evolution of SSF bioreactor. Innovative bioreactor design is needed to provide precise control, maximize yields, and enhance cost-effectiveness, thereby facilitating the application of SSF in large-scale industrial production. Furthermore, depending on the choice of starter culture and plant protein substrate, a fermented product may exhibit different protein content, protein digestibility, anti-nutritional factor content, bioactive peptides, and amino acid composition. Achieving the desirable combination of these properties requires further research, especially to develop large-scale applications of SSF for plant-based protein production. In conclusion, SSF holds the potential to improve the nutritional value and biological activity of plant proteins in a low-cost and high-yield manner, although more research is needed to identify the specific proteins and peptides that have contributed to the nutritional value and biological activities of the fermented foods.

Author contribution

Xiaoyu Feng: Conceptualization, Investigation, Writing – Original Draft Pangzhen Zhang: Conceptualization, Investigation, Resources, Writing – Review & Editing, Supervision. Ken Ng: Writing – Review & Editing, Supervision. Said Ajlouni: Writing – Review & Editing, Supervision. Zhongxiang Fang: Conceptualization, Resources, Writing – Review & Editing, Supervision.

Acknowledgments

The authors thank the support of University of Melbourne Research Scholarship.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the Agrifutures Australia.