Metaproteomics insights into fermented fish and vegetable products and associated microbes

Highlights • Increasing global population means higher demand for healthy food.• Fish and vegetables are healthy foods, but overproduction leads to spoilage.• Fermentation of fish/vegetables elongate their shelf lives, improved flavour and functions.• Microbes associated with Fish/vegetable fermentation produce health conferring peptides.• There is little review on peptides elicited during fish/vegetable fermentations.


Introduction
One of the main pivots of human health is the type and quantity of diet they consume. Fish and vegetables are known health-conferring sources of food (Olovo, Udoekong, & Akan, 2021;Méndez & Pazos, 2017); however, due to their high nutritional contents and water activity, they are easily degraded, thus lowering their shelf-life and potentially becoming vectors for pathogenic microorganisms. As the world's population burgeons toward the 9.6 billion projected mark by 2050, the demand for food would also increase. To sustain the huge food demand and keep populations healthy, there must be sufficient healthy food supplies (Chaudhary, Bhalla, Patiyal, Raghava, & Sahni, 2021). The overproduction of food to meet huge demands could also lead to spoilage, especially in places without proper storage facilities. Microbes inherent in foods are versatile; some spoil foods, while others help preserve raw foods through fermentation processes, given suitable environmental conditions. Microbes self-perpetuate while interacting with food molecules. During fermentation, they elongate the endproducts shelf life, improve their nutritional contents, aroma, taste, and texture relative to the raw food material (s) (Voidarou et al., 2021;Marco et al., 2017). According to Voidarou et al. (2021), Louis Pasteur defined fermentation as 'life without oxygen'-a process that yields endproducts like CO 2 , ethanol, organic acid, and other organic molecules: important health and industrial products.
Fermentation processes can either be controlled (novel, modern, or industrial, with active allochthonous microbes) or uncontrolled (traditional, natural, or spontaneous, with active autochthonous microbes) and result in the biotransformation of raw food components (Marco et al., 2017). The numerous possible end-products, their bioactivities, and food security seem to be driving forces for the widening applications of fermentation processes in modern times. Functional microbes produce peptides that improve the organoleptic, preservative, digestive, anti-oxidative, probiotic, anti-microbial, anti-toxin, and antianti-nutrient outcomes in fermented foods (Voidarou et al., 2021;Sharma, Garg, Kumar, Bhatia, & Kulshrestha, 2020); these attributes make fermented end-products good natural therapeutics, and can be utilized to curb westernized diets-induced disease conditions like obesity and cardiovascular diseases.
It is common knowledge that consumers of fermented products consume 'live microbes' and transformed molecules together; and that these components confer health (Voidarou et al., 2021;Marco et al., 2017). Fermentation has surpassed the 'preservation' narrative; therefore, it is imperative to study these molecules, elucidate their characteristics, interactions, and find possible applications, especially in the formulation of good foods. Proteomics biochemically measures low molecular weight compounds like peptides, amino acids, aldehydes, organic acids, and amines (Wang, Xia, Gao, Xu, & Jiang, 2017), helps monitor certain of these molecules as fermented food's safety and quality biomarkers (Méndez & Pazos, 2017). There are numerous metaproteomic studies on fermented dairy and certain meat products, but works on fermented fish and vegetable products are scanty. Fish (unsaturated fatty acids like omega fatty acids) and vegetables (bioactive substances like polyphenols) are known for their high nutritional contents (Olovo, Udoekong, & Akan, 2021;Méndez & Pazos, 2017), fermentation can yield even healthier end-products. In this review, we outlined modern proteomic methods and their usefulness; reviewed literature works on fermented fish and vegetable peptides; microbes associated with fish and vegetable fermentations; and challenges encountered in proteomic studies.

Metaproteomic analysis methods
Studies on fermented food proteins-metaproteomic analysis allow for the accurate identification and quantification of proteins; and provides details on their authenticity, origin, biological activities, allergenicity, and sensory properties. Scientists use information gotten from such studies either for descriptive or/and comparison purposes: descriptive analysis provides a crucial understanding of metabolic activities of the microbial communities under specific conditions (Heyer et al., 2019), while comparative analysis simultaneously elucidates the taxonomic composition and functionality of microbial communities in different fermentation processes, and micro-ecology (Heyer et al., 2019). The characterization of proteins occurs at genomic, transcriptomic, and post-transcriptional levels, while their metaproteomic profiles fall under quality, quantity, and functions categories (Ji et al., 2017).
The quantitative approach provides the relative abundance information of a specific protein while focusing on identifying and characterizing the complete protein set present in the fermented food, including their post-translational modifications (PTMs). On the other hand, the functional approach deals with the functional interaction(s) between proteins or between a protein and other molecules. The qualitative properties of proteins are retrieved from either a non-assembled metagenome database, an assembled metagenome database, or a built taxonomy database (Geron, Werner, Wattiez, Lebaron, & Matallana-Surget, 2019).
The workflow for the metaproteomic analysis of fermented fish and vegetables is as illustrated in Fig. 1a. It shows protein extraction from fermented foods and subsequent purification processes. Proteins digested into fragmented peptides are then analyzed with different highperformance separation techniques such as one-dimension and multidimension chromatography, two-dimension gel electrophoresis, and high-resolution mass spectrometry. These methods monitor protein compositions in fermented foods and the changes during the fermentation process (Ortea, O'Connor, & Maquet, 2016). Proteomic studies reveal processes that identified proteins go through and can be a useful tool for toxin, allergen, nutritional value, and storage authentications.

Total protein extraction method
Total protein extraction from fermented foods of interest is the first key step in metaproteomic studies. Factors like differences in cell wall structures and inherent microbial cell lysis resistance level affect protein extraction (Zhang et al., 2018). For fish and vegetable products, another important factor aside from the characteristics of the microbial community is the heterogeneity in salinity and alkalinity of fish and vegetables, respectively (Liu et al., 2021a). These microbial and food factors affect the kind and quality of the extractable proteins to varying degrees.
The basic extraction protocol for proteins involves the physical disruption of cells, dissolution with buffer, purification by precipitation, and protein lysis. The buffer, precipitant, and lysate choices should be considered carefully as they are important for the release of proteins from fermenting microbes (Hayoun et al., 2019). Detergents such as sodium dodecyl sulfate (SDS) and 3-(3-Cholamidopropyl) dimethylammonium)-1-propane sulfonate (CHAPS) show high efficiencies with protein lysis and improvements can be achieved when incorporated with grinding, bead beating, and ultrasonication (Kunath et al., 2019). However, important factors like substrates, products, and repeatability should be considered when choosing methods.

Separation/purification of extracted proteins
The separation and purification of peptides follow after the extraction process, and several methods can be employed just before analyzing the peptides with mass spectrometers. Some of the methods are:

Gel-based separation
Gel separation is one of the commonest methods used for protein separation, uses a 2D polyacrylamide gel electrophoresis/differential ingel electrophoresis (2D-PAGE/DIGE), and yields about 2000 discrete protein spots after staining (Anguraj Vadivel, 2015). Protein expression herein is a factor of the spot size and intensity. Gel separations are done either in single (1D), two (2D), or three (3D) dimensions. The single dimension method of Gel separation bases on the protein's molecular weight (using sodium dodecyl sulfate-polyacrylamide gel electrophoresis SDS-PAGE). The two-dimension method bases on the protein's isoelectric point (using immobilized pH gradient gel strips) and molecular weight (SDS-PAGE). Although the 1D and 2D methods are common, their inability to detect post-translational modifications (as with fish and vegetable proteins) (Arya, Prakash, Sougrakpam, & Deswal, 2021) is a major setback. Colignon, Raes, Dieu, Delaive, and Mauro (2013) adverted the setback by developing the three-dimension method. The 3D uses isoelectric focusing and sample fractionation followed by two consecutive separations with SDS-PAGE, with two different buffer systems. The 3D separation method evades the co-migration interferences that affect protein resolution while providing a wide range of applications in quantitative profiling of complex proteomes and identifying posttranslational modification (Colignon et al., 2013). In addition, another method developed for easy peptide recovery uses a disulfide-containing analog of bis-acrylamide called bis-acrylylcystamine (BAC). Peptides released from the gel can be enhanced by adding tris-(2-carboxyethyl) phosphine (TCEP) for a complete dissolution with BAC-cross-linked acrylamide gel (Takemori, Takemori, Ishizaki, & Hasegawa, 2014). BAC gels can be used for improved complex membrane proteins recovery before mass spectroscopic analysis since low quantity proteins cannot be detected. Incorporating BAC gel with 2D and 3D separation methods, protein recovery, identification, and separation resolution from in-gel digestion is improved.

Gel-free separation
Gel-free separation methods solve most of the limitations with the gel-based method and can also serve as a suitable complement depending on the fermented food type and target protein. The most common approach used is the multidimensional protein identification (MudPIT) with strong cation-exchange (SCX) fractionation, reversedphase (RP) chromatography, and tandem mass spectrometer (MS/MS). MudPIT is an important tool in food proteomic researches and had been used in the study of pumpkin (Cucurbita maxima) and lettuce (Lactuca sativa) (Won et al., 2010). Another comparative approach is incorporating anion/cation exchange reversed-phase chromatography to chromatographically separate proteins (Quan, Feng, Lui, Shi, & Chu, 2017). This combination achieves a 2D separation. Wu, Li, Huang, and Zhang (2021) identified certain proteins with this method and reported that the method was efficient with a high degree of automation and accurate protein information.

Other approach(es)
There are other recently reported methods with efficient protein extract digestion yield and reduced application time. Two examples of these methods are the suspension-trapping filter-based approach (S-Trap) and solid-phase-enhanced sample preparation (SP3). The SP3 method is more efficient in speed and peptide delivery (Hayoun et al., 2019). However, other methods such as molecular imprinting, microfluidic chip, magnetic separation, reverse micelles, and crystallization also have high precision and efficiency when combined with other separation technologies .

MS detection
After the use of gel-based or gel-free methods to sufficiently reduce sample complexities, the peptides are mixed and analyzed using mass spectroscopy. In proteomic studies, the most used ionization methods are electrospray ionization (ESI)  and matrix-assisted laser desorption/ionization (MALDI) (Kailasa et al., 2020). However, electrospray ionization is employed more due to its high-throughput secondary spectra of peptides. Directly coupling liquid chromatography (LC) with ESI-MS achieves a highly automated detection of peptides in fermented food (Liu, Qin, Mazhar, & Miao, 2021b;Ortea et al., 2016), whether targeted or non-targeted (Hart-Smith, Reis, Waterhouse, & Wilkins, 2017). Targeted quantitative metaproteomics has gained popularity when used with mass spectrometry-based protein quantification because it can detect and analyze specific sets of proteins with high sensitivity, accuracy, and reproducibility . According to Hart-Smith et al. (2017), targeted proteomic assays can be accomplished through the acquisition of peptide MS/MS data using selected reaction monitoring (SRM), parallel reaction monitoring (PRM), or targeted data acquisition (TDA). Also, a targeted proteomic assay can be achieved via comprehensive data-independent acquisition (DIA) strategies, e.g., sequential window acquisition of all theoretical spectra (SWATH), which produces extractable quantitative data using 10,000 proteins assay libraries (Sim et al., 2020).
Non-targeted quantitative proteomics has been explored more in foods. It is a more exploratory, data-dependent acquisition (DDA) using LC-MS/MS analysis. The highest abundance of peptide ions from full MS scans is selected for MS/MS (Yesiltas et al., 2021). Herein, highly abundant microbial proteins (>1% relative abundance by riBAQ) are selected for bioinformatic analysis, and the full-length sequences from microbial and food proteins were analyzed using EmulsiPred (htt ps://github.com/MarcatiliLab/EmulsiPred). Other methods that can employ non-targeted approaches include isobaric tags for relative and absolute quantitation (iTRAQ), and tandem mass tags or label-free. These methods were recently employed for the protein profiling of certain food substances (like commercial soybean milk and quinoa seeds) (Galindo-Luján et al., 2021).

MS data analysis
In proteomic studies, MS/MS is commonly used for protein identification compared with peptide mass fingerprinting. For fermented fish and vegetable proteins, the use of this procedure becomes important because there is limited information in the public database on related proteins. Fig. 1b shows the schematic illustration for the bottom-up and top-down MS-based metaproteomics approaches suitable for use in fermented fish and vegetable protein researches. With the top-down proteomics, intact or large protein fragments are directly subjected to gasphase fragmentation for MS/MS analysis; by contrast, the bottom-up approach is widely used for protein identification by MS. Complexmixtures, or purified proteins are subjected to proteolytic cleavage; while MS or MS/MS is used to analyze resulting peptides. The commonly used software for peptide and protein identification and quantification are categorized based on their functions (Chen, Hou, Tanner, & Cheng, 2020), as presented in Fig. 1C. SwissProt, NCBInr, PIR, and EMBL are a few public databases tools; however, they do not include fermented fish and vegetable peptide data. The unavailability of fermented fish and vegetable protein data increases the risk of misidentifying newly extracted peptides from these fermented foods. Available protein databases are either generated from shotgun metagenomic data or are pseudo-metagenomics databases constructed from proteins obtained from a public database according to the microbial community structure obtained by 16S rRNA analysis (Muth, Renard, & Martens, 2016). Recently, transcriptome-derived protein databases are gaining increasing attention as manually curated food proteomics databases such as FermFooDb (https://webs.iiitd.edu.in/raghava/fermfoodb/) (Chaudhary et al., 2021). FermFooDb comprises 2205 entries with relevant fields like peptide sequence, Mass and IC50, food source, functional activity, fermentation conditions, starter culture, testing conditions of sequences in vitro or in vivo, type of model, and method of analysis.

Bioinformatics analysis
The bottom-up proteomic approach requires taxonomy-specific peptides for microbial community identification, quantification, functional annotation, metabolic, and biological classification. Thus, bioinformatics data is an integral part of metaproteomics (Krutz, 2019); some analytical tools used include the Protein ANalysis Through Evolutionary Relationships (PANTHER); Database for Annotation, Visualization and Integrated Discovery (DAVID); Ingenuity Pathway Analysis (IPA, QIA-GENs Redwood City, www.qiagen.com/ingenuity); Gene Ontology (GO); Cluster of Orthologous Groups (COG); and Kyoto Encyclopedia of Genes and Genomes (KEGG) (http://geneontology.org/) (Liu et al., 2021b;Heyer et al., 2017). Also, the protein-protein interactions databases, like the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING); Molecular INTeraction database (MINT); IntAct, Microbial Protein Interaction Database (MPIDB) as extensively described by Calderón-González, Hernández-Monge, Herrera-Aguirre, and Luna-Arias (2016) are also used. Among the commonly employed bioinformatics tools, the GO provides controlled, structured vocabularies and classifications that focus on the knowledge of gene products' roles in cells. These gene products are classified either as cellular components (CC), biological processes (BP), and molecular functions (MF) (Galindo-Luján et al., 2021). The COG represents a phylogenetic classification of proteins from completely sequenced genomes used to distinguish evolutionary relationships between orthologs and paralogs (Heyer et al., 2017), while KEGG analytical tool groups microbial genes and proteins according to their molecular levels and functional information (L et al., 2021b).

Proteomic analysis of fish
Fish consumption is growing rapidly at an average annual per capita fish consumption rate of 9.0-17.8 kg (FAO, SOFIA, 2014). The high consumption volume is due to extensive outdoor fish cultivation that meets high demands at very cheap rates. Fish is a cheap source of food and contains a high amount of nutrients such as proteins, vitamins, minerals, long-chain polyunsaturated fatty acids (PUFs), eicosapentaenoic (EPA), docosahexaenoic acid (DHA), peptides, selenium, and taurine; these nutrients help prevent and treat metabolic diseases (Méndez & Pazos, 2017). According to Jacek (2005), the high nutritional content and water activity of fish becomes a disadvantage for its shelf-life. In addition, fish muscle proteins are heat-labile, have an abundance of proteolytic enzymes, and aids the proliferation and survival of psychrophilic microbes. These factors combined lower fish's shelf life when compared to terrestrial animals' meat.
The shelf life of harvested fish can be extended either by salting, smoking, or drying; however, when left for extended periods, the activities of co-functioning multiple microorganisms and their enzymes ferment the minimally processed fish-yielding unique organoleptic properties and extending the fish product's shelf-life (Ji et al., 2017). The microorganisms and their enzymes convert certain organic materials into simpler compounds associated with unique aromas, texture, and flavour characteristics found in fermented fish products. There are two classifications for fermented fish according to Zang, Xu, Xia, and Regenstein (2019) (a) by the nature of their end products; (i) whole or sliced, but the fish still retains its structure, (ii) paste, here the fish is fermented in paste form, and (iii) sauce, here the fish is fermented in a broth or liquid and (b) by the fermentation method adopted; (i) traditional or spontaneous, here natural or autochthonous microbes biotransform the fish and (ii) starter culture, here cultured or allochthonous microbes bio-transform the fish.
Different organoleptic outcomes for fermented fish, depending on countries of origin, fish types, and environmental factors (process) (Waisundara, Jayawardena, & Watawana, 2016). Modern sciences like microbiology and biotechnology help elucidate microbial biotransformations of raw or minimally processed food items like fish. The bacteria Photobacterium profundum was the most abundant identified microbe in fish fermentation studies conducted by Ji et al. (2017). Functional microbes involved in fermentative processes yield useful end-products during their primary and secondary metabolic activities; their actions on fish could either be acidification (or carbohydrate catabolism) -that yields anti-microbial peptides (extends shelflife), gelation of myofibrillar and sarcoplasmic proteins of the musclespeptides that change the elasticity, cohesion, and hardness of endproducts, and the degradation of proteins and lipids that yields myriad of peptides that allude to flavourful, tasty compounds and easily digestible and absorbable nutrients (Zang et al., 2019). Table 1 shows peptides identified from fermented fish.

Proteomic analysis of vegetables
Vegetables, just like fish products, are also considered a healthbeneficial diet type; they have high nutrient and water activity content. These factors make them easily liable to microbial and enzymatic degradations (Sajjad, Rasool, Ahmad Fazili, & Ahmed Bhat, 2000), hence their short shelf life. Numerous works on plant-sourced foods show that they contain various bioactive substrates, either nutritive or anti-nutritive (Sharma et al., 2020;Awak, Udofia, Akan, Uffia, & Udoekong, 2017). Notwithstanding, the microorganisms involved in plant-food fermentation increase the nutritional contents of fermented plant products by (a) increasing the amount and bioavailability of nutrients and (b) enhancing the density of nutrients (Nkhata, Ayua, Kamau, & Shingiro, 2018). They enhance the density of nutrients by synthesizing promoters that aid with nutrient adsorption, influence the uptake of nutrients via the mucosa membrane, pre-digest the food components, and reduce or degrade the anti-nutrient contents of the parent food. Bacteria from the Lactococcus family are the most identifiable microbes associated with plant fermentation.
According to Voidarou et al. (2021), fermentation, an age-long process, is gaining keen interest. Many beneficial end-product possibilities are obtainable as long as microorganisms and food substrates are placed side by side in conducive environments (anaerobic). The endproducts from vegetable-carbohydrate fermentation are as varied as many types of vegetables and varieties of fermenting microorganisms (Table 1). Plant carbohydrate fermentation entails accepting electrons by organic molecules (pyruvate or acetyl CoA) and reactions that lead to the formation of important peptides. Peptides from fermented vegetables can be used as microbial energy sources; howbeit, they are not the preferred energy molecule. Instead, they are used for the production of hormones, enzymes, and haemoglobin; needed for cell growth and repairs, the normal functioning of the muscles, nerve signaling, and immunity (Marsh, Munn, & Baines, 2013), making plant-sourced peptides an important set of macromolecules for body functions. Fermented vegetable proteins are linked with decreased risk for metabolic diseases (Marsh et al., 2013); this is due to plant-sourced diets having very low saturated fats, cholesterol-free, good sources of antioxidants, high fiber contents, and haem iron. Table 1 enlists works with immunomodulatory benefits of peptides from fermented vegetables.

Proteomics of microbes associated with fermented foods
Metaproteomic studies of any fermented food give insights into microbial type, community interactions, typical roles, and expressed protein molecules (Ji et al., 2017). The demand for fermented products with high consistency by urban dwellers is turning food industries into controlled fermentation processes; however, rural dwellers still prefer traditional methods (Tamang et al., 2020, Tamang, Shin, Jung, & Chae, 2016. About 90% of consumed fermented products are prepared uncontrolled in homes utilizing inherent food microbes. The mixture is left on its own or given conducive environmental conditions to thrive. Traditionally fermented foods are generally simple, require simple ingredients, minimal preparations, and processing (Marco et al., 2017); however, the diversity of fermenting microbial succession involved is very complex, notwithstanding.
In every natural microbe-food substrate mixture, there is also a mixture of functional and non-functional microorganisms. The functional microbes are responsible for the biotransformation of the chemical constituents in food substrates (Tamang, Shin, Jung, & Chae, 2016). The fermentation process, whether traditional or industrial, follow three distinguished units of operations thus: (a) pre-treatment of the food substrates, which includes transportation, salting, grading, washing, sorting, mixing, etc. (b) the bioprocessing that utilizes suitable microbes/enzymes that bio-transform, synthesize, remove, degrade, etc., certain substrates and (c) bioprocessing atmosphere, that includes cooling, freezing, and heating.
Stability in the composition of microbes is an important factor, as alterations in their diversity during the process might yield noticeable differences in organoleptic properties/quality-even with the same food substrate. Therefore, to have consistent, high-quality, safe, and good sensory fermented end-product(s), the microbial composition must be stable and resilient (Marco et al., 2017). Modern fermentation has progressed towards end-product consistencies, with certain process modifications. Portions of previously fermented foods (like fruits or malted cereals with fermenting microbes) can be added to raw food substrates or new food batches to initiate fermentation. This method termed back slopping is frequently used to standardized microbial fermented end-products (Tamang et al., 2020). Standardized starter cultures allow for consistent fermented end-products, reduced spoilage, increased food safety, and ensures large-scale production of fermentation end-products (Tamang et al., 2020). Just as with those functional microorganisms in traditional/natural fermentation, starter microbial cultures transform the raw foods and yield more desirable/healthier end-products.
Fermented foods (fish and vegetables inclusive) are studied using different methods. The principal is the conventional microbiological culture-dependent methods, where isolates are identified using phenotypic and biochemical characterizations. However, microbes are now identified using molecular methods like metagenomics and massive sequencing. Lactic acid bacteria (LABs), such as Leuconostoc, Lactobacillus, and Weissella species (Chen, Chen, and Lei, 2017), are the prominent bacterial species that help ferment vegetables. According to Lee, Jung, & Jeon (2015), genomics and other technologies have been used to study the dynamic microbial communities and metabolic changes during fermentations. The functions of the peptides make fermentation end-products a vital patterned food type. Yang, Fan, and Xu (2020) identified about 2,175 proteins in fermented Siniperca chuatsi, 1,217 of which were involved in metabolic pathways, while 352 were associated with amino acid metabolism. Certain microbes such as Streptococcus sp., Bacillus sp., Escherichia sp., and Pseudoalteromonas sp. possess about 63 amino acids that are degradation-related, all of which generate aromatic compounds. These compounds are responsible for the unique taste, flavour, and organoleptic attributes of fermented foods (Yang et al., 2020).

Challenges and future perspectives
Some challenges hamper the application of metaproteomic analysis on traditionally fermented foods (fish and vegetables); these include but are not limited to the redundancy of protein identifications, impurities, and complexities of food samples, paucity of genome sequences essential for their protein identifications. Additionally, uneven distribution of species, extensive fluctuations in expression levels of proteins in microorganisms, and significant genetic varieties within microbial communities are other major challenges in metaproteomic studies (Simon & Daniel, 2011). Furthermore, identifying proteins relying on a metagenomics database derived from the same sample and any sequence cross-contamination or submission errors onto the metagenomics database may compromise homologous protein identification (Pible & Armengaud, 2015). Therefore, care must be exercised throughout the Energy metabolism and amino acid metabolism Ji et al. (2017).

Shewanella violacea
Cold shock-like protein CspA Genetic information processing and cellular processes Ji et al. (2017).

Shewanella violacea
Cold shock-like protein CspA Genetic information processing and cellular processes Ji et al. (2017).

Siniperca chuatsi
Vibrio campbellii DNA-directed RNA polymerase subunit beta Genetic information processing and cellular processes Ji et al. (2017).

Siniperca chuatsi
Vibrio cholerae DNA-directed RNA polymerase subunit alpha Genetic information processing and cellular processes Ji et al. (2017).

Fermented Vegetable Products Vegetable Product
Fermenting microbes Identified proteins and peptides Function Reference
Dichloromethane, chlorophyll, carotenoids, phenolics and vitamin C, capsaicin, quercetin, and 3-(4 ′ -hydroxyl-3 ′ ,5 ′ -dimethoxyphenyl) propionic acid Anti-oxidative activity Woo, Kim, Noh, and Song (2017); Hwang and Song (2001). entire process. Other concerns are adequate and comprehensive protein separation, meta-information from online sources, and redundant protein groupings. These also need attention as they can affect the outcomes. Advanced methods need to be employed for the extraction, identification, and verification methods of metaproteomic studies to be effective. Although omics technologies have offered unique datasets at a variety of molecular levels (Lagier et al., 2018), metaproteomics and other omics approaches, on the other hand, cannot fully reveal the presence and growth of a microbial community. Due to the vast extreme complexities of samples from traditionally fermented fish and vegetables, systematic biology can provide a whole picture of activities and higher-level biological relationships by merging multi-omics data and bioinformatics technology to link cause and effect. Consequently, the molecular nature of biological activities is revealed (Cocolin et al., 2018), thereby obtaining new functional microorganisms as well as functional metabolites in fermented fish and vegetables. Furthermore, for effectiveness, metaproteomics is not a stand-alone technique, it should be combined with cytometry, microscopy, meta-transcriptomics, metabolomics, and metagenomics for thorough investigations and understanding of microbial populations and metabolic models (Vilanova & Porcar, 2016).

Leuconostoc citreum
Protein modification and interaction should be another point of interest, especially with traditional fermentation (Gagnaire, Jardin, Jan, and Lortal, 2009). Additional researches into regulatory links between proteins' post-translational modifications and metabolites, as well as proteins interactions, can help to elucidate the regulatory mechanisms of proteins and product quality in traditionally fermented foods. These interactions and modifications are capable of altering fermentation outcomes and processes.
The development of advanced software tools with features to handle enormous datasets and are user-friendly can greatly improve metaproteomic analyses. The provision of cheaper metaproteomic analytic tools, just like DNA sequencing, would make analysis commonplace (Chiapello, Zampieri, & Mello, 2020) and enhance the listing of more novel proteins never before reported/explored in traditional or controlled fermentations.

Conclusion
Fermentation is an important process, not just because it preserves food. However, it further produces various molecules and metabolites that make the end-products healthier than the initial raw food substrates, all thanks to versatile microorganisms and conducive environments. Some of the end-product proteins are signature organoleptic, preservative, and anti-microbial peptides. Therefore, increased detailed data from studies are needed to elucidate the microecology responsible for the production of peptides, their accurate identification, characteristics, and functions in fermented fish and vegetables (Tamang et al., 2020). The elucidation would enable a more controlled production of these important peptides on a large scale.
Fermentation is a process that will always be with us. With the arrays of microbes identified in fermented fish and vegetable products (although studies on fermented fish peptides are few), researchers can enlist important molecules associated with anti-microbial, anti-obesity, anti-oxidative, anti-hypertensive, even protein molecules that inhibit the activities of ACE-a host receptor that viral agents attach for host entry and replication. Nutraceutical firms can harvest these proteins and formulate them into good foods. Food and pharmaceutical scientists can solve the need for natural therapeutics to abate certain diseases via fermentation and harvesting these sets of peptides from healthy raw foods (fish and vegetables).

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.  (2007); Sura, Garg, and Garg (2001).