Download

RESEARCH ARTICLE

Identification and characterization of novel bioactive peptide from red seaweed (Pyropia vietnamensis) proteins

Nur Iliana Basri1, Amiza Mat Amin1,2*, Fisal Ahmad1,2

1Faculty of Fisheries and Food Science Universiti Malaysia Terengganu (UMT), Terengganu, Malaysia;

2Functional Food RIG, Food Security in Changing Climate SIG, Food Security Research Cluster, Universiti Malaysia Terengganu (UMT), Terengganu, Malaysia

Abstract

Pyropia vietnamensis is one of the most abundant seaweeds in the Indo-Pacific region. This study aimed to perform an in silico evaluation of P. vietnamensis proteins as potential precursors of bioactive peptides and to determine the novel peptide in terms of its half maximal inhibitory concentration (IC50) and the stability under a controlled laboratory environment (in vitro) toward the dominant biological activity. The proteomic profiles of P. vietnamensis proteins were determined using LC-MS/MS analysis. Next, five proteins were chosen and employed for in silico analysis using the BIOPEP-UWM database. The in vitro characterizations of novel peptides were carried out using a dipeptidyl peptidase-IV (DPP-IV) inhibitor screening assay kit. In silico analysis revealed that DPP-IV and angiotensin-converting enzyme (ACE) inhibitors were the most potential bioactive peptides in P. vietnamensis proteins. Calpain 2, papain, pepsin (pH > 2), and stem bromelain were predicted as the enzymes with the most potential to produce DPP-IV and ACE inhibitors. The novel peptides predicted were CFA, ACF, RFPS, DEWG, NYCL, CVPR, and DACF. The synthesized CVPR with an IC50 of 0.66 mg/ml exhibited stability at pH 3–7, 30–50°C, and resisted gastrointestinal digestion. This study revealed P. vietnamensis proteins could offer health benefits due to their therapeutic potential with sustainable industrial applications on functional foods, nutraceuticals, and pharmaceuticals.

Key words: ACE inhibitory peptide, antidiabetes, antihypertensive, DPP-IV inhibitory peptide, in silico

*Corresponding Author: Amiza Mat Amin, Functional Food RIG, Food Security in Changing Climate SIG, Food -Security Research Cluster, Universiti Malaysia Terengganu (UMT), 21030 Kuala Nerus, Terengganu, Malaysia. Email: ama@umt.edu.my

Academic Editor: Jelena Popović-Djordjević, PhD., Department for Food Technology and Biochemistry, Faculty of Agriculture, University of Belgrade, Belgrade, Serbia

Received: 27 September 2024; Accepted: 5 January 2025; Published: 3 February 2025

DOI: 10.15586/qas.v17i1.1525

© 2025 Codon Publications
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0). License (http://creativecommons.org/licenses/by-nc-sa/4.0/)

Introduction

Seaweeds are classified into red, green, and brown based on their pigmentations, nutrients, and chemical compositions (Sultana et al., 2023). They have been consumed as food mainly in China, Japan, and the Republic of Korea. According to El-Beltagi et al. (2022), seaweeds are increasingly utilized in the medical field due to their diverse structural features and novel biological activities including antihypertensive, antidiabetic, antimicrobial, anticancer, and antioxidant properties. Pyropia, formerly known as Porphyra, accounted for more than 95% of cultivated seaweed globally in 2019, and Malaysia is one of the top 10 producers. Several Pyropia species are used to prepare commercial food called “nori,” which is widely consumed by the Japanese. Bioactive peptides are protein fragments composed of 2–20 amino acids that exhibit health-promoting properties in the human body (Agirbasli and Cavas, 2017). The protein content of red seaweeds is the highest (up to 47%) compared to green (9–26%) and brown seaweeds (3–15%) (Peñalver et al., 2020). Hence, red seaweeds could be a good precursor for the generation of bioactive peptides. The identification of bioactive peptides from food proteins can be determined either via in vitro or in silico approach. This study utilized an in silico approach using the BIOPEP-UWM database that contained information on bioactive peptides including their protein sequences and biological properties. The in silico method is important in the prediction of potential biological activity and enzyme substrate to release specific bioactive peptides.

In the advanced research of bioactive peptides, in silico methods enable efficient screening, design, and characterization of peptides. The integration of in silico with experimental methods may accelerate the development of bioactive peptides for diverse applications, such as functional foods, nutraceuticals, and pharmaceuticals. The key applications of the in silico method include peptide discovery and screening, peptide design and optimization, molecular docking and dynamic simulation, and structure-activity relationship (SAR) analysis. Potential bioactive peptides released from the proteins can be commercialized as functional foods, nutraceuticals, and pharmaceuticals targeted for certain bioactivity. For example, Nori S peptide (Shirako Co., Ltd., Tokyo, Japan) and Wakame peptide jelly (Riken Vitamin Co., Ltd., Tokyo, Japan) are functional products from seaweed produced in Japan with FOSHU-approved antihypertensive claims (Hayes and Tiwari, 2015).

Generally, this study was conducted to fill gaps in current knowledge that limit the full understanding and application of seaweed-derived peptides in managing metabolic disorders. For example, there is limited diversity in peptide identification and isolation where only a small fraction of the vast diversity of seaweed peptides has been explored especially DPP-IV inhibition. Also, the variability in levels of bioactivity between seaweed species is poorly understood, for instance, Palmaria palmata possessed potent DPP-IV inhibitory activity (Harnedy et al., 2015). Hence, it is anticipated to observe the factors influencing this variability including environmental conditions and extraction method of Pyropia vietnamensis. Additionally, this study paves the way for a deeper understanding of peptide stability under physiological conditions such as pH, temperature, and gastrointestinal digestion which affects their potential bioavailability and efficacy. Importantly, the advanced screening techniques used in this study are valuable in research directions to address gaps. Recently, the antioxidant and antimicrobial properties of P. vietnamensis have been reported by Bhatia et al. (2021) in vitro. To the best of our knowledge, no study has been reported on proteomic profiles of P. vietnamensis proteins and the identification of bioactive peptides from it using in silico approach. Therefore, this study aims to determine the proteomic profiles of P. vietnamensis proteins, to identify its most promising bioactive peptides using an in silico approach and to perform in vitro characterization of a synthesized novel peptide.

Materials and Methods

Materials and chemical reagents

Around 1 kg of P. vietnamensis was collected from the upper intertidal zone at Tok Jembal Beach, Kuala Nerus, Terengganu, Malaysia during the monsoon season (low tide). Liquid nitrogen was purchased from MCY Marine (Kuala Terengganu, Terengganu, Malaysia). Phenol, dithiothreitol (DTT), urea, pancreatin (EC232.468.9), formic acid, bovine trypsin (proteomic grade), and dipeptidyl peptidase IV (DPP-IV) Inhibitor Screening Kit were purchased from Sigma-Aldrich (St. Louis, Missouri, USA). Pepsin (EC3.4.23.1), ammonium acetate, and methanol were purchased from R&M Chemicals (Essex, UK). The Coomassie (Bradford) Protein Assay Kit was purchased from Thermo Fisher (Rockford, Illinois, USA). All other chemicals used were of analytical grades.

Preparation of sample

A schematic diagram of the methodology has been illustrated in Figure 1. Firstly, the mature P. vietnamensis was carefully selected and collected by removing the dried and premature ones. The sample was thoroughly rinsed with clean seawater to eliminate debris, sand, epiphytes, and other extraneous substances and was delivered to the laboratory in an ice box. In the laboratory, P. vietnamensis were rinsed with distilled water and dried with tissue paper to remove excess water. The cleaned seaweed samples were deep-frozen in liquid nitrogen for rapid freezing and crushed directly in liquid nitrogen into a fine powder using a prechilled mortar and pestle, before freezing at −80°C. Next, acetone-dried powder (AcDP) of seaweed was prepared according to Awang et al. (2010) with some modifications. Seaweed powder (±5 g/tube) was washed with ice-cold acetone (10 mL/tube) using vortex for 5 min and was then centrifuged at 10,000 × g for 5 min at 4°C before the supernatant was discarded. This washing step was carried out until the color of ice-cold acetone was clear to ensure that the contaminants, for example, lipids, polysaccharides, phenolic compounds, and pigments, were completely removed from the sample, to facilitate protein extraction. In this study, the washing step was repeated nine times to get a clear color of ice-cold acetone. Next, the remaining ice-cold acetone in the resulting pellet was allowed to evaporate at room temperature. AcDP was stored at −80°C in a sealed container for further analysis.

Figure 1. A schematic diagram of the methodology.

Protein extraction

Protein extraction was performed using the phenol extraction methanol–ammonium acetate precipitation (PHE) method as described by Carpentier et al. (2005) and Awang et al. (2010) with some modifications. AcDP (1.0 g) was resuspended in 10 mL of extraction buffer (50 mM Tris-HCl pH 8.5, 0.5 M EDTA, 100 mM KCl, 1% v/v DTT, 30% w/v sucrose) and was mixed using vortex mixer for 30 s. Next, ice-cold Tris-buffered (pH 8.0) phenol (10 mL) was added, and the mixture was mixed using a vortex mixer for 15 min, followed by centrifugation (10,000 × g for 3 min) at 4°C. The phenolic phase was collected and reextracted with an equal volume of extraction buffer. The phenolic phase after centrifugation was allowed to precipitate overnight with five times the volume of cooled 100 mM ammonium acetate in methanol at −20°C. Then, the supernatant was discarded after centrifugation (10,000 × g, 30 min at 4°C). The resulting pellet was washed twice with ice-cold acetone containing 0.2% v/v DTT and between the two rinsing steps, the sample mixture was incubated for 1 h at −20°C. During each wash, the pellet was vortexed for 10 min at 4°C to remove pigments and other interfering compounds properly. The pellets were air-dried at room temperature (26°C ± 1). The pellets were resuspended in 300 μL resolubilization buffer containing 7 M urea, 2 M thiourea, 4% v/v CHAPS, and 1% v/v DTT. The sample mixture was gently vortexed every 15 min for 1 h at room temperature. The supernatant was collected after centrifugation at 10,000 × g for 30 min at 20°C and was stored at −80°C for further analysis. Total soluble protein in the supernatant was determined using the Bradford assay (Bradford, 1976), on 96-well microtiter plates by referring to the Thermo Scientific Coomassie (Bradford) Protein Assay Kit Instructions. It was found that the soluble protein content of P. vietnamensis was 14.90 mg/mL (n = 4).

LC-MS/MS analysis and database searching

Before LC-MS/MS analysis, the extracted protein sample was digested using the trypsin in-solution digestion method (Kinter and Sherman, 2005). The sample was then mixed with 100 µL of 0.1% formic acid in deionized water. The solution was filtered using a 0.45 µm regenerated cellulose (RC) membrane syringe filter (Sartorius AG, Goettingen, Germany). The LTQ-Orbitrap Velos Pro mass spectrometer (Thermo Fisher Scientific, CA, USA) coupled with Easy-nLC II nano-liquid chromatography system was used to perform the analysis. The eluent was sprayed into the mass spectrometer at 2.1 kV (source voltage) at 220°C. Full scan mass analysis was performed from 300–2000 m/z at a resolving power of 60,000 (at m/z 400, FWHM; 1-s acquisition) with data-dependent MS/MS analyses (ITMS) triggered by the eight most abundant ions from the parent mass list of predicted peptides with rejection of singly and unassigned charge state. Collision-induced dissociation (CID) was applied as the fragmentation technique. Collision energy was set at 35. The sample was analyzed in duplicate readings.

PEAKS Studio software (PEAKS, version 7.5, Bioinformatics Solution, Waterloo Canada) was used to perform de novo sequencing and database searching. The peptide sequences were matched with proteins from the UniProtKB (Seaweed and Algae) database. Parent mass and precursor mass tolerance were set at 0.1 Da. The false discovery rate (FDR) of protein and peptide detection was set at <0.1% to increase confidence. In addition, the significant score of −10l gP for proteins greater than 20 was used for protein acceptance. At the same time, the minimum number of the unique peptide was set as to 1.

Protein sequences of red seaweed (P. vietnamensis)

Five proteins from LC-MS/MS results were selected based on the highest confidence level (−10l gP) values (240.79–162.24) and the protein sequences were confirmed with the Universal Protein Knowledgebase (UniProtKB, https://www.uniprot.org/) library database.

In silico analysis using BIOPEP-UWM database

Evaluation of P. vietnamensis as a potential precursor for bioactive peptides

The probability of five selected P. vietnamensis proteins to liberate bioactive peptides was analyzed using the BIOPEP-UWM database. Accessed on January 14, 2023, this database contains 4623 acknowledged bioactive peptides with 62 biological activities. The number of potential bioactive peptides of each protein was computed using “profiles of potential biological activity” in the database and the predicted biological activity was analyzed altogether. The data obtained were tabulated in Microsoft Excel 2021 and the predicted fragments of specific biological activity to be released from the selected proteins were counted manually. The biological activity that gave the highest number of fragments was chosen to be reported.

The frequency of occurrence of fragments with given activity (A) of P. vietnamensis proteins was calculated by the database using the following formula, where a is the number of fragments of given activity in a protein sequence, and N is the number of amino acid (AA) residues in the protein chain. The total frequencies of occurrences of fragments (∑A) for each protein sequence were also calculated by the database.

A=aN

The BIOPEP-UWM database also shows the potential biological activity of the protein (B) of proteins and was calculated by using the following formula, where a is the number of fragments of given activity in a protein sequence, and N is the number of amino acid (AA) residues in the protein chain.

B=i=1kaiEC50i*N

In silico proteolysis

Pyropia vietnamensis proteins were subjected to in silico proteolysis using BIOPEP’s enzyme-action tool. Figure 2 shows the flowchart of in silico proteolysis conducted in this study. Each protein sequence was hydrolyzed individually by 33 different enzymes contained in the database. The potential enzymes to release bioactive peptides with diverse biological activities from the proteins were selected to report including pancreatic elastase, papain, ficin, leukocyte elastase, stem bromelain, calpain 2, and pepsin (pH > 2). Enzyme combinations were performed among those seven selected enzymes. The efficiency of the released fragments was measured based on the frequency of occurrence of fragments by selected enzymes (AE) using the following formula, where d is the number of fragments with a given activity in the protein sequence released by enzymes, and N is the number of amino acid residues in protein chains. The sum of frequencies of occurrences (∑AE) of all P. vietnamensis proteins for DPP-IV inhibitory and ACE inhibitory activities using a single and combination of two or three enzymes was summarized in Figure 3.

Figure 2. Scheme of in silico proteolysis using enzyme-action tool of BIOPEP-UWM database.

Figure 3. The sum of frequencies of occurrences (∑AE) of DPP-IV inhibitor and ACE inhibitor using single and combination enzymes.

AE=dN

In silico screening and characterization of novel peptides

The peptide fragments predicted to be released from P. vietnamensis proteins with known biological activity were counted manually. BIOPEP-UWM shows the fragments with the activities available in the database. In this study, sequences of three and four amino acid residues were screened and considered for potential biological activity. The bioinformatics tool used to measure the bioactivity score for peptide fragments was PeptideRanker which is accessible at http://distilldeep.ucd.ie/PeptideRanker. It is based on the prediction and ranking of the likelihood of a peptide exerting bioactive properties based on an N-to-1 neural network algorithm. According to Coscueta et al. (2022), a peptide that scores above the PeptideRanker threshold (0.5) is identified as bioactive. Thus, novel peptides with a bioactivity score >0.5 were considered for analysis as potentially bioactive. In silico characterizations in terms of water solubility, resistance to digestion, toxicity, allergenicity, and IC50 were conducted to determine the potential characteristics of novel peptides. The solubility of novel peptides in water was predicted using PepCalc at http://pepcalc.com. Gastrointestinal digestion was predicted using PeptideCutter, available at http://webexpasy.org/peptide_cutter. Three different enzymes; chymotrypsin-low specificity, chymotrypsin-high specificity, pepsin (pH 1.3), pepsin (pH > 2), and trypsin were used to evaluate the resistance of the novel peptides against digestion. Next, ToxinPred and AllerTop were utilized to determine the toxicity and allergenicity of the novel peptides, which are freely accessible at http://www.imtech.res.in/raghava/toxinpred/ and https://www.ddg-pharmfac.net/AllerTOP, respectively.

In vitro analyses

The selected novel tetrapeptide (CVPR) was synthesized to further analyze in vitro. The synthesis was not performed by the authors of the article but outsourced to the commercial company Apical Scientific Sdn. Bhd. (Malaysia). This synthetic peptide with purity of 99.1% was used to determine its DPP-IV inhibitory activity and stability.

DPP-IV inhibitory activity

The activity of synthetic DPP-IV inhibitory peptide was determined using a DPP-IV Inhibitor Screening Kit (Sigma-Aldrich, USA) in 96-well plates. All assays were performed in triplicate (n = 3) and Sitagliptin was used as a positive control. The DPP-IV inhibitory activity was determined by plotting % inhibition as a function of the concentration of the test compound. The values were expressed as the mean of half maximal inhibitory concentration (IC50) that inhibits DPP-IV activity by 50% ± standard deviation. The DPP-IV inhibitory activity was calculated as follows, where FLU1 is the fluorescence at T1; FLU2 is the fluorescence at T2; SlopeSM is the slope for the inhibitor sample and SlopeEC is the slope for enzyme control.

Slope=FLU2FLU1T2T1
Relative inhibition=SlopeECSlopeSMSlopeEC×100%

Stability of a synthetic DPP-IV inhibitory peptide

The stability of a synthetic DPP-IV inhibitory peptide against pH, temperature, and simulated gastrointestinal digestion was determined according to the method by Kong et al. (2021) and Walsh et al. (2004) with minor modifications. For pH stability, the peptide was diluted with deionized water to a working concentration of 2 mg/mL. Next, 10 mL of peptide solution was adjusted to different pH values (3.0, 5.0, 7.0, 9.0, or 11.0, respectively) and incubated in a water bath at 25°C for 30 min. The pH of the treated samples was immediately adjusted to pH 7. The sample without pH treatment was used as a control.

For temperature stability, the peptide was diluted with deionized water and made up to 2 mg/mL working concentration. Next, 10 mL of peptide solution was heated at different temperatures (30°C, 50°C, 70°C, or 90°C, respectively) in a water bath (Memmert GmbH, Germany) for 30 min. All samples were immediately cooled in iced water. The sample without heat treatment (25°C) was used as a control.

For stability against simulated gastrointestinal digestion (SGID), the peptide was diluted to 10 mg/mL working concentration and reduced to pH 2.0 using 1 N HCl. Then, pepsin was added to the peptide solution with a ratio of an enzyme to the substrate of 1:40 (w/w,) and the reaction mixture was incubated in a water bath shaker for 90 min at 37°C. The pH of the reaction mixture was then adjusted to pH 7.5 by the addition of 1 M sodium hydroxide (NaOH) and was further incubated with pancreatin at an enzyme-to-substrate ratio of 1:25 (w/w) in a water bath shaker at 37°C. After 150 min, the reaction mixture was exposed to 80°C for 20 min to stop the digestion. The DPP-IV inhibitory activity was measured and expressed as the activity (%) relative to that without any treatment (control, 100%) for all of the assays above.

Statistical analysis

All results were presented as mean ± standard deviation. Data were subjected to One-Way analysis of variance (ANOVA) by using SPSS Version 20.0 (SPSS Inc., Chicago). Significant difference between treatments were analyzed further using least significant difference (LSD) at 0.05 significant level.

Results

Potential bioactive peptides from P. vietnamensis proteins

Protein sequences in FASTA format and common features for selected proteins from P. vietnamensis are listed in Table 1. The amino acid residues of these proteins range from 161–172 with a molecular mass of 17,464 – 18,201 Da. Protein C-phycocyanin beta chain had the highest amino acid residues (172) and the highest molecular mass (18,201 Da). Table 2 shows the total number of potential bioactive peptides from P. vietnamensis proteins.

Table 1. Accession numbers for protein database and sequences of protein from seaweed.

Protein (Accession number) Sequence Amino acid length Molecular mass (Da)
R-phycoerythrin alpha chain (P51368) MKSVITTTISAADAAGRFPSSSDLESVQGNIQRAAARLEAAEKLASNHEAV VKEAGDACFAKYSYLKNPGEAGDSQEKVNKCYRDVDHYMRLVNYCLVVGGTGP VDEWGIAGAREVYRTLNLPTSAYVASFAFARDRLCVPRDMSAQAGVEYAGNLDYIINSLC 164 17,698
Phycocyanin alpha subunit (A0A141SF63) MKTPITEAIASADSQGRFLSNGELQAINGRYQRAAASLGAARSLTNNAQRLITGAAQS VYTKFPYVTQMPGPTYASSAIGKAKCARDIGYYLRMVTYCLVVGATGPMDEYLVAG LEEINRSFELSPSWYVEALQYIKGSHGLSGQIGNEANVYLDYAINTLS 162 17,464
Allophycocyanin alpha chain (P59856) MSIVTKSIVNADAEARYLSPGELDRIKSFVLSGQRRLRIAQILTDNRERIVKQGGQQ LFQKRPDVVSPGGNAYGEEMTATCLRDLDYYLRLVTYGIVAGDVTPIEEIGLVGVKE MYNSLGTPISGVAEGVKCMKSVACSLLAGEDSAEAGFYFDYTLGAMQ 161 17,509
C-phycocyanin beta chain (Q1XDB0) MLDAFAKVVAQADARGEFLSNTQLDALSSMVAEGNKRLDVVNKINSNASAIVTNSA RALFAEQPQLIQPGGNAYTSRRMAACLRDMEIVLRYVSYAMIAGDSSVLDDRCLNGLRE TYQALGTPGSSVSVAVQKMKEASVALANDLTGITQGDCSALIAELGSYFDRAAVSVV 172 18,201
Allophycocyanin beta chain (P59857) MQDAITSVINAADVQGKYLDDSSVEKLRGYFQTGELRVRAAATIAANAATIIKESVAKS LLYSDITRPGGNMYTTRRYAACIRDLDYYLRYATYGMLAGDPSILEERVLNGLK ETYNSLGVPIGATIQAILAMKEVTISLVGPDAGKEMGLYFDYICSGLS 161 17,484

Table 2. Total number of potential bioactive peptides from P. vietnamensis proteins.

Proteins Number of activities Number of potential bioactive peptides
DPP-IV inhibitor ACE inhibitor DPP-III inhibitor Anti- oxidative Alpha-glucosidase inhibitor Other activities
R-phycoerythrin alpha chain 15 103 83 15 7 6 26
Phycocyanin alpha subunit 19 99 86 12 18 5 41
Allophycocyanin alpha chain 13 97 77 16 9 3 34
C-phycocyanin beta chain 14 105 71 14 6 2 24
Allophycocyanin beta chain 16 105 84 18 15 2 41
Total number of potential bioactive peptides 509 401 75 55 18 166

ACE, angiotensin-converting enzyme; DPP-III, dipeptidyl peptidase-III; DPP-IV, dipeptidyl peptidase-IV

Table 3 shows the frequency of occurrence (A) and potential biological activity (B) values of bioactive peptides from P. vietnamensis proteins. DPP-IV inhibitor exhibited a higher A value for all selected proteins, followed by ACE inhibitor. Therefore, this study focused on the DPP-IV inhibitor and ACE inhibitor as the main biological activity to be released from P. vietnamensis proteins. The total frequency of occurrences of bioactive fragments (∑A) among five proteins ranged from 1.3081 to 1.6646. Based on the data, the allophycocyanin beta chain showed the highest value followed by the phycocyanin alpha subunit, allophycocyanin alpha chain, r-phycoerythrin alpha chain, and c-phycocyanin beta chain. The highest A value of the DPP-IV inhibitor in the allophycocyanin beta chain was observed to contribute to the majority of ∑A value. The database also presented the potential biological activity of protein (B). In contrast, the B value of the ACE inhibitor was higher than that of the DPP-IV inhibitor.

Table 3. The frequency of occurrence (A) and the potential biological activity (B) of bioactive pepides from P. vietnamensis proteins.

The frequency of occurrence of bioactive peptides (A) (The potential biological activity of bioactive peptides [B])
Protein ∑A DPP-IV inhibitor ACE inhibitor DPP-III inhibitor Antioxidative Alpha-glucosidase inhibitor
R-phycoerythrin alpha chain 1.4635 0.6280 (0.0002) 0.5061 (0.0214) 0.0915 0.0427 0.0366
Phycocyanin alpha subunit 1.6175 0.6111 (0.0001) 0.5309 (0.0625) 0.0741 0.1111 0.0370
Allophycocyanin alpha chain 1.4845 0.6025 (0.0002) 0.4783 (0.0160) 0.0994 0.0559 0.0373
C-phycocyanin beta chain 1.3081 0.6105 (0.0004) 0.4128 (0.0064) 0.0814 0.0349 0.0291
Allophycocyanin beta chain 1.6646 0.6522 (0.0002) 0.5217 (0.0108) 0.1118 0.0932 0.0311

*The value of the potential biological activity of the protein (B) for specific activities was rounded off to the fourth decimal place. ∑A, sum of the frequency of occurrence of bioactive peptides; ACE, angiotensin-converting enzyme; DPP-III, dipeptidyl peptidase-III; DPP-IV, dipeptidyl peptidase-IV

In silico proteolysis of P. vietnamensis proteins

Enzymatic hydrolysis is the most common method to release peptides from parent proteins, enabling them to exhibit biological functions. This method requires the use of one or more enzymes to react with sample proteins to release bioactive peptides. In silico proteolysis was conducted to predict the potential proteolytic enzymes capable of producing peptides with specific biological activity. In this study, selected proteins were subjected individually to 33 different enzymes contained in the BIOPEP-UWM database using an enzyme-action tool (not shown). Among the 33 different enzymes, seven enzymes namely pancreatic elastase, papain, ficin, leukocyte elastase, stem bromelain, calpain 2, and pepsin (pH > 2) were identified as potential candidates for releasing bioactive peptides from the proteins, mainly DPP-IV inhibitors. Hence, these enzymes were selected for further reporting. Enzymatic action generated multiple peptides from each protein. Table 4 shows the number of potential DPP-IV inhibitory and ACE inhibitory fragments to be released from P. vietnamensis proteins by specific enzymes. These peptides, which have been reported and exist in the database, were predominantly dipeptides. In addition, tripeptides (GPV) and tetrapeptides (ASL, AEL, SVYT) were also found. Among the selected enzymes, calpain 2 and pepsin (pH > 2) were predicted to release the most peptide fragments for both DPP-IV inhibitors and ACE inhibitors.

Table 4. The number of potential DPP-IV inhibitors and ACE inhibitors to be released by selected enzymes using BIOPEP’s enzyme-action tool.

Enzyme DPP-IV inhibitor ACE inhibitor
Pancreatic elastase 70 47
Papain 78 52
Ficin 55 38
Leukocyte elastase 62 36
Stem bromelain 77 58
Calpain 2 98 63
Pepsin (pH >2) 109 61

The efficiency of selected enzymes in releasing bioactive peptides was measured based on the frequency of release fragments with a given activity by selected enzymes (AE) and the relative frequency of release fragments with a given activity by selected enzymes (W). Figure 3 illustrates the ∑AE values of DPP-IV inhibitor and ACE inhibitor after undergoing in silico proteolysis using a single and combination of enzymes. Among the single enzyme treatments, calpain 2 was predicted to be the most effective in releasing DPP-IV inhibitory peptides with higher A values, while papain seems to be promising in producing ACE inhibitory peptides. Moreover, pepsin (pH > 2) and stem bromelain were also observed as potential enzymes capable of releasing peptides with both DPP-IV and ACE inhibitory activity. Enzyme-action tool allows the proteolysis to be simulated using a combination of enzymes up to three enzymes. As a result, papain + calpain 2, papain + pepsin (pH > 2), and pancreatic elastase + papain + ficin were the potential combinations predicted to produce DPP-IV inhibitory peptides with higher A values compared to the other enzyme combinations. In contrast, combinations of papain + stem bromelain, papain + pepsin (pH > 2), and pancreatic elastase + papain + ficin had the potential to release ACE inhibitory peptides.

In silico screening and characterization of novel peptides

In this study, unidentified tripeptide and tetrapeptide were screened as novel peptides. The novel peptides could be produced during the digestion of the parent protein (Akbarian et al., 2022) and could possess potent biological activities, allergenic properties, or toxic properties once absorbed into the body cell. Nine tripeptides (QRA, CFA, RDV, DHY, ESV, ACF, AYV, DMS, DYI) and 21 tetrapeptides (RFPS, NHEA, KNPG, QEKV, NKCY, DEWG, RDRL, EVYR, CVPR, NIQR, DACF, DVDH, AQAG, IINS, QGNI, NYCL, SVIT, SNHE, AVVK, EVYR, VPRD) were unreported peptides to be released by the activity of specific enzymes. Narrowing down, seven novel peptides were selected for further investigation as they are potentially bioactive based on the bioactivity score (>0.5) calculated using the PeptideRanker tool, with the most likely being CFA, ACF, and DACF (Table 5). This online tool calculates bioactivity scores based on the neural network model. This model is trained on a dataset containing peptides with experimentally verified bioactivities and inactive peptides. It enables the tool to learn patterns and features that are associated with bioactivity. The model then evaluates the peptide sequence and assigns a bioactivity probability score between 0 and 1, with 0 indicating a low likelihood of bioactivity, while 1 indicates a high likelihood of bioactivity. Commonly, peptides with scores above 0.5 are considered to have significant bioactivity potential (Coscueta et al., 2022).

Table 5. The potential novel peptides to be released from P. vietnamensis proteins.

Peptides CFA (Cys-Phe-Ala) ACF (Ala-Cys-Phe) RFPS (Arg-Phe-Pro-Ser) DEWG (Asp-Glu-Trp-Gly) NYCL (Asn-Tyr-Cys-Leu) CVPR (Cys-Val-Pro-Arg) DACF (Asp-Ala-Cys-Phe)
Potential biological activity ACE/DPP-IV inhibitor ACE inhibitor ACE/DPP-IV inhibitor ACE/DPP-IV inhibitor ACE/DPP-IV inhibitor ACE/DPP-IV inhibitor ACE inhibitor
Bioactivity score 0.941 0.975 0.807 0.516 0.736 0.544 0.923
Enzymes Pancreatic elastaste, leukcoyte elastaste Papain, calpain 2 Pancreatic elastaste Pancreatic elastaste, stem bromelain Leukcoyte elastaste, stem bromelain Papain, calpain 2 Ficin
MW (g/mol) 339.41 339.41 505.57 505.48 511.59 473.59 454.5
Water solubility Poor Poor Good Good Poor Good Good
Isoelectric point 2.9 2.97 10.55 0.62 2.86 9.21 0.76
Net charge at pH 7 −0.1 −0.1 1 −2 −0.1 0.9 −1.1
*Resistance to digestion No No No No No No No
Toxicity Nontoxic Nontoxic Nontoxic Nontoxic Nontoxic Nontoxic Nontoxic
Allergenicity probability Nonallergenic Nonallergenic Allergen Nonallergenic Allergic Nonallergenic Nonallergenic

*Resistance to digestion using enzymes: chymotrypsin-low specificity, chymotrypsin-high specificity, pepsin (pH 1.3), pepsin (pH >2), and trypsin.

Based on the results, the water solubility of CFA, ACF, and NYCL was poor, which may be due to their hydrophobic residues. Interestingly, all peptides exhibited nontoxic and nonallergenic properties, except for RFPS and NYCL. Nevertheless, all of them are predicted to degrade following enzymatic digestion. Furthermore, this study primarily focused on DPP-IV inhibitory activity, and the potential novel peptide was chosen for in vitro analysis. Among the novel peptides, CVPR was chosen to be further characterized in vitro due to its good water solubility, which would facilitate the experiment and indicate its potential availability in human blood after crossing cell membranes. This criterion is important for the peptide to exhibit biological effects in the human body. Although CFA and ACF were most likely to be bioactive based on the bioactivity score (which is closer to 1), their poor water solubility may challenge the experiment and limit peptide transfer into the bloodstream. A specific solvent may be required during analysis and the peptide solution may influence the DPP-IV inhibitory activity making it inconvenient to work with. Generally, the nontoxic and nonallergenic properties predicted from CVPR were important criteria that were considered.

In vitro characterization and stability of DPP-IV inhibitory peptide

The molecule with the sequence CVPR was synthesized in an attempt to evaluate its DPP-IV inhibitory activity and its stability against pH, temperature, and simulated gastrointestinal digestion (SGID). First, the inhibitory activity of the DPP-IV inhibitory peptide was measured at different peptide concentrations (Figure 4a). Based on the result, the percentage of inhibition increased significantly with higher peptide concentrations, demonstrating the influence of peptide concentration on the inhibitory activity. The IC50 of DPP-IV inhibitory peptide was calculated as 1393.61 ± 28.72 µM, similar to Atlantic salmon (Salmo salar) gelatin hydrolysate with IC50 of 1547.15 ± 34.15 µM as reported by Neves et al. (2017). According to Liu et al. (2019), it is not the length of the peptide that influences the DPP-IV inhibitory activity, but the composition and sequence of amino acid residues might correlate with DPP-IV inhibitory properties. Furthermore, the effect of pH ranges from 3.0 to 11.0 on the relative DPP-IV inhibitory activity of the peptide is depicted in Figure 4b. The inhibitory activity was observed to reduce approaching strong alkaline condition (pH > 9.0), indicating that the peptide favors an acidic condition with the highest relative DPP-IV inhibitory activity at pH 3.0 (80.10% ± 4.64) and 5.0 (86.04% ± 3.96). The impacts of different temperatures (30, 50, 70, 90, and 100°C) on the relative DPP-IV inhibitory activity are illustrated in Figure 4c. The highest relative DPP-IV inhibitory activity was observed at 30°C (82.19% ± 3.30). Upon increasing temperature, the inhibitory activity of the peptide was gradually decreased. Lastly, the peptide was subjected to two stages of simulated digestion to determine its stability against the activity of gastrointestinal enzymes. Based on Figure 4d, the relative DPP-IV inhibitory activity of the peptide was maintained before or after in vitro gastrointestinal digestion.

Figure 4. (A) DPP-IV inhibitory activity of peptide at different concentrations; (B) pH stability at pH 3.0, 5.0, 7.0, 9.0, 11.0; (C) Thermal stability at different temperatures (30, 50, 70, 90, 100°C) after 30 min of incubation period; (D) Simulated gastrointestinal stability of peptide against pepsin and pepsin + pancreatin. Data shown are means ± standard deviation of triplicates, different letters indicate significant differences at P < 0.05.

Discussion

In this study, the proteomic analysis revealed P. vietnamenis contained mostly phycobiliproteins. This correlates to a previous study that reported the most abundant protein in red seaweed is phycobiliproteins which consist of phycoerythrin, phycocyanin, and allophycocyanin (Cotas et al., 2020). In the BIOPEP-UWM database, “A” value(s) indicates the frequency of specific proteins containing encrypted bioactive peptides (Minkiewicz et al., 2019). Based on the results, DPP-IV inhibitory activity was predicted as the most potential bioactivity to be released from P. vietnamensis proteins followed by ACE inhibitory activity. Similar values of ∑A (1.129–1.482), ADPP-IV inhibitor (0.586–0.688), and AACE inhibitor (0.336–0.518) were reported from Gracilaria changii (Sharmin et al., 2022). These results are supported by in vitro findings of Malaysian seaweeds that possessed potent DPP-IV inhibitory activity (Chin et al., 2014). In contrast, green seaweed Caulerpa spp. was predicted with high antihypertensive properties with a frequency value of ACE inhibitor in the range of 0.3822–0.4330 than DPP-IV inhibitor with a frequency value of 0.0550–0.0714 (Agirbasli and Cavas, 2017).

DPP-IV is a serine protease that cleaves two amino acids at the N-terminal, proline, and alanine that possess vital functions in the insulin secretion mechanism. The incretin hormones, glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide 1 (GLP-1), are accountable for the role of maintaining the blood glucose level (BGL) by increasing insulin secretion, β-cell mass, decreasing glucagon secretion, and shifting the gastric emptying rate. However, they had a short half-life of less than 2 min and were quickly metabolized by DPP-IV enzyme. DPP-IV inhibition is a primary target in the therapy of type 2 diabetes mellitus (T2DM) which is defined as a failure of the body to govern blood glucose levels efficiently, resulting in impaired insulin secretion as well as insulin resistance. Numerous food proteins have been reported to release DPP-IV inhibitory peptides including from seaweed species such P. palmata, Porphyra dioica, Ulva spp., and Gracilaria opuntia (Cermeño et al., 2019; Cian et al., 2022; Harnedy et al., 2015; Makkar and Chakraborty, 2016). On the other hand, ACE is important in blood pressure management in the renin–angiotensin (RAS) and the kallikrein–kinin systems. The main role of ACE is to convert inactive angiotensin-I to active vasoconstrictor angiotensin-II in the RAS, which results in a rise in blood pressure due to constrictions of blood vessels. The secondary role of ACE is to inactivate the vasodilator bradykinin which is responsible for promoting vasodilation and lower blood pressure. This dual function allows ACE to regulate blood pressure by promoting vasoconstriction (via angiotensin II) and reducing vasodilation (via bradykinin degradation). In people with hypertension, the presence of ACE inhibitors is crucial to lowering angiotensin II levels and increasing bradykinin, resulting in vasodilation and reduced blood pressure. To date, seaweed species Acrochaetium sp. and Gracilariopsis lemaneiformis have been reported to possess antihypertensive activity (Deng et al., 2018; Windarto et al., 2022).

In silico proteolysis was utilized as an initial mining of bioactive peptide production as well as to overcome the challenges of in vitro and in vivo enzymolysis which are time-consuming, costly, and laborious processes (Agyei et al., 2018). In this study, single enzyme treatment was predicted to give a higher score of ∑AE compared to proteolysis with a combination of enzymes. The lower ∑AE values obtained may be due to the action of enzyme combinations that would damage the original inhibitory peptides as mentioned by Klompong et al. (2007), while less or no bioactive peptides may be obtained from extensive proteolysis. The results suggested calpain 2, papain, pepsin (pH > 2), and stem bromelain as potential enzymes to release peptides with DPP-IV and ACE inhibitory activity. The results were consistent with the published study whereby calpain 2 had maximum AE for DPP-IV inhibitor from H. stipulacea (Kandemir-Cavas et al., 2019). Moreover, papain has been identified as an efficient protease for generating ACE inhibitory peptides from Ulva lactuca protein extract (Garcia-Vaquero et al., 2019) and exhibited a great likelihood to release renin inhibitory peptides in silico (Pihlanto and Mäkinen, 2017). Meanwhile, pepsin was also reported to release peptides with high antihypertensive activity (Baba et al., 2021). In silico proteolysis using pepsin (pH > 2) produced the highest number of bioactive fragments including ACE inhibitor and DPP-IV inhibitor from Ulva lactuca. In addition, stem bromelain has been reported to produce a significant amount of peptide fragments (>100), thus the potential to release bioactive peptides from seaweed (Garcia-Vaquero et al., 2019; Sharmin et al., 2022). Eventually, this in silico work helps the researcher to screen for potential enzymes to produce peptides with desired bioactivity.

In the peptide chain, certain amino acids play an important role in inhibiting enzymes such as DPP-IV and ACE due to their specific chemical properties and interactions with enzyme active sites. These amino acids influence the peptide’s binding affinity, stability, and inhibitory activity. Specifically, the presence of proline residues at specific positions will make the peptide resistant to DPP-IV’s and ACE’s cleavage. In contrast, the presence of alanine residues contributes to the peptide’s affinity with the DPP-IV active site. Moreover, hydrophobic amino acids, for example, valine and tryptophan enhance the hydrophobic interactions with DPP-IV and ACE, respectively. Basic amino acids such as arginine and lysine are also critical for DPP-IV inhibition where they form strong ionic or hydrogen bonds with DPP-IV. On the other hand, sulfur-containing amino acids including cysteine and methionine enhance and stabilize the peptide–enzyme interaction. These structural–functional relationships are important for the therapeutic and functional applications of bioactive peptides.

Novel tripeptides and tetrapeptides with significant bioactivity potential were reported in this study, and the ones with potential DPP-IV inhibitors were chosen for in vitro characterization. Additionally, solubility is one of the crucial aspects to influence the peptides’ transfer into the blood, while poor solubility could result in less absorption. Hence, CVPR was chosen to further investigate due to its potency to act as a DPP-IV inhibitor, exhibiting good water solubility, no potential toxicity, and nonallergenic properties. CVPR had proline residues in the peptide chain which may exhibit potent DPP-IV inhibitory activity (Cermeño et al., 2019). In this study, novel peptides were predicted to degrade during digestion. Fundamentally, PepCalc uses simplified models that might not account for the full range of biological factors influencing peptide stability. This model relies on known cleavage motifs and does not fully capture the complex interplay between peptide structure, enzyme specificity, and the environment. Also, it might assume the peptide is fully linear and accessible to enzymes, leading to an overestimation of its susceptibility to cleavage. Hence, it must be aware that this computational tool can sometimes produce false positives for degradation susceptibility due to biases in their training datasets. Nevertheless, this limiting factor could be overcome by an oral delivery system either through the linkage of peptide drugs with lipophilic or hydrophilic moieties, or by encapsulation.

In vitro characterization was carried out to determine the activity and stability of novel peptides associated with the dominant bioactivity predicted in silico which is DPP-IV inhibitory activity. To the best of our knowledge, tetrapeptide CVPR has not been previously identified as a DPP-IV inhibitory peptide. CVPR exhibited IC50 of 1393.61 ± 28.72 µM, lower than that of the proline-containing peptide of PP and GGPAGPAV with IC50 of 4343.48 ± 29.78 and 8139.11 ± 134.68 µM, respectively (Neves et al., 2017). This suggests that P. vietnamensis has higher DPP-IV inhibitory activity than Salmo salar. Nevertheless, P. palmata exhibited potent activity of ILAP, LLAP, and MAGVDHI with IC50 values of 43.40 ± 1.40, 53.67 ± 0.82, 159.37 ±13.67 µM, respectively (Harnedy et al., 2015). Currently, gliptins are the most potent commercial DPP-IV inhibitors approved by the Food and Drug Administration (FDA), while there are 10 DPP-IV inhibitor drugs still undergoing clinical trials (Dahlén et al., 2022). According to Gupta et al. (2018), the efficiency of gliptins measured as IC50 are as of the followings; sitagliptin = 0.684 µM, saxagliptin = 0.707 µM, and vildagliptin = 2.286 µM. Figure 5 compares IC50 values between peptide and the established DPP-IV inhibitors. It is noted that lower IC50 values indicate potency.

Figure 5. The comparison of IC50 values between peptide and the established DPP-IV inhibitors.

CVPR developed potential inhibitory effects against DPP-IV compared to other peptides with much higher IC50 values such as spinach dry-cured ham, milk, and skin or scale collagen-derived peptide with IC50 of more than 10,000, 11,000, and 20,000 µM, respectively (Gallego et al., 2014; Hatanaka et al., 2014; Nongonierma and FitzGerald, 2013). It is important to note that in vitro IC50 does not always predict in vivo efficacy accurately. The physiological environment can significantly influence a peptide’s activity, and factors such as digestion, absorption, distribution, metabolism, and excretion play crucial roles in determining the actual bioactivity within the body. Generally, peptides with higher IC50 values may still contribute to cumulative inhibitory effects when present in complex food matrices, despite their lower individual potency. In the development of nutraceutical products, the activity of CVPR provides a natural and less potent alternative to synthetic drugs, making them suitable for preventive approaches in managing prediabetes or mild hyperglycemia. Interestingly, CVPR may also serve as lead compounds for developing more potent and specific DPP-IV inhibitors and can be used in combination therapies to enhance the efficacy of existing antidiabetic medications.

Nevertheless, analysis of CVPR in animal models and in vivo is required to validate its efficiency as a DPP-IV inhibitory peptide for further practical application and commercialization. Previously studied, various dietary proteins found that certain plant-based proteins and hemoglobin exhibited higher DPP-IV inhibitory activity in vivo compared to in vitro results. For example, the plasma DPP-IV activity in rats was decreased significantly in the presence of peas, gluten, hemoglobin, and gelatin compared to DPP-IV inhibitory activity in vitro (Fleury et al., 2022). Moreover, walnut and tea-derived peptides were observed to improve glucose metabolism in mice in vivo (Lu et al., 2019; Mu et al., 2024) This suggests that peptides derived from these proteins may have enhanced efficacy in living systems, potentially due to factors such as improved absorption or stability.

The stability of the peptide against pH, temperature, and SGID was examined. The results of pH stability were similar to the previous finding by Xiang et al. (2021) where peptides derived from discarded shrimp heads showed a significant decrease in bioactivity under an adverse alkaline condition. This may be because the peptide is further degraded into an inactive fragment under an extremely alkaline condition, or the pH adjustment might alter the charges in the peptide which eventually affects the activity of the peptide. Furthermore, the thermal stability results are comparable to that of walnut-derived peptides which exhibited good DPP-IV inhibitory activity across different temperatures (Kong et al., 2021). Stability testing of bioactive peptides concerning pH and temperature is crucial for their successful application, especially in nutraceuticals. By testing bioactive peptides under varying pH and temperatures, stability ensures that peptides retain their functional properties, for example, the ability to inhibit enzymes such as DPP-IV or ACE under diverse physiological and processing conditions. Moreover, the thermal and pH stability extends the shelf life of the products, since peptides are often exposed to high temperatures during processing for example, pasteurization, and storage, and may encounter varying pH levels in formulations for example, acidic in beverages and basic in certain supplements. Furthermore, nontoxicity and nonallergenicity are critical for the safe and effective use of peptides in functional foods. Functional foods are intended to provide health benefits beyond basic nutrition, and any potential adverse effects, such as toxicity or allergenicity, can undermine their purpose and marketability. Nontoxic peptides ensure the food does not cause harmful effects, such as cytotoxicity, organ damage, or disruption of normal physiological functions, while nonallergenic peptides minimize the risk of allergic reactions, which can range from mild symptoms to severe. Additionally, these properties ensure regulatory compliance and product acceptance due to functional integrity suitable for human consumption.

For the bioactive peptides to exert their physiological effects within the gastrointestinal tract, circulatory system, and specific cells, the bioactive peptides are anticipated to be resistant to hydrolysis by digestive enzymes. Figure 4 shows that no significant difference was observed in the DPP-IV inhibitory activity after simulated digestion, indicating that the peptide may resist gastrointestinal digestion. This finding aligns with the results of Harnedy et al. (2015), whereby ILAP, LLAP, and MAGVDHI from P. palmata were observed to survive an in vitro simulated digestion. Other food-derived peptides that have been reported to resist gastrointestinal transit include VPLVM and VPYPQ (Pei et al., 2022; Zheng et al., 2019). These findings suggest that most VP-containing peptides may exhibit stability following SGID. Comparing both in silico and in vitro results, discrepancies may arise between the two approaches. This is because an in silico study is a prediction made based on the in silico tools, while an in vitro study gives a more real finding in in vitro conditions. CVPR might be predicted to degrade by the PepCalc app due to the presence of residues like arginine, which are common targets for proteases. However, the structural features of CVPR for example, the rigidity introduced by proline, potential protective effects of cysteine, and overall conformation may render it more resistant to enzymatic digestion in vitro than computational models can predict. Also, in silico tools often simulate digestion under standardized conditions for maximum degradation efficiency. Meanwhile, the differences in in vitro conditions can influence enzyme activity and the interaction of CVPR with other components in the digestion medium may reduce enzyme access to cleavage sites. Hence, CVPR is not affected by SGID and thus maintains its DPP-IV inhibitory activity. This highlights the importance of validating in silico predictions with in vitro and in vitro assays to account for real-world complexities.

Conclusions

This study found that DPP-IV and ACE inhibitory activities were the most promising bioactivities predicted to be released from P. vietnamensis proteins, suggesting their potential to inhibit enzymes involved in the pathologies of diabetes and hypertension. To the best of our knowledge, this is the first study on DPP-IV and ACE inhibitory activity from P. vietnamensis proteins via an in silico approach. This approach has provided valuable insights for the researcher to predict the most potential bioactivity from P. vietnamensis. It has narrowed the selection of potential proteolytic enzymes that could release peptides with a high frequency of DPP-IV and ACE inhibitory activity. Further research on mechanistic studies should be carried out to elucidate the binding interactions between peptides and DPP-IV and ACE through molecular docking and simulation. Also, conducting in vivo validation through animal and human studies to assess the real-world efficacy, safety, and pharmacokinetics of these peptides would be useful in designing future work. P. vietnamensis is an attractive renewable resource making it potential for commercial production. It is abundant and sustainable where they are widely available and can be cultivated sustainably in marine environments, with minimal reliance on freshwater or arable land. Furthermore, P. vietnamensis offers a wide range of bioactive compounds potential for functional foods, nutraceuticals, and pharmaceuticals. Currently, there are growing consumer preference for natural and sustainable bioactive compounds, thus boosting the commercial viability of seaweed peptides. Nevertheless, the production of bioactive peptides of P. vietnamensis at a commercial scale while maintaining their activity and cost-effectiveness is a significant challenge. Hence, the strategies for scalability need to be considered including the optimization of cultivation by the use of biotechnological advances, such as selective breeding and genetic engineering, to improve seaweed yields and peptide content. Moreover, the adoption of advanced scalable extraction techniques to improve yield and reduce costs, as well as the exploration of bioreactor-based methods for controlled peptide production is one of the crucial aspects to be taken into account. Finally, the collaborative efforts between academic institutions, industry, and governments to fund research and accelerate the development of scalable methods would benefit both industries and consumers worldwide.

Authors’ Contributions

All the data is available with the authors and shall be provided upon request.

Data Availability Statement

All the data is available with the authors and shall be provided upon request.

Conflicts of Interest

The authors declare no conflicts of interest relevant to this article’s content. The funders had no role in the design of the study, collection, analyses or interpretation of data, writing of the manuscript or in the decision to publish the results.

Funding

This research was funded by the Ministry of Higher Education Malaysia through the Fundamental Research Grant Scheme (FRGS) (FRGS/1/2018/WAB01/UMT/02/4; Vot. No. 59505).

REFERENCES

Agirbasli, Z., and Cavas, L., 2017. In silico evaluation of bioactive peptides from the green algae Caulerpa. Journal of Applied Phycology 29(3): 1635–1646. 10.1007/s10811-016-1045-7

Agyei, D., Tsopmo, A., and Udenigwe, C.C., 2018. Bioinformatics and peptidomics approaches to the discovery and analysis of food-derived bioactive peptides. Analytical and Bioanalytical Chemistry 410(15): 3463–3472. 10.1007/s00216-018-0974-1

Akbarian, M., Khani, A., Eghbalpour, S., and Uversky V.N., 2022. Bioactive peptides: Synthesis, sources, applications, and proposed mechanisms of action. International Journal of Molecular Sciences 23(3): 1445. 10.3390/ijms23031445

Awang, A., Karim, R., and Mitsui, T., 2010. Proteomic analysis of Theobroma cacao pod husk. Journal of Applied Glycoscience 57(4): 245–264. 10.5458/jag.57.245

Baba, W.N., Baby, B., Mudgil, P., Gan, C.Y., Vijayan, R., and Maqsood, S., 2021. Pepsin generated camel whey protein hydrolysates with potential antihypertensive properties: Identification and molecular docking of antihypertensive peptides. LWT 143 (1): 111135. 10.1016/j.lwt.2021.111135

Bhatia, S., Sharma, A.K., De La Cruz, C.B., Chaugule, B., and Al-Harrasi, A., 2021. Nutraceutical, antioxidant, antimicrobial properties of Pyropia vietnamensis (tanaka et Pham-Hong Ho) J.E. Sutherl. et monotilla. Current Bioactive Compounds 17 (2): 151–164. 10.2174/1573407216666200227094935

Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72 (1–2): 248–254. 10.1016/0003-2697(76)90527-3

Carpentier, S.C., Witters, E., Laukens, K., Deckers, P., Swennen, R., and Panis, B., 2005. Preparation of protein extracts from recalcitrant plant tissues: An evaluation of different methods for two-dimensional gel electrophoresis analysis. Proteomics 5(10): 2497–2507. 10.1002/pmic.200401222

Cermeño, M., Stack, J., Tobin, P.R., O’Keeffe, M.B., Harnedy, P.A., Stengel, D.B., et al. 2019. Peptide identification from a Porphyra dioica protein hydrolysate with antioxidant, angiotensin converting enzyme and dipeptidyl peptidase IV inhibitory activities. Food & Function 10(6): 3421–3429. 10.1039/c9fo00680j

Chin, Y.X., Lim, P.E., Maggs, C.A., Phang, S.M., Sharifuddin, Y., and Green, B.D., 2014. Anti-diabetic potential of selected Malaysian seaweeds. Journal of Applied Phycology 2(5): 2137–2148. 10.1007/s10811-014-0462-8

Cian, R.E., Nardo, A.E., Garzón, A.G., Añon, M.C., and Drago, S.R., 2022. Identification and in silico study of a novel dipeptidyl peptidase IV inhibitory peptide derived from green seaweed Ulva spp. hydrolysates. LWT 154: 112738. 10.1016/j.lwt.2021.112738

Coscueta, E.R., Batista, P., Gomes, J.E., da Silva, R., and Pintado, M.M., 2022. Screening of novel bioactive peptides from goat casein: In silico to in vitro validation. International Journal of Molecular Sciences 23(5): 2439. 10.3390/ijms23052439

Cotas, J., Leandro, A., Pacheco, D., Gonçalves, A.M., and Pereira, L., 2020. A comprehensive review of the nutraceutical and therapeutic applications of Red Seaweeds (Rhodophyta). Life (Basel, Switzerland) 10(3): 19. 10.3390/life10030019

Dahlén, A.D., Dashi, G., Maslov, I., Attwood, M.M., Jonsson, J., Trukhan, V., et al. 2022. Trends in Antidiabetic Drug Discovery: FDA approved drugs, new drugs in clinical trials and global sales. Frontiers in Pharmacology. 12: 807548. 10.3389/fphar.2021.807548

Deng, Z., Liu, Y., Wang, J., Wu, S., Geng, L., Sui, Z., et al. 2018. Antihypertensive effects of two novel angiotensin I-converting enzyme (ACE) inhibitory peptides from Gracilariopsis lemaneiformis (Rhodophyta) in spontaneously hypertensive Rats (SHRs). Marine Drugs 16(9): 299. 10.3390/md16090299

El-Beltagi, H.S., Mohamed, A.A., Mohamed, H.I., Ramadan, K.M., Barqawi, A.A., and Mansour, A.T., 2022. Phytochemical and potential properties of seaweeds and their recent applications: A review. Marine Drugs 20(6): 342. 10.3390/md20060342

Fleury, L., Deracinois, B., Dugardin, C., Nongonierma, A.B., FitzGerald, R.J., Flahaut, C., et al. 2022. In vivo and in vitro comparison of the DPP-IV inhibitory potential of food proteins from different origins after gastrointestinal digestion. International Journal of Molecular Sciences 23(15): 8365. 10.3390/ijms23158365

Gallego, M., Aristoy, M.C., and Toldrá, F., 2014. Dipeptidyl peptidase IV inhibitory peptides generated in Spanish dry-cured ham. Meat Science. 96(2): 757–761. 10.1016/j.meatsci.2013.09.014

Garcia-Vaquero, M., Mora, L., and Hayes, M., 2019. In vitro and in silico approaches to generating and identifying angiotensin-converting enzyme I inhibitory peptides from green macroalga Ulva lactuca. Marine Drugs 17(4): 204. 10.3390/md17040204

Gupta, A., Jacobson, G.A., Burgess, J.R., Jelinek, H.F., Nichols, D.S., Narkowicz, C.K., et al. 2018. Citrus bioflavonoids dipeptidyl peptidase-4 inhibition compared with gliptin antidiabetic medications. Biochemical and Biophysical Research Communications 503(1): 21–25. 10.1016/j.bbrc.2018.04.156

Harnedy, P.A., O’Keeffe, M.B., and FitzGerald, R.J., 2015. Purification and identification of dipeptidyl peptidase (DPP) IV inhibitory peptides from the macroalga Palmaria palmata. Food Chemistry 172: 400–406. 10.1016/j.foodchem.2014.09.083

Hatanaka, T., Kawakami, K., and Uraji, M., 2014. Inhibitory effect of collagen-derived tripeptides on dipeptidylpeptidase-IV activity. Journal of Enzyme Inhibition and Medicinal Chemistry 29(6): 823–828. 10.3109/14756366.2013.858143

Hayes, M., and Tiwari, B., 2015. Bioactive carbohydrates and peptides in foods: An overview of sources, downstream processing steps and associated bioactivities. International Journal of Molecular Sciences 16(9): 22485–22508. 10.3390/ijms160922485

Kandemir-Cavas, C., Pérez-Sanchez, H., Mert-Ozupek, N., and Cavas, L., 2019. In silico analysis of bioactive peptides in invasive sea grass Halophila stipulacea. Cells 8(6): 557. 10.3390/cells8060557

Kinter, M., and Sherman, N.E., 2005. Protein sequencing and identification using tandem mass spectrometry. New York: John Wiley & Sons.

Klompong, V., Benjakul, S., Kantachote, D., and Shahidi, F., 2007. Antioxidative activity and functional properties of protein hydrolysate of yellow stripe trevally (Selaroides leptolepis) as influenced by the degree of hydrolysis and enzyme type. Food Chemistry 102(4): 1317–1327.

Kong, X., Zhang, L., Song, W., Zhang, C., Hua, Y., Chen, Y., et al. 2021. Separation, identification and molecular binding mechanism of dipeptidyl peptidase IV inhibitory peptides derived from walnut (Juglans regia L.) protein. Food Chemistry 347: 129062. 10.1016/j.foodchem.2021.129062

Liu, R., Cheng, J., and Wu, H., 2019. Discovery of food-derived dipeptidyl peptidase IV inhibitory peptides: A Review. International Journal of Molecular Sciences 20(3): 463. 10.3390/ijms20030463

Lu, Y., Lu, P., Wang, Y., Fang, X., Wu, J., and Wang, X., 2019. A novel dipeptidyl peptidase IV inhibitory tea peptide improves pancreatic β-cell function and reduces α-cell proliferation in streptozotocin-induced diabetic mice. International Journal of Molecular Sciences 20(2): 322. 10.3390/ijms20020322

Makkar, F., and Chakraborty, K., 2016. Antidiabetic and anti-inflammatory potential of sulphated polygalactans from red seaweeds Kappaphycus alvarezii and Gracilaria opuntia. International Journal of Food Properties 20(6): 1326–1337. 10.1080/10942912.2016.1209216

Minkiewicz, P., Iwaniak, A., and Darewicz, M., 2019. BIOPEP-UWM database of bioactive peptides: Current opportunities. International Journal of Molecular Sciences 20(23): 5978. 10.3390/ijms20235978

Mu, X., Li, D., Xiao, R., Guan, K., Ma, Y., Wang, R., et al. 2024. Discovery of novel dipeptidyl peptidase-IV inhibitory peptides derived from walnut protein and their bioactivities in vivo and in vitro. Current Research in Food Science 9: 100893. 10.1016/j.crfs.2024.100893

Neves, A.C., Harnedy, P.A., O’Keeffe, M.B., Alashi, M.A., Aluko, R.E., and FitzGerald, R.J., 2017. Peptide identification in a salmon gelatin hydrolysate with antihypertensive, dipeptidyl peptidase IV inhibitory and antioxidant activities. Food Research International 100(Pt 1): 112–120. 10.1016/j.foodres.2017.06.065

Nongonierma, A.B., and FitzGerald, R.J., 2013. Inhibition of dipeptidyl peptidase IV (DPP-IV) by tryptophan containing dipeptides. Food & Function 4(12): 1843–1849. 10.1039/c3fo60262a

Pei, J., Liu, Z., Pan, D., Zhao, Y., Dang, Y., and Gao, X., 2022. Transport, stability, and in vivo hypoglycemic effect of a broccoli-derived DPP-IV inhibitory peptide VPLVM. Journal of Agricultural and Food Chemistry 70(16): 4934–4941. 10.1021/acs.jafc.1c08191

Peñalver, R., Lorenzo, J.M., Ros, G., Amarowicz, R., Pateiro, M., and Nieto, G., 2020. Seaweeds as a functional ingredient for a healthy diet. Marine Drugs 18(6): 301. 10.3390/md18060301

Pihlanto, A., and Mäkinen, S., 2017. The function of renin and the role of food-derived peptides as direct renin inhibitors. In: Tolekova, A.N. Renin-angiotensin system—Past, present and future. InTech. 10.5772/intechopen.69513

Sharmin, K.N., Amiza, M.A., Ahmad, F., Razali, S.A., and Hashim, F., 2022. In silico analysis of Gracilaria changii proteins for potential bioactive peptides. IOP Conference Series: Earth and Environmental Science 967(1): 12017. 10.1088/1755-1315/967/1/012017

Sultana, F., Wahab, M.A., Nahiduzzaman, M., Mohiuddin, M., Iqbal, M.Z., Shakil, A., et al. 2023. Seaweed farming for food and nutritional security, climate change mitigation and adaptation, and women empowerment: A review. Aquaculture and Fisheries 8(5): 463–480. 10.1016/j.aaf.2022.09.001

Walsh, D.J., Bernard, H., Murray, B.A., MacDonald, J., Pentzien, A.K., Wright, G.A., et al. 2004. In vitro generation and stability of the lactokinin β-lactoglobulin fragment. Journal of Dairy Science 87(11): 3845–3857. 10.3168/jds.S0022-0302(04)73524-9

Windarto, S., Lee, M.C., Nursyam, H., and Hsu, J.L., 2022. First report of screening of novel angiotensin-I converting enzyme inhibitory peptides derived from the red alga Acrochaetium sp. Marine Biotechnology (NY) 24(5): 882–894. 10.1007/s10126-022-10152-w

Xiang, X., Lang, M., Li, Y., Zhao, X., Sun, H., Jiang, W., et al. 2021. Purification, identification and molecular mechanism of dipeptidyl peptidase IV inhibitory peptides from discarded shrimp (Penaeus vannamei) head. Journal of Chromatography B, Analytical Technologies in the Biomedical and Life Sciences 1186: 122990. 10.1016/j.jchromb.2021.122990

Zheng, L., Xu, Q., Lin, L., Zeng, X.-A., Sun, B., and Zhao, M., 2019. In vitro metabolic stability of a casein-derived dipeptidyl peptidase-IV (DPP-IV) inhibitory peptide VPYPQ and its controlled release from casein by enzymatic hydrolysis. Journal of Agricultural and Food Chemistry 67(38): 10604–10613. 10.1021/acs.jafc.9b03164

Supplementary