https://orcid.org/0000-0001-8044-9446
Figures
Abstract
Inhibitors of enzymes involved in carbohydrate digestion may be a potential option for glycemic control in Diabetes Mellitus. This study aimed to evaluate the effect of the trypsin inhibitor isolated from tamarind seed (Tamarindus indica L.) (TTI) on α-amylase. After confirmation of the obtaining and characterization of the TTI, the in vitro inhibitory activity of the TTI against α-amylase was analyzed. The interaction of the modeled structures’ theoretical TTI (TTIp 56/287) and five of its derived peptides with α-amylase was also evaluated in silico using Docking and Molecular Dynamics, and their functional properties were examined. The Interaction Potential Energy (IPE) and the main interactions of the peptide-α-amylase complex were described using three-dimensional representations. TTI presented 100% antitryptic activity and a molecular mass of approximately 21 kDa. In vitro, inhibition of α-amylase was higher than 37%. These results were corroborated by computational analyses, which demonstrated strong interaction between the TTIp 56/287 complex and its peptides with the enzyme. The Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) analyses showed good stability. IPE was −705.08 kJ/mol for DTVHDTDGQVPL and −584.11 kJ/mol for TIAPACAPKPAR. Electrostatic interactions stand out, especially the salt bridge, between the main residues that interacted in the complexes (DTVHDTDGQVPL, TIAPACAPKPAR, and TVSQTPIDIPIGLPVR). Additionally, the bioactive potential predicted two candidates with good stability, a long half-life, and bioactivity in an intestinal simulation environment. This is the first report of tamarind trypsin inhibitor or its peptides inhibiting α-amylase. Thus, the amino acid sequences DTVHDTDGQVPL and TIAPACAPKPAR were revealed as candidates that could be tested for action against α-amylase and possibly for glycemic control.
Citation: de Souza LAL, de Souza Sena CP, de Macêdo Oliveira FC, Serquiz RP, Aguiar AJFC, de Souza Nascimento AM, et al. (2025) In vitro and molecular modeling insights into α-amylase inhibition by tamarind seed-derived trypsin inhibitor: Implications for hyperglycemic control. PLoS One 20(9): e0333289. https://doi.org/10.1371/journal.pone.0333289
Editor: Yusuf Oloruntoyin Ayipo, Kwara State University, NIGERIA
Received: June 25, 2025; Accepted: September 11, 2025; Published: September 29, 2025
Copyright: © 2025 de Souza et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Ours submission contains all raw data required to replicate the results of your study (Supporting information - Compressed/ZIP File Archive). All relevant data are within the paper and its Supporting information files.
Funding: This project was funded by the National Council for Scientific and Technological Development (CNPq), grant number 303094/2022-2, and by the Coordination for the Improvement of Higher Education Personnel (CAPES – Finance Code 001).The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Obesity is considered a chronic non-communicable disease (NCD) and has become a global public health problem [1]. Lifestyle (diet and physical activity) is one of the main factors that influence the progression of obesity [2]. Additionally, this condition can lead to excessive fat storage and an increase in adipose tissue, which leads to a moderate chronic inflammatory process, which triggers a range of complications, including insulin resistance (IR), type 2 Diabetes Mellitus (T2DM), cardiovascular diseases, cancer, and hepatic steatosis, among other related diseases [3].
Diabetes Mellitus (DM) affects people of all ages, and over the years, the number of cases worldwide has been increasing dramatically. By 2045, it is estimated that 783 million people worldwide will have the disease, affecting approximately 1 in 8 adults, including type 1 DM (T1DM) and T2DM, as well as other unclassified types of Diabetes [4]. T1DM and T2DM are the most common types related to the absence of insulin production and gradual loss of adequate insulin secretion, respectively. This fact results in difficulty in controlling blood glucose [5]. Maintaining control of DM has been one of the most significant challenges, and therefore, making regular physical activity and a healthy diet are necessary for satisfactory blood glucose control [6].
Nevertheless, drug therapy is often indispensable. On the other hand, undesirable effects, such as nausea, abdominal distension, diarrhea, and liver damage, which can be caused by the use of a variety of therapeutic agents, are consolidated in the treatment of DM, such as acarbose, voglibose, and miglitol [6,7]. This drives the search for less aggressive alternatives, preferably natural and with a low risk of toxicity [7]. From this perspective, dietary proteins stand out as excellent sources of bioactive peptides. These peptides, which are products of protein proteolytic hydrolysis, have been highlighted in the food and health industries as they demonstrate versatility in treating diseases [8,9]. Therefore, identifying peptides with these properties may be a promising strategy for treating DM.
Considering that the increase in blood glucose, common in T2DM, is related to the carbohydrate digestion process, enzyme inhibitors involved in this process have also been evaluated to control or reduce the glycemic response [10]. Among the protease inhibitors, the trypsin inhibitor isolated from tamarind seeds has been the target of research conducted by the Nutrition and Bioactive Substances for Health (NutriSBioativoS) research group at UFRN, Brazil, in various forms, including partially purified (TTI), purified (TTIp), and nanoencapsulated (ECW) forms. It has shown bioactive potential in preclinical studies [11] and significant effects on blood glucose control [12,13].
Regarding TTI, Medeiros et al. [14] determined the amino acid sequence of this inhibitor, as well as its three-dimensional structure definition. In the same study, the theoretical model of TTI, number 56, conformation number 287 (TTIp 56/287), demonstrated the best stability, which allowed the performance of in silico analyses. Subsequently, in silico analysis revealed that TTI and its derived peptides have great potential for treating T2DM as they interact with the insulin receptor (IR) [12,14]. Furthermore, in a preclinical study, TTI reduced fasting blood glucose levels in Wistar rats with T2DM [12]. However, it was unclear how TTI acted, and it remains unknown whether it can inhibit α-amylase, a recognized therapeutic target for the treatment of T2DM. Although studies on plant-derived α-amylase inhibitors exist [9,10], the use of trypsin inhibitors extracted from tamarind seeds (Tamarindus indica L.) remains little explored in the literature.
In this context, in silico studies have emerged as a promising tool for identifying and validating therapeutic targets, as well as discovering bioactive molecules with a wide range of health applications [12]. The study combines in vitro assays with computational modeling (docking and molecular dynamics), which strengthens the robustness of the results and aims to prospect peptides derived from TTI. Most studies in this area are limited to basic laboratory tests or computational modeling alone; few integrate both in a complementary manner. This study investigated the possibility that TTI and/or its peptides inhibit α-amylase in vitro and interact with it in silico, a recognized therapeutic target for controlling T2DM. To the best of our knowledge, this is the first report of tamarind trypsin inhibitor or its peptides inhibiting α-amylase.
2. Methodology
2.1. Obtaining trypsin inhibitor from tamarind seeds (TTI)
The tamarind fruit (Tamarindus indica L.) was purchased from a local store in Natal/RN, Brazil, and registered in the National System for Genetic Heritage Management and Associated Knowledge (SisGen) under the number AF6CE9C. The TTI was obtained in accordance with the recommendations of Carvalho et al. (2016) [15].
The Bradford (1976) [16] methodology was used for protein quantification. The test was performed in triplicate to evaluate the inhibition against trypsin, as described by Kakade, Simons, and Liener (1969) [17]. To assess the degree of purity and estimate the molecular mass of the CE (crude extract), F1 (Fraction 1), F2 (Fraction 2), and TTI proteins, the methodology developed by Laemmli (1970) [18] was used.
2.2. In vitro inhibition of α-amylase
The inhibitory activity of TTI against α-amylase was performed according to the methodology initially described by Telagari and Hullatti (2015) [19] and adapted by Oliveira et al. (2024) [20]. In a 96-well microplate, a reaction mixture containing 15 μL of α-amylase (0.125 mg/mL), 30 μL of sodium phosphate buffer (0.1 M, pH = 6.9) and 30 μL of samples (at concentrations of 0.3, 0.6 and 1.5 mg/mL) was added, which were pre-incubated at 37 °C for 20 minutes. After preincubation, 30 μL of 0.125% soluble starch in 0.1 M sodium phosphate buffer (pH 6.9) was added as a substrate, and the mixture was incubated for another 30 min at 37 °C. Then, 75 μL of 3,5-dinitrosalicylic acid solution was added and heated at 100 °C in a water bath for 10 min. After adding 150 μL of water, the absorbance was measured at 540 nm. In the negative control, samples were not added, and the reaction volume was completed with the same buffer used in the test. Three acarbose solutions (at the same concentrations as the samples, 0.3, 0.6, and 1.5 mg/mL) were used as positive controls. The concentration range was based on literature [19,20]. The percentage of α-amylase inhibition was calculated using Equation 1.
2.3. In silico study of the interaction of TTIp 56/287 and its derivative peptides with α-amylase
2.3.1. Molecular docking between TTI 56/287 and α-amylase.
The interaction between the trypsin inhibitor isolated from tamarind seeds (TTI) and the α-amylase enzyme was analyzed computationally in this study. For this purpose, the theoretical model number 56 and conformation number 287 of TTI (TTIp 56/287) were used [14]. The molecular model of α-amylase was obtained from the RCSB Protein Data Bank, PDB ID 5VA9 [21].
Molecular docking was performed to predict the preferred conformations of the ligand molecules [TTIp 56/287 in the binding site of the target macromolecule (α-amylase)]. The binding site that anchors the inhibitors was defined based on the PDB ID 5VA9 structure obtained by Goldbach et al. (2019) [21]. Molecular docking was performed to predict the preferred conformations of the ligand molecules [TTIp 56/287 in the binding site of the target macromolecule (α-amylase)]. The binding site that anchors the inhibitors was defined based on the PDB ID 5VA9 structure obtained by Goldbach et al. (2019) [21]. Docking was conducted with the High Ambiguity Driven Biomolecular DOCKing server (HADDOCK 2.4), which has been extensively applied to protein–peptide systems [https://doi.org/10.1002/prot.25802]. The resulting complexes were ranked according to the HADDOCK Score (HS), a weighted combination of van der Waals, electrostatic, desolvation, and restraint violation energies, together with a buried surface area term. Complexes with lower HS values were considered to represent more favorable binding poses and were selected for further analyses [22].
Conformational models of the TTIp 56/287 and α-amylase interaction were derived from ten molecular docking simulations, systematically exploring the enzyme’s binding site by varying the ligand’s amino acid sequence. The objective of this strategy was to contemplate the surfaces with the greatest steric probability of coupling occurring.
2.3.2. Obtaining, predicting, modeling, and selecting three-dimensional structures of peptides derived from TTIp 56/287.
Peptides were generated in silico by enzymatic cleavage of the protein sequence reported by Gomes et al. (2024) [23], using trypsin and chymotrypsin virtual digestion via the Peptide Cutter tool (https://www.expasy.org/resources/peptidecutter). The molecular structures of the resulting peptide sequences were modeled using the Pep-Fold3 server [24]. The three-dimensional structures of the peptides containing five or more amino acid residues were refined and optimized through molecular dynamics simulations using the GROningen Machine for Chemical Simulations (GROMACS) software [25], version 2023.1, implemented with the CHARMM36 force field [26] and employing the Transferable Intermolecular Potential 3 Point (TIP3P) water model [27].
Peptides were then subjected to site-directed molecular docking against the α-amylase binding site. Based on an arbitrary criterion, the three peptides with the best HS scores were selected for subsequent molecular dynamics simulations. To complete a set of five peptides for these simulations (arbitrary choice), two additional peptides, predicted to possess antidiabetic bioactivity, were selected using the AntiDMPred computational predictor [28], a machine learning-based tool.
2.3.4. Molecular dynamics simulation of TTIp 56/287-derived peptides and α-amylase.
Molecular dynamics simulations were performed using the GROningen Machine for Chemical Simulations (GROMACS) software [25], version 2023.1, implemented with the CHARMM36 force field [26]. The complex systems were solvated in a dodecahedral box using the Transferable Intermolecular Potential 3 Point (TIP3P) water model [27]. The systems were neutralized by adding Na+ and Cl- counterions at a concentration of 0.15 mol/L. The Leap-Frog algorithm solved the motion equations [29] with an integration interval of 2.0 fs. Long-range interactions were modeled using the Particle-Mesh Ewald sum (PME) [30] with a cut-off of 1.2 nm. Van der Waals interactions were also calculated using the same limit. This value was chosen to ensure that the radius did not extend beyond the distance between the protein-ligand complex and the walls of the simulation box, which is essential for accurately capturing long-range interactions. Bonds involving hydrogen atoms were constrained using the Linear Constraint Solver for Molecular Simulations (LINCS) algorithm [31].
Systems were minimized by the steepest descent algorithm for 10,000 steps with a tolerance of 10 kJ mol − 1 nm − 1 and equilibrated in two steps of 1 ns, the first in the NVT ensemble (closed systems with constant volume and temperature) and the second in the NPT ensemble (closed systems with constant pressure and temperature). The temperature was controlled at 310.15 K (37 °C) by the V-rescale thermostat [32], while the pressure was controlled by the Parrinello–Rahman barostat [33] (for the NPT ensemble). The total simulation time in the production dynamics was 300 ns for all systems. This period of time was determined through Root Mean Square Deviation (RMSD) analysis. The simulation was run until the RMSD reached a steady state, characterized by its oscillation around a mean value without any significant upward or downward trend. A minimum stability period of 200 nanoseconds (ns) was used to ensure sufficient data for statistical analysis. The simulations were conducted using physiological conditions to model the biological system accurately.
2.3.5. Analysis of molecular dynamics simulations.
The stability of the complexes was evaluated during and after the molecular dynamics simulations through Root Mean Square Deviation (RMSD), Root Mean Square Fluctuation (RMSF), and Interaction Potential Energy (IPE), as well as analysis of B-factor structures. RMSF, IPE, and B-factor were calculated from the final 50 ns of the simulation. The interaction analysis included the calculation of the IPE and a description of the main interactions: intermolecular hydrogen bonds (conventional and carbon), electrostatic (charge-charge, salt bridge, pi–cation, pi–anion), and hydrophobic (pi-pi, pi–pi-t, amide–pi, alkyl, pi–sigma and pi–alkyl).
Images of peptide-protein interactions were produced by BIOVIA Discovery Studio software and the 3D Protein Imaging server [34]. Graphs were generated by Graphing, Advanced Computation, and Exploration of Data (Grace) [33].
2.3.6. Prediction of bioactivity and cell penetration capacity, half-life in gut-like environment.
Peptides selected for computational modeling with length > 10 aa (since the responsible server requires sequences longer than 10 aa) were screened for their bioactive potential using the PeptideRanker bioactive peptide prediction server [35]. Peptides with a score ≥ 0.5 in PeptideRanker were classified as potentially bioactive. The membrane translocation potential of the peptides was evaluated using the CellPPD online server [36].
The computational prediction of the half-life (MV) of the peptides was performed via the HLP server, and the time and stability were verified. Peptides with MV < 0.1 seconds were classified as ‘low stability,’ those with MV > 0.1 and < 1.0 seconds were classified as ‘normal stability’, and those with MV > 1.0 seconds were classified as ‘high stability.’
2.4. Statistical analysis
Data were expressed as mean ± standard deviation (SD) of three independent determinations. Statistical analyses were performed using Analysis of Variance (ANOVA) followed by Tukey’s post hoc test. Results were considered statistically significant at p < 0.05 (95% confidence level), using GraphPad Prism version 9.4.0.
3. Results
3.1. Obtaining trypsin inhibitor isolated from tamarind
The chromatographic profile of F2 (protein fraction saturated with 30–60% ammonium sulfate) was demonstrated (Fig 1A), and the isolation of TTI was visualized on a 12% SDS-PAGE gel (Fig 1B), showing a predominant protein band of approximately 21 kDa. TTI presented 0.7 mg of proteins and 100% inhibition of trypsin (393.47 IU/mg) in 70 g of tamarind seed flour.
3.2. In vitro inhibition of α-amylase
Regarding in vitro inhibition of α-amylase, TTI, and the control drug (acarbose), a commercial synthetic α-amylase inhibitor were tested at concentrations of 0.3 mg/mL, 0.6 mg/mL, and 1.5 mg/mL. It was observed that 100% inhibition of α-amylase by acarbose occurred at the highest concentration (1.5 mg/mL). In addition, a significant reduction (p < 0.05) in inhibition was observed when the acarbose concentration was reduced. Unlike what was observed with acarbose, TTI presented lower inhibition (approximately 37.3%), with no statistically significant differences, at the three concentrations tested (Fig 2). The percentage inhibition of ITT and acarbose were significantly different (p < 0.05), and the inhibition of acarbose was higher against α-amylase.
3.3. In silico study of the interaction of TTIp 56/287 and derivative peptides with α-amylase
3.3.1. Molecular docking TTIp 56/287 – α-amylase.
The results obtained through the HADDOCK 2.4 server identified the preferred conformations of the TTIp 56/287 – α-amylase complexes. Ten amino acid sequences contained in TTIp 56/287 were chosen to guide the molecular docking studies: sequence 1 (DTVHDTDGQV), encompassing residues from ASP001 to VAL010; sequence 2 (PLNNAGQYYI), from PRO011 to ILE020; sequence 3 (LPAQQGKGGG), from LEU021 to GLY029; sequence 4 (LGLSNDDDGN), from LEU031 to ASN040; sequence 5 (PIDIPIGLPVRFS), from PRO049 to SER061; sequence 6 (TTALSLNIEFTI), from THR070 to ILE081; sequence 7 (EKGYTVPKLSDDF), from GLU101 to PHE113; sequence 8 (SSAAPFKLKQFEEDYKLVYCSK), from SER115 to LYS135; sequence 9 (SESGERKCVDLG), from SER136 to GLY147; and sequence 10 (KKVDEESSEEWSIV), from LYS171 to VAL184. Thus, we have the TTIp 56/287 structure, highlighting the amino acid sequences selected for the molecular docking study with α-amylase (PDB ID 5VA9) (Fig 3).
The parameters provided by the HADDOCK Web Server for the most stable clusters of the TTIp 56/287 – α-amylase complexes were evaluated (Table 1). The number of structures obtained for each complex, the Haddock Score (HS), the RMSD, the relative energies of the van der Waals (EVWD) and electrostatic Coulombic (EELEC) interactions, and the Z-Score were obtained. Through these data, it was observed that regions 2 (PLNNAGQYYI) and 1 (DTVHDTDGQV) presented the best HS, −151.5 ± 3.0 and −143.0 ± 4.9, respectively, and were chosen as the best fitting orientations of TTIp 56/287 – α-amylase. More negative HS values indicated stronger and more energetically favorable binding.
The molecular model structures of the TTIp 56/287 – α-amylase complexes were generated (Table 1) in molecular surface representation (Fig 4). Regions 2 (PLNNAGQYYI) and 1 (DTVHDTDGQV) were chosen as the best-fitting orientations of TTIp 56/287 – α-amylase.
3.3.2. Obtaining, predicting, modeling, and selecting the three-dimensional structures of peptides derived from TTIp 56/287.
Five of the 13 peptides evaluated were selected and used in this study and are listed with their respective Haddock Score (HS) (Table 2). This factor was decisive in the selection of the peptides since the more negative the HS value, the higher the ligand-protein affinity. The energy and scoring parameters for the most stable clusters of the peptide-α-amylase complexes were also presented (Table 2). In decreasing order of HS: peptide 1 DTVHDTDGQVPL (12 aa; HS = −79.2 ± 6.3); peptide 2 TIAPACAPKPAR (12 aa; HS = −73.1 ± 9.3); peptide 3 TVSQTPIDIPIGLPVR (16 aa; HS = −72.8 ± 3.9); peptide 4 DEQSSEK (7 aa; HS = −41.1 ± 2.2); peptide 5 ILPAQQGK (8 aa; HS = −40.6 ± 3.5).
The last two peptides, DEQSSEK and ILPAQQGK, were chosen using the AntiDMPred software. This tool employed machine learning to predict antidiabetic potential by selecting candidates based on a probability threshold (pt) optimized for specificity and sensitivity. Notably, the pt values for DEQSSEK and ILPAQQGK were the highest (0.59), indicating strong potential [28].
The structural representation of surfaces and cartoon of the peptide-α-amylase complexes were also presented (Fig 5). As previously mentioned, five peptides with the most stable conformations were selected based on molecular docking results. It was found that the first three peptides presented (DTVHDTDGQVPL, TIAPACAPKPAR, and TVSQTPIDIPIGLPVR) exhibited the best HS values (<−72) and suggested the most stable docking.
Regarding the molecular dynamics results, the RMSD plot over the 300 ns simulation time (Fig 6) allowed us to evaluate the conformational stability of the complexes. Using the initial structure as a reference, the RMSD values were calculated based on the α-carbons. The RMSD analysis indicated that α-amylase maintained structural stability across all complexes, with RMSD values consistently below 2.5 nm. In contrast, the peptides exhibited notable conformational fluctuations, ranging from 0.1 to 0.7 nm, highlighting the dynamic behavior of TIAPACAPKPAR and ILPAQQGK.
In addition to the RMSD, the RMSF data (Fig 7) allowed us to evaluate the local conformational fluctuation of the complexes. The RMSF plot showed that the peptides TIAPACAPKPAR (Fig 7d) and ILPAQQGK (Fig 7b) presented the highest rigidity, followed by DTVHDTDGQVPL (Fig 7a). From RMSF, it can be observed that the conformational fluctuations remained within a range of up to 0.4 nm.
A comprehensive analysis of intermolecular interactions between the peptides and α-amylase was conducted, yielding detailed Interaction Potential Energy (IPE) data (Fig 8). Interaction potential energies (IPE) were calculated from the sum of short-range Coulomb and Lennard-Jones potentials using data collected over the final 50 ns of the simulation (Table 3). The IPE values demonstrated that longer-chain peptides exhibited stronger attractive energies. Specifically, the peptide TVSQTPIDIPIGLPVR showed the most negative total energy, at −824.53 ± 74.43 kJ/mol. This was followed by DTVHDTDGQVPL, with a total energy of −705.08 ± 43.24 kJ/mol, and TIAPACAPKPAR, which presented −584.11 ± 53.26 kJ/mol. The Coulomb energy component exhibited greater variability across the complexes compared to the Lennard-Jones component. This suggests that long-range electrostatic interactions play a crucial role in stabilizing the peptides at the enzymatic binding site. Notably, complexes TVSQTPIDIPIGLPVR and DTVHDTDGQVPL displayed significant Coulomb potentials of −528.21 ± 74.43 kJ/mol and −410.41 ± 43.24 kJ/mol, respectively.
Analysis of the interaction potential energy profiles (Fig 8) indicated a clear predominance of Coulombic interactions (blue line), reflecting the electrostatic potential, over Lennard-Jones interactions (red line), which characterized van der Waals forces. At the binding site, these forces contributed differentially to the stability of the complexes, varying in intensity according to the participating residues. The black line indicates the total interaction energy, which is the sum of the Coulomb and Lennard-Jones interactions.
The temporal evolution of the three-dimensional structures of the five α-amylase-peptide complexes across a 300 ns molecular dynamics simulation, with snapshots provided at 75 ns intervals, was demonstrated (Fig 9). The final conformational snapshots of peptides DTVHDDTDGQVPL, TVSQTPIDIPIGLPVR, and TIAPACAPKPAR, as determined by molecular dynamics simulations (Figs 10–12, respectively, using stick model representations). These figures showcase the peptides (green) positioned within the α-amylase interaction pocket alongside the interacting enzyme residues (red). The key interaction regions were further detailed [Figs 10b–12b, segmented into four distinct areas (i-iv)].
Analysis of the DTVHDDTDGQVPL-α-amylase complex (Fig 10a) revealed significant Coulomb interactions, including charge-charge (salt bridges) and hydrogen bonds. Residues ASP01 (−95.17 kJ/mol), HIS04 (−71.87 kJ/mol), ASP05 (−116.26 kJ/mol), ASP07 (−127.19 kJ/mol), and LEU12 (−121.60 kJ/mol) exhibited the highest energetic contributions to the intermolecular stability of the complex. Fig 10b presents specific intermolecular interactions. In region i, ASP01 (peptide) formed a 1.71 Å salt bridge with ILE148 (enzyme) and a 2.84 Å hydrogen bond with GLU149 (enzyme), while. ASP01-ILE148 pair formed a conventional 1.84 Å hydrogen bond. In region ii, the peptide interacted with the enzyme through residues HIS04, ASP05, and ASP07, maintaining a conventional hydrogen bond between the peptide-enzyme residue pairs HIS04-HIS201 (1.96 Å) and ASP07-ASP300 (1.48 Å); as well as charge-charge interactions between ASP05-LYS200 and ASP07-LYS200 (1.91 and 5.53 Å). In regions iii and iv, conventional hydrogen bonds were detected between the residue pairs GLN09-GLN063 (2.90 Å) and LEU12-SER108 (1.63 and 1.99 Å), respectively, and a hydrogen-carbon bond between LEU12-VAL107 (2.81 Å).
In the TVSQTPIDIPIGLPVR-α-amylase complex (Fig 11a), key stabilizing residues were identified, with ARG16 exhibiting the highest interaction energy (−106.63 kJ/mol), followed by GLN04 (−60.39 kJ/mol), ASP08 (−46.76 kJ/mol), ILE11 (−47.29 kJ/mol), VAL15 (−40.06 kJ/mol) and LEU13 (−32.31 kJ/mol).
Region i, illustrates the primary attractive interactions that stabilized the peptide-enzyme complex (Fig 11b). Specifically, conventional hydrogen bonds are observed between GLN04 and TRP059 (2.02 Å) and GLN04 and ASP353 (2.54 Å), while carbon-hydrogen bonds were formed between GLN04 and HIS305 (2.89 Å) and GLN04 and ASP352 (2.36 Å). Panels ii and iv highlight crucial interactions in this area of the pocket, such as conventional hydrogen bonds maintained between the pairs ILE11-GLY308 (2.56 Å) and ARG16-TRP284 (2.56 Å), in addition to the hydrogen-carbon bond ARG16-LYS261 (2.67 Å). Region iii was characterized by van der Waals interactions and alkyl-type interactions between the residue pairs LEU13-LYS257 (4.79 Å), LEU13-ALA260 (4.55 Å), VAL15-LYS257 (4.63 Å), and VAL15-LYS261 (5.26 Å).
In the TIAPACAPKPAR-α-amylase complex, key interactions included: a pi-alkyl interaction between PRO10 and TYR151; a conventional hydrogen bond between PRO10 and LYS200; a salt bridge (1.73 Å) and a conventional hydrogen bond (1.83 Å) between ARG12 and ASP300. Besides, two salt bridges between ARG12 and ASP356; pi-alkyl interactions between PRO04 and TRP058, and between PRO04 and TRP059; a conventional hydrogen bond between ILE02 and GLN063; amide-pi stacking between ILE02 and TRP059; and two pi-alkyl interactions between ALA03 and TRP059 (Fig 12a).
It is crucial to consider the relevant influence of electrostatic interactions in the case of salt bridges in the TIAPACAPKPAR-α-amylase complex. The electrostatic interactions mediated by the ARG12 residue of the peptide and the ASP300 and ASP356 residues of the enzyme played a crucial role in stabilizing the salt bridges within the complex, leading to a substantial reduction in the complex’s potential energy, measured at −244.23 kJ/mol (Fig 12b).
The computational prediction of the bioactivity of the selected peptides was also analyzed (Table 4). The peptides TIAPACAPKPAR and DTVHDTDGQVPL stood out in terms of bioactive potential (PeptideRanker) and half-life analyses, demonstrating high stability.
4. Discussion
The search for therapeutic alternatives for the management of Diabetes Mellitus (DM) has generated intense interest due to the impact of this disease, especially when related to persistent hyperglycemia [37]. Several approaches have been studied, including blocking or inhibiting therapeutic targets such as α-amylase, an enzyme responsible for carbohydrate digestion. In this context, peptide drugs have gained prominence due to their greater potency, tissue specificity, and reduced side effects [38]. For example, between 2015 and 2019, the Food and Drug Administration (FDA) approved 15 peptide drugs, corresponding to 7% of the total approvals, thereby reinforcing the potential of these compounds in modern therapeutics [39].
Recent advances in peptide screening and computational biology approaches have facilitated the development of new peptide drugs, allowing the screening and identification of active molecules through bioinformatics tools [23,37]. Many studies performed in silico have been subsequently confirmed in vivo, reinforcing the validity of computational findings [23,37]. Thus, the results of the present study were presented with the perspective that the trypsin inhibitor isolated from tamarind seed (Tamarindus indica L.), named TTI, could be a potential candidate for inhibiting α-amylase. In this study, TTI and especially two TTI-derived peptides, in vitro and in silico, showed the highest potential to inhibit α-amylase.
The obtaining of TTI was reproduced and confirmed through 12% SDS-PAGE stained with Coomassie Brilliant Blue G250, revealing protein bands with a predominance of molecular mass around 21 kDa. These data on the isolation and characterization of TTI agree with previous studies, such as those carried out by Carvalho et al. (2016) [15], Costa et al. (2018) [40], Costa et al. (2022) [41], and Lima et al. (2022) [42]. After obtaining, the in vitro α-amylase inhibition test was performed, demonstrating an inhibition of 37.3%. Related studies using seed protein hydrolysates, such as those from Luffa cylindrica [43], revealed inhibitions of up to 28% (when cleaved with trypsin) and 25% (with pepsin). In comparison, Awosika and Aluko (2019) [44] reported an inhibitory activity of 30.39% with yellow pea (Pisum sativum L.) protein hydrolysates. Thus, a similarity is observed between the inhibitory percentages found in these studies and the value obtained for the TTI. Furthermore, in this in vitro study, TTI was isolated and not completely purified.
Although acarbose, a purified synthetic inhibitor specific for α-amylase, presents up to 100% inhibition, TTI, a multifunctional natural protein with 100% antitryptic activity, achieved almost 40% inhibition. This fact sparked interest in investigating TTI-derived peptides, as it is known that isolated or purified peptides tend to bind more easily to the enzyme’s active site due to their size and more intense chemical interactions. In contrast, hydrolysates contain a wide variety of peptides and present weaker enzymatic binding capacity [38].
If TTI were purified, it would likely have exhibited greater affinity for α-amylase and, consequently, a higher inhibition rate. However, purified TTI has a molecular weight of 19 kDa [14]. Therefore, prospecting for TTI-derived peptides with affinity for α-amylase appears to be a more promising strategy.
Given this perspective, the study investigated, through computational tools, the interactions between the theoretical TTI (TTIp 56/287) and α-amylase using molecular modeling techniques. Ten different docking sequences were suggested to elucidate conformations with higher binding affinities. These analyses revealed that variations of TTIp 56/287 in three-dimensional space resulted in the identification of specific domains that demonstrate enhanced interaction with the target enzyme. This variety of docking possibilities reinforced the strategy of prospecting peptides derived from TTIp 56/287 as potential α-amylase inhibitors.
The molecular docking methodologies employed, notably through the HADDOCK 2.4 server, identified the preferred conformations of the complexes formed between TTIp 56/287 and α-amylase. And between the derived peptides and the enzyme. In docking studies for macromolecules, such as proteins, it is essential to previously orient an active region for anchoring in the binding site [44].
Based on the scoring results, the conformation generated using sequence 2 (PLNNAGQYYI) demonstrated significantly greater stability, indicated by the highest HS value, followed by the conformation derived from sequence 1 (DTVHDTDGQV). These data suggest that the derived peptides have an affinity for α-amylase, forming stable and energetically favorable interactions. Based on these results, five peptides derived from TTIp 56/287, previously indicated by Gomes et al. [23], were selected so that their complexes with α-amylase could be subjected to molecular dynamics simulations and the binding free energy calculation.
Molecular dynamics simulations were performed for 300 ns for each of the five peptide-α-amylase complexes (using the structure of human pancreatic α-amylase, PDB ID: 5VA9). Analysis of the RMSD throughout the simulation revealed that, after an initial period of significant fluctuations, the values stabilized, ranging from 0.08 to 0.3nm, after approximately 100 ns. This stabilization indicates that the systems reached a state of dynamic equilibrium, which confers reliability to the conformations sampled in the last 100 ns for subsequent analyses, such as calculating the binding free energy [45,46].
The RMSF analysis highlighted that the peptides TIAPACAPKPAR and ILPAQQGK were the most stable, in terms of rigidity, followed by DTVHDTDGQVPL. These three peptides showed a greater propensity to form hydrogen bonds [45]. While TIAPACAPKPAR showed a slightly better structural alignment with reference conformation, DTVHDTDGQVPL presented the best HS value, as indicated in Table 2. These observations highlight the complex interaction between hydrogen bond formation, structural deviations, and stability. This underscored the importance of visualizing the three-dimensional intermolecular interactions between peptides and α-amylase. It is worth noting that, in a study by Alhawday et al. (2024) [47], RMSF values for α-amylase inhibitory peptides reached up to 3.5 nm, suggesting that TTI-derived peptides 56/287 exhibit superior rigidity and stability in enzyme inhibition, which may consequently impact the inhibitory potential and activity of these peptides.
In the IPE analysis, the results indicated that the interaction energy is more evenly distributed along the peptide chain in the complexes formed by DTVHDTDGQVPL-α-amylase and TVSQTPIDIPIGLPVR-α-amylase. In contrast, other complexes, even without complete uniformity in the distribution of favorable interactions, showed residues with significant energy contributions, as observed in the TIAPACAPKPAR-α-amylase complex [45]. It is essential to consider that Coulomb interactions, of the charge-charge type and the salt bridges formed between charged amino acids (such as ARG, ASP, GLU, HIS, and LYS), played a crucial role in the stabilization of ligand-protein complexes together with conventional hydrogen bonds and hydrogen-carbon bonds.
Furthermore, mapping of the enzyme residues most involved in the interactions highlighted the key roles of GLN036, SER108, GLU149, LYS200, ASP300, ASP352, ASP353, ASP356, ARG195, HIS305, and GLY304. These findings corroborate previous studies that employed in silico approaches to investigate peptides derived from natural or non-natural sources with inhibitory potential against α-amylase [44,47]. Based on the study of the interaction energies per residue, it is proposed to modify the polarity of the central regions of DTVHDTDGQVPL and TIAPACAPKPAR to maximize the attractive intermolecular interactions in these regions, thereby conferring a greater binding affinity toward α-amylase.
Unlike previous studies that focused solely on molecular docking, the present research advanced the analysis by employing molecular dynamics simulations, extending up to 300 ns for each peptide-α-amylase complex. This approach enabled a detailed investigation of the key amino acid residues in the enzyme’s catalytic site, which are critical for substrate binding. The insights gained from this analysis may contribute to the more efficient design of analogous peptides with enhanced α-amylase inhibitory activity [45].
Complementary studies, such as that of Alhawday et al. (2024) [47], who developed and evaluated isoxazolidine derivatives as potential α-amylase inhibitors, reinforced the importance of these approaches. In their work, the synthesized compounds demonstrated promising in vitro inhibitory activities, with IPE values below −400 kJ/mol, high inhibitory potency (approximately 50%), and stability, evidencing a correlation between in silico data and experimental results.
In the therapeutic context, the greater the inhibitory potential of a molecule on α-amylase, the greater its efficacy in treating hyperglycemia associated with DM [48]. Acarbose, although a synthetic inhibitor with excellent inhibitory results, is related to adverse effects such as abdominal pain, flatulence, and diarrhea. This limitation reinforces the need to seek therapeutic alternatives with a lower incidence of side effects [49].
In silico analyses of the bioactive potential revealed that DTVHDTDGQVPL and TIAPACAPKPAR peptides exhibited superior stability and a prolonged half-life compared to TVSQTPIDIPIGLPVR. In particular, the TIAPACAPKPAR peptide achieved the highest score on the PeptideRanker server (exceeding 0.6), standing out as the candidate with the greatest predicted bioactive potential. One of the challenges in applying peptides with biological action is the low stability during the digestive process, resulting from the action of intestinal proteases [50].
Therefore, a half-life prediction study was conducted in a simulated intestinal environment, where all the peptides analyzed showed normal or high stability. The stability of peptides in a simulated intestinal environment is crucial for oral applications. It is rarely evaluated in initial peptide prospecting studies, thereby serving as a differentiator of this study. However, assessing peptides with fewer than 10 amino acids was impossible. This limitation is inherent to the tool used, and therefore, more robust analyses are necessary in the future to confirm the data presented. To date, there are no studies in the literature that confirm the reliability of these predictors for α-amylase inhibitory peptides or whether similar predictions have been experimentally validated in prior literature.
In summary, the prospecting of bioactive TTI-derived peptides has proven to be a promising alternative for treating T2DM, as it effectively inhibits α-amylase from a natural trypsin inhibitor source. The data indicated that the TTIp 56/287 complex (evaluated in vitro and silico) and the peptides DTVHDTDGQVPL and TIAPACAPKPAR (evaluated in silico) interact effectively with the enzyme. These peptides were assessed in silico, without an assessment of toxicity, immunogenicity, or off-target effects. In addition, none were tested in vitro or in vivo. Therefore, the peptide synthesis, possible delivery systems, optimization of peptide length, and its action on α-amylase must be biologically validated in the future.
These results, following in vitro and in vivo validation, will confirm whether these peptides might reduce dosing frequency, whether there is a risk of non-specific inhibition of other digestive enzymes, and potential effects on postprandial glucose curves compared to existing drugs. Therefore, this study, still hypothetically, suggests the potential of these hydrolysates to be used as functional ingredients in the development of nutraceutical products or even drugs intended for the control and treatment of T2DM, offering an alternative with a lower risk of adverse effects compared to the synthetic inhibitors currently available.
5. Conclusion
The present study evaluated the in vitro inhibitory and interaction effects of TTI and/or its peptides on α-amylase. Regarding computational analyses, the docking and molecular dynamics simulation results showed that TTIp 56/287 and its peptides interacted stably and with affinity with the α-amylase molecule.
From the RMSD, RMSF, and IPE, it was observed that the peptides DTVHDTDGQVPL and TIAPACAPKPAR demonstrated the most promising results. This observation was further supported by bioactivity, which revealed the high structural stability of these two molecules. Therefore, these peptide sequences are highlighted as promising candidates for future studies aimed at validating their hypoglycemic potential through α-amylase inhibition, along with an acceptable safety profile. Consequently, they are excellent candidates for future in vitro and in vivo analyses.
Acknowledgments
To the Laboratory of Chemistry and Function of Bioactive Proteins, the Laboratory of Computational Chemistry, and the High-Performance Processing Center (NPAD) at Federal University of Rio Grande do Norte, for the computational facilities required to perform molecular simulations.
References
- 1.
Brasil. Pesquisa Nacional de Saúde: 2019: Atenção Primária à Saúde e Informações Antropométricas. Rio de Janeiro: IBGE; 2020.
- 2. Schmitt LO, Gaspar JM. Obesity-induced brain neuroinflammatory and mitochondrial changes. Metabolites. 2023;13(1):86.
- 3. Crujeiras AB, Cordero P, Garcia-Diaz DF, Stachowska E, González-Muniesa P. Molecular basis of the inflammation related to obesity. Oxid Med Cell Longev. 2019;2019:5250816. pmid:30911347
- 4.
International Diabetes Federation. IDF diabetes atlas. 10 ed. International Diabetes Federation; 2021.
- 5. American Diabetes Association Professional Practice Committee. 2. classification and diagnosis of diabetes: standards of medical care in diabetes-2022. Diabetes Care. 2022;45(Suppl 1):S17–38. pmid:34964875
- 6.
Sociedade Brasileira de Diabetes SBD. Diretriz da Sociedade Brasileira de Diabetes. Brasil: Sociedade Brasileira de Diabetes (SBD); 2019.
- 7. Holt RIG, DeVries JH, Hess-Fischl A, Hirsch IB, Kirkman MS, Klupa T, et al. The management of type 1 diabetes in adults. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2021;44(11):2589–625. pmid:34593612
- 8. Ashaolu TJ. Antioxidative peptides derived from plants for human nutrition: their production, mechanisms and applications. Eur Food Res Technol. 2020;246:853–65.
- 9. Möller NP, Scholz-Ahrens KE, Roos N, Schrezenmeir J. Bioactive peptides and proteins from foods: indication for health effects. Eur J Nutr. 2008;47(4):171–82. pmid:18506385
- 10. Farias TC, de Souza TS, Fai AE, Koblitz MG. Critical review for the production of antidiabetic peptides by a bibliometric approach. Nutrients. 2022;14(20).
- 11. Verhamme IM, Leonard SE, Perkins RC. Proteases: pivot points in functional proteomics. Methods Mol Biol. 2019;1871:313–92. pmid:30276748
- 12. Aguiar AJFC, de Queiroz JLC, Santos PPA, Camillo CS, Serquiz AC, Costa IS, et al. Beneficial effects of tamarind trypsin inhibitor in Chitosan-Whey protein nanoparticles on hepatic injury induced high glycemic index diet: a preclinical study. Int J Mol Sci. 2021;22(18):9968. pmid:34576130
- 13. Costa I, Lima M, Medeiros A, Bezerra L, Santos P, Serquiz A, et al. An insulin receptor-binding multifunctional protein from tamarindus indica L. Presents a hypoglycemic effect in a diet-induced type 2 diabetes-preclinical study. Foods. 2022;11(15):2207. pmid:35892791
- 14. de Medeiros AF, de Souza BBP, Coutinho LP, Murad AM, Dos Santos PIM, Monteiro N de KV, et al. Structural insights and molecular dynamics into the inhibitory mechanism of a Kunitz-type trypsin inhibitor from Tamarindus indica L. J Enzyme Inhib Med Chem. 2021;36(1):480–90. pmid:33491503
- 15. Carvalho FMC, Lima VCO, Costa IS, Luz ABS, Ladd FVL, Serquiz AC, et al. Anti-TNF-α agent tamarind kunitz trypsin inhibitor improves lipid profile of wistar rats presenting dyslipidemia and diet-induced obesity regardless of PPAR-γ induction. Nutrients. 2019;11(3):512. pmid:30818882
- 16. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem. 1976;72:248–54. pmid:942051
- 17. Kakade M, Simons N, Liener I. An evaluation of natural vs. synthetic substrates for measuring the antitryptic activity of soybean samples. Cereal Chem. 1969;46:518–26.
- 18. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227(5259):680–5. pmid:5432063
- 19. Telagari M, Hullatti K. In-vitro α-amylase and α-glucosidase inhibitory activity of Adiantum caudatum Linn. and Celosia argentea Linn. extracts and fractions. Indian J Pharmacol. 2015;47(4):425–9. pmid:26288477
- 20. Oliveira FC de M, Holanda TMV, de Assis CF, Xavier Júnior FH, de Sousa Júnior FC. Flours from Spondias mombin and Spondias tuberosa seeds: physicochemical characterization, technological properties, and antioxidant, antibacterial, and antidiabetic activities. J Food Sci. 2024;89(1):342–55. pmid:38126119
- 21. Goldbach L, Vermeulen BJA, Caner S, Liu M, Tysoe C, van Gijzel L, et al. Folding then binding vs folding through binding in macrocyclic peptide inhibitors of human pancreatic α-amylase. ACS Chem Biol. 2019;14(8):1751–9. pmid:31241898
- 22. Van Zundert GCP, Rodrigues JP, Trellet M, Schmitz C, Kastritis PL, Karaca E, et al. The HADDOCK2.2 web server: user-friendly integrative modeling of biomolecular complexes. J Mol Biol. 2016;428(4):720–5.
- 23. Francisca T. Gomes A, de Medeiros WF, Bezerra LL, Beatriz S. Luz A, de Sousa Junior FC, de L. Vale SH, et al. Interaction between insulin receptor and a peptide derived from a trypsin inhibitor purified from tamarind seed: An in silico screening of insulin-like peptides. Arabian J Chem. 2024;17(6):105780.
- 24. Lamiable A, Thévenet P, Rey J, Vavrusa M, Derreumaux P, Tufféry P. PEP-FOLD3: faster de novo structure prediction for linear peptides in solution and in complex. Nucleic Acids Res. 2016;44(W1):W449-54. pmid:27131374
- 25. Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, et al. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX. 2015;1(2):19–25.
- 26. Huang J, MacKerell AD Jr. CHARMM36 all-atom additive protein force field: validation based on comparison to NMR data. J Comput Chem. 2013;34(25):2135–45. pmid:23832629
- 27. Jorgensen WL, Chandrasekhar J, Madura JD, Impey RW, Klein ML. Comparison of simple potential functions for simulating liquid water. J Chem Phys. 1983;79(2):926–35.
- 28. Chen X, Huang J, He B. AntiDMPpred: a web service for identifying anti-diabetic peptides. PeerJ. 2022;10:e13581. pmid:35722269
- 29. Van Gunsteren WF, Berendsen HJC. A leap-frog algorithm for stochastic dynamics. Mol Simul. 1988;1(3):173–85.
- 30. Darden T, York D, Pedersen L. Particle mesh Ewald: an N⋅log(N) method for Ewald sums in large systems. J Chem Phys. 1993;98(12):10089–92.
- 31. Hess B, Bekker H, Berendsen HJ, Fraaije JG. LINCS: A linear constraint solver for molecular simulations. J Comput Chem. 1998;18(12):1463–72.
- 32. Bussi G, Donadio D, Parrinello M. Canonical sampling through velocity rescaling. J Chem Phys. 2007;126(1):014101. pmid:17212484
- 33. Nosé S, Klein ML. Constant pressure molecular dynamics for molecular systems. Mol Phys. 1983;50(5):1055–76.
- 34. Tomasello G, Armenia I, Molla G. The Protein Imager: a full-featured online molecular viewer interface with server-side HQ-rendering capabilities. Bioinformatics. 2020;36(9):2909–11. pmid:31930403
- 35. Mooney C, Haslam NJ, Pollastri G, Shields DC. Towards the improved discovery and design of functional peptides: common features of diverse classes permit generalized prediction of bioactivity. PLoS One. 2012;7(10):e45012. pmid:23056189
- 36. Amorim-Carmo B, Daniele-Silva A, Parente AMS, Furtado AA, Carvalho E, Oliveira JWF, et al. Potent and Broad-Spectrum Antimicrobial Activity of Analogs from the Scorpion Peptide Stigmurin. Int J Mol Sci. 2019;20(3):623. pmid:30709056
- 37. Medeiros I, Gomes AFT, Oliveira E Silva EG, Bezerra IWL, da Silva Maia JK, Piuvezam G, et al. Proteins and peptides studied in silico and in vivo for the treatment of diabetes mellitus: a systematic review. Nutrients. 2024;16(15):2395. pmid:39125276
- 38. Antony P, Vijayan R. Bioactive peptides as potential nutraceuticals for diabetes therapy: a comprehensive review. Int J Mol Sci. 2021;22(16):9059. pmid:34445765
- 39. De La Torre BG, Albericio F. Peptide therapeutics 2.0. Molecules. 2020;25:2293.
- 40. Costa ROA, Matias LLR, Passos TS. Safety and potential functionality of nanoparticles loaded with a trypsin inhibitor isolated from tamarind seeds. Future Foods. 2020;1–2:100001.
- 41. Costa IS, Medeiros AF, Carvalho FMC, Lima VCO, Serquiz RP, Serquiz AC, et al. Satietogenic protein from tamarind seeds decreases food intake, leptin plasma and CCK-1r gene expression in obese Wistar Rats. Obes Facts. 2018;11(6):440–53. pmid:30537704
- 42. Lima VCO, Luz ABS, Amarante M do SM, Lima MCJS, Carvalho FMC, Figueredo JBS, et al. Tamarind multifunctional protein: safety and anti-inflammatory potential in intestinal mucosa and adipose tissue in a preclinical model of diet-induced obesity. Obes Facts. 2021;14(4):357–69. pmid:34256373
- 43. Arise RO, Idi JJ, Mic-Braimoh IM, Korode E, Ahmed RN, Osemwegie O. In vitro Angiotesin-1-converting enzyme, α-amylase and α-glucosidase inhibitory and antioxidant activities of Luffa cylindrical (L.) M. Roem seed protein hydrolysate. Heliyon. 2019;5(5):e01634. pmid:31193002
- 44. Awosika TO. Inhibition of the in vitro activities of α-amylase, α-glucosidase and pancreatic lipase by yellow field pea (Pisum sativum L.) protein hydrolysates. Int J Food Sci Technol. 2019;54:2021–34.
- 45. Abdullahi M, Uzairu A, Shallangwa GA, Mamza PA, Ibrahim MT, Ahmad I, et al. Structure-based drug design, molecular dynamics simulation, ADMET, and quantum chemical studies of some thiazolinones targeting influenza neuraminidase. J Biomol Struct Dyn. 2023;41(23):13829–43. pmid:37158006
- 46. Du Z, Comer J, Li Y. Bioinformatics approaches to discovering food-derived bioactive peptides: reviews and perspectives. TrAC Trends Anal Chem. 2023;162:117051.
- 47. Alhawday F, Alminderej F, Ghannay S, Hammami B, Albadri AEAE, Kadri A, et al. In silico design, synthesis, and evaluation of novel enantiopure isoxazolidines as promising dual inhibitors of α-amylase and α-glucosidase. Molecules. 2024;29(2):305. pmid:38257218
- 48. Siow HL, Lim T, Gan CY. Development of a workflow for screening and identification of αlfa-amylase inhibitory peptides from food source using an integrated bioinformatics-phage display approach: case study–cumin seed. Food Chem. 2017;2(14):67–76.
- 49. Moein S, Pimoradloo E, Moein M, Vessal M. Evaluation of antioxidant potentials and α-amylase inhibition of different fractions of labiatae plants extracts: as a model of antidiabetic compounds properties. Biomed Res Int. 2017;2017:7319504. pmid:29082253
- 50. Manzoor M, Singh J, Gani A. Exploration of bioactive peptides from various origin as promising nutraceutical treasures: In vitro, in silico and in vivo studies. Food Chem. 2022;373(Pt A):131395. pmid:34710682