Effects of thermal processing on protein quality and bioactivity of tilapia: Explored by in vitro and computer simulated digestion

Abstract

Thermal processing plays a crucial role in tilapia consumption. To investigate the effects of heat treatment on tilapia protein quality and biological activity, this study employed three methods—steaming, microwaving, and baking to process tilapia protein, followed by static INFOGEST In Vitro digestion. Protein quality assessment indicators included degree of hydrolysis, molecular weight distribution, amino acid composition, and scoring. Bioactivity was verified through In Vitro experiments and computer simulations. Results indicate that thermal processing significantly enhances protein digestibility and amino acid availability by moderately altering protein structure and exposing enzymatic cleavage sites. Among the treatments, baked achieved the highest degree of hydrolysis, essential amino acid index, and biological value; while steaming exhibited the highest amino acid release rate and essential amino acid score. Concurrently, thermal processing significantly enhanced tilapia protein bioactivity: Baked markedly increased hypoglycemic activity, with the α-amylase and α-glucosidase inhibitory activities after gastrointestinal enzyme hydrolysis in the baked group reaching 51.48% and 51.74%, respectively, significantly higher than the 42.72% and 41.17% in the fresh group. Steaming substantially elevated antioxidant activity, with ABTS radical scavenging activity reaching 69.00% compared to 57.96% in the fresh group.

Graphical abstract

• •Baking caused the most hydrolysis; steaming caused the most amino acid release.
• •Thermally processed group had higher protein amino acid scores than raw.
• •Computer-simulated cleavage improved the biological activity of tilapia proteins.
• •Baking caused the most hypoglycaemia and steaming the most antioxidant activity.
• •Thermal processing improves tilapia protein nutrition and biological activity.

1. Introduction

Proteins are of critical importance in the context of human health. The quality of a protein is determined by the composition of its amino acids, its efficiency in digestion, and its absorption rate of the amino acids and peptides produced during digestion. These factors collectively affect the protein's ability to perform its designated physiological function (Marinangeli et al., 2021). The efficiency of proteins in meeting the body's amino acid needs can be measured by examining their amino acid composition (Jones et al., 2020). Meanwhile, protein digestibility is a crucial aspect in determining what proportion of the consumed protein is absorbed, thus providing insight into the recommended daily protein intake for individuals (Berardy et al., 2019). Proteins exhibit a high degree of diversity, with differences between sources in amino acid composition and digestibility (Adhikari et al., 2022). As stated by Nosworthy et al. (2023), plant proteins are characterized by the absence of certain essential amino acids and a reduced level of digestibility. This phenomenon can be attributed to the presence of antinutritional agents.

Tilapia has been identified as a highly protein-rich foodstuff, with the potential to meet daily nutritional requirements while providing sufficient essential amino acids, thus making it a valuable economic asset in global aquaculture (Zhang et al., 2020). Thermal processing plays an important role in tilapia consumption. Moderate exposure to heat and moisture has been shown to induce changes in proteins that can be structural, conformational, and oxidative in nature. It has been shown that this process causes the protein structure to partially unfold, exposing active sites (Li, Sun, et al., 2021) that can interact with digestive enzymes. Jiang et al. (2021) studied Pacific oysters and concluded that all cooking methods increase protein digestibility. Nevertheless, it has been demonstrated that excessive heat treatment has the capacity to modify the binding sites of proteins to proteohydrolases and reduce their susceptibility to digestive proteases by polymerizing the proteins, making them more compact. Filgueras et al. (2011) concluded in their study of American ostrich that cooking reduces the sensitivity of myofibrillar proteins to pepsin. This decline in sensitivity results in a reduction in digestibility, which, in turn, leads to a decline in the eating and dietary value of heat-treated meat.

In addition to their nutritional role, proteins from tilapia are a potential source of bioactive peptides, particularly those with antioxidant and angiotensin-I-converting enzyme (ACE) inhibitory activities, and researchers are exploring ways to predict these properties. Virtual hydrolysis of the BIOPEP-UWM database has been applied by researchers. Wang, Qi, et al. (2025) through the integration of computational tools and in vitro experiments, identified novel potential ACE inhibitory peptides derived from proteins of four tuna species (Thunnus maccoyii). Chen et al. (2023) developed an activity-directed enzymatic screening method for chia seed press proteins through computational digestion.

The advantage of this virtual hydrolysis method lies in its ability to rapidly screen large protein sequences for potential bioactive peptide fragments that may be released after enzymatic cleavage. This provides a theoretical basis for identifying promising bioactive peptides before conducting costly and time-consuming In Vitro or in vivo experiments. Nevertheless, there is a lack of comprehensive documentation regarding the specific bioactive properties  of tilapia protein hydrolysates, particularly their antioxidant, angiotensin-converting enzyme (ACE) inhibitory, and anti-diabetic activities in both computer simulations and In Vitro experiments. This study aims to determine the effects of heat on tilapia protein and predict enzyme digestibility. This objective will be validated through In Vitro experiments. This approach will facilitate an accurate assessment of the nutritional value by providing insights into the products of tilapia protein digestion resulting from different thermal processing treatments. This study offers novel insights into the mechanisms through which diverse thermal processing procedures impact the In Vitro digestion and bioactivity of tilapia.

2. Material and methods

2.1. Material

Fresh tilapia, CR Vanguard Supermarket, Guangzhou, China, mass 500 ± 50 g/tail; α-glucosidase (1000 U/mg), α-amylase (2000 U/mg), pepsin (3000 U/mg), trypsin (250 U/mg) were from Macklin (Shanghai, China); OPA, DNS reagent, DPPH from Solarbio (Nanjing, China); PNPG from Yuan Ye (Shanghai, China); all reagents were from Macklin unless otherwise stated. All reagents are of analytical grade purity.

2.2. Preparation of protein samples

Fresh tilapia samples (n = 24 fish) harvested across the entire dorsal muscle were trimmed into fillets of uniform size (approximately 150 ± 10 g in weight and 2.0 ± 0.3 cm in thickness). These fillets were then randomly allocated to four different groups (n = 6 per group): the raw group (TF), the steaming group (TS), the baking group (TG), and the microwave group (TM). To precisely monitor the heat treatment process, a thermocouple temperature probe (Type T, 0.5 mm diameter) is inserted into the center of the fish fillet. This records real-time changes in core temperature during heating, ensuring uniform heating across all sections. The three heat treatments followed Li, Ye, and Singh (2021) method with slight changes:(1) Steaming: the fish fillets were placed into the steamer after the water in the steamer pot had been brought to boiling point, and then steaming was performed at 100 °C for 10 min, after which the surface was sucked off after the fish fillets were removed from the steamer. (2) Baking:  The fish fillets were baked  in a preheated oven at 160 °C for 20 min (3) microwave: Place the fish fillets in a 500 W microwave oven and heat for 5 min. After cooling at room temperature for 10 min, store the sample in a − 80 °C freezer bag. Thaw at 20 °C for 1 h before proceeding to the next step in the experimental protocol. All treatments were performed in three independent experimental replicates  (n = 3), meaning the entire process from fish allocation to cooking and analysis was repeated on three separate occasions with new fish samples.

2.3. Extraction of tilapia m yofibrillar protein

2 g of chopped meat samples from each group were weighed to an accuracy of 0.01 g. Following the method of Fu, Hao, et al., 2025with minor modifications, prepare the extraction buffer by adding 6 mol/L NaCl and 10.0 mM/L PMSF to 50 mM Tris-HCl buffer (pH 7.2). Subsequently, immerse the sample in an ice bath for 25–30 min. Centrifuge at 12,000 rpm for 15 min at 4 °C. Pool the resulting pellet into a fibrin solution.

2.4. Static in vitro digestion simulation

The study followed the standard INFOGEST protocol with SGF & SIF prepared as previously detailed by Brodkorb et al. (2019). The process of gastric digestion was initiated by combining 10 mL of a 5 mg/mL protein solution with 10 mL of prepared SGF at pH = 2, followed by the introduction of pepsin was then added to achieve an enzyme activity of 2000 U/mL. The mixture was incubated in a constant-temperature shaker at 37 °C and 160 rpm for 2 h. The digested solution was collected and tested for G120 content. For the enteric digestion stage, take 10 mL of gastric digest, inactivate it with boiling water, and gradually raise the pH of the mixture to 7.0 using 1 M NaOH. Combine it with 10 mL of simulated intestinal fluid (SIF) at pH 7 in a volumetric flask. Add trypsin to achieve an enzyme activity of 100 U/mL, simultaneously adding bile salts to reach a concentration of 10 mM. Finally, add CaCl₂ solution to a final concentration of 0.6 mM. Incubate the final mixture in a constant-temperature shaker at 160 rpm for 4 h. after which it became collected and subjected to measurement to produce the sample designated I240. Prior to the initiation of the digestion process, the sample was designated G0. Prior to further examination, all samples were maintained at a temperature of −80 °C.

2.5. Determination of hydrolysis(DH)

The degree of hydrolysis measurements were made with modified o-phthalaldehyde (OPA) assays. 2 mL of anhydrous ethanol were used to dissolve 80 mg of OPA, then transferred to 75 mL of deionised water (Sodium tetraborate 3.8 g, sodium dodecyl sulphonate 0.1 g). Subsequently, 88 mg of dithiothreitol were introduced into 100 mL of the OPA agent, which were then fixed. Subsequently, 10 μL of the specimen or control being examined was combined with 250 μL of the OPA agent referenced above. The reaction was then permitted to proceed for a duration of 15 min. The absorbancy value (340 nm) was measured using a Multiskan™ FC (Guangzhou, China), a leucine calibration range of 0–10 mM was employed for the purpose of analysis. The free NH₂ group value was determined using the following eq. (1):

$$\mathit{DH}\left(\%\right)=\frac{{\mathit{NH}}_1-{\mathit{NH}}_2}{{\mathit{NH}}_3-{\mathit{NH}}_2}\times 100$$ (1)

where NH₁ is indicative of the concentration of free amino groups at differing stages throughout the process of digestion. NH₂ is representative of the concentration of free amino groups prior to the onset of digestion(G0). NH₃ is the concentration of total free amino groups following the acid hydrolysis process (6 M HCl, 110 °C, 24 h).

2.6. Determination of molecular weight distribution(MW)

The molecular weight distribution of peptides was determined during digestion. Following the method of Fu, Li, et al. (2025), this measurement was performed using an LC-20 A liquid chromatograph (Tokyo, Japan). Size exclusion chromatography was employed using a TSKgel G2000 SWXL column with an inner diameter of 7.8 mm, column length of 300 mm, and particle size of 5 μm. The mobile phase consisted of a mixture of 45% acetonitrile and 55% water, adjusted with 0.1% trifluoroacetic acid. Elution proceeded at a constant flow rate of 0.5 mL/min with a 20 μL injection volume, detected at 220 nm. Calibration curves were constructed using cytochrome C, peptidase, and vitamin B12 to calculate the molecular weight distribution of peptides in the samples.

2.7. Determination of free amino acids

The amino acid composition of the specimens was analyzed by means of an L-8900 Amino Acid Auto Analyser (Hitachi, Japan. The specimen solution was subjected to filtration using a 0.22-μm-pore-size filter prior to its injection into the analyser. The amino acids present in each sample were determined by retention time, as well as peak area, in comparison to those of a standard amino acid mixture.

2.8. Calculation of amino acid scores

As posited by Yin et al. (2024), standardised amino acid analysis emerged in the context for measuring egg protein quality. This study references the FAO/WHO standard whole egg protein profile (Food and Agriculture Organization of the United Nations (FAO), 2013) and employs the essential amino acid score (EAAS) and essential amino acid index (EAAI) metrics to evaluate the protein quality of tilapia under different conditions. The biological value (p-BV) of the four assays was predicted using the methodology of Abdo et al. (2024). The calculation of EAAS and EAAI was performed using eqs. (2), (3), respectively, and the calculation of p-BV was conducted according to eq. (4):

$$\text{EAAS}=\frac{\mathrm{mg}\text{of}\mathrm{EAA}\text{in}1\mathrm{g}\text{of sample protein}}{\mathrm{mg}\text{of}\mathrm{EAA}\text{in}1\mathrm{g}\text{of whole}\mathrm{egg}\text{protein}}\times 100$$ (2)

$$\mathit{EAAI}=\sqrt[n]{\frac{{\mathit{Lyu}}_s}{{\mathit{Lyu}}_r}\times \frac{{\mathit{Tyr}}_s}{{\mathit{Tyr}}_r}\times \cdots \times \frac{{\mathit{His}}_s}{{\mathit{His}}_r}}$$ (3)

Where r denotes the content of essential amino acids in the reference protein (mg/g); n denotes the number of amino acids calculated; s denotes the content of essential amino acids in the sample protein (mg/g).

$$p-\mathit{BV}=1.09\times \mathit{EAAI}-11.70$$ (4)

2.9. Computer simulation of enzyme digestion

In the UniProtKB protein database (https://www.uniprot.org/), enter the corresponding species code and protein name abbreviation based on the specific species or protein. Myosin and actin are the two most abundant proteins in fish skeletal muscle, and thus they were selected as representatives to reflect the primary characteristics of the overall hydrolysate, with accession numbers P68264, P68143, I3KU64, and I3KU65. The “Enzyme Action” tool available on the BIOPEP-UWM server (http://www.uwm.edu.pl/biochemia/index.php/en/biopep) was used for computer digestion, following the method of Koh et al. (2022) and maintaining consistency with In Vitro simulated digestion experiments. This involved simulating gastrointestinal digestion of each protein sequence through the sequential application of endoproteases: (1) Pepsin (EC 3.4.23.1; pH 1.3) (2) Trypsin (EC 3.4.21.4). Use the BIOPEP-UWM server (http://www.uwm.edu.pl/biochemia/index.php/en/biopep) to paste the amino acid sequence of the target protein into the designated input box, select the enzyme and enzyme combination, and then run the simulation. Finally, use the “Potential Bioactivity Profile” tool to analyze the potential bioactivity of all obtained peptides.

2.10. Determination of antioxidant activity

2.10.1. DPPH free radical scavenging activity

It is evident that the methodology established previously by Wang et al. (2024) has been followed to a great extent, with only minor adaptations being made. The DPPH solution was meticulously crafted in anhydrous ethanol, with a concentration that was precisely calibrated at 0.2 mg/mL. Then, 0.5 mL of solutions G120 and I240, respectively, were added to the DPPH solution for 30 min in the light. Subsequently, the samples were submitted for a centrifugal procedure at a velocity setting of 8000 rotations per minute for a temporal span of exactly five minutes. The initial step involved the addition of a volume equivalent to 200 μL of each solution to a 96-well plate. This was done using Multiskan™ FC (Guangzhou, China), and the initial extinction coefficient (A₁) was measured at a wavelength of 517 nm. The A₂ measurement was conducted following the combination of 1 mL of the solution with 2 mL of anhydrous ethanol, whilst the A₃ measurement was taken at 517 nm after the addition of 1 mL of distilled waters and 2 mL of the DPPH solution. The results were calculated by means of the following formula (6):

$$\text{DPPH}\left(\%\right)=\left(1-\frac{A_1-A_2}{A_3}\right)\times 100\%$$ (6)

2.10.2. ABTS+ free radical scavenging capacity

As Wang, Zhang, et al. (2025) describe, an aqueous dispersion of ABTS at a concentrations of 7 mM was added to an aqueous dispersion of persulfate at a constant concentration of 2.45 mM. The mixture was then left under dark conditions for 12 h to promote the generation of ABTS radical cations. The subsequent dilution of the sample with distilled water resulted in an optical density of 0.70 ± 0.02 at 734 nm. The pipette is used to transfer 1 mL of G120 and I240, following which 2 mL of diluted ABTS radical cation solution (set as A₁) are added. The mixture is then agitated and subjected to reaction for 6 min at room temperature in an environment devoid of light. Subsequent to the completion of the reaction, the turbidimetric measurement of the sample at 735 nm is facilitated by means of the Multiskan™ FC enzyme labeller (Guangzhou, China). A Multiskan™ FC Enzyme Labeler (Guangzhou, China) was employed to conduct absorbance measurements at a wavelength of 735 nm. Distilled water was utilised as an uncontaminated control, designated A₀, while potassium persulfate solution provided an additional uncontaminated control, termed A₂. The corresponding blank control absorbances were calculated according to Eq. (7).

$$\text{ABTS}\left(\%\right)=\left(1-\frac{{\mathrm{A}}_1-{\mathrm{A}}_2}{{\mathrm{A}}_0}\right)\times 100\%$$ (7)

2.10.3. Hydroxyl radical scavenging activity

The test solutions (G120 and I240) were prepared using a test tube. Thereafter, 1 mL of 0.15 mM FeSO₄, 0.4 mL of 2 mM salicylic acid ethanol and 1 mL of 6 mM H₂O₂ were introduced (the ‘control’ was distilled water and the ‘blank group’ was the test solution with distilled water). The solutions were subsequently subjected to rigorous agitation, followed by placement in a 37 °C thermostatic water bath for a duration of 30 min. Thereafter, the solutions were permitted to attain a state of equilibrium for a period of 20 min, culminating in a cooling process to the ambient temperature of 25 °C. Subsequently, an absorption wavelength of 510 nm was utilised to conduct the measurements on sample group B. The results were calculated by means of the following formula (8):

$$\mathrm{OH}·\left(\%\right)=\frac{A-\left(B-C\right)}{A}\times 100$$ (8)

2.10.4. Determination of total reducing power

The working solution of Ferric Reactive Azide Parameter (FRAP) was formulated by combining 300 mM acetic acid solution, 10 mM TPTZ-HCl solution, and 20 mM ferrous chloride solution, in the proportion of 10:1:1 (v/v/v). The mixture was subsequently exposed to an ambient temperature of 37 °C in an incubator for a period of 10 min prior to utilization. Subsequently, 5 μL of G120 and I240 and 180 μL of FRAP reaction solution was dispensed into a designated 96-well enzyme plate. The reaction was then allowed to proceed at 37 °C for a period of 5 min. Absorbances at 593 nm were then determined and used to create a Trolox calibration curve (0–900 mM). Consequently, the potential of the compounds to reduce iron was determined through the use of iron-reducing antioxidants, with the expression of the antioxidant capacity being given as the quantity of mM of equivalent Trolox per mM of the compound under investigation.

2.11. Determination of hypoglycaemic activity

2.11.1. α-glucosidase inhibitory activity

The inhibitory potential of α-glucosidase (AG) has been appraised by means of the methodology originally established by Kamal et al. (2018). The experiment involved the preparation of a 0.75 millimole solution of PNPG and a 0.2 millimole solution of a-glucosidase, prepared in 100 millimole phosphate buffer solution at a pH of 6.8. Subsequently, 40 μL of solutions G120 and I240, and 40 μL of a 5 mg/mL enzyme solution, were added to 1 mL of buffer. The resultant mixture was then subjected to temperature elevation to 37 °C for a period of 5 min. Then, 20 μL of 5 mM/L PNPG was added. The reaction was continued at 37 °C for 30 min to obtain the sample set A₁，A₂ was prepared without samples. The samples were prepared without AG Blank A₀. By means of a Multiskan™ FC enzyme marker (Guangzhou, China) at a wavelength of 405 nm. The absorbances were calculated using the following formula (9):

$$\text{Inhibition rate}\left(\%\right)=\left(1-\frac{A_1-A_2}{A_0}\right)\times 100\%$$ (9)

2.11.2. α-amylase inhibitory activity

As Tolmie et al. (2021) asserted, alpha-amylase (AA) inhibition was gauged through the implementation of specific modifications. The experiment was conducted by subjecting G120 and I240 (0.5 mL each) to a thermal treatment at 37 °C within a circulating water bath, in conjunction with 5 mg/mL of α-amylase (0.5 mL). The duration of this process was set to five minutes. A solution of dissolved starch (1%) was then added to the mixture, with the volume of this solution being 0.5 mL. The preparation should then be heated to 37 °C for 30 min. The addition of 1 mL of DNS chromogen followed by 1 mL of DNS colour developer stopped the reaction. The degree of absorption of the reaction solutions was monitored at a 540 nm wavelength after boiling for 10 min in a water bath. The absorption was calculated using eq. (10) for groups A₁, A₀ (the blank group, with ultrapure water instead of the sample), and A₂ (the control group, with phosphate buffer instead of the enzyme solution).

$$\text{Inhibition rate}\left(\%\right)=\left(1-\frac{A_1-A_2}{A_0}\right)\times 100\%$$ (10)

2.12. Determination of ACE inhibitory activity

The determination of ACE inhibitory activity was performed according to the method described by Yang et al. (2024). A solution of 5.82 mM/L histidine-histamine-leucine (HHL) and an ACE solution with an enzyme activity of 0.1 U/mL were prepared using a borate buffer solution (containing 0.3 mol/L NaCl) at pH 8.3 and 0.1 mol/L. In a 1.5-mL centrifuge tube, combine 100 μL of the 5.82 mM/L HHL solution with 20 μL of the sample. The mixture was then subjected to preheating in a 37 °C water bath for a period of five minutes, after which 10 μL of the 0.1 U/mL ACE solution was added to initiate the reaction. The mixture should then be subjected to static incubation in a water bath maintained at a temperature of 37 °C for a period of 60 min. Subsequently, 150 μL of 1 mol/L HCl should be added to terminate the reaction. Thereafter, 1 mL of ethyl acetate should be added, and the hippuric acid generated by the reaction should be extracted via vortex shaking. Centrifuge at 4000 ×g at 4 °C for 10 min, separate the upper layer of ethyl acetate, and add 1 mL of ultrapure water via vortex shaking to ensure complete dissolution. Finally, the absorbances of the solution at 228 nm were measured using the UV-1100 Ultraviolet Spectrophotometer (Shanghai, China) as A₁. The blank group A₀ was prepared by the addition of 150 μL of 1 mol/L HCl prior to the reaction, thus terminating it prematurely, while the control group A₂ utilised ultrapure water in lieu of the sample. The following formula (11) was utilised to calculate the extinction coefficient:

$$\text{Inhibition rate}\left(\%\right)=\left(\frac{A_2-A_1}{A_2-A_0}\right)\times 100\%$$ (11)

2.13. Statistics and analyses

The Levene test was used to verify the homogeneity of variances, and the Shapiro-Wilk test was used to verify the normal distribution of the data. The analysis was conducted using the method of analyzing variances (one-way ANOVA) as well as the comparison of means (Duncan's method of multiple comparisons) technique, with SPSS 21.0 employed as the statistical software. The statistical significance of a given variable was determined by the p-value, which was set at P < 0.05. The generation of graphs was facilitated by R4.4.1 and Origin 2024 software.

3. Results and discussion

3.1. Protein digestion characteristics

3.1.1. DH

As demonstrated in Fig. 1A-B, the degree of protein hydrolysis is affected by heating. Fig. 1A shows that during the stomach digestion stage, the degree of protein degradation of G120 exhibited statistically significant differences (P < 0.05) among the various specimens exposed to different treatment modalities. Fig. 1B illustrates the extent of protein hydrolysis occurring during the small intestine stage (I240). Data indicates that the degree of protein hydrolysis in the I240 stage following heat processing was significantly higher than in the fresh group. This is significant because it demonstrates that heating treatment exerts a substantial influence on the degree of protein hydrolysis. The outcomes of this investigation are of considerable significance in comparison with the raw sample group (P < 0.05). This occurrence may be attributed to the denaturing effect of heating on protein conformation, resulting in the exposure of active sites during the enzymatic digestion of fish protein. This, in turn, has been shown to increase the availability of protease and peptide binding sites, thereby facilitating the enzymatic hydrolysis reaction (Liu et al., 2020). It was determined through rigorous analysis that the extent of hydrolysis of TG was precisely 72.61%, which was significantly higher than that observed for TS and TM (P < 0.05). Elevated temperatures may cause slight alterations in protein structure and partial unfolding, thereby increasing their susceptibility to proteases (Liu & Xiong, 2000). At the same time, a study by Liu et al. (2025) concluded that microwave heating causes hydrophobic amino acids buried inside proteins to initially become exposed but later be reburied, promoting intermolecular aggregation of proteins and resulting in a reduced hydrolytic degree.

Fig. 1.

(A) Hydrolysis degree of tilapia protein after gastric digestion under different heating conditions; (B) Hydrolysis degree of tilapia protein after enteric digestion under different heating conditions; (C) Molecular weight distribution of tilapia protein after digestion under different heating conditions, Letters indicate significant differences in the content of the same molecular weight range at different digestion stages within the same sample; G0: undigested stage; G120: gastric digestion stage; I240: enteric digestion stage. The different letters indicate significant differences (P < 0.05).

3.1.2. MW

It is noteworthy that the molecular mass distribution can provide a more intuitive reflection of the level of protein degradation. The results of the molecular mass distribution analysis of hydrolysates subjected to different thermal treatment conditions are shown in Fig. 1C. In the undigested state (G0), the predominant peptide distribution fell within the range of 10,000 Da and 6000–10,000 Da. However, as the digestion process progressed, a noticeable decline in the percentage of high-molecular-mass peptides corresponded to an increase in the proportions of lower masses. It has been demonstrated that the percentage of peptides with a mass between 3000 and 6000 Da, resulting from heat treatment, is significantly higher in G120 than in the control group (P < 0.05). As indicated by the statistical analysis, there was a marked increase in the proportion of peptides with a molecular mass of less than 1000 Da at I240, compared to the raw group (P < 0.05). For samples with higher hydrolysis levels, the molecular weight distribution of peptide segments shifted significantly toward the low molecular weight region. This finding aligns with the trend in hydrolysis degree, indicating that a certain degree of heat treatment promotes the release of oligopeptides(Fu, Xiang, et al., 2025).

3.2. Release of free amino acids

As demonstrated in Figs. 2A-C, the dynamics of free amino acids in tilapia myofibrillar protein during simulated digestion are illustrated. Fig. 2A shows that a total of 24 amino acids, including 16 non-essential and 8 essential amino acids, were identified as by-products of tilapia protein digestion under diverse experimental conditions. The most abundant amino acids were identified as arginine and glutamic acid, while tryptophan was determined to be the most prevalent essential amino acid. As the digestion stage progressed, there was a gradual accumulation of total amino acids (Fig. 2B). Notably, at the conclusion of the intestinal digestion (I240), the free amino acid (FAA) releases of TF, TS, TG, and TM reached 102.07, 123.51, 115.61, and 113.62 mg/g, respectively. Li et al. (2022) reported consistent findings, showing that the content of certain components in tilapia fillet samples processed using steam or bake techniques was significantly higher compared to the control group that used unprocessed methods. This may be attributed to the fact that steaming (moist heat treatment) gently and uniformly denatures proteins, minimizing excessive aggregation or surface hardening. This provides greater substrate accessibility for intestinal enzymes, enabling them to hydrolyze peptides more thoroughly into free amino acids. Concurrently, the essential amino acid content released from protein exhibited a marked increase compared to the G0 level during the G120 and I240 stages, as evident in Fig. 2C. The findings suggest that the release of FAA from protein hydrolysis products following heating significantly exceeded that observed prior to heating. This observation signifies that exposure to elevated temperatures for an extended duration fosters the transformation and deposition of FAAs.

Fig. 2.

(A) Bubble plots of free amino acid release after tilapia protein digestion under different heating conditions; (B) total free amino acid release after tilapia protein digestion under different heating conditions; (C) release of essential amino acids after tilapia protein digestion under different heating conditions. The different letters indicate significant differences (P < 0.05).

3.3. Amino acid score

In evaluating tilapia protein quality under diverse thermal processing conditions, amino acid scores were utilised to assess protein quality and determine their efficacy in meeting the amino acid demands for optimal organismal health. EAAS, EAAI, and BV are shown in Figs. 3A–C, with specific data presented in Table S1. It is imperative to acknowledge the indispensable role of essential amino acids in human nutrition under all conditions. These nutrients are of particular significance in the context of the quality of food proteins. Consequently, it has been determined that the EAAS content exerts a considerable impact on the nutritive worth of proteins, as illustrated in Fig. 3A. At stage G0, the EAAS of the three thermal processing modalities was lower than that of TF. However, as digestion progressed, the tilapia protein sites exhibited proteinaceous binding and amino acid release increased, resulting in a substantial EAAS increase at the G120 and I240 stages. This finding suggests that thermal processing can significantly enhance the nutritional value of tilapia proteins.The EAAI has been demonstrated to be a reliable metric for the evaluation of protein quality, with higher values corresponding to proteins that are particularly nutritious (Shi et al., 2020). As demonstrated in Fig. 3B, TF exhibited the highest EAAI at the G0 stage, indicating that heating leads to protein aggregation, resulting in amino acids being concentrated within the protein. As the reaction progressed, the EAAI values of TS, TG, and TM were higher than those of TF at the G120 and I240 stages, indicating that the nutritional quality of fresh tilapia protein was relatively lower than that of heat-processed protein. However, the results of microwave treatment were significantly lower than those of the other two groups. This may be due to the formation of uneven partial aggregation structures in the proteins within the samples. Previous research by Cao et al. (2019) supports this finding, as they observed that microwave-heated samples exhibited distinct protein aggregation. This finding indicates that the nutritional quality of raw tilapia proteins was comparatively worse than that of thermally processed proteins.P-BV, an indicator of the efficacy of the body in extracting proteins from the diet (Zeng et al., 2013), was also measured and exhibited a high degree of correlation with the EAAI, as shown in Fig. 3C. In this figure, TG had the highest p-BV value of 65.36%, followed by TS with 61.72%. These results indicate that thermal processing improved the protein quality of tilapia protein, with steaming and baking being the most effective.

Fig. 3.

(A) Essential amino acid score (EAAS) of tilapia proteins under different heat treatment conditions; (B) essential amino acid index (EAAI) of tilapia proteins under different heat treatment conditions; and (C) predicted biological value (BV) of tilapia proteins under different heat treatment conditions. TF: Fresh Tilapia Protein, TS: Steamed Tilapia Protein, TG: Baked Tilapia Protein, and TM: Microwave Tilapia Protein.

3.4. Computer simulation of enzyme digestion

A selection of four proteins from tilapia was made from the UniProtKB database. Subsequently, these four proteins were submitted to computer-assisted proteolysis using trypsin (EC 3.4.21.4) and pepsin (EC 3.4.23.1; pH 1.3), with the BIOPEP-UWM server employed as the computational framework. Due to variations in the cleavage sites of the enzymes, different amino acid sequences were generated. Integrating these computational outcomes with the enzymatic cleavage characteristics of the proteases facilitates the prediction of peptide activity based on the position and recurrence of residue types within peptides (Minkiewicz et al., 2019). However, simulation results can only predict amino acid sequences and cannot predict amino acid types or structures. They require validation through experimental data, though current validation may be insufficient. Future research can enhance the accuracy and reliability of simulation results through further experimental validation and model optimization. The peptide sequences of the four proteins are displayed in Figs. 4C-D. After removing duplicates, a total of 168 peptides were obtained, including 72 dipeptides, 42 tripeptides, and 16 tetrapeptides. The potential bioactivities of all obtained peptides were analyzed by the BIOPEP-UWM server, as illustrated in Figs. 4A-B. At the G120 stage, the activities of the peptides were initially released, with ACE inhibitor and dipeptidyl peptidase IV inhibitor being the main active components. As the experiment progressed to the I240 stage, the ACE inhibitory activity increased threefold compared to the G120 stage, and the antioxidant activity was significantly improved. It is acknowledged that bioactive peptides typically contain basic amino acids (Lys and Arg) and hydrophobic amino acids (Leu, Gly, Pro, Ile, Val, Trp, and Phe), among others (Luo et al., 2022). Notably, multiple predicted peptides exhibiting angiotensin-converting enzyme (ACE) inhibitory activity are rich in hydrophobic amino acids (e.g., Val, Leu, Pro, Phe) and are located within the protein's hydrophobic regions. Additionally, peptides containing Tyr, His, or Trp are frequently predicted as antioxidant sequences, consistent with the mechanism linking radical scavenging capacity to aromatic electron donor properties. As demonstrated in Fig. 2A, the content of these amino acids increased significantly as the digestion stage progressed, reaching a peak at the I240 stage. This development aligns with the anticipated trend. Future research can enhance the accuracy and reliability of simulation results through further experimental validation and model optimization, such as verifying simulated active peptide segments via LC-MS/MS.

Fig. 4.

(A) Predicted bioactivity maps of the gastric digestion stage after computer-simulated enzymatic digestion; (B) Predicted bioactivity maps of the intestinal digestion stage after computer-simulated enzymatic digestion; (C) Amino acid sequences and active peptide segmentation sites of the gastric digestion stage after computer-simulated enzymatic digestion; (D) Amino acid sequences and active peptide segmentation sites of the intestinal digestion stage after computer-simulated enzymatic digestion.

3.5. Determination of antioxidant activity

3.5.1. DPPH free radical scavenging capacity

This is due to the established role of DPPH as a tool in the analysis of antioxidants, which is employed to evaluate the capacity of compounds to eliminate free radical species. DPPH radicals possess the ability to capture electrons or hydrogen atoms, thereby transforming into stable molecules and exhibiting a reduction in absorption at 517 nm. The demonstration provided in Fig. 5A indicates a clear capacity for the scavenging of DPPH radicals within tilapia protein hydrolysates. The DPPH radical scavenging capacity of four distinct protein groups demonstrated a significant increase, with a p-value less than 0.05, during the transition from the G0 to I240 phase. Specifically, the clearance rates for the TF, TS, TG, and TM groups at the I240 stage reached 48.69%, 56.91%, 57.67%, and 54.85%, respectively, representing a significant increase compared to their G0 stage values (12.24%, 9.43%, 8.24%, and 7.25%, respectively).Concurrently, the levels of amino acids (including lysine, alanine, and glycine) experienced a considerable surge. These products have demonstrated significant anti-radical properties (Ramírez et al., 2021). This study demonstrated that the DPPH radical quenching rate of TF was greater in comparison to the other three groups at both the G0 and G120 stages. However, following intestinal digestion (I240), the rate for TS, TG, and TM increased significantly, exceeding 48.69% for TF. In a separate study, Adamska and Felisiak (2025) found that microwaving and baking increased the antioxidant activity of Sulphur Shelf by approximately 10% and 25%, respectively, which is consistent with the findings of this paper.

Fig. 5.

(A) DPPH radical scavenging capacity of tilapia proteins at each stage of digestion under different heat treatment conditions; (B) ABTS radical scavenging capacity of tilapia proteins at each stage of digestion under different heat treatment conditions; (C) OH- radical scavenging capacity of tilapia proteins at each stage of digestion under different heat treatment conditions; (D) total reducing capacity of tilapia proteins at each stage of digestion under different heat treatment conditions. (D) Total reducing capacity of tilapia protein at each stage of digestion under different heat treatment conditions. The different letters indicate significant differences (P < 0.05). TF: Fresh Tilapia Protein, TS: Steamed Tilapia Protein, TG: Baked Tilapia Protein, and TM: Microwave Tilapia Protein.

3.5.2. ABTS free radical scavenging capacity

ABTS assays are available for both lipophilic and hydrophilic antioxidant systems. As demonstrated in Fig. 5B, the capacity to scavenge free radicals during In Vitro digestion is typically higher in the intestine than in the stomach. This discrepancy can be attributed to the increased synthesis of new peptide fragments, particularly during the gastric phase (Magouz et al., 2023). Furthermore, an investigation into the antioxidant activity of tilapia proteins subjected to three different heating regimens revealed significant variations. Notably, raw materials exhibited the strongest capacity in the undigested product at the G0 phase. However, TS and TG exhibited a more pronounced scavenging of free radicals at the G120 and I240 phases (P < 0.05), suggesting that the application of heat promoted the release of antioxidant-active peptides. This finding corroborates the observations reported by Rao et al. (2020), who established that heat treatment increased antioxidant activity in egg white proteins.

3.5.3. OH •free radical scavenging capacity

An assessment was conducted to determine the presence of antioxidant capacity, with the OH• radical scavenging capacity being measured in the process. Furthermore, the investigation sought to ascertain the substance's ability to protect. As demonstrated in Fig. 5C, at the G0 stage, TF exhibited a significantly higher OH• radical scavenging capacity (17.91%) in comparison to the other three groups. However, at the G120 stage, the capacities of TS and TG were significantly higher than that of TF (37.56%), while the OH• radical scavenging capacity of TM (36.66%) remained lower than that of TF. At the I240 stage, the OH• radical scavenging capacity of all heat-treated samples was higher than that of TF (60.59%), indicating that heat treatment may cause the breakage of disulfide bonds, hydrophobic bonds, and hydrogen bonds in proteins. This reduces the hydrogen bonding between N—H and C O groups within the molecule, resulting in fewer disulfide bonds and lower protein aggregation (Guo et al., 2019). This, in turn, provides more enzyme-cleavage binding sites, releases more physiologically active amino acids (Xie et al., 2019), and offers more radical reaction sites. The ABTS free radical scavenging capacity showed consistent results.

3.5.4. Total reducing capacity

T-AOC is key to evaluating food safety and health monitoring by characterizing the ability of antioxidants to scavenge reactive oxygen species. As demonstrated in Fig. 5D, the reducing potential of the four samples showed an increasing trend as the digestion reaction progressed, with a significant (P < 0.05) increase in the total reducing potential of the thermally processed samples compared to the raw samples at the I240 stage. This may be due to the formation of non-enzymatic browning products with antioxidant capacity, primarily melanoidins. These brown, high-molecular-weight polymers are end-products of the Maillard reaction. Their antioxidant activity is believed to stem from their complex chemical structure, which can act as a metal chelator and a free radical scavenger. Therefore, the thermal processing not only alters protein structure but also generates these novel bioactive compounds, contributing to the overall antioxidant profile of the cooked tilapia (Pérez-Burillo et al., 2019). Moreover, it is noteworthy that hydrolysates have the potential to contain amino acids, peptides, or other molecules that possess hydrogen-donating capacity. This property contributes to the formation of more stable products when they react with free radicals. Consequently, this reaction effectively terminates the free radical chain reaction (Borges et al., 2022).

3.6. Determination of hypoglycaemic activity

3.6.1. α-glucosidase inhibitory activity

The evaluation of the subject's capacity to reduce glucose levels In Vitro was undertaken through the measurement of its α-glucosidase inhibitory action. As shown in Fig. 6A, the α-glucosidase activities of TS, TG, and TM were significantly lower than that of TF by 10.32% at the G0 phase. However, this trend was reversed at the G120 and I240 phases, with all three exceeding that of TF by 42.72%. This is likely due to the fact that the enzymatically cleaved small molecule peptides competed for active sites with the enzyme, preventing carbohydrate attack and hindering hydrolysis, thereby preventing the entry of carbohydrates into the bloodstream (Mwalimu & Packirisamy, 2023). Among the three distinct groups of samples subjected to varying thermal processing conditions, the activity of TG reached 51.48%, a figure that was significantly higher compared to those observed in the TS and TM groups. This outcome suggests that thermal processing conditions may have a minor impact on the inhibition of glucosidase. This claim is corroborated by research undertaken by Paul et al. (2024), which determined that the heat treatment temperature of fruit samples exhibited a divergent effect on the augmentation of glucosidase inhibition.

Fig. 6.

(A) α-Glucosidase inhibitory activity of tilapia proteins at various stages of digestion under different heat treatment conditions; (B) α-Amylase inhibitory activity of tilapia proteins at various stages of digestion under different heat treatment conditions. The different letters indicate significant differences (P < 0.05). TF: Fresh Tilapia Protein, TS: Steamed Tilapia Protein, TG: Baked Tilapia Protein, and TM: Microwave Tilapia Protein.

3.6.2. α-amylase inhibitory activity

The hypoglycemic activity of the four samples was further assessed by analyzing their inhibitory effect on alpha-amylase, as both alpha-amylase and alpha-glucosidase, as carbohydrate hydrolases, can retard the increase in blood glucose and inhibit the development of postprandial hyperglycemia. As demonstrated in Fig. 6B, the α-amylase inhibiting activities of TS and TG before digestion (G0) were not significantly different but were significantly lower than that of TF (12.35%), while TM had the lowest activity among the four groups at 9.07%. At the G120 stage, the activities of the TG and TS groups increased considerably (P < 0.05), possibly due to the low molecular weight (MW) and high density characteristics that prevent enzyme digestion and influence the enzyme's active site (Ge et al., 2023). The α-amylase inhibitor properties observed in all samples were found to reach their maximum at the I240 stage. Furthermore, the activities of TS, TG, and TM were found to be substantially higher than that of TF, the lowest of the four groups, with TM being significantly higher than TF, proving that thermal processing could enhance the α-amylase inhibitory activity (Ge et al., 2023).

3.7. ACE inhibitory activity

ACE is a zinc protease that has been shown to convert the inactive form of decapeptide into octapeptide, an effective vasoconstrictor (Carrera-Alvarado et al., 2024). Additionally, ACE has been demonstrated to inactivate the catalytic function of bradykinin, thereby playing a crucial role in blood pressure regulation. Subsequent analysis of the inhibitory effects of the four samples on ACE was conducted to evaluate their blood pressure-regulating activity. As demonstrated in Fig. 7, all samples exhibited low ACE inhibitory activity prior to digestion (G0), with the heat-processed group showing significantly lower activity than the fresh group. However, at the G120 stage, the activity of the TS and TG groups significantly increased (P < 0.05), markedly higher than that of TF. Some scholars have hypothesized that the ACE inhibitory activity of peptides may be closely related to their specific amino acid composition (Jamdar et al., 2012). The content of free amino acids (see Fig. 2) in the processed group was significantly higher than that in the fresh group, which may be one of the reasons for the increased activity. The ACE inhibitory activity of all samples reached its peak at the I240 stage, showing widespread ACE inhibition (>60%). However, a significant increase in ACE inhibitory activity was observed in the TS, TG, and TM groups when compared to the TF group, suggesting that thermal processing significantly enhances ACE inhibitory activity. Current research has demonstrated that thermal processing not only alters nutritional components but also induces protein unfolding and aggregation (Zhang & Chang, 2019). This finding indicates that thermal processing facilitates the release of active peptides and amino acids, a conclusion that is corroborated by the results presented in Section 3.1.

Fig. 7.

ACE inhibitory activity of tilapia proteins at various stages of digestion under different heat treatment conditions. The different letters indicate significant differences (P < 0.05). TF: Fresh Tilapia Protein, TS: Steamed Tilapia Protein, TG: Baked Tilapia Protein, and TM: Microwave Tilapia Protein.

3.8. Bioactivity analysis

In order to comprehensively and visually assess the bioactivities of tilapia proteins, we used a spider-web plot method to quantify physiological activities, including antioxidant and hypoglycemic activities. As demonstrated in Figs. 8A-B, at the G120 stage (Fig. 8A), among the tilapia proteins subjected to four distinct treatments, TG exhibited the greatest α-glucosidase and α-amylase inhibitory capacities, as well as the highest ACE inhibitory capacity, while TS possessed the highest ABTS, OH• radical trapping, and total reduction capacities. At the I240 stage (Fig. 8B), TG exhibited the greatest α-amylase and α-glucosidase inhibitory activities, while TS exhibited the greatest ABTS, DPPH, and OH• radical scavenging capacities. As illustrated in Fig. 2A, there was a significant increase in amino acid release of tilapia protein following steaming and baking compared to the raw control group. This assertion is corroborated by the findings of Gallego et al. (2020), who demonstrated that Phe, Pro, His, Trp, and Tyr possess the capacity to enhance scavenging activity by stabilizing the free radical structure through various forms of interaction and through the hydrophobic and spatial characteristics of the moieties. As illustrated in Fig. 1B, the molecular weight (MW) distribution of the proteins shifted toward lower molecular weight fractions as the digestion progressed, with all samples exhibiting peak activity at the I240 stage. This demonstrates that low molecular weight protein peptides can exhibit higher activity, emphasizing the importance of peptide size in bioactivity (Yu et al., 2025). Consequently, the augmented level of activity post-digestion can be ascribed to the release of elevated quantities and levels of peptides, or to the intrinsic properties of the freshly synthesized peptides (Zaky et al., 2022). In general, the steamed and baked groups demonstrated the highest bioactivity, with the microwave group exhibiting the next highest level and the raw group showing the lowest level.

Fig. 8.

(A) Radar plot of bioactivity of tilapia protein during gastric digestion phase; (B) Radar plot of bioactivity of tilapia protein during enteric digestion phase.

3.9. Relevance analysis

To further elucidate the impact of amino acid composition on the biological activity of tilapia protein, a Pearson correlation analysis was conducted. As illustrated in Fig. 9A, following gastric digestion, amino acid scores and related parameters exhibited a significant influence on biological activity (P < 0.05). Specifically, a significant positive correlation was identified between free amino acids (FAA) and antioxidant activity indicators (P < 0.05). This aligns with previous studies confirming that protein hydrolysates rich in Ala, Leu, and Pro exhibit potent free radical scavenging activity via direct electron transfer (Ketnawa et al., 2018). From the G0 to G120 stage, the substantial increase in essential amino acid release corresponded with a concurrent enhancement in biological activity. Correlation analysis at the I240 stage (Fig. 9B) revealed that the degree of hydrolysis significantly affects biological activity (P < 0.05). Furthermore, both FAA and the essential amino acid index (EAAI) demonstrated strong correlations with DPPH and ABTS radical scavenging activities, as well as ACE inhibitory activity (P < 0.05), with the strength of these positive correlations increasing alongside the elevated activity levels. Additionally, as shown in Fig. 3, biological value (BV) is robustly correlated with amino acid release and scores. Collectively, these findings indicate that as the digestion process progresses, the protein quality, digestibility, and biological activity of all four samples gradually improve.

Fig. 9.

Correlation analysis between digestive properties, amino acid scores and bioactivity of tilapia proteins. (A) Gastric digestion stage; (B) Intestinal digestion stage.

4. Conclusion

This study compared the protein quality and bioactivity of tilapia under raw and three thermal processing conditions. It was demonstrated that varying thermal processing conditions exerted divergent effects on protein quality and bioactivity, with the baked condition exhibiting optimal digestibility and the steam condition generating the most significant release of essential amino acids. Results from computer simulations and In Vitro assays showed that thermal processing significantly enhanced the bioactivity of tilapia proteins, with baking processing having the highest α-amylase and α-glucosidase inhibitory activities and steaming having the highest scavenging capacity for ABTS, DPPH, and OH radicals. In summary, thermal processing leads to a decrease in molecular weight while increasing the content of bioactive amino acids in the hydrolysate. Therefore, thermal processing plays a crucial role in enhancing protein nutrition, and appropriate thermal processing conditions can maximize the biological activity of tilapia protein. However, computer simulations and In Vitro models cannot fully replicate the complex physiological environment within the body. The true physiological effects still require validation through animal studies or clinical trials to provide more compelling evidence for the development of functional foods.

CRediT authorship contribution statement

Zirui Fu: Writing – original draft, Methodology, Formal analysis, Data curation. Ya Wei: Writing – original draft, Methodology, Investigation. Chunsheng Li: Validation, Methodology, Data curation. Huan Xiang: Supervision, Resources, Data curation. Jianwei Cen: Validation, Supervision, Methodology. Shengjun Chen: Supervision, Resources. Yongqiang Zhao: Validation, Supervision, Resources. Xiao Hu: Validation, Methodology. Yanyan Wu: Validation, Resources, Investigation. Jun Li: Validation, Resources, Methodology. Hui Huang: Writing – review & editing, Writing – original draft, Resources, Investigation, Funding acquisition. Shuxian Hao: Writing – review & editing, Visualization, Supervision, Resources, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material

Data availability

Data will be made available on request.