Effect of pH extraction on protein content
BSA equivalent concentration in the liquid phase at different pH, showed a high alkaline medium, enabling more protein to be dissolved. However, an increase in pH might harm the essential amino acids might reduce digestibility and biological value (Gerzhova et al. 2016). So, pH = 10.0 was used for the protein extraction. Traditionally alkaline extraction is followed by isoelectric precipitation. Our results showed that the pattern of protein solubility of pomegranate seed proteins was U-shaped with an apparent dip in solubility around pH 4–5, especially at pH 4.4. These results are consistent with the reports of most research in which the isoelectric pH range for most plant proteins was considered to be at pH = 4–5 (Gerzhova et al. 2016; Olivares-Galván et al. 2020). According to the results, the pomegranate seed protein concentrate contains high protein content (70.2 ± 0.36%).
Optimization of enzymatic hydrolysis using RSM
The pomegranate seed protein was hydrolyzed using alcalase, to obtain the optimum condition of hydrolysis. Alcalase is an endopeptidase and a low-specificity food-grade protease to releases several short peptides at a cost-effective rate (Olivares-Galván et al. 2020).
The selected experimental combinations were presented in Table 2. For the response variables, second-order models were developed that included constant, linear, quadratic, and interaction terms.
Table 2
Central composite arrangement for independent variables
| Variable Levels | Response |
Run | X1 | X2 | X3 | Y1 | Y2 |
1 | 50 | 105 | 2 | 87.1 | 0.512 |
2 | 45 | 180 | 1 | 83.9 | 0.52 |
3 | 50 | 180 | 2 | 86 | 0.5 |
4 | 50 | 105 | 2 | 85.6 | 0.5 |
5 | 50 | 105 | 2 | 86 | 0.52 |
6 | 45 | 30 | 1 | 82.9 | 0.49 |
7 | 55 | 30 | 1 | 78 | 0.42 |
8 | 55 | 105 | 2 | 77.9 | 0.44 |
9 | 45 | 105 | 2 | 83 | 0.51 |
10 | 50 | 30 | 2 | 86.7 | 0.516 |
11 | 45 | 180 | 3 | 82.1 | 0.502 |
12 | 50 | 105 | 2 | 88.6 | 0.51 |
13 | 45 | 30 | 3 | 80 | 0.48 |
14 | 50 | 105 | 2 | 84.7 | 0.519 |
15 | 55 | 180 | 1 | 75.4 | 0.39 |
16 | 50 | 105 | 2 | 86.6 | 0.511 |
17 | 55 | 180 | 3 | 70 | 0.4 |
18 | 50 | 105 | 3 | 84.8 | 0.523 |
19 | 55 | 30 | 3 | 71 | 0.41 |
20 | 50 | 105 | 1 | 87 | 0.513 |
X1 (temperature, °C), X2 (time, min), X3 (enzyme to substrate ratio, %) and their responses Y1(DPPH scavenging, %) and Y2 (Ferric reducing power, absorbance at 700 nm)
The results of the analysis of variance (ANOVA) demonstrate that all statistical models are significant (p < 0.01). The lack of fit tests, which indicate the fitness of the models obtained, was non-significant. The coefficients of variance were less than 10% which meant that the models were considered reproducible and can be used to optimize hydrolysis conditions (Table 3).
Table 3
ANOVA for the response surface quadratic polynomial model
| Y1 | Y2 |
Source | F Value | p-value | coefficient | | F Value | p-value | coefficient |
Model | 33.52** | < 0.0001 | | | 37.10** | < 0.0001 | |
X1 | 89.51** | < 0.0001 | -3.96 | | 175.16** | < 0.0001 | -0.044 |
X2 | 0.082 NS | 0.7802 | -0.12 | | 0.014 NS | 0.9070 | -0.0004 |
X3 | 21.26** | 0.0010 | -1.93 | | 0.29 NS | 0.6017 | -0.0018 |
X1X2 | 3.20 NS | 0.1038 | -0.84 | | 9.49* | 0.0116 | -0.011 |
X1X3 | 4.23 NS | 0.0667 | -0.96 | | 0.88 NS | 0.3707 | + 0.0035 |
X2X3 | 0.52 NS | 0.4873 | + 0.34 | | 0.16 NS | 0.6963 | + 0.0015 |
X12 | 70.08** | < 0.0001 | -6.68 | | 52.79** | < 0.0001 | -0.046 |
X22 | 0.96 NS | 0.3504 | -0.78 | | 4.34 NS | 0.0638 | -0.013 |
X32 | 2.38 NS | 0.1538 | -1.23 | | 0.26 NS | 0.6185 | -0.00327 |
Lack of Fit | 0.94 NS | 0.5284 | | | 3.26 NS | 0.1104 | |
R-Squared | 0.9679 | | | | 0.9709 | | |
Adj R-Squared | 0.9390 | | | | 0.9447 | | |
C.V. % | 1.61 | | | | 2.18 | | |
**: p < 0.01, *: p < 0.05, NS: Non-significant, X1: Temperature, X2: Time, X3: Enzyme to Substrate ratio, Y1: DPPH scavenging power, Y2: ferric reducing power, C.V.: Coefficient of variance |
Regression coefficients showed a significant effect (p < 0.01) of temperature (X1) in linear and quadratic terms and E/S ratio (X3) in linear terms on the DPPH scavenging power values. The coefficient of determination values was 0.967. The predicted R2 of 0.868 agrees with the adjusted R2 of 0.939. Also, regression coefficients showed that temperature had a linear and a quadratic highly significant effect on the ferric reducing ability values, and the interaction terms of temperature and time (X1X2) had a significant (p < 0.05) effect. The coefficient of determination values was 0.970. The predicted R2 of 0.805 is reasonable, with the adjusted R2 of 0.944.
A high R2 value, a lack-of-fit that was non-significant, and an adjusted R2 near 1 suggest these models are applicable for explaining a relationship between the independent variables and the responses.
Analysis of response surface graphs
According to the model, three-dimensional response surface graphs were drawn by varying two independent variables and keeping another independent variable at the central point (Fig. 1).
To investigate the effect of temperature as one of the independent variables, in Fig. 1, diagrams (a) and (b), which respectively show the interaction of temperature-time, and temperature-ratio of the enzyme to the substrate on DPPH scavenging activity, and Also, diagrams (e) and (d), which respectively show the interaction of temperature-time, and temperature-ratio of the enzyme to the substrate on ferric reducing power were studied. Studies have shown that enzymatic reactions are affected by temperature. The optimal temperature for enzyme activity can differ for different substrates, enzyme stability, substrate availability, and by-product formation may affect enzymatic reaction rate (Singh et al. 2018). In current study, Up to 48.8°C, the DPPH scavenging and ferric reducing power increased but then decreased rapidly. In general, the hydrolysis rate goes up with temperature because of exposed peptide bonds. At a lower temperature, the reduced antioxidant activity could be due to incomplete hydrolysis, and at a higher temperature, it could be due to thermal denaturation and the loss of enzyme activity (Singh et al. 2018; Wang and Shahidi 2018). Similar behavior was observed for the proteolysis of bovine plasma protein (Seo et al. 2015), turkey meat (Wang and Shahidi 2018), and Mantle of Cuttlefish (Hamzeh et al. 2019).
A non-linear effect of time on DPPH scavenging power and ferric reducing power was observed in the range of 30–180 min by studying the corresponding interaction diagrams (Fig. 1 (a, c, d, f)). An increase in DPPH scavenging activity and ferric reducing power is achieved by increasing reaction time up to certain levels and then slightly decreasing. This may be because longer treatment results in the hydrolysis of the antioxidant peptides into the production of hydrophilic smaller peptides, which have a lower ability to quench a hydrophobic DPPH molecule and deactivate the enzyme during a long period (Zhuang et al. 2013; Singh et al. 2018; Hamzeh et al. 2019). Similar dependence has been observed for hydrolytic reactions of Mantle of Cuttlefish (Hamzeh et al. 2019), rice bran (Singh et al. 2018), and turkey meat (Wang and Shahidi 2018). The optimal enzymatic hydrolysis period was demonstrated to be 97.5 min by this study.
As shown in Figs. 1 (b, c, e, f), the non-linear effect on DPPH scavenging activity and ferric reducing power, was revealed by the ratio of E/S. Increasing the E/S ratio (up to 4) led to hydrolysates with higher antioxidant activity, but increasing the E/S ratio further did not result in a significant increase in antioxidant activity (P > 0.05).
Increased enzyme concentration increases the availability of enzyme active sites, leading to significant degradation of the proteins and cleavage of peptide bonds (Kurozawa et al. 2008). As E/S ratios increase to a certain extent, excessive hydrolysis decreases the chances of enzyme-substrate interactions, eventually leading to a stationary phase without evidence of hydrolysis. The profile could result from substrate oversaturation resulting in inhibition of the enzyme or auto-digestion (Singh et al. 2018). Enzymatic hydrolysis of flying squid muscle protein (Fang et al. 2012), corn gluten meal (Zhuang et al. 2013), and porcine liver (Maluf et al. 2020) exhibited similar patterns. This study showed that the optimal enzyme/substrate ratio for enzymatic hydrolysis was 1.3% (v/w).
Validation Test
Based on the combination of all the optimal regions, the maximum antioxidant activity of hydrolysates was achieved at a temperature of 48.8°C with a reaction time of 97.50 min and an E/S ratio of 1.3% (w/w). Polynomial models predicted optimal values of DPPH scavenging power and ferric reducing power at the suggested conditions. Samples were analyzed under optimal conditions to obtain experimental values. The results are shown in Table 4. The response predicted by the model showed close agreement with the experimental data. Thus, optimum conditions of hydrolysis were validated by RSM.
Table 4
Experimental and corresponding predicted values for DPPH scavenging power and Ferric reducing power using optimum values of independent variables (n = 3).
Enzyme | Temperature 0C | Time min | E/S ratio %w/w | DPPH scavenging power | Ferric reducing power |
Predicted | Experimental | | Predicted | Experimental |
Alcalase | 48.8 | 97.5 | 1.3 | 87.87 a | 88 ± 0.97a | | 0.523 b | 0.50 ± 0.83b |
Those with different letters are significantly different, with p < 0.05. Comparisons were made between the observed and predicted values for each correspondent response.
Degree of Hydrolysis
The degree of hydrolysis (DH) is an essential factor in tracing or controlling protein hydrolysis reactions, as it is related to chain lengths and peptide cleavage rates. Furthermore, DH is proportional to peptide size or structure, affecting amino acid exposures, biological activities, and the taste of peptides (Akbarbaglu et al. 2019; Cotabarren et al. 2019; Fathi et al. 2022). The DH is determined by the patterns of cleavage and enzyme specificities involved in the hydrolysis. Hydrolysis occurs when a hydrolyzing enzyme is accessible to the scissile peptide bonds. Similarly, the enzyme's affinity for binding substrates, the shape, and structure of its active sites, and how the peptide bond is oriented also play a part (Mirzapour et al. 2016). According to the results, DH of pomegranate seed protein with alcalase is 36 ± 1.2%. Similar results have been reported in the hydrolysis of Iranian wild almond (Mirzapour et al. 2016) and flaxseed protein (Akbarbaglu et al. 2019).
EC50
Antioxidant activity of the pomegranate seed protein and its hydrolysates (produced in optimized conditions) was evaluated by the ABTS, DPPH radicals scavenging, and Fe2+ chelating methods. the dose-response curve was used to estimate the EC50 value of samples. As an indicator of antioxidant capacity, EC50 is a concentration of samples that can scavenge 50% of total radicals. Thus, a lower EC50 value indicates better free radical scavenging abilities. As shown in Table 5 for all methods, hydrolysates showed a significantly lower EC50 than non-digested proteins. Other studies also have demonstrated enhanced antioxidant activity by the enzyme hydrolysis of food proteins for a variety of reasons, including the size of peptides or general hydrophobicity (Cotabarren et al. 2019).
Table 5
Comparison of EC50 using different antioxidant test
Test | Non-hydrolyzed | Alcalase hydrolysate | Ascorbic Acid |
DPPH radical scavenging activity (mg/ml) | 1.2 ± 0.07 a | 0.18 ± 0.015 b | 0.0094 ± 0.004 c |
ABTS radical scavenging activity (mg/ml) | 1.3 ± 0.15 a | 0.4 ± 0.08 b | 0.027 ± 0.001 c |
Fe chelating (mg/ml) | 1.03 ± 0.3 a | 0.22 ± 0.07 b | 0.12 ± 0.06 b |
There is a significant difference between those with different letters (p < 0.05) |
SDS-PAGE patterns of hydrolysates
The molecular weight patterns were characterized by SDS-PAGE. The electrophoretic profile of denatured pomegranate seed protein resulted in bands spanning mainly at 15–75 kDa. The majority showed a molecular weight higher than 20 kDa, and two strong bands of intensity between 20 to 25 kDa and 35 to 48kDa were identified (Fig. 3). Smaller peptide bands and lower band intensity, along with losing some of the peptide fractions, confirm the efficacy of alcalase at cleaving. The electrophoretic patterns of the hydrolysates show peptides at a molecular weight below 11 kDa. The method couldn't separate peptides smaller than 11 kDa.
Other researchers showed that alcalase could hydrolyze the proteins and produce small peptides. The enzyme hydrolysis of peptides releases bioactive peptides with a variety of physiological properties, and these low molecular weight peptides are capable of passing through the intestine and demonstrating biological properties (Mirzapour et al. 2016; Singh et al. 2018; Zang et al. 2019; Teshnizi et al. 2020).
Surface hydrophobicity
The Surface hydrophobicity of pomegranate seed protein and its hydrolysate were compared. ANS was used as a fluorescent probe to measure the fluorescence intensity. As shown in Fig. 4, the enzymatic hydrolysis greatly (P < 0.05) improved the surface hydrophobicity.
A similar trend was found in the enzymatic hydrolysis of tomato seed protein [9]. pomegranate seed protein has a low hydrophobic value since it is composed of intact and folded proteins that have the majority of hydrophobic residues in the core site rather than on the surface for stability. Enzymatic hydrolysis may affect the surface hydrophobicity, so the increase in surface hydrophobicity may be explained by the exposure of buried hydrophobic groups. As a result, when polypeptide chains are broken down, nonpolar amino acids are exposed, causing an increase in hydrophobicity. Hydrophilic acidic and basic amino acids, when located close together, can neutralize each other electrostatically, which also results in greater hydrophobicity. In hydrolyzed proteins, this factor might explain the high surface hydrophobicity. However, longer hydrolysis times can reduce the hydrophobicity by removing nonpolar amino acids. Furthermore, it is thought that the high hydrophobicity of peptides enhances their ability to scavenge free radicals (Meshginfar et al. 2018; Zang et al. 2019). Our results showed that produced hydrolysates have a higher hydrophobicity and higher antioxidant power than intact protein.
Scanning electron microscopy
Figure 5 shows SEM images of hydrolyzed and unhydrolyzed samples. unhydrolyzed protein exhibited complex structures comprised of random sheets of different sizes and shapes. According to SEM images, the protein degraded into small fragments and looser structure with many folds after enzyme hydrolysis, resulting in smaller particles compared with untreated samples with the same SEM parameters. The results are similar to other studies that reported protein had degraded into small fragments, and particle size had been reduced after enzyme treatment (Islam et al. 2021; Fathi et al. 2022).