3.1. Chemical composition of quinoa protein concentrate and hydrolysis degree
The isolation procedure of quinoa proteins from quinoa flour involves the solubilization of proteins in alkali and the isoelectric precipitation at acidic pH [21, 28]. Table 1 shows the proximal composition of QPC extracted at pH 9.0 and precipitated at pH 4.5. The protein content was close to that informed by Aluko & Monu [23] and Rueda et al. [22] (∼60%) and lower than the value obtained by Abugoch et al. [29] (∼77%); all extracted at pH 9.0. Other researchers showed higher protein content in quinoa protein isolates, reaching values between 75 and 89%, with an extraction pH of 10.0 and 9.5, respectively [30, 31]. The ash content doubled the one reported by Abugoch et al. [29]; while the mean value informed by Kaspchak et al., X. Wang et al. and L. Wang et al. [30–32] was around 2%. Besides, for these authors, moisture oscillated between ∼3 and 5%; while the value reported by Abugoch et al. [29] was close to our findings (∼6%). Fat content was near to Aluko & Monu's [23] findings (9.95%), while for other researchers fat content was less than 2% [31, 33] and around 3% [34]. Nonfiber carbohydrate content was 8%, lower than the value reported by Abugoch et al. [29], at pH 9 (18.8%); while total dietary fiber reached 10%.
Table 1
Chemical composition of quinoa protein concentrate (QPC)
Moisture (g/100g) | 7.2 ± 0.5 |
Ash (g/100g*) | 5.99 ± 0.02 |
Protein (g/100g*) | 58 ± 1 |
Fat (g/100g*) | 10.7 ± 0.3 |
Dietary fiber (g/100g*) | 10.1 ± 0.1 |
Carbohydrates (g/100g*) | 8.0 ± 2.0 |
*dry basis |
Table 1
The extent of proteolysis was determined by the DH. The DH of the QPH samples was 30 ± 4%. The SDS-PAGE patterns of QPC and QPH were reported in our previous work [21].
As regards QPH, the protein content has not been determined because it is expected to be similar to that of the concentrate, as the Kjeldahl method is strictly dependent on total organic nitrogen content, wherein protein structure will not interfere with the accuracy of protein determination [35].
3.2. Antioxidant capacity of quinoa protein hydrolysates
3.2.1. Iron-(II) chelating activity
Chelation of metals has an antioxidant effect due to iron being implicated in Fenton reactions and catalyzes the generation of reactive oxygen species, promoting lipid oxidation and oxidative damage at different levels [36]. The iron-(II) chelating capacity of QPH was 74 ± 2%, significantly higher than for QPC (62 ± 3%) (p = 0.004). Depending on the protein source, enzyme type, and the hydrolysis treatment conditions, the chelating properties may be favored or unfavored [24]. The metal chelating activity of polypeptides may be attributed to the interaction with charged amino acid residues, or through iron binding by peptides [37]. In our case, hydrolysis might expose histidine, aspartic acid, and glutamic acid residues which were recognized as the main determinants in iron chelating activity [16, 38, 39], which may explain the increase in the activity.
Regarding other vegetable protein sources, Pantoa et al. [40] studied the bioactivities of young rice hydrolysates and they showed that the iron-chelating activities of alcalase hydrolysates (DH ∼15–30%) were considerably higher than native protein, regardless of the maturation stage and hydrolysis time, but with a high correlation with DH. Besides, bromelian treated-black bean protein hydrolysates (4 h) had a higher capacity to bind ferrous ions than unhydrolyzed samples which required much higher concentrations to chelate iron [16]. Y. Zhang et al. [41] also studied the iron-chelating peptides from mung beans. Their results indicated that the mung bean hydrolysates prepared by alcalase had high ferrous ion chelating ability (50.25 ± 0.64, DH 25%) as compared to those obtained by other enzymes like papain and neutral protease [41]. Alcalase can effectively cleave peptide bonds to increase the number of amino and carboxyl groups, generating more negative charges, and leading to the binding of iron (II) through electrostatic interactions [42].
3.2.2. Reducing power
The reducing power lies in the capacity to donate electrons to a stable substrate, interrupting the free radical chain reaction [43]. As shown in Table 2, the reducing power of QPH was significantly higher than for QPC (p < 0.000) and this capacity had a concentration-dependent effect for all the samples except for BHT, as shown by an increase in the absorbance at 700 nm (p < 0.000). A similar trend was observed in rice bran protein hydrolysates (protease from Bacillus licheniformis, DH 6–12%) where their ability to stabilize radical species was more pronounced as the concentration of hydrolysates increased from 5 g/L to 20 g/L [44]. In addition, Liu et al. [45] studied soy protein isolates hydrolyzed with alcalase and flavourzyme and showed that the absorbance increased from around 0.2 and 0.4 to 0.6 and above 1.0 for a concentration from 10 to 40 mg/mL of hydrolysates and peptide fractions (MW 1–5 Da), respectively. Besides, Moghadam et al. [46] evaluate the potential antioxidant activity of the walnut protein and its hydrolysates using the reducing power test. Their results showed that the reducing ability of walnut protein increased through enzymatic hydrolysis with trypsin from 0 to 120 min. The higher reducing power of the hydrolysates may be related to the increased usage of hydrogen ions generated by cleaved peptides and to free amino acids and the length and composition of peptides released [43, 47, 48].
Table 2
Reducing power (Abs at 700 nm) of quinoa protein concentrate (QPC) and hydrolysates (QPH)
Sample | Concentration (% w/v) |
| 0.5 | 1 | 2 |
QPC | 0.505 ± 0.001cB | 0.76 ± 0.05bB | 0.84 ± 0.06aB |
QPH | 0.777 ± 0.009cA | 0.92 ± 0.04bA | 1.095 ± 0.007aA |
BHT* | 1.22 ± 0.03b | 1.29 ± 0.01ab | 1.36 ± 0.01a |
*Butylhydroxytoluene (BHT) |
Different capital letters in the same column indicate significant differences (p < 0.05) between samples with and without enzymatic treatment. Different lowercase letters in the same row indicate significant differences (p < 0.05) between QPC, QPH or BHT concentrations. |
The high capacity of QPH to reduce free radicals highlights their potential to fulfill the increased demand for natural alternatives to synthetic antioxidants in the food industry. Despite the high effectiveness of BHA and BHT, among others, to slow down lipid oxidation, they are being replaced due to harmful health effects [49]. According to Table 2, the addition of BHT exhibits greater reducing power compared to the QPC and QPH samples. However, it is worth noting that the BHT concentrations tested would not be suitable for food formulation as they exceed the permitted limits [50]. In this regard, the reducing power found in the hydrolysates becomes relevant, as they offer a natural alternative to the use of BHT.
Table 2
3.3. Determination of the antioxidant capacity of emulsion gels
3.3.1 ABTS
Figure 1 shows the ABTS antioxidant activity of EG and EGH. ABTS mean value for extracts from EGH was 21 ± 3 µmol Trolox equivalents/g sample, significantly higher than that obtained from EG which was 11 ± 4 (p = 0.000). The unfolding of the protein structure through hydrolysis treatment can expose hidden amino acids increasing the oxidant quenching potential [9]. Olivera-Montenegro et al. [51] studied the effect of two extraction pretreatments on the antioxidant activity of quinoa protein and their hydrolysates. They found a significant increase in the antioxidant activity in ABTS assay with the enzymatic hydrolysis using Bacillus spp. (DH ~ 24%), being around 40% higher than native protein for conventional extraction. Mudgil et al. [35] also explored the bioactive properties of hydrolysates produced from quinoa protein isolates. Their results showed that after hydrolysis with different enzymes (bromelain, chymotrypsin, and protease; DH > 50%), a significant increase in the ABTS radical scavenging activity was achieved; while unhydrolyzed samples had the lowest antioxidant activity. Similar findings were obtained by Rueda et al. [22] who demonstrated that hydrolysis of quinoa protein concentrates with alcalase and protamex significantly increased the ABTS measurements at different hydrolysis times (from 30 to 180 min; DH around 20–30%), but the usage of flavourzyme reduced the antioxidant activity. Research results highlight that several factors influence the antioxidant activity of quinoa protein hydrolysates and can be responsible for the different findings. These factors include enzyme type and specificity, ratio of enzyme to substrate, sequential use of enzymes, reaction time and temperature, and size and structure of peptides [9, 11, 52].
Figure 1
3.3.2 DPPH
Results from the DPPH assay are shown in Fig. 1. The mean value of extracts obtained from EGH was 12 ± 2 µmol Trolox equivalents/g sample. The extracts derived from the EG also exhibited the capacity to reduce DPPH radicals but to a lesser extent than those emulsion gels formulated with QPH (p = 0.000). Figure 1 also shows that the increase in the QPH concentration significantly increased the antioxidant activity (p < 0.000).
Similar behavior was observed for quinoa protein concentrate dispersions after hydrolysis by pancreatin enzyme at different times. Results showed that the hydrolysates obtained (DH about 20%) significantly reduced DPPH radicals compared to unhydrolyzed samples, mainly those obtained from 6 h hydrolysis [10]. The improvement in the ability of hydrolysates to reduce DPPH free radicals may be due to the disruption of native protein structure and the exposure of inner amino acids and patches, becoming them more capable of interacting with the DPPH radicals, which are hydrophobic [43, 53].
Galante el a. [54] also formulated acid-induced gels with quinoa protein concentrates and their hydrolysates obtained using a serin protease from Aspergillus niger. According to their results, all the samples exhibited high antioxidant activity against DPPH radicals. Still, there were no significant differences among the different hydrolysis times evaluated (from 0 to 180 min, DH from 8 to 17%).
It is important to bear in mind that even though the antioxidant activity determined by DPPH method was lower than the ABTS results, as was expected for the lower solubility of proteins/peptides in ethanol, it was observed a high correlation between both assessments (Pearson's correlation coefficient 0.925, IC 95% for p 0.806; 0.972, p < 0.000).
3.4. Oxidative stability
Lipid oxidation commonly occurs in oil-in-water emulsion systems and causes a decrease in both nutritional and sensorial product quality [55]. Figure 2 shows that QPH were more effective in depressing the oxidation of high-oleic sunflower oil than QPC. After 30 days of storage, EGH showed lower levels of malondialdehyde (p < 0.000) and the concentration of this secondary oxidation product decreased as increased QPH concentration (p < 0.000). Based on the results presented in section 3.2, the QPH demonstrated better iron chelation/reducing characteristics compared to QPC. Therefore, the enhancement in lipid stability could be attributed to the fact that QPH adsorbed at droplet surfaces or dispersed in the aqueous phase of the EGH, which could repulse or chelate metals and thus improve the oxidative stability of the systems.
The greater oxidative stability presented by the hydrolysates compared to the non-hydrolyzed proteins has also been reported by other authors. Cheetangdee & Benjakul [44] investigated the effect of rice bran hydrolysates (viscozyme-L from Aspergillus sp., α-amylase from A. oryzae and protease from Bacillus licheniformis, DH 0 to 12%) on lipid oxidation of oil-in-water emulsions stored for up to 15 days. They concluded that the oxidative stability of soy oil became higher than the control without hydrolysis, and also the efficiency in retarding oxidation was hydrolysate concentration-dependent, as shown by the lower thiobarbituric reactive substances obtained. Moreover, Gomes & Kurozawa [53] studied the functional and antioxidant properties of rice protein hydrolysates with alcalase and flavourzyme (DH 1–10%). They found that rice peptides derived from alcalase treatment were more effective in reducing linseed oil oxidation when compared with the protein hydrolyzed with flavourzyme. That may be due to the release of peptides by the action of alcalase, with the exposure of amino acids with more antioxidant capacity at the surface, as well as the generation of new antioxidant peptides [56]. This strong inhibition capacity on lipid oxidation may be related to the amphiphilic behavior of peptides which could distribute at both the dispersed and the aqueous phases of emulsions [57]. Then, the inhibition effect of protein hydrolysates in oil-in-water emulsions depends on several factors like enzyme type, amino acid profile, polypeptide sequences as well as the distribution of hydrolysates in the aqueous phase or the interfacial area [57, 58].
Figure 2
3.5. Rheological properties
At the beginning of the time sweep test, samples showed typical fluid-like behavior (emulsion systems), where Gʺ was greater than Gʹ. After the addition of CaCl2, at the 50 s, all emulsions underwent sol to gel transition where both moduli instantaneously increased and reached a plateau value. At this time Gʹ values were above Gʺ, indicating a solid-like behavior and thus, a protein network formation (Fig. 3). EGH reached significantly lower Gʹmax values than the EG (p = 0.000). This finding may be related to the lower molecular weight of the hydrolysates which led to a random aggregation of the suspension resulting in a coarser and weaker gel structure. This behavior was observed in soy protein gels where the addition of 0.5 and 0.75% of hydrolysates resulted in a weakening of the gel mesh [59], and egg white hydrolysate gels [60]. Moreover, Fig. 3B also shows that Gʹmax values were higher as QPH concentration increased from 0.5 to 2% (p = 0.000), which could be attributed to a more pronounced crosslinking of the continuous phase when the concentration of these peptides increased in the medium. Felix et al. [61] found a similar protein concentration dependence in quinoa protein gels, where the weakest gels were obtained at the lowest concentrations of the polypeptides (Gʹ at 1 Hz ∼15 Pa and ∼4 Pa for gels with 300 and 150 g/kg quinoa protein, respectively).
Figure 3
A stress sweep test was conducted on gelled samples. Gʹ and Gʺ as a function of strain amplitude are reported in Fig. 4. All emulsion gels showed similar behavior. At small strains, both moduli were independent of the strain amplitude; where the physical bonding network seems to be in a dynamic equilibrium and the sample structure behaves as a viscoelastic solid [62]. This test also confirmed that the 0.3% strain, chosen for time, frequency, and temperature sweep tests, was within the linear viscoelastic region. As the strain increased beyond the linear viscoelastic region, both Gʹ and Gʺ started to decrease markedly indicating the onset of structural breakdown, where the rate of breaking bonds increases faster than the rate of reformation. This behavior is typical for a colloidal gel and was also observed in thermally-induced quinoa protein gels [63, 64] and in soy protein isolate emulsion gels induced by NaCl [65].
Figure 4
The viscoelastic response of all the emulsion gels as a function of frequency is shown in Fig. 5. The Gʹ was greater than the Gʺ throughout the entire frequency range, which means that the elastic component was dominant compared to the viscous character, thus deformation in the linear range was mainly recoverable. Besides, both Gʹ and Gʺ were only weakly dependent on the frequency, which is indicative of typical gel-like behavior [66].
Figure 5
To further evaluate the strength of the emulsion gels, the frequency dependence of G′ of the samples was evaluated with the power-law model. The resulting parameters (kʹ and nʹ) are summarized in Table 3. For all samples the correlation coefficient R2 ≥ 0.96, reflecting that this model fitted well for the Gʹ behavior. The k′ magnitudes show the strength of the interactions in the emulsion gels [67] and were significantly higher for EG (p = 0.002). Hydrolysis treatment may alter the hydrophobic strength and the electrostatic charges between protein-protein and protein-fat bindings, thus weakening the gel matrix [68]. Besides, kʹ values showed that as the QPH concentration increased, the Gʹ values increased, being EGH with 1% and 2% of QPH those which reached the highest Gʹ values (p = 0.001). The exponent n magnitude can provide data about the physical properties of the emulsion gel networks [69]. When exponent n = 0 denotes an elastic covalent gel, exponent n > 0 indicates that the emulsion is composed of non-covalent physical crosslinking [70]. nʹ values here obtained were not zero (though very low), which suggests that non-covalent interactions play a role in the structure of the emulsion gels [71]. However, this physical gel behavior does not mean there are no covalent bonds [72]. nʹ values were significantly higher for the EGH (p = 0.001).
Table 3
Rheological parameters for emulsion gels with 0.5, 1, and 2% (w/v) of quinoa protein concentrate (EG) and quinoa protein hydrolysates (EGH)
Sample/Concentration | kʹ (Pa.sn) | nʹ | Gʹ (kʹ.ωn)* | R2 |
EG-0.5 | 1357 ± 52aA | 0.0555 ± 0.0001cB | 1542 ± 59aA | 0.96 |
EG-1 | 1300 ± 18aA | 0.0585 ± 0.0006bB | 1487 ± 19aA | 0.96 |
EG-2 | 1341 ± 25aA | 0.0640 ± 0.0009aB | 1555 ± 32aA | 0.97 |
EGH-0.5 | 109 ± 2bB | 0.061 ± 0.007bA | 126 ± 4bB | 0.96 |
EGH-1 | 167.8 ± 0.6aB | 0.071 ± 0.003abA | 197.8 ± 0.8aB | 0.98 |
EGH-2 | 184 ± 12aB | 0.079 ± 0.005aA | 221 ± 13aB | 0.98 |
*ω = 10 rad.s− 1 |
Different capital letters in the same column indicate significant differences (p < 0.05) between samples with and without enzymatic treatment. Different lowercase letters in the same column indicate significant differences (p < 0.05) between QPC or QPH concentrations. |
Table 3
Figure 6 shows the effect of temperature on the modification of the viscoelastic properties of the EG and EGH at different QPC or QPH concentrations. In all the emulsion gels, Gʹ and Gʺ display a slow upward trend throughout the entire course of heat treatment, which were almost linear within the 20–80 ºC temperature range. The increase in temperature may promote the strengthening of interactions between the protein segments, which can arrange covalent and non-covalent bonds, which cause a three-dimensional network and the reinforcement of the gelled samples [73, 74]. In addition, the EG displayed the highest G′ and G′′ values during the thermal treatment compared to EGH. A similar trend was observed in soy protein hydrolysates produced by alcalase during 5 to 180 min, followed by cross-linking with transglutaminase (DH 1.37–6.61%), but substantially lower final G′ values were achieved (around 10 Pa) [75].
Figure 6
Besides, the variation of Gʹ (after and before thermal treatment, ΔGʹ) was significantly lower for the EGH than for EG (p < 0.000) indicating a lesser reinforcement of the gel matrix. Partial hydrolysates are probably composed of a mixture of peptides/proteins capable of gel and non-gelling peptides. Thus, to reach a similar content of gelling peptides, it is necessary to have a higher concentration of these hydrolysates in the samples in comparison with those formulated with native protein [76], to achieve a similar network strengthening during heat treatment. With the highest concentration of QPH (2%), the final Gʹ reached in EGH, after the thermal treatment, was around 80 Pa below the final Gʹ reached in EG with the lowest concentration of QPC (0.5%). Although the strengthening of the EGH was lesser than EG, it is noteworthy to observe that the gel structure remained stable through the entire temperature range. Thermal treatments could induce intermolecular modifications in the emulsion gel structure. Hence, the assessment of the thermal stability of the emulsion gels which will be used in the formulation of products that need heating is of great interest (Zhu et al., 2022).