2023 年 29 巻 1 号 p. 57-69
The optimal preparatory conditions identified using random-centroid optimization revealed that the maximum hydroxyl radical antioxidant activity (63.4 ± 4.8 µmol of gallic acid equivalent/g of protein) was achieved at 70 °C; relative humidity, 45%; reaction time, 47.92 h; and glucose to soybean 11S globulin ratio, 10.66:1 (w/w). The average particle size of the oil-in-water emulsions prepared using optimally glycated, unreacted, and native soybean 11S globulin was 1.68 ± 0.11, 1.81 ± 0.18, and 2.36 ± 0.42 µm, respectively. The emulsion stability of the optimally glycated and unreacted soybean 11S globulin was superior to that of native soybean and did not exhibit complete separation after four weeks. When a sufficient amount of optimally glycated soybean 11S globulin was present in the emulsion, oil oxidation was significantly suppressed, suggesting the potential use of optimally glycated soybean 11S globulin as an antioxidant.
Soy proteins are widely used in the food industry because of their excellent nutritional value, processability, and biological functionality (Yamauchi, 1994), and several researchers have recognized that they acquire functional alterations (e.g., improved solubility, foaming, emulsifying ability, and emulsifying stability) by glycation using the Maillard reaction (Achouri et al., 2005; Tian et al., 2011; Li et al., 2019). On the other hand, in several further studies (Nishimura et al., 2011; Isono et al., 2012; Nishimura and Saeki, 2016; Nishimura and Saeki, 2018a; Nishimura and Saeki, 2018b; Nishimura et al., 2019; Nishimura et al., 2020; Nishimura and Saeki, 2021), we demonstrated that protein glycation confers antioxidant capacity. These findings led us to consider the development of emulsifiers with antioxidant capacity or natural product-derived antioxidants by glycation of soybean 11S globulin using the Maillard reaction. Indeed, in a previous study (Nishimura et al., 2022), glucose-conjugated soybean 11S globulin was prepared by mixing soybean 11S globulin and glucose at a ratio of 1:11.7 (w/w) for up to 72 h at 52 °C and 38% relative humidity (RH), resulting in several functional alterations such as improved solubility in low ionic strength medium (pH 6.4), antioxidant ability, and emulsifying properties. Based on these results, the glycation of 11S globulin could produce a novel emulsifier with antioxidant properties.
Several studies have employed various types of reducing sugars as carbonyl compounds, and chicken myofibrillar proteins as amino compounds to investigate the preparatory conditions for glycated chicken myofibrillar proteins. Moreover, a random centroid optimization (RCO) program (Nakai, 1990; Dou, et al., 1993; Nakai et al., 2009) was used to efficiently identify the preparative conditions for glycated chicken myofibrillar proteins that exhibit the strongest antioxidant capacity and high solubility in low ionic strength medium (Nishimura and Saeki, 2016; Nishimura and Saeki, 2018a; Nishimura and Saeki, 2018b; Nishimura et al., 2019; Nishimura et al., 2020; Nishimura and Saeki, 2021). Glycated soybean 11S globulin has been observed to also acquire antioxidant capacity (Nishimura et al., 2022); and following acquisition of sufficient antioxidant capacity by glycation, it could be considered as an emulsifier with antioxidant properties.
This study sought to reveal the conditions for preparing glucose-conjugated soybean 11S globulin with maximum antioxidant activity using the RCO program, in which four factors were considered: temperature, RH, reaction time, and glucose to soybean 11S globulin mixing ratio. The emulsifying properties of optimally glycated soybean 11S globulin were then assessed, and methyl linoleate was used to determine the effect of glycated soy protein on the suppression of oil autooxidation when used as an emulsifier or antioxidant.
Materials and chemicals Soybeans were purchased from the store Fujiya Katsuobushi (Kyoto, Japan). Gallic acid (GA) was obtained from ChromaDex Inc. (Irvine, CA, USA). All other chemicals were of reagent grade, and were obtained from Nacalai Tesque, Inc. (Kyoto, Japan) or Wako Pure Chemicals Industries, Ltd. (Osaka, Japan).
Preparation of defatted soybeans and soybean 11S globulin Defatted soybeans and soybean 11S globulin were prepared as reported in Nishimura et al. (2022). The soybean 11S globulin immediately after its preparation was considered as the native one.
RCO program The RCO program (Nakai, 1990; Dou et al., 1993; Nakai et al., 2009; Nishimura and Saeki, 2016; Nishimura and Saeki, 2018a; Nishimura and Saeki, 2018b; Nishimura et al., 2019; Nishimura et al., 2020; Nishimura and Saeki, 2021) was used to determine the optimal conditions for preparing glycated soybean 11S globulin with maximum hydroxyl radical (·OH) scavenging ability (HORAC). Briefly, four experimental parameters, temperature (40–70 °C), RH (30–80%), reaction time (5–75 h), and glucose to soybean 11S globulin mixing ratio [5–15:1 (w/w)], were entered into the RCO program. Nine sets of experimental conditions within a certain search range for each factor were provided by a regulated random design. After the experiments, the residual ratios of ·OH obtained from the random search were entered into the program as responses. Based on the random search results, four sets of experimental conditions were presented for the centroid search, and the obtained response values were entered once again into the RCO program. Maps were drawn for approximating the response surface and optimizing the set of factor values according to the random and centroid search results.
Glycation of soybean 11S globulin The prepared soybean 11S globulin solution was mixed with glucose at ratios predetermined by the RCO vertex. A subsequent operation was performed as previously described (Nishimura et al., 2022) to obtain the lyophilized protein powder. The Maillard reaction between soybean 11S globulin and glucose was performed by incubating the lyophilized powder under the 13 experimental conditions defined by the RCO. An incubator/humidity cabinet (KCL-2000A, Tokyo Rikakikai Co., Ltd., Tokyo, Japan) was used to control the temperature and RH.
Only the lyophilized prepared soybean 11S globulin maintained under optimal conditions was considered as the unreacted soybean 11S globulin.
Preparation of samples for HORAC measurement The samples for HORAC measurement were prepared using a previously described protocol (Nishimura et al., 2022).
HORAC measurement The HORAC of each protein sample was measured using a chemiluminescence reader (AccuFLEX Lumi400, Aloka Co. Ltd., Tokyo, Japan) based on the Fenton reaction, where ·OH reacts with luminol, resulting in light emission (Yildiz and Demiryürek, 1998; Parejo et al., 2000). The specific method is described in Nishimura et al. (2022). Thirteen samples prepared according to their respective RCO vertices were used to calculate the residual ratio of ·OH. The antioxidant activity was measured as the relative light unit of a sample/relative light unit of the control × 100. These values were used to evaluate the RCO. The glycated soybean 11S globulin that displayed the highest antioxidant capacity is considered as the optimally glycated soybean 11S globulin.
Solubility of glycated soybean 11S globulin After glycation, the protein powder was immediately mixed with 15 mM sodium phosphate buffer (pH 7.0) and incubated at 4 °C overnight. After reaching a final protein concentration of 1.5 mg/mL, the mixture was dispersed using a high-speed blender (T-10 basic Ultra-Turrax, IKA-Labortechnik, Staufen, Germany) at 13 500 rpm for 0.5 min; this step was repeated once. A portion of this solution was separated as a total protein (A) solution, and the supernatant water-soluble protein (B) solution was separated via centrifugation at 32 000 × g for 30 min at 4 °C. All of the above operations were performed at a temperature of 4 °C or lower. To separate unreacted sugars and proteins, a 15% trichloroacetic acid solution was added to 1 mL of A and B protein solutions to achieve a collection concentration of 7.5%. The solution was maintained at room temperature for 30 min and then centrifuged at 1 000 × g for 30 min. The precipitates were air-dried overnight and dissolved in 1 M NaOH solution. Thereafter, the protein concentrations were determined using the Biuret method (Gornall et al., 1949). The solubilities of A and B proteins in each solvent were calculated using Equation 1.
 |
Determination of available lysine and fructosamine The available lysine (Hernandez and Alvarez-Coque, 1992; Saeki, 1997) and fructosamine (Johnson et al., 1983) were quantified to monitor the progress of the Maillard reaction under optimal conditions for soybean 11S globulin and glucose. The specific procedure is described elsewhere (Nishimura et al., 2022). The assays for determining available lysine and fructosamine concentrations were performed immediately after sample preparation.
Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) analysis The total protein solutions used in the solubility assay were mixed with equal volumes of 125 mM Tris-HCl (pH 6.8) containing 8 M urea, 4% SDS, and 40 mM N-ethylmaleimide. The mixture was allowed to stand overnight at room temperature. 2-Mercaptoethanol (2-ME) was added to each mixture to a final concentration of 10% (v/v), and the mixture was boiled for 3 min. Electrophoresis was carried out on a 10% acrylamide slab gel according to Laemmli's method (Leammli, 1970). Thereafter, staining was performed with Coomassie Brilliant Blue R-250.
Emulsifying properties The emulsifying properties of native, optimally glycated, and unreacted soybean 11S globulin were measured according to methods detailed in Nishimura et al. (2022).
Thiobarbituric acid (TBA) test
Emulsion preparation Oil-in-water (O/W) emulsions were prepared using a 5 wt.% oil phase and 95 wt.% aqueous phase. The aqueous phase consisted of native, optimally glycated, and unreacted soybean 11S globulin solutions (10 mg/mL protein in 15 mM sodium phosphate buffer, pH 7.0) or surfactant solution (2 wt.% Tween 20 in 15 mM sodium phosphate buffer, pH 7.0). Methyl linoleate was used as the oil phase. The oil and aqueous phases were mixed and homogenized using a high-speed blender (Microtec Co., Ltd., Chiba, Japan) at 15 000 rpm for 3 min. The average droplet diameter was further reduced using an ultrasonic homogenizer (Nihonseiki Kaisha, Tokyo, Japan) at maximum power for 2 min.
Emulsion dilution and initiation of lipid oxidation The O/W emulsions prepared as described above were diluted 16.7-fold with phosphate buffer (15 mM sodium phosphate buffer, pH 7.0). To determine lipid oxidation inhibition, the soybean 11S globulin solutions (native, optimally glycated, and unreacted) (3 mg/mL) were diluted. Dilution is convenient for the subsequent TBA tests, and additionally is required to minimize the oil droplet interactions to clearly assess the effect of FeSO4 on oil droplets. The oil phase concentration in the diluted emulsions for the TBA test was approximately 0.3% (w/w). The addition of 5 µM FeSO4 initiated the oxidation of methyl linoleate in the emulsions. The sample emulsions were subdivided into small portions (2 mL/tube) and stored at 50 °C for 7 d. The samples were removed at predetermined intervals, and the oxidation of methyl linoleate was measured using TBA tests.
Measurement of TBA-reactive substances (TBARS) TBA was measured using the method described by Buege and Aust (1978). An aliquot (0.5 mL) of oxidized emulsion was mixed with 2.0 mL TBA solution (0.375% TBA, 15% trichloroacetic acid, and 0.04% butylated hydroxytoluene in 0.25 N HCl) and 0.5 mL sodium phosphate buffer (15 mM, pH 7.0). The mixture was heated to 95 °C for 15 min. After cooling, the mixture was centrifuged at 1 700 × g for 15 min. The absorbance of the supernatant was measured at a wavelength of 532 nm using a spectrometer. The results are expressed as the amount of TBARS. The value was calculated as malondialdehyde equivalents based on the molecular absorbance of malondialdehyde (i.e., 1.56 × 105).
Protein content determination When measuring the amount of available lysine and emulsifying properties, the contents of the prepared and glycated soybean 11S globulin were determined as described by Lowry et al. (1951). The biuret method (Gornall et al., 1949) was used for the other assays, with bovine serum albumin used as a protein standard.
Statistical analyses Statistical analyses were performed using the results obtained from three independent glycated soybean 11S globulin samples. The results are presented as mean ± standard deviation of three replicates. The differences were derived using Student's t-test for comparisons between two groups and one-way analysis of variance (ANOVA), followed by Tukey's multiple comparisons test for multiple groups. All statistical analyses were performed using Microsoft Excel ver. 2016, with Ekuseru-Toukei 2010 (Social Survey Research Information Co., Tokyo, Japan). Statistical significance was set at p < 0.05.
Determination of optimal glucose-conjugated soybean 11S globulin As outlined in Table 1, RCO was used to determine the optimal conditions for producing glucose-conjugated soybean 11S globulin with the greatest ·OH scavenging ability, based on variations in temperature, RH, reaction time, and glucose to soybean 11S globulin mixing ratio (w/w). The parameters for each experiment were calculated using the RCO program (Nakai, 1990; Dou et al., 1993; Nakai et al., 2009; Nishimura and Saeki, 2016; Nishimura and Saeki, 2018a; Nishimura and Saeki, 2018b; Nishimura et al., 2019; Nishimura et al., 2020; Nishimura and Saeki, 2021) and implemented for different preparations of the glycated soybean 11S globulin. Briefly, there were nine sets of experimental parameters proposed in the random research, and the residual ratio of ·OH, which was considered as the response value for RCO, was obtained for each set of factor values. The obtained response values were included in an orderly manner into the RCO program for further centroid research, and four new sets of factor values were proposed for residual ratio of ·OH. After conducting 13 experiments during the first cycle of RCO, the minimum residual ratio of ·OH for the first cycle random research was 7.6%. Therefore, the experiments were stopped at that point and mapped to determine the optimum combination of the variables for minimum residual ratio of ·OH.
| Vertex No. | Temperature (°C) | RH (%) | Reaction Time (h) | Glucose mixing ratio to soybean 11S globulin (w/w) | Evaluation (residual ratio of ·OH) (%) | Solubility (%) |
|---|---|---|---|---|---|---|
| 1 | 55 | 40 | 23.71 | 12.59 | 85.8 | 93.2 |
| 2 | 48 | 50 | 72.31 | 9.87 | 75.4 | 73.0 |
| 3 | 70 | 45 | 47.92 | 10.66 | 7.6 | 82.6 |
| 4 | 56 | 73 | 51.09 | 5.64 | 46.3 | 92.7 |
| 5 | 68 | 38 | 8.75 | 10.7 | 59.2 | 94.8 |
| 6 | 46 | 56 | 47.17 | 12.87 | 96.4 | 84.7 |
| 7 | 54 | 76 | 53.33 | 11.69 | 77.5 | 91.2 |
| 8 | 43 | 44 | 72.02 | 5.22 | 94.6 | 89.8 |
| 9 | 53 | 39 | 15.95 | 14.64 | 96.5 | 94.5 |
| 10a) | 61 | 52 | 45.02 | 9.22 | 32.4 | 88.6 |
| 11a) | 62 | 58 | 40.27 | 9.67 | 30.3 | 94.0 |
| 12a) | 57 | 61 | 56.16 | 9.46 | 42.4 | 93.7 |
| 13a) | 60 | 52 | 45.58 | 10.73 | 33.4 | 82.7 |
All data were mapped, as shown in Fig. 1, which enables visualization of the experimental response surface and indicates data trends (Nakai, 1990; Dou, et al., 1993; Nakai et al., 2009). The arrow on each map shows the optimum value for each factor. The minimum residual ratio of ·OH was 7.6% when the four factor values were maintained at a temperature of 70 °C, RH of 45%, reaction time of 47.92 h, and glucose to soybean 11S globulin mixing ratio (w/w) of 10.66:1. In general, trend lines presented on each map can serve as an indicator of the optimum condition, and lower lines closer to the best response are more reliable in determining the trend than upper lines, in the case of minimization. The glycated soybean 11S globulins obtained from all vertices were highly soluble in low ionic strength solutions (pH 7.0) (Table 1).
Mapping results of the experiments subjected to random-centroid optimization (RCO) for scavenging ·OH. Evaluation: The comparative emission intensity of ·OH produced by the Fenton reaction in the presence of glycated soybean 11S globulin with glucose was used for the evaluation (residual ratio of ·OH) of antioxidative activity. The vertex that provided the smallest evaluation was sought. (A) Temperature. (B) Relative humidity. (C) Reaction time. (D) Glucose mixing ratio of soybean 11S globulin (w/w). (●) represents each vertex. The lines indicate the probable trends. The arrows in each graph indicate the best results.
HORAC value of the optimal glucose-conjugated soybean 11S globulin The inhibition curve for the HORAC values associated with optimally glycated soybean 11S globulin is shown in Fig. 2. Nishimura et al. (2022) reported that native soybean 11S globulin barely demonstrated HORAC in the range of 0.66–1.36 mg/mL. In contrast, the optimally glycated soybean 11S globulin reduced the production of ·OH with increasing protein concentration, and the reduction ratio reached 42.6 ± 3.6% (n = 3) at a protein concentration of 0.12 mg/mL. The half-maximal inhibitory concentration (IC50) of the optimally glycated soybean 11S globulin was 0.096 ± 0.002 mg/mL, as shown in Fig. 2.
Effect of the concentration of optimally glucose-conjugated soybean 11S globulin with glucose on hydroxyl radical (·OH) scavenging ability (HORAC).
The antioxidative abilities of the optimally glucose-conjugated soybean 11S globulin with glucose in 15 mM sodium phosphate buffer (pH 7.0) against ·OH were measured using the fluorescence method. The values are expressed as mean ± standard deviation (n = 3).
The HORAC of the optimally glycated soybean 11S globulin was converted into GA equivalent activity (µmol/mL) based on the IC50 value of GA [6.4 ± 0.1 nmol/mL (Nishimura and Saeki, 2018a)]; hence, a value of 63.4 ± 4.8 µmol of GA equivalent/g of protein was obtained. This value was approximately 7.7-fold higher than that (8.2 ± 1.6 µmol of GA equivalent/g of protein) of glycated soybean 11S globulin obtained after the 72 h reaction time employed in Nishimura et al. (2022), and about 6.5-fold higher than that (9.7 ± 0.7 µmol of GA equivalent/g of protein) of glucose-conjugated chicken myofibrillar proteins (Nishimura et al., 2019).
Boxin et al. (2002) measured the HORAC values of common fruit extracts. Comparing those values with the HORAC values of optimally glucose-conjugated soybean 11S globulin obtained in this study, the antioxidant capacity was found to be approximately two-fold greater than that of sour cherry or grape, and two-thirds greater than that of blueberry, although its value was about one-fifth that of elderberry extract.
The brown color in the optimally glycated soybean 11S globulin was darker than that observed in Nishimura et al. (2022), suggesting that the browning reaction was well advanced in the optimal reaction. Therefore, many antioxidant substances, such as melanoidin, are produced. Such production could depend on temperature. In general physicochemical reactions, the reaction rate increases 2–3-fold when the temperature increases by 10 °C. Considering this temperature coefficient (Q10), the Maillard reaction can be assumed to proceed 4–9-fold faster under optimal conditions than that found in a previous study (Nishimura et al., 2022). Although the reaction time under the optimal conditions was shortened from 72 to 47.92 h, the reaction proceeded faster owing to the increased reaction temperature, resulting in a greater antioxidant capacity. Furthermore, the solubility of the optimally glycated soybean 11S globulin in water at pH 7.0 was 82.6% (Table 1). Based on these results, this soybean could be used as an emulsifier.
Changes in the amount of available lysine and fructosamine in soybean 11S globulin during glycation The available lysine rapidly decreased during the first 6 h to 20.0 ± 3.9% (n = 3). However, the value gradually decreased to 14.6 ± 3.0% (n = 3) at 47.92 h (Fig. 3A). In contrast, the quantity of fructosamine rapidly increased during the first 6 h to 186.5 ± 15.2 µmol/g of protein (n = 3). Thereafter, the value gradually decreased to 109.2 ± 20.6 µmol/g of protein (n = 3) at 47.92 h (Fig. 3B).
Changes in the amounts of available lysine and fructosamines of the optimally glucose-conjugated soybean 11S globulin with glucose.
Lyophilized soybean 11S globulin-glucose mixtures at a weight ratio of 1:10.66 (w/w) were maintained at 70 °C and 45% RH for up to 47.92 h. The levels of available lysine (A) and fructosamine (B) were measured. The values are expressed as mean ± standard deviation (n ≥ 3).
During the reaction from 0 to 6 h, the available lysine content decreased (Fig. 3A), whereas the fructosamine content increased (Fig. 3B). As shown in a previous study (Nishimura et al., 2022), this finding indicates that the initial stage of the Maillard reaction occurred rapidly, and the lysine residue of soybean 11S globulin was conjugated with glucose to rapidly produce fructosamine. The decrease in fructosamine after 6 h could be caused by further progression of the Maillard reaction to an intermediate stage.
The relationship between changes in the amount of available lysine and fructosamine suggests the creation of glucose-conjugated soybean 11S globulin; however, this was not determined. Accordingly, conjugation was determined by SDS-PAGE analysis.
SDS-PAGE analysis of the optimal glucose-conjugated soybean 11S globulin The optimally glycated soybean 11S globulin was subjected to SDS-PAGE during the reaction (Fig. 4). The molecular weight of the soybean 11S globulin calculated from the electrophoretic mobility was about 52 000–53 000 (Fig. 4A, 0 h). The electrophoretic mobility of soybean 11S globulin gradually decreased as the reaction progressed, as has been observed in the glycation process of chicken myofibrillar proteins so far (Isono et al., 2012), and the molecular weight of this major band reached 56 000–57 000 after 6 h (Fig. 4A, 6 h). Figure 3A shows that some 80% of the available lysine was bound to glucose after 6 h. In addition, Yamauchi reported that 100 g of soybean 11S globulin contains 5.7% lysine (1984). From these results, it is calculated that approximately 2 500 of glucose is bound to 1 mol of soybean 11S globulin during the 6-h reaction, reaching about 55 000–56 000. This value is close to the value of 56 000–57 000, indicating that glucose was conjugated with soybean 11S globulin in the Maillard reaction.
Changes in the protein subunit during the Maillard reaction.
Changes in soybean 11S globulin levels in the presence of glucose were monitored in 15 mM sodium phosphate buffer (pH 7.0). Soybean 11S globulin reacted for up to 47.92 h and was then subjected to SDS-PAGE analysis in the presence (B) or absence (A) of 10% 2-ME.
The addition of 2-ME (Fig. 4B) resulted in the separation of the acidic and basic subunits, as the bond between acidic and basic subunits is attributed to SS bonding (Mikami et al., 2010). Moreover, decreases in electrophoretic mobility were observed in both subunits, suggesting that glycation occurred in both the acidic and basic subunits.
The density of the bands tended to decrease with time. This may be due to the polymerization of proteins by covalent bonds (other than SS bonds) over time, forming huge polymers that do not even enter the stacking gels.
The above results indicate that glycation occurred properly in the Maillard reaction, resulting in a strong antioxidant capacity. If the optimal glucose-conjugated soybean 11S globulin has emulsifying properties, it can be used as an emulsifier with antioxidant properties. Thus, the use of a glycated protein as an emulsifier could reduce the amount of other antioxidant additives. Accordingly, the emulsifying properties of the optimal glucose-conjugated soybean proteins were examined.
Emulsifying properties The emulsifying properties of the native, optimally glycated, and unreacted soybean 11S globulin were measured at pH 7.0 under low ionic strength. The emulsifying properties were estimated using two criteria: emulsifying ability and emulsion stability.
Emulsifying ability The O/W emulsions were prepared using three types of soybean 11S globulin solutions with corn oil, and their particle sizes were measured. The average particle size of the native soybean 11S globulin (immediately after its preparation) was the same as that reported in the previous study (Nishimura et al., 2022).
The average particle size of the emulsions prepared using native soybean 11S globulin was 2.36 ± 0.42 µm (n = 3) (Fig. 5A), while that of the emulsions prepared with optimally glycated and unreacted soybean 11S globulin was 1.68 ± 0.11 µm (n = 3) and 1.81 ± 0.18 µm (n = 3), respectively (Fig. 5B and C). The average particle size of the emulsion prepared using optimally glycated soybean 11S globulin was significantly smaller than that of the emulsion prepared using native soybean (p < 0.05) (Fig. 5A and B). Although there were no significant differences between the average particle sizes of the emulsions prepared using native soybean 11S globulin and the unreacted soybean (Fig. 5A and C), the average particle size of the emulsion prepared using unreacted soybean 11S globulin appeared to be smaller than that obtained using the native soybean. This significant reduction (Fig. 5A and B) and the tendency for a smaller (Fig. 5A and C) average particle size of the emulsion were caused by structural changes in the soybean 11S globulin, which could be attributed to the gradual loosening or unfolding of the acidic and basic subunit levels (or at the quaternary level of soybean 11S globulin) as glycation proceeded. The subunits were found to rapidly spread at the interface, resulting in high emulsification efficiency and smaller particles. On the other hand, during emulsification, once fine oil droplets are formed due to shear forces, further agitation causes the oil droplets to fuse and induce larger oil droplets. In this process, the glycated protein prevents the oil droplets from reassembling due to its sugar chains; thus, the small oil droplets are maintained and the particle size may be reduced. Notably, the smaller the particle size of the emulsion droplet, the better the emulsion (Prak et al., 2015). Accordingly, the emulsifying activity of optimally glycated and unreacted soybean 11S globulin was superior to that of the native soybean.
Particle size distributions of the emulsion of soybean 11S globulin.
The emulsifying ability of the proteins was analyzed by measuring the particle size distribution and calculating the mean droplet diameter of the emulsion samples using a light-scattering instrument. A, B, and C show the particle size distributions of the emulsion prepared using native soybean 11S globulin and optimally glycated and unreacted soybean, respectively. The values represent mean ± SD of three independent experiments.
Emulsion stability The stability of the emulsions prepared with native, optimally glycated, and unreacted soybean 11S globulin was determined. The emulsions were poured into vials (35.0 × 81.5 mm) and sealed. Thereafter, each state was visually observed at 4 °C for up to 4 weeks (Fig. 6).
Emulsion stability of soybean 11S globulin.
The stability of the emulsions was analyzed by sealing the vials containing the emulsions and maintaining them at 4 °C without agitation, and performing visual observation immediately after emulsification, and at 1 and 4 weeks. The emulsion stability of native (A), optimally glycated (B), and unreacted soybean 11S globulin (C) was determined via three independent experiments. The average emulsion stability of each soybean 11S globulin sample is shown.
Immediately after emulsification, all emulsions were dispersed and stabilized. However, the emulsion of native soybean 11S globulin completely separated into water and cream phases after one week. Phase separation of the optimally glycated soybean 11S globulin and unreacted soybean was clearly not observed after 1 week. Similar results were obtained for up to four weeks. These results indicate that optimally glycated soybean 11S globulin and unreacted soybeans efficiently formed more stable emulsions than the native soybean.
The decrease in particle size due to glycation (Fig. 5B) could have been caused by the rapid spread of subunits at the interface. Although the unreacted soybean 11S globulin emulsion resulted in a smaller particle size than the native soybean, the process evidently differed from that of optimally glycated soybean 11S globulin. A temperature of 70 °C is low for the complete denaturation of soybean protein [the denaturation temperature is above 90 °C (Morita, 2006)]. Thus, the protein may have been partially denatured during the reaction, resulting in adequate denaturation and a good balance between the hydrophobic and hydrophilic parts, thereby improving emulsification. The hydrophobic part of the molecule may be adequately exposed to the surface, enabling easier contact between oil and the hydrophobic part of the molecule. If the degree of denaturation is too high, the hydrophobic moieties will be overexposed, causing the molecules to aggregate and precipitate, thereby reducing emulsification. However, as the solubility of unreacted soybean 11S globulin was 32.5 ± 3.5% (n = 3), it was not suitable for practical use owing to its insolubility in water.
Peng et al. (2018) reported improved emulsifying properties of soybean 11S globulin conjugated with soy soluble polysaccharides using the Maillard reaction. Glucose introduced into protein molecules protrudes into the aqueous phase to form a thick hydration layer, which precludes close contact with oil droplets, thereby preventing their aggregation and coalescence (Dickinson, 2003; O'Mahony et al., 2017).
The smaller the average particle size, the more stable the emulsion (Wang et al., 2019). Thus, improvement in the emulsion stability of optimally glycated and unreacted soybean 11S globulin (Fig. 6) could depend on the increase in the number of smaller particles. Based on these results, we determined whether soybean could act as an emulsifier that suppresses lipid oxidation.
Lipid oxidation in emulsions when the samples were used as emulsifiers TBA tests were performed to determine the extent of lipid peroxidation. Figure 7 shows the results of the emulsions tested with Tween 20 and native, optimally glycated, and unreacted soybean 11S globulin. When oxidation began with the addition of FeSO4, the TBARS of the Tween 20 emulsion increased up to day 3 and reached 9.7 ± 0.3 µM (n = 3). Thereafter, the value decreased, and was 5.0 ± 0.3 µM (n = 3) after day 7. However, the TBARS of native, optimally glycated, and unreacted soybean 11S globulin emulsions gradually increased over time, reaching 10.6 ± 0.3, 10.0 ± 0.2, and 10.8 ± 0.3 µM (n = 3) after day 7, respectively. A significant difference was found between the TBARS values of native and optimally glycated soybean 11S globulin emulsions on day 3 at a 1% risk rate. This suggests that optimal glycated soybean 11S globulin suppressed oil oxidation on day 3 when used as an emulsifier. Thereafter, the TBARS values of optimally glycated soybean 11S globulin emulsions were similar to those of native soybean 11S globulin emulsions after 5 and 7 days, suggesting less degradation of the TBA products. The antioxidant capacity of optimally glycated soybean 11S globulin did not suppress oil oxidation in the long term when used as an emulsifier.
Inhibitory effects on the lipid oxidation of several types of soybean 11S globulin used as emulsifiers.
Emulsions were prepared using Tween 20 (■), native soybean 11S globulin (□), optimally glycated (○), and unreacted soybean (●) as emulsifiers. These emulsions were stored at 50 °C for 7 d, and then lipid oxidation was conducted. The extent of oxidation was evaluated using TBA tests. The TBA test value is expressed as the equivalent of malondialdehyde (TBARS). ** represents significant differences between values (p < 0.01).
The difference between the change in the curve for Tween 20 and the other three samples might be related to the difference between the oil droplet surface covered by a small-molecule surfactant versus a protein or glucose; this may be due to the effect of the adsorption layer on the diffusion of radicals and oxygen. Accordingly, whether oil oxidation could be suppressed when optimally glycated soybean 11S globulin was used as an antioxidant was assessed in a subsequent study.
Lipid oxidation in emulsions when samples were used as antioxidants TBA tests were conducted to determine whether the oil would be less oxidized when soybean 11S globulin was used as the antioxidant (Fig. 8). Tween 20 was used for the emulsification. When the addition of FeSO4 began the oxidation reaction, the TBARS in the presence of 15 mM sodium phosphate buffer (pH 7.0), and the native and unreacted soybean 11S globulin gradually increased up to day 3, reaching 9.7 ± 0.3, 9.2 ± 0.5, and 8.6 ± 0.7 µM (n = 3), respectively. Thereafter, all values began to decrease, with 5.0 ± 0.3, 5.2 ± 0.6, and 5.2 ± 0.1 µM (n = 3) obtained after day 7, respectively. There were no significant differences between the TBARS values of the three samples. Although the trend in the changes of TBARS in the optimally glycated soybean 11S globulin emulsion was similar to those in the other three samples, the values following days 3 and 5 were 4.9 ± 1.1 and 4.0 ± 1.8 µM (n = 3), respectively. These values were significantly lower than those of the other three samples at a 1% risk rate, indicating that the optimally glycated soybean 11S globulin suppressed oil oxidation. Although optimally glycated soybean 11S globulin has emulsifying properties, when used as an emulsifier at a protein concentration of some 0.57 mg/mL at the experimental final condition, it did not suppress oil oxidation in the long term. However, when a sufficient amount of optimally glycated soybean 11S globulin was present in the emulsion at a concentration of about 2.9 mg/mL at the experimental final condition, oil oxidation was significantly suppressed. These results suggest that optimally glycated soybean 11S globulin has beneficial antioxidant properties.
Inhibitory effects of several types of soybean 11S globulin on lipid oxidation.
Emulsions were prepared using Tween 20, diluted 16.7 times with 15 mM sodium phosphate buffer (pH 7.0) (■), 0.3% native (□), 0.3% optimally glycated (○), and 0.3% unreacted soybean 11S globulin (●). All emulsions were stored at 50 °C for 7d. The extent of oxidation was evaluated using the TBA tests. The TBA test value is expressed as the equivalent of malondialdehyde (TBARS). ** represents significant differences between values (p < 0.01).
Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research (C) (No. 18K02195) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
Conflict of interest There are no conflicts of interest to declare.
gallic acid
HORAChydroxyl radical antioxidant capacity
IC50half-maximal inhibitory concentration
2-ME2-mercaptoethanol
·OHhydroxyl radical
PAGEpolyacrylamide gel electrophoresis
RCOrandom-centroid optimization
RHrelative humidity
SDSsodium dodecyl sulfate
TBAthiobarbituric acid
TBARSTBA-reactive substances
