In this study, the walnuts were purchased from the Winsuk County Woody Grain and Oil Forestry (Xinjiang, China). The soybean meal was acquired from Shandong Yuwang Group (Shandong, China). The 2 × Laemmli sample buffer, colored pre-stained proteins molecular weight standards, and pre-prepared gels were purchased from Shanghai Biyuntian Biotechnology Co., Ltd. (Shanhai, China). Methanol, ethanol and acetonitrile were purchased from MREDA Technology Co., Ltd. (HPLC grade, Beijing, China). Hydrochloric acid (HCl) and sodium hydroxide (NaOH) were purchased from Sinopharm Chemical Reagent Co., Ltd (analytical grade, Shanghai, China).

In this study, analysis of protein subunits by SDS-PAGE. The SDS-PAGE analysis was performed according to a previous report with slight modifications (27). In this study, the samples were prepared in a loading buffer containing 0.1 M Dithiothreitol (DTT) and incubated in a water bath at 95°C for 10 min. The electrophoretic gel was run at 120 V for 80 min after 10 uL of sample was loaded to electrophoretic gel. Then, the electrophoretic separation gel was stained with a Coomassie brilliant blue staining solution for 120 min, and the excess dye solutions were removed with the Coomassie brilliant blue destaining solution. Finally, the protein bands were photographed by the chemical gel imaging system Bio-Rad (Bio-Rad, Hercules, USA).

In this study, the solubility of proteins was determined using the NSI. After preparing the samples (as described in the section “2.2. Sample preparation”), the nitrogen content of the precipitated fraction was determined using the Kjeldahl method (28). The protein conversion factor was 5.3 and 6.25 for WPI and SPI, respectively. According to the SDS-PAGE results, no significant bands were detected in the SPI precipitated. Therefore, the NSI only referred to the WPI solubility. The NSI of WPI was calculated using the following Eq. 1:

Where P_WPI was the mass of initial WPI (g), P_residue was the mass of residual protein (g).

In this study, the particle size of composite nanoparticles were analyzed by DLS. DLS measurements were carried out by Nano-ZS90 (Malvern, Malvern, UK). The samples were diluted in ultrapure water to a final protein concentration of 0.1% (w/v) and were analyzed at 25°C. The refractive indices of the proteins and the dispersion medium were 1.450 and 1.330, respectively.

In this study, the analysis of protein-protein interactions were performed by fluorescence chromatography. The samples were adjusted to a final protein concentration of 0.1% (w/v) with ultrapure water, and the intrinsic fluorescence spectra of complex nanoparticles were measured at room temperature using a FS5 fluorescence spectrometer (Edinburgh Instruments, Livingston, UK). An excitation wavelength of 280 nm was applied in this study, and the emission spectra was recorded between 300 and 400 nm. The theoretical fluorescence (Theor.) spectra were obtained by adding the fluorescence spectra of individual WPI and SPI with different concentrations in the samples, on the other hand, the experimental fluorescence (Exp.) spectra corresponded to the fluorescence spectra of complexed nanoparticles (29). In this study, the significance of hydrogen bonding, hydrophobic interactions, and electrostatic interactions in the interaction between WPI and SPI was also evaluated. For that, WPI and SPI were dissolved in 36 mL of distilled water and processed according to the methodology described in the section “2.2. Sample preparation.” Then, 4 mL of 0.1 M thiourea, SDS, and NaCl solutions (with a final concentration of 10 mM) were added to the protein samples. Next, the fluorescence spectra were collected using the same methodology used to prepare the previous samples.

Structural studies on composite nanoparticles were performed using the fluorescent probe ANS. This probe can specifically bind to the hydrophobic domain of the protein, being capable of detecting changes in the protein microenvironment. Briefly, the samples were diluted to a final protein concentration of 0.05% (w/v), then 10 μL of 8 M ANS was added to 4 mL of sample, and the reaction was completed kept in the dark for 15 min. Next, the samples were excited at 390 nm wavelength, and the fluorescence intensities were measured between 400 and 600 nm.

In this study, the analysis of protein secondary structure was performed by CD spectra. To determine the CD spectra of composite nanoparticles, a MOS-450 spectrometer (BioLogic Science Instruments, Ltd., Claix, France) was used. In this study, the samples were collected at different pH values during the pH-cycle process and diluted with ultrapure water. The wavelength range of far-UV CD was obtained between 190 and 250 nm, and the protein concentration was 0.05% (w/v). The near-UV CD was composed of wavelengths between 250 and 320 nm, and the protein concentration used was 0.01% (w/v). Distilled water was used as a blank solution for the determination.

In this study, the stability of composite nanoparticles were analyzed by zeta-potential. Zeta-potential measurements were carried out by Nano-ZS90 (Malvern, Malvern, UK). For this study, the samples were adjusted in ultrapure water to a final protein concentration of 0.1% (w/v) and analyzed at 25°C. The refractive indices of the proteins and the dispersion medium were 1.450 and 1.330, respectively.

In this study, the stability of composite nanoparticles were analyzed by the H₀. The H₀ parameter was determined using the ANS fluorescent probe method (30). The samples were diluted with ultrapure water to obtain different protein concentrations ranging between 0.0125 and 0.1% (w/v). After 4 mL of the samples were mixed with 10 μL of 8 mmol/L ANS solution and placed in the dark for 15 min. Then, the fluorescence intensity of composite nanoparticles was determined by FS5 fluorescence spectrometer at excitation wavelength of 390 nm and an emission wavelength of 484 nm. In this study, the H₀ parameter was determined by the fluorescence intensity and protein concentration slope.

To determine the amino acid composition of the nanoparticles (g/100g protein), the proteins were hydrolyzed with hydrochloric acid. The co-assembled protein samples were placed into a sealed tube, and 10 ml of 6 mol/L hydrochloric acid solution was added. After neutralization with nitrogen, the tubes were sealed and hydrolyzed at 110 °C for 24 h. The supernatant was composed of a protein hydrolysate and determined by high-performance liquid chromatography. An Agilent 1200 liquid chromatograph equipped (Agilent Technologies Inc., Santa Clara, CA, USA) with a Zorbax Eclipse-AAA column (4.6 × 150 mm, 3.5 μm, Agilent Technologies Inc., Santa Clara, CA, USA) was employed to analyzing. The mobile phase was divided into phase A (pH 4.8, 0.04 mol/L NaH2PO4) and phase B (methanol: acetonitrile: ultrapure water (45:45:10 v/v/v)). The liquid phase conditions were as follows: flow rate, 1 mL/min; UV detection wavelength, 338 nm; injection volume, 1 mL; the column temperature was 40°C. The essential amino acid index (EAAI) was calculated according to formula 2 and corresponded to the geometric mean of the essential amino acid content and the corresponding amino acid content in the whole egg protein (31) (2):

Where the subscript w indicates the protein sample, s indicates the whole egg protein, and n indicates the number of amino acids.

At least three independent replicate experiments were performed for each determination of the sample to obtain the mean. Statistical results were expressed as mean ± standard deviation (SD) and analyzed using SPSS 20.0 software. Significant differences (p < 0.05) between groups were determined by one-way ANOVA variance.

In this study, WPI and SPI were co-solubilized at pH 12.0 and then neutralized and collected to prepare the soluble composite nanoparticles (Figure 1A). The results demonstrated that WPI aggregated and precipitated at the bottom in an aqueous solution before the SPI addition (Figure 1B). However, after adding SPI and using the pH-cycle technology, the resulting composite nanoparticles could dissolve in water (Figure 1B). Furthermore, composite nanoparticles with WPI: SPI = 1:0.5 and 1:1 still maintained colloidal stability in neutral water systems (Figure 1C).

As described above, the subunit information of the protein molecular weight was analyzed by the SDS-PAGE technology. The results obtained from the supernatants revealed that characteristic subunits of WPI and SPI were detected in composite nanoparticles. Furthermore, these subunit bands of SPI became more pronounced with the addition of SPI (Figure 2A, lanes 4–11). However, no characteristic subunit bands of SPI were observed in the SDS-PAGE performed with the precipitated fraction, indicating that only WPI was present after precipitation (Figure 2B, lanes 4–11). The composite nanoparticles in the supernatant retained the characteristic subunit bands of WPI and SPI after adding SPI and using the pH-cycle technology. These results suggest that most subunits of WPI and SPI were involved in the process of co-assembly to prepare composite nanoparticles and remained intact during the process. It has been described that conventional methods significantly degraded or removed most of the subunits of WPI, requiring a large quantity of energy and tedious labor and consequently leading to reduced applicability in the industry (34). This made the protocol significantly different from the conventional method. To investigate the changes in the solubility of walnut protein, the NSI was measured.

Studies described that protein solubility, a crucial functional property, could affect other several functional properties of proteins (35). The NSI of WPI in the composite nanoparticles was the percentage of WPI dissolved in solution relative to the initial WPI used in the reaction. Since the characteristic subunits of WPI were only present in the precipitate of composite nanoparticles (Figure 2B), the percentage of dissolved WPI can be obtained by calculating the difference between the total added amount and the content in precipitation. As shown in Figure 2C, the NSI of WPI with or without the pH-cycle treatment was approximately 20.09 and 12.64%, respectively. However, adding SPI induced an increase (30.71% at a WPI: SPI ratio of 1:0.01) in the NSI of WPI in the composite nanoparticles. Additionally, the NSI of WPI increased to 88.53% after adding SPI at a WPI: SPI ratio of 1:1, which was 7 times higher than in WPI nanoparticles without the pH-cycle treatment. Furthermore, the results showed that the WPI insolubility was significantly inhibited after the preparation of composite nanoparticles using the pH-cycle technology and adding SPI. On the other hand, most of the subunits in each protein were essentially unchanged in the composite nanoparticles (Figure 2A), demonstrating that the technology could improve the protein solubility and maintain molecular integrity. Since this study focused on WPI, SPI was added in an amount not exceeding WPI to avoid interferences with the structural analysis of composite nanoparticles. In this study, the morphological and structural properties of composite nanoparticles were evaluated to understand the formation mechanism of composite nanoparticles. The morphological of the composite nanoparticles were evaluated in the subsequent section.

The microscopic morphology of the protein and the composite nanoparticles can be visualized by TEM analysis. A significant aggregation was observed in TEM due to the high hydrophobicity of WPI (Figure 3A). Furthermore, an irregular aggregation was detected, leading to low protein solubility at pH 7.0, consistent with previous studies (36). On the other hand, SPI was well-dispersed in solution with uniform particle sizes at pH 7.0 (Figure 3B). Interestingly, the aggregation of proteins was reduced in the presence of the WPI/SPI = 1:0.1 ratio (Figure 3C). However, gradual, smaller, more dispersed, and homogeneous composite nanoparticles were obtained when the concentration of soy protein was enhanced (Figures 3D, E). To further investigate the structure of composite nanoparticles, AFM combined with the DLS technology was used. Therefore, the AFM results showed that an increase in SPI contributed to the dispersion of composite nanoparticles and the formation of uniform particles with a smaller size. These results agreed with the TEM results described above (Figure 4). The AFM results indicated that the composite nanoparticles collapse into nanoscale spherical structures on the mica sheet surface. The particle size of protein nanoparticles ranged between 80 and 250 nm, and a decrease was observed after adding SPI, which was in agreement with the TEM and AFM results (Figure 4D). Contrarily, the particle size was slightly larger than the size observed in TEM due to the crumpling of the sample that occurred during the drying pre-treatment in TEM. According to the microscopic and DLS analyses, it was possible to observe that when the two proteins interacted, a new protein structure was formed (30). To explore the mechanism behind the phenomenon of protein structure change, the structural properties of the composite nanoparticles were determined. The protein-protein interactions were elaborated.

Some studies revealed that substances with good water solubility must be self-repelling and inhibit aggregation. Therefore, the hydrophobicity can modulate their physical stability, solubility, aggregation tendency, and adsorption behavior (50). Due to the macromolecular nature of proteins, surface hydrophobicity can significantly alter the protein function compared to the total hydrophobicity (51). In the present study, surface hydrophobicity and zeta-potential were used to investigate the surface properties of the composite nanoparticles. The results in Figure 7A showed that the surface hydrophobicity of the composite nanoparticles decreased after adding SPI. However, an opposite trend was observed in the absolute zeta-potential values (Figure 7B). This result indicates that the interaction between the two proteins can lead to a gradual increase in the anti-folding of WPI, which continuously inhibited the continuously inhibits the formation of hydrophobic groups and the exposure of charged protein groups, creating repulsive force on the surface to improve the solubility and stability of the composite nanoparticles.

Additionally, this study investigated changes in the surface hydrophobicity and zeta-potential of the composite nanoparticles when the pH decreased. Figure 7C demonstrated that hydrophobic regions were continuously formed during the folding process. Furthermore, the surface hydrophobicity of composite nanoparticles was lower than in WPI, indicating that the interaction between WPI and SPI significantly inhibited the formation of hydrophobic regions and maintained the relative unfolding of the proteins conformation. The absolute values of zeta-potential continuously decreased with the decrease of pH, indicating that the charged groups exposed to pH 12 were encapsulated inside the complex structure (Figure 7D). The results showed that the decrease of surface hydrophobicity and the increase of net surface charge were crucial factors for good dispersion of the composite nanoparticles. Finally, the amino acid composition and functional properties were investigated.

WPI and SPI are considered excellent sources of amino acids for humans, being widely available in the diet (52), mainly WPI. The Food and Agriculture Organization of the United Nations and the World Health Organization (FAO/WHO) defined that nine essential amino acids are essential in dietary habits, including His, Ile, Leu, Lys Met + Cys, Phe + Tyr, Thr, Trp, and Val, to maintain normal body functions and health. For preschool children, the recommended levels of each amino acid are 19, 28, 66, 58, 25, 63, 34, 11, and 35 mg/g protein, respectively, and for adults, 16, 13, 19, 16, 17, 19, 9, 5, and 13 mg/g protein, respectively. The complex protein nanoparticles obtained in this study showed an impaired amino acid composition with EAAI values ranging from 71.73 to 74.61. On the other hand, SPI and WPI exhibited EAAI values of 70.72 and 78.25, respectively (Table 1). Therefore, it is possible to infer that the design of novel soluble protein complex structures composed of multiple proteins could be an effective strategy to obtain protein products with balanced nutritional composition and functionality.

Under external energy input, proteins can reduce the oil-water interfacial tension and induce the formation of emulsions in oil-water mixtures. Moreover, the emulsification performance was closely related to the hydrophilic-hydrophobic balance of the molecular structure (20). Therefore, solubility and hydrophobicity are crucial parameters to modulate the emulsification effect when proteins are used as interfacial stabilizers. The EAI and ESI of WPI were significantly lower than in SPI, demonstrating the limitation of the lack of solubility on its ability to adsorb at the oil-water interface (Supplementary Figures 3A, B). Additionally, the solubility and surface hydrophobicity of WPI was improved by the interaction with SPI, and the EAI and ESI of the composite nanoparticles were significantly higher than in WPI (Supplementary Figures 3A, B). Moreover, the co-assembled structure was relatively unfolded, and the structure was rearranged to expose the internal groups capable of adsorbing to the surface of oil droplets. Therefore, an improvement in the hydrophilic-lipophilic balance of the protein particles could occur, leading to an increase in the EAI in WPI (53). This result showed that solubility is highly relevant for protein emulsification. The increase in ESI occurred due to the interaction of WPI and SPI, resulting in the formation of a new three-dimensional network spatial structure. Therefore, the protein contains all the proprieties to adhere to the oil droplet’s surface. And the composite nanoparticles with a certain amount of surface charge are adsorbed to the oil-water interface to maintain the emulsion stability by the electrostatic repulsion between the interface-interface and then prevent the occurrence of oil droplet agglomeration (54).

Protein foaming properties can significantly alter the value for application in foamy foods, such as ice cream. Furthermore, some biological parameters, including protein solubility, molecular flexibility, hydrophobicity, and charge density, also significantly modulate the foaming ability (FA) (55). Additionally, changes in protein structure, size, and interactions could lead to alterations in the surface tension, viscoelasticity, and surface rheological properties and consequently affect the foaming stability (FS) (56). Taking this into account, the FA and FS parameters were evaluated in this study. The FA of the composite nanoparticles was significantly improved compared to WPI. However, the FS was boosted less (Supplementary Figures 2C, D). The results indicate that the interaction between WPI and SPI formed a new three-dimensional hydrophilic structure. The structure showed a relatively unfolded state, enhanced protein flexibility and solubility, and enhanced interaction between protein and water molecules, promoting protein adsorption at the liquid-air interface. All these changes could induce a significant improvement in the FA (57). Compared to WPI or SPI alone, the FS of the composite nanoparticles was enhanced because WPI and SPI interacted to form a new three-dimensional structure, exposing some of the hydrophobic groups. On the other hand, inter-protein forces were enhanced, resulting in rapid adsorption and formation of a stronger protein membrane at the air-water interface, enhancing the FS (55, 58). Overall, the composite nanoparticles showed improved functional properties compared to WPI or SPI alone.