Duckweed powder (13 g, dry weight basis) was immersed in 430 g distilled water for 12 h to reach a moisture content of approximately 80 ± 1% (w.b.). This soaking condition provided a rehydrated duckweed suspension with a weight ratio of rehydrated duckweed powder to extract aqueous media of 1/6. For duckweed protein extract, or DPE, produced with ultrasound-assisted extraction (i.e., UAAE and UAWE), the rehydrated duckweed suspension was subjected to cell disruption using an ultrasonicator (Sonics & Materials, VCX-750 Vibra-cell, Oxon, United Kingdom). A 25-mm diameter titanium alloy tip was used for cell disruption. The sample was treated with ultrasonication at an operating condition of 20 kHz and a 70% amplitude, which provided a power of approximately 562.5 W for 18 min. This operating condition resulted in an energy intensity of 107 W/cm². The ultrasonication was performed in cycles, with each duty cycle consisting of 30 s of ultrasonication followed by a 10-s pause. To prevent protein from thermal denaturation, the temperature of the duckweed suspension was controlled at approximately 10 ± 1°C by placing the glass beaker containing the sample suspension in an ice slurry bath. After the ultrasonication, the sample had a pH of 6.6 ± 0.1.

In this study, duckweed protein extracts (DPEs) were produced using three different extraction techniques: alkaline extraction (AE), ultrasonication-assisted alkaline extraction (UAAE), and ultrasonication-assisted water extraction (UAWE). Table 1 shows DPEs preparation methods and abbreviation of the methods used in this study. For UAAE and UAWE samples, the rehydrated duckweed suspensions were subjected to ultrasonication to break the cells, followed by alkaline (for UAAE) or water (for UAWE) extraction, respectively. Protein extraction of samples AE and UAAE was performed using an alkali extraction method. The pH of the rehydrated duckweed suspension was adjusted to 8.5 with 1.0 and 0.5 M NaOH, and stirred continuously for 2 h at room temperature (25 ± 1°C). The mixture was then centrifuged at 16,500 × g for 15 min at 4°C, and the supernatant was filtered through a two-layer cheesecloth. The pH of the supernatant was then lowered to 4.5 using 1 and/or 0.5 N HCl. The acidified sample was stirred for 30 min at room temperature before being centrifuged at 16,500 × g for 15 min to obtain a DPE pellet. For the UAWE sample, the same procedure was followed, except that the alkalization step was omitted. The resulting protein pellet was pre-frozen at − 20°C and then freeze-dried at − 35°C under a vacuum pressure of 20 Pa for 30 h (Christ, Alpha 1–4 LSCPlus, Germany) to obtain a freeze-dried sample with a final moisture content of 5.0 ± 0.5% (w.b.).

Production yield, protein yield, and protein recovery were calculated to discuss protein extractability. The production yield was the mass of the resulting DPE compared to the total mass of duckweed used for extraction. Protein yield was the protein mass in the resulting DPE compared to the total mass of duckweed used for extraction. Protein recovery was the protein mass in the resulting DPE compared to the total protein mass contained in duckweed used for extraction. All values were obtained using a dry weight basis and with triplicate measurements.

The protein content was evaluated based on the nitrogen combustion technique and the Dumas nitrogen analysis method of the sample was evaluated based on the studies of Grossmann et al. (2019) and Nieuwland et al. (2021). Calibration curves were prepared using EDTA with a known nitrogen content of 9.58%. Crude protein content was calculated using a nitrogen-to-protein conversion factor of 6.25 (Appenroth et al., 2017; EFSA, 2021; Duangjarus et al., 2022). Other compositions, including lipids, total dietary fibers, ash, and carbohydrates, were performed following the methods of Grossmann et al. (2019) and Nieuwland et al. (2021). The remaining percentage was indicated as the carbohydrate content.

To determine the amino acid profile of duckweed protein extracts (DPE), a microwave extraction method was used with slight modifications to the procedure described by Moldoveanu (2005). The DPE was weighed for 19 ± 1 mg, dissolved in 10 mL of 6 N HCl containing 0.1% (v/v) phenol, and then flushed with nitrogen. The mixture was subjected to microwave digestion at 130°C for 30 min (MultiWave3000, Anton Paar GmbH, Graz, Austria) and then cooled down to ambient temperature. The digested sample was filtered through a 0.45-μm PTFE membrane. Finally, the amino acid in the DPE sample was analyzed via ion exchange chromatography using an amino acid analyzer (Biochrome 30 + , Biochrom, Cambridge, United Kingdom).

Dried duckweed and duckweed protein extracts (DPEs) were dissolved in 5%SDS and heated at 95°C for 5 min. The protein solution was mixed with a sample buffer containing 0.125 M Tris–HCl (pH 6.8), 4% SDS, 20% glycerol, and 10% BME- at a ratio of 1:1. SDS-PAGE was performed according to the method of Tang and Yongsawatdigul (2020). Sample solutions (10 μg/μL) were loaded onto 10, 12.5, and 15% polyacrylamide gel and run at 60 V for 5 min, followed by 120 V for 90 min. The gel was then stained using 0.125% Coomassie Brilliant Blue R-250, and destained using a mixed solution of 25% methanol and 10% acetic acid. TriColor Broad Protein Ladder (3.5–245 kDa) was used to evaluate protein mass.

The examination method was performed similar to the method of Cui et al. (2020) with a slight modification. Protein solubility of the samples in DI water having pH in the range of 3.0–9.0 was examined. The samples were then centrifuged at 16,500 × g for 15 min, and the protein content in the supernatant was measured using the Carbon/Nitrogen Analyzer method (LECO, CN828, United States). The solubility profile of DPE was calculated as the percentage of protein in the supernatant compared to the total amount of initial protein.

WHC and OHC were determined using the method of Stone et al. (2015) with some modifications. Duckweed protein extract (0.1 g) was mixed with 10 mL of distilled water or soybean oil in a pre-weighed centrifuge tube and vortexed at high speed for 30 s. The sample was then incubated at room temperature for 30 min. Next, the hydrated sample was centrifuged at 15,000 × g for 20 min. The supernatant was gently decanted, the precipitate was collected and reweighed. WHC and OHC were calculated using Eqs. (1, 2), respectively.

The emulsifying capacity and stability of the samples were determined according to the method described by Duangjarus et al. (2022) with a slight modification. DPE dissolved in distilled water at a concentration of 1% (w/v) was prepared and the pH was adjusted to 7.0 and 8.5. The fraction that was completely soluble in water was obtained by centrifugation, and the supernatant or DPE solution was collected for the test. Ten milliliters of the DPE solution were then homogenized with 10 mL of sunflower oil using a high-shear homogenizer (Ultra-Turrax IKA, T25, Staufen, Germany) for 1 min at room temperature. An additional 10 mL of sunflower oil was added, and the mixture was homogenized again at the same homogenization speed for 2 min. The homogenized sample was left to stand at ambient temperature for 24 h. The height of the emulsion layer (H_EL) and the total height of the sample (H_TS) were measured at t₀ and t₂₄. Emulsion capacity (EC) was calculated at t₀ by dividing H_EL0 by H_TS0. Emulsion stability (ES) was calculated at t₂₄ by determining EC₂₄ and EC₀, and then dividing EC₂₄ by EC₀. Both ES and EC were expressed as a percentage.

All analyses of the DPEs properties were conducted at least in duplicate. The experimental data is presented as X ¯ ± S D . To assess the differences among the data, an analysis of variance (ANOVA) was conducted, and Tukey’s range test was performed at a confidence level of 95% ( p < 0.05 ) using the SPSS package software.

The extraction of duckweed protein was studied using different duckweed-to-aqueous solution ratios (DSR) in the range of 1/6 to 1/25, as reported by Nieuwland et al. (2021), Duangjarus et al. (2022), and Inguanez et al. (2023). However, comparing the results of these studies can be challenging due to variations in the types of duckweed and moisture content used.

In our study, duckweed powder with an 80 ± 1% moisture content after 12-h rehydration was used as a starting material. Therefore, the DSR used in the studies of Nieuwland et al. (2021) and Duangjarus et al. (2022) were estimated at 80% moisture content and were found to be equivalent to 1/5.5 and 1/20, respectively. A DSR range of 1/6 to 1/20 was thus set to determine its effect on duckweed protein extractability.

As shown in Figure 3, the protein content, protein yield, and protein recovery of DPEs increased with an increase in the aqueous solution ratio up to 15. However, a further increase did not significantly improve extractability. The production yield of the DPEs was not considerably affected by the DSR within the studied range (p ≥ 0.05). The studies of Potin et al. (2019), Bello et al. (2023), and Shen et al. (2023) reported similar results, showing that extraction efficiency and yield increased as the solid-to-liquid ratio used in the UAE process increased. This could be explained because the higher solvent ratio helps dissolve the extracted compound, making the mixture less dense and promoting the cavitation effect, which improves the extraction efficiency.

As above reported, when a DSR ratio was used after 1/15, the resulting extraction yields were ineffective. This observation is consistent with the findings of Sun et al. (2018) and Liao et al. (2022), who also noted that using too much solvent in the UAE process can compromise extraction yields beyond a certain limit. This might be due to the excessive solvent volume that compromised the acoustic energy density, leading to a reduction in the cavitation effect and extraction efficiency. Additionally, using higher solvent ratios in an industrial-scale operation is impractical due to lower production capacity, higher effluents, and increased operating costs. The extraction ratio has been shown to affect the chemical composition of the extracted proteins, which can in turn affect the techno-function of the DPE. Based on these findings, DSRs of 1/6 and 1/15 were selected to prepare DPE for further analysis.

As presented in Table 2, there was no significant difference in the protein and lipid contents of the DPE extracted at DSR of 1/6 and 1/15. However, increasing the aqueous solution ratio from 6 to 15 resulted in a considerable reduction in total dietary fiber from 8.93 ± 0.16 to 1.39 ± 0.01%. It was noted that hemicellulose and some dietary fibers were dissociated in an alkali solution (Maphosa and Jideani, 2016). Therefore, increasing the alkali solution ratio reduced the dietary fiber content of the DPEs. Additionally, the ash content of the DPE produced at the DSR of 1/15 was doubled compared to the DPE produced at the DSR of 1/6, which could be attributed to the higher amount of NaOH used for extraction. The results demonstrated the impact of the DSR on regulating the chemical composition and playing a role in the particle size distribution of the protein extracts.

Table 3 shows techno-functional properties of DPEs produced at two different DSRs (1/6 and 1/15). The properties were evaluated in terms of water holding capacity (WHC), oil holding capacity (OHC), and emulsifying properties (emulsifying capacity or EC and emulsifying stability or ES). The results indicate that the DSR values in the range of 1/6 to 1/15 did not affect DPE’s functional properties. As a result, the DSR of 1/6 was selected for DPE preparation in subsequent studies.

The current study employed a combining method of alkaline extraction (AE) and UAE, so-called UAAE, to produce the duckweed protein extract (DPE) and compared it with AE and UAWE. Table 4 shows the production yield, protein yields, and protein recovery of DPE produced using AE, UAAE, and UAWE methods. The result shows that UAAE provided the highest production yield of 16.2%, followed by UAWE (11.9%) and AE (7.6%). Similarly, the protein yield and protein recovery were ranked from the highest to the lowest as UAAE, UAWE, and AE. For protein recovery, it indicates the efficiency of the processing to extract protein from raw material and recover it into the protein extracts. Based on the results, UAAE had the highest protein recovery of 34.2%, followed by UAWE with 24.6%, and AE with 17.1%. Comparing AE to UAAE extraction, both production yield and protein recovery of the DPE were improved by 2.1 times.

Besides, to understand how US and AE affect the extractability of duckweed proteins, we compared the protein yield obtained from each extraction method with its production yield. The extraction method exhibited a protein yield-to-production yield ratio close to 1.0, indicating that the technique provided DPE with high protein purity. It was found that the ratio number ranked from highest to lowest corresponded to AE (0.73) to UAAE (0.64) and UAWE (0.62) for the protein yield-to-production-yield ratio. The finding indicated that ultrasonication was crucial in promoting both protein and non-protein extraction, while alkaline treatment was necessary in facilitating protein extraction. Increasing the pH of the suspension undergoing ultrasonication extraction to an alkaline pH might provide a positive synergistic effect on protein extraction. The protein recovery and production yield of Wolffia protein achieved in this study were generally higher than in previous studies (Kingwascharapong et al., 2021; Duangjarus et al., 2022).

Chemical composition of DPEs obtained through different extractions, namely AE, UAAE, and UAWE, was investigated. The results presented in Table 5 showed that UAAE and UAWE exhibited lipid content approximately 2.3-fold higher than that obtained from AE. Despite having a lower protein content than DPE obtained via AE, the DPE produced using UAAE and UAWE exhibited significantly higher protein recovery due to the high protein extractability promoted by ultrasonication. These results suggest that ultrasonication was important in releasing proteins and other co-extracted components from the duckweed cells into the extraction aqueous media. This phenomenon was attributed to the high energy intensity of ultrasound waves that generate cavitation force, causing the breaking of cells and releasing various components from the duckweed cells. The high energy and shear force also facilitated the rupture or dissociation of long-chain molecules to shorter-chain molecules, which could subsequently form a complex with protein and/or co-extracted molecules (Kingwascharapong et al., 2021; Sert et al., 2022).

Furthermore, the results revealed that AE and UAAE contained higher amounts of dietary fiber than UAWE. This was because alkaline could extract dietary fibers from plant cells within a temperature range of 4–85°C (Devi et al., 2023; Liu et al., 2023). This mechanism involves the dissociation of hydrogen and covalent bonds, as well as the ester linkage of cellulose and hemicellulose components by the hydroxyl ion (Maphosa and Jideani, 2016; Tejada-Ortigoza et al., 2016). Therefore, AE and UAAE, which were extracted in an alkali solution, contained higher amounts of dietary fibers than UAWE.

Figure 4A shows the amino acid composition of three different DPEs: AE, UAAE, and UAWE. The five main amino acids (AAs) found in all three DPEs were Glu, Asp., Leu, Arg, and Ala. The amino acid content of the DPEs produced with UAE applied (i.e., UAAE and UAWE) was considerably increased. However, it was observed that sulfur amino acids (SAA) were not detected in UAWE, which was produced without alkaline extraction. In contrast, SAAs were detected in both DPEs produced using AE and UAAE, with a slightly higher amount compared to the raw material. These findings suggest a significant role of alkaline treatment in SAA recovery.

As recommended by FAO (2013), the values of EAAs recommended intake for preschool-aged children for His, Ile, Leu, Lys, Thr, Trp, Val, SAAs, and AAA (aromatic amino acids) are 15, 30, 59, 45, 23, 6, 39, 22, and 38 mg/g_protein, respectively. Notably, the application of UAAE led to an overall enhanced EAA value that meets the requirement. All EAAs of the DPE were even higher than the values recommended for adults.

Furthermore, ultrasonication increased the moles of polar and nonpolar AA by approximately 2.0–2.2 folds, as shown in Figure 4B. AE also enhanced the moles of polar and non-polar amino acids by about 1.8 and 2.2 folds, respectively. The extent and balance between polar and non-polar amino acids played a crucial role in determining the protein solubility, water-, and oil-holding capacity, and emulsifying properties of the protein extracts. The impact was investigated and reported in 3.3.6–3.3.7. This study reported for the first time the role of UAE and AE in promoting the extraction of different profiles of amino acids from duckweed. The new findings provide valuable data for future studies seeking to optimize the extraction of protein from food sources and develop protein extracts with improved nutritional value.

The protein profiles of duckweed powder (raw material) and DPEs produced using AE, UAAE, and UAWE analyzed by SDS-PAGE are shown in Figure 5. The results showed that duckweed powder exhibited more prominent bands at 25, 43, and 50 kDa, while UAWE showed more significant bands at 10, 25, and 50 kDa. On the other hand, UAAE exhibited stronger bands at 25, 50, and 240 kDa with a weaker band at 60 kDa. DPE obtained via AE showed stronger bands at 10, 50, and 240 kDa and a weaker band at 60 kDa. Notably, the band at 25 kDa was not detected in the AE sample.

Compared to UAAE and AE, UAWE exhibited a much more intense band at 50 kDa. However, proteins with Mw of 60 kDa and higher than 240 kDa were not detected. This observation suggested that the alkaline extraction induced changes in protein structure and composition, leading to isomerization and/or cross-linking of the proteins/peptide chains in the subsequent step of acidic precipitation. This was supported by several larger molecules found on the top of the gel in AE and UAAE.

Moreover, it was noticed that the band intensity of small protein molecules (10 kDa) was weaker in the UAAE sample as compared to the AE. This could be attributed to protein degradation caused by the high-intensity ultrasound. The 25-kDa band was detected in all samples produced with UAE applied, indicating that UAE induced protein–protein molecular interaction. These findings are consistent with previous studies (Kingwascharapong et al., 2021; Sert et al., 2022), which reported that ultrasonication induced molecular interactions of proteins-proteins and proteins-other components.

The functional properties of proteins in food are significantly influenced by their particle size. The size of protein particles in the bulk phase varies depending on the pH of the solution. Table 6 presents the volume median, Dv(50), or average particle size, of DPEs produced using AE, UAAE, and UAWE. The results showed that the Dv(50) of all three DPEs significantly increased when the pH decreased from 8.5 to 7.0 and 4.0 (p < 0.05). Additionally, it was observed that the use of UAE to produce DPE resulted in a decrease in the DPE’s particle size, regardless of whether alkaline extraction was used or not. The smaller particle size was attributed to the high shear force and cavitation generated by ultrasonication, which broke down protein molecules and other components. These findings were consistent with previous studies on the use of UAE to produce other plant proteins, as reported by Kingwascharapong et al. (2021) and Sengar et al. (2022). The study suggests that using UAE can effectively produce DPE with smaller particle sizes, which may in turn improve functional properties for various food applications. However, further study is needed to determine the optimal UAE conditions and particle size effects on DPE’s functional properties.

Surface net charge plays a crucial role in determining protein functional properties. The zeta potential (ζ) values of all DPEs were measured at pH 8.5, 7.0, and 4.0 and are presented in Table 6 The zeta potential of three different DPEs in aqueous solutions at three different pH levels (8.5, 7.0, and 4.0) were all negative and significantly different from each other (p < 0.05). As the pH decreased from 8.5 to 4.0, the negative values of the DPEs decreased This could be attributed to the protonation of the carboxyl and amino groups of the protein, caused by shifting pH from alkaline to acidic values (Ladjal-Ettoumi et al., 2015).

The DPE solution with higher absolute zeta potential values suggested the greater electrostatic repulsion of the molecules, which tended to have less aggregation, thereby exhibiting smaller particles as previously reported. The higher absolute zeta potential values observed in the DPE solution suggested greater electrostatic repulsion among the molecules. This, in turn, led to reduced aggregation and resulted in smaller particles, as previously reported. The result was consistent with many studies that applied ultrasonication to other types of protein extraction, which found that UAE played an important role in particle size reduction and surface charge property enhancement of the protein extracts (Li et al., 2020; Sengar et al., 2022; Liu et al., 2023; Rao et al., 2023).

Figure 6 shows the solubility of DPEs produced using AE, UAAE, and UAWE. The solubility of proteins was observed to be at its minimum when the pH was around 4, and it increased as the pH either decreased or increased. The pH is known as the isoelectric point (pI), where a protein’s net charge becomes zero.

The result of this study was consistent with the existing report on the relationship between pH and protein solubility. According to Anoop et al. (2023), leaf proteins typically have a pI between 4.0 and 6.0. The pI of all DPEs examined in this study was found to be approximately 4.0. When the pH deviated from the pI value, the solubility of these proteins significantly increased. This can be explained by the increase in the net charge on the protein surface, which results in stronger repulsive electrostatic forces between protein molecules and their interaction with water.

The solubility of UAAE was significantly lower than AE. This could be because proteins were aggregated and/or denatured during the high-intensity ultrasonication. Additionally, the solubility of DPE produced via UAAE was lower than that prepared by UAWE in both acidic and alkaline conditions. Furthermore, it was noteworthy that the poorer solubility could be attributed to the high amount of dietary fibers co-extracted with the proteins during the alkaline extraction, as indicated by the higher dietary fiber content in UAAE sample (Table 5). According to Turck et al. (2021), more than 90% of dietary fibers found in duckweed are water-insoluble, which could account for the lower solubility of the DPE produced via the UAAE approach. The solubility of protein is a critical factor in maintaining the stability of emulsions, as higher solubility allows proteins to effectively dissociate and migrate to the water/oil interface and provide stabilization for oil droplets (McClements et al., 2022). However, extracts with a high proportion of dietary fibers can help stabilize the emulsion system by increasing aqueous phase viscosity. A similar finding was observed in the study of He et al. (2020).

The emulsifying capacity and stability of protein indicate the protein’s ability to form and stabilize emulsion, respectively. The former capacity involves the ability of the protein molecules to rapidly adsorb on the oil droplet surfaces upon homogenization. At the same time, the latter property is associated with the protein’s ability to prevent or retard the oil droplet aggregation after forming the emulsion (McClements et al., 2022). Emulsion-based foods need both properties but with different balances between the two properties depending on the type of food. The effects of the ultrasound and alkaline treatments on emulsifying properties were investigated. The emulsifying properties, including emulsifying capacity (EC) and emulsifying stability (ES) of the resulting DPEs, are presented in Table 7. Additionally, the appearance of all emulsion samples after storing for 1 and 24 h is presented in Supplementary Figure S1.

All three DPEs demonstrated 100% emulsifying capacity (EC) at pH 7.0 and 8.5, with emulsion stability (ES) of 90–92% after 24 h. While the protein content of UAAE and UAWE was lower than AE, their emulsifying properties were not significantly different ( p ≥ 0.05 ). Moreover, the protein solubility of UAAE was lower than the other two DPEs. Despite these differences, the emulsifying properties of all three DPEs showed no significant difference ( p ≥ 0.05 ). According to the study of McClements et al. (2022), smaller protein particles tend to migrate faster towards the oil–water interface, which promotes them to effectively stabilize the oil droplets. This is likely the reason for the comparatively high EC of the UAAE, as the protein particles in the sample were smaller, and therefore more effective at stabilizing the emulsion though the UAAE sample had lower protein content than the AE sample.

The emulsifying capacity (EC) of all three DPEs was 100% at pH 7.0 and 8.5, with emulsion stability (ES) ranging between 90 and 92% after 24 h. Although the protein content of DPE produced by UAAE and UAWE was lower than that obtained via AE, their emulsifying properties were not significantly different ( p ≥ 0.05 ). Furthermore, the protein solubility of DPEs produced using UAAE was lower than the other two DPEs. Despite these variations, all three DPEs showed no significant difference in their emulsifying properties ( p ≥ 0.05 ). As previously mentioned, DPE produced via UAAE had smaller particles at pH 8.5. Smaller protein particles typically exhibit better emulsifying capacity and stability than larger particles, as they migrate quickly to the oil–water interface and effectively stabilize the oil droplet. This could contribute to the good emulsifying property of the DPE produced using UAAE.

The ES at pH 7.0 of DPE produced via UAAE and UAWE were 91.53 and 90.88%, respectively, whereas the ES at pH 8.5 of the UAAE and UAWE were 92.87 and 91.3%, respectively. The ES of three commercial SPIs at pH 7.0 was 90.32–91.30%, and at pH 8.5 was 91.63–94.12%. The ES of UAAE and UAWE were comparable to the commercial SPIs (p ≥ 0.05). However, DPE produced via AE exhibited a lower ES than the commercial SPI at pH 7.0 and 8.5 (p < 0.05). It is worth noting that despite having lower protein content than SPI, all DPEs produced via UAE had comparable emulsifying properties to SPI. This finding implied that the emulsifying properties of the DPEs were not attributed to only the protein content. However, other components like carbohydrates and dietary fibers in the DPEs also co-contributed to improve emulsifying properties. Typically, dietary fibers and carbohydrates added into the emulsion system have been noted to play a role in emulsion capacity and stability improvement. These are generated via Pickering emulsion formation and enhance the continuous phase’s viscosity to retard the oil droplet aggregation, respectively. The DPE produced using ultrasonication proved to have excellent potential for emulsion application and exhibited emulsifying properties comparable to commercial SPI.

WHC and OHC are terms used to describe the ability of solid or semi-solid materials to retain water and oil in the sample under the influence of external forces, pressure, or heat applied. WHC plays a crucial role in determining the juiciness of foods, and it also impacts weight loss and cooking yield in food processes. As shown in Table 7, the WHC of three DPEs (AE, UAAE, and UAWE) was similar, ranging from approximately 3.5–3.8 g_Water/g_Sample. It is important to note that the testing method and conditions can significantly affect WHC values, so comparisons with other studies must be made with caution.

An article reviewed by Ma et al. (2022) presented experimental data on the WHC of various plant protein extracts. Their results were obtained through an experimental set comparable to ours. The report showed that depending on the type of protein extract, WHC ranged from 0.8–1.0, 1.3–4.9, and 1.1–7.9 g_Water/g_Sample for protein flour, protein concentrates, and protein isolates, respectively. Protein contents ranged from 23–27%, 63–77%, and 81–98% dry mass. All DPEs had WHC values comparable to most plant protein concentrates and many protein isolates. The DPE produced through UAAE and UAWE had even higher WHC than other protein concentrates.

Regarding OHC, Table 7 shows that DPEs produced through UAAE exhibited slightly higher OHC than the other two DPEs and three commercial soy protein isolates (SPI#1, #2, and #3). The OHC values (g_Oil/g_Sample) of DPEs were found to be significantly different (p < 0.05) and ranked from highest to lowest as UAAE (5.2) > UAWE (4.4) > AE (3.3). The three commercial SPIs had OHC values of about 4.2–4.3 g_Oil/g_Sample. OHC of UAWE was not significantly different from that of three commercial SPI. Typically, proteins’ hydrophilic and polar amino acid residues enhance their water-holding ability. On the other hand, proteins with high levels of hydrophobic amino acid residues tend to have good OHC (Tang et al., 2021; Ma et al., 2022). As previously reported, the DPE produced through UAAE and UAWE contained high levels of non-polar amino acids. Moreover, dietary fibers are known to improve WHC and OHC (Mehta et al., 2013; Zhang et al., 2020). As previously mentioned, AE resulted in the DPE with high dietary fiber. An analysis of the correlation among the content of dietary fibers, the amount of polar and non-polar amino acids, and water- and oil-holding capacity was conducted and is shown in Supplementary Figure S2. The results strongly suggest that the good OHC of the DPE produced via UAAE in this work was not solely due to its protein content. The amounts of non-polar amino acids and dietary fiber contents also played important roles in promoting OHC.

This finding is consistent with the reports of Miravalles et al. (2019)and Ma et al. (2022), which stated that the WHC and OHC of protein extracts are not always determined by their protein content alone. They are also influenced by the proportion and extent of hydrophilic and hydrophobic amino acid residues on the protein surface and other co-extracted components. The DPE produced using UAAE had a high ratio of non-polar amino acids, total amino acids, and dietary fiber content. As a result, it exhibited the highest OHC compared to the other two DPEs.

The limited techno-functional properties of most plant proteins, such as poor solubility and emulsifying properties, have been reported as factors that can restrict their use in food products (Colgrave et al., 2021). However, this study demonstrated that the DPE produced via UAAE shows promise as a possible alternative for food product applications such as bakeries, emulsion-based foods, and beverages.