Extraction and Nutritional Value of Soybean Meal Protein Isolate

Abstract

The increasing protein demand driven by global population growth necessitates the search for an alternative protein source. Soybean meal (SM), with approximately 47–49% proteins, is a viable option. Soymeal protein isolate (SMPI) is a cost-efficient protein source with a well-balanced amino acid profile, making it suitable for addressing this demand. This study attempts to address the lack of information regarding the extractability and nutritional potential of SMPI obtained utilizing a weak base and recoverable solvent. In this work, the structural and compositional studies of SMPI, as well as the aqueous extractability of ammonium hydroxide (NH₄OH), are investigated. Furthermore, we examined the effects of heat treatment during oil extraction, where a comparison between protein isolates from soymeal and whole soybeans was carried out. The maximum extraction yield of 24.1% was achieved by increasing the concentration of NH₄OH from 0.25–1%. In comparison to the soy protein isolate from whole soybeans (SPI), the compositional analysis report revealed that SMPI had higher levels of crude protein, moisture, and ash content, but lower levels of fat and carbohydrates. Furthermore, the examination of the amino acid composition confirms the existence of vital amino acids in SMPI. The amino acid score indicates that methionine, lysine, and threonine are the limiting amino acids. SMPI and SPI share structural and functional group similarities, as demonstrated by Fourier-transform infrared spectroscopy. Gel electrophoresis using sodium dodecyl sulfate–polyacrylamide shows that the protein molecular weight distributions of SPI and SMPI are similar. This in-depth evaluation emphasizes the advantages of SM by advocating its application in other sectors beyond conventional animal feed, such as nutritional supplements and bio-based products, and by improving the environmental sustainability and global food chains.

1. Introduction

Glycine max or soybeans are among the most valuable agricultural products in the world [1]. Even though soybeans can be processed into a broad variety of foods, such as tofu, soy milk, soy flour, and soy nut butter, a greater percentage of the soybeans that are produced in the United States are crushed into edible oil [2]. Soybean oil, extensively used for the production of sustainable diesel, makes up around 20% of the weight of a bushel of soybeans (equivalent to around 12 lbs./1.55 gal); the remaining residue is soybean meal (SM) [3]. To optimize oil extraction from soybeans and enhance the quality of both the oil product and the final soybean meal, a thermal treatment is applied [4]. This crucial process used during the mechanical pressing or solvent extraction of oil facilitates the breakdown of the oil body structures to release the oil, reducing the oil viscosity leading to a high oil yield and lower moisture content, particularly in solvent extraction [4,5]. Also, high-quality meals are produced when SM antinutritional factors (ANFs) are inactivated with a heat treatment [6]. The rapid expansion of soybean oil processing facilities in the United States is causing a significant rise in the availability of soybean meal (SM). This suggests that SM can be easily sourced for a variety of applications including animal feed, feedstock for organic fertilizers, dietary supplements, and bio-based products. Moreover, SM protein is regarded as high-quality due to its abundance of essential amino acids such as lysine, leucine, and tryptophan [7]. Extensive research has shown that SM has a high level of protein (47–49%), making it a candidate for an alternative protein source [8].

The market value of SM is lower than that of protein isolates because of ANFs such as trypsin inhibitors which limit the use of SM in animal feed [9]. Several studies have thoroughly demonstrated the adverse effects of ANFs on the development and functioning of the gastrointestinal tract, particularly in juvenile animals, resulting in decreased nutritional absorption and hindered growth [10]. Also, the richness in the essential amino acids of SM is measured by its protein content, which affects the price of the product [11]. An increase of 1% in the crude protein content of SM can result in a monetary yield of USD 10.27 per metric ton (MT) for swine and USD 12.62 per MT for poultry [11]. Therefore, developing highly effective and efficient extraction techniques has the potential to enhance its economic value by reducing ANFs, improving protein quality and promoting better nutritional benefits [12,13]. The study of the nutritional and structural characteristics of SMPI will expand its applicability in a variety of contexts by offering end users insightful information. In terms of simplicity and the cost-effectiveness, solvent extraction techniques show great potential when compared to other extraction methods like enzyme-aided, ultrasound-assisted extraction (sonication), high-pressure processing (HPP), and aqueous two-phase systems (ATPS) [14,15]. Conversely, research has recognized the use of a recoverable solvent (NH₄OH) as an environmentally friendly method for plant protein extraction [16]. Since SM has undergone processing steps such as oil extraction, desolventizing, and drying, its structural and compositional properties might have been altered compared to that of soybeans [17].

In an effort to fully utilize SM, the research has mostly concentrated on lowering ANFs while ignoring other essential characteristics and practical uses. Although the use of SM in animal nutrition and aquaculture is well-documented, its wide use is restricted due to the inadequate investigation of the effects of heat treatment on the functional and structural characteristics of its protein component [18,19]. The quality of the meal that is produced, especially the protein’s nutritional content, can be greatly impacted by the heat treatment applied to soybeans during the oil extraction process [20]. The protein structure of the meal may be compromised, and some amino acids may become less available if the oil extraction process is performed with the involvement of excessive heat [21]. Without considering the potential usefulness of SM’s protein isolate as dietary supplements for humans, food additives, and ecologically friendly bio-based goods, its application is restricted to the traditional limits of animal feed due to our lack of knowledge regarding the nutritional qualities and safe processing of this material [22]. Several benefits could be unlocked if future research focuses on improving SM protein characterization and extraction methods to fill these limitations. Also, optimizing the utilization of SM proteins will also minimize wastage and promote a sustainable agricultural methodology [23,24].

2. Materials and Methods

2.1. Materials

Defatted soymeal and whole soybean seeds were provided by the Pilot Plant Unit of the Agricultural and Biosystem Department, North Dakota State University. Both samples were ground with a commercial grinder before use and soymeal with a sieve mesh size of 0.425 mm was used. All chemicals and reagents used for the study were of analytical grade; Novozymes A/S (Bagsvaerd, Denmark) supplied the enzymes used and a mini-protean precast gel and protein standard was purchased from Bio-Rad (Bio-Rad Laboratories California, Hercules, CA, USA).

2.2. Extraction of Protein Isolates

SMPI was obtained according to the process described by Adeniyi et al., 2023 [25]. SM protein was solubilized by dissolving the meal in an aqueous alkaline solution in a 1:10 sample-to-solvent ratio with different concentrations of NH₄OH. The range of the solvent concentration variables measured are 0, 0.25, 0.5, 0.75, and 1%, respectively. Then, 2.5 g of soybean meal powder was dispersed in 25 mL of each solvent concentration. The dissolving solutions were placed in a water bath at 52.5 °C and 130 rpm for 12 h [16]. The soluble fraction was obtained by centrifugation at 1500 rpm for 10 min, followed by decantation. A further analysis was carried out on the supernatant using an isoelectric point to precipitate the protein. With 1N HCl, the pH of the supernatant was adjusted to between 4.5 and 5.0 [26], and the mixture was then allowed to precipitate overnight. The residue collected by centrifugation at 1000 rpm for 10 min was rinsed with distilled water and oven-dried for 12 h at 60 °C using a Binder ED-56 oven dryer (Horsham, PA, USA). All extraction experiments were completed in triplicate. SPI was extracted using the same procedure, but with a slight variation in the concentration of NH₄OH. Soybean powder was dispersed in 0.5% NH₄OH to solubilize the protein. This concentration was reported by Bello et al., 2023 [16], as the optimum concentration for the extraction of SPI. Figure 1 shows a graphical illustration of the alkaline/isoelectric point extraction of SMPI. The total dry matter yield was determined using Equation (1) as a percentage of the total dry matter after extraction, TDM_o, to the total dry matter in the initial suspension, TDM_s [27].

$$TDM\left(\mathrm{\%}\right)={\displaystyle \frac{{TDM}_o}{{TDM}_s}}\times 100$$

(1)

2.3. Proximate Analysis

2.3.1. Moisture Content

The moisture contents were determined according to the AOAC 925.10 standard techniques [28,29]. Briefly, 1 g of sample was weighed in a crucible and oven-dried at 105 °C for 24 h. The weight of the empty crucible was recorded before samples were added for drying. The percentage of the moisture content was determined by Equation (2).

$$\%Mc={\displaystyle \frac{Mw}{Sw}}\ast 100$$

(2)

Mw = Sw − (Dc − Ec)

Mw = moisture weight, Sw = sample weight, Ec = empty crucible, and Dc = dried sample and crucible weight

2.3.2. Ash Content

The determination of the total ash content of the samples was performed by incineration using Thermos Fisher Scientific Thermolyne Benchtop TM 1100 °C Muffle Furnace (Vernon Hills, IL 60061, USA). For this, 2 g of each sample was weighed in a crucible and placed in the muffle furnace at 575 °C for 24 h. The samples were allowed to cool in a desiccator, and the Total ash was calculated using Equation (3).

$$\%ash={\displaystyle \frac{M_2-M_1}{M\mathrm{s}}}\times 100$$

(3)

M₂ = Weight of ash. M₁ = Weight of crucible. Ms = Weight of sample.

2.3.3. Fat Content

To determine the total fat content, 2.00 g of the homogenized sample was subjected to continuous lipid extraction for 5 h in a Soxhlet extractor using petroleum ether as the solvent [28].

2.3.4. Total Protein

The protein was determined in triplicate by measuring the total nitrogen content with Thermos scientific Flash Smart™ Elemental Analyzer using the Dumas method. For this, 2–3 mg of the sample was combusted at a high temperature in an oxygen atmosphere. Through subsequent oxidation and reduction tubes, nitrogen was quantitatively converted to N₂, measured by a thermal conductivity detector. The nitrogen content was then converted into the protein content using a conversion factor of 6.25, commonly used for various sample types [30].

2.3.5. Total Carbohydrate

The arithmetic difference method was used to estimate the samples’ total carbohydrate content. Using the equation below, the proximate composition values of other parameters were subtracted from 100 to estimate the amount of available carbohydrates.

%Carbohydrate = 100 − (%moisture + %Fat + %protein + %ash)

2.4. Enzymatic Hydrolysis of Soybean Meal Protein Isolates

Soybean meal protein hydrolysate was prepared according to the method described by [31,32]. Three different protease enzymes, namely Alcalase, Neutrase, and Pepsin, were used for the experiment. About 5 g of each sample of protein isolate was dissolved in PBS in a 1:10 solid-to-solvent ratio. The pH of the suspension was then adjusted to the optimal pH for each enzyme using 1 M NaOH. The suspension was allowed to stabilize for 2 h in a water bath at 40 °C. Subsequently, a 5% enzyme concentration was added to the mixture, and the reaction was continued by placing it in a water bath for 24 h at 40 °C and 100 rpm. The reaction was halted by heating at 90 °C for 10 min. The resulting mixture was centrifuged at 4000 rpm for 10 min to obtain the hydrolysate.

Amine Quantification

The total amines present in the hydrolysate were quantified using a similar method described by [33]. The 2,4,6-Trinitrobenzene Sulfonic Acid (TNBS) assay method was employed for estimating the total amount of amine present in the sample. For this, 0.25 mL of TNBS (0.01% w/v) was added to 0.5 mL of each sample and the mixture was vortexed. Afterwards, the solution was incubated in a water bath for 2 h at 37 °C. To each sample, 0.125 mL of 1 N HCl and 0.25 mL (about 0.01 oz) of 10% SDS were added to stop the reaction. The resultant solution’s absorbance was measured at a wavelength of 335 nm. Prior to the runoff of the experiment, glycine standard (20 μg/mL) was prepared by dissolving 0.2 g of glycine in 10 mL of distilled water and then diluted as needed. In 0.1 M sodium bicarbonate reaction buffer (pH 8.5), the soy samples were dissolved. The standard curve generated was used in calculating the concentration of the extracted amine in each sample.

2.5. FTIR

Using a Fourier-transform infrared (FTIR) spectrometer coupled with an attenuated total reflectance (ATR) accessor (Thermos Scientific Nicolet 8700, Waltham, MA, USA), dried SMP and SPI were analyzed on a diamond plate at a resolution of 4 cm⁻¹. The spectra were collected at a frequency region of 4000–600 cm⁻¹ and 64 scans were recorded for each sample. The amide II region of the FTIR spectra, usually ranging from 1500 to 1600 cm⁻¹, was extracted using Origin Pro 2024b Learning Edition (Origin Lab Corporation, Northampton, MA, USA). The spectra were subjected to second derivative calculations to enhance the peak resolution. Deconvolution was performed using the ‘Peak Analyzer’ tool, utilizing a Gaussian fitting model to analyze and separate overlapping peaks. The relative amount of different secondary structures in the protein was estimated by calculating the areas under the deconvoluted peaks. All selected peaks were deconvoluted in triplicate.

2.6. SDS-PAGE Electrophoresis

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) was carried out on Mini-protean precast gel (Bio-Rad Laboratories, Hercules, CA, USA). Bio-Rad Precision Plus Protein Standards (10–250 kDa) were used as molecular markers according to the method of Laemmli 1970. All samples were prepared by dilution with an equal volume of Laemmli sample buffer. The gels were run in mini-protean electrophoresis cells at 150 V for 40 min or until the dye reached the bottom of the gel and then stained with Coomassie blue dye R-250. The bands were visualized after staining and washing using distilled water with three to four changes until the bands were clearly visible.

2.7. Amino Acid Profiling of SMPI

The amino acid profile was carried out at the University of Missouri’s AESCL. Cation-exchange chromatography (cIEC-HPLC) with post-column ninhydrin derivatization and quantification was used as the approved technique for amino acid analysis.

2.8. Differential Scanning Measurement

The thermal properties of SMPI and SPI were evaluated using a TA DSC 2500 thermal instrument (TA Instruments New Castle, DE 19720, USA) according to the method described by Das et al. with some modifications [34]. Approximately 6–8 mg of protein samples was weighed in an aluminum pan using a Mettler Toledo Analytical Plus precision balance (Mettler-Toledo, LLC Polaris Parkway, Columbus, OH, USA). The initial weight of the aluminum pan was measured before and after the sample was added. The pan was hermetically sealed and heated at a rate of 10 °C/min between 20 °C and 130 °C. An empty pan was hermetically sealed and used as a reference. The onset temperature, peak temperature, and transition enthalpy were obtained from the DSC curve using TRIOS V5.6.0.87 software.

2.9. Statistical Analysis

Experiments on the extractability of SMPI, Amine quantification, and proximate analysis were carried out in triplicates, and the results were reported as means of the three values. Data were analyzed using One-way ANOVA (see Supplementary Materials). Tukey’s comparison Test was used to determine significant differences between the means. A two-sample t-test was employed to assess significant differences between parameters. A statistical analysis was performed using Minitab 21.4.3.0 statistical software with the significance level set at p < 0.05.

3. Results and Discussion

3.1. Extractability of Soybean Meal Protein

The rigidity and complexity of the plant cell walls trap proteins within the matrix, making extraction difficult. Previous research has indicated that alkaline conditions are more effective than neutral or acidic conditions for extracting plant protein [35] because they solubilize proteins and assist in the breaking down of cell wall structures in plant materials, releasing trapped proteins [36]. Under alkaline conditions, the net negative charge on soymeal globulin proteins increases, reducing protein–protein interactions and enhancing solubilization [37]. The influence of severe pH, which arises under alkaline conditions, can cause the breakdown of hydrogen, amide, and disulfide bonds [37]. However, extraction at a higher pH can disrupt the structure of lysine and cysteine amino acids and negatively affect protein digestibility. These observations underscore the efficiency of mild alkaline conditions, particularly NH₄OH, in breaking down plant cell walls to improve plant protein extractability [24].

In this study, SMPI was extracted using NH₄OH at different concentrations (0–1%) and precipitated using isoelectric points. Unlike NH₄OH solutions, Figure 2 shows that no protein was extracted in water (0% NH₄OH). This is because a large percentage of total SM proteins (65–85%) are globulin proteins. The highest protein yield (24.1%) was obtained at a 1% NH₄OH concentration followed by a 22.9% protein yield at 0.25% NH₄OH. Increasing the concentration from 0.25% to 0.5% NH₄OH did not result in a higher yield. One-way ANOVA (p < 0.05) suggests that varying the concentrations of NH₄OH significantly influence the protein extract yield. The Tukey pairwise comparison test did not indicate any significant difference in the protein yield from 0.25–1% NH₄OH. Although higher alkaline conditions enhanced the solubility and yield of SMPI, a high pH can also result in excessive denaturation and loss of function properties of the protein [38]. Therefore, we selected a 0.25% NH₄OH concentration suitable for SMPI extraction.

The observed difference in the protein yield, as shown in Figure 2, can be attributed to a change in the alkaline conditions [39]. At a 0% NH₄OH concentration, SM proteins, which contain hydrophobic regions and complex tertiary structures, are not easily solubilized by water alone, as it cannot disrupt the stabilizing forces created by disulfide bonds and other interactions. Similar findings by Gerzhova et al. revealed that the extractability of proteins from canola meal increased, with an extraction yield of approximately 27% at pH 10 and 58% at pH 12 [27]. These findings highlight the effectiveness of the alkaline extraction method [40]. Extremely high alkaline levels can permanently alter the structure of proteins, leading them to precipitate or clump together and reducing the production of soluble, functional proteins [41].

3.2. Proximate Composition

Table 1. Proximate analysis of Soymeal Protein and Soybean Protein Isolates.

Parameters	SMPI (%)	SPI (%)
Moisture content a	14 ± 0.58	11 ± 1.00
Ash b	2.12 ± 2.47	1.66 ± 0.28
Fat a	0.02 ± 0.00	8.67± 0.01
Protein a	68 ± 4.50	56.8 ± 0.80
Carbohydrate a	15.86 ± 3.01	21.87 ± 0.34

Value shown as mean ± standard deviation. Parameters that do not share a letter show no statistical difference between the means.

shows the proximate composition of SMPI and SPI obtained through alkaline extraction and isoelectric precipitation. The results reveal that SMPI contains 14% moisture, 2.12% ash, 0.02% fat, 68% protein, and 16% carbohydrates, while SPI contains 11% moisture, 1.66% ash, 8.67% fat, 57% protein, and 22% carbohydrates.

The moisture content of SMPI is higher than that of SPI. This can be caused by a number of variables such as irregularities in the oil extraction procedure, a less rigorous drying regimen intended to maintain the functionality of the proteins, and the porous nature of soybean meal, which permits it to hold on to more moisture than soybeans [42]. On the other hand, hydrophobic interactions within SPI may have an impact on the reduced moisture content reported in SPI by lowering moisture retention [43]. A high contrast in the protein content is evident between SMPI (68 ± 4.5%) and SPI (56.8 ± 0.8%). The substantial variability observed in SMPI, as indicated by the standard error, may be due to the impact of the processing conditions. Oil extraction removes a significant amount of fat, producing more protein concentrate products [44,45]. According to Deleu et al. [41], the protein content of isolates of soymeal ranges from 61% to 91%, while L’ho-cine et al. used an aqueous extraction method to obtain an 84% protein content. Similar to this, by maximizing processing parameters utilizing alkali and isoelectric point procedures, Sethi et al. observed a greater value of more than 90% dried-weight protein isolates from khesari dhal, a rich protein source among pulses [46,47,48].

SMPI, with the highest protein content from our results, exhibits a notably lower protein content (68 ± 4.5%) compared to other protein isolates having over 90% protein. This emphasizes the presence of non-protein substances. Another factor that may affect extraction is the conditions. NH₄OH creates a milder extraction condition that effectively protects the protein’s native structure and functions because it is less basic than NaOH [49]. The optimization of the extraction and further purification of protein isolates could enhance the value of the protein contents.

Moreover, the ash content in SPI is lower than the ash content in SMPI. Although, the standard error of the ash content of SMPI (±2.47) is considerably greater, suggesting a greater level of variability in the data. The observed differences, which are not statistically significant (p > 0.05), may be attributed to the inclusion of non-meal substances, such as external particles, during the processing of SM. Since the ash content quantifies the mineral content in the sample [50], a lower ash content in SPI and SMPI indicates a greater level of protein purity. SPI had a higher fat content than SMPI. The increased fat content may be attributed to processing methods or SPI’s inherent composition. Technically, lipids can impact the overall moisture content by repelling water, potentially resulting in lower moisture levels in protein isolates [16].

The carbohydrate content in SMPI is 15.86%, while SPI, on the other hand, had a carbohydrate content of 21.87%. This can be attributed to presence of glycoproteins, such as lectin (soybean agglutinin) in soybeans resulting in the breakdown of glycans under heat processing and a mild alkaline treatment [51]. The breaking of glycosidic linkages that results from combining a mild alkaline treatment with heat treatments like drying during the production of SM may have a substantial effect on glycoproteins. This process results in the separation and degradation of carbohydrates. Consequently, SMPI has fewer carbohydrates in comparison to SPI derived from whole soybeans, which has a higher percentage of its glycoprotein intact. The difference in carbohydrate contents detected between the two isolates indicates a significant variation in their individual nutritional characteristics.

3.3. Evaluation of SMPI Digestibility Using Various Industrial Proteases

Protein digestibility is defined as the proportion of proteins that digestive enzymes can hydrolyze and absorb as amino acids or other nitrogen compounds [52]. There is a high demand for easy-to-hydrolyze plant-based proteins. Although soybeans are the largest source of plant protein, the constituent proteins have low digestibility, primarily due to antinutritional factors (ANFs), which are protease inhibitors. Nevertheless, many soybean processing methods, particularly heat treatment during oil extraction, can deactivate ANFs [53,54]. Although heat treatment is known to benefit soybean protein digestibility, the effect of thermal processing on the digestibility of soymeal protein depends on the optimal conditions and nature of the SM protein. While the inactivation of ANFs is beneficial and could increase protein digestibility, excessive heat can reduce the availability of some essential amino acids, potentially generating new types of ANFs such as Millard products [55]. Moreover, there are limited studies confirming that heat treatment alone will simultaneously increase the digestibility of SM protein.

To investigate the digestibility of SM protein, we performed the in intro hydrolysis of SMPI with three known proteases—Alcalase, Neutrase, and Pepsin—using SPI as the control to isolate the effect of several processes involved in soymeal production on the digestibility of SMPI. The result, expressed as the degree of hydrolysis (DH), shows that the digestibility of SPI is significantly higher than that of SMPI, with pepsin having the highest hydrolytic effects of the three enzymes (Figure 3). This result is in contrast with the findings that the protein digestibility of soybean protein increases with various processing methods, including those involved in the oil extraction process that generates SM [56,57]. The joint effect of most processing methods, especially the heat treatment of soybeans, is the reduction of ANFs [58]. All processing methods are reported to reduce protease inhibitors (ANFs) such as trypsin, hemagglutinins, and chymotrypsin inhibitors [59,60].

Eliminating protease inhibitors in soybean protein may not increase the digestibility of SM protein. It should be noted that some inhibitors, such as Bowman–Birk inhibitors (BBIs), are relatively heat-stable due to their multiple disulfide bridges that block the activity of proteases at independent binding sites [61]. Heat-stable inhibitors like BBIs would be present in SM produced from processing that involved less soybean heating (like solvent extraction). This is usually the case as a compromise to keep the protein quality intact while thermally destroying protease inhibitors. Although the chronic consumption of SM protein with residual levels of heat-stable protease inhibitors will unlikely have any negative health impacts on humans and animals, it could have some pharmacological effects [62]. Similarly, the pretreatment of soybean biomass before oil extraction, leading to SM production and the subsequent protein isolation from SM, can influence protein digestibility [63]. This study reveals that the processing steps for producing SM, which primarily focus on oil extraction, may not always optimize the conditions for improving SM protein digestibility, as the product of interest is the oil rather than the meal. The resulting SM could have experienced structural modifications, as well as the loss of components that enhance digestibility. A proximate analysis demonstrates that the cleavage of the glycosidic link in glycoprotein results in the degradation of carbohydrates. This modification may affect the accessibility of peptide bonds that are specifically targeted by protease enzymes in SMPI. As a result, the SM that is produced may have lower protein digestibility because of these structural alterations and the removal of beneficial elements.

3.4. FTIR Spectroscopy

The functional groups typically present in a soybean protein, such as amide groups (amides I, II, and III), are the distinctive backbone of protein structures [64]. These functional groups are important for the structure, function, and interaction of the protein with other molecules. However, the protein isolates may have had these functional groups changed by a lot of SM processing. We explore the use of FTIR spectroscopy to evaluate the changes that may have occurred in SMPI’s structure by comparing it to the SPI. FTIR measures the absorbance of infrared radiation by the sample material, causing a molecular vibration; the resulting spectrum shows the sample’s molecular fingerprint [57]. The FTIR spectrum of SPI and SMPI, recorded in the range of 3915 to 600 cm⁻¹, as shown in Figure 4 provides insights into their molecular structures and functional groups. As shown in the proximate analysis, SMPI has a unique OH peak at 3505 cm⁻¹, which suggests the presence of hydroxyl groups. This highlights information about SMPI’s moisture content [65]. Additionally, peaks observed at 3125 cm⁻¹ and 1625 cm⁻¹ indicate N-H stretching vibration and a C=O carboxyl group, respectively. These peaks are frequently associated with the amine group, hydrogen bond, and amide I band. This band is crucial for assessing the secondary structure of proteins. These findings align with the FTIR composition of soybeans, which are known to contain significant amounts of carbohydrates and lipids [26,28]. However, the FTIR analysis of SPI reveals a distinct set of peaks indicative of its composition. The distinct peak at 2924 cm⁻¹ indicates the presence of C-H stretching vibrations associated with lipids, typically present in higher concentrations in whole soybeans than soymeal [34]. Also, a distinct peak of NH observed at the broad absorption band of 3039 cm⁻¹ suggests the presence of amino acid symmetric stretching vibrations [66]. Furthermore, peaks at 1743 cm⁻¹ and 1626 cm⁻¹ suggest C=O stretching vibrations and amide groups, respectively, further confirming the presence of proteins.

Despite these differences, the FTIR spectra of SPI and SMPI share common functional groups. PO2—at 1235 cm⁻¹—indicates phospholipids, essential components of cell membranes in both soybeans and soybean meal [66,67]. The presence of methyl (CH₃) groups at 1395 cm⁻¹ suggests the presence of lipids, consistent with the composition of both samples. Additionally, the presence of CO-O-C at 1158 cm⁻¹ indicates the presence of ester linkages, commonly found in triglycerides [68], further supporting the lipid content in both samples. Based on these FTIR results, it was observed that SPI comprises a mixture of carbohydrates, lipids, and proteins, with a relatively higher proportion of carbohydrates and lipids than SMPI. SMPI, however, seems enriched in proteins and carbohydrates, with a lower proportion of lipids. The presence of amide groups in SMPI suggests a higher concentration of proteins, likely present in the form of residual meal after oil extraction from soybeans. Also, distinct NH peaks in SMPI indicate a higher protein concentration than SPI. The FTIR results provide valuable insights into the structural composition of both SPI and SMPI, which can further aid their characterization and utilization in various applications.

3.5. SDS-PAGE Banding Pattern of Soybean and Soybean Meal Protein Isolates

The thermal processing of soybeans during the extraction of oil may impact the quality of the resulting meal, particularly the nutritional value of the meal’s protein. High temperatures during the extraction procedure can decrease the presence of certain amino acids and alter the protein structure of the meal [69]. We used sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) to analyze SMPI and SPI in order to investigate any potential alterations in the protein molecular weight that may have occurred during the soybean oil extraction process. We observed similar banding patterns for both SMPI and SPI using two protein concentrations (50 mg/mL and 25 mg/mL). Figure 5 (Panel A) shows the banding pattern of 50 mg/mL of SMPI and SPI in lane 2 and lane 3, respectively, while Panel B shows the corresponding 25 mg/mL samples. In the 50 mg/mL samples, the protein molecular weight ranges from 250 kD to 20 kD (Panel A), whereas the range is from 150 kD to about 35 kD in the 25 mg/mL isolates (Panel B).

The appearance of visible and undegraded bands in both concentrations of SMPI revealed the presence of globulin proteins of different molecular weights. Both concentrations of SMPI showed visible and undegraded bands, indicating no changes to the meal protein. In the same way, the 50 mg/mL SPI pattern showed that the soybean used in this study had not been broken down and still had the main globular proteins that are found in soybeans, such as glycinin and β-conglycinin. Our results show that these globular proteins are not clearly visible at low-protein concentrations, as demonstrated by their absence in the 25 mg/mL sample. When up to 5 mg/mL protein concentrations were used in both samples, globular proteins with molecular weights between about 95 kD and 20 kD were visible [70]. Therefore, since the samples used in this study were not degraded, the comparable isolate band patterns at varying concentrations show that the application of heat during the extraction of soybean oil has a minimal influence on the protein composition of the resultant meal.

3.6. Deconvolution of Amide II Regions of FTIR spectra

To identify the secondary structure of SMPI and SPI, the amide regions, such as α-helices, β-sheets, and random coils, were identified by deconvoluting the amide II region of FTIR. As the amide II region of SMPI and SPI is highly sensitive to secondary structures, the deconvolution of this area was carried out to examine their structural configurations. This particular region serves as an important marker of protein structural conformation, which is predominantly reflected by the N-H bending and C-N stretching vibrations within the peptide bond. The deconvoluted spectra of SPI and SMPI are shown in Figure 6 and Figure 7 respectively with prominent peaks observed within the 1550 cm⁻¹ to 1602 cm⁻¹ range. Gaussian fits were used to resolve overlapping peaks, revealing the underlying elements that support the overall absorption process. Using the OriginPro 2024b software, we calculated the area under each peak as a percentage to illustrate how each peak contributed to this region.

As shown in Figure 6, four peaks at 1582 cm⁻¹, 1564 cm⁻¹, 1550 cm⁻¹, and 1602 cm⁻¹ were observed from the deconvoluted curve of SPI. The peak observed at 1582 cm⁻¹ is the most prominent, comprising 54% of the total area (as shown in Figure 8). As this particular amide region is attributed to β-sheet structures, which are commonly linked to the stability and rigidity of proteins, we suspect the presence of a sheet like structure held together by hydrogen bonds. Although α-helices contribute to the structural pattern of proteins, the smaller peak at 1550 cm⁻¹, which occupies 4% of the total area, suggests a minimal contribution from α-helical structures. The other peaks at 1564 cm⁻¹ and 1602 cm⁻¹ can be attributed to random coils (occupying about 17% of the area) and aggregated β-sheets (occupying 22.9% of the area), respectively.

Similarly, the SMPI samples show the same number of peaks; however, the structural contributions of these peaks differ from those of the SPI samples. Approximately 59% of the total area is occupied by the major peak linked to β-sheets in SMPI, indicating a more efficiently arranged β-sheet structure. Furthermore, the peak observed at 1564 cm⁻¹, which accounts for around 30% of the overall area, suggests the presence of significant random coils, which enhances the flexibility of the protein. Moreover, the peak linked to the α-helical structure represents more than 6% of the whole area, surpassing the 4% observed in SPI. Finally, the peak located at 1604 cm⁻³, which accounts for 4% of the overall area, indicates the possible existence of small aggregated β-sheets or extended structures. The results emphasize that secondary protein structures exist and are important for understanding the structure and operation of SMPI and SPI.

3.7. Differential Scanning Calorimetry (DSC)

The simultaneous breaking of intra- and intermolecular interactions within a protein can result in a shift from its native conformation to a denatured state [71]. This process of protein structure breakdown requires energy to disrupt the bonds stabilizing the proteins. Differential scanning calorimetry detects this energy utilization as endothermic peaks, which corresponds to the temperature at which protein loses its structural integrity [71].

Figure 9 illustrates the thermogram for SPI and SMPI. The thermogram for SMPI reveals a single broad endothermic peak centered at 106.96 °C, along with a shoulder at 89.64 °C, and a thermal transition onset temperature of 10.56 °C, with an enthalpy of 158.00 J/g. Similarly, a broader peak at 88.29 °C with an enthalpy of 190.22 J/g is observed for SPI samples. Previous studies have demonstrated that both β-conglycinin and glycinin of SPI exhibit endothermic peaks at 72–80 °C and 80 °C–94 °C, respectively [72,73]. Our findings are in agreement with these studies, as the broader peak in SPI indicates the presence of both 7S and 11S globulin proteins, with the observed denaturing temperature falling close to the expected range.

The endothermic peak and high denaturing temperature in SMPI at 106.96 °C suggest the presence of a highly stable, aggregated glycinin protein structure, which is likely due to a strong intermolecular interaction within the protein aggregates. Moreover, the shoulder of SMPI endothermic peak at 89 °C could signify β-conglycinin denaturation with slight stabilization. Additionally, a protein structure with high thermal stability is likely attributed to the presence of β-sheets, which contribute to the stability through extensive hydrogen bonding. These findings are consistent with thermal characteristics of SMPI, as reported in previous studies. Das et al., in the study of the thermal properties of pH-adjusted SMPI, observed the denaturing temperature and major endothermic peak to fall within the range of 64–102 °C [34]. Similarly, Pereira Souza et al. reported the denaturing temperature to be 141 °C and 193 °C [72]. The differences in the composition of raw materials and additional processing steps involved in the production of soymeal such as heat treatment can have a stabilizing effect on SMPI, pushing its denaturing temperature higher than that of SPI from whole soybeans.

3.8. Amino Acid Composition

Plant protein isolates are widely recognized as a highly beneficial dietary source of essential amino acids (EAA). However, the amino acid profile of plant protein isolates varies from one plant source to another, often due to variability in production processes.

Table 2. Amino acid composition of SMPI and some plant protein isolates, including FAO/WHO EAAs’ requirements for adults and children.

Amino Acid	SMPI	Soy a		Pea a	Wheat a	FAO/WHO (1985)
						Child	Adult
Essential
Phenylalanine + Tyrosine	7.94	5.4		6.3	6.1	6.3	1.9
Histidine	2.12	1.5		1.6	1.4	1.9	1.6
Isoleucine	4.3	1.9		2.3	2.0	2.8	1.3
Leucine	6.61	5		5.7	5.0	6.6	1.9
Lysine	4.8	3.4		4.7	1.1	5.8	1.6
Methionine + cysteine	1.65	0.5		0.5	1.4	2.5	1.7
Threonine	2.81	2.3		2.5	1.8	3.4	0.9
Valine	4.07	2.2		2.7	2.3	3.5	1.3
Tryptophan	1.25	1.6		0.8	1.2	1.1	0.5
∑EAA	35.55	23.8		27.1	22.3	33.9	12.7
Non-essential
Alanine	3.13	2.8		3.2	1.8
Arginine	6.47	4.8		5.9	2.4
Glycine	3.44	2.7		2.8	2.4
Proline	4.53	3.3		3.1	8.8
Serine	3.78	3.4		3.6	3.5
Glutamic acid	17.79	12.4		12.9	26.9
Aspartic acid	10.35
Hydroxyproline	0.06
Hydroxylysine	0.03
Taurine	0.04
Lanthionine	0.19
Ornithine	0.27
∑NEAA	50.08
Hydrophobic AAs	32.96	23		24.6	27.9
Uncharged polar	10.55	8.1		8.9	8.4
Basic	13.39	9.7		12.2	4.9
Acid	17.78

Values are represented in w/w% = grams per 100 g of samples. ^a Cited from Gorissen et al., 2018 [76]. FAO/WHO EAAs’ requirements for adults and children [75].

shows the amino acid profile (g/100 g of protein) of SMPI compared to the previously published amino acid profiles of protein isolates from soybeans, peas, and wheat, along with FAO/WHO-suggested [74,75] amino acid requirements. The EAA content of SMPI is 35.55 g/100 g, which is higher than soy, pea, and wheat protein isolates, which have an EAA content of 22.51, 23.8, 27.1, and 22.3 w/w%, respectively. This means that SMPI has a well-balanced amino acid composition. As shown in

Table 3. Nutritional evaluation of SMPI.

Composition	EAA (mg/g)	Reference Pattern (mg/g) [72]	AAS (%)
Histidine	21.2	19	111.6
Lysine	48	58	82.8
Tryptophan	12.5	11	113.6
Phenylalanine + Tyrosine	79.4	63	126
Methionine + Cystine	16.5	25	66
Threonine	28.1	34	82.2
Leucine	66.1	66	100.2
Isoleucine	43	28	153.6
Valine	40.7	35	116.3

, glutamic acid is the most common amino acid in all samples. This aligns with previous research on plant protein isolates. Aspartic acid and leucine are the next most prevalent in SMPI. Interestingly, SMPI showed the highest level of hydrophobic amino acids, 32.96 w/w%, indicating its poor solubility in water.

Compared to the FAO/WHO requirements (Table 2), the EAA content of all the protein isolates is generally higher or similar, except for threonine, lysine, and methionine, which fall short of the recommended levels. This suggests that SMPI is the most suitable source of essential amino acids among the evaluated plant protein isolates. However, we should make efforts to supplement diets with extra lysine, methionine, and threonine to meet the FAO/WHO criteria. The amino acid score (AAS), as shown in Protein Table 3, is an important metric for assessing a protein’s nutritional value and quality. Based on the test protein’s amino acid profile in comparison to the suggested reference pattern by FAO/WHO, the AAS identifies and reflects the limiting essential amino acids in the protein source [75]. The calculated AAS score identifies methionine + cysteine, threonine, and lysine as the limiting amino acids, with an AAS score of less than 100, in agreement with previous studies [72].

4. Conclusions

In this study, we examined the extractability of SMPI using a weak solvent (NH₄OH) across five different concentrations. We also hypothesize that the heat treatment used in soybean processing during oil extraction influences the protein composition of the resulting soymeal. We tested this hypothesis by looking at the molecular, structural, and compositional properties of protein isolates obtained from whole soybeans and soymeal. These analyses were carried out using FTIR spectroscopy, SDS-PAGE banding patterns, and other close-by analytical methods. We conducted a complete amino acid profiling of soymeal protein isolates to examine their nutritional value. This study further demonstrates the efficient use of NH₄OH in protein extraction from soymeal. Although there was no significant increase in the protein yield within the concentration range of 0.25–1%, we suggest that 0.25% NH₄OH is sufficient for maximizing protein extraction, improving the cost-effectiveness and minimizing chemical usage in the extraction process. The proximate analysis revealed that soybean meal has a higher protein concentration compared to whole soybeans, with soymeal protein having about 68%. Although this value does not meet the standard value of protein isolate (80–90%), it does signify the availability of non-protein materials in the extracted proteins.

Contrary to previous opinions that heat treatment would increase the protein digestibility of soybean meal protein, this study established that SPI has a higher digestibility than SMPI. We suspect this could be related to the cleavage of the glycosidic bonds in the carbohydrate portion of glycoprotein, resulting in the degradation of carbohydrates. This modification, caused by excessive heat treatment, may affect the accessibility of peptide bonds that are specifically targeted by protease enzymes. Consequently, SMPI may exhibit lower protein digestibility because of these structural alterations and the removal of beneficial elements. Further analysis will be required to ascertain the digestibility of soymeal. SDS page electrophoresis reveals similar molecular structures and gel banding patterns, demonstrating that heat treatment during oil extraction does not affect the protein quality of soybean meal. FTIR demonstrates the presence of proteins and carbohydrates in both SMPI and SPI. Unique peaks at 3505 and 2924 indicate an abundance of moisture and lipids, respectively, supporting claims from a proximate analysis. The several processes involved in the production of SM may not have an impact on the stability of the protein, as evidenced by the deconvolution of the amide II region of the SMPI FTIR spectra. Its superior stability over SPI is further demonstrated by the significant quantity of the β-sheet structure that exists in SMPI, making up about 59% of the overall area. One feature that makes them suitable for a wide range of functional and dietary uses is their highly stable and structured structure.

The differential scanning calorimetry (DSC) analysis complements these findings by highlighting the thermal stability of SMPI. The higher denaturation temperature of 106.96 °C for SMPI, compared to 88.29 °C for SPI, reflects the protein’s superior thermal stability. We inferred that SMPI’s overall nutritional quality and usefulness are greatly enhanced by the substantial hydrogen bonding within the β-sheet structures, which is likely responsible for its stability. The measured thermal properties show consistency with prior research and indicate that the heat treatment and subsequent processing of SMPI have a beneficial effect on its stability and nutritional content. The amino acid profile shows that soymeal protein possesses a well-balanced amino acid composition, with the presence of all essential amino acids, although limited in methionine, lysine, and threonine. These findings provide insights into using soymeal to improve global food systems and promote environmental sustainability.

It is imperative for future investigations to prioritize the optimization of extraction methods by examining the impact of other processing parameters such as temperature, pH, and the ratio of solids to solvents on the extract yield. Additionally, it is important to investigate how different combinations of enzymes work together to improve protein breakdown and digestibility, which would make them more useful in food and other industries.