Sonication integrated with micronization: structural modification to improve the in-vitro digestibility, functional, and antioxidant properties of red pepper seed protein isolate
Muhammad Faisal Manzoor, Muhammad Waseem, Tazeddinova Diana, Noman Walayat, Tosheva Zilolakhon Abduvalievna, Rana Muhammad Aadil, Shahzad Hussain, Zahoor Ahmed, Murtaza Ali, Xin-An Zeng, Abderrahmane Aït-Kaddour
TL;DR
This study shows how combining micronization and ultrasound can improve the properties of red pepper seed protein for use in food and supplements.
Contribution
The novel contribution is the combined use of micronization and ultrasound to modify red pepper seed protein structure and enhance its functional and antioxidant properties.
Findings
Combined treatment increased RPSPI stability, solubility, and emulsifying activity.
Structural changes included increased random coil and α-helix content, confirmed by FTIR.
Antioxidant and in vitro digestibility properties were significantly enhanced.
Abstract
The study aimed to extract protein from red pepper seeds and to probe the effects of micronization (MN) at 15,000 rpm for 3 and 6 min, ultrasound (US) at 720 W, 40 kHz for 10 and 15 min, and their combined impact on red pepper seed protein isolate (RPSPI). The combined treatment (MN 6 min and US 10 min) substantially reduced particle size and increased −ve zeta potential, thereby significantly increasing RPSPI stability compared to untreated and individual treatments. Structural results showed significant molecular changes: increased free-SH content and surface hydrophobicity, and decreased intrinsic fluorescence intensity, indicating improved exposure of buried residues and partial unfolding. Fourier Transform Infrared peak spectra results verified secondary structural modification, evidenced by a significant increase in random coil and α-helix content. The disruption and significant…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsProteins in Food Systems · Food composition and properties · Microbial Inactivation Methods
Introduction
1
The growing demand for plant-based proteins as substitutes for animal-derived proteins has spurred research efforts to valorize agro-industrial by-products [1]. The red pepper seed (RPS) (Capsicum annuum L.) protein is an underutilized by-product of the food processing and spice industries. RPS are a good source of proteins and bioactive compounds; however, their direct use in food formulations is limited. It is mainly associated with its compact structure, relatively low digestibility, and limited solubility, which limit its nutritional and techno-functional performance [2].
Nowadays, non-thermal physical technologies have gained attention as valuable tools for plant protein modification and for improving quality without affecting its thermosensitive bioactive compounds [1]. So, ultrasound (US) and micronization (MN) treatments can modify protein structure, improve the exposure of reactive groups, and enhance digestive and antioxidant properties. US-induced acoustic cavitation generates localized shear forces and microjets, thereby disrupting aggregates, unfolding protein chains, and increasing surface area to expose reactive groups [3]. The literature reports that the US can improve the techno-functional properties (foaming, emulsification, and solubility), in vitro digestibility, and antioxidant activity of plant proteins, including pea [4], faba bean [5], and soy protein isolates [6], by increasing molecular flexibility or causing partial denaturation. In contrast, MN decreases particle size to the submicron or micron range, thereby increasing specific surface area by disrupting physical linkages and altering structural properties within protein matrices [7]. MN of plant protein isolates, such as pea [8], soy [9], sorghum and millet [10], has been shown to improve functional properties, including hydration and dispersion stability by exposing surface-active sites and reducing particle size. These structural alterations also increase the accessibility of bound antioxidant and enzymatic sites, as reported in pea [8] and soy protein isolates [9], which exhibit improved antioxidant properties, solubility, and potential enzyme interactions compared to native proteins. The individual MN may have a limited impact on protein arrangements if it remains aggregated or tightly bound [9].
The growing interest in MN and the US; their combined processing remains unexplored, specifically from a mechanistic perspective. The existing literature has mainly focused on determining the individual processing effects of US-assisted extraction to improve the functional properties of plant proteins, assuming additive outcomes [3], [11]. The present work introduces MN as a strategic pretreatment before US processing, aiming to facilitate matrix accessibility and particle size distribution, and thereby enhancing protein susceptibility to US-induced cavitation, microstreaming and shear forces. This interplay is expected to increase controlled protein unfolding, exposure of hydrophobic and hydrophilic regions, interactions with antioxidant radicals and digestive enzymes, compared to individual treatments and the control sample [12], [13]. To date, no study has investigated the combined effect of MN and US on particle size reduction and acoustic cavitation, which have been explained for these underutilized sources.
The study objective arises from the growing global demand for sustainable, novel, and functional plant-based proteins, as well as for the valorization of by-products. During spice processing, RPS are produced in bulk quantities and remain extremely underexploited despite their high nutritional values. Improving its nutritional and techno-functional properties through novel processing techniques offers a pathway to develop value-added ingredients for the food industry and achieve waste-minimization goals. The novelty of the study lies not only in selecting an underutilized protein source (pepper seed) as a sustainable raw material, but also in the conceptual integration of the US and MN as a combined physical alteration approach, rather than as individual processing steps. The study aims to link particle-level structural breakdown with molecular-level conformational modifications, providing a new direction for the design of sequential non-thermal treatments to improve plant protein functionality.
Materials and methods
2
Sample preparation and defatting
2.1
Dried red pepper seeds (RPS) were obtained from Guangzhou, China, and ground into powder using a high-speed crusher (DFY-500C, Wenling, China). The obtained RPS flour was passed through a 40-mesh sieve to remove coarse particles. Then RPS flour was added to n-hexane at a 1:5 w/v ratio and stirred continuously for 24 h. The obtained defatted powder was first dried at room temperature, then in a hot-air oven at 50 °C for 2 h. The dried, defatted powder was sealed in polyethene bags and stored at room temperature until further use.
Preparation of red pepper seed protein isolate (RPSPI)
2.2
The RPSPI was prepared using the alkali and acid precipitation method reported by Deng, Du, Liu, Xiong, Wang, Rao, Liu, Zhao and Liao [14]. Defatted RPS powder (200 g) was dispersed in distilled water (4 L) at a 1:20 w/v ratio, and the pH of the mixture was adjusted to 11 with 1 M NaOH. After that, the mixture was stirred gently for 7 h at ambient temperature, then centrifuged (Centrifuge Instrument Co., Ltd., Changsha, Xiangyi, China) (9000×g) at 4 ℃ for 15 min, and filtered. The supernatant pH was adjusted to 4.5 with 1 M HCl, and the mixture was again centrifuged (9000×g) at 4 ℃ for 15 min. The obtained precipitates were washed with deionized water, then neutralized to pH 7, dissolved in 100 mL of deionized water, and dialyzed for 24 h at 4 ℃ against Milli-Q ultrapure water using a 3,500 Da molecular weight cutoff membrane. The RPSPI solution was freeze-dried at −80 ℃ for 48 h and stored at −20 ℃ until further processing.
Micronization (MN) treatment
2.3
The RPSPI dispersion (7:1 mg/mL) was prepared and treated with high-speed mechanical shearing using a T 25 ULTRA TURRAX® (Guangzhou Shenhua Biotechnology, China) at 15,000 rpm for 3 and 6 min. The 6 min treatment was applied in 2 × 3 min bursts, with 30 sec of cooling between bursts to avoid over-processing and heat. The samples treated for 3 and 6 min were coded MN3 and MN6, respectively.
Ultrasound (US) treatment
2.4
The RPSPI dispersion (7:1 mg/mL) was prepared, and a 40 mL sample was added to a flat-bottom flask (100 mL). Then sonication was done in an ultrasound bath (Skyman Corporation, JP-070S, China). The sample was treated for 10 and 15 min at 40 kHz, with a power of 720 W at 30 ± 2 °C. The treatment temperature was maintained by continuous water circulation at a 0.5 L/min flow rate. US treatment resulted in total energy inputs of 432 and 648 kJ for 10 and 15 min, respectively, whereas the control sample processed without US had zero energy input. The US-treated sample was coded as US10 and US15. For US treatment, the parameters were finalized based on relevant literature and preliminary trials to achieve effective protein structural alteration while maintaining protein stability.
Combined treatment
2.5
For combined treatment, the RPSPI dispersion (7:1 mg/mL) was prepared. First, MN was performed (3 and 6 min), followed by US treatment (10 and 15 min). The combined treatment samples were coded as MN3 + US10, MN6 + US10, MN3 + US15, and MN6 + US15 (sample codes are further described in the footnotes to Table 1, Table 2). After treatments, all samples were cooled in a water bath and then lyophilized for further analysis.Table 1. Effect of MN, US, and combined treatments on secondary structure content (%) of RPSPI.Treatmentsβ-Sheetα-helixβ-turnRandom coilUT21.82 ± 0.16^b^27.95 ± 0.08^e^22.29 ± 0.09^b^27.94 ± 0.30^d^MN322.01 ± 0.19^a^27.06 ± 0.15^ef^23.43 ± 0.13^a^27.50 ± 0.25^de^MN621.90 ± 0.15^b^28.03 ± 0.12^d^22.02 ± 0.18^b^28.05 ± 0.21^c^US1020.95 ± 0.18^c^29.19 ± 0.21^c^21.45 ± 0.11^c^28.41 ± 0.08^c^US1520.10 ± 0.11^c^29.85 ± 0.16^bc^20.73 ± 0.17^d^29.32 ± 0.13^b^MN3 + US1019.03 ± 0.19^d^30.35 ± 0.14^b^20.59 ± 0.08^d^30.03 ± 0.11^ab^MN6 + US1018.77 ± 0.17^e^31.28 ± 0.13^a^19.10 ± 0.10^e^30.85 ± 0.14^a^MN3 + US1518.21 ± 0.21^e^30.40 ± 0.16^ab^21.27 ± 0.07^c^30.12 ± 0.12^a^MN6 + US1519.16 ± 0.20^d^29.48 ± 0.13^c^22.35 ± 0.15^b^29.01 ± 0.18^b^Different letters show significant differences (p < 0.05) among treatments in the same column.UT: Untreated sample, MN3: Micronize for 3 mins, MN6: Micronize for 6 mins, US10: Ultrasound treatment for 10 mins, and US15: Ultrasound treatment for 50 mins, MN3 + US10: Micronize for 3 mins then ultrasonication for 10 mins, MN6 + US10: Micronize for 6 mins then ultrasonication for 10 mins, MN3 + US15: Micronize for 3 mins then ultrasonication for 15 mins, and MN6 + US15: Micronize for 6 mins then ultrasonication for 15 mins.Table 2. Impact of MN, US, and combined treatments on antioxidant properties (%) and IVPD (%) of RPSPI.TreatmentsDPPH scavenging ability (%)ABTS scavenging ability (%)∙OH scavenging ability (%)IVPD (%)UT20.23 ± 0.23^g^31.34 ± 0.28^f^24.04 ± 0.24^g^80.11 ± 0.34^d^MN322.12 ± 0.19^f^33.12 ± 0.25^e^26.22 ± 0.18^f^81.87 ± 0.45^c^MN625.32 ± 0.32^e^35.02 ± 0.21^d^27.10 ± 0.22^e^82.43 ± 0.51^c^US1026.63 ± 0.34^de^35.82 ± 0.22^d^27.73 ± 0.19^e^83.78 ± 0.41^bc^US1527.12 ± 0.21^d^37.67 ± 0.15^c^28.62 ± 0.12^de^84.96 ± 0.63^b^MN3 + US1029.92 ± 0.24^cd^38.74 ± 0.27^bc^30.12 ± 0.17^c^86.12 ± 0.56^ab^MN6 + US1032.18 ± 0.37^a^41.56 ± 0.23^a^33.01 ± 0.27^a^87.85 ± 0.78^a^MN3 + US1531.46 ± 0.17^b^37.34 ± 0.19^c^31.21 ± 0.13^b^86.96 ± 0.39^ab^MN6 + US1530.41 ± 0.26^c^39.23 ± 0.32^b^29.82 ± 0.11^d^84.12 ± 0.28^b^Different letters show significant differences (p < 0.05) among treatments in the same column.UT: Untreated sample, MN3: Micronize for 3 mins, MN6: Micronize for 6 mins, US10: Ultrasound treatment for 10 mins, and US15: Ultrasound treatment for 50 mins, MN3 + US10: Micronize for 3 mins then ultrasonication for 10 mins, MN6 + US10: Micronize for 6 mins then ultrasonication for 10 mins, MN3 + US15: Micronize for 3 mins then ultrasonication for 15 mins, and MN6 + US15: Micronize for 6 mins then ultrasonication for 15 mins.
Characterization of structural properties of RPSPI
2.6
Free sulfhydryl (−SH) contents
2.6.1
The RPSPI sample was measured for free-SH content using Ellman’s method as documented by Manzoor, Waseem, Diana, Wang, Ahmed, Ahmed, Ali and An-Zeng [3]. The lyophilized RPSPI solutions were added to Tris-glycine buffer (0.086 M Tris, 0.09 M glycine, and 4 mM EDTA, pH 7). The admixture was incubated at ambient temperature for 1 h in a water bath shaker. Thereafter, the sample mixture was centrifuged at 10,000×g for 15 min, and 3.0 mL of the supernatant was collected, mixed with 30 μL of prepared Ellman’s reagent [4 mg 5, 5′-dithiobis (2-nitrobenzoic acid) in Tris-glycine buffer], and incubated for 20 min. Absorbance of the sample and reagent was recorded using a UV spectrophotometer (Amodel Onlab, EU-2600, Shanghai, China) at A_412_ nm for estimation of free-SH content expressed as μmol/g of protein with a molar extinction coefficient of 13,600 M^−1^ cm.
Surface hydrophobicity (H0)
2.6.2
The H_0_ for the RPSPI sample was estimated using the protocol outlined by Gulzar, Martín-Belloso and Soliva-Fortuny [15]. The RPSPI sample was diluted using the clean distilled water (Conc. 0.1–0.5 mg/mL). Then, an accurate volume of 20 μL of 8-anilinonaphthalene-1-sulfonic acid (ANS) (i.e., 8 mmol/L) was thoroughly mixed in the diluted 5 mL RPSPI solution. Afterwards, the fluorescence intensity was measured on a fluorescence spectrophotometer (Amodel Onlab, EU-2600, Shanghai, China) at 360 and 463 nm for excitation and emission, respectively. Consequently, the H_0_ index was estimated by measuring the first slope of fluorescence intensity versus protein concentration.
Particle size and zeta-potential
2.6.3
RPSPI particle size and zeta potential were determined by adopting the method of Lu, Xiong, Yao, Zhang and Wang [16], with minor amendments. The 7 mg/mL RPSPI solution was prepared and centrifuged at 12,000×g for 15 min at 10 °C. The 1 mL of the centrifuged supernatant was diluted to a final concentration of 0.625 mg/mL. The measurement was performed with a Zetasizer Nano (Zetasizer Ultra, Malvern Panalytical, UK) with a refractive index (γ) of 1.330 at 25 °C.
Fourier-transformed infrared spectroscopy (FTIR)
2.6.4
The secondary structure of RPSPI samples was measured using FTIR spectroscopy [17]. Each RPSPI lyophilized sample powder was mixed with potassium bromide (KBr) at a 1:100 (w/w) ratio and pressed into pellets. The RPSPI samples were characterized on an FTIR spectrometer (Spectrum 3 FT-IR spectrometer, PerkinElmer, USA) over 4000–400 cm^−1^, at 4 cm^−1^ resolution. After that, the secondary structure content of all samples was determined with PeakFit 4.12 software (Systat, CA, USA).
Intrinsic fluorescence spectra
2.6.5
RPSPI samples were analyzed for fluorescence spectra using a fluorescence spectrometer (Hitachi F-4600, Kyoto, Japan) according to the method validated by Deng, Du, Liu, Xiong, Wang, Rao, Liu, Zhao and Liao [14], with slight modifications. Before analysis, each RPSPI sample was uniformly adjusted to a protein concentration of 0.3 mg/mL. Scanning conditions were set to 700 V, 1200 nm/min, and 280 nm excitation. However, the emission wavelengths were set to 290 and 500 nm, and the slit widths were set to 2.5 nm for both the excitation and emission.
Scanning electron microscopy (SEM)
2.6.6
The surface structural changes in RPSPI samples were investigated using a field-emission scanning electron microscope (Thermo Fisher Scientific, Quattro S, Brno S.R.O., Vlastimila Pecha, Czech Republic) with slight modification. The RPSPI samples were attached to double-sided sticky tape affixed to the SEM stubs and coated with 8 nm of gold. Gold was then sputtered onto the RPSPI sample to prevent charge accumulation using the Cressington sputter coater (Model 108 auto) for 210 s, at 20 mA. However, the acceleration voltage was set to 10 kV, and RPSPI samples were measured in the SEM at a magnification power of 1000X.
Crystallography
2.6.7
The RPSPI sample’s X-ray diffraction (XRD) patterns were measured using a D8 DISCOVER diffractometer (Bruker, AXS, Germany) with Cu Kα radiation (λ = 1.5406 Å). The diffractograms of each sample were recorded between 10° and 90° with a counting time of 0.5 s and a step size of 0.05°.
Differential Scanning Calorimeter (DSC)
2.6.8
RPSPI samples were subjected to thermal characterization measurements according to the method of Wang, Ma, Wang, Zhao and Liao [18] using a DSC (DSC-8000, PerkinElmer, USA). In protocol, 2 mg of each RPSPI sample was combined with 6 μL of deionized water, then sealed in a hermetic aluminium pan. Thereafter, each RPSPI sample was equilibrated for 10 min at 30 °C and heated from 30 to 150 °C at 10 °C/min, with each sample measured in triplicate. The empty aluminium pan was used as a reference. Subsequently, the enthalpy change (ΔH), temperature (T_0_), and denaturation temperature (T_d_) were recorded using DSC (TA Instruments, New Castle, DE, USA).
Characterization of functional properties of RPSPI
2.7
Turbidity
2.7.1
The RPSPI samples were tested for turbidity using the method of Manzoor, Waseem, Diana, Wang, Ahmed, Ahmed, Ali and An-Zeng [3], with slight modifications. Spectrophotometric (Amodel Onlab, EU-2600, Shanghai, China) absorbances of each RPSPI sample, supernatant, and deionized water (i.e., reagent blank) were recorded at 400 nm.
Solubility
2.7.2
The solubility of each RPSPI sample was determined using the protocol described by Wang, Zhou, Wang, Wang, Jiang, Liu and Yu [19]. About 10 mg/mL of each RPSPI sample was diluted in distilled water, mixed, and centrifuged at 10,000×g for 15 min. Before and after centrifugation, each RPSPI sample dispersion was tested for protein content by adopting the biuret method with bovine serum albumin as a standard. However, the solubility was determined via a formula as given below;
Oil holding capacity (OHC) and water holding capacity (WHC)
2.7.3
For WHC and OHC, 1.0 g of each RPSPI sample was mixed with clean distilled water and rapeseed oil (40 mL each). The resultant solution mixture was thoroughly mixed separately for 1 min. After thorough mixing, the admixtures were kept at room temperature (21–24 °C) for 6 h, then centrifuged at 3000 × g for 30 min at 20 °C.
where: W0 = Isolates’ mass, tube, and absorbed oil or water; W1 = Protein isolates’ mass and tube; W3 = Mass of RPSPI.
Foaming characteristics (FC) and foaming stability (FS)
2.7.4
RPSPI samples’ FC and FS were measured in accordance with the method of Loushigam and Shanmugam [20]. Precisely measured 1 % (w/v) of each RPSPI sample solution (pH 7) was mixed with clean distilled water for 60 min at 25 ℃. Thereafter, about 15 mL of protein admixture was homogenized for 3 min at room temperature. Total volume (mL) was estimated using a graduated cylinder before and after the whipping. The foam layer volume was calculated at 15 and 30 min relative to the samples’ initial foam volume to estimate FC and FS:
Emulsion activity index (EAI) and stability index (ESI)
2.7.5
The RPSPI sample was measured for EAI and ESI using the recent method of Manzoor, Waseem, Diana, Wang, Ahmed, Ahmed, Ali and An-Zeng [3]. Briefly, 3 mL (0.1 % (w/v) of RPSPI sample was poured into 3 mL of pepper seed oil and homogenized for 2 min at 12000 rpm. Afterwards, an accurate volume of 50 μL from the emulsion was added to 5 mL (0.1 %) of sodium dodecyl sulfate (SDS) and mixed thoroughly. For EAI, the sample absorbance (A_0_) was recorded at 500 nm against a 0.1 % SDS solution as a reagent blank. After 10 min, the method was performed again for sample absorbance (A_10_) for determining the ESI using the following formulae;
where: ϕ is oil (%) volume fraction and T, D, and C are constant (2.303), concentration of proteins (g/mL), and dilution ratio.
Characterization of antioxidant properties of RPSPI
2.8
DPPH radical scavenging activity
2.8.1
The protocol of Cai, Huang and Wang [21] was followed to determine the DPPH radical-scavenging activity of RPSPI. Absorbance of samples, reagent blank, and control was taken at 517 nm, DPPH (%) calculations were made using the following equation;
where: A0 = Absorbances of control upholding distilled water and DPPH reagent solution; Ai = Absorbances of DPPH solution mixed sample; A_j_ = Absorbances of the ethanol mixed sample.
ABTS radical scavenging activity
2.8.2
The method of Yang, Cai, Yan, Tian, Du and Wang [22] was followed to estimate ABTS antioxidant activity in RPSPI samples. Absorbances of the sample and reagent blank were measured at 734 nm, and calculations were made using the equation given below;
where: A_1_ = Absorbances of ABTS solution containing sample; A_0_ = Reagent blank (distilled water) containing ABTS solution.
Hydroxyl radical scavenging activity
2.8.3
OH radical-scavenging activities of the RPSPI sample were estimated using the approach outlined by Manzoor, Waseem, Diana, Wang, Ahmed, Ahmed, Ali and An-Zeng [3]. Spectrophotometer (Amodel Onlab, EU-2600, Shanghai, China) absorbances of each sample, and the reagent blank were taken at 510 nm, and final values were calculated as;
where: A_0_ = Absorbance of reagent blank (distilled water); A_1_ = Absorbance of RPSPI sample; A_2_ = Absorbance of control (H_2_O_2_)
In-vitro protein digestibility (IVPD)
2.9
The IVPD of each RPSPI sample was estimated using the method of Vashishth, Semwal, Naika, Sharma and Kumar [23], with minor amendments. The RPSPI suspensions (7 mg/mL) were adjusted to pH 8.0 with 0.1 M NaOH, and then the multienzyme solution (1.5 mL) was added, thoroughly mixed, incubated for 10 min at 37 °C in a water bath, and the pH decline was measured. The multienzyme solution (pH 8) was prepared with protease (1.3 mg/mL), chymotrypsin (3.1 mg/mL), and trypsin (1.6 mg/mL), and Casein was used as a control. IVPD was calculated as follows:
where: Xf indicates the pH value of the solution (after 10 min).
Statistical analysis
2.10
All RPSPI samples were analyzed in triplicate (n = 3), and the acquired results were expressed as Mean ± S.D. Results were statistically analyzed (Minitab 14.1, Minneapolis, USA) by one-way ANOVA at p < 0.05. Data were plotted using OriginPro (8.0, Minneapolis, USA).
Results and Discussion
3
Impact of US, MN, and combined treatments on structural properties of RPSPI
3.1
Free sulfhydryl (−SH) group
3.1.1
The –SH group of RPSPI is a key marker of structural modifications, revealing the formation or disruption of disulfide bonds during US, MN, and their combined treatments by exposing sulfur-containing amino acids, especially cysteine residues [24]. Results indicate that the untreated RPSPI showed –SH contents of 2.61 ± 0.12 µmol/g, as presented in Fig. 1A. MN treatments moderately increased the –SH content to 3.96 ± 0.09 µmol/g, while US treatments significantly (p < 0.05) increased it to 4.72 ± 0.13 µmol/g compared to untreated and micronized RPSPI. The combined treatment of (MN6 + US10) yielded a significantly higher (p < 0.05) SH content (5.21 ± 0.14 µmol/g), suggesting a combined improvement in the exposure of the –SH group (Fig. 1A). After US treatments, an increase in –SH contents may be associated with acoustic cavitation, which generates shear forces, microstreaming, and localized turbulence. It’s ultimately directed to partially unfold the structure of RPSPI and expose the buried sulfur-containing amino acids to the surface. Moreover, shear forces generated by US waves and cavitation can break –SS bonds, thereby increasing –SH content. The increases in –SH content, indicative of protein unfolding, may improve functional attributes [24]. Yao, Li, Wu, Martin and Ashokkumar [25] reported a significant increase in –SH content in US-treated hempseed protein isolate correlated with cavitation-induced disruption and unfolding of non-covalent interaction.Fig. 1. Free sulfhydryl group and surface hydrophobicity (A), Particle size and Zeta Potential (B) of RPSPI after MN, US, and combined treatments.
After MN treatment, particle size reduction by high-energy was directed to improve the –SH contents, which was assumed to be to a lesser extent than US. It's associated with increased solvent accessibility and surface area, enabling quantification of –SH contents previously immersed within the matrix. Moreover, MN may induce mild structural modifications without significant denaturation, as evidenced by a moderate increase in –SH content [7]. The combined US-MN treatments showed the most significant increase (p < 0.05) in –SH content, due to a combined effect arising from the sequential actions of MN and US. MN first improves the exposure of reactive groups by reducing the particle size of RPSPI, breaking down residual aggregates, and increasing surface area. Later, the US engages in an intractable conflict with a smaller, more exposed, resulting in the disaggregation and partial unfolding of RPSPI. A similar study by Kamani, Semwal and Meera [12] reported a combined effect of US combined with physical modification of black gram protein, resulting in elevated –SH levels and enhanced protein functionality. Results indicate that prolonged combined treatments lead to oxidation of –SH groups to –S–S– bonds and produce aggregation via hydrophobic interactions, ultimately reducing the –SH contents.
Surface hydrophobicity (H0)
3.1.2
H_0_ indicates the hydrophobic amino acid quantity in protein exposed on the surface and is a sensitive index of unfolded/tertiary structural modification in protein [11]. In our study, the H_0_ of the untreated RPSPI was 1770 ± 32, while MN induced a moderate increase in H_0_ up to 1972 ± 41. After MN treatment, the rise in H_0_ is associated with a significant reduction in particle size, thereby improving solvent accessibility and reducing aggregation, ultimately exposing more hydrophobic domains at the surface. Hao, Zhang, Yang, Zhang, Wu, Liu, Sui and Zhang [9] reported that high-speed shearing homogenization significantly increased the H_0_ of soybean protein isolate. US treatment also significantly increased (p < 0.05) the H_0_ to 2120 ± 46, higher than in the untreated and micronized samples, as shown in Fig. 1A. After US treatments, the observed increase is likely due to shear forces induced by cavitation, which disrupt the internal hydrophobic core arrangement, partially unfolding the protein, and exposing buried hydrophobic residues to the solution [26]. Our results align with an increase in H_0_, reported for whey protein isolate [11] and amaranth protein isolate [27], due to protein unfolding resulting from changes in tertiary structure induced by US.
In contrast, combined (MN + US) treatments induced a highly significant increase (p < 0.05) in H_0_ to 2335 ± 44, indicating a 32 % increase over the untreated sample. The combined effect of MN + US showed the highest yield. Firstly, the MN decreases the overall particle size and narrows residual aggregates, thereby promoting maximal exposure of hydrophobic domains. The US disaggregates proteins and disrupts their tertiary structure, thereby exposing hydrophobic moieties further. In all samples, this increase in H_0_ is associated with changes in tertiary structure, as evident in FTIR/fluorescence spectra (Fig. 2, Fig. 3) and a significant decrease in particle size (Fig. 1B). Moreover, a reduction in H_0_ of samples subjected to prolonged US and MN combined might be due to reaggregation of hydrophobic patches and partially unfolding protein molecules, which extend non-covalent bonds and ultimately protect the hydrophobic regions [27]. However, our chosen combined treatment parameters are still beneficial compared with untreated and individual treatments.Fig. 2. Particle Size Distribution (A) Fluorescence intensity (B) of RPSPI after MN, US, and combined treatments.Fig. 3FT-IR spectrum of RPSPI after MN, US, and combined treatments (A&B).
Average particle size (PS) and zeta potential
3.1.3
Results revealed that untreated RPSPI had a zeta potential of −53.38 ± 0.38 mV and volume-mean PS 554.9 ± 6 nm. The MN individual significantly shifted the zeta potential to −57.43 ± 0.28 mV and decreased PS to 405.3 ± 7 nm (Fig. 1B). During MN, the PS reduction is associated with high collision forces and mechanical shear, which disrupts RPSPI aggregates into smaller, uniform particles. Moreover, it increases the particular surface area and exposes charged amino and carboxyl groups buried earlier in the protein aggregates. It intensifies electrostatic repulsion between particles, thereby increasing the magnitude of zeta potential and colloidal stability [10]. Earlier, Zhao, Yuan, Yuan, Zhao, Kang, Zhu, He and Ma [28] reported a reduction in PS and an increase in zeta potential due to mechanical shear-induced dissociation of aggregates, partial unfolding, and electrostatic repulsion in quinoa protein after MN.
In comparison, US treatment alone more significantly shifted the zeta potential (−57.67 ± 0.29 mV), decreasing PS to (345.1 ± 6 nm) than MN and the untreated sample. US-induced cavitation reduces PS because collapsing microbubbles produce strong micro-jets and shear that break large RPSPI aggregates into smaller residues. Concurrently, these US-induced mechanical impacts uncover buried charged (–NH_3_^+^, –COO^–^) groups, thereby improving their surface ionization, the absolute value of the zeta potential, colloidal stability, and electrostatic repulsion. Our findings align with those of Kahraman, Petersen and Fields [29], who reported a significant reduction in PS and an increase in the absolute value of zeta potential in hemp concentrate under US treatment. The combined treatment (MN → US) showed a significant shift in absolute zeta potential (−59.46 ± 0.32 mV), exhibited stronger electrostatic repulsion, and induced the smallest PS (225.8 ± 5 nm). These modifications correspond to increased RPSPI functionality and dispersion stability. Combined treatments initially decreased PS via strong collisions, cavitation, and shear forces, thereby disrupting large protein aggregates and unmasking buried charged sites, improving zeta potential. In contrast, combined treatment prolonged processing induces protein re-aggregation through excessive energy input, hydrophobic interactions, and protein unfolding, ultimately generating larger particles and decreased surface charge stability, revoking the earlier effects. Our findings align with those of Shi, Liu, Hu, Gao, Qayum, Bilawal, Munkh-Amgalan, Jiang and Hou [13], who reported a significant decrease in PS under the combined effects of homogenization and US in whey protein isolate.
The particle size distribution in Fig. 2A exhibited that MN and US treatments, individually and in combination, significantly change PS. The untreated sample showed a broad bimodal distribution with large protein aggregates. After MN, US, and combined treatment, the broader peak with lower intensity moved toward smaller diameters, revealing disruption of protein aggregates and good colloidal stability. These results also supported the findings on zeta’s potential.
Secondary structure (FTIR spectra)
3.1.4
The secondary structural changes in RPSPI after the MN, US, and combined treatments were investigated by FT-IR, and the spectra revealed characteristic Amide bands that show significant confirmation changes, as presented in Fig. 3 A&B. These distinctive absorption peaks mainly correspond to the polypeptide backbone of central amide regions. The Amide A region, ranging from 3500 to 3000^−1^, has a unique peak associated with stretching vibrations of O-H and N-H bonds in the polypeptide backbone, which are mainly linked with the H-bonds. All the treated samples showed a broad peak in this region, except for the combined treatment (MN → US) at prolonged treatment time. Structure-specific fingerprints describing important Amide I band were recorded between 1702–1590 cm^−1^, centralized at 1660 cm^−1^, while secondary peaks were observed between 1586–1485 cm^−1^ for Amide II, centred at 1542 cm^−1^, and for Amide III centred at 1237 cm^−1^. The Amide I region is primarily linked with C
<svg xmlns="http://www.w3.org/2000/svg" version="1.0" width="20.666667pt" height="16.000000pt" viewBox="0 0 20.666667 16.000000" preserveAspectRatio="xMidYMid meet"><metadata> Created by potrace 1.16, written by Peter Selinger 2001-2019 </metadata><g transform="translate(1.000000,15.000000) scale(0.019444,-0.019444)" fill="currentColor" stroke="none"><path d="M0 440 l0 -40 480 0 480 0 0 40 0 40 -480 0 -480 0 0 -40z M0 280 l0 -40 480 0 480 0 0 40 0 40 -480 0 -480 0 0 -40z"/></g></svg>
O stretching, Amide II N–H bending and C–N stretching vibrations, while Amide III C–N and C–O stretching and N-H and OC–N bending vibrations [30]. Results revealed that the shape and position of the FT-IR spectral peaks of RPSPI were virtually unchanged among all treatments. In contrast, changes in the absorbance of RPSPI spectra at specific wavenumbers were clearly observed (Fig. 3A and B). This absorbance variation shows a systematic relationship with different treatments; during MN, US, and combined treatments, absorbance increases significantly across all the wavenumbers compared to the untreated sample. In contrast, it reduces with combined treatment and prolonged treatment duration.
A deconvolution was performed for detailed examination of secondary structure changes in the Amide I region, which showed a significant increase (p < 0.05) in the structure of random coil and α-helix, while a significant (p < 0.05) reduction in β-turn and β-sheet contents, as presented in Table 1. The most significant increase in random coil and α-helix and a decrease in β-turn and β-sheet was observed in the combined treatment (MN5 + US10) due to a combined effect. These results suggested that physical processing caused the rearrangement and unfolding of RPSPI-ordered structures. In the region, the observed band at ∼1660 cm^−1^ indicated that the α-helix is stabilized by intramolecular H-bonding of the peptide backbone, which is usually broken down or altered under potent cavitational or mechanical forces. An increase in the α-helix content indicated initial unfolding followed by partial refolding of RPSPI. The concurrent increase in random coil content supports the idea that physical processing caused partial denaturation of RPSPI, exposed hydrophobic residues, and changed intact β-structures into flexible arrangements (1640–1620 cm^−1^). Mechanistically, US-induced microbubbles form, grow, and implosively collapse, generating localized high shear forces, temperatures, and pressures that transiently break non-covalent interactions and H-bonds within proteins, leading to random coil formation and the unfolding of β-sheets [4]. The MN-induced intense turbulent flow and shear gradients can cause unfolding and mechanical fragmentation of macromolecules [9]. While integrated (MN → US) processing impacts become effective, US cavitation increases dispersion and disaggregation initiated by MN, providing peptide bonds with uniform solvent exposure and modifying secondary structures [31].
A decrease in β-turn and β-sheet contents revealed weakening of van der Waals and H-bond interactions that strengthen their arrangements, and an improvement in backbone mobility. This change might be linked to improved functional properties and solubility, as β-sheet disruption enhances the accessibility of hydrophilic groups and improves molecular flexibility through US and MN-induced conformational changes. Earlier, Hao, Zhang, Yang, Zhang, Wu, Liu, Sui and Zhang [9] and Chen, Guo, Wang, Yang, Chen, Wen, Yang and Kan [32] reported increases in α-helix and random coil and a decrease in β-structures, due to partial refolding and unfolding of the backbone in soy protein isolate and Qingke protein after US and homogenization. The Amide II region peak (1542 cm^−1^) also confirms these rearrangements, as this region derives from C–N stretching and N–H bending; intensity changes or shifts exhibited alteration in H-bonding geometry. Furthermore, the noticeable peaks at 1237 cm^−1^ (Amide III) revealed modification in N–H in-plane bending with C–N stretching coupled, align with reduced β-order and increased random coil [33].
Furthermore, the amide I band broadening and shift (around 1660 cm^−1^) show the breakdown of internal H-bonds and exposure of CO groups to the aqueous solution. Also, the Amide III band enhancement around 1237 cm^−1^ supports greater backbone flexibility and conformational destabilization. Overall, these changes enhance the foaming and emulsifying properties of RPSPI because the partially unfolded proteins more readily align at water–oil interfaces and form viscoelastic films.
Intrinsic fluorescence spectroscopy (FS)
3.1.5
Intrinsic FS is the most frequently used technique to investigate the tertiary structure of proteins because fluorescence emission generally arises from aromatic amino acid residues, specifically tryptophan, whose λ_max_ (emission maxima) and intensity are sensitive to folding state and polarity of their microenvironment [34]. The slight decrease in fluorescence intensity (FI) was observed in MN, US, and combined treatments compared to the untreated sample in the following manner: Untreated > MN > US > Combined treatment without any shift to the peak position, as presented in Fig. 2B. Results indicate that λ_max_ did not change significantly after MN and US treatments, while a slight change was observed after the combined treatments, showing a more quenching chromophore environment without noticeable modification in polarity.
In untreated RPSPI, high-intensity spectra showed a comparatively compact confirmation due to tryptophan buried in the hydrophobic core, which prevented quenching. After MN treatment, a decrease in FI suggested conformational rearrangement or partial unfolding, promoting the exposure of aromatic amino acid residues to the microenvironment and increased non-radiative quenching. It's due to shear and turbulence forces that break down weak hydrophobic interactions, H-bonds, and van der Waals interactions, causing protein unfolding or disaggregation [35]. Due to cavitation, a decrease in FI after the US induces bubbles that burst violently, generating local and micro-jet turbulence, ultimately promoting protein aggregation or unfolding, depending on duration and power. Earlier, Wang, Yang, Sun, Zhu, Lian, Dai, Xu, Tong, Wang and Jiang [17] reported a similar decline in FI for soy protein Isolate after US treatment, associated with increased exposure of tryptophan residues and improved quenching by H_2_O molecules. Even if the λ_max_ did not shift, the FI decrease indicates a decline in quantum yield related to the possible formation of intra- or intermolecular aggregates and solvent exposure.
The combined treatments showed the lowest FI and a slight change in λ_max_, proposing a combined effect of structural breakdown. The successive application of MN and then US induced a more pronounced physical rearrangement and disintegration of protein regions, leading to maximum aggregation or unfolding. It enhances the accessibility of chromophores to polar quenchers and reduces the efficiency of radiative emission. Earlier, in whey protein isolate, Shi, Liu, Hu, Gao, Qayum, Bilawal, Munkh-Amgalan, Jiang and Hou [13] reported a similar combined effect decline in FI. Mechanistically, no significant change in λ_max_ is proposed; the local polarity of tryptophan residues did not change drastically, while a decrease in FI shows that the RPSPI tertiary structure became less compact or looser. Earlier, a decline in FI correlates with decreased tertiary structure compactness and improved H_0_ [36]. FT-IR outcomes firmly endorsed that these tertiary structural modifications described a significant increase in random coil and α-helix contents and a decrease in β-turn and β-sheet contents, suggesting disruption of ordered H-bonding and partial unfolding. These results showed significant changes in tertiary structure, influencing techno-functional characteristics such as emulsification, solubility and digestibility.
Surface morphology (SEM)
3.1.6
The illustrative SEM images of RPSPI treated with MN, US, and their combined treatments reveal distinct morphological alterations compared to untreated RPSPI, as shown in Fig. 4. The untreated RPSPI surface exhibited a compact, irregular, larger lamellar and aggregated structure. While MN, US, and their combined treatment of RPSPI resulted in a significant breakdown of these compact aggregates, yielding smoother, fragmented, and smaller particles with improved surface porosity, as shown in Fig. 4. These microstructural changes align with the results of particle size reduction and increased foaming, solubility, and emulsifying properties.Fig. 4. Illustrative SEM images of RPSPI after MN, US, and their combined treatments. UT: Untreated sample, MN3: Micronize for 3 mins, MN6: Micronize for 6 mins, US10: Ultrasound treatment for 10 mins, and US15: Ultrasound treatment for 50 mins, MN3 + US10: Micronize for 3 mins then ultrasonication for 10 mins, MN6 + US10: Micronize for 6 mins then ultrasonication for 10 mins, MN3 + US15: Micronize for 3 mins then ultrasonication for 15 mins, and MN6 + US15: Micronize for 6 mins then ultrasonication for 15 mins.
The observed structural modification, particularly loosening, revealed the reorganization and partial unfolding of RPSPI molecules during processing. MN loosened the RPSPI structural arrangement due to a decrease in particle size; these modifications were possibly caused by cessation of linkage, followed by shearing treatment. So, the MN processing breaks the globular structure and prevents aggregate formation by reducing particle size [10]. Similar outcomes were reported for MN protein isolated from sorghum and pearl millet [10] and from soy [9]. US-induced cavitation produced intense localized high temperatures and micro-shear forces, breaking van der Waals forces, hydrophobic interactions, and H-bonds. It leads to the breakdown of RPSPI aggregates into smaller particles, exposing hydrophilic groups and thereby increasing solubility by promoting protein-water interactions [37]. Moreover, micrographs confirm the FTIR outcomes, which showed a decrease in β-turn and β-sheet with an improvement in random coil and α-helix structures. A reduction in β-turns and β-sheets indicates partial RPSPI denaturation, which helps generate a more porous, finer microstructure by inducing a more flexible molecular arrangement [38]. Conformational flexibility promotes interfacial rearrangement and adsorption during foaming and emulsification, thereby yielding better functional properties, as shown in Fig. 6, Fig. 7. Also, a decline in FI suggests a more polar environment due to the exposure of tryptophan residues, corroborating the tertiary structure modification observed in SEM micrographs, which showed surface fragmentation and roughening [17]. SEM micrographs of the combined treatment, specifically MN integrated with US for 15 min, showed decreased surface smoothness and partial reaggregation of finer particles, implying exposure-induced molecular re-association. This outcome aligns with the decrease in foaming, emulsifying, and solubility properties observed after prolonged treatments, which is probably associated with protein–protein relinkage through disulfide or hydrophobic interactions. So, MN combined with US for a 10 min treatment emerges as optimal for getting a structural breakdown without further aggregation.Fig. 5X-ray diffraction (XRD) spectra (A), and Thermal properties (B) of RPSPI after MN, US, and combined treatments.Fig. 6. Solubility and Turbidity (A), WHC and OHC (B) of RPSPI after MN, US, and combined treatments.Fig. 7FC and FS (A), EAI and ESI (B) of RPSPI after MN, US, and combined treatments.
Crystallography (XRD)
3.1.7
XRD analysis was performed to assess crystalline structure modifications in RPSPI treated with MN, US, and their combined treatments. XRD spectra shown in Fig. 5A revealed two peaks near 2θ = ∼20◦ and 2θ = ∼10° for all RPSPI-treated samples; these are labelled as crystalline regions II and I, suggestive of ordered protein structural domains [15]. Upon MN treatment, the higher-angle peak shifts slightly from ∼19.94° (untreated) to ∼19.91°, and the lower-angle peak shifts from ∼9.98° to ∼9.71°. In US treatment, the intensity of higher-angle peaks decreased by ∼19.86°, while that of lower-angle peaks was reduced by up to ∼9.59°. In combined treatment, the higher and lower-angle peaks shift significantly up to ∼19.64° and ∼9.38°, respectively. The most significant effect was observed with the combined treatment of MN6 + US10. Moreover, after a combined US treatment for 15 min, there was a slight increase in crystallinity, as evidenced by increased intensity and peak sharpening.
The diffraction peaks revealed that the native RPSPI contains aggregated or semi-ordered domains, in line with other protein XRD results associating residual order with functionality [39]. A progressive decrease in intensity suggests that MN treatment partially disrupts or weakens intermolecular interactions (H-bonds and hydrophobic interactions) within crystalline domains. It indicates the loss of relative crystallinity. The reduction in crystal size is inversely associated with protein particle size (PS). In our results, a decrease in PS of RPSPI was observed following physical treatment, confirming a loss of relative crystallinity. During US treatment, shear forces, microstreaming, and cavitation expose hydrophobic regions and unfold protein aggregates, thereby altering tertiary and secondary structures. Our results showed that the US reduced the intensity of higher and lower-angle peaks and broadened them, indicating disruption of a more crystalline domain, increasing the d-spacing and decreasing ordering [15]. Under combined treatment, a more significant impact was observed with the sequence shear + US, which first disrupts larger aggregates via MN and then via US finer-ordered domains. Prolonged combined treatment, a slight increase in peak intensity indicating re-ordering or partial re-aggregation of RPSPI under higher energy input; this phenomenon was verified by Xu, You, Kashenye, Zheng, Li, Zhang and Yang [4], where prolonged US treatment causes reaggregation.
Our XRD results are consistent with fluorescence and FT-IR data: a decline in fluorescence intensity indicates a modification of the tertiary structure, and FT-IR shows an increase in random coil and α-helix content and a decrease in β-turn and β-sheet content. Mechanistically, a reduction in intensity and broadening of the d-spacing in XRD patterns indicate that β-sheet-rich aggregates are unfolded into coils and helices, and disruption of the tertiary structure. A reduction of β-sheet content associated with loss of ordered arrangement and the emergence of random coil indicates a more disordered conformation.
Thermal properties
3.1.8
The thermal behavior of the RPSPI, evaluated by DSC, under MN, US, and combined treatments, is shown in Fig. 5B as DSC thermograms. The thermograms showed that the RPSPI underwent significant changes in its enthalpy change (ΔH) and denaturation temperature (Td) after the MN, US, and combined treatments. The untreated RPSPI sample showed an endothermic transition with T_0_ of 49.60 °C, Td of 83.72 °C, and ΔH of 121.75 J/g. After MN treatment, significant changes were observed with T_0_ up to 55.71 °C, Td up to 86.95 °C, and ΔH up to 116.64 J/g compared to the untreated sample. MN-induced intense cavitation microflows, turbulence, and mechanical shear disrupt intramolecular hydrophobic interactions, H-bonds, and some weaker disulfide linkages within the protein structure, leading to partial unfolding, decreased structural compactness, and reduced aggregate size [12], [40]. Our results align with those of Thakur, Sharma, Khatkar and Sharma [10], who reported a slight increase in Td and a decrease in ΔH after homogenization, due to the formation of less-ordered, smaller aggregates and the breakdown of tertiary structures (disintegration of the linkage).
After US treatment, a significant decrease in Td to 85.20 °C, and ΔH up to 115.50 J/g were observed compared to the untreated sample. US-induced cavitation produced localized transient temperature–pressure fluctuations, microjets, and high shear that break weak interactions, expose hydrophobic regions, cause partial unfolding, and facilitate small-scale rearrangement [12], [41]. This ordered-domain disruption indicates some stabilizing interactions must be disrupted during DSC denaturation, thereby resulting in the lower enthalpy. Moreover, ΔH suggests the amount of heat required to denature the protein, reflecting the net energy required to disrupt endothermic hydrogen bonds and exothermic hydrophobic interactions. Our results align with those of Habib, Singh, Ahmad, Jan, Gupta, Jan and Bashir [42], who reported a significant decrease in the enthalpy change and denaturation temperature after 20 min of US processing of pumpkin seed protein isolate. Mechanistically, it indicates US-induced breakdown of hydrophobic interactions and H-bonds, facilitating loosening of the native structure and partial unfolding. It's shown that less energy is required to cause thermal denaturation, indicating reduced structural compactness and improved molecular disorder. Similarly, a decrease in ΔH for US-treated faba bean protein was reported by Gulzar, Martín-Belloso and Soliva-Fortuny [15].
The combined treatment exhibited the most significant reduction in ΔH (121.75 → 98.25 J/g), showing the lowest enthalpic demand for denaturation. Mechanistically, MN first exposes hidden domains, loosens the structure, and decreases aggregate size; then US further breaks down weak covalent bonds (SH or S–S interactions), exposes regions, and rearranges the protein network. So, the combined treatment increases structural breakdown and thus reduces ΔH the most. Moreover, upon prolonged combined treatment, a slight rise in ΔH (98.25 → 104.63 J/g) was noted. That might be due to partial re-stabilization or reaggregation of protein networks: they may have formed intermolecular cross-links or small aggregates, which slightly increase the energy barrier to denaturation. Similar results were reported after prolonged US, leading to increased aggregation and thermal stability of whey-protein [43]. The DSC results obtained for the thermal transition behavior of RPSPI can be used to tailor its applications and functionality.
Impact of US, MN, and combined treatments on functional properties of RPSPI
3.2
Solubility and turbidity
3.2.1
The untreated RPSPI showed solubility of 49.4 ± 0.39 % and turbidity of 0.88 ± 0.01, indicating the presence of large aggregates and poor dispersion. While MN moderately improved solubility to 59.8 ± 0.58 % and reduced turbidity to 0.75 ± 0.01. MN mechanically disrupts aggregates via turbulence and intense shear, improving molecular solvation and flexibility, reducing the particle size, ultimately enhancing protein solubility and decreasing turbidity. It can also expose polar/charged residues, producing strong electrostatic repulsion due to the high zeta potential (−ve charge), which prevents reaggregation and provides a stable solution [28]. Our results align with those of Thakur, Sharma, Khatkar and Sharma [10], who reported a significant increase in solubility and a decrease in turbidity after high-shear homogenization of sorghum and pearl millet proteins.
In contrast, the US alone significantly increased (p < 0.05) the solubility of RPSPI to 66.8 ± 0.47 %, and decreased turbidity to 0.64 ± 0.01, as shown in Fig. 6A. US-induced acoustic cavitation produces microjets and localized shear that break non-covalent (electrostatic and hydrogen) bonds, which hold protein aggregates closely. It leads to exposure of hydrophilic residues, partial unfolding, and disaggregation of larger aggregates into smaller soluble units. Earlier, Gao, Rao and Chen [44] reported a significant increase in solubility through cavitation-induced disaggregation of pea protein isolate to 50 %. Xu, You, Kashenye, Zheng, Li, Zhang and Yang [4] reported that a significant decrease in turbidity after US treatment of sunflower meal protein isolate was associated with the breakdown of protein–protein interactions, which convert the large aggregates into small molecules, ultimately forming a less turbid isolate. Moreover, the increase in solubility and reduction in turbidity after US treatment are due to increased particle quantities after particle size reduction, as evident in Fig. 1B.
The synergetic impact of the combined (MN → US) treatment resulted in a significant increase in solubility (76.9 ± 0.72 %) and a significant decrease in turbidity to 0.51 ± 0.01. PS results showed a significant reduction (Fig. 1B), with the soluble fraction improving by 55 % compared with the untreated sample. Firstly, MN partially unfolded proteins and decreased PS, enhancing accessibility of hydrophobic regions and enabling more uniform penetration of US energy. Then, during the US stage, cavitation produces localized shear gradients and microjets that expose hydrophilic groups and disrupt residual aggregates, thereby enhancing protein-water interactions. The noted reduction in particle size and turbidity confirms this disaggregation. Zhao, Huang, McClements, Liu, Wang and Liu [40] found that combined homogenization and US increased pea protein solubility and decreased turbidity**.** Fig. 6A shows that prolonged combined treatment resulted in reduced solubility and increased turbidity. The mechanical fragmentation from MN can expose reactive thiols and hydrophobic regions. Then, US cavitation, radical formation, and local heat promote covalent cross-linking, hydrophobic re-association, disulfide reshuffling, insoluble aggregates, and re-aggregation, which scatter light, leading to higher turbidity and lower solubility.
Water holding capacity (WHC) and oil holding capacity (OHC)
3.2.2
MN and US significantly (p < 0.05) modified the oil- and water-holding properties of RPSPI via structural changes that altered surface accessibility of non-polar and polar residues (Fig. 6B). Individual MN improved the WHC and OHC of the RPSPI compared with the untreated sample. MN intense mechanical shear unfolds macromolecular structures, breaks down protein aggregates, and exposes hydrophilic groups to build H-bonds with surrounding water [9]. Moreover, reduced PS enhances surface area and generates several open, porous networks that can entangle more water [10]. Similar structural changes also improved OHC by exposing hydrophobic regions and increasing the flexibility of unfolded protein chains, facilitating potent physical entrapment and lipid interaction [45]. Our results align with Thakur, Sharma, Khatkar and Sharma [10], who reported a significant increase in OHC of sorghum and pearl millet protein, and Magalhães, Maurílio, de Souza, Tribst and Leite Júnior [45] reported a significant rise in WHC and OHC of pea protein after high shear homogenization.
Individual US treatments significantly (p < 0.05) increased the OHC and WHC of RPSPI up to 5.76 ± 0.07 g/g and 1.94 ± 0.03 g/g, respectively, compared to MN and the untreated sample. US cavitation generates microturbulence, transient shear, and local heating, thereby exposing polar (−NH_2_, −OH, −COOH) groups and unfolding buried globular regions. Forming H-bonds with surrounding H_2_O increases the WHC. Concurrently, partial unfolding of the protein produces a more flexible molecular matrix that retains water via entanglement and capillary forces [4]. In the case of OHC, cavitation and mechanical forces of US break non-covalent interactions, such as hydrophobic interactions and H-bonds, thereby exposing the non-polar amino acids (isoleucine, phenylalanine, and leucine) on the surface. These hydrophobic components improve the protein's affinity for oil via physical restraint in surface cavities and van der Waals interactions, resulting in better OHC. Moreover, US-generated cavitation produces a flexible tertiary structure that increases internal gaps and surface area, which can immobilize or absorb oil droplets within the protein moieties [5]. Our results align with Xu, You, Kashenye, Zheng, Li, Zhang and Yang [4], and Badjona, Bradshaw, Millman, Howarth and Dubey [5], who reported a significant increase in WHC of pea protein isolate and OHC faba bean protein isolate after US treatment.
Combined (MN → US) treatments induced a highly significant (p < 0.05) increase in WHC (2.78 ± 0.04 g/g) and OHC (8.22 ± 0.07 g/g) compared to individual treatments and untreated samples. The increase is initially due to a synergetic effect that reduces PS, protein unfolding, and the exposure of hydrophobic and hydrophilic residues. Prolonged treatments intensified shear forces and cavitation, facilitating hydrophobic aggregation, disulfide bond formation, and protein–protein interactions, leading to buried active binding sites, decreased surface flexibility, and a compacted protein network, ultimately in reduced OHC and WHC. Earlier, Shi, Liu, Hu, Gao, Qayum, Bilawal, Munkh-Amgalan, Jiang and Hou [13] reported a similar trend in whey protein isolate following combined homogenization and US treatment.
Foaming capacity (FC) and foaming stability (FS)
3.2.3
MN, US, and combined treatments induced a significant increase (p < 0.05) in foaming capacity and stability (%) of RPSPI compared to the untreated sample, as shown in Fig. 7A. The most significant increase in foaming properties was observed in the combined treatment (MN6 + US10), while the combined treatment at US 15 min showed a modest decrease in foam formation and its stability. Moll, Salminen, Griesshaber, Schmitt and Weiss [46] reported that MN improved foaming properties by breaking down loose aggregates, decreasing particle size, and increasing surface-active protein residues that quickly adsorb to newly generated air–water interfaces during whipping. Reduce partial unfolding and particle size induced by strong shear, unmask hydrophobic residues, and enhance interfacial activity, which promotes air incorporation and better foam volumes. Earlier, Thakur, Sharma, Khatkar and Sharma [10] reported a significant increase in foaming properties after MN in sorghum and pearl millet proteins.
During US treatment, an increase in foaming properties occurs due to cavitation-induced microstreaming, shock waves, and transient high shear, which expose hydrophobic residues, partially unfold proteins, and form small soluble aggregates, ultimately increasing interfacial elasticity [3]. Progressively, these changes improve the interface adsorptive flux and increase the adsorbed films’ viscoelasticity, enhancing foamability and bubble coalescence resistance. Our FTIR results support these changes in foaming properties, decreases in α-helix content facilitate the rearrangement of proteins at interfaces and the formation of stable foam films [47]. Earlier, Manzoor, Waseem, Diana, Wang, Ahmed, Ahmed, Ali and An-Zeng [3] reported that US-treatment increased the foaming properties of pepper seed protein isolate by inducing cavitation, aggregate disruption and protein unfolding. These interactions enhanced interfacial activity and surface hydrophobicity, enabling rapid adsorption and generating a stable viscoelastic film around air bubbles. Gulzar, Martín-Belloso and Soliva-Fortuny [15] reported a significant increase in the foaming properties of faba bean protein isolate after US treatments.
The highest increase during combined treatment by collective effect, as MN scatter and partially de-aggregates and enhances exposed surface area, then US unfolds proteins and induces finely soluble aggregates that produce an elastic interfacial network [12]. Decrease in foaming properties during prolonged combined treatment, due to intense treatment, which causes excessive hydrophobic exposure and irreversible aggregation, decreasing the mobility and solubility of proteins, and reducing rapid adsorption and film formation [47]. Also, the formation of insoluble large aggregates minimizes the ability to induce an elastic, cohesive interfacial layer, which is detrimental to foam stability.
Emulsifying activity index (EAI) and emulsifying stability index (ESI)
3.2.4
The RPSPI EAI and ESI after MN, US, and combined treatment were significantly increased (p < 0.05) compared to the untreated sample, as shown in Fig. 7B. The most significant increase was observed in the combined treatment (MN6 + US10), while a slight reduction was observed at US 15 min combined with MN. Mechanistically, MN increased EAI and ESI by partially unfolding the tertiary structure, reducing particle size, and improving H_0_ and surface charge. These protein structural modifications accelerate diffusion to newly generated water–oil interfaces and sustain interfacial films, forming uniformly coated small droplets and better EAI/ESI [12]. Earlier, Thakur, Sharma, Khatkar and Sharma [10] also reported a significant increase in the emulsion capacity and stability of sorghum protein after MN treatment.
During the US, an implosive bubble burst induces intense transient high temperatures, microjets, and local shear, exposing buried hydrophobic residues, unfolding globular domains, and decreasing particle size [3]. The more partially unfolded and flexible protein molecules rapidly adsorb at the water–oil interface, forming an interfacial viscoelastic layer that prevents coalescence, and thereby improves long-term stability and interfacial area (m^2^/g) [48]. Kang, Zhang, Guo, Lei and Yang [49] reported that US-induced cavitation enhanced the emulsification properties of chickpea protein isolate by inducing partial protein unfolding. It increased the molecular flexibility and surface hydrophobicity, ultimately facilitating oil–water interface adsorption and stabilizing the emulsion system. Our results align with those of Gulzar, Martín-Belloso and Soliva-Fortuny [15], who reported a significant increase in EAI and ESI of faba bean protein isolate after US treatment. A combined model of MN and US showed better results, because initial droplet breakage and efficient macro-dispersion provided by MN, then US effectively alters protein conformation and refines droplet size to increase interfacial activity. A slight decline in EAI and ESI in combined treatment is associated with over-processing-induced ordered self-assembly and aggregation. Intense processing facilitates the formation of high-order aggregates, such as amyloid-like structures and irreversible intermolecular β-stacking, which increase particle size, isolate surface-active monomers, and reduce the amount of protein available to adsorb interfaces [50].
Impact of US, MN, and combined treatments on antioxidant abilities of RPSPI
3.3
The antioxidant activities (ABTS, DPPH, and ∙OH scavenging) of RPSPI after MN, US, and combined treatment are presented in Table 2. The untreated sample exhibited ABTS (31.34 ± 0.28 %), DPPH (20.23 ± 0.23 %), and ∙OH scavenging ability (24.04 ± 0.24 %). The MN and US significantly increased the antioxidant properties, while the most significant increase (p < 0.05) was observed in combined treatment (MN6 + US10) of ABTS up to 41.56 ± 0.23 %, DPPH up to 32.18 ± 0.37 %, and ∙OH scavenging ability (33.01 ± 0.27 %). Mechanistically, the MN disrupts protein aggregates, partially unfolds them, decreases particle size, and exposes buried amino acid residues (Tyr, Trp, Cys) that can donate hydrogen atoms or electrons to neutralize radicals. These exposed hydrophobic regions also increase the rate of radical kinetics. Thus, MN does not induce excessive peptides or fragmentation release; the increase in antioxidant abilities remains moderate. Earlier, Zhao, Zhang, Li and Dong [51] reported an increase in the antioxidant activity of soy protein isolate following high-pressure homogenization.
During US treatment, microstreaming and cavitation induce partial fragmentation, depolymerization of aggregates, conformational changes in proteins, and enhance the exposure and surface area of bioactive moieties. This way increases mass transfer, contact between reactive sites and radicals, and may produce a small peptide fragment with strong antioxidant ability [3]. Our results align with Hu and Li [6], who reported that US pretreatment significantly increased the ABTS and DPPH scavenging abilities of soy protein isolate nanofibrils, due to improved exposure of Trp residues and enhanced crystallinity, enabling better radical electron donation.
The combined treatment (MN → US) sequence showed the highest antioxidant-scavenging abilities. Mechanistically, MN first induced size reduction and disruption of aggregates, followed by US-induced disruption of residual aggregates and increased exposure of active sites. The literature also reported that the combined effect of MN with US significantly improves the functional properties and solubility of the protein isolate [52]. When MN was combined with US (15 min), a slight decline in antioxidant abilities was observed. Mechanistically, prolonged treatment may oxidize amino acid side chains (e.g., thiol or phenolic groups) or form insoluble aggregates, unfolding reactive fragments into less available forms, and thereby decreasing the number of available radical-scavenging sites. Zhao, Zhang, Li and Dong [51] reported a decrease in DPPH activity of soy protein isolate when combined homogenization and enzymatic hydrolysis were prolonged, due to a reduction in reactive sites resulting from irreversible degradation of the protein structure.
Impact of US, MN, and combined treatments on IVPD of RPSPI
3.4
The IVPD values in Table 2 indicated a slight increase in individual MN and US as compared to the Untreated sample. In contrast, the combined treatment (MN3 + US10) showed a significant (p < 0.05) enhancement compared to all other treatments. The increase in digestibility was due to mechanical breakdown from MN and protein unfolding from US-induced cavitation, which significantly modifies protein structure, thereby facilitating enzyme access. The combined effect of integrated treatment induced extensive alterations in tertiary and secondary structures, as evidenced by other findings, including increased α-helix and random coil content, enhanced functional properties, and reduced particle size. An increase in random coil and α-helix content, along with tertiary structural changes, indicates a partially unfolded, more flexible conformation. Random coil enhancement indicates exposure of buried peptide bonds and loosening of the structure, thereby enabling enzyme attack. At the same time, the α-helix increased in the treated sample due to the conversion of rigid β-sheet into helices, which are more susceptible and loosely packed to enzymatic unfolding. The outcomes are also backed by the lower ΔH values, which indicate lower structural stability. Altogether, these structural changes are aligned with the observed improvement in IVPD. Earlier, an increase in IVPD was reported in sorghum and pear millet after MN [10] and chickpea protein isolate [49]. Combined treatment of MN, especially with a US 15 min treatment, slightly decreased IVPD. It is more likely associated with intermolecular interactions and protein aggregation that hide enzyme-accessible regions. Moreover, secondary structures, partial refolding, or excessive unfolding may create more stable sites, thereby reducing exposure to enzymatic hydrolysis. Earlier, Yildiz [53] reported that the combined application of high-pressure processing and US, along with pH-shifting, increased protein digestibility. The combined effects reduced particle size, disrupted quaternary structure, and modified secondary structure. These interplays expose cleavage sites and molecular flexibility to digestive enzymes, reducing the steric interference linked with native protein aggregates.
Mechanistic explanation: sonication combined with micronization to structural and functional modification
3.5
The combined impact of MN and US effectively modified the RPSPI through complementary physical mechanisms, as shown in Fig. 8. Firstly, MN disrupts the ordered crystalline structure, reduces particle size, increases surface area and exposes buried functional groups. This MN pre-conditioning disrupts intermolecular non-covalent interactions such as hydrophobic aggregation and H-bonding, making proteins susceptible to succeeding physical treatments. Subsequently, the US operates at the molecular level via acoustic cavitation, characterized by the generation, growth and collapse of microbubbles. US-induced cavitation generates localized microjets, shear forces, transiently elevated pressure–temperature gradients and intense micro-mixing, further promoting protein unfolding. These interplays facilitate controlled conformational rearrangements and partial dissociation of aggregates without substantial denaturation, preserving primary structure while altering tertiary and secondary structures. In combination, MN increases the effectiveness of US through dispersion uniformity, particle hydration, and US energy transmission. The fine protein particle, pre-disrupted by MN, responds efficiently to acoustic cavitation, exposes reactive function groups and increases molecular flexibility, such as charged groups and free −SH groups, thereby facilitating increased protein-oil and −water interactions. The combined processing consequently improves solubility, gelation behavior, foaming capacity, antioxidant properties and in-vitro digestibility. Overall, MN-US integration offers a practical physical approach to modifying plant protein structure–function relationships while sustaining protein sustainability and quality.Fig. 8. Mechanistic pathway of the combined impact of MN and US on structural modification of RPSPI to improve functional properties.
Conclusion
4
In the present study, we applied the MN, US, and their combined treatments to RPSPI, studying the impact on structural modifications, to improve in-vitro digestibility, functional, and antioxidant properties. The results showed that combined treatment (MN + US) on RPSPI produced significant modification in its tertiary structure and conformation. Mechanistically, MN (surface abrasion, mechanical milling, and impact) and cavitation-induced mechanical forces (shock waves and microjets) act in combination to decrease particle size and induce heat and high local shear. It generates loosening of the tertiary structure and partial unfolding, exposing −SH (free thiol) groups and hydrophobic regions. Furthermore, it increased surface area, changed the −ve surface charge, and exposed newly formed groups, accelerating adsorption at oil–water and air–water interfaces, and improving emulsifying and foaming properties. Better dispersion and smaller particles also reduce turbidity and enhance solubility, while reassembly and unfolding under shear result in local ordering and a recorded slight increase in crystallinity. These results clarify the multifaceted role of non-thermal technologies (MS & US) and provide a theoretical framework for converting an extremely underexploited by-product protein (RPSPI) into a value-added ingredient by improving its in vitro digestibility, functional and antioxidant properties. Although the integrated MN and US processing showed better results, the current study was limited to in vitro evaluation and laboratory-scale processing. Future studies should focus on treatment optimization, allergenicity assessment, in vivo digestibility, access scalability under industrial processing and performance analysis of modified RPSPI in real food systems.
CRediT authorship contribution statement
Muhammad Faisal Manzoor: Writing – original draft, Software, Investigation, Formal analysis, Data curation, Conceptualization. Muhammad Waseem: Writing – original draft, Software. Tazeddinova Diana: Investigation, Data curation. Noman Walayat: Investigation, Data curation. Tosheva Zilolakhon Abduvalievna: Writing – review & editing, Data curation. Rana Muhammad Aadil: Writing – review & editing, Data curation. Shahzad Hussain: Writing – review & editing, Data curation. Zahoor Ahmed: Writing – review & editing, Data curation. Murtaza Ali: Writing – review & editing, Data curation. Xin-An Zeng: Visualization, Supervision, Funding acquisition. Abderrahmane Aït-Kaddour: Writing – review & editing, Visualization, Funding acquisition.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Rout S.Gupta R.K.Sharma S.Srivastav P.P.Recent trends in the modification of plant proteins by non-thermal techniques: impact on functional properties and their application in food systems Eur. Food Res. Technol.2025122
- 2Dobón-Suárez A.Zapata P.J.García-Pastor M.E.A comprehensive review on characterization of pepper seeds: unveiling potential value and sustainable agrifood applications Foods 14202519694050949710.3390/foods 14111969 PMC 12154291 · doi ↗ · pubmed ↗
- 3Manzoor M.F.Waseem M.Diana T.Wang R.Ahmed Z.Ahmed I.A.M.Ali M.An-Zeng X.Ultrasound-assisted modification to improve the red pepper seed protein isolate structural, functional, and antioxidant properties Int. J. Biol. Macromol.309202514315410.1016/j.ijbiomac.2025.14315440233908 · doi ↗ · pubmed ↗
- 4Xu G.You W.Kashenye B.N.Zheng H.Li R.Zhang Q.Yang Y.Ultrasound treatment on commercial pea protein isolates systems: effect on structure, rheology and gelling properties Food Chem.464202514190810.1016/j.foodchem.2024.14190839520888 · doi ↗ · pubmed ↗
- 5Badjona A.Bradshaw R.Millman C.Howarth M.Dubey B.Optimization of ultrasound-assisted extraction of faba bean protein isolate: structural, functional, and thermal properties. Part 2/2Ultrason. Sonochem.110202410703010.1016/j.ultsonch.2024.107030 PMC 1137825039153419 · doi ↗ · pubmed ↗
- 6Hu A.Li L.Effects of ultrasound pretreatment on functional property, antioxidant activity, and digestibility of soy protein isolate nanofibrils Ultrason. Sonochem.90202210619310.1016/j.ultsonch.2022.106193 PMC 957904536257213 · doi ↗ · pubmed ↗
- 7Gu S.Li J.Zhang Y.Xia J.Guo C.Hu Z.Jin W.Shen W.Micronization of defatted Rice bran dietary Fiber and its particle size effect on the structure and physicochemical properties of Myofibrillar proteins Food Chem.202514654510.1016/j.foodchem.2025.14654541038048 · doi ↗ · pubmed ↗
- 8Li C.Mc Clements D.J.Dai T.Deng L.Feng Z.Li T.Liu C.Chen J.Enhancing the dispersibility of commercial pea protein ingredients using stirred media milling: potential mechanisms of action Food Hydrocoll.1452023109130