Ultrasonic and Glycation-Modified Soy Protein Isolate Delivery System Enhances the Antioxidant Activity of Antrodia cinnamomea Triterpenoids

Abstract

Antrodia cinnamomea is a rare medicinal and edible macrofungus, and its triterpenoids (ACT, A. cinnamomea triterpenoids) exhibit notable hepatoprotective, antioxidant, anticancer, and immunomodulatory activities. However, their poor aqueous solubility and low dispersibility in aqueous media have limited their practical applications. In this study, the conditions for ultrasonic treatment and xylo-oligosaccharide (XOS)-mediated glycation for soy protein isolate (SPI) were optimized; ACT was then encapsulated into the modified SPI carrier to prepare XOS-SPI-ACT nanoparticles. The delivery system was systematically characterized in terms of encapsulation efficiency (74.22 ± 2.15)%, drug-loading capacity (71.19 ± 4.67)%, storage stability, thermal stability, Fourier transform infrared (FTIR) spectroscopy, UV fluorescence spectroscopy, circular dichroism (CD) spectroscopy, and surface morphological features. The results showed that ACT was effectively embedded in XOS-SPI to form a stable complex with excellent thermal stability and favorable storage stability over a 28-day period. The in vitro antioxidant activities of XOS-SPI-ACT, XOS-SPI, and free ACT were comparatively evaluated. XOS-SPI-ACT exhibited significantly higher scavenging capacities against DPPH radicals, ABTS radicals, hydroxyl radicals, and superoxide anions, as well as higher FRAP values (94%, 74%, 75%, 68%, and 2 mmol/g), compared with free ACT (48%, 17%, 21%, 32%, and 1 mmol/g). Furthermore, XOS-SPI-ACT effectively inhibited lipid peroxidation in the β-carotene/linoleic acid oxidation model, with an overall antioxidant performance of 72%, markedly higher than the 20% of free ACT. This study effectively improves the aqueous solubility and dispersibility of ACT, expands their application potential, and provides a foundation for developing ACT-based natural antioxidants and functional foods.

1. Introduction

Antrodia cinnamomea, a rare medicinal and edible fungus of the phylum Basidiomycota and genus Antrodia, has been reported to exhibit hepatoprotective, anticancer, anti-inflammatory, antioxidant, antiviral, antihypertensive, immunomodulatory, and gut microbiota-modulating activities [1]. Triterpenoids are the predominant bioactive components in the fruiting bodies of A. cinnamomea. Lanostane- and ergostane-type triterpenoids are the most abundant subtypes and are considered the main contributors to its characteristic intense bitterness [2]. A. cinnamomea triterpenoids (ACT) also show a wide range of biological activities. In oncology, Antcin C induces tumor cell apoptosis by inhibiting the PI3K/AKT and MAPK signaling pathways, and by downregulating matrix metalloproteinase (MMP)-2 and MMP-9 to suppress invasion and metastasis [3]. In hepatoprotective research, Antcin A alleviates carbon tetrachloride-induced hepatic injury in mice by reducing the levels of pro-inflammatory cytokines and restoring superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) activities through attenuation of the MAPK/NF-κB signaling cascade [4]. Immunomodulatory studies further show that ACT bidirectionally repolarizes pro-inflammatory M1 macrophages toward the reparative M2 phenotype, balances inflammatory responses, and enhances natural killer (NK) cell cytotoxicity, thereby improving antiviral immunity [5].

Despite these activities, the poor aqueous solubility of ACT results in low dispersibility in physiological environments, which restricts its efficacy. Current delivery strategies for fungal triterpenoids mainly include cyclodextrin inclusion complexes, liposomal formulations, and polymeric nanoparticles [6,7,8]. However, these strategies are often associated with several drawbacks, including complex preparation procedures, the employment of synthetic polymers with potential toxicological risks, poor food compatibility, high production costs, and relatively low drug-loading capacities [7,9]. Nanocarrier technologies can improve the solubility, stability, and targeting ability of natural active ingredients through physical encapsulation and covalent conjugation. Soy protein isolate (SPI) is widely used as a carrier matrix due to its biocompatibility and ease of modification [10]. Nevertheless, native SPI typically exhibits a limited loading capacity for hydrophobic bioactive compounds, and thus requires further structural modification to improve its delivery efficiency [11]. Ultrasonic treatment, characterized by high efficiency, controllability, and simple operation, is widely used to regulate the emulsifying, structural, and gelation properties of soy proteins [12]. Notably, although previous studies have applied ultrasonic treatment to soy protein modification, its application in SPI-based delivery systems for triterpenoid encapsulation remains largely underexplored. Glycation is also an effective strategy for improving protein functionality. Through the Maillard reaction, reducing sugars covalently cross-link with protein chains to generate protein–saccharide conjugates, enabling broader applications of SPI in the food industry [13]. However, the combined application of ultrasonic treatment and glycation for improving the delivery efficiency of hydrophobic triterpenoids via SPI-based carriers was few reported. Therefore, the present study was designed to develop a novel SPI delivery system modified by ultrasonic treatment and XOS-mediated glycation, aiming to improve the aqueous solubility, storage stability, and in vitro antioxidant activity of ACT.

Oxidative stress in food systems is triggered by the excessive accumulation of reactive oxygen species (ROS) and free radicals, which subsequently induces lipid peroxidation, protein oxidation, and overall quality deterioration of food products. Such oxidative damage not only reduces the nutritional value and induces off-flavors in food products but also shortens their shelf life [14]. Natural antioxidants, such as triterpenoids, can scavenge free radicals and inhibit lipid peroxidation, thus maintaining the physicochemical stability of food products and prolonging their shelf life [15]. Although ACT has been shown to possess considerable in vitro antioxidant activity [16], its poor aqueous solubility and high environmental sensitivity limit its direct application in food processing and preservation systems. Therefore, the aim of this study was to develop an ultrasonication and glycation-modified SPI delivery system for the encapsulation of ACT. The primary goals were to enhance the antioxidant activity of ACT and broaden its potential applications in the development of functional foods.

2. Materials and Methods

2.1. Materials

The fruiting bodies of Antrodia cinnamomea (type strain ATCC 200183, naturally cultivated on Cinnamomum kanehirae wood blocks for one year) were purchased from Fujian Haotian Biotechnology Co., Ltd. (Fujian, China). Soy protein isolate (SPI, protein content ≥ 90%) was purchased from Beijing Dingguo Biotechnology Co., Ltd. (Beijing, China). Sodium dodecyl sulfate (SDS), o-phthaldialdehyde (OPA), and β-mercaptoethanol were purchased from Sigma-Aldrich (St. Louis, MO, USA). Xylo-oligosaccharides (XOS, degree of polymerization 2–7, reducing sugar content ≥ 95%), AB-8 macroporous, boric acid, vanillin, glacial acetic acid, perchloric acid, oleanolic acid, absolute ethanol, potassium dihydrogen phosphate (KH₂PO₄), and dipotassium hydrogen phosphate (K₂HPO₄) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Simulated saliva, pepsin (from porcine gastric mucosa), and trypsin (from bovine pancreas) were purchased from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China). All chemical reagents used in this study were of analytical grade and used without further purification.

2.2. Extraction and Purification of ACT

The fruiting bodies of A. cinnamomea were dried at 45 °C, pulverized, and subjected to reflux extraction with anhydrous ethanol at a solid-to-liquid ratio of 1:50 (g dry powder per mL solvent) for 4 h (5 g of dried powder extracted with 250 mL of anhydrous ethanol). The extract was concentrated to approximately one-fifth of its original volume and dried in a constant-temperature oven at 75 °C to obtain the crude ACT extract. The crude ACT extract was dispersed in deionized water at a solid-to-liquid ratio of 1:10 (w/v; ~3 g of crude ACT dispersed in 30 mL of deionized water). The aqueous dispersion was mixed with petroleum ether at a volume ratio of 1:1 (v/v) in a sealed conical flask, and the mixture was magnetically stirred at 200 rpm in a 65 °C water bath for 20 min to minimize solvent evaporation. After stirring, the mixture was allowed to stand for 10 min to achieve complete phase separation, and the upper petroleum ether layer was discarded. This defatting procedure was repeated twice. The pH of the defatted aqueous phase was adjusted to 2.5 using 1 mol/L hydrochloric acid (HCl). The AB-8 macroporous adsorption resin was pretreated by soaking in anhydrous ethanol for 24 h, then packed into a glass chromatography column (2.0 cm × 30 cm) and rinsed with distilled water until no ethanol odor was detectedable. The pretreated resin column was loaded with the defatted sample solution at a volume of 0.5 column volumes (CV). After static adsorption for 5 h to reach adsorption equilibrium, the column was sequentially rinsed with 3 CV of ultrapure water and 3 CV of 20% (v/v) ethanol to eliminate non-specifically adsorbed impurities. Finally, the target ACT compounds were eluted with 6 CV of anhydrous ethanol at a flow rate of 1 mL/min. The eluate was concentrated to approximately one-fifth of its original volume under reduced pressure and dried at 75 °C to yield purified ACT. The total triterpenoid content in the purified ACT was determined via colorimetry using a vanillin-glacial acetic acid-perchloric acid system with oleanolic acid (Sigma, St. Louis, MO, USA, CAS: 508-02-1, ≥98%, HPLC grade) as the standard reference substance. The samples were reacted with the vanillin-glacial acetic acid-perchloric acid reagent at 60 °C for 45 min, and the absorbance of the resulting mixture was measured at 550 nm using an Evolution 200 UV-Vis spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) [17]. Additionally, SPI and XOS exhibited no significant absorbance at a wavelength of 550 nm, thus excluding their interference with the colorimetric assay.

2.3. Optimization of Modification Conditions for SPI

2.3.1. Optimization of the Ultrasonic Modification Conditions

Ultrasonic treatment was applied to a 20 mg/mL SPI solution using an ultrasonic processor (Sonics VCX750, Sonics & Materials, Inc., Newtown, CT, USA) operating at 20 kHz with a 13 mm probe; the treatment conditions were 50 mL of SPI solution in a 100 mL glass beaker, 15 mm probe immersion depth, and an ice-water bath for temperature control. To optimize the ultrasonic power, the SPI solution was adjusted to pH 7.0 and subjected to ultrasonic treatment at 500, 600, 700, 800, and 900 W for 10 min under a pulsed mode (3 s on, 3 s off). After each ultrasonic treatment, the solubility of SPI was determined directly at the corresponding initial pH (no additional pH adjustment was performed). To optimize the ultrasonic duration, the SPI solution (pH 7.0) was ultrasonically treated at a fixed power of 700 W for 5, 10, 15, 20, and 25 min, respectively. After each treatment, the solubility of SPI was determined directly at the corresponding initial pH (no additional pH adjustment was performed). To optimize the ultrasonic pH, the pH of the SPI solution was adjusted to 7.0, 8.0, 9.0, 10.0, or 11.0, then each pH-adjusted solution was ultrasonically treated at 700 W for 10 min. After each treatment, the solubility of SPI was determined directly at the corresponding initial pH (no additional pH adjustment was performed). Finally, based on the solubility values of SPI, the optimal ultrasonic power, treatment duration, and pH value for SPI modification were determined [18].

2.3.2. Optimization of Glycation Modification Conditions

Ultrasonically modified SPI (Ult-SPI) was mixed with XOS at mass ratios of 1:1, 1:2, 1:3, 1:4, and 1:5 (Ult-SPI:XOS), and the mixture was dissolved in 10 mmol/L phosphate buffer (pH 7.4). The resulting solution was incubated with constant stirring in a sealed glass vessel in a 90 °C water bath for 2 h, followed by freeze-drying. Ult-SPI and XOS were then mixed at a fixed mass ratio of 1:2, and the mixture was dissolved in 10 mmol/L phosphate buffer (pH 7.4). The solution was incubated with constant stirring in a sealed glass vessel in water baths at 50, 60, 70, 80, and 90 °C for 2 h, respectively, followed by freeze-drying. Finally, Ult-SPI and XOS were mixed at a mass ratio of 1:2, and the mixture was dissolved in 10 mmol/L phosphate buffer (pH 7.4). The solution was incubated with constant stirring in a sealed glass vessel in a 90 °C water bath for 30, 60, 90, 120, and 150 min, respectively. Subsequently, the resulting glycation reaction product was dialyzed against ultrapure water using a 1 kDa molecular weight cut-off dialysis membrane to remove unreacted XOS, and then freeze-dried. After each glycation treatment, the degree of grafting (DG) and browning intensity (BI) of the reaction products were measured to identify the optimal Ult-SPI:XOS mass ratio, glycation temperature, and reaction duration [18].

2.3.3. Determination Methods for Modified SPI Properties

(1)

Solubility

The pH of a 10 mg/mL SPI solution was adjusted to 8.0 using 0.5 mol/L NaOH and equilibrated for 30 min in a magnetically stirred water bath at 25 °C. The mixture was centrifuged at 5000× g for 15 min at 4 °C, and the protein concentration in the supernatant was measured using a BCA protein assay kit. The solubility was calculated using Equation (1):

$$\mathrm{S}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y} (\%)={\displaystyle \frac{{\mathrm{C}}_1}{{\mathrm{C}}_0}}\times 100\%$$

(1)

where C₁ is the protein concentration determined in the supernatant, and C₀ is the initial SPI concentration.

(2)

Degree of Grafting (DG)

The free amino groups were quantified using the o-phthaldialdehyde (OPA) method. The OPA working reagent was prepared by dissolving 40 mg of OPA in 1 mL of methanol, followed by the sequential addition of 2.5 mL of 20% (w/v) SDS solution, 25 mL of 0.1 mol/L boric acid buffer, and 100 μL of β-mercaptoethanol. The final volume of the reagent was adjusted to 50 mL with ultrapure water, and the reagent was stored in the dark at 4 °C until use. A calibration curve was constructed using L-lysine as the standard reference substance at concentrations ranging from 0.1 to 1.0 mmol/L. For each sample, two measurements were performed: (1) sample reaction system (200 μL sample solution + 4 mL OPA working reagent), and (2) sample blank system (200 μL sample solution + 4 mL PBS, pH 7.4). All reaction mixtures were vortexed thoroughly, incubated at 35 °C for 2 min, and the absorbance of each mixture was measured at 340 nm using a UV-Vis spectrophotometer. The net absorbance (A_net) for each sample was calculated using Equation (2):

A_net = A_reaction − A_blank

(2)

The free amino group content of protein samples was calculated from the L-lysine calibration curve and expressed as millimoles lysine equivalent per gram of protein (mmol Lys eq/g). The degree of grafting (DG) was calculated using Equation (3):

$$\mathrm{D}\mathrm{G} (\%)={\displaystyle \frac{{\mathrm{A}}_0-{\mathrm{A}}_1}{{\mathrm{A}}_0}}\times 100\%$$

(3)

where A₀ is the free amino group content of unmodified SPI or Ult-SPI, and A₁ is free amino group content of glycated XOS-SPI conjugates.

(3)

Browning Intensity

According to Sun et al. [19], 2 mL of XOS-SPI solution was mixed with 1 mL of 20% (w/w) SDS and 1 mL of 0.1 mol/L sodium borate, vortexed, and the absorbance at 420 nm was measured.

(4)

Extrinsic Fluorescence Spectroscopy

As described by Philippe et al. [20], 4 mL of SPI or XOS-SPI solution (1 mg/mL, 5 mmol/L phosphate buffer, pH 7.0) was mixed with 20 μL of 8 mmol/L ANS. Fluorescence emission spectra (350–550 nm) were recorded using an excitation wavelength of 347 nm and excitation/emission slit widths of 5 nm.

(5)

Surface Hydrophobicity Index

According to Chelh et al. [21], 4 mL of SPI or XOS-SPI at concentrations of 0.002, 0.004, 0.006, 0.008, and 0.01 mg/mL (10 mmol/L phosphate buffer, pH 7.0) was mixed with 20 μL of 8 mmol/L ANS. After vortexing, the fluorescence intensity was measured at an excitation wavelength of 390 nm and an emission wavelength of 470 nm, with slit widths of 5 nm. The initial slope of the fluorescence intensity versus protein concentration plot was defined as the surface hydrophobicity index (H₀).

2.4. Preparation of XOS-SPI-ACT Nanoparticles

2.4.1. Optimization of the Mass Ratio of ACT to XOS-SPI

A 20 mg/mL SPI solution (pH 10) was ultrasonically treated at 700 W for 10 min, followed by glycation with XOS at an Ult-SPI:XOS mass ratio of 1:2 at 90 °C for 2 h to obtain XOS-SPI. Ethanol-dissolved ACT and XOS-SPI were then mixed at mass ratios of 1:1, 2:1, 1:2, 1:4, or 1:8 and magnetically stirred at room temperature for 3 h, followed by overnight incubation at 4 °C; a macroscopic gel was formed during the incubation period. The resulting gel was then mechanically homogenized at 11,000 rpm for 3 min using a high-speed homogenizer, followed by further dispersion via ultrasonication (600 W, 3 s on/3 s off pulsed mode, 15 min) in an ice-water bath to maintain the sample temperature below 25 °C (to prevent protein denaturation), thus fabricating XOS-SPI-ACT nanoparticles. The nanoparticles were freeze-dried and stored at 4 °C until use.

2.4.2. Determination of Encapsulation Efficiency and Drug Load Capacity

XOS-SPI-ACT nanoparticles solutions with varying ACT contents were centrifuged at 12,000 rpm (rotor radius of 8.8 cm) and 4 °C for 15 min, and 500 μL of the resulting supernatant was collected for subsequent analysis. To eliminate potential interference from ethanol and the sample matrix, a matrix-matched blank control was prepared by using XOS-SPI conjugate solution with the same ethanol concentration as that in the sample supernatant. The mass of free ACT in each sample was calculated from the standard curve described in Section 2.1 and recorded as M_free. The encapsulation efficiency (EE) and drug-loading capacity (LC) were calculated as follows [22]:

$$\mathrm{E}\mathrm{n}\mathrm{c}\mathrm{a}\mathrm{p}\mathrm{s}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e} (\mathrm{E}\mathrm{E},\%)={\displaystyle \frac{{\mathrm{M}}_{\_\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}-{\mathrm{M}}_{\_\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{e}}}{{\mathrm{M}}_{\_\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}}}\times 100\%$$

(4)

$$\mathrm{D}\mathrm{r}\mathrm{u}\mathrm{g}\mathrm{-}\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{c}\mathrm{a}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{y} (\mathrm{L}\mathrm{C},\%)={\displaystyle \frac{{\mathrm{M}}_{\_\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}-{\mathrm{M}}_{\_\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{e}}}{{{\mathrm{M}}_{\_\mathrm{X}\mathrm{O}\mathrm{S}-\mathrm{S}\mathrm{P}\mathrm{I}}+\mathrm{M}}_{\_\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}-{\mathrm{M}}_{\_\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{e}}}}\times 100\%$$

(5)

where M_total is the total mass of added ACT (mg), M_free is the mass of free ACT in the supernatant (mg), and M_XOS-SPI is the mass of XOS-SPI added to the sample (mg).

2.4.3. Determination of Particle Size, PDI, and Zeta Potential

According to Chen et al. [23], the particle size, polydispersity index (PDI), and zeta potential of nanoparticles with different ACT mass ratios were determined using an ES90 Nano Malvern particle analyzer (Malvern Instruments, Malvern, UK) at a sample concentration of 1 mg/mL (dissolved in 5 mmol/L phosphate-buffered saline (PBS), pH 7.0). The instrument parameters were set as follows: refractive index of 1.45, absorption coefficient of 0.001, viscosity of 0.8872 cp, and temperature of 25.0 ± 0.2 °C.

2.5. Characterization of XOS-SPI-ACT Nanoparticles

2.5.1. Fourier Transform Infrared Analysis

As described by Liu et al. [24], an appropriate amount of the sample powder (SPI, XOS-SPI or XOS-SPI-ACT) was analyzed by Fourier transform infrared (FTIR) spectroscopy using a Nicolet iS50 FTIR spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). The parameters were as follows: temperature 25 °C, scanning range 500–4000 cm⁻¹, resolution 4 cm⁻¹, wavenumber accuracy 0.01 cm⁻¹, and 32 scans.

2.5.2. UV-Vis Absorption Spectroscopy

According to Zhang et al. [25], 1 mL of SPI, XOS-SPI, or XOS-SPI-ACT solution was diluted to 1 mg/mL with ultrapure water, centrifuged at 8000× g and 4 °C for 15 min, and 4 mL of the resulting supernatant was transferred to a 1 cm quartz cuvette. UV-Vis absorption spectra were recorded in the wavelength range of 190 to 400 nm using a UV-1900i UV-Vis spectrophotometer (Shimadzu Corporation, Kyoto, Japan). Ultrapure water was used as the blank control, and the transmittance was calibrated to 100% at 280 nm prior to measurement.

2.5.3. Circular Dichroism Spectroscopy Analysis

Following the method of Fu et al. [26], 1 mL of a 0.1 g/L SPI, XOS-SPI, or XOS-SPI-ACT solution (10 mmol/L PBS, pH 7.0) was analyzed using a J-810 circular dichroism spectrometer (JASCO Corporation, Tokyo, Japan). The operating parameters were as follows: cuvette path length 0.1 cm, scanning speed 100 nm/min, response time 0.125 s, and bandwidth 1 nm. The mean residual ellipticity (θ) is expressed in deg·cm²·dmol⁻¹.

2.5.4. Scanning Electron Microscope Analysis

The freeze-dried sample (SPI, XOS-SPI, or XOS-SPI-ACT) was sputter-coated with a thin gold layer using a Leica ACE600 high-vacuum sputter coater (Leica Microsystems, Wetzlar, Germany). The samples were then observed using a Gemini SEM 300 field-emission scanning electron microscope (Carl Zeiss, Oberkochen, Germany) at an accelerating voltage of 20 kV.

2.5.5. Differential Scanning Calorimeter Analysis

According to Zeng et al. [27], the thermal stability of the XOS-SPI-ACT nanoparticles was assessed using a DSC 8500 differential scanning calorimeter (PerkinElmer, Waltham, MA, USA). An empty aluminum pan was sealed and equilibrated at 20 °C. A 2.5 mg aliquot of the sample powder was placed into another aluminum pan, compacted, and sealed. An empty sealed pan served as the blank control. Scanning was conducted from 20 to 200 °C at a rate of 10 °C/min. Thermograms were analyzed to determine the peak temperature (Td).

2.6. Determination of In Vitro Antioxidant Activity

Stock solutions of SPI, XOS-SPI and XOS-SPI-ACT were prepared in 10 mmol/L phosphate buffer (pH 7.4) at concentrations ranging from 10 to 40 mg/mL (i.e., 10, 15, 20, 25, 30, 35, and 40 mg/mL). Free ACT was dissolved in anhydrous ethanol at the same concentration range to prepare stock solutions. The corresponding phosphate buffer or anhydrous ethanol was used as the blank control, while vitamin C (VC) solutions at the corresponding concentrations were used as the positive control. All experiments were performed in triplicates.

2.6.1. DPPH Radical-Scavenging Activity

The assay was performed as described by Sako et al. [28]. A total of 2.0 mL of the sample solution was mixed with 2.0 mL of 0.2 mmol/L 1,1-diphenyl-2-picrylhydrazyl (DPPH) solution. After incubation in the dark for 30 min, the absorbance was measured at 517 nm. The DPPH scavenging capacity was calculated using Equation (6):

$$\mathrm{D}\mathrm{P}\mathrm{P}\mathrm{H} \mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{-}\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{c}\mathrm{a}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{y} (\%)=(1-{\displaystyle \frac{{\mathrm{A}}_1-{\mathrm{A}}_2}{{\mathrm{A}}_0}})\times 100\%$$

(6)

where A₀ is the absorbance of the blank control (DPPH solution); A₁ is the absorbance of the sample mixture (sample solution + DPPH solution); A₂ is the absorbance of the sample blank (sample solution), which is used to correct for interference from sample color and turbidity.

2.6.2. ABTS Radical-Scavenging Activity

The ABTS working solution was prepared as described by Sotek et al. [29]. A volume of 0.1 mL of the sample solution was mixed with 3.9 mL of the ABTS working solution, incubated in the dark for 5 min, and the absorbance was recorded at 734 nm. The scavenging capacity was calculated using Equation (7):

$$\mathrm{A}\mathrm{B}\mathrm{T}\mathrm{S} \mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{-}\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{c}\mathrm{a}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{y}(\%)=(1-{\displaystyle \frac{{\mathrm{A}}_1}{{\mathrm{A}}_0}})\times 100\%$$

(7)

where A₀ and A₁ are the absorbances of the blank control and sample, respectively.

2.6.3. Hydroxyl Radical-Scavenging Activity

The hydroxyl radical-scavenging assay was performed as described by Eremia et al. [30] with minor modifications. Hydroxyl radicals (·OH) were generated by mixing 0.9 mL of 1 mmol/L FeSO₄ solution (dissolved in acetate buffer, pH 3.5) and 0.9 mL of 1 mmol/L H₂O₂ solution. Then, 200 μL of the sample solution was added to the reaction system, and the mixture was incubated in the dark at room temperature for 1 h to facilitate hydroxyl radical scavenging by the sample. Subsequently, 2 mL of 5 mmol/L salicylic acid ethanol solution was added to capture the residual hydroxyl radicals, and the mixture was incubated at 37 °C for 10 min to induce color development. The color reaction was terminated by cooling the mixture in an ice-water bath for 5 min. The absorbance was measured at 510 nm using ultrapure water as the reference. The blank control contained acetate buffer instead of the sample solution. The scavenging capacity was calculated using Equation (8):

$$· \mathrm{O}\mathrm{H} \mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{-}\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{c}\mathrm{a}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{y} (\%)={\displaystyle \frac{{\mathrm{A}}_{\_\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}}-{\mathrm{A}}_{\_\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}}}{{\mathrm{A}}_{\_\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}}}}\times 100\%$$

(8)

where A_blank and A_sample are the absorbances of the blank control and the sample, respectively.

2.6.4. Superoxide Anion-Scavenging Capacity

The superoxide anion-scavenging assay was conducted as described by Wang et al. [31]. A total of 6 mL of Tris-HCl buffer (pH 8.2) was mixed with 0.5 mL of the sample solution and pre-incubated at 37 °C for 10 min. A 7 mmol/L pyrogallol solution was freshly prepared in 10 mmol/L HCl to prevent autoxidation immediately prior to the assay. After adding 1 mL of the pre-warmed (37 °C) pyrogallol solution, the mixture was incubated at room temperature for 4 min, and the reaction was terminated with 500 μL of concentrated HCl. The absorbance was measured at 320 nm, with a blank control prepared using Tris-HCl buffer. The scavenging capacity was calculated using Equation (9):

$$\mathrm{S}\mathrm{u}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{x}\mathrm{i}\mathrm{d}\mathrm{e} \mathrm{a}\mathrm{n}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{-}\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{c}\mathrm{a}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{y} (\%)={\displaystyle \frac{{\mathrm{A}}_{\_\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}-{\mathrm{A}}_{\_\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}}}{{\mathrm{A}}_{\_\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}-{\mathrm{A}}_{\_\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}}}}\times 100\%$$

(9)

where A_sample, A_control, and A_blank are the absorbances of the sample mixture, positive control, and blank control, respectively.

2.6.5. β-Carotene Bleaching Inhibitory Activity

The β-carotene bleaching inhibitory assay was performed as described by Alam et al. [32]. Briefly, 5 mg of β-carotene was dissolved in 10 mL of chloroform, followed by the sequential addition of 250 μL of linoleic acid and 3 mL of Tween-20 (emulsifier). After thorough vortexing, the chloroform was removed under reduced pressure at 40 °C, and 250 mL of oxygen-saturated ultrapure water was added to form an emulsion. A 5 mL aliquot of the emulsion was mixed with 1 mL of the sample solution and incubated at 50 °C for 2 h. Absorbance at 470 nm was measured every 20 min. A blank control was prepared by replacing the sample solution with an equal volume of anhydrous ethanol to match the solvent composition of the samples. The β-carotene bleaching rate (Rate) and inhibition rate of lipid peroxidation were calculated using Equations (10) and (11), respectively:

$$\mathrm{R}\mathrm{a}\mathrm{t}\mathrm{e}={\displaystyle \frac{{\mathrm{A}}_0-{\mathrm{A}}_{\mathrm{t}}}{\mathrm{t}}}$$

(10)

$$\mathrm{I}\mathrm{n}\mathrm{h}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} (\%)=(1-{\displaystyle \frac{{\mathrm{R}\mathrm{a}\mathrm{t}\mathrm{e}}_{\_\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}}}{{\mathrm{R}\mathrm{a}\mathrm{t}\mathrm{e}}_{\_\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}}}})\times 100\%$$

(11)

where A₀ is the initial absorbance of the mixture at 0 min, and A_t is the absorbance of the mixture at time t.

2.6.6. Ferric-Reducing Antioxidant Power

The ferric-reducing antioxidant power (FRAP) assay was performed as described by Morales et al. [33]. The FRAP working reagent was freshly prepared by mixing 300 mmol/L acetate buffer (pH 3.6), 10 mmol/L TPTZ solution (dissolved in 40 mmol/L HCl), and 20 mmol/L FeCl₃·6H₂O solution at a volume ratio of 10:1:1 (v/v/v). The mixture was incubated at 37 °C for 5 min before use. In a 96-well microplate, 100 μL of FRAP working reagent was mixed with 20 μL of the sample solution, and the total volume was adjusted to 200 μL with methanol. After incubation at room temperature for 5 min, the absorbance of each well was measured at 593 nm using an Infinite F50 microplate reader (TECAN, Männedorf, Switzerland). A calibration curve was constructed using FeSO₄ standard solutions (0–1 mmol/L), and the results were expressed as mmol Fe²⁺ equivalents per gram of sample.

2.7. Statistical Analysis

All experimental data are presented as mean ± standard deviation (SD) derived from three independent experiments, with each experiment conducted in three technical replicates using independently prepared nanoparticle batches. Graphs were plotted using Origin 2021 software. Statistical analyses were performed using SPSS 26.0 software. Two-group comparisons were analyzed using Student’s t-test, multiple comparisons were performed via one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test, and Pearson’s correlation analysis was used to evaluate linear relationships. Statistical significance was set at p < 0.05.

3. Results and Discussion

3.1. Ultrasonic Treatment and Xylooligosaccharide Glycation-Modified SPI

3.1.1. Ultrasonic Modification of SPI

As shown in Figure 1A, the solubility of SPI increased progressively with increasing ultrasonic power, reaching a maximum of 82.9% at 700 W, and then declined at higher power levels. This bell-shaped trend reflects the formation and collapse of microbubbles during sonication, which generates localized shear forces and shock waves. These effects disrupt the intermolecular interactions within SPI and enhance particle dispersion [34]. A similar pattern was observed for the treatment time (Figure 1B): the solubility peaked at 98.6% after 10 min of sonication but decreased with longer exposure. Short-term sonication promotes the partial unfolding of SPI molecules, exposing the buried hydrophilic side chains and improving solvent compatibility. However, prolonged cavitation can lead to random aggregation or fragmentation of polypeptide chains, thereby reducing solubility [35].

Figure 1. Effects of ultrasonic power (A), duration (B), and pH (C) on the solubility of SPI. Note: SPI, soybean protein isolate; Different lowercase letters above the columns indicate significant differences (p < 0.05) by one-way ANOVA and Tukey’s post hoc test. Data are mean ± SD (n = 3).

The solubility of SPI also presented a bell-shaped response to pH, gradually increasing up to pH 10 and then decreasing under more alkaline conditions (Figure 1C). The solubility reached 99.1% at pH 10, while stronger alkaline conditions disrupted the molecular integrity of SPI and promoted unfolding or random cross-linking of protein chains, forming insoluble aggregates [36]. Based on the results shown in Figure 1A–C, the optimal ultrasonic modification conditions were identified as 700 W power, 10 min treatment duration, and pH 10. Under these conditions, ultrasonicated SPI (Ult-SPI) achieved maximum solubility.

3.1.2. XOS Glycation of Ult-SPI

During the Maillard reaction, both the DG and BI increased with the XOS mass ratio (Figure 2A). The increase in DG is attributed to more frequent collisions between the free amino groups on Ult-SPI and carbonyl groups at the reducing ends of XOS, increasing the likelihood of covalent interactions [37]. However, excessive browning is undesirable because it imparts unfavorable color and sensory properties to the final product [38]. Thus, optimization requires balancing the grafting efficiency with controlled browning.

Figure 2. Effects of Ult-SPI/XOS mass ratio (A), reaction temperature (B), and reaction duration (C) on degree of grafting and browning intensity. Note: SPI, soybean protein isolate; Ult-SPI, the SPI modified by ultrasonication; XOS, xylo-oligosaccharides. Different lowercase letters above the columns indicate significant differences (p < 0.05) by one-way ANOVA and Tukey’s post hoc test.

Glycation efficiency was also significantly affected by the reaction temperature and duration (Figure 2B,C). DG increased steadily with increasing temperature, whereas BI peaked at 80 °C and decreased at 90 °C (Figure 2B). The reduction in BI at 90 °C may be related to temperature-induced aggregation of Ult-SPI, which could lower the number of free amino groups available for reaction and limit Maillard progression [39]. Nevertheless, this proposed mechanism needs further verification via analytical techniques such as particle size analysis. Increasing the reaction duration from 0 to 120 min significantly increased DG, after which it plateaued (Figure 2C). The BI showed a similar trend during the first 90 min; however, at 120 min, the BI decreased relative to that at 90 min and 150 min. This temporary decrease was attributed to the formation of stable Ult-SPI–XOS conjugates, which consume reactive carbonyl groups and reduce pigment formation [40,41].

Based on these results, the optimal glycation conditions were determined to be an Ult-SPI/XOS mass ratio of 1:2, reaction temperature of 90 °C, and reaction duration time of 120 min.

3.2. Surface Hydrophobicity of XOS-SPI

As shown in Figure 3A, native SPI showed no characteristic emission peak in the wavelength range of 400–600 nm, whereas XOS-SPI exhibited a strong fluorescence emission peak at 460 nm. This fluorescence signal indicates an enhanced binding of ANS to the hydrophobic regions of the protein, suggesting that the combined modification process induced the exposure of hydrophobic groups previously embedded in the SPI protein matrix. Consistently, XOS-SPI showed significantly higher surface hydrophobicity than native SPI (Figure 3B), further confirming the exposure of hydrophobic groups previously embedded in the native protein matrix [42]. Together, these results provide direct evidence of successful XOS conjugation and the accompanying structural rearrangement of SPI.

Figure 3. Exogenous fluorescence spectra (A) and surface hydrophobicity indices (B) of SPI and XOS-SPI. Note: SPI, soybean protein isolate; XOS, xylo-oligosaccharides; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with XOS. Different lowercase letters above the columns indicate significant differences (p < 0.05) by one-way ANOVA and Tukey’s post hoc test.

3.3. Preparation of XOS-SPI-ACT Micelles

3.3.1. Effect of ACT-to-XOS-SPI Mass Ratio on EE

The mass ratio of ACT to XOS-SPI significantly affected the EE of the nanoparticles. As shown in Table 1, the mass ratio of 1:2 (w/w) resulted in the highest EE of (74.22 ± 2.15) %, along with an LC of (26.74 ± 0.30) %. Increasing the concentration of XOS-SPI within a suitable range improved ACT solubilization and binding efficiency. However, excessive XOS-SPI led to stronger intermolecular cross-linking, which hindered ACT entrapment and reduced EE [43]. Therefore, an ACT-to-XOS-SPI mass ratio of 1:2 (w/w) was selected for the subsequent experiments.

Table 1. Effect of ACT-to-XOS-SPI mass ratio on encapsulation efficiency.

Mass Ratio (ACT: XOS-SPI)	Encapsulation Efficiency (%)
2:1	48.87 ± 2.66 d
1:1	54.51 ± 3.72 c
1:2	74.22 ± 2.15 a
1:4	64.11 ± 3.7 b
1:8	32.81 ± 2.48 e

Note: ACT, Antrodia cinnamomea triterpenoids; SPI, soybean protein isolate; XOS, xylo-oligosaccharides; XOS-SPI, the SPI modified via ultrasonication and glycosylation with XOS; Different lowercase letters indicate significant differences (p < 0.05) between groups by one-way ANOVA and Tukey’s post hoc test; Data are mean ± SD (n = 3).

3.3.2. Effect of ACT-to-XOS-SPI Mass Ratio on Storage Stability

Storage stability is a major determinant of the practical applicability of protein micelles in food systems. Stable protein micelles must withstand processing conditions and maintain their integrity during long-term storage and transportation [44]. The effect of the ACT-to-XOS-SPI mass ratio on micellar stability was assessed by monitoring the particle size, polydispersity index (PDI), and zeta potential after 7 and 28 days of storage at 4 °C. Smaller particle sizes reduce sedimentation tendencies; lower PDIs indicate better dispersibility; and higher absolute zeta potentials reflect stronger electrostatic stabilization [45].

As shown in Figure 4A, ultrasonic-assisted glycation reduced the hydrodynamic diameter of XOS-SPI to 194.63 ± 1.11 nm, compared with 200.83 ± 0.75 nm for native SPI. This reduction is attributed to ultrasonic cavitation and intermolecular repulsion generated during glycation, both of which limit droplet coalescence [46]. At an ACT-to-XOS-SPI mass ratio of 1:2, the nanoparticles displayed the lowest PDI and the largest absolute zeta potential, indicating the highest colloidal stability of the nanoparticles. The negative surface charge generated a strong electrostatic repulsion, preventing aggregation and maintaining uniform dispersion [47].

Figure 4. Particle size (A), PDI (B), and zeta potential (C) of samples stored for 0, 7, and 28 days. Note: PDI, polydispersity index; SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T1 to X-S-T5 represent nanoparticles with ACT:XOS-SPI mass ratios of 2:1, 1:1, 1:2, 1:4, and 1:8, respectively. Different lowercase letters above columns in the same group indicate significant differences (p < 0.05) between samples by one-way ANOVA and Tukey’s post hoc test. Data are mean ± SD (n = 3).

After 7 days of storage at 4 °C, no significant changes in particle size and PDI were observed for any formulation (Figure 4B), which demonstrated favorable short-term storage stability of the nanoparticles. After 28 days of storage, XOS-SPI-ACT nanoparticles prepared at a mass ratio of 1:2 exhibited only a 2.51% increase in particle size, a 2.76% increase in PDI, and a 26.82% decrease in zeta potential from −12.31 ± 0.66 mV to −8.99 ± 0.61 mV (Figure 4C). These changes were minimal among all the tested ratios, which indicated the superior long-term storage stability. This favorable storage stability was attributed to electrosteric stabilization, whereby the conformational rearrangement of XOS-SPI molecules exposes hydrophobic residues and enhances the hydrophobic interactions between XOS-SPI and ACT, while the hydrophilic XOS moiety forms a steric hindrance layer on the particle surface. The resulting interfacial network slows the structural relaxation during storage [48]. Thus, considering EE, LC, and micellar stability, the optimal ACT-to-XOS-SPI mass ratio was determined to be 1:2.

3.4. Structural Characterization of SPI, XOS-SPI and XOS-SPI-ACT Nanoparticles

3.4.1. FTIR and UV-Vis Absorption Spectroscopy

As shown in the FTIR spectra (Figure 5A), the absorption intensity of XOS-SPI in the 3700–3200 cm⁻¹ region increased markedly compared with that of native SPI, indicating glycation-induced structural changes in the protein [49]. The enhanced O–H stretching vibration near 3200 cm⁻¹ may be attributed to the introduction of hydroxyl groups from XOS, as well as alterations in the intramolecular hydrogen bonding networks of SPI. In the fingerprint region (800–1500 cm⁻¹), XOS-SPI-ACT displayed characteristic band shifts and variations in relative absorption intensities compared to native SPI and XOS-SPI. The spectral variations in the 800–1260 cm⁻¹ range were consistent with the incorporation of carbohydrate moieties from XOS, which confirmed the successful modification of SPI by XOS glycation.

Figure 5. FTIR (A) and UV-Vis absorption (B) spectra of different samples. Note: FTIR, Fourier transform infrared; UV-Vis, Ultraviolet–Visible; SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T1 to X-S-T5 represent nanoparticles with ACT:XOS-SPI mass ratios of 2:1, 1:1, 1:2, 1:4, and 1:8, respectively; X-S-T = XOS-SPI-ACT, representing nanoparticles with ACT:XOS-SPI mass ratios of 1:2.

ACT exhibited a characteristic band at 2930 cm⁻¹ corresponding to saturated C–H stretching [50] and a strong peak at 1678 cm⁻¹ associated with α,β-unsaturated carbonyl groups [51]. In the FTIR spectrum of XOS-SPI-ACT, the 3000–3600 cm⁻¹ region was dominated by a broad O–H stretching band centered at approximately 3200 cm⁻¹, whereas the ACT-specific 2930 cm⁻¹ band was barely visible. This observation may indicate that ACT was successfully embedded within the XOS-SPI matrix, although additional evidence (e.g., in vitro release studies or direct microscopic imaging) is required to confirm its precise localization, as the exact spatial distribution and encapsulation degree cannot be elucidated solely by FTIR data.

The UV-Vis absorption spectra (Figure 5B) showed an obvious spectral shift for XOS-SPI compared with native SPI. A new absorption maximum at 270–280 nm was observed in XOS-SPI, which may be related to the formation of early Maillard reaction products such as Schiff-base (-RC=N-) [52], although the contribution from protein conformational changes and microenvironmental alterations of aromatic amino acids cannot be excluded. The increased absorbance reflects a higher density of UV-active glycation products and greater molar absorptivity [53]. After ACT encapsulation, the absorbance of XOS-SPI-ACT decreased with a blue shift, indicating structural reorganization of the conjugate and a more polar microenvironment around the remaining chromophores.

3.4.2. Circular Dichroism Spectroscopy of XOS-SPI-ACT Nanoparticles

The Cotton effects in the CD spectra reflect the differences in the secondary structure and overall molecular conformation of proteins [54]. In the far-UV region (190–250 nm, Figure 6A), native SPI displayed a negative Cotton band at 210 nm, which is a characteristic feature of α-helical-rich protein structures. This negative Cotton band red-shifted to 217 nm for XOS-SPI, which indicated glycation-induced conformational changes in the SPI molecular structure. For XOS-SPI-ACT, the characteristic negative molar ellipticity was significantly diminished and lacked a distinct minimum, suggesting a further disruption of the SPI secondary structure upon ACT loading. In the near-UV region (250–400 nm, Figure 6A), distinct spectral variations were also observed, which reflected alterations in the microenvironment of aromatic amino acid residues in SPI. Collectively, these results demonstrate that both XOS glycation and ACT encapsulation induced significant structural rearrangements in SPI molecules, which confirmed the stable interaction between ACT and the XOS-SPI carrier matrix.

Figure 6. Circular dichroic diagram (A) and secondary structure (B) of different samples. Note: SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T = XOS-SPI-ACT, representing nanoparticles with ACT:XOS-SPI mass ratios of 1:2.

Quantitative analysis of the SPI secondary structure (Figure 6B) showed that α-helix stability depends on uninterrupted intramolecular hydrogen bonding and hydrophobic interactions [55]. Ultrasonic treatment reduced the α-helix content of SPI via cavitation and shear effects, with a portion of the lost α-helix structure being converted into β-turns and random coils [56]. Following XOS glycation, the carbohydrate chains from XOS coated the protein surface and reduced the direct effect of shear forces, thereby mitigating the disruption of the α-helix structure. β-sheets, which rely heavily on interstrand hydrogen bonding, were more sensitive to shear forces, leading to a decrease in β-sheet content after ultrasonic treatment, with a portion of the lost β-sheet structure converted into random coils.

3.4.3. Scanning Electron Microscopy and Thermal Stability of XOS-SPI-ACT Nanoparticles

SEM micrographs of the freeze-dried samples were acquired to characterize the morphological changes in SPI-based materials after modification (Figure 7A). At magnifications of 100×, 500×, and 1000×, native SPI presented loosely aggregated, porous microspheres with an irregular particle size distribution. After ultrasonic treatment and glycation, the original spherical morphology was disrupted, producing sheet-like fragments owing to shear forces and cavitation. Following ACT encapsulation, XOS-SPI-ACT exhibited a rough, reticulated surface, indicating stable interactions between ACT and the XOS-SPI matrix and the corresponding restructuring of the protein network. It should be noted that these SEM micrographs reflect the morphology of freeze-dried samples, which may differ from the morphological state of the nanoparticles in aqueous dispersion.

Figure 7. SEM micrograph (A) and DSC thermogram (B) of SPI, X-S, and X-S-T. Note: SEM, scanning electron microscopy; DSC, differential scanning calorimetry; SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T = XOS-SPI-ACT, represent nanoparticles with ACT:XOS-SPI mass ratios of 1:2.

DSC provides insights into protein thermal transitions, conformational changes, aggregation behavior, and overall thermal stability [57,58]. The denaturation peak temperature (Td) is positively correlated with the strength of intramolecular interactions; the higher the Td value, the greater the resistance to denaturation [57]. As shown in Figure 7B, native SPI exhibited a major endothermic transition with a denaturation Td of approximately 90 °C, which corresponds to the thermal denaturation of SPI proteins, primarily the glycinin (11S) globulin fraction. XOS-SPI exhibited higher heat flow values over the entire temperature range in comparison with native SPI, with its major endothermic transition occurring at a higher temperature (Td = 102.78 °C), which indicated that ultrasonic-assisted XOS glycation enhanced the thermal stability of SPI. This enhancement is attributed to the increased β-sheet content, as the ordered, hydrogen-bonded β-strands strengthen intramolecular interactions and raise the energy barrier for unfolding [58]. After ACT encapsulation, the Td of XOS-SPI-ACT decreased slightly to 98.78 °C but remained substantially higher than that of native SPI, indicating that the nanoparticles maintained strong thermal stability.

3.5. In Vitro Antioxidant Capacity of XOS-SPI-ACT Nanoparticles

3.5.1. DPPH and ABTS Radical-Scavenging Capacity

As shown in Figure 8A, the DPPH radical-scavenging capacity of the modified SPI (XOS-SPI) was significantly higher than that of the native SPI, and XOS-SPI-ACT exhibited substantially greater activity than both free ACT and XOS-SPI. This suggests that the sugar chain structure of XOS improves ACT stability or provides additional active sites, thereby enhancing the radical-scavenging efficiency [59]. Concentration-dependent analysis (Figure 8B) revealed that the DPPH radical-scavenging rate of XOS-SPI-ACT reached 94% at 25 mg/mL, which was significantly higher than that of free ACT at the same concentration. This result demonstrates that XOS-SPI encapsulation effectively enhanced the in vitro antioxidant activity of ACT under the present experimental conditions.

Figure 8. The DPPH and ABTS free radical scavenging ability of different samples. Note: (A) DPPH radical scavenging capacities of SPI, XOS-SPI, XOS-SPI-ACT, and ACT. XOS-SPI-ACT was tested at 25 mg/mL, while the other samples were adjusted to equivalent concentrations, and VC was used at 25 mg/mL; (B) DPPH radical scavenging capacities of XOS-SPI-ACT at various concentrations; (C) ABTS radical scavenging capacities of SPI, XOS-SPI, XOS-SPI-ACT, and ACT. XOS-SPI-ACT was tested at 20 mg/mL, while the other samples were adjusted to equivalent concentrations, and VC was used at 20 mg/mL; (D) ABTS radical scavenging capacities of XOS-SPI-ACT at various concentrations; DPPH, 1,1-Diphenyl-2-picrylhydrazyl; ABTS, 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid); VC, vitamin C; SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T = XOS-SPI-ACT, represent nanoparticles with ACT:XOS-SPI mass ratios of 1:2. Different lowercase letters above the columns indicate significant differences (p < 0.05) by one-way ANOVA and Tukey’s post hoc test.

The ABTS assay yielded similar results. XOS-SPI-ACT exhibited rapid and strong ABTS radical-scavenging capacity, markedly higher than that of both XOS-SPI and free ACT (Figure 8C). The concentration-response curve (Figure 8D) indicates that XOS-SPI-ACT achieved more than 70% inhibition at 10 mg/mL, and the scavenging effect approached saturation at higher concentrations, reaching 74% at 20 mg/mL. This high efficiency suggests that XOS-SPI-ACT may be suitable for applications that require rapid antioxidant action during critical stages of food processing. However, it is worth noting that some natural mushrooms species, such as Daedalea quercina, have been reported to exhibit ABTS radical-scavenging activity with an IC₅₀ of only 389.20 μg/mL [60], indicating that the scavenging efficiency of the present ACT delivery system can be further improved.

3.5.2. Hydroxyl Radical and Superoxide Anion Scavenging Capacity

The hydroxyl radicals (·OH) are among the most reactive oxygen species; therefore, their scavenging efficiency is a key indicator of antioxidant performance [61]. As shown in Figure 9A, SPI and XOS-SPI alone displayed moderate ·OH scavenging rates (50.06% and 60.26%, respectively), whereas free ACT performed poorly (22.02%). After encapsulation, XOS-SPI-ACT achieved a significantly higher scavenging rate of 75.72%, indicating a synergistic effect between the protein matrix and ACT. Concentration-dependent curves (Figure 9B) showed that XOS-SPI-ACT reached approximately 65% scavenging at 10 mg/mL and increased to 75% at 25 mg/mL, demonstrating strong ·OH neutralization even at low concentrations. This potent ·OH radical-scavenging activity of XOS-SPI-ACT highlights its potential applications in the development of natural antioxidant formulations. Above 30 mg/mL, the scavenging rate plateaued, likely reflecting the equilibrium between radical generation and quenching kinetics [62].

Figure 9. The hydroxyl radical and superoxide anion scavenging ability of different samples. Note: (A) Hydroxyl radical scavenging capacities of SPI, XOS-SPI, XOS-SPI-ACT, and ACT. XOS-SPI-ACT was tested at 30 mg/mL, while the other samples were adjusted to equivalent concentrations, and VC was used at 30 mg/mL; (B) Hydroxyl radical scavenging capacities of XOS-SPI-ACT at various concentrations; (C) Superoxide anion scavenging capacities of SPI, XOS-SPI, XOS-SPI-ACT, and ACT. XOS-SPI-ACT was tested at 25 mg/mL, while the other samples were adjusted to equivalent concentrations, and VC was used at 25 mg/mL; (D) Superoxide anion scavenging capacities of XOS-SPI-ACT at various concentrations; ·OH, hydroxyl radicals; O₂⁻, superoxide anion; VC, vitamin C; SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T = XOS-SPI-ACT, represent nanoparticles with ACT:XOS-SPI mass ratios of 1:2. Different lowercase letters above the columns indicate significant differences (p < 0.05) by one-way ANOVA and Tukey’s post hoc test.

The superoxide anion (O₂⁻) scavenging rate of XOS-SPI-ACT was also significantly higher than that of XOS-SPI and free ACT (Figure 9C), again demonstrating synergistic enhancement. Concentration-dependent analysis (Figure 9D) revealed a smaller increase, from 65.51% at 10 mg/mL to 69.21% at 25 mg/mL, indicating a relatively concentration-insensitive response to superoxide anions. Despite this modest increase, the consistently high scavenging activity highlights the potential of nanoparticles in antioxidant applications.

3.5.3. β-Carotene Bleaching Inhibition

In the β-carotene/linoleic acid system, XOS-SPI-ACT achieved more than 70% inhibition of β-carotene bleaching (Table 2), which was substantially higher than the approximately 20% achieved by free ACT. This suggests that encapsulation improves the dispersibility and interfacial activity of ACT in lipid-rich environments. Concentration-dependent analysis (Table 2) showed an inhibition rate of 72% at a 20 mg/mL concentration. Although the inhibitory effect gradually declined with time, XOS-SPI-ACT exhibited a slower decay than the other samples, demonstrating a more persistent antioxidant response throughout the experimental period.

Table 2. The β-carotene bleaching inhibition capacity of different samples and concentrations.

Sample		β-Carotene Bleaching Inhibition Ratio (%)
		0 min	30 min	60 min	90 min	120 min
Name	SPI	42.42 ± 1.73 a	40.68 ± 1.52 a	39.96 ± 1.56 ab	37.8 ± 1.56 bc	37.08 ± 0.17 c
	X-S	67.17 ± 0.85 a	65.41 ± 0.71 b	64.43 ± 0.64 b	61.73 ± 0.66 c	58.92 ± 0.68 d
	X-S-T	72.72 ± 3.23 a	70.13 ± 1.02 ab	69.57 ± 1.04 ab	68.47 ± 1.05 b	68.61 ± 1.05 b
	ACT	22.42 ± 0.65 a	21.29 ± 1.89 ab	20.9 ± 1.94 ab	18.19 ± 1.95 bc	15.54 ± 1.95 c
	VC	91.99 ± 2.30 a	91.13 ± 2.45 a	88.36 ± 2.21 ab	85.81 ± 2.24 b	85.14 ± 2.26 b
Concentration of X-S-T (mg/mL)	10	59.55 ± 3.03 a	57.12 ± 0.96 ab	55.01 ± 2.79 b	49.60 ± 2.78 c	46.67 ± 2.43 c
	15	66.56 ± 3.14 a	63.76 ± 0.99 ab	61.58 ± 1.00 bc	59.39 ± 1.01 c	56.12 ± 1.34 d
	20	72.72 ± 3.23 a	70.13 ± 1.02 ab	69.57 ± 1.04 ab	68.47 ± 1.05 b	68.61 ± 1.05 b
	25	80.33 ± 1.24 a	77.93 ± 1.06 b	75.09 ± 1.07 c	73.98 ± 1.07 cd	72.80 ± 1.08 d
	30	82.23 ± 1.27 a	80.38 ± 1.30 ab	78.21 ± 1.32 bc	77.57 ± 1.33 c	75.19 ± 1.33 d

Note: VC, vitamin C; SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T = XOS-SPI-ACT, represent nanoparticles with ACT:XOS-SPI mass ratios of 1:2; X-S-T was determined at 20 mg/mL, while the other samples were adjusted to equivalent concentrations, and VC was used at 20 mg/mL. Different lowercase letters indicate significant differences among different time points (p < 0.05) by one-way ANOVA and Tukey’s post hoc test. Data are presented as mean ± SD (n = 3).

3.5.4. Total FRAP of XOS-SPI-ACT Nanoparticles

The FRAP assay provides an integrated measure of the electron-donating capacity. As shown in Figure 10A, XOS-SPI-ACT exhibited the highest FRAP value among all samples (excluding the VC positive control), significantly exceeding those of free ACT, XOS-SPI, and native SPI. This demonstrates that XOS-SPI-ACT has the strongest overall reducing ability. The concentration-dependent profiles (Figure 10B) showed a steady increase in FRAP values up to 25 mg/mL, followed by a gradual decline at concentrations above this level, yielding a bell-shaped response. This decrease in FRAP values at higher concentrations may be attributed to assay saturation, increased sample turbidity, or inner filter effects, rather than a genuine reduction in the intrinsic antioxidant capacity of XOS-SPI-ACT. Further systematic investigations are required to elucidate the underlying molecular mechanisms responsible for this phenomenon. Nevertheless, a distinct synergistic interaction between XOS-SPI and ACT was evident in the composite system, whereby the XOS-SPI protein matrix enhances the stability and dispersibility of ACT, while the reductive hydroxyl groups of XOS improve the electron-transfer efficiency of the antioxidant system [63]. The dose-responsive antioxidant profile of XOS-SPI-ACT highlights its potential as a customizable antioxidant system suitable for diverse application scenarios.

Figure 10. Total antioxidant capacity of different samples. Note: (A) Total antioxidant capacities of SPI, X-S, X-S-T, and ACT. X-S-T was tested at 25 mg/mL, while the other samples were adjusted to equivalent concentrations, and VC was used at 25 mg/mL; (B) Total antioxidant capacities of X-S-T at various concentrations; FRAP, ferric-reducing antioxidant power; VC, vitamin C; SPI, soybean protein isolate; X-S = XOS-SPI, the SPI modified via ultrasonication and glycosylation with xylo-oligosaccharides; ACT, Antrodia cinnamomea triterpenoids; X-S-T = XOS-SPI-ACT, represent nanoparticles with ACT:XOS-SPI mass ratios of 1:2. Different lowercase letters above the columns indicate significant differences (p < 0.05) by one-way ANOVA and Tukey’s post hoc test.

4. Conclusions

In this study, SPI was modified using a combined strategy of ultrasonication and XOS-mediated glycation, and the resulting XOS-SPI was used to encapsulate ACT at a mass ratio of 1:2. The as-prepared XOS-SPI-ACT nanoparticles exhibited an EE of (74.22 ± 2.15)% and an LC of (26.74 ± 0.30)%, also displaying favorable stability over a 28-day storage period. In vitro antioxidant assays showed that, compared with free ACT, the scavenging capacities of XOS-SPI-ACT against DPPH, ABTS, hydroxyl, and superoxide anion radicals increased by 95.83%, 335.29%, 257.14%, and 112.50%, respectively, whereas its total FRAP increased by 100.00%. This XOS-SPI-based encapsulation strategy successfully entrapped ACT within the nanoparticle matrix, which not only improved the aqueous solubility and dispersibility of ACT but also significantly enhanced its in vitro antioxidant activity, thereby providing a promising platform for the development of stable natural antioxidant formulations.

From a practical perspective, the raw materials employed herein (SPI and XOS) are food-grade, inexpensive, and readily available, thus presenting favorable potential for economic viability. The preparation procedure incorporates ultrasonication and glycation, both of which are scalable techniques with established industrial applications. Nevertheless, further optimization of process parameters (e.g., continuous flow ultrasonication) is still required to enhance production efficiency and lower energy consumption toward industrial-scale production. Regarding practical applicability, XOS-SPI-ACT nanoparticles could potentially be incorporated into diverse functional food products, including beverages, dairy products, and solid formulations, although their stability, sensory properties, and bioavailability in real food systems warrant further investigation.

It should be emphasized that only the in vitro antioxidant activity of XOS-SPI-ACT was evaluated in the present study. Future research should extend to cellular and animal models to validate the in vivo efficacy of these nanoparticles, as well as assess their performance in actual food matrices. Furthermore, the long-term stability and potential toxicity of the nanoparticles need to be systematically investigated prior to commercial application.