Enhancing the preservation of chicken breast using a curdlan/BSA gel interacted with citrus flavonoids: Interaction mode and dual-component protection

Abstract

This study developed bovine serum albumin (BSA)-curdlan (Cn) gels infused with flavonoid extract (Fs) from orange peels to enhance chicken breast's quality. Characterization via fluorescence intensity (FI), FTIR, UV–vis, SEM, and XPS confirmed effective flavonoid integration and stable structure. Among variants, the BSA-Cn-Fs-9 gel showed the highest antioxidant activity (DPPH 40.30% and ABTS 93.72%) and sustained the release rate of Fs extract (72.91%). BSA-Cn-Fs-9 gel showed strong inhibitory effects against bacteria, including S. aureus (12.7 mm), B. cereus (13.4 mm), E. coli (11.4 mm), and L. monocytogenes (16 mm), and against fungi, including P. expansum (35.7%) and A. westerdijkiae fc-1 (57.44%). Applied to chicken, BSA-Cn-Fs-9 gel inhibits microbial growth, maintains a pH of 6.5, and reduces lipid oxidation (TBRAS at 0.29 mg/kg). The findings showed that the synthesized gel could serve as an eco-friendly gel material with strong potential for practical food safety monitoring and for reducing food contamination.

Highlights

• •The gel matrix controlled the release of flavonoid extracts.
• •Improving gel texture properties with flavonoids for improved bioactivity.
• •The synthesized gel showed an inhibitory effect on L. monocytogenes.
• •The synthesized gel showed an inhibitory effect on A. westerdijkiae.
• •The synthesized gel preserved chicken breast quality.

1. Introduction

Food products are susceptible to microbial and fungal spoilage, which affects their shelf life and causes significant economic losses for producers. Furthermore, the use of large quantities of chemical additives in food preservation may also raise health and consumer-acceptance concerns (Mia et al., 2025). Therefore, the use of natural extracts with biological properties that can preserve food without compromising human health has attracted considerable interest in food-preservation research (Anas et al., 2019; Mia et al., 2025). Plant extracts have previously demonstrated their ability to inhibit bacteria and molds, thereby extending food shelf life (Batiha et al., 2020; Khan et al., 2014). Orange peel is a rich source of flavonoids, and orange-peel flavonoid extract (Fs) has been reported to exhibit antibacterial and antifungal activity (Naseer et al., 2026; Saleem et al., 2023). Extracts from orange flavedo and albedo have demonstrated antibacterial and antioxidant effects (Mohammed et al., 2024). Reported mechanisms include increased cytoplasmic-membrane permeability, intracellular leakage, and inhibition of microbial growth (Faleye et al., 2023; Liu et al., 2026). Although Fs is promising for food-processing, packaging, and postharvest applications, its sensitivity to light, oxygen, metal ions, pH, and heat remains a major limitation (Yuan et al., 2024).

Gels are three-dimensional polymer networks formed through physical or chemical cross-linking (Lu et al., 2026) and are used in tissue engineering (Lu et al., 2024), food-packaging applications (Chi et al., 2025; Ding et al., 2025), and the pharmaceutical industry (Xiao et al., 2024). Natural gels derived from polysaccharides and proteins are widely used in functional foods because of their non-toxicity, biocompatibility, biodegradability, and edibility (Hu et al., 2024; Teng et al., 2024; Yu et al., 2026). In this context, curdlan (Cn) is a bioactive polysaccharide composed of approximately 300–500 glucose units linked by glycosidic bonds (Yuan et al., 2021). Cn and its derivatives have been reported to exhibit antifungal, immunomodulatory, antiviral, anticancer, and antibacterial activities (Chen & Wang, 2020). Accordingly, curdlan gels have been developed as delivery matrices for bioactive compounds, including tannic acid (Zhou et al., 2022) and polyphenols (Li, Dong, et al., 2024).

BSA is a 66-kDa protein capable of heat-induced gelation (Shen et al., 2013). Heat treatment can produce an irreversible BSA gel that binds bioactive ingredients within a composite matrix (Hamedi et al., 2022; Shi et al., 2024). However, single-component BSA gels may show limited thermal stability, gel strength, and water retention (Liu et al., 2021). Combining proteins with polysaccharides can strengthen the network through non-covalent interactions with flavonoids and may improve their retention and long-term activity. Altan et al. (2018) revealed that the interaction of carvacrol with zein and polylactic acid enhanced the stability of the conjugate system by forming hydrogen bonds between amino groups in zein and hydroxyl groups in carvacrol. Such gel networks may protect flavonoids from temperature, pH, and oxygen while modulating their release (Araiza-Calahorra et al., 2020; Liu et al., 2021).

Polysaccharide-protein gels loaded with bioactive compounds can serve as delivery matrices, antimicrobial preservatives, and antioxidant systems in foods. In this context, Yan et al. (2018) reported that encapsulated quercetin in chitosan hydrochloride and carboxymethyl chitosan hydrogel showed DPPH radical-scavenging activity of up to 90%, exceeding that of free quercetin. Whey protein isolate-lotus root amylopectin hydrogels were formed to integrate quercetin with enhanced phytochemical stability (Liu et al., 2020). Moreover, Hu et al. (2018) reported that epigallocatechin gallate-loaded lysozyme amyloid fibril hydrogels exhibited antibacterial activity against both Gram-positive and Gram-negative bacteria. In the present study, curdlan and BSA were combined as a dual-component matrix intended to retain Fs through non-covalent interactions while allowing gradual release. Therefore, this study explored the antimicrobial and antioxidant capacities of a bovine serum albumin (BSA)-curdlan (Cn) gel integrating flavonoid extract (Fs) from orange peel. Multispectral techniques, including fluorescence intensity (FI), Fourier transform infrared spectroscopy (FTIR), UV–visible spectroscopy, and X-ray photoelectron spectroscopy (XPS), were used to examine interactions during gel formation. The characterization and stability of the designed gel were investigated using scanning electron microscopy (SEM), X-ray diffraction (XRD), and differential scanning calorimetry (DSC). We hypothesized that the BSA-Cn matrix would support Fs retention and release, thereby combining antioxidant and antimicrobial functions during refrigerated chicken breast storage.

2. Materials and methods

2.1. Materials

Curdlan, 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), 2,2-diphenyl-1-picrylhydrazyl (DPPH), and bovine serum albumin (BSA) were obtained from Macklin Biochemical Co., Ltd. (Shanghai, China). Ethanol, sodium chloride (NaCl), phosphate-buffered saline (PBS), potato dextrose agar (PDA), nutrient agar, trichloroacetic acid, mannitol egg-yolk polymyxin agar, violet red bile agar, ALOA agar, and Baird-Parker agar supplemented with egg-yolk tellurite were of analytical grade. UPLC-MS/MS-grade methanol and acetonitrile were obtained from Sigma-Aldrich (USA), formic acid from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China), and flavonoid analytical standards from Yuanye Bio-Technology Co., Ltd. (Shanghai, China).

2.2. Extraction purification, and determination of flavonoid extract

For the extraction assay, 20 g of orange peel powder was mixed with 200 mL of 80% EtOH at a ratio of 1:10 (w/v). The mixture was stirred at 200–500 rpm for 6 h at room temperature. The extract was filtered and centrifuged at 8000 rpm for 10 min to obtain a clear supernatant. The extract was concentrated using a rotary evaporator until the volume reached 10 mL. The flavonoid extract was purified using an AB-8 resin column and eluted with 80% EtOH according to the method reported by Salah et al. (2025), which was subjected to targeted UPLC-MS/MS analysis of a panel of 46 flavonoids using a Waters Acquity UPLC system coupled to an AB SCIEX 5500 triple-quadrupole mass spectrometer. An Acquity UPLC HSS T3 column (1.8 μm, 2.1 × 100 mm) was maintained at 40 °C. mobile phase A consisted of water containing 0.1% formic acid, whereas mobile phase B consisted of acetonitrile containing 0.1% formic acid. The flow rate was maintained at 0.3 mL/min, the injection volume was 6 μL, and the total run time was 18 min. The gradient program was as follows: 0–1 min, 10% B; 1–14 min, 10–90% B; 14–15 min, 90% B; 15–15.10 min, 90–10% B; and 15.10–18 min, 10% B for column re-equilibration. Mass-spectrometric detection was conducted using an electrospray-ionization source in multiple-reaction-monitoring (MRM) mode. The source conditions were as follows: curtain gas, 35 arbitrary units; collision gas, 9 arbitrary units; ion-spray voltage, 4500 V; source temperature, 450 °C; ion-source gas 1, 45 arbitrary units; and ion-source gas 2, 45 arbitrary units. The chromatographic peaks were integrated using MultiQuant software, and concentrations were calculated using the one-point internal-standard method. Compound assignments were based on the retention times and compound-specific MRM transitions established using the corresponding analytical standards.

2.3. Preparation of BSA-Cn gels loaded with flavonoid extract (Fs)

The heat-set gel was produced using the thermal-gelation method, as described by Hou et al. (2024). BSA (10 g) was dissolved in 100 mL of 1 mM NaCl solution (pH 7). The protein suspension was mixed for 3 h, then kept at 4 °C overnight to ensure complete solubilization. NaCl provided ionic strength by shielding charged groups, limiting aggregation, and promoting BSA hydration. Subsequently, 2 g of Cn (2%) was dissolved into the BSA solution, and the mixture was re-stirred at 1300–1500 rpm for 4 h. The heat treatment induced gelation at 81 °C for 10 min, followed by immediate cooling in an ice bath to 25 °C. The gel was then stored at 4 °C overnight and designated as the BSA-Cn gel. Fs was first dissolved in a small volume of food-grade ethanol and gradually added to the BSA-Cn solution to obtain final Fs concentrations of 5, 7, or 9 mg/mL. The mixture was homogenized at 10,000 ×g for 5 min to form a stable pre-emulsion. The final ethanol concentration did not exceed 5%; the mixture was then stirred at 800 rpm and maintained at 5 °C to promote interaction between Fs and the BSA-Cn matrix. Gelation was then induced using the same heating and cooling procedure. The gels were then stored at 4 °C overnight and designated BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9, respectively.

2.4. Molecular interactions within the gels

2.4.1. UV–visible spectroscopy

To analyze protein structural changes during gelation, samples were first prepared by homogenizing the gel in a phosphate buffer (pH 7.4) at 12,000 ×g for 15 min at 4 °C. UV–visible spectra were recorded using a double-beam spectrophotometer (UV-8000, Metash Instruments, Shanghai, China) over 200–400 nm at room temperature, using a 1-cm quartz cuvette. To minimize the impact of light scattering from residual turbidity, all measurements included a baseline correction using a blank containing only the gel matrix. Additionally, absorbance at 600 nm was monitored for each sample to quantify any remaining turbidity. This approach ensured that the recorded spectra reflected true protein absorbance rather than artifacts from light scattering (Cai et al., 2019).

2.4.2. FTIR spectroscopy

Interactions between Fs and the BSA-Cn gel were examined by FTIR spectroscopy (Ding et al., 2024). Spectra were recorded from 4000 to 1000 cm⁻¹ using an FTIR spectrometer (Nicolet iS50, Thermo Fisher Scientific, Waltham, MA, USA).

2.4.3. Fluorescence spectroscopy

Intrinsic fluorescence was measured using a Cary Eclipse spectrofluorometer (Varian, Palo Alto, CA, USA) equipped with a front-face accessory to minimize excitation- and emission-path distortions caused by sample turbidity (Salah & Xu, 2021). Gel samples were gently dissolved in 6 M urea to release protein fractions, followed by centrifugation at 12,000 ×g for 15 min to obtain a clear supernatant. Aliquots were transferred to a 1-mm-path-length quartz cuvette, and spectra were collected at room temperature with an excitation wavelength of 280 nm and emission spectra recorded from 300 to 500 nm. All fluorescence spectra were normalized against a blank gel matrix treated identically but without protein. This protocol, including sample preparation and instrumental settings, was designed to ensure that the observed fluorescence changes were attributable to protein structural transitions rather than to scattering or polarization artifacts inherent to gel systems. Fluorescence quenching was calculated using Eq. (1), where F₀ is the fluorescence intensity of BSA in the reference sample and F is the fluorescence intensity of the corresponding ligand-containing sample:

$$\mathit{Quenching}\left(\%\right)=\frac{\text{Fluorescence intensity of protein}-\text{Fluorescence intensity of protein ligands}}{\text{Fluorescence intensity of protein}}x100$$ (1)

2.4.4. XPS analysis

The gel samples were characterized by X-ray photoelectron spectroscopy (XPS; ESCALAB 250Xi, Thermo Fisher Scientific, Waltham, MA, USA). The C1s, O1s, and N1s signals were analyzed at their respective binding energies, and the spectra were processed using Thermo Avantage software (Salah et al., 2024).

2.4.5. Docking analysis

Discovery Studio 2.5 (Accelrys Software Inc., San Diego, CA, USA) and the LibDock algorithm were used to model interactions among Cn, BSA, and a representative citrus flavonoid. The 2D-dimensional structure of neohesperidin (PubChem CID: 442439) and kaempferide (PubChem CID: 5281666) was selected as a representative citrus flavanone based on our chromatographic results (Table S1 and Fig. S1). The curdlan structure was obtained from PubChem (SID 481108482). The 3D- structure of BSA was obtained from the RCSB Protein Data Bank. Hydrogen atoms were added and crystallographic water molecules were removed before docking. Potential binding cavities on BSA were identified before docking. Docking simulations were performed separately for neohesperidin and kaempferide using LibDock protocol, which represents the protein as rigid while allowing each ligand to be flexible. For each flavonoid, the highest-ranking LibDock pose was selected for visualization. LibDock scores are dimensionless ranking outputs and were interpreted separately from the calculated interaction energies (kcal/mol); neither metric was treated as an experimental binding free energy. Comparative heat maps for BSA, curdlan, neohesperidin, and kaempferide were generated with the Heatmapper web server using a pairwise function. Docking poses for both flavonoids were visualized using MOE 2015 (Khalifa et al., 2021). The docking analysis was therefore used qualitatively to identify plausible contacts rather than to confirm extract composition or thermodynamic binding strength.

2.5. Characterization of BSA-Cn-Fs gels

2.5.1. SEM and EDX

Scanning electron microscopy (SEM) images of the gels were obtained using a Verios G4 UC microscope (Thermo Fisher Scientific, Brno, Czech Republic). A 3 mg sample was mounted on an aluminum stub, and an accelerating voltage of 5 kV was used. During the SEM process, oxygen, carbon, and nitrogen were analyzed by energy-dispersive X-ray spectroscopy (EDX) (Tian et al., 2024).

2.5.2. Color profile

The color profile after incorporating Fs into the BSA-Cn gel matrix was investigated using a MiniScan XE Plus 4500 L colorimeter (HunterLab, Reston, VA, USA). Color coordinates were expressed in CIE L*, a*, and b*, representing lightness, the red-green axis, and the yellow-blue axis, respectively (Salah et al., 2024).

2.5.3. Texture profile analysis (TPA)

Texture profile analysis was performed using a TA.XT Plus texture analyzer (Stable Micro Systems Ltd., Godalming, Surrey, UK) equipped with a 5-kg load cell (Teng et al., 2024).

2.6. Stability analysis

2.6.1. XRD

The crystalline and amorphous structures of the gels were examined using an X-ray diffractometer (Rigaku Corp., Tokyo, Japan). Approximately 50 mg of each sample was scanned from 5° to 90° at 4° min⁻¹, with the tube current and voltage set to 40 mA and 40 kV, respectively (Li, Dong, et al., 2024).

2.6.2. DSC

Thermal profiles were recorded using a differential scanning calorimeter (Setaram Instrumentation, Caluire, France). Approximately 20 mg of powdered gel was hermetically sealed in an aluminum pan and heated from 30 to 350 °C (Liu et al., 2023).

2.7. Dual-action effects of BSA-Cn-Fs gels

2.7.1. Release behavior of Fs

Fs release from the gels was determined by comparing the amount released into the medium with the initial amount loaded (Nalini et al., 2022; Su et al., 2024). Hydrogel samples were reconstituted in 5 mL of PBS (pH 7.4) at 37 °C with continuous stirring. The analysis was initiated by withdrawing 3-mL aliquots for absorbance measurement at 280 nm. Following aliquot removal, an equivalent volume of fresh PBS (3 mL) was immediately returned to each sample to maintain a constant total volume. The experiment was performed in triplicate to minimize variation, and results were expressed as mean ± SD. The release behavior of flavonoids from prepared gels was calculated using Eq. (2):

$$\mathit{Release\; profile}\left(\%\right)=\frac{\text{Released amount of flavonoids}}{\text{The intial amount of flavonoids}}x100$$ (2)

2.7.2. Antioxidant activity

The antioxidant capacity of the gels was measured using standard DPPH and ABTS radical-scavenging assays following Salah et al. (2020). For the DPPH assay, absorbance was measured at 517 nm using a microplate reader (Multiskan GO, Thermo Fisher Scientific, Waltham, MA, USA).

$$\mathit{DPPH\; Scavenging\; capacity}\left(\%\right)=\frac{\mathrm{B}-\mathrm{G}}{\mathrm{B}}x100$$ (3)

B denotes the blank absorbance, and G denotes the gel-sample absorbance.

For the ABTS assay, the working solution was adjusted to an absorbance of 0.70–0.80 at 734 nm. Sample (0.8 mL) was mixed with 0.2 mL of working solution, incubated at 25 °C for 10 min, and measured at 734 nm using a microplate reader (Multiskan GO, Thermo Fisher Scientific, Waltham, MA, USA).

$$\mathit{ABTS\; Scavenging\; ability}\left(\%\right)=\frac{\mathrm{B}-\mathrm{G}}{B}x100$$ (4)

B denotes the blank absorbance, and G denotes the gel-sample absorbance.

2.7.3. Antimicrobial effects

The antibacterial and antifungal activities of the gel samples were evaluated using standardized agar diffusion methods (Zhou et al., 2022). For the antibacterial assay, the following strains were utilized: B. cereus (ATCC 14579), E. coli (ATCC 700728), L. monocytogenes EGD-e (obtained from the BeNa-Culture Collection, China), and S. aureus (ATCC 25923, obtained from the China Center of Industrial Culture Collection, Beijing, China). Briefly, 100 μL of each fresh bacterial culture was spread evenly over nutrient agar plates. Sterile 6-mm paper discs were immersed in the respective gels for 2 min, placed on the inoculated agar surface, and incubated at 37 °C for 24 h. The antibacterial activity was quantified by measuring the diameter (mm) of the inhibition zones around all discs. No antibiotic positive control was included; therefore, inhibition-zone diameters were used only for relative comparison among gel formulations and not as an absolute benchmark of antimicrobial potency.

For the antifungal assay, P. expansum 3.3703 (isolated from moldy apples) and A. westerdijkiae strain fc-1 (obtained from the Chinese Academy of Agricultural Sciences, Beijing, China) were used. The fungi were cultured on potato dextrose agar (PDA). Spore suspensions were harvested with sterile water, quantified using a hemocytometer, and standardized to 1 × 10⁶ spores/mL. The gels were applied to inoculated PDA plates by the well-diffusion method, and the plates were incubated at 25 °C for 7 d. Growth inhibition (%) was calculated according to Salah et al. (2025):

$$\mathit{Inhibition}\left(\%\right)=\frac{\text{Control zone}-\text{Treatment zone}}{\text{Control zone}}x100$$ (5)

2.8. Application of gels to chicken breast preservation

The chicken breasts were divided into three groups, each containing three 25-g pieces that were subsequently sealed in sterile plastic packs (Chen et al., 2024). The marination conditions were as follows: 50 mL of treatment solution was applied for 5 min, corresponding to a 2:1 volume-to-mass ratio (mL/g). Group I received distilled water (control), group II received BSA-Cn gel, and group III received BSA-Cn-Fs-9 gel. To evaluate the preservation effects of the synthesized gels on chicken breast, pH, the lipid oxidation indicator (TBARS), and total colony count were determined at 0, 3, 6, 9, and 12 d of storage at 4 °C. The chicken samples were not artificially inoculated; the microbial counts therefore represented the naturally occurring microbiota of the samples. Direct marination was used as a proof-of-concept application model rather than as a finalized commercial or clean-label format.

2.8.1. pH of marinated chicken breast

One gram of chicken breast was homogenized for 5 min at 15,000 rpm with 10 mL of distilled water using an XHF-DY high-speed disperser (Ningbo Scientz Biotechnology Co., Ltd., Ningbo, China). The pH was measured using a PHS-25 pH meter (Shanghai Yidian Scientific Instruments Co., Ltd., Shanghai, China) (Zhao, Huang, et al., 2022).

2.8.2. TBARS value of marinated chicken breast

A 10-g portion of chicken breast was added to 50 mL of 7.5% trichloroacetic acid. The mixture was incubated at 40 °C for 1 h. To initiate the reaction, thiobarbituric acid reagent (5 mL) was thoroughly mixed with the same volume of chicken breast filtrate, and the mixture was heated at 100 °C for 20–30 min. The resulting solution was centrifuged, and the supernatant was analyzed by UV–visible spectroscopy (UV-8000, Metash Instruments, Shanghai, China). Absorbance was recorded at 532 nm. The TBARS value was calculated according to Qin et al. (2025): TBARS (mg/kg) = (A₅₃₂ / M_s) × 9.48 (6)

M_s denotes the mass of chicken breast used in the assay, and A₅₃₂ denotes the absorbance at 532 nm.

2.8.3. Total colony count

One gram of chicken breast was homogenized for 5 min at 15,000 rpm in 10 mL of sterile water using an XHF-DY high-speed disperser (Ningbo Scientz Biotechnology Co., Ltd., Ningbo, China). The homogenate was serially diluted 10-fold in sterile saline, and 100 μL of an appropriate dilution was spread evenly on nutrient agar plates (Kim et al., 2022). The plates were incubated at 37 °C for 24 h, and total viable colonies were counted. Counts of B. cereus, E. coli, L. monocytogenes EGD-e, and S. aureus were determined using mannitol egg-yolk polymyxin agar at 30 °C, violet red bile agar at 37 °C, ALOA agar at 37 °C, and Baird-Parker agar supplemented with egg-yolk tellurite at 30 °C, respectively.

2.9. Statistical analysis

All experiments were performed in triplicate, and results are presented as mean ± standard deviation (SD). Data were analyzed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA). Differences among means were assessed by one-way analysis of variance (ANOVA) followed by Duncan's multiple-range post hoc test, with statistical significance set at p < 0.05.

3. Results and discussion

3.1. Interactions and stability of BSA-Cn and Fs

Targeted UPLC-ESI-QQQ-MS/MS yielded quantifiable signals for 35 of the 46 monitored compounds in the AB-8-purified orange-peel extract (Table S1 and Fig. S1). Kaempferide was the most abundant targeted analyte at 78,252.01 ng/g (retention time, 7.34 min), followed by neohesperidin at 43,641.60 ng/g (5.18 min). These two compounds were therefore treated as the key representative flavonoids in the subsequent docking analysis.

In general, the Fs spectrum comprises two characteristic absorption bands. Band I is associated with the cinnamoyl system (the B ring and part of the C ring) and commonly occurs at 295–360 nm, whereas band II is associated with the benzoyl system (the A ring and part of the C ring) at 260–290 nm, as previously reported by Zhao, Meng, et al. (2022). Fig. 1A revealed that BSA-Cn spectra show two distinct bands; the band I detected at 203 nm refers to the framework of BSA, while the weak absorption band II detected at 277 nm relates to the residue amino acids, including tryptophan, tyrosine, and phenylalanine (Alam et al., 2015). The Fs spectrum (9 mg/mL) showed maximum at 330 and 279 nm. The BSA-Cn-Fs-9 spectrum retained the Fs-associated band near 330 nm, supporting incorporation of Fs into the gel matrix. The aromatic-residue band also shifted slightly from 277 to 279 nm, indicating a change in the local protein environment (Wang et al., 2022). Siddiqui et al. (2018) reported that vanillin binding increased the absorbance intensity of the BSA spectrum.

Fig. 1.

The Uv-absorption (A) and FTIR spectral (B) of BSA-Cn and BSA-Cn-Fs-9 gels. The fluorescence intensities of BSA, BSA—Cn, BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9 were also presented (C). The XRD (D) and DSC (E) peaks of Cn, BSA—Cn, and BSA-Cn-Fs-9 with their enthalpy and anomalies analysis.

FTIR spectroscopy was used to examine the intermolecular interaction between BSA-Cn gel and Fs (Fig. 1B). The BSA-Cn spectrum showed an O—H stretching band at 3323.78 cm⁻¹ and an amide I band (C O stretching) at 1635.91 cm⁻¹, the latter reflecting the secondary structure of BSA (Li & Hao, 2015). After Fs incorporation, both bands shifted. For instance, the BSA-Cn-Fs-9 spectrum showed a shift of the O—H band to the lower wavenumber of 3278.38 cm⁻¹, while the amide I band shifted slightly to 1635.70 cm⁻¹, consistent with the involvement of hydrogen bonding between the BSA-Cn gel and Fs. Liu et al. (2023) reported that the interaction between curdlan and soy protein isolate is strongly influenced by hydrogen-bond formation. Li, Dong, et al. (2024) reported that the interaction of polyphenols with curdlan shifted characteristic bands through hydrogen-bond interactions. Together, these shifts support hydrogen-bond involvement in the formation of the BSA-Cn-Fs-9 matrix (Liu et al., 2023).

Fluorescence intensity (FI) was employed to assess the binding interaction between BSA, Cn, Fs, and their conjugates. The intrinsic fluorescence data (Fig. 1C) showed that the formation of the BSA-Cn complex induced a 28.42% quenching, accompanied by a minor bathochromic shift in the emission maximum from 341 nm to 344 nm. Subsequent addition of Fs into the BSA-Cn system resulted in a concentration-dependent and pronounced quenching of fluorescence, with the emission peak undergoing a substantial red shift to 420 nm. At the maximum Fs concentration tested (9 mg/mL), the fluorescence was entirely quenched, with an efficiency of 97.44%. These changes are consistent with an altered local environment around BSA chromophores after Fs incorporation (Salah et al., 2020).

Fig. 1D showed that BSA displayed two characteristic diffraction peaks at 2θ positions of 9.40° and 20.75° (Data not shown). The XRD graphs of Cn and Fs-9 show one distinct peak at 19.65 and 20.90°, respectively. The BSA-Cn and BSA-Cn-Fs-9 gels show two diffraction peaks at 2θ positions of (9.05, 19.48°) and (9.34, 20.26°), respectively, indicating the occurrence of an amorphous state (Salah et al., 2024). The lower intensity of the BSA-Cn-Fs-9 pattern indicated reduced molecular ordering after Fs incorporation (Tian et al., 2024). Li, Dong, et al. (2024) demonstrated that the interaction of curdlan with polyphenols induced the transition from a more ordered to a more amorphous state. The XRD pattern therefore indicates that the loading of Fs-9 decreased the molecular order of BSA—Cn, leading to a reduction in the crystalline phase. This transition might help to enhance the bioaccessibility and sustain the activity of flavonoids in the long term (Li, Xu, et al., 2024).

Fig. 1E presents the thermal attributes of Cn, BSA—Cn, and BSA-Cn-Fs-9. The observed endothermic transitions between 86.7 °C and 101 °C in the DSC thermograms are characteristic of moisture evaporation, consistent with thermal behavior reported for similar hydrated biopolymer systems (Patel et al., 2024). Cn displayed endothermic peaks at 224.68 and 297.37 °C (Li, Dong, et al., 2024; Sun et al., 2011). After gel formation, the BSA-Cn transitions shifted to 228.73 and 317.00 °C, indicating altered thermal behavior of the composite matrix. The BSA-Cn-Fs-9 composite showed major transitions at 230.12 and 315.01 °C, compared to the result of BSA or Fs-9 (Fig. S2). Relative to Cn, the higher-temperature transitions of both composite gels indicate modification of the thermal response, consistent with non-covalent interactions between flavonoid hydroxyl groups and protein carbonyl-containing groups, as supported by prior findings (Ni et al., 2025).

XPS provided further evidence of compositional changes after gel formation (Fig. 2A). The spectra for the BSA-Cn gel showed characteristic peaks in the C1s region corresponding to C—C (282.77 eV), C C (284.26 eV), and C–O/C-N/C-O-C/C=O (285.90 eV) bonds, while O1s and N1s peaks indicated the existence of H-C–O, C—O, and C—N bonds, respectively. The flavonoid-loaded BSA-Cn-Fs-9 gel exhibited a similar profile but with enhanced signal intensities for the C C component at 284.26 eV and the O1s C—O component at 530.84 eV (Su et al., 2024). Li, Zhao, et al. (2024) reported that the XPS of chitosan-tannin (CS-TA) hydrogel displayed a new peak at 288.8 eV, resulting from an H-bond formed between the OH group of TA and the amino group of CS. Huang et al. (2024) reported that the interaction of phenolic acids with chitosan increased the peak at 284.92 eV, a result similarly observed in our XPS data. During hydrogel formation, hydroxyl groups may form hydrogen bonds with amino-acid groups in BSA, leading to further changes in the element content. The BSA-Cn-Fs-9 gel showed an increase in the O1s atomic percentage and a decrease in the C1s atomic percentage, consistent with an increased contribution from oxygen-containing groups after Fs incorporation, as shown in Fig. 2B. These elemental changes support the incorporation of Fs into the BSA-Cn matrix.

Fig. 2.

The XPS spectrum (A) atomic perecent (%) (B) of BSA—Cn, BSA-Cn-Fs-9, and Fs-9 gels.

Figs. 3A-B shows the interaction simulation between Cn, BSA, and the two representative flavonoids, kaempferide and neohesperidin. BSA-Cn depicts noncovalent association and identifies Tyr161, Lys162, Glu182, Arg185, and Leu115. The binding between Cn and BSA was reported to involve 9 amino acid residues, with a dimensionless LibDock score of −4.11 and a calculated interaction energy of −1.77 kcal/mol. Across the kaempferide interaction maps, predicted contacts involved Tyr149, Trp213, Arg217, Leu237, Ile263, Ala290, Ser343, and Leu346 with LibDock score of −4.54 and a binding energy of −1.91 kcal/mol. Across the neohesperidin interaction maps, labeled contacts included Trp213, Arg194, Arg293, Lys221, His241, Leu259, Ile263, Ser343, Leu346, and Leu368 with LibDock score of −4.02 and a binding energy of −1.98 kcal/mol. The heat maps showed distinct patterns for BSA, Cn, kaempferide, and neohesperidin, qualitatively consistent with the spectroscopic evidence of interaction (Fig. 1). Hydrophobic and electrostatic contacts may also contribute to the modeled complexes. Taken together, the data support the plausible incorporation of both flavonoids into the BSA-Cn matrix through noncovalent interactions, particularly H-bonding and hydrophobic contacts. Accordingly, the physical properties of the gel matrix were examined next.

Fig. 3.

The 2D- and 3D- docking analysis of BSA binding with Cn and kaempferide (A) or neohesperidin (B) showing the structure of each component, surface binding analysis, protein-ligands noncovalent interaction, hydrophobicity, electrostatic forces, and heat map analysis.

3.2. Characterization of BSA-Cn-Fs gel

The morphology of the prepared gel and Fs extract was investigated, as shown in Figs. 4A-C. The morphology of BSA-Cn showed a semi-spherical shape and uniform particles. Similar spherical or semi-spherical morphologies have been reported for BSA-carboxymethyl-curdlan and BSA-curdlan systems (Klimek et al., 2019; Rafigh et al., 2016), while soy protein isolate-curdlan gels have shown a regular, compact matrix (Liu et al., 2023). This morphology may reflect reorganization of curdlan helices during gelation (Xiao et al., 2017). Fs appeared as irregular, plate-like particles distributed heterogeneously (Fig. 4C). After Fs incorporation, BSA-Cn-Fs-9 retained semi-spherical particles but formed a more compact structure, indicating that Fs modified the microstructure generated during gelation (Li et al., 2025). The mean particle size increased from 0.87 μm for BSA-Cn (range, 0.4–1.6 μm) to 1.28 μm for BSA-Cn-Fs-9 (range, 0.5–3.0 μm). EDX analysis of C, N, and O is presented in Figs. S3A—C. The C and O atomic percentages followed the order Fs-9 < BSA-Cn < BSA-Cn-Fs-9; the higher C and O contributions in BSA-Cn-Fs-9 were consistent with the altered elemental composition expected after Fs incorporation.

Fig. 4.

SEM clicks of BSA-Cn (A), BSA-Cn-Fs-9 (B), and Fs-9 (C), as well as their surface plot and densiometric analysis. CrP: Crust porosity; CP: Core porosity; SO: Surface openings; FD: Fractal dimension.

Food texture and mouthfeel are influenced by the structures of proteins, polysaccharides, and their conjugates (Tao et al., 2022). The texture-profile parameters of the prepared gels are shown in Fig. S4A. Across BSA—Cn, BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9, hardness decreased from 557 to 169 g, chewiness from 105 to 8.5, springiness from 23% to 11%, and gumminess from 470 to 88. All parameters were highest for BSA—Cn. Potato protein-curdlan gel strength has likewise been reported to decrease under alkaline conditions (Yu et al., 2026), whereas addition of 4% curdlan improved the strength and elasticity of whey protein isolate gels (Hu et al., 2024). In the present system, the reductions after Fs incorporation may reflect disruption of protein-polysaccharide associations and reorganization of the gel network (Hou et al., 2024).

The color profiles of the BSA—Cn, BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9 gels were quantified and are presented in Fig. S4B. Fs incorporation increased L* and b* and decreased a*, indicating a lighter, more yellow appearance. The BSA-Cn-Fs gels consequently exhibited an orange‑tinted, gel-like appearance (Fig. S4C). These color changes provide additional evidence that Fs was distributed within the BSA-Cn matrix. Together, the texture and color data show that Fs incorporation modified the physical properties and appearance of the gels; their functional properties were therefore evaluated next.

3.3. Dual-component protective functions of the gels

3.3.1. Radical-scavenging activity and cumulative release profile

Flavonoid antioxidant capacity depends on concentration and molecular structure, and DPPH is commonly used to assess radical-scavenging activity in gel systems (Bahadori et al., 2017). The scavenging activity of gels against the free radicals of DPPH and ABTS was determined, as shown in Fig. 5A. DPPH-scavenging activity was 25.3 ± 2.79% for BSA—Cn, 31.49 ± 1.31% for BSA-Cn-Fs-5, 36.46 ± 1.61% for BSA-Cn-Fs-7, and 40.3 ± 2.7% for BSA-Cn-Fs-9. The corresponding ABTS values were 81.43%, 92.87%, 93.00%, and 93.73%, respectively. Using p < 0.05 as the significance threshold, the letters in Fig. 5A show that BSA-Cn-Fs-9 had the highest numerical DPPH value but did not differ significantly from BSA-Cn-Fs-7; similarly, the three Fs-loaded gels did not differ significantly in ABTS activity. Thus, the tested range showed a concentration-related numerical increase without establishing a statistically optimized global maximum. The BSA-Cn matrix may facilitate Fs dispersion and increase contact with radicals (Liang et al., 2021). Activity is associated with phenolic hydroxyl groups in Fs and antioxidant amino-acid residues in BSA (Kitts, 2005; Li, Dong, et al., 2024). Consistent with this interpretation, Cai et al. (2019) found that DPPH-scavenging activity of ferulic acid-grafted curdlan depended on ferulic acid loading, and Wang et al. (2021) reported enhanced activity after grafting ferulic acid onto carboxylated curdlan. Because functional activity depends on Fs availability, release from the gel matrix was evaluated over 24 h (Fig. 5B). Cumulative release reached 65.60% for BSA-Cn-Fs-5, 67.91% for BSA-Cn-Fs-7, and 72.91% for BSA-Cn-Fs-9. All gels showed gradual release, with a marked increase around 10–12 h followed by continued release to 24 h. Endpoint letters in Fig. 5B indicate that BSA-Cn-Fs-9 released significantly more Fs than BSA-Cn-Fs-7 and BSA-Cn-Fs-5 (p < 0.05). BSA-Cn-Fs-9 was therefore the best-performing formulation within the tested 5–9 mg/mL range, although formal optimization over a broader design space remains necessary.

Fig. 5.

The antioxidant activity versus DPPH and ABTS radicals (A), as well as their flavonoid release profile (%) (B) of BSA—Cn, BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9 gels. Values represent mean and ± standard deviation of three replicates. Different letters in the same column indicate significant differences among results (p < 0.05).

3.3.2. Antimicrobial activity

Flavonoids can exert antibacterial effects by disrupting membrane integrity, virulence, and energy metabolism (Faleye et al., 2023); accordingly, the BSA-Cn-Fs gels were evaluated for antibacterial activity. Figs. 6A-B demonstrate the inhibition ability of prepared gels against E. coli, S. aureus, B. cereus, and L. monocytogenes. The inhibition zones for L. monocytogenes were found to be 8.7 ± 0.01 mm for (BSA—Cn), 12.05 ± 0.02 mm for (BSA-Cn-Fs-5), 13.4 ± 0.06 mm for (BSA-Cn-Fs-7), and 16 ± 0.12 mm for (BSA-Cn-Fs-9), as shown in Fig. 6B. The gels' ability against S. aureus was demonstrated by the zone of inhibition values of 10.2 ± 0.03, 11.09 ± 0.08, 11.7 ± 0.02, and 12.7 ± 0.05 mm in BSA—Cn, BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9, orderly. Their ability against B. cereus was demonstrated by zone of inhibition values of 8.5 ± 0.01, 10.6 ± 0.05, 12.01 ± 0.03, and 13.4 ± 0.02 mm, respectively. Moreover, the synthesized BSA-Cn-Fs-9 gel, at the maximum amount of Fs extract, significantly (p < 0.05) exhibited antibacterial activity against four bacteria in the following order: E. coli < S. aureus < B. cereus < L. monocytogenes. The diameters of the antibacterial zones measured 16.0 ± 0.12, 13.4 ± 0.02, 12.7 ± 0.05, and 11.4 ± 0.05 mm, respectively, indicating that Fs incorporated in BSA-Cn gel significantly (p < 0.05) influenced the antibacterial activity. Lai et al. (2024) demonstrated the antibacterial activity of flavonoids extracted from navel orange peel against Gram-negative bacteria (Xanthomonas citri subsp. citri) and Gram-positive bacteria (Meticillin-resistant Staphylococcus aureus). Moreover, Lin et al. (2025) reported that the loaded carvacrol in a curdlan hydrogel exhibited good antimicrobial activity against the most common spoilage bacteria in fish.

Fig. 6.

The In vitro antibacterial activity (A) and their inhibitions zone diameter (B) of BSA—Cn, BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9 gels against B. ceraus, L. monocytogenes, S. aureus, and E. coli after 24 h. The In vitro antifungal activity (C) and their inhibition (%) after 3 and 7 d (D) of BSA—Cn, BSA-Cn-Fs-5, BSA-Cn-Fs-7, and BSA-Cn-Fs-9 gels against A. westerdijkiae, and P. expansum. Values represent mean and ± standard deviation of three replicates.

Furthermore, the antifungal activity of the prepared gels against P. expansum and A. westerdijkiae was shown in Figs. 6C-D. After 3d incubation time, the inhibition ability (%) against P. expansum was 11.1 ± 0.74, 26.26 ± 1.74, 29.29 ± 1.74, and 33.33 ± 0.01%, according to the sequence of prepared gels. While their inhibition ability (%) against A. westerdijkiae were 4.93 ± 2.13, 44.44 ± 3.7, 50.61 ± 2.13, and 55.55 ± 0.01, respectively. After a 7-day incubation time, the inhibition ability (%) of gels against P. expansum was 14.27 ± 1.06, 21.43 ± 0.26, 24.99 ± 0.93, and 35.70 ± 2.27%. Their inhibition ability (%) against A. westerdijkiae were 20.54 ± 2.06, 40.4 ± 1.26, 51.07 ± 1.08, and 57.44 ± 1.85, respectively. After 3 d, the inhibition ability (%) of BSA-Cn-Fs-9 against P. expansum and A. westerdijkiae was 33.33 ± 0.01 and 55.55 ± 0.01%, respectively. After 7 d, the inhibition ability (%) was 35.7 ± 2.27% and 57.44 ± 1.85%, respectively. The highest antifungal activity was observed for the BSA-Cn-Fs-9 gel, which showed inhibition of A. westerdijkiae and P. expansum. This outcome shows that the integration of Fs enhanced the antimicrobial activity of BSA-Cn gel by up to 36%. The prepared gel was effective in inhibiting bacteria and fungi, suggesting strong potential to preserve the quality of chicken breast.

3.4. Effects of gels on chicken breast quality

Fig. 7 The total colony count (A), the enumeration of B. cereus (B), E. coli (C), L. monocytogenes (D) S. aureus (E), the TBARS value (F), and the pH (G) of marinated chicken breast (H) were determined under different storage times (0, 3, 6, 9, and 12 d). At 0 days, the total bacteria count was 3.21 log₁₀ CFU/mL in the control group (Fig. 7A). The growth trend of the total bacteria count increased from 3.21 log₁₀ CFU/mL to 7.82 log₁₀ CFU/mL, indicating the contamination of chicken breast at the end of storage time. At 12 days, the total bacteria count was 6.98 log₁₀ CFU/mL in the BSA-Cn gel group, which was lower than that in the control group. Moreover, the total bacteria count was 5.94 log₁₀ CFU/mL in the BSA-Cn-Fs-9 gel group, demonstrating the inhibitory effect of the gel against microorganisms on the surface of chicken breast at the end of storage time. Therefore, the BSA-Cn-Fs-9 gel maintained the total number of colonies under the acceptable limit of the hygiene standard for fresh poultry meat (Si et al., 2024). The B. cereus count of chicken breast treated by BSA-Cn-Fs-9 gel was reduced from 6.23 log₁₀ CFU/mL to 4.11 log₁₀ CFU/mL at 12 d of storage time (Fig. 7B). The E. coli count of chicken breast treated by BSA-Cn-Fs-9 gel was reduced from 6.11 log₁₀ CFU/mL to 3.98 log₁₀ CFU/mL at 12 d of storage time (Fig. 7C). The growth inhibition trend was observed in the total counts of L. monocytogenes (Fig. 7D), and S. aureus (Fig. 7E), confirming that the inhibitory effect of BSA-Cn-Fs-9 gel. These data indicated that the BSA-Cn-Fs-9 gel preserved chicken breast against food spoilage bacteria under refrigeration conditions (Chen, Hu, Zheng, Wang, and Xu, 2024). Fig. 7F exhibits alterations in TBARS values under different treatments by synthesized gels during the storage time. The TBARS values of chicken breast marinated in the BSA-Cn-Fs-9 gel were significantly lower than those of the control group (p < 0.05), implying that incorporating the flavonoid extract into the gel matrix effectively prevented lipid oxidation. Fig. 7G shows a change in pH under different treatments using the synthesized gels over storage time. The results showed that the pH of chicken breasts (control) increased from 6.02 to 7.24 at the storage time of 0 d to 12 d, demonstrating a deterioration in the freshness of chicken breast samples. Firstly, the pH of the prepared hydrogel is higher than that of the control sample. This is attributed to the alkaline pH of the BSA-Cn gel matrix. However, the pH of the BSA-Cn-Fs-9 gel is lower than that of the control group, due to the acidic nature of the Fs extract in the gel. The increase in the pH of chicken breast during storage time could be related to microbial contamination. Notably, the pH of chicken breasts marinated in BSA-Cn gel increased from 6.19 to 6.90. The pH of chicken breasts marinated in BSA-Cn-Fs-9 gel increased from 5.73 to 6.58. This indicated that the BSA-Cn-Fs-9 gel significantly controlled the pH rise of chicken breasts during storage (p < 0.05), compared with the control group. Si et al. (2024) reported that the hydrogel formed from antimicrobial peptides, bacterial cellulose, and polyvinyl alcohol controlled the pH of chicken breast meat during storage. This lower pH might be linked to the inhibitory effect of the BSA-Cn-Fs-9 gel on microbial growth by inhibiting the formation of alkaline nitrogen compounds during meat deterioration (Lu et al., 2020). Taken together, the BSA-Cn-Fs-9 gel shows that it has maintained the quality of chicken breasts by inhibiting microbial growth, controlling pH, and reducing lipid oxidation.

Fig. 7.

The chicken breast quality attributes of the total colony count (A), the enumeration of B. cereus(B), E. coli(C), L. monocytogenes(D)S. aureus(E), the TBARS value (F), and the pH (G) of marinated chicken breast (H) were determined under different storage times (0, 3, 6, 9, and 12 d).

4. Conclusion

Heat-induced BSA-Cn gels successfully incorporated orange-peel Fs, and the spectroscopic and structural results were consistent with non-covalent interactions, particularly hydrogen bonding, within the composite matrix. Among the tested formulations, BSA-Cn-Fs-9 produced the highest numerical antioxidant values, the highest 24-h cumulative release, and the strongest antimicrobial performance; however, it should be regarded as the best-performing formulation within the tested range rather than as a statistically optimized global optimum. In the refrigerated chicken model, the treatment slowed microbial growth, limited the pH increase, and reduced TBARS relative to the untreated control through 12 d. The study is demonstrated by UPLC ESI–QQQ–MS/MS to screen the flavonoid extract profiling. However, the qualitative interpretation of low-magnitude docking energies, release testing only at pH 7.4, and the absence of antibiotic positive control are needed to be further studied. In addition, direct marination with a BSA-containing gel requires sensory, allergen-labeling, regulatory, and clean-label assessment. Future work should quantify individual flavonoids with authentic standards, optimize loading and gel composition, evaluate release under food-relevant pH conditions, include reference antimicrobials, and test the formulation as an edible coating or active-packaging component.

CRediT authorship contribution statement

Mahmoud Salah: Writing – review & editing, Writing – original draft, Methodology. Mohamed F. Hassan: Visualization, Software. Ibrahim Khalifa: Visualization, Software. Mohammed Mansour: Writing – review & editing. Bokai Hu: Writing – review & editing. Yuanyuan Liu: Validation. Qiujin Zhu: Supervision, Resources.

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.

Appendix A. Supplementary data

Supplementary material

Data availability

Data will be made available on request.