Enhancing garlic-clove shaped gel properties of mandarin fish (Siniperca chuatsi) myofibrillar protein with prickly ash extracts: Underlying roles of polyphenols

Abstract

The potential role of prickly ash extracts (PAEs) in enhancing garlic-clove shaped protein gel (GSPG)-forming capacity of mandarin fish (Siniperca chuatsi) myofibrillar protein (MP) was studied. PAEs with 0%, 20%, 60%, and 100% ethanol were utilized to systematically analyze changes in the physicochemical properties, water retention, chemical forces, and protein structure of GSPG. Results demonstrated that PAE60 (38.09 mg gallic acid equivalent/g) significantly improved gel strength and water-holding capacity compared to the control (P < 0.05). The immobilized water (23.04% decrease of relaxation time), hydrogen bonds (56.74% increase), disulfide bonds (174.31% increase), and β-sheet (49.42% increase) became significantly better than the control (P < 0.05). The gel microstructure had more orderly and compact networks with relatively increases of fractal dimension (1.83%), porosity (22.32%), and number of pores (23.66%) with statistical significances (P < 0.05). The findings could lay a theoretical groundwork for improving garlic-clove shaped fish protein with plant polyphenols.

Graphical abstract

Highlights

• •Garlic-clove shaped protein gel (GSPG) property was newly applied to fish protein.
• •Prickly ash polyphenol effectively enhanced GSPG formation of myofibrillar protein.
• •Prickly ash extract with 60% ethanol (PAE60) exerted the highest polyphenol yield.
• •PAE60 modulated the immobilized water, hydrogen and disulfide bonds, and β-sheet.
• •The study lays out a new strategy for fish GSPG's development in the food industry.

1. Introduction

Mandarin fish (Siniperca chuatsi) is widely distributed in East Asia, with China leading its global production, which reaches more than 400,000 tons per year and is valued at approximately 30 billion yuan (Huang et al., 2024; Jiang et al., 2023). Fermented mandarin fish, a traditional Chinese fish product, is characterized by its unique elastic texture, forming “garlic cloves-like meat” that is related to the heat-induced aggregation of myofibrillar protein (MP) (Guo et al., 2025; Zhou et al., 2023). Reflecting its substantial output, the fermented mandarin fish industry is experiencing significant growth in China, particularly within Anhui Province. In 2023, the province's annual processing volume approached 55,000 tons, generating an output value exceeding 5 billion yuan (Guo et al., 2025).

MP in fish muscle is recognized as the essential functional protein that determines heat-induced gelation properties (Ni et al., 2025). The formation of desirable fish protein texture is closely associated with favorable changes of MP in the water retention, chemical forces, and protein structure (Zhang et al., 2015). MP exerts the capacity to stabilize water within the fish protein gel matrix, thereby affecting the textural characteristics (Zhang et al., 2020; Zhao et al., 2020). Chemical forces such as disulfide bonds, hydrogen bonds, and hydrophobic interactions differently drive the structural transformation of MP to a viscoelastic texture (Zhao et al., 2024). Notably, MP governs the thermally induced gelation that is critical for the generation of garlic-clove shaped protein gel (GSPG) (Zhou et al., 2023).

Prickly ash, referring to the pericarp of Zanthoxylum bungeanum, is about 4⁓5 mm in diameter, slightly cracked red (red pepper) or green (green pepper) spherical peel, containing a black seed. Widely used as a spice in Asian food processing, prickly ash also is abundant in plant polyphenols such as phenolic acids and flavonoids, with studies reporting a polyphenol content of up to 54.54 mg GAE/g (Jing et al., 2021; Li et al., 2020). Gelation of MP, influencing the texture characteristics of the fish products, is a significant functional property for fish protein (Yu et al., 2022). Enhancement of gelling properties for MP can be achieved by optimizing protein microenvironment and binding substances, among which plant polyphenols are one of important modulating factors (Huang et al., 2022; Singh et al., 2020). Integrating natural polyphenols into MP has been shown to improve the parameters such as gel strength, texture, and water-holding capacity (WHC) (Xu et al., 2021). Moreover, polyphenols have the potential to induce the partial unfolding of protein structures, promote MP cross-linking, and thereby enhance gelation properties (Ni et al., 2025).

Prickly ash is added during the liquid-state fermentation, making mandarin fish successfully ferment in a suitable environment (Chen et al., 2023). Our previous studies optimized key microenvironmental parameters (e.g., acidification, ionic strength, and moisture content) for the fish muscle protein (Zhou et al., 2023). Besides, prickly ash is commonly used to enhance aroma and inhibit microbial growth in fermented fish products (Chen et al., 2023; Wang et al., 2021). However, the potential role of prickly ash polyphenols in reinforcing MP gel properties and consequently enhancing GSPG-forming capacity remains unclear.

This study evaluated the physicochemical properties (gel strength and texture), water retention (WHC and low-field nuclear magnetic resonance (LF-NMR)), chemical forces (non-covalent interactions, disulfide bonds and sulfhydryl groups), and protein structure (Fourier transform infrared spectra (FTIR) and scanning electron microscopy (SEM)) to elucidate the mechanism by which prickly ash extracts (PAEs) influence mandarin fish MP gel properties and GSPG-forming potentials.

2. Materials and methods

2.1. Materials

Prickly ash was procured from Yonghui Supermarket (Hefei, China). Gallic acid (GA), Folin-Ciocalteu's phenol reagent, and 5,5′-dithio-bis (2-nitrobenzoic acid) (DTNB), were obtained from Sigma-Aldrich, Shanghai, China. Sodium dodecyl sulfate (SDS) and other chemicals and reagents were purchased from Aladdin Co. Ltd, Shanghai, China. Cultured mandarin fish (600 ± 100 g) were sourced from Chizhou, Anhui Province, China, and humanely slaughtered as directed by Zhou et al. (2023). Followed a 20-min bath in ice-water (2.4 kg ice: 3.6 L water: 1 kg fish), the fish were euthanized by a blunt force trauma to the head in full compliance with EU Directive 2010/63 on the ethical standards for scientific animal use, with the approval from the Welfare and Ethics Committee for Animal Experimentation of the Anhui Academy of Agricultural Sciences (No. AAAS2024–24), and were stored at 4 °C for further processing.

2.2. Extraction of prickly ash polyphenol under different ethanol concentrations

Preparation of PAEs was based on the protocol outlined by Sun et al. (2020) with proper modifications. Under 70 °C, the prickly ash was desiccated to a constant weight, then pulverized and passed through a 20-mesh sieve. Each 10 g portion of the dried powder was blended with 100 mL of ethanol at 0%, 20%, 60%, and 100% (v/v) concentrations and underwent a 24-h extraction at room temperature, followed by ultrasonication for 30 min. After centrifugation at 5000 ×g for 30 min, the supernatant fluid was gathered, and the solid residues underwent two additional extractions under identical conditions. No. 9 filter paper and a 0.45 μm organic membrane were used to filter the combined supernatants sequentially. After rotary evaporation at 50 °C, a freeze dryer was used to lyophilize the extracts. The extracts created with 0%, 20%, 60%, and 100% ethanol were designated as PAE0, PAE20, PAE60, and PAE100, respectively, and were stored at −40 °C for usage. The Folin-Ciocalteu method was carried out to quantify the polyphenol content in the extracts with GA as the standard (Jing et al., 2021). The contents were presented as milligrams of gallic acid equivalent (GAE) per gram of dry weight of the PAEs.

2.3. Extraction of mandarin fish MPs

Mandarin fish MPs were extracted following the method stated by Wang et al. (2020). Back muscles from mandarin fish were minced using a food processor (JYS-A800, Joyoung Company Limited, Jinan, Shandong, China) and homogenized with quadruple volume of cold extraction buffer (50 mM NaH₂PO₄/Na₂HPO₄, 100 mM NaCl, 2 mM MgCl₂, 1 mM EDTA, pH 7.0) for 2 min. After centrifugation (3500 ×g, 15 min, 4 °C), the precipitate was collected and underwent two washes following the previously mentioned procedure. Ultimately, the sediment was washed three times with quadruple volume of 100 mM NaCl solution to obtain fresh fish protein paste. Before the last centrifugation, the protein suspension was filtered with a 4-layer filter cloth to remove the connective tissue.

2.4. Preparation of GSPG using heat-induced MP-PAE gels

GSPG was prepared based on the process outlined by Wang et al. (2020) and Zhou et al. (2023) with proper modifications. MPs (fresh fish protein paste) were dissolved in a solution (pH 6.0) comprising 0.6 M NaCl and 20 mM Na₂HPO₄/NaH₂PO₄ at a 45 mg/mL concentration. The suspension was divided into five portions: four were supplemented with 0.25% PAE0, PAE20, PAE60, and PAE100 (w/w, based on protein), respectively, while one served as a control without PAE addition. After equilibration (4 °C, 4 h) and centrifugation (3500 ×g, 5 min, 4 °C), MP gels were processed by incubating at 40 °C for 30 min, then heating at 90 °C for 30 min, cooling in ice water for 2 h, and storing at 4 °C. Finally, the MP gel was sliced into garlic clove shapes (thickness, 5 mm), similar to the garlic clove-like meat found in fermented mandarin fish, for further detection.

2.5. Gel strength

A TA-XT Plus texture analyzer (Stable Micro Systems Ltd., Godalming, UK) was employed to assess gel strength of the MP gels, following the procedure stated by Zhou et al. (2023) with proper adjustments. A P/0.5S spherical probe with the following settings was used for the measurement: test speed of 1.00 mm/s; compression distance of 3.00 mm; return speed of 10.00 mm/s; return distance of 15.00 mm; constant force of 1.00 g.

2.6. Texture profile analysis (TPA)

The TPA of the MP gels, containing hardness, springiness, cohesiveness, gumminess, chewiness, and resilience, were evaluated according to the method stated by Zhou et al. (2023) with minimal tweaks. The texture analyzer with a P/36R probe was applied and the detailed parameters were listed below: trigger force, 5.00 g; test speed, 5.00 mm/s; post-test speed, 5.00 mm/s; pre-test speed, 1.00 mm/s; compression distance, 3.00 mm; interval time, 5.00 s.

2.7. WHC

WHC for the MP gels was assayed following a previously utilized approach (Yang et al., 2024). The MP gel was wrapped in three layers of filter paper after accurate weighing (m₁), and centrifuged at 6000 ×g for 10 min at 4 °C. Subsequently, the resulting mass of MP gel (m₂) was measured with precision. The WHC calculation was performed using the equation below:

$$\mathrm{WHC}\left(\%\right)=m₂/m₁\times 100\%$$ (1)

2.8. LF-NMR

An LF NMR analyzer (MesoMR23-060H-I, Suzhou Newmark Analytical Instruments Co., Suzhou, China) was employed to gauge the spin-spin relaxation times (T₂) based on a method sketched by Zhou et al. (2023). The MP gel was loaded into a nuclear magnetic resonance tube and analyzed following the Carr-Purcell-Meiboom-Gill pulse sequence with these settings: 22 MHz frequency, 1800 ms waiting period, 0.2 ms echo interval, 8 scans conducted, and 4000 echoes collected.

2.9. Non-covalent interactions

Non-covalent interactions were analyzed following a previously reported method (Zhang et al., 2023). For each treatment, the MP gels were homogenized with four solutions, containing SA of 0.05 M NaCl, SB of 0.6 M NaCl, SC of 0.6 M NaCl in 1.5 M urea, and SD of 0.6 M NaCl in 8 M urea, at a ratio of 1 g: 10 mL. The mixtures were incubated at 4 °C for 1 h. After centrifugation (10,000 ×g, 4 °C, 15 min), the biuret reagent served to assess the supernatant's protein content. The variation in protein concentration across specific solutions, including ionic bonds (SB-SA), hydrogen bonds (SC-SB), and hydrophobic interactions (SD-SC), was used to characterize the contributions of non-covalent interactions. The results are represented in terms of the percentage of all non-covalent interactions.

2.10. Disulfide bonds and sulfhydryl groups

The NTSB (2-nitro-5-thiosulphobenzoate) and DTNB assays were used to assess the concentrations of disulfide bonds and sulfhydryl groups, respectively, adopting the technique outlined by Zhang et al. (2018). 0.5 mL of the SD supernatant (4 mg/mL) was mixed with 4.5 mL of buffer A (1% SDS, 0.2 M Tris-HCl, 0.1 M Na₂SO₃, 3 mM EDTA, 8 M urea, pH 8.0, for disulfide bond determination) or buffer C (1% SDS, 0.2 M Tris-HCl, 3 mM EDTA, 8 M urea, pH 8.0, for sulfhydryl group determination) thoroughly. Subsequently, 4 mL of this mixture was combined with 0.4 mL of buffer B (10 mM NTSB, 1% SDS, 0.2 M Tris-HCl, 0.1 M Na₂SO₃, 3 mM EDTA, 8 M urea, pH 9.5, for disulfide bond determination) or buffer D (10 mM DTNB, 10 mM Tris-HCl, pH 8.0, for sulfhydryl group determination), and kept at 40 °C for 30 min. A UV-5500 spectrophotometer (Shanghai Metash Instruments Co., Ltd., Shanghai, China) was utilized to measure the absorbance at 412 nm. The content calculation was performed using the equation below:

$$\text{Concentration}\left(\mathrm{mol}/{10}^5\mathrm{g}\right)=\frac{A\times D}{B\times C}$$ (2)

where A, B, C, D, respectively, is the absorbance at 412 nm, the protein content, a molar extinction coefficient (13,600 M⁻¹ cm⁻¹), and the dilution ratio.

2.11. FTIR

For FTIR analysis, the MP gels were lyophilized and pulverized with KBr in a 1:100 (g/g) ratio. An FTIR spectrometer (Nicolet iS20, Thermo Fisher Scientific, Massachusetts, USA) captured the spectra over 4000⁓500 cm⁻¹, with 32 scans at 4 cm⁻¹ resolution. Protein secondary structures were examined with the help of PeakFit software (version 4.12; Seasolve, Framingham, MA, USA).

2.12. SEM

To study the microstructure of the MP gels, a scanning electron microscope (Gemini 300, Carl Zeiss AG, Jena, Germany) was employed, adhering to the procedure stated by Zhou et al. (2023) with some adjustments. Cubes measuring 3 × 3 × 3 mm were cut from MP gels, immersed in glutaraldehyde (2.5%, v/v) for 24 h, and rinsed thrice with phosphate buffer (0.1 M, pH 7.0). Dehydration was performed with 30%, 50%, 70%, 80%, and 90% ethanol for 15 min per step, followed by a 30-min treatment with anhydrous ethanol. A critical point dryer (K850, Quorum Technologies, East Sussex, UK) was used to dry the dehydrated samples and coat them with gold before scanning electron microscopy.

Microscopy images were examined as per the protocol specified by Zhang et al. (2020). ImageJ software (version 1.5.2a) with the Fractal Box Count plug-in was utilized to threshold and binarize the micrographs. The fractal dimension (Df), porosity, and number of pores were calculated. The box-counting technique involved overlaying grids of progressively smaller dimensions onto the image and tallying the grid boxes that contained pixels for each grid scale. The Df was derived using the scaling rule, and calculated using these equations:

$$D=-\log \mathit{N\epsilon }/\log \epsilon$$ (3)

$$\mathrm{Df}=D+1$$ (4)

where Nε and ε represent the count of boxes in the grid at a specific scale containing pixels, and the corresponding scale, respectively.

2.13. Statistical analysis

All the measurements were performed three times, and data of the experiment are presented as mean ± standard deviations. The mean comparisons were performed using one-way analysis of variance, followed by Duncan's multiple analysis at the 5% level using SPSS (version 22.0, IBM Co., New York, USA).

3. Results and discussion

3.1. Polyphenol contents of PAEs

The prickly ash was treated with ethanol at 0%, 20%, 60%, and 100% (v/v) concentrations. As illustrated in Fig. 1, the corresponding polyphenol contents could be observed in PAEs by the used ethanol concentrations. The trend of polyphenol yields with ethanol concentrations from 0% to 100% was first increased and then decreased. The maximum yield of 38.09 mg GAE/g was achieved at the 60% ethanol concentration, which, significantly, was greater than that of the others (P < 0.05). This finding was similar to the previous research on coconut husk, which exhibited that phenolic content extracted with 60% ethanol was significantly higher than that with 80%, 40%, or 100% ethanol (Buamard & Benjakul, 2015). Increased concentration of ethanol could facilitate the breakage of cell membrane and enhance phenolic solubility, while excessive ethanol concentration might result in protein coagulation, impeding the diffusion of phenolics (Latif et al., 2024). Furthermore, absolute ethanol (100% ethanol concentration) was reported to relatively strengthen the hydrogen bonding between polyphenols and macromolecules, producing an inhibitory effect on polyphenol extraction (Ma et al., 2025).

Fig. 1.

Polyphenol yield of prickly ash (Zanthoxylum bungeanum) with different ethanol concentrations. Values are mean ± standard deviations (n = 3). Values with the same letters are not significantly different (P > 0.05). GAE represents gallic acid equivalent.

3.2. Changes in physicochemical properties of MP-PAE gels

3.2.1. Gel strength

The prepared GSPG samples using MP with PAEs are shown in Fig. 2A. The shape and the thickness were simulated to match the parameters of “garlic cloves-like meat” found in fermented mandarin fish (Zhou et al., 2023). The gel strength of the GSPG samples were presented in Fig. 2B. After adding PAE0 or PAE20, the gel strength significantly became stronger as compared to that of the control (without PAE addition) (P < 0.05). Furthermore, PAE60 exhibited excellent gel strength value, significantly higher than other treatments (P < 0.05), which was generally consistent with the results reported by Buamard and Benjakul (2015). Previous investigations have also illustrated that some polyphenols can enhance the strength of MP gels to different extents (Jiang et al., 2025; Li et al., 2021). Ni et al. (2025) thought that polyphenols exert the capacity to decrease the degradation of MPs during thermal gelation, thereby maintaining the integrity of protein network structure. However, the gel strength with PAE100 was relatively weaker than that of PAE60 (P < 0.05), which might result from the interference of co-extracted fat-soluble impurities in the ethanol extract on the protein matrix formation (Yan et al., 2020; Zhang et al., 2024).

Fig. 2.

Changes in physicochemical properties of mandarin fish MP gels with PAEs by GSPG (A), gel strength (B), and TPA (C—H). Values are mean ± standard deviations (n = 3). Values with the same letters are not significantly different (P > 0.05) for each detected parameter. MP, PAEs, GSPG, and TPA, respectively, represent myofibrillar protein, prickly ash extracts, garlic-clove shaped protein gel, and texture profile analysis.

3.2.2. TPA

TPA values have become an important criterion for testing gel quality in processed fish products (Zhou et al., 2023). The textural characteristics of the mandarin fish MP gel (GSPG samples), as affected by various PAEs, are shown in Figs. 2C–2H, including hardness, springiness, cohesiveness, gumminess, chewiness, and resilience, respectively. In Fig. 2C, PAE significantly enhanced the hardness of the MP gel, and the PAE60 gel presented a markedly higher hardness value than other treatments (P < 0.05). This textural superiority existed in springiness (Fig. 2D), gumminess (Fig. 2F), and chewiness (Fig. 2G). However, cohesiveness (Fig. 2E) and resilience (Fig. 2H) differed from them, showing no statistically significant variation between the PAE60 gel and the control (P > 0.05). The dose has been proven to affect the extent to which polyphenols improve the texture characteristics of MP-rich surimi gels (Buamard & Benjakul, 2015; Singh et al., 2020). In short, the variation of polyphenol contents in PAEs (Fig. 1) was considered to be the suggested reason affecting the texture performance of the MP gel or the GSPG.

3.3. Changes in water retention of MP-PAE gels

3.3.1. WHC

WHC represents how well protein gels can hold water and is regularly applied to assess the yield and quality of gel products (Zhang et al., 2020). Fig. 3A shows the changes in WHC of the MP gel with PAEs. Compared with the control group, the WHC with PAEs was increased to different extents, while PAE60 exhibited the highest WHC (P < 0.05). Previous studies have demonstrated that MP gels treated with polyphenol compounds relatively exhibited higher WHC (Jiang et al., 2025; Li et al., 2021; Ni et al., 2025). The water retention of stabilized gel structure is thought to be facilitated by non-covalent interactions (especially hydrogen bonds) induced by phenolic compounds (Jiang et al., 2025; Xu et al., 2021). Furthermore, the differences in the degree of WHC improvement among the four PAE groups might be due to the variations in their polyphenol contents (Fig. 1). Guo et al. (2024) found that higher WHC was observed in MP gel with greater polyphenol incorporation.

Fig. 3.

Changes in water retention of mandarin fish MP gels with PAEs. A, B and C correspond to WHC, LF-NMR spectra and water status, respectively. Values are mean ± standard deviations (n = 3). Values with the same letters are not significantly different (P > 0.05) for each detected parameter. MP, PAEs, WHC, and LF-NMR, respectively, represent myofibrillar protein, prickly ash extracts, water-holding capacity, and low-field nuclear magnetic resonance.

3.3.2. LF-NMR

Using LF-NMR to measure T₂ relaxation time of hydrogen protons can efficiently evaluate how water molecules are distributed and migrated within gel systems (Li et al., 2021). Fig. 3B presents the T₂ spectra of mandarin fish MP gels containing PAEs. Two types of water, including bound water (T₂₁, 0.1–1 ms) and immobilized water (T₂₂, 40–400 ms), were observed, whilst free water was not detected under the LF-NMR experimental conditions. A similar pattern of water distribution in mandarin fish sample was reported by Zhou et al. (2023).

Water behavior in protein gels follows distinct patterns: bound water is intimately connected to macromolecules such as proteins and polysaccharides, making it immobile, while immobilized water is physically trapped within the gel network and exhibits restricted movement (Zhao et al., 2020). In Fig. 3C, the T₂₂ peak area proportion remained over 98% of the total peak area in all groups, suggesting that immobilized water was predominant in mandarin fish MP gels. PAE60 gel exhibited the shortest T₂₂ relaxation time (84.66 ms) across all samples. This parameter, as outlined by Wang et al. (2020), conversely correlates with water mobility, with shorter times indicating tighter water confinement and more homogeneous gel structures. Therefore, the reduced relaxation time of PAE60 could indicate the enhanced water retention within the gel network, probably resulting from the polyphenol-participated improvements in MP cross-linking (Xu et al., 2021). Li et al. (2021) reported that a type of polyphenol increased the exposure sites of active groups in MPs, which might enhance the interactions between MPs and phenolic compounds. These findings were mainly in agreement with the WHC's results (Fig. 3A), further illustrating the potential role of prickly ash polyphenols in improving gel water retention.

3.4. Changes in chemical forces of MP-PAE gels

3.4.1. Non-covalent interactions

The key non-covalent interactions, containing ionic bonds, hydrogen bonds, and hydrophobic interactions, contribute to the generation of protein gels, affecting their structure and properties (Zhang et al., 2023). In Fig. 4A, hydrophobic interactions in mandarin fish MP gels accounted for 86.71% of the total non-covalent bonds in the control group, followed sequentially by hydrogen bonds (9.07%) and ionic bonds (4.22%). A comparable result was observed in heat-induced MP-rich carp surimi gels, as the literature stated by Zhang et al. (2018). In short, the primary force behind the formation of heat-induced MP gels from mandarin fish is hydrophobic interactions, which are also essential in supporting the characteristics of the gel. Zhang et al. (2018) thought that, when gelation occurs, proteins unfold and expose hydrophobic amino acids that cluster into myosin heavy chain groups, which quickly enhances hydrophobic interactions.

Fig. 4.

Changes in chemical forces of mandarin fish MP gels with PAEs. A, and B correspond to non-covalent interactions, and disulfide bond and sulfhydryl contents, respectively. Values are mean ± standard deviations (n = 3). Values with the same letters are not significantly different (P > 0.05) for each detected parameter. MP and PAEs, respectively, represent myofibrillar protein and prickly ash extracts.

The addition of PAEs in this study differently reduced the proportion of hydrophobic interactions at PAE20, PAE60, and PAE100, while the hydrophobic interactions of PAE60 were predominantly the non-covalent interactions (accounting for 81.15%) in mandarin fish MPs. Previous studies have reported that hydrophobic interactions are essential molecular forces in forming protein-polyphenol conjugates (Li et al., 2021; Quan et al., 2019). However, the reduction phenomena might be due to the quantity and type of phenolic molecular in PAEs, which determine their function in protein (Pan et al., 2020).

In Fig. 4A, Hydrogen bonds, as compared to the control, were significantly strengthened by using different PAEs (P < 0.05), and reached the greatest value with PAE60 by attaining 1.57 times of the control. Jiang et al. (2025) reported that phenolic compounds enhanced the hydrogen bonds in Goose MP gel, which is primarily in agreement with the observed results above. Phenolic compounds bound with proteins via hydrogen bonds, applying their hydroxyl groups to be the main binding locations (Quan et al., 2019; Xu et al., 2021). Li et al. (2021) thought that phenolic compounds can unfold protein structures, revealing tyrosine residues as strong hydrogen bond acceptors, which allow phenolic hydroxyl groups to bind and form hydrogen bonds.

Furthermore, the addition of PAEs increased the proportion of ionic bonds in mandarin fish MP gels at PAE20 and PAE60. Positively charged groups of proteins, such as ε-amino groups in lysines, is thought to react with negatively charged hydroxyl groups present in polyphenols to form ionic interactions (Quan et al., 2019).

3.4.2. Disulfide bonds and sulfhydryl groups

During the thermal gelation of fish proteins, the generation and transformation of disulfide bonds and sulfhydryl groups play a vital role (Li et al., 2021). Fig. 4B depicts the variations in disulfide bonds and sulfhydryl groups in mandarin fish MP with PAEs. There was an approximate upward trend in disulfide bond content from the control to PAE0, PAE20, and PAE60, with sulfhydryl content trending downward generally. Similar polyphenol-dose-dependent reductions of sulfhydryl groups were reported in previous studies, which might be due to the generation of C—S bonds through the addition reactions of oxidized phenolic compounds with sulfhydryl groups in peptide chains (Guo et al., 2024; Huang et al., 2022). Moreover, abundant hydroxyl groups in phenolic compounds have been demonstrated to interact with sulfhydryl groups, resulting in the corresponding decrease (Guo et al., 2024). The growth of disulfide bonds commonly indicated the improved gel structure or properties (Jiang et al., 2025; Xu et al., 2021). Studies have reported that oxidized polyphenols can promote the alteration of sulfhydryl groups into disulfide bonds while being reduced and might induce conformational changes in proteins, facilitating the oxidation of sulfhydryl groups (Balange & Benjakul, 2009; Zhao et al., 2021). Phenolics might be oxidized spontaneously during PAE preparation or MP gelation, contributing to the cross-linking of mandarin fish MPs (Singh et al., 2020).

3.5. Changes in protein structure of MP-PAE gels

3.5.1. FTIR

FTIR analysis is usually utilized to assess changes in secondary structures of proteins by identifying their characteristic absorption peaks (Zhou et al., 2023). Analyzing the secondary structures of proteins relies heavily on the amide I region, with absorption peaks at 1650⁓1660, 1600⁓1640, 1660⁓1695, and 1640⁓1650 cm⁻¹, respectively, corresponding to the α-helix, β-sheet, β-turn, and random coil structures (Zhang et al., 2023). Fig. 5A depicts the secondary structures of mandarin fish MP gels with PAEs. The β-sheet was dominant conformation in the MP gel of mandarin fish, accounting for more than 30%, which was contributed from the α-helix into β-sheet during the process of thermal gelation partially (Jiang et al., 2025). The group with PAEs added, as compared to the control, had diminished contents of α-helix, β-turn, and random coil, and an increased β-sheet content (P < 0.05). Similar unfolding-promoting effect of phenolic compounds was also reported in previous studies (Huang et al., 2022; Xu et al., 2021). Pan et al. (2020) demonstrated that the α-helix structure was stabilized by the hydrogen bonding networks between amino acid backbones chiefly. However, polyphenols with their profuse hydroxyl groups could compete for these binding sites, potentially destabilizing the α-helix configuration and inducing partial protein unfolding (Guo et al., 2024). Concurrently, the aggregation induced by disulfide bonds could also increase β-sheet content, because these bonds might reorganize polypeptide chains into extended and interlinked structures (Wang et al., 2020).

Fig. 5.

Changes in the structure of mandarin fish MP gels with PAEs. A, B—F, G, H, and I correspond to secondary structure, microstructures, Df, porosity, and number of pores, respectively. Values are mean ± standard deviations (n = 3). Values with the same letters are not significantly different (P > 0.05) for each detected parameter. MP, PAEs, and Df, respectively, represent myofibrillar protein, prickly ash extracts, and fractal dimension.

PAE60 possessed the highest β-sheet content (45.30%), exhibiting marked distinctions from other PAEs (P < 0.05). The structural change from α-helix conformation to the β-sheet could promote an increase of juiciness by supplying additional binding sites for water molecules, as confirmed by both the improved WHC (Fig. 3A) and enhanced gel strength (Fig. 2B) (Wang et al., 2020). Additionally, the reduced random coil content with PAE60 could further enhance the gel network's orderliness and stability (Huang et al., 2022).

3.5.2. SEM

SEM analysis shows the critical microstructural differences in mandarin fish MP gel samples (Figs. 5B-5F), with quantitative comparisons of Df (Fig. 5G), porosity (Fig. 5H), and number of pores (Fig. 5I). PAE incorporation improved the microstructures of the MP gels to different extents (Figs. 5B-5F), as demonstrated by more compact and organized networks with smaller voids and a continuous matrix, the observation of which is majorly consistent with the results of previous researches (Guo et al., 2024; Jiang et al., 2025; Ni et al., 2025). Similarly, phenolic compounds in coconut husk ethanolic extracts have enhanced protein strand density, connectivity, and rigidity in surimi gel networks (Buamard & Benjakul, 2015; Singh et al., 2020).

Df is commonly used to analyze food structure, with higher values indicating more complex and orderly structures characterized by enhanced protein cross-linking (Zhang et al., 2020). In Fig. 5G, the Df of PAE60, significantly, was better than the other treatments (P < 0.05), and similar results could be seen in the porosity (Fig. 5H) and the number of pores (Fig. 5I). The results might be due to the phenolic-induced cross-linking with proteins through covalent interactions and hydrogen bonds (Singh et al., 2020; Yang et al., 2025). Jiang et al. (2025) suggested that hydroxyl groups of polyphenols interact with the amino and carboxyl groups in MPs, leading to the formation of dense gel networks with superior texture properties. Furthermore, denser gel networks could retain more water (Guo et al., 2024), which coincided with the observed improvement in WHC (Fig. 3A).

3.6. Underlying roles of prickly ash polyphenols in the enhancement of GSPG using MP with PAE

The PAE60 polyphenols (38.09 mg GAE/g) significantly enhanced the values of gel strength and WHC, respectively, by 459.05% (Fig. 2B) and 15.95% (Fig. 3A) compared to the control (P < 0.05). As their corresponding results, T₂₂ relaxation time in Fig. 3C (23.04% decrease), hydrogen bonds in Fig. 4A (56.74% increase), disulfide bonds in Fig. 4B (174.31% increase), and β-sheet structures in Fig. 5A (49.42% increase) with statistically significances (P < 0.05) were proved to play important roles in the enhanced gel properties. Meanwhile, the PAE60 gel microstructure exhibited more orderly and compact three-dimensional networks with an excellent Df of 2.836 (1.83% increase), a porosity of 52.07 (22.32% increase), and a number of pores of 308.33 (23.66% increase) in Figs. 5B-5I (P < 0.05).

Based on the significant results, Fig. 6 shows a suggested mechanism elucidating how polyphenols from prickly ash enhance gel properties of mandarin fish MPs using GSPG model. Polyphenols of prickly ash might promote the MP's unfolding in secondary structures and the transformation from α-helix conformation to the β-sheet by disrupting the MPs' hydrogen bonds that stabilize the α-helix structures. The structural change exposed active groups within the peptide chains, which might act as hydrogen bond acceptors and interact with phenolic hydroxyl groups to form hydrogen bonds. Furthermore, the conformational changes in proteins might promote the oxidation of sulfhydryl groups into disulfide bonds. Therefore, mandarin fish MPs become highly cross-linked, aggregating into a dense and uniform gel network with better textural properties. Furthermore, water mobility might be restricted within the compact structure, resulting in an advanced capacity for holding water.

Fig. 6.

Proposed mechanism of prickly ash polyphenols in the formation of GSPG using MP with PAE. GSPG, MP, and PAE, respectively, represent garlic-clove shaped protein gel, myofibrillar protein, and prickly ash extract.

4. Conclusion

This study demonstrated that prickly ash ethanolic extracts significantly enhance the gel properties of mandarin fish MP under GSPG model, with the optimal effect observed at 60% ethanol. PAE60 (38.09 mg gallic acid equivalent/g) significantly enhanced gel strength (459.05% increase) compared to the control (P < 0.05). Additionally, a 15.95% increase in WHC was observed, which was related to the reduced mobility of immobilized water (a 23.04% decrease of relaxation time) with statistical significances (P < 0.05). Furthermore, the addition of PAE60 facilitated the development of hydrogen bonds (56.74% increase) and disulfide bonds (174.31% increase), enhancing the MP cross-linking. FTIR analysis revealed that polyphenols in prickly ash may facilitate the α-helix unfolding of MPs, thus raising the β-sheet content by 49.42% (P < 0.05). Meanwhile, the gel network was more compact and uniform with relatively increases of fractal dimension (1.83%), porosity (22.32%), and number of pores (23.66%). Therefore, PAEs rich in polyphenols could serve as natural additives to enhance the gel qualities of mandarin fish MP, which is beneficial for the formation of the unique “garlic cloves-like meat”. Also, this research could provide novel strategies for the mandarin fish processing industry by modulating MP function to create novel fish protein foods. Future research is underway to investigate the impact of various phenolic molecular in PAEs on the gel properties of mandarin fish's GSPG, with an explanation provided by the MP gelation mechanism.

CRediT authorship contribution statement

Wanmei Li: Writing – review & editing, Writing – original draft, Visualization, Methodology, Data curation, Conceptualization. Yingqin Zhou: Validation, Investigation, Formal analysis. Haiqiu Wei: Visualization, Validation, Investigation. Kexin Xiong: Validation, Methodology. Yan Yan: Validation, Methodology. Jingjing Huang: Validation, Methodology. Ningning Xie: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

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.

Data availability

Data will be made available on request.