Effect of ethanol concentration on conjugates of gallic acid and whey protein isolate: structural and functional properties

Abstract

This study investigated the impact of varying ethanol concentrations on the functionality and structure of conjugates of gallic acid (GA) and whey protein isolate (WPI). The solubility, total phenolic content, and sulfhydryl group content of the conjugates significantly decreased with increasing ethanol concentration. The structure of the conjugates was altered as indicated by changes in their infrared and fluorescence spectra. By measuring the emulsification and foaming properties, we found that ethanol improved the functional properties of the conjugates. The study examined the influence of ethanol on the conjugates of polyphenolic compounds with whey protein, broadening the scope of research on protein modification. Furthermore, it provided a scientific basis for the future development of novel functional foods.

Graphical Abstract

Open in new tabDownload slide

Introduction

Whey protein isolate (WPI) contains a variety of beneficial proteins, including β-lactoglobulin, α-lactalbumin, bovine serum albumin, and immunoglobulin (Wang et al., 2020b). Its high solubility, foaming, emulsification, gelling and slimy taste make it a popular functional food additive in the food formulation. Because hydrophobic residues are generally buried within the compact rigid globular protein structures of the intact form of WPI, they cannot effectively perform their functional characteristics, thus restricting their utilisation in the field of food (Feng et al., 2021).

In recent research, it has been found that ethanol can cause protein denaturation, alter protein secondary structure, and promote protein aggregation (Kundu et al., 2017). Due to its lower polarity and dielectric constant compared to water, ethanol is more susceptible to intermolecular dissociation. This can lead to the weakening or destruction of non-covalent bonds maintained by hydrogen bonds and hydrophobic interactions, leading to protein denaturation and aggregation (Peng et al., 2020). The conformation of the globular proteins was altered following treatment with varying concentrations of ethanol, and the denaturation of the proteins was most pronounced at approximately 50%. Hoon Lee et al. (2024) prepared mealworm protein treated with various concentrations of ethanol. As the concentration of ethanol increased, the secondary structure of the protein was affected, and the number of α-helices decreased significantly with the increase in ethanol concentration.

The interaction of polyphenols and proteins has attracted considerable attention from the scientific community. When polyphenols undergo oxidation to form quinones, they can interact covalently with proteins and affect their emulsification, foaming and gel ability (Cheng et al., 2024). Existing studies have focused on the modification of proteins by ethanol or polyphenols, respectively. However, the effect of ethanol concentration on polyphenol protein conjugates has not been reported. Therefore, the purpose of this paper is to study the effects of different ethanol concentrations on the structure and functional properties of polyphenol-protein conjugate. This study was of great significance to the understanding of ethanol and gallic acid (GA)-protein conjugates and provided a theoretical basis for the development of new food additives in the future.

Materials and methods

Materials

Whey protein isolate (WPI) (protein content ≥90%) was bought from the American Hilmar Cheese Company (USA). All reagents were of analytical grade.

Preparation of WPI-GA-E conjugates

WPI conjugates were prepared in accordance with previous studies, with a few modifications (Man et al., 2024). WPI and GA were added to 0%, 10%, 20%, 40%, and 60% (v/v) ethanol solutions, the pH was then adjusted to 9 using the 2 M NaOH solution, resulting in the final WPI concentration of 3% (w/v) and total phenolic content of 200 μmol L⁻¹. The resulting solution was stored at 4 °C and stirred continuously for 24 h to promote protein hydration. When the solution was back at room temperature, the protein mixture was subject to dialysis to remove free polyphenols, then freeze-dried to remove excess water completely.

Determination of total phenolic content

Conjugates were measured for their total phenolic content by Folin Ciocalteu method (Hao et al., 2022). Taken 1 mL of sample solution and mixed it evenly and fully with 2.5 mL of Folin Ciocalteu reagent. Stood in the dark for 3 min and 1 mL of 15% Na₂CO₃ solution was added, ensuring that it was completely dissolved in the solution. Finally, they were placed in the dark again and continued to react for 30 min. The absorbance value was then measured at a wavelength of 760 nm using the spectrophotometer (UV-2450, Shimadzu, Japan). Standard curves were plotted by using different concentrations of GA. The phenolic content of conjugates was determined by calculating from the standard curve:

where N represented the dilution factor and C denoted the concentration of conjugate.

Determination of free sulfhydryl (R-SH) and disulfide bond (R-S-S-R') groups content

The levels of total and free sulfhydryl (SH) groups in WPI-GA-E conjugates were determined by Ellman method (Zhao et al., 2012). Determination of free sulfhydryl groups: 1 mL of sample solution was added to 5 mL of Tris-Gly buffer solution, followed by 0.1 mL of Ellman's reagent. The mixture was then incubated at 25 °C for 15 min. The absorbance of the sample was measured at 412 nm. The solution without Ellman's reagent was used as a blank control. Determination of total sulfhydryl groups: 1 mL of sample solution was added to 5 mL of Tris-gly-8 M Urea-0.5% SDS buffer solution, followed by 0.1 mL of Ellman's reagent. The mixture was then incubated at 25 °C for 15 min. The absorbance of the sample was measured at 412 nm. The disulfide bond content was calculated by the content of total sulfhydryl and free sulfhydryl.

where A₄₁₂ is the absorbance at 412 nm, 73.53 is the conversion factor, and C is the sample concentration.

Ultraviolet absorption spectroscopy

Ultraviolet absorption spectroscopy was analysed by the following methods (Zhang et al., 2019). The UV absorption spectroscopy of conjugate solution (1.0 mg mL⁻¹) was recorded in the wavelength range of 200–360 nm by using the UV–vis spectrophotometer and the scan speed was set at medium.

Fourier-transform infrared spectra and secondary structure analysis (FTIR)

The alterations in the secondary structure of WPI-GA-E conjugates were detected by infrared spectroscopy (Yan & Zhou, 2021). The freeze-dried sample was mixed with KBr at 1:100, crushed and carefully pressed to make a thin layer, and then scanned with an infrared spectrophotometer (FTIR-8400S, Shimadzu, Japan) under the conditions of 400–4000 cm⁻¹ and 4 cm⁻¹ for a total of 16 times to acquire the infrared spectrum of the sample.

Intrinsic fluorescence spectra analysis

The WPI-GA-E conjugates were scanned with the fluorescence spectrophotometer (Lumina, Thermo Fisher Scientific, USA) to obtain the intrinsic fluorescence spectrum (Jia et al., 2022). The corresponding sample solution was diluted to 0.1 mg mL⁻¹, and the fluorescence intensity was measured. The instrument parameters were: the scanning emission spectrum was 300 ~ 400 nm, the scanning speed was 1200 nm min⁻¹, the voltage was 500 mV, and the excitation and emission slit width was 5 nm, respectively.

Measurement of surface hydrophobicity (H₀)

The H₀ of the WPI-GA-E conjugates were determined by the method of 1-anilinaphthalene-8-sulfonic acid (ANS) described (Han et al., 2023). The WPI-GA-E conjugate was diluted with phosphate buffers so that its final concentrations were 0.05, 0.1, 0.15, 0.20 and 0.25 mg mL⁻¹. Four millilitres of the WPI conjugate solution and 40 μL of 8 mM ANS solution were mixed together uniformly. The fluorescence intensity of the solution was measured by a fluorescence spectrophotometer at the excitation wavelength of 390 nm and the emission wavelength of 470 nm. The H₀ of conjugates calculated from the slope of the curve of fluorescence intensity and protein concentration.

Zeta potential, particle size measurement and polymer dispersity index(PDI)

Zeta potential, particle size, and PDI measurements were conducted using a Zetasizer Nano (ZEN3700, Malvern, Britain) (Xiong et al., 2016). The WPI-GA-E conjugates were diluted to a concentration of 0.1 mg mL⁻¹ in phosphate buffer prior to analysis in order to minimise the potential for multiple particle effects.

The functional properties of WPI-GA-E conjugates

Solubility

The solubility of the WPI-GA-E conjugates was determined following the method of Zhang et al. (2021) (Zhang et al., 2021). Weighed 1 g of the conjugate freeze-dried powder and dissolved it in 100 mL of NaOH solution with pH 9, the total protein content was the theoretical concentration of 1 g per 100 mL and then stirred with a magnetic field for 40 min, the supernatant was obtained after centrifugation for 15 min. The absorbance at 595 nm was determined by the Coomassie blue dye-binding method. The standard protein solution with different concentrations of bovine serum protein was prepared and the standard curve was drawn. The protein concentration of the conjugate was calculated by the standard protein concentration equation, and the result was compared with the theoretical concentration of the protein, and the solubility of the conjugate was obtained:

Emulsifying activity index (EAI) and emulsion stability index (ESI)

The EAI and ESI of the WPI-GA-E conjugates were conducted following the method (Xia et al., 2010). Dilute the sample to a concentration of 1 mg mL⁻¹ using a phosphate buffer solution. Then, 5 mL of the above solution and 15 mL of soybean oil were homogenised in a high-speed homogeniser at 13 600 r.p.m. for 2 min. Finally, samples are taken immediately. Used the spectrophotometer to measure the absorbance at 500 nm, that was A₀. After standing for 30 min, the absorbance was measured again, which was A₃₀:

where N represented the dilution ratio: 250, C denoted the protein concentration in the sample solution, and φ indicated the fraction of the oil phase: 0.25, T represented the time difference.

Foaming capacity (FC) and foam stability (FS)

The FC and FS of the conjugates were determined using the method (Wang et al., 2017). The samples (3 mg mL⁻¹, 30 mL) were homogenised at 17 600 r.p.m. for 2 min, and the height of the homogenised solution was recorded as V₀. After standing for 30 min, the height was recorded again as V₃₀:

where 30 represented the liquid level before homogenization.

Rheological properties

The rheological properties of the WPI-GA-E conjugates were determined by a rheometer (MCR102e, Anton Paar, Austria) based on the previous method (Tao et al., 2023). The WPI-GA-E conjugates were prepared with deionised water to 10 mg mL⁻¹. Take an appropriate amount of samples to be tested and put them on the rheometer platform. The solution to be tested was evenly dispersed on the rheometer platform. Selected a parallel plate with a diameter of 50 mm and set a gap of 0.5 mm. The relation between viscosity and shear rate of the sample was measured in the range of 1–100 s⁻¹.

Statistical analysis

All experiments were repeated three times. Univariate analysis of variance (Anova) and minimum significant difference (LSD) tests were used to test for homogeneity of variance, where P < 0.05 was considered statistically significant. Plotting was performed using Origin 2023.

Results and discussion

Total phenolic content

Gallic acid (GA) contains phenolic hydroxyl groups which are the reactive sites for binding to proteins (Dai et al., 2022). Ethanol unfolded the protein structure and broke the hydrogen bonds and hydrophobic bonds between polyphenols and gallic acid, so the total phenolic content decreased with the increase of ethanol concentration. However, a high concentration of ethanol caused protein unfolding or aggregate behaviour, it can not break the bond between protein and gallic acid, so the total phenolic content of WPI-GA-E-60 increased (Fig. 1).

Free sulfhydryl, total sulfhydryl content and disulfide bond content

Ethanol unfurled the protein structure and exposed the free sulfhydryl groups. The hydroxyl group of phenolic compounds is high activity, and the phenolic hydroxyl group can bind closely with the protein, make the protein cross-linked, gather a large number of disulfide bonds, and promote the stability of the protein structure (Man et al., 2024). Ethanol can break the binding of protein to gallic acid, and the rate of gallic acid binding to protein to form a disulfide bond is lower than the exposure rate of ethanol to the mercapto group. However, as the concentration of ethanol increased, the number of free sulfhydryl groups increased, and these sulfhydryl groups reacted with the hydroxyl group on the gallic acid to form disulfide bonds, so the number of disulfide bonds in the conjugate increased, while the number of free sulfhydryl groups decreased. The addition of ethanol accelerated the oxidation of cysteine, resulting in the reduction of protein total sulfhydryl content (Fig. 2).

Ultraviolet absorption spectrum analysis

As shown in Fig. 3, with the increase of ethanol concentration, the UV characteristic peak intensity increased first and then decreased. Ethanol caused the protein structure to unfold, exposing tryptophan residues buried inside the protein, and thus exhibiting the highest UV absorption intensity. However, when the ethanol concentration increases, these tryptophan residues react nucleophilise with polyphenols, causing the chromophore to embed in the chromophore of the protein, thereby reducing the UV absorption intensity (Singh et al., 2010).

Fourier transform infrared spectroscopy (FTIR)

FTIR is a widely utilised tool for studying the structure of biomolecules. It measures changes in the intensity of protein absorption or emission at different wavelengths, revealing key changes in their secondary structure, such as conformational changes like α-helix, β-folding, and random coiling. FTIR spectra of the WPI-GA-E conjugates are presented in Fig. 4. The WPI-GA-E conjugates had three bands close to 1530 cm⁻¹, 1650 cm⁻¹, and 3300 cm⁻¹, which were associated with N-H, C-O, and -OH stretching (Antony et al., 2022). Compared with the spectra of the WPI-GA-E-0, the absorption peaks related to amide A in the spectra of WPI-GA-E-20 and WPI-GA-E-40 moved from 3292.25 cm⁻¹ to 3300.71 cm⁻¹ and 3302.81 cm⁻¹, respectively, indicating that hydrogen bonds were formed in the conjugate. This finding was in line with the research consistent with Ke et al. (2023), where they observed that the peak of amide I experienced a redshift, suggesting a potential interaction between free amino groups and polyphenols. WPI-GA-E-20 and WPI-GA-E-40 had higher stability than other conjugates. As the concentration of ethanol increased, the amide A band shifted towards a lower wavelength, indicating the presence of an increased number of hydrogen bonds in the sample. This was the primary force responsible for maintaining the stability of the protein structure (Xin et al., 2023). In addition, covalent cross-linking of C-N or C-S bonds can be formed between proteins and polyphenols, and the amide II band of the WPI-GA-E conjugates moved from 1528 cm⁻¹ to 1532 cm⁻¹ and 1536 cm⁻¹, indicating that both N-H and C-N bonds participated in the reaction, indicating electrostatic and hydrophobic interactions between the complex (Li et al., 2024b).

Secondary structures of WPI-GA-E conjugates

This specific region of the amide I band (1600–1700 cm⁻¹) significantly influences the secondary structure of proteins. Bands at 1660–1695 cm⁻¹, 1640–1650 cm⁻¹, 1610–1640 cm⁻¹, and 1650–1660 cm⁻¹ represent β-sheets, random coils, α-helixes, and β-turns (Wang et al., 2020a). Table 1 shows the percentage of secondary structure content of WPI-GA-E conjugates.

The content of β-sheets and random coils in the WPI-GA-E conjugates were not significantly different (P >0.05). The α-helixes content increased first and then decreased when compared with the WPI-GA-E-0. The reason for the increase of α-helix content may be that the polarity of alcohols is lower than that of water, and the solution environment with low polarity is more conducive to the formation of intramolecular hydrogen bonds, similar to the results of Lei et al. (2021), the appropriate concentration of ethanol can effectively promote the formation of protein α-helix structure. The solubility of high concentration ethanol to polyphenols became poor, which changed the binding mode of polyphenols. Similar to the trend of Xia et al. (2010), they observed that the α-helical structure in egg yolk gel samples treated with 80% ethanol showed a gradual reduction trend and severe aggregation under high ethanol concentration conditions. Under the influence of high-concentration ethanol, the protein molecules undergo reassembly and form a dense granular structure.

The reason why WPI-GA-E-60 had the lowest α-helix content and the highest β-angle content was that α-helix and β-angle can be transformed into each other through the conformational change of protein molecules. The polarity of ethanol was low, and the environment of protein in low polarity was conducive to the formation of hydrogen bonds, which promoted the transformation of β-angle to α-helix structure.

Intrinsic fluorescence spectra analysis

Intrinsic fluorescence spectroscopy, primarily based on the presence of aromatic amino acid residues, is a conventional method for evaluating alterations in the tertiary structure of proteins (Zhao et al., 2022). Tryptophan is a protein that exhibits high sensitivity to the polarity of its microenvironment. By changing the microenvironment of amino acid residues, ethanol exposes tryptophan to the polarity environment (Zhou et al., 2016), which can cause protein aggregation and conformational change, and therefore change the fluorescence intensity (Zhong et al., 2023).

Ethanol treatment did not change the shape of the intrinsic fluorescence profiles of proteins. With the increase in ethanol concentration, the fluorescence intensity of the conjugate exhibited a pattern of initially increasing and then decreasing. The fluorescence intensity of WPI-GA-E-20 was higher than that of other groups, indicating that ethanol could open the structure of proteins and expose amino acid residues (Feng et al., 2021). However, when the concentration of ethanol is too high, it will stimulate the covalent reaction between aromatic groups and polyphenols, alter the spatial conformation of protein structure, and result in the coverage of chromophores, thereby reducing fluorescence intensity. Ethanol made the protein structure unfold, but WPI-GA-E-0 was not affected by ethanol, so the fluorescence intensity of WPI-GA-E-0 was lower than that of the ethanol treatment group (Fig. 5).

Surface hydrophobicity (H₀) analysis of WPI-GA-E

Surface hydrophobicity (H₀) can reveal the exposure of hydrophobic regions on the surface of proteins, providing a crucial scientific foundation for investigating protein conformation, folding mechanisms, and interface behaviour (Li et al., 2024a). As shown in Fig. 6, the surface hydrophobicity of the conjugate increases first and then decreases with the increase of ethanol concentration, which may be because ethanol causes protein molecules to unfold, and there are enough initial buried hydrophobic groups exposed to the aqueous phase, but excessive ethanol concentration can lead to the destruction of the protein's hydrophobic region, resulting in a decrease in its H₀ (Feng et al., 2021). In addition, polyphenols can bind to hydrophobic regions of proteins and occupy binding sites for ANS (Chandrapala et al., 2011). The introduction of gallic acid hydroxyl group can react with the groups (such as the amino group and sulfhydryl group) on the side chain of protein, thereby reducing the surface hydrophobicity (Ke et al., 2023).

Zeta potential and polymer dispersity index (PDI)

Zeta potential is a crucial parameter for measuring indexes to describe the stability of colloidal dispersions and surface charge of colloids in solution. Figure 7a shows that the WPI-GA-E conjugates were negatively charged. Ethanol opens the protein structure and exposes the internal sulfhydryl groups. The reaction of gallic acid with these sulfhydryl groups can change the surface charge distribution of proteins, and further affect the zeta potential of proteins. Ethanol concentration has a certain effect on zeta potential, but it is also limited by protein concentration, pH and other factors, and there is no obvious trend of zeta potential under different concentrations of ethanol.

As depicted in Fig. 7b, the concentration of ethanol had little effect on the Z-average of protein, mainly because the average size of protein particles was mainly affected by the properties and structure of the protein itself, while the concentration of ethanol mainly affected the solubility and precipitation behaviour of protein, rather than directly changing the size of protein particles.

Ethanol can change the polarity and permittivity of the solution, thus affecting the conformation and stability of the protein. Ethanol affects the folding state of the protein, resulting in changes in its physical properties and an increase in the polymer dispersity index (PDI). With the further increase of ethanol concentration, it causes more serious damage to the structure of protein, resulting in the decline of its stability, and thus the decrease of PDI.

Particle size distribution analysis

The results in Fig. 8 show that WPI-GA-E-0 was a single peak, and the WPI-GA-E conjugate changed from a single peak to a multi-peak after the addition of ethanol, indicating that the addition of ethanol enhanced the electrostatic repulsion between proteins, resulting in the inability of proteins to aggregate. Gallic acid is easily oxidised to form quinone, which reacts with amino acids on the side chain of protein, increasing the particle size of the conjugate, thereby changing the particle size distribution of the conjugate (Man et al., 2024). However, with the increase of ethanol concentration, the total phenolic content decreased, and the reaction with amino acids decreased, causing the conjugate to form bimodal or even multi-modal, which was similar to the results of Cheng et al. (2024), they found that particle size of soybean protein isolate (SPI) changed with the change of phlorotannins (PT) concentration, and SPI was a bimodal structure, and the peak structure changed from bimodal to unimodal when it was covalently combined with PT.

Functional properties of WPI-GA-E conjugates

Solubility analysis

Solubility plays a crucial role in the functionality of proteins in food applications. With the increase of ethanol concentration, the number of water molecules in the solvent decreases, and the solubility of water to protein molecules is weakened, and the increase of ethanol concentration changes its charge state and spatial structure, further causing protein aggregation, resulting in the solubility of conjugate decreases with the increase of ethanol concentration (Liu et al., 2024). In addition, the solubility of proteins is affected by many factors, the most important of which is the surface hydrophobicity of proteins. Previous experiments had measured that WPI-GA-E-0 had the lowest surface hydrophobicity and the least hydrophobic region on the surface of protein molecules, resulting in an increase in protein solubility, so the solubility of WPI-GA-E-0 was higher than that of other ethanol treatment groups (Fig. 9).

Emulsifying activity index (EAI) and emulsion stability index (ESI) analysis

Emulsifying activity index (EAI) reflects the ability of a protein to form a film at the oil–water interface, while emulsion stability index (ESI) measures the stability of the emulsion, with increases or decreases in EAI and ESI often occurring simultaneously, reflecting the overall properties of the protein during emulsification (Han et al., 2022). As shown in Fig. 10, with the increase of ethanol concentration, the EAI and ESI of the conjugate exhibited the trend of first increasing and then decreasing. This change is mainly due to the fact that ethanol makes the protein structure unfold, enhances the interfacial adsorption of the protein, and improves the EAI of the conjugate. However, high concentrations of ethanol can cause proteins unfolding, appear to aggregate and reduce the EAI of conjugates.

Foaming capacity (FC) and foam stability (FS) analysis

As can be seen from Fig. 11, FC and FS of WPI-GA-E-0 were lower than other groups (P < 0.05), and the interaction between ethanol and water significantly reduced the surface tension of the solution, promoting foam formation and stabilising on the liquid surface. Among them, the FC and FS of WPI-GA-E-20 were the highest, because they had a higher H₀. These hydrophobic groups enhanced the interfacial activity of WPI, which was more conducive to the formation of foam (Oboroceanu et al., 2014). However, under high concentration of ethanol, the formation of macromolecular polymers will produce a steric effect, resulting in the decline of FC and FS. This resembled research conducted by Hoon Lee et al. (2024), they found that the foaming ability of mealworm protein (ETMP) treated with ethanol was greatly affected by the ethanol concentration. Among them, ETMP-20 with the lowest H₀ had the lowest foam capacity, and ETMP-80 with the highest H₀ had the highest foam capacity. In addition, the interaction between polyphenols and proteins changed the interfacial properties of protein membranes, forming a stable interfacial membrane at the gas–liquid interface, the total phenolic content of WPI-GA-E-60 was higher than that of WPI-GA-E-40 in previous experiments. Therefore, the FC and FS of WPI-GA-E-60 are higher than that of WPI-GA-E-40.

Rheological analysis

Rheological properties play a crucial role in determining the viscosity, elasticity, and other characteristics of food raw materials (Ling et al., 2024). Figure 12 illustrated all conjugates exhibited shear rate-dependent thinning behaviour, mainly due to high speed shear leading to internal structural rupture (Wang et al., 2021). Due to ethanol's ability to reduce interfacial tension and intermolecular repulsion, the apparent viscosity of the conjugate decreased over the entire shear rate range as the ethanol concentration increased, except for WPI-GA-E-20, it showed higher readings than all the samples. Due to previous experiments, the surface hydrophobicity of WPI-GA-E-20 was the highest. More hydrophobic groups were exposed under this condition, and a three-dimensional network structure was formed between proteins, which reduced fluidity, that was, the connection between protein molecules was tighter and the apparent viscosity increased.

Conclusions

The results show that ethanol concentration can change the structural and functional properties of WPI-GA conjugate. Ethanol can open the structure of proteins, resulting in the total sulfhydryl content, solubility and zeta potential of conjugated compounds as a whole. With the increase of ethanol concentration, proteins appear to unfold or aggregate, leading to the decrease of total phenol content and the number of disulfide bonds in conjugates and then increase. The results of the fluorescence spectrophotometer showed that the internal fluorescence intensity and surface hydrophobicity of conjugated compounds increased under the treatment of ethanol. The results show that ethanol can improve the emulsification, foaming and foam stability of conjugates, and the 20% ethanol concentration has the greatest effect on the conjugates' functional properties. And the rheological property is also reflected in this concentration, the apparent viscosity is the highest. The ability to improve the functional properties of conjugates is slightly reduced at high concentrations of ethanol, so reasonable control of the effect of ethanol concentration on the structure and functional properties of proteins is a necessary condition for expanding its application as a food additive in the food industry.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (No. 31860431).

Author contributions

Ruihua Zhang: Writing – original draft; methodology; investigation; conceptualization. Mingyan Ai: Visualization; formal analysis. Shenghuizi Chen: Project administration; validation. Shuting Li: Formal analysis. Chunlan Zhang: Writing – review and editing. Zhiqiang Zhou: Funding acquisition; formal analysis. Jiankang Lu: Project administration; supervision; conceptualization; resources; writing – review and editing.

Ethical approval

Ethics approval was not required for this research.

Peer review

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/ijfs.17561.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References