Ultrasound-assisted enzymatic extraction of jujube (Ziziphus jujuba Mill.) polysaccharides: Extraction efficiency, antioxidant activity, and structure features

Graphical abstract

Abstract

This study investigated the effect of ultrasound-assisted enzymatic extraction (UAEE) on the extraction efficiency, antioxidant activity, and structural properties of jujube polysaccharide (JPS), with hot water extraction (HWE), ultrasound-assisted extraction (UAE), and enzymatic-assisted extraction (EAE) serving as controls. Optimal extraction conditions were determined through a multi-index weighted scoring method that comprehensively accounted for yield, duration, and antioxidant activity. Results demonstrated that the JPS yield obtained by UAEE at 22/33 kHz was 10.5 % to 16.3 % higher than those achieved by the other methods, significantly enhancing antioxidant activity. Monosaccharide composition analysis revealed that UAEE increased the content of key mono-sugars in JPS. Additionally, assessments of molecular weight distribution, zeta potential, and rheological properties showed that UAEE reduced the molecular weight and apparent viscosity of JPS, resulting in a looser structural configuration. These structural modifications were observed in scanning electron microscope (SEM) images, which revealed a filamentous branched morphology in JPS obtained through UAEE. Further observations using the atomic force microscope (AFM) indicated that the polysaccharide chains extracted by UAEE were shorter in length, lower in height, and free from aggregation.

1. Introduction

Jujube (Ziziphus jujuba Mill.), a mature fruit of the genus Ziziphus within the Rhamnaceae family, is widely cultivated in the provinces of Xinjiang, Hebei, Shandong, and Shanxi of China [1]. According to statistics from the Ministry of Agriculture, the cultivated area for jujube in China reached approximately 25,000 square kilometers in 2021, with a total yield of 7,345,300 tons, accounting for 98 % of the global production of jujube fruits [2]. As a food and medicinal product with homology, jujube is recognized for its substantial nutritional and medical values, including its ability to regulate immunity, nourish Qi and blood, and provide anti-aging benefits, all of which play a crucial role in healthcare and disease prevention [3]. These health benefits are mainly attributed to the abundant presence of bioactive substances in jujube, including polysaccharides, polyphenols, pentacyclic triterpenoids, flavonoids, and cyclic adenosine phosphate [4]. With the improvement of living standards and the expansion of the jujube market, a wide array of jujube-based products has emerged, encompassing dried jujube, jujube puree, and jujube wine. However, these products are primarily basic processed goods, while the development of more profound and more intensive jujube processing remains limited. Extracting and utilizing the bioactive components of jujube in functional food processing can significantly enhance its comprehensive utilization value. Numerous studies have demonstrated that polysaccharides extracted from jujube possess potent antioxidant activity and are closely associated with its pharmacological effects, such as antiviral, anti-tumor, and immune-enhancing properties. Thus, it is imperative to further explore the development and utilization of natural antioxidants, particularly jujube polysaccharides, to mitigate risks to human health [5], [6], [7], [8].

Selecting an appropriate extraction method is crucial for achieving high yield and quality jujube polysaccharides (JPS). The conventional hot water extraction (HWE) method is widely employed for JPS extraction due to its simplicity and cost-effectiveness. However, this method is still limited in some aspects. For instance, the extraction yield of intracellular polysaccharides from jujube using HWE is relatively low, with only a small fraction of the polysaccharides being extracted [9]. Typically, the yield of JPS extracted by HWE depends mainly on the extraction time and temperature. Prolonged high-temperature extraction can lead to polysaccharide degradation and a consequent reduction in their biological activity [10]. Besides, polysaccharides obtained through HWE often contain impurities, such as pigments and proteins, necessitating additional time for subsequent separation and purification. Enzymatic-assisted extraction (EAE) is another commonly employed method for JPS extraction, where enzymes such as cellulase or pectinase are utilized to break down the jujube cell wall, thereby facilitating the release of polysaccharides. EAE operates under mild reaction conditions, minimizing damage to the polysaccharide structure and producing high yield. However, the high cost of enzymes restricts their application in industrial production [11]. Consequently, an increasing number of researchers are exploring novel extraction methods for polysaccharides, including those assisted by physical fields such as ultrasound and microwave. Each extraction method has advantages and disadvantages, and combining multiple techniques has become a prominent trend in polysaccharide extraction to obtain more active polysaccharides and enhance extraction yield. The synergistic effects generated by combining the strengths of different extraction methods can effectively improve the extraction efficiency and bioactivity of the polysaccharides [8], [10]. Nevertheless, studies focusing on the physicochemical characteristics, antioxidant activity, and digestive properties of JPS obtained through combined extraction methods remain relatively scarce.

Ultrasound technology, known for its green, safe, and efficient nature, has been widely adopted across various food processing units, particularly in extraction. Compared to traditional chemical and biological methods, ultrasound-assisted extraction (UAE) offers continuous or simultaneous enhancement through multiple physicochemical mechanisms, including mechanical action, cavitation, and thermal effects [12]. During extraction, acoustic cavitation generates micro-steaming and micro-turbulence in the liquid medium, resulting in strong mechanical disturbances that intensify interparticle collisions, material decomposition, and localized rupture [13], [14]. On the other hand, the reduction in particle size induced by ultrasound promotes mass transfer and increases the contact surface between the solid and liquid phases, thereby facilitating the dissolution of target components within the sample matrix [15]. Liu et al. [6] utilized UAE to extract proteins from walnut dregs, observing a 7.3 % increase in protein yield and a 4.77 % increase in purity compared to alkaline extraction while reducing the extraction time by 50 % [5]. Similarly, Beaudor et al. [16] compared the efficiency of UAE and conventional extraction (CE) for recovering phenolics from spent coffee grounds (SCGs), finding that UAE, under optimal conditions, achieved over 83 % phenolic recovery from SCGs, with 30 % improvement over the conventional method [16]. Ultrasound-assisted enzymatic extraction (UAEE) is a conjunction method that applies ultrasound under mild enzymatic hydrolysis conditions, and the “addictive effect” of combining ultrasound with enzymes effectively hydrolyzes the cell walls of fruit and vegetable tissues to release target components while maintaining their structure and bioactivities, proving more effective than pure extraction methods. Saeed et al. [17] used UAEE to extract active ingredients (phenolics and flavonoids) from Gymnema sylvestre and obtained an extraction rate three times higher than that of the traditional maceration method [17]. Lin et al. [18] discovered that under optimal UAEE conditions, the yield of Shatian pomelo peel polysaccharides (StPP) reached 30.1310 %, with the antioxidant activity of the obtained StPP surpassing that obtained through the HWE method [18].

Hence, the main objective of this work was to evaluate the effects of different methods (HWE, UAE, EAE, and UAEE) on the extraction of polysaccharides from jujube fruits. The extraction yield, processing time, and antioxidant activities (including DPPH, ABTS, hydroxyl radical scavenging ability, and total reducing power) of the extracted JPS were assessed using a multiple-index weighted scoring method to determine the optimal extraction method. Changes in the antioxidant activity of JPS following simulated in vitro gastric and gastrointestinal digestion were examined to provide a fundamental basis for the subsequent development of oral products containing jujube polysaccharides. Additionally, the structural properties and thermal stability of JPS were also studied.

2. Materials and methods

2.1. Materials and chemicals

Fresh jujube (Zizyphus jujuba Mill., variety: Huizao) fruits were harvested and purchased from Xinjiang, China, and jujube pits were manually removed before experiments. DPPH (1,1-diphenyl-2-picrylhydrazyl), ABTS (2,2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid)), and pectinase were purchased from Macklin Biochemical Technology Co., Ltd (Shanghai, China). Other chemicals used in this study were all analytical grades purchased from Sinopharm (Shanghai, China).

2.2. Single-factor experiments

The appropriate amount of fresh jujube fruit was weighed and subjected to single-factor experiments to determine the extraction temperature, the type of enzyme used, the amount of enzyme added, and the solid–liquid ratio. The single-factor experimental conditions were 45 °C, 50 °C, 55 °C, 60 °C, 65 °C, and 70 °C for extraction temperature; cellulase, papain, pectinase, cellulase + pectinase (1:1), cellulase + papain (1:1), and pectinase + papain (1:1) for enzyme type selection; 0.4 %, 0.6 %, 0.8 %, 1 %, and 1.2 % for enzyme dosage; 1:5, 1:10, 1:15, 1:20, and 1:25 for solid–liquid ratio. Subsequent extraction experiments of JPS with four different methods were built on the conditions selected in the single-factor experiments.

2.3. Extraction of JPS by different methods

2.3.1. Hot water extraction

A certain weight of fresh jujube fruit was homogenized with deionized water (60 °C) at a ratio of 1:10 (w/v) in a food blender (JYL-C012, Joyoung Co., Ltd., China, rotation speed: 21000 rpm) for 1 min. Based on pre-tests, the temperature of the extracted water was set to 60 °C, and the homogenate was transferred to a beaker and placed in the water bath at 60 °C, stirring for polysaccharide extraction. After a period of time, the homogenate was centrifuged at 4000 rpm for 15 min at ambient temperature. The supernatant was collected, lyophilized, and labeled as JPS-HWE.

2.3.2. Ultrasound-assisted extraction

UAE extraction experiments were performed using a self-designed multi-frequency ultrasound device described in a previous study [19]. A certain weight of fresh jujube fruit was homogenized with deionized water (60 °C) at a ratio of 1:10 (w/v) in a food blender (JYL-C012, Joyoung Co., Ltd., China) for 1 min. The homogenate was transferred to a beaker and placed in the ultrasound chamber at 60 °C with stirring. The immersion depth of the beaker was 10–12 cm in the ultrasound chamber, which was in the middle of the chamber, to ensure that the homogenate received the ultrasound waves to the greatest extent. After a period of time, the homogenate was centrifuged at 4000 rpm for 15 min at ambient temperature. The supernatant was collected, lyophilized, and labeled as JPS-UAE.

2.3.3. Enzymatic-assisted extraction

A certain weight of fresh jujube fruit was homogenized with deionized water (60 °C) at a ratio of 1:10 (w/v) in a food blender (JYL-C012, Joyoung Co., Ltd., China) for 1 min. Based on pre-tests, 1 % (w/v) of pectinase was selected for the EAE extraction test. The beaker containing jujube homogenate and pectinase was placed in a water bath at 60 °C for polysaccharide extraction. After a period, the mixture solution was boiled in hot water (100 °C) for enzyme inactivation. Subsequently, the mixture was centrifuged at 4000 rpm for 15 min at ambient temperature, and the supernatant was lyophilized and labeled as JPS-EAE.

2.3.4. Ultrasound-assisted enzymatic extraction

UAEE extraction experiment used the same ultrasonic device in UAE. A certain weight of fresh jujube fruit was homogenized with deionized water (60 °C) at a ratio of 1:10 (w/v) in a food blender (JYL-C012, Joyoung Co., Ltd., China) for 1 min. Pectinase (1 %, w/v) was added to the homogenate, and the mixture was placed in the ultrasound chamber for polysaccharide extraction at 60 °C. Mono-frequency (22, 33, and 40 kHz), dual-frequency (22/33, 33/40, and 22/40 kHz), and tri-frequency (22/33/40 kHz) were selected to perform the UAEE experiment. The acoustic energy density of the 7 frequency modes was kept constant during the test. In detail, the power used for mono-frequency modes was the maximum output (300 W), and the power of each generator was set to half and one-third of the maximum power for dual- and tri-frequency modes, respectively. After the extraction experiment, extracts were centrifuged at 4000 rpm for 15 min at ambient temperature, and supernatants were lyophilized and labeled as JPS-UAEE 22 kHz, JPS-UAEE 33 kHz, JPS-UAEE 40 kHz, JPS-UAEE 22/33 kHz, JPS-UAEE 33/40 kHz, JPS-UAEE 22/40 kHz, JPS-UAEE 22/33/40 kHz, respectively.

All extraction tests (HWE, UAE, EAE, and UAEE) were conducted for 30 min and were repeated at least three times.

2.4. Extraction yield

The extraction yield of JPS obtained from different extraction methods was calculated as follows:

$$Extractionyield\left(\%\right)=\frac{M_1}{M_0}\times 100\%$$ (1)

where M₀ is the dry mass of jujube fruits, and M₁ is the weight of lyophilized crude polysaccharides.

2.5. Antioxidant activity assay

2.5.1. Determination of DPPH radical-scavenging assay

The DPPH radical-scavenging activity was measured using the method of Liu et al. [20] with some modifications [20]. An appropriate volume of JPS solution and 0.1 mM DPPH ethanol solution were mixed and kept dark at 37 °C for 30 min. The absorbance of the solution was measured at 517 nm, and absolute ethanol instead of JPS solution was set as the control. The DPPH scavenging activity was calculated as follows:

$$DPPHradicalscavengingactivity\left(\%\right)=\left(1-\frac{A_1}{A_0}\right)\times 100\%$$ (2)

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

2.5.2. Determination of ABTS radical-scavenging ability

The ABTS radical-scavenging activity was measured using the method of Wang et al. [21] with some modifications [21]. Appropriate ABTS powder was weighed accurately and mixed with deionized water to form a 7 mM ABTS stock solution. Then, the stock solution was mixed with 2.45 mM K₂S₂O₈ solution in equal volume and left the mixture in the dark for 12 h. The mixture was diluted with absolute ethanol to give an absorbance value of 0.7 ± 0.02 at 734 nm to obtain the ABTS working solution. Extracted JPS solution (2 mL) was mixed with ABTS working solution in equal volume in the dark at room temperature for 30 min. The absorbance was recorded at 734 nm, and the distilled water replaced the JPS solution was set as the control. The ABTS scavenging activity was calculated as follows:

$$\mathit{ABTSradicalscavengingactivity}(\%)=\left(1-\frac{A_1}{A_0}\right)\times 100\%$$ (3)

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

2.5.3. Determination of hydroxyl radical scavenging ability

The determination of hydroxyl radical scavenging ability was based on the method of Zhou et al. (2022) with some modifications[22]. An appropriate volume of JPS solution (2 mL) with a series of concentrations was mixed with 1 mL 9 mM FeSO₄ and 1 mL 9 mM salicylic acid–ethanol solution. Subsequently, 1 mL of freshly prepared H₂O₂ solution (8.8 mM) was added and left the mixture in the dark at 37 °C for 30 min. The absorbance was measured at 510 nm. The hydroxyl radical scavenging activity was calculated as follows:

$$\mathit{Hydroxylradicalscavengingactivity}(\%)=\left(1-\frac{A_1-A_2}{A_0}\right)\times 100\%$$ (4)

where A₁ is the absorbance of the JPS sample, and A₂ and A₀ are the absorbance values measured by replacing distilled water with H₂O₂ and the sample solution, respectively.

2.5.4. Determination of total reducing power

The determination method of total reducing power was referred to by Zhu et al. [23] and slightly modified [23]. An appropriate volume of JPS solution (1 mL) with different concentrations was prepared and mixed with 2.5 mL of 0.2 M phosphate buffer (pH 6.6) and 2.5 mL of 1 % (w/v) K₃Fe(CN)₆ solution. The mixture was reacted in a water bath at 50 °C for 20 min. After cooling, 10 % (v/v) trichloroacetic acid (TCA) was added to the above mixture and centrifuged at 3000 rpm for 10 min at room temperature. The supernatant was collected and taken 2 mL mixed with 2 mL deionized water and 0.4 mL 0.1 % (w/v) FeCl₃ solution. The mixture was homogenized well and reacted at room temperature for 10 min. The absorbance value was recorded at 700 nm, and distilled water replaced the sample as the control.

2.6. Jujube polysaccharide extraction level assay

A multi-index weighted comprehensive scoring method was selected to evaluate the extraction level of jujube polysaccharide obtained by different extraction methods, which was referred to the study of Liu et al. [6] with some modifications Liu et al. [6]. The polysaccharide yield, extraction time, and antioxidant activity of JPS were selected as the main indicators. According to the importance of each indicator, the yield, extraction time, and antioxidant activity were assigned with different weight coefficients, with the weight coefficient of yield set at 0.5, extraction time set at 0.1, and antioxidant activity set at 0.4. In detail, the JPS antioxidant activity of this study was evaluated by four methods, and the corresponding weight coefficient of each method was 0.1. The score for each experiment was calculated as follows:

$$\mathit{Score}=\left(\frac{A}{A_{\mathit{Max}}}\times 100\right)\times 0.5-\left(\frac{B}{B_{\mathit{Max}}}\times 100\right)\times 0.1-\left(\frac{C}{C_{\mathit{Max}}}\times 100\right)\times 0.1-\left(\frac{D}{D_{\mathit{Max}}}\times 100\right)\times 0.1-\left(\frac{E}{E_{\mathit{Max}}}\times 100\right)\times 0.1-\left(\frac{F}{F_{\mathit{Max}}}\times 100\right)\times 0.1$$ (5)

where A-E represent the indicators of JPS yield, extraction time, IC₅₀-DPPH, IC₅₀-ABTS, IC₅₀-hydroxyl radical, and EC₅₀ (total reducing power), respectively, and Max in the subscript is the maximum value of each indicator.

2.7. Simulated in vitro gastric and gastrointestinal digestion

The simulated in vitro gastric and gastrointestinal digestion of extracted JPS with different methods was carried out based on the study of Ketnawa et al. [24] with some modifications [24]. Simulated gastric digestion: The JPS solutions were preheated at 37 °C for 30 min and adjusted the pH value to 1.5 using 1.0 M HCl solution. Subsequently, 4 % (E/S) pepsin was added, and the mixture was incubated at 37 °C for 2 h in a shaker. After enzyme inactivation and cooling, the pH value of the above mixture was adjusted to 7.0 using 1.0 M NaOH and centrifuged at 12000 rpm for 10 min. The supernatant was collected as the gastric digestion of JPS. Simulated gastrointestinal digestion: the pH value of the obtained gastric digestion solution was adjusted to 7.5 using 1.0 M NaOH, and 2 % (E/S) pancreatin was added, and the mixture was incubated at 37 °C for 4 h. After enzyme inactivation and cooling, pH value of the above mixture was adjusted to 7.0 using 1.0 M NaOH and centrifuged at 12000 rpm for 10 min. The supernatant was collected as the gastrointestinal digestion of JPS.

The changes in antioxidant activity before and after digestion were evaluated by DPPH free radical scavenging capacity.

2.8. Purification of JPSs extracted by different methods

The obtained crude JPS solutions were evaporated to 1/5 of the original volume using a rotary evaporator (R-100, BUCHI Ltd), and subsequently, 4 times the volume of absolute ethanol was added and placed at 4℃ overnight for precipitation. Then, the solution was centrifuged at 5000 rpm for 15 min, and the precipitate was collected. The supernatant repeated the above steps several times and collected the precipitates. The collected and pooled precipitates were reconstituted with distilled water in a ratio of 1:2 (w/v). The solubilized JPS was mixed with Sevage reagent at 4:1 (v/v) 7 times to remove protein, then treated with 30 % H₂O₂ to decolorize. The obtained samples were dialyzed for 48 h using the boiled and cleaned dialysis bags (3500 Da), and the dialyzed water was changed every 12 h. The obtained dialysate was lyophilized for subsequent determination of JPS structural properties.

2.9. Determination of monosaccharide composition

The monosaccharide composition of JPS was determined by an HPLC with pre-column derivatization of the PMP method based on the study of Fang et al. [25] with minor modifications [25]. An aliquot (2.0 mg) of lyophilized JPS was hydrolyzed using 2 mL of 2 M trifluoroacetic acid (TFA) at 105℃ for 6h. The remaining TFA was blown by nitrogen gas after cooling to room temperature and washed with methanol until the TFA was removed entirely. The hydrolyzed samples and standard monosaccharides were derivatized and filtered with 0.45 μm aqueous membrane for measurement. A Zorbax Eclipse XDB-C18 column (4.6 mm × 250 mm, 10 μm, Agilent, USA) was used. The injection volume was 20 μL, and the column temperature was 25℃. Phosphate-buffered saline (0.05 M, pH 6.7) and acetonitrile at a flow rate of 1 mL/min were used as eluents. Mixtures of monosaccharides (rhamnose (Rha), arabinose (Ara), galactose (Gal), glucose (Glu), xylose (Xyl), mannose (Man), and glucuronic acid (GluA)) at a series of concentrations were used as external standards for sugar identification and quantitation. Three replications were conducted for each test.

2.10. Molecular weight and zeta-potential analysis

The weight-average molecular weight (Mw), number-average molecular weight (Mn), and molecular weight distribution (Mw/Mn) of JPS were determined using SEC–MALLS (DAWN HELEOS II, λ = 658 nm: Wyatt Technologies Corporation, USA) on an Agilent 1100 system equipped with two SEC columns (OHpak SB-806 M HQ and SB-805 HQ, Shodex, Japan) following the method of Yan et al. [26].

Zeta potentials of JPS solutions were measured using dynamic light scattering (DLS) on a Malvern Zetasizer, Nano ZS90 (Malvern Instruments Ltd, UK) at ambient temperature.

2.11. Fourier transform infrared (FT-IR) spectroscopic analysis

The FT-IR spectra of JPSs were recorded using a Nicolet IS50 FTIR spectrometer (Thermo Inc., Waltham, MA, USA) equipped with a deuterated triglycine sulfate detector. For all samples, 32 scans were performed at 4 cm⁻¹ resolution at 600–4000 cm⁻¹. The obtained spectrograms were analyzed by OMNIC (Thermo Inc.).

2.12. Thermogravimetric analysis

The thermal stability of JPSs was determined by a thermal gravimetric analyzer (STA 499C, Netzsch Inc., Germany). JPS samples were weighed to 5 mg, and the empty crucible was set as the reference. Measurements were conducted in a dynamic inert nitrogen atmosphere at 40 mL/min, and the linear heating rate was 10 ℃/min from 30 °C to 600 °C. All runs were performed in triplicate.

2.13. Rheological characteristics

The apparent viscosity of JPS solutions was measured using the method adopted by Yu et al. [27] with minor modifications [27]. An aliquot (15.0 mg) of JPS was dissolved in deionized water to prepare a 5 mg/mL solution. The solution was stirred well for 4 h at room temperature and placed at 4℃ overnight. Rheological analysis was conducted using a Discovery HR-1 hybrid rheometer (TA Instruments, New Castle, USA) with a parallel plate (40 mm diameter, 1 mm gap dimension) at room temperature.

2.14. Scanning electron microscopy (SEM)

The microstructure of JPS extracted by different methods was photographed using a Hitachi S-3400 N scanning electron microscope (Hitachi Inc., Tokyo, Japan). The details followed the method of Wu et al. [28].

2.15. Atomic force microscopy (AFM) analysis

The morphology observation of JPSs was carried out using a nano V electronic atomic force microscope (Bruker Inc., Germany) equipped with a Si3N4 probe in a tapping mode. In detail, 20 μL of JPS solution (10 μg/mL) was uniformly pipetted onto a freshly cleaved mica plate and was air-dried in a super-clean bench for 4 h at ambient temperature.

2.16. Statistical analysis

All experiments were performed at least in triplicate, and data were expressed as the mean ± standard deviation (SD). The differences in results were evaluated using a one-way analysis of variance (ANOVA) with Duncan’s multiple-range tests. All statistical tests were analyzed using SPSS 26.0 (SPSS Inc., USA) at the significance levels of p = 0.05.

3. Results and discussion

3.1. Results of single-factor experiments

Fig. 1 displays the impact of four selected factors on the extraction rate of jujube polysaccharides. The extraction rate increased with temperature, reaching its maximum at 60℃ (50.37 %). This enhancement may be attributed to the thermal disruption of the polysaccharide structure within the jujube cell wall, facilitating the release of JPS. However, further increasing the temperature from 60 °C to 70 °C showed no significant change in the extraction rate. Considering yield and energy consumption, a temperature of 60 °C was chosen for JPS extraction.

Fig. 1.

Effect of four selected factors on the extraction yield of jujube polysaccharides (A: Extraction temperature; B: Enzyme combination; C: Enzyme addition amount; D: Solid-liquid ratio) and extraction rate of jujube polysaccharides obtained under the optimal conditions (E: extraction rate; F: maximum yield).

The effects of cellulase, papain, pectinase and their combination on the polysaccharide extraction rate are presented in Fig. 1B. Cellulase and pectinase act on the cellulose and pectin in the cell wall of jujube fruits, respectively, disrupting the structural network and releasing polysaccharides. On the other hand, papain cleaves peptide bonds of proteins in the cell wall, further aiding in the release of JPS. Among the enzymes tested, pectinase working alone achieved the highest yield (51.61 %), significantly (p < 0.05) higher than that of other enzymes, either alone or in combination. This superior performance of pectinase is likely due to its ability to degrade pectin, an “adhesive” in the cell wall and middle lamella. By destabilizing the supporting structure of the jujube cell wall, pectinase enhances the release of JPS. Therefore, pectinase was selected for use in the EAE and UAEE experiments.

After determining the pectinase, Fig. 1C shows the effect of the enzyme dosage on the JPS extraction rate. The yield significantly (p < 0.05) increased as the enzyme dosage rose from 0.4 % to 1 %, plateauing thereafter at 1.2 %. This trend likely results from insufficient enzyme at dosages below 1 %, leading to incomplete degradation of the jujube cell wall. Conversely, excessive enzyme dosage may cause over-cleavage of the glycosidic bonds in polysaccharides, reducing the JPS extraction yield. Consequently, a pectinase dosage of 1 % was selected for subsequent experiments.

Once the extraction time, the type of enzyme used, and the enzyme dosage were determined, the solid–liquid ratio of jujube fruit to water was screened in Fig. 1D. As shown, the polysaccharides yield increased initially then decreased with the solid–liquid ratio, reaching the maximum value (51.34 %) at the ratio of 1:10. This trend may be related to excessive solvent (water) leading to a decrease in substrate concentration, thus enlarging the concentration gradient and adversely affecting JPS solubilization. Hence, a solid–liquid ratio of 1:10 was selected for this study.

3.2. Effect of different extraction methods on the yield of JPS

Fig. 1E exhibits the changes in polysaccharide yield against extraction time using HWE, UAE, EAE, and UAEE methods. Under the tested conditions, the JPS yield initially increased but decreased after reaching a critical point. This decline could be attributed to the breakage of glycosidic bonds within JPS, causing polysaccharide decomposition and a consequent reduction in yield [29]. Compared to the HWE method, employing ultrasonic waves significantly enhanced polysaccharide extraction, particularly when combined with pectinase, indicating the synergistic effect of ultrasound and enzyme in this process.

Fig. 1F is plotted by the maximum yield in Fig. 1E, highlighting significant differences in JPS yield among extraction methods, with HWE yielding the lowest, followed by UAE. The polysaccharide yield obtained by UAE increased by 5.25 % compared to HWE, attributed to the cavitation and mechanical effects of ultrasonication promoting polysaccharide precipitation [30]. UAEE and EAE methods produced significantly higher JPS yields than HWE and UAE, with varying results depending on the selected frequency modes. Among tested frequency modes, mono-frequency at 22 kHz (72.8 %) and 33 kHz (72.5 %) achieved the highest JPS yields, showing a 23.69 % increase over HWE and 7.27 % over EAE. Ultrasound likely enhanced extraction by directly disrupting the polysaccharide network within the jujube fruit cell wall, loosening the structure for easier extraction. Additionally, ultrasonication improved the interaction between pectinase and substrates, facilitating jujube polysaccharide degradation. The synergistic effect of ultrasound and enzyme boosted the extraction of polysaccharides from jujube fruits. This finding aligns with a study by Yang et al. [31], where UAEE (with α-amylase) increased rice protein extraction by 147.79 % compared to the alkaline control, also due to the synergistic effect of ultrasound and enzyme [31].

No significant difference in JPS extraction rates was observed under mono-frequency (40 kHz), dual-frequency (22/33, 22/40, and 33/40 kHz), and tri-frequency (22/33/40 kHz) compared to EAE. This absence of difference may result from the simultaneous frequencies negatively impacting the pectinase structure, thus reducing extraction efficiency. In a previous study, we investigated the inactivation of horseradish peroxidase (HRP) using the same multi-frequency ultrasound device and found that dual- and tri-frequency modes were more effective than mono-frequency modes for inactivating HRP [32]. Similarly, Zhang et al. [33] reported that dual- and tri-frequency modes generated higher noise signal intensity and cavitation bubble pulsation compared to mono-frequency, using a polyvinylidene fluoride (PVDF) sensor [33]. The lower JPS extraction rate under dual- and tri-frequency modes compared to EAE could be due to the enhanced cavitation effect causing polysaccharide degradation during UAEE.

3.3. Antioxidant activity of JPS

The scavenging ability of JPSs extracted by different methods on DPPH free radicals is shown in Fig. 2A, indicating that all JPS samples exhibited substantial DPPH radical scavenging ability, following a concentration-dependent trend. The scavenging ability was stabilized at a JPS concentration of 6 mg/mL, reaching a maximum capacity of 77.98 %. The IC₅₀ values (Fig. 2B) ranged from 3.65 ± 0.05 to 5.68 ± 0.21 mg/mL, signifying the concentration at which 50 % of the DPPH radicals were scavenged. Notably, JPSs extracted using the 22 kHz, 22/33 kHz, and 22/40 kHz UAEE methods significantly lowered the IC₅₀ value than other samples, with the possible reason being that UAEE under these frequency modes induced more severe degradation of JPS during extraction, altering its molecular weight, thus enhancing its hydrogen ion donation ability and increasing its DPPH scavenging efficiency. Fig. 2C illustrates the ABTS radical scavenging capacity of the extracted JPSs. A significant increase in scavenging ability was observed with increasing JPS concentrations from 0.2 to 1.0 mg/mL, with the JPS extracted by the 22/33 kHz UAEE method exhibiting the highest ABTS scavenging capacity of 91.82 %, which is 19.53 % higher than the HWE method. The IC₅₀ values (Fig. 2D) further confirm that JPSs obtained using the 40 kHz, 22/33 kHz, and 33/40 kHz UAEE methods showed the highest ABTS radical scavenging ability.

Fig. 2.

The scavenging ability and IC₅₀/EC₅₀ values of jujube polysaccharides extracted by different methods (A and B: DPPH; C and D: ABTS; E and F: Hydroxyl radicals; G and H: reducing power).

Hydroxyl radicals are a typical reactive oxygen species (ROS) produced by the human body, which can react with various biomolecules in cells, leading to apoptosis or organ damage [34]. Similar to the DPPH and ABTS assays, the scavenging rate of hydroxyl radicals gradually increased with rising sample concentrations. At 12 mg/mL, JPS obtained from 22/33 kHz UAEE presented the highest scavenging capacity (66.85 ± 0.99 %), 29.88 % higher than HWE. Consistent with the ABTS results, the IC₅₀ values (Fig. 2F) demonstrated that JPSs obtained from 40 kHz, 22/33 kHz, and 33/40 kHz UAEE showed superior hydroxyl radical scavenging ability. The reducing power of polysaccharides correlates with their antioxidant activity [10]. As depicted in Fig. 2G, the reducing power of JPS extracted by various methods increased with JPS concentrations ranging from 2 to 10 mg/mL, with UAEE-extracted samples rendering the highest reducing power. At a concentration of 10 mg/mL, JPS obtained from 22/33 kHz UAEE showed significantly higher absorbance compared to other samples, followed by 22/40 kHz and 22 kHz, all surpassing the HWE method. The EC₅₀ value, representing the JPS concentration corresponding to an absorbance of 0.5, further demonstrated that JPS extracted by the 22 kHz, 22/33 kHz, and 22/40 kHz UAEE methods possessed potent reducing power.

Taken together, JPS obtained by UAEE presented higher antioxidant activity compared to traditional HWE, UAE, and EAE methods, especially at the dual frequency of 22/33 kHz, probably due to the synergistic effect of pectinase and ultrasonication at this frequency, which positively influenced the structure of the jujube polysaccharides.

3.4. Evaluation of JPS extraction level

A multi-index weighted comprehensive scoring method was adopted to assess the efficacy of four extraction methods for jujube polysaccharides. Different weighting coefficients were assigned to experimental indicators (yield, duration to maximal extraction yield, and four antioxidant activities) to reflect their importance. A higher calculated score denotes that the extraction method is more suitable for JPS. This purpose-driven multi-index weighted scoring method effectively mitigates the one-sided impact of single factors on experimental results, rendering the results more comprehensive and systematic [6], [35]. The weighted score values of each index for the different extraction methods are shown in Table 1. Notably, the JPS yield and its antioxidant activity did not exhibit a consistent pattern. For example, while the highest polysaccharide yield was achieved with 22 kHz UAEE, its corresponding antioxidant activity was lower than that of other methods. This discrepancy may be due to factors influencing antioxidant capacity, such as molecular weight and chain conformation [36]. This underscores that it is unreasonable to determine the optimal method and conditions for JPS with high antioxidant activity based solely on one index (e.g., yield). Therefore, the experimental results were comprehensively analyzed by assigning different weighting coefficients, which revealed that the 22/33 kHz UAEE method attained the highest score of 15.0597, followed by the 22 kHz and 22/40 kHz UAEE methods, with scores of 11.7327 and 10.5163, respectively. In contrast, the HWE method received the lowest score. This alignment with the experimental results suggests that the polysaccharides obtained by UAEE at 22/33 kHz exhibited a higher extraction yield and maintained better antioxidant activity.

Table 1.

Multi-index weighted comprehensive scores of jujube polysaccharides under different extraction methods.

Extraction methods	Yield (%)	Duration (min)	IC50-DPPH (mg/mL)	IC50-ABTS (mg/mL)	IC50-Hydroxyl radicals (mg/mL)	EC50 (mg/mL)	Scores
JPS-HWE	58.64	15	5.24	0.59	11.32	7.53	−3.90
JPS-UAE	61.73	20	5.31	0.41	11.24	7.80	−1.55
JPS-EAE	67.90	20	5.17	0.54	11.79	9.20	−1.18
JPS-UAEE 22 kHz	72.84	15	3.77	0.52	9.92	6.74	11.73
JPS-UAEE 33 kHz	72.53	15	5.40	0.57	10.73	8.44	5.18
JPS-UAEE 40 kHz	67.59	15	4.64	0.43	12.24	8.84	3.92
JPS-UAEE 22/33 kHz	68.21	10	3.65	0.44	7.86	6.00	15.06
JPS-UAEE 33/40 kHz	68.83	10	5.68	0.44	8.64	7.59	9.42
JPS-UAEE 22/40 kHz	67.28	10	3.84	0.59	8.38	6.55	10.52
JPS-UAEE 22/33/40 kHz	66.05	15	4.85	0.55	10.58	7.61	3.16

3.5. Effect of simulated in vitro gastric and gastrointestinal digestion on the antioxidant activity of JPS

To further explore the alterations in the antioxidant activity of JPS, a simulated gastrointestinal digestion was conducted to evaluate its efficacy and provide a theoretical basis for applying and developing antioxidant food products. As shown in Fig. 3A, the DPPH radical scavenging rate of JPS significantly increased after digestion with gastric fluid compared to pre-digestion samples, with the substantial increase observed in JPS obtained by the UAEE method, showing an improvement of 8.6 % to 18.9 %. This enhancement is likely due to the acidic pH environment, which disrupts the molecular chains of polysaccharides, leading to a reduction in molecular weight and an increase in reducing ends, contributing to increased antioxidant activity. This result is in line with the findings of Chen et al. [37], who reported a significant decrease in the molecular weight of mulberry polysaccharides following simulated in vitro gastrointestinal digestion [37]. Notably, there was no significant difference in antioxidant activity among JPSs obtained through different frequency modes of the UAEE method. Besides, JPSs subjected to intestinal fluid digestion showed no significant differences in antioxidant activity compared to gastric fluid digestion, except for the 22 kHz UAEE sample.

Fig. 3.

DPPH scavenging ability of jujube polysaccharides extracted by different methods after simulated gastrointestinal digestion (A), FT-IR spectra (B), apparent viscosity (C), and thermogravimetric curves (D) of jujube polysaccharides extracted by different methods (Lowercase, uppercase, and lowercase letters with subscript in Fig. 3A represent the statistical analysis of the three groups (before digestion, after digestion with gastric fluid, and after gastrointestinal digestion); * represent significant differences between the gastric fluid digestion and gastrointestinal digestion groups).

3.6. Monosaccharide composition

The monosaccharide composition of JPS obtained through various extraction methods is detained in Table 2. All seven tested monosaccharides were present in the JPS extracted by different methods, confirming their heteropolysaccharide nature. Among these, Gal and Ara were the predominant monosaccharide units of JPS, followed by Rha. It is hypothesized that the structure of JPS is primarily composed of arabinogalactan, which is consistent with previous studies by Ji et al. [4], who also identified Ara and Gal as major components of jujube polysaccharides, albeit in differing proportions. These variations are likely due to differences in geographical origin, jujube fruit variety, and extraction methodologies. As shown in Table 2, the contents of Gal, Ara, and Rha in JPS were significantly increased by the 22/33 kHz UAEE method compared to other extraction techniques. This increase can be attributed to the cavitation effect induced by ultrasonication, which, in conjunction with the pectinase hydrolysis of the UAEE method, disrupted the network structure of the jujube cell wall and improved the efficiency of the enzymatic reaction, significantly enhancing the extraction yield and purity of JPS, which reflected on the elevated contents of the main monosaccharides.

Table 2.

Monosaccharide composition of jujube polysaccharides extracted by different methods.

Extraction methods	Man (mg/mL)	Rha (mg/mL)	Glu (mg/mL)	Gal (mg/mL)	Xyl (mg/mL)	Ara (mg/mL)	GluA (mg/mL)
JPS-HWE	4.36 ± 0.08b	26.74 ± 1.02ab	25.06 ± 0.53c	57.24 ± 0.93b	8.13 ± 1.36b	44.12 ± 0.20a	3.11 ± 0.19a
JPS-UAE	4.26 ± 0.01b	23.94 ± 0.93a	19.62 ± 0.70b	52.59 ± 1.45a	6.97 ± 1.19b	42.54 ± 0.72a	2.74 ± 0.22a
JPS-EAE	3.35 ± 0.02a	28.55 ± 1.41bc	13.26 ± 0.69a	52.90 ± 1.92a	0.37 ± 0.07a	53.52 ± 2.91b	3.07 ± 0.19a
JPS-UAEE 22/33 kHz	4.22 ± 0.09b	31.31 ± 0.43c	14.09 ± 0.35a	61.29 ± 1.00c	0.19 ± 0.01a	58.53 ± 0.72c	4.07 ± 0.11b

Note: Values in the same column with different letters differ significantly (p < 0.05; one-way ANOVA followed by Duncan’s test).

3.7. Molecular weight and zeta-potential

The molecular weight (Mw), number-average molecular weight (Mn), and Mw/Mn ratio of JPSs are shown in Table 3. The table reveals that the Mw of JPS varied depending on the extraction methods. Specifically, the Mw and Mn of JPS extracted via HWE were 7.46 × 10⁵ and 7.51 × 10⁵, respectively, whereas these values decreased to 5.60 × 10⁵ and 6.00 × 10⁵, respectively, for JPS obtained through UAE. This reduction suggests that the polysaccharides extracted by UAE had a smaller molecular weight, probably due to the sonication-induced cavitation and mechanical effects destroying the spatial structure of JPS, thus reducing the molecular weight [38]. Moreover, the Mw of JPS obtained by EAE and UAEE was significantly lower than that obtained by HWE, possibly because of the synergistic effect of ultrasound and pectinase, which caused more extensive damage to the spatial structure of JPS and the cleavage of intramolecular and intermolecular glycosidic bonds. Generally, a higher molecular weight of plant polysaccharides represents a more complex structure and richer functional activity, while the increased viscosity of such polysaccharides can limit their application. Conversely, plant polysaccharides with relatively low molecular weights offer advantages in bioactivity owing to faster mass transfer and higher bioavailability. For instance, Wang et al. [39] found that okra polysaccharides obtained through a three-phase separation method, which included low molecular weight polysaccharides, exhibited superior in vitro antioxidant capacity [39]. This could also explain the high antioxidant activity observed in JPS extracted by UAEE in this study. The polydispersity coefficient, represented by the Mw/Mn ratio, reflects the degree of molecular weight distribution; a lower Mw/Mn ratio indicates a more uniform molecular weight distribution. According to Hwang et al. [40], polysaccharides with a high Mw and an Mw/Mn of less than 2 have good homogeneity and are less prone to aggregation [40]. In this study, the Mw/Mn ratios of JPS were all below 2, with no significant differences observed in the polydispersity coefficient across the different extraction methods.

Table 3.

Molecular weight and zeta potential of jujube polysaccharides extracted by different methods.

Extraction methods	Mn (×105 Da)	Mw (×105 Da)	Mw/Mn	Zeta potential (mv)
JPS-HWE	7.460(±0.244 %)	7.511(±0.266 %)	1.007(±0.361 %)	−30.77 ± 3.27c
JPS-UAE	5.999(±0.166 %)	6.003(±0.106 %)	1.001(±0.234 %)	−24.56 ± 1.87b
JPS-EAE	1.003(±0.578 %)	1.065(±0.713 %)	1.061(±0.918 %)	−29.86 ± 0.77c
JPS-UAEE 22/33 kHz	1.217(±0.226 %)	1.234(±0.259 %)	1.014(±0.344 %)	−16.23 ± 0.94a

Note: Values in the same column with different letters differ significantly (p < 0.05; one-way ANOVA followed by Duncan’s test).

The zeta potential of JPS obtained through different extraction methods is depicted in Table 3, with all samples displaying a dominance of negative charges. Compared to HWE, the absolute value of the zeta potential of JPS was significantly reduced in samples extracted via UAE and UAEE, with UAEE exhibiting a greater reduction. It indicates that JPS became looser packed, leading to a more stable solution under the synergistic effect of ultrasound and pectinase, which agrees with the observed Mw results. Furthermore, according to Fang et al. [25], the absolute value of the zeta potential of polysaccharide solution was positively correlated with their uronic acid content [25]. Our findings supported this observation, indicating that JPS extracted by UAEE had the highest uric acid content.

3.8. FT-IR spectroscopy

The FT-IR spectra of JPSs, as shown in Fig. 3B, reveal similar characteristic absorption peaks across various extraction methods, indicative of a consistent preliminary structure of JPS. Specifically, the broad absorption peak at 3321 cm⁻¹ corresponded to the stretching vibration of the O-H bond, which was broadened due to the increase in intramolecular and intermolecular hydrogen bonding. The peak at 2935 cm⁻¹ was related to the stretching vibration of C-H bonds, including CH, CH₂, and CH₃. The absorption peaks observed at 1602 cm⁻¹ and 1412 cm⁻¹ were attributed to the C=O stretching and C-C bending, respectively. Additionally, the peaks at 1156 cm⁻¹ and 1015 cm⁻¹ accorded with the C-O-H and C-O-C glycosidic bonds of the pyranose ring [41]. Besides, the peak at 890 cm⁻¹ might refer to the presence of β-glycosidic bonds in the JPSs [38].

3.9. Apparent viscosity behavior

As depicted in Fig. 3C, the apparent viscosity of JPSs extracted by different methods decreased with increasing shear rate, displaying pseudo-plastic fluid behavior with shear-thinning tendencies. This trend is mainly because of the disruption of polysaccharide molecule entanglements under heightened shear force. Once a critical shear rate is reached, the apparent viscosity of JPS stabilizes, exhibiting Newtonian fluid behavior. This may result from the shear-induced breakage of inter- and intramolecular glycosidic bonds, weakening intermolecular associations [42]. Among the extraction methods, JPS obtained by HWE exhibited the highest apparent viscosity at equivalent shear rates, followed by UAE and EAE. Conversely, JPS extracted using the UAEE at 22/33 kHz showed the lowest viscosity. This finding is consistent with Yu et al. [27], who reported that apparent viscosity may correlate with molecular weight, particle size, and specific surface area [27]. As observed in this study, the reduced molecular weight of JPS obtained by UAEE suggests enhanced solubility and fluidity, making it particularly suitable for applications in the food and biomedical industries.

3.10. Thermogravimetric analysis

The thermogravimetric (TG) curves (Fig. 3D) recorded the weight loss of JPS extracted by different methods during the heating process, displaying a consistent three-stage degradation pattern. During heating, JPS underwent successive phases, including water evaporation, glycosidic and hydrogen bond breaking, structural skeleton decomposition, and transformation of residual substances, all contributing to the gradual reduction in mass. The initial mass loss (ca. 20 %) occurred between 30 and 150℃, primarily due to the volatilization of both free and bound water within the JPS [43]. Subsequently, the prominent degradation appeared in the second stage (150-500℃), with a polysaccharide mass loss of about 68 %, largely attributed to the breakage of linkages and bonds and the degradations of thermally sensitive functional groups within JPS, including the cleavage of C-C and C-O bonds [25]. As the temperature rose to 600℃, the mass loss of JPS gradually decreased, with the order of residual mass being UAEE (10.4 %) > EAE>UAE>HWE (5.59 %). Although the four extraction methods exhibited slight variations in mass loss and decomposition rates, all demonstrated good thermal stability.

3.11. Microstructure observation

The surface morphology of JPS extracted by different methods is illustrated in Fig. 4. JPS extracted by HWE presented a flake-like structure with a smooth surface characterized by interconnections or aggregates. Conversely, JPS obtained through the UAE method showed a mesh-like morphology with pores of varying sizes on the surface, along with instances of rupture, likely induced by the cavitation and mechanical effects of ultrasonication, which cause molecular destruction in JPS (D. [44]. This phenomenon aligns with the findings by Yan et al. [45], who observed similar results in polysaccharides derived from Phellinus linteus mycelia treated with ultrasound [45]. The surface morphology of EAE-extracted JPS showed no noticeable pores but exhibited fractured and branched flake-like structures, likely due to the enzymolysis of pectinase on the jujube fruit cell wall. Compared to HWE, the surface structure of JPS extracted by 22/33 kHz UAEE had obvious differences. The synergistic action of ultrasonication and enzyme led to the disruption of intra- and intermolecular glycosidic bonds, breaking down the lamellar structure and aggregates of JPS into small segments with filamentous branching, presenting a loose reticulate morphology. This observation is consistent with the molecular weight and zeta potential mentioned above.

Fig. 4.

Scanning electron micrographs and AFM chromatograms of jujube polysaccharides extracted by different methods.

3.12. AFM observation

JPSs extracted using different methods were observed using AFM to gain deeper insights into the chain structure. As shown in Fig. 4, polysaccharides obtained via HWE predominantly exhibited long, straight chains with multiple flexible branches, which were locally aggregated, with a molecular height close to 2.7 nm. JPS extracted through UAE showed breakage on the long straight chains, resulting in a reduced molecular height of 1.2 nm. In contrast, the molecular height of JPS extracted by EAE was similar to HWE but exhibited noticeable depolymerization, resulting in a highly branched structure. After UAEE extraction by UAEE at 22/33 kHz, the JPS structure appeared looser than that of UAE, with long straight chains splitting into short segments. The aggregation phenomenon was eliminated, and the molecular height decreased. This change is primarily attributed to the synergistic effects of sonication and enzyme action, which cleaved glycosidic bonds and intermolecular linkages, resulting in the fracture of long straight chains of JPS. Meanwhile, this process restricted the aggregation of glycan chains and prevented the formation of entanglements. Hua et al. [38] reported similar findings when treating lentinan with ultrasound at different frequencies, observing a transformation of the polysaccharide structure into short-branched chains after sonication, which is consistent with our results [38].

4. Conclusions

In this study, we comparatively investigated the effects of different extraction methods on the extraction yield and antioxidant activity of JPS, with optimal conditions determined using a multiple-index weighted scoring method that considered the extraction yield, duration, and antioxidant activities. The structural properties of the obtained JPS were also examined. The results showed that the 22/33 kHz UAEE method achieved the highest comprehensive score. Compared to HWE, UAEE significantly increased the extraction yield of JPS by 16.32 %, with the extracted samples exhibiting good antioxidant activity. Even after in vitro digestion simulation, the JPS retained high antioxidant activity. Monosaccharide composition and content analyses showed that while the UAEE method had no evident impact on the monosaccharide composition, it notably increased the contents of the dominant sugar units (Gal, Ara, and Rha). A comprehensive analysis of molecular weight, zeta potential, and apparent viscosity displayed that the UAEE method significantly decreased the Mw of obtained JPS and improved the fluidity of the JPS solution, leading to increased solubility and reduced apparent viscosity. The various extraction methods had no remarkable effect on the functional group structure and thermal stability of JPS. Further, SME graphs verified these findings, showing that the flake-like morphology of UAEE-extracted JPS was fragmented into small segments with filamentous branches, presenting a loose reticular structure. AFM analysis further indicated that the chain structure of JPS obtained by UAEE was shortened, with a decreased molecular height and no obvious aggregate formation. These results suggest that UAEE can be a potent technique for polysaccharide extraction from jujube fruits, enhancing extraction yield while preserving antioxidant activity. Implementing this technique in industrial production could facilitate the advanced processing of jujube fruits and foster new product development, significantly increasing industrial benefits.

CRediT authorship contribution statement

Yiting Guo: Writing – review & editing, Writing – original draft, Methodology, Funding acquisition, Formal analysis, Conceptualization. Shenao Nan: Writing – original draft, Validation, Investigation, Formal analysis. Chengcheng Qiu: Writing – original draft, Validation, Investigation, Data curation. Chenyu Song: Validation, Investigation. Bengang Wu: Writing – review & editing, Methodology, Funding acquisition, Conceptualization. Yanhua Tang: Writing – review & editing, Visualization, Resources. Lifang Cheng: Visualization, Resources. Haile Ma: Supervision, Project administration.

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.