Extraction method shapes the structure and fermentation behavior of white kidney bean episperm polysaccharides in gut microbiota of the elderly Authors: Yimei Hu, Xinyang Li, Caiping Duan, Yanbo Wang, Bei Wang, Chunli Kong, Paul de Vos Published: 2026 ABSTRACT AbstractAging is often accompanied by gut microbiota dysbiosis. Dietary supplementation with polysaccharides offers a promising, cost-effective approach to restore microbial balance and support healthy aging. This study investigated structural characteristics and in vitro fermentation behavior of white kidney bean episperm polysaccharides extracted using ultrasound-assisted acid (AEP), enzyme (EEP), and chelator (CEP) methods. Extraction method markedly influenced structure: AEP and CEP were enriched in xylose and arabinose, while EEP was rich in glucose and xylose. Morphologically, AEP exhibited a compact, porous structure. Fermentation profiles revealed that AEP notably enriched Bifidobacterium and produced the highest levels of butyric acid. Xylose in AEP supported sustained carbohydrate availability throughout fermentation, while CEP was rapidly degraded. All fractions increased total SCFAs and reduced bSCFAs. These findings suggest that white kidney bean episperm polysaccharides, particularly AEP, have strong prebiotic potential for modulating elderly gut microbiota and may serve as a promising ingredient targeting age-related microbial dysbiosis.Keywords: White kidney bean, Extraction, Episperm polysaccharides, Elderly, Gut microbiota Introduction As global aging accelerates, the number of elderly is estimated to reach 1.4 billion by 2030. In recent years, common health conditions such as increased intestinal permeability, chronic inflammation, neurodegenerative disorders, and metabolic syndrome have been associated with gut microbiota dysbiosis in elderly populations. This issue has attracted global attention due to its potential to substantially escalate both public health challenges and economic burdens (Ghosh et al., 2022). Numerous studies have shown that gut microbiota undergoes significant changes with aging, typically characterized by low abundance of beneficial bacteria like Bifidobacterium and high abundance of pathogens such as Enterobacteriaceae (Qi et al., 2023). Dietary supplement with prebiotics is one promising approach to regulate gut microbiota composition and metabolites, thereby improving overall health status of the elderly. Polysaccharides, as an important type of prebiotics, have been widely recognized to stimulate gut microbiota growth and activity of the elderly. For example, supplement of Oudemansiella raphanipes polysaccharides was proved to increase the abundance of beneficial bacteria Dubosiella, Clostridiales, and Prevotellaceae, and decrease the abundance of harmful bacteria Desulfovibrio. Such effects were resulted from microbial metabolites SCFAs, with benefits to strengthen age-related gut barrier dysfunction and cognitive decline (Ren et al., 2024). Similarly, Codonopsis pilosula polysaccharides could repair intestine oxidative and inflammatory stress and liver damage induced by aging, which were achieved through a so-called gut microbiota associated enterohepatic axis (Zou et al., 2023). However, these beneficial effects of polysaccharides were highly dependent on their chemical structures, such as molecular weight, monosaccharides composition, esterification degree, etc. (Zheng, Li, et al., 2024). Previous studies have shown that both the botanical source and the extraction method are critical determinants of the chemical structure and resulting biological activity of polysaccharides. These structural differences can influence their prebiotic potential, particularly in elderly populations where modulation of the gut microbiota may help counteract age-associated dysbiosis and its related health consequences. As reported, polysaccharides from the peel and stem of different varieties of pitaya differed in monosaccharides composition, molecular weight, and surface morphology. These structural differences might induce their variations in antioxidant capacity (Zheng, Pi, et al., 2024). Even when derived from the same source, polysaccharides can exhibit substantial structural variation depending on the extraction approach. For example, Zhang et al. compared Tricholoma mongolicum Imai polysaccharides obtained using five different extraction techniques and found that ultrasound-assisted extraction yielded fractions enriched in low-molecular-weight monosaccharides, which were associated with the highest antioxidant activity (Zhang et al., 2022). Kidney bean (Phaseolus vulgaris L.) is one of the most widely cultivated legume crops globally and contains 15–35% non-digestible polysaccharides (Kong et al., 2024). These polysaccharides represent an interesting and relatively low-cost dietary source for modulating gut microbiota in elderly populations. However, the influence of different extraction approaches on their structural characteristics and associated bioactivities has not yet been systematically investigated. Therefore, here we used ultrasound-assisted extraction to isolate polysaccharides from white kidney bean episperm using different methods including acid, enzymatic, and chelating agent-based. Structural characteristics including monosaccharide composition, molecular weight, functional group composition, fine structure, and microstructure were tested. Commercially available sunflower calathide pectic polysaccharide was taken as control. To investigate whether these molecules can be fermented by gut microbiota of the elderly and possible effects to regulate the gut microbiota, we performed in vitro fermentation with pooled fecal inoculum of the elderly over 65 years old. Digesta at 0, 12, 24, and 36 h were collected upon fermentation, and polysaccharides degradation, gut microbiota composition, and organic acid production were also investigated. Materials and methods Materials White kidney beans were harvested in Liaoning Province, China. Sunflower calathide pectic polysaccharides (SCP) and carbazole were purchased from Sigma Aldrich. Monosaccharide standards, including arabinose (Ara), rhamnose (Rha), galactose (Gal), fucose (Fuc), glucose (Glc), xylose (Xyl), galacturonic acid (GalA), and glucuronic acid (GlcA), were purchased from Sigma Chemical Co. (St. Louis, MO, USA). All chemicals and reagents were of analytical grade. Extraction of kidney bean episperm polysaccharides Kidney bean episperm polysaccharides were extracted according to the method described by Gutohrlein et al. (Gutöhrlein et al., 2020) with minor modifications. Polysaccharides extracted with different methods were defined as follows: acid extractable polysaccharides (AEP), enzyme extractable polysaccharides (EEP), and chelator extractable polysaccharides (CEP). Extraction procedures using acid were conducted as follows: First, kidney bean episperm powder was dissolved in deionized water (1:15, w/v), and then citric acid was used to adjust the pH to 2.0, and ultrasonication was performed at room temperature for 30 min, followed by stirring at 70 °C for 3 h. During the extraction process, the beaker was covered with plastic wrap to minimize solvent evaporation. Following extraction, filtration and centrifugation, the supernatant was concentrated by evaporation to approximately one-third of its original volume. Seventy percent ethanol was then added to the filtrate, which was stored at 4 °C for 12 h. Polysaccharides were collected by centrifugation at 16770 g for 10 min at 4 °C. A second alcohol precipitation was performed for another 12 h, followed by two to three washes with anhydrous ethanol to remove remaining impurities. The wet polysaccharides were dried in a hot air oven at 55 °C until a constant weight was obtained. Extraction was repeated using cellulase and ammonium oxalate, following the same procedure as described above, with the modifications outlined below. For cellulase extraction, kidney bean episperm powder was suspended in deionized water at a ratio of 1:15 (w/v). Three percent cellulase was added, and the mixture was stirred at a constant temperature of 55 °C for 3 h. The temperature was then increased to 100 °C and maintained for 10 min to inactivate the enzyme. For ammonium oxalate extraction, kidney bean episperm powder was suspended in a 50 mM ammonium oxalate solution at a ratio of 1:15 (w/v) prior to extraction. Physicochemical characterization of kidney bean episperm polysaccharides Monosaccharides composition and content Monosaccharides composition and content were determined by ion chromatograph (DIONEX ICS-3000 USA, Diane Corporation, USA) on a CarboPacTMPA20 column (3 × 150 mm, Thermofisher) using a pulsed ampere detector (5000, Thermo Fisher, Waltham, USA) as described by Xu et al. (Xu et al., 2021). A sample of polysaccharides and standard monosaccharides (10 mg each) was mixed with 4 mL of trifluoroacetic acid (TFA, 2 mol/L). Air was removed by flushing with nitrogen for 1 min. The mixture was hydrolyzed at 120 °C for 1 h. After cooling to room temperature, TFA was removed by nitrogen evaporation, and ultrapure water was added to a final volume of 10 mL. The solution was filtered through a 0.2 μm membrane prior to analysis. For chromatographic separation, mobile phases A, B, and C were ultrapure water, 250 mM NaOH, and 1 M sodium acetate, respectively. The injection volume was 10 μL, the column temperature was maintained at 35 °C, and the flow rate was 0.5 mL/min. Polysaccharide degradation during fermentation was determined using the same method. The utilization rate of monosaccharides was determined as the followed equation: Degradation rate%=C0−C36C0×100 (1)where C0 and C36 were the contents of monosaccharides at the beginning (T0) and the end (T36) of fermentation. Molecular weight Molecular weights of weight-average of molar mass (Mw) and number-average of molar mass (Mn) were analyzed with high performance gel permeation chromatography coupled with laser light scattering (SEC-MALLS, DAWN HELEOS, Wyatt Technologies Co., Santa Barbara, CA, USA). SEC-MALLS analysis was performed using an Agilent 1260 series HPLC system (Agilent Technologies, Palo Alto, CA, USA) with Optilab rEX refractive index detector (RID, DAWN EOS, Wyatt Technologies Co., Santa Barbara, CA, USA) and Shodex806 column (8 × 300 mm, Japan). Samples were filtered through a 0.45 μm membrane and eluted with 0.1 M sodium chloride to obtain a final concentration of 2 mg/mL. The mobile phase consisted of 0.1 M sodium chloride solution. The injection volume was 100 μL, and the flow rate was 0.5 mL/min. Data were analyzed using Astra software, version 6.1.7 (Wyatt Technologies Co., Santa Barbara, CA, USA). Structural characteristics of kidney bean episperm polysaccharides Fourier transform infrared spectroscopy (FTIR) FTIR spectra was conducted using a PerkinElmer FTIR spectrometer (PerkinElmer Spectrum 400 FT-IR, Waltham, MA, USA). Samples were dried in an oven at 105 °C for 6 h to completely remove moisture and reduce impurities. The dried material was passed through a 200-mesh sieve and finely ground with pure potassium bromide at a ratio of 1:100 (w/w). The mixture was placed in a mold and pressed into a transparent sheet using a hydraulic press. Infrared spectra were recorded over the wavelength range of 4000 to 400 cm−1 with 32 scans. Data were collected and processed using OMMIC software, version 8.2 (Thermo Nicolet Co., USA). Small angle X-ray scattering (SAXS) Samples were completely dissolved at a concentration of 1% (w/v) and centrifuged at 16770 g for 30 min at room temperature to remove insoluble material. An aliquot of 80 μL of the supernatant was placed in the detection container, sealed, and subjected to vacuum to remove air bubbles and minimize background scattering. Small-angle X-ray scattering (SAXS) measurements were performed using an Anton Paar instrument (Graz, Austria) at 25 °C, with a sample-to-detector distance of 540.74 mm. Data were collected using three exposures of 300 s each and analyzed with GIFT software (Zhang et al., 2023). Scanning electron microscopy (SEM) Morphological differences among samples were analyzed by SEM (FEI Quanta 250, Thermo Fisher). Samples were mounted on the specimen holder using double-sided carbon tape and coated with gold under vacuum. Images were taken at an acceleration potential of 5 kV with magnifications of 10,000 × . In vitro fecal fermentation of kidney bean episperm polysaccharides Preparation of fecal inoculum and fermentation Fecal samples were collected from three healthy elderly individuals (two males and one female; aged 65–70 years) and three healthy adults (one male and two females; aged 25–26 years) who had not received antibiotic treatment for at least 3 months. The studies involving human participants were reviewed and approved by the Ethical Committee for Scientific Research, School of Food and Health, Beijing Technology and Business University (NO. 253 of year 2025). All participants provided written informed consent, adhering to relevant laws and institutional guidelines. Fresh fecal samples were collected in self-sealing bags, and the air was removed before sealing. Fecal inocula were prepared by homogenizing the samples in sterile phosphate-buffered saline (PBS, pH 7.4) to obtain a 10% (w/v) suspension. The homogenate was magnetically stirred for 15 min and filtered through a triple layer of sterile gauze. Three replicates of fecal inocula were prepared from each donor. All procedures were conducted in an anaerobic workstation (gas phase: 80% N₂, 10% CO₂, 10% H₂). The basal nutrient medium (1 L) was prepared as follows: 2.0 g peptone, 2.0 g yeast extract, 0.1 g NaCl, 40.0 mg K₂HPO₄, 40.0 mg KH₂PO₄, 10.0 mg MgSO₄, 10.0 mg CaCl₂, 2.0 g NaHCO₃, 0.5 g cysteine-HCl, 0.5 g bile salts, 2.0 mL Tween 80, 5.0 μL vitamin K₁, and 20.0 mg hemin dissolved in 6 mol/L NaOH. The pH was adjusted to 7.4 using 0.1 mol/L HCl. For each fermentation, 50 mg of sample was added to 1 mL of fecal inoculum and 9 mL of the basal nutrient medium. Three independent vessels were prepared for fermentation in each group, and fermentation samples were collected at 0, 12, 24, and 36 h. The samples were centrifuged at 3000 g for 10 min at 4 °C, and the sediment was rapidly frozen in liquid nitrogen, and stored at −80 °C for microbial analysis. The supernatant was collected in Safe lock microcentrifuge tubes to determine carbohydrate degradation and short chain fatty acids (SCFAs). These two samples were heated in a water bath at 100 °C for 10 min to inactivate the enzymes, and then stored at −80 °C. DNA extraction and 16S rRNA gene sequencing of gut microbial analysis DNA was extracted from the sample according to manufacturer's instructions using a DNA extraction kit (Omega Bio Tek, USA). Qubit dsDNA HS Assay Kit and Qubit 4.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, Oregon, USA) were used to determine DNA concentration of samples, and the quality was evaluated by 1% agarose gel electrophoresis. Universal primers 27F: AGRGTTTGATYNTGGCTCAG and 1492R: TASGGHTACCTTGTTASGACTT were used to amplify full-length 16S rRNA genes from genomic DNA. Multiplex sequencing was performed using forward and reverse 16S primers, with PacBio barcode sequences specific to the sample at the tail of the primers. UCHIME algorithm (v8.1) was used to detect, remove chimeric sequences, and obtain pure readings. The sequences with a similarity of ≥97% were clustered into the same operational taxonomic unit (OTU) using USEARCH (v10.0) clustering. According to the Naive Bayes classifier in QIIME2, all selected representative readings were annotated using the SILVA database with a confidence threshold of 70%. Short chain fatty acids (SCFAs) SCFAs were determined using headspace solid-phase microextraction-gas chromatography–mass spectrometry (Agilent 7890 A, Agilent) (Chen et al., 2020). A 1 mL sample was mixed with 104 μL of 6 M hydrochloric acid and 105 μL of a 500 ppm internal standard solution containing 2-ethylbutyric acid. External standards consisted of acetic acid, propionic acid, isobutyric acid, butyric acid, isovaleric acid, and valeric acid at concentrations ranging from 10 to 1000 ppm. Volatile compounds were adsorbed using a solid-phase extraction (SPE) head at 40 °C for 30 min and subsequently thermally desorbed at the injection port for 3 min. Separation was performed on a DB-WAXUI column (30 m × 0.25 mm × 0.25 μm) with helium as the carrier gas at a flow rate of 1 mL/min. The oven temperature was programmed as follows: 50–100 °C from 0 to 10 min at 5 °C/min, 100–120 °C from 10 to 14 min at 5 °C/min, 120–150 °C from 14 to 39 min at 2 °C/min, and 150–220 °C from 39 to 56 min at 10 °C/min. The inlet and ion source temperatures were both maintained at 250 °C. Statistical analysis All experiments were conducted in triplicates, and the data are expressed as mean ± standard deviation (SD). Statistical differences were analyzed at a significance level of 5% using analysis of variance (ANOVA) and Tukey test in SPSS 21 (IBM, New York, USA). The results were plotted using Origin 9.0 (OriginLab Inc., Northampton, Massachusetts, USA). Results and discussion Monosaccharides composition analysis Chemical structure properties are important for biological activities of polysaccharides, which are usually reflected by their monosaccharide composition and molecular weight (Zheng, Li, et al., 2024). Here, we first investigated monosaccharide composition of white kidney bean episperm polysaccharide under different extraction methods, i.e. acid extractable polysaccharides (AEP), enzyme extractable polysaccharides (EEP), and chelator extractable polysaccharides (CEP). Commercially available sunflower calathide pectic polysaccharides (SCP) were also tested. As shown in Table 1, monosaccharides of SCP, AEP, EEP, and CEP mainly contained (GalA), xylose (Xyl), glucose (Glc), arabinose (Ara), galactose (Gal), rhamnose (Rha), fucose (Fuc), and gluconic acid (GlcA). Molar proportions of the detected monosaccharides differed substantially between the samples, reflecting the impact of extraction method. SCP was characterized by a high abundance of GalA and Rha, indicative of an acidic polysaccharide structure rich in homogalacturonan or rhamnogalacturonan regions. Gal and Ara were present in moderate amounts, while Glc and Xyl were detected in relatively low proportions. This profile is consistent with a pectic polysaccharide framework in which GalA and Rha form the backbone and Gal and Ara occur primarily in branched side chains. Table 1The physicochemical properties, monosaccharides composition (molar proportion, %) and molecular parameters of SCP, AEP, EEP, and CEP.MonosaccharideSCPAEPEEPCEPPurity (%)79.45 ± 1.5682.24 ± 0.5079.23 ± 1.2374.68 ± 0.82Yield (%)–9.35 ± 1.515.13 ± 0.567.62 ± 1.59 Monosaccharides composition (molar proportion, %)Fuc1.060.360.4tRha17.813.102.831.32Ara10.2418.949.7415.57Gal14.849.3111.135.58Glc3.451.5251.921.45Xyl2.9563.4720.7773.44GalA48.772.842.832.26GlcA0.890.460.380.38 Molecular parametersMw (kDa)217.00 ± 0.0182.50 ± 0.0177.40 ± 0.0036.40 ± 0.01Mn (kDa)253.00 ± 0.02163.00 ± 0.01148.00 ± 0.0139.00 ± 0.02Mw/Mn1.14 ± 0.011.65 ± 0.011.32 ± 0.011.04 ± 0.02SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides; Fuc, fucose; Rha, rhamnose; Ara, arabinose; Gal, galactose; Glc, glucose; Xyl, xylose; GalA, galacturonic acid; GlcA, Glucuronic acid; t, trace. The physicochemical properties, monosaccharides composition (molar proportion, %) and molecular parameters of SCP, AEP, EEP, and CEP. SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides; Fuc, fucose; Rha, rhamnose; Ara, arabinose; Gal, galactose; Glc, glucose; Xyl, xylose; GalA, galacturonic acid; GlcA, Glucuronic acid; t, trace. In contrast, AEP and CEP displayed a markedly different composition, with Xyl and Ara as the predominant monosaccharides, followed by Gal, Rha, GalA, and Glc. The predominance of Xyl and Ara suggests that these fractions are enriched in xylan-arabinose-rich backbones, likely associated with arabinoxylan or arabinogalactan-type structures typical of hemicellulosic cell wall components. Ara in these fractions is expected to occur primarily in the furanose form, and its glycosidic linkages are known to be labile under acidic hydrolysis (Patil et al., 2012), which may contribute to its elevated release during analysis. Similarly, the high xylose content may be explained by the susceptibility of Xyl-containing glycosidic bonds to acid cleavage. Acidic and chelator-based extraction conditions, as applied in AEP and CEP recovery, can also disrupt metal ion cross-links (e.g. calcium bridges) between polysaccharide chains in the cell wall, thereby enhancing polysaccharide release (Yang et al., 2022). EEP presented yet another profile, with Glc as the dominant monosaccharide, followed by Xyl, Gal, Ara, GalA, and Rha. All polysaccharide fractions contained only minor amounts of Fuc and GlcA. Collectively, these results indicate that polysaccharides from white kidney bean episperm are predominantly heteropolysaccharides. Although certain chemical characteristics were shared across fractions, the observed differences in monosaccharide molar ratios highlight notable variations in molecular structure and suggest that extraction method plays a decisive role in determining both the composition and the potential functional properties of the recovered polysaccharides. Molecular weight and homogeneity In addition to compositional differences, the molecular weight (Mw) profiles of the polysaccharides were also strongly influenced by the extraction method (Table 1). The Mw values of AEP and EEP were 82.5 and 77.4 kg/mol, respectively, both substantially higher than that of CEP (36.4 kg/mol). A similar trend was observed for the number-average molecular weight (Mn). The observed patterns are consistent with the mechanistic effects of the extraction methods. Acid extraction, particularly under strong acidic conditions, can hydrolyze insoluble polysaccharides into smaller soluble fragments while still retaining relatively high Mw fractions (Căpriță & Căpriță, 2011). Enzymatic extraction with cellulase disrupts the cell wall matrix by hydrolyzing cellulose and hemicellulose networks, thereby releasing polysaccharides that may have been physically trapped within the wall structure (Dranca & Oroian, 2019). Chelator-based extraction with ammonium oxalate disrupts calcium-mediated cross-links among polysaccharide chains in the cell wall matrix. While milder chelating conditions can help preserve polymer integrity (Fernandes et al., 2020), higher concentrations or prolonged exposure can lead to partial depolymerization of pectin. In this study, ammonium oxalate extraction at 50 mM yielded the lowest Mw value, which may indicate a stronger disruption of the cell wall matrix compared with both acidic and enzymatic treatments. The homogeneity of the polysaccharide fractions, expressed as the polydispersity index (Mw/Mn), was 1.65 for AEP, 1.32 for EEP, and 1.04 for CEP, indicating that AEP had the broadest molecular weight distribution. Together, these results show that extraction method not only determines the monosaccharide composition and structural backbone features of white kidney bean episperm polysaccharides, but also exerts a pronounced effect on their molecular size distribution, which may in turn influence their physicochemical behavior and biological functionality. Fourier transform-infrared Fourier transform-infrared (FT-IR) spectroscopy further characterized functional groups and compositions of white kidney bean episperm polysaccharides. As shown in Fig. 1, absorption peaks near 3400 cm−1 and 2900 cm−1 were typical features of polysaccharides. The broad and intense peaks near 3400 cm−1 were mainly attributed to stretching vibration of O—H, implying abundant hydroxyl in polysaccharides molecule. While absorption peaks near 2900 cm−1 were produced by C—H vibrations (CH, CH2, and CH3), which reflected hydrocarbon skeleton characteristics of polysaccharides (Zheng, Pi, et al., 2024). The wavelength range within 1750–1600 cm−1 reflected typical signal of C Created by potrace 1.16, written by Peter Selinger 2001-2019 O vibration, indicating existence of galacturonic acid. More specifically, absorption peaks around 1730 cm−1 and 1640 cm−1 were generated by stretching vibrations of COOR and COO−, respectively, further revealing characteristic structures of esterification and carboxylation in polysaccharides molecule. FT-IR spectra of SCP, AEP, EEP and CEP. SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides. Besides, absorption bands around 1415 cm−1 and 1370 cm−1 demonstrated bending vibration of C—H, which gave more detail of the molecular vibration characteristics of polysaccharides. Strong absorption peaks within the range of 1200–1000 cm−1 were derived from pyranose skeleton in monosaccharide configuration of the polysaccharides. The final absorption peaks near 890 cm−1 were clear signal of β - glycosidic bonds, indicating consistency of glycosidic bond types among different extraction approaches (Zheng, Li, et al., 2024). Overall, although the wavelength of certain typical bands varied slightly under different extraction methods, the polysaccharides molecules exhibited similar infrared spectral characteristics. The results demonstrated that SCP, AEP, EEP, and CEP had similar chemical compositions, as evidenced by monosaccharides. Conformation analysis by small-angle X-ray scattering Innovative detection technique of small-angle X-ray scattering (SAXS) was applied to investigate fine structure of the polysaccharides at nanometer scale. SAXS patterns of all polysaccharides displayed similar overall distribution, but the scattering intensity of CEP was lower than EEP, AEP, and SCP (Fig. 2A). Notably, the scattering peak of SCP shifted toward higher q-values, which was correlated with its highest molecular weight or implied its complex branched structure (Chivero et al., 2014). SAXS analysis of SCP, AEP, EEP and CEP. (A) SAXS intensity curves. (B) Kratky plots of SAXS scattering profiles. (C) Normalized P(r) curves. SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides. Effects of different extraction methods on polysaccharides conformation could be evaluated by Kratky plots (Fig. 2B), with abilities to characterize the spatial structure and highlighting scattering features. Here, Kratky curves of SCP, AEP, EEP, and CEP displayed distinct but clear peaks at 0.045, 0.042, 0.051, and 0.058 nm, respectively. At these peak points, the polysaccharides chains exhibited coiled conformation, with SCP having the highest degree of coiling. Such types of Kratky plots indicated the formation of three-dimensional structures and folded conformations for all polysaccharides. This is different from the Kratky plots reported for soybean polysaccharides, which implied extended chain conformations without any clear peaks, exhibiting a rigid rod-like structure (Chivero et al., 2014). SAXS curves also allow quantification of polysaccharides particle size (Fig. 2C). This could be done through fitting the pair-distance distribution function P(r), which represents all distances between macromolecules, and the particle size is determined by calculating the model-independent maximum dimension (Dmax). The most frequently occurred particles size were 48.10, 50.11, 42.46, and 36.87 nm, and the corresponding Dmax values of SCP, AEP, EEP, and CEP were 119.19, 129.04, 108.23, and 97.97 nm, respectively. A comparison of these Dmax values showed that CEP had the lowest Dmax, indicating the most compact conformation, while AEP and SCP exhibited a relatively extended conformation. As illustrated previously, a more compact molecular conformation could promote bacterial adhesion to the polysaccharides surface, thereby enhancing fermentation efficiency (Tingirikari, 2018). The results may suggest the advantage of CEP in gut microbiota utilization. Scanning electron microscopy The structural morphologys of SCP, AEP, EEP, and CEP were observed under scanning electron microscopy (SEM). As shown in Fig. 3, the surface of SCP showed smooth and continuous feature, indicating its structural integrity. In comparison, AEP, EEP, and CEP showed much rougher and cracked structures, with large differences from SCP. The energy generated by ultrasound with elevated temperature and pulse-pressure could penetrate the sample and induce rupture of polysaccharides, representing as coarse surface structure. Besides, polysaccharides extracted under different methods also differed in morphology by SEM observation. AEP and EEP exhibited aggregated characteristics, with AEP having more compact and evenly distributed porous structure, while EEP exhibited more heterogeneous and relatively larger opening morphology. It has been shown that aggregation features of polysaccharides might be attributed to the presence of carboxyl and hydroxyl groups, and intermolecular interactions could promote the tightly organized structure (Ji et al., 2021). Differently, CEP exhibited compact packed lamellar structure, which would enlarge surface area, with more exposure of polar groups such as hydroxyl and carboxyl groups, and other water binding sites. Such difference might be due to the lower molecular weight of CEP than AEP and EEP, as validated in the study of Wang et al. (Wang et al., 2021). Our results demonstrated that extraction methods could significantly influence surface morphology of polysaccharides, such as shape, damage degree, particle size, and contact area, with possibilities to give different access to enzymatic degradation and gut microbial utilization. SEM images of SCP, AEP, EEP and CEP at 10000× magnification. SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides. Dynamic changes in monosaccharide composition during fermentation Polysaccharides provide important fermentation basis for gut microbiota, and the fermentation capacity is dependent on monosaccharides composition and structural characters. As stated above, white kidney bean polysaccharides extracted by different methods varied in structural features, which may suggest different utilization rates by gut microbiota. Therefore, we first investigated utilization rate of monosaccharides of SCP, AEP, EEP, and CEP by gut microbiota of the elderly. The changes of monosaccharides with fermentation are shown in Table 2. Gut microbiota started to utilize Glc, Gal, and Ara already at T12, and these monosaccharides were consumed by 87.02%, 56.07%, and 71.88% at T36, respectively. As reported, monosaccharides, especially neutral sugars in the side chain of polysaccharides were easily to be utilized during fermentation (Yu et al., 2022), indicating that Glc, Gal, and Ara might be linked at the side chain of SCP. The top most abundant monosaccharides GalA and Rha in SCP were consumed from T24, and were only utilized by 26.04% and 26.87% at T36. This might be due to GalA exists at the backbone of SCP and belongs to acidic monosaccharide, which was speculated to be utilized after neutral sugars. Table 2Monosaccharide degradation of SCP, AEP, EEP, and CEP by fermentation with gut microbiota of the elderly over 65 years old.PolysaccharidesFermentation time (h)Monosaccharide content (ug/mL)FucRhaAraGalGlcXylGalAGlcASCP01.3923.6311.3818.0014.631.84225.322.19121.4324.3410.7417.3611.162.18230.912.00241.3522.814.3510.354.691.95191.571.80361.1217.283.207.911.901.73166.651.59AEP01.3411.1540.2641.7620.43285.3434.055.08121.5714.1825.9730.395.65322.4529.504.75241.0913.7011.2216.872.99263.9016.004.13361.0212.447.5514.223.7453.3613.602.95EEP02.1819.5749.9664.49196.52127.4666.296.00121.9217.3645.5061.6286.87110.7745.704.48241.7817.4028.6649.923.9486.6537.424.30362.1620.2819.7047.534.4788.3742.874.12CEP00.816.0924.0530.2417.11281.1837.284.37120.555.2511.8913.003.92210.2229.452.88240.465.686.867.970.83171.9722.933.05360.344.534.345.910.59106.5910.472.57SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides; Fuc, fucose; Rha, rhamnose; Ara, arabinose; Gal, galactose; Glc, glucose; Xyl, xylose; GalA, galacturonic acid; GlcA, Glucuronic acid. Monosaccharide degradation of SCP, AEP, EEP, and CEP by fermentation with gut microbiota of the elderly over 65 years old. SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides; Fuc, fucose; Rha, rhamnose; Ara, arabinose; Gal, galactose; Glc, glucose; Xyl, xylose; GalA, galacturonic acid; GlcA, Glucuronic acid. Differently, in AEP, EEP, and CEP, monosaccharides of Glc, Ara, Gal, and GalA were already utilized from T12, and were gradually degraded with fermentation by gut microbiota of the elderly, but varied in utilization extent. By the end of fermentation at T36, these monosaccharides were consumed as high as 96.54%, 81.96%, 80.44%, and 71.92% in CEP, 81.69%, 81.25%, 65.95%, and 60.07% in AEP, and 97.72%, 60.58%, 26.29%, and 35.34% in EEP, respectively. The results may suggest that CEP was more readily utilized by gut microbiota of the elderly. The primary monosaccharides in AEP and CEP were Xyl and Ara, with molar ratios reaching 82.41% and 89.01%, respectively (p < 0.05). Therefore, we speculated that AEP and CEP were heteropolysaccharides primarily composed of xylan. As the top most abundant monosaccharides in AEP and CEP, the consumption of Xyl in AEP was delayed until T24, and was only utilized by 7.51%, but it was vastly degraded by 81.30% at T36. It may imply that Xyl in AEP provided a continuous carbohydrate source for gut microbiota. The delayed utilization of Xyl may be due to the fact that the gut microbiota first utilized neutral sugars such as Ara, which exists on the polysaccharide side chain and is more easily hydrolyzed. Xyl on the main chain was then slowly consumed, thereby providing continuous carbon source to support Xyl-degrading bacterial growth, and also avoiding the problem of excessive gas production caused by rapid fermentation. Differently, Xyl in CEP was already degraded by 25.23% at T12, and the degradation kept increasing to 38.84% and 62.09% at T24 and T36, respectively. Furthermore, the less abundant Rha in AEP and EEP remained unchanged, indicating limited contribution for gut microbiota growth. While the less abundant Rha, GlcA, and Fuc were even consumed by 25.60%, 41.22%, and 57.34%, respectively. The rapid and high utilization extent of monosaccharides in CEP might be due to its low molecular weight. Similar results were obtained in gut fermentation of blackberry polysaccharides that lower molecular-weight fragments were much easier to be hydrolyzed and utilized by gut microbiota (Li et al., 2022). In addition, the rapid utilization of monosaccharides might also be related to molecular conformation (Tingirikari, 2018). CEP exhibited the smallest Dmax value, indicating its most compact molecular conformation, which resulted in superior fermentation performance, followed by EEP. Although AEP exhibited a relatively looser conformation, its SEM results revealed a uniformly distributed porous structure, which was more conducive to polysaccharides fermentation (Fang et al., 2021). Such different utilization levels of these monosaccharides among SCP, AEP, EEP, and CEP may contribute to different gut microbial communities and metabolites. Regulation of gut microbiota at the genus level Gut microbiota plays a pivotal role in human health, which was shown to be deteriorated with aging process and has been linked to increased risk for aging-associated disorders (Qi et al., 2023). Gut microbiota composition could be significantly regulated by dietary polysaccharides. Therefore, we next investigated the effects of AEP, EEP, and CEP on gut microbiota of the elderly through in vitro fermentation using fecal samples of the elderly. Commercially available SCP was taken as sample reference. At the initial stage of fermentation (T0), microbiome compositions of the elderly (CO) were dominated by beneficial bacteria, such as Blautia_A, Bifidobacterium, Gemmiger_A, Faecalibacterium and Clostridium_T genera, with a total abundance of 54.73%, while genera containing opportunistic or potentially pathogenic bacteria including Escherichia, Collinsella, Klebsiella, and Ruminococcus_E genera only reached 17.66%. By supplementary of the above polysaccharides, gut microbiota compositions were modified to different extents. SCP induced a low relative abundance of Escherichia of 51.70% at T12, and increased to 66.55% and 66.92% at T24 and T36, respectively. Although Collinsella and Ruminococcus_E were reduced from 6.17% and 3.91% at T12 to 4.09% and 1.89% at T36, these were still higher than the abundances in CO group (p < 0.01). During the late stage of fermentation, the abundance of Klebsiella in CO group gradually increased, reaching 27.46% at T36 (Fig. S1F). However, the relative abundance of Klebsiella was only 0.55% in SCP fermentation at T36. Similarly, AEP, EEP, and CEP also induced a rather low level of Klebsiella of 0.14%, 0.05%, and 0.11% at T36, respectively, which were dramatically lower than CO without polysaccharides supplement (p < 0.05) (Fig. S1H-J). Klebsiella is one type of opportunistic pathogen. Its excessive proliferation can serve as one important indicator of intestinal microecological imbalance and immune aging. It has been reported that Klebsiella, usually found in gut microbiota of the elderly with osteoporosis, frailty, hypertension, etc., negatively influences individual's health through triggering gut barrier dysfunction and immune system inflammation (Sun et al., 2024). Our results indicated that in despite of extraction methods, white kidney bean episperm polysacchrides all showed inhibitory effects on genera of potential pathogens members of the elderly similar to commercial available polysaccharides. Extraction approaches, however, induced different abundances of beneficial genera in batch fermentation of the elderly. For example, the abundance of Bifidobacterium in CO dropped to 2.16% at T36, while the relative abundance of Bifidobacterium reached 7.79% at T12 in SCP fermentation, and was slightly decreased to 5.57% at T36 (Fig. S1K, L). Differently, although both AEP and EEP induced only 3.11% and 2.28% of Bifidobacterium at T12, these were increased to 12.88% and 11.28% at T36, respectively (p < 0.05) (Fig. S1M, N). Such effect was not observed in CEP fermentation. The abundance of Bifidobacterium shows significant reduction with age, and has been closely associated with immunosenescence, chronic low-grade inflammation and cognitive decline. AEP and EEP interventions that promoted Bifidobacterium may contribute to alleviate cognitive impairment, sarcopenia and inflammation in the elderly (Baek et al., 2023). CEP specifically enriched Clostridium_T from 0.52% at T12 to 21.68% at T36, showing a significant increase compared to the abundance of the CO group at T36 (p < 0.05) (Fig. S1T). Clostridium_T was reported to produce butyric acids effectively, which may suggest potential of CEP to regulate a balanced gut microbiota and improve gut barrier of the elderly (Lee et al., 2016). These differences in gut microbiota regulation may be related to structural characteristics such as the monosaccharides composition and molecular weight of polysaccharides. Based on the monosaccharide composition of the polysaccharides, we speculated that AEP and CEP may be xylan composed of xylose. Xylan can be utilized by various gut microbiota, including Clostridium, Bifidobacterium, and Bacteroides. Consistent with previous findings, our research indicates that xylan may serve as carbon source for promoting the growth of butyrate producing bacteria such as Clostridium and Bifidobacterium (Zhang et al., 2019). Xylanases and β-xylosidases secreted by these bacterial genera could synergistically hydrolyze the xylan backbone, breaking down xylan into xylose monomers, thereby providing a continuous carbon source for the bacterial growth (de Mello Capetti et al., 2024). Simultaneously, xylan promoted the production of SCFAs to exert prebiotic effects upon bacterial degradation. From the perspective of molecular weight, CEP, with the lowest molecular weight, demonstrated greater activity in promoting Bifidobacterium growth, and the finding was corroborated by Huang et al. (Huang et al., 2024). As shown in Fig. 5A, among all fermentation groups containing different polysaccharides, AEP had particularly significant effect on gut microbiota diversity. The Chao1 index in AEP significantly dropped to the lowest level at T36 compared to T0 (p < 0.05), indicating that microbial species richness in the AEP group was significantly lower. At the same time, the Simpson index and Shannon index in AEP also exhibited similar trends, while these were not observed in other polysaccharide groups. In consistent with previous observation that single polysaccharides may selectively enrich the existing microbiota rather than introducing new species or altering the original microbiota, which may result in no change or a decrease in α diversity (Liu et al., 2025). Our results here indicated that the reduction in microbial α diversity by AEP might be due to the proliferation of specific bacterial populations. To thoroughly evaluate the effects of different extracts on the gut microbiota, linear discriminant analysis (LEfSE) was performed (Fig. 5B). CO group was dominated by Faecalibacterium and Ruminococcus. The results clearly showed that Klebsiella dominated CO group at T36, while Aerococcus was present at high abundance in EEP group. In the CEP group, the abundance of certain genera within Clostridiales were mostly enriched, indicating that CEP promoted these bacterial groups. We found that the levels of Klebsiella bacteria, which were enriched in the CO group after fermentation, were significantly reduced in the white kidney bean episperm polysaccharide fermentation groups, indicating that white kidney bean episperm polysaccharides have an inhibitory effect on potential pathogens. Comparison of gut microbiota diversity at the genus level among different groups and LEfSe analysis. (A) α diversity analysis. (B) LEfSe analysis. CO was set as a sample control with pooled fecal samples of the elderly without any carbohydrate supplement. SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides. Dynamic changes in SCFAs and bSCFAs production during fermentation As an outcome of fermentation, distribution of SCFAs and bSCFAs act as an important sign for gut microbial metabolic activities for utilizing carbohydrates and protein in colon. SCFAs, mainly acetic, propionic, and butyric acids are primary gut microbial fermentation products of dietary fiber, and play crucial roles in metabolism and immune regulation. While bSCFAs, especially isobutyric acid and isovaleric acid, are demonstrated to be hazardous to gut epithelium, and to be positively associated with aging when in high proportions (Rios-Covian et al., 2020). As shown in Fig. 6A, acetic acid was the most abundant SCFAs among all groups. It was already rapidly accumulated at high level in AEP, EEP, and CEP fermentation at T12, which were significantly higher than SCP and control (CO) without any carbohydrate supplementation (p < 0.05). AEP and CEP fermentation kept increasing acetic acid to 6.53 and 5.41 mmol/L until T24, and were slightly decreased to 5.52 and 4.67 mmol/L, respectively at T36. While acetic acid in EEP was consistently increased to 7.9 mmol/L at T36. Although acetic acid was also gradually increased in SCP with fermentation, it was much lower than that in AEP, EEP, and CEP (p < 0.05). The difference might be due to the higher relative abundance of beneficial bacteria of Bifidobacterium and Clostridium_T in AEP, CEP, and EEP than SCP. These bacteria were reported to produce acetic acid, which are generally considered as major SCFAs to inhibit enteropathogenic infections, enhance gut barrier function, and alleviate age-related dysfunctions (Ma et al., 2023). Concentration of short chain fatty acids (SCFAs) and branched short-chain fatty acids (bSCFAs) after fermentation of extracted polysaccharides with SCP, AEP, EEP and CEP at T0, T12, T24, and T36. (A) Acetic acid. (B) Butyric acids. (C) Propionic acids. (D) Valeric acids. (E) Isobutyric acid. (F) Isovaleric acid. (G) Total SCFAs. (H) Total bSCFAs. (I) Total SCFAs / Total bSCFAs. Different capital letters indicate significant differences (p < 0.05) among different groups. Different lowercase letters indicate significant differences (p < 0.05) with fermentation time. SCP, sunflower calathide pectic polysaccparides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides. Acetic acid act as important intermediate product for butyric acid production. The decrease of acetic acid in AEP and CEP in later period of fermentation might be ascribed to the synthesis to butyric acid. It was evidenced by the content of butyric acid in AEP and CEP (Fig. 6B) that were significantly increased to 6.49 and 6.14 mmol/L at T36, respectively, and were dramatically higher than other groups (p < 0.05). Butyric acid is not only an important energy source for gut epithelium, but also enhances intestinal barrier function, and alleviates inflammation related to leaky gut syndrome that usually occur in the elderly (Rashidah et al., 2022). The high production of butyric acid in AEP and CEP might be a result of utilization of Xyl and dynamic changes of Bifidobacteria and Clostridium_T in AEP and CEP (Fu et al., 2019). The butyrate producing bacteria of Clostridium_T was already significantly enriched at T24 in AEP (Fig. 4A), which might start to utilize Ara and Xyl, with accumulation of butyric acid at T24 in AEP. As Xyl provided continuous carbon source, Bifidobacteria was significantly enriched at T36 in AEP (Fig. 4A), which produced more butyric acid. The significant increase of butyric acid in CEP at T36 was also consistent with the accumulation of Clostridium_T in CEP at T36. Relative abundance of gut microbiota at genus level by fermentation of SCP, AEP, EEP and CEP with pooled fecal samples of the elderly over 65 years old. (A) Relative abundance of each genus by fermentation of SCP, AEP, EEP and CEP at T0, T12, T24, and T36. (B) Heatmap of the relative abundance distribution of the genus by fermentation of SCP, AEP, EEP and CEP at T0, T12, T24, and T36. CO was set as a sample control with pooled fecal samples of the elderly without any carbohydrate supplement. SCP, sunflower calathide pectic polysaccharides; AEP, acid extractable polysaccharides; EEP, enzyme extractable polysaccharides; CEP, chelator extractable polysaccharides. Differently, propionic acid and valeric acid were produced in a relatively low amount across all groups (Fig. 5C and D), which might limit their biological functions. For example, despite consistent increase of propionic acid were observed for SCP and EEP during fermentation, the contents were only 0.27 and 0.40 mmol/L at T36, respectively, which were much lower than those in CO. The contents of valeric acid were in a much lower level, and remained around 0.06 mmol/L for all samples. Currently, biological functions of propionic acid remained controversial. They were reported to improve glucose and lipid metabolism, inhibit fat accumulation, and regulate immune activities (Markowiak-Kopeć & Śliżewska, 2020). Meanwhile, side effects were also observed when propionic acid were in excessive amount, such as increased apoptosis, hepatic oxidative stress, and inflammatory events by entering the liver (Alhusaini et al., 2023). In contrast, bSCFAs including isobutyric acid and isovaleric acid in SCP, AEP, EEP, and CEP were synthesized in relatively low amount as compared to CO (p < 0.05) in later fermentation stages at T24 and T36. Isobutyric acid and isovaleric acid are mainly metabolites from the fermentation of essential amino acids of valine and leucine. Such types of fermentation usually occur when dietary carbohydrates are lacking or exhausted (Rios-Covian et al., 2020). Therefore, the low contents of bSCFAs in SCP, AEP, EEP, and CEP than in CO might be attributed to the consistent supply of polysaccharides, which could be explained by the gradual consumption of each monosaccharides until the end of fermentation (Table 2). Furthermore, higher levels of Blautia in CO might be another reason for raising bSCFAs production as reported previously (Procházková et al., 2023). To sum up, total SCFAs including acetic acid, butyric acid, propionic acid, and valeric acid in SCP, AEP, EEP, and CEP were all significantly increased with fermentation (p < 0.05). White kidney bean polysaccharides, especially AEP and CEP showed much higher total SCFAs than EEP and SCP (p < 0.05). This could be explained by the differences in molecular weight. Among all polysaccharides, CEP exhibited a significantly lower molecular weight than AEP and EEP. Studies indicated that low-molecular-weight polysaccharides were more readily utilized by gut microbiota than high-molecular-weight polysaccharides, causing the proliferation of SCFA-generating bacteria and producing more SCFAs, thus resulted in better intestinal fermentability (Zhao et al., 2021). Meanwhile, SCP, AEP, EEP, and CEP produced much less total bSCFAs in later period of fermentation at T24 and T36. As a result, the ratios of total SCFAs/bSCFAs in CO kept decreasing with fermentation, indicating stronger protein fermentation over carbohydrates. It was demonstrated that bSCFAs accumulation might be negative factors for gut inflammation, acute ischemic stroke, and metabolic disorder that are usually occur in the elderly populations (Ratajczak et al., 2021). However, this tendency was reversed by carbohydrates supplementation. SCFAs/bSCFAs in SCP remained unchanged, indicating a balanced carbohydrates and protein fermentation. AEP kept increasing SCFAs/bSCFAs, and EEP and CEP increased the ratio first, followed by slight decrease, implying that gut microbiota shifted from carbohydrates to partial protein fermentation. However, the ratios of AEP, EEP, and CEP were all significantly higher than CO and SCP, indicating stronger overall carbohydrates utilization by gut microbiota of the elderly, and stronger prebiotic potential to benefit the health of the elderly. Conclusion In summary, polysaccharides from white kidney bean episperm were heteropolysaccharides, and structural characteristics were highly dependent on the extraction methods. Our results indicated that monosaccharides in AEP and CEP were mainly Xyl and Ara, while EEP was Glc and Xyl, which might be associated with their higher molecular weight. AEP showed an evenly distributed porous micro-structure, which may give more access to gut microbiota utilization. In vitro fermentation model with the elderly gut microbiota showed that AEP, CEP, and EEP all significantly inhibited gut pathogens despite of extraction methods. Meanwhile, AEP enhanced the relative abundance of Bifidobacterium, and stimulated the production of SCFAs, especially butyric acid, and increased the ratio of SCFAs/bSCFAs with fermentation. However, the study has certain limitations. More donors could be recruited for fermentation to avoid inter-individual variability. Additionally, ex vivo fermentation models cannot fully simulate the human gut environment or the interactions among the gut microbiota and various organs. In vivo study is recommended for better translation of these results. Furthermore, 16S rRNA analysis has limited resolution, making it difficult to distinguish closely related species, and the whole-genome metagenomic sequencing could be applied. Overall, these findings may suggest the prebiotic potential of white kidney bean episperm polysaccharides, especially AEP, for modulating gut microbiota of the elderly, and provide theoretical foundation for developing dietary interventions toward the elderly. CRediT authorship contribution statement Yimei Hu: Writing – review & editing, Writing – original draft, Software, Methodology, Investigation, Formal analysis, Data curation. Xinyang Li: Software, Investigation, Formal analysis, Writing – review & editing. Caiping Duan: Validation, Software, Methodology, Investigation. Yanbo Wang: Writing – review & editing, Supervision, Resources, Conceptualization. Bei Wang: Writing – review & editing, Supervision, Resources, Conceptualization. Chunli Kong: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Paul de Vos: Writing – review & editing, 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. Acknowledgment The study was supported by the National Natural Science Foundation of China (32201954) and Research Foundation for Youth Scholars of Beijing Technology and Business University (RFYS2025). Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2026.104079. Contributor Information Bei Wang, Email: wangbei@th.btbu.edu.cn. Chunli Kong, Email: kongc0606@btbu.edu.cn. Supplementary data Supplementary material Data availability Data will be made available on request. References Associated Data Supplementary Materials Supplementary material Data Availability Statement Data will be made available on request.