Structural analysis and prebiotic properties of the polysaccharides produced by Lactiplantibacillus plantarum YT013

Abstract

Exopolysaccharides (EPS) from Lactiplantibacillus plantarum YT013 were isolated, structurally characterized and evaluated for gastrointestinal stability and in vitro prebiotic potential. YT013-EPS-1 (57.51 kDa) demonstrated superior resistance to simulated gastrointestinal digestion and maintained structural integrity relative to YT013-EPS-2 (149.3 kDa). Both EPSs exhibited significantly increased antioxidant activity after digestion relative to their undigested forms. In an in vitro mouse fecal fermentation model, YT013-EPS-1 significantly increased SCFA production and promoted beneficial bacteria such as Phascolarctobacterium and Lactobacillus, while reducing potentially pathogenic genera such as Escherichia-Shigella and Acinetobacter. These microbial and metabolic shifts suggest that YT013-EPS-1 influences gut microbiota and promotes microbial metabolites associated with intestinal barrier function and host metabolic regulation. Notably, YT013-EPS-1 exerted more pronounced modulatory effects on gut microbiota composition and SCFA profiles than YT013-EPS-2. Collectively, these findings highlight the potential of YT013-EPS-1 as a prebiotic ingredient for modulating gut microbiota in vitro.

Graphical abstract

Highlights

• •Two EPS from L. plantarum YT013 were isolated and structurally identified.
• •YT013-EPS-1 exhibited superior gastrointestinal stability than YT013-EPS-2.
• •YT013-EPS-1 selectively promoted gut probiotics with potential prebiotic effects.

1. Introduction

Lactic acid bacteria (LAB) are a group of Gram-positive microorganisms extensively utilized in the production of fermented foods and classified as Generally Recognized as Safe (GRAS) (Abdul Hakim, Xuan, & Oslan, 2023; Ju et al., 2024). During the initial phase of fermentation, LAB metabolize carbohydrates into large quantities of organic acids (such as lactic acid and acetic acid) and produce other metabolites, including bacteriocins, ethanol, and exopolysaccharidesthat contribute to food preservation, texture, and flavor development (Abdul Hakim et al., 2023). Among these bioactive compounds, exopolysaccharides (EPS) have garnered considerable interest due to their structural complexity and functional properties. EPS are high molecular weight biopolymers secreted into the surrounding environment during metabolic processes (Caggianiello, Kleerebezem, & Spano, 2016). They can be classified into homopolysaccharides or heteropolysaccharides, with molecular weights (Mw) ranging from 10³ to 10⁷ Da (Yang, Jiang, & Tian, 2023). Structural variations in Mw, monosaccharide composition, glycosidic linkages, and branching patterns critically influence their physicochemical characteristics and bioactivities (Zhang et al., 2024). In recent years, EPS derived from LAB have been extensively studied due to their distinctive structural characteristics and diverse biological activities. These include antioxidant effects, cholesterol-lowering properties, anti-tumor activities, immunomodulatory functions, intestinal barrier repair, and prebiotic effects (Dedhia, Marathe, & Singhal, 2022). These multifaceted characteristics highlight the substantial potential for applications of LAB-derived EPS in functional foods, cosmetics, and biomedicine (Yang et al., 2023).

Moreover, recent studies have revealed that microbial EPS, such as those produced by lactic acid bacteria (e.g., Lactobacillus plantarum, Bifidobacterium longum), which typically consist of heteropolysaccharides rich in glucose, galactose, and rhamnose, exhibit resistance to digestion by gastric and small intestinal fluids, allowing them to reach the colon where they serve as substrates for gut microbiota (Fei et al., 2024; Miyamoto et al., 2023; Zhang et al., 2024). This process not only modulates the gut microbial community but also promotes the production of short-chain fatty acids (SCFAs) and other metabolites, which have been associated with improved intestinal barrier function, modulation of immune responses, attenuation of inflammation, and enhanced metabolic homeostasis (Kalyanaraman, Cheng, & Hardy, 2024; Shoaf, Mulvey, Armstrong, & Hutkins, 2006). Therefore, EPS are increasingly recognized as prebiotic candidates (Gao et al., 2024; Huang et al., 2022; Tang et al., 2022). Accumulating evidence suggests that the prebiotic properties of EPS are closely dependent on their intricate structural characteristics (Y. Liu, Wang, & Wu, 2021). Previous studies have identified molecular weight, glycosidic bond, and monosaccharide composition as key determinants of polysaccharide fermentation by gut microbiota (Surayot et al., 2014). However, research on novel purified EPS fractions from LAB and their prebiotic effects remains limited. Further investigation is imperative to elucidate the detailed structure of homogeneous LAB-derived EPS, particularly those with clear fermentation characteristics. Furthermore, understanding the structural transformations of EPS during digestion and fermentation is crucial for unveiling their mechanisms of action in the gastrointestinal tract. Lactiplantibacillus plantarum YT013, a LAB strain isolated from serofluid dish, a traditional fermented vegetable widely consumed in northwest China, was previously identified by our research group (Zhang et al., 2022). Crude EPS (YT013-EPS) extracted from this strain exhibited strong anti-cancer activity by inducing apoptosis in AGS cells (Zhang et al., 2022). However, the structural features of its purified EPS fractions and their impact on gut microbiota remain unexplored, which is of growing interest given the central role of gut microbiota in regulating host metabolism, immune responses, and maintaining intestinal barrier integrity. Given the potential of LAB-derived EPS as functional biomolecules, a systematic investigation of YT013-EPS is warranted.

In this study, two homogeneous exopolysaccharides, YT013-EPS-1 and YT013-EPS-2, were isolated from Lactiplantibacillus plantarum YT013 and structurally characterized by analyzing their chemical composition, monosaccharide profile, molecular weight, functional groups, and surface morphology. Their prebiotic potential was evaluated using in vitro gastrointestinal digestion and fecal fermentation models, focusing on digestibility, antioxidant activity, as well as modulatory effects on gut microbiota. These findings contribute to the understanding of LAB-derived EPS and support future research into their functional relevance.

2. Materials and methods

2.1. Materials

Lactiplantibacillus plantarum YT013, isolated from serofluid dish–a traditional fermented food–as described in our previous study (Zhang et al., 2022), is preserved in the China Center for Type Culture Collection (CCTCC, M2018775). Pepsin, yeast extract, agar, and other MRS medium components were obtained from Solarbio Science & Technology Co., Ltd. (Beijing, China). Monosaccharide standards were purchased from Sigma-Aldrich Chemical Co., Ltd. (St. Louis, MO, USA). Inulin and SCFA standards were obtained from Macklin Biochemical Co., Ltd. (Shanghai, China). Other reagents, including hydrochloric acid, potassium chloride, and sodium chloride, were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China), and all were of analytical grade.

C57BL/6 mice (n = 12; six males and six females; 8 weeks old, 18–22 g) were purchased from Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences.

2.2. Extraction, isolation and purification of EPS

Exopolysaccharides were extracted from L. plantarum YT013 following the procedure described in our previous work. The strain was activated for two generations, then inoculated at a 5 % (v/v) concentration into Man-Rogosa and Sharp (MRS) broth, followed by cultivation at 37 °C for 36 h. After centrifugation, the supernatant was filtered, concentrated under reduced pressure, sequentially mixed with three volumes of absolute ethanol and maintained at 4 °C overnight. Subsequently, the precipitate was collected via centrifugation, redissolved in deionized water, and treated with Sevage reagent (chloroform/n-butanol, 4:1, v/v) to remove proteins. After removing organic solvents by reduced-pressure concentration, the solution was dialyzed (MWCO 8–14 kDa) against deionized water for 72 h and lyophilized to yield crude polysaccharides (YT013-EPS-T).

The crude EPS was purified by DEAE-52 anion exchange cellulose chromatography using a stepwise elution with deionized water, followed by NaCl solutions (0.1, 0.3, and 0.5 M) at a flow rate of 1 mL/min. The eluate was collected in 8 mL fractions and monitored using the phenol–sulfuric acid method. Fractions exhibiting similar elution profiles were combined, dialyzed and lyophilized, resulting in two purified EPS fractions, designated YT013-EPS-1 and YT013-EPS-2.

2.3. Characterization of YT013-EPS-1 and YT013-EPS-2

2.3.1. Chemical composition analysis

Total carbohydrate content of YT013-EPS fractions was determined by the phenol–sulfuric acid method with glucose as the standard. Uronic acid content was measured using the m-hydroxy diphenyl colorimetric method, using D-galacturonic acid as a standard. Protein content was measured using BCA commercial kit, with bovine serum albumin (BSA) as the standard.

2.3.2. Ultraviolet-visible (UV–vis) spectrometry analysis

UV–Vis spectra of YT013-EPS-1 and YT013-EPS-2 (2 mg/mL) was recorded using a UV-1750 spectrophotometer (Shimadzu, Japan). Baseline correction was performed using deionized water as the blank.Spectra were scanned from 200 to 800 nm to evaluate the presence of proteins and nucleic acids in the polysaccharide samples.

2.3.3. Molecular weight analysis

The molecular weights of YT013-EPS-1 and YT013-EPS-2 were measured via size-exclusion chromatography equipped with an Ultrahydrogel™ column (7.8 × 300 mm, Waters, USA), a refractive index detector (RID), and a multi-angle laser light scattering (MALLS). The samples (50 μL of 1 mg/mL) were filtered through a 0.45 μm cellulose filter and eluted with ultrapure water at 0.5 mL/min. The refractive index increment (dn/dc) was determined to be 0.133 mL/g at 660 nm. Molecular weight was calculated using the Zimm method with ASTRA software (Wyatt Technology Corp., USA).

2.3.4. Monosaccharide composition

Monosaccharide composition was analyzed by pre-column derivatization with 1-phenyl-3-methyl-5-pyrazolone (PMP) followed by high-performance liquid chromatography (HPLC) method. Polysaccharide solutions (2 mL, 2 mg/mL) were hydrolyzed with 4 M trifluoroacetic acid (TFA) at 121 °C for 2 h. The hydrolysates were washed with methanol three times to remove residual TFA, then dried under nitrogen gas and redissolved in 2.0 mL distilled water. The derivatization was performed by adding 2.0 mL of 0.5 M PMP-methanol solution and 1.0 mL of 0.2 M NaOH, followed by incubation at 70 °C for 1 h. The reaction was neutralized with 0.2 M HCl, and the aqueous phase was extracted with chloroform three times, filtered through a 0.22 μm membrane, and analyzed by HPLC (Thermo U3000 system) with a diode array detector (DAD) and a ZORBAX Eclipse XDB-C18 column (5 μm, 4.6 × 250 mm). The analysis conditions were as follows: injection volume of 10 μL, mobile phase consisting of 50 mM phosphate buffer (pH 6.7) and acetonitrile (83:17, v/v), with flow rate of 1.0 mL/min, column temperature maintained at 30 °C, and DAD detection wavelength at 245 nm. To confirm hydrolysis completeness, polysaccharide samples underwent repeated hydrolysis under identical conditions. Consistent monosaccharide profiles analyzed by HPLC confirmed complete hydrolysis.

2.3.5. Fourier-transform infrared (FT-IR) spectrometry analysis

FT-IR spectra were recorded using a NEXUS 670 spectrometer (Thermo Nicolet, USA). Dried samples (2 mg) were mixed with 200 mg of KBr powder and compressed into thin pellets. The FT-IR spectrum was scanned over a wavenumber of 4000–400 cm⁻¹. KBr was regarded as the blank group.

2.3.6. Congo red assay

The presence of a triple-helix conformation was evaluated using the Congo red assay as described by Guo et al. (Guo et al., 2021). In brief, 80 μmol/L Congo red reagent was mixed with polysaccharide solutions (1.0 mg/mL) at a 1:2 ratio. NaOH solution (1 M) was then added to achieve final NaOH concentration ranging from 0.1 to 0.5 M. Subsequently, UV–Vis spectra was recorded from 400 to 600 nm to determine the maximum absorption wavelength (λ_max). Deionized water was used as a blank control.

2.3.7. Scanning electron microscopy (SEM)

The surface morphology of YT013-EPS-1 and YT013-EPS-2 was observed using SEM system (JSM-6701F, JEOL, Japan) with an accelerating voltage of 20 kV. The dried samples were mounted on conductive adhesive and then sputter-coated with a thin layer of gold. Morphology of the polysaccharides was observed at magnifications of 500×, 1000×, and 5000 ×.

2.4. Evaluation of antioxidant activities

Polysaccharide solutions were prepared at concentrations of 0.25, 0.5, 1, 2, 3, 4, 5, and 10 mg/mL, with vitamin C (Vc) as the positive control in all antioxidant assays.

• a.DPPH free-radical scavenging activity

The DPPH· free radical scavenging ability of the samples was performed using a modified method based on the protocol reported (Bhanja, Maity, Rout, Sen, & Patra, 2022). Briefly, 20 μL of the polysaccharide solution at various concentrations was co-incubated with 180 μL of a 0.40 mmol/L DPPH solution in ethanol for 30 min, protected from light. The absorbance of the mixture was then measured at 517 nm. Deionized water served as blank control instead of samples. The DPPH scavenging activity was calculated using the following equation:

$$\text{DPPH free}-\text{radical scavenging activity}\left(\%\right)=\left(1-\left(A_1-A_2\right)/A_0\right)\times 100\%.$$

Where A₀ and A₁ represent the absorbance of the blank control and samples (or Vc), while A₂ refers to the absorbance of the sample using anhydrous ethanol instead of the DPPH solution.

• b.ABTS free-radical scavenging activity

The ABTS· free radical scavenging ability of samples was assessed using a modified method based on the method reported (Singh, Jagtap, Rajpurohit, & Singh, 2025). In general, 1.4 mmol/L ABTS solution was mixed with 0.49 mmol/L K₂(SO₄)₂ in a 1:1 ratio and incubated protected from light for 16 h to generate ABTS radicals. The UV absorption at 734 nm was adjusted to 0.70 ± 0.02 by adding 5 mmol/L phosphate buffer (PBS, pH 7.4) to prepare the ABTS· working solution. For the measurement, 20 μL of the polysaccharide solution was combined with 180 μL of the ABTS· working solution and incubated protected from light for 6 min. Absorbance was measured at 734 nm. The ABTS scavenging activity was calculated using the following formula:

$$\text{ABTS free}-\text{radical scavenging activity}\left(\%\right)=\left(1-\left(A_1-A_2\right)/A_0\right)\times 100\%$$

Where A₀ and A₁ were the absorbance of the blank control and samples (or Vc), while A₂ represents the absorbance of the sample using deionized water instead of the ABTS· working solution.

• c.Hydroxyl radical scavenging activity

For the hydroxyl radical assay (Liang et al., 2024), a reaction mixture consisting of 1 mL of sample solution, 1 mL of 6 mmol/L FeSO₄, 1 mL of 3 % H₂O₂, and 1 mL of 9 mmol/L salicylic acid-ethanol solution was incubated at 40 °C for 1 h. Absorbance was measured at 510 nm. The scavenging activity was calculated using the following equation:

$$\text{Hydroxyl radical scavenging activity}\left(\%\right)=\left(1-\left(A_1-A_2\right)/A_0\right)\times 100\%$$

Where A₀ and A₁ were the absorbance of the blank control and samples (or Vc), while A₂ refers to the absorbance of the sample using deionized water instead of the H₂O₂.

2.5. Simulated gastrointestinal digestion

2.5.1. Simulated gastric digestion of polysaccharides

In vitro simulated gastric digestion was performed in accordance with the method reported by Chuan Liu (C. Liu et al., 2025), with slight modifications. The gastric electrolyte solution was prepared by mixing 34.5 mL of KCl solution (0.5 M), 4.5 mL of KH₂PO₄ solution (0.5 M), 62.5 mL of NaHCO₃ solution (1 M), 59 mL of NaCl solution (2 M), 2 mL of MgCl₂·6H₂O solution (0.15 M), 2.5 mL of (NH₄)₂CO₃ solution (0.5 M), 6.5 mL of HCl solution (6 M), and 15 μL of CaCl₂·2H₂O (0.3 M). The pH was then adjusted to 3.0 with 0.1 M HCl. During the subsequent 3 h of incubation, the pH of the digestion system was monitored every 30 min using a calibrated digital pH meter and remained stable at approximately pH 3.0. Subsequently, 1.25 mL of pepsin (80,000 U/mL) was added to 28.75 mL of gastric electrolyte solution to obtain artificial gastric fluid. Next, 30 mL of polysaccharides solution (10 mg/mL) was mixed with 30 mL of artificial gastric fluid, and then placed in a water bath shaker at 37 °C for 3 h. After incubation, the gastric digestion solution was heated at 100 °C for 5 min to inactivate the enzyme. Deionized water, prepared as described above, served as the control.

2.5.2. Simulated digestion of polysaccharides in the small intestine

To prepare the small intestinal electrolyte solution, 34 mL of KCl solution (0.5 M), 4 mL of KH₂PO₄ solution (0.5 M), 212.5 mL of NaHCO₃ solution (1 M), 48 mL of NaCl solution (2 M), 5.5 mL of MgCl₂·6H₂O solution (0.15 M), 3.5 mL of HCl solution (6 M), and 12.5 μL of CaCl₂·2H₂O (0.3 M) were mixed thoroughly. After the simulated gastric digestion, the pH of the mixture was adjusted to 7.0 with 0.1 M NaOH. During the subsequent incubation, the pH of the digestion system was monitored every 30 min using a calibrated digital pH meter and remained stable at approximately pH 7.0. Subsequently, 6.25 mL of bile salt solution (38.4 mg/mL) and 12.5 mL of pancreatic enzyme solution (800 U/mL) were added to 25 mL of the prepared intestinal electrolyte solution to obtain artificial intestinal fluid. An equal volume (25 mL) of the gastric digested sample was then mixed with the artificial intestinal fluid (1:1, v/v). The resulting mixture was incubated at 37 °C for 4 h. After incubation, the intestinal digestion mixture was heated at 100 °C for 5 min to terminate the reaction. Deionized water was used as a control to replace the artificial intestinal fluid.

2.6. In vitro fecal fermentation

2.6.1. Preparation of fecal microbiota and fermentation medium

The fermentation medium was prepared according to the previous report with minor modifications (Li et al., 2024). Briefly, the medium was composed of 2.0 g of peptone, 2.0 g of yeast extract, 0.1 g of NaCl, 0.04 g of K₂HPO₄, 0.04 g of KH₂PO₄, 0.01 g of MgSO₄·7H₂O, 2.0 g of NaHCO₃, 0.01 g of CaCl₂·2H₂O, 2 mL of Tween 80, 0.5 g of bile salts, 0.025 g of hemin, 0.5 g of L-cysteine, and 10 μL of vitamin K₁, dissolved in 1 L of distilled water. The pH was adjusted to 7.0 and sterilized at 121 °C for 15 min. The fresh fecal samples were collected immediately from 12 specific-pathogen-free (SPF) 8-week-old C57BL/6 mice (six males and six females, body weight between 18 and 22 g). The collected feces were resuspended in 0.1 M PBS (pH 7.2) in a ratio of 1:6 (w/v), followed by homogenization and filtration to obtain fecal suspension.

All animal procedures were reviewed and approved by the Animal Ethics Committee of Lanzhou University (Approval No. 2024-10-23) and were conducted in accordance with institutional guidelines.

2.6.2. Preparation of fermentation samples

YT013-EPS-1 or YT013-EPS-2 served the only carbon source in the treatment groups. The final fermentation mixture (32 mL) contained 8 mL of the prepared fecal suspension, 4 mL of YT013-EPS-1 or YT013-EPS-2 solution (25 mg/mL), and 20 mL of sterilized fermentation medium. The mixture was placed in a 37 °C anaerobic chamber containing anaerobic agents and anaerobic indicators. Deionized water mixed with fecal suspension was used as the blank control group (BLK), while inulin served as the positive control (INU) under the same conditions. Each fermentation condition was conducted in biological triplicate (n = 3) to ensure reproducibility and statistical reliability. Samples (2 mL) were collected at 0 and 48 h for subsequent analysis.

2.7. Analysis of digestion and fermentation liquid

2.7.1. Determination of pH, reducing sugars and total carbohydrate content

The pH of the samples during the fermentation process was measured using a pH meter. The levels of residual total carbohydrates and reducing sugars during digestion and fermentation were quantified using the phenol‑sulfuric acid method and the 3,5-dinitrosalicylic acid (DNS) assay, with glucose serving as the standard.

2.7.2. Analysis of Mw and monosaccharide composition

The Mw and monosaccharide composition of digested and fermented YT013-EPS-1 and YT013-EPS-2 were analyzed as described in 2.3.3, 2.3.4, respectively.

2.7.3. Evaluation of antioxidant activities

The ABTS· and hydroxyl radical scavenging activities of the digested YT013-EPS-1 and YT013-EPS-2 samples were evaluated according to the protocols described in Section 2.4. To account for the lower concentration of polysaccharides after digestion, samples were adjusted to final concentrations of 0.1, 0.2, 0.5, 0.75, and 1 mg/mL. Vitamin C was used as a positive control. All assays were performed in triplicate.

2.7.4. SCFAs analysis

The concentrations of short-chain fatty acids (SCFAs) were analyzed using a gas chromatography-triple quadrupole mass spectrometry (GC–MS/MS) system consisting of an Agilent 7890B gas chromatograph coupled with a 7000 D mass spectrometric detector (Agilent Technologies, Santa Clara, CA, USA). Fecal fermentation products were centrifuged at 12,000 × g for 10 min, the supernatants were mixed with 200 μL of 20 % phosphoric acid solution (v:v). After centrifugation, the obtained supernatants were filtered through a 0.45 μm filter membrane, followed by the addition of 40 μL of 2-methylvaleric acid used as an internal standard at a final concentration of 40 μg/mL. The GC–MS/MS conditions were as follows: the column was HP-INNOWAX (30 m × 0.25 mm × 0.25 μm); 1 μL of sample was injected in split mode (10:1); nitrogen was used as the carrier gas at a flow rate of 1.0 mL/min. The heating program commenced at 100 °C for 3 min, then increased by 5 °C/min to reach 150 °C, followed by an increase of 20 °C/min to 200 °C, which was maintained for 5 min. The electron energy was set at −70 eV.

2.7.5. Assessment of microbiota changes

Genomic DNA was extracted from the 48 h fermentation samples of each group, followed by amplification of the V3 + V4 region of the 16S rDNA using specific primers with barcodes for microbial community analysis. The primer sequences used were as follows: 341F (CCTACGGGNGGCWGCAG) and 806R (GGACTACHVGGGTATCTAAT). After polymerase chain reaction (PCR) amplification of the target 16S rDNA fragments, the amplicons were purified using a gel purification method. The purified amplicons were then ligated with sequencing adapters to construct the sequencing library. Finally, high-throughput sequencing of the library was performed using the Illumina sequencing platform. PCR amplification, sequencing library construction, and sequencing were conducted with technical support from Gene Denovo Biotechnology Co., Ltd. (Guangzhou, China). After quality control and chimera removal, each sample retained between 98,845 and 117,985 effective reads, with an average of approximately 109,000 high-quality reads per sample. The effective read ratios ranged from 90.57 % to 94.93 %, indicating sufficient sequencing depth and data quality for reliable microbial community analysis (Supplementary Table 5).

2.8. Statistical analysis

All experimental results were statistically analyzed using SPSS 29.0.1.0 software. Graphs were generated using GraphPad Prism 10.0 and Origin 2024. The experiments were independently repeated three times, and the results are expressed as mean ± standard deviation (SD). Statistical significance of differences between groups was determined using one-way analysis of variance (ANOVA) and Student's t-test.

3. Results and discussion

3.1. Isolation and purification of YT013-EPS-1 and YT013-EPS-2

The crude EPS fraction, designated as YT013-EPS, was obtained through fermentation of MRS broth with L. plantarum YT013. Following ethanol precipitation, the crude EPS yield reached 660 ± 3.15 mg per liter of fermentation broth, equivalent to 0.066 % (w/v) relative to the initial culture volume. To further investigate the structural characteristics of YT013-EPS, fractionation of individual polysaccharide components was performed. YT013-EPS was separated using DEAE-cellulose 52 anion exchange chromatography, yielding two sub-fractions: YT013-EPS-1 (57.78 ± 6.97 % of the total YT013-EPS) and YT013-EPS-2 (17.38 ± 3.19 % of the total YT013-EPS), which were eluted with 0 M and 0.1 M NaCl, respectively (Fig. 1A). These two fractions were further purified by dialysis. After lyophilization, YT013-EPS-1 exhibited a white, loosely organized lamellar structure with large flakes, whereas YT013-EPS-2 appeared as a white, soft, and fluffy texture.

Fig. 1.

Extraction, purification, and physicochemical characterization of EPS from Lactiplantibacillus plantarum YT013. (A) Purification of YT013-EPS using DEAE-52 column chromatography; (B) UV–Vis spectra; (C) SEC chromatogram; (D) FT-IR spectra; (E) HPLC chromatogram; (F) Congo red analysis; (G–I) ABTS·, ·OH and DPPH· scavenging capacity of YT013-EPS-1 and YT013-EPS-2. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.2. Structural properties of YT013-EPS-1 and YT013-EPS-2

3.2.1. Chemical composition and mw determination

The chemical compositions of YT013-EPS-1 and YT013-EPS-2, including total carbohydrate, protein, and uronic acid contents, were determined. As shown in Table 1, the total carbohydrate content was 87.38 ± 2.69% for YT013-EPS-1 and 82.19 ± 3.15% for YT013-EPS-2, while the total soluble protein content was 2.48 ± 0.091% and 1.75 ± 0.073%, respectively. The purity of YT013-EPS-1 and YT013-EPS-2 was evaluated by UV–Vis spectroscopic analysis (Fig. 1B). Neither sample exhibited absorption peaks at 260 or 280 nm, indicating the absence of nucleic acid and protein contamination. Furthermore, results revealed that YT013-EPS-1 contained no detectable uronic acids, whereas YT013-EPS-2 exhibited a uronic acid content of 5.23 ± 0.77%, suggesting that YT013-EPS-2 is an acidic polysaccharide.

Table 1.

Analysis of chemical composition of YT013-EPS-1 and YT013-EPS-2.

Sample	YT013-EPS-1	YT013-EPS-2
Extraction yield (%)	57.78 ± 6.97	17.38 ± 3.19
Total carbohydrate content (wt%)	87.38 ± 2.69	82.19 ± 3.15
Protein (wt%)	2.48 ± 0.091	1.75 ± 0.073
Uronic acid (wt%)	N.D.	5.23 ± 0.77
Glucose, Glc	62.92 %	18.30 %
Mannose, Man	32.89 %	10.55 %
Arabinose, Ara	1.45 %	N.D.
Galactose, Gal	1.51 %	5.90 %
Xylose, Xyl	1.22 %	N.D.
Rhamnose, Rha	N.D.	61.86 %
Glucuronic acid, GlcA	N.D.	3.39 %
Glucuronic acid, GalA	N.D.	1.99 %

Note: N.D. represents “No Data Provided,” indicating that no data were detected.

The homogeneity and molecular weight (Mw) distribution of YT013-EPS-1 and YT013-EPS-2 were analyzed using multi-detector size-exclusion chromatography (SEC). As shown in Fig. 1C, both YT013-EPS-1 and YT013-EPS-2 exhibited a single, broad, and symmetrical peak, indicating they were pure and relatively homogeneous. The average Mw of YT013-EPS-1 and YT013-EPS-2 was calculated to be 57.51 kDa and 149.3 kDa, respectively. Furthermore, the polydispersity indices (Mw/Mn) for YT013-EPS-1 and YT013-EPS-2 were calculated to be 2.08 and 3.74, respectively, suggesting a wide Mw distribution. Polysaccharides with broader Mw distributions tend to undergo slower and more sustained microbial fermentation in the colon, as chains of varying lengths are degraded at different rates (Gu et al., 2018). This can prolong the production of SCFAs, potentially enhancing the prebiotic potential of the EPS. Therefore, the observed PDI values support the functional relevance of YT013-EPS-1 and YT013-EPS-2, confirming their suitability for further structural and biological evaluation.

3.2.2. FT-IR analysis

The FT-IR spectrum is a valuable tool for identifying the primary functional groups and chemical bonds in polysaccharides. The FT-IR spectra of YT013-EPS-1 and YT013-EPS-2 (Fig. 1D) exhibited characteristic vibration peaks associated with polysaccharides. A broad and intense band near 3300 cm⁻¹ was attributed to O-H stretching vibrations, which also reflected extensive intra- and intermolecular hydrogen bonding. Such hydrogen bonding networks are known to contribute to the structural stability and physicochemical properties of exopolysaccharides, including solubility and gel-forming capacity. Enhanced peaks around 2900 cm⁻¹ were attributed to the C-H stretching vibration of the methyl group within the sugar rings. Additionally, a strong absorption band between 1000 and 1200 cm⁻¹ was indicative of C-O bending vibrations from C–O–H or C–O–C bonds, suggesting the presence of a pyranose ring structure. Both YT013-EPS-1 and YT013-EPS-2 also displayed strong absorption bands near 1660 cm⁻¹, ascribed to the stretching vibrations of C O resulting from the scissoring vibrations of bound water. For YT013-EPS-2, a strong absorption peak at 1247.2 cm⁻¹ was observed, corresponding to the stretching vibration of the deprotonated carboxyl group (COO⁻), indicating the characteristic absorbance band of acid polysaccharides. The weak absorption peak at 861.1 cm⁻¹ in YT013-EPS-2 was associated with α-glycosidic bonds. Furthermore, the band at 968.1 cm⁻¹ was attributed to the α-1,6-glycosidic linkage in YT013-EPS-2, suggesting that this fraction contains α-glycosidic linkages between sugar units within a pyranose ring configuration. In contrast, YT013-EPS-1 exhibited an absorption peak at 923.1 cm⁻¹, characteristic of a β-glycosidic bond, which was absent in YT013-EPS-2. β-glycosidic linkages were generally more resistant to enzymatic hydrolysis in the upper gastrointestinal tract, allowing the polysaccharide to reach the colon relatively intact. This structural feature may confer increased fermentation resistance, resulting in a more sustained microbial degradation process and potentially prolonged SCFA production. Additionally, the slight absorption at 847.4 cm⁻¹ in YT013-EPS-1 was likely due to the presence of α-glycosidic bonds or mannose.

3.2.3. Monosaccharide composition analysis

The monosaccharide profiles and composition of YT013-EPS-1 and YT013-EPS-2 were analyzed and presented in Fig. 1E and Table 1. The analysis revealed that both YT013-EPS-1 and YT013-EPS-2 are heteropolysaccharides. YT013-EPS-1 consisted of five monosaccharides, including glucose, mannose, arabinose, galactose, and xylose, in a molar ratio of 1.07: 0.85: 0.03: 0.03: 0.03. In addition, YT013-EPS-2 was primarily composed of rhamnose, glucose, mannose, galactose, glucuronic acid, and galacturonic acid, with a molar ratio of 0.30: 0.06: 0.05: 0.03: 0.01: 0.01. The presence of galacturonic acid and glucuronic acid in YT013-EPS-2 suggested its acidic nature, which aligns with previous reports on acidic polysaccharides.

3.2.4. Congo red assay

The Congo red assay was employed to identify the triple-helix conformation of polysaccharides. Under alkaline conditions, polysaccharides with a triple-helix structure interact with Congo red, resulting in a characteristic shift in the maximum absorption wavelength (λ_max) of Congo red. At low concentrations of NaOH, the polysaccharide adopts an ordered triple-helix conformation, to which Congo red binds, leading to a redshift in λ_max. With increasing NaOH concentration, the triple-helix structure is gradually converted into a random coil due to the breaking of hydrogen bonds, leading to a sharp blue shift in λ_max. As presented in Fig. 1F and Supplementary Fig. 1, the λ_max of the YT013-EPS-2-Congo red complexes decreased gradually with increasing NaOH concentrations. Similar to the trend observed with ddH₂O (blank control), the YT013-EPS-2-Congo red complex exhibited a continuous decrease in λ_max with increasing NaOH concentration, indicating the absence of a triple-helix structure in YT013-EPS-2. In contrast, the YT013-EPS-1-Congo red complex showed a notable red shift in λ_max at lower NaOH concentrations (0 M to 0.2 M), followed by a significant blue shift when the NaOH concentration exceeded 0.2 M, demonstrating that YT013-EPS-1 possesses a triple-helix structure. The presence of a triple-helix structure implies a more ordered and compact tertiary conformation, which enhances intermolecular hydrogen bonding, thereby improving physicochemical stability, including thermal resistance and enzymatic tolerance. Moreover, such conformations have been reported to resist rapid degradation in the gastrointestinal tract, potentially leading to more sustained fermentation and prolonged SCFA production in the colon. Therefore, the presence of a triple-helix structure in YT013-EPS-1 suggests its potential for greater physiological stability and prebiotic functionality compared to the more flexible, non-helical YT013-EPS-2.

3.2.5. SEM analysis

Scanning electron microscopy (SEM) provides an effective technique for characterizing the surface morphology of polysaccharides. Representative SEM images of YT013-EPS-1 and YT013-EPS-2 at various magnifications are shown in Fig. 2. At low magnifications (500×), both EPS fractions exhibited similar irregular flake-like appearances. However, at higher magnifications, distinct morphological differences became apparent. At 1000× and 5000× magnifications, YT013-EPS-1 displayed a compact texture with a uneven surface, covered with irregularly distributed large pores, suggesting a highly branched, entangled and porous structure. In contrast, YT013-EPS-2 exhibited a rough surface with minor wrinkles and honeycomb-like formations, indicative of a lower degree of branching compared to YT013-EPS-1. These morphological differences likely arise from variations in the degree of branching, which affect intermolecular interactions and the overall polymerization of the macromolecules.

Fig. 2.

SEM images. (A-C) YT013-EPS-1; (D-F) YT013-EPS-2. Magnifications of 500×, 1000×, and 5000 ×.

3.3. Antioxidant activity of YT013-EPS-1 and YT013-EPS-2 in vitro

The in vitro antioxidant properties of YT013-EPS-1 and YT013-EPS-2 were evaluated by assessing their radical scavenging activities against ABTS, DPPH, and hydroxyl radicals. As shown in Fig. 1 (G-I), both fractions exhibited varying scavenging abilities depending on radical type. Notably, both YT013-EPS-1 and YT013-EPS-2 showed relatively strong scavenging effects against ABTS and hydroxyl radicals but weak activity toward DPPH radicals, with maximal scavenging rates of 6.40 % and 13.95 %, respectively, at 10.00 mg/mL. The IC₅₀ values of YT013-EPS-1 and YT013-EPS-2 were 2.173 mg/mL and 13.85 mg/mL, respectively, for the ABTS radical scavenging capacity, and 7.267 mg/mL and 1.218 mg/mL for the hydroxyl radical scavenging capacity. Compared with vitamin C (ABTS, IC₅₀ = 0.102 mg/mL; OH, IC₅₀ = 0.167 mg/mL), both EPSs exhibited moderate antioxidant activities, consistent with the IC₅₀ range reported for many natural polysaccharides. Interestingly, YT013-EPS-1 was more effective against ABTS radicals, which reflects its potential to neutralize a broad spectrum of reactive oxygen species, possibly due to structural features, e.g., electron-donating hydroxyl groups. In contrast, YT013-EPS-2 showed stronger hydroxyl radical scavenging, suggesting a greater capacity to mitigate highly reactive species involved in cellular oxidative damage. These distinct radical-specific antioxidant profiles support the potential of YT013-EPSs as natural antioxidant candidates for further functional evaluation.

3.4. Simulated gastrointestinal digestion of YT013-EPS-1 and YT013-EPS-2

The release of reducing sugars during digestion is often used as an indirect indicator of polysaccharide degradation. As shown in Supplementary Table 1, both YT013-EPS-1 and YT013-EPS-2 exhibited a significant increase in reducing sugar content following simulated gastric digestion (P < 0.0001), likely due to the acidic environment facilitating partial cleavage of glycosidic bonds. A further increase was observed after intestinal digestion (P < 0.0001), potentially due to the enzymatic activities in the small intestine, which may have triggered the release of oligosaccharide monomers from the polysaccharides. While these findings suggest that both EPS fractions undergo partial degradation under simulated gastrointestinal conditions, we acknowledge that reducing sugar levels alone do not directly confirm glycosidic bond cleavage. Therefore, the results are presented as indirect evidence of digestive susceptibility, and further analyses are needed to elucidate the specific degradation mechanisms.

Mw is a critical structural characteristic of polysaccharides and serves as an indicator of their susceptibility to degradation during digestion. To further assess structural changes, Mw shifts of YT013-EPSs during digestion were analyzed via SEC. As shown in Fig. 3 (A-B), the SEC analysis revealed that during gastric digestion, the Mw of the main peak of YT013-EPS-1 did not exhibit significant changes, with only a minor fraction of low Mw fragments observed. During intestinal digestion, the main peak shifted slightly to the right but remained predominantly within the high Mw region, with an increased presence of low Mw fragments. In contrast, as illustrated in Fig. 3 (D-E), YT013-EPS-2 exhibited a significant reduction in its main peak Mw after gastric digestion, accompanied by an increase in low Mw fragments. Following further intestinal digestion, the proportion of low Mw fragments further increased, indicating more pronounced degradation. These results suggest that under simulated gastrointestinal conditions, the acidic polysaccharide YT013-EPS-2 is more susceptible to enzymatic degradation, whereas the neutral polysaccharide YT013-EPS-1 demonstrates higher resistance to digestive processes.

Fig. 3.

Digestive and fermentation properties of YT013-EPSs. (A-C) Mw distribution of YT013-EPS-1 after gastric digestion, intestinal digestion and fecal fermentation; (D-F) Mw distribution of YT013-EPS-2 after gastric digestion, intestinal digestion and fecal fermentation; (G) Changes in total carbohydrates during fermentation; (H) Changes in reducing sugars during fermentation; (H) pH value at different time points during in vitro fermentation. Different lowercase letters represent significant differences (P < 0.05) and the same letter represents no significant difference (P > 0.05).

To further investigate polysaccharide stability, monosaccharide composition was analyzed pre- and post-digestion (Fig. 4 and Supplementary Table 2). The monosaccharide composition of both YT013-EPS fractions remained largely consistent throughout the digestive process, although slight variations in their molar ratios were observed. For YT013-EPS-1, the molar ratio of glucose, mannose, arabinose, galactose, and xylose was 3.08: 0.94: 0.10: 0.12: 0.06 following gastric-intestinal digestion. Similarly, for YT013-EPS-2, the molar ratio of glucose, mannose, rhamnose, galactose, glucuronic acid,and galacturonic acid shifted to 1.02: 0.69: 0.15: 0.04: 0.01: 0.01 after digestion. A similar increase in glucose content during digestion has also been reported in previous studies (Li et al., 2024). This phenomenon may be attributed to the selective cleavage of specific glycosidic bonds, leading to the relative enrichment of glucose as the predominant product under enzymatic action. These findings align with the Mw data, further confirming that YT013-EPS-2 is more susceptible to digestive degradation than YT013-EPS-1.

Fig. 4.

Monosaccharide composition of YT013-EPS-1 and YT013-EPS-2 during gastrointestinal digestion. (0 h and GI represent the monosaccharide composition before and after gastrointestinal digestion, respectively.).

Overall, analyses of Mw distribution, reducing sugar release, and monosaccharide composition collectively suggest that YT013-EPS-1 exhibits greater structural stability and stronger resistance to gastric and intestinal digestion, whereas YT013-EPS-2 was more susceptible to degradation. These differences likely reflect variations in glycosidic linkages, molecular conformation, and monosaccharide composition between the two EPS fractions.

3.5. Effect of digestion on antioxidant activities of YT013-EPS-1 and YT013-EPS-2

The antioxidant capacities of YT013-EPS-1 and YT013-EPS-2 were significantly enhanced following in vitro gastrointestinal digestion. As shown in Supplementary Table 3, the radical scavenging activities against ABTS· and ·OH radicals in all post-digestion samples increased in a concentration-dependent manner. For YT013-EPS-1, the IC₅₀ value for ABTS· and ·OH scavenging decreased from 2.173 mg/mL to 0.433 mg/mL and from 7.267 mg/mL to 0.056 mg/mL, respectively (P < 0.0001). Similarly, YT013-EPS-2 exhibited marked reductions in IC₅₀ values, from 13.85 to 0.353 mg/mL for ABTS· (P < 0.0001), and from 1.218 to 0.001 mg/mL for ·OH (P < 0.0001). These findings indicate that the antioxidant activities of YT013-EPS-1 and YT013-EPS-2 were substantially enhanced following simulated gastrointestinal digestion in vitro. The observed enhancement in antioxidant activity is likely attributable to digestion-induced structural changes, such as the breakdown of polymers into low-molecular-weight fragments and the exposure of functional groups (e.g., hydroxyl and carboxyl). These alterations may increase conformational flexibility and improve the accessibility of active sites for radical neutralization. Therefore, gastrointestinal digestion appears to potentiate the antioxidant efficacy of YT013-EPSs, supporting their application in functional foods.

3.6. Effects of YT013-EPS-1 and YT013-EPS-2 on gut microenvironment

3.6.1. Changes in mw, residual carbohydrate, and reducing sugars

To elucidate the degradation and utilization patterns of YT013-EPS-1 and YT013-EPS-2 by gut microbiota, changes in molecular weight (Mw), residual carbohydrates, and reducing sugar concentrations were analyzed. As shown in Fig. 3C and Fig. 3F, significant shifts in the Mw distribution were observed for both YT013-EPS-1 and YT013-EPS-2 after 48 h of in vitro fermentation, with the appearance of low-Mw peaks in the SEC chromatograms. This indicates that the gut microbiota are capable of depolymerizing both EPS fractions into lower Mw fragments. Notably, the main peak of YT013-EPS-1 exhibited an overall shift toward a lower Mw, with no distinct peaks observed in the high Mw region. This indicates that YT013-EPS-1 undergoes more extensive degradation and is more readily degraded and utilized by gut microbiota compared to YT013-EPS-2.

Additionally, the residual carbohydrate content in both YT013-EPS fractions decreased significantly after fermentation (Fig. 3G), suggesting effective catabolism by the intestinal microbiota. Notably, the residual carbohydrate for YT013-EPS-1 (28.19 ± 1.85 %) and YT013-EPS-2 (25.31 ± 1.97 %) was substantially lower (P < 0.0001) than that of the BLK group (86.71 ± 0.98 %), yet showed a slight difference compared to inulin (22.45 ± 1.95 %). This indicates that the gut microbiota demonstrated varying efficiencies in utilizing YT013-EPS-1, YT013-EPS-2, and inulin, likely due to differences in polysaccharide structure and composition.

Furthermore, as illustrated in Fig. 3H, after 48 h of fermentation, the reducing sugar concentration significantly decreased for YT013-EPS-1 from 0.93 ± 0.01 mg/mL to 0.25 ± 0.02 mg/mL (P < 0.0001), and for YT013-EPS-2 from 0.86 ± 0.04 mg/mL to 0.24 ± 0.01 mg/mL (P < 0.0001). Based on these observations, it is hypothesized that during the initial stages of fermentation, the reducing sugar content may have increased due to polysaccharide degradation. However, this increase was followed by a reversal, as microorganisms consumed reducing sugars as metabolic carbon sources at a rate exceeding their production, leading to a net decrease.

3.6.2. Dynamic changes in monosaccharide compositions

To further investigate the fermentation behavior of YT013-EPS-1 and YT013-EPS-2, dynamic changes in monosaccharide composition during in vitro fermentation were evaluated (Supplementary Fig. 2 and Table 2). Although the overall monosaccharide profiles of the fermented samples were comparable to those of the unfermented groups, notable shifts in individual monosaccharide levels were observed. Notably, the glucose level decreased by 68.27 % and 56.57 % in the YT013-EPS-1 and YT013-EPS-2 groups, respectively, indicating that glucose may be preferentially utilized by gut microbes. Interestingly, the relative mannose content increased by 59.89 % and 59.01 % in YT013-EPS-1 and YT013-EPS-2, respectively. Based on the previous study, this increase could be attributed to microbial enzymes selectively degrading glucose-rich regions within the polysaccharides, thereby exposing or releasing mannose residues previously embedded within the matrix. Furthermore, a significant reduction in glucuronic acid content was observed in YT013-EPS-2 following fermentation. Although direct enzymatic assays were not performed in this study, this reduction may be related to the activity of β-glucuronidase, a microbial enzyme known to cleave glucuronic acid from glycosidic linkages (Kumari et al., 2025; Song, Feng, Liu, Duan, & Zhang, 2025). These changes highlight the shift in substrate preference during fermentation, reflecting the dynamic activity of the gut microbiota. It could be inferred that gut microbiota might be prone to break down glucoside linkage in YT013-EPS-1 and YT013-EPS-2, and glucose was the predominant monosaccharide consumed.

Table 2.

Concentration of SCFAs at different time points during in vitro fermentation.

SCFAs (mmol/L)	Sample	Anaerobic fermentation time (h)
		0	48
Acetic acid	BLK	2.07 ± 0.06	8.32 ± 0.33d
	YT013-EPS-1		34.55 ± 0.64a
	YT013-EPS-2		25.95 ± 0.23b
	INU		23.43 ± 0.71c
Propionic acid	BLK	N.D.	1.69 ± 0.02d
	YT013-EPS-1		5.92 ± 0.01b
	YT013-EPS-2		4.31 ± 0.02c
	INU		6.28 ± 0.08a
n-Butyric acid	BLK	N.D.	0.43 ± 0.03d
	YT013-EPS-1		3.87 ± 0.06b
	YT013-EPS-2		3.26 ± 0.07c
	INU		4.15 ± 0.05a
i-Butyric acid	BLK	N.D.	0.38 ± 0.07d
	YT013-EPS-1		0.98 ± 0.08a
	YT013-EPS-2		0.67 ± 0.01b
	INU		0.55 ± 0.01c
n-Valeric acid	BLK	N.D.	0.97 ± 0.01c
	YT013-EPS-1		1.42 ± 0.02b
	YT013-EPS-2		1.30 ± 0.02b
	INU		2.39 ± 0.02a
total SCFAs	BLK	2.07 ± 0.06	12.13 ± 0.15d
	YT013-EPS-1		47.02 ± 1.30a
	YT013-EPS-2		35.90 ± 0.72c
	INU		39.52 ± 0.46b

Note: (1) N.D. stands for “No Data Provided,” indicating that no data were detected. (2) Different lowercase letters indicate significant differences (P < 0.05), while the same letter indicates no significant difference (P > 0.05).

3.6.3. Change in pH and production of SCFAs

Intestinal microbiota can metabolize polysaccharides as essential carbon sources, generating acidic metabolites under anaerobic conditions, thereby lowering the environmental pH. Consequently, pH changes serve as crucial indicators for monitoring polysaccharide degradation and utilization during in vitro fermentation. As presented in Fig. 3I, the initial pH values across all test groups were above 7.00. After 48 h of anaerobic fermentation, the pH of the BLK group decreased slightly, probably resulting from the fermentation of proteins or other components in the culture medium. In contrast, a more pronounced pH decline was detected in the INU group, decreasing to 5.56 ± 0.03 (P < 0.0001). Similarly, notable pH reductions were observed in the YT013-EPS-1 and YT013-EPS-2 groups, with final pH values decreasing to 5.34 ± 0.02 (P < 0.0001) and 6.08 ± 0.01 (P < 0.0001), respectively. The pH decline in the YT013-EPS-1 group was more substantial than that in the YT013-EPS-2 group, likely due to differences in the accumulation of SCFAs or other acidic metabolites during fermentation.

To further investigate the capacity of gut microbiota to metabolize the two YT013-EPS fractions, GC–MS/MS was employed to analyze the SCFA profiles. As shown in Table 2, after 48 h of fermentation,the total SCFA levels in the YT013-EPS-1 (47.02 ± 1.30 mmol/L) and YT013-EPS-2 (35.90 ± 0.72 mmol/L) groups increased significantly compared to the BLK group (2.07 ± 0.06 mmol/L) (P < 0.0001). Moreover, the total SCFA content in the YT013-EPS-1 group exceeded that of the INU group (39.52 ± 0.46 mmol/L), indicating that YT013-EPS-1 possesses superior fermentability compared to the well-recognized prebiotic, inulin. Acetic acid, propionic acid, and n-butyric acid emerged as the predominant metabolites, while i-butyric acid, n-valeric acid, and i-valeric acid were present at relatively low concentrations. Specifically, the concentrations of acetic acid, propionic acid, and n-butyric acid in the YT013-EPS-1 group increased significantly (P < 0.0001) to 34.55 ± 0.64 mmol/L, 5.92 ± 0.01 mmol/L and 3.87 ± 0.06 mmol/L, respectively. In the YT013-EPS-2 group, the corresponding SCFA concentrations reached 25.95 ± 0.23 mmol/L, 4.31 ± 0.02 mmol/L and 3.26 ± 0.07 mmol/L (P < 0.0001). These variations in SCFA profiles between the two EPS fractions are likely attributable to differences in substrate composition and microbial response. N-butyric acid production is primarily driven by the fermentation of mannose, galactose, and galacturonic acid, whereas propionic acid production is predominantly supported by glucose and mannose fermentation.

Differences in fermentation outcomes between YT013-EPS-1 and YT013-EPS-2 may be partly attributed to structural characteristics. FT-IR spectra indicated that YT013-EPS-1 features β-glycosidic linkages (923.1 cm⁻¹), which are typically more resistant to host digestion and thus more likely to reach the colon intact. YT013-EPS-2, in contrast, exhibited signals characteristic of α-glycosidic bonds and uronic acids, which may influence solubility and charge distribution. Additionally, SEM analysis suggested that YT013-EPS-1 possesses a more compact structure with porous microfeatures, potentially indicating higher branching, enhanced water retention, and greater microbial accessibility. These structural features may partly explain the enhanced SCFA production and microbial activity observed; however, further in-depth studies are required to establish causality.

3.6.4. Effect on fecal microbiota

High-throughput sequencing was performed to assess the regulatory effects of YT013-EPS-1 and YT013-EPS-2 on intestinal microbiota. Alpha diversity, reflecting bacterial community richness and diversity, was evaluated using the Shannon, Simpson, ACE, and Chao1 indices. As shown in Table 3, the Coverage index exceeded 0.998 for all samples, confirming sufficient sequencing depth of samples. The YT013-EPS-1 and INU groups exhibited a significant increase in the Shannon and Simpson indices compared to the BLK group (P < 0.001), indicating improved microbial diversity and evenness. In contrast, the Chao1 and ACE indices were significantly lower in the YT013-EPS-1 (P < 0.01), while changes in the YT013-EPS-2 and INU groups were not statistically significant. This apparent discrepancy between richness and diversity may be due to the selective enrichment of specific taxa capable of utilizing polysaccharides, as has been reported in other in vitro studies (Luo et al., 2023). Principal Component Analysis (PCA) revealed distinct clustering patterns among treatment groups (Supplementary Fig. 3), with PC1 and PC2 contributing 42.80 % and 31.46 % of the total variation, respectively. The YT013-EPS-1 group showed clear separation from the BLK group, with partial overlap with the INU group, indicating a shift in microbial structure that resembles the pattern induced by inulin. The YT013-EPS-2 group, in contrast, clustered more closely with the BLK group.

Table 3.

Effect of YT013-EPS-1 and YT013-EPS-2 on the alpha-diversity.

	Chao1	Ace	Shannon	Simpson	Goods_coverage
BLK	478.21 ± 18.66a	489.69 ± 26.35a	4.03 ± 0.11c	0.83 ± 0.01b	99.91 ± 0.0002 %a
INU	474.59 ± 2.37a	480.13 ± 19.86a	4.88 ± 0.26a	0.92 ± 0.05a	99.90 ± 0.0001 %a
YT013-EPS-1	433.33 ± 11.01b	441.55 ± 23.18b	4.53 ± 0.32b	0.91 ± 0.02a	99.89 ± 0.0001 %a
YT013-EPS-2	476.10 ± 17.83a	478.56 ± 12.94a	4.17 ± 0.27c	0.81 ± 0.01b	99.88 ± 0.0002 %a

Note: Different lowercase letters indicate significant differences (P < 0.05), while the same letter indicates no significant difference (P > 0.05).

At the phylum level, the dominant bacterial phyla across all treatment groups included Pseudomonadota, Bacteroidota, Bacillota and Actinomycetota, collectively constituting over 90 % of the microbiota (Fig. 5A). Notably, the relative abundance of Pseudomonadota was significantly decreased in the YT013-EPS-1 and INU groups compared to the BLK group (P < 0.0001), reaching 22.81 % and 26.93 %, respectively. In contrast, the YT013-EPS-2 group exhibited a markedly higher abundance of Pseudomonadota, reaching 52.45 %, which was significantly greater than that in the other groups (P < 0.0001). Given that certain Pseudomonadota species have been linked with intestinal dysbiosis (Park et al., 2025), the observed reduction may reflect a shift toward a potentially more balanced microbial community. Furthermore, the relative abundance of Bacteroidota and Bacillota significantly increased in both YT013-EPS-1 and YT013-EPS-2 groups compared to the BLK group (P < 0.0001), suggesting that these EPSs may serve as carbon and energy sources supporting their growth. The enrichment of Bacteroidota may contribute to polysaccharide degradation through increasing the production of depolymerases and glycoside hydrolases, thereby facilitating SCFA formation (Rad et al., 2025). After 48 h of fermentation, Bacillota abundance also rose markedly in the YT013-EPS-1 (15.00 %, P < 0.0001), YT013-EPS-2 (11.29 %, P < 0.05), and INU (42.10 %, P < 0.0001) groups versus the BLK group (7.71 %). Given its known role in SCFA production from complex carbohydrates, the increase in Bacillota likely reflects its active participation in EPS and inulin metabolism, in line with prior reports on the effects of dietary polysaccharides on gut microbiota (Yüksel, Voragen, & Kort, 2024). A similar beneficial function has been reported for Actinomycetota. After fermentation, a significant increase in Actinomycetota abundance was observed in both the YT013-EPS-1 and INU groups (P < 0.001). Actinomycetota are considered probiotic-related taxa, associated with cholesterol regulation, intestinal protection, and immune modulation (Owens, Ahmed, Lang Harman, Stewart, & Mori, 2024). Taken together, these findings suggest that supplementation with YT013-EPSs and inulin selectively promotes the growth of specific gut microbiota, with Bacillota and Bacteroidota likely serving as key contributors to YT013-EPS degradation and utilization. Moreover, the modulatory effect of YT013-EPS-1 on gut microbiota shows a similar trend to the well-recognized prebiotic, inulin. Notably, although YT013-EPS-1 and inulin induced similar shifts in gut microbiota composition, the distinct SCFA profiles likely reflect differences in species-level enrichment and polysaccharide structure that affect fermentation efficiency.

Fig. 5.

Effects of YT013-EPS-1 and YT013-EPS-2 on intestinal flora. (A) phylum level; (B) genus level; (C) LefSe analysis.

The top 30 genera affected by different substrates are shown in Fig. 5B, with key differential taxa identified via LEfSe analysis (Fig. 5C). Among these, the top 10 genera with the highest relative abundance included Escherichia-Shigella, Dubosiella, Bacteroides, Parabacteroides, Turicimonas, Phascolarctobacterium, Candidatus_Saccharimonas, Alistipes, Acinetobacter, and Lactobacillus. Compared with the BLK group, the relative abundance of Escherichia-Shigella, a group associated with gastrointestinal infection (Wu, Xiong, Zhan, & Xu, 2024), was significantly reduced in the YT013-EPS-1 and INU groups (P < 0.0001). Additionally, Dubosiella was enriched in YT013-EPS-1 (P < 0.001) and INU groups (P < 0.0001), a genus linked to SCFA production and modulation of the immune system, including enhancement of Treg/Th17 balance and intestinal barrier integrity (Pan et al., 2025). Bacteroides and Parabacteroides were also enriched, particularly in the YT013-EPS-1 and YT013-EPS-2 groups. These genera are well known for their roles in complex polysaccharide metabolism and SCFA production (Duan et al., 2024). Notably, this increased abundance of Parabacteroides may be attributed to the presence of multiple polysaccharide utilization loci (PULs), conferring a competitive advantage in the intestinal microbial ecosystem by facilitating the degradation of complex polysaccharides (Imanbayev et al., 2024). Phascolarctobacterium, significantly increased (P < 0.001 for all treatments) across all EPS-treated and INU groups, is capable of converting succinate to propionate and has been associated with microbial stability and potential inhibition of pathogens like Clostridioides difficile (N. H. S. Chu, Chow, & Chan, 2024). Importantly, the relative abundance of Acinetobacter, a conditionally pathogenic bacterium, was significantly reduced following EPS treatment (P < 0.0001). Lactobacillus, a genus frequently enriched by dietary fiber and often associated with health-promoting properties such as lactic acid production and pathogen inhibition, was significantly increased after 48 h fermentation (P < 0.001 for all treatments), in agreement with prior in vitro fermentation studies.

Taken together, these in vitro findings indicate that YT013-EPS-1 and YT013-EPS-2 selectively influenced the composition of gut microbiota, including increased relative abundance of several genera previously associated with beneficial metabolic functions. However, further in vivo studies are necessary to verify their prebiotic potential and health relevance. The species-level relative abundance of the gut microbiota is shown in Supplementary Table 4.

3.7. Clinical potential and future perspectives

YT013-EPS-1, a structurally complex branched heteropolysaccharide, exhibits promising prebiotic potential through its ability to selectively modulate gut microbiota and promote sustained SCFA production. In vitro fermentation data revealed enrichment of beneficial SCFA-producing genera such as Bacteroides, Parabacteroides, Phascolarctobacterium, and Lactobacillus, coupled with the suppression of opportunistic pathogens like Escherichia-Shigella and Acinetobacter. These microbial shifts suggest a possible relevance to dysbiosis-associated conditions such as inflammatory bowel disease (IBD), metabolic syndrome, and functional gastrointestinal disorders; however, this remains to be verified in vivo. Compared to rapidly fermentable prebiotics such as inulin, YT013-EPS-1 displayed a more gradual fermentation pattern in vitro, which may reduce gastrointestinal intolerance symptoms such as bloating and gas, common side effects associated with prebiotic intake. This is particularly relevant given the precedent of lactulose, a clinically used prebiotic known to modulate gut microbiota but often limited by dose-dependent tolerance issues, as described by Chu et al. (N. Chu, Ling, Jie, Leung, & Poon, 2022). Nevertheless, it is important to acknowledge that the current evidence is derived exclusively from in vitro models. While such models offer high controllability and reproducibility, they cannot fully replicate the complex, dynamic interactions within the human gastrointestinal tract. Therefore, future research should prioritize well-designed human clinical trials to assess efficacy, safety, and tolerability. Additionally, mechanistic studies utilizing omics approaches (e.g., metagenomics, metabolomics) may provide insights into the molecular pathways involved. Optimized delivery strategies, such as microencapsulation or co-administration with compatible probiotics, may further enhance its functional stability and bioavailability. Overall, these findings position YT013-EPS-1 as a promising prebiotic candidate with translational potential, warranting further clinical and mechanistic validation.

4. Conclusion

This study comprehensively investigated the structural characteristics, gastrointestinal stability, and prebiotic potential of two exopolysaccharide fractions, YT013-EPS-1 and YT013-EPS-2, isolated from Lactiplantibacillus plantarum YT013. Structural analysis revealed that YT013-EPS-1 is a neutral heteropolysaccharide, while YT013-EPS-2 is an acidic heteropolysaccharide, with Mw of 57.51 kDa and 149.3 kDa, respectively. Compared with YT013-EPS-2, YT013-EPS-1 exhibited greater resistance to simulated gastrointestinal digestion, maintaining its structural integrity and demonstrating enhanced antioxidant activity despite undergoing partial structural modifications. During in vitro fecal fermentation, both YT013-EPS-1 and YT013-EPS-2 were effectively metabolized by gut microbiota, leading to a significant increase in SCFA production, particularly acetic, propionic, and butyric acids. Microbiota composition analysis demonstrated that supplementation with YT013-EPSs selectively modulated the gut microbial community, promoting the proliferation of beneficial bacteria such as Phascolarctobacterium, Lactobacillus, Bacteroides, and Parabacteroides, while reducing the abundance of potentially pathogenic genera like Escherichia-Shigella and Acinetobacter. Notably, YT013-EPS-1 exhibited superior gastrointestinal stability and prebiotic properties compared to YT013-EPS-2. These findings highlight the potential of YT013-EPSs, particularly YT013-EPS-1, as promising functional ingredients with prebiotic effects, contributing to modulating gut microbiota composition and SCFA production under controlled in vitro conditions. However, it is important to note that these results are based on in vitro experiments; therefore, further in vivo studies are necessary to validate these findings and determine their relevance to gut health and microbiota homeostasis.

Author contribution

Min Chen: Conceptualization, Methodology, Investigation, Data curation, Writing - Original draft. Hua Zhang: Data curation, Writing - Review & Editing. Rui Yang: Methodology, Formal analysis, Visualization. Xinhang Wang: Validation, Writing - Review & Editing. Liqian Du: Investigation, Data curation. Yuqi Yue: Data curation, Methodology. Xin Ma: Writing - Review & Editing, Visualization. Fanting Zhao: Resources, Formal analysis. Yingjie Liu: Resources, Formal analysis. Zhongkun Zhou: Data curation, Methodology. Yunhao Ma: Investigation, Data curation. Baizhuo Zhang: Data curation, Methodology. Yuanchun Zhao: Data curation, Methodology. Yan Jin: Visualization, Formal analysis. Hongmei Zhu: Writing - Review & Editing. Peng Chen: Supervision, Project administration, Writing - Review & Editing, Funding acquisition.

CRediT authorship contribution statement

Min Chen: Writing – original draft, Methodology, Investigation, Data curation, Conceptualization. Hua Zhang: Writing – review & editing, Data curation. Rui Yang: Visualization, Methodology, Formal analysis. Xinhang Wang: Writing – review & editing, Validation. Liqian Du: Investigation, Data curation. Yuqi Yue: Methodology, Data curation. Xin Ma: Writing – review & editing, Visualization. Fanting Zhao: Resources, Formal analysis. Yingjie Liu: Resources, Formal analysis. Zhongkun Zhou: Methodology, Data curation. Yunhao Ma: Investigation, Data curation. Baizhuo Zhang: Methodology, Data curation. Yuanchun Zhao: Methodology, Data curation. Yan Jin: Visualization, Formal analysis. Hongmei Zhu: Writing – review & editing. Peng Chen: Writing – review & editing, Supervision, Project administration, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material

Data availability

Data will be made available on request.