Animal experiments were conducted in accordance with the procedures and guidelines set forth by the Animal Care and Use Committee of the Institute of Subtropical Agriculture, Chinese Academy of Sciences, and were approved under the designation ISA 2020-18.

Dried B. papyrifera powder (0.5 kg), sourced from the local market, was soaked in 95% ethanol at a ratio of 1:10 (w/v) and stored at room temperature for one week. The supernatant was collected and filtered. This procedure was performed three times, and the supernatant was concentrated using a vacuum rotary evaporator (RE 2008, Shengye, Shanghai, China) paired with a cooling system (DLSB-10L-10, Yuhua, Hunan, China). The concentrated extract was then decolorized with petroleum ether and further concentrated by evaporation, culminating in freeze drying.

The levels of total flavonoids, total polyphenols, total sugar, and total antioxidant capacity were determined following the protocols associated with the relevant kits obtained from Suzhou Keming Biotechnology Co., Ltd. (Suzhou, China). The active compounds within BPE were examined using liquid chromatography-tandem secondary mass spe ctrometry (LC–MS/MS). The instrument parameters, running conditions, and analysis protocols were aligned with our previously published methods (Zhai et al., 2021). An UltiMate 3000 (Thermo Fisher Scientific, Waltham, MA, USA) equipped with a C18 chromatographic column was used. Mobile phases A and B consisted of 100% acetonitrile and 0.1% formic acid and H₂O and 0.1% formic acid, respectively. The elution gradient ran as follows: A: 0 min–7 min, 5%–50%; 7 min–8 min, 50%–75%; 8 min–9 min, 75%–80%; 9 min–11 min, 80%–90%; 11 min–15 min, 90%–95%; 15 min–20 min, 95%. The flow rate was maintained at 0.2 mL/min. MS/MS, equipped with Q-Exactive Focus (Thermo Fisher Scientific, Waltham, MA, USA), was configured as follows: ion source: ESI, atomization temperature: 300°C, sheath gas pressure: 40 ARB, auxiliary gas pressure: 10 ARB, transmission capillary temperature: 320°C, scanning mode: full scan, resolution 35,000, in-source CID: 0 ev, ddms2, resolution 17,500, HCD stepped NCE 10, 30, 50. The top 20 metabolites of BPE were identified.

The inhibitory effects of BPE were determined according to a previously described method (Niu et al., 2019b). A total of six indicator pathogens implicated in animal intestinal diseases, including E. coli W25K, E. coli ATCC 35150, E. coli 8700 1.558, Salmonella typhimurium, Salmonella enteritis CMCC S007, and Clostridium perfringens ATCC 9120, were employed. Overnight cultures of these pathogens were prepared in LB (Luria–Bertani) medium. The final concentrations of BPE were adjusted to 25 mg/mL, 50 mg/mL, and 100 mg/mL in LB solution containing 1 × 10⁶ CFU/mL of the pathogens, and 1 mL of this solution was added to the wells of a 24-well plate (JET, Guangzhou, China). The plates were then incubated for 24 h at 37 °C. Samples were extracted at 0 h, 4 h, 6 h, 8 h, 12 h, and 24 h for viable cell number determination (Log₁₀ CFU/mL) using an agar plating method. Distilled water served as a control, and the experiments were performed in triplicate.

A total of 36 female C57BL/6 mice (20 g–25 g, aged 11 weeks) were acquired from the STJ Laboratory Animal Co., Ltd. (Hunan, China) for the purpose of this study. The mice were maintained under controlled conditions: temperature (22°C –25°C), humidity (55%–75%), with 12-hour light-dark cycles, and unrestricted access to both food and drinking water. Following a one-week acclimation period, the mice were randomly assigned to four distinct treatment groups: the control group (Con), BPE group (BPE), dextran sulfate sodium group (DSS), and DSS + BPE group (DSS + BPE). Each group contained three replicates, and each replicate consisted of three mice that were housed in a single cage.

For the mice in the Con and BPE groups, distilled drinking water was provided ad libitum, while the mice in the DSS and DSS + BPE groups were administered drinking water infused with 2% (w/v) DSS (36,000 M. Wt.–50,000 M. Wt., MP Biomedicals, Solon, OH, USA) over 7 days ( Figure 2A ). Concurrently, the mice in the BPE and DSS + BPE groups were subjected to oral gavage with 200 mg/kg body weight of BPE, whereas mice in the Con and DSS groups received an equivalent volume of PBS daily. The adopted dosage of BPE in the mouse experiment was described in an earlier study (Xu et al., 2022). All mice were subjected to weight assessment and daily evaluation of the diarrhea index. On Day 11, following three additional days of feeding, the mice were humanely sacrificed to facilitate sample collection. The sampling methodology was consistent with that described in an earlier study (Zhai et al., 2021). The mice were fasted for 6 h and were then euthanized by intraperitoneal injection of 2% pentobarbital sodium (45 mg/kg body weight). Blood samples were collected via enucleation of the eyeball. The distal ileum (near the cecum to 2 cm) and colon were collected and fixed in 4% neutral polyformaldehyde for intestinal histological analysis or frozen in liquid nitrogen for real-time quantitative PCR (RT-qPCR) analysis. Colonic contents were collected for the analysis of the intestinal microbiota.

Intestinal histological alterations were evaluated by hematoxylin and eosin (H&E) staining. In brief, neutral polyformaldehyde-fixed ileum and colon samples were dehydrated using ethanol and xylene, followed by paraffin embedding. Paraffin-embedded tissues were then sectioned into 5 μm slices, deparaffinized, hydrated, and stained with H&E. Villus length (VL) and crypt depth (CD) were measured using an Olympus microscope paired with VistarImage software (Olympus, Japan) (Zhai et al., 2019).

The protocol for colon immunostaining was performed in accordance with previously established methods (Zhai et al., 2021). Following deparaffinization, rehydration, and serum blocking following antigen retrieval, the tissue sections were treated with primary antibodies. After overnight incubation at 4°C, the tissues were exposed to secondary antibodies and incubated in the dark at room temperature for 50 min. The slides were then treated with DAPI solution for 10 min and stored in the dark. A Nikon Eclipse C1 fluorescence microscope equipped with a Nikon DS-Us imaging system was used for image capture and analysis.

Quantitative real-time PCR (RT-qPCR) was performed to assess colon gene expression associated with antioxidant activity, inflammation, and epithelial barrier integrity following an established method (Zhai et al., 2021). The mRNA of colon samples was extracted using a columnar RNA extraction kit (Magen, Guangzhou, China), and the concentration was measured using a NanoDrop 2000C spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). RNA was reverse-transcribed into cDNA using a commercial kit (CW2582M, CWBIO, Jiangsu, China). The primers used in this study were designed and synthesized by Shanghai Shenggong Biotechnology Co., LTD (Shanghai, China). The primer information is presented in Supplementary Table 1 . RT-qPCR was conducted using a FastSYBR kit (CW0955M, CWBIO) following the manufacturer’s protocol and run in an ABI 7500 FAST system (Applied Biosystems Instruments, Thermo Fisher Scientific, United States) under standard running conditions. qRT-PCR was performed using the Ct value as the PCR threshold cycle number at the endpoint. The relative expression levels of target genes were normalized to the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GADPH) using the ΔΔCT method (Ru et al., 2022).

Genomic DNA (gDNA) extraction, quality assessment, sequencing, and related analyses were performed in accordance with previously described methods (Niu et al., 2019a). Briefly, gDNA was extracted using the CTAB method (Niu et al., 2019a) and the concentration was quantified using a NanoDrop 2000. Purity and quality were determined using 1% agarose gel electrophoresis. The V4 hypervariable region of 16S rRNA was amplified using universal primers 515F and 806R. PCR and sequencing were performed on an Illumina NovaSeq 6000 platform using the commercial services of Norson Bioinformatics Co., Ltd. (Beijing, China). Sequence clusters exhibiting 97% similarity were identified as identical operational taxonomy units (OTU) utilizing UParse (version v 1.9.1). The Silva database (http://www.arb-silva.de/) with the Mothur algorithm was used to annotate the classification information of representative sequences. Alpha diversity indexes, such as Observed_species, Chao1, Simpson, and Shannon, were analyzed in QIIME (Version 1.7.0) and visualized using R software (Version 2.15.3). Principal component analysis at the OTU level was calculated in QIIME (Version 1.9.1).

Untargeted metabolomic analyses of colonic samples were performed using an LC–MS/MS platform. Data were interpreted using the commercial services of Novogene Biotech Co., Ltd. (Beijing, China). Colonic sample preparation and LC–MS/MS analysis conditions were implemented in accordance with a previously described method (Lu et al., 2019). Metabolites with VIP >1.5, P-value <0.05, and fold change (FC) >2 were designated as differentially expressed metabolites (DEMs). The identified metabolites were functionally annotated using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database and the annotated metabolites were mapped to the KEGG pathway database. The enrichment of metabolic pathways among the differential metabolites was subsequently performed.

The data generated in the mouse study were subjected to statistical analysis by one-way analysis of variance (ANOVA) using SPSS (version 20, SPSS Inc., Chicago, IL, USA). Tukey’s HSD multiple range test was used to ascertain statistically significant disparities (P <0.05) in the mean responses to distinct treatments. Data are presented as the mean ± standard error (SEM) of the mean of triplicate groups. Notations such as * denotes P <0.05, ** represents P <0.01, and *** indicates P <0.001. GraphPad Prism 7 was used for figure creation.

The total contents of flavonoids, polyphenols, and sugars in BPE were quantified as 67.72 mg/g, 343.62 mg/g, and 181.81 mg/g, respectively. The scavenging activity against superoxide radicals was 45.53%, and the principal bioactive compounds are detailed in Table 1 . Within BPE, the six predominant bioactive compounds, quinic acid, geniposidic acid, 1-caffeoyiquinic acid, cryptochlorogenic acid, caffeic acid, and citric acid, each constituted more than 1%.

The in vitro inhibitory activities of BPE against several prevalent intestinal pathogens were evaluated ( Figure 1 ). Three different concentrations of BPE were found effective in impeding the growth of C. perfringens ATCC 13124 ( Figure 1C ), S. typhimurium CMCC 50115 ( Figure 1D ), and S. enterica subsp. enterica ATCC 9120 ( Figure 1E ), without affecting the growth of enterotoxigenic E. coli W25K ( Figure 1A ), and P. aeruginosa SSI 82000 ( Figure 1B ). Only the highest concentration of BPE (100 mg/mL) demonstrated an antagonistic effect on the growth of Salmonella paratyphi B ATCC 50094 ( Figure 1F ).

A mouse model of colitis was established by administering 2% DSS in drinking water over a 7-day period ( Figure 2A ). The effect of BPE on both normal (PBS gavage) and DSS-treated mice was explored. Compared to the control (Con) treatment, DSS treatment led to a marked reduction in body weight ( Figure 2B ), augmentation of liver and spleen indices ( Figure 2D ), and rapid onset of diarrhea ( Figure 2C ). While co-treatment with BPE did not counteract weight loss, it moderately mitigated diarrhea severity and liver weight gain compared with DSS treatment alone. The singular BPE treatment did not cause a notable change in body weight or organ index but induced slight diarrhea after 5 days of gavage. Colon length did not differ significantly among the four treatments ( Figure 2E ).

The effect of BPE on the intestinal morphology of both normal and DSS-treated mice was examined ( Figure 3 ). DSS treatment severely destroyed the colonic structure, which BPE co-treatment did not ameliorate by BPE cotreatment. Isolated BPE treatment caused minor damage to the colonic structure. In terms of ileal structure, no conspicuous differences were observed among the four treatments ( Figure 3A ). From numerical statistics, neither single nor co-BPE treatment led to significant alterations in villus length (VL), crypt depth (CD), and VL/CD ratio within the ileum and colon in comparison with the Con and DSS treatments.

We investigated the effects of BPE on colonic gene expression related to antioxidant activity, inflammation, and barrier function ( Table 2 ). DSS treatment markedly downregulated the relative gene expression of CYP1A1, Nrf2, and TLR4, yet co-treatment with BPE did not suppress the expression of these genes when compared with Con treatment. A single BPE treatment did not also cause significant changes in any of the tested genes. Furthermore, the impact of BPE on the expression of proteins associated with antioxidant and barrier functions was assessed using colon immunofluorescence ( Figure 4 ). In comparison to Con treatment, DSS treatment significantly reduced the protein expression of Nrf2, claudin-1, occludin, and ZO-1, and BPE co-treatment did not restore these levels. Treatment with isolated BPE did not induce any notable differences in these proteins. For AHR protein expression, a single BPE treatment increased its expression, as did co-treatment with BPE and DSS. These results demonstrate that BPE was not efficacious in rectifying DSS-induced dysfunctions related to intestinal antioxidant and barrier functionalities at both the gene and protein expression levels.

The 16S rDNA sequencing methodology was used to assess the influence of BPE on the colonic microbiota. In comparison with the control (Con) treatment, there was a significant reduction in ACE- and Chao1-based α-diversity of the colonic microbiota in mice treated with DSS. This effect was reversed by BPE co-treatment; however, individual administration of BPE did not induce notable differences in α-diversity indices ( Table 3 ). Regarding the similarity of colonic microbiota among the four treatments, the PCoA plot illustrated a distance order of DSS + BPE > DSS > BPE when compared with the Con treatment ( Figure 5A ). A Venn diagram displayed 626 core OTUs among the four treatments, with 218, 93, 123, and 71 unique OTUs identified in the Con, BPE, DS S, and BPE + DSS groups, respectively ( Figure 5B ). The colonic bacterial composition was profiled at both phylum ( Figure 5C ) and genus ( Figure 5D ) levels. At the phylum level, Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria were the predominant phyla across all treatments. In comparison with Con treatment, DSS induced a significant decrease in the abundance of Firmicutes and Tenericutes, accompanied by an increase in Bacteroidetes abundance. Conversely, BPE supplementation in DSS-treated mice did not affect the altered phyla. A single BPE treatment did not alter the dominant phyla relative to those in the Con treatment. At the genus level, Dubosiella and Turicibacter were the most dominant genera among the four treatments. Compared with the Con treatment, DSS treatment resulted in a significant surge in Bacteroides abundance, an effect that was not restored by BPE co-treatment. Individual BPE treatments notably increased Faecalibaculum abundance, but co-treatment with BPE decreased it. Relative to the Con and DSS groups, BPE addition significantly augmented the abundance of Akkermansia.

Untargeted metabolomics was used to investigate the effects of BPE in normal and DSS-treated mice ( Figure 6 ). In contrast to the Con group, individual BPE treatments led to more clustered colonic metabolites, comprising 161 differentially expressed metabolites (DEMs) such as aspartic acid, cysteine-S-sulfate, 3,3-Dimethylglutaric acid, N-acetylonithine, lyxose, and vanillin. These mainly corresponded to the enriched pathways of alanine, histidine, arginine, and proline metabolism. DSS treatment induced a completely distinct metabolite profile, encompassing 218 DEMs, including flavin adenine dinucleotide, 3,3-Dimethylglutaric acid, aspartic acid, uric acid, 2(13)-Dihome, and N-Isovalerylglycine. These primarily belong to the enriched pathways involving bile secretion, steroid hormone biosynthesis, ferroptosis, and arachidonic acid metabolism. Supplementation with BPE in DSS-treated mice failed to restore the altered metabolite profile, with only 24 DEMs detected, such as 2-Hydroxyphenylacetic acid and arabitol, mainly corresponding to a metabolic pathway.