T. chebula fruits were purchased from the Hehua Chi Medicine Market in Chengdu, Sichuan Province, and the sample was preserved and identified by Dr. Gao Zhou from Hubei University of Technology (sample number: 20211213). Dextran Sulfate Sodium Salt (DSS) (MW: 36000∼50000) was purchased from MP Biomedicals company (Santa Ana, California, United States). 5-Aminosalicylic acid was purchased from Shanghai Yi’en Chemical Technology Co., Ltd. (Shanghai, China). CAT, GSH, MDA, GSH assay kits, and staining reagents were purchased from Beyotime Biotechnology (Shanghai, China). All analytical standards were purchased from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China). TLR4, β-actin antibody, and horseradish peroxidase-conjugated anti-rabbit IgG purchased from Servicebio Biotechnology Co., Ltd. (Wuhan, China). MyD88, NF-κBp65 antibody purchased from Proteintech Group (Wuhan, China). Methanol and acetonitrile were chromatographically pure from Thermo Fisher Scientific Inc. (Shanghai, China).

Forty male SPF-grade C57BL/6 mice, weighing 20–25 g, were purchased from Hubei Provincial Center for Disease Control and Prevention. The animals received food and water freely at a room temperature of 23°C ± 2°C and a humidity of 55% ± 5% for 5 days of accommodation before the experiments. All experimental operations were carried out under the supervision of the IACUC (Institutional Animal Care and Use Committee) of the Hubei University of Technology, and experimental manipulations followed the relevant regulations.

Crushed the fruits of T. chebula with a grinder, weighed 200 g powder, extracted three times with ten times volume of 95% ethanol by ultrasonication, and concentrated by rotary evaporation after the extraction. A certain amount of ethanolic extract of T. chebula be diluted using pure water and sequentially extracted with ten times the volume of petroleum ether, dichloromethane, ethyl acetate, and n-butanol, the extracted solutions concentrated by rotary evaporation and dried to obtain five extract fractions.

Using Thermo Fisher U3000 high-performance liquid chromatography system with a C¹⁸ column (COSMOSIL, 4.6 mm × 250 mm, 5 μm) was used to distinguish components from the total 95% ethanol extract and five fractions of T. chebula and then characterize the main active ingredients in TCEA by using six standard compounds. The injection volume was 10 μL, the flow rate was set to 1 mL/min, the column temperature was kept at 30°C, and the wavelength of the UV detector was set at 275 nm. The mobile phase comprised solvent A (0.1% formic acid in water) and solvent B (acetonitrile). The gradient elution was constant at 5% solvent B for 5 min, followed by a rise from 5% to 10% in 6 min, from 10% to 15% in 5 min, and kept at 15% for 4 min. In the next 9 min, solvent B rose to 18%, then continued to rise to 20%, maintained for 14 min, and finally, solvent B returned to 5% from 20% within 3 min, followed by a gradient equilibration at 5% for 4 min.

Forty C57BL/6 male mice were randomly divided into five groups with eight mice in each group: Control group; DSS group; 5-ASA group (200 mg/kg); TCEA low-dose group (100 mg/kg); and TCEA high-dose group (200 mg/kg). All groups, except the control group, were given 2.5% DSS freely for six consecutive days to build an acute UC model. The 5-ASA group and the TCEA-treated group were intervened according to the above doses by intragastric administration once daily. In contrast, the control and DSS groups were given the same volume of saline. On day 7, DSS was stopped provided in all groups, and saline, 5-ASA, and TCEA were continued to be administered to corresponding groups. On day 8, all mice were sacrificed by carbon dioxide (Figure 3A).

From day 1 to day 8, relevant indices were recorded daily, including body weight, diet consumption, water consumption, fecal traits, and blood in the stool. The disease activity index (DAI) can calculate as follows: DAI = (weight change score + fecal trait score + blood in stool score)/3 (Cooper et al., 1993). In the fecal occult blood test, the score of visible blood in stool was 4, and the rest of the stool was detected by the o-tolidine method. DAI scoring criteria are shown in Table1 below:

Large intestines were collected, photographed, and measured for length after sacrificing the mice. Colons without the cecum were washed with sterile saline to remove feces and stored in a tube for further experiments. About 1 cm of middle colon tissue was cut off and fixed with 4% paraformaldehyde. The fixed colonic tissues were dehydrated, embedded, and sectioned. Hematoxylin-eosin (HE), alcian blue (AB), and periodic acid-Schiff (PAS) staining were employed, respectively, to monitor the morphological changes among the five groups. The staining results will be photographed using a light microscope (Olympus, Japan). The score for the pathology of mice colon tissue was made by referring to several kinds of literature (Wu et al., 2020; Pan et al., 2022; Xiong et al., 2022), as detailed in Table 2.

Colonic tissues were homogenized on ice, and the protein concentration of the homogenate of each sample was determined using the Komas Brilliant Blue method after centrifuging at 10,000 g for 10 min. The level of CAT, GSH, MDA, and SOD was quantified by commercial kits to determine the degree of oxidative damage of colonic tissues in each group.

QRT-PCR detected TLR4/MyD88/NF-κB pathway-related genes transcription in mice colonic tissues. Colonic tissue was cut up into small pieces with scissors and homogenized thoroughly by an electric homogenizer. RNA was extracted from colonic tissue with Trizol reagent. The purity and concentration of the extracted RNA were determined using NanoDrop 2000 (Thermo Fisher Scientific, United States). The RNA was reverse transcribed to cDNA using a reverse transcription kit (Vazyme, China). Then qRT-PCR was performed according to the instructions of AceQ Universal SYBR qPCR Master Mix (Vazyme, China). qRT-PCR reaction procedures were set as one cycle at 95°C for 5 min and 40 cycles at 95°C for 10 s and 60°C for 30 sec. Calculations were performed using the 2^−ΔΔCT method and β-actin as an internal reference. The primer sequences of the relevant genes are shown in Table 3.

PBS buffer containing PMSF and phosphatase inhibitor was added to colonic tissue to extract total proteins, and the Komas Brilliant Blue method was used to determine the protein content of supernatant after the colonic tissue was homogenized on ice and centrifuged at 12,000 g for 10 min. The extracted proteins were mixed by loading buffer and boiled for 10 min to denaturation. Extracted proteins were separated by sodium dodecyl sulfate-polyacrylamide electrophoresis (SDS-PAGE) on a 10% polyacrylamide gel. The proteins in the gel were then transferred to a nitrocellulose (NC) membrane, and then the membrane was blocked with 5% skimmed milk for 1.5 h. After incubation of the primary antibody overnight at 4°C, the membrane was washed three times with TBST. After incubating with horseradish peroxidase (HRP) labeled secondary antibody for 1 h at room temperature, the membrane was washed three times with TBST again for further exposure. Finally, protein visualization analysis was performed by the Enhanced chemiluminescence (ECL) system, and the results were quantified using ImageJ software.

Colon tissues were rinsed with sterile saline, and intestinal contents were collected in sterilized EP tubes. Based on several results, such as colon length, weight, and tissue staining, we found that the TCEA-L group significantly affected DSS-induced UC. Therefore we chose the TCEA-L group for the subsequent intestinal flora analysis. Six samples in control, DSS, and TCEA-L groups were randomly selected for the next step of the intestinal flora study. After extracting each sample’s DNA, the DNA’s purity and concentration were analyzed by agarose gel electrophoresis. The V3-V4 variable region was amplified through PCR technology using diluted DNA as a template. PCR products were purified and normalized to equal concentrations after they qualified. NovaSeq 6000 was employed for sequencing after library construction. The original tags data was obtained after splicing the reads of each sample using FLASH, and the quality control of original tags data was finally carried out to gain effective tags. The effective tags of all samples were clustered by the Uparse algorithm, with 97% of the consistency will become the operational taxonomic units (OTU). The Venn diagram, sample richness and diversity, phylum and genus level difference, principal component, LEfSe, and function prediction among the three groups were then analyzed by Qiime, R, LEfSe, and other software. The whole service was done by Metware Biotech (Wuhan, Hubei, China).

Data are shown as means ± SD from three to eight replicate experiments using different mice. Real-time PCR data represent the mean ± SE for triplicate measurements on each tissue, with tissue sets isolated from three mice. Experimental data were analyzed using SPSS 26.0 (Chicago, IL, United States) software. Statistical differences between multiple groups were analyzed via one-way analysis of variance (ANOVA) followed by the least significant difference (LSD) test for comparisons. p-values less than 0.05 were considered statistically significant differences.

The extraction rate of T. chebula powder by ultrasonic extraction was 42.55%. After extraction with different solvents, the ethyl acetate fraction showed the highest extraction rate, and the petroleum ether fraction showed the lowest extraction rate, 44.479%, and 0.093%, respectively. The remaining parts included dichloromethane, n-butanol, and aqueous fractions, extracted at 0.640%, 15.882%, and 24.628%, respectively. By comparing the HPLC profiles of the five fractions, we observed that the components in petroleum ether, dichloromethane, and aqueous phase were relatively few. Still, the components mostly clustered in the ethyl acetate fraction compared with other fractions, as shown in Figure 1. To find out which main components are contained in the ethyl acetate extract, we performed qualitative analysis by standard compounds. It can be seen from Figure 2 that the ethyl acetate fraction contains chebulic acid, gallic acid, corilagin, chebulagic acid, ellagic acid, and chebulinic acid, where chebulinic acid, gallic acid, and chebulagic acid showed relatively high abundance under a UV absorption wavelength of 275 nm.

After 3-day modeling, the feces of the DSS group began to be shapeless, and the DAI score increased significantly. After 7-day modeling, the mice in the DSS group were listless and appeared to have visible symptoms, including severe diarrhea and stool bleeding. At the same time, the DAI score reached its peak. Compared with the DSS group, mice in 5-ASA and TCEA-administered groups showed more activities and lower DAI scores. Meanwhile, the body weight of the mice in the DSS group decreased significantly throughout the experiment, with the highest rate of weight loss reaching 28.93%. 5-ASA and TCEA administration groups also had different degrees of body weight loss, but it was better than the DSS group (Figures 3B, C). Colon shortening is an essential feature of ulcerative colitis. Our results indicated that the colon length of the control group is the longest, with a reduction of more than 25% in the DSS group compared to the control group.5-ASA and different doses of TCEA treatments could ameliorate the colon shortening caused by DSS modeling (Figure 3D).

HE staining results can be observed from Figure 4A, which demonstrated that the colonic mucosal, submucosal, and muscle layers were intact, the crypt surface was regular, the crypt was neatly arranged, and goblet cells were abundant in the control group mice. On the contrary, significant damage to the mucosal layer of colonic tissue, obvious destruction of glands and crypt, goblet cell loss, and inflammatory cell infiltration were observed in DSS group mice. According to the colon histopathological score (Figure 4B), it was found that the score of the DSS group was more than ten times higher than that of the control group, presenting an obvious upward trend, while the scores of the positive drug group and the TCEA treatment group decreased to a certain extent, among which the downward trend of TCEA-L was more evident. Mucus from goblet cells plays a vital role in intestinal protection. From the results of AB-PAS staining, we could see that the colon of the control group was not only full of mucus but also rich in goblet cells. After DSS induction, both mucus secretion and goblet cells were significantly reduced. The intervention of 5-ASA and TCEA alleviated the colon injury caused by DSS, and the number of goblet cells increased significantly with partial recovery of mucus secretion (Figures 4C, D).

Colonic tissue is associated with oxidative stress indicators of results, as shown in Figure 5. The contents of CAT, GSH, and SOD in the DSS group were significantly decreased, while the level of MDA was specifically reflected in the sharp increase in the DSS group. These changes were reversed after 5-ASA or TCEA treatment. Our results confirmed that TCEA increases the levels of CAT and GSH (Figures 5A, B) and inhibits MDA production in the colonic tissue (Figure 5C). Besides, although TCEA intervention reduced SOD levels to some extent, no significant difference was detected (Figure 5D). Overall, TCEA administration could relieve on oxidative stress, in which TCEA-H group was more effective.

TLR4/NF-κB signaling pathway is one of the vital signaling pathways mediating inflammation and plays a key role in the occurrence and development of UC (Wu et al., 2020). As shown in Figure 6, we determined the mRNA expression levels of TLR4, MyD88, and NF-κB p65 by qRT-PCR to detect whether the mechanism of action of TCEA’s anti-inflammatory effect involved in TLR4/NF-κB signaling pathway. Our results showed apparent increases in the mRNA expressions of TLR4, MyD88, and NF-κB p65 after DSS induction. Nevertheless, after different doses of TCEA intervention, the above genes’ expressions were reversed to the control level, especially the TCEA-H group, which showed the best reverse effect (Figures 6A–C). We also validate the expression of TLR4, MyD88, and NF-κB p65 at the protein level. Western blot showed that the expression of TLR4, MyD88, and NF-κB p65 in colon tissue of the DSS group increased sharply, and their expressions were gradually inhibited in 5-ASA, TCEA-L, and TCEA-H groups compared with those of DSS group (Figures 6D–G).

It was found that intestinal flora plays a non-negligible role in the occurrence of UC, so we exploited the 16S sequencing technique to test whether TCEA alters the intestinal flora. The Venn diagram demonstrated the changes in the number of intestinal microbes among three groups; the OTU numbers of intestinal flora in control, DSS, and TCEA-treated groups were 1190, 1263, and 1171, respectively (Figure 7A), which indicated that compared with the control group, DSS treatment increased the total number of intestinal microbes in the DSS group, while TCEA treatment inhibited the trend. In addition, the principal component analysis (PCA) results on the OTU level suggested a clear divergence in the microbial community composition between the control, DSS, and TCEA treatment groups. Still, the composition was more similar between the control and TCEA treatment groups (Figure 7B). α-diversity can reflect the richness and diversity of the sample microbial communities, where Chao1 and ACE indices reflect the richness and Shannon and Simpson’s index reflects diversity. These four indexes in the DSS group all showed an uptrend. Still, the Chao1 and ACE indexes were decreased after TCEA treatment, indicating that the intestinal flora richness of mice was significantly reduced (Figures 7C, D). However, Shannon and Simpson’s index had no significant difference after TCEA treatment showing the diversity changed weakly (Figures 7E, F).

To clarify which bacteria make the discrepancy between the different groups, the composition of the microbial community was contrasted at the phylum and genus levels. At the phylum level, the composition of the flora was dominated by the Bacteroidota and Firmicutes, accounting for more than 75% of the species’ abundance. Compared with the control group, the DSS-induced group showed a decrease in the relative abundance of the Bacteroidota and an increase in the relative abundance of the Firmicutes, but TCEA treatment suppressed these changes. Meanwhile, the relative abundance of Proteobacteria in the DSS and control group assumed bulky diversity (Figure 7G). Figure 7H demonstrates the differences in microbial communities at the genus level between the control, DSS, and TCEA -treated groups; we can see that Bacteroides, Alistipes, and Anaeroplasma grown evidently in the DSS group, while Helicobacter and Alloprevotella presented a downward trend. TCEA treatment reversed this phenomenon by increasing the relative abundances of Helicobacter and Alloprevotella and reducing the relative lots of Bacteroides, Alistipes, and Anaeroplasma.

LEfSe analysis was used further to explore the bacteria with significant differences among the groups. A total of 35 bacteria showed obvious changes among the control, DSS, and TCEA treatment groups, and the evolutionary branching diagram shows the major microorganisms in each group, with the diameter size of the circles on each stratum proportional to the relative abundance size (Figures 8A, B). Functional prediction based on PICRUSt2 function for each sample population revealed that the control group was more positively correlated with genetic information processing, environmental information processing, and human disease but negatively correlated with metabolic and organismal systems. However, the DSS group showed the opposite correlation compared to the control group, while the TCEA treatment group showed a similar correlation to the control group (Figure 8C).

To discuss whether changes in the gut microbiota in mice are associated with UC-related biomarkers, Pearson correlation analysis was used to analyze microbiota at species level with DAI, colon length, histological score, GSH, and inflammatory factor levels (Figure 9). These results indicated that Bacteroides_stercorirosoris, Bacteroides_acidifaciens, Clostridiales_bacterium_CIEAF_020 were negatively correlated with GSH and colon length. However, these three bacteria showed a significant positive correlation with other parameters. Moreover, Desulfovibrio_fairfieldensis, Bacteroides_sartorii, and Bacteroides_thetaiotaomicron exhibited an obvious positive correlation with IL-1β, while Clostridium_leptum and Christensenella_minuta exhibited an obvious positive correlation with IL-6. The above evidence suggests the nonnegligible role of intestinal flora in the remission of UC.