Ginsenoside Rg1 (batch number: 16080053, purity: 99.91%) was purchased from Chengdu Purifa Technology Development Co., Ltd. (Chengdu, China). Dextran sodium sulfate (DSS; molecular weight: 36-50 kDa; batch number: 160110) was purchased from MP Biomedicals (California, USA). Y27632 (batch number: 129830-38-2) was purchased from APE&BIO (Texas, USA).
Male specific pathogen-free (SPF) BALB/c mice (8 - 9 weeks, 22 ± 2 g) were purchased from Hunan Slack Jingda Experimental Animal Co., Ltd. (Changsha, China) (Animal Certificate Number: SCXK (Xiang) 2019-0004). Reproduce freely with standard diet and tap water according to the Institutional Animal Care and Use Committee at the animal facility of Traditional Chinese Medicine (Nanchang, China). This protocol (license number: JZ2019-235) was approved by the Institutional Animal Care and Use Committee (IACUC) of Jiangxi University of Traditional Chinese Medicine.
DSS-induced colitis and treatment
All mice were divided randomly into 4 groups (10 mice per group): Control, DSS, DSS+Rg1 and DSS+Rg1 group. Male BALB/c mice of the DSS, DSS+Rg1, DSS+Y27632 groups received 3% (wet/vol) DSS in drinking water for 7 days, then normal water for 7 days, followed by administration of 2% (wet/vol) DSS for 7 days. Mice of the Control group received normal drinking water. At the beginning of the 8th day, mice in the DSS+Rg1 group were orally administrated with 200 mg/kg/day ginsenoside Rg1, mice in the DSS+Y27632 group were orally administrated with 10 mg/kg/day Y27632, and mice in the DSS and Control groups were treated with the equal volume of normal saline. Throughout the study, all mice were weighed once daily (09:00), and monitored daily for diarrhea, hematochezia, hunched posture and hair loss.
Mice were anesthetized deeply with 20% pentobarbital sodium , the abdominal cavity was quickly opened, the colon tissue was directly separated, the length and weight of the colon were measured, and the colon index = colon weight/mouse weight × 100%.
Pathological histology analysis
The proximal colon of mouse was fixed in 4% polyformaldehyde solution for 24 hours. The tissue was dehydrated with gradient ethanol, transparent with xylene, embedded in paraffin, and finally cut into 4 μm-thickness slices. The sections were stained with hematoxylin and eosin (Solarbio, Beijing, China), and and images were collected under the optical microscope (Lecia, Wetzlar, Germany) for pathological analysis. The pathological injure of the colon were blindly assessed by two different pathologists, including inflammatory cell infiltration and tissue damage . The scoring of inflammatory cell infiltration was evaluated as 0 (rare inflammatory cells in the lamina propria) to 3 (transmural extension of the infiltration of inflammatory cells), and tissue damage was evaluated ranging from 0 (no mucosal damage) to 3 (extensive mucosal damage and extension through deeper structures of the bowel wall).
Enzyme-linked immunosorbent assay (Elisa)
The colon tissue (100 mg) was collected and lysised by RIPA (radio immunoprecipitation assay) solution at a ratio of 1:10, incubated at 4 ℃ for 1 hour, homogenized by ultrasonic homogenizer for 20 min, centrifuged at 4 ℃ for 15 min. Total protein in each sample was quantified by a total protein detection kit (Aidlab Biotechnlologies Co., Ltd., Beijing, China), and then diluted in 1× PBS to a final concentration of 3000 ng/mL And then the supernatant was obstained and detected. The concentrations of IL-6, IL-33, CCL-2, TNF-α, IL-4 and IL-10 were measured by commercial Elisa kits (Invitrogen, Calif., USA) according to the manufacturer’s protocol, and then the optical density (OD) values at 450 nm was detected using a microplate reader (Thermo, Varioskan, MA, United States). Then, each cytokine was quantified basally based on a standard curve established using an Elisa kit.
Peripheral blood was collected in anticoagulated tubes, and mononuclear cells were isolated for further analysis. These cell samples were resuspended in RPMI 1640 and incubated in 5% CO2 at 37 ℃ for 3 hours. All samples were detected on a FACS calibur analyser (Becton-Dickinson, Mountain View, CA). Then, these cells were incubated with an Fcγ receptor-blocking mAb (CD16/32; BioLegend, San Diego, CA, USA) for 15 minutes at 4 °C. Subsequently, for surface antigen detection, the cells were shielded from light and labeled with with PE-conjugated anti-CD163 (1:100), PE-Cyanine7-conjugated anti-CD206 (Invitrogen, Calif., 1: 100), Alexa Fluor 488-conjugated anti-iNOS (Invitrogen, Calif., USA, 1: 200), PE/Cyanine7-conjugated anti-CD284 (Biolegend, Calif., USA, 1: 100), Alexa Fluor® 647-conjugated anti-F4/80 (BD Biosciences, Franklin Lakes, USA, 1: 100), PE-conjugated anti TIM-1 (BD Biosciences, Franklin Lakes, USA, 1: 100) and PerCP-Cy™5.5-conjugated anti-CD11b (BD Biosciences, Franklin Lakes, NJ, USA, 1: 100) antibodies. All data were analyzed using FlowJo 7.6.1 software (TreeStar, Ashland, OR, USA), and the inactive cells were excluded by gating.
The normalized supernatants (5 μg/μL) of colonic tissues were prepared as described in 2.6. An equivalent amount of protein in each sample was fractionated onto sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to a polyvinylidene fluoride (PVDF) membrane with a Bio-Rad Western blot apparatus. The PVDF membranes were blocked with 5% fat-free milk then incubated overnight with the following primary antibodies at 4°C. The primary antibodies were Nogo B (Abcam, ab47085, 1:2500), MIF (Abcam, ab187064, 1:2500), Pim-1 (CST 3247s, 1:1000), TLR2 antibody (Abcam, ab16894, 1:500), Rock1 (Abcam, ab45171, 1:1000), RhoA (Abcam, ab187027, 1:7500). These membranes were treated with the corresponding secondary antibody Goat-anti-Rabbit lgG (HRP) (Abcam, ab205718, 1:5000), Goat-anti-mouse lgG (HRP) (Abcam, ab205719, 1:10000) for 1-2 h at room temperature. Subsequently, these membranes were visualized with ECL western blot substrate. The specific protein bands were scanned with a UVP Chen Studio (Analytik Jena, Germany) and quantified using Image-Pro Plus 6.0 software (Media Cybernetic, MD, United States).
Microbial diversity analysis
The stool samples were collected on day 21 and immediately stored at −80℃ for bacterial DNA extraction. The microbial diversity analysis was entrusted to Majorbio Bio-Pharm Technology (Shanghai, China). Total bacterial DNA was extracted from fecal samples according to QIAamp DNA Stool Mini Kit (Qiagen, Valencia, CA, USA) instructions. The V3-V4 region of the bacterial 16S rRNA gene was amplified with primers: 338F (5’-ACTCCTACGGGAGGCAGCAG-3’) and 806R (5’-GGACTACHVGGGTWTCTAAT-3’). PCR products were purified using the AxyPrep DNA gel extraction kit (Axygen, Union City, USA) and quantified by Qubit (Invitrogen, USA). Then, the qualified libraries were sequenced using Illumina miseq platform. Operational taxonomic units (OTUs) were clustered with a 97% similarity cut-off using UPARSE (version 7.1). The taxonomy of each 16S rRNA gene sequence was assigned by the ribosomal database project (RDP) Classifier algorithm (http://rdp.cme.msu.edu/) against the SILVA (SSU123) 16S rRNA database using a confidence threshold of 70% . Alpha diversity analysis (Mothur, version v.1.30.1) was used to evaluate the Chao1 abundance and the Shannon index. Principal coordinates analysis (PCoA) was performed using Mothur, and statistical analysis was performed based on the values of PC1 . Linear discriminant analysis (LDA) coupled with effect size (LEfSe) measurements (based on non-parametric factorial Kruskal–Wallis sum-rank test and the Wilcoxon rank-sum test) was used to identify taxa that were significantly different (biomarkers) between groups, with P < 0.05 and an LDA score threshold of 4 . Microbial difference analysis, correlation analysis, and co-occurrence network analysis were performed using I-sanger (Majorbio Bio-Pharm Technology Co. Ltd.; www.i-sanger.com). 
Data were expressed as the mean ± SEM (Standard error of mean). Statistical analyses were carried out using GraphPad Prism 7.0 software (San Diego, CA, USA). One-way analysis of variance (ANOVA) followed by the Tukey test for multiple comparisons were performed to determine significance. All p-values less than 0.05 were considered to be statistically significant.