Animals

Male Wild-type C57BL/6 mice (6–8 weeks, 23 ± 2 g) and Male tuft cell deficient (Pou2f3−/−) C57BL/6 mice (6–8 weeks old, 23 ± 2 g) came from the Laboratory Animal Center of Nanjing Medical University (Jiangsu Province, China). Mice were housed under specific pathogen-free (SPF) conditions (21 ± 2 ℃, 12 h: 12 h dark–light cycle. Wild type mice were divided into five groups: Control (Con), DSS, DSS + 200 mg/kg berberine, DSS + berberine + 20 mg/kg U73122 and berberine (n = 8 in each group). Tuft cell deficient mice were divided into three groups: Pou2f3−/−, Pou2f3−/− + DSS, Pou2f3−/− + DSS + berberine (n = 8 in each group).

Berberine (C20H18NO4+, purity: 99.70%, Selleck Chemicals, 2086–83-1, Houston, TX, USA) was administered orally one day in advance and once every two days. Mice were effectively given 3% DSS (Material Pioneer Inc, 160,110, Oklahoma, USA) orally by drinking water for seven consecutive days to create an established mouse model of acute colitis, which displayed symptoms such diarrhea, prostration, and bloody stools. The experimental intervention of U73122 (MedChemExpress, New Jersey, USA) was performed by gavage. Berberine and U73122 were dissolved in carboxymethyl cellulose sodium (CMC-Na, 0.5%, Macklin, 9004–32-4). The control mice were orally administered with solvent as a control.

Disease activity index score

Weight loss, stool consistency, and bleeding were assessed according to Millar et al. Methods [50, 51].

Histopathological examination

Colons were collected and fixed in 4% PFA for 48 h prior to paraffin embedding. Haematoxylin and eosin (H&E) were used to stain three-μm-thick sections following standard protocols and the degree of inflammation was assessed through a previously described histological colitis score. The scores ranged from 0 to 6 (total score) representing the degree of tissue damage and inflammatory infiltration. 0 = no inflammation or mucosal injury, 1 = slight accumulation of inflammatory cells or discrete mucosal injury, 2 = inflammatory infiltration of submucosa or surface mucosal erosions and focal ulcers, 3 = aggravated infiltration or extensive mucosal damage.

Immunofluorescence and immunohistochemistry

Colons were cut into 3 µm slices for immunohistochemistry. The sections were blocked with donkey serum containing 0.3% triton X-100 for 2 h at 21–25 ℃, after which antigens were unmasked with a 0.01 mol L−1 citrate buffer solution at 91–95 ℃ for 3 min. Slices were then incubated with anti-Dclk1 antibody (Abcam, Cambridge, ab37994, MA, USA), anti-MUC2 antibody (Proteintech, 27,675–1-AP), anti-ChgA antibody (Abcam, Cambridge, ab283265, MA, USA), anti-α-gustducin antibody (Santa cruz, sc-518163), anti-Dclk1 antibody (Proteintech, 68,234–1-Ig) and anti-Trpm5 antibody (Proteintech, 18,027–1-AP), overnight at 4 ℃. After incubation, slices were washed thrice (5 min per wash) with phosphate-buffered saline (PBS) and incubated with fluorescein isothiocyanate (FITC)-labelled anti-rabbit IgG antibody (Thermo Fisher Scientific, A-21206, PA, USA) for 2 h at 22–25 ℃. Finally, the sections were sealed with an anti-fluorescence quenching tablet containing DAPI (4′,6-diamidino-2-phenylindole, Solarbio, Beijing, China) for nuclei staining and photographed under a fluorescence microscope (400 × magnification, ZEISS, Jena, Germany). The frequency of DCLK1+ tuft cells and ChgA+ enteroendocrine cell were determined by employing Image J software. The positive staining area of MUC2+ goblet cells was counted using Image J software. The percentage of α-gustducin+Dclk1+ cells and Trpm5+Dclk1+ cells in Dclk1+ positive cells in the colon was enumerated and subjected to analysis.

Colon sections were de-paraffinized in xylene, rehydrated using a graduated alcohol-to-water ratio and blocked as previously mentioned. The slides were then incubated with the primary antibodies anti-Occludin (abcam, ab216327, MA, USA) and Claudin-1 (Santa Cruz, sc81796, MA, USA), overnight at 4 °C. After primary antibody incubation, slides were washed thrice with PBS-T and then incubated with streptavidin–horseradish peroxidase (Absin Bioscience Inc, Shanghai, China) for 40 min at 22–25 ℃, followed by an incubation with the substrate 3,3′-diaminobenzidine (Shanghai Gene Company, Shanghai, China) for 5–15 min. Haematoxylin was used as a counterstain. Image acquisition was performed using a Zeiss microscope (ZEISS, Jena, Germany).

Culture of colonic organoids and immunofluorescence staining

The cultivation and handling of colonic organoids were performed as described [52]. Colon tissues were taken from mice with 4 weeks age, cut longitudinally, remove feces, clean with cold PBS solution at least 15 times. Incubate the tissue at 4 °C on a rotator at 20 rpm for 30 min in 25 mL 2.5 mmol/L cold EDTA solution. Then, resuspend the tissue in 10 mL cold PBS and filter the upper liquid through a 70 μm filter membrane into a 15 mL centrifuge tube. Repeat the previous step six times, labeling the tubes 1–6. Take 500 μL of each tube and transfer it to 12-well plates under a microscope to observe the quality and number of crypts. Count and record the best result, which is 500 crypts per well. Resuspend the Matrigel (Corning, NY, USA) mixed culture medium and form a hemispherical shape on the plate. Add 500 μL of culture medium (Stem Cell Technologies, Vancouver, BC, Canada) to each well after the gel solidifies. Crypts usually form within 5–7 days. On day 6, the organoids were treated with berberine at concentrations of 10 μM for 48 h. For immunofluorescence staining, the Matrigel-containing organoids were placed in a refrigerator for 15 min to dissolve the Matrigel. Then they were centrifuged at 3000 g for 10 min and fixed in 1 mL PFA for 45 min. Staining was performed using anti-DCLK1/MUC2/ChgA antibody, DAPI stain (Vazyme, Nanjing, China), following the aforementioned methods.

ELISA

The levels of IL-25 in the mouse colonic organoid culture supernatants were determined using a mouse IL-25 ELISA kit (MULTI SCIENCES, EK2179, Hangzhou, China).

RNA extraction and quantitative real-time polymerase chain reaction(qRT-PCR)

Colon tissues were homogenised in TRIzol before total RNA extraction using a RNA extraction kit (Thermo Fisher Scientific, Waltham, MA, USA), according to manufacturer’s instructions. The RNA purity and concentration was determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific) and RNA was converted in cDNA by reverse transcription. Thereafter, quantitative real-time PCR was carried out using a iQTM SYBR® Green Supermix (Vazyme Biotech, Nanjing, China) on a BioRad CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA). Relative gene expression was calculated using the comparative cycle threshold (CT) method. β-actin was used as a housekeeping gene to normalise gene expression. The primer sequences used in this study are listed in Additional File 2: Table 1.

Extraction of lymphocytes from the colonic lamina propria and spleen

The large intestine was prepared for analysis by removal of fat and lymph nodes from its surface, followed by a longitudinal incision to extract its content. The tissue was then cut to 1–2 cm pieces, incubated for 10 min at 22–25 ℃ with 1 mmol L−1 DTT (Ditiotreitol, Aladdin, 3483–12-3, USA) at 250 rpm, and then further digested for 10 min at 37 °C at 200 rpm with PBS containing 30 mmol L−1 EDTA (Aladdin, 139–3–33, USA) and 10 mmol L−1 HEPES (Aladdin, H109408, USA). An additional digestion step was conducted using RPMI1640 media (Sigma R8758, USA) containing 150 g/mL of DNase I (Sigma, 9003–98-9, USA) and 150 μg/mL of collagenase VIII (Sigma, 9001–12-1, USA), in an incubator with 5% CO2 at 37 °C for 75 min. The homogenate was collected, shaken vigorously until fragmentation by a vortex, and centrifuged at 1800 rpm for 5 min at 22–25 ℃. The supernatant was removed. The cells were reconstituted in 4 mL of 40% Percoll, and 80% Percoll (2.5 mL) (univ, 17,089,109–1, Shanghai, China) was gradually added to the bottom of the tube while rotating at 2500 g at 22–25 ℃. Twenty minutes of centrifugation was spent using a gradient centrifugation. The cells in the middle layer were collected, counted, and characterized using flow cytometry. Splenic lymphocytes were extracted with reference to previous studies [53].

Flow cytometry analysis

The lymphocytes collected from the large intestine were incubated with PerCP-Cy5.5 anti-mouse CD3 (BD, 551,163, New York, USA), FITC anti-mouse CD4 (Biolegend, 100,406, Beijing, China), PE anti-mouse IFN-γ (eBioscience™, 12–7311-82, USA), APC anti-mouse IL-4 (eBioscience™, 17–7041-82, USA) and BV421 anti-mouse IL-17A (BD, 563,354, New York, USA); APC anti-mouse CD45 (Biolegend, 157,606, Beijing, China), FITC anti-mouse lineage (eBioscience™, 22–7770-72, USA), PE anti-mouse KLRG1 (BD, 561,621, New York, USA), BV421 anti-mouse GATA3 (Biolegend, 653,814, Beijing, China) and acquired in a BD FACSVerseTM cytometer (BD Biosciences, Heidelberg, Germany). Results were analysed using the FlowJo® software (Treestar Inc., San Carlos, CA, USA).

Statistical analysis

All of the experiments were performed at least three times independently. Data were presented as mean ± SEM. Paired two-tailed Student's t-test was utilized for the comparison of two groups; one-way or two-way analysis of variance (ANOVA) followed by Tukey's post-hoc test was used for the comparison of more than two groups. Statistical significance was defined as P < 0.05, and was performed using GraphPad Prism software (version 8.0, San Diego, CA).