Animal and treatment

C57BL/6 mice, aged 6–8 weeks, were purchased from the Chinese University of Hong Kong. The animal welfare and experiments procedures were strictly followed the guidelines approved by the Ethics Review Committee of Macau University of Science and Technology. The mice were maintained in an automated, environmentally controlled setting featuring a 12-hour light-dark cycle and provided unlimited access to MilliQ water and PicoLab® Rodent Diet (sourced from LabDiet, USA).

A total of 32 mice were employed and randomly assigned to four groups: the control (Ctrl) group, the D. vulgaris (DV) group, the Model (DSS) group, and the D. vulgaris + DSS (DSS + DV) group. Prior to modeling, mice in the DV and DSS + DV groups underwent transplantation of 2 × 10⁸ CFU (colony forming units) of D. vulgaris suspended in phosphate-buffered saline (PBS) via gavage for a week (from day 1 to 7). Subsequently, mice in the DV, DSS, DSS + DV groups were administered the drinking water with 1.8% DSS (weight/volume) for ten days (from day 8 to 17) to induce colitis. Notably, the transplantation of D. vulgaris by gavage continued throughout the modeling period. Finally, a one-week recovery period (from day 18 to 24) was established to investigate the impact of persistent D. vulgaris infection (Fig. 1a).

Throughout the entire experiment, parameters including food consumption, body weight, and the disease activity index (DAI, a composite score based on body weight loss, occult blood presence, and stool consistency) were monitored at intervals of two to three days. DAI scores were evaluated following the previous studies [16, 17]. Stool consistency: normal (score = 0), loose (score = 1), diarrhea (score = 2). Blood loss: no loss (score = 0), gross bleeding (score = 1). Appearance: normal (score = 0), hunched (score = 1), starey coat (score = 2), lethargic (score = 3). Body weight (%) loss: none (score = 0), 0–5 (score = 2), 5–10 (score = 2), 10–15 (score = 3), > 15 (score = 4).

In the experiment of evaluating the impact sodium hydrosulfide (NaHS) on GM composition, mice were administered with NaHS (7.5 mg/kg) using intraperitoneal injection for 15 consecutive days. On the 15th day, feces from every single mouse were collected and kept in -80℃ refrigerator for further experiments.

The culture of bacteria

D. vulgaris (DSM 1744) was purchased from China General Microbiological Culture Collection Center. Akkermansia muciniphila (BAA-2869) was purchased from American Type Culture Center. Bacteria was cultured in the anaerobic chamber (Whitley A35 Workstation, UK) with 37℃ and 65% humidity. The value of OD 600 nm from micro-spectrometer was used to determine the density of the bacteria.

Samples collection and DNA extraction

Half of the mice from every group were sacrificed under CO2 anesthetization on the day of 17th and 24th, respectively. And then open the belly to isolate the cecum and colon for further experiment. The isolated colon was dipped into 2 ml cold PBS to collect the colon content. About 2 cm of the colon were separated and kept in the formalin for histology staining. Mucosa from the mice colon and the cecum were collected and kept in the − 80℃ refrigerator for further experiment.

Histology and immunohistochemical staining

Paraffin sections 5 μm thick were used for hematoxylin and eosin (H&E) staining, Alcian blue staining and immunohistochemical (IHC). Briefly, the section slides were immersed in xylene to remove paraffin, followed by a graded series of ethanol washes to rehydrate the tissue. To enhance antigen accessibility, microwave-mediated heat-induced epitope retrieval was employed. Subsequently, the IHC targets were specifically probed using antibodies against E-Cadherin (diluted 1:200, #3195S, Cell Signaling Technology), occludin (diluted 1:200, #404,700, Invitrogen), and ZO-1 (diluted 1:200, #617,300, Invitrogen). For chromogen development, the LSAB-HRP kit (K0679, DAKO) was utilized, facilitating the visualization of antibody-bound antigens. The section slides were counterstained with a nuclear stain by hematoxylin. After dehydrated, the tissue sections were mounted and viewed under a Leica microscope with a Leica camera (DFC310 FX) and Leica Application Suite software (Version 4.4.0, Switzerland).

Measurement of SCFAs

Five microliters of 4-Cl-phenylalanine (0.3 mg/mL) was added to the tested samples (50 µl serum), and then 50 µl of the mixture was added into 200 µl of the chilled MeOH. The mixture was centrifuged at 13,000 rpm for 5 min at 4 °C. The supernatant water layer of the mixture was carefully isolated, with the extraction procedure being performed in duplicate. Finally, the supernatant was dried under the nitrogen stream, derived, and then analyzed by HPLC-Q-TOF/MS as previously described [18].

Total RNA extraction and quantitative reverse transcription polymerase chain reaction (qRT-PCR)

Total RNA was isolated from mucosa using TRNeasy Mini Kit (QIAGEN) following the manufacturer’s instructions. The concentration of RNA was quantified using NanoDrop 2000 C spectrophotometer (Thermo, USA). Then, the first-strand cDNA was synthesized from 2 µg of total RNA using Transcriptor Universal cDNA Master Kit (Roche). qRT-PCR was performed to determine the differentially expressed level of mRNA using Applied Biosystems ViiA™ 7 PCR system (Carlsbad, CA, USA). The specific primer sequences are listed in Table S1. β-actin was used as an internal control. The qRT-PCR was performed using Power SYBR®Green PCR Master Mix (Applied Biosystems Inc., Carlsbad, CA, USA) as described [19].

DNA extraction and 16S rRNA full-length sequencing

Fecal samples from the 17th day were collected and extracted genomic DNA using QIAamp DNA Stool Mini Kit (QIAGEN) following the guidelines by the manufacturer. The purified DNA was kept in the − 80℃ refrigerator, pending 16 S rRNA amplicon sequencing.

The 16S rRNA gene PCR primers with barcode on the forward primer were used in a 35-cycle PCR using the HotStarTaq Plus Master Mix Kit (Qiagen). After amplification, PCR products are checked in 2% agarose gel to determine the success of amplification and the relative intensity of bands. Samples are pooled together in equal proportions based on their molecular weight and DNA concentrations. The PCR pool is then purified using Ampure PB beads (Pacific Biosciences).

The SMRTbell libraries (Pacific Biosciences) are prepared following the manufacturer’s user guide and sequencing performed on the PacBio Sequel following the manufacturer’s guidelines. After completion of initial DNA sequencing, Circular Consensus Sequencing (CCS) was analyzed using PacBio’s CCS algorithm. The CCS algorithm aligns the subreads individually from each template to generate consensus sequences thereby correcting the stochastic errors generated in the initial analysis. In summary, the CCS sequencing data is depleted of barcodes, oriented 5’ to 3’, sequences < 150 bp removed, and sequences with ambiguous base calls removed. Operational taxonomic units (OTUs) were defined by clustering at 3% divergence (97% similarity). Final OTUs were taxonomically classified using BLASTn against a curated database derived from NCBI (www.ncbi.nlm.nih.gov).

The test of H2S from cecum content

In the anaerobic chamber, 100 mg of cecal content was placed in an autoclaved tube with 300 µL of PBS. After mixture, 100 µL solution was transferred into a well of a 96-well plate, one well per sample. The lid of the 96-well plate was embedded with agarose containing 0.1 M lead acetate. The reaction principle involved the formation of PBS, a brown to black precipitate, through the reaction between S2- and Pb2+. This reaction took place within the 96-well plate in the anaerobic chamber for 2 h at 37℃ and 65% humidity. After the completion of the reaction, the agarose-embedded lid was visualized using the Gel Doc XR + system (Bio-Rad) and subsequently digitized for analysis using Image J 1.54d (National Institute of Health, USA) [20].

Statistical analysis

One-way ANOVA (for parametric data) and Kruskal–Wallis tests (for non-normal data) were performed to observe significantly among the groups using GraphPad Prism 8.4.3 (GraphPad Software, USA) [21]. The composition of microbiota was analyzed using R platform with vegan package (2.5.7, DOI: https://doi.org/10.32614/CRAN.package.vegan), pheatmap (1.0.12, DOI: https://doi.org/10.32614/CRAN.package.pheatmap), ggplot2 (3.3.2, DOI: https://doi.org/10.32614/CRAN.package.ggplot2), dplyr (1.0.1, DOI: https://doi.org/10.32614/CRAN.package.dplyr), ggpubr (0.4.0, DOI: https://doi.org/10.32614/CRAN.package.ggpubr), scales (1.1.1, DOI: https://doi.org/10.32614/CRAN.package.scales), grid (4.0.2, DOI: https://doi.org/10.32614/CRAN.package.grid), and tidyverse (1.3.1, DOI: https://doi.org/10.32614/CRAN.package.tidyverse). The R packages of vegan, ggplot2, and scales were used to perform alpha diversity analysis. The R packages of pheatmap, scales, and dplyr were used to perform the heatmap analysis. The value of p < 0.05(*), p < 0.01(**) and p < 0.001 (***) denote for all of the statistical analyses.