Animals used in this study included C57BL/6 WT adult male mice weighing 25–30 g, and C57BL/6 TLR4 gene-deleted type TLR4${}^{-/-}$ mice [22] weighing 24–33 g (Model Animal Research Center of Nanjing University, Nanjing, China). The TLR4${}^{-/-}$ mice were fertile with no developmental abnormalities. All the mice were raised in specific pathogen-free (SPF) conditions in the Laboratory Animal Center of the Naval Medical University (Shanghai, China). Approval for the research was granted by the Ethics Committee of the Naval Medical University per the Declaration of Helsinki.

The DSS-induced colitis model in all WT and TLR4${}^{-/-}$ mice was constructed by dividing them into WT and TKR4${}^{-/-}$ groups with 12 mice in each group to observe their survival time. For a period of seven days, the mice were administered 1.5% DSS [23] (40,000 kDa; ICN Biochemicals, Anaheim, California, USA) in their drinking water before switching to ordinary water. After the seven-day DSS treatment, the survival status of the mice in the two groups was observed daily. Six mice in each group received 1.5% DSS orally for seven days and were used to observe the general biological indicators including daily weight change and food and water intake (recorded on day five) of the two groups. Six mice in each group receiving 1.5% DSS orally for five days were used for colon bleeding observation, peripheral blood analysis, and histopathological examination. Six mice in each group receiving 1.5% DSS orally for seven days were used to examine the fecal samples by means of 16S rRNA analysis.

On day five after DSS treatment, about 2 mm of the mouse tail was cut off, from which about 40 µL blood was sampled, and then transferred to a K2-EDTA containing Eppendorf (EP) tube, followed by repeated inversions of the tube. Blood cells were counted utilizing a small animal blood cell counter.

The mice were euthanized via cervical dislocation on the third or fifth day following DSS treatment. and the intestinal tissues were isolated for colon bleeding observation. The dehydration of the gut tissues was conducted in an ascending series of ethanol, cleared, paraffin-embedded, and serially sliced into 2–7 micron-thick sections utilizing a rotatory microtome (Aihua, Tianjin, China). Hematoxylin-eosin (H & E) staining was commonly used to stain the deparaffinized sections [24].

Six WT mice and six TLR4-KO mice were classified into two groups and orally administered with 1.5% DSS. Following DSS treatment, the mice in both groups were separately transferred to 12 clean cages lined with sterile filter paper on the seventh day. Immediately following defecation, fecal samples were placed in sterile centrifuge tubes and kept at –80 °C for later examination and transplantation [25].

Extraction of total genomic DNA through the DNA Extraction Kit (Tiangen Biotechnology Company, Beijing, China) was conducted following the instructions of the kit. DNA was quantified and its quality was examined through NanoDrop and agarose gel, respectively. Dilution of the extracted DNA to 1 ng/µL concentration was done and was subsequently kept at –20 °C until further use. Barcoded primers and Takara Ex Taq (Takara, Tokyo, Japan) were used in PCR to amplify the bacterial 16S rRNA genes using the diluted DNA as a template. Concerning the amplification of the 16S rRNA genes’ variable regions V3–V4, the universal primers 343 F and 798 R were employed. The primers were connected with an Illumina sequencing adapter with the reverse primer containing a sample barcode. PCR products were purified and the concentrations were adjusted for sequencing on an Illumina Miseq PE300 system (Oebiotech, Shanghai, China). Utilizing their distinctive barcodes, the samples’ raw sequencing reads were sorted. After removing the barcode and linker, they were subjected to primer sequencing and clustering to generate operational taxonomic units (OTUs) using the software V search with a 97% similarity threshold [26]. All sequences were classified through the NCBI BLAST and SILVA databases [27]. R software (R-4.2.2, Auckland, New Zealand) was employed to carry out the 16S rRNA analysis of the fecal samples. The raw data of 16S rRNA analysis obtained in this study are available in the Gene Expression Omnibus (GEO) database with accession number GSE227565.

Statistical treatment.Comparisons between TKR4${}^{-/-}$ and relevant control WT mice groups with DSS treatment were conducted by means of the Student’s t-test analysis of variation. Survival differences between TKR4${}^{-/-}$ and WT mice groups with DSS treatment were evaluated utilizing Kaplan–Meier plus Cox Regression Analysis. Statistical significance was associated with a p value 0.05. The QIIME package was utilized for the selection of the representative reading of each OTU. Estimator calculation ($\alpha$-diversity and $\beta$-diversity) was conducted by the MOTHUR program [28]. Adonis and Anosim were used to assess the presence of any remarkable variation in the WT and TLR4 KO groups. ANOVA and LEfSe methods were utilized to examine the species with considerable biological duplication variation in the statistical group. ANOVA was employed to examine OTU and species with remarkable variations at the phyla, class, order, family, genus, and species. LEfSe analysis reveals the composition of different species in two or more groups of biological communities. Whereas the species were assessed through LEfSe analysis for remarkable variation at the phyla, class, order, family, genus levels.

First, the function of TLR4 in DSS-induced intestinal injury was explored. It was found that the mortality in TLR4-KO mice was notably elevated in contrast to WT mice (p = 0.0137) (Fig. 1A). There were remarkable variations in the dynamic change of body weight between the two groups on day seven. TLR4-KO mice showed considerably more body weight loss than WT controls subsequent to orally administered 1.5% DSS (Fig. 1B). Simultaneously, the protective function of TLR4 against DSS-induced acute intestinal damage was analyzed using some basic daily behavioral parameters. The water and food consumption of WT and TLR4-KO mice treated with 1.5% DSS-supplemented water was recorded for seven consecutive days per mouse. The result showed that no significant difference in the water-drinking (p = 0.8899) and food-ingestion (p = 0.8207) abilities between TLR4-KO and WT mice at day 1 (Fig. 1C,D). However, the food-ingestion (p = 0.0011) and water-drinking (p = 0.0027) abilities of TLR4-KO mice began decreasing significantly from day five in comparison with those of the WT mice (Fig. 1E,F).

Fig. 1.

Severe fatality and morbidity in TLR4-KO mice subsequent to orally administered 1.5% DSS. (A) WT and TLR4-KO mice were orally administered with 1.5% DSS for seven days. The survival duration was documented (p = 0.0137). (B) Body weight change (%) = (weight at day X / weight at day 0) $\times$ 100%. (C) Water consumption (mL/mouse) and (D) food weight (g/mouse) of WT and TLR4-KO mice were recorded on day one (p = 0.8899, p = 0.8207). (E) Water consumption (mL/mouse) and (F) food weight (g/mouse) of WT and TLR4-KO mice were recorded on day five (p = 0.0011, p = 0.0027). *p 0.05, denotes remarkable variation from the WT control group; NS, No significant difference detected. Kaplan–Meier plus Cox Regression Analysis was used in (A). Student’s t-test analysis was used in (B–F). Abbreviations: DSS, Dextran Sulfate Sodium Salt; WT, wild-type; TLR4-KO, TLR4 knockout.

Morphological changes of the mice in the two groups were determined after oral administration of 1.5% DSS. Intestinal bleeding was detected in TLR4-KO and WT mice on days three (p = 0.0002) and five (p = 0.00015) respectively, but the severity of symptoms was increased in the TLR4-KO group in comparison to the WT group (Figs. 2A,3A). Consistent with colon bleeding, TLR4-KO mice showed more severe anemia by quantifying the peripheral red blood cell count on day five (RBC Count: p = 0.0093; Hematocrit: p = 0.0054) in both groups (Fig. 2B,C). Simultaneously, intestinal pathologic evaluation was performed on day five after oral administration of 1.5% DSS. It was found that intestinal inflammation and tissue necrosis in TLR4-KO mice were more severe than those in WT mice (Fig. 3B). All the above results demonstrated that TLR4-KO elevated the severity of intestinal epithelial injury in DSS-induced enteritis in mice.

Fig. 2.

Elevated susceptibility of TLR4-KO mice to contract anemia. (A) Stool bleeding Score on days three and five. The following is the stool bleeding score: 0 = absence of any gross blood visible in the entirety of the colon; 1 = presence of gross blood in 1/3 of the colon; 2 = 2/3; 3 =$>$ 2/3 of the colon. (B) RBC count in peripheral blood of the mice on day five subsequent to orally administered 1.5% DSS (p = 0.0093). (C) The hematocrit concentration in the peripheral blood of the mice on day five of orally administered 1.5% DSS (p = 0.0054). p 0.05, denoting remarkable variation from the WT control group. Student’s t-test analysis was utilized in this section. Abbreviation: RBC, red blood cell.

Fig. 3.

Colon histopathology of the two mouse groups. (A) Photographs of bleeding colon from WT and TLR4-KO mice on day seven of orally administered 1.5% DSS. (B) H&E staining photographs of the colon in the two mice groups seven days following 1.5% DSS-infused drinking water treatment.

Differences in the composition of the microbiota of the gut in fecal samples were compared through sequencing 16S rRNA gene between TLR4-KO and WT mice at day seven after orally administered 1.5% DSS using Operational Taxonomic Unit (OTU) [29] partitioning with 97% sequence similarity in gut microbiota. From the flower plot, shared intestinal microbiota OTUs of all fecal samples were found to be 42, the number on the petal indicated the total OTUs of each sample minus the number of common OTUs (Supplementary Fig. 1). The abundance of each OTU was analyzed and the first 50 with the most abundant OTU were selected, an evolutionary tree was established, and the abundance of OTU in different samples was displayed with a heatmap graph (Supplementary Fig. 2). $\alpha$-diversity [30] is defined as the microbiota diversity within a community. The diversity of microbiota has two aspects: one is the type of bacteria, and the other is the bacteria abundance uniformity. No significant difference in $\alpha$-diversity value in the intestinal microbiota was observed between TLR4-KO and WT mice (Supplementary Fig. 3A–C). $\beta$-diversity [31] refers to the diversity of microbiota between different communities. $\beta$-diversity includes principal component analysis (PCA), principal coordinates analysis (PCoA), and Unweighted Pair Group Method with Arithmetic mean (UPGMA) analysis. PCA and PCoA present visual coordinates for the similarity or variation of data. These results demonstrated that the dots of the six WT and six TLR4 KO samples were close to the respective dots in the same group, while the dots in the TLR4-KO group were further away from those in the WT group (Fig. 4A,B). UPGMA revealed similarities between the samples or groups through the distance matrix algorithm. The circular tree was established by UPGMA analysis, showing that the gut microbiota composition varied considerably in the two groups (Fig. 4C). Hence, the intestinal microbiota composition in $\beta$-diversity varied considerably between TLR4-KO and WT mice after 1.5% DSS treatment.

Fig. 4.

$\beta$-diversity analysis on intestinal microbiota of WT and TLR4-KO mice. (A) Principal component analysis (PCA) analysis of gut bacteria data. (B) Gut bacteria principal coordinates analysis (PCoA) analysis. (C) The circular tree by Unweighted Pair Group Method with Arithmetic mean (UPGMA) analysis.

According to the microbiota taxonomic levels of phylum, class, order, family, and genus, the relative abundance of each sample at various levels was recorded. The average number of reads per sample was listed in Supplementary Table 1. After a seven-day treatment with 1.5% DSS, the dominant phylum, class, order, family, and genus in WT mice were Bacteroidetes, however, they were considerably reduced in TLR4-KO mice. Simultaneously, the phylum Proteobacteria, the class Gammaproteobacteria (Fig. 5A,B), the order Enterobacteriales, the family Enterobacteriaceae, and the genus Escherichia/Shigella and Rikenellaceae were increased in TLR4-KO mice in contrast to WT controls (Supplementary Fig. 4A–C). Furthermore. ANOVA statistical analysis was carried out on OTUs and the five aforementioned phylogenetic levels, and the first ten different OUTs (species abundances) were selected for relative abundance boxplot analysis for the sake of quickly obtaining intra- and inter-group differences between the dominant and differential species. The results of ANOVA difference statistics analysis showed at the family level abundant Enterobacteriaceae, Moraxellaceae, Deferribacteraceae, Desulfovibrionaceae, Rikenellaceae, Xanthomonadaceae, and Pseudomonadaceae in gut microbiota in TLR4-KO mice, and abundant Bacteroidales, Pepococcaceae, Alcaligenaceae and Porphyromonadaceae in gut microbiota in WT mice following DSS treatment (Fig. 6A,B). The linear discriminant analysis (LDA) coupled with effect size measurements (LEfSe) analysis indicated that the gut microbiota genera were differentially numerous between WT and TLR4-KO mice after orally administered 1.5% DSS. Gladography revealed a considerably increased abundance of Bacteroidetes at the genus level in WT mice, whereas an increased abundance of Deferribacteres, Tenerlcutes, and Proteobacteria was determined in TLR4-KO mice (Fig. 7).

Fig. 5.

Gut microbiota variation in TLR4-KO and WT mice after DSS treatment at levels of phylum and class. (A) The relative abundance of the main phylum. (B) The relative abundance of the main class.

Fig. 6.

Statistical analysis of microbial multivariate variables between TLR4-KO and WT mice following orally administered 1.5% DSS. (A,B) ANOVA statistical analysis of differences between TLR4-KO and WT mice following oral administration of 1.5% DSS at the family level.

Fig. 7.

Differential enrichment of multiple taxa in the TLR4-KO and WT control mice feces as per coupled with effect size measurements (LEfSe) analysis.