DSS (MW. 36000–50000) was bought from MP Biomedicals (USA). GA (Cat. No. S30153 98.5% for in vivo, B20851 ≥ 98% for in vitro) and 5-aminosalicylic acid (5-ASA, Cat. No. S30083) were purchased from Yuanye (Shanghai, China). Lipopolysaccharides (LPS, Cat. No. L2880) were bought from Sigma-Aldrich (Germany).

Male BALB/c mice (SPF class, 6–8 weeks old, weighing 18–22 g) were provided by the Experimental Animal Center, Shanghai University of TCM. All the mice were housed in a temperature- and humidity-controlled environment (six mice per cage, 20°C ± 2°C, and 40%–60% humidity) with a 12-h light/dark cycle to acclimatize for one week with free access to food and water. All animal experiment protocols were approved by the Animal Research Welfare Council of Shanghai University of Traditional Chinese Medicine (Ethics No. PZ200120).

The mice were divided into Control group, Model group, DSS+5-ASA group, DSS+40 mg/kg GA group, DSS+80 mg/kg GA group, and DSS+120 mg/kg GA group. To generate colitis, 3.5% DSS was orally administered to the mice through drinking water for 7 days (except Control group). GA at 40 mg/kg, 80 mg/kg, 120 mg/kg, and 100 mg/kg and 5-ASA (dissolved in 0.5% carboxy-methylcellulose sodium) were orally administered to the mice from day 7 to day 12. 5-ASA acted as the positive control. On day 13, via ether anesthesia, all mice were euthanized to collect blood samples and colon tissues. The blood samples of the mice were collected by removing the eyeball (Li et al., 2020).

The tissues were fixed in 4% paraformaldehyde for 24 h, cut into 5-µm sections after dehydration, and embedded in paraffin. HE reagents were used to stain the colonic sections. Histological scores were evaluated by assessing the indices for changes in the epithelium and for cell infiltration (Dohi et al., 2005; Yu et al., 2021). The following were observed: 1) no notable signs of inflammation; 2) very low levels of leukocyte infiltration; 3) low levels of leukocyte infiltration; 4) high levels of leukocyte infiltration, moderate vascular density, and thickening of the colon wall; and 5) transmural infiltration, loss of goblet cells, severe vascular density, and thickening of the colon wall. Tissue sections were visualized and photographed under a microscope using a camera system (Nikon DS-U3, Japan).

Colonic tissue samples and sera were collected from BALB/c mice. The expressions of TNF-α, IL-33, interleukin-18 (IL-18), IL-1β, and IFN-γ in serum and in the lysis buffer of the colon tissue were determined by ELISA (Merck, Germany) according to the manufacturer’s instructions.

The protocols and the reagents of RT-qPCR have been described in detail previously (Yu et al., 2021). Total RNA was obtained from colon tissues using a TRIzol reagent (Biotime, China) following the manufacturer’s instructions. After RNA concentration was determined, complementary DNA (cDNA) was generated using an RT-PCR reagent (Thermo Fisher Scientific Inc., Vilnius, Lithuania). Next, RT-PCR was conducted using the FastStart Universal SYBR Green Master (Rox) (Roche, Basel, Switzerland) through a LightCycler 96 q-PCR system (Roche, Basel, Switzerland). The glyceraldehyde 3-phosphate dehydrogenase (GAPDH) acted as a control for the total mRNA amount. The results were detected using the 2^−△△CT method. The PCR primer sequences in this work are shown in Table 1.

Immunohistochemistry (IHC) was performed following previously described methods (Yu et al., 2021). The colonic tissue was fixed in 4% neutral formalin, dehydrated with increasing concentrations of ethanol and dimethyl benzene, embedded in paraffin, and cut into 5-μm sections. The paraffin section was mounted on slides, cleaned, and hydrated. The section was treated with a buffered blocking solution (3% bovine serum albumin in PBS) for 25 min. Then, the sections were incubated with a primary antibody against NLRP3 (1:100) at 4°C for 24 h and incubated with a secondary antibody at room temperature for 50 min. The reaction was blocked by washing with PBS for 5 min. Images were acquired through a microscope using a camera system (Nikon DS-U3, Japan).

RAW264.7 cell lines were bought from the Chinese Academy of Sciences (Shanghai, China). Dulbecco’s modified Eagle medium (DMEM, Sigma, United States) supplemented with penicillin (100 U/mL), streptomycin (100 μg/mL), and 10% fetal bovine serum (FBS, Sigma, United States) was used for culture at 37°C with 5% CO₂ in an incubator. Cell lines were collected until the logarithmic growth phase for the experiment.

For the toxicology experiment, RAW264.7 cell lines were seeded onto a 96-well plate at 5 × 10³ cells per well and cultured overnight. Then RAW264.7 cells were incubated with serial concentrations of GA dissolved in 0.1% dimethylsulfoxide (DMSO) for 24 h. The concentrations of GA were 12.5 μM, 25 μM, 50 μM, 100 μM, and 200 μM. Next, 10 μL of Cell Counting Kit-8 (CCK-8, Dojindo, Japan) solution was added to each well. The culture plate was placed in the incubator for 2 h, and the absorbance of the viable cells was detected (450 nm) using a Multiskan FC microplate reader (Thermo Scientific, United States).

RAW264.7 cells were inoculated into 6-well plates for 24 h and then treated with 50 μM GA, 100 μM GA, and 200 μM GA and 1 μg/mL LPS for 24 h. The amounts of NO release were determined by a nitric oxide assay kit (Beyotime, China). In brief, 50 μL standard and cell supernatants were added to the 96-well plate and mixed with 50 μL Griess Reagent I and 50 μL Griess Reagent II at room temperature, respectively. The OD value was examined by a microplate reader at 562 nm.

Total protein was extracted by cell samples and colonic tissue of different treatment groups in the lysis buffer. The cells were collected and incubated in an RIPA buffer (with the addition of phosphorylation inhibitor and protease inhibitor) for 20 min in an ice bath. After centrifugation at 15,000 rpm for 15 min at 4°C, the upper-clarified lysate was collected. For the colon tissue, 700 μL RIPA lysate was added to a 100-mg sample to extract the protein. Protein concentration was determined using a BCA kit (Beyotime Biotechnology).

For western blotting, the extracted protein of each group was separated by 8%–12% SDS-PAGE and transferred onto a polyvinylidene fluoride membrane. The membrane was blocked with a 5% BSA solution. Then, the membrane was cut into different blots and probed with specific primary antibodies (Table 2) overnight at 4°C (12 h). Next, the membrane was washed with 1 × PBST and incubated with secondary antibodies (Beyotime Biotechnology) at room temperature for 1 h. Then, it was incubated with an enhanced chemiluminescent substrate (Beyotime Biotechnology). In this experiment, β-actin, α-tubulin, HSP90, and GAPDH were used as an internal control to confirm that the amounts of loaded protein were equal. All experiments were repeated at least three times. Western blotting plots were quantified using ImageJ software (United States).

All experiments were performed at least three times, and the results were expressed as mean ± SD. The gray value of the protein was quantified by ImageJ software. Data were analyzed using the software SPSS version 25.0 and illustrated using Graphpad Prism 8.0. The differences between the two groups were analyzed with t-test. ANOVA was performed for multiple comparisons. p < 0.05 was considered statistically significant.

The experiments were designed and carried out as described in the section entitled “Methods and materials” (Figure 1A). Compared with the Control group, a severe weight loss of mice was observed in the DSS group (p < 0.001). GA treatment at 120 mg/kg attenuated DSS-induced weight loss (p < 0.05) (Figure 1B). Similarly, DSS typically caused colonic shortening (p < 0.001), while 120 mg/kg GA ameliorated this change (p < 0.05) (Figure 1C). HE staining of colon sections in the DSS group showed an enormous inflammatory cell infiltration with epithelial cell damage, edematous, hemorrhagic, and ulcerated groups compared with the Control group. 5-ASA significantly alleviated these symptoms induced by DSS (p < 0.01), while GA (40 mg/kg, 80 mg/kg, and 120 mg/kg) restored colon damage, improved crypt architecture, relived congestion and edema (Figure 1D), and reduced the histological scores remarkably (p < 0.01) (Figure 1E).

The secretions of TNF-α, IL-33, IL-18, IL-1β, and IFN-γ in both the serum and colon tissues of mice in the DSS group were significantly upregulated compared with the Control group (Figure 2), indicating DSS-induced inflammation. Compared with the DSS group, 5-ASA and GA markedly inhibited the secretions of TNF-α, IL-33, IL-18, IL-1β, and IFN-γ in both the serum and colon tissues.

DSS-induced murine colitis showed many pathological features on UC in humans, especially type I inflammatory response. Compared with the Control group, the mRNA expression of NLRP3, ASC, Caspase-1, Caspase-4, IL-1β, and IL-18 significantly increased in the DSS group (Figure 3). 5-ASA and GA dramatically lowered the upregulated mRNA expressions of these inflammasomes induced by DSS (Figure 3), indicating 5-ASA and GA were effective.

We have demonstrated that GA reduced the levels of TNF-α, IL-33, IL-18, IL-1β, and IFN-γ in the serum and colon tissues, which provides the possibility of suppressing the activation of NLRP3 inflammasomes. We also detected a relative downstream protein expression of NLRP3. IHC results illustrated that the expressions of NLRP3 were hardly found in the Control group but were significantly elevated in the DSS group, and after treatment of GA, the expression of NLRP3 was reduced (Figure 4A). The results of western blotting revealed that GA treatment notably repressed the over-expression of NLRP3 and Caspase-1 induced by DSS, leading to IL-1β secretion reduction (Figure 4B). Those results indicated that GA might have an anti-inflammatory ability via inhibiting the activation of the NLRP3 inflammasome.

The CCK-8 assay indicated that the concentrations of GA applied in the subsequent experiments exhibited no evident influence on RAW264.7 cell viability (Figure 5A). NO release was remarkably increased in the presence of LPS compared with untreated cells (p < 0.001) and was suppressed after GA treatment (Figure 5B). Similarly, the protein expressions of NLRP3, iNOS, and COX2 were dramatically upregulated with LPS stimulation (Figure 5C). GA treatment significantly lowered the LPS-stimulated elevation of NLRP3 at 200 μM (p < 0.05) and dose-dependently downregulated LPS-induced iNOS and COX2 expressions (p < 0.01). Further investigation demonstrated that GA significantly reduced the phosphorylation of p65-NF-κB (p < 0.05) and ERK (p < 0.01) in the RAW264.7 cell primed by LPS (Figure 5D).