A total of 50 male C57BL/6 mice (age, 8 weeks; weight, 20 g; SPF grade) expressing natural killer (NK)1.1 were provided by Shanghai SLAC Laboratory Animal Co., Ltd., and housed in a specific-pathogen-free environment under normal conditions of humidity (50±5%) and temperature (25±2°C) with a 12-h light/dark cycle and access to food and water ad libitum. The mucosal immune response of UC is a Th2 lymphocytic-like response mediated by NKT cells. Therefore, NK1.1-expressing mice were used as the research subjects. Subsequently, on day 1, the mice were anesthetized with 35–40 mg/kg pentobarbital sodium intraperitoneally and shaved on the back between the shoulders, subsequently a 2×2 cm area of skin was exposed, which was treated with 200 µl sensitizing solution (Sigma-Aldrich; Merck KGaA). On the day 2, the animals developed an allergic reaction. On day 5, fasting was initiated. On day 6, following pentobarbital sodium anesthesia, a 3.5 F hose, lubricated with sterile paraffin oil (Sangon Biotech Co., Ltd.) was slowly inserted into the anus of the mouse (distance, 4 cm). A total of 150 µl oxazolone gavage solution (Sigma-Aldrich; Merck KGaA) was slowly injected and the catheter was removed. The head of the animal was kept down in a vertical position for 60 sec in order to leave the oxazolone solution in the intestinal lumen. All protocols were reviewed and approved by the ethics committee of Shanghai University of Traditional Chinese Medicine (approval no. 0015; Shanghai, China).

PD (16) was made up of 30 g radix pulsatillae, 9 g cortex fraxini, 5 g cortex phellodendri and 12 g rhizoma coptidis (all purchased from Longhua Hospital Shanghai University of Traditional Chinese Medicine; Shanghai, China). To obtain the water-soluble extracts, the samples were extracted using ethanol, and then concentrated, dried and powdered (17). The samples extracted by alcohol were autoclaved and stored at 4°C. The mice were randomly divided into the following five groups, with 10 mice in each group: i) PD group; ii) oxazolone-induced colitis group; iii) IL-13 (1 mg/kg body weight) intervention using anti IL-13 (PeproTech; cat. no. 500-P178) group; iv) 5-ASA (cat. no. 461814-25G; Sigma-Aldrich; Merck KGaA; 20 mg/g body weight) positive control group; and v) negative control group (equal volume saline gavage). The PD extracts (20 mg/g body weight) were administered intragastrically to the oxazolone-induced colitis mice once a day for 7 days.

The disease activity index (DAI) was used to evaluate the grade and extent of intestinal inflammation according to an established index system (18). During the treatment, the body weight, stool characteristics and the degree of bloody stool were recorded daily and were subsequently assessed using a DAI scoring system. The body weight change in each mouse was calculated using the following formula: DAI=[(percent weight loss + diarrheal stool score + bloody stool score)]/3. The scores ranged from 0 to 4, including the following three parameters: Weight loss, stool consistency and rectal bleeding.

RT-qPCR was used to evaluate mRNA expression in different colon tissue groups. The extraction of total RNA was performed using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instruction. Next, purified RNA was subjected to RT using the PrimeScript RT reagent kit (Takara Bio, Inc.). RT-qPCR was performed using SYBR Green qPCR Master Mix (Takara Bio, Inc.) as follows: Initial degeneration at 95°C for 30 sec, followed by 40 cycles of 95°C for 5 sec and 60°C for 30–60 sec. The relative expression of mRNA (normalized against actin) was calculated by the 2^−ΔΔCq method (19).

The following primers were used for each target gene: Beclin-1 forward, 5′-CCCAGCCAGGATGATGTCTAC-3′ and reverse, 5′-AGGTCTCCAGTGACCTTGAGT-3′; LC3B forward, 5′-CCGTAGTTCGCTGTACGAGG-3′ and reverse, 5′-AACTCACGTCGGATGTCCAG-3′; PI3K forward, 5′-CTGGAATGTGTGGCTGGAGT-3′ and reverse, 5′-AGGAGGAAGCGGTGGTCTAT-3′; Akt forward, 5′-ATGAACGACGTAGCCATTGTG-3′ and reverse, 5′-TTGTAGCCAATAAAGGTGCCAT-3′; mTORC1 forward, 5′-ATCGTGCTGTTGGGTGAGAG-3′ and reverse, 5′-TGGATCTCCAGCTCTCCGAA-3′; autophagy-related protein 13 (ATG13) forward, 5′-TGGACCATCAACCGGAAACT-3′ and reverse, 5′-CAAGGGTATGCAGCTGTCCA-3′; zona occludens protein 1 (ZO-1) forward, 5′-CATGGCCGGGAATGATGAGA-3′ and reverse, 5′-TGAAACTCTTGCCTCGTCCG-3′; claudin-2 forward, 5′-GATCGGCTCCATCGTCAGC-3′ and reverse, 5′-TGTTGGGTAAGAGGTTGTTTTCC-3′; and GAPDH forward, 5′-AGGTCGGTGTGAACGGATTTG-3′ and reverse, 5′-TGTAGACCATGTAGTTGAGGTCA-3′.

ELISA kits (cat. no. ab219634; Abcam) were used to detect the amount of secretory IL-13 in the serum and the expression of myeloperoxidase (MPO; cat. no. PA5-16672, Gibco; Thermo Fisher Scientific, Inc.) in the colon tissues of the mice, according to the manufacturer's instructions. The OD value was measured at 450 nm using a microplate reader (Thermo Fisher Scientific, Inc.) and the concentration was calculated using a standard curve.

Colonic mucosal NKT cell infiltration was observed using H&E staining. The Chiu's scoring system was used to quantitatively determine the histological scores of the intestine (20). Following 7 days of growth, the mice were sacrificed by decapitation and 10 cm colonic mucosa was removed from four mice in each group, washed with PBS and fixed in 10% natural buffered formalin at room temperature for 30 min. The fixed tissues were embedded in paraffin, cut into tissue sections (5-µm thick) and stained with H&E (25°C; 3 h; Beyotime Institute of Biotechnology). The slides were examined with a light microscope (Nikon Corporation) and the data derived were qualitatively evaluated with QWin software 3.1 (Leica Microsystems, Inc.) using the following parameter: NKT cell infiltration.

Frozen colonic tissues in liquid nitrogen were mechanically homogenized (20% w/v) in lysis buffer (pH 7.4) containing 0.1 mM EDTA, 20 mM HEPES, 12.5 mM MgCl₂, 150 mM 0.1% Nonidet P40, NaCl, 1 mM dithiothreitol, 0.2 mM phenylmethylsulfonyl fluoride, and 1 µg/ml leupeptin, aprotinin and pepstatin. Homogenates were sonicated (120 Hz; 4°C; 20 min) and subsequently centrifuged for 10 min at 12,000 × g at 4°C. Protein concentration was assessed using a BCA assay kit (Beyotime Institute of Biotechnology). Equivalent quantities (20 µg) of protein from the cell lysates were loaded onto 15% gels (Beyotime Institute of Biotechnology), separated via SDS-PAGE and then separated proteins were transferred onto polyvinylidene fluoride membranes (EMD Millipore). The membranes were blocked with 5% non-fat milk in Tris-buffered saline (25°C; 30 min) at room temperature and subsequently incubated with primary antibodies (4°C; overnight) against Beclin-1 (1:1,000; cat. no. 3738; Cell Signaling Technology, Inc.), LC3B (1:1,000; cat. no. 43566; Cell Signaling Technology, Inc.), PI3K (1:1,000; cat. no. 4255; Cell Signaling Technology, Inc.), Akt (1:1,000; cat. no. 4691; Cell Signaling Technology, Inc.), mTORC1 (1:1,000; cat. no. 10441-1-AP; ProteinTech Group, Inc.), phosphorylated (p)-MTORC1 (1:1,000; cat. no. 5536; Cell Signaling Technology, Inc.), ATG13 (1:1,000; cat. no. 13273; Cell Signaling Technology, Inc.), Occludin (1:1,000; cat. no. 91131; Cell Signaling Technology, Inc.), ZO-1 (1:1,000; cat. no. ab59720; Abcam), claudin-2 (1:1,000; cat. no. 48120; Cell Signaling Technology, Inc.) and GAPDH (1:1,000; cat. no. 5174; Cell Signaling Technology, Inc.) overnight at 4°C. The following day, the membranes were rinsed using TBS with 1% Tween-20 three times for 8 min each, followed by incubation at room temperature for 1 h with goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (1:1,000; Cell Signaling Technology, #7074, Inc.). The membranes were washed three times and detected using ECL reagent (1:1,000; Cell Signaling Technology, Inc.). Finally, the densitometry was analyzed by Image J v1.8.0.112 software (National Institutes of Health).

LC3B expression levels were analyzed by immunofluorescence in colonic tissues. The 2-µm paraffin-embedded sections of the colon tissue sections (4% PFA; 25°C; 24 h) were dewaxed, rehydrated with gradient alcohol, subjected to antigen repair (100°C) and subsequently rinsed three times with 0.01M PBS (25°C) with 0.1% Tween-20 (PBST). The sections were blocked (25°C) with 5% BSA and 0.3% Triton X-100 for 1 h, placed in a wet box for an additional 1 h and incubated with a primary antibody against LC3B (1:1,000; cat. no. 43566; Cell Signaling Technology, Inc.) at 4°C overnight. The following day, the samples were washed three times with PBS for 5 min each time and then was incubated with Alexa-488-conjugated donkey secondary antibody against rabbit antibody (1:3,000; Cell Signaling Technology, Inc.; cat. no. 4412). The nuclei were counterstained with 10 µg/ml DAPI (25°C; 3 min) at room temperature. The images were captured using an Olympus fluorescent microscope (Olympus Corporation).

The results are presented as the mean ± standard deviation. All data presented are representative of ≥3 experiments. Statistical analysis was conducted using GraphPad Prism 6.2 software (GraphPad Software, Inc.) and significance was determined using two-way ANOVA and Dunnett's multiple comparisons test. P<0.05 was considered to indicate a statistically significant difference.

In order to investigate the therapeutic effect of PD on colitis, the present study induced colitis in a mouse model by skin pre-sensitization and colonic administration of oxazolone. The results indicated that the mice injected with oxazolone developed diarrhea, bloody stool and weight loss. Histological changes, such as monocyte infiltration, intestinal wall thickening, mucosal hyperplasia and distortion of crypt structures, were observed in the model group compared with the normal control group (Fig. 1A). Concomitantly, it was found that PD significantly reduced H&E and DAI scores compared with the control (Tables I and II). These effects were also noted with IL-13 intervention and oral administration of 5-ASA (Fig. 1A-C). These findings indicated that PD mitigated the severity of oxazolone-induced colitis in mice.

The length of the colon is considered an indirect indicator of inflammation and is negatively correlated with the severity of colitis (21). In addition, the effects of PD were investigated on inflammation of oxazolone-induced colitis. The colon tissues were significantly longer and lighter in the normal control, IL-13 intervention, 5-ASA and PD groups compared with those of the model group; Fig. 2A-C). The accumulation of MPO in the colon is a marker of neutrophil entry into the tissue and IL-13 is an inflammatory cytokine characteristic of UC. The ELISA results indicated that PD caused a significant reduction in the levels of MPO in the colon tissues and a significant decrease in serum IL-13 levels compared with the model group (Fig. 3A and B). The same results were found in the IL-13 and 5-ASA groups compared with model group (Fig. 3A and B). These findings illustrated that PD could inhibit oxazolone-induced colitis inflammation.

To assess the levels of autophagy in the colon tissues, western blotting, immunofluorescence and RT-qPCR assays confirmed that the percentage of cells with low LC3 fluorescence intensity was significantly higher compared with that of the normal control group (Fig. 4A and B). The mRNA and protein expression levels of Beclin1 and LC3 were upregulated in the model group compared with control (Fig. 5A-E). However, Beclin1 and LC3 expression levels significantly decreased when the mice were administered PD, 5-ASA or IL-13 (Fig. 5A-E). This indicated that PD significantly inhibited tissue autophagy in oxazolone-induced colitis. In order to further investigate the molecular mechanism of action of PD, the expression levels of the associated molecules in the PI3K-Akt-mTORC1 signaling pathway were investigated via western blotting and RT-qPCR assays. The results demonstrated that PI3K, Akt, mTORC1 and ATG13 expression levels were significantly downregulated in the PD, 5-ASA or IL-13 intervention groups (Fig. 6A-D). In addition, western blot analysis indicated that the of p-AKT, p-mTORC1 and ATG13 expression levels were increased in the model group compared with the control group; however, expression levels were decreased in the PD, 5-ASA and IL-13 groups compared with the model group. There was no change in the expression levels of Akt or mTORC1.

(Fig. 6E and F). Therefore, these results suggested that the protective effects of PD against oxazolone-induced colitis were exerted via the regulation of the PI3K-Akt-mTORC1 signaling pathway.

Several studies have shown that the expression levels of tight junction proteins found between intestinal epithelial cells, are affected in inflammatory bowel disease, leading to alterations in tight junction structure and function (10,22). Therefore, to determine whether PD maintains the integrity of the intestinal barrier, occludin and ZO-1 expression levels were detected. The data indicated that they were decreased, whereas claudin-2 levels were increased compared with model (Fig. 7A-C). The relative mRNA expression levels of these markers including occludin and ZO-1were altered compared with model (Fig. 7A) following PD treatment. These results suggested that PD protected the intestinal barrier by altering the expression levels of the colonic tissue tight junction proteins, which were induced by inflammatory stimuli.