RAW264.7 cells, a macrophage line of mouse, was obtained from Cell Bank of Academy of Sciences (Shanghai, Chain). After detection of mycoplasma-free contamination, the cells were seeded in DMEM medium (Gibco, Carlsbad, CA, United States) with 10% fetal bovine serum, 200 μg/ml L-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin and cultured at 37°C, 5% CO₂.

M1 macrophages were generated by stimulation with 100 ng/mL LPS (Escherichia coli, Sigma-Aldrich, United States) and 10 ng/mL IFN-γ (Sino Biological, Shanghai, China) for 24 h. RAW264.7 cells were induced with 10 ng/mL IL-4 and 10 ng/mL IL-13 (both purchased from Sino Biological) for 24 h to generate of M2 macrophages. LBP and LPS/IFN-γ treated simultaneously RAW264.7 cells for 24 h, or LBP alone treated RAW264.7 cells for 24 h. Fludarabine and AS1517499 (both purchased from TargetMol, Shanghai, China) were employed to validate the STAT1and STAT6 signaling pathways.

The procedure of all animal experiment was approved by the Institutional Animal Care and Use Committee (IACUC) of Shaanxi Normal University (permit number: SCXK (Shan) 2021-002), China. The mice were kept in an environment with the controlled temperature, humidity, and light (23°C–24°C, 55%–65%, the cycle of 12/12 h light/dark) and provided with adequate water and food.

C57BL/6 male mice (22–27 g, 8 weeks old) were weighed. Then they were divided to five groups randomly (n = 5 in each group). There are group A, NC (negative control); group B, NC + LBP 200 mg/kg; group C, DSS; group D, DSS + LBP 100 mg/kg (low-dose), and group E, DSS + LBP 200 mg/kg (high-dose), respectively. The mice from groups A and B were given drinking water for 0–9 days. The mice from groups C, D, and E were administered 2.5% DSS (MW50000; MP Biomedicals, Irvine, CA, United States) in water of drinking for 0–7 days to build the experimental colitis murine model.

LBP was purchased from Solarbio Science and Technology Co., Ltd. (Cat No. SP9311, Beijing, China), derived fromLycium fruits with water extract-alcohol precipitation method. Total sugar content was determined by the phenol-sulfuric acid method using glucose as standard (Dubois et al., 1956). The content of total sugar was 93.2%. The 200 and 100 mg/kg LBP were used in animal experiments according to the previous literatures (Zhang et al., 2020; Xie et al., 2021). The mice in groups D, B, and E were given every day 100 mg/kg or 200 mg/kg LBP (dissolved in 1×PBS) by gavage at a fixed time from the second day of the DSS perform, and the mice in groups A and C were given the same amount of 1×PBS by gavage until the ninth day.

Each mouse of every group was under close, daily observation for the body weight, the consistency of stool and hematochezia. The DAI was calculated based on the loss of weight, the consistency of stool and hematochezia, based on the scoring standards in the literature (Shon et al., 2015). The mice were euthanized on day nine.

In RAW264.7 cells, the influence of LBP on the viability was tested by the CCK-8 kit (HB-CCK8-1, Hanbio Biotechnology, Shanghai, China). Briefly, 5 × 10³ cells/well in 100 μL of cell suspension were plated in the 96-well plate. 200, 400, 800, or 1,600 μg/mL LBP were used to treated the cells. After incubating for 24 h at 37°C, 5% CO₂, discarded the medium, then 100 μL of fresh medium and 10 μL of CCK8 reagent were added for another 4 h incubation followed by the absorbance reading at 450 nm using the microplate reader (Thermo, United States). The cell viability of each group was calculated according to the following formula: the cells viability (%) = (OD_LBP -OD _blank)⁄(OD _control -OD _blank) × 100.

The Total RNA from various groups of RAW264.7 cells was extracted with TRIzol ^® reagent (Takara, Dalian, China). Then 500 ng total RNA was used for the reverse transcription experiment with the Prime Script™ RT Reagent Kit (Takara). In a system of ABI 7900HT Fast Real-Time PCR, the PCR Kit (SYBR green), (QIAGEN, Hilden, Germany) was used to perform the qPCR experiment. Each test was repeated at least three times. The mRNA expression level of all genes tested was normalized with ß-actin as the standard using the 2^−△△Ct method. The main primers for qPCR are shown in Table 1.

The cells were washed and fixed using paraformaldehyde at a concentration of 4%, then incubated with the diluted primary antibodies at 4°C overnight. The primary antibodies used are anti-NOS2 and anti-Arg-1 (rabbit polyclonal, 1:300, both from Proteintech, Chicago, IL, United States), p-STAT1 (rabbit polyclonal, 1:200, Affinity, Shanghai, China). After washing of cells with 1×PBS, cells were incubated with goat anti-rabbit IgG Dylight 594 or Dylight 488 (1:500, both from Abbkine, Wuhan, China) for 1 h in the dark. Representative pictures were captured with a Zeiss Imager M2 microscope and were analyzed using Image J software (version: v1.8.0).

Mice in the five groups were anesthetized with ether, and their colons were removed and fixed with paraformaldehyde at a concentration of 4%. Then, the tissues were paraffin-embedded to make wax blocks and cut into 5 mm sections. Next, the H&E staining were performed after the sections were dried. The specific staining procedure was performed according to our previous staining procedure (Mao et al., 2017).

Immunofluorescence double-staining in mouse colon tissue was performed. Paraffin-embedded slides were performed the quenching of dewaxed, rehydrated, antigen repair, the block of endogenous peroxidase activity and non-specific antigens. Next, the slides were co-incubated with the following primary antibodies: anti-F4/80 (rabbit polyclonal, 1:10,000), anti-NOS2, and anti-Arg-1 (rabbit polyclonal, 1:1,500) (all purchased from Servicebio Co., Ltd., Wuhan, China); and p-STAT1, p-STAT6 (rabbit polyclonal, 1:200) (both from Affinity, Shanghai, China). Then the slides were co-incubated with the corresponding HRP labeled secondary antibody for 50 min at room temperature. Next, the corresponding opal tyramines signal amplification (TSA) stain working solution (Cat: G1226, purchased from Servicebio Co., Ltd., Wuhan, China) was dripped onto the tissue to ensure complete tissue coverage, and incubated for 10 min at room temperature. After the second run, DAPI was used to stain nucleus. Representative pictures of H&E and immunofluorescence double-staining were captured with a Zeiss Imager M2 microscope. The number of positive cells was counted in five randomly selected areas under fluorescence microscope.

Mouse colon tissues or RAW264.7 cells were added with an appropriate amount of RIPA buffer, followed by protease inhibitors and phosphatase inhibitors, and lysed on ice for 40 min. Then, nuclear and cytoplasmic proteins from RAW264.7 cells were separated using the Nuclear and Cytoplasmic Extraction Kit. After protein quantification, the same amount of protein was added to each gel well, and separated by 10% resolving gel and 5% stacking gel and transferred onto PVDF membranes. After blocking with 5% BSA for 2–3 h, incubated PVDF membranes with NOS2, Arg-1 (rabbit polyclonal, 1:500), Lamin A/C (rabbit polyclonal, 1:1,000) (all purchased from Proteintech, Chicago, IL, United States); and p-STAT1, STAT1, p-STAT6, and STAT6 (rabbit polyclonal, 1:500) (all from Affinity, Shanghai, China); ß-actin (rabbit polyclonal, 1:1,000) (from Santa Cruz, Dallas, TX, United States) at 4°C overnight. Next, the membranes were disposed with HRP-labeled Goat IgG antibody (anti-rabbit, 1:10,000) (from Zhong Shan, Beijing, China) for 1 h. Then the membranes were washed with 1×PBST, and incubated in a photoluminescence solution for imaging. The bands of the protein were analyzed by Image J software (version: v1.8.0).

Comparisons between the two groups were performed using Student’s t-test. One-way ANOVA was used to compare three or more groups. Statistical analysis of the obtained data was performed using GraphPad Prism 8.0 software, and the data are reported as the mean ± SD. In all figures, p < 0.05 indicates that the difference was statistically significant.

To study whether LBP has a protective role for IBD, a murine colitis model of DSS-induced was established. The execution process of the animal experiment is shown in Figure 1A. In the process of the experiment, the DSS-treated mice exhibited obvious symptoms of the weight loss, diarrhea, and bloody stool. In contrast with the DSS group, the high dose of LBP (DSS + LBP 200 mg/kg) group was found to significantly protect from weight loss (Figure 1B) and atrophy of the colon (Figures 1C, D), and reduce the DAI score in mice with colitis (Figure 1E). However, the treatment group with low dose of LBP (DSS + LBP 100 mg/kg) was also found to obviously protect from the weight loss in mice relative to the DSS group on day seven after DSS induction (Figure 1B).

On the basis of assessing the symptoms of colitis in mice, the histological morphology of the colon tissue was further investigated. The results indicated that the DSS + LBP 200 mg/kg group showed significantly less disordered tissue structure, and the infiltration of inflammatory cells and histological damage as significantly lower, as was the histological score in contrast with the DSS group (Figures 1F, G).

In addition, a great number of macrophages were found in the colon tissue of mice DSS- induced. The previous researches have shown that the polarization of macrophages played a dominant role in the onset and outcome of IBD (Zhu et al., 2016; Zhou et al., 2019). To further investigate whether LBP alleviated IBD by regulating macrophage polarization, the protein level of NOS2 and Arg-1 were examined by Western blot in colon tissues. It was found that the protein level of NOS2 in the DSS + LBP 200 mg/kg group was markedly decreased in contrast with the DSS group (Figures 1H, I), while the protein level of Arg-1 was obviously increased in contrast with the DSS group (Figures 1H, J), suggesting that LBP may protect against IBD by regulating macrophage polarization.

Next, immunofluorescence double-staining was performed to further detect whether LBP treatment affects the changes in the quantity of M1 and M2 macrophages in the colon tissues of mice with colitis. The results showed that 100 and 200 mg/kg LBP treatment both significantly reduced the quantity of F4/80⁺ NOS2⁺ M1 macrophages, and the effect of 200 mg/kg LBP was more obvious (Figure 2A). 200 mg/kg LBP treatment obviously increased F4/80⁺ Arg-1⁺ M2 macrophages (Figure 2B) in the colonic tissue of mice with colitis. The quantitative results of the number of F4/80⁺ NOS2⁺ M1 and F4/80⁺ Arg-1⁺ M2 macrophages in Figures 2A, B were shown in Figures 2C, D. The results further proved that LBP could protect against IBD by regulating macrophage polarization.

The results above suggested that LBP could alleviate the symptoms of DSS-induced colitis by potentially regulating the polarization of macrophages. In order to further research the potential mechanisms of LBP regulating macrophage polarization, the experiments were further performed using RAW264.7 macrophage cell line for the purpose in vitro.

To optimize the dose of LPB in vitro experiment, the different doses of LBP were tested in vitro by the evaluation of the viability of RAW264.7 cells by CCK8 assay. The results indicated that all concentrations of LBP tested were found to have no toxicity to the RAW264.7 cells (Figure 3A). Then, the influence of different concentrations of LBP on the protein level of NOS2 in RAW264.7 cells with M1 phenotype was detected. It was found that 400 μg/mL LBP had a significant role in reducing the expression of NOS2 (Figures 3B, C). Therefore, 400 μg/mL of LBP was selected as the final working concentration for subsequent experiments.

Then, 400 μg/ml as the treatment concentration, the phenotypic experiments on the effect of LBP on the polarization of macrophage were performed in RAW264.7 cells. The immunofluorescence results revealed that the stimulation of LPS/IFN-γ obviously increased the expression of NOS2, however, this elevated expression was significantly suppressed in the presence of LBP (Figure 4A). The quantification of the mean fluorescence intensity of Figure 4A each group is shown in Figure 4B. Differently, the expression of Arg-1 was obviously decreased after the stimulation of LPS/IFN-γ, however, the expression of Arg-1 was increased when RAW264.7 cells were subjected to both LBP and LPS/IFN-γ (Figure 4C). The quantification of the mean fluorescence intensity of Figure 4C each group is shown in Figure 4D.

To further explore the effect of LBP on macrophage polarization, the mRNA and protein levels of macrophage polarization-related genes were examined in RAW264.7 cells. The results revealed that the expression of NOS2 and IL-6 were obviously increased after stimulation using LPS/IFN-γ, while the expressions of both were significantly decreased after treatment with LBP (Figure 5A). The expression of IL-10 was increased after stimulation with LPS/IFN-γ + LBP compared with no LBP treatment (Figure 5A). The result of Western blot demonstrated that the protein level of NOS2 was obviously advanced, and the protein level of Arg-1 was markedly reduced after stimulation with LPS/IFN-γ. Treatment with LBP resulted in higher Arg-1 expression and lower NOS2 expression (Figure 5B). The quantification of protein bands in each group is shown in Figure 5C.

Next, IL-4/IL-13 were used to induce M2 polarization and stimulated M2 RAW264.7 cells using LPS/IFN-γ to observe the transition from M2 to M1 with or without LBP. The result of a qPCR showed that after LPS/IFN-γ treatment of M2 RAW264.7 cells, the mRNA level of NOS2 and IFN-γ were obviously enhanced, while the expression of Ym1 was markedly attenuated. However, in the presence of LBP, the mRNA level of NOS2 and IFN-γ were obviously reduced, and expression of Ym1 was increased (Figure 5D). Western blot results showed that M2 RAW264.7 cells after stimulation with LPS/IFN-γ showed higher NOS2 expression and lower Arg-1 expression. In the presence of LBP, the elevated expression of NOS2 was suppressed, whereas the expression of Arg-1 was significantly elevated (Figure 5E). The quantification of protein bands in each group is shown in Figure 5F. Thus, LBP can inhibit the repolarization of RAW264.7 cells from M2 to M1, which is beneficial for maintaining the M2-like phenotype.

It has been documented that STAT1 and STAT6 signaling pathways involves the polarization of M1/M2-like macrophage. To investigate whether LBP affects macrophage polarization by STAT1 and STAT6 signaling pathways, the effect of LBP on the STAT1 and STAT6 pathways was investigated in RAW264.7 cells. First, we detected the effect of LBP on p-STAT1 by cellular immunofluorescence (Figure 6A). The results of mean fluorescence intensity of total p-STAT1 (Figure 6B) and mean fluorescence intensity nucleoplasm ratio (Figure 6C) revealed that the protein level of total p-STAT1 and p-STAT1 in the nuclei were obviously increased after LPS/IFN-γ treatment, suggesting the activation of STAT1 signaling. In the presence of LBP, the protein level of total p-STAT1 and p-STAT1 in the nuclei were obviously decreased.

Western blot results indicated that the protein level of p-STAT1 in the cytosol and nuclei of RAW264.7 cells were obviously increased after LPS/IFN-γ treatment. When RAW264.7 cells were subjected to both LBP and LPS/IFN-γ, the expression of p-STAT1 in the cytosol (Figures 6D, E) and nuclei (Figures 6D, F) was markedly reduced.

LBP was found to be beneficial in maintaining the M2-like phenotype (Figure 5). Furthermore, the activation of the STAT6 signal pathway is required for maintaining the M2-like phenotype (Zhou et al., 2020). To test whether LBP maintained M2-like phenotype through the STAT6 signaling pathway, the protein level of p-STAT6 in the cytosol and nuclei of RAW264.7 cells with or without LBP treatment was tested via Western blot. The results proved LBP did increase the expression of p-STAT6 in the cytosol and nuclei (Figure 6G). The quantification of the protein bands of p-STAT6 in the cytosol and nuclei was shown in Figures 6H, I. At the same time, the level of p-STAT1 was detected in RAW264.7 cells with or without LBP treatment, and the results showed that treating RAW264.7 cells with LBP alone did not affect the level of p-STAT1 in the cytosol (Supplementary Figure S1A, B) and nuclei (Supplementary Figure S1A, C).

Subsequently, we investigated whether the downregulation effect of LBP on LPS/IFN-γ-induced M1 macrophages depends on the STAT1 pathway. For this purpose, a potent inhibitor of p-STAT1 (Fludarabine) was administrated in LPS/IFN-γ-stimulated RAW264.7 cells with or without LBP, and the protein level of p-STAT1 and NOS2 were detected by Western blot. The results showed that LBP could significantly decreased the level of p-STAT1 and NOS2, suggesting that LBP inhibits the M1 phenotype by inhibiting the STAT1 signaling pathway. Moreover, when Fludarabine dramatically reduced the protein level of p-STAT1, the inhibitory effect of LBP on M1 macrophage phenotype was significantly disrupted (Figure 7A). The quantification of the protein bands of p-STAT1 and NOS2 was shown in Figures 7B, C.

We also investigated whether the effect of LBP on M2 macrophages depends on the STAT6 signaling pathway by using an effective inhibitor of p-STAT6 (AS1517499) in IL4/IL13-stimulated RAW264.7 cells with or without LBP. Western blot analysis showed that LBP could significantly increase the protein level of p-STAT6 and Arg-1, suggesting that LBP promotes the M2 phenotype by promoting the STAT6 signaling pathway. Moreover, the effect of LBP on promoting M2 phenotype disappeared when p-STAT6 was significantly inhibited (Figure 7D). The quantification of the protein bands of p-STAT6 and Arg-1 was shown in Figures 7E, F. Taken together, these data indicate that LBP regulates M1/M2 macrophage polarization through STAT1 and STAT6 signaling pathways.

We had demonstrated in vitro that LBP regulated M1/M2 macrophage polarization by modulating the STAT1 and STAT6 pathways, and also found that LBP could regulate macrophage polarization in DSS-induced colitis mice by the detection of the expression of M1 or M2 marker in colon tissue (Figures 1H–J; Figure 2). In order to further verify whether LBP regulated M1/M2 macrophage polarization by STAT1 and STAT6 signaling pathways in vivo, the immunofluorescent double-staining of F4/80 and p-STAT1 or p-STAT6 were performed to detect the level of p-STAT1 and p-STAT6 in intestinal macrophages of mice. The results indicated that the quantity of F4/80⁺ p-STAT1⁺ cells in the DSS + LBP 100 mg/kg group and the DSS + LBP 200 mg/kg group was obviously reduced in contrast with the DSS group (Figure 8A), while the quantity of F4/80⁺ p-STAT6⁺ cells in DSS + LBP 200 mg/kg group was obviously increased in contrast with the DSS group (Figure 8B). The quantification of results of the number of F4/80⁺ p-STAT1⁺ cells, and F4/80⁺ p-STAT6⁺ cells in Figures 8A, B were revealed in Figures 8C, D. This result showed that LBP regulated STAT1 and STAT6 pathways in vivo, which was consistent with the results in vitro.

Based on the results in vitro and in vivo, the mechanisms of LBP regulating macrophage polarization was summarized in Figure 9. In the STAT1 and STAT6 pathways, LPS and IFN-γ can activate the STAT1 pathway, which promotes M1-like polarization of macrophage; IL-4 and IL-13 activate the STAT6 signaling pathway, which promotes M2-like polarization of macrophage. LBP inhibited M1 macrophage polarization by reducing the level of p-STAT1 and promoted M2 macrophage polarization by increasing the level of p-STAT6 in vivo and in vitro, and resulting in the alleviation of DSS-induced IBD in mice.