Evodiamine (>98% HPLC) was purchased from Chengdu Puruifa Biotechnology Co., Ltd. (Chengdu, China) (Shen et al., 2019). DSS (36–50 kDa) was obtained from MP Biomedicals (Aurora, OH). Phorbol 12-myristate 13-acetate (PMA), lipopolysaccharide (LPS), adenosine triphosphate (ATP), DSS (5 kDa), 5-aminosalicylic acid (5-ASA) and 4′,6-diamidino-2-phenylindole (DAPI) were bought from Sigma–Aldrich (St. Louis, MO, United States). ELISA kits for murine MPO, IL-1β and IL-18 were purchased from Multi Sciences Biotech Co., Ltd. (Hangzhou, China), ELISA kits for human IL-1β and IL-18 were obtained from Excell Biotech Co., Ltd. (Taicang, China). RPMI-1640 medium, fetal bovine serum (FBS) (Gibco, United States), anti-NLRP3, anti-Caspase1 and anti-IL-1β antibodies were purchased from ABclonal Technology (ABclonal, Boston, United States), anti-ASC, anti-cleaved caspase1 and anti-LC3 antibodies were bought from cell signaling technology (CST, Danvers, United States), anti-ASC (mouse) antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, United States), anti-CD11b and anti-LAMP-1 antibodies were obtained from Abcam technology (Cambridge, United Kingdom), anti-GAPDH and anti-P62 antibodies were purchased from Proteintech group (Chicago, United States), HRP-conjugated secondary antibody, Alexa Fluor-488 conjugated goat anti-rabbit or anti-mouse IgG (H + L) secondary antibodies and Alexa Flour-594 conjugated goat anti-rabbit IgG (H + L) secondary antibody were bought from Invitrogen (Carlsbad, United States).

Human THP-1 cells were grown in RPMI-1640 medium, supplemented with 10% (v/v) fetal bovine serum, 100 mg/ml streptomycin and 100 U/ml penicillin under a humidified environment with 5% CO₂ at 37°C.

C57BL/6J mice (male, 6–8 weeks old, 18–22 g) were purchased from Model Animal Research Center of Nanjing University (Nanjing, China). All mice were housed in the experimental animal facility, water and food were available ad libitum. Mice were randomly assigned to control (vehicle; 0.5% sodium carboxymethylcellulose, CMC-Na), DSS-treated, evodiamine-treated (20, 40, and 60 mg/kg), evodiamine-only (60 mg/kg) and 5-ASA-treated (100 mg/kg, positive control) groups. Acute colitis was induced by 3% DSS (36–50 kDa) in drinking water for 7 days, then change treated water to normal water for another continuously 3 days before sacrifice (Okumura et al., 2016; Chen H. et al., 2017). Evodiamine and 5-ASA dissolved in the sodium solution of carboxymethyl cellulose (CMC-Na, 0.5%) and treated by oral gavage once a day from day 1 to day 10. All animal welfare and experiment procedures were carried out rigid according to guideline of European Directive 2010/63/EU and approved by the Nanjing University Animal Care and Use Committee (NJU-ACUC).

The severity of colitis was evaluated by colon length, the DAI, histologic analyses, and MPO activity (Tasaka et al., 2015). Body weight, fecal blood loss and stool consistency were monitored daily. Briefly, the disease activity index (DAI) was calculated as the sum of three parameters as follows: body weight loss (0, ≤1%; 1, 1–5%; 2, 5–10%; 3, 10–15%; 4 ≥ 15%), diarrhea (0, normal; 2, loose stools; 4, watery diarrhea) and blood in the stool (0, no bleeding; 2, slight bleeding; 4, gross bleeding) (Song and Park, 2014), also shown in Table 1. After animals were sacrificed, part of the colon was dissected. To evaluate the degree of morphometric inflammation, the length between the ileocecal junction and the anal verge were measured. The distal part of colon was fixed in 4% paraformaldehyde, embedded in paraffin and sliced into 5 μm-thick sections. The sections were routinely stained with hematoxylin and eosin (H&E) (Hayashi et al., 2014). Formalin-fixed hematoxylin tissue sections were evaluated by an experienced pathologist microscopically in a blinded way. Severity of inflammation, crypt damage and ulceration were simultaneously evaluated (Wu et al., 2012).

Myeloperoxidase (MPO) level was determined by ELISA to evaluate neutrophil infiltration into inflamed colonic mucosa. Briefly, colon tissues were weighed and homogenized in cold phosphate buffer saline (PBS) for 5% colonic tissue homogenate (Mei et al., 2019). Supernatants were collected to determine MPO levels according to the manufacturer’s instructions after they were centrifuged at 12,000 g at 4°C for 15 min. The results were presented as absorbance per mg of tissue.

Colonic tissues of the mice were weighed and homogenized in cold PBS for 5% colonic tissue homogenate. To obtain the supernatants, homogenate was centrifuged at 12,000 g at 4°C for 15 min. Protein concentration in the supernatants was determined by BCA protein assay, and cytokines in the colonic tissue homogenate were analyzed by ELISA kits. For in vitro assay, THP-1 cells were seeded in 12-well plates and the supernatants were harvested after stimulation. IL-1β and IL-18 were detected using ELISA kits according to the manufacturer’s instructions.

Paraffin-embedded colonic tissues were stained with CD11b antibody to analyze inflammatory cell infiltration. The sections were deparaffinized in xylene and rehydrated in a decreased alcohol gradient and washed with ddH₂O. After unmasking antigens, colon sections were blocked with 5% BSA and incubated with anti-CD11b antibody (1:200) at 4°C overnight, and FITC labeled secondary antibody at room temperature for 1 h. Finally, sections were counterstained with DAPI. Images were acquired by an Olympus FV3000 confocal microscope.

The distal colon was removed, washed with ice-cold PBS, and homogenized in lysis buffer containing 1% Triton X-100, 10 mM Tris (pH 7.4), 1 mM EDTA, 1 mM EGTA, 150 mM NaCl, protein inhibitor cocktail (1:100) and Phenylmethylsulfonyl fluoride (PMSF) by ultrasonic splitting on ice. After centrifugation at 12,000 g for 30 min at 4°C, the supernatants were collected and protein concentrations were determined using a BCA protein assay kit (Pierce, Rockford, IL), then were stored at −80°C till for use. The lysates were mixed with Laemmli sample buffer and heated at 95°C for 5 min. The samples were subjected to electrophoresis on 10% SDS-polyacrylamide gels and transferred electrophoretically to PVDF membranes (Millipore, Billerica, MA). The membranes were blocked with TBST (0.1% Tween 20 in Tris-buffered saline) containing 5% skimmed milk at room temperature for 2 h, incubated with the primary antibodies at a predetermined dilution at 4°C overnight. Subsequently, the membranes were washed three times with TBST, incubated with the horseradish peroxidase (HRP)-conjugated secondary antibody at room temperature for 2 h, and then the membranes were visualized with ECL detection reagent.

THP-1 cells were seeded in 6-well plates and stimulated as indicated. Cells were harvested by scraping in 1 ml cold PBS containing 2 mM EDTA, and centrifuged at 1,500 g for 5 min at 4°C. Pellets were re-suspended the in 500 μl cold buffer A (20 mM HEPES-KOH, pH 7.5, 10 mM KCL, 1.5°mM MgCl_2, 1 mM EDTA, 1 mM EGTA, and 320 mM sucrose) and cell were lyzed by shearing 30 times through 21-gauge needle. Subsequently, the lysates were centrifuged at 1,800 g at 4°C for 8 min to remove the bulk nuclei and unbroken cells (Okada et al., 2014). The supernatants were diluted in CHAPS buffer [20 mM HEPES-KOH (pH 7.5), 5 mM MgCl₂, 0.5 mM EGTA, 0.1 mM PMSF, and 0.1% CHAPS] and centrifuged at 5,000 g for 8 min at 4°C to pellet ASC oligomers (Lorden et al., 2017). Then the pellets were resuspended in 50 μl CHAPS buffer with 2 mM disuccinimidyl suberate to cross-link at RT for 30 min and pelleted by centrifugation at 5,000 g for 8 min. The crosslinked pellets were collected for ASC oligomerization assay by immunoblotting (Chen L. et al., 2017).

The differentiated THP-1 cells were cultured on coverslips in 24-well plate and treated with LPS (100 ng/ml, 3 h), ATP (1 mM, 1 h) (Qin et al., 2016) and evodiamine (10 μM, 1 h). The cells were fixed in 4% paraformaldehyde for 30 min, washed for three times with PBS for 5 min, permeabilized with 0.2% Triton X-100 for 10 min, washed for three times, blocked with 5% BSA for 2 h (Wang et al., 2015). Cells were incubated with the primary anti-LC3 (1:150), anti-LAMP-1 (abcam, 1:100), anti-caspase1 (Proteintech, 1:150) or/and anti-ASC (CST, 1:200) overnight at 4°C. Washed for three times with PBST, cells were incubated with Alexa Flour 488 fluorescein-conjugated secondary antibody or the Alexa Flour 594 fluorescein-conjugated secondary antibody for 2 h (Thomas et al., 2010). The coverslips were counterstained with DAPI and images were taken by an Olympus FV3000 confocal microscope.

Statistical significance for multiple comparisons was performed by one-way ANOVA and Newman-Keuls analysis using Prism 5.0 (GraphPad Software). A p value <0.05 was considered statistically significant. All data are indicated as means ± SEM and from at least three independent experiments.

The structure of evodiamine is presented in Figure 1A. Colitis is characterized by significant diarrhea, loose feces, and visible fecal blood, resulting in a remarkable decrement in body weight (Liu W. et al., 2013). To evaluate the preventive effect of evodiamine on ulcer colitis, we performed the experimental colitis by DSS as showed in Figure 1B. Body weight of colitis mice induced by DSS decreased. After mice were treated by evodiamine for 10 days, colons length recovered, and the loss of body weight was improved (show in Figures.1C–E). To evaluate whether there is a toxic effect of evodiamine at the maximum concentration of 60 mg/kg, we administrated evodiamine alone to mice without DSS stimulation. There is no visible difference in body weight changes and colon length between evodiamine (60 mg/kg) and control mice. It suggested that the maximum dose of evodiamine was safe enough for treatment in this study.

To further confirm the anti-inflammatory effect of evodiamine on colitis, we observed pathological changes including mucosal damage, necrosis, the loss of tissue architecture, edema, and infiltration of inflammatory cells such as neutrophils and monocytes in colon tissues of mice by H&E staining and histological analysis. As shown in Figure 2A, pathological changes significantly improved by evodiamine dose-dependently (Figures 2A,B). DAI also deceased by evodiamine (Figure 2C). We continued to examine MPO levels, the important inflammatory index in colons of mice. MPO level in mice colon treated by evodiamine was much lower than that in the DSS group with the treatment of vehicle (Figure 2D). It suggested that attenuated ulcerative colitis injury by evodiamine resulted from the anti-inflammatory effect.

It was reported that the occurrence of colitis is associated with the secretion of IL-1β and IL-18 (Zaki et al., 2010), our data also demonstrated that secretion of IL-1β and IL-18 increased in colitis mice colon. Levels of IL-1β and IL-18 in the serum with DSS stimulation were decreased by evodiamine (Figures 3A,B). To further confirm the anti-inflammatory effect of evodiamine, we examined macrophage infiltration in the colon sections using immunofluorescence. Macrophages indicated by CD11b were detected in the mucosa of colitis mice treated by vehicle, while little CD11b were observed in evodiamine-treated colonic mice samples. It demonstrated that macrophages infiltration was inhibited by evodiamine (Figure 3C).

Inflammatory cytokine IL-1β and IL-18 were active after their inactive cytoplasmic precursor pro-IL-1β and pro-IL-18 were cleaved by caspase-1 (Liu W. et al., 2013). We continued to examine whether cleaved caspase-1 can be regulated by evodiamine. Cleaved-caspase-1 in colitis mice colon was markedly decreased by evodiamine, so was the ratio of cleaved caspase one to pro-caspase 1 (shown in Figures 3D–F), it suggested that caspase-1 activation in vivo was inhibited by evodiamine. It was consistent with the decrement of IL-1β and IL-18.

After Pro-caspase-1 bound to ASC, it was autoclaved and the mature active form cleaved caspase-1 produced (Liu W. et al., 2013). Our data indicated that ASC in the colon was decreased after the intervention of evodiamine, while there was no significant alteration in NLRP3 expression (shown in Figures 3D,G,H).

NLRP3 inflammasome was combined by ASC, NLRP3 and caspase-1 when cells face environmental stress. Under the stimulation of LPS and ATP, inflammasome was assembled, then caspase-1 was activated. As a result, the secretion of IL-1β and IL-18 was increased. To confirm whether evodiamine effects on NLRP3 inflammasome, we examined levels of IL-1β and IL-18 in THP1 cells by Elisa. As shown in Figures 4A,B, IL-1β and IL-18 remarkably increased in THP1 cells with the stimulation of LPS and ATP; the increment was significantly inhibited by evodiamine dose-dependently. To address how evodiamine act on NLRP3 inflammasome, we firstly examined NLRP3 expression using immunoblot. There is no remarkable alteration in NLRP3 in THP1 cells (Figures 4C,G), the level of ASC was up-regulated in the stimulation of DSS (Figures 4C,F). The decrement of IL-1β and IL-18 suggested the suppression of NLRP3 inflammasome, we further determine ASC oligomerization, the vital step for NLRP3 inflammasome activation (Lu et al., 2014; Chen L. et al., 2017).

As shown in Figure 4H, ASC oligomer was enhanced by ATP and LPS, and the increment was decreased by evodiamine. Bryan et al. reported that in response to ATP, ASC translocate from the nucleus, co-localizes with cytoplasmic caspase-1 (Bryan et al., 2009; Liu W. et al., 2013), then caspase-1 auto-cleaves and releases from the complex. Our immunoblot and immunofluorescence analysis showed that autocleavage of caspase-1 was inhibited by evodiamine (Figures 4C–E,I) so that colocalization of ASC and caspase-1 were interrupted, release of activated caspase-1 lessened. These data suggest that the assembly of NLRP3 inflammasome was inhibited by evodiamine via the suppression of ASC oligomerization.

Autophagy and inflammasome interacts (Yuk and Jo, 2013; Kim et al., 2016). Pro-inflammatory cytokine upstream of the NFκB signaling cascade is in a state of inaction or reduced activity when autophagy activated. We examined the activity of p-p65NFκB and p-IκB which are critical protein in NFκB signaling cascade. As shown in Figure 5A, evodiamine decreased the phosphorylation of p65NFκB and IκB, it suggested that autophagy is likely activated. To further explore whether autophagy was able to be regulated by evodiamine, we investigated the degradation of p62 to confirm whether a complete autophagic flux occurred in evodiamine treatment condition. P62 protein in colon of colitis mice significantly decreased by evodiamine (shown in Figure 5B). Next, we examined the effect of evodiamine on p62 and LC3 at continuous time. Expression of p62 was suppressed by evodiamine time-dependently (shown in Figure 5C). At the same time, there is a slight increment in LC3-II expression induced by evodiamine. It suggested that autophagy in THP1 cells enhanced by evodiamine.

We also assessed LC3 expression in the cell lysates of mice colon using immunoblot analysis. LC3-I is cytosolic, while the hallmarks of autophagy, LC3-II is membrane- bound. We found that LC3-II protein levels increased in colitis mice colon after evodiamine treatment (Figure 5B). It implies that evodiamine increased autophagosome formation. We continue to address whether NLRP3 inflammasome and autophagy were simultaneously regulated by evodiamine. Co-localization of LC3 and ASC, which was able to be regarded as the intracellular sensor of NLRP3 was detected. As shown in Figure 5D, the overlap between ASC and LC3 were enhanced after the treatment of evodiamine. It suggested that the regulation of NLRP3 inflammasome induced by evodiamine might be via the intervention of autophagy.

Increment in autophagosome formation or decrement in autophagosome degradation cause enhanced autophagosome accumulation. To study how evodiamine regulate autophagy, we observed levels of LC3-II and p62 in THP1 cells co-incubated with bafilomycin A1 (Baf A1), blocking the fusion of autophagolysosome to inhibit the degradation, and 3-methyladenine (3-MA), inhibiting the autophagosome formation (Figure 6A). The expression of LC3-II increased when THP1 cells were incubated with evodiamine in the presence of Baf A1 (shown in Figure 6B). There is no significant effect on the expression of LC3-II induced by evodiamine in the presence of 3-MA. These data suggested that autophagosome degradation was significantly inhibited and the formation of autophagosome is slightly influenced by evodiamine.

We continued to investigate the effect of evodiamine on inflammasome activity after autophagy was regulated by Baf A1 and 3-MA. The expression of cleaved caspase1, the marker of inflammasome activation, was increased by LPS and ATP, the overexpression was suppressed by evodiamine. Cleaved caspase-1 could not be inhibited by evodiamine in the presence of autophagy inhibitors such as Baf A1 and 3-MA (Figure 6C). These data indicated that inflammasome activation was inhibited by evodiamine in an autophagy-dependent manner. Once autophagosome mature, the lysosomal and autophagosome fuse to form autophagolysosome, result in the degradation of autophagosome (Ganley et al., 2011). To confirm the autophagolysosome formation, we further detected the co-localization of ASC and LAMP-1. Shown in Figure 6D, in cells incubated with evodiamine, immunostaining indicated that ASC surrounded by LAMP-1 which is the lysosome marker, suggesting that autophagosomes fused with inflammasome components and lysosomal. These data demonstrated that the formation of autophagolysosome was enhanced and degradation was inhibited by evodiamine.