C57BL/6 mice (male, 28–35 g) were purchased from Cyagen Biosciences Inc. (Suzhou, China). All mice were maintained under specific pathogen-free conditions at tge Experimental Animal Center, Dalian Medical University (Certificate of Conformity SYXK Liao 2013–0006). The study was approved by the Dalian Medical University Animal Care and Ethics Committee (No: AEE20046). The animals had ad libitum access to food and water for 2 weeks prior to experimentation in laboratory conditions of 23°C, 12 h/12 h light/dark, and 50% humidity. No animals died before the study was started.

Kumatakenin (PubChem CID: 5318869) of ≥95% purity and sulfasalazine (SASP, PubChem CID: 5359476) were obtained from Sigma-Aldrich (St. Louis, MO, United States of America). Anti-enolase (Eno3, ab232759), anti-iron regulatory protein (IRP1, ab183721), anti-4-HNE (ab48506), and anti-GAPDH (ab9485) antibodies were obtained from Abcam (Hong Kong, China). Other reagents were purchased from Sigma-Aldrich.

Acute colitis was induced in C57BL/6 mice with 2.5% dextran sulfate sodium (DSS) dissolved in sterile distilled water for 9 days. Mice were randomly assigned to eight groups of 12 animals each. Control mice in group I were given sterile distilled water. Groups II, III, IV, and V were treated with DSS for 9 days to induce colitis. Group II was a DSS colitis model control. Mice in groups III, IV, and V drank sterile distilled water containing DSS and were treated by intragastric administration of 150 mg/kg SASP, low-dose 25 mg/kg kumatakenin, or high dose 100 mg/kg kumatakenin (dissolved in sterile distilled water) for 9 consecutive days, respectively.

Groups Ⅵ, Ⅶ, and Ⅷ were administered Eno3 inhibitor ENOblock (1 mg/kg) by intraperitoneal injection once a day for 3 days to inhibit Eno3 expression in vivo. Group Ⅵ served as the Eno3 depletion control. Following pretreatment with ENOblock, Group Ⅶ was given DSS for 9 consecutive days. Mice in group Ⅷ received water containing DSS and were intragastrically administered 100 mg/kg kumatakenin (dissolved in sterile distilled water) for 9 consecutive days in the presence of Eno3 depletion. SASP is an anti-inflammatory drug used to treat IBD and was a positive control for the effects of kumatakenin on colitis.

Colon histology, morphology, and the severity of inflammation were assessed by colonoscopy, HE staining, and colon length sections. Disease activity index, rectal bleeding, diarrhea, and daily food intake/body weight were recorded.

Disease activity index (DAI) scores have historically correlated well with pathological findings in a DSS-induced model of IBD (Ghia et al., 2008). The DAI is the combined score of weight loss, stool consistency, and bleeding. Scores were defined as follows: body weight loss, 0, no loss; 1, 5%–10%; 2, 10%–15%; 3, 15%–20%; 4, 20%; stool consistency, 0, normal; 2, loose stool; 4, diarrhea; bleeding, 0, no blood; 2, presence of bleeding; and 4, gross bleeding.

An enzyme-linked immunosorbent assay (ELISA) was performed to measure the production of colonic mediators and cytokines. Harvested colon tissue specimens were cut into 5 mm pieces, processed, and mounted on slides for immunohistochemical analysis.

The viability of MODE-K cells was determined by measuring cellular metabolism using 96-well plates. A total of 1 × 10⁶ MODE-K cells was seeded into 96-well plates in 10% FBS medium per well and incubated overnight (37°C, 5% CO₂). Medium was replaced the next day with DMEM containing different concentrations of kumatakenin (10 or 80 μM) in the presence of different doses of erastin (0.1, 0.2, 0.5, 1, or 10 μM). After treatment for 48 h, cells were incubated with CCK8 solution for 4 h. Finally, absorbance was measured at 450 nm using a microplate reader.

Colonic tissue from mice in different groups was fixed with glutaraldehyde (2.5%), and then post-fixed in osmium tetroxide (2%) with potassium ferrocyanide (1.6%) in sodium cacodylaten (0.1 mol/L). Colonic tissue samples were then cut to 5 mm³ sections, stained with 2% uranyl acetate, dehydrated in increasing concentrations of acetone (50, 70, 80, and 90%) twice, and embedded in eponate. Sections were further cut into 80 nm sections, and stained with hematoxylin and eosin (HE), uranyl acetate (2%), and lead citrate. A Hitachi H7600 transmission electron microscope was used to capture images in the microscope core.

MODE-K cells in different groups were incubated with BODIPY 581/591 C11 (5 μM) at 37 °C for 30 min. Digested solutions were strained and pelleted to isolate cells. Cells were washed, resuspended in phosphate-buffered saline (PBS), transferred through a cell strainer for flow cytometry (The BD Accuri™ C6 Plus).

Total RNA from colonic tissues from the DSS control and DSS plus kumatakenin group were extracted using TRIzol reagent according to the manufacturer’s instructions. Using the protocol for the mRNA-Seq sample preparation kit, a final cDNA library was established. Mean insert size for the paired-end libraries was 150 bp. The RNA_seq data presented in this study have been deposited in Sequence Read Archive database with the accession number PRJNA913780 (https://dataview.ncbi.nlm.nih.gov/object/PRJNA913780?reviewer=idsh27k1p7lf43eqduqn2b0hsg).

To assess iron at the tissue level, colonic tissue (10 mg) from mice in different groups was isolated and washed with cold PBS. Homogenization was performed using a homogenizer in 4–10 volumes of Iron Assay Buffer using a Dounce on ice, with 10–15 passes. Tissue homogenates were centrifuged at 16,000 × g for 10 min to remove insoluble materials, and the supernatant was prepared for iron determination following the iron assay (Abcam, Cambridge, UK; ab83366).

To assess iron at the cellular level, 10⁶ cells were washed with PBS and lysed in 50 μL of lysis buffer containing 1% Triton X-100. Cell lysates were centrifuged at 15,000 rpm for 10 min, and the supernatants were used to measure total iron labile (Fe²⁺ and Fe³⁺) according to the manufacturer’s instructions. The Quantichrom iron assay (BioAssay Systems, Hayward, CA, United States of America; DIFE-250) was used for studying cellular iron levels at 590 nm.

MODE-K cells incubated with complete DMEM were treated with actinomycin D for 0, 3, or 6 h. Total RNA was collected with TRIzol reagent, and mRNA levels were analyzed by qPCR. qPCR reactions were performed in triplicate using Monet Mix and specific assay kits. Amplification was performed for 45 cycles with enzyme activation at 95°C for 30 min followed by denaturation at 60°C.

A total of 30 nmol/L of Eno3 and IRP1 siRNAs (GENERAL BIOL, Chuzhou, Anhui) was introduced into MODE-K cells by transient transfection with Lipo2000 according to the manufacturer’s instructions. Cells were collected 48 h after transfection and levels of Eno3 and IRP1 expression were measured.

The 3D structure of Eno3 (PDB ID: 2XSX) was downloaded from the Protein Data Bank. As previously described (He et al., 2020), the 3D structure of kumatakenin was initially constructed using Chem3D software (Cambridge Soft, Waltham, MA, United States of America), and optimized with Discovery Studio software (BIOVIA Inc., San Diego, CA, United States of America) at CHARMmforce field. Finally, the CDOCKER protocol of Discovery Studio was used to analyze the interaction of kumatakenin and Eno3.

Animal experiments, in vitro experiments, and data analysis were conducted according to a single-blind study design. One-way analysis of variance was used for multiple group comparisons; multiple comparison between groups was performed using the S-N-K method. Two-tailed unpaired Student’s t-tests were used to evaluate between two group comparisons. Data are expressed as means ± standard deviation (SD). The data were normally distributed and the groups had equal variances. All experiments were repeated at least three times. p-values less than 0.05 were considered statistically significant.

Cytotoxicity of kumatakenin in MODE-K cells was measured to verify the effective concentration of kumatakenin in vitro. Our results showed that kumatakenin exerted no significant cytotoxicity on MODE-K cells at the dose range of 5–80 μΜ (Supplementary Figure S1A). At 10–120 mg/kg, gastric administration of kumatakenin induced no significant alteration of body weight (Supplementary Figure S1B), disease activity index (Supplementary Figure S1C), colonic proinflammatory cytokine TNF-α production (Supplementary Figure S1D), serum FD-4 content (Supplementary Figure S1E), colonic iron level (Supplementary Figure S1F) or colonic expression of Eno3 (Supplementary Figure S1G) in normal mice.

As shown in Figures 1B, C, colonoscopy and HE staining showed mucosal bleeding, vascular patterns, infiltration of inflammatory cells, and atrophy of intestinal villi in colonic tissues from DSS-treated mice. Excessive intestinal inflammation was also observed after DSS treatment, which was characterized by short colon length (Figure 1C), DAI (Figure 2E), rectal bleeding (Figure 2F), stool consistency score (Figure 2G), and developed body weight/food intake loss (Figures 2H, I). These data confirmed successful establishment of a colitis model.

Oral administration of kumatakenin and SASP alleviated mice colitis symptoms, including restoration of mucosal bleeding, vascular patterns, infiltration of inflammatory cells, and atrophy of intestinal villi. Increased colon length, decreased DAI score, alleviation of diarrhea and rectal bleeding, and increased body weight/food intake was observed in kumatakenin-treated colitis mice. Overall, these data showed that kumatakenin significantly ameliorated colitis in a DSS-induced model.

Intestinal inflammation is related to increased neutrophil infiltration and accumulation of proinflammatory cytokines and mediators (Liu et al., 2022). Figure 2 shows that MPO activity (Figure 2A), pro-inflammation cytokine TNF-α/IL-6 (Figures 2B, C) production in colonic tissues, and FD-4 levels (Figure 2D) in serum samples from colitis mice were significantly higher than those from normal mice; these effects could be reversed by kumatakenin or SASP. These data suggested that kumatakenin reduced colonic inflammation and restored barrier function in DSS-treated mice.

Enhanced ferroptosis in colonic epithelial cells was observed in chemically induced colitis, and inhibition of ferroptosis has been shown to be beneficial for the treatment of IBD (Mayr et al., 2020). Thus, mitochondrial morphology, colonic iron level, and lipid peroxidation products malonaldehyde and 4-hydroxynonenal (HNE) contents were measured in colonic epithelial cells from colitis.

Transmission electron microscopy (TEM, Figure 3A) showed the presence of pyknosis of mitochondria in colonic epithelial cells from colitis mice. ELISA (Figure 3B) showed increased colonic iron levels in mice after DSS treatment. In addition, IHC and ELISA showed that increased expression of 4-HNE and higher MDA content were observed in DSS-induced colitis mice (Figures 3C, D). These results suggested that increased ferroptosis was determined in colonic epithelial cells in colitis mice. Kumatakenin administration exerted significant anti-ferroptosis effects in epithelial cells from colitis mice, including restoration of mitochondrial morphology, reduced iron level, and decreased expression of MDA and 4-HNE. The effects of kumatakenin on ferroptosis were also measured in vitro. CCK-8 results showed that erastin induced significant cell death; this effect was reversed by kumatakenin (Figure 3E). Flow cytometry results showed increased lipid peroxidation level in the erastin-treated group, which could also be reversed by kumatakenin (Figure 3F). Taken together, kumatakenin treatment exerted significant anti-ferroptosis effects in epithelial cells.

To measure the underlying mechanism of kumatakenin-exerted suppression of ferroptosis, RNA-seq was performed. Data showed that 797 genes were upregulated and 565 genes were downregulated significantly (P_adj < 0.05, |log₂foldchange|>1) in the kumatakenin-treated group compared with the DSS control group. The top 20 most differentially expressed genes between the DSS control and DSS + kumatakenin groups were chosen for further investigation (Figure 4A). The most enriched pathways of DEGs, including iron ion transport (GO: 0006826) and iron ion homeostasis (GO:0055072) pathways, as determined by Gene Ontology (GO) analysis, are shown in Figure 4B qPCR results further verified that kumatakenin significantly increased enolase (Eno3) expression in colitis mice (Figure 4C).

To further verify whether Eno3 was involved in kumatakenin-induced anti-ferroptosis effects, protein expression of Eno3 in kumatakenin-treated colitis mice was measured. Western blotting and IHC confirmed that Eno3 expression in colitis mice was significantly lower than the vehicle group; this effect was reversed by treatment with kumatakenin (Figures 4D,E).

Furthermore, the Eno3 inhibitor ENOblock significantly reversed kumatakenin-induced anti-colitis effects, including restoration of mucosal bleeding, vascular patterns and loose stools, restoration of intestinal villi, increased colon length, and decreased DAI score (Figures 5A–C). ENOblock treatment also reversed kumatakenin-induced suppression of ferroptosis, characterized by restoration of mitochondrial morphology and reduction of colonic iron levels in colitis mice (Figures 5D, E). These data showed that kumatakenin inhibited ferroptosis in epithelial cells from colitis mice in an Eno-3-dependent manner.

Small interfering mediated Eno-3 gene knockdown in MODE-K cells was performed to further measure the mechanism underlying kumatakenin-induced suppression of iron levels and ferroptosis in epithelial cells. Our results showed that kumatakenin reversed the increased iron levels and lipid peroxidation induced by erastin. Knockdown of Eno-3 significantly abolished kumatakenin-exerted suppression of iron content and lipid peroxidation level (Figures 6A–C), consistent with the in vivo results.

A previous study showed that upregulation of Eno family members led to inactivation of iron-regulatory protein 1 (IRP1), decreased iron levels, and suppression of ferroptosis (Zhang et al., 2022). ENO, as an RNA-binding protein, binds the IRP1-5′ untranslated region (UTR), promoting degradation of IRP1 mRNA (Zhang et al., 2022). Given that IRP1 serves as an important factor involved in iron homeostasis (Huynh et al., 2019), we further measured the effects of kumatakenin on mRNA stability and protein expression of IRP1 in colitis mice. Our results showed that mRNA and protein expression of IRP1 in colonic tissues was significantly increased in colitis mice (Figures 6D,E) compared with the vehicle group. Kumatakenin treatment inhibited the accumulation of IRP1 in the DSS-treated group. RNA stability assays showed that erastin treatment significantly inhibited IRP1 mRNA degradation in MODE-K cells (Figure 6F), which could be reversed by kumatakenin treatment.

To further confirm whether kumatakenin decreased IRP1 expression by regulation of Eno3, we measured the effects of kumatakenin on IRP1 in the erastin-treated group. Our results showed that IRP1 expression was increased after erastin treatment, and that this effect could be reversed by kumatakenin. Knockdown of Eno3 significantly blocked kumatakenin-exerted regulation of IRP1 (Figure 6G), indicating kumatakenin inhibited IRP1 expression by activating Eno3. Furthermore, knockdown of Eno3 significantly abolished kumatakenin-mediated regulation of iron and lipid reactive oxygen species (ROS) levels in epithelial cells; these effects were reversed by genetic knockdown of IRP1 (Figures 6H, I). Taken together, these data showed that kumatakenin inhibited cellular iron levels and suppressed ferroptosis in epithelial cells by regulating the Eno-3-IRP1 axis.

Molecular docking experiments were performed to investigate the activation mechanism between kumatakenin and Eno3. Figure 7 shows that kumatakenin can bind Eno3. Kumatakenin formed three hydrogen bonds with the residues in Eno3. Hydroxyl groups at C-1′, C-2″, and C-2″ formed hydrogen bonds with Ala46, Ser37, and Arg15, respectively. Thus, Ala46, Ser37, and Arg15 were identified as contact sites in Eno3 bound to kumatakenin.