Lumefantrine was purchased from Aladdin (Shanghai, China), Camptothecin (CPT) from Macklin (Shanghai, China), YK-4-279 from MedChemExpress (Shanghai, China), dextran sulfate sodium (DSS) from MP Biomedicals (Shanghai, China). The primary antibodies used were: Claudin-1 (Abcam, ab15098, Cambridge, United Kingdom), ZO-1 (CST, 8193P, Massachusetts, United States), p-p65 (Proteintech, 82335-1-RR, Wuhan, China), p65 (CST, 8242S, Massachusetts, United States), p-IκB (Proteintech, 82349-1-RR, Wuhan, China), IκB (Proteintech, 66418-1-Ig, Wuhan, China), β-Actin (Proteintech, 66009-1-Ig, Wuhan, China), and FLI-1 (Santa Cruz, sc-365294, Shanghai, China). HRP-conjugated secondary antibodies anti-mouse/rabbit IgG were purchased from ZSGB-Bio (Beijing, China) for IHC, and CST (Massachusetts, United States) for WB.

Female C57BL/6J mice were purchased from Chengdu Dossy experimental animals Co., Ltd. All procedures were approved by the Institutional Animal Care and Use Committee of Chengdu Medical College. 8-week-old (19–21 g) mice were housed under SPF conditions (22°C ± 1°C, 55% ± 5% humidity, 12-h light/dark cycle) with a 7-day acclimatization period prior to experimentation. And then 18 mice were randomly assigned to three experimental groups: (1) control group receiving standard drinking water, (2) DSS-induced colitis group administered 3% dextran sulfate sodium (DSS) in drinking water, and (3) treatment group receiving both DSS and daily oral gavage of Lumefantrine (20 mg/kg body weight) administered as a suspension in sterile distilled water. Throughout 7-day induction and treatment phase, clinical parameters - weight loss, stool consistency, and rectal bleeding - were recorded daily and quantified via standardized scoring criteria. Terminal procedures were performed on day 7 under ether anesthesia. Blood samples collected by eyeball excision were centrifuged at 3000 rpm for 10 min to isolate serum, which was subsequently stored at −80°C. Colons were excised from the ileocecal junction to the anal verge for precise length measurement, followed by systematic tissue sampling - distal colon segments were snap-frozen in liquid nitrogen for molecular analyses while middle colon sections were processed for histological evaluation.

Disease progression was quantified using a validated Disease Activity Index (DAI) scoring system (Mukherjee et al., 2024), which evaluates three parameters: percentage weight loss, fecal bleeding severity, and stool consistency. The composite DAI score represents the arithmetic mean of these three components, with higher scores indicating more severe colitis. Detailed scoring criteria are presented in Supplementary Table S1.

Mid-colon segments were fixed in 4% paraformaldehyde, paraffin-embedded, and sectioned at 5 μm thickness for hematoxylin and eosin (H&E) staining using stain kit (Solarbio, G1120, Beijing, China). Histopathological assessment was performed by a board-certified pathologist using the standardized criteria outlined in Supplementary Table S2.

Immunohistochemistry was performed as previously described. Briefly, antigen retrieval was conducted using citrate buffer (pH 6.0) at 95°C for 20 min. Endogenous peroxidase activity was quenched with 3% H₂O₂ for 15 min. Sections were incubated with primary antibodies at 4°C overnight. Sequential incubation with horseradish peroxidase (HRP)-conjugated secondary antibodies (anti-mouse/rabbit IgG; 1:200 dilution) was performed at 37°C for 30 min. Diaminobenzidine (DAB) (Beijing ZSGB-BIO, Beijing, China) chromogen was used for signal development, with hematoxylin counterstaining.

Serum levels of IL-1β, IL-6, and TNF-α were measured using commercial ELISA kits (Cusabio, Wuhan, China) according to the manufacturer’s protocol. Optical density was measured at 450 nm using a microplate reader (Molecular Devices, United States).

Total RNA was extracted from colon tissues using TRIzol reagent (Tiangen Biotech, Beijing, China), followed by cDNA synthesis with Reverse Transcriptase (Vazyme Biotech, Nanjing, China). Primers were synthesized by Tsingke Biotech Co., Ltd. (Beijing, China), and the sequences are listed in Supplementary Table S3. Quantitative PCR amplification was conducted using the following protocol: initial denaturation at 95°C for 15 min, followed by 40 cycles of 95°C for 10 s (denaturation) and 60°C for 30 s (annealing/extension), according to the manufacturer’s instructions for SYBR Green I master mix (Roche Diagnostics, Branchburg, United States). β-Actin served as the endogenous control. Relative gene expression was calculated via 2^−ΔΔCt method.

Computational docking was performed to predict Lumefantrine-FLI-1 interactions. The FLI-1 tertiary structure was modeled using AlphaFold2. The PDB file of FLI-1 was uploaded to the PlayMolecule platform (https://playmolecule.com/) to predict potential ligand-binding pockets. Docking simulations were performed using SailVian, and binding stability was assessed through 100 ns molecular dynamics simulations.

Cell lysates were incubated with epoxy-activated μSphere magnetic beads (Tianyan Biotech, Wuxi, China) pre-conjugated with 20 μM Lumefantrine (4°C, 12 h rotation). After washing (3× PBS), bound proteins were eluted (0.5 mol/L NaCl) and analyzed by Western blotting. Control groups received ethanolamine-blocked beads without compound.

The Chinese Academy of Sciences Cellular Bank (Shanghai, China) provided IEC-6 and NCM460 cell lines. Both cell lines were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS), cultured in a humidified incubator (37°C, 5% CO₂, 95% air). Cells were subcultured at 80%–90% confluence using 0.25% trypsin-EDTA.

Cells were seeded in 96-well plates at 5,000-10,000 cells/well and allowed to adhere for 24 h. Following treatment with Lumefantrine (0–40 μM) for 24/48 h, 10 μL CCK-8 reagent (Labgic, Beijing, China) was added to each well. After 2–4 h incubation protected from light, absorbance was measured at 450 nm using a microplate reader (Molecular Devices, United States). Cell viability (%) was calculated.

IEC-6 was treated with 20 μg/mL LPS for 48 h, and then treated with Lumefantrine, Camptothecin (CPT) or YK-4-279 for another 24 h. Treated cells were lysed in RIPA buffer containing protease/phosphatase inhibitors (Roche, Penzberg, Germany). Lysates were centrifuged, and supernatants were quantified via BCA assay (Thermo Fisher Scientific, United States). Proteins with equal amounts (20 μL) were electrophoretic separated on 10% SDS-PAGE gels and transferred to PVDF membranes (Merck Millipore, Darmstadt, Germany). After blocking with 5% non-fat milk/TBST (1 h, RT), membranes were incubated with primary antibodies (1:1000) overnight at 4°C, followed by HRP-conjugated secondary antibodies (anti-mouse/rabbit IgG, 1:5,000) for 2 h at RT. Signals were developed using ECL Prime (Labgic, Beijing, China) and quantified via ImageJ software.

All experiments were repeated independently at least three times. All data are represened as mean values ±standard deviation (S.D.). GraphPad Prism 9.0 performed unpaired t-tests. Significance thresholds: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns = not significant (p ≥ 0.05).

Daily monitoring revealed significant weight loss in DSS-treated mice compared to controls (Figure 1A). Lumefantrine administration markedly attenuated weight reduction. Disease activity index (DAI) scores progressively increased in model mice but were suppressed by Lumefantrine treatment (Figure 1B). Colon shortening induced by DSS was partially reversed in the treatment group (Figure 1C). Histopathological analysis demonstrated severe mucosal erosion and inflammatory infiltration in DSS mice, whereas Lumefantrine preserved epithelial continuity and reduced crypt damage (Figure 1D). Correspondingly, immunohistochemical and Western blot analyses demonstrated that DSS challenge significantly reduced colonic Claudin-1 and ZO-1 expression, indicative of impaired epithelial barrier integrity (Figures 1E,F). Lumefantrine administration restored these tight junction proteins, correlating with preserved mucosal architecture.

DSS challenge significantly upregulated colonic mRNA expression of pro-inflammatory cytokines IL-1β, IL-6, Tnf-α, along with inflammatory mediators Cox-2 and iNos. Lumefantrine administration markedly attenuated these elevations (Figures 2A–E). Serum ELISA quantification revealed corresponding decreases in protein concentrations: IL-1β, IL-6 and TNF-α all declined following Lumefantrine intervention (Figures 2F–H).

The three-dimensional structure of FLI-1 predicted via the UniProt database (Figure 3A) and Lumefantrine (Figure 3B) were subjected to molecular docking. The 2D interaction diagram (Figure 3C) revealed that Lumefantrine binds to FLI-1 through hydrogen bonds and van der Waals forces, with the strongest binding affinity observed at arginine 123 (Arg123) (ΔEtotal = −167.90 kcal/mol). PlayMolecule combined with SailVian predicted three potential binding sites for Lumefantrine on FLI-1 (Figures 3D–F). The optimal docking pose exhibited a binding score of −7.9, comparable to positive control molecules TK216 (−7.9) and YK-4-279 (−7.6).

Molecular dynamics simulations revealed stable complex formation (Figure 3G): Root mean square deviation (RMSD) stabilized at 2.0–2.2 nm after 20 ns equilibration (Figure 3H). Root mean square fluctuation (RMSF) analysis indicates residue 20-60/220-240 flexibility, while active site conformation preserved (Figure 3I). The radius of gyration (Rg) of the protein stabilized between 2.4–2.8 nm, reflecting stable folding influenced by secondary structure (Figure 3J). Decreased solvent-accessible surface area (SASA) indicates a tightly packed, stable complex (Figure 3K). Total binding energy was calculated to be −45.25 kcal/mol, dominated by nonpolar energy (−38.7 kcal/mol).

To experimentally validate the Lumefantrine-FLI-1 interaction predicted by computational analyses, we performed affinity-based pull-down assays. FLI-1 was selectively enriched in Lumefantrine-conjugated μSphere bead lysates compared to GST-tagged negative controls (Figures 4A,B), confirming direct binding between Lumefantrine and FLI-1. Western blot analysis further revealed FLI-1 suppression by Lumefantrine, comparable to the inhibitory effects of Camptothecin (CPT), a known FLI-1 antagonist (Figure 4C).

CCK-8 assays demonstrated negligible cytotoxicity at concentrations below 40 μM during 24-h exposure in intestinal epithelial cells IEC-6 and NCM460 (Figures 5A,B). Thus, 20 μM Lumefantrine was employed for subsequent mechanistic investigations. Western blot analysis demonstrated that Lumefantrine significantly suppressed LPS-induced phosphorylation of both p65 and IκB (Figure 5C). Notably, Lumefantrine exhibited NF-κB inhibitory efficacy comparable to established FLI-1 inhibitors YK-4-279 and CPT, which served as positive controls in this study. Critically, when FLI-1 was inhibited by CPT, Lumefantrine’s suppression of p-p65 and p-IκBα was abolished (Figure 5D), confirming that its NF-κB inhibition operates specifically through FLI-1 targeting.

Building upon FLI-1’s regulatory role in NF-κB signaling, we explored its therapeutic relevance in IBD. Analysis of public transcriptome datasets revealed significantly elevated FLI-1 expression in intestinal tissues from IBD patients compared to healthy controls (Figure 6A). Consistently, our DSS-induced colitis model showed increased colonic FLI-1 expression, which was effectively suppressed by Lumefantrine treatment (Figure 6B). These results establish FLI-1 as a promising therapeutic target in IBD, with Lumefantrine exerting its anti-inflammatory effects through direct FLI-1 binding and subsequent suppression of the NF-κB signaling cascade.