The DC was synthesized and identified in our lab (purity > 98%). DC was dissolved in DMSO (Sigma-Aldrich, United States) to prepare a solution of 10 mM and stored at −20 °C. Rog (rosiglitazone) was purchased from Dalian Meiluo Biotechnology Co., Ltd. GW9662 were purchased from Selleck Chemicals (S2915). NCM460, HT-29 and HCT-116 cells were kindly provided by Stem Cell Bank, Chinese Academy of Sciences, and were cultured in McCoy’s 5A medium (12330031, Gbico) supplemented with 10% fetal bovine serum (16140071, Gbico) at 37 °C and 5% CO₂.

Cell viability was measured using the CCK-8 assay. Cells were seeded in 96-well plates at a density of 1 × 10⁴ cells/well for 24 h. After treatment with the indicated concentration of DC for 24 h, then 10 μL of CCK-8 solution was added, and the plate was incubated for 1 h at 37 °C. Absorbance was measured at 450 nm using a BioTek Spectrum spectrophotometer (Thermo Scientific, United States). The IC₅₀ value was calculated using GraphPad Prism7 software.

HT-29 and HCT-116 cells were planted on 6-well plates at a density of 1,000 cells/well for 24 h. The DC at different concentrations (0, 0.75, 1.5, and 3.0 μM) were added and cultured for 24 h. The culture medium was refreshed every 2 days to maintain growth for 10 days. The colonies were then washed with PBS, fixed with 4% paraformaldehyde, and stained with 0.1% crystal violet for 20 min at room temperature, followed by washing cells 3 times with PBS. Finally, the number of cell clones was counted using ImageJ software.

Apoptosis was quantified by using an apoptosis detection kit (BD Biosciences, Tokyo, Japan) according to the manufacturer’s instructions. Cells (2 × 10⁵/mL) were seeded into 6-well plates and were treated with DC (0, 0.75, 1.5, and 3.0 μM) for 24 h. After incubation, cells were washed twice with PBS before harvesting and re-suspended in binding buffer. Annexin V-FITC/PI staining was added to the cell suspension and incubated for 15 min in the dark at 37 °C. Cells were analyzed using an ACEN flow cytometer. Early and late apoptosis were summed to calculate the apoptotic rate.

Cell cycle was detected using a cell cycle analysis kit (BD Biosciences, United States) according to the manufacturer’s instructions. Cells (2 × 10⁵/mL) were seeded into 6-well plates and cultured for 24 h, and then were treated with the indicated DC concentration for 24 h. Then cells were washed twice with PBS, and then fixed in 70% ethanol for overnight at 4 °C. After that, cells were then stained with PI and analyzed cell cycle phase using an ACEN flow cytometer (ACEN, NovoCyte).

HT-29 cells (1 × 10⁴ cells/well) were co-transfected with 0.2 μg PPRE-TK-Luc reporter plasmid, and phRL-CMV Renilla luciferase in 24-well plates using Lipofectamine 2000 (Invitrogen, 11668019) for 24 h. Co-transfected cells were then treated with Rog or DC with or without GW9662 for 24 h. Cells were harvested, and luciferase activity was measured using the dual-Luciferase reporter assay system (Promega, E1980) according to the manufacturer’s instructions. Firefly luciferase values were divided by Renilla luciferase values to control for transfection efficiency.

Molecular docking was performed using AutoDock Vina 1.1.2 software (The Scripps Research Institute, La Jolla, CA, United States). Default settings and the Vina scoring function were employed. The crystal structure of PPARγ was obtained from the Protein Data Bank (PDB ID: 2PRG). Ligands and water molecules were removed from the crystal structures of the protein, and hydrogen atoms were added. Analysis and visual exploration of the ligand-protein interactions of the docking poses was performed using Discovery Studio 2020 software (Dassault Systems BIOVIA, San Diego, CA, United States, 2020).

Lipofectamine™ RNAiMAX (Invitrogen, 13778150) was purchased from Thermo-Fisher Scientific. The anti-PPARγ siRNA (si-PPARγ: 5-CCAAGUUUGAGUUUGCUGUdTd-3) (Li et al., 2014) and negative control siRNA (si-NC: 5-CCUAGUAUGACUAAGCUGUdTd-3) were designed and synthesized by GenePharma (Shanghai, China). HT-29 cells were transiently transfected for 48 h according to the manufacturer′s instructions (Lifetechnologies). After 48 h of transfection, the cells were examined and used for the subsequent assay.

Four-week-old male Balb/c nude mice were obtained from BiKai Biotechnology Co., Ltd. (Shanghai, China) and were housed and maintained under specific-pathogen-free (SPF) conditions in facilities approved by the Animal Ethics Committee of the Naval Medical University. All animal experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and approved by the Animal Ethics Committee of the Naval Medical University, China (Approval No. SMMU-2022-13). Each animal was injected subcutaneously with HT-29 cells (2 ×10⁶ cells/100 μL) into the right flanks of the nude mice. When the tumors reached a size of 100 mm³, the mice were randomly divided into 3 groups (6 mice/group) and treated with saline containing 0.9% sodium chloride (vehicle) or DC at doses of 1.5 mg/kg and 3.0 mg/kg by intraperitoneal injection once a day for 14 consecutive days. Tumor size was measured every other day using calipers, and tumor volume was calculated according the formula: tumor volume (mm³) = (tumor length) × (tumor width)²/2. Mice were sacrificed after 14 days of treatment. All procedures were conducted in accordance with the accepted guidelines for the use and care of laboratory animals. The tumors were harvested, photographed, weighed, and then stored at −80 °C for subsequent experiments.

Heart, liver, spleen, lung, kidney and tumor tissues from nude mice were fixed in 4% paraformaldehyde and embedded in paraffin. The paraffin-embedded tumor tissue sections (5 μm) were deparaffinized and rehydrated before staining with eosin and hematoxylin. The images were captured by using a light microscope (Leica, DMi8).

Deparaffinized and rehydrated tissue sections were permeabilized with 0.5% Triton X-100 and incubated with normal goat serum for 1 h at 37 °C. All washes between each step were performed with TBS. Detection of apoptotic cells was performed using the tunnel assay kit (Servicebio, G1501) and the proliferation marker was examined with the Ki67 antibody (Servicebio, GB121141) in tumor tissues according to the manufacturer’s instructions, and nuclei were counterstained with 5 μg/mL DAPI (Servicebio, G1012) for 5 min at 37 °C. Images were captured using a light microscope (Leica, DMi8).

Drug-treated cells and tumor tissues from xenograft mice were harvested and washed twice with ice-cold PBS. Cells and tumor tissues were lysed with RIPA buffer (Beyotime, P0013C) containing 1x protease inhibitors (Roche) and centrifuged (12,000 g for 15 min) at 4 °C. The supernatant was collected and the protein concentration was determined by the BCA assay (Beyotime, P0011). Equivalent amounts of protein (15–30 μg) were loaded and separated on 10% SDS-PAGE gels. After electrophoresis, protein bands were transferred to PVDF membranes (Bio-Rad) and blocked with 5% non-fat milk for 1 h at 37-°C. Bands were incubated overnight with an appropriate primary antibody at 4-°C. The primary antibodies were as follows: cleaved caspase 3 (CST, 9,661), Bax (Abcam, ab53154), Bcl-2 (Abcam, ab196495), CDK2 (CST, 18,048), Cyclin A2 (CST, 67,955), and β-actin (CST, 3,700). The next day, the membrane was washed three times with TBST for 5 min each and incubated again with the secondary antibody for 1 h at room temperature. β-Actin was also loaded as a control. Images were acquired from LI-COR (Lincoln, NE, United States). Band intensity was quantified using ImageJ software.

Data were calculated as the mean ± SD of at least three independent experiments. Statistical analysis was performed using the GraphPad Prism 7.0 software (San Diego, CA, United States). Student’s t-test was used to compare differences between two groups. Differences between multiple groups were analyzed by one-way ANOVA with Tukey’s post hoc test. Differences were considered to be significant when p < 0.05.

To investigate the antitumor effect of DC on human tumor cell lines. A cytotoxicity assay was conducted using CCK-8. In previous experiments, our results have shown that DC has the strongest cytotoxicity in HT-29 cells, followed by MCF-7 and DU145 tumor cells (Supplementary Figure S1). So, human colon cell lines were chosen as experimental cells. As shown in Figure 1B, DC inhibited the proliferation of HT-29 and HCT-116 colorectal tumor cells, as well as NCM460 normal colon mucosal cells, in a concentration-dependent manner, with the IC₅₀ values of 2.52 ± 1.34 μM, 1.48 ± 2.72 μM and 8.09 ± 1.67 μM, respectively, after incubation for 24 h. This result demonstrated that, compared to the colorectal tumor cells, DC exhibited lower cytotoxicity in the normal NCM-460 cells, with 3.21- to 5.47-fold higher viability in normal cells than in tumor cells, as shown in the dose-response curves. After DC treatment at 3.0 μM, the significant morphological changes were also observed, including the gradually shrinking cell bodies, larger gap, and small vesicles filled in the cytoplasm (Figure 1C). Additionally, to examine the ability of the cell lines to form a colony, HT-29 and HCT-116 cells were seeded and incubated with different concentrations of DC for 10 days. The results showed that colony formation of HT-29 and HCT-116 cells was suppressed by DC treatment in a dose-dependent manner. The rates of colony formation after DC (0, 0.75, 1.5 and 3.0 μM) treatment were 100.00% ± 3.55%, 44.16% ± 4.97%, 13.75% ± 1.54% and 8.95% ± 1.09% in HT-29 cells and were 100.00% ± 12.45%, 60.80% ± 8.11%, 9.51% ± 2.45% and 3.13% ± 0.85% in HCT-116 cells, respectively (Figure 1D). Collectively, DC exerted a potent inhibitory effect on the proliferation of human colon cell lines.

Induction of apoptosis in cancer cells is one of the major strategies for the development of antitumor drugs. To determine whether growth inhibition of DC was associated with cell apoptosis, HT-29 and HCT-116 cells were treated with DC (0, 0.75, 1.5, and 3.0 μM) for 24 h and then apoptosis rates were analyzed by flow cytometry. The results showed that DC obviously induced apoptosis rates to 52.67% ± 0.80% in HT-29 and 33.09% ± 1.09% in HCT-116 after 3.0 μM DC treatment, respectively (Figures 2A,B). The apoptotic induce effects were also verified by Western blotting, in which the protein expression levels of cleaved caspase 3 and Bax were elevated, while the protein level of Bcl-2 was significantly decreased following the indicated DC treatment (Figures 2C,D). These results revealed that DC could markedly induce colon tumor cell apoptosis.

Cell cycle dysregulation is crucial for the aberrant proliferation of tumor cells. To confirm the relationship between the growth inhibition of DC and cell cycle arrest, HT-29 and HCT-116 cells were treated with DC (0, 0.75, 1.5, and 3.0 μM) for 24 h and the cell percentage of each cycle phase was analyzed by flow cytometry. Our results showed that DC significantly induced cell cycle arrest at S phase, and the percentage in S phase was obviously higher in the 3.0 μM DC-treated group compared with the control group in HT-29 (46.53% ± 0.78 versus 37.39% ± 0.83) and HCT-116 (41.21% ± 1.25 versus 29.55% ± 0.29) cells, respectively. And compared with those of the control group, the percentage in G₀/G₁ phase was reduced in the HT-29 and HCT-116 cells, and the G₂/M phase was elevated in the HT-29 and HCT-116 cells. The results suggested that DC induced cell cycle arrest at the S phase in a dose-dependent manner (Figures 3A,B). Furthermore, the protein expressions of the crucial mitotic signaling pathway in S phase arrest (CDK2 and cyclin A2), which are involved in the progression from S to G2/M phase were measured. The results demonstrated that the expressions of CDK2 and cyclinA2 were significantly reduced by DC at 3.0 μM in HT-29 and HCT-116 cells, further confirming that DC could arrest the cell cycle at S phase in colon cells (Figures 3C,D).

To test whether the antitumor effect of DC could be mediated by PPARγ transcription, a cell highly expressing endogenous PPARγ was chosen from the HT-29 and HCT-116 cells. As shown in Figure 4A, HT-29 expressed a higher PPARγ protein than HCT-116 cells, thus being chosen for subsequent experiments. Moreover, the luciferase reporter was constructed to examine PPARγ transcriptional activity following DC or rosiglitazone (Rog, a known PPARγ agonist, positive control) treatment. The results indicated that treatment with DC after 24 h, the transcriptional level of PPARγ in HT-29 cells was significantly enhanced by 1.81-fold (P < 0.01) than the control cells. To confirm the above observation, a selective PPARγ antagonist GW9962 was co-cultured with DC in HT-29, the result exhibited that the DC activated transcriptional activity of PPARγ was markedly attenuated, suggesting that the antitumor mechanism of DC might be associated with PPARγ activation (Figure 4B).

Docking studies were employed to explore the possibility of DC binding to PPARγ and the potential binding mode between PPARγ (PDB ID: 2PRG) and DC using AutoDock Vina 1.1.2 software. The results indicated that DC can tightly occupy the ATP binding site of PPARγ as shown in Figures 4C–F, with the score of −7.9 kcal/mol. The ester carbonyl group and ketone carbonyls of the five-membered ring of DC can strongly interact with the residues Arg²⁸⁸ and Ser³⁴² in the hinge domain of PPARγ via two hydrogen bonds, respectively. Additionally, the seven-membered ring, double bond and methyl of DC can form hydrophobic interaction with Cys²⁸⁵, Leu³³⁰, Leu³³³, and Met³⁶⁴ residues.

To verify the role of PPARγ in DC-mediated proliferation inhibition, HT-29 cells were co-treated with DC and GW9662 (10.0 μM, a PPARγ antagonist). As shown in Figures 5A,B, compared to DC-treated group, the inhibitory effect of co-treatment was markedly reduced, suggesting that GW9662 blocked the growth inhibition of DC in HT-29 cells. Similarly, the co-treatment of DC and GW9662 could increase the number of HT-29 cell clones, indicating that PPARγ antagonist could abrogate the inhibition of DC on HT-29 cell colony formation (Figures 5C,D).

Previous experiments have revealed that PPARγ activation could induce cell apoptosis. To explore whether PPARγ activation was involved in the DC-induced apoptosis of HT-29 cells, the apoptosis rate was tested following co-treatment. Compared to the DC-treated group, the apoptotic rates of HT-29 cells co-treated with DC and GW9662 were sharply decreased from 35.59% ± 2.83% to 23.06% ± 1.60%, indicating that GW9662 treatment attenuated DC-induced apoptosis of HT-29 cells (Figures 5E,F). In the cell cycle assays, the cell proportion at S phase was obviously reduced from 42.89% ± 1.11% (3.0 μM DC) to 38.67% ± 1.43% (3.0 μM DC + GW9662) and that at G2/M phase from 25.75% ± 0.37% (3.0 μM DC) to 22.85% ± 0.55% (3.0 μM DC + GW9662), while the proportion at G0/G1 was markedly increased from 31.36% ± 0.90% to 38.48% ± 1.38%, suggesting that GW9662 treatment abrogated DC-induced S phase arrest of HT-29 cells (Figures 5G,H). Consistent with the results from the flow cytometry assay, Western blot showed that the protein levels of cleaved caspase-3 and Bax were decreased and those of Bcl-2, CDK2 and cyclin A2 were increased in HT-29 cells after co-treatment of DC and GW9662 (Figures 5I,J), suggesting that GW9662 treatment decreased DC-induced apoptosis and cell cycle-related protein expression in HT-29 cells.

To further confirm the role of PPARγ activation in the antitumor activity of DC, a si-PPARγ HT-29 cell line was constructed using RNA interference technology. The knockdown efficiency of PPARγ in HT-29 cells was examined using WB in si-NC HT-29 and PPARγ siRNA HT-29 cells. The results indicated that the endogenous PPARγ protein expression in si-PPARγ HT-29 cells were efficiently weakened compared to the control cells (Figures 6A,B). The si-NC HT-29 and si-PPARγ HT-29 cells were examined for cell viability, cell apoptosis, and cell cycle arrest by DC treatment.

Cell viability assay showed that the anti-proliferation effect of DC was markedly reduced in si-PPARγ HT-29 cells (Figure 6C), suggesting that the knockdown of PPARγ protein could reverse DC-mediated growth inhibition in HT-29 cells.

The apoptosis induction of DC against PPARγ siRNA HT-29 cells was examined by flow cytometry. As shown in Figures 6D,E, after DC treatment at the same concentrations, the apoptosis rate of si-PPARγ HT-29 cells (21.68% ± 1.69%) were markedly lower than that of si-NC HT-29 cells (33.24% ± 0.83%). In the cell cycle assay, compared to si-NC HT-29 cells, the proportion of PPARγ siRNA HT-29 cells at S phase were significantly reduced from 34.38% ± 1.78% to 28.8% ± 0.45%, while the proportion at G2/M was markedly increased from 22.73% ± 0.86% to 30.96% ± 1.37% after 3.0 μM DC treatment. The proportion of G0/G1 phase showed no obvious change (Figures 6F,G). This result suggested that the knockdown of PPARγ was able to reverse DC-mediated cell cycle arrest in HT-29 cells. Western blot showed that the protein levels of cleaved caspase-3 and Bax were decreased and the Bcl-2, CDK2, and cyclin A2 were increased in si-PPARγ HT-29 when compared with those of si-NC HT-29 group (Figures 6H,I), revealing that the knockdown of PPARγ could reverse DC-mediated apoptosis and cyclin-related protein expressions in HT-29 cells.

The anti-tumor efficacy of DC was further examined using a HT-29 cell xenograft mouse model. When the tumor volume reached 100 mm³, the 0.9% sodium chloride or DC at doses of 1.5 mg/kg and 3.0 mg/kg was administered by intraperitoneal injection daily for 14 days. Mice were weighed, and tumor volumes were measured at the beginning of each treatment. The results revealed that administration of different doses of DC did not change the body weight of mice compared to the control group (Figure 7A). The tumor volume and weight were approximately 709.6 mm³ and 1.04 g in the control group, while the tumor volume and weight significantly decreased to 402.4 mm³, 315.6 mm³ and 0.78 g, 0.61 g following 1.5 mg/kg, 3.0 mg/kg DC treatment after 14 days (Figures 7B–D), suggesting that DC could suppress colon cancer growth in vivo.

HE, Tunel, and ki67 analyses were performed to observe the morphological changes of tumors. HE staining revealed large areas of necrosis in the tumor tissues of the DC-treated group, while no necrosis or only mild necrosis was observed in the control group. Tunel and ki67 experiments indicated that Tunel-positive cells (green fluorescence) were increased, and ki67-positive cells indicate proliferation was reduced after the mice were treated with 1.5 and 3.0 mg/kg DC in the tumor tissues (Figures 7E–G), respectively. The expression levels of apoptotic proteins were detected by Western blot. Consistent with the in vitro results, the protein expression levels of cleaved caspase 3 and Bax were significantly elevated and the Bcl-2 was reduced in the tumors after the indicated DC treatment (Figures 7H,I). These results showed that the number of apoptotic cells dramatically increased in the DC-treated group but not in the control group. In addition, no significant changes were observed in the histological morphology of the heart, liver, spleen, lung, and kidney of DC-treated mice (Figure 7J), indicating that DC was not obviously toxic to normal tissues in vivo.