Human AML cells: KG-1(No.CCL-246.1), HL-60 (No.CRL-3306), KASUMI-1 (No.CRL-2724), KASUMI-6 (No.CRL-2775), murine AML cell line: C1498 (No.TIB-49), all purchased from American Type Culture Collection (ATCC) Cell Bank, United States. SPF NCG mice, 4-5 weeks old, weighing 16–18 g, were purchased from Beijing Vital River Laboratory Animal Co., Ltd. Animal production license number: SCXK (Beijing) 2021-0011.

Reagents and drugs: fetal bovine serum (No.A5669701), RPMI1640 medium (No.11875119) were purchased from Gibco, United States; penicillin-streptomycin solution (No.C0222), cell counting kit-8 (No.C0038) and BCA protein concentration quantitative kit (No.P0010S) were purchased from Shanghai Beyotime Biotechnology. Hematoxylin-eosin (HE) staining kit (No. G1120) was purchased from Soleibao Biotechnology Company, Beijing, China. 5-bromo-2-deoxyuracil (EDU) kit (No. KGA9602-100), human interleukin-6 (IL-6), IL-1β, Tumor necrosis factor -α (TNF-α) enzyme-linked immunosorbent assay (ELISA) kit (No. KGC1111-48, KGC1103-48, KGC1122-48), flow cytometry Annexin-v-FITC kit (No. KGA1101-100) were purchased from China Jiangsu Keygen Biological Co., Ltd.

Mouse hemoglobin (Hb) ELISA kit (No. LE-M1587) was purchased from Hefei Lyle Biotechnology Co., Ltd., China, and mouse erythropoietin (Epo) ELISA kit (No.XG-E989530) was purchased from Shanghai Xige Biotechnology Co., Ltd., China. human B-cell lymphoma-2 (BCL2) antibody (No.12789-1-AP), human Bcl-2 Associated X protein (BAX) antibody (No. 50599-2-Ig), human Poly ADP-ribose polymerase (PARP) antibody (No.13371-1-AP), ATG5 Monoclonal antibody (No. 66744-1-Ig) and human Cleaved caspase3 antibody (No.19677-1-AP) were purchased from Wuhan Proteintech Biological Co., Ltd. Polydatin (No.HY-N0120A) was purchased from MCE Biological Co., Ltd. The purity of Polydatin (batch number: HY-N0120A-29) was 99.57%.

Instruments: DILITCEN22 desktop centrifuge, Suzhou Beirui Instrument Co., Ltd., China; hBS-ScanX full wavelength microplate reader, Nanjing Detieer Experimental Equipment Co., Ltd., China; dYJ-905 inverted metallographic microscope, Shanghai Dianying Optical Instrument Co., Ltd., China; Invitrogen iBright all-round gel imager, Thermo Fisher Science and Technology Co., Ltd., United States; FACSCanto II flow cytometer, BD Biomedical, United States; 1290 LC-MS, 1290 UHPLC with 6230TOF MS System, Agilent, United States.

The cells were cultured in RPMI1640 medium supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin at 37°C with 5% CO₂. Upon reaching a cell density exceeding 80%, the cells were harvested by centrifugation and passaged at a ratio of 1:3.

KG-1, HL-60, C1498, KASUMI-1, and KASUMI-6 cells were plated in 96-well plates at a density of 8 × 10³ cells per well. PD stock solution was prepared by dissolving in dimethyl sulfoxide (DMSO) according to the manufacturer’s instruction. Following a 12-hour incubation period to allow for cell adherence, the PD stock solution was diluted with culture medium to achieve working concentrations of 1, 5, 10, 20, 40, and 100 μM. The cells were then exposed to these varying concentrations of PD for 48 h (Li et al., 2017). Subsequently, 10 μL of cell counting kit 8 (CCK-8) solution was added to each well, followed by a 2-hour incubation in darkness. Absorbance at a wavelength of 450 nm was measured using an enzyme labeling instrument, and the IC50 value was calculated using GraphPad Prism software.

The KG-1, HL-60, C1498, KASUMI-1, and KASUMI-6 cells were seeded in 96-well plates at a density of 3 × 10³ per well. After 12 h of adherence, the cells were treated with PD (0, 10, 20, and 40 μM), respectively, with 5 replicate wells in each group. At 24, 48, 72, and 96 h after drug treatment, 10 μL of CCK-8 solution was added to each well, which was then incubated in the dark for 2 h. The absorbance was then measured at a wavelength of 450 nm. The cell proliferation rate was calculated according to the following formula: The cell proliferation rate is calculated as follows: (Absorbance value of experimental group - Absorbance value of blank well)/(Absorbance value of control group − Absorbance value of blank well) × 100%.

KG-1, HL-60, C1498, KASUMI-1 and KASUMI-6 were seeded in 6-well plates at a density of 1 × 10⁶ per well. After adherent for 12 h, the cells were treated with PD (0, 10, 20, and 40 μM), respectively. After 48 h, the total RNA was extracted with Trizol. cDNA was synthesized using a reverse transcription kit according to the manufacturer’s instructions. RT-qPCR was then conducted with a SYBR Prime Script RTPCR kit utilizing custom primers (in Supplementary Table S1). The relative expression levels of mRNA were quantified using the 2^{∧ −ΔΔCt} method.

KASUMI-1 cells were seeded in 96-well plates (or 6-well plates) at a density of 3 × 10³ (or 1 × 10⁶) per well. After 12 h of attachment, the cells were treated with PD (0, 10, 20, and 40 μM) for 48 h, respectively. After treatment, the relevant experimental tests were performed. The CCK-8 experimental steps were performed according to the above description.

The EDU assay was carried out according to the protocol from the EDU kit manual after treatment. In summary, the EDU reagent was diluted with culture medium to create a working solution, and then was added to the cells for incubation for 6 h. After the incubation, 4% paraformaldehyde solution was added to fix the cells, and 1% Triton X-100 solution was used for membrane penetration. The Click-iT EdU reaction reagent was prepared according to the EDU kit manual and added to the cell culture plate for further incubation for 30 min. Finally, the cells were observed under a fluorescence inverted microscope, and photographs were taken for record.

Following drug treatment, the procedures outlined in the Annexin V/PI Apoptosis Detection Kit manual were adhered to. This included the collection and washing of cells and supernatant with PBS solution three times, followed by the addition of 5 μL of Annexin V solution and 5 μL of PI solution. Subsequently, the samples were thoroughly mixed and incubated at room temperature in the absence of light for 15–30 min. The samples were then analyzed using appropriate equipment, and the resulting data were processed utilizing Flow jo10.0 software.

The total protein of the cells was extracted and quantified using a BCA kit after treatment. Subsequently, according to the sample order, 40 μg samples were fractionated using 10% SDS-polyacrylamide gel electrophoresis (PAGE) and subsequently transferred onto a PVDF membrane. Following blocking with 5% non-fat milk, the membrane was incubated overnight at 4°C with the respective primary antibody solution, which was diluted at a ratio of 1:1000. This was followed by a 5-hour incubation with the appropriate secondary HRP-conjugated antibodies (1:8000). Ultimately, the protein signal was visualized using an ECL detection kit, and the strip’s gray value was analyzed using ImageJ software.

After the cells were treated with drugs, the operation was performed according to the instructions of the ELISA detection kit. In short, the cell supernatant solution was collected and added to the pre-coated well plate. After the incubation, the antibody labeled with the enzyme was washed and added to continue the incubation with the specific antibody. The color reaction was performed, and the absorbance was measured by a microplate reader.

In terms of mechanism analysis, the cells were divided into 0 μM group, 20 μM PD group, ATG5 autophagy inhibitor (agent-82,10 μM) group, and 20 μM PD + ATG5 autophagy inhibitor combination group (agent-82,10 μM) group. After 48 h of treatment, the cells were tested according to the CCK-8, EDU, ELISA, Western blot and flow cytometry operation steps described above.

NCG mice used to construct KASUMI-1 subcutaneous transplantation tumor model. All mice were housed in a sterile mouse house environment, alternating day and night every 12 h, and were given adequate food and drinking water. After 1 week of adaptation, the AML model was established by subcutaneous injection of KASUMI-1 cells (2 × 10⁷) in the right limb of mice. When the tumor volume reached 100 mm³, the mice were treated with drugs. PD stock solution was dissolved in 0.9% saline solution to prepare the treatment solutions. The mice in each experimental group received the PD solution via intragastric gavage once daily at dosages of 50 mg/kg, 100 mg/kg, and 200 mg/kg, respectively. The control group was given an equal amount of DMSO diluted in saline solution (Zhang et al., 2019; Li et al., 2023). During the treatment, the tumor volume and the body weight of the mice were measured every 3 days. The tumor volume was calculated according to the formula V = L × W2/2, where L represents the tumor length and W represents the tumor width. After 4 weeks, peripheral blood was extracted from the tail vein of mice, and 2 mL was mixed with 38 mL Turk blood diluent, and then the white blood cell count (WBC) was observed under a microscope. The remaining blood was centrifuged at 12,000 rpm for 5 min, and the supernatant was collected. The expression levels of hemoglobin (Hb) and erythropoietin (Epo) were detected using a whole blood automatic analyzer. After that, the mice were euthanized, and the lung, kidney, spleen and liver tissues of the mice were collected, and then the tissue samples were stained with hematoxylin and eosin. All experimental animals were euthanized for cervical dislocation with a high concentration of CO₂, and the animal experiments were approved by the Ethics Committee for the Institutional Animal Care and Ethics Committee of The Second Affiliated Hospital, University of South China (Animal Ethics 230319).

After the lung, kidney, spleen and liver tissues of mice were fixed and embedded, they were cut into 6 μm slices by frozen section machine. The operation was carried out according to the instructions of HE staining kit, and the pathological structure of the tissues was analyzed by taking photos under the microscope.

The three-dimensional structure of PD and ATG5 protein was obtained by PubChem. The active binding sites of PD and ATG5 were defined by AutoDock software, and the network was generated.

The real-time drug concentration of PD in the blood of healthy mice was analyzed by high-performance liquid chromatography (HPLC), with six mice in each group undergoing alternate blood collection. The blood was collected at the following time points: 0 min, 5 min, 15 min, 30 min, 1 h, 2 h, 4 h, 8 h, 16 h, and 24 h. The following parameters were used for the determination: flow rate: The flow rate was set at 0.1 mL/min, the temperature at 25°C, the column at Bonus-PR, and the mobile phase comprised 20 mM sodium phosphate buffer (pH2.5)-acetonitrile (90:10).

All experiments were repeated more than 3 times, and data analysis was performed using GraphPad Prism (version 9.0). All results are presented as mean ± standard error (SEM). The difference between the two groups was performed using a 2-tailed Student's t-test. One-way analysis of variance (ANOVA) was used for comparison between multiple groups. P < 0.05 was considered significant.

The study investigated the drug sensitivity and cytotoxic effects of PD on acute myeloid leukemia (AML) cell lines. The IC50 values of PD for KG-1, HL-60, C1498, KASUMI-1 and KASUMI-6 cells were determined to be 33.21, 32.39, 30.09, 25.88, and 35.02, respectively (Figure 1A). PD exhibited a reduction in cell viability in AML cells, with the KASUMI-1 cell line demonstrating the lowest proliferative viability and highest drug sensitivity (Figure 1B). Furthermore, RT-qPCR analysis revealed a significant upregulation of ATG5 expression in KASUMI-1 cells (Figure 1C). This resulted in the choice of the KASUMI-1 cell line for subsequent experimental investigations.

CCK-8 assay demonstrated that PD reduced KASUMI-1 cell viability in a dose-dependent manner at different times (Figure 2A). And the analysis of EDU assay showed that the red fluorescence intensity of the cells was significantly lower than that of the control group with the increase of PD concentration (Figures 2B, C), indicating that PD significantly inhibited KASUMI-1 cell proliferation. Additionally, the levels of IL-6, TNF-α, and IL-1β expression were assessed following treatment with PD. The results demonstrated a clear dose-dependent reduction in the expression of IL-6, TNF-α, and IL-1β (Figures 2D–F). These findings suggest that PD has the potential to effectively suppress the proliferation and inflammatory factor expression of AML cells in an in vitro setting.

To assess the impact of PD on the apoptosis of AML cells in an in vitro setting, flow cytometry and western blot experiments were conducted subsequent to PD administration. The findings indicated a notable increase in the rate of apoptosis in KASUMI-1 cells following PD treatment, as determined through flow cytometry analysis, in a manner that was dependent on the dosage administered (Figures 3A, B). Consistently, western blot analysis demonstrated that PD incubation decreased the expression of the anti-apoptotic protein Bcl-2, while increasing the expression levels of the pro-apoptotic signals Bax, cleaved caspase-3, and cleaved PARP (Figure 3C).

The Autodock software was utilized to predict the binding activity between PD and ATG5. The analysis revealed multiple amino acid binding sites and binding energies below −5 kJ/mol (Figure 4A), suggesting a strong binding affinity between PD and ATG5. Furthermore, the results indicated that ATG5 may serve as a potential target for PD. Subsequent western blot analysis demonstrated a dose-dependent induction of autophagy in KASUMI-1 cells following PD treatment, as evidenced by the upregulation of ATG5, ATG7, Beclin1, and Microtubule-associated protein 1A/1B-light chain 3 (LC3) (Figures 4B, C). Various concentrations of PD have been found to have a notable impact on the ubiquitination process of ATG5, with a concentration of 40 μM demonstrating a significant reduction in the ubiquitination level of ATG5 (Figure 4D). The collective findings suggest that PD plays a role in decreasing the ubiquitination level of ATG5, consequently enhancing the protein stability of ATG5, increasing the expression of ATG5, and facilitating the initiation of autophagy (Figure 4E).

To elucidate the precise mechanism by which PD acts against AML tumors in vitro, functional recovery experiments were conducted utilizing ATG5 inhibitors. The suppressive impact of PD on EDU positivity in KASUMI-1 cells was attenuated when co-administered with an ATG5 inhibitor (agent-82) (Figures 5A, B). ELISA analysis revealed that the combined treatment of agent-82 and PD reversed the inhibitory effects of PD on the expression of IL-6, IL-1β, and TNF-α (Figure 5C). Flow cytometry results indicated that the co-treatment of agent-82 and PD led to a significant decrease in the apoptosis rate of KASUMI-1 cells compared to those treated with PD alone (Figures 5D, E). Additionally, the combined treatment reversed the upregulation of pro-apoptotic proteins Bax, cleaved caspase3, and PARP, as well as the downregulation of the anti-apoptotic protein Bcl2 induced by PD treatment alone (Figures 5F, G).

Furthermore, we conducted additional experiments to determine if agent-82 could reverse the autophagy-inducing effects of PD on KASUMI-1 cells. Western blot analysis revealed a significant decrease in the expression of ATG5 following treatment with agent-82, as well as a decrease in the expression of other autophagy-related proteins including ATG7, Beclin1, and LC3 when compared to the control group (Figures 6A, B). Interestingly, when PD was combined with agent-82 treatment, the expression of ATG5, ATG7, Beclin1, and LC3 was further suppressed (Figures 6B–E). The aforementioned findings indicate that PD may potentially act as an anti-AML agent through the modulation of the ATG5-mediated autophagy pathway.

The impact of PD on acute myeloid leukemia (AML) was further explored through in vivo studies utilizing NCG mice xenografts harboring KASUMI-1 cells. Analysis of Figures 7A–C indicated a significant reduction in tumor growth, volume, and weight following PD treatment. Kaplan-Meier survival analysis demonstrated a higher survival rate in mice treated with PD compared to the control group receiving DMSO throughout the 30-day observation period (Figure 7D). Interestingly, no significant difference in body weight changes was observed between the DMSO and PD treatment groups (Figure 7E). Additionally, the immunofluorescence findings demonstrated a significant induction of apoptosis in AML cells in vivo by PD (Figures 7F, G). Immunohistochemical analyses further validated the upregulation of Cleaved caspase3 and LC3 proteins, and downregulation of Ki67 protein in AML cells in vivo following PD treatment (Figure 7H). These results suggest that PD exhibits anti-tumor properties through the stimulation of cell autophagy in AML nude mouse models, without causing evident toxicity or adverse effects in mice.

Hematoxylin and eosin staining indicated that PD did not have a significant impact on the histological changes of the heart, liver, spleen, kidney, and lung compared to the DMSO group (Figure 8A). A pharmacokinetic analysis was conducted to assess the metabolism of different doses of PD in mice (Figure 8B). Consistent with its inhibitory effects on AML cells in vitro, PD led to a dose-dependent decrease in white blood cell count in vivo (Figure 8C). Subsequent studies revealed a significant upregulation of hemoglobin and erythropoietin expression in mice following treatment with PD (Figures 8D, E).