Plant extracts (Table SI) were obtained from the Korea Plant Extract Bank at the Korea Research Institute of Bioscience and Biotechnology. Each extract was dissolved in DMSO.

Human AML cell lines HL-60, U937 and THP-1 were purchased from Korean Cell Line Bank (Seoul, Korea) and cultured in RPMI-1640 medium (Cytiva, cat. no. SH30027.01) supplemented with 10% FBS (HyClone, Cytiva; cat. no. SH30919.03), 1% HEPES (cat. no. 15630-080), 1% penicillin/streptomycin (cat. no. 15140-122) and 1% L-glutamine (all Gibco, cat. no. 25030-81; Thermo Fisher Scientific, Inc.) at 37°C in a 5% CO₂ incubator. U937 cell line was authenticated using STR profiling by Cosmogenetech Co., Ltd. Normal bone marrow cells from wild-type C57BL/6 mice were cultured in RPMI-1640 medium with 10% FBS as aforementioned. Norchelerythrine and peltatoside were purchased from ChemFaces (cat. no. CFN92737 and CFN70318). Phorbol-12 myristate-13 acetate (PMA) (cat. no. P8139) and ROS scavenger N-acetylcysteine (NAC) was purchased from Sigma-Aldrich (cat. no. A9165; Merck KGaA).

The induction of AML cell differentiation was determined by assessing CD11b and CD14 expression on the cell surface. Briefly, cells were treated with DMSO, CIP (20 μg/ml for 96 h), norchelerythirne (2, 5 or 10 μM for 96 h) or Corydalis speciosa Maxim. (20 μg/ml for 96 h) at 37°C, harvested and washed with PBS, followed by staining with FITC-conjugated CD11b (cat. no. 101206) or PE-cy7 conjugated CD11b (both Biolegend, Inc.; cat. no. 101216) or PercP-Cy5.5-conjugated CD14 (BD Biosciences, cat. no. 550787) for 1 h at 4°C in the dark. To assess the effect of norchelerythrine on normal hematopoiesis in mice, bone marrow cells were treated with 5 μM norchelerythrine at 37°C for 96 h. Cells were harvested and washed with PBS, then stained with PE-conjugated CD45 (eBioscience, cat. no. 12-0451-82; Thermo Fisher Scientific, Inc.), PE-Cy7-conjugated CD11b (BioLegend, Inc.; cat. no. 101216), APC-conjugated Ly-6G (BD Biosciences, cat. no. 560599) and FITC-conjugated CD14 (BioLegend, Inc.; cat. no. 123307) for 1 h at 4°C in the dark. The samples were analyzed using a FACSDiva Fusion Flow Cytometer (BD Biosciences). At least 10,000 cells were analyzed for each data point. Data analysis was carried out using FlowJo software version 10.8.1 (Treestar Inc.).

Total RNA was isolated from U937 cells using Tri-RNA reagent (Favorgen, cat. no. FATRR001) and cDNA was synthesized using the PrimeScript RT reagent kit (Takara Bio, Inc., cat. no. RR047A) according to the manufacturer's protocol. Equal amounts of cDNA were used for transcript PCR amplification, which was performed using TOPreal qPCR PreMIX SYBR Green with low ROX (Enzynomics Co., Ltd.; cat. no. RT500M). The thermocycling conditions were as follows: 95°C for 15 min followed by 40 cycles of 95°C for 15 sec, 59°C for 30 sec and 72°C for 30 sec. Table I lists the primers used. Relative gene expression was analyzed using the 2-∆∆Cq method (41).

Cell proliferation rates were determined using trypan blue exclusion assay (42). The cells (2×10⁵ cells/well) were seeded into a 24-well plate and treated with CIP (0, 10 or 20 μg/ml), norchelerythrine (0, 5 or 10 μM) or peltatoside (0, 500 nM, 1, 2 or 5 μM) at 37°C for 24-120 h. Cells were stained with 0.4% Trypan blue (Gibco; Thermo Fisher Scientific, Inc.; cat. no. 15250061) for 3 min at room temperature every day, and the number of viable cells was counted using a hemocytometer.

Cell viability was assessed using CellTiter 96 AQueous MTS assay. The cytotoxicity of CIP and Corydalis speciosa Maxim. extract in human AML cell lines (HL-60 and THP-1) was tested by seeding cells in a six-well plate at a density of 4×10⁵/well followed by treatment with CIP or Corydalis speciosa Maxim. extracts (20 μg/ml) at 37°C for 5-8 days. The cytotoxicity of CIP was identified by plating cells in a 96-well plate at a density of 3×10⁴ AML cells (HL-60 and U937) or 5×10⁵ normal bone marrow cells/well and treating them with CIP (0, 50, 100, 200 or 300 μg/ml) at 37°C for 24 h. The cytotoxic effect of norchelerythrine and peltatoside was examined by treating HL-60, U937 and THP-1 cells with norchelerythrine (0 or 10 μM for 72 or 96 h) or peltatoside (0, 500 nM, 1, 2 or 5 μM for 48, 72 h or 96 h) at 37°C. MTS reagent (Promega Corporation; cat. no. G1112) was added for 4 h at 37°C. The absorbance at 450 nm was measured using a GloMax Microplate multi-mode reader (Promega Corporation) (43).

Trypan blue stain was used to determine the cell apoptosis rate. HL-60 and THP-1 cells were plated in six-well plates at a density of 4×10⁵ cells/well and treated with Corydalis incisa (Thunb.) Pers. or Corydalis speciosa Maxim. extract (20 μg/ml) at 37°C for 5-8 days. The cells were stained with 0.4% trypan blue (Gibco, Thermo Fisher Scientific, Inc.; cat. no. 15250061) for 3 min at room temperature. The number of positively stained cells was counted using a hemocytometer (44).

HL-60 cells (1×10⁴ cells/well) were seeded in a 12-well plate in methylcellulose (Methocult H4100, StemCell Technologies) supplemented with 10% FBS and 1% penicillin/streptomycin, as previously described (43) and treated with CIP (2 μg/ml) at 37°C for 11 days. The colonies containing >50 cells were counted manually 11 days after plating using an Olympus CX31microscope (Olympus Corporation) at 400× magnification.

CIP was analyzed using Waters AQUITYTM UPLC system equipped with an XEVO-QTOF mass detector. ACQUITY UPLC BEH C18 column was used, with two mobile phases containing water with 0.1% formic acid (solvent A) and acetonitrile with formic acid (solvent B). The mobile phase was delivered at a flow rate of 0.4 ml/min, and a 1 μl injection volume was used. The elution gradient was as follows: 1, B 8%; 13, B 8-40%; 2.5, B 40%; 0.5, B 40-100%; 2.5, B 100%; 0.5, B 100-8%; 2 min B 8%. XEVO-QTOF mass detector featured electrospray ionization. The analysis was conducted in negative and positive ion modes in the range of 100-1,500 m/z. N₂ was used as the desolvation gas. The desolvation temperature was set to 350°C at a flow rate of 800 l/h with a source temperature of 110°C. The capillary and cone voltages were set to 300 and 40 V, respectively. CIP was prepared (3 mg/ml) in methanol.

Intracellular ROS were detected by MitoSOX Red staining. Following treatment with 10 μM norchelerythrine at 37°C for 48 h, cells were harvested and washed with PBS, followed by staining with MitoSOX Red (MedChemExpress, cat. no. HY-D1055) for 20 min at 37°C in the dark. The cells were washed with PBS and analyzed using a FACSDIVA fusion Flow Cytometer (BD Biosciences). At least 10,000 cells were analyzed for each data point. Data analysis was performed using FlowJo software version 10.8.1 (Treestar Inc.).

HL-60 and U937 cells were treated with norchelerythrine (5 or 10 μM at 37°C for 6, 12 or 48 h) and washed in PBS. The cells lysed in RIPA buffer (Elpis Biotechnology, cat. no. EBA-1149) containing 1 Na-vanadate, 50 β-glycerophosphate disodium salt (both Sigma-Aldrich; Merck KGaA; cat. no. G9422), 142 β-mercaptomethanol (Bioworld Technology, Inc.; cat. no. 41300000-1) and 5 mM EDTA and ProteaseArrest (Thermo Scientific, Inc.; cat. no. 87786). Protein concentrations were measured using BCA assay kit (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The samples were boiled at 100°C for 10 min, 25 μg proteins loaded on 6 or 15% polyacrylamide gels and transferred to Immobilon-P transfer membrane. The membranes were blocked for 1 h at room temperature in 0.1% TBST with 1% BSA (MP Biomedicals, LLC; cat. no. 160069) and incubated with primary antibodies overnight at 4°C, washed three times for 5 min in TBST, and incubated with HRP-conjugated anti-mouse or anti-rabbit secondary antibodies for 1 h at room temperature. The membranes were washed three times for 10 min in TBST. The protein bands were detected using chemiluminescent substrate (EzWestLumi plus; ATTO corporation) and visualized using the Luminograph II (ATTO Corporation). The proteins were quantified using ImageJ 1.54 g (National Institutes of Health). The primary antibodies against p21 (cat. no. #2947S), γH2AX (cat. no. #9718S), Ataxia telangiectasia mutated (ATM) (cat. no. #2873S) and phosphorylated ATM (all 1:1,000; #5883S) were purchased from Cell Signaling Technology, Inc. The primary antibody against β-actin (1:5,000; cat. no. sc-47778) and the anti-mouse secondary antibody (cat. no. sc-516102, 1:5,000) was obtained from Santa Cruz Biotechnology, Inc. The anti-rabbit secondary antibody (cat. no. A120-101P, 1:5,000) was purchased from Bethyl Laboratories, Inc. (45).

Pusan National University Hospital (Busan, South Korea) provided the bone marrow samples from 4 male patients with AML, collected between August 2023 and November 2023, after obtaining informed written consent from patients and approval from the institutional review board of Pusan National University Hospital (approval no. IRB 2403-010-137). The mean age of the patients was 55 years (range, 21-80 years). Table SII lists the mutation profiles of patients.

CIP-resistant cell line was generated by seeding HL-60 cells (2×10⁵ cells/well) in 12-well plates followed by treatment with DMSO or 2 μg/ml CIP at 37°C for 96 h. Medium was removed, and the cells were cultured for an additional 48 h in RPMI-1640 medium supplemented with 10% FBS, 1% HEPES, 1% penicillin/streptomycin and 1% L-glutamine. Subsequently, the cells were treated with 4, 6, 8, 10, 12, 15, 18 and 20 μg/ml CIP, respectively, in the same manner. This cycle was repeated for over 6 months. Resistance was confirmed by FACS analysis of CD11b and cell counting.

Total RNA from HL-60-DMSO and HL-60-CIP cell lines was extracted using Tri-RNA reagent (Favorgen, FATRR001). DNA was removed using Ambion AM1906 DNase Treatment and Removal Reagents (Ambion) according to the manufacturer's instructions. RNA quality was assessed by Agilent 4200 TapeStation System (Agilent Technologies, Inc.). QuantSeq-3′mRNA-seq was performed by ebiogen, Inc. Briefly, RNA library construction was performed using the Quant-Seq 3′ mRNA-Seq V2 library prep kit FWD with UDI 12 nt Sets A1-A4, (UDI12A_0001-0384). 384 preps (cat. no. 193.384; Lexogen GmbH). The loading concentration of the final library was >4 nM and was measured using Qubit (Thermo Fisher Scientific Inc.). High-throughput single-end 75 bp sequencing was conducted using NextSeq 500/550 (Illumina, Inc.) using a NextSeq 500/550 High Output Kit v2.5 (75 cycles, cat. no. 20024906; Illumina, Inc.). Gene ontology (GO) and differentially expressed genes (DEG) were analyzed using ExDEGA version 5.0 (ebiogen, Inc.). Differentially expressed genes with fold change >2 and P<0.05 were selected. A heatmap was generated using MultiExperiment Viewer 4.9.0 (sourceforge.net/projects/mev-tm4/).

All data are presented as the mean ± SD. Statistically significant differences were calculated using non-parametric Mann-Whitney U or unpaired t-test or one-way ANOVA test with Tukey's post hoc test using Prism version 5.03 software (GraphPad Software, Inc.; Dotmatics). All experiments were repeated ≥3 times. P<0.05 was considered to indicate a statistically significant difference.

A total of 100 plant extracts native to South Korea were screened morphologically based on the presence of adherent cells, after treatment with 20 μg/ml plant extracts for 96 h, which indicated the induction of differentiated cells. Four extracts that promoted the formation of adherent cells were then analyzed for CD11b expression, a myeloid differentiation marker, following exposure to 20 μg/ml plant extracts for 96 h. Extracts of CIP were most effective in increasing CD11b expression (Table SI). Therefore, the present study focused on CIP extracts for further characterization. U937 AML cell line was exposed to CIP extracts (20 μg/ml) to confirm the ability of CIP to induce AML differentiation, which significantly increased CD11b expression (Fig. 1A). In addition, increases in size (forward scatter) and granularity (side scatter) are commonly observed as the cells differentiate (46), which was the case following CIP treatment (Fig. 1B). PMA can induce differentiation of AML cells and make them adherent and elongated (47). Co-treatment of CIP notably increased the number of adherent/elongated cells compared with PMA alone, suggesting that CIP enhances PMA-induced differentiation of THP-1 AML cells (Fig. 1C).

The present study analyzed expression of genes related to myeloid differentiation using RT-qPCR. Consistent with the data that CIP efficiently induces AML differentiation, myeloid differentiation markers (colony stimulating factor 1 receptor, integrin αM and CD14) were upregulated and negative regulators of myeloid differentiation (Myb and Myc) were downregulated following CIP treatment. Furthermore, MAFB, a transcription factor key for early myeloid and monocytic differentiation (48), exhibited increased expression following exposure to CIP. In addition, lysozyme and MMP9, genes associated with differentiated cell functions such as neutrophils and macrophages, showed elevated expression in response to CIP treatment. These results collectively showed that CIP effectively stimulates differentiation in human AML cells (Fig. 1D).

HL-60 and U937 AML cells were exposed to CIP (0, 10 or 20 μg/ml), followed by cell counting every 24 h to determine differentiation by CIP would affect proliferation in AML. CIP significantly decreased the cell number in a dose-dependent manner in HL-60 and U937 cells (Fig. 2A). Consistently, CIP decreased viability and increased the apoptosis rates in HL-60 and THP-1 cells (Fig. 2B and C). CIP extracts (2 μg/ml) had an inhibitory effect on the colony formation in AML (Fig. 2D). In addition, CIP exhibited more toxicity in HL-60 and U937 AML cells than normal bone marrow cells (Fig. S1), suggesting that the cytotoxic effect of this extract may be specific to AML cells with minimal impact on normal cells. These data collectively indicated that CIP extracts have anti-proliferative and pro-apoptotic effects by triggering differentiation in AML cells.

CIP-resistant cells were generated to understand CIP-induced differentiation and identify potential targets of CIP (49). Human AML cell line HL-60 was treated with DMSO or increasing CIP concentrations (2-20 μg/ml) for 6 months. Following exposure to CIP (20 μg/ml), CD11b expression notably increased in HL-60-DMSO but increased to a lesser extent in HL-60-CIP cells (Fig. 3A and B). HL-60-DMSO cells were sensitive to CIP treatment, leading to a decrease in cell number, while HL-60-CIP cells were resistant to CIP, showing no decrease in cell number. These data showed that HL-60-CIP cells are resistant to CIP-induced differentiation and cell cycle arrest (Fig. 3A and B). Furthermore, when HL-60-CIP cells were cultured in the absence of CIP, they remained resistant to CIP for 3 weeks, suggesting that resistance was achieved through stable genetic alteration (Fig. S2).

The genes involved in CIP-induced AML differentiation were identified at the genome level using Quantseq 3′mRNA-seq using HL-60-DMSO and HL-60-CIP cells. The genes regulated by CIP were enriched in 'hematopoietic progenitor cell differentiation', 'myeloid cell development' and 'myeloid cell differentiation' (Fig. 3C). RNA-seq analysis showed that genes that promote leukemogenesis and inhibit myeloid differentiation, such as myeloperoxidase (50), proteoglycan 2 (51), insulin-like growth factor 2 mRNA-binding protein 1 (52), tribbles homolog 1 (53), dachshund homolog 1 (54), ikaros family zinc finger 2 (55), preferentially expressed antigen in melanoma (56), CD81 (57), CCAAT/enhancer binding protein δ (58) and CD84 (59), were upregulated in CIP-resistant HL-60 cells, validating that HL-60-CIP cells are resistant to differentiation-inducing agents by regulating genes involved in modulating AML differentiation (Fig. 3D-F).

The compounds in CIP that induce differentiation in AML cells were identified and quantified by UPLC-QTOF-MS on the CIP extract. Two notably peaks were identified and labeled peltatoside and norchelerythrine (Fig. 4). Peltatoside did not have any significant effect on the viability or proliferation of U937 and THP-1 cells (Fig. S3A and B). These findings suggested that peltatoside is not the differentiation-inducing compound present in CIP. Norchelerythrine significantly increased the fluorescence intensity of CD11b and CD14 in HL-60 and U937 cells (Fig. 5A). Furthermore, treatment of HL-60 and U937 with norchelerythrine resulted in significantly decreased cell viability and proliferation in a dose-dependent manner (Fig. 5B and C). Moreover, norchelerythrine decreased the number of viable cells in primary AML cells from patients harboring IDH1, ETS variant transcription factor 6, Lysine methyltransferase 2A (KMT2A) and CCAAT/enhancer binding protein α mutation (Figs. 5D and S4). Myeloid lineage cells distribution was not altered by norchelerythrine, suggesting that norchelerythrine did not have any effect on normal hematopoiesis in mice (Fig. S5).

The effects of Corydalis genus plant extract on AML cells were examined based on the hypothesis that plants of the same genus have comparable compound compositions and similar biological effects. C. speciosa Maxim. extract increased expression of the CD11b surface marker (Fig. S6A). Similarly to CIP, the extract of C. speciosa Maxim. decreased cell viability and increased the rate of apoptosis (Fig. S6B and C). Consistent with these results, norchelerythrine, but not peltatoside, was also observed in the UPLC-QTOF-MS chromatogram of the C. speciosa Maxim. extract (Fig. S7). These findings suggest that norchelerythrine may be one of the major compounds in CIP responsible for promoting cell differentiation and inhibiting cell proliferation in AML.

ROS influence cellular signaling processes key for cell proliferation and differentiation. In particular, ROS concentration plays a regulatory role in the differentiation of cells during hematopoiesis (60,61). The mechanism of norchelerythrine-induced differentiation was examined using a fluorescent probe (MitoSOX Red) to detect ROS levels in the presence or absence of norchelerythrine. The cells treated with norchelerythrine had higher intracellular ROS levels than those treated with DMSO (Fig. 6A). Previous studies have suggested that alkaloids promote ROS accumulation, leading to DNA damage (62-64). DNA damage in human and murine myeloid leukemia cells is associated with myeloid differentiation and cell cycle arrest (65,66). Therefore, the present study examined whether norchelerythrine activates DNA damage response via ROS production. Norchelerythrine treatment led to increased levels of H2AX phosphorylation at Ser139 (γH2AX), which is a marker of DNA double-strand breaks. In addition, levels of phosphorylated ATM and p21, but not total ATM, increased following exposure to norchelerythrine (Fig. 6B). The ROS scavenger NAC was used to determine if ROS accumulation mediated differentiation of AML cells. NAC treatment reduced the differentiation induced by norchelerythrine significantly (Fig. 6C), suggesting that ROS generated by norchelerythrine were involved in AML differentiation. These findings suggest that ROS and DNA damage response are critical in norchelerythrine-induced differentiation.