Cell lines and culture

The human tumor cell lines and 293FT cells were obtained from the Cell Bank of the Chinese Academy of Sciences. CMK, MEG01, M07e, U937, and K562 cells were cultured in RPMI medium (Vivacell, C3001-0500, Shanghai, China) supplemented with 1% penicillin/streptomycin (P/S, Beyotime, C0222 ) and 10% fetal bovine serum (FBS, Vivacell, C04001-050). In addtion, M07e cells were treated with 10 ng/ml recombinant human GM-CSF (E.coli) (Novoprotein, C003, Suzhou, China). 293FT cells were cultured in high-glucose DMEM medium (Vivacell, C3113-0500) supplemented with 1% P/S and 10% FBS. All the cell lines were maintained in a humidified atmosphere (37℃, 5%CO2). All the cell lines were subjected to short tandem repeat (STR) identification, and the identification results are shown in Supplementary Material 1.

Cell counting kit-8 (CCK-8) assay

To assess the effect of indisulam on cell viability, CMK, MEG01, M07e, U937, and K562 cells were seeded in 96-well plates at a density of 1 × 104 cells/well. Indisulam (MedChemExpress, HY-13650, NJ, USA) was dissolved and diluted with dimethyl sulfoxide (DMSO, Sigma-Aldrich, D2650, MO, USA). After indisulam treatment for 72 h, cell proliferation ability and viability were detected via a CCK-8 kit (APExBIO, K1018, TX, USA) according to the manufacturer's instructions. Each concentration was tested in three independent experiments. The short hairpin RNA (shRNA)-treated cells were seeded in 96-well plates at a density of 1 × 103 cells/well and were tested every two days for a total of one week after puromycin selection. The absorbance at 450 nm was measured with a spectrometer (Thermo, MA, USA), and the ability of proliferation and viability were quantified via Graph Prism 9.0 (GraphPad Software Inc., San Diego, CA, USA). Three technical replicates were performed for each experiment.

Soft agar colony formation

To detect the proliferative ability and invasiveness, the cells were treated or transduced as described above and 1 × 104 cells/well were inoculated in a soft agar medium seeded into 6-well plates. Every 1 to 2 days, 100 to 200 μl complete medium was added to support nutrients and prevent gel cracking. After approximately 2 to 3 weeks, the cells were fixed overnight with 4% paraformaldehyde (Beyotime, P0099). The samples were subsequently stained with Giemsa solution (Beyotime, C0131) and photographed. Finally, the number of colonies in each group was calculated as the number of cells that formed colonies \(\div\) the total number of cells × 100%. The data are representative of 3 technical replicates.

Lentivirus preparation and infection

The shRNAs targeting RBM39 and ZMYND8 were inserted into the pLKO.1-puro lentiviral vector (IGE Biotechnology Ltd., Guangzhou, China) containing puromycin resistance. To prepare lentiviruses, we purchased the envelope plasmid and packaging plasmid from Addgene (pMD2. G: #12259; psPAX2: #12260; Cambridge, MA, USA). Then, we transfected the purified plasmids together with pMD2. G and psPAX2 into 293FT cells via polyethyleneimine (PEI) (Sigma-Aldrich, 49553-93-7) at a ratio of 4:1:3 according to the manufacturer’s protocol. After 6 h of transfection, replace the medium with fresh medium. The supernatants were collected at 48 h and 72 h after transfection and filtered through a 0.45 μm filter. To concentrate the lentivirus, a quarter volume of PEG8000 (Sigma-Aldrich, P5413) was then mixed with lentivirus and incubated on a shaker at 4 °C. After 24 h, the samples were centrifuged at 4000 g for 20 min. The resulting pellet was finally resuspended by small volume of Phosphate-buffered saline (PBS). When performing lentivirus transfection, AMKL cells were incubated with concentrated lentivirus in the presence of 1 μg/mL hexadimethrine bromide (Polybrene) (Sigma-Aldrich, H9268) for 24 h. After transfection, stable cell lines were selected by puromycin (Beyotime, ST551) for 3 days. The sequences of shRNA are listed in Supplementary Table S1.

Real-time quantitative PCR (RT-qPCR)

To detect the mRNA expression level, RNA was extracted via a Total RNA Isolation Kit (Vazyme, RC112-01; Nanjing, China) according to the manufacturer's instructions. Total RNA was then reverse transcribed into cDNA via a reverse transcription kit (Applied Biosystems, 94404, CA, USA). RT-qPCR was performed via LightCycler 480 SYBR Green I Master Mix (Roche, 4887352001, Basel, Switzerland) in a LightCycler 480 Real-Time System (Roche). The relative mRNA expression of the target genes was subsequently calculated via the \(^}\) method, with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression used as a control. Three technical replicates were performed for each experiment. The primers used in this study are shown in Supplementary Table S2.

RNA splicing analysis and verification

Following the alignment of the RNA-seq data, rMATS (4.1.2) software was used to analyze alternative splicing events, including skipped exons (SEs), retained introns (RIs), mutually exclusive exons (MXEs), alternative 5' splice sites (A5SSs), and alternative 3' splice sites (A3SSs). The splicing events were visualized via the IGV Genome browser (version 2.19.2). Significant splicing events were subsequently chosen based on the criteria of a false discovery rate (FDR) < 0.05 and an absolute difference in inclusion level (IncLevel) > 0.2. Total RNA from vehicle- or indisulam- treated cells were extracted by fastPure Cell/Tissue Total RNA Isolution Kit V2 (Vazyme, RC112-01; Nanjing, China). Then, cDNA was synthesized from 1000 ng of total RNA using the reverse transcription kit (Applied Biosystems, 94404, CA, USA) in a 25 μl reaction on an ABI PCR instrument (Thermo Fisher, Applied Biosystems). To confirm the RNA splicing events, PCR was performed to detect SE of EZH2 and MXE of ZMYND8. cDNA was amplified using 2 × Taq Plus Master Mix II (Vazyme, P213) and subjected to electrophoresis on 1.2% agarose gel at 130 V for 30 min. Finally, the gel was visualized using a gel imager (UVITECT, FireReader, UK) under ultraviolet light. The PCR primers used are listed in Supplementary Table S2.

Western blot analysis

On the first day, AMKL cells were collected after the corresponding treatment and washed with PBS before lysis using RIPA lysis buffer (Beyotime, P0013B). The extracted protein concentration was quantified to be 10 mg/ml. Proteins of the same mass were separated via electrophoresis, transferred to polyvinylidene fluoride (PVDF) membranes (GVS, 1212639, Zola Predosa, Italy), and blocked with 5% skim milk. Then, primary antibodies were added, and the samples were incubated overnight at 4 °C with shaking. On the second day, the PVDF membranes were incubated with secondary antibodies at room temperature for 1 h. After the PVDF membranes were wetted with ECL luminescence solution (Millipore, WBKLS0500, USA), the membranes were visualized with an AI600 image gel imaging analyzer (GE, MA, USA). GAPDH was used as a loading control. The raw images of western blots are provided in the supplementary material. The antibody details are shown in Supplementary Table S3.

RNA sequencing (RNA-seq)

To detect the mechanically induced effects of indisulam on AMKL, CMK cells treated with indisulam were collected and sent to Novogene Bioinformatics Technology Co., Ltd. (Beijing, China) for sequencing. Libraries were constructed from total RNA, and the resulting mRNA fragments were used as templates to synthesize the first strand of cDNA via the M-MuLV reverse transcriptase system, followed by the second strand of cDNA from dNTPs. The purified and screened cDNAs were subjected to PCR amplification, and the resulting products were subsequently purified to obtain a library. DESeq2 software (version 1.40.0) was used for differential expression analysis between the two comparison combinations. The sequencing results are shown in Supplementary Table S4.

Quantitative proteomicsProtein extraction

Proteomic analysis was performed by Jingjie Biotechnology Co., Ltd. (Hangzhou, China). Samples were suspended in lysis buffer (8 M urea, 1% protease inhibitor) and lysed by ultrasonic disruption. The total protein was collected by centrifuge at 4 °C, 12,000 g for 10 min and proceed to concentration determination using BCA assay.

Trypsin digestion

For digestion, equal amounts of protein were taken from each sample and adjusted to the same volume with lysis buffer. Trichloroacetic Acid was added to a final concentration of 20%, vortexed, and incubated at 4 °C for 2 h to precipitate proteins. After centrifugation (4500g, 5 min), the supernatant was discarded, and the pellet was washed 2–3 times with cold acetone. After drying the precipitate, 200 mM Triethylammonium Bicarbonate was added to reach the final concentration. The precipitate was sonicated to disperse, and trypsin was added at a 1:50 (protease:protein, m/m) ratio for overnight digestion. Dithiothreitol was added to a final concentration of 5 mM, and reduction was carried out at 56 °C for 30 min. Iodoacetamide was then added to a final concentration of 11 mM, and the mixture was incubated at room temperature in the dark for 15 min.

LC–MS/MS analysis

The peptides were dissolved in mobile phase A of the liquid chromatography system and loaded onto the Evotip according to the manufacturer's instructions. The peptides were then separated using the Evosep One ultra-high-performance liquid chromatography system, utilizing the pre-set 60 SPD method. After separation, the peptides were ionized in the Capillary ion source and injected into the timsTOF Pro 2 mass spectrometer for data acquisition. The data were acquired using the data-independent parallel accumulation serial fragmentation mode. FDR was adjusted to < 1%. The proteins with a P-value < 0.01 and a log2-fold change between the two groups, either < − 0.9 or > 1.11, were considered significantly different in expression. These significant proteins were then selected and verified via Western blot analysis.

Apoptosis and cell cycle assays

We collected AMKL cells and washed them with precooled PBS. Then, the cells were suspended in 1 × binding buffer and stained with fluorescein isothiocyanate (FITC)-Annexin V antibody (BD Biosciences, 556420, NJ, USA) and propidium iodide (PI) solution (BD Biosciences, 556463) for apoptosis analysis. For cell cycle analysis, the cells were harvested, washed with precooled PBS, and fixed in 75% ethanol for 24 h. The cells were processed via the Cell Cycle and Apoptosis Assay Kit (Beyotime, C1052) following the manufacturer's instructions. Both the apoptosis and cell cycle data were processed via flow cytometry (Beckman Gallios™ Flow Cytometer; Beckman, USA). Each experiment was performed three technical replicates.

Generation of CRISPR–Cas9 DCAF15 knockout

The Cas9 gene was introduced into CMK and MEG01 cell lines through lentiviral transduction and selected with 500 μg/mL geneticin (MedChemExpress, HY-108718) for one week to establish stable Cas9-expressing cell lines. Small guide oligos (sgRNAs) targeting DCAF15 were then cloned and inserted into the Lenti-CRISPR plasmid, which was purchased from GENECHEM (Shanghai, China). Cells expressing the Cas9 gene were transduced with either sgRNA-DCAF15 or nontargeting control sgRNA (sgNC) lentiviral particles. Lentivirus preparation was conducted by GENECHEM via previously established protocols. The sequences of the sgRNAs are listed in Supplementary Table S1.

In vivo experiments

All experiments related to animals were approved by the ethics committee of the Animal Care Committee of Soochow University (CAM-SU-AP#: JP-2018-1). NSG female mice aged 4–6 weeks were randomly divided into two or three groups. One hundred microliters of a cell suspension prepared in PBS (1 × 106 CMK cells expressing firefly luciferase) was injected into each mouse via the tail vein. For drug treatment, indisulam was dissolved in 20% SBE-β-CD (MedChemExpress, HY-17031) in saline and administered via intraperitoneal injection (12.5 mg/kg/day) or vehicle for 7 days. After 10 days of injection, in vivo imaging was conducted by an imaging system (Berthold, NightOWL II LB 983, Germany) every 2 to 3 days. The fluorescence intensity, body weight, adverse skin events, hair condition, and incidence of diarrhea were monitored during the treatment period. At the end of the study, the paraffin-embedded tissue blocks from the liver, spleen, and hind limbs were prepared for hematoxylin and eosin (HE) staining and immunohistochemistry (IHC) with a Ki67-antihuman antibody (Servicebio, GB12114, China). All mice were euthanized before reaching a 20% body weight loss.

Statistical analysis

Statistical analysis was performed using GraphPad Prism software (version 9.0) and R (version 4.2.2). The Mann–Whitney U test was used to compare the two groups' differences, and the Pearson method was used for correlation analysis. In the case of three groups, one-way ANOVA was performed. The survival time was analyzed using the Log-Rank test with P < 0.05 considered statistically significant. The following symbols indicated the data significance: * P< 0.05, ** P < 0.01, *** P < 0.001, and **** P < 0.0001.