Primary AML cells and AML cell lines
Human BM samples were obtained from a cohort of 24 individuals with AML and five healthy volunteer donors from 2007 to 2016 at the Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences. The sample collection procedures were in accordance with guidelines of the Helsinki Declaration, and written informed consents were obtained from all patients and healthy donors after approval of the Ethics Committee of our hospital. All patients were reevaluated and met the 2016 WHO diagnostic criteria for AML. BM mononuclear cells (MNCs) were isolated by standard Ficoll-Plaque density gradient separation procedure and frozen viably. The clinical characteristics of AML patients in this study were listed in Table S1 (Online Resource 1).
The AML cell lines which were used in this study were purchased from the American Type Culture Collection. Kasumi-1, a human AML cell line with t(8;21) translocation, was maintained in RPMI1640 (Gibco, Carlsbad, USA) with 20% fetal bovine serum (FBS, Gibco, Carlsbad, USA) and 1% penicillin/streptomycin (P/S) (Beyotime, Shanghai, China). THP-1, a human monocytic cell line derived from an acute monocytic leukemia patient, was cultured in RPMI1640 supplemented with 10% FBS and 1% P/S. KG-1 and KG-1a cells, two cell lines derived from a patient with AML, were cultured in IMDM supplemented with 20% FBS and 1% P/S. All cells were cultured at 37 °C in a 5% CO2 atmosphere.
MNCs from primary AML samples were seeded in 6-well plates at a concentration of 3 × 105 cells/mL for 2 mL per well then treated with 5.5 μM GSK-J4 (Sigma-Aldrich, USA) or solvent control DMSO (Sigma-Aldrich, USA) for 24 h. Cells were then collected and counted by Trypan Blue (Sigma-Aldrich, USA) staining to determine the alive cell numbers. Cell survival after the 24-h drug treatment was calculated by the percentage of live cells in GSK-J4 treated group divided by that in DMSO group for every single sample. For leukemia cell lines, we seeded Kasumi-1, THP-1, KG-1 and KG-1a cells in 96-well plates (3 × 104 cells/well). After a 72-h treatment of GSK-J4 with various concentrations, Cell Counting Kit-8 (CCK-8, Dojindo Laboratories, Japan) solution was added and then incubated at 37 °C in a 5% CO2 incubator for additional 4 h. Absorbance was measured using Micro-plate Reader (Synergy H4, BioTek, USA) at the absorbance of 450 nm. The data was graphically displayed and the IC50 was determined using GraphPad Prism 5 software (GraphPad Software, USA).
Colony-forming cell assay
Primary MNCs of AML patients or leukemic cell lines were cultured in the presence or absence of 5.5 μM GSK-J4 for 24 h, harvested and washed twice with PBS. Then the treated and untreated primary MNCs, or AML cell lines, were re-counted and plated into 24-well plates in 0.5 mL methylcellulose medium (MethoCult H4434, StemCell Technologies, Canada) at cell densities of 2 × 105 cells/mL or 1 × 103 cells/mL, respectively. Cells were cultured for 14 days then the types and numbers of colonies were identified and numerated under a microscope.
Cell growth, apoptosis and cell-cycle analyses
Kasumi-1 cells were cultured in 24-well plates with or without GSK-J4 for 7 days to test the effect of GSK-J4 on cell growth. Each condition of the cultures was triplicated at each time point and cell numbers were counted manually every day. Kasumi-1 cells were collected and treated with 5.5 μM GSK-J4 or solvent control (DMSO) for 24 h to determine if GSK-J4 affect apoptosis of AML cells. Apoptosis assay was performed with JC-1 staining (AAT Bioquest, Sunnyvale, CA, USA) to detect the mitochondrial membrane potential (MMP) according to the manufacturer’s protocol. JC-1 is capable of selectively entering mitochondria, and reversibly forming JC-1 aggregates upon mitochondria membrane polarization that causes shifts in emitted light from green (emission of JC-1 monomeric form, detected through the FITC channel) to red (emission of JC-1 aggregate form, detected through PE channel). Apoptotic cells demonstrate a decrease in MMP and the increased emitted green fluorescence can be detected by flow cytometry. To determine the impact of GSK-J4 on cell cycle, Kasumi-1 cells were treated by 5.5 μM GSK-J4 for 24 h, followed by BrdU and 7-AAD (BrdU Flow Kits, BD Biosciences, USA) double staining. The percentage of cells in different cell-cycle status was determined by flow cytometry.
Combination index determination
The combination index (CI) values for GSK-J4 and cytosine arabinoside (Ara-C) (Pfizer, New York, USA) were calculated by median dose-effect analysis (assuming mutual exclusivity) using CalcuSyn Version 2.1 (Biosoft, Ferguson, MO) software. CI values less than 1.0 represent a synergistic interaction of the combination treatment (Fiskus et al. 2014; Kojima et al. 2008).
Chromatin immunoprecipitation (ChIP)-qPCR
The chromatin immunoprecipitation (ChIP) assay was performed with EZ-Magna ChIP™ A-Chromatin Immunoprecipitation Kit (Millipore, Temecula, USA) according to the manufacturer’s recommendations. Gene expression was measured by real-time quantitative PCR (qPCR) as we previously described (Si et al. 2016). The primers used in qPCR were shown in Table S2 and S3 (Online Resource 1).
Western blot assay
Whole-cell lysates of GSK-J4 or DMSO treated cells were extracted by SDS loading buffer (Beyotime, China) and subjected to western blot analyses. The antibodies used for western blot include H3 (ab1791, Abcam, England) and H3K27me3 (C36B11, Cell Signaling Technology, USA). These two antibodies along with normal rabbit IgG (2729, Cell Signaling Technology, USA) were also used in ChIP assays. All the experiments were repeated at least twice with similar results.
RNA-sequencing and bioinformatic analyses
RNA-sequencing experiments for Kasumi-1 cells treated with DMSO or 5.5 µM GSK-J4 were performed by Novogene (Beijing, China). Briefly, total RNAs were isolated from Kasumi-1 cells (three biological replicates per condition) using TRIzol reagent (Life Technologies, USA), and the RNA qualifications were confirmed by 1% agarose gels and the Bioanalyzer 2100 system (Agilent Technologies, CA, USA). Sequencing libraries were generated using NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (NEB, USA) following manufacturer’s protocols. After cluster generation, the libraries were sequenced on an Illumina Hiseq platform, and 150 bp paired-end reads were generated. The split read aligner TopHat (v2.0.12) and Bowtie2 (v2.2.3) were used to align the reads to the appropriate genomes. HTSeq (v0.6.1) was used to count the read numbers mapped to each gene and FPKM method was used for determining gene expression levels. Differential expression analysis of GSK-J4 and DMSO groups was performed using the DESeq R package (1.18.0). Genes with an adjusted p value < 0.05 and fold change > 1.6 found by DESeq were assigned as differentially expressed. Gene Ontology (GO) enrichment analysis of differentially expressed genes was implemented by the GOseq R package and the KOBAS software was used to test the statistical enrichment of differentially expressed genes in KEGG pathways.
Mouse model of human AML
To validate the anti-leukemic effect of GSK-J4 in vivo, 1 × 107 Kasumi-1 cells were injected via tail vein into 5-week-old female NOD-Prkdcem26Cd52Il2rgem26Cd22/Nju (NCG) mice (Nanjing Biomedical Research Institute of Nanjing University, China) following a sub-lethal irradiation at a dose of 200 cGy. GSK-J4 (50 mg/kg) or solvent control (1.5% DMSO in 0.9% saline) was administered into the recipient mice via peritoneal injection once a day for 4–6 weeks (5-day injection and 2-day break for each week). The final concentration of diluted GSK-J4 was 5 mg/mL and the injection volume was ~ 300 μL. Mice were humanely killed in accordance with Institutional Animal Care and Use Committee (IACUC) protocols at 10 weeks. BM cells (mixed cells from tibias and femurs) were flushed and suspended in PBS for phenotypic analysis by flow cytometry. BM, spleens, livers and kidneys were fixed in 10% Accustain Formalin Solution (Sigma-Aldrich, USA) and embedded in paraffin for hematoxylin and eosin (H&E) staining. All animal experiments were approved by the Animal Research Committee of our hospital.
BloodSpot database was used to retrieve mRNA expression data for the expression patterns of KDM6B in AML patients and normal hematopoietic stem cells (HSCs) (http://servers.binf.ku.dk/bloodspot/) (Bagger et al. 2016). The relationship between KDM6B gene expression and prognosis in patients with AML was examined using the PrognoScan database (http://www.abren.net/PrognoScan/) (Mizuno et al. 2009). The patient samples were split into high and low KDM6B expression groups with the optimal cutoff point determined by the minimum p value approach. The p value was calculated by the log-rank test and corrected according to the database description.
Experimental data was reported as mean with standard error of the mean (SEM), unless otherwise indicated. Unpaired t test or Mann–Whitney test was used for two group comparisons, and ANOVA analysis was used to determine differences among three or more groups. A p value of < 0.05 was considered significant.