Background

The EP300-ZNF384 fusion gene is a common genomic lesion in patients with acute B lymphoblastic leukemia (B-ALL) and B/M mixed phenotype acute leukemia (MPAL) [1, 2]. It is generated by the t(12; 22) (p13; q13) chromosome translocation fusing EP300 to ZNF384 [1]. In B-ALL, ZNF384 can fuse with 22 partner genes, including EP300, TCF3, TAF15, and CREBBP [3]. Among these, EP300 is recognized as the most common fusion partner of ZNF384, found in approximately 3.7% of patients with B-ALL [4]. Most EP300-ZNF384-positive cases showed a typical immunophenotype (weak or absent expression of CD10 and aberrant expression of at least one myeloid antigen; CD13 or CD33) [5, 6]. Patients with EP300-ZNF384 achieve more favorable clinical outcomes than those patients without it [5, 7].

In comparison to B-ALL with other genomic lesions, EP300-ZNF384-positive B-ALL is characterized by a distinct gene expression signature predominantly enriched in the upregulation of Janus kinase/signal transducers and activators of transcription (JAK/STAT) and cell adhesion pathways, and downregulation of the cell cycle and DNA repair pathways [7, 8]. Expression of EP300-ZNF384 blocked the differentiation of mouse Lin− c-KitLow cells into CD19+ B cells in vitro [9]. Mice transplanted with EP300-ZNF384-expressing bone marrow (BM) cells develop acute monoblastic leukemia with defects in B cell development, characteristic leukocytosis, anemia, splenomegaly, and accumulation of blast cells in the BM [9]. Expression of EP300-ZNF384 in pro-B cells resulted in the development of pre-B cell ALL and differentiation arrest at the pre-B stage in a mouse transplantation model [10]. These results indicated that EP300-ZNF384 is an oncogenic driver of B-ALL. Emerging studies have demonstrated that EP300-ZNF384 promotes B cell proliferation, inhibits the differentiation of B cell, and maintains the hematopoietic stem cell (HSC) signature by activating the expression of CLCF, BTLA, GATA3, ID2, and SALL4 [6, 11, 12]. However, the oncogenic mechanisms underlying the functions of EP300-ZNF384 remain largely unknown.

IL3Rα (also known as CD123) is the α subunit of the interleukin 3 (IL-3) heterodimeric cell-surface receptors, which is widely expressed in hematopoietic stem/progenitor cells, monocytes, megakaryocytes, B-lymphocytes, and plasmacytoid dendritic cells [13,14,15]. Upon binding IL-3, IL3Rα recruits the βc subunit to form a functional IL3R complex, which regulates the proliferation, survival, and differentiation of hematopoietic cells [16,17,18,19]. IL3Rα is highly expressed in various types of leukemia, such as acute myeloid leukemia (AML), B-ALL, and blastic plasmocytoid dendritic neoplasms and has been recognized as a marker for leukemic stem cells in AML [20,21,22,23]. Elevated expression of IL-3Rα is associated with enhanced blast proliferation and a poor prognostic phenotype in patients with AML, indicating that CD123 is a promising therapeutic target for leukemia therapy [24,25,26]. Multiparameter immunophenotypic analysis demonstrated that patients with ZNF384 rearrangements, including EP300-ZNF384, had significantly higher CD123 than those without a ZNF384 fusion, suggesting that CD123 is a biomarker for ZNF384-related fusion [27]. These observations suggest the possible involvement of the IL3RA signal in EP300-ZNF384-primed B-ALL.

In this study, we investigated the functional roles of IL3RA signaling pathway in EP300-ZNF384-induced leukemogenesis and identified the regulatory mechanism of EP300-ZNF384 fusion protein that is responsible for IL3RA transcriptional activation. We further evaluated therapeutic potential of doxorubicin in EP300-ZNF384 positive B-ALL cells in vitro and in vivo, pointing to that targeted therapies involving anthracyclines may produce a favorable outcome in patients with EP300-ZNF384.

Materials and methods

Sample preparation and cell cultures

Primary samples were collected from patients diagnosed with B-ALL, according to the French-American-British classification, at the Department of Hematology of the Second Hospital of Dalian Medical University. Peripheral blood mononuclear cells (PBMCs) were obtained from healthy individuals for BM transplantation and were used to purify CD19+ cells by EasySep™ Human CD19 Positive Selection Kit II (Cat. 17,854, Stemcell Technologies, Vancouver, BC, Canada) according to the manufacturer’s instructions. All patients and donors provided written consent for the protocols, which were approved by the Institutional Review Board and Medical Science Ethics Committee of Dalian Medical University in accordance with the Declaration of Helsinki (2023-XWLW NO.2). HEK-293T cells were purchased from the American Type Culture Collection. Human B-cell acute lymphoblastic leukemia cell lines NALM-6 and BALL-1 were purchased from iCell Bioscience Inc. (Shanghai, China). The human leukemia cell line KG-1α and the Burkitt’s lymphoma cell line Daudi were obtained from Professor Ying Lu (Institute of Dermatology, Xinhua Hospital, School of Medicine, Shanghai Jiao Tong University, China). Primary PBMCs were cultured in Iscove’s modified Dulbecco’s media (IMDM) supplemented with 20% fetal bovine serum (FBS, Gibco, Waltham, MA, USA). HEK-293T cells were cultured in DMEM supplemented with 10% FBS (Gibco). NALM-6, BALL-1, and Daudi cells were cultured in RPMI 1640 media supplemented with 10% FBS (Gibco). KG-1α was cultured in IMDM media supplemented with 20% FBS (Gibco). All cells were maintained in a humidified 5% CO2 atmosphere at 37 °C.

Reagents and antibodies

Antibodies against ZNF384 (Cat. Ab176689; Abcam, Cambridge, UK), p-STAT5 (Cat. 9314; CST, Danvers, MA, USA), STAT5 (Cat. 94,205; CST), FLAG (Cat. F1804; Merck, Darmstadt, Germany), and GAPDH (Cat. 60004-1-Ig; Proteintech Group, Rosemont, IL, USA) were used for western blot analysis. The following antibodies were used for flow cytometry analysis: APC mouse anti-human CD123 (Cat. 560087; BD Biosciences, Franklin Lakes, NJ, USA), APC mouse IgG2a, κ Isotype Control (Cat. 555576; BD Biosciences), APC Mouse Anti-Human CD19 (Cat. 340437; BD Biosciences), and an IgG1 kappa isotype control (Cat. 340442, BD Biosciences). The doxycycline hyclate was purchased from Meilunbio (Cat. MB1088; Dalian, China). Recombinant human IL-3 was purchased from PeproTech (Cat. 200-03; Cranbury, NJ, USA). Doxorubicin hydrochloride (Cat. T1020), dexamethasone (Cat. T1076) and vincristine sulfate (Cat. T1270) were purchased from TOPSCIENCE (Singapore). Puromycin was purchased from Selleck Chemicals (Cat. S7417; Houston, TX, USA). Enzyme-linked immunosorbent assay (ELISA) kit for IL-3 was purchased from Cloud-Clone Corp. (Cat. SEA076Hu; Houston, TX, USA). Pegaspargase was purchased from Jiangsu Hengrui Pharmaceutical Co., Ltd.

RNA library preparation, sequencing, and processing

Total RNA was extracted from cryopreserved mononuclear cells (MNCs) of EP300-ZNF384 positive B-ALL patients (n = 2) and healthy donors (n = 8) using an AllPrep DNA/RNA Mini Kit (Cat. 80,204; Qiagen, Hilden, Germany). After quality control of RNA concentration and purity, libraries were prepared according to the protocol of the TruSeq RNA Sample Preparation Kit (Illumina, San Diego, CA, USA), and library quality was assessed using a Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA). Whole transcriptome sequencing was performed on a NovaSeq platform with a paired-end 150-bp read-length by the Novogene Company (Beijing, China). All sequencing data were mapped to the reference genome (hg38) using STAR (v2.7.6a) [28], and gene expression levels were measured as reads per kilobase per million mapped reads (RPKM) obtained using the Cufflinks package (v2.2.1) [29] guided by transcript coordinates according to the gene annotation format (GTF) file from GENCODE (Release 27, GRCh38).

Public RNA-seq data analysis

To obtain the expression features of EP300-ZNF384-positive B-ALL patients, the expression matrix was downloaded from Therapeutically Applicable Research to Generate Effective Treatments ALL Phase II (TARGET-ALL-P2: https://ocg.cancer.gov/programs/target/), and subtype information was obtained from Gu et al. [30]. The TARGET-ALL-P2 cohort included 204 patients with B-ALL, consisting of seven EP300-ZNF384-positive patients, seven other ZNF384 rearrangement patients, and 197 EP300-ZNF384-negative patients. We compared the expression profiles between EP300-ZNF384-positive and -negative patients using the “limma” package (v3.46.0) in R (http://cran.r-project.org/, v3.6), and the cutoffs for significantly differential expressed genes (DEGs) were set as fold change (FC) > 2 and P < 0.01. Gene set enrichment analysis (GSEA) was performed between EP300-ZNF384-positive and -negative patients using the Molecular Signatures Database (MSigDB; http://www.broadinstitute.org/gsea/msigdb/index.jsp). Volcano plot and heatmap were drawn using the “ggplot2” R package (v3.2.1).

Plasmids

The EP300-ZNF384 fusion gene and wild-type ZNF384 were cloned from B-ALL patient samples using the Phanta Max Master Mix (Cat. P525; Vazyme, Nanjing, China) and inserted into the pLVX plasmid to construct pLVX-EP300-ZNF384, pLVX-EP300-ZNF384-3×flag, pLVX-ZNF384, and pLVX-ZNF384-3×flag respectively. For construction of the luciferase reporter vectors, a series of the sequences with different 5’ flanking regions of the IL3RA promoter were amplified from human genomic DNA with indicated primers and inserted into the pGL3-basic plasmid using the ClonExpress Ultra One Step Cloning Kit (Cat. C115; Vazyme). The EP300-ZNF384-binding site was mutated using the Mut Express II Fast Mutagenesis Kit V2 (Cat. C214; Vazyme) to construct mutant reporter plasmids (-234/-222 mutation and − 234/-222 deletion). Individual clones of each construct were confirmed by DNA sequencing. The primer sequences are listed in Table S1.

ShRNA

The sequences targeting IL3RA (shIL3RA-1: GCAGT GAACA ATAGC TATTG C; shIL3RA-2: GGATT CATGA CGTGG ATTTC T; shIL3RA-3: GGTGT CGTTT CGATG ACATC T; shIL3RA-4: GGAAC GTACA CAGTA CAAAT A) were cloned into the pLKO lentiviral vector (Addgene, Watertown, MA, USA). An empty pLKO lentiviral vector was used as a negative control.

Lentivirus packaging and transfection

HEK-293T cells were co-transfected with the lentiviral constructs, packaging plasmid (psPAX2) and envelope plasmid (pMD2.G) at a ratio of 4:3:1 using jetPRIME® (Polyplus, Illkirch, France). The viruses were collected, concentrated, and added to the cells using polybrene (Cat. HY-112,735; MedChemExpress, Monmouth Junction, NJ, USA) for 24 h. The infected cells were further selected with 2 µg/mL puromycin (Cat. S7417; Selleck) or 400 µg/mL hygromycin B (Cat. ST1389; Beyotime, Shanghai, China), to obtain stably transfected cells.

Flow cytometry

The cells were washed with a staining solution (2% FBS in PBS) and resuspended in a staining solution containing fluorochrome-conjugated antibodies for 30 min. Propidium iodide (PI) was added to exclude dead cells from analysis. After staining, the cells were washed and analyzed using a Cytoflex flow cytometer (Beckman Coulter, Brea, CA, USA). Analysis was performed using CytExpert and FlowJo software. For xenograft sample analysis, BM cells were flushed from the femurs of mice and subjected to red blood cell lysis (Cat. R1010; Solarbio, Beijing, China) according to the manufacturer’s protocol. BM cells were then strained through a 70-µm mesh cell strainer and immunophenotyped using standard methodology.

Western blot analysis

Fresh cells were lysed on ice in radioimmunoprecipitation assay buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate and 1 mM phenylmethylsulfonyl fluoride) containing protease and phosphatase cocktail inhibitors (Cat. HY-K0010 and Cat. HY-K0021; MedChemExpress) for 30 min. The lysates were then cleared by centrifuging at 12,000 g for 10 min. The protein concentration was quantified using the Bradford assay. Protein lysates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to a nitrocellulose membrane (Millipore, Burlington, MA, USA), and immunoblotted with specific antibodies. Protein expression was detected using a ChemiDoc XRS Imaging System (Bio-Rad Laboratories; Hercules, CA, USA) with an enhanced chemiluminescence reagent (Advansta, San Jose, CA, USA).

RNA extraction and RT-PCR/qPCR

Total RNA was extracted from cultured cells or patient samples using AG RNAex Pro Reagent (Cat. AG21102; Accurate Biology, Changsha, China), and RT was performed using the HiScript II Q RT SuperMix kit (Cat. R223; Vazyme) according to the manufacturer’s instructions. QPCR was performed on a Bio-RAD CFX96 fluorescence quantitative PCR instrument by using the ChamQ Universal SYBR qPCR Master Mix (Cat. Q711; Vazyme) and the relative expression of the interested genes was calculated by the 2−ΔΔCt method. Human ACTB was used as an internal control. The primer sequences are listed in Table S2.

Cell line-derived xenograft (CDX) models

The NOD⁃PrkdcscidIl2rgnull (NYG) mice were purchased from Liaoning Changsheng Biotechnology co., Ltd. (China). All animal studies were approved by the Institutional Animal Care and Use Committee of Dalian Medical University and carried out in accordance with established institutional guidelines and approved protocols (AEE22111). NALM-6-EP300-ZNF384 and NALM-6-EP300-ZNF384/shIL3RA cells (5 × 105) were intravenously injected into NYG mice (5-8 weeks old). Mice were fed doxycycline (Dox, 2 mg/mL)-containing water supplemented with 3% sucrose, which was replaced every 2 d. Mice were weighed daily. Four mice were sacrificed 20 d post-transplantation to monitor engraftment. The remaining six mice were euthanized after signs of disease (10% weight loss, disruption of locomotor coordination, lack of grooming, and lethargy), and hematopoietic tissues (BM from femurs, spleen, and liver) were harvested. For the doxorubicin susceptibility analysis, NALM-6-vector, NALM-6-EP300-ZNF384, and NALM-6-ZNF384 cells (5 × 105) were intravenously injected into NYG mice. After 10 d, doxorubicin (8 mg/kg) was administered intraperitoneally. The experiments were terminated on day 17, and hematopoietic tissues (BM from the femurs and liver) were harvested.

Hematoxylin and eosin (HE) staining and Wright–Giemsa staining

Spleen and liver tissues were fixed with 10% neutral-buffered formalin for 24 h. Tissue sections were prepared, stained with HE staining solution, and examined by a pathologist blinded to the experimental groups. Representative images were acquired at ×200 magnification using an Olympus BX63 microscope equipped with a DP73 camera (Olympus, Tokyo, Japan). BM smears were prepared from mice and stained with Wright-Giemsa stain solution (Cat. G1020; Solarbio) according to the manufacturer’s protocol. The cell morphology was observed under an Olympus BX63 microscope.

Dual luciferase assay

HEK-293T cells were co-transfected with 200 ng of pLVX-EP300-ZNF384, pLVX-ZNF384, or the pLVX empty vector in combination with 200 ng of pGL3-IL3RA promoter and 10 ng pRL-TK renilla plasmid using the jetPRIME® (Polyplus). After transfection for 24 h, both firefly luciferase and Renilla luciferase activities were detected with an EnSpire Multimode Plate Reader (PerkinElmer, Waltham, MA, USA) using the Dual-Luciferase Reporter Assay System (Cat. E1910; Promega, Madison, WI, USA) according to the manufacturer’s instructions. Firefly luciferase activity was normalized to Renilla luciferase activity to control transfection efficiency. Primers used for the luciferase constructs are listed in Table S1.

Chromatin immunoprecipitation (ChIP)

The pLVX-EP300-ZNF384-3×flag, pLVX-ZNF384-3×flag and pLVX-3×flag plasmids were respectively transfected into HEK-293T cells for 48 h using jetPRIME® (Polyplus). ChIP assay was conducted using the ChIP-IT ® Express kit (Cat. 53009; Active Motif, Carlsbad, CA, USA), according to the manufacturer’s specifications. The cells were cross-linked with 1% formaldehyde (Cat. 28908; Thermo Fisher Scientific) and lysed with lysis buffer. Cross-linked chromatin was sheared into fragments with an average size of 300-1500 bp and immunoprecipitated using antibodies against flag and IgG (Cat. A7028; Beyotime). Specific PCR amplification was conducted using primers compatible with IL3RA or MMP7 promoter sequences. The primer sequences are listed in Table S3.

ELISA

CD19+ cells were infected with lentiviruses overexpressing the empty vector, EP300-ZNF384, or ZNF384 in serum-free medium. Cell culture media were collected at 48 h and the secretion of IL-3 was determined using an ELISA kit for IL-3 (Cat. SEA076Hu; Cloud-Clone Corp.) according to the manufacturer’s instructions.

Cell viability assay

Cell viability was determined using the CCK-8 assay. Briefly, NALM-6 cells harboring the empty vector, EP300-ZNF384, or ZNF384 were seeded into a 96-well plate at a density of approximately 2 × 104 cells per well and treated with the indicated drug concentrations for 48 h. At the end of the experiment, 10 µL of CCK-8 reagent (Cat. MA0218; Meilunbio) was added to each well and incubated at 37 °C for 2 h. Optical densities at 450 nm were determined using a Multiscan Spectrum SpectraMax M2 (Molecular Devices, San Jose, CA, USA). The relative viability values of cells were calculated.

Methylcellulose-based colony forming cell (CFC) assay

Single-cell suspensions were plated in 96-well plates (Corning) at a density of 200 cells/well and cultured in IMDM medium supplemented with 10% FBS (Gibco) and 1.27% methylcellulose (Sigma). The cells were treated as indicated. After 10 d, the colonies were photographed and counted under an inverted microscope (Olympus).

Apoptosis analysis

An Annexin V-FITC/PI Apoptosis Kit (Cat. E-CK-A211; Elabscience) was used to detect the apoptosis-inducing effects of doxorubicin in NALM-6 cells. The cells were resuspended in a 500 µL 1× binding buffer with 5 µL Annexin V-FITC and 5 µL PI and incubated at room temperature for 15 min in dark. Apoptosis was detected using flow cytometer (CytoFLEX, Beckman Coulter). The results were analyzed using FlowJo software.

Statistical analysis

Statistical analyses were performed using the GraphPad Prism 8 software (GraphPad Software Inc., La Jolla, CA, USA). The unpaired Student’s t-test was used to perform statistical analyses between two groups. One-way analysis ANOVA was used for comparisons between two groups in multiple groups. Survival data were analyzed using the log-rank test, and survival curves were assessed using the Kaplan-Meier method. The levels of significance were set at *P < 0.05, **P < 0.01, and ***P < 0.001. The data generated and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Results

EP300-ZNF384 promotes the expression of IL3RA in leukemia cells

To fully understand the oncogenic mechanism underlying EP300-ZNF384, we compared the expression profiles of EP300-ZNF384 fusion-positive B-ALL patients to negative patients using publicly available data from TARGET (n = 204). The volcano plot displays differentially expressed genes (DEGs), including 346 upregulated and 216 downregulated genes in EP300-ZNF384-positive patients (Fig. 1A). Functional annotation analysis of DEGs revealed that the top three enriched Kyoto Encyclopedia of Genes and Genomes pathways were hematopoietic cell lineage, cytokine-cytokine receptor interaction, and the JAK/STAT signaling pathway (Fig. 1B). Using GSEA, we identified that the cytokine-cytokine receptor interaction pathway was significantly upregulated in the EP300-ZNF384-positve patient samples (Fig. 1C). Furthermore, we found that 12 cytokine-cytokine receptor interaction pathway genes were significantly upregulated in EP300-ZNF384-positive patients (Fig. 1A and D). These data indicate that the EP300-ZNF384 fusion may promote the expression of key cytokines or cytokine receptors. In particular, both IL3 and IL3RA were significantly upregulated (Fig. 1A and D), suggesting that IL3RA signaling may be a downstream effector of the EP300-ZNF384 fusion protein.

Fig. 1
figure 1

EP300-ZNF384 promotes the expression of IL3RA in B-cell acute lymphoblastic leukemia (B-ALL). (A) Volcano plot of comparative analysis between EP300-ZNF384 positive (n = 7) and negative (n = 190) patients in the TARGET-ALL-P2 cohort. Upregulated (n = 346) and downregulated (n = 216) genes are colored by red and green (fold change > 2 and P < 0.01). The labeled texts showed the upregulated genes in the cytokine-cytokine receptor interaction pathway. (B) Bar plot of enriched Kyoto Encyclopedia of Genes and Genomes gene sets using the differentially expressed genes (fold change > 2 and P < 0.01) from the comparison of the EP300-ZNF384 positive and negative patients in the TARGET-ALL-P2 cohort. (C) Representative GSEA plot of EP300-ZNF384-positive patients compared with EP300-ZNF384-negative patients. The normalized enrichment score (NES) and nominal P-values are shown in the graph. (D) Heatmap of 12 upregulated genes in the cytokine-cytokine receptor interaction pathway in EP300-ZNF384-positive patients. IL3 and IL3RA are labeled with blue. (E) The empty vector, EP300-ZNF384, and ZNF384 were overexpressed in NALM-6 cells through lentivirus-mediated gene transfer. Puromycin (2 µg/mL)-selected cells were collected for RT-qPCR. The expression of EP300-ZNF384 together with those of 12 upregulated genes in the cytokine and cytokine receptor interaction pathway were determined relative to ACTB expression. Data represent the mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001. (F) Immunoblotting analysis of the EP300-ZNF384 fusion and wild-type ZNF384 proteins in the empty vector, EP300-ZNF384, and ZNF384-overexpressing NALM-6 cells. (G) Surface expression of CD123 was evaluated by flow cytometry on NALM-6 cells with the empty vector, EP300-ZNF384 fusion gene, and ZNF384. The proportion of CD123-positive cells and the median fluorescence intensity of CD123 were quantified. *P < 0.05, ***P < 0.001

Full size image

To confirm the specific cytokine signaling pathway involved in EP300-ZNF384-induced leukemogenesis, the EP300-ZNF384 fusion gene, wild-type ZNF384 (ZNF384), and empty vector (vector) were expressed in NALM-6 cells. The expression of EP300-ZNF384 was confirmed by RT-qPCR (Fig. 1E) and the 130 kDa EP300-ZNF384 fusion protein was detected by immunoblotting for ZNF384 (Fig. 1F). RT-qPCR was performed to evaluate the expression of the 12 upregulated genes within the cytokine-cytokine receptor interaction pathway. Overexpression of EP300-ZNF384 resulted in significant upregulation of FLT3, IL3RA, IL3, ACVR1B, TNFRSF10A, and CLCF1 and significant downregulation of IL9R and CXCL3 in NALM-6 cells (P < 0.05, Fig. 1E). Among the six upregulated genes, FLT3 and CLCF1 have been previously reported to be upregulated in EP300-ZNF384-positive cells (P < 0.001). Notably, the overexpression of EP300-ZNF384 led to the most significantly upregulated expression of IL3RA (∼ 6.5-fold, P < 0.0001) in NALM-6 cells (Fig. 1E). Similar results were also observed in BALL-1, Daudi, and KG-1α cells with EP300-ZNF384, supporting that EP300-ZNF384 fusion protein promotes the expression of IL3RA in both B and myeloid leukemia cell lines (Fig. S1A-C). Further, we detected the expression of IL3Rα (CD123) on the cell membrane by flow cytometry. We detected a basal expression of CD123 (9.28%) in vector expressing NALM-6 cells (Fig. 1G). In response to EP300-ZNF384 expression, we observed a 2.3-fold increase in the CD123+ subpopulation and a 1.29-fold increase in CD123 fluorescence intensity in NALM6 cells (Fig. 1G). A similar observation was made for BALL-1 cells (Fig. S1D). Moreover, we observed that EP300-ZNF384 promotes the expression of IL3 in NALM-6 cells (∼ 2.6-fold, p < 0.0001, Fig. 1E), but not in BALL-1, Daudi, and KG-1α cells. These results demonstrate that EP300-ZNF384 promoted the expression of IL3RA in a wide range of leukemia cells.

Increased expression of IL3RA by EP300-ZNF384 promotes B-ALL cell proliferation with or without IL-3

IL3Rα functions as a specific receptor of the IL-3 ligand, which is necessary for IL-3 activated cell proliferation [24]. To investigate whether increased expression of IL3RA was responsible for IL-3 induced cell proliferation of EP300-ZNF384-positive cells, we performed a CFC assay. In the absence of IL-3, both EP300-ZNF384 and ZNF384 enhanced the colony forming ability of NALM-6 cells, as evidenced by elevated colony number (average colony number: Vector, 0.5, EP300-ZNF384, 11, ZNF384, 7.25) and size (average colony size: Vector, 23.13 μm, EP300-ZNF384, 31.2 μm, ZNF384, 30.99 μm) (Fig. 2A-C). Notably, EP300-ZNF384-expressing NALM-6 cells displayed stronger colony-forming ability than ZNF384-expressing NALM-6 cells, as evidenced by significantly increased colony number (P < 0.0001). In the presence of IL-3, the colony formation was significantly elevated in both EP300-ZNF384- and vector-expressing NALM-6 cells (average colony number: Vector, 8, EP300-ZNF384, 21.75, ZNF384, 7.75; average colony size: Vector, 30.94 μm, EP300-ZNF384, 43.78 μm, ZNF384, 30.09 μm). NALM-6 cells harboring EP300-ZNF384 showed the strongest colony-forming ability in response to IL-3 treatment (Fig. 2B-C). Western blotting confirmed that IL-3 specifically increased the phosphorylation of STAT5 in EP300-ZNF384-positive NALM-6 cells (Fig. S2A). These data indicate that EP300-ZNF384 induced IL3RA elevation to make it sensitive to IL-3 stimulation, resulting in augmented activation of STAT5 and enhanced colony formation in B-ALL cells.

Fig. 2
figure 2

Increased expression of IL3RA by EP300-ZNF384 promotes B-cell acute lymphoblastic leukemia (B-ALL) cell proliferation with or without IL-3. (A) NALM-6 cells overexpressing EP300-ZNF384, ZNF384, and the empty vector were subjected to a colony forming cell (CFC) assay in the presence or absence of IL-3 (10 ng/mL). Scale bar: 50 μm. Data derived from A was used for quantitative analysis of colony size (B) and colony number (C). *P < 0.05, **P < 0.01, ***P < 0.001, ns, no significance. (D) The NALM-6 cells were co-transfected with the shRNA vector (shIL3RA or empty vector) and overexpression vector (EP300-ZNF384 or empty vector) via lentivirus-mediated gene transfer. Hygromycin B (400 µg/mL) and puromycin (2 µg/mL) double selected cells were subjected to CFC assay. Scale bar: 100 μm. (E) Data derived from D were used for quantitative analysis of colony size (diameter) and number. *P < 0.05, **P < 0.01, ***P < 0.001, ns, no significance

Full size image

To understand the functional involvement of IL3RA in EP300-ZNF384-positive B-ALL cells, we used a doxycycline-inducible shRNA expression system to knock down IL3RA (shIL3RA-1, shIL3RA-2, shIL3RA-3, and shIL3RA-4) in NALM-6 cells. An empty vector served as a control. The knockdown efficiency of IL3RA was assessed by RT-qPCR. Doxycycline supplementation significantly reduced the expression of IL3RA in both shIL3RA-2- and shIL3RA-4-expressing NALM-6 cells (Fig. S2B). Then, an EP300-ZNF384-expressing vector as well as control vector was transfected into shIL3RA-expressing NALM-6 cells. We performed a CFC assay to evaluate the proliferative ability of EP300-ZNF384- or vector-expressing NALM-6 cells with or without IL3RA. Due to the basal expression of CD123, IL-3 modestly promote the proliferation of vector expressing NALM-6 cells (Fig. S2C). The knockdown of IL3RA (shIL3RA-2 and shIL3RA-4) by doxycycline only induced a modest reduction in colony formation in empty vector expressing-NALM-6 cells, with or without IL-3 (middle and right panels of Fig. S2C). In pLKO and EP300-ZNF384-expressing NALM-6 cells, IL-3 significantly upregulated colony formation (left panel of Fig. 2E). Consistent with this, interference with IL3RA completely abolished the colony formation ability of EP300-ZNF384-expressing NALM-6 cells with or without IL-3 (middle and right panels of Fig. 2E). In shIL3RA-4-expressing NALM-6 cells, we confirmed that doxycycline supplementation efficiently reduced the expression of IL3RA induced by EP300-ZNF384 (Fig. S2D). Furthermore, silencing of IL3RA abrogated the IL-3-induced phosphorylation of STAT5 at Y694 in EP300-ZNF384 expressing NALM-6 cells (Fig. S2E). These results collectively indicate that EP300-ZNF384-expressing cells acquired a growth advantage by upregulating the expression of IL3RA.

Conditional knockdown of IL3RA impairs the expansion of EP300-ZNF384 positive B-ALL cells in mice

To further define the oncogenic role of IL3RA in EP300-ZNF384-expressing B-ALL cells in vivo, EP300-ZNF384-positive NALM-6 cells expressing shIL3RA-4 or a control vector were transplanted into immunodeficient NYG mice (Fig. 3A). Mice that received EP300-ZNF384-expressing NALM-6 cells rapidly developed leukemia, with a median survival time of 26 d. However, IL3RA depletion remarkably extended the survival of recipient mice, with a median survival of 36 d (n = 6, P < 0.01; Fig. 3B). Four mice were sacrificed at 20 d after transplantation, and hematopoietic tissues, such as BM from the femurs, spleen, and liver were harvested for leukemia expansion and infiltration analysis. EP300-ZNF384 positive NALM-6 cells harboring shIL3RA exhibited a much lower leukemia burden in the mouse BM than cells harboring the empty vector EP300-ZNF384 (n = 4, P = 0.0001, Fig. 3C-D). Consistently, our results showed that IL3RA depletion markedly reduced the infiltration of EP300-ZNF384-expressing NALM-6 cells into the spleen and liver (Fig. 3E-G). Morphological analysis revealed excessive numbers of immature blasts in the BM smears of mice that received EP300-ZNF384-expressing NALM-6 cells. In the shIL3RA group, both mature myeloid cells and immature blasts were observed, suggesting that IL3RA depletion impaired the expansion of EP300-ZNF384-expressing NALM-6 cells in the BM and the infiltration in liver and spleen. Collectively, these results indicate that the expression of IL3RA is crucial for the propagation of EP300-ZNF384-expressing NALM-6 cells in vivo.

Fig. 3
figure 3

Conditional knockdown of IL3RA impairs the engraftment of EP300-ZNF384-positive B-cell acute lymphoblastic leukemia (B-ALL) cells in mice. (A) The schematic diagram of the in vivo experiment. NALM-6 cells expressing EP300-ZNF384 or EP300-ZNF384 and shIL3RA were transplanted into immune-deficient NOD⁃PrkdcscidIl2rgnull mice. Mice were then treated with doxycycline (Dox, 2 mg/mL) in drinking water. After 20 d, four mice were sacrificed and femur bone marrow cells were collected for leukemia cell analysis. The remaining six mice were euthanized upon signs of disease. (B) Kaplan–Meier survival curve of mice injected with EP300-ZNF384 or EP300-ZNF384 and shIL3RA-expressing NALM-6 cells. **P < 0.01 (C) hCD19 fractions within the femur bone marrow of each mouse. (D) Column plots summarizing data from all animals analyzed as in C. ***P < 0.001. (E) Changes in the spleen in EP300-ZNF384-expressing NALM-6-transplantable mice in response to IL3RA knockdown. (F) Spleen weights are presented by column plots. ***P < 0.001. (G) The leukemic invasions in spleen and liver were analyzed by hematoxylin and eosin staining. Scale bar: 50 μm. (H) Wright–Giemsa staining of BM cells from mice with NALM-6 cells expressing EP300-ZNF384 or EP300-ZNF384 and shIL3RA. Scale bar: 20 μm

Full size image

EP300-ZNF384 fusion protein transactivates the promoter activity of IL3RA

ZNF384 is a C2H2-type zinc-finger transcription factor. Thus, there is reason to believe that EP300-ZNF384 may target sites carrying the ZNF384 DNA-binding motif. To determine whether EP300-ZNF384 regulates IL3RA gene expression at transcriptional level, the promoter of IL3RA (from − 1293 to + 110) was cloned into the pGL3 reporter construct. Dual-luciferase reporter constructs were co-transfected with EP300-ZNF384, ZNF384, or an empty vector into HEK-293T cells. We found that the luciferase activity was increased in HEK-293T cells bearing EP300-ZNF384 and ZNF384 compared to the empty vector (Fig. 4A). The EP300-ZNF384 fusion displayed stronger transcriptional activity of IL3RA than wild-type ZNF384, indicating that the fusion of EP300 to ZNF384 enhanced the transcriptional activity of ZNF384 on IL3RA promoter. We found that EP300-ZNF384 activated IL3RA transcription in a dose-dependent manner, further demonstrating that the EP300-ZNF384 fusion protein transactivated the promoter activity of IL3RA (Fig. 4B).

Fig. 4
figure 4

EP300-ZNF384 fusion protein transactivates the promoter activity of IL3RA (A) EP300-ZNF384- or ZNF384-expressing vectors were co-transfected with the PGL3-IL3RA-promoter and pRL-TK into HEK-293T cells. After 24 h, dual luciferase assay was performed. Data represents the mean ± SD. **P < 0.01, ***P < 0.001. (B) Increasing doses of EP300-ZNF384-expressing vectors and the luciferase reporter vectors were co-transfected into HEK-293T cells. Luciferase analysis for the IL3RA promoter was performed. ***P < 0.001. (C) Diagram showing the putative binding sites of ZNF384 in IL3RA promoter (orange). (D) Truncated mutations of the IL3RA promoter are shown. Numbering is indicated with respect to the transcriptional start site. (E) Luciferase analysis for the truncated IL3RA promoters was performed. *P < 0.05, **P < 0.01, ***P < 0.001, ns, no significance. (F) Schematic presentation of the IL3RA promoter mutations used in this study. The mutated IL3RA promoter reporter along with pRL-TK were co-transfected with the empty vector, EP300-ZNG384, or ZNF384-expressing vectors into HEK-293T cells for 24 h. Dual-luciferase assay was performed. ***P < 0.001, ns, no significance. (G) HEK-293T cells were transfected with flag-tagged EP300-ZNF384- and ZNF384- expressing vectors for 48 h. Chromatin fragments of these cells were immunoprecipitated with a flag or nonspecific antibody (ns-Ab). PCR was conducted with primers corresponding to the genomic IL3RA and MMP7 promoter sequences. MMP7 served as a positive control

Full size image

To determine the region of IL3RA promoter regulated by the EP300-ZNF384 fusion, we performed motif prediction using the JASPAR database and found that EP300-ZNF384 could potentially bind to three regions; namely -234 bp/-223 bp, -735 bp/-731 bp, and -1057 bp/-1035 bp; in the human IL3RA promoter (Fig. 4C). Dual luciferase analysis showed that EP300-ZNF384 failed to activate the transcription at the IL3RA promoter when the − 331 bp/-215 bp region of IL3RA promoter was deleted (Fig. 4D, E). To identify the critical nucleotides within this region, deleted or substituted mutations of the -234 bp/-222 bp were introduced into the luciferase reporter vector (the upper panel of Fig. 4F). Both deletion of the -234 bp/-222 bp and replacement of the sequence with CCTGTGTCAGAG shifted the promoter from active to inactive status (the lower panel of Fig. 4F). The results of ChIP assay further indicated that EP300-ZNF384 interacted with the -234 bp/-222 bp region of the IL3RA promoter (Fig. 4G). These data collectively demonstrate that the EP300-ZNF384 fusion protein transactivates the promoter activity of IL3RA by binding to the − 234 bp/-222 bp region.

EP300-ZNF384 elevates the expression of IL3Rα (CD123) in B precursor cells from healthy individuals

To fully understand whether EP300-ZNF384 promotes the expression of IL3RA in the initial stages of leukemogenesis, hCD19+ B precursor cells were isolated from healthy individual-derived PBMCs and transfected with lentiviruses expressing EP300-ZNF384, ZNF384, or the vector. Compared with the empty vector, EP300-ZNF384 strikingly promoted the expression of IL3RA in hCD19+ cells derived from three healthy individuals (Fig. 5A). The CD123 expression was then determined by flow cytometry. Less than 1% of the samples was stained with negative isotype controls, which was set as the gating region. In comparison with the empty vector, EP300-ZNF384 elevated the expression of CD123 on cell surface in hCD19+ cells derived from 5 healthy individuals (Fig. 5B and Fig. S3A-B).

Fig. 5
figure 5

EP300-ZNF384 elevates the expression of CD123 in B precursor cells. (A) CD19-positive cells were enriched with magnetic beads from peripheral blood mononuclear cells of healthy volunteers. Empty vector, EP300-ZNF384-, and ZNF384-expressing vectors were transfected into CD19-positive cells by lentivirus-mediated gene transfer. After 48 h, cells were collected for reverse-transcription quantitative PCR assay to assess the transcription of IL3RA. Data are represented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ns, no significance. (B) Representative flow cytometry dot plots illustrating CD123 fractions in CD19 positive B precursor cells in response to EP300-ZNF384 or ZNF384 expression

Full size image

Our previous RNA sequencing analysis demonstrated that the expression of IL3 was significantly upregulated in B-ALL patients with EP300-ZNF384 fusion lesions compared to that in B-ALL patients without EP300-ZNF384 fusion lesions (Fig. 1D). We demonstrated that EP300-ZNF384 induced the expression of IL3 in NALM-6 cells. However, the expression of IL3 is poorly detected by RT-qPCR in other leukemia cell lines. We then determined expression patterns during hematopoiesis using the BloodSpot website (https://servers.binf.ku.dk/bloodspot/). In the normal human hematopoiesis (DMAP) database, IL3 was mildly expressed in early- and pro-B cells (Fig. S3C). Furthermore, we performed RT-qPCR and ELISA assays to investigate whether EP300-ZNF384 induced the expression of IL3 in CD19+ B precursor cells. Our data demonstrate that both the expression and secretion of IL3 were promoted by EP300-ZNF384 in B precursor cells (Fig. S3D, E). Our results demonstrated that EP300-ZNF384 promotes B-ALL cell expansion by activating the expression of IL3Rα and the secretion of IL-3 at the initial stage of leukemogenesis. Further, this implies that EP300-ZNF384 boosts IL3RA autocrine signaling to confer a growth advantage to B-ALL cells.

EP300-ZNF384 positive B-ALL cells are more sensitive to doxorubicin in vitro and in vivo

A previous clinical report showed that approximately 87.5% of patients who received VDCLP or a VDCLP-like regimen of induction chemotherapy achieved complete remission [31]. We speculate that chemotherapeutic agents within the VDCLP regimen (vincristine, doxorubicin, L-asparaginase, and dexamethasone were used in this research) might sensitize B-ALL leukemia cells with EP300-ZNF384. We then performed the CCK-8 assay to assess the sensitivity of NALM-6 cells bearing the EP300-ZNF384 fusion, wild-type ZNF384, or Vector to chemotherapeutic agents. Our results revealed that both doxorubicin and pegaspargase preferentially inhibited the cell viability of EP300-ZNF384 expressing-NALM-6 cells in a dose-dependent manner (Fig. S4A-D). Of note, doxorubicin reduced the viability of EP300-ZNF384-expressing NALM-6 cells by about 10% over a wide range of concentrations (0.078–0.31 µM), as compared with cells with the empty vector (Fig. S4A). Furthermore, EP300-ZNF384-expressing NALM-6 cells were selectively sensitive to doxorubicin in the presence or absence of IL-3 (Fig. 6A). Apoptosis was determined by dual staining with FITC-labeled Annexin V and PI. Our results demonstrated that doxorubicin significantly induced the apoptosis of EP300-ZNF384 expressing-NALM-6 cells, as evidenced by the elevated Annexin V+ proportion (Fig. 6B, C). Next, we examined the in vivo therapeutic potential of doxorubicin by treating NYG mice transplanted with NALM-6 cells expressing the EP300-ZNF384, ZNF384, or empty vectors (Fig. 6D). In line with our ex vivo observations, EP300-ZNF384-expressing NALM-6 cells responded remarkably to doxorubicin treatment in vivo, resulting in a significantly decreased percentage of hCD19+ cells in mouse BM and a reduced leukemia burden in the liver (Fig. 6E, F, and S5). These findings support the therapeutic value of doxorubicin in targeting EP300-ZNF384 positive B-ALL cells.

Fig. 6
figure 6

EP300-ZNF384 positive B-cell acute lymphoblastic leukemia (B-ALL) cells are more sensitive to doxorubicin in vitro and in vivo. (A) Empty vector-, EP300-ZNF384-, and ZNF384-expressing NALM-6 cells were treated with increasing doses of doxorubicin for 48 h in the presence or absence of IL-3. Cell viability was determined by CCK-8 assay. Data are represented as mean ± SD. ***P < 0.001. (B) Empty vector-, EP300-ZNF384-, and ZNF384-expressing NALM-6 cells were treated with indicated doses of doxorubicin for 48 h. Cell apoptosis was detected by dual Annexin V-FITC/PI staining. Representative flow cytometry dot plots are presented (C) Column plots represents the proportion of Annexin V-positive cells. Data were represented as mean ± SD (n = 3). Unpaired t-test. *P < 0.05, ***P < 0.001. (D) Schematic describing generation of CDX mouse model by engrafting NOD⁃PrkdcscidIl2rgnull (NYG) mice with NALM-6 cells harboring the empty vector, EP300-ZNF384, or ZNF384, followed directly by in vivo treatment with doxorubicin (8 mg/kg, intraperitoneally). (E) Percentage of human CD19+ subpopulations in the bone marrow of NYG mice engrafted with empty vector or EP300-ZNF384 or ZNF384 expressing NALM-6 cells, followed by in vivo doxorubicin treatment. (F) Leukemia involvement in the liver was determined by H&E staining. Scale bar indicates 50 μm

Full size image

Discussion

The EP300-ZNF384 fusion gene is one of the most common genetic lesions within B-ALL patients and induces leukemogenesis in murine transplantation models [9, 10]. Here, we demonstrated that the EP300-ZNF384 fusion protein substantially induces the expression of IL3Rα, which contributes to an acquired growth advantage of B-ALL cells in the presence or absence of IL-3. Mechanistically, the EP300-ZNF384 fusion protein transactivates the promoter activity of IL3RA by binding to an A-rich sequence localized at -222/-234 of the IL3RA gene. We further demonstrated that doxorubicin treatment sensitized EP300-ZNF384 expressing B-ALL cells.

IL3Rα (also known as CD123) regulates the proliferation, survival, and differentiation of hematopoietic cells upon binding with IL-3. IL3Rα is highly expressed in multiple types of leukemia such as AML and ALL and has been recognized as a specific marker for leukemic stem cells in AML patients [20,21,22,23, 32]. The elevated expression of IL-3Rα is associated with enhanced blast proliferation and a poor prognostic phenotype in patients with AML [24,25,26]. Our current work disclosed that EP300-ZNF384 increased the expression of IL3Rα, which resulted in an enhanced proliferative activity of B-ALL cells. IL-3 is a tightly regulated pleiotropic cytokine produced mainly by activated T lymphocytes [33]. Numerous studies demonstrated that IL-3 promotes proliferation and colony formation of primary leukemic cells in vitro, supporting that aberrant IL-3 signal plays significant roles in the process of leukemogenesis [34, 35]. In addition, we showed that EP300-ZNF384 promoted the expression and production of IL-3 in B precursor cells, indicating that both autocrine and paracrine IL-3 may be related to the enhanced growth advantage of EP300-ZNF384-positive B-ALL cells. Collectively, our findings suggest the oncogenic mechanism underlying EP300-ZNF384 is involved in aberrant IL3Rα signaling.

The expression of IL3RA is regulated by transcription factors and epigenetic regulators [36]. A recent study revealed that the NUP98-NSD1 fusion gene induces CD123 expression in 32D cells [37]. Clinical data showed that patients with ZNF384 rearrangements have significantly higher CD123 expression than those with non-ZNF384 rearrangements (BCR-ABL, KMT2A rearrangement, and other B-ALL) [27]. To date, there has been no evidence that oncogenic fusion proteins directly regulate the expression of IL3RA. In the present study, we demonstrated that EP300-ZNF384 transcriptionally promotes the expression of IL3RA, which enhanced the proliferative activity of B-ALL cells. We identified a EP300-ZNF384 fusion protein binding sequence at the IL3RA promoter and proved the fusion of EP300 with ZNF384 increased the transaction activity at the IL3RA promoter, leading to a much higher expression of IL3Rα on B-ALL cell membrane. Our findings suggest that CD123 is a direct downstream target of the EP300-ZNF384 fusion protein and could serve as a valuable marker of EP300-ZNF384 fusion. Furthermore, our data confirmed that the previously proposed FCM scoring system, including CD10, CD13, CD33, and CD123, is reliable for distinguishing ZNF384 rearrangements from non-ZNF384 rearrangements with high sensitivity and specificity.

Targeting CD123 via CAR-engineered T cells, neutralizing antibodies, and IL3RA antibodies coupled to toxins has been shown to have beneficial effects on the survival of AML models [38,39,40,41]. The CD123-targeted antibody-drug conjugate demonstrated promising selective activity in preclinical models of B-ALL [42]. Our recent study showed that depletion of IL3RA completely abolished the growth advantage conferred by EP300-ZNF384 in B-ALL cells and impaired the engraftment of EP300-ZNF384-positive B-ALL cells in mice. Although current work did not discuss the anti-leukemic activity of CD123-targeted therapy, given the specific expression pattern of CD123 in tumors with EP300-ZNF384 rearrangement, our results support CD123 as a promising therapeutic target for patients with EP300-ZNF384 fusion, suggesting that IL3Rα-targeting therapeutics may be effective for patients with EP300-ZNF384 rearrangement. Studies on the anti-leukemia activity of CD123-targeted therapeutics are urgently needed in the next few years.

ALL treatment has mainly been based on chemotherapy and hematopoietic stem cell transplantation (HSCT) in high risk [43]. B-ALL patients with EP300-ZNF384 have been found to achieve a higher remission rate than those without it [5, 7]. Thus, we speculated that certain agents within the chemotherapy regimens may sensitize EP300-ZNF384-positive leukemia cells. We performed drug screening using vincristine, doxorubicin, L-asparaginase, and dexamethasone (VDCLP) regimen and found that doxorubicin displayed surprising inhibitory effects on EP300-ZNF384 positive cells. Doxorubicin is an anthracycline antibiotic used in a wide range of cancers, including carcinomas, sarcomas, and hematological cancers [44]. Doxorubicin reportedly intercalates with their aglycone moieties between neighboring DNA base pairs, ultimately ceasing the replication and transcription of DNA [45]. Given its high affinity for DNA, doxorubicin may interrupt the growth of EP300-ZNF384-positive leukemia cells in vitro and in vivo (Fig. 6). Previous transcriptome profiling studies have reported that the DNA repair pathway is significantly decreased in EP300-ZNF384 positive patients [8]. It is likely that doxorubicin exerts a therapeutic effect by inhibiting the transcription of IL3RA and weakening the DNA damage repair process. Our results indicated that B-ALL patients with EP300-ZNF384 may benefit from chemotherapy regimens that include anthracyclines. Furthermore, an optimized chemotherapy protocol, including high-dose and intensive delivery of doxorubicin, may achieve better outcomes in patients with EP300-ZNF384. Collectively, our findings indicate that targeted therapies involving anthracyclines and CD123-targeted agents may produce a favorable outcome in patients with EP300-ZNF384 and could be developed as a standard treatment for such cases.

Notably, this study has limitations that can be overcome in further research. First, we did not explain why EP300-ZNF384 positive B-ALL cells are selectively sensitive to doxorubicin; thus, further research should investigate the mechanism of doxorubicin action on EP300-ZNF384 positive B-ALL cells. Moreover, we did not evaluate the anti-leukemic role of targeting CD123 in EP300-ZNF84 positive B-ALL; thus, further research should explore therapeutic potential of CD123-targeted therapies including CAR-engineered T cells, neutralizing antibodies, and IL3RA antibodies coupled to toxins. Further research should also attempt to explore the regulatory mechanism of EP300-ZNF384 on IL3 expression.