Aberrant activation of RPB1 is critical for cell overgrowth in acute myeloid leukemia

Experimental Cell Research xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Experimental Cell Research journal homepage: www.elsevier.com/locate/yexcr

Aberrant activation of RPB1 is critical for cell overgrowth in acute myeloid leukemia Qingfeng Yua,b,1, Ying Xua,1, Haifeng Zhuangc,1, Zhaoxing Wua, Lei Zhanga, Jinfan Lid, Linlin Yanga, Bowen Wua, Ping Wanga, Xuzhao Zhanga, Xiaoxian Gane, Yun Lianga, Shu Zhenga, Xiaofang Yua,c, Ying Gua,f,∗∗, Rongzhen Xua,g,∗ a Department of Hematology and Cancer Institute (Key Laboratory of Cancer Prevention and Intervention, China National Ministry of Education), The Second Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou, 310009, China b Department of Hematology, the First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China c Department of Hematology, the First Affiliated Hospital of Zhejiang Chinese Medical University, Hangzhou 310009, China d Department of Pathology, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang 310009, China e Zhejiang Academy of Medical Sciences, Hangzhou, 310012, China f Institute of Genetics, Zhejiang University and Department of Genetics, Zhejiang University School of Medicine, Hangzhou, 310058, China g Institute of Hematology, Zhejiang University, Hangzhou, 310009, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Acute myeloid leukemia RPB1 POLR2A Oncogene Apoptosis

Acute myeloid leukemia (AML) is a group of highly aggressive malignancies with a 5-year overall survival of less than 40%. Cell overgrowth with defective apoptosis is a hallmark of AML, but little is known about how it occurs. Here, we show that aberrant activation of the largest subunit of RNA polymerase II (RPB1) encoded by POLR2A gene is critically involved in this hallmark. We retrospectively analyzed the expression profiles of POLR2A and RPB1 in a panel of AML cell lines, primary AML patients and peripheral blood samples. Meanwhile, correlation analysis was used to explore the correlation between the expression of RPB1 with tumor burden and overall survival time in untreated AML samples. RNA-Seq approach was performed to identify the differentially expressed genes between RPB1 silencing AML cells with control cells after knocking out RPB1. Furthermore, orthotopic AML models were established with RPB1 silencing and control cells to investigate the effects of RPB1 protein level on leukemia cell growth. In most AML patients, RPB1 was aberrantly activated and closely associated with poor prognosis, but not in normal hematopoietic cells. Global transcriptomic analysis revealed that POLR2A knockout strongly impaired growth of AML cells by selectively depleting a substantial set of AMLrelated oncogenic and anti-apoptosis genes such as MYC, RUNX2, MEIS1, CDC25A and BCL-2. Silencing RPB1 by genetic technology led to a potent regression of human refractory AML in mouse models. These findings reveal that dysregulated RPB1 is a central oncogenic hub that drives overgrowth by hijacking an array of oncogenic and anti-apoptosis factors. Targeting RPB1 is a potential therapeutic for treating AML.

1. Introduction Acute myeloid leukemia (AML) is a group of highly aggressive malignancies with molecular and clinical heterogeneity, which is characterized by a blockade in differentiation of hematopoietic stem cells and a clonal expansion of myeloid blasts in the bone marrow and peripheral blood. Although a number of therapies such as the standard

“7 + 3” induction therapy (DA regimen) are effective in killing leukemic cells in AML, the vast majority of AML patients suffer from lethal disease relapse within 3 years [1–4]. One of the biggest challenges in treating AML is the inability of conventional therapies to kill rapidly proliferating tumor cells with defective apoptosis, a universal hallmark of AML as other cancers [5]. Thus, elucidation of the molecular mechanisms by which cancer cells maintain rapidly proliferating ability

∗ Corresponding author. Department of Hematology and Cancer Institute (Key Laboratory of Cancer Prevention and Intervention, China National Ministry of Education), The Second Affiliated Hospital, College of Medicine, Zhejiang University. Tel.: +86 571 87214404. ∗∗ Corresponding author. Department of Hematology and Cancer Institute (Key Laboratory of Cancer Prevention and Intervention, China National Ministry of Education), The Second Affiliated Hospital, College of Medicine, Zhejiang University. Tel.: +86 571 87214404. E-mail addresses: [email protected] (Y. Gu), [email protected] (R. Xu). 1 These authors have contributed equally to this work.

https://doi.org/10.1016/j.yexcr.2019.111653 Received 30 July 2019; Received in revised form 25 September 2019; Accepted 27 September 2019 0014-4827/ © 2019 Published by Elsevier Inc.

Please cite this article as: Qingfeng Yu, et al., Experimental Cell Research, https://doi.org/10.1016/j.yexcr.2019.111653

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

2.3. Tissue culture

and drug resistance is expected to provide a basis for developing more potent and safer therapies for curing AML. Our previous studies have identified oncogenic Shp2 [6], CaMKIIγ [7,8]and the viral Np9 [9] oncogene as critical regulators for overgrowth and apoptosis resistance of myeloid leukemia. Moreover, we have demonstrated that CaMKIIγ plays a critical role in stabilization of the c-Myc protein in T cell lymphoma [10]. We hypothesize that there may be a central oncogenic hub that drives overgrowth and drug resistance by regulating an array of oncogenic and anti-apoptosis factors. Transcription is a “vital process” required for all living cells [11]. However, cancer cells need high levels of transcription for maintaining rapid proliferation and survival, and dysregulated transcription is universal in cancer [11–13]. For instance, RNA polymerase I (RNPI) transcriptional overactivity was thought to be essential for the survival of hematological tumor cells and can be therapeutically targeted in vivo [14,15]. RNPIII activity is increased in tumor cells compared with normal cells [16]. In contrast to RNPI and RNP III overactive in cancer, heterozygous loss of POLR2A, which encodes the largest subunit of RNPII (RPB1), is frequently observed in human colorectal and prostate cancers [17,18], although it is thought to be required for transcribing oncogenic genes and anti-apoptotic factors maintaining rapid growth and apoptosis resistance of tumor cells [19–21]. In addition, POLR2A mutation was also reported in a distinct subset of meningiomas [22]. However, RPB1 overactivity has not previously been linked to human AML and other cancer. Here, we analyzed the expression profiles of POLR2A mRNA and itscoding protein RPB1 in AML, as well as their correlations with outcome of AML patients. RNA-Seq was performed to identify RPB1-mediated genes. We used genetic approaches to investigate the role and mechanism of RPB1 in the growth of cancer cells in refractory AML models in vitro and in vivo. Using this strategy, we have identified and validated dysregulated RPB1 as a key oncogenic hub that drives overgrowth by coordinated hijacking an array of oncogenic and anti-apoptosis genes. We further demonstrate that targeting RPB1 is a potential therapeutic for treating AML.

The AML cell lines KG-1, KG-1a were cultured in Iscove's modified Dulbecco medium, Kasumi-1, HL-60, NB4, MOLM-13, THP-1 were maintained in RPMI 1640 media. HEK293 and 293T cells were cultured in Dulbecco's modified Eagle medium. All media were supplemented with 10% fetal Bovine serum (FBS) at 37 °C, 5% CO2 humidified incubator. 2.4. Isolation mononuclear cells and CD34+ cells Granulocyte colony stimulating factor (G-CSF) mobilized bone marrow stem cell samples were obtained as excess material from normal donors for allogeneic transplantation. AML primary samples and healthy donors were obtained following informed consent and with ethical approved (#IR2018001163) from the ethics committees of the Second Affiliated Hospital, Zhejiang University School of Medicine. Isolation mononuclear cells (MNCs) were obtained from AML primary samples, healthy peripheral blood donors, bone marrow cells and cord blood samples by density gradient centrifugation using Ficoll reagent. The research was approved by the ethics committees of the Second Affiliated Hospital, Zhejiang University School of Medicine. CD34+ cells were isolated by positive selection for the cell surface marker CD34 using MojoSort™ Nanobeads (BioLegend) in accordance with the manufacturer's instructions. 2.5. Animal transplantation and in vivo drug studies

2. Material and methods

Animal studies were approved by the Zhejiang Chinese Medical University Animal Care and Welfare Committee (#10616). For human refractory AML orthotopic model and drug treatment: 2 × 106 MOLM13 cells with Dox-inducible POLR2A-KO vector or control cells were injected into female NSG mice (7-weeks) through the tail vein. After detecting obvious tumor signal by bioluminescence imaging using an IVIS 100 bioluminescence/optical imaging system (Xenogen, Alameda, CA), mice received DOX to initiate POLR2A silencing via oral gavage. For Cytarabine treatment, the mice were received either vehicle or Cytarabine (50 mg/kg) via intraperitoneal injection once a day for 7 days.

2.1. Construction of POLR2A sgRNA lentivirus

2.6. Reagents and antibodies

Lentivirus vector pCW-Cas9 (doxycycline inducible; #5066) and pLX-sgRNA (lentiviral vector, #50662) were purchased from Addgene. The lentivirus vector and lentivirus packaging vectors psPAX2 and pMD2.G were transfected into 293T cells using PolyJet™ DNA In Vitro Transfection Reagent (Signagen). Reagent (Signagen). Lentiviral supernatant was produced and harvested after 48–72 h. The sequences of sgRNA to target human POLR2A as follows: 5′-GCTTCAGTTCATCCGGACTC-3’; 5′-CAGGGGGTGATTGA GCGGAC-3’; 5′-CGGCCTCCCTCAGTCGTCTC-3’; 5′-GGCCGCTGCCAAA CATGTGC-3’.

Doxorubicin, Blasticidin S, puromycin and DMSO were purchased from Sigma Aldrich. Methyl-thiazol-tetrazolium (MTT) were purchased from Sangon Biotech. RPB1 (F-12) antibody was purchased from Santa Cruz Biotechnology. The RPB1 (phosphor S2 193468), c-Myc (32072), Bcl-2 (182858) antibodies were purchased from Abcam. Anti-GAPDH (60004-1-Ig) was purchased from Proteintech Group and the horseradish peroxidase HRP-conjugated secondary antibodies were acquired from Huabio. 2.7. Colony-formation assay Cells (50–200 cells/well) were cultured in a 6-well plate using RPMI-1640 medium at 37 °C, 5% CO2 humidified incubator. Following the formation of sufficiently large clones, cells were stained with 0.1% crystal violet. The number of colony-forming units were generated and counted.

2.2. Construction of POLR2A KO cells To generate Molm-13 cells with POLR2A knockout, Molm-13/Teton Cas9 cells were generated by transducing Molm-13 cells with pCWCas9 lentivirus. Subsequently, Molm-13/Tet-on Cas9 cells were transduced with POLR2A sgRNA lentivirus. After POLR2A KO lentivirus infection for 48 h, then the cells were selected with puromycin and Blasticidin S. Individual cells were plated in 96-well plates. Each subsequent clone was divided into two parts, the expression of POLR2A in determined by western blotting for a portion. Clones with loss of POLR2A protein expression were further identified by Sanger sequencing of the POLR2A genomic locus.

2.8. Apoptosis and cell cycle analysis Apoptosis was detected by Annexin V-APC/7-AAD double staining with the AnnexinV-APC/7-AAD Apoptosis Detection Kit (BD Biosciences) according to the manufacturer's instructions and analyzed with Canto-Ⅱ (BD Biosciences) using FlowJo software. For cell cycle analysis, cells were fixed in 70% ethanol before staining and treated 2

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

Fig. 1. POLR2A and RPB1 is aberrantly activated in AML, but not in normal hematopoietic cells. (A, B, D, E) Western blot analyses of RPB1 protein levels in various AML cell lines (A), RPB1 expression in primary AML samples (B), RPB1 expression in normal peripheral blood samples (D), RPB1 expression in the cord blood samples (E).(C, I) comparisons of POLR2A mRNA and RPB1 protein levels among various AML cell lines, normal PB, cord blood samples and primary AML samples by Western blot (C) and qRT-PCR (I) analyses. (F) Immunofluorescence staining analyses of RPB1 protein levels in AML cell line (MOLM-13), primary AML samples (AML-M5) and normal hematopoietic stem/progenitor cells. (G, H) Western blot analyses the expression of phosphorylated and total RPB1 in AML cell lines, quiescent or mobilized human hematopoietic stem cells, S1 (CD34+ primary AML-M5 sample), S2 (CD34+ KG-1a cells), S3 (CD34+ cord blood sample), S4 (mobilized human hematopoietic stem/progenitor sample), S5 (human hematopoietic stem/progenitor sample). Quantification of POLR2A mRNA expression by qPCR, levels were normalized to an endogenous control, β-Actin. The results were repeated at least three times. Differences with a p value < 0.05 were considered statistically significant. * for p < 0.05; ** for p < 0.01; *** for p < 0.001.

with propidium iodide. Samples were analyzed by using a BD FACS Canto-Ⅱ flow cytometer and ModFit software.

Thermo scientific) containing protease and phosphatase inhibitor (1861281, Thermo scientific), transferred to PVDF membranes (BioRad). The bound antibodies were visualized using Super signal reagents (Thermo Fisher Scientific) and quantified with imageJ software (https://imagej.nih.gov/ij/).

2.9. Western blotting Cells were collected and lysed in protein extraction reagent (78501, 3

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

2.10. Real Time-PCR analysis (qRT-PCR)

3. Results

Total RNA was extracted by using Trizol (Invitrogen) according to the manufacturer's instructions. 500 ng of total RNA was reverse transcribed to cDNA using PrimeScriptTMⅡ 1st strand cDNA Synthesis Kit (Takara). qRT-PCR were performed with SYBR® Premix Ex TaqTM II (TliRNaseH Plus) (Takara) on 7500 Real-Time PCR Systems (Applied Biosystems, USA). Gene expression level was determined by using the ΔΔcycle threshold method normalized to β-Actin. qRT-PCR primers used were as follows: POLR2A forward, 5′-TTGTATCCGTACCCACA GCA-3’; POLR2A reverse, 5′- CATGATCAGCTCCCCATTCT-3’; MYC forward, 5′- GTCAAGAGGCGAACACACAAC-3’; MYC reverse, 5′- TTG GACGGACAGGATGTATGC-3’; Bcl-2 forward, 5′- GTGGATGACTGAGT ACCTGAACCG-3’; Bcl-2 reverse, 5′- AGAGTCTTCAGAGACAGCCAG GAG-3’; CDC25A forward, 5′- CTCCTCCGAGTCAACAGATTCA-3’; CDC25A reverse, 5′- CAACAGCTTCTGAGGTAGGGA-3’; GADD45A forward, 5′-TCGGCTGGAGAGCAGAAGACC-3’; GADD45A reverse, 5′-ACA TCTCTGTCGTCGTCCTCGTC-3’; TP53 forward, 5′-GAGGTTGGCTCTGA CTGTACC-3’; TP53 reverse, 5′-TCCGTCCCAGTAGATTACCAC-3’; ACTIN forward, 5′-ACTCTTCCAGCCTTCCTTCC-3’; ACTIN reverse, 5′AGCACTGTGTTGGCGTACAG-3’. All experiments were performed in triplicate in a 20 μl reaction volume.

3.1. RPB1 is aberrantly activated in AML cells but not in normal hematopoietic cells RPB1 is defined as the heart of the cellular transcription machinery due to its requirement for transcribing all protein-coding genes [17,22,23]. To investigate the dependency of AML cells and normal hematopoietic cells on RPB1, then we analyzed RPB1 expression in pretreatment BM samples from 113 AML patients and 10 healthy PB samples. The median values of RPB1 analyzed quantitatively by imageJ in AML cell lines, primary AML patients, cord blood (CB) and peripheral blood (PB) samples were 1.49148, 0.788934, 0.0280763 and 0.048144 (Fig. 1A, B, D, E), respectively. To confirm these observations, we used immunofluorescence staining to stain RPB1 protein in AML cells and normal hematopoietic stem/progenitor cells (HSCs). Likewise, RPB1 protein was abundantly detected in both AML cell lines (MOLM-13) and primary leukemia cells (AML-M5), but very low expression was observed in normal hematopoietic stem/progenitor cells and cord blood cells (Fig. 1F, Figs. S1A–D). RPB1 contains a flexible C-terminal domain (CTD) comprising multiple repeats of the heptad consensus sequence YSPTSPS in humans. Phosphorylation of CTD residues is thought to recruit multiple co-factors that directly binds to RNAPII at sites of transcription [24]. We analyzed the phosphorylated RPB1 (p-RPB1) levels and found that the pRPB1 levels were also highly expressed in AML cell lines (Fig. 1G). Interestingly, CD34+ AML cells also expressed high levels of p-RPB1, whereas both t-RPB1 and p-RPB1 were significantly at lower levels in quiescent or mobilized human hematopoietic stem/progenitor cells (Fig. 1H). To address whether RPB1 was dysregulated in AML at the transcription level, POLR2A mRNA expression was retrospectively analyzed from 147 patients with newly diagnosed AML, 8 AML cell lines and 20 healthy PB samples as a control. The levels of POLR2A mRNA expression were significantly higher in primary AML patients than those in controls (median quantification of mRNA expression after normalization: 1.33566 versus 0.79482) (Fig. 1I). The median values of AML cell lines, primary AML samples and normal peripheral blood samples were 1.27974, 1.33566 and 0.79482, respectively. However, in contrast to previous reports that the expression levels of PORL2A and RPB1 were tightly correlated with its gene copy number in human colorectal carcinoma [17], no closely correlations were found between POLR2A mRNA and RPB1 protein levels in AML patients, suggesting the presence of post-transcriptional and post-translational events. To further validate above observations, we analyzed the POLR2A mRNA expression data from the GEPIA dataset with 5602 samples across 31 cancer types. Consistent with our observations, POLR2A mRNA was significantly overexpressed in AML samples (LAML) (N = 173) when compared with normal blood cell samples (N = 70) (Fig. 2A) (median log2 Fold Change values 6.5838 versus 5.4643). To examine whether aberrant activation of POLR2A was limited in AML, we analyzed 30 other types of tumors for POLR2A mRNA levels in TCGA dataset. We found that cholangiocarcinoma (CHOL) and thymoma (THYM) also exhibited significant increased POLR2A mRNA levels (Fig. 2B and C), but POLR2A mRNA levels in CHOL and THYM were lower than those in AML patients (Fig. S2A). By contrast, we observed that adrenocortical carcinoma (ACC), testicular germ cell tumors (TGCT) and uterine corpus endometrial carcinoma (UCS) exhibited decreased POLR2A mRNA levels as compared to corresponding non-tumor samples (Fig. 2D, E, F). Unexpectedly, there were no significant differences of POLR2A mRNA levels between tumor and nontumor tissues, among 25 other common types of tumors (Fig. S2B). These results indicate that aberrant activation of RPB1 occurs predominantly in AML, CHOL and THYM, but not in most other types of tumors. Unexpectedly, RPB1 protein is expressed at lower levels in normal hematopoietic stem/progenitor cells.

2.11. RPB1 confocal microscopy Cells were washed with PBS twice, and then fixed on the coverslips in 4% paraformaldehyde for 20 min. After permeabilized with 0.1%Trition X-100 and washed with PBST buffer, the cell samples were blocked in PBS with 5% bovine serum albumin (BSA) for 1 h, incubated with primary antibody RPB1 for overnight at 4 °C, washed with PBST buffer and treated with goat anti-mouse secondary antibody labeled with Rhodamine. Images were acquired using a Zeiss Confocal Laser Scanning Microscope 710 (LSM710, Germany).

2.12. RNA sequencing Sequencing libraries were prepared by using the NEBNext® Ultra™ RNA Library Prep Kit and sequenced on an Illumina Hiseq 4000. A total of 150 bp paired-end reads were aligned to the genome using HISAT2 version 2.1.0 and counted by using HTSeq. Differential expression was calculated utilizing the DEGseq package in R (version 1.28.0). Gene FPKMs were computed by summing the FPKMs of transcripts in each gene group.

2.13. Bioinformatics mining Gene Expression Profiling Interactive Analysis (GEPIA) database (http://gepia.cancer-pku.cn/) was mined to predict the POLR2A expression level between the tumor samples and paired normal tissues. Gene expression correlation analysis was performed for giving sets of TCGA expression data by GEPIA. The POLR2A mRNA expression on AML subtype compared to other malignant hematopoiesis subtypes or normal hematopoietic cells were analyzed by BloodSpot database (www.bloodspot.eu).

2.14. Statistical analysis Statistical analysis (Student's t-test, Mann-Whitney U test, Pearson correlation and log-rank test) were performed using Prism 7 (GraphPad Software). Differences with a p value < 0.05 were considered statistically significant. Differences were labeled as follows: * for p < 0.05; ** for p < 0.01; *** for p < 0.001. 4

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

Fig. 2. POLR2A is deregulated in AML, CHOL, THYM, ACC, TGCT and UCS. (A, B, C) TCGA data analysis reveals aberrantly activation of POLR2A mRNA levels in AML, CHOL and THYM. (D, E, F) TCGA data analysis reveals decreased POLR2A mRNA levels as compared to corresponding non-tumor samples in the ACC, TGCT, and UCS.

AML patients (RPB1low ≤ median) had a significantly longer overall survival than RPB1high patients (Fig. 3C). Since the expression of POLR2A was frequently up-regulated in AML, we next compared the expression of POLR2A among AML subtypes, other hematological malignancies and normal hematopoietic cells using BloodSpot database [25]. We found no lineage differences in POLR2A expression among AML subtypes and ALL (Fig. 3D, Table S1). These observations indicate that the high RPB1 level is not only positively with tumor burden but also an adverse prognostic factor in AML patients.

3.2. RPB1 correlates with tumor burden and poor prognosis in AML patients To gain insight into the clinical importance of our findings, we then analyzed RPB1 levels in relation with white blood cell (WBC) levels in various AML cases. We observed that RPB1 levels correlated with hyperproliferative phenotype of AML in a retrospective cohort of 75 AML patients at diagnosis (Fig. 3A, R = 0.6018). For statistical analysis, patients were divided into high- and low-RPB1 expression groups (cut off value: median). In the primary cohort, patients with high level of RPB1 (RPB1high) expression had higher white blood cell numbers than RPB1low patients (median WBC: 51.25 × 109/L versus 30 × 109/L) (Fig. 3B), indicating that RPB1 levels positively correlated with the malignant proliferative potential of leukemia cells in AML patients. Notably, we observed that RPB1low

3.3. RPB1 is critically required for survival, proliferation and cell-cycle progression of AML cells To reveal the biological basis of RPB1 dependency of AML cells, we 5

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

Fig. 3. RPB1 levels are correlated positively with tumor burden and poor prognosis in AML patients. (A) Correlation analysis between RPB1 and WBC levels in AML patients by Western blotting. (B) Comparison of WBC levels between RPB1-high and -low AML patients. (C) Kaplan-Meier survival curves for RPB1-high and -low AML patients. (D) The POLR2A mRNA expression relative to the nearest normal counterpart in different AML subtypes, ALL and normal hematopoietic cells from BloodSpot analysis.

POLR2A-KO significantly decreased colony numbers of AML cells (Fig. 4E and F). In addition, we also examined cell-cycle progression and observed that silence of RPB1 resulted in an accumulation of AML cells in S phase (46.22% versus 27.07%), suggesting that RPB1 is essential for S phase transition and cell cycle progression (Fig. 4G and H).

performed genetic silence of RPB1 using doxycycline (Dox) -inducible CRISPR/Cas9 mediated POLR2A knockout (POLR2A-KO) system. We observed that RPB1 level in POLR2A-KO cells was reduced to 6.81% compared to wild-type (WT) cells at the fourth day after Dox induction (Fig. 4A) and RPB1 silence dramatically decreased viability of MOLM13 cells (Fig. 4B). In parallel, The MOLM-13 cells displayed potent apoptosis (26.1% versus 5.6%) at the fourth day after Dox induction (Fig. 4C and D). Consistently, colony forming assay results showed that

6

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

Fig. 4. POLR2A is critically required for proliferation, cell-cycle progression and viability of AML Cells. (A) Western blotting analysis of RPB1 levels in MOLM-13 cells after DOX-induced POLR2A-KO. (B) Comparison of proliferation curves of MOLM-13 cells after DOX-induced POLR2A-KO with control. (C, D) Apoptosis analysis in MOLM-13 cells after DOX-induced POLR2A-KO. (E, F) Representative images and quantification of colony numbers in MOLM-13 cells after DOX-induced POLR2AKO. (G, H) representative images and quantification of cell cycle by the FCM in MOLM-13 cells after DOX-induced POLR2A-KO.

identify its potential target genes in human AML cells. This strategy identified a total of 1836 differentially expressed genes (DEGs) on the base of at least > 1.5-fold as compared with controlled cells (GSE126000). We observed that 995 (54.19%) of these genes were down-regulated, in which approximately 10% (93 genes) were associated with cancer (Fig. S3), including AML-associated oncogenic genes involved in cell cycle (Fig. 5A) and anti-apoptosis (drug resistance) (Fig. 5B), such as oncogenic transcription factors MYC [26–30], MEIS1

3.4. RPB1 coordinately regulates an array of oncogenic and anti-apoptosis genes Given that RPB1 is a key enzyme in transcribing mRNAs, but little is known, on a global level, about it regulates what mRNA species in cancer cells and how these mRNAs affect the cancer cell fate. To better understand the role of dysregulated RPB1 in pathogenesis of AML, we use CRISPR/Cas9 mediated POLR2A-KO and RNA-Seq approach to

7

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

Fig. 5. RPB1 selectively regulates oncogenic genes and anti-apoptosis genes. (A, B) Heatmaps show RPB1-regulated cell cycle- (A) and apoptosis- (B) gene expression profiles in MOLM-13 cells after POLR2A-KO compared with control. (C) Western blotting analysis of RPB1 and c-Myc and Bcl-2 protein levels in MOLM-13 cells after POLR2A-KO. (D) Heatmap reveals c-myc-regulated gene expression profile in MOLM-13 cells after POLR2A-KO compared with control. (E, F) TCGA data analysis shows the correlation between POLR2A mRNA levels with c-myc and Bcl-2 mRNA levels in AML patients and normal whole blood. Correlation is shown using R and significance was determined using a Spearman correlation.

8

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

refractory AML leukemia model. Because human AML MOLM-13 cells expressed high level of RPB1, harbors FLT3-ITD mutant and exhibits resistance to Cytarabine (Ara-C) treatment (Fig. S6), we established human refractory AML orthotopic model in NSG mice using AML MOLM-13-luciferin cells with Dox-inducible POLR2A-KO vector or control vector. Consistent with the in vitro results, a significant reduction in tumor signal was observed with RPB1 silencing compared with control on the 7th day after POLR2A-KO initiation (Fig. 6). These observations clearly show that silencing of RPB1 by POLR2A-KO could efficiently eliminate most of refractory AML cells in vivo.

[31–33], RUNX2 [34,35], and FLT3 [36], cell cycle regulators CDC25A and CDC25C, and anti-apoptosis genes Bcl-2, and Bcl-11A [37–41]. To confirm these observations, we selected 5 genes to determine their levels by qRT-PCR. c-Myc, CDC25A and Bcl-2 were chosen as downregulation representatives of oncogene, cell cycle regulator and antiapoptosis factor, respectively. Meanwhile TP53 and GADD45A were chosen as representatives of cell proliferation suppressors, respectively. qRT-PCR results were consistent with RNA-Seq analysis (Fig. S4). These findings indicate that RPB1 is preferentially required for the transcription of oncogenic and anti-apoptosis genes, but not for all proteincoding genes. We next focused on cancer-associated gene ontology (GO) analysis to define consensus pathways that may be affected by silencing of RPB1. As expected, we found that RPB1 modulated DEGs were predominantly associated with cell cycle, metabolism, apoptosis, biosynthesis of amino acids, DNA replication, transcriptional imbalance, viral carcinogenesis, protein processing in the endoplasmic reticulum, platinum drug resistance, and a number of cancer-related signaling pathways such as fox-, p53-, motor-, PI3K-Akt-, MAPK-, PPAR-, NKkappa B- Wnt-, TNF-, and Jak-STAT signaling pathways. (Fig. S5). These observations reveal that RPB1-regulated genes are involved in most aspects of cancer cells, such as cell growth, cell cycle, metabolism, apoptosis, and drug resistance.

4. Discussion In this study, we demonstrate that the POLR2A-encoding RPB1 is a critical driver of cell overgrowth and apoptosis resistance in AML. In contrast to heterozygous loss of POLR2A in human colorectal carcinoma and prostate cancer [20,21], we found that both POLR2A mRNA and RPB1 were aberrantly activated in the majority of AML patients. Notably, we show that RPB1 is essential for survival, growth and cell cycle progression of AML cells and its level is correlated with tumor burden and prognosis in AML. Surprisingly, compared to the commonly accepted belief that RPB1 is thought to be required for all proteincoding genes and should be abundantly expressed in all living cells [11,13,17,18,22,23], we demonstrate that RPB1 protein expressed at lower levels in normal hematopoietic cells. Intriguingly, we also find that aberrant activation of RPB1 occurred predominantly in AML, CHOL and THYM, but not in most other types of cancers. Our global transcriptomic analysis revealed that 54.19% of differentially expressed genes is down regulated after silencing RPB1 in MOLM-13 cells. Among these genes, approximately 10% of genes are involved in cell proliferation and drug resistance (apoptosis resistance), including oncogenic transcription factors c-MYC, RUNX2, FLT3, and MEIS1, cell cycle regulators CDC25A and CDC25C, and anti-apoptosis genes such as Bcl-2, and Bcl-11A. Of particular note, we find that both Myc and Bcl-2 genes, which are two master oncogenic factors associated with rapid proliferation of AML [26–28,37–45], are RPB1 target genes. Cancer cells require high levels of oncoproteins to maintain rapid proliferation and constitutive expression of antiapoptotic genes to resist oncogene-induced apoptosis [19]. For example, overexpression of Myc protein can promote cell apoptosis, which could be blocked by Bcl-2 expression. However, little is known about who regulates the transcription balance of oncogenes and anti-apoptotic genes in AML. According to our data, we propose the following mechanistic model for RPB1-driven overgrowth and drug resistance of AML: RPB1 is a central oncogenic hub that coordinates upregulates an array of oncogenic genes and anti-apoptosis genes, which drives unstrained proliferation. Finally, we demonstrate that high level of RPB1 is selectively required for overgrowth and apoptosis resistance of AML cells but not for normal hematopoietic cells, which is consistent with previous reports that inhibition of RNPII in untransformed cells like Rat-1 or human AG 1522 fibroblasts resulted not in apoptosis but in growth arrest [19]. These data suggest that targeting RPB1 might be a potential therapeutic for treating AML. Indeed, we demonstrate that targeting RPB1 by genetic technology selectively depletes AML-related oncogenes and antiapoptosis factors, and potent regresses human refractory AML in orthotopic AML mouse models.

3.5. RPB1 is critically required for the transcription of c-Myc and Bcl-2 genes MYC is required for proliferation of most hematopoietic progenitor populations and cells with a phenotype of hematopoietic stem cells (HSCs) [26]. Deregulated MYC expression is a hallmark of AML [3,27,30]. Both qRT-PCR and WB analysis results showed that POLR2AKO significantly decreased mRNA and proteins of c-Myc and Bcl-2 (Fig. S4, Fig. 5C). Importantly, we found a distinct set of c-Myc target genes downregulated in POLR2A Silencing cells (Fig. 5D). Consistently, TCGA dataset analysis revealed that POLR2A mRNA levels was positively correlated with c-Myc mRNA levels in AML patients and normal whole blood (Fig. 6E, R = 0.67). A similar effect was also observed for Bcl-2 (Fig. 6F, R = 0.57), which is a critical node in apoptosis signaling pathways in AML [37–44]. These data indicate that targeting RPB1 could simultaneously suppress both MYC oncogenic-and Bcl-2 antiapoptosis- signaling pathways. 3.6. Dysregulated RPB1 is a potential therapeutic target in AML To understand whether dysregulated RPB1 could be a therapeutic target for refractory/relapsed AML, we examined the effect of RPB1 silencing on the growth of AML cells in vivo using a pre-established

Availability of data and materials RNA-seq data are available at GEO under accession number GSE126000.

Fig. 6. Reduction of RPB1 regresses human refractory AML in orthotopic and xenograft mouse models. Representative bioluminescent images of xenograft tumors derived from orthotopically implanted MOLM-13 cells expression control and Dox-inducible POLR2A-KO. Silencing of RPB1 on the 7th day by POLR2A-KO leads potent regression of human AML MOLM-13 cells in the orthotopic mouse model.

Funding This work was supported in part by the National Natural Science Foundation of China (81470306, 81670138 and 81870111), the Special 9

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

project foundation of the State Administration of traditional Chinese Medicine (JDZX2015114), the Natural Science Foundation of Zhejiang Province (LY19H290003).

(2005) 3142–3149. [7] Y. Gu, T. Chen, Z. Meng, Y. Gan, X. Xu, G. Lou, H. Li, X. Gan, H. Zhou, J. Tang, G. Xu, L. Huang, X. Zhang, Y. Fang, K. Wang, S. Zheng, W. Huang, R. Xu, CaMKII gamma, a critical regulator of CML stem/progenitor cells, is a target of the natural product berbamine, Blood 120 (2012) 4829–4839. [8] Y. Gu, W. Zheng, J. Zhang, X. Gan, X. Ma, Z. Meng, T. Chen, X. Lu, Z. Wu, W. Huang, R. Xu, Aberrant activation of CaMKIIgamma accelerates chronic myeloid leukemia blast crisis, Leukemia 30 (2016) 1282–1289. [9] T. Chen, Z. Meng, Y. Gan, X. Wang, F. Xu, Y. Gu, X. Xu, J. Tang, H. Zhou, X. Zhang, X. Gan, C. Van Ness, G. Xu, L. Huang, X. Zhang, Y. Fang, J. Wu, S. Zheng, J. Jin, W. Huang, R. Xu, The viral oncogene Np9 acts as a critical molecular switch for coactivating beta-catenin, ERK, Akt and Notch1 and promoting the growth of human leukemia stem/progenitor cells, Leukemia 27 (2013) 1469–1478. [10] Y. Gu, J. Zhang, X. Ma, B.W. Kim, H. Wang, J. Li, Y. Pan, Y. Xu, L. Ding, L. Yang, C. Guo, X. Wu, J. Wu, K. Wu, X. Gan, G. Li, L. Li, S.J. Forman, W.C. Chan, R. Xu, W. Huang, Stabilization of the c-myc protein by CAMKIIgamma promotes T cell lymphoma, Cancer Cell 32 (2017) 115–128 e117. [11] C. Villicana, G. Cruz, M. Zurita, The basal transcription machinery as a target for cancer therapy, Cancer Cell Int. 14 (2014) 18. [12] A. Augert, D. MacPherson, Treating transcriptional addiction in small cell lung cancer, Cancer Cell 26 (2014) 783–784. [13] M.J. Bywater, R.B. Pearson, G.A. McArthur, R.D. Hannan, Dysregulation of the basal RNA polymerase transcription apparatus in cancer, Nat. Rev. Cancer 13 (2013) 299–314. [14] M.J. Bywater, G. Poortinga, E. Sanij, N. Hein, A. Peck, C. Cullinane, M. Wall, L. Cluse, D. Drygin, K. Anderes, N. Huser, C. Proffitt, J. Bliesath, M. Haddach, M.K. Schwaebe, D.M. Ryckman, W.G. Rice, C. Schmitt, S.W. Lowe, R.W. Johnstone, R.B. Pearson, G.A. McArthur, R.D. Hannan, Inhibition of RNA polymerase I as a therapeutic strategy to promote cancer-specific activation of p53, Cancer Cell 22 (2012) 51–65. [15] N. Hein, D.P. Cameron, K.M. Hannan, N.N. Nguyen, C.Y. Fong, J. Sornkom, M. Wall, M. Pavy, C. Cullinane, J. Diesch, J.R. Devlin, A.J. George, E. Sanij, J. Quin, G. Poortinga, I. Verbrugge, A. Baker, D. Drygin, S.J. Harrison, J.D. Rozario, J.A. Powell, S.M. Pitson, J. Zuber, R.W. Johnstone, M.A. Dawson, M.A. Guthridge, A. Wei, G.A. McArthur, R.B. Pearson, R.D. Hannan, Inhibition of Pol I transcription treats murine and human AML by targeting the leukemia-initiating cell population, Blood 129 (2017) 2882–2895. [16] S. Cabarcas, L. Schramm, RNA polymerase III transcription in cancer: the BRF2 connection, Mol. Cancer 10 (2011) 47. [17] Y. Liu, X. Zhang, C. Han, G. Wan, X. Huang, C. Ivan, D. Jiang, C. Rodriguez-Aguayo, G. Lopez-Berestein, P.H. Rao, D.M. Maru, A. Pahl, X. He, A.K. Sood, L.M. Ellis, J. Anderl, X. Lu, TP53 loss creates therapeutic vulnerability in colorectal cancer, Nature 520 (2015) 697–701. [18] Y. Li, Y. Liu, H. Xu, G. Jiang, K. Van der Jeught, Y. Fang, Z. Zhou, L. Zhang, M. Frieden, L. Wang, Z. Luo, M. Radovich, B.P. Schneider, Y. Deng, Y. Liu, K. Huang, B. He, J. Wang, X. He, X. Zhang, G. Ji, X. Lu, Heterozygous deletion of chromosome 17p renders prostate cancer vulnerable to inhibition of RNA polymerase II, Nat. Commun. 9 (2018) 4394. [19] C. Koumenis, A. Giaccia, Transformed cells require continuous activity of RNA polymerase II to resist oncogene-induced apoptosis, Mol. Cell. Biol. 17 (1997) 7306–7316. [20] S.K. Radhakrishnan, A.L. Gartel, A novel transcriptional inhibitor induces apoptosis in tumor cells and exhibits antiangiogenic activity, Cancer Res. 66 (2006) 3264–3270. [21] E.C. Cho, B. Mitton, K.M. Sakamoto, CREB and leukemogenesis, Crit. Rev. Oncog. 16 (2011) 37–46. [22] V.E. Clark, A.S. Harmanci, H. Bai, M.W. Youngblood, T.I. Lee, J.F. Baranoski, A.G. Ercan-Sencicek, B.J. Abraham, A.S. Weintraub, D. Hnisz, M. Simon, B. Krischek, E.Z. Erson-Omay, O. Henegariu, G. Carrion-Grant, K. Mishra-Gorur, D. Duran, J.E. Goldmann, J. Schramm, R. Goldbrunner, J.M. Piepmeier, A.O. Vortmeyer, J.M. Gunel, K. Bilguvar, K. Yasuno, R.A. Young, M. Gunel, Recurrent somatic mutations in POLR2A define a distinct subset of meningiomas, Nat. Genet. 48 (2016) 1253–1259. [23] J. Odawara, A. Harada, T. Yoshimi, K. Maehara, T. Tachibana, S. Okada, K. Akashi, Y. Ohkawa, The classification of mRNA expression levels by the phosphorylation state of RNAPII CTD based on a combined genome-wide approach, BMC Genomics 12 (2011) 516. [24] M.A. Sdano, J.M. Fulcher, S. Palani, M.B. Chandrasekharan, T.J. Parnell, F.G. Whitby, T. Formosa, C.P. Hill, A novel SH2 recognition mechanism recruits Spt6 to the doubly phosphorylated RNA polymerase II linker at sites of transcription, Elife 6 (2017). [25] F.O. Bagger, D. Sasivarevic, S.H. Sohi, L.G. Laursen, S. Pundhir, C.K. Sonderby, O. Winther, N. Rapin, B.T. Porse, BloodSpot: a database of gene expression profiles and transcriptional programs for healthy and malignant haematopoiesis, Nucleic Acids Res. 44 (2016) D917–D924. [26] A. Wilson, M.J. Murphy, T. Oskarsson, K. Kaloulis, M.D. Bettess, G.M. Oser, A.C. Pasche, C. Knabenhans, H.R. Macdonald, A. Trumpp, c-Myc controls the balance between hematopoietic stem cell self-renewal and differentiation, Genes Dev. 18 (2004) 2747–2763. [27] C. Bahr, L. von Paleske, V.V. Uslu, S. Remeseiro, N. Takayama, S.W. Ng, A. Murison, K. Langenfeld, M. Petretich, R. Scognamiglio, P. Zeisberger, A.S. Benk, I. Amit, P.W. Zandstra, M. Lupien, J.E. Dick, A. Trumpp, F. Spitz, Author Correction: a Myc enhancer cluster regulates normal and leukaemic haematopoietic stem cell hierarchies, Nature 558 (2018) E4. [28] L.A. A, D. Tolomeo, I. Cifola, M. Severgnini, A. Turchiano, B. Augello, G. Squeo, D.A.P., D. Traversa, G. Daniele, A. Lonoce, M. Pafundi, M. Carella, O. Palumbo,

Ethics approval and consent to participate Animal studies were approved by the Zhejiang Chinese Medical University Animal Care and Welfare Committee (#10616). AML primary samples and healthy donors were obtained following informed consent and with ethical approved (#IR2018001163) from the ethics committees of the Second Affiliated Hospital Zhejiang University School of Medicine. For original data, please contact [email protected]. cn. Author contributions R.Z.X conceived of the study, initiated, designed, and supervised the experiments. R.Z.X, Y.G. and R.Z.Y wrote the manuscript. Q.F.Y, Y.X, Z.X.W, H.F.Z, L.Z, J.F.L, L.L.Y, B.W.W, P.W, X.Z.Z, and Y.L performed experiments. X.X.G, X.F.Y and S.Z supervised the experiments. Declaration of competing interest All authors declare that they have no competing interests. Acknowledgements The authors thank Dr. Jiwei Li (Shanghai Lifegenes Biotechnology CO., Ltd) for providing technical assistance of bioinformatics analysis. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.yexcr.2019.111653. References [1] J.W. Tyner, C.E. Tognon, D. Bottomly, B. Wilmot, S.E. Kurtz, S.L. Savage, N. Long, A.R. Schultz, E. Traer, M. Abel, A. Agarwal, A. Blucher, U. Borate, J. Bryant, R. Burke, A. Carlos, R. Carpenter, J. Carroll, B.H. Chang, C. Coblentz, A. d'Almeida, R. Cook, A. Danilov, K.T. Dao, M. Degnin, D. Devine, J. Dibb, D.K.t. Edwards, C.A. Eide, I. English, J. Glover, R. Henson, H. Ho, A. Jemal, K. Johnson, R. Johnson, B. Junio, A. Kaempf, J. Leonard, C. Lin, S.Q. Liu, P. Lo, M.M. Loriaux, S. Luty, T. Macey, J. MacManiman, J. Martinez, M. Mori, D. Nelson, C. Nichols, J. Peters, J. Ramsdill, A. Rofelty, R. Schuff, R. Searles, E. Segerdell, R.L. Smith, S.E. Spurgeon, T. Sweeney, A. Thapa, C. Visser, J. Wagner, K. Watanabe-Smith, K. Werth, J. Wolf, L. White, A. Yates, H. Zhang, C.R. Cogle, R.H. Collins, D.C. Connolly, M.W. Deininger, L. Drusbosky, C.S. Hourigan, C.T. Jordan, P. Kropf, T.L. Lin, M.E. Martinez, B.C. Medeiros, R.R. Pallapati, D.A. Pollyea, R.T. Swords, J.M. Watts, S.J. Weir, D.L. Wiest, R.M. Winters, S.K. McWeeney, B.J. Druker, Functional genomic landscape of acute myeloid leukaemia, Nature 562 (2018) 526–531. [2] T. Farge, E. Saland, F. de Toni, N. Aroua, M. Hosseini, R. Perry, C. Bosc, M. Sugita, L. Stuani, M. Fraisse, S. Scotland, C. Larrue, H. Boutzen, V. Feliu, M.L. NicolauTravers, S. Cassant-Sourdy, N. Broin, M. David, N. Serhan, A. Sarry, S. Tavitian, T. Kaoma, L. Vallar, J. Iacovoni, L.K. Linares, C. Montersino, R. Castellano, E. Griessinger, Y. Collette, O. Duchamp, Y. Barreira, P. Hirsch, T. Palama, L. Gales, F. Delhommeau, B.H. Garmy-Susini, J.C. Portais, F. Vergez, M. Selak, G. DanetDesnoyers, M. Carroll, C. Recher, J.E. Sarry, Chemotherapy-resistant human acute myeloid leukemia cells are not enriched for leukemic stem cells but require oxidative metabolism, Cancer Discov. 7 (2017) 716–735. [3] H.B. Maganti, H. Jrade, C. Cafariello, J.L. Manias Rothberg, C.J. Porter, J. YockellLelievre, H.L. Battaion, S.T. Khan, J.P. Howard, Y. Li, A.T. Grzybowski, E. Sabri, A.J. Ruthenburg, F.J. Dilworth, T.J. Perkins, M. Sabloff, C.Y. Ito, W.L. Stanford, Targeting the MTF2-MDM2 Axis sensitizes refractory acute myeloid leukemia to chemotherapy, Cancer Discov. 8 (2018) 1376–1389. [4] A.L. Boyd, L. Aslostovar, J. Reid, W. Ye, B. Tanasijevic, D.P. Porras, Z. Shapovalova, M. Almakadi, R. Foley, B. Leber, A. Xenocostas, M. Bhatia, Identification of chemotherapy-induced leukemic-regenerating cells reveals a transient vulnerability of human AML recurrence, Cancer Cell 34 (2018) 483–498 e485. [5] P.A. Cassier, M. Castets, A. Belhabri, N. Vey, Targeting apoptosis in acute myeloid leukaemia, Br. J. Canc. 117 (2017) 1089–1098. [6] R. Xu, Y. Yu, S. Zheng, X. Zhao, Q. Dong, Z. He, Y. Liang, Q. Lu, Y. Fang, X. Gan, X. Xu, S. Zhang, Q. Dong, X. Zhang, G.S. Feng, Overexpression of Shp2 tyrosine phosphatase is implicated in leukemogenesis in adult human leukemia, Blood 106

10

Experimental Cell Research xxx (xxxx) xxxx

Q. Yu, et al.

[29]

[30]

[31]

[32]

[33]

[34]

[35]

A. Dolnik, D. Muehlematter, J. Schoumans, N. Van Roy, G. De Bellis, G. Martinelli, G. Merla, L. Bullinger, C. Haferlach, C.T. Storlazzi, MYC-containing amplicons in acute myeloid leukemia: genomic structures, evolution, and transcriptional consequences, Leukemia 32 (2018) 2152–2166. A. Nebbioso, V. Carafa, M. Conte, F.P. Tambaro, C. Abbondanza, J. Martens, M. Nees, R. Benedetti, I. Pallavicini, S. Minucci, G. Garcia-Manero, F. Iovino, G. Lania, C. Ingenito, V. Belsito Petrizzi, H.G. Stunnenberg, L. Altucci, c-Myc modulation and acetylation is a key HDAC inhibitor target in cancer, Clin. Cancer Res. 23 (2017) 2542–2555. Y. Pikman, A. Puissant, G. Alexe, A. Furman, L.M. Chen, S.M. Frumm, L. Ross, N. Fenouille, C.F. Bassil, C.A. Lewis, A. Ramos, J. Gould, R.M. Stone, D.J. DeAngelo, I. Galinsky, C.B. Clish, A.L. Kung, M.T. Hemann, M.G. Vander Heiden, V. Banerji, K. Stegmaier, Targeting MTHFD2 in acute myeloid leukemia, J. Exp. Med. 213 (2016) 1285–1306. Z. Li, P. Chen, R. Su, C. Hu, Y. Li, A.G. Elkahloun, Z. Zuo, S. Gurbuxani, S. Arnovitz, H. Weng, Y. Wang, S. Li, H. Huang, M.B. Neilly, G.G. Wang, X. Jiang, P.P. Liu, J. Jin, J. Chen, PBX3 and MEIS1 cooperate in hematopoietic cells to drive acute myeloid leukemias characterized by a core transcriptome of the MLL-rearranged disease, Cancer Res. 76 (2016) 619–629. H.J. Ferreira, H. Heyn, M. Vizoso, C. Moutinho, E. Vidal, A. Gomez, A. MartinezCardus, L. Simo-Riudalbas, S. Moran, E. Jost, M. Esteller, DNMT3A mutations mediate the epigenetic reactivation of the leukemogenic factor MEIS1 in acute myeloid leukemia, Oncogene 35 (2016) 3079–3082. Y. Ogawara, T. Katsumoto, Y. Aikawa, Y. Shima, Y. Kagiyama, T. Soga, H. Matsunaga, T. Seki, K. Araki, I. Kitabayashi, IDH2 and NPM1 mutations cooperate to activate hoxa9/meis1 and hypoxia pathways in acute myeloid leukemia, Cancer Res. 75 (2015) 2005–2016. P.B. Staber, P. Zhang, M. Ye, R.S. Welner, E. Levantini, A. Di Ruscio, A.K. Ebralidze, C. Bach, H. Zhang, J. Zhang, K. Vanura, R. Delwel, H. Yang, G. Huang, D.G. Tenen, The Runx-PU.1 pathway preserves normal and AML/ETO9a leukemic stem cells, Blood 124 (2014) 2391–2399. K. Morita, C. Tokushige, S. Maeda, H. Kiyose, M. Noura, A. Iwai, M. Yamada, G. Kashiwazaki, J. Taniguchi, T. Bando, M. Hirata, T.R. Kataoka, T. Nakahata, S. Adachi, H. Sugiyama, Y. Kamikubo, RUNX transcription factors potentially

[36]

[37]

[38]

[39] [40] [41]

[42] [43]

[44] [45]

11

control E-selectin expression in the bone marrow vascular niche in mice, Blood Adv. 2 (2018) 509–515. X. Jiang, P.Y. Mak, H. Mu, W. Tao, D.H. Mak, S. Kornblau, Q. Zhang, P. Ruvolo, J.K. Burks, W. Zhang, T. McQueen, R. Pan, H. Zhou, M. Konopleva, J. Cortes, Q. Liu, M. Andreeff, B.Z. Carter, Disruption of wnt/beta-catenin exerts antileukemia activity and synergizes with FLT3 inhibition in FLT3-mutant acute myeloid leukemia, Clin. Cancer Res. 24 (2018) 2417–2429. D.M. Moujalled, G. Pomilio, C. Ghiurau, A. Ivey, J. Salmon, S. Rijal, S. Macraild, L. Zhang, T.C. Teh, I.S. Tiong, P. Lan, M. Chanrion, A. Claperon, F. Rocchetti, A. Zichi, L. Kraus-Berthier, Y. Wang, E. Halilovic, E. Morris, F. Colland, D. Segal, D. Huang, A.W. Roberts, A.L. Maragno, G. Lessene, O. Geneste, A.H. Wei, Combining BH3-Mimetics to Target Both BCL-2 and MCL1 Has Potent Activity in Pre-clinical Models of Acute Myeloid Leukemia, (2018) Leukemia. H.E. Ramsey, M.A. Fischer, T. Lee, A.E. Gorska, M.P. Arrate, L. Fuller, K.L. Boyd, S.A. Strickland, J. Sensintaffar, L.J. Hogdal, G.D. Ayers, E.T. Olejniczak, S.W. Fesik, M.R. Savona, A novel MCL1 inhibitor combined with venetoclax rescues venetoclax-resistant acute myelogenous leukemia, Cancer Discov. 8 (2018) 1566–1581. M. Konopleva, A. Letai, BCL-2 inhibition in AML: an unexpected bonus? Blood 132 (2018) 1007–1012. D.C. Altieri, AML therapy: wake up the guardian and cut loose the executioners, Cancer Cell 32 (2017) 719–720. R. Pan, V. Ruvolo, H. Mu, J.D. Leverson, G. Nichols, J.C. Reed, M. Konopleva, M. Andreeff, Synthetic lethality of combined bcl-2 inhibition and p53 activation in AML: mechanisms and superior antileukemic efficacy, Cancer Cell 32 (2017) 748–760 e746. D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell 144 (2011) 646–674. A.R. Delbridge, S. Grabow, A. Strasser, D.L. Vaux, Thirty years of BCL-2: translating cell death discoveries into novel cancer therapies, Nat. Rev. Cancer 16 (2016) 99–109. P. Juin, O. Geneste, F. Gautier, S. Depil, M. Campone, Decoding and unlocking the BCL-2 dependency of cancer cells, Nat. Rev. Cancer 13 (2013) 455–465. K.H. Khoo, C.S. Verma, D.P. Lane, Drugging the p53 pathway: understanding the route to clinical efficacy, Nat. Rev. Drug Discov. 13 (2014) 217–236.