Search

Mutations of the JAK2 gene are frequent aberrations in the aging hematopoietic system and in myeloid neoplasms. While JAK-inhibitors efficiently reduce hyperinflammation induced by the constitutively active mutated JAK2 kinase, the malignant clone and abundance of mutated cells remains rather unaffected. Here, we sought to assess for genetic vulnerabilities of JAK2-mutated clones. We identified lysine-specific demethylase KDM4C as a selective genetic dependency that persists upon JAK-inhibitor treatment. Genetic inactivation of KDM4C in human and murine JAK2-mutated cells resulted in loss of cell competition and reduced proliferation. These findings led to reduced disease penetrance and improved survival in xenograft models of human JAK2-mutated cells. KDM4C deleted cells showed alterations in target histone residue methylation and target gene expression, resulting in induction of cellular senescence. In summary, these data establish KDM4C as a specific dependency and therapeutic target in JAK2-mutated cells that is essential for oncogenic signaling and prevents induction of senescence.

Persistence of malignant clones is a major determinant of adverse outcome in patients with hematologic malignancies. Despite the fact that the majority of patients with acute myeloid leukemia (AML) achieve complete remission after chemotherapy, a large proportion of them relapse as a result of residual malignant cells. These persistent clones have a competitive advantage and can re-establish disease. Therefore, targeting strategies that specifically diminish cell competition of malignant cells while leaving normal cells unaffected are clearly warranted. Recently, our group identified YBX1 as a mediator of disease persistence in JAK2-mutated myeloproliferative neoplasms. The role of YBX1 in AML, however, remained so far elusive. Here, inactivation of YBX1 confirms its role as an essential driver of leukemia development and maintenance. We identify its ability to amplify the translation of oncogenic transcripts, including MYC, by recruitment to polysomal chains. Genetic inactivation of YBX1 disrupts this regulatory circuit and displaces oncogenic drivers from polysomes, with subsequent depletion of protein levels. As a consequence, leukemia cells show reduced proliferation and are out-competed in vitro and in vivo, while normal cells remain largely unaffected. Collectively, these data establish YBX1 as a specific dependency and therapeutic target in AML that is essential for oncogenic protein expression.

Acute myeloid leukemia (AML) patients suffer dismal prognosis upon treatment resistance. To study functional heterogeneity of resistance, we generated serially transplantable patient-derived xenograft (PDX) models from one patient with AML and twelve clones thereof, each derived from a single stem cell, as proven by genetic barcoding. Transcriptome and exome sequencing segregated clones according to their origin from relapse one or two. Undetectable for sequencing, multiplex fluorochrome-guided competitive in vivo treatment trials identified a subset of relapse two clones as uniquely resistant to cytarabine treatment. Transcriptional and proteomic profiles obtained from resistant PDX clones and refractory AML patients defined a 16-gene score that was predictive of clinical outcome in a large independent patient cohort. Thus, we identified novel genes related to cytarabine resistance and provide proof of concept that intra-tumor heterogeneity reflects inter-tumor heterogeneity in AML.

In an effort to identify novel drugs targeting fusion-oncogene–induced acute myeloid leukemia (AML), we performed high-resolution proteomic analysis. In AML1-ETO (AE)-driven AML, we uncovered a deregulation of phospholipase C (PLC) signaling. We identified PLCgamma 1 (PLCG1) as a specific target of the AE fusion protein that is induced after AE binding to intergenic regulatory DNA elements. Genetic inactivation of PLCG1 in murine and human AML inhibited AML1-ETO dependent self-renewal programs, leukemic proliferation, and leukemia maintenance in vivo. In contrast, PLCG1 was dispensable for normal hematopoietic stem and progenitor cell function. These findings are extended to and confirmed by pharmacologic perturbation of Ca++-signaling in AML1-ETO AML cells, indicating that the PLCG1 pathway poses an important therapeutic target for AML1-ETO+ leukemic stem cells.

Resistance towards cancer treatment represents a major clinical obstacle, preventing cure of cancer patients. To gain mechanistic insights, we developed a model for acquired resistance to chemotherapy by treating mice carrying patient derived xenografts (PDX) of acute lymphoblastic leukemia with widely-used cytotoxic drugs for 18 consecutive weeks. In two distinct PDX samples, tumors initially responded to treatment, until stable disease and eventually tumor re-growth evolved under therapy, at highly similar kinetics between replicate mice. Notably, replicate tumors developed different mutations in TP53 and individual sets of chromosomal alterations, suggesting independent parallel clonal evolution rather than selection, driven by a combination of stochastic and deterministic processes. Transcriptome and proteome showed shared dysregulations between replicate tumors providing putative targets to overcome resistance. In vivo CRISPR/Cas9 dropout screens in PDX revealed broad dependency on BCL2, BRIP1 and COPS2. Accordingly, venetoclax re-sensitized derivative tumors towards chemotherapy, despite genomic heterogeneity, demonstrating direct translatability of the approach. Hence, despite the presence of multiple resistance-associated genomic alterations, effective rescue treatment for polychemotherapy-resistant tumors can be identified using functional testing in preclinical models.

Acute leukemias require more accurate and effective treatments, especially upon disease relapse. In search for novel therapeutic targets for acute leukemias, we established a pipeline for CRISPR-Cas9 mediated functional genomic screens which harbor the ability to elegantly increase our knowledge about vulnerabilities and gene dependencies. For a highly patient-related setting, we performed CRISPR knockout (KO) dropout screens in patient-derived xenograft (PDX) models in vivo, combining the advantages of studying an individual patient´s tumor cell in the physiologic in vivo bone marrow microenvironment. Serially transplantable PDX models were lentivirally transduced to stably express Cas9. A customized CRISPR-Cas9 library targeting about 100 genes addressing surface molecules was designed, cloned and transduced into two PDX models of acute lymphoblastic leukemia (ALL). Enriched PDX ALL cells were transplanted into NSG mice and grown until advanced disease stage. Input versus end stage cells were subjected to next generation sequencing, followed by data analysis using MAGeCK algorithm. Data analysis revealed commonly depleted as well as sample-specific depleted genes between the two PDX models tested, and CXCR4 and ITGB1 were the top commonly depleted genes from the screens. For target validation, single sgRNA were cloned into the knockout vector; concomitant expression of recombinant fluorochromes allowed competitive growth assays in PDX models in vivo, comparing cells with KO of interest versus control KO in the same animal. In vivo competitive assay showed that both PDX models clearly depended on both CXCR4 and ITGB1, validating an essential function for both genes in the two PDX ALL models. Of note, disintegrin and metalloproteinase domain-containing protein 10 (ADAM10) was among the list of dropout genes in the screens. ADAM10 is known for its role in the central nervous system and considered as a therapeutic target in Alzheimer disease, but poorly studied in the context of leukemia. In competitive validation assays in vivo, ADAM10 KO population showed a clear growth disadvantage compared to control KO cells in a number of PDX models of ALL, but also acute myeloid leukemia (AML); in some PDX models, ADAM10 KO cells were completely abrogated, indicating that ADAM10 plays an essential role for different types of acute leukemias and represents a yet unknown vulnerability. To better characterize the role of ADAM10 for the clinical situation, we performed further in vivo assays with PDX models. Re-expression of ADAM10 in ADAM10 KO PDX cells could partially rescue the phenotype in an in vivo competitive reconstitution assay, unequivocally proving ADAM10 essentiality in ALL cells. Interestingly, a similar rescue assay expressing a ADAM10 variant lacking the disintegrin domain resulted in the same phenotypical compensation, highlighting essentiality of the enzymatic but not adhesion domain of ADAM10 in tumor engraftment and growth in BM. Important for translating the molecular insights into clinical use, PDX cells treated with an ADAM10 chemical inhibitor ex vivo, showed reduced tumor engraftment capacity compared to the vehicle treated cells, suggesting a role for ADAM10 in tumor-niche interactions and homing to the bone marrow. Further, we performed limiting dilution transplantation assays to determine stem cell frequencies; ADAM10 loss resulted in reduced stemness and a reduced number of leukemia-initiating cells compared to control KO cells, indicating that ADAM10 is essential also in leukemia stem cells. Of clinical relevance, ADAM10 KO significantly sensitized leukemic cells towards treatment of mice with the routine chemotherapeutic drug Cyclophosphamide in vivo, suggesting a putative synergistic effect when addressing ADAM10 as a therapeutic target. In summary, our data revealed ADAM10 as attractive novel vulnerability in acute leukemias with essential function for the tumor-niche interaction, leukemia stem cells and anti-leukemia treatment. ADAM10 might be addressed as therapeutic target to treat acute leukemias in the future. Disclosures No relevant conflicts of interest to declare.

Activating mutations of FMS-like tyrosine kinase 3 (FLT3), notably internal tandem duplications (ITDs), are associated with a grave prognosis in acute myeloid leukemia (AML). Transforming FLT3ITD signal transduction causes formation of reactive oxygen species (ROS) and inactivation of the protein-tyrosine phosphatase (PTP) DEP-1/PTPRJ, a negative regulator of FLT3 signaling. Here we addressed the underlying mechanisms and biological consequences. NADPH oxidase 4 (NOX4) messenger RNA and protein expression was found to be elevated in FLT3ITD-positive cells and to depend on FLT3ITD signaling and STAT5-mediated activation of the NOX4 promoter. NOX4 knockdown reduced ROS levels, restored DEP-1 PTP activity and attenuated FLT3ITD-driven transformation. Moreover, Nox4 knockout (Nox4(-/-)) murine hematopoietic progenitor cells were refractory to FLT3ITD-mediated transformation in vitro. Development of a myeloproliferative-like disease (MPD) caused by FLT3ITD-transformed 32D cells in C3H/HeJ mice, and of a leukemia-like disease in mice transplanted with MLL-AF9/ FLT3ITD-transformed murine hematopoietic stem cells were strongly attenuated by NOX4 downregulation. NOX4-targeting compounds were found to counteract proliferation of FLT3ITD-positive AML blasts and MPD development in mice. These findings reveal a previously unrecognized mechanism of oncoprotein-driven PTP oxidation, and suggest that interference with FLT3ITD-STAT5-NOX4-mediated overproduction of ROS and PTP inactivation may have therapeutic potential in a subset of AML.

Nucleophosmin (NPM1) gene mutations are among the most studied genetic lesions in acute myeloid leukemia (AML) and are associated with a good prognosis. The frequency of exon 12 NPM1 mutation in ad...

In different types of myeloid leukemia, increased formation of reactive oxygen species (ROS) has been noted and associated with aspects of cell transformation, including the promotion of leukemic cell proliferation and migration, as well as DNA damage and accumulation of mutations. Work reviewed in this article has revealed the involvement of NADPH oxidase (NOX)-derived ROS downstream of oncogenic protein–tyrosine kinases in both processes, and the related pathways have been partially identified. FMS-like tyrosine kinase 3 with internal tandem duplications (FLT3-ITD), an important oncoprotein in a subset of acute myeloid leukemias, causes activation of AKT and, subsequently, stabilization of p22phox, a regulatory subunit for NOX1-4. This process is linked to ROS formation and DNA damage. Moreover, FLT3-ITD signaling through STAT5 enhances expression of NOX4, ROS formation, and inactivation of the protein–tyrosine phosphatase DEP-1/PTPRJ, a negative regulator of FLT3 signaling, by reversible oxidation of its catalytic cysteine residue. Genetic inactivation of NOX4 restores DEP-1 activity and attenuates cell transformation by FLT3-ITD in vitro and in vivo. Future work is required to further explore these mechanisms and their causal involvement in leukemic cell transformation, which may result in the identification of novel candidate targets for therapy.

The family of cold shock proteins (CSPs) is highly conserved and consists of 8 members, including Ybx1-3, Csde1 and Lin28. Ybx1 is a multifunctional DNA/RNA binding protein that modulates gene transcription and translation during inflammation and malignant transformation. Recently, our group identified Ybx1 as a mediator of Jak2 signaling in MPN that protects Jak2-mutated cells from Jak-inhibitor induced apoptosis. In a recently published genome wide CRISPR-Cas9 dropout screen in AML cell lines, depletion of Ybx1 resulted in the highest dropout indices compared to other CSP members, with strongest dependencies in cell lines harboring MLL-rearrangements. Protein expression of Ybx1 in healthy individuals (n=10), primary MDS (n=54) and AML (n=58) bone marrow (BM) biopsies, revealed high protein expression in the majority of AML and MDS cases. Consistently, gene expression data revealed high mRNA expression of Ybx1 in AML samples compared to normal controls. Genetic inactivation of Ybx1 in human AML cell lines by RNAi resulted in reduced proliferative capacity. Therefore, we sought to investigate the requirement for Ybx1 in malignant transformation. We used BM cells from a previously published conventional knockout (ko) mouse model (Lu et al., 2005) in which homozygous deletion is embryonically lethal due to brain malformation. We sorted Lineage-Sca1+Kit+ (LSK-) cells from the BM of heterozygous (Ybx1+/-) and wildtype (Ybx1+/+) mice. Cells were retrovirally infected with either MLL-AF9 (MA9) or HoxA9 and Meis1a (HA9M1) to assess for disease development by serial plating in methylcellulose. Haploinsufficiency for Ybx1 in MA9- or HA9M1 transformed cells limited re-plating capacity to 2-4 rounds. When we injected 2,5x 104 MA9-infected LSK cells into sublethally irradiated recipient mice, recipients of MA9-Ybx1+/- cells (n=8) and MA9-Ybx1+/+ (n=10) showed development of AML. However, recipients of MA9-Ybx1+/- cells had a significant delay in AML development (median survival 67.5 days for Ybx1+/+ versus 101.5 days for Ybx1+/- animals, p=0.0078**). This effect appeared even more pronounced when 1x 106 whole BM cells were transplanted into sublethally irradiated secondary recipients. Besides a significant delay in AML development (median survival 37.5 days for recipients of MA9-Ybx1+/+ versus 79 days for MA9-Ybx1+/- BM, p=0.0042**), disease penetrance was reduced by 40%, indicating that haploinsufficiency for Ybx1 impairs development of MA9 driven AML. In contrast, immunophenotypic abundance of stem- and progenitor cells in Ybx1+/+ versus Ybx1+/- animals revealed comparable numbers in all relevant subpopulations. Serial competitive transplantation of Ybx1+/+ and Ybx1+/- BM into primary and secondary recipient animals showed no competitive disadvantage or lack of self-renewal capacity of Ybx1+/- cells. To address the question whether Ybx1 may also be essential for maintenance of AML, we used RNAi to deplete Ybx1 in already established MA9 driven AML. LSK cells from BL/6 mice transformed with MA9 were injected into primary recipient mice. After AML onset, MA9-LSK cells were sorted and infected with either one of 3 shRNAs against Ybx1 or non-targeting (NT-) control. Lentiviral knockdown of 40% reduced colony formation by more than 50% but did not limit the re-plating capacity in vitro. When injected into sub-lethally irradiated recipient mice, lentiviral knockdown (kd) of Ybx1 resulted in a significant delay in AML development (median survival 39.5 days for NT-control versus 53 days for Ybx1 kd, p=0.0446*). To validate our findings, we used a newly generated conditional ko mouse model for Ybx1, in which exon 3 coding for the cold-shock domain is deleted by activation of an Mx1-Cre-recombinase following pIpC administration. Preliminary results provide first evidence that genetic deletion of Ybx1 after onset of MA9 driven leukemia resulted in improved survival of primary recipient (median survival 73 versus 83 days) and a reduced penetrance in secondary recipient mice. Taken together our results may provide first evidence for a functional role of Ybx1 in MLL-AF9 driven AML. As Ybx1 seems to be dispensable for normal hematopoietic cells, these findings may offer a potential therapeutic index. Experiments to assess for the requirement for Ybx1 in maintenance of murine and human AML as well as analysis on proteomic and transcriptional changes following Ybx1 deletion are currently under way. Disclosures No relevant conflicts of interest to declare.

Acute myeloid leukemia (AML) is a clinically and biologically complex and heterogeneous hematopoietic neoplasm. Recent advances in acute myeloid leukemia (AML) biology have lead to prognosticate and predict treatment outcome in AML based on molecular markers. Mutations in NPM1, CEBPA are considered good prognosis and High BAALC, ERG & MN1 expression associate with worse outcome in AML patients treated with standard chemotherapy. Although many efforts have been made to identify genetic mutations and modulated gene expression levels that can be used to predict outcomes in patients with AML, the association between these prognostic markers has not been evaluated. We have reported previously that the NPM1 mutated patients have significantly high dCK and hENT1 gene expression (involved in cytarabine metabolism) and low ABCG2 and ABCB1transporter expression (Abraham et al, ASH abstracts; Nov 2011; 118: 3481 and Nov 2012; 120: 143), suggesting that the good prognostic nature of this mutation is possibly due to the better metabolism and transport of the chemotherapeutic drugs used in induction therapy. We extended this study to look for association between NPM1/FLT3 mutation status and the RNA expression of other good or poor prognostic markers in patients with AML. We prospectively included 274 adult patients with AML in this study. The median age was 42 years (range 16-74y). AML was diagnosed according to the FAB and WHO classifications. There were 238 patients with de novo AML; Secondary AML -6; Therapy related AML- 2 and Relapsed AML-28. Bone marrow cytogenetics and immunophenotyping analysis was available for all patients at diagnosis and/or relapse. Diagnostic bone marrow MNCs were isolated by ficoll- density gradient centrifugation and stored in trizol reagent for RNA expression and mutation detection. RT-PCR was used to screen AML-ETO and Inv 16, and the expression of BAALC, ERG1, MN1, CXCR4 and WT1were analyzed using RQ-PCR. NPM1-c, FLT3 ITD and TKD were screened using DNA PCR followed by gene-scan, sequencing or RFLP methods. The basic demographics and the frequency of the markers are listed in Table 1. When analyzed separately in normal karyotype AML (NK-AML), the frequencies of the mutations were: NPM1: 52.2%; FLT3-ITD: 24%; TKD: 4.3%. When the RNA expression of BAALC, WT1, ERG1, CXCR4 and MN1 was compared in patients with NPM1 or FLT3 mutations, we noticed that patients with NPM1 had significantly low expression of BAALC, MN1 and ERG1 while those with FLT3 mutations (ITD or TKD) had higher expression of these genes (Figure1). There was no significant association with CXCR4 or WT1 expression and these mutations. When analyzed separately in the normal karyotype AML, these associations were still significant. In addition, the relapsed patients had significantly higher expression of BAALC, MN1, and ERG1 RNA compared to de-novoAML cases (data not shown). To conclude, we show that NPM1 or FLT3 mutations acquire the prognostic significance due to several factors including BAALC, ERG1 and MN1expression levels in addition to drug metabolizing enzymes’ and drug transporter expression. These factors must be taken into consideration when attempting to personalize chemotherapy in AML. Disclosures: No relevant conflicts of interest to declare.