Search

Acute megakaryoblastic leukemia (AMKL) represents ∼10% of pediatric acute myeloid leukemia cases and typically affects young children (

Leukemia is a complex genetic disease caused by errors in differentiation, growth, and apoptosis of hematopoietic cells in either lymphoid or myeloid lineages. Large-scale genomic characterization of thousands of leukemia patients has produced a tremendous amount of data that have enabled a better understanding of the differences between adult and pediatric patients. For instance, although phenotypically similar, pediatric and adult myeloid leukemia patients differ in their mutational profiles, typically involving either chromosomal translocations or recurrent single-base-pair mutations, respectively. To elucidate the molecular mechanisms underlying the biology of this cancer, continual efforts have been made to develop more contextually and biologically relevant experimental models. Leukemic cell lines, for example, provide an inexpensive and tractable model but often fail to recapitulate critical aspects of tumor biology. Likewise, murine leukemia models of leukemia have been highly informative but also do not entirely reproduce the human disease. More recent advances in the development of patient-derived xenografts (PDXs) or human models of leukemias are poised to provide a more comprehensive, and biologically relevant, approach to directly assess the impact of the in vivo environment on human samples. In this review, the advantages and limitations of the various current models used to functionally define the genetic requirements of leukemogenesis are discussed.

Chromosomal translocations involving the KMT2A gene are among the most common genetic alterations found in pediatric acute myeloid leukemias although the molecular mechanisms that initiate the disease remain incompletely defined. To elucidate these initiating events we used a human model system of acute myeloid leukemia driven by the KMT2A-MLLT3 (KM3) fusion. More specifically, we investigated changes in DNA methylation, histone modifications, and chromatin accessibility at each stage of our model system and correlated these with expression changes. We observed the development of a pronounced hypomethyl - ation phenotype in the early stages of leukemic transformation after KM3 addition along with loss of expression of stem-cell-associated genes and skewed expression of other genes, such as S100A8/9, implicated in leukemogenesis. In addition, early increases in the expression of the lysine demethylase KDM4B was functionally linked to these expression changes as well as other key transcription factors. Remarkably, our ATAC-sequencing data showed that there were relatively few leukemia-specific changes and that the vast majority corresponded to open chromatin regions and transcription factor clusters previously observed in other cell types. Integration of the gene expression and epigenetic changes revealed that the adenylate cyclase gene ADCY9 is an essential gene in KM3-acute myeloid leukemia, and suggested the potential for autocrine signaling through the chemokine receptor CCR1 and CCL23 ligand. Collectively, our results suggest that KM3 induces subtle changes in the epigenome while co-opting the normal transcriptional machinery to drive leukemogenesis.

Hematopoietic clones with leukemogenic mutations arise in healthy people as they age, but progression to acute myeloid leukemia (AML) is rare. Recent evidence suggests that the microenvironment may play an important role in modulating human AML population dynamics. To investigate this concept further, we examined the combined and separate effects of an oncogene (c-MYC) and exposure to interleukin-3 (IL-3), granulocyte-macrophage colony-stimulating factor (GM-CSF), and stem cell factor (SCF) on the experimental genesis of a human AML in xenografted immunodeficient mice. Initial experiments showed that normal human CD34+ blood cells transduced with a lentiviral MYC vector and then transplanted into immunodeficient mice produced a hierarchically organized, rapidly fatal, and serially transplantable blast population, phenotypically and transcriptionally similar to human AML cells, but only in mice producing IL-3, GM-CSF, and SCF transgenically or in regular mice in which the cells were exposed to IL-3 or GM-CSF delivered using a cotransduction strategy. In their absence, the MYC+ human cells produced a normal repertoire of lymphoid and myeloid progeny in transplanted mice for many months, but, on transfer to secondary mice producing the human cytokines, the MYC+ cells rapidly generated AML. Indistinguishable diseases were also obtained efficiently from both primitive (CD34+CD38−) and late granulocyte-macrophage progenitor (GMP) cells. These findings underscore the critical role that these cytokines can play in activating a malignant state in normally differentiating human hematopoietic cells in which MYC expression has been deregulated. They also introduce a robust experimental model of human leukemogenesis to further elucidate key mechanisms involved and test strategies to suppress them.

Acute myeloid leukemia (AML) is an aggressive form of blood cancer defined by the uncontrolled proliferation and clonal expansion of immature myeloblast cells in the blood and bone marrow, leading to hematopoietic failure. Despite the use of aggressive and cytotoxic standard-of-care drugs, patients often relapse and succumb to the disease partially due to the inability of medically unfit patients to withstand the cytotoxic treatments, regrowth from minimal residual disease and the chemo-resistant nature of leukemic stem cells (LSCs) which can remain in a quiescent state and reside in a protective bone marrow niche. Hence, novel therapies targeting unique leukemic stem cell biology are highly needed to eliminate and avoid reoccurrence. High-throughput screens of human AML LSCs are not performed due to technical issues such as low LSC frequency within primary samples, an inability to purify LSCs, and the difficulty maintaining and expanding primary patient samples and LSCs in vitro. We were able to optimize conditions for a 4-week in vitro large-scale expansion (>600 million bulk) of the primary human AML sample 8227 (OCI-AML-8227), functionally validated to be enriched for LSCs in long-term xenotransplant assays (Eppert et al., 2011). These optimized conditions enabled the isolation and maintenance of the LSC-containing fraction for a chemical screen. We isolated the CD34+ LSC-containing fraction (>90% purity) and performed a high-throughput screen of 11,166 chemical molecules using a CellTiter Glo assay followed by a counter screen against normal CD34+ cord blood (CB) hematopoietic stem and progenitor cells. From this HT screen, a total of 61 hits had >70% inhibition on CD34+ 8227 cells and We then performed dose response assays for each candidate compounds and confirmed 35 potent anti-LSC compounds with IC 50 < 1 μM. This refined the types of compounds to including anti-apoptotic inhibitors, GSK inhibitors, protease inhibitors, metabolism inhibitors, HDAC inhibitors, BET inhibitors, nucleic acid synthesis inhibitors, cell cycle inhibitors and Wnt/β-catenin inhibitors. This is interesting as some of the classes of these compounds (inhibitors of GSK, BET, nucleic acid synthesis, Wnt/β-catenin and metabolism) have been shown to target bulk and leukemic stem cells in AML in vitro and in vivo. We now aim to examine LSC eradication in a panel of genetically defined primary AMLs confirmed through in vitro and in vivo assays. Our goal is to be able to understand and establish the molecular mechanisms and biomarkers on primary functional LSCs. Disclosures No relevant conflicts of interest to declare.

Acute myeloid leukemias (AMLs) result from a series of genetic events occurring in a stem or progenitor hematopoietic cell that gives rise to their clonal expansion and an impaired capacity to differentiate. To circumvent the genetic heterogeneity of AML patient cohorts, we have developed a model system, driven by the MLL-AF9 (MA9) oncogene, to generate multiple human leukemias using progenitor cells from a single healthy donor. Through stepwise RNA-sequencing data generated using this model and AML patients, we have identified consistent changes associated with MA9-driven leukemogenesis and demonstrate that no recurrent secondary mutations are required. We identify 39 biomarkers whose high expression level is specific to this genetic subtype of AML and validate that many of these have diagnostic utility. We further examined one biomarker, the receptor tyrosine kinase (RTK) RET, and show through shRNA knockdowns that its expression is essential for in vivo and in vitro growth of MA9-AML. These results highlight the value of novel human models of AML derived from single donors using specific oncogenic fusions to understand their biology and to uncover potential therapeutic targets.

Acute myeloid leukemia (AML) is a disease that results from the uncontrolled growth of primitive myeloid progenitor cells that are unable to undergo terminal differentiation. To better understand the genetic and epigenetic changes involved in leukemogenesis, we use a human model leukemia system where cord blood donor cells from a single donor are transduced with the human KMT2A-MLLT3 gene fusion to produce leukemias. We have generated multiple model leukemias and compared these to genetically matched pediatric AML patient samples that has allowed us to characterize the step-wise epigenetic and splicing changes that accompany leukemogenesis, allowing us to build an integrated view of this process. Our analysis of these data have revealed a number of candidate biomarker genes that are expressed in KMT2A translocated AMLs but not normal blood cells, some of which are important for leukemic growth. These candidate genes, which have validated diagnostic value, have also been used as targets for monoclonal antibody generation in order to assess their value as potential immunotherapeutics. Having established the value of this approach, we are now expanding our models to include additional oncogenic drivers in order to elucidate the differences in patient outcomes associated with the presence of specific fusions. Moreover, we have recently leveraged these model leukemias and matched patient samples to perform a high-throughput small molecule screen to identify novel therapeutic vulnerabilities and deconvolute the role of secondary mutations in pediatric patients. The establishment of this chemogenomic pipeline to generate, characterize, and then screen model leukemias will provide not only insight into the molecular mechanisms involved in this disease, but also how they can be rationally targeted to improve patient outcomes.

Acute myeloid leukemias (AMLs) with translocations of the mixed lineage leukemia (MLL/KMT2A) gene are common in young patients and are generally associated with poor clinical outcomes. The molecular biology of MLL fusion genes remains incompletely characterized and is complicated by the fact that more than 100 different partner genes have been identified in fusions with MLL. The continuously growing list of MLL fusions also represents a clinical challenge with respect to identification of novel fusions and tracking of the fusions to monitor progression of the disease after treatment. Recently, we have developed a novel single-donor model leukemia system that permits the development of human AML from normal cord blood cells. Gene expression analysis of this model and of MLL-AML patient samples has identified a number of candidate biomarker genes with highly biased expression on leukemic cells. Here, we present data demonstrating the potential clinical utility of several of these candidate genes for identifying known and novel MLL fusions.

Background: Current evidence suggests that genetic and epigenetic abnormalities drive the development of human Acute Myeloid Leukemias (AMLs). However, whether these are sufficient to establish a permanent, self-sustaining AML population, and the potential role of shared perturbed downstream pathways is unknown. We hypothesized that a modest upregulated expression of MYC might play such a role given its commonly increased expression in many AML patients' cells. To test this hypothesis, we assessed the dynamics and types of cells produced in sublethally irradiated NOD-Rag1-/--IL2Rγc-/-(NRG) mice transgenically producing human IL3, GM-CSF and SCF (NRG-3GS mice) following their transplantation with freshly isolated subsets of normal CD34+ cord blood (CB) cells that were first lentivirally transduced with a human MYC cDNA. Results: FACS and Western blot analyses indicated this produced a 2 to 5-fold increase in MYC mRNA and protein levels in MYC-transduced CD34+ CB cells, and 21/22 NRG-3GS mice injected with ≥6,500 of these cells developed a fatal human AML population within 7 weeks. Histological analysis of their bone marrow and spleen cells showed both contained a prominent human CD123+CD33+CD15±CD34-CD14-CD19-CD3- blast population. Additional limiting dilution transplants showed that both the CD34+CD38- cells (enriched for hematopoietic stem cells) and the more differentiated CD34+ GMPs were similarly highly susceptible (at frequencies of 1/14 and 1/46, respectively) and, in both cases, generated progeny that could initiate serially transplantable leukemias with the same phenotypic and transcriptomic features. Comparison to normal CB cells indicated these most closely resembled GMPs, and comparison to pediatric AML patient samples indicated a similarity to myelomonocytic leukemias with enhanced MYC expression. Interestingly, 14 sublethally irradiated NRG mice (the parental strain not producing human 3GS) transplanted with matched aliquots of CD34+ MYC-transduced cells regenerated a normal spectrum of CD19+ lymphoid cells, CD14+ and CD15+ GM cells and readily detectable CD34+ cells for up to 32 weeks of follow-up with no evidence of leukemogenesis. However, transfer of these regenerated human cells into secondary NRG-3GS mice, even after this extended period, enabled their rapid production of a lethal human AML in all 5 mice tested. In contrast, matched aliquots transplanted into 5 NRG recipients produced declining grafts of normal cells. This finding was then exploited to determine which growth factors were responsible for activating the AML program by transplanting NRG mice with CD34+ CB cells transduced with MYC and just a single growth factor, or all 3 as a positive control. In this set of experiments, a lethal human AML was obtained when MYC was paired with human IL3 or GM-CSF (or all 3 together), but not with SCF (or no growth factors). Conclusion: We report here a new in vivo model of MYC-induced human myeloid leukemogenesis that produces a serially transplantable AML closely resembling human pediatric myelomonocytic leukemias with elevated MYC expression. The rapidity, consistency, and high frequency of this transformation process obtained by transducing late granulopoietic as well as early types of normal human CD34+ progenitor cells makes this system highly attractive for future mechanistic and therapeutic testing experiments. The discovery that MYC deregulation alone generates a stable "latent program" that can be rapidly activated by exposure to exogenous growth factors typical of inflammatory states also raises intriguing questions about the potential role of such events in the genesis of AML populations that arise in patients. Disclosures Beer: Karus therapeutics Ltd.: Employment.

Acute myeloid leukemias (AML) comprise a genetically diverse group of human hematologic malignancies with a generally poor prognosis and modes of pathogenesis that have been difficult to investigate. Historic evidence and more recent findings have identified hematopoietic stem cell programs as major targets of mutations that predispose to and/or initiate a multistep process of transformation. However, details of the changes involved and how they interact in emerging human leukemic cells have been particularly challenging to characterize. This is due to an emerging appreciation of the heterogeneity in events that impact the differentiation processes of normal human hematopoietic cells, as well as difficulties in recreating the full leukemogenic alterations in the actual human cells affected, to avoid discrepancies inherent in studying model organisms. Here we describe several models of de novo human leukemogenesis that illustrate their power and the novel results they can generate. These include our recent discovery of a novel “latent” leukemic state that can be obtained in MYC-transduced human cord blood cells regenerating a normal spectrum of lymphoid and myeloid progeny in transplanted immunodeficient mice, but that can then be rapidly activated to an aggressive form of AML by in vivo exposure to a single human growth factor. These findings portend the future utility of de novo human leukemogenesis models as new platforms for elucidating shared molecular mechanisms responsible for different stages of human leukemogenesis that may be initiated by different mutations or exposure to different microenvironmental conditions. The flexibility and consistency of these models also make them attractive for identifying and testing new treatment strategies targeting mechanisms required for disease manifestation.

The three-amino-acid loop extension (TALE) class homeodomain proteins MEIS1 and PKNOX1 (PREP1) share the ability to interact with PBX and HOX family members and bind similar DNA sequences but appear to play opposing roles in tumor development. Elevated levels of MEIS1 accelerate development of HOX- and MLL-induced leukemias, and this pro-tumorigenic property has been associated with transcriptional activity of MEIS1. In contrast, reduction of PKNOX1 levels has been linked with cancer development despite the absence of an identifiable transactivating domain. In this report, we show that a chimeric protein generated by fusion of the MEIS1 C-terminal region encompassing the transactivating domain with the full-length PKNOX1 (PKNOX1-MC) acquired the ability to accelerate the onset of Hoxa9-induced leukemia in the mouse bone marrow transduction/transplantation model. Gene expression profiling of primary bone marrow cells transduced with Hoxa9 plus Meis1, or Hoxa9 plus Pknox1-MC revealed perturbations in overlapping functional gene subsets implicated in DNA packaging, chromosome organization, and in cell cycle regulation. Together, results presented in this report suggest that the C-terminal domain of MEIS1 confers to PKNOX1 an ectopic transactivating function that promotes leukemogenesis by regulating expression of genes involved in chromatin accessibility and cell cycle progression.

It is believed that hemopoietic stem cells (HSC), which colonize the fetal liver (FL) rapidly, expand to establish a supply of HSCs adequate for maintenance of hemopoiesis throughout life. Accordingly, FL HSCs are actively cycling as opposed to their predominantly quiescent bone marrow counterparts, suggesting that the FL microenvironment provides unique signals that support HSC proliferation and self-renewal. We now report the generation and characterization of mice with a mutant allele of Baf250a lacking exons 2 and 3. Baf250aE2E3/E2E3 mice are viable until E19.5, but do not survive beyond birth. Most interestingly, FL HSC numbers are markedly higher in these mice than in control littermates, thus raising the possibility that Baf250a determines the HSC pool size in vivo. Limit dilution experiments indicate that the activity of Baf250aE2E3/E2E3 HSC is equivalent to that of the wild-type counterparts. The Baf250aE2E3/E2E3 FL-derived stroma, in contrast, exhibits a hemopoiesis-supporting potential superior to the developmentally matched controls. To our knowledge, this demonstration is the first that a mechanism operating in a cell nonautonomous manner canexpand the pool size of the fetal HSC populations.

Acute myeloid leukemia (AML) occurs when hematopoietic progenitor cells acquire genetic defects blocking the regulation of normal growth and differentiation. Although recurrent translocations have been identified in AML, almost half of adult AML patients present with a normal karyotype (NK-AML). While cell line models exist to study AML, they frequently have abnormal/unstable karyotypes, while primary cells from NK-AML patients are difficult to maintain in vitro. Here we provide a thorough molecular characterization of a recently established cell line, CG-SH, which has normal cytogenetics, representing a useful new model for NK-AML. Using high-throughput DNA sequencing, we first defined the genetic background of this cell line. In addition to identifying potentially deleterious SNVs in genes relevant to AML, we also found insertions in both GATA2 and EZH2, two genes previously linked to AML. We further characterized the growth of this model system in vitro with a cytokine mix that promotes faster cell growth. We assessed gene expression changes after the addition of cytokines to the culture media and found differential expression in genes implicated in proliferation, apoptosis and differentiation. Our results provide a detailed molecular characterization of genetic defects in this cell line derived from an NK-AML patient.

The GATA2 gene encodes a zinc-finger transcription factor that acts as a master regulator of normal hematopoiesis. Mutations in GATA2 have been implicated in the development of myelodysplastic syndrome and acute myeloid leukemia (AML). Using RNA sequencing we now report that GATA2 is either mutated with a functional consequence, or expressed at low levels in the majority of normal karyotype AML (NK-AML). We also show that low-GATA2-expressing specimens (GATA2(low)) exhibit allele-specific expression (ASE) (skewing) in more than half of AML patients examined. We demonstrate that the hypermethylation of the silenced allele can be reversed by exposure to demethylating agents, which also restores biallelic expression of GATA2. We show that GATA2(low) AML lack the prototypical R882 mutation in DNMT3A frequently observed in NK-AML patients and that The Cancer Genome Atlas AML specimens with DNMT3A R882 mutations are characterized by CpG hypomethylation of GATA2. Finally, we validate that several known missense single-nucleotide polymorphisms in GATA2 are actually loss-of-function variants, which, when combined with ASE, represent the equivalent of homozygous GATA2 mutations. From a broader perspective, this work suggests for the first time that determinants of ASE likely have a key role in human leukemia.

Histone methylation is a dynamic and reversible process proposed to directly impact on stem cell fate. The Jumonji (JmjC) domain-containing family of demethylases comprises 27 members that target mono-, di-, and trimethylated lysine residues of histone (or nonhistone) proteins. To evaluate their role in regulation of hematopoietic stem cell (HSC) behavior, we performed an in vivo RNAi-based functional screen and demonstrated that Jarid1b and Jhdm1f play opposing roles in regulation of HSC activity. Decrease in Jarid1b levels correlated with an in vitro expansion of HSCs with preserved long-term in vivo lymphomyeloid differentiation potential. Through RNA sequencing analysis, Jarid1b knockdown was associated with increased expression levels of several HSC regulators (Hoxa7, Hoxa9, Hoxa10, Hes1, Gata2) and reduced levels of differentiation-associated genes. shRNA against Jhdmlf, in contrast, impaired hematopoietic reconstitution of bone marrow cells. Together, our studies identified Jarid1b as a negative regulator of HSC activity and Jhdmlf as a positive regulator of HSC activity.

Pediatric acute megakaryoblastic leukemia (AMKL) accounts for 10% of childhood acute myeloid leukemia (AML) cases and remains a high fatality cancer. CBFA2T3-GLIS2, NUP98-KDM5A, RBM15-MKL1 and MLL gene rearrangements are recurrent aberrations that are mutually exclusive and found at similar frequencies in half the cases of pediatric AMKL. The recently identified CBFA2T3-GLIS2 and NUP98-KDM5A chimeric oncogenes are associated with inferior outcomes (overall survival, OS: ~30%) compared to patients harboring the RBM15-MKL1 gene fusion (OS ~70%). To investigate NUP98-KDM5A driven leukemogenesis, human cell lines and mouse models were engineered using overexpression of the chimeric oncogene in CD34+cord blood (CB) stem/progenitor cells. To this end, the cDNA of the NUP98-KDM5A fusion, consisting of the 5' portion of the nuclear pore protein nucleoporin 98 (NUP98) gene fused to the 3' portion of the histone lysine demethylase 5A (KDM5A) encoding gene, was cloned into a MNDU lentiviral vector carrying a GFP reporter gene. Using optimized culture conditions, 10,000 freshly isolated CB-CD34+ (day 0) cells were seeded in multiple wells in vitro and transduced with either NUP98-KDM5A or control (CTL) vectors. Initial gene transfer rates were 8.2 ± 0.95% (expt.1, n=3) and 22.9 ± 4.16% (expt.2, n=12) for NUP98-KDM5A and 48 ± 0.04% (expt.1, n=3) and 76.9 ± 6.36% (expt.2, n=4) for CTL conditions. Overexpression (OE) of NUP98-KDM5A led to maturation block and increased cell proliferation in vitro as compared to CTL, as assessed by cytological examination of cells, serial tracking of increasing CD34+GFP+ cell fractions and lack of a c-KIThi differentiated mast cell population. This was observed by day 25 of culture in 66% of the wells in a first experiment (2/3), and in 92% (11/12) of the wells in a confirmatory study. An overriding NUP98-KDM5A CD34+GFP+ immature cell population was present by day 42 of culture (up to 94% of cells), while CD34+GFP+cells were significantly decreased in the CTL conditions (0.3 to 8%), where mast cells predominated. Cell lines were maintained in culture for 180 days in the first experiment (day 145 in ongoing expt.2) and aliquoted at regular intervals for immunophenotypic assessment, ARN/DNA extraction, cryopreservation, and cytogenetics studies. Clonogenic activity was markedly increased in NUP98-KDM5A-OE cells (colony forming cell frequency of 1:3 vs 1:200 in CTL, at day 88 of culture). Transcriptomic profiling of selected cell lines closely matched that of NUP98-KDM5A bearing pediatric AMKL samples. Notably, there was an up-regulation of HOXA and HOXB gene clusters, as well as other transcription factors, epigenetic regulators, cell surface receptors and kinases. No additional driver mutations were detected in the cell lines or patient samples (n=2). Epigenetic characterisation of 2 cell lines is ongoing (H3K4, H3K9, H3K27 and H3K79 methylation status) as the chimeric oncogene encodes for the C-terminal portion of a H3K4 histone demethylase (KDM5A). Xenotransplantation of 75% of day 7 cells in immunodeficient mice resulted in the development of overt AMKL in 1 of 3 mice after 32 weeks. Recipient mouse bones were white and brittle, and the marrow cavity infiltrated by 30% hCD45loCD61+GFP+ leukemic blasts, with typical megakaryoblastic morphology. The leukemic blasts were also detected in blood (5%), and in enlarged spleen (0.2%). Secondary transplantation of isolated AMKL cells (from bone marrow and spleen) was performed, along with expression profiling by RNA sequencing. In an ongoing confirmatory experiment, 4 out of 12 mice displayed hCD45+NUP98-KDM5A-GFP+ cells (0.1, 0.2, 0.5, and 1.4%) in blood at 16 weeks, while 1 out of 4 CTL mice displayed 0.4% hCD45+-GFP+ engraftment at the same time point. Analysis of lentiviral integration patterns of genomic DNA of both cell lines and xenografts by southern blot is ongoing to determine clonality. Overexpression of the NUP98-KDM5A chimeric oncogene in cord blood CD34+ cells can thus be used to engineer human cell lines and xenograft models to study this poor prognostic subgroup of pediatric AMKL. By crossing expression profiles of patient samples, cell lines and xenograft models, a list of potential NUP98-KDM5A direct targets was elaborated. The engineered cell lines pave the way to genetic and chemical screens, to identify leukemia specific functional dependencies and drug targets, to ultimately improve patient outcome. Disclosures No relevant conflicts of interest to declare.

Ly49-mediated recognition of MHC-I molecules on host cells is considered vital for natural killer (NK)–cell regulation and education; however, gene-deficient animal models are lacking because of the difficulty in deleting this large multigene family. Here, we describe NK gene complex knockdown (NKCKD) mice that lack expression of Ly49 and related MHC-I receptors on most NK cells. NKCKD NK cells exhibit defective killing of MHC-I–deficient, but otherwise normal, target cells, resulting in defective rejection by NKCKD mice of transplants from various types of MHC-I–deficient mice. Self–MHC-I immunosurveillance by NK cells in NKCKD mice can be rescued by self–MHC-I–specific Ly49 transgenes. Although NKCKD mice display defective recognition of MHC-I–deficient tumor cells, resulting in decreased in vivo tumor cell clearance, NKG2D- or antibody-dependent cell-mediated cytotoxicity–induced tumor cell cytotoxicity and cytokine production induced by activation receptors was efficient in Ly49-deficient NK cells, suggesting MHC-I education of NK cells is a single facet regulating their total potential. These results provide direct genetic evidence that Ly49 expression is necessary for NK-cell education to self–MHC-I molecules and that the absence of these receptors leads to loss of MHC-I–dependent “missing-self” immunosurveillance by NK cells.

The molecular mechanisms regulating self-renewal of leukemia stem cells remain poorly understood. Here we report the generation of 2 closely related leukemias created through the retroviral overexpression of Meis1 and Hoxa9. Despite their apparent common origin, these clonal leukemias exhibit enormous differences in stem cell frequency (from 1 in 1.4, FLA2; to 1 in 347, FLB1), suggesting that one of these leukemias undergoes nearly unlimited self-renewal divisions. Using next-generation RNA-sequencing, we characterized the transcriptomes of these phenotypically similar, but biologically distinct, leukemias, identifying hundreds of differentially expressed genes and a large number of structural differences (eg, alternative splicing and promoter usage). Focusing on ligand-receptor pairs, we observed high expression levels of Sdf1-Cxcr4; Jagged2-Notch2/1; Osm-Gp130; Scf-cKit; and Bmp15-Tgfb1/2. Interestingly, the integrin beta 2-like gene (Itgb2l) is both highly expressed and differentially expressed between our 2 leukemias (∼ 14-fold higher in FLA2 than FLB1). In addition, gene ontology analysis indicated G-protein-coupled receptor had a much higher proportion of differential expression (22%) compared with other classes (∼ 5%), suggesting a potential role regulating subtle changes in cellular behavior. These results provide the first comprehensive transcriptome analysis of a leukemia stem cell and document an unexpected level of transcriptome variation between phenotypically similar leukemic cells.

Background Recent development in sequencing technologies with deep coverage for mutation analysis has enabled the identification of clonal architecture in some cancers. RAS mutations are observed in a large proportion of MLL leukemias. Our hypothesis is that determination of RAS mutation status in MLL leukemias should provide insights into the clonal make up of this disease and clues about the nature of clones that overcome therapy. Methods We combined exome and transcriptome sequencing in 32 adult MLL leukemias and results were compared to our cohort of 48 normal karyotype (NK) AML. Exome capture and paired-end sequencing (2 x 100bp, Illumina HiSeq 2000) were performed using TruSeq (Illumina) protocols. Mean coverage was 165X for transcriptome and 42X for exome. Initial analysis was focused on 25 known AML-associated genes and excluded all other novel mutations. Average transcriptome and exome coverage for N/KRAS alleles was 287X (25-846) and 42X (9-151), respectively. Clones were defined based on the identification of N/KRAS mutations in at least 1% of the reads. Results Figure 1 shows mutation status, MLL partners and FAB classification for each MLL leukemia. No mutations were observed in NPM1, FLT3 (ITD), CEBPA (biallelic), RUNX1, DNMT3A, IDH1, KIT, BCOR, SF3B1, U2AF1 or RAD21. On average, 1 mutated gene (range: 0-4) per sample was found compared to 3 (range 0-5) in NK-AML (p < 0.0001). We observed that 13/32 MLL leukemias (which include 2 paired samples) harbored N/KRAS mutations. There were no association between RAS mutation status and MLL partner, FAB classification, age, white blood cell count and overall survival. RAS mutations were found in 15% of NK-AML which contained on average 2.3 additional mutations in leukemia-associated genes compared to only 0.3 (p Excluding 2 paired relapse specimens, a total of 24 N/KRAS mutated clones were identified in 11 of the 30 MLL leukemias. The first sample included 5 clones each containing different NRAS mutations (e.g. G13R, G13D, etc.) contributing to 17, 9, 4, 2 and 2 % of the reads. Since RAS mutations are mostly heterozygous, we estimated that the contribution of each clones varied between 34 (i.e. 17% x 2) to 4%. A similar analysis revealed 4 clones in another specimen, contributing to 22, 12, 12 and 4 % of the cells. In 4 additional samples, the proportions of N/KRAS mutated clones were 1) 42, 38 and 8% 2) 92, 4 and 2 %, 3) 78 and 12% and 4) 52 and 32%, establishing that 20% (6/30) of these MLL leukemias were oligo- to polyclonal. In comparison, our NK-AML cohort of 48 patients included 7 specimens mutated for N/KRAS in which a total of 10 different clones were identified for an average of 0.2 RAS mutated clones per NK-AML versus 0.8 in MLL leukemias (p=0.007). This result further strengthens the hypothesis that RAS and MLL-fusion genes are strong collaborators in human AML. Grossmann et al recently showed that RAS mutated clones can be lost at relapse (Leukemia, 2013), possibly suggesting that other genes are at play in collaborating with MLL-fusions and causing drug resistance. To identify such genes, we further analyzed paired diagnosis and relapse samples in 2 patients. In the first patient, the KRAS mutation that was found in 66% of the cells at diagnosis was identified in all cells at relapse. In the second patient, while KRAS G12V and G12D mutations were found in 78% and 12 % of the cells at diagnosis, only the G12V clone was detected in 100% of the cells at relapse indicating in vivo clonal selection in both cases. We then performed a comparative analysis of mutated/wild type allele ratios for other coding genes. This analysis enabled us to identify a subset of mutations in candidate genes that are present at relapse in the dominant clone but that were undetectable or at lower frequency at presentation, indicating they might be specifically involved into occurrence of relapse (i.e. drug resistance). Conclusion NRAS and KRAS are mutated in 37% of MLL leukemias in this cohort. In contrast to NK-AML, these leukemias are frequently oligo- to polyclonal and contain few additional mutations suggesting that RAS activation may be sufficient to induce AML in the presence of MLL fusions. Evidence from our limited number of relapse patients, and that of others, suggests that RAS does not confer drug resistance which could be explained by novel mutations in genes that were specifically detected in the dominant clones at relapse. Disclosures: No relevant conflicts of interest to declare.

Hematopoiesis is sustained by a pool of multipotent hematopoietic stem cells (HSCs) that have the capacity to differentiate into cells of all blood cell lineages. The pool of long-lived HSCs is maintained throughout life by the self-renewal ability of HSCs. New evidence suggests the process of alternative splicing is an important regulator of the maturation and activation of blood and immune effector cells. It is presently estimated that almost all multi-exon genes in human genome undergo alternative pre-mRNA splicing, and aberrant splicing has been linked to a variety of human pathologies. However, the role that pre-mRNA splicing may have for HSCs behaviour remains largely unexplored. Heterogeneous nuclear ribonucleoprotein L (hnRNPL) is an RNA-binding protein that regulates alternative splicing by binding exonic splicing silencers elements (ESS) resulting in exon exclusion from the mature mRNA. RT-PCR analyses showed that hnRNPL is expressed in early stages of hematopoiesis including HSCs and lineage restricted hematopoietic progenitors. To test the role of hnRNPL in hematopoietic differentiation, we have generated conditional deficient mice, since a constitutive deletion of hnRNPL results in early embryonic lethality. Animals carrying two hnRNPL-floxed alleles (hnRNPLfl/fl) can be deleted at adult stage by the pIpC inducible MxCre transgene or by the VavCre transgene, which is expressed in all hematopoietic cells starting at embryonic stage E14. VavCre+hnRNPLfl/fl mice were not viable and did not progress further in their development than embryonic stage E17.5 and ablation of hnRNPL by pIpC injection caused a high rate of mortality in adult MxCre+hnRNPLfl/fl mice compared to control animals. Both the fetal liver (FL) of VavCre+hnRNPLfl/fl mice and the bone marrow (BM) of adult MxCre+hnRNPLfl/fl mice had a significantly reduced cellularity. Furthermore, flow cytometric analysis revealed in both FL and BM a significant reduction in frequency and absolute numbers of all mature blood cells, the lymphoid and myeloid precursors, CLPS, CMPs and GMPs and to a lesser extent the erythroid/megakaryocytic precursors (MEPs). Methylcellulose and both competitive and non-competitive transplantation assays demonstrated that HSCs lacking hnRNPL cannot generate lineage-committed progenitors and have lost their self-renewal capacity and reconstitution potential. A genome-wide analysis of mRNA expression and splicing through next-generation RNA sequencing of wild-type (WT) or VavCre+hnRNPLfl/fl E14.5 Lin- c-kit+ fetal liver cells (FLCs) revealed that hnRNPL deficiency affects not only alternative splicing but also gene expression levels in hematopoietic progenitors. In the absence of hnRNPL, genes implicated in regulating apoptosis, DNA damage response and cell division where found up-regulated in Lin- c-kit+ FLCs. Among those genes, many were p53 effector genes such as Cdkn1a, Ccng1, Trp53inp1, TrailR2, Bax and Zmat3. In addition genes that are known to be required for normal hematopoiesis and HSCs functions such as Gfi1, CD34, Csfr1, Egr1 and Runx1 were found down-regulated in those cells. Further analyses by qPCR and Western blots confirmed those findings and also showed that the level of p53 protein expression was upregulated in VavCre+hnRNPLfl/fl FLCs although the mRNA level is the same as in the WT cells suggesting that hnRNPL affects p53 mRNA translation efficiency. Similarly, several genes found differentially spliced are implicated in cell cycle progression or required for normal hematopoiesis in FL such as Bcl11a, Cdk4, Ccnd2 and TRP53bp1. These results together with an increased level of Reactive Oxygen Species (ROS) and elevated levels of phosphorylated histone H2AX (γ-H2AX, a sensor for double strand DNA breaks) suggest that hnRNPL regulates the activation of a p53 dependent DNA damage response pathway in hematopoietic stem cells. As a consequence loss of hnRNPL results in a loss of hematopoietic stem and progenitor cells. Our data also suggest that hnRNPL does not only regulate alternative splicing but also expression levels of a set of specific effector genes involved in HSC survival, proliferation, ultimately affecting self-renewal. Disclosures: No relevant conflicts of interest to declare.

Author(s): Belanger, S; Tu, MM; Rahim, MMA; Mahmoud, AB; Patel, R; Tai, LH; Troke, AD; Wilhelm, BT; Landry, JR; Zhu, Q; Tung, KS; Raulet, DH; Makrigiannis, AP

In a previous gain-of-function screen, we identified 18 nuclear factors that enhance mouse hematopoietic stem cell (HSC) activity in vitro. Of these factors, the majority was believed to augment HSC function intrinsically. In the current study, we investigated the mechanisms of action of the previously identified agonists of HSC activity and tested whether human HSCs are also responsive to these factors. Our results unexpectedly revealed that the majority of the identified factors confer a competitive advantage to mouse HSCs in a non-cell-autonomous manner. Five of these factors, namely FOS, SPI1, KLF10, TFEC, and PRDM16, show robust transcriptional cross-regulation and are often associated with osteoclastogenesis. These findings define at least one novel non-cell-autonomous network in engineered niches. Surprisingly, and in contrast to their important effect on mouse HSCs, all engineered niches failed to significantly enhance the activity of human HSCs. This last finding further supports a lack of conservation in determinants that control HSC self-renewal in mouse versus human cells.

Abstract 1325 New innovative strategies are needed to expand human hematopoietic stem cells (HSCs) for clinical applications. In this perspective, we recently developed an in vitro/in vivo screening strategy, which revealed 18 nuclear factors that enhance HSC activity (Deneault et al., Cell 2009). During these experiments, mouse HSCs were kept for 12 days in mini-cultures that included viral producer cells for each tested factor. Interestingly, 4 of the 18 hits identified in this initial screen, i.e., Fos, Tcfec, Hmgb1 and Sfpi1 operated through non-HSC autonomous (NHA) mechanisms: expanded HSCs were not infected with retrovirus. This suggested that the transfected viral producer cells (feeder cells) produced membrane-bound or soluble molecules that promote expansion of HSCs introduced in these cultures. We now provide evidence that seven additional factors, i.e., Smarcc1, Vps72, Sox4, Klf10, Ski, Prdm16 and Erdr1 significantly enhance HSC activity through NHA mechanisms, hereafter called “NHA factors”. Moreover, we found that Vps72, Fos and Klf10 also promote expansion of human HSCs by NHA mechanisms. Interestingly at least for feeders expressing Klf10, we observed that physical contact between HSCs and these engineered support cells was not necessary, suggesting the presence of secreted molecule(s) in the medium. Expression profiling was next performed using mRNA extracted from feeder cells transduced with each of the NHA factors. Firstly, we shed light on the transcriptional make up and potential convergence of signaling pathways in the engineered feeder cells: our results reveal two independent but interconnected transcriptional regulatory subnetworks. Strikingly, some constituents of the subnetworks, i.e., Sfpi1, Fos, Klf10 and Tcfec (Mitf-related) have previously been shown to play critical roles in the regulation of osteoclasts, which are a myeloid-derived population of cells residing in the HSC bone marrow niche. In addition, all of the NHA factors act in concert to increase Prdm16 expression levels in a range from 2.5- to 54.6-fold. For this reason, Prdm16 clearly holds the position of central hub of the osteoclastic network. However, Tcfec represents the ultimate downstream effector of this pathway as Prdm16 elevates its expression levels by 18.9-fold. In parallel, Prdm16 can also increase the expression of Sfpi1, which in turn can upregulate Tcfec expression up to 874-fold. Moreover, upregulated mRNA targets corresponding to factors that are secreted or associated with the plasma membrane were considered as potential candidate agonists of HSC self-renewal. A high degree of overlap was observed between the sets of proteins produced by feeder cells engineered to overexpress Sfpi1, Fos, Klf10, Tcfec or Ski. These secreted and membrane bound proteins include Agt, Aspn, Ogn, Ptgds, Nckap1l, Rgs16 and Lcn2. Studies are ongoing to characterize the contribution of these newly identified NHA proteins in HSC expansion. Validation of these NHA proteins in human HSC expansion will have a clear potential for translational medicine. Disclosures: No relevant conflicts of interest to declare.

Abstract 1587 Although important efforts have been invested in the discovery of genes that regulate normal or leukemic hematopoietic stem cells (HSC) self-renewal, the number of validated candidates remains low, due largely to the unavailability of functionally pure stem cell populations. Moreover, it is often difficult to identify the normal counterpart cell from which leukemia originated, further complicating studies based on comparative gene expression. In this study, we used a series of recently characterized Hoxa9 + Meis1 acute myeloid leukemias (AML) derived from fetal liver (FL) cells (Wilhelm BT et al., submitted). These leukemias are remarkably similar in several aspects including their L-HSC frequency (between ∼1 in 100 to 350) except for one leukemia (FLA2) in which 70% of the cells show repopulation ability (i.e., L-HSC). We reasoned that comparative mRNA profiling of FLA2 to the phenotypically similar FLB1 (0.3% L-HSC) might identify genes uniquely associated with L-HSC self-renewal. We observed a 2–3-fold upregulation of Gpx3 in FLA2, which was confirmed by qRT-PCR. In accordance with this, all 14 of the tested Gpx3 promoter region CpG sequences were methylated in FLB1 and hypomethylated in FLA2 cells. The higher expression of GPx3 in FLA2 was confirmed at the protein level and reflected in elevated glutathione peroxidase activity in comparison to FLB1. Importantly, we also observed in FLA2 a relative reduction in reactive oxygen species (ROS) level (DCFDA) and a concomitant decrease in p38 MAPK activation (western blot and mass spectrometry). The correlation of Gpx3 levels with L-HSC frequency could be reflective of their functional dependence on this enzyme. FLA2 cells being difficult to manipulate ex vivo, to address this we utilized retroviruses encoding shRNAs and a GFP reporter to explore the in vivo function of FLA2 cells with downregulated Gpx3. The decrease in percentage of GFP+ donor cells when leukemia became apparent (∼19 days) from that of populations initially transplanted, was 4-fold higher following Gpx3 knockdown in comparison to shLuciferase transduction. Moreover, those shGpx3 infected FLA2 remaining at day 19 displayed a 3-fold decrease in GFP mean fluorescence intensity relative to their control counterparts. These results show that GFPhigh cells were selectively depleted, and suggest that Gpx3 is critical for the competitiveness of L-HSCs. Because redox metabolism has been implicated in HSC self-renewal, we also analyzed its expression and function in normal HSC to gain further insight into the role of GPx3 in stem cell activity. Interestingly, compared to FL-HSCs, isolated 3 and 4 week bone marrow (BM), HSCs exhibited a 39- and 6-fold decrease in Gpx3 expression, respectively. A correlation of Gpx3 levels with enhanced self-renewal was also observed in vitro as overexpression of several nuclear determinants of HSC expansion such as Hoxb4, NA10HD, Klf10 and Prdm16 promoted Gpx3 expression by 3.2 to 19.2-fold. We next infected BM cells enriched for HSCs with retroviruses carrying shRNAs to Gpx3. shRNA targeting of Gpx3 dramatically inhibited hematopoietic reconstitution. Transplantations of sublethally irradiated recipients indicated that Gpx3 knockdown significantly impaired both early and late donor-derived hematopoiesis. These results suggest that GPx3 is critical for repopulation mediated by both short and long-term repopulating cells. In reciprocal gain-of-function experiments, Lin-CD150+CD48- cells engineered to overexpress Gpx3, showed a marked competitive advantage over controls when transplanted following a 7-day ex vivo culture step. Insertional mutagenesis was ruled out as proviral integration analyses of six recipients confirmed polyclonal hematopoiesis. Moreover, some mice were in part reconstituted by the same clones, indicating that self-renewal occurred in vitro prior to transplantation. Phenotypic analysis of late-transplant hematopoietic tissues showed that Gpx3-transduced cells contributed to lymphoid and myeloid repopulation, confirming their multipotentiality. Together, these results indicate that Gpx3 enhances HSC expansion ex vivo possibly through modulation of self-renewal activity. In conclusion, a unique model of primary L-HSC was exploited to identify Gpx3 as a critical determinant for the competitiveness of L-HSCs and complementary experiments demonstrated a key role for this gene in normal HSC self-renewal. Disclosures: No relevant conflicts of interest to declare.

Introduction Acute leukemia is characterized by abnormalities affecting the self-renewal of hematopoietic stem cells and differentiation of myeloid or lymphoid lineages. Our lab is focused on developing a better understanding of the biology of acute leukemias involving the KMT2A-MLLT3 fusion that is commonly found in pediatric leukemias. Methods/Results By transducing healthy CD34+ cord blood cells with the KMT2A-MLLT3 fusion and xenografting them into immunodeficient mice, we can generate human acute leukemias where the genetic background of the initial donor is defined. We have used this model system to interrogate the epigenetic changes that occur at the earliest stages of leukemogenesis, something which is not possible with patient samples. Interestingly, two distinctive DNA methylation patterns were observed in model myeloid and lymphoid samples, although acute leukemias can arise from the same molecular events. Furthermore, our ChIP-seq and ATAC-seq data revealed that the KMT2A-MLLT3 induction has a significant impact on chromatin rearrangement and distribution of histone marks. By correlating these epigenetic data with gene expression data from patient and model leukemic cells, we identified novel potential biomarkers that are specifically highly expressed and epigenetically regulated in KMT2A-rearranged leukemias. The pharmacological inhibition of one of these genes, encoding a GPCR, leads to significant apoptosis in leukemic cell lines highlighting its essential role in leukemogenesis. Conclusion An epigenetic reprogramming is required for KMT2A-MLLT3 acute leukemia development. This study also highlights novel biomarkers that are not only of potential clinical interest, but also likely reflect important aspects of the underlying biology of KMT2A-translocated leukemias.