Search

Our goal is to identify oncogenic loci in regions of recurrent DNA copy number alterations in cancer. Constitutional trisomy 21 (Down syndrome) carries a 20-fold increased risk of B-ALL, and chr.21 gains are the most common acquired aneuploidy in B-ALL. Interstitial amplification in the chr.21q22 region (iAMP21) is also a recurrent finding in B-ALL and carries a poor prognosis. However, the gene(s) on chr.21 responsible for this association remain unclear. We studied the Ts1Rhr mouse, which carries germline triplication of 31 genes homologous to human chr.21q22. Chr.21q22 triplication was sufficient to promote B cell autonomous self-renewal and maturation defects, and cooperated with BCR-ABL or CRLF2 with JAK2 R683G to accelerate leukemogenesis. Chr.21q22 triplication also resulted in histone H3K27 hypomethylation at gene promoters, and the expression signature of triplicated B cells was enriched for genes targeted by polycomb repressor complex 2 (PRC2), which trimethylates H3K27. Thus, chr.21q22 triplication may deregulate B cell development by causing H3K27 hypomethylation at genes critical for progenitor cell growth. In support of this hypothesis, pharmacologic inhibition of PRC2 function was sufficient to confer self-renewal in wild-type B cells, while inhibition of H3K27 demethylases blocked self-renewal induced by chr.21q22 triplication. In three independent B-ALL cohorts, PRC2/H3K27 gene signatures distinguished leukemias with +21 from those without, validating the same biology in human disease. One of the 31 triplicated genes, HMGN1, encodes a nucleosome binding protein known to modulate chromatin structure and facilitate transcriptional activation. When we overexpressed HMGN1 in BaF3 proB cells, H3K27me3 decreased proportionally to the level of overexpression. We next knocked down each of the 31 triplicated genes with lentivirally-expressed shRNAs (5 per gene) and assessed the effects on growth of Ts1Rhr and wild-type primary B cells. Strikingly, Hmgn1 was the top scoring gene and all 5 hairpins targeting Hmgn1 were depleted in the assay. Finally, we studied transgenic mice (HMGN1_OE) that overexpress human HMGN1 (∼2-fold total overexpression). HMGN1_OE mice had a defect in B cell maturation, increased proB colony forming capacity, and a transcriptional signature overlapping with that of triplication of all 31 Ts1Rhr genes. In a bone marrow transplant model driven by BCR-ABL, recipients of HMGN1_OE bone marrow developed B-ALL with decreased latency (median 33 days vs not reached) and increased penetrance (17/18 vs 4/17 mice died by 80 days; leukemia-free survival difference P Citation Format: Andrew A. Lane, Bjoern Chapuy, Charles Y. Lin, Trevor Tivey, Hubo Li, Elizabeth Townsend, Diederik van Bodegom, Tovah A. Day, Shuo-Chieh Wu, Huiyun Liu, Akinori Yoda, Gabriela Alexe, Anna Schinzel, Timothy J. Sullivan, Sebastien Malinge, Jordan Taylor, Kimberly Stegmaier, Jacob Jaffe, Michael Bustin, Geertruy te Kronnie, Shai Izraeli, Marian Harris, Kristen Stevenson, Donna Neuberg, Lewis B. Silverman, Steven E. Sallan, James E. Bradner, William C. Hahn, John D. Crispino, David Pellman, David M. Weinstock. Triplication of HMGN1 promotes B cell acute lymphoblastic leukemia (B-ALL) through suppression of H3K27me3. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 433. doi:10.1158/1538-7445.AM2014-433

A recent study has demonstrated that Notch is a positive regulator of murine megakaryopoiesis.[1][1] In this issue of Blood , Poirault-Chassac et al report the opposite: Notch inhibits the terminal differentiation of human megakaryocytes.[2][2] Does this discrepancy arise from species-specific

Members of the GATA family of zinc-finger transcription factors have critical roles in a variety of cell types. GATA-1, GATA-2 and GATA-3 are required for proliferation and differentiation of several hematopoietic lineages, whereas GATA-4, GATA-5 and GATA-6 activate cardiac and endoderm gene expression programs. Two GATA cofactors have recently been identified. Friend of GATA-1 (FOG-1) interacts with GATA-1 and is expressed principally in hematopoietic lineages, whereas FOG-2 is expressed predominantly in heart and brain. Although gene targeting experiments are consistent with an essential role for FOG-1 as an activator of GATA-1 function, reporter assays in transfected cells indicate that FOG-1 and FOG-2 can act as repressors. We have cloned a Xenopus laevis homologue of FOG that is structurally most similar to FOG-1, but is expressed predominantly in heart and brain, as well as the ventral blood island and adult spleen. Ectopic expression and explant assays demonstrate that FOG proteins can act as repressors in vivo, in part through interaction with the transcriptional co-repressor, C-terminal Binding Protein (CtBP). FOG may regulate the differentiation of red blood cells by modulating expression and activity of GATA-1 and GATA-2. We propose that the FOG proteins participate in the switch from progenitor proliferation to red blood cell maturation and differentiation.

[Comment on Tober et al, page 1433, and][1][Ghinassi et al, page 1460][2]Megakaryocytes and erythroid cells share a common precursor, the megakaryocyte/erythroid progenitor (MEP). Two reports in this issue of Blood provide new insights into megakaryocyte ontogeny and the plasticity of MEPs. Shortly

Acquired somatic mutations in the transcription factor GATA1 are a defining feature of acute megakaryocytic leukemia in children with Down syndrome. A new study shows that similar inherited GATA1 mutations do not promote leukemia in the absence of trisomy 21 but lead to defects in multiple hematopoietic lineages.

Comment on Ge et al, page [1570][1] Children with Down syndrome (DS) show a remarkable predisposition to develop acute megakaryocytic leukemia. In this issue of Blood, Ge and colleagues identify a set of genes that are differentially expressed in DS versus non-DS megakaryocytic leukemias. These

We have recently shown that small molecule inhibitors of Aurora A kinase (AURKA) including dimethyfasudil and MLN827 induce polyploidization and differentiation of normal and malignant megakaryocytes, as evidenced by increased DNA content and upregulation of the cell surface markers CD41 and CD42. Furthermore, our pre-clinical studies demonstrate that MLN8237 shows potent anti-leukemia activity and anti-myelofibrotic activity in MPNs by promoting polyploidization and terminal differentiation of abnormal megakaryocytes. To determine the mechanism by which AURKA inhibitors ameliorate the leukemic and myelofibrotic phenotypes, we have examined the functional requirement for Aurka in adult hematopoiesis. To circumvent the embryonic lethality of the germline knockout and study the requirement for AURKA in adult hematopoiesis, we induced deletion in Aurka floxed mice with Mx1-Cre. Complete loss of AURKA caused a rapid and profound defect in hematopoiesis, with the mice developing pancytopenia and marked hypocellularity of the bone marrow. Notably, we observe an increase in the proportion of CD41 and CD42-positive megakaryocytes in contrast to the deficiency of myeloid and lymphoid cells in the bone marrow. To determine whether the observed defects are cell autonomous, we transplanted Aurkafl/fl, Mx1-Cre bone marrow cells to lethally irradiated recipients, confirmed engraftment, and induced deletion with pIpC. Upon AURKA deletion, the transplanted mice develop an identical phenotype to the one observed in the conditional knockout mice, demonstrating a cell autonomous requirement for AURKA during hematopoiesis. Moreover, in competitive transplantation experiments, Aurka-/- cells are selectively lost and fail to contribute to hematopoiesis. Mechanistically, loss of AURKA led to a significant degree of apoptosis in hematopoietic cells, likely due to mitotic catastrophe resulting from impaired chromosome segregation. To bypass the requirement for AURKA in progenitor cells and examine AURKA function in the megakaryocyte lineage, we deleted AURKA in megakaryocytes ex vivo by infecting Aurka floxed bone marrow with MSCV-Cre. We found that deletion of AURKA resulted in increased CD41 and CD42 expression as well as increased ploidy of the megakaryocyte fraction. Together these results are consistent with a selective differentiation effect of AURKA deficiency on megakaryocytes. To investigate whether AURKA modulates differentiation of megakaryocytes through interactions with lineage specific transcription factors, we performed co-immunoprecipitation experiments between AURKA and megakaryocyte transcription factors containing consensus AURKA phosphorylation sites in primary megakaryocytes. Results confirmed a robust interaction between AURKA and p45 NF-E2. Additionally, we found by an in vitro kinase assay that AURKA phosphorylates p45 NF-E2 on the S170 residue, as the S170A mutant of NF-E2 cannot be phosphorylated by AURKA. To explore the functional importance of NF-E2 phosphorylation at S170, we created a phosphomimetic mutant, S170D, and assayed its ability to induce megakaryocyte differentiation in primary cultures. While overexpression of wild-type NF-E2 significantly increased the expression of the megakaryocyte differentiation markers CD41 and CD42, overexpression of the S170D mutant was markedly less effective in promoting megakaryocyte differentiation, indicating that the loss of phosphorylation promotes NF-E2 activity and that NF-E2 activity may be suppressed by AURKA phosphorylation. Finally, to determine if AURKA inhibitors promote megakaryocyte differentiation by allowing activation of NF-E2, we knocked-down down NF-E2 in megakaryocytic cell lines and subsequently treated with diMF or MLN8237. Strikingly, cells with NF-E2 knocked-down displayed significantly less CD41 and CD42 expression in response to AURKA inhibitor treatment compared to control knockdowns. This experiment demonstrates that NF-E2 is required for the differentiation effect caused by AURKA inhibition. Taken together, our data show that Aurora A kinase is required for adult hematopoiesis and that Aurora A regulates NF-E2 function during megakaryocyte differentiation. Our results also reveal that inhibition of AURKA kinase activity couples polyploidization and terminal differentiation of megakaryocytes. Disclosures: Crispino: Sanofi: Research Funding.

We recently reported that the induction of polyploidization of malignant megakaryocytes shows great promise as a new therapy for acute leukemia. Polyploidization inducers such as dimethylfasudil (diMF) and MLN8237, both of which target Aurora A kinase (AURKA), induce proliferation arrest, polyploidization, expression of megakaryocyte differentiation markers and apoptosis of leukemic megakaryocytes in vitro and in vivo. Since megakaryocytes in primary myelofibrosis (PMF) show impaired polyploidization and maturation, and likely directly contribute to the disease, we predicted that polyploidization inducers would provide a new therapeutic strategy. To determine the effect of these compounds on the growth of MPN cells, we first treated the JAK2 V617F mutant megakaryocytic SET2 cell line with varying doses of MLN8237 and diMF. Both compounds effectively and dose dependently inhibited proliferation, induced polyploidization and upregulation of lineage specific markers CD41 and CD42, and increased apoptosis. Furthermore, MLN8237 synergized with ruxolitinib to induce apoptosis of the SET2 cells and also potently induced growth arrest of JAK2 inhibitor persistent SET2 cells. We observed a similar polyploidization and differentiating activity of MLN8237 and diMF on megakaryocytes derived from primary human PMF progenitors. The ability of these agents to induce polyploidization was specific, as the non-megakaryocyte fractions of the cultures were not affected. Next, we assayed the activity of polyploidization inducers on progression of MPNs in two mouse models: JAK2V617F conditional knockin mice and mice engrafted with MPLW515L expressing bone marrow progenitors. Of note, spleens from both mouse models displayed a robust increase in both total and phosphorylated forms of AURKA relative to control animals, further suggesting that AURKA is a rational target in this disease. We first assayed the activities of MLN8237 and diMF in the MPLW515L bone marrow transplantation model. Recipient mice develop a rapid MPN characterized by leukocytosis, thrombocytosis and bone marrow fibrosis. Both MLN8237 and diMF reduced the disease burden, as evidenced by significant reductions in the liver and spleen weights, white cell counts and platelet counts. Both compounds also led to a significant decrease of fibrosis in the bone marrow, diminished infiltration of megakaryocytes and granulocytes in the liver, and a profound reduction in the numbers of megakaryocytes within the spleen. Moreover, plasma levels of TGF-β a known myelofibrogenic cytokine, were decreased by more than 3-fold by the drug treatment. Both diMF and MLN8237 led to selective polyploidization of megakaryocytes in the spleen as well as marked reductions in the levels of p-AURKA. Of note, neither agent affected the extent of phosphorylation of STAT3 or STAT5. Therefore, we tested whether the combined use of a JAK inhibitor and a polyploidy inducer would show enhanced activity in vivo. Indeed, the combination of MLN8237 and ruxolitinib led to greater reductions in tumor burden in the MPLW515L mouse model than either agent alone. Similar results were obtained using the JAK2V617F knock-in model. To further validate our conclusion that AURKA is a target in PMF, we infected Aurkafl/fl floxed bone marrow progenitors with MPLW515L and transplanted the cells to irradiated recipients. Excision of both alleles of Aurka by Cre mediated recombination completely resolved the disease, while heterozygous deletion of Aurka significantly reduced the disease burden, in a manner similar to treatment with MLN8237. Given that heterozygous deletion of Aurka does not alter normal hematopoiesis in mice, the fact that a 50% reduction in kinase expression was associated with a significant decrease in disease burden suggests that there is an effective therapeutic window in which AURKA inhibitors will be effective against MPN while sparing normal hematopoiesis. Although JAK inhibitors provide symptomatic relief, it is becoming clear that they are not curative. Thus, there is an urgent need to develop new agents to use in combination with JAK inhibitors. Our data reveal that inducing polyploidization and differentiation of dysplastic megakaryocytes in PMF ameliorates features of the disease both in vitro and in vivo. These results support the initiation of clinical studies that combine a JAK inhibitor with an AURKA inhibitor. Disclosures: Crispino: Sanofi: Research Funding.

The X-linked blood transcription factor GATA1 is required for the survival and maturation of erythroid cells. Loss of Gata1 causes profound anemia and related mid-gestation lethality in mouse models and human GATA1 mutations are associated with congenital dyserythropoietic anemia, congenital thrombocytopenia, porphyria, Diamond-Blackfan anemia (DBA) and acute megakaryoblastic leukemia in children with Down syndrome (DS-AMKL). In DBA, DS-AMKL, and a subset of congenital anemia, GATA1 mutations cause the exclusive production of a shorter GATA1 isoform, termed GATA1s. The GATA1s isoform retains both zinc fingers and binds DNA and cofactors accordingly, but lacks the N-terminal 83 residues of full-length GATA1. Intriguingly, exclusive production of GATA1s promotes AMKL in the context of Down syndrome (trisomy 21), while disomic individuals harboring the same type of mutations suffer from anemia and severe red cell defects. A mutation causing exclusive production of Gata1s in mice does not appear to affect adult hematopoiesis, but has been shown to cause a marked expansion of megakaryocytes in the fetal liver. The effects of the replacement of Gata1 with Gata1s during fetal liver erythropoiesis, however, remain uncharacterized. To investigate the effects of exclusive Gata1s production during fetal hematopoiesis, we performed comprehensive phenotypic and mechanistic studies using Gata1s knock-in embryos. We found significant changes in myelo-erythroid differentiation beyond the known expansion of megakaryocytes. Flow cytometric analysis revealed altered erythroid differentiation at embryonic days 12.5 and 14.5, evidenced by decreased Ter119 and increased CD71 expression, characteristic of delayed erythroid maturation. Changes in expression of myelo-erythroid progenitor commitment markers were also discovered. Specifically, we observed marked decreases in pre-CFU-E and CFU-E committed progenitors and an increase in pre-GM and pre-MegE populations. Our findings are consistent with a bias towards megakaryopoiesis at the expense of erythroid commitment caused by the expression of Gata1s in place of full-length Gata1. A shift in differentiation was also observed in the embryonic granulocyte/ macrophage lineage, with an increased generation of macrophages with fewer developing granulocytes. Mechanistically, we found that expression of erythroid -specific Gata1 target genes such as Alas2, Slc4a1 and Klf1 are markedly reduced in the Gata1 knock-in erythroid cells, indicating that Gata1s is a less effective transcriptional activator than full-length Gata1. In particular, given that Klf1 functions to promote erythroid specification downstream of the megakaryocyte-erythrocyte progenitor, our results suggest that Gata1s may promote megakaryocyte differentiation at the expense of erythroid differentiation, in part, by failing to activate Klf1. Taken together, our studies demonstrate a stage-specific requirement for full-length Gata1 during embryonic erythropoiesis. Furthermore, erythroid defects associated with exclusive production of Gata1s in humans may result from an incomplete activation of the erythroid transcriptional program. Disclosures: Crispino: Sanofi: Research Funding.

The dual specificity tyrosine-regulated kinase 1A (DYRK1A) phosphorylates a growing number of proteins to influence various aspects of cell biology, including survival, apoptosis, proliferation, and neuronal development and function. Encoded on human chromosome 21, the DYRK1A locus resides in the Down syndrome critical region (DSCR) and has been implicated in the pathogenesis of neurocognitive deficits that result from trisomy 21. We recently demonstrated a prominent tumor-promoting role for DYRK1A in acute megakaryoblastic leukemia in children with DS (DS-AMKL). This discovery prompted us to ask whether DYRK1A is important for normal hematopoiesis, and if additional types of leukemia result from enhanced DYRK1A activity. Apart from our study in DS-AMKL, the involvement of DYRK1A in normal and malignant hematopoiesis has not been described. Dyrk1a germline knockout mice die mid-gestation; therefore, we generated conditional knockout (CKO) mice to allow inducible deletion of the kinase in hematopoietic cells with Mx1-Cre. In the bone marrow, CKO mice exhibit significant reductions of the hematopoietic stem cell-enriched LSK and common lymphoid progenitor populations (5-fold and >20-fold reductions vs. controls, respectively). In B cell lineage, CKO mice have 2-fold reductions of small pre-B and immature B cells, and a 5-fold reduction of mature recirculating B cells. In the thymus, CKO mice show dramatic loss of CD4+/CD8+double positive cells (DP, >30-fold fewer at 4 weeks). Furthermore, DYRK1A-deficient cells fail to reconstitute lymphoid cells in competitive bone marrow transplantation experiments and give rise to virtually no pre-B cell colonies in ex vivo assays. In contrast to these lymphoid defects, myeloid development is largely normal, with slightly increased numbers of erythroid precursors, granulocytes, and monocytes. To understand the mechanisms by which DYRK1A controls lymphopoiesis, we performed microwestern blot arrays. This approach allowed us to simultaneously screen >200 antibodies to identify differentially expressed proteins between CKO and control lymphoid cells. We found increased levels of cell cycle regulatory proteins, such as CDK1, suggesting that DYRK1A modulates cell cycle status during lymphoid development. Indeed, cell cycle analysis revealed that there were 20-30% fewer quiescent (G0) cells among CKO small pre-B cells and DP thymocytes, two developmental stages characterized by exit from the cell cycle following a proliferative burst. Among its targets, DYRK1A is known to contribute to DREAM (DP, Rb-like, E2F, and MuvB) complex assembly via phosphorylation of LIN52. The DREAM complex, not previously described in hematopoiesis, promotes quiescence by repressing the transcription of numerous cell cycle-associated genes during G0. RNA sequencing analysis in flow-purified G0 cells confirmed that DREAM complex targets are markedly de-repressed in the affected CKO populations, highlighting the importance of DYRK1A/DREAM complex interaction during lymphoid cell cycle exit. Both B- and T-lineage ALL express high levels of DYRK1A relative to other tumor types. Having observed that DYRK1A is essential for lymphoid cells, we next asked whether targeting DYRK1A with small molecule inhibitors would be an effective therapy for ALL. We found that EHT 1610, a highly selective DYRK1 inhibitor (Leblond B., Casagrande A.-S., Désiré L., Foucourt A., Besson T., DYRK1 inhibitors and uses thereof, WO 2013026806, 2013.02.28), dose-dependently induced apoptosis in B- and T-ALL cell lines and primary human pediatric ALL samples. Thus far, all cell lines and all primary samples tested were exquisitely sensitive to EHT 1610. Moreover, EHT 1610 induced apoptosis of primary ALL cells that were resistant to cytarabine, suggesting that DYRK1A inhibitors may be used in combination with standard ALL therapies for refractory or relapsed cases. Together, our data establish novel essential roles for DYRK1A in both normal and malignant lymphoid development and provide rationale for the design of DYRK1A-targeted ALL therapies. Disclosures: Leblond: Diaxonhit Therapeutics: Employment. Casagrande:Diaxonhit Therapeutics: Employment. Désiré:Diaxonhit Therapeutics: Employment. Crispino:Sanofi: Research Funding.

The full set of consensus sequences at the 5' splice site is recognized during splicing of pre-mRNA in extracts depleted of U1 snRNP. High concentrations of HeLa SR proteins or purified SC35 alone promote the splicing of specific RNA substrates, bypassing the requirement for U1 snRNP in formation of the U2 snRNP-pre-mRNA complex. Under these conditions, mutations in the substrate that increase the sequence complementarity between U6 snRNA and the 5' splice site region can facilitate splicing. This provides additional strong evidence that U1 snRNP is not essential for splicing. Thus, the consensus sequence at the 5' splice site is probably recognized twice during splicing of most introns; however, some pre-mRNAs could potentially be processed in the absence of interactions with U1 snRNP in regions of the nucleus containing high concentrations of SR protein.

Abstract 3450 Our goal is to generate sufficient PLTs from ex vivo-generated MKs for clinical utility in PLT transfusions. A critical step in this process begins with ex vivo-generated hMKs and deriving clinically relevant PLTs. We reported that infused mature, high-ploidy murine (m) MKs derived from fetal liver (FL) cells increased PLT counts in recipient mice in a clinically relevant fashion, thus avoiding the need to generate ex vivo functional PLTs. We examined whether this strategy applies to hMKs derived from FL cells (gestational age, 17–21 weeks) and bone marrow (BM) cells in a xeno-transfusion model using immunodeficient NOD/SCID/IL-2Rγcnull (NSG) mice. Infused hPLTs isolated from blood had a half-life (T1/2) of 10 hours (hrs), compared to 24 hrs for infused murine PLTs. The hPLTs were functional in NSG mice as demonstrated by their incorporation into growing thrombi in situ. Both hFL hematopoietic mononuclear cells and hBM-CD34+ cells were cultured in serum-free media supplemented with optimized cytokine cocktails to generate hMKs. In contrast to the murine studies where the FL cell-derived mMKs were the most efficient source of derived mPLTs, FL cell-derived hMKs had low ploidy (0% ≥ 8N ploidy), gave rise to ∼16 PLTs/infused hMK, and had a short T1/2 (6 hrs). In contrast, 17% of hBM cell-derived MKs had a ploidy of ≥ 8N, and after infusion into NSG mice, resulted in a wave of MKs transiently entrapped in the pulmonary microvasculature and then over ∼0.5–3 hrs released PLTs with a T1/2 of 10 hrs, comparable to infused hPLTs. Maximally, we achieved a level of 5% of circulating total PLTs being derived from human cells with ∼32 PLTs/infused hMK. These hPLTs were normal in size, displayed normal levels of surface markers, were functional, and incorporated into growing thrombi. One strategy to increase hPLT yield is to expose developing MKs to drugs reported to increase MK maturation, thrombopoiesis, and/or facilitate hematopoietic progenitor cell expansion. Such drugs include dimethylfasudil (diMF) (an inhibitor of several kinases involved in polyploidization), UNC0638 (a G9a histone methyltransferase inhibitor), SR1 (an AhR antagonist), and nicotinamide (a sirtuin histone/protein deacetylases inhibitor). Although diMF promoted size and polyploidization of hMKs, diMF markedly worsened yield of PLTs/infused hMK and decreased PLTs T1/2 in vivo. UNC0638 led to significant cell expansion, but lowered hMKs ploidy and PLTs/infused hMK yield. Nicotinamide increased maturation, size and polyploidization of hMKs, but PLT release following MK infusion needs further study. Of note, SR1 that has been reported to promote the expansion of human HSC, not only increased size and ploidy of hMKs, but also hPLT release in vitro and in vivo. SR1-treated hMKs resulted in a 3-fold increased yield of normal size, T1/2 and functional PLTs/infused hMK compared to a DMSO-treated control. In summary, like mMKs, infused hMKs into mice release PLTs in the pulmonary vasculature though at a lower efficiency. Released hPLTs were functional and T1/2 was as expected. diMF enhanced MK ploidy, but worsened PLT yield and T1/2, while an AhR antagonist SR1 that also improved MK ploidy appears to markedly enhance yield of PLT/infused hMK, while maintaining T1/2. The ability of SR1 to enhance PLT release from induced pluripotent stem cells (iPSCs)-derived MKs remains to be tested, but this drug appears to be a strong candidate for a therapeutic strategy to take ex vivo-grown hMKs and generate PLTs in clinical relevant numbers. Disclosures: No relevant conflicts of interest to declare.

Abstract 1201 In acute megakaryocytic leukemia (AMKL), there is a failure of megakaryocytes to differentiate, become polyploid and stop dividing. We used an integrated screening approach that included chemical, proteomic and genetic screens to identify small molecules and their targets that control polyploidization and differentiation of normal and malignant megakaryocytes. We identified several small molecule inducers of polyploidy and used siRNA and proteomic target ID approaches to determine the cellular targets of the lead small molecule dimethylfasudil (diMF). Aurora Kinase A (AURKA) was identified as one of the top targets of diMF. AURKA is an attractive target in AMKL for several reasons. First, AURKA is overexpressed in AMKL cells. Second, at the biologically effective doses used in our cell-based assays, AURKA inhibition was selective for the megakaryocyte lineage. Third, AURKA inhibition by diMF or the selective AURKA inhibitor MLN8237 increased MK polyploidy, induced features of differentiation, blocked proliferation of AMKL blasts, and improved survival in an AMKL mouse model. AURKA is important in mitotic spindle assembly, mitosis, chromosomal alignment and segregation. Moreover, it is required for embryonic development, as Aurka−/− embryos fail to grow beyond the blastocyst stage. However, the extent to which AURKA is necessary for steady state hematopoiesis in adults is unknown. To investigate the necessity of AURKA in hematopoiesis, we utilized a conditionally targeted strain of mice (Aurkaflox/flox). To delete AURKA in megakaryocytes ex vivo, Aurkaflox/flox bone marrow cells were expanded, transduced with a retrovirus expressing Cre and GFP, and then cultured in the presence of THPO for 72 hours. We found that deletion of AURKA resulted in increased CD41 and CD42 expression as well as increased DNA content. Assays for apoptosis by Annexin V staining of Aurkaflox/flox cells infected with Cre also showed increased apoptosis in AURKA-deleted cells at 24 and 48 hours. To delete AURKA in vivo, we crossed Aurkaflox/flox mice to MX1-Cre mice and injected wild-type, heterozygous and homozygous floxed mice expressing MX1-Cre with pIpC every other day for six days. We found that deletion of AURKA in hematopoietic progenitors leads to pancytopenia, profound bone marrow defects and death within two weeks. Colony formation assays showed significantly decreased myeloid, erythroid and megakaryocyte colony formation with AURKA deficiency. Bone marrow histology displayed markedly hypocellular marrow, but curiously, flow cytometry revealed a significant increase in the percentage of CD41 and CD42 positive cells. This observation suggests that AURKA normally acts to restrain terminal differentiation of megakaryocytes and is consistent with the CD41 and CD42 inducing ability of AURKA inhibitors. To confirm that AURKA is the key target of our recently identified polyploidy inducers, we assayed the effects of diMF and MLN8237 on Aurka+/+, Aurka+/− and Aurka−/− megakaryocytes. 300 nM diMF and 100 nM MLN8237, concentrations that strongly induce polyploidy, did not increase MK polyploidization in Aurka−/− MKs. diMF and MLN8237 treatment increased polyploidy in Aurka+/− MKs with no significant difference in comparison to Aurka+/+ MKs. We also assayed the ability of wild-type or the T217D mutant of AURKA, which is resistant to inhibition by MLN8237, to reduce the induction of polyploidy caused by diMF and MLN8237 upon overexpression. CMK cells were infected with viruses harboring wild-type or T217D AURKA, treated with DMSO, 3 μM diMF or 30 nM MLN8237 for 72 hours, and then evaluated for DNA content. The increase in polyploidization induced by both compounds was significantly decreased in cells overexpressing the T217D mutant of AURKA. With overexpression of the wild-type AURKA, there was a trend towards reduction in polyploidy, but more variable effects and no significant difference. Thus, AURKA T217D overexpression reduced the ability of diMF and MLN8237 to induce polyploidization, consistent with our conclusion that diMF targets AURKA. Together, our data support a role of AURKA in megakaryocyte polyploidization and differentiation and show that AURKA is required for steady state hematopoiesis. The results also show that AURKA is the key target of diMF in the induction of polyploidization of megakaryocytes and support the development of Aurora A kinase inhibitors in clinical trials for AMKL. Disclosures: No relevant conflicts of interest to declare.

Abstract 115 Extra copies of chromosome 21 (polysomy 21) is the most common somatic aneuploidy in B-cell acute lymphoblastic leukemia (B-ALL), including >90% of cases with high hyperdiploidy. In addition, children with Down syndrome (DS) have a 20-fold increased risk of developing B-ALL, of which ∼60% harbor CRLF2 rearrangements. To examine these associations within genetically defined models, we investigated B-lineage phenotypes in Ts1Rhr mice, which harbor triplication of 31 genes syntenic with the DS critical region (DSCR) on human chr.21. Murine pro-B cell (B220+CD43+) development proceeds sequentially through “Hardy fractions” defined by cell surface phenotype: A (CD24−BP-1−), B (CD24+BP-1−) and then C (CD24+BP-1+). Compared with otherwise isogenic wild-type littermates, Ts1Rhr bone marrow harbored decreased percentages of Hardy fraction B and C cells, indicating that DSCR triplication is sufficient to disrupt the Hardy A-to-B transition. Of note, the same phenotype was reported in human DS fetal liver B-cells, which have a block between the pre-pro- and pro-B cell stages (analogous to Hardy A-to-B). To determine whether DSCR triplication affects B-cell proliferation in vitro, we analyzed colony formation and serial replating in methylcellulose cultures. Ts1Rhr bone marrow (B6/FVB background) formed 2–3-fold more B-cell colonies in early passages compared to bone marrow from wild-type littermates. While wild-type B-cells could not serially replate beyond 4 passages, Ts1Rhr B-cells displayed indefinite serial replating (>10 passages). Ts1Rhr mice do not spontaneously develop leukemia, so we utilized two mouse models to determine whether DSCR triplication cooperates with leukemogenic oncogenes in vivo. First, we generated Eμ-CRLF2 F232C mice, which express the constitutively active CRLF2 mutant solely within B-cells. Like Ts1Rhr B-cells, (but not CRLF2 F232C B-cells) Ts1Rhr/CRLF2 F232C cells had indefinite serial replating potential. In contrast with Ts1Rhr B-cells, Ts1Rhr/CRLF2 F232C B-cells also engrafted into NOD.Scid.IL2Rγ−/− mice and caused fatal and serially transplantable B-ALL. Second, we retrovirally transduced BCR-ABL1 into unselected bone marrow from wild-type and Ts1Rhr mice and transplanted into irradiated wild-type recipients. Transplantation of transduced Ts1Rhr cells (106, 105, or 104) caused fatal B-ALL in recipient mice with shorter latency and increased penetrance compared to recipients of the same number of transduced wild-type cells. By Poisson calculation, the number of B-ALL initiating cells in transduced Ts1Rhr bone marrow was ∼4-fold higher than in wild-type animals (1:60 vs 1:244, P=0.0107). Strikingly, transplantation of individual Hardy A, B, and C fractions after sorting and BCR-ABL1 transduction demonstrated that the increased leukemia-initiating capacity almost completely resides in the Ts1Rhr Hardy B fraction; i.e., the same subset suppressed during Ts1Rhr B-cell differentiation. To define transcriptional determinants of these phenotypes, we performed RNAseq of Ts1Rhr and wild-type B cells in methylcellulose culture (n=3 biologic replicates per genotype). As expected, Ts1Rhr colonies had ∼1.5-fold higher RNA abundance of expressed DSCR genes. We defined a Ts1Rhr signature of the top 200 genes (false discovery rate (FDR) 2,300 signatures in the Molecular Signatures Database (MSigDB) C2 Chemical and Genetic Perturbations with the Ts1Rhr signature identified enrichment in multiple gene sets of polycomb repressor complex (PRC2) targets and H3K27 trimethylation. Most notably, SUZ12 targets within human embryonic stem cells were more highly expressed in Ts1Rhr cells (P=1.2×10−6, FDR=0.003) and the same SUZ12 signature was enriched in patients with DS-ALL compared to non-DS-ALL (P=0.007). In summary, DSCR triplication directly suppresses precursor B-cell differentiation and promotes B-cell transformation both in vitro and by cooperating with proliferative alterations such as CRLF2 activation and BCR-ABL1 in vivo. Pharmacologic modulation of H3K27me3 effectors may overcome the pro-leukemogenic effects of polysomy 21. Disclosures: No relevant conflicts of interest to declare.

Abstract 2867 Primary myelofibrosis (PMF) is a clonal hematologic malignancy, which results from the transformation of a pluripotent hematopoietic progenitor cell. A major consequence of this transformation is increased hematopoiesis and an overproduction of abnormal blood cells. PMF is associated with bone marrow fibrosis, extramedullary hematopoiesis, increased numbers of circulating CD34+ cells, splenomegaly, and a propensity to evolve to AML. Patients also display anemia and thrombocytopenia and harbor abnormal, immature megakaryocytes (Mks) in their bone marrow and spleen. PMF patients can present well known mutations including JAK2V617F (65%), MPL (10%), TET2 (17%), CBL (6%), IDH (4%,), which are not specific to the disease and are also present in polycythemia vera, essential thrombocythemia and AML. We hypothesize that the genetic events associated with PMF, including MPL and JAK2 mutations, contribute to defects in Mk maturation, but that additional changes are needed to explain the striking abnormalities seen in PMF relative to the other myeloproliferative diseases. Although there have been studies to examine the aberrant gene expression program of CD34+ cells of PMF patients, we chose to examine the changes that occur in gene expression specifically in Mks as a way to better understand their abnormal differentiation and to determine their contribution to the disease. Primary CD34+ cells from PMF patients and healthy donors were cultivated in serum free media supplemented with recombinant TPO, BSA, liposomes, insulin and transferrin to support the growth of Mks. After 10 days of differentiation, we evaluated the cultures for proliferation, apoptosis and differentiation by flow cytometry. We found that PMF specimens gave rise to a lower percentage of mature (CD41+CD42+) cells as compared to healthy donors, but showed, a lower ploidy level, a greater proliferation and increased survival. These observations are consistent with the clinical observations that PMF bone marrow is characterized by an increased number of immature, dysplastic Mks. We used flow cytometry to collect two populations of cells for analysis: immature CD41+CD42− Mks, and CD41+CD42+ mature MKs. After sorting, we extracted RNA and performed whole genome microarray analysis with Illumina Human HT12-v4 arrays on cohorts of PMF and control specimens. Gene expression data were analyzed by GeneSpring and Gene Set Enrichment Analysis (GSEA). We found that the CD41+CD42− MKs derived from PMF progenitors showed reduced expression of GATA1 as compared to control cells, as expected based on previous study by Dr. Alessandro Vannuchi. GeneSpring analysis revealed that myeloid transcription factors, including CEBPa, GFI1, and SPI1 (PU.1), which are not expressed in normal MKs, are strikingly and significantly overexpressed in PMF samples. Moreover, c-myb, which regulates the erythroid/Mk cell fate decision, FOG-1 and AML1, are also overexpressed in PMF Mks. This aberrant myeloid gene expression program in PMF Mks is reminiscent of a similar defect we observed in Mks with reduced expression of GATA-1 and GATA-2. We predict that reduced levels of GATA-1 protein in PMF Mks, as reported by Dr. Alessandro Vannucchi and colleagues, is in part responsible for the aberrant growth and differentiation of the PMF Mks. Our data support the model that PMF Mks are defective in their ability to properly regulate expression of hematopoietic regulators. Further analysis by GSEA revealed that hematopoietic and cytokine pathways are among those that are highly enriched in PMF Mks. We recently reported that the molecules dimethylfasudil (diMF) and MLN9237 are able to selectively increase ploidy, Mk surface marker expression, and apoptosis of malignant Mks. We treated Mks derived from PMF progenitor cells with diMF and observed a high increase in polyploidization accompanied with a reduction of Mks proliferation. Thus, diMF is able to partially restore Mk differentiation of PMF cells, supporting the testing of polyploidy inducers in myelofibrosis patients. Disclosures: Pardanani: Sanofi-Aventis: Clinical trial support Other; YM BioSciences: Clinical trial support, Clinical trial support Other; Bristol-Myers Squibb: Clinical trial support, Clinical trial support Other.

Abstract 854 GATA-1-deficiency causes dysregulation of multiple genes and leads to aberrant megakaryopoiesis. How the dysregulated genes may contribute to abnormal megakaryocyte development and differentiation is largely unknown. In this study, we focused on Pstpip2, a gene that was upregulated in GATA-1low and GATA-1s mutant megakaryocytes. A strong GATA-1 binding site in the intron 1 of Pstpip2 gene was previously identified by GATA-1 ChIP-Seq studies. These observations strongly suggest that Pstpip2 is a GATA-1 target gene. By using ChIP-PCR we showed that GATA-1 indeed bound to Pstpip2 intron 1 region in megakaryocytes. We also observed that Pstpip2 was expressed in myeloid cells and hematopoietic progenitors, but not lymphocytes. Noticeably, two cell lines with GATA-1 deficiencies, G1ME and CMK, which approximate the MEP and megakaryoblast stages, respectively, showed moderate expression of Pstpip2. Moreover, K562 cells displayed significant upregulation of PSTPIP2 and concomitant downregulation of GATA-1 during TPA-induced megakaryocytic differentiation. To study its function in megakaryocytic differentiation, we modulated PSTPIP2 expression by overexpression or shRNA knockdown. Ectopic expression of PSTPIP2 in K562 or CMK cells impaired TPA-induced megakaryocytic differentiation as evidenced by decreased CD41 expression. In contrast, downregulation of PSTPIP2 by shRNA promoted CD41 expression and increased polyploidy. These findings suggest that failure to suppress PSTPIP2 due to loss of GATA-1 may contribute to abnormal megakaryopoiesis. To probe the mechanism by which PSTPIP2 inhibits megakaryocytic differentiation, we tested whether the interaction with its partner PEST-PTP or phosphorylation of key tyrosine residues in the protein by Src kinase is required. We created two mutants, W232A and Y323/333/F, which block its interaction with PEST-PTP or prevent tyrosine phosphorylation, respectively. Similar to WT PSTPIP2, both mutants retained the ability to inhibit megakaryocytic differentiation suggesting that inhibition of megakaryocytic differentiation does not depend on the PEST-PTP interaction or Src kinase phosphorylation. Separately, we have noted that PSTPIP2 also interacts with Fyn, Grb2, and Shc, suggesting that PSTPIP2 may affect signal transduction. Indeed, we found that ERK phosphorylation was downregulated by PSTPIP2. Since THPO/MPL signaling via JAK/STAT is the physiological pathway in megakaryocyte maturation rather than BCR-ABL signaling seen in K562 cells, we next analyzed the effect of PSTPIP2 on MPL signaling in G1ME cells. Consistent with our findings in K562 and CMK cells, both WT and mutant PSTPIP2 inhibited CD41 expression. Interestingly, ectopic expression of WT PSTPIP2 also caused cell cycle arrest at S phase in G1ME cells. PSTPIP2 also suppressed Erk and Shc phosphorylation. Considering the important role of MAPK/ERK signaling in megakaryocyte development, we believe that PSTPIP2 acts as a negative regulator of THPO-MPL signaling to suppress TPO-stimulated MAPK/ERK activation and subsequently inhibit megakaryopoiesis. Finally, we confirmed the function of PSTPIP2 in megakaryocyte differentiation using mouse bone marrow progenitor cells. Ectopic expression of PSTPIP2 significantly downregulated CD41 expression, reduced CFU-GM and CFU-Mk numbers, and reduced polyploidy, while knockdown of PSTPIP2 exhibited the opposite effects. Taken together, we have identified a new negative regulator of THPO-MPL signaling that is inhibited by GATA-1 in normal megakaryocyte development. Upregulation of PSTPIP2 may contribute to abnormal megakaryopoiesis observed in GATA-1-deficient megakaryocytes. Disclosures: No relevant conflicts of interest to declare.

Individual small nuclear ribonucleoproteins (snRNPs) U1, U2, and U4/U6 were removed from nuclear extracts of HeLa cells by antisense affinity depletion. Addition of a highly purified preparation of SR proteins fully restored splicing activity in reactions depleted of U1 snRNP but did not reconstitute splicing in reactions depleted of the other snRNPs. Affinity selection experiments revealed that spliceosomes lacking U1 snRNA formed in the U1 snRNP-depleted reactions reconstituted with SR proteins. Thus, high concentrations of SR proteins facilitate the assembly of precursor messenger RNA (pre-mRNA) into a spliceosome in the absence of interactions with U1 snRNP.

Abstract 3862 Somatic mutations in JAK2 and MPL are associated with the BCR-ABL negative myeloproliferative neoplasms (MPNs) including polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF). While oral JAK2 inhibitors improve peripheral blood counts and splenomegaly, these drugs show varying effects on JAK2 mutant allele burden and do not eliminate the malignant clone in humans or in animal models of MPN. Although much of the research to date has focused on JAK/STAT signaling, other pathways downstream of the class I cytokine receptors, including PI3K/AKT and ERK are also activated in MPNs. Our hypothesis is that persistent activation of these signaling pathways contributes to the progression of myeloproliferative neoplasms (MPNs). Multiple studies have shown that overexpression of activated JAK2 and MPL mutants (e.g. JAK2 V617F or MPL W515L) in primary murine bone marrow cells causes robust AKT, STAT3, and STAT5, JNK, ERK phosphorylation. To determine the extent to which these signaling pathways are involved in the disease, we cultured JAK2 V617F and MPL W515L expressing clones of the G1ME and 6133 megakaryocytic cell lines with a panel of small molecule kinase inhibitors. We discovered that inhibition of the PI3K/AKT and MAPK/JNK signaling pathways with triciribine and SP600125 respectively, potently suppressed growth of these cells by inducing G1 arrest and apoptosis. In contrast, inhibitors of Ras, ERK, Src family tyrosine kinase, and PKC failed to significantly inhibit proliferation of the JAK2 or MPL mutant expressing cells. Murine bone marrow cells transduced with MPL W515L show a dramatic expansion of megakaryocytes (CFU-MKs) and this expansion was abrogated in the presence of PI3K/AKT inhibitors suggesting a requirement for this pathway in aberrant megakaryocyte expansion. Since AKT inhibition showed the strongest effect, we assayed the activity of MK-2206, a potent and selective allosteric AKT inhibitor, on multiple models of MPNs. We discovered that MK-2206 induced proliferative arrest and apoptosis accompanied by suppression of PI3K/AKT signaling in G1ME and 6133 cells expressing MPLW515L. MK-2206 also potently inhibited liquid culture growth and colony formation of PMF patient CD34+ cells in vitro. Finally, in preliminary murine transplants experiments with MPLW515L expressing bone marrow progenitors, treatment with MK-2206 led to significant reductions in peripheral blood leukocytosis and extramedullary hematopoiesis. Together, these findings demonstrate that the PI3K/AKT axis represents a rational target for therapy in human myeloproliferative neoplasms. Disclosures: No relevant conflicts of interest to declare.

Abstract 64 Megakaryocytes are one of the few cell types that undergo a modified form of the cell cycle termed endomitosis, in which cells skip the late stages of mitosis to become polyploid. Murine and human megakaryocytes commonly reach modal ploidy states of 32N and 16N, respectively, and can sometimes achieve DNA contents as high as 64N. Polyploidization is associated with upregulation of megakaryocyte lineage specific genes, proplatelet formation and expression of genes related to apoptosis. RNA expression array studies have shown that high ploidy states are strongly correlated with megakaryocyte differentiation and maturation. Importantly, the choice of a megakaryocyte to undergo polyploidization and differentiation is inextricably linked to exit from the proliferative cell cycle. Given that megakaryocytes in patients with essential thrombocythemia are hyperproliferative and that those in primary myelofibrosis fail to undergo normal differentiation or polyploidization, we hypothesized that small molecule inducers of polyploidization would drive these cells to exit the proliferative cell cycle and undergo terminal differentiation or death. In collaboration with the Broad Institute, we performed a high throughput screen and identified small molecules that induce polyploidization and proliferative arrest of malignant megakaryocytes, including those that express MPLW515L and JAK2 V617F. We have shown that these compounds, including the Rho kinase inhibitor dimethylfasudil (diMF), selectively increase polyploidization, expression of megakaryocyte cell surface markers, and apoptosis of murine and human megakaryocytic cell lines and primary cells. Furthermore, diMF blocked the growth of primary human AMKL blasts both in vitro and in vivo. With respect to MPNs, diMF showed potent activity against megakaryocytic cell lines and primary cells expressing either JAK2 V617F or MPL W515L alleles commonly associated with these disorders. diMF inhibited proliferation, induced polyploidization and upregulation of lineage specific markers CD41 and CD42, and increased apoptosis of megakaryocytes transduced with JAK2 V617F or MPLW515L. diMF also significantly reduced megakaryocyte colony forming units (CFU-MK), and induced polyploidization and differentiation of bone marrow and fetal liver megakaryocytes from Gata1 mutant mice, which develop a PMF-like disease. Given that diMF induces growth arrest, polyploidization and apoptosis of cells that express activated mutants of JAK2 and MPL, we predicted that it, as well as other small molecule inducers of polyploidy, would be efficient at restraining aberrant megakaryocyte proliferation in both PMF and ET. To assay the effectiveness of diMF in these disorders, we treated peripheral blood mononuclear cells from patients with either PMF or ET with diMF and monitored growth and maturation of megakaryocytes. We discovered that diMF induced polyploidization and subsequent apoptosis of both types of MPN primary samples. diMF also reduced CFU-MK of these MPN patient samples. Next, we assessed the activity of diMF in a model of MPN in which congenic recipients of MPLW515L transduced Balb/C bone marrow cells develop a rapid MPN characterized by leukocytosis, thrombocytosis, bone marrow fibrosis, and death. diMF led to a significant decrease of fibrosis in the bone marrow, diminished infiltration of megakaryocytes and granulocytes in the liver, and a profound reduction in the numbers of megakaryocytes within the spleen of a mouse model of PMF. diMF also led to a significant reduction in the platelet count and a trend towards decreased white cell count, with no effect on hematocrit. Overall, diMF results were comparable to intermediate doses of INCB16562. These encouraging results strongly suggest that diMF induces a decline in megakaryocyte lineage, which leads to reduction in platelet count, and support pre-clinical development of diMF for megakaryocytic subtypes of MPNs. Of note, diMF did not inhibit the phosphorylation of Stat5 or Stat3, suggesting that it acts through a mechanism distinct from JAK2 inhibitors. Interestingly, combination of diMF with a selective JAK2 inhibitor greatly enhanced the efficacy of diMF to inhibit proliferation and induce apoptosis in MPLW515L transduced megakaryocytic cell line. These data support combining JAK inhibition and induction of megakaryocyte polyploidy as a new therapeutic strategy for MPNs. Disclosures: No relevant conflicts of interest to declare.

Abstract 3451 Down syndrome (DS) is associated with an increased frequency of Acute Megakaryoblastic Leukemias (AMKL). In the context of DS, AMKL is usually preceded by the development of a transient myeloproliferative disorder (TMD). Up to 10% of DS children may develop this disorder, and approximately 20–30% of those with TMD go on to develop AMKL, which may or may not be preceded by a return to normal hematopoiesis. While trisomy 21 (tri21) alone is not sufficient to develop AMKL, it is clear that the acquisition of somatic mutations in GATA1 is a requirement for the development of DS-AMKL. Much has been studied in relation with this genetic mutation, however not much is known about the epigenetic changes accompanying the development of AMKL in DS patients. Given the step-wise nature of the development of DS-AMKL, we hypothesized that distinct and cumulative epigenetic changes would be associated with each of the phases of the disease. In order to address this hypothesis we performed DNA methylation profiling of a cohort of 7 primary DS-AMKL samples using the latest design of the HELP microarray assay, which covers ∼230,000 CpG sites at ∼22,000 RefSeq promoter regions. In order to better understand how aberrant epigenetic patterns become established during the development of this disease we compared these DNA methylation profiles to those of mononuclear cells from normal fetal liver (FL-MNC) (n=4), tri21 FL-MNC (n=4) and TMD blasts (n=6). We identified a set of 711 unique genes that displayed significant loss of methylation in Tri21 FL-MNC with respect to control FL-MNC (FDR < 5% and methylation delta ≥ 25%). However, when tri21 FL-MNC were compared to the GATA1s positive TMD blasts, we observed that TMD samples acquired significant hypermethylation of 346 unique genes at the same significance cutoff. A direct comparison of the genes affected by these two opposing waves of methylation changes demonstrated that different sets of genes were being targeted at each stage. Next we compared the DNA methylation profiles of TMD and DS-AMKL blasts, and found them to be virtually identical, indicating that an epigenetic imprint becomes established at the TMD phase, which is still present at the leukemic phase. Finally we compared DS-AMKL to a cohort of 8 non-DS AMKL, and observed that DS-AMKL samples display significant hypomethylation of 267 unique genes, indicating that these two diseases are not only genetically but also epigenetically distinct. In determining the location of the epigenetic changes acquired at the different stages of the disease we found that, similar to what had been previously reported for gene expression, DNA methylation changes did not localize solely to chr 21, but that these changes were distributed along all chromosomes. However, detailed analysis of chr 21 revealed that changes on this chromosome preferentially targeted a specific region on the long arm, the Down Syndrome Critical Region (DSCR). This region becomes significantly hypomethylated in Tri21 FL-MNC and remains hypomethylated through all the stages of the disease. Several groups have sought to identify candidate genes on the DSCR that may contribute to the pathogenesis of the disease. Two genes, BACH1 and CXADR, have been identified in common by two different studies as being overexpressed in DS-AMKL vs. Non-DS AMKL. Both of these genes displayed aberrant hypomethylation at their promoter regions in Tri21 FL MNC, TMD and DS-AMKL samples: such hypomethylation was absent from all the control samples. Finally, given that GATA1 mutations are a requirement for the development of DS-AMKL, we compared the different aberrant DNA methylation signatures to the normal GATA1 targets identified by ChIP sequencing experiments, and found 20–25% overlap. In summary, we have completed the first comprehensive DNA methylation study of the developmental phases of DS-AMKL and have demonstrated that a) specific epigenetic changes are associated with the different stages of the disease, b) these changes occur in ‘waves' of sequential hypo and hypermethylation, c) aberrant DNA methylation on chr 21 preferentially targets the DSCR and results in aberrant overexpression of the associated genes, and d) epigenetic profiles of TMD and DS-AMKL are virtually identical. These observations lead us to believe that an epigenetic memory established at the TMD stage may be at least in part contributing to the development of full-blown AMKL at a later time point. Disclosures: No relevant conflicts of interest to declare.

Abstract 2321 Enucleation of erythroblasts is the final step of differentiation before reticulocytes enter circulation in mammals. In culture, enucleation efficiencies vary widely, between 0 and 90%: improving this process is a necessary step for large-scale production of reticulocytes for transfusion. Recently, we demonstrated that the formation, movement and coalescence of vesicles and vacuoles play a key role in enucleation. Moreover, we showed that vacuolin-1, a small molecule that induces the formation and fusion of vacuoles, increases the enucleation of primary erythroblasts. Separately, we recently discovered that the anti-apoptotic and chromosomal passenger protein survivin plays a novel role in enucleating erythroblasts. Survivin interacts with clathrin and EPS-15, proteins that play important roles in vesicle trafficking, and together these proteins contribute to enucleation. In order to further understand the specific role played by survivin in vesicle trafficking and enucleation, we treated fetal liver Ter119- cells derived from survivin fl/fl mice with TAT-Cre to excise the survivin gene and assess the consequences of survivin loss on enucleation. We observed a significant reduction in the extent of enucleation upon deletion of survivin. Of note, survivin depleted cells harbored fewer numbers of vacuoles compared to control cells, and the vacuoles that were present were smaller than normal. Survivin deleted cells also lacked autophagolysosomes, which are formed by fusion of late endosomes/lysosomes to autophagosomes. Since vacuolin-1 is known to induce fusion of endosomes and lysosomes, we hypothesized that vacuolin-1 would rescue this enucleation defect observed in survivin KO erythroblasts. Indeed, vacuolin-1 reversed the inhibitory effect of survivin KO on enucleation. Surprisingly, we also observed that vacuolin-1 promoted the proliferation of human erythroblasts at the CFU-E stage, leading to a 4-fold increase in the percentage of glycophorin-positive cells on day 12. Thus, we conclude that vacuolin-1 acts at multiple stages of differentiation to improve the production of erythrocytes in vitro. Finally, to gain insights into the mechanism of action of the “hit” vacuolin-1, and to identify more potent small molecules that enhance enucleation in vitro, we designed and studied a further small set of vacuolin-1 structural analogs (N=12). Four vacuolin-1 analogs increase enucleation of erythroblasts with greater potency than that of vacuolin-1, providing us with initial insights into the small molecule structure-activity relationships that might govern this process. We are currently assaying the activity of these next generation compounds in erythroblast cultures as a means to improve production of red blood cells for human use. Disclosures: No relevant conflicts of interest to declare.

Survivin is a 16.5 kD protein that is highly expressed in most cancer cells. Although it is also expressed in embryonic tissues and proliferating adult cells, it may be a promising target for cancer therapy if it has distinct functions in normal versus malignant cells. It is well established that survivin has an essential role in dividing cells as a member of the chromosome passenger complex (CPC). We recently showed that survivin is highly expressed in orthochromatic erythroblasts, cells that have exited the cell cycle and undergoing enucleation. Based on this observation, we predicted that survivin would play a novel role in maturation of these late erythroblasts. To discover the function of survivin in these post-mitotic cells, we first analyzed the localization of survivin and other members of the CPC by immunofluorescence. We discovered that survivin is expressed in cytoplasm, but did not colocalize with aurora B kinase or INCENP in primary enucleating erythroblasts. Next, we overexpressed survivin in the murine erythroleukemia (MEL) cell line, which undergoes terminal differentiation but does not enucleate, and quantified the effect on maturation by flow cytometry. In contrast to control MEL cells, which down-regulate survivin and fail to enucleate, nearly 15% of the survivin overexpressing cells showed nuclear extrusion. To obtain insight into this novel effect of survivin, we purified survivin protein complexes from MEL cell lysates and performed mass spectrometry. We found that EPS-15 (EGFR pathway substrate clone 15) and clathrin, two proteins that are involved in endocytic vesicle trafficking, interact with survivin. We confirmed these interactions by immunoprecipitation and immunofluorescence studies in primary human erythroblasts. Next, we individually knocked down survivin and EPS-15 in primary human erythroblast cultures and demonstrated that both proteins are required for optimal enucleation. Moreover, deletion of the survivin gene in fetal liver erythroblasts isolated from survivin floxed embryos, reduced the enucleation efficiency without affecting cell survival or differentiation. In order to determine how survivin and EPS15 contribute to enucleation, we performed light microscopic and ultrastructural examination of primary murine red cells undergoing terminal maturation. We observed that loss of survivin was associated with a decrease in the number of large cellular vacuoles within erythroblasts. Furthermore, by immunogold-EM, we found that survivin co-localized with EPS-15 within LAMP1-positive vacuoles in late erythroblasts. Finally, we observed that vacuolin-1, a small molecule that increases the number and size of vacuoles, rescued the enucleation defect of survivin-/- erythroblasts. Together, these results suggest that survivin plays an important and novel role in vesicle trafficking within maturing red blood cells. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2984. doi:10.1158/1538-7445.AM2011-2984

Abstract 89 Megakaryocytes undergo a unique mode of cell cycle called endomitosis, which is characterized by repeated rounds of DNA synthesis without cell division. Normally, polyploidization is tightly coupled with expression of megakaryocyte specific genes and terminal differentiation. Although there have been several reports of differences between the proliferative and endomitotic cell cycles, the precise mechanisms that regulate the switch to polyploidization and terminal differentiation remain unknown. To identify genes that play a role in this process, we performed a high throughput shRNA screen assaying for genes whose decreased expression induce polyploidization in CMK megakaryocytic leukemia cell line. We used a kinase biased library of shRNAs, reasoning that targeting the kinome would identify pathways or complexes important for this switch. After transduction of with the lentiviral shRNA library, CMK cells were selected with puromycin, fixed and stained with Hoechst dye. A high content imaging assay using CellProfiler software was employed to analyze the DNA content of microscopic images of cells. By this approach, we screened 775 genes, 650 encoding kinases, with at least 5 hairpins per gene. In parallel, we performed an otherwise identical screen (i.e., modifier screen) that included sub-maximal doses of dimethylfasudil, a drug that we have previously shown to promote megakaryocyte polyploidization. To reduce the off-target effects of shRNA, we only considered genes that scored with two or more shRNAs. In comparison to non-targeted shRNA controls, we considered the top 5% of genes in either the direct or modifier screens to be hits. By this analysis, we identified 71 genes whose loss promoted polyploidization. Pathway analysis revealed that the majority of these genes clustered into five groups. Among these clusters, one was notable for containing AURKA (aurora kinase A), AURKB (aurora kinase B), CCNB1 (cyclin B1), CDK2 and PLK1 (polo-like kinase). The screen revealed that knockdown of Aurora B kinase significantly increased polyploidization in the absence of diMF (p Disclosures: No relevant conflicts of interest to declare.

Abstract 2604 GABP transcription factor has been implicated in the regulation of genes that are required for normal megakaryocytic differentiation. Megakaryocytes express several related ets factors, including Fli-1, ets2, and GABP, and it has been unclear if any single ets factor plays a non-redundant role in these cells. The tetrameric GABP transcription factor complex contains two molecules of GABPα, which binds DNA, and two molecules of GABPβ, which encodes the transcription activation domain. We created mice with loxP recombination sites which flank exons that encode Gabpa ets-related DNA-binding domains (floxed Gabpa, or Gabpa fl/fl), and bred them to mice that carry Mx1-Cre. In response to injection with the synthetic polynucleotide, pIC, these mice express Cre recombinase and efficiently delete Gabpa; these animals are referred to as knock-out (KO) mice. Control mice, which carry floxed Gabpa but lack Mx1-Cre, were treated identically with pIC. Platelet counts of KO mice declined to less than 50,000 within nine days, while platelet counts in control mice were unaffected. One half of KO mice died within two weeks of Gabpa deletion due to widespread visceral hemorrhage. Histologic examination of the bone marrow and spleen reveals a loss of megakaryocytes in KO mice, compared to control animals. Residual megakaryocytes in KO mice exhibit increased expression of platelet-specific antigens, CD41 and CD42, and a significant increase of DNA ploidy. Because Gabpa KO mice died with a striking loss of megakaryocytes and platelets, yet megakaryocytic differentiation appeared to be unimpaired, we sought to better define the nature of this defect. Bone marrow from Gabpa fl/fl mice was infected with a retrovirus that expresses Cre and green fluorescent protein (GFP), or control virus that expresses only GFP; grown for three days in liquid culture conditions that foster megakaryocytic differentiation; and analyzed for CD41 and CD42 expression, ploidy, and apoptosis. Gabpa was efficiently deleted by the Cre-bearing virus, and Gabpa deletion was associated with increased expression of CD41 and CD42, and increased DNA ploidy. However, Gabpa deletion was also associated with increased megakaryocytic-associated apoptosis, and in vitro megakaryocyte colony formation was dramatically reduced in Gabpα null cells. In summary, deletion of Gabpa in mice is associated with plummeting platelet counts, widespread visceral hemorrhage, and a loss of splenic and bone marrow megakaryocytes. In vitro analysis demonstrates intact megakaryocytic differentiation and a profound loss of megakaryocytic progenitor cells. The increased expression of megakaryocytic antigens and DNA ploidy may indicate that Gabpa deletion enhances megakaryocytic differentiation or, alternatively, it may represent selective loss of more immature megakaryocytic cells following Gabpa disruption. Data that directly test these alternative hypotheses will be presented. In summary, we demonstrate that GABP plays a non-redundant role in megakaryocyte development, that GABP is required for the proliferation of committed megakaryocytic progenitors, but that GABP is not required for the later stages of megakaryocytic maturation. Disclosures: No relevant conflicts of interest to declare.

Abstract 3643 GATA-1 is a zinc finger transcription factor that regulates the differentiation of megakaryocytes and erythrocytes from the megakaryocyte-erythrocyte progenitor (MEP). Mutations in GATA1 are associated with hematologic malignancies of these two related lineages. Acquired mutations that lead to expression of the short isoform of GATA-1, termed GATA-1s, are associated with Acute Megakaryocytic Leukemia in children with Down syndrome (DS-AMKL). Moreover, inherited mutations in the N-finger of GATA1, such as V205M, cause a set of related diseases characterized by dyserythropoietic anemia and thrombocytopenia. These latter mutations disrupt recruitment of the essential cofactor FOG-1 and thus promote disease by interfering with the ability of the GATA-1:FOG complex to properly regulate gene expression. Despite the fact that V205 lies along the surface of the zinc finger opposite to DNA, previous studies suggest that FOG-1 may regulate the chromatin binding activity of GATA-1 at a subset of sites. In contrast to V205 mutation, the precise mechanisms by which GATA-1s contributes to disease is poorly understood. Previous studies have shown that GATA-1s uncouples megakaryocyte proliferation from differentiation, likely by an inability of GATA-1s to properly repress expression of a subset of GATA-1 target genes. We hypothesized that both inherited and acquired GATA1 mutations contribute to disease by interfering with not only target gene activation or repression, but also with GATA-1 chromatin binding. In order to define the chromatin binding activity of GATA-1 and its disease associated mutants, we performed chromatin immunoprecipitation coupled with next generation sequencing (ChIP-Seq) for wild-type GATA-1, GATA-1s and GATA-1V205G in the G1ME cell line. G1ME cells, which were derived from GATA-1 null ES cells, approximate an MEP in that they can differentiate into either erythroid cells (in the presence of EPO) or megakaryocytes (in the presence of TPO) upon reconstitution with GATA-1. We expressed GATA-1, GATA-1s, or GATA-1V205G in G1ME cells by retroviral transduction and subjected the cells to ChIP for GATA-1. The resulting DNA was sequenced on the Illumina GAII, yielding 11.8M, 10.4M, and 8.5M uniquely mapped reads. Analysis of the datasets using QuEST yielded 2367, 963, and 4130 peaks, respectively, with an FDR of Disclosures: No relevant conflicts of interest to declare.

Abstract 3263 Background: Despite the recent advances in chemotherapy for acute lymphoblastic leukemia (ALL), the development of drug resistance and long-term side effects of current treatments warrant new treatment modalities. Survivin/BIRC5, an inhibitor of apoptosis protein, is critical for the survival and proliferation of cancerous cells, is expressed in AML and ALL cells, and has been implicated in leukemia relapse. In the present study, we test the hypothesis that survivin is critical to the pathway of self-renewal of drug-resistant ALL cells. Methods: For gain of function studies, primary ALL cells were transduced with lentiviral survivin IRES GFP reporter (Survivin GFP) or empty GFP as a control. For loss of function studies, an inducible lentiviral shRNA vector expressing tRFP upon induction with doxycyclin was used. For in vitro evaluation of Survivin, CFU assays were used to monitor self-renewal capability, MTT assays and Trypan blue counts for viability determination were used for drug testing. For in vivo experiments, we used a NOD/SCID IL2Rγ−/- xenograft model with patient-derived ALL cells. Results: Survivin overexpression in primary ALL cells led in vitro to 4-fold more colonies than control in primary and secondary CFU assays and to increased resistance against Vincristine, Dexamethasone and L-Asparaginase (VDL) compared to controls (p Conclusion: Taken together, we show that Survivin is a key component in primary drug resistant ALL cells and adjuvant specific targeting of Survivin using shRNA or EZN3042 has the potential to eradicate relapse of leukemia. Disclosures: Yang: Amgen Inc: Employment, Equity Ownership.

Abstract 469 Children with Down syndrome (DS) have a 500-fold increased risk of Acute Megakaryoblastic Leukemia (DS-AMKL). In addition to trisomy 21, DS-AMKL blasts harbor somatic mutations in the gene encoding GATA-1, an essential transcriptional regulator of erythroid and megakaryocytic differentiation. In every case, the GATA1 mutations lead to a block in expression of the full-length protein, but allow for the translation of a shortened isoform named GATA-1s, which lacks the N-terminal transcriptional activation domain. Recent studies have shown that mutations in of JAK2, JAK3 and MPL are another feature of AMKL. Whereas the impact on hematopoiesis of individual GATA-1s, JAK2 V617F and MPL W515L mutants is relatively well known, the specific effects of trisomy 21, the initiating event in DS leukemogenesis, remains unclear. Here, we used two different experimental models to identify specific genes on human chromosome 21 (Hsa21) that participate in leukemogenesis: human DS-AMKL cell lines and the Ts1Rhr mouse strain, which is trisomic for only 33 orthologs of Hsa21 genes. Comprehensive analysis of hematopoiesis in progressively more complex mouse models of DS disorders, including Ts1Rhr mice, Ts1Rhr/KI-Gata-1s compound mutant mice, and recipients of Ts1Rhr/KI-Gata-1s hematopoietic progenitors expressing MPL W515L, revealed that the presence of trisomy 21 impacts the phenotype of each genetic background. Ts1Rhr mice develop a progressive myeloproliferative disease with megakaryocytic features while Ts1Rhr/KI-Gata1s double transgenic mice develop thrombocytosis at an earlier age compared to Gata1s mutant or Ts1Rhr mice alone. However neither the Ts1Rhr nor the compound mutant mice develop AMKL. In stark contrast, expression of MPL W515L in Ts1Rhr/KI-Gata1s hematopoietic progenitors led to a rapid lethal megakaryoblastic disorder characterized by profound marrow and splenic fibrosis and the presence of immature megakaryoblasts in lethally irradiated recipients. Of note, transplantation of MPL W515L expressing singly mutant Ts1Rhr or KI-Gata-1s progenitors failed to cause a similar disease. These findings show that three oncogenic events are necessary and sufficient to induce a megakaryoblastic leukemia in recipient mice. In parallel, to identify specific genes on Hsa21 that contribute to AMKL, we performed shRNA screening in DS-AMKL cell lines by knocking down the human orthologs of the 33 genes trisomic in the Ts1Rhr mice. We then evaluated the consequence of knockdown on megakaryocyte proliferation, survival and/or differentiation. This study revealed that altered expression of ERG, DYRK1A, CHAF1B and HLCS affected megakaryocyte growth. Analysis of their expression in human samples suggest that these four genes likely play a role in etiology of DS-AMKL. Moreover expression evaluation of those genes in our in vivo models of 1, 2 or 3 genetic events revealed that ERG, a known megakaryoblastic oncogene, and DYRK1A are the most prominent genes implicated in these megakaryocytic disorders. DYRK1A (dual specificity tyrosine-regulated kinase 1A) has been functionally implicated in many developmental disorders associated with DS and shown to be constitutively phosphorylated/activated in human AMKL cell lines. We found that overexpression of DYRK1A cooperates with Gata1s to lead to a increase in megakaryocyte growth in liquid culture and in colony assays in vitro. In summary, we show that partially trisomic Ts1Rhr mice display progressive defects in the megakaryocytic compartment, that mutagenesis of Gata1 in the Ts1Rhr background exacerbates the Ts1Rhr phenotype, and that addition of an activating mutation of MPL leads to a fulminant megakaryocytic leukemia with myelofibrosis. Moreover, among the 33 genes present in 3 copies in the Ts1Rhr background, we identify DYRK1A as a likely megakaryoblastic oncogene in DS-AMKL. Disclosures: No relevant conflicts of interest to declare.

Abstract 2605 GATA family transcription factors play critical roles in various mammalian developmental processes, including hematopoiesis. In particular, GATA-1 expression is necessary for proper terminal differentiation of mast cells, red blood cells, eosinophils, and megakaryocytes. GATA-2 is required for proliferation and survival of hematopoietic stem and progenitor cells, and is also expressed in erythroid precursors, mast cells, and early megakaryocytes. In developing erythrocytes, GATA-2 and GATA-1 are responsible for temporal control of a multi-factor transcriptional regulatory network that involves (a) GATA-2 positively regulating its own gene transcription, (b) GATA-2 positively regulating the expression of the Gata1 gene, (c) GATA-1 positively regulating its own gene transcription, and (d) GATA-1 negatively regulating Gata2 gene transcription. During this sequence of events, a “GATA switch” occurs, wherein GATA-1 replaces GATA-2 at canonical GATA binding sites within the regulatory regions of the Gata2 and Gata1 genes, as well as at many other genomic loci that encode genes responsible for proliferation or differentiation of erythroid progenitors. Similarly, in early megakaryocytic progenitors, GATA-2 promotes proliferation and suppresses expression of alternative-lineage genes; subsequent activation of GATA-1 precipitates terminal differentiation with concomitant downregulation of proliferative genes and activation of megakaryocyte-specific genes. The presence or role of a GATA switch in megakaryocytes has not yet been formally investigated. To address the role of the GATA switch in megakaryocytic differentiation, we performed massively parallel sequencing of chromatin immunoprecipitation (ChIP-Seq) material for GATA-2 and GATA-1 before or after GATA-1 restoration in the GATA1-null megakaryocytic progenitor cell line, G1ME. We obtained 22 million unique GATA-2 tags and 10 million unique GATA-1 tags and identified 14985 and 5102 high-confidence GATA-2 and GATA-1 binding sites, respectively. Additionally, we used 13 million tags from ChIP for H3K4me3 to identify 24909 genomic sites enriched for the presence of trimethylated lysine-4 on histone H3. Trimethylated H3K4 marks nearly half of all GATA-1 bound sites and one-third of GATA-2 bound sites. Over 40% of the sites bound by GATA-1 in differentiating G1ME cells were also bound by GATA-2 in proliferating G1ME cells, indicating that a GATA switch does indeed occur during megakaryocyte development. Coordinated analyses of these occupancy data with previously published gene expression datasets show that the lists of bound genes are significantly enriched for differentially expressed genes and the data depict a generally antagonistic relationship between GATA-2 and GATA-1. Interestingly, we find that even among genes that don't contain GATA switch sites, greater than 40% of those bound by GATA-1 were also occupied by GATA-2 at distinct sites. To further characterize the occupied loci, we surveyed the genomic regions bound by GATA-1 and GATA-2 to detect motifs enriched in the sequences surrounding the peak calls. As expected, we found that over 80% contained the canonical WGATAR binding motif. In contrast to reports of motifs enriched in GATA-1 ChIP studies in erythroid cells, we failed to observe significant enrichment of LRF binding motifs. Rather, the GATA-1 and GATA-2 bound regions in megakaryocytes are strongly enriched for motifs that match the binding sites for Ets family transcription factors. Finally, we have found that these genomic regions are indeed occupied by one or more Ets factors in proliferating G1ME cells. Together, these data establish the presence of a GATA switch in megakaryocyte development and provide novel insights into coordinated gene regulation by GATA factors and the differences between the closely related erythroid and megakaryocyte lineages. Disclosures: No relevant conflicts of interest to declare.

Abstract 87 Survivin is a member of the inhibitor of apoptosis protein (IAP) family that antagonizes caspases and has been implicated in regulation of apoptosis, cell division and cell cycle both in cancer cells and normal tissues. Survivin is essential for activation of Aurora kinase that phosphorylates Histone H3, an event required for transcriptional regulation and cytokinesis. Although there is no evidence that Survivin directly regulates gene transcription, Bir1, a C. elegans homologue of Survivin, regulates transcription, most likely through Histone phosphorylation by Aurora kinase. We previously showed that Survivin is expressed and growth factor regulated in human CD34+ cells (Fukuda & Pelus. Blood 2001 & 2002). Antagonizing Survivin impairs production of mouse bone marrow hematopoietic progenitor cells in vitro (Fukuda et al. Blood 2004) and conditional Survivin gene deletion in vivo in mice leads to bone marrow ablation as a result from loss of hematopoietic stem and progenitor cells (Leung et al. JEM 2007). However, little is known about signaling pathways downstream of Survivin in hematopoietic stem cells (HSC). Herein, we investigated the mechanism of action of Survivin by creating Survivin gene signature network following gene deletion in HSC and by functional rescue on impaired HSC activity induced by Survivin deletion. Survivin deletion in Tamoxifen-Cre Survivinflox/flox mice led to significant changes in 2,228 mRNA in marrow CD34neg, c-kit+, Sca-1+, lineage− cells. Molecular network of the genes affected by Survivin deletion in HSC were created using DAVID (Huang et al. Nature Protocol 2009) and Cytoscape (Cline et al. Nature Protocol 2007). Among the signaling pathways defined by KEGG (www.kegg.org), phosphatidylinositol signaling pathway was most significantly affected in HSC by Survivin gene deletion (P These results suggest that Evi-1 lies down stream of Survivin in HSC and that repression of Evi-1 expression is a functional consequence of Survivin ablation rather than an indirect effect resulting from apoptosis and/or cell cycle alteration of HSC induced by Survivin deletion. Our data indicate that Survivin regulates HSC proliferation at least through Evi-1. Alteration of genes with multiple cellular functions as a consequence of Survivin deletion and the presence of a functional network downstream of Survivin suggest that Survivin regulates multiple effector pathways in HSC, independent of its activity as a caspase inhibitor. Disclosures: No relevant conflicts of interest to declare.

Abstract 1602 The hematopoietic transcription factor Ikaros regulates the development of lymphoid cells. In particular, Ikaros has been shown to suppress expression of Notch target genes in developing thymocytes. In addition, Ikaros-deficient animals fail to develop B cells and instead display a T cell lymphoproliferative disorder. Furthermore, alterations in Ikaros gene expression are associated with human lymphoid neoplasms and loss of Ikaros is associated with progression of myeloproliferative neoplasms to acute myeloid leukemia. However, the role of Ikaros in normal and malignant myelopoiesis is not well characterized. Therefore, we first analyzed myeloid development in Ikaros deficient mice and observed extramedullary hematopoiesis in the spleen and a striking thrombocytosis in the peripheral blood of 8–9 week-old Ikaros-null mice (2×106 platelets/μl, compared to 0.5×106 platelets/μl in wild-type littermates). Flow cytometry, histology and colony forming assays revealed that the bone marrow and spleen of young Ikaros-null mice harbored a substantial increase in the numbers of megakaryocytes and myeloid cells. We next investigated how Ikaros activity is reduced during terminal differentiation. Previous reports have shown that GATA1-deficient and GATA1s mutant megakaryocytes, which are associated with Down syndrome acute megakaryoblastic leukemia (DS-AMKL), express aberrantly high levels of Ikaros, suggesting that Ikaros is a target of GATA-1 repression during terminal megakaryocyte differentiation. By chromatin immunoprecipitation assays, we found multiple sites in the Ikaros locus that are bound by GATA-2 in proliferating progenitors and by GATA-1 in maturing megakaryocytes. Furthermore, we discovered that GATA-1s fails to occupy these conserved Ikaros loci. Together these results strongly suggest that GATA-1, but not GATA-1s, displaces GATA-2 from the Ikaros gene during differentiation and that this GATA switch leads to repression of Ikaros, in turn allowing for terminal maturation of megakaryocytes. Since Notch participates in specification of the megakaryocyte lineage, we then asked whether Ikaros inhibited megakaryopoiesis by interfering with Notch signaling. Retroviral transduction of full-length Ikaros in murine common myeloid progenitors, but not in LSK cells, reduced megakaryocyte development upon Notch stimulation in OP9-DL1 stromal co-cultures. Ectopic expression of Ikaros in the 6133 cell line, which models the t(1;22) subtype of AMKL and shows aberrant Notch target genes activation by the OTT-MAL fusion oncogene, inhibited proliferation and induced apoptosis. Genome wide expression profiling in transduced 6133 cells confirmed that Ikaros alters expression of several genes involved in the Notch and growth factor signaling pathways. These results indicate that Ikaros restricts megakaryocyte development and inhibits proliferation of OTT-MAL-transformed AMKL cells at least in part by suppressing Notch signaling. Together, our results suggest that a functional antagonism between the Notch pathway and Ikaros controls normal megakaryocyte development and that this axis is deregulated in AMKL, contributing to aberrant expansion of immature megakaryoblasts. Disclosures: No relevant conflicts of interest to declare.

Abstract 792 Acute megakaryoblastic leukemia (AMKL) is a rare subtype of acute myeloid leukemia characterized by expansion of immature megakaryocytes and bone marrow myelofibrosis. AMKL is frequently associated with chromosomal abnormalities, such as trisomy 21 or t(1;22), which leads to expression of the OTT-MAL fusion protein. Mutations in GATA1 are present in nearly all cases of Down syndrome AMKL, while mutations in JAK3, MPL, KIT, and FLT3 are associated with a smaller subset of patients. Since most patients with AMKL face a very poor prognosis, new therapies are desperately needed. Given that megakaryocytes undergo polyploidization during the normal course of differentiation, we hypothesized that small molecule inducers of polyploidization would drive megakaryoblasts to exit the proliferative cell cycle and induce terminal differentiation. These agents would thus serve as targeted differentiation therapeutics for AMKL in a manner analogous to ATRA for treatment of acute promyelocytic leukemia. To identify small molecules that induce megakaryocyte polyploidization, we incubated the CMK megakaryoblastic cell line, which harbors both GATA1 and JAK3 mutations, with libraries of kinase or histone deacetylase inhibitors, natural products, and small molecules derived from diverse oriented synthesis. After three days, we stained the cells with Hoechst dye, imaged plates with the ImageXpress Micro microscope and converted image files into quantitative values of DNA content with Cell Profiler software. By screening 10,000 compounds, we identified 207 molecules that significantly and reproducibly increased polyploidization. Among the compounds scoring in this screen were microtubule disrupting and stabilizing agents and actin disrupting agents, which are expected to cause alterations in spindle formation or cytokinesis and result in polyploidization. A more interesting subset of compounds led to robust polyploidization, with DNA contents reaching 32N and 64N, and also to induction of megakaryocyte differentiation. We have focused our efforts on the most potent compound of the latter class, dimethylfasudil (diMF), a known Rho kinase inhibitor. diMF blocked proliferation and simultaneously induced marked polyploidization, differentiation and apoptosis of human and murine megakaryoblastic leukemia cell lines as well as primary murine and human megakaryocyte progenitors. diMF also inhibited megakaryocyte colony formation and induced polyploidization and differentiation of GATA-1 deficient and GATA-1s knock-in fetal liver megakaryocytes, which mimic the aberrant megakaryopoiesis seen in infants with DS. Importantly, diMF did not induce polyploidization of the CD41-negative fraction of primary murine or human ex vivo cultures, revealing its specificity for megakaryocytes. When administered to mice by oral gavage, diMF selectively affected megakaryocytes within the bone marrow and did not alter peripheral blood counts. diMF also effectively inhibited AMKL development in mice transplanted with MPL mutant, OTT-MAL expressing cells: 50% and 30% of the recipients treated with 66 mg/kg and 33 mg/kg diMF bid respectively, remain disease free after three months whereas all placebo treated mice succumbed to AMKL within 3 weeks. To identify the targets of diMF, we used a combination of the SILAC (stable isotope labeling with amino acids in cell culture) proteomic approach and Ambit's KinomeScan. These unbiased assays identified multiple binding targets of diMF, including JAK3, MAP4K2, and Rho kinase 1 (ROCK1). Using shRNA mediated knock-down, we found that decreased expression of ROCK1, but not ROCK2, led to increased sensitivity of CMK cells to diMF induced polyploidization. Knockdown of ROCK1 alone also induced features of megakaryocyte differentiation such as enhanced CD41 expression. These studies implicate ROCK1 in regulation of the switch between proliferation and differentiation of megakaryocytes and show that diMF acts through inhibition of ROCK1. Taken together, our work demonstrates that diMF and other polyploidy inducing agents discovered in our screen represent potential targeted agents for AMKL. Furthermore, since diMF acted similarly on megakaryoblasts with the various genetic lesions associated with each of the different classes of AMKL, we predict that this approach will be an effective therapeutic strategy for all categories of megakaryocytic neoplasms. Disclosures: No relevant conflicts of interest to declare.

Abstract 475 ERG is an oncogene located on the long arm of human chromosome 21 that encodes an ETS transcription factor that has been reported to be involved in normal and aberrant megakaryopoiesis (Salek-Ardakani et al Cancer Res. 2009;69:4665; Stankiewicz et al Blood 2009;113:3337). High ERG expression has also been reported associated with poor prognosis of cytogenetically normal AML (Marcucci et al J Clin Oncol. 2007;25:3337). Thus, deciphering the molecular targets and oncogenic pathways activated by overexpressed ERG protein is likely to lead to more effective therapy for ERG-related myeloid leukemias. Towards these goals we have combined in-vitro and in-vivo approaches. We have created transgenic mice expressing the ERG3 hematopoietic isoform under the VAV promoter. All these mice die from invasive acute megakaryocytic or undifferentiated myeloid leukemia by the age of 5 months. The leukemias are transplantable, and primary growth factor dependent leukemic cell lines, suitable for pharmacological and molecular studies, have been established. To identify ERG target genes and proteins, we have analyzed gene expression in two “mirror-image” cellular systems – shRNA mediated knockdown of ERG in Meg01 megakaryocytic leukemia cells and overexpression of ERG in K562 erythroleukemia cells. Gene Set Enrichment Analysis (GSEA) demonstrated that the “ERG gene-associated expression signature” in these cell lines is enriched with genes that were also identified in primary human AML characterized by ERG overexpression. By chromatin immunoprecipitation, we then identified specific, direct ERG targets and confirmed their expression in samples from ERG transgenic leukemias and primary human “ERG-overexpressing” AMLs. Surprisingly, we observed that the lymphoid kinase Bruton agammaglobulinemia tyrosine kinase (BTK) is an ERG direct target that is upregulated in both ERG overexpressing mouse and human leukemias. Preliminary experiments have demonstrated activity of a BTK inhibitor on ERG transgenic and ERG overexpressing human AML cells. ERG overexpression also induced marked activation of RAS signaling as supported by elevated levels of RAS-GTP and its downstream targets and significant growth inhibitory activity caused by a novel RAS inhibitor, farnesylthiosalicylic acid (Salirasib), that disrupts the spatiotemporal localization of active Ras (Rotblat et al Methods Enzymol. 2008;439:467) was observed. Dramatic growth inhibitory activity has also been induced by two novel compounds that induce megakaryocytic polyploidization, suggesting a potential role for “differentiation” therapy of ERG-related leukemias. Thus we have created a mouse model for ERG-related AML and characterized several genes and biochemical pathways that are susceptible to pharmacological inhibition. Our studies are not only likely to decipher the mechanisms by which ERG overexpression contributes to leukemogenesis, but also provide a platform for preclinical studies of novel therapeutics of these poor prognosis leukemias. Disclosures: No relevant conflicts of interest to declare.

Abstract 3651 Poster Board III-587 The Ets transcription factor PU.1 is a master regulator absolutely required for the differentiation of monocytes, macrophages, and B cells in the fetal liver and in the adult bone marrow. PU.1 drives hematopoietic differentiation partly through direct protein-protein interactions with other transcription factors, such as the AP-1 transcription factor c-Jun. We have shown that c-Jun can be recruited to promoters which do not include AP-1 binding sites, such as the MCSFR promoter, and act as a PU.1-dependent co-activator. To address the functional importance of this interaction, we identified and studied PU.1 point mutants that lost the capability to physically interact with c-Jun while retaining normal DNA binding affinity. These mutants failed to efficiently transactivate a PU.1 target reporter, and, more importantly, were unable to induce monocyte/macrophage differentiation of the PU.1-deficient immature myeloid 503 cell line. Subsequently, we have generated knock-in mouse models harboring these single point mutations by means of homologous recombination. The mutant mice phenotypically resemble PU.1-deficient mice, have an early block in hematopoiesis, and die perinatally. We show that the mutant PU.1 mRNA and protein is expressed in long-term and short-term hematopoietic stem cells, but that the maturation into lymphoid primed multipotent progenitor (LMPP) and later progenitor populations is severely blocked, leading to an almost complete loss of mature B, T and myeloid cells. Collectively, our data strongly suggest that the PU.1-c-Jun interaction is crucial for normal PU.1 function in vivo during murine fetal hematopoiesis. Disclosures: No relevant conflicts of interest to declare.

Abstract 1896 Poster Board I-919 Somatic mutations in JAK2 and MPL are associated with BCR-ABL negative myeloproliferative neoplasms (MPNs) including polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF). Mutations in these genes have also been identified in rare cases of acute megakaryoblastic leukemia (AMKL). To investigate the role of activated JAK2 and MPL signaling in aberrant megakaryopoiesis associated with MPN and AMKL, we overexpressed JAK2 V617F or MPL W515L in primary murine hematopoietic progenitor cells. We found that expression of mutant MPL, but not JAK2, led to a dramatic increase in the size and number of megakaryocyte colonies. Overexpression of constitutively activated alleles of STAT3 or STAT5 did not, however, enhance megakaryocyte colony formation. Moreover, knock-down of STAT3 or STAT5 failed to inhibit the growth of JAK2 V617F or MPL W515L transduced clones of the G1ME megakaryocytic cell line. Together, these observations argue against the simple model of JAK2 and MPL mutants conferring cytokine independent growth and excessive proliferation solely by inducing STAT signaling. Since TPO is known to activate additional signaling pathways, including MAPK/ERK and PI3K/AKT, we hypothesized that alterations in pathways other than JAK/STAT contribute to MPNs. To determine which of these signaling pathways are involved, we cultured JAK2 V617F and MPL W515L expressing clones of the G1ME and 6133 megakaryocytic cell lines with a panel of small molecule kinase inhibitors. We discovered that inhibition of the PI3K/AKT and MAPK/JNK signaling pathways with triciribine and SP600125 respectively, dramatically suppressed growth of these cells by inducing G1 arrest and apoptosis. Of note, the inhibitory effect of these two small molecules was more pronounced than that caused by JAK2 inhibitor II. Inhibitors of Ras, ERK, Src family tyrosine kinase, and PKC all failed to significantly inhibit proliferation of the JAK2 or MPL mutant expressing cells. Since triciribine showed the strongest activity against the mutant cell lines, we next assayed its activity on a mouse model of MPL-induced AMKL, on primary murine cells expressing MPL W515L and on human MPN specimens. Twice daily injection with 0.1mg/kg of triciribine significantly delayed AMKL development in mice transplanted with 6133/MPL mutant cells. Triciribine also repressed CFU-Mk formation from MPL W515L transduced mouse bone marrow as well as BFU-E formation from human PV patient samples. These findings demonstrate that AKT is a potential target for therapy of human myeloproliferative neoplasms. Disclosures: Levine: Novartis: Research Funding; TargeGen: Consultancy.

Specific genes that contribute to Transient Myeloproliferative Disorder (TMD) and DS Acute Megakaryocytic Leukemia (AMKL) remain largely undefined. To date, only the X-linked transcription factor GATA1, which is mutated in these disorders, has been implicated in contributing to DS hematopoietic malignancies. We recently reported that overexpression of either of the chromosome 21 ETS family members ERG or ETS2 led to a dramatic increase in megakaryopoeisis from wild-type or Gata1-knockdown fetal liver hematopoietic progenitors. Furthermore, we demonstrated that expression of ETS2 or ERG caused a significant increase in megakaryocyte colony formation and led to immortalization of Gata1-knockdown fetal liver hematopoietic progenitors in vitro. In new studies to define the mechanisms by which these ETS proteins promote megakaryopoiesis, we discovered that the ETS family member FLI1, which shares significant homology with ERG, significantly increased CD41 and CD42 expression and the degree of polyploidization of megakaryocytes in vitro in a manner that was nearly identical to that of ectopic ERG. Furthermore, FLI1 also immortalized Gata1-knockdown fetal liver progenitors. To characterize the effects of ectopic ETS family members on transcription, we performed genome wide expression profiling using RNA isolated from fourth generation immortalized Gata1-knockdown cells expressing ETS2, ERG, FLI1 or GFP alone. Principal component analysis showed that the ERG and FLI1 gene expression patterns clustered tightly with one another and were well separated from both the MIGR1 and ETS2 signatures. Unsupervised hierarchical clustering further revealed that 726 genes were differentially expressed in FLI1, ERG, ETS2 immortalized cells compared to MIGR1 control cells. Of these, genes associated with the erythroid lineage, including Klf1, Lmo2, glycophorin A, Egr1, Ank1 and Eraf were strongly repressed. These findings are consistent with recent publications showing that FLI1 promotes the development of megakaryocytes at the expense of red blood cells and suggest that ERG and ETS2 mimic FLI1 in our assays. Notably, we did not observe significant increases in the expression of Hox genes or Meis1, genes whose expression is often associated with transformed hematopoietic cells. However, we did detect increased expression of Bmi1, which has an essential role in regulating the proliferative capacity of both normal and malignant hematopoietic progenitors, as well as increased expression of Jak2 and Stat5, two genes whose activation is associated with myeloproliferative disease and megakaryocytic transformation, in ETS-family transduced cells. Next, to determine whether ETS-family expression affected the immunophenotype of Gata1-knockdown fetal liver progenitors, immortalized cells were assessed for cell surface expression of c-kit and CD41. ERG and FLI1 expressing colonies included a substantial proportion of CD41+/c-kit- cells, indicative of megakaryocytic maturation. Furthermore, colonies formed from progenitors transduced with ERG were comprised of greater than 90% CD41+ cells, confirming that nearly all of the cells within the immortalized colonies are of the megakaryocyte lineage. Finally, because our gene array data showed increased Jak2 and Stat5 mRNA levels in ETS-family transduced cells, and activating mutations in JAK3 are associated with AMKL, we assessed whether JAK/STAT signaling was altered in the immortalized cells. Using flow cytometry, we found that cells within the ERG and FLI1 cultures consistently showed significant increases in phospho-STAT3 staining in comparison to the MIGR1 control. The increases in STAT3 phosphorylation suggest that ERG and FLI1 promote immortalization of Gata1-knockdown fetal liver progenitors in part by activation of the JAK/STAT pathway. Together our data support the hypothesis that an increased dosage of ETS factors, including ETS2 and ERG, contribute to a pro-megakaryocytic phenotype and to the leukemogenic process in DS TMD and AMKL.

The mechanisms that drive the final step in the development of reticulocytes from nucleated erythroblasts remain largely undefined. We recently reported that the chromosome passenger complex (CPC) protein survivin is required for the maintenance of hematopoietic stem cells and for the proliferation and survival of erythroid progenitors. Several lines of evidence suggest that survivin also plays an additional role late in erythropoiesis. First, 20% of survivin heterozygous mice showed a decrease in the proportion of enucleated cells relative to nucleated erythroblasts. Second, survivin is highly expressed in orthochromatic erythroblasts undergoing enucleation, where it partially co-localizes with actin at the junction between the extruding nucleus and cytoplasm. Third, overexpression of survivin promoted enucleation during DMSO-induced murine erythroleukemia (MEL) cell differentiation. Since survivin plays an essential role during normal mitosis and because enucleation has been postulated to occur by asymmetric cytokinesis, we investigated the link between cytokinesis and enucleation. Using primary erythroblast cultures, we found that chemical inhibitors of cytokinesis failed to prevent enucleation. Furthermore, we could not detect co-localization of survivin with aurora-B or INCENP, two other members of the CPC, in enucleating primary human erythroblasts. To determine the mechanism by which survivin contributes to enucleation, we utilized an in vivo biotinylation strategy to isolate survivin protein complexes from extracts of MEL cells. Using mass spectrometry, we discovered that survivin complexes include clathrin and its partner EPS15 (EGFR pathway substrate clone 15). We confirmed the endogenous interaction between these three proteins by co-immunoprecipitation. Treatment of primary erythroblast cultures with an inhibitor of endocytosis led to a decrease in enucleation efficiency. In addition, siRNAs targeting survivin, EPS15, or clathrin each reduced the degree of enucleation of primary human erythroblasts, confirming a requirement for this protein complex in enucleation. Together, our data demonstrate that survivin promotes enucleation by a novel clathrin-mediated process. Furthermore, our data help explain electron microscopy (EM) observations that enucleation is accompanied by an accumulation of vesicles that coalesce to form U-shaped channels at the junction between the nucleus and plasmalemma of the erythroblast (Simpson and Kling, JCB35:237, 1967). These EM studies along with our new data suggest that vesicular trafficking may be one of the major driving forces in enucleating erythroblasts.

GATA2 is an essential transcription factor that regulates multiple aspects of hematopoiesis. Mutations in GATA2 are associated with blast crisis phase of chronic myelogenous leukemia (CML) while overexpression of GATA2 is a hallmark of Down syndrome acute megakaryoblastic leukemia (DS-AMKL), a malignancy that is defined by the combination of trisomic chromosome 21 and a GATA1 mutation. Here, we show thatGATA2 is essential for megakaryocyte development from both wild-type and GATA-1mutant hematopoietic progenitors. Furthermore, we reveal that GATA2 is indispensable for cell cycle progression and that its expression cooperates with activated JAK3 to promote TPO-independent expansion of megakaryocytes. Genome-wide microarray analysis revealed that GATA2 regulates a wide set of genes including cell cycle regulators and lineage-specific genes that are essential for terminal differentiation of megakaryocytes. Of note, a subset of genes within the GATA2 signature is contained within human DS-AMKL patient specimens. These observations implicate overexpression of GATA2 in the aberrant gene expression and proliferation of AMKL. Taken together, our data demonstrate thatGATA2 is a critical regulator of normal and malignant megakaryopoiesis.

Survivin is a member of chromosome passenger complex, which plays an important role in chromosome alignment, separation and cytokinesis. We recently reported that survivin is necessary for the proliferation and survival of hematopoietic stem and progenitor cells. Furthermore, we previously showed that reduced levels of survivin expression facilitates megakaryocyte development, whereas elevated levels of survivin inhibit their maturation and polyploidization. However, the extent to which survivin is necessary for polyploidization and terminal differentiation of committed megakaryocytes remains unclear. To determine whether survivin is required for megakaryocyte and platelet biogenesis, we mated mice with floxed alleles of survivin (sur fl/fl) to mice that express Cre recombinase under the control of the PF4 promoter. Compound mutant animals appeared grossly normal and harbored normal platelet counts. Furthermore, survivin deleted and control littermates displayed similar expression of CD41 and CD42, as well as similar DNA content within the CD41+ population. The only significant difference detected was an increase in annexin V staining of CD41+ cells within the bone marrow of the mice with survivin deletion. Analysis of DNA extracted from these bone marrows showed no evidence of the survivin deletion, indicating that the surviving cells all escaped excision. These in vivo findings are consistent with a requirement for survivin in the survival or proliferation of megakaryocyte progenitors. Next, to induce megakaryocyte development ex vivo, we cultured bone marrow from surv fl/fl mice in vitro in the presence of TPO. Using this approach, we were able to induce survivin deletion in 75% of the cells as evidenced by PCR. Despite the deletion of survivin, polyploidization of the ex vivo generated megakaryocytes was unaffected. Finally, we induced deletion of survivin by retroviral infection of surv fl/fl progenitors with MSCV-Cre and found that megakaryocyte polyploidization was actually increased in the excised population. Taken together, our results suggest that survivin is not required for polyploidization, but is necessary for proliferation of megakaryocyte progenitors.

Hematopoietic cell fate is determined by combinatorial interactions between nuclear proteins that activate and repress transcription. This principle is illustrated by the transcription factors GATA-1 and PU.1, which promote erythro-megakaryocytic and granulocyte-macrophage development respectively. These two proteins interact physically to cross-antagonize each other’s activities, creating regulatory loops for hematopoietic differentiation. Previously, we showed that loss of GATA-1 causes expansion of bipotental megakaryocyte erythroid progenitors (MEPs) from embryonic stem cells or fetal liver derived hematopoietic progenitors. These cells, termed G1ME, for GATA-1-null megakaryocyte-erythroid, proliferate continuously in culture and differentiate into committed megakaryocytes and erythroblasts when GATA-1 activity is restored. G1ME cells express GATA-2, a GATA-1-related protein normally found in multipotential hematopoietic progenitors and stem cells. These cells also express moderate levels of PU.1 mRNA, approximately 1/3 of that expressed in the myeloid cell line 416B. Upon retroviral restoration of GATA-1, GATA-2 is downregulated and PU.1 mRNA decreases rapidly. Microarray analysis of GATA-1-rescued G1ME cells revealed repression of PU.1 and many of its downstream target genes, raising the possibility of direct PU.1/Sfpi1 gene repression by GATA-1. Chromatin immunoprecipitation (ChIP) studies identified two GATA factor-binding sites at the PU.1/Sfpi1 locus. In the absence of GATA-1, when the PU.1/Sfpi1 gene is active, these sites are occupied by GATA-2. Retrovirally expressed GATA-1 replaces GATA-2 at these sites, repressing PU.1 transcription during concomitant erythro-megakaryocytic maturation. These findings resemble the “GATA-factor switch” described at other loci such as Gata2 and Kit where GATA-2 and GATA-1 compete for the same cis elements to activate and repress transcription respectively. To test this, we used siRNA to repress GATA-2 expression in G1ME cells by about 60%. Strikingly, this caused PU.1 to be upregulated 4-fold, indicating that GATA-2 also represses PU.1/Sfpi1, but to a lesser extent than GATA-1. Moreover, G1ME cells expressing GATA-2 siRNA differentiated into macrophages, as evidenced by morphology, expression of numerous cell-type specific markers and massive induction of macrophage specific genes including myeloperoxidase, Mac-1, and C/EBPα. Our findings illustrate two new insights into the transcriptional control of hematopoietic cell differentiation: First, cross-antagonism between GATA-1 and PU.1 not only occurs at the level of protein-protein interaction, but also through direct transcriptional repression. Second, in addition to having opposite effects on transcription of the same target gene as described previously, GATA-2 and GATA-1 can act cooperatively and successively to exert repressive effects of different magnitudes that gradually restrict gene expression during hematopoietic development. In this model, hematopoietic progenitors express GATA-2 and low levels of PU.1 that maintain the multipotential state but are not sufficient for myelopoiesis. Repression of GATA-2 in the absence of GATA-1 raises PU.1 levels to stimulate granulocyte-macrophage development. In contrast, activation of GATA-1 causes PU.1 to be fully repressed, promoting erythrocyte and megakaryocyte differentiation. Our data illustrate how lineage fate and hematopoietic differentiation are influenced by the stoichiometry between GATA-1, GATA-2, and PU.1 in multipotential progenitors.

ETS2 and ERG are two ETS family transcription factors encoded on chromosome 21 that have been implicated in leukemia and other types of cancer. In the setting of trisomy 21 (e.g. in individuals with Down syndrome) there is a significant increase in the propensity for leukemia, including AMKL (acute megakaryoblastic leukemia). It has been theorized that genes located on chromosome 21 may contribute to the predisposition of children with Down syndrome for pediatric AMKL, possibly by collaborating with mutations in the transcriptional regulator GATA1. To test the oncogenic potential of these genes in the hematopoietic system, ETS2 and the ERG isoform, ERG1, were over-expressed in both wild-type and GATA1 knockdown (ΔneoΔHS) fetal liver hematopoietic progenitors and the effects on growth and differentiation were studied. We discovered that over-expression of either ETS2 or ERG1 in wild-type progenitors increased the formation of CD41 positive megakaryocytes by 2–3 fold in liquid culture. However, whereas ETS2 appeared to primarily promote the early stages of megakaryopoiesis, ERG1 over-expression led to a dramatic rise in the percentage of cells expressing CD42, a marker of more mature megakaryocytes. Ectopic ERG1 expression also affected the ploidy profile, with nearly all cells containing ≥4N DNA content. Quantitative real time RT-PCR analysis of lineage specific genes confirmed that expression of megakaryocyte genes, including GPIIb (CD41), platelet factor 4 (PF4), and ß1-tubulin, was significantly increased in undifferentiated cells over-expressing either ETS2 or ERG1. These results show that over-expression of ETS2 or ERG1 rapidly initiates the megakaryocyte program. Over-expression of either ETS2 or ERG1 in GATA1 knockdown fetal liver cells similarly led to a 2.5–5 fold increase in the formation of CD41 positive megakaryocytes in liquid culture; however, neither ETS2 or ERG1 could circumvent the differentiation block that is characteristic of the GATA1 knockdown strain, including the inability to form proplatelet extensions. We next assessed the capacity of ETS2 or ERG1 over-expression to alter the colony forming ability of wild-type and GATA1 knockdown fetal liver progenitors. Both ETS2 and ERG1 expression led to a significant increase in CFU-MK derived from wild-type progenitors, as evidenced by an increase in the number of acetylcholinesterase positive colonies. In contrast, neither gene altered the formation of CFU-MK when over-expressed in the GATA1 knockdown progenitors, which already show a prominent enhancement in megakaryocyte colony formation over wild-type fetal liver cells. Finally, serial re-plating assays performed with GATA1 knockdown fetal liver cells showed that ectopic expression of either ETS2 or ERG1 could immortalize hematopoietic progenitors, whereas only ERG1 could enhance colony re-plating of wild-type cells. These findings provide evidence for a specific synergy between GATA-1 deficiency and ETS2 over-expression and suggest a distinct oncogenic potential for each factor. Overall, these results in primary murine fetal liver cells demonstrate that ETS2 and ERG1 promote the megakaryocyte lineage and facilitate immortalization of GATA1 knockdown cells, consistent with a potential oncogenic role in Down syndrome AMKL for these two ETS family genes located on chromosome 21.

Enucleation is the final step in the development of reticulocytes and this process is unique to mammalian red blood cells among vertebrates. Even though enucleation of erythroblasts was structurally studied by electron microscopy half a century ago, little is known about the underlying mechanism. Initially it was thought that enucleation involved degradation of DNA and other apoptotic processes. More recently it has been shown that enucleation of erythroid progenitors does not require major apoptotic machinery including caspase-3 activation (Krauss et al, 2005; Yoshida et al, 2005). Alternatively, it has been suggested that enucleation in erythroblasts is a result of asymmetric cytokinesis. In support of this later model, survivin, a regulator of chromosomal segregation and cytokinesis, which also has an anti apoptotic role, is highly expressed in orthochromatic erythroblasts, cells at the stage of differentiation just prior to enucleation (Gurbuxani et al, 2005). Moreover conditional heterozygous deletion of survivin in hematopoietic cells caused a decrease in enucleated cells in the spleen (Leung et al, 2007). This has led us to hypothesize that survivin has a role in enucleation of erythroblasts. To test this prediction, we first over expressed survivin in MEL (murine erythroleukemia) cells and then subjected them to DMSO induced differentiation. Normally MEL cells decrease their survivin expression level upon differentiation and at the same time undergo limited enucleation (

Survivin, the smallest member of the Inhibitor of Apoptosis (IAP) family of proteins, is expressed in proliferating cells, including fetal tissues, T-lymphocytes and hematopoietic progenitors. Survivin is also highly overexpressed in a wide spectrum of human cancers. Numerous studies have implicated survivin as a key mediator of the spindle assembly checkpoint and cytokinesis, while others have suggested an additional role as an inhibitor of apoptosis. In order to precisely examine the role of survivin in hematopoiesis in vivo, we induced acute deletion of the survivin gene in mice harboring either heterozygous or homozygous floxed alleles of survivin and the MX1-Cre transgene with polyinosinic-polycytidylic acid (pI-pC). Complete loss of survivin in homozygous floxed, MX1-Cre animals resulted in rapid death, which was associated with near total ablation of the bone marrow and marked hypocellularity of the spleen. The peripheral blood of these mutants displayed pancytopenia, including reticulocytopenia, and the presence of abnormal erythrocytes. Hematopoietic progenitors were markedly decreased, including c-kit+ cells, as well as CD71hiTer119hi erythroid precursors. Analysis of the cell cycle parameters revealed that a significant proportion of erythroid progenitors accumulated in G1, while others became polyploid. These defects could not be rescued by overexpression of the anti-apoptotic protein Bcl-2, consistent with a primary function of survivin in cell cycle progression. Bone marrow transplantation studies further confirmed that there exists a cell autonomous requirement for survivin in hematopoietic progenitors. Surprisingly, acute heterozygous deletion of survivin led to defects in erythropoiesis in approximately 20% of the animals. Upon pI-pC treatment, affected survivin heterozygous floxed mice with the MX1-Cre transgene exhibited a striking decrease in CD71−Ter119hi cells at the terminal stages of maturation. Morphological analysis of erythroid cells from neonatal spleens of these affected mice revealed a dramatic reduction in enucleated erythrocytes in the CD71−Ter119hi population compared to control animals. In addition, the majority of these erythroid cells were megaloblastic, with less condensed nuclear chromatin and nuclear-cytoplasmic dysynchrony. These observations suggest that there is a dose dependence for survivin. We predict that haploinsufficiency of survivin specifically affects “differentiation divisions” that occur concomitant with terminal maturation of red blood cells. Taken together, we show that homozygous deletion of survivin results in a dramatic loss of hematopoietic progenitors and rapid death, while heterozygous deletion leads to a selective defect in red blood cell development.

GATA-1 is a critical transcription factor that governs megakaryocyte development and differentiation. GATA-1-deficient megakaryocytes proliferate excessively and fail to undergo terminal differentiation, likely as a consequence of an aberrant gene expression program. In our studies to identify key signal transduction pathways that are dysregulated in the absence of GATA-1, we observed that STAT1, a gene of unknown function in megakaryocytes, was significantly downregulated in GATA-1-deficient megakaryocytes. Surprisingly, ectopic expression of STAT1, or IRF-1, a downstream effector of STAT1 signaling, promoted megakaryocytic differentiation of G1ME cells, a GATA-1-null and TPO-dependent erythromegakaryocytic cell line. Both STAT1- and IRF-1-transduced G1ME cells, but not GFP-expressing control cells, displayed multiple markers of terminal differentiation, including expression of CD42, as well as a significant fraction of polyploid cells. BrdU labeling assays confirmed that these transduced cells continued to effectively synthesize DNA, suggesting that STAT1 signaling switches these cells from mitosis to endomitosis. STAT1 and IRF-1 also modulated TPO signaling evidenced by enhanced STAT3 and STAT5 phosphorylation. Since STAT1 signaling is activated by IFN-g, we further investigated whether IFN-g-treatment also led to polyploidization of G1ME cells. IFN-g effectively promoted polyploidization of both G1ME cells and primary bone marrow-derived CD41+ cells in the absence of TPO. Finally, overexpression of STAT1 in primary bone marrow cells isolated from GATA-1 knockdown mice also promoted polyploidization in the absence of TPO. Taken together, these data reveal a GATA-1-independent mechanism by which STAT1 signaling promotes megakaryocyte differentiation.

Megakaryocytes, which fragment to give rise to platelets, undergo a unique form of cell cycle, termed endomitosis, to become polyploid and terminally differentiate. During this process, cells transverse the cell cycle but the late stages of mitosis are bypassed to lead to accumulation of DNA up to 128N. While the mechanisms of polyploidization in megakaryocytes are poorly understood, a few cell cycle regulators, such as cyclin D3, have been implicated in this process. Hematopoietic transcription factors, including GATA-1 and RUNX1 are also essential for polyploidization, as both GATA1-deficient and RUNX1-null megakaryocytes undergo fewer rounds of endomitosis. Interestingly, GATA-1 deficient megakaryocytes are also smaller than their wild-type counterparts. However, the link between transcription factors and the growth and polyploidization of megakaryocytes has not been established. In our studies to identify key downstream targets of GATA-1 in the megakaryocyte lineage, we discovered that the cell cycle regulators cyclin D1 and p16 were aberrantly expressed in the absence of GATA-1: cyclin D1 expression was reduced nearly 10-fold, while that of p16ink4a was increased 10-fold. Luciferase reporter assays revealed that GATA-1, but not the leukemic isoform GATA-1s, promotes cyclinD1 expression. Consistent with these observations, megakaryocytes that express GATA-1s in place of full-length GATA-1 are smaller than their wild-type counterparts. Chromatin immunoprecipitation studies revealed that GATA-1 is bound to the cyclin D1 promoter in vivo, in primary fetal liver derived megakaryocytes. In contrast, GATA-1 is not associated with the cyclin D1 promoter in erythroid cells, which do not become polyploid. Thus, cyclin D1 is a bona fide GATA-1 target gene in megakaryocytes. To investigate whether restoration of cyclin D1 expression could rescue the polyploidization defect in GATA-1 deficient cells, we infected fetal liver progenitors isolated from GATA-1 knock-down mice with retroviruses harboring the cyclin D1 cDNA (and GFP via an IRES element) or GFP alone. Surprisingly, expression of cyclin D1 did not increase the extent of polyploidization of the GATA-1 deficient megakaryocytes. However, co-overexpression of cyclin D1 and Cdk4 resulted in a dramatic increase in polyploidization. Consistent with the model that cyclinD:Cdk4/6 also regulates cellular metabolism, we observed that the size of the doubly infected cells was also significantly increased. Finally, in support of our model that cyclin D:Cdk4/6 kinase activity is essential for endomitosis, we discovered that introduction of wild-type p16 TAT fusion protein, but not a mutant that fails to interact with Cdk4/6, significantly blocked polyploidization of primary fetal liver derived megakaryocytes. Taken together, our data reveal that the process of endomitosis and cell growth relies heavily on cyclinD:Cdk4/6 kinase activity and that the maturation defects in GATA-1 deficient megakaryocytes are due, in part, to reduced Cyclin D1 and increase p16 expression.

Loss of a whole chromosome 5 or a deletion of the long arm of this chromosome, del(5q), is observed in 10% of patients with a myelodysplastic syndrome (MDS) or acute myeloid leukemia arising de novo, and in 40% of patients with therapy-related MDS or AML (t-MDS/t-AML). To identify a leukemia-related gene on chromosome 5, we previously delineated a 970 kb segment of 5q that is deleted in all patients examined, and prepared a genomic contig and transcript map of this region. Mutation analysis of 20 candidate genes within the commonly deleted segment did not reveal inactivating mutations in the remaining alleles, nor was there evidence of transcriptional silencing via DNA methylation. These observations are compatible with a haploinsufficiency model in which loss of one allele of the relevant gene(s) perturbs cell fate. One candidate gene is EGR1, which encodes a zinc finger transcription factor that is a member of the WT1 family of transcription factors. EGR1 has been shown to regulate hematopoietic cytokine levels (IL3, GM-CSF). Mouse embryo fibroblasts that are null or hemizygous for Egr1 bypass senescence and have apparently immortalized growth characteristics, consistent with loss of a tumor suppressor gene. Subsequent studies revealed that Egr1 is a transcriptional activator of both the p53 and p21Cip1/Waf1 genes during the stress response as well as during senescence, thereby representing a critical mechanism for controlling proliferation, growth arrest, differentiation, and apoptosis. Loss of Egr1 function may allow hematopoietic stem cells to bypass p53-mediated senescence or apoptosis, thereby contributing to leukemogenesis. To evaluate the role of Egr1 in hematopoiesis, we obtained Egr1+/− mice from J. Milbrandt (Washington University). Egr1−/− mice are viable, with mild post-natal growth retardation and infertility. Complete blood counts and body weight were collected for wild-type, Egr1+/−, and Egr1−/− mice every 6 weeks over the course of one year. The Egr1−/− mice display elevated white blood cell counts, elevated lymphocytes, and decreased neutrophil counts, and are unable to maintain normal RBC counts, Hb, and Hct as compared to wild-type and heterozygous mice. To examine the role of Egr1 in erythropoiesis, we evaluated the erythropoietic response to phenylhydrazine-induced hemolytic anemia by treating Egr-1-deficient mice with standard doses of phenylhydrazine (60 mg/kg). Egr1+/− and Egr1−/− mice were unable to respond to the anemia and died with 2 days of the treatment, whereas wild type mice recovered fully within 12 days of the treatment. Egr1-deficient mice treated with N-nitroso-N-ethylurea (ENU), a potent DNA alkylating agent, develop MDS or T cell lymphoma. Egr1-deficient mice developed lymphomas at an increased frequency and rate compared to wild-type animals, indicating that Egr1 cooperates with other mutations in the genesis of hematopoietic neoplasms. MDS was seen only in the Egr1-deficient mice, and is characterized by elevated white blood cell counts, anemia, and thrombocytopenia, with ineffective erythropoiesis in the bone marrow and spleen. Together, these results suggest a role for Egr1 in murine erythropoiesis and implicate EGR1 in the development of myeloid leukemias characterized by abnormalities of chromosome 5.

Trisomy 21 causes human Down syndrome (DS), a heterogeneous group of phenotypes including marked predisposition to leukemia. Children with Down syndrome are 500 times more likely to develop acute megakaryoblastic leukemia (AMKL) than other children. Furthermore, non-DS children with AMKL often have acquired trisomy 21 in their leukemic clones, suggesting that trisomy 21 in hematopoietic cells contributes to leukemic transformation. To better understand the impact of trisomy 21 on blood cell homeostasis and leukemia, we studied hematopoiesis in the Ts65Dn mouse model of Down syndrome. Ts65Dn mice harbor a segmental trisomy for mouse chromosome 16, homologous to human chromosome 21, and display many of the phenotypes associated with human DS, including craniofacial anomalies and learning deficits. To define the hematopoietic parameters for this strain, we performed monthly complete blood counts for a cohort of trisomic mice and their disomic littermates and discovered the development of progressive thrombocytosis and mild anemia in trisomic animals. Increased numbers of CD41+ megakaryocytes with lower modal ploidy were detected in the bone marrow and spleen of Ts65Dn mice as early as three months of age. Over time, expansion of the megakaryocyte population was accompanied by a decrease in TER119+ cells in the bone marrow, myelofibrosis, splenomegaly, and extramedullary hematopoiesis. Colony forming assays demonstrated increased colony forming ability in the spleens of trisomic mice along with variable decreased hematopoiesis in the bone marrow. Further, characterization of stem cells in the bone marrow indicated a hyperproliferative stem cell population. Importantly, the mice did not develop malignant leukemia by the age of 18 months and no mutations were found in the blood transcription factor GATA1, which is commonly affected in human AMKL. While Ts65Dn mice do not develop the AMKL seen in humans with DS, our results indicate that, trisomy 16 can cause hyperproliferation of the myeloid lineages, extramedullary hematopoiesis, and bone marrow fibrosis in mice. Additionally, these results suggest that trisomy 21 in humans may initiate a similar process in hematopoietic stem cells, which may contribute to leukemogenesis. Unexpectedly, this phenotype also bears significant resemblance to the human myeloproliferative disease chronic idiopathic myelofibrosis (CIMF). These findings may provide insight into the origins and progression of human myeloid diseases, including AMKL and CIMF.

Down syndrome-associated acute megakaryocytic leukemia (DS-AMKL) is a complex malignancy that evolves in hematopoietic progenitors with trisomy 21 that acquire a somatic mutation in the blood transcription factor GATA1. The mechanistic relationship between these two genetic factors that leads to leukemia is poorly understood. In order to study the interplay between trisomy 21 and GATA1 mutations, we are developing a mouse model of this malignancy. The Ts65Dn mouse, which contains a segmental trisomy for mouse chromosome 16, homologous to human 21, has been reported to display several of the cognitive and craniofacial phenotypes seen in humans with DS, but the hematopoietic system has not been assessed as a model for blood development in humans with DS. We have evaluated adult hematopoiesis in the Ts65Dn strain by comparing monthly complete blood counts (CBC) of peripheral blood from 14 trisomic and 20 disomic littermates. Similar to humans with DS, Ts65Dn trisomic mice display persistent erythrocyte macrocytosis, with values at the high end of the normal range. Trisomic mice also harbor decreased numbers of red blood cells, mildly elevated platelet counts, a higher percentage of monocytes and a lower hemoglobin concentration. Interestingly infants with DS frequently display thrombocytosis. In addition, we have characterized fetal liver hematopoiesis in the Ts65Dn strain by FACS analysis of hematopoietic precursors and by performing colony assays. In general, we did not detect any significant differences in erythroid, myeloid, or megakaryocytic colony formation between trisomic or disomic fetuses. Likewise flow cytometry for CD34, TER119, and CD41 demonstrated overall similar numbers of cells in these compartments for Ts65Dn mice and disomic littermates. However, one of seven trisomic embryos displayed a significant increase in the proportion of CD34+ cells with concomitant decrease in both Ter119+ and CD41+ populations. In addition, cells from this fetal liver gave rise to seven-fold and three-fold increases in BFU-E and CFU-Mk colonies respectively, with no change in the CFU-GM. Although the sample size is small, these findings suggest that a subset of Ts65Dn trisomic fetuses exhibit aberrant hematopoiesis. Taken together, our study indicates that the Ts65Dn trisomic mouse may be an excellent model to study human DS hematopoiesis.

Summary Tyrosine kinases are aberrantly activated in numerous malignancies, including acute myeloid leukemia (AML). To identify tyrosine kinases activated in AML, we developed a screening strategy that rapidly identifies tyrosine-phosphorylated proteins using mass spectrometry. This allowed the identification of an activating mutation (A572V) in the JAK3 pseudokinase domain in the acute megakaryoblastic leukemia (AMKL) cell line CMK. Subsequent analysis identified two additional JAK3 alleles, V722I and P132T, in AMKL patients. JAK3 A572V , JAK3 V722I , and JAK3 P132T each transform Ba/F3 cells to factor-independent growth, and JAK3 A572V confers features of megakaryoblastic leukemia in a murine model. These findings illustrate the biological importance of gain-of-function JAK3 mutations in leukemogenesis and demonstrate the utility of proteomic approaches to identifying clinically relevant mutations.