Search

The ontogeny of the hematopoietic system consists of two broad programs. The first, an HSC-independent program, consists of overlapping waves of primitive, erythro-myeloid (EMP), and some lymphoid progenitors. HSC-independent hematopoiesis is required for normal fetal development, and provides self-renewing tissue-resident macrophage populations that persist in the adult. This is followed by the emergence of an HSC-dependent program that arises from arterial vessels within the body of the embryo. The overlapping emergence and lineage output of HSC-independent and HSC-derived hematopoiesis raises important questions regarding the identity and potential functional differences of their mature progeny. However, the transition from HSC-independent to HSC-derived hematopoiesis in the murine fetus remains incompletely characterized, particularly since the maturing erythroid, megakaryocytic and myeloid progeny of EMP and HSCs are currently not easily distinguishable. Additionally, lineage-tracing approaches have been challenging because they have relied largely on the temporal induction of promoters that are expressed both in HSC-independent progenitors and in HSCs, which have significant temporal overlap in their developmental emergence and result in incomplete or in mixed labeling. To help resolve this question, we have developed Mds1 CreERT2 mice, utilizing the first transcription start site of MECOM gene, which is expressed in HSC and emerging HSC (Yuasa et al., 2005 EMBO; Hou et al. 2020 Cell Research; Zhu et al. 2020, Blood). When mated with Rosa-YFP reporter mice and induced at E9.5 with tamoxifen, this construct lineage-traces pre-HSCs present in the E11.5 AGM region, as well as HSCs in the fetal liver and adult marrow. Importantly, no labeling of primitive erythroid cells, primitive macrophage-derived microglia, EMP, or EMP-derived cells in the E11.5 or E12.5 fetal liver was detected with tamoxifen induction at either E9.5 or E8.5. Analysis of E9.5 tamoxifen-treated Mds1 CreERT2Rosa26 LSL-YFP embryos indicates that HSCs have begun to generate small numbers of differentiating erythroid, myeloid and lymphoid progeny in the liver between E12.5 and E14.5. By E16.5, a significant proportion of differentiating erythroid, myeloid and B-cell lineage cells in the liver are HSC-derived, and HSCs have now begun to contribute erythroid and myeloid cells to the rapidly expanding pool of circulating blood cells. In the adult, we found increasing contributions of HSCs to macrophages in liver, lung and kidney. Interestingly, the majority of F4/80+ cells in the adult bone marrow and spleen were also lineage-traced in these mice. Thus, HSCs ultimately provide the majority of adult marrow macrophages that go on to self-maintain in the adult marrow (Hashimoto et al., 2013, Immunity). The Mds1 CreERT2 mouse model will serve as a useful to deconvolute the complexity of hematopoiesis as it unfolds in the embryo and functions postnatally. Disclosures Palis: Rubius Therapeutics: Consultancy.

Background: Activation of MDS1-EVI1 Complex (MECOM) via proviral insertion at its gene locus (Morishita et al., 1988) or through chromosomal rearrangements at 3q26 (Suzukawa et al., 1994; Tang et al., 2019) results in marked overexpression of EVI1, and is predominantly associated with myeloid malignancies (Secker-Walker et al., 1995), especially high-risk acute myeloid leukemia (AML) (Barjesteh van Waalwijk van Doorn-Khosrovani et al., 2003). We previously showed that EVI1 overexpression skews hematopoiesis toward the myeloid lineage via the upregulation of a master regulator of myelopoiesis PU.1 encoded by SPI1 (SFPI1) (Ayoub et al., 2018), which explains the high cooccurrence of 3q abnormalities in myeloid malignancies. However, the role of EVI1 in reduced survival and high rates of relapse in AML patients is still unknown, and therapies specific for EVI1+ AML are absent due to the lack of EVI1 specific targets. Here we describe a novel mechanism of action of EVI1 in controlling leukemic stem cell quiescence by upregulating an essential member of the cyclin-dependent kinase inhibitor (cip/kip) gene family: CDKN1C/P57Kip2. Results: We first identified the upregulation of CDKN1C in the presence of EVI1 overexpression using RNA-seq on sorted hematopoietic stem and progenitor leukemia cells from our previously published EVI1+ leukemia mouse model (EVI1TO) (Ayoub et al., 2018). Since CDKN1C/ P57 expression has been linked to high relapse rates in AML patients following chemotherapy (Radujkovic et al., 2016), we confirmed the upregulation of CDKN1C using RNA-seq on sorted populations (mononucleated cells (MNCs), leukemia stem cells (LSCs), and minimal residual disease (MRD)) from activated-MECOM AML patients in comparison to nonactivated-MECOM high risk AML patients (67 samples from 27 patients). Additionally, we found EVI1 overexpression produces a reversible block in differentiation and engraftment that can be rescued by ablating EVI1 overexpression or by additional mutations. In an effort to identify EVI1-induced transcriptional regulations for CDKN1C, we performed chromatin immunoprecipitation and sequencing (ChIP-seq) and assayed for transposase accessible Chromatin (ATAC-seq) using our in vivo and in vitro EVI1 overexpression models. ChIP-seq for EVI1 showed a binding site for EVI1 located 300 kb from CDKN1C transcription start site (TSS), and ATAC-seq in EVI1 overexpressing AML showed an open chromatin pattern at the CDKN1C TSS only in the presence of EVI1 overexpression. We are presently investigating the molecular mechanism involved in the transcriptional regulation of CDKN1C in the presence of EVI1 overexpression. Conclusions: Our data suggest a correlation between EVI1 and CDKN1C expression in high risk AML with 3q abnormalities and provide insights into a potential molecular mechanism for MECOM activation in controlling leukemic stem cell quiescence and resistance to chemotherapy by upregulating CDKN1C. Disclosures Andreeff: Daiichi-Sankyo; Jazz Pharmaceuticals; Celgene; Amgen; AstraZeneca; 6 Dimensions Capital: Consultancy; Daiichi-Sankyo; Breast Cancer Research Foundation; CPRIT; NIH/NCI; Amgen; AstraZeneca: Research Funding; Centre for Drug Research & Development; Cancer UK; NCI-CTEP; German Research Council; Leukemia Lymphoma Foundation (LLS); NCI-RDCRN (Rare Disease Clin Network); CLL Founcdation; BioLineRx; SentiBio; Aptose Biosciences, Inc: Membership on an entity's Board of Directors or advisory committees; Amgen: Research Funding.

While hematopoietic stem cells (HSCs)-intrinsic effects of aging have been explored, less is known about how HSC support is altered by the aged bone marrow microenvironment (BMME). To assess the role of the BMME in HSC aging, we compared the BMME in young (6-12 weeks) and aged (20-24 months) male mice and young (50 YO) human volunteers. Aged mice had remodeling of the BMME, with expansion of the marrow cavity and vascular volume compared to young mice. BMME constituents were redistributed within two distinct anatomic regions, namely endosteal bone-associated (BA) and marrow-associated (MA) cells. BA cells in aged mice contained fewer phenotypic mesenchymal/osteoblastic progenitors, with reduction in their ability to constitute colony forming units (CFUs). CFU loss was also observed in aged human volunteers. Aged murine MA had significant expansion of dysfunctional mesenchymal stem cells (MSCs) and activated macrophages (MΦ). Increased MΦ were also detected in aged human marrows. Following this in vivo characterization, we developed an ex vivo co-culture system to determine if aged murine BMME cells could impart aging characteristics to young HSCs. Young murine HSCs co-cultured with aged MA cells acquired phenotypic properties of aged HSCs, including increased CD41+ expression. Single cell RNA sequencing of Long Term-HSCs (LT-HSCs) from young and aged mice also identified upregulation of integrin-β3 (CD61) as a novel marker of aged LT-HSCs. Subsequent flow cytometry analysis confirmed the increase in CD61+ expression in vivo in aged HSCs. Importantly, aged MA - but not BA cells - also increased CD61+ expression in young HSCs ex vivo, highlighting the region-specific remodeling of the BMME that occurs with age. We then used a reductionist approach to identify targetable cellular and molecular regulators of the region-specific BMME-induced HSC aging. CD45+ and Ter119+ depletion in aged MA cells did not induce CD41+ expression in young HSCs, suggesting that a critical BMME component responsible for non-cell-autonomous HSC aging is present within the hematopoietic pool. Since marrow MΦ can regulate HSCs, we co-cultured aged MA MΦ with young MA and found that aged MΦ were sufficient to increase CD41+ expression in young HSCs. The addition of aged MΦ also expanded young MSCs, demonstrating that MΦ orchestrate both BMME remodeling and HSC aging. We next aimed to explore mechanisms by which aged MA MΦ impart aging characteristics to HSCs. Transcriptional analysis of murine MA MΦ demonstrated an increase in inflammatory activation in aged mice compared to young mice. This finding was also present in aged human MΦs. Among the inflammatory signals, interleukin-1β (IL-1β) was identified to be necessary and sufficient to mediate the aging effect of aged MA MΦ on young HSCs. Transcriptional analysis also revealed downregulation of phagocytic programs in aged MA MΦ compared to young MA MΦ. Supporting the transcriptional data, aged MA MΦs cultured in vitro demonstrated impaired ability to engulf senescent neutrophils compared to young MA MΦ. Bone marrow MΦ continuously remove large quantities of senescent neutrophils through phagocytosis, a process also known as efferocytosis. Complementing the in vitro findings, in vivo testing demonstrated that young MA MΦ are primarily responsible for engulfing senescent neutrophils and that aged MA MΦ had reduced engulfment of senescent neutrophils. No phagocytic defect was identified in aged BA MΦ, highlighting the regionalization of MΦ function within the BMME that is differentially impacted with age. Consistent with the systemic impact of the efferocytic defect of aged MA MΦ, aged mice had increased levels of circulating senescent neutrophils and. Moreover, neutrophils from aged mice had increased caspase-1 activity, a signal required for IL-1β activation. Together, these data provide evidence that aging differentially remodels two anatomically distinct BMMEs. Regional specialization of marrow MΦ was differentially impacted by aging and induced aging characteristics in HSCs. We propose that impaired removal of senescent neutrophils by aged MA MΦ increases IL-1β production, leading to local inflammation and disrupted BMME and HSC function in aged mice. Strategies aimed at restoring healthy efferocytic activity as well as diminishing IL-1β production or function could therefore reduce the aging effect on HSCs by rejuvenating the BMME. Disclosures Liesveld: Onconova: Honoraria; Seattle Genetics: Honoraria.

A subgroup of leukemogenic mixed-lineage leukemia (MLL) fusion proteins (MFPs) including MLL-AF9 activates the Mecom locus and exhibits extremely poor clinical prognosis. Mecom encodes EVI1 and MDS1-EVI1 (ME) proteins via alternative transcription start sites; these differ by the presence of a PRDI-BF1-RIZ1 (PR) domain with histone methyltransferase activity in the ME isoform. Using an ME-deficient mouse, we show that ME is required for MLL-AF9-induced transformation both in vitro and in vivo. And, although Nup98-HOXA9, MEIS1-HOXA9, and E2A-Hlf could transform ME-deficient cells, both MLL-AF9 and MLL-ENL were ineffective, indicating that the ME requirement is specific to MLL fusion leukemia. Further, we show that the PR domain is essential for MFP-induced transformation. These studies clearly indicate an essential role of PR-domain protein ME in MFP leukemia, suggesting that ME may be a novel target for therapeutic intervention for this group of leukemias.

Abnormal expression of the transcription factor EVI1 through chromosome 3q26 rearrangements has been implicated in the development of one of the most therapeutically challenging high-risk subtypes of acute myeloid leukemia (AML). Here we integrated genomic and metabolic screening of hematopoietic stem cells to reveal that EVI1 overexpression altered cellular metabolism. A pooled shRNA screen targeting metabolic enzymes identified the ATP-buffering, mitochondrial creatine kinase CKMT1 as a druggable dependency in EVI1-positive AML. Of 18 screened AML cell lines harboring various genetic alterations, only the four EVI1-expressing lines exhibited markedly elevated CKMT1 protein expression and activity. Treatment of this cell line panel with either CKMT1-targeting shRNAs or cyclocreatine, an analog of the CKMT1 substrate creatine and inhibitor of the creatine biosynthesis pathway, showed that elevated CKMT1 protein expression correlated with sensitivity to CKMT1 pathway inhibition. Consistent with these data, flow cytometry analysis of a panel of 68 unselected primary AML patient specimens revealed that the four leukemias with the highest levels of EVI1 expression also had elevated CKMT1 protein levels and enhanced sensitivity to cyclocreatine treatment. We next established that enforced EVI1 expression increased CKMT1 protein and mRNA levels and that three independent shRNA molecules targeting EVI1 drastically reduced CKMT1 expression in two EVI1-positive AML cell lines. A luciferase-based reporter system established that RUNX1 represses CKMT1 expression through direct binding to its promoter. ChIP-qPCR approaches were then applied to dissect the sequential events involved in EVI1-induced CKMT1 upregulation and the possible role of RUNX1 as an intermediate. In both primary AML samples and cell lines, we determined that EVI1 represses RUNX1 expression by directly binding to its promoter. This, in turn, eliminates repressive RUNX1 binding at the CKMT1 promoter and thereby promotes CKMT1 expression. Based on these data, we explored the relationship between EVI1 and RUNX1 expression with CKMT1 mRNA levels in two AML transcriptional datasets (GSE14468 and GSE10358). We divided these cohorts into four subgroups with high versus low expression of EVI1 and RUNX1. Consistent with our mechanistic analysis, primary AML samples within the EVI1high/RUNX1low subgroup were significantly more likely to express high levels of CKMT1 than AML samples in the other three subgroups. CKMT1 promotes the metabolism of arginine to creatinine. To determine the effect of CKMT1 suppression on this pathway, we measured the metabolic flux of stable-isotope labeled L-arginine 13C6 through creatine synthesis using mass spectrometry. CKMT1-directed shRNAs or cyclocreatine selectively decreased intracellular phospho-creatine and blocked production of ATP by mitochondria. Salvage of the creatine pathway by exogenous phospho-creatine restored normal mitochondrial function, prevented the loss of viability of human EVI1-positive AML cells induced by cyclocreatine or CKMT1-directed shRNAs, and maintained the serial replating activity of Evi1-transformed bone marrow cells. Primary human EVI1-positive AML is frequently associated with somatic NRAS mutations. Thus, to investigate whether EVI1 over-expression sensitizes primary AMLs to CKMT1 inhibition in vivo, we transplanted primary NrasG12D mutant AMLs with and without elevated Evi1 expression into congenic recipient mice. In this system, Ckmt1 knockdown did not significantly alter the outgrowth of control Nras mutant AML cells compared to a shControl (63% versus 71%). By contrast, NrasG12D AML cells characterized by elevated Evi1 expression were profoundly depleted by Ckmt1 suppression to 2% versus 58% in shControl recipients. Consistent with these results, pharmacologic or genetic inhibition of the CKMT1-dependent pathway blocked disease progression and prolonged the survival of mice injected with human EVI1-positive cells but not with EVI1-negative cells, without noticeable cytotoxic effect on normal murine cells. In conclusion, we have integrated "omic" approaches to identify CKMT1 as a druggable liability in EVI-positive AML. This study supports a potential therapeutic avenue for targeting the creatine kinase pathway in EVI1-positive AML, which remains one of the worst outcome subtypes of AML. Disclosures DeAngelo: Incyte: Consultancy; Novartis: Consultancy; Celgene: Consultancy; Amgen: Consultancy; Baxter: Consultancy; Pfizer: Consultancy; Ariad: Consultancy. Stone:Pfizer: Consultancy; Agios: Consultancy; Jansen: Consultancy; Celator: Consultancy; Merck: Consultancy; Amgen: Consultancy; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Consultancy; Novartis: Consultancy; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees; Xenetic Biosciences: Consultancy; Sunesis Pharmaceuticals: Consultancy; Seattle Genetics: Consultancy; Roche: Consultancy; Juno Therapeutics: Consultancy; ONO: Consultancy.

The hematopathology subcommittee of the Mouse Models of Human Cancers Consortium recognized the need for a classification of murine hematopoietic neoplasms that would allow investigators to diagnose lesions as well-defined entities according to accepted criteria. Pathologists and investigators worked cooperatively to develop proposals for the classification of lymphoid and nonlymphoid hematopoietic neoplasms. It is proposed here that nonlymphoid hematopoietic neoplasms of mice be classified in 4 broad categories: nonlymphoid leukemias, nonlymphoid hematopoietic sarcomas, myeloid dysplasias, and myeloid proliferations (nonreactive). Criteria for diagnosis and subclassification of these lesions include peripheral blood findings, cytologic features of hematopoietic tissues, histopathology, immunophenotyping, genetic features, and clinical course. Differences between murine and human lesions are reflected in the terminology and methods used for classification. This classification will be of particular value to investigators seeking to develop, use, and communicate about mouse models of human hematopoietic neoplasms.

Ecotropic Viral Integration Site 1 (EVI1) is a zinc finger oncoprotein implicated in a subtype of acute myeloid leukemia associated with poor prognosis. Leukemic cells overexpressing EVI1 display a block in myeloid differentiation and resistance to apoptosis, both of which are reversed with EVI1 shRNA knockdown. Previous mutagenesis studies have shown that DNA binding via the first set of zinc fingers (ZF1) is critical for leukemic transformation by EVI1. The mediators between EVI1 DNA-binding and leukemogenesis are mostly unknown. Using chromatin immunoprecipitation and sequencing (ChIP-Seq) in myeloid leukemic cells, we have recently identified an evolutionarily conserved EVI1 binding site 35 kilobases downstream of Cebpa, encoding the CCAAT enhancer-binding protein alpha, a master regulator of early granulopoiesis. In order to obtain a better understanding of the potential regulatory role of this region in leukemogenesis, we used genome database analysis to see what other transcription factors or epigenetic marks are present. ChIP-Seq data deposited in the UCSC databank indicates the presence of CTCF, GATA2 and CEBPβ, as well as TAL1, p300, and GATA1 in leukemic cells, the latter three of which are known to bind together in a complex. This data suggests that the Cebpa +35 region serves an important role in regulating hematopoietic development and Cebpa transcription. In order to characterize the binding of EVI1 ZF1 to the Cebpa +35 site, we performed methylation interference assays and oligonucleotide mutagenesis assays. EVI1 ZF1 binds within this +35kb control region to the sequence TGACAGTGACGG with 133% affinity relative to the canonical motif. In an effort to assess any biological effects linked to EVI1 binding at the Cebpa +35 motif, we conducted EVI1 overexpression studies in the EML hematopoietic progenitor cell line. While control EML cells exhibited marked upregulation of Cebpa upon induction, EVI1-transduced cells displayed a significant suppression of Cebpa transcript. To determine if this suppression requires DNA binding, we tested a mutant of EVI1 that is defective in DNA binding (EVI1R205N); this was unable to suppress Cebpa induction. To test whether EVI1 can bind to the +35 KB site, we performed ChIP-PCR, which indeed showed EVI1 occupancy at the +35 KB site. In conclusion, we have shown evidence that EVI1 suppresses Cebpa transcription in early hematopoietic progenitor cells, and that the EVI1 binding site at Cebpa +35 serves an important function in myeloid maturation. This regulatory node represents a possible therapeutic target for the treatment of acute myeloid leukemia. Disclosures: No relevant conflicts of interest to declare.

Abstract 3652 Poster Board III-588 SRF is a MADS-box transcription factor first identified as an inducer of immediate-early gene expression, such as c-fos, Junb, Fosb and Egr1, in response to cytokines. It plays a critical role in cardiac and smooth muscle development and differentiation by regulating cell cycle, apoptosis, cell growth, and differentiation. SRF regulates expression of contractile and cytoskeletal genes. Its function in hematopoiesis has not yet been revealed. While no hematopoietic disease specific mutations of SRF have been identified, the region on chromosome 6 where the SRF gene is located, is a site of frequent chromosomal aberrations in MDS. MKL1, a target of the t(1;22) translocation in acute megakaryoblastic leukemia, is a cofactor for SRF mediated gene activation. In published work (Cheng EC et al., Blood 2009;113:2826), we showed that MKL1 expression increases with normal megakaryocytic (Mk) differentiation, that it promotes Mk differentiation by increasing the expression of Mk specific genes such as CD42b, and that mice lacking MKL1 have decreased platelet counts while showing increased numbers of low ploidy Mk in the BM. In addition, we showed that the effects of MKL1 on Mk differentiation require SRF. In order to test the role of SRF in Mk development, we crossed PF4-Cre mice (Tiedt R et al., Blood 2007;109:1503), which express Cre recombinase in cells committed to the Mk lineage, to SRFF/F mice (Miano JM et al., PNAS 2004;101:17132) in which functional SRF is no longer expressed after Cre-mediated excision. Our findings are quite surprising, as knockout of SRF in the Mk lineage leads to a far more profound phenotype than MKL1 KO. PF4-Cre x SRFF/F (PF4-Cre/SRF) mice are born with a normal mendelian frequency, but have significant macrothrombocytopenia. The platelet counts in WT and KO mice are 703 ± 33 × 103/μl vs 460 ± 23 × 103/μl, respectively. Despite the decreased platelet number, the BM has increased numbers and percentages of CD41+ Mk (WT: 0.41 ± 0.06% and KO: 1.92 ± 0.12%) with a predominance of Mk with hypolobated nuclei. Spleens of PF4-Cre/SRF mice show significantly increased numbers of Mk (mean of 4 and 15 Mk per high power field for WT and KO spleens, respectively). Ploidy as an assay of Mk maturation in the BM is significantly reduced in SRF −/− Mk. The percentage of CD41+ cells in SRF KO BM that is 2N is nearly 2-fold higher, and the percentage with greater than 8N ploidy is decreased by 70%. Hematopoietic stem and progenitor populations in PF4-Cre/SRF BM are unaffected, as expected due to stage and lineage specific knockout of SRF. In contrast, there is an increase in Mk progenitors in vivo in PF4-Cre/SRF mice, reflected both by FACS analysis and CFU-Mk in vitro analysis. The mechanism by which SRF disrupts Mk maturation was studied by assessment of Mk morphology, and gene expression analysis. Mk lacking SRF expression have abnormal stress fiber formation based on phalloidin staining, a phenotype parallel to that observed in mice lacking myosin heavy chain 9 (MYH9) expression, and MYH9 expression was decreased in SRF KO Mk. In summary, the data show that SRF is critical for normal Mk maturation including polyploidization and platelet production, and the mechanism by which loss of SRF disrupts megakaryocytopoiesis is, at least in part, by loss of normal expression of known targets of SRF, such as cytoskeletal genes including MYH9. Disclosures: No relevant conflicts of interest to declare.

The EVI1 gene at chromosome 3q26 is associated with acute myeloid leukemogenesis, due to both chromosomal rearrangement and to overexpression in the absence of rearrangement. Some rearrangements such as t(3;3) and inv(3) result in overexpression of EVI1 protein, while translocation t(3;21) yields an AML1-MDS1-EVI1 (AME) fusion protein. EVI1 possesses two zinc finger domains, an N-terminal domain with fingers 1–7, which binds to GACAAGATA, and a C-terminal domain (fingers 8–10) which binds GAAGATGAG. Inhibition of EVI1 function with a small molecule compound may provide a targeted therapy for EVI1-expressing leukemias. As a first step towards inhibiting the leukemogenic function of EVI1, we performed structure-function studies on both EVI1 and AME protein to determine what domains are critical for malignant transformation activity. Assays were Rat1 fibroblasts in a soft agar colony forming assay for EVI1; primary bone marrow cells in a serial replating assay for AME. Both assays revealed that mutation of arginine 205 in zinc finger 6 of EVI1, which completely abrogates sequencespecific DNA binding via the N-terminal zinc finger domain, resulted in complete loss of transforming activity; mutations in other domains, such as the C-terminal zinc finger domain, CtBP binding domain, and the domains of AML1 had less of an effect or no effect on transforming activity. In an effort to inhibit EVI1 leukemogenic function, we developed a polyamide, DH-IV-298, designed to block zinc fingers 1–7 binding to the GACAAGATA motif. DNAseI footprinting revealed a specific interaction between DH-IV-298 and the GACAAGATA motif; no significant interaction was observed elsewhere; a mismatch polyamide failed to footprint at equivalent concentrations; and DH-IV-298 failed to bind to a control DNA lacking the GACAAGATA motif. Electromobility shift assay showed that, at a 1:1 polyamide:DNA ratio, DH-IV-298 lowered EVI1:DNA affinity by over 98%, while mismatch was significantly less effective (74% reduction). To assess the effect of DH-IV-298 on EVI1 binding to DNA in vivo, we performed CAT reporter assays in a NIH-3T3-derived cell line with a chromosome-embedded tet-inducible EVI1-VP16 as well as a EVI1-responsive CAT reporter. Removal of tetracycline resulted in a four-fold increase in CAT activity that was completely blocked by DH-IV-298. The mismatch polyamide was significantly less effective than DH-IV-298. Further studies are being performed to assess the effect on endogenous gene expression, and on growth of leukemic cells that express EVI1. These studies provide evidence that a cell permeable small molecule compound may effectively block the activity of a leukemogenic transcription factor.

Retroviral insertion at the Evi1 locus causes high level of expression of the gene in myeloid neoplasms in mice and in nonmalignant expansions of myeloid cells in humans and monkeys. In these settings, it is suggested that EVI1 confers a survival advantage on myeloid cells. Here, we investigate the survival phenotype in DA-1 cells, a leukemic cell line with provirally activated Evi1. We report that short hairpin-mediated suppression of Evi1 in DA-1 cells induces apoptosis via the intrinsic/mitochondrial pathway: DNA fragmentation and histone release are both induced, as is reduction in mitochondrial membrane potential. In addition, procasapses 3 and 9, but not caspase 8 or Bid, are cleaved following Evi1 knockdown; phosphoAKT remains unchanged. Furthermore, mRNA expression profiling following Evi1 suppression show transcriptional changes in several apoptotic regulators, including a 3.5-fold decrease in Bcl2a1 (bfl/A1), a prosurvival member of the Bcl-2 family. To assess whether EVI1 regulates Bcl2a1 expression in a cell type other than DA-1 cells, we transduced primary murine Lin-/Sca-1+/c-Kit+ cells with EVI1 via retroviral vector, and then assessed the expression of Bcl2a1. This revealed that EVI1 induced a more than five-fold increase in Bcl2a1 mRNA expression, as measured by quantitative PCR. Furthermore, transduction of primary murine bone marrow cells with retroviruses bearing either Bcl2a1 or Evi1 resulted in a significant decrease in spontaneous apoptosis, as assessed by the activity of caspases 3 and 7. To further test if Bcl2a1 is necessary for leukemic transformation by Evi1, we assessed the ability of Evi1 to confer serial replating ability on primary bone marrow cells from Bcl2a1−/− mice. Bone marrow was harvested from Bcl2a1−/− and C57BL6 mice and transduced with retrovirus containing either no gene or Evi1. While in C57BL6 mice, Evi1 induced a significant increase in colonies, bone marrow from Bcl2a1−/− mice were resistant to transformation by Evi1. To show that this was due to the lack of Bcl2a1, the gene was added back via retrovirus. While Bcl2a1 by itself did not induce significant number of colonies over vector, when introduced into Bcl2a1−/− cells together with Evi1, there was a significant increase in colony formation. These data indicate that transformation of bone marrow cells by Evi1 depends on the presence of the Bcl2a1 gene. We further show EVI1 can transcriptionally upregulate a BCL2A1::luc reporter that harbors 1.37 kb of the human BCL2A1 upstream sequence. Our analysis of the Bcl2a1 promoter indicates that the effect of EVI1 is likely indirect. Based on our findings, we propose that EVI1 acts to block apoptosis in DA-1 cells by transcriptionally activating Bcl2a1.

The MDS1-EVI1 locus is of considerable interest due to its role in myeloid malignancies and dysplasias. It is well established now that the locus has two different transcription start sites (TSS) located 0.5 Mb apart, and these have the capacity to encode different isoforms, which variably contain zinc finger DNA binding domains and a SET-like domain that may have histone modifying ability. In order to better understand the biological role of this locus, we knocked in a lacZ allele into the Mds1 (upstream) TSS by homologous recombination in ES cells and created mice harboring this modified allele (K. Lezon-Geyda, S. Lin, G. Steele-Perkins et al, in preparation). By staining for beta-galactosidase activity, we documented the distribution of Mds1 activity during embryonic development and in the adult. During development, five major organ systems showed expression: musculoskeletal, renal, cardiac, neural, and hematopoietic, and in the latter three, there was a striking and highly specific spatiotemporal pattern of expression suggesting that Mds1-Evi1 plays important regulatory roles. In the developing heart, staining was seen in the anterior heart field specifically during the formation of the cardiac outflow tract, with significant spatiotemporal overlap with Mef2c, which encodes an important cardiac transcriptional regulatory protein. Thereafter, expression in the heart is very low. Beta-gal staining in the hematopoietic system in the embryo is limited to the clusters of nascent hematopoietic progenitors that develop at day 9.5 p.c. in the ventrolateral wall of the dorsal aorta and bud into the vascular lumen. Strikingly, we see no staining in other endothelium, nor in the fetal liver of 12.5–14.5 day embryos, wherein the majority of fetal hematopoiesis takes place. In adult bone marrow, there is beta-gal activity exclusively in the lin− c-kit+ Sca1+ progenitor population, with all of the beta-gal-positive cells being in the progenitor pool, and nearly all of the progenitor cells staining. While homozygous mice are viable, they are small, kyphotic, and have a shortened lifespan. Morphologic and quantitative analysis of the peripheral blood failed to reveal any significant abnormality. To assess the function of the hematopoietic system more rigorously, competitive repopulations of homozygous Mds1-deficient marrow progenitors with wildtype progenitors were performed. Within several weeks after transplant, the Mds1−/− cells were undetectable in the recipients, revealing that the homozygous Mds1-null bone marrow progenitors are deficient in their repopulating ability. To identify what function Mds1-Evi1 plays in hematopoetic cells, we used shRNA to suppress its expression in the myeloid cell lines 32Dcl3 and DA-1. This revealed an increase in steady state levels of cell death, as documented by histone release, TUNEL staining, and caspase activation. These data suggest that a primary role for the Mds1-Evi1 locus in hematopoietic cells is to promote their survival, thus allowing normal expansion at the progenitor stage.

Mds1-Evi1 is a complex locus with two distinct transcriptional start sites, four possible polyadenylation sites, and nineteen exons spread over half a megabase that encodes at least six protein isoforms containing variable numbers of DNA-binding zinc finger motifs. In mouse and some other species, the locus is a common site of proviral insertion in myeloid leukemia, while in human it is a site of nonrandom chromosomal rearrangement in myelodysplasia, AML, and CML blast crisis. One such translocation is t(3;21), encoding the AML1-MDS1-EVI1 (AME) fusion protein. We have characterized the RNA transcripts and proteins generated by Mds1-Evi1 in murine leukemic cells bearing Evi1 proviral insertion using three different antisera specific for the N-terminal and C-terminal domains of EVI1, as well as MDS1, and find consistent overexpression of three protein isoforms of 135, 123, and 103 kDa (EVI1A, B, and C, respectively) while other isoforms, including MDS1-EVI1, are not normally expressed. However, by knocking lacZ into the Mds1 first exon we showed that it is normally expressed in hematopoietic progenitors, similar to Evi1. The N-terminal zinc finger domain present in the EVI1A and B isoforms binds to the GACAAGATA motif with high affinity, and single missense mutations in zinc finger six abrogate high affinity binding. Using two different bioassays, we have demonstrated that cellular transformation by either the EVI1A isoform or by AME is dependent on DNA binding via this N-terminal domain, thus indicating that transformation is dependent on the action of this domain on specific target genes. To determine the biological effect of Evi1 on leukemic cell growth, short hairpin RNAs were used to suppress Evi1, and this resulted in slowing of the cell cycle and prolongation of the G2/M phase. RNA expression profiling revealed significant changes in gene expression accompanied Evi1 suppression, including some involved in cell cycle control. Upon long term Evi1 suppression, numerous markers of myeloid differentiation were induced, indicating the acquisition of a differentiation-like phenotype. These data indicate an effect of EVI1 on both cell cycle and differentiation, and suggest that the effect on cell cycle is direct. To gain further insight into the nature of EVI1 target genes, we utilized tet-regulated expression of Evi1A, and identified a set of putative targets of the A isoform that includes genes directly involved in cell cycle. However, the fold changes are relatively minor, and although the kinetics of activation suggests these are direct targets, this has not been conclusively demonstrated. We predict on the basis of these data that EVI1 expression modulates several key regulators of cell cycle and this results in an acceleration of cell cycle. With high level EVI1 expression, block to differentiation is induced as well.