ALBERT

Description: The requirement of c-Myb during erythropoiesis spurred an interest in identifying c-Myb target genes that are important for erythroid development. Here, we determined that the neuropeptide neuromedin U (NmU) is a c-Myb target gene. Silencing NmU, c-myb, or NmU's cognate receptor NMUR1 expression in human CD34+ cells impaired burst-forming unit-erythroid (BFU-E) and colony-forming unit-erythroid (CFU-E) formation compared with control. Exogenous addition of NmU peptide to NmU or c-myb siRNA-treated CD34+ cells rescued BFU-E and yielded a greater number of CFU-E than observed with control. No rescue of BFU-E and CFU-E growth was observed when NmU peptide was exogenously added to NMUR1 siRNA-treated cells compared with NMUR1 siRNA-treated cells cultured without NmU peptide. In K562 and CD34+ cells, NmU activated protein kinase C-βII, a factor associated with hematopoietic differentiation-proliferation. CD34+ cells cultured under erythroid-inducing conditions, with NmU peptide and erythropoietin added at day 6, revealed an increase in endogenous NmU and c-myb gene expression at day 8 and a 16% expansion of early erythroblasts at day 10 compared to cultures without NmU peptide. Combined, these data strongly support that the c-Myb target gene NmU functions as a novel cofactor for erythropoiesis and expands early erythroblasts.

Description: The c-myb proto-oncogene encodes a transcription factor, Myb, which is essential for normal hematopoiesis. Myb may also play a role in leukemogenesis but possible mechanisms remain ill defined. To gain further insights into this issue, we sought to identify Myb regulated genes in human myeloid leukemic cells utilizing a tamoxifen inducible expression system and a microarray approach. Myb function was conditionally abrogated by tamoxifen following infection of K562 cells with a bicistronic retroviral vector MIGR1 which had c-myb’s DNA binding domain (DBD) and the Drosophila engrailed protein transcription repressor domain (MERT) fused to a modified estrogen receptor that binds tamoxifen. MERT was subcloned upstream of an IRES-EGFP cassette in MIGR1 to allow for FACS purification on the basis of GFP expression. To identify Myb regulated genes, purified K562-MERT cells were exposed to 1 μM tamoxifen or ethanol (control) for three days, processed for hybridization to the microarray gene chip and analyzed by software algorithms from Incyte and Arrayex. When endogenous Myb activity was suppressed by MERT, 105 genes out of 10,000 genes on the microarray chip changed 〉 2-fold in expression. Of these 105 genes, 34 increased their expression 〉2-fold while 70 decreased their expression 〉2-fold. Since Myb expression is elevated in leukemic cells, we hypothesized that Myb functions in malignant hematopoietic cells to induce the expression of genes that are essential for their maintenance and survival. Therefore, we focused on those genes that decreased in expression when Myb activity was inhibited by MERT. Among the most repressed was cdc7 (2.8-fold decrease), an intra-S-phase regulator. To verify the microarray data, we utilized real-time PCR to quantitate the expression of cdc7 in our K562-MERT cells. Cdc7 expression decreased 5-fold in tamoxifen treated K562-MERT cells relative to control cells, which is consistent with the microarray data. We then performed chromatin immunoprecipitation (ChIP) experiments to determine whether cdc7 is a direct target of Myb. When the chromatin from untreated K562 and K562-MERT cells was immunoprecipitated with anti-c-Myb, we observed one PCR product using a primer pair that flanked the Myb binding sites in the promoter region of cdc7. This same result was observed in our positive control ChIP experiment in which the chromatin was immunoprecipitated with anti-acetyl histone H4, indicating that the region of the cdc7 promoter containing the Myb binding sites is poised for transcription. When Myb transactivation activity was inhibited by MERT in K562-MERT cells, no PCR product was observed following chromatin immunoprepitation with anti-c-Myb. These results strongly suggest that cdc7 is a direct gene target of c-Myb in malignant hematopoietic cells. Investigation of the transcriptional regulation of cdc7 in hematopoietic cells may yield new clues to Myb’s role in leukemogenesis.

Description: The c-myb proto-oncogene encodes a transcription factor, Myb, which is essential for the growth and survival of normal and malignant hematopoietic cells. We, and others, have previously shown that malignant hematopoietic cells are more dependent on c-Myb function than are normal hematopoietic cells. Based on these findings, we hypothesized that c-Myb regulates a unique set of genes in leukemic cells that are required for their growth. To identify Myb gene targets, we performed a transcriptome analysis using human myeloid leukemic cells engineered to express a conditionally active dominant negative Myb (MERT). Analysis of the microarray data derived from these experiments revealed that when Myb activity was inhibited, neuromedin U (NmU), a neuropeptide involved in energy homeostasis, decreased in expression 5 fold compared to control cells, a result that was confirmed by quantitative real-time PCR. Combined, the microarray and quantitative real-time PCR data suggested that Myb directly regulates NmU gene expression in hematopoietic cells. To address this question in the absence of a formally defined human NmU promoter, we examined the DNA sequence upstream of the predicted transcription start site (as noted in Genbank accession #NM_006681) for potential Myb transcription factor binding motifs. After scanning the DNA sequence (~2kb) upstream of the predicted transcription start site, eleven potential Myb response elements (MREs) were identified. Of these MREs, five were identified as canonical (PyAAC(G/C)G). Our search also identified potential AML1, PU.1, CBP, STAT3, and STAT5 binding motifs within the human NmU promoter region. To determine if any of the potential MREs within the NmU promoter were functional, we first completed in vitro assays using luciferase reporter constructs followed by in vivo assays using chromatin immunoprecipitation (ChIP) assays. The luciferase reporter constructs were generated after we determined the actual transcription start of human NmU by primer extension assays. Using a Fam-labeled NmU specific primer that annealed proximal to the predicted transcription start site, we observed a 20-nucleotide difference between the predicted and actual transcription start of NmU. When all eleven potential MREs within the NmU promoter were upstream of luciferase, a 6-fold increase in luciferase activity was observed compared to empty vector. We next systematically mutated the MREs to determine which one(s) Myb bound directly. Thus far, the in vitro luciferase assay has identified MREs at −446 and −626, which are proximal to NmU’s transcription start as important for Myb-mediated expression. To determine the physiologic relevance of our in vitro studies, we performed ChIP assays. When chromatin from K562 cells, a human myeloid leukemia cell line, was immunoprecipitated with anti-c-Myb, we observed the expected PCR product using primer pairs that flanked select MREs. These same results were obtained in our positive control ChIP experiment in which the chromatin was immunoprecipitated with anti-acetyl histone 4 indicating that the promoter region of NmU is poised for transcription. Further characterization of the regulation of NmU gene expression in normal and malignant hematopoietic cells may yield new clues to Myb’s role in leukemogenesis and could suggest new therapeutic targets in human leukemia cells.

Description: Abstract 1241 Erythropoiesis is a multi-step process during which hematopoietic stem cells terminally differentiate into red blood cells (RBCs). Erythropoietin (EPO) is the only known cytokine regulator of terminal erythroid differentiation. Previously, we reported that the neuropeptide, neuromedin U (NmU), which interacts with NmU receptor type 1 (NMUR1), functions as a novel extracellular cofactor with EPO to promote the expansion of early erythroblasts, which are CD34−, CD71+, glycophorin A (GlyA)dim(Gambone et al, Blood. 2011). Here, we describe studies to understand the mechanism whereby NmU augments EPO effects on erythroid cell growth. EPO triggers Janus kinase (Jak)-2 dependent activation of signal transducer and activator of transcription (STAT) 5 and phosphatidylinositol 3-kinase (PI3K) to promote the proliferation and/or survival of erythroid progenitor cells. We hypothesized that NmU peptide would cooperate with EPO to promote the proliferation of early erythroblasts through STAT5 and/or PI3K activation. To address this hypothesis, we cultured primary human CD34+ cells in 2-stage liquid culture with IL-3, IL-6, and stem cell factor (SCF) from day 0 to day 6. On day 6, 2U/mL of EPO was added, and the cells were cultured for an additional 5 days to expand erythroid progenitors. On day 11, cells were briefly serum starved and then stimulated with EPO and/or NmU in the absence or presence of a Jak-1/2 inhibitor. Activation of STAT5 and S6, a surrogate marker for PI3K activation, were assessed by phospho-flow in ERY3 (CD34−, CD71+, GlyA+) and ERY4 (CD34−, CD71dim, GlyA+) cells. As expected, EPO alone activated STAT5 and S6 in ERY3 cells only, and the presence of a Jak-1/2 inhibitor diminished STAT5 activation. Interestingly, STAT5 and S6 were activated by NmU peptide alone in ERY3 and ERY4. Surprisingly, in the presence of a Jak-1/2 inhibitor, NmU peptide, which binds to NMUR1 a G-protein coupled receptor, did not activate STAT5 or S6 in ERY3 or 4 cells, suggesting that NmU functions through a JAK kinase in erythroid cells. No additive or synergistic activation of STAT5 and S6 is observed in the presence of both EPO and NmU peptide when EPO was used at a dose of 2 U/mL. The mechanism whereby NmU activates a JAK dependent signaling pathway is under investigation. Preliminary evidence suggests that EPO induces the physical association of NMUR1 with EPO receptor (EPOR). Taken together, we propose that NmU is a neuropeptide expressed in bone marrow cells that cooperates to regulate erythroid expansion during early erythropoiesis through the activation of cytokine receptor like signaling pathways and perhaps through direct interaction with EPOR. NmU may be useful in the clinical management of anemia in patients unresponsive to EPO or other erythroid-stimulating agents. Disclosures: No relevant conflicts of interest to declare.

Description: The c-myb proto-oncogene has been implicated in leukemogenesis, but possible mechanisms remain ill defined. To gain further insight to this process, we used transcript profiling in K562 cells expressing a dominant-negative Myb (MERT) protein. A total of 105 potential Myb gene targets were identified. Neuromedin U (NmU), a peptide affecting calcium transport, underwent the greatest expression change (∼ 5-fold decrease). To verify a linkage between c-myb and NmU, their mRNA levels were quantitated using real-time polymerase chain reaction in primary acute myeloid leukemia (AML) and acute lymphoid leukemia (ALL), as well as normal hematopoietic cells. We found that c-myb was elevated in AML and ALL samples, but NmU expression was increased only in AML cells. Significantly, only AML cells expressed the cognate receptor of NmU, NMU1R, suggesting the presence of a novel autocrine loop. We examined this possibility in detail. Exogenous NmU “rescued” growth suppression in K562-MERT cells and stimulated the growth of primary AML cells. Short interfering RNA “knockdown” of NmU in K562 cells arrested cell growth. Exposing Indo-1–labeled K562 cells to NmU induced an intracellular Ca++ flux consistent with engagement of the NMU1R. Combined, these results suggest that NmU expression is related to Myb and that the NmU/NMU1R axis constitutes a previously unknown growth-promoting autocrine loop in myeloid leukemia cells.

Description: Abstract 3876 The proto-oncogene c-myb encodes the transcription factor c-Myb, which is predominantly expressed in immature hematopoietic cells where it plays an obligate role in definitive hematopoiesis. Given the critical functions of c-Myb in lineage commitment, proliferation, and differentiation, c-Myb regulatory factors are of great interest but remain incompletely defined. In recent years, c-Myb has been shown to regulate the expression of microRNA (miRNA) molecules in hematopoietic cells. MiRNA molecules are noncoding RNA molecules that are 21–23 nucleotides in length and function to hybridize to the 3′UTR region of its target mRNA to stimulate/repress translation or induce mRNA degradation. For example, in hematopoietic cells, miR-15a and c-Myb form an autoregulatory negative feedback loop in that over-expression of miR-15a in hematopoietic cells was determined to block erythroid and myeloid colony formation. In megakaryocytes, the hormone thrombopoietin induced miR-150 expression which subsequently functioned to degrade c-myb mRNA through direct interaction with c-myb's 3′-UTR. Our studies have focused on determining the physiologic function of the neuroendocrine Neuromedin U (NmU) during the early stages of erythropoiesis, because we recently determined that silencing NmU in primary human CD34+ cells impairs burst-forming units-erythroid and colony-forming unit-erythroid formation. In subsequent studies, we determined that c-Myb directly interacts with the NmU promoter at Myb Response Elements (MREs) distal to its transcription start site. Also, the expression profiles of NmU and c-myb are similar in CD34+ cells cultured under erythroid inducing conditions for 10 days, and silencing c-myb expression in hematopoietic cells inhibits NmU expression. To gain insight into the regulatory mechanism involved in NmU expression during the early stages of erythropoiesis, we hypothesized that miRNA molecules regulated by c-Myb would inhibit NmU expression through a negative feedback loop. To address this hypothesis, we first scanned the 3′-UTR of NmU and identified 24 different miRNA molecules predicted to interact with NmU's 3′-UTR. Second, we used luciferase reporter assays to determine which of the miRNA molecules interacted with NmU's 3′-UTR. Of the three miRNA molecules we tested, miR-101 directly interacted with NmU's 3′-UTR in a dose-dependent manner. Third, we determined the expression profile of miR-101 in primary CD34+ cells cultured under erythroid inducing conditions. The gene expression of miR-101 was inversely correlated with NmU and c-myb. Finally, because miR-101 contained 6 MREs, we determined the ability of c-Myb to directly interact with the promoter of miR-101 using chromatin immunoprecipitation (ChIP) assays. Using primers that flank the MREs proximal to miR-101's transcription start site, we observed a greater than 2-fold increase in the amplification of DNA recovered from ChIP assays completed with c-Myb antibody compared to ChIP assays completed with irrelevant antibody. Studies are underway to confirm by luciferase-reporter assays that c-Myb directly binds to and transactivates the miR-101 promoter. Collectively, these data identify a regulatory loop comprised of c-Myb, NmU, and miR-101 that could be of potential importance during human erythropoiesis. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 3160 Recently, we reported that the neuropeptide, neuromedin U (NmU), functions as a novel extracellular cofactor with erythropoietin (EPO) to promote the expansion of early human erythroblasts. Because the expression of NmU is important during the early stages of erythropoiesis, we aimed to understand its temporal regulation during erythroid development. Although we have demonstrated that NmU is a target of the erythroid transcriptional regulator, c-Myb, our understanding of NmU regulation is incomplete. We hypothesized that microRNA (miRNA) molecules function to regulate NmU expression at the post-transcription level during erythropoiesis. Upon sequence analysis of the 3'-UTR of NmU using microCosm in the miRBase Targets database, 20 different miRNA molecules were predicted to interact with NmU's 3'-UTR. Among the 20 different miRNA molecules predicted to interact with NmU's 3'UTR, miR-101 was of interest, because in an independent study, its expression was elevated as measured by microarray analyses from primary human CD34+ cells cultured under erythroid inducing conditions. To determine the ability of miR-101 to directly interact with the 3'UTR of NmU, we used luciferase reporter assays. In a dose-dependent manner, miR-101 directly interacted with NmU's 3'-UTR. Also, 24-hours post-nucleofection of miR-101 into K562 cells, a hematopoietic cell line, the expression of NmU was decreased compared to control. Over-expression of miR-101 in primary human CD34+ cells decreased the growth of colony-forming unit-erythroid (CFU-E) ∼50% compared to control cells. In the presence of exogenously added NmU peptide, CFU-E growth from CD34+ cells over-expressing miR-101 was rescued to the level observed with control miRNA treated cells. To further determine the relationship between NmU, EPO, and miR-101, we cultured primary human CD34+ cells using a 2-phase liquid culture condition to induce erythroid development. During the first phase (days 0–6), the cells were cultured with IL-3, IL-6, and stem cell factor (SCF). The second phase of the erythroid inducing culture conditions began on day 6 when EPO was added to the culture. Erythroid differentiation was monitored using flow cytometry and fluorescent conjugated antibodies against CD34, transferrin receptor (CD71), and glycophorin A (GlyA). In parallel, primary cells were collected at regular intervals during culture to measure the expression of NmU mRNA and miR-101 by real time PCR (RT-PCR). Under our erythroid inducing culture conditions, NmU expression peaked between days 4 and 6 (before adding EPO) and between days 10 to 12. Also, between days 10 to 12 of culture in erythroid inducing conditions, we observed a dramatic increase in cell proliferation. Between days 13 to 15, cell proliferation reached a plateau, and the expression of miR-101 peaked. Erythroid progenitors purified from cord blood mononuclear cells by cell sorting revealed that NmU expression peaked in CD34-, CD71+, GlyA- (ERY2) cells, which is in good agreement with an independent microarray study, and miR-101 expression was not detected. By contrast, in CD34-, CD71lo, GlyA+ (ERY4) cells, miR-101 expression peaked while NmU expression decreased to the level observed in CD34-, CD71-, GlyA- cells. Combined, these data identify NmU as a novel miR-101 target and indicate that miR-101 regulates the temporal expression of NmU during the later stages of erythropoiesis. We hypothesize that the miR-101/NmU axis is a critical modulator of erythroid cell expansion that augments the effects of erythropoietin. Disclosures: Carroll: Glaxo Smith Kline, Inc.: Research Funding; Sanofi Aventis Corporation: Research Funding; TetraLogic Pharmaceuticals: Research Funding; Agios Pharmaceuticals: Research Funding.

Description: Abstract 673 Inhibition of lymphocyte trafficking early after allogeneic stem cell transplantation (SCT) could limit T cell interactions with antigen-presenting cells and migration to target tissues. This represents a novel strategy to prevent GvHD without interfering with GvL activity. CCR5 is a chemokine receptor expressed on effector T-cells and immature dendritic cells and binds 3 ligands - CCL3, CCL4 and RANTES (CCL5). Accumulating evidence from animal models and clinical observations implicates CCR5 as pivotal in the pathogenesis of GvHD. Genomic analyses suggest that the same CCR5 polymorphisms that confer resistance to HIV infection also correlate with a lower susceptibility to acute GvHD. Maraviroc (MVC; Selzentry®, Pfizer) is the first oral CCR5 antagonist in clinical use. We hypothesized that modulating T-cell trafficking early after allogeneic SCT via CCR5 blockade would limit GvHD. We therefore performed preclinical and clinical testing of MVC as GvHD prophylaxis. Our goals were to 1) determine in vitro activity of MVC on chemotaxis, 2) determine the feasibility, safety and appropriate dose of MVC as part of GvHD prophylaxis, and 3) demonstrate biological activity of MVC through immune pharmacodynamic assays. In vitro, MVC fully inhibited CCR5 internalization by CCL3 and RANTES even at concentrations as low as 1 μM. Using RANTES as a chemotactic trigger, MVC caused dose-dependent inhibition of lymphocyte chemotaxis by up to 53% at MVC 1mM. To address concerns that MVC might impair hematopoiesis, we demonstrated that CCR5 was not expressed on the surface of bone marrow- and peripheral blood-derived CD34+ cells. Moreover, when CD34+ cells were plated in methylcellulose, formation of CFU-GEMM and CFU-GM was not affected by the presence of MVC 1μM; CFU-E and BFU-E were slightly decreased compared to controls. Based on these and other data, we enrolled 19 pts in a phase I/II study of reduced intensity conditioned allogeneic SCT with MVC GvHD prophylaxis. Pts received fludarabine 120mg/m2 and IV busulfan 6.4 mg/kg followed by peripheral blood stem cells from matched related (n=6), matched unrelated (n=10) and 1-antigen mismatched unrelated (n=3) donors. In addition to standard GvHD prophylaxis with tacrolimus and methotrexate, MVC at escalating dose levels was given from day -2 to +30. Median age was 63 (range 21–74). Indications for SCT were AML/MDS (9), NHL (4), myelofibrosis (2), CLL, aplastic anemia, Hodgkin lymphoma and myeloma (1 each). Pharmacokinetic analysis on 6 pts at each dose revealed that the 300 mg and 150 mg bid dose levels resulted in mean Cavg of 536 and 118 ng/ml, respectively. 3/6 patients at 150mg did not reach the targeted minimum Cavg (100 ng/ml), while the 300mg dose level resulted in adequate Cavg in 6/6 patients and was used as the phase II dose. MVC was well tolerated; 3 pts did not complete the entire course because of transient LFT abnormalities (1) or mucositis (2). The median time to ANC〉500/μL was 15 d (range 10–21) and to platelets〉20k/μL was 13 d (range 11–24) with no graft rejections. The median donor chimerism at day 100 was 97% (range 83–100%). A day 100-landmark analysis in evaluable pts demonstrated that the cumulative incidence of acute GvHD grade 2–4 was 27% (grade 3–4; 9%) in this high-risk population. Importantly, by day 100 all cases of acute GvHD involved only the skin without liver or intestinal involvement. At a median follow up of 186 days, 3/19 patients relapsed (2 AML, 1 NHL) and 6/19 patients died (3 disease-related, 1 neutropenic sepsis, 1 SOS, 1 unrelated). There were no GvHD-related deaths. To explore potential mechanisms, we tested the capacity of patient serum to inhibit CCR5 internalization and chemotaxis. Patient serum from multiple time points (trough, 1, 2, 3, 4, 6 hr post dose) effectively prevented internalization of CCR5 by RANTES. In addition, in vitro chemotaxis of normal donor T-cells in response to RANTES was significantly impaired in the presence of patient serum from day 0 (on MVC) as compared to day 60 (off MVC). In summary, inhibition of lymphocyte trafficking to peripheral tissues represents a novel strategy to modulate and possibly reduce acute GvHD in allogeneic SCT. MVC at 300mg bid was well tolerated and biologically active in pharmacodynamic assays. Patients receiving MVC exhibited limited GvHD by day 100 without excessive relapses. The phase II portion of the trial is ongoing. Disclosures: Off Label Use: Off label use of maraviroc (Selzentry) will be discussed. Frey:Pfizer, Inc.: Speakers Bureau. Vonderheide:Pfizer, Inc.: Research Funding. Porter:Pfizer, Inc.: Research Funding.

Description: The c-myb proto-oncogene encodes a transcription factor, Myb, which is essential for the growth and proliferation of normal and malignant hematopoietic cells. We, and others, have previously shown that malignant hematopoietic cells are much more dependent on c-myb function than are normal hematopoeitic cells, and that transient disruption of c-myb expression causes malignant cells to undergo apoptosis while normal cells are relatively spared. Based on these findings, we hypothesized that c-myb regulates a unique set of genes in leukemic cells that are required for cell survival. To identify Myb gene targets, we performed a transcriptome analysis using human myeloid leukemic cells engineered to express a conditionally active dominant negative Myb (MERT). Analysis of the microarray data revealed that when Myb activity was inhibited by tamoxifen in MERT cells, CDC7, an intra-S phase regulator, decreased in expression 2.8-fold compared to untreated control cells. To verify this, we utilized real-time PCR to quantitate the expression of CDC7, and found that it decreased 5-fold in tamoxifen treated MERT cells relative to control cells. In aggregate, the microarray and real-time PCR data suggested that Myb directly regulates CDC7 gene expression in hematopoietic cells. To address this question in the absence of a formally defined human CDC7 promoter, we examined the DNA sequence upstream of the predicted transcription start site (as noted in Genbank accession # AY585721) for potential Myb transcription factor binding motifs. After scanning the DNA sequence (~3kb) upstream of the predicted transcription start site, nine potential Myb response elements (MREs) were identified. The CDC7 sequences from mouse, chimp, and yeast were also analyzed for MREs and compared to those present in the putative human CDC7 promoter to identify conserved MREs. Using this strategy, we also identified potential AML1, PU.1, CBP, STAT3, and STAT5 binding motifs within the human CDC7 promoter region. To determine if any of the potential Myb binding sites with the CDC7 promoter were actually utilized in vivo, we carried out chromatin immunoprecopitation (ChIP) assays. When the chromatin from untreated MERT cells was immunoprecipitated with anti-c-Myb, we observed one PCR product using a primer pair that flanked each conserved MRE. These same results were obtained in our positive control ChIP experiment in which the chromatin was immunoprecipitated with anti-acetyl histone 4. When Myb transactivation activity was inhibited in tamoxifen treated MERT cells, no PCR product was detected following chromatin immunoprecipitation with the anti-Myb antibody suggesting that the ChIP binding results were not due to artifact. We have just completed a primer extension assay with a Fam-labeled primer that flanked the predicted CDC7 promoter region and will use the resulting sequence data to identify the actual CDC7 transcriptional start site. We will also shortly complete identification of functional regions within the human CDC7 promoter through use of Luciferase reporter assays. Investigation of the transcriptional regulation of CDC7 in hematopoietic cells may yield new clues to Myb’s role in leukemogenesis.