ALBERT

Description: Abstract 3626 One of the major pathways involved in the angiogenic process is the VEGF/VEGFR signaling pathway. VEGFA was found to be an independent prognostic factor for therapeutic outcome in AML (de Bont et al, 2002;Aguayo et al, 1999). Moreover, it has been shown that after chemotherapy the expression level of VEGFR2 is restored to normal in the bone marrow of AML patients in complete remission (Padro et al, 2002). Both VEGF and its receptors Flt-1(VEGFR1) and KDR (VEGFR2) are expressed by leukemic blasts which results in both a paracrine and an autocrine VEGF/VEGFR2 pathway in AML (Dias et al, 2001). In response to VEGF stimulation, VEGFR2 has been shown to transmit intracellular signals leading to activation of multiple downstream signaling pathways including the mitogen-activated protein (MAP) kinases, JAK)STAT, and Phosphoinositide 3-kinases (PI3 kinases). In AML it has been described that simultaneous activation of these pathways results in a poor prognosis (Kornblau et al, 2006). Chromosomal translocations involving the Mixed Lineage Leukemia (MLL) gene at locus 11q23 are associated with a poor outcome. These MLL translocations represent 15–20% of pediatric acute myeloid leukemia's. In this study we show that VEGFR2 expression is significantly higher on blasts of 11q23 translocated acute myeloid leukemia's compared to blasts of AML patients with another karyotype. In addition, using peptide arrays and proteome profiler arrays we have found that the VEGF receptor signaling pathway and part of the MAP kinase signaling pathway are active and or phosphorylated in blasts of 11q23 translocated patients (Fig. 1). Proteins from these active pathway are potential targets in the treatment of pediatric 11q23 translocated AML. To further investigate these potential targets we have performed ex vivo drug target assays using an inhibitor and an antibody against two key proteins from the major signal transduction pathways, VEGFR2 antibody (IMC1121b) and MEK inhibitor (U0126). No effect of VEGFR2 inhibition was measured in two AML cell lines (HL-60 and THP-1). Whereas, MEK inhibition had an effect on both cell lines, LC50 8.7 μM and 7.5 μM for HL-60 and THP-1 respectively. Cell survival of the THP-1 cell line (containing a 11q23 translocation) was substantially inhibited after addition of a combination therapy with an inhibitor against MEK and a antibody against VEGFR2 (LC50 2.9 μM). In contrast, the VEGFR2 antibody in combination with the MEK inhibitor had only a marginal effect on the HL-60 cell line (LC50 7.7 μM). Future experiments to investigate the effect of the combination therapy on primary patient material are underway. Overall these data show for the first time that targeting multiple active pathways in pediatric 11q23 AML could result in promising opportunities concerning future therapy options. Figure 1. Provisional scheme of active kinases and phosphorylated proteins in AML. Green-white, active kinase; Yellow, phosphorylated protein; Green-Yellow, active kinase and phosphorylated protein; thick lining, bold receptors and, bold cell cycle proteins, kinase activity higher in AML cells compared to normal bone marrow. Red arrows, site of inhibition. Figure 1. Provisional scheme of active kinases and phosphorylated proteins in AML. Green-white, active kinase; Yellow, phosphorylated protein; Green-Yellow, active kinase and phosphorylated protein; thick lining, bold receptors and, bold cell cycle proteins, kinase activity higher in AML cells compared to normal bone marrow. Red arrows, site of inhibition. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 2338 Cancer arises when somatic cells are able to escape the restraints that normally withhold them from unlimited expansion. Cancer progression is thought to be the net result of signaling through various protein-kinase mediated networks driving cell proliferation and survival. The kinome networks can be affected by numerous factors; including acquired or selected mutations as well as environmental cross talk. Additionally, the loss of phosphatases could be a causative factor for activation of multiple tyrosine kinases as well (Cell. 2011 Mar 4;144(5):703–18). Deregulated kinase activity is frequently observed in leukemia, leading to induced proliferation, migration, survival, and chemotherapy resistance of leukemic cells. (Leuk Lymphoma. 2011 Jan;52(1):122–30, Leukemia. 2005 Apr;19(4):586–94, Exp Hematol. 2005 Jun;33(6):660–70) However, single kinase-targeted cancer therapies can default when cancer cells bypass through alternative routes, facilitating therapeutic resistance. In order to circumvent the constraints given by an inihibitor, we need to monitor kinome reprogramming upon mono-treatments to develop the most successful combination therapy approach for disease specific subgroups, as poor prognostic MLL-rearranged AML. Rational designs of kinase inhibitor or RTK antibody combinations require a high-throughput measurement of kinome and proteome activity signatures within this patient subgroup. Figure 1 outlines our approach to explore the intracellular signaling networks and study the dynamic changes resulting in reprogramming of the kinome network, with the goal to define combinational therapeutic strategies. In this study, we succeeded using combined high-throughput approaches for kinomic and proteomic profiling to identify specific aberrant kinase signatures in MLL-rearranged AML as compared to NBM (Fig. 1B/C). The altered activated kinase signatures of a comprehensive set of MLL-rearranged AML patient samples resulted in a detailed map of the overall kinase activity and phosphorylation of signal transduction molecules, which allowed the selection of possible druggable targets i.e. MEK, JNK, and CREB (Fig. 1B/C). Pharmacological MEK inhibition in primary MLL-rearranged AML demonstrated to be most successful in reducing the AML cell survival, without showing cytotoxicity in NBM (mean MEK inhibitor IC50 of 3.5μM +/- 0.7μM in primary MLL-rearranged AML versus a mean IC50 〉50μM in NBM), whereas for CREB and JNK inhibitors MLL-rearranged AML cells were equally affected as NBM cells. Dynamic kinome reprogramming of signaling networks in response to MEK therapy did occur, by inducing the activation of RTKs to bypass the initial MEK inhibitory effects in a MLL-rearranged AML cell line. The dynamic escape mechanism allowed us to predict and test the efficacy of novel combination strategies. Combined MEK and VEGFR-2 inhibition demonstrated to induce cell death sufficiently in MLL-rearranged AML (Fig. 1D). This advantageous strategy allows rational design of successful and selective combination therapies for specific target inhibitors. Figure 1. Kinomics, proteomics, and signaling dynamics for novel combination therapies in MLL-rearranged AML (A) Overall study design for novel combination therapies (B) Constructed kinome trees showing the overall kinase derived peptide activity in 15 individual primary MLL-rearranged AML samples versus 5 NBM samples to identify the distribution and the overlap of kinase activity. Combined kinome trees revealed induced JNK annotated peptide activity in MLL-rearranged AML. (C) Representative protein kinase phosphorylation arrays of primary MLL-rearranged AML (n=9 in total) versus NBM (n=4 in total). Phosphoproteomic analysis identified induced phosphorylation levels of MEK and CREB protein kinases in MLL-rearranged AML. (D) Evaluation of AML cell survival of an anticipated successful combined therapeutic strategy in MLL-rearranged AML, showing the synergistic beneficial effect of combined MEK and VEGFR-2 inhibition in MLL-rearranged AML cell line THP-1. Figure 1. Kinomics, proteomics, and signaling dynamics for novel combination therapies in MLL-rearranged AML . / (A) Overall study design for novel combination therapies (B) Constructed kinome trees showing the overall kinase derived peptide activity in 15 individual primary MLL-rearranged AML samples versus 5 NBM samples to identify the distribution and the overlap of kinase activity. Combined kinome trees revealed induced JNK annotated peptide activity in MLL-rearranged AML. (C) Representative protein kinase phosphorylation arrays of primary MLL-rearranged AML (n=9 in total) versus NBM (n=4 in total). Phosphoproteomic analysis identified induced phosphorylation levels of MEK and CREB protein kinases in MLL-rearranged AML. (D) Evaluation of AML cell survival of an anticipated successful combined therapeutic strategy in MLL-rearranged AML, showing the synergistic beneficial effect of combined MEK and VEGFR-2 inhibition in MLL-rearranged AML cell line THP-1. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 5235 Aberrant Ephrin signaling has been shown to be an important pathway that contributes to the pathogenesis of many solid tumors (Surawska et al. Cytokine & Growth factor reviews 2004). Deregulated ephrin receptor (Eph) and ligand (Efn) expression is often associated with poor prognosis in solid tumors. Ephrin receptor and ligand overexpression can result in tumorigenesis through induced tumor growth, tumor cell survival, angiogenesis and metastasis (Surawska et al. Cytokine & Growth Factor Reviews 2004; Campbell et al. Curr. Isues Mol. Biol. 2008; Chen et al. Cancer Research 2008). In normal cells Eph receptors and ligands play key roles in vascular patterning, where they function in endothelial cell migration, and proliferation (Adams et al, Genes Dev. 1999; Zhang et al., Blood 2001). Thus far particularly EphB4 receptor and ephrin-B2 ligand have been implicated in the process of normal angiogenesis. In acute myeloid leukemia (AML) patients it was found that bone marrow biopsies at diagnosis exhibited enhanced microvessel density (MVD) (de Bont ES et al., BJH 2001; Byrd JC et al., Blood 2002; Padro et al., Blood 2000). Normal hematopoietic stem cells (HSCs) express the following mRNA transcripts ephrin receptors EphA1, EphA2, EphB2, and EphB4 and ephrin ligands EfnA3, EfnA4, and EfnB2. Moreover, overexpression of EphB4 receptor in HSCs (from cord blood) resulted in enhanced differentiation towards megakaryocytes (Wang et al. Blood 2002). In AML cell lines there is a common co-expression on protein level observed between EphB4 receptor and ephrin-B2 ligand. Recently, an aberrant DNA methylation of ephrin receptors and ligands was described in acute lymphocytic and myelocytic leukemia cell lines (Kuang et al. Blood 2010). In addition, restoration of EphB4 expression in an acute lymphoid leukemia cell line resulted in reduced proliferation and apoptotic cell death. These data suggests that the ephrin signaling pathway might play an important role in leukemia. In a previous study we have found high kinase activity of EphB receptors and high phosphorylation levels of EphB receptors in AML samples, as measured using kinase arrays and proteome profiler arrays. In this study, we have found extensive membrane expression of EphB1 on AML cell lines and primary AML blasts. To identify the role of Ephrin signaling in AML, two AML cell lines THP-1 and HL60 with an EphB1 membrane expressing cell percentage of 70% and 20% respectively were chosen for stimulation with Ephrin-B1 ligand. Treatment of these cell lines with Ephrin-B1 ligand resulted in a decreased proliferation 30% in THP-1 cells versus 22% in HL60 cells and increased apoptosis 23% in THP-1 cells and 4% in HL60 cells. Of note, the most prominent effect of Ephrin-B1 stimulation was found in THP-1 cells, this cell line contained a higher percentage of EphB1 membrane expressing cells. We further investigated the mechanism through which EphB1 reduces leukemic cell growth and induces leukemic cell death in THP-1 cells. Westernblot analysis of cell cycle regulators showed that expression of the anti-apoptotic protein BCL2 is reduced upon Ephrin-B1 ligand stimulation and the expression of the pro-apoptotic protein BAX is induced. In addition, mRNA expression of the cell cycle inhibitor of cell cycle progression p21 was found to be 2,5 fold upregulated in ephrin-B1 ligand treated cells compared to untreated control cells. MGG stainings of Ephrin-B1 treated cells revealed multiple cells with two nuclei in both THP-1 and HL60 cells. These results indicate that a high percentage of AML cells express EphB1 receptor on the membrane and that stimulation of these cells with Ephrin-B1 ligand results in reduced leukemic growth and increased cell death. EphrinB1 activation in AML deserves further investigation considering EphB1 as a putative new treatment option for AML patients. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 1368 New treatment options are necessary to improve survival rates for patients with Acute Lymphoblastic Leukemia (ALL), especially for patients with unfavorable prognostic predictors. As a new therapeutic approach specific protein kinase inhibitors are being developed that can down-regulate vital signaling pathways in leukemic blasts (McCubrey et al, 2008). The main goal of the present study is to obtain a better understanding of the kinase signaling pathways active in ALL cells and to identify potential targets for therapeutic intervention, To identify active signaling pathways in ALL we have used kinase activity arrays containing 1024 peptides representing all major signaling pathways and human proteome profiler arrays containing 46 phospho-antibodies on lysates of primary ALL blasts. In 20 patient samples a total of 10.6% 109(1024) peptides were found to be phosphorylated in 90% of the samples. About 46% 50(109). Activities for kinases including PKC, PKA, Akt, CAMK2, CDC2, CDK2, ERK, GSK3beta, JAK and MAPK were detected in these lysates. The human proteome profiler array demonstrated high levels of protein phosphorylation of CREB and RSK. We constructed a provisional signal transduction scheme of active kinases and phosphorylated proteins in ALL cells (Fig. 1A). Consistent with earlier reports, we identified a prominent role for the Raf/MEK/ERK and the PI3K/Akt/mTOR pathways in these ALL cells. Based on this provisional signal transduction scheme we composed a list of possible new druggable targets. Two proteins were selected for further investigation, CREB and RSK. Inhibition of RSK by the p90 RSK inhibitor BI-D1870 had no effect on cell viability as measured with WST-1 cell viability assay in ALL cell lines. Interestingly, inhibition of CREB by the CREB inhibitor KG-501 showed a dose- and time-dependent decrease in cell viability in all cell lines tested (LC50 values after 24h: Jurkat: 18.55 mM, Molt 4: 13.02 mM, RCH-ACV: 38.11 mM, and RS4;11 45.36 mM (Fig. 1B). LC50 values after 48h: Jurkat: 7.36 mM, Molt 4: 6.53 mM, RCH-ACV: 31.73 mM, RS4;11 36.66 mM (Fig. 1C)). In addition, apoptosis measured by AnnexinV/ PI staining showed an increased percentage of apoptotic cells in a dose- and time-dependent manner in all cell lines upon treatment with the CREB inhibitor (apoptosis after 24h: Jurkat 35.83% to 79.7%, Molt 4: 12.19% to 48.5%), RCH-ACV 11.30% to 45.9%, and RS4;11 9.84% to 19.16. Apoptosis after 48h: Jurkat 53.40% to 86.4%, Molt 4: 27.70% to 92.9%, RCH-ACV 14.07% to 63.32%, and RS4;11 7.11% to 20.75%) (Fig. 1D). To investigate the downstream effect of CREB inhibition we measured the mRNA expression of a know CREB target gene: BCL-2. Upon inhibition of CREB (50 mM KG-501) mRNA levels of BCL-2 were found to be significantly decreased compared to vehicle treated cells. In conclusion we have identified the transcription factor CREB in vitro as a potential druggable target for ALL. It is known that CREB plays an important role as a downstream target of hematopoietic growth factor signaling in hematopoiesis (Cheng et al, 2008). Based on these results, we propose CREB as a promising potential druggable target in ALL. Figure 1. (A) Provisional signal transduction scheme of active kinases and phosphorylated proteins in ALL. Green: active kinase; Yellow: phosphorylated protein; Green-Yellow: active kinase and phosphorylated protein. (B) Cell viability percentages plotted against concentration of KG-501 (mM) after 24h. (C) Cell viability percentages plotted against concentration of KG-501 (mM) after 48h. (D) Representative flow cytometric dot-plots of AnnexinV/ PI flow cytometry, inhibition of CREB induced a dose- and time-dependent apoptosis in the Molt 4 cell line. Disclosures: No relevant conflicts of interest to declare.

Description: Acute myeloid leukemia (AML) is a heterogeneous disease, characterized by a multitude of genetic events. Activating mutations in receptor tyrosine kinases have been identified in 50% of AML primary blasts, and deregulation of one or more signal transduction pathways including the JAK/STAT, RAS/Raf/MEK/ERK, and PI3K/AKT pathways is common (Kornblau et al., 2006). High-throughput procedures which generate comprehensive descriptions of cellular signaling without a priori assumptions in each sample, would enable us to directly assess a broader range of targets for future treatment strategies. In the present study we identified kinase activity profiles in 22 pediatric leukemia samples (6 acute myeloid leukemia, 9 acute lymphoid leukemia, 3 Philadelphia positive acute lymphoid leukemia and 5 chronic myeloid leukemia) and in various normal tissues such as colon and kidney. Peptide phosphorylation profiles were determined using the Pamchip® tyrosine kinase micro array system (Pamgene International B.V., ‘s Hertogenbosch, the Netherlands). This array consists of 144 peptides representing key phosphorylation sites of proteins known to be involved in signal transduction processes. We generated a comprehensive description of the phosphotyrosine proteome of all patient samples. Within the variety of profiles in the various leukemia samples, peptide corresponding with phosphorylation consensus sequences for MAP kinases showed remarkable high levels of phosphorylation, whereas in various normal tissue types substantial lower activity was observed on these substrates. These results imply activated MAPK signaling to be a prominent characteristic of leukemia as already described (Towatari et al., 1997). A differential phosphorylation pattern was seen for the RON peptide (macrophage stimulating 1 receptor (c-met-related tyrosine kinase)), only phosphorylated in one of the AML samples, 4 out of 5 CML samples and one Ph+ ALL sample. According to Phospho-ELM and the literature (Follenzi et al., 2000) this tyrosine kinase residue can be phosphorylated via autophosphorylation by RON kinase or transphosphorylation by MET kinase (met proto-oncogene (hepatocyte growth factor receptor)).To verify the role of MET kinase in AML, a cell survival assay was performed with the selective MET kinase inhibitor PHA 665752. Dose dependent decrease in cell survival was observed in three primary AML samples tested with LC50 concentrations ranging from 2 μM to 5 μM. In conclusion, this study describes a new high throughput technique to generate tyrosine phospho-proteomes or even kinome profiles of various leukemic samples in the future. With this technique we found MET to be a new therapeutic target in AML, and it demonstrates the importance of MAP kinase signaling in various leukemic samples. In the era of a rapidly increasing number of small molecule inhibitors this technique will enable us to rapidly identify new potential targets in different kinds of leukemia.