|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200397Ka 4500
controlfield tag 001 002029118
007 cr mnu|||uuuuu
008 090916s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002892
Bewry, Nadine N.
STAT3 contributes to resistance towards BCR-ABL inhibitors in a bone marrow microenvironment model of drug resistance in chronic myeloid leukemia cells
h [electronic resource] /
by Nadine N. Bewry.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 149 pages.
Dissertation (Ph.D.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: Imatinib mesylate (imatinib) represents a potent molecularly targeted therapy against the oncogenic tyrosine kinase, BCR-ABL. Although imatinib has shown considerable efficacy against chronic myeloid leukemia (CML), displaying high rates of complete hematological and complete cytogenetic responses, treatment with imatinib is not curative and overtime advanced-stage CML patients often become refractory to further treatment. Acquired resistance to imatinib has been associated with mutations within the kinase domain of BCR-ABL, BCR-ABL gene amplification, leukemic stem cell quiescence as well as over-expression of the multidrug resistance (MDR1) gene. However, in vitro resistance models often fail to consider the role of the tumor microenvironment in the emergence of the imatinib-resistant phenotype. The bone marrow is the predominant microenvironment of CML and is a rich source of both soluble factors and extracellular matrixes, which may influence drug response.To address the influence of the bone marrow microenvironment on imatinib sensitivity, we utilized an in vitro co-culture bone marrow stroma model. Using a transwell system, we demonstrated that soluble factors secreted by the human bone marrow stroma cell line, HS-5, were sufficient to cause resistance to apoptosis induced by imatinib in CML cell lines. We subsequently determined that culturing CML cells in HS-5-derived conditioned media (CM) inhibits apoptosis induced by imatinib and other second generation BCR-ABL inhibitors. These data suggest that more potent BCR-ABL inhibitors will not overcome resistance associated with the bone marrow microenvironment. Additionally, we determined that CM increases the clonogenic survival of CML cells following treatment with imatinib. HS-5 cells are reported to express several cytokines and growth factors known to activate signal transducer and activator of transcription 3 (STAT3).Given its crucial role in the survival of hematopoietic cells, we asked whether, 1) CM derived from HS-5 cells can activate STAT3 in CML cells and 2) does activation of STAT3 confer resistance to BCR-ABL inhibitors. We demonstrated that exposure of the CML cell lines, K562 and KU812, to CM caused an increase in phospho-Tyr STAT3, while no increases in phospho-Tyr STAT5 were noted. Moreover, resistance was associated with increased levels of the STAT3 target genes, Bcl-xl, Mcl-1 and survivin. Furthermore, reducing STAT3 levels with siRNA sensitized K562 cells cultured in CM to imatinib-induced cell death (p<0.05, Student's t-test). Importantly, STAT3 dependency was specific for cells grown in CM, as reducing STAT3 levels in regular growth conditions had no effect on imatinib sensitivity.Together, these data support a novel mechanism of BCR-ABL-independent imatinib resistance and provide preclinical rationale for using STAT3 inhibitors to increase the efficacy of imatinib within the context of the bone marrow microenvironment.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Lori Hazlehurst, Ph.D.
x Molecular Medicine
t USF Electronic Theses and Dissertations.
STAT3 Contributes to Resistance Towards BCR-ABL Inhibit ors in a Bone Marrow Microenvironment Model of Drug Resistance in Ch ronic Myeloid Leukemia Cells by Nadine N. Bewry A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Molecular Medicine College of Medicine University of South Florida Major Professor: Lori Hazlehurst, Ph.D. George Blanck, Ph.D. William Dalton, Ph.D. Andreas Seyfang, Ph.D. Larry Solomonson, Ph.D. Date of Approval: December 2, 2008 Keywords: imatinib, nilotinib, dasatinib, STAT5, hemat opoiesis, cytokines Copyright 2009 Nadine N. Bewry
i TABLE OF CONTENTS List of Tables...................................... ................................................... ...............vi List of Figures..................................... ................................................... ..............vii List of Abbreviations ............................... ................................................... ..........x Abstract ............................................ ................................................... ..........xiii Chapter I: Introduction............................. ................................................... ........1 Hematopoiesis...................................... ................................................... ..1 Types of Leukemia..................................... .....................................3 Normal vs. Leukemic Hematopoietic Stem Cells............ .................4 Chapter II: Chronic Myeloid Leukemia and the Philadel phia Chromosome ......................................... ................................................... ..7 The BCR-ABL Protein ............................... ..............................................12 Structure of c-BCR Protein........................... .................................13 Structure of c-ABL Protein .......................... .................................14 One BCR-ABL Oncogene, 3 BCR-ABL Proteins Isoforms................ .......16 Chapter III: BCR-ABL Pathogenesis................... ...............................................19 The Role of Modular Domains........................ .........................................19 Activation of Multiple Signal Transduction Pathways.... ...........................21 Constitutive Mitogenic Activation: The Ras/MEK Pathway... .........22
ii Inhibition of Apoptosis............................ .......................................25 The PI3/AKT Pathway................................ ........................25 The JAK/STAT Pathway................................ .....................28 Decreased Adhesion to Bone Marrow Stroma and Extracellular Matrices............................... ...............................31 Chapter IV: Treatment for Chronic Myeloid Leukemia... .....................................34 Before Signal Transduction Inhibitors................ ......................................34 Busulfan............................................ ............................................34 Hydroxyurea.......................................... ........................................35 Recombinant Interferon-Alpha........................ ..............................35 Allogeneic Stem Cell Transplantation................. ..........................36 Rationally Designed BCR-ABL Signal Transduction Inhib itors................37 Imatinib Mesylate................................... .......................................37 Imatinib Resistant Chronic Myeloid Leukemia............ ...................41 Point Mutations within the BCR-ABL Kinase Domain............................................. .......................42 BCR-ABL Gene Amplification................................. .44 Overexpression of Drug Transporters................... ...45 Nilotinib ......................................... ...............................................46 Dasatinib........................................... ............................................47 Chapter V: Minimal Residual Disease Chronic Myeloid Leu kemia....................49 Failure to Kill Leukemic Stem Cells.................. ........................................49
iii Pre-existing Mutations within the BCR-ABL Kinase Doma in....................50 Epigenetic Contributions............................ ..............................................51 Tumor Microenvironment ............................. ...........................................52 Chapter VI: The Role Signal Transducers and Activators of Transcription (STATs) in Chronic Myeloid Leukemia Oncogenesis and Drug Resistance............................................ ................................................... ..54 The Role of STAT3.................................. ................................................56 The Role of STAT5.................................. ................................................56 Chapter VII: Objectives.............................. ................................................... ....58 Chapter VIII: Materials and Methods................. .................................................60 Cell Cultures...................................... ................................................... ...60 Generation of Conditioned Media (CM)............... ....................................60 Drugs and Reagents.................................. ..............................................61 Preparation of Lysates and Western Blotting............ ...............................61 Apoptosis Assay....................................... ...............................................61 siRNA Transfection.................................... ..............................................62 Bromodeoxyuridine (BrdU) Antibody Staining........... ...............................62 Clonogenic Assay...................................... ..............................................62 Preparation of Nuclear Extracts..................... ..........................................63 Electrophoretic Mobility Shift Assay (EMSA).............. .............................63 MTT Assay............................................ ................................................... 63 Statistical Analysis.................................. .................................................64
iv Chapter IX: Results................................. ................................................... .........65 Co-culture Bone Marrow Stromal Model Protects K562 CML Cells from Imatinib-Induced Apoptosis..................... ........................................65 Characterizing Conditioned Media.................... .......................................67 Collecting Conditioned Media Beyond 3 Hours does not Provide Greater Protection against Imatinib-Induced Cel l Death.............................................. .........................................67 Conditioned Media Stored for up to One Week Still Pr ovided Protection against Imatinib-Induced Cell Death........ ...............68 Serum is not Required for Production of the Protective Soluble Factor(s) Found in HS-5-Derived Conditioned Media.............................................. .........................................70 Conditioned Media does not Convey a Growth Advantage to K562 CML Cells..................................... ..................................71 Heat-Inactivated Conditioned Media does not Protect K56 2 Cells against Imatinib-Induced Cell Death ............ ..................74 HS-5-Derived Conditioned Media Protects K562 and KU812 CML Cell Lines from Death Induced by Imatinib Mesylate..... ..........................76 Conditioned Media from Non-Stromal Cell Lines do not Protect K562 CML Cells from Death Induced by Imatinib Mesylate ....................81 Conditioned Media Protects K562 CML Cells from Death I nduced by 2nd Generation BCR-ABL Inhibitors, Nilotinib and D asatinib..............82
v Conditioned Media Activates STAT3 in K562 and KU812 C ML Cell Lines.............................................. ................................................... .......84 STAT3 Activation in CML Cells Is BCR-ABL-Independent... ....................86 STAT5 Activation is BCR-ABL-Dependent ................ .............................88 STAT3 Activation is not SRC-Dependent................. ...............................89 Protein Expression Levels of STAT3 Downstream Targets are Increased in K562 Cells Cultured in Conditioned Media. .........................90 Reducing STAT3 Levels with siRNA Increases Sensitivity to Imatinib Mesylate in Conditioned Media.............. ....................................93 Addition of GM-CSF, IL-6 and VEGF to Regular Media Induces the Imatinib-Resistant Phenotype Associated with Conditione d Media.........99 Chapter X: Discussion and Future Direction............. .......................................103 Literature Cited................................... ................................................... ...........122 Presentation of Studies............................. ................................................... ....147 About The Author................................... .................................................En d Page
vi LIST OF TABLES Table 1 Types of Leukemia .................................... .....................................4 Table 2 Three Phases of Chronic Myeloid Leukemia................ .................11 Table 3 Some of BCR-ABLs Substrates.......................... ..........................12 Table 4 The BCL-2 Family of Proteins........................ ...............................30 Table 5 Definition of the Types of Responses to Chronic Myeloid Leukemia Treatment.................................. ...................................37 Table 6 Profile of BCR-ABL Signal Transduction Inhibitors.... ...................40 Table 7 Some Imatinib-Resistant BCR-ABL Kinase Mutations..... .............43
vii LIST OF FIGURES Figure 1 The Process of Hematopoiesis.......................... .............................3 Figure 2 The Philadelphia (Ph) Chromosome................... ...........................9 Figure 3 Structural Domains of the c-BCR Protein............... .......................14 Figure 4 Structural Domains of the c-ABL Protein............... ........................16 Figure 5 1 BCR-ABL Oncogene, 3 BCR-ABL Proteins Isoforms................ .18 Figure 6 Structure of the p210 BCR-ABL Protein............... .........................21 Figure 7 Mechanisms of Malignant Cell Transformation by BCR-ABL. .......22 Figure 8 BCR-ABL Activates the Ras/MEK Mitogenic Signal Transduction Pathway................................. ..................................24 Figure 9 BCR-ABL Activates the PI3K/AKT Anti-Apoptotic Signal Transduction Pathway................................. ..................................27 Figure 10 Summary of Signal Transduction Pathways Activated in BCR-ABL-Mediated Leukemogenesis....................... ....................32 Figure 11 Mechanism of Imatinib Mesylate Action in Chronic Myeloid Leukemia Cells...................................... ........................................41 Figure 12 Gene Amplification Due to the Presence of Double Minut es.........45 Figure 13 Cytokine Receptor Families............................ ..............................55 Figure 14 Co-culture Bone Marrow Transwell Stromal Model....... ................67
viii Figure 15 The Effects of Conditioned Media Collected at Various Times on Imatinib-Induced Cell Death in K562 CML Cell s............68 Figure 16 The Effects of Storage on 3-Hour-derived Conditioned Media......69 Figure 17 The Effects of Serum-Free Conditioned Media on Imatin ibInduced K562 Cell Death............................. .................................71 Figure 18 The Effects of Conditioned Media on K562 Cell Prolife ration and DNA Synthesis................................... ....................................73 Figure 19 Heat-Inactivated Conditioned Media does not Protect aga inst Imatinib Mesylate-Induced Cell Death................. ..........................75 Figure 20 HS-5-Derived Conditioned Media Protects K562 and KU18 2 CML Cells from Death Induced By Imatinib............. .....................78 Figure 21 Conditioned Media from Non-Stroma Cell Lines Does not Protect K562 CML Cells from Death Induced by Imatinib.. ...........82 Figure 22 HS-5-Derived Conditioned Media Protects K562 CML Cell s from Death Induced by 2nd Generation BCR-ABL Inhibito rs, Nilotinib and Dasatinib............................. .....................................84 Figure 23 STAT3 Phospho-Y705 is Increased in K562 and KU812 CML Cells Cultured in Conditioned Media............ ........................86 Figure 24 Basal Phospho-Y705 STAT3 is Increased in K562 CML Cells Cultured in Conditioned Media and is Sustained i n the Presence of Imatinib................................. ....................................87
ix Figure 25 K562 Cells Cultured in Conditioned Media Show Equal Levels of Phospho-Tyr STAT5 and Equal Inhibition of Phospho-Tyr STAT5 After Imatinib Treatment.......... ....................89 Figure 26 Dasatinib Inhibited Phospho-SRC Activity but not Phospho Y705 STAT3 Activity.................................. ...................................90 Figure 27 The Effects of Conditioned Media on STAT3 Downstream Targets, Bcl-Xl, Mcl-1 and Survivin.................. .............................92 Figure 28 Reducing STAT3 Levels With siRNA Reverses Imatinib Resistance in K562 Cells Cultured in Conditioned Media. ............96 Figure 29 The Effects of GM-CSF-, IL-6-, Or VEGF-Supplemented Regular Media on Imatinib-Sensitivity in K562 CML Ce lls..........101 Figure 30 Proposed Mechanisms of Resistance to BCR-ABL Inhibitors in Chronic Myeloid Leukemia Cells.................... .........................121
x LIST OF ABBREVIATIONS ABL Abelson kinase AKT Protein kinase B ALL Acute lymphoblastic leukemia AP Accelerated phase Bcr Breakpoint cluster region BME Bone marrow microenvironment BP Blast crisis phase CaLB Calcium-dependent lipid binding CD Cluster of differentiation CML Chronic myeloid leukemia CNL Chronic neutrophilic leukemia CP Chronic phase CRP Complement-regulatory proteins ECM Extracellular matrix GAB2 GRB-2-associated binding protein GAP GTPase activating protein GEF Guanine-nucleotide exchange factor GM-CSF Granulocyte macrophage-colony stimulating factor
xi GRB-2 Growth factor-binding protein 2 HSC Hematopoietic stem cell IC 50 Half-maximal inhibitory concentration IkB Inhibitor of kappa B IL Interleukin IM Imatinib mesylate JAK Janus kinase kb Kilobases kD Kilodalton MAPK Mitogen activated protein kinase MDR1 Multidrug resistance-1 drug transporter MRD Minimal residual disease NF-kB Nuclear factor kappa B NK cells Natural killer cells PDK1 Phosphoinositide-dependent protein kinase 1 PH Pleckstrin homology PI Phosphotidylinositides PI3K Phosphoinositide 3-kinase PMN or PML Polymorphonuclear leukocytes qRT-PCR Quantitative real time-polymerase chain reactio n SFK SRC family kinase SCID Severe combined immunodeficient
xii STAT Signal transducer and activator of transcription VEGF Vascular endothelial growth factor WBC White blood cell
xiii STAT3 Contributes to Resistance Towards BCR-ABL Inhibit ors in a Bone Marrow Microenvironment Model of Drug Resistance Nadine N. Bewry ABSTRACT Imatinib mesylate (imatinib) represents a potent molecu larly targeted therapy against the oncogenic tyrosine kinase, BCR-ABL. Although imatinib has shown considerable efficacy against chronic myeloid leukemia ( CML), displaying high rates of complete hematological and complete cytoge netic responses, treatment with imatinib is not curative and overtime a dvanced-stage CML patients often become refractory to further treatment. Acquired resistance to imatinib has been associated with mutations within the kinase domain of BCRABL, BCR-ABL gene amplification, leukemic stem cell quiescence as well as over-expression of the multidrug resistance (MDR1) gene However, in vitro resistance models often fail to consider the role of th e tumor microenvironment in the emergence of the imatinib-resistant phenotype. The bone marrow is the predominant microenvironment of CML and is a rich sour ce of both soluble factors and extracellular matrixes, which may influence dr ug response. To address the influence of the bone marrow microenvironme nt on imatinib
xiv sensitivity, we utilized an in vitro co-culture bone marrow stroma model. Using a transwell system, we demonstrated that soluble factors secre ted by the human bone marrow stroma cell line, HS-5, were sufficient to cause resistance to apoptosis induced by imatinib in CML cell lines. We subse quently determined that culturing CML cells in HS-5-derived conditioned m edia (CM) inhibits apoptosis induced by imatinib and other second generati on BCR-ABL inhibitors. These data suggest that more potent BCR-ABL inhibitors will not overcome resistance associated with the bone marrow microenvironm ent. Additionally, we determined that CM increases the clonogenic survival of CML cells following treatment with imatinib. HS-5 cells are reported to express several cytokines and growth factors known to activate signal transducer and a ctivator of transcription 3 (STAT3). Given its crucial role in the survival of hemat opoietic cells, we asked whether, 1) CM derived from HS-5 cells can activate STAT3 in CML cells and 2) does activation of STAT3 confer resistance to BCR-ABL inh ibitors. We demonstrated that exposure of the CML cell lines, K562 and KU812, to CM caused an increase in phospho-Tyr STAT3, while no increase s in phospho-Tyr STAT5 were noted. Moreover, resistance was associated wit h increased levels of the STAT3 target genes, Bcl-xl, Mcl-1 and survivin. Fu rthermore, reducing STAT3 levels with siRNA sensitized K562 cells cultured in CM to imatinibinduced cell death (p<0.05, Students t-test). Importa ntly, STAT3 dependency was specific for cells grown in CM, as reducing STAT3 leve ls in regular growth conditions had no effect on imatinib sensitivity. Toge ther, these data support a
xv novel mechanism of BCR-ABL-independent imatinib resista nce and provide preclinical rationale for using STAT3 inhibitors to increase the efficacy of imatinib within the context of the bone marrow microenvironment
1 CHAPTER I INTRODUCTION Hematopoiesis Although an estimated 1 x 10 10 red blood cells and 1 x 10 9 white blood cells are produced per hour, mature blood cells have a li mited lifespan of only a few hours to a few days and, therefore, require continu ous production. Hematopoiesis is the highly regulated process of blood cell production from hematopoietic stem cells (HSCs) within the bone marrow ( Figure 1). This process, which continues throughout adulthood, first beg ins in the embryonic yolk sack during the first weeks of embryonic development, proce eds in the liver and then the spleen. In adults, the bone marrow becomes th e major site of hematopoiesis and ceases in the liver and spleen. The pr imary locations of HSC production are femurs, hip, ribs, sternum, as well as oth er bones. HSCs are small, non-adherent, rounded cells that possess a rounded nucleus and can be identified by their low cytoplasm-to-nu cleus ratio. Characterized by their multipotency and high replicativ e and differentiation capacity, HSCs are capable of giving rise to all types of blood cells, which are divided into three distinct lineages: lymphoid, myeloid and erythroid. The lymphoid lineage of HSCs is composed predominantly of T-ce lls and B-cells, a
2 category of leukocytes or white blood cells (WBCs) called lymp hocytes. Although they are the smallest WBCs, they are the cornerstones of th e immune system, playing vital roles in the cell-mediated and humoral components of adaptive immunity, respectively. This lineage also consists of natura l-killer (NK) cells, which plays a central role in defending the body from bo th tumors and virally infected cells. The myeloid lineage produces another cate gory of leukocytes that include granulocytes, megakaryocytes and macrophages. Granul ocytes, also known as polymorphonuclear leukocytes (PMN or PML) because o f the varying shapes of the nucleus, consist of: 1) basophils, the least common of the granulocytes that play a role in regulating the immune inflammatory response to allergens and drugs; 2) eosinophils, which also respond to allergens and are involved in defending against infection and parasites; an d 3) neutrophils, the most abundant WBCs in the body that contain granules of bacteria-killing enzymes in their cytoplasm and function as phagocytes, respondi ng primarily to acute inflammation due to bacterial infection or fungi Megakaryocytes produce blood platelets, which are small cells necessary for norma l blood clotting. Macrophages are phagocytes involved in both innate and ada ptive immunity. The third lineage of HSCs, the erythroid lineage, gives rise to oxygen-carrying red blood cells (RBC), which make up 45% of blood volume.
3 Figure 1 The Process of Hematopoiesis All of the cellular components of the blood are derived from hematopoietic stem cells. Blood ce lls are divided into three lineages: lymphoid, myeloid and erythroid. Types of Leukemia Hematopoiesis is highly regulated by a balance of growt h and death signals that determine the fate of each cell in the he matopoietic system. Consequently, mutations of stem cells can disrupt the hema topoietic system to confer a growth advantage. This deregulated hematopoie sis creates an imbalance between cell production, destruction and diffe rentiation and can lead to a number of hematologic diseases. Leukemia, the abno rmal proliferation of
4 blood cells in the bone marrow, results in deregulated blood cell production and cell differentiation. Leukemia is classified as acute or chronic based on path ological and clinical characteristics, such as the speed with which the disea se progresses and the phenotype of the affected cells. While acute leukem ia is distinguished by the rapid increase of WBCs, an increase in mature, yet abnor mal blood cells typifies chronic leukemia. Leukemia can be further sub-divided as lymphoid or myeloid based on the lineage of the blood cells that are affect ed (Table 1). Normal vs. Leukemic Hematopoietic Stem Cells Normal HSCs are a rare, quiescent homogeneous cell popul ation whose production is very tightly regulated. Additional character istics of stem cells include their multipotent differentiation capacity an d their ability to self-renew .
5 Based on the latter characteristic, HSCs can be subdivide d by their stem cell hierarchy, or the degree to which they are able to self -renew, as shortor longterm repopulating cells . Despite major advances in our understanding of stem cells, little is known about their role of stem cells in human malignancies. T here is a growing body of evidence that lends support to the existence of cancer ste m cells (CSCs), a small, primitive subpopulation of cancer cells that escape normal control and are capable of initiating, propagating and maintaining t he cancer cell population . Studies show that CSCs have been implicated in a number of human malignancies, including breast and brain tumors [4, 5]. Among the CSC subpopulation, studies have identified an d characterized cancer-causing leukemic stem cells (LSCs). These cells share many of the canonical properties of normal HSCs, such as multipotency, the ability to selfrenew, and extensive proliferative capacity [6, 7]. Ho wever, unlike normal HSCs, these LSCs display a reduced or absent exogenous growth f actor-dependence . Instead, these cancer-causing leukemic cells are capab le of generating their own growth signals, most notably via the autocrine produ ction of cytokines such as interleukein-3 (IL-3) and/or granulocyte colony-stimul ating factor (GM-CSF) . Early studies using immunodeficient mice as xenotranspla ntation models demonstrated the existence of leukemia-initiating stem cells in CML patients . Like normal stem cells, these cells lack lineage marker s (Lin-) and are
6 distinguished as follows: CD34+CD38Thy-1+Lin(Baum C, 1992), where CD34 is a single-pass transmembrane glycoprotein that function s as a cell-cell adhesion factor and is expressed early in hematopoietic tissues ; CD38 is an early cell surface marker for white blood cells  and C D90 or Thy-1 is the thymocyte (T-cell precursor) marker . Research done in these early CML models confirmed that while chronic phase CD34+ CML pat ient cells contained both normal and LSC fractions, patients in blast crisis p hase CML have a greater preponderance of LSCs . CML LSCs are not composed o f a functionally homogeneous cell population as previously believed, but like normal HSCs, are comprised of a heterogeneous pool of short-term and lon g-term repopulating cells [18-20]. Addressing the failures of cell cycle-active chemotherapeut ic agents to eradicate CSCs in vivo and in vitro poses a challenge for CML patient care, as primitive quiescent cells are resistant to imatinib. Thi s resistant, quiescent LSC phenotype coupled with an augmented stem cell populati on, an enhanced selfrenewal capacity and decreased growth-factor dependence/i ncreased autocrine secretion of growth factors may contribute to chemo-resista nce in CML patients and subsequent relapse.
7 CHAPTER II CHRONIC MYELOID LEUKEMIA AND THE PHILADELPHIA CHR OMOSOME Currently within the United States every 1 in 4 death s is due to cancer . Among cancers, chronic myeloid leukemia (CML) is poss ibly the most wellstudied. CML is a myeloproliferative disorder that is ch aracterized by the neoplastic transformation and malignant expansion of pl uripotent hematopoietic stem cells (HSCs) within the bone marrow resulting in can cer of the white blood cells. It accounts for 7% to 20% of all cases of leukemia and according to the America Cancer Society estimates, in 2008 there will be 4 ,830 newly diagnosed cases of CML with 450 of those resulting in deaths. Whil e it affects all age groups, it is predominantly affects middle-aged to elde rly individuals at a rate of 1-2 per 100,000. CML is the first disease to be linked to a clear, consistent genetic abnormality, the Philadelphia (Ph) chromosome which was first described in 1960 by Peter Nowell of the University of P ennsylvania and David Hungerford of the Fox Chase Cancer Center, two scientist s from Philadelphia, Pennsylvania . The team described a tiny acrocentric chro mosome in cells cultured from the blood of seven patients harboring th e disease. This chromosome later became the unique genetic signature a nd diagnostic clinical marker of CML patients 
8 The Ph chromosome is formed by the reciprocal translocat ion and fusion between the long arms of chromosomes 9 and 22 [t(9,22)( q34;q11)] [24-26] (see Figures 2A and B). Chromosome 9 encodes the proto-oncog ene ABL, while chromosome 22 encodes the breakpoint cluster region gene ( bcr ), a serine/threonine kinase named after the site where it is located. This translocation event results in the formation of the chim eric oncogene, BCR-ABL which gives rise to an aberrant and deregulated, constit utively active tyrosine kinase capable of activating numerous downstream targets.
9 Figure 2: The Philadelphia (Ph) chromosome CML is the first disease to be linked to a clear and consistent genetic abnormality, t he Philadelphia chromosome. ( A ) Human karyotype, with chromosomes 9 and 22 highlighted; ( B )
10 Formation of the Philadelphia chromosome. The Philade lphia chromosome is formed by the head-to-tail reciprocal translocation be tween the long arms of chromosome 9, which contains the proto-oncogenic ABL non-receptor tyrosine kinase, and chromosome 22, which contains the bcr serine/threonine kinase [t(9:22)(q34:11)]. CML is divided into three phases based on clinical charact eristics that such as the amount of blast cells (immature, abnormal whi te blood cells) in the blood and bone marrow and the severity of symptoms prese nted (see Table 2). The initial chronic phase (CP), diagnosed in approxima tely 95% of patients, has an average duration of 4 to 6 years and is often asymptom atic, or patients may experience mild symptoms of fatigue, anemia, splenomelga ly, upper abdominal pain or mass. Less common presentations may include fever, w eight loss, anorexia or gout. Chronic phase CML is characterized by a n overproduction of immature myeloid and mature granulocytes in the spleen, peripheral blood and bone marrow; however, the cells retain the ability to d ifferentiate and function normally. The lack of therapeutic intervention and the presence of additional genetic and/or epigenetic defects cause the disease to pro gress to an accelerated phase (AP), characterized by the presence of 10 20% primitive blast cells in the peripheral blood and bone marrow. When t he disease advances to the blast-crisis phase it resembles acute leukemias and i s characterized by the presence of >20% undifferentiated blasts in the periph eral blood and bone marrow. At this grave stage patients have a life expectan cy of 3-6 months.
11 While the cause of the initial translocation event which gives rise to CML remains unclear, studies have shown that translocation can be induced by ionizing radiation (IR) . Additionally, the physica l distance between the BCR gene and the ABL gene in human hematopoietic progenitor cells is shorter than might be expected by chance and could favor translocatio n between the two genes [28, 29].
12 The BCR-ABL Protein BCR-ABLs oncogenic capacity is mediated by its multiple m odular domains inherited from BCR and ABL that facilitate di verse protein-protein interactions (some of BCR-ABLs protein targets are descri bed in Table 3).
13 Structure of c-BCR Protein The 160-kd (p160) breakpoint cluster region protein, B CR, is a large multidomain, ubiquitously expressed serine/threonine kinase com prised of 1,271 amino acids (Figure 3). At its NH 2 -terminus, BCR has a coiled-coil/oligomeric domain which enables the protein to dimerize in vivo . The serine/threonine kinase domain defines the proteins function by phosphoryl ating substrate on either serine or threonine residues. Known substrates of BCR include Bap-1, a member of the 14-3-3 family of proteins [30, 31] and BCR itself . Within the center of the protein is the Rho/guanine nucleotide e xchange factor (Rho/GEF) domain that is specific for the Rho-subfamily of small G TPases, RhoA and CDC42. Both small GTPases regulate actin cytoskeleton rea rrangement and control cell functions such as cell migration, morphology and cell cycle progression. The Rho/GEF domain also interacts with and activates transcription factors such as NF-kB . This domain is followed by the P leckstrin homology (PH)/Calcium-dependent lipid binding (CaLB) domain that binds phospholipids, such as phosphotidylinositides (PI). This interaction pla ys an important role in signal transduction by localizing the protein to cellul ar membranes and activating second messenger. The carboxyl-terminus of the protein cont ains a Rac-GTPase activating protein (Rac-GAP) domain , a small GTPa se belonging to the Ras super-family that plays a role in regulating actin polyme rization. Additionally, BCRs tyrosine 177 (Y177) residue is an autophosphorylati on site that is essential for binding to the adaptor molecule, GRB-2 and activating the mitogen
14 activated protein kinase (MAPK) signal transduction path way. BCR-ABL autophosphorylation also occurs on Y283, Y328 and Y360 [35-37]. Figure 3: Structural Domains of The c-BCR Protein There are several functional domains found within BCR: At its N-terminus lies an oligomerization domain; this is followed by a serine/threonine kinase d omain; a Rho/Guanine nucleotide exchange factor (Rho/GEF) domain; a Pleckstri n homology (PH) domain and a Rac-GTPase activating protein (Rac-GAP) do main. Key residues include Y177 which enables mitogenic activation of the M APK pathway via binding of the adaptor molecule GRB-2 Structure of c-ABL Protein The c-ABL gene was initially identified as the human homologue o f the Abelson murine leukemia virus (A-MuLV), which is the acut e transforming retrovirus encoding the v-ABL oncogene . It is located on chromosome 9q34, spans 230 kilobases (kb) and possesses 11 exons that encodes a 145 kilodalton (kD) (1,097 amino acids) non-receptor tyrosine kinase. I n addition to c-ABL, the ABL family of non-receptor tyrosine kinases is also compr ised of ARG (ABLrelated gene, designated ABL2). Like BCR, ABL is expressed ubiquitously within the cell, but predominates within the nucleus and cytoplasm. Its diverse localization within the cell reflects its diverse functions, which are mediated by its protein-pr otein interactions.
15 These interactions facilitates ABLs involvement in numero us cellular processes, including cellular response to DNA damage and genotoxi c stress [32, 39-41], cell cycle regulation [28, 42] and rearrangement and cell mi gration . There are two isoforms designated 1b and 1a, which are generated by alternat ive splicing of the first exon (Figure 4) and are driven by their own promoters. Differing only in their NH 2 -terminus sequences, the 1b isoform encodes a 6.5 kb mRNA, while the 1a isoform encodes a slightly shorter 5 kb transcript t hat is 19 amino acids shorter than 1b. Within the amino-terminus of the 1b isoform lies a myri stoylation site which enables plasma membraneattachment. This site is missing in the 1a isoform. Further within the NH 2 -terminus are three SRC-homology domains, SH1-SH3. The SH1 tyrosine kinase domain defines protein s function. It contains two key residues: Y412, the site of phosphorylation in t ransformed cells, and Y393, the major site of autophosphorylation within th e proto-oncogenes kinase domain. Both the SH2 and SH3 domains are involved in m ediating proteinprotein interactions: the SH2 domain binds to the phosp hotyrosine regions of target proteins, while the SH3 domain recognizes proli ne residues. Within the center of the protein is a proline-rich region (P) whi ch, conversely, binds the SH3 domains of other protein substrates . Towards the ca rboxyl-terminus there are three (NLS) [42, 45], a DNA binding domain (DNA BD) , a nuclear exporting signal (NES), and an actin-binding domain which mediates binding to globular/ monomeric (G) and filamentous (F) actin.
16 Figure 4: Structural Domains of The c-ABL Protein There are several functional domains found within ABL: a myristoylation si te; SRC-homology (SH) domains SH1-3; four proline-rich regions (P); three nu clear localization signal (NLS) regions; a DNA binding domain (DNA BD; an actin binding domain and a nuclear export signal (NES). One BCR-ABL Oncogene, 3 BCR-ABL Proteins Isoforms Breakpoints within the BCR gene results in primary fusion BCR-ABL transcripts with the same portion of ABL sequence in the carboxyl-terminus, but include different amounts of bcr sequence at the NH2-terminus (see Figure 5A). Unlike ABL, which has only one breakpoint within a region greater t han 300 kb at its 5 end, BCRs breakpoints are localized within three breakpoint clust er regions (bcr) resulting in three BCR-ABL fusion prote ins varying in size from 190 to 230 kD (Figure 5B). Each isoform is associated with a different type of leukemia. In 95% of CML patients and 1% Ph-positive acute lymphob lastic leukemia (ALL) patients, the breakpoint occurs within the major breakpoint cluster region of BCR (M-bcr), a 5.8 kb area stretching across exons b12-b16. These patients
17 possess the p210 BCR-ABL isoform, which is formed when eit her exon b12 or b13 of BCR fuses with exon a2 of ABL. Most ALL patients possess the breakpoint within the minor bcr region (mbcr), which spans a 54.5 kb area between exons b1 and b2 The resulting p190 BCR-ABL fusion protein is characteristic of ALL. The p230 BCR-ABL isoform is seen in patients with chroni c neutrophilic leukemia (CNL), a rare disease in which too many stem cel ls develop into neutrophils. CNL may stay the same or progress quickly into acute leukemia. The recently characterized mu-bcr ( -bcr), is associated with this form of leukemia and is located downstream of exon b19. The research presented here is focused on the role of p2 10 BCR-ABL in CML pathogenesis.
18 Figure 5: 1 BCR-ABL Oncogene, 3 BCR-ABL Proteins Is oforms Fusion transcripts contain the same portion of the ABL within the carboxyl-terminus but different amounts of BCR sequence at the amino-terminus depending on where the breakpoints occur within BCR ( A ), Breakpoint locations within the ABL and BCR genes. 3 breakpoint cluster regions (bcr) are found wit hin the BCR gene: the major bcr (M-bcr), the minor bcr (m-bcr), and the mu bcr ( -bcr). Breakpoints within each of these regions give rise to a different f orm of leukemia. ABL has a >300 kb breakpoint region between exons a1 and a2. ( B ), Structure of the chimeric BCR-ABL transcripts. The p190 BCR-ABL isoform is found in Phpositive ALL, the p210 isoform is found in CML, and the p230 isoform is found in CNL (Clinical Cancer Research Vol. 8, 2177-2187, July 20 02).
19 CHAPTER III BCR-ABL PATHOGENESIS The Role of Modular Domains Features of key domains and motifs contribute to p210 BC R-ABLmediated malignant transformation in CML patients (Fi gure 6A). The oligomerization domain of BCR performs several crucial r oles: 1) it enables the chimeric BCR-ABL proteins to form dimers (or tetramers) and to transautophosphorylate, which abrogates the need for regula tory external kinases (see Figure 6B); 2) it is an essential activator of ABL ki nase activity ; and, 3) it promotes the association of BCRABL with F-actin which f acilitates cytoskeletal rearrangement, cell spreading and cell migration  The GRB-2 binding site at tyrosine 177 activates both the MAPK survival pathway [46, 47], and the phosphatidylinositol 3-kinase (PI3K) anti-apoptotic pat hway . Purified ABL is kinase-active, therefore regulatory cisand trans-acting elements are believed to be involved in its constitutive inhibition. Intermolecular interactions within the first three tandem domains of A BL facilitate their assembly into an auto-inhibitory structure that negatively regul ates the protein. The SH3 domain interacts with Pro1124 within the SH2-linker r egion and results in a clamp structure that confines the kinase in an inactive co nfirmation [44, 48].
20 Furthermore, the myristoylation site binds to the tyros ine-kinase domain of ABL, operating as a latch that reinforces the SH3SH2 cla mp [48, 49]. Studies show that the SH3 domain is intrinsically capable of suppressi ng the transforming ability of ABL and its deletion or mutation fully activa tes the kinase . Additionally, a number of proteins also bind to the SH 3 domain of ABL and serve as negative regulators of the proto-oncoprotein. ABL in teractor proteins 1 and 2 (Abi-1 and Abi-2) bind to ABLs SH3 domain and activat e the inhibitory function of this domain [50-52]. Additionally Pag/Msp23, anothe r ABL SH3-binding protein, dissociates from ABL under oxidative stress and facilitates its tyrosine kinase activity . Fusion of BCR with ABL breaks the intermolecular bonds within ABLs SH domains and disrupts its constitutively inhibited regulato ry configuration. Furthermore, activated ABL tyrosine kinase triggers the ubiquitin-proteasomemediated destruction of the Abi proteins that normally antagonize its oncogenic potential . Collectively, deregulation of BCR and ABLs multiple do mains is a major contributing factor in CML pathogenesis.
21 Figure 6: Structure of p210 BCR-ABL Protein ( A ), BCR-derived sequences and ABL-derived sequences. Fusion of BCR to ABL is centra l to the pathogenesis of CML as it not only disrupts ABLs regula tory configuration but also disrupts BCR-ABLs localization within the cell. ( B ), Location of transautophosphorylation sties within BCR-ABL. Proteins are able to dimerize via their oligomerization domains and trans-self-phosphorylate on both Y177, a key binding site for GRB-2 and Y1124 within the activation loop of ABLs kinase domain. Activation of Multiple Signaling Pathways BCR-ABL activates numerous pathways commonly used in hematop oietic growth factor receptor signaling. These activated pathwa ys contribute to CML
22 pathogenesis by causing 1) constitutive mitogenic activatio n, inhibition of apoptosis and decreased adhesion to bone marrow stroma extracellular matrices (Figure 7). Figure 7: Mechanisms of Malignant Cell Transformati on by BCR-ABL BCRABL-mediated malignant transformation in CML pathoge nesis occurs via three major mechanisms: constitutive mitogenic activation, inhibi tion of apoptosis and decreased adhesion to bone marrow stroma extracellular matrices. Constitutive Mitogenic Activation: The Ras/MEK Path way The Ras/extracellular-signal-regulated kinase (ERK) (R as/ERK) transmits extracellular signals into the nucleus and results in gen e transcription [55-57]. The autophosphorylated Y177 residue within BCR-ABLs first exon serves as a docking site for growth factor-binding protein 2 (GRB2)s SH2 domain and
23 facilitates signal transduction (Figure 8). GRB-2 form s a complex with the guanine nucleotide exchange factor, son of sevenless (SOS) protein, which stabilizes RAS in its active GTP-bound conformation and induces its activation. RAS activation can also be mediated by three other adapt or molecules and BCRABL substrates, SH2-containing protein (Shc), Crk-like pr otein (Crkl) and phospho-protein p62 DOK [58-60]. RAS activation initiates a cascade of signaling events invo lved in CML pathogenesis. Activated Raf-1 (MAP KKK) phosphorylates MEK1/2 (MAP KK) on both serine and threonine residues; MEK1/2 then phosph orylates and activates the extracellular-signal regulated kinases, ERK1/2 (MAP K), which results in the phosphorylation and activation of the transcription factor Elk1. Upon activation, Elk-1 translocates to the nucleus and binds the serum resp onse factor (SRF) transcription factor. This ternary complex then binds the serum response element (SRE) of the immediate early genes and proto-oncogenes, c-Fos and c-Jun, to direct their transcriptional activation and regulation. Together, c-Fos and c-Jun DNA-binding proteins heterodi merize to form the AP-1 transcription factor which upregulates the tr anscription of genes involved in cell proliferation, cell differentiation an d mediating growth-factor independence. These genes include the G1-cyclin dependen t kinases (CDKs) ; CDK regulator, cyclin D1 ; c-Jun ; the a nti-apoptotic protein Bcl-xL ; granulocyte-macrophage colony-stimulating factor (G M-CSF) hematopoietic cytokine ; the second messenger, cyclic adenosine monoph osphate (cAMP)
24 ; and cAMP response element binding protein (CREB) . The upregulation of these genes is characteristic of CML pathogenesis. Figure 8: BCR-ABL Activates the Ras/MEK Mitogenic S ignal Transduction Pathway. Autophosphorylation of BCR-ABL on Y177 results in it s interaction with the adaptor molecule, GRB-2 GRB-2 interacts with the guanine nucleotideexchange factor, SOS, which in turn mediates RAS activat ion. The adaptor molecules Shc and Crkl can also mediate RAS activation. RA S is coupled to the mitogen activated protein kinase (MAPK) pathway by the se rine/threonine kinase, Raf. Raf in turn phosphorylates mitogen activated and extracellular-signal regulated kinase kinases 1 and 2 (MEK1/2). This lead to a phosphorylation and activation of and extracellular-signal regulated kinase kinases 1 and 2 (Erk1/2) and the subsequent activation of the transcription factor Elk.
25 Inhibition of Apoptosis The PI3/AKT Pathway The lipid kinase phosphoinositide 3-kinase (PI3K) r epresents a family of cytosolic, intracellular signaling proteins whose deregula tion is associated with malignant cell transformation . Class I PI3Ks, which are commonly activated in CML, are heterodimeric molecules that are composed of a regulatory and a catalytic subunit and can be divided into two subclasses, class 1A and B. While class1A consists of an 85 kD regulatory subunit (p85) and a 110 kD catalytic subunit (p110), class1B consists of only a 110 kD catalytic sub unit (p110 /PI3K ). The class 1A PI3K pathway is activated either in one o f two ways: 1) the p85 regulatory subunit binding to an activated t yrosine residue on the activated interleukin (IL)-3 receptor, or 2) by the p1 10 subunit binding to activated Ras. Once activated, PI3K translocates to the cell plasma m embrane where it phosphorylates the membrane phospholipid PI(4,5)P2 to p roduce PI(3,4,5)P3. This in turn activates phosphoinositide-dependent protei n kinase 1 (PDK1) and facilitates its membrane localization where it phosphoryl ates and activates protein kinase B (AKT), PI3K primary downstream effecto r protein. Activation of PI3K/AKT survival pathway is essential to BC R-ABLmediated leukemogenesis  (Figure 9). BCR-ABL inter acts indirectly with the p85 subunit of PI3K through the adaptor molecules GRB2 and SHC to stimulate the constitutive activation of the p110 catalytic subunit . This interaction is mediated by BCR-ABLs recruitment of GRB-2-associated bi nding protein
26 (GAB2) through GRB-2. GAB2 then activates the p85 regu latory subunit of PI3K, leading to constitutive AKT activation, which can also b e achieved by SHCs binding to the SH2 domain within BCR-ABLs carboxyl-te rminus. Mutations to the p85 subunit of PI3K resulted in the i nhibition of BCR-ABLdependent growth in hematopoietic CML cells, which hig hlights the role of PI3K in BCR-ABL-mediated pathogenesis . Additionally, BCR-ABL upregulates the transcription of the p110 subunit of PI3K in several CML cell lines. Constitutive activation of the PI3K/AKT pathway suppresses apoptosis in several ways. AKT phosphorylates the pro-apoptotic protein Bad, which is scavenged by the cytosolic protein 14-3-3 and neutralized. T his decreases apoptosis by preventing Bad from binding and inhibitin g the anti-apoptotic protein Bcl-xL . AKT promotes cell survival by phosphorylating and suppressing the activity of FKHRL1/FoxO3, a member of the 14-3-3 Fork -head family of transcription factors. Phosphorylation by AKT sequesters Fo xO3 in the cytoplasm, preventing it from translocating to the nucleu s where it activates genes necessary for cell death, such as Bad and Bim [73, 7 4]. AKT-induced phosphorylation of the initiator death caspase, caspase9, also decreases apoptosis by inhibiting its proteolytic activities direct ly . The ubiquitously expressed serine/threonine kinase glycogen synthase kinase -3 (GSK-3), another key downstream target of AKT, also induces apoptosis in tu mor cells . However, induction of apoptosis by GSK-3 is inhibited w hen it is phosphorylated by AKT. Lastly, AKT activates the nuclear factor-kB (NF-kB) transcription-factor,
27 which is involved in the transcription of anti-apoptoti c genes. Normally NF-kB is held in the cytoplasm in an inactive complex with its i nhibitor, IkB. However, phosphorylation of NF-kB by AKT releases the transcription factor and enables it to perform its anti-apoptotic gene transcription . Figure 9: BCR-ABL Activates the PI3K/AKT Anti-Apopt otic Signal Transduction Pathway. By recruiting GRB-2 -associated binding protein (GAB2) or the adaptor molecule Shc, BCR-ABL activates the PI3K/ AKT pro-survival pathway; a process which can also be mediated by the signa ling adaptor protein Crk. This interaction facilitates PI3K/AKT mediated an ti-apoptotic signaling in by phosphorylation and inhibition of: 1) the pro-apoptot ic protein Bad, making it incapable of suppressing the activity of Bcl-xL; 2) FKHR L1/FoxO3 making it incapable of translocating to the nucleus to activate p ro-apoptotic proteins; 3) the cysteine protease, caspase-9; 4) the serine/threonine kina se glycogen synthase kinase-3 (GSK-3).
28 The JAK/STAT Pathway Signal transducer and activators of transcription (STATs) were originally characterized as latent cytoplasmic transcription factors. The y are members of a family transcription factors that are involved in normal cellular response to cytokines, such as cell differentiation. There are seven STAT proteins: STAT1, STAT2, STAT3, STAT4, STAT5 (which consists of two closely related proteins, STAT5A and STAT5B) and STAT6. They have diverse biolo gical roles and control critical cellular functions including cycle progre ssion, cell proliferation, survival, differentiation, apoptosis and oncogenesis. STAT-mediated signal transduction occurs in several steps. Extracellular binding of cytokines to their cognate receptors, followe d by receptor dimerization, leads to the engagement of associated Janus kinases (JAKs), a family of four cytoplasmic non-receptor tyrosine kinases (JAK1, JAK2, JAK3 a nd Tyk2), to the receptors cytoplasmic tail. JAKs then become activated by aut o-phosphorylation, causing them to phosphorylate tyrosine residues within the receptor s cytoplasmic tails. These receptor-associated phosphotyrosine residues serve as docking sites that recruit latent cytoplasmic STAT monomer s through its SH2 domain. Further tyrosine phosphorylation by JAKs leads to activation of STAT molecules, which in turn induces the formation of STAT h omoor heterodimers through interaction of the phosphotyrosine of one mol ecule with the SH2 domain of the other molecule. STATs can also be activated direc tly by growth factor receptors, such as epidermal growth factor receptor (EGFR) platelet derived
29 growth factor receptor (PDGFR) and GM-CSF receptor (or CD116), as well as by cytoplasmic kinases and oncoproteins, such as SRC and ABL. Act ivated STAT dimers are translocated into the nucleus via importin / as well as Ran-GDP. Once inside the nucleus activated STAT molecules bind a unique cytokine inducible sequence named GAS ( -interferon activated sequences) within the promoter of target genes, thereby activating gene tran scription. The activation of STATs is required for repopulation of the stem cell poo l, as well as cell proliferation and differentiation in response to ext ernal stimuli. Several STATs, particularly STATs 1, 3 and 5, are constitu tively activated in several types of cancers. Typically STAT1 is involved in cell growth arrest [78, 79], in promoting apoptosis  and is implicated as a tumor suppressor . STAT3 and STAT5 are involved in cellular transformation cell cycle progression and preventing apoptosis [82-84]. Aberrant activation of cytokine-mediated signal transduc tion pathways involving STATs are capable of conferring cytokine-indepen dent growth to leukemia cells . In CML, BCR-ABL can induce phosphory lation and activation of STAT5 via its SH2 and SH3 domains, hijacking normal cellular functions by circumventing the need for cytokine-mediated STAT5 activat ion. BCR-ABL kinase-dependent STAT5 activation in CML cells has been shown to be an essential component of CMLs pathogenesis [86, 87] (see F igure 10). Cells with constitutive STAT activation promote malignant cell proliferation by upregulating the gene expression cyclin D1, a cell cycl e regulator that controls
30 cell cycle progression from G1 to S phase, and c-Myc, a key t ranscription factor that promotes cell cycle progression by inhibiting the exp ression of p21, a cyclindependent kinase inhibitor [88, 89]. c-Myc is also capable of upregulating the expression cyclins. Additionally, constitutively activated STAT s inhibits apoptosis by upregulating the expression of members of the Bcl-2 family of anti-apoptotic proteins Bcl-xL, Bcl-2 and Mcl-1 (see Table 4 for list of proand anti-apoptotic Bcl-2 family of proteins). The combined effects of 1) deregulated cell cycle progress ion and, 2) reduced apoptosis, results in uncontrolled cell prolife ration and malignant cell transformation characteristic in CML pathogenesis. The BCL-2 Family of Proteins. The Bcl-2 (B-cell lymphoma-2) family of proteins governs mitochondrial outer membrane permeability (MOMP ). They are categorized functionally by their ability to promote or i nhibit cell death (proor anti-apoptotic) and structurally by their Bcl-2 homology ( BH) domains. Each protein possesses at least one of four of the characteristi c BH domains, named
31 BH1-4, and can therefore be subdivided further into t hose that have multidomain and those that have BH3-only domain. With the exception of Mcl-1 and BLF/A1, which have three conserved BH domains, the anti-apoptotic family of proteins possesses all four domains. Conversely, with the exception of Bax, Bak and Bok which are multidomain proteins, all the pro-apoptotic family of proteins are BH3only domain-containing proteins. TM, transmembrane. Decreased Adhesion to Bone Marrow Stroma and Extrac ellular Matrices During hematopoiesis, cell differentiation and prolif eration is tightly regulated through specific interactions between HSCs and bone marrow stroma. Transformed CML progenitor cells escape normal negative regulatory signals and exhibit enhanced motility and perturbed adhesion p roperties. Rho GTP activating proteins (GTPases) are important regulators o f cell motility, cell adhesion, actin assembly and cell migration . They play important roles within the cell by transducing extracellular regulatory signals to effector molecules (reviewed in . Rac GTPases, Rac 1-3, are members of the Rho GTPase family and cycle between GDPand GTP-bound states. BCRABL activates RhoA and Rac1 GTPases via BCRs Rho/GEF domain and ABL s Rac GAP domain respectively. Additionally, Rac1 and Rac2 are signi ficantly upregulated in CD34+ chronic phase CML cells . BCR-ABL recruits and activates Rac1 through its interaction with its hematopoietic-specific GEF, Vav, which leads to alterna tions in actin assembly. Deregulation of actin assembly enhances the motility of C ML cells and facilitates their diminished adhesion to bone marrow stroma . CML cells also display
32 increased expression of focal adhesion proteins, such as fo cal adhesion kinase (FAK), talin and paxillin, when compared to normal ce lls. These focal adhesion proteins play important roles in cell motility and survi val, and their increased expression enable CML cells to escape the normal negative regulatory signals provided by component of the BME . Additionally, C ML cells express an adhesion-inhibitory variant of 1 integrin that is not found in normal progenitor cells . Taken together, BCR-ABL-mediated cytoskeletal rearrange ment contributes to CML pathogenesis. Figure 10: Summary of Signal Transduction Pathways Activated in BCRABL-Mediated Leukemogenesis. BCR-ABL is capable of phosphorylating numerous substrates to activate signal transduction pathways sh ared by cytokine
33 receptors and are involved in the growth and differen tiation of hematopoietic cells. These activated signal transduction pathways include: Ras/MEK, which results in promitotic transcriptional gene regulation; PI3/AKT; which results in suppression of apoptosis; JAK/STAT5, which results in prosurvival gene transcription; Rac-GTPase activation, which results in cytoske letal reorganization and anchorage-independent growth.
34 CHAPTER IV TREATMENT FOR CHRONIC MYLOID LEUKEMIA Before Signal Transduction Inhibitors Prior to the development of targeted therapy against CML, therapeutic options included cytotoxic drugs such as busufan, hydroxyurea recombinant interferon (rIFN ), bone marrow transplant (BMT) and allogeneic stem cel l transplantation (allo-SCT). Busulfan Busulfan is an inexpensive, oral alkylating agent that produces DNAcrosslinks resulting in the inhibition of DNA and protein synthesis. It produces periods of hematologic response in patients and is often restricted to those who are intolerant or resistant to hydroxyurea (see Table 5 for a list of definitions of the types of response to CML treatment). While it lower s white blood cell (WBC) count, its myelosuppressive effects were often delayed and p ersistent, which complicated therapy. Additional draw-backs to using busulfa n include hyperpigmentation and interstitial lung disease [96, 9 7].
35 Hydroxyurea Hydroxyurea is also a DNA-synthesis inhibitor that achieved hematologic remission in patients within 1-2 months of use and is nor mally well tolerated in CML patients. Although it has been shown to be better than busulfan, both drugs rarely induced cytogenic responses. Neither busulfan nor hydr oxyurea prevented disease progression from chronic phase to acute phase in CML patients. Recombinant Interferon-Alpha Recombinant interferon-alpha (IFN ) treatment produces cytogenetic response in 70% of patients, with 20% of those patient s experiencing complete cytogenic response. Studies show that rIFN works well against CML alone or in combination with cytarabine, another DNA-damaging agen t. It increased patients survival rates and is the treatment of choice for patient s who do not have a HLAmatched bone marrow donor or are too old for BMT [98 99]. Under rIFN treatment patient survival at 5 years was 57%, compared to 42% in the other previously discussed treatment groups (CML Trailists Coor perative Group, J Nat Cancer Inst. 1997; 89:1616-1620). Unfortunately, rIFNa treatment is limited by its toxicity profile, with 15-25% of patients discontinuing treatment due to intolerable side effects and another 35-50% requiring dose reductio n due to poor drug tolerance. Side effects include anorexia, fever, chills an d postnasal drip.
36 Allogeneic Stem Cell Transplantation At present, only allogeneic stem cell transplantation (a llo-SCT) is an effective and curative treatment option for CML patien ts, providing long-term disease eradication . Stem cells are derived from the bone marrow or peripheral blood of donors. Results are best for patie nts in chronic phase when compared to those in accelerated or blast crisis phase. Th e long term survival rate using this treatment method is 50-60% for CML chr onic phase patients, 1520 for accelerated phase patients, and <10% for blast c risis phase patients . Treatment-related morbidity and mortality rates in all o-SCT patients are high and often associated with organ toxicity, complica tions from infection and chronic graft-versus-host disease (GVHD) . Due to the difficulty in finding a healthy, young HLA-matched sibling or donor, allo-SCT i s often reserved for patients under the age of 55 years old who are unrespon sive to secondary BCRABL inhibitors.
37 Rationally Designed BCR-ABL Signal Transduction Inh ibitors Imatinib Mesylate Constitutive BCR-ABL kinase activity in HSCs is an essential component of CML pathogenesis. The identification of BCR-ABL as the hallmark oncogenic transforming protein in CML made it an ideal target for drug development. This gave rise to rationally designed, small bioavailable sign al transduction inhibitors (STI) specific for the tyrosine kinase domain of BCR-ABL Imatinib mesylate (imatinib, also known as Gleevec; formerly STI571 and CG P 57158; Novartis Pharmaceuticals Corporation, Basel, Switzerland) was ide ntified as a lead
38 compound in a high throughput in vitro screening for tyrosine kinase inhibitors (TKIs). The 2-phenylaminopyrimidine-derivative was desi gned to bind the ATP binding site of ABL kinases (see Figure 11) While imati nib is also capable of binding to other kinases, such as the stem cell factor (SC F) cytokine receptor, ckit, and platelet-derived growth factor receptor (PDGFR ), it is specific for ABL oncoproteins, including c-ABL and ETV1-ABL. Functionally, imatinib easily transverses the cell membra ne and selectively binds the tyrosine kinase domain of BCR-ABL. This dom ain consists of a bi-lobed structure in which Mg-ATP is located in a deep cleft betw een the aminoand carboxyl-terminal lobes. BCR-ABL cycles between two distin ct active and inactive states, acting as a molecular switch depending on the conformation of its activation loop which contains the highly conserved aspa raginephenylalanine-glycine (DFG) motif. In an active kinase conformation, Y1294 in the activation loop becomes phosphorylated and takes on a DFG-out configuration, in which these conserved residues are flipped out of their usual posi tion. This open conformation provides a platform for substrate binding and activation. However, in the inactive kinase conformation, this DFG mo tif is folded into the ATP-binding site and makes a channel opening outsi de the threonine 315 gatekeeper residue . This residue is essential as it is located near the ABL catalytic domain in the middle of the imatinib bindi ng site and controls access to the hydrophobic region of the enzymatic active site. The i nactive conformation of
39 BCR-ABL causes an auxiliary binding site within the ATP -binding pocket to open, which facilitates imatinib binding by the formation of hydrogen bonds. Interaction with imatinib traps and stabilizes BCR-ABL tyrosine kinase in an inactive, closed conformation, which terminates its kinase activ ity and inhibits its autoand substrate phosphorylation . With has an IC 50 value of 0.5 M, imatinib has been shown to induce apoptosis in BCR-ABL+ cell lines as well as in primary l eukemia cells from CML patients  (see Table 6 for a profile of BCR-ABL inhibitors). Cells undergo apoptosis within 48 to 72 hours following imatinib tr eatment. Chronic phase CML patients who had failed rIFN -therapy and were treated with 400 mg/day of imatinib did not progress t o accelerated or blast crisis phase. Additionally, 95% of achieved complete hematologi c response and 60% had a major cytogenic response (MCR). A long-term follow -up of these patients showed a 13% increase in the MCR rate to 73%, with 63 % of patients having a CCR. Studies done in accelerated-phase CML patients were encouraging, though less dramatic. Eighty two percent of these patient s showed hematologic responses, with MCR seen at 24% and CCR at 17%. Imatinibs efficacy has revolutionized CML patient care ([ 104, 106]. By obstructing BCR-ABLs tyrosine kinase domain and its abili ty to commandeer signal transduction pathways crucial for CML pathogenesi s, imatinib inhibits BCR-ABLs role in cell transformation at micromolar conce ntrations.
40 Yet despite the efficacy of imatinib, CML patients in ad vanced stages of the disease (whose in accelerated or blast crisis phase) eve ntually display inadequate responses to treatment and often experience relapse within a year .
41 Figure 11: Mechanism of Imatinib Action in Chronic Myeloid Leukemia Cells ( A ), CML pathogenesis is mediated by the constitutive acti vation of the BCR-ABL oncoprotein, which binds ATP to phosphorylate and activate numerous substrates involved signal transduction and malignant cell transformation. ( B ), Chemical formula of imatinib (STI-571). ( C ), Imatinib works as a competitive inhibitor and ATP-mimic that selectively competes for and binds to the ATP binding site/tyrosine kinase (TK) domain of BCR-ABL. ( D ), Ribbon representation of the three dimensional crystal structure of the catalytic domain of the ABL tyrosine kinase in complex with imatinib (figure taken from Wikimedia Commons). Imatinib Resistant Chronic Myeloid Leukemia In vitro resistance models have played a critical role in identif ying various mechanisms of primary (acquired) resistance that thwart the effects of imatinib
42 treatment. Patient relapse is associated to three main mechanisms of imatinib resistance: 1) point mutations within BCR-ABLs tyrosine kinase domain that (a) directly interfere with the ability of imatinib to bi nd to BCR-ABL kinase domain, or (b) impair the ability of BCR-ABL to achieve the inact ive conformation required for imatinib binding [108-111]; 2) bcr-ABL gene amplification, and 3) overexpression of drug transporters. Point Mutations within the BCR-ABL Kinase Domain Gorre et al. was the first to demonstrate the relationship between a point mutation at ABL kinase domain and imatinib resistance [1 08]. The most resistant kinase mutation occurs when threonine 315 gatekeeper resi due changes to isoleucine (T315I). This change directly blocks the abili ty of imatinib to bind the kinase domain while still retaining BCR-ABLs capacity to bind ATP and catalyze substrate phosphorylation. The T315I point mutation cau ses a shift in the equilibrium between BCR-ABLs inactive and active states, restoring BCR-ABL kinase activity. This can be attributed to T315I-mutation -specific changes in phosphorylation pattern within the kinase domain. This m utation results in a shift in the phosphorylation of two key tyrosine residues wit hin the ATP binding loop that confers an oncogenic fitness advantage to cells possessin g this mutation . Imatinib mesylate resistance is commonly associated with mut ations that alter the flexibility of the BCR-ABL kinase domain to adopt the inactive conformation needed for imatinib binding. A summary o f some of these well-
43 characterized mutations is provided in Table 7. Like T31 5I, these mutations result from amino acid substitutions and consist of approximat ely 25 amino acids residues distributed throughout the ABL kinase domain that provide varying degrees of imatinib resistance. In some patients with sta ble chronic phase CML a number of these mutations are associated with disease pr ogression to advanced stages . The most commonly characterized mutations tha t account for 60 70% of all kinase mutations affect residues Gly250, Tyr 253, Glu255, Met351 and Phe359.
44 BCR-ABL Gene Amplification Gene amplification of target proteins is often charact eristic of malignant cell transformation and is a frequently used mechanism to generate drug resistant neoplastic cells. Studies show that resistance to imatinib also occurs through BCR-ABL gene amplification. Imatinib-resistant CML cells have displayed higher levels of BCR-ABL protein expression an d phosphorylation when compared to parental, imatinib-sensitive cells due to robust overexpression of BCR-ABL mRNA transcripts . Studies also show BCR-ABL gene amplification linked to the presence of additional copi es of the gene in some CML patients . In other patients, unusual karyotyping using banding cytogenetics coupled with the overexpression of BCR-ABL led to the discovery of multiple double minutes which are small fragments of extrachrom osomal DNA encoding the BCR-ABL gene  (Figure 12). Disease progression from chronic to accelerated or blast cr isis phase in CML patients is associated with an increase in BCR-ABL transcripts during imatinib treatment [108, 117]. This increase in transcr ipts correlates with enhanced CML cell survival determined by their ability to overcome the apoptotic effects of imatinib. These results underscore the fact tha t BCR-ABL gene amplification is an important mechanism of acquired resi stance to imatinib.
45 Figure 12: Gene Amplification Due to the Presence o f Double Minutes FISH showing cells in metaphase. Chromosomes stained in red and double minutes stained in green (left) and all nuclear DNA stained b lack (right). In both panels, the double minutes are indicated by arrow. Image taken from Wikipedia. This work is in the public domain in the United States because it is a work of the United States Federal Government. Overexpression of Drug Transporters Cancer cells exposed to chemotherapeutic agents for an ext ended period of time often develop a multidrug resistance (MDR) ph enotype that is characterized by overexpression of the multidrug resistanc e1(MDR1; also called P-glycoprotein (P-gy), ABCB1 and CD243) efflux pump. T he broad substrate specificity of MDR1 enables it to regulate the distribut ion and bioavailability of numerous drugs. In CML patients this results in an excessive efflux of imatinib, rendering the chemotherapeutic agent ineffective. The importance of MDR1 in imatinib resistance is refle cted in one recent study that demonstrated that polymorphisms of the MDR1 ge ne are indicative of the type of molecular response CML patients may have to im atinib treatment
46 . MDR1. Retroviral-mediated transfection of BCRABL+ cells with the MDR1 gene greatly decreased its sensitivity to the apoptotic e ffects of imatinib, a phenotype that was could be reversed with the addition o f MDR1 pump modulators . Additionally, cell lines resistant to imatinib were capable of growing continuously in its presence due to increased ex pression of the MDR1 gene. Furthermore, Dulucq et. al. also revealed that the presence of MDR-1 polymorphisms correlated to suboptimal patient response to imatinib treatment . Similarly, patients with CML and treated with high dose imatinib displayed drug clearance and drug resistance in a manner consisten t with MDR1-mediated elimination . Nilotinib The need to address imatinib-resistant CML ushered in an active area of research focused on the development of second generation i nhibitors. The observation that imatinib-resistance is predominantly a ssociated with mutations within BCR-ABL kinase domain led to the development of other small-molecule ABL kinase inhibitors with less stringent binding require ments and that are more effective in combating resistance. One such inhibitor is nilotinib (AMN107, Novartis Pharma ceuticals Corporation, Basel, Switzerland) (AMN107; Novartis), an other phenylaminopyrimidine derivative that followed a rati onal drug design based on the crystal structures of imatinib in complexes with ABL. It was developed by modifying the N -methylpiperazine ring within the imatinib molecule. This
47 modification facilitates greater hydrogen-bond interact ions with wild-type and many imatinib-resistant kinase mutants by providing a be tter topological fit to the kinase domain. Clinical studies show that nilotinib is ac tive against 32 of the 33 mutant forms of BCR-ABL [121, 122]. It is approximate ly 30-fold more potent than imatinib as an ABL inhibitor, with an IC 50 value of 30 nM. However, like imatinib, it possesses no significant activity against the T 315I mutant protein. Nilotinib is effective in patients who are unresponsive to treatment with imatinib. When CML patients in accelerated-phase were treated with nilotinib they experienced a hematologic response rate of 47% and a MCR rate of 29%, with a 12-month survival rate of 79% . Clinical st udies show that nilotinib is also well-tolerated at doses of up to 600 mg daily [124 ]. Dasatinib The need to address imatinib-resistant CML also resulte d in the development of dasatinib (BMS-354825; Bristol-Myers Squi bb), a dual BCRABL/SRC family kinase inhibitor (SFK: Fyn, Yes, SRC, Lyn Lck, Fgr, Blk, Hck). It also has additional activity against c-Kit and PDGFR. Da satinib is clinically more active than imatinib and nilotinib against both wild-t ype and mutant BCR-ABL proteins responsible for drug resistance [121, 125]. Dasatinib has less stringent conformational requirements for interaction with BCR-ABL kinase domain than nilotinib. Consequentl y, it binds both the active and inactive conformations. Dasatinib induces complete hematologic remissions and major cytogenetic responses in imatinib-resi stant CML patients in
48 all three phases, chronic, accelerated, and blast phases. Re sponse rates show that 95% of patients with chronic-phase CML and 82% o f patients in acceleratedphase CML respond to dasatinib treatment during a >12 months median followup timeframe . However, like imatinib and niloti nib, dasatinib has no effect against the T315I mutation and nearly all blast crisis p hase patients experience relapse in less than a year of treatment
49 CHAPTER V MINIMAL RESIDUAL DISEASE CHRONIC MYELOID LEUKEMIA Despite the efficacy of imatinib, nilotinib and dasatini b in achieving high rates of cytogenetic and hematological responses to varyin g degrees, they are unable to override all forms of drug-resistant kinase m utations. Overtime some CML patients, particularly those in advanced stages of t he disease, become refractory to further treatment. In almost all patient s the BCR-ABL transcript persists below the level of cytogenetic detection while un dergoing treatment, indicating the presence of minimal residual disease (MR D) CML. This suggests that treatment with imatinib and more potent BCR-AB L inhibitors is not sufficient to eradicate this disease. Several factors have been implicated in contributing to M DR CML: 1) the failure to kill leukemic stem cells, 2) BCR-ABL gene amplification, 3) the presence of pre-existing BCR-ABL kinase mutations, 4) epi genetic changes, and 5) the tumor microenvironment. Failure to Kill Leukemic Stem Cells The identification of a rare population of primitive quiescent BCR-ABL+ cells with an intrinsic insensitivity to imatinib treatm ent lends support this model [127-129]. In vitro studies of quiescent CD34+ leukemic cells revealed that th ey
50 were resilient against high-dose imatinib treatment of up to 10-folds higher than the normal dosage. Additionally, while imatinib supp ressed the proliferation of leukemic colony-forming cells (CFCs) and long-term culture -initiating cells (LTCICs) it did not induce apoptosis in these primitive cell s. Furthermore, residual BCR-ABL+ stem cells persisted in the bone marrow of CML patients who had achieved CCR with imatinib. Th is indicates that primitive BCR-ABL+ CML cells that are capable of enter ing the cell cycle could repopulate the bone marrow compartment to facilitate patient relapse when imatinib is withdrawn. The insensitivity of leukemic stem cells to the apoptotic e ffects of imatinib, coupled with their self-renewing capacity are importan t contributing factors to MRD CML. Pre-existing Mutations within the BCR-ABL Tyrosine Kinase Domain In one study 12% of CML patients who had no cytogenetic r esponse to imatinib therapy and who did not display bcr-ABL gene amplification possessed rare cells that had point mutations within the kinase d omain at the time of diagnosis. This included the highly resistant Thr315Ile mutation . Additionally, BCR-ABL+ imatinib-naive cells from CML patients possessed the Glu255Lys point mutation, which is involved in preventing the BCR-ABL kinase from adopting the inactive conformation needed for im atinib interaction . Pre-existing mutations in BCR-ABL+ cells could facilitat e their outgrowth during therapy owing to the selective pressure of imat inib. The fitness
51 advantage conferred to these mutated cells would enable a clonal selection of the minor population of cells carrying the mutation. Epigenetic Contributions While the t(9;22) translocation that produces the BCRABL oncogenic tyrosine kinase is the major transforming even in CML, additional molecular and/or epigenetic changes have been implicated in CML oncogenesis. These changes are also associated with the transition from chro nic to blast crisis phase that accompanies the emergence of MRD CML and patient r elapse. Disease progression in CML patients has been linked to in crease de novo DNA-methylation of the ABL promoter contained within a CpG island, which is associated with ABL gene silencing [132, 133]. This suggests that the course of the disease may be documented by the extent of CpG methyl ation. During chronic phase CML the expanded granulocytic linea ge still retains its ability to differentiate. However, the accelerated a nd blast crisis phases are characterized by decreased cell differentiation and the s ubsequent accumulation of immature hematopoietic cells. Molecular changes result ing in differentiation arrest are associated with this transition. Gain-of-fun ction mutations in GATA-2, a gene that serves as a negative regulator of hematopoieti c stem progenitor cell differentiation, resulted in its enhanced transactivation capacity and augmented inhibition of hematopoietic cell differentiation [134 ].
52 Tumor Microenvironment The bone marrow microenvironment (BME) provides CML cel ls direct interaction with stromal cells and extracellular matrices (ECM), which may modulate drug response. We reported previously that adh esion to fibronectin was sufficient to protect K562 CML cells from imatinib-ind uced cell death (Damiano [135, 136]. Similar studies reveal that direct contact of leukemic cells with the human stromal cell line, HS-5, significantly increased cell proliferation, viability and colony formation . Additionally, leukemic-stroma l cell interactions enhanced in vitro leukemic cell survival while attenuating chemotherapy-in duced cell killing. Using a co-culture model consisting of le ukemic cells and bone marrow-derived stromal cells, studies show that this inter action resulted in increased activation of the several signal transduction pa thways involved in cell survival, including PI3K/AKT, Ras/ERK and STAT3 . The BME is also a rich source of cytokines and growth factor s that could also influence drug response. While the retroviral tran sduction of BCR-ABL into severe combined immunodeficient (SCID) bone marrow pro genitor cells grown healthy mice resulted in rapid, fatal, imatinib-resistan t leukemia, its transduction into similar cells lacking the cytokine receptor common -chain resulted in imatinib-sensitivity. This suggests that cytokines within t he microenvironment are responsible for conferring imatinib resistance . S tudies to identify cytokines involved in modulating imatinib sensitivity revealed that autocrine secretion of GM-CSF was involved. Its secretion activates the JAK/STAT5 p athway and
53 protects imatinib-naive and CML progenitor cells from de ath induce by BCR-ABL inhibitors . Additionally, mutated CML cells were able to spread resistance to non-mutated cells through overexpression of IL-3, wh ich resulted in the activation of the Ras/MEK and JAK2/STAT5 signal transdu ction pathways . This highlights the importance of the bone marrow micr oenvironment in modulating imatinib response. This resistance can be confe rred by either direct adhesion to bone marrow stroma or by the secretion of b one-marrow derived soluble factors.
54 CHAPTER VI THE ROLE SIGNAL TRANSDUCERS AND ACTIVATORS OF TRANSCRIPTION (STATS) IN CHRONIC MYELOID LEUKEMIA ONCOGENESIS AND DRUG RESISTANCE Receptor engagement by hematopoietic cytokines results in their phosphorylation and the activation of JAKs. These receptors consist of a unique ligand-binding subunit and a signal transducing subuni t  (see Figure 13 for cytokine receptor families). STATs are activated transientl y by numerous cytokines via JAKs , as well as SFKs . Various types of hematologic malignancies are associated with the aberrant, constitutive, cytokine-mediated activation of STAT proteins. In CML, BCR-ABL activates signal transduction pathways shared by man y hematopoietic cytokines, including the STAT3 and STAT5 signaling path ways.
55 Figure 13: Cytokine-Receptor Families Cytokine-receptors consist of a ligandbinding subunit and a signal-transducing subunit. Hem atopoietic cytokine receptors can be categorized based on their signaling subu nit as follows: 1) those that signal through a single subunit, which incl udes growth hormone (GH), prolactin (PRL), erythropoietin (EPO); granulocyte colon y-stimulating factor (GCSF), and thrombopoietin )THPO); 2) those that signal through the gp130 or gp 130-related subunit, which includes interleukins (IL)-6, 11 and 12, leukemia inhibitory factor (LIF), oncostatin M, ciliary neurotro phic factor (CTNF), and cardiotropin-1 (CT-1); 3). those that signal through t he gp140 -subunit or a ligand-binding subunit, which includes IL-3 and 5, and granulocyte-macr ophage colony-stimulating factor (GM-CSF); 4) those that signa l through a common chain ( C ) and a or ligand-binding subunits, which includes IL-2,4,7,9, a nd 15; 5) those that signal through two more subunits, such as t he interferon(IFN) subfamily of receptors, which includes IFNRI and II.
56 The Role of STAT3 STAT3 governs signal transduction in growth factor-media ted control of hematopoiesis and myeloid cell differentiation. Its acti vation up-regulates the expression of genes associated with cell survival and proli feration including Bcl-x, Mcl-1, and cyclin D1 [89, 145, 146]. Persistent STAT3 a ctivation has been detected in a variety of hematopoietic malignancies and so lid tumors [147-149]. The enforced expression of BCR-ABL in primitive embryoni c stem cells (ESCs) resulted in their persistent STAT3 activation and cause d the cells to retain their primitive morphology by inhibiting their differentiati on . The increased activation of STAT3 in primary leukemia blast cells lends su pport to its involvement in disease progression of CML patients and th eir transition from chronic phase to blast crisis phase . Taken together, these observations highlight the role o f amplified STAT3 activation in the pathogenesis of CML and underscore its involvement in imatinib resistance. The Role of STAT5 STAT5 has gained prominence because it is often constitu tively activated in a wide variety of human cancers, ranging from solid tu mors seen prostate, lung, breast and colon cancers, to hematologic malignanci es, such as myeloproliferative disorders [152-155]. It is activated b y non-receptor tyrosine kinases, such as BCR-ABL and SFKs. Like STAT3, STAT5 activa tion upregulates the expression of Bcl-x, Mcl-1, and cyclin D1.
57 BCR-ABL-dependent STAT5 activation was observed in severa l leukemic cell lines possessing the Ph+ clone, but not untransforme d cells or BCR-ABLnegative leukemia cells [154, 156]. Furthermore, the e xpression of BCR-ABL in IL-3/GM-CSF-dependent leukemic cells resulted in aberra nt STAT5 activation and growth factor independence . Similarly, constit utive activation of STAT5 by the SFK, Hyk, resulted in cytokine-independent cell gro wth in cytokinedependent leukemic cells . STAT5 activation is also a ssociated with chronicto-blast crisis phase transition in leukemia patients. A s indicated previously, in up to 95% of AML blast cells constitutive activation of STAT3 and STAT5 could be detected . Interestingly, increased STAT5 activation in BCR-ABL-posi tive leukemic cells and STAT5s ability to confer resistance to dasatini b has been shown to be directly correlated with leukemic cell density . This recent study revealed that an increase in the density of leukemic cells resulted in a decrease in dasatinibs efficacy against STAT5-mediated signaling. Re sults indicate that this observation may attribute to the upregulation of STAT5 -regulated anti-apoptotic proteins. Taken together, this demonstrates that constitutive STAT5 activation promotes CML pathogenesis and imatinib resistance by upregulating the expression of anti-apoptotic proteins, Bcl-x, Mcl-1 and cyclin D1. Additionally, enhanced STAT5 activation is capable of mediating cytokine -independent growth in leukemic cells, leading to increased cell survival.
58 CHAPTER VII OBJECTIVES The BCR-ABL fusion gene is critical for the development of CML. Treatment with imatinib is highly effective against CM L; however treatment response in patients is compromised by the emergence of ima tinib resistance. In vitro resistance models often fail to consider the role of t he tumor microenvironment in the emergence of imatinib resistance The predominant microenvironment of CML is the bone marrow, a rich source of both soluble factors and extracellular matrixes which may influence dru g response. This study was developed to address the influence of the bone marro w microenvironment on imatinib sensitivity. We utilized an in vitro co-culture bone marrow stroma model to demonstrate that stable soluble factors secreted by the HS-5 human bone marrow stroma cell line were sufficient to protect K562 CML cells against imatinib-induced cell death. Subsequently, we observed t hat HS-5-derived conditioned media (CM) also provided protection against cell death. Previous studies showed that HS-5 stromal cells secrete cytokines and growth factors involved in the growth and differentiation of hematop oietic stem cells . Some of these cytokines are capable of activating signal trans duction pathways involved in myeloid differentiation. Investigations into the activation of survival
59 pathways that reconstitute BCR-ABL signaling in CM may p rovide novel insights into the mechanism of environment-mediated drug resistan ce (EMDR). One such pathway being investigated is the signal transducer and activator of transcription 3 (STAT3) survival pathway. A review of the current lit erature led to the following hypothesis: Bone marrow-derived soluble factors in conditioned media contribute to the failure of BCR-ABL inhibitors to eradicate mini mal residual disease CML. In order to refute or support this hypothesis, the foll owing objectives were formulated: 1. To determine if HS-5-derived CM protects CML cell li nes from death induced by second generation BCR-ABL inhibitors 2. To determine whether HS-5-derived CM increased the proliferation and clonogenic survival of CML cells 3. To determine whether HS-5-derived CM mediate the activation of STAT3 and STAT5 cell survival pathways in CML cells
60 CHAPTER VIII MATERIALS AND METHODS Cell Cultures Human CML K562 and KU812 cells (obtained from the Ame rican Type Culture Collection) were cultured in RPMI 1640 supplemented wi th 10% fetal bovine serum, 1% L-glutamine, and penicillin/streptomycin [reg ular medium (RM); Life Technologies] at 37C in 5% CO 2 in a humidified incubator. The human stromal cell line HS-5 (also obtained from the American Type Cu lture Collection) was maintained under the same conditions. Generation of Conditioned Media (CM) CM was generated by culturing 5 x 10 5 HS-5 cells/mL in RM overnight in a humidified atmosphere at 37C with 5% CO 2 to achieve 75% to 80% confluency. Subsequently, the medium was removed and the cells were i ncubated in fresh medium for 3 hours. The supernatant was then collected and cleared of contaminating cells by centrifuging at 2,000 rpm for 5 m inutes. Supernatant was aliquoted and stored at -80C for future use.
61 Drugs and Reagents Imatinib mesylate and nilotinib (AMN107), supplied by N ovartis Pharma, were dissolved in DMSO as a 10 mM stock solution and stored in al iquots at -20C. Dasatinib, supplied by Bristol-Myers Squibb, was treated s imilarly. Treatment with Tyrosine Kinase Inhibitors Preparation of Lysates and Western Blotting K562 or KU812 cells were cultured at a density of 4 x 1 0 5 /mL in either RM or CM for various time points in the presence or absence of eit her a matched vehicle control (DMSO) or drug (imatinib or dasatinib). Subseq uently, cells were pelleted and lysed in radioimmunoprecipitation assay lysis buffer supp lemented with phosphatase and protease inhibitors (1 mM Na 3 VO 4 1 mM NaF, 1 mM phenylmethylsulfonyl fluoride, 10 g/mL leupeptin, an d 10 g/mL aprotinin). Insoluble materials were removed by centrifugation at 4 C for 20 minutes at 15,000 rcf Antibodies used for Western blotting were anti-STAT3, anti-STAT3 (Tyr705), anti-Bcl-xl, anti-Mcl-1, anti-Survivin, and p-SRC (all from Cell Signaling Technologies). Apoptosis Assay Apoptosis was measured using the Annexin V-FITC apopto sis detection kit (Alexis Biochemicals) according to the manufacturers recomm endations. Data were acquired using CellQuest Pro software version 4.0. 2 (BD Biosciences) and analyzed using FlowJo software version 7.2.2 (Tree Star ).
62 siRNA Transfection K562 cells cultured in RM were transiently transfected w ith either STAT3 siRNA (siGENOME ON-TARGETplus SMARTpool reagent; Dharmacon) or a control siRNA (ONTARGETplus siCONTROL Nontargeting reagent; D harmacon) using Amaxa Nucleofector methodology (Amaxa). Briefly, 1 x 1 0 6 K562 cells were transfected per manufacturers instructions. Thirty-six hou rs post-transfection, the process was repeated to ensure continued suppression of STA T3 levels. Transfected K562 cells (2.0 x 10 5 /mL) were then cultured in RM or CM for 3 hours then treated with 250 nM imatinib or a vehicle control (DMSO). Cell death was measured using Annexin V apoptosis assay as described pr eviously. Bromodeoxyuridine (BrdU) Antibody Staining 2.5 x 10 5 K562 cells/mL were cultured in RM or CM for 24 hours t hen pulsed with 30 g/mL 5-bromo-2-deoxyuridine (BrdU) for 30 minute s. Cells were fixed with ethanol and BrdU incorporation was detected using FITCconjugated anti-BrdU antibody (Calbiochem, Gibbstown, NJ). Data acquisition a nd analysis were done using fluorescence-activated cell sorting (FACS). Clonogenic Assay K562 cells were diluted to a concentration of 10,000 cel ls/mL in RM or CM for 6 hours in the absence or presence of imatinib and incubat ed at 37C with 5% CO 2 Cells are transferred to 0.3% final agar solution reconstituted in RM or CM. Once the agar solidifies, cells were allowed to incubate for 8-10 additional days. Cell colonies (>50 cells) were then counted on 2 mm g rid culture dish (Corning).
63 Preparation of Nuclear Extracts Nuclear extracts were prepared by resuspending 5 x 106 K5 62 cells in hypotonic buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KC l, 0.5 mM dithiothreitol, 0.2 mM PMSF, 2 g/mL leupeptin, 0.15 g/mL aprotini n) followed by incubation on ice 5 minutes. Cells were centrifuged for 10 seconds a t 14000 rpm and washed once with hypotonic buffer. The nuclei were collect ed by centrifugation for 5 minutes at 14000 rpm and washed once with hypoton ic buffer. The nuclear pellet was resuspended in 50 L of hypertonic buffer (2 0 mM HEPES-KOH, pH 7.9, 25% glycerol, 450 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM dithiothreitol, 0.5 mM PMSF, 2 g/mL leupeptin, 0.1 5 g/mL aprotinin), incubated for 30 minutes at 4C followed by centrifugation for 2 minutes at 14000 rpm. Electrophoretic Mobility Shift Assay (EMSA) A total of 2 L nuclear extract was incubated with a d ouble-stranded 32P-labeled oligonucleotide (1 ng) sis-inducible element (SIE), (5TCGAGTATTTCCCAGAAAAGGAACAGCT-3 and its complement) i n 10 L binding buffer (25 mM HEPES, pH 7.9, 100 M ethylene glycol tetraacetic acid (EGTA), 200 M MgCl2, 500 M dithiothreitol, 1 g/mL BSA, 0.2 g/mL poly dI:dC) for 15 minutes at room temperature. Protein-D NA complexes were detected by autoradiography and quantified using a phosp hoimager. MTT Assay. Incubate 50,000 K562 cells/ mL for 3 hours in either re gular media or conditioned media. Add varying concentrations of imatinib. Incubate drug-treated cells for 72
64 hours at 37C, 5% CO 2 Add MTT dye and incubate for 3 hours at 37C, 5% CO 2 Read absorbance at 540 nm on a spectrophotometer. Statistical Analysis A preliminary examination of the data was done using descriptive summary analysis and the Anderson-Darling statistic. Significance testing of the doseresponse apoptosis assays was done using analysis of covariance. The Students t test was used for the siRNA experiments to co mpare the logtransformed fold change values, with the null hypothesi s being a log (fold change) = 0 (or conversely the fold-change = 1). A signifi cance level of 0.05 was considered statistically significant for all tests. The geom etric means of the four ratios for siRNA experiments and the 95% confidence int ervals were also calculated
65 RESULTS Co-culture Bone Marrow Stromal Model Protects K562 CML Cells from Imatinib-Induced Apoptosis Previous work done by Torok-Storb et. al. revealed that HS-5 human stromal cells are capable of producing cytokines and growt h factors that are involved in the growth and differentiation of HSC and are also capable of supporting the ex vivo expansion of both immature and mature progenitor ce lls . To test our hypothesis that the bone marrow mi croenvironment contributes to the failure of BCR-ABL inhibitors to eradicate min imal residual disease CML, we utilized a co-culture bone marrow microenvironment tr answell model system and took advantage of HS-5 stroma cells ability to suppor t HSC and their progenitors. We wanted to determine whether the bone marrow microenvironment protects the K562 CML cell line, which o riginated from the pleural effusion of a CML female patient in termina l blast crises phase, from death induced by imatinib. Using this co-culture model, we first wanted to delineat e the contributions of (1) HS-5-derived soluble factors alone, and (2) dire ct adhesion to HS-5 cells on imatinib resistance in K562 CML cells treated with 1 M of imatinib (Figure 14). Our control samples consisted of K562 cells cultured in regular media (RM).
66 Here we saw approximately 35% cell death. Next, we det ermined the effects of transiently secreted soluble factors on K562 cells culture d in the upper chamber of the transwell (TSF), and here we saw protection fro m cell death at approximately 10%. We then examined the effects that d irect adhesion to HS-5 stroma cells as well as secreted soluble factors on imatin ib-induced K562 cell death (SF+A). Here we observed an even greater protect ion from cell death at 5%. Lastly, we examined the effects of HS-5-derived stab le soluble factors alone on apoptosis by using HS-5 conditioned media (CM). Her e we saw 15% cell death. Taken together, these data revealed that the coculture model was not required for the production of the protective soluble f actors produced by HS-5. It also revealed that HS-5 conditioned media contained stab le soluble factors that provided protection against imatinib-induced cell de ath. Therefore, to simplify our bone marrow stroma model, we utilized HS-5-derived cond itioned media in our subsequent experiments.
67 Figure 14: Co-culture Bone Marrow Transwell Stromal Model ( A ) Co-culture transwell model system of K562 chronic myeloid leukemia (CM L) cultured in: i) regular media (RM: RPMI + 10% FBS, 1% Penicillin /S treptomycin); ii) the upper chamber of the transwell system with HS-5-secreting trans ient soluble factors (TSF) or with direct adhesion to HS-5 stroma cells (SF+ A); iii) .HS-5-derived conditioned media (CM: RM collected after 3-hour incuba tion with HS-5 cells). ( B ) Annexin V apoptosis assay of conditions described in A. Characterizing Conditioned Media Collecting Conditioned Media beyond 3 Hours does no t Provide Greater Protection against Imatinib-Induced Cell Death To verify whether 3-hour HS-5-derived conditioned medi a provided the best protection against imatinib-induced cell death, we generated the conditioned media as described before, however, instead of only colle cting at 3 hours it was collected at varying time-points: 1, 3, 6 and 24 hours.
68 As seen in Figure 15, our data revealed that condition ed media generated beyond 3 hours did not provide greater protection fro m death induced by 500 nM of imatinib. Figure 15: The Effects of Conditioned Media Collect ed at Various Times on Imatinib-Induced Cell Death in K562 CML Cells. HS-5 stromal cells cultured to 75-80% confluency in regular media were incubated for 1, 3, 6, 24 hours in fresh media. Cell death was measured using the Annexin V apo ptosis assay following treatment with 500 nM of imatinib for 48 hours. Resu lts were analyzed using flow cytometry. Conditioned Media Stored for up to One Week Still P rovided Protection against Imatinib-Induced Cell Death To determine the effects of storage on the protective characteristic of 3 hour-derived conditioned media, we cultured K562 cells in regular media and
69 conditioned media that was freshly collected, one day and one week old. Oneday and one-week old conditioned media were stored at 80C. After treating K452 cells with 500 nM of imatinib, ou r results were analyzed with Annexin V apoptosis assay. Figure 16 reveal s that 3-hour-derived conditioned media that was stored at -80C for up to o ne week could still protect K562 CML cells from death induced by imatinib. This confi rmed that HS-5derived soluble factor(s) in the conditioned media is s table and does not have a short half-life. To ensure consistency and preserve the in tegrity of our data, we did not store conditioned media for longer than one w eek. Figure 16: The Effects of Storage on 3-Hour-Derived Conditioned Media. To make the time-dependent conditioned media, HS-5 stroma l cells were grown to 75-80% confluency in regular media (RM) for 3 hours. Me dia was then collected and used fresh, after one day of storage at -80C or af ter one week of storage. Cell death was measured using the Annexin V apoptosis a ssay following
70 treatment with 500 nM of imatinib for 48 hours. Resu lts were again analyzed using flow cytometry. Serum is not Required for Production of the Protect ive Soluble Factor(s) Found in HS-5-Derived Conditioned Media We next wanted to determine if HS-5 human stromal ce lls require serum to produce the activate component in conditioned media. To answer this question we examined whether serum-free conditioned media (SFCM) can still protect K562 cells from apoptosis induced by imatinib. HS-5 strom al cells were allowed to incubate for 3 hours in regular media void of serum This SF-CM was then collected and supplemented with 10% FBS then utilized in our experiments. As before, K562 cells were cultured in regular media (RM) conditioned media (CM) and SF-CM. Our results were analyzed using the Annexin V apoptosis assay followed by flow cytometry analysis. As seen in Figure 17, HS-5 stromal cells were capable of producing their protective soluble factor(s) in the absence of FBS. Add itionally, since media containing serum is more difficult to fractionate, these data demonstrates that future studies to identify the protective component of con ditioned media using fractionation can be readily accomplished with this SF-CM
71 Figure 17: The Effects of Serum-Free Conditioned Me dia on ImatinibInduced K562 Cell Death. To make the time-dependent serum-free (SF) conditioned media, HS-5 stromal cells were grown to 7580% confluency in regular media (RM) for 3 hours. Media was then collecte d and 10% fetal calf serum (FBS) was added. SF-CM was used fresh, after one d ay of storage at 80C or after one week of storage. Cell death was measu red using the Annexin V apoptosis assay following treatment with 500 nM of ima tinib for 48 hours. Results were again analyzed using FACS. Conditioned Media does not Convey a Growth Advantag e to K562 CML Cells Given our previous results which show that conditioned m edia protects K562 cells from death induced by imatinib within our bo ne marrow stromal model, we decided to investigate whether conditioned is capable of increasing the rate of K562 cell proliferation and DNA synthesi s.
72 Using trypan blue exclusion staining method, Figure 18A shows that when cells were counted over the course of six days, there was no increase in cell proliferation in K562 cells regardless of culture condi tions. To verify these results, the rate of DNA synthesis, which is i ndicative of cell proliferation, was also measured via BrdU incorpora tion staining and detection with a FITC-conjugated anti-BrdU antibody. F igure 18B shows that DNA synthesis is also not increased in conditioned media. Figure 18C is a representative of three independent BrdU incorporation staining assays.
74 Figure 18: The Effects of Conditioned Media on K562 Cell Proliferation and DNA Synthesis. ( A ) 50,000 K562 cells were cultured in regular media or conditioned media and total cell count was determined e very 24 hours using trypan blue exclusion staining. Results and standard d eviation represents the average of 4 independent experiments. ( B ) The rate of DNA synthesis was determined by 5-bromo-2-deoxyuridine (BrdU) antibody st aining. 2.5 x 10 5 K562 cells/mL were cultured in regular media or conditioned media for 24 hours then pulsed with 30 g/mL of BrdU for 30 minutes. Cells wer e then fixed with ethanol and BrdU incorporation was detected using FITC-conjugat ed anti-BrdU antibody. Data acquisition and analysis were done using FACS. ( C ) Representative figure of BrdU incorporation staining (n=3). Heat-Inactivated Conditioned Media does not Protect against Imatinib Mesylate-Induced Cell Death Cytokines are proteins that initiate cellular communicatio n. To determine if the HS-5-derived protective soluble factor(s) found in con ditioned media is a cytokine we cultivated K562 CML cells in regular media and heat-inactivated
75 conditioned media, treated them with 500 nM imatinib for 48 hours then performed an Annexin V apoptosis to determine cell de ath. Results were analyzed via FACS. Figure 19 shows that when K562 CML cells were cultured i n heatinactivated conditioned media there was no protection fr om death induced by imatinib. This demonstrates that the protective soluble f actor in HS-5 conditioned media is a protein. 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 500 nM IM% Imatinib-specific apoptosis RM CM CM-Heat Figure 19: Heat-Inactivated Conditioned Media does not Protect K562 Cells Against Imatinib-Induced Cell Death. K562 cells were cultured for 3 hours in either regular media or conditioned media that was he at-inactivated at 95C for 10 minutes. Cells were treated with 500 nM imatinib o r 0.1% DMSO for 48 hours. Cell death was measured using Annexin V apoptosis assay fo llowed by fluorescence-activated cell sorting analysis.
76 HS-5-Derived Conditioned Media Protects K562 and KU 812 CML Cell Lines from Death Induced by Imatinib Mesylate Results from our co-culture bone marrow stromal model ha d revealed that HS-5 conditioned media alone can also protect K562 CM L cells from imatinibinduced cell death. We wanted to determine whether t he effects we saw was cell line specific by doing a dose response profile. To addre ss this question, we performed an imatinib-dose response profile using the C ML cell lines, K562 cells and KU812, which is a pre-basophilic cell line that ori ginated from the peripheral blood of a CML patient in blast crisis phase. Cells were cultured in regular media or conditioned and treated with increasing doses of im atinib (0, 125, 250, 500 nM imatinib) for 48 hours. The results of the Annexin V apoptosis revealed that ex posing either K562 or KU182 cells to conditioned media for 3 hours before drug treatment was sufficient to inhibit apoptosis induced by imatinib (F igures 20A and B) and suggested that the protective effects of conditioned medi a is not cell-line specific. Additionally, we performed a clonogenic assay, or colony fo rmation assay, to determine whether conditioned media increased the ability of K562 cells to divide and form colonies. The clonogenic assay is an in vitro cell survival assay based on the ability of a single cell to grow into a co lony. It is a measure of cancer cells ability to repopulate and form colonies in semisolid media. The colony is defined to consist of at least 50 cells.
77 First, we wanted to determine the optimal K562 cell c oncentration to use for the clonogenic assay. To determine this, we performe d clonogenic assays using varying concentrations of K562 cells to obtain a cel l-density profile. As shown in Figure 20C, in the absence of imatinib, cond itioned media did not increase the clonogenic survival of K562 cells across the di ffering cell concentrations indicated. Additionally, in the absence of imatinib, the initial cell concentrations of 2,500 and 5,000 cells provided too few colonies at the end of 10 days to justify the use of these concentrations. Therefor e, we utilized the initial cell concentration of 10,000 cells in our subsequent clon ogenic assays to ensure that there would be enough viable colonies to count, e ven after high-dose imatinib treatment. Figure 20D shows that when K562 cells are treated with varying concentrations of imatinib, the clonogenic survival of tho se cells cultured in conditioned media was increased across all imatinib concen trations when compared to those control cells cultured in regular media Taken together, these data indicate that not only does conditioned media prot ect from imatinib-induced apoptosis or cell death but also results in a higher pe rcentage of cells capable of dividing and repopulating. Finally, as shown in Figure 20E, removal of K562 cells fr om conditioned media into regular media partially reverses resistance t o imatinib-induced cell death. This suggests that imatinib-resistance in CML cells occurs in the presence
78 of HS-derived soluble factors and in the absence of the soluble factors CML cells become re-sensitized to the apoptotic effects of imatinib
79 Figure 20: HS-5-Derived Conditioned Media Protects K562 and KU182 CML Cells from Death Induced by Imatinib. ( A ) K562 or ( B ) KU182 cells were cultured in either regular media or conditioned media for 3 h and before the addition of various concentrations of imatinib or 0.1% DMSO for 36 hours. Cell
80 death was measured using Annexin V apoptosis assay followe d by fluorescenceactivated cell sorting analysis. % Drug-specific apoptosis was calculated by subtracting the background cell death in control DMSO-tr eated cells from drug treated cell death. A Conditioned media significantly protects K562 cells fr om imatinib-mediated cell death (P < 0.05, analysis of co variance). B Conditioned media significantly protects KU182 cells from imatinib-med iated cell death (P < 0.05, analysis of covariance). C Cell-density profile of K562 cells shows similar clonogenic survival in regular media and conditioned med ia. K562 cells were cultured in regular media or conditioned media withou t imatinib. After 6 hours incubation at 37C, 5% CO 2 cells were cultured in 0.3% agar made of regular media or conditioned media and containing their respect ive imatinib concentrations. Cells were allowed to incubate at 37C, 5% CO 2 for 10 days. Cell colonies (>50 cells) were counted. Representative figure in triplicate (n = 3 independent experiments). The starting concentration o f 10,000 cells was used in subsequent clonogenic assays. D Conditioned media increases the clonogenic survival of K562 CML cells treated with imatinib. K562 cells were cultured in regular media or conditioned media and treated with increasing concentrations of imatinib as depicted. After 6 hours incubation at 37C 5% CO 2 cells were cultured in 0.3% agar made of regular media or condit ioned media and containing their respective imatinib concentrations. Cells were allowed to incubate at 37C, 5% CO 2 for 10 days. Cell colonies (>50 cells) were counted. Representative figure in triplicate (n = 3 independen t experiments). E Removal of conditioned media partially reverses resistance to ima tinib (P < 0.05, Student t test). K562 cells were grown in either conditioned me dia or regular media for 3 hours. Following 3 hours, appropriate samples were remo ved from conditioned media and placed in regular media for an additional 3 hours. Cells were treated with 500 nmol/L imatinib for 36 hours and apoptosis w as measured by Annexin V positivity. Representative figure of experiments done i n triplicates.
81 Conditioned Media from Non-Stromal Cell Lines do no t Protect K562 CML Cells from Death Induced by Imatinib Mesylate Some cells are capable of secreting their own soluble f actors via an autocrine loop and condition their own media. We inves tigated whether the protection against cell death that we observed using HS -5-derived conditioned media was specific to this human stromal cell type. To add ress this, we generated conditioned media from several other non-stro mal cells. K562 CML cells, U937 lymphoma cells and 8226 myeloma cells were incub ated in regular media for three hours, after which their conditioned m edia was collected. To perform the experiment, K562 cells were cultured for t hree hours in regular media and the conditioned media of the different cell lines previously described. The cells were then treated with 500 nM of imatinib f or 48 hours. An Annexin V apoptosis assay was performed followed by flow cytometry an alysis. As you can see in Figure 21, there was no protection from imatinibinduced cell death in K562 cells cultured in its own cond itioned media, as well as in conditioned media from U937 and 8226 cell lines. Th is suggests that the secreted, protective soluble factor(s) responsible for confe rring the imatinibresistant phenotype are specific to bone marrow stromal cells.
82 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 0100200300400500600 Imatinib (nM)% Imatinib-specific apoptosis RM CM CM-8226 CM-U937 CM-K562 Figure 21: Conditioned Media from Non-Stroma Cell L ines does not Protect K562 CML Cells from Death Induced by Imatinib. K562 cells were cultured for 3 hours in either regular media or conditioned media generated from K562, U937 and 8226 cell lines. Cells were treated with 500 nM im atinib or 0.1% DMSO for 48 hours. Cell death was measured using Annexin V apo ptosis assay followed by fluorescence-activated cell sorting analysis. Conditioned Media Protects K562 CML Cells from Deat h Induced by 2nd Generation BCR-ABL Inhibitors, Nilotinib and Dasati nib One of the shortcomings of imatinib therapy stems from th e fact that, over the course of treatment, many patients develop resistan ce which results in relapse and disease progression. Studies have shown that imatinib resistance is
83 often caused by point mutations in the BCR-ABL tyrosine kinase domain [113, 161]. This discovery has led to the design and developme nt of more potent, second generation kinase inhibitors that target imatini b-resistant BCR-ABL mutants. Dasatinib (Sprycel, BMS-354825, Bristol-Myers Squ ibb) and nilotinib (AMN107) represent two potent second-generation BCR-AB L inhibitors, both of which are more potent than imatinib and show considerab le efficacy against most of the well-characterized BCR-ABL mutants. Both imatini b and nilotinib bind to the inactive conformation of BCR-ABL, although nilotin ib was designed to provide a better topological fit to the enzymes tyrosine kinase d omain and binds with approximately thirty times higher affinity than imatini b. In contrast, dasatinib binds both the active and inactive conformations of BCR-ABL kina se due to its less stringent conformational requirements for binding and it is a less selective kinase inhibitor, targeting SRC kinase family members as well. It is approximately 300 times potent than imatinib against wild-type BCR-ABL e xpressing cells and, therefore, inhibits BCR-ABL kinase activity at low-nano molar concentration (between 0.75 to 1.0 nM). As shown in Figures 22A and B, conditioned media protect s from both nilotiniband dasatinib-induced cell death. These dat a indicate that the development of more potent BCR-ABL inhibitors will no t circumvent resistance associated with exposure of CML cells to soluble factors pro duced by the bone marrow microenvironment.
84 Figure 22 : HS-5-Derived Conditioned Media Protects K562 CML Ce lls from Death Induced by 2nd Generation BCR-ABL Inhibitors, Nilotinib And Dasatinib K562 cells were allowed to incubate for either 48 ho urs (nilotinib) or 24 hours (dasatinib). Conditioned media significantl y protects K562 cells from (A) nilotinib-mediated cell death (P < 0.05, analysis of co variance) and (B) dasatinibmediated cell death (P < 0.05, analysis of covariance). Representative graph in triplicates (n = 2 independent experiments) Conditioned Media Activates STAT3 in K562 and KU812 CML Cell Lines Signal transducers and activators of transcription (STATs) are members of a family of transcription factors that were originally characterized as mediators of cytokineand growth factor-induced signaling. Many of th e growth factors and cytokines reported to be expressed in the HS-5 cell line a re known activators of STAT3 and STAT5, including granulocyte-macrophage colon y-stimulating factor, G-CSF, interleukin-6, and vascular endothelial growth factor [162-167]. Thus, to
85 determine the mechanism(s) of drug resistance observed in conditioned media, we decided to exam the roles of both STAT3 and STAT5 within our bone marrow stromal model. We first examined the effects of conditio ned media on STAT3 activation. We wanted to determine whether culturing K562 cells in conditioned media caused an increase in the phosphorylation of STAT 3. To address this question, we performed Western blotting of K562 and KU812 CML cell lines cultured in regular media or condi tioned media across various time points and used antibodies to probe for pTyr 705-STAT3. Total STAT3 and -actin serve as loading controls. As shown in Figures 23A and B, a rapid and sustained increase in pTyr705-STAT3 levels is obs erved across the indicated time points in both K562 and KU182 cells cultu red in conditioned media as compared to those cells cultured in regular media. T his suggests that our drug-resistant phenotype may be due to the anti-apopto tic affects of increased STAT3 activation and also demonstrates that our results w ere not cell-line specific.
86 Figure 23: STAT3 Phospho-Y705 is Increased in K562 and KU812 CML Cells Cultured in Conditioned Media. ( A ) K652 cells were cultured in regular media or conditioned media for the time points indicated. Ce lls were collected, lysed, and analyzed for either pTyr705 or total STAT3 via West ern blotting. -actin was used as a loading control. ( B ) KU812 cells were cultured in RM or CM for the time points indicated. Cells were collected, lysed, and analyzed for either pTyr705 or total STAT3 via Western blotting. -actin was used as a loading control. Representative figure (n = 3 independent expe riments). STAT3 Activation in CML Cells is BCR-ABL-Independen t Previous studies have shown that BCR-ABL activates STAT3 by phosphorylating Ser727 via the MEK pathway . To determine whether conditioned media-mediated STAT3 activation in CML cell s is BCR-ABL dependent, we cultured K562 cells in regular media or conditioned media for 3
87 hours, then inhibited BCR-ABL signaling by treating cel ls with varying doses of imatinib for 48 hours. Cells were subsequently lysed fol lowed by Western blotting for p-Y705 STAT3 and total STAT3 as a loading control We see in Figure 24 that treating K562 cells cultured in conditioned media with increasing concentrations of imatinib did not re duce STAT activation. In fact, there is an increase in the basal levels of p-Y705 STAT 3 in cells grown in conditioned media when compared to those cultured in r egular media and there is also sustained STAT3 activation in the presence of ima tinib. This data lends greater support to our previous findings that our dru g-resistant phenotype may be due to the anti-apoptotic affects of increased, aberran t STAT3 activation in CML cells cultured in conditioned media. Figure 24: Basal Phospho-Y705 STAT3 is Increased in K562 CML Cells Cultured in Conditioned Media and is Sustained in t he Presence of Imatinib K562 cells were cultured in regular media or conditio ned media for 3 hours prior to treating cells with varying doses of imat inib for 48 hours and followed by Western blotting for pSTAT3 and total STA T3.
88 STAT5 activation is BCR-ABL-dependent Previous work done by de Groot revealed that BCR-ABL-me diated STAT5 activation can contribute to K562 leukemic cells transform ation . To investigate the potential contribution of STAT5 activat ion in our drug resistant phenotype seen in conditioned media, we decided to also examine whether STAT5 activation in our model is BCR-ABL dependent. As before, we cultured K562 cells in regular media or conditioned media for 3 hours, inhibited BCR-ABL signaling with varying doses of imatinib for 48 hours an d performed Western blotting for p-Y694 STAT5, with total STAT5 used as a loading control. In contrast to our observations with STAT3, culturing K562 cells in conditioned media did not increase the basal levels of p STAT5 (Figure 25). Furthermore, pSTAT5 levels were equally inhibited rega rdless of the culture condition. This indicates that STAT5 activation is BCR-AB L-dependent and eliminates STAT5 as a mediator in the drug-resistant ph enotype conveyed by conditioned media.
89 Figure 25: K562 Cells Cultured in Conditioned Media Show Equal Levels of Phospho-Tyr STAT5 and Equal Inhibition of Phospho-T yr STAT5 after Imatinib Treatment. K562 cells were cultured in regular media or conditi oned media for 3 hours; cells were treated with varying doses of imatinib for 48 hours and probed for pSTAT5 and total STAT5 via Western blo tting. STAT3 Activation is not SRC-Dependent Previous studies have shown that STAT3 activation can occur via activation of c-SRC, a member of the proto-oncogenic tyr osine kinases . To examine whether STAT3 activation within our model is a lso being mediated by SRC, we cultured K562 cells in regular media or HS-5 -derived conditioned media, then treated the cells with increasing concentr ations of the dual BCRABL/SRC-kinase inhibitor, dasatinib, for 24 hour, foll owed by Western blotting for pY705 STAT3 and p-SRC. Total STAT3 and GAPDH were u sed as loading controls. Work done in our lab revealed that while dasatinib i nhibited p-SRC activity, it did not have any inhibit p-Y705 STAT3 activ ity (Figure 26). This suggests that STAT3 activation in conditioned media is also independent of SRC activity.
90 Figure 26: Dasatinib Inhibited Phospho-SRC Activity but not Phospho-Y705 STAT3 Activity. Treating K562 CML cells with dasatinib inhibited level s of pSRC family members but did not attenuate the levels o f pSTAT3 when K562 cells were cultured in CM. K562 cells were grown in regular media or conditioned media, and then treated with increasing doses of the dual BCR-ABL/SRC-kinase inhibitor, dasatinib, for 24 hours. This was followed by Western blotting for pY705 STAT3 and p-SRC. Total STAT3 and GAPDH were use d as loading controls. Representative figure (n = 3 independent expe riments). Protein Expression Levels of STAT3 Downstream Targe ts are Increased in K562 Cells Cultured in Conditioned Media We next examined whether increased activation of STAT3 corresponded with an increased expression of its downstream targets, B cl-xL, Mcl-1 and survivin. Previous reports indicated that inhibition of BCR-ABL results in decreased expression of STAT5-regulated genes including Bcl-xL (2 4). As shown in Figure 27, while basal levels of Bcl-xL, Mcl-1 and survivin wer e not increased in K562 cells cultured in conditioned media compared to those cult ured in regular media
91 (0 nM imatinib), when those cells were treated with im atinib in the context of conditioned media these protein expression levels were sust ained. Specifically, we observed that, in the presence of 500 nmol/L imatin ib, Bcl-xl, Mcl-1, and survivin were increased by 4.19 1.82, 3.48 1.84, and 7.0 3.41, respectively (n = 3 independent experiments) when cells were culture d in conditioned media compared with cells cultured in regular media. These dat a suggest that activation of STAT3 may be responsible for persistent expression of STAT-regulated genes despite inhibition of STAT5 following imatinib treat ment.
92 Figure 27: The Effects of Conditioned Media on STAT 3 Downstream Targets, Bcl-Xl, Mcl-1 and Survivin. K562 CML cells were cultured in regular media or conditioned media for 3 hours then treated w ith imatinib in a dosedependent manner for 48 hours. Western blot analysis wa s done with specific antibodies to Bcl-xl, Mcl-1, and survivin as described in Materials and Methods. Western blots showing: ( A ) Bcl-xl; ( B ), Mcl-1; and ( C ), survivin. -actin was used as a loading control. Representative figure (n = 3 inde pendent experiments).
93 Reducing STAT3 Levels with siRNA Increases Sensitiv ity to Imatinib Mesylate in Conditioned Media SiRNA technology was used to determine the causative role for activation of STAT3 with respect to mediating resistance to imatin ib when K562 cells were cultured in conditioned media. Figure 27A depicts the timeline used in performing these time-course experiments. As shown in Figure 28B, STAT3 was reduced at 84 hours an d remained reduced for 120 hours. Based on these time-course experi ments, apoptotic assays were designed as follows: K562 CML cells were transfe cted with either a control siRNA or STAT3 siRNA and this process was repeate d 36 hours later to ensure that STAT3 gene expression remained silenced; 84 hours following imatinib treatment, control siRNA and STAT3 siRNA-tra nsfected cells were treated with either 250 nmol/L imatinib or vehicle con trol for 36 hours (120 h after initial transfection with siRNA) and apoptosis was dete cted by Annexin V binding. Because some tumor types require STAT3 for survival, we fi rst analyzed whether reducing STAT3 levels with siRNA was sufficient to indu ce apoptosis when cells were cultured in either regular media or conditioned media. As shown in Figure 28C, reducing STAT3 levels (120 hours time point) wit h siRNA did not cause apoptosis in either growth condition. However, reducin g STAT3 levels sensitized K562 to imatinib-induced cell death when K562 cells we re grown in conditioned media but not in regular media (see Figure 28D for a representative figure).
94 In Figure 28E we graphed the geometric mean of the r atios of each group depicted, showing the 95% confidence intervals and the P -values. We used the geometric mean instead of the arithmetic mean because wi th the arithmetic mean values below 1 are limited to numbers between 1 a nd 0. But values above 1 are limited to 1 and infinity. Therefore, using th e geometric mean enabled us to address this asymmetry. The first group we analyzed was those cells transfected with the control siRNA and treated with imatinib reg ardless of culture conditions. Our results in Figure 28E show that in the control si RNA-transfected cells, there was a significant difference in imatinib-specific cell death seen in K562 cells cultured in regular when compared to those cultured in conditioned media. We next analyzed the STAT3 siRNA-transfected cells that wer e treated with imatinib in either regular media or conditioned media. Our r esults show that when STAT3 is knocked-down using siRNA there was no significant diff erence in imatinibinduced apoptosis in cells cultured in regular media co mpared to those cultured in conditioned media. In essence, we were able to reverse the drug-resistant phenotype seen in conditioned media by knocking-down STAT 3. We then analyzed those cells cultured in regular media, regardle ss of their transfection state. As seen in Figure 28E, our results show that when cultured in regular media, whether or not STAT3 is knocked-down, there was n o significant difference in apoptosis induced by imatinib. Lastly, we analyzed those cells cultured in conditioned media, regardless of their tr ansfection state. Our results
95 show that when cultured in conditioned media and STAT3 is knocked-down, there was a significant difference in apoptosis induced by imatinib. Taken together, these data indicate that STAT3 contrib utes to imatinib resistance only when cells are cultured within the context of bone marrow stroma-derived conditioned media. Furthermore, these da ta suggest that STAT3 can compensate for BCR-ABL survival signals and thus repr esents a potential BCR-ABL-independent mechanism of drug resistance.
98 Figure 28: Reducing STAT3 Levels with siRNA Reverse s Imatinib Resistance in K562 Cells Cultured in Conditioned Me dia. K562 cells were cultured in either regular media or conditioned media and treated with either siRNA to STAT3 or control siRNA. ( A ), Time line for conducting siRNA experiments. ( B ), STAT3 knockdown was confirmed using Western blotting an d STAT3 was noted to be maximally decreased at 84 hours a nd remained reduced for at least 120 hours (n = 5 independent experiment s). ( C ), reducing STAT3 levels was not sufficient to cause cell death in cells cultu red in regular media or conditioned media. Mean SD of 5 independent experi ments. ( D reducing STAT3 levels enhances sensitivity to imatinib when K562 ce lls are cultured in conditioned media. Apoptosis of K562 cells cultured in r egular media or conditioned media, treated with either STAT3 siRNA or control siRNA, in the presence or absence of 250 nmol/L imatinib was done usin g the Annexin V detection of apoptotic cells and fluorescence-activated cell sorting analysis. % Specific apoptosis was calculated by subtracting background c ell death from imatinib-mediated cell death. ( E ), Geometric mean of the 5 combined independent experiments showing the 95% confidence inte rvals and the corresponding P-values. The geometric mean was calculated a s the 5th root of
99 the product of all the members of the data set in all 5 experiments. The formula is depicted as follows: where a represents the data of each experiments (1-5). We decided a priori that a p-value <0.05 is statistically significant. Representative figure in triplicates (n = 5 independent experiments). Addition of GM-CSF, IL-6 and VEGF to Regular Media Induces the ImatinibResistant Phenotype Associated with Conditioned Med ia Previous studies highlighted the involvement of multip le cytokines in mediating drug-resistance in CML cells and demonstrate t he involvement of more than one cytokine in stroma-mediated protection of leu kemic cells . Therefore, we decided to investigate the role of key cyt okines, GM-CSF, IL-6 and VEGF, on our conditioned media-induced imatinib-resi stant phenotype within our model system. We wanted to determine whether the addit ion of these STAT3activating cytokines to regular media induce imatinib-res istance in K562 CML cells. To answer this question we cultured K562 cells for 3 hours in regular media containing varying concentrations of GM-CSF (0.125 2.0 ng/mL), IL-6 (0.25 4.0 ng/mL) or VEGF (3.0 48.0 ng/mL), treated the cells with vehicle control or 500 nmol/L imatinib for 48 hours and analyzed our resul ts via Annexin V apoptosis assay followed by FACS analysis. Our preliminary results indicate that the addition of these cytokines to regular media induced the imatinib-resistant phenotype that has been observed in K562 CML cells cultured in conditioned media. Figure s 29A and B show that 2.0 ng/mL GM-CSF and 4.0 ng/mL IL-6, respectively, indu ced the conditioned
100 mediaassociated imatinib-resistant phenotype to K562 cells cultured in regular media. At 48.0 ng/mL VEGF partially restores this pheno type to K562 cells cultured in regular media (Figure 28C). Together, th ese results suggest that multiple cytokines, including GM-CSF, IL-6 and VEGF, may be involved in mediating imatinib resistance to CML cells within the context of the bone marrow stroma.
102 Figure 29: The Effects of GM-CSF-, IL-6-, or VEGF-S upplemented Regular Media on Imatinib-Sensitivity in K562 CML Cells. K562 CML cells were cultured for 3 hours in regular media supplemented wit h varying concentrations of GM-CSF, IL-6 and VEGF as indicated added. After t reatment with DMSO vehicle control or 500 nmol/L imatinib, cells were all owed to incubate 48 hours at 37C, 5% CO 2 Cell death was measure using AnnexinV apoptosis assay an d data was analyzed using FACS. Representative figure (n = 2 independent experiments).
103 CHAPTER X DISCUSSION AND FUTURE DIRECTION Over 40 years ago ground-breaking research led to the i dentification of the BCR-ABL chimeric oncoprotein as the initial transformin g event in chronic myeloid leukemia (CML). This discovery revolutionized the treatment of CML and made BCR-ABL an ideal target for drug development. T his gave rise to rationally designed, small molecule signal transduction inhibitors ( STI) specific for the tyrosine kinase domain of BCR-ABL, whose constitutive act ivation is the hallmark of BCR-ABL-mediated cell transformation. The novel BCR-ABL tyrosine kinase (TK) inhibitor, im atinib mesylate (imatinib; IM), represents the first molecularly targete d therapy in patient care and has become the gold standard in the treatment for CML patients. By working as an ATP-mimic that selectively competes for and binds to t he TK domain of BCR-ABL, imatinib essentially stabilizes the oncoprotein in an inactive conformation that inhibits its enzymatic and transforming activities. The remarkable efficacy of this targeted-therapy approach to CML treatment is evidenced by the fact that imatinib produces a complete hematological response rate, which described the amount of normal leukocyte count in the peripheral blood, of 95% in newly diagnosed patients. Furthermore it produces complete
104 cytogenetic response, which is the absence of detectable Phpositive cells from more than 20 bone marrow cells in metaphase, in 90% o f patients with minimal toxicity. However, despite imatinibs efficacy and the efficacy of mor e potent BCRABL inhibitors such as nilotinib and dasatinib, the emer gence drug-resistant CML challenges us to re-visit our understanding of the mecha nisms involved in CML disease occurrence, progression and drug-resistance. Clin ical data arising from studies targeting BCR-ABL inhibitors indicate that alth ough BCR-ABL inhibitors are very effective, targeted therapy using a single age nt does not eliminate minimal residual disease (MRD). This is due to the obse rvation that BCR-ABLpositive clones are still detectable by quantitative real time-PCR in patients undergoing treatment with these inhibitors. Furthermo re, these findings are consistent with clinical data indicating that patients t hat discontinue imatinib therapy rapidly relapse to a tumor burden which is as gre at or greater then before treatment, suggesting that imatinib does not induce cell death in the stem cell population or the population capable of self-renewal [172, 173]. Imatinib was specifically designed to bind to and block t he ATP-binding pocket of the BCR-ABL kinase domain while in an inactive conformation. Mutations within the kinase domain disrupt that conform ation and thwart the ability of imatinib to bind. Therefore, the predomin ant and most characterized mechanism of imatinib-resistance is the emergence of clo nes possessing point mutations within the kinase domain of BCR-ABL. These o bservations led to the
105 development of second generation BCR-ABL inhibitors, ni lotinib and dasatinib, which are 30 and 300 times more potent than imatinib, respectively. However, despite the efficacy of these inhibitors, their use has not remedied the problem of all chemo-resistant CML associated with kinase mutations. Some patients present with indomitable pre-existing mutations, such as T315I, which these inhibitors cannot eradicate, while other patients acquir e this mutation and the resistant phenotype during the course of treatment with these inhibitors. Furthermore, recent evidence reveals that mutations of th e BCR-ABL kinase domain occur in approximately 40% to 60% of imatinib-r esistant patients . Therefore, approximately 50% of these imatinib-resist ant CML patients do not have kinase domain mutations but are resistant to imati nib by alternative means. This suggests that BCR-ABL-independent mechanisms could con tribute to imatinib-resistance and underscore the need to identify and target BCR-ABLindependent pathways. Models used to investigate the mechanisms of drug resistance seen in CML often highlight the cell-autonomous modes of resist ance: mutations within the tyrosine kinase domain of BCR-ABL itself that eith er block imatinib binding or impair the ability of the kinase to assume the correct co nformation required for imatinib to bind. But these models often fail to consid er the role of the surrounding cells and tissues, the microenvironment, on the cancer itself and the role of this microenvironment on the emergence of the tumor drug-resistance phenotype. However, recent studies have draw attention t o the involvement of
106 the tumor microenvironment in providing cancer cells an escape from chemotherapy-induced cell death. While studying drug res istance in Ph-positive acute lymphoblastic leukemia (ALL), William et al reported on a nonautonomous mechanism of drug resistance involving cytokin e signaling within the hematopoietic microenvironment . This study revealed that the contribution of BCR-ABL-independent mechanisms was capable of protect ing CML cells from imatinib-induced cell death. Therefore, the use of an in vitro non-autonomous model of drug resistance may provide further insights in to the role of host-derived stimuli on CML cells in conferring resistance to BCR-ABL inhibitors and may provide a greater comprehension of how to combat this mechanism of resistance. To investigate the role of the bone marrow microenviro nment on resistance to BCR-ABL inhibitors, we utilized an in vitro bone marrow stromal model that exposes CML cells to multiple soluble factors p roduced by bone marrow stroma cells. The human bone marrow stromal cell line, HS-5, served as the model of the bone marrow microenvironment as it c losely approximates the bone marrow-associated cytoprotection observed in drug-trea ted leukemia patients. HS-5 is fibroblastic and has been shown to secre te numerous cytokines that are capable of supporting the ex vivo expansion of mature and immature hematopoietic cells and their progenitors (. Two hematopoietic CML cell lines were used to model the disease: K562 cells origina ted in the bone marrow, are highly undifferentiated and were derived from the pleural effusion of a CML
107 patient in blast crisis phase; KU812 cells are pre-basophi lic and originated from the peripheral blood of a blast-crisis-phase CML patien t. Previous work done in our laboratory revealed that adh esion of K562 CML cells to fibronectin could provide cell adhesion mediate d-drug resistance (CAMDR) against imatinib-induced cell death through anti-a poptotic integrin-mediated signaling . Similarly, our initial set of experim ents using a co-culture transwell model system revealed that direct adhesion of K5 62 cells to HS-5 stoma cells conferred protection against cell death induce d by imatinib, possibly through a similar mechanism. Interestingly, however, was the observation that HS-5-derived soluble factors were also capable of modulat ing imatinib response in CML cells and confer resistance. This data indicates tha t both direct contact with the bone marrow stroma and soluble factors produced by supporting stroma cells play a role in the emergence of the CML drug-res istant phenotype. This data also lends support to previous observations that, in addition to targeted therapy against BCR-ABL, the tumor-microenvironmental i nteractions may play a crucial role in protecting CML cells from the anti-apopt otic and anti-proliferative effects of imatinib. To delineate the contributions of bone marrow stroma-derived soluble factors on resistance to BCR-ABL inhibitors and t o simplify our model the data presented in this dissertation were obtained from the use of HS-5-derived conditioned media. It has been demonstrated here that imatinib induces a h igh rate of cell death in CML cells cultured in regular media (RM: RPMI + 10% fetal bovine
108 serum). However, using the same media conditioned for th ree hours by the HS-5 human stroma cells (CM) is sufficient to cause resistance t o imatinib-induced cell death in K562 and KU182 CML cells. Since HS-5 stromal ce lls produce cytokines that support the growth and differentiation of HSCs, this data suggests that HS-5derived soluble factor(s) may be involved in mediating th is imatinib-resistant phenotype in CML cells. Furthermore, HS-5 conditioned media could be stored at -80C and retain its ability to produce imatinib resis tance in these CMLS cells, while heat-inactivated conditioned media lost this abil ity. These data indicate that the soluble factor(s) conferring resistance to imatinib i s quite stable and capable of preserving its function during the freezing and tha wing process. Additionally, heat-inactivation of conditioned media would destroy i ts protein components, which includes growth factors and cytokines. The observati on that heatinactivated conditioned media restores imatinib-sensitivi ty to CML cells demonstrates that the protective component in conditioned media is a protein. Based on findings of similar studies, the hematopoietic cytokine granulocyte macrophage-colony stimulating factor (GM-CSF) and interl eukin 6 (IL-6) are two potential cytokines that may be involved in mediating the BCR-ABL-independent imatinib-resistant phenotype seen in our model. While HS-5-derived conditioned media increased the clon ogenic survival of K562 cells, exposure of CML cells to the soluble facto rs in conditioned media did not lead to an increase in cell proliferation. Th is indicates that the soluble factor(s) mediate imatinib-resistance through cell survi val pathways. There are
109 several well-characterized cytokine-receptor-activated cell surv ival pathways that may be involved in promoting clonogenic survival of these cells. The Janus activated kinase (JAK)/STAT signal transduction pathway is important in converting cytokine-receptor signals into downstream surviva l signals and is often constitutively activated in hematopoietic malignanci es. Studies show that in BCR-ABL-expressing cells, constitutive activation of JAKs is associated with constitutive STAT3 Try-705 phosphorylation. Studies show that STAT5 is also constitutively activated in CML cells. Other survival pathways t hat may enhance the clonogenic survival of CML cells are the anti-apopto tic phosphatidylinositol 3kinase (PI3K)/AKT signaling pathway, whose activation has been shown to not only be involved in BCR-ABL-mediated cell transformation but also resistance to BCR-ABL inhibitors. Additionally, BCR-ABL also signals t hrough the extracellular signal regulated kinase 1/2 (ERK-1/2) pathway and has been implicated in the anti-apoptotic activities of BCR-ABL. Data presented here revealed that when CML cells were n o longer exposed to the protective soluble factor(s) associated with HS-5 conditioned media and were re-cultured in regular media, the im atinib-resistant phenotype diminished. This demonstrates that withdrawal of CML ce lls from bone marrow stroma-secreted soluble factor(s) was sufficient to re-sensi tize these cells to the apoptotic effects of imatinib. It is possible that this o bservation is attributable to the termination of cytokine-receptor-mediated signalin g. Furthermore, this data reveals a direct correlation between the presence of HS-5 -derived soluble factors
110 and the emergence of imatinib resistance, an observation that has also been highlighted in other studies. The hypothesis within this dissertation is based on the im portance of soluble factors specific to the bone marrow microenvironm ent in promoting chemoresistance in CML cells. To accurately test this hypothesi s it was important to investigate whether K562 CML cells could condition it s own media via autocrine secretion of its own soluble factors and induce chemoresistance. Additionally, we investigated whether other non-stromal cell lines were capable of conveying protection against cell death induced by imatini b. In either case, if there is protection against cell death induced by BCR-AB L inhibitors then this would demonstrate that the bone marrow microenvironmen t may not be responsible for providing chemo-resistance to CML cells a nd would discredit its role in modulating drug response within our model. Co nditioned media from K562 CML cells and two hematopoietic cell lines, U937 and 82 26, were generated in the same manner as HS-5-derived conditioned media: the se cells were cultured in regular media for 3 hours, enabling the media to become conditioned by the different cell lines; the conditioned media was then co llected by centrifugation to remove debris and utilized in subsequent experiments. K56 2 CML cells were cultured in regular media and several types of conditione d media, as previously described. The cells were treated with imatinib for 4 8 hours to induce cell death, which was measured using the Annexin V apoptosis assay. Ou r data revealed that of the different conditioned media generated, o nly HS-5-derived conditioned
111 media was capable of conveying imatinib-resistance in K56 2 CML cells. This data validated the unique role and importance of solub le factors specific to the bone marrow stroma in protecting CML cells from death i nduced by imatinib. There are numerous bone marrow stroma-derived soluble factors and cytokines involved in the growth and differentiation hema topoietic cells that may also be crucial in conferring resistance to BCR-ABL inhi bitors. According to Torok-Storb (1999), the human HS-5 stroma cell produ ces a cocktail of cytokines that include IL-6, GM-CSF, G-CSF and VEGF. It is con ceivable that these cytokines could confer protection against imatinib-induced cell death to varying degrees either individually on in combination with each other. In fact, while studies by Wang et al focused on the role of the autocrine secretion of GM -CSF in mediating resistance to BCR-ABL inhibitors in CML cells , Weisberg et al. emphasized the involvement of the bone marrow stroma as a whole in mediating protection against CML cell death by using a cocktail of stroma-derived cytokines . Our data from the bone marrow stromal model lends cre dence to the involvement of survival signaling mediated by stroma-deri ved soluble factors in the emergence of chemo-resistant CML. This is also eviden ced by our observations that neither nilotinib nor dasatinib are capable of overriding the protective effects of HS-5-secreted soluble factors on K5 62 CML cells. This demonstrates that even though one inhibitor was designed to bind more stringently to the tyrosine kinase domain of BCR-ABL a nd the other was
112 designed to recognize and bind to both its active and in active conformations, the development and use of these more potent, rationally-d esigned inhibitors was not sufficient to address this mechanism of resistance involving the bone marrow microenvironment. There is growing interest in research to target components of the hematopoietic tumor microenvironment as a way to en hance the efficacy of targeted therapy against CML. While research that addre sses BCR-ABLindependent mechanisms of imatinib resistance have gained momentum, there are also concerns about the effects that targeting the b one marrow microenvironment may have on normal hematopoiesis and su rrounding cells. Studies suggest that the best initial approach is to tar get individual signal transduction pathways that are constitutively activated in ma lignant hematopoietic cells. BCR-ABLs modular domains make it capable of activating s everal signal transduction pathways that can contribute to enhanced CML cell survival, including Ras/ERK, PI3K/AKT, and STAT5 signal transdu ction pathways. These pathways can also be activated by external signals includi ng growth factors, cytokines, and interactions with extracellular matrices. Fu rthermore, studies suggest that these pathways can be reconstituted in a BCRABL-independent manner by these external signals. Our data supports this observation and reveals that HS-5-derived soluble factors are capable of BCR-AB L-independent constitutive STAT3 activation via Tyr-705 phosphorylation Studies show that constitutive STAT3 Tyr-705 phosphorylation and activation is mediated by
113 activated JAKs. Activated STAT3 homoor heterodimerizes an d translocates to the nucleus where it leads to the transcriptional activ ation of STAT3 target antiapoptotic genes. Studies show that cytokines that are capable of STAT3 act ivation include IL-6, vascular endothelial growth factor (VEGF), and GM -CSF. Here we demonstrate that not only does the exposure of either K 562 or KU812 CML cells to conditioned media derived from HS-5 cells resulted i n the rapid and sustained increase in phospho-Y705 STAT3 levels, but we also show th at this exposure to HS-5-derived soluble factors also led to an increase in the protein expression levels of STAT3 downstream targets, Bcl-xL, Mcl-1 and surv ivin. This suggests that HS-5-derived soluble factors may be involved in the activation of STAT3, which in turn resulted in STAT3-mediated expression of these anti-apoptotic proteins. Furthermore, the addition of imatinib does not reduce phospho-Y705 STAT3 levels or the levels of its downstream targets. This demonstrates that activation of STAT3 is indeed independent of BCR-ABL a ctivity and highlights the importance of BCR-ABL-independent mechanisms in facilit ating the imatinibresistant phenotype. This data also contradicts an earli er observation made by Coppo et. al. which showed that the constitutive phosphorylation of S TAT3 on Tyr-705 was dependent on BCRABL . This differenc e may be attributable to the fact that they utilized a hematopoietic cell li ne that was retrovirally transduced with a p210-BCR-ABL expressing vector as oppose to utilizing a cell line obtained from CML patients.
114 Since studies show that STAT3 activation can also be accompl ished through the activities of SRC family kinases (SFKs), it was important to determine whether the increase in BCR-ABL-independent STAT3 activity was attributable to SFKs. Our data showed that when K562 CML cells were treated with the dual BCR-ABL/SFK inhibitor, dasatinib, sustai ned phosphorylation of STAT3 was still observed in those cells cultured in HS-5-de rived conditioned media, while phospho-SRC levels were attenuated. This observation eliminates the involvement of SFKs in the increased activation of ST AT3 that we observed in our model and lends greater support to the involveme nt of HS-5 derived soluble factors in mediating STAT3 activation, possible th rough cytokine receptor-engagement and JAKs activation. While investiga ting the role of STAT3 activation in K562 CML cells exposed to HS-5 soluble facto rs and assessing the mechanism(s) involved in its activation, one interestin g and consistent observation was that the use of increasing concentratio ns of imatinib resulted in increased STAT3 activation. In examining this further, other studies have reported that BCR-ABL can activate STAT5 and STAT1 ind ependently of the activation of JAKs . Furthermore, a dominant-negat ive STAT5 was shown to inhibit colony formation of K562 cells and blocking BCR -ABL was shown to inhibit STAT5-dependent DNA binding as well as STAT5 target g enes, Bcl-xL and Mcl-1 [169, 176-180]. Bcl-xl, Mcl-1, and cyclin D1 represent ge nes that can be regulated by either STAT5 or STAT3 activation. Therefo re, based on these observations, we propose that within the bone marrow mi croenvironment STAT3
115 can compensate for BCR-ABL-dependent activation of STAT5 -dependent genes, which are critical for CML cell survival, when STAT5 act ivity is blocked by BCRABL inhibitors. In support of this hypothesis, our data also indicate that when K562 cells were treated with imatinib, the levels of Bc l-xL, survivin, and Mcl-1 were increased when cells were cultured in conditioned me dia compared with cells cultured in regular media. Together, these data suggest that, depending on culture conditions, K562 CML cells may have plasticity with respect to STATdependency. To determine the causal role of STAT3 activation in our drug-resistant phenotype associated with conditioned media, we reduced STAT3 levels using siRNA technology. Here we show that this reduction in STA T3 levels led to the enhanced sensitivity of K562 CML cells to imatinib-indu ced cell death. Importantly, this increased imatinib-sensitivity in CML cells was observed only within the context of the bone marrow stroma model, a s reducing STAT3 levels in regular media did not result in increased cell death. These data highlight the contribution and significance of BCR-ABL-independent S TAT3 activation in reconstituting BCR-ABL-mediated survival signaling event s that promote CML progression. Furthermore, our data indicate that whil e the initial BCR-ABL insult that initiated the disease within the primitive hemato poietic stem cell population is crucial to CML pathogenesis, clearly soluble factors uniqu e to the hematopoietic microenvironment are capable of protecting CML cells from the apoptotic effects of BCR-ABL inhibitors. To address the question recently p osed by Charles
116 Sawyers in his news brief entitled Where lies the blame for resistance tumor or host? , the synthesis of data presented here, alo ng with data presented in previous studies, suggest that both the tumor and the ho st are responsible. Consequently, the use of BCR-ABL inhibitors alone is n ot sufficient to address this type of resistance, which is associated with BCR-ABL -independent STAT3 activation. While BCR-ABL inhibitors target the fusio n oncoprotein, it does not target STAT3. Therefore, these data also provide preclinical rationale for targeting STAT3 in order to increase the efficacy of BC R-ABL inhibitors in treating drug-resistant CML cells. To that end, a mult i-targeted approach to CML patient care that involves the combined use of BCR-ABL in hibitors and STAT3 inhibitors may prove more effective in addressing this type of drug-resistant CML. In vitro studies to identify compounds that are highly selective fo r STAT3 have highlighted the effectiveness of cisplatin/IS3 295  as well as cucurbitacin Q  in inhibiting STAT3 activation. Bone marrow stroma-derived conditioned media consists of a complex mixture of cytokines and other soluble factors that can be source of survival signals affecting cell differentiation, expansion and sur vival. The functional redundancy of many of these soluble factors and their sig naling receptors may converge to activate STAT3 and circumvent cell death induce d by BCR-ABL inhibitors. Since several studies have highlighted the in volvement of GM-CSF, IL6 and VEGF in the emergence and progression of hemato logical malignancies, we examined the contribution of these cytokines in our d rug-resistant phenotype.
117 We wanted to determine whether exposure of CML cells t o these cytokines can modulate their response BCR-ABL inhibitors. Here, our data revealed that when these cytokines were added to regular media in varying concentrations a dosedependent imatinib resistant phenotype emerged. This i s evidenced by the steady reduction in imatinib-induced CML cell deaths, whi ch correlated with increased cytokine concentrations. The addition of 2.0 ng /mL of GM-CSF, 4.0 ng/mL of IL-6 or 48.0 ng/mL of VEGF to regular medi a initiated the return of the imatinib-resistant phenotype associated with HS-5 condi tioned media. These results, which are consistent with observations made in si milar studies, indicate that these cytokines may be working individually or coopera tively to protect CML cells from death induced by BCR-ABL inhibitors. Furthermore, data from other studies revealed that cond itioned media from malignant cell lines expressing high levels of pho spho-STAT3 also contained cytokines that are capable of stimulating STAT3 activation, such as IL6. To determine the involvement of cytokines such as IL-6, GM-CSF and VEGF within our model, their activity in conditioned media could also be impeded using blocking antibodies. For example, IL-6 signal transduct ion, which activates STAT3 through the glycoprotein (gp) 130/JAK pathway and is notable for its pleiotropic tumor-promoting activities, could be blocked using a gp130-blocking antibody or an IL-6-blocking antibody (such as BR-3 and 522, respectively; provided by Cell Sciences, Inc.). Alternatively, to validat e the results obtained from the use of cytokine-blocking antibodies, the gene expression of these
118 cytokines in HS-5 cells could also be sequentially silenced us ing siRNA technology. To identify the specific soluble factor(s) causative for i matinib resistance and STAT3 activation within our model, one global a pproach is to utilize multiplexed bead-based assays that efficiently detect an d quantify several cytokines simultaneously (Invitrogen). These assays consist of captureantibodies conjugated to biologically inert beads and ar e designed to identify various soluble factors, including cytokines, with enhanced speed and accuracy. Furthermore, since our data revealed that HS-5 cells do not require serum to produce the protective soluble factors, we could identif y the active soluble factor(s) in HS-5 conditioned media that confers imatin ib resistance by fractionating the SF-CM into its protein components and performing the Annexin V apoptosis assay on each fraction. Fractionation could be accomplished in several ways. Centricon spin columns (Miliipore) can separat e the components of conditioned media by molecular weight. Additionally, fr actionation can also be accomplished using liquid chromatography or mass spectromet ry. Constitutive STAT3 activation is associated with several hum an malignancies, including solid tumors and hematological d iseases. Interestingly, we observed that reducing STAT3 levels sensitized imati nib-resistant CML cells to its apoptotic effects. It is possible that while BCR -ABL-mediated oncogenic signaling events are being impeded by imatinib, the no n-autonomous activation of STAT3 by soluble factors within the bone marrow micro environment facilitates
119 these signal transduction events (see Figure 30 for propo sed model). Additionally, our data provides preclinical rationale to support the use of STAT3 inhibitors as BCR-ABL-inhibitor sensitizers and for cond uction validating our observations in studies using primary patient specimens. On e set of experiments would be to confirm data presented in other studies t hat show increased STAT3 activation in primary CML patient samples when compared t o healthy bone marrow donors. Furthermore, cells from CML patient can be used to delineate whether small molecule inhibitors targeting STAT3, such as phosphotyrosyl peptides , are capable of enhancing the efficacy o f BCR-ABL inhibitors using in vivo models. Furthermore, patients samples could be used to examine whether STAT3 activation enables CML cells to escape complement-dependen t cytotoxicity via CD46 expression. It has been reported that STAT3 activa tion is capable of inducing the expression of CD46, one of the complemen t-regulatory proteins (CRPs) expressed on the surface of normal and transformed cells. While complement proteins target bacteria and cancerous cells f or lysis , cancer cells often escape elimination by complement proteins. Th is escape mechanism is due in part to the expression of this membrane-boun d CRP, which inactivates complement components. It would be of interest to e whe ther increased STAT3 activation in patient samples also resulted in this escape mechanism. Studies show that BCR-ABL-positive MRD in some imatini b-resistant CML cells are characterized by increased activation of the PI3K /AKT and Ras/MEK
120 signal transduction pathways. These pathways are capable of mediating cell survival by transmitting signals from multiple cell surfa ce receptors to transcription factors within the nucleus. However, the me chanisms responsible for the enhanced activation of these survival pathways rem ain poorly understood. Future studies may include investigating whether the PI3 K/AKT and Ras/MEK signal transduction remains activated in imatinib-treate d CML cells cultured in conditioned media. Furthermore, a SCID-hu in vivo model could be implemented to validate the involvement of these pathways in modulati ng imatinib response in CML cells. In conclusion, while imatinib and other BCR-ABL inhi bitors are highly effective in treating CML patients and addressing mechan isms of acquired resistance, the high rates of patient relapse and the persistence of MRD CML challenges us to continuously expand our understanding of CML pathogenesis and seek innovative solutions to combat this disease. Data p resented here demonstrate that targeted therapy using a single agen t does not eradicate drug resistance associated with the bone marrow microenvironm ent. However, the use of combinational therapy that targets BCR-ABL as wel l as stroma-derived soluble factors promises to increase the efficacy of BCR-ABL inhibitors by sensitizing cells to their apoptotic effects. This accomplish ment would translate into increased remission rates among CML patients and f urther revolutionize oncology therapeutics. Additionally, these data demonstrat e the importance of bone marrow stroma models that consider the tumor micr oenvironment in
121 identifying novel targets that enhance the efficiency of existing targeted drug therapies. Figure 30: Proposed Mechanisms of Resistance to BCR -ABL Inhibitors in Chronic Myeloid Leukemia Cells Data from our bone marrow stroma model show increased and sustained STAT3 activation in CML cell s. Bone marrow stroma cells secrete numerous soluble factors that are cap able of STAT3 activation. Additionally, soluble factors within the bone marrow microenvironment may be capable of reconstituting BCR-ABL downstream signa ling events, such as the Ras/MEK and PI3K/AKT cell survival signal transd uction pathways.
122 LITERATURE CITED 1. Weissman, I.L., Stem cells: units of development, units of regeneration, and units in evolution. Cell, 2000. 100 (1): p. 157-68. 2. Guenechea, G., et al., Distinct classes of human stem cells that differ in proliferative and self-renewal potential. Nat Immunol, 2001. 2 (1): p. 75-82. 3. Bruce, W.R. and H. Van Der Gaag, A Quantitative Assay for the Number of Murine Lymphoma Cells Capable of Proliferation in Vivo. Nature, 1963. 199 : p. 79-80. 4. Al-Hajj, M., et al., Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A, 2003. 100 (7): p. 3983-8. 5. Singh, S.K., et al., Identification of human brain tumour initiating cells. Nature, 2004. 432 (7015): p. 396-401. 6. Guan, Y., B. Gerhard, and D.E. Hogge, Detection, isolation, and stimulation of quiescent primitive leukemic progenitor cel ls from patients with acute myeloid leukemia (AML). Blood, 2003. 101 (8): p. 3142-9. 7. Krivtsov, A.V., et al., Transformation from committed progenitor to leukaemia stem cell initiated by MLL-AF9. Nature, 2006. 442 (7104): p. 818-22.
123 8. Maguer-Satta, V., et al., BCR-ABL expression in different subpopulations of functionally characterized Ph+ CD34+ cells from patien ts with chronic myeloid leukemia. Blood, 1996. 88 (5): p. 1796-804. 9. Jiang, X., et al., Autocrine production and action of IL-3 and granulocyte colony-stimulating factor in chronic myeloid leukemia. Proc Natl Acad Sci U S A, 1999. 96 (22): p. 12804-9. 10. Holyoake, T.L., et al., Cell separation improves the sensitivity of detecting rare human normal and leukemic hematopoietic cells in vi vo in NOD/SCID mice. Cytotherapy, 2000. 2 (6): p. 411-21. 11. Wang, J.C., et al., High level engraftment of NOD/SCID mice by primitive normal and leukemic hematopoietic cells from patients wit h chronic myeloid leukemia in chronic phase. Blood, 1998. 91 (7): p. 2406-14. 12. Lapidot, T., et al., A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature, 1994. 367 (6464): p. 645-8. 13. Bonnet, D. and J.E. Dick, Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoie tic cell. Nat Med, 1997. 3 (7): p. 730-7. 14. Simmons, D.L., et al., Molecular cloning of a cDNA encoding CD34, a sialomucin of human hematopoietic stem cells. J Immunol, 1992. 148 (1): p. 267-71. 15. Deaglio, S., K. Mehta, and F. Malavasi, Human CD38: a (r)evolutionary story of enzymes and receptors. Leuk Res, 2001. 25 (1): p. 1-12.
124 16. Ades, E.W., et al., Isolation and partial characterization of the human homologue of Thy-1. J Exp Med, 1980. 151 (2): p. 400-6. 17. Sirard, C., et al., Normal and leukemic SCID-repopulating cells (SRC) coexist in the bone marrow and peripheral blood from C ML patients in chronic phase, whereas leukemic SRC are detected in blast crisis. Blood, 1996. 87 (4): p. 1539-48. 18. Cozzio, A., et al., Similar MLL-associated leukemias arising from selfrenewing stem cells and short-lived myeloid progenitors. Genes Dev, 2003. 17 (24): p. 3029-35. 19. Jamieson, C.H., et al., Granulocyte-macrophage progenitors as candidate leukemic stem cells in blast-crisis CML. N Engl J Med, 2004. 351 (7): p. 657-67. 20. Neering, S.J., et al., Leukemia stem cells in a genetically defined murine model of blast-crisis CML. Blood, 2007. 110 (7): p. 2578-85. 21. Jemal, A., et al., Cancer statistics, 2008. CA Cancer J Clin, 2008. 58 (2): p. 71-96. 22. Nowell, P.C. and D.A. Hungerford, Chromosome studies on normal and leukemic human leukocytes. J Natl Cancer Inst, 1960. 25 : p. 85-109. 23. Rudkin, C.T., D.A. Hungerford, and P.C. Nowell, DNA Contents of Chromosome Ph1 and Chromosome 21 in Human Chronic Gran ulocytic Leukemia. Science, 1964. 144 : p. 1229-31.
125 24. Rowley, J.D., Letter: A new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine flu orescence and Giemsa staining. Nature, 1973. 243 (5405): p. 290-3. 25. Bartram, C.R., et al., Translocation of c-ab1 oncogene correlates with the presence of a Philadelphia chromosome in chronic myelocytic l eukaemia. Nature, 1983. 306 (5940): p. 277-80. 26. Groffen, J., et al., Philadelphia chromosomal breakpoints are clustered within a limited region, bcr, on chromosome 22. Cell, 1984. 36 (1): p. 93-9. 27. Deininger, M.W., et al., Selective induction of leukemia-associated fusion genes by high-dose ionizing radiation. Cancer Res, 1998. 58 (3): p. 421-5. 28. Kozubek, S., et al., Distribution of ABL and BCR genes in cell nuclei of normal and irradiated lymphocytes. Blood, 1997. 89 (12): p. 4537-45. 29. Neves, H., et al., The nuclear topography of ABL, BCR, PML, and RARalpha genes: evidence for gene proximity in specific p hases of the cell cycle and stages of hematopoietic differentiation. Blood, 1999. 93 (4): p. 1197-207. 30. McWhirter, J.R., D.L. Galasso, and J.Y. Wang, A coiled-coil oligomerization domain of Bcr is essential for the tran sforming function of Bcr-Abl oncoproteins. Mol Cell Biol, 1993. 13 (12): p. 7587-95. 31. Reuther, G.W., et al., Association of the protein kinases c-Bcr and Bcr-Abl with proteins of the 14-3-3 family. Science, 1994. 266 (5182): p. 129-33.
126 32. Laneuville, P., Abl tyrosine protein kinase. Semin Immunol, 1995. 7 (4): p. 255-66. 33. Montaner, S., et al., Multiple signalling pathways lead to the activation of the nuclear factor kappaB by the Rho family of GTPases. J Biol Chem, 1998. 273 (21): p. 12779-85. 34. Diekmann, D., et al., Bcr encodes a GTPase-activating protein for p21rac. Nature, 1991. 351 (6325): p. 400-2. 35. Pendergast, A.M., et al., SH1 domain autophosphorylation of P210 BCR/ABL is required for transformation but not growth factor independence. Mol Cell Biol, 1993. 13 (3): p. 1728-36. 36. Puil, L., et al., Bcr-Abl oncoproteins bind directly to activators of the Ras signalling pathway. EMBO J, 1994. 13 (4): p. 764-73. 37. Liu, J., Y. Wu, and R.B. Arlinghaus, Sequences within the first exon of BCR inhibit the activated tyrosine kinases of c-Abl and the Bcr-Abl oncoprotein. Cancer Res, 1996. 56 (22): p. 5120-4. 38. Abelson, H.T. and L.S. Rabstein, Influence of prednisolone on Moloney leukemogenic virus in BALB-c mice. Cancer Res, 1970. 30 (8): p. 2208-12. 39. Mayer, B.J. and D. Baltimore, Mutagenic analysis of the roles of SH2 and SH3 domains in regulation of the Abl tyrosine kinase. Mol Cell Biol, 1994. 14 (5): p. 2883-94. 40. Yuan, Z.M., et al., p73 is regulated by tyrosine kinase c-Abl in the apoptotic response to DNA damage. Nature, 1999. 399 (6738): p. 814-7.
127 41. Li, B., et al., Distinct roles of c-Abl and Atm in oxidative stress response are mediated by protein kinase C delta. Genes Dev, 2004. 18 (15): p. 1824-37. 42. Kipreos, E.T. and J.Y. Wang, Cell cycle-regulated binding of c-Abl tyrosine kinase to DNA. Science, 1992. 256 (5055): p. 382-5. 43. Woodring, P.J., T. Hunter, and J.Y. Wang, Regulation of F-actindependent processes by the Abl family of tyrosine kinases. J Cell Sci, 2003. 116 (Pt 13): p. 2613-26. 44. Smith, K.M., R. Yacobi, and R.A. Van Etten, Autoinhibition of Bcr-Abl through its SH3 domain. Mol Cell, 2003. 12 (1): p. 27-37. 45. Taagepera, S., et al., Nuclear-cytoplasmic shuttling of C-ABL tyrosine kinase. Proc Natl Acad Sci U S A, 1998. 95 (13): p. 7457-62. 46. Pendergast, A.M., et al., BCR-ABL-induced oncogenesis is mediated by direct interaction with the SH2 domain of the GRB-2 a daptor protein. Cell, 1993. 75 (1): p. 175-85. 47. Sattler, M., et al., Critical role for Gab2 in transformation by BCR/ABL. Cancer Cell, 2002. 1 (5): p. 479-92. 48. Nagar, B., et al., Structural basis for the autoinhibition of c-Abl tyrosine kinase. Cell, 2003. 112 (6): p. 859-71. 49. Hantschel, O., et al., A myristoyl/phosphotyrosine switch regulates c-Abl. Cell, 2003. 112 (6): p. 845-57.
128 50. Shi, Y., K. Alin, and S.P. Goff, Abl-interactor-1, a novel SH3 protein binding to the carboxy-terminal portion of the Abl pr otein, suppresses v-abl transforming activity. Genes Dev, 1995. 9 (21): p. 2583-97. 51. Dai, Z. and A.M. Pendergast, Abi-2, a novel SH3-containing protein interacts with the c-Abl tyrosine kinase and modulates c-Abl transforming activity. Genes Dev, 1995. 9 (21): p. 2569-82. 52. Cicchetti, P., et al., Identification of a protein that binds to the SH3 reg ion of Abl and is similar to Bcr and GAP-rho. Science, 1992. 257 (5071): p. 803-6. 53. Wen, S.T. and R.A. Van Etten, The PAG gene product, a stress-induced protein with antioxidant properties, is an Abl SH3-bin ding protein and a physiological inhibitor of c-Abl tyrosine kinase activity. Genes Dev, 1997. 11 (19): p. 2456-67. 54. Dai, Z., et al., Oncogenic Abl and Src tyrosine kinases elicit the ubiquitin dependent degradation of target proteins through a R as-independent pathway. Genes Dev, 1998. 12 (10): p. 1415-24. 55. Nishida, E. and Y. Gotoh, The MAP kinase cascade is essential for diverse signal transduction pathways. Trends Biochem Sci, 1993. 18 (4): p. 128-31. 56. Marshall, C.J. and S.J. Leevers, Mitogen-activated protein kinase activation by scrape loading of p21ras. Methods Enzymol, 1995. 255 : p. 273-9.
129 57. Lewis, T.S., P.S. Shapiro, and N.G. Ahn, Signal transduction through MAP kinase cascades. Adv Cancer Res, 1998. 74 : p. 49-139. 58. Oda, T., et al., Crkl is the major tyrosine-phosphorylated protein in neutrophils from patients with chronic myelogenous leukemi a. J Biol Chem, 1994. 269 (37): p. 22925-8. 59. Pelicci, G., et al., Constitutive phosphorylation of Shc proteins in human tumors. Oncogene, 1995. 11 (5): p. 899-907. 60. Bhat, A., et al., Interactions of p62(dok) with p210(bcr-abl) and Bcr-Ablassociated proteins. J Biol Chem, 1998. 273 (48): p. 32360-8. 61. Hennigan, R.F. and P.J. Stambrook, Dominant negative c-jun inhibits activation of the cyclin D1 and cyclin E kinase complexes. Mol Biol Cell, 2001. 12 (8): p. 2352-63. 62. Kidd, M., et al., Global expression analysis of ECL cells in Mastomys natalensis gastric mucosa identifies alterations in the AP1 pathway induced by gastrin-mediated transformation. Physiol Genomics, 2004. 20 (1): p. 131-42. 63. Lunec, J., et al., Redox-regulation of DNA repair. Biofactors, 2003. 17 (14): p. 315-24. 64. Manicassamy, S., et al., Protein kinase C-theta-mediated signals enhance CD4+ T cell survival by up-regulating Bcl-xL. J Immunol, 2006. 176 (11): p. 6709-16.
130 65. Lane, S.J., et al., Corticosteroid-resistant bronchial asthma is associated with increased c-fos expression in monocytes and T lymphocytes J Clin Invest, 1998. 102 (12): p. 2156-64. 66. Proffitt, J., et al., An ATF/CREB-binding site is essential for cell-specific and inducible transcription of the murine MIP-1 beta cyt okine gene. Gene, 1995. 152 (2): p. 173-9. 67. Sanyal, S., et al., AP-1 functions upstream of CREB to control synaptic plasticity in Drosophila. Nature, 2002. 416 (6883): p. 870-4. 68. Sattler, M., et al., Thrombopoietin induces activation of the phosphatidylinositol-3' kinase pathway and formation of a complex containing p85PI3K and the protooncoprotein p120CBL. J Cell Physiol, 1997. 171 (1): p. 28-33. 69. Kharas, M.G., et al., Ablation of PI3K blocks BCR-ABL leukemogenesis in mice, and a dual PI3K/mTOR inhibitor prevents expansion of human BCRABL+ leukemia cells. J Clin Invest, 2008. 118 (9): p. 3038-50. 70. Ren, S.Y., et al., Intrinsic regulation of the interactions between the SH 3 domain of p85 subunit of phosphatidylinositol-3 kinase a nd the protein network of BCR/ABL oncogenic tyrosine kinase. Exp Hematol, 2005. 33 (10): p. 1222-8. 71. Hickey, F.B. and T.G. Cotter, Identification of transcriptional targets associated with the expression of p210 Bcr-Abl. Eur J Haematol, 2006. 76 (5): p. 369-83.
131 72. Datta, S.R., et al., Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell, 1997. 91 (2): p. 231-41. 73. Brunet, A., et al., Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell, 1999. 96 (6): p. 857-68. 74. Stahl, M., et al., The forkhead transcription factor FoxO regulates transcription of p27Kip1 and Bim in response to IL-2. J Immunol, 2002. 168 (10): p. 5024-31. 75. Cardone, M.H., et al., Regulation of cell death protease caspase-9 by phosphorylation. Science, 1998. 282 (5392): p. 1318-21. 76. Pap, M. and G.M. Cooper, Role of glycogen synthase kinase-3 in the phosphatidylinositol 3-Kinase/Akt cell survival pathway. J Biol Chem, 1998. 273 (32): p. 19929-32. 77. Agarwal, A., et al., The AKT/I kappa B kinase pathway promotes angiogenic/metastatic gene expression in colorectal cancer b y activating nuclear factor-kappa B and beta-catenin. Oncogene, 2005. 24 (6): p. 102131. 78. Bromberg, J.F., et al., Transcriptionally active Stat1 is required for the antiproliferative effects of both interferon alpha and interferon gamma. Proc Natl Acad Sci U S A, 1996. 93 (15): p. 7673-8. 79. Grimley, P.M., et al., Prolonged STAT1 activation related to the growth arrest of malignant lymphoma cells by interferon-alpha. Blood, 1998. 91 (8): p. 3017-27.
132 80. Sironi, J.J. and T. Ouchi, STAT1-induced apoptosis is mediated by caspases 2, 3, and 7. J Biol Chem, 2004. 279 (6): p. 4066-74. 81. Rocnik, J.L. and D.G. Gilliland, Cell-autonomous and -nonautonomous contributions of STAT1 in murine models of tumorigenesi s. Cancer Cell, 2006. 10 (1): p. 1-2. 82. Stout, B.A., et al., IL-5 and granulocyte-macrophage colony-stimulating factor activate STAT3 and STAT5 and promote Pim-1 and cyclin D3 protein expression in human eosinophils. J Immunol, 2004. 173 (10): p. 6409-17. 83. Lee, I.H., et al., Inhibition of interleukin 2 signaling and signal transd ucer and activator of transcription (STAT)5 activation during T cell receptormediated feedback inhibition of T cell expansion. J Exp Med, 1999. 190 (9): p. 1263-74. 84. Buettner, R., L.B. Mora, and R. Jove, Activated STAT signaling in human tumors provides novel molecular targets for therapeutic in tervention. Clin Cancer Res, 2002. 8 (4): p. 945-54. 85. Shuai, K., et al., Constitutive activation of STAT5 by the BCR-ABL oncogene in chronic myelogenous leukemia. Oncogene, 1996. 13 (2): p. 247-54. 86. Nieborowska-Skorska, M., et al., Signal transducer and activator of transcription (STAT)5 activation by BCR/ABL is dependent on intact Src
133 homology (SH)3 and SH2 domains of BCR/ABL and is requi red for leukemogenesis. J Exp Med, 1999. 189 (8): p. 1229-42. 87. Kisseleva, T., et al., Signaling through the JAK/STAT pathway, recent advances and future challenges. Gene, 2002. 285 (1-2): p. 1-24. 88. Matsumura, I., et al., Transcriptional regulation of the cyclin D1 promoter by STAT5: its involvement in cytokine-dependent growth of hematopoietic cells. EMBO J, 1999. 18 (5): p. 1367-77. 89. Magne, S., et al., STAT5 and Oct-1 form a stable complex that modulates cyclin D1 expression. Mol Cell Biol, 2003. 23 (24): p. 8934-45. 90. Raftopoulou, M. and A. Hall, Cell migration: Rho GTPases lead the way. Dev Biol, 2004. 265 (1): p. 23-32. 91. Blanchard, J.M., Small GTPases, adhesion, cell cycle control and proliferation. Pathol Biol (Paris), 2000. 48 (3): p. 318-27. 92. Diaz-Blanco, E., et al., Molecular signature of CD34(+) hematopoietic stem and progenitor cells of patients with CML in chroni c phase. Leukemia, 2007. 21 (3): p. 494-504. 93. Daubon, T., et al., Differential motility of p190bcr-abland p210bcr-abl expressing cells: respective roles of Vav and Bcr-Abl GEFs. Oncogene, 2008. 27 (19): p. 2673-85. 94. Gordon, M.Y., et al., Adhesive defects in chronic myeloid leukemia. Curr Top Microbiol Immunol, 1989. 149 : p. 151-5.
134 95. Zhao, R.C., et al., Gene therapy for chronic myelogenous leukemia (CML): a retroviral vector that renders hematopoietic progenito rs methotrexateresistant and CML progenitors functionally normal and no ntumorigenic in vivo. Blood, 1997. 90 (12): p. 4687-98. 96. Liu, P., et al., Activating mutations of Nand K-ras in multiple myeloma show different clinical associations: analysis of the Eastern Cooperative Oncology Group Phase III Trial. Blood, 1996. 88 (7): p. 2699-706. 97. Oliner, H., et al., Interstitial pulmonary fibrosis following busulfan thera py. Am J Med, 1961. 31 : p. 134-9. 98. Garcia-Manero, G., et al., Treatment of Philadelphia chromosome-positive chronic myelogenous leukemia with weekly polyethylene glycol formulation of interferon-alpha-2b and low-dose cytosi ne arabinoside. Cancer, 2003. 97 (12): p. 3010-6. 99. Bonifazi, F., et al., Chronic myeloid leukemia and interferon-alpha: a study of complete cytogenetic responders. Blood, 2001. 98 (10): p. 3074-81. 100. Thomas, E.D., et al., Marrow transplantation for the treatment of chronic myelogenous leukemia. Ann Intern Med, 1986. 104 (2): p. 155-63. 101. Gratwohl, A., et al., Indications for haemopoietic precursor cell transplants in Europe. European Group for Blood and Marrow Tran splantation (EBMT). Br J Haematol, 1996. 92 (1): p. 35-43. 102. Gratwohl, A., et al., Risk assessment for patients with chronic myeloid leukaemia before allogeneic blood or marrow transplant ation. Chronic
135 Leukemia Working Party of the European Group for Blood and Marrow Transplantation. Lancet, 1998. 352 (9134): p. 1087-92. 103. Manley, P.W., et al., Imatinib: a selective tyrosine kinase inhibitor. Eur J Cancer, 2002. 38 Suppl 5 : p. S19-27. 104. Druker, B.J., et al., Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med, 1996. 2 (5): p. 561-6. 105. le Coutre, P., et al., In vivo eradication of human BCR/ABL-positive leukemia cells with an ABL kinase inhibitor. J Natl Cancer Inst, 1999. 91 (2): p. 163-8. 106. Buchdunger, E., et al., Inhibition of the Abl protein-tyrosine kinase in vitro and in vivo by a 2-phenylaminopyrimidine derivative. Cancer Res, 1996. 56 (1): p. 100-4. 107. Sawyers, C.L., et al., Imatinib induces hematologic and cytogenetic responses in patients with chronic myelogenous leukemia in m yeloid blast crisis: results of a phase II study. Blood, 2002. 99 (10): p. 3530-9. 108. Gorre, M.E., et al., Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science, 2001. 293 (5531): p. 876-80. 109. Hofmann, W.K., et al., Ph(+) acute lymphoblastic leukemia resistant to the tyrosine kinase inhibitor STI571 has a unique BCR-ABL ge ne mutation. Blood, 2002. 99 (5): p. 1860-2.
136 110. Nardi, V., M. Azam, and G.Q. Daley, Mechanisms and implications of imatinib resistance mutations in BCR-ABL. Curr Opin Hematol, 2004. 11 (1): p. 35-43. 111. Deininger, M., E. Buchdunger, and B.J. Druker, The development of imatinib as a therapeutic agent for chronic myeloid leuke mia. Blood, 2005. 105 (7): p. 2640-53. 112. Skaggs, B.J., et al., Phosphorylation of the ATP-binding loop directs oncogenicity of drug-resistant BCR-ABL mutants. Proc Natl Acad Sci U S A, 2006. 103 (51): p. 19466-71. 113. Shah, N.P., et al., Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imat inib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell, 2002. 2 (2): p. 117-25. 114. le Coutre, P., et al., Induction of resistance to the Abelson inhibitor STI571 in human leukemic cells through gene amplification. Blood, 2000. 95 (5): p. 1758-66. 115. Sirulink, A., R.T. Silver, and V. Najfeld, Marked ploidy and BCR-ABL gene amplification in vivo in a patient treated with STI57 1. Leukemia, 2001. 15 (11): p. 1795-7. 116. Morel, F., et al., Double minutes containing amplified bcr-abl fusion gene in a case of chronic myeloid leukemia treated by imatinib. Eur J Haematol, 2003. 70 (4): p. 235-9.
137 117. Campbell, L.J., et al., BCR/ABL amplification in chronic myelocytic leukemia blast crisis following imatinib mesylate administr ation. Cancer Genet Cytogenet, 2002. 139 (1): p. 30-3. 118. Dulucq, S., et al., Multidrug resistance gene (MDR1) polymorphisms are associated with major molecular responses to standard-dose imatinib in chronic myeloid leukemia. Blood, 2008. 112 (5): p. 2024-7. 119. Mahon, F.X., et al., MDR1 gene overexpression confers resistance to imatinib mesylate in leukemia cell line models. Blood, 2003. 101 (6): p. 2368-73. 120. Burger, H., et al., Chronic imatinib mesylate exposure leads to reduced intracellular drug accumulation by induction of the ABCG 2 (BCRP) and ABCB1 (MDR1) drug transport pumps. Cancer Biol Ther, 2005. 4 (7): p. 747-52. 121. Weisberg, E., et al., Characterization of AMN107, a selective inhibitor of native and mutant Bcr-Abl. Cancer Cell, 2005. 7 (2): p. 129-41. 122. O'Hare, T., et al., AMN107: tightening the grip of imatinib. Cancer Cell, 2005. 7 (2): p. 117-9. 123. le Coutre, P., et al., Nilotinib (formerly AMN107), a highly selective BCRABL tyrosine kinase inhibitor, is active in patients with i matinib-resistant or -intolerant accelerated-phase chronic myelogenous leukemia. Blood, 2008. 111 (4): p. 1834-9.
138 124. Kantarjian, H., et al., Nilotinib in imatinib-resistant CML and Philadelphia chromosome-positive ALL. N Engl J Med, 2006. 354 (24): p. 2542-51. 125. Shah, N.P., et al., Overriding imatinib resistance with a novel ABL kinase inhibitor. Science, 2004. 305 (5682): p. 399-401. 126. Talpaz, M., et al., Dasatinib in imatinib-resistant Philadelphia chromosome-positive leukemias. N Engl J Med, 2006. 354 (24): p. 253141. 127. Graham, S.M., et al., Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood, 2002. 99 (1): p. 319-25. 128. Holtz, M.S., et al., Imatinib mesylate (STI571) inhibits growth of primitive malignant progenitors in chronic myelogenous leukemia th rough reversal of abnormally increased proliferation. Blood, 2002. 99 (10): p. 3792-800. 129. Bhatia, R., et al., Persistence of malignant hematopoietic progenitors in chronic myelogenous leukemia patients in complete cytogeneti c remission following imatinib mesylate treatment. Blood, 2003. 101 (12): p. 4701-7. 130. Roche-Lestienne, C., et al., Several types of mutations of the Abl gene can be found in chronic myeloid leukemia patients resistant to STI571, and they can pre-exist to the onset of treatment. Blood, 2002. 100 (3): p. 1014-8.
139 131. Hofmann, W.K., et al., Presence of the BCR-ABL mutation Glu255Lys prior to STI571 (imatinib) treatment in patients wit h Ph+ acute lymphoblastic leukemia. Blood, 2003. 102 (2): p. 659-61. 132. Zion, M., et al., Progressive de novo DNA methylation at the bcr-abl locus in the course of chronic myelogenous leukemia. Proc Natl Acad Sci U S A, 1994. 91 (22): p. 10722-6. 133. Asimakopoulos, F.A., et al., ABL1 methylation is a distinct molecular event associated with clonal evolution of chronic myeloid leukemi a. Blood, 1999. 94 (7): p. 2452-60. 134. Zhang, S.J., et al., Gain-of-function mutation of GATA-2 in acute myeloid transformation of chronic myeloid leukemia. Proc Natl Acad Sci U S A, 2008. 105 (6): p. 2076-81. 135. Damiano, J.S., L.A. Hazlehurst, and W.S. Dalton, Cell adhesion-mediated drug resistance (CAM-DR) protects the K562 chronic myelogen ous leukemia cell line from apoptosis induced by BCR/ABL inhi bition, cytotoxic drugs, and gamma-irradiation. Leukemia, 2001. 15 (8): p. 1232-9. 136. Hazlehurst, L.A., R.F. Argilagos, and W.S. Dalton Beta1 integrin mediated adhesion increases Bim protein degradation and contribut es to drug resistance in leukaemia cells. Br J Haematol, 2007. 136 (2): p. 269-75. 137. Garrido, S.M., et al., Acute myeloid leukemia cells are protected from spontaneous and drug-induced apoptosis by direct contact wit h a human
140 bone marrow stromal cell line (HS-5). Exp Hematol, 2001. 29 (4): p. 44857. 138. Tabe, Y., et al., Activation of integrin-linked kinase is a critical prosurvival pathway induced in leukemic cells by bone marrow-derived str omal cells. Cancer Res, 2007. 67 (2): p. 684-94. 139. Williams, R.T., W. den Besten, and C.J. Sherr, Cytokine-dependent imatinib resistance in mouse BCR-ABL+, Arf-null lymphob lastic leukemia. Genes Dev, 2007. 21 (18): p. 2283-7. 140. Wang, Y., et al., Adaptive secretion of granulocyte-macrophage colonystimulating factor (GM-CSF) mediates imatinib and nilo tinib resistance in BCR/ABL+ progenitors via JAK-2/STAT-5 pathway activation. Blood, 2007. 109 (5): p. 2147-55. 141. Liu, J., et al., BCR-ABL mutants spread resistance to non-mutated cells through a paracrine mechanism. Leukemia, 2008. 22 (4): p. 791-9. 142. Baker, S.J., S.G. Rane, and E.P. Reddy, Hematopoietic cytokine receptor signaling. Oncogene, 2007. 26 (47): p. 6724-37. 143. Flores-Morales, A., et al., In vitro interaction between STAT 5 and JAK 2; dependence upon phosphorylation status of STAT 5 and JAK 2. Mol Cell Endocrinol, 1998. 138 (1-2): p. 1-10. 144. Yu, C.L., et al., Enhanced DNA-binding activity of a Stat3-related prote in in cells transformed by the Src oncoprotein. Science, 1995. 269 (5220): p. 81-3.
141 145. Gesbert, F. and J.D. Griffin, Bcr/Abl activates transcription of the Bcl-X gene through STAT5. Blood, 2000. 96 (6): p. 2269-76. 146. Epling-Burnette, P.K., et al., Inhibition of STAT3 signaling leads to apoptosis of leukemic large granular lymphocytes and decrea sed Mcl-1 expression. J Clin Invest, 2001. 107 (3): p. 351-62. 147. Song, L., et al., Activation of Stat3 by receptor tyrosine kinases and cytokines regulates survival in human non-small cell carcinoma cells. Oncogene, 2003. 22 (27): p. 4150-65. 148. Garcia, R., et al., Constitutive activation of Stat3 by the Src and JAK tyrosine kinases participates in growth regulation of hum an breast carcinoma cells. Oncogene, 2001. 20 (20): p. 2499-513. 149. Catlett-Falcone, R., et al., Constitutive activation of Stat3 signaling confers resistance to apoptosis in human U266 myeloma cells. Immunity, 1999. 10 (1): p. 105-15. 150. Coppo, P., et al., Constitutive and specific activation of STAT3 by BCRABL in embryonic stem cells. Oncogene, 2003. 22 (26): p. 4102-10. 151. Spiekermann, K., et al., Constitutive activation of STAT3 and STAT5 is induced by leukemic fusion proteins with protein tyrosine k inase activity and is sufficient for transformation of hematopoietic p recursor cells. Exp Hematol, 2002. 30 (3): p. 262-71.
142 152. Kotecha, N., et al., Single-cell profiling identifies aberrant STAT5 activation in myeloid malignancies with specific clinical and biologic correlates. Cancer Cell, 2008. 14 (4): p. 335-43. 153. Fantin, V.R., et al., Constitutive activation of signal transducers and activators of transcription predicts vorinostat resistance in cutaneous Tcell lymphoma. Cancer Res, 2008. 68 (10): p. 3785-94. 154. Carlesso, N., D.A. Frank, and J.D. Griffin, Tyrosyl phosphorylation and DNA binding activity of signal transducers and activators of transcription (STAT) proteins in hematopoietic cell lines transformed by Bcr/Abl. J Exp Med, 1996. 183 (3): p. 811-20. 155. Ilaria, R.L., Jr. and R.A. Van Etten, P210 and P190(BCR/ABL) induce the tyrosine phosphorylation and DNA binding activity of mult iple specific STAT family members. J Biol Chem, 1996. 271 (49): p. 31704-10. 156. van der Plas, D.C., et al., Interleukin-7 signaling in human B cell precursor acute lymphoblastic leukemia cells and murine BAF3 cells invo lves activation of STAT1 and STAT5 mediated via the interle ukin-7 receptor alpha chain. Leukemia, 1996. 10 (8): p. 1317-25. 157. Donato, N.J., et al., Down-regulation of interleukin-3/granulocytemacrophage colony-stimulating factor receptor beta-chain i n BCR-ABL(+) human leukemic cells: association with loss of cytokine-mediate d Stat-5 activation and protection from apoptosis after BCR-ABL i nhibition. Blood, 2001. 97 (9): p. 2846-53.
143 158. Poincloux, R., et al., Tyrosine-phosphorylated STAT5 accumulates on podosomes in Hck-transformed fibroblasts and chronic myeloid leukemia cells. J Cell Physiol, 2007. 213 (1): p. 212-20. 159. Nam, S., et al., Dasatinib (BMS-354825) inhibits Stat5 signaling associated with apoptosis in chronic myelogenous leukemia cell s. Mol Cancer Ther, 2007. 6 (4): p. 1400-5. 160. Torok-Storb, B., et al., Dissecting the marrow microenvironment. Ann N Y Acad Sci, 1999. 872 : p. 164-70. 161. O'Hare, T., A.S. Corbin, and B.J. Druker, Targeted CML therapy: controlling drug resistance, seeking cure. Curr Opin Genet Dev, 2006. 16 (1): p. 92-9. 162. Ahmed, S.T. and L.B. Ivashkiv, Inhibition of IL-6 and IL-10 signaling and Stat activation by inflammatory and stress pathways. J Immunol, 2000. 165 (9): p. 5227-37. 163. Bartoli, M., et al., VEGF differentially activates STAT3 in microvascular endothelial cells. FASEB J, 2003. 17 (11): p. 1562-4. 164. Dani, C., et al., Paracrine induction of stem cell renewal by LIF-deficient cells: a new ES cell regulatory pathway. Dev Biol, 1998. 203 (1): p. 149-62. 165. Mangan, J.K., et al., Granulocyte colony-stimulating factor-induced upregulation of Jak3 transcription during granulocytic dif ferentiation is mediated by the cooperative action of Sp1 and Stat3. Oncogene, 2006. 25 (17): p. 2489-99.
144 166. Miranda, M.B., et al., Cytokine-induced myeloid differentiation is dependent on activation of the MEK/ERK pathway. Leuk Res, 2005. 29 (11): p. 1293-306. 167. Schuringa, J.J., et al., Constitutive Stat3, Tyr705, and Ser727 phosphorylation in acute myeloid leukemia cells caused by the autocrine secretion of interleukin-6. Blood, 2000. 95 (12): p. 3765-70. 168. Coppo, P., et al., BCR-ABL activates STAT3 via JAK and MEK pathways in human cells. Br J Haematol, 2006. 134 (2): p. 171-9. 169. de Groot, R.P., et al., STAT5 activation by BCR-Abl contributes to transformation of K562 leukemia cells. Blood, 1999. 94 (3): p. 1108-12. 170. Danhauser-Riedl, S., et al., Activation of Src kinases p53/56lyn and p59hck by p210bcr/abl in myeloid cells. Cancer Res, 1996. 56 (15): p. 3589-96. 171. Weisberg, E., et al., Stromal-mediated protection of tyrosine kinase inhibitor-treated BCR-ABL-expressing leukemia cells. Mol Cancer Ther, 2008. 7 (5): p. 1121-9. 172. Cortes, J. and M.E. O'Dwyer, Clonal evolution in chronic myelogenous leukemia. Hematol Oncol Clin North Am, 2004. 18 (3): p. 671-84, x. 173. Michor, F., et al., Dynamics of chronic myeloid leukaemia. Nature, 2005. 435 (7046): p. 1267-70. 174. Guilhot, F., et al., Dasatinib induces significant hematologic and cytogenetic responses in patients with imatinib-resistant o r -intolerant
145 chronic myeloid leukemia in accelerated phase. Blood, 2007. 109 (10): p. 4143-50. 175. Roecklein, B.A. and B. Torok-Storb, Functionally distinct human marrow stromal cell lines immortalized by transduction with the human papilloma virus E6/E7 genes. Blood, 1995. 85 (4): p. 997-1005. 176. Horita, M., et al., Blockade of the Bcr-Abl kinase activity induces apoptosis of chronic myelogenous leukemia cells by suppressing signal tr ansducer and activator of transcription 5-dependent expression of Bcl-xL. J Exp Med, 2000. 191 (6): p. 977-84. 177. Huang, M., et al., Inhibition of Bcr-Abl kinase activity by PD180970 blocks constitutive activation of Stat5 and growth of CML cells. Oncogene, 2002. 21 (57): p. 8804-16. 178. Aichberger, K.J., et al., Identification of mcl-1 as a BCR/ABL-dependent target in chronic myeloid leukemia (CML): evidence for coo perative antileukemic effects of imatinib and mcl-1 antisense oligo nucleotides. Blood, 2005. 105 (8): p. 3303-11. 179. de Groot, R.P., et al., STAT5-Dependent CyclinD1 and Bcl-xL expression in Bcr-Abl-transformed cells. Mol Cell Biol Res Commun, 2000. 3 (5): p. 299-305. 180. Dumon, S., et al., IL-3 dependent regulation of Bcl-xL gene expression by STAT5 in a bone marrow derived cell line. Oncogene, 1999. 18 (29): p. 4191-9.
146 181. Sawyers, C.L., Where lies the blame for resistance--tumor or host? Nat Med, 2007. 13 (10): p. 1144-5. 182. Turkson, J., et al., A novel platinum compound inhibits constitutive Stat3 signaling and induces cell cycle arrest and apoptosis of mali gnant cells. J Biol Chem, 2005. 280 (38): p. 32979-88. 183. Sun, J., et al., Cucurbitacin Q: a selective STAT3 activation inhibitor wit h potent antitumor activity. Oncogene, 2005. 24 (20): p. 3236-45. 184. Turkson, J., et al., Phosphotyrosyl peptides block Stat3-mediated DNA binding activity, gene regulation, and cell transformat ion. J Biol Chem, 2001. 276 (48): p. 45443-55. 185. Liszewski, M.K. and J.P. Atkinson, Membrane cofactor protein (MCP; CD46). Isoforms differ in protection against the classical pathway of complement. J Immunol, 1996. 156 (11): p. 4415-21.
147 PRESENTATION OF STUDIES Publications resulting from these studies include one fi rst-authored paper in press, and one manuscript in press. These results were also p resented as an oral presentation at the American Association of Cancer Research 2008 Annual Meeting. These presentations and papers are listed below Bewry NN Rajesh R. Nair, Michael F. Emmons, David Boulware, Lo ri A. Hazlehurst. 2008. STAT3 contributes to resistance towards BCR-ABL inhibitors in a bone marrow microenvironment model of drug resist ance (Manuscript in press) Bewry NN Rajesh R. Nair, Michael F. Emmons, Lori A. Hazlehurst 2008. Bone marrow stromal cells activate STAT3 and confer resistance t o BCR-ABL inhibitors in K562 CML cells. American Association of Can cer Research Annual Meeting; Minisymposia: Molecular Mechanisms of Drug Resist ance, San Diego, CA Bewry NN, Rajesh R. Nair, Michael F. Emmons, Lori A. Hazlehurst. 2007. Role of the Bone Marrow Microenvironment and STAT3 Activation in the Imatinib-
148 resistant Phenotype of K562 Chronic Myeloid Leukemia (CM L) Cells. Seminar, Department of Molecular Medicine, USF, Tampa, FL Bewry NN, Rajesh R. Nair, Michael F. Emmons, Lori A. Hazlehurst. 2007. Role of the Bone Marrow Microenvironment in Mediating Drug R esistance in Chronic Myeloid Leukemia (CML). Department of Molecular Medi cine Scientific Retreat, Brookville, FL Bewry NN Rajesh R. Nair, Michael F. Emmons, Lori A. Hazlehurst 2007. Imatinib Resistance in Chronic Myeloid Leukemia (CML). McKnight Doctoral Fellowship Mid-Year Research & Writing Conference, Tam pa, FL Bewry NN Emmons M., Dalton W., Hazlehurst L. 2007. Contrib ution of the bone marrow microenvironment in mediating resistance to BcrAbl inhibitors in Chronic Myeloid Leukemia (CML). Moffitt Scientific Retreat, T ampa, FL.
149 ABOUT THE AUTHOR Nadine N. Bewry completed her undergraduate studies at Tennessee State University with a Bachelors of Science degree in B iology. In 2004 she entered the University of South Florida as a doctorate student in the Medical Sciences Program at the College of Medicine. During he r matriculation, Nadine was a recipient of the Diversity Student Success Fellowship and the Genshaft Family Doctoral Fellowship. Additionally, Nadine was the recipient of the American Association for Cancer Researchs Brigid G. Leven thal Scholar-inTraining Award. She was also invited to give an oral presentation at AACRs 2008 Annual Meeting at the Minisymposia on the Molecula r Mechanisms of Drug Resistance. Nadine has published two manuscripts, one in the Journal of Biological Chemistry (2007) and the other in Molecular Cancer Therapeutics (2008), which is based on her doctoral research. Nadines two wonderful children, Yishmael and Tovliyah, are her blessing, as they keep her happy, motivated and energized.