CLONALITY AND CYCLING STATUS OF LEUKEMIC PROGENITORS FROM PATIENTS WITH ACUTE MYELOID LEUKEMIA (AML) By YINGHUI GUAN B.Sc (Ecology and Environmental Biology), Peking University, 1992 M.Sc. (Biochemistry and Molecular Biology), Peking University, 1995 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Pathology and Laboratory Medicine We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 2002 UBC Rare Books and Special Collections - Thesis Authorisation Form Page 1 of 1 In presenting this thesis i n p a r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of this thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of ytCH^of&*J& j-CK. The University of B r i t i s h Columbia Vancouver, Canada http://www.library.ubc.ca/spcoll/thesauth.html 4/2/2003 Abstract We have detected a high concentration of cytogenetically normal long term culture-initiating cells (LTC-IC) in the peripheral blood (PB) of 12 patients with newly-diagnosed AML. To determine if these progenitors originated from normal polyclonal hematopoiesis the PCR-based human androgen receptor allele (HUMARA) assay was used to study PB cells from 5 female patients with cytogenetically abnormal AML. In 4/5 samples cytogenetics and clonality data from LTC indicate that a substantial number of normal polyclonal hematopoietic progenitors often persist in A M L PB at diagnosis. However, the fifth A M L sample contrasted with the others since the HUMARA assay demonstrated that LTC-derived colonies were predominantly clonal although the majority of the progenitors were cytogenetically normal. Thus, it appears that the abnormality detected on routine cytogenetics is not a reliable marker of the leukemic clone in this case. The second goal of this thesis was to characterize cycling status of different leukemic progenitors. An overnight 3H-thymidine (3H-Tdr) suicide assay was used to analyze the proliferative status of malignant progenitors detected in CFC, LTC-IC and NOD/SCID mouse leukemia initiating cell (NOD/SL-IC) assays from the peripheral blood of 15 patients with newly-diagnosed AML. FISH analysis of colonies from CFC and LTC-IC assays confirmed that most cytogenetically abnormal CFC and LTC-IC from the 15 samples as well as cytogenetically normal LTC-IC detected were actively cycling. However, the same assay has demonstrated that some A M L LTC-IC and most NOD/SL-IC were largely quiescent. The growth factor responsiveness of quiescent A M L cells was then studied and their functional properties compared to those of AML cells in active cell cycle. Hoechst 33342/Pyronin Y staining and FACS sorting were used to isolate Go cells in from the Gi and S/G2+M subpopulations of peripheral blood cells from 4 newly diagnosed A M L patients. Go ii A M L cells from these patients exhibited a strong tendency to enter active cell cycle when cultured even in the absence of supportive growth factors. LTC-ICs were easily detected among cycling A M L cells that had been cultured for 72 hours. In contrast, the exit from Go dramatically reduced the ability of A M L progenitors to engraft in mouse bone marrow. iii Table of Contents ABSTRACT II TABLE OF CONTENTS IV LIST OF TABLES IX LIST OF FIGURES XI LIST OF ABBREVIATIONS XIII ACKNOWLEDGMENTS XV CHAPTER 1 INTRODUCTION 1 1.1 NORMAL AND LEUKEMIC HEMA TOPOIESIS 2 1.1.1 Normal hematopoiesis and functional assays 2 1.1.1.1 Colony forming cell (CFC) assay 3 1.1.1.2 Long term culture-initiating cell assay 4 1.1.1.3 In vivo repopulation assay with immunodeficient mice 5 1.1.2 Acute myeloid leukemia (AML): clinical features and cell biology 6 1.1.2.1 General characteristics of A M L 6 1.1.2.2 Clonal origin of AML and leukemic hierarchy 6 1.1.3 Markers of clonality 9 1.1.3.1 XCIP and clonality markers on X chromosome 9 1.1.3.2 Interpretation of XCIP results 14 1.2 DEREGULATED GENE EXPRESSION IN AML 16 1.2.1 Cytogenetic abnormalities in AML patients 16 1.2.2 Oncogenic transcription factors in AML 20 iv 1.2.3 Autocrine production of growth factors in AML 22 1.2.4 Mutations in growth factor receptors 24 1.2.5 Abnormalities of genes involved in apoptosis 27 1.3 CELL CYCLE DEREGULATION IN AML 28 1.3.1 The mitotic cell cycle 28 1.3.2 Cell cycle analyses 28 1.3.2.1 Detection of DNA replication 28 1.3.2.2 Quantitation of cellular DNA and RNA 29 1.3.2.3 Studying cell cycle with markers of proliferation and cell cycle progression 30 1.3.3 Cell-cycle regulatory elements involved in AML 30 1.3.4. Cycling status and responses to growth factors of normal and leukemic hematopoietic progenitors 34 1.3.4.1 Cell cycle status and responses to growth factors of normal hematopoietic progenitors 34 1.3.4.2 Cell cycle status and deregulated proliferation of A M L 36 1.4 RATIONALE AND THESIS HYPOTHESES 37 CHAPTER 2 MATERIALS AND METHODS 40 2.1 AML CELLS 41 2.2 AML COLONY FORMING ASSAY. 41 2.3 AML LONG-TERM CULTURE ASSAY 42 2.4 GENERATION OFBONE MARROW FIBROBLASTS (BMF) FROM PATIENT BM CELLS 42 2.5 SHORT-TERM IN VITRO CULTURE. 42 2.6 EXTRACTION OF GENOMIC DNA FROM BULK AML CELLS AND METHYLCELLULOSE COLONIES 43 2.7 PCR BASED FLUORESCENT HUMARA ASSAY 43 2.8 GENESCANANALYSIS 44 2.9 QUANTITATION OF RELATIVE AMOUNTS OFPCR PRODUCTS 45 2.10 OVERNIGHT 3H-TDR SUICIDE ASSAY 46 2.11 FISH. 47 2.11.1 Metaphase Preparation 47 2.11.2 Slide Preparation 47 2.11.3 FISH probes 48 2.11.4 Slide Hybridization and Visualization 50 2.12 ANIMALS 51 2.13 TRANSPLANTATION OF AML CELLS IN NOD/SCID MICE 52 2.14 DETECTION OF HUMAN CELL ENGRAFTMENT IN MICE WITH FLOW CYTOMETRY 52 2.15 HOECHST 33342 (HST) AND PYRONIN Y (PY) STAINING. 55 2.16 HIGH-RESOLUTION CELL CYCLE ANALYSIS USING KI-67 ANTIBODY AND 7-AAD 55 2.17 STIMULA TION OF G0 CELLS INTO MITOTIC CELL CYCLE 57 2.18 RT-PCR ANALYSES 57 2.19 SOUTHERN BLOTTING 58 2.19.1 Gel Transfer 58 2.19.2 Probes 58 2.19.3 Probe Hybridization 59 2.20 ENZYME LINKED IMMUNOSORBENT ASSA Y (ELISA) AND BIO-A CTIVITYASSA Y 59 2.21 CALCULATIONS AND STATISTICAL ANALYSES 60 2.21.1 Calculations for % kill of A M L progenitors 60 2.21.2 statistical analysis 61 CHAPTER 3 POLYCLONAL NORMAL HEMATOPOIETIC PROGENITOR CELLS IN PERIPHERAL BLOOD OF PATIENTS WITH AML 62 3.1 INTRODUCTION. 63 3.2 RESULTS 65 vi 3.2.1 A M L Patients with Cytogenetically Normal LTC-ICs 65 3.2.2 Clonality of PB mononuclear cells and B M F from A M L patients 67 3.2.3 Clonality of CFC and LTC-IC in A M L PB 69 3.2.4 Purging A M L samples with growth factors 74 3.3 DISCUSSION. 76 CHAPTER 4 CHARACTERIZATION OF THE CYCLING STATUS OF LEUKEMIC PROGENITORS FROM PATIENTS WITH A M L 80 4.1 INTRODUCTION. 81 4.2 RESULTS 83 4.2.1 Maintenance of A M L progenitors in Short Term Culture 83 4.2.2 Cycling Status of Fresh and Cryopreserved A M L - C F C 87 4.2.3 Cycling Status of A M L LTC-IC 90 4.2.4 Cycling status of NOD/SL-IC 93 4.3 DISCUSSION. 95 CHAPTER 5 ISOLATION AND STIMULATION OF QUIESCENT PRIMITIVE LEUKEMIC PROGENITOR CELLS FROM PATIENTS WITH A M L 99 5.7 INTRODUCTION. 100 5.2 RESULTS 103 5.2.1 Isolation of A M L cells in Go, Gi and S/G 2+M phases using Hst/PY staining 103 5.2.2 Comparison of cycling status of A M L CFC and LTC-IC determined by 3H-Tdr suicide assay and Hst/PY sorting 108 5.2.3 NOD/SL-IC activity was restricted to G 0 A M L cells 110 5.2.4 The spontaneous entry of Go cells into cell cycle upon in vitro culture 113 5.2.5 Autocrine growth factor production and/or internal tandem duplication (ITD) associated constitutive activation of the Flt-3 receptor is related to autonomous proliferation of A M L cells 118 vii 5.3 DISCUSSION. 121 CHAPTER 6 CONCLUSIONS AND FUTURE PROSPECTIVES 126 REFERENCES 134 viii List of Tables Chapter 1 Table 1 Polymorphic loci commonly used in analysis of XCIP 10 Table 2 Prognostic classification of cytogenetic abnormalities, their association with FAB subtypes and frequencies in AML patients 18 Chapter 3 Table 3 Clinical and in vitro LTC characteristics of AML patients 66 Table 4 Comparison of the clonality of PB mononuclear cells and with corresponding bone marrow fibroblasts (BMF).... 69 Table 5 Clonality of CFC colonies 70 Table 6 Clonality of LTC-IC colonies 71 Table 7 Purging leukemic progenitors and expanding normal progenitors from AML patients with long-term culture and growth factors 75 Chapter 4 Table 8 The clinical characteristics of the AML patients 84 Table 9 Maintenance of A M L progenitors (CFC and LTC-IC) overnight in serum free medium supplemented with growth factors 85 Table 10 Comparison of maintenance (% input) of different A M L progenitors 86 Table 11 Comparison of the cycling status of AML-CFC between fresh and cryopreserved samples 87 Table 12 The cycling status of the AML-CFC 89 Table 13 Cycling status of 5- and 8-week AML LTC-IC 91 Table 14: The effect of over night exposure on cycling status of A M L CFC and AML LTC-IC 92 Table 15 Cycling status of NOD/SL-IC 94 Table 16 Comparison of # of abnormal/ # of total cells tested (% abnormal) in diagnostic B M and among human CD45+ engrafted in NOD/SCID mice 8 weeks after injection 94 ix Chapter 5 Table 17 The clinical characteristics of the AML patients 105 Table 18 Fractionation of AML PB cells by Hst/PY staining 106 Table 19 % Purity of different cell cycle subpopulations determined by anti-Ki-67 antibody and 7-AAD 107 Table 20 Engraftment of different Hst/PY stained cell cycle subpopulations in NOD/SCID mice 112 Table 21 Stimulation of AML G 0 cells to enter cell cycle 115 Table 22 Maintenance of AML CFC and AML LTC-IC capacity as A M L cells exit G 0 116 X List of Figures Chapter 1 Figure 1 Normal and leukemic hematopoietic hierarchies 8 Figure 2 HUMARA clonality assay 13 Figure 3 The structure of Flt-3 receptor and mutations found in A M L patients 26 Figure 4 Cell cycle regulation 32 Chapter 2 Figure 5. The relationship between the relative amounts of two samples of DNA in a mixture and the peak area ratio (PAR) of the PCR products amplified from this mixture 46 Figure 6 Detection and determination of human cell engraftment in NOD/SCID mice 54 Figure 7 Reanalysis of cells stained with Hst/PY to determine their cell cycle status according to their staining with anti-Ki-67-FITC and 7-AAD 56 Chapter 3 Figure 8 Genescan profiles of PCR products of genomic DNA from AML blasts and bone marrow fibroblasts (BMF) from patient 5 68 Figure 9. Representative Genescan profiles of LTC-IC-derived colonies from patient 5 72 Chapter 5 Figure 10 Hst/PY sorting 104 Figure 11 CFC and LTC-IC distribution in different cell cycle subpopulations 109 Figure 12 Short-term SFM culture of Go cells with and without growth factors 114 Figure 13 Reduced capacity to engraft NOD/SCID mice as Go A M L cells enter into active cell cycle 117 Fi gure 14 Growth factor mRNA expression in AML Blasts 119 Figure 15 Flt-3 ITD in A M L patient samples 120 xi Chapter 6 Figure 16 Hypothetical model for the cycling status and cell cycle regulation of different AML progenitors 132 xii List of Abbreviations 3H-Tdr tritiated thymidine 4- HC 4-hydroperoxycyclophosphamide 5- FU 5-fiuorouracil AML acute myeloid leukemia AML-CFC acute myeloid leukemia colony forming cells A M L LTC-IC acute myeloid leukemia long-term culture initiating cells AR androgen receptor BAC bacteria artificial chromosome BFU-E burst forming unit erythroid B M bone marrow BSA bovine serum albumin CFU-GEMM colony forming unit granulocyte, erythrocyte, macrophage, megakaryocyte CFU-GM colony forming unit granulocyte macrophage CFU-S colony forming unit spleen CGH comparative genome hybridization cGy centiGrey Ci Curie CML chronic myeloid leukemia dATP deoxyadenosine triphosphate dCTP deoxycytosine triphosphate dGTP deoxyguanine triphosphate DIG digoxigenin DMEM Dulbecco's modified Eagle's medium dTTP deoxythymidine triphosphate dUTP deoxyuridine triphosphate EDTA ethylene diamine tetraacetic acid Epo erythropoietin FAB French American British FACS flurorescence activated cell sorting FCS fetal calf serum FISH fluorescence in situ hybridization FITC fluorescein isothiocyanate FL flt-3 ligand GAPDH glyceraldehyde-3 -phosphate-dehydrogenase G-CSF granulocyte colony stimulating factor GM-CSF granulocyte-macrophage colony stimulating factor Gy Grey HCP hematopoietic cell phosphatase HFN Hanks balanced salt solution with 2%FCS and 0.02% Sodium Azide HSC hematopoietic stem cell Hst hoechst 33342 HUMARA human androgen receptor allele ITD internal tandem duplication IL-1 interleukin 1 IL-2 interleukin 2 xiii IL-3 interleukin 3 IL-6 interleukin 6 IMDM Iscove's modified Dulbecco's medium ' L ' allele: leukemia related allele Lin" lineage negative LTC long-term culture LTC-IC long-term culture initiating cell MPB mobilized peripheral blood NBM normal bone marrow NOD/SCID non-obese diabetic/severe combined immunodeficient NOD/SL-IC NOD/SCID mouse leukemia initiating cell PAR peak area ratio PBS phosphate buffered saline PI propidium iodide PNM phosphate nonidet P-40 buffer with skim milk powder PY pyronin Y Rb retinoblastoma RT-PCR reverse transcriptase polymerase chain reaction SC-IC suspension culture initiating cells SCID sever combined immunodeficient SKY spectrum karyotyping SDS sodium dodecyl sulphate s.d. standard deviation SF steel factor SFM serum free medium TE Tris EDTA buffer TGF p transforming growth factor beta TNFa tumor necrosis factor alpha Tpo thrombopoietin Xm maternal X chromosome Xp paternal X chromosome XCJP X chromosome inactivation pattern YAC yeast artificial chromosome XIV Acknowledgments First, I would like thank my supervisor, Dr. Donna Hogge, for giving me the opportunity to work on these projects, and for her many years of patient guidance and support with which I have come this far from knowing nothing about leukemias. I would also like to thank Gitte Gerhard, for her great technical expertise and organizational skills in the lab, as well as her being such a good friend with me. Thank you to other members who have worked in Hogge's lab, Michaela, Hiro, and Laurie, for providing a friendly lab atmosphere. I thank the members of my supervisory committee, Dr. Jurgen Vielkind, Dr. Gerry Krystal, Dr. Hermann Ziltener and Dr. Calvin Roskelly, for giving me advice and keeping me on track throughout the years. I would also like to thank Dr. Heather Sutherland, Suzanne Vercauteren and Sharon Louise for useful suggestions on our joint lab meetings. Thank you, Xiao yan, for your help on so many things. Janet, I appreciate your kind effort on proofreading my thesis. I would like thank my parents, for always supporting me for whatever I do and their unbelievable help on taking care of my daughter, Melody. A big thank to my friends inside and outside Terry Fox, for all the chats we had on our work and life, and all the fun we had, whether it be having lunch in parks around the corner, or having a girls' only candle light party, Eleni, Silvia, Nadine and Vivian, Lixin and Grace. I would like to thank my little daughter for providing me with constant inspiration to do my best. Last but not the least, I would like to thank my dear husband for his love, for always encouraging me and being there with me in good times and bad. XV Chapter 1 Introduction 1 1.1 Normal and leukemic hematopoiesis 1.1.1 Normal hematopoiesis and functional assays Every day, billions of blood cells with different morphologies and functions die and are replaced. These cells are highly differentiated and therefore not capable of proliferation. Instead, their continuous production throughout life relies upon the differentiation and proliferation of a rare population of very primitive hematopoietic progenitor cells called "hematopoietic stem cells" (HSCs) present at the highest concentration in the bone marrow (Eaves & Eaves, 1988). The essential properties of HSCs are their extensive proliferative capacity and their ability to both reproduce their own numbers and differentiate through a hierarchical progression of less primitive lineage-committed progenitor cells to ultimately give rise to mature lymphoid and myeloid blood cells. The lineage-committed progenitor cells have a reduced capacity to proliferate and limited, if any, self-renewal capacity as compared to HSCs (Eaves & Eaves, 1988). In spite of the large number of mature blood cells that can arise from HSCs, the cell number in the steady state blood of a normal individual is maintained at a relatively constant level. Cell proliferation is balanced by differentiation and apoptosis (Ogawa, 1993). However, under conditions of stress and increased demand for blood cells (e.g. following hemorrhage or infection), the output of blood cells rapidly increases. This output is regulated by various growth factors, cytokines and cellular components present in the bone marrow microenvironment. In this microenvironment, there is a network comprised of stromal cells (fibroblasts, endothelial cells and adipocytes) as well as macrophages and lymphocytes (Senesebe et al., 1997). This network interacts with progenitors including HSCs and secretes positive growth factors such as colony-stimulating factors, tyrosine kinase receptor ligands, or negative regulators such as transforming growth factor p (TGF-P) and tumor necrosis factor (TNF) (Sensebe et al., 1997; Ogawa, 1993). 2 These growth factors/cytokines can act independently or synergistically with others to maintain or stimulate lineage commitment of HSCs by binding to cell surface receptors. Some cytokines are known to have various biological functions on different types of progenitors, but how they elicit these functions and what signal transduction pathways they initiate are not completely understood. However, there are many examples in which disruption of these pathways has led to deregulated proliferation, blocked differentiation or decreased cell death and ultimately resulted in hematological disorders (Radich et al., 1990; Preisler et al., 1997). Since hematopoietic progenitors are rare cells and morphologically indistinct, their identification relies primarily upon in vitro and in vivo functional assays (Eaves et al, 1997). There are three types of functional assays for human hematopoietic progenitors which will be used in the experiments described in this thesis: the colony forming cell assay, the long term culture-initiating cell assay and in vivo mouse repopulation assay. 1.1.1.1 Colony forming cell (CFC) assay When low concentrations of bone marrow or blood cells are plated in semi-solid methylcellulose medium supplemented with serum and specific recombinant growth factors critical for the differentiation of different lineages, discrete colonies of mature cells with distinctive morphologies will form after 10-14 days. The majority of colonies obtained under routine conditions are derived from single progenitors (Dube et al , 1981; Gandini and Gartler, 1969). The morphology and size of colonies reflect lineage commitment and proliferative potential of the hematopoietic progenitor from which they arose. These progenitors are called colony forming cells (CFC) or colony forming units (CFU). Sometimes, "mixed" colonies containing two or more cell types are present in cultures. The progenitors of these mixed colonies are thought to be more primitive than progenitors producing single-lineage colonies. Similarly, the larger the colony size the greater the proliferation potential and the less differentiated the progenitor from which the colony is derived. 3 1.1.1.2 Long term culture-initiating cell assay Another in vitro assay which detects hematopoietic progenitor cells that appear to be more primitive than CFCs is the long term culture initiating cell (LTC-IC) assay (Sutherland et al., 1990; Ploemacher et al., 1991). The observation that granulocytes and macrophages can be produced continuously for many months in cultures established by seeding unseparated marrow cells at high concentrations in liquid medium containing serum and hydrocorticosone led to the development of the long-term bone marrow culture (LTC) system (Dexter et al., 1977; Gartner and Kaplan, 1980). Over the initial 2 or 3 weeks in LTC, the nonhematopoietic stromal elements present in bone marrow form an adherent layer with which hematopoietic cells associate. In murine LTC, the stromal cell layer provides growth factors that help sustain the survival and proliferation of very primitive hematopoietic cells including those with in vivo lymphomyeloid repopulating activity (Fraser et al., 1992). In human LTC, the progenitor sustaining the output of CFCs from these cultures for 5 or more weeks is called the long term culture initiating cell (LTC-IC) (Sutherland et al., 1989). Each LTC-IC on average gives rise to more than 1 daughter CFC. Further evidence that LTC-ICs are a more primitive population than CFCs came from cell surface phenotyping studies in which LTC-ICs could be partially separated from CFCs by their higher expression of CD34 and lower expression of HLA-DR and CD38 (Sauvageau et al., 1994; Sutherland et al., 1989). These phenotypic features are similar to those progenitors capable of in vivo repopulation (Baum et al., 1992; Huang and Terstappen, 1994; Terstappen et al., 1991). LTC can also be initiated by coculturing hematopoietic cells, including highly purified progenitors with pre-established stromal feeder layers. Mouse stromal cell lines that have been engineered to produce several early acting human cytokines (e.g. interleukin 3 (IL-3), steel factor (SF), etc.) can replace human bone marrow stromal cells in supporting the growth and proliferation of hematopoietic progenitor cells in LTCs (Hogge et al., 1996; Otsuka et al., 1991; Sutherland et al., 1993). In one study, the optimal feeders for normal LTC-IC detection consists 4 of a combination of murine M2-10B4 fibroblasts (a cloned line of mouse bone marrow origin; (Lemoine et al., 1988)) engineered to produce high level of human granulocyte colony stimulating factor (G-CSF) and IL-3 and mixed 1:1 with Sl/Sl fibroblasts (an embryonic cell line derived from Sl/Sl mice, which are defective for the gene encoding SF (Sutherland et al, 1993)) engineered to produce soluble SF. The pre-established feeder layer permits CFC outputs to be determined by the number of LTC-IC in the input test cell suspension independent of the concentration of co-existing stromal cell precursors. 1.1.1.3 In vivo repopulation assay with immunodeficient mice In recent years, the most widely used method for quantifying human HSC candidates is to transplant human hematopoietic cells into sublethally irradiated immunodeficient mice (e.g. severe combined immunodeficient (SCID) or non-obese diabetic SCID (NOD/SCID) mice). Human hematopoietic cells will home to the marrow of these animals where they proliferate and differentiate to produce large numbers of lymphoid and myeloid progeny (Cashman et al., 1997a; Cashman et al , 1997b; Dick, 1996; Nolta et al., 1994; Pflumio et al , 1996; van der Loo et al., 1998). In addition to the T and B lymphocyte deficiencies seen in SCID mice, NOD/SCID mice also have decreased natural killer cell activity, macrophage function and complement activity (Shultz et al., 1995). As compared to SCID animals, the immunodeficiencies of NOD/SCID mice enable smaller numbers of human cells to engraft without the requirement for exogenous administration of human growth factors (Cashman et al., 1997b; Lapidot et al , 1994). Because of the increased sensitivity and reproducibility with which human cells can be detected in NOD/SCID recipients, it has been possible to use these animals to quantify human lymphomyeloid repopulating cells at limiting dilutions. The self-renewal capacity of these primitive progenitors can also be demonstrated by their ability to engraft secondary mice. Cell sorting studies also demonstrated that progenitors are enriched, similar to LTC-IC, in CD34+CD38" cells which lack markers of lineage commitment (Bhatia et al., 1997). Although 5 NOD/SCID engrafting cells and LTC-IC may be overlapping populations, they are likely not completely equivalent since the former are present at lower frequency in unseparated bone marrow, are relatively resistant to retrovirus transduction (Larochelle et al., 1996) and are quiescent (Glimm et al, 2000; Gothot et al., 1998). 1.1.2 Acute myeloid leukemia (AML): clinical features and cell biology 1.1.2.1 General characteristics of AML A M L is a clonal, malignant disease that originates from the transformation of a single hematopoietic progenitor cell. Malignant myeloblasts in which differentiation is relatively blocked accumulate in the bone marrow and peripheral blood of patients. Based on evidence of lineage differentiation observed in the morphology, histochemistry and cell surface phenotypes of leukemic blasts, A M L is categorized into the subtypes Mo-M 7 according to the French American British (FAB) classification (Schumacher, 1990). The disordered growth of these leukemic blasts suppresses normal blood cell production resulting in anemia, thrombocytopenia and neutropenia. Patients experience fatigue, weakness and weight loss as a result of anemia; easy bruising, gum bleeding and hemorrhage as a result of thrombocytopenia; and infections as a result of reduced levels of granulocytes. AML accounts for 90% of acute leukemia in adults. The median age of onset is 64 years. Although most younger patients will enter a complete remission with conventional chemotherapy and many can be cured, for patients older than 50 years of age the complete remission rate is only 30-50% (Leith et al., 1997). The 5-year survival among 553 patients over 55 years of age treated with Eastern Cooperative Oncology Group chemotherapy protocols between 1989 to 1997 was approximately 12% (Rowe, 2001). 1.1.2.2 Clonal origin of AML and leukemic hierarchy The clonal origin of the malignant cells in AML has been firmly established by previous studies on nonrandom chromosome abnormalities and X-chromosome inactivation pattern in 6 leukemic blasts (Yunis et al., 1981; Fialkow et al., 1981; Allen et al , 1992; Busque and Gilliland, 1998; Delabesse et al., 1995; Fearon et al., 1986; Gale et al., 1996). Recent studies on A M L leukemic progenitors have also suggested that in most cases A M L is originated from a primitive hematopoietic progenitor. This leukemic 'stem cell' is capable of limited differentiation to more mature leukemic progenitor subtypes and finally to leukemic blasts that have very limited proliferative potential (Sutherland et al , 1996) (Figure 1). The heterogeneity seen among blasts from different patients reflects the variable capacities of these transformed leukemic progenitors to differentiate. The hypothesis that this heterogeneity is determined by the specific genetic rearrangement in the leukemic clone rather than by the degree of commitment of the transformed cell is supported by the finding that the ability to initiate the long-term engraftment in NOD/SCID mice was found exclusively in the CD34+CD38" subset of leukemic cells from patients with a variety of AML subtypes (Bonnet and Dick, 1997). The in vitro and in vivo functional assays used to detect normal progenitors can be similarly used to detect malignant progenitors. For example, some leukemic cells can form blast cell colonies in colony assay and are thus called A M L CFCs, and others can survive at least 5 weeks in long-term cultures supported by pre-established feeder cells and generate A M L CFC carrying the cytogenetic abnormality characteristic of the leukemic clone. These latter leukemic progenitors were thus called AML long-term culture initiating cells (AML LTC-ICs) (Ailles et al., 1997). A M L progenitor cells detected in NOD/SCID mouse transplantation experiments have also been characterized recently (Ailles et al., 1999a; Bonnet and Dick, 1997). These NOD/SCID mouse leukemia-initiating cells (NOD/SL-ICs), like most normal NOD/SCID mouse engrafting cells and normal and AML LTC-IC, are CD34+CD38" and typically lack expression of CD71 and HLADR (Blair et al., 1998; Kawagoe et al., 1999). However, NOD/SL-ICs also typically lack the expression of Thy-1 and c-kit receptor (Blair et al., 1997; Blair and Sutherland, 2000), markers that are found on normal NOD/SCID mouse engrafting cells. The proliferation and 7 differentiation of these leukemic progenitors in vivo generates a leukemia in mice that retains many of the clinical characteristics of the disease in human patients (Ailles et al., 1999a; Bonnet and Dick, 1997). NOD/SCID engrafting ct /LTC-IC CFC Normal (D) Mature cells © m © NOD/SL-IC/ AML LTC-IC Leukemic (ID / \ / \ (D) (ID AML CFC (JJ) (D / \ / \ / \ / \ AML blasts Figure 1 Normal and leukemic hematopoietic hierarchies 8 1.1.3 Markers of clonality Since leukemic cells originate from a single transformed progenitor cell, analysis of clonality of a cell population plays an important role in understanding the disease state at the time of presentation and sheds light on events that take place in initiation and progression of the disease. To detect a monoclonal population in a polyclonal cell background requires a marker system that enables the progeny of the originally transformed cell to be recognized. A wide variety of assays are now available. The clonality markers include the somatic cytogenetic alterations specific for the disease, leukemic associated alterations such as the activation of oncogenes and inactivation of tumor suppressor genes and markers on the X chromosome which can be distinguished by the study of the X chromosome inactivation pattern (XCIP). Cytogenetic abnormalities and genetic changes in AML will be discussed in 1.2. In this section I will focus on the clonality assays based on the detection of XCIP. 1.1.3.1 XCIP and clonality markers on X chromosome It is now over 40 years since Dr Lyon first described the mosaicism of the X chromosome genes affecting coat color in mice (Lyon, 1961). The observation led to the hypothesis that the equivalence of expression of genes on the X chromosome between females and males is achieved through the inactivation of one of the two X chromosomes in early female embryogenesis. The X chromosome which remains active in each cell after inactivation is randomly determined and is faithfully transmitted to its progeny through all the following somatic divisions. Every female is therefore mosaic for cells with inactivated paternal X chromosome (Xp) and cells with inactivated maternal X chromosome (Xm); with approximately half of the cells having the inactive Xp and the remainder of the cells with the inactive Xm (Gale, 1999). The single cell origin of malignant neoplasms predicts that all malignant cells should have the same XCIP. In contrast, normal tissues in a female derived from a variety of normal precursor cells should be polyclonal and heterogeneous for their XCIP. Methods to 9 evaluate XCIP have two basic requirements: first, the ability to distinguish between the maternal and paternal X chromosomes, which can be achieved by the identification of polymorphic alleles of genes on the X chromosome, and second, a means to differentiate the alleles on the inactive from those on the active X chromosome (Busque & Gilliland, 1998). The pioneering work of Beutler and Fialkow (Beutler, 1962; Fialkow, 1972) using analysis of expression of isoenzymes of glucose-6-phosphate dehydrogenase (G-6-PD), laid the foundation for utilizing XCIP to determine the clonal origin of human tumors. However, electrophoretically distinguishable G-6-PD isozymes are heterozygous in only 30-35% of African American females. Alternatively, investigators use DNA and RNA polymorphisms as listed in Table 1 to analyze XCIP. Because 90% of women are heterozygous for informative polymorphisms in one or more of these genes, these assays are now applicable to most females. Table 1 Polymorphic loci commonly used in analysis of XCIP, adapted from (Gale, 1999) and (Busque and Gilliland, 1993) Target Gene/Locus Polymorphism Techniques % sample heterozygo sity Protein glucose-6-phosphate Amino acid substitution Protein gel 30-35 dehydrogenase electrophoresis (African (G-6-PD) American) DNA Phosphoglycerate kinase RFLP Southern blotting, 30-40 (PGK) PCR Hypoxanthine RFLP Southern blotting «17 phosphoribosyl transferase (HPRT) DXS255 (M27p) VNTR Southern blotting «78 Human androgen receptor VNTR PCR 70-90 allele (HUMARA) RNA G-6-PD RFLP RT-PCR «27 HUMARA VNTR RT-PCR «90 Abbreviations: RFLP, restriction fragment length polymorphism; VNTR, variable number tandem repeat; RT-PCR, reverse transcriptase PCR; 10 a. DNA analysis It is well known that loss of activity of many genes, including those on the X chromosome, are accompanied by methylation of cytosine residues in the promoter region (Heard et al., 1997). The variability of this methylation pattern thus allows the active and the inactive X chromosomes to be distinguished. The differential methylation can be identified by digestion of DNA with methylation-sensitive restriction enzymes such as Hpa II (which cuts at C N I ' C G G ) and Hha I (which cuts at GCG^C). In general, cytosine residues are methylated on the inactive allele and unmethylated on the active allele (Heard, et al, 1997). However, the methylation pattern on each allele can be more complex and heterogeneous. It is especially important for PCR analysis that the polymorphic site(s) and the differentially methylated site(s) are close to each other on the DNA sequence. The methylation sites should be within the range of PCR amplification and their methylation pattern should strictly correlate with the XCIP (Busque & Gilliland, 1998). Based on these standards, suitable polymorphic genes for PCR analysis include phosphoglycerate kinase (PGK), hypoxanthine phosphoribosyl transferase (HPRT) and the human androgen receptor allele (HUMARA). In 1987, Vogelstein first described DNA clonality analysis using restriction fragment length polymorphisms (RFLP) and methylation patterns of HPRT and PGK genes on the X chromosome (Vogelstein et al., 1987). In PCR-based clonality assay, two aliquots of DNA are amplified for each sample. One aliquot is left undigested as the control to demonstrate the presence of different sized PCR products each generated by one allele. In the other aliquot, the DNA is digested with either Hpa II or Hha I prior to PCR amplification. The enzyme will cut all the unmethylated alleles so that PCR performed after digestion using a 5' primer upstream of the restriction sites only amplifies a fragment from the uncut, inactive alleles. In case of PGK and HPRT, the amplified PCR products need to be further digested with restriction enzymes to allow size separation of the two alleles. The PCR products of HUMARA, however, contain 11-31 11 (CAG) repeats, which can be separated without further processing (Allen et al., 1992). The quantitation of the relative expression of the two alleles in PCR-base clonality assays can be achieved using a labeled primer and densitometric analysis (Allen et al., 1992; Delabesse et al., 1995). An illustration of the PCR-based clonality assay using HUMARA is shown in Figure 2. Problems with DNA analyses can be caused by incomplete DNA digestion with the restriction enzymes and complex methylation patterns sometimes seen in malignant samples and after in vitro culture. For example, DNA methylation has been found to decrease in cultured normal mouse, hamster and human fibroblasts (Wilson and Jones, 1983). In addition, hypermethylation of both alleles of the M27P gene has also been reported in leukemic samples (Gale et al., 1996). Recently, Uchida et al (Uchida et al., 2000) have developed a new method to avoid the inaccurate results sometimes generated by incomplete enzyme digestion. In this new method, unmethylated cytosines are modified by sodium bisulfate and changed into uracil, whereas methylated cytosines remain unchanged, producing sequence differences between the active and the inactive AR alleles. Using primers specific for the modified methylated and unmethylated alleles, the methylation status of alleles on Xp and Xm can be distinguished. 12 Hpa II or Hha I, 37°C, O/N • M M 1 PCR; Genescan clonal polyclonal clonal Figure 2 HUMARA clonality assay M : maternal androgen receptor allele; P: paternal androgen receptor allele; H : inactive; • : active. 13 b. R N A analysis Recently, many researchers have tried to use RNA to identify XCIP instead of DNA to circumvent the problems caused by complex DNA methylation patterns. Genes suitable for this analysis must contain a polymorphism in their coding sequences, and they must be expressed in cells of interest. G-6-PD was re-explored for RNA clonality analysis and the heterozygosity rate increased to 27% in the general female population (Beutler and Kuhl, 1990). The AR gene, however, is only suitable for the mRNA assay in some tissues (Busque et al., 1994). In addition to the advantage of not being influenced by the complex methylation patterns that can complicate DNA analyses, transcription assays can also identify clonality of cells that lack nuclei such as reticulocytes and platelets. However, reverse transcriptase-PCR (RT-PCR), which is used in the mRNA clonality assay, is a sensitive technique. The requirement for cell purity is thus high, and this can be especially demanding for hematological samples. For instance, red blood cell contamination will not affect a DNA analysis for neutrophil preparations but will affect an RNA-based analysis. 1.1.3.2 Interpretation of XCIP results Although X-chromosome linked clonality analysis has the advantage of not needing prior knowledge of a specific cytogenetic abnormality, the results must be interpreted carefully. Since X-chromosome inactivation is random and the inactivation occurring in one cell is not affected by inactivation in other cells, a normal polyclonal cell population in a female should contain 50% cells with Xp in the active state and 50% of cells with Xm in the active state. A shift in this ratio may thus indicate the presence of a clonal cell population among normal cells. However, studies have shown that there is a considerable variation in XCIPs of hematopoietic cells among hematologically normal females. About 20-25% of blood or bone marrow samples from normal females show a skewed XCIP (more than 75% of cells show inactivation of the maternal or 14 paternal X chromosome) (Fey et al., 1994; Gale et al., 1994). This phenomenon can be explained by two potential factors. One is the primary constitutive skewing that may occur in early embryogenesis. This is thought to be related to the limited (as few as 8 to 16) number of cells committed to hematopoiesis and present at the time of random X-chromosome inactivation (Busque and Gilliland, 1993). According to this hypothesis, the smaller the number of the cells present at the time of inactivation, the greater the possibility of deviation of XCIP from the 1:1 ratio. In addition, a constitutively skewed XCIP may also be genetically determined. Primary gene inactivation in mice is determined by different alleles of a gene called X-chromosome controlling elements (Xce) located within the X-inactivation center (Cattanach, 1975). Although there is still no evidence of human homologues of the Xce gene, Pegoraro et al (Pegoraro et al., 1997) reported a family with a high incidence of imbalanced XCIP and found that the affected individuals had a large segment of DNA deleted in the Xq28 region. Plenge et al (Plenge et al., 1997) provided further evidence for a genetic cause of skewed XCIP by finding a rare point mutation in the X-inactivation specific transcript minimal promoter in two different families that both showed preferential inactivation of the X chromosome carrying the mutation. However, more commonly it appears that skewing of XCIP can increase with age (Busque et al., 1996; Fey et al., 1994; Gale et al., 1996). In one study, the frequency of skewing in myeloid cells of normal older females (>75 years of age) was 56% whereas in women younger than 50, it was only 22% (Gale et al., 1996). The skewing may not necessarily reflect a preleukemic state but rather a stochastic change in stem-cell usage. For example, the stem cell pool may only be maintained by a small number of stem cells in elderly people. The rest of the stem cell clones may be lost due to terminal differentiation (Gale, 1999). Both the constitutional and age-related skewings have to be taken into consideration when evaluating XCIPs. A second clonality assay is often needed as a reference. Cells from tissues not affected by the disease are required as controls to demonstrate the constitutional XCIP for 15 comparison with the cell population being tested. Although PCR based XCIP assays can detect of as few as 100 clonal cells, the sensitivity of X-chromosome linked clonality assay still does not allow the detection of minimal residual disease in a large normal cell background. For example, a clone occupying 50% of a normal tissue would be required to shift the XCIP from 1:1 to 1:3. 1.2 Deregulated gene expression in AML Over 50% of A M L patients have nonrandom, leukemia specific chromosome abnormalities that are detectable by various cytogenetic assays (Look, 1997). Many genetic changes caused by these chromosome rearrangements as well as submicroscopic abnormalities such as point mutations have been found to be involved in the deregulated growth of A M L leukemic blasts. These mutations frequently target genes that are involved in the regulation of gene transcription and signal transduction as well as in cell proliferation and differentiation. In the following pages, I will first discuss the cytogenetic abnormalities in AML and then focus on specific genes implicated in leukemogenesis including transcription factors, autocrine production of growth factors or cytokines, growth factor receptors and genes associated with cell death. 1.2.1 Cytogenetic abnormalities in AML patients Karyotypic abnormalities in malignant cells from AML patients correlate with FAB subtypes and provide the most powerful prognostic indicator of therapeutic responses and overall survival. Table 2 shows the prognostic classification of cytogenetic abnormalities based on two clinical trials carried out by the Medical Research Council of the United Kingdom on different age groups (Grimwade et al., 2001; Grimwade et al, 1998). Patients with translocations t(8;21), t(15;17) and inversion (16) have a relatively favorable prognosis, characterized by low drug resistance and reduced risk for relapse and superior overall survival. In contrast, the presence of complex cytogenetic changes, monosomy of chromosome 5 (-5) or deletion (del) of 5q, 16 chromosome 3q abnormalities and -7 was found to be associated with poorer overall survival. Other patients who do not fall into either group were classified in the intermediate prognostic group (Grimwade et al., 1998). Favorable prognostic cytogenetics were less frequent in patients over 55 years of age, whereas the presence of poor prognostic cytogenetic abnormalities was more common in these older patients (Grimwade et al., 2001). 17 Table 2 Prognostic classification of cytogenetic abnormalities, their association with FAB subtypes and frequencies in AML patients FAB subtypes Frequency (%) (younger/older)* Abnormalities Prognosis Favorable t(8;21) M2 8/2 t(15;17) M3 12/4 inv(16) M4eo 4/1 Intermediate Abnl lq23, e.g. t(9;l I)(p22;q23) M4orM5 4/1 normal 42/48 +8 9/10 +21 3/2 +22 1/1 del (7q) 2/4 del (9q) 2/2 Adverse Complex abnormalities 6/13 -7 4/8 -5/del (5q) 3/12 Abn (3q) 3/3 Abbreviations: t: translocation; inv: inversion; abn: abnormal; del: deletion : younger: children and adults up to 55 years of age; older: adults more than 55 years old 18 Due to the close association between cytogenetics and clinical outcomes, the traditional G-banding technique has become an essential part of the routine diagnostic and prognostic evaluation for A M L patients. However, the interpretation of chromosome banding patterns for cytogenetics requires highly skilled professionals and is labor-intensive. Moreover, only metaphase cells can be analyzed, which is unsuitable for analyzing non-dividing cells. In the 1980s, fluorescent in situ hybridization (FISH) was developed to improve the efficiency of chromosome analysis (Pinkel et al., 1986). FISH is based on the principle that a single-stranded DNA probe labeled with either biotin or digoxigenin or directly conjugated to a fluorochrome can anneal to complementary nuclear DNA of the target cells affixed on a microscopic slide. After annealing, the probe is detected by attachment of fluorochrome-labeled antibodies against biotin (streptavidin) or digoxigenin (anti-digoxigenin) and visualized by fluorescence microscopy. DNA FISH probes derived from cloning breakpoints of different translocations and inversions have facilitated the identification of genes that play a critical role in the transformation of normal cells to leukemic cells. Since FISH can detect chromosome rearrangements in both interphase and metaphase cells, large numbers of cells and samples can be analyzed. The use of FISH as a complement to traditional G-banding has provided a powerful diagnostic tool for known chromosome aberrations in hematopoietic disorders (Bernard and Berger, 2000). More recently, other molecular genetic methods including spectral karyotyping, comparative genome hybridization, PCR and microarrays have also been developed to both improve the sensitivity of cytogenetic assays and pinpoint genes involved in chromosome changes (Bernard and Berger, 2000; Lichter et al., 2000; Schrock and Padilla-Nash, 2000). However, chromosome changes may not necessarily be a reliable marker of the entire leukemic clone. According to the model of AML leukemogenesis proposed by Jacobson et al (Jacobson et al., 1984), the development of leukemia goes through two major steps. At first, the transformation occurs in a single hematopoietic progenitor cell. The chromosome structures are 19 not significantly changed by the transformation. Instead, point mutations, which are undetectable as obvious chromosomal abnormalities, may lead to the changes of expression of certain genes critical for proliferation, differentiation and/or apoptosis. The clone derived from the transformed progenitor can therefore gain a growth advantage over other clones. Only after additional transformation events, visible chromosomal changes may occur. Therefore, cytogenetic changes may be secondary to the initial transforming events associated with neoplastic proliferation. Supporting this model, Fialkow and other researchers (Jacobson et al., 1984; Fearon et al., 1986; Fialkow et al., 1987) found that a clonal population regenerated the bone marrow of a few AML patients who entered clinical remission. Interestingly, all clonal remission patients the studied had the same X-chromosome inactivation phenotype as the original leukemic cells. Clonal remission was taken by these researchers as evidence for re-emergence a "preleukemic" clone after chemotherapy, consistent with the multi-step leukemogenesis model (Fialkow et al., 1987). 1.2.2 Oncogenic transcription factors in AML Somatically acquired translocations and inversions occur in up to 45% of patients with AML (Look, 1997). The most frequent targets of these structural rearrangements are transcription factors. Translocations often fuse two genes that are otherwise physically separate from each other. Although few downstream targets have been identified for these chimeric transcription factors, much has been learned about the role of these fusion proteins in human leukemia. The transcription factor complex of core binding factor a (CBFa or AML1) and P (CBFP) is the most common target of chromosome translocations in acute leukemia. AML1 is disrupted by the A M L 1-Eight-twenty-one (ETO) fusion in translocation (8;21) as well as by fusions with other genes in chromosome translocations. The CBFP-smooth muscle myosin heavy chain (SMMHC or MYH11) fusion in inversion 16 (q22;pl3) occurs in 12% of AMLs (Look, 1997). Mice with engineered homologous deletions of either the AML1 or CBFP gene lack fetal liver hematopoiesis and myeloid and erythroid progenitors derived from definitive hematopoiesis. 20 These animals die early in embryogenesis (Okuda et al., 1996; Sasaki et al., 1996; Wang et al., 1996b; Wang et al., 1996c), indicating the crucial role of this transcription factor complex in hematopoiesis. Mice heterozygous for AML1-ETO or CBF|3-MYH11 fusion genes demonstrate a similar deficiency in fetal liver hematopoiesis and a phenotype nearly identical to what is seen in AML1 and CBFp knock-out mice (Castilla et al., 1996; Yergeau et al., 1997). Thus the fusion proteins appear to act as dominant negative inhibitors of normal CBF function. Translocations and deletions involving chromosome band 1 lq23 disrupt the myeloid lymphoid leukemia (MLL) gene by fusing the N-terminal region of the M L L gene with various partner genes. These are among the most frequent structural chromosome rearrangements in AML. AF-9 on 9p21 is the most common partner of MLL fusions (Dimartino & Cleary, 1999). An AF-9/MLL fusion protein analogous to that produced by translocation between chromosomes 9p21 and 1 lq23 in human AML was created in mice by using homologous recombination to insert the AF-9 gene into one allele of the MLL locus. Mice expressing the fusion protein developed A M L which resembled the disease seen in human A M L patients with translocation (9;11) (Corral et al., 1996). Another example suggesting that the relationship between structural chromosome changes and the pathogenesis of A M L is chromosome translocations involving the retinoic acid receptor a (RARa) gene on chromosome 17q21. Although over 90% of cases of acute promyelocyte leukemia (APL) or M3 subtype of A M L are characterized by a translocation between chromosome 15q22 at the site of the promyelocytic leukemia (PML) gene and the RARa locus (Grignani et al., 1994), variants in which other partner genes are rearranged and fused to the RARa gene have also been described (reviewed in (Alcalay et al., 2001)). PML-RARa retains many functions of wild type RARa but it interferes with the RAR-RXR pathway by acting as a dominant negative inhibitor (Piazza et al., 2001). PML has features of a tumor suppressor, 21 therefore disruption of the normal PML pathway by the fusion protein may also contribute to leukemogenesis (Wang et al., 1998b). Mice transgenic for the PML/RARct fusion protein display leukemia with features common to human APL and an accumulation of promyelocyte precursors in the bone marrow (Brown et al., 1997). The functional properties of other RARoc fusion proteins are largely unknown. Transgenic mice for these fusion genes develop various types of leukemia of which the majority are associated with disordered myelopoiesis (Alcalay et al, 2001). The identification and cloning of the aforementioned and other translocation-induced fusion molecules and the creation of in vitro and in vivo models to study the functions of these chimeric genes have provided many important clues to understand the initiation and progression of AML. 1.2.3 Autocrine production of growth factors in AML. Using blast cell CFC assays, Reilly et al (Reilly et al., 1989) observed that blasts from over 70% of A M L patients exhibited different levels of colony growth when cultured in the absence of added growth factors or cytokines that have been shown to support the growth of normal as well as A M L CFC. In a study carried out with a series of 50 patients, Hunter et al found that patients with blasts capable of growing autonomously in semi-solid media had a much lower (56%) complete remission rate compared to patients whose leukemic cells needed growth factors to grow in methylcellulose (94%) (Hunter et al., 1993). Autonomous growth is often related to the autocrine secretion of growth factors by leukemic blasts. Young et al (Young and Griffin, 1986) found that granulocyte macrophage colony stimulating factor (GM-CSF) mRNA could be detected in half of the AML blast samples, while in normal hematopoietic cells, GM-CSF was only expressed in activated T lymphocytes. The same group later identified the expression of granulocyte colony stimulating factor (G-CSF) and macrophage colony stimulating factor (M-CSF) in a significant proportion of AML patients (Young et al., 1988). The simultaneous expression of all three growth factors was also found in many cases. Many other researchers 22 have confirmed these findings and observed autocrine production of other cytokines including interleukin l(IL-l), interleukin 6 (IL-6) and tumor necrosis factor-a (TNF-a) (reviewed in Russell, 1992). More recently, expression of Steel factor (SF) and the ligand for fms like tyrosine kinase receptor-3 (Flt-3L or FL) has been detected in primary A M L leukemic blasts (Cole et al., 1996; Zheng, 2001). Although AML progenitors from many patients respond to IL-3 (Miyauchi et al., 1988; Vellenga et al., 1987; Hoang et al., 1988; Piacibello et al., 1995; Siitonen et al., 1996), none of the 28 samples tested by Oster had detectable levels of mRNA for the interleukin 3 (IL-3) gene (Oster et al., 1989). However, a more recent study demonstrated IL-3 mRNA expression in 3/15 A M L patient samples tested (Ailles et al., 1999a). Interestingly, expression of functional receptors for GM-CSF, G-CSF, IL-3, SF and FL have been demonstrated in either immortalized leukemic cell lines or primary AML cells (Broudy et al., 1992; Budel et al., 1989a; Budel et al., 1989b; Park et al., 1989; Carow et al., 1996). In particular, in a recent study by Zheng et al, Flt-3 was found to be expressed concomitantly with its ligand FL in all the 71 AML patients tested (Zheng, 2001). The fact that autonomous growth of leukemic blasts could be inhibited by neutralizing antibodies against GM-CSF and IL-ip (Reilly et al, 1989; Young et al, 1986) and the extensive expression of growth factor receptors on A M L blasts suggests that extracellular autocrine and paracrine loops can cause autonomous proliferation of A M L cells. Majka et al (Majka et al., 2001) have recently found that normal CD34+ cells from B M or PB, which represent a relatively primitive population in normal individuals, can also secrete numerous growth factors including SF and FL. These endogenous growth factors were found to protect both normal CD34+ and leukemic blasts from apoptosis (Russell et al., 1995). This suggests that autocrine production of some growth factors may be a common feature of undifferentiated hematopoietic cells rather than a unique characteristic of leukemic blasts. 23 1.2.4 Mutations in growth factor receptors Receptors for FL (Flt-3) and SF (c-kit) belong to the Class III receptor tyrosine kinase family. Their structure includes an extracellular ligand binding region, transmembrane (TM) and juxtamembrane (JM) domains, a split tyrosine kinase domain (TK1 and TK2) and a C-terminal domain (Figure 3)(Lyman, 1998). Growth factor binding to the extracellular domain of these receptors leads to receptor dimerization and phosphorylation of receptor downstream signaling molecules which are responsible for cell proliferation and activation (Gilliland & Griffin, 2002). Amplification or somatic mutation of these receptors may increase their downstream signaling and cause cancers. Flt-3 is the most commonly mutated gene in AML. About 30% of patients have mutations on Flt-3 (Gilliland & Griffin, 2002). One of the mutations discovered recently in AML is an internal tandem duplication (ITD) of part of the JM region (exons 11 and 12) of the Flt-3 receptor (Nakao et al., 1996). This duplication is always in-frame and exists in a head-to-tail fashion. The length of the duplication is between 20 and 200 base pairs. The Flt-3 ITD occurs in about 22% of A M L cases (Kiyoi et al., 1999; Rombouts et al , 2000) and more often in older patients (Stirewalt et al., 2001). It is associated with increased peripheral blood leukemic blast cell counts, lower complete remission rates and reduced leukemia-free survival especially in pediatric AML patients (Kiyoi et al., 1997; Rombouts et al., 2000) (Meshinchi et al., 2001). In one study, Flt-3 ITD positive AML patient samples were found to engraft to higher levels in NOD/SCID mice as compared to ITD negative samples (Rombouts et al., 2000). This finding strengthens the relationship between Flt-3 ITD mutation and poor clinical outcome since AML samples with cytogenetic abnormalities associated with poor prognosis also grow better in NOD/SCID mice than those with good prognostic chromosomal changes (Ailles et al., 1999a; Bonnet and Dick, 1997). Mutations of Flt-3 gene also happen in the TK domains. In particular, the substitution mutation at aspartic acid residue 835 (D835) was found in 7% of A M L patients 24 (Yamamoto et al., 2001) and are associated with a reduced event free survival that is similar to that seen with the Flt-3 ITD (Yamamoto et al., 2001). In vitro studies have shown that both ITD and D835 mutations result in constitutive activation of the Flt-3 receptor (Kiyoi et al., 1998; Yamamoto et al., 2001). ITD mutations can initiate ligand-independent dimerization and tyrosine phosphorylation and activation of signal transducers and activators of transcription (Stats), Ras/ mitogen-activated protein kinase and phosphatidylinositol 3' kinase pathways (Gilliland & Griffin, 2002). The overexpression of these signaling molecules has also been observed in A M L patients (Fair et al., 1988; Radich et al., 1990; Bos et al., 1987; Hayakawa et al., 1998; Towatari et al., 1997; 1998; Xia et al., 1998). In addition, the abnormal expression of such signaling proteins is associated with the Flt-3 ITD mutation. For example, in one study, 47% of the cases with activated Stat 5 were also Flt-3 ITD positive (Birkenkamp et al., 2001). Moreover, Flt-3 ITD transfected 32D or BaF3 cell lines showed constitutive phosphorylation of Stat5 (Hayakawa et al., 2000; Mizuki et al., 2000). Although Flt-3 ITD has been primarily associated with human AML, the expression of this mutant receptor alone is not sufficient to induce a similar leukemia in murine system. The expression of Flt-3 ITD in primary mouse bone marrow cells only induced a myeloproliferative disorder characterized by high white blood cell counts in mice (Kelly et al., 2002a). Secondary mutations are required to cause AML. Flt-3 ITD have been reported in patients with chromosome translocations such as t(8;21), inv (16), t(15;17) and llq23 abnormalities (Gilliland & Griffin, 2002), suggesting the fusion proteins as results from these translocations may cooperate with the mutant Flt-3 receptor to induce AML. The results from a recent study have strongly supported this hypothesis. In this study, transplantation of Flt-3 ITD transduced bone marrow cells from PML/RARa transgenic mice into lethally irradiated mice can induce APL like disease and the disease latency of these transplanted mice is shorter than that of PML-RARa transgenic mice (Kelly et al, 2002 b). 25 Deletions, point mutations and insertions in the SF receptor (c-kit) gene have also been found in less than 10% of AML patients, and leukemic blasts from these patients often have a mast cell morphology (Gari et al., 1999; Sperr et al., 1998). D835 Figure 3 The structure of Flt-3 receptor and mutations found in AML patients (Lyman, 1998; Gilliland & Griffin, 2002) The structure of Flt-3 consists of an extracellular ligand binding domain, a transmembrane (TM) domain, a juxtamembrane domain (JM), two tyrosine kinase domains (TK1 and TK2) and a cytoplasmic domain (C).The ITD mutation happens in the JM domain, while D835 mutation occurs in TK2 domain. 26 1.2.5 Abnormalities of genes involved in apoptosis Cellular susceptibility to apoptosis is regulated by a number of proteins including the p53 tumor suppressor gene, Bcl-2 and Bax. The mutations and abnormal expression of these genes have been found in many AML patients. Mutations of the p53 gene are rarely seen in A M L patients. However, the nonmutated p53 from some leukemia samples has a conformation similar to the mutated protein and this is associated with poor prognosis in AML patients (Zheng et al., 1999; Zhu et al., 1993). Bcl-2 is a mitochondrial protein that suppresses and delays the onset of programmed cell death following growth factor deprivation, while Bax is a pro-apoptotic protein that has been shown to be essential for the initiation of cell death after ultraviolet-induced damage or DNA double-strand breaks (Wei et al., 2001). Bcl-2 expression was found in leukemic blasts from AML patients (Bradbury et al., 1997) with higher expression in CD34+ than in CD34" leukemic samples. High expression of Bcl-2 was associated with both a low complete remission rate after chemotherapy and a shorter overall survival as compared to Bcl-2 negative cases (Campos et al., 1993). Moreover, upregulation of Bcl-2 and/or downregulation of Bax have been shown to be involved in the anti-apoptotic effect of GM-CSF and FL on A M L cells (Klampfer et al., 1999; Lisovsky et al., 1996). Although mice overexpressing Bcl-2 only developed a disease analogous to human chronic myelomonocytic leukemia, when they were crossed with Fas (which is a receptor in a pathway inducing apoptosis) deficient mice, 15% of the mice with both mutations reproducibly developed a disease similar to a subtype of human A M L (AML-M2) (Traver et al., 1998). The transcription factor nuclear factor-KB ( N F - K B ) may also play a role in reduced cell death of leukemic blasts. Activation of N F - K B has an antiapoptotic effect (Wang et al., 1996a; Wang et al., 1998a) while inhibition of N F - K B induces apoptosis (Kim et al , 2000) in several types of cancer. Recently, in a study conducted by Guzman et al, N F - K B was shown to be constitutively 27 activated in CD34+CD38" AML cells but not in CD34+ cells from normal B M (Guzman et al., 2001). Thus, various mechanisms that can delay or inhibit apoptosis may play a role in the development of AML. 1.3 Cell cycle deregulation in AML 1.3.1 The mitotic cell cycle There are four distinct phases in the cell cycle that ultimately lead to cell division or mitosis and the production of genetically identical daughter cells: Gi, a growth phase with high metabolic activity; S, the DNA synthesis phase; G2, a short gap phase; and M (mitosis), a period when a cell with two sets of chromosomes divided between two daughter cells. In early Gi, in the absence of growth factors or serum, cells may enter a resting phase called Go. Go cells can be stimulated to enter Gi by the addition of growth factors or serum (Figure 4). Several checkpoints in the cell cycle ensure the genetic integrity of daughter cells and allow regulation of the cell cycle by extracellular signals. An important checkpoint between early and late Gi ensures the genetic fidelity of cells and prevents damaged cells from entering S phase. Once cells pass the restriction point, they are irreversibly committed to the next division cycle, independent of stimulation by growth factors. Beyond this point, cells are also refractory to growth inhibitors. This is why most of the cell cycle studies on malignant cells have focused on the G| restriction point and the molecules affecting Gi progression. Another important replication checkpoint is at the boundary between S and G2 phase. 1.3.2 Cell cycle analyses 1.3.2.1 Detection of DNA replication S phase occupies about 50% of the cell cycle in actively proliferating cells. The proportion of cells in this phase can thus be used as an indicator of the number of cycling cells in a population. High specific activity tritiated thymidine ( H-Tdr) is often used to detect cells in S phase. The 28 radio actively labelled thymidine is incorporated into DNA strands during DNA synthesis. The decay of 3 H causes multiple aberrations in chromosomes including double or single strand breaks, base destruction, chromatid gaps, breakage of glycoside-phosphate ester linkages, chromatid interchanges and cross-linking between strands of DNA. When cells attempt to repair this damage and join the broken ends, many other chromosome aberrations may be formed, such as translocations, inversions, rings, etc.(Hori and Nakai, 1978). This will ultimately lead to cell death. A 20 minute exposure to high-specificity 3H-Tdr is sufficient to kill all S phase cells, which makes this technique an efficient way to detect actively cycling cells. Bromodeoxyuridine (BrdUrd), a thymidine analog, can also be used as an S phase indicator. Other cytotoxic drugs such as cytosine arabinoside (Ara-C) (Lacombe et al., 1992) measure cycling cells similarly by killing S phase cells through various mechanisms. 1.3.2.2 Quantitation of cellular DNA and RNA One of the disadvantages of most S phase specific methods for cell cycle analysis is that cycling cells are killed, which makes it impossible for their functional properties to be studied. Alternatively, cells can be stained with non-toxic fluorescent DNA and RNA dyes and then separated into subpopulations in different phases of the cell cycle according to their differing DNA/RNA content by flow cytometry. The most popular DNA/RNA dye combination is Hoechst 33342 (Hst)/Pyronin Y (PY). Hst is known to facilitate relatively reliable flow cytometric measurement of the DNA content of living cells. PY is an efficient fluorescent as well as an absorption stain for RNA (Shapiro, 1981). PY binds preferentially to G-C rather than to A-T pairs whereas Hst only binds to A-T triplets and attaches in the outer groove of the DNA helix. Thus, there should be little interference between these two dyes (Muller and Gautier, 1975). Hst/PY staining can also be used in conjunction with fluoresceinated antibodies for detection of cell surface antigens in living and fixed cells (Shapiro, 1981). Quiescent (Go) cells, under normal conditions, have a diploid DNA content detected by Hst staining and a low RNA content leading 29 to minimal staining with PY. When cells are stimulated to enter the Gi, DNA content is still diploid but gene transcription, RNA synthesis and PY staining increase. Cells accumulate increased DNA content throughout S phase and have tetraploid DNA as well as increased RNA content in G 2 and M . Other DNA and RNA dyes that have been used in combination to distinguish different phases of the cell cycle include propidium iodide, 7-aminoactinomycin-D (7-AAD), acridine orange, etc. (Darzynkiewicz, 1988). 1.3.2.3 Studying cell cycle with markers of proliferation and cell cycle progression Detection of mRNA and protein for genes that are differentially expressed during specific phases of the cell cycle can also be used to determine the cycling status of cells. Proliferating cell nuclear antigen (PCNA) is a nuclear DNA binding protein that specifically binds to DNA initiation and repair sites during DNA synthesis (Lacombe and Belloc, 1996). PCNA is expressed throughout the cell cycle. However, after unbound PCNA is eluted anti-PCNA antibodies can be used to detect S phase cells. Another widely used proliferation marker is K i -67. It is also a nuclear DNA binding protein for which expression increases during exit from Go to Gi. Analysis of the expression of genes involved in cell cycle regulation has also been used in cell cycle analysis, for example, the cyclin D family including D1-D3, cyclin E, cyclin A, cyclin B, Rb protein and p53 (Keyomarsi and Pardee, 1993; Lacombe and Belloc, 1996). 1.3.3 Cell-cycle regulatory elements involved in A M L Cell cycle progression is highly regulated by a series of proteins and enzymes. The majority of cell cycle studies have used fibroblasts as target cells. Only recently have efforts been made to understand the cell cycle machinery operating in hematopoietic cells. Figure 4 shows simplified diagram of cell cycle regulation. Two major stimulatory components of cell cycle progression are cyclin dependent kinases (CDKs) and cyclins. The complexes comprised of cyclin D/CDK4/6, cyclin E/CDK2 and cyclin A/CDK2 are assembled and activated consecutively to drive cell cycle progression. Cyclin D/CDK4/6 and cyclin E/CDK2 complexes phosphorylate the 30 members of retinoblastoma protein (pRb, RBI) family. The complex that pRb forms with E2F halts cell cycle progression at the Gi checkpoint. However, after it is phosphorylated, pRb can no longer bind to E2F. Uncomplexed E2F transactivates multiple genes involved in the synthesis of DNA (Dictor, Ehinger et al. 1999). The stimulatory components of the cell cycle machinery are also regulated by inhibitory enzymes and proteins, including cyclin-dependent kinase inhibitors (CDKIs) such as p21, p27 and p57, members of the p21 family. p21 family inhibits the activity of cyclin E/CDK2 and cyclin A/CDK2. Other inhibitors that affect Gi cell progression include the above mentioned pRb protein and the upstream regulator of the p21 family, p53 (Figure 4) (Dao and Nolta, 1999; Dictor et al., 1999). Growth factor signaling positively regulates the transition from Go to early Gi through the cyclin D/CDK4/6 complex. The strength of this signal determines whether early Gi phase cells transit from growth factor dependence to growth factor independence in late Gi or exit the cell cycle and enter Go. Several negative cell cycle regulators also influence this process, pi6, a member of the INK4 family, in physiological settings, sequesters the CDK4/6 complex, keeping cells in a quiescent state. In addition, pi 30, another member of the pRb family, by forming a complex with E2F directly or indirectly represses the transcription of immediate early genes (e.g. nuclear receptor N10 protein, members of Jun family, members of the Fos family, c-myc etc.) that are involved in progression from Go into active cell cycle. (Andreeff, 1986; Ho and Dowdy, 2002) (Bravo, 1990). 31 Growth factors" Genotoxic insults -Growth factors-INK4 family (pl6,pl5,pl8orpl9) 1 ?' + Cyclin D/—— • | CDK4/6 5 3 ^ p21 family H (P21,p27,p57) CyclinE/ CDK2 It ^ p21/ PCNA Growth factor dependent (reversible) Growth factor independent (irreversible) Figure 4 Cell cycle regulation (Dao and Nolta, 1999; Dictor et al., 1999; Ho and Dowdy, 2002) 32 Many of these cell cycle regulatory genes are mapped to chromosome regions that are involved in the re-occurring translocations and deletions in acute myeloid leukemia (Dictor et al, 1999). For example, members of the pRb family, cyclin A, CDK2 and 6, E2F1 though 5, and members of p21 (p21 and p27) and LNK4 (pi5, pl6 and pi9) families have all been found to be involved in either homozygous deletions or balanced chromosome rearrangements in AML (Dictor et al, 1999; Sherr, 1996). Some of these mutations have been correlated with patient prognosis. For example, patients with deletions involved with pi6 and/or pi5 genes have significantly shorter complete remissions and lower event free and overall survivals as compared to patients without these abnormalities (Faderl et al., 2000). However, in general, mutations of these cell cycle modulators are unusual in A M L (Baghdassarian and Ffrench, 1996). Ogawa et al only found only 2 cases with homozygous deletions of the pl6 gene in 134 A M L patients (Ogawa et al , 1995). On the other hand, the expression level of these regulators is often different in leukemic than in normal cells. Low expression of both pRb and p53 is found in about 20% and 10% of A M L samples, respectively (Kornblau et al., 1998; Sugimoto et al., 1991). Decreased expression of these two key cell cycle regulators has been correlated with a reduced rate of complete remission and shortened overall survival (Kornblau et al., 1998; Wattel et al., 1994). However, in other studies, the expression of p53 was found to be normal or occasionally elevated in most A M L patients (Zheng et al., 1999) (Konikova and Kusenda, 2001). Elevated expression of cyclin E was also observed in 27% AML samples and was accompanied by the enhanced expression of p27 (Iida et al , 1997). Although patients with high p27 expression had significantly increased disease free survival, no correlation has been found so far between cyclin E expression and prognosis (Yokozawa et al., 2000). So far, there are few studies of the expression of other cell cycle regulatory proteins in AML (Radosevic et al., 2001). 33 1.3.4. Cycling status and responses to growth factors of normal and leukemic hematopoietic progenitors 1.3.4.1 Cell cycle status and responses to growth factors of normal hematopoietic progenitors Under steady state conditions, the need to maintain a stem cell reservoir should dictate that HSCs remain quiescent. Many lines of evidence have suggested that normal primitive stem cells from bone marrow are in fact largely found in Go/Gi. For example, primitive cells from murine and human bone marrow have shown resistance to cytotoxic drugs such as 5-fluorouracil (5-FU) and 4-hydroxycyclophophamide, respectively, which selectively kill cells in S phase (Berardi et al., 1995; Lerner and Harrison, 1990). Purified stem cell candidates from different tissue origins have also proven to be quiescent. Reems and Torok-Storb reported that 96.1% of human bone marrow CD34 +CD38 l o w were in G 0 /G| while among CD34 +CD38 h , g h cells 12.9% were in S phase. Since it is commonly accepted that high expression of CD34 and lack of expression of CD38 are the features of multipotent HSCs, they concluded that most primitive stem cells were deeply quiescent (Reems and Torok-Storb, 1995). Later, using different cell sorting strategies, other groups reported similar results (Bradford et al., 1997; Jordan et al., 1996; Leemhuis et al., 1996). Most of these studies had exposed cells to cytotoxic drugs such as Ara C or reagents such as 3H-Tdr for a short period to characterize their cycling status. These techniques can only discriminate cells in S phase from those in the remaining stages of the cell cycle and are relatively insensitive for the detection of small populations of cycling or quiescent cells. When techniques such as Ki-67 staining or RNA staining with acridine orange, which can discriminate Go and Gi, were used, many purified HSC candidates from bone marrow were found to be in Gi but not in Go (Jordan et al., 1996; Lemoli et al., 1997; Ponchio et al., 1995). Even the most primitive HSC population, according to Bradford et al (Bradford et al., 1997), could complete one cell cycle in about 30 days. Thus, as would be expected from the need for ongoing replacement of mature blood cells in steady state hematopoiesis and the need for effective and 34 rapid regeneration of more mature cells in case of a sudden blood cell loss, a significant number of HSCs from bone marrow are actively cycling or available for recruitment into the cell cycle when appropriately stimulated. HSCs from peripheral blood, either in steady state or mobilized by growth factors such as G-CSF or after myelosuppressive chemotherapy, are more deeply arrested in Go than HSCs from bone marrow. Using the overnight 3H-Tdr analysis, Ponchio et al determined that only 19±8% LTC-IC from steady state peripheral blood were actively cycling as compared to 85±5% of LTC-IC from bone marrow (Ponchio et al., 1995). There also seemed to be more primitive progenitors in Gi and fewer in S in mobilized peripheral blood (MPB) than in bone marrow, although the proportion of the Go phase cells were similar from these two sources (Lemoli et al., 1998; Ponchio et al., 1995; Uchida et al., 1997; Roberts and Metcalf, 1995). Very few CFC and LTC-IC from umbilical cord blood were found to be in S phase (Lucotti et al., 2000). Evidence of a large quiescent fraction within the compartment of pluripotent progenitors detectable in standard in vitro colony assays (so called colony forming units-granulocyte-erythroid-monocyte-megakaryocyte or CFU-GEMM) has existed for almost two decades (Cashman et al., 1985; Fauser and Messner, 1979). More than 90% these CFU-GEMM from bone marrow survived short-term exposure to 3H-Tdr while a larger proportion of more committed progenitors such as colony forming units-granulocyte-macrophage (CFU-GM) and burst-forming units-erythroid (BFU-E) from bone marrow were in S phase (20-30%) (Aglietta et al., 1989; Fauser and Messner, 1979). In contrast, in steady state peripheral blood, different types of circulating CFCs were largely quiescent (Ponchio et al., 1995). Although CFCs from mobilized peripheral blood were shown to be more actively cycling than CFCs from steady state blood (Ponchio et al., 1997), they were still more quiescent than their bone marrow counterparts (Cashman et al., 1997b). The different microenvironmental factors to which various tissues are exposed may explain these differences in cell cycle status. 35 The proportion of actively cycling cells decreases with the differentiation of committed progenitor cells to morphologically distinguishable precursors (myeloblasts, promyelocytes and myelocytes) (Aglietta et al, 1989). During the late stages of differentiation, mature cells are irreversibly arrested in Go. Quiescent HSCs and progenitor cells can be recruited into the cell cycle by different cytokines. While some cytokines are primarily stimulatory or inhibitory, most have more complex functions in regulating the proliferation of progenitor cells. For example, IL-4 has been shown to be both stimulatory and inhibitory for hematopoietic proliferation (Rennick et al., 1987). GM-CSF, which stimulates the proliferation of neutrophil progenitors, also regulates the cell cycle of dormant primitive progenitors (Ogawa, 1993). Additionally, the functions of different growth factors likely overlap with each other due to the structural homology of receptors for these cytokines (Ihle et al., 1994). Some growth factors are more critical in hematopoiesis than others. Studies using knock-out mice provided direct evidence that SF, FL and thrombopoietin (Tpo) are limiting factors (Carver-Moore et al., 1996; Mackarehtschian et al., 1995; Miller et al., 1997) in mouse hematopoiesis, whereas GM-CSF, IL-3 and IL-5 might not be so critical (Nishinakamura et al., 1996; Stanley et al., 1994). It has been demonstrated that both the types and concentrations of cytokines can determine whether human bone marrow LTC-IC will undergo self-renewal or differentiation (Metcalf, 1980; Hirayama et al., 1994; Yonemura et al., 1996; Zandstra et al., 1997), although the mechanism involved remains unknown. 1.3.4.2 Cell cycle status and deregulated proliferation of A M L In 1978 Minden et al. (Minden et al., 1978) used the 3H-Tdr suicide technique to demonstrate that leukemic CFCs from AML patients were more actively cycling than normal CFCs in the peripheral blood. However, Terpstra et al (Terpstra et al., 1996) demonstrated that the cell cycle status of AML-cobblestone area forming cells (CAFC), which are believed to represent more primitive leukemic progenitors than AML CFC (Breems et al., 1994), was more heterogeneous. 36 31-82% of 6 week A M L CAFCs were resistant to the cytotoxic drug 5-FU (Eaves et al., 1998; Terpstra et al., 1996). In addition, they found that most cells from 2 A M L patients that were capable of engrafting SCID mouse bone marrow were resistant to killing by 5-FU (Terpstra et al., 1996). Consistent with this study, Jordan et al have also found that the CD34+CD38" leukemic cell population, which is enriched for SCID and NOD/SCID mouse repopulating cells, is largely in Go (Jordan, 1999). Despite their aggressive behavior in vivo, AML progenitors from about 70% of leukemia specimens depend on growth factors and cytokines to proliferate in vitro. The effects of growth factors on A M L blasts are also complex and interactive. For example, TNF alpha has been found to synergize with IL-3 and GM-CSF to stimulate the proliferation of leukemic cells from AML. However, it inhibits the induced growth of leukemic blasts by G-CSF and SF in vitro (Carter et al., 1996). IL-1 induced proliferation of AML cells and stimulated the endogenous secretion of GM-CSF, IL-6 and TNF in most A M L cases tested. IL-4 can suppress the induction of autocrine production of growth factors but it also has an additive effect with IL-1 on stimulating proliferation of A M L cells (Wagteveld et al., 1992). There is a significant heterogeneity among A M L blasts with regard to their responses to various external stimuli, which may be explained by the specific genetic abnormalities present in the leukemic clone (Rowe and Liesveld, 1999). Nevertheless, these various bodies of data from different investigators suggest that primitive leukemic progenitors from AML patients might have a different cycling status than their normal counterparts. Thus, characterization of the cell cycle status of leukemic 'stem cell candidates' such as A M L LTC-IC and NOD/SL-IC is important for our further understanding of the biology of A M L and the factors which influence the initiation and progression of the disease. 1.4 Rationale and Thesis Hypotheses In the long-term culture system originally designed for detecting normal HSCs, many cytogenetically normal colonies have been observed in cultures initiated with bone marrow or 37 peripheral blood cells from patients with A M L (Ailles et al., 1997; Coulombel et al., 1985; Guan and Hogge, 2000). Based on the multistep leukemogenesis model, although karyotypically normal, these colonies could nevertheless originate from a clonal neoplastic progenitor population in which chromosome structures have not yet been affected by the transformation events. However, many previous morphology studies and our own observations support the hypothesis that these karyotypically normal colonies may be derived from residual truly normal hematopoiesis (Ailles et al , 1997; Bernstein et al., 1992). The unusually high concentration of these normal progenitors in the peripheral blood of A M L patients suggests that mobilization of these cells from the bone marrow to the circulation is part of the disease process (Ailles et al., 1997; Bernstein et al., 1992). To test this latter hypothesis and investigate the clonal origin of these seemingly normal progenitors, we used the HUMARA X-chromosome linked clonality assay in combination with FISH to study LTC-ICs from several female AML patients from whom a large number of cytogenetically normal colonies had previously been detected in LTC. Chapter 3 shows the results of this study. The second major focus of this thesis is to investigate the cycling status of primitive AML progenitor cells. As aforementioned in 1.3.3, many cell cycle regulatory genes are mutated or have abnormal expression in patients with AML. Our hypothesis is that these genetic changes will lead to aberrant cell cycle status in AML progenitors and that this is of much more importance to the growth of the leukemic clone than the cycling of leukemic blasts which have very limited proliferative potential. The cell cycle state of A M L LTC-IC and A M L cells capable of repopulating the newly-developed mouse model NOD/SCID have not been systematically studied and compared with that of AML CFCs. The overnight 3H-Tdr suicide assay has been useful in comparing the proliferative status of LTC-IC from steady state blood of normal individuals and from peripheral blood of patients with chronic myeloid leukemia (CML) (Eaves et al., 1998; Ponchio et al , 1995). Hence, in Chapter 4,1 will present the results of studies using 38 the overnight 3H-Tdr suicide assay to assess the cycling status of A M L CFC, AML LTC-IC and NOD/SL-IC in a series of newly-diagnosed A M L patients. FISH was performed to verify the leukemic origin of these progenitors and to allow the cycling state of cytogenetic ally normal and abnormal CFC and LTC-IC to be compared. We found that most NOD/SL-ICs and many A M L LTC-ICs from newly-diagnosed AML patients were quiescent as indicated by their survival after overnight exposure to 3H-Tdr. Since most cytotoxic chemotherapeutic drugs act on cycling cells, these quiescent progenitors may represent a reservoir of chemotherapy drug resistant cells. In order to further study the distributions of each of three AML progenitors in different phases of the cell cycle, we first isolated cells in Go, Gi and S/G2+M using the Hst/PY DNA/RNA staining method. Sorted cells were then plated into three in vitro and in vivo functional assays. Subsequently, we plated quiescent A M L cells into short-term suspension culture in an effort to stimulate them to enter the cell cycle. We found that many Go AML cells spontaneously exit Go and enter active cell cycle within 24 hours of culture initiation even in the absence of exogenous growth factors. This promoted us to study the mechanism of this autonomous proliferation of quiescent leukemic cells. The samples used in these experiments were screened for the presence of the Flt-3 ITD mutation and for autocrine production of growth factors previously identified as being produced by some A M L blasts, including IL-3, GM-CSF, G-CSF, SF and FL. Results from these studies are described in Chapter 5. Based on these data, a model of the cycling of different AML progenitors is proposed to explain how leukemic blasts gain a significant growth advantage over normal hematopoietic cells in vivo. Possible future directions for studies of leukemic cell cycle regulation and signaling are discussed under "Conclusions and Future Prospectives". 39 Chapter 2 Materials and Methods 40 2.1 AML cells Peripheral blood (PB) and bone marrow (BM) samples were obtained from newly diagnosed AML patients after obtaining informed consent and with the approval of the Clinical Screening Committee for Research Involving Human Subjects of the University of British Columbia. Light density mononuclear cells were isolated by Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) density gradient separation and cryopreserved in DMEM with 50% fetal calf serum (FCS) and 10% dimethylsulfoxide (DMSO). All AML samples were evaluated at diagnosis for blast morphology, histochemical staining characteristics and cell surface phenotypes following which they were classified by the clinical diagnostic laboratory into French-American-British subtypes. Standard metaphase cytogenetic analysis was performed on diagnostic B M specimens by the clinical cytogenetics laboratories. 2.2 AML Colony Forming Assay Freshly isolated and thawed, cryopreserved cells were assayed immediately or after overnight culture for A M L CFC. Cells were plated at 0.008 to 2xl0 5 cells/ml in methylcellulose-based medium (0.92% methylcellulose, 30% FCS, 2mmol/l L-glutamine, 10"4 mol/L (3-mercaptoethanol, 1% bovine serum albumin (BSA) in a-medium (StemCell Technologies, Vancouver, Canada) supplemented with the following growth factors: 3 units/ml erythropoietin (StemCell), 20 ng/ml each of IL-3, IL-6 (Terry Fox Lab., Vancouver, Canada), G-CSF and GM-CSF (Novartis, Basel, Switzerland), 50 ng/ml each of SF and FL (Immunex, Seattle, WA). Colonies were scored using an inverted microscope after 10 to 14 days of culture at 37 °C (Ailles et al., 1997). 41 2.3 AML Long-Term Culture Assay 4-5xl06 unsorted or Ix loMxlO 6 sorted A M L cells in 1 ml Myelocult media (StemCell) with 10~6 mol/L solucortef (Sigma- Aldrich Canada Ltd., Oakville, Ontario, Canada) and 100 ng/ml SF were seeded onto a confluent layer of irradiated (80Gy) mouse embryonic fibroblasts (Sl/Sl) which had been engineered to produce human IL-3 (S1/S1-J-IL-3) in a 35 mm tissue culture dish. These fibroblast feeder layers secrete 16 ng/ml human IL-3 into their growth medium as determined by the ability of this medium to stimulate 3H-Tdr incorporation into Mo7e cells (Avanzi et al., 1988; Otsuka et al., 1991). LTCs were incubated at 37°C in 5% C 0 2 with weekly half medium changes. SF was added to the cultures twice weekly to maintain a final concentration of 100 ng/ml. After 5 or 8 weeks, both the adherent and nonadherent cells were harvested (the former by trypsinization), pooled and plated into methylcellulose CFC assay as described above. 2.4 Generation of Bone Marrow Fibroblasts (BMF) from Patient BM Cells To generate BMF, cryopreserved B M cells from A M L patients were thawed and plated at <2xl05 cells/ml in Iscove's medium with 20% FCS in tissue culture dishes. The medium was replaced every 4-5 days until a confluent monolayer of fibroblasts was obtained. Cells were then trypsinized and pelleted by centrifugation prior to DNA extraction. 2.5 Short-term in vitro culture A M L cells were cultured in serum free medium containing 20% serum substitute BIT (containing 5% BSA, 50ug/ml bovine pancreatic insulin and lmg/ml human transferrin, StemCell Technologies Inc., Vancouver, Canada), 2mmol/L L-glutamine and 2xl0"5 mol/L P-mercaptoethanol supplemented with various growth factors combinations. After 1-10 days of 42 culture, cells were harvested with a cell scraper and Iscove's medium containing 5% FCS. Cells were pelleted and plated into various assays. 2.6 Extraction of Genomic DNA from Bulk AML Cells and Methylcellulose Colonies A M L peripheral blood cells and BMF were first lysed at 37°C overnight with a lysis buffer containing TNE (10 mmol/L Tris, 150mmol/L NaCl and lOmmol/L EDTA), 0.5% sodium dodecyl sulfate (SDS) and 0.2mg/ml proteinase K and then incubated overnight at 37 °C. Proteinase K was inactivated by heating to 95°C for 10 minutes following which DNA was extracted with phenol once followed by phenol/chloroform (1:1) and chloroform once each and precipitated with ethanol, 3mol/L sodium acetate (pH5.2) and 40ug glycogen as DNA carrier. DNA pellets were then washed with 70% ethanol and dissolved in lx TE (lOmmol/L Tris, 1 mmol/L EDTA). Individual colonies from the direct CFC assay and methylcellulose assays of LTCs were plucked into microcentrifuge tubes containing 1ml PBS using a modified fine-tip Pasteur pipet. After centrifugation, cell pellets were resuspended in 200 ul cell lysis buffer. After overnight incubation at 37 °C or 3 hours at 55 °C, genomic DNA was extracted with the above mentioned method and dissolved in 50 ul Hha I digestion buffer (New England Biolabs, Ontario, Canada). 2.7 PCR based fluorescent HUMARA assay Two aliquots, 7ul each of the colony DNA samples or lOng each of the genomic DNA from frozen A M L cells or BMF, were incubated overnight at 37°C. One aliquot was digested with 20 units of the methylation sensitive restriction enzyme Hha I (New England Biolabs), the other was left undigested. After 10 minutes incubation at 95°C, 5ul of the Hha I digest or the undigested control mixture was added to 0.75ul DMSO, 2.5ul lOxPCR buffer (Life Technologies, Burlington, Ontario, Canada), 0.75ul 50mmol/L MgCl 2 (Life Technologies), 0.5ul lOmmol/L 43 dNTP and 12.5pmol each of the sense and antisense primers (Life Technologies). Finally, 0.5 unit of Platinum Taq DNA polymerase (Life Technologies) and water were added to make up a final volume of 25ul. The first round of the nested PCR reaction was done using the following external primers: sense HUMARA I 5'-GGAAGTAGGTGGAAGATTCAGCCA-3' (nucleotides 130-153 of human androgen receptor (AR) sequence in GenBank, accession number M21748 (Allen et al., 1992)) and anti-sense HUMARA IV 5'-GCTGTGAAGGTTGCTGTTCCTCAT -3'(nucleotides 485-508 of human AR). DNA was amplified for 30 cycles (45 seconds at 94°C, 30 seconds at 60°C and 30 seconds at 72°C) with an initial denaturing step (3 minutes at 94°C). The second round of amplification was carried out with 1/100th of the first round products using the same conditions for 40 cycles and a pair of internal primers: HUMARA II 5'-TCCAGAATCTGTTCCAGAGCGTGC-3' (sense, nucleotides 230-253) and HUMARA III 5'-TACGATGGGCTTGGGGAGAACCAT-3' (anti-sense, nucleotides 434-445). The 5'end of HUMARA II was labelled with fluorescein (Life Technologies). 2.8 Genescan Analysis 2u,l of each PCR product was added to 11.1 ul Template Suppression reagent and 0.4ul Genescan 500 TAMRA size standards (ABI Systems, Foster City, USA). The mixture was then heat denatured for 2.5 minutes at 95°C, cooled on ice and electrophoretically separated on an ABI 310 DNA sequencer. The resulting fluorescein-labelled DNA peaks were quantified using GeneScan® 2.1 software. For analysis of colony DNA PCR products, only results from digested samples where a single DNA peak was obtained on the GeneScan profile were included. Samples showing signals representing both of the alleles after Hha I digestion were considered ambiguous (perhaps the result of incomplete digestion of colony DNA or contamination of DNA from one colony with that from another) and the results were discarded. 44 2.9 Quantitation of relative amounts of PCR products In order to demonstrate that PCR amplification of 2 different AR alleles would be linear and thus allow quantitation of the relative amounts of each, DNA from two male PB samples (A and B) from which the PCR products of the AR alleles could be easily distinguished were mixed in different proportions. After PCR, the amount of amplified product from each allele was calculated from the area of the highest peak in the Genescan profile from that allele. The value obtained when the peak area of PCR products for allele A was divided by the peak area for allele B was plotted against the proportion of sample A DNA that had been mixed with sample B DNA in the original mixture. As shown in Figure 5, there was a linear relationship (slope = 1.375) and a close correlation between these values (r=0.95, P=0.004). However, in each of 2 experiments there was a modest overamplification of the smaller AR allele fragment (B) as compared to the larger (A). In subsequent experiments using genomic DNA from A M L blasts or B M fibroblasts, the allele that was predominantly amplified from A M L blast samples digested with Hhal was designated as the leukemia associated allele or " L " allele. The other allele was designated as the normal allele or " N " allele. The ratio of the peak area of amplified AR PCR products from samples digested with Hhal was calculated by dividing the value for the allele ' L ' by that for the allele ' N ' . This L:N ratio was then corrected for the difference in efficiency of amplification of the 2 alleles. This is done by dividing the L:N ratio from the digested samples with the ratio (L:N) obtained for PCR products amplified from DNA that had not been digested with Hhal. Corrected peak area ratio (PAR)= PAR of Hhal digested sample PAR of undigested control sample 45 Ratio of genomic DNA amount A:B Figure 5. The relationship between the relative amounts of two samples of DNA in a mixture and the peak area ratio (PAR) of the PCR products amplified from this mixture. Two male genomic DNA samples A and B were mixed in different proportions as shown on the X axis. Nested PCR was performed on these mixtures. The PAR for each mixture was determined using GeneScan 2.1 software as shown on Figure 2 and described in the Methods. The ratio of the starting DNA A:B ratio was then plotted against the mean PAR of the PCR products A:B (A=229bp, B=210 bp) from two independent experiments. Linear regression was estimated using Prism 2 software (slope=1.375). 2.10 Overnight 3H-Tdr Suicide Assay Fresh or thawed, cryopreserved AML cells were resuspended at 2xl06cells/ml in serum free media. In most experiments lOOng/ml human Steel Factor (SF), 20ng/ml IL-3 and 20 ng/ml G-CSF were added to the over night culture. However, in some experiments the growth factors were varied as indicated in Chapter 4. Equal volumes of this cell suspension were plated into 100mm Petri dishes (Corning, NY, USA) in the presence or absence of 20uCi/ml of high specific activity 3H-Tdr (25u,Ci/mmol: Amersham, Oakville, Canada) and incubated at 37°C in an atmosphere of 5% C O 2 in air for 16-24 hour. 400u.g/ml cold thymidine (Sigma-Aldrich Canada Ltd., Oakville, Ontario, Canada) was added to medium after the overnight exposure to H-Tdr. Cells were then washed with 40u,g/ml cold thymidine in Iscove's medium with 2% FCS and resuspended in Myelocult (StemCell Technologies, Vancouver, Canada) medium prior to plating 46 in A M L colony forming cell (AML CFC), long term culture-initiating cell (AML LTC-IC) assays or injected into NOD/SCID mice. 2.11 FISH 2.11.1 Metaphase Preparation Although most FISH analysis was performed on interphase cells, metaphase preparation was occasionally needed to confirm the specificity of probes or for probes which were only suitable for metaphase FISH. lxlO 7 thawed A M L cells were cultured in 10ml of Iscove's medium with 20% FCS, lOng/ml IL-3, lOOng/ml SF and lOOng/ml FL, while 0.6ml whole blood from a normal individual was cultured in 10ml of RPMI medium containing 15% FCS, 2mmol/L L-glutamine and 0.1ml phytohemagglutinin leukocyte conditioned medium (StemCell). After 72 hours, 60ul of 10ug/ml of colcemid (Life Technologies) was added to the medium and cells were incubated for an additional 10 minutes. After centrifugation, the cell pellet was resuspended in 15ml hypotonic KC1 solution (0.075 mol/L) and incubated in 37 °C water bath for 10 minutes. This treatment was repeated once or twice to completely lyse the cytoplasmic membranes. The remaining nuclei were then fixed with 3:1 methanol:acetic acid. For methylcellulose colonies, 2 days before the intended date for plucking colonies onto slides, 2 drops of 3 umol/L fluorodeoxyuridine and 120 umol/L uridine in Iscove's medium were added to each 35 mm dish to block DNA synthesis. The dishes were incubated at 37 °C for 16-24 hours, and then the blocking of DNA synthesis was released with 2 drops per dish of 300 mmol/L thymidine in Iscove's medium. Four hours later, 4 drops of colcemid were added to each dish which were then incubated for another 24 hours. 2.11.2 Slide Preparation Fixed cells were dropped onto plain glass slides to allow spreading of individual metaphases and discrimination of different chromosomes. The slides were air-dried and stored in -20 °C 47 freezer if not immediately used for FISH. Individual colonies with and without metaphase preparation were plucked from methylcellulose assays into hypotonic KC1 solution (0.075 mol/L) using modified fine-tip Pasteur glass pipets. After 10 minutes, hypotonic KC1 was removed and nuclei were fixed onto multiwell glass slides (Celline Assoc., New Field, NJ). Bone marrow smears from mice were prepared for FISH by cutting one end off the femur and smearing the cut end on a glass slide. The slide was then fixed in 3:1 methanol: acetic acid for 10 minutes before FISH. 2.11.3 FISH probes DNA probes used for FISH included two DNA plasmids containing the whole chromosome 8 library (8-painting, a gift from Dr. D. Pinkel, University of California at San Francisco) and chromosome 8 centromere specific sequence (D8Z2;ATCC, Rockville, MD), yeast artificial chromosome (YAC) clones for the chromosomel3 centromere (909E9 or 74E3), chromosome 16q22 (959E9) and 16pl3 (854E2) (from CEPH), a bacteria artificial chromosome (BAC) clone (b-IL9 or b-393) for chromosome 5q31 (gifts from Dr. L. Nagarajan, University of Texas, MD Anderson Cancer Center) and DNA containing the 9q34(c-abl) sequences from COS cells (a gift from Dr. E. Henske, Harvard Medical School). The 1 lq23(MLL) probe labeled with digoxigenin (DIG, Digoxigenin-11-dUTP; Boehringer-Mannheim, Laval, Canada) was purchased from Oncor (Gaithersb.erg, MD). The 8-painting and D8Z2 plasmid DNA were labeled with DIG using nick translation. lOOul of nick translation reaction buffer contained 5ug of plasmid DNA with lx nick translation buffer (50mmol/L Tris, pH8.0; 5mmol/L MgCl 2 , 0.1 mmol/L p-mercaptoethanol, 10 ixg/ml BSA, 30 umol/L each of dATP, dGTP and dCTP, 10 umol/L dTTP), lx enzyme mix (0.05 unit/ul DNA polymerase I (pol I, Life Technologies), 7.5xl0"4 unit/u.1 DNAse I (Life Technologies), 5 mmol/L Tris, pH 7.5, 0.5 mmol/L magnesium acetate, 0.01 mmol/L phenylmethylsulfonyl 48 fluoride, 0.1 mmol/L P-mercaptoethanol, 5% glycerol.lO ng/ml BSA), 30 mmol/L DIG-dUTP (Boehringer-Mannheim) and 0.02 unit/ml DNA pol I. This reaction mix was incubated at 16 °C for 90 minutes before lOul stop buffer and 190ul distilled water were added. The labeled DNA was then precipitated with 30ul of 3mol/L sodium acetate, pH 5.2, and 750ul cold 100% ethanol and incubated on dry ice for 15 minutes. The DNA was then pelleted, washed, dried and resuspended in 30ul of water. The concentration of the labeled probe was estimated by running 5ml of nick translation product in a 1% agarose gel and comparing the brightness of the product with the standard bands produced by 0.5p.g of A, phage DNA digested with Hind III (Life Technologies). Human DNA from YAC, BAC or COS cells was first amplified with inter-Alu PCR, using primers 5 '-TCCCAAAGTGCTGGGATTACA-3' and 5'-CTGCACTCCAGCCTGGG-3' (Lengauer, Genomics, 1992). The PCR reaction consisted of 100-200ng DNA, lxPCR buffer (Life Technologies PCR buffer containing 20mmol/L Tris-HCl, pH8.4, 50mmol/L KC1), 50mmol/L of each deoxynucleoside triphosphate (dNTP; Life Technologies), 3mmol/L MgCl2, 25pmol of each primer, 5units of Taq DNA polymerase (Life Technologies) and water to make a final volume of 50ul. PCR was performed in an Ericomp Twinblock thermal cycler. 30 cycles of 96 °C for 1 minute, 37 °C for 30 seconds and 72 °C for 6 minutes were performed after an initial denaturing step (96 °C for 1 minute). Finally, the reaction mix was incubated at 72 °C for an additional 6 minutes. 5ul of PCR product was run on a 1% agarose gel to assess the success of PCR amplification. The PCR products usually appeared as a smear due to the variable size of fragments between two Alu sites in the human genomic DNA. The rest of the PCR product was then labeled with nick translation using the method described for plasmid probes. 49 2.11.4 Slide Hybridization and Visualization Microscope slides with unsorted cells, sorted cells, methylcellulose colonies or bone marrow smears were pretreated in 2xSSC (lxSSC contains 0.15 mol/L sodium chloride and 0.015 mol/L sodium citrate, pH7.0) at 37 °C for at least half an hour, denatured in 70% formamide in 4xSSC at 72 °C for 2 minutes and dehydrated for 2 minutes each in a series of 70%, 80%, 90% and 100% ethanol. The DIG labeled chromosome 8-painting probe, at a concentration of 2ng/ul in Master Mix #1 (50% formamide, 10% wt/vol dextran sulfate, 2xSSC) and 4u.g human Cot-1 DNA (Life Technologies), was denatured at 72 °C for 5 minutes and incubated for an hour at 37 °C to allow hybridization of the Cot-1 DNA to highly repetitive DNA sequences in the probe. The labeled 8 centromeric D8Z2 probe, also used at 2ng/ul, was denatured in 55% formamide, 10% wt/vol dextran sulfate, lxSSC and 100u.g/ml salmon sperm DNA and directly mounted onto slides. Probes made from inter-Alu PCR and nick translation were hybridized at 20-30ng/u.l in Master Mix #1 with human Cot-1 DNA at 37°C for half an hour. The M L L (1 lq23) probe from Oncor was placed at 37 °C for 5 minutes and then applied to slides. In all cases for hybridization to target cells, 2-3 u.1 of denatured probe were applied to each well of multiwell slides and 10ul were applied to plain glass slides. Slides were then covered with glass coverslips, sealed with rubber cement and incubated at 37 °C overnight. The post-hybridization washing steps for the different probes are as follows: For the chromosome 8 painting probe: at 45°C, twice with 55% formamide in 4xSSC (NaCl, 0.6M, Na 3 C6H507.2H 2 0, 0.06M, pH7) for 15 minutes, 2xSSC for 15 minutes, 0.5x SSC for 5 minutes and O.lx SSC for 5 minutes. For the chromosome 8 centromere probe: at 45°C, 50% formamide in 4xSSC for 15 minutes, twice in 55% formamide in 4xSSC for 15 minutes, 2xSSC for 15 minutes, and twice in O.lxSSC for 15 minutes. For the 9q34 probe: at 43°C, three times in 50% formamide in 4xSSC for 5 minutes and three times in 2xSSC for 5 minutes. For the MLL 50 (1 lq23) Oncor probe, 13-centromere, 16q22, 16pl3 and 5q31 PCR probes: at 45°C, twice with 50% formamide in 4xSSC for 15 minutes and 2xSSC for 15 minutes. The slides were then incubated with 1% BSA in 4xSSC for 5 minutes at room temperature following which lOOul solution of 8 ug/ml sheep anti-DIG-fluorescein isothiocyanate (FITC) antibody (Boehringer-Mannheim) in 4xSSC with 1% BSA was evenly spread on each slide with a piece of parafilm. After incubation at 37 °C for 1 hour in the dark the parafilm was removed from each slide which was subsequently washed in 4xSSC, 0.1% Triton X-100 in 4xSSC and PN buffer (0.1 mol/L NaH2P04/Na2HP04, pH8.0, 0.1% Nonidet-P40), each at room temperature for 10 minutes. The fluorescent signals were then amplified by incubating slides with 30ug/ml rabbit anti-sheep FITC (Vector Labs, Burlingame, CA) in PNM buffer (PN buffer with 5 % skim milk powder) for a further hour. Finally, slides were washed three times 5 minutes each in PN buffer at room temperature and once 5 minutes in 0.1% Triton X-100 in 4xSSC and counterstained with 0.1-0.5 u-g/ml propidium iodide (PI) in antifade (200 mmol/L 1.4-diazabicyclo-[2,2,2]-octaine, 2mmol/L Tris, pH 8.0, 90% glycerol). Slides were viewed on a Zeiss Axioplan fluorescence microscope using double or triple bandpass filters to allow simultaneous visualization of FITC and PI signals (Omega Optical Inc., Battleboro, VT). 2.12 Animals NOD/SCID mice were bred and housed in pathogen-free conditions in the British Columbia Cancer Research Center Joint Animal Facility according to protocols approved by the Animal Care Committee of the University of British Columbia. 51 2.13 Transplantation of AML cells in NOD/SCID Mice 8-10 week old mice were irradiated with 350 cGy from a 1 3 7Cs source. Within 24 hours of irradiation, l x l 0 7 unsorted A M L cells previously incubated with or without 3H-Tdr in serum free media or Ixl0 3-8xl0 5 sorted or cultured cells together with lxlO 6 carrier cells (irradiated human B M cells) per mouse were injected via the tail vein. During the entire period after irradiation, the mice were supplied with acidified drinking water with lOOmg/L ciprofloxacin (Bayer AG, Leverkusen, Germany). After 4 and /or 8 weeks, bone marrow aspiration was performed on mice anesthetized by isoflurane (Abbott Laboratories Limited, Saint-Laurent, Quebec) gas supplied by a vaporizer (Ohio Medical Products, Madison, WI). After disinfecting the knee joint with 70% ethanol, a 22-gauge needle on an empty syringe was inserted between the two femur condyles through an angle parallel to the femur axis. With the needle inside the femoral shaft, bone marrow was aspirated into the syringe and subsequently rinsed out with 50% FCS in a medium (Verlinden et al., 1998). Eight to 12 weeks post injection, mice were sacrificed by C O 2 inhalation. Bone marrow was removed from the four long bones of the limbs (femurs and humeri) by flushing them with a medium with 50% FCS. For one of the patient samples, blood cells were also collected by cardiac puncture and spleen cells were obtained by mincing and grinding the tissue in medium. 2.14 Detection of Human Cell Engraftment in Mice with Flow Cytometry An aliquot of 200ul of the cells from mouse tissues were treated with 4ml 7% ammonium chloride to lyse red blood cells (Stem Cell Technologies), placed on ice for 20 minutes, then spun down and re-suspended in 1 lOul Hanks' balanced salt solution with 2% FCS, 0.02% Sodium Azide (HFN) and 5% human serum. Each sample was split into two flow cytometry tubes and placed on ice for 10 minutes. To block mouse nonspecific Fc receptors, 40ul 1 TOO 52 anti-mouse IgG Fc receptor monoclonal antibody (2.4 G2, produced from a hybridoma provided by Dr. Steven Szilavassy) was added to each tube and cells were further incubated on ice for 10 minutes. One aliquot of each sample was incubated with mouse anti-human CD45 conjugated with fluorescein (FITC) (Terry Fox Laboratory, Vancouver, Canada), and the other aliquot with fluoresceinated mouse IgGl (Beckon Dickinson Immunocytometry Systems, San Jose, CA) as an isotype control. After 30 minutes on ice, cells were washed with HFN, spun down and stained with 2 u-g/ml PI in HFN. Finally, unbound PI was washed away and replaced with 250ul fresh HFN. Flow cytometry analysis was performed on a Beckon Dickinson FACScan or FACSort flow cytometer. A gate was set to exclude at least 99.9% of cells labeled with the isotype control, and the percent of CD45+ cells was then determined using the same gate after excluding nonviable cells. The fluorescence activated cell sorting (FACS) profiles of marrow cells from a control mouse that had not been injected with human cells and of normal human peripheral blood cells stained with human anti-CD45 antibody were used as negative and positive standards, respectively, to set quadrant regions to distinguish CD45 negative and positive populations (see Figure 6). The percent CD45+ cells from control mice and from the aliquots stained with isotype control was always below 0.1%. The percent human cell engraftment in mouse tissue samples was calculated as the percentage of the CD45-FITC stained cells in the lower right quadrant region subtracted with the percentage of the isotype control FITC conjugate stained cells in the same region. 53 control mouse test mouse PC I A c CD45-FITC control human B i f f D Mr • 0.03 IgGl-FITC test mouse . ' ^ $ $ ^ • 2 2 . 9 7 CD45-FITC CD45-FITC Figure 6 Detection and determination of human cell engraftment in NOD/SCLD mice The CD45-FITC profiles for control mouse bone marrow and normal human peripheral blood cells are used as negative and positive controls to set quadrants to distinguish CD45" and CD454 populations, as shown in A and B respectively. For each sample, the percent human cell engraftment was determined by subtracting the percentage of the lower right quadrant in the isotype control profile (C) from that of the same region in the CD45 profile (D). For example, the percent engraftment in this representative mouse B M sample is 22.97-0.03=22.94. 54 2.15 Hoechst 33342 (Hst) and Pyronin Y (PY) Staining To evaluate cell cycle status with Hoechst 33342 (Hst) and Pyronin Y (PY) staining of DNA and RNA, respectively, the technique of Gothot et al was used with minor modifications (Gothot et al., 1998). Thawed, cryopreserved cells were incubated overnight in serum free medium to re-activate RNA synthesis. Cells were then harvested with a cell scraper and resuspended in HFN with Hst at a final concentration of 5ug/ml. After incubation at 37 °C for 45 minutes, concentrated PY was added at lug/ml followed by another 45 minutes incubation at 37 °C. Cells were washed twice with the same concentrations of Hst and PY in HFN and a third time in HFN with Hst, PY and 2ug/ml of PI. Finally, cells were resuspended in HFN with Hst and PY and sorted on a FACStar Plus (Becton Dickinson, San Jose, CA). 2.16 High-resolution cell cycle analysis using Ki-67 antibody and 7-AAD Cells were analyzed for Ki-67 expression and DNA content as recently described (Gothot et al., 1997; Jordan et al., 1996) with some modifications. As described in section 1.3.2.3, Ki-67 is a nuclear antigen of which expression increases when cells exit Go phase and enter into Gi phase of the cell cycle, so it was used to distinguish quiescent cells from cycling cells. 7-AAD staining can distinguish S/G2+M cells with an increased DNA content from Go and Gi cells. Sorted cells were washed and resuspended in 1ml phosphate-buffered saline (PBS) with 0.4% formaldehyde. After 30 minutes on ice, an equal volume of PBS +0.2% Triton X-100 was added and cells were left in 4 °C overnight. Cells were then washed and resuspended in PBS+1% FCS and stained with FITC conjugated anti-Ki-67 or an isotype control antibody (clone MIB-1; Beckman-Coulter) for 1 hour on ice. Finally, cells were washed, resuspended in PBS+1% FCS containingljag/ml 7-AAD (Sigma) and kept on ice for at least 3 hours. Stained cells were analyzed using a FACScan flow cytometer (BDIS, San Jose, CA) using FL-1 and FL-3 channels for FITC-Ki-67 and 7-AAD, respectively (see Figure 7). 55 o B 98.5.3 c • • Si-.**, •-:. . v .;yV^ .;:'^ :-.-. • 7 - A A D Figure 7 Reanalysis of cells stained with Hst/PY to determine their cell cycle status according to their staining with anti-Ki-67-FITC and 7-AAD. Cells from patient 10 were sorted into fractions containing Go, Gi and S/G2+M cells according to Hst/PY staining and then fixed with 0.4% formaldehyde and permeabilized with 0.2% Triton X-100 before being stained with anti-Ki-67-FITC and 7-AAD. As expected, 88.67% of the cells in Go determined by Hst/PY staining expressed low levels of Ki-67 and had diploid DNA content (A). 98.53% of the cells in Gi according to the Hst/Py technique expressed Ki-67 but still contained diploid DNA, i.e. were confirmed to be in Gi(B). 75.44% of the S/G2+M cells as determined by Hs/PY staining were also in these phases of the cell cycle as determined by their high expression of Ki-67 and greater than diploid DNA content (C). 56 2.17 Stimulation of Go cells into Mitotic Cell Cycle A M L cells and normal lineage depleted bone marrow cells were stained with Hst and PY and sorted into different phases of the cell cycle as described above. Cells with the lowest staining with Hst and PY were collected as Go cells and cultured in 2ml serum free medium (SFM) with or without lOOng/ml SF, lOOng/ml FL and 20ng/ml IL-3 for 24 or 72 hours. The cultured cells were harvested with a cell scraper and either re-stained with Hst and PY for FACS analysis or plated into CFC or LTC-IC assays or injected into NOD/SCID mice. 2.18 RT-PCR Analyses To detect growth factor gene expression by AML blasts, total RNA was extracted with Trizol (Life Technologies) from aliquots of 1,000 light density mononuclear cells. cDNA was synthesized by reverse transcription using random hexamer (Sigma) priming and reverse transcriptase (Superscript II from Gibco-BRL). 5ul of total cDNA was used for a 50ul total volume primer specific PCR amplification for IL-3, SF, G-CSF, granulocyte macrophage colony stimulating factor (GM-CSF) and Flt-3 ligand (FL) sequences. The PCR cocktails also contained 20mmol/L Tns-HCl (pH 8.4), 50 mmol/L KC1, 2 mmol/L MgCl 2 , 200 umol/L of each dNTP (Gibco-BRL), 5 units of Taq DNA Polymerase (Gibco-BRL) and 20 pmol of specific primers for IL-3 (5'-GCTCCCATGACCCAGACAACGTCC-3' and 5'-CAGATAGAAC-GTCAGTTTCCTCCG-3'), SF (5 '-GGATCTGCAGGAATCGTGTGACTA-3' and 5'-CTTCAGGAGTAAAGAGCCTGGGTT-3'), G-CSF (5'-CTGTGGACAGTGCAGGAAG-CCACC-3' and 5'-GCTGGGCAAGGTGGCGTAGA-ACGC-3'), GM-CSF (5'-CAGG-AGGCCCGGCGTC-TCCTGAAC-3' and 5'-ACAAGCAGAAAGTCCTTCAGGTTC-3'), FL (5'-AACAACCTATCTCCTCCTGCT-3' and 5'-GGCACATTTGGTGACAAAGTG-3') and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (5'-GTCTTCACCACCATGGAGAAGG-3' and 5'-GCCTGCTTCACCACCTTCTTGA-3'). The 57 expected sizes of amplified DNA fragments for these genes were 345 bp (IL-3), 335 bp (SF), 547 bp (G-CSF), 290 bp (GM-CSF), 306 bp (FL) and 493 bp (GAPDH), respectively. Forty cycles of PCR(94 °C for 30 sec, 62 °C fori min and 72 °C for 1 min) were performed after the initial denaturation (94 °C, 5 min). PCR products were detected by electrophoresis in 1.2% agarose gels followed by Southern blotting. To detect ITD of the Flt-3, cDNA was synthesized as described above from 1 ug of total RNA extracted from l-5xl0 6 AML cells with Trizol reagent. 0.8 u.1 of cDNA was used in a 50ul PCR reaction containing 20mmol/L Tris-HCl (pH 8.4), 50mmol/L KC1, 1.5mmol/L MgCl 2 , 2.5 units of Taq DNA Polymerase and 40 pmol of specific primers for the Flt-3 (5'-TGTCGAGCAGTACTCTAAACA-3' and 5'-ATCCTAGTACCTTCCCAAACTC-3'). After an initial denaturation (94 °C, 5 min), 35 cycles (94 °C for 30 sec, 62 °C for 45 sec and 72 °C for 1 min) were performed and followed by an extension step (72 °C, 10 min). PCR products were electrophoresed in 3% agarose gels at 90 V for 4.5 h. 2.19 Southern Blotting 2.19.1 Gel Transfer Ten microliters of PCR products was electrophoresed in a 1.2% agarose gel. Following electrophoresis, the gel was soaked in 0.4N NaOH for at least 5 minutes. After at least 2 hours of capillary transfer of DNA fragments from the denatured gel to Zeta-Probe nylon membrane using 0.4N NaOH, the membrane was fixed in an UV cross-linker for 2-5 minutes and stored in -20 °C freezer if not immediately used for hybridization. 2.19.2 Probes cDNA probes for human IL-3, SF, G-CSF, GM-CSF and FL were released from plasmid vectors using restriction endonuclease digestion. The probe fragments were isolated by agarose gel electrophoresis. The DNA fragments were recovered by electro-elution and purified with 58 phenol/chloroform. DNA recovery was estimated by comparison of band densities on agarose gel electrophoresis. For GAPDH, products of GAPDH specific PCR were used as the hybridization probe. 2.19.3 Probe Hybridization Prior to hybridization, the membrane was soaked in 2xSSC at room temperature and then pre-hybndized at 60 °C in 4.6x SSC, 7.6% formamide, 1.5mmol/L EDTA, 0.86% skim milk, 0.8% SDS, 380 ug/ml salmon sperm DNA and 7.5% dextran sulphate for 1-2 hours. Radiolabeled probes were prepared by incubating 50ng of the denatured DNA fragment in the presence of 20 umol/L each of dATP, dGTP and dTTP, 3 units of the Klenow fragment of polymerase I and 50uCi of [32P]dCTP (Amersham Pharmacia) using a random priming kit (Invitrogen). Unbound [32P]dCTP was removed using a Sephedex G-50 column and the specific radioactivity of the probe was measured by a scintillation counter (BioScan QC-2000) (InterScience Inc., Markham, ON). A maximum of 2xl0 7 cpm/lOOul of labeled probe was added to the pre-hybridization solution and the membrane hybridized overnight at 60 °C. The hybridized membrane was washed at 60 °C for 1 h in a solution of 0.3x SSC and 0.1% SDS and autoradiography was performed by exposing the membrane to Kodak XAR film (Eastman Kodak, Rochester, NY) at -70 °C for 0.5-3 h. 2.20 Enzyme Linked Immunosorbent Assay (ELISA) and Bio-Activity Assay Media were collected from low-density AML PB cells after they were incubated in SFM without growth factors for 72 hours (at a concentration of 2x106 cells/ml) and concentrated by Centricon-10 concentrators (Amicon Inc, Beverly, MA) to about 4-fold. The secretion of IL-3, SF and GM-CSF was evaluated by a sandwich enzyme immunoassay technique with Quantikine kits (R& D) according to the manufacturer's protocols. To detect FL, 1, 2 and 4 times dilution of concentrated conditioned media were incubated with 4xl04BaF3-Flt-3 cells (Immunex) plated in 59 round-bottom 96-well plates. After 3 days at 37°C, 1 u.Ci of 3H-Tdr was added to each well. After 6 hour additional incubation, cells were harvested onto a glass fiber filter using LKB Betaplate cell harvester (Wallac, Turku, Finland), and 3H-Tdr incorporation was counted in a LKB Betaplate liquid scintillation counter (Wallac). Recombinant FL standards were set up in triplicate and the test medium dilutions were tested in duplicate. Concentration of FL in the conditioned media was calculated referring to the standard curve created with 0.02-10ng/ml human recombinant FL (Terry Fox Lab). The lower limit of sensitivity of this assay is approximately 40 pg/ml. 2.21 Calculations and Statistical Analyses 2.21.1 Calculations for % kill of AML progenitors After overnight culture with or without 3H-Tdr, equal proportions of the original cultures were plated in CFC, LTC-IC or NOD/SCID mouse assay without correction for any change in cell numbers during the initial 16 to 24 hour culture period. The % kill of A M L CFC and LTC-IC was calculated using the following formula: % kill = 1YCFC/106 cells cultured without 3H-TdrV(CFC /106 cells cultured with 3H-Tdr)l CFC/106 cells cultured without 3H-Tdr For calculation of the proportion of cytogenetically normal or abnormal progenitors which were cycling, the number of CFC in direct methylcellulose assay or assays from LTC of cells which had or had not been exposed to 3H-Tdr was multiplied by the % normal or abnormal colonies detected by FISH. These values were then used to calculate the % kill of the two populations using the formula above. Cells treated and untreated (control) with 3H-Tdr were each injected into a cohort of 4-6 mice. The % kill of NOD/SL-IC was calculated using the mean percentages of human cell engraftment from both groups of mice, based on the following formula: 60 % kill = [(mean % engraftment of the control group)-(mean % engraftment of the 3H-Tdr group)! mean % engraftment of the control group 2.21.2 statistical analysis The statistical significance of differences between groups was calculated by % analysis or Student t test as indicated. 61 Chapter 3 Polyclonal normal hematopoietic progenitor cells in peripheral blood of patients with AML 62 3.1 Introduction The clonal origin of the malignancies described by the term A M L was initially suggested by the discovery of acquired, nonrandom chromosomal abnormalities in leukemic blasts (Yunis et al., 1981). In parallel studies the phenomenon of X-chromosome inactivation was used to demonstrate the clonality of malignant cells from female AML patients heterozygous for 2 variants of G-6-PD (Fialkow et al., 1981). More recently alternative X-chromosome inactivation (XCI)-based clonality assays which are applicable to a larger proportion of the female population have been developed (Fearon et al, 1986; Allen et al., 1992; Delabesse et al., 1995; Gale et al., 1996; Busque and Gilliland, 1998). In particular, the human androgen receptor (AR) locus contains a highly polymorphic locus in which variable numbers of CAG repeats are found. The frequencies of the different alleles of this gene, as determined by the numbers of these repeats, is such that more than 90% of Caucasian females are heterozygous. In addition, upstream of and proximal to these CAG repeats, a number of cleavage sites for methylation sensitive restriction enzymes are found which are differentially methylated on the inactive vs active X chromosomes. The PCR-based HUMARA assay which exploits these polymorphisms in the AR gene and has been extensively used to study normal and malignant hematopoietic cells including individual blast colonies (Tilley et al., 1989; Allen et al., 1992; Delabesse et al., 1995). We have used LTC of AML cells with supportive feeder layers and cytokines to characterize the functional and phenotypic properties of AML LTC-IC. These progenitors were originally defined by their ability to give rise to more than one CFC progeny which carried the cytogenetic abnormality characteristic of the leukemic clone after 5 weeks in culture (Ailles et al., 1997). In the course of these studies we also identified a number of cytogenetically normal progenitors in A M L blood and bone marrow in spite of the apparent overwhelming presence of malignant blasts in these tissues (Ailles et al., 1997; Feuring-Buske and Hogge, 2001). In some cases we were able to separate these normal and abnormal cells by flow cytometry (Blair et al., 1997; 63 Feuring-Buske and Hogge, 2001) . Other investigators have also documented the presence of cytogenetically normal cells in LTC of A M L samples even though such cells could not be detected in direct CFC assay (Chang et al., 1986; Coulombel et al., 1985; Firkin et al., 1990). G-6-PD enzyme analysis has also been used to demonstrate polyclonal hematopoiesis in LTC from several AML patients (Singer, 1988). However, these studies and our own also suggested that there was heterogeneity among AML samples with respect to the relative number of normal progenitors present. In addition, in none of these reports were both cytogenetics and clonality analysis performed. Some of the karyotypically normal colonies we detected in assays from AML LTC were morphologically abnormal (Ailles et al., 1997). In addition, cytogenetically and/or morphologically normal but nevertheless clonal remissions are well documented in occasional AML patients (Fearon et al., 1986; Fialkow et al., 1987). Thus, it seemed possible that, at least in some cases, karyotypically normal LTC-IC from A M L patients could be part of the malignant clone. The current study was undertaken to determine if the cytogenetically normal progenitors detected in LTC of A M L cells were derived from normal polyclonal hematopoiesis. Clarification of this issue was of interest both to further our understanding of the interaction between normal and malignant hematopoiesis in the leukemic host and of practical concern for ex vivo purging and stem cell expansion protocols. In this study we performed the HUMARA assay on leukemic blasts and, when possible, normal B M fibroblasts as well as individual colonies derived from LTC and CFC assays from newly-diagnosed female AML patients. The A M L samples selected included 3 with the inv (16) and related chromosomal abnormalities and 2 additional cases where we had previously found the frequency of cytogenetically normal LTC-IC to be particularly high (Ailles et al., 1997; Feuring-Buske et al., 2000). The data demonstrate that although leukemic blasts as well as a substantial proportion of directly clonogenic cells from these A M L samples were clonal, in most patient samples the cytogenetically normal LTC-IC were polyclonal and 64 thus truly normal. An ex vivo purging strategy combining the current LTC and a growth factor combination that only supports the expansion of normal LTC-IC but not leukemic progenitors from patients with chronic myeloid leukemia (CML)(Petzer et al , 1997) was used to selectively expand normal progenitors from four A M L patients. 3.2 Results 3.2.1 AML Patients with Cytogenetically Normal LTC-ICs From the current and our previous LTC experiments, a relatively high concentration of cytogenetically normal CFC were detected in LTC initiated with A M L cells from 12 newly-diagnosed patients whose diagnostic marrow samples showed cytogenetic abnormalities (patients 1-12 in Table 3). Although LTCs were initiated with A M L samples in which 31-100% of the cells were karyotypically abnormal, after 5 weeks only 0-56% of the LTC-derived colonies were still cytogenetically abnormal (p<0.001). LTC-ICs from 6/12 A M L samples were exclusively cytogenetically normal. The clinical characteristics and the cytogenetic results from bone marrow cells at diagnosis and colonies after LTC of these patients were listed in Table 3 65 > -a c 15 o O C D I D o u O ^> ' C <u y i u H X) o 1-a, X' GO fe CO £ — o <D s u .O es CO o ' I—I <D C S o o b s (U c o PQ o a CQ < fe "3 fe 1 -<D cs £ C D CS o ^ O fe oo C D . £ £ C D -£ o cs oo o oo C D C D - - i cs fe cr (N O r- oo C D »—* —^ O ur, o o o\ >D oo !£},oo o ^ _ cs cs t~-o ^ rD A u-^ o cs 1/1 o o ON ^ o o w >D o o — rs ^ O ° o g >D p (N ^ ^ ^ O C D cs vo CS I/) I D CS I D C3 >D LD ^ ^ ^ r^-fefefefe fe fe 5! s 00 ON 00 t r i (N (N n — cs m cs C D •vT C D o ^ *3 <0 ^ CD CS I D CS I D V D 00 O N 2 ~ 2 3.2.2 Clonality of PB mononuclear cells and BMF from A M L patients The HUMARA assay was first used to identify 5 female A M L samples where there was at least a 6 base pair difference in the sizes between the PCR fragments amplified from the two AR alleles (equivalent to 2 CAG repeats, as shown in Table 2) so that electrophoretic separation could be reliably obtained (Delabesse et al., 1995; Wu et al., 1999). Al l five samples (patient 1-5 in Table 3) had cytogenetic changes identified on diagnostic B M samples which were detectable by FISH, including 3 with inv(16) or related abnormalities (Table 3). BMF from 3 of the 5 patients were available to act as normal control cells. After digestion of genomic DNA from PB mononuclear cells with the methylation sensitive restriction enzyme Hha I, predominant PCR amplification of only one of the two AR alleles was observed from all 5 samples consistent with the presence of a monoclonal population of leukemic blasts (Figure 8). As shown in Table 4, the corrected allele ratio (L:N) for all the PB mononuclear cell samples was greater than 10:1, suggesting that >80% of the cells in these populations were part of the malignant clone. 67 1 N J N Figure 8 Genescan profiles of PCR products of genomic DNA from A M L blasts and bone marrow fibroblasts (BMF) from patient 5. Genomic DNA from leukemic blasts or BMF was incubated with or without Hha I prior to PCR amplification. PCR products were detected using the Genescan program in the ABI Prism 310 genetic analyzer. A and B, Genomic DNA from blasts without (A) or with Hha I (B) digestion showing a clonal pattern; C and D, Genomic DNA from the BMF digested without (C) or with Hha I (D) showing a polyclonal pattern. The shaded area in figure A indicates the "Peak area" for the ' L ' allele that was used to calculate the Peak area ratio as described in Figure 5 and Chapter 2. L,leukemia associated allele; N, normal allele 68 Table 4 Comparison of the clonality of PB mononuclear cells and with corresponding bone marrow fibroblasts (BMF) Corrected allele ratio Allele L:N Patient HUMARA alleles (bp) PB BMF 1 223*/232 ND + 2 215*/221 13:1 2:1 3 237*/243 105:1 1:1 4 221*/239 >20:11 ND 5 221*/236 23:1 2:1 * size of the L allele. + not determined. £ The amplification of the ' N ' allele was not detectable above background fluorescence. This background fluorescence was used to calculate a minimum estimate of the allele ratio L:N. ** The peak area ratio for the 2 AR alleles after Hhal digestion was corrected for the difference in the efficiency of amplification of the 2 alleles in the PCR reaction by subtracting the peak area ratio obtained for the 2 alleles after PCR without prior Hhal digestion. In contrast to the PB mononuclear cells, the corrected allele ratios of the PCR products from BMF from 3 of the patients ranged between 1:1 and 1:2 after Hhal digestion indicating that these cells were polyclonal (Figure 8, Table 4) (Gale et al., 1996). 3.2.3 Clonality of CFC and LTC-IC in AML PB As shown on Table 5, a proportion of colonies detected in direct CFC assays of A M L PB were cytogenetically abnormal by FISH from 3 of the 4 patient samples where colony growth was obtained. HUMARA assays demonstrated a predominance of CFC in which the leukemia-associated allele (L) was amplified from colony DNA after Hhal digestion. Al l colonies from patient 5 which were analyzed by FISH were cytogenetically normal and HUMARA assays 69 demonstrated relatively equal numbers of colonies with amplified DNA from each of the 2 AR alleles. Thus, the FISH and HUMARA data are consistent in demonstrating the presence of CFC derived from the leukemic clone from the former 3 patients while suggesting that PB CFC from patient 5 are predominantly normal. Table 5 Clonality of CFC colonies # of colonies with amplification of allele Patient CFC #/105 cells FISH % normal (# normal/# total analyzed) L N 1 4700 82 (28/34) 40 0 2 1115 74 (14/19) 38 15 3 735 38 (8/21) 22 2 4 5 ND ND ND* 5 43 100 (19/19) 20 29 ND = not determined. The HUMARA assay was also used to analyze colonies plucked from methylcellulose assays of 5-week-old LTC of A M L PB cells from each of the 5 patients (Table 6). The resulting data are compared with FISH analysis of colonies from the same assays on Table 5. Although up to 100% of the B M cells were reported as being karyotypically abnormal on conventional cytogenetic analysis and the PB mononuclear cell HUMARA assay demonstrated a large clonal population of cells for all 5 of these samples, in 4 cases (samples 2, 3, 4, and 5) (Table 4) only cytogenetically normal LTC-derived colonies were detected. Consistent with these FISH results, clonality analysis demonstrated PCR amplification of the N and L alleles of the AR from a similar proportion of LTC-IC-derived colonies for these 4 patient samples (Table 6, Figure 9). The predicted percentage of normal, polyclonal LTC-IC in these patients thus ranges from 73-70 86% assuming that all of the colonies from which the N allele were amplified as well as a similar number of the colonies from which the L allele was amplified are normal. Table 6 Clonality of LTC-IC colonies # of colonies with amplification of allele Patient LTC-IC-derived CFC #/106 cells FISH % normal (# normal/# total analyzed) L N 1 9601 85 (17/20) 46 4 2 50 100 (19/19) 25 19 3 2050 100 (20/20) 28 16 4 10 100 (17/17) 33 19 5 202 100 (26/26) 32 22 71 A B C 1 J / L Figure 9. Representative Genescan profiles of LTC-IC-derived colonies from patient 5. A, genomic DNA from a colony without Hha I digestion showing the two AR alleles of patient 5. B, genomic DNA from a colony digested with Hha I showing amplification of the L allele. C, genomic DNA from a colony digested with Hha I showing amplification of the N allele. 72 These data are in contrast to those derived from PB cells from patient 1. In this case B M cytogenetics was abnormal in only 38% of cells although the B M was replaced with more than 90% blasts at the time of this analysis. Similarly, all 40 CFC analyzed with the HUMARA assay showed amplification of the ' L ' AR allele although only 6 of 34 (18%) CFC were cytogenetically abnormal (Table 5). The results of cytogenetic analysis of LTC-derived colonies were similar with only 3 of 20 showing the expected chromosome abnormality. However, the HUMARA assay detected amplification of the ' L ' allele from 46 of 50 (92%) of colonies suggesting that the large majority of LTC-IC detected in this patient sample were part of the malignant clone although cytogenetic abnormalities were not seen. Nevertheless, the presence of a small number of colonies where the ' N ' allele was amplified indicates that residual normal hematopoiesis was detectable even in this patient. PB cells from patient 1 were also injected into NOD/SCLD mice. Eight weeks after intravenous injection of 107 cells, 60% of the cells in mouse B M were of human origin. HUMARA assay performed on CD45+ human cells isolated from the B M of these mice amplified only the ' L ' allele after Hha I digestion suggesting that these engrafted cells were derived from the malignant clone as were the majority of progenitors detected in vitro from this patient. When considering the results from all 5 patient samples, the proportion of CFCs that were cytogenetically normal was significantly smaller than that proportion of LTC-ICs and a smaller proportion of normal, polyclonal CFC than LTC-IC was also detected (p<0.05). 73 3.2.4 Purging A M L samples with growth factors As shown in the above experiments, the cytogenetically normal colonies derived from LTC initiated with most AML patients are truly normal. In subsequent experiments we tested the possibility of selectively expanding the numbers of these normal progenitors in AML samples using a growth factor combination previously shown to expand primitive normal progenitors from normal individuals and patients with chronic myeloid leukemia (CML) (Petzer et al., 1997). The four A M L samples selected had a relatively high concentration of cytogenetically abnormal CFC or LTC-IC progenitors detected among input cells compared with other samples shown in Table 3. Cells from these samples were cultured in SFM containing SF, FL, IL-3, IL-6 and G-CSF for up to 10 days before being plated into CFC and LTC assays. As shown in Table 7, culture for 10 days in SFM eliminated all cytogenetically abnormal LTC-ICs from these 4 samples even in the absence of growth factors. However, only normal LTC-ICs from patient 11 could be expanded by the growth factor combination. In the other three cases, the addition of growth factors to SFM did not significantly change the numbers of cytogenetically normal LTC-IC detected (p>0.05). The proportions of cytogenetically abnormal CFCs detected from these 3 patients samples were also reduced in these SFM cultures. However, in only one case (patient 12) were the numbers of cytogenetically normal CFCs expanded. 74 l -o o o £3 o i d ca <u 3 IQ O IQ o I * 00 a '•S Q I C D IQ o Q fe O fe o o 2 fe O + fe o o fe o + fe o o § ^ 1 7" o O C 3 £ o a cs o .£ N O * oo 'r—\* a C N C N C N © O © c '-4—» fe r--© C D O ^ W W C N O C D N O I D <—i U fe r j U H fe oo 8 O C D N O d O 2 S2* ^ 22 U U ^ fe o fe ON * * NO ,_ C D d I D od d o o C N O N O C N O N I D ^—^ o N O o I D O N 0 0 O N O C N I D C D u u *T fe u U H fe fe U I O -13 I fe C D fe o o <U s "5b c o C N h i fe -a fe oo O fe o a CD "5b e o o fe -4—» • i-H fe - H 00 £ fe ,fe a fe + ^ - I O cS ,03 O P > W u u ^ fe u U H fe C N t i l o 2 — fe e 2 fe oo fe O o 2 2 o c .s £ £ <u C fe .22 H 3.3 Discussion In previous studies we have demonstrated the outgrowth of cytogenetically normal hematopoietic colonies from LTC initiated with cells from newly-diagnosed A M L patients (Ailles et al., 1997; Feuring-Buske et al, 2000). We also observed that the concentration of these apparently normal LTC-IC in AML peripheral blood was 10- to more than 100-fold increased as compared to that found in normal steady state blood (Ailles et al., 1997). The large numbers of these circulating progenitors in AML patients whose normal blood and marrow elements were largely replaced by malignant blasts suggested that these LTC-IC may be part of the malignant clone although they lacked the expected chromosomal marker. This possibility was supported by the findings from other investigators who had demonstrated clonal hematopoiesis in LTC from some A M L patients (Bernstein et al., 1992; Singer et al., 1988). However, the fact that progenitors detected in LTC of several other AML samples in these same studies appeared to be polyclonal suggested that, at least in some cases, the large numbers of cytogenetically normal LTC-IC we were detecting in our patients were truly normal and perhaps were in some way 'mobilized' into the peripheral blood as part of the leukemic process. We were particularly interested in certain subtypes of AML, such as those carrying the inv (16) and related abnormalities, where the cytogenetically normal colonies derived from LTC also appeared to be morphologically normal and their numbers could be expanded using certain combinations of feeders and cytokines which support hematopoiesis from normal individuals (Ailles et al., 1997; Hogge et al., 1997). For the current study we chose to combine cytogenetic analysis by FISH with clonality analysis using the HUMARA assay. This latter assay was chosen because more than 90% of females are known to be heterozygous for different AR alleles which can be detected by this PCR-based technique (Busque and Gilliland, 1998). As shown by previous investigators as well 76 as by the current data, the technique can be used to detect and quantitate clones within a polyclonal population and to analyze small DNA samples such as those derived from hematopoietic colonies (Figures 2.1 and 3.1) (Busque and Gilliland, 1998; Delabesse et al, 1995). As expected, in our AML samples the leukemic blasts were a clonal population with little or no evidence of cells expressing the AR allele PCR product that was not associated with the leukemic clone. In contrast, for the 3 patients where marrow samples were available, BMF were clearly polyclonal. The deviation from the theoretical L:N allele ratio of 1:1 detected in the HUMARA assay of BMF from patients 2 and 5 is most likely explained by a mild skewing of the Lyonization ratio which is known to occur commonly in women of all ages (Busque et al., 1996; Gale et al., 1997). However, the alternative possibility, i.e. that there was some contamination of these samples with malignant cells, cannot not be completely excluded. Both FISH and clonality data from CFC assays demonstrated that a substantial portion of directly clonogenic cells from patients 1, 2, and 3 were malignant while those from patient 5 were not. However, it is also clear that most of the CFC from patient 1 that failed to show cytogenetic abnormalities were part of the leukemic clone since the ' L ' AR allele fragment was amplified from all of the 40 colonies analyzed (Table 5). Similar data was obtained when cells from patient 1 were placed in LTC. Although only a small proportion of LTC-derived colonies were cytogenetically abnormal, the leukemia-associated AR PCR product was detected in 46/50 CFC analyzed by HUMARA assay. Thus, like several previous case reports, data from patient 1 demonstrate that clonal hematopoiesis can predominate in LTC as well as CFC assays from A M L patients (Bernstein et al., 1992; Singer et al., 1988). In addition, the current example demonstrates that this may be true even when chromosome abnormalities detected in the diagnostic B M sample cannot be found in these progenitors. This conclusion is likely valid even though we were unable to eliminate the possibility of extreme skewing of Lyonization (>90% inactivation of one X chromosome in normal hematopoietic tissue) in this individual since 77 normal tissue from her was not available. Previous reports suggest that the possibility of such extreme skewing in a woman of her age would be approximately 10% (Fey et al., 1994; Gale et al., 1997). The remaining 4 patient samples studied in this report exhibit a different pattern of clonality among LTC-IC. For each of these no cytogenetically abnormal LTC-IC-derived colonies could be detected and the 2 AR allele fragments were amplified from relatively equal numbers of such colonies. In patients 2 and 3, this contrasts with the results shown on Table 5 from the direct CFC assay where colonies derived from the malignant clone were easily detected. Thus, among these AML patients (which included 3 with chromosomel6q abnormalities), the karyotypic abnormality is a reliable marker of the malignant clone and LTC-IC that are cytogenetically normal are predominantly derived from normal polyclonal hematopoiesis. Combined with our previous results, these data suggest that a large reservoir of normal hematopoietic progenitors is present in patients with certain subtypes of AML at diagnosis in spite of the profound cytopenias observed in these individuals and the replacement of most functioning B M with malignant blasts (Ailles et al., 1997; Feuring-Buske and Hogge, 2001). They also suggest that efforts to purge, purify or expand normal hematopoietic progenitors for therapeutic purposes is more likely to be successful from patients with certain subtypes of AML than others. The sensitivity of the X-chromosome inactivation analysis does not allow us to eliminate the possibility of a small malignant clone admixed with a primarily polyclonal progenitor population (Busque and Gilliland, 1998; Mach-Pascual et al., 1998). However, when combined with the detection of exclusively cytogenetically normal colonies in the LTC assays, the results suggest that malignant LTC-IC were present at very low frequency in patients 2, 3, 4, and 5 in our study. Interestingly, a proportion of AML samples, including some of those from patients with inv(16), also grow relatively poorly in NOD/SCID mice indicating that factors which create 78 a competitive advantage for these leukemic clones in human patients have not yet been duplicated in current in vitro or in vivo assay systems (Ailles et al., 1999a). Certain combinations of cytokines in serum free medium can expand numbers of normal bone marrow LTC-IC by more than 50 fold over 10 days in culture (Zandstra et al., 1997). These same conditions also increased numbers of normal LTC-IC isolated from the peripheral blood of CML patients. However, leukemic LTC-ICs could no longer be detected at the end of the 10 day culture period, consistent with a relatively reduced self-renewal potential for the leukemic progenitors (Petzer et al., 1997). We were interested in the possibility that AML LTC-ICs might have a similar proliferative disadvantage which would allow normal progenitors present in patient PB to be selectively expanded with the same cocktail of cytokines in SFM. We found that 10 days in suspension culture could completely eliminate the cytogenetically abnormal LTC-ICs from all four A M L samples tested even in the absence of growth factors. However, a 5-fold expansion of normal LTC-IC was only seen with one patient sample, which is much less than the greater than 50-fold expansion that has been demonstrated for LTC-ICs from normal bone marrow under the same conditions. The numbers of normal LTC-ICs after 10 day suspension culture from the other 3 AML patient samples were either unchanged or reduced as compared to input levels. Whether the reduced expansion potential of normal LTC-ICs from AML PB is related to a general feature of circulating or mobilized LTC-IC or this is a unique feature of residual circulating normal progenitors in A M L patients, what are the factors that influence the heterogeneity in the growth properties of AML samples and how to use such knowledge to improve our ability to analyze and potentially treat this disease are all questions that will need to be addressed in future studies. 79 Chapter 4 Characterization of the cycling status of leukemic progenitors from patients with AML 80 4.1 Introduction In spite of the abnormalities in hematopoietic cell differentiation which are obvious in the blast cells from patients with AML, the malignant clone appears to be maintained by a hierarchy of progenitor cell types very similar to that seen in normal hematopoiesis (Ailles et al., 1999a; Bonnet and Dick, 1997; Buick et al., 1977; Haase et al., 1995; McCulloch, 1983; Sutherland et al., 1996) . These malignant progenitors include cells which form colonies of blasts in semisolid media when provided with a source of hematopoietic growth factors (AML-CFC), more primitive and rare cells which initiate long term malignant hematopoiesis when co-cultured with a supportive fibroblast feeder layer and cytokines (AML LTC-IC) and leukemic cells that can initiate long-term engraftment in immunodeficient NOD/SCID mice (NOD/SL-IC) (Ailles et al., 1997; Bonnet and Dick, 1997; Buick et al., 1977). The AML LTC-IC are at least 10-fold less frequent in the malignant cell population than the directly clonogenic AML-CFC (Ailles et al., 1997). The frequency of NOD/SL-IC ranges between 0.7-45 per 107 leukemic blasts, which is 200-800 fold lower than the frequency of AML LTC-IC in the same cell population. Each NOD/SL-IC produces at least 106 leukemic blasts and many A M L CFC and LTC-IC (Ailles et al., 1999a). In spite of these differences in the frequency and some functional properties AML LTC-IC and NOD/SL-IC share similar cell surface phenotypes. They are both highly enriched among the CD34+ CD38" subpopulation of leukemic cells from most A M L subtypes and lack expression of CD71, HLA-DR, c-kit and Thy-1 (Blair et al., 1997; Blair et al., 1998; Blair and Sutherland, 2000; Bonnet and Dick, 1997; Sutherland et al., 1996). Thus it is likely that the LTC and NOD/SCID mouse assays detect similar, if not identical, populations of primitive leukemic cells. The apparent competitive advantage that malignant hematopoiesis exhibits over normal blood cell production in AML patients cannot be explained by the proliferation of leukemic blasts, most of which are terminally differentiated. However, it has been demonstrated that most 81 AML-CFC are in active cell cycle while normal CFC and LTC-IC, particularly in steady state peripheral blood, are largely quiescent (Eaves and Eaves, 1988; Minden et al., 1978; Ponchio et al., 1995). It appears likely that the more active proliferation of leukemic progenitors contributes to the dominance of the leukemic clone in vivo. The molecular basis for this continuous leukemic progenitor proliferation has not been completely elucidated but, at least in some cases, may be related to autocrine or paracrine growth factor production by AML blasts (Oster et al., 1989; Young and Griffin, 1986). The cytokines produced by leukemic cells permit a degree of factor independent AML-CFC growth in vitro and appear to facilitate engraftment and proliferation of AML cells in vivo in immunodeficient mice (Ailles et al., 1999a). Thus, it seems likely that escape from normal growth controls through this and perhaps other mechanisms is a feature of malignant progenitors throughout the leukemic cell hierarchy. The current study was undertaken to characterize the proliferative status of AML LTC-IC and NOD/SL-IC, the most primitive AML progenitors that can be detected in vitro and in vivo, respectively, and to compare the cycling status of these primitive progenitors to the more differentiated AML-CFC and, where possible, normal progenitors from the same patient sample. For these studies an overnight 3H-Tdr suicide assay was used. In this technique, the period of exposure to 3H-Tdr is extended from the usual 20 minutes to 16 to 24 hours to allow all cells in the cell cycle to pass through S-phase during the incubation period. The results of previous studies with normal CFC and leukemic cell lines are consistent with the prediction that the maximum cell kill observed would increase from the 50% typically seen with the 20 minute technique to 100% with the more prolonged 3H-Tdr incubation (Ponchio et al., 1995). The variability inherent in the LTC-IC and NOD/SCID engraftment assays makes the increased differential cell kill obtained with 16 to 24 hour 3H-Tdr incubation necessary to allow cycling and noncycling populations of LTC-IC to be reliably distinguished. Using this assay Ponchio et al (Ponchio et al., 1995) have shown that normal LTC-IC from steady state peripheral blood are 82 quiescent while the cycling status of those in the bone marrow is more heterogeneous. In both cases, however, 72 hours exposure to appropriate growth factors was sufficient to trigger quiescent LTC-IC into active cell cycle. Using this same technique we now demonstrate that a large proportion of leukemic LTC-IC as well as CFC in the peripheral blood of patients with newly-diagnosed AML are in active cell cycle, while the majority of NOD/SL-ICs are quiescent. 4.2 Results 4.2.1 Maintenance of A M L progenitors in short term culture In initial experiments the ability of serum free culture conditions to maintain A M L CFC and LTC-IC numbers at input levels overnight was tested. A M L PB cells from 15 untreated patients as shown in Table 8 were first cultured in serum free medium with SF, G-CSF and IL-3 since this growth factor combination had been formerly shown to maintain normal CFC and LTC-IC without a significant change in their numbers for up to 24 hours (Ponchio et al., 1995). Table 9 shows that on average the numbers of A M L CFC and LTC-IC did not change significantly from input values over 16-24 hours under these culture conditions in the absence of 3H-Tdr. The mean±s.d. percent recoveries of CFC and week 5 LTC-IC among 15 A M L samples were 100+55 and 111±55, respectively. Subsequently, the effects of five different growth factor combinations and serum free medium alone were compared on progenitor numbers from 3 of the A M L samples (from patients 6, 7and 17). Al l of these conditions appeared to be equivalent for short-term maintenance of AML progenitors (Table 9). In subsequent experiments to assess the cycling status of different leukemic progenitors, the originally tested conditions using SF, G-CSF and IL-3 were used to allow direct comparison with the previously reported experiments with normal peripheral blood (Ponchio et al., 1995). 83 o y* a PC oo o - B G u 60 a ca s o a u o 0 0 I—I rD C N cr C D C N cr <D cr O N <u u C o O o is 43 is G G G u u u o o o o o o o 0 0 C D cr >/•> CQ £ G bfl O rD + X r - " <-D cr CN © CN X N O " rD CN r - " CN -•—-. rM - 1 . 4q N O T—1 + cr 0 0 r + rD CN X T 3 oo" dd X ca N O " > X N O " cr o CN T3 cr m cr CN CN N O " cr o CN "5 CN CN 0 0 + X 0 0 + ^ X + X" oo" CN + oo" + X oo" A C D C D cr rD T 3 x" X N O " o G X X N O " N O N O N O N O •a J i u CQ , ^ o o I D O N s — ' o *—i CN c~ m J CN i n CN o O N A_, O N O CN C D ID CN O N t~ CD CN CN~ ID ID CD *D O N O N ^—^ * — ' N O t~ CD CN >D I ID CN N O CD ID — 1 CD — ID CN CN CN C3 ( 3 ID ID 2 2 ; g 2 2 2 2 2 : 2 2 2 § 2 2 2 1) 00 U H 2 2 2 2 2 2 2 0 0 * D C D N O N O C N r -O N I D 0 0 N O O N > D 0 0 >D N O 0 0 U - U -o N O C D C D C D 3 G c3 N O r- O O O N o CN CN o o I * 03 Ci a Ci I* a s Ci 00 3 1-3 o X! *0 Ci o3 "3 a 3 oo I— o o o o c + + + CO + H + cos ( x CO + fe + + ' fe + o + col 3 o o l/~> +1 ON o +1 ON +1 oo <N +1 ON O ON +1 ON +1 o i n +1 O o U fe U m +1 +1 oo oo +1 m ON +1 oo in m oo +1 © 0 0 in +1 in in in +1 U I U 00 00 o3 o -4—• T3 eni sse 00 Ci o l-l Q. a. X Ci o Oi l-03 mb ults 3 00 3 Ci XI -4—* 00 00 awin din awin ivi X! < Ci <K X) 03 -a >> 15 lcula -4—' 03 lcula edi ca a « i c s-Ci ye Ci 03 l- 00 3 00 » o3 Ci o i-i Ci £ 00 • 00 l i O O XI C i whe Ci <5 O Ci -*-» o3 Ci 00 T3 S i i-Hi Ci -a qui x> a Ci 00 Ci (3 3 lest S-i lest O o lest » -J—» 'a a Oi o3 00 00 00 pro pro of Ci 00 J-i 3 X! Ci ex X> _3 >. 3 X> 3 o Ci u-00 Ci 3 Q 00 o3 o input± ercenl after input± a ed 0 s -The :tect ean # I D -a a a "oo 3 o C N fe" CO U I O a "5b 3 O C N 3 3 Ci Ci 3 3 O Xj «N O ^ T Ci J 1 a 3 o oo 3 O g o , , i i „ § fe 8 <•> J1 © "c3 ^ x r .a 03 00 O a o T 3 oo 3 w .2 o Ci CO ^ II 5 co H Previously, Ailles, L. et al (Ailles et al., 1999) have shown that 24 hour culture of A M L cells in vitro with or without the addition of growth factors was sufficient to compromise the detection of NOD/SL-IC significantly, while A M L CFC and LTC-IC from the same samples were maintained at input numbers. In our experiments, two time points were chosen to compare the maintenance and cycling status of A M L LTC-IC and NOD/SL-IC in the vitro in the serum free conditions, 4 or 5 weeks and 8 weeks. Five weeks is a commonly used endpoint for the LTC-IC assay, and at week 8, maximal engraftment in NOD/SCID mice can be achieved for most AML samples (Ailles et al., 1999a). Table 10 shows the results of the 3 different functional assays with 6 samples (1, 17-21). As expected, most AML CFC and LTC-IC detected either at week 5 or 8 were well maintained during the 16 hours of serum free suspension culture in serum free medium with growth factors. However, under the same conditions the engraftment levels of cultured human A M L cells in NOD/SCID mice decreased significantly in 3 out of 6 samples compared to uncultured (Table 10) (p<0.05, paired Student t test) Table 10 Comparison of maintenance (% input) of different A M L progenitors 5 week 8 week 4 week 8 week Patient CFC LTC-IC LTC-IC NOD/SL-IC NOD/SL-IC 1 90 174 51 43 28 17 79 178 324 128 100 18 267 73 33 not done 7 19 122 86 127 23 82 20 80 87 145 34 30 21 100 106 277 121 99 Mean+/-s.d. 123±72 117+47 160+118 70±50 58±41 86 4.2.2 Cycling Status of Fresh and Cryopreserved AML-CFC To determine the cycling status of leukemic progenitors, light density, mononuclear cells from the peripheral blood of newly-diagnosed AML patients were incubated in the presence or absence of high specific activity 3H-thymidine for 16-24 hours. In initial experiments, the effect of cryopreservation on the cycling activity of AML-CFC was evaluated. As shown in Table 11 A M L CFC from these initial 3 samples were largely in active cell cycle when freshly isolated from A M L patient blood as expected from the previous work of other investigators (Minden et al., 1978). Furthermore, neither their numbers nor their cycling status was significantly altered by prior cryopreservation (p=0.52 and 0.053, respectively, by paired Student t test). Table 11 Comparison of the cycling status of AML-CFC between fresh and cryopreserved samples CFC/106 cells Patient Control +JH-Tdr % kill by 3H-Tdr A - fresh 1200 175 85 frozen 575 50 91 B - fresh 260 155 40 frozen 375 180 52 C - fresh 13175 7325 45 frozen 16500 6575 60 87 In subsequent experiments using cryopreserved specimens, AML-CFC from all 15 patient samples listed in Table 8 were studied as shown on Table 12. The mean ± s.d. % kill for all 15 samples was 60±28. Nine of these 15 samples showed more than 50% progenitor cell kill following the overnight 3H-Tdr exposure while in 6 samples (patients 10, 11, 12, 14, 18 and 20) the majority of AML-CFC were quiescent. In 3 of these latter 6 all the quiescent CFC could be shown to carry the cytogenetic abnormality expected in the malignant clone. Interestingly, incubation of cells from patient 12 with 3H-Tdr allowed some cytogenetically normal CFC to be detected which were not seen in the absence of 3H-Tdr and there was a significant (p<0.05, %2) increase in the proportion of cytogenetically normal colonies for patient 7 in the presence of H-Tdr. When only the relative numbers of cytogenetically abnormal CFC in the suicide assays with or without 3H-Tdr were considered, a higher % kill of malignant progenitors was calculated for these 2 samples and from patient 15. However, on average the mean + s.d. % kill of cytogenetically abnormal AML-CFC was 61±28, i.e. not significantly different from the value calculated for % kill of total CFC. 88 u fe u <D x j o oo -4—• OS -4—» 00 00 #G "o >% o ID X H X os H l x 1 5 g >^  U < 00 — H - l CO £2 s 3-X> oo •a- O N — i o O N o O N N O O N I T ) —< N O o O N N O O N O N N O O N (N O N 0 0 t o ,—N . — v ^—^ —v o o o o oo o o o o o o (N o T—< o N O »—4 o N O i--« * — ' N ' m I T ) o o oo 1 <n m to CN oo ~* oo 0 0 in <N o 0 0 N O 0 0 oo in o N O o •3-o o t o O N O 0 0 O in o r -oo — i N O m oo in O N 0 0 ° O N ° i n if; — N O r ^ Co* 0 0 o 1 o o o o o o o O N - o m o m JN m -— ' — i m m o m o in o o o m oo „ <N ~ °° N O 8 -2 So o m N O o o oo m O 2 § oo O o © — 0 0 CN| +1 o N O x i 00 e <D > o 1-T3 H I o G O oj CD 3 O Xi 1/1 )-< o -4-» o X! + CO CD ID a oo X) T3 1) s--4—» o oo 3 o <u X! 1-- o H I K o -*—» OJ t-00 O cx X ID o X -a 3 x i O fe u 3 o G 03 O '-*-» ID G CD OO O o o G o t l o cx o 1-cx <a x i (D O G <D o -3 * S 00 >—< ™ x> g ^ G Xj in ^ < ^ O N oo 4.2.3 Cycling Status of AML LTC-IC After overnight culture with or without exposure to 3H-Tdr A M L cells from the same 15 patients were plated in 5-week LTC-IC assays. As shown in Table 13 the mean % kill±s.d. for the total LTC-IC populations detected in these assays was 65±35 indicating a predominantly cycling progenitor population. In fact, in only 3 cases was the % kill detected less than 50%. As expected from our previous studies of A M L samples in LTC (Ailles et al., 1997; Coulombel et al., 1985), in 7 patient samples a significant number of cytogenetically normal LTC-IC could be detected. The mean % kill by 3H-Tdr of these seemingly normal progenitors (78% ± 23) was comparable to that of the cytogenetically abnormal AML LTC-IC (87% ± 13). However, in 2 cases (cells from patients 11 and 15) exposure to 3H-Tdr increased the proportion of cytogenetically normal LTC-IC significantly (p<0.02 by x )• To test whether or not the presence of growth factors in the overnight incubation with H-Tdr would affect the cycling status of AML CFC or 5-week LTC-IC, cells from 2 patient samples were incubated for 16 hours in serum free medium either with or without the 'standard' growth factor cocktail (SF + G-CSF + IL-3). As shown on Table 14, there was no reduction in the % kill of either progenitor cell type when the growth factors were removed. The % kill of A M L LTC-IC detected after 8 weeks in culture was similar to the % kill of 5-week LTC-IC in each of the patients tested (p>0.05, r=0.96)(Table 13) 90 E-I 00 •o -a V o ft U U H — J 53 y u H J u C 3 O _ & s < (JH -5 ^ < w l"7< u "u £ 2 a v 6 Q Q Q Q Q Q Q Q Q Q Q Q O N N O N O o o +1 OO 00 oo oo —. CN ro '—' o Q Q Q Q Q Q Q Q Q N O O N OO in — m N O o\ O N N O N O oo N O "3" O N O N '3 OO O N — ' O N t-~ O O N £ 2 ^ 2 O N O N r-N O —* ^ s N O <-<-> o *-—' 00 © Q 00 o O N O N N O O N O N N O O N © N O oo +1 oo in +1 I "~> N O o O oo o § 2 O N S oo r- O N °2 — —' <N O in in o i n O r~ OO -H i n o m in O JN JN ^5 O Q I T ) o N O O N O o o l~~ O § m w > ' C !Q go oo O in N O N O N O o o O N o O O O r-< o N O x m —1 ^ m T J -o o O o NO O oo ON i n O OO NO NO o o +1 c ea u IS DO > o -o H I X l+H o c o T3 ID X 3 o x in o o t 3 X X ! o + fe 00 3 •3 ID (D I  S S (D CO ID X i T3 H m O -4—» ID 1 -3 00 O PH X ID 3 O X i 3 o3 U 1—1 I u H i—l o 3 X ) 03 03 O "S 3 <D 00 O o X ) CM -8 -a H I X n fe i u H fe ^1 U fe u u o o D r—< I u H fe + o •~ 3 o U O u fe u -a I H + i<3 CO c O T3 C o o 3 o CD '-3 fe NO ON C D - - H C D C D 00 I D C N I D 2 ^ VO ^ o r - ON i CD fe o o fe + CD + 00 l > ON 00 ID oo t--o o ON ON O CN *D O rD CD i 00 CD O L D I D > D CD CN ON fe O 3 o fe + CD + 00 CN a o CN fe 00 U 6 u CO O oo c o CN 2 »—l O 3 <£ c J= £ £ 2 8 ^ bO 00 • S ^ > <u ^ -*-» S I fe « 00 2 ^ fe 1 §• 3 co "•3 g CD S CD 53 cS s 3 g CD C CO H fe" I O oo •5 CD £ 00 C N O N 4.2.4 Cycling status of NOD/SL-IC After overnight culture with or without exposure to 3H-Tdr, A M L cells from 7 patients (1,10,17-21) were injected intravenously into NOD/SCID mice. After 4 and 8 weeks, bone marrow was harvested either by aspiration or sacrificing the mice (see 2.13, Chapter 2). The % human cell engraftment in mouse bone marrow was determined using FACS analysis for human CD45+ cells. The mean percent engraftment from the 3H-Tdr group and the control group were used to calculate % kill of NOD/SL-IC based on the assumption that there is a linear relationship between the number of AML NOD/SL-IC injected and the % of human A M L cells detected in mouse bone marrow. Comparison of engraftment in NOD/SCID mice who had received H-Tdr treated and untreated cells 4 or 8 weeks previously revealed that NOD/SL-IC were largely quiescent (<50% kill) in 6 of 7 patient samples with no significant difference detected between the 2 time points (p>0.05, r=0.75)(Table 15). FISH was performed on at least 100 CD45+ cells sorted from the bone marrow of mice injected with cells from patients 1 and 10 and the % cytogenetically abnormal cells was compared with that detected in the diagnostic marrow aspirates from the patients (Table 16). The proportion of cytogenetically abnormal CD45+ cells in the marrow of mice receiving 3H-Tdr treated and untreated A M L cells was similar to that detected in the original patient marrow aspirate (Table 16). Thus 3H-Tdr treatment did not selectively kill cytogenetically normal or abnormal cells detected in NOD/SCID mice. In comparing the % kill of AML LTC-IC (Table 13) and NOD/SL-IC (Table 15) detected at the 4-5 week and 8-week time points it is apparent that A M L LTC-IC from patients 1,17 and 20 contrast with NOD/SL-IC from the same samples in being predominantly in active cell cycle. In contrast, both progenitor populations from patients 10, 18 and 19 were largely quiescent while those from patient 21 were actively cycling. 93 Table 15 Cycling status of NOD/SL-IC Mean ±s.d.% CD45 + cells of mice injected with lx lO 7 c on Day 0 (# of mice tested) % kill by 3 H-Tdr 4-week 8 -week 4-week NOD/SL-IC 8-week NOD/SL-IC Patient Control* + 3H-Tdr Control + JH-Tdr 1 14±6 (6) 12±6 (6) 1 0 ± 1 0 ( 5 ) 7±4 (5) 17 26 10 2+1(6) 2±2 (6) 1±1 (6) 1+1 (6) 0 38 17 53±29 (6) 40±47 (5) 93±9 (5) 97±3 (4) 24 0 18 ND ND 4±8 (6) 2±3 (6) ND 43 19 23+13 (6) 10+13 (6) 81 + 13 (6) 64±22 (6) 58 20 20 2±2 (4) 4±5 (4) 8±8 (4) 7±8 (3) 0 16 21 18±7 (4) 3±2 (4) 19+13 (4) 4±2 (4) 82* 81* ND: not done * : % CD45+ cells of mice which were injected with cells cultured in the serum free medium (SFM) growth factors without the addition of H-Tdr overnight. KV: The percentage of engraftment in mice injected with patient cells treated with 3H-Tdr was significantly lower than the control group injected with untreated cells. Table 16 Comparison of # of abnormal/ # of total cells tested (% abnormal) in diagnostic B M and among human CD45+ engrafted in NOD/SCID mice 8 weeks after injection Patient Patient B M metaphases Control group JH-Tdr group at diagnosis 1 6/16 (38) 46/203 (23) 57/243 (24) 10 16/16 (100) 98/112 (88) 290/346 (84) 94 4.3 Discussion We have previously demonstrated that co-culture of leukemic blasts from patients with A M L for at least 5 weeks with supportive fibroblast feeder layers and cytokines allows identification and quantitation of a rare, primitive malignant progenitor analogous to normal long term culture-initiating cells. We have termed these cells A M L LTC-IC (Ailles et al., 1997). Recently, primitive leukemic progenitors have also been identified by their ability to initiate long-term engraftment in immunodeficient NOD/SCID mice (Ailles et a l , 1999a; Bonnet and Dick, 1997). Most of the AML LTC-IC and nearly all the NOD/SL-IC are found within the CD34+CD38" fraction of leukemic cells and thus have a cell surface phenotype similar to normal LTC-IC and primitive normal cells that initiate growth in immunodeficient mice (Bonnet and Dick, 1997; Kawagoe et al., 1999; Sutherland et al., 1996). It is thus likely that at least some of the primitive malignant cells detected in the LTC and NOD/SCID mouse systems can maintain leukemic hematopoiesis in human patients. To improve our understanding of leukemic stem cell biology and potentially aid in the rational design of new therapeutic regimens we sought to identify differences in the growth characteristics of different A M L progenitors and their normal cell counterparts. One possibility that might explain the growth advantage that malignant hematopoiesis has over normal blood cell production in vivo is an increased proliferative fraction among the leukemic progenitor cell compartment. In fact, it has been previously demonstrated that AML progenitors which form colonies in growth factor-supplemented methylcellulose are a more actively cycling population than normal CFC (Minden et al., 1978). Our goal in the current study was to extend these observations to the more primitive leukemic progenitor cell compartment represented by AML LTC-IC and NOD/SL-IC. Test cells were exposed to 3H-Tdr overnight to maximize our ability to discriminate predominantly cycling from quiescent progenitor populations in both assays. Since the duration of one mitotic cycle in eukaryotic cells 95 is 16 to 24 hours, exposure of cells to 3H-Tdr for this time period allows all cells in different phases of active cell cycle to pass through S phase where the labeled nucleoside can be incorporated into newly-synthesized DNA and the cell subsequently killed (Ponchio et al., 1995). In previous studies where this method was validated for use with appropriate normal human hematopoietic progenitors controls showed that there was no effective reutilization of H-Tdr with the over night incubation. The concentration of 3H-Tdr used and the period of exposure was sufficient to obtain a maximum kill value of 100% as compared to the more traditional assay using a 20 minute 3H-Tdr exposure where the average maximum cell kill is 50% (Ponchio et al., 1995). In the current study we demonstrated that neither the number of AML LTC-IC or their cycling status changed over the 16 to 24 hour incubation period whether or not growth factors were present in the culture medium. Furthermore, prior cryopreservation also did not change the cycling status of A M L CFC allowing this assay to be performed on frozen samples from our stored bank of leukemic cells. In previous studies, using the same 3H-Tdr technique employed in these experiments, both directly clonogenic cells and LTC-IC from normal steady state peripheral blood were found to be quiescent (mean % kill of both progenitor cell types <20%) (Ponchio et al., 1995). Interestingly, LTC-IC from normal steady state bone marrow or mobilized peripheral blood have a larger proliferative fraction and exposure of both CFC and LTC-IC from normal blood or marrow to appropriate cytokines for a minimum of 72 hours will trigger the entry of most of these progenitors into active cell cycle (Ponchio et al., 1995; Ponchio et al., 1997). Thus, many AML progenitors in the blood of untreated patients contrast with normal blood progenitors by being in active cell cycle and thus appear more similar to normal marrow progenitors in this regard. As demonstrated in Chapter 3, most of the cytogenetically normal LTC-IC detected in the blood of many newly-diagnosed and untreated AML patients are derived from residual normal polyclonal hematopoiesis. In these current experiments it is interesting that the majority of 96 cytogenetically normal LTC-IC detected in A M L patient blood were actively proliferating, in contrast to the highly quiescent state of LTC-IC found in normal steady state blood. It is possible that factors prevalent in the leukemic process in vivo, such as autocrine growth factor production by the leukemic blasts, stimulate the proliferation (or mobilization from the marrow) of normal LTC-IC (Oster et al., 1989; Young and Griffin, 1986). In contrast to the large proportion of actively cycling A M L CFC and LTC-IC we identified in most patient samples, 6 of 7 AML samples contained largely quiescent NOD/SL-IC. More than 10% of A M L CFC and LTC-IC also survived the overnight exposure to 3H-Tdr in 10 out the 15 samples tested (Tables 12 and 13). Particularly, A M L LTC-IC from patients 10 and 19 were exclusively quiescent. Some of the progenitors that survived exposure to 3H-Tdr carried the cytogenetic marker characteristic of the malignant clone in all the AML-CFC assays and 6 of the 8 AML LTC-IC assays. These data demonstrate a reservoir of quiescent leukemic progenitors that may be resistant to standard chemotherapeutic agents. They further suggest that efforts to trigger this population to enter into cell cycle may enhance the efficacy of such regimens. Experiments to determine if growth factors which were shown to stimulate the short-term proliferation of AML blasts and support AML CFC growth in methylcellulose assay (Ailles et al., 1997) can stimulate the entry of quiescent AML CFC and LTC-IC into active cell cycle will be described in Chapter 5. Other investigators have suggested that at least some of the AML progenitors which initiate cobblestone areas in LTC and growth in severe combined immunodeficient (SCLD) mice may be quiescent further supporting the relevance of our study (Terpstra et al., 1996). The current data for the cycling status of AML CFC and LTC-IC contrast with those obtained using the same 3H-Tdr suicide assay on peripheral blood progenitors from patients with CML where >99% kill of both leukemic CFC and LTC-IC was obtained (Ponchio et al., 1995). The variability in the proliferative status of AML CFC and LTC-IC from different patient 97 samples as compared to the consistent and homogeneously cycling status of analogous cells from patients with CML no doubt reflects the greater heterogeneity in the recognized cellular and molecular changes associated with the former as compared to latter disease states. 98 Chapter 5 Isolation and Stimulation of Quiescent Primitive Leukemic Progenitor Cells from Patients with AML 99 5.7 Introduction In Chapter 4, we demonstrated that a significant proportion of A M L LTC-ICs from some patients were actively cycling which was in contrast to the LTC-IC from normal steady state peripheral blood (Guan and Hogge, 2000; Ponchio et al, 1995). However, using the overnight 3H-Tdr suicide assay (Ponchio et al, 1995), the majority of NOD/SL-ICs were found to be quiescent. In fact, we also observed quiescent AML CFC and LTC-IC from a significant proportion of patients (discussed in 4.3). These resting progenitors may contribute to the resistance to chemotherapeutic regimens seen in some patients. Stimulation of these progenitors into active cell cycle should enhance their sensitivity to such drugs which are effective against actively proliferating cells. We have taken advantage of the Hst/PY DNA/RNA staining technique which allows the separation of cells into rather pure Go, Gi and S/G2+M subpopulations to isolate quiescent AML cells. This method has been successfully used for the characterization of the cell cycle status and isolation of quiescent primitive progenitors from both normal hematopoietic and CML cells (Gothot et al., 1997; Gothot et al., 1998; Holyoake et al., 1999). Gothot et al (Gothot et al., 1997) demonstrated that in both B M and mobilized peripheral blood (MPB) CD34+ cells, LTC-IC activity was higher in the Go than in the Gi subpopulation. Go CD34+ cells also had higher proliferative potential than Gi CD34+ cells. Although CD34+ cells which reentered Go after in vitro division stimulated by growth factors regained some LTC-IC ability, the frequency of these progenitors was lower than in the originally quiescent CD34+ cells (Gothot et al., 1997). In a subsequent study by the same group, CD34+ cells from MPB in different phases of the cell cycle were injected into NOD/SCID mice. The repopulating cells were found to be predominantly in Go- Short-term activation with growth factors stimulated these quiescent cells to enter Gi but also largely depleted their repopulating capacity (Gothot et al., 1998). In contrast, both Go and Gi CD34+ leukemic cells from CML patients could engraft NOD/SCID P2 microglobulin knock-out 100 mice for as long as 6 weeks (Holyoake et al, 1999). In addition, when cultured in the absence of added growth factors Go CML cells rapidly begin to proliferate and demonstrate autocrine production of IL-3 and G-CSF as they traversed from Go to Gi (Holyoake et al., 2001). Many growth factors are capable of stimulating the proliferation of A M L leukemic blasts. Ailles et al found that more than 60% of AML blast samples responded to stimulation with the single factors IL-3, SF or FL as demonstrated by 2-4 fold increase of 3H-Tdr incorporation as compared to cultures without growth factors. More than 90% of the 53 samples tested responded to two or three of these factors in combination and the response was the greatest with all three factors. In addition, in A M L CFC assays, over 90% of the patient samples also responded to IL-3, SF, FL or G-CSF added to the methylcellulose media (Ailles et al., 1997). AML-CFC from many AML samples show factor independent growth which appears to correlate with autocrine growth factor production by these cells. Expression of many growth factors, including IL-3, GM-CSF, G-CSF, SF and FL has been well demonstrated in AML blasts (Ailles et al, 1999a; Cole et al , 1996; Young et al., 1988; Young and Griffin, 1986; Zheng, 2001). Recently, an internal tandem duplication (ITD) in the juxtamembrane (JM) domain of the Flt-3 receptor has been found in 20-25% of AML samples. This mutation results in constitutive phosphorylation and activation of the receptor and enhanced engraftment in NOD/SCED mice (Kiyoi et al., 1998; Rombouts et al., 2000). It might also deregulate cell cycling through activation of downstream effectors in the Flt-3 signaling pathways. The purpose of the current study was to determine, firstly, if the progenitors detected among quiescent A M L cells were functionally different from those detected among cycling cells and if these quiescent cells would respond to growth factor stimulation. We were also interested to know if growth factor independent proliferation of Go A M L cells would be seen as it had been in CML and whether or not autocrine stimulation or Flt-3 ITD mutation might be responsible. 101 In this chapter, Go AML cells from 4 patient samples selected where 3H-Tdr suicide assays had demonstrated a significant number of quiescent AML-CFC, LTC-IC and NOD/SL-IC were first isolated with the Hst/PY staining method. The progenitor content of sorted Go AML cells was then compared with that of Gi and S/G2+M cells and the results were further compared with those obtained from with the 3H-Tdr suicide assays from Chapter 4. Subsequently, Go A M L cells were exposed to a combination of growth factors (IL-3, SF and FL) in short term (24 -72 hours) suspension culture and the cycling status and functional properties of the stimulated and unstimulated cells were compared. Autocrine growth factor production and the presence of the ITD mutation of the Flt-3 receptor gene were also studied in these patient samples. 102 5.2 Results 5.2.1 Isolation of A M L cells in Go, Gi and S/G2+M phases using Hst/PY staining A M L cells from 4 patients (Table 17) were stained with Hst/PY and 3 different subpopulations were isolated. Cells in Go have the lowest DNA and RNA content and hence the lowest Hst and PY staining. When Go cells transit to G\, DNA content remains the same, but RNA synthesis and PY staining increases. S/G2+M cells which have undergone DNA synthesis have both high Hst and high PY staining (Figure 10). The proportion of cells in the different cell cycle phases was similar among the 4 patient samples tested (Table 18). The vast majority of the A M L cells were in G! phase (mean % G\± s.d.: 92.7±1.4). The mean % of A M L cells in Go± s.d was 3.4±0.8, which was much lower than the corresponding percentages of Go cells detected in two normal bone marrow samples depleted of mature cells positive for lineage markers (NBM Lin") and containing 28% and 47% of CD34+ cells in marrows A and B, respectively (% Go cells 39.6% and 44.6% for NBM Lin" A and B, respectively). A mean±s.d. of 3.9%±0.7% of the AML cells were in S/G2+M phase, which was similar to the corresponding fraction from the two NBM Lin" samples. FISH demonstrated similar percentages of cytogenetically abnormal cells in the 3 cell cycle subpopulations from patients 1 and 10 and the values were also similar to the results from diagnostic bone marrow samples of these 2 patients (Table 18). 103 G,+S/G,+M > PY G0+G, , , S/G,+M B Hst Hst Figure 10 Hst/PY sorting Cells were stained with Hst/PY and sorted into different cell cycle subpopulations. Go and Gi were discriminated using PY histogram (A) where there are two distinctive peaks representing low and high RNA contents, respectively. Similarly, Gi and S/G2+M phases were separated according to Hst histogram (B) with low and high DNA contents, respectively. A representative Hst/PY dot profile from patient 10 was shown in C 104 Table 17 The clinical characteristics of the A M L patients. Age(yr) FAB Type WBC (% blasts)* Cytogenetics from B M Patient /Sex (xl09/L) (% abnormal) FISH probe^  1 58 F M4 370 (46) 47, XY, +13 (38) 13 centromere 10 21 M M l 252 (90) 46, XY, add (6)(p23), 1 lq23 or MLL t(6;ll)(q27;23)(100) 17 64 F M4 151(75) Normal 19 60 F M5a 47(92) Normal WBC, total PB white blood cell count at diagnosis. The probe used is described in detail in Materials and Methods. 105 Table 18 Fractionation of AML PB cells by Hst/PY staining % of total gated cells (% cytogenetically abnormal) Patient Go G, S/G2+M 1 3.3 (31) 93.6 (29) 3.1 (35) 10 4.6(91) 90.6 (89) 4.8 (90) 17 2.8 (-) 93.2 (-) 4.0 (-) 19 2-9 (-) 93.3 (-) 3.7 (-) Mean + s.d. 3.4±0.8 92.7+1.4 3.9±0.7 NBM Lin" * A 39.6 54.0 6.4 B 44.6 48.2 7.2 *: abbreviation for normal bone marrow lineage depleted samples. 106 The success with which pure populations of cells in the different stages of the cell cycle had been isolated was re-examined by staining Hst/PY sorted cell fractions with an antibody against Ki-67, a nuclear antigen expressed only in cycling cells, and 7-AAD, a DNA dye (Figure 7 and Table 19). On average, the proportion of Hst/PY sorted Go cells which were Ki-67 negative and had low DNA content by 7-AAD staining was 76% (range=46-97%), and the proportion of G\ cells according to Hst/PY staining which were Ki-67 positive and had low DNA content was 82% (range=72-99%). A mean (range) of 57% (43-75%) of S/G2+M phase cells as determined by Hst/PY staining were also in S/G2+M when Ki-67 expression and 7-AAD staining were used for this analysis. Those cells not in S/G2+M subpopulation by the latter analysis appeared to be in Gi. These data confirm that sorting cells by Hst/PY staining enriched cells in Go by greater than 10-fold (Tables 18 and 19). Table 19 % Purity of different cell cycle subpopulations determined by anti-Ki-67 antibody and 7-AAD Patient Go G, S/G2+M 1 89 99 75 10 97 80 68 17 46 79 43 19 72 72 44 Mean±s.d. 76±22 82+11 57±17 107 5.2.2 Comparison of cycling status of AML CFC and LTC-IC determined by 3H-Tdr suicide assay and Hst/PY sorting Cells from each of the 4 patients sorted into 3 subpopulations according to Hst/PY staining were plated into CFC and LTC-IC assays. The results were compared with the cycling status of CFC and LTC-IC at week 5 determined by 3H-Tdr suicide assay as described in 4.2.2 and 4.2.3 (shown in Figure 11). In patients 1,17 and 19 where most of the AML-CFC were found to be actively cycling by the 3H-Tdr suicide assay, over 99% of CFCs were in the G\ and S/G2M subpopulations by Hst/PY staining and less than 1% of them were in Go- In contrast, AML-CFC from patient 10 showed only 19% kill by 3H-Tdr indicating that most of the progenitors were quiescent. However, none of the Hst/PY stained and sorted cells formed colonies in CFC assay in either of the 2 experiments initiated with cells from this patient. The vast majority of AML LTC-ICs from patient 1 and 17 were actively cycling as determined by 3H-Tdr suicide assay. Consistent with this result, nearly all the LTC-ICs from both patients were detected in cultures initiated with Gi and S/G2+M cells from the Hst/PY sort. In patient 10 and 19 where LTC-IC were only detected in the Go subpopulation, there was no kill of these progenitors by H-Tdr. Thus, the cycling status of AML CFC and LTC-IC determined by Hst/PY sorting was consistent with that generated from 3H-Tdr suicide assays for the 4 patient samples. 108 % S/G2+M 03 o o CFC LTC-IC CFC LTC-IC CFC LTC-IC CFC LTC-IC % kill by 3H-Tdr : 61 99 19 99 97 91 Patient 10 17 19 Figure 11 CFC and LTC-IC distribution in different cell cycle subpopulations AML cells were stained with Hst/PY and sorted into 3 different subpopulations as shown on Figure 10 according to their DNA and RNA content. Sorted cells were then plated into CFC and LTC-IC assays. The relative number of CFC or LTC-IC derived colonies contributed by each subpopulation from lxlO 6 total cells within all 3 sorting gates was calculated by multiplying the percentage of this subpopulation in 106 total gated cells by the number of colonies produced by 106 cells from the subpopulation. For example, relative # of CFC from Go in 106 total gated cells=%Go in total gated cells x # of CFC colonies/106 Go cells. The total number of progenitors was the sum of relative numbers of progenitors from all three subpopulations. % of total progenitors in each subpopulation was then calculated by dividing the relative number of colonies from this subpopulation by the total number of progenitors. These results were compared with those from the 3H-Tdr suicide assay as shown in Table 12 and 13. 109 5.2.3 NOD/SL-IC activity was restricted to G 0 A M L cells A M L cells from each of the Hst/PY stained and sorted subpopulations were injected at different cell doses into cohorts of 2 to 3 NOD/SCID mice (Table 20). At 4 and 8 weeks post-injection, bone marrow aspirates were performed and cells were recovered and stained with anti-human CD45 antibody to monitor the human cell engraftment in these mice. At week 12, animals were all sacrificed and their bone marrow was harvested for analysis. Engraftment of unsorted cells from all 4 patient samples was detected at all 3 time points although the proportion of human cells present in mouse B M varied from <1% to >90% after i.v. injection of 1-I0xl06 AML cells. The engraftment level in mice at week 4 post-injection was always lower than the other two time points for each of the 4 samples, so Table 20 shows only the results at week 8 or 12 post-injection when the engraftment was highest. Sorted cells from patient 1 showed significant engraftment only at week 12. 4 of 5 mice injected with 0.2 or 2xl0 5 Go cells and 1 of the 2 mice injected with 106 Gi cells showed engraftment at that time point. However, the proportion of human cells detected in the mouse injected with Gi cells was comparable to that detected in mice injected with 50-fold fewer Go cells. Similar results were obtained with sorted cells from patient 10 although in this case the level of engraftment detected with Go cells was even low at week 12 and human cells could not be detected in the B M of any mice injected with Gi cells. Highest level of engraftment of Go cells from patients 17 and 19 was detected at week 8 and week 12, respectively. The level of engraftment achieved with this cell fraction was comparable to that obtained with 10 to 25-fold higher dose of unsorted cells at 8-week and 12-week time points. Low level engraftment (1.7-2.3%) of Gi cells from patient 17 was detected at week 8 with cell doses 4-fold higher than that of Go cells which resulted in 20 to 40-fold higher level of engraftment. No engraftment was detected in any of the 17 mice injected with S/G2+M cells from any of the 4 patient samples at any time point. These results are 110 consistent with those generated with the overnight H-Tdr suicide assay which demonstrated that most NOD/SL-IC from these patient samples are quiescent (Table 15) 111 Table 20 Engraftment of different Hst/PY stained cell cycle subpopulations in NOD/SCID mice (8-12 week post-injection) patient cell dose/mouse (xlO 5) % C D 45 + 1 - unsorted 10 0.4; 0.44 5 0.34; 0 1 0;0 - Go 2 7.2; 0.44 0.2 0.21;0.20;0 0.02 0;0;0 - G , - .10 0; 0.27 - 1 0;0 0.1 0;0 - S/G 2 +M ' 2 0;0 0.02 0 10 - unsorted 100 0.58; 0.35; 0; 0.10; 2 - Go 8 0.12; 0.18 1 0.22; 0 0.1 0;0 - G , 8 0 4 0 1 0 - S/G 2 +M 1 0;0 0.1 0;0 17 - unsorted 10 99; 86; 77 1 8.2; 19; 22 - Go 1 •91; 44; 74 0.1 40; 39 0.01 0; 0.12 - G , 4 2.3; 1.7 1 0;0.19 0.1 0;0 - S/G 2 +M 0.2 0;0 0.1 0,0 19 - unsorted. 50 93; 83; 2.9 5 14; 2.7; 0.15 - Go 2 80; 81 0.1 1.7; 0.19; 0.29 0.0! 0;0;0 G , iO 0 1 0; 0 0.1 - 0;0 - S/G-.+M 2 ' 0;0 0.2 0;0 0.02 0;0 *: '0'% engraftment is defined for any levels of engraftment below 0.1% 112 5.2.4 The spontaneous entry of Go cells into cell cycle upon in vitro culture The previous experiments had demonstrated a substantial proportion of primitive AML progenitors detected in NOD/SCID mice or LTC to be in Go. In subsequent experiments efforts were made to trigger AML Go cells to enter active cell cycle in serum free culture with growth factors. Go cells were isolated by Hst/PY staining from the 4 A M L samples and plated in serum free suspension culture either with or without lOOng/ml SF, lOOng/ml FL and 20ng/ml IL-3 for 24 or 72 hours. At both time points, cells were harvested and re-stained with Hst and PY. Stained cells were analysed by FACS for cell cycle distribution and in CFC, LTC and NOD/SCID mouse assays. Interestingly, in serum free medium (SFM) alone, a significant proportion of quiescent A M L cells from each of the 4 patients spontaneously exit Go and enter into Gi or S/G2+M (Figure 12 and Table 21). A mean (range) of 17% (1.2-37%) of initially quiescent AML cells were still in Go after 72 hours in culture. In growth factor-supplemented cultures, Go AML cells began cycling more quickly so that by 72 hours the mean (range) proportion of Go cells was 3% (0-7.6%). In contrast, in the absence of growth factors under the same serum free conditions, the majority of Go cells from lineage depleted bone marrow of 2 normal individuals (NBM Lin" A and B) remained quiescent for 72 hours (Figure 12, Table 21). Table 22 shows the comparison of AML-CFC and LTC-IC detected among Go cells before and after culture with and without growth factors. A M L CFC and LTC-IC, including those from the 2 samples where LTC-IC progenitors were initially quiescent (patients 10 and 19 as shown in Table 13), were detected among cells that were in active cell cycle after being cultured for 72 hours at similar or increased numbers as compared to uncultured Go cells (Table 22). In contrast, exit from Go after 72h in culture was associated with reduced ability of AML cells from 2 patients to engraft NOD/SCID mice. Detectable but reduced engraftment for up to 12 weeks was seen with cells from patient 17 that were cultured without growth factors and where 38% of cells remained in Go. However, when the same cells were cultured with growth factors only 5.6% remained in Go and no 113 engraftment could be detected. With Go cells from patient 19 culture with and without the growth factors reduced the % of cells in Go to less than 1% and no engraftment was seen in any mice injected with these cultured cells (Figure 13). On the other hand, mice injected with the same or fewer numbers of uncultured Go cells showed easily detectable engraftment (range of human CD45+ cells 1-98%) at the same time point. Time 0 72 hours no GFs 72 hours+SF+FL+IL-3 AML G 0 cells 93 J (patient 1) NBM Lin" G 0 cells 48 (A) • Hst Figure 12 Short-term SFM culture of Go cells with and without growth factors Go cells were isolated from 4 A M L patient samples and then incubated in SFM with and without lOOng/ml SF, lOOng/ml FL and 20ng/ml IL-3 for 72 hours. G 0 cells from two N B M Lin" samples were only incubated in SFM for 3 days without growth factors. The proportions of 3 subpopulations from two representative experiments with A M L and N B M Lin" samples, respectively, determined by Hst/PY staining at the time of sorting and after SFM culture are shown on the above FACS profiles. 114 Table 21 Stimulation of AML Go cells to enter cell cycle % Go* of total Patient Condition Time 0 24 hours 72 hours No GF SF+FL+IL-3 84.4±19.6 2.1 1.1 1.2+ 0.1 0.6± 0.4 10 No GF SF+FL+IL-3 99.8+0 80.0 66.7 37.0+4.2 0±0 17 No GF SF+FL+IL-3 98.7+1.0 22.4 12.4 16.8+14.4 7.6+0.5 19 No GF SF+FL+IL-3 93.4+6.8 3.1 1.3 11.9+15.3 2.0+2.0 NBM Lin" A N B M Lin" B No GF No GF ND' ND $ 95.9 99.9 88.3 98.3 *: The percent Go at time 0 was determined by re-analyzing sorted Go cells according to Hst/PY staining. The proportions of Go cells at 24hours and 72hours were determined by re-staining the cultured cells with Hst/PY. Each number at Time 0 and 72 hours for A M L samples is represented as mean ± s.d. from two independent experiments while numbers at 24 hours for AML samples and at 24- and 72- hour time points for NBM Lin" Go cells are from one representative experiment. $: ND: not done 115 Table 22 Maintenance of A M L CFC and AML LTC-IC capacity as A M L cells exit G 0 # of colonies/10 input TimeO 72 hours no GF 72 hours+SF+FL+IL-3 Patient 1 % G 0 96.7 1.2 0.8 CFC 4700 51500 29000 LTC-IC 1406 74400 57806 10 % G 0 99.8 34 0 CFC ND ND ND LTC-IC 241 640 447 17 % G 0 98.0 6.6 7.2 CFC 800 NG 3750 LTC-IC 3400 60258 50700 19 % G 0 96.1 1.1 0.6 CFC ND ND ND LTC-IC 10220 32941 9333 ND: not done; NG: no growth 116 ON fe CN CN in o o o + fe + fe 00 n fe fe a C N o r- a o <U o 6 O H fe < N _ O C D fe C N rr_ fe oo >- fe O ( N o o <D o s a o o oo o NO ON 0 0 NO ON 1) o w 1> o o a O N O <D •4-> 1-1 00 f> 00 cd fe d .s 5.2.5 Autocrine growth factor production and/or internal tandem duplication (ITD) associated constitutive activation of the Flt-3 receptor is related to autonomous proliferation of AML cells To investigate the mechanisms which might explain why quiescent leukemic cells would spontaneously exit Go, AML cells from these 4 patients were studied for the presence of autocrine growth factor production and/or ITD mutation of the Flt-3 receptor. Expression of at least 3 growth factor genes was detected in mRNA from AML blasts of every patient sample tested, 4 samples positive for IL-3, 2 for SF, 3 for GM-CSF and 4 for FL, whereas only FL and low level of SF transcripts were seen in 1 of the 2 NBM lin" cells (Figure 13). The expression of G-CSF was not detectable in any AML sample or normal control cells. To determine if expression at the RNA level resulted in production of bioactive growth factors, ELISA analysis was performed on growth media from cultured AML cells. The expression of IL-3 and GM-CSF was detected at >7.8 and 40-760 pg/ml, respectively, in medium from all samples where the corresponding growth factor mRNA could be detected by RT-PCR. However, ELISA could not detect soluble SF protein from either patient 17 or 19 in which SF mRNA had been observed. To detect bioactive FL, AML cell growth medium was used to stimulate the FL-responsive cell line, BaF3-Flt-3 (Shibayama et al., 1998). The limit of sensitivity of the assay was 40pg/ml and low level FL activity in this range was all that could be detected in the test medium. However, both SF and FL are known to exist in membrane bound form which would not have been detected in either the ELISA or BaF3-Flt-3 assay (Brasel et al., 1995; Lyman et al., 1994; Martin et al., 1990). No difference of growth factor production was seen between Go and cycling cells in any case (data not shown). Interestingly, the Flt-3 ITD was detected by RT-PCR in 3 of 4 AML blast cell RNA samples (Figure 14). Both autocrine growth factor production and the Flt-3 ITD mutations seen in these AML samples might be responsible for the autonomous proliferation of the quiescent A M L cells from these samples. 118 RT: IL-3 SF G-CSF GM-CSF FL o a r S 10 Patient 17 71, / \ / \ / \ / \ / \ • H I N B M Lin" A B / \ / \ + - + -GAPDH Figure 14 Growth factor mRNA expression in AML Blasts RT-PCR was carried out with total cDNA synthesized from10,000 A M L blast or normal B M cells using a specific primer for each growth factor. Southern blots of PCR products were hybridized with the corresponding cytokine-specific probes. For each cytokine, the positive control was a murine cell line transfected with the corresponding human cytokine cDNA. 119 Figure 15 Flt-3 ITD in A M L patient samples Total RNA was extracted from l-5xl0 6 blast cells. PCR was performed on total cDNA with specific primers flanking exon 11 and 12 of the Flt-3 receptor where the ITD usually occurs. The positive control was an A M L patient sample known to contain the Flt-3 ITD mutation and the negative control was an ALL sample who did not have ITD mutation. 120 5.3 Discussion As shown in Chapter 4, although many AML CFC and LTC-IC were found to be actively cycling, in every case some progenitors of both kinds were quiescent. Moreover, the NOD/SL-IC from 6 of 7 samples were largely quiescent, further indicating the presence of a reservoir of primitive progenitors among noncylcing leukemic cells. The Hst/PY staining technique allows the identification and separation of Go and cycling cells. With this method, viable cells can be divided into three distinct subpopulations. Go cells contain 2n DNA and low RNA content (Hst low PY l 0 W ) , Gi cells have 2n DNA but higher RNA content (Hst l o w/PY h i g h) while S/G2+M cells contain >2n DNA and high RNA content (Hst h , g h/PY h , g h). Analysis of the progenitor content of these 3 cell subpopulations in AML samples confirmed the cycling status of A M L CFC and LTC-IC which had been previously determined by 3H-Tdr suicide assay. Similarly, NOD/SL-ICs were found almost exclusively in the Go subpopulation which was consistent with the low % kill of NOD/SL-IC by 3H-Tdr in these same patient samples. Primitive quiescent leukemic progenitors have also been observed by other researchers. Terpstra et al used fluorouracil treatment to selectively kill cycling cells and demonstrate that A M L cells that initiate growth in SCID mice and in long-term culture were spared (Terpstra et al., 1996). Using 4', 6-diamidoino-2-phenylindol and Ki67 staining, Jordan et al demonstrated that the vast majority of CD34+CD38" cells from A M L patients were in Go (Jordan, 1999). Holyoake et al recently also identified a rare population of quiescent cells from CML patient samples (Holyoake et al., 1999). In current experiments, the cycling status of A M L LTC-IC and NOD/SL-IC was compared. A significantly larger proportion of the former than the latter progenitor cell type was actively cycling in 2 of 4 patients (pt. 1 and 17). These data are consistent with results from our previous study which demonstrated that NOD/SL-ICs are relatively resistant to retroviral transduction (as are all quiescent cells) while AML CFC and LTC-IC are not (Ailles et al., 1999b). The 121 difference in cycling status between leukemic progenitor populations detected in the LTC and NOD/SCID assays suggest that the 2 assays detect somewhat different (but perhaps overlapping) cell populations in spite of the similarity of the cell surface phenotypes so far described for AML LTC-IC and NOD/SL-IC (Blair et al., 1997; Blair et al., 1998; Blair and Sutherland, 2000; Sutherland et al., 1996) In each of the 4 AML samples tested, Go cells spontaneously entered cell cycle within 3 days in culture even in the absence of growth factor supplements. This behavior is similar to the spontaneous proliferation previously observed for Go CML cells (Holyoake et al., 2001). The addition of IL-3, SF and FL to the culture stimulated A M L Go cells to enter cell cycle more quickly and completely. Our results seem to contradict the observation of Jordan et al who found that only 5-8% of leukemic CD34+CD38" cells entered Gi or S phase after 48 hour incubation in SFM with a similar growth factor cocktail (Jordan et al, 1999). There are several possible explanations for this difference. Firstly, functional progenitor assays were not performed in the latter report (Jordan et al, 1999). Although most A M L LTC-IC and NOD/SL-IC are expected to be among CD34+CD38" A M L cells, the frequency of these progenitors in even this highly purified population can be quite low. In addition, AML LTC-IC may not be found exclusively in the CD34rCD38" fraction from some patient samples (Sutherland et al., 1996) and the engraftment of CD34" cells in NOD/SCID mice can also be observed for some A M L patients (Blair et al., 1998). Thus, assessment of the cycling status of primitive cells using cell surface phenotyping may yield different results than functional assays. Secondly, there was only a small number of patient samples studied in both our current experiments and this previous report, which suggests that patient selection may have influenced both data sets. Al l 4 samples tested in the current experiments were from A M L patients with poor or intermediate prognosis according to their diagnostic cytogenetic abnormalities. In addition, 3 of 4 samples carried the Flt-3 ITD mutation which has also been associated with a poor prognosis (Rombouts et al., 2000) as has 122 high presenting white blood cell count which was also seen in 3 of 4 A M L patients. We and others have demonstrated that high level engraftment of AML samples in NOD/SCID mice is also a feature of poor prognosis leukemias (Ailles et al., 1999a), including those with Flt-3 ITD mutation (Rombouts, Blokland et al. 2000) and which exhibit autocrine growth factor production (Russell et al., 1995). Thus, the findings in the current report may pertain to a subset of AMLs. However, this enhances rather that diminishes the clinical relevance of our study since this subset of AMLs are patients in which more understanding of leukemic stem cell biology is most needed to aid in the design of new therapeutic strategies. Unfortunately, all 4 of the patients from whom samples were obtained for our experiments have subsequently died of their leukemia. A small fraction of cells (0.6-7.6%) from 3 of the 4 patients remained quiescent at the end of 72 hours in liquid suspension culture even with growth factor stimulation. Elevated expression of wild-type Wilms' tumor gene 1 (WT1) has been shown to maintain CD34+CD38" stem cells in a quiescent state (Ellisen et al., 2001) and its expression has been identified in almost all leukemic cells, independent of disease type (Inoue et al., 1994; Oka et al., 2000). Increased expression of WT1 in leukemic progenitor cells relative to the total population of malignant cells might make them less likely to exit Go and therefore be more resistant to chemotherapy. Exit from Go dramatically reduced the ability of A M L progenitors to engraft in mouse bone marrow while their ability to read out in CFC and LTC-IC assays was not affected. Others have demonstrated similar difficulties with normal NOS/SCID mouse repopulation cells (Gothot et al., 1998). This suggests that factors which determine the ability of normal stem cells to home to and engraft in the B M microenvironment may also affect malignant stem cells. In addition, the low engraftment of CML CD34 cells in NOD/SCID mice compared to an almost complete replacement of mouse bone marrow by blasts from many AML samples may perhaps be explained by the higher proportion of actively cycling CML progenitors (Eaves et al., 1998; Holyoake et al., 1999). 123 Autocrine growth factor production was observed for leukemic blasts from all 4 samples studied here. Included among the factors studied was FL for which RNA was detected in each sample. This result is consistent with a recent study showing FL expression in all 71 AML patient samples tested (Zheng, 2001). Normal CD34+ cells from PB and B M have also been previously found to produce FL (Majka et al., 2001), a finding which was confirmed in one of the two NBM Lin" control samples we tested. However, quiescent N B M Lin" CD34 enriched cells from this sample did not show the spontaneous proliferation observed in leukemic cells. Hence, F L expression may not be critical for the autonomous entry into cell cycle of the quiescent leukemic cells, particularly in samples positive for the Flt-3 ITD which can induce constitutive activation of Flt-3 signaling pathways (Kiyoi et al, 1998). Our current study has provided further insight into the pathogenesis of A M L in which leukemic blasts might be maintained primarily by the 'differentiation' of a population of quiescent leukemic progenitor cells. The deregulation caused by autocrine production of growth factors (Russell et al., 1995), by mutations of growth factor receptors or by increased expression of transcription factors such as N F - K B (Guzman et al., 2001) may stimulate these primitive malignant cells to enter cell cycle from a quiescent state more easily than their normal counterparts. These features together with the more actively cycling status of mature leukemic progenitors such as AML C F C which are the progeny of the relatively quiescent leukemic stem cells, permit leukemic blasts to eventually outnumber normal hematopoietic cells in vivo. Nevertheless, a population of deeply quiescent leukemic stem cells exists in some patients that is refractory to short-term growth factor stimulation and may allow the leukemic clone to escape intensive chemotherapy in vivo. When cytotoxic drugs are removed, these resting leukemic cells can enter active cell cycle and cause leukemic relapse. In addition to more active proliferation, other differences between leukemic and normal hematopoietic stem cells, such as a lower frequency of apoptotic cell death (Russell et al., 1995) or shorter cell cycle duration (Strom et al., 124 2000), may also be necessary to allow AML progenitors to prevail over their normal counterparts in human patients. 125 Chapter 6 Conclusions and Future Prospectives 126 One possible treatment for A M L patients is myeloablative therapy followed by autologous stem cell transplantation. A major concern about autologous transplantation is that leukemic cells may not be completed eradicated from the autologous grafts or there may exist a 'hidden' clonal pre-leukemic progenitor cell population that will reinitiate frank leukemia and cause relapse. Cytogenetic analyses can detect minimal residual leukemic cells with known chromosome abnormalities. However, in the absence of a chromosome marker, alternative strategies are required to investigate the clonal origin of potentially neoplastic cell populations. Our experiments on the clonality of A M L LTC-IC using the HUMARA assay have provided strong evidence that most of the cytogenetically normal progenitors detected in this system are polyclonal and truly normal. The high concentration of these normal progenitors that have been observed in the blood of some A M L patients suggests that relatively normal numbers of regenerating hematopoietic progenitors will be present after chemotherapy induced remission and available for harvest for subsequent autologous transplant. These results also suggest the possibility of purging patient blood of residual leukemic progenitors ex vivo and potentially expanding normal progenitor numbers with suitable cytokine cocktails. However, our preliminary experiments showed that cytokine response could be heterogeneous among different patient samples (Table 7). Thus, an optimal growth factor combination may need to be selected for different patients. Recently, cytogenetically normal cell populations have been isolated from A M L patients by FACS sorting according to different cell surface phenotypes. For example, Blair et al (Blair et al., 1997) found that most colonies derived from 2-8 week old suspension cultures initiated with the CD34+ Thy-1+ fraction of AML patient cells did not have the cytogenetic abnormalities seen in diagnostic bone marrow samples. The same group of researchers later also found that leukemic progenitor cells that were capable of engrafting the NOD/SCID mice did not express c-kit, unlike normal NOD/SCID mouse engrafting cells (Blair 127 and Sutherland, 2000). More recently, Feuring-Buske detected exclusively cytogenetically normal cells in a CD34+CD38" subpopulation that was capable of effluxing the DNA dye, Hoechst 33342, from 9 AML patients, although, at diagnosis, cells from these patients were 12-100% cytogenetically abnormal (Feuring-Buske and Hogge, 2001). Thus, cell sorting strategies have been developed to purify apparently normal progenitor cells from A M L patient samples. However, HUMARA analysis of cytogenetically normal progenitors from some patients such as patient 1 described in Chapter 3 suggests that in some cases there may be a pre-leukemic LTC-IC population without cytogenetic changes which is nevertheless part of the malignant clone. Precaution is required to identify cases similar to patient 1 prior to isolation and expansion of normal progenitors from these patients for autologous transplantation. Our long-term culture studies also suggest that there may be a relationship between cytogenetic abnormalities and other prognostic features and the presence of normal progenitors in A M L patients. We have previously observed that most patients with an inversion of chromosome 16 or related abnormalities, which are commonly associated with a good prognosis, produced very few colonies in long-term cultures and these colonies were almost 100% normal (Table 3). However, assessment of the relationship between the presence of a large reservoir of normal polyclonal LTC-IC at diagnosis and clinical prognosis would require the study of much larger numbers of patients. Cell cycle status is an important feature of a cell population. Many cancer cells gain a growth advantage over normal cells through deregulated proliferation. Most A M L leukemic blasts have only limited proliferative potential and the duration of each cell cycle in blast cells and AML CFCs seems to be similar to that of analogous normal hematopoietic cells (Andreeff, 1986; Minden et al., 1983). However, our results suggest that the increased proportion of actively cycling leukemic progenitor cells might explain the overwhelming growth of malignant blasts in vivo. The results from Chapter 4 have clearly demonstrated that more A M L CFC and LTC-IC are in active cell cycle than their normal counterparts in steady-state peripheral blood. There are 128 several possible mechanisms to explain their active proliferative state. One is that the deregulated expression of cell cycle regulatory proteins such as pRb, p53 and cyclin E and autocrine production of stimulatory growth factors (Kornblau et al., 1998; Sugimoto et al., 1991). The activation of the MAP kinase pathway through either Ras mutation or the Flt-3 ITD might also be responsible (Hayakawa et al., 1998; Towatari et al., 1997). Targeting of these candidate genes with specific inhibitors might allow the development of novel therapeutic strategies. Another possible mechanism for the actively cycling status of A M L progenitors is the mobilization of both leukemic and normal progenitors into the PB which might be induced by the leukemic process. Normal mobilized CFC and LTC-ICs were previously found to be more actively cycling populations than the same progenitors in steady state blood (Ponchio et al., 1997). This is similar to our finding in Chapter 4 where similar proportion of cytogenetically normal LTC-ICs from A M L patients were in active cell cycle as that seen in cytogenetically abnormal LTC-IC. Autocrine production of various growth factors by leukemic cells may be able to mimic the situation created when exogenous administration of growth factors such as G-CSF, GM-CSF or SF mobilizes normal progenitors from the marrow microenvironment. These autocrine factors potentially function not only to initiate the mobilization process but also to transmit signals via networks composed of adhesion molecules, growth factor receptors, intracellular signal transduction pathways and matrix metalloproteinases (MMPs). For example, the administration of G-CSF significantly decreases expression of adhesion molecules such as very late antigen (VLA)-4, VLA-5, leukocyte function associated molecule (LFA) 1 and LFA 3 (Fu and Liesveld, 2000). Therefore, the expression of G-CSF as well as its receptor in A M L may contribute to the enhanced migration of normal and leukemic progenitors to the periphery. Elevated expression of matrix metalloproteinases 2 and 9 and endo-beta-D-glucuronidase (heparanases) have also been found in AML cells. These enzymes can degrade basement membrane proteins, thereby facilitating transendothelial migration of hematopoietic cells. The increased expression of these 129 genes has been related to the invasiveness of the leukemia (Bitan et al., 2002; Janowska-Wieczorek et al., 1999). It may also correlate with the mobilization of normal as well as leukemic progenitors into the PB. Further investigation of mechanisms involved in progenitor mobilization in A M L patients and how these different mechanisms interact in leukemia will be helpful. In contrast to many AML CFCs and LTC-ICs, AML LTC-IC from some patients and NOD/SL-ICs from most patients tested were quiescent as shown in both Chapter 4 and Chapter 5. The quiescence of NOD/SL-IC partly explains why it was relatively difficult to retrovirally transduce these cells as compared to AML CFC and LTC-IC in the previous gene transfer experiments carried out in our laboratory (Ailles et al., 1999b). To genetically modify the most primitive quiescent leukemic stem cells, one should consider using lentiviral or adenoviral vectors which do not require the target cells be dividing for successful transduction to occur. In spite of their originally quiescent state, after short-term in vitro culture in serum free medium even without growth factors, most of these primitive leukemic cells rapidly went into active cell cycle. Recruitment of quiescent leukemic cells into cell cycle did not affect their ability to grow in long-term cultures but substantially impaired their ability to engraft NOD/SCID mice. We performed two experiments to attempt to explain the mechanism of this autonomous activation of proliferation in quiescent AML cells. Autocrine production of at least 3 of the selection of growth factors tested including IL-3, SF, GM-CSF and FL were found in all 4 patient samples tested. In addition, 3 of the 4 samples contained Flt-3 ITD mutations. Still unexplained is the mechanism whereby quiescent AML cells are triggered to enter into active cell cycle after engraftment in NOD/SCID mice. Kinetic and cycling studies for different A M L progenitors that home to the NOD/SCID bone marrow after injection are necessary to clarify this issue. Understanding homing and engrafting processes of AML cells in NOD/SCID mice would allow better utilization of this animal model to study AML and manipulate leukemic stem cells. It 130 would also be interesting to explore the role of other mechanisms in the spontaneous entry into cell cycle of these quiescent A M L cells. For example, the downregulation and inactivation of early cell cycle modulating genes such as the INK4 family and the pl30/E2F complex (see Figure 4) in A M L cells may render quiescent cells more easier to enter Gi. In addition, hematopoietic cell phosphatase (HCP) or Src homology region 2 domain-containing phosphatase 1 interferes with signaling of a broad spectrum of growth factor and cytokine receptors, including SF, GM-CSF, M-CSF, IL-3 and Epo by dephosphorylation of the associated growth factor receptor binding protein-2 or Janus family tyrosine kinase 2. It thereby decreases cell response to these stimulatory growth factors and maintains cells in Go (Ihle et al., 1994; Y i and Ihle, 1993)(Figure 3). Recently, Beghini et al (Beghini et al., 2000) found an aberrant splicing in the mRNA of the N terminal src homology 2 domain in HCP in several A M L cases. The level of the aberrant mRNA was higher before treatment than at remission in these patients, suggesting the aberrant HCP RNA was produced by the leukemic clone (Beghini et al., 2000). The downregulation of HCP and abnormal HCP protein in quiescent A M L progenitors may also be responsible for their autonomous entry into cell cycle as we have observed. Based on these results, a model of the cell cycle regulation of leukemic progenitor cells can be proposed. Similar to normal HSCs, the most primitive leukemic stem cells in AML patients are quiescent. Their quiescence may be maintained by the negative regulators such as TGF-P that act on normal HSCs in the bone marrow microenvironment. However, it is also likely that the excessive growth of leukemic blasts in patient bone marrow and periphery send out negative signals to prevent the entry of quiescent A M L cells into cell cycle. p27 and p53, which are two negative cell cycle regulators, have already been found to have elevated expression in AML blasts (Iida et al., 1997; Konikova and Kusenda, 2001). More extensive studies on the expression of early cell cycle regulators in AML blasts will be helpful for further understanding of the in vivo behavior of leukemic stem cells. To increase the number of leukemic blasts, quiescent 131 primitive A M L progenitor cells must be recruited into cell cycle possibly through a reduction of negative cell cycle regulators secreted by leukemic blasts and their own autocrine production of stimulatory growth factors. However, in our experiments there were always some residual quiescent cells in most A M L samples studied even in the presence of growth factors at the end of a 72-hour culture. These deeply quiescent cells might be part of the reservoir of leukemic cells that cause relapse. More mature leukemic progenitors such as some LTC-IC and CFC may be responsible for the rapid expansion of leukemic blast cell numbers since they are more continuously in active cell cycle. The proposed model for the cycling status of different leukemic progenitors is outlined in Figure 16. NOD/SL-IC/ A M L LTC-IC A M L CFC A M L blasts Regulators from B M microenvironment, e.g. TGF-p Autocrine production e.g.GM-CSF, G-CSF IL-3, SF, FL, etc. I 00 u Figure 16 Hypothetical model for the cycling status and cell cycle regulation of different A M L progenitors To evaluate this model of deregulated proliferation in A M L progenitors, many important questions must be answered, for example: Do negative cell cycle regulatory proteins such as p53 or p27 inhibit the spontaneous entry of quiescent leukemic cells into cell cycle? If so, can the suppressed quiescent cells maintain their ability to engraft NOD/SCID mice and initiate growth 132 in CFC and LTC assays to the same extent as the freshly isolated quiescent cells? Do the deeply quiescent leukemic cells that are resistant to short-term exposure to growth factors express autostimulatory growth factors to the same extent as the quiescent cells that can spontaneously enter into cell cycle? Can we stimulate these deeply quiescent cells in vitro with other growth factor combinations? Are they normal or leukemic? Do they have elevated expression of negative regulators such as WT1 or HCP? Such studies would provide important insights into the understanding of A M L disease and new clues for potential therapeutic strategies. 133 References Aglietta, M. , Piacibello, W., Sanavio, F., Stacchini, A., Apra, F., Schena, M. , Mossetti, C , Carnino, F., Caligaris-Cappio, F. and Gavosto, F. (1989) Kinetics of human hemopoietic cells after in vivo administration of granulocyte-macrophage colony-stimulating factor. J Clin Invest, 83, 551-7. Ailles, L.E., Gerhard, B. and Hogge, D.E. (1997) Detection and characterization of primitive malignant and normal progenitors in patients with acute myelogenous leukemia using long-term coculture with supportive feeder layers and cytokines. Blood, 90, 2555-64. Ailles, L.E., Gerhard, B., Kawagoe, H. and Hogge, D.E. (1999a) Growth characteristics of acute myelogenous leukemia progenitors that initiate malignant hematopoiesis in nonobese diabetic/severe combined immunodeficient mice. Blood, 94, 1761-72. Ailles, L.E., Humphries, R.K., Thomas, T.E. and Hogge, D.E. (1999b) Retroviral marking of acute myelogenous leukemia progenitors that initiate long-term culture and growth in immunodeficient mice. Exp Hematol, 27, 1609-20. Alcalay, M. , Orleth, A., Sebastiani, C., Meani, N. , Chiaradonna, F., Casciari, C., Sciurpi, M.T., Gelmetti, V., Riganelli, D., Minucci, S., Fagioli, M . and Pelicci, P.G. (2001) Common themes in the pathogenesis of acute myeloid leukemia. Oncogene, 20, 5680-94. Allen, R.C., Zoghbi, H.Y., Moseley, A.B., Rosenblatt, H.M. and Belmont, J.W. (1992) Methylation of Hpall and Hhal sites near the polymorphic CAG repeat in the human androgen-receptor gene correlates with X chromosome inactivation. Am J Hum Genet, 51, 1229-39. Andreeff, M. (1986) Cell kinetics of leukemia. Semin Hematol, 23, 300-14. Avanzi, G.C., Lista, P., Giovinazzo, B., Miniero, R., Saglio, G., Benetton, G., Coda, R., Cattoretti, G. and Pegoraro, L. (1988) Selective growth response to IL-3 of a human leukaemic cell line with megakaryoblastic features. Br J Haematol, 69, 359-66. Baghdassarian, N. and Ffrench, M. (1996) Cyclin-dependent kinase inhibitors (CKIs) and hematological malignancies. Hematol Cell Ther, 38, 313-23. Baum, C M . , Weissman, I.L., Tsukamoto, A.S., Buckle, A.M. and Peault, B. (1992) Isolation of a candidate human hematopoietic stem-cell population. Proc Natl Acad Sci USA, 89, 2804-8. Beghini, A., Ripamonti, C.B., Peterlongo, P., Roversi, G., Cairoli, R., Morra, E. and Larizza, L. (2000) RNA hyperediting and alternative splicing of hematopoietic cell phosphatase (PTPN6) gene in acute myeloid leukemia. Hum Mol Genet, 9, 2297-304. Berardi, A.C., Wang, A., Levine, J.D., Lopez, P. and Scadden, D.T. (1995) Functional isolation and characterization of human hematopoietic stem cells. Science, 267, 104-8. Bernard, O.A. and Berger, R. (2000) Location and function of critical genes in leukemogenesis inferred from cytogenetic abnormalities in hematologic malignancies. Semin Hematol, 37, 412-9. 134 Bernstein, I.D., Singer, J.W., Smith, F.O., Andrews, R.G., Flowers, D.A., Petersens, J., Steinmann, L., Najfeld, V., Savage, D., Fruchtman, S. and et al. (1992) Differences in the frequency of normal and clonal precursors of colony- forming cells in chronic myelogenous leukemia and acute myelogenous leukemia. Blood, 79, 1811-6. Beutler, E. and Kuhl, W. (1990) The NT 1311 polymorphism of G6PD: G6PD Mediterranean mutation may have originated independently in Europe and Asia. Am J Hum Genet, 47, 1008-12. Beutler, E., Yeh, M. , Fairbanks, V. F. (1962) Normal human female as a mosaic fo X-chromosome activity: Studies using the gene for G6PD deificiency as a marker. Proc Natl Acad Sci USA, 48, 9-16. Bhatia, M. , Wang, J.C., Kapp, U., Bonnet, D. and Dick, J.E. (1997) Purification of primitive human hematopoietic cells capable of repopulating immune-deficient mice. Proc Natl Acad Sci USA, 94, 5320-5. Birkenkamp, K.U., Geugien, M. , Lemmink, H.H., Kruijer, W. and Vellenga, E. (2001) Regulation of constitutive STAT5 phosphorylation in acute myeloid leukemia blasts. Leukemia, 15, 1923-31. Bitan, M. , Polliack, A., Zecchina, G., Nagler, A., Friedmann, Y., Nadav, L., Deutsch, V., Pecker, I., Eldor, A., Vlodavsky, I. and Katz, B.Z. (2002) Heparanase expression in human leukemias is restricted to acute myeloid leukemias. Exp Hematol, 30, 34-41. Blair, A., Hogge, D.E., Ailles, L.E., Lansdorp, P.M. and Sutherland, H.J. (1997) Lack of expression of Thy-1 (CD90) on acute myeloid leukemia cells with long-term proliferative ability in vitro and in vivo. Blood, 89, 3104-12. Blair, A., Hogge, D.E. and Sutherland, H.J. (1998) Most acute myeloid leukemia progenitor cells with long-term proliferative ability in vitro and in vivo have the phenotype CD34(+)/CD71(-)/HLA-DR. Blood, 92, 4325-35. Blair, A. and Sutherland, H.J. (2000) Primitive acute myeloid leukemia cells with long-term proliferative ability in vitro and in vivo lack surface expression of c-kit (CD117). Exp Hematol, 28, 660-71. Bonnet, D. and Dick, J.E. (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med, 3, 730-7. Bos, J.L., Verlaan-de Vries, M. , van der Eb, A.J., Janssen, J.W., Delwel, R., Lowenberg, B. and Colly, L.P. (1987) Mutations in N-ras predominate in acute myeloid leukemia. Blood, 69, 1237-41. Bradbury, D.A., Zhu, Y . M . and Russell, N.H. (1997) Bcl-2 expression in acute myeloblastic leukaemia: relationship with autonomous growth and CD34 antigen expression. Leuk Lymphoma, 24, 221-8. Bradford, G.B., Williams, B., Rossi, R. and Bertoncello, I. (1997) Quiescence, cycling, and turnover in the primitive hematopoietic stem cell compartment. Exp Hematol, 25, 445-53. 135 Brasel, K., Escobar, S., Anderberg, R., de Vries, P., Gruss, H J . and Lyman, S.D. (1995) Expression of the flt3 receptor and its ligand on hematopoietic cells. Leukemia, 9, 1212-8. Bravo, R. (1990) Genes induced during the G0/G1 transition in mouse fibroblasts. Semin Cancer Biol, 1, 37-46. Breems, D.A., Blokland, E.A., Neben, S. and Ploemacher, R.E. (1994) Frequency analysis of human primitive haematopoietic stem cell subsets using a cobblestone area forming cell assay. Leukemia, 8, 1095-104. Broudy, V.C., Smith, F.O., Lin, N. , Zsebo, K.M., Egrie, J. and Bernstein, I.D. (1992) Blasts from patients with acute myelogenous leukemia express functional receptors for stem cell factor. Blood, 80, 60-7. Brown, D., Kogan, S., Lagasse, E., Weissman, I., Alcalay, M. , Pelicci, P.G., Atwater, S. and Bishop, J.M. (1997) A PMLRARalpha transgene initiates murine acute promyelocytic leukemia. Proc Natl Acad Sci USA, 94, 2551-6. Budel, L.M., Touw, LP., Delwel, R., Clark, S.C. and Lowenberg, B. (1989a) Interleukin-3 and granulocyte-monocyte colony-stimulating factor receptors on human acute myelocytic leukemia cells and relationship to the proliferative response. Blood, 74, 565-71. Budel, L.M., Touw, LP., Delwel, R. and Lowenberg, B. (1989b) Granulocyte colony-stimulating factor receptors in human acute myelocytic leukemia. Blood, 74, 2668-73. Buick, R.N., Till, J.E. and McCulloch, E.A. (1977) Colony assay for proliferative blast cells circulating in myeloblasts leukaemia. Lancet, 1, 862-3. Busque, L. and Gilliland, D.G. (1993) Clonal evolution in acute myeloid leukemia. Blood, 82, 337-42. Busque, L. and Gilliland, D.G. (1998) X-inactivation analysis in the 1990s: promise and potential problems. Leukemia, 12, 128-35. Busque, L., Mio, R., Mattioli, J., Brais, E., Blais, N., Lalonde, Y., Maragh, M . and Gilliland, D.G. (1996) Nonrandom X-inactivation patterns in normal females: lyonization ratios vary with age. Blood, 88, 59-65. Busque, L., Zhu, J., DeHart, D., Griffith, B., Willman, C , Carroll, R., Black, P.M. and Gilliland, D.G. (1994) An expression based clonality assay at the human androgen receptor locus (HUMARA) on chromosome X. Nucleic Acids Res, 22, 697-8. Byrne, J.L. and Marshall, C.J. (1998) The molecular pathophysiology of myeloid leukaemias: Ras revisited. Br J Haematol, 100, 256-64. Campos, L., Rouault, J.P., Sabido, O., Oriol, P., Roubi, N. , Vasselon, C , Archimbaud, E., Magaud, J.P. and Guyotat, D. (1993) High expression of bcl-2 protein in acute myeloid leukemia cells is associated with poor response to chemotherapy. Blood, 81, 3091-6. 136 Carow, C.E., Levenstein, M. , Kaufmann, S.H., Chen, J., Amin, S., Rockwell, P., Witte, L., Borowitz, M.J., Civin, C.I. and Small, D. (1996) Expression of the hematopoietic growth factor receptor FLT3 (STK- 1/Flk2) in human leukemias; Blood, 87, 1089-96. Carter, A., Haddad, N., Draxler, I., Israeli, E., Raz, B. and Rowe, J.M. (1996) Expression and role in growth regulation of tumour necrosis factor receptors p55 and p75 in acute myeloblastic leukaemia cells. Br J Haematol, 92, 116-26. Carver-Moore, K., Broxmeyer, H.E., Luoh, S.M., Cooper, S., Peng, J., Burstein, S.A., Moore, M.W. and de Sauvage, FJ . (1996) Low levels of erythroid and myeloid progenitors in thrombopoietin-and c- mpl-deficient mice. Blood, 88, 803-8. Cashman, J., Bockhold, K., Hogge, D.E., Eaves, A.C. and Eaves, C J . (1997a) Sustained proliferation, multi-lineage differentiation and maintenance of primitive human haemopoietic cells in NOD/SCID mice transplanted with human cord blood. Br J Haematol, 98, 1026-36. Cashman, J., Eaves, A.C. and Eaves, C.J. (1985) Regulated proliferation of primitive hematopoietic progenitor cells in long-term human marrow cultures. Blood, 66, 1002-5. Cashman, J.D., Lapidot, T., Wang, J.C., Doedens, M. , Shultz, L.D., Lansdorp, P., Dick, J.E. and Eaves, C.J. (1997b) Kinetic evidence of the regeneration of multilineage hematopoiesis from primitive cells in normal human bone marrow transplanted into immunodeficient mice. Blood, 89, 4307-16. Castilla, L.H., Wijmenga, C , Wang, Q., Stacy, T., Speck, N.A., Eckhaus, M. , Marin-Padilla, M. , Collins, F.S., Wynshaw-Boris, A. and Liu, P.P. (1996) Failure of embryonic hematopoiesis and lethal hemorrhages in mouse embryos heterozygous for a knocked-in leukemia gene CBFB-MYH11. Cell, 87, 687-96. Cattanach, B.M. (1975) Control of chromosome inactivation. Annu Rev Genet, 9, 1-18. Chang, J., Coutinho, L., Morgenstern, G., Scarffe, J.H., Deakin, D., Harrison, C , Testa, N.G. and Dexter, T.M. (1986) Reconstitution of haemopoietic system with autologous marrow taken during relapse of acute myeloblastic leukaemia and grown in long-term culture. Lancet, 1, 294-5. Coffer, P.J., Koenderman, L. and de Groot, R.P. (2000) The role of STATs in myeloid differentiation and leukemia. Oncogene, 19, 2511-22. Cole, S.R., Aylett, G.W., Harvey, N.L., Cambareri, A.C. and Ashman, L.K. (1996) Increased expression of c-Kit or its ligand Steel Factor is not a common feature of adult acute myeloid leukaemia. Leukemia, 10, 288-96. Corral, J., Lavenir, I., Impey, H., Warren, A.J., Forster, A., Larson, T.A., Bell, S., McKenzie, A.N., King, G. and Rabbitts, T.H. (1996) An M11-AF9 fusion gene made by homologous recombination causes acute leukemia in chimeric mice: a method to create fusion oncogenes. Cell, 85, 853-61. 137 Coulombel, L., Eaves, C , Kalousek, D., Gupta, C. and Eaves, A. (1985) Long-term marrow culture of cells from patients with acute myelogenous leukemia. Selection in favor of normal phenotypes in some but not all cases. J Clin Invest, 75, 961-9. Dao, M. and Nolta, J. (1999) Molecular control of cell cycle progression in primary human hematopoietic stem cells: methods to increase levels of retroviral- mediated transduction. Leukemia, 13, 1473-80. Darzynkiewicz, Z. (1988) Cellular RNA content, a feature correlated with cell kinetics and tumor prognosis. Leukemia, 2, 777-87. Delabesse, E., Aral, S., Kamoun, P., Varet, B. and Turhan, A.G. (1995) Quantitative non-radioactive clonality analysis of human leukemic cells and progenitors using the human androgen receptor (AR) gene. Leukemia, 9, 1578-82. Dexter, T.M., Allen, T.D. and Lajtha, L.G. (1977) Conditions controlling the proliferation of haemopoietic stem cells in vitro. J Cell Physiol, 91, 335-44. Dick, J.E. (1996) Normal and leukemic human stem cells assayed in SCID mice. Semin Immunol, 8, 197-206. Dictor, M. , Ehinger, M. , Mertens, F., Akervall, J. and Wennerberg, J. (1999) Abnormal cell cycle regulation in malignancy. Am J Clin Pathol, 112, S40-52. DiMartino, J.F. & Cleary M. L. (1999) Mil arrangements in haematological malignancies: lessons from clinical and biological studies. Br J Haematol, 106, 614-26. Dube, I.D., Eaves, C.J., Kalousek, D.K. and Eaves, A.C. (1981) A method for obtaining high quality chromosome preparations from single hemopoietic colonies on a routine basis. Cancer Genet Cytogenet, 4, 157-68. Eaves, A . C , Barnett, M.J., Ponchio, L., Cashman, J.D., Petzer, A.L. and Eaves, CJ . (1998) Differences between normal and CML stem cells: potential targets for clinical exploitation. Stem Cells, 16, 77-83; discussion 89. Eaves. C , Miller, C , Cashman, J., Conneally, E., Petzer, A., Zanstra, P. and Eaves, A. (1997) Hematopoietic stem cells: inferences from in vivo asssays. Stem Cells, 15 suppl. 1, 1-5. Eaves, C J . and Eaves, A.C. (1988) Regulation of hemopoietic progenitor cell proliferation. Behring Inst Mitt, 85-92. Edwards, A., Hammond, H.A., Jin, L., Caskey, C.T. and Chakraborty, R. (1992) Genetic variation at five trimeric and tetrameric tandem repeat loci in four human population groups. Genomics, 12, 241-53. Ellisen, L.W., Carlesso, N., Cheng, T., Scadden, D.T. and Haber, D.A. (2001) The Wilms tumor suppressor WT1 directs stage-specific quiescence and differentiation of human hematopoietic progenitor cells. EmboJ, 20, 1897-909. 138 Faderl, S., Kantarjian, H.M., Estey, E., Manshouri, T., Chan, C.Y., Rahman Elsaied, A., Kornblau, S.M., Cortes, J., Thomas, D.A., Pierce, S., Keating, M J . , Estrov, Z. and Albitar, M. (2000) The prognostic significance of pl6(rNK4a)/pl4(ARF) locus deletion and MDM-2 protein expression in adult acute myelogenous leukemia. Cancer, 89, 1976-82. Farr, C J. , Saiki, R.K., Erlich, H.A., McCormick, F. and Marshall, C.J. (1988) Analysis of RAS gene mutations in acute myeloid leukemia by polymerase chain reaction and oligonucleotide probes. Proc Natl Acad Sci USA, 85, 1629-33. Fauser, A. A. and Messner, H.A. (1979) Proliferative state of human pluripotent hemopoietic progenitors (CFU- GEMM) in normal individuals and under regenerative conditions after bone marrow transplantation. Blood, 54, 1197-200. Fearon, E.R., Burke, P.J., Schiffer, C.A., Zehnbauer, B.A. and Vogelstein, B. (1986) Differentiation of leukemia cells to polymorphonuclear leukocytes in patients with acute nonlymphocytic leukemia. N Engl J Med, 315, 15-24. Feuring-Buske, M. , Frankel, A., Gerhard, B. and Hogge, D. (2000) Variable cytotoxicity of diphtheria toxin 388-granulocyte-macrophage colony-stimulating factor fusion protein for acute myelogenous leukemia stem cells. Exp Hematol, 28, 1390-400. Feuring-Buske, M . and Hogge, D.E. (2001) Hoechst 33342 efflux identifies a subpopulation of cytogenetically normal CD34(+)CD38(-) progenitor cells from patients with acute myeloid leukemia. Blood, 97, 3882-9. Fey, M.F., Liechti-Gallati, S., von Rohr, A., Borisch, B., Theilkas, L., Schneider, V., Oestreicher, M. , Nagel, S., Ziemiecki, A. and Tobler, A. (1994) Clonality and X-inactivation patterns in hematopoietic cell populations detected by the highly informative M27 beta DNA probe. Blood, 83, 931-8. Fialkow, P.J. (1972) Use of genetic markers to study cellular origin and development of tumors in human females. Adv Cancer Res, 15, 191-226. Fialkow, P.J., Singer, J.W., Adamson, J.W., Vaidya, K., Dow, L.W., Ochs, J. and Moohr, J.W. (1981) Acute nonlymphocytic leukemia: heterogeneity of stem cell origin. Blood, 57, 1068-73. Fialkow, P.J., Singer, J.W., Raskind, W.H., Adamson, J.W., Jacobson, R.J., Bernstein, I.D., Dow, L.W., Najfeld, V. and Veith, R. (1987) Clonal development, stem-cell differentiation, and clinical remissions in acute nonlymphocytic leukemia. N EnglJ Med, 317, 468-73. Firkin, F.C., Birner, R., Russell, S.H. and Garson, O.M. (1990) Contrasting patterns of neoplastic cell behaviour in long-term culture of bone marrow from patients with acute leukaemia and myelodysplastic disorders. A survey of responses in 31 cases with cytogenetic determination of neoplastic status of cultured cells in 17 studies. Br J Haematol, 75, 476-84. Fraser, C.C., Szilvassy, S.J., Eaves, C.J. and Humphries, R.K. (1992) Proliferation of totipotent hematopoietic stem cells in vitro with retention of long-term competitive in vivo reconstituting ability. Proc Natl Acad Sci USA, 89, 1968-72. 139 Fu, S. and Liesveld, J. (2000) Mobilization of hematopoietic stem cells. Blood Rev, 14, 205-18. Gale, R.E. (1999) Evaluation of clonality in myeloid stem-cell disorders. Semin Hematol, 36, 361- 72. Gale, R.E., Fielding, A.K., Harrison, C.N. and Linch, D.C. (1997) Acquired skewing of X-chromosome inactivation patterns in myeloid cells of the elderly suggests stochastic clonal loss with age. Br J Haematol, 98, 512-9. Gale, R.E., Mein, C A . and Linch, D.C. (1996) Quantification of X-chromosome inactivation patterns in haematological samples using the DNA PCR-based HUMARA assay. Leukemia, 10, 362- 7. Gale, R.E., Wheadon, H., Boulos, P. and Linch, D.C. (1994) Tissue specificity of X-chromosome inactivation patterns. Blood, 83, 2899-905. Gandini, E. and Gartler, S.M. (1969) Glucose-6-phosphate dehydrogenase mosaicism for studying the development of blood cell precursors. Nature, 224, 599-600. Gari, M., Goodeve, A., Wilson, G., Winship, P., Langabeer, S., Linch, D., Vandenberghe, E., Peake, I. and Reilly, J. (1999) c-kit proto-oncogene exon 8 in-frame deletion plus insertion mutations in acute myeloid leukaemia. Br J Haematol, 105, 894-900. Gartner, S. and Kaplan, H.S. (1980) Long-term culture of human bone marrow cells. Proc Natl Acad Sci USA, 11, 4756-9. Gilliland, D. G. and Griffin, J. D. (2002) Role of FLT3 in leukemia. Curr Opin Hematol, 9, 274-81. Glimm, H., Oh, I.H. and Eaves, C J . (2000) Human hematopoietic stem cells stimulated to proliferate in vitro lose engraftment potential during their S/G(2)/M transit and do not reenter G(0). Blood, 96,4185-93. Gothot, A., Pyatt, R., McMahel, J., Rice, S. and Srour, E.F. (1997) Functional heterogeneity of human CD34(+) cells isolated in subcompartments of the GO / G l phase of the cell cycle. Blood, 90, 4384-93. Gothot, A., van der Loo, J.C, Clapp, D.W. and Srour, E.F. (1998) Cell cycle-related changes in repopulating capacity of human mobilized peripheral blood CD34(+) cells in non-obese diabetic/severe combined immune-deficient mice. Blood, 92, 2641-9. Grignani, F., Fagioli, M. , Alcalay, M. , Longo, L., Pandolfi, P.P., Donti, E., Biondi, A., Lo Coco, F. and Pelicci, P.G. (1994) Acute promyelocytic leukemia: from genetics to treatment. Blood, 83, 10-25. Grimwade, D., Walker, H., Harrison, G., Oliver, F., Chatters, S., Harrison, C.J., Wheatley, K., Burnett, A.K. and Goldstone, A.H. (2001) The predictive value of hierarchical cytogenetic classification in older adults with acute myeloid leukemia (AML): analysis of 1065 patients entered into the United Kingdom Medical Research Council AML11 trial. Blood, 98, 1312-20. 140 Grimwade, D., Walker, H., Oliver, F., Wheatley, K., Harrison, C., Harrison, G., Rees, J., Hann, I., Stevens, R., Burnett, A. and Goldstone, A. (1998) The importance of diagnostic cytogenetics on outcome in AML: analysis of 1,612 patients entered into the MRC A M L 10 trial. The Medical Research Council Adult and Children's Leukaemia Working Parties. Blood, 92, 2322-33. Guan, Y. and Hogge, D.E. (2000) Proliferative status of primitive hematopoietic progenitors from patients with acute myelogenous leukemia (AML). Leukemia, 14, 2135-41. Guzman, M.L., Neering, S.J., Upchurch, D., Grimes, B., Howard, D.S., Rizzieri, D.A., Luger, S M . and Jordan, C.T. (2001) Nuclear factor-kappaB is constitutively activated in primitive human acute myelogenous leukemia cells. Blood, 98, 2301-7. Haase, D., Feuring-Buske, M., Konemann, S., Fonatsch, C , Troff, C , Verbeek, W., Pekrun, A., Hiddemann, W. and Wormann, B. (1995) Evidence for malignant transformation in acute myeloid leukemia at the level of early hematopoietic stem cells by cytogenetic analysis of CD34+ subpopulations. Blood, 86, 2906-12. Hayakawa, F., Towatari, M. , Iida, H., Wakao, H., Kiyoi, H., Naoe, T. and Saito, H. (1998) Differential constitutive activation between STAT-related proteins and MAP kinase in primary acute myelogenous leukaemia. Br J Haematol, 101, 521-8. Hayakawa, F., Towatari, M. , Kiyoi, H., Tanimoto, M. , Kitamura, T., Saito, H. and Naoe, T. (2000) Tandem-duplicated Flt3 constitutively activates STAT5 and MAP kinase and introduces autonomous cell growth in IL-3-dependent cell lines. Oncogene, 19, 624-31. Heard, E., Clerc, P. and Avner, P. (1997) X-chromosome inactivation in mammals. Annu Rev Genet, 31, 571-610. Hirayama, F., Clark, S.C. and Ogawa, M. (1994) Negative regulation of early B lymphopoiesis by interleukin 3 and interleukin 1 alpha. Proc Natl Acad Sci USA, 91, 469-73. Ho, A. and Dowdy, S.F. (2002) Regulation of G(l) cell-cycle progression by oncogenes and tumor suppressor genes. Curr Opin Genet Dev, 12, 47-52. Hoang, T., Haman, A., Goncalves, O., Wong, G.G. and Clark, S.C. (1988) Interleukin-6 enhances growth factor-dependent proliferation of the blast cells of acute myeloblastic leukemia. Blood, 72, 823-6. Hogge, D.E. (1994) Cytogenetics and oncogenes in leukemia. Curr Opin Oncol, 6, 3-13. Hogge, D.E., Ailles, L.E. and Gerhard, B. (1997) Cytokine responsiveness of primitive progenitors in acute myelogenous leukemia. Leukemia, 11, 2220-1. Hogge, D.E., Lansdorp, P.M., Reid, D., Gerhard, B. and Eaves, C.J. (1996) Enhanced detection, maintenance, and differentiation of primitive human hematopoietic cells in cultures containing murine fibroblasts engineered to produce human steel factor, interleukin-3, and granulocyte colony-stimulating factor. Blood, 88, 3765-73. 141 Holyoake, T., Jiang, X., Eaves, C. and Eaves, A. (1999) Isolation of a highly quiescent subpopulation of primitive leukemic cells in chronic myeloid leukemia. Blood, 94, 2056-64. Holyoake, T.L., Jiang, X., Jorgensen, H.G., Graham, S., Alcorn, M.J., Laird, C., Eaves, A.C. and Eaves, C.J. (2001) Primitive quiescent leukemic cells from patients with chronic myeloid leukemia spontaneously initiate factor-independent growth in vitro in association with up-regulation of expression of interleukin-3. Blood, 97, 720-8. Hori, T.A. and Nakai, S. (1978) Unusual dose-response of chromosome abberrations induced in human lymphocytes by very low dose exposures to tritium. Mutat Res, 50, 101-10. Huang, S. and Terstappen, L.W. (1994) Lymphoid and myeloid differentiation of single human CD34+, HLA-DR+, CD38- hematopoietic stem cells. Blood, 83, 1515-26. Hunter, A.E., Rogers, S.Y., Roberts, LA., Barrett, A.J. and Russell, N. (1993) Autonomous growth of blast cells is associated with reduced survival in acute myeloblasts leukemia. Blood, 82, 899-903. Ihle, J.N., Witthuhn, B.A., Quelle, F.W., Silvennoinen, O., Tang, B. and Y i , T. (1994) Protein tyrosine phosphorylation in the regulation of hematopoiesis by receptors of the cytokine-receptor superfamily. Blood Cells, 20, 65-80. Iida, H., Towatari, M. , Tanimoto, M. , Morishita, Y., Kodera, Y. and Saito, H. (1997) Overexpression of cyclin E in acute myelogenous leukemia. Blood, 90, 3707-13. Inoue, K., Sugiyama, H., Ogawa, H., Nakagawa, M. , Yamagami, T., Miwa, H., Kita, K., Hiraoka, A., Masaoka, T., Nasu, K. and et al. (1994) WT1 as a new prognostic factor and a new marker for the detection of minimal residual disease in acute leukemia. Blood, 84, 3071-9. Jacobson, R.J., Temple, M.J., Singer, J.W., Raskind, W., Powell, J. and Fialkow, P.J. (1984) A clonal complete remission in a patient with acute nonlymphocytic leukemia originating in a multipotent stem cell. N Engl J Med, 310, 1513-7. Janowska-Wieczorek, A., Marquez, L.A., Matsuzaki, A., Hashmi, H.R., Larratt, L.M., Boshkov, L.M., Turner, A.R., Zhang, M.C., Edwards, D.R. and Kossakowska, A.E. (1999) Expression of matrix metalloproteinases (MMP-2 and -9) and tissue inhibitors of metalloproteinases (TIMP-1 and -2) in acute myelogenous leukaemia blasts: comparison with normal bone marrow cells. Br J Haematol, 105, 402-11. Jordan, C.T., Phillips, G. L., Rizzieri, D. A., Luger, S. M. , Grimes, B., Upchurch D. (1999) Cell cycle analyses and apoptosis in primitive leukeumic and normal cells. Blood, 94, 687a. Jordan, C.T., Yamasaki, G. and Minamoto, D. (1996) High-resolution cell cycle analysis of defined phenotypic subsets within primitive human hematopoietic cell populations. Exp Hematol, 24, 1347-55. Kawagoe, H., Humphries, R.K., Blair, A., Sutherland, H.J. and Hogge, D.E. (1999) Expression of HOX genes, HOX cofactors, and MLL in phenotypically and functionally defined subpopulations of leukemic and normal human hematopoietic cells. Leukemia, 13, 687-98. 142 Kelly, L. M. , Liu, Q., Kutok, J. L., Williams, I. R., Boulton, C. L. and Gilliland, D. G. (2002a) FLT3 internal tandem duplication mutations associated with human acute myeloid leukemia induce myeloproliferative disease in murine bone marrow transplant model. Blood, 99, 310-18. Kelly, L. M. , Kutok, J. L., Williams, I. R., Boulton, C. L., Amaral, S. M. , Curley, D. P., Ley, T. J. and Gilliland, D. G. (2002b) PML/RARalpha and FLT3-ITD induces an APL-like disease in a mouse model. Proc Natl Acad Sci USA, 99, 8283-8. Keyomarsi, K. and Pardee, A.B. (1993) Redundant cyclin overexpression and gene amplification in breast cancer cells. Proc Natl Acad Sci USA, 90, 1112-6. Kim, J.Y., Lee, S., Hwangbo, B., Lee, C.T., Kim, Y.W., Han, S.K., Shim, Y.S. and Yoo, C.G. (2000) NF-kappaB activation is related to the resistance of lung cancer cells to TNF-alpha-induced apoptosis. Biochem Biophys Res Commun, 273, 140-6. Kiyoi, H., Naoe, T., Nakano, Y., Yokota, S., Minami, S., Miyawaki, S., Asou, N. , Kuriyama, K., Jinnai, I., Shimazaki, C., Akiyama, H., Saito, K., Oh, H., Motoji, T., Omoto, E., Saito, H., Ohno, R. and Ueda, R. (1999) Prognostic implication of FLT3 and N-RAS gene mutations in acute myeloid leukemia. Blood, 93, 3074-80. Kiyoi, H., Naoe, T., Yokota, S., Nakao, M. , Minami, S., Kuriyama, K., Takeshita, A., Saito, K., Hasegawa, S., Shimodaira, S., Tamura, J., Shimazaki, C., Matsue, K., Kobayashi, H., Arima, N. , Suzuki, R., Morishita, H., Saito, H., Ueda, R. and Ohno, R. (1997) Internal tandem duplication of FLT3 associated with leukocytosis in acute promyelocytic leukemia. Leukemia Study Group of the Ministry of Health and Welfare (Kohseisho). Leukemia, 11, 1447-52. Kiyoi, H., Towatari, M. , Yokota, S., Hamaguchi, M. , Ohno, R., Saito, H. and Naoe, T. (1998) Internal tandem duplication of the FLT3 gene is a novel modality of elongation mutation which causes constitutive activation of the product. Leukemia, 12, 1333-7. Klampfer, L., Zhang, J. and Nimer, S.D. (1999) GM-CSF rescues TF-1 cells from growth factor withdrawal-induced, but not differentiation-induced apoptosis: the role of BCL-2 and MCL-1. Cytokine, 11, 849-55. Konikova, E. and Kusenda, J. (2001) P53 protein expression in human leukemia and lymphoma cells. Neoplasma, 48, 290-8. Kornblau, S.M., Andreeff, M. , Hu, S.X., Xu, H.J., Patel, S., Theriault, A., Roller, C , Kantarjian, H., Estey, E., Deisseroth, A.B. and Benedict, W.F. (1998) Low and maximally phosphorylated levels of the retinoblastoma protein confer poor prognosis in newly diagnosed acute myelogenous leukemia: a prospective study. Clin Cancer Res, 4, 1955-63. Kuiper, G.G., Faber, P.W., van Rooij, H.C., van der Korput, J.A., Ris-Stalpers, C , Klaassen, P., Trapman, J. and Brinkmann, A.O. (1989) Structural organization of the human androgen receptor gene. J Mol Endocrinol, 2, Rl-4. Lacombe, F. and Belloc, F. (1996) Flow cytometry study of cell cycle, apoptosis and drug resistance in acute leukemia. Hematol Cell Ther, 38, 495-504. 143 Lacombe, F., Belloc, F., Dumain, P., Puntous, M. , Lopez, F., Bernard, P., Boisseau, M.R. and Reiffers, J. (1992) Quantitation of resistance of leukemic cells to cytosine arabinoside from BrdUrd/DNA bivariate histograms. Cytometry, 13, 730-8. Lapidot, T., Sirard, C , Vormoor, J., Murdoch, B., Hoang, T., Caceres-Cortes, J., Minden, M. , Paterson, B., Caligiuri, M.A. and Dick, J.E. (1994) A cell initiating human acute myeloid leukaemia after transplantation into SCLD mice. Nature, 367, 645-8. Larochelle, A., Vormoor, J., Hanenberg, H., Wang, J.C., Bhatia, M. , Lapidot, T., Moritz, T., Murdoch, B., Xiao, X.L., Kato, I., Williams, D.A. and Dick, J.E. (1996) Identification of primitive human hematopoietic cells capable of repopulating NOD/SCID mouse bone marrow: implications for gene therapy. Nat Med, 2, 1329-37. Leemhuis, T., Yoder, M.C., Grigsby, S., Aguero, B., Eder, P. and Srour, E.F. (1996) Isolation of primitive human bone marrow hematopoietic progenitor cells using Hoechst 33342 and Rhodamine 123. Exp Hematol, 24, 1215-24. Leith, CP., Kopecky, K.J., Godwin, J., McConnell, T., Slovak, M.L., Chen, I.M., Head, D.R., Appelbaum, F.R. and Willman, C.L. (1997) Acute myeloid leukemia in the elderly: assessment of multidrug resistance (MDR1) and cytogenetics distinguishes biologic subgroups with remarkably distinct responses to standard chemotherapy. A Southwest Oncology Group study. Blood, 89, 3323-9. Lemoine, F.M., Humphries, R.K., Abraham, S.D., Krystal, G. and Eaves, C J . (1988) Partial characterization of a novel stromal cell-derived pre-B-cell growth factor active on normal and immortalized pre-B cells. Exp Hematol, 16, 718-26. Lemoli, R.M., Tafuri, A., Fortuna, A., Catani, L., Rondelli, D., Ratta, M . and Tura, S. (1998) Biological characterization of CD34+ cells mobilized into peripheral blood. Bone Marrow Transplant, 22 Suppl 5, S47-50. Lemoli, R.M., Tafuri, A., Fortuna, A., Petrucci, M.T., Ricciardi, M.R., Catani, L., Rondelli, D., Fogli, M. , Leopardi, G., Ariola, C. and Tura, S. (1997) Cycling status of CD34+ cells mobilized into peripheral blood of healthy donors by recombinant human granulocyte colony-stimulating factor. Blood, 89, 1189-96. Lerner, C. and Harrison, D.E. (1990) 5-Fluorouracil spares hemopoietic stem cells responsible for long-term repopulation. Exp Hematol, 18, 114-8. Lichter, P., Joos, S., Bentz, M. and Lampel, S. (2000) Comparative genomic hybridization: uses and limitations. Semin Hematol, 37, 348-57. Lisovsky, M. , Estrov, Z., Zhang, X., Consoli, U., Sanchez-Williams, G., Snell, V., Munker, R., Goodacre, A., Savchenko, V. and Andreeff, M. (1996) Flt3 ligand stimulates proliferation and inhibits apoptosis of acute myeloid leukemia cells: regulation of Bcl-2 and Bax. Blood, 88, 3987-97. Liu, K.D., Gaffen, S.L. and Goldsmith, M.A. (1998) JAK/STAT signaling by cytokine receptors. Curr Opin Immunol, 10, 271-8. 144 Look, A.T. (1997) Oncogenic transcription factors in the human acute leukemias. Science, 278, 1059-64. Lotem, J. and Sachs, L. (1992) Hematopoietic cytokines inhibit apoptosis induced by transforming growth factor beta 1 and cancer chemotherapy compounds in myeloid leukemic cells. Blood, 80, 1750-7. Lowe, S.W., Schmitt, E.M., Smith, S.W., Osborne, B.A. and Jacks, T. (1993) p53 is required for radiation-induced apoptosis in mouse thymocytes. Nature, 362, 847-9. Lucotti, C , Malabarba, L., Rosti, V., Bergamaschi, G., Danova, M. , Invernizzi, R., Pecci, A., Ramajoli, I., Perotti, C., Torretta, L., De Amici, M., Salvaneschi, L. and Cazzola, M. (2000) Cell cycle distribution of cord blood-derived haematopoietic progenitor cells and their recruitment into the S-phase of the cell cycle. Br J Haematol, 108, 621-8. Lyman, S.D. (1998) Biologic effects and potential clinical applications of Flt3 ligand. Curr Opin Hematol, 5, 192-6. Lyman, S.D., James, L., Johnson, L., Brasel, K., de Vries, P., Escobar, S.S., Downey, H., Splett, R.R., Beckmann, M.P. and McKenna, H.J. (1994) Cloning of the human homologue of the murine flt3 ligand: a growth factor for early hematopoietic progenitor cells. Blood, 83, 2795-801. Lyon, M.F. (1961) Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature, 190, 372-373. Mach-Pascual, S., Legare, R.D., Lu, D., Kroon, M. , Neuberg, D., Tantravahi, R., Stone, R.M., Freedman, A.S., Nadler, L.M. , Gribben, J.G. and Gilliland, D.G. (1998) Predictive value of clonality assays in patients with non-Hodgkin's lymphoma undergoing autologous bone marrow transplant: a single institution study. Blood, 91, 4496-503. Mackarehtschian, K., Hardin, J.D., Moore, K.A., Boast, S., Goff, S.P. and Lemischka, LR. (1995) Targeted disruption of the flk2/flt3 gene leads to deficiencies in primitive hematopoietic progenitors. Immunity, 3, 147-61. Majka, M. , Janowska-Wieczorek, A., Ratajczak, J., Ehrenman, K., Pietrzkowski, Z., Kowalska, M.A., Gewirtz, A.M., Emerson, S.G. and Ratajczak, M.Z. (2001) Numerous growth factors, cytokines, and chemokines are secreted by human CD34(+) cells, myeloblasts, erythroblasts, and megakaryoblasts and regulate normal hematopoiesis in an autocrine/paracrine manner. Blood, 97, 3075-85. Martin, F.H., Suggs, S.V., Langley, K.E., Lu, H.S., Ting, J., Okino, K.H., Morris, C.F., McNiece, I.K., Jacobsen, F.W., Mendiaz, E.A. and et al. (1990) Primary structure and functional expression of rat and human stem cell factor DNAs. Cell, 63, 203-11. McCulloch, E.A. (1983) Stem cells in normal and leukemic hemopoiesis (Henry Stratton Lecture, 1982). Blood, 62, 1-13. 145 Meshinchi, S., Woods, W.G., Stirewalt, D.L., Sweetser, D.A., Buckley, J.D., Tjoa, T.K., Bernstein, I.D. and Radich, J.P. (2001) Prevalence and prognostic significance of Flt3 internal tandem duplication in pediatric acute myeloid leukemia. Blood, 97, 89-94. Metcalf, D. (1980) Clonal analysis of proliferation and differentiation of paired daughter cells: action of granulocyte-macrophage colony-stimulating factor on granulocyte-macrophage precursors. Proc Nad Acad Sci USA, 11, 5327-30. Miller, C.L., Rebel, V.I., Helgason, CD. , Lansdorp, P.M. and Eaves, C J . (1997) Impaired steel factor responsiveness differentially affects the detection and long-term maintenance of fetal liver hematopoietic stem cells in vivo. Blood, 89, 1214-23. Minden, M. , Major, P., Griffin, J., Wu, A. and Kufe, D. (1983) Generation time of leukaemic blast progenitor cells. Cell Tissue Kinet, 16, 577-82. Minden, M.D., Till, J.E. and McCulloch, E.A. (1978) Proliferative state of blast cell progenitors in acute myeloblasts leukemia (AML). Blood, 52, 592-600. Miyauchi, J., Kelleher, C.A., Wong, G.G., Yang, Y . C , Clark, S.C, Minkin, S., Minden, M.D. and McCulloch, E.A. (1988) The effects of combinations of the recombinant growth factors GM-CSF, G- CSF, IL-3, and CSF-1 on leukemic blast cells in suspension culture. Leukemia, 2, 382-7. Mizuki, M. , Fenski, R., Halfter, FL, Matsumura, L, Schmidt, R., Muller, C , Gruning, W., Kratz-Albers, K., Serve, S., Steur, C , Buchner, T., Kienast, J., Kanakura, Y., Berdel, W.E. and Serve, H. (2000) Flt3 mutations from patients with acute myeloid leukemia induce transformation of 32D cells mediated by the Ras and STAT5 pathways. Blood, 96, 3907-14. Muller, W. and Gautier, F. (1975) Interactions of heteroaromatic compounds with nucleic acids. A - T- specific non-intercalating DNA ligands. Eur J Biochem, 54, 385-94. Nakao, M. , Yokota, S., Iwai, T., Kaneko, H., Horiike, S., Kashima, K., Sonoda, Y., Fujimoto, T. and Misawa, S. (1996) Internal tandem duplication of the flt3 gene found in acute myeloid leukemia. Leukemia, 10, 1911-8. Nishinakamura, R., Miyajima, A., Mee, P.J., Tybulewicz, V.L. and Murray, R. (1996) Hematopoiesis in mice lacking the entire granulocyte-macrophage colony- stimulating factor/interleukin-3/interleukin-5 functions. Blood, 88, 2458-64. Nolta, J.A., Hanley, M.B. and Kohn, D.B. (1994) Sustained human hematopoiesis in immunodeficient mice by cotransplantation of marrow stroma expressing human interleukin-3: analysis of gene transduction of long-lived progenitors. Blood, 83, 3041-51. Ogawa, M. (1993) Differentiation and proliferation of hematopoietic stem cells. Blood, 81, 2844-53. Ogawa, S., Hangaishi, A., Miyawaki, S., Hirosawa, S., Miura, Y., Takeyama, K., Kamada, N., Ohtake, S., Uike, N., Shimazaki, C. and et al. (1995) Loss of the cyclin-dependent kinase 4-inhibitor (pi 6; MTS1) gene is frequent in and highly specific to lymphoid tumors in primary human hematopoietic malignancies. Blood, 86, 1548-56. 146 Oka, Y., Udaka, K., Tsuboi, A., Elisseeva, O.A., Ogawa, H., Aozasa, K., Kishimoto, T. and Sugiyama, H. (2000) Cancer immunotherapy targeting Wilms' tumor gene WT1 product. J Immunol, 164, 1873-80. Okuda, T., van Deursen, J., Hiebert, S.W., Grosveld, G. and Downing, J.R. (1996) AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell, 84, 321-30. Oster, W., Cicco, N.A., Klein, H., Hirano, T., Kishimoto, T., Lindemann, A., Mertelsmann, R.H. and Herrmann, F. (1989) Participation of the cytokines interleukin 6, tumor necrosis factor-alpha, and interleukin 1 -beta secreted by acute myelogenous leukemia blasts in autocrine and paracrine leukemia growth control. J Clin Invest, 84, 451-7. Otsuka, T., Thacker, J.D., Eaves, C.J. and Hogge, D.E. (1991) Differential effects of microenvironmentally presented interleukin 3 versus soluble growth factor on primitive human hematopoietic cells. J Clin Invest, 88, 417-22. Park, L.S., Waldron, P.E., Friend, D., Sassenfeld, H.M., Price, V., Anderson, D., Cosman, D., Andrews, R.G., Bernstein, I.D. and Urdal, D.L. (1989) Interleukin-3, GM-CSF, and G-CSF receptor expression on cell lines and primary leukemia cells: receptor heterogeneity and relationship to growth factor responsiveness. Blood, 74, 56-65. Pegoraro, E., Whitaker, J., Mowery-Rushton, P., Surti, U., Lanasa, M . and Hoffman, E.P. (1997) Familial skewed X inactivation: a molecular trait associated with high spontaneous-abortion rate maps to Xq28. Am J Hum Genet, 61, 160-70. Petzer, A.L., Eaves, C.J., Barnett, M.J. and Eaves, A.C. (1997) Selective expansion of primitive normal hematopoietic cells in cytokine- supplemented cultures of purified cells from patients with chronic myeloid leukemia. Blood, 90, 64-9. Pflumio, F., Izac, B., Katz, A., Shultz, L.D., Vainchenker, W. and Coulombel, L. (1996) Phenotype and function of human hematopoietic cells engrafting immune- deficient CB17-severe combined immunodeficiency mice and nonobese diabetic-severe combined immunodeficiency mice after transplantation of human cord blood mononuclear cells. Blood, 88, 3731-40. Piacibello, W., Fubini, L., Sanavio, F., Brizzi, M.F., Severino, A., Garetto, L., Stacchini, A., Pegoraro, L. and Aglietta, M. (1995) Effects of human FLT3 ligand on myeloid leukemia cell growth: heterogeneity in response and synergy with other hematopoietic growth factors. Blood, 86,4105-14. Piazza, F., Gurrieri, C. and Pandolfi, P.P. (2001) The theory of APL. Oncogene, 20, 7216-22. Pinkel, D., Straume, T. and Gray, J.W. (1986) Cytogenetic analysis using quantitative, high-sensitivity, fluorescence hybridization. Proc Natl Acad Sci USA, 83, 2934-8. Plenge, R.M., Hendrich, B.D., Schwartz, C , Arena, J.F., Naumova, A., Sapienza, C , Winter, R.M. and Willard, H.F. (1997) A promoter mutation in the XIST gene in two unrelated families with skewed X-chromosome inactivation. Nat Genet, 17, 353-6. 147 Ploemacher, R.E., van der Sluijs, J.P., van Beurden, C.A., Baert, M.R. and Chan, P.L. (1991) Use of limiting-dilution type long-term marrow cultures in frequency analysis of marrow-repopulating and spleen colony-forming hematopoietic stem cells in the mouse. Blood, 78, 2527-33. Ponchio, L., Conneally, E. and Eaves, C. (1995) Quantitation of the quiescent fraction of long-term culture-initiating cells in normal human blood and marrow and the kinetics of their growth factor-stimulated entry into S-phase in vitro. Blood, 86, 3314-21. Ponchio, L., Eaves, C , Hogge, D., Perotti, C , Torretta, L., Cerani, P., Salvaneschi, L., Ascari, E. and Cazzola, M . (1997) Both cycling and noncycling primitive progenitors continue to be mobilized into the circulation during the leukapheresis of patients pretreated with chemotherapy and G-CSF. Br J Haematol, 99, 394-402. Preisler, H.D., Bi, S., Venugopal, P. and Raza, A. (1997) Cytokines, molecular biological abnormalities, and acute myelogenous leukemia. LeukRes, 21, 299-312. Radich, J.P., Kopecky, K.J., Willman, C.L., Weick, J., Head, D., Appelbaum, F. and Collins, S.J. (1990) N-ras mutations in adult de novo acute myelogenous leukemia: prevalence and clinical significance. Blood, 76, 801-7. Radosevic, N. , Delmer, A., Tang, R., Marie, J.P. and Ajchenbaum-Cymbalista, F. (2001) Cell cycle regulatory protein expression in fresh acute myeloid leukemia cells and after drug exposure. Leukemia, 15, 559-66. Reems, J. A. and Torok-Storb, B. (1995) Cell cycle and functional differences between CD34+/CD38hi and CD34+/381o human marrow cells after in vitro cytokine exposure. Blood, 85, 1480-7. Reilly, I.A., Kozlowski, R. and Russell, N.H. (1989) Heterogeneous mechanisms of autocrine growth of A M L blasts. Br J Haematol, 72, 363-9. Rennick, D., Yang, G., Muller-Sieburg, C , Smith, C , Arai, N. , Takabe, Y. and Gemmell, L. (1987) Interleukin 4 (B-cell stimulatory factor 1) can enhance or antagonize the factor-dependent growth of hemopoietic progenitor cells. Proc Natl Acad Sci USA, 84, 6889-93. Roberts, A.W. and Metcalf, D. (1995) Noncycling state of peripheral blood progenitor cells mobilized by granulocyte colony-stimulating factor and other cytokines. Blood, 86, 1600-5. Rombouts, W.J., Blokland, I., Lowenberg, B. and Ploemacher, R.E. (2000) Biological characteristics and prognosis of adult acute myeloid leukemia with internal tandem duplications in the Flt3 gene. Leukemia, 14, 675-83. Rowe, J.M. (2001) Current standard therapy of adult acute myeloid leukemia, Acute Myeloid Leukemia. Hematology 2001, American Society of Hematology Education Program Book, 62-68. Rowe, J.M. and Liesveld, J.L. (1999) Hematopoietic growth factors and acute leukemia. Cancer Treat Res, 99, 195-226. 148 Russell, N.H. (1992) Autocrine growth factors and leukaemic haemopoiesis. Blood Rev, 6, 149-56. Russell, N.H., Hunter, A.E., Bradbury, D., Zhu, Y .M. and Keith, F. (1995) Biological features of leukaemic cells associated with autonomous growth and reduced survival in acute myeloblastic leukaemia. Leuk Lymphoma, 16, 223-9. Sasaki, K., Yagi, H., Bronson, R.T., Tominaga, K., Matsunashi, T., Deguchi, K., Tani, Y., Kishimoto, T. and Komori, T. (1996) Absence of fetal liver hematopoiesis in mice deficient in transcriptional coactivator core binding factor beta. Proc Natl Acad Sci USA, 93, 12359-63. Sauvageau, G., Lansdorp, P.M., Eaves, C.J., Hogge, D.E., Dragowska, W.H., Reid, D.S., Largman, C., Lawrence, H.J. and Humphries, R.K. (1994) Differential expression of homeobox genes in functionally distinct CD34+ subpopulations of human bone marrow cells. Proc Natl Acad Sci USA, 91, 12223-7. Schrock, E. and Padilla-Nash, H. (2000) Spectral karyotyping and multicolor fluorescence in situ hybridization reveal new tumor-specific chromosomal aberrations. Semin Hematol, 37, 334-47. Schumacher, H.R. (1990) Acute Leukemia, Approach to Diagnosis. Igaku-Shoin Ltd, New York, NY. ' Sensebe, L., Deschaseaux, M. , Li , J., Herve, P. and Charbord, P. (1997) The broad spectrum of cytokine gene expression by myoid cells from the human marrow microenvironment. Stem Cells, 15, 133-43. Shapiro, H.M. (1981) Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and pyronin Y. Cytometry, 2, 143-50. Sherr, C.J. (1996) Cancer cell cycles. Science, 274, 1672-7. Shibayama, H., Anzai, N. , Ritchie, A., Zhang, S., Mantel, C. and Broxmeyer, H.E. (1998) Interleukin-3 and Flt3-ligand induce adhesion of Baf3/Flt3 precursor B- lymphoid cells to fibronectin via activation of VLA-4 and VLA-5. Cell Immunol, 187, 27-33. Shultz, L.D., Schweitzer, P.A., Christianson, S.W., Gott, B., Schweitzer, I.B., Tennent, B., McKenna, S., Mobraaten, L., Rajan, T.V., Greiner, D.L. and et al. (1995) Multiple defects in innate and adaptive immunologic function in NOD/LtSz-scid mice. J Immunol, 154, 180-91. Sieff, C A . (1994) Hematopoietic cell proliferation and differentiation. Curr Opin Hematol, 1, 310-20. Siitonen, T., Lundstrom, M. , Zheng, A., Savolainen, E.R. and Koistinen, P. (1996) Effect of mast cell growth factor on clonogenic blast cell growth in acute myelogenous leukemia. Ann Hematol, 73, 71-8. Singer, J.W., Bernstein, I.D., Davis, P.C., Walters, T.R., Raskind, W.H. and Fialkow, P.J. (1988) The effect of long-term marrow culture on the origin of colony-forming cells in acute 149 myeloblastic leukemia: studies of two patients heterozygous for glucose-6-phosphate dehydrogenase. Leukemia, 2, 148-52. Sperr, W.R., Walchshofer, S., Horny, H.P., Fodinger, M. , Simonitsch, I., Fritsche-Polanz, R., Schwarzinger, I., Tschachler, E., Sillaber, C , Hagen, W., Geissler, K., Chott, A., Lechner, K. and Valent, P. (1998) Systemic mastocytosis associated with acute myeloid leukaemia: report of two cases and detection of the c-kit mutation Asp-816 to Val. Br J Haematol, 103, 740-9. Stanley, E., Lieschke, G.J., Grail, D., Metcalf, D., Hodgson, G., Gall, J.A., Maher, D.W., Cebon, J., Sinickas, V. and Dunn, A.R. (1994) Granulocyte/macrophage colony-stimulating factor-deficient mice show no major perturbation of hematopoiesis but develop a characteristic pulmonary pathology. Proc Natl Acad Sci U S A, 91, 5592-6. Stirewalt, D.L., Kopecky, K.J., Meshinchi, S., Appelbaum, F.R., Slovak, M.L., Willman, C.L. and Radich, J.P. (2001) FLT3, RAS, and TP53 mutations in elderly patients with acute myeloid leukemia. Blood, 97, 3589-95. Strom, D.K., Nip, J., Westendorf, J.J., Linggi, B., Lutterbach, B., Downing, J.R., Lenny, N. and Hiebert, S.W. (2000) Expression of the AML-1 oncogene shortens the G(l) phase of the cell cycle. J Biol Chem, 275, 3438-45. Su, X., Zhou, T., Wang, Z., Yang, P., Jope, R.S. and Mountz, J.D. (1995) Defective expression of hematopoietic cell protein tyrosine phosphatase (HCP) in lymphoid cells blocks Fas-mediated apoptosis. Immunity, 2, 353-62. Sugimoto, K., Toyoshima, H., Sakai, R., Miyagawa, K., Hagiwara, K., Hirai, H., Ishikawa, F. and Takaku, F. (1991) Mutations of the p53 gene in lymphoid leukemia. Blood, 77, 1153-6. Sutherland, H.J., Blair, A. and Zapf, R.W. (1996) Characterization of a hierarchy in human acute myeloid leukemia progenitor cells. Blood, 87, 4754-61. Sutherland, H.J., Eaves, C.J., Eaves, A . C , Dragowska, W. and Lansdorp, P.M. (1989) Characterization and partial purification of human marrow cells capable of initiating long-term hematopoiesis in vitro. Blood, 74, 1563-70. Sutherland, H.J., Hogge, D.E. and Eaves, C J . (1993) Growth factor regulation of the maintenance and differentiation of human long-term culture-initiating cells (LTC-IC). Leukemia, 7Suppl2, S122-5. Sutherland, H.J., Lansdorp, P.M., Henkelman, D.H., Eaves, A.C. and Eaves, C J . (1990) Functional characterization of individual human hematopoietic stem cells cultured at limiting dilution on supportive marrow stromal layers. Proc Natl Acad Sci USA, 87, 3584-8. Terpstra, W., Ploemacher, R.E., Prins, A., van Lorn, K., Pouwels, K., Wognum, A.W., Wagemaker, G., Lowenberg, B. and Wielenga, J.J. (1996) Fluorouracil selectively spares acute myeloid leukemia cells with long- term growth abilities in immunodeficient mice and in culture. Blood, 88, 1944-50. 150 Terstappen, L.W., Huang, S., Safford, M. , Lansdorp, P.M. and Loken, M.R. (1991) Sequential generations of hematopoietic colonies derived from single nonlineage-committed CD34+CD38-progenitor cells. Blood, 11, 1218-27. Tilley, W.D., Marcelli, M. , Wilson, J.D. and McPhaul, M J . (1989) Characterization and expression of a cDNA encoding the human androgen receptor. Proc Natl Acad Sci USA, 86, 327-31. Towatari, M. , Iida, H., Tanimoto, M. , Iwata, H., Hamaguchi, M . and Saito, H. (1997) Constitutive activation of mitogen-activated protein kinase pathway in acute leukemia cells. Leukemia, 11, 479-84. Traver, D., Akashi, K., Weissman, I.L. and Lagasse, E. (1998) Mice defective in two apoptosis pathways in the myeloid lineage develop acute myeloblastic leukemia. Immunity, 9, 47-57. Tsui, F.W. and Tsui, H.W. (1994) Molecular basis of the motheaten phenotype. Immunol Rev, 138, 185-206. Uchida, N., He, D., Friera, A.M., Reitsma, M., Sasaki, D., Chen, B. and Tsukamoto, A. (1997) The unexpected G0/G1 cell cycle status of mobilized hematopoietic stem cells from peripheral blood. Blood, 89, 465-72. Uchida, T., Ohashi, H., Aoki, E., Nakahara, Y., Hotta, T., Murate, T., Saito, H. and Kinoshita, T. (2000) Clonality analysis by methylation-specific PCR for the human androgen- receptor gene (HUMARA-MSP). Leukemia, 14, 207-12. Van Der Lely, N. , Poddighe, P., Wessels, J., Hopman, A., Geurts van Kessel, A. and De Witte, T. (1995) Clonal analysis of progenitor cells by interphase cytogenetics in patients with acute myeloid leukemia and myelodysplasia. Leukemia, 9, 1167-72. van der Loo, J.C., Hanenberg, H., Cooper, R.J., Luo, F.Y., Lazaridis, E.N. and Williams, D.A. (1998) Nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mouse as a model system to study the engraftment and mobilization of human peripheral blood stem cells. Blood, 92, 2556-70. Vellenga, E., Ostapovicz, D., O'Rourke, B. and Griffin, J.D. (1987) Effects of recombinant IL-3, GM-CSF, and G-CSF on proliferation of leukemic clonogenic cells in short-term and long-term cultures. Leukemia, 1, 584-9. Verlinden, S.F., van Es, H.H. and van Bekkum, D.W. (1998) Serial bone marrow sampling for long-term follow up of human hematopoiesis in NOD/SCID mice. Exp Hematol, 26, 627-30. Vogelstein, B., Fearon, E.R., Hamilton, S.R., Preisinger, A . C , Willard, H.F., Michelson, A.M., Riggs, A.D. and Orkin, S.H. (1987) Clonal analysis using recombinant DNA probes from the X-chromosome. Cancer Res, 47, 4806-13. Wagteveld, A.J., Esselink, M.T., Limburg, P., Halie, M.R. and Vellenga, E. (1992) The effects of IL-1 beta and IL-4 on the proliferation and endogenous secretion of growth factors by acute myeloblastic leukemic cells. Leukemia, 6, 1020-4. 151 Wang, C.Y., Mayo, M.W. and Baldwin, A.S., Jr. (1996a) TNF- and cancer therapy-induced apoptosis: potentiation by inhibition of NF-kappaB. Science, 274, 784-7. Wang, C.Y., Mayo, M.W., Korneluk, R.G., Goeddel, D.V. and Baldwin, A.S., Jr. (1998a) NF-kappaB antiapoptosis: induction of TRAF1 and TRAF2 and c-IAPl and c- IAP2 to suppress caspase-8 activation. Science, 281, 1680-3. Wang, Q., Stacy, T., Binder, M., Marin-Padilla, M. , Sharpe, A.H. and Speck, N.A. (1996b) Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous system and blocks definitive hematopoiesis. Proc Natl Acad Sci USA, 93, 3444-9. Wang, Q., Stacy, T., Miller, J.D., Lewis, A.F., Gu, T.L., Huang, X., Bushweller, J.H., Bories, J.C., Alt, F.W., Ryan, G., Liu, P.P., Wynshaw-Boris, A., Binder, M. , Marin-Padilla, M. , Sharpe, A.H. and Speck, N.A. (1996c) The CBFbeta subunit is essential for CBFalpha2 (AML1) function in vivo. Cell, 87, 697-708. Wang, Z.G., Delva, L., Gaboli, M. , Rivi, R., Giorgio, M. , Cordon-Cardo, C., Grosveld, F. and Pandolfi, P.P. (1998b) Role of PML in cell growth and the retinoic acid pathway. Science, 279, 1547-51. Wattel, E., Preudhomme, C., Hecquet, B., Vanrumbeke, M. , Quesnel, B., Dervite, I., Morel, P. and Fenaux, P. (1994) p53 mutations are associated with resistance to chemotherapy and short survival in hematologic malignancies. Blood, 84, 3148-57. Wei, M.C., Zong, W.X., Cheng, E.H., Lindsten, T., Panoutsakopoulou, V., Ross, A.J., Roth, K.A., MacGregor, G.R., Thompson, C.B. and Korsmeyer, S.J. (2001) Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science, 292, 727-30. Wheatley, D. (1982) Cell Growth and Division. E. Arnold Ltd., London. Wilson, V.L. and Jones, P.A. (1983) DNA methylation decreases in aging but not in immortal cells. Science, 220, 1055-7. Wu, CD. , Wickert, R.S., Williamson, J.E., Sun, N . C , Brynes, R.K. and Chan, W.C. (1999) Using fluorescence-based human androgen receptor gene assay to analyze the clonality of microdissected dendritic cell tumors. Am J Clin Pathol, 111, 105-10. Xia, Z., Baer, M.R., Block, A.W., Baumann, H. and Wetzler, M . (1998) Expression of signal transducers and activators of transcription proteins in acute myeloid leukemia blasts. Cancer Res, 58,3173-80. Yamamoto, Y., Kiyoi, H., Nakano, Y., Suzuki, R., Kodera, Y., Miyawaki, S., Asou, N., Kuriyama, K., Yagasaki, F., Shimazaki, C , Akiyama, H., Saito, K., Nishimura, M. , Motoji, T., Shinagawa, K., Takeshita, A., Saito, H., Ueda, R., Ohno, R. and Naoe, T. (2001) Activating mutation of D835 within the activation loop of FLT3 in human hematologic malignancies. Blood, 97, 2434-9. 152 Yergeau, D.A., Hetherington, C.J., Wang, Q., Zhang, P., Sharpe, A.H., Binder, M. , Marin-Padilla, M. , Tenen, D.G., Speck, N.A. and Zhang, D.E. (1997) Embryonic lethality and impairment of haematopoiesis in mice heterozygous for an AML1-ETO fusion gene. Nat Genet, 15, 303-6. Y i , T. and Ihle, J.N. (1993) Association of hematopoietic cell phosphatase with c-Kit after stimulation with c-Kit ligand. Mol Cell Biol, 13, 3350-8. Yokozawa, T., Towatari, M. , Iida, H., Takeyama, K., Tanimoto, M. , Kiyoi, H., Motoji, T., Asou, N., Saito, K., Takeuchi, M. , Kobayashi, Y., Miyawaki, S., Kodera, Y., Ohno, R., Saito, H. and Naoe, T. (2000) Prognostic significance of the cell cycle inhibitor p27Kipl in acute myeloid leukemia. Leukemia, 14, 28-33. Yonemura, Y., Ku, H., Hirayama, F., Souza, L.M. and Ogawa, M . (1996) Interleukin 3 or interleukin 1 abrogates the reconstituting ability of hematopoietic stem cells. Proc Natl Acad Sci U S ,4,93,4040-4. Young, D.C, Demetri, G.D., Ernst, T.J., Cannistra, S.A. and Griffin, J.D. (1988) In vitro expression of colony-stimulating factor genes by human acute myeloblasts leukemia cells. Exp Hematol, 16, 378-82. Young, D.C. and Griffin, J.D. (1986) Autocrine secretion of GM-CSF in acute myeloblasts leukemia. Blood, 68, 1178-81. Yunis, J.J., Bloomfield, C D . and Ensrud, K. (1981) Al l patients with acute nonlymphocytic leukemia may have a chromosomal defect. N Engl J Med, 305, 135-9. Zandstra, P.W., Conneally, E., Petzer, A.L., Piret, J.M. and Eaves, C J . (1997) Cytokine manipulation of primitive human hematopoietic cell self- renewal. Proc Natl Acad Sci USA, 94, 4698-703. Zheng, A., Castren, K., Saily, M. , Savolainen, E.R., Koistinen, P. and Vahakangas, K. (1999) p53 status of newly established acute myeloid leukaemia cell lines. Br J Cancer, 79, 407-15. Zheng, R., Baldwin, B. R., Levis, M. , FaiTse, K., Malehorn, M.T., Gorin, N . C , Small, D. (2001) Co-expression of the FLT3 receptor and its ligand FL, by primary A M L cells. Blood, 98, 90a. Zhu, Y.M. , Bradbury, D. and Russell, N. (1993) Expression of different conformations of p53 in the blast cells of acute myeloblasts leukaemia is related to in vitro growth characteristics. Br J Cancer, 68, 851-5. 153