UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of CD44 in the regulation of hematopoiesis Ghaffari, Saghi 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1996-147487.pdf [ 9.24MB ]
JSON: 831-1.0087664.json
JSON-LD: 831-1.0087664-ld.json
RDF/XML (Pretty): 831-1.0087664-rdf.xml
RDF/JSON: 831-1.0087664-rdf.json
Turtle: 831-1.0087664-turtle.txt
N-Triples: 831-1.0087664-rdf-ntriples.txt
Original Record: 831-1.0087664-source.json
Full Text

Full Text

T H E R O L E O F CD44 IN THE REGULATION OF HEMATOPOIESIS by SAG HI GHAFFARI A THESIS SUBMITTED IN PARTIAL FULFILLMENT O F THE REQUIREMENTS 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 June 1996 © Saghi Ghaffari, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it 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. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. ' Department of *?p$kr* ^ ^ V M -The University of British Columbia Vancouver, Canada Date I ^ u , V j DE-6 (2/88) Abstract The daily production of the large number of circulating blood cells is the result of a controlled process called hematopoiesis. In humans, adult hematopoiesis takes place in the bone marrow microenvironment where the regulated production and presentation of both positive and negative-act ing factors control hematopoiet ic cell proliferation, differentiation and survival. Many cell surface molecules, including growth factor receptors and adhesion molecules are involved in these processes, although their relative importance remains to be understood. In the present thesis, I have focused on an examinat ion of the potential role of the CD44 family of adhesion molecules in the regulation of primitive hematopoietic cell proliferation and differentiation. CD44 is a widely expressed multifunctional cell surface glycoprotein which has been implicated in many different adhesion-dependent processes. In addit ion, expression of particular isoforms of CD44 has been associated with malignant transformation and/or the acquisit ion of metastatic potential. The expression of CD44 isoforms was examined on primitive normal and myeloid leukemic hematopoiet ic progenitor cells. These studies have shown that the differentiat ion-associated changes of CD44 isoform expression is altered during leukemogenesis. In addit ion, monoclonal antibodies to distinct CD44 epitopes were found to have either inhibitory, enhancing, or no effect on normal , but not on CML, stromal-dependent hematopoiesis. These results provide evidence of early differentiation-associated changes in CD44 expression during normal hematopoiesis which may be deregulated in chronic phase CML as well as in a variety of other myeloid leukemias. They also identify CD44 as a putative regulatory component in the interactions of hematopoiet ic cells with st roma, and suggest that this function is absent in primitive CML hematopoiet ic cells. Future investigations of molecular mechanisms involved in the regulation of CD44 l igand-binding and its i i downst ream effects should provide insights into the biological observat ions descr ibed at the cellular level in this thesis. i i i TABLE OF CONTENTS Page A B S T R A C T ii TABLE O F CONTENTS iv LIST OF TABLES vii LIST OF FIGURES ix LIST OF ABBREVIATIONS x A C K N O W L E D G M E N T S xi C H A P T E R 1 INTRODUCTION 1.1 Regulation of Hematopoiesis 1 1.1.1 Hierarchical Organization of The System 2 1.1.2 Mechanisms of Regulation 6 1.1.3 Biology of Myeloid Leukemias 8 Chronic Myeloid Leukemia 8 Acute Myeloid Leukemia 11 1.2 Bone Marrow Microenvironment 12 1.2.1 Cellular Components 12 1.2.2 Growth Factors and Inhibitors 14 1.2.3 Extracellular Matrix 17 Fibronectin 18 Thrombospondin 19 Collagen and Laminin 20 Tenascin 20 Glycosaminoglycans 21 1.2.4 Adhesion Molecules 25 Integrins 26 Selectins 28 Immunoglobul in Supergene Family 29 c-Kit and Steel Factor 30 Other Cell Surface Receptors 30 1.2.5 Long-term Marrow Cultures as a Model 32 1.2.6 Abnormal Interactions in Leukemia 33 1.3 CD44 36 1.3.1 Structure and Regulation of Expression 37 C D 4 4 G e n e 37 CD44 Protein 40 Membrane-Bound CD44 40 Soluble CD44 42 Cell Surface Distribution 42 iv 1.3.2 CD44 Ligands and Interactions with Extracellular Matrix 44 Hyaluronan 44 Other CD44 Ligands 45 Regulation of Expression and Ligand-Binding 47 1.3.3 Cellular Functions 51 Lymphocytes 51 T Lymphocytes 52 B Lymphocytes 54 Natural Killer Cells 55 Monocytes 55 Primitive Hematopoietic Cells 55 1.3.4 Expression and Role of CD44 Variants 57 Associat ion of CD44H With Mal ignancy 58 CD44 Variants in Malignancy 59 CD44 Variants and Normal Cells 61 1.3.5 Summary 62 1.4 Thesis Objectives 64 C H A P T E R 2 MATERIALS AND METHODS 2.1 Cells 2.1.1 Normal Bone Marrow 67 2.1.2 Mobil ized Peripheral Blood 67 2.1.3 A M L and CML Samples 67 2.1.4 Cell Lines 68 2.2 Antibodies 69 2.3 Hyaluronan-FITC 2.3.1 Preparation 70 2.3.2 HA-FITC Binding Assay and Protein Kinase C 70 2.4 Staining and Flow Cytometry 72 2.5 RT-PCR analysis 73 2.6 Hematopoiet ic Progenitor Cultures and Assays 2.6.1 Colony-Forming Cell Assays 74 2.6.2 Initiation and Maintenance of Human Marrow Adherent Layers 74 2.6.3 Long-Term Culture Initiating-Cell Assay 75 2.6.4 Ant ibody Treatment in vitro 76 V C H A P T E R 3 DIFFERENTIATION-ASSOCIATED C H A N G E S IN CD44 ISOFORM EXPRESSION DURING N O R M A L HEMATOPOIESIS AND THEIR ALTERATION IN CHRONIC MYELOID LEUKEMIA 3.1 Introduction 7 7 3.2 Results 3.2.1 Altered CD44 Expression on Phenotypically Defined Populations of Primitive CML Cells 79 3.2.2 Differential Expression of CD44 on Different Normal Progenitor Populations 84 3.2.3 Expression of CD44 on CML Progenitors is Altered 86 3.3 Discussion 91 C H A P T E R 4 ALTERED PATTERNS OF CD44 EPITOPE EXPRESSION IN HUMAN CHRONIC AND A C U T E MYELOID LEUKEMIA 4.1 Introduction 97 4.2 Results 4.2.1 Altered Patterns of CD44 Epitope Expression on Primary Human Leukemic Cells 98 4.2.2 Variable Expression of CD44 Epitopes on Human Myeloid Leukemia Cell Lines 103 4.2.3 CD44 Isoform Expression and HA-Binding Following TPA Stimulation 103 4.3 Discussion 106 C H A P T E R 5 STROMAL-DEPENDENT HEMATOPOIESIS IN C U L T U R E S O F CHRONIC MYELOID LEUKEMIA CELLS IS UNAFFECTED BY ANTI-CD44 ANTIBODIES THAT INHIBIT O R STIMULATE NORMAL HEMATOPOIETIC CELLS 5.1 Introduction 111 5.2 Results 5.2.1 Characterization of the Reactivity of Two Novel Epitope-Specific Ant i -CD44 mAbs (7f4 and 8d8) with Normal and Leukemic (CML) Hematopoietic Cells 112 v i 5.2.2 Opposite Effects of Epitope-Specif ic Ant i -CD44 Antibodies on the Production and Maintenance of Normal Bone Marrow Clonogenic Cells and LTC-IC i n L T C 113 5.2.3 Lack of Ant i -CD44 Epitope-Specif ic Effects on Primitive Neoplastic Hematopoietic Progenitors f rom Chronic Myeloid Leukemic Patients in LTC 118 5.3 Discussion 121 C H A P T E R 6 S U M M A R Y A N D PERSPECTIVES 125 REFER EN C ES 135 v i i L is t o f T a b l e s Page Table 1 - Compar isons of the distribution of total CD34+ cells and the total l ight-density cells by their levels of expression of CD44 and Thy-1 in normal BM versus CML PB 83 Table 2 - Progenitor distributions in subpopulat ions of CD34+ cells in normal BM by their levels of expression of CD44 and Thy-1 87 Table 3 - Progenitor distributions in subpopulat ions of CD34+ C M L PB cells by their levels of expression of CD44 and Thy-1 89 Table 4 - Progenitor distributions in circulating subpopulat ions of chemotherapy and growth factor mobil ized CD34+ cells 92 Table 5 - Initial W B C and clonogenic progenitor concentrat ions in fresh or cryopreserved PB samples from patients with CML 93 Table 6 - Features of the leukemic patients and their samples used in these studies 99 Table 7 - Distribution of CFC between adherent and non adherent fractions of LTC 117 v i i i L is t o f F igu res Page Figure 1 - Hierarchy of hematopoietic cells and their clonogenic assays 3 Figure 2 - Bone marrow microenvironment 13 Figure 3 - Structure of the CD44 gene and the corresponding protein 38 Figure 4 - Binding of HA-FITC to CD44R1 transfected C O S cells 71 Figure 5 - Gates used for analysis of antibody reactivity with normal BM cells 81 Figure 6 - Expression of CD44 versus Thy-1 on CML PB CD34+ cells 82 Figure 7 - RT-PCR analysis of 2G1+ cells 85 Figure 8 - Distribution of hematopoietic progenitors according to their level of CD44 expression 88 Figure 9 - Progenitor distribution in CML PB according to CD44 expression 90 Figure 10 - Scatter plot of CD44 epitopes on leukemic cells 100 Figure 11 - Histograms of CD44 epitopes on leukemic cells 102 Figure 12 - Effect of TPA on CD44 expression and HA-FITC binding of human myeloid cell lines 104 Figure 13 - Opposing effects of ant i -CD44 monoclonal antibodies on hematopoiesis in 5 week long term culture 114 Figure 14 - Ant i -CD44 monoclonal antibodies affect hematopoiesis in 1 week long-term culture 116 Figure 15 - Lack of effect of ant i -CD44 monoclonal antibodies on hematopoiesis in liquid culture 119 Figure 16 - Absence of effect of ant i -CD44 monoclonal antibodies on C M L hematopoiesis in long-term culture 120 Figure 17 - Effects of ant i -CD44 monoclonal antibodies on hematopoiesis in vitro 128 Figure 18 - Mitogenic versus cytoskeletal pathways 131 Figure 19 - Cytoskeletal alterations in BCR-ABL transformed cells 133 ix Lis t o f A b b r e v i a t i o n s ABL: Abelson A M L : Acute Myeloid Leukemia ALL: Acute Lymphoid Leukemia BCR: Breakpoint Cluster Region B M : Bone Marrow CD: Cluster of Differentiation CD44v: CD44 Variant (carrying addit ional alternatively spliced exons) CFU-S: Colony Forming Unit-Spleen CFC: Colony-Forming-Cel l CML: Chronic Myeloid Leukemia C R U : Competi t ive Repopulat ing Unit C S F - 1 : Macrophage-Colony-Stimulating-Factor CTL: Cytotoxic T Lymphocyte D M E M : Iscove's Modif ied Dulbecco's Medium ECM: Extracellular Matrix EGF: Epidermal Growth Factor EPO: Erythropoietin FITC: Fluorescein Isothiocyanate FA: Focal Adhesion FACS: Fluorescence Activating Cell Sort ing FAK: Focal Adhesion Kinase FCS: Fetal Calf Serum b-FGF: Basic-Fibroblast Growth Factor FGF: Fibroblast Growth Factor FSC: Forward Scatter G A G : Glycosaminoglycan G-CSF: Granulocyt ic-Colony Stimulating Factor GM-CSF: Granulocytic Macrophage-Colony Stimulating Factor GTP: Guanine Nucleotide Tr iphosphate G V H : Graft Versus Host GY: Gray HA: Hyaluronan or Hyaluronic Acid HBS: Hyaluronan Binding Site HBSS: Hank's Balanced Salt Solution HEV: High Endothel ial Venule HFN: Hanks' , FCS, N a N 3 ICAM-1 : Intercellular Adhesion Molecule-1 Ig: Immunoglobul in IL: Interleukin IFN-y: Interferon-y KD: Kilodalton L F A - 1 : Leukocyte Funct ion Antigen-1 LIF: Leukemia Inhibitory Factor LTC: Long Term Culture LTC-IC: Long Term Culture-Initiating Cell mAb: Monoclonal Ant ibody MIP-1a: Macrophage Inf lammatory Protein-1oc PB: Peripheral Blood PBS: Phosphate Buffered Saline PCR: Polymerase Chain Reaction PE: Phycoerythrin PI: Propidium Iodide PKC: Protein Kinase C PMA: Phorbol Myristate Acetate RPMI: Roswell Park Memorial Institute RT-PCR: Reverse Transcr iptase-Polymerase Chain Reaction SDS: Sodium Dodecyl Sulfate S E M : Standard Error of the Mean SF: Steel Factor S I : Steel SSC: Side Scatter TGF-(3: Transforming Growth Factor-p TNFoc: Tumor Necrosis Factor-oc TPA: tetradecanoyl-phorbol-13-acetate T P O : Thrombopoiet in VLA: Very Late Ant igen W B C : White Blood Cell A C K N O W L E D G M E N T S I would like to thank all of the people who made this thesis possible. All of those who participated in thinking and discussing, those who gave me their t ime, those who offered me their valuable technical help, those who supported and encouraged me during the difficult t imes and those who shared with me moments of laughter. It has been a unique, unforgettable experience to work in the Terry Fox Laboratory and to know Canadians who are the most welcoming people that I have met since having left Iran. XI Chapter 1 Introduction 1.1 R e g u l a t i o n of H e m a t o p o i e s i s Hematopoiesis is the process by which mature blood cells of distinct l ineages are produced from pluripotent hematopoietic stem cells. In mammal ian development, the first known sites of hematopoiesis are the yolk sac blood islands. Recently, the almost simultaneous appearance of hematopoiet ic cells within the embryo itself (in the so-called aort ic-gonadal-mesonephros, or A G M , region) has also been described. Hematopoiesis then shifts to the fetal liver and spleen and finally to the bone marrow where hematopoiesis normally cont inues throughout adult life (Zon, 1995). The final products of this process, the mature blood cells, carry out highly specialized and vital functions. These include the transport of oxygen and carbon dioxide by erythrocytes, blood clotting by platelets, innate immunity mediated by macrophages, granulocytes and natural killer cells, and antigen-specif ic immune responses by B and T lymphocytes. These cells circulate in large numbers and because of their relatively short life-spans (a few days to weeks) , they need to be constantly replaced. This, in turn, requires the coordinated differentiation and extensive amplif ication of the multiple l ineages of blood cell types from the small stem cell compartment on a daily basis in addit ion to the activation of cells with immune and/or inf lammatory functions (Orkin, 1995). Hematopoiesis is thus a highly complex and dynamic process both during development and later, with many factors contributing to its control. The present section will review various aspects of hematopoietic cell regulation and their alterations in human leukemia in addition to the organization of the hematopoiet ic system. l 1.1.1 H ie ra rch ica l O r g a n i z a t i o n of the S y s t e m The hematopoiet ic system has historically been subdivided into myeloid and lymphoid compartments. All of the myeloid cells are produced in the bone marrow. These include cells of the erythroid, granulocytic and megakaryocyt ic l ineages and their progeny (i.e. erythrocytes, granulocytes, macrophages, and platelets). The lymphoid compartment gives rise to T and B lymphocytes and NK cells which are produced and/or found to various extents in the spleen, thymus, bone marrow, and lymph nodes. The differentiation of all these different cell types is a multistep process usually spanning many cell divisions. Each lineage is therefore visual ized as being composed of a series of overlapping but distinct stages which differ in terms of their potential to proliferate, their responsiveness to different regulators, and their phenotype. The most primitive compartment (referred to as the stem cell compartment) is composed of pluripotent cells that are individually capable of generat ing all of the hematopoietic cell l ineages and also of sustaining their production for extensive periods of t ime through their capacity to execute self-renewal divisions. When these cells become restricted to a particular l ineage they also become more restricted in their proliferation capacity. These cells are commonly referred to as hematopoietic progenitors although they retain many characterist ics possessed by multipotent cells. For example, they both express several of the same surface antigens such as the CD34 molecule and by light microscopy are not morphologically distinguishable. Their recognition is, however, possible by their abilities to generate specific types of colonies in vitro under appropriate condit ions. Lineage-restricted progenitors also preferentially express distinct antigens that may be used for their differential purif ication. When these cells divide, they differentiate into the morphological ly recognizable cells that are the immediate precursors of specific types of mature blood cells (Figure 1). 2 Figure 1 . H ie ra rchy of H e m a t o p o i e t i c Cel ls a n d the i r C l o n o g e n i c A s s a y s . Cells are shown in terms of their CD34 antigenic profile and the assays used to detect them. 3 Over the years, several in vitro and in vivo hematopoiet ic assays have been developed which have al lowed cells with distinct hematopoiet ic properties to be separated and examined (Sutherland et al. , 1991). Different types of l ineage-restricted and some multipotent hematopoietic progenitors are identified by their ability to form colonies (of at least 20 to 50 cells) in a semi-sol id medium (usually containing either methylcellulose or soft agar) supplemented with appropriate nutrients and specific growth factors. After a defined period of growth, which is characteristic for each progenitor and reflects its proliferative capacity under condit ions of optimal stimulation, each colony will be composed of terminally differentiating cells of recognizable morphology. Thus when , for example, a heterogeneous population of cells such as those present in a sample of bone marrow is placed in a colony assay, colonies of different size and containing different cell types appear at different t imes after plating. The longer the period of t ime after plating before the cells in the colony become morphological ly recognizable, the more primitive the cell f rom which the colony has ar isen. The number of cells contained within the colony when it has achieved such maturity provides a measure of its proliferative potential and the type of cells detected in the colony indicates its differentiation potential. Progenitors that give rise to colonies of granulocytes and macrophages are called granulocyte-macrophage colony-forming units or C F U - G M . These C F U - G M include cells with a broad range of proliferative capacit ies. In addition CFU-GM further differentiate to become either CFU-G or CFU-M. Burst-forming units-erythroid (BFU-E) represent a primitive type of erythroid restricted progenitors. They produce multifocal colonies (bursts) containing many clusters of hemoglobinized red cells. The multifocal morphology of these colonies facilitates estimating their size and BFU-E have been subdivided into primitive and mature BFU-E according to whether they produce more or less than 8 erythroblast clusters. Each such cluster is produced by a progenitor called a 4 colony-forming unit-erythroid (CFU-E) which forms single or double clusters of 8-64 hemoglobinized red cells. Megakaryocyte progenitors have also been classif ied into megakaryocyte burst-forming units (BFU-Mk) and megakaryocyte colony-forming units (CFU-Mk) using similar principles. Colony-forming cells that produce mixtures of these lineages under the same culture condit ions also exist and are referred to as colony-forming unit-granulocyte, erythrocyte, macrophage, megakaryocyte (CFU-GEMM). Although they are multipotent, most C F U - G E M M lack significant self-renewal ability and are separable f rom stem cells (Sutherland et al . , 1991). Recent studies suggest that the latter may not form colonies in semi-sol id cultures either because they cannot divide under such condit ions or because the growth factors they require have not been identif ied. However, it is now believed that at least some stem cells can be maintained and st imulated to proliferate in a liquid culture system referred to as long-term cultures (LTC). Hematopoiesis in this system is establ ished in the presence of an adherent layer of stromal cells. These stromal cells can support the generation of clonogenic hematopoiet ic cells for long periods of t ime (at least two months) during which the hematopoiet ic cells are in constant or intermittent turn over (Eaves et al. , 1993). This f inding led to the suggest ion that the initial cell activated must have included a very primitive hematopoiet ic cell which was then responsible for the subsequent sustained output of clonogenic progenitors detectable after being plated in secondary colony assays. The original very primitive cell thus identified has been named a LTC-init iating cell or LTC-IC. LTC-IC can be quanti tated by limiting dilution assays or, based on knowledge of the average number of c lonogenic cells produced per LTC-IC after 5 weeks by simply dividing the total number of clonogenic cells produced by a test input innoculum by that number (Sutherland et al . , 1989, 1990). 5 Several lines of evidence indicate that LTC-IC share many characterist ics with murine competit ive long-term in vivo repopulating cells (referred to as CRU) . In mice, C R U are considered to represent stem cells since they are capable of repopulat ing both the lymphoid and myeloid hematopoiet ic compartments of irradiated recipients for long periods of t ime (at least 5 weeks in mice) (Szilvassy et al. , 1989a; 1989b). Other hematopoietic cells that appear to be intermediate between C R U , LTC-IC and progenitors with in vitro colony-forming capacity have also been character ized. One of these is the spleen colony-forming unit (CFU-S), a cell that is capable of making colonies of up to 10 7 mature cells in the spleen of an irradiated mouse after either 8-9 days (CFU-S day 8-9) or 12 days (CFU-S day 12). Blast-colony-forming cells, another type of cell in this category, are identified by their failure to differentiate until after they have produced colonies of 20 to a few hundred undifferentiated blast cells which may require up to 21 days of incubation in semi-solid medium even when all of the growth factors needed for their future differentiation are already present. The term "primitive hematopoiet ic cells" is often used to encompass all of the cells descr ibed above. However, it is important to appreciate the fact that many of these cells can be biologically distinguished from one another. 1.1.2 M e c h a n i s m s of Regu la t i on Sustained hematopoiesis is required through an individual's lifetime to cont inuously replace the mature blood cells because of their short half-life. This requires the maintenance of a sufficient pool of pluripotent hematopoiet ic s tem cells by balancing the production of new stem cells with the production of progenitors with more restricted developmental potentials. Hematopoiet ic s tem cells possess the ability to migrate both out of and into sites within the organism that will support these processes. In addit ion, hematopoiet ic s tem cell behaviour is regulated by a variety of specific genes that control their internal funct ions. 6 Hematopoiesis is thus controlled by the combined effects of extracellular signals emanat ing from the environment in which the stem cell is located and by molecules contained within the nucleus of the cell that then activate l ineage-specific genes. Although it is not clear how a multipotential cell chooses a single pathway of differentiation, a body of data suggests that extracellular signals (reviewed in 1.2) largely regulate the cell cycle status and progression of hematopoiet ic progenitors, whereas their choice of differentiation pathway may be predominant ly controlled by transcription factors (Orkin, 1995). The loss of individual transcription factors in genetically manipulated mice has been found to result in various l ineage-specific hematopoiet ic defects. Animals deficient for the erythroid specific factor GATA-1 were character ized by the absence of primitive and definitive erythroid development (Pevny et al. , 1995). Similarly, the absence of the Ikaros gene was found to specifically result in a lack of T, B, and NK cell development (Georgopolous et al . , 1994). Mice missing PU.1 gene lack cells of disparate l ineages, including both myeloid and lymphoid cells (Scott et al., 1994). Interestingly, the absence of some more broadly expressed transcription factors were found to result in l ineage-selective deficits. E2A knockout mice exhibit a specific absence of B-lymphopoiesis at an early stage, whereas the development of other t issues in which E2A is normally expressed is unaltered (Bain et al., 1994). Similarly, the loss of tal-1/SCL, a factor also expressed in several l ineages, results in the selective failure of erythropoiesis at the yolk sac stage (Warren et al., 1994). Activation of a pathway may result in the selective suppression of another. Forced differentiation along megakaryocyt ic l ineage upon GATA-1 expression is paralleled with suppression of myeloid l ineage (Visvader and Adams, 1993). Similarly the expression of Ikaros gene has been suggested to negatively regulate the myeloid differentiation as it promotes lymphoid development (Georgopolous et al. , 1994). 7 These few examples, together with many others not ment ioned here, illustrate the importance of transcription factors in the regulation of hematopoiesis. Hematopoiesis is, however, controlled at multiple levels. The regulatory role of other elements such as growth factors, growth factor receptors, the extracellular matrix, and adhesion molecules will be discussed below in Sect ion 1.2. 1.1.3 B i o l o g y of M y e l o i d L e u k e m i a s As discussed above, specific genes must be turned on and off in both temporal and l ineage-specific manner for normal hematopoiesis to occur. Interruption of these processes leads to an interference with the normal differentiation process. As a result, maturation arrest and/or cellular t ransformation may occur. In addit ion, leukemic cells may show abnormali t ies in their responses to, or dependence on, interactions with stimulatory or inhibitory cytokines that regulate their normal counterparts (Metcalf, 1993). Leukemia has thus been considered to be an uncontrolled proliferation or expansion of hematopoiet ic cells that do not retain the normal capacity to differentiate into mature blood cells, and may affect all l ineages including lymphoid cells (Sawyers et al . , 1991). This section will be restricted to an overview of myeloid leukemias. C h r o n i c M y e l o i d L e u k e m i a (CML) C M L is a multi l ineage clonal mal ignancy believed to arise as a result of a rearrangement of genetic material on chromosomes 9 and 22 in a pluripotent hematopoiet ic stem cell. This usually involves a reciprocal translocation (t9:22;q34:q11) and causes the formation of an abnormally shortened chromosome 22, called the Philadelphia Chromosome (Ph). During this translocat ion, the first exon of abl on chromosome 9 is replaced by various 5' bcr exons on chromosome 22 according to the positions of the specific breakpoints 8 in both of these genes. Two alternative chimeric Bcr-Abl oncogenic products of 210 KD (p210 Bcr-Abl) a n c j 135 KD (p185 Bcr-Abi)> associated primarily with CML and acute lymphoid leukemia (ALL), respectively, are generated. In CML, the Bcr-Abl chimeric fusion gene is found in hematopoietic cells of all l ineages. The clinical features of CML reflect an expansion of the myeloid compartment, with retention of full differentiation, during the initial (chronic) phase of the disease. This initial phase is eventually fol lowed by progression to acute myeloid leukemia (AML) which is characterized by the appearance of a significant populat ion of undifferentiated blast cells usually with a lymphoid or myeloid phenotype. This later stage of the disease (known as blast crisis) is often accompanied by the presence of additional secondary mutations which may involve p53 or Ras (Sawyers et al . , 1991). The abl gene encodes for a nuclear tyrosine kinase with src homology 2 and 3 (SH2 and SH3) domains. Upregulation of the kinase activity of Abl , essential for the oncogenicity of Bcr-Abl, depends upon the upstream bcr sequences fused to abl and the number of bcr exons involved (Lugo et al . , 1990; Muller et al. , 1991 ; Whirter et al., 1991). A large body of evidence has linked Bcr-Abl oncogenici ty to the specific stimulation of the Ras/ MAP kinase/ Jun kinase pathway (Pendergast et al. , 1993; Puil et al. , 1994; Skorski et al . , 1994; Tauchi et al. , 1994; Gishizky et al . , 1995; Raitano et al. , 1995; Sawyers et al. , 1995). Recent evidence also suggests that there are redundant pathways stimulating Ras through Bcr-Abl and that these pathways may be used differentially by fibroblasts and hematopoietic cells (Goga et al. , 1995). The oncogenic potential of Bcr-Abl fusion proteins has been demonstrated in a variety of cells in culture (McLaughlin et al., 1987; Lugo et al. , 1990). A direct role of Bcr-Abl in the development of P h + CML or ALL has also been shown by gene transfer and transgenic models (Daley et al. , 1990; Heisterkamp et al . , 1990). 9 Primitive hematopoiet ic cells in the blood and bone marrow of CML patients have been found to exhibit an unregulated state of proliferation (Eaves et al. , 1986). These cells do not respond to the inhibitory effect of the stromal adherent layer in LTC and cycle continuously (Eaves et al. , 1986; 1991b). Particularly, CML progenitors have been found to ignore the inhibitory effect of the growth factor MIP-1a which partially regulates the cycling status of primitive normal hematopoiet ic cells (Eaves et al. , 1993). However, these cells are inhibited by TGF-(3 to the same extent as normal cells (Cashman et al . , 1990; 1992). These f indings suggest that Bcr-Abl t ransformed cells have the ability to bypass the signaling through MIP-1a. Despite the anemia characteristic of untreated patients, the erythroid progenitor compartment is amplif ied in C M L patients and some of these cells are able to mature in the absence of erythropoietin (EPO) suggesting a major differentiation block at the level of C F U -E (Eaves and Eaves, 1979). A more recent study suggest that the EPO-independent erythroid colony-formation in CML is dependent on the presence of SF and directly involves the Bcr-Abl kinase activity (Issaad and Vainchenker, 1994). In t ransformed fibroblast cells, however, Bcr-Abl abrogates the anchorage dependence but not the growth factor requirements of these cells (Renshaw et al. , 1995). It was also found that although the number of C M L LTC-IC is about 20 fold lower in the marrow compared to normal individuals, the number of LTC-IC in the blood of CML patients is up to 1 0 5 higher suggest ing that the blood of CML patients is enriched for LTC-IC as compared to marrow (Udomsakdi et al. , 1992a, 1992b, 1992c). These data indicate that a defect in adhesive properties of Bcr-Abl transformed cells may contribute to the C M L pathogenesis. 10 A c u t e M y e l o i d L e u k e m i a (AML) In contrast to the molecular homogeneity of CML (and Ph+ ALL) , it is now appreciated that A M L comprises a heterogeneous family of malignant hematological disorders characterized both by an increase in the number of circulating white blood cells and a defect in their normal maturation and funct ion. These may appear de novo (without clinical recognition of a preleukemic phase) or may represent a progressed state of less acute human blood disorders, such as those described as myelodysplasia or, as discussed above, CML. These observat ions have suggested that full leukemic transformation requires defects in both growth and differentiation. The leukemic cell in one subset of acute leukemia may fail to express any known markers of differentiation and it has been postulated that these types of leukemia may result f rom all of the mutat ions occurr ing in a very primitive hematopoietic cell (Sawyers et al. , 1991). Mutat ion, deletion or rearrangement of various genes encoding for growth factors, growth factor receptors, and differentiation-specific transcription factors have all been found to be causative of and/or associated with different types of A M L in humans or in animal models. Alteration of these genes disrupt the normal development and maturation of hematopoietic cells, perhaps by blocking differentiation. Cooperat ion between genes that selectively affect cell growth, differentiation or survival may also result in a fully leukemic phenotype (Cline, 1994). Development of quantitative in vivo and in vitro assays along with genetic characterizat ion of myeloid leukemic subtypes will allow the detect ion, measurements , functional analysis and manipulation of specific primitive hematopoiet ic cell compartments of leukemic cells (Lapidot et al . , 1994). n 1.2 B o n e M a r r o w M i c r o e n v i r o n m e n t Although some primitive hematopoietic cells circulate, hematopoiesis normally occurs only in restricted sites in the bone marrow in humans and in the bone marrow and spleen of mice. Intravenously injected stem cells "home" to the marrow of irradiated recipients to initiate regeneration of hematopoiesis. After ectopic implantation of marrow and splenic t issue, reconstitution of s t roma precedes hematopoiesis (Tavassoli , 1989). The interactions between cell surface receptors including growth factor receptors and adhesion molecules expressed on both stroma and primitive hematopoietic cells and molecules present in the bone marrow microenvironment are believed to regulate hematopoiesis in vivo (Figure 2). These interactions also assure the regulation of the movement and circulation of hematopoietic cells within and in between the hematopoiet ic sites. Many components of these interactions have been identif ied, however, their relative importance in processes regulating hematopoiesis are poorly understood. This section will review current understanding of the relative contributions of the major components of the bone marrow microenvironment to the regulation of hematopoiet ic cell proliferation and differentiation. 1.2.1 Cel lu la r C o m p o n e n t s Human hematopoiesis takes place in the cavities of flat bones and the trabecular regions at the end of long bones. Hematopoiet ic cells are localized in the extravascular spaces between marrow vascular sinuses. The sinus wall forms a barrier, called the marrow-blood barrier, between the hematopoiet ic compartment and the circulation. This barrier is composed of an inner layer of endothel ial cells and an abluminal coat of cells described as adventit ial reticular cells. Developing blood cells must cross the sinus wall to enter the circulation. Bone marrow st roma consists of non-hematopoiet ic cells including f ibroblasts, 12 F igu re 2 . B o n e M a r r o w M i c r o e n v i r o n m e n t . Putative and known adhesion molecules expressed on primitive hematopoietic cells are shown. Stromal cells secrete growth factors, ECM proteins, and proteoglycans. Proteoglycans bind, stabilize, and present growth factors to the relevant cells. 13 macrophages, adipocytes, endothelial cells, and reticular cells, which together produce an important extracellular matrix (ECM) network. These cells are also involved in the production of cytokines active on hematopoiet ic cells (Lichtman, 1981). Stromal cells have been identified by their morphology, cell surface molecule expression, and enzymatic activities (Lichtman, 1984). Primitive hematopoietic cells may be found closely associated with stromal cells (Figure 2) , as has been described for erythroid progenitors and macrophages in so-called erythroblastic islands (Undritz, 1950) or in bone marrow cell aggregates (Funk et al, 1994). Adipocytes, f ibroblasts and macrophages have also been found to associate with one another in clusters that remain intact even after bone marrow aspiration (Blazsek et al. , 1990). Al though these clusters of cells are not enriched in their content of hematopoiet ic cells, they may produce large quantities of growth factors, such as GM-CSF. Interestingly, these stromal cell complexes are attached to glycosaminoglycans that are known to have a role in presenting growth factors such as G M - C S F (Clark et al . , 1992). 1.2.2 G r o w t h Fac to rs a n d Inh ib i to rs The proliferation, survival and differentiation of hematopoiet ic cells is control led in large part by the regulated production and release of a group of glycoproteins termed growth factors. Hematopoietic growth factors provide the growth-promoting and differentiation-inducing or inhibitory stimuli that control these responses in hematopoietic cells. A large number of these factors have been identified and their genes cloned. During the past decade, molecules involved in intracellular pathways tr iggered by growth factor-receptor binding and their effects on cycling or differentiation of hematopoietic cells have also been identif ied. The effect of growth factors may be early, acting on primitive hematopoiet ic progenitors or late, acting on more mature l ineage-restricted 14 progenitors. Many growth factors may also exhibit synergistic or at least addit ive effects on hematopoiet ic targets and most hematopoietic growth can, in fact, affect cells of several l ineages (Metcalf, 1989; Moore, 1991 ; Ogawa, 1993). Hematopoiesis can be sustained in long-term marrow cultures in the absence of added growth factors. However, many such growth factors are produced endogenously by the stroma. In fact, mRNA for several st imulating and inhibitory growth factors have been detected in bone marrow derived stromal cells and their proteins have been found in the stromal culture supernatants. These factors include colony-stimulating factor-1 (CSF-1), granulocyte-CSF (G-CSF) , granulocyte-macrophage-CSF (GM-CSF), interleukin-1 (IL-1), IL-6, IL -11 , leukemia inhibitory factor (LIF), b-basic-fibroblast growth factor (b-FGF), Steel factor (SF), macrophage inf lammatory protein-1a (MIP-1a) and transforming growth factor-p (TGF-p) (Anderson et al. , 1990; Broxmeyer et al. , 1990; Cashman et al . , 1990; Gabbianell i et al. , 1990; Paul et al. , 1990; Eaves et al. , 1991a; Wetzler et al. , 1991b). Surprisingly, IL-3, which is a potent factor for st imulating the proliferation of multipotential hematopoiet ic cells, does not appear to be normally produced by stromal cells (Li and Johnson, 1985; Eaves et al . , 1991a). Except for b-FGF and C S F - 1 , all of the factors listed above have been found to affect, either positively or negatively, the proliferation of primitive mult ipotent hematopoiet ic cells (Verfaillie et al., 1994a). Inhibitory factors of hematopoiesis include TNF-oc and interferons that appear to be l ineage-non specific, affecting progenitors at different stages of differentiation (Raefsky et al. , 1985; Akahane et al., 1987), as well as MIP-1a and TGF-p. MIP-1a and TGF-p are specific to primitive hematopoietic cells and have been thought mostly responsible for the quiescent state of many primitive hematopoiet ic cells (Cashman et al. , 1990; Broxmeyer et al. , 1990). Interestingly, IL-3 and SF have been reported to support also the survival of primitive hematopoiet ic cells 15 (Bodine et al . , 1991 ; Katayama et al. , 1993). In contrast to what their names suggest, neither G-CSF nor GM-CSF effects are restricted to granulocyte and/or macrophage precursors. Indeed, it is now clear that both of these factors may synergize with others to stimulate the proliferation of primitive multipotent as well as other types of l ineage-restricted hematopoietic cells. Growth factors may be found in the bone marrow microenvironment in diffusible, soluble form or, alternatively, may be presented to such cells in an ECM-bound form. In addit ion, SF exists in both diffusible and membrane-bound forms. Interestingly, it has been reported that membrane-bound SF may be a significantly more potent hematopoiet ic stimulating factor than soluble SF (Toskoz et al. , 1993). It seems likely that bone marrow stromal cells also produce both the recently c loned flk2/flt3 ligand and thrombopoiet in (TPO; also known as mpl l igand). T P O appears to be a primary regulator of megakaryocy te differentiation (Bartley et al. , 1994; de Sauvage et al., 1994; Lok et al., 1994). Both flk2/flt3 l igand and T P O have been found to be potent stimulators of the proliferation of primitive multipotent hematopoietic cells (Lyman et al . , 1993; Hannum et al. , 1994; Pe tze re t al. , 1996b). An interesting model which supports a role of the s t roma in regulating hematopoiesis in vivo is provided by Steel (SI) and W mutant mice. Mutat ions at either the W or SI locus appear to have similar effects on the development of hematopoiet ic, melanocytes and germ cell l ineages although the underlying mechanisms are different (Witte, 1990). Molecular analysis of the W and SI locus have identified W is an allele of the c-Kit gene which codes for a tyrosine kinase receptor closely related to the receptors for colony stimulating factor-1 (CSF-1R) and platelet-derived growth factor receptor (PDGF-R) (Chabot et al. , 1988). The SI gene encodes a soluble factor that functions as a l igand for the c-Kit receptor (Copeland et al. , 1990; Zsebo et al. , 1990; Huang et al. , 1990). In addit ion, because the SI gene product can be membrane-bound, its receptor, c-16 kit, can also function as an adhesion molecule (see below). These f indings explain earlier observations demonstrat ing that transplantation of t issue f ragments f rom either +/+ or W mutant mice into SI mice cured the hematological abnormali t ies of the SI recipient mice (Bernstein, 1970) and intravenous injection of SI mutant hematopoiet ic cells into W mutant corrected their anemia, whereas W mutant hematopoiet ic cells injected into SI mutant recipients failed to correct the anemia of the latter (McCulloch et al. , 1965). Interestingly, several growth factors such as G-CSF, SF, IL-7 and MIP-1a have been found to enhance the release of primitive hematopoiet ic progenitors, including those with repopulating ability, into the peripheral blood fol lowing their administrat ion in vivo (Welte et al., 1987; Cohen et al., 1987; Bodine et al. , 1993; Briddell et al . , 1993; Krzysztof et al. , 1995; Lord et al. , 1995). This is usually accompanied by an increase in the production and release of large numbers of mature circulating WBCs. Although the exact mechanism of peripheral blood progenitor mobil ization by growth factors in unknown, it is commonly thought that this effect of growth factors in vivo may be mediated through the modulat ion of adhesive interactions between primitive hematopoietic cells and the molecular components of the bone marrow microenvironment. 1.2.3 Ex t race l lu la r Mat r ix (ECM) In addit ion to growth factors, bone marrow stromal cells produce a number of proteins that form a complex ECM network. These proteins include several col lagens (Type I, III, IV, V) , glycoproteins (f ibronectin, thrombospondin, hemonect in, laminins, tenascin), and glycosaminoglycans or GAG (hyaluronic acid (HA), heparan sulfate, dermatan sulfate and chondroit in sulfate). This ECM provides a dynamic physical intercellular bridge and assists in facilitating cell-cell communicat ion by conveying information to hematopoietic cells through specif ic cel l-surface receptors, and by co-localizing and stabilizing growth factor 17 interactions between hematopoietic cells and stromal elements. For example TGF-p can bind several ECM proteins such as f ibronectin, col lagen type IV, thrombospondin and the core protein or GAG chains of proteoglycans (Massague, 1992; Butzow et al. , 1993). Several lines of evidence suggest that the E C M may control the proliferation, survival and differentiation of various cell types (Juliano and Haskill, 1993), specific examples relevant to hematopoiet ic system are discussed below. F i b r o n e c t i n Fibronectin is a ubiquitous ECM glycoprotein of 450 KD composed of two similar subunits joined together by disulfide bonds. Various isoforms of f ibronectin are generated by alternative splicing. A 75-120 KD proteolytic f ragment in the center of the molecule contains the "cell binding domain". The min imum ligand-recognit ion site within this domain is the 3 amino acid sequence arginine-glycine-aspart ic acid (RGD). Cells bind to the RGD sequence through integrin receptors, mostly through a5p1 and less frequently av|33 or Gpl l - l l la (Pytela et al. , 1985; Ruoslahti and Piersbacher, 1987). Adhesion to f ibronectin can, however, occur in an RGD-independent manner through binding to a region in the 33/66 KD proteolytic f ragment of the carboxy-terminal heparin-binding domain. This region harbors a sequence known as l l l -CS. A peptide containing the l l l -CS sequence present in certain isoforms of f ibronectin, and known as C S -1, contains the leucine-aspartic acid-valine amino acid sequence (LDV) which is known to interact with a4(31 integrin (Guan and Hynes, 1990). Two f lanking regions of C S - 1 , known as FN-C/H I and FN-C/H II also interact with oc4[31 integrin, or with membrane-bound proteoglycans (Haugen et al. , 1990). Several lines of evidence indicate that the ability of hematopoiet ic cells to adhere to f ibronectin is regulated during their differentiation. Early work by Patel and Lodish (1984) showed that adhesion to f ibronectin changes during the 18 differentiation and maturation of erythroid cells and that differentiation was arrested in the late erythroblast stage when erythroid cells were induced to differentiate in the absence of f ibronectin. In addit ion, the adhesion of hematopoiet ic progenitors, including LTC-IC, to various f ragments of f ibronectin is differentially regulated with maturation (Coulombel et al. , 1988; Vui l let-Gaugler et al. , 1990; Verfaill ie et al. , 1991 ; Wil l iams et al., 1991 ; Papayannopoulou and Brice; 1992; Kerst et al. , 1993; Verfaill ie et al. , 1994b). T h r o m b o s p o n d i n Thrombospondin is a 450 KD glycoprotein composed of three identical subunits cross-l inked at their amino terminal heparin-binding domain by disulfide bonds. This protein is produced by f ibroblasts, platelets and endothelial cells and is involved in cell adhesion and binding to glycosaminoglycans and f ibr inogen (Lawler and Hynes, 1987). Cell surface proteoglycans can bind to the heparin-binding domain of thrombospondin (Asch et al. , 1990). Thrombospondin serves as an adhesive ligand for a variety of hematopoietic progenitors (CFU-G E M M , BFU-E and CFU-GM) (Long and Dixit, 1990). During their differentiation, adhesion to thrombospondin is lost totally for mature erythroid cells and partially for granulocytes. This suggests that the receptor for thrombospondin on hematopoiet ic progenitors is differentially downregulated during the differentiation of hematopoietic cells along particular pathways (Bornstein, 1992). Interestingly, adhesion of hematopoietic progenitors to immobil ized thrombospondin in the presence of immobil ized SF increases dramatical ly the response of progenitors to additional cytokines, such as IL-3 and GM-CSF. This has led to the suggest ion that adhesion to thrombospondin may generate a signal that modulates the response of progenitors to cytokines (Long et al. , 1992). 19 C o l l a g e n a n d L a m i n i n Collagens and laminin are found in large amounts in the bone marrow. The most common forms of collagen are col lagen types I, II, III, and IV. These f ibrous proteins are found in the extracellular space where col lagen types l-lll constitute the structural backbone. Col lagen type IV is assembled in a sheet-l ike mesh work and makes up a major part of basal membranes (Verfaillie et al, 1994a). Laminin is an 850 KD complex of cross-shaped heterotr imers. Seven different assembly forms have been detected. Laminin has several funct ional domains and can bind to collagen type IV, various cell surface molecules including CD44 (see below), and proteoglycans (Yoder and Wil l iams, 1995). The exit of mature hematopoietic cells f rom the bone marrow into the peripheral blood requires that they traverse the basal membranes of endothelial cells. Granulocytes have integrin receptors that may facilitate this process (Liesveld et al. , 1991). However, more primitive hematopoiet ic cells do not express these receptors and have not been found to bind to col lagen or laminin structures. It is therefore not clear whether either col lagen or laminin play a role in localization, proliferation and/or differentiation of primitive hematopoiet ic cells. T e n a s c i n The tenascin family consists of three members known as tenascin-C, tenascin-R, and tenascin-X (Erickson, 1993). The tenascins are large mult idomain glycoproteins consisting of an EGF-like domain, a f ibronectin type III domain, and a fibrinogen-like terminal knob. Different isoforms of tenascin-C are also generated by alternative splicing. Tenascin-C can be detected in the bone marrow between developing hematopoietic cells and also in LTC, where it is deposi ted in an extracellular meshwork overlaying the stromal cells. Tenascin-C can act as both an anti-adhesive and an adhesive molecule. Various human 20 myeloid cell lines have been shown to bind to tenascin-C, but the nature of their l igands or the binding of more primitive progenitors to tenascin-C is not clear. The synthesis of tenascin-C by bone marrow stromal cells is strongly inf luenced by the presence of glucocorticoids. Addit ion of glucocort icoids to LTC either at the t ime of their initiation or after they have been establ ished downregulates dramatically the expression of tenascin-C (Erickson, 1993). The influence that glucocorticoids have been reported to have on LTC (Croisille et al . , 1994) may therefore be mediated at least partly through modulat ion of tenascin-C expression. G l y c o s a m i n o g l y c a n s (GAGs) GAGs are negatively charged polysaccharides composed of repeating disaccharide units. Except for HA, all GAGs are sulfated molecules and in t issues are found covalently attached to a core protein (in this form they are called proteoglycans) (Lindhal and Hook, 1978). GAGs include heparin, heparan sulfate, dermatan sulfate, keratan sulfate, chondroit in sulfate, and HA, all of which are detected in the bone marrow microenvironment (except for keratan sulfate). Various growth factors and components of the ECM can show a high affinity for binding GAGs. The ECM proteins with such binding sites include f ibronect in, col lagen, laminin, and thrombospondin. Among growth factors, f ibroblast and endothelial cell growth factors, in particular, have very high binding affinities for GAGs. The binding of GAG to various molecules appears to be charge dependent since this binding is largely modulated by the degree of sulfation of the GAG. For example, the more highly sulfated the G A G , the better it will bind to f ibronectin. The binding of GAG to ECM is also affected by the size of the GAG and/or by the number of GAG chains attached to the same core protein, which increases the number of potential GAG interaction sites. 21 The core protein of these proteoglycans are also capable of interacting with other molecules (Ruoslahti, 1989). Binding of growth factors to proteoglycans or GAGs is thought to have an important regulatory role. For example FGF or TGF-p both bind signaling receptors with high affinity. However, the same growth factors also bind with low affinity to cell surface proteoglycans that can not transmit signals but somehow modulate the ability of the growth factor or the signaling receptor to generate a biological response (Schlessinger et al . , 1995). Several mechanisms have been proposed. Proteoglycans may stabilize the growth factor and protect it f rom proteolytic degradat ion. They may also concentrate a given growth factor at particular sites, thus potentially enhancing its efficacy. Alternatively, growth factors bound to E C M proteoglycans may serve as a reservoir for growth factors, that can be slowly released by enzymatic degradation of the proteoglycans. The binding of a growth factor to a proteoglycan may also induce a conformational change in the growth factor that enhances its receptor affinity (Schlessinger et al. , 1995). In turn, some growth factors, such as TGF-p, can induce the production of GAGs and proteoglycans. For example, some of the growth regulatory activities of TGF-p have been attributed to this function (Ruoslaht i ; 1989; Faassen et al. , 1993). GAGs are produced by both hematopoietic and stromal cells and are present in abundance in LTC (Gallagher et al. , 1983; Spooncer et al . , 1983; Yoder and Wil l iams, 1995). Some GAGs such as HA and heparan sulfate appear to be involved in the adhesion of some primitive hematopoiet ic progenitors to s t roma in LTC (Gordon et al., 1988). Glucocort icoids, which are necessary for sustained myelopoiesis in LTC, have been found to increase the sulfation of GAGs such as heparan sulfate LTC in addition to their effect on tenascin-C, as noted above (Gallagher et al., 1983). Increased sulfation of heparan sulfate and other GAGs in glucocorticoid treated LTC is thought to 22 enhance the binding of primitive hematopoietic progenitors to s t roma via heparan sulfate, by increasing the binding affinity of heparan sulfate for its l igand(s) (Siczkowski et al. , 1992). This may, at least partly, explain the effect of glucocort icoids on hematopoiesis in LTC (Croisille et al. , 1994). Heparan su l fa te has been reported to bind several growth factors including IL-3, GM-CSF, FGF, and MIP-1p (Gordon et al. , 1987; Roberts et al . , 1988; Rapraeger et al. , 1991 ; Tanaka et al. , 1993) and is found mainly in the adherent layer of LTC (Gallagher et al., 1993). Heparan sulfate may enhance hematopoiesis in LTC by binding both primitive progenitors and growth factors that can stimulate them, by concentrating these growth factors at the site of their cellular receptors, or by enhancing the attachment of primitive cells to the stromal feeder layer. Similarly, c h o n d r o i t i n su l fa te GAGs have been found to st imulate hematopoiesis in LTC. Chondroit in sulfate is the major sulfated GAG found in the LTC (mostly in the medium) (Gallagher et al. , 1983). An increased synthesis of chondroit in sulfate in LTC was found to correlate with an enhancement of CFU-S and CFU-GM production in LTC (Spooncer et al. , 1983). The major chondroit in sulfate proteoglycan found on primitive hematopoiet ic progenitors is CD44 (Verfaillie et al . , 1994b). HA is a major component of the extracellular matrix that has been commonly found to surround migrating and proliferating cells, particularly in embryonic t issues (Underhil l , 1992). HA is a non-sulfated GAG composed of linear polymers of disaccharide repeats [glucuronic acid [3(1 -3) A/-acetyl g lucosamine, up to 50,000 repeats] and differs f rom other GAGs in that it is not covalently l inked to a core protein. The p linkages between the saccharides and the presence of hydrogen bonds allow each molecule of HA to form a long rod-like structure (Laurent and Fraser, 1992). The large number of hydrophil ic residues on HA allow it to bind a large amount of water and this results in the 23 production of a highly hydrated, relatively rigid matrix through which cells are able to migrate. This property of HA may be due to its effect on phosphorylat ion of focal adhesion proteins including p p 1 2 5 F A K via the cell surface receptor R H A M M that has been implicated in cell locomotion (Hall et al. , 1994). HA can bind to the surface of cultured cells with a relatively high affinity and cells in LTC are found coated with HA (Underhil l and Toole, 1979; Gallagher et al. , 1983) . In addit ion, the production and local accumulat ion of HA have been reported to character ize or to correlate with the degree of invasion of some carc inomas (Knudson et al. , 1989), although these results remain controversial (Underhil l and Toole, 1982). Foca l A d h e s i o n s (FA) FA are discrete p lasma membrane assemblies in which complexes of cytoskeletal and membrane components are tightly associated with the underlying E C M . In these regions, the plasma membrane is special ized at its cytoplasmic face for anchoring stress fibers, the large bundle of microf i laments that are prominent in many cultured cells (Burridge et al. , 1988). A variety of molecules including integrin adhesion and growth factor receptors as well as cytoskeletal proteins, have been found to converge at the FA site (Craig and Johnson, 1996). One of the major cytoplasmic proteins found at the site of FA is a tyrosine kinase of 125 KD named focal adhesion kinase ( p p 1 2 5 F A K ) (Schaller et al . , 1992). This protein is phosphorylated when cells bind f ibronectin v ia integrins and/or when cells are transformed by oncogenes such as Bcr-Abl or v-src with which it may also physically associate (Guan and Shal loway, 1992; Hanks et al . , 1992; Schlaepfer et al., 1994; Salgia et al. , 1995a; 1995b; 1996). p p 1 2 5 F A K has also been found associated with a number of cytoskeletal proteins including paxillin (Schaller and Parsons, 1994). It has been reported recently 24 that the formation of FA complexes also requires the activity of small GTP-binding proteins Rac and Rho (Nobes and Hall, 1995; Hotchin and Hall, 1995). Al though the increasing volume of information on ECM signaling through FA and the link between p p 1 2 5 F A K and other cellular proteins have improved our understanding of how ECM may contribute to the cellular behavior such as cell shape, migration and proliferation, the relative importance of these molecules in specific organs remains to be investigated. In particular, most, if not all of these studies have been restricted to fibroblasts and epithelial cells. 1.2.4 A d h e s i o n M o l e c u l e s Normal traffic of primitive progenitor cells into and out of their sites of production requires interactions of adhesion ligands with their receptors on both stromal cells and primitive hematopoietic cells. Primitive hematopoiet ic cells with either short or long term regenerating capacity can adhere avidly to both plastic (Gordon, 1994) and stromal cells in vitro (Mauch et al. , 1980; Kerk et al. , 1985; Gordon et al. , 1990; Verfaillie et al. , 1990; Trevisan and Iscove; 1995). In vivo these cells are also found tightly associated with stromal cells in the bone marrow (Funk et al. , 1994). Nevertheless, primitive hematopoiet ic cells can leave the marrow and circulate in the peripheral blood. Conversely, the peripheral intravenous administration of primitive hematopoiet ic cells causes them to "home" specifically to the bone marrow in humans and mice, and in spleen in mice. Stromal cells, including endothelial cells, express a variety of adhesion receptors including V C A M - 1 , CD34, P E C A M - 1 , CD44, ICAM-1 , M a c - 1 , and E-selectin, whose expression may all be at least partly controlled by growth factors (Simmons et al. , 1992; Liesveld et al. , 1994; Rafii et al. , 1994; Jacobsen et al. , 1996). In fact, various cytokines such as TNF-a , IL-3, GM-CSF and SF have been found to upregulate the hematopoietic or stromal cell expression and/or 25 binding affinity of adhesion molecules such as VCAM-1 or a4(31 and a5(31 (Simmons et al . , 1992; Levesque et al. , 1995). A l though the cell surface expression of several adhesion molecules on primitive hematopoiet ic cells has also been documented (Figure 2) , much less is. known about the function of these receptors in interactions with stromal l igands and components of the ECM. In fact, the presence of these receptors on the surface of the cells does not always indicate an adhesive function since many adhesion receptors need to be activated to be functional. The fol lowing section will focus on adhesion molecules with characterized function in primitive hematopoiet ic cells. I n t e g r i n s The integrins comprise a large family of divalent-dependent heterodimer adhesion receptors composed of an a subunit noncovalently at tached to a p subunit, and they are variably expressed on a wide range of cells (Hynes, 1992). There are 8 known p subunits and 14 known a subunits. Several a subunits share the same (3 subunit (for example a subunits in VLA family share the same p i integrin). The function of these molecules has been largely character ized on leukocytes. The intracellular domain is heterodimer-specif ic, suggest ing a unique function for each integrin combinat ion. The cytoplasmic domain of the p i integrin is associated with the cytoskeleton. Binding of p1 to its l igand may generate an intracellular signaling cascade event known as "outside in" signal ing, and also may be responsible for modulat ing the affinity of the heterodimer complex binding to its l igand in response to other activating pathways and known as "inside out" signaling. In many cases integrin activation is necessary for their modulation from a low affinity to a high affinity binding-competent state. 26 Integrins such as LFA-1 (P2ol), VLA-4 (a4p1) and VLA-5 (a5p1) are expressed on the surface of normal human CD34+ bone marrow cells (Liesveld et al . , 1993). In addit ion, it has been shown that LFA-1 is specifically expressed on l ineage commit ted progenitors and not on LTC-IC and may contribute to the adhesion of the former to the stroma (Gunji et al. , 1992). a4(31 integrin is an adhesion molecule that appears to be important in mediat ing the interactions of primitive hematopoietic progenitors with s t roma. Addit ion of antibodies to a4p1 completely inhibits lymphopoiesis and retards myelopoiesis in LTC (Miyake et al. , 1991). a4p1 and a 5 p i are both expressed on human primitive and commit ted hematopoietic progenitors and are the principal mediators of adhesion of these cells to f ibronectin (Verfaillie et al. , 1991 ; Wil l iams et al. , 1991 ; Simmons et al., 1992; Teixido et al., 1992; Kerst et al. , 1993; Liesveld et al. , 1993). However, oc4pi also mediates the adhesion of human commit ted progenitors to stromal cells through binding of VCAM-1 (Simmons et al. , 1992). It has been shown that the stromal-st imulated development of erythroid progenitors is dependent, at least partly, on a variation in their expression of a 4 p i and a 5 p i and the adhesive interactions of these integrins with the respective ligand(s) (Coulombel et al. , 1988; Rosemblatt et al. , 1991 ; Papayannopoulou and Brice, 1992; Yanai et al . , 1994). a 4 p i may also cooperate with CD44 in mediating the adhesion of l ineage-restricted progenitors to f ibronectin (Verfaillie et al., 1994b). a4pi - f ib ronect in binding, that mediates the adhesion of commit ted progenitors to stromal cells, also inhibits the proliferation of these progenitors (Hurley et al. , 1995). It has also been reported that interactions of the oc5pi integrin on hematopoietic progenitors with f ibronectin inhibits the proliferation of the growth factor-dependent (GM-CSF or SF) cell line M 0 7 e and induces programmed cell death (Sugahara et al. , 1994). These observat ions suggest that the binding of hematopoietic cells to f ibronectin 27 via a4(31 and a5|31 may also control hematopoietic cell proliferation versus survival in the bone marrow. However, further investigations with primary primitive cells are clearly required. a4p i - f ib ronect in and a4(31-VCAM-1 interactions have also been implicated in mediat ing the in vivo homing of primitive hematopoiet ic progenitors to the marrow and spleen. The same molecular mechanisms may also be involved in the mobil ization of progenitors to the blood (Will iams et al. , 1991 ; Papayannopoulou and Nakamoto, 1993; Papayannopoulou et al . , 1995). Taken together, these data suggest that primitive hematopoiet ic cells bind st roma, at least partly, through interactions of integrins with their respective l igands expressed in st roma. These interactions may be partly control led by the production of growth factors, which can also regulate the cycl ing status of primitive progenitors. Se lec t i ns Selectins are a family of cell adhesion molecules that contain an amino-terminal C-type ( C a + + dependent) lectin domain, an EGF repeat, and a discrete number of modules similar to those found in certain complement-binding proteins. Selectins mediate leukocyte-endothelial cell interactions by binding to specific carbohydrate ligands (Bevi lacqua and Nelson, 1993). The L-selectin homing receptor is expressed on leukocytes, lymphocytes and hematopoiet ic progenitors, P-selectin is found on platelets and endothelial cells, and E-selectin is expressed on endothelial cells, specifically fol lowing their exposure to cytokines such as T N F - a or IL-1 (Varki, 1994). Less is known about the expression of selectins on primitive hematopoiet ic progenitors. However, L-selectin is known to be present on primitive CD34+ bone marrow hematopoietic cells and may contribute to their homing to the marrow after transplantation (Dercksen et al. , 1995). The addit ion 28 of anti-L-selectin antibodies results in agglutination of nonadherent cells and a slower rate of mature myeloid cell production in LTC (Miyake et al. , 1990a). CD34+ cells, including l ineage-restricted progenitors, express the P-selectin glycoprotein l igand (PSGL-1) and can bind to P-selectin (Zannett ino et al. , 1995). Mice deficient for both E-selectin and P-selectin exhibit an elevation in their circulating numbers of mature leukocytes in addition to an increase in their splenic content of CFU-GM and BFU-E (Frenette et al . , 1996). I m m u n o g l o b u l i n S u p e r g e n e Fami ly ( IgSF) This superfamily of cell surface molecules is characterized by the presence 90-100 amino acid domain first characterized in immunoglobul in (Ig). Some members of this family, such as LFA-3, exist as molecules covalently l inked to a phosphatidylinositol structure at the cell surface. Many Ig family members are involved in both cell-cell communicat ion and antigen recognit ion. One member of this family V C A M - 1 , is a cytokine-inducible cell adhesion molecule expressed by stromal and endothelial cells and is involved in their ability to bind to hematopoietic progenitors. Another molecule, I C A M - 1 , has been well characterized for its ability to bind to L F A - 1 . ICAM-1 is expressed on a subset of l ineage-restricted hematopoietic progenitors and its expression appears to be lost with maturation (Arkin et al. , 1991). The adhesive function of ICAM-1 on these progenitors is less well characterized. Thy-1 is a glycosyl phosphatidylinositol (GPI)-l inked member of the IgSF. This cell surface glycoprotein is expressed at high levels on T lymphocytes and neurons and at low levels on certain very primitive hematopoiet ic cells. It has been suggested that Thy-1 is an adhesion molecule on thymocytes and neurons (He et al. , 1991), however, the role of Thy-1 as an adhesion molecule on more primitive hematopoiet ic cells has not been clarif ied. It has been shown that CD34+ cells expressing Thy-1 are particularly enriched for LTC-IC, therefore 29 Thy-1 has been used as a marker for the purification of primitive hematopoiet ic progenitors (Craig et al . , 1993). c-Ki t a n d Steel Fac to r c-kit is expressed on both primitive hematopoietic progenitors and stromal cells (Berardi et al. , 1995, Funk et al. , 1995). SF exists in both soluble and membrane-bound forms (Flanagan et al., 1991). In addition to st imulating proliferation, by binding to c-kit, the membrane-bound form of S F may also funct ion as an adhesion molecule. For example, it has been shown that the adhesion of human megakaryocytes to bone marrow stromal f ibroblasts that express the membrane bound form of SF is mediated partly through interactions of S F with c-kit expressed on megakaryocytes and results also in the proliferation of megakaryocytes (Avraham et al. , 1992). SF has also been found to modulate the avidity of a4[31 and a5(31 integrins on hematopoiet ic cell lines including megakaryoblast ic cells (Kovach et al. , 1995). Following its binding to SF, c-kit activates mast cell adhesion to f ibronectin v ia a5(31 integrin and this function requires activation of the c-kit tyrosine kinase activity (Dastych and Metcalfe, 1994; Kinashi and Springer, 1994). The adhesion of primitive hematopoiet ic cells to stroma may also be mediated partly by c-kit (Kodama et al . , 1994). O the r Cel l Su r face R e c e p t o r s Several other cell surface molecules are also expressed on CD34+ and primitive hematopoiet ic progenitor cells, some of which are used commonly in the purif ication and selection of primitive hematopoietic progenitors (for example antibodies to the cell surface receptors CD34, CD45, T h y - 1 , CD38 have been used in various experiments presented in this thesis). However, it is not clear 30 whether these molecules serve a significant, if any, adhesive function on primitive hematopoiet ic cells. CD34 is a heavily O-glycosylated mucin expressed by all primitive hematopoiet ic progenitors and endothelial cells. Two different species generated by alternative splicing have been descr ibed. Despite the small size of the CD34+ compartment of normal marrow ( -1%) , this includes all of the progenitors detectable by quantitative functional assays. This makes CD34 a useful marker of primitive hematopoietic progenitors for all cell separat ion and purif ication studies (Strauss et al. , 1986; Krause et al. , 1996). It has been reported that the murine endothelial CD34 binds L-selectin (Baumhueter et al. , 1993). More recently, CD34 has also been implicated in the regulation of adhesive interactions of primitive hematopoietic progenitors with s t roma (Healy et al. , 1995). However, the ligand of CD34 in these interactions does not appear to be L-selectin and remains to be def ined. Mice deficient in CD34 are viable but exhibit an overall reduction in their hematopoiet ic progenitor activity (Cheng et al., 1996). It is not yet known whether this defect is due to abnormal adhesive interactions of these cells. CD45, or the leukocyte common antigen, is a family of glycoproteins that are expressed at different levels by the majority of hematopoietic cells. Several isoforms of this molecule are generated by alternative splicing. Some of these are differentially expressed on primitive hematopoietic progenitors of various l ineages (Lansdorp et al. , 1990). The overall low level of CD45 expression on primitive hematopoiet ic progenitors within the CD34 compartment has made this molecule a useful tool for the selection of different types of hematopoiet ic progenitor cells. CD45 is a phosphatase whose function as an essential component of ant igen-induced signaling events in lymphocytes is well establ ished. Several lines of evidence indicate that CD45 may also funct ion as an adhesion molecule on lymphocytes (Brady-Kalnay and Tonks, 1995). Using a 31 primitive hematopoiet ic cell line (FDCP-mix) cocultured with stromal cells it has been shown that CD45 on FDCP-mix may participate in cell adhesion by binding to heparan sulfate on the st roma (Coombe et al. , 1994). However, an adhesive funct ion of CD45 on primitive normal hematopoietic cells has not been reported. CD38 is a 46 KD type II t ransmembrane glycoprotein, expressed on T and B lymphocytes and primitive l ineage-restricted hematopoiet ic cells. Al though the function of this molecule is not clear, CD38 has been implicated in several diverse activities including cell proliferation and cell adhesion (Malavasi et al . , 1994). CD34+CD38- cells are enriched for LTC-IC and ant i-CD38 antibodies have therefore been used for the purification of primitive hematopoiet ic cells (Sauvageau et al. , 1994). However, the function of this molecule on hematopoiet ic progenitors remains unknown. 1.2.5 L o n g - T e r m M a r r o w Cu l tu res as a Mode l The original development of murine bone marrow LTC by Dexter and col leagues has since contributed considerably to our understanding of the role of the bone marrow microenvironment in hematopoiesis in vivo (Dexter et al. , 1977). In contrast to in vitro clonogenic assays using semi-solid media (Metcalf, 1977), the development, differentiation, and maintenance of primitive hematopoiet ic cells, including C R U (Fraser et al. , 1992), occurs in close proximity of these cells with the adherent layer of stromal cells. Interestingly, semi-sol id media do not support the proliferation of LTC-IC even when supplemented with growth factors that can stimulate LTC-IC proliferation and survival (Petzer et al. , 1996a). Detailed biochemical and microscopic studies performed on stromal cells in the adherent fraction of LTC have led to the conclusion that these latter cells are similar to bone marrow stromal cells in vivo in terms of their composi t ion, ECM, and growth factor producing abilities (Dexter, 1982). 32 Thus, the adherent layer of the LTC mimics many features of the bone marrow microenvironment in vivo and over the years it has been used as a model to examine interactions of primitive hematopoietic cells with var ious components of the bone marrow microenvironment. Many of the data presented above, have, in fact, been generated by analysis of variously manipulated LTC. In establ ished LTC, primitive hematopoietic cells are localized in the adherent fraction in juxtaposit ion to the stromal cells whereas the more mature hematopoiet ic cells are released into the nonadherent fraction (Mauch et al. , 1980; Dexter, 1982; Coulombel et al. , 1983, Kerk et al . , 1985). Al though the formation of an adherent stromal layer is necessary to support the hematopoiesis that occurs in LTC (Dexter, 1982), actual physical contact between hematopoiet ic and stromal cells is not required (Verfaillie et al. , 1992; Zandstra et al. , 1994). In fact, the support provided by the stromal cells can be replaced by their condit ioned media (Verfaillie et al. , 1993), al though the identity of the active factors has not yet been establ ished. However, primitive hematopoiet ic cells in vivo are found to be physically associated with stromal cells (Funk et al. , 1994). 1.2.6 A b n o r m a l I n te rac t i ons in L e u k e m i a Primitive CML progenitors have been found to have a decreased ability to adhere to normal s t roma (Gordon et al. , 1987; Verfaillie et al . , 1992). This may contribute to their release into the peripheral blood in vivo. However, recent studies have shown that elevated numbers of normal LTC-IC may also be present in the blood of CML patients suggesting a reactive process rather than an intrinsic defect as the underlying mechanism. Examinat ion of the cellular and ECM composit ion of the bone marrow of patients with C M L has demonstrated few differences with that of normal bone marrow (Verfaillie et al. , 1994a). Al though Bcr-Abl-posit ive cells (likely 33 macrophages) are found in the stroma of CML bone marrow, the fibroblast-stromal components are not part of the malignant clone and no difference has been found in the cytokine production of chronic phase CML stromal cells compared to normal stroma (Otsuka et al. , 1991; Bhatia et al. , 1995). In the later stages of the disease, endogenous IL-1 production may be increased (Wetzler et al. , 1991a) similar to observations of patients with other types of acute leukemia (Hogge, 1994). C M L progenitors have been found to have a decreased ability to bind to f ibronectin al though they adhere normally to basement membrane components (Verfaillie et al. , 1992). Similarly, as compared to normal cells, CML progenitors express higher levels of the laminin and col lagen ligands, the a2 and a 6 integrins, as well as a nonfunctional (31 integrin (Bhatia et al. , 1994). However, the level of expression of other f ibronectin l igands, i.e. the (31 as well as the a4 and oc5 integrins, is normal. CML progenitors have also been reported to lack expression of a phosphatidylinositol- l inked cell adhesion molecule and LFA-3 (Gordon et al. , 1991 ; Upadhyaya et al., 1991). Treatment with interferon a which in some patients can induce a normalization of the peripheral blood count with some restoration of normal hematopoiesis, appears to induce the normal expression and/or function of one or more of these adhesion molecules (Upadhyaya et al. , 1991 ; Bhatia et al. , 1994). Recent molecular f indings of the cytoskeletal and focal adhesion state of Bcr-Abl t ransformed cells may be relevant to understanding the adhesive defects of neoplastic CML progenitors. Several lines of evidence indicate that in Bcr-Abl t ransformed C M L cells, the mitogenic Ras pathway is constitutively activated (Pendergast et al., 1993; Puil et al. , 1994; Tauchi et al. , 1994, Sawyers et al. , 1995). Recent data suggest that the small Ras related GTP-binding proteins of Rho family are required for Ras transformation (Ridley and Hall, 1992; Ridley et al. , 1992; Qiu et al. , 1995a, 1995b; Khosravi-Far et al. , 1995). In Bcr-Abl 34 t ransformed cells several cytoskeletal and focal adhesion proteins including paxillin and p p 1 2 5 F A K are constitutively and specifically tyrosine phosphorylated (Gotoh et al. , 1995; Salgia et al. , 1995a; 1995b; 1996). In addit ion, in these cells, paxillin is found to be physically associated with Bcr-Abl (Salgia et al. , 1995b). Several integrins, including (31, have been found in the focal adhesion, in some cases physically associated with p p 1 2 5 F A K (Craig and Johnson, 1996). The integrin-mediated activation of the Ras/MAP kinase pathway has also been found to involve p p 1 2 5 F A K (Schlaepfer et al., 1994). Taken together these data suggest that in Bcr-Abl t ransformed cells one or more of Rho pathways regulating the cytoskeletal rearrangement may be altered and contribute to the adhesive defects of these cells and their state of t ransformation. Alteration of the adhesive properties of neoplastic CML progenitors may, therefore, be the result of an abnormal function of cytoskeletal and FA proteins, including p i integrin or other adhesive molecules that are normally l inked to the cytoskeletal proteins. Bcr-Abl is directly linked to the cytoskeletal protein actin (McWhirter and Wang , 1993), and the presence of several motifs within Bcr have been found to be involved in actin binding or to have various activities towards the small GTP-binding proteins of the Rho family Rac, Rho and Cdc42 that are involved in cytoskeletal rearrangement (Diekmann et al. , 1991 ; McWhirter and W a n g , 1991 ; Chuang et al . , 1995). The presence, or alternatively, the alteration of theses motifs in the Bcr-Abl chimeric protein may lead to cytoskeletal abnormali t ies and ultimately defective adhesive properties. However, these molecular changes have not yet been investigated in Bcr-Abl t ransformed cells of hematopoiet ic origin. In addit ion, the potential separate role of Bcr in the alterations of the cytoskeleton observed in Bcr-Abl t ransformed cells has not been evaluated. In contrast to stromal layers established from marrow samples f rom C M L patients, those f rom some A M L patients have been found to be defective in their 35 cellular components and their ability to support normal hematopoiesis. Inhibitory factors have also been found in the supernatant of such AML-der ived stromal layers (Greenberger, 1992; Mayani et al. , 1992). FACS analysis has shown that CD34+ cells f rom A M L patients may not express ICAM-1 and variably express the oc2(31 and (32 integrins (Reuss-Borst et al. , 1992; Liesveld et al. , 1993). A M L cells have also been found to bind fibroblasts via a5(31 (Kortlepel et al. , 1993). In all cases examined, A M L cells have been found to express high levels of CD44 (Kortlepel et al., 1993; Liesveld et al. , 1993). A lack of detai led information on the adhesion molecules on leukemic A M L cells or their function is largely due to the heterogeneous nature of A M L and the difficulties encountered in obtaining reproducible quantitative data for defined compartments of A M L progenitor cells. 1.3 CD44 CD44 is a widely expressed cell surface glycoprotein involved in diverse biological processes, some of which are associated with malignant t ransformat ion. The complexity of CD44 function is matched by and possibly related to the complex structure of the CD44 gene. The presence of multiple variably spliced exons enables the production of a large number of isoforms, which vary even further according to their post-translational modif ications. Not surprisingly, these features of CD44 have captured a broad interest in the scientific communi ty and have stimulated a large and rapidly expanding literature dealing with this molecule. A comprehensive review of CD44 is thus well beyond the scope of this thesis. Accordingly, the remainder of this introductory chapter will focus on a discussion of that which is known about C D 4 4 relevant to its involvement in the regulation of normal and leukemic hematopoiesis. 36 1.3.1 S t r u c t u r e a n d Regu la t i on of E x p r e s s i o n C D 4 4 G e n e CD44 is encoded in both the human and murine genomes by a single copy gene that is highly conserved between the two species (Lesley et al . , 1993a). In humans this gene is located on the short arm of chromosome 11 (Goodfel low et al, 1982; Franke et al. , 1983) and in mice it is found on chromosome 2 (Colombatt i et al., 1982). The human CD44 gene (shown diagramatical ly in Figure 3) comprises approximately 60 Kb of genomic DNA and contains twenty exons (Screaton et al. , 1992; Tolg et al. , 1993). Twelve of the twenty exons are known to undergo alternative splicing (Figure 3), and eighteen distinct transcripts have thus far been described (Screaton et al. , 1992; Tolg et al . , 1993). The extracellular region of the standard (s) form of CD44 (CD44s; also referred to as the hematopoietic form, CD44H) is encoded in its entirety by exons that do not appear to undergo alternative splicing (designated as exons 1s to 7s) and are therefore found in all forms of CD44. The t ransmembrane region is encoded by the exon 8s and the intracellular region by either exon 9s or 10s, the latter two exons being alternatively spliced to generate either a short (9s; 3 amino acids) or a long (10s; 70 amino acid) cytoplasmic tail (Goldstein and Butcher, 1990). In addit ion to the standard exons, there are ten other alternatively spl iced exons (1v to 10v; v indicating "variant") that variably insert sequences within a single site in the extracellular domain of CD44 (between exons 5s and 6s) to generate a large number of higher molecular weight isoforms (Figure 3) . The occasional use of an alternate splice donor site in exon 5s and splice acceptor site in exon v3 further add to the variety of CD44 transcripts (Screaton et al. , 1992; Cooper et al. , 1992; Tolg et al. , 1993). The human v1 exon has been found to contain an in-frame stop codon (Screaton et al., 1992). 37 EXTRACELLULAR VARIABLE DOMAIN TAIL EXONS < • l l l l l l l l l l HI 3 d 2 A b : Hya lu ronan -b i n d i n g s i te CHO-B— <§) CHO CHO CHO CD44 F igu re 3. S t r u c t u r e o f t he CD44 Gene a n d t h e C o r r e s p o n d i n g P r o t e i n . Potential sites of carbohydrate attachment are indicated as CS (chondroit in sulfate) and CHO (Asparagine-l inked glycans). T M designates the exon encoding the t ransmembrane region. 38 The promoters of both the human and mouse CD44 genes have been found to possess several putative c/'s-acting elements. Most of these are spaced somewhat differently in the two species, although the TATA box and an AP-1 site are located at nearly identical positions with respect to the coding region of the CD44 gene (Herrlich et al., 1993). It has been reported that either overexpression of the oncogenes jun or ras or t reatment of cells with phorbol ester results in an enhancement of CD44 transcript ion (Herrlich et al. , 1993). The AP-1 site has been found important in this induction since a single point mutation within the active AP-1 site impaired responsiveness to transcriptional activation by jun, ras, and phorbol ester (Herrlich et al. , 1993). Studies of human T cell lines have found overexpression of a murine homeobox gene HlxXo have an effect (reducing) on the expression of CD44 on these cells which was associated with a concomitant reduction in their adhesiveness. This has led to the more general suggest ion that homeobox transcript ion factors may have a role in the regulation of CD44 transcript production (Allen and Adams, 1993). Little is known about the mechanisms that regulate the splicing of the CD44 transcript. Overexpression of ras, which is associated with increased promoter activity, has also been found to be associated with a moderate increase in the ratio of CD44v/CD44s transcripts as well as low level expression of the v6 epitope, suggest ing a possible role for ras in the regulation of CD44 transcript splicing (Hofmann et al. , 1993; Jamal et al. , 1994). The generat ion of the v6 isoform of CD44 in activated T cells has also been found to coincide with changes in the expression of newly identified splicing factors, al though the role of these factors in v6 isoform generation is not yet clear (Screaton et al . , 1995). Two allelic forms of the mouse CD44 gene, also called Pgp-1 (phagocyte glycoprotein-1) have been identified and have been designated Pgp-1.1 (found in BALB/c mice) and Pgp-1.2 (C3H and C57BL/6 mice) (Hughes and August , 39 1981 ; Trowbr idge et al, 1982; Hughes et al, 1983; Nottenburg et al . , 1989; Zhou et al . , 1989). Amino acids 1-187 of murine CD44 are 8 5 % identical to the human and baboon proteins and an even greater identity between the three species is found in their t ransmembrane and cytoplasmic regions (90-98% identical) (Nottenburg et al. , 1989; Zhou et al., 1989). This contrasts with the sequence def ined by amino acids 187-259 (the membrane proximal extracellular region) which is strikingly dissimilar between mouse and primates (35% identical) (Nottenburg et al. , 1989; Zhou et al., 1989). CD44 Pro te in M e m b r a n e - B o u n d CD44 The molecular cloning of CD44 has revealed several domains present within this molecule (shown diagramatically in Figure 3). CD44s (which contains no variable regions) encodes a 341 amino acid t ransmembrane protein with a predicted molecular weight of 37-38 Kd and containing an extracellular domain of 248 amino acids, a t ransmembrane spanning domain of 21 amino acids, and a 72 amino acid cytoplasmic domain (Stamenkovic et al. , 1989; Goldstein et al. , 1989; Nottenburg et al. , 1989; Zhou et al. , 1989). Al though rare, a CD44 isoform with a t runcated cytoplasmic domain of only 3 amino acids has also been identified (Goldstein and Butcher, 1990). As discussed in the previous sect ion, higher molecular weight isoforms can also be generated by the insertion of various sequences encoded by alternatively spliced exons within a unique site in the membrane proximal region of the extracellular domain (Dougherty et al. , 1 9 9 1 ; Hofmann et al. , 1991 ; Stamenkovic et al., 1991 ; Jackson et al. , 1992). A stretch of 90 relatively hydrophobic residues comprises the amino-terminus of the CD44 polypeptide. This amino-terminal domain has homology to carti lage link proteins and other proteoglycans, and mediates interactions of 40 CD44 with glycosaminoglycans. In particular, this region binds to HA (Goldstein et al. , 1989; Nottenburg et al. , 1989; Stamenkovic et al. , 1989; Zhou et al. , 1989) in spite of the fact that it lacks the conserved basic residues found in carti lage link and other proteoglycan core proteins traditionally thought to be involved in HA recognit ion (Neame and Barry, 1993). The carti lage link homologous region of CD44 contains six cysteine residues that may be linked by disulfide bounds to form a single globular domain, although the crystal structure of CD44 has not yet been determined. The entire extracellular portion of human CD44 contains six potential sites for A/-linked glycosylation, five of which lie in the carti lage link homologous region (Stamenkovic et al. , 1989; Goldstein et al., 1989). The membrane proximal region of CD44 contains several serine and threonine residues that may serve as sites of O-linked glycosylation as well as four serine-glycine motifs that may function as sites of chondroit in sulfate or heparan sulfate at tachment (Goldstein et al. , 1989; Stamenkovic et al. , 1989; Brown et al . , 1991). Al though the core CD44 polypeptide is approximately 37 kD, the most abundant form of CD44 expressed on hematopoietic cells (CD44H) has an apparent molecular weight of 85-90 Kd (Jalkanen et al. , 1987; Underhil l et al. , 1987; Wayner and Carter, 1987). Variable degrees of glycosylation and chondroit in and heparan sulfate attachments are thought to account for this addit ional molecular mass (Lesley et al. , 1993). Moreover, a diversity in the types of post-translational modifications may contribute to the production of cell type-specif ic forms of mature CD44 (Carter and Wayner, 1988; Jalkanen et al. , 1988; Kansas et al. , 1989; Naujokas et al. , 1993). The cytoplasmic region of CD44 contains six serine residues that are subject to phosphorylat ion (Lesley et al., 1993). Five of these serine residues are highly conserved among all mammals . 41 S o l u b l e CD44 Several groups have reported the presence of substantial concentrat ions of a soluble form of CD44 (up to 5 u,g/ml) in the p lasma of normal individuals (Picker et al . , 1989; Bazil and Horejsi, 1992; Yang and Binns, 1993; Katoh et al. , 1994; Ristamaki et al., 1994). Al though the cellular origins and the mechanisms that generate these soluble molecules are unclear, the shedding of CD44 f rom the surface of both cell lines and primary cells has been described (Bazil and Horejsi, 1992; Yang and Binns, 1993; Katoh et al. , 1994). The treatment of peripheral blood cells with phorbol myristate acetate (PMA), TNF-oc, or ant i -CD44 antibodies (mAb MEM-85) (Bazil and Horejsi, 1992) as well as the intraperitoneal injection of ant i -CD44 antibodies (mAb IRAWB14) have all been found to result in an increased release of CD44 from the cell surface (Camp et al. , 1993). The presence of soluble CD44 in the circulation also appears to be inf luenced in vivo by both tumor growth and immune activity (Katoh et al. , 1994). Al though the role of soluble CD44 has not yet been determined, this molecule retains its functional activity as demonstrated by its ability to bind HA (Katoh et al. , 1994). Cel l S u r f a c e D i s t r i b u t i o n The independent cloning of CD44 by several groups (Stamenkovic et al. , 1989; Goldstein et al. , 1989; Nottenburg et al. , 1989; Zhou et al . , 1989; Idzerda et al . , 1989) led to the rapid recognition of its identity with many leukocyte surface antigens previously characterized by their reactivity with a variety of different antibodies [Pgp-1 , Ln(Lu)-related p80, ECM III, Hermes, HUTCH-1 and gp-85] (Underhil l et al. , 1987; Carter and Wayner, 1988; Gallatin et al . , 1989; Kansas et al. , 1989; Pals et al. , 1989; Picker et al. , 1989). Given the initial identification of CD44 as a leukocyte surface protein, its expression on lymphocytes and macrophages has, not surprisingly, been particularly well character ized. However, CD44 has also been detected on many other cell types 42 including f ibroblasts, epithelial cells, and keratinocytic cells, as well as the majority of other types of blood cells (Lesley et al. , 1993). Historically, the 85-90 KD CD44 molecule found on the surface of lymphocytes has been referred to as the hematopoiet ic form (CD44H). A similar form of CD44 is found on the majority of hematopoiet ic cells as well as on fibroblasts and astrocytes. Because of the broad t issue distribution of this form of CD44, this molecule has been referred to as the common form of CD44. CD44 splice variants, primarily containing exons v8-v10, have been found to be expressed as of day 6.5 during murine embryogenesis (Fenderson et al . , 1993; Ruiz et al. , 1995). Fenderson et al. (1993) have also examined the distribution of HA during embryogenesis by means of a soluble CD44-immunoglobul in fusion protein. This immunohistochemical analysis has found HA to also be present beginning at approximately day 6 in the basement membranes of the primitive ecto- and endoderm within the yolk sac cavity. Mult ipotent hematopoiet ic stem cells isolated from day 11 mouse embryos were found to express no detectable CD44 protein, as opposed to the earlier embryonic t issues (Huang and Auerbach, 1993). Similar results were obtained f rom studies of mult ipotent undifferentiated embryonic stem (ES) cells, al though two CD44 transcripts were detected (Haegel et al. , 1994). However, the differentiation of ES cells in culture was shown to be associated with de novo expression of exon v10-containing-CD44 isoforms at both the RNA and surface protein levels (Haegel et al., 1994). Several CD44 species capable of binding HA have also been found in the st roma of the chorionic villi of human placenta (St. Jacques et al., 1993). The distribution of CD44 during development has generally been found to mirror that of HA, which has led to the suggest ion that CD44-mediated interactions may be involved in the process of embryogenesis (Underhil l et al. , 1992; Ruiz et al. , 1995). 43 1.3.2 CD44 L i g a n d s a n d In te rac t ions w i t h Ex t race l lu la r Mat r ix H y a l u r o n a n As discussed in section, the homology of the amino terminal region of CD44 to carti lage link proteins and proteoglycans initially suggested that CD44, like these related molecules, may have HA binding activity (Goldstein et al. , 1989; Stamenkovic et al. , 1989). In addit ion, Underhil l et al. (1987) had descr ibed a HA receptor similar to CD44 with respect to its size, cellular distr ibution, and cytoskeletal interactions (Culty et al. , 1990). Concurrent studies found the adhesive interactions of CD44+ B cells with stromal cell lines to be HA-mediated and inhibitable with ant i-CD44 antibodies (Miyake et al. , 1990b). Formal proof of the specificity of CD44 for HA was the demonstrat ion that a soluble CD44- immunoglobul in fusion protein bound to rat lymph node high endothelial venules in a HA-dependent manner (Aruffo et al. , 1990). Interestingly, an ant i-CD44 mAb (Hermes-3) that specifically blocks the binding of lymphocytes to mucosal HEV was not found to interfere with stromal cel l-CD44 binding to purified HA (Jalkanen et al. , 1987; Goldstein et al. , 1989). Nevertheless, the treatment of HEV or stromal cells with hyaluronidase as well as the addit ion of soluble HA or ant i-CD44 antibodies (mAbs KM201 and K-3) have been shown to specifically and completely inhibit their interaction with lymphocytes (Aruffo et al. , 1990; Culty et al. , 1990; Miyake et al. , 1990a; 1990b). Site-directed mutagenesis and truncation studies of CD44 have identified two clusters of basic amino acid residues, one within and one outside of the carti lage link homologous region, to be essential for HA-binding (Peach et al. , 1993; Yang et a l . , 1994). Recently, CD44 and two related HA-binding molecules (RHAMM and Link protein) have been found to share the same HA-binding motif [B(X 7 )B] which is located within each of these two clusters (B representing one of 44 two basic amino acid residues, arginine or lysine, that f lank a stretch of 7 non-acidic amino acids) (Yang et al. , 1994). The HA-binding potential of higher molecular weight isoforms of CD44 has been a subject of some controversy. This has emanated from contradictory reports describing either the inability (Stamenkovic et al. , 1991) or the ability (He et al. , 1992; Dougherty at al. , 1994) of exon v8-v10-containing isoforms to bind to HA. This issue has been carefully addressed in a recent study by Bennett et al . (1995b) which found that al though several CD44 variants are able to bind HA, this interaction is much weaker than that mediated by CD44H. The reduced HA binding of higher molecular weight isoforms was shown to be due, at least in part, to O-linked carbohydrate moieties which are added to the region encoded by variably spliced exons, rather than to the additional peptide component of these glycoproteins. Glycosylation of CD44 therefore appears to be an important factor in the regulation of its l igand-binding activity (Bennett et al . , 1995b). Al though HA is present at high concentrat ions in the extracellular matrix of the bone marrow (Gallagher et al. , 1983; Spooncere t al. , 1983; Underhil l , 1992), evidence of HA binding to CD44 expressed on primitive hematopoiet ic cells is l imited. As will be discussed in Section, the amount of CD44 expressed on a given cell does not necessari ly correlate with its HA-binding activity. Rather, this appears to be regulated by other mechanisms which remain poorly def ined. Other CD44 L i g a n d s In addit ion to HA, CD44 has the ability to bind several other extracellular matrix proteins such as type I and type VI collagen (Wayner and Carter, 1987; Faassen et al . , 1992), f ibronectin, laminin (Jalkanen and Jalkanen, 1992), mucosal vascular addressin (Streeter et al. , 1988; Goldstein et al. , 1989), and CD44 itself (St John et al. , 1990; Droll et al. , 1995). CD44 seems to have a low 45 affinity for type I col lagen, and at least in some cells only the chondroit in sulfate-bound form of CD44 may be responsible for such binding (Faassen et al. , 1992). Chondroi t in sul fate-bound CD44 also mediates interactions with f ibronectin by binding to the heparin-binding site of this molecule (Jalkanen and Jalkanen, 1992). Overexpression of CD44 has been found to promote homotypic aggregat ion (St. John et al., 1990), and ant i-CD44 treatment (mAb H4C4) of hematopoiet ic cells has also been shown to have this same effect (Belitsos et al. , 1990). Homotypic aggregation of hematopoietic cells and fibroblasts may also occur through CD44-HA interactions if the cells express CD44 and are coated with HA (Lesley et al. , 1990). Recently such homotypic aggregation has been shown to involve CD44 binding to itself (Droll et al. , 1995). A recently described ligand for CD44 is the heavily glycosylated secreted glycoprotein, serglycin (Toyama-Sorimachi and Miyasaka, 1994). The expression of serglycin seems to be restricted to cells of the yolk sac, hematopoiet ic t issues, and some tumor cell lines (Stevens et al. , 1988). Serglycin has been implicated in stabilizing and packaging basically charged proteases, cytolytic proteins, and cytokines within secretory granules, and may also have a role in their transport to the extracellular environment. The CD44 binding elements on serglycin are glycosaminoglycans containing chondroit in-4-sulfate. Interestingly, the binding site of serglycin on CD44 appears to localize to the amino terminal region, close to and/or overlapping with the binding site of HA (Toyama-Sor imachi et al. , 1995). Thus an ant i-CD44 antibody recognizing the HA-binding site of CD44 can inhibit its ability to bind serglycin and, conversely, serglycin strongly inhibits the binding of CD44 to HA (Toyama-Sor imachi et al . , 1995). In addit ion, serglycin appears to bind an activated form of CD44 on cytotoxic T cell (CTL) clones, as has been previously described for HA, and this interaction results in an increased cytolytic activity (Toyama-Sor imachi et a l . , 1995). 46 Another newly documented ligand of CD44 is the cytokine Osteopont in or ETA-1 (early T lymphocyte activation 1) (Weber et al., 1996). ETA-1 is an extracellular phosphoprotein secreted by activated T cells, osteoblasts, macrophages and other cell types (Patarca et al. , 1990). ETA-1 was found to bind specifically to CD44 and this interaction was blocked by antibodies recognizing the HA binding site of CD44 (mAbs KM81 and IM7). Interactions between CD44 and E T A - 1 , but not CD44 and HA, were found to be involved in the chemotact ic movement of CD44 transfected cells (Weber et al. , 1996). R e g u l a t i o n of E x p r e s s i o n a n d L i g a n d - B i n d i n g Cellular regulation of the l igand-binding function of adhesion receptors controls the t iming and specificity of cell-cell and cell-substrate interactions. CD44 shares this property of regulated function with many other adhesion molecules. Moreover, the broad distribution of CD44 and its l igands necessitates a tight regulation of CD44-l igand-binding abilities. Studies on the regulation of CD44 binding activity have raised the possibility that the cytoplasmic domain of CD44 may contribute to its functional state v ia interactions with other molecules. The CD44 cytoplasmic domain is known to associate with actin and several cytoskeletal proteins including ezrin, radixin and moesin of ERM family and ankyrin (Lacy and Underhil l , 1987; Carter and Wayner , 1988; Kalomiris and Bourguignon, 1988, Tsuki ta, 1994), as well as certain signal t ransducing proteins such as protein kinase C (PKC; Carter and Wayner, 1988; Kalomiris and Bourguignon, 1989). In addit ion, the cytoplasmic domain of purified CD44 has been found to exhibit a GTP-binding and GTP-ase activity that enhanced its interaction with ankyrin (Lokeshwar and Bourguignon, 1992). Deletion and truncation studies have shown that the absence of the cytoplasmic domain of CD44 impairs its HA-binding ability and also abrogates its 47 binding to ankyrin (Lesley et al., 1992, 1993, 1995, Lokeshwar, 1994). However, the exact role of this domain in regulating CD44-HA binding and / or CD44-cytoskeleton association is still uncertain. In the absence of the cytoplasmic domain, CD44 activating antibodies have been found to reestablish CD44-HA binding (Thomas et al., 1992, Lesley et al., 1993b; Hathcock et al., 1993; Perschl et al . , 1995). It appears that activating antibodies do so without altering the level of CD44 expression on the cell surface. Rather, this ant ibody-induced activation seems to result f rom an induced conformational change in the CD44 molecule or a change in their cell surface distribution. This may involve (or lead to) clustering of multiple CD44 molecules and / or their association with other cell surface or intracellular components since monovalent Fab f ragments of these "activating" ant i -CD44 antibodies were much less efficient in the level of HA-binding activity they were able to stimulate as compared to intact antibody. These results suggest that activation of the HA binding function of CD44 requires multivalent binding and/or cross-l inking of CD44 to form multimeric aggregates on the cell surface (Perschl et al., 1995). Addit ional studies have shown that these activating antibodies could induce HA binding activity on prefixed cells that are permeable to trypan blue and propidium iodide indicating that intracellular signaling events were not necessary (Perschl et al. , 1995). The fact that cells transfected with a truncated form of CD44 can bind to immobil ized HA (Lesley et al. , 1992) indicates that immobil ized HA may itself activate the HA-binding function of CD44 possibly by stabilizing low affinity interactions through a clustering of CD44 molecules into multivalent receptors. The impairment of ankyrin-binding in cytoplasmic truncated form of CD44 suggests that intracellular binding of the cytoplasmic domain of CD44 to ankyrin may be required for its extracellular adhesion funct ion. The phosphorylat ion of the cytoplasmic domain of CD44 appears to be important for CD44-HA binding. Point mutations of either of the serine residues 48 (ser 325 or ser 327) that are found within the cytoplasmic domain of CD44 expressed on T cells abol ished both the CD44-mediated HA-binding activity of these cells as well as their ability to undergo l igand-induced receptor modulat ion. In transfected cells, wild-type CD44 accumulates in molecular clusters as a result of l igand binding whereas mutated CD44 redistributes to a single polar cap (Pure et al . , 1995). It has been proposed that phoshorylation of serine residues (Carter and Wayner , 1988) within the cytoplasmic domain of CD44 can regulate its associat ion with the cytoskeleton (Kalomiris and Bourguignon, 1989; Camp et al., 1991). However, it is not clear whether phosphorylat ion occurs in the cytoplasmic domain of CD44 and regulates its HA-binding and cytoskeleton associat ion or whether the phosphorylat ion of cytoskeletal proteins associated with the CD44 cytoplasmic domain are involved in the regulation of CD44- l igand interaction. These may also be subject to variation according to type of cell studied. In particular, in migrating macrophages, the majority of CD44 has been found to be phosphorylated but unbound to the cytoskeleton (Camp et al. , 1991 b). In addit ion, mutation of the serine residues (ser 323 and ser 325) within the CD44 cytoplasmic domain that are responsible for the majority of CD44 phosphorylat ions seen in epithelial cells does not affect the localization of CD44 nor its associat ion with the cytoskeleton in epithelial cells (Neame and Isacke, 1993). This has led to the hypothesis that phosphorylat ion of a linker protein, for example ERM family members (Tsukita et al., 1994), and not CD44 itself may regulate the cytoskeletal associations of CD44 (Neame and Isacke, 1992; 1993). Phosphorylat ion of cytoskeletal proteins associated with the CD44 cytoplasmic domain may regulate the clustering of CD44 on the cell surface and increase its affinity for HA-binding. The PKC-induced phosphorylation of the intracytoplasmic tail of CD44 has been associated with an enhancement of its ankyrin-binding funct ion in vitro 49 (Kalomiris and Bourguignon, 1989). It has also been found that phorbol ester t reatment of some cells can increase their CD44-mediated HA-binding ability. However, it is not yet clear whether this is simply a consequence of an induced increase in the level of CD44 expression or whether PKC activation of phorbol ester also has a direct effect on the phosphorylat ion of the cytoplasmic domain of CD44 and its subsequent association with the cytoskeleton (Lesley et al. , 1990; Hyman et al . , 1991 ; Kalomiris and Bourguignon, 1989; Murakami et al . , 1994; Morimoto et al. , 1994). Taken together, these data support the hypothesis that phosphorylat ion of residues within the cytoplasmic domain of CD44 may be a prerequisite for enabl ing CD44 to bind soluble HA and PKC activation may enhance HA binding of CD44 by upregulating the expression of CD44 and / or by inducing the phosphorylat ion of cytoskeletal proteins that bind CD44 cytoplasmic domain. The cytoskeleton appears to play an important role, in regulating CD44- l igand binding, al though the exact mechanism of action is still unknown. An intact cytoskeleton may be required, at least in some cells, for the redistribution and change of CD44 molecules on the cell surface from a low affinity to a high avidity CD44- l igand binding. Alternatively, the rearrangement of cytoskeletal proteins may be necessary for the phosphorylat ion of the cytoplasmic domain of CD44 that has been found to be important for CD44-l igand binding. It has been hypothesized that the HA-binding function of CD44 is either constitutive, inducible, or non-existent on a given CD44-expressing cell and studies of variant clones derived from a single parental cell line has demonstrated that all of these possibilities can exist (Lesley et al. , 1993b). In addit ion it has been shown that N-linked glycosylation of the amino terminal region of CD44 can reduce its recognition of HA (Lesley et al., 1995; Katoh et al, 1995; Bennett et al . , 1995b). These f indings provide an explanation for previous observat ions showing an inverse relationship between the higher HA binding 50 ability of IL-5-activated B cells and their decreased glycosylation of surface CD44 (Hathcock et al. , 1993). Thus, post-translational modif ications, and in particular glycosylation of CD44, may regulate CD44 binding to its l igand and this, in turn, can be affected by even a single mutation in the hyaluronan recognition site of CD44. The level of CD44 expression, as well as the type and isoform of CD44 expressed are also subject to regulation and vary according to cell type and its state of activation and / or proliferation and / or differentiation (these effects will be discussed in Sections and below, and also later in Sect ions 1.3.3 and 1.3.4). Alternative splicing of CD44 has also been descr ibed in different cells (Lesley et al., 1993a; and in 1.3.4). However, the exact mechanism of regulation of CD44 transcription and splicing is unknown. All of these may contribute to the variation in the CD44-l igand-binding ability. For example phorbol ester stimulation of PKC induces or increases CD44 expression in different cells resulting in an enhancement of CD44-l igand binding and/or may induce phosphorylat ion of cytoskeletal proteins involved in the regulation of CD44-HA binding (Lesley et al., 1990, Murarakami et al. , 1994; Morimoto et al . , 1994; and data presented in this thesis). 1.3.3 Cel lu la r F u n c t i o n s L y m p h o c y t e s CD44 has been implicated in numerous adhesion-dependent funct ions. For example, ant i -CD44 antibodies have been found to differentially and specifically block lymphocyte-binding to lymph nodes, mucosal (Hermes-3) or synovial high endothelial venules (Jalkanen et al., 1987; 1988). In fact, it was the ability of the Hermes-3 antibody to specifically inhibit the binding of lymphocytes to mucosal high endothelial venules that was subsequent ly 51 exploited in a strategy to clone the CD44 cDNA (Goldstein et al. , 1989). However, others have since found that CD44 is not involved in the normal trafficking of circulating lymphocytes, although it may be necessary for leukocyte extravasation into inf lammatory sites involving nonlymphoid t issue (Camp et al. , 1993). T L y m p h o c y t e s CD44 has been described as a differentiation, adhesion, and cost imulatory molecule on T cells (Haynes et al., 1989; Lesley et al. , 1993). Prothymocyte progenitors that home to and repopulate the thymus can be discerned by their high level cell surface expression of CD44 (Trowbridge et al . , 1982; W u et al. , 1993; Marquez et al. , 1995). CD44 has also been identified as a T cell differentiation marker expressed by both immature thymocytes and mature activated memory T cells (Trowbridge et al., 1982; Lynch et al. , 1987; Lynch and Ceredig, 1988; Lesley et al. , 1988; Horst et al. , 1990, Godfrey et al . , 1993; W u et al. , 1993; Marquez et al., 1995). The differentiation of CD4+CD8+ thymocytes into CD4CD8" populations is concomitant with a decrease in CD44 expression which is then fol lowed by an upregulation during the subsequent differentiation into mature CD4 or CD8 single positive T cells (Ceredig, 1987; Lynch et al. , 1987; Lynch and Ceredig, 1988; Butterfield et al. , 1989; Godfrey et al . , 1993). The activation of mature T cells with PMA also results in a modulat ion of C D 4 4 expression and / or association with cytoskeleton (Haegel and Ceredig, 1991 ; Geppert and Lipsky, 1991). Ant igen cross-l inking studies with either ant i-CD3 or ant i -CD2 antibodies combined with various ant i-CD44 antibodies (mAbs H90, N IH44-1 , A3D8, and A1G3) has been shown to result in enhanced T cell proliferation, suggest ing that CD44 may function as a costimulatory molecule (Huet et al, 1989; Shimizu et al. , 1989; Denning et al. , 1990; Seth et al., 1991). However, an individual ant i -CD44 52 antibody (mAb 212.3) has been found to inhibit ant i -CD3-induced activation of T cells (Rothman et al. , 1991). The signif icance of these opposing functions of ant i -CD44 antibodies remains unclear. Interestingly, the engagement of CD44 by antibodies or HA-binding has been found to inhibit apoptosis of T cells induced by ant i-CD3 mAbs or dexamethasone without any change in the level of bcl-2 expression (Ayroldi et al. , 1995). The triggering of T cells through CD44 also promotes their homotypic adhesion via LFA-1 through a process that requires PKC activation and an intact cytoskeleton (Koopman et al. , 1990). CD44-mediated T cell activation has also been shown to increase binding to kerat inocytes and endothelial cells (Bruynzeel et al., 1993; Toyama-Sor imachi et al . , 1993). CD44 expression has been found to be upregulated on C D 8 + T cells during an allogeneic response (Mobley and Dailey, 1992) and protective C D 8 + T cell clones against malaria (in contrast to nonprotective clones) were found to express high levels of CD44 that, when cross-l inked with ant i -CD44 antibodies (mAbs KM201 and KM703) , induce LFA-1-mediated homotypic aggregat ion (Rodrigues et al. , 1992). In addition to its role in mounting a cytolytic T cell response, the activation of T cells through CD44 also appears to be involved in st imulating IL-2 production and the induction of T cell proliferation through a tyrosine kinase-dependent pathway similar to the one seen fol lowing activation via CD3 (Galandrini et al. , 1993). The In vivo activation of CD44 on lymphocytes has been demonstrated in a murine al logeneic response. About 1/3 of splenic T cells isolated 6-9 days post- immunizat ion were found to express very high levels of CD44 and to bind HA (Lesley et al. , 1994). This HA-binding fraction included both CD4 and CD8 single positive T cells and contained almost all of the cytotoxic activity. However, it appears that HA recognition may not be directly involved, since cytotoxicity 53 could not be inhibited by HA or by ant i-CD44 antibodies recognizing the HA-binding site (Lesley et al. , 1994). B L y m p h o c y t e s CD44 has also been found to be a differentiation and activation marker on B cells which is upregulated with maturation (Kansas and Dailey, 1989; Murakami et al . , 1991a; 1991b; Kincade, 1992; Camp et al . , 1991a; Hathcock et al., 1993). The in vitro polyclonal stimulation of B cells through their cell surface immunoglobul in (Ig) induces a substantial increase in CD44 expression. Short and long-term ant igen-primed and Ig-secreting B cells have also been found to express high levels of CD44, whereas naive non-primed B cells express very low levels of CD44 (Camp et al. , 1991a). Similarly, B cells cultured for several days in the presence of Interleukin 5 (IL-5) results in the generat ion of daughter populat ions that differ markedly in their expression of CD44, with the more highly proliferating and differentiated B cells expressing higher levels of CD44 (Murakami et al. , 1990). Moreover, only this subpopulat ion was found to use CD44 to bind HA on coated dishes (Murakami et al. , 1990). Act ivated B cells f rom mice undergoing graft-versus-host (GVH) reactions, and not f rom normal mice, have been shown to express very high levels of CD44 and demonstrate HA-binding activity (Murakami et al. , 1991a). Interestingly, these cells are also the ones that are activated and produce immunoglobul in (Murakami et al . , 1991a; 1991b). Taken together, these f indings suggest that CD44 is an adhesion molecule that allows activated B cells to interact with HA. The importance of HA has been indicated by the f inding that both the addit ion of exogenous HA and pretreatment of either the stromal cells or the B cells with hyaluronidase can prevent their interaction with one another (Miyake et al. , 1990b). 54 Natura l Ki l ler Cel ls CD44 may also play a role in the function of NK cells. This was first suggested by studies in which the treatment of dogs with a monoclonal ant ibody (mAb S5) that recognizes CD44 facilitated their ability to be successful ly t ransplanted with bone marrow from an unrelated donor (Sandmaier et al. , 1990). Subsequent studies indicated that the ant i-CD44 antibody treatment rendered the recipient NK cells radiation-sensitive, thereby improving the engraftment of foreign cells (Tan et al. , 1993). This activation of NK cells was found to result in increased effector-target conjugate formation and TNF-oc production (Campanero et al. , 1991). A role for CD44 in NK cell ontogeny has been suggested by studies in LTC that have found a requirement for CD44-HA interactions for the subsequent generation of mature cytotoxic effector cells f rom pre-NK precursors (Delfino et al. , 1994). M o n o c y t e s The activation of monocytes through an ant i-CD44 antibody or hyaluronan-binding has been found to stimulate the release of a number of cytokines (Denning et al. , 1990; Webb et al, 1990; Noble et al . , 1993). This activation of monocytes was shown to enhance their cooperat ion with T cells by facil itating cell-cell binding and by inducing the monocytic production of IL-1 and T N F - a (Denning et al. , 1990; Webb et al, 1990). The ant i-CD44 tr iggering of monocytes was found to be dependent upon cross linking of the antigen (Webb et al, 1990). Pr im i t i ve H e m a t o p o i e t i c Cel ls CD44 has been found on all types of mature blood cells (Lesley et al. , 1993) and is specifically involved in mediating the adhesion of murine platelets to HA (Koshiishi et al., 1994). Murine CD44 has been shown to be present on 55 more primitive hematopoiet ic cells, including progenitors detectable in vivo as day-10 CFU-S and those capable of forming colonies of granulocytes and macrophages in vitro (Trowbridge et al. , 1982; Hughes et al. , 1983; Spangrude et al. , 1989). Antibody-staining and cell sorting of human bone marrow has documented the presence of CD44 on the majority of light density mononuclear, CD34+ cells and on all types of l ineage-restricted progenitors (Lewinsohn et al. , 1990; Kansas et al. , 1990 and data presented in this thesis). In addit ion, Kansas et al . (1990) have found CD44 to be downregulated during erythroid differentiation. CD44 has been thought to be actively involved in regulating hematopoiesis since it was found that the addition of an ant i -CD44 ant ibody (mAb KM201) could inhibit the production of nonadherent cells in murine B lymphoid (Whitlock-Witte) or myeloid (Dexter) LTC (Miyake et al. , 1990a). It has also been reported that this profound inhibition occurred only if the ant i -CD44 antibody was added within the first week of initiating the cultures (Kincade, 1992), indicating that the antibody might be interfering with a crucial CD44-dependent interaction of primitive hematopoietic cells with stromal cells that occurs early in LTC. The antibody responsible for this effect, K M 2 0 1 , has subsequent ly been found to recognize the HA binding site of CD44, thereby implicating HA in this interaction (Miyake et al. , 1990b). In the human system, the addit ion of ant i -CD44 antibodies to LTC has subsequent ly been found to reduce the production of granulopoietic colony-forming cells (CFU-GM) after 5 weeks (Gunji et al, 1992). Interestingly, the antibody used in these human studies was found to bind within the proteoglycan homologous domain of CD44, but does not inhibit HA recognition (Culty et al. , 1990; Peach et al. , 1993). Specific binding of CD44 on myelomonoblast ic cell lines (KG-1 and KG-1a) to immobil ized HA has also been reported. This binding could be prevented when the cells were treated with PKC inhibitors and enhanced by treatment with PKC stimulators, indicating that myeloid CD44 binding to HA likely requires PKC 56 (Morimoto et al. , 1994). However, the mechanism of PKC activity on CD44-ligand binding is still unclear. CD44 has been found to be the major chondroit in-sulfate-modif ied proteoglycan that is immunoprecipi tated from a population of bone marrow hematopoiet ic cells that is enriched for short term lineage-restricted progenitors (Verfaillie et al. , 1994b). This is of particular interest in v iew of studies demonstrat ing the binding of purified CD44 to the FN-C/H II f ragment of f ibronectin through a chondroit in sulfate-dependent, CD44 core- independent mechanism (Verfaillie et al., 1994b). Moreover, the adhesion of clonogenic human hematopoiet ic progenitors to the heparin-binding site, and not the CS-1 site, of f ibronectin can be inhibited by their treatment with ant i -CD44 antibodies. On the other hand, CD44 does not appear to be solely responsible for the adhesion of these cells to C S - 1 , although it may cooperate with VLA-4 in this process (Verfaillie et al. , 1994b). 1.3.4 E x p r e s s i o n a n d Role of CD44 Var ian ts The most common CD44 isoform produced is an 85-90 KD protein, a l though, as already discussed, higher molecular weight isoforms are generated by alternative splicing in a variety of cells (Lesley et al. , 1993). During the past f ive years both the cellular distribution of these isoforms and the c i rcumstances under which they are generated have been major areas of investigation for many groups interested in understanding the functional role of a CD44 repertoire. In this sect ion, some of the evidence describing the involvement of CD44 isoforms in mal ignancy will first be reviewed and, subsequently, proposed roles and c i rcumstances for their generation and acquisit ion of particular functional propert ies will be discussed. CD44 is highly expressed on several tumor cell lines (Knudson et al. , 1989; Underhil l et al . , 1992) and its l igand, HA, has been found in large amounts 57 in several tumors, particularly surrounding metastatic cells. Moreover, interactions between fibroblasts and human tumor cells has been found to st imulate the production of HA (Knudson et al. , 1984). It has therefore been proposed that CD44 may cooperate with HA in facilitating the spread and migration of t ransformed cells. A s s o c i a t i o n of CD44H w i t h M a l i g n a n c y The common form of CD44 as well as the spliced variants have been implicated in mal ignancy. HA-binding of CD44 has often been implicated in these observat ions. For example, the CD44" Burkitt lymphoma Namalwa stably transfected with CD44H cDNA was found to display an increased tumorigenici ty and metastatic behavior in immunoincompetent athymic nude mice (Sy et al . , 1 9 9 1 ; 1992). The tumor formation was suppressed in the presence of soluble CD44H fused to immunoglobul in and coinjected with transfected CD44H+ Namalwa cells suggesting that CD44H was directly responsible for tumor formation (Sy et al. , 1992). The adhesion, motility and invasive behavior of a highly metastatic mouse melanoma cell line on collagen type I also appeared to be mediated through the chondroit in sulfate moiety of CD44H (Fassen et al. , 1992). Migration of human melanoma cells on hyaluronan was correlated with CD44 expression, and competed by soluble hyaluronan or CD44 (Thomas et al. , 1992; 1993). CD44 has been found to function in matrix assembly by organizing, anchoring and binding hyaluronan (Knudson et al . , 1993). The CD44 function in matrix assembly could be of great importance for migrating and proliferating cells. In fact, it has been demonstrated that the capacity of CD44 to mediate tumor cell at tachment to hyaluronan directly determines the rate of formation of tumor mass in vivo (Bartolazzi et al. , 1994). Transformation of colon mucosa to carc inoma has been reported to be correlated with downregulat ion of C D 4 4 H (Tanabe et al. , 1993; Takahashi et al. , 1995). The reintroduction of 58 C D 4 4 H by stably transfection into colon carc inoma cells reduced their growth rate and tumorigenicity in in vitro and in vivo studies (Takahashi et al . , 1995). Both in vitro and in vivo growth correlated with the ability of CD44H to bind hyaluronan and required the CD44 cytoplasmic domain (Takahashi et al. , 1995). A large number of clinical reports have established the association and / or correlat ion between the high expression or conversely the lack of expression of CD44 with progression, tumor metastasis and prognosis in different solid tumors (Cooper and Dougherty, 1995). CD44 Var ian ts in M a l i g n a n c y By cloning CD44 Stamenkovic et al . discovered a distinct CD44 variant highly expressed in carcinomas and only weekly in normal epithelial t issues (CD44E, Stamenkovic et al. , 1989; Brown et al. , 1991). CD44E and a highly related molecule CD44R1 carried an additional 132 amino acids in the membrane proximal region of the extracellular domain and was encoded by the alternatively spliced exons v8-v10 (Stamenkovic et al. , 1989; Dougherty et al . , 1991). Several hematopoietic cell lines (but not all) derived f rom patients with different types of leukemia and lymphoma were found to express the variant-containing isoforms (Stamenkovic et al. , 1989; Dougherty et al. , 1991). CD44E has subsequent ly been found to have also distinct functional properties (Stamenkovic et al. , 1991 ; Thomas et al. , 1992; Bartolazzi et al. , 1994). Two CD44 clones isolated from a metastatic mutant of a rat pancreatic carc inoma cell line (BSp73) contained additional amino acid sequences in the juxta-membrane of the extracellular domain of CD44 molecule and were encoded by alternatively spliced exons v4-v7 or v6-v7( Gunthert et al . , 1 9 9 1 ; Herrl ich et al . , 1993). Interestingly, the stable transfection of CD44-var iant containing isoform cDNAs into the parental non metastatic rat carc inoma cells that express only CD44H and not the additional isoforms, conferred high 59 metastatic potential to the cells in vivo (Gunthert et al. , 1991 ; Herrlich et al . , 1993). The metastatic potential of these isoforms did not require the cytoplasmic domain of CD44 (Herrlich et al. , 1993). Coinjection of antibodies directed against the metastatic-specif ic domain of CD44 isoform retarded or prevented the metastatic spread in vivo (Reber et al. , 1990; Seiter et al. , 1993). Variant containing CD44 isoforms were found in different human tumor cell lines (Hofmann et al . , 1991 ; Brown et al. , 1991). In contrast, the human exon v6 containing CD44 isoforms were found to be specifically downregulated during malignant transformation of tumors of squamocel lular origin (Salmi et al . , 1993). In human colorectal neoplasia, the expression of CD44 variants including v6 containing isoforms and CD44R1 (v8-v10) were mostly found on invasive and metastatic carcinomas but only on a limited subset of normal epithelial t issues and suggested a poor prognosis (Heider et al., 1993; Finn et al. , 1994). These studies initiated several others to establish the importance and/or utility of CD44 variant as a potential neoplastic marker in different mal ignancies. Using exon-specif ic primers and primers designed to amplify all variant-containing CD44 by RT-PCR, and also exon-specif ic mAbs and using FACS, it has been possible to generate a large amount of information regarding the expression of alternate CD44 in human primary and secondary solid tumors and their association or correlation with different parameters including aggressiveness and prognosis of various forms of solid tumors (Jalkanen et al. , 1990; Matsumura and Tarin, 1992; Heider et al., 1993; Mulder et al . , 1994; Kaufman et al . , 1995; Matsumura et al., 1995). CD44 expression has also been found to correlate with the metastatic potential and poor prognosis in non-Hodgkin's lymphomas (Jalkanen et al. , 1990). Addit ional CD44 isoforms including v6 containing variants were also expressed in aggressive lymphomas and v6 has been found as an independent 60 prognosis factor for non-Hodgkin's lymphomas (Salles et al. , 1993; Koopman et al., 1993; S taude re t al. , 1995). Taken together, these studies suggest that alteration of CD44 variant expression is a common feature of malignant cells. However, the exact mechanism of the contribution of CD44 to the malignancy is not clear. CD44 Var ian ts a n d N o r m a l Cel ls Although the expression of abundant amounts of CD44 variants carrying addit ional alternative sequences has been associated with mal ignancy, their expression has also been found in normal t issues (Lesley et al . , 1993a). CD44 variants carrying the metastasis (v6) and/or (v9)-associated sequences were transiently expressed upon maturation or activation of B and T lymphocytes and macrophages (Arch et al. , 1992; Koopman et al. , 1993; Salles et al. , 1993; Haegel et al . , 1993; Mackay et al. , 1994). Immunostaining detect ion of v6-containing CD44-variant expression on the epithelial cells of new born rats and adult humans revealed it to be distinct f rom the expression of the common form of C D 4 4 (Wirth et al. , 1993; Salmi et al. , 1993). Using exon-specif ic mAbs and immunohistochemical staining, CD44 carrying alternative-specific sequences (CD44v3-CD44v10) were also found to be differentially expressed on epithelial cells f rom a variety of normal t issues (Mackay et al. , 1994; Patel et al . , 1994). Activation of T cells with Interferon-y or T N F - a upregulated the expression of v6 and v9-carry ing-CD44 variants respectively (Mackay et al. , 1994). Upon activation with IFN-y, epithelial cells and astrocytes express v6-containing variants of CD44 (Haegel et al., 1993; Mackay et al. , 1994). CD44 isoforms expressed specifically in activated cells may also seek new ligands for example CD44R1(v8-v10) isoforms have been found to bind high affinity CD44R1 rather than C D 4 4 H and to homoaggregate in a HA-independent fashion (Droll et al. , 1995). However, HA-binding-activated T cells in vivo expressed only the 61 common CD44 and not any higher molecular weight variant as detected by RT-PCR (Lesley et al. , 1994). Taken together these data suggest that during tumorigenesis and mal ignancy, t ransformed cells may reactivate the expression of gene segments that serve highly special ized functions during development and in adult t issues. The generat ion of CD44 isoforms may simply accompany malignant t ransformat ion. Alternatively, highly proliferative / activated cells expressing large amount of CD44v isoforms may be the cells that preferentially undergo malignant transformation. Their expression of CD44 variants in addit ion to other adhesion molecules may therefore give a growth advantage to these cells as compared to their normal counterparts. Variably spliced exons modify CD44 molecule by offering new sites for O-l inked glycosylation and glycosaminoglycan attachment, and this addit ion of highly glycosylated sequences into the extracellular domain of CD44 may reduce the HA binding function of the molecule (Bennett et al . , 1995b; Lesley et al . , 1995; Katoh et al. , 1995). However in its g lycosaminoglycan-bound form, CD44 may also function as a proteoglycan in immobil izing and presenting growth factors (Tanaka et al. , 1993; Bennett et al., 1995a; Jackson et al. , 1995). Interestingly, the v3-containing CD44 variants were found attached to chondroit in sulfate and heparan sulfate (CD44v3-v10 and CD44v3,v8-v10) and were then able to present heparin-binding growth factors (Bennett et al. , 1995a; Jackson et al . , 1995). 1.3.5 S u m m a r y CD44 is a family of adhesion molecules with a wide range of funct ions in different cells. Complex mechanisms of regulation are necessary for the control of CD44- l igand interactions. Splicing represents one level of regulation of CD44. The expression of spl iced variants seems to follow a restricted pattern as 62 compared to the common CD44, and may also be more limited to highly proliferative and / or differentiated cell populations. However, at a molecular level, almost nothing is known about either the mechanism of splicing or how the l igand binding of CD44 splice variants is altered, although cytoplasmic phosphorylat ion events, effects on the cytoskeleton, the presence of other (adhesion) molecule(s), the cell surface distribution and conformat ion of CD44, glycosylation of CD44 and other post-translational modifications of the molecule, as well as changes in the state of activation, differentiation and / or proliferation are all implicated. 63 1.4 Thesis Objectives The overall objective of my thesis was to investigate how interactions between hematopoiet ic cells and components of the extracellular matrix might contr ibute to regulating the proliferation and differentiation of primitive hematopoiet ic cells. At the t ime this project was initiated, the expression of several adhesion molecules on primitive hematopoietic and stromal cells had been reported, but their potential function in affecting hematopoiesis was less well character ized. The availability of in vitro and in vivo assays for quantitative and discriminating measurements of different types of primitive human hematopoiet ic cells together with the development of various purification strategies enabl ing their routine isolation also made functional studies of def ined subpopulat ions of primary primitive hematopoietic cells possible. In this thesis I have focused on an examinat ion of the potential role of CD44 in interactions between primitive hematopoiet ic cells and the extracellular matrix. This was prompted by the knowledge that CD44 is an adhesion molecule that had already been implicated as having important functions in different cell types of mesenchymal origin. Variations in the expression of CD44 [the common form and higher molecular weight isoforms including the human CD44R1 (v8-v10)] , had also been linked with malignancies and metastasis. Several distinct features of CD44 indicated that this molecule might also be relevant to the regulation of primitive hematopoietic cells. High expression of CD44 on a majority of hematopoiet ic cells and the regulation of CD44 expression with differentiation of hematopoiet ic cells through the myeloid and lymphoid l ineages, evidence of the ability of CD44 to bind to components of the extracellular matrix and an inhibitory effect of ant i-CD44 antibodies on murine hematopoiesis in LTC all suggested a role of CD44 in hematopoiesis. 64 The specific aim of my project was to investigate the potential involvement of CD44 in interactions of primitive hematopoietic cells with stroma and to explore potential alterations in CD44 expression or function in leukemia. As a first step, I therefore initiated an examination of the cell surface expression of CD44 on various subpopulat ions of primitive hematopoiet ic cells with def ined progenitor activities. These studies included an analysis of the most primitive human hematopoiet ic cells detectable in vitro, namely LTC-IC, present in normal bone marrow or CML PB samples. The question asked was: "do normal LTC-IC express CD44 on their surface and if so, how does the level of CD44 expression on these cells compare to that seen on other normal and C M L hematopoiet ic progenitors?" A related question was to examine the cell surface expression of v10 variant of CD44 in normal and leukemic hematopoiet ic cells. The results of these studies are presented in Chapter 3. In addit ion, a more detailed analysis of the expression of various CD44 epitopes on various subpopulat ions of leukemic cells f rom patients with more aggressive forms of leukemia (i.e., CML in blast crisis and AML) was then investigated, as well as the binding of myeloid leukemic cells to hyaluronan. The results of these studies are summarized in Chapter 4 . Since these studies indicated that CD44 is expressed on LTC-IC and the expression of CD44 isoforms is deregulated in leukemia, it was of interest to investigate the potential involvement of CD44 in interactions of normal and leukemic LTC-IC with stromal cells. This was performed using the LTC system as a model of the bone marrow microenvironment. Involvement of CD44 in LTC-IC interactions with stromal cells was investigated using a panel of different anti-CD44 antibodies to evaluate their effect (s) on the presence / production and maintenance of normal vs leukemic CFC and LTC-IC in LTC after varying periods of incubation. The specific questions addressed in these latter exper iments were: "How does the number of normal CFC and LTC-IC present / 65 produced in LTC treated with ant i-CD44 antibodies compare to untreated cultures after varying periods of t ime and how does this number compare to the relative production of leukemic CFC and LTC-IC in the presence of antibodies?", and "Can the effects seen with ant i-CD44 antibodies in the presence of s t roma, be reproduced also in the absence of st roma?" The results of these studies are reported in Chapter 5 of this thesis. On the basis of these f indings, I have developed a possible model to describe how CD44 might be involved in causing the effects I have observed. This model is presented and discussed in Chapter 6. 66 Chapter 2 Materials and Methods 2.1 Cells 2.1.1 N o r m a l B o n e M a r r o w All samples of human peripheral blood (PB) and bone marrow (BM) were obtained from informed and consenting individuals. Samples of normal BM were f rom harvests taken for allogeneic transplantation or f rom normal cadaveric t issue. All analyses were performed on cells in the light density fraction (<1.077g/cm3) isolated using Ficoll-hypaque (Pharmacia LKB, Uppsala, Sweden) . 2.1.2 Mob i l i zed Per iphera l B l o o d Human normal PB samples were obtained from informed and consent ing individuals. These samples were obtained from leukapheresis harvests collected f rom patients with hematological diseases in remission (two with Hodgkins disease and one AML) . The two patients with Hodgkins disease received chemotherapy (cyclophosphamide: 7g/m2) fol lowed by administration of interleukin-3 (IL-3, 2.5-5 u.g/Kg/day) and GM-CSF (5 j ig/Kg/day) until the leukaphereses were completed. The A M L patient received G-CSF (12 u,g/Kg/day) for 5 days fol lowed by 2 days of leukapheresis. 2.1.3 A M L a n d C M L S a m p l e s Heparinized PB samples were also obtained with informed consent f rom normal donors and patients with newly diagnosed (untreated) A M L or P h + CML in chronic or accelerated phase, or in blast crisis. In almost all cases, primary cells were cryopreserved and stored at -70°C in a mixture of 9 0 % fetal calf serum (FCS; StemCell 67 Technologies, Vancouver, Canada) and 10% dimethyl sulfoxide prior to analysis. In order to ensure that the results of analysis of CML samples could be attr ibuted to leukemic cells, these studies were restricted to PB samples f rom Ph+ C M L patients in chronic, accelerated phase or blast crisis with high white blood counts (WBC) at the t ime the samples were taken. In the functional studies of the chronic phase of CML only samples with clonogenic cells present at frequencies at least 25-fold and/ or LTC-IC present at f requencies at least 2000 fold above the mean for f resh normal blood were used for evaluat ion. Ph positivity of CML samples, CFC produced in LTC and LTC-IC derived CFC was verified by cytogenetic analysis at the initiation of LTC and after one and 5 weeks in the primary LTC and after 5 weeks in the secondary LTC. All analyses of primary cells were performed on light density (<1.077g / c m 3 ) blood or marrow cells isolated using Ficoll-hypaque (Pharmacia LKB, Uppsala, Sweden) , or a subsequent ly purified subpopulat ion of CD34+ cells. 2.1.4 Cel l L ines KG 1a (myelomonoblastic) cells (Koeffler et al., 1980) were maintained in Iscove's modif ied Dulbecco's medium (DMEM) supplemented with 1 0 % FCS (Hyclone, Utah, U.S.A.), K562 (erythroleukemic) cells (Lozzio etal . , 1981) in RPMI 1640 10% FCS, M 0 7 e (megakaryoleukemic) cells (Avanzi et al., 1988) in DMEM containing 10 % FCS, 5 x 10 " 5 M (3-Mercaptoethanol, 10% 5637 cell condit ioned medium (Welte et al., 1985) and 5 ng/ml of purified human IL-3, TF-1 (erythroleukemic) cells (Kitamura et al., 1989) in D M E M containing 10% FCS supplemented with 5 x 10 " 5 M (3-Mercaptoethanol and 5 ng/ml of purif ied human GM-CSF, and M B - 0 2 (erythroleukemia) cells (Morgan et al., 1991) in RPMI supplemented with 10% heat- inact ivated-FCS, 5 % human serum, 1 mM pyruvate, and 5 ng/ml of GM-CSF. 68 2.2 Antibodies Murine lgG1 mAbs specific for human CD34 (8G12) (Schmitt et al. , 1991), Thy-1 (5E10) (Craig et al. , 1993), and an epitope located on the common region of different human CD44 isoforms ( 3 d 2, which recognizes the hyaluronan-binding site) (Dougherty et al. , 1994), and murine lgG2a mAb specific for human CD45 (9.1) were purified f rom hybr idoma culture supernatants using Protein G Sepharose 4 Fast Flow (Pharmacia LKB). 8G12 labeled with Cyanine-5 (Cy5) or with phycoerythrin (PE; Pharmingen, San Diego, CA). 5E10 was labeled with PE (Pharmingen, San Diego, CA). CD38-PE was purchased from (Becton Dickinson & Co; San Jose, CA). In phenotypic studies antibodies (except for 3c12) were used directly as hybr idoma culture supernatants without further purif ication. 3c12 was used at 1ug/ml. An irrelevant murine lgG1 mAb (anti-dextran) was used as the negative isotype control in all staining experiments (Thomas et al . , 1992). Fluorescein isothiocyanate (FITC) and PE-conjugated goat F ( a b ' ) 2 ant i -mouse IgG (H+L) (GAM) were purchased from Caltag Laboratories (South San Francisco, CA). 2G1 is another murine lgG1 mAb which reacts with a CD44 epitope encoded by exon v10 (cells expressing exon v10-containing CD44 isoforms are referred to as 2G1+ o r v 1 0 + cells) (Dougherty et al. , 1994). 7f4 and 8d8 are murine lgG1 mAbs which recognize differentially expressed CD44 epitopes (present on some activated cells) (Dougherty et al. , 1995). Reactivity with C O S cells t ransfected with, and transiently expressing various CD44 isoforms conf i rmed that 7f4 and 8d8 epitopes are present on a common region shared by all isoforms (Dougherty, personal communicat ion, 1996). F (ab ' ) 2 f ragments of 3c12 and 8G12 were prepared by pepsin treatment. The success of this t reatment was verif ied by SDS-PAGE and the activity of the ant ibody f ragments was tested by FACS. In all in vitro functional studies 20 u.g /ml of purif ied antibodies were used. 69 2.3 Hyaluronan-FITC 2.3.1 P repa ra t i on HA-FITC preparation was made with potassium HA from human umbilical cord, and FITC (Sigma Chemicals, St. Louis, MO) as previously descr ibed (de Belder & Wik, 1975). Absence of free FITC was verified by thin layer chromatography. Preparat ions were pooled, resuspended in water, lyophilized and stored at -20°C. The specificity of HA-FITC recognition was verified by analysis of specific binding to DEAE-dextran transfected COS-1 cells (Hammarskjold et al. , 1986) transiently expressing CD44H or CD44R1 (Figure 4) . 2.3.2 HA-FITC B i n d i n g A s s a y a n d Pro te in K inase C HA-FITC B i n d i n g A s s a y . Cells were washed three t imes in phosphate buffered saline (PBS), incubated with HA-FITC at 2.5 ug/ml in Hank's balanced salt solution (HBSS) for 30 minutes on ice and then washed in HFN. In double staining exper iments with ant i -CD34 mAb, 8G12-PE and HA-FITC were added to the incubation tube at the same t ime. In experiments where unconjugated primary mAbs were used, HA-FITC was added last, after the antibody staining was complete. To compete HA-FITC binding with ant i -CD44 mAb, cells were first incubated with 3c12 at 25 ug/ml in HFN on ice for 30 minutes, washed twice with PBS and incubated with HA-FITC. To compete HA-FITC with unlabeled HA, cells were first washed three t imes in PBS, and then incubated with HA (100 ug/ml) and HA-FITC (2.5 ug/ml) on ice. Cells were subsequent ly washed twice in HFN, the second wash containing 2 ug/ml propidium iodide (PI, S igma Chemicals, St Louis, MO) to stain dead cells. T P A T r e a t m e n t a n d Pro te in K inase C ac t i va t i on . Cells were washed in PBS three t imes prior to their incubation for 20 hours at 37°C with or without 60 ng/ml of 1 2 - 0 -70 Vector LU Q_ !j HA-FITC CD44-R1 CD44-R1 NIL HA-FITC Figure 4. B i n d i n g of HA-FITC t o CD44R1 T r a n s f e c t e d COS Cel ls . C O S cells were transfected with either vector alone (CDM8) or CD44R1 /CDM8 plasmid DNA and analyzed 3 days post-transfection for binding of HA-FITC and the anti-v10 mAb 2G1 as indicated. 71 tetradecanoyl-phorbol-13-acetate (TPA) (Sigma) in the standard growth medium used for their propagation (Castagna et al., 1982). 2.4 Staining and Flow Cytometry Cells were washed twice and resuspended in HBSS with 2 % FCS and 0.02% sodium azide (NaN 3 ) (HFN). All staining procedures were performed at a cell concentrat ion of 10 7 ce l ls /ml . Cells were first incubated with HFN containing 5 % human serum at room temperature to block Fc receptors, then washed twice with HFN and incubated with an ant i-CD44 mAb ( 3 d 2, 2 G 1 , 7f4, 8d8) or the lgG1 control mAb for 30 minutes at 4°C, fol lowed by two washes with HFN. Samples were then resuspended in a 1:50 dilution of GAM-FITC in HFN, incubated for 30 minutes at 4°C, washed twice and incubated for another 30 minutes at 4°C with 200 ug/ml of the irrelevant mouse IgG mAb to block residual GAM-FITC (the use of a high concentrat ion of this antibody was necessary to block the residual GAM-FITC on the surface of cells expressing high numbers of CD44 molecules). The ant i -CD34 mAb (8G12-Cy5 or 8G12-PE at 10 ng/ml) and 5 ng/ml of either anti-Thy-1 mAb (5E10-PE) or ant i -CD38-PE were then added to this solution without further washing in the presence of the irrelevant mouse lgG1 for another 30 minutes at 4°C. Cells were subsequent ly washed twice in HFN, the second wash containing 2 ug/ml of PI. Analysis of cells was performed on a FACSort (Becton Dickinson & Co., St. Jose, CA) . Sort ing of cells was performed on a FACStar Plus (Becton Dickinson & Co., St Jose, CA) equipped with a 5-W Argon and a 30-mW helium neon laser (Spectra-physics, Mountain View, CA). In all experiments, gates were set to exclude dead cells (PI+) as well as most erythrocytes and granulocytes as defined by their forward light scatter (FSC) and side scatter (SSC) characterist ics. Sorted cells were col lected in D M E M containing 2 % FCS and were kept on ice until plated. 72 2.5 RT-PCR Analysis The fol lowing polymerase chain reaction (PCR) primers corresponding to CD44 sequences located in the 5' common region and in the 3' alternatively spliced exon v10 were used: 5 ' (5 ' -TGTACATCAGTCACAGACCT-3 ' ) and 3'(5'-AGGAAC G ATTG AC ATTAG AG-3 ' ) . Total RNA was isolated using TRIzol Reagent (Gibco BRL, Gaithersburg, MD). The first cDNA strand was generated in a 50 |LLI reaction containing 5 jxg of total RNA, 0.5 mM dNTPs, 200 pmoles of random hexamers, 4 m M dithiothreitol, 200 units of superscript reverse transcriptase (Gibco BRL) and 2.5 units of human placental RNase inhibitor (Gibco BRL). The reaction was incubated at 42°C for one hour fol lowed by 5 minutes at 95°C to inactivate the enzyme. 10 JJ.I of this reaction was then subjected to PCR in a 100 |il volume of 25 m M KCI, 1.5 m M MgCl2, 0.2 m M dNTPs, 100 pmoles of each primer and 2.5 units of Taq polymerase. 30 cycles were carried out as fol lows: 30 seconds denaturat ion at 94°C; 30 seconds anneal ing at 48°C and 3 minutes extension at 72 °C. This was fol lowed by a final extension at 72°C for 5 minutes. The PCR products were separated on a 2 % agarose gel, transferred to a nylon membrane, and hybridized to a 196 bp v10-specif ic probe obtained by PCR of the CD44R1 cDNA (Dougherty et al . , 1991) using the primers 5 ' (5 ' -TAGGAATGATGTCACAGGTG - 3') and 3'(5'-AGGAA C G ATTG AC ATTAG AG-3 ' ) . Hybridization was performed overnight at 60°C in 6 x SSC (1 x SSC is 0.15 M NaCI, 0.015 M sodium citrate, pH=7.5), 1 % Sodium Dodecyl Sulfate (SDS), 0 .02% polyvinylpyrrolidone, 0 .02% ficoll, 0 .02% BSA, 10 | ig/ml of denatured salmon sperm DNA and 1 0 5 cpm/ml of denatured probe. The filter was then washed at a final concentrat ion of 0.1 x SSC, 1 % SDS at 65°C. Autoradiography was performed at room temperature with a Kodak XAR-5 fi lm for 14 hours. 73 2.6 Hematopoietic Progenitor Cultures and Assays 2.6.1 C o l o n y - F o r m i n g Cel l A s s a y s Cells f rom primary samples or LTC harvests were assayed for clonogenic erythropoietic (BFU-E, CFU-GM) and multi l ineage (CFU-GEMM) progenitors in Iscove's DMEM-based methylcellulose cultures containing 3 units/ml of human erythropoiet in and 20 ng/ml each of G-CSF (Amgen, Thousand Oaks, CA) , GM-CSF (Sandoz, Basel, Switzerland), IL-3 (Sandoz), IL-6 (obtained from the supernatant of C O S cells transfected with a full copy human IL-6 c-DNA, D. Hogge, Terry Fox Laboratory) and 50 ng/ml of SF (Amgen). The methodology and criteria for hematopoiet ic colony generation and recognition were the same as previously descr ibed (Cashman et al., 1985). 2.6.2 In i t ia t ion a n d Ma in tenance of H u m a n M a r r o w A d h e r e n t Laye rs Human feeder layers of stromal cells were prepared as previously descr ibed (Sutherland et al. , 1989). Briefly, 2 to 4 weeks before setting up the experiment, human BM mononuclear cells were resuspended at 1 0 6 / ml in LTC [an enriched alpha medium containing 40 mg/L inositol, 400 mg/L glutamine, 12.5% horse serum, 12.5% FCS, 10-4 M 2-mercaptoethanol (STI) to which freshly dissolved hydrocort isone sodium hemisuccinate [Sigma] was added just before use to give a final concentrat ion of 10-6 M] and plated in Corning t issue culture dishes (Corning Glassworks, Corning, NY). These cultures were subsequently maintained as for regular LTC. 48 to 72 hours before the experiment, adherent cells f rom these LTC were trypsinized, irradiated with 15 grays and plated at 1.5 x 1 0 5 cells / ml in LTC medium. Mouse marrow-derived fibroblast cell line M210B4 and human feeder layer have been previously found to support the maintenance of human LTC-IC to the same extent (Sutherland et al. , 1989). In addit ion, M210B4 that had been genetically engineered to produce human G-CSF, IL-3 and SF have been found to support 74 maintenance of LTC-IC to even a greater extent than normal human feeder layers of stromal cells (Sauvageau et al. , 1994). Growth factor producing M210B4 cells were used to evaluate the number of LTC-IC in normal and CML specimens (chapter 3). To evaluate the effect of addition of antibody in LTC human feeder layers were used (chapter 5). In these latter experiments, the secondary LTC-IC assays were establ ished on M210B4 cells. 2.6.3 L o n g - T e r m Cu l tu re In i t ia t ing-Cel l A s s a y The general procedure used for LTC-IC assays has also been descr ibed in detail previously (Sutherland et al., 1990). Briefly, test cells were resuspended in long-term medium and then seeded onto semi-confluent feeder layers of irradiated (80Gy) stromal cells. These LTC-IC assay cultures were then maintained at 37°C and fed weekly by replacement of half of the medium containing half of the nonadherent cells with the same volume of fresh longterm medium. After a total of 5 or 6 weeks according to the type of study, the nonadherent cells were removed, washed and combined with the trypsinized and suspended adherent layer cells. These pooled cells were then plated in methylcellulose assays as described above. The total number of clonogenic cells (BFU-E plus CFU-GM plus CFU-GEMM) present at 6 weeks provides a relative measure of the number of LTC-IC originally present in the test suspension (Sutherland et al . , 1990). Limiting dilution exper iments have shown that under the condit ions used here for LTC-IC detect ion, on average one LTC-IC will produce approximately 8 clonogenic cells (Petzer et al. , 1996) and this was therefore the value used to derive the absolute LTC-IC frequencies reported. In LTC-IC assays where no colonies were detected, the minimum number of LTC-IC that could have been detected was calculated by assuming that one colony had been produced by the entire aliquot of cells evaluated in the final methylcel lulose assays. This value (instead of zero) was then used to derive an estimate of the upper limit of the mean ± standard error of the mean (SEM) number of LTC-IC in groups where one or more 75 values were below the limit of detection. In such cases the mean is indicated as < x ± S E M . In some of studies (presented in chapter 5) the adherent and non adherent cells of 1 and 5 week primary LTCs carried in parallel were assessed separately for their c lonogenic and LTC-IC contents in a secondary semi solid methylcel lulose and long-term culture respectively. The secondary LTC established after 1 or 5 weeks al lowed the detect ion of the presence or production of LTC-IC after 1 week and 5 weeks. 2.6.4 A n t i b o d y T r e a t m e n t in v i t r o In order to saturate all antigen sites, cells were first preincubated with 20 ug/ml of mAbs for 30 minutes on ice in their own medium prior to culture. Ant ibody coated light density mononuclear cells ( 1 0 6 / ml) or CD34+CD38- cells (500 cells/ ml) were plated in parallel either in LTC or liquid suspension culture supplemented with 50 ng/ml of SF and 20 ng/ml of IL-6. The progenitor contents of liquid suspension cultures and a proport ion of LTC were assessed as described above 1 week later. Ant ibodies were added at each weekly half medium change in the primary 5 week LTC. 76 Chapter 3 Differentiation-Associated Changes in CD44 Isoform Expression During Normal Hematopoiesis and Their Alteration in Chronic Myeloid Leukemia data presented in this chapter is included in the fol lowing manuscript: Ghaffari , S., Dougherty, G.J., Lansdorp, P.M., Eaves, A.C. , and Eaves, C.J. (1995). Differentiation-Associated Changes in CD44 Isoform Expression During Normal Hematopoiesis and Their Alteration in Chronic Myeloid Leukemia. Blood 86, 2976-2985. 3 . 1 . I n t r o d u c t i o n In adult life, the proliferation and differentiation of primitive hematopoiet ic progenitors is normally restricted to the bone marrow where these cells interact with var ious stromal cells of nonhematopoiet ic origin as well as with a complex mixture of ECM components . Such interactions are believed to regulate the accessibil i ty of these cells to stimuli that control their viability, cell cycle progression and their movement into and out of the circulation. These concepts are based in part on the identification on the surface of primitive hematopoietic cells of cytokine receptors and adhesion molecules with affinities for various membrane or ECM-bound l igands (Verfaillie et al. , 1994a; Berardi et al . , 1995). In addit ion, in vivo experiments have shown that particular l igands, or antibodies for specific cell adhesion molecules, may stimulate the rapid exodus of primitive hematopoietic cells f rom the marrow into the blood or conversely, may alter the homing of primitive hematopoietic cells into the marrow (Will iams et al. , 1991 ; Molineux et al. , 1991 ; Bodine et al. , 1993; Papayannopoulou and Nakamoto, 1993). Neoplastic transformation may also alter the adhesive characterist ics of primitive hematopoietic cells with associated changes in their turnover and t issue distribution. For example, as ment ioned earlier in this thesis, in 77 CML, the leukemic progenitors exhibit abnormal adhesive properties, are found at abnormal ly elevated levels in the blood and are able to establish hematopoiesis in extramedul lary sites (Gordon et al. , 1987; Gordon et al., 1991 ; Udomsakdi et al . , 1992; Verfail l ie e t a l . , 1992). CD44 is expressed on the surface of many cells including representatives of all hematopoiet ic cell l ineages (Trowbridge et al., 1982; Kansas et al. , 1990; Lewinsohn et al., 1990; Lesley et al. , 1993). Hyaluronan is the most widely recognized l igand of CD44, but evidence of binding to f ibronectin, col lagen, serglycin and osteopontin has also been reported (Underhil l , 1992; Toyama-Sor imachi et al. , 1995; Weber et al. , 1996). The common form of CD44 expressed on hematopoiet ic cells (CD44H) is a 85-90 KD glycoprotein. A large number of higher molecular weight isoforms may also be produced in specific cell types or under specific condit ions as a result of the alternative splicing of at least 10 contiguous exons (v1-v10) within the C D 4 4 gene (Screaton et al. , 1992; Mackay et al. , 1994). CD44R1 is one of several v10-containing CD44 cDNAs. It was cloned f rom the KG-1a leukemic cell line and contains in the extracellular region of the molecule an insertion of 132 amino acids encoded by exons v8, v9 and v10 (Dougherty et al. , 1991). This isoform corresponds to the epithelial form of CD44 (CD44E) f rom which it differs by only 3 amino acids (Stamenkovic et al. , 1991). CD44R2 is a v10-containing isoform of CD44 that shares only the last 69 amino acids present in the unique region of C D 4 4 R 1 . Al though changes in CD44 isoform expression including have been found to characterize various metastasizing cells such as those present in certain aggressive lymphomas (Horst et al. , 1990; Salles et al. , 1993; Koopman et al. , 1993) the expression of CD44 isoforms in human leukemia was not known. Evidence that CD44 may be involved in the regulation of early stages of normal hematopoiesis has been suggested by the observation that primitive clonogenic cells express C D 4 4 (Lewinsohn et al . , 1990) and that the addit ion of ant i -CD44 monoclonal antibodies to longterm marrow cultures (LTC) can cause a marked and sustained decrease in the number of mature cells subsequent ly found in the 78 nonadherent fraction of these cultures (Miyake et al., 1990; Gunji et al. , 1992). In order to assess whether the inhibitory effect of ant i -CD44 antibodies was taking place at the level of very primitive hematopoietic cells, it was important to investigate whether LTC-IC indeed express CD44. Similarly, in spite of the well known abnormal adhesive properties of C M L cells, potential alterations in their expression of CD44 have not been invest igated. The purpose of the studies presented in this Chapter was to determine whether there are differences in the levels and / or isoform expression of CD44 during the earliest stages of primitive normal hematopoietic cell development and whether these processes may be altered in the corresponding stages of leukemic hematopoiesis characterist ic of cells produced in patients with CML. 3.2 R e s u l t s 3.2.1 A l t e r e d CD44 E x p r e s s i o n o n P h e n o t y p i c a l l y De f ined P o p u l a t i o n s of P r im i t i ve C M L Cel ls In order to compare the patterns of expression of CD44, CD34 and Thy-1 on primitive normal and neoplastic (Ph+) CML cells, I isolated various fractions of light density bone marrow cells f rom normal individuals and compared their staining profiles with those obtained for cells in the light density fraction of PB f rom a series of newly d iagnosed C M L patients with high W B C counts. The choice of this source of CML cells for these comparisons was based on previous data showing that the light density cells in the PB of chronic phase CML patients with high W B C counts, on average, will contain a >10-fold increase in all types of leukemic (Ph+) progenitors (both clonogenic cells and LTC-IC) such that they approach or even exceed the f requency of the same types of primitive cells in the light density fraction of normal BM cells (Eaves and Eaves, 1987; Udomsakdi et al., 1992). 79 Figure 5 shows representative dot plots for a normal BM sample and Figure 6 shows analogous plots for a representative CML PB sample stained in the same way. These illustrate the general f inding for all samples studied that > 9 0 % of the light density cells in both normal BM and CML PB were CD44+. In addit ion, amongst the CD34+ populat ions present in these samples, no CD44- cells were detectable. From these analyses, two additional similarities between normal BM and CML PB were consistently noted. First, the relative proportion of CD34+ cells that were also Thy-1 + was approximately the same. Second, the proportion of (total) CD34+ cells in C M L PB expressing very high (1000 x background, designated CD44+++) , as compared to the proport ion expressing intermediate (100 x background, CD44++) levels of CD44 appeared indistinguishable from those characteristic of the CD34+ populat ion in normal BM (Table 1). However, in the CML samples, the proportion of Thy-1 + cells that were CD44+++ was significantly higher than in normal BM (0.4 ± 0 . 1 % of CD34+ cells in normal BM compared to 1.2 ± 0 .2% of CD34+ cells in CML PB, p<0 .01 , Student 's t test). In addit ion, I looked within the entire light density fraction of normal BM and CML PB, as well as within their respective CD34+ subpopulat ions, for cells expressing exon v10-containing CD44 isoforms. For this the 2G1 mAb, which specifically recognizes an epitope present on the extracellular portion of CD44 isoforms containing the amino acids encoded by the v10 exon, was used. In normal BM, 2G1+ cells were relatively rare, compris ing from 4 to 8 % of the total light density fraction and < 1 % of the CD34+ cells. In all of the 6 CML PB samples analyzed, 2G1+ cells were much more prevalent (up to 3 0 % of the light density cell fraction) and included cells that were also expressing CD34 (up to 7%) although there was marked patient to patient variability in the proport ion of CD34+ cells that were also 2G1+. Nevertheless, because of the large elevations in total numbers of light density myeloid cells in the blood of these patients, these f indings suggest that there is a marked increase in the absolute number of 2G1 + hematopoiet ic cells in patients with CML. On the other hand, in both normal BM and 80 A B • • • i ' • - • -SSC CD44 Figure 5. Gates u s e d fo r a n a l y s i s of a n t i b o d y reac t i v i t y w i t h n o r m a l B M ce l l s . BM cells were gated to exclude dead cells (A) , and cells with high SSC and very low FSC (B) .The gating used to select cells with high C D 3 4 expression is shown in Panel C. The gating used to subdivide the CD34+ cells according to their expression of CD44 and Thy-1 is shown in Panel D. The four quandrants shownin Panel D represent the four CD34 subpopulat ions analyzed functionally and these were designated as fol lows: C D 4 4 + + T h y - 1 " , C D 4 4 + + T h y - 1 + , C D 4 4 + + + T h y - 1 " and C D 4 4 + + + T h y - 1 + . The same gates were used for C M L PB and mobil ized blood (MOB). 81 1 0 4 V 1 0 3 i I _ . I • • " t - • • I -1 QO I _ i — — n I I _ I 1 I I _ I i l l • ml 10° 10 1 1 0 2 1 0 3 1 0 4 C D 4 4 F igu re 6. E x p r e s s i o n of CD44 v e r s u s Thy-1 o n C M L PB C D 3 4 + ce l l s . Cells shown are located within the CD34+ gate shown in Figure 5. 82 T a b l e 1 . C o m p a r i s o n s o f t h e d i s t r i b u t i o n o f to ta l C D 3 4 + ce l l s a n d t h e to ta l l i gh t d e n s i t y ce l l s a c c o r d i n g t o the i r leve ls o f e x p r e s s i o n o f CD44 a n d Thy-1 in n o r m a l BM v e r s u s C M L PB* Population Origin nt C D 4 4 + + C D 4 4 + + + Evaluated of Cells Thy-1 + Thy-1" TOTAL Thy-1 + Thy-1" TOTAL Total LDF Normal BM 6 3 ± 1 80 ± 1 82 + 1 0.4 + 0.1 17+1 18 ± 1 CML 6 3 + 1 77 + 3 80 + 1 1.2 + 0.2 1 8 ± 3 19± 1 PB CD34 + Normal BM 6 0.2 +0.1 4 ± 2 4 + 2 0.05 ±0.02 0.9 ±0 .3 1 ±0 .3 CML 6 0.4 ±0.1 11 ± 8 11 ± 8 0.3 ± 0.02 4 ± 3 4 ± 3 PB * Values shown are the mean ± SEM of the individual values for the various populat ions evaluated expressed in each case as a % of the total light density fraction (LDF). f Number of samples, for 6 normal individuals and 5 C M L patients. One of the samples f rom the first CML patient was studied twice (before and after thawing, see Table 5). 83 C M L PB, all of the 2G1+ cells appeared to be Thy-1- - FACS analysis further showed that the light density 2G1+ cells from either normal BM or CML PB had low SSC propert ies and intermediate to high FSC characteristics. Upon sort ing and May-Grunwald Giemsa staining, all 2G1+ cells, regardless of their origin, appeared to be exclusively myeloid cells (and did not include erythroblasts or other recognizable cell types). In functional assays of the sorted CD34+ 2G1 + CML cells, no clonogenic progenitors of any kind was detected. In order to determine the nature of the v10-containing CD44 isoform (s), expressed by 2G1+ cells, RT-PCR was performed. Primers were designed to amplify all v10-containing CD44 isoforms, but to exclude all other sequences lacking this exon. This was achieved by selecting a 5' primer from a region common to all CD44 isoforms, and a 3' primer in the last of the alternatively spliced exons (v10). This analysis yielded two major bands of approximately 680 and 480 bp (Figure 7). These correspond to the expected sizes of the CD44 isoforms R1 and R2 (672 and 481 bp, respectively), which have been previously demonstrated in various hematopoiet ic cells. (Dougherty et al. , 1991) 3.2.2 D i f fe rent ia l E x p r e s s i o n of CD44 o n Di f ferent N o r m a l P r o g e n i t o r P o p u l a t i o n s In v iew of the extensive variation in the level of CD44 expression seen on the CD34+ cells present in normal BM, it was of interest to determine whether any l ineage-associated changes in CD44 expression might be demonstrable. To examine this possibility, light density CD34+ cells f rom 6 normal BM samples were sorted according to their expression of very high (+++) or intermediate levels (++) of CD44 and detectable versus undetectable levels of T h y - 1 . Cells f rom each of the four subpopulat ions thus obtained (illustrated in Figure 5) were then assayed in both clonogenic and LTC-IC assays. The resultant frequencies of the various progenitor 84 F igu re 7. RT-PCR a n a l y s i s of 2 G 1 + ce l l s . (A) Primers ( ^ z ^ : ) were designed to amplify all v10-containing CD44 isoforms and to exclude all isoforms lacking this exon. The shaded area corresponds to the t ransmembrane domain. Human v1 contains an in-frame stop codon. (B) PCR products were separated on a 2 % agarose gel, transferred to a nylon membrane and hybridized to a v10-specif ic probe (see Materials and Methods). The arrows indicate bands corresponding to expected reaction products for CD44R1 and CD44R2. 85 types measured are shown in Table 2. The greatest enrichment of BFU-E was obtained in the CD44++Thy-1 - fraction and to a lesser extent in the CD44++Thy-1 + fract ion. Very few BFU-E were found in the CD44+++ fractions (either Thy-1 + or T h y - 1 F i g u r e 8). In contrast, comparing the absolute numbers of colonies recovered I found that both CFU-GM and LTC-IC were more heterogeneous in their levels of CD44 expression (Table 2 and Figure 8). Most CFU-GM were T h y - 1 - and further separat ion of these cells according to their level of CD44 expression did not result in a selective enr ichment of a subpopulat ion of T h y - 1 - CFU-GM. Some (8 + 4 %) C F U - G M were also Thy-1+. These appeared to be confined primarily to the CD44++ populat ion of CD34+ cells. As found previously, LTC-IC were usually more highly enriched in the Thy-1 + as opposed to the T h y - 1 - fractions of CD34+ normal BM cells. (Craig et al. , 1993). However, in terms of total LTC-IC yields, a substantial proport ion (68 ± 13% ) of all LTC-IC were Thy -1 " . Figure 8 shows a comparison of the relative distribution of each of these normal progenitor types between the CD44++ and the CD44+++ fractions. The difference in the ratio of CD44++ to CD44+++ progenitors between CFU-GM (or LTC-IC) and BFU-E was statistically significant (p<0.02 in both cases; Student 's f tes t ) . 3.2.3 E x p r e s s i o n of CD44 o n C M L P r o g e n i t o r s is A l t e red PB samples from 5 CML patients were analyzed to determine the level of expression of Thy-1 and CD44 on various primitive leukemic cell populat ions. The frequencies of C F U - G M and BFU-E in each of the 4 CD34+ subpopulat ions def ined by differences in Thy-1 and CD44 expression (gated as indicated in Figure 6) are shown in Table 3. The relative recoveries of these progenitors in the CD44++ versus the CD44+++ fractions by comparison to normal progenitors are shown in Figure 9. In order to control for the fact that CML cells might exhibit features unique to circulating and/or activated/mobil ized progenitors, a series of PB harvests col lected after t reatment of remission A M L or Hodgkins patients with chemotherapy and 86 Tab le 2. P r o g e n i t o r d i s t r i b u t i o n s in s u b p o p u l a t i o n s of C D 3 4 + ce l l s in n o r m a l BM a c c o r d i n g t o the i r leve ls of e x p r e s s i o n of CD44 a n d Thy-1 Type of Sample Number of Progenitors per 105 Cells* Progenitor No. Evaluated CD44++ CD44+++ Thy-1 + Thy-1 " Thy-1 + Thy-1 " BFU-E 1 5000 7500 N D t <500 2 ND 8000 ND 800 3 1300 5000 <1000 600 4 2100 5700 ND 500 5 2500 13000 ND 500 6 ND 9700 ND 500 mean 2700 8100 <1000 <600 ± S E M ± 8 0 0 ± 1200 ± 4 0 CFU-GM 1 3800 15000 ND 12000 2 ND 13000 ND 8500 3 23000 19000 15000 20000 4 9000 6700 ND 4500 5 7700 6200 ND 8000 6 ND 9500 ND 13000 mean 11000 11000 ND 11000 ± S E M ± 4 2 0 0 ± 2 0 0 0 ± 2 2 0 0 LTC-IC± 1 ND ND ND ND 2 18000 900 500 700 3 1800 80 <500 100 4 1600 200 1800 300 5 3300 900 13000 800 6 2900 800 <300 200 mean 5600 600 <3300 400 ± S E M ± 3 2 0 0 ± 2 0 0 ± 2 5 0 0 ± 100 * For the population analysed, t Not done. t Measured as the total clonogenic output in LTC after 6 weeks divided by 8. 87 100 0 > o u CD DC 50 H BFU-E CFU-GM LTC-IC TOTAL CELLS F igu re 8. D i s t r i b u t i o n o f h e m a t o p o i e t i c p r o g e n i t o r s a c c o r d i n g t o the i r level of CD44 e x p r e s s i o n . Compar ison of the distribution of different types of progenitors between the C D 3 4 + C D 4 4 + + ( • ) and C D 3 4 + C D 4 4 + + + ( • ) fractions of normal light density marrow cells with the light scatter characterist ics shown in Figure 5. Values shown are the mean ± SEM of measurements of the relative recovery of progenitors f rom each of 6 experiments calculated in each case by mult iplying the percentages of cells retrieved in each fraction by the corresponding progenitor enr ichment observed in that fraction and then normalizing the data to 100% for the number of progenitors detected in the total CD34+ populat ion. 88 Tab le 3. P r o g e n i t o r d i s t r i b u t i o n s in s u b p o p u l a t i o n s of C D 3 4 + C M L PB ce l l s a c c o r d i n g t o the i r leve ls of e x p r e s s i o n of CD44 a n d Thy-1 Type of Patient Number of Progenitors per 105 Ce l l s f Progenitor No.* Evaluated CD44++ CD44+++ Thy-1 + Thy-1 " Thy-1 + T h y - 1 ' BFU-E 1a* N D * 13000 ND 7100 1b 4800 5000 5300 3500 2 7000 11000 4300 700 3 5300 17000 3500 <500 4 7300 23000 7900 2500 5 7200 15000 ND 1800 mean 6300 14000 5300 <2700 ± S E M ± 5 0 0 ± 2 5 0 0 ± 1000 ± 1000 CFU-GM 1a ND 900 ND 11000 1b 9000 2700 19000 18000 2 13000 6700 12000 16000 3 9900 3800 24000 13000 4 21000 6000 19000 20000 5 12000 4100 ND 23000 mean 13000 4000 19000 17000 ± S E M ± 2 1 0 0 ± 9 0 0 ± 2 5 0 0 ± 1800 * 1a and 1b refer to fresh and thawed cells f rom the same original sample (see Table 5) t For the populat ion analysed. $ Not done. 89 TOTAL MNC CO Q O "I * ' * ' I SSC CD34 + CELLS • • f§P'... • CD44 100 75 H 50 H 25 H t t ND III CM MB NM BFU-E CM MB NM CFU-GM CM MB NM LTC-IC CM MB NM TOTAL CELLS Figure 9. P rogen i to r d i s t r i b u t i o n in C M L PB a c c o r d i n g t o CD44 e x p r e s s i o n . Comparison of the distribution of BFU-E, C F U - G M , and LTC-IC in the C D 3 4 + C D 4 4 + + P ) and C D 3 4 + C D 4 4 + + + ( • ) fractions of CML PB (CML, n >3), mobil ized blood (MOB, n=3) and normal bone marrow (NBM, n=6). Error bars indicate one S E M above the mean . 90 administrat ion of G-CSF or GM-CSF and IL-3 were also analyzed. Dot plots of CD44 versus Thy-1 expression by the CD34+ cells in these leukapheresis samples were not noticeably different (data not shown) from those seen for normal BM (Figure 5) or CML PB (Figure 6). Similar CD44 gates were therefore used to compare the distribution of different progenitor types amongst the 2 subpopulat ions of interest (defined by their levels of CD44 expression) in the CML PB and mobil ized normal PB samples. The results of these studies are shown in Table 4 and in Figure 9. The analyses of the CML samples revealed a number of interesting f indings. First, as can be seen in Table 3, both Thy-1 + and T h y - 1 - BFU-E and CFU-GM were readily and consistently detected in CML PB. Although genotyping studies of the colonies produced by these sorted progenitors were not performed, it is unlikely f rom the number originally present in the samples used (Table 5), that any of these sorted progenitors contained a significant proportion of normal cells. Moreover, in a more extensive analysis of CD34+ subpopulat ions of CML progenitors (Petzer et al. , 1996b) the presence of a significant population of Ph+ Thy-1 + CFU-GM in C M L PB has been directly demonstrated. Finally, as shown in Figure 9, the proportion of CFU-GM and LTC-IC found in the CD44+++ fraction of CD34+ CML PB cells was consistently and significantly (p<0.01) elevated by comparison to CFU-GM and LTC-IC in normal BM. Interestingly, this was not true for the BFU-E present in CML PB. Moreover, the alteration in CD44 expression by CML CFU-GM was not found to be a general feature of C F U - G M that had been mobil ized into the circulation since the ratio of CD44++ to CD44+++ C F U - G M in leukapheresis harvests of "normal" cells was not different f rom that characteristic of CFU-GM in normal BM. 3 . 3 D i s c u s s i o n The production of different CD44 isoforms in conjunction with variable degrees of glycosylation and chondroit in sulfate attachment are thought to explain the diversity of adhesion-dependent processes in which CD44 has been implicated (Underhil l , 1992; 91 Tab le 4. P r o g e n i t o r d i s t r i b u t i o n in c i r c u l a t i n g s u b p o p u l a t i o n s o f c h e m o t h e r a p y a n d g r o w t h fac to r m o b i l i z e d CD34* ce i l s Progenitor Evaluated Sample No. CD44+ + CD44+++ FQ/10 5* Enricht t Recovery (%)* FQ/10 5* Enricht t Recovery (%)* BFU-E 1 2000 4 86 610 1 14 2 4300 73 97 290 5 3 3 13000 230 97 1200 21 3 mean ±SEM 6400 + 3300 100 + 66 93 ± 4 680 ±260 9 ± 6 7 ± 4 CFU-GM 1 6000 5 73 4300 4 27 2 1900 38 84 800 17 16 3 2100 49 62 4100 97 38 mean + SEM 3300 ± 1300 31 ± 13 73 + 6 3100 ±1100 39 ± 2 9 27 ± 6 * Frequency (FQ) per 105 cells for the population analysed. t Calculated by dividing the frequency per 105 sorted cells by the f requency per 105 unsorted light density cells in each individual experiment, t Calculated as described in the caption to Figure 8. 92 Tab le 5. In i t ia l W B C a n d c l o n o g e n i c p r o g e n i t o r c o n c e n t r a t i o n s in f r e s h or c r y o p r e s e r v e d per iphera l b l o o d s a m p l e s f r o m pa t ien ts w i t h C M L Sample W B C Status BFU-E CFU-GM No. (x 1fj6 per ml) when u s e d f (per ml) (per ml) 1a* 180 F 7200 2700 1b 180 T 7400 2200 2 250 T 400,000 150,000 3 450 T 150,000 380,000 4 350 T 970,000 190,000 5 300 T 780,000 120,000 * Progenitor values for samples 1 a and 1 b are from the same patient, t Fresh (F), or f rozen and subsequently thawed (T) samples. 93 Lesley et al. , 1993). In this chapter I have described further evidence that CD44 may be involved in the regulation of early stages of hematopoiesis based on the demonstrat ion of differentiation and transformation - associated changes in the expression of this gene product on primitive normal and CML cells. Analysis of the entire CD34+ population in normal BM and CML PB indicated that CD44- cells are not present at a detectable level within this fraction, in contrast to the total light density cell fraction of which a small percent (up to approximately 10%) may be identified as C D 4 4 - . Functional studies showed that the LTC-IC in normal marrow express intermediate to very high levels of CD44, thus mirroring the heterogeneous pattern of CD44 expression also exhibited by normal CFU-GM. In contrast, BFU-E in normal BM represent a more homogenous population of cells in terms of their pattern of CD44 expression. It will be interesting to determine whether the subset of C F U - G M expressing very high levels of CD44 are those that are able to bind immobi l ized hyaluronan (Smadja-Joffe et al. , 1994) or that cooperate with the OC4P1 integrin in binding to the C-terminal heparin-binding domain of f ibronectin (Verfaillie et al . , 1994b). Kansas et al. (1990) have also reported changes in CD44 expression as hematopoiet ic cells mature, al though in their studies CD44 was seen to be downregulated at later stages of myeloid and erythroid cell maturation. The pattern of CD44 expression on primitive C M L cells was found to differ in two respects f rom that exhibited by normal BM or mobil ized PB cells. First, a significantly larger proportion of C M L C F U - G M and LTC-IC were found to express CD44 at a very high level, and second, a subset of CD34+ cells expressing an exon v10-containing CD44 isoform(s) that was not detectable in normal BM could be readily detected in all CML PB samples analyzed. Since all normal samples used in this work were frozen, the changes in the pattern of CD44 expression on CML cells could not be attributed to a general f reeze-thaw artifact. Whether or not these changes may contribute to the abnormal adhesive properties previously described for CML cells remains to be establ ished. 94 The mechanism(s) underlying the increased expression of CD44 on CML C F U -G M and LTC-IC as well as the increased expression of exon v10-containing CD44 isoform(s) on more mature CML cells, including some within the CD34+ populat ion, is also unknown. In particular, it is not clear whether these increases are accompanied by concomitant changes in the expression of other CD44 isoforms. Activation is one of multiple mechanisms of CD44 isoform expression described in different cell types (Lesley et al. , 1993). The increase observed here in expression of CD44 on C M L C F U - G M and LTC-IC may, therefore, be related to their constitutively act ivated state, (Camp et al . , 1991 ; Bodine et al. , 1994) for example, through Bcr-Abl st imulation of the Ras pathway as a result of the association of p 2 1 0 B c r _ A b l with Grb-2/Sos (Pendergast et al . , 1993; Puil et al. , 1994; Tauchi et al., 1994). The product of the bcr-ablgene has also been found to inhibit p120 GAP activity (Skorski et al. , 1994) which would favor the formation of GTP-bound Ras and hence its accumulat ion in an activated form. The results of Bcr-Abl transfection experiments have also implicated p 2 1 0 B c r - A b l in the constitutive activation of Ras (Skorski et al. , 1994) and several studies have indicated that Ras activation may alter CD44 expression, both by modulat ing CD44 promoter activity as well as via mechanisms that control CD44 transcript splicing (Hofmann et al., 1993; Jamal et al. , 1994; Penno et al., 1994). It is interesting to note that modulat ions of CD44 expression were not observed in cells obtained f rom patients treated with IL-3 in spite of the described ability of this cytokine to transiently activate the Ras pathway (Satoh et al. , 1991). Taken together, these f indings suggest multiple mechanisms by which the observed changes in CD44 expression seen at different levels of granulopoiesis in CML could be related to the expression of the Bcr-Abl gene in these cells, al though they do not explain why such changes should be restricted to the granulopoietic l ineage. The identification of a small subset of maturing myeloid cells in normal BM that appear to express different exon v10-containing isoforms of CD44 is concordant with previous reports demonstrat ing that a change in CD44 isoform expression is not 95 uniquely associated with malignant transformation (Arch et al. , 1992; Mackay et al. , 1994) but may be more closely correlated with changes in properties that also occur during the development or functional activation of normal cells (Bennett et al . , 1995a; Jackson et al. , 1995). The observed increase in expression of exon v10-containing CD44 isoform(s) on CML cells could simply reflect a selective amplif ication in CML of a cell type that normally expresses these isoforms. Alternatively, it could reflect a direct effect of malignant transformation on cells that would not normally express exon v10. Support for the latter possibility has been recently suggested by the f inding of alterations in CD44 expression on primary cells f rom patients with other types of (acute) myeloid leukemia (data presented in Chapter 4). In summary, this study has documented the presence of CD44 on LTC-IC in normal human BM and provided evidence of a heterogeneity in the level of the CD44 expressed on these very primitive cells that is also shared by normal C F U - G M . In contrast, most normal BFU-E express on average a lower level of CD44. Evidence for the expression of different v10-containing CD44 isoforms in normal human marrow cells has also been provided. In CML, there is a disproport ionate increase in the type of C F U - G M and LTC-IC that express very high levels of CD44. These f indings, taken together with the abnormal production by CD34+ CML cells of exon v10-containing isoform(s) of CD44, suggest that the control of CD44 mRNA production and processing may be influenced both by normal differentiation mechanisms as well as those that mediate leukemic transformation. 96 Chapter 4 Altered Patterns of CD44 Epitope Expression in Human Chronic and Acute Myeloid Leukemia data presented in this chapter is included in the fol lowing manuscript: Ghaffari , S., Dougherty, G.J., and Eaves, C.J. Altered Patterns of CD44 Epitope Expression in Human Chronic and Acute Myeloid Leukemia (submitted for publication) 4.1 I n t r o d u c t i o n CD44 is highly expressed on the majority of normal hematopoiet ic cells thus far examined (Trowbridge et al. , 1982; Kansas et al., 1990; Lewinsohn et al. , 1990), including those with functional properties of primitive progenitors (data presented in Chapter 3). The level, type and ligand binding ability of CD44 normally varies according to the state of proliferation, differentiation and/or activation of the cell , reviewed in (Lesley et al . , 1993). For example, during the early differentiation of primitive hematopoiet ic cells, changes in the level of CD44 expression accompany commitment to the erythroid lineage (Chapter 3; Kansas et al. , 1990). Similarly, fol lowing the activation of normal T and B cells, another isoform of CD44, CD44R1 generated by alternative splicing of the v8-v10 exons, is expressed (Dougherty et al. , 1991 ; Murakami et al. , 1991 ; Stamenkovic et al., 1991 ; Hathcock et al. , 1993). Changes in CD44 expression have also been associated with a variety of mal ignant phenotypes (Matsumura and Tarin, 1992; Cooper and Dougherty, 1995) including cells f rom patients with A M L and CML (Chapter 3; Kortlepel et al . , 1993). However, the possibility that different populations of human leukemic cells may show changes in the isoforms, or epitopes, or the hyaluronan (HA)-binding capacity of the CD44 molecules they express has not been investigated previously. To address these questions, I utilized a panel of 4 different CD44 97 epitope-specif ic monoclonal antibodies (mAbs) to compare their reactivities with normal human bone marrow (BM) cells and with a variety of primary samples and establ ished lines of human myeloid leukemic origin. Three of the epitopes studied are present on a region of the CD44 molecule that is common to most isoforms. One of these antibodies can block HA binding, although its expression does not indicate HA-binding competence (Dougherty et al. , 1994). The other two have been associated with T cell activation (Dougherty et al. , 1995). The fourth epitope analyzed is encoded by exon v10 and hence allows the identif ication of cells expressing v10-containing isoforms of CD44 (Dougherty et al. , 1994). 4.2 Resu l t s 4.2.1 A l t e r e d Pat te rns of CD44 Ep i tope E x p r e s s i o n o n P r imary H u m a n L e u k e m i c Cel ls PB cells f rom a total of 17 patients with CML (12 in chronic phase studied at diagnosis, 2 later in accelerated phase and 3 at the t ime of diagnosis of blast crisis) and 13 patients with newly diagnosed A M L (all FAB categories except M2) as well as 6 normal BM samples were used to investigate leukemia-associated changes in CD44 expression. FACS analyses of both light density cells and the CD34+ subpopulat ions stained with the 4 different ant i -CD44 epitope-specif ic mAbs descr ibed in the Introduction and Material and Methods were performed. The clinical data for the patients studied are summarized in Table 6 and the results of the FACS analyses are summar ized in Figures 10A and 10B. More than 9 0 % of the light density cells and all of the CD34+ cells in every sample studied showed positive staining with 3c12, the mAb that recognizes the CD44 HA-binding site. This is similar to what I descr ibed in Chapter 3 for normal BM cells and for PB cells f rom a smaller series of chronic phase C M L patients. In that chapter 3 also reported that reactivity of light density normal BM 98 Tab le 6. Fea tures of t h e l eukemic pa t ien ts a n d the i r s a m p l e s u s e d in t h e s e s t u d i e s D isease W B C / L % C D 3 4 + FAB W B C / L % C D 3 4+ C M L s t a t u s x 1 0 9 ce l ls A M L t y p e x 1 09 ce l l s 1 CP 180 3 1 REABIT 86 7 2 CP 260 27 2 M1 58 50 3 CP 450 17 3 M1 200 3 4 CP 350 59 4 M1 153 2 5 CP 58 0.6 5 M1 255 95 6 CP 300 16 6 M3 44 0.3 7 CP 27 ND 7 M4 88 22 8 CP 114 25 8 M4 12 26 9 CP 250 56 9 M4 24 8 10 CP 220 69 10 M4 65 70 11 CP 300 12 11 M5 103 0.3 12 CP 50 55 12 M5 262 23 13 AP 36 64 13 M6 27 50 14 AP 36 14 15 BC 94 78 16 BC 93 79 17 BC 110 70 CP = chronic phase, AP = accelerated phase (included with CCP patients in Figures 10 and 11), BC = blast crisis, REABIT: refractory anemia with excess of blasts in transformation 99 2G1 7 f 4 8d8 1 0 0 jgio LU > CO 2 1 0.1 - - • • - 1 -• • • I " • • • • l a • • 1 -- 1 I • • 1 - - • • • • • — • - •• • • • • - •• • • • • - • • • • • • • • • 1 • • • _ • • — B — B— • NBM (6) CCP* (15) CBC (3) AML* (13) NBM (6) B 2G1 CCP* CBC (12) (3) 7 f 4 AML* (13) NBM (5) CCP (12) CBC (3) AML (13) 8d8 100 CBC (3) AML 0) F igu re 10. Scat te r p lo t of CD44 e p i t o p e s o n l e u k e m i c ce l l s . Proportion of positive cells in the light density fraction (A) or the CD34+ subpopulat ion (B) of cells f rom a series of normal BM samples or PB samples from different C M L (CCP = C M L in chronic phase, CBC = CML in blast crisis) and A M L patients as assessed after staining of the cells with mAbs directed against different CD44 epitopes. The number of patients studied in each category is shown in brackets. Solid symbols refer to the % of positive cells over the background (99% of cells stained with isotype control antibody) where such cells were detected. The open symbols refer to patients where such cells were not detectable. * Indicates a significant difference (P<0.05 by the Mann-Whi tney U test) f rom normal BM. 100 cells with the 2G1 (v10-specific) mAb is restricted to a relatively small (2-10%) subpopulat ion which does not have detectable progenitor activity, whereas the proport ion of 2G1+ chronic phase CML cells is higher (up to 3 0 % of the total light density population and up to 10% of the CD34+ cells). It was, therefore, of particular interest to examine the reactivity of 2G1 with cells f rom patients with more aggressive leukemias (i.e., A M L and CML in blast crisis). As shown in Figure 10, the proportion of 2G1+ light density cells f rom the A M L patients studied was highly variable but, on average, was significantly higher (p<0.05) than in normal BM even when this comparison was restricted to cells that also expressed CD34+. Interestingly, a significantly increased (p<0.05) proport ion of CD34+ cells f rom CML patients in chronic or accelerated phase were also 2G1+ whereas in the 3 samples from CML patients in blast crisis, no 2G1+ CD34+ cells were detected. In the case of 7f4 (one of the activation epitopes), only a small percentage (0.1-5%) of light density normal BM cells showed positive staining and none of these coexpressed CD34, whereas this epitope was detected on a significantly larger (p<0.05) proport ion of the light density cells f rom both A M L patients (where up to 9 0 % of cells were 7f4+) and patients with chronic or accelerated phase CML (where up to 6 0 % of the cells were 7f4+). Moreover, the 7 f 4 + cells in the leukemic samples included a substantial fraction of the CD34+ population (2-40%). On the other hand, 8d8 (the other activation epitope) was found to be expressed on a high proportion of the light density cells in normal BM and this did not appear to be altered in the categories of leukemia studied. Figure 11 shows examples of the overall level of expression of CD44 (on a per cell basis) as indicated by staining with 3c12, as well as the level of expression of each of the 3 epitopes detected by the same mAbs used in Figure 10. The samples shown were selected for the presence of a significant population of positive cells in instances where such examples were found. These profiles serve to illustrate the general f inding that the wide patient-to-patient variation seen in the frequency of epitope-posit ive leukemic cells (Figure 10) was not associated with obvious differences in the level of 101 2G1 7f4 8d8 3 d 2 N B M 0) XI £ 3 C M L « Q) > m A M L B N B M C M L A M L n / \ Log Fluorescence 2G1 7f4 8d8 XI E 3 z o O > Q) DC A 3 d 2 i \ i \ Log Fluorescence F i g u r e 1 1 . H i s t o g r a m s of CD44 e p i t o p e s o n l e u k e m i c ce l l s . Examples of histo-grams of the expression (in samples of patients' cells that stained positive with the ant i -CD44 mAb indicated-solid profile) overlaid on the staining profile for the isotype control ant ibody (open profile) of light density cells (A), or CD34+ cells (B) f rom normal BM or PB f rom CML and A M L patients (See also the caption to Figure 10). 102 expression of any of the CD44 epitopes examined as compared to normal cells. However, because of the known heterogeneity in the populations being compared, these studies obviously do not preclude the possibility that such differences might be revealed by the simultaneous use of other antibodies to allow further discrimination of minor subpopulat ions. 4.2.2 Var iab le E x p r e s s i o n of CD44 Ep i topes o n H u m a n M y e l o i d L e u k e m i a Cel l L i n e s Using the same four different CD44 epitope-specif ic mAbs I next surveyed the staining profiles of cells f rom 5 established human myeloid leukemic cell lines that differ in their expression of features of granulopoietic, erythroid or megakaryopoiet ic differentiation. Representative staining profiles (up to 6 analyses per line) are shown in Figure 12A. All cell lines tested except K562 were consistently found to express high levels (3 logs over the background) of CD44 as assessed by their binding of 3c12 (data not shown) . The other 4 cell lines also reacted reproducibly (albeit to different extents) with 2G1 as well as with 7f4 and 8d8. To determine whether any of these staining patterns could be related to an ability of the cells to bind HA, addit ional cells f rom each line were stained with a soluble HA-FITC preparation. Only the two cell lines that expressed the highest levels of CD44 (KG 1a and M 0 7 e ) showed readily detectable binding of soluble HA which could be blocked by 3c12 as well as unlabelled HA to demonstrate the involvement of CD44. The other two cell lines, TF-1 and M B - 0 2 , al though positive for CD44, did not show detectable HA binding. As suggested by previous studies (Dougherty et al. , 1991), K562 cells showed no evidence of CD44 expression and thus served as a useful negative control in these exper iments. 4.2.3 CD44 I s o f o r m E x p r e s s i o n a n d H A - B i n d i n g F o l l o w i n g T P A S t i m u l a t i o n The expression of the epitopes detected by 2G1 and 7f4 on KG1a, M 0 7 e , TF-1 and M B 0 2 cells, as well as on primary human myeloid leukemic cells, suggested that 103 8G12 2G1 7f4 KG1A K562 M 0 7 E TF-1 M B 0 2 Si £ 3 0> o a> re ^ — H A - F I T C — • 8d8 o n l y + + un labe l l ed 3c12 HA L o g F l u o r e s c e n c e B 8G12 2G1 7f4 8d8 ^ — H A - F I T C — • o n l y + + un labe l l ed 3 d 2 HA K562 M 0 7 e v n E 3 0> o > re o cc h /Ml L l ' V A A. / t L o g F l u o r e s c e n c e 104 c L U 0. 00 T5 00 10" 103 < O 102 101 J 10" 10° 101 102 HA-FITC 103 10" 102 103 HA-FITC 10" F igu re 12. E f fec t o f T P A o n CD44 e x p r e s s i o n a n d HA-FITC b i n d i n g o f h u m a n m y e l o i d cel l l ines . (A) Human myeloid cell lines (KG1a: myelomonocyt ic, K562: erythroleukemic, M 0 7 e : megakaryoblast ic, T F - 1 : erythroleukemic, M B - 0 2 : erythroleukemic) were stained with ant i-CD34 mAb (8G12) or mAbs directed against different CD44 epitopes (2G1, 7f4, 8d8 - solid profiles, overlaid on the isotype control staining- open profiles) and/or HA-FITC (HA-FITC only-solid profile) overlaid on the control unstained cells (HA-FITC only-open profi les). The absence of CD34 expression on M 0 7 e cells reported by others to be present (Liesveld et al 1993; Levesque et al 1995) was conf i rmed by Northern blot analysis (data not shown). The specificity of the HA binding was tested by compet ing HA-FITC with unlabel led HA (+ unlabelled HA - solid profile) overlaid on both control unstained cells and on cells stained with HA-FITC (+ unlabelled HA - open profi les). The role of CD44 in mediat ing the HA binding detected was demonstrated by its inhibition in the presence of 3c12 (+ 3c12 - solid profiles) overlaid on both cells stained with isotype control ant ibody and on cells stained with HA-FITC (+ 3c12 -open profi les). ND = not done. (B) Effect of 20 hours of TPA treatment (60 ng/ml) of K562 cells and M 0 7 e cells on their staining with ant i-CD44 antibodies or HA-FITC (same descriptors as for panel A, except that a third, open profile, has been added to each of the data sets for 2 G 1 , 7f4 and 8d8 staining to indicate the results obtained for parallel staining of control (non-TPA treated) cells in the same experiment). (C) Dot plot of TPA-treated KG1a cells after double staining with either 2G1/GAM-PE or 8d8/GAM-PE and HA-FITC. 105 this might be indicative of an activated state shared by t ransformed myeloid cells and not by normal BM cells. In addit ion, it is known that the ability of CD44 to bind HA can be regulated separately from its expression (Lesley et al. , 1990; Lesley et al . , 1994; Bennett et al. , 1995b; Cooper and Dougherty, 1995; Katoh et al . , 1995; Lesley et al. , 1995) and may be elicited in CD44+ cells only after their activation by exposure to anti-CD44 antibodies or TPA (Lesley et al., 1993). It was, therefore, of interest to use these cell lines to investigate whether TPA treatment of leukemic cells might alter the expression of different CD44 epitopes on these cells and, if so, whether their HA-binding capacity would be affected. As shown in Figure 12B, TPA treatment for 20 hours induced the expression of CD44 on the previously negative K562 cells and also conferred on them the ability to bind soluble HA. TPA treatment of M 0 7 e cells also increased their overall level of CD44 expression (~ 3-fold) and HA-binding ability. However, CD44 expression on both KG1a and TF-1 cells was not significantly altered by TPA treatment (data not shown) nor did TPA treatment induce TF-1 cells to bind soluble HA. TPA treatment also did not significantly change the pre-existing HA-binding ability of KG1a cells but may have increased the affinity of HA binding since an increased amount of either 3c12 antibody or unlabelled HA was required to block the binding of labeled HA (as compared with control cells that had not been treated with TPA, data not shown). Interestingly double staining of TPA-st imulated KG1a cells with HA-FITC and either 2G1-GAM-PE or 8d8-GAM-PE showed that HA-binding to these cells increased as the level of expression of these 2 CD44 epitopes increased on the surface of these cells (Figure 12C). Doubly stained M 0 7 e cells also showed a concordant increase in HA-binding and CD44 epitope expression (data not shown) . 4.3 D i s c u s s i o n Specific increases or decreases in the expression of CD44 have been found to characterize many types of malignant cells (Cooper and Dougherty, 1995). Studies of t ransformed hematopoiet ic cells are more limited; however, in non-Hodgkin 's 106 l ymphomas, the presence of large amounts of CD44 protein on the cell surface has been correlated with a rapidly disseminating phenotype and poor prognosis (Horst et al. , 1990; Salles et al. , 1993; Koopman et al. , 1993). Overexpression of experimental ly introduced cDNAs for specific CD44 isoforms has also been found to confer metastatic potential (Gunthert et al. , 1991 ; Hofmann et al. , 1991 ; Cooper & Dougherty, 1995). HA, a natural l igand of CD44, is found at high concentrat ions in the extracellular matrix of numerous t issues where it surrounds migrating and proliferating cells (Underhil l , 1992). Such associat ions reinforce the notion that this l igand-receptor pair is involved in the dissemination of malignant cells. Given the well known ability of human leukemic populations to exit in large numbers f rom the BM into the PB, I have initiated a series of studies to investigate a potential role of deregulated CD44 activity in this process. In the present chapter I confirm the ubiquitous expression of the CD44 gene in normal human marrow cells and in primary leukemic cells obtained from patients with a spectrum of different myeloid leukemias, as well as on a number of established human myeloid leukemic cell l ines. However, I also demonstrate the differential expression on these cells of CD44+ epitopes that have not been previously examined. In addit ion, different leukemic cell l ine-specific effects of phorbol ester treatment on the expression and funct ion of the HA-binding region of CD44 have been revealed. The types of leukemia studied ranged f rom chronic phase CML, where differentiation per se is minimally perturbed, to a variety of FAB subtypes of A M L where there may be little evidence of any normal differentiation. In most leukemic samples, there was a significantly higher proport ion of cells expressing the 2 CD44 epitopes that were rarely found on normal BM cells, particularly when cells coexpressing CD34+ were compared. One of these is the epitope recognized by the 7f4 mAb. This epitope is found on only a small subset of light density cells in normal BM and does not include those with progenitor activity or any other CD34+ cells. In contrast, this epitope was found to be expressed at a high 107 level on the CD34+ cells in most of the primary leukemic samples studied and in all of those obtained from patients with AML. 7f4 recognizes an epitope that is not expressed on resting CD44+ T cells but is induced by activation of T cells upon PHA stimulation (Dougherty et al. , 1995). Similarly, expression of this epitope was found to be induced on the A M L cell line SR91 upon stimulation with TNF-oc (Dougherty et al. , 1995). It has, therefore, been suggested that expression of this epitope is dependent on conformational changes in or post-translational modifications of CD44. Activation of CD44 upon st imulation of a permissive cell may lead to an attachment of the cytoplasmic domain of CD44 to cytoskeletal components leading to their clustering on the cell surface and hence acquisit ion of a higher avidity/affinity for HA (Lesley et al. , 1993). Al though it is not clear that phorbol ester treatment affects the phosphorylat ion of CD44 in intact cells (Camp et al . , 1991 ; Lesley et al. , 1993), it has been proposed that PKC activation by TPA stimulation may result in phosphorylat ion of the cytoplasmic domain of CD44 and its subsequent binding to cytoskeletal proteins in vitro (Kalomiris & Bourguignon, 1988; Kalomiris & Bourguignon, 1989; Lokeshware t al. , 1994; Morimoto et al. , 1994). The presence of the 7f4 epitope on CD44 molecules expressed on CD34+ hematopoiet ic cells in leukemic but not normal samples suggests the presence of an "activated" form of CD44 on leukemic CD34+ cells. However, the individual levels characterist ic of any particular line or sample did not indicate any obvious pattern and there was also no clear associat ion between any of the epitopes expressed and the type of leukemia, the stage of differentiation, or the CD34+ or blast cell content of the primary leukemic samples analyzed. Some of the heterogeneity observed in the f requency of cells expressing particular CD44 epitopes on leukemic cells f rom different patients may be the result of changes in transcription of the CD44 gene (Hofmann et al., 1993; Herrl ich et al . , 1993; Jamal et al . , 1994) or in the post-translational modification of the CD44 molecule (Katoh et al. , 1995; Lesley et al., 1995; Bennett et al . , 1995b), either of which might 108 alter the adhesive properties of the CD44 expressed on the cell surface. Future studies may help to distinguish between these alternatives. In addit ion, variable amplif ication in different patients of leukemic cells expressing particular features of subsets of primitive normal BM cells could also contribute to a heterogeneous pattern of CD44 binding. Whether or not the observed changes are simply indicative of (or even irrelevant to) rather than part of the abnormal cell trafficking that character izes all of the myeloid leukemias examined here, including chronic phase CML, will also have to await further study using functional endpoints to identify potential consequences of altered CD44 epitope expression. Such studies using the long-term culture system as a model of the marrow microenvironment were undertaken and the results are presented in the next Chapter. As a further approach towards the investigation of the role of altered CD44 expression in the abnormal properties of human myeloid leukemic cells, I examined the HA-binding capacity of various myeloid leukemic cell lines and determined the effect of activation of PKC on their expression of CD44 and its HA-binding capacity. These studies showed an increase in CD44-mediated HA-binding activity in association with an increase in CD44 expression in K562 and M 0 7 e cells. KG1a cells showed an increase in CD44-mediated binding of soluble HA, as shown previously for immobil ized HA (Morimoto et al. , 1994), but without a detectable change in CD44 expression. Taken together, these findings suggest that PKC is likely to be involved in regulating CD44 levels and function in myeloid cells. In conclusion, I have further characterized CD44 expression on normal human marrow cells and have obtained new evidence of a deregulat ion in the expression of specific CD44 epitopes in human leukemia. In both primary samples and establ ished lines of human myeloid leukemic cells this included cells expressing the CD34 ant igen, which is generally considered as a phenotypic marker of primitive hematopoietic cells. In particular, the consistent increase in expression of the 7f4 epitope (which has been associated with T cell 109 activation) on CD34+ cells f rom leukemic patients and, in particular, f rom with AML, suggests a common functional alteration of properties that characterize primitive normal hematopoietic cells. no Chapter 5 Stromal-Dependent Hematopoiesis in Cultures of Chronic Myeloid Leukemia Cells is Unaffected by Anti-CD44 Antibodies that Inhibit or Stimulate Normal Hematopoietic Cells data presented in this chapter is included in the fol lowing manuscript: Ghaffari , S., Dougherty, G.J., Eaves, A.C., and Eaves, C.J. Stromal-Dependent Hematopoiesis in Cultures of Chronic Myeloid Leukemia Cells is Unaffected by Ant i -CD44 Antibodies that Inhibit or Stimulate Normal Hematopoiet ic Cells (to be submitted for publication) 5.1 I n t r o d u c t i o n Adult hematopoiesis takes place in the bone marrow where hematopoiet ic cells interact with a complex molecular environment. This includes interactions with l igands and receptors expressed on the surface of adjacent stromal cells as well as the multiple components of the extracellular matrix produced by these cells (Verfaillie et al. , 1994a). Such interactions are believed to modulate the viability, cycling status and movement of hematopoiet ic ells and thereby regulate their rate of amplif ication and ultimate output of mature progeny (Verfaillie et al., 1994a). The LTC system appears to be a useful model for analyzing the molecular mechanisms involved since this system is character ized by the presence of a stromal cell-containing adherent layer that can both sustain and regulate hematopoiesis for extensive periods of t ime (Eaves and Eaves, 1987). Moreover, it has been shown that the defective cycl ing control exhibited by the primitive neoplastic progenitors in patients with CML is reproduced in the adherent layer of LTC containing these cells (Eaves et al. , 1986). In Chapters 3 and 4, I showed that the expression of CD44 is altered on primitive CML progenitors (CFC, LTC-IC and CD34+cel ls) . The variety of regulatable binding properties of CD44, together with the observation that antibodies specific for in the HA-binding site of CD44 can inhibit hematopoiesis in normal murine LTC (Miyake et al. , 1990a; 1990b) suggested that CD44 may play a key role in facil itating the regulatory effects of stromal cells on the hematopoietic cells with which they interact. To investigate this hypothesis, I have examined and compared the effects of different ant i -CD44 monoclonal antibodies (mAb) on normal and leukemic (CML) hematopoiesis in a number of in vitro systems. The mAb studies included three (3c12, 7f4, 8d8) of the four mAbs used in studies presented in Chapters 3 and 4. 5.2 R e s u l t s 5.2.1 Charac te r i za t i on of the Reac t iv i t y of T w o Nove l Ep i tope -Spec i f i c A n t i - C D 4 4 m A b s (7f4 a n d 8d8) w i t h Norma l a n d L e u k e m i c (CML) H e m a t o p o i e t i c Ce l ls Assessment of the progenitor content of the positive and negatively stained fractions of light density normal BM showed that both CFC and LTC-IC were all 7 f4 _ as were most of the other cells, including those expressing CD34. In contrast, 8d8, which is expressed on a large proportion of light density and CD34+ normal BM cells (~ 50%) was found to react with - 3 0 - 5 0 % of the CFC compartment but < 5% of the LTC-IC in the same suspensions. As reported in Chapter 3, all CFC and LTC-IC express 3c12. Recovery values after staining showed no evidence of ant i -CD44 specific losses of progenitor activity (or viability). A similar analysis of CML progenitors was carried out using cryopreserved samples of PB from chronic phase patients with high W B C counts previously shown to contain exclusively Ph+ CFC and LTC-IC (Petzer et al. , 1996c). As reported in Chapter 4, a larger proportion of neoplastic LTC-IC (or CFU-GM) expressed very high levels of CD44 than is characteristic of normal LTC-IC (or C F U -GM) . Since the staining profiles of CML and normal CD34+ cells with 7f4 and 8d8 was not significantly different more detailed analyses of the reactivity of these two mAbs with functionally defined subpopulat ions of CD34+ cells were not undertaken. 112 5.2.2 O p p o s i t e Ef fec ts of Ep i tope-Spec i f i c An t i -CD44 A n t i b o d i e s o n t h e P r o d u c t i o n a n d Ma in tenance o f N o r m a l B o n e M a r r o w C l o n o g e n i c Cel ls a n d LTC-IC in LTC The effect of the addition of three ant i-CD44 mAbs (described in Chapters 3 and 4) on hematopoiesis was tested in vitro. 3c12 recognizes an epitope involved in hyaluronan-recognit ion on CD44 (Dougherty et al. , 1994), and 7f4 and 8d8 recognize distinct epitopes on the common region of CD44 in activated cells (Dougherty et al. , 1995). 3c12 epitope is present on all clonogenic cells including LTC-IC (results presented in Chapter 3), In a first series of experiments, potential effects of each of the three CD44" epitope-specif ic mAbs studied on normal LTC-IC maintenance and/or production of CFC progeny in LTC was assessed. 20 | ig/ml of purified mAbs (or F (ab ' ) 2 f ragments) were added to the culture medium at the t ime the test cells were initially seeded onto preestabl ished, irradiated adherent feeder layers and weekly thereafter at the t ime of each weekly change of half of the medium. Five weeks later the nonadherent and adherent fractions of each culture were harvested and an aliquot of each plated in secondary assays (in the absence of Abs) to measure the total number of CFC and LTC-IC present. As can be seen in Fig 13 (open bars) in the presence of the ant ibody against the HA-binding site of CD44 (3c12), CFC generat ion in these cultures was markedly (~50-fold) and specifically inhibited (relative to control cultures) to which no Ab was added, (p<0.001, one-tailed Mest n=6) since, in parallel cultures, the addition of other mAbs of the same (e.g., ant i-CD34, 8G12), or a different isotype (ant i-CD45, 9.4) reactive with both LTC-IC and their progeny or not (anti-dextran, DX1) has no such effect. Moreover, the inhibitory effect of 3c12 was similar in cultures initiated with unseparated (light density) BM cells (Figure 13A) or a derived CD34++CD38" fraction enr iched in its LTC-IC content (Sauvageau et al. , 1994) (Figure 13B) and was not diminished when F (ab ' ) 2 f ragments were used instead of intact Abs (Figure 13B) nor when the concentrat ion of 3c12 was decreased 100-fold (to 200 ng/ml). The inhibitory 113 _J o cc I -z o o LL o 300 H 200 H 100H ANTIBODY: 3c12 SPECIFICITY 600-8G12 DX1 CD45 CD34 Dex t ran B O CC I -z o o u. o 500-400-300-200: 100H X H N D ND ND ND ANTIBODY: 3c12 QMQ 3 c 1 2 F(ab')2 8 d 8 7f4 8G12 + 8G12 p f .L^ DX1 8d8 t ( a o >2 SPECIFICITY: •CD44 MD34- Dex t ran F igure 13. O p p o s i n g e f fec ts of an t i -CD44 m o n o c l o n a l a n t i b o d i e s o n h e m a t o p o i e s i s in 5 w e e k l ong t e r m cu l tu re . Total CFC Q ) and LTC-IC ( | ) present at 5 weeks in LTC initiated with normal bone marrow light density mononuclear cells (A) or highly purified CD34+CD38- cells (B) established on human feeder layer of stromal cells. In all experiments the number of clonogenic cells produced in the presence of mAbs is reported as a percentage of control cultures that were established in the absence of any antibodies. All anti-CD44 antibodies were of lgG1 isotype. Control mAbs, anti-dextran and anti-CD34 (DX1 and 8G12) are also lgG1 and anti-CD45 (9.4) is lgG2a. Dextran is not expressed by mammalian cells, CD34 is expressed by all hematopoietic clonogenic cells and CD45 is expressed by majority of hematopoietic cells including those with clonogenic activity. Values shown are the mean ± SEM of measurements evaluated in at least 3 independent experiments. ND, not done. 114 effect of 3 d 2 must have occurred at the level of LTC-IC since the number of these cells present in the same 5 week-old LTC was also markedly decreased (p<0.001, one-tailed f-test, Figure 13, solid bars). In light of these observat ions, the f inding that the addit ion of 8d8, either alone at 20 | ig/ml or together with 7f4 (Figure 13), or even alone at 500 ng/ml, significantly enhanced CFC production (p<0.001, one-tai led Mest n=6) in the same experiments is particularly striking. A similar effect of 8d8 on LTC-IC maintenance was also observed, but these were more viable and did not reach statistical signif icance. To determine whether the opposite effects of 3 d 2 and 8d8 seen on LTC-IC funct ion after 5 weeks in LTC were unique to this progenitor population or whether they might also extend to later stages of hematopoiesis, additional experiments of the same design were carried out but progenitor assays were performed at the end of one rather then 5 weeks. Not only were similar effects observed (Figure 14), it is interesting to note that both the inhibitory actions of 3 d 2 and the stimulatory actions of 8d8 on CFC and LTC-IC numbers showed the same differences previously noted in terms of the average level of CD44 expression on these populations (Chapter 3), i.e. greater effects being associated with higher surface expression of CD44 which is characteristic of both LTC-IC and their commit ted granulopoietic progeny (CFU-GM) by compar ison to primitive erythroid progenitors (BFU-E) (Chapter 3). On the other hand, there was no evidence that simple exposure of LTC-IC to 3 d 2 altered their differentiation potential since the - 9 : 1 ratio of CFU-GM to BFU-E in their 5 week progeny CFC generated subsequent ly in the absence of any ant i-CD44 Abs remained unchanged. To investigate the possibility that the observed effects of 3 d 2 or 8d8 might be explained by a change in the proportion of LTC-IC and/or CFC retained in the adherent layer, the distribution of these cells between the adherent and nonadherent fractions of the cultures was examined after both one and 5 weeks. However, no evidence of any such effect was seen (Table 7). 115 250 200 150 100 4 50 A O DC O O 0 3 d 2 7f4 8d8 7f4 9.4 8G12 DX1 8d8 B ANTIBODY: 3c12 7f4 8d8 9.4 8G12 DX1 SPECIFICITY: CD44 CD45 CD34 Dextran Figure 14. An t i -CD44 m o n o c l o n a l a n t i b o d i e s af fect h e m a t o p o i e s i s in 1 w e e k l o n g - t e r m cu l t u re . CFC (A) and LTC-IC (B) present at 1 week in the nonadherent ( N A f J ) and adherent ( A D H H ) fractions of LTC of normal bone marrow light density mononuclear cells established on human feeder layer of stromal cells. Values determined as in Figure 13. 116 Tab le 7. D i s t r i b u t i o n of CFC b e t w e e n adheren t a n d n o n a d h e r e n t f r a c t i o n s of LTC Week 1 CFC W e e k 5 CFC No A b 0.3 ± 0.02 0.42 ± 0.02 8d8 0.42 ± 0.01 0.3 ± 0.02 7f4 0.38 ± 0.01 0.44 ± 0.05 7f4 + 8d8 0.46 ± 0 0.33 ± 0.04 8G12 0.43 ± 0 . 0 2 0.5 ± 0.06 The ratio of CFC in the non adherent fraction of LTC was measured (values shown the mean of at least 3 independent experiments ± SEM) 117 To determine whether any of the ant i-CD44 Ab effects observed on normal hematopoiesis in LTC might reflect direct independent effects on the progenitors, themselves, the same mAbs were added to parallel suspension cultures of either light density or CD34++CD38" normal BM cells which did not contain a preestabl ished feeder layer but to which relatively low concentrat ions of soluble Steel factor (50 ng/ml) and IL-6 (20 ng/ml) were added instead as a substitute mechanism of support ing their viability and proliferation. As shown for the cultures initiated with light density cells, al though the numbers of LTC-IC and CFC detected 1 week later in these suspension cultures were similar to those measured under control LTC condit ions, both the 3c12 and 8d8 effects disappeared in the absence of st roma (Figure 15A). The same result was obtained for the cultures initiated with CD34++CD38" cells (Figure 15B). 5.2.3 L a c k of An t i -CD44 Ep i tope-Spec i f i c E f fec ts o n Pr im i t i ve N e o p l a s t i c H e m a t o p o i e t i c P r o g e n i t o r s f r o m C h r o n i c Mye lo id L e u k e m i c Pa t ien ts in LTC In Chapter 3 and 4 I described the alterations of CD44 epitopes expressed on primitive hematopoiet ic leukemic cells f rom CML patients and also demonstrated that the level of CD44 expression on CFU-GM and LTC-IC and not on BFU-E is very high as compared to their normal counterparts. In view of the above evidence that CD44 is involved in the stromal cel l-mediated regulation of normal hematopoiesis, it was interesting to examine how each of the mAbs studied might affect P h + hematopoiesis under similar condit ions. Accordingly, a series of experiments were then carried in which light density cells from 3 CML patients with high W B C and exclusively Ph+ progenitors were cultured for one or 5 weeks on normal marrow adherent layers in the presence or absence of various mAbs. As shown in Figure 16, neither the inhibitory effects of 3c12 nor the stimulatory effects of 8d8 obtained on normal marrow progenitors were seen with the exception of the 5 week LTC-IC values where some slight inhibition was apparent. However, it should be noted that Ph+ LTC-IC decline much more rapidly in vitro then do their normal counterparts (Petzer et al. , 1996c). As 118 A 200 ANTIBODY: 3c12 7f4 8d8 9.4 8G12 SPECIFICITY: - 4 CD44 • CD45 CD34 F igu re 15. L a c k of e f fec t of an t i -CD44 m o n o c l o n a l a n t i b o d i e s o n h e m a t o p o i e s i s in l i qu id cu l tu re . Total CFC ( f j ) and LTC-IC ( | ) generated after one week from samples of total light density mononuclear cells (A) or highly purified CD34+CD38- cells (B) in suspension cultures containing IL-6 and Steel factor. Values determined as in Figure 13. 119 o 15CH •\oo-i 50 H B O cc H o o 250 200 A 150H 100H ANTIBODY: 3 d 2 7f4 8d8 8G12 SPECIFICITY: CD44 • CD34 Figure 16. A b s e n c e of e f fec t of an t i -CD44 m o n o c l o n a l a n t i b o d i e s o n C M L h e m a t o p o i e s i s in l ong - te rm cu l tu re . Total CFC ( f j ) and LTC-IC fl) generated from light density mononuclear cells of PB of 3 CML patients after 1 week (A) and 5 weeks (B) in LTC established on human marrow feeder layer of stromal cells. Values determined as in Figure 13. 120 a result, the absolute numbers of LTC-IC present in the LTC of Ph+ cells, even in the absence of any Abs, was much lower than in the normal LTC and the accuracy of any potentially inhibitory effects on this populations would therefore be correspondingly reduced. 5.3 D i s c u s s i o n The localization of adult hematopoiesis in the bone marrow microenvironment al lows a direct interaction of primitive hematopoietic cells with various components of the extracellular matrix. These interactions are believed to regulate the cycl ing status of primitive hematopoiet ic cells and their subsequent differentiation. Many growth factors and adhesion molecules are involved in these interactions, however their relative importance remains to be defined (Verfaillie et al. , 1994a). Mitogenic and adhesion signaling pathways that regulate cell proliferation, differentiation, p lasma membrane attachment and migration meet in cytoskeleton and the l inkage between these pathways has just begun to be understood (Ridley, 1994; Gumbiner, 1996). Ant i -CD44 monoclonal antibodies have been found to have an inhibitory or activating functional effects by modulating CD44 conformation, distribution, l igand-binding or signaling in a variety of cell systems (Zheng et al. , 1995). These include among others cell adhesion, lymphocyte activation and growth factor product ion, GTP-binding protein and GTP-ase activity (Lesley et al. , 1993a). CD44 isoforms can also function as proteoglycans in binding and presenting growth factors (Bennett et al. , 1995a; Jackson et al. , 1995). Variant forms of CD44, exhibit ing unique l igand-binding and functional properties, are also produced during cell differentiation, t ransformation or cell cultured under specific condit ions (Lesley et al., 1993a). The cytoplasmic domain of CD44 has been found associated with PKC and several cytoskeletal proteins such as ankyrin, actin and ERM (ezrin, radixin, moesin) family members (Lacy and Underhil l , 1987; Carter and Wayner, 1988; Kalomiris and Bourguignon, 1988; Tsukita et al. , 1994). Assembly of CD44 cytoplasmic domain with 121 cytoskeletal proteins and phosphorylat ion of these complexes that may occur during cell differentiation, regulate the cell surface distribution and hyaluronan-binding of CD44 (Lesley et al. , 1992; 1993; Perschl et al. , 1995; Pure et al. , 1995). Al though CD44 is expressed at high levels on both primitive hematopoiet ic and stromal cells, and CD44 on myeloid cells has clearly the ability to bind hyaluronan (Chapter 4; Morimoto et al. , 1994), the binding of primary primitive hematopoiet ic cells to hyaluronan through CD44 is not clear. Moreover, the mechanism regulating CD44-hyaluronan-binding of myeloid hematopoietic cells is still unknown (Lesley et al . , 1993). Ant ibodies that recognize the hyaluronan-binding site of CD44 have been previously reported to inhibit the production of mature cells in murine lympho-myeloid LTC (Miyake et al. , 1990a; 1990b). Interestingly, an antibody that recognizes an epitope within the proteoglycan homologous region but distinct f rom the hyaluronan-binding site of CD44 (Culty et al., 1990; Peach et al. , 1993) has also been reported to have an inhibitory effect on hematopoiesis in LTC (Gunji et al. , 1992). Moreover, antibodies that recognize the hyaluronan-binding site of CD44 have also been found to inhibit CD44-binding to other ligands (e.g. serglycin or osteopontin) (Toyama-Sor imachi et al. , 1995; Weber et al. , 1996). In this chapter I have examined the effects of addition of various newly descr ibed ant i -CD44 mAbs on the production of CFC and LTC-IC at 1 and 5 weeks in LTC of human hematopoiet ic cells initiated on preestablished irradiated feeder layer of human stromal cells. I have found that ant i-CD44 antibodies can have multiple effects, i.e. either a posit ive, a negative or no effect, on the production or maintenance of LTC-IC in LTC. Al though some direct effect of ant i -CD44 mAbs on primitive hematopoiet ic cell differentiation and / or proliferation is not ruled out, it is clear that CD44 has multiple functions in hematopoiet ic-stromal interactions involved in LTC hematopoiesis. The repeated addit ion of 3c12 mAb that recognizes the hyaluronan-binding site of CD44 resulted in a complete inhibition of production of CFC in 5 week LTC. The inhibitory 122 effect of 3c12 was not mediated through non-specific Fc receptor binding since similar effect was seen with 3c12 whole molecule or F (ab ' ) 2 f ragments. The inhibitory effect of 3c12 was occurring at the level of LTC-IC since no CFC was found in the secondary 5 week LTC initiated at the end of the first 5 weeks in the absence of antibodies. These results also indicated that 3c12 did not retain LTC-IC in the adherent fraction of LTC. A single addit ion of 3c12 resulted also in a similar inhibition of the production of CFC and LTC-IC after one week. These results indicate that CD44-l igand interactions through the hyaluronan-binding site of CD44 are important for proliferation and differentiation of the most primitive hematopoietic cell in LTC. In contrast addition of 8d8 mAb that recognizes an epitope within the common region of CD44 distinct f rom 3c12 and not involved in hyaluronan-binding of CD44, enhanced the CFC output after one and five weeks in LTC. 8d8 and 3c12 added together did not prevent significantly the inhibitory effect of 3c12 (data not shown). On the other hand the st imulating effect of 8d8 on hematopoiet ic cell proliferation and differentiation may be explained at different levels. 8d8 may induce homotypic aggregation of CD44 on hematopoiet ic and stromal cells and increase the efficiency of interactions of the two cell types. 8d8 may enhance the proteoglycan function of CD44 and therefore increase its ability to bind and to present growth factors. Alternatively 8d8 may induce stromal cells to produce st imulating growth factors, as it has been shown for other ant i -CD44 mAbs (Webb et al . , 1990). As I have described in Chapters 3 and 4, cells f rom patients with chronic myeloid leukemia express, in addition to an abnormally high level of CD44 on CFU-GM and LTC-IC, new forms of CD44 on their cell surface. In order to assess the function of CD44 in interactions of primitive CML hematopoietic cells with s t roma I investigated the effect of ant i -CD44 antibodies on the production of Ph+ CML CFC and LTC-IC in LTC. Interestingly 3 d 2 and 8d8 did not affect the production of CFC and LTC-IC at one week. 8d8 did not affect the production of CFC in 5 week LTC initiated with Phu-C M L samples and the repeated addition of 3 d 2 inhibited only minimally the 123 maintenance of Ph+ CFC produced in a secondary 5 week LTC. These results suggested that primitive hematopoietic cells of CML samples can ignore the effect of ant i -CD44 antibodies and the CD44 pathway is altered in these cells. Primitive Ph+ hematopoietic CML cells have been described to exhibit several adhesive defects (Verfaillie et al. , 1994a). Mitogenic pathway and adhesion signaling are linked at the level of small GTP-binding proteins of Rho family (Ridley and Hall, 1992; Z igmond, 1996). Various motifs within Bcr protein may be involved in act in-f i lament binding of Bcr-Abl (Ridley, 1994; McWhirter and Wang , 1991) . In addit ion, in Bcr-Abl t ransformed cells, the components of focal adhesion, that is the site of cell-extracellular matrix attachment and actin f i lament-adhesion molecules assembly (Gumbiner, 1996), such as focal adhesion kinase ( p 1 2 5 F A K ) (Gotoh et al. , 1995) and cytoskeletal proteins paxillin, talin, vinculin and tensin are all phosphorylated and some have been found associated with Bcr-Abl (Gotoh et al. , 1995; Salgia et al . , 1995a; 1995b). The assembly of focal adhesion is regulated by small GTP-binding proteins of Rho family (Ridley and Hall, 1992). Al though CD44 has not been found in the focal adhesion, the involvement of GTP-binding proteins of Rho family in the regulation of CD44 adhesive system may be considered (Craig and Johnson, 1996). The abnormal phosphorylat ion and assembly of cytoskeletal proteins in addit ion to an abnormal ly high expression of CD44 may explain the abnormal function of CD44 pathway in primitive hematopoiet ic Bcr-Abl t ransformed progenitor cells. In conclusion, I have found that CD44 may have multiple functions in the normal adhesive interactions of primitive hematopoietic cells with st roma. The coincident f inding of an upregulation in the expression of CD44 on very primitive hematopoiet ic neoplastic cells in CML (Chapter 3), and an associated failure of these cells to respond to ant i-CD44 epitope specific interference (both positive and negative) of stromal-cell mediated regulation support of primitive stages of normal hematopoiesis in vitro, suggests an involvement of CD44 in the pathogenesis of this disease in vivo. 124 Chapter 6 Summary and Perspectives Hematopoiesis is a complex and dynamic process that assures the daily product ion, replenishment, and release of mature hematopoiet ic cells in the peripheral blood. Regulation of hematopoiesis occurs at multiple levels with multiple cross-talks. In adults, hematopoiesis occurs in the bone marrow microenvironment where close proximity of stromal elements largely regulates hematopoiet ic cell behaviour (Tavassoli , 1989). It seems clear that adhesion molecules expressed on both hematopoietic and stromal cells, in cooperat ion with growth factor receptors and ECM, assure not only the cellular contact but actively participate in the regulation of cellular behaviour (Juliano and Haskil l , 1993). In the present thesis I have focused on an examinat ion of the potential role of the CD44 family of adhesion molecules in the control of hematopoiet ic cell proliferation and differentiation. Bone marrow hematopoiesis is the result of a balanced regulation of both positive and negative-acting factors, and these studies have implicated CD44 in this process. I have first shown that CD44 expression is downregulated during the development of the erythroid l ineage (Chapter 3). Al though the majority of hematopoiet ic progenitors, including LTC-IC, express the same level of CD44, about 3 0 % of CFU-GM and LTC-IC, but not BFU-E, express higher levels (10 fold higher) of CD44. It is not known whether these subsets of CFU-GM and LTC-IC are any different in their proliferation or differentiation capacity. It is interesting, however, that in CML where clonogenic cells are in a highly proliferative state, the majority (up to 70%) of CFU-GM and LTC-IC express the same very high levels of CD44 (Chapter 3). Al though signaling through CD44 has not been extensively studied, a recent report on the involvement of tyrosine kinases in the signaling of CD44 in T cells (Taher et al. , 1996) indicates that 125 CD44 may be directly involved in cell proliferation. It would therefore be interesting to examine the proliferative potential of different subsets of hematopoiet ic cells according to their level of CD44 expression, in particular, to compare the proliferative activity of CFU-GM and LTC-IC expressing very high levels to the rest of these subpopulat ions. During the past few years a large body of data has been generated showing that higher molecular weight CD44 isoforms are produced in malignant cells, some of which are correlated with poor prognosis (Cooper and Dougherty, 1995). Data presented in this thesis show that different forms of CD44 are also produced in human leukemias, with a pattern distinct f rom their normal counterparts (Chapter 4). The frequency of expression of CD44 v10 as well as an act ivated form of CD44 expressing the 7f4 epitope is increased in both total and CD34+ mononuclear cells in human leukemias (Chapter 4) . The expression of v10 containing CD44, may confer new ligand binding abilities to leukemic cells and allow them to leave the normal sites of hematopoiesis. These cells may also exhibit proliferative advantages over the normal cells for example by no longer being subject to the negative regulation of the BM ECM. However, the expression of these isoforms does not seem to be restricted to malignant cells since these isoforms are also produced by up to 10% of normal bone marrow mononuclear cells that do not include cells with progenitor activity (Chapter 3). These f indings, along with recently reported data by other investigators, indicate that expression of higher molecular weight CD44 isoforms is not only a feature of mal ignancy. It is, however, possible that CD44 isoform expression characterizes highly differentiative cells. The function of CD44 isoforms in cell migration is still not clear. In particular, the ultimate functional roles of CD44-HA-binding are not known and it is not clear how the HA-binding of CD44 contributes to the cell migration and metastasis. 126 The expression of a specific, activated CD44 epitope on leukemic cells may indicate an activated state of CD44 on these cells. This epitope was specifically expressed on TNF-oc treated leukemic cell lines and not on untreated cells (Dougherty et al. , 1995). The expression of this epitope may also indicate glycosylation alterations that often accompany leukemogenesis or some conformational change of the CD44 molecule. Ant ibodies recognizing the HA-binding site of CD44 have been previously reported to inhibit the production of mature cells in the nonadherent fraction of murine LTC (Miyake et al. , 1990a; 1990b). I have shown that this effect is taking place at the level of LTC-IC since the 3c12 mAb that interferes with the HA-binding of CD44 inhibits completely and irreversibly the production of LTC-IC in human LTC after 5 weeks and more than 6 0 % of LTC-IC after 1 week (Chapter 5). In contrast to the effect seen with 3c12, the addition of 8d8 mAb was found to increase the production of primitive hematopoietic progenitors in a st roma-dependent manner. Interestingly, the primitive Bcr-Abl t ransformed cells in C M L were not affected by the presence of 3c12 nor 8d8 in 5 week LTC (Chapter 5) . Al though HA is highly synthesized in both the bone marrow ECM and in LTC, and hematopoiet ic cells have been found to be coated with HA in LTC, little direct evidence of HA-binding by primary hematopoietic cells exists. My work, as well as that of other investigators, indicates that hematopoiet ic cell lines can bind HA, al though the molecular mechanisms regulating this binding are unknown. CD44-HA binding may be important for stromal dependent hematopoiesis, and activated primitive hematopoietic cells may require HA-binding for their non-directional movement (Figure 17). This interaction may also confer a certain shape to the cells indispensable for their proliferation. It is known that certain ECM elements such as fibronectin can regulate hematopoietic cell proliferation/differentiation and/or survival (Patel and Lodish, 1984; Sugahara et al . , 1994). The interaction of primitive hematopoietic cell CD44 with HA in 127 Hematopoietic Progenitor Enhancement (8d8) or prevention (3c12) of CD44 binding to its ligand (HA, CD44,...). Activation through CD44 resulting in production of stimulating (8d8) or inhibitory (3c12) growth factor(s). Induction (8d8) or prevention (3c12) of growth factor binding to CD44. F igu re 17. E f fec ts o f An t i -CD44 M o n o c l o n a l A n t i b o d i e s o n H e m a t o p o i e s i s in v i t r o . Three putative mechanist ic explanations of the positive (8d8) and negative ( 3 d 2) effects of these antibodies on hematopoiesis in vitro. 128 response to mitogenic signals may be a weak and transient interaction that is tightly regulated. Alternatively, 3c12 may induce a stromal cell to release an inhibitory factor or prevent the presentation of a stimulatory factor by a CD44 proteoglycan to the primitive hematopoiet ic cell as it has been observed with other ant i -CD44 mAbs (Webb et al. , 1990; Noble et al. , 1993). Over the past few years, the importance of proteoglycans in the binding, stabil ization, presentation of growth factors, and the regulation of growth factor receptor-binding has attracted the attention of many investigators (Ruoslahti , 1989; Schlessinger et al., 1995). The core protein of proteoglycans is covalently at tached to GAGs, which are produced at high concentrat ions in both the bone marrow ECM and LTC. An increase in the synthesis of a particular G A G , chondroit in sulfate, is correlated with an increase in the production of primitive hematopoiet ic cells, including CFU-S, in LTC (Gallagher et al. , 1983a; 1983b; Spooncer et al . , 1983). The proteoglycan form of CD44, modif ied by chondroit in sulfate, has been found to be expressed on hematopoietic progenitor cells (Verfaillie et al . , 1994b). CD44 has also been shown to bind growth factors, a property it shares with other proteoglycans (Tanaka et al . , 1993; Bennett et al . , 1995a; Jackson et al. , 1995). An increase in the chondroit in sulfate-modif ied form of CD44 may be induced by the 8d8 antibody, thereby introducing a conformational change in the CD44 molecule and activating its proteoglycan state on primitive hematopoiet ic cells in LTC (Figure 17). In this case the chondroit in sulfate modif ied form of CD44 should bind HA with a lower affinity since the glycosylated CD44 has a lower binding affinity for HA. The lack of an 8d8 effect in liquid culture supplemented with growth factors, may be due to the fact that in these condit ions, growth factor receptors are already saturated. Alternatively, 8d8 may be inducing stromal cells to produce an activating factor as it has been previously reported for other mAbs directed against CD44 (Webb et al. , 1990; 129 Noble et al . , 1993). CD44, which is capable of binding to itself (Droll et al. , 1995), is expressed on both primitive hematopoietic cells and stromal cells and ant i -CD44 antibodies partially dissociate naturally occurring aggregates of these cell types (Funk et al., 1994). This aggregation may be mediated by HA which is making a bridge between different cell types or by homoaggregat ion of CD44 on different cell types (Lesley et al . , 1990). It is possible that 8d8 is increasing the affinity of CD44 on hematopoiet ic and/or stromal cells for either HA, CD44 or other l igands expressed on opposing cells. Cell proliferation and division is the result of a dynamic cooperat ion between various pathways controll ing the cytoskeletal organization and gene expression (Gumbiner, 1996; Craig and Johnson; 1996). In response to environmental signals, the cell changes in both shape and its degree of at tachment to the ECM substratum (Figure 18). The harmony in cross-talk between pathways is crucial for cell division to occur. Recent evidence indicates that growth factor-receptor binding resulting in gene expression and mitogenesis by signaling through alternative Ras pathways, also tr iggers different cytoskeletal pathways (Ridley et al. , 1992; Ridley and Hall, 1992). These signals act through one or more members of the Rho family of small GTP-binding proteins, including Cdc42, Rac and Rho (Nobes and Hall, 1995). Each of these proteins regulates a unique morphological change that involves rearrangement of actin f i laments (F-actin). Cdc42 regulates the f i lopodia, Rac regulates the membrane ruffles, and Rho regulates the stress fibers (Zigmond, 1996). Al though activated Ras induces the reorganization of the cytoskeleton, the control of this pathway is independent of the mitogenic activity of Ras (Ridley and Hall, 1992; Joneson et al . , 1996). The GTP-binding proteins have also been found to regulate the binding of cytoskeletal proteins to actin f i laments (Craig and Johnson, 1996). Several cytoskeletal proteins are specifically phosphorylated by signals generated during 130 Control of Cell Growth and Gene Expression Control of Cytoskeletal Reorganization, Cell Adhesion, Shape and Moti l i ty ECM ECM Growth factor .Focal Adhesio Assembly I Actin Stress Fibre Formation Actin Membrane Ruffling Filopodia N U C L E U S F igu re 18. M i t o g e n i c v e r s u s Cy toske le ta l P a t h w a y s . I3l cell at tachments to the ECM. These result in a clustering of adhesion receptors and cytoskeletal proteins in FA sites. Rho proteins are necessary for stress fiber formation and FA. Integrins are the major adhesion receptors in the FA (Craig and Johnson, 1996; Gumbiner, 1996). However, integrins mediate f irm at tachment and transient lower binding affinity may be mediated by selectins and CD44. Parallel association of these adhesion receptors and cytoskeletal proteins may occur in other sites of cell attachment to the ECM (Craig and Johnson, 1996, Z igmond, 1996). In this regard the fact that on primitive hematopoiet ic cells the binding to f ibronectin appears be the result of a cooperat ion between CD44 and (31 integrins is of particular interest (Verfaill ie et al. , 1994b). It is also interesting that ERM family member proteins that are reported to bind CD44 cytoplasmic domain, regulate the lamell ipodia and f i lopodia format ion. These proteins are found in membrane ruffles and are substrate for several kinases (Zigmond, 1996). In Bcr-Abl t ransformed cells, the mitogenic Ras pathway is constitutively activated (Figure 19) (Pendergast et al. , 1993; Puil et al. , 1994; Tauchi et al. , 1994; Skorski et al., 1994; Gishizky et al. , 1995; Raitano et al. , 1995; Sawyers et al. , 1995). In addit ion, several cytoskeletal proteins of FA are constitutively phosphorylated, indicating that one or more pathways controll ing actin reorganization have been impaired (Gotoh et al. , 1995; Salgia et al. , 1995a; 1995b; 1996). Activation of Ras has been found associated with an increase CD44 promoter activity and a moderate increase in the ratio of CD44v/CD44s transcripts as well as low level expression of the v6 epitope, suggest ing a possible role for ras in the regulation of CD44 transcript splicing (Hofmann et al . , 1993; Jamal et al . , 1994, Penno et al. , 1994). It is likely, that Bcr-Abl t ransformation results not only in the upregulation of CD44 expression and splicing via Ras pathway, but also in the activation of cytoskeletal pathways involving GTP-binding proteins of the Rho family. Adhesion of CD44 to the 132 Control of Cell Growth and Gene Expression Control of Cytoskeletal Reorganization, Cell Adhesion, Shape and Motility F igu re 19. Cy toske le ta l A l t e r a t i o n s in B C R - A B L T r a n s f o r m e d Ce l ls . Activation of the Ras Pathway in Bcr-Abl transformed cells may result in the upregulation of CD44 isoform expression observed in primitive hematopoietic progenitors. CD44 - ligand binding and CD44 signaling have both been found to require an intact cytoskeleton in some cells. In Bcr-Abl transformed cells, several cytoskeletal proteins have been found constitutively phosphorylated and at least one of them (Paxillin) has been found physically associated with Bcr-Abl. In Bcr-Abl transformed cells pathways controlling the reorganization of cytoskeleton may be altered (constitutively activated). This would result in non functional adhesion molecules whose ligand-binding and / or signaling require an intact cytoskeleton. 133 ECM, and in particular CD44-HA binding, requires an intact cytoskeleton that is not provided in Bcr-Abl t ransformed cells. CD44 binding functions and / or signaling pathways are therefore likely impaired in Bcr-Abl t ransformed C M L cells. The model I propose (Figure 19) is that mitogenic signals st imulate alternative Rho pathways that control the reorganization of the cytoskeleton and CD44-HA binding, both of which are important for primitive hematopoiet ic cell division. This however, may be a very transient interaction and highly regulated. Interruption of this interaction (i.e. by 3c12 mAb) in a primitive hematopoiet ic cell which has received a mitogenic signal would cause the cell to undergo apoptosis. When cells are activated, CD44 would be rearranged on the cell surface via cytoskeleton rearrangements and its binding affinity would be increased. 8d8 may induce HA-binding or alternatively the proteoglycan form of CD44 may be activated in the presence of 8d8 and this would contr ibute to the CD44 potential to present growth factors. Bcr-Abl t ransformed cells may ignore these effects since both mitogenic and cytoskeletal pathways are constitutively act ivated. Taken together these studies underline the involvement of CD44 in normal processes that regulate hematopoiet ic/stroma-ECM interactions. 134 References Aizawa, S. and Tavassol i , M. (1988). Detection of membrane lectins on the surface of hemopoiet ic progenitor cells and their changing pattern during differentiation. Exp. Hematol . 16, 325-329. Akahane, K., Hosoi, T., Urabe, A., Kawakami, M., and Takaku, F. (1987). Effects of recombinant human tumor necrosis factor (rhTNF) on normal human and mouse hemopoiet ic progenitor cells. Int. J . Cell Cloning. 5, 16-26. Al len, J.D. and Adams, J .M. (1993). Enforced expression of Hlx homeobox gene prompts myeloid cell maturation and altered adherence properties of T cells. Blood 81, 3242-3251 . Anderson, D.M., Lyman, S.D., Baird, A., Wignal l , J.M., Eisenman, J . , Rauch, C , March, C.J., Boswell , H.S., Gimpel, S.D., Cosman, D., and Wil l iams, D.E. (1990). Molecular cloning of mast cell growth factor, a hematopoiet in that is active in both membrane bound and soluble forms. Cell 63, 235-243. Arch, R., Wirth, K., Hofmann, M., Ponta, H., Matzku, S., Herrl ich, P., and Zoller, M. (1992). Participation in normal immune responses of a metastasis- inducing splice variant of CD44. Science 257, 682-685. Arkin, S., Naprstek, B., Guarini, L., Ferrone, S., and Lipton, J .M. (1991). Expression of intercellular adhesion molecule-1 (CD54) on hematopoietic progenitors. Blood 77, 948-953. Aruffo, A., Stamenkovic, I., Melnick, M., Underhil l , C.B., and Seed, B. (1990). CD44 is the principal cell surface receptor for hyaluronate. Cell 61, 1303-1313. Asch , A.S. , Tepler, J . , Silbiger, S., and Nachman, R.L. (1990). Cellular at tachment to thrombospondin: cooperat ive interactions between receptor systems. J . Biol. Chem. 266, 1740-1745. Avanzi , G.C., Lista, P., Giovinazzo, B., Miniero, R., Saglio, G., Benetton, G., Coda, R., Cattorett i , G., and Pegoraro, L. (1988). Selective growth response to IL-3 of a human leukaemic cell line with megakaryoblast ic features. Br. J . Haematol . 69, 359-366. Avraham, H., Scadden, D.T., Chi, S., Broudy, V.C., Zsebo, K.M., and Groopman, J.E. (1992). Interaction of human bone marrow fibroblast with megakaryocytes: role of the c-kit l igand. Blood 80, 1679-1684. Ayroldi , E., Cannari le, L., Migliorati, G., Bartoli, A., Nicoletti, I., and Riccardi, C. (1995). CD44 (Pgp-1) inhibits CD3 and dexamethasone- induced apoptosis. Blood 86, 2672-2678. 135 Bain, G., Maandag, E.C., Izon, D.J., Amsen, D., Kruisbeek, A .M. , Weintraub, B.C., Krop, I., Schlissel, M.S., Feeney, A.J . , and van Roon, M. (1994). E2A proteins are required for proper B cell development and initiation of immunoglobul in gene rearrangements. Cell 79, 885-892. Bartley, T.D., Bogenberger, J . , Hunt, P., Li, Y.S. , Lu, H.S., Mart in, F., Chang, M.S., Samal , B., Nichol, J.L., Swift, S., Johnson, M.J., Hsu, R.Y., Parker, V.P., Suggs, S., Skrine, J.D., Merewether, L.A., Clogston, C , Hsu, E., Hokom, M.M., Hornkohl , A., Choi , E., Pangel inan, M., Sun, Y., Mar, V., McNinch, J . , Simonet, L., Jacobsen, F., Xie, C , Shutter, J . , Chute, H., Basu, R., Selander, L., Troll inger, D., Sieu, L., Padil la, D., Trai l , G., Elliott, G., Izumi, R., Covey, T., Crouse, J . , Garcia, A., Xu , W., del Casti l lo, J . , Biron, J . , Cole, S., Hu, M.C.T., Pacifici, R., Ponting, I., Saris, C , W e n , D., Yung, Y.P., Lin, H., and Bosselman, R.A. (1994). Identification and cloning of a megakaryocyte growth and development factor that is a ligand for the cytokine receptor Mpl. Cell 77, 1117-1124. Bartolazzi, A., Peach, R., Aruffo, A., and Stamenkovic, I. (1994). Interaction between CD44 and hyaluronate is directly implicated in the regulation of tumor development. J . Exp. Med. 180, 53-66. Baumhueter , S., Singer, M.S., Henzel, W., Hemmer ich, S., Renz, M., Rosen, S.D., and Lasky, L.A. (1993). Binding of L-selectin to the vascular sialomucin CD34. Science 262, 436-438. Bazil, V. and Horejsi, V. (1992). Shedding of the CD44 adhesion molecule f rom leukocytes induced by ant i-CD44 monoclonal antibody simulating the effect of a natural receptor l igand. J . Immunol. 149, 747-753. Belitsos, P.C., Hildreth, J.E.K., and August, J.T. (1990). Homotypic cell aggregat ion induced by ant i -CD44(Pgp-1) monoclonal antibodies and related to CD44(Pgp-1) expression. J . Immunol . 144, 1661-1670. Bennett, K.L., Jackson, D.G., Simon, J.C., Tanczos, E., Peach, R., Modrel l , B., Stamenkovic, I., P lowman, G., and Aruffo, A. (1995a). CD44 isoforms containing exon v3 are responsible for the presentation of heparin-binding growth factor. J . Cell Biol. 128, 687-698. Bennett, K.L., Modrel l , B., Greenfield, B., Bartolazzi, A., Stamenkovic, I., Peach, R., Jackson, D.G., Spring, F., and Aruffo, A. (1995b). Regulation of CD44 binding to hyaluronan by glycosylation of variably spliced exons. J . Cell Biol. 131, 1623-1633. Berardi, A.C. , W a n g , A., Levine, J.D., Lopez, P., and Scadden, D.T. (1995). Functional isolation and characterization of human hematopoietic s tem cells. Science 267, 104-108. Bernstein, S.E. (1970). Tissue transplantation as an analytic and therapeutic tool in hereditary anemias. A m . J . Surg. 119, 448 -451 . 136 Bevi lacqua, M P . and Nelson, R.M. (1993). Selectins. J . Clin. Invest. 91, 379-387. Bhat ia, R., Wayner, E.A., McGlave, P.B., and Verfaill ie, C M . (1994). Interferon-oc restores normal adhesion of chronic myelogenous leukemia hematopoiet ic progenitors to bone marrow stroma by correcting impaired (3-| integrin receptor funct ion. J . Clin. Invest. 94, 3 8 4 - 3 9 1 . Bhat ia, R., McGlave, P.B., Dewald, G.W., Blazar, B.R., and Verfaill ie, C M . (1995). Abnormal function of the bone marrow microenvironment in chronic myelogenous leukemia: role of malignant stromal macrophages. Blood 85, 3636-3645. Blazsek, I., Misset, J.L., Benavides, M., Comisso, M., Ribaud, P., and Mathe, G. (1990). Hematon; a multicellular functional unit in normal human bone marrow: structural organizat ion, hemopoietic activity, and its relationship to myelodysplasia and myeloid leukemias. Exp. Hematol . 18, 259-265. Bodine, D.M., Crosier, P.S., and Clark, S . C (1991). Effects of hematopoiet ic growth factors on the survival of primitive stem cells in liquid suspension culture. Blood 78, 914-920. Bodine, D.M., Seidel, N.E., Zsebo, K.M., and Orlic, D. (1993). In vivo administration of s tem cell factor to mice increases the absolute number of pluripotent hematopoiet ic s tem cells. Blood 82, 445-455. Bodine, D.M., Seidel, N.E., Gale, M.S., Nienhuis, A.W., and Orlic, D. (1994). Efficient retrovirus transduct ion of mouse pluripotent hematopoietic stem cells mobi l ized into the peripheral blood by treatment with granulocyte colony-stimulating factor and stem cell factor. Blood 84, 1482-1491. Bornstein, P. (1992). Thrombospondins: structure and regulation of expression. FASEB J . 6, 3290-3299. Brady-Kalnay, S.M. and Tonks, N.K. (1995). Protein tyrosine phosphatases as adhesion receptors. Curr. Opin. Cell Biol. 7, 650-657. Briddell, R.A., Hartley, C.A., Smith, K.A., and McNiece, I.K. (1993). Recombinant rat s tem cell factor synergizes with recombinant human granulocyte colony-st imulat ing factor in vivo in mice to mobilize peripheral blood progenitor cells that have enhanced repopulating potential. Blood 82, 1720-1723. Brown, T.A., Bouchard, T., St John, T., Wayner, E., and Carter, W . G . (1991). Human kerat inocytes express a new CD44 core protein (CD44E) as a heparan-sulfate intrinsic membrane proteoglycan with additional exons. J . Cell Biol. 113, 2 0 7 - 2 2 1 . Broxmeyer, H.E., Sherry, B., Lu, L , Cooper, S., Oh, K.O., Tekamp-Olson, P., Kwon, B.S., and Cerami , A. (1990). Enhancing and suppressing effects of recombinant 137 murine macrophage inf lammatory proteins on colony formation in vitro by bone marrow myeloid progenitor cells. Blood 76, 1110-1116. Bruynzeel , I., Koopman, G., Liesbeth, M.H., van der Raaij, L.M.H., Pals, S T . , and Wi l lemze, R. (1993). CD44 antibody stimulates adhesion of peripheral blood T cells to kerat inocytes through the leukocyte funct ion-associated antigen-1/intercellular adhesion molecule-1 pathway. J . Invest. Dermatol . 100, 424-428. Burridge, K., Fath, K., Kelly, T., Nuckolls, G., and Turner, C. (1988). Focal adhesions: t ransmembrane junctions between the extracellular matrix and cytoskeleton. Annu. Rev. Cell Biol. 4, 487-525. Butterf ield, K., Fathman, C.G., and Budd, R.C. (1989). A subset of memory C D 4 + helper T lymphocytes identified by expression of Pgp-1 . J . Exp. Med. 169, 1461-1466. Butzow, R., Fukushima, D., Twardzik, D.R., and Ruoslahti, E. (1993). A 60-kD protein mediates the binding of transforming growth factor-p to cell surface and extracellular matrix proteoglycans. J . Cell Biol. 122, 721-727. Camp, R.L., Kraus, T.A., Birkeland, M.L., and Pure, E. (1991a). High levels of CD44 expression distinguish virgin from antigen-primed B cells. J . Exp. Med. 173, 763-766. Camp, R.L., Kraus, T.A., and Pure, E. (1991b). Variations in the cytoskeletal interaction and posttranslational modification of the CD44 homing receptor in macrophages. J . Cell Biol. 115, 1283-1292. Camp, R.L., Scheynius, A., Johansson, C , and Pure, E. (1993). CD44 is necessary for opt imal contact allergic responses but is not required for normal leukocyte extravasat ion. J . Exp. Med. 178, 497-507. Campanero, M.R., Pulido, R., Alonso, J.L., Pivel, J.P., Pimentel-Muinos, F.X., Fresno, M., and Sanchez-Madr id, F. (1991). Down-regulat ion by tumor necrosis factor -a of neutrophil cell surface expression of the sialophorin CD43 and the hyaluronate receptor CD44 through a proteolytic mechanism. Eur. J . Immunol . 21, 3045-3048. Carter, W . G . and Wayner, E.A. (1988). Characterization of the class III col lagen receptor, a phosphorylated, t ransmembrane glycoprotein expressed in nucleated human cells. J . Biol. Chem. 263, 4193-4201 . Cashman, J . , Eaves, A.C. , and Eaves, C.J. (1985). Regulated proliferation of primitive hematopoiet ic progenitor cells in long-term human marrow cultures. Blood 66, 1002-1005. Cashman, J.D., Eaves, A.C., Raines, E.W., Ross, R., and Eaves, C.J. (1990). Mechanisms that regulate the cell cycle status of very primitive hematopoiet ic cells in long-term human marrow cultures. I. Stimulatory role of a variety of mesenchymal cell activators and inhibitory role of TGF-p. Blood 75, 96 -101 . 138 Cashman, J.D., Eaves, A.C., and Eaves, C.J. (1992). Granulocyte-macrophage colony-st imulat ing factor modulat ion of the inhibitory effect of t ransforming growth factor-p on normal and leukemic human hematopoietic progenitor cells. Leukemia 6, 886-892. Chabot, B., Stephenson, D.A., Chapman, V.M. , Besmer, P., and Bernstein, A. (1988). The proto-oncogene c-kit encoding a t ransmembrane tyrosine kinase receptor maps to the mouse W locus. Nature 335, 88-89. Cheng, J . , Baumhueter, S., Cacalano, G., Carver-Moore, K., Thibodeaux, H., Thomas, R., Broxmeyer, H.E., Cooper, S., Hague, N., Moore, M., and Lasky, L.A. (1996). Hematopoiet ic defects in mice lacking the sialomucin CD34. Blood 87, 479-490. Chuang, T.H. , Xu , X., Kaart inen, V., Heisterkamp, N., Groffen, J . , and Bokoch, G.M. (1995). Abr and Bcr are multifunctional regulators of the Rho GTP-binding protein family. Proc. Natl. Acad. Sci. U. S. A. 92, 10282-10286. Clark, B.R., Gallagher, J.T., and Dexter, T .M. (1992). Cell adhesion in the stromal regulations of haemopoiesis. Clin. Haematol . 5, 619-652. Cline, M.J. (1994). The molecular basis of leukemia. N. Engl. J . Med. 330, 328-336. Cohen, A .M. , Zsebo, K.M., Inoue, H., Hines, D., Boone, T.C., Chazin, V.R., Tsai , L., Ritch, T., and Souza, L.M. (1987). In vivo stimulation of granulopoiesis by recombinant human granulocyte colony-stimulating factor. Proc. Natl. Acad. Sci. USA 84, 2484-2488. Colombatt i , A., Hughes, E.N., Taylor, B.A., and August, J.T. (1982). Gene for a major cell surface glycoprotein of mouse macrophages and other phagocytic cells is on chromosome 2. Proc. Natl. Acad. Sci. U. S. A. 79, 1926-1929. Coombe, D.R., Watt, S.M., and Parish, C.R. (1994). Mac-1 (CD11b/CD18) and CD45 mediate the adhesion of hematopoietic progenitor cells to stromal cell e lements via recognit ion of stromal heparan sulfate. Blood 84, 739-752. Cooper, D.L., Dougherty, G., H a m , H.-J., Jackson, S., Baptist, E.W., Byers, J . , Datta, A., Phill ips, G., and Isola, N.R. (1992). The complex CD44 transcriptional unit: I. Alternative splicing of three internal exons generates the epithelial form of CD44. B iochem. Biophys. Res. Commun. 182, 569-578. Cooper, D.L. and Dougherty, G.J. (1995). To metastasize or not? Selection of CD44 splice sites. Nature Medicine 1, 635-637. Copeland, N.G., Gilbert, D.J., Cho, B.C., Donovan, P.J., Jenkins, N.A., Cosman, D., Anderson, D., Lyman, S.D., and Wil l iams, D.E. (1990). Mast cell growth factor maps near the steel locus on mouse chromosome 10 and is deleted in a number of steel alleles. Cell 63, 175-183. 139 Coulombel , L , Eaves, A.C. , and Eaves, C.J. (1983). Enzymatic treatment of long-term human marrow cultures reveals the preferential location of primitive hemopoiet ic progenitors in the adherent layer. Blood 62, 291-297. Coulombel , L , Vuillet, M.H., Leroy, C , and Tchernia, G. (1988). Lineage- and stage-specific adhesion of human hematopoietic progenitor cells to extracellular matr ices f rom marrow f ibroblasts. Blood 71, 329-334. Craig, S.W. and Johnson, R.P. (1996). Assembly of focal adhesions: progress, paradigms, and portents. Curr. Opin. Cell Biol. 8, 74-85. Craig, W., Kay, R., Cutler, R.L., and Lansdorp, P.M. (1993). Expression of Thy-1 on human hematopoiet ic progenitor cells. J . Exp. Med. 777, 1331-1342. Croisil le, L., Auffray, I., Katz, A., Izac, B., Vainchenker, W., and Coulombel , L. (1994). Hydrocort isone differentially affects the ability of murine stromal cells and human marrow-der ived adherent cells to promote the differentiation of CD34 + 7CD38" long-term culture-initiating cells. Blood 84, 4116-4124. Culty, M., Miyake, K., Kincade, P.W., Silorski, E., Butcher, E.C., and Underhil l , C. (1990). The hyaluronate receptor is a member of the CD44 (H-CAM) family of cell surface glycoproteins. J . Cell Biol. 111, 2765-2774. Daley, G.Q., Van Etten, R.A., and Balt imore, D. (1990). Induction of chronic myelogenous leukemia in mice by the P2iobcr/abl gene of the Philadelphia chromosome. Science 247, 824-830. Dastych, J . and Metcalfe, D.D. (1994). Stem cell factor induces mast cell adhesion to f ibronectin. J . Immunol . 152, 213-219. de Belder, A . N . and Wik, K.O. (1975). Preparation and properties of f luorescein-labelled hyaluronate. Carbohydr. Res. 44, 251-257. de Sauvage, F.J., Hass, P.E., Spencer, S.D., Malloy, B.E., Gurney, A.L., Spencer, S.A., Darbonne, W.C. , Henzel, W.J. , Wong , S.C., Kuang, W.J . , Oles, K.J., Hultgren, B., Solberg Jr, L.A., Goeddel , D.V., and Eaton, D.L. (1994). Stimulation of megakaryocytopoiesis and thrombopoiesis by the c-MpI l igand. Nature 369, 533-538. Delfino, D.V., Patrene, K.D., DeLeo, A.B. , DeLeo, R., Herberman, R.B., and Boggs, S.S. (1994). Role of CD44 in the development of natural killer cells f rom precursors in long-term cultures of mouse bone marrow. J . Immunol. 152, 5171-5179. Denning, S.M., Le, P.T., Singer, K.H., and Haynes, B.F. (1990). Ant ibodies against the CD44 p80, lymphocyte homing receptor molecule augment human peripheral blood T cell act ivation. J . Immunol. 144, 7-15. Dercksen, M.W., Gerri tsen, W.R., Rodenhuis, S., Dirkson, M.K.A., Slaper-Cortenbach, I .CM. , Schaasberg, W.P. , Pinedo, H.M., von dem Borne, A.E.G.K., and van der 140 Schoot, C.E. (1995). Expression of adhesion molecules on C D 3 4 + cells: C D 3 4 + L-select in + cells predict a rapid platelet recovery after peripheral blood stem cell t ransplant ion. Blood 85, 3313-3319. Devaraj , P.E., Foroni, L , Sekhar, M., Butler, T., Wright, F., Mehta, A., Samson, D., Prentice, H.G., Hoffbrand, A.V., and Seeker-Walker, L.M. (1994). E2A/HLF fusion cDNAs and the use of RT-PCR for the detection of minimal residual disease in t(17;19)(q22;p13) acute lymphoblastic leukemia. Leukemia 8, 1131-1138. Dexter, T .M. , Al len, T.D., and Lajtha, L.G. (1977). Condit ions controll ing the proliferation of hemopoiet ic stem cells in vitro. J . Cell Physiol. 91, 335-344. Dexter, T .M. (1982). Stromal cell associated haemopoiesis. J . Cell Physiol. [Suppl]. 7, 87-94. Diekmann, D., Brill, S., Garrett, M.D., Totty, N., Hsuan, J . , Monfr ies, C , Hall, C , Lim, L., and Hall, A. (1991). Bcr encodes a GTPase-act ivat ing protein for p21rac. Nature 351, 400-402. Dougherty, G.J., Lansdorp, P.M., Cooper, D.L., and Humphr ies, R.K. (1991). Molecular cloning of CD44R1 and CD44R2, two novel isoforms of the human CD44 lymphocyte "homing" receptor expressed by hemopoietic cells. J . Exp. Med. 174, 1-5. Dougherty, G.J., Cooper, D.L., Memory, J.F., and Chiu, R.K. (1994). Ligand binding specificity of alternatively-spliced CD44 isoforms. Recognit ion of hyaluronan by C D 4 4 R 1 . J . Biol. Chem. 269, 9074-9078. Dougherty, G.J., Dougherty, S T . , Dirks, J.F., Chui , R.K., Peters, C.E., and Droll, A. (1995). Regulat ion of the functional activity and ligand binding specificity of the adhesion protein CD44. Trends Glycosci. Glycotechnol. 7, 45-56. Droll, A., Dougherty, S T . , Chiu, R.K., Dirks, J.F., McBride, W.H. , Cooper, D.L., and Dougherty, G.J. (1995). Adhesive interactions between alternatively spliced CD44 isoforms. J . Biol. Chem. 270, 11567-11573. Eaves, A .C. and Eaves, C.J. (1979). Abnormali t ies in the erythroid progenitor compartments in patients with chronic myelogenous leukemia (CML). Exp. Hematol . 7, 65-75. Eaves, A . C , Cashman, J.D., Gaboury, L.A., Kalousek, D.K., and Eaves, C.J. (1986). Unregulated proliferation of primitive chronic myeloid leukemia progenitors in the presence of normal marrow adherent cells. Proc. Natl. Acad. Sci. U. S. A. 83, 5306-5310. Eaves, C.J. and Eaves, A.C. (1987). Cell culture studies in CML. In Baill ieres Clinical Hematology. Anonymouspp. 931 -961 . 141 Eaves, C.J., Cashman, J.D., Kay, R.J., Dougherty, G.J., Otsuka, T., Gaboury, L.A., Hogge, D.E., Lansdorp, P.M., Eaves, A.C. , and Humphries, R.K. (1991a). Mechanisms that regulate the cell cycle status of very primitive hematopoietic cells in long-term human marrow cultures. II . Analysis of positive and negative regulators produced by stromal cells within the adherent layer. Blood 78, 110-117. Eaves, C.J., Cashman, J.D., Sutherland, H.J., Otsuka, T., Humphries, R.K., Hogge, D.E., Lansdorp, P.M., and Eaves, A.C. (1991b). Molecular analysis of primitive hematopoiet ic cell proliferation control mechanisms. Ann. N. Y. Acad. Sci. 628, 298-306. Eaves, C.J., Suther land, H.J., Udomsakdi , C , Lansdorp, P.M., Szi lvassy, S.J., Fraser, C C , Humphr ies, R.K., Barnett, M.J., Phillips, G.L., and Eaves, A.C. (1992). The human hematopoiet ic stem cell in vitro and in vivo. Blood Cells 18, 301-307. Eaves, C , Udomsakdi , C , Cashman, J . , Barnett, M., and Eaves, A. (1993). The biology of normal and neoplastic stem cells in CML. Leuk. Lymphoma. 11, 245-253. Eaves, C. and Eaves, A. (1994). Differential manipulation of normal and chronic myeloid leukemia s tem cell proliferation in vitro. Blood Cells 20, 83-95. Eaves, C.J., Cashman, J.D., and Eaves, A.C. (1993). Control mechanisms for primitive hemopoiet ic cells. In M. Guigon, F.M. Lemoine, N. Dainiak, A. Schechter, and A. Najman, eds. (Montrouge, France: Colloques INSERM/John Libbey Eurotext Ltd.), pp. 123-131 . Erickson, H.P. (1993). Tenascin-C, tenascin-R and tenascin-X: a family of talented proteins in search of functions. Curr. Opin. Cell Biol. 5, 869-876. Faassen, A.E. , Schrager, J.A., Klein, D.J., Oegema, T.R., Couchman, J.R., and McCarthy, J.B. (1992). A cell surface chondroit in sulfate proteoglycan, immunological ly related to CD44, is involved in type I col lagen-mediated melanoma cell motility and invasion. J . Cell Biol. 116, 521-531 . Faassen, A.E. , Mooradian, D.L., Tranquil lo, R.T., Dickinson, R.B., Letourneau, P . C , O e g e m a , T.R., and McCarthy, J.B. (1993). Cell surface CD44-related chondroit in sulfate proteoglycan is required for transforming growth factor-p-st imulated mouse melanoma cell motility and invasive behavior on type 1 col lagen. J . Cell Sci. 105, 5 0 1 -5 1 1 . Fenderson, B.A., Stamenkovic, I., and Aruffo, A. (1993). Localization of hyaluronan in mouse embryos during implantation, gastrulation and organogenesis. Differentiation 54, 85-98. Finn, L., Dougherty, G., Finley, G., Meisler, A., Becich, M., and Cooper, D.L. (1994). Alternative splicing of CD44 pre-mRNA in human colorectal tumors. B iochem. Biophys. Res. C o m m u n . 200, 1015-1022. 142 Flanagan, J.G., Chan, D.C., and Leder, P. (1991). Transmembrane form of the kit l igand growth factor is determined by alternative splicing and is missing in the S l d mutant. Cell 64, 1025-1035. Francke, U., Foellmer, B.E., and Haynes, B.F. (1983). Chromosome mapping of human cell surface molecules: monoclonal ant i -human lymphocyte antibodies 4F2, A3D8, and A 1 G 3 define antigens controlled by different regions of chromosome 1 1 . Somat. Cell Mol. Genet. 9, 333-344. Fraser, C.C., Szilvassy, S.J., Eaves, C.J., and Humphries, R.K. (1992). Proliferation of totipotent hematopoiet ic stem cells in vitro with retention of long-term competi t ive in vivo reconstituting ability. Proc. Natl. Acad. Sci. U. S. A. 89, 1968-1972. Frenette, P.S., Mayadas, T.N. , Rayburn, H., Hynes, R.O., and Wagner , D.D. (1996). Susceptibi l i ty to infection and altered hematopoiesis in mice deficient in both P- and E-selectins. Cell 84, 563-574. Funk, P.E., Kincade, P.W., and Witte, P.L. (1994). Native associat ion of early hematopoiet ic cells and stromal cells isolated in bone marrow cell aggregates. Blood 83, 361-369. Funk, P.E., Stephan, R.P., and Witte, P.L. (1995). Vascular cell adhesion molecule 1-positive reticular cells express interleukin-7 and stem cell factor in the bone marrow. Blood 86, 2661-2671 . Gabbianel l i , M., Sargiacomo, M., Pelosi, E., Testa, U., Isacchi, G., and Peschle, C. (1990). "Pure" human hematopoiet ic progenitors: Permissive action of basic f ibroblast growth factor. Science 249, 1561 -1564. Galandrini , R., Albi, N., Tripodi, G., Zarcone, D., Terenzi , A., Moretta, A., Grossi , C.E., and Velardi , A. (1993). Antibodies to CD44 trigger effector funct ions of human T cell c lones. J . Immunol . 150, 4225-4235. Gallagher, J.T., Spooncer, E., and Dexter, T .M. (1983). Role of the cellular matrix in haemopoiesis I. Synthesis of glycosaminoglycans by mouse bone marrow cell cultures. J . Cell Sci. 63, 155-171. Gallatin, W.M. , Wayner, E.A., Hoffman, P.A., St.John, T., Butcher, E.C., and Carter, W . G . (1989). Structural homology between lymphocyte receptors for high endothel ium and class III extracellular matrix receptor. Proc. Natl. Acad. Sci. U. S. A. 86, 4654-4658. Georgopoulos, K., Bigby, M., Wang, J.H. , Molnar, A., W u , P., Winandy, S., and Sharpe, A. (1994). The ikaros gene is required for the development of all lymphoid l ineages. Cell 79, 143-156. Geppert , T.D. and Lipsky, P.E. (1991). Associat ion of various T cell-surface molecules with the cytoskeleton. J . Immunol. 146, 3298-3305. 143 Gishizky, M.L., Cortez, D., and Pendergast, A .M. (1995). Mutant forms of growth factor-binding protein-2 reverse BCR-ABL- induced transformation. Proc. Natl. Acad . Sci. U. S. A. 92, 10889-10893. Godfrey, D.I., Kennedy, J . , Suda, T., and Zlotnik, A. (1993). A developmental pathway involving four phenotypically and functionally distinct subsets of CD3-CD4-CD8- triple negative adult mouse thymocytes defined by CD44 and CD25 expression. J . Immunol . 750, 4244-4252. Goga, A., McLaughl in, J . , Afar, D.E.H., Saffran, D.C., and Witte, O.N. (1995). Alternative signals to RAS for hematopoietic transformation by the BCR-ABL oncogene. Cell 82, 981-988. Goldstein, L.A., Zhou, D.F.H., Picker, L.J., Minty, C.N., Bargatze, R.F., Ding, J.F., and Butcher, E.C. (1989). A human lymphocyte homing receptor, the Hermes ant igen, is related to carti lage proteoglycan core and link proteins. Cell 56, 1063-1072. Goldstein, L.A. and Butcher, E.C. (1990). Identification of mRNA that encodes an alternative form of H-CAM (CD44) in lymphoid and nonlymphoid t issues. Immunogenet ics 32, 389-397. Goodfel low, P.N., Banting, G., Wiles, M.V., Tunnaclif fe, A., Parkar, M., So lomon, E., Dalchau, R., and Fabre, J.W. (1982). The gene, MIC4, which controls expression of the antigen defined by monoclonal antibody F10.44.2, is on human chromosome 1 1 . Eur. J . Immunol . 12, 659-663. Gordon, M.Y., Dowding, C.R., Riley, G.P., Goldman, J.M., and Greaves, M.F. (1987). Altered adhesive interactions with marrow stroma of haematopoiet ic progenitor cells in chronic myeloid leukaemia. Nature 328, 342-344. Gordon, M.Y., Riley, G.P., and Clarke, D. (1988). Heparan sulfate is necessary for adhesive interactions between human early hemopoiet ic progenitor cells and the extracellular matrix of the marrow microenvironment. Leukemia 2, 804-809. Gordon, M.Y., Clarke, D., Atkinson, J . , and Greaves, M.F. (1990). Hemopoiet ic progenitor cell binding to the stromal microenvironment in vitro. Exp. Hematol . 18, 837-842. Gordon, M.Y., Atkinson, J . , Clarke, D., Dowding, C.R., Goldman, J .M. , Grimsley, P.G., Siczkowski , M., and Greaves, M.F. (1991). Deficiency of a phosphatidyl inositol-anchored cell adhesion molecule influences haemopoiet ic progenitor binding to marrow st roma in chronic myeloid leukemia. Leukemia 5, 693-698. Gordon, M.Y. (1994). Plastic-adherent cells in human bone marrow generate long-term hematopoiesis in vitro. Leukemia 8, 865-870. 144 Gotoh, A., Miyazawa, K., Ohyashiki , K., Tauchi , T., Boswell , H.S., Broxmeyer, H.E., and Toyama, K. (1995). Tyrosine phosphorylat ion and activation of focal adhesion kinase (p125 F A K ) by BCR-ABL oncoprotein. Exp. Hematol . 23, 1153-1159. Greenberger, J.S. (1992). Is the marrow stroma of A M L patients a " leukemic" stroma? Exp. Hematol . 20, 1041-1042. Grzegorzewski , K.J., Komschl ies, K.L., Jacobsen, S.E.W., Ruscetti , F.W., Keller, J.R., and Wiltrout, R.H. (1995). Mobil ization of long-term reconstituting hematopoiet ic stem cells in mice by recombinant human interleukin 7. J . Exp. Med. 181, 369-374. Guan, J.L. and Hynes, R.O. (1990). Lymphoid cells recognize an alternatively spliced segment of f ibronectin via the integrin receptor a 4 p Y Cell 60, 5 3 - 6 1 . Guan, J.L. and Shal loway, D. (1992). Regulation of focal adhesion-associated protein tyrosine kinase by both cellular adhesion and oncogenic t ransformat ion. Nature 358, 690-692. Gumbiner, B.M. (1996). Cell adhesion: the molecular basis of t issue architecture and morphogenesis. Cell 84, 345-357. Gunji , Y., Nakamure, M., Hagiwara, T., Hayakawa, K., Matsushita, H., Osawa, H., Nakayoshi , K., Nakauchi , H., Yanagisawa, M., Muira, Y., and Suda, T. (1992). Expression and function of adhesion molecules on human hematopoiet ic s tem cells: C D 3 4 + L F A - 1 ' cells are more primitive than C D 3 4 + L F A - 1 + cells. Blood 80, 429-436. Gunthert, U., Hofmann, M., Rudy, W., Reber, S., Zoller, M., Haubmann, I., Matzku, S., Wenze l , A., Ponta, H., and Herrl ich, P. (1991). A new variant of glycoprotein CD44 confers metastatic potential to rat carc inoma cells. Cell 65, 13-24. Gunthert , U. (1993). CD44: a multitude of isoforms with diverse functions. Curr. Top. Microbiol. Immunol. 184, 47-63. Haegel , H., Dierich, A., and Ceredig, R. (1994). CD44 in differentiated embryonic stem cells: surface expression and transcripts encoding multiple variants. Dev. Immunol . 3, 239-246. Haegel , H. and Ceredig, R. (1991). Transcripts encoding mouse CD44 (Pgp-1 , Ly-24) ant igen: strain variation and induction by mitogen. Eur. J . Immunol . 21, 1549-1553. Hall, C.L., W a n g , C , Lange, L.A., and Turley, E.A. (1994). Hyaluronan and the hyaluronan receptor R H A M M promote focal adhesion turnover and transient kinase activity. J . Cell Biol. 126, 575-588. Hall, C.L., Yang, B., Yang, X., Zhang, S., Turley, M., Samuel , S., Lange, L.A., W a n g , C , Curpen, G.D., Savani , R.C., Greenberg, A .H. , and Turley, E.A. (1995). Overexpression of the hyaluronan receptor RHAMM is t ransforming and is also required for H-ras transformation. Cell 82, 19-28. 145 Hammarskjo ld, M.L., Wang , S.C., and Klein, G. (1986). High-level expression of the Epstein-Barr virus EBNA1 protein in CV1 cells and human lymphoid cells using a SV40 late replacement vector. Gene 43, 41-50. Hanks, S.K., Calalb, M.B., Harper, M.C., and Patel, S.K. (1992). Focal adhesion protein-tyrosine kinase phosphorylated in response to cell at tachment to f ibronectin. Proc. Natl. Acad . Sci. U. S. A. 89, 8487-8491. Hannum, C., Culpepper, J . , Campbel l , D., McClanahan, T., Zurawski , S., Bazan, J.F., Kastelein, Ft., Hudak, S., Wagner, J . , Mattson, J . , Luh, J . , Duda, G., Mart ina, N., Peterson, D., Menon, S., Shanafelt, A., Muench, M., Kelner, G., Namikawa, R., Rennick, D., Roncarolo, M.G., Zlotnik, A., Rosnet, O., Dubreuil , P., B i rnbaum, D., and Lee, F. (1994). Ligand for FLT3/FLK2 receptor tyrosine kinase regulates growth of haematopoiet ic stem cells and is encoded by variant RNAs. Nature 368, 643-648. Hathcock, K.S., Hirano, H., Murakami, S., and Hodes, R.J. (1993). CD44 expression on activated B cells. J . Immunol. 151, 6712-6722. Haugen, P.K., McCarthy, J.B., Skubitz, A.P.N. , Furcht, L.T., and Letourneau, P.C. (1990). Recognit ion of the A-chain carboxy-terminal heparin binding region of f ibronectin involves multiple sites: two recognition sequences act independently to promote neural cell adhesion. J . Cell Biol. 111, 2733-2742. Haynes, B.F., Telen, M.J., Hale, L.P., and Denning, S.M. (1989). CD44- A molecule involved in leukocyte adherence and T-cell activation. Immunol. Today 10, 423-428. He, H., Naquet, P., Caillol, D., and Pierres, M. (1991). Thy-1 supports adhesion of mouse thymocytes to thymic epithelial cells through a C a 2 + independent mechanism. J . Exp. Med. 173, 515-518. He, Q., Lesley, J . , Hyman, R., Ishihara, K., and Kincade, P.W. (1992). Molecular isoforms of murine CD44 and evidence that the membrane proximal domain is not critical for hyaluronate recognit ion. J . Cell Biol. 119, 1711-1719. Healy, L., May, G., Gale, K., Grosveld, F., Greaves, M., and Enver, T. (1995). The stem cell ant igen CD34 functions as a regulator of hemopoiet ic cell adhesion. Proc. Natl. Acad . Sci. U. S. A. 92, 12240-12244. Heider, K.-H., Hofmann, M., Hors, E., van den Berg, F., Ponta, H., Herrl ich, P., and Pals, S T . (1993). A human homologue of the rat metastasis-associated variant of CD44 is expressed in colorectal carcinomas and adenomatous polyps. J . Cell Biol. 120, 227-233. Heisterkamp, N., Jenster, G., ten Hoeve, J . , Zovich, D., Pattengale, P.K., and Groffen, J . (1990). Acute leukemia in bcr/abl transgenic mice. Nature 344, 251-253. 146 Herrl ich, P., Rudy, W., Hofmann, M., Arch, R., Zoller, M., Zawadzki , V., Tolg, C , Hekele, A., Koopman, G., Pals, S., Heider, K.-H., S leeman, J . , and Ponta, H. (1993). CD44 and splice variants of CD44 in normal differentiation and tumor progression. In Cell Adhesion Molecules. M.E. Hemler and E. Mihich, eds. (New York: Plenum Press), pp. 265-288. Hofmann, M., Rudy, W., Zoller, M., Tolg, C , Ponta, H., Herrl ich, P., and Gunthert, U. (1991). CD44 splice variants confer metastatic behavior in rats: homologous sequences are expressed in human tumor cell lines. Cancer Res. 51, 5292-5297. Hofmann, M., Rudy, W., Gunthert, U., Zimmer, S.G., Zawadzki , V., Zoller, M., Lichtner, R.B., Herrl ich, P., and Ponta, H. (1993). A link between ras and metastatic behavior of tumor cells: ras induces CD44 promotor activity and leads to low-level expression of metastatic-specif ic variants of CD44 in CREF cells. Cancer Res. 53, 1516-1521. Hogge, D.E. (1994). Cytogenetics and oncogenes in leukemia. Curr. Opin. Oncol . 6, 3-13. Hook, M. (1984). Cell-surface glycosaminoglycans. Annu. Rev. Biochem. 53, 847-869 Horst, E., Meijer, C.J., Radaskiewicz, T., van Dongen, J.J., Pieters, R., Figdor, C.G., Hoof tman, A., and Pals, S T . (1990). Expression of a human homing receptor (CD44) in lymphoid mal ignancies and related stages of lymphoid development. Leukemia 4, 383-389. Hotchin, N.A. and Hall, A. (1995). The assembly of integrin adhesion complexes requires both extracellular matrix and intracellular rho/rac GTPases. J . Cell Biol. 131, 1857-1865. Huang, E., Nocka, K., Beier, D.R., Chu, T.Y., Buck, J . , Lahm, H.W., Wellner, D., Leder, P., and Besmer, P. (1990). The hematopoietic growth factor KL is encoded by the SI locus and is the l igand of the c-kit receptor, the gene product of the W locus. Cell 63, 225-233. Huang, H. and Auerbach, R. (1993). Identification and characterization of hematopoiet ic s tem cells f rom the yolk sac of the early mouse embryo. Proc. Natl. Acad. Sci. USA 90, 10110-10114. Huet, S., Groux, H., Cail lou, B., Valentin, H., Prieur, A .M. , and Bernard, A. (1989). CD44 contributes to T cell activation. J . Immunol. 143, 7 9 8 - 8 0 1 . Hughes, E.N., Colombatt i , A., and August, J T . (1983). Murine cell surface glycoproteins. J . Biol. Chem. 258, 1014-1021. Hughes, E.N. and August, J T . (1981). Characterization of p lasma membrane proteins identified by monoclonal antibodies. J . Biol. Chem. 256, 664 -671 . 147 Hurley, R.W., McCarthy, J.B., and Verfaill ie, C M . (1995). Direct adhesion to bone marrow st roma via f ibronectin receptors inhibits hematopoiet ic progenitor proliferation. J . Cl in. Invest. 96, 511-519. Hyman, R., Lesley, J . , and Schulte, R. (1991). Somatic cell mutants distinguish CD44 expression and hyaluronic acid binding. Immunogenet ics 33, 392-395. Hynes, R.O. (1992). Integrins: versatility modulat ion, and signaling in cell adhesion. Cell 69, 11-25. Idzerda, R.L., Carter, W.G. , Nottenburg, C , Wayner, E.A., Gallatin, W . M . , and St.John, T. (1989). Isolation and DNA sequence of a cDNA clone encoding a lymphocyte adhesion receptor for high endothel ium. Proc. Natl. Acad. Sci. U. S. A. 86, 4659-4663. Issaad, C. and Vainchenker, W. (1994). Growth of erythroid colonies in chronic myelogenous leukemia is independent of erythropoietin only in the presence of steel factor. Blood 84, 3447-3456. Jackson, D.G., Buckley, J . , and Bell, J. I . (1992). Multiple variants of the human lymphocyte homing receptor CD44 generated by insertions at a single site in the extracellular domain. J . Biol. Chem. 267, 4732-4739. Jackson, D.G., Bell, J. I . , Dickinson, R., T imans, J . , Shields, J . , and Whitt le, N. (1995). Proteoglycan forms of the lymphocyte homing receptor CD44 are alternatively spliced variants containing the v3 exon. J . Cell Biol. 128, 673-685. Jacobsen, K., Kravitz, J . , Kincade, P.W., and Osmond, D.G. (1996). Adhesion receptors on bone marrow stromal cells: in vivo expression of vascular cell adhesion molecule-1 by reticular cells and sinusoidal endothel ium in normal and y-irradiated mice. Blood 87, 73-82. Ja lkanen, S., Bargatze, R.F., Toyos, J . , and Butcher, E.C. (1987). lymphocyte recognit ion of high endothel ium: antibodies to distinct epitopes of an 85-95kD glycoprotein antigen differentially inhibit lymphocyte binding to lymph node, mucosal , or synovial endothelial cells. J . Cell Biol. 705, 983-990. Ja lkanen, S., Jalkanen, M., Bargatze, R., Tammi , M., and Butcher, E.C. (1988). Biochemical properties of glycoproteins involved in lymphocyte recognit ion of high endothelial venules in man. J . Immunol. 747, 1615-1623. Ja lkanen, S., Joensuu, H., and Klemi, P. (1990). Prognostic value of lymphocyte homing receptor and S phase fraction in non-hodgkin's lymphoma. Blood 75, 1549-1556. Ja lkanen, S. and Jalkanen, M. (1992). Lymphocyte CD44 binds the COOH-terminal heparin-binding domain of f ibronectin. J . Cell Biol. 116, 817-825. 148 Jamal , H.H., Cano-Gauci , D.F., Buick, R.N., and Filmus, J . (1994). Act ivated ras and src induce CD44 overexpression in rat intestinal epithelial cells. Oncogene 9, 417-423. Joneson, T., White, M.A., Wigler, M.H., and Bar-Sagi, D. (1996). Stimulation of membrane ruffling and MAP kinase activation by distinct effectors of RAS. Science 2 7 7 , 8 1 0 - 8 1 2 . Jul iano, R.L. and Haskill, S. (1993). Signal transduction from the extracellular matrix. J . Cell Biol. 120, 577-585. Kalomiris, E.L. and Bourguignon, L.Y.W. (1988). Mouse T lymphoma cells contain a t ransmembrane glycoprotein (GP85) that binds ankyrin. J . Cell Biol. 106, 319-327. Kalomiris, E.L. and Bourguignon, L.Y.W. (1989). Lymphoma protein kinase C is associated with the t ransmembrane glycoprotein, GP85, and may function in GP-85-ankyrin binding. J . Biol. Chem. 264, 8113-8119. Kansas, G.S., W o o d , G.S., and Dailey, M.O. (1989). A family of cel l-surface glycoproteins defined by a putative anti-endothelial cell receptor antibody in man. J . Immunol . 142, 3050-3057. Kansas, G.S., Muirhead, M.J., and Dailey, M.O. (1990). Expression of the CD11/CD18, leukocyte adhesion molecule 1, and CD44 adhesion molecules during normal myeloid and erythroid differentiation in humans. Blood 76, 2483-2492. Kansas, G.S. and Dailey, M.O. (1989). Expression of adhesion structures during B cell development in man. J . Immunol. 142, 3058-3062. Katayama, N., Clark, S.C., and Ogawa, M. (1993). Growth factor requirement for survival in cell cycle dormancy of primitive murine lymphohematopoiet ic progenitors. Blood 81, 610-616. Katoh, S., McCarthy, J.B., and Kincade, P.W. (1994). Characterizat ion of soluble CD44 in the circulation of mice: levels are affected by immune activity and tumor growth. J . Immunol . 153, 3440-3449. Katoh, S., Zheng, Z., Oritani, K., Shimozato, T., and Kincade, P.W. (1995). Glycosylat ion of CD44 negatively regulates its recognition of hyaluronan. J . Exp. Med. 182, 419-429. Kaufmann, M., Heider, K.-H., Sinn, H.-P., von Minckwitz, G., Ponta, H., and Herrl ich, P. (1995). CD44 variant exon epitopes in primary breast cancer and length of survival. Lancet 345, 615-619. Kaushansky, K. (1995). Thrombopoiet in: the primary regulator of platelet product ion. Blood 86, 4 1 9 - 4 3 1 . 149 Kerk, D.K., Henry, E.A., Eaves, A . C , and Eaves, C.J. (1985). Two classes of primitive pluripotent hemopoiet ic progenitor cells: Separat ion by adherence. J . Cell Physiol. 725, 127-134. Kerst, J .M. , Sanders, J.B., and Slaper-Cortenbach, I . C M . (1993). a 4 p i and a5p1 are differentiationally expressed during myelopoiesis and mediate the adherence of human CD34++ cells to f ibronectin in an activation-dependent way. Blood 81, 3 4 4 - 3 5 1 . Khosravi-Far, R., Solski, P.A., Clark, G.J., Kinch, M.S., and Der, C.J. (1995). Activation of R a c 1 , RhoA, and mitogen-activated protein kinases is required for Ras transformat ion. Mol. Cell Biol. 75, 6443-6453. Kinashi, T. and Springer, T.A. (1994). Steel factor and c-kit regulate cell-matrix adhesion. Blood 83, 1033-1038. Kincade, P.W. (1992). Cell interaction molecules and cytokines which participate in B lymphopoiesis. Baill ieres. Clin. Haematol . 5, 575-598. Ki tamura, T., Tojo, A., Kuwaki , T., Chiba, S., Miyazono, K., Urabe, A., and Takaku, F. (1989). Identification and analysis of human erythropoietin receptors on a factor-dependent cell line, T F - 1 . Blood 73, 375-380. Klein, G. (1995). The extracellular matrix of the hematopoiet ic microenvironment. Experient ia 57, 914-925. Knudson, W., Biswas, C , and Toole, B.P. (1984). Interactions between human tumor cells and fibroblasts stimulate hyaluronate synthesis. Proc. Natl. Acad. Sci. USA 6767-6 7 7 1 . Knudson, W., Biswas, C , Li, X.-Q., Nemec, R.E., and Toole, B.P. (1989). The role and regulation of tumour-associated hyaluronan. In The biology of hyaluronan. AnonymousWi ley, Chichester: John Wiley & Sons, Ciba Foundation Sympos ium 143, pp. 150-169. Knudson, W., Bartnik, E., and Knudson, C.B. (1993). Assembly of pericellular matrices by COS-7 cells transfected with CD44 lymphocyte-homing receptor genes. Proc. Natl. Acad. Sci. U. S. A. 90, 4003-4007. Kodama, H., Nose, M., Niida, S., Nishikawa, S., and Nishikawa, S.I. (1994). Involvement of the c-kit receptor in the adhesion of hematopoiet ic s tem cells to stromal cells. Exp. Hematol . 22, 979-984. Koeffler, H.P., Bill ing, R., Lusis, A.J . , Sparkes, R., and Golde, D.W. (1980). An undifferentiated variant derived f rom the human acute myelogenous leukemia cell line (KG-1). Blood 56, 265-273. 150 Koopman, G., van Kooyk, Y., de Graaff, M., Meyer, C.J.L.M., Figdor, C.G., and Pals, S T . (1990). Triggering of the CD44 antigen on T lymphocytes promotes T cell adhesion through the LFA-1 pathway. J . Immunol. 145, 3589-3593. Koopman, G., Heider, K.H., Horst, E., Adolf, G.R., van den Berg, F., Ponta, H., Herrl ich, P., and Pals, S T . (1993). Activated human lymphocytes and aggressive non-Hodgkin's lymphomas express a homologue of the rat metastasis-associated variant of CD44. J . Exp. Med. 177, 897-904. Kort lepel, K., Bendal l , L.J., and Gottl ieb, D.J. (1993). Human acute myeloid leukaemic cells express adhesion proteins and bind to bone marrow fibroblast monolayers and extracellular matri proteins. Leukemia 7, 1174-1179. Koshiishi, I., Shizari, M., and Underhil l , C.B. (1994). CD44 can mediate the adhesion of platelets to hyaluronan. Blood 84, 390-396. Kovach, N.L., Lin, N., Yednock, T , Harlan, J .M., and Broudy, V.C. (1995). Stem cell factor modulates avidity of a4(31and a5(31 integrins expressed on hematopoiet ic cell l ines. Blood 85, 159-167. Krause, D.S., Fackler, M.J., Civin, C.I., and May, W.S. (1996). CD34: structure, biology, and clinical utility. Blood 87, 1-13. Lacy, B.E. and Underhil l , C.B. (1987). The hyaluronate receptor is associated with actin f i laments. J . Cell Biol. 105, 1395-1404. Lansdorp, P.M., Suther land, H.J., and Eaves, C.J. (1990). Selective expression of CD45 isoforms on functional subpopulat ions of C D 3 4 + hemopoiet ic cells f rom human bone marrow. J . Exp. Med. 172, 363-366. 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 SCID mice. Nature 367, 645-648. Laurent, T.C. and Fraser, J.R.E. (1992). Hyaluronan. FASEB J . 6, 2397-2404. Lawler, J . and Hynes, R.O. (1987). Structural organization of the thrombospondin molecule. Sem Thromb Hemost 13, 245-254. Lesley, J . , Schulte, R., Trotter, J . , and Hyman, R. (1988). Qualitative and quantitative heterogeneity in Pgp-1 expression among murine thymocytes. Cell Immunol . 112, 40-54. Lesley, J . , Schulte, R., and Hyman, R. (1990). Binding of hyaluronic acid to lymphoid cell lines is inhibited by monoclonal antibodies against Pgp-1 . Exp. Cell Res. 187, 224-233. 151 Lesley, J . and Hyman, R. (1992). CD44 can be activated to function as an hyaluronic acid receptor in normal murine T cells. Eur. J . Immunol. 22, 2719-2723. Lesley, J . , He, Q., Miyake, K., Hamann, A., Hyman, R., and Kincade, P.W. (1992). Requirements for hyaluronic acid binding by CD44: A role for the cytoplasmic domain and activation by antibody. J . Exp. Med. 175, 257-266. Lesley, J . , Hyman, R., and Kincade, P.W. (1993a). CD44 and its interaction with extracellular matrix. Adv. Immunol. 54, 271-335. Lesley, J . , Kincade, P.W., and Hyman, R. (1993b). Ant ibody- induced activation of the hyaluronan receptor function of CD44 requires multivalent binding by ant ibody. Eur. J . Immunol . 23, 1902-1909. Lesley, J . , Howes, N., Perschl, A., and Hyman, R. (1994). Hyaluronan binding function of CD44 is transiently activated on T cells during an in vivo immune response. J . Exp. Med. 180, 383-387. Lesley, J . , Engl ish, N., Perschl, A., Gregoroff, J . , and Hyman, R. (1995). Variant cell lines selected for alterations in the function of the hyaluronan receptor CD44 show differences in glycosylat ion. J . Exp. Med. 182, 431-437. Levesque, J.P., Leavesley, D.I., Niutta, S., Vadas, M., and S immons, P.J. (1995). Cytokines increase human hemopoietic cell adhesiveness by activation of very late antigen (VLA)-4 and VLA-5 integrins. J . Exp. Med. 181, 1805-1815. Lewinsohn, D.M., Nagler, A., Ginzton, N., Greenberg, P., and Butcher, E.C. (1990). Hematopoiet ic progenitor cell expression of the H-CAM (CD44) homing-associated adhesion molecule. Blood 75, 589-595. Li, C.L. and Johnson, G.R. (1985). Stimulation of mult ipotential, erythroid and other murine haematopoiet ic progenitor cells by adherent cell lines in the absence of detectable mult i -CSF (IL-3). Nature 316, 633-636. L ichtman, M.A. (1981). The ultrastructure of the hemopoiet ic environment of the marrow: A review. Exp. Hematol . 9, 391-410. Lichtman, M.A. (1984). The relationship of stromal cells to hemopoiet ic cells in marrow. In Long-Term Bone Marrow Culture. D.G. Wright and J.S. Greenberger, eds. (New York: Alan R. Liss, Inc.), pp. 3-29. Liesveld, J.L., Winslow, J.M., Kempski , M.C., Ryan, D.H., Brennan, J.K., and Abboud, C.N. (1991). Adhesive interactions of normal and leukemic human CD34+ myeloid progenitors: role of marrow stromal, fibroblast, and cytomatrix components . Exp. Hematol . 19, 63-70. 152 Liesveld, J.L., Winslow, J.M., Frediani, K.E., Ryan, D.H., and Abboud, C.N. (1993). Expression of integrins and examination of their adhesive function in normal and leukemic hematopoiet ic cells. Blood 81, 112-121. Liesveld, J.L., Frediani, K.E., Harbol, A.W., DiPersio, J.F., and Abboud, C.N. (1994). Characterizat ion of the adherence of normal and leukemic CD34+ cells to endothelial monolayers. Leukemia 8, 2111-2117. Lindahl, U. and Hook, M. (1978). Glycosaminoglycans and their binding to biological macromolecules. Annu. Rev. Biochem. 47, 385-417. Lok, S., Kaushansky, K., Holly, R.D., Kuijper, J.L., Lofton-Day, C.E., Oort, P.J., Grant, F.J., Heipel, M.D., Burkhead, S.K., Kramer, J .M., Bell, L.A., Sprecher, C.A., B lumberg, H., Johnson, R., Prunkard, D., Ching, A.F.T., Mathewes, S.L., Bailey, M.C., Forstrom, J.W., Buddie, M.M., Osborn, S.G., Evans, S.J., Sheppard, P.O., Presnell , S.R., O'Hara, P.J., Hagen, F.S., Roth, G.J., and Foster, D.C. (1994). Cloning and expression of murine thrombopoiet in cDNA and stimulation of platelet production in vivo. Nature 369, 565-568. Lokeshwar, V.B. , Fregien, N., and Bourguignon, L.Y.W. (1994). Ankyrin-binding domain of CD44 (GP85) is required for the expression of hyaluronic acid-mediated adhesion funct ion. J . Cell Biol. 126, 1099-1109. Lokeshwar, V.B. and Bourguignon, L.Y.W. (1992). The lymphoma t ransmembrane glycoprotein GP85 (CD44) is a novel guanine nucleotide-binding protein which regulates GP85 (CD44)- ankyrin interaction. J. Biol. Chem. 267, 22073-22078. Long, M.W. (1992). Blood cell cytoadhesion molecules. Exp. Hematol . 20, 2 8 8 - 3 0 1 . Long, M.W., Briddell, R., Walter, A.W., Bruno, E., and Hoffman, R. (1992). Human hematopoiet ic stem cell adherence to cytokines and matrix molecules. J . Clin. Invest. 90, 251-255. Long, M.W. and Dixit, V .M. (1990). Thrombospondin functions as a cytoadhesion molecule for human hematopoiet ic progenitor cells. Blood 75, 2311-2318. Lord, B.I., Wool ford, L.B., Wood, L.M., Czaplewski , L.G., McCourt , M., Hunter, M.G., and Edwards, R.M. (1995). Mobilization of early hematopoietic progenitor cells with BB-10010: a genetically engineered variant of human macrophage inf lammatory protein-1a. Blood 85, 3412-3415. Lozzio, B.B., Lozzio, C.B., Bamberger, E.G., and Feliu, A.S. (1981). A multipotential leukemia cell line (K-562) of human origin. Proc. Soc. Exp. Biol. Med. 166, 546-550. Lugo, T.G., Pendergast, A .M. , Muller, A.J . , and Witte, O.N.,(1990). Tyrosine kinase activity and transformation potency of bcr-abl oncogene products. Science 247, 1079-1082. 153 Lyman, S.D., James, L., Vanden Bos, T., de Vries, P., Brasel, K., Gliniak, B., Hol l ingsworth, L.T., Picha, K.S., McKenna, H.J., Splett, R.R., Fletcher, F.A., Maraskovsky, E., Farrah, T., Foxworthe, D., Wil l iams, D.E., and Beckmann, M.P. (1993) . Molecular cloning of a ligand for the flt3/flk-2 tyrosine kinase receptor: A proliferative factor for primitive hematopoietic cells. Cell 75, 1157-1167. Lynch, F., Chaudhr i , G., Al lan, J.E., Doherty, P.C., and Ceredig, R. (1987). Expression of Pgp-1 (or Ly24) by subpopulat ions of mouse thymocytes and activated peripheral T lymphocytes. Eur. J . Immunol. 17, 137-140. Lynch, F. and Ceredig, R. (1988). Ly-24 (Pgp-1) expression by thymocytes and peripheral T cells. Immunol . Today 9, 7-10. Mackay, C.R., Terpe, H.J., Stauder, R., Martson, W.L., Stark, H., and Gunthert, U. (1994) . Expression and modulation of CD44 variant isoforms in humans. J . Cell Biol. 124, 71-82. Malavasi , F., Funaro, A., Roggero, S., Horenstein, A., Calosso, L., and Mehta, K. (1994) . Human CD38: a glycoprotein in search of a funct ion. Immunology Today 15, 95-97. Marquez, C , Ttr igueros, C , Fernandez, E., and Toribio, M.L. (1995). The development of T and non-T cell l ineages f rom CD34+ human thymic precursors can be t raced by the differential expression of CD44. J . Exp. Med. 181, 475-483. Maru, Y., Wit te, O., and Shibuya, M. (1996). Deletion of ABL SH3 domain reactivates de-ol igomerized BCR-ABL for growth factor independence. FEBS Lett. 379, 244-246. Massague, J . (1992). Receptors for the TGF-b family. Cell 69, 1067-1070. Matsumura, Y., Sugiyama, M., Hayle, A.J . , Robinson, P., Smith, J.C., and Tar in, D. (1995) . Unusual retention of introns in CD44 gene transcripts in bladder cancer provides new diagnostic and clinical oncological opportunit ies. J . Pathol. 177, 11-20. Matsumura, Y. and Tarin, D. (1992). Signif icance of CD44 gene products for cancer diagnosis and disease evaluation. Lancet 340, 1053-1058. Mauch, P., Greenberger, J.S., Botnik, L., Hannon, E., and Hel lman, S. (1980). Evidence for structured variation in self-renewal capacity within long-term bone marrow cultures. Proc. Natl. Acad. Sci. U. S. A. 77, 2927-2930. Mayani , H., Guilbert, L.J., Clark, S.C., Belch, A.R., and Janowska-Wieczorek, A. (1992). Composi t ion and functional integrity of the in vitro hemopoiet ic microenvironment in acute myelogenous leukemia: Effect of macrophage colony-stimulating factor. Exp. Hematol . 20, 1077-1084. 154 McCul loch, E.A., Siminovitch, L., Till, J.E., Russell, E.S., and Bernstein, S.E. (1965). The cellular basis of the genetically determined hemopoietic defect in anaemic mice of genotype Sl7Sl d Blood 26, 399-410. McLaughl in, J . , Chianese, E., and Witte, O.N. (1987). In vitro transformation of immature hematopoiet ic cells by the p210 BCR/ABL oncogene product of the Phi ladelphia chromosome. Proc. Natl. Acad. Sci. U. S. A. 84, 6558-6562. McWhirter, J.R. and Wang , J.Y.J. (1991). Activation of tyrosine kinase and microf i lament-binding functions of c-ab/by bcrsequences in bcr/ablfusion proteins. Mol. Cell Biol. 11, 1553-1565. McWhirter, J.R. and Wang , J.Y.J. (1993). An actin-binding function contr ibutes to transformation by the Bcr-Abl oncoprotein of Philadelphia chromosome-posi t ive human leukemias. E M B O J . 12, 1533-1546. McWrighter, J.R. and Wang , J.Y.J. (1991). Activation of tyrosine kinase and microf i lament-binding functions of c-ab/by bcr sequences in bcr-abl fusion proteins. Mol . Cell Biol. 11, 1553-1565. Metcalf, D. (1977). Hemopoietic Colonies. In vitro cloning of normal and leukemic cells (Berlin Heidelberg: Springer-Verlag). Metcalf, D. (1989). The molecular control of cell division, differentiation commitment and maturat ion in haemopoiet ic cells. Nature 339, 27-30. Metcalf, D. (1993). Hematopoiet ic regulators: redundancy or subtlety? Blood 82, 3515-3523. Miyake, K., Medina, K.L., Hayashi, S.I., Ono, S., Hamaoka, T., and Kincade, P.W. (1990a). Monoclonal antibodies to Pgp-1/CD44 block lympho-hemopoiesis in long-term bone marrow cultures. J . Exp. Med. 171, 477-488. Miyake, K., Underhil l , C.B., Lesley, J . , and Kincade, P.W. (1990b). Hyaluronate can funct ion as a cell adhesion molecule and CD44 participates in hyaluyronate recognit ion. J . Exp. Med. 172, 69-75. Miyake, K., Weissman, I.L., Greenberger, J.S., and Kincade, P.W. (1991). Evidence for a role of the integrin VLA-4 in lympho-hemopoiesis. J . Exp. Med. 173, 599-607. Mobley, J.L. and Dailey, M.O. (1992). Regulation of adhesion molecule expression by CD8 T cells in vivo. J . Immunol. 148, 2348-2356. Molineux, G., Migdalska, A., Szmitkowski, M., Zsebo, K., and Dexter, T .M. (1991). The effects on hematopoiesis of recombinant stem cell factor (l igand for c-kit) administered in vivo to mice either alone or in combinat ion with granulocyte colony-st imulat ing factor. Blood 78, 961-966. 155 Moore, M.A.S. (1991). Clinical implications of positive and negative hematopoiet ic stem cell regulators. Blood 78, 1-19. Morgan, D.A., Gumucio, D .L , and Brodsky, I. (1991). Granulocyte-macrophage colony-st imulat ing factor-dependent growth and erythropoiet in- induced differentiation of a human cell line MB-02. Blood 78, 2860-2871 . Morimoto, K., Robin, E., Le Brousse-Kerdi les, M.C., Li, Y., Clay, D., Jasmin, C , and Smadja-Joffe, F. (1994). CD44 mediates hyaluronan binding by human myeloid KG1A and KG1 cells. Blood 83, 657-662. Mulder, J.R., Kruyt, P.M., Sewnath, M., Oost ing, J . , Seldenrijk, C.A., We idema, W.F. , Offerhaus, G.J.A., and Pals, S T . (1994). Colorectal cancer prognosis and expression of exon-v6-containing CD44 proteins. Lancet 344, 1470-1472. Muller, A.J . , Young, J.C., Pendergast, A .M. , Pondel, M., Landau, N.R., Li t tman, D.R., and Witte, O.N. (1991). BCR first exon sequences specifically activate the BCR/ABL tyrosine kinase oncogene of Philadelphia chromosome-posi t ive human leukemias. Mol. Cell Biol. 7 7, 1785-1792. Murakami , S., Miyake, K., June, C.H., Kincade, P.W., and Hodes, R.J. (1990). IL-5 induces a Pgp-1 (CD44) bright B cell subpopulat ion that is highly enr iched in proliferative and Ig secretory activity and binds to hyaluronate. J . Immunol . 745, 3618-3627. Murakami , S., Miyake, K., Abe, R., Kincade, P.W., and Hodes, R.J. (1991a). Characterizat ion of autoantibody-secreting B cells in mice undergoing st imulatory (chronic) graft-versus-host reactions. Identification of a CD44hi population that binds specifically to hyaluronate. J . Immunol. 146, 1422-1427. Murakami , S., Miyake, K., Kincade, P.W., and Hodes, R.J. (1991b). Functional role of CD44 (Pgp-1) on activated B cells. Immunol. Res. 10, 15-27. Murakami, S., Shimabukuro, Y., Miki, Y., Saho, T , Hino, E., Kasai, D., Nozaki , T , Kusumoto, Y., and Okada, H. (1994). Inducible binding of human lymphocytes to hyaluronate via CD44 does not require cytoskeleton association but does require new protein synthesis. J . Immunol. 752, 467-477. Naujokas, M.F., Morin, M., Anderson, M.S., Peterson, M., and Miller, J . (1993). The chondroit in sulfate form of invariant chain can enhance stimulation of T cell responses through interaction with CD44. Cell 74, 257-268. Neame, S.J. and Isacke, C M . (1992). Phosphorylat ion of CD44 in vivo requires both Ser323 and Ser325, but does not regulate membrane localization or cytoskeletal interaction in epithelial cells. EMBO J. 7 7, 4733-4738. 156 Neame, S.J. and Isacke, C M . (1993). The cytoplasmic tail of CD44 is required for basolateral localization in epithelial MDCK cells but does not mediate associat ion with the detergent- insoluble cytoskeleton of f ibroblasts. J . Cell Biol. 121, 1299-1310. Nobes, C D . and Hall, A. (1995). Rho, Rac, and Cdc42 GTPases regulate the assembly of mult imolecular focal complexes associated with actin stress f ibers, lamell ipodia, and f i lopodia. Cell 81, 53-62. Noble, P.W., Lake, F.R., Henson, P.M., and Riches, D.W.H. (1993). Hyaluronate activation of CD44 induces insulin-like growth factor-1 expression by a tumor necrosis factor-a-dependent mechanism in murine macrophages. J . Clin. Invest. 91, 2368-2377. Nottenburg, C , Rees, G., and St.John, T. (1989). Isolation of mouse CD44 cDNA: structural features are distinct f rom the primate of cDNA. Proc. Natl. Acad. Sci. U. S. A. 86, 8521-8525. Ogawa, M. (1993). Differentiation and proliferation of hematopoiet ic s tem cells. Blood 81, 2844-2853. Orkin, S.H. (1995). Transcription factors and hematopoietic development. J . Biol. Chem. 270, 4955-4958. Otsuka, T., Eaves, C.J., Humphries, R.K., Hogge, D.E., and Eaves, A.C. (1989). Lack of evidence for abnormal autocrine or paracrine mechanisms underlying the uncontrol led proliferation of primitive chronic myeloid leukemia progenitor cells. Leukemia 5, 861-868. Pals, S T . , Hogervorst, F., Keizer, G.D., Thepen, T , Horst, E., and Figdor, C. (1989). Identification of a widely distributed 90-kDa glycoprotein that is homologous to the hermes-1 human lymphocyte homing receptor. J . Immunol. 143, 851-857. Papayannopoulou, T., Craddock, C , Nakamoto, B., Priestley, G.V., and Wolf, N.S. (1995). The V L A / V C A M - 1 adhesion pathway defines contrasting mechanisms of lodgement of t ransplanted murine hemopoietic progenitors between bone marrow and spleen. Proc. Natl. Acad. Sci. USA 92, 9647-9651. Papayannopoulou, T. and Brice, M. (1992). Integrin expression profiles during erythroid differentiation. Blood 79, 1686-1694. Papayannopoulou, T. and Nakamoto, B. (1993). Peripheralization of hemopoiet ic progenitors in primates treated with ant i-VLA 4 integrin. Proc. Natl. Acad . Sci. U. S. A. 90, 9374-9378. Patarca, R., Wei , F.-Y., Singh, P., Morasso, M.I., and Cantor, H. (1990). Dysregulated expression of the T cell cytokine Eta-1 in CD4" 8" lymphocytes during the development of murine auto immune disease. J . Exp. Med. 172, 1177-1183. 157 Patel, D.D., Hale, L.P., Whichard, L.P., Radcliff, G., Mackay, C.R., and Haynes, B.F. (1995). Expression of CD44 molecules and CD44 ligands during human thymic fetal development: expression of CD44 isoforms is developmental ly regulated. Int. Immunol . 7, 277-286. Patel, V.P. and Lodish, H.F. (1984). Loss of adhesion of murine erythroleukemia cells to f ibronectin during erythroid differentiation. Science 224, 996-998. Paul, S.R., Bennet, F., and Calvett i , J.A. (1990). Molecular cloning of a cDNA encoding interleukin 1 1 , a stromal cell-derived lymphopoietic and hematopoiet ic cytokine. Proc. Natl. Acad . Sci. USA 87, 7512-7516. Peach, R.J., Hol lenbaugh, D., Stamenkovic, I., and Aruffo, A. (1993). Identification of hyaluronic acid binding sites in the extracellular domain of CD44. J . Cell Biol. 122, 257-264. Pendergast, A .M. , Quil l iam, L.A., Cripe, L.D., Bassing, C.H., Dai, Z., Li, N., Batzer, A., Rabun, K.M., Der, C.J., Schlessinger, J . , and Gishizky, M.L. (1993). BCR-ABL- induced oncogenesis is mediated by direct interaction with the SH2 domain of the GRB-2 adaptor protein. Cell 75, 175-185. Penn, P.E., J iang, D.Z., Fei, R.G., Sitnicka, E., and Wolf, N.S. (1993). Dissecting the hematopoiet ic microenvironment. IX. Further characterization of murine bone marrow stromal cells. Blood 81, 1205-1213. Penno, M.B., August, J.T., Baylin, S.B., Mabry, M., Linnoila, R.I., Lee, V.S. , Croteau, D., Ling Yang, X., and Rosada, C. (1994). Expression of CD44 in human lung tumors. Cancer Res. 54, 1381-1387. Perschl, A., Lesley, J . , English, N., Trowbridge, I., and Hyman, R. (1995). Role of CD44 cytoplasmic domain in hyaluronan binding. Eur. J . Immunol . 25, 4 9 5 - 5 0 1 . Petzer, A.L., Hogge, D.E., Lansdorp, P.M., Reid, D.S., and Eaves, C.J. (1996a). Self-renewal of primitive human hematopoietic cells (long-term-culture-init iating cells) in vitro and their expansion in defined medium. Proc. Natl. Acad. Sci. U. S. A. 93, 1470-1474. Petzer, A.L. , Eaves, C.J., Lansdorp, P.M., Ponchio, L., Barnett, M., and Eaves, A.C. (1996b). Characterizat ion of primitive subpopulat ions of normal and leukemic cells present in blood of patients with newly diagnosed as well as establ ished chronic meyloid leukemia. Blood {in press) Petzer, A.L., Zandstra, P.W., Piret, J .M., and Eaves, C.J. (1996c). Differential cytokine effects on primitive (CD34 + CD38) human hematopoietic cells: novel responses to flt3-l igand and thrombopoiet in. J . Exp. Med. (in press) 158 Pevny, L , Lin, C.S., D'Agati, V., Simon, M.C., Orkin, S.H., and Costantini , F. (1995). Development of hematopoietic cells lacking transcription factor G A T A - 1 . Development 727,163-172. Picker, L.J., de los Toyos, J . , Telen, M.J., Haynes, B.F., and Butcher, E.C. (1989). Monoclonal antibodies against the CD44 [ln(Lu)-re\a\ed p80], and Pgp-1 antigens in man recognize the hermes class of lymphocyte homing receptors. J . Immunol . 742, 2046-2051 . Puil, L., Liu, J . , Gish, G., Mbamalu, G., Bowtel l , D., Pelicci, P.G., Ar l inghaus, R., and Pawson, T. (1994). Bcr-Abl oncoproteins bind directly to activators of the ras signalling pathway. E M B O J. 13, 764-773. Pure, E., Camp, R.L., Peritt, D., Panettieri, A.R., Jr., Lazaar, A.L., and Nayak, S. (1995). Defective phosphorylat ion and hyaluronate binding of CD44 with point mutat ions in the cytoplasmic domain. J. Exp. Med. 757, 55-62. Pytela, R., Pierschbacher, M.D., and Ruoslahti, E. (1985). Identification and isolation of a 140 kd cell surface glycoprotein with properties expected of a f ibronectin receptor. Cell 40, 191-198. Qiu, R.G., Chen, J . , Kirn, D., McCormick, F., and Symons, M. (1995a). An essential role for Rac in Ras transformation. Nature 374, 457-459. Qiu, R.G., Chen, J . , McCormick, F., and Symons, M.A. (1995b). Role for Rho in Ras transformat ion. Proc. Natl. Acad. Sci. U. S. A. 92, 11781-11785. Raefsky, E.L., Platanias, L.C., Zoumbos, N.C., and Young, N.S. (1985). Studies of interferon as a regulator of hematopoietic cell proliferation. J . Immunol . 735, 2507-2512. Rafii, S., Shapiro, F., Rimarachin, J . , Nachman, R.L., Ferris, B., Weksler, B., Moore, M.A.S., and Asch, A .S. (1994). Isolation and characterization of human bone marrow microvascular endothelial cells: hematopoietic progenitor cell adhesion. Blood 84, 10-19. Raitano, A.B. , Halpern, J.R., Hambuch, T.M., and Sawyers, C.L. (1995). The Bcr-Abl leukemia oncogene activates Jun kinase and requires Jun for t ransformat ion. Proc. Natl. Acad . Sci. U. S. A. 92, 11746-11750. Rapraeger, A . C , Krufka, A., and Olwin, B.B. (1991). Requirement of heparin-sulphate for b-FGF mediated fibroblast growth and myoblast differentiation. Science 252, 1705-1708. Raskind, W . H . and Fialkow, P.J. (1987). The use of cell markers in the study of human hematopoiet ic neoplasia. Adv. Cancer Res. 49, 127-167. 159 Reber, S., Matzku, S., Gunthert, U., Ponta, H., Herrl ich, P., and Zoller, M. (1990). Retardation of metastatic tumor growth after immunizat ion with metastasis-specif ic monoclonal antibodies. Int. J . Cancer 46, 919-927. Renshaw, M.W., McWhirter, J.R., and Wang , J.Y.J. (1995). The human leukemia oncogene bcr-abl abrogates the anchorage requirement but not the growth factor requirement for proliferation. Mol. Cell Biol. 15, 1286-1293. Reuss-Borst, M.A., Steinke, B., Waller, H.D., Buhring, H.J., and Muller, C A . (1992). Phenotypic and clinical heterogeneity of CD56-posit ive acute nonlymphoblast ic leukemia. Ann . Hematol . 64, 78-82. Ridley, A .J . , Paterson, H.F., Johnston, C.L., Diekmann, D., and Hall, A. (1992). The small GTP-binding protein rac regulates growth factor- induced membrane ruffling. Cell 70, 401-410. Ridley, A .J . (1994). Membrane ruffling and signal t ransduct ion. Bioessays 16, 321-327. Ridley, A .J . and Hall, A. (1992). The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70, 389-399. Ristamaki, R., Joensuu, H., Salmi, M., and Jalkanen, S. (1994). Serum CD44 in malignant lymphoma: an association with treatment response. Blood 84, 238-243. Roberts, R., Gallagher, J . , Spooncer, E., Al len, T.D., Bloomfield, F., and Dexter, T .M. (1988). Heparan sulphate bound growth factors: A mechanism for stromal cell mediated haemopoiesis. Nature 332, 376-378. Rodrigues, M., Nussenzweig, R.S., Romero, P., and Zavala, F. (1992). The in vivo cytotoxic activity of C D 8 + T cell clones correlates with their levels of expression of adhesion molecules. J . Exp. Med. 175, 895-905. Rosales, C , O'Brien, V., Kornberg, L , and Juliano, R. (1995). Signal t ransduct ion by cell adhesion receptors. Biochim. Biophys. Acta 1242, 77-98. Rosemblatt , M., Vuil let-Gaugler, M.H., Leroy, C , and Coulombel , L. (1991). Coexpression of two fibronectin receptors, VLA-4 and VLA-5, by immature human erythroblastic precursor cells. J. Clin. Invest. 87, 6 -11 . Rossbach, H.C., Kr izanac-Bengez, L , Santos, E.B., Gooley, T.A., and Sandmaier, B.M. (1996). An antibody to CD44 enhances hematopoiesis in long-term marrow cultures. Exp. Hematol . 24, 221-227. Rothman, B.L., Blue, M.L., Kelley, K.A., Wunderl ich, D., Mierz, D.V., and Aune, T .M. (1991). Human T cell activation by OKT3 is inhibited by a monoclonal ant ibody to CD44. J . Immunol . 147, 2493-2499. 160 Ruiz, P., Schwarzler, C , and Gunthert, U. (1995). CD44 isoforms during differentiation and development. Bioessays 17, 17-24. Ruoslahti , E. (1989). Proteoglycans in cell regulation. J . Biol. Chem. 264, 13369-13372. Ruoslahti , E. (1991). Integrins. J . Clin. Invest. 87, 1-5. Ruoslahti , E. and Piersbacher, M.D. (1987). New perspectives in cell adhesion: RGD and integrins. Science 238, 491-497. Salgia, R., Brunkhorst, B., Pisick, E., Li, J.-L., Lo, S.H., Chen, L.B., and Griff in, J.D. (1995a). Increased tyrosine phosphorylat ion of focal adhesion proteins in myeloid cell lines expressing p210BCR/ABL. Oncogene 11, 1149-1155. Salgia, R., Li, J.-L., Lo, S.H., Brunkhorst, B., Kansas, G.S., Sobhany, E.S., Sun, Y., Pisick, E., Hallek, M., Ernst, T., Tantravahi, R., Chen, L.B., and Griffin, J.D. (1995b). Molecular cloning of human paxillin, a focal adhesion protein phosphorylated by P 2 1 0 B C R / A B L j B j 0 | C h e m 270, 5039-5047. Salgia, R., Sattler, M., Pisick, E., Li, J.L., and Griffin, J.D. (1996). p 2 1 0 B C R / A B L induces format ion of complexes containing focal adhesion proteins and the protooncogene product p 1 2 0 c C b l . Exp. Hematol . 24, 310-313. Salles, G., Zain, M., J iang, W., Boussiotis, V.A., and Shipp, M.A. (1993). Alternatively spliced CD44 transcripts in diffuse large-cell lymphomas: Characterizat ion and compar ison with normal activated B cells and epithelial mal ignancies. Blood 82, 3539-3547. Salmi, M., Gron-Virta, K., Sointu, P., Grenman, R., Kalimo, H., and Jalkanen, S. (1993). Regulated expression of exon v6 containing isoforms of CD44 in Man: downregulat ion during malignant transformation of tumors of squamocel lu lar origin. J . Cell Biol. 122, 431-442. Sandmaier , B.M., Storb, R., Appelbaum, F.R., and Gallatin, W . M . (1990). An antibody that facil itates hematopoiet ic engraftment recognizes CD44. Blood 76, 630-635. Satoh, T., Nakafuku, M., Miyaj ima, A., and Kaziro, Y. (1991). Involvement of ras p21 protein in signal-transduction pathways from interleukin 2, interleukin 3, and granulocyte/macrophage colony-stimulating factor, but not f rom interleukin 4 . Proc. Natl. Acad . Sci. U. S. A. 88, 3314-3318. 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 C D 3 4 + subpopulat ions of human bone marrow cells. Proc. Natl. Acad. Sci. U. S. A. 91, 12223-12227. 161 Sawyers, C.L., Denny, C.T., and Witte, O.N. (1991). Leukemia and the disruption of normal hematopoiesis. Cell 64, 337-350. Sawyers, C.L., McLaughl in, J . , and Witte, O.N. (1995). Genetic requirement for Ras in the transformation of f ibroblasts and hematopoietic cells by the bcr-abl oncogene. J . Exp. Med. 181, 307-313. Schaller, M.D., Borgman, C.A., Cobb, B.S., Vines, R.R., Reynolds, A.B. , and Parsons, J.T. (1992). pp125 F A K , a structurally distinctive protein-tyrosine kinase associated with focal adhesions. Proc. Natl. Acad. Sci. USA 89, 5192-5196. Schaller, M.D. and Parsons, J.T. (1994). Focal adhesion kinase and associated proteins. Curr. Opin. Cell Biol. 6, 705-710. Schlaepfer, D.D., Hanks, S.K., Hunter, T., and van der Geer, P. (1994). Integrin-mediated signal transduction linked to Ras pathway by G R B 2 binding to focal adhesion kinase. Nature 372, 7 8 6 - 7 9 1 . Schlessinger, J . , Lax, I., and Lemmon, M. (1995). Regulation of growth factor activation by proteoglycans: what is the role of the low affinity receptors? Cell 83, 357-360. Schmitt, C , Eaves, C.J., and Lansdorp, P.M. (1991). Expression of CD34 on human B cell precursors. Clin. Exp. Immunol. 85, 168-173. Scott, E.W., S imon, M.C., Anastasi , J . , and Singh, H. (1994). Requirement of transcription factor PU.1 in the development of multiple hematopoiet ic l ineages. Science 265, 1573-1577. Screaton, G.R., Bell, M.V., Jackson, D.G., Cornelis, F.B., Gerth, U., and Bell, J. I . (1992). Genomic structure of DNA encoding the lymphocyte homing receptor CD44 reveals at least 12 alternatively spliced exons. Proc. Natl. Acad . Sci. U. S. A. 89, 12160-12164. Screaton, G.R., Caceres, J.F., Mayeda, A., Bell, M.V., Plebanski, M., Jackson, D.G., Bell, J. I . , and Krainer, A.R. (1995). Identification and characterization of three members of the human SR family of pre-mRNA splicing factors. EMBO J. 14, 4336-4349. Seiter, S., Arch, R., Reber, S., Komitowski, D., Hofmann, S., Ponta, H., Herrl ich, P., Matzku, S., and Zoller, M. (1993). Prevention of tumor metastasis formation by anti-variant CD44. J . Exp. Med. 177, 443-455. Seth, A., Gote, L., Nagarkatt i , M., and Nagarkatt i , P.S. (1991). T-cell-receptor-independent activation of cytolytic activity of cytotoxic T lymphocytes mediated through CD44 and gp90 M E L " 1 4 . Proc. Natl. Acad. Sci. USA 88, 7877-7881 . 162 Shimizu, Y., van Seventer, G.A., Siraganian, R., Wahl , L., and Shaw, S. (1989). Dual role of the CD44 molecule in T cell adhesion and activation. J . Immunol . 143, 2457-2463. Shiota, Y., Minguell , J . , Zanjani, E.D., and Tavassol i , M. (1992). Induction of upmodulat ion of homing receptors in cloned hemopoietic progenitors by growth factors. Bone Marrow Transplant 9, 123-127. Siczkowski , M., Clarke, D., and Gordon, M.Y. (1992). Binding of primitive hematopoiet ic progenitor cells to marrow stromal cells involves heparan sulfate. Blood 80, 912-919. Siczkowski , M., Amos, T.A.S., and Gordon, M.Y. (1993). Hyaluronic acid regulates the function and distribution of sulfated glycosaminoglycans in bone marrow stromal cultures. Exp. Hematol . 21, 126-130. S immons, P.J., Masinovsky, B., Longenecker, B.M., Berenson, R., Torok-Storb, B., and Gallatin, W . M . (1992). Vascular cell adhesion molecule-1 expressed by bone marrow stromal cells mediates the binding of hematopoietic progenitor cells. Blood 80, 388-395. S immons, P.J. and Torok-Storb, B. (1991). Identification of stromal cell precursors in human bone marrow by a novel monoclonal antibody, S T R O - 1 . Blood 78, 55-62. Skorski , T., Kanakaraj , P., Ku, D.H., Nieborowska-Skorska, M., Canaani , E., Zon, G., Perussia, B., and Calabretta, B. (1994). Negative regulation of p120GAP GTPase promoting activity by p210bcr/abl: implication for ras-dependent Philadelphia chromosome positive cell growth. J . Exp. Med. 179, 1855-1865. Smadja-Joffe, F., Legras, S., Li, Y., Clay, D., Robin, E., Morimoto, K., LeBousse-Kerdiles, M.C., and Jasmin, C. (1994). Hematopoietic progenitors f rom normal human bone marrow adhere to hyaluronan (HA). This adhesion is mediated by CD44 and involves preferentially granulo-macrophagic progenitors. Blood 84, 416a. Spangrude, G.J., Klein, J . , Heimfeld, S., Aihara, Y., and Weissman, I.L. (1989). Two monoclonal antibodies identify thymic-repopulat ing cells in mouse bone marrow. J . Immunol . 142, 425-430. Spooncer, E., Gallagher, J.T., Krizsa, F., and Dexter, T .M. (1983). Regulation of haemopoiesis in long-term bone marrow cultures. IV. Glycosaminoglycan synthesis and the stimulation of haemopoiesis by (3-D-Xylosides. J . Cell Biol. 96, 510-514. St Jacques, S., Dadi, H.K., and Letarte, M. (1993). CD44 in human placenta: localization and binding to hyaluronic acid. Placenta 14, 25-39. St John, T., Meyer, J . , Idzerda, R., and Gallatin, W . M . (1990). Expression of CD44 confers a new adhesive phenotype on transfected cells. Cell 60, 45-52. 163 Stamenkovic, I., Amiot, M., Pesando, J.M., and Seed, B. (1989). A lymphocyte molecule implicated in lymph node homing is a member of the carti lage link protein family. Cell 56, 1057-1062. Stamenkovic, I., Aruffo, A., Amiot, M., and Seed, B. (1991). The hematopoiet ic and epithelial forms of CD44 are distinct polypeptides with different adhesion potentials for hyaluronate-bearing cells. EMBO J. 10, 343-348. Stauder, R., Eisterer, W., Thaler, J . , and Gunthert, U. (1995). CD44 variant isoforms in non-hodgkin's lymphoma: A new independent prognostic factor. Blood 85, 2885-2899. Stevens, R.L., Avraham, S., Gartner, M.C., Bruns, G.A.P., Austen, K.F., and Weis, J .H. (1988) . Isolation and characterization of a cDNA that encodes the peptide core of the secretory granule proteoglycan of human promyelocytic leukemia HL-60 cells. J . Biol. Chem. 263, 7287-7291 . Strauss, L.C., Rowley, S.D., La Russa, V.F., Sharkis, S.J., Stuart, R.K., and Civin, C.I . (1986). Antigenic analysis of hematopoiesis. V. Characterizat ion of My-10 antigen expression by normal lymphohematopoiet ic progenitor cells. Exp. Hematol . 14, 878-886. Streeter, P.R., Berg, E.L., Rouse, B.T.N., Bargatze, R.B., and Butcher, E.C. (1988). A t issue-specif ic endothelial cell molecule involved in lymphocyte homing. Nature 331, 41-46. Sugahara, H., Kanakura, Y., Furitsu, T., Ishilhara, K., Oritani, K., Ikeda, H., Ki tayama, H., Ishikawa, J . , Hashimoto, K., Kanayama, Y., and Matsuzawa, Y. (1994). Induction of programmed cell death in human hematopoietic cell lines by f ibronectin v ia its interaction with very late antigen 5. J . Exp. Med. 179, 1757-1766. Suther land, 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, 1563-1570. Suther land, H.J., Lansdorp, P.M., Henkelman, D.H., Eaves, A . C , and Eaves, C.J. (1990) . Functional characterization of individual human hematopoiet ic s tem cells cultured at limiting dilution on supportive marrow stromal layers. Proc. Natl. Acad . Sci. U. S. A. 87, 3584-3588. Suther land, H.J., Eaves, A . C , and Eaves, C.J. (1991). Quantitat ive assays for human hemopoiet ic progenitor cells. In Bone Marrow Processing and Purging: A Practical Guide. A.P. Gee, ed. (Boca Raton: CRC Press Inc), pp. 155-171. Sy, M.S., Guo, Y.-J . , and Stamenkovic, I. (1991). Distinct effects of two CD44 isoforms on tumor growth in vivo. J . Exp. Med. 174, 859-866. Sy, M.S., Guo, Y.-J . , and Stamenkovic, I. (1992). Inhibition of tumor growth in vivo with a soluble CD44- immunoglobul in fusion protein. J . Exp. Med. 176, 623-627. 164 Szilvassy, S.J., Fraser, C.C., Eaves, C.J., Lansdorp, P.M., Eaves, A . C , and Humphr ies, R.K. (1989a). Retrovirus-mediated gene transfer to purified hemopoiet ic s tem cells with long-term lympho-myelopoiet ic repopulating ability. Proc. Natl. Acad. Sci. U. S. A. 86, 8798-8802. Szi lvassy, S.J., Lansdorp, P.M., Humphries, R.K., Eaves, A . C , and Eaves, C.J. (1989b). Isolation in a single step of a highly enriched murine hematopoiet ic s tem cell populat ion with competit ive long-term repopulating ability. Blood 74, 930-939. Taher, T.E.I . , Smit, L., Giff ioen, A.W., Schilder-Tol, E.J.M., Borst, J . , and Pals, S T . (1996). Signaling through CD44 is mediated by tyrosine kinases. J . Biol. Chem. 271, 2863-2867. Takahashi , K., Stamenkovic, I., Cutler, M., Saya, H., and Tanabe, K.K. (1995). CD44 hyaluronate binding influences growth kinetics and tumorigenicity of human colon carc inomas. Oncogene 11, 2223-2232. Tan , P.H.S., Santos, E.B., Rossbach, H . C , and Sandmaier, B.M. (1993). Enhancement of Natural Killer activity by an antibody to CD44. J . Immunol . 750, 812-820. Tanaka, Y., Adams, D.H., Hubscher, S., Hirano, H., Siebenlist, U., and Shaw, S. (1993). T-cell adhesion induced by proteoglycan-immobil ized cytokine MIP-1(3. Nature 361, 79-82. Tauchi , T., Boswell , H.S., Leibowitz, D., and Broxmeyer, H.E. (1994). Coupl ing between p210bcr-abl and She and Grb2 adaptor proteins in hematopoiet ic cells permits growth factor receptor- independent link to Ras activation pathway. J . Exp. Med. 779, 167-175. Tavassol i , M. (1989). Handbook of the Hemopoietic Microenvironment (NJ: Humana Press). Teixido, J . , Hemler, M.E., Greenberger, J.S., and Anklesaria, P. (1992). Role of (31 and [32 integrins in the adhesion of human CD34 h l stem cells to bone marrow st roma. J . Clin. Invest. 90, 358-367. Thomas, L., Byers, H.R., Vink, J . , and Stamenkovic, I. (1992). C D 4 4 H regulates tumor cell migration on hyaluronate-coated substrate. J . Cell Biol. 7 75, 971-977. Thomas, L., Etoh, T., Stamenkovic, I., Mihm Jr., M.C., and Byers, H.R. (1993). Migration of human melanoma cells on hyaluronate is related to CD44 expression. J . Invest. Dermatol . 700, 115-120. Thomas , T.E. , Abraham, S.J.R., Otter, A.J . , Blackmore, E.W., and Lansdorp, P.M. (1992). High gradient magnetic separation of cells on the basis of expression levels of cell surface ant igens. J . Immunol. Methods 754, 245-252. 165 Toksoz, D., Zsebo, K.M., Smith, K.A., Hu, S., Brankow, D., Suggs, S.V., Mart in, F.H., and Wil l iams, D.A. (1993). Support of human hematopoiesis in long-term bone marrow cultures by murine stromal cells selectively expressing the membrane-bound and secreted forms of the human homolog of the steel gene product, s tem cell factor. Proc. Natl. Acad . Sci. U. S. A. 89, 7350-7354. Tolg, C , Hofmann, M., Herrl ich, P., and Ponta, H. (1993). Splice choice f rom ten variant exons establishes CD44 variability. Nucleic Acids Res. 21, 1225-1229. Toskoz, D., Zsebo, K.M., and Smith, H.A. (1993). Support of murine hematopoiesis in long-term bone marrow cultures by murine stromal cells selectively expressing the membrane bound and secreted forms of the human homolog of the steel gene product, s tem cell factor. Proc. Natl. Acad. Sci. USA 89, 7350-7354. Toyama-Sor imachi , N., Miyake, K., and Miyasaka, M. (1993). Activation of CD44 induces ICAM-1/LFA-1- independent, C a 2 + , Mg 2 + - independent adhesion pathway in lymphocyte-endothel ial cell interaction. Eur. J . Immunol. 23, 439-446. Toyama-Sor imachi , N., Sor imachi, H., Tobita, Y., Ki tamura, F., Yagi ta, H., Suzuki , K., and Miyasaka, M. (1995). A novel ligand for CD44 is serglycin, a hematopoiet ic cell l ineage-specif ic proteoglycan. J . Biol. Chem. 270, 7437-7444. Toyama-Sor imachi , N. and Miyasaka, M. (1994). A novel l igand for CD44 is sulfated proteoglycan. Int. Immunol . 6, 655-660. Trevisan, M. and Iscove, N.N. (1995). Phenotypic analysis of murine long-term hemotopoiet ic reconstituting cells quantitated competit ively in vivo and comparsion with more advanced colony-forming progeny. J . Exp. Med. 181, 93-103. Trowbr idge, I.S., Lesley, J . , Schulte, R., Hyman, R., and Trotter, J . (1982). Biochemical characterizat ion and cellular distribution of a polymorphic, murine cell-surface glycoprotein expressed on lymphoid t issues. Immunogenet ics 15, 299-312. Tsai , S., Sieff, C.A., and Nathan, D.G. (1986). Stromal cel l-associated erythropoiesis. Blood 67, 1418-1426. Tsuki ta, S., Oishi, K., Sato, N., Sagara, J . , and Kawai, A. (1994). ERM family members as molecular linkers between the cell surface glycoprotein CD44 and act in-based cytoskeletons. J . Cell Biol. 126, 391 -401 . Udomsakdi , C , Eaves, C.J., Lansdorp, P.M., and Eaves, A.C. (1992a). Phenotypic heterogeneity of primitive leukemic hematopoietic cells in patients with chronic myeloid leukemia. Blood 80, 2522-2530. Udomsakdi , C , Eaves, C.J., Swolin, B., Reid, D.S., Barnett, M.J., and Eaves, A.C. (1992b). Rapid decline of chronic myeloid leukemic cells in long-term culture due to a defect at the leukemic stem cell level. Proc. Natl. Acad. Sci. U. S. A. 89, 6192-6196. 166 Udomsakdi , C , Lansdorp, P.M., Hogge, D.E., Reid, D.S., Eaves, A . C , and Eaves, C.J. (1992c). Characterizat ion of primitive hematopoietic cells in normal human peripheral b lood. Blood 80, 2513-2521 . Underhil l , C. (1992). CD44: the hyaluronan receptor. J . Cell Sci . 103, 293-298. Underhil l , C.B., Green, S.J., Comogl io, P.M., and Tarone, G. (1987). The hyaluronan receptor is identical to a glycoprotein of Mr 85, 000 (gp85) as shown by a monoclonal ant ibody that interferes with binding activity. J . Biol. Chem. 262, 13142-13146. Underhil l , C.B. and Toole, B.P. (1979). Binding of hyaluronate to the surface of cultured cells. J . Cell Biol. 82, 475-484. Underhil l , C.B. and Toole, B.P. (1982). Transformat ion-dependent loss of the hyaluronate-containing coats of cultured cells. J . Cell Physiol. 110, 123-128. Undritz, E. (1950). Die regionalen monocyten der blutkorperchnnester. Folia Haematol . (Leipz). 70, 32-42. Upadhyaya, G., Guba, S . C , Sih, S.A., Feinberg, A.P., Talpaz, M., Kantarj ian, H.M., Deisseroth, A.B. , and Emerson, S.G. (1991). Interferon-alpha restores the deficient expression of the cytoadhesion molecule lymphocyte function antigen-3 by chronic myelogenous leukemia progenitor cells. J . Clin. Invest. 88, 2131-2136. Varki , A. (1994). Selectin l igands. Proc. Natl. Acad. Sci. U. S. A. 92, 7390-7397. Verfail l ie, C , Blakolmer, K., and McGlave, P. (1990). Purified primitive human hematopoiet ic progenitor cells with long-term in vitro repopulating capacity adhere selectively to irradiated bone marrow stroma. J . Exp. Med. 172, 509-520. Verfail l ie, C M . , McCarthy, J.B., and McGlave, P.B. (1991). Differentiation of primitive human mult ipotent hematopoietic progenitors into single l ineage clonogenic progenitors is accompanied by alterations in their interaction with f ibronec. J . Exp. Med. 174, 693-703. Verfail l ie, C M . , McCarthy, J.B., and McGlave, P.B. (1992). Mechanisms underlying abnormal trafficking of malignant progenitors in chronic myelogenous leukemia. J . Clin. Invest. 90, 1232-1241. Verfail l ie, C M . (1993). Soluble factor(s) produced by human bone marrow st roma increase cytokine-induced proliferation and maturation of primitive hematopoiet ic progenitors while preventing their terminal differentiation. Blood 82, 2045-2053. Verfail l ie, C , Hurley, R., Bhatia, R., and McCarthy, J.B. (1994a). Role of bone marrow matrix in normal and abnormal hematopoiesis. Crit. Rev. Oncol . Hematol . 16, 201-224. 167 Verfail l ie, C M . , Benis, A., Lida, J . , McGlave, P.B., and McCarthy, J.B. (1994b). Adhesion of commit ted human hematopoietic progenitors to synthetic peptides from the C-terminal heparin-binding domain of f ibronectin: cooperat ion between the integrin a4p1 and the CD44 adhesion receptor. Blood 84, 1802-1811. Visvader, J . and Adams, J .M. (1993). Megakaryocyt ic differentiaion induced in 416B myeloid cells by GATA-2 and GATA-3 transgenes or 5-azacytidine is tightly coupled to GATA-1 expression. Blood 82, 1493-1501. Vuil let-Gaugler, M.H., Breton-Gorius, J . , Vainchenker, W., Guichard, J . , Leroy, C , Tchernia, G., and Coulombel , L. (1990). Loss of attachment to f ibronectin with terminal human erythroid differentiation. Blood 75, 865-873. Warren, A.J . , Col ledge, W.H. , Carl ton, M.B., Evan, M.J., Smith, A .J . , and Rabbitts, T .H. (1994). The oncogenic cysteine-rich LIM domain protein rbtn2 is essential for erythroid development. Cell 78, 45-57. Wayner, E.A. and Carter, W.G. (1987). Identification of multiple cell adhesion receptors for col lagen and f ibronectin in human f ibrosarcoma cells possessing unique a and common p subunits. J . Cell Biol. 705, 1873-1884. Webb, D.S.A., Shimizu, Y., van Seventer, G.A., Shaw, S., and Gerrard, T . L (1990). LFA-3, CD44, and CD45: Physiologic triggers of human monocyte T N F and IL-1 release. Science 1295-1297. Weber, G.F., Ashkar, S., Glimcher, M.J., and Cantor, H. (1996). Receptor- l igand interaction between CD44 and Osteopontin (Eta-1). Science 277, 509 Welte, K., Platzer, E., Lu, L., Gabri love, J.L., Levi, E., Merte lsmann, R., and Moore, M.A.S. (1985). Purification and biochemical characterization of human pluripotent hematopoiet ic colony-stimulating factor. Proc. Natl. Acad. Sci. U. S. A. 82, 1526-1530. Welte, K., Bonil la, M.A., Gillio, A.P., Boone, T . C , Potter, G.K., Gabri love, J . L , Moore, M.A.S., O'Reilly, R.J., and Souza, L.M. (1987). Recombinant human granulocyte colony-st imulat ing factor. Effects on hematopoiesis in normal and cyclophosphamide-treated primates. J . Exp. Med. 755, 941-948. Wetzler, M., Kurzrock, R., Lowe, D.G., Kantarj ian, H., Gut terman, J .U. , and Talpaz, M. (1991a). Alteration in bone marrow adherent layer growth factor expression: A novel mechanism of chronic myelogenous leukemia progression. Blood 78, 2400-2406. Wetzler, M., Talpaz, M., Lowe, D.G., Baiocchi, C , Gut terman, J.U. , and Kurzrock, R. (1991b). Constitutive expression of leukemia inhibitory factor RNA by human bone marrow stromal cells and modulation by IL -1 , T N F - a and TGF-p. Exp. Hematol . 19, 347-352. 168 Wight, T .N. , Kinsella, M.G., Keating, A., and Singer, J.W. (1986). Proteoglycans in human long-term bone marrow cultures: biochemical and ultrastructural analyses. Blood 67, 1333-1343. Wil l iams, D.A., Rios, M., Stephens, C , and Patel, V.P. (1991). Fibronectin and VLA-4 in haematopoiet ic stem cel l-microenvironment interactions. Nature 352, 4 3 8 - 4 4 1 . Wir th, K., Arch, R., Somasundaram, C , Hofmann, M., Weber, B., Herrl ich, P., Matzku, S., and Zoller, M. (1993). Expression of CD44 isoforms carrying metastasis-associated sequences in newborn and adult rats. Eur. J . Cancer 29A, 1172-1177. Witte, O.N. (1990). Steel locus defines new multipotent growth factor. Cell 63, 5-6. W u , L , Kincade, P.W., and Shortman, K. (1993). The CD44 expressed on the earliest intrathymic precursor population functions as a thymus homing molecule but does not bind to hyaluronate. Immunol. Lett. 38, 69-75. Yanai , N., Sekine, C , Yagita, H., and Obinata, M. (1994). Role for integrin very late activation antigen-4 in stromal-dependent erythropiesis. Blood 83, 2844-2850. Yang, B., Yang, B.L., Savani , R.C., and Turley, E.A. (1993). Identification of a common hyaluronan binding motif in the hyaluronan binding proteins R H A M M , CD44 and link protein. E M B O J. 13, 286-296. Yang, H. and Binns, R.M. (1993). Isolation and characterization of the soluble and membrane-bound porcine CD44 molecules. Immunology 78, 547-554. Yoder, M.C. and Wil l iams, D.A. (1995). Matrix molecule interations with hematopoiet ic s tem cells. Exp. Hematol . 23, 961-967. Zandstra, P.W., Eaves, C.J., and Piret, J .M. (1994). Expansion of hematopoiet ic progenitor cell populations in stirred suspension bioreactors of normal human bone marrow cells. Biotechnology 12, 909-914. Zannett ino, A.C.W., Berndt, M . C , Butcher, C , Butcher, E . C , Vadas, M.A., and S immons, P.J. (1995). Primitive human hematopoietic progenitors adhere to P-selectin (CD62P). Blood 85, 3466-3477. Zheng, Z., Katoh, S., He, Q., Oritani, K., Miyake, K., Lesley, J . , Hyman, R., Hamik, A., Parkhouse, M.E., Farr, A.G. , and Kincade, P.W. (1995). Monoclonal antibodies to CD44 and their influence on hyaluronan recognit ion. J . Cell Biol. 120, 485-495. Zhou, D.F.H., Ding, J.F., Picker, L.J., Bargatze, R.F., Butcher, E . C , and Goeddel , D.V. (1989). Molecular cloning and expression of Pgp-1 : the mouse homolog of the human H-CAM (Hermes) lymphocyte homing receptor. J . Immunol. 143, 3390-3395. Z igmond, S.H. (1996). Signal transduction and actin f i lament organizat ion. Curr. Opin. Cell Biol. 8, 66-73. 169 Zon, L.I. (1995). Developmental biology of hematopoiesis. Blood 86, 2 8 7 6 - 2 8 9 1 . Zsebo, K.M., Wypych, J . , McNiece, I.K., Lu, H.S., Smith, K.A., Karkare, S.B., Sachdev, R.K., Yuschenkoff , V.N. , Birkett, N.C., Wil l iams, L.R., Satyagal, V .N. , Tung , W., Bosselman, R.A., Mendiaz, E.A., and Langley, K.E. (1990). Identif ication, purif ication, and biological characterization of hematopoietic s tem cell factor f rom buffalo rat liver-condit ioned medium. Cell 63, 195-201. 170 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items