Biology of the hemopoietic microenvironment Mayani H, Guilbcrt LJ, Janowska-Wieczorck A. Biology of the hcmopoictic microenvironment. Eur J Haematol 1992: 49: 225-233. 0 Munksgaard 1992. Abstract: I n adult mammals, hemopoiesis takes place primarily in the bone marrow. The steady-state production of blood cells depends to a large extent on the interaction between hemopoietic stem/progenitor cells (HPC) and the different components of the microenvironment present in the medullary cavity. During the last three decades, in vivo and in vitro studies have allowed significant advances in understanding of the biology of such a hemopoietic microenvironment. Although not evident in histological sections, it is well known that the hemopoietic microenvironment is a highly organized structure that regulates the location and physiology of HPC. The hemopoictic microenvironment is composed of stromal cells (fibroblasts, macrophages, endothelial cells, adipocytes), accessory cells (T lymphocytes, monocytes), and their products (extracellular matrix and cytokines). Microenvironmental cells can regulate hemopoiesis by interacting directly (cell-to-cell contact) with HPC and/or by secreting regulatory molcculcs that influence, in a positive or negative manner, HPC growth. Recent in vitro studies suggest that functional abnormalities of the hemopoietic microenvironment may be implicated in the manifestation of certain hematological disorders such as aplastic anemia, and acute and chronic myelogenous leukemia. Thus, the charactcrization of the structure and function of the human hemopoietic microenvironment may have relevance in understanding and treating different hematological disorders.
Introduction
Effective hemopoiesis is the result of the interplay between hemopoietic stem/progenitor cells and a supporting stroma localized in very specific organs (1-4). In normal adult mammals, more than 95 of the hemopoietic activity occurs in the bone marrow. In fact, this is the only site in which myelopoiesis, erythropoiesis, and lymphopoiesis proceed simultaneously. This has led to the conclusion that local tissue influences critical for hemopoiesis operate primarily in the medullary cavity ( 5 ) . Under certain conditions, the spleen can also act as a hemopoietic organ (l), indicating that suitable conditions for hemopoiesis are also present in it. During the last three decades, extensive studies, both in vivo and in vitro, have been focused on the structural and physiological characterization of the environment in which hemopoiesis takes place. With the development of experimental systems for the inv i m culture of stromal cells (4, 6), and the cloning of the genes for several hemopoietic cytokines (7-9) significant advances in understanding the way in which the hemopoietic microenvironment regulates blood cell production have been made. The present article intends to present an overview on the struc-
Hector Mayani, Larry J. Guilbert and Anna Janowska-Wieczorek Departments of Medicine and Immunology, University of Alberta and Canadian Red Cross Blood Center, Edmonton, Alberta, Canada
Key words: bone marrow - cytokines extracellular matrix - hemopoiesis - stroma
Correspondence: Anna Janowska-Wieczorek M.D., Ph.D., Div. Clinical Hematology, Department of Medicine, University of Alberta, 8249-1 14 Street, Canada T6G 2R8. Phone: (403)431-0202; FAX: (403)43 1-046 1. Accepted for publication 28 August 1992
ture and composition of the hemopoietic microenvironment and the different ways in which it regulates blood cell production. Particular emphasis is placed on the hemopoietic microenvironment in humans under both normal and pathological conditions. Evidence for a hemopoietic microenvironment
The first evidence for the presence of specific conditions, in the bone marrow and spleen, necessary for hemopoietic development came from experiments in which mouse marrow cells were injected into syngeneic, irradiated mice. The first capillary bed those cells encounter is in the lungs, but the lungs do not become hemopoietic, nor d o the skeletal muscles or other body parts to which those injected cells that escape the lungs may be shunted next. Hemopoiesis is established exclusively in the bone marrow and, to a lesser extent, in the spleen (in the form of macroscopic colonies (10). Later on, it was observed that the hemopoietic colonies developed in the spleen were primarily erythroid, whereas those developed in the marrow were mainly granulocytic (1 1, 12). Furthermore, it was also shown that, among the colonies developed in the spleen, those 225
Mayani et al. developed in the red pulp were almost exclusively erythroid, whereas the granulocytic colonies developed under the splenic capsule and the trabeculae
MEDULLARY CAVITY
(1 3).
Microgeographical regions have also been described in the bone marrow. Lord and colleages have shown that the spatial distribution of stem cells, as well as of erythroid and myeloid committed progenitors in the mouse bone marrow, is not random (14, 15). They suggest that the hernopoietic tissue possesses a well-defined structure, although not evident in histological sections, that is relevant to the proliferation and differentiation of hernopoietic cells. Based on both in vivo and in vitro evidence, Schofield has proposed the existence of a stem cell “niche”, considered to be a functional compartment of marrow strorna that isolates conditions for self replication from conditions conducive to differentiation (16). ACCESSORY CELLS
Definition of the hemopoietic microenvironment
The marrow stromal tissue is a network of cells (fibroblast, endothelial cells, macrophages, adipocytes) and extracellular matrix (collagen, laminin, fibronectin, proteoglycans) which physically supports the hemopoietic cells and influences their proliferation and differentiation (17). Stromal cells have been considered to be organized into hemopoietic microenvironments that support blood cell development. The term hemopoietic microenvironment was created to distinguish different microgeographic stroma1 cell influences within an organ on the pluripotent stem cells, as opposed to whole organ differences, each of which is composed of the sum of its stromal microenvironments ( 18). The hemopoietic microenvironment, however, is not easy to define. If it is taken to be a broad term describing those cells that regulate hemopoietic development, other cells in addition to the stromal cells may be components of it. This is, in fact, the case for T lymphocytes, which are known to regulate hemopoiesis by secreting multiple hemopoietic cytokines (19). Thus, using this definition, stromal cells contribute to the hemopoietic microenvironment, but not all microenvironment cells are strornal cells. In the present article, we define the hemopoietic microenvironment as the local network of stromal cells (jibroblast, macrophages, endothelial cells, udipocytes), accessory cells (T lymphocytes and monocytes) and their products (extracellularmatrix and cytokines), capable qf injuencing the self renewal, proliferation and diferentiation of hemopoietic stem/ progenitor cells (see Fig. 1). The rest of this review will be focused to the structure and physiology of the hemopoietic microenvironment of the bone marrow. 226
v;(
Fig. 1. Schcmatic representation of the different components of the hemopoietic microenvironment and the way they influence the proliferation and differentiation of hemopoielic stem/progenitor cells (HPC). Stromal cells (fibroblasts, macrophages, endothelial cells, adipocytes) can interact with HPC via direct cell-to-cell contact (a). They can also produce proteins that constitute the extracellular matrix (b), which, in turn, interact with HPC (c). On the other hand, stromal cells are also a major source of cytokines (d), which may remain as cell-associated molecules and be presented to the H P C during the direct interaction between stromal cells and HPC, or may be secreted and act on their target cclls (e). Cytokines can also be retained by the extracellular matrix and be presented to the HPC in that way (f). Accessory cells (T lymphocytes and monocytcs) influence hemopoiesis by the production of cytokines (g). Finally, it is important 10 note that stroma1 and accessory cells not only produce cytokines, but are also regulated by them (hj).
Composition of the hemopoietic microenvironment in vivo
Reticular cells are the major cellular components of the bone marrow stroma (20) and there are two types, adventitial and fibroblastic. Adventitial reticular cells are located around venous sinuses forming a layer that partially covers the endothelium’s abluminal side. Morphological studies indicate that one of their major roles is the regulation of the migration of mature blood cells from the marrow to the circulation (21). Fibroblastic reticular cells, on the other hand, are located within the marrow hemopoietic cord; their thin cytoplasmic processes envelop maturing hemopoietic cells in the marrow compartment (22). These fibroblastic cells are usually in close contact with immature granulocytic cells, which suggests that there are functional interactions between the two cell types (23,24). Macrophages are the second major cellular component of the marrow stroma (2 1). Topographically,
The hemopoietic microenvironment they are found in three different sites in the marrow: as central macrophages in the erythroblastic islands, perisinal macrophages on the abluminal side of the sinus endothelium, and also dispersed between the developing hemopoietic cells. The central macrophage is usually surrounded by maturing red cells (erythroblasts), thus forming an anatomic unit commonly referred to as an erythroblastic island (25). It has been suggested that the central macrophage plays a key role in erythroid maturation. The possibility exists that macrophages of the erythroblastic island produce or process factors that are then transferred directly to the erythroblasts through the membrane channels of communication (22). Hypoxia, which stimulates red cell production, also leads to an increase in the central macrophage’s size and its phagocytic activity (2 1). Perisinal macrophages cover varying proportions of the sinus endothelium’s abluminal surface. Their cytoplasmic protrusions penetrate the endothelium, engulf senescent or defective red cells, and remove them from the circulation (21). Endothelial cells and udipocytes are also major components of the marrow stroma (21). Endothelial cells are the primary cell type forming the blood vessels of the marrow. Adipocytes, on the other hand, occupy the majority of the marrow cavity in adult mammals (26). Their role in hemopoiesis is not clear, although in vitro studies suggest that they are capable of producing hemopoietic growth factors (27). Marrow stromal cells produce and secrete macromolecules that form a highly organized structure known as the extrucellulur matrix (28-30). The major components of the extracellular matrix are (i) collagens, fibrillar proteins produced by reticular and endothelial cells, (ii) fibronectin, a glycoprotein produced by reticular cells and macrophages, (iii) laminin, another large glycoprotein produced by endothelial cells, and (iv) proteoglycans, composed of a core protein to which one or more glycosaminoglycans are attached. In vitro studies indicate that the extracellular matrix of the bone marrow not only acts as a physical support for the hernopoietic elements, but also plays a dynamic role in the regulation of hemopoiesis (30). The vast majority of the medullary cavity i y occupied by maturing granulocytic cells which, upon /zzchzhg their final stage, will be released into the circulation in a process mediated by endothelial and reticular cells (21). T o date, however, the role of this granulocytic pool in blood cell production is not clear. Some have suggested that the size of the granulocytic pool regulates the production of granulocyte progenitors in an inversely proportional way; that is to say, when the granulocytic pool is low, production of granulocyte progenitors is enhanced, and vice ver-
sa (1). This may suggest that such a granulocyte pool could also be considered as part of the hemopoietic microenvironment. To date, however, there is no clear evidence that maturing granulocytes are directly implicated in the regulation of progenitor cell proliferation. Composition of the hemopoietic microenvironment in vitro
Among the different in vitro models that have been developed to study the hemopoietic microenvironment, the long-term marrow culture (LTMC) system seems to be the one that most closely reproduces the conditions observed in vivo. Both murine and human hemopoiesis can be sustained for several weeks in LTMC (31, 32). This is due to the development of an adherent cell layer, consisting of different cell types, which provides the physical support and the hemopoietic cytokines (soluble and cell-associated proteins) necessary for the growth of hemopoietic progenitors (4). It is noteworthy that the composition of the adherent cell layer of LTMC is very similar to the bone marrow stroma. The major components of such an adhering layer are fibroblasts that are capable of producing collagen types I and I l l as well as fibronectin (33). These cells are the most numerous, comprising 45-55% of the total cell number (34). Macrophages are also present in significant levels, comprising 25-352) of the adherent cells (34). Endothelial cells, on the other hand, comprise 5-202) of the total cell number (35). Adipocytes are usually observed in both murine and human LTMC, although their presence may depend on certain culture conditions (4). Lymphocytic as well as myeloid cells have also been observed in the adhering layers of LTMC, although at reduced levels (34). Like its in vivo counterpart, the extracellular matrix developed in LTMC consists of collagen I, 111, and IV, laminin, fibronectin and proteoglycans (33, 36,37). Recent studies by Coulombel and colleagues have shown that erythroid progenitors adhere in vitro to extracellular matrices derived from marrow fibroblasts and that such an interaction is mediated by fibronectin (38). The same group of investigators has also demonstrated that terminal erythroid differentiation results in loss of attachment to fibronectin, providing a potential mechanism for the release of mature red blood cells from the bone marrow into the circulation (39). More recent studies by Koenigsmann et al. (40) have shown that myeloid, as well as erythroid progenitor cells have the ability to bind to collagen I, suggesting that this protein may play a role in the localization of committed myeloid and erythroid progenitors within the bone marrow. 221
Mayani et al. Origin of the microenvironment cells
The origin of the cells of the hemopoietic microenvironment has been a subject of great controversy during the last decade. As demonstrated by several investigators, bone marrow macrophages as well as circulating monocytes and T lymphocytes, are derived from the pluripotent hemopoietic stem cell (41). The origin of the stromal cells, on the other hand, is not so clear. Among the so-called “non-hemopoietic” cells of the marrow microenvironment, fibroblasts are the ones that have been best characterized in terms of their origin; thus, we will focus this subsection on these cells. Since the early work of Friedenstein and colleagues (42), and years later by Castro-Malaspina and collaborators (43), it became clear that fibroblasts derive from adherent, nonphagocytic progenitors capable of forming fibroblastic colonies in v i m (colony-forming unit-fibroblast, or CFU-F). By using a complement-mediated cytotoxicity assay, it was shown that CFU-F are different from hemopoietic progenitors (CFU-C), which may also imply different origins (43). Very recently, Simmons and TorokStorb demonstrated that CFU-F express a cellsurface antigen, absent on any other hemopoietic progenitor, recognized by the monoclonal antibody STRO-1 (44). Interestingly, cells giving rise to adipocytes and endothelial cells were also positive for STRO-1 (44). The same investigators have performed extensive phenotypical studies on CFU-F and found that, although these cells share some characteristics with the hemopoietic progenitors such as the expression of the CD34 antigen, they are distinct cell types (45). In vitro studies in human LTMC have suggested the existence of a pluripotent cell in the bone marrow that gives rise to hemopoietic cells and to their stromal microenvironment (46). Such an idea has been supported by in vitro studies in rats showing that the marrow stroma grown in culture contains cells with hemopoietic potential (47). These reports seem to give support to the study of Keating and colleagues, who found that the human hemopoietic microenvironment developed in vitro after bone marrow transplantation is of donor origin (48). However, more recent studies have indicated that the human in vitro microenvironment after bone marrow transplantation is, with the exception of macrophages, of host origin (49-5 I), suggesting independent origins for hemopoietic and stromal cells. The apparent discrepancies between the above studies may be due to differences in the techniques used to assess cultured cells and the type of cells analyzed. For instance, Laver and colleagues (50) used chromosomal analysis, which determines the origin of stroma1 cells capable of undergoing mitosis, whereas 228
Keating and colleagues (48), by determining the presence of the Y-body chromosome, analyzed nondividing cells. Thus, whether hemopoietic and stromal (fibroblastic) cells derive from a common or from independent stem cells is still an open question. Interaction between hemopoietic cells and their microenvironment
Microenvironment cells can influence hemopoiesis, either in a positive or in a negative way, by different mechanisms (Fig. 1). These include (i) direct cell-tocell contact, (ii) the secretion of proteins that constitute the extracellular matrix, and (iii) the production of soluble and cell-associated cytokines (52). During the last few years, it has become evident that more than one of these mechanisms can operate simultaneously in the interaction between microenvironmental and hemopoietic cells. Direct cell-to-cell contact between stromal and hemopoietic cells has been demonstrated, both in vivo and in vitro, by morphological studies using electron microscopy (53). These studies have shown the preferential association of fibroblasts with granulocytic cells, and macrophages with erythroid cells. This had led to the suggestion that those interactions are necessary to the optimal development of hemopoietic cells. Functional studies have indicated that murine hemopoietic stem cells (CFU-S) need to interact physically with stromal cells in order to grow in LTMC (54). Similar requirements for cell-to-cell contact have been reported for the growth of human primitive hemopoietic cells (CFU-Blast) in v i m (55). In keeping with those observations is the fact that, when human bone marrow cells are cultured in LTMC, the great majority of the hemopoietic progenitors are located in the adherent layer that develops after 3 weeks (56, 57), which supports the idea that they interact with their surrounding stroma. More recently, long-term culture-initiating cells (LTC-IC), the most suitable candidate for human hemopoietic stem cells described to date, have also been shown to be preferentially located in association with adherent stromal cells (58). Microenvironment cells are a major source of hemopoietic cytokines. They produce hemopoietic stimulators such as GM-CSF, G-CSF and M-CSF, and inhibitors, such as TNF,, TGFP, and IFN, (52). Some of these cytokines are produced only by activated microenvironment cells (T lymphocytes and macrophages); thus, they have been postulated to play a major role on blood cell production under situations of acute stress, such as infection (52). Others are constitutively produced by stromal cells. Indeed, biologically active M-CSF and G-CSF have been found in the supernatant of human LTMC at
The hemopoietic microenvironment IL-1 EZSF IL-6
--\-,
IFNT PDGF TGFp TNFlY
IL-1 SCF IL-6 IFNp
k::
IL-8 M-CSF G-CSF GM-CSF
,-,
IL-1 IL-4 IL-8 TGFP TNFa M-CSF
IL-1 PDGF
M-CSF
IL-1- IL-10 GM-CSF G-CSF IFNa IFN4 IFNT TG FB TNFa
-b
T
@--b
IL-I - IL-6 IL-9 IL-10 $$;CSF LIF TNFa -. . _
Fig. 2. Microenvironmental cells not only produce hemopoietic cytokines, but are also regulated by them. The figure shows some of the cytokines that are produced by, and act on, fibroblasts (F), macrophages (M), and T lymphocytes (T). IL, interleukin; CSF, colony-stimulating factor; G, granulocyte; M, macrophage, GM, granulocyte-macrophage; PDGF, platelet-derived growth factor; TGFP, transforming growth factor$; TNF,, tumor necrosis factor-a, SCF, stem cells factor (c-kit ligand); IFN, interferon; LIF, leukemia inhibitory factor.
levels similar to those found in vivo (32). Due to the fact that microenvironmental cells not only produce hemopoietic cytokines, but are also regulated by them (Fig. 2), the interactions between these cells, cytokines and hemopoietic cells are rather complex. In order to illustrate this point, we will present some examples. M-CSF is a lineage-specific growth factor, produced by microenvironmental cells, that stimulates the proliferation, maturation and function of monocytic cells (7-9, 59). When M-CSF is added to semisolid cultures of human bone marrow cells, it acts directly on macrophagic progenitors, stimulating their growth. Interestingly, when recombinant human M-CSF is added to human LTMC at a concentration 3- to 4-fold higher than the endogenous concentration, hemopoiesis is inhibited. This is due to the production by macrophages of a soluble activity, that contains TNF,, capable of inhibiting the growth of multipotential, myeloid and erythroid progenitors (60,61). Thus, M-CSF acts both as a positive (when acting directly on progenitor cells) and negative (when interacting with stromal cells) regulator of hemopoiesis.
Studies by Cashman and colleagues have shown that immature erythroid progenitors present in the nonadherent fraction of human LTMC are actively cycling at all times. In contrast, the same cell population is in a quiescent state when localized in the adherent layer (62). These authors have presented evidence indicating that stromal cells keep closely located hemopoietic progenitors out of cycle (Gc,)by secreting TGFP (63, 64). On the other hand, cytokines such as IL-1 or P D G F stimulate stromal cells to produce stimulatory cytokines, such as GM-CSF, and these, in turn, stimulate the hemopoietic progenitors to proliferate (63,64). Thus, the balance between the levels of hemopoietic stimulators and inhibitors seems to be of critical importance in the regulation of hemopoietic progenitor cell proliferation. The effects of GM-CSF, G-CSF, and IL-3 in human LTMC provide another example for the role of the hemopoietic microenvironment in the regulation of hemopoiesis in vitro. Addition of these factors into human LTMC results in increased levels of progenitor cells (65-67). However, when these factors are constitutively produced by genetically engineered fibroblasts at concentrations similar to those observed when the recombinant proteins were added to the cultures, their effects are significantly enhanced (66,67). These observations suggest that the way in which the hemopoietic stimulators are presented to the progenitor cells may determine the magnitude of the biological response. Interestingly, recent reports indicate that factors like GM-CSF are produced by stromal cells and then selectively retained by the glycosaminoglicans of the extracellular matrix (36, 68), thus suggesting that the extracellular matrix may play a key role in the presentation of cytokines to the hemopoietic cells. Hemopoietic cytokines may also be presented to hemopoietic cells during the direct contact with stromal cells. Indeed, cellassociated forms of factors like M-CSF and Stem Cell Factor (c-kit ligand) have been described (69, 70). The actual mechanisms by which microenvironmental cells regulate hemopoietic stem/progenitor cell proliferation and differentiation in vivo are not known in detail. It is thought that they are similar to those observed in in vitro systems (e.g. LTMC); however, it must be kept in mind that the in vivo situation is much more complex than the most complex in vitro model, hence, the development of in vivo models in which the interaction between hemopoietic and microenvironmental cells can be studied is absolutely necessary. In this regard, the SCID mouse model may be of considerable relevance.
229
Mayani et al. The hemopoietic microenvironment in hematological diseases
The fact that microenvironment cells regulate blood cell production suggests that alterations within the hemopoietic microenvironment may be involved in the development and/or progression of hematological diseases. Indeed, several in vitro studies have demonstrated functional abnormalities of the hemopoietic microenvironment in certain hemopathies. Ershler and colleagues reported the case of a woman with congenital hypoplastic anemia (Diamond-Blackfan syndrome) with marked erythroid hypoplasia. When her bone marrow cells, obtained by aspiration, were cultured in methylcellulose, an exuberant erythroid growth was observed. The erythroid nature of the colonies developed in methylcellulose was confirmed through quantitation of heme synthesis and globin mRNA accumulation in a liquid culture system. In contrast, when whole bone marrow fragments were similarly cultured, no appreciable hemoglobin synthesis was observed. Their results suggested that, in that particular patient, hypoplastic anemia resulted from an unfavorable hemopoietic microenvironment (7 1). Different investigators have observed reduced numbers of fibroblastic progenitors (CFU-F) in cultures of bone marrow from patients with acute myelogenous leukemia (AML) (72-75). Interestingly, the results of Nagao et al. suggested that the suppression of CFU-F formation in leukemic cultures was through humoral factors secreted by leukemic cells (73), implying that the reduced levels of C F U - F are the result, and not the cause, of the disease. On the other hand, a functional defect in AML-derived fibroblastic cells has been observed by Greenberg and colleagues (76). They found that, in contrast to normal marrow-derived fibroblasts, marrow fibroblasts obtained from untreated AML patients failed to enhance granulopoiesis in culture (76), thus suggesting that not only quantitative but also qualitative abnormalities may exist in AML-derived fibroblasts. Functional deficiencies in AML-derived bone marrow macrophages have also been documented. Greenberg and collaborators observed a reduced production of colony-stimulating activity by macrophages from AML patients as compared to macrophages from normal subjects (77). More recently, Mayani et al. have reported functional differences in LTMC between macrophages derived from normal and AML bone marrow: normal marrow-derived stromal adherent layers developed in LTMC showed a much better capacity to sustain normal hemopoiesis in chimeric LTMC than adherent layers derived from AML bone marrow (78). This correlated with the production by AML stromal cells of a soluble activity capable of inhibiting the growth of hemopoietic progenitors. Further studies by the same group
230
demonstrated that such an activity, which is tnainly produced by AML-derived macrophages, is in fact TNF, (61). Normal bone marrow macrophages do not seem to produce T N F a unless stimulated by CSF-1. Interestingly, CSF-1 downregulates the production of TNF, by AML-derived macrophages (6 1). The authors have suggested that macrophages developed in LTMC from AML bone marrow are functionally abnormal. This idea is supported by the fact that macrophages developed in vitro from AML peripheral blood or bone marrow have been shown to be derived from AML blasts (79, 80). l n vitro studies using two-stage LTMC have provided evidence that stromal adherent layers established from patients with chronic myelogenous leukemia (CML) show a functional defect; that is to say, they were not able to sustain the growth of hemopoietic progenitors derived from normal bone marrow (8 1). Functional abnormalities (i.e. reduced production of colony-stimulating activity) in bone marrow stromal cells from patients with aplastic anemia have also been reported (82). In contrast to the reports mentioned above, other studies indicate that the in v i m developed hemopoietic microenvironment from patients with hematological disorders such as aplastic anemia (83, 84), primary myelofibrosis ( 8 5 ) , and myelodysplasia (86) is functionally normal, suggesting that the origin and development of such diseases are exclusively due to alterations within hemopoietic cells. It is possible that the apparently contradictory results that have been reported, particularly regarding the functional integrity of the hemopoietic microenvironment in aplastic anemia, are due to differences in the experimental models that have been used. Thus, whether the hemopoietic microenvironment is involved in the development and/or progression of certain hemopathies is still an open issue. Effects of chemotherapy and radiation on the hemopoietic microenvironment
Chemotherapy and radiation are currently used to treat different forms of cancer, including leukemia. During the last decade, however, several studies have been reported indicating that both chemical agents and radiation can alter the functional integrity of the marrow microenvironment. This has been documented in murine as well as in human systems (8789). The numbers of fibroblast progenitors in mice treated with busulfan or cyclophosphamide have been found to be significantly lower than the numbers observed in control animals (90). On the other hand, studies using the LTMC system have shown that adherent layers from normal donor mice were able to support comparable levels of hemopoiesis
The hemopoietic microenvironment
(80% of the adequate control) when inoculated with bone marrow cells from normal mice. In contrast, when stroma from busulfan-treated marrow was inoculated with marrow cells from normal mice the levels of hemopoietic progenitors were only 20-40% of the levels of hemopoietic progenitors seen in cultures containing normal stroma (9 l). Similar results were observed in studies of patients in remission from acute lymphoblastic leukemia (ALL) after being treated with chemo- and radiotherapy. Testa and colleagues have shown that adherent layers froin such patients were able to support only 50% of hemopoietic growth as compared to adherent layers from normal subjects (92). Radiation has also been found to induce significant damage to the marrow stroma, especially at doses above 10 Gy (93). Some studies indicate that stromal cells are sensitive to doses as low as 2.55 Gy. However, injury at this level of exposure can be detected only in in vitro systems, since stromal cells in vivo have a large capacity for repair. Single exposure in vivo to doses of 40 Gy results in irreparable injury to the stroma and to permanent aplasia (93). At present, several approaches are being explored in order to define new strategies for the treatment of neoplasms without altering the hemopoietic microenvironment. In this context, the use of specific cytokine combinations in conjunction with mild doses of chemo- and/or radiotherapy may prove useful. Concluding remarks
Understanding the biology of the hemopoietic microenvironment is of great relevance in hematology since the optimal production of blood cells depends to a large extent on the interaction between hemopoietic stem/progenitor cells and their surrounding environment. During the last two decades, major advances have been made in the structural and physiological characterization of the hemopoietic microenvironment. However, two major questions remain unanswered. Firstly, are stromal (fibroblastic) and hemopoietic cells derived from a common progenitor? Studies supporting and rejecting such a possibility have been reported and no definite answer has yet been found. Secondly, although it seems clear that functional abnormalities in the hemopoietic microenvironment exist in certain hematological diseases, it is not known to what extent such abnormalities contribute to the development and/or progression of the disease. There is no doubt that in the next few years the answers to these and other important questions will be obtained. This, in turn, will have a major impact in understanding the basic biology of hemopoiesis and the way in which certain hemopathies develop.
Acknowledgements Research in the authors’ laboratories is funded by grants from the Alberta Cancer Board (AJW) and the National Cancer Institute of Canada (LJG). Hector Mayani was supported by 21 scholarship from the National University of Mexico.
References 1. METCALFD, MOOREMAS. Haemopoietic cells. Amstcrdam: North-Holland Publishing Co, 1971. 2. WOLF NS. The hemopoietic microenvironmcnL. Clin Hacmatol 1979: 8: 469-500. 3. DORSHKIND K. Regulation of hemopoicsis by bone marrow stromal cells and their products. Annu Rev Ininiunol 1990: 8: 111-137. 4. GREENBERCER JS. The hemopoietic microenvironment. Crit Rev Oncol Hernatol 1991: 11: 65-84. 5 . TAVASSOLI M, FRIEDENSTEIN A. Hemopoietic stromal niicroenvironment. Am J Hematol 1983: 15: 196-203. 6. DEXTER TM. SPOONCER E, SIMMONS P. ALLENTD. Longterm marrow culture: An overview of techniques and experience. New York: Alan R Liss Inc. 1984: p. 57-96. 7. SIEFFCA. Hematopoietic growth factors. J Clin Invest 1987: 79: 1549-1557. 8. CLARKSC, KAMEN K. The human hcinopoietic colonystimulating factors. Science 1987: 236: 1229- 1237. U, LANTZM, RICHTERJ. The re9. OLSSONI, GULLBERG ceptors for regulatory molecules of hemopoiesis. Eur J Haematol 1992: 48: 1-9. 10. TILLJE, MCCULLOCHEA. A direct measurement of the radiation sensitivity of normal niouse bone marrow cells. Rad Res 1961: 14: 213-222. 1 I . WOLF NS, TRENTIN J. Effects of hemopoietic organ stroma on differentiation of pluripotent stem cells. J Exp Mcd 1968: 127: 205-214. J. Determination of bone marrow stem cell differ12. TRENTIN entiation by stromal hemopoietic inductive microenvironments (HIM). Am J Pathol 1971: 65: 621-628. 13. TRENTIN J . Influence of hemopoietic organ stroina on stem cell differentiation. In: GORDON AS, ed. Regulation of hemopoiesis. New York: Meredith, 1970: 16 1- 186. 14. LORD BI, TESTANG, HENDRYJH. The relative spatial distribution of CFU-S and CFU-C in the mouse femur. Blood 1915: 46: 65-72. 15. FRASSONI F, TESTANG, LORD €31. The relative spatial distribution of erythroid progenitor cells (BFUe and CFUe) in the normal mouse femur. Cell Tissue Kinet 1982: 15: 447455. 16. SCHOFIELDR. The relationship between the splccn colonyforming cells and the haemopoietic stem cell. Blood Cells 1918: 4: 7-25. 17. OWENM. Marrow stromal stem cells. J Cell Sci 1988: suppl 10: 63-76. 18. TRENTIN J. Hemopoietic microenvironments. In: TAVASSOLI M, ed. Handbook of the hemopoietic microenvironment. Humana Press, Inc. 1989: 1-86. 19. GILLISS. T cell-derived lymphokines. In: PAULLWE, ed. Fundamental Immunology. New York: Ravcn Press, 1989: 621-637. 20. TAVASSOLI M, YOFFEYJ M . Bone niarrow: Structure and function. New York: Alan R Liss, Inc., 1983: 47-64. M. Cellular components of stroma in v i w in com21. SHAKLAI parison with in vitro systems. In: TAVASSOLI M, ed. Handbook of the hemopoietic microenvironment. Huinana Press Inc. 1989: 219-251. 22. SHAKLAI M, TAVASSOLI M. Cellular relationship in the rat bone marrow studied by freeze-fracture and lanthanum i n -
23 1
Mayani et al.
23.
24. 25. 26. 27.
28. 29. 30.
31. 32. 33. 34. 35.
36. 37.
38.
pregnation thin sectioning electron microscopy. J Ultrastruc R ~ 1979: s 69: 343-361. LAPUSHIN KW. T R E N T IJ.NIdentification of stromal elements in erythroid and neutrophil granuloid spleen colonies: Light and electron microscopy study. Exp Haematol 1977: 5: 505-522. WESTEN H, BAINTON DF. Association of alkalinephosphate-positive reticulum cells in bone marrow with granulocytic precursors. J Exp Med 1979: 150: 919-937. BESSISM, BRETON-GORIUS J. Iron metabolism in the bone marrow as seen by electron microscopy. Blood 1962: 19: 635-663. TAVASSOLI M. Marrow adipose cells and hemopoiesis. An interpretative review. Exp Hcmatol 1984: 12: 139. LANOTTEM, METCALFD, DEXTERTM. Production of monocyte/macrophage colony-stimulating factor by preadypocyte cell lines derived from murine marrow stroma. J Cell Physiol 1982: 112: 123-127. URlsT M. Bone: Formation by autoinduction. Science 1965: 150: 893-899. REDDIA, HUGGINS C. Biochemical sequences in the transformation of normal fibroblasts in adolescent rats. Proc Natl Acad Sci 1972: 69: 1601-1605. ZUCKERMAN KS, PRINCECW, GAYS. The hemopoietic extracellular matrix. In: TAVASSOLI M, ed. The handbook of the hemopoietic microenvironment. Humana Press Inc 1989: 399-432. DEXTERTM, ALLENTD, LAJTHAL. Conditions controlling the proliferation of hemopoietic stem cells in vitro. J Cell Physiol 1977: 91: 335-344. JANOWSKA-WIECZOREK A, TUPAS J, FRIESENI, et al. Kinetics of G-CSF and CSF-I production in human long-term marrow cultures. Blood 1988: 72 suppl 1: 122a. BENTLEYSA. Collagen synthesis by bone marrow stromal cells: A quantitative study. Br J Haematol 1982: S O : 491-497. ANDREONI C, MOREAUI, RIGALD. Long-term culture of human bone marrow. I. Characterization of adherent cells in flow cytometry. Exp Hematol 1990: 18: 431-437. BENTLEY SA. The role and composition of the adherent layer in long-term bone marrow culture. In: WRIGHTD G & GREENBERGER J, eds. Long-term bone marrow culture. Alan R liss Inc, New York: 1984: 141-156. KEATINGA, GORDONMY. Hierarchical organization of hemopoietic microenvironments. Leukemia 1988: 2: 766-769. WIGHTTN, KINSELLAMG, KEATINGA, SINGERJW. Proteoglycans in human long-term marrow cultures: Biochemical and ultrastructural analyses. Blood 1986: 67: 13331343. COULOMBEL L, VUILLETM, LEROYC, TCHERNIA G. Lineage and stage specific adhesion of human hemopoictic progenitor cells to extracellular matrix from marrow fibroblasts. Blood 1988: 71: 329-334.
39. VUILLET-GAUGLER MH. BRETON-GORIUS J, VAINCHENK E R W, el a[. Loss of attachment of fibronectin with terminal human erythroid differentiation. Blood 1990: 76: 865873. 40. KOENIGSMAN M, GRIFFIN JD, DICARLO J, CANNISTRA SA. Myeloid and erythroid progenitor cells from normal bone marrow adhere to collagen type I. Blood 1992: 79: 657-665. 41. SPANGRUDE GJ. Hematopoietic stem-cell differentiation. Curr Op Immunol 1991: 3: 171-178. AJ, CHAILAKHYAN KK, LALYKINA KS. 42. FRIEDENSTEIN The development of fibroblast colonies in monolayer cultures of guinea pig bone marrow and spleen colonies. Cell Tissue Kinet 1970: 3: 393-403. 43. CASTRO-MALASPINA H, GAY RE, RESNICK G, et al. Characterization of human bone marrow fibroblast colony-forming cells (CFU-F) and their progeny. Blood 1980: 56: 289-301. 44. SIMMONS P, TOROK-STORB B. Identification of stromal cell
232
45. 46.
47.
48.
49.
precursors in human bone marrow by a novel monoclonal antibody. STRO-I. Blood 1991: 78: 55-62. SIMMONS P, TOROK-STORB B. CD34 expression by stronial precursors in normal human adult bone marrow. Blood 1991: 78: 2848-2853. SINGER JW, KEATINC A, CUTTNER J. et d.Evidence for a stem cell common LO hematopoiesis and its in vitro microenvironment: Studies of patients with clonal hemopoietic neoplasia. Leuk Res 1984: 8: 535-545. ISLAM A, GONGJK, HENDERSON ES. Direct evidence for a stem cell common to hemopoiesis and its in vitro microenvironment: Studies on syngeneic (inbred) wistar furth rats. J Med 1988: 19: 119-136. KEATING A, SINGERJ, KILLENP, et al. Evidence for donor origin of the in vitro hematopoietic microenvironment following marrow transplantation in man. Nature 1982: 298: 280283. SIMMONS P, PRZEPIORKA D, THOMAS ED, ct al. Host origin of marrow stromal cells following allogeneic bone marrow transplantation. Nature 1987: 328: 429-432.
50. LAVERJ, JHANWAR SC, O'REILLY,CASTRO-MALASPINA H. Host origin of human hemopoietic microenvironment following allogeneic bone marrow transplantation. Blood 1987: 70: 1966-1968. 51. AGEMATSUK, NAKAHORIY. Recipient origin of bone marrow-derived fibroblastic stromal cells during all periods following bone marrow transplantation in humans. Br J Haematol 1991: 79: 359-365. B. Cellular interactions. Blood 1988: 72: 37352. TOROK-STORB 385. 53. WEBS L. The hemopoietic microenvironment of the bone marrow: An ultrastructural study of the stroma in rats. Anat Rec 1976: 186: 161-184. 54. ROBERTS RA, SPOONCER E, PARKINSON EK, et al. Metabolically inactive 3T3 cells can substitute for marrow stromal cells to promote the proliferation and development of multipotent haematopoietic stem cells. J Cell Physiol 1987: 132: 203-214. 55. GORDON MY, DOWDING R, RILEYGP, GREAVES MF. Characterization of stroma-dependent blast colony-forming cells in human bone marrow. J Cell Physiol 1987: 130: 150156. 56. COULOMBEL L, EAVESAC, EAVESCJ. Enzymatic treatment of long-term human marrow cultures reveals the preferential location of primitive hemopoietic progenitors in the adherent layer. Blood 1983: 62: 291-297. 57. MAYANIH, GUILBERTLJ, JANOWSKA-WIECZOREK A. Modulation of erythropoiesis and myelopoiesis by exogenous erythopoietin in human long-term marrow cultures. Exp Hematol 1990: 18: 174-179. 58. VERFAILLIE C, BLAKOLMER K, MCGLAVE P. Purified primitive human hematopoietic progenitor cells with long-term in vitrn repopulating capacity adhere selectively to irradiated bone marrow stroma. J Exp Med 1990: 172: 509-520. 59. STANLEY ER, GUILBERT LJ, TUSHINSKY RJ, BARTELMEZ S. CSF-1: A mononuclear phagocyte lineage specific hemopoietic growth factor. J Cell Biochem 1983: 21: 151-159. 60. MAYANIH, GUILBERTLJ, CLARK SC, JANOWSKAWIECZOREK A. Inhibition of hematopoiesis in normal human long-term marrow cultures treated with recombinant human macrophage colony-stimulating factor. Blood 199 1: 78: 65 1657. 61. MAYANI H, GUILBERTLJ, SYCH I, JANOWSKAWIECZOREK A. Production of tumor necrosis factor-cc in human long-term marrow cultures from normal subjects and patients with acute myelogenous leukemia: Effect of recombinant macrophage colony-stimulating factor. Leukemia 1992 (in press). 62. CASHMAN J, EAVESAC, EAVESCJ. Regulated proliferation
The hemopoietic microenvironment
63. 64.
65.
66.
67.
of primitive heniatopoietic progenitor cells in long-term human marrow cultures. Blood 1985: 66: 1002-1005. EAVESAC, EAVESCJ. Maintenance and proliferation control of primitive hemopoietic progenitors in long-term cultures of human marrow cells. Blood Cells 1988: 14: 355-368. CASHMAN J, EAVES AC, RAINESEW, et al. Mechanisms that regulate the cell cycle status of very primitive hematopoietic cells in long-term human marrow cultures. I. Stimulatory role of a variety of mesenchymal cell activators and inhibitory role of TGFP. Blood 1990: 75: 96-101. COUTINHO L, WILL A, RADFORDJ, et al. Effects of recombinant human granulocyte colony-stimulating factor (CS F), human granulocyte-macrophage-CSF, and gibbon interleukin-3 on hematopoiesis in human long-term marrow culture. Blood 1990: 75: 2 1 18-2 129. HOGGED. CASHMAN J, HUMPHRIES RK, et al. Differential and synergistic effects of human granulocyte-macrophage colony-stimulating factor and human granulocyte colonystimulating factor on hematopoiesis in human long-term marrow cultures. Blood 1991: 77: 493-499. OTSUKAT, THACKER D, HOGGED. The effects of interleukin 6 and interleukin 3 on early hematopoietic events in long-term cultures of human marrow. Exp Hematol 1991: 19: 1042-1048.
68. GORDONMY, RILEY GP, WATT SM, GREAVESMF. Compartmentalization of a haematopoietic growth factor (GM-CSF) by glycosaminoglycans in the bone marrow microenvironment. Nature 1987: 326: 403-405. 69. STEIN J, BORZILLO GV, RETTENMIER CW. Direct stimulation of cells expressing receptors for macrophage colony stimulating vactor (CSF-1) by a plasma membrane bound precursor of human CSF-1. Blood 1990: 76: 1308-1314. 70. FLANAGAN JG, LEDERP. The kit ligand: a cell surface molecule altered in steel mutant fibroblasts. Cell 1990: 63: 185- 194. 71. ERSHLERWB, ROSSJ, FINLAY J, et al. Bone marrow microenvironment defect in congenital hypoplastic anemia. N Engl J Med 1980: 302: 1321-1327. S, MOTOMURA S. IBAYASHI H. Differentiation of 72. KANEKO human bone marrow-derived fibroblastoid colony forming cells (CFU-F) and their roles in haemopoiesis in vitro. Br J Haematol 1982: 51: 217-225. 73. NAGAO T. YAMAUCHI K, KOMATSUDA M, et al. Inhibition of human bone marrow fibroblast colony formation by leukemic cells. Blood 1983: 62: 1261-1265. 74. WLODARSKI KH, JANOWSKA-WIECZOREK A. Precursor cells of fibroblast colony forming unit in the normal and pathologic human bone marrow and in the peripheral blood. Folia Haematol 1984: 5: 686-691. 75. HIRATA J, KATSUNO M, KANEKOS, et al. Clinical signlficance of human bone marrow stromal cell colonies in acute leukemias. Leuk Res 1986: 12: 1441-1445. 76. GREENBERG BR, WILSON FD, WOO L. Granulopoietic effects of human bone marrow fibroblastic cells and abnormalities in the granulopoietic microenvironment. Blood 198 1: 58: 557-556. 77. GREENBERC PL, MARAB, HELLERP. Marrow adherent cell colony-stimulating activity produciion in acute myeloid leukemia. Blood 1978: 52: 362-378. 78. MAYAN1 H, GUILBERT LJ, CLARK SC, BELCH AR, JANOWSKA-WIECZOREK A. Composition and functional in-
79. 80.
81.
82. 83.
84.
85.
86.
87.
88. 89.
90.
91. 92.
93.
tegrity of the in vitro hemopoietic microenvironment in acute myelogenous leukemia: Effect of macrophage colonystimulating Factor. Exp Hematol 1992: 20: 1077-1 084. LANGLEY GR, SMITHLJ, MCCULLOCH EA. Adherent cells in cultures in blast progenitors in acute myeloblastic leukemia. Leuk Res 1986: 10: 953-959. MIYAUCHI J, WANGC, KELLEHER CA, el al. The effects of recombinant CSF-1 on the blast cells of acute myeloblastic leukemia in suspension culture. J Cell Physiol 1988: 135: 55-62. TAKAHASHI M, KEATINCA, SINGER JW. A functional defect in irradiated adherent layers from chronic myelogenous leukemia long-term marrow cultures. Exp Hematol 1985: 13: 926-93 1. JUNEJA HS, GARDNER FH. Functionally abnormal marrow stromal cells in aplastic anemia. Exp Hematol 1985: 13: 194199. GORDONMY, GORDON-SMITH EC. Bone marrow fibroblastoid colony-forming cells (F-CFC) in aplastic anaemia: colony growth and stimulation of granulocyte-macrophage (GM-CFC). Br J Haematol 1980: 49: 465-477. MARSHJCW, CHANGJ, TESTANG, Hows JM, DEXTER TM. Zn vitro assessment of marrow “stem cell” and stromal cell function in aplastic anemia. Br J Haematol 1991: 78: 258-267. HOTTAT, MURATET, YAMAOH, YAMADA H, SAITOH. Granulopoiesis-supporting effects of marrow stromal cells in continuous culture from patients with primary myelofibrosis. Br J Haematol 1986: 64: 669-673. COUTINHO LH, GEARY CG, CHANGJ, et al. Functional studies of bone marrow hemopoietic and stromal cells in the myelodysplastic syndrome (MDS). Br J Haeinatol 1990: 75: 16-25. WATHENLM, KNAPPSA, DEGOWINRL. Suppression of marrow stromal cells and microenvironmental damage following sequential radiation and cyclophosphamide. 1nt J Radiat Oncol Biol Phys 1981: 7: 935-941. TESTANG, HENDRYJH, MOLINEUXG. Long-term bone marrow damage in experimental systems and in patients after radiation or chemotherapy. Anti-Cancer Res 1985: 5: 101. TESTANG, HENDRYJH, MOLINEUX G. Long-term bone marrow damage after cytotoxic treatment: Stem cells and NG, GALERP, eds. Hematomicroenvironment. In: TESTA poiesis: Long-term effects of chemotherapy and radiation. New York: Marcel Dekker Inc., 1988: 75-91. MOLINEUX G, TESTA NG, MASSA G, SCHOFIELD R. An analysis of hemopoietic and microenvironmental populations of mouse bone marrow after treatment with busulfan. Biomedicine 1986: 40: 2 15-220. ANDERSON RW, MATHEWS KI, CROUSEDA, SHARPJG. In vitroevaluation ofhematopoiesis in mice treated with busulphan or nitrogen mustard. Biomedicine 1982: 36: 149-152. TESTANG. BHAVNANI M, WILL A, M O R R I S JONESPH. Long-term bone marrow damage after treatment for acute lymphoblastic leukemia. In: TESTA NG, GALERP, eds. Hematopoiesis: Long-term effects of chemotherapy and radiation. New York: Marcel Dekker Inc., 1988: 279-287. KNOSPEWH. Long-term bone marrow damage after irradiation. In: TESTA NG, GALERP, eds. Hematopoiesis: Longterm effects of chemotherapy and radiation. New York: Marcel Dekker Inc. 1988: 93-130.
233
Introduction
Effective hemopoiesis is the result of the interplay between hemopoietic stem/progenitor cells and a supporting stroma localized in very specific organs (1-4). In normal adult mammals, more than 95 of the hemopoietic activity occurs in the bone marrow. In fact, this is the only site in which myelopoiesis, erythropoiesis, and lymphopoiesis proceed simultaneously. This has led to the conclusion that local tissue influences critical for hemopoiesis operate primarily in the medullary cavity ( 5 ) . Under certain conditions, the spleen can also act as a hemopoietic organ (l), indicating that suitable conditions for hemopoiesis are also present in it. During the last three decades, extensive studies, both in vivo and in vitro, have been focused on the structural and physiological characterization of the environment in which hemopoiesis takes place. With the development of experimental systems for the inv i m culture of stromal cells (4, 6), and the cloning of the genes for several hemopoietic cytokines (7-9) significant advances in understanding the way in which the hemopoietic microenvironment regulates blood cell production have been made. The present article intends to present an overview on the struc-
Hector Mayani, Larry J. Guilbert and Anna Janowska-Wieczorek Departments of Medicine and Immunology, University of Alberta and Canadian Red Cross Blood Center, Edmonton, Alberta, Canada
Key words: bone marrow - cytokines extracellular matrix - hemopoiesis - stroma
Correspondence: Anna Janowska-Wieczorek M.D., Ph.D., Div. Clinical Hematology, Department of Medicine, University of Alberta, 8249-1 14 Street, Canada T6G 2R8. Phone: (403)431-0202; FAX: (403)43 1-046 1. Accepted for publication 28 August 1992
ture and composition of the hemopoietic microenvironment and the different ways in which it regulates blood cell production. Particular emphasis is placed on the hemopoietic microenvironment in humans under both normal and pathological conditions. Evidence for a hemopoietic microenvironment
The first evidence for the presence of specific conditions, in the bone marrow and spleen, necessary for hemopoietic development came from experiments in which mouse marrow cells were injected into syngeneic, irradiated mice. The first capillary bed those cells encounter is in the lungs, but the lungs do not become hemopoietic, nor d o the skeletal muscles or other body parts to which those injected cells that escape the lungs may be shunted next. Hemopoiesis is established exclusively in the bone marrow and, to a lesser extent, in the spleen (in the form of macroscopic colonies (10). Later on, it was observed that the hemopoietic colonies developed in the spleen were primarily erythroid, whereas those developed in the marrow were mainly granulocytic (1 1, 12). Furthermore, it was also shown that, among the colonies developed in the spleen, those 225
Mayani et al. developed in the red pulp were almost exclusively erythroid, whereas the granulocytic colonies developed under the splenic capsule and the trabeculae
MEDULLARY CAVITY
(1 3).
Microgeographical regions have also been described in the bone marrow. Lord and colleages have shown that the spatial distribution of stem cells, as well as of erythroid and myeloid committed progenitors in the mouse bone marrow, is not random (14, 15). They suggest that the hernopoietic tissue possesses a well-defined structure, although not evident in histological sections, that is relevant to the proliferation and differentiation of hernopoietic cells. Based on both in vivo and in vitro evidence, Schofield has proposed the existence of a stem cell “niche”, considered to be a functional compartment of marrow strorna that isolates conditions for self replication from conditions conducive to differentiation (16). ACCESSORY CELLS
Definition of the hemopoietic microenvironment
The marrow stromal tissue is a network of cells (fibroblast, endothelial cells, macrophages, adipocytes) and extracellular matrix (collagen, laminin, fibronectin, proteoglycans) which physically supports the hemopoietic cells and influences their proliferation and differentiation (17). Stromal cells have been considered to be organized into hemopoietic microenvironments that support blood cell development. The term hemopoietic microenvironment was created to distinguish different microgeographic stroma1 cell influences within an organ on the pluripotent stem cells, as opposed to whole organ differences, each of which is composed of the sum of its stromal microenvironments ( 18). The hemopoietic microenvironment, however, is not easy to define. If it is taken to be a broad term describing those cells that regulate hemopoietic development, other cells in addition to the stromal cells may be components of it. This is, in fact, the case for T lymphocytes, which are known to regulate hemopoiesis by secreting multiple hemopoietic cytokines (19). Thus, using this definition, stromal cells contribute to the hemopoietic microenvironment, but not all microenvironment cells are strornal cells. In the present article, we define the hemopoietic microenvironment as the local network of stromal cells (jibroblast, macrophages, endothelial cells, udipocytes), accessory cells (T lymphocytes and monocytes) and their products (extracellularmatrix and cytokines), capable qf injuencing the self renewal, proliferation and diferentiation of hemopoietic stem/ progenitor cells (see Fig. 1). The rest of this review will be focused to the structure and physiology of the hemopoietic microenvironment of the bone marrow. 226
v;(
Fig. 1. Schcmatic representation of the different components of the hemopoietic microenvironment and the way they influence the proliferation and differentiation of hemopoielic stem/progenitor cells (HPC). Stromal cells (fibroblasts, macrophages, endothelial cells, adipocytes) can interact with HPC via direct cell-to-cell contact (a). They can also produce proteins that constitute the extracellular matrix (b), which, in turn, interact with HPC (c). On the other hand, stromal cells are also a major source of cytokines (d), which may remain as cell-associated molecules and be presented to the H P C during the direct interaction between stromal cells and HPC, or may be secreted and act on their target cclls (e). Cytokines can also be retained by the extracellular matrix and be presented to the HPC in that way (f). Accessory cells (T lymphocytes and monocytcs) influence hemopoiesis by the production of cytokines (g). Finally, it is important 10 note that stroma1 and accessory cells not only produce cytokines, but are also regulated by them (hj).
Composition of the hemopoietic microenvironment in vivo
Reticular cells are the major cellular components of the bone marrow stroma (20) and there are two types, adventitial and fibroblastic. Adventitial reticular cells are located around venous sinuses forming a layer that partially covers the endothelium’s abluminal side. Morphological studies indicate that one of their major roles is the regulation of the migration of mature blood cells from the marrow to the circulation (21). Fibroblastic reticular cells, on the other hand, are located within the marrow hemopoietic cord; their thin cytoplasmic processes envelop maturing hemopoietic cells in the marrow compartment (22). These fibroblastic cells are usually in close contact with immature granulocytic cells, which suggests that there are functional interactions between the two cell types (23,24). Macrophages are the second major cellular component of the marrow stroma (2 1). Topographically,
The hemopoietic microenvironment they are found in three different sites in the marrow: as central macrophages in the erythroblastic islands, perisinal macrophages on the abluminal side of the sinus endothelium, and also dispersed between the developing hemopoietic cells. The central macrophage is usually surrounded by maturing red cells (erythroblasts), thus forming an anatomic unit commonly referred to as an erythroblastic island (25). It has been suggested that the central macrophage plays a key role in erythroid maturation. The possibility exists that macrophages of the erythroblastic island produce or process factors that are then transferred directly to the erythroblasts through the membrane channels of communication (22). Hypoxia, which stimulates red cell production, also leads to an increase in the central macrophage’s size and its phagocytic activity (2 1). Perisinal macrophages cover varying proportions of the sinus endothelium’s abluminal surface. Their cytoplasmic protrusions penetrate the endothelium, engulf senescent or defective red cells, and remove them from the circulation (21). Endothelial cells and udipocytes are also major components of the marrow stroma (21). Endothelial cells are the primary cell type forming the blood vessels of the marrow. Adipocytes, on the other hand, occupy the majority of the marrow cavity in adult mammals (26). Their role in hemopoiesis is not clear, although in vitro studies suggest that they are capable of producing hemopoietic growth factors (27). Marrow stromal cells produce and secrete macromolecules that form a highly organized structure known as the extrucellulur matrix (28-30). The major components of the extracellular matrix are (i) collagens, fibrillar proteins produced by reticular and endothelial cells, (ii) fibronectin, a glycoprotein produced by reticular cells and macrophages, (iii) laminin, another large glycoprotein produced by endothelial cells, and (iv) proteoglycans, composed of a core protein to which one or more glycosaminoglycans are attached. In vitro studies indicate that the extracellular matrix of the bone marrow not only acts as a physical support for the hernopoietic elements, but also plays a dynamic role in the regulation of hemopoiesis (30). The vast majority of the medullary cavity i y occupied by maturing granulocytic cells which, upon /zzchzhg their final stage, will be released into the circulation in a process mediated by endothelial and reticular cells (21). T o date, however, the role of this granulocytic pool in blood cell production is not clear. Some have suggested that the size of the granulocytic pool regulates the production of granulocyte progenitors in an inversely proportional way; that is to say, when the granulocytic pool is low, production of granulocyte progenitors is enhanced, and vice ver-
sa (1). This may suggest that such a granulocyte pool could also be considered as part of the hemopoietic microenvironment. To date, however, there is no clear evidence that maturing granulocytes are directly implicated in the regulation of progenitor cell proliferation. Composition of the hemopoietic microenvironment in vitro
Among the different in vitro models that have been developed to study the hemopoietic microenvironment, the long-term marrow culture (LTMC) system seems to be the one that most closely reproduces the conditions observed in vivo. Both murine and human hemopoiesis can be sustained for several weeks in LTMC (31, 32). This is due to the development of an adherent cell layer, consisting of different cell types, which provides the physical support and the hemopoietic cytokines (soluble and cell-associated proteins) necessary for the growth of hemopoietic progenitors (4). It is noteworthy that the composition of the adherent cell layer of LTMC is very similar to the bone marrow stroma. The major components of such an adhering layer are fibroblasts that are capable of producing collagen types I and I l l as well as fibronectin (33). These cells are the most numerous, comprising 45-55% of the total cell number (34). Macrophages are also present in significant levels, comprising 25-352) of the adherent cells (34). Endothelial cells, on the other hand, comprise 5-202) of the total cell number (35). Adipocytes are usually observed in both murine and human LTMC, although their presence may depend on certain culture conditions (4). Lymphocytic as well as myeloid cells have also been observed in the adhering layers of LTMC, although at reduced levels (34). Like its in vivo counterpart, the extracellular matrix developed in LTMC consists of collagen I, 111, and IV, laminin, fibronectin and proteoglycans (33, 36,37). Recent studies by Coulombel and colleagues have shown that erythroid progenitors adhere in vitro to extracellular matrices derived from marrow fibroblasts and that such an interaction is mediated by fibronectin (38). The same group of investigators has also demonstrated that terminal erythroid differentiation results in loss of attachment to fibronectin, providing a potential mechanism for the release of mature red blood cells from the bone marrow into the circulation (39). More recent studies by Koenigsmann et al. (40) have shown that myeloid, as well as erythroid progenitor cells have the ability to bind to collagen I, suggesting that this protein may play a role in the localization of committed myeloid and erythroid progenitors within the bone marrow. 221
Mayani et al. Origin of the microenvironment cells
The origin of the cells of the hemopoietic microenvironment has been a subject of great controversy during the last decade. As demonstrated by several investigators, bone marrow macrophages as well as circulating monocytes and T lymphocytes, are derived from the pluripotent hemopoietic stem cell (41). The origin of the stromal cells, on the other hand, is not so clear. Among the so-called “non-hemopoietic” cells of the marrow microenvironment, fibroblasts are the ones that have been best characterized in terms of their origin; thus, we will focus this subsection on these cells. Since the early work of Friedenstein and colleagues (42), and years later by Castro-Malaspina and collaborators (43), it became clear that fibroblasts derive from adherent, nonphagocytic progenitors capable of forming fibroblastic colonies in v i m (colony-forming unit-fibroblast, or CFU-F). By using a complement-mediated cytotoxicity assay, it was shown that CFU-F are different from hemopoietic progenitors (CFU-C), which may also imply different origins (43). Very recently, Simmons and TorokStorb demonstrated that CFU-F express a cellsurface antigen, absent on any other hemopoietic progenitor, recognized by the monoclonal antibody STRO-1 (44). Interestingly, cells giving rise to adipocytes and endothelial cells were also positive for STRO-1 (44). The same investigators have performed extensive phenotypical studies on CFU-F and found that, although these cells share some characteristics with the hemopoietic progenitors such as the expression of the CD34 antigen, they are distinct cell types (45). In vitro studies in human LTMC have suggested the existence of a pluripotent cell in the bone marrow that gives rise to hemopoietic cells and to their stromal microenvironment (46). Such an idea has been supported by in vitro studies in rats showing that the marrow stroma grown in culture contains cells with hemopoietic potential (47). These reports seem to give support to the study of Keating and colleagues, who found that the human hemopoietic microenvironment developed in vitro after bone marrow transplantation is of donor origin (48). However, more recent studies have indicated that the human in vitro microenvironment after bone marrow transplantation is, with the exception of macrophages, of host origin (49-5 I), suggesting independent origins for hemopoietic and stromal cells. The apparent discrepancies between the above studies may be due to differences in the techniques used to assess cultured cells and the type of cells analyzed. For instance, Laver and colleagues (50) used chromosomal analysis, which determines the origin of stroma1 cells capable of undergoing mitosis, whereas 228
Keating and colleagues (48), by determining the presence of the Y-body chromosome, analyzed nondividing cells. Thus, whether hemopoietic and stromal (fibroblastic) cells derive from a common or from independent stem cells is still an open question. Interaction between hemopoietic cells and their microenvironment
Microenvironment cells can influence hemopoiesis, either in a positive or in a negative way, by different mechanisms (Fig. 1). These include (i) direct cell-tocell contact, (ii) the secretion of proteins that constitute the extracellular matrix, and (iii) the production of soluble and cell-associated cytokines (52). During the last few years, it has become evident that more than one of these mechanisms can operate simultaneously in the interaction between microenvironmental and hemopoietic cells. Direct cell-to-cell contact between stromal and hemopoietic cells has been demonstrated, both in vivo and in vitro, by morphological studies using electron microscopy (53). These studies have shown the preferential association of fibroblasts with granulocytic cells, and macrophages with erythroid cells. This had led to the suggestion that those interactions are necessary to the optimal development of hemopoietic cells. Functional studies have indicated that murine hemopoietic stem cells (CFU-S) need to interact physically with stromal cells in order to grow in LTMC (54). Similar requirements for cell-to-cell contact have been reported for the growth of human primitive hemopoietic cells (CFU-Blast) in v i m (55). In keeping with those observations is the fact that, when human bone marrow cells are cultured in LTMC, the great majority of the hemopoietic progenitors are located in the adherent layer that develops after 3 weeks (56, 57), which supports the idea that they interact with their surrounding stroma. More recently, long-term culture-initiating cells (LTC-IC), the most suitable candidate for human hemopoietic stem cells described to date, have also been shown to be preferentially located in association with adherent stromal cells (58). Microenvironment cells are a major source of hemopoietic cytokines. They produce hemopoietic stimulators such as GM-CSF, G-CSF and M-CSF, and inhibitors, such as TNF,, TGFP, and IFN, (52). Some of these cytokines are produced only by activated microenvironment cells (T lymphocytes and macrophages); thus, they have been postulated to play a major role on blood cell production under situations of acute stress, such as infection (52). Others are constitutively produced by stromal cells. Indeed, biologically active M-CSF and G-CSF have been found in the supernatant of human LTMC at
The hemopoietic microenvironment IL-1 EZSF IL-6
--\-,
IFNT PDGF TGFp TNFlY
IL-1 SCF IL-6 IFNp
k::
IL-8 M-CSF G-CSF GM-CSF
,-,
IL-1 IL-4 IL-8 TGFP TNFa M-CSF
IL-1 PDGF
M-CSF
IL-1- IL-10 GM-CSF G-CSF IFNa IFN4 IFNT TG FB TNFa
-b
T
@--b
IL-I - IL-6 IL-9 IL-10 $$;CSF LIF TNFa -. . _
Fig. 2. Microenvironmental cells not only produce hemopoietic cytokines, but are also regulated by them. The figure shows some of the cytokines that are produced by, and act on, fibroblasts (F), macrophages (M), and T lymphocytes (T). IL, interleukin; CSF, colony-stimulating factor; G, granulocyte; M, macrophage, GM, granulocyte-macrophage; PDGF, platelet-derived growth factor; TGFP, transforming growth factor$; TNF,, tumor necrosis factor-a, SCF, stem cells factor (c-kit ligand); IFN, interferon; LIF, leukemia inhibitory factor.
levels similar to those found in vivo (32). Due to the fact that microenvironmental cells not only produce hemopoietic cytokines, but are also regulated by them (Fig. 2), the interactions between these cells, cytokines and hemopoietic cells are rather complex. In order to illustrate this point, we will present some examples. M-CSF is a lineage-specific growth factor, produced by microenvironmental cells, that stimulates the proliferation, maturation and function of monocytic cells (7-9, 59). When M-CSF is added to semisolid cultures of human bone marrow cells, it acts directly on macrophagic progenitors, stimulating their growth. Interestingly, when recombinant human M-CSF is added to human LTMC at a concentration 3- to 4-fold higher than the endogenous concentration, hemopoiesis is inhibited. This is due to the production by macrophages of a soluble activity, that contains TNF,, capable of inhibiting the growth of multipotential, myeloid and erythroid progenitors (60,61). Thus, M-CSF acts both as a positive (when acting directly on progenitor cells) and negative (when interacting with stromal cells) regulator of hemopoiesis.
Studies by Cashman and colleagues have shown that immature erythroid progenitors present in the nonadherent fraction of human LTMC are actively cycling at all times. In contrast, the same cell population is in a quiescent state when localized in the adherent layer (62). These authors have presented evidence indicating that stromal cells keep closely located hemopoietic progenitors out of cycle (Gc,)by secreting TGFP (63, 64). On the other hand, cytokines such as IL-1 or P D G F stimulate stromal cells to produce stimulatory cytokines, such as GM-CSF, and these, in turn, stimulate the hemopoietic progenitors to proliferate (63,64). Thus, the balance between the levels of hemopoietic stimulators and inhibitors seems to be of critical importance in the regulation of hemopoietic progenitor cell proliferation. The effects of GM-CSF, G-CSF, and IL-3 in human LTMC provide another example for the role of the hemopoietic microenvironment in the regulation of hemopoiesis in vitro. Addition of these factors into human LTMC results in increased levels of progenitor cells (65-67). However, when these factors are constitutively produced by genetically engineered fibroblasts at concentrations similar to those observed when the recombinant proteins were added to the cultures, their effects are significantly enhanced (66,67). These observations suggest that the way in which the hemopoietic stimulators are presented to the progenitor cells may determine the magnitude of the biological response. Interestingly, recent reports indicate that factors like GM-CSF are produced by stromal cells and then selectively retained by the glycosaminoglicans of the extracellular matrix (36, 68), thus suggesting that the extracellular matrix may play a key role in the presentation of cytokines to the hemopoietic cells. Hemopoietic cytokines may also be presented to hemopoietic cells during the direct contact with stromal cells. Indeed, cellassociated forms of factors like M-CSF and Stem Cell Factor (c-kit ligand) have been described (69, 70). The actual mechanisms by which microenvironmental cells regulate hemopoietic stem/progenitor cell proliferation and differentiation in vivo are not known in detail. It is thought that they are similar to those observed in in vitro systems (e.g. LTMC); however, it must be kept in mind that the in vivo situation is much more complex than the most complex in vitro model, hence, the development of in vivo models in which the interaction between hemopoietic and microenvironmental cells can be studied is absolutely necessary. In this regard, the SCID mouse model may be of considerable relevance.
229
Mayani et al. The hemopoietic microenvironment in hematological diseases
The fact that microenvironment cells regulate blood cell production suggests that alterations within the hemopoietic microenvironment may be involved in the development and/or progression of hematological diseases. Indeed, several in vitro studies have demonstrated functional abnormalities of the hemopoietic microenvironment in certain hemopathies. Ershler and colleagues reported the case of a woman with congenital hypoplastic anemia (Diamond-Blackfan syndrome) with marked erythroid hypoplasia. When her bone marrow cells, obtained by aspiration, were cultured in methylcellulose, an exuberant erythroid growth was observed. The erythroid nature of the colonies developed in methylcellulose was confirmed through quantitation of heme synthesis and globin mRNA accumulation in a liquid culture system. In contrast, when whole bone marrow fragments were similarly cultured, no appreciable hemoglobin synthesis was observed. Their results suggested that, in that particular patient, hypoplastic anemia resulted from an unfavorable hemopoietic microenvironment (7 1). Different investigators have observed reduced numbers of fibroblastic progenitors (CFU-F) in cultures of bone marrow from patients with acute myelogenous leukemia (AML) (72-75). Interestingly, the results of Nagao et al. suggested that the suppression of CFU-F formation in leukemic cultures was through humoral factors secreted by leukemic cells (73), implying that the reduced levels of C F U - F are the result, and not the cause, of the disease. On the other hand, a functional defect in AML-derived fibroblastic cells has been observed by Greenberg and colleagues (76). They found that, in contrast to normal marrow-derived fibroblasts, marrow fibroblasts obtained from untreated AML patients failed to enhance granulopoiesis in culture (76), thus suggesting that not only quantitative but also qualitative abnormalities may exist in AML-derived fibroblasts. Functional deficiencies in AML-derived bone marrow macrophages have also been documented. Greenberg and collaborators observed a reduced production of colony-stimulating activity by macrophages from AML patients as compared to macrophages from normal subjects (77). More recently, Mayani et al. have reported functional differences in LTMC between macrophages derived from normal and AML bone marrow: normal marrow-derived stromal adherent layers developed in LTMC showed a much better capacity to sustain normal hemopoiesis in chimeric LTMC than adherent layers derived from AML bone marrow (78). This correlated with the production by AML stromal cells of a soluble activity capable of inhibiting the growth of hemopoietic progenitors. Further studies by the same group
230
demonstrated that such an activity, which is tnainly produced by AML-derived macrophages, is in fact TNF, (61). Normal bone marrow macrophages do not seem to produce T N F a unless stimulated by CSF-1. Interestingly, CSF-1 downregulates the production of TNF, by AML-derived macrophages (6 1). The authors have suggested that macrophages developed in LTMC from AML bone marrow are functionally abnormal. This idea is supported by the fact that macrophages developed in vitro from AML peripheral blood or bone marrow have been shown to be derived from AML blasts (79, 80). l n vitro studies using two-stage LTMC have provided evidence that stromal adherent layers established from patients with chronic myelogenous leukemia (CML) show a functional defect; that is to say, they were not able to sustain the growth of hemopoietic progenitors derived from normal bone marrow (8 1). Functional abnormalities (i.e. reduced production of colony-stimulating activity) in bone marrow stromal cells from patients with aplastic anemia have also been reported (82). In contrast to the reports mentioned above, other studies indicate that the in v i m developed hemopoietic microenvironment from patients with hematological disorders such as aplastic anemia (83, 84), primary myelofibrosis ( 8 5 ) , and myelodysplasia (86) is functionally normal, suggesting that the origin and development of such diseases are exclusively due to alterations within hemopoietic cells. It is possible that the apparently contradictory results that have been reported, particularly regarding the functional integrity of the hemopoietic microenvironment in aplastic anemia, are due to differences in the experimental models that have been used. Thus, whether the hemopoietic microenvironment is involved in the development and/or progression of certain hemopathies is still an open issue. Effects of chemotherapy and radiation on the hemopoietic microenvironment
Chemotherapy and radiation are currently used to treat different forms of cancer, including leukemia. During the last decade, however, several studies have been reported indicating that both chemical agents and radiation can alter the functional integrity of the marrow microenvironment. This has been documented in murine as well as in human systems (8789). The numbers of fibroblast progenitors in mice treated with busulfan or cyclophosphamide have been found to be significantly lower than the numbers observed in control animals (90). On the other hand, studies using the LTMC system have shown that adherent layers from normal donor mice were able to support comparable levels of hemopoiesis
The hemopoietic microenvironment
(80% of the adequate control) when inoculated with bone marrow cells from normal mice. In contrast, when stroma from busulfan-treated marrow was inoculated with marrow cells from normal mice the levels of hemopoietic progenitors were only 20-40% of the levels of hemopoietic progenitors seen in cultures containing normal stroma (9 l). Similar results were observed in studies of patients in remission from acute lymphoblastic leukemia (ALL) after being treated with chemo- and radiotherapy. Testa and colleagues have shown that adherent layers froin such patients were able to support only 50% of hemopoietic growth as compared to adherent layers from normal subjects (92). Radiation has also been found to induce significant damage to the marrow stroma, especially at doses above 10 Gy (93). Some studies indicate that stromal cells are sensitive to doses as low as 2.55 Gy. However, injury at this level of exposure can be detected only in in vitro systems, since stromal cells in vivo have a large capacity for repair. Single exposure in vivo to doses of 40 Gy results in irreparable injury to the stroma and to permanent aplasia (93). At present, several approaches are being explored in order to define new strategies for the treatment of neoplasms without altering the hemopoietic microenvironment. In this context, the use of specific cytokine combinations in conjunction with mild doses of chemo- and/or radiotherapy may prove useful. Concluding remarks
Understanding the biology of the hemopoietic microenvironment is of great relevance in hematology since the optimal production of blood cells depends to a large extent on the interaction between hemopoietic stem/progenitor cells and their surrounding environment. During the last two decades, major advances have been made in the structural and physiological characterization of the hemopoietic microenvironment. However, two major questions remain unanswered. Firstly, are stromal (fibroblastic) and hemopoietic cells derived from a common progenitor? Studies supporting and rejecting such a possibility have been reported and no definite answer has yet been found. Secondly, although it seems clear that functional abnormalities in the hemopoietic microenvironment exist in certain hematological diseases, it is not known to what extent such abnormalities contribute to the development and/or progression of the disease. There is no doubt that in the next few years the answers to these and other important questions will be obtained. This, in turn, will have a major impact in understanding the basic biology of hemopoiesis and the way in which certain hemopathies develop.
Acknowledgements Research in the authors’ laboratories is funded by grants from the Alberta Cancer Board (AJW) and the National Cancer Institute of Canada (LJG). Hector Mayani was supported by 21 scholarship from the National University of Mexico.
References 1. METCALFD, MOOREMAS. Haemopoietic cells. Amstcrdam: North-Holland Publishing Co, 1971. 2. WOLF NS. The hemopoietic microenvironmcnL. Clin Hacmatol 1979: 8: 469-500. 3. DORSHKIND K. Regulation of hemopoicsis by bone marrow stromal cells and their products. Annu Rev Ininiunol 1990: 8: 111-137. 4. GREENBERCER JS. The hemopoietic microenvironment. Crit Rev Oncol Hernatol 1991: 11: 65-84. 5 . TAVASSOLI M, FRIEDENSTEIN A. Hemopoietic stromal niicroenvironment. Am J Hematol 1983: 15: 196-203. 6. DEXTER TM. SPOONCER E, SIMMONS P. ALLENTD. Longterm marrow culture: An overview of techniques and experience. New York: Alan R Liss Inc. 1984: p. 57-96. 7. SIEFFCA. Hematopoietic growth factors. J Clin Invest 1987: 79: 1549-1557. 8. CLARKSC, KAMEN K. The human hcinopoietic colonystimulating factors. Science 1987: 236: 1229- 1237. U, LANTZM, RICHTERJ. The re9. OLSSONI, GULLBERG ceptors for regulatory molecules of hemopoiesis. Eur J Haematol 1992: 48: 1-9. 10. TILLJE, MCCULLOCHEA. A direct measurement of the radiation sensitivity of normal niouse bone marrow cells. Rad Res 1961: 14: 213-222. 1 I . WOLF NS, TRENTIN J. Effects of hemopoietic organ stroma on differentiation of pluripotent stem cells. J Exp Mcd 1968: 127: 205-214. J. Determination of bone marrow stem cell differ12. TRENTIN entiation by stromal hemopoietic inductive microenvironments (HIM). Am J Pathol 1971: 65: 621-628. 13. TRENTIN J . Influence of hemopoietic organ stroina on stem cell differentiation. In: GORDON AS, ed. Regulation of hemopoiesis. New York: Meredith, 1970: 16 1- 186. 14. LORD BI, TESTANG, HENDRYJH. The relative spatial distribution of CFU-S and CFU-C in the mouse femur. Blood 1915: 46: 65-72. 15. FRASSONI F, TESTANG, LORD €31. The relative spatial distribution of erythroid progenitor cells (BFUe and CFUe) in the normal mouse femur. Cell Tissue Kinet 1982: 15: 447455. 16. SCHOFIELDR. The relationship between the splccn colonyforming cells and the haemopoietic stem cell. Blood Cells 1918: 4: 7-25. 17. OWENM. Marrow stromal stem cells. J Cell Sci 1988: suppl 10: 63-76. 18. TRENTIN J. Hemopoietic microenvironments. In: TAVASSOLI M, ed. Handbook of the hemopoietic microenvironment. Humana Press, Inc. 1989: 1-86. 19. GILLISS. T cell-derived lymphokines. In: PAULLWE, ed. Fundamental Immunology. New York: Ravcn Press, 1989: 621-637. 20. TAVASSOLI M, YOFFEYJ M . Bone niarrow: Structure and function. New York: Alan R Liss, Inc., 1983: 47-64. M. Cellular components of stroma in v i w in com21. SHAKLAI parison with in vitro systems. In: TAVASSOLI M, ed. Handbook of the hemopoietic microenvironment. Huinana Press Inc. 1989: 219-251. 22. SHAKLAI M, TAVASSOLI M. Cellular relationship in the rat bone marrow studied by freeze-fracture and lanthanum i n -
23 1
Mayani et al.
23.
24. 25. 26. 27.
28. 29. 30.
31. 32. 33. 34. 35.
36. 37.
38.
pregnation thin sectioning electron microscopy. J Ultrastruc R ~ 1979: s 69: 343-361. LAPUSHIN KW. T R E N T IJ.NIdentification of stromal elements in erythroid and neutrophil granuloid spleen colonies: Light and electron microscopy study. Exp Haematol 1977: 5: 505-522. WESTEN H, BAINTON DF. Association of alkalinephosphate-positive reticulum cells in bone marrow with granulocytic precursors. J Exp Med 1979: 150: 919-937. BESSISM, BRETON-GORIUS J. Iron metabolism in the bone marrow as seen by electron microscopy. Blood 1962: 19: 635-663. TAVASSOLI M. Marrow adipose cells and hemopoiesis. An interpretative review. Exp Hcmatol 1984: 12: 139. LANOTTEM, METCALFD, DEXTERTM. Production of monocyte/macrophage colony-stimulating factor by preadypocyte cell lines derived from murine marrow stroma. J Cell Physiol 1982: 112: 123-127. URlsT M. Bone: Formation by autoinduction. Science 1965: 150: 893-899. REDDIA, HUGGINS C. Biochemical sequences in the transformation of normal fibroblasts in adolescent rats. Proc Natl Acad Sci 1972: 69: 1601-1605. ZUCKERMAN KS, PRINCECW, GAYS. The hemopoietic extracellular matrix. In: TAVASSOLI M, ed. The handbook of the hemopoietic microenvironment. Humana Press Inc 1989: 399-432. DEXTERTM, ALLENTD, LAJTHAL. Conditions controlling the proliferation of hemopoietic stem cells in vitro. J Cell Physiol 1977: 91: 335-344. JANOWSKA-WIECZOREK A, TUPAS J, FRIESENI, et al. Kinetics of G-CSF and CSF-I production in human long-term marrow cultures. Blood 1988: 72 suppl 1: 122a. BENTLEYSA. Collagen synthesis by bone marrow stromal cells: A quantitative study. Br J Haematol 1982: S O : 491-497. ANDREONI C, MOREAUI, RIGALD. Long-term culture of human bone marrow. I. Characterization of adherent cells in flow cytometry. Exp Hematol 1990: 18: 431-437. BENTLEY SA. The role and composition of the adherent layer in long-term bone marrow culture. In: WRIGHTD G & GREENBERGER J, eds. Long-term bone marrow culture. Alan R liss Inc, New York: 1984: 141-156. KEATINGA, GORDONMY. Hierarchical organization of hemopoietic microenvironments. Leukemia 1988: 2: 766-769. WIGHTTN, KINSELLAMG, KEATINGA, SINGERJW. Proteoglycans in human long-term marrow cultures: Biochemical and ultrastructural analyses. Blood 1986: 67: 13331343. COULOMBEL L, VUILLETM, LEROYC, TCHERNIA G. Lineage and stage specific adhesion of human hemopoictic progenitor cells to extracellular matrix from marrow fibroblasts. Blood 1988: 71: 329-334.
39. VUILLET-GAUGLER MH. BRETON-GORIUS J, VAINCHENK E R W, el a[. Loss of attachment of fibronectin with terminal human erythroid differentiation. Blood 1990: 76: 865873. 40. KOENIGSMAN M, GRIFFIN JD, DICARLO J, CANNISTRA SA. Myeloid and erythroid progenitor cells from normal bone marrow adhere to collagen type I. Blood 1992: 79: 657-665. 41. SPANGRUDE GJ. Hematopoietic stem-cell differentiation. Curr Op Immunol 1991: 3: 171-178. AJ, CHAILAKHYAN KK, LALYKINA KS. 42. FRIEDENSTEIN The development of fibroblast colonies in monolayer cultures of guinea pig bone marrow and spleen colonies. Cell Tissue Kinet 1970: 3: 393-403. 43. CASTRO-MALASPINA H, GAY RE, RESNICK G, et al. Characterization of human bone marrow fibroblast colony-forming cells (CFU-F) and their progeny. Blood 1980: 56: 289-301. 44. SIMMONS P, TOROK-STORB B. Identification of stromal cell
232
45. 46.
47.
48.
49.
precursors in human bone marrow by a novel monoclonal antibody. STRO-I. Blood 1991: 78: 55-62. SIMMONS P, TOROK-STORB B. CD34 expression by stronial precursors in normal human adult bone marrow. Blood 1991: 78: 2848-2853. SINGER JW, KEATINC A, CUTTNER J. et d.Evidence for a stem cell common LO hematopoiesis and its in vitro microenvironment: Studies of patients with clonal hemopoietic neoplasia. Leuk Res 1984: 8: 535-545. ISLAM A, GONGJK, HENDERSON ES. Direct evidence for a stem cell common to hemopoiesis and its in vitro microenvironment: Studies on syngeneic (inbred) wistar furth rats. J Med 1988: 19: 119-136. KEATING A, SINGERJ, KILLENP, et al. Evidence for donor origin of the in vitro hematopoietic microenvironment following marrow transplantation in man. Nature 1982: 298: 280283. SIMMONS P, PRZEPIORKA D, THOMAS ED, ct al. Host origin of marrow stromal cells following allogeneic bone marrow transplantation. Nature 1987: 328: 429-432.
50. LAVERJ, JHANWAR SC, O'REILLY,CASTRO-MALASPINA H. Host origin of human hemopoietic microenvironment following allogeneic bone marrow transplantation. Blood 1987: 70: 1966-1968. 51. AGEMATSUK, NAKAHORIY. Recipient origin of bone marrow-derived fibroblastic stromal cells during all periods following bone marrow transplantation in humans. Br J Haematol 1991: 79: 359-365. B. Cellular interactions. Blood 1988: 72: 37352. TOROK-STORB 385. 53. WEBS L. The hemopoietic microenvironment of the bone marrow: An ultrastructural study of the stroma in rats. Anat Rec 1976: 186: 161-184. 54. ROBERTS RA, SPOONCER E, PARKINSON EK, et al. Metabolically inactive 3T3 cells can substitute for marrow stromal cells to promote the proliferation and development of multipotent haematopoietic stem cells. J Cell Physiol 1987: 132: 203-214. 55. GORDON MY, DOWDING R, RILEYGP, GREAVES MF. Characterization of stroma-dependent blast colony-forming cells in human bone marrow. J Cell Physiol 1987: 130: 150156. 56. COULOMBEL L, EAVESAC, EAVESCJ. Enzymatic treatment of long-term human marrow cultures reveals the preferential location of primitive hemopoietic progenitors in the adherent layer. Blood 1983: 62: 291-297. 57. MAYANIH, GUILBERTLJ, JANOWSKA-WIECZOREK A. Modulation of erythropoiesis and myelopoiesis by exogenous erythopoietin in human long-term marrow cultures. Exp Hematol 1990: 18: 174-179. 58. VERFAILLIE C, BLAKOLMER K, MCGLAVE P. Purified primitive human hematopoietic progenitor cells with long-term in vitrn repopulating capacity adhere selectively to irradiated bone marrow stroma. J Exp Med 1990: 172: 509-520. 59. STANLEY ER, GUILBERT LJ, TUSHINSKY RJ, BARTELMEZ S. CSF-1: A mononuclear phagocyte lineage specific hemopoietic growth factor. J Cell Biochem 1983: 21: 151-159. 60. MAYANIH, GUILBERTLJ, CLARK SC, JANOWSKAWIECZOREK A. Inhibition of hematopoiesis in normal human long-term marrow cultures treated with recombinant human macrophage colony-stimulating factor. Blood 199 1: 78: 65 1657. 61. MAYANI H, GUILBERTLJ, SYCH I, JANOWSKAWIECZOREK A. Production of tumor necrosis factor-cc in human long-term marrow cultures from normal subjects and patients with acute myelogenous leukemia: Effect of recombinant macrophage colony-stimulating factor. Leukemia 1992 (in press). 62. CASHMAN J, EAVESAC, EAVESCJ. Regulated proliferation
The hemopoietic microenvironment
63. 64.
65.
66.
67.
of primitive heniatopoietic progenitor cells in long-term human marrow cultures. Blood 1985: 66: 1002-1005. EAVESAC, EAVESCJ. Maintenance and proliferation control of primitive hemopoietic progenitors in long-term cultures of human marrow cells. Blood Cells 1988: 14: 355-368. CASHMAN J, EAVES AC, RAINESEW, et al. Mechanisms that regulate the cell cycle status of very primitive hematopoietic cells in long-term human marrow cultures. I. Stimulatory role of a variety of mesenchymal cell activators and inhibitory role of TGFP. Blood 1990: 75: 96-101. COUTINHO L, WILL A, RADFORDJ, et al. Effects of recombinant human granulocyte colony-stimulating factor (CS F), human granulocyte-macrophage-CSF, and gibbon interleukin-3 on hematopoiesis in human long-term marrow culture. Blood 1990: 75: 2 1 18-2 129. HOGGED. CASHMAN J, HUMPHRIES RK, et al. Differential and synergistic effects of human granulocyte-macrophage colony-stimulating factor and human granulocyte colonystimulating factor on hematopoiesis in human long-term marrow cultures. Blood 1991: 77: 493-499. OTSUKAT, THACKER D, HOGGED. The effects of interleukin 6 and interleukin 3 on early hematopoietic events in long-term cultures of human marrow. Exp Hematol 1991: 19: 1042-1048.
68. GORDONMY, RILEY GP, WATT SM, GREAVESMF. Compartmentalization of a haematopoietic growth factor (GM-CSF) by glycosaminoglycans in the bone marrow microenvironment. Nature 1987: 326: 403-405. 69. STEIN J, BORZILLO GV, RETTENMIER CW. Direct stimulation of cells expressing receptors for macrophage colony stimulating vactor (CSF-1) by a plasma membrane bound precursor of human CSF-1. Blood 1990: 76: 1308-1314. 70. FLANAGAN JG, LEDERP. The kit ligand: a cell surface molecule altered in steel mutant fibroblasts. Cell 1990: 63: 185- 194. 71. ERSHLERWB, ROSSJ, FINLAY J, et al. Bone marrow microenvironment defect in congenital hypoplastic anemia. N Engl J Med 1980: 302: 1321-1327. S, MOTOMURA S. IBAYASHI H. Differentiation of 72. KANEKO human bone marrow-derived fibroblastoid colony forming cells (CFU-F) and their roles in haemopoiesis in vitro. Br J Haematol 1982: 51: 217-225. 73. NAGAO T. YAMAUCHI K, KOMATSUDA M, et al. Inhibition of human bone marrow fibroblast colony formation by leukemic cells. Blood 1983: 62: 1261-1265. 74. WLODARSKI KH, JANOWSKA-WIECZOREK A. Precursor cells of fibroblast colony forming unit in the normal and pathologic human bone marrow and in the peripheral blood. Folia Haematol 1984: 5: 686-691. 75. HIRATA J, KATSUNO M, KANEKOS, et al. Clinical signlficance of human bone marrow stromal cell colonies in acute leukemias. Leuk Res 1986: 12: 1441-1445. 76. GREENBERG BR, WILSON FD, WOO L. Granulopoietic effects of human bone marrow fibroblastic cells and abnormalities in the granulopoietic microenvironment. Blood 198 1: 58: 557-556. 77. GREENBERC PL, MARAB, HELLERP. Marrow adherent cell colony-stimulating activity produciion in acute myeloid leukemia. Blood 1978: 52: 362-378. 78. MAYAN1 H, GUILBERT LJ, CLARK SC, BELCH AR, JANOWSKA-WIECZOREK A. Composition and functional in-
79. 80.
81.
82. 83.
84.
85.
86.
87.
88. 89.
90.
91. 92.
93.
tegrity of the in vitro hemopoietic microenvironment in acute myelogenous leukemia: Effect of macrophage colonystimulating Factor. Exp Hematol 1992: 20: 1077-1 084. LANGLEY GR, SMITHLJ, MCCULLOCH EA. Adherent cells in cultures in blast progenitors in acute myeloblastic leukemia. Leuk Res 1986: 10: 953-959. MIYAUCHI J, WANGC, KELLEHER CA, el al. The effects of recombinant CSF-1 on the blast cells of acute myeloblastic leukemia in suspension culture. J Cell Physiol 1988: 135: 55-62. TAKAHASHI M, KEATINCA, SINGER JW. A functional defect in irradiated adherent layers from chronic myelogenous leukemia long-term marrow cultures. Exp Hematol 1985: 13: 926-93 1. JUNEJA HS, GARDNER FH. Functionally abnormal marrow stromal cells in aplastic anemia. Exp Hematol 1985: 13: 194199. GORDONMY, GORDON-SMITH EC. Bone marrow fibroblastoid colony-forming cells (F-CFC) in aplastic anaemia: colony growth and stimulation of granulocyte-macrophage (GM-CFC). Br J Haematol 1980: 49: 465-477. MARSHJCW, CHANGJ, TESTANG, Hows JM, DEXTER TM. Zn vitro assessment of marrow “stem cell” and stromal cell function in aplastic anemia. Br J Haematol 1991: 78: 258-267. HOTTAT, MURATET, YAMAOH, YAMADA H, SAITOH. Granulopoiesis-supporting effects of marrow stromal cells in continuous culture from patients with primary myelofibrosis. Br J Haematol 1986: 64: 669-673. COUTINHO LH, GEARY CG, CHANGJ, et al. Functional studies of bone marrow hemopoietic and stromal cells in the myelodysplastic syndrome (MDS). Br J Haeinatol 1990: 75: 16-25. WATHENLM, KNAPPSA, DEGOWINRL. Suppression of marrow stromal cells and microenvironmental damage following sequential radiation and cyclophosphamide. 1nt J Radiat Oncol Biol Phys 1981: 7: 935-941. TESTANG, HENDRYJH, MOLINEUXG. Long-term bone marrow damage in experimental systems and in patients after radiation or chemotherapy. Anti-Cancer Res 1985: 5: 101. TESTANG, HENDRYJH, MOLINEUX G. Long-term bone marrow damage after cytotoxic treatment: Stem cells and NG, GALERP, eds. Hematomicroenvironment. In: TESTA poiesis: Long-term effects of chemotherapy and radiation. New York: Marcel Dekker Inc., 1988: 75-91. MOLINEUX G, TESTA NG, MASSA G, SCHOFIELD R. An analysis of hemopoietic and microenvironmental populations of mouse bone marrow after treatment with busulfan. Biomedicine 1986: 40: 2 15-220. ANDERSON RW, MATHEWS KI, CROUSEDA, SHARPJG. In vitroevaluation ofhematopoiesis in mice treated with busulphan or nitrogen mustard. Biomedicine 1982: 36: 149-152. TESTANG. BHAVNANI M, WILL A, M O R R I S JONESPH. Long-term bone marrow damage after treatment for acute lymphoblastic leukemia. In: TESTA NG, GALERP, eds. Hematopoiesis: Long-term effects of chemotherapy and radiation. New York: Marcel Dekker Inc., 1988: 279-287. KNOSPEWH. Long-term bone marrow damage after irradiation. In: TESTA NG, GALERP, eds. Hematopoiesis: Longterm effects of chemotherapy and radiation. New York: Marcel Dekker Inc. 1988: 93-130.
233
