UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Molecular mechanisms regulating the accumulation and functional activity of tumour-associated macrophages Dougherty, Shona Thomson 1995

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

Item Metadata

Download

Media
831-ubc_1995-982799.pdf [ 12.76MB ]
Metadata
JSON: 831-1.0088837.json
JSON-LD: 831-1.0088837-ld.json
RDF/XML (Pretty): 831-1.0088837-rdf.xml
RDF/JSON: 831-1.0088837-rdf.json
Turtle: 831-1.0088837-turtle.txt
N-Triples: 831-1.0088837-rdf-ntriples.txt
Original Record: 831-1.0088837-source.json
Full Text
831-1.0088837-fulltext.txt
Citation
831-1.0088837.ris

Full Text

Molecular mechanisms regulating the accumulation and functional activity of tumour-associated macrophages by SHONA THOMSON DOUGHERTY M.B. Ch.B. University of Edinburgh, Scotland, 1985 A THESIS SUBMITTED IN PARTIAL FULHLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Pathology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA APRIL 1995 © Shona Thomson Dougherty, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ffl 7H O L O ^ V The University of British Columbia Vancouver, Canada Date %b* OJ/J /?f$T DE-6 (2/88) ii ABSTRACT Most solid tumour masses also contain a significant number of macrophages and other host-derived immune cells. In previous studies, such infiltrating cells have been shown to exhibit a wide range of biological activities in vitro, some of which, if they were also to occur in vivo, could have a profound effect on tumour growth and/or metastasis. At present, the molecular mechanisms that regulate the accumulation and functional activity of tumour-associated macrophages remain poorly defined. One would anticipate, however, that factors which enhance the recruitment of monocytes or other more primitive macrophage precursor cell types into a tumour site from the peripheral blood, or which enhance the survival, proliferation and/or differentiation of these cells within the tumour microenvironment, might function to increase the total number of macrophages present within a tumour. The work described in this thesis was designed to test this hypothesis through the identification of specific tumour cell-derived factors that could play a role in regulating the accumulation of tumour-associated macrophages. Macrophage progenitor cells with extensive proliferative capacity are known to circulate and have been identified within various histologically distinct tumour types. As a first step towards trying to define the mechanisms that lead to an accumulation of macrophages in tumours, an effort was made to determine whether tumour cells alone are able to provide all of the signals necessary to support the proliferation and differentiation of myeloid progenitor cells in vitro. Using nylon wool non-adherent bone marrow cells as a source of progenitors, tumour cells derived from the murine fibrosarcoma Fsa-N were shown to be fully capable of maintaining the output of clonogenic cells for up to 4 weeks in vitro in the absence of any other exogenous source of growth factors active on these cells. This activity was, however, not restricted to tumour cells as non-transformed fibroblasts derived from a number of non-hemopoietic tissues had a similar supportive capacity. In order to identify some of the factors that may play an important role in this process, fibroblastoid cell lines derived from a number of mutant mouse strains were also tested for their myelosupportive capacity. Interestingly, no difference was observed in the output of clonogenic cells in cultures containing embryonic Sl/Sl and littermate-derived +/+ fibroblasts indicating that Steel factor is not necessary for the sustained production of the progenitor cell type measured in this assay system. In contrast, fibroblastoid cells derived from macrophage-colony stimulating factor (M-CSF)-deficient op/op mice were found to be compromised in their supportive capacity relative to equivalent cells derived from phenotypically normal op/+ littermate controls. Taken together, the data from these experiments suggest that tumour cells and other non-transformed fibroblastoid cells are capable of maintaining myeloid progenitor cells in vitro and that M-CSF can play an important role in this process. In order to define the contribution of M-CSF to the accumulation and functional activity of tumour-associated macrophages in vivo, fibroblastoid cell lines derived from both M-CSF-deficient op/op mice and their phenotypically normal op/+ littermate controls were transformed with a retroviral vector encoding Polyoma large T. The lines produced, designated op/opPy and op/+Py, were then inoculated subcutaneously into SCID mice and both the proportion and phenotype of the macrophages present within the tumours that developed was determined. Since tumours derived from both cell types were found to contain a similar percentage of macrophages it would appear that M-CSF does not play an important role in determining the macrophage content of at least these particular tumours. M-CSF does, however, play a role in regulating their functional activity. Thus, while the macrophages present in M-CSF-deficient op/opPy tumours expressed the same level of 114/A10 mRNA as macrophages present within op/+Py tumours, they expressed much lower levels of Interleukin-lp (IL-lp), Tumour Necrosis Factor-a (TNF-a) and Fc Receptor (FcR)yll mRNA. The molecular mechanisms responsible for maintaining the viability of tumour-associated macrophages in the absence of M-CSF remain to be determined; while preliminary evidence was obtained to indicate that op/op and op/opPy cells may elaborate a soluble factor distinct from M-CSF and GM-CSF that can promote monocyte survival in vitro. Finally, experiments to define the molecular mechanisms that regulate the production and functional activity of TNF-a within the tumour microenvironment were initiated. Two fibroblastoid tumour cell lines, Fsa-R and Fsa-N, were inoculated subcutaneously into syngeneic LPS hyporesponsive (Lpsd) C3H/HeJ and normal (Lpsn) C3H/HeN mice and tumour growth, macrophage content, and the production of TNF-a by tumour-associated macrophages was determined. Fsa-N and Fsa-R tumours grew equally well in both mouse strains and contained almost exactly the same proportion of macrophages. However, the macrophages present within the tumours grown in C3H/HeJ mice produced 5-10 fold less TNF-a than equivalent cells present within tumours grown in C3H/HeN mice. These data suggest that the mechanisms that operate within the tumour microenvironment to induce the production of TNF-a act, at least in part, via the same signal transduction pathway that is defective in Lps^ C3H/HeJ mice. Moreover, it appears that the differences in the level of TNF-a produced within tumours grown in C3H/HeN and C3H/HeJ mice, while seemingly dramatic, are insufficient to alter either tumour growth rate or macrophage content. V TABLE OF CONTENTS Page Abstract ii Table of Contents v List of Tables ix List of Figures x List of Abbreviations xii Acknowlegdements xv Chapter 1 Introduction 1 1. Overview 1 2. Origin of tissue macrophages 2 2.1. Monocyte development in the bone marrow 2 2.2 Recruitment of monocytes into peripheral tissues 6 2.3 Localized proliferation of macrophage precursor cells in peripheral tissues 9 3. Macrophage heterogeneity 10 3.1 Differentiation 11 3.2 Modulation by local microenvironment 13 3.3 Sublineages 14 4. Macrophage activation 15 4.1 Molecular mechanisms regulating macrophage activation 16 4.2 Macrophage cytotoxic activity 19 4.3 Macrophage-induced stimulation of tumour cell proliferation 23 5. Tumour-associated macrophages 24 5.1 Correlation between macrophage infiltration and prognosis 24 5.2 Correlation between macrophage infiltration and tumour metastasis 26 5.3 Correlation between the macrophage content of tumours and regression 29 6. Accumulation of macrophages in tumours 32 6.1 Mediators of acute inflammation 34 6.2 Immune mechanisms 35 6.3 Tumour-derived chemotactic factors 36 6.4 Tumour-derived chemokinetic factors 38 6.5 Depression of macrophage chemotactic and inflammatory responses by tumour-derived factors 40 vi 7. Thesis objectives 43 8. References 45 Chapter 2 Characterization of the myelo-supportive capacity of fibroblastoid cell lines 73 1. Introduction 73 2. Materials and methods 76 2.1 Animals 76 2.2 Cell lines 76 2.3 Expression vectors 77 2.4 Isolation of bone marrow cells 78 2.5 Isolation of bone marrow-derived hemopoietic progenitor cells 78 2.6 Preparation of bone marrow-derived and fibroblast adherent layers 78 2.7 Preparation of conditioned media (CM) 79 2.8 CFU-M assay 80 2.9 RT-PCR analysis of M-CSF transcripts 80 2.10 Assay for competitive repopulating cells (CRU) 81 2.11 CFU-S assay 81 2.12 CFU-F assay 82 2.13 Determination of the supportive capacity of adherent cell lines 82 2.14 CFC assay 82 3. Results 83 3.1 Isolation of a stromal cell depleted, bone marrow-derived hemopoietic cell population that can be used to assay the myelo-supportive capacity of various stromal cell types 83 3.2 Validation of the use of CFC output from NW-non-adherent cells to quantify the supportive function of stromal cell population 85 3.3 Comparison of the myelo-supportive function of fibroblasts from different sources 89 3.4 Characterization of op/op and op/+ fibroblast cell lines 89 3.5 Myelo-supportive ability of op/op and op/+ cell lines 92 4. Discussion 97 5. References 101 Chapter 3 Role of M-CSF in regulating the accumulation and functional activity of tumour-associated macrophages 106 1. Introduction 106 Vll 2. Materials and methods 107 2.1 Animals 107 2.2 Cell lines 107 2.3 Monoclonal antibodies (mAb) 108 2.4 Expression vectors 108 2.5 Preparation of conditioned media (CM) 109 2.6 Macrophage colony assay 109 2.7 Generation of op/opPy and op/+Py tumours 109 2.8 Indirect immunoperoxidase staining 110 2.9 Northern blot analysis 110 2.10 cDNA probes 111 2.11 Monocyte survival assay 111 2.12 GM-CSF bioassay 111 3. Results 112 3.1 Generation and characterization of polyoma transformed op/op and op/+ cell lines 112 3.2 Role of tumour cell-derived M-CSF in regulating the accumulation and functional activity of tumour-associated macrophages 115 3.3 Production of a monocyte survival factor by op/op fibroblasts 120 4. Discussion 125 5. References 128 Chapter 4 Molecular mechanisms regulating TNF-a production by tumour-associated macrophages 132 1. Introduction 132 2. Materials and methods 133 2.1 Animals 133 2.2 Tumour cell lines 134 2.3 Monoclonal antibodies (mAbs) 134 2.4 Preparation of tumour cell suspensions 134 2.5 Isolation of tumour-associated macrophages 135 2.6 Indirect immunoperoxidase staining 135 2.7 Northern blot analysis 135 2.8 Western blot analysis 136 2.9 Purification of tumour cells from enzymatically disaggregated tumour cell suspensions 136 2.10 Preparation of tumour cell conditioned media (CM) 137 Vll l 2.11 Bioassay of GM-CSF 137 2.12 Isolation of a murine TNF-a cDNA 137 2.13 Generation of retroviral vector JmTNF-a 138 2.14 Infection of Fsa-N cells with JmTNF-a 138 2.15 Quantiatation of TNF-a production by transduced Fsa-N cells 138 2.16 Tumourigenicity of Fsa-NJneo and Fsa-NJmTNF-a cells 139 3. Results 139 3.1 Growth of Fsa-N and Fsa-R tumours in C3H/HeN and C3H/HeJ mice 139 3.2 Macrophage content of Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice 139 3.3 TNF-a production by tumour-associated macrophages present in Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice 139 3.4 Production of GM-CSF by TNF-a- and IL-1 p-stimulated Fsa-N cells 143 3.5 Production of GM-CSF by tumour cells purified from Fsa-N tumours grown C3H/HeN and C3H/HeJ mice 147 3.6 Introduction and expression of the murine TNF-a gene in Fsa-N tumour cells 147 3.7 Growth of Fsa-NJneo and Fsa-NJmTNF-a cells in vivo 147 4. Discussion 152 5. References 155 Chapter 5 Conclusions and future directions 160 IX LIST OF TABLES Page Table 1. Frequency, enrichment and recovery of different cell types in normal murine bone marrow after separation on a NW column 87 Table 2. Number of clonogenic progenitor cells produced in co-cultures containing various fibroblast cell lines 90 Table 3. Number of clonogenic progenitor cells produced in co-cultures containing op/op and op/+ fibroblast feeders 97 LIST OF FIGURES  P a 8 e Figure 1. Total CFC production of cultures initiated with fresh bone marrow and NW-non-adherent cells plated directly onto plastic 84 Figure 2. Total CFC production of cultures initiated with NW-non-adherent cells plated onto pre-established LTC adherent layers and onto plastic 86 Figure 3 Linear relationship of CFC production and the number of adherent supporting cells in the feeder layers of co-cultures 88 Figure 4. Requirement for M-CSF or GM-CSF in the proliferation and differentiation of CFU-M 91 Figure 5. Morphology of macrophage colonies grown in M-CSF or GM-CSF 93 Figure 6. Northern blot analysis of M-CSF mRNA expression in cell lines derived from op/op mice and their phenotypically normal (+/?) littermates 94 Figure 7. M-CSF production by cell lines derived from op/op mice and their phenotypically normal (+/?) littermates 95 Figure 8. RT-PCR analysis of M-CSF transcripts 96 Figure 9. Northern blot analysis of Polyoma large T mRNA expression in op/opPy and op/+Py cells 113 Figure 10. Expression of the Polyoma large T antigen in op/opPy and op/+Py cells 114 Figure 11. M-CSF production by op/opPy and op/+Py cells 116 Figure 12. Northern blot analysis of M-CSF mRNA expression in op/opPy and op/+Py cells 117 Figure 13. M-CSF production by pCDM8- and pCDM8.mM-CSF-transfected op/opPy cells 118 Figure 14. Growth of op/opPy and op/+Py tumours 119 Figure 15. Macrophage content of op/opPy and op/+Py tumours 121 Figure 16. Northern blot analysis of 114/A10, FcRYII, IL-lp and TNF-a mRNA Figure 17. Promotion of monocyte survival in vitro by M-CSF and GM-CSF 123 expression in op/opPy and op/+Py tumour-associated macrophages 122 Figure 18. Promotion of monocyte survival in vitro by op/op and op/+ conditioned media 124 Figure 19. Growth of Fsa-N and Fsa-R tumours in C3H/HeN and C3H/HeJ mice 140 Figure 20. Macrophage content of Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice 141 Figure 21. Northern blot analysis of TNF-a mRNA expression in Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice 142 Figure 22. Western blot analysis of TNF-a protein expression in Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice 144 Figure 23. Production of GM-CSF by TNF-a- and IL-lp-stimulated Fsa-N cells 145 Figure 24. Lack of synergy between TNF-a and IL-ip in the stimulated production of GM-CSF by Fsa-N cells 146 Figure 25. Isolation of tumour cells from Fsa-N tumours grown in C3H/HeN and C3H/HeJ mice 148 Figure 26. GM-CSF production by tumour cells purified from Fsa-N tumours grown in C3H/HeN and C3H/HeJ mice 149 Figure 27. Northern blot analysis of TNF-a mRNA expression in Fsa-NJneo and Fsa-NJmTNF-a cells 150 Figure 28. Growth of Fsa-NJneo and Fsa-NJmTNF-a cells in vivo 151 Xll LIST OF ABBREVIATIONS ADCC oMEM BCG CD cDNA CFC CFU-F CFU-G CFU-GM CFU-M CFU-S cGy CM CO2 C. parvum CRU DMEM DTH EDTA FcRYII FCS FCI FMLP 5FU G-CSF GM-CSF Y-ray Gy HBSS HPP-CFC ICAM-1 ICAM-2 IFN-Y antibody-dependent cell-mediated cytotoxicity alpha-minimum essential medium Bacille-Calmette-Gueurie cluster differentiation cloned DNA colony forming cell colony forming unit-fibroblast colony-forming unit-granulocyte colony-forming unit-granulocyte/macrophage colony-forming unit-macrophage colony-forming unit-spleen centigray conditioned medium carbon dioxide Corynebacterium parvum competitive repopulating unit Dulbeccos minimum essential medium delayed-type hypersensitivity ethylenediaminetetraacetic acid Fc receptor gamma-II fetal calf serum fetal clone 1 F-met-leu-phe 5-fluorouracil granulocyte colony-stimulating factor granulocyte/macrophage colony-stimulating factor gamma-rays Gray Hanks balanced salt solution high proliferative potential-colony forming cell Intercellular adhesion molecule-1 Intercellular adhesion molecule-2 interferon-gamma intrahepatic interleukin xni Th2 intraperitoneal intravenous Jzen. 1-based retroviral vector containing murineTNF-a cDNA Jzen.l-based retroviral vector containing neomycin-resistance cDNA kilodaltons kilovoltage potential Leukocyte function-associated antigen-1 Leukocyte function-associated antigen-3 lipopolysaccharide LPS hyporesponsive gene LPS normoresponsive gene long-term culture long-term culture-initiating cell monoclonal antibody macrophage chemotactic protein-1 N-acetyl-L-alanyl-D-isoglutamine macrophage inhibitory factor nature killer cells nitric oxide nylon wool cell line derived from homozygous mice carrying osteopetrosis gene op/op cell line expressing Polyoma Large T antigen cell line derived from heterozygous mice carrying osteopetrosis gene op/+ cell line expressing Polyoma large T peripheral blood leukocytes phosphate buffered saline platelet-derived growth factor purified protein derivative Roslin Park medium reverse transcriptase-merase chain reaction Homozygous Steel locus mutation Normal homozygous littermate control Tumour-associated macrophages Transforming growth factors-beta T helper type 1 lymphocytes T helper type 2 lymphocytes Tumour necrosis factor-alpha xiv UV ultraviolet VPF vascular permeablity factor XV ACKNOWLEDGEMENTS I wish to express my gratitude to Connie Eaves for being my supervisor and for her support and guidance, to Dr. Dana Devine, (Department of Pathology, UBC), Dr. Dagmar Kalousek (Cytogenetics, B.C. Childrens Hospital), Dr. Fumio Takei (Department of Pathology, UBC), Dr. Don Brunette (Department of of Oral Biology, UBC) for helpful guidance over the years, to Grace Lima and Jessica Maltman for expert technical assistance, Fred Jensen and his team for taking care of the animals used in my experiments, to The National Cancer Institute of Canada for financial support, to my daughters, Sarah and Karen, for their patience, and to Graeme, my husband and friend without whose support and encouragement, I could not have begun, let alone come this far. 1 CHAPTER 1 Introduction 1.0 Overview Macrophages are evolutionally one of the oldest identifiable types of defensive cells in multicellular organisms. A great deal is known about their role in health, development and certain infectious and other disease states. Scientists have long been curious about their presence within solid tumour masses and the function that they play. Recently, much interest has been shown in the possibility of augmenting some of their documented antitumour activities. Central to our ability to exploit this avenue for therapeutic gain, is an understanding of the mechanisms that are involved in the accumulation and functional activity of tumour-associated macrophages. The term "macrophage", from the Greek meaning "big eater", was originally coined more than 100 years ago by the Russian embryologist Hie Metchnikoff, and used by him to describe the large phagocytic mononuclear cells that he first observed in the coelomic space of sea urchin embryos and subsequently in the circulation, tissues and body cavities of a variety of mammalian species including man (1). In 1924, Aschoff assigned macrophages to the Reticuloendothelial System along with reticular cells, endothelial cells, fibroblasts, histiocytes and monocytes (2). However, following the realization that monocytes and macrophages share a common bone marrow precursor, both cell types, together with their committed progenitors (monoblasts and promonocytes), were instead grouped into what is now known as the Mononuclear Phagocyte System (3). The term 'resident macrophage* is frequently used to describe the macrophages present within normal, non-inflamed tissues. These same cells may also be observed in small numbers in an inflammatory exudate where they are sometimes called 'normal macrophages'. 'Elicited macrophages' are those that have been attracted to a given site in response to a particular stimulus and may show considerable heterogeneity. 'Activated macrophages' demonstrate an increase in 2 one or more of their functional activities, or in the appearance of a new functional activity. Both 'resident' and 'elicited' macrophages can become activated (4). 2.0 Origin of tissue macrophages It is now generally accepted that under normal steady state conditions most tissue macrophages originate from pluripotent hemopoietic stem cells located in the bone marrow (5). These cells divide and differentiate giving rise to progeny that are heterogeneous in nature and variably committed to produce cells of the lymphoid and myeloid lineages (6). The most immature cell thus far described that appears to be irreversibly restricted to macrophage production is the monoblast (7). Murine monoblasts have a cell cycle time of about 12 hours (5) and division of one cell produces two promonocytes which, in turn, divide after about 16 hours to give two monocytes (5). These proliferation and differentiation events are absolutely dependent, at least in vitro, on the presence of certain growth factors including M-CSF and GM-CSF (8). Monocytes do not remain in the bone marrow for long and are rapidly released into the circulation. In the mouse, their half-life in the blood is about 17 hours (9). Under steady-state conditions a small proportion of these cells are believed to migrate into the tissues where they differentiate under the influence of local microenvironmental signals to produce characteristic tissue macrophages (9). Cytokines and chemotactic factors that are released at sites of inflammation can act on adjacent vascular endothelial cells inducing the expression of adhesion proteins thereby dramatically upregulating the recruitment of monocytes from the circulation (10,11). 2.1 Monocyte development in the bone marrow In adult animals the hemopoietic stem cells from which monocytes and tissue macrophages are ultimately derived are located primarily in the bone marrow. In the absence of unique morphologic and phenotypic characteristics suitable for the direct identification of hemopoietic stem cells, these cells are defined operationally by their ability to reconstitute, and sustain for extended periods of time, all of the hemopoietic compartments of lethally irradiated recipients, 3 including cells of both the lymphoid and the myeloid lineages (12, 13). They are capable of self renewal (14), and in addition, produce progeny that become progressively more limited in their proliferative capacity and differentiation potential. As is the case for pluripotent hemopoietic stem cells, detection of these various derivative progenitor cell types relies on the use of functional assays that measure their developmental potential. The spleen colony-forming cell (CFU-S), which gives rise to a macroscopic colony on the surface of the spleen upon injection into an irradiated recipient, was the first pluripotent progenitor cell type to be identified using this type of functional approach (12, 15). Subsequently, it was shown that cells identifiable as CFU-S are biologically heterogeneous and can be further subdivided based on the time after injection in which the macroscopic spleen colonies derived from these cells first appear. Thus CFU-S that generate macroscopic colonies within 8-9 days may represent a more mature cell than CFU-S that produce colonies that only appear 12-14 days after injection (16-18). The competitive repopulating unit (CRU) assay provides a methodology for quantitating pluripotent murine hemopoietic cells that are able to repopulate both the lymphoid and myeloid compartments of lethally irradiated mice when injected at limiting dilution (19). CRU occur at a frequency of 1 in 104 nucleated adult murine marrow cells (19) which is slightly lower than the frequency of CFU-S (3 per 104). CRU are believed to be present at low numbers in the circulation but can be readily detected there following a variety of systemic treatments including chemotherapy or specific growth factor administration. Although CRU and some Day 12-14 CFU-S share many properties resulting in their frequent co-purification (20,21), the extent to which these populations overlap remains controversial (22). Several years ago Nakahata and Ogawa (23, 24) described in vitro conditions that allow the clonal growth of a cell with stem cell-like properties. This so-called 'blast colony-forming cell' forms colonies which, at an early stage will contain 20-100 undifferentiated blast cells most of which can form secondary colonies of mature cells when transferred to fresh culture medium. These secondary colonies have been found to contain cells of different myeloid lineages in varying combinations (23, 25-30). Although the presence of CRU amongst the in vitro generated progeny of blast colony-forming cells has not been reported, recent modifications to the culture conditions 4 employed have made it possible to show that a proportion of blast colony-forming cells have B lymphopoietic as well as multi-lineage myeloid potential. The murine high proliferative potential colony-forming cell (HPP-CFC) is another operationally defined hemopoietic progenitor cell with extensive proliferative capability demonstrable in vitro (31). HPP-CFC are relatively resistant to the effects of 5-fluorouracil (5-FU) in vivo (31,32) indicating that they are a non-cycling population. They form colonies containing several thousand macrophages in cultures stimulated by a combination of cytokines, usually including M-CSF (33). Presumably these cells are closely related to, or overlap with, those described as blast colony-forming cells in replating experiments. However, HPP-CFC are recognized to encompass a spectrum of progenitor stages and their precise relationship to other cell types within the hemopoietic hierarchy is not certain. Another primitive progenitor cell that can be detected by an in vitro assay is the long-term culture-initiating cell (LTC-IC) or cobblestone area-forming cell (CAFC) (34). The LTC-IC assay allows an in vitro estimate of the marrow repopulating ability of a cell population. The assay uses a miniturized stroma-dependent bone marrow culture to detect the presence of primitive cells with the ability to initiate long-term cultures. Limiting dilutions of cells to be assayed are co-cultured with pre-established irradiated stromal layers in microliter wells and the formation of cobblestone areas by the LTC-IC, is observed. The frequency of LTC-IC is then calculated using Poisson statistics, and correlates with the production of secondary granulocyte/macrophage colony-forming units in the adherent layer of individual wells. Lineage-restricted hemopoietic progenitor cells are, like their pluripotent precursors, not morphologically recognizable and have therefore also been more readily studied using functional rather than cytological or immunophenotyping methods for their identification. The first in vitro hemopoietic progenitor cell assay to be developed (35, 36) demonstrated that bone marrow cells suspended in semi-solid agar cultures in the presence of 'feeder cells' formed colonies containing granulocytes and/or macrophages. The cells giving rise to these colonies were termed CFU-GM. Subsequently, it was shown that the feeders could be replaced by conditioned media and eventually by specific growth factors, three of which came to be known as M-CSF, G-CSF and 5 GM-CSF reflecting the cellular composition of the colonies they induce (37). Similar types of semi-solid culture systems containing appropriate growth factors have also been devised to allow detection of lineage-restricted progenitors for each of the other major myeloid pathways. Unlike their more immature progenitors, monoblasts can be identified and isolated based on their expression of unique morphological and phenotypic characteristics. They produce detectable amounts of lysozyme and non-specific esterase (NSE), have IgG receptors and can phagocytose IgG-coated erythrocytes, but not C3b-coated erythrocytes, and only rarely ingest opsonized bacteria (38). They do not pinocytose to any significant extent. The progeny of monoblasts, promonocytes, are also positive for lysozyme and NSE and have in addition, peroxidase-positive granules. The majority have IgG and C3b receptors and can ingest IgG-coated erythrocytes and opsonized bacteria but relatively few C3b-coated erythrocytes (38, 39). They can pinocytose. The murine promonocyte has a diameter on light microscopy of 12-18 nm with several characteristic immature and mature azurophilic granules within its cytoplasm and a deeply indented and irregular nucleus. On electron microscopy promonocytes have distinctive cytoplasmic granules (very different from granulocytic granules), well developed golgi, plentiful ribosomes and a moderate number of mitochondria. The nucleus is of moderate size, kidney- or horse shoe-shaped, with one or more nucleoli and sometimes slight peripheral chromatin condensation (40). Monocyte maturation from promonocytes involves an increase in nuclear chromatin condensation, a decrease in both ribosome numbers and the amount of rough endoplasmic reticulum. By light microscopy, murine monocytes have a diameter of 12-15 ^ m with an eccentric, kidney-shaped, round or irregular nucleus containing fine chromatin and occupying 50% of cell. The cytoplasm contains fine granules, large azurophilic granules and numerous clear vacuoles. By electron microscopy, 1 or 2 small nucleoli are identifiable, surrounded by chromatin. Abundant cytoplasm is seen with scattered rough endoplasmic reticulum and prominent golgi (near nuclear indentations). There are scattered small vesicles, numerous in the golgi region and at the cell surface, which represent pinocytotic vacuoles. There are abundant small mitochondria, 6 inconspicuous bundles of fibrils (often perinuclear) and granules 0.05-0.2 \xm in diameter which contain dense, homogeneous material with a surrounding membrane (40). 2.2 Recruitment of monocytes into the peripheral tissues An essential step in the entry of monocytes into the tissues under both steady state and inflammatory conditions requires that these cells adhere to and then migrate through the vascular endothelium (41). At present, the molecular mechanisms that regulate this process remain poorly defined. In contrast, the events that regulate the transendothelial migration of lymphocytes and neutrophils have been much better characterized and shown to depend upon a multistep adhesion cascade that involves the sequential and co-ordinated interaction between numerous receptor-ligand pairs (42, 43). Briefly, this model predicts that the initial binding of leukocytes to vascular endothelial cells is mediated primarily by members of the selectin gene family which recognize and bind fucosylated carbohydrate ligands, especially structures containing sialyl-Lewisx (44, 45). Leukocyte L-selectin plays a central role in this process, as it may act both as a constitutively active lectin recognizing various endothelial cell ligands including GlyCAM-1 (46), CD34 (47), and MAdCAM-1 (48), and as a structure presenting sialyl-Lewisx to both P- and E-selectin, which are transiently induced on endothelial cells in response to inflammatory cytokines such as IL-lp and TNF-a (44, 45). Although of low affinity, selectin-mediated adhesion events are sufficient to allow cells to partly resist the shear force exerted by the flow of blood causing them to slow down and "roll' along the surface of the vessel (49, 50). If appropriate signals are present, leukocytes stop rolling and attach firmly to the endothelial cell surface. This process appears to be mediated by leukocyte integrins (42, 43). Depending on the cell type and target tissue involved, several different integrins may be used includinga4pi which can recognize and bind both VCAM-1 and the CS-1 containing splice variant of fibronectin (42, 51); a4p7 which binds MAdCAM-1 (25), and cqp2 (LFA-1) which binds ICAM-1 and ICAM-2 (42, 51). Although ICAM-1 is not constitutively expressed on endothelial cells, its production occurs following stimulation of these cells with inflammatory cytokines (42, 51). ICAM-2 is constitutively expressed on endothelial cells (52) and may function as a ligand for cqp2 in non-7 inflamed tissue sites. A third LFA-1 ligand, designated ICAM-3, has recently been identified, but this molecule does not seem to be present on endothelium (27). Although constitutively expressed, leukocyte integrins need to be activated before they can bind their respective ligands (26). While it is unlikely that they are the only molecules involved, chemoattractants belonging to the chemokine family have been shown to upregulate integrin function on neutrophils and T cells (28, 29). The presence of heparin-binding sites on these molecules suggests another potential adhesive mechanism by which they could be retained on the endothelial cell surface (29). The final step in the adhesion cascade involves the transendothelial migration of leukocytes and their subsequent movement into the tissues. Once again, integrins, particularly the fibronectin receptor a5Pi, appear to play an important role in this process (53). In general, monocytes appear to adhere better to non-stimulated endothelial cells than do either lymphocytes or granulocytes (41, 54-57). The majority of monocytes express L-selectin, and under conditions of shear, the initial attachment of these cells to endothelial cells could be substantially inhibited by monoclonal antibodies against this molecule (58). In contrast, under non-shear conditions, antibodies against L-selectin had little effect, while binding was inhibited 40-70% by monoclonal antibodies belonging to CDlla, CDllb or CD18 (54, 55, 57). These data suggest an important role for the fc integrins LFA-1 and Mac-1 in mediating the firm attachment of monocytes to unstimulated endothelium. However, antibodies against these molecules, even when present together, are unable to completely block monocyte adherence, suggesting that other cell surface adhesion proteins must also be involved in this process (54, 55, 57, 59). The ligand recognized by LFA-1 and Mac-1 on the endothelial cell surface is also unclear. Although unstimulated endothelial cells have been found to express low levels of ICAM-1 (60), monoclonal antibodies against this molecule did not inhibit monocyte adhesion (61). On the other hand, although unstimulated endothelial cells do express ICAM-2 (52), it is unknown whether this molecule is involved in monocyte adhesion. Monocytes also express a third 02 integrin, CDllc/CD18 (pl50,95) (62, 63) which has been shown to play an important role in monocyte attachment to, and migration on, plastic surfaces (64). However, contradictory results have been obtained concerning its possible involvement in 8 the adhesion of monocytes to unstimulated endothelium (65-67). pl50,95 is expressed at far higher levels on tissue macrophages than on circulating monocytes (68). Thus it is tempting to speculate that this molecule is mainly involved in events that occur after transendothelial migration and following differentiation of monocytes into tissue macrophages. Recent studies have demonstrated that much of the ^-independent adhesion of monocytes to unstimulated endothelium could be blocked by monoclonal antibodies against the 04 or Pi chains of VLA-4 (69). In contrast to the situation seen with activated endothelium, however, monocyte attachment to unstimulated endothelium was not inhibited by monoclonal antibodies against VCAM-1 (69), suggesting that in this particular instance, VLA-4 binds to some as yet unknown ligand. Large numbers of monocytes accumulate at sites of inflammation in vivo. Several in vitro studies have demonstrated that monocytes show increased adhesion to endothelial cells that have been stimulated with various inflammatory cytokines including TNF-a, IL-1 or IFN-V(61). Moreover, the monocytes that bound to such stimulated endothelial cells rapidly developed a stretched or spread morphology that was not seen on equivalent unstimulated endothelial cells (70). Using a combination of cytokines that induce a different spectrum of adhesion proteins on endothelial cells, and a panel of blocking monoclonal antibodies, it has been shown that much of the enhanced binding of monocytes to stimulated endothelium can be ascribed to an interaction of monocyte VLA-4 with VCAM-1 on the endothelial cell surface (71). A role for E-selectin has also been proposed (61, 63, 72). However, the induction of E-selectin on endothelial cells appears to peak 4 hours after stimulation with cytokines, returning to basal levels by 24 hours (73), while maximal binding of monocytes occurs at 24 hours (61), a finding that would appear to argue against an involvement of E-selectin. Moreover, the entry of monocytes into sites of inflammation similarly peaks between 24 and 48 hours (74). It should be noted, however, that the expression of E-selectin may be more persistent in vivo than in vitro (75). Less is known about the molecules that mediate the subsequent transendothelial migration of monocytes. Time lapse video studies have revealed that the initial site of monocyte attachment to endothelial cells is random. However, they then rapidly migrate and accumulate at points of contact 9between adjacent endothelial cells (54). It is interesting to note in this regard that a recentlydiscovered adhesion protein designated PECAM-1 or CD31 (76) also localizes to the intercellular junctions of cultured endothelial cells (77). In vivo, however, PECAM-1 appears to be evenly distributed on the surface of endothelial cells, although it does redistribute to endothelial cell junctions following stimulation with TNF-a(78), suggesting a possible mechanism by which the recruitment of monocytes into sites of inflammation could be regulated. Various other molecules have been suggested to play a role in monocyte transendothelial migration. In particular, it has been demonstrated that although monoclonal antibodies against the hyaluronan-binding domain of CD44 do not interfere with monocyte attachment to endothelial cells, they do profoundly inhibit transendothelial migration (Dougherty, G.J., personal communication). In summary, as with other leukocyte cell types, an interacting cascade of adhesion proteins appear to regulate the entry of monocytes into the tissues under both steady state and inflammatory conditions. 2.3 Localized proliferation of macrophage precursor cells in the peripheral tissues Cell marking studies suggest that most tissue macrophages are derived from circulating monocytes. Nevertheless, a number of studies have demonstrated that a small proportion of 'macrophages' isolated from sites other than the bone marrow, such as the peritoneal cavity, blood, lung, liver, thymus and lymph nodes, are capable of proliferation in the presence of appropriate colony-stimulating factors (79-83). The precise position of these macrophage colony-forming cells (CFU-M) within the hemopoietic hierarchy remains to be determined, but it is likely that they represent myeloid progenitor cells that have left the bone marrow and entered the peripheral tissues. It is likely that they are more mature than bone marrow CFU-GM as they can survive an initial absence of colony stimulating factors, they only proliferate in vitro after a delay of one to two weeks, and then generate only small colonies composed solely of macrophages (8). The quantitative significance of the local proliferation of myeloid progenitor cells in the generation of tissue macrophages in vivo has been the subject of considerable debate (5, 84). The percentage of macrophages which incorporate 3H-thymidine in vivo varies from one tissue to 10 another. For example, only approximately 5% of peritoneal macrophages will incorporate this label (85), while in the lung, this figure may be as high as 20% (86, 87). Consequently, it seems likely that the extent to which local proliferation contributes to the macrophage pool will depend largely on which tissue is being examined. The macrophages that accumulate at inflammatory sites are also derived mainly from blood monocytes (88). However, depending on the nature of the inflammatory stimulus used, a variable proportion of these cells may proliferate in situ (89). Nevertheless given the magnitude of the monocyte influx, it seems unlikely that local proliferation contributes significantly to the total number of macrophages that present within inflammatory sites, at least early in the response. 3.0 Macropha£e heterogeneity Macrophages were originally considered to be mainly scavenging cells but it is now appreciated that they can perform a large number of diverse and, on occasion, mutually antagonistic functions, many of which are essential to the appropriate development and well being of the host. Thus while macrophages appear on one hand, to constitute a major line of host defense against the establishment and spread of microbial (90-92) and neoplastic (93-96) disease, they may also support the growth of certain microorganisms (97) and tumour cells (98-101). Similarly, although macrophages have been shown to act as accessory cells in the induction of humoral and cell-mediated immune responses (102, 103), they may also inhibit the proliferation and differentiation of both lymphocytes (104-106) and other cell types (107, 108). Other important functions ascribed to macrophages include the secretion of numerous biologically active substances (109) and various homeostatic activities unrelated to host defense such as the removal of senescent red blood cells, turnover of lung surfactant, and tissue repair and remodeling (110, 111). They are also involved in the regulation of hemopoiesis and angiogenesis, and are thought to play an important role in reproduction and embryogenesis. For example, the macrophage mediator TNF-a has been shown to be secreted by spermatids (112) and to inhibit the production of testosterone by the Leydig cells in the testes (113). In the female, TNF-a can be detected in the follicular cells of 11 both healthy and atretic ovarian follicles as well as in oocytes (114). TNF-a mRNA also has been detected in murine embryonic tissue and in fetal thymus and spleen (115). Antibodies directed against TNF-a injected into mice in utero Or just after birth, have been shown to have profound effects on growth as well as causing marked atrophy of certain organs, i.e. thymus, spleen and lymph nodes with an associated lymphopenia (115). Although all macrophages show common morphological, biochemical or functional characteristics they are by no means uniform in these respects. As long ago as 1925, Sabin et al. (116) using supravital staining techniques described the similarity between resident murine macrophages in the spleen, peritoneal cavity and liver and pointed out that such cells differ in many respects from the monocyte-derived exudate macrophages elicited by intraperitoneal (i.p.) administration of blood. The development in recent years of various assays of macrophage function, the identification of cell surface and intracellular antigens using heteroantisera and monoclonal antibodies, and the use of assays which identify and quantitate intracellular and extracellular enzyme activities, has produced a wealth of data documenting macrophage heterogeneity (117-119). Thus it seems that for almost every characteristic which has been examined, macrophages can show diversity of expression. The precise molecular mechanisms responsible for the generation of this heterogeneity remain to be determined. What little evidence there is suggests that both differentiation processes and their modulation by local microenvironmental stimuli play major roles. The possible existence of macrophage sublineages derived from distinct bone marrow precursors poses an additional element of complexity to this question. 3.1 Differentiation While certain macrophage characteristics are stably expressed during the differentiation of these cells from bone marrow precursors, the expression of others features depends upon differentiation stage and may vary in both a qualitative and quantitative manner. For example, during the differentiation of macrophages from bone marrow precursors in vitro, the expression of receptors for C3b, the ability to pinocytose horse radish peroxidase and lysozyme activity all 12 increase (7, 120). Bursuker et all. (88) have suggested that the expression of the lysosomal enzyme p-galactosidase may be an even more useful indication of macrophage differentiation stage, since the activity of this enzyme increases during differentiation but, unlike many other characteristics (e.g. C3b expression), p-galactosidase levels remain unaltered when macrophages are exposed to environmental stimuli (88). As they differentiate macrophages may also express different functional characteristics. For example, the cells exhibiting a natural killer-like activity which can be isolated from a Corynebacterium parvum-elicited peritoneal exudate or which can be generated in cultures of bone marrow cells were identified as promonocytes. More mature myeloid cells did not perform this function (121). Differentiation of monocytes into characteristic tissue macrophages either in vivo or in vitro is accompanied by a considerable increase in cell size (122). It is possible to exploit this fact to separate macrophages into functionally distinct subpopulations and thereby demonstrate a correlation between differentiation stage and functional capability (122, 123). For example, only small, presumably poorly differentiated, macrophages were found to express MHC Class II determinants and present macrophage-bound antigen to primed T cells in an antigen-specific proliferation assay. It was therefore proposed that as macrophages differentiate they lose MHC Class II expression and accessory cell function (122, 123). Taken together, these findings suggest that the differentiation of macrophages may involve the generation of a number of functionally distinct phenotypes which together contribute to their apparent heterogeneity. Within a particular (geographically localized) macrophage population, some cells would likely be poorly differentiated having just arrived from the blood stream, whereas others might have been resident in the tissue for a longer period of time. Inflammatory agents can produce dramatic alterations in the functional characteristics of macrophage populations. It is possible that this is achieved, at least in part, by causing an influx of relatively immature cells into a site, thereby altering the proportion of cells in different differentiation stages (88). Proliferation of immature exudate macrophages may add further to the degree of heterogeneity. It has been reported that the expression of several cell surface components varies with the cell cycle 13 (124, 125). The heterogeneity of Fc receptor expression in the macrophage tumour cell line P388Di and other cell lines may be due to cell cycle variation (126). Neuman and Sorg (127) have suggested that primary macrophages also express distinct functions at different stages in the cell cycle and that stimulation of proliferation could thus contribute to macrophage heterogeneity. 3.2 Modulation by the microenvironment The term "modulation" is used to describe those induced changes in macrophage morphology, biochemistry or function that result from continuous exposure to a particular signal but, when the signal is removed/the changes are reversed and the cell rapidly returns to its former state (128). Macrophages from different tissues vary considerably with respect to a large number of characteristics and modulation is thought to be the major source of this heterogeneity. This concept is based on the assumption that essentially identical cells enter tissues and that the heterogeneity observed between macrophages at different sites results from the exposure of the incoming cells to site-specific microenvironmental signals. For example, alveolar macrophages function in a high oxygen environment, are largely aerobic and generate their energy by oxidative phosphorylation. In contrast, peritoneal macrophages function in a low oxygen environment, are largely anaerobic and generate their energy by glycolysis. Consequently, these two populations differ in their expression of key enzymes of the oxidative phosphorylation and glycolysis metabolic pathways (129-131). Bar-Eli et al. (132) have shown that the progeny of single bone-marrow macrophage precursors exhibit different metabolic enzyme profiles depending upon culture conditions. If cultured in a high oxygen environment, they expressed the enzymes of oxidative phosphorylation. If cultured in a low oxygen environment, they expressed the enzymes of glycolysis. These environmentally-induced changes were also reversible, i.e., cells adapted to a low oxygen tension were exposed to high oxygen conditions showed a time-dependent recovery in the expression of the enzymes of oxidative phosphorylation. In as much as enzyme expression is readily altered by the cellular environment it is a modulated characteristic. Many other aspects of the heterogeneity observed between macrophages 14 at different sites may similarly be the result of environmental modulation. Macrophage MHC Class II expression and cytotoxic activity are two obvious examples. Although a feature of modulated changes is that they are reversible, it is possible to imagine a situation where microenvironmental signals exert such a profound effect on cellular physiology that some induced characteristics might appear permanent. The expression of a number of important macrophage characteristics, however, may not depend exclusively on the nature of the microenvironmental stimuli present. For example, the degree of macrophage heterogeneity which results from the administration of inflammatory stimuli may be explained not only in terms of the nature of the modulating signals generated but also in terms of the responsiveness of the cells to these signals. Macrophage responsiveness is likely to depend both on differentiation stage and previous exposure to modulating signals. 3.3 Sublineages The possible existence of true sublineages of macrophages derived from distinct bone marrow precursors provides an additional level of complexity to the subject of macrophage heterogeneity. These precursors may produce cells with a preferred tissue destination and functional capabilities, and stimuli such as those generated by inflammation may lead to the preferential involvement of certain sublineages. The evidence in support of the existence of distinct macrophage sublineages is not conclusive. However, it does appear that macrophage precursor cells can be heterogeneous with respect to physical, biochemical and functional characteristics (133-136) and even clonal subpopulations of macrophages derived from individual bone marrow precursors may express different phenotypes. For example, Claesson and Olsson (137) found that only selected colonies of murine bone marrow-derived macrophages, expressed strong natural cytoxicity for tumour targets. The authors did not, however, examine the differentiation stage of the cells in each colony. Previously, promonocytes have been identified as the cells responsible for natural cytoxicity (138). It is therefore interesting to note that colonies consisting exclusively of promonocytes have been described in cultures of murine bone marrow (139). 15 Bursuker and Goldman (124) have shown that in cultures of bone marrow from normal mice, macrophage clones express either high or low 5'-nucleotidase activity. The expression of the ectoenzyme 5'-nucleotidase has proven a useful marker to distinguish resident and inflammatory peritoneal macrophages which express high and low activity, respectively (124, 125). These results have been taken to suggest that resident and inflammatory macrophages may arise from distinct bone marrow precursors (124). This conclusion was reinforced by the finding that in cultures of bone marrow from animals undergoing an inflammatory reaction, the percentage of macrophage clones expressing low 5'-nucleotidase activity was significantly increased (125). In this instance, there appeared to be no major differences in the degree of differentiation between bone marrow-derived macrophages from normal or inflammation bearing animals as determined by the expression of p-galactosidase, although clonal expression of this enzyme was not simultaneously examined. Thus the progeny of individual macrophage precursors appear to retain certain distinctive characteristics during their differentiation in vitro. Assuming that there is no internal regulation within a clone that results in the collective expression of a particular set of characteristics, then these experiments imply the existence of distinct macrophage sublineages. The actual number of such sublineages remains to be determined. Although it seems unnecessary to postulate that many will be found and heterogeneity is likely to be generated mainly by differentiation and/or microenvironmental modulation of one, or at the most, a few highly versatile macrophage lineages. 4.0 Macrophage activation Normal resident tissue macrophages can phagocytose and kill pyogenic bacteria such as Staphylococcus aureus and Escherichia coli (140). There are, however, a number of diverse microbial pathogens that can survive inside macrophages and eventually kill them through unrestricted intracellular multiplication. Such organisms include the bacteria Brucella, Listeria, Mycobacteria, Salmonella and Legionella (90,141-143), and the protozoa Leishmania (144, 145), Toxoplasma (146) and Trypanosome (147). Animals can, however, acquire impressive levels of 16 immunity to infection with these intracellular parasites (148). Such immunity was found not to be passively transferable by serum (149), but was related to the development of an enhanced microbidical activity by host macrophages (90, 141). The term 'activated macrophage' was introduced by Mackaness (141) for the purpose of describing the intrinsic adaptive changes that enable macrophages to express enhanced antimicrobial activity. It is now evident that 'activated' macrophages obtained from animals with acquired immunity to infection with intracellular parasites not only show enhanced antimicrobial activity against the infecting organism but are also able to non-specifically kill antigenically unrelated organisms (142), virally infected cells (150) and transformed cells (151). Inflammatory stimuli induce dramatic changes in the morphological, biochemical and functional characteristics of macrophage populations. The response generated depends upon the nature of the inflammatory stimulus used. Moreover, even within a population of macrophages elicited by an inflammatory stimulus, not all are activated in the strict sense (152) of having acquired the ability to kill intracellular parasites or tumour cells (147, 153). In addition, different subpopulations of 'activated, macrophages may be differentially isolated (154). 4.1 Molecular mechanisms regulating macrophage activation The activation of macrophages depends upon an interacting cascade of signals (155, 156). LPS is a potent stimulator of many of the early events involved in this process, such as the production and secretion of TNF-a, IL-1 and IL-6 (157, 158). However, LPS alone does not generally induce effector function unless the cells have first been primed, for example, by treatment with IFN-y(159, 160). It has been suggested that one way in which IFN-y may mediate its effects is to induce the expression on macrophages of receptors for TNF-a (161, 162). With the appearance of these receptors, TNF-a produced in response to LPS can act in an autocrine fashion amplifying cytokine secretion and triggering a series of signal transduction events that lead ultimately to the acquisition of anti-microbial and/or anti-tumour function (163, 164). The involvement of TNF-a in these events is supported by the observation that recombinant TNF-a can replace LPS in the induction of effector functions in macrophages primed with IFN-y (163, 17 164). Moreover, antibodies against TNF-a have been shown to inhibit LPS-mediated induction of effector functions in macrophages primed with IFNj/(165). Although LPS may play a role in macrophage activation in response to bacterial pathogens, it is unlikely to be present within tumours or within other non-septic inflammatory sites. T cells belonging to the T helper type 1 (Thl) subset are, however, capable of producing both TNF-a and IFN-y in response to antigen presented in association with appropriate Class II determinants and could, therefore, provide the activating signal usually ascribed to LPS (166, 167). Indeed, culture supernatants conditioned by T helper type 1 lymphocytes (Thl cells) have been shown to activate anti-microbial and anti-tumour functions in macrophages and this activity could be reduced or abrogated by antibodies directed against IFN-y or TNF-a (166). Culture supernatants from T helper type 2 lymphocytes (Th2 cells) which do not secrete either IFN-y or TNF-a, could not activate macrophages by this mechanism (168). Several groups have reported evidence to suggest that activation of macrophages by T cells may require direct contact between the two cell types (166,168-170). Although the molecular mechanism underlying such a requirement is unclear, it should be noted that TNF-a is present on some Thl cells as a transmembrane protein (170-172). Moreover, Thl cells expressing membrane-bound TNF-a have been shown to activate macrophage anti-leishmania activity more effectively than either soluble TNF-a (170) or Thl cells that did not express membrane bound TNF-a (169). Th2 cells do not express TNF-a and were unable alone to fully activate resting macrophages, but were able to do so if the macrophages involved were first primed with IFN-y (166). This activity required direct interaction between the Th2 cells and the macrophages and was not mediated by culture supernatants conditioned by Th2 cells (168). The initial interaction between Th2 cells and macrophages was antigen-specific and MHC-restricted (166). The subsequent activation of macrophages was not, and could even be mediated by paraformaldehyde fixed cells (168). Although the molecular mechanisms by which Th2 cells mediate their macrophage-activating function remain unresolved, most studies have focused on the potential involvement of signal transduction pathways triggered via cell surface adhesion proteins. Although the interaction between p39 and its ligand CD40 has been implicated in T cell-B cell signaling 18 (173), it appears that CD40 is not expressed on inflammatory macrophages (174). Macrophages do express B7.2 and can be induced to express B7.1 by treatment with IFN-y (175), raising the possibility that interactions between CD28 and one or other of these two molecules could perhaps play a role in macrophage activation. Macrophages also express LFA-3, and ICAM-1 and ICAM-2, which can interact, respectively, with the T cell molecules CD2 and LFA-1. LFA-1 interactions with ICAM-1 have been implicated in B cell activation (176) and crosslinking of LFA-3 as well as CD44 and CD45 have been shown to induce human monocytes to secrete TNF-a (177). Since no difference was observed in the ability of Thl and Th2 cells to activate B cells in a contact-dependent manner (178), it is likely that contact-dependent activation of macrophages may be similarly mediated by both T helper cell subsets. Although Th2 cells are clearly able to deliver cognate activating signals to IFN-y stimulated macrophages in vitro (178), it appears that it is primarily Thl cells that are responsible for macrophage activation and the development of cell-mediated responses in vivo (155). Indeed, Th2 cells can produce a number of cytokines including IL-4 and IL-10 that are potent inhibitors of macrophage activation (155, 179-181). Both of these molecules need to be present at the initiation of macrophage activation and are thought to mediate their effects largely by abrogating the production of TNF-a (179, 180, 182). Transforming growth factor (TGF)-p which is produced by many different cell types including tumour cells, can also inhibit macrophage activation (181). This molecule, however, does not appear to inhibit TNF-a production, but rather abrogates the ability of IFN-y to prime macrophages (183). The simultaneous requirements for effective activation signals and competent macrophages at each stage in the activation process forms the basis of the control mechanism regulating macrophage effector function. Both the responsiveness of macrophages to activation signals and the final cytolytic state are short-lived functions which are rapidly lost with time and cannot subsequently be regained (184-187). Thus activated macrophages will only persist if activation stimuli and responsive macrophages are constantly replaced. 19 4.2 Macrophage cytotoxic activity Activated macrophages may exhibit a spectrum of cytotoxic reactions against tumour cells in vitro ranging from transient inhibition of cell growth to irreversible lysis (188-190). The final outcome of the interaction between macrophages and tumour cells in vitro presumably depends on the state of activation of the macrophage population* the sensitivity of the tumour targets and the conditions employed in the assay system. It is clear that within a population of activated macrophages not all of the cells are able to lyse tumour targets (153, 147). Moreover, the expression of cytotoxic activity is not a property of cells sharing particular physical, morphological or biochemical characteristics but varies with the system under investigation. In some cases, small cells appear to be particularly cytotoxic (191), while in others, large cells are more effective (192, 193). The cytotoxic activity of activated macrophages is immunologically non-specific and such cells will efficiently kill a variety of syngeneic, allogeneic and xenogeneic neoplastic targets while leaving non-neoplastic cells unharmed (194). The mechanism by which activated macrophages discriminate between neoplastic and non-neoplastic cells is not clear. Macrophage-mediated cytoxicity, at least in vitro appears to occur independently of tumour cell immunogenicity, drug sensitivity, invasiveness and metastatic potential (194-196). For example, B16 melanoma variant cell lines of high or low metastatic potential which were either susceptible or resistant to syngeneic T cell-mediated lysis were all equally susceptible to lysis in vitro by lymphoidne-activated macrophages (194). Similarly cell lines derived from the UV-2237 fibrosarcoma which differed greatly in invasive and metastatic potential in vivo (197), as well as immunogenicity and susceptibility to NK cell-mediated lysis (198) were all killed by activated macrophages in vitro (195). Attempts to isolate a tumour cell variant line resistant to macrophage cytolysis by repeated selection of the small percentage of B16 melanoma or UV-2237 fibrosarcoma target cells which survived exposure to activated macrophages in vitro proved unsuccessful (195). Cell lines resistant to cytolysis by T cells (B16) or NK cells (UV-2237) were readily obtained by selection with the appropriate effector cells. However, all these variants were still susceptible to macrophage-mediated cytolysis. These studies suggest that cytolytic macrophages recognize some tumour cell 20 characteristic which is intimately linked to the malignant phenotype and distinct from that recognized by T cells or NK cells. Macrophage-mediated cytolysis has generally been shown to be contact dependent, although phagocytosis does not seem to be involved (199, 200). Macrophages activated in vivo by BCG infection or pyran copolymer (201-203) or in vitro by lymphokines (201, 204) exhibited strong selective binding to neoplastic target cells. Examination of the interface between activated macrophages and attached targets by transmission electron microscopy revealed a zone of very close apposition between the two cells (202, 205). In contrast, the binding between non-activated macrophages and neoplastic target cells was weak, low-level and mostly non-specific (201-204). Macrophages can also adhere to certain normal cell types including T cells (206) and endothelial cells (77), although such adhesion does not generally trigger phagocytosis or cell killing. They can, however, bind and phagocytose cells undergoing apoptosis and have thus been suggested to play an important role in embryogenesis, tissue remodelling, thymic education, and in the clearance of dead and dying granulocytes from sites of inflammation (207). Target cell binding, although necessary, is not in itself sufficient for macrophage-mediated cytotoxicity. Macrophages primed in vivo by BCG or in vitro by lymphokines bound neoplastic target cells efficiently but did not lyse them until the macrophages were pulsed with traces of LPS (208,205). Similarly, activated macrophages cultured overnight lost the capacity to lyse tumour cells although they were still able to efficiently bind the target cells (208). A number of biologically active substances secreted by activated macrophages have been proposed to mediate the cytotoxic activity of these cells (209). Particular interest in this regard has focused on TNF-aand nitric oxide (NO) (210-212). Although activated macrophages secrete high levels of TNF-a (213, 214) and may also express a 26 kD membrane bound form of the molecule (215-217), it is clear, that not all signals that induce macrophages to secrete TNF-a also induce these cells to become cytotoxic, nor are all cell lines that are sensitive to macrophage-mediated killing also killed by TNF-a, even when the cytokine is added at very high concentration (218). In independent studies, Hibbs et all. (219) demonstrated that the ability of activated macrophages to induce tumour cell cytostasis required the presence of L-arginine. Subsequently, it was shown that 21 this amino acid is required for the synthesis of nitrate/nitrite by various mammalian cells including macrophages (220). NO was shown to be an essential intermediate in this process (220, 221). Moreover, NO alone was found to be directly cytostatic for tumour cells (221) and analogues of L-arginine or inhibitors of NO synthase (222) consistently inhibit or abolish macrophage-mediated cytostasis (222, 223). NO appears to mediate its effects via the nitrosylation of iron-sulfur centers in enzymes that become inactivated when iron is released, reducing the sulfur to iron ratio (224, 225). These enzymes include the oxidoreductases of complexes I and II of the electron transport chain, aconitase which is involved in the Krebs citric acid cycle, and ribonucleotide reductase, the rate limiting enzyme for DNA synthesis (219, 221, 226). NO clearly plays an important role in macrophage-mediated cytostasis, but its role in target cell lysis is not established (221, 227). Although the murine mastocytoma cell line P815 is strikingly sensitive to killing by NO (228, 229), many other lines are resistant (209). Thus the broad tumouricidal activity of activated macrophages is difficult to reconcile with the limited number of target cells that appear sensitive to TNF-a and NO. One possibility is that both these molecules may interact to induce macrophage-mediated killing. Indeed, Higuchi et al. (229) demonstrated the almost complete inhibition of tumour cell cytotoxicity by murine macrophage hybridomas in the presence of both L-arginine free medium and anti-TNF antiserum (229). Depending on the tumour target used, either treatment alone was generally very much less effective. Similar results have been obtained using BCG-activated macrophages (211). These findings have promoted Higuchi and co-workers to propose that the damage caused to biochemical pathways by NO may increase the sensitivity of target cells to TNF-a-mediated killing (229). In this regard it is interesting to note that mitochondria have been suggested to be important targets of TNF-a-mediated effects based on morphological evidence of damage early in cytotoxic responses (230), impaired respiration in TNF-a-treated cells (230-232) and the protective effects of Manganese-dependent superoxide dismutase (Mn-SOD) (233). A number of other molecules, in addition to TNF-a and NO, have been implicated in macrophage killing. Activated macrophages have been shown, for example, to secrete large amounts of lysosomal enzymes (234). Hibbs (153) observed that lysosomal material was 22 transferred from activated macrophages to attached tumour cell targets and proposed that secreted lysosomal enzymes were producing the resultant cytolytic effect, probably through their hydrolytic activity. However, the role of lysosomal enzymes in target cell killing has been questioned by the finding that the addition of even quite high concentrations of lysosomal enzymes to culture medium usually enhanced rather than inhibited tumour cell growth (235). Clearly if lysosomal enzymes mediate macrophage cytotoxic activity, it must be by a mechanism other than a direct attack on the exterior surface of the target cell. Activated macrophages also secrete reactive oxygen metabolites (236). Several lines of evidence suggest that these may play an important part in the killing of tumour cell targets. Firstly, target cells are susceptible to lysis by oxygen metabolites in the amounts released from activated macrophages in response to phorbol myristate acetate (PMA) (237). Secondly, deprivation of either oxygen or glucose prevents PMA-induced oxygen metabolite release from activated macrophages and greatly reduces target cell killing (203, 238). Finally, exogenous scavengers of oxygen metabolites such as catalase or thioglycollate broth, abolish PMA-induced target cell killing by activated macrophages (238, 239). In other studies, (240-242) a cytotoxic principle present in culture supernants conditioned by activated macrophages has been characterized as a neutral serine proteinase with a molecular weight of approximately 40,000 daltons. However, it should be noted that inhibitors of neutral proteinase activity may exert numerous effects including the suppression of reactive oxygen metabolite release by activated macrophages (243). Activated macrophages also secrete neutral proteinases and complement component C3 (143). Neutral proteinases can cleave C3 to give C3a and C3b fragments. It has been suggested that C3a is cytotoxic for tumour target cells (244). However work by Goodman et al. (245) has cast serious doubt on the cytotoxic capacity of C3a and clearly much more work needs to be carried out to clarify the importance, of C3a in macrophage-mediated target cell killing. Another suggestion is that arginase is responsible for the cytolytic capacity of activated macrophages (246). Arginase was shown to be released by macrophages activated in vitro by 23 zymosan or LPS but not by normal macrophages (246). It is presumed that arginase exerts its cytolytic effect by preventing the target cell from obtaining sufficient arginine (246). Stadecker et al. (247) suggested that the nucleoside thymidine is responsible for target cell killing. Thymidine is synthesized and secreted by activated macrophages (247) but few target cells are likely to be susceptible to lysis at the concentrations attained. In summary, it appears that lysis of target cells by activated macrophages occurs in at least three stages. In the first stage, activated macrophages bind to target cells. This leads to the establishment of an area of contact between the two cells and the triggering of metabolic events within the macrophage. In the second stage, macrophages secrete lytic effector substances into the region of contact. It is likely that several such substances are secreted and the relative importance of each may vary depending on the target cells or environmental conditions. The contact area presumably acts to concentrate these lytic substances and protect them from inhibitors present in the extracellular compartment. Finally, the third stage involves target cell injury and eventual loss of viability. 4.3 Macrophage-induced stimulation of tumour cell proliferation It is apparent that under some conditions, immune reactions may stimulate rather than inhibit the proliferation of tumour cells (248). Normal peritoneal macrophages have been shown to enhance the growth of lymphoid tumour cells which, by themselves, grew poorly in vitro (249-251). Similarly, Olivotto and Bomford (252) reported that cultured murine peritoneal macrophages stimulated 3H-thymidine uptake by tumour cells. Moreover, enhancement of tumour cell proliferation is not limited only to resident macrophages. Krahenbuhl et al. (253) showed that peritoneal macrophages from animals injected with Corynbacterium parvum or killed Toxoplasma gondii greatly enhanced 3H-thymidine uptake by L-929 cells at low effector to target cell ratios. Similar results have been reported for proteose-peptone activated macrophages (254). The in vivo significance of the tumour growth-promoting activity of macrophages remains to be elucidated. Hewlett et al. (251) suggested that macrophages enhance tumour cell proliferation in vitro simply by "improving" culture conditions. However, Prehn (248) proposed that 24 immunostimulation of tumour cell proliferation plays an important part in the early stages of tumour development in vivo. In this connection it is interesting to note that in one study treatment of mice with the anti-macrophage agent silica was reported to inhibit the growth of subcutaneously implanted Lewis lung carcinoma cells (255). 5.0 Tumour-associated macrophages 5.1 Correlations between macrophage infiltration and prognosis Pathologists have long been aware that most solid tumour masses, irrespective of histological type, contain a significant infiltrate of various host-derived cells, in particular macrophages. Indeed, initial reports demonstrating the presence of such cells first appeared during the latter half of the 19th century (256). At that time, the gross similarity between the tumour-associated infiltrate and that present within sites of chronic inflammation gave rise to the belief that tumours develop at sites of previous chronic inflammation. Handley (257) was the first to suggest a potentially beneficial role for the patient of the mononuclear cell infiltration of tumours. He studied a single patient with disseminated melanoma and proposed that "the abundant round cell infiltration and the absence from this region of a network of small permeated lymphatics are clear indications that reparatory processes, inadequate for cure, are not wanting in melanotic sarcoma." Subsequently, many groups have attempted to establish a correlation between mononuclear cell infiltration of tumours and prognosis. MacCarty (258), for example, investigated the relationship between various histologically recognizable tumour features and postoperative survival in a group of 293 cases consisting of cancer of the stomach, breast and rectum. He reported a three-fold increase in postoperative survival provided that immune cell infiltration, tumour cell differentiation, fibrosis and hyalinization were all present within the primary tumour. Immune cell infiltration alone was also associated with improved survival, although to a lesser extent. In another retrospective study, MacCarty and Mahle (259) investigated the postoperative life span of 200 gastric cancer patients staged as to degree of spread. The results obtained also indicated that for the small number of patients studied with limited glandular involvement there was 25 a marked survival advantage if tumour cell differentiation and mononuclear cell infiltration were present. Underwood (256) similarly reviewed over 30 publications dealing with non-lymphoid tumours which have appeared since 1921, and once again noted a positive association between lymphoreticular infiltration' and an improved prognosis, although there were several exceptions. One of the strongest arguments in favor of an important role of mononuclear cell infiltration in tumour growth comes from studies on medullary carcinoma of the breast. These tumours display highly "malignant" cytological features and, in general, would be expected to have poor long-term survival rates. Moore and Foote (260) identified 52 cases of medullary carcinoma in a study of 1,000 patients who had undergone radical mastectomy. In most of these, the tumours were shown to contain a dense lymphoid and macrophage infiltrate. Moreover, as a group, medullary carcinoma patients had a much higher 5 year survival rate compared to patients with other types of breast cancers. Only 6 (11.5%) died within 5 years, a figure approximately five times better than was seen at the time for other breast cancers. In addition, although 22 (42.3%) of these patients had axillary metastasis at operation, 11 (50%) were well after 5 years compared with a 27% 5 year survival in other types of breast cancer with axillary involvement. Unfortunately, this study was retrospective and used historical controls for survival that appear at odds with more recent and extensive studies (261). This early work is, however, important because the majority of the patients did not receive any therapy other than surgery (unlike most more recent investigations) which may, in turn, have influenced their outcome. In more recent studies, monoclonal antibodies have been used to better characterize the nature of the host cell infiltrate present within spontaneously arising human solid tumours. In particular the macrophage content of human breast, colorectal and cervical cancers (262-265), melanomas (266) and brain tumours (267) has been extensively studied. The total number of macrophages, and of activated macrophages expressing CR1, infiltrating breast and colorectal cancer appears to be significantly increased in comparison to normal tissues, particularly in more advanced stages of the disease (262, 264). In contrast, the number of T cells did not differ significantly between normal and malignant tissue. Similar results were also obtained in melanoma 26 (266), suggesting that at least in some tumours, macrophages may actually assist rather than inhibit tumour growth. The precise localization of macrophages within the tumour mass also may be an important variable affecting outcome. Indeed, with respect to colorectal carcinoma, and in contrast to the studies described above, it has been suggested that the presence of increased numbers of macrophages at the invasive edge of a tumour indicates more favorable prognosis (268). These data indicate further that the macrophages present at the tumour margin may differ functionally from those found within the body of the tumour. In summary then, the relationship between macrophage infiltration and tumour prognosis still remains unclear. Although most studies suggest that the presence of large numbers of macrophages correlates with improved prognosis, there are many exceptions. Some authors who have been disappointed by inconsistent results have rejected the concept entirely (269-271). However, it would indeed be surprising if it were always possible to correlate a complex biological phenomenon such as survival with any single characteristic in all tumours. In particular, given the fact that macrophages can perform functions that can both inhibit and promote tumour growth, it may be necessary to consider not only total number of macrophages present within a tumour mass, but also the number of each functionally distinct macrophage subset. 5.2 Correlations between macrophage infiltration and tumour metastasis Evidence that mononuclear cell infiltration of tumours influences their capacity for metastasis is circumstantial. It has been a frequent finding in experimental models that tumours with the lowest incidence of metastases tend to contain the highest numbers of host cells, particularly macrophages. Such a correlation has been demonstrated for rat sarcomas (272), murine sarcomas (273), murine lymphomas (274) and rat gliomas and sarcomas (275, 276). There are, of course, also many exceptions to this correlation. For instance, Eccles (277) found groups of Bittner factor-positive murine mammary adenocarcinomas and chemically-induced squamous cell carcinomas which showed little or no metastases and contained low to moderate numbers of macrophages. It must be remembered, however, that metastasis is a complex 27 biological phenomenon which depends on both the intrinsic characteristics of the tumour cells as well as the host immune response. Presumably in some circumstances tumour characteristics alone may prevent or limit the possibility of spontaneous metastases developing (278). Thus, while low levels of infiltrating cells are not necessarily associated with tumours of high metastatic potential, there are few examples in animal studies, of tumours containing a dense host cell infiltrate successfully producing high incidences of overt metastasis. Although these observations do not establish causality they are strongly suggestive of a role for tumour-associated host cells, particularly macrophages, in reducing metastasis. Although no mutant murine strain or human completely deficient in macrophages has yet been described, a number of agents selectively toxic for macrophages have been identified and these have been widely used to study the effect of macrophage depletion in vivo. Although the reduction in macrophage number and/or function achieved in these studies was both modest and transient, the results obtained support an important role for macrophages in inhibiting tumour growth and metastasis in vivo. Thus, treatment of mice with the macrophage toxins carageenan or silica was shown to shorten the period of tumour latency observed following exposure to UV irradiation (279). Similarly, in transplantable tumour systems, macrophage depletion, again using carageenan or silica, was found to increase the incidence of both spontaneous (280) and experimental metastasis (281). For some tumours, depletion of T cells in vivo has been shown to cause a considerable decrease in the number of tumour-associated macrophages and an associated increase in the incidence and/or distribution of metastases. This effect has been documented for rat sarcomas grown in animals T cell-depleted by thymectomy plus whole body irradiation (272) or Cyclosporin A treatment (282) and in nu/nu recipients (283, 284). Similar results were also obtained for murine sarcomas or lymphomas grown in congenitally athymic animals (277) or Cyclosporin A treated animals (282) and murine mammary adenocarcinomas grown in animals T cell-depleted by thymectomy and irradiation (285). Reconstitution of T cell-depleted animals restored macrophage number to the levels seen in control animals (285-287). Thus it appears that manipulation of host T cell responses can reduce both tumour-associated macrophage number and metastasis, suggesting 28 that both of these parameters may be related. It should, however, be noted that the processes used to alter the macrophage content of a tumour frequently alter other factors that might influence metastasis, in particular lymphocyte-mediated responses. Assuming that tumour-associated macrophages may be important in controlling metastasis, how might they produce such an effect? Certainly, if it were possible to prevent malignant cells from leaving the primary tumour site, metastasis would not occur. It is tempting to consider that high levels of tumour-associated macrophages, while failing to stop local growth may still prevent or inhibit tumour cell spread. One might then expect macrophages from highly immunogenic, non-metastatic tumours to be more cytotoxic than those from weakly immunogenic, metastatic tumours. In this regard, it is interesting that Mantovani (273) found that the highly immunogenic non-metastatic murine FS6 sarcoma contained more than twice the number of macrophages than its weakly immunogenic, spontaneously metastatic subline mFS6. In addition, while macrophages from the FS6 tumour inhibited the growth and DNA synthesis of tumour cells in vitro, macrophages obtained from the mFS6 tumour non-specifically enhanced the proliferative activity of tumour target cells. Opposite results have, however, been found in other tumour systems. For example, macrophages isolated from the highly immunogenic, non-metastatic rat HSBPA fibrosarcoma were poorly cytotoxic in vitro (272). In contrast macrophages from the poorly immunogenic, metastatic rat HSN and ASBP1 fibrosarcomas exhibited strong non-specific cytotoxic activity (288). Similarly, Loveless and Heppner (289) compared the macrophage content and function of a series of transplanted murine mammary tumours originally derived from a single spontaneously arising tumour and found no correlation between macrophage content and metastatic potential. Moreover, macrophages from every metastatic tumour used were found to be cytotoxic whereas only 6 out of 17 non-metastatic tumours contained cytotoxic macrophages. Berg (290) investigated the relationship between inflammation and survival in patients with large anaplastic breast tumours. He demonstrated that 73% of patients who survived more than 5 years after removal of the primary tumour had shown marked round cell (mainly plasma cell) infiltration at the junction of tumour and normal breast while only 30% of patients who survived less than three years had exhibited a similar infiltration. Berg concluded that the association 29 between good prognosis and inflammation was most likely secondary to the presence in tumours of a dense host cell infiltrate which tended to reduce the metastatic potential of the tumours. Lauder et al. (291) suggested that the close association of macrophages and tumour cells observed in primary breast carcinoma may have retarded tumour cell emigration and contributed to the lack of metastasis in those patients with prominent host cell infiltrates. Similarly, Pelouze and Bonenfant (292) showed that gastric carcinomas infiltrated by high numbers of host cells produced fewer nodal metastases and an increased duration of survival as compared to similar tumours infiltrated by low numbers of host cells (277). 5.3 Correlations between the macrophage content of tumours and regression The spontaneous regression of established primary neoplasms in humans is a relatively rare event. There have, however, been infrequent observations of spontaneous regression of malignant melanoma in which partial or total disappearance of primary tumours has been clearly documented. Tumour regression was frequently shown to be associated with the presence of a dense host cell infiltrate (270, 293), although other studies failed to find such an association (294, 295). Tumour cell imprint preparations from a regressing melanotic lesion revealed numerous large macrophages in juxtaposition to melanoma cells (296). Epstein et al. (297) reported a series of melanoma patients with skin metastases in whom removal of a primary melanoma was associated with the appearance of multiple prominent halo nevi. However, on this occasion lymphocytes and not macrophages were observed around typical degenerating melanoma cells. In two well established murine tumour systems, the mammary adenocarcinoma T1699 (298) and the Moloney virus-induced sarcoma (299, 300), spontaneous regression was associated with the accumulation of lymphocytes, macrophages and granulocytes. Both tumour-associated lymphocytes and macrophages exhibited direct cytotoxic activity in vitro. Moreover, macrophages from regressing T1699 tumours were also capable of killing tumour targets by antibody-dependent cell-mediated cytotoxicity (ADCC) (301). For some regressing tumours, histological studies have revealed a close association between infiltrating macrophages and tumour cells. For example, in a regressing transplanted rat tumour, 30 numerous macrophages were observed with long processes closely apposed to tumour cells, with all stages of phagocytosis of apparently intact tumour cells evident (302). Analogous appearances have been noted in regressing leukemic intradermal tumours and hepatomas (303, 304). If tumour-associated macrophages do have the potential to destroy neoplastic cells, then it would be logical to assume that experimental procedures which increase either their number, or functional activity might have an inhibitory effect on tumour growth. Indeed, treatment of tumour masses by intralesional injection of BCG or C. parvum is based on the assumption that these procedures will increase both the number and activity of tumour-associated macrophages. Moreover, such treatments have on occasion been shown to produce dramatic regression of animal (305, 306) and human (307-309) tumours. In addition, several studies have demonstrated that bacterial adjuvants injected intra-tumourally or admixed with a tumour cell inoculum greatly increased the macrophage content of certain rat or murine sarcomas and produced a decrease in the number or incidence of spontaneous metastases (277). Moore and McBride (310) reported that regression of a murine methylcholanthrene-induced fibrosarcoma following a single intravenous (i.v.) injection of C. parvum was accompanied by the appearance within the tumour of a population of small highly 'activated' macrophages. There are several reports demonstrating inhibition of both primary and metastatic tumour growth in mice following the adoptive transfer of syngeneic macrophages that had been activated in vitro using lymphokines and/or other stimuli (311-317). Human tumour xenografts growing in nu/nu mice were similarly shown to regress following i.v. administration of IFN-y-activated human monocyte-derived macrophages, (318) or human peritoneal macrophages obtained from renal patients receiving continuous ambulatory peritoneal dialysis (319). Cytotoxic macrophages have also been studied as potential mediators of regression of established human malignancies (320-322). In these initial studies, large numbers of mononuclear cells purified from the peripheral blood of patients by leukapheresis and further separated from contaminating erythrocytes and granulocytes by Ficoll-Hypaque gradient centrifugation were cultured for 7 days in hydrophobic Teflon bags in the presence of recombinant human IFN-y (200 u/ml). Activated macrophages present within the heterogeneous cell population obtained were then 31 purified by centrifugal elutriation and reinfused i.v., intraperitoneally (i.p.) or intrahepatically (i.h.) into patients. Approximately 13-79% of the number of monocyte/macrophages originally seeded were recovered following culture. The recovered cells expressed the maturation-associated surface molecules MAX, CD16, CD51 and CD71 and were cytotoxic for allogeneic tumour cells targets in vitro. They also secreted large quantities of TNF-a, IL-6 and GM-CSF following stimulation with LPS (321). In a Phase 1 clinical trial, 15 patients with a variety of malignancies were treated by the i.v. or i.p. administration of these activated macrophages. The adoptively transferred cells were well tolerated although side effects included low grade fever, induction of intravascular coagulation, and abdominal discomfort (following i.p. administration). High levels of IL-6 could be detected in the serum and ascitic fluid of treated patients. In three of seven patients with malignant ascites that received macrophages i.p., a clinical response consisting of the disappearance of tumour cells and a substantial decrease in the level of tumour-associated biochemical markers present within the ascitic fluid were noted. Further studies with larger patient groups and appropriate controls will be required to established the clinical potential of this approach. Liposomes constitute a convenient carrier to target biologically active materials to macrophages. Following i.v. administration, the vast majority (80-90%) of liposomes are taken up by macrophages (323). Moreover, liposome-encapsulated mediators of activation such as lymphokines or the synthetic agent N-acetyl-L-alanyl-D-isoglutamine (MDP) are able to induce cytotoxic activity in macrophages in vivo and in vitro (324). It has been reported that multiple i.v. injections of liposomes containing lymphokines (323, 325-328) or MDP (326, 328-330) resulted in significant destruction of established murine melanoma lung metastases. It was assumed that the mechanism of this effect involves the localization of activated macrophages within the metastatic lesions (324). Obrist and Sandberg (331) reported that the number of macrophages infiltrating peritoneal hepatomas was significantly elevated by the i.v. injection of a covalent conjugate of the chemotactic peptide, FMLP and a monoclonal antibody reactive with cell surface antigens 32 expressed by the hepatoma cells. This is a particularly interesting study as it suggests an approach whereby macrophages can be specifically attracted to a tumour mass. Whether accumulation of host cells within a tumour actually causes regression or is only circumstantial is often not clear. Treatment of sarcoma-bearing mice with daily azathioprine injections resulted in an increase in the proportion of tumour-associated T cells and macrophages. However, individually, neither macrophages nor lymphocytes were cytotoxic in vitro (332). Thus it remains questionable as to whether the observed tumour regression was the result of cell-mediated cytotoxicity or of drug action alone. Similarly, it was recently reported that during cyclophosphamide-induced tumour regression, there was an accumulation and infiltration of macrophages throughout the tumour mass (333, 334). However, it was apparent in these studies that although the tumour mass became smaller, there was an actual increase in the number of tumourogenic cells within the regressing mass and most tumours eventually showed renewed growth (335). In summary, it seems that although tumour regression is commonly accompanied by a dense host cell infiltrate particularly macrophages, there is insufficient evidence at present to suggest that such host cells always cause the regression. 6.0 Accumulation of macrophages in tumours It is generally accepted that although different tumours may contain different numbers of macrophages, once established, the macrophage content of a given tumour remains fairly constant during passage from one syngeneic host to another (336-338). Thus, while primary chemically-induced tumours initially often contain a high percentage of macrophages, this number typically decreases sharply during early passage, eventually leveling off at a figure which may be maintained through many subsequent in vivo passages (336-338). In this connection, it is interesting to note that the immunogenicity of chemically-induced tumours may also decrease during early passage (339-341). The most likely explanation of these results is that repeated transplantation selects tumour cells particularly well adapted to growth under the set conditions that 33 transplantation imposes and that such cells elicit, by some well regulated process, a characteristic host cell infiltrate (338). The host cell origin of tumour-associated macrophages has been demonstrated in several studies. Tumour cell suspensions depleted of macrophages prior to inoculation into syngeneic hosts by treatment with anti-macrophage serum and complement (336) or by long-term in vitro culture (336, 342, 343) still developed a characteristic macrophage infiltrate, suggesting that tumour-associated macrophages were not derived from cells included in the tumour inoculum. Confirmation of the host origin of tumour-associated macrophages was obtained by growing parental strain tumours in Fl hybrid hosts (337, 342, 344, 345). For example, CBA (H-2k) methylcholanthrene-induced fibrosarcoma cells were injected into (C3H x DBA/2)F1 (H-2k x H-2d) recipients. Host cells infiltrating the tumour expressed H-2k and H-2d antigens since histocompatibility antigens are codominant, and were identified with anti-H-2d antisera. Tumour cells on the other hand expressed only H-2k antigens and did not react with anti-H-2d antisera (342). The current dogma concerning the origin of macrophages in normal tissues or inflammatory sites suggests that these cells are derived mainly from blood monocytes, although in some circumstances, in situ macrophage proliferation may also occur. The accumulation of macrophages within tumours presumably occurs by the same mechanism. Influx of monocytes into a tumour site undoubtedly occurs (346, 347). However, identification of recently arrived monocytes in tumour cell suspensions by peroxidase staining, revealed very little staining that could not be accounted for by neutrophil contamination (348). This implies a very rapid differentiation of tumour-infiltrating monocytes into macrophages (349). The extent of in situ macrophage proliferation is more uncertain. Following i.v. injection of 3H-thymidine, labelled macrophages were found in situ (350). In addition, in the presence of CSF, tumour-associated macrophages may form colonies in vitro (351). However, preliminary data on labelling indices following i.v. injection of 3H-thymidine were shown to be too low to account for the observed increase in macrophage number as the tumour grows (348). In addition, tumour cell suspensions pulsed with 3H-thymidine in vitro showed a similar low macrophage labelling index (348). In conclusion, it seems likely that 34 most tumour-associated macrophages are derived from an influx of blood monocytes into the tumour site although some in situ macrophage proliferation may also occur. The relative contribution of each of these mechanisms presumably varies between different tumours. Some tumours have been shown to produce CSF (352) and with these, local macrophage proliferation may contribute significantly to the total macrophage content. Numerous authors have investigated the degree of macrophage infiltration of tumours as a function of time after injection of tumour cells. It appears from these studies that the kinetics of macrophage infiltration may also vary from one tumour to another. In some, the percentage of macrophages peaked early in the growth of the tumour before declining to a level which was maintained throughout the remaining period of growth (343, 353). With other tumours, the same percentage of macrophages was found throughout the period of growth (336). In contrast, the macrophage content of several sarcomas and carcinomas has been shown to remain relatively constant until a critical tumour size is reached, after which time it decreased substantially, perhaps reflecting diminished infiltration of blood monocytes (343,353-355). Several possible mechanisms may contribute to the accumulation of macrophages in tumours. 6.1 Mediators of acute inflammation Monocyte migration has been shown to occur in response to factors released during imflammation. Many factors have been shown to have this effect in non-tumour systems e.g. bacterial iV-formyl peptides, C5a, and thrombin, as well as growth factors and cytokines released from platelets, neutrophils, lymphocytes and monocytes themselves e.g. TGF-p, PDGF, TNF-a, TNF-p, GM-CSF, platelat-activating factor and leukotriene B4. However, Evans (336) reported no apparent correlation between the macrophage content of tumours and the degree of necrosis or neutrophil infiltration. In addition the anti-inflammatory drug indomethacin failed to affect the macrophage content of a variety of rat sarcomas and carcinomas (356). Similarly, the non-steroidal anti-inflammatory agent flubriprofen which effectively inhibits prostaglandin synthesis did not alter either the migration of labelled monocytes into rat sarcomas or the number of macrophages found 35 within these tumours (277). Thus, on the basis of this evidence, it seems unlikely that macrophages are attracted to a tumour site by mediators of acute inflammation. 6.2 Immune mechanisms It has been reported that the levels of macrophages associated with tumours appears to be related to the immunogenicity of tumour cells, such that highly immunogenic tumours have a relatively high macrophage content, and weakly immunogenic tumours a low content (272, 275, 336, 343). Such correlations were most clearly seen when tumours of similar aetiology, histology and host origin were compared (272). Of particular interest in this respect are pairs of tumours where the macrophage content of an immunogenic line and its less immunogenic derivative were compared. For example, the macrophage content of the strongly immunogenic benzo(a)pyrene-induced sarcoma FS6 was around 44% while the weakly immunogenic mFS6 subline contained only 19% (273). Similar results were obtained using the L5718Y Eb and ESb lymphomas (357). The correlation between tumour immunogenicity and macrophage content strongly suggests that macrophage accumulation is sustained by stimuli which are immunological in nature. This hypothesis is supported by studies which demonstrated a significant reduction in the macrophage content of various sarcomas and carcinomas grown in T cell depleted (272, 285, 358) or nude (284, 359) animals. In addition, administration of Cyclosporin A, a potent immunosuppressive agent with selective inhibitory effects on T cell (360) decreased both the migration of labelled monocytes into rat sarcomas and the number of macrophages found within these tumours (277). However, the suggested involvement of specific anti-tumour immune responses in the accumulation of macrophages in tumours is not supported by all studies. Evans and Lawler (361) for instance, found no correlation between immunogenicity and macrophage content in an extensive study of 33 primary methylcholanthrene-induced murine sarcomas. In addition, Szymaniec and James, (362) found that the levels of tumour-associated macrophages in a murine sarcoma were unaffected by transplantation into thymectemized animals. Similar findings have 36 recently been reported for other methylcholanthrene-induced murine sarcomas transplanted into T cell depleted (353, 361) or nu/nu (348) mice. Thus it appears that for some, but not all, tumours, there is a good correlation between immunogenicity and macrophage content. However, this is by no means always the case, and for other tumours, different mechanisms may be more important. More studies are needed before it can be ascertained whether the relationship between tumour immunogenicity and macrophage content applies to most, or to only a few tumours. 6.3 Tumour-derived chemotactic factors The accumulation of leukocytes at inflammatory foci is part of the host defense response to many noxious agents. One of the mechanisms which leads to such localization is the elaboration of molecules chemotactic for leukocytes. These factors diffuse out from the site of release and leukocytes then migrate up this concentration gradient to form the inflammatory focus. Neutrophils, basophils, eosinophils and macrophages all respond to various chemotactic stimuli. Factors which preferentially attract a particular type of leukocyte have been described. Thus the nature of chemotactic factors may determine the cellular composition of inflammatory foci. There have been a few reports indicating that tumour cells may elaborate factors that are chemotactic for macrophages. Meltzer et al. (363) found that culture superaatants from 4 different murine methylcholanthrene-induced fibrosarcomas contained factors that were chemotactic for resident and BCG-activated peritoneal macrophages but not for peritoneal granulocytes. Activated peritoneal macrophages were more responsive to the tumour-derived chemotactic factors than were resident peritoneal macrophages. Minimal levels of chemotactic activity were found in supernatants from syngeneic murine embryo fibroblast cultures. Similarly, in an extensive study Bottazzi et al. (364) found chemotactic activity in the culture supernatants of most murine and human sarcomas and carcinomas irrespective of their histology or origin (chemically-induced or spontaneous). Leukemias and lymphomas on the other hand were either inactive or elaborated a low chemotactic activity. The elaboration of chemotactic factors by human tumour cells in vitro was confirmed by Peri et al. (365) who demonstrated that culture supernatants from the human sarcoma cell line 8387 37 and the human melanomas 1080/2 and SKBR-3 were chemotactic for human monocytes whereas the K562, Raji and CEM lymphoma lines produced little chemotactic activity. Several tumour cell-derived factors chemotactic for monocytes have recently been identified. Among the best characterized of these is monocyte chemotactic protein-1 (MCP-1). MCP-1 is also known as tumour-derived chemotactic factor (TDCF) and monocyte chemotactic and activating factor (MCAF) (366-369) and constitutes the human homologue of the murine gene JE (370). It is a member of the chemokine family and is chemotactic for monocytes but not lymphocytes or granulocytes (371). It interacts with a receptor that is coupled to a pertussis toxin-sensitive G-protein to induce a rapid increase in intracellular free calcium (372). Supporting an important role for this molecule in the recruitment of tumour-associated macrophages, initial studies demonstrated a correlation between the level of MCP-1 produced by certain tumour cells and their macrophage content (373,374). Clones of a murine sarcoma that differed in their expression of MCP-1 mRNA were similarly shown to exhibit corresponding differences in their macrophage content (375). Moreover, transduction of CHO cells or murine melanoma cells with the MCP-1 gene greatly increased the number of macrophages present within the tumours produced from these cells following inoculation into either nu/nu or syngeneic recipients (376,377). MCP-1 is, however, not the only factor involved in regulating the accumulation of tumour-associated macrophages, and in some studies no obvious correlation was observed between MCP-1 production and the macrophage content of a tumour. Two additional tumour-derived chemotactic proteins related to MCP-1, designated MCP-2 and MCP-3, have recently been identified although their role in the recruitment of tumour-associated macrophages remains to be determined (378). Macrophage colony-stimulating factor (M-CSF) and Granulocyte-macrophage colony-stimulating factor (GM-CSF) are produced by certain tumour cells and can act as specific chemoattractants (379, 380). Several different tumour cells have been shown to produce a 46kD protein known as vascular permeability factor (VPF) or vascular endothelial growth factor (VEGF) (381) which is related to platelet-derived growth factor (PDGF), and is chemotactic for monocytes in vitro. 38 6.4 Tumour-derived chemokinetic factors Several authors have described tumour cell-derived factors that can act on macrophages in vitro to increase their rate of random translational movement (chemokinesis). For example, Meltzer et al. (188) using time lapse cinemicrographic analysis observed that the rate of translational movement of BCG-activated murine peritoneal macrophages co-cultured with neoplastic cells was four times that observed if the macrophages were co-cultured with non-neoplastic cells derived from the same cloned embryo cell line. The translational movement of resident peritoneal macrophages, on the other hand, was less than that of activated macrophages and was unaltered by the addition of neoplastic or non-neoplastic cells. Snodgrass et al. (382, 383) obtained similar results for resident and pyran copolymer-activated murine peritoneal macrophages co-cultured with Lewis lung carcinoma cells. In their studies, activated macrophages alone moved almost three times faster than resident macrophages. The rate of translational movement of resident macrophages was unaltered by co-culture with neoplastic Lewis lung carcinoma cells or with non-neoplastic murine embryo fibroblasts. Activated macrophages on the other hand moved three times faster when coincubated with, but not actually in contact with, carcinoma cells. However, when such a macrophage came into contact with a tumour cell, its rate of lateral movement slowed to one comparable with activated macrophages cultured alone. In contrast, activated macrophages co-cultured with non-neoplastic murine embryo fibroblasts showed no alteration in their rate of translational movement. These results suggest that neoplastic cells elaborate chemokinetic factors which act to increase the rate of translational movement of activated macrophages. Indeed, Snodgrass et al. (382, 384) and Schuller et al. (385) have shown that addition of cell-free culture supernatant from Lewis lung carcinoma cells greatly enhanced the translational movement of pyran copolymer-activated murine peritoneal macrophages while supernatant from primary murine embryo fibroblast cultures had no effect. The rate of translational movement of resident peritoneal macrophages was either unaltered (382, 384) or inhibited (385) by addition of supernatant from Lewis lung carcinoma cell cultures. 39 In a study, Lane et al. (386) found that culture supernatants from six different syngeneic and allogeneic murine tumour cell lines all increased the rate of translational movement of Corynebacterium parvw/w-activated murine peritoneal macrophages while supernatants from normal murine embryo fibroblast cultures had no effect The chemokinetically active component of Lewis lung carcinoma conditioned medium was partially characterized and found to be a trypsin-sensitive, heat-stable, high molecular weight (300,000-400,000 dalton) factor that exhibited no chemotactic activity. The factor increased the rate of translational movement of C. parvum and pyran copolymer-activated macrophages but had no effect on oyster glycogen- or thioglycollate-elicited macrophages. Reports on the molecular size of chemokinetic factors isolated from other tumours have been rather vague. Usually they were described as non-dialyzable indicating a molecular weight of greater than 12,000 daltons (387). Since all of the tumour-derived chemokinetic factors which have been described have a molecular weight above 12,000 daltons it is possible that all these molecules are related. Several factors with chemokinetic activity have been isolated from other sources. Aaskov and Anthony (388) described a high molecular weight (> 150,000 daltons) factor in culture supernatants of BCG-stimulated human peripheral blood lymphocytes which enhanced the movement of normal murine splenic macrophages in the macrophage inhibitory factor (MIF) assay but had no chemotactic activity. A similar factor was found in supernatants of PPD-stimulated lymphocytes (389). This factor was heat-stable, had an electrophoretic mobility in the gamma-globulin region and increased the random movement of human buffy coat leukocytes from a well cut in agarose gel. Thus it appears that the physiochemical features of tumour-derived chemokinetic factors are not unusual when compared with factors from other sources. At present however, the precise nature of chemokinetic factors remains to be elucidated. In this connection, it is interesting to note that cell coat glycocalyx glycoproteins have a molecular weight range of 200,000 to 500,000 daltons (390, 391) which is fairly similar to that suggested for chemokinetic factors. 40 In summary, it appears that at least in vitro, neoplastic cells may elaborate factors which can enhance the random movement of activated macrophages while having little or no effect on non-activated macrophages. If these factors are produced in vivo and have similar effects on monocytes, then they could served to specifically increase the rate of migration of activated cells out of the vascular system in response to an appropriate stimulus, while having little or no effect on non-activated cells. In this way such factors might influence not only the macrophage content of tumours, but also their activity. 6.5 Depression of macrophage chemotactic and inflammatory responses by tumour-derived factors Patients with advanced malignant disease are often unable to generate delayed type hypersensitivity (DTH) reactions to antigens to which they had previously been sensitized (392) or to produce normal cellular exudates in response to mild skin abrasions (393) or inflammatory stimuli (394). Equivalent results have been obtained in animal studies using both spontaneously arising (395, 396) and transplanted (355, 397-400) tumours. In most instances the anergy associated with tumour burden appears to be the result of a macrophage rather than a lymphocyte defect. Lymphocytes from tumour-bearing animals have usually been found to manifest normal in vitro responses to both antigens and mitogens (401). On the other hand, macrophage chemotactic responsiveness in vitro may be greatly reduced (402-408), and such inhibition only becomes evident late in tumour progression and rapidly disappears if the tumour is excised (409). The most likely interpretation of these results is that tumours elaborate factors which inhibit the response of macrophages to chemotactic stimuli. Cell-free homogenates of transplanted tumours (405,410, 411), tumour ascitic fluid (412), plasma or urine from tumour-bearing animals (413), and tumour cell culture supernatants (414, 415) have all been shown to inhibit macrophage chemotaxis in vitro and/or depress macrophage accumulation at inflammatory sites in vivo. Similarly, Cianciolo et al. (416) found inhibitory activity in all of 17 malignant effusions from patients with various types of neoplasms, whereas 41 effusions from 17 patients with non-malignant diseases possessed no significant inhibitory activity. The physicochemical nature of tumour-derived anti-chemotactic anti-inflammatory factors has been investigated in several systems. Pike and Snyderman (405) identified an anti-inflammatory factor with a molecular weight of between 6,000-10,000 daltons in sonicates of 4 histologically distinct tumours. Cianciolo et al. (396) found a similar factor in spontaneously arising murine carcinomas. Nelson and Nelson (387,397) reported that murine tumour cell culture supernatants contained both low (10,000 daltons) and high (>10,000 daltons) molecular weight factors which could inhibit the early phase of a prolonged DTH reaction in vivo and macrophage migration and motility in vitro. Similarly, Norman and Cornelius (415) found that the P815 mastocytoma and a murine methylcholanthrene-induced fibrosarcoma both produced at least two soluble anti-inflammatory factors that inhibited macrophage accumulation in vivo when injected into syngeneic recipients. One factor was a low molecular weight peptide (<1,000 daltons) while the other had a molecular weight of between 30,000-100,000 daltons. Cianciolo et al. (396) found that the anti-chemotactic factor present in human malignant effusions was a polypeptide of molecular weight 15,000-70,000 daltons. Interestingly this factor also reacted with monoclonal antibodies against P15E, a structural protein of type C retroviruses (416). Subsequently, it was demonstrated that purified P15E itself functioned as a potent anti-inflammatory agent when injected into normal mice and that both human and murine tumours produced P15E-related antigenic material (417, 418). Moreover, antibodies against P15E were shown to abrogate the anti-inflammatory activity of bovine tumour extracts (419) and that growth of transplanted murine tumours could be significantly delayed in vivo by treatment with monoclonal antibodies against P15E (419). Lindvall and Sjogren (420) reported that antibodies against P15E also inhibited the growth of rat yolk sac tumours. Peptide mapping studies identified a 26 amino acid domain that appears to be responsible for much of the biological activity of P15E (421). Interestingly, this sequence is conserved within a large number of murine, feline, bovine, avian, simian, and even human retroviruses (421). More importantly perhaps, there appears to be significant homology between the P15E peptide and a 42 region present within the immunoregulatory protein TGF-p (422). A peptide corresponding to this region of TGF-p is able to mimic the immunosuppressive activity of TGF-p (422). Moreover, at least one monoclonal antibody that blocks the anti-inflammatory activity of P15E is also able to bind TGF-p (423). Whether the anti-inflammatory molecule recognized by antibodies to PI5E is TGF-p or the product of endogenous retroviral sequences remains to be determined. The molecular mechanisms by which these tumour-derived anti-chemotactic anti-inflammatory factors mediate their effects on macrophages has received little attention. Normann and Sorkin (414), however, found that culture supernatants from two rat tumour cell lines inhibited both macrophage accumulation in vivo in response to i.p. injection of peptone and the chemotactic responsiveness of peptone elicited-peritoneal macrophages in vitro if mixed with the chemotactic agent or cells. If the responding macrophages were incubated with culture supernatants for a short period of time and then washed their chemotactic responsiveness was also greatly reduced. The supernatants had no effect on macrophage viability and repetitive absorptions with macrophages depleted culture supernatants of their capacity to inhibit the chemotactic responsiveness of fresh macrophages. Other groups obtained similar results using sonicated murine tumour cell dialysates (409) and human malignant effusions (416). These results suggest that tumour-derived anti-chemotactic factors act by binding to the macrophage cell surface. The events which occur following this interaction remain to be elucidated. The part played by tumour-derived anu-chemotactic/anti-inflammatory factors in determining the macrophage content of tumours is difficult to evaluate. Although the effects of these inhibitors only become systemically evident late in tumour growth, it is possible that within the locality of the tumour mass functionally significant levels of inhibitors may be found at a much earlier stage. However no study to date has attempted to correlate the production of anti-inflammatory factors by tumour cells with the macrophage content of the tumour. Moreover, no one has yet examined the production of both chemotactic and anti-chemotactic factors within the same tumour system. If, indeed, certain tumours do produce both chemotactic and anti-chemotactic factors, the two acting together could perhaps provide a rate-limiting mechanism that would control the macrophage content of the tumour. 43 7.0 Thesis Objectives As discussed above, most solid tumour masses appear to contain a significant number of macrophages. Moreover, these cells have been shown to be capable of performing a wide range of functions in vitro, some of which, if they were also to occur within the tumour microenvironment, might be expected to have a profound effect on tumour growth and/or metastasis. Based on such findings, a number of groups are currently exploring the therapeutic potential of systemically activating macrophages so as to enhance their anti-tumour activity. It is clear, however, that before the benefit of such an approach can be fully realized it will be necessary to better understand the molecular mechanisms that regulate the accumulation and functional activity of tumour-associated macrophages. While it is generally agreed that factors that enhance the recruitment of monocytes and/or other more primitive macrophage precursor cells into a tumour site from the peripheral blood, or which enhance the survival, proliferation and/or differentiation of these cells would function to increase the total number of macrophages present within a tumour, the actual factors involved in these complex biological processes remain poorly defined. The overall goal of this thesis was to identify some of these factors and determine the role that they may play in regulating the accumulation of macrophages at tumour sites. Although macrophage progenitor cells with extensive proliferative capacity have been identified within various histologically distinct tumour types, the contribution that the proliferation and differentiation of such cells makes to the total macrophage content of a tumour remains unclear. To begin to address this question, my first major objective was to determine whether tumour cells alone are capable of providing all of the signals necessary to support the proliferation and differentiation of myeloid progenitor cells. To do this, a co-culture assay was developed, based on the long-term culture system, to allow quantitative comparisons of the relative myelosupportive capacity of various cell lines to be made. Using this same assay system and cell lines derived from various mutant mouse strains, I hoped to be able to identify some of the cytokines that might be involved in this process. The development, validation and use of this assay system is described in Chapter 2. 44 Based on the results of these studies which demonstrated an important role for M-CSF in the maintenance of myeloid progenitor cells in vitro, my next objective was to test the hypothesis that this particular cytokine may also have a role in regulating the accumulation and/or functional activity of tumour-associated macrophages in vivo. In order to do this, fibroblastoid cell lines derived from M-CSF-deficient op/op mice and their phenotypically normal op/+ littermate controls were transformed with a retroviral vector encoding Polyoma large T and the macrophage content of the tumours produced following inoculation of these cell lines into SCID mice was determined. I then planned to use Northern blot analysis to further phenotype the macrophages present within both tumour types. The results of these studies are presented in Chapter 3. Among the cytokines known to be both regulated by M-CSF and to play an important role in recruiting macrophages into sites of inflammation is TNF-a. Thus the third and final objective addressed in this thesis was to determine whether TNF-a is produced within the tumour microenvironment, and if it is, how is its production controlled and what role might it have in regulating the recruitment of macrophages into tumour sites. The experimental approach I planned to employ was to characterize both TNF-a production and the macrophage content of identical tumours grown in syngeneic Lpsn C3H/HeN and L/?s^C3H/HeJ mice. The results of these studies are presented in Chapter 4. Together, these objectives were anticipated to serve as a logical first step towards the larger goal of elucidating the full complex network of interactions that contributes to the regulation of the number and functional activity of tumour-associated macrophages. Only through a better understanding of these events will it be possible to fully exploit the potential of macrophages in gene therapy- and/or immunotherapy based approaches to the treatment of cancer. 8 . References 1. Karnovsky, M. L. Metchinoff in Messina: a century of studies on phagocytosis. N Engl J Med. 1981; 304: 1178-1180. 2. Aschoff, L. Das reticulo-endotheliale system. Ergeb Inn Med Kingerheilkd. 1924; 26: 1-118. 3. van Furth, R.; Cohn, Z. A.; Hirsch, J. G.; Humphrey, J. H.; Spector, W. G.; Langevoort, H. L. The mononuclear phagocyte system: a new classification of macrophages, monocytes and their precursor cells. Bull WHO. 1972; 46: 845-852. 4. van Furth, R. Origin and turnover of monocytes and macrophages. Curr Top Pathol. 1989; 79: 125-150. 5. van Furth, R. Development of mononuclear phagocytes. Forster, O.; Landy, M., Eds. Heterogeneity of Mononuclear Phagocytes. : Academic Press, London; 1981: 3-10. 6. Lemischka, I. R.; Raulet, D. H.; Mulligan, R. C. Developmental potential and dynamic behaviour of hematopoietic stem cells. Cell. 1986; 45: 917-927. 7. Goud, T. J. L. M.; van Furth, R. Proliferative characteristics of monoblasts grown in vitro. J Exp Med. 1975; 142: 1200-1217. 8. Metcalf, D. Regulation of macrophage production. Adv Exp Med Biol. 1982; 155: 33-48. 9. van Furth, R.; Cohn, Z. A. The origin and kinetics of mononuclear phogocytes. J Exp Med. 1968; 128: 415-435. 10. Rothlein, R.; Dustin, M. L.; Marlin, S. D.; Springer, T. A. A human intercellular adhesion molecule (ICAM-1) distinct from LFA-1. J Immunol. 1986; 137: 1270-1274. 11. Dustin, M. L.; Rothlein, R.; Bhan, A. K.; Dinarello, C. A.; Springer, T. A. Induction by IL-1 and interferon-y: tissue distribution, biochemistry and function of a natural adherence molecule (ICAM-1). J Immunol. 1986; 137: 245-254. 12. Wu, A. M.; Till, J. E.; Siminovitch, L.; McCulloch, E. A. Cytological evidence for a relationship between normal hematopoietic colony-forming cells and cells of the immune system. J Exp Med. 1968; 127: 455-463. 13. Lemischka, I.; Raulet, D. H.; Mulligan, R. C. Developmental potential and dynamic behaviour of hematopoietic stem cells. Cell. 1986; 45: 917-927. 14. Lajtha, L. G. Stem Cell Concepts. Differentiation. 1979; 14: 23-34. 15. Till, J. E.; McCulloch, E. A. A direct measurement of the radiation sensitivity of normal bone marrow cells. Radiat Res. 1961; 14: 213-222. 16. Hodgson, G. S.; Bradley, T. R. Properties of hematopoietic stem cells surviving 5-fluorouracil treatment: evidence for a pre-CFU-S cell. Nature. 1979; 281: 381-382. 17. Van Zant, G. Studies of hematopoietic stem cells spared by 5-fluorouracil. J Exp Med. 1984; 159: 679-690. 46 18. Magli, M. C.; Iscove, N. N.; Odartchenko, N. Transient nature of early hematopoietic spleen colonies. Nature. 1982; 295: 527-529. 19. Svilvassy, S. J.; Humphries, R. K.; Lansdorp, P. M ; Eaves, A. C ; Eaves, C. J. Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy. Proc Natl Acad Sci USA. 1990; 87: 8736-8740. 20. Szilvassy, S. J.; Lansdorp, P. M.; Humphries, R. K.; Eaves, A. E.; Eaves, C. J. Blood. Isolation in a single step of a highly enriched murine hematopoietic stem cell population with competitive long-term repopulating ability. 1989; 74: 930-939. 21. Spangrude, G. J.; Heimfeld, S.; Weissman, I. L. Purification and characterization of mouse hematopoietic stem cells. Science. 1988; 241: 58-62. 22. Ploemacher, R. E.; Brons, R. H. C. Separation of CFU-S from primitive cells responsible for reconstitution of the bone marrow hematopoietic stem cell compartment following irradiation: evidence for a pre-CFU-S cell. Exp Hematol. 1989; 17: 263-266. 23. Nakahata, T.; Ogawa, M. Hemopoietic colony-forming cells in umbilical cord blood with extensive capability to generate mono- and multipotential hemopoietic progenitors. J Clin Invest. 1982; 70: 1324-1328. 24. Nakahata, T.; Ogawa, M. Identification in culture of a new class of hemopoietic colony-forming units with extensive capability to self-renew and generate multipotential colonies. Proc Natl Acad Sci USA. 1982; 79: 3843-3846. 25. Berlin, C ; Berg, E. L.; Briskin, M. J.; Andrew, D. P.; Kilshaw, P. J.; Holzmann, B.; Weissman. I.L.; Hamann, A.; Butcher, E. C. a 4 ^ intergin mediates lymphocyte binding to the mucosal vascular addressin MAdCAM-1. Cell. 1993; 74: 185-195. 26. Springer, T. A. Adhesion receptors of the immune system. Nature. 1990; 346: 425-433. 27. de Fougerolles, A. R.; Qin, X.; Springer, T. A. Characterization of the function of ICAM-3 and comparison with ICAM-1 and ICAM-2 in immune responses. J Exp Med. 1994; 179: 619-629. 28. Miller, M. D.; Krangel, M. S. Biology and biochemistry of the chemokines: a family of Chemotactic and inflammatory cytokines. CRC Crit Rev Immunol. 1992; 12: 17-46. 29. Tanaka, Y.; Adams, D. H.; Hubscher, S.; Hirano, H.; Siebenlist, U.; Shaw, S. T-cell adhesion induced by proteoglycan-immobilized cytokine MIPl-p. Nature. 1993; 361: 79-82. 30. Suda, T.; Suda, J.; Ogawa, M.; Ihle, J. N. Permissive role of interleukin-3 (IL-3) in proliferation and differentiation of multipotential hemopoietic progenitors in culture. J Cell Physiol. 1985; 124: 182-190. 31. Bradley, T. R.; Hodgson, G. S. Detection of primitive macrophages progenitor cells in mouse bone marrow. Blood. 1979; 54: 1446-1450. 32. Boswell, H. S.; Wade, P. M.; Quesenberry, P. J. Thy-1 antigen expression by murine high-proliferative capacity hematopoietic progenitor cells. J Immunol. 1984; 133: 2940-2949. 47 33. McNiece, I. K.; Kriegler, A. B.; Hapel, A. J.; Fung, M. C ; Young, I. G.; Bradley, T. R.; Hodgson, G. S. Recombinant interleukin-3 exhibits synergistic factor activity. Cell Biol Int Rep. 1984; 8: 812. 34. Ploemacher, R. E.; van der Sluijs, J. R; Voerman, J. S. A.; Brons, N. H. C. An in vitro limiting-dilution assay of long-term repopulating hematopoietic stem cells in the mouse. Blood. 1989; 74: 2755-2763. 35. Pluznik, D. H.; Sachs, L. The cloning of normal 'mast' cells in tissue culture. J Cell Comp Physiol. 1966; 66: 319-324. 36. Bradley, T. R.; Metcalf, D. The growth of mouse bone marrow cells in vitro. Aust J Exp Biol Med Sci. 1966; 44: 287-300. 37 Metcalf, D.; Merchav, S. Effects of GM-CSF deprivation on granulocytes and monocytes. J Cell Physiol. 1982; 112: 411-418. 38. van Furth, R.; Diesselhoff-den Dulk, M. M. C.; van Zwet, Th L.; Raeburn, J. A.; Crofton, R.; Blusse van Oud Alblas, A. Characteristics, origin and kinetics of human and murine mononuclear phagocytes. Mononuclear phagocytes. Functional aspects, van Furth, R., Ed. : Martinus Nijhoff, The Hague, Boston, London; 1980. 39. van Furth, R.; Diesselhoff-den Dulk, M. M. C. The kinetics of promonocytes and monocytes in the bone marrow. J Exp Med. 1970; 132: 813-828. 40. Auger, M. J.; Ross, J. A. The biology of the macrophage. The Macrophage. Lewis, C. E.; McGee, J. O'D, Eds. : Oxford University Press, New York; 1992: 3-74. 41. Pawlowski, N. A.; Abraham, E. L.; Pointer, S.; Scott, W. A.; Cohn, Z. A. Human monocyte endothelial cell interaction in vitro. Proc Natl Acad Sci USA. 1985; 821: 8208-8212. 42. Springer, T. A. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell. 1994; 76: 301-314. 43. Ager, A. Lymphocyte recirculation and homing: roles of adhesion molecules and chemoattractants. Trends Cell Biol. 1994; 4: 326-333. 44. Lasky, L. A. Selectins: interpreters of cell-specific carbohydrate information during inflammation. Science. 1992; 258: 964-969. 45. Rosen, S. D. Cell surface lectins in the immune system. Semin Immunol. 1993; 5: 237-247. 46. Lasky, L. A.; Singer, M. S.; Dowbenko, D.; Imai, Y.; Henzel, W. J.; Grimley, C.; Fennie, C ; Gillett, N.; Watson, S. R.; Rosen, S. D. An endothelial ligand for L-selectin is a novel mucin-like molecule. Cell. 1992; 69: 927-938. 47. Baumhueter, S.; Singer, M. S.; Henzel, W.; Hemmerich, S.; Renz, M.; Rosen, S. D.; Lasky, L. A. Binding of L-selectin to the vascular sialomucin, CD34. Science. 1993; 262: 436-438. 48. Berg, E. L.; McEvoy, L. M.; Berlin, C ; Bargatze, R. F.; Butcher, E. C. L-selectin-mediated lymphocyte rolling on MAdCAM-1. Nature. 1993; 366: 695-698. 48 49. Lawrence, M. B.; Springer, T. A. Leukocytes roll on a selectin at physiologic flow rates: distinction from and prerequisite for adhesion through integrins. Cell. 1991; 65: 859-873. 50. Ley, K.; Gaehtgens, P.; Fennie, C ; Singer, M. S.; Lasky, L. A.; Rosen, S. D. Lectin-like cell adhesion molecule 1 mediates leukocyte rolling in mesenteric venules in vivo. Blood. 1991; 77: 2553-2555. 51. Bevilacqua, M. P. Endothelial-leukocyte adhesion molecules. Annu Rev Immunol. 1993; 11:767-804. 52. de Fougerolles, A. R.; Stacker, S. A.; Schwarting, R.; Springer, T. A. Characterization of ICAM-2 and evidence for a third counter-receptor for LFA-1. J Exp Med. 1991; 174: 253-267. 53. Szekanecz, Z.; Humphries, M. J.; Ager, A. Lymphocyte adhesion to high endothelium is mediated by two beta 1 integrin receptors for fibronectin, alpha 4 beta 1 and alpha 5 beta 1. J Cell Biol. 1992; 101: 885-894. 54. Beekhuizen, H.; Corsel-van Tiburg, A. J.; van Furth, R. Characterization of monocyte adherence to human macrovascular and microvascular endothelial cells. J Immunol. 1990; 145: 510-518. 55. Carlos, T. M.; Dobrina, A.; Ross, R.; Harlan, J. M. Multiple receptors on human monocytes are involved in adhesion to cultured human endothelial cells. J Leuk Biol. 1990; 48: 451-456. 56. Hakkert, B. C ; Rentenaar, J. M.; van Aken, W. G.; Roos, D.; van Mourik, J. A. A three-dimensional model system to study the interactions between human leukocytes and endothelial cells. Eur J Immunol. 1990; 20: 2775-2781. 57. Muller, W. A.; Weigl, S. A. Monocyte-selective transendothelial migration: dissection of the'binding and transmigration phases by an in vitro assay. J Exp Med. 1992; 176: 819-828. 58. Spertini, O.; Luscinskas, F. W.; Gimbrone, M. A.; Tedder, T. F. Monocyte attachment to activated vascular endothelium in vitro is mediated by leukocyte adhesion molecule-1 (L-selectin) under nonstatic conditions. J Exp Med. 1992; 175: 1789-1792. 59. Hakkert, B. C ; Kuijpers, T. W.; Leeuwenberg, J. F. M.; van Mourik, J. A.; Roos, D. Neutrophil and monocyte adherence to and migration across monolayers of cytokine-activated endothelial cells: the contribution of CD 18, ELAM-1, and VLA-4. Blood. 1991;78:2721-2726. 60. Dustin, M. L.; Rothlein, R.; Bhan, A. K.; Dinarello, C. A.; Springer, T. A. Induction of IL-1 and interferon, tissue distribution, biochemistry, and function of a natural adherence molecule. J Immunol. 1986; 137: 245-254. 61. Beekhuizen, H.; Corsel-van Tilburg, A. J.; Blokland, I.; van Furth, R. Characterization of the adherence of human monocytes to cytokine-stimulated human macrovascular endothelial cells. Immunology. 1991; 74: 661-669. 62. Lanier, L. L.; Arnaout, M. A.; Schwarting, R.; Warner, N. L.; Ross, G. D. pl50,95, a third member of the LFA-1/CR3 polypeptide family identified by anti-LeuM5 monoclonal antibody. Eur J Immunol. 1985; 15: 713-718. 49 63. Springer, T. A.; Miller, L. J.; Anderson D.C. pl50,95, the third member of the Mac-1, LFA-1 human leukocyte adhesion glycoprotein family. J Immunol. 1986; 136: 240-245. 64. Keizer, G. D.; Borst, J.; Visser, W.; Schwarting, R.; de Vries, J. E.; Figdor, C. G. Membrane glycoprotein p 150,95 of human cytokine T cell clones is involved in conjugate formation with target cells. J Immunol. 1987; 138: 3130-3136. 65. te Velde, A. A.; Keizer, G. D.; Figdor, C. G. Differential function of LFA-1 family of molecules (CD 11 and CD 18) in adhesion of monocytes to melanoma and endothelial cells. Immunology. 1987; 61: 261-267. 66. Wallis, W. J.; Beatty, P. G.; Ochs, H. D.; Harlan, J. M. Human monocyte adherence to cultured vascular endothelium: monoclonal antibody-defined mechanisms. J Immunol. 1985; 135: 2323-2330. 67. Mentzer, S. J.; Crimmins, M. A. V.; Burakoff, S. J.; Faller, D. V. Alpha and beta subunits of the LFA-1 membrane molecule are involved in human monocyte-endothelial cell adhesion. J Cell Physiol. 1987; 130: 410-415. 68. Miller, L. J.; Schwarting, R.; Springer, T. A. Regulated expression of the Mac-1, LFA-1, pl50,95 glycoprotein family during leukocyte differentiation. J Immunol. 1986; 137: 2891-2900. 69. Chuluyan, H. E.; Issekutz, A. C. VLA-4 integrin can mediate CD1 l/CD18-independent transendothelial migration of human monocytes. J Clin Invest. 1993; 92: 2768-2777. 70. Beekhuizen, H.; Verdegaal, E. M. E.; Blokland, I.; van Furth, R. Contribution of ICAM-1 and VCAM-1 to the morphological changes in monocytes bound to human venous endothelial cells stimulated with rIL-4 or rIL-la. Immunol. 1992; 77: 469-472. 71. Beekhuizen, H.; van Furth, R. Monocyte adherence to human vascular endothelium. J Leukocyte Biol. 1993; 54: 363-378. 72. Walz, G.; Aruffo, A.; Kolanus, W.; Bevilacqua, M.; Seed, B. Recognition by ELAM-1 of the sialyl-Lex determinant on myeloid and tumor cells. Science. 1990; 250: 1132-1135. 73. Bevilacqua, M. P.; Pober, J. S.; Mendrick, D. L.; Cotran. R.S.; Gimbrone, M. A. Identification of an inducible endothelial-leukocyte adhesion molecule. Proc Natl Acad Sci USA. 1987; 84: 9238-9242. 74. Issekutz, T. B.; Issekutz, A. C ; Movat, H. Z. The in vivo quantitation and kinetics of monocyte migration into acute inflammatory tissue. Am J Path. 1981; 103: 47-55. 75. Cotran, R. S.; Gimbrone, M. A.; Bevilacqua, M. P.; Mendrick, D. L.; Pober, J. S. Induction and detection of a human endothelial activation antigen in vivo. J Exp Med. 1986; 164: 661-666. 76. Newman, P. J.; Beradt, M. C ; Gorski, J.; White II, G. C ; Lyman, S.; Paddock, C ; Muller, W. A. PECAM-1 (CD31) cloning and relation to adhesion molecules of the immunoglobulin gene superfamily. Science. 1990; 247: 1219-1222. 77. Muller, W. A.; Ratti, C. M.; McDonnell, S. L.; Cohn, Z. A. A human endothelial cell-restricted, externally disposed plasmalemmal protein enriched in intercellular junctions. J Exp med. 1989; 170: 399-414. 50 78. Ioffreda, M. D.; Albelda, S. M.; Elder, D. E.; Radu, A.; Leventhal, L. C.; Zweiman, B.; Murphy, G. F. TNFa induces E-selectin expression and PECAM-1 (CD31) redistribution in extracutaneous tissues. Endothelium. 1993; 1: 47-54. 79. Lin, H-S. Colony formation in vitro by mouse blood monocytes. Blood. 1977; 49: 593-598. 80. MacVittee, T. J.; McCarthy, K. F. The detection of in vitro monocyte-macrophage colony-forming cells in mouse thymus and lymph nodes. J Cell Physiol. 1977; 92: 203-208. 81. McCarthy, K. F.; MacVittee, T. J. Velocity sedimentation studies on monocyte-macrophage colony-forming cells from marrow, spleen, blood and peritoneal exudate. J Reticuloendothel Soc. 1978; 24: 263-269. 82. MacVittee, T. J.; Porvaznik, M. Dectection of in vitro macrophage colony-forming cells (M-CFC) in mouse bone marrow, spleen and peripheral blood. J Cell Physiol. 1978; 97: 305-313. 83. Chen, D-M; Lin, H-S; Stahl, P.; Stanley, E. R. Clonal growth in vitro by mouse Kupffer cells. Exp Cell Res. 1979; 121: 103-109. 84. DeBakker, J. H.; Daems, W. Th. The heterogeneity of mouse peritoneal macrophages. Forster, O.; Landy, M., Eds. Heterogeneity of Mononuclear Phagocytes.: Academic Press, London; 1981: 11-13. 85. van Furth, R. Vane, J. R.; Fereira, J. R. Mononuclear Phagocytes in Inflammation.: Springer Verlag, Berlin-Heidelberg-New York ; 1978: 69-108. 86. Lin, H-S; Kuhn, C ; Kup, T. Clonal growth of hamster free alveolar cells in soft agar. J Exp Med. 1975; 142: 877-886. 87. Chen, D-M; Lin, H-S. Interaction of murine colony stimulating factor (CSF-1) with alveolar mononuclear phagocytes. Exp Cell Res. 1982; 140: 323-329. 88. Bursuker, I.; Rhodes, J. M.; Goldman, R. p-galactosidase-an indicator of the maturation state of mouse and human mononuclear phagocytes. J Cell Physiol. 1982; 112: 385-390. 89. Dienstman, S. R.; Defendi, V. Necessary and sufficient conditions for recruitment of macrophages into the cell cycle. Exp Cell Res. 1978; 115: 191-199. 90. Mackaness, G. B. The immunological basis for acquired cellular resistance. J Exp Med. 1964; 120: 105-120. 91. Ruskin, J.; Mcintosh, M.; Remington, J. S. Studies on the mechanism of resistance to phylogenetically diverse organisms. J Immunol. 1969; 103: 252-259. 92. Steigbigel, R. T.; Lambert, L. H.; Remington, J. S. Phagocytic and bactericidal properties of normal human monocytes. J Clin Invest. 1974; 53: 131-142. 93. Adams, D. O.; Snyderman, R. Do macrophages destroy nascent tumors? J Natl Cancer Inst. 1979; 62: 1341-1345. 94. Adams, D. O. Macrophage activation and secretion. Fed Proc. 1982; 41: 2193-2197. 51 95. Adams, D. O.; Johnson, W. J.; Marino, P. A. Mechanisms of target recognition and destruction in macrophage-mediated tumor cytotoxicity. Fed Proc. 1982; 41: 2212-2221. 96. Zoler, M. L. Marshaling macrophages against metastases. JAMA . 1983; 249: 1690-1695. 97. Mackaness, G. B. Resistance to intracellular infection. J Infect Dis. 1971; 123: 439-445. 98. Noar,'D. Suppressor cells: permitters and promoters of malignancy. Adv Cancer Res. 1979; 29: 45-125. 99. Gablzon, A.; Leibovich, S. J.; Goldman, R. Contrasting effects of activated and nonactivated macrophages from tumor-bearing mice on tumor growth in vivo. J Natl Cancer Inst. 1980; 65: 913-920. 100. Gorelik, E.; Wiltrout, R. H.; Brunda, M. J.; Holden, H. T.; Herberman, R. B. Augmentation of metastasis formation by thioglycollate-elicited macrophages. Int J Cancer. 1982; 29: 575-581. 101. Prehn, R. T. Immunological surveillance versus immunological stimulation of oncogenesis-a formal proof of the stimulation hypothesis. Adv Exp Med Biol. 1982; 155: 77-85. 102. Unanue, E. R. The regulatory role of macrophages in antigenic stimulation. Adv Immunol. 1972; 15: 95-165. 103. Howie, S. M.; McBride, W. H. Tumor-specific T helper activity can be abrogated by two distinct suppressor cell mechanisms. Eur J Immunol. 1982; 12: 671-675. 104. Kirchner, R ; Muchmore, A. V.; Chused, T. M.; Holden, H. T.; Herberman, R. B. Inhibition of proliferation of lymphoma cells and T lymphocytes by suppressor cells from spleens of tumor-bearing mice. J Immunol. 1975; 114: 206-210. 105. Baird, L. G.; Kaplan, A. M. Macrophage regulation of mitogen-induced blastogenesis. I. Demonstration of inhibitory cells in the spleens and peritoneal exudates of mice. Cell Immunol. 1977; 28: 22-35. 106. Unanue, E. R. The regulatory role of macrophages in antigenic stimulation. II. Symbiotic relationship between lymphocytes and macrophages . Adv Immunol. 1981; 31: 1-136. 107. Jones, J. T.; McBride, W. H.; Weir, D. M. The in vitro killing of syngeneic cells by peritoneal cells from adjuvant-stimulated mice. Cell Immunol. 1975; 18: 375-383. 108. Keller, R. Susceptibility of normal and transformed cell lines to cytostatic and cytocidial effects exerted by macrophages. J Natl Cancer Inst. 1976; 56: 369-374. 109. Nathan, C. F.; Murray, H. W.; Cohn, Z. A. The macrophage as an effector cell. N Engl J Med. 1980; 304: 622-626. 110. Douglas, S. D. Mononuclear phagocytes and tissue regulatory mechanisms. Dev Comp Immunol. 1980;4:7-10. 111. Silverstein, S. C. Membrane receptors and the regulation of mononuclear phagocyte effector functions. Adv Exp Med Biol. 1982; 155: 21-31. 52 112. De, S. K.; Chen, H-L; Pace, J. L.; Hunt, J. S.; Terranova, P. F.; Enders, G. C. Expression of tumor necrosis factor-alpha in mouse spermatogenic cells. Endocrinology. 1993; 133: 389-396. 113. Xiong, Y.; Hales, D. B. The role of tumor necrosis factor-alpha in the regulation of mouse Leydigcell steroidodenesis. Endocrinology. 1993; 132: 2438-2444. 114. Chen, H-L; Marcinkiemicz, J. L.; Sancho-Tello, M ; Hunt, J. S.; Terranova, P. F. Tumor necrosis factor-alpha gene expression in mouse oocytes and folicular cells. Biol Reprod. 1993; 48: 707-714. 115. Sylvie de Kossodo, B.; Grau, G. E.; Daneva, T.; Pointaire, P.; Fossati, L.; Ody, C ; Zapf, J.; Piguet, P-F; Gaillard, R. C ; Vassalli, P. Tumor necrosis factor a is involved in mouse growth and lymphoid tissue development. J Exp Med. 1992; 176: 1259-1264. 116. Sabin, F. R.; Doan, C. A.; Cunningham, B. S. Discrimination of two types of phagocytic cells in the connective tissues by the supravital technique. Contrib Embryol. 1925; 16: 125-162. 117. Walker, W. S. Functional heterogeneity of macrophages. Nelson, D. S., Ed. Immunobiology of the Macrophage.: Academic Press, New York ; 1976: 91-110. 118. Hoppner, K. E.; Wood, P. R.; Nelson, D. S. Macrophage heterogeneity. Vox Sang. 1979; 36: 257-274. 119. Walker, W. S. Macrophage heterogeneity. Adv Exp Med Biol. 1982; 155: 435-443. 120. Goud, T. J. L. M.; Schotte, C.; van Furth, R. Identification and characterization of the monoblast in mononuclear phagocyte colonies grown in vitro . J Exp Med. 1975; 142: 1180-1199. 121. Lohmann-Matthes, M-L; Sun, D. M.; Zahringer, M. Maturation of macrophages and macrophage mediated cytotoxicity. Heterogeneity of Mononuclear Phagocytes. Forster, O.; Landy, M., Eds. : Academic Press, London; 1981: 486-495. 122. Lee, K-C. On the origin and mode of action of functionally distinct macrophage subpopulations. Mol Biol Chem. 1980; 30: 39-55. 123. Lee, K-C; Kay, J.; Wong, M. Separation of functionally distinct subpopulations of Corynebacterium parvum -activated macrophages with predominantly stimulatory or suppressive effects on cell mediated cytotoxic T cell response. Cell Immunol. 1979; 42: 28-41. 124. Bursuker, I.; Goldman, R. Distinct bone marrow precursors for mononuclear phagocytes expressing high and low 5'-nucleotidase activity. J Cell Physiol. 1982; 112: 237-242. 125. Bursuker, I.; Goldman, R. On the origin of macrophage heterogeneity: a hypothesis . J Reticuloendothel Soc. 1983; 33: 207-220. 126. Gandour, D. M.; Walker, W. S. Macrophage cell cycling: influence on Fc receptors and antibody dependent phagocytosis. J Immunol. 1983; 130: 1108-1112. 127. Neumann, C.; Sorg, C. Differential induction of plasminogen activator and interferon production in murine macrophages. Heterogeneity of Mononuclear Phagocytes. Forster, O.; Landy, M., Eds. : Academic Press, London; 1981: 368-376. 53 128. Dougherty, G. J.; McBride, W. H. Macrophage heterogeneity. J Clin Lab Immunol. 1984; 14: 1-11. 129. Oren, R. A.; Farnham, A.; Saito, K.; Milofsky, E.; Karnovsky, M. 1. Metabolic patterns in three types of phagocytozing cells. J Cell Biol. 1963; 17: 487-501. 130. Karnovsky, M. L.; Simmons, S.; Glass, E. A.; Shafer, A. W.; Hart, P. D. Metabolism of macrophages. Mononuclear Phagocytes, van Furth, R., Ed.: Blackwell, Oxford; 1970: 102-117. 131. Simons, L. M.; Robin, E. D.; Phillips, J. R.; Acevedo, J.; Axline, S. G.; Theodore, J. Enzymatic basis for bioenergetic differences of alveolar versus peritoneal macrophages and enzyme regulation by molecular O2. J Clin Invest. 1977; 59: 443-448. 132. Bar-Eli, M.; Territo, M. C ; Cline, M. J. The progeny of a single progenitor cell can develop characteristics of either a tissue or an alveolar macrophage. Blood. 1981; 57: 95-98. 133. Metcalf, D.; McDonald, H. R. Heterogeneity of in vitro colony- and cluster-forming cells in the mouse marrow: segregation by velocity sedimentation. J Cell Physiol. 1975; 85: 643-653. 134. Byrne, P.; Heit, W.; Kubanek, B. The in vitro differentiation of density sub-populations of colony-forming cells under the influence of different types of colony-stimulating factor. Cell Tissue Kinet. 1977; 10: 341-351. 135. Bol, S.; Williams, N. The maturation state of three types of granulocyte/macrophage progenitor cells from mouse bone marrow. J Cell Physiol. 1980; 102: 233-243. 136. Nicola, N. A.; Burgess, A. W.; Johnson, G. R.; Metcalf, D.; Battye, F. L. Differential expression of lectin receptors during hemopoietic differentiation: enrichment for granulocyte-macrophage progenitor cells. J Cell Physiol. 1980; 103: 217-237. 137. Claesson, M. H.; Olsson, L. Autoreactive natural killer-like cells from agar-cloned murine bone marrow cells. Nature. 1980; 283: 578-580. 138. Lohmann-Matthes, M-L; Sun, D. M.; Zahringer, M. Maturation of macrophages and macrophage mediated cytotoxicity. Heterogeneity of Mononuclear Phagocytes. Forster, O.; Landy, M., Eds. : Academic Press, London; 1981: 486-495. 139. Buhles, W. C. Studies on mononuclear phagocyte progenitor cells: morphology of cells and colonies in liquid culture of mouse bone marrow. J Reticuloendothel Soc. 1979; 25: 363-378. 140. van Zwet, T. L.; Thomson, J.; van Furth, R. Effect of glucocorticosteroids on the phagocytosis and intercellular killing by peritoneal macrophages. Infect Immun. 1975; 12: 699-705. 141. MacKaness, G. B. Cellular resistance to infection. J Exp Med. 1962; 116: 381-406. 142. MacKaness, G. B. The monocyte in cellular immunity. Semin Hematol. 1970; 7: 172-184. 143. Nathan, C. F.; Murray, H. W.; Cohn, Z. A. The macrophage as an effector cell. N Engl J Med. 1980; 303: 622-626. 54 144. Farah, F. S.; Samra, S. A.; Nuwayri-Salti, N. The role of macrophages in cutaneous leishmaniasis. Immunology. 1975; 29: 755-764. 145. Behin, R.; Mauel, J.; Biroum-Noerjasin; Rowe, D. S. Mechanisms of protective immunity in experimental cutaneous leishmaniasis of the guinea pig. II. Selective destrucion of different Leishmania species in activated guinea pig and mouse macrophages. Clin Exp Immunol. 1975; 20: 351-358. 146. Remington, J. S.; Krahenbuhl, J. L.; Mendenhall, J. W. A role for activated macrophages in resistance to infection with toxoplasma. Infect Immun. 1972; 6: 829-834. 147. Hibbs, J. B. Jr.; Taintor, R.; Chapman. H.A.Jr.; Weinberg, J. Macrophage tumor killing: influence of the local microenvironment. Science. 1977; 197: 279-282. 148. Cohn, Z. A. Activation of mononuclear phagocytes: fact, fancy, and future. J Immunol. 1978; 121: 813-816. 149. Miki, K.; MacKaness, G. B. The passive transfer of acquired resistance to Listeria monocyogenes. J Exp Med. 1964; 120: 93-104. 150. Rodda, S. J.; White, D. O. Cytotoxic macrophages-a rapid non-specific response to viral infection. J Immunol. 1976; 117: 2067-2072. 151. Alexander, P.; Evans, R. Endotoxin and double stranded RNA render macrophages cytotoxic. Nature. 1971; 232: 76-78. 152. Ogmundsdottir, H. M.; Weir, D. M. Mechanisms of macrophage activation. Clin Exp Immunol. 1980; 110: 223-234. 153. Hibbs, J. B. Jr. Discriminating between neoplastic and nonneoplastic cells in vitro by activated macrophages. J Natl Cancer Inst. 1974; 53: 1487-1492. 154. Wing, E. J.; Gardner, J. D.; Ryning, F. W.; Remington, J. S. Dissociation of effector functions in populations of activated macrophages. Nature. 1977; 268: 642-644. 155. Adams, D. O.; Hamilton, T. A. Cell biology of macrophage activation. Ann Rev Immunol. 1984;2:283-318. 156. Paulnock, D. M. Macrophage activation by T cells. Curr Opinion Immunol. 1992; 4: 344-349. 157. Beutler, B.; Cerami, A. The biology of cachetin/TNF: a primary mediator of the host response. Ann Rev Immunol. 1989; 7: 625-656. 158. Sherry, B.; Cerami, A. Cachectin/tumour necrosis factor exerts endocrine, paracrine, and autocrine control of inflammatory responses. J Cell Biol. 1988; 107: 1269-1277. 159. Drysdale, B. E.; Agaarwal, S.; Shin, H. S. Macrophage mediated tumoricidal activity: mechanisms of activation and cytotoxicity. Prog Allergy. 1988; 40: 111-161. 160. De Maeyer, E.; De Maeyer-Guignard, J. Interferon-y. Curr Opin Immunol. 1992; 4: 321-326. 161. Tsujimoto, M.; Yip, Y. K.; Vilcek, J. Interferon-y enhances expression receptors for tumor necrosis factor. J Immunol. 1986; 136: 2441-2444. 55 162. Ruggiero, V.; Tavernier, J.; Fiers, W.; Baglioni, C. Induction of the synthesis of tumor necrosis factor receptors by interferon-y. J Immunol. 1986; 136: 2445-2450. 163. Munoz Fernandez, M. A.; Fernandez, M. A.; Fresno, M. Synergism between tumor necrosis factor-a and interferon^ on macrophage activation for the killing of intracellular Trypanosoma cruzy through a nitric oxide-dependent mechanism. Eur J Immunol. 1992; 22: 301-307. 164. Stout, R. D.; Suttles, J. Evidence for involvement of TNFa in the induction phase and IFNp in the effector phase of antiproliferative activity of splenic macrophages. Cell Immunol. 1992; 139: 363-374. 165. Langermans, J. A.; van der Hulst, M. E.; Nibbering, P. H.; Hiemstra, P. S.; Fransen, L.; van Furth, R. IFN-y-induced L-arginine-dependent Toxoplasmastatic activity in murine peritoneal macrophages is mediated by endogenous tumor necrosis factor-a. J Immunol. 1992; 148: 568-574. 166. Stout, R. D.; Bottomly, K. Antigen specific activation of effector macrophages by interferon-/ producing (Thl) T cell clones. Failure of IL-4 producing (Th2) T cell clones to activate effector function in macrophages. J Immunol. 1989; 142: 760-765. 167. Mosmann, T. R.; Coffman, R. L. Heterogeneity of cytokine secretion patterns and functions of helper T cells. Adv Immunol. 1989; 46: 111-148. 168. Stout, R. D.; Suttles, J. T cell-macrophage cognate interaction in the activation of macrophage effector function by Th2 cells. J Immunol. 1993; 150: 5330-5337. 169. Sypek, J. P.; Wyler, D. J. Antileishmanial defense in macrophages triggered by tumor necrosis factor expressed on CD4+ T lymphocyte plasma membrane. J Exp Med. 1991; 174: 755-759. 170. Birkland, T. P.; Sypek, J. P.; Wiler, D. J. Soluble TNF and membrane TNF expressed on CD4+ T lymphocytes differ in their ability to activate macrophage antileishmanial defense. J Leuk Biol. 1992; 51: 296-299. 171. Ju, S. T.; Ruddle, N. H.; Strack, P.; Dorf, M. E.; De Kruyff, R. H. Expression of two distinct cytolytic mechanisms among murine CD4 subsets , J Immunol. 1990; 144: 23-31. 172. Liu, C.; Detmers, P. A.; Jianfg, S.; Young, J. D. Identification and characterization of a membrane-bound cytotoxin of murine cytolytic lymphocytes that is related to tumor necrosis factor/cachectin. Proc Natl Acad Sci USA. 1989; 86: 3286-3290. 173. Noewelle, R. J.; Roy, M ; Shepherd, D. M ; Stamenkovich, I.; Ledbetter, J. A.; Aruffo, A. A 39-kDa protein on activated helper T cells binds CD40 and transduces the signal for cognate activation of B cells. Proc Natl Acad Sci USA. 1992; 89: 6550-6554. 174. Clark, E. A. CD40: A cytokine receptor in search of a ligand. Tissue Antigens. 1990; 36: 33-36. 175. Freedman, A. S.; Freeman, G. J.; Rhynhart, K.; Nadler, L. M. Selective induction of B7/BB-1 on interferon-y stimulated monocytes: a potential mechanism for amplification of T cell activation through the CD28 pathway. Cell Immunol. 1991; 137: 429-437. 176. Lohoff, M.; Kock, A.; Rollinghoff, M. Two signals are involved in polyclonal B cell stimulation by T helper type 2 cells: a role for LFA-1 molecules and interleukin-4. Eur J Immunol. 1992; 22: 599-602. 177. Webb, D. S. A.; Shimizu, Y.; van Seventer, G. A.; Shaw, S.; Gerrard, T. L. LFA-3, CD44, and CD45: physiologic triggers of human monocytes TNF and IL-1 release. Science. 1990; 249: 1295-1297. 178. Noewelle, R. J.; Daum, J.; Bartlett, W. C.; McCann, J.; Shepard, D. M. Cognate interactions between helper T cells and B cells. V. Reconstitution of helper T cell function using purified plasma membranes from activated Thl and Th2 helper T cells and lymphokines. J Immunol. 1991; 146: 1118-1124. 179. Oswald, I. P.; Wynn, T. A.; Sher, A.; James, S. L. Interleukin-10 inhibits macrophage microbicidal activity by blocking the endogenous production of tumor necrosis factor-a required as a costimulatory factor for interferon-y-induced activation. Proc Natl Acad Sci USA. 1992; 89: 8676-8680. 180. Sher, A.; Gazzinelli, R. T.; Oswald, I. P.; Celerici, M.; Kullberg, M.; Pearce, E. J.; Berzofsky, J. A.; Mosman, T. R.; James, S. L.; Morse, H. C. Role of T-cell derived cytokines in the downregulation of immune responses in parasitic and retroviral infection. Immunol Rev. 1992; 127: 183-204. 181. Oswald, I. P.; Gazzinelli, R. T.; Sher, A.; James, S. L. IL-10 synergizes with IL-4 transforming growth factor-p to inhibit macrophage cytotoxic activtity. J Immunol. 1992; 148: 3578-3582. 182. De Waal Malefyt, R.; Yssel, H.; Roncarolo, M-G; Spits, H.; De Vries, J. E. Interleukin-10. Curr Opin Immunol. 1992; 4: 314-320. 183. Pinson, D. M.; LeClaire, R. D.; Lorsbach, R. B.; Parmely, M. J.; Russel, S. W. Regulation by transforming growth factor-pl of expression and function of the receptor for IFN-y on mouse macrophages. J Immunol. 1992; 149: 2028-2034. 184. Ruco, L. P.; Meltzer, M. S. Macrophage activation for tumor cytotoxicity: induction of tumoricidal macrophages by supernants of PPD-stimulated Bacillus-Calmette-Guerin-immune spleen cell culture. J Immunol. 1977; 119: 889-896. 185. Meltzer, M. S.; Ruco, L. P.; Leonard, E. J. Macrophage activation for tumor cytotoxicity: mechanisms of macrophage activation by lymphokines. Adv Exp Med Biol. 1980; 121: 381-398. 186. Meltzer, M. S.; Occhionero, M.; Ruco, L. P. Macrophage activation for tumor cytotoxicity: regulatory mechanisms for induction and control of cytotoxic activity. Fed Proc. 1982; 41: 2198-2205. 187. Meltzer, M. S. Tumor cytotoxicity by lymphokine-activated macrophages: development of macrophage tumoricidal activity requires a sequence of reactions. Lymphokines. 1981; 3: 319-332. 188. Meltzer, M. S.; Tucker, R. W.; Breuer, A. C. Interaction of BCG-activated macrophages with neoplastic and nonneoplastic cell lines in vitro: cinemicrographic analysis. Cell Immunol. 1975; 17: 30-42. 57 189. Keller, R. Susceptibility of normal and transformed cell lines to cytostatic and cytocidal effects exerted by macrophages. J Natl Cancer Inst. 1976; 56: 369-374. 190. Kaplan, A. M ; Walker, P. M.; Morahan, P. S. Tumor cell cytotoxicity versus cytostasis of pyran activated macrophages. Modulation of Host Immune Resistance. Chirigos, M. A., Ed. Washington, D.C: Fogarty International Centre Proceedings, US Government Printing Office; 1977: 277-286. 191. Ruco, L. P.; Meltzer, M. S. Macrophage activation for tumor cytotoxicity: development of macrophage cytotoxic activity requires completion of a sequence of short lived intermediary reactions. J Immunol. 1978; 121: 2035-2042. 192. Lee, K-C; Berry, D. Functional heterogeneity in macrophages activated by Corynebacterium parvum: characterization of subpopulations with different activities in promoting immune responses and suppressing tumor cell growth. J Immunol. 1977; 118: 1530-1540. 193. Lee, K-C; Wong, M.; Mclntyre, D. Characterization of macrophage subpopulations responsive to activation by endotoxin and lymphokines. J Immunol. 1981; 126: 2474-2479. 194. Fidler, I. J.; Raz, A. The induction of tumoricidal capacities in mouse and rat macrophages by lymphokines. Lymphokines, Volume 3. Pick, E., Ed.: Academic Press, New York ; 1981: 345. 195. Fidler, I. J.; Poste, G. Macrophage-mediated destruction of malignant tumor cells and new strategies for the therapy of metastatic disease. Springer Semin Immunopath. 1982; 5: 161-174. 196. Loveless, S. E.; Heppner, G. H. Tumor-associated macrophages of mouse mammary tumors. I. Differential cytotoxicity of macrophages from metastatic and non-metastatic tumors. J Immunol. 1983; 131: 2074-2078. 197. Fidler, I. J.; Cifone, M. A. Properties of metastatic and nonmetastatic cloned subpopulations of an ultraviolet light-induced murine fibrosarcoma of recent origin. Am J Pathol. 1979; 97: 633-648. 198. Hanna, N.; Fidler, I. J. Role of natural killer cells in the destruction of circulating tumor emboli. J Natl Cancer Inst. 1980; 65: 801-809. 199. Evans, R.; Alexander, R. Mechanisms of extracellular killing of nucleated mammalian cells by macrophages. Immunobiology of the Macrophage. Nelson, D. S., Ed.: Academic Press, New York; 1976: 536-576. 200. Hibbs, J. B. Jr.; Chapman, H. A. Jr.; Weinberg, J. B. The macrophage as an antineoplastic surveillance: biological perspectives. J Reticuloendothel Soc. 1978; 24: 549-570. 201. Marino, P. A.; Adams, D. O. Interaction of Bacillus Calmette-Guerin-activated macrophages and neoplastic cells in vitro. I. Conditions of binding and its selectivity. Cell Immunol. 1980; 54: 11-25. 202. Marino, P. A.; Adams, D. O. Interaction of Bacillus-Calmette-Guerin-activated macrophages and neoplastic cells. II. The relationship of selective binding to cytolysis. Cell Immunol. 1980; 54: 26-35. 58 203. Marino, P. A.; Whisnant, C.; Adams, D. O. Binding of BCG-activated macrophages to tumor targets. Selective inhibition by membrane preparations from homologous and heterologous neoplastic cells. J Exp Med. 1981; 154: 77-87. 204. Piessens, W. F. Increased binding of tumor cells by macrophages activated in vitro with lymphocyte mediators. Cell Immunol. 1978; 35: 303-317. 205. Adams, D. O.; Johnson, W. J.; Marino, P. A. Mechanisms of target recognition and destruction in macrophage-mediated tumor cytotoxicity. Fed Proc. 1982; 41: 2212-2221. 206. Unanue, E. R. The regulatory role of macrophages in antigenic stimulation. Adv Immunol. 1972; 15: 95-165. 207. Savill, J.; Fadok, V.; Henson, P.; Haslett, C. Phagocyte recognition of cells undergoing apoptosis. Immunol Today. 1993; 14: 131-136. 208. Adams, D. O.; Marino, P. A. Evidence for a multistep mechanism of cytolysis by BCG-activated macrophages: the interrelationship between the capacity for cytolysis, target binding and secretion of cytolytic factor. J Immunol. 1981; 126: 981-987. 209. Schwamberger, G.; Flesch, I.; Ferber, E. Tumoricidal effector molecules of murine macrophages. Pathol Biol. 1991; 59: 248-253. 210. Klostergaard, J. Macrophage tumoricidal mechanisms. Res Immunol. 1993; 144: 274-276. 211. Klostergaard, J.; Leroux, M. E.; Hung, M-C. Cellular models of macrophage tumoricidal effector mechanisms in vitro. Characterization of cytolytic responses to tumor necrosis factor and nitric oxide pathways in vitro. J Immunol. 1991; 147: 2802-2808. 212. Urban, J. L.; Shepard, H. M.; Rothstein, J. L.; Sugarman, B. J.; Schreiber, H. Tumor necrosis factor: a potent effector molecule for tumor cell killing by activated macrophages. Proc Natl Acad Sci USA. 1986; 83: 5233-5237. 213. Mannel, D. N.; Moore, R. N.; Mergenhagen, S. E. Macrophages as a source of tumoricidal activity (tumor-necrotizing-factor). Infect Immun. 1980; 30: 523-530. 214. Beutler, B.; Cerami, A. Tumor necrosis, cachexia, shock and inflammation: a common mediator. Ann Rev Biochem. 1988; 57: 505-518. 215. Decker, T.; Lohmann-Matthes, M-L; Gifford, G. E. Cell-associated tumor necrosis factor (TNF) as a killing mechanism of activated cytotoxic macrophages. J Immunol. 1987; 138: 957-962. 216. Klostergaard, J.; Stoltje, P. A.; Kull, F. C. Jr. Tumoricidal effector mechanisms of murine BCG-activated macrophages: role of TNF in conjugation-dependent and conjugation-independent pathways. J Leukocyte Biol. 1990; 48: 220-228. 217. Bakouche, O.; Ichinose, Y.; Heicappell, R.; Fidler, I. J.; Lachman, L. B. Plasma membrane-associated tumor necrosis factor. A non-integral membrane protein possibly bound to its own receptor. J Immunol. 1988; 140: 1142-1147. 218. Keller, R.; Keist, R.; Wechsler, A.; Leist, T. P.; van der Meide, P. H. Mechanism of macrophage-mediated tumor cell killing: a comparative study of the role of reactive nitrogen intermediates and tumor necrosis factor. Int J Cancer. 1990; 46: 682-689. 59 219. Hibbs, J. B. Jr.; Vavrin, Z.; Taintor, R. R. L-arginine is required for expression of the activated macrophage effector mechanism causing selective metabolic inhibition of target cells. J Immunol. 1987; 138: 550-565. 220. Tayeh, M. A.; Marietta, M. A. Macrophage oxidation of L-arginine to nitric oxide, nitrite and nitrate. Tetrahydrobiopterin is required as a cofactor. J Biol chem. 1989; 264: 19654-19658. 221. Hibbs, J. B. Jr.; Taintor, R. R.; Vavrin, Z.; Rachlin, E. M. Nitric oxide: a cytotoxic activated macrophage effector molecule. Biochem Biophys Res Commun. 1988; 157: 87-94. 222. Keller, R.; Geiges, M.; Keist, R. L-arginine-dependent reactive nitrogen intermediates as mediators of tumor cell killing by activated macrophages. Cancer Res. 1990; 50: 1421-1425. 223. Keller, R.; Keist, R.; Klauser, S.; Schweiger, A. The macrophage response to bacteria: flow of L-arginine through the nitric oxide and urea pathways and induction of tumoricidal activity. Biochem Biophys Res Commun. 1991; 177: 821-827. 224. Lancaster, J. R. Jr.; Hibbs, J. B. Jr. EPR demonstration of iron-nitrosyl complex formation by cytotoxic activated macrophages. Proc Natl Acad Sci USA. 1990; 87: 1223-1227. 225. Pellat, C ; Henry, Y.; Drapier, J. C. IFN-gamma-activated macrophages: detection by electron paramagnetic resonance of complexes between L-arginine-derived nitric oxide and non-heme iron proteins. Biochem Biophys Res Commun. 1990; 166: 119-125. 226. Lepoivre, M.; Chenais, B.; Yapo, A.; Lemaire, G.; Thelonder, L.; Tenu. J.P. Alterations of ribonucleotide reductase activity following induction of the nitrate-generating pathway in adenocarcinoma cells. J Biol Chem. 1990; 265: 141-143. 227. Stuehr, D. J.; Nathan, C. F. Nitric oxide. A macrophage product responsible for cytostasis and respiratory inhibition in tumor target cells. J Exp Med. 1989; 169: 1543-1555. 228. Keller, R.; Keist, R. Abilities of activated macrophages to manifest tumoricidal activity and to generate reactive nitrogen intermediates: a comparative study in vitro and ex vivo. Biochem Biophys Res Commun. 1989; 164: 968-973. 229. Higuchi, M.; Higashi, N.; Taki, H.; Osawa, T. Cytolytic mechanisms of activated macrophages. Tumor necrosis factor and L-arginine-dependent mechanisms act synergistically as the major cytolytic mechanisms of activated macrophages. J Immunol. 1990; 144: 1425-1431. 230. Matthews, N.; Neale, M. L. Studies on the mode of action of tumor necrosis factor on tumor cells in vitro. Lymphokines. 1987; 14: 223-252. 231. Lancaster, J. R. Jr.; Laster, S. M.; Gooding. L.R. Inhibition of target cell mitochondrial electron transfer by tumor necrosis factor. FEBS Letters. 1989; 248: 169-174. 232. Schulze-Osthoff, K.; Bakker, A. C ; Vanhaesebroeds, B.; Beyaert, R.; Jacob, W. A.; Fiers, W. Cytotoxic activity of tumor necrosis factor is mediated by early damage of mitochondrial function. J Biol Chem. 1992; 267: 5317-5323. 60 233. Wong, G. H. W.; Elwell, J. H.; Oberley, L. W.; Goeddel, D. N. Manganous superoxide dismutase is essential for cellular resistance to cytoxicity of tumor necrosis factor. Cell. 1989;58:923-931. 234. Davies, P.; Allison, A. C. Secretion of macrophage enzymes in relation to the pathogenesis of chronic inflammation. Immunobiology of the Macrophage. Nelson, D. S., Ed.: Academic Press, New York; 1976: 427-461. 235. Vischer, T. L.; Bretz, U.; Baggiolini, M. In vitro stimulation of lymphocytes by neutral proteinases from human polymorphonuclear leukocyte granules. J Exp Med. 1976; 144: 863-872. 236. Johnston, R. B. Jr.; Godzik, C. A.; Cohn, Z. A. Increased superoxide anion production by immunologically activated and chemically elicited macrophages. J Exp Med. 1978; 148: 115-127. 237. Nathan, C. F.; Brukner, L. H.; Silverstein, S. C ; Cohn, Z. A. Extracellular cytolysis by activated macrophages and granulocytes. I. Pharmacologic triggering of effector cell and the release of hydrogen peroxide. J Exp Med. 1979; 149: 84-99. 238. Nathan, C. F.; Silverstein, S. C ; Brukner, L. H.; Cohn, Z. A. Extracellular cytolysis by activated macrophages and granulocytes. J Exp Med. 1979; 149: 100-113. 239. Nathan, C.; Cohn, Z. A. Role of oxygen dependent mechanisms in antibody-induced lysis of tumor cells by activated macrophages. J Exp Med. 1980; 152: 198-208. 240. Adams, D. O. Effector mechanisms of cytolytically-activated macrophages. I. Secretion of neutral proteinases and effect of protease inhibitors. J Immunol. 1980; 124: 286-292. 241. Adams, D. O.; Kao, K. J.; Farb, R; Pizzo, S. V. Effector mechanisms of cytolytically-activated macrophages. II. Secretion of a cytolytic factor by activated macrophages and its relationship to secreted neutral protease. J Immunol. 1980; 124: 293-300. 242. Johnson, W. J.; Whisnant, C. C ; Adams, D. O. The binding of BCG-activated macrophages to tumor targets stimulates secretion of cytolytic factor. J Immunol. 1981; 127: 1787-1792. 243. Kitagawa, S.; Takaku, F.; Sakamoto, S. Evidence that proteases are involved in superoxide production by human polymorphonuclear leukocytes and monocytes. J Clin Invest. 1980; 65: 74-81. 244. Ferluga, J.; Schorlemmer, H. U.; Baptista, L. C ; Allison, A. C. Cytolytic effects of the complement cleavage product C3a. Br J Cancer. 1976; 34: 626-634. 245. Goodman, M. G.; Weigle, W. C ; Hugli, T. E. Inability of C3a anaphylatoxin to promote cellular lysis. Nature. 1980; 283: 78-80. 246. Currie, C. A. Activated macrophages kill tumour cells by releasing arginase. Nature. 1978; 273: 758-759. 247. Stadecker, M. J.; Calderon, J.; Karnovsky, M. L.; Unanue, E. R. Synthesis and release of thymidine by macrophages. J Immunol. 1977; 119: 1738-1743. 248. Prehn, R. T. Do tumors grow because of the immune response of the host? Transplantation. 1976; 28: 34-42. 61 249. Namba, Y.; Hanaoka, M. Immunocytology of cultured IgM-forming cells in the mouse. I. Requirement of phagocytic cell factor for the growth of IgM-forming tumor cells in tissue cultutre. J Imunol. 1972; 109: 1193-1200. 250. Nathan, C. F.; Terry, W. D. Differentiatial stimulation of murine lymphoma growth in vitro by normal and BCG-activated macrophages. J Exp Med. 1975; 142: 887-902. 251. Hewlett, G.; Opitz, H-G; Schlumberger, H. D.; Lemke, H. Growth regulation of a murine lymphoma line by a 2-mercaptoethanol or macrophage activated serum factor. Eur J Immunol. 1977; 7: 781-785. 252. Olivotto, M.; Bomford, R. In vitro inhibition of tumor growth and DNA synthesis by peritoneal and lung macrophages from mice injected with Corynebacterium parvum Int J Cancer. 1974; 13: 478-488. 253. Krahenbahl, J. L.; Lambert, L. H. Jr.; Remington, J. S. Effect of Corynebacterium parvum treatment and Toxoplasma gondii infection on macrophage-mediated cytostasis of tumour target cells. Immunology. 1976; 31: 837-846. 254. Keller, R. Cytostatic elimination of syngeneic rat tumor cells in vitro by nonspecifically activated macrophages. J Exp Med. 1973; 138: 625-644. 255. Sadler, T. E.; Jones, P. D. E.; Castro, J. E. Effect of altered phagocytic activity on the growth of primary and metastatic tumour. James, K.; McBride, W. H.; Stuart, A., Eds. The Macrophage and Cancer.: Published by the editors, Edinburgh; 1977: 155-162. 256. Underwood, J. C. W. Lymphoreticular infiltration of human tumours: prognostic and biological implications, a review. Br J Cancer. 1974; 30: 538-542. 257. Handley, W. S. The pathology of melanotic growths in relation to their operative treatment. Lancet. 1907; 1: 927-953. 258. MacCarty, W. C. Factors which influence longevity in cancer. Ann Surg. 1922; 76: 9. 259. MacCarty, W. C.; Mahle, A. E. Relation of differentiation and lymphocytic infiltration to postoperative longevity in gastric carcinoma. J Lab Clin Med. 1921; 6: 473-480. 260. Moore, O. S.; Foote, F. W. Jr. The relatively favorable prognosis of medullary carcinoma of the breast. Cancer. 1949; 2: 635-642. 261. Morrison, A. S.; Black, M. M.; Lowe, C. R. Some international differences in histology and survival in breast cancer. J Cancer. 1973; 2: 261-267. 262. Lwin, K. Y.; Zuccarini, O.; Sloane, J. P.; Beverley, P. C. L. An immunohistological study of leukocyte localization in benign and malignant breast tissue. Int J Cancer. 1985; 36: 433-438. 263. Koch, B.; Giedl, J.; Hermanek, P.; Kalden, J. R. The analysis of mononuclear cell infiltrations in colorectal adenocarcinoma. J Cancer Res Clin Oncol. 1985; 109: 142-151. 264. Allen, C. A.; Hogg, N. Association of colorectal tumor epithelium expressing HLA-D/DR with CD8-positive T-cells and mononuclear phagocytes. Cancer Res. 1987; 47: 2919-2923. 62 265. Tay, S. K.; Jenkins, D.; Maddox, P.; Hogg, N.; Singer, A. Tissue macrophage response in human papillomavirus infection and cervical intraepithelial neoplasia. Br J Obstet Gynaecol. 1987; 94: 1094-1097. 266. Brocker, E. B.; Zwadlo, G.; Holzmann, B.; Macher, E.; Sorg, C. Inflammatory cell infiltrates in human melanoma at different stages of tumor progression. Int J Cancer. 1988; 41: 562-567. 267. Shinonaga, M.; Chang, C. C ; Suzuki, N.; Sata, M.; Kuwabara, T. Immunohistological evaluation of macrophage infiltrates in brain tumors. Correlation with peritumoral edema. J Neurosurg. 1988; 68: 259-265. 268. Jass, J. R.; Love, S. B.; Northover, J. M. A. A new prognostic classification of rectal cancer. Lancet. 1987; 1 (8545): 1303-1306. 269. Champion, H. R.; Wallace, I. W.; Prescott, R. J. Histology in breast cancer diagnosis. Br J Cancer. 1972; 26: 129-138. 270. McGovern, V. J.; Mihm Jr, M. C ; Bailly, C ; Booth, J. C ; Clark Jr, W. H.; Cochran, A. J.; Hardy, E. G.; Hicks, H. D.; Levene, A.; Lewis, M. G.; Little, J. H.; Milton, G. W. The classification of malignant melanoma and its histologic reporting. Cancer. 1973; 32: 1446-1457. 271. Morrison, A. S.; Black, M. M.; Lowe, C. R. Some international differences in histology and survival in breast cancer. J Cancer. 1973; 2: 261-267. 272. Eccles, S. A.; Alexander, P. Macrophage content of tumours in relation to metastatic spread and host immune reaction. Nature. 1974; 250: 667-669. 273. Mantovani, A. Effects of in vitro tumor growth on murine macrophages isolated from sarcoma lines differing in immunogenicity and metastasizing capacity. Int J Cancer. 1978; 22: 741-746. 274. Davey, G. C ; Currie, G. A.; Alexander P. Immunity as the predominant factor determining metastasis by murine lymphomas. Br J Cancer. 1979; 40: 590-596. 275. Morantz, R. A.; Wood, G. W.; Foster, M.; Clark, M.; Gollahon, K. Macrophages in experimental and human brain tumors. Part 1: Studies of the macrophage content of experimental rat brain tumors of varying immunogenicity. J Neurosurg. 1979; 50: 298-304. 276. Moore, K.; Moore, M. Intra-tumor host cells of transplanted rat neoplasms of different immunogenicity. Int J Cancer. 1977; 19: 803-813. 277. Eccles, S. A. Host immune mechanisms in the control of tumor metastasis. Haskill, S., Ed. Tumour Immunity in Prognosis. The Role of Mononuclear Cell Infiltration. New York, Basel: Marcel Dekker; 1982: 37-74. 278. Sugarbaker, E. V. Cancer metastasis: a product of tumor-host interactions. Curr Probl Cancer. 1979; 3: 1-59. 279. Norbury, K. C ; Kripke, M. L. Ultraviolet-induced carcinogenesis in mice treated with silica, trypan blue or pyran copolymer. J Reticuloendothel Soc. 1979; 26: 827-837. 63 280. Jones, P. D.; Castro, J. E. Immunological mechanisms in metastatic spread and the antimetastatic effects of C. parvum. Br J Cancer. 1977; 35: 519-527. 281. Mantovani, A.; Giavazzi, R.; Polentarutti, N.; Spreafico, F.; Garattini, S. Divergent effects of macrophage toxins on growth of primary tumors and lung metastases in mice. Int J Cancer. 1980; 25: 617-620. 282. Eccles, S. A.; Heckford, S. E.; Alexander, P. Effect of cyclosporin A on the growth of spontaneous metastasis of syngeneic animal tumours. Br J Cancer. 1980; 42: 252-259. 283. Eccles, S. A. Host immune mechanisms in the control of tumor metastasis. Haskill, S.,.Ed. Tumour Immunity in Prognosis. The Role of Mononuclear Infiltration. Haskill, S., Ed. Marcel Dekker Inc: New York; 1982: 37-74. 284. Eccles, S. A.; Styles, J. M.; Hobbs, S. M.; Dean, C. J. Metastasis in the nude rat associated with the lack of immune response. Br J Cancer. 1979; 40: 802-805. 285. Parthenais, E.; Haskill, J. S. Immunity to the T1699 murine mammary tumor. II. Thymic influence on the in situ inflammatory response, metastatic growth and invasiveness. J Immunol. 1979; 123: 1334-1338. 286. Eccles, S. A. Macrophages and cancer. Castro, J. E., Ed. Immunological Aspects of Cancer. MTP Press: Lancaster, England; 1978: 123-154. 287. Pimm, M. V.; Baldwin, R. W. Immunology and immunotherapy of experimental and clinical metastasis. Baldwin, R. W., Ed. The Secondary Spread of Cancer. New York: Academic Press; 1978: 163-209. 288. Evans, R. Macrophage-mediated cytotoxicity: its possible role in rheumatoid arthritis. Ann N.Y. Acad Sci. 1975; 256: 275-287. 289. Loveless, S. E.; Heppner, G. H. Tumor-associated macrophages of mouse mammary tumors. I. Differential cytotoxicity of macrophages from metastatic and non-metastatic tumors. J Immunol. 1983; 131: 2074-2078. 290. Berg, J. W. Inflammation and prognosis in breast cancer; a search for host resistance. Cancer. 1959; 12: 714-720. 291. Lauder, I.; Aherne, W.; Stewart, J.; Sainsbury, R. Macrophage infiltration of breast tumors: a prospective study. J Clin Pathol. 1977; 30: 463-468. 292. Pelouze, G.; Bonenfant, J-L. Signification possible de la presence de cellales epitheliodes dans le cancer de l'estromae. Union Med Can. 1979; 108: 348-353. 293. Burg, G.; Braun-Falco, O. The cellular stromal reaction in malignant melanoma. A cytochemical investigation. Arch Dermatol Forsch. 1972; 245: 318-333. 294. Jones, W. M.; Williams W.J.; Roberts, M. M.; Davies, K. Maligant melanoma of the skin: prognostic value of clinical features and the role of treatment in 111 cases. Br J Cancer. 1968; 22: 437-441. 295. Cochran, A. J. Histology and prognosis in malignant melanoma. J Pathol. 1969; 97: 459-468. 64 296. The, T. H.; Eibergen, R.; Lamberts, H. B.; Oldhoff, J.; Ploeg, E. v. d.; Schraffordt Koops, H.; Nieweg, H. O. Immune phagocytosis in vivo of human malignant melanoma cells. Acta Med Scand. 1972; 192: 141-144. 297. Epstein, W. L.; Sagebeil, R.; Spitler, L.; Wybran, J.; Reed, W. B.; Blois, M. S. Halo nevi and melanoma. JAMA. 1973; 225: 373-377. 298. Haskill, J. S.; Yamamura, Y.; Radov, L. Host responses within solid tumors: non-thymus-derived specific cytotoxic cells within a murine mammary adenocarcinoma. Int J Cancer. 1975; 16: 798-809. 299. Russell, S. W.; Cochrane, C. G. The cellular events associated with regression and progression of murine (Moloney) sarcomas. Int J Cancer. 1974; 13: 54-63. 300. Holden, H. T.; Haskill, J. S.; Kirchner, H.; Herberman, R. B. Two functionally distinct anti-tumor effector cells isolated from primary murine sarcoma virus-induced tumors. J Immunol. 1976; 117: 440-446. 301. Haskill, J. S.; Fett, J. W. Possible evidence for antibody-dependent macrophage-mediated cytotoxicity directed against murine adenocarcinoma cells in vivo. J Immunol. 1976; 117: 1992-1998. 302. Carr, I.; Underwood, J. C. E.; McGinty, F.; Wood, P. The ultrastructure of the local lymphoreticular response to an experimental neoplasm. J Pathol. 1974; 113: 175-182. 303. Snodgrass, M. J.; Hanna, M. G. Ultrastructural studies of histiocyte-tumor cell interactions during tumor regression after intralesional injection of Mycobacterium bovis. Cancer Res. 1973; 33: 107-116. 304. Feldman, D. G.; Ehrenreich, T.; Gross, L. Electron microscopic study of the growth and regression of leukemic intradermal tumors in guinea pigs. Cancer Res. 1974; 34: 901-914. 305. Zbar, B.; Bernstein, I. D.; Bartlett, G. L.; Hanna, M. G.; Rapp, H. T. Immunotherapy of cancer: regression of intradermal tumors and prevention of growth of lymph node metastasis after intralesional injection of living Mycobacterium bovis. J Natl Cancer Inst. 1972; 49: 119-130. 306. Likhite, V. V. Rejection of tumors and metastases in Fischer 344 rats following intratumoral administration of killed Corynebacterium parvum . Int J Cancer. 1974; 14: 684-690. 307. Morton, D. L.; Eilber, F. R; Malmgren, R. A.; Wood, W. C. Immunologic factors which influence response to immunotherapy in malignant melanoma. Surgery. 1970; 68: 158-164. 308. Morton, D. L.; Eilber, F. R.; Holmes, E. C ; Hunt, J. S.; Cetcham, A. S.; Silverstein, M. J.; Sparks, F. C. BCG immunotherapy of malignant melanoma: summary of a seven year experience. Ann Surg. 1974; 180: 635-643. 309. Holmes, E. C ; Morton, D. L.; Eilber, F. R.; Golub, S. E.; Sulit, H. L. Immunotherapy in malignant melanoma. Natl Cancer Inst Monograph. 1976; 44: 85-97. 310. Moore, K.; McBride, W. H. The activation state of macrophage subpopulations from a murine fibrosarcoma. Int J Cancer. 1980; 26: 609-615. 65 311. Fidler, I. J. Inhibition of pulmonary metastasis by intravenous injection of specifically activated macrophages. Cancer Res. 1974; 34: 1074-1078. 312. Liotta, L. A.; Gattozzi, C.; Kleinerman, J.; Saidel, G Reduction of tumour cell entry into vessels by BCG-activated macrophages. Br J Cancer. 1977; 36: 639-641. 313. Den Otter, W.; Dullens, H. F. J.; van Loveren, H.; Pels, E. Antitumor effects of macrophages injected into animals. James, K.; McBride, W. H.; Stuart, A., Eds. The Macrophage and Cancer.: Published by the editors, Edinburgh; 1977: 119-141. 314. Fidler, I. J.; Barnes, Z.; Fogler, W. E.; Kirsh, R.; Bugelski, P.; Poste G. Involvement of macrophages in the eradication of established metastases following intravenous injection of liposomes containing macrophage activators. Cancer Res. 1982; 42: 496-501. 315. Wang, B. S.; Lumanglas, A. L.; Durr, F. E. Immunotherapy of a murine lymphoma by adoptive transfer of syngeneic macrophages activated by Bisantrene. Cancer Res. 1986; 46: 503-506. 316. Bartholeyns, J.; Lombard, Y.; Dumont, S.; Hartmann, D.; Chokry, M.; Giaimis, J.; Kaufmann, S.; Poindron, P. Immunotherapy of cancer: experimental approach with activated macrophages proliferating in culture. Cancer Detect Prev. 1988; 12: 413-420. 317. Bartholeyns, J.; Lombard, Y.; Poindron, P. Immunotherapy of murine sarcoma by adoptive transfer of resident peritoneal macrophages proliferating in culture. Anticancer Res. 1988; 8: 145-151. 318. Bartholeyns, J.; Lopez, M.; Andreesen, R. Adoptive immunotherapy of solid tumors with activated macrophages: experimental and clinical results. Anticancer Res. 1991; 1: 1201-1204. 319. Ben-Efraim, S.; Tah, C ; Fieren, M. J.; Van dem Bemd, G. J.; Bonta, I. L. Inflammation amplifies the antitumor cytostasis by human peritoneal macrophages. Med Oncol Tumor Pharmacother. 1994; 8: 87-94. 320. Andreesen, R.; Scheibenbogen, C ; Brugger, W.; Krause, S.; Meerpohl, H. G.; Leser, H. G.; Engler, H.; Lohr, G. W. Adoptive transfer of tumor cytotoxic macrophages generated in vitro from circulating blood monocytes: a new approach to cancer immunotherapy. Cancer Res. 1990; 50: 7450-7456. 321. Brugger, W.; Scheibenbogen, C ; Krause, S.; Andreesen, R. Large-scale production of human tumor cytotoxic macrophages grown from blood monocytes of cancer patients . Cancer Detect Prev. 1991; 15: 407-412. 322. Andreesen, R.; Hennemann, B. Adoptive immunotherapy with autologous macrophages: current status and future perspectives. Pathobiol. 1991; 59: 259-263. 323. Poste, G.; Bucana, C ; Raz, A.; Bugelski, P.; Kirsh, R.; Fidler, I. J. Analysis of the fate of systemically administered liposomes and implications for their use in drug delivery. Cancer Res. 1982; 42: 1412-1422. 324. Fidler, I. J.; Poste, G. Macrophage-mediated destruction of malignant tumor cells and new strategies for the therapy of metastatic disease. Springer Semin Immunopathol. 1982; 5: 161-174. 66 325. Fidler, I. J. Therapy of spontaneous metastases by intravenous injection of liposomes containing lymphokines. Science . 1980; 208: 1469-1471. 326. Fidler, I. J.; Raz, A.; Fogler, W. E.; Kirsh, R.; Bugelski, P.; Poste, G. Design of liposomes to improve delivery of macrophage-augmenting agents to alveolar macrophages . Cancer Res. 1980; 40: 4460-4466. 327. Hart, I. R.; Fogler, W. E.; Poste, G.; Fidler, I. J. Toxicity studies of liposome-encapsulated immunomodulators administered intravenously to dogs and mice. Cancer Immunol Immunother. 1981; 10: 157-166. 328. Fidler, I. J.; Barnes, Z.; Fogler, W. E.; Kirsh, R.; Bugelski, P.; Poste, G. Involvement of macrophages in the erradication of established metastases following intravenous injection of liposomes containing macrophage activators. Cancer Res. 1982; 42: 496-501. 329. Fidler, I. J. The in situ induction of tumorcidal activity in alveolar macrophages by liposomes containing muramyldipeptide is a thymus-independent process. J Immunol. 1981; 127: 1719-1720. 330. Fidler, I. J.; Sone, S.; Fogler, W. E.; Barnes, Z. L. Erradication of spontaneous metatases and activation of alveolar macrophages by intravenous injection of liposomes containing muramyldipeptide. Proc Natl Acad Sci USA. 1981; 78: 1680-1684. 331. Obrist, R.; Sandberg, A. L. Enhancement of macrophage invasion of tumors by administration of chemotactic factor-antitumor antibody conjugates. Cell Immunol. 1983; 81: 169-174. 332. Evans, R. Tumor macrophages in host immunity to malignancies. Fink, M. A., Ed. The Macrophage in Neoplasia. New York: Academic Press; 1976: 27-42. 333. Evans, R. Macrophage accumulation in primary and transplanted tumors growing C5-deficient B 10.D2/oSn mice. Int J Cancer. 1980; 26: 227-229. 334. Evans, R.; Madison, L. D.; Eidlen, D. M. Cyclophosphamide-induced changes in the cellular composition of a murine methylcholanthrene-induced tumor and their relation to bone marrow and blood leukocyte levels. Cancer Res . 1980; 40: 395-402. 335. Evans, R. Macrophages and neoplasms: new insights and their implication in tumor biology. Cancer Metastasis Rew. 1982; 1: 227-239. 336. Evans, R. Macrophages in syngeneic animals. Transplantation. 1972; 14: 468-473. 337. Pross, H. F.; Kerbel, R. S. An assessment of intratumor phagocytic and surface marker bearing cells in a series of autochthonous and early passaged chemically induced murine sarcomas. J Natl Cancer Inst. 1976; 57: 1157-1163. 338. McBride, W. H.; Woodruff, M. F. A.; Forbes, G. H.; Moore, K. Effect of C. parvum on the number and activity of macrophages in primary and transplanted murine fibrosarcomas. Br J Cancer. 1982; 46: 448-451. 339. Prehn, R. T.; Min, J. M. Immunity to methylcholanthrene-induced sarcomas. J Natl Cancer Inst. 1957; 18: 769-778. 67 340. Globerson, A.; Feldman, M. Antigenic specificity of benzo(a)pyrene-induced sarcomas. J Natl Cancer Inst. 1964; 32: 1229-1243. 341. Menard, S.; Colnaghi, M. I.; Della-Porta, G. In vitro demonstration of tumor-specific common antigens and embryonal antigens in murine fibrosarcomas induced by 7,12-dimethyl-benz(a)anthracene. Cancer Res. 1973; 33: 478-481. 342. Kerbel, R. S.; Pross, H. F. Fc receptor-bearing cells as a reliable marker for quantitation of host lymphocytes and tumor cells. Cell. 1974; 3: 105-112. 343. Moore, K.; Moore, E. M. Intra-tumor host cells of transplanted rat neoplasms of different immunogenicities. Int J Cancer. 1977; 19: 802-813. 344. Kerbel, R. S.; Pross, H. F.; Elliot, E. V. Origin and partial characterization of Fc receptor-bearing cells found within experimental carcinomas and sarcomas. Int J Cancer. 1975; 15: 918-932. 345. Kerbel, R. S. Resistance of activated macrophages to H2-antibody-mediated cytotoxicity and Fc rosette inhibition. Nature. 1976; 259: 226-228. 346. DeLusto, E ; Haskill, J. S. In situ cytotoxic cells in a methylcholanthrene-induced tumor. J Immunol. 1978; 121: 1007-1009. 347. Talmadge, J. E.; Key, M.; Fidler, I. J. Macrophage content of metastatic and nonmetastatic rodent neoplasms. J Immunol. 1981; 126: 2245-2248. 348. Evans, R. Macrophages and neoplasms: new insights and their implication in tumor immunobiology. Cancer Metastasis Rev. 1982; 1: 227-239. 349. Kaizer, L.; Lala, P. K. Post-mitotic age of mononuclear cells migrating into TA-3(St) solid tumours. Cell Tissue Res. 1977; 10: 279-288. 350. Key, M.; Haskill, J. S. Macrophage-mediated, antibody-dependent destruction of tumor cells: in vitro identification of an in situ mechanism. J Natl Cancer Inst. 1981; 66: 102-110. 351. Stevens, T. C ; Currie, G. A.; Peacock, J. H. Repopulation of X-irradiated Lewis lung carcinoma by malignant cells and host-macrophage progenitiors . Br J Cancer. 1978; 38: 573-582. 352. Milas, L.; Faykus, M. H. Jr.; McBride, W. H.; Hunter, N.; Peters, L. T. Concomitant development of granulocytes and enhancement of metastases formation in tumor-bearing mice. Clin Exp Metastasis. 1984; 2: 181-190. 353. Evans, R.; Eilden, D. M. Macrophage accumulation in transplanted tumors is not dependent on host immune responsiveness or presence of tumor-associated rejection antigens. J Reticulendothel Soc. 1981; 30: 425-437. 354. Haskill, J. S.; Proctor, J. W.; Yamamura, Y. Host responses within solid tumors. I. Monocytic effector cells within rat sarcomas. J Natl Cancer Inst. 1975; 54: 387-392. 355. Normann, S. J.; Cornelius, J. Concurrent depression of tumor macrophage infiltration and systemic inflammation by progressive cancer growth. Cancer Res. 1978; 38: 3453-3459. 68 356. Moore, M.; Moore, K. Kinetics of macrophage infiltration of experimental rat neoplasms. James, K.; McBride, W. H.; Stuart, A., Eds. The Macrophage and Cancer. : Published by the editors, Edinburgh; 1977: 330-339. 357. Schirrmacher, V.; Shantz, G.; Clauer, K.; Komitowski, D.; Zimmermann, H-P; Lohmann-Matthes, M-L. Tumor metastasis and cell-mediated immunity in a model system in DBA/2 mice. I. Tumor invasiveness in vitro and metastasis formation in vivo. Int J Cancer. 1979; 23: 233-244. 358. Alexander, P.; Eccles, S. A.; Gauci, C. L. The significance of macrophages in human and experimental tumors. Ann N.Y. Acad Sri. 1976; 276: 124-133. 359. Stutman, O. Delayed tumour appearance and absence of regression in nude mice infected with murine sarcoma virus. Nature. 1975; 253: 141-144. 360. White, D. J.; Plumb, A. M.; Pawelec, G.; Brons, G. Cyclosporin A: an immunosuppressive agent preferentially active against proliferating T cells. Transplantation. 1979; 27: 55-58. 361. Evans, R.; Lawler, E. M. Macrophage content and immunogenicity of C57BL/6J and BALB/cByJ methylcholanthrene-induced sarcomas. Int J Cancer. 1980; 26: 831-835. 362. Szymaniec, S.; James, K. Studies on the Fc receptor bearing cells in a transplanted methylcholanthrene induced mouse fibrosarcoma Br J Cancer. 1976; 33: 36-50. 363. Meltzer, M. S.; Stevenson, M. M.; Leonard, E. J. Characterization of macrophage chemotaxins in tumor cell cultures and comparison with lymphocyte-derived chemotactic factors. Cancer Res. 1977; 37: 721-725. 364. Bottazzi, B.; Polentanutti, N.; Balsani, A.; Boraschi, D.; Ghezzi, P.; Salmona, M.; Mantovani, A. Chemotactic activity for mononuclear phagocytes of culture supernatants from murine and human tumor cells: evidence for a role in the regulation of macrophage content of neoplastic tissues. Int J Cancer. 1983; 31: 55-63. 365. Peri, G.; Biondi, A.; Bottazzi, B.; Polentarutti, N.; Bordignon, C ; Mantovani, A. Tumoricidal and immunoregulating activity of macrophages from human ovarian carcinomas. Adv Exp Med Biol. 1982; 155: 737-743. 366. Yoshimura, T.; Robinson, E. A.; Tanaka, S.; Tanaka, S.; Appella, E.; Kuratsu, J.; Leonard, E. J. Purification and amino acid analysis of two human glioma-derived monocyte chemoattractants. J Exp Med. 1989; 169: 1449-1459. 367. Matsushima, K.; Larsen, C. G.; DuBois, C. G.; Oppenheim, J. J. Purification and characterization of a novel monocyte chemotactic and activating factor produced by a human myelomonocytic cell line. J Exp Med. 1989; 169: 1485-1490. 368. Van Damme, J.; Decock, B.; Lenaerts, J. P.; Conings, R.; Bertini, R.; Mantovani, A.; Billiau, A. Identification by sequence analysis of chemotactic factors for monocytes produced by normal and transformed cells stimulated with virus, double-stranded RNA or cytokine. Eur J Immunol. 1989; 19: 2367-2373. 369. Bottazzi, B.; Colotta, F.; Sica, A.; Nobili, K; Mantovani, A. A chemoattractant expressed in human sarcoma cells (tumor-derived chemotactic factor, TDCF) is identical to monocyte chemoattractant protein-1/monocyte chemotactic and activating factor (MCP-1/MCAF). Int J Cancer. 1990; 45: 795-797. 370. Rollins, B. J.; Morrison, E. D.; Stiles, C. D. Cloning and expression of JE, a gene inducible by platelet-derived growth factor and whose product has cytokine-like properties. Proc Natl Acad Sci USA. 1988; 85: 3738-3742. 371. Oppenheim, J. J.; Zachariae, C. O. C ; Mukaida, N.; Matsushima, K. Properties of the novel proinflammatory supergene "intercrine" cytokine family. Ann Rev Immunol. 1991;9:671-648. 372. Kelvin, D. J.; Michiel, D. F.; Johnston, J. A.; Lloyd, A. R.; Sprenger, H.; Oppenheim, J. J.; Wang, J-M. Chemokines and serpentines: the molecular biology of chemokine receptors. J Leukocyte Biol. 1993; 54: 604-612. 373. Mantovani, A.; Ming, W. J.; Balotta, C.; Abdeljalil, B.; Bottazzi, B. Origin and regulation of tumor-associated macrophages: the role of tumor-derived chemotactic factor. Biochem Biophys Acta. 1986; 865: 59-67. 374. Bottazzi, B.; Polentarutti, N.; Acero, R.; Balsari, A.; Boraschi, D.; Ghezzi, P.; Salmona, M.; Mantovani A. Regulation of the macrophage content of neoplasms by chemoattractants. Science. 1983; 220: 210-212. 375. Walter, S.; Bottazzi, B.; Govoni, D.; Colotta, F.; Mantovani, A. Macrophage infiltration and growth of sarcoma clones expressing different amounts of monocyte chemotactic protein/JE. Int J Cancer. 1991; 49: 431-435. 376. Rollins, B. J.; Sunday, M. E. Suppression of tumor formation in vivo by expression of the JE gene in malignant cells. Mol Cell Biol. 1991; 11: 3125-3131. 377. Bottazzi, B.; Walter, S.; Govoni, D.; Colotta, F.; Mantovani, A. Monocyte chemotactic cytokine gene transfer modulates macrophage infiltration, growth, and susceptibility to IL-2 therapy of a murine melanoma. J Immunol. 1992; 148: 1280-1285. 378. van Damme, J.; Proost, P.; Lenaerts, J. P.; Opdenakker, G. Structural and functional identification of two human, tumor-derived monocyte chemotactic proteins (MCP-2 and MCP-3) belonging to the chemokine family. J Exp Med. 1992; 176: 59-65. 379. Wang, J-M; Griffin, J. D.; Rambaldi, A.; Chen, Z. G.; Mantovani, A. Induction of monocyte migration by recombinant macrophage colony-stimulating factor. J Immunol. 1988; 141: 575-579. 380. Fu, Y. X.; Cai, J. P.; Chin, Y. H.; Watson, G. A.; Lopez, D. M. Regulation of leukocyte binding to endothelial tissues by tumor-derived GM-CSF. Int J Cancer. 1992; 50: 585-588. 381. Dvorak, H. F.; Nagy, J. A.; Dvorak, A. M. Structure of solid tumors and their vasculature: implications for therapy with monoclonal antibodies. Cancer Cells. 1991; 3: 77-85. 382. Snodgrass, M. J.; Kaplan, A. M.; Harris, T. M. Macrophage mobility in response to carcinoma cells and their soluble by-products. J Reticuloendothel Soc. 1976; 20: 57a, Abstract 113. 383. Snodgrass, M. J.; Harris, T. M.; Geeraets, R.; Kaplan, A. M. Mobility and cytotoxicity of activated macrophages in the presence of carcinoma. J Reticuloendothel Soc. 1977; 22: 149-157. 70 384. Snodgrass, M. J.; Harris, T. M.; Kaplan, A. M. Chemokinetic response of activated macrophages to soluble products of neoplastic cells. Cancer Res. 1978; 38: 2925-2929. 385. Schuller, G. B.; Snodgrass, M. J.; Kaplan, A. M. Macrophage motility and chemotaxis in response to culture fluids from tumor cell lines. Fed Proc. 1978; 37: 1578, abstract 1697. 386. Lane, R. D.; Kaplan, A. M.; Snodgrass, M. J.; Szakal, A. K. Characterization of a macrophage chemokinetic factor in tumor cell culture media. J Reticuloendothel Sod 1982; 31: 171-184. 387. Nelson, M.; Nelson, D. S. Macrophages and resistance to tumours. I. Inhibition of delayed-type hypersensitivity reactions by tumour cells and by soluble products affecting macrophages. Immunology. 1978; 34: 277-290. 388. Aaskov, J. G.; Anthony, H. M; Assay of human lymphokines in vitro. Evidence for a migration stimulation factor (MstF) which interferes with the migration inhibition assay. Aust J Exp Biol Med Sci. 1976; 54: 527-529. 389. Weisbart, R. H.; Bluestone, R.; Goldberg, L. S.; Pearson, C. M. Migration enhancement factor: a new lymphokine. Proc Natl Acad Sci USA. 1974; 71: 875-879. 390. Codington, J. F.; Sanford, B. H.; Jeanloz, R. W. Glycoprotein coat of the TA3 cell. Isolation and partial characterization of a sialic containing glycoprotein fraction. Biochem. 1972; 11: 2559-2564. 391. Slayter, H. S.; Codington, J. F. Size and configuration of glycoprotein fragments cleaved from tumor cells by proteolysis. J Biol Chem. 1973; 248: 3405-3410. 392. Southam, C. M. The immunologic status of patients with non-lymphomatous cancer. Cancer. 1968; 28: 1433-1440. 393. Dizon, Q.; Southam, C. M. Abnormal cellular response to skin abrasion in cancer patients. Cancer. 1963; 16: 1288-1292. 394. Johnson, M. W.; Maibach, H. I.; Salmon, S. E. Skin reactivity in patients with cancer. Impaired delayed hypersensitivity or faulty inflammatory response. N Engl J Med. 1971; 284: 1255-1257. 395. Normann, S. J.; Schardt, M.; Sorkin, E. Anti-inflammatory effect of spontaneous lymphoma in SJL/J mice. J Natl Cancer Inst. 1979; 63: 825-833. 396. Cianciolo, G. J.; Herberman, R. B.; Snyderman, R. Depression of murine macrophage accumulation by low molecular weight factors derived from spontaneous mammary carcinoma. J Natl Cancer Inst. 1980; 65: 829-834. 397. Bernstein, I. D.; Zbar, B.; Rapp, H. J. Impaired inflammatory response in tumor-bearing guinea pigs. J Natl Cancer Inst. 1972; 49: 1641-1647. 398. Normann, S. J.; Sorkin, E. Cell-specific defect in monocyte function during tumor growth. J Natl Cancer Inst. 1976; 57: 135-140. 399. Snydermann, R.; Pike, M. C ; Blaylock, B. L.; Weinstein, P. Effect of neoplasms on inflammation: depression of macrophage accumulation after tumor implantation. J Immunol. 1976; 116: 585-589. 71 400. Normann, S. J.; Schardt, M.; Sorkin, E. Cancer progression and monocyte inflammatory dysfunction: relationship to tumor excision and metastasis. Int J Cancer. 1979; 23: 110-113. 401. Cianciolo, G. J.; Snyderman, R. Neoplasia and mononuclear phagocyte function. Herberman, R. B.; Friedman, H., Eds. The Reticuloendothelial System. A Comprehensive Treatise. Vol. 5, Cancer. : Plenum Press, New York; 1983: 193-216. 402. Hausman, M ; Brosman, S.; Fahey, J. L.; Snyderman, R. Defective mononuclear leukocyte chemotactic activity in patients with genitourinary carcinoma. Clin Res . 1973; 21: 646a. 403. Hausman, M ; Brosman, S.; Snyderman, R.; Mickey, M. R.; Fahey, J. Defective monocyte function in patients with genitourinary carcinoma. J Natl Cancer Inst. 1975; 55: 1047-1054. 404. Boetcher, D. A.; Leonard, E. J. Abnormal monocyte chemotaxis response in cancer patients. J Natl Cancer Inst. 1974; 52: 1091-1099. 405. Pike, M. C ; Snyderman, R. Depression of macrophage function by a factor produced by neoplasms: a mechanism for abrogation of immune surveillance. J Immunol. 1976; 117: 1243-1249. 406. Rubin, R. H.; Cosimi, A. B.; Goetzl, E. J. Defective human mononuclear leukocyte chemotaxis as an index of host resistance to malignant melanoma. Clin Immunol Immunopathol. 1976; 6: 376-388. 407. Snyderman, R.; Seigler, H. F.; Meadows, L. Abnormalities of monocyte chemotaxis in patients with melanoma: effects of immunotherapy and tumor removal. J Natl Cancer Inst. 1977; 58: 37-41. 408. Snyderman, R.; Meadows, L.; Holder, W.; Wells, S. Jr. Abnormal monocyte chemotaxis in patients with breast cancer: evidence for a tumor-mediated effect. J Natl Cancer Inst. 1978; 60: 737-740. 409. Cianciolo, G. J.; Snyderman, R. Effects of neoplasms on mononuclear phagocytes. Tumor Immunity in Prognosis. The Role of Mononuclear Infiltration. Haskill, S., Ed.: Marcel Dekker Inc., New York; 1982: 151-174. 410. Snyderman, R.; Pike, M. C. An inhibitor of macrophage chemotaxis produced by neoplasms. Science. 1976; 192: 370-372. 411. Cheung, H. T.; Cantarow, W. D.; Sundharadas, G. Characteristics of a low molecular weight factor extracted from mouse tumors that affects in vitro properties of macrophages. Int J Cancer. 1979; 23: 344-352. 412. Normann, S. J. Tumor cell threshold required for suppression of macrophage inflammation. J Natl Cancer Inst. 1978; 60: 1091-1096. 413. Snyderman, R.; Cianciolo, G. J. Further studies of a macrophage chemotaxis inhibitor (MCI) produced by neoplasms: murine tumors free of lactic dehydrogenase virus produce MCI. J Reticuloendothel Soc. 1979; 26: 453-458. 414. Normann, S. J.; Sorkin, E. Inhibition of macrophage chemotaxis by neoplastic and other rapidly proliferating cells in vitro. Cancer Res. 1977; 37: 705-711. 415. Normann, S. J.; Cornelius, J. Characterization of anti-inflammatory factors produced by murine tumor cells in culture. J Natl Cancer Inst. 1982; 69: 1321-1327. 416. Cianciolo, G.; Hunter, J.; Silva, J.; Haskill, J. S.; Snyderman, R. Inhibitors of monocyte responses to chemotaxins are present in human cancerous effusions and react with monoclonal antibodies to the P15(E) structural protein of retroviruses. Clin Invest. 1981;68:831-844. 417. Cianciolo, G. J.; Lostrom, M. E.; Tarn, M.; Snyderman, R. Murine malignant cells synthesize a 19,000-dalton protein physiochemically and antigenically related to the immunosuppressive retroviral protein P15E. J Exp Med. 1983; 158: 885-900. 418. Cianciolo, G. J.; Phipps, D.; Snyderman, R. Human malignant and mitogen-transformed cells contain retroviral P15E-related antigen. J Exp Med. 1984; 159: 964-969. 419. Nelson, M.; Nelson, D. S.; Spradbro, P. B.; Kuchroo, V. K.; Jennings, P. A.; Cianciolo, G. J.; Snyderman, R. Successful tumor immunotherapy: possible role of antibodies to antiinflammatory factors produced by neoplasms. Clin Exp Immunol. 1985; 61: 109-117. 420. Lindvall, M.; Sjogren, H. O. Inhibition of rat yolk sac tumor growth in vivo by monoclonal antibody to the retroviral molecule P15E. Cancer Immunol Immunother. 1991; 33: 21-27. 421. Cianciolo, G. J.; Kipnis, R. J.; Snyderman, R. Similarity between P15E of murine and feline leukemia viruses and P21 of HTLV. Nature. 1984; 311: 515. 422. Cianciolo, G. J. Inhibition of lymphocyte proliferation by a synthetic peptide corresponding to a region of transforming growth factor beta homologous to CKS-17, an immunosuppressive retroviral-related peptide. Clin Res. 1990; 38: 325A. 423. Cianciolo, G. J. Antiinflammatory effects of neoplasms. Res Immunol. 1993; 144: 268-270. 73 CHAPTER 2 Characterization of the mvelo-supportive capacity of fibroblastoid cell lines 1 . Introduction While it is generally agreed that the overall macrophage content of a tumour is determined both by the recruitment of monocytes from the circulation and the local proliferation and differentiation of macrophage progenitor cells within the tumour microenvironment (1-3), at present, neither the relative importance of each of these mechanisms, nor the precise molecular events involved, have been adequately defined. What is clear, however, is that while different experimental tumours may vary greatly in their macrophage content (1), the percentage of macrophages present within a particular tumour remains fairly constant both during periods of logarithmic growth and upon transplantation to secondary recipients (1-3). Moreover, metastatic tumour deposits frequently contain the same proportion of macrophages as the primary tumours from which they were derived (4). Taken together these findings suggest that signals derived from tumour cells themselves, rather than tumour-associated tissue-specific stromal elements, are primarily responsible for determining the macrophage content of a particular tumour. The long-term culture system (LTC) originally developed by Dexter and colleagues (5) provides in vitro conditions that support the continued proliferation and differentiation of primitive murine hemopoietic cells for several months, dependent only on the weekly provision of new growth medium. The sustained hemopoietic activity of these cultures is associated with the formation of a heterogeneous adherent layer containing many cells of nonhemopoietic origin (fibroblasts, adipocytes, and endothelial cells) as well as large numbers of macrophages (5,6). All of these cell types are thought to be representative of the "stromal" cells that provide a suitably supportive microenvironment within the bone marrow and spleen but normally not elsewhere. The LTC system has thus been used extensively as an in vitro model with which to investigate the cellular interactions that may regulate hemopoiesis in vivo. 74 A number of studies have shown that at least some of the hemopoiesis-supporting functions of the LTC adherent layer can be replaced by subcultured marrow fibroblasts (7) or various fibroblast cell lines (8-14). A large array of hemopoietic growth factors are known to be produced within murine LTC (15-18) and some of these can either alone or in combination stimulate both early and late stages of hemopoiesis (19). Recently, two of these, interleukin-6 (IL-6) and Steel factor/Stem cell factor (SCF), both of which are produced by fibroblasts, have been implicated as endogenous stimulators of the later stages of hemopoiesis in this system based on the inhibitory effects observed following the addition to LTC of neutralizing antibodies to these factors (20, 21). In addition, adhesive interactions involving CD44 (Pgp-1), or the binding of VLA-4 to VCAM-1 have been demonstrated to occur in the LTC system and to influence the ability of hemopoietic cells to proliferate and differentiate under these conditions (22,23). As a first step toward determining the possible contribution of local macrophage progenitor cell proliferation to the total macrophage content of a tumour, and to define some of the essential signals involved in this process, fibroblastoid tumour cells, and other non-transformed fibroblasts obtained from both the hemopoietic and non-hemopoietic tissues of various normal and mutant murine strains, were tested for their ability to replace the supportive function of the LTC adherent layer in the LTC system. Recent studies have indicated that murine cells capable of long-term repopulation in vivo copurify with cells defined as LTC-initiating cells (LTC-IC) on the basis of their ability to generate clonogenic progenitors after a minimum of 4 weeks in vitro in the presence of competent feeder layers (24, 25). We therefore chose this latter in vitro endpoint to quantify the supporting capability of various types of feeder layers. To facilitate these studies, murine bone marrow cells were separated on a nylon wool (NW) column to obtain a non-adherent population of primitive hemopoietic cells depleted of cells capable of generating a functional LTC adherent layer (26, 27) and thus ideal as an indicator population for quantitating the supportive activity of various candidate cell types. Initial experiments were undertaken to establish the utility of a co-culture assay using such indicator cells. It was then used to evaluate the supportive capacity of various fibroblastoid cell lines. Cell lines derived from two different mutant murine strains (op/op and Sl/Sl), and their 75 littermate controls, were used to investigate the role of their respective deficient factors in the support of hemopoietic progenitors. The Sl/Sl mouse carries a mutation on chromosome 10 which results in the absence of functional SCF (28). SCF is the ligand for the c-kit receptor, and is required during normal development of the embryo, being involved in cellular migration and establishment of migrated cells. It has also been shown to play a role in the proliferation, differentiation and survival of progenitor cells of a variety of lineages. The op/op mouse carries a point mutation in the macrophage colony-stimulating factor (M-CSF) gene located on chromosome 3, resulting in the insertion of a thymidine base at a point 262 bp downstream of the ATG start codon. This causes a frame shift generating a premature stop codon 27 bp further downstream (29). Cells from these mice contain only the 4.6 mRNA species of M-CSF whereas cells from their normal littermates produce the 4.6 and 2.3 species (30). The op/op mice are totally deficient in production of M-CSF which results in their bones being osteopetrotic. They are easily distinguishable from their phenotypically normal littermates by 10 days of age by the absence of incisors, dome-shaped skull, short tail and smaller size. These mice overall have a shorter life span and are poor breeders. The results obtained indicate that both transformed and non-transformed fibroblastoid cells alone, irrespective of the tissue from which they were derived, are capable of supporting hemopoiesis, although these cells were on average, only 16 to 26% as effective on a per cell basis as primary heterogeneous LTC adherent layers. No significant difference in the supportive capacity of embryonic Sl/Sl fibroblasts as compared to littermate-derived +/+ fibroblasts was detected using this assay. Fibroblastoid cells derived from M-CSF-deficient op/op mice were, however, severely compromised in their supportive ability, suggesting an important role for M-CSF in this process. 76 2 . Materials and Methods 2.1 Animals (C57BL/6J x C3H/HeJ)Fl (B6C3F1) and C3H/HeJ mice were bred and maintained in the Joint Animal Facility of the British Columbia Cancer Research Centre. Except where specified these mice were used in all experiments between 6 and 12 weeks of age. (C57BL/6J x C3HeB/FeJ)F2 B6C3Fe (osteopetrotic) op/op homozygotes and phenotypically normal littermate controls together with foster mothers, were obtained from Jackson Laboratories (Bar Harbor, ME.). 2.2 Cell lines M2-10B4 and M1B1 are murine fibroblast cell lines that were isolated previously in the Terry Fox Laboratory from the adherent layer of the same LTC initiated with adult B6C3Fi marrow (31). GB1/6 cells were obtained from Dr. J. Greenberger (University of Pittsburgh, Pittsburgh, PA) and were derived from a LTC adherent layer of B6 origin (14). WC-Re-Sl/Sl and WC-Re-+/+ cell lines were derived respectively from Sl/Sl murine embryos and their +/+ littermate controls and were obtained from Dr. D. Cook (University of North Carolina, NC.) (32). The murine fibroblast cell line NIH/3T3 and the SV40 transformed simian fibroblastoid cell line COS7 (33), were obtained from American Type Culture Collection (ATCC; Rockville, MD). Fsa-N is a fibroblastoid cell line derived from a fibrosarcoma that arose spontaneously in the shielded limb of an irradiated C3Hf/Sed/Kam mouse (34). Tumours derived from this cell line contain 75-80% macrophages. The cell lines 1G8, 1H1, S5-2 and AHSpCL were generated as part of this study. 1G8 and 1H1 were derived directly from the bone marrow of B6C3Fi murine fetuses removed on the 14th day of gestation. S5-2 and AHSpCL were derived from B6C3Fi spleens. These cell lines were established by placing pieces of tissue in medium containing FCS and the resulting outgrowth of fibroblasts were maintained by regular passage using 0.25% trypsin in citrate saline (Stem Cell Technologies Inc.) when they were confluent. 77 Fibroblastoid cell lines were also generated from the lungs of op/op mice and their phenotypically normal littermates by a similar technique. Tissue was pooled from two mice per group in order to increase the likelihood of successfully establishing a permanent cell line. The lungs were finely minced using surgical blades and the resulting pieces suspended in 10 ml of RPMI+20% FCS, plated in a 10 cm tissue culture dish (Falcon 3003; Becton Dickinson Labware, Mississauga, Ontario) and incubated at 37°C in a humidified atmosphere containing 5% CO2 in air. Again, an outgrowth of fibroblastoid cells was evident after about one week, and the medium along with non-adherent cellular debris was removed and 10 ml fresh RPM+20% FCS added. As the fibroblast layers became confluent they were maintained by regular passage using 0.25% trypsin in citrate saline (Stem Cell Technologies Inc.), and the FCS concentration in the medium was reduced to 10%. All of these cell lines were maintained in RPM 1640 medium (Stem Cell Technologies Inc., Vancouver, BC) containing 15%o fetal calf serum (FCS) (Hyclone, Logan, UT) (RPMI+15% FCS) at 37°C, in a humidified atmosphere containing 5% CO2 in air, except for the cell lines generated from the op/op mice and their phenotypically normal littermate controls which were maintained in RPMI+10% FCS, and WC-Re-Sl/Sl and WC-Re-+/+, Fsa-N, NIH/3T3 and COS7 which were maintained in Dulbecco's minimum essential medium (DMEM)+10% FCS, and GB1/6 cells which were maintained in RPMI+10% FCS supplemented with 10"5M hydrocortisone and 10_4M mercaptoethanol and cultured at 33°C. 2.3 Expression vectors Plasmid pGEM2MCDF10 containing a 4 kb murine M-CSF cDNA isolated from the pre-B-lymphocyte cell line 70Z/3 (35) was obtained from ATCC. A Dralll-Sacl fragment containing the entire coding region but lacking extensive 5' and 3' untranslated sequences was isolated, the ends blunted with T4 DNA polymerase, and the fragment obtained ligated into the Hindi site of pUC13/18. Following digestion with Xbal, the fragment obtained was ligated into the Xbal site of the episomal expression vector pCDM8 (36) generating the plasmid pCDM8.mM-CSF. pCDM8 contains both SV40 and Polyoma origins of replication and can thus replicate to high copy number 78 within the nucleus of both SV40 transformed simian cell lines (e.g. COS7 cells) and Polyoma transformed murine cell lines (e.g. op/opPy) (36). pRC3.mGM-CSF, a pCDM8-based expression vector containing the full length murine GM-CSF cDNA was a generous gift from Dr. Graeme J. Dougherty (Terry Fox Laboratory, B.C. Cancer Research Centre). 2.4 Isolation of bone marrow cells For each experiment in which bone marrow cells were used, these were pooled from at least 3 B6C3F1 mice. Cells were obtained by removing the femurs, flushing them with RPM+5% FCS using a sterile 3 ml syringe and a 21 gauge needle, followed by gentle dispersion by repeated passage through the needle. They were then counted and maintained on ice until used. 2.5 Isolation of bone marrow-derived hemopoietic progenitor cells The procedure used was a modification of that originally proposed by Julius et al. (37). Briefly, one gram of nylon wool (NW; Fenwal Laboratories, IL) was packed into a 6 ml syringe giving a column 3 cm high. After autoclaving, it was saturated with RPM+5% FCS and incubated for 1 h at 37°C. 1 x 108 freshly isolated murine bone marrow cells in 1 ml of RPMI+5% FCS were then added to the top of the column and incubated for a further 1 h at 37°C after which, non-adherent cells were eluted with 10 ml of warmed (37°C) RPM+5% FCS. 2.6 Preparation of bone marrow-derived and fibroblast adherent layers Adherent bone marrow-derived stromal cell layers were established by placing 2 x 106 fresh bone marrow cells in 1 ml of myeloid long-term medium (LTM; Stem Cell Technologies, Inc.,Vancouver, B.C.) into each well of a 24-well plate (Nunc). LTM consisted of an alpha-based medium supplemented with 12.5% fetal calf serum, 12.5% horse serum, 2-mercaptoethanol at 10" 4 M, inositol at 40 mg/L and folic acid at 10 mg/L to which freshly dissolved hydrocortisone 21-hemisuccinate (Sigma Chemicals, Co., St. Louis, MO) to give a final concentration of 10"6 M. Cultures were maintained at 33°C, in a humidified atmosphere of 5% CO2 in air for 3-4 weeks 79 before use with weekly replacement of half of the medium. Control adherent layers devoid of viable hemopoietic progenitors were obtained by exposing 3 week or older cultures to 15 Gy of y-rays (137Cesium, 400 cGy/min). To prepare fibroblast adherent layers, 3 x 105 cells were added to the wells of a 24-well plate (Nunc), incubated for approximately 18 h to allow the cells to adhere and spread out, then the plates were irradiated with a single fraction of y-rays from a 137Cesium source. The radiation dose used (30-100 Gy) varied with each cell line and was that which in initial studies inhibited cell growth while maintaining as high a level of cell viability as possible. Viability was determined by trypan blue exclusion. 2.7 Preparation of conditioned media (CM) COS7 cells were transfected with plasmid DNA by electroporation using the Bio-Rad Gene-Pulsar System (Bio-Rad). Briefly, log phase cells were trypsinized and resuspended in phosphate buffered saline (PBS) at a final concentration of 5 x 106 cells/ml. 0.4 ml aliquots were mixed with approximately 2.5 \ig of plasmid DNA (CDM8, pCDM8.mM-CSF or pRC3.mGM-CSF), transferred to a 0.4 cm cuvette and electroporated at 280 volts with a capacitance setting of 250 microfarads. The time constants obtained ranged from 6.5 to 7.4 ms. After electroporation, cells were incubated on ice for 5 min then diluted in 30 ml DMEM+10% FCS and plated in a 15 cm tissue culture dish (Falcon 3025; Becton Dickinson). Seventy-two hours later the supernatants were removed, centrifuged at 1200rpm for 10 min to remove cellular debris, filtered through a 0.45 nM filter (Syrfil-MF; Costar Corp., Cambridge, MA) and stored at -20°C until required. Aliquots were thawed only once and any unused material discarded. To obtain conditioned media, 10 cm tissue culture dishes (Falcon 3003; Becton Dickinson) were seeded with 1 x 106 cells in 10 ml DMEM+10% FCS. Cultures were incubated for 48 h, at •J 37°C and the supernatants collected, processed and stored as described above. 80 2.8 CFU-M assay Peripheral blood was collected from the inferior vena cava of anaesthetized C3H/HeJ mice using a 1 ml syringe fitted with a 21 gauge needle, and sodium heparin (Stem Cell Technologies Inc.) added at a final concentration of 10 u/ml to prevent clotting. The blood was diluted 1:4 in PBS and 8 ml gently layered over 4 ml Lympholyte M (Cedarlane, Hornby, Ontario) in a 15 ml conical centrifuge tube (Falcon 2096; Becton Dickinson). Tubes were centrifuged at 1800 rpm for 20 min at room temperature and the peripheral blood leukocytes (PBL) located at the blood/Lympholyte M interface collected, washed once with Hank's balanced salt solution (HBSS) and resuspended in alpha modified Eagles medium (aMEM; Stem Cell Technologies Inc.) containing 20% fetal clone I (FCI; Hyclorie) (aMEM+20% FCI). 1 x 104 PBL with or without varying concentrations of recombinant M-CSF or GM-CSF (Genzyme Corp., Boston, MA) or dilutions of CM (see above) were added in a final volume of 1 ml aMEM+20% FCI to the wells of a 24-well plate (Falcon 3047; Becton Dickinson). The plates were incubated for 10-12 days at 37°C, and the number of macrophage colonies produced quantitated by washing the wells twice with PBS, and then staining the adherent cells for approximately 10 min with 1% methylene blue in methanol. Colonies containing >10 cells were counted using a inverted phase microscope. In order to confirm that the cells present within these adherent colonies were indeed macrophages, representitive wells were harvested by incubation with PBS containing 2 mM EDTA and cytospin preparations prepared and stained with a panel of mAbs using an indirect immunoperoxidase technique as described in Chapter 3. 2.9 RT-PCR analysis of M-CSF transcripts cDNA was prepared from 5 \ig of total RNA isolated from cell lines using the Pharmacia First Strand Synthesis Kit (Pharmacia, Baie d'Urfe, Quebec, Canada). Subsequent PCR was carried out exactly as described by the manufacturer using the following primer pair: 5' M-CSF (5'-GCTCTAGAGCTGCCCGTATGACCGCGCG-3') and 3* M-CSF (5'-GCTCTAGAGGGG GTGTTGTCTTTAAAGC-3'). Both primers contain added 5' Xba-1 restriction sites. The samples 81 were placed in a thermal cycler (Biosycler, BIOS, New Haven, CT) and cycled 30 times at 95°C for 30 s, 55°C for 30 s, and 72°C for 1 min. PCR products were separated on a 1% agarose gel, purified using a Geneclean II Kit (BIO 101 Inc., Vista, CA), digested with Xba-1 and ligated into the Xba-1 site of pUC19. Double stranded templates were sequenced by the dideoxy chain termination method (38), using T7 DNA polymerase and the reaction conditions suggested by the manufacturer (Sequenase Version 2.0 DNA Sequencing Kit, USB, Cleveland, OH). 2.10 Assay for Competitive repopulating cells (CRU) This assay was performed essentially as described previously (39, 40). Briefly, varying numbers of test cells of male B6C3FJ origin were injected intravenously together with 2 x 105 compromised female 'helper' bone marrow cells into lethally irradiated female B6C3F! mice (8.5 Gy, 250 kVp x-ray). Mice were killed 5 weeks later and DNA prepared from their bone marrow and thymus to determine whether repopulation of these tissues with male cells was detectable (>5% of total DNA present) by Southern blot analysis using a Y-specific probe. The frequency of CRU in the original test cell suspension assayed was then calculated as previously described from the proportions of negative animals in each dose group using Poisson statistics and the method of maximum likelihood (40). 2.11 CFU-S assay CFU-S were measured by injecting lethally irradiated B6C3Fj mice (8.5 Gy 250 kVp x-ray) intravenously with appropriate dilutions of cells in 0.2 ml phosphate buffered saline (PBS) + 2% FCS (41). Controls received only 0.2 ml of PBS+2% FCS. Recipients were killed 12 days later and their spleens placed in Tellyesniczky's Fixative (18 parts 70% ethanol, 1 part glacial acetic acid and 1 part formalin). Macroscopic colonies in the spleens were counted with the aid of a simple magnifying glass (x 5). 82 2.12 CFU-F assay CFU-F were assayed by plating varying concentrations of test cells in 10 cm tissue culture dishes (Becton-Dickinson) in 10 ml of LTM (Stem Cell Technologies Inc.) and incubating at 33°C for 10 days with a full change of media on day 5. On day 10 all media was removed from the dishes and these were washed twice with PBS and adherent cells fixed and stained with 1% methylene blue in methanol. Colonies of fibroblasts were identified morphologically and scored using an inverted microscope. 2.13 Determination of the supportive capacity of adherent cell lines 1 x 105 NW-separated cells in 1 ml of LTM were added per well and cultured for 4 weeks either alone, with pre-established irradiated LTC adherent layers or with irradiated fibroblast cell lines. The co-cultures were incubated at 33°C and maintained by removal of half of the culture medium and its replacement with fresh medium each week. The co-cultures were harvested at the end of 4 weeks by removing the non-adherent cells, washing the adherent layer with aMEM and then incubating with trypsin. The non-adherent and adherent cell fractions plus washings from 3 replicate wells were pooled and the cell concentration adjusted to obtain a single cell suspension containing 1 x 106 cells per ml. An aliquot of this was then plated in methylcellulose assay medium to determine the CFC content of the culture. 2.14 CFC assay 0.3 ml of the cell suspension to be asssayed were added to 3.0 ml of methylcellulose medium (0.8% methylcellulose in a-medium supplemented with 30% FCS, 2% pokeweed mitogen-stimulated spleen cell conditioned medium, 10% glutamine, 10% agar-stimulated human leukocyte conditioned medium, 1% BSA and 10"4M 2-mercaptoethanol, and 3 u/ml of human erthyropoietin) (Stem Cell Technologies Inc.) and the mixture vortexed. Two 1.1 ml aliquots were then plated in separate 35 mm Petri dishes (Stem Cell Technologies Inc.), covered and placed in a 10 cm Petri dish along with 4 ml of distilled water in a third open 35 mm dish to prevent drying. Cultures were incubated for 14 days at 37°C in a humidified atmosphere of 5% CO2 in air, and 83 standard morphological criteria used to identify erythroid (from BFU-E), granulopoietic (from CFU-GM) and mixed colonies (from CFU-GEMM). The colony numbers obtained were combined to yield total clonogenic progenitor (CFC) values. 3 .0 Results 3.1 Isolation of a stromal cell depleted, bone marrow-derived hemopoietic cell population that can be used to assay the myelo-supportive capacity of various stromal cell types In order to ultimately determine the myelo-supportive capacity of various fibroblastoid cell lines, it was necessary in the first instance to develop a simple and reproducible means of isolating large numbers of hemopoietic progenitor cells, depleted of stromal elements, that could be used as an indicator cell population. Passage of bone-marrow cells over a NW column proved the most practical. The extent to which stromal cells capable of forming a functional adherent layer were depleted by passage of bone marrow cells over a NW column was first assessed by plating serial dilutions of fresh bone marrow cells or NW-non-adherent bone marrow cells, either directly onto plastic, or onto pre-established LTC adherent layers, and the clonogenic cell content of these cultures assessed 4 weeks later. As shown in Figure 1, in the absence of a pre-established feeder, fresh bone marrow cells plated at concentrations as low as 2 x 105 cells per 2 cm2 culture (1 ml) could still form a functionally intact adherent layer able to support the production of CFC from the LTC-IC contained within the original input inoculum. However, in parallel cultures initiated with up to 107 NW-non-adherent cells no adherent layers were established and no CFC were produced. This occurred only when cultures were initiated with as many as 107 NW-non-adherent cells and then, in only 1 of 3 such experiments. The difference between the number of input NW-non-adherent cells and fresh bone marrow cells required to support the generation of an equivalent number of clonogenic progenitors 4 weeks later in the absence of a pre-established feeder, indicates that the NW separation technique reduced the concentration of cells capable of forming a functionally supportive adherent layer by at least 50-fold. Importantly, experiments in which NW-84 i |2 10 10" 10' Number of Cells Seeded 10* Figure 1. Total CFC production of cultures initiated with fresh bone marrow and NW-non-adherent cells plated directly onto plastic. Varying numbers of fresh bone marrow cells (-•-, -*.-, -•-), and NW-non-adherent cells (-©-, -*-, -D-) were seeded directly onto plastic. After 4 weeks of culture, the number of CFC present was determined. Each point represents the average CFC determinations from 2 samplings per culture. Each pair of symbols (e.g. -•-, -n-) represents a separate experiment. 85 non-adherent bone marrow cells plated onto pre-established feeder layers were shown to produce large numbers CFC confirmed that the NW-non-adherent cell population still contained LTC-IC capable of responding in the presence of appropriate stromal derived signals.(Figure 2 and Table 1). The NW-non-adherent cell population was also compared with fresh bone marrow in terms of its content of a variety of primitive hemopoietic cell types for which quantitative assays are available, including in vitro clonogenic cells, Day 12 CFU-S and CRU as well as LTC-IC. As shown in Table 1, all types of progenitors evaluated were similarly enriched in the NW-non-adherent population, approximately 2-fold relative to fresh marrow cells, although recoveries were consistently poor at approximately 10-20%. Taken together, these findings demonstrate that the NW separation technique is a simple and reproducible method for obtaining a population of cells that contains at least twice the frequency of the most primitive hemopoietic cell types (i.e., LTC-IC and CRU, as well as Day 12 CFU-S) present in fresh marrow but which is depleted at least 50-fold of cells that can generate a supportive layer essential for the production of clonogenic progeny from LTC-IC over a 4 weeks period. 3-2 Validation of the use of CFC output from NW-non-adherent cells to quantify the supportive function of stromal cell populations Next, a series of experiments was undertaken to evaluate the relationship between the number of supportive adherent cells present at the time of initiating co-cultures with a fixed number of LTC-IC (i.e., 105 NW-adherent cells) and the CFC yield 4 weeks later using irradiated cells from pre-established LTC adherent layers a source of supportive cells. As shown in Figure 3, the number of CFC present was found to be a linear function (R=0.971) of the number of irradiated adherent cells initially added to the cultures, indicating the ability of this assay to quantitate cells able to support the generation of CFC from LTC-IC. It should be noted that although all categories of colonies (erythroid, granulopoietic and mixed) were seen and included in the data shown for CFC output after 4 weeks in LTC, more than 90% of these (and in the co-cultures described below) were of the CFU-GM type. Thus the 86 i & 1 0 4 i io-10 2 , l O 1 ^ 10' io-TTUT l l l l | I I I I l l l l | 10S i o 6 I -OiQAr-l i i i I I I H | 10* J 10" 10 Number of Cells Seeded 10* Figure 2. Total CFC production of cultures initiated with NW-non-adherent cells plated onto pre-established LTC adherent layers and onto plastic. Varying numbers of NW-non-adherent cells were seeded onto pre-established LTC adherent layers (-•-, -A-, -•-) and directly onto plastic (-©-, -•-, -D-) . After 4 weeks of culture, the number of CFC present was determined. Each point represents the average CFC determinations from 2 samplings per culture. Each pair of symbols (e.g. -•-, -•-) represents a separate experiment. 87 Table 1. Frequency, enrichment and recovery of different cell types after separation of normal murine bone marrow on a NW column. Cell Type Assayed Total Cells CRU@ CFU-S* LTC-IC® CFC CFU-F Frequency in Unseparated Marrow (per 105 cells) NA 10 7 ± 1.0 4.5 ± 0.45 333 ±167 2 Frequency (per 105 cells) NA 22.7 24.3 ± 4.3 10.4 ± 1.9 361 ±194 <0.016 NW-non-adherent cells Enrichment NA 2.3 3.6 ± 1.6 2.3 ± 0.6 1.5 ± 0 . 1 <0.0083 % Recovery 7.4 ± 0.7 16.8 20.4 ± 3 . 9 17.0 ± 4.8 13.7 ± 1.9 <0.062 No, . o f Expts* (14) (3) (2) (3) (9) (3) The frequency of each cell type was determined on freshly isolated murine bone marrow cells before and after separation on a NW column as described in the Methods. Values shown are the mean ± SEM from the number of separate experiments shown in brackets. No. of Expts# = number of experiments NA = not applicable @ = experiments conducted by Steve Szilvassy and Saghi Ghaffari (Terry Fox Laboratories) CFU-S* = day 12 CFU-S 88 0 f O o o 1 0 S ION 1 0 2 ^ 10 1 0 i i i i i 111 n i i i i 111 n i i i i 111 n 10 3 10 4 10 5 10( Number of Feeder Cells Seeded Figure 3. Linear relationship of CFC production and the number of adherent supporting cells in the feeder layers of co-cultures. 2 x 105 NW-non-adherent cells were co-cultured with varying numbers of irradiated adherent cells derived from pre-established LTC adherent layers. After 4 weeks, the CFC output was determined using the methylcellulose clonogenic assay. In each experiment, the CFC determination was made from 2 samplings per culture. Results show the mean ± SEM for 3 separate experiment. (R=0.971). 89 supportive function measured was primarily a reflection of that required for the generation of monopoietic progenitor cells. 3.3 Comparison of the myelo-supportive function of fibroblasts from different sources Fibroblast cell lines derived from a variety of tissues were assayed for their ability to substitute for LTC adherent layers in supporting the production of clonogenic progenitors from the LTC-IC present in NW-non-adherent bone marrow cell populations. Five different fibroblast cell lines originally derived from murine bone marrow (M2-10B4, M1B1 GB1/6, 1G8, 1H1) and 2 from spleen (S5-2 and AHspCL) were all found to be able to support the generation of CFC although they were only between 15% and 26% as effective as whole LTC adherent layers in this regard (Table 2). Several other fibroblast cell lines derived from pooled fetal tissues (NIH/3T3, WC-Re-+/+ and WC-Re-Sl/Sl) were also able to support the production of clonogenic cells at equivalent levels. This finding is of interest since WC-RE-S1/S1 fibroblasts are genetically incapable of producing Steel factor. Fsa-N, a transformed fibroblastoid cell line that generates tumours containing 75-80% macrophages, was equivalent to the other fibroblast cell lines tested in its supportive activity. SV40 transformed simian COS7 cells, however, showed a markedly reduced supportive capacity (7% of the LTC adherent layer controls). 3.4 Characterization of op/op and op/+ fibroblast cell lines. In order to define the possible involvement of M-CSF in the myelo-supportive capacity of fibroblastoid cells, lines were established from the lungs of op/op mice and their phenotypically normal littermates and tested for their ability to produce M-CSF. Mice which are homozygous for the op gene are easily distinguished from phenotypically normal littermates by their smaller size, dome-shaped skulls and lack of incisors. The proliferation and differentiation of peripheral blood-derived CFU-M in vitro is critically dependent upon the presence of the hemopoietic growth factors M-CSF or GM-CSF. In the absence of one or other of these molecules no macrophage colonies are generated (Figure 4). In contrast, when present at an optimal concentration, both factors appear capable of inducing 90 Table 2. The relative supportive capacity of various fibroblast cell lines for clonogenic progenitor cells Feeder Origin of Cells Relative Supportive Capacity Fsa-N M2-10B4 M1B1 GBL/6 1G8 1H1 S5-2 AHSpCL WC-Re- +/+ WC-Re- Sl/Sl NIH/3T3 COS7 murine fibrosarcoma murine marrow fibroblasts murine spleen fibroblasts murine embryo fibroblasts simian kidney fibroblasts 21.0 18.5 23.1 16.7 20.1 25.8 23.3 15.1 16.8 18.7 25.9 7.0 +/-+/-+/-+/-+/-+/-+/-+/-+/-+/-+/-+/-3.5 2.0 1.8 1.9 2.8 4.5 3.3 2.5 1.4 3.3 8.2 3.3 (4) (4) (5) (2) (7) (4) (2) (7) (6) (10) (4) (4) Fibroblast cell lines were co-cultured with 1 x 105 NW-non-adherent cells for 4 weeks and the total number of CFC present then determined from 2 samplings per culture. Results are expressed as a percentage, +/- SEM, of the total CFC number detected in control cultures containing an irradiated pre-established LTC adherent layer. The number of separate experiments from which the data was obtained is shown in brackets. All values are significantly different from the controls (=100%) (p< 0.001), and the value for COS7 cells is less than all other fibroblast lines tested in this series (p< 0.08). P values were obtained using Student's t-test. 91 i> iO a • • MM 2. cs 5 « i % z ' « . V) 0 v m* n o - c - o « Z 600-400-200-1 00 5 a u <yj 0 U :•:•:•:•:•: •vi"*-***: •*•*•*•*•* vXvifevlv! X'X'X'XvX' !.*•*•*•*•*.*.*.*.*.*• •X*X*X*XvX '.•.*••••.•.•.*.*.*.*. >:?:•:?:$:$ T §111 *•*•*•*•*•*•*•.•.•.*.* vX*X*X*X*X • X*X*X *.*•*•*•* • XvXvX*Xv . * . * • * • * • * . * » * • * • * • * • * i i u. u. CO CO ^ s> ? s-s o 1 * O v) O O u Conditioned medium Figure 4. Requirement for M-CSF or GM-CSF in the proliferation and differentiation of CFU-M. Murine PBL were incubated in aMEM+10% FCI containing 10% COS.CDM8, COS.mM-CSF or COS.mGM-CSF conditioned medium and the number of macrophage colonies generated, was determined on day 10-12 following fixation and staining in a solution of 1% methylene blue in methanol. Results shown represent the mean of 3 values ± SEM from 3 separate experiments. 92 approximately 1:200 peripheral blood leukocytes purified by centrifugation over Lympholyte M to produce adherent macrophage colonies which, by day 14, contain 20-200 cells (Figure 4). Although there are clear differences in the morphology of the adherent cells present within the colonies induced by M-CSF and GM-CSF (Figure 5), in both instances, >95% of the cells produced are CD18-, CD44- and CD45-positive and express the macrophage markers Mac-1 (CD1 lb) and F4/80 (data not shown). The cell lines derived from both the op/op mice and their phenotypically normal littermates expressed approximately equal levels of M-CSF mRNA as determined by Northern blot analysis (Figure 6). However, only medium conditioned by the cell line derived from the phenotypically normal littermates was able to support the generation of macrophage colonies in vitro from peripheral blood precursor cells (Figure 7). Medium conditioned by the op/op cell line had essentially no supportive activity even when tested at a high concentration. In agreement with previous reports, sequence analysis of RT-PCR products demonstrated that the M-CSF gene present within a transformed derivative of the op/op cell line (see Chapter 3) contained an additional thymidine residue 262 bp downstream of the ATG start codon. The presence of this additional base produces a shift in reading frame and that generates a premature stop codon 27 bp down stream of the point of insertion (Figure 8). Similar RT-PCR analysis of a transformed derivative of the cell line derived from the phenotypically normal littermates revealed the presence of both wild-type and mutant M-CSF transcripts indicating that at least one of the animals from which this line had been derived was a heterozygote (op/+) (Figure 8). Taken together, these data confirm that the op/op fibroblastoid cell line utilized in this study was deficient in its production of biologically active M-CSF. 3-5 Mvelo-supportive ability of op/op and op/+ cell lines The op/op and op/+ cell lines were then assayed for their ability to support the production of clonogenic progenitors from the LTC-IC present in NW-non-adherent bone marrow cell populations. As shown in Table 3, M-CSF deficient op/op cells were markedly reduced in their supportive activity in comparison to the op/+ cells. It should be noted, however, that the op/+ cells «» '-* # ff • • * # * * w 93 Figure 5. Morphology of macrophage colonies grown in M-CSF or GM-CSF. Murine PBL were incubated in aMEM+10% FCI containing (a) 10% COS.mM-CSF, or (b) 10% COS.mGM-CSF conditioned medium. The macrophage colonies generated after 10-12 days were photographed following fixation and staining in a solution of 1% methylene blue in methanol. 94 9.49 kb -7.46 kb -4.40 kb -2.37 kb -1.35 kb -0.24 kb -<>• a o + o Figure 6. Northern blot analysis of M-CSF mRNA expression in cell lines derived from op/op mice and their phenotypically normal (+/?) littermates. 10 ng of total cellular RNA isolated from cell lines derived from op/op mice and their phenotypically normal (+/?) littermates was run in each lane and filters hybridized with a 3 2P-labeled probe that encompassed the entire coding region of the murine M-CSF cDNA. 95 500-1 a CQ o ft. 2 O u o +/? op/op r 2 4 6 8 10 % conditioned medium 12 Figure 7. M-CSF production by cell lines derived from op/op mice and their phenotypically normal (+/?) littermates. Murine PBL were incubated for 10-12 days in aMEM+10% FCI containing various concentrations of medium conditioned by cell lines derived from op/op mice and their phenotypically normal (+/?) littermates. The number of macrophage colonies generated was determined following fixation and staining in a solution of 1% methylene blue in methanol in dulpicate cultures. Results shown here are from a single representative experiment. 96 • +/7-PY M-CSF RT-PCR products - GCC TTT TTT CTG GTA CAA GAC ATA ATA GAT GAG - GCC TTT TTT CTG GTA CAA GAC ATA ATA GAT GAG - GCC TTT TTT CTG GTA CAA GAC ATA ATA GAT GAG - GCC TTT TTT phT GGT ACA AGA CAT AAT AGA |TGA| G TjCT GGT ACA AGA CAT AAT AGA TGA G — GCC TTT TTT op/op-PY M-CSF RT-PCR products GCC TTT TTT GCC TTT TTT GCC TTT TTT GCC TTT TTT GCC TTT TTT TCT GGT ACA AGA CAT AAT AGA TCT GGT ACA AGA CAT AAT AGA TCT GGT ACA AGA CAT AAT AGA TCT GGT ACA AGA CAT AAT AGA TJCT GGT ACA AGA CAT AAT AGA TGA TGA TGA TGA TGA G G G G G Figure 8. RT-PCR analysis of M-CSF transcripts in cell lines derived from op/op mice and their phenotypically normal (+/?) littermates M-CSF cDNAs were amplified by RT-PCR, subcloned into pUC19 and individual clones sequenced. Table 3. The relative supportive capacity op/op and op/+ fibroblast feeders for clonogenic progenitor cells Cell line Tissue Origin Relative Supportive Capacity Expt 1 Expt 2 op/op lung 0.46 0.66 op/+ " 4.3 7.5 Op/op and op/+ cell lines were co-cultured with 1 x 105 NW-non-adherent cells for 4 weeks and the total number of CFC present was determined from 2 samplings per culture. Results are expressed as a percentage of the total CFC number detected in control cultures containing an irradiated pre-established LTC adherent layer of B6C3Fi origin. The values obtained in each of 2 separate experiments are shown. 97 a were, themselves, poorly supportive relative to other fibroblastoid cells, perhaps reflecting their heterozygous genotype. 4.0 Discussion Previous studies have shown that some of the supportive functions of LTC adherent layers generated from unseparated bone marrow can be replaced by fibroblastoid cells subcultured from the adherent layer (7) or by various fibroblast cell lines (8-14). The assay for supportive activity described here utilizes a 4 week period of incubation prior to assessment of the number of CFC produced since this has been shown to measure the functional output of an initial (LTC-IC) population with phenotypic properties similar to the cells that are capable of long-term in vivo repopulation (CRU) (25, 42). In the present study we have shown that the LTC-IC content of normal murine marrow cells is doubled when these cells are passed through a NW column. At the same time, this procedure decreases the frequency of CFU-F by at least 300 fold and the myelo-supportive capacity of the cell suspension by a factor of at least 50. Thus NW separation provides a convenient and highly effective method for obtaining a population of indicator LTC-IC essentially free of cells capable of generating a supportive feeder layer. This makes them ideally suited for investigations of the cell types (or factors produced by them) that stimulate the production of clonogenic cells from murine LTC-IC. Our finding that various fibroblastoid cells can serve as a substitute for LTC adherent layers for the support of monopoietic precursors, albeit less effectively on a per cell basis, came somewhat as a surprise in view of published data implying that this property might be relatively rare (20). The observation that Sl/Sl and +/+ fibroblast cell lines do not differ significantly in their supportive capacity would suggest further that Steel factor does not play an essential role in this process, at least as defined using this particular assay system. It is appreciated, however, that assuming a standard deviation of ± 6% in the supportive capacity of each of the two cell populations, then at least 27 experiments would need to be performed in order to ensure that the 98 probability of finding the mean supportive capacity of the cell lines to be identical, when they in fact differ from one another by 1 standard deviation, would be less than 5%. The conclusion that Steel factor is not essential for the maintenance of LTC-IC in the murine LTC system is in agreement with the findings of Kodama et al. (20) and Wineman et al. (13). Similar observations have also been made in the human system where CFC production from cell populations that were highly enriched in their LTC-IC content were shown to be supported approximately equally well by a variety of murine fibroblasts including those of Sl/Sl origin (14, 32,43). In addition, the latter studies revealed an equivalent human LTC-IC supportive activity of specific factors of both fibroblast and non-fibroblast origin (32). Alternative modes of primitive hemopoietic cell stimulation has also been well established in the murine system from studies of the response of 5-Fluorouracil-resistant blast colony-forming cells (19). Thus it is possible that murine LTC-IC might show similar responses to different factors produced by established fibroblast cell lines by comparison to fibroblast-containing primary LTC adherent layers. Recent studies in the human system have shown that the regulation of analogously defined LTC-IC may also differ according to whether LTC-IC maintenance or production of progeny CFC is measured (32). Similar findings have been obtained from studies of the in vitro responses of a rare subpopulation of murine bone marrow that is highly enriched in CRU (44) and LTC-IC (45). For example, in the presence of Steel factor and IL-6 these cells are stimulated over a 3 to 4 week period to produce a large (1000-fold) expansion of cells with the original (SCA-1+ Lin-WGA+) phenotype without, however, any significant overall change in CRU numbers (44). In addition, cells defined operationally as LTC-IC may include biologically distinct subpopulations, although it has not yet been possible to identify differences in LTC-IC phenotype that are related to consistent differences in their functional potential (46). Nevertheless, CFC output from murine LTC-IC may be regulated somewhat differently from other LTC-IC functions (e.g., CFU-S output). Further studies to compare such endpoints would be necessary to address these possibilities. In previous studies, cells from the non-adherent fraction of established LTC were frequently used for this purpose, or CFC (or CFU-S) maintenance was followed for shorter periods, typically only 2 to 3 weeks. Since it is known that the non-adherent fraction of LTC is highly depleted of the most 99 primitive hematopoietic cells (47-49) (presumably due to the longer period of time allowed for cell adherence and the different molecules available for adherent cell attachment), it is possible that the activities measured here may not be the same as those previously ascribed to other fibroblast lines. At present there is little information as to what the factor(s) that stimulate either murine or human LTC-IC might be. In spite of previous data suggesting that sustained hemopoiesis in LTC requires physical contact between the supportive elements and their targets (50), more recent studies, at least in the human system, indicate that this is not the case (51, 52). Moreover, certain soluble recombinant factors have been identified that can mimic the adherent layer of LTC in supporting the production of CFC in vitro for periods of at least 3 weeks (32, 53). On the other hand, the production of membrane or matrix-bound factors by stromal cells (fibroblasts) and whole LTC adherent layers has also been documented (54-58). Previous studies of the activities present in media conditioned by M2-10B4 cells failed to detect any IL-3, GM-CSF, G-CSF, IL-4 or IL-6 (31,43) suggesting that none of these factors are necessarily involved in the mechanism(s) by which such fibroblasts support CFC production from murine LTC-IC. When added to methylcellulose assays of normal murine bone marrow, M2-10B4 conditioned media provided a weak stimulation of macrophage colony formation (consistent with some M-CSF production by M2-10B4 cells) but no other colony type (unpublished findings). The observation that M-CSF deficient op/op fibroblasts exhibited a very poor myelo-supportive capacity suggests either a direct role for this cytokine in determining the number of CFC obtained from LTC-IC when these are co-cultured with fibroblasts in the absence of added growth factors, or an indirect effect of M-CSF on other cell types which, in turn, are involved in the generation of CFC. In addition to fibroblasts, the adherent layer of the murine LTC contains by 4 weeks, a significant number of macrophages (5, 16). These cells are known to produce a variety of soluble mediators which can be shown to alter the expression of growth factors and adhesion molecules on fibroblasts. For example, macrophage-derived products like IL-1 and TNF-a have been shown to influence the expression of ICAM-1, GM-CSF and G-CSF in fibroblasts and endothelial cells (59, 60), as well as cells present within murine LTC adherent layers (17, 61). These cytokines can also modulate the composition of the extracellular matrix (59, 62). The greater supportive capacity of 100 primary LTC adherent layers observed here by comparison to any fibroblast cell line thus far tested may be a direct (or indirect) consequence of a decreased number of macrophages in adherent layers established initially from pure fibroblast cell populations. For example, fibroblasts have been shown to be capable of secreting many factors that are involved in the support of hemopoiesis in vitro particularly following appropriate induction. Thus reduced exposure to such inducing signals might account for their limited capacity to fully replace the function of heterogeneous LTC adherent layers, In summary, I have described a method for quantifying the net capacity of various adherent cell types to support the generation of hemopoietic clonogenic cell progeny from LTC-IC using NW-non-adherent normal murine bone marrow cells as a sensitive indicator population. The assay demonstrates a linear dose response relationship between the number of feeder cells initially seeded and the CFC output after 4 weeks of culture, and has been used to demonstrate the equivalent, albeit lineage-restricted, supportive capacity of a variety of normal and transformed fibroblastoid cell lines. Furthermore, analysis of the relative supportive capacity of fibroblastoid cells from mice that represent natural homozygous Steel factor and M-CSF knockouts (i.e. Sl/Sl and op/op mice, respectively) by comparison to their +/+ and op/+ controls have provided evidence that M-CSF but not Steel factor may play an essential role in the mechanism(s) by which fibroblasts support the generation of monopoietic cells from LTC-IC in vitro. 101 5 . References 1. Evans, R. Macrophages in syngeneic animal tumours. Transplantation. 1972; 14: 468-473. 2. McBride, W. H. Phenotype and functions of intratumoral macrophages. Biochem Biophy Acta. 1986; 865: 27-41. 3. Mantovani, A.; Ming, W. J.; Balotta, C.; Abdeljalil, B.; Bottazzi, B. Origin and regulation of tumor-associated macrophages: the role of tumor-derived chemotactic factor. Biochem Biophys Acta 1986; 865: 59-67. 4. Walter, S.; Bottazzi, B.; Govoni, D.; Colotta, F.; Mantovani, A. Macrophage infiltration and growth of sarcoma clones expressing different amounts of monocyte chemotactic protein/JE. Int J Cancer. 1991; 49: 431-435. 5. Dexter, T. M.; Allen, T. D.; Lajtha, L. G. Conditions controlling the proliferation of hemopoietic stem cells in vitro. J Cell Physiol. 1977; 91: 335-344. 6. Perkins, S.; Fleischman, R. A. Hematopoietic microenvironment. Origin, lineage, and transplantability of the stromal cells in long-term bone marrow cultures from chimeric mice. J Clin Invest. 1988; 81: 1072-1080. 7. Brockbank, K. G. M.; de Jong, J. P.; Piersma, A. H.; Voerman, J. S. A. Hemopoiesis on purified bone-marrow-derived reticular fibroblasts in vitro. Exp Hematol. 1986; 14: 386-394. 8. Collins, L. S.; Dorshkind, K. A stromal cell line from myeloid longterm bone marrow cultures can support myelopoiesis and B lymphopoiesis. J Immunol. 1987; 138: 1082-1087. 9. Kodama, H-A; Amagai, Y.; Koyama, H.; Kasai, S. A new preadipose cell line derived from newborn mouse calvaria can promote the proliferation of pluripotent hemopoietic stem cells in vitro. J Cell Physiol. 1982; 112: 89-95. 10. Leung, L. C ; Johnson, G. R. In vitro maintenance of hemopoietic stem cells with lymphoid and myeloid repopulating ability by a cloned murine adherent bone marrow cell line. Exp Hematol. 1987; 15: 989-984. 11. Hunt, P.; Robertson, D.; Weiss, D.; Rennick, D.; Lee, F.; Witte, O. N. A single bone marrow-derived stromal cell type supports the in vitro growth of early lymphoid and myeloid cells. Cell. 1987; 48: 997-1007. 12. Song, Z. X.; Shadduck, R. K.; Innes, D. J. Jr.; Waheed, A.; Quesenberry, P. J. Hematopoietic factor production by a cell line (TC-1) derived from adherent murine marrow cells. Blood. 1985; 66: 273-281. 13. Wineman, J. P.; Nishikawa, S-I; Muller-Sieburg, C. E. Maintenance of high levels of pluripotent hematopoietic stem cells in vitro: Effect of stromal cells and c-kit. Blood. 1993;81:365-372. 14. Anklesaria, P.; Klassen, V.; Sakakeeny, M. A.; FitzGerald, T. J.; Harrison, D.; Rybak, M. E.; Greenberger, J. S. Biological characterization of cloned permanent stromal cell lines from anemic Sl/Sld mice and +/+ littermates. Exp Hematol. 1987; 15: 636-644. 102 15. Eliason, J. F.; Thorens, B.; Kindler, V.; Vassalli, P. The roles of granulocyte-macrophage colony-stimulating factor and interleukin 3 in stromal cell-mediated hemopoiesis in vivo. Exp Hematol. 1988; 16: 307-312. 16. Gimble, J. M.; Pietrangeli, C.; Henley, A.; Dorheim, M. A.; Silver, J.; Namen, A.; Takeichi, M.; Goridis, C.; Kincade, P. W. Characterization of murine bone marrow and spleen-derived stromal cells: Analysis of leukocyte marker and growth factor mRNA transcript levels. Blood. 1989; 74: 303-311. 17. Kittler, E. L. W.; McGrath, H.; Temeles, D.; Crittenden, R. B.; Kister, V. K.; Quesenberry, P. J. Biologic significance of constitutive and subliminal growth factor production by bone marrow stroma. Blood. 1992; 79: 3168-3178. 18. Gutierrez-Ramos, J. C.; Palacois, R. In vitro differentiation of embryonic stem cells into lymphocyte precursors able to generate T and B lymphocytes in vivo . Proc Natl Acad Sci USA. 1992; 89: 9171-9175. 19. Ogawa, M. Differentiation and proliferation of hematopoietic stem cells. Blood. 1993; 81: 2844-2853. 20. Kodama, H.; Nose, M.; Yamaguchi, Y.; Tsunoda, J.; Suda, T.; Nishikawa, S.; Nishikawa, S. In vitro proliferation of primitive hemopoietic stem cells supported by stromal cells: Evidence for the presence of a mechanism(s) other than that involving c-kit receptor and its ligand. J Exp Med. 1992; 176: 351-361. 21. Hudak, S.; Thompson-Snipes, L.; Rocco, C ; Jackson, J.; Pearce, M.; Rennick, D. Anti-IL-6 antibodies suppress myeloid cell production and the generation of CFU-c in long-term bone marrow cultures. Exp Hematol. 1992; 20: 412-417. 22. Miyake, K.; Weissman, I. L.; Greenberger, J. S.; Kincade, P. W. Evidence for a role of the integrin VLA-4 in lympho-hemopoiesis. J Exp Med. 1991; 173: 599-607. 23. Miyake, K.; Medina, K. L.; Hayashi, S-I; Ono, S.; Hamaoka, T.; Kincade, P. W. Monoclonal antibodies to Pgp block lympho-hemopoiesis in long-term bone marrow cultures. J Exp Med. 1990; 171: 477-488. 24. Ploemacher, R. E.; Van Der Sluijs, J. P.; Voerman, J. S. A.; Brons, N. H. C. An in vitro limiting-dilution assay of long-term repopulating hematopoietic stem cells in the mouse. Blood. 1989; 74: 2755-2763. 25. Weilbaecher, K.; Weissman, I.; Blume, K.; Heimfeld, S. Culture of phenotypically defined hematopoietic stem cells and other progenitors at limiting dilution on Dexter monolayers. Blood. 1991; 78: 945-952. 26. Piersma, A. H.; Brockbank, K. G. M.; Ploemacher, R. E.; van Vliet, E.; Brakel-van Peer, K. M. J.; Visser, P. J. Characterization of fibroblastic stromal cells from murine bone marrow. Exp Hematol. 1985; 13: 237-243. 27. Boswell, H. S.; Albrecht, P. R.; Shupe, R. E.; Williams, D. E.; Burgess, J. Role of stromal populations in hemopoietic stem cell proliferation. I. Physically distinct subpopulations of hemopoietic stem cells and stromal progenitors determine long-term culture hemopoiesis. Exp Hematol. 1987; 15: 46-53. 28. Witte, O. N. Steel locus defines new multipotent growth factor. Cell. 1990; 63: 5-6. 103 29. Yoshida, H.; Hayashi, S-I; KUnisada, T.; Ogawa, M ; Nishikawa, S.; Okamura, H.; Sudo, T.; Shultz, L. D.; Nishikawa, S-I. The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature. 1990; 345: 442-444. 30. Wiktor-Jedrzejczak, W.; Bartocci, A.; Ferante, A. W. Jr.; Ahmed-Ansari, A.; Sell, K. W.; Pollard, J. W.; Stanley, E. R. Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse. Proc Natl Acad Sci USA. 1990; 87: 4828-4832. 31. Lemoine, F. M.; Humphries, R. K.; Abraham, S. D. M.; Krystal, G.; Eaves, C. J. Partial characterization of a novel stromal cell-derived pre-B-cell growth factor active on normal and immortalized pre-B cells. Exp Hematol. 1988; 16: 718-726. 32. Sutherland, H. J.; Hogge, D. E.; Cook, D.; Eaves, C. J. Alternative mechanisms with and without Steel factor support primitive human hematopoiesis. Blood. 1993; 81: 1465-1470. 33. Gluzman, Y. SV40-transformed simian cells support the replication of early SV40 mutants. Cell. 1981; 23: 175-182. 34. McBride, W. H. Phenotype and function of intratumoural macrophages. Biochem Biophys Res Commun. 1986; 865: 27-41. 35. Ladner, M. B.; Martin, G. A.; Noble, J. A.; Wittman, V. P.; Warren, M. K.; McGrogan, M ; Stanley, E. R. cDNA cloning and expression of murine macrophage colony-stimulating factor from L929 cells. Proc Natl Acad Sci USA. 1988; 85: 6706-6710. 36. Seed, B. An LFA-3 cDNA encodes a phospholipid-linked membrane protein homologous to its receptor CD2. Nature. 1987; 329: 840-842. 37. Julius, M. H.; Simpson, E.; Herzenberg, L. A. A rapid method for the isolation of functional thymus-derived murine lymphocytes. Eur J Immunol. 1973; 3: 645-649. 38. Hattori, M.; Sakaki, Y. Dideoxy sequencing method using denatured plasmid templates. Anal Biochem. 1985; 152: 232-238. 39. Szilvassy, S. J.; Lansdorp, P. ML; Humphries, R. K.; Eaves, A. C ; Eaves, C. J. Isolation in a single step of a highly enriched murine hematopoietic stem cell population with competitive long-term repopulating ability. Blood. 1989; 74: 930-939. 40. Szilvassy, S. J.; Humphries, R. K.; Lansdorp, P. M.; Eaves, A. C ; Eaves, C. J. Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy. Proc Natl Acad Sci USA. 1990; 87: 8736-8740. 41. Till, J. E.; McCulloch, E. A. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat Res. 1961; 14: 213-222. 42. Eaves, C. J.; Cashman, J. D.; Eaves, A. C. Methodology of long-term culture of human hemopoietic cells. J Tissue Culture Methods. 1991; 13: 55-62. 43. Sutherland, H. J.; Eaves, C. J.; Lansdorp, P. M.; Thacker, J. D.; Hogge, D. E. Differential regulation of primitive human hematopoietic cells in long-term cultures maintained on genetically engineered murine stromal cells. Blood. 1991; 78: 666-672. 104 44. Rebel, V. I.; Dragowska, W.; Eaves, C. J.; Humphries, R. K.; Lansdorp, P. M. Amplification of Sca-1+ Lin- WGA+ cells in serum-free cultures containing Steel factor, interleukin-6, and erythropoietin with maintenance of cells with long-term in vivo reconstituting potential. Blood. 1994; 83: 128-136. 45. Lemieux, M. E.; Rebel, V. I.; Lansdorp, P. M ; Eaves, C. J. Quantitation and characterization of murine lympho-myeloid progenitors in vitro using a modification of the LTC-IC assay. Blood. 1993; 82 (suppl 1): 13a. 46. Sutherland, H. J.; Lansdorp, P. M ; Henkelman, D. H.; Eaves, A. C ; Eaves, C. J. Functional characterization of individual human hematopoietic stem cells cultured at limiting dilution on supportive marrow stromal layers. Proc Natl Acad Sci USA . 1990; 87: 3584-3588. 47. Mauch, P.; Greenberger, J. S.; Botnik, L.; Hannon, E.; Hellman, S. Evidence for structured variation in self-renewal capacity within long-term bone marrow cultures. Proc Natl Acad Sci USA. 1980; 77: 2927-2930. 48. Pistoia, V.; Ghio, R.; Roncella, S.; Cozzolino, F.; Zupo, S.; Ferrarini, M. Production of colony-stimulating activity by normal and neoplastic human B lymphocytes. Blood. 1987; 69: 1340-1347. 49. Harrison, D. E.; Lerner, C. P.; Spooncer, E. Erythropoietic repopulating ability of stem cells from long-term marrow culture. Blood. 1987; 69: 1021-1025. 50. Bentley, S. A. Close range cell: cell interaction required for stem cell maintenance in continuous bone marrow culture. Exp Hematol. 1981; 9: 308-312. 51. Verfaillie, C. M. Direct contact between human primitive hematopoietic progenitors and bone marrow stroma is not required for long-term in vitro hematopoiesis. Blood. 1992; 79: 2821-2826 52. Zandstra, P. W.; Eaves, C. J.; Piret, J. M. Expansion of hematopoietic progenitor cell populations in stirred suspension bioreactors of normal human bone marrow cells. Biotechnology. 1994; (in press). 53. Migliaccio, G.; Migliaccio, A. R.; Valinsky, J.; Langley, K.; Zsebo, K.; Visser, J. W. M.; Adamson, J. W. Stem cell factor induces proliferation and differentiation of highly enriched murine hematopoietic cells. Proc Natl Acad Sci USA. 1991; 88: 7420-7424. 54. Broxmeyer, H. E.; Maze, R.; Miyazawa, K.; Carow, C ; Hendrie, P. C ; Cooper, S.; Hangoc, G.; Vadhan-Raj, S.; Lu, L. The kit receptor and its ligand, steel factor, as regulators of hemopoiesis. Cancer Cells. 1991; 3: 480-487. 55. Lyman, S. D.; James, L.; Vanden Bos, T.; de Vries, P.; Brasel, K.; Gliniak, B.; Hollingsworth, L. T.; Picha, K. S.; McKenna, H. J.; Splett, R. R.; Fletcher, F. A.; Maraskovsky, E.; Farrah, T.; Foxworthe, D.; Williams, D. E.; Beckmann, M. P. Molecular cloning of a ligand for the flt3/flksine kinase receptor: A proliferative factor for primitive hematopoietic cells. Cell. 1993; 75: 1157-1167. 105 56. Hannum, C ; Culpepper, J.; Campbell, D.; McClanahan, T.; Zurawski, S.; Bazan, J. R; Kastelein, R.; Hudak, S.; Wagner, J.; Mattson, J.; Luh, J.; Duda, G.; Martina, N.; Peterson, D.; Menon, S.; Shanafelt, A.; Muench, M.; Kelner, G.; Namikawa, R.; Rennick, D.; Roncarolo, M-G; Zlotnik, A.; Rosnet, O.; Dubreuil, R; Birnbaum, D.; Lee, F. Ligand for FLT3/FLK2 receptor tyrosine kinase regulates growth of haematopoietic stem cells and is encoded by variant RNAs. Nature. 1994; 368: 643-648. 57. Gordon, M. Y.; Riley, G. P.; Watt, S. M.; Greaves, M. F. Compartmentalization of a haematopoietic growth factor (GM-CSF) by glycosaminoglycans in the bone marrow microenvironment. Nature. 1987; 326: 403-405. 58. Roberts, R.; Gallagher, J.; Spooncer, E.; Allen, T. D.; Bloomfield, F.; Dexter, T. M. Heparan sulphate bound growth factors: A mechanism for stromal cell mediated haemopoiesis. Nature. 1988; 332: 376-378. 59. Dinarello, C. A. Interleukin-1 and interleukin-1 antagonism . Blood. 1991; 77: 1627-1652. 60. Beutler, B.; Cerami, A. Cachectin and tumor necrosis factor as two sides of the same biological coin. Nature. 1986; 320: 584-588. 61. Lovhaug, D.; Pelus, L. M.; Nordlie, E. M.; Boyum, A.; Moore, M. A. S. Monocyte-conditioned medium and interleukin 1 induce granulocyte-macrophage colony-stimulating factor production in the adherent cell layer of murine bone marrow cultures. Exp Hematol. 1986; 14: 1037-1042. 62. Ito, A.; Sato, T.; Iga, T.; Mori, Y. Tumor necrosis factor bifunctionally regulates matrix metalloproteinases and tissue inhibitor of metalloproteinases (TIMP) production by human fibroblasts. FEBS Letters. 1990; 269: 93-95. 106 CHAPTER 3 Role of M-CSF in regulating the accumulation and functional activity of tumour-associated macrophages 1.0 Introduction Tumour-associated macrophages are derived from both the differentiation of monocytes recruited into a tumour site from the peripheral blood (1, 2) and the localized proliferation of macrophage progenitor cells within the tumour microenvironment (3, 4). Thus factors that are chemotactic or chemokinetic for monocytes and/or macrophage progenitor cell types, or which promote the survival, proliferation and/or differentiation of these cells might be expected to increase the total number of macrophages present within a tumour (2). As shown in Chapter 2, at least some fibroblastoid tumour cells appear capable of providing all of the signals necessary for the sustained output of macrophage progenitor cells in vitro from a more primitive hemopoietic precursor. Moreover, fibroblast-derived macrophage colony stimulating factor (M-CSF) appeared to be involved in this process. M-CSF is produced at high levels by many histologically distinct tumour cell types. In addition to its effects on granulopoietic progenitor cells it has been shown to function both as a monocyte-specific chemoattractant (5), and to promote the survival and differentiation of monocytes in vitro (6) and in vivo (7, 8). Importantly, M-CSF has also been shown to directly 'activate' macrophages (9-13) and to 'prime' these cells so that they can respond to other stimuli (14). Phase 1 clinical trials exploring the anti-tumour activity of M-CSF are presently underway in a number of centers (15, 16). In order to better define the role played by M-CSF in regulating the accumulation and functional activity of tumour-associated macrophages, fibroblastoid cell lines derived from both M-CSF-deficient op/op mice and their phenotypically normal op/+ littermate controls were transformed with a retroviral vector encoding Polyoma large T. These lines were then inoculated subcutaneously into SCID mice and both the proportion and phenotype of the macrophages present 107 within the tumours produced, was determined. The data obtained suggest that at physiological concentrations, tumour cell-derived M-CSF does not play a major role in determining the macrophage content of at least these tumours, although it does appear to modulate the functional activity of these cells. Thus although macrophages present in M-CSF deficient op/opPy tumours express the same level of 114/A10 mRNA as macrophages present within op/+Py tumours, they express far lower levels of IL-lp, TNF-a and FcRyll mRNA. Evidence is also presented that fibroblastoid cells may elaborate a soluble factor distinct from M-CSF and GM-CSF that can promote monocyte survival in vitro. 2.0 Materials and methods 2.1 Animals BALB/c-nu/nu breeders originally obtained Harlan Sprague Dawley (Indianapolis, IN), BALB/c-SCID breeders originally obtained from Dr. J. Dick (Hospital for Sick Children, Toronto, ON) and C3H/HeJ breeders originally obtained from Jackson Laboratories, were bred in the Joint Animal Facility of the British Columbia Cancer Research Center where all animals were maintained. 2.2 Cell lines The SV40 transformed simian fibroblastoid cell line COS7 (17) was obtained from the American Type Culture Collection (ATCC; Rockville, MD). The derivation and characterization of the op/op and op/+ fibroblastoid cell lines is described in detail in Chapter 2. These cells were transformed with an ecotropic retroviral vector designated CTV-4Py encoding the Polyoma Large T antigen (a generous gift from Dr. Robert Kay, Terry Fox Laboratory). Briefly, subconfluent monolayers of the op/op and op/+ cells were incubated with cell free viral supematants containing 5 ng/ml polybrene (Sigma) as previously described (18). Ten days later the cells were harvested by trypsinization and inoculated subcutaneously into nu/nu 108 recipients to select for tumourigenic cells. Tumours were excised, enzymatically digested in a mixture of Dispase and collagenase as previously described (19), and transformed cell lines re-established in vitro by culturing and expanding the enzymatically disaggregated tumour cell suspensions in Dulbecco's minimum essential medium (DMEM) (Stem Cell Technologies Inc.) containing 10% FCS (Hyclone) (DMEM+10% FCS). 2.3 Monoclonal antibodies (mAbs') Hybridomas secreting mAbs directed against the cc-subunit of Mac-1 (TIB 128) (20), the F4/80 antigen (HB198) (21), and the hapten TNP (TIB 191), were obtained from ATCC. The mAb 114/A10 is directed against a 150,000 kD antigen expressed on IL-3-dependent mast cell lines, primitive hemopoietic cells and mature tumour-associated macrophages and has been described in detail (22-24). The mAb PAb 1614 reacts with both the Polyoma large T and SV40 large T antigens (25). It was a generous gift from Dr. Jurgen Vielkind (Environmental Carcinogenesis Unit, B.C. Cancer Research Centre). All hybridomas were maintained in DMEM+10% FCS and tissue culture supernatants obtained as previously described (26). 2.4 Expression vectors The pGEM2MCDF10 plasmid containing a 4 kb murine M-CSF cDNA isolated from the pre-B-lymphocyte cell line 70Z/3 (27) was obtained from ATCC. A Dralll-Sacl fragment containing the entire coding region but lacking extensive 5' and 3' untranslated sequences was isolated, the ends blunted with T4 DNA polymerase, and the fragment obtained ligated into the Hindi site of pUC13/18. Following digestion with Xbal, the fragment obtained was ligated into the Xbal site of the episomal expression vector pCDM8 (28) generating the plasmid pCDM8.mM-CSF. pCDM8 contains both SV40 and Polyoma origins of replication and can replicate to high copy number within the nucleus of both SV40 transformed simian cell lines (eg. COS7 cells) and Polyoma transformed murine cell lines (eg. op/opPy) (28). 109 pRC3.mGM-CSF, a pCDM8-based expression vector containing the full length murine GM-CSF cDNA was a generous gift from Dr. Graeme J. Dougherty (Terry Fox Laboratory, B.C. Cancer Research Centre). 2.5 Preparation of conditioned media (CM1 COS7 cells and op/opPy cells were transfected with plasmid DNA by electroporation using the Bio-Rad Gene-Pulsar System (Bio-Rad). Briefly, cells were trypsinized and resuspended in PBS at a final concentration of 5 x 106 cells/ml. 0.4 ml aliquotes were mixed with approximately 2.5 ng of plasmid DNA (CDM8, pCDM8.mM-CSF or pCDM8.mGM-CSF), transferred to a 0.4 cm cuvette and electroporated at 280 volts with a capacitance setting of 250 microfarads. The time constants obtained ranged from 6.5 to 7.4 ms. After electroporation cells were incubated on ice for 5 min then diluted in 30 ml DMEM+10% FCS and plated in a 15 cm tissue culture dish (Falcon 3025; Becton Dickinson). Supernatants were removed 72 h later, centrifuged at 250 g for 10 min to remove cellular debris, filtered through a 0.45 |iM filter (Syrfil-MF; Costar Corp., Cambridge, MA) and stored at -20°C until required. Aliquots were thawed only once and any unused material discarded. To obtain op/op, op/+, op/opPy and op/+Py conditioned media, 10 cm tissue culture dishes (Falcon 3003; Becton Dickinson) were seeded with 1 x 106 cells in 10 ml DMEM+10% FCS. Cultures were incubated for 48 h, at 37°C in 5% humidified CC^and the supernatants collected, processed and stored as described above. 2.6 Macrophage colony assay CFU-M were assayed as previously described in Chapter 2. 2-7 Generation of op/opPv and op/+Pv tumours To produce experimental tumours with which to investigate the role played by tumour cell-derived M-CSF in regulating the recruitment and functional activity of tumour-associated 110 macrophages, 1 x 106 in vitro cultured op/opPy or op/+Py cells harvested by trypsinization were injected subcutaneously into the flank of SCID recipients. 2.8 Indirect immunoperoxidase staining Indirect immunoperoxidase staining was used to confirm the expression of the Polyoma large T antigen in retrovirally transduced cells, and to determine the number of macrophages present within enzymatically disaggregated tumour cell suspensions. Briefly, cytospin preparations were air dried, fixed in acetone for 5 min, and incubated for 30 min at room temperature with 50 \il of the appropriate mAb tissue culture supernatant. After extensive washing in HBSS, the cytospins were incubated for a further 30 min with 50 \il of a 1:2000 dilution of horseradish peroxidase-conjugated goat anti-rat IgG (Sigma Chemical Company, St Louise, MO) (TIB 128, HB198, 114/A10) or a 1:100 dilution of horseradish peroxidase-conjugated rabbit anti-murine IgG (DAKO, Dimension Laboratories Inc., Mississauga, Ontario) (PAb 1614, TIB 191). After washing in HBSS, the reaction was developed by incubating the slides for 5 min at room temperature in PBS containing 0.06% (w/v) 3-3'-diaminobenzidine (Sigma) and 0.012% (v/v) H2Q2 (Sigma). 2.9 Northern blot analysis Total cellular RNA was isolated from parental and transformed op/op and op/+ cell lines and enymatically disaggregated tumour cell suspensions, using the guanidine isothiocyanate/CsCl method as previously described (29). Ten \ig of RNA were electrophoresed through a 1% (w/v) agarose gel containing 5% (v/v) de-ionized formaldehyde , transferred to a nylon membrane (Zeta-Probe GT, Bio-Rad Laboratories Ltd., Mississauga, Ontario) and cross-linked by exposure to ultraviolet radiation (Stratalinker; Stratagene, La Jolla, CA). Filters were pre-hybridized for 1 h at 42°C in 500 mM sodium phosphate buffer, pH 7.2, 50% (v/v) formamide, 5% sodium dodecyl sulfate, 1 mM EDTA, and lmg/ml bovine serum albumin (Fraction V; Sigma) and then hybridized for 16 h at 42°C in the same solution containing denatured 32P-labelled probes prepared using the I l l Pharmacia Ready-To-Go Oligonucleotide Labelling Kit as per the manufacturers instructions. Filters were washed and developed as previously described (30). 2.10 cDNA probes Full length cDNA probes for murine IL-lp, TNF-a, FcRyll, 114/A10 and actin were obtained from Dr. Graeme Dougherty in the Terry Fox Laboratory. 2.11 Monocyte survival assay Peripheral blood was collected from C3H/HeJ mice as described above and diluted 1:4 in PBS. 10 ml aliquots were gently layered over 30 ml of 65% Percoll (Pharmacia) and centrifuged at 300 g for 20 min at room temperature. The cells located at, and immediately below, the blood/Percoll interface were collected, washed once in PBS and resuspended in aMEM+20% FCI. Based on nuclear morphology and expression of the Mac-1 antigen, greater than 80% of these cells appeared to be monocytes (data not shown). 1 x 104 cells were added to the wells of a flat bottomed 96 well plate (Falcon 3072, Becton Dickinson) together with various concentrations of conditioned media in a final volume of 100 \il and the plates incubated at 37°C, 5% CO2 for 48 h. Tissue culture supematants were then gently removed and replaced with 100 1^ of HBSS containing 1 ^ g/ml Calcein AM (Molecular Probes, Eugene, OR). Following incubation at 37°C for 30 min, the supematants were discarded and number of fluorescent viable cells determined using a Fluorescence plate reader (Millipore, Malborough, MA). Calcein AM is a fluorogenic substrate that is cleaved only in viable cells to form the esterase product calcein, a green fluorescent embrane-impermeant product (31,32). 2.12 GM-CSF bioassav 1 x 105 op/opPy and op/+Py cells were added to the wells of a 24 well plate and incubated overnight. The tissue culture medium was then discarded and replaced with 1 ml of fresh DMEM+10% FCS containing various concentrations of recombinant murine TNF-a (Cedarlane Laboratories, Hornby, Ontario) or human IL-lp (R&D Systems, Minneapolis, MN). 48 h later the 112 medium was removed, microfuged briefly to remove any cellular debris, and the level of GM-CSF determined as follows. Briefly, serial dilutions of the conditioned media or recombinant GM-CSF (Genzyme) were prepared in the wells of flat bottomed 96 well plate. 1 x 104 GM-CSF-dependent B6SUtA cells were then added to each well and the plates incubated for 48 h at 37°C. 1 ^Ci (37 Bq) 3H-thymidine was added to each well and the plates incubated for a further 12 h. Wells were harvested using a multiwell cell harvester (LKB, Gaithersburg, MD). Results are expressed as mean counts per minute (cpm) for quadruplicate cultures. 3 .0 Results 3.1 Generation and characterization of Polyoma transformed op/op and op/+ cell lines. To generate tumourigenic cell lines that could be used to determine the role played by tumour cell-derived M-CSF in regulating the accumulation and/or functional activity of tumour-associated macrophages, subconfluent monolayers of op/op and op/+ fibroblasts were infected with an ecotropic retroviral vector, designated CTV-4Py, that encodes the Polyoma large T antigen. Ten days later, numerous foci (>100) of transformed cells were evident in dishes infected with the CTV-4Py vector but not in equivalent dishes infected with the control CTV-4 vector lacking Polyoma-derived sequences (data not shown). The cells from infected and control cultures were then harvested and inoculated separately (subcutaneously) into BALB/c-nu/nu recipients (3 mice per group). Nu/nu recipients were used to avoid rejection of tumour cells expressing highly antigenic Polyoma-derived proteins. By day 40, tumours had developed in all of the mice that received CTV-4Py infected op/op or op/+ fibroblasts but not in any of the mice that received •control CTV-4 infected cells (data not shown). The tumours that developed were then excised, enzymatically dissaggregated and Polyoma transformed op/op (op/opPy) and op/+ (op/+Py) cell lines re-established in vitro. Northern blot analysis (Figure 9) and indirect immunoperoxidase staining (Figure 10), confirmed that both of these cell lines, but not the parental (non-transformed) op/op and op/+ cell lines, express Polyoma large T. 113 * a ~a "a o o 9.49 kb . 7.46 kb - • • 4.40 kb - *m m 2.37 kb - p W Polyoma 1.35 kb - ^ # early region 0.24 kb -V t* Actin Figure 9. Northern blot analysis of Polyoma large T mRNA expression in op/opPy and op/+Py cells. Ten ng of total cellular RNA isolated from op/opPy and op/+Py cells were run in each lane and the filters hybridized with a 32p-Iabeled cDNA probe encompassing the entire Polyoma early region. 114 Figure 10. Expression of the Polyoma large T antigen in op/opPy and op/+Py cells. in Cytospin preparations of (a) op/op, (b) op/opPy, (c) op/+ and (d) op/+Py cells were fixed i acetone and stained for Polyoma large T antigen expression using an indirect immunoperoxidase technique. 115 To confirm that the op/opPy cell line remains deficient in its production of M-CSF, op/opPy and op/+Py cells were next tested for their ability to induce the proliferation of CFU-M in vitro. As shown in Figure 11, the transformed cell lines resemble their non-transformed counterparts in that medium conditioned by the op/+Py cell line was able to induce the generation of macrophage colonies from peripheral blood leukocytes while medium conditioned by the op/opPy cell line could not. Northern blot analysis indicated that the op/+Py andop/opPy cell lines also still expressed approximately equal levels of M-CSF mRNA (Figure 12). To confirm that the major reason for the inability of the op/opPy cells to induce the proliferation of peripheral blood CFU-M was the absence of functional M-CSF, and to rule out other possibilities, such as the production by these cells of soluble mediators that block the action of M-CSF or other hemopoietic growth factors, op/opPy cells were transfected with a pCDM8-based episomal expression vector encoding murine M-CSF and supernatants conditioned by these cells or by control pCDM8 transfected cells were tested for their ability to support the proliferation and/or differentiation of CFU-M in vitro. As shown in Figure 13, medium conditioned by op/opPy cells transfected with pCDM8.M-CSF was fully capable of generating large numbers of macrophage colonies. Control supernatants conditioned by op/opPy cells transfected with the pCDM8 vector alone had no supportive capacity. 3 2 Role of tumour cell-derived M-CSF in regulating the accumulation and functional activity of tumour-associated macrophages In order to define the role of tumour cell-derived M-CSF in regulating the accumulation and functional activity of tumour-associated macrophages, 1 x 106 op/opPy or op/+Py cells were inoculated subcutaneously into SCID mice and both the size and weight of the tumours generated and their macrophage content determined on day 28. SCID mice were used at this point because they were more readily available than nu/nu mice and would also not reject the tumour cells. As shown in Figure 14, op/+Py tumours were significantly larger than equivalent op/opPy tumours. The macrophage content of op/opPy or op/+Py tumours was determined by indirect immunoperoxidase staining of cytospin preparations of enzymatically disaggregated tumour cell 116 4) J= — Q. C3 O 0 . E 2 O 4) i_ *c « o .a — s 2 3 z 60-1 5 0 -40-30-20-10-T mi .+ O ! T 1 1 1 1 1 ii a. .+ "a. o a. I Source of conditioned medium Figure 11. M-CSF production by op/opPy and op/+Py cells. 1 x 104 PBL were incubated with 10% conditioned medium from op/op, op/+, op/opPy and op/+Py cells. The number of macrophage colonies containing >10 cells was determined on day 10-12 following fixation and staining in a solution of 1% methylene blue in methanol in duplicate cultures. The results shown here represent the mean ± SEM from 3 separate experiments. 117 in vivo unselected selected 9.49 kb -7.46 kb " 4.40 kb -2.37 kb -1.35 kb -0.24 kb -I 1 I 1 + .o + o + g" O O O O O O J | | M H JMl^M Mk^mmu Y* T^ T^ Figure 12. Northern blot analysis of M-CSF mRNA expression in op/opPy and op/+Py cells. Ten ng of total cellular RNA isolated from op/op and op/+ cells, unselected CTV-4Py transduced op/op and op/+ cells, and tumourigenic in vivo selected op/opPy and op/+Py cells were run in each lane and filters hybridized with a32P-labeled cDNA probe encompassing the entire coding region of the murine M-CSF. 118 200-2 -c a. o o % conditioned medium op/op-PY CDM8.mM-CSF O— op/op-PY CDM8 Figure 13. M-CSF production by pCDM8- and pCDM8.mM-CSF-transfected op/opPy cells. 1 x 104 PBL were incubated with various concentrations of medium conditioned by pCDM8- and pCDM8.mM-CSF-transfected op/opPy cells. The number of macrophage colonies containing >10 cells was determined on day 10-12 following fixation and staining in a solution of 1% methylene blue in methanol. Each point represents the mean ± SD of duplicate samples from a single representative experiment. 119 WO ^^ +* J= ec o> £ u 3 O £ 3 H 1.8-1.6-1.4-1.2-1-0.8-0.6-0.4-0.2-o-Figure 14. Growth of op/opPy and op/+Py tumours. 1 x 106 op/opPy and op/+Py cells were inoculated subcutaneously into the flank of BALB/c SCID mice. The weight of the tumours generated was determined on day 28. Results are the mean weights, ± SD, of tumours grown in 3 mice per group. op/+Py tumours are significantly larger than op/opPy tumours (p<0.05, Student's t test). I w i t i • i i i 11 111 T opPy op/+Py 120 suspensions using a panel of anti-macrophage monoclonal antibodies. As shown in Figure 15, although the proportion of Mac-1-positive and F4/80-positive cells is somewhat higher in op/+Py tumours, both tumour types contain significant numbers of macrophages and the proportion of 114/A10-positve cells was remarkably similar. Interestingly however, the macrophages present within op/op/Py and op/+Py tumours exhibited obvious morphological differences indicating that although M-CSF may not play a major role in regulating the accumulation of tumour-associated macrophages it does, nevertheless, have an effect on the phenotype and/or function of these cells. Tumour-associated macrophages derived from op/opPy tumours were smaller, rounder and more monocyte-looking than the macrophages derived from op/+Py tumours which were more spread out, irregular and mature looking. Indeed, in support of such a role, Northern blot analysis demonstrated that although macrophages present within op/opPy and op/+Py tumours expressed similar levels of 114/A10 mRNA, the macrophages present within the op/opPy tumour expressed far lower levels of IL-lp and TNF-a and FcRyll mRNA (Figure 16). 3.3 Production of a monocyte survivial factor by op/op fibroblasts In the absence of added hemopoietic growth factors, murine peripheral blood monocytes die rapidly in vitro. Indeed, after 48 h of culture <10% of the starting population remain adherent and viable (Figure 17). Addition of M-CSF or GM-CSF prevented such cell death (Figure 17). In addition, M-CSF but not GM-CSF induced dramatic morphological changes in the adherent cells (data not shown). Interestingly, medium conditioned by op/op and op/opPy cell lines also enhanced monocyte survival (Figure 18). As these conditioned media do not contain detectable levels of M-CSF or GM-CSF (as determined by their ability to support the proliferation of either peripheral blood-derived CFU-M (Figure 11) or the GM-CSF-dependent cell line B6SUtA) (data not shown), it appears that the op/op cell line produces another, as yet undefined, soluble factor that can promote monocyte survival in vitro. 121 50 40 ^ 30 w 4* J* 1 20 a. 10 o Figure 15. Macrophage content of op/opPy and op/+Py tumours. Acetone fixed cytospin preparations of enzymatically disaggregated tumour cell suspensions were stained with mAbs directed against the a-subunit of Mac-1 (TIB 128), the F4/80 antigen (HB198) and 114/A10 using an indirect immunoperoxidase technique. Each experiment was run in duplicate and each point represents the mean ± SEM of at least 3 experiments. In each experiment at least 200 cells were counted. The percentage of Mac-1-positive and F4/80-positive cells present in op/+Py tumours was significantly higher (p<0.05) than in op/opPy tumours. There was no significant difference in the percentage of 114/A10-positive present in op/opPy and op/+Py tumours (p=0.14). P values were obtained using Student's t test. XXXXXj /////< •xxxxx,. 7////J /////-'XXXXX* /////< /////v xxxxx* xxxxx-xxxxx> xxxxx, XXXXX.I xxxxxx^ "XXXXX^ / / / / / i '/////^  /////, 'XXXXX* xxxxx, ////A XXXXX* 7////. " " - -XX xx //, . 'XX •••xxx ••xxxx '•••••X •••••• '•••••X '•••XXX ••••XX '•••XXX '••••XX '•••XXX ••••XX •••XXX •••••• •••••• '•••XXX •••••• •••XXX •••••X • *J - * " • * ' • op/+Py 0 op/opPy 2 ! ^ • » — s "• o = S v S 1 PH a fc &. + ,0 + ,0 "e, "a, "o. "a o o o o 114/A10 € • * » FcRyll s 2 IL-ip TNF-a It Actin 122 Figure 16. Northern blot analysis of 114/A10, FcRyll, IL-lp and TNF-a mRNA expression in op/opPy and op/+Py tumour-associated macrophages. Ten ng of total cellular RNA isolated from op/opPy and op/+Py cell lines and tumours were run in each lane. Filters were hybridized with 32P-labeled cDNA probes encompassing the entire coding regions of murine 114/A10, FcRyll, IL-lp, TNF-a and actin, 123 c 3 Urn U w it W C <U u VI <u u o 3 6000-1 5000-4 0 0 0 -3000 -2000 -1000-£ 70 Q O V3 o (J % T • Conditioned media i I o u Figure 17. Promotion of monocyte survival in vitro by M-CSF and GM-CSF 1 x 104 peripheral blood monocytes purified by Percoll gradient centrifugation were added to each well of a 96 well plate together with 10% medium conditioned for 72 h by 1 x KPlml pCDM8-, pCDM8.mM-CSF- or pRC3.mGM-CSF-transfected COS7 cells, and the number of viable monocytes was determined 48 h later following calcein labelling using a fluorescent plate reader for duplicate wells. Each point represents the mean ± SEM of triplicate experiments. 2000 - i 124 ~jjj 1500 H s >> 95 u w W e u I_ O 3 1000-5 0 0 -I' Z T o CO o T CO U 8 1— a o "a. o m> • : • * • : a . o >> a . a . o »». a o T ->» + a. o Conditioned medium Figure 18. Promotion of monocyte survival in vitro by op/op and op/+ conditioned media 1 x 104 peripheral blood monocytes purified by Percoll gradient centrifugation were added to each well of a 96 well plate together with 10% medium conditioned for 72 h by 1 x ICfi/ml pCDM8-, pCDM8.mM-CSF- or pRC3.mGM-CSF-transfected COS7 cells and op/op, op/+, op/opPy or op/+Py cells. The number of viable monocytes was determined 48 h later following calcein labelling using a fluorescent plate reader. Each point represents the mean ± SEM of triplicate experiments. 125 4 .0 Discussion Several tumour cell-derived factors chemotactic for monocytes have recently been identified (33-37) and shown to play an important role in regulating the accumulation of tumour-associated macrophages (38-44). These include MCP-1, MCP-2, MCP-3, and VPF. However, there is also evidence that these are not the only molecules involved in this complex process (41). The major aim of the present study was to determine what role, if any, tumour cell-derived M-CSF may play in determining the total macrophage content of a tumour. M-CSF is an attractive candidate in this regard. It is constitutively produced at a high level by many histologically distinct human and animal tumours (2) and has been shown to function both as a monocyte-specific chemoattractant (5), and to promote the survival, proliferation and differentiation of monocytes and other macrophage precursor cells in vitro (6) and in vivo (7, 8). The results presented in this Chapter demonstrate that tumours initiated from M-CSF-deficient op/opPy fibroblastoid cells contain almost as many macrophages as similar tumours initiated from M-CSF-producing op/+Py cells. These findings argue against a prerequisite role of M-CSF for the accumulation of  macrophages within tumours. This conclusion, however, contrasts with recent studies in which the introduction of the human M-CSF gene into the murine plasmacytoma cell line J558L was shown to dramatically increase the number of macrophages present within tumours generated from these cells following inoculation into syngeneic BALB/c mice (45). Growth of these M-CSF transduced tumours was ghtly inhibited and attempts to activate the tumour-associated macrophages by the systemic administration of LPS and/or IFN-y were unsuccessful. Although the reason for the difference between these results and the data presented in this Chapter remain to be determined, they may simply reflect the fact that the very high and non-physiological level of M-CSF produced by transfected J588L cells is sufficient to induce dramatic increases in both the number and functional activity of circulating monocyte/macrophages in tumour-bearing animals (45). Moreover, it should be noted, tlpit since the animals used in these gene transfer studies were immunocompetent, the 126 possibility that immune responses directed against human M-CSF may have influenced the nature of the host cell infiltrate present within the genetically engineered tumours cannot be ruled out. The data presented in this Chapter also demonstrate, that like their human counterparts, murine monocytes rapidly die in the absence of added hemopoietic growth factors. Although the molecular nature of the molecule(s) that can maintain monocyte viability within M-CSF-deficient op/opPy tumours remain to be determined, there are several obvious possibilities. Firstly, host-derived stromal elements or immune cells could perhaps provide sufficient amounts of M-CSF to maintain monocyte/macrophage viability. Arguing against this possibility, however, is the observation, discussed further below, that macrophages present within tumours initiated from op/opPy cells express far lower levels of several M-CSF inducible mRNA transcripts, including IL-lp, TNF-cc and FcRyll, than equivalent cells present within op/+Py tumours. Thus if M-CSF is produced within op/opPy tumours it must be assumed that the levels generated, while sufficient to maintain monocyte/macrophage viability at a level similar to that seen in op/+Py tumours, are nevertheless insufficient to induce equivalent levels of IL-lp, TNF-a and FcRyll gene expression within these cells. GM-CSF can also promote monocyte survival in vitro. Although most fibroblastoid cell lines do no$ constitutively produce GM-CSF, many can be induced to do so following stimulation with the macrophage-derived cytokines IL-lp, TNF-a or IFN-y. Arguing against a role for this molecule in the accumulation of tumour-associated macrophages, however, was the finding that GM-CSF was not constitutively produced at detectable levels by either op/opPy or op/+Py cells, nor could these cells be readily induced to secrete this cytokine following stimulation with IL-lp or TNF-a (see Chapter 4). Moreover, Northern blot analysis failed to demonstrate the presence of significant qualities of GM-CSF mRNA within CD45-ve/CD 18-ve tumour cells purified from either op/+Py or op/opPy tumours (data not shown). A final possibility is that op/opPy cells may elaborate a soluble mediator distinct from M-CSF or GM-CSF that can promote monocyte survival in vitro. Evidence supporting the existence of such a mdiecule can be inferred from the observation that mice deficient in both M-CSF and GM-CSF still contain significant numbers of circulating monocytes and near normal numbers of 127 macrophages in at least certain tissues (46). In the present study, we provide provisional data that op/opPy cells may also produce such an unknown factor. Although tumour cell-derived M-CSF does not appear to play a critical role in regulating the accumulation of tumour-associated macrophages, it can have an effect on the phenotype of these cells. This statement is based on the finding that macrophages present with op/+Py tumours were morphologically distinct, and expressed higher levels of IL-lp, TNF-cc and FcRyll mRNA than those present within op/opPy tumours. All three of these latter genes have been shown previously to be induced by M-CSF in vitro. In addition, op/+Py tumours also appeared to grow somewhat faster than op/opPy tumours initiated with the same number of cells. How differences in tumour growth relate to the altered functional capabilities of the macrophages present within these tumorus remains to be determined. It is interesting to note, however, that TNF-a which is expressed at a higher level within the faster growing op/+Py tumours, has been suggested to play an important role in the induction of tumour angiogenesis in vivo (47-52). 128 5 . References 1. Mantovani, A. Tumor-associated macrophages. Curr Opin Immunol. 1990; 2: 689-692. 2. Mantovani, A.; Bottazzi, B.; Colotta, F.; Sozzani, S.; Ruco, L. The origin and function of tumor-associated macrophages. Immunol Today. 1992; 13: 265-270. 3. Stewart, C. C. Local proliferation of mononuclear phagocytes in tumors. J Reticuloendothel Soc. 1983;34:23-27. 4. Evans, R.; Cullen, R. T. In situ prolferation of intratumor macrophages. J Leukocyte Biol. 1984; 35: 561-572. 5. Wang, J. M.; Griffin, J. D.; Rambaldi, A.; Chen, Z. G.; Mantovani, A. Induction of monocyte migration by recombinant macrophage colony-stimulating factor. J Immunol. 1988; 141: 575-579. 6. Tushinski, R. J.; Oliver, I. T.; Guilbert, L. J.; Tynan, R W.; Warner, J. R.; Stanley, E. R. Survival of mononuclear phagocytes depends upon a lineage-specific growth factor that the differentiated cells selectively destroy. Cell. 1982; 48: 997-1004.  7. Chikkappa, G.; Broxmeyer, H. E.; Cooper, S.; Williams, D. E.; Hangoc, G.; Greenberg,  M. L.; Waheed, A.; Shadduck, R. K. Effect in vivo of multiple injections of purified and recombinant human macrophage colony-stimulating factor to mice. Cancer Res. 1989; 49: 3558-3561.   8. Munn, D. H.; Garnick, M. B.; Cheung, N. V. Effects of parenteral recombinant human macrophage colony-stimulating factor on monocyte number, phenotype, and antitumor cytotoxicity in nonhuman primates. Blood. 1990; 75: 2042-2047.  9. Wing, E. J.; Waheed, A.; Shadduck, R. K.; Nagle, L. S.; Stephenson, K. Effect of colony stimulating factor on murine macrophages. Induction of antitumor activity. J Clin Invest. l982; 69: 270-276. 10. Wing, E. J.; Ampel, N. M.; Waheed, A.; Shadduck, R. K. Macrophage colony-stimulating factor (M-CSF) enhances the capacity of murine macrophages to secrete oxygen reduction -products. J Immunol. 1985; 135: 2052-2056. 11. Nakoinz, I.; Ralph, P. Stimulation of macrophage antibody-dependent killing of tumor targets by recombinant lymphokine factors and M-CSF. Cell Immunol. 1988; 116: 331-366. 12. Munnr3p. H.; Cheung, N. K. Antibody-dependent antitumor cytotoxicity by human monocytes cultured with recombinant macrophage colony-stimulating factor. Induction of efficient antibody mediated antitumor cytotoxicity not detected by isotope release assays. J Exfftfed. 1989; 170: 511-526. 13. Sanda, M. G.; Bolton, E ; Mule, J. J.; Rosenberg, S. A. In vivo administration of recombinant macrophage colony-stimulating factor induces macrophage-mediated antibodv-dependent cytotoxicity of tumor cells. J Immunotherapy. 1992; 12: 132-137. 14. Warren, M. K.; Ralph, P. Macrophage growth factor CSF-1 stimulates human monocyte production of interferon, tumor necrosis factor and colony stimulating activity. J Immunol. 1986; 137: 2281-2285. 129 15. Bajorin, D. R; Jakubowski, A.; Cody, B. Recombinant macrophage colony stimulating factor (rhM-CSF): a phase I clinical trial in patients with metastatic melanoma. Proc Am Soc Clin Oncol. 1990; 9: 183-185. 16. Bajorin, D. F.; Cheung, N-K V.; Houghton, A. N. Macrophage colony-stimulating factor: biological effects and potential applications for cancer therapy. Blood. 1991; 82: 42-48. 17. Gluzman, Y. SV40-transformed simian cells support the replication of early SV40 mutants. Cell. 1981; 23: 175-182. 18. Hughs, P. F. D.; Thacker, D.; Hogge, D.; Sutherland, H. J.; Thomas T.E.; Lansdorp, P. M.; Eaves, C. J.; Humphries, R. K. Retroviral gene transfer into primitive normal and leukemic progenitor cells using clinically applicable procedures. J Clin Invest. 1992; 89: 1817. 19. Moore, K.; McBride, W. H. The activation state of macrophage subpopulations from a murine fibrosarcoma. Int J Cancer. 1980; 26: 609-615. 20. Ho, M. K.; Springer, T. A. Mac-1 antigen: quantitative expression in macrophage populations and tissues, and immunofluorescent localization in the spleen. J Immunol. 1982; 128: 2281-2286. 1. Austyn, J. M.; Gordon, S. F4/80, a monoclonal antibody directed specifically against the mouse macrophage. Eur J Immunol. 1981; 11: 805-815. ougherty, G. J.; Dougherty, S. T.; Kay, R. J.; Lansdorp, P.; Humphries, R. K. Identification and characterization of 114/A10, an antigen highly expressed on the surface of murine myeloid and erythroid progenitor cells and IL-3-dependent cell lines. Exp Hematol. 1989; 17: 877-882. 23. Dougherty, G. J.; Kay, R. J.; Humphries, R. K. Molecular cloning of 114/A10, a cell ^ 4|urface antigen containing highly conserved repeated elements, which is expressed by marine hemopoietic progenitor cells and interleukin-3-dependent cell lines. J Biol Chem. 1W9; 264: 6509-6514. 24. Kay, R.; Dougherty, G. J.; Humphries, R. K. Heterogeneous modifications of the 114/A10 protein of interleukin-3-dependent cells are concentrated in a highly repetitive amino-ferminal domain. J Biol Chem. 1990; 265: 4962-4968. ' ' •% 25. Ball, R. K.; Birgit, S.; Quellhorst, S.; Brandner, G.; Braun, D. G. Monoclonal antibodies against simian virus 40 nuclear large T tumour antigen: epitope mapping, papova virus cross-reaction and cell surface staining. EMBO J. 1984; 3: 1485-1491. 26. HarlowfE.; Lane D. Antibodies. A Laboratory Manual.: Cold Spring Harbour laboratory; 27. Ladner,jM. B.; Martin, G. A.; Noble, J. A.; Wittman, V. P.; Warren, M. K.; McGrogan, M.; Stanley, E. R. cDNA cloning and expression of murine macrophage colony-stimtilating factor from L929 cells. Proc Natl Acad Sci USA. 1988; 85: 6706-6710. :# 28. Seed, B. Aj| LFA-3 cDNA encodes a phospholipid-linked membrane protein homologous to its receptor CD2. Nature. 1987; 329: 840-842. 130 29. Chirgwin, J. M.; Przybyla, A. E.; MacDonald, R. J.; Rutter, W. J. Isolation of biologically active ribonucleic acid from sources enriched in riboneclease. Biochemistry. 1979; 18: 5294-5297. 30. Kay, R. J.; Boissy, R. J.; Russnak, R. H.; Candido, E. P. M. Efficient transcription of a Caenorhabditis elegans heat shock gene pair in mouse fibroblasts is dependent on multiple promoter elements which can function bidirectionally. Mol Cell Biol. 1986; 6: 3134-3143. 31. MacCoubrey, I. C ; Moore, P. L.; Haugland, R. P. Quantitative fluorescence measurements of cell viability (cytotoxicity) with a multi-well plate scanner. J Cell Biol. 1990; 111: Abstract 303. 32. Moore, P. L.; MacCoubrey, I. C ; Haugland, R. P. A rapid pH insensitive, two color fluorescence viability (cytotoxicity) assay. J Cell Biol. 1990; 111: Abstract 304. 33. Matsushima, K.; Larsen, C. G.; DuBious G.C.; Oppenheim, J. J. Purification and characterization of a novel monocyte chemotactic and activating factor produced by a human myelomonocytic cell line. J Exp med. 1989; 169: 1485-1490. 34. Van Damme, J.; Decock, B.; Lenaerts, J. P.; Conings, R.; Bertini, R.; Mantovani, A.; Billiau, A. Identification by sequence analysis of cheomtactic factors for monocytes produced by normal and transformed cells stimulated with virus, double standed RNA or cytokine. Eur J Immunol. 1989; 19: 2367-2373. 35. Yoshimura, T.; Robinson, E. A.; Tanaka, S.; Appella, E.; Kuratsu, J, I.; Leonard, E. J. Purification and amino acid analysis of two human glioma-derived monocyte chemoattractants. J Exp Med. 1989; 169: 1449-1459. 36. Bottazzi, B.; Colotta, E ; Sica, A.; Nobili, N.; Mantovani, A. Chemoattractant expressed in human sarcoma cells (tumor-derived chemotactic factor, TDCF) is identical to monocyte chemoattractant protein-1 /monocyte chemotactic and activating factor (MCP- 1/MCAF). '% J Cancer. 1990; 45: 795-797. 37. Rollijp; B. J.; Morrison, E. D.; Stiles, C. D. Cloning and expression of JE, a gene inducible by platelet derived growth factor and whose product has cytokine-like properties. Proc * Jatl Acad Sci USA. 1988; 85: 3738-3742. 38. Bottazzi, B.; Polentarutti, N.; Acero, R.; Balsari, A.; Boraschi, D.; Ghezzi, P.; Salmona, M.pMantovani, A. Regulation of the macrophage content of neoplasms by chemoattractants. Science. 1983; 220: 210-212. i 39. Bottazzij, B.; Polentarutti, N.; Balsari, A.; Boraschi, D.; Ghezzi, P.; Salmona, M.; Mantovani, A. Chemotactic activity for mononuclear phagocytes of culture supernatants fromimurine and human tumor cells: evidence for a role in the regulation of the macrophage content of neoplastic tissues. Int J Cancer. 1983; 31: 55-63. 40. Benomaff A.; Ming, W. J.; Taraboletti, G.; Ghezzi, P.; Balotta, C ; Cianciolo, G. J.; Snyderman, R.; Dore, J. F.; Mantovani, A. Chemotactic factor and P15E-related chemotaxis inhibitor in human melanoma cell lines with different macrophage content and tumorigenicity in nude mice . J Immunol. 1987; 138: 2372-2379. 41. Mantovani, A.; Ming, W. J.; Balotta, C ; Abdeljalil, B.; Bottazzi, B. Origin and regulation of tumor-associated macrophages: the role of tumor-derived chemotactic factor. Biochem Biophys Acta. 1986; 865: 59-67. 131 42. Walter, S.; Bottazzi, B.; Govoni, D.; Colotta, F.; Mantovani, A. Macrophage infiltration and growth of sarcoma clones expressing different amounts of monocyte chemotactic protein/JE. Int J Cancer. 1991; 49: 431-435. 43. Rollins, B. J.; Sunday, M. E. Suppression of tumor formation in vivo by expression of the JE gene in malignant cells. Mol Cell Biol. 1991; 11: 3125-3131. 44. Bottazzi, B.; Walter, S.; Govoni, D.; Colotta, F.; Mantovani, A. Monocyte chemotactic cytokine gene transfer modulates macrophage infiltration, growth, and susceptibility to IL-2 therapy of a murine melanoma J Immunol. 1992; 148: 1280-1285. 45. Dorsch, M.; Hock, H.; Kunzendorf, U.; Diamantstein, T.; Blankenstein, T. Macrophage colony-stimulating factor gene transfer into tumor cells induces macrophage infiltration but not tumor suppression. Eur J Immunol. 1993; 23: 186-190. 46. Lieschke, G. L.; Stanley, E.; Grail, D.; Hodgson, G.; Sinickas, V.; Gall, J. A. M.; Sinclair, R. A.; Dunn, A. R. Mice lacking both macrophage- and granulocyte-macrophage colony-stimulating factor have macrophages and coexistant osteopetrosis and severe lung disease. Blood. 1994; 84: 27-35. 47. Montrucchio, G.; Lupia, E.; Battaglia, E.; Passerini, G.; Bussolino, F.; Emanuelli, G.; Camussi, G. Tumor-necrosis factor a-induced angiogenesis depends upon in situ platelet-activating factor biosynthesis. J Exp Med. 1994; 180: 377-382. 48. Klagsbrun, M.; D'Amore, P. A. Regulators of angiogenesis. Annu Rev Physiol. 1991; 53: 217-237. 49. Folkman, J.; Klagsbrun, M. Angiogenic factors. Science. 1987; 235: 442-447. 50. Tracey, K.; Vlassara, H.; Cerami, A. Cachectin/tumor necrosis factor. Lancet. 1989; 20: 1122-1126. 51. Fra|er-Schroder, M.; Risau, W.; Hallmann, R.; Gautschi, P.; Bohlen, P. Tumor-necrosis factor type a, a potent inhibitor of endothelial cell growth in vitro, is angiogenic in vivo . Proc Natl Acad Sci USA. 1987; 84: 5277-5281. 52. Leibovich, S. J.; Polverini, P. J.; Shepard, H. M.; Wiseman, D. M.; Shively, V.; Nuseir, ,N. Macrophage-induced angiogenesis is mediated by tumor-necrosis factor-a . Nature. ' 1987; 239: 630-632. 132 CHAPTER 4 Molecular mechanisms regulating TNF-a production by tumour-associated macrophages 1.0 Introduction A number of recent studies have demonstrated the production of TNF-a by macrophages present within both human and animal solid tumour masses (1, 2). At present, however, neither the functional importance of intratumoural TNF-a, nor the molecular mechanisms that regulate its production, are well characterized. TNF-a is a pleotropic cytokine that has been shown to perform a large number of diverse and on occasion mutually antagonistic functions, some of which might be expected to have profound effects on tumour growth if they were to occur within the tumour microenvironment (3, 4). For example, TNF-a is directly cytotoxic or cytostatic for certain human and animal tumour cell lines in vitro (5) but may also enhance the proliferation of others(6, 7). In addition, it can induce the expression on endothelial cells of various adhesion proteins involved in the recruitment of leukocytes to sites of inflammation (8,9) and at low concentrations may stimulate angiogenesis through the production of platelet-activating factor (10-12). At higher concentrations, however, TNF-a may instead increase vascular permeability (13, 14) and endothelial procoagulant activity (15) thereby causing hemorrhagic necrosis (16). Finally, although expression of TNF-a inhibits the growth of certain tumour cells in vivo (17, 18), it has also been shown to increase the metastatic potential of others (6). The mutant murine strain C3H/HeJ, first described by Heppner and Weiss in 1965 (19), has proven an invaluable tool with which to investigate the molecular mechanisms that regulate the production and functional activity of TNF-a (20). This particular strain carries a mutation in the LPS response gene (Lps), located on the short arm of chromosome 4 between Mup-1 and Ps (21,22), and as a consequence produces greatly reduced quantities of TNF-a (and IL-1) following 133 stimulation with endotoxin (20). While the precise nature of this mutation remains to be determined, available evidence points toward a defect in the signal transduction pathway triggered by the binding of LPS to its cell surface receptor(s) (20). In order to begin to define the molecular mechanisms that regulate the production of TNF-a within the rumour microenvironment and to determine the possible functional importance of this molecule, two fibroblastoid tumour cell lines, Fsa-R and Fsa-N, were inoculated subcutaneously into syngeneic Lpsd C3H/HeJ and Lpsn C3H/HeN mice. Tumour growth, macrophage content, and the production of TNF-a by tumour-associated macrophages was then determined. The results obtained indicate that although both Fsa-N and Fas-R tumours grew equally well in both murine strains and contained almost exactly the same proportion of macrophages, the macrophages present within tumours grown in C3H/HeJ mice expressed 5-10 fold less TNF-a than equivalent cells present within tumours grown in C3H/HeN mice. These data suggest that the mechanisms that operate within the tumour microenvironment to induce the production of TNF-a act, at least in part, via the same signal transduction pathway that is defective in Lpsd C3H/HeJ mice. Moreover, it appears that the differences in the level of TNF-a produced within tumours grown in C3H/HeN and C3H/HeJ mice, while seemly dramatic, are insufficient to alter either tumour growth rate or macrophage content 2 .0 Materials and Methods 2.1 Animals C3H/HeN (Lpsn) and C3H/HeJ (Lpsd) mice were obtained from Charles River (Quebec) and Jackson Laboratories (Bar.Harbor, ME), respectively. Both strains were bred and maintained in the Joint Aninfal Facility of the British Columbia Cancer Research Center and used when they were between 6 and 12 weeks of age. 134 2.2 Tumour cell lines Fsa-N is a fibrosarcoma that arose spontaneously in the shielded limb of an irradiated C3Hf/Sed/Kam mouse (23). Fsa-R is a chemically-induced fibrosarcoma that was generated in a C3Hf/Sed/Kam mouse by treatment with methylcholanthrene (22). Both tumours grow progressively in syngeneic C3H mice and are well characterized with respect to the systemic and local immune responses that they elicit (24-28). Cell lines were established from enzymatically disaggregated tumour cell suspensions as previously described (24) and maintained in Dulbecco's minimum essential medium (DMEM) (Stem Cell Technologies Inc.) containing 10% Fetal Calf Serum (FCS, Hyclone, Logan, UT).  2.3 Monoclonal Antibodies (mAbs) mAb YE1.21, directed against murine CD45, was obtained from Dr. F. Takei (Terry Fox Laboratory, BC Cancer Research Centre). mAbs TIB213, directed against the cc-subunit of murine LFA-1 (CDlla), TIB128, directed aganist the a-subunit of Mac-1 (CDllb) (29), and HB198, directed against the murine macrophage antigen F4/80 (30), were all obtained from ATTC. mAb 114/A10, which recognizes an antigen expressed on murine myeloid cells (31-33)) was obtained from Dr. Graeme Dougherty (Terry Fox Laboratory). 2.4 Preparation of tumour cell suspensions Tumours were initiated by the subcutaneous inoculation of 5 x 105 cultured tumour cells into the left flank. Tumour-bearing animals were killed by cervical dislocation on day 28 and their tumours removed under sterile conditions, weighed, finely minced, and disaggregated by vigorous agitation for lh at room temperature in Hanks' balanced salt solution (HBSS) containing in 0.5% (w/v) Dispase (Grade II, Boehringer Mannhein, West Germany) and 0.01% (w/v) collagenase (Sigma Chemical Co., St. Louis, MO) (24). The resulting cell suspension was washed three times in HBSS before use. 135 2.5 Isolation of tumour-associated macrophages Enzymatically disaggregated tumour cell suspensions were resuspended at 1 x 106 cells/ml in HBSS+20% FCS and 0.5% (w/v) Dispase. Twenty-five ml aliquots of each cell suspension were then added to 15 cm diameter Integrid tissue culture dishes (Becton Dickinson, Sunnyvale, CA) and incubated at 37°C for 30 min. Non-adherent cells were removed by vigorous pipetting and extensive washing in HBSS. Immunohistochemical staining (as described below) with mAbs YE1.21, TIB213, TIB 128, F4/80 and 114/A10 indicated that greater than 95% of cells still adherent to the plastic surface under these conditions were macrophages (data not shown). 2.6 Indirect immunoperoxidase staining Cytospin preparations were air dried, fixed in acetone for 5 min, and incubated for 30 min at room temperature with 50 |xl of the appropriate mAb tissue culture supernatant. After extensive vykshing in HBSS, the cytospins were incubated for a further 30 min with 50 ,^1 of a 1:2000 dilution of horseradish peroxidase-conjugated goat anti-rat IgG (Sigma). After washing in HBSS, the reaction was developed by incubating the slides for 5 min at room temperature in phosphate buffered saline (PBS) containing 0.06% (w/v) 3-3'-diaminobenzidine (Sigma) and 0.012% (v/v) H2Q2 (Sigma). Positively staining cells were easily identified by the presence of the characteristic brown colpur associated with the cell. 2.7 Northern blot analysis Total^ellular RNA was using the guanidine isothiocyanate/CsCl method as described (34). Ten jig of Rf<JA were electrophoresed through a 1% (w/v) agarose gel containing 5% (v/v) de-ionized formaldehyde, transferred to a nylon membrane (Zeta-Probe GT, Bio-Rad Laboratories Ltd., Mississauga, Ontario) and cross-linked by exposure to ultraviolet radiation (Stratalinker; Stratagene, La Jolla, CA). Filters were pre-hybridized for 1 h at 42°C in 500 mM sodium phosphate buffer, pH 7.2, 50% (v/v) formamide, 5% sodium dodecyl sulfate, 1 mM EDTA, and 1 J mg/ml bovine serum albumin (Fraction V; Sigma) and then hybridized for 16 h at 42°C in the same solution containing denatured 32P-labelled probes prepared using the Pharmacia Ready-To-Go 136 Oligonucleotide Labelling Kit as per the manufacturer's instructions. Filters were washed and developed as described (35). 2.8 Western blot analysis Enymatically disaggregated tumour cell suspensions were resuspended at 1 x 107 cell/ml in PBS containing 1% (v/v) NP40,5 mM EDTA and 10 mM phenylmethylsulfonyl fluoride. Lysates were incubated on ice for 5 min, microcentrifuged for 5 min at 4°C to pellet nuclei and other cell debris, and stored at -70°C until required. Total cellular proteins were separated under reducing conditions on a 15% sodium dodecyl sulphate polyacrylamide gel (SDS-PAGE) and transferred to nitrocellulose membranes (Bio Rad) as previously described (28). Non-specific binding sites were blocked by incubating the membranes overnight in PBS containing 5% (w/v) milk protein. After washing in HBSS, the blots were incubated for 4 h at room temperature with a 1:100 dilution of jgplyclonal rabbit anti-murine-TNF-a (Genzyme Corp, Boston, MA), washed three more times in HBSS, and incubated for a further 1 h at room temperature with a 1:2000 dilution of horseradish peroxidase-conjugated goat anti-rabbit IgG (Sigma) in DMEM+5% FCS. After extensive washing in HBSS the blot was developed in PBS containing 0.06% (w/v) 3'-3-diaminobenzidine and 0.012% (v/v) H 2 0 2 . "'•} 2.9 Purification of tumour cells from enzymatically disaggregated tumour cell suspensions 1 xHp7 enzymatically disaggregated tumour cells were resuspended in 3 ml of a mAb tissue culture supernatant pool containing equal amounts of TIB213, TIB 128 and YE1.21, and incubated on ice for 3p min. The cells were washed three times with HBSS+2% FCS and incubated for a 'I further 30 rr|n on ice in the dark with 1ml of fluoroscein isothiocyanate-conjugated goat anti-rat IgG F(ab')2 fragment (FITC-GaRIgG) (Cooper Biomedical, West Chester, PA) diluted 1:100 in HBSS+2% FCS. The cells were then washed three times with HBSS+2% FCS, resuspended in HBSS+2% FCS containing 1 ng/ml propidium iodide and sorted into positively staining hemopoietic cell populations and negatively staining tumour cell populations using a FACSort (Becton Dickyfson, Mount View, California). V 137 2.10 Preparation of tumour cell conditioned media (CM) 1 x 105 tumour cells in a final volume of 1 ml DMEM+10% FCS containing various concentrations of recombinant murine TNF-a (Cedarlane Laboratories, Hornby, Ontario) or human IL-lp(R&D Systems, Minneapolis, MN) were added to the wells of a 24-well plate (Falcon 3047: Becton Dickson) and incubated at 37°C for 48 h. Conditioned media were then removed, centrifuged at 250 g for 10 min to remove cellular debris, filtered through a 0.45 \iM filter (Syrfil-MF; Costar Corp., Cambridge, MA) and stored at -20°C until required. Aliquots were thawed once only prior to testing. 2.11 Bioassav for GM-CSF 104 GM-CSF-dependent B6SUtA cells in 100 jd of DMEM+10% FCS were placed in each well of a flat bottomed 96-well tissue culture plate (Falcon 3072: Becton Dickson) together with serial dilutions of recombinant GM-CSF or tumour cell conditioned media and incubated for 48 h at 37°C. 1 nCi (37 Bq) of 3H-thymidine was then added to each well and the plates incubated for a further 12 h. Wells were harvested using a multiwell cell harvester (LKB, Gaithersburg, MD). -Results are expressed as mean counts per minute (cpm) for quadruplicate cultures. 2.12 Isolation of a murine TNF-a cDNA A full length murine TNF-a cDNA was generated by RT-PCR. Briefly, total cellular RNA was isolated from Fsa-N tumours grown in a C3H/HeN mice as described above (34). cDNA was synthesizednising the Pharmacia First Strand Synthesis Kit (Pharmacia, Baie d'Urfe, Quebec) and PCR was capied out exactly as described by the manufacturer using the following primer pair: 5' mTNF-a (5VGGTCTAGACACCATGAGCACAG-3') and 3' mTNF-a (5'-GGTCTAGAACA CCCATTCCCTTCAC-3'). Both primers contain added 5' Xbal restriction sites. The samples were placed in a thermal cycler (Biosycler, BIOS, New Haven, CT) and cycled 30 times at 95°C for 30 s, 55°C for 30 s, and 72°C for 1 min. PCR products were separated on a 1% agarose gel, purified using a Geneclean II Kit (BIO 101 Inc., Vista, CA), digested with Xbal and ligated into nded templates were sequenced by the dideoxy chain 138 termination method (36), using T7 DNA polymerase and the reaction conditions suggested by the manufacturer (Sequenase Version 2.0 DNA Sequencing Kit, USB, Cleveland, OH). 2.13 Generation of retroviral vector JmTNF-a To construct the retroviral vector JmTNF-a, the full length murine TNF-a cDNA was first of all inserted into the Xbal site of the plasmid pTZ19R/Tk-neo. A Smal-Hindlll fragment containing both the TNF-a cDNA and Tk-neo was then isolated and cloned into Hpal-Hindlll cut Jzen.l (37). The plasmid obtained was transfected into the ecotropic packaging line GP+E-86 (38) by calcium phosphate precipitation and transfected cells selected using G418 (0.5 mg/ml active weight) (Gibco Life Technologies Inc., Grand Island, New York, USA). 2.14 Infection of Fsa-N cells with JmTNF-a Fsa-N cells were infected with the JmTNF-a virus or with control Jneo virus by incubating subconfluent monolayers of tumour cells with cell-free viral supernatants containing 5 ng/ml spolybrene, as previously described (39). Infected cells were selected and maintained in medium containing G418 (0.3 mg/ml active weight). To maintain any heterogeneity present within the starting population groups of at least 100 clones were pooled for further study. 2.15 Quantitation of TNF-a production by transduced Fsa-N cells TNF-a levels in supernants of retrovirally-transduced Fsa-N cells were determined by a commerciaFELISA kit using the procedure described by the manufacturer (Endogen). 2.15 Tumourigenicitv of Fsa-NJneo and Fsa-NJmTNF-a cells The tumourigenicity of Fsa-NJneo and Fsa-NJmTNF-a cells was assessed by determining the number of cells that had to be injected in order to generate tumors in 50% of subcutaneous sites (TD50 assay). Varying sizes of inocula were injected into a minimum of 10 sites in a minimum of 5 mice. Animals were examined 3 times per week and tumour take scored as positive when a 139 palpable mass of >5 mm in diameter was detected. TD50 values were then calculated using Logit analysis. 3 .0 Results 3.1 Growth of Fsa-N and Fsa-R tumours in C3H/HeN and C3H/HeJ mice 5 x 105 Fsa-N and Fsa-R tumour cells were inoculated subcutaneously into the left flank of recipient animals. Tumours were still growing rapidly at the time of their excision on day 28. Each tumour was weighed and as shown in Figure 19, there was no significant difference in the size of these two tumours in each of the two syngeneic murine strains. 3.2 Macrophage content of Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice Indirect immunoperoxidase staining was used to determine the macrophage content of Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice. The data obtained indicate that while Fsa-N tumours contain almost three times as many macrophages as Fsa-R tumours, there was no difference in the macrophage content of either tumour type when grown in C3H/HeN mice as compared to C3H/HeJ mice (Figure 20). 3.3 TNF-a production by tumour-associated macrophages present in Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice Northern blot analysis was used to determine the production of TNF-a mRNA by the macrophages present within Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice. As shown in Figure 21, substantially less TNF-a mRNA appears to be present in whole 140 EJ C3H/HeN • C3H/HeJ Fsa-N Fsa-R Tumour cell line Figure 19. Growth of Fsa-N and Fsa-R tumours in C3H/HeN and C3H/HeJ mice. 5 x 105 tumour cells were inoculated subcutaneously into the left flank of recipient animals. Tumours were excised on day 28 and weighed. Results show the mean weight ± SD of tumours grown in 3 mice per group. For both tumour cell lines no significant difference in tumour weight was observed in C3H/HeN and C3H/HeJ animals 141 80-i E3 C3H/HeN i i C3H/HeJ Fsa-N Fsa-R Tumour cell line Figure 20. Macrophage content of Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice. The percentage of macrophages present in enzymatically disaggregated tumour cell suspensions prepared from Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice was determined by indirect immunperoxidase staining using mAb TIB 128 (anti-Mac-1). Each determination was made on duplicate preparations for each experiment. The data shown represents the mean ± SEM of at least 5 independent tumours. 200 cells were counted on each slide. There is no significant difference between the macrophage content of Fsa-N and Fsa-R tumours grown in either of the two murine strains 142 Fsa-N Fsa-R u 3 O S 3 <D "o Z *-, <U <L> DC X •o a «s 8 S3 • 3 O S 3 H Z <u £ CO 00 J3 s e« s •—> <u X #9 • • TNF-a Actin l s S 3 <u - s o • i Z -> <u <u DC X C5 3 i u 3 O S 3 z <u DC ex s o CTj s !—> <u E Figure 21. Northern blot analysis of TNF-a mRNA expression in Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice. Ten ng of total cellular RNA isolated from whole enzymatically disaggregated Fsa-N and Fsa-R tumours grown in C3H/HeN or C3H/HeJ mice, or macrophages isolated from these tumours by brief adherence to plastic, were run in each lane and the filters hybridized with a32P-labeled probe encompassing the entire coading region of the murine TNF-a cDNA. To confirm equal loading, blots were stripped and reprobed with a murine actin probe. 143 enzymatically disaggregated tumour cell suspensions prepared from Fsa-N tumours grown in C3H/HeN mice as compared to equivalent cell suspensions prepared from tumours grown in C3H/HeJ mice. Similar differences in TNF-a expression were also observed in macrophages purified from these tumours by adherence to plastic. Moreover, although little TNF-a mRNA could be detected in whole enzymatically disaggregated Fsa-R tumour cell suspensions (reflecting the lower macrophage content of these tumours), macrophages purified from Fsa-R tumours grown in C3H/HeJ mice also contained greatly reduced quantitities of TNF-a mRNA (Figure 21). In order to test whether the differences in the amount of TNF-a mRNA found in the macrophages present in tumours grown in C3H/HeN and C3H/HeJ mice resulted in similar differences in the amount of TNF-a protein produced, lysates prepared from enzymatically disaggregated tumour cell suspensions were examined for TNF-a expression by Western blot analysis (Figure 22). Both the 26 kD and 17 kD TNF-a species were readily detected in the lysates of Fsa-N tumours grown in C3H/HeN mice, but little if any TNF-a protein could be detected in lysates prepared in a similar way from tumours grown in C3H/HeJ mice. The level of TNF-a in lysates prepared from Fsa-R tumours grown in C3H/HeN mice was low as expected from the lower macrophage content of these tumours, and no TNF-a was detectable in lysates prepared from the Fsa-R tumours that had been grown in C3H/HeJ mice. 3.4 Production of GM-CSF by TNF-a- and IL-lB-stimulatedFsa-N cells As shown in Figure 23, Fsa-N tumour cells do not constitutively produce GM-CSF although they can be readily induced to do so by stimulation with TNF-a or IL-lp. While a dose response is obtained with these cytokines over the range of 1-10 ng/ml, above this level, little additional GM-CSF is produced. Moreover, there does not appear to be a synergistic interaction between TNF-a and IL-lp in the induction of GM-CSF by Fsa-N cells (Figure 24). 144 43 kD - • 29 kD - I 18kD-H 14kD~H 6kD-B 3kD-B o fl I — » GO' tin sfJmTNF Fsa-N S w w PL, ffi ffi - • - .- ~ - + ... i#£5P>*"*"*»-. < , ' % " • ? Figure 22. Western blot analysis of TNF-a protein expression in Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice. Lysates of enzymatically disaggregated tumour cell suspensions prepared from Fsa-N and Fsa-R tumours grown in C3H/HeN and C3H/HeJ mice were separated on a 15% SDS-PAGE gel, transferred to nitrocellulose, probed with a polyclonal rabbit anti-murine TNF-a antiserum followed by a peroxidase conjugated goat anti-rabbit IgG and the reaction developed in PBS containing 0.06% 3'-3 diaminobenzidine and 0.012% H2O2. E 50000 "I 40000 ~ 30000 -2 0 0 0 0 -10000 -145 1 10 100 IL-lb (ng/mi) 1000 40000 -i TNF (ng/ml) Figure 23. Production of GM-CSF by TNF-a- and IL-ip-stimulated Fsa-N cells 1 x 105 Fsa-N cells were incubated together with various concentrations of (a) IL-lp or (b) TNF-a and conditioned supernatants removed 48 h later and assayed for their ability to support the proliferation of the GM-CSF-dependent cell line B6SUtA. Each point represents the mean ± SD of quadruplicate wells. 146 50000 1 40000 -30000 -20000 " 1 0 0 0 0 -10 100 TNF (ng/ml) H o 0 i • io 0 100 IL-1 (ng/ml) Figure 24. Lack of synergy between TNF-a and IL-lp in the stimulated production of GM-CSF by Fsa-N cells 1 x 105 Fsa-N cells were incubated together with various concentrations of IL-lp and TNF-a. Conditioned supernatants were removed 48 h later and assayed for their ability to support the proliferation of the GM-CSF-dependent cell line B6SUtA. Each point represents the mean± SD of quadruplicate wells. 147 3.5 Production of GM-CSF bv tumour cells purified from Fsa-N tumours grown in C3H/HeN and C3H/HeJ mice Purified tumour cells were isolated from enzymatically disaggregated Fsa-N tumour cell suspensions by staining with a pool of mAbs directed against determinants expressed on hemopoietic cells (TIB213/CD18, TIB128/CDllb and YE1.21/CD45) and sorting the negatively stained cell population on the FACS (Figure 25). Medium conditioned by these purified tumour cells for 48 h was then tested using a sensitive bioassay for the presence of GM-CSF. As shown in Figure 26, Fsa-N tumour cells purified from tumours grown C3H/HeN and C3H/HeJ mice both produced low and approximately equal amounts of GM-CSF. 3.6 Introduction and expression of the murine TNF-a gene in Fsa-N tumour cells Retroviral-mediated gene transfer was used to introduce and express the murine TNF-a cDNA in the Fsa-N tumour cell line. Northern blot analysis confirmed the presence within the transduced cells of a major RNA species of approximately 4.4 kb that hybridized with a murine TNF-a probe (Figure 27) corresponding to the size expected for the full length retroviral transcript driven off the 5' LTR of JmTNF-a. Western blot analysis further demonstrated the presence within Fsa-NJmTNF-a cells but not control Fsa-NJneo cells of TNF-a proteins of approximately 26 kD and 17 kD (Figure 22). Moreover, using a sensitive ELISA technique, 1 ml of tissue culture medium conditioned by 105 Fsa-NJmTNF-a cells for 48 h was shown to contain approximately 50 ng/ml TNF-a. 3.7 Growth of Fsa-NJneo and Fsa-NJmTNF-a cells in vivo Although Fsa-NJneo and Fsa-NJmTNF-a cells exhibited similar growth rates in vitro, their growth in vivo was dramatically and reproducibly different. Thus while inoculation of only 2.8 x 103 Fsa-NJneo was sufficient to induce tumour growth in 50% of mice, 5.3 x 106 Fsa-NJmTNF-a cells were needed to produce a similar tumour take (Figure 28). Moreover, those 148 UNSORTED SORTED 3 C ^ i C3H/HeN C3H/HeJ 1 — - -ie-Fluorescence intensity Figure 25. Isolation of tumour cells from Fsa-N tumours grown in C3H/HeN and C3H/HeJ mice. Purified tumour cells were isolated from enzymatically disaggregated tumour cell suspensions by staining with a pool of mAbs directed against determinants expressed on hemopoietic cells (CD 18, CD l ib and CD45) and sorting the negatively stained tumour cell population using a FACS. 149 8000 -| -a HeJ -» HeN %CM Figure 26. GM-CSF production by tumour cells purified from Fsa-N tumours grown in C3H/HeN and C3H/HeJ mice. 1 x 105 Fsa-N tumour cells isolated from C3H/HeN and C3H/HeJ mice were added in a final volume of 1ml to each well of a 24 well plate. Conditioned supematants were removed 48 h later and assayed for their ability to support the proliferation of the GM-CSF-dependent cell line B6SUtA. Each point represents the mean ± SD of quadruplicate wells. 8 PU o I S 1 2 2 i i tU PU 9.49 kb -7.46 kb " 4.40 kb - m T N F - a 2.37 kb " 1.35 kb ~ 0.24 kb " Actin 150 Figure 27. Northern blot analysis of TNF-a mRNA expression in Fsa-NJneo and Fsa-NJmTNF-a cells. Ten \ig of total cellular RNA isolated from Fsa-NJneo and Fsa-NJmTNF-a cells were run in each lane and the filters hybridized with a 32P-labeled probe encompassing the entire coding region of the murine TNF-a cDNA. To confirm equal loading of RNA, blots were stripped and reprobed with a 1800 bp Pstl fragments derived from the plasmid pAl-actin. 151 100 HI < o S h-h-Z LU O cr 111 a. Fsa-NJmTNF-a LOG CELL NUMBER INJECTED 1 o Figure 28. Growth of Fsa-NJneo and Fsa-NJmTNF-a cells in vivo. Various numbers of Fsa-NJneo and Fsa-NJmTNF-a cells were injected into a minimum of 10 sites in a minimum of 5 C3H/HeN mice. Animals were examined 3 times a week and tumour take scored as positive when a palpable mass of >5 mm in diameter was detected at any time within 2 months of injection. TD50 values were then calculated using Logit analysis.  1 5 2 Fsa-N tumours that did grow, grew very slowly and exhibited a high rate of spontaneous n). 4 .0 Discussion Tumour-associated macrophages and lymphocytes have both been shown to produce TNF-a (1, 2). However, at present, the molecular mechanisms that regulate the production and functional activity of this molecule within the microenvironment of tumours remain poorly defined. In the present study, we demonstrate that macrophages present within tumours grown in Lpsd CSH/HeJ mice express substantially lower quantities of TNF-a mRNA and protein than the macrophages present in similar numbers within tumours derived from the same initial cell populations but grown in Lpsn C3H/HeN mice. These data indicate that the signals that operate in situ to induce the production of TNF-a by tumour-associated macrophages act, at least in part, via the same pathway that is defective in C3H/HeJ mice. Inspite of intense investigation, the molecular nature of the Lpsd defect has remained elusive (20). Cells from C3H/HeN and C3H/HeJ appear to bind LPS equally well (40, 41) and do not differ m Jheir expression of any of the specific LPS-binding proteins that have been identified to date (42-4W: Attention has therefore focused on the possibility that C3H/HeJ mice may be defective in a signal transduction pathway triggered by LPS binding. In this regard, recent studies by Ding et al. (49) point toward a relationship between microtubules and the product of the Lps gene. These authors demonstrated that taxol, a potent -inhibitor of microtubule depolymerization, resembles LPS in its ability to induce TNF-a release and down-regulate the expression of TNF-a receptors on the surface of macrophages. Using St nbred strains of mice a direct correlation was demonstrated between the ability of macrophages to down-regulate their TNF-a receptors in response to taxol, and their responsiveness to.L^S. Thus it was concluded that taxol responsiveness is closely linked to the Lps gene and that microtubules or microtubule-associated proteins may be involved in LPS her these molecules themselves or other downstream proteins regulated 153 by microtubule depolymerization may be defective in Lpst* mice. In this regard, it is interesting to note that Shinji et al. (50) have demonstrated that the reorganization of microfiliments normally seen after stimulation of macrophages with LPS does not occur in macrophages isolated from M C3H/He^ mice. Other studies have suggested that a GTP-binding (G) protein, or the coupling of this i molecul©4o the LPS receptor, may be defective in Lps^mice. This conclusion is based largely on the observation that Pertussis toxin, which inactivates the a subunit of Gj and other G proteins ia ADP-ribosylation, can inhibit B cell and macrophage responses to LPS. This is further supported by recent studies suggesting a role for G[2 in the stimulation of U937 cells by LPS (52). 3||the other hand, Shinomiya and Nakano (53) have demonstrated that the calcium ionophore, (187, stimulates IL-1 production in Lpsn, but not Lpsd, macrophages although intracellular free calcium was increased in both cell types. Further studies provided evidence that macrophages from £3H/HeJ mice may lack a particular calmodulin-binding protein (54). The molecular mechanisms that operate within tumours to switch on TNF-a production Femain to be determined. Direct contact between tumour cells and macrophages has been shown to induce the production of TNF-a in some systems (55). This effect can be blocked by antibodies fflv ) ) suggesting the involvement of this molecule in the process. Crosslinking of own to induce human monocytes to secrete TNF-a (56). Moreover, r cells, including those used in this study, secrete high levels of hyaluronan gand for CD44. Indeed hyaluronan itself has been shown to induce macrophages to producrTNF-a. s also been shown to be induced by ionizing radiation (58) and hypoxia (59) ression of the gene may be controlled by the redox potential of the cell. Both -R tumours are hypoxic when grown in subcutaneous sites with mean pC>2 mHg respectively (Personal communication, Dr. Dai Chaplin, CRC Gray lesex, England). Interestingly, it has recently been demonstrated ion is a potent inducer of TNF-a in C3H/HeN mice, very little TNF-a is ed C3H/HeJ mice (Dougherty, et al, manuscript in preparation). These 1 5 4 data sugges| that C3H/HeJ mice are defective in their response to reactive oxygen intermediates. Whether they're similarly defective in their response to hypoxia remains to be determined. Given tHe wide and varied effects of TNF-a, the fact that tumours producing quite different levels ofJthe cytokine grow at the same rate and contain the same proportion of tumour-associated macrophages is intriguing. The simplest explanation for this finding, is that the level of TNF-a in C3H/HeJ tumours is already optimal and that additional cytokine produced within urs has no further effect. In support of this possibility, in vitro studies nstrated that maximum levels of GM-CSF were produced following stimulation of Fsa-N th TNF-a at 10 ng/ml and that no additional GM-CSF was produced when TNF-a was t up to 100 ng/ml. Similarly, tumour cells isolated from Fsa-N tumours grown in /HeN and C3H/HeJ mice expressed approximately equal amounts of GM-CSF. Alternatively, the cells present within the tumour microenvironment may simply be ry to the effects of TNF-a. It has recently been demonstrated that the cytotoxic activity of is severely inhibited at oxygen tensions similar to those found within the tumours used in dy (i.e., <2%) (60, 61). Moreover, many of the signals (including hypoxia) that induce hages to produce TNF-a also induce the production of soluble TNF-a receptors that may  activity of the molecule (59). This may be a major mechanism by which the ais regulated in vivo. In this latter model, introduction of the TNF-a cDNA into its tumour growth because it results in TNF-a expression without the co-incident 155 5. References 1 Rubbert A • Manger, D.; Lang, N.; Calden, I. R.; Platzer, E. Functional characterization of ' tumour "infiltrating lymphocytes, lymph node lymphocytes and peripheral lymphocytes. Int J Cancer. 1991; 49: 25-31. 2 Pusztai, L.; Clover, L. M.; Cooper, K.; Starkey, P. M.; Lewis, C. E.; McGee, J. CD. Expression of tumour necrosis factor a and its receptors in carcinoma of the breast. Br J • Cancer. 1994;70:289-292. 3 Goeddle, D. V.; Aggarwal, B. B.; Gray, P. W.; Leung, D. W.; Nedwin, G.E.; Palladino M. A.; Patton, J. S. Tumor necrosis factors: gene structure and biological activities. Cold p r i n g Harbor Symp Quant Biol. 1986; 51: 597-609. cey, K.; Vlassara, H.; Cerami, A. Cachectin/tumor necrosis factor. Lancet. 1989; 20: 1122-i'l26.  Ruggiero, V.; Latham, K.; Baglioni, C. Cytostatic and cytotoxic activity of tumor necrosis factor on human cancer cells. J Immunol. 1987; 138: 2711-2717.  Sugarman, B. J.; Aggarwal, B. B.; Hass, P. E.; Figari, I. S.; Palladino, M. A.; Shepard H. ' M Recombinant human tumor necrosis factor-a: effects on proliferation of normal and transformed cells in vitro. Science. 1985; 230: 943-945.  Kirstein, M.; Baglioni, C. Tumor necrosis factor stimulates proliferation of human osteosarcoma cells and accumulation of c-myc messenger RNA. J Cell Physiol. iy»»; 134: 479-485. I Springer T A Traffic signals for lymphocyte recirculation and leukocyte emigration: the  multistep paradigm. Cell. 1994; 76: 301-314. 9.#Bevilacqua, M. P. Endothelial-leukocyte adhesion molecules. Annu Rev Immunol. 1993; 11: '•^A^i^-804. 10 Frater-Jbhroder, M.; Risau, W.; Hallmann, R.; Gautschi, P.; Bohlen, P. Tumor-necrosis faltor type a, a potent inhibitor of endothelial cell growth in vitro, is angiogenic in vivo . Ppc Natl Acad Sci USA. 1987; 84: 5277-5281. 11 Leibolich, S. J.; Polverini, P. J.; Shepard, H. M.; Wiseman, D. M.; Shively, V.; Nuseir, NTMacrophage-induced angiogenesis is mediated by tumor-necrosis factor-a . Nature. 1987; 239: 630-632. 12. Montrufchio, G.; Lupia, E.; Battaglia, E.; Passerini, G.; Bussolino, F.; Emanuelli, G.; Can?ussi, G. Tumor-necrosis factor a-induced angiogenesis depends upon in situ platllet-activating factor biosynthesis. J Exp Med. 1994; 180: 377-382. I 13. Beutler, B»; Cerami, A. Cachectin and tumor necrosis factor as two sides of the same biologic^teoin. Nature. 1986; 320: 584-588. 14. Brett, J.; GerlachgH.; Nawroth, P.; Steinberg, S.; Godman, G.; Stern, D. Tumor necrosis factor /cacKelin increases permeability of endothelial cell monolayers by a mechanism involvingjfegulatory G proteins. J Exp Med. 1989; 169: 1977-1991. 156 15. Bevilacqua, M.; Prober, J.; Majeau, G.; Fiers, W.; Cotran, R.; Gimbrone, M. Recombinant TNF induces procoagulant activity in endothelium. Proc Natl Acad Sci USA. 1986; 83: 4533-4537. 16. Carswell, E. A.; Old, L. J.; Kassel, R. L.; Green, S.; Fiore, N.; Williamson, B. An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci USA. 1975; 72: 3666-3670. 17. Teng, M. N.; Park, H. K.; Koeppen, K. J.; Tracey, K. J.; Fendly, B. M.; Schreiber, H. Long-term inhibition of tumor grow by tumor necrosis factor in the absence of cachexia or tumor immunity. Proc Natl Acad Sci USA. 1991; 88: 3535-3539. 18. Asher, A. L.; Mule, J. J.; Kasid, A.; Restifo, N. P.; Salo, J. C ; Reichert, C. M.; Jaffe, G.; Fendly, B.; Kriegler, M.; Rosenberg, S. A. Murine tumor cells transduced with the gene for tumor necrosis factor-a. Evidence for a paracrine immune effect of tumor necrosis factor against tumors. J Immunol. 1991; 146: 3227-3234. 19. Heppner, G.; Weiss, D. W. High susceptibility of strain A mice to endotoxin and endotoxin-red blood cell mixtures. J Bacterid. 1965; 90: 696-703. 10. Sultzer, B. M.; Castagna, R.; Bandekar, J.; Wong, P. Lipopolysaccharide nonresponder cells: the C3H/HeJ defect. Immunobiology. 1993; 187: 257-271. 1 . Watson, J.; Largen, M.; McAdam, K. P. W.J. Genetic control of endotoxic responses in mice. J Exp Med. 1978; 147: 39-49. Watson, J.; Kelly, K.; Largen, M.; Taylor, B. A. The genetic mapping of a defective LPS response gene in C3H/HeJ mice. J Immunol. 1978; 120: 422-424. 23. McBride, W. H. Phenotype and function of intratumoural macrophages. Biochem Biophys Res Commun. 1986; 865: 27-41. 24g|Howie, S.; McBride, W. H. Tumour specific T helper activity can be abrogated by two distinct suppressor cell mechanisms. Eur J Immunol. 1982; 12: 671-675. 25. McBride, W. H.; Howie, S. Induction of tolerance to a murine fibrosarcoma in two zones of db'sage-the involvement of suppressor cells. Br J Cancer. 1986; 53: 707-711. i 26. Dougherty, G. J.; McBride, W. H. Immunoregulating activity of tumor-associated macrophages. Cancer Immunol Immunother. 1986; 23: 67-72. 27. Dougherty, G. J.; McBride, W. H. Accessory cell activity of tumor associated macrophages. 'J Natl Cancer Inst. 1986; 76: 541-548. I 28. Hakomori, S.; Kannagi, R. Carbohydrate antigens in higher animals. Handbook of Experimental Immunology, Volume 1. Weir, D. M.; Herzenberg, L. A.; Blackwell, C ; Herlenberg, L. A., Eds.: Blackwell Scientific, Edinburgh; 1986: 9.1. > 29. Ho, M. K.;l3eringer, T. A. Mac-1 antigen: quantitative expression in macrophage population^and tissues, and immunofluorescent localization in the spleen. J Immunol. 1982; 128:,2fi81-2286. 30. Austyn, J. M.^Gordon, S. F4/80, a monoclonal antibody directed specifically against the mouse macrophage. Eur J Immunol. 1981; 11: 805-815. 157 31 Doueherty G. J.; Dougherty, S. T.; Kay, R. J.; Lansdorp, P.; Humphries, R. K Identification and characterization of 114/A10, an antigen highly expressed on the surface of murine myeloid and erythroid progenitor cells and IL-3-dependent cell lines. Exp Hematol. 1989; 17: 877-882. 32 Dougherty, G. J.; Kay, R. J.; Humphries, R. K. Molecular cloning of 114/A10, a cell surface antigen containing highly conserved repeated elements, which is expressed by murine hemopoietic progenitor cells and interleukin-3-dependent cell lines. J Biol Chem. 1989; 264: 6509-6514. 33 Kav R • Doueherty, G. J.; Humphries, R. K. Heterogeneous modifications of the 114/A10 ' protein of interleukin-3-dependent cells are concentrated in a highly repetitive amino-terminal domain. J Biol Chem. 1990; 265: 4962-4968. 34. Oiirgwin, J. M.; Przybyla, A. E.; MacDonald, R. J.; Rutter W. I Isolation of biologically active ribonucleic acid from sources enriched in nboneclease. Biochemistry. 19/9,18. 5294-5297. 35 Kav R J • Boissy, R. J.; Russnak, R. H.; Candido, E. P. M. Efficient transcription of a Caenorhabditis elegans heat shock gene pair in mouse fibroblasts is dependent on multiple promoter elements which can function bidirectionally. Mol Cell Biol. 1986; 6: 3134-3143. 36 Hattori M.; Sakaki, Y. Dideoxy sequencing method using denatured plasmid templates. Anal  Bio'chem. 1985; 152:232-238.  Laker, G; Stocking, C ; Bergholz, U.; Hess, N.; de Lamarter J. F.; Ostertag, W Autocrine stimulation after transfer of the granulocyte/macrophage colony-stimulating factor gene and autonomous growth are distinct but independent steps in the oncogenic pathway. Proc Natl Acad Sci USA. 1987; 84: 8458-8462. Markowitz D.; Goff, S.; Bank, A. A safe packaging line for gene transfer: separating viral genes on two different plasmids. J Virol. 1988; 62: 1120-1124. 39 Hughs, P. F. D.; Thacker, J. D.; Hogge, D.; Sutherland, H. J.; Thomas, T. E.; Lansdorp, P*M • Eaves, C. J.; Humphries, R. K. Retroviral gene transfer into primitive nonnaland leukemic progenitor cells using clinically applicable procedures. J Clin Invest. 1992; 89: 1817-1824. 40 Gregory, S. H.; Simmerman, D. H.; Kern, M. The lipid A moiety of lipopolysaccharide is spfcifcially bound to B cell subpopulations of responder and nonresponder animals . J Immunol. 1980; 125: 102-107. 41 Eldridge J. H.; Jacobs, D. M. Surface phenotype of LPS-binding murine lymphocytes. Proc Soc'Exp Biol Med. 1984; 175: 458-467. 42 Lei M.*G.; Morrison, D. C. Specific endotoxin lipopolysaccharide-binding proteins on murine splenocytes. I. Detection of lipopolysaccharide-binding sites on splenocytes and spleno%te subpopulations . J Immunol. 1988; 141: 996-1005. 43. Lei, M. G.; Mofrison, D. C. Specific endotoxin lipopolysaccharide-binding proteins on murine'splen^ytes. II. Membrane localization and binding characteristics. J Immunol. 1988; 141: 1006-1011. 158 44. Golenbock, D. T.; Hampton, R. Y.; Raetz, C. R. H.; Wright, S. D. Human phagocytes have multiple lipid A-binding sites. Infect Immunol. 1990; 58: 4069-4075 45. Hampton, R. Y.; Golenbock, D. T.; Penman, N.; Krieger, M.; Raetz, C. R. H. Recognition and plasma clearance of endotoxin by scavenger receptors. Nature. 1991; 352: 342-344. 46. Wright, S. D.; Ramos, R. A.; Tobias, P. S.; Ulevitch, R. J.; Mathison, J. C. CD 14, a receptor for complexes of lipopolysaccharide (LPS) and LPS binding protein. Science. 1990;249:1431-1433 47. Schumann, R. R.; Leong, S. R.; Flaggs, G. W.; Gray, P. W.; Wright, S. D.; Mathison, J. C.; Tobias, P. S.; Ulevitch, R. J. Structure and function of lipopolysaccharide binding protein. Science. 1990; 249: 1429-1431. 48. Kirikae, T.; Kirikae, F.; Schade, F. U.; Yoshida, M.; Kondo, S.; Hisatsune, K.; Nishikawa, S-I; Rietschel, E. Th. Detection of lipopolysaccharide-binding proteins on membranes of murine lymphocytes and macrophage-like cell lines. FEMS Microbiol Immunol. 1991; 3: 327-336. 49. Ding, A. H.; Porteu, F.; Sanchez, E.; Nathan, C. F. Shared actions of endotoxin and taxol on TNF receptors and TNF release. Science. 1990; 248: 370-372. 50. Shinji, H.; Kaiho, S.; Nakano, T.; Yoshida, T. Reorganization of microfilaments in macrophages after LPS stimulation. Exp Cell Res. 1991; 193: 127-133. 51. Jakway, J. P.; DeFranco, A. L. Pertussis toxin inhibition of B cell and macrophage responses to bacterial lipopolysaccharide. Science. 1986; 234: 743-746. 52. Daniel-Issakani, S.; Spiegel, A. M.; Strulovici, B. Lipopolysaccharide response is linked to the GTP binding protein, Gi2, in the promonocytic cell line U937. J Biol Chem. 1989; 264:20240-20247. 5 3 ^ Shinomiya, H.; Nakano, M. Calcium ionophore A23187 does not stimulate lipopolysaccharide nonresponsive C3H/HeJ peritoneal macrophages to produce in|erleukin 1. J Immunol. 1987; 139: 2730-2736. 54. Terad% Y.; Shinomya, H.; Nakano, M. Defect of calmodulin-binding protein in expression of mterleukin-1 p gene by LPS-non-responder C3H/HeJ mouse macrophages. Biochem Biophys Res Comm. 1989; 158: 723-726. 55. Zembala, M.; Siedlar, M.; Ruggiero, I.; Wieckiewicz, J.; Mytar, B.; Mattei, M.; Collizzi, V. the MHC class-II and CD44 molecules are involved in the induction of tumour necrosis factor (TNF) gene expression by human monocytes stimulated with tumour cells . Int J Career. 1994; 56: 269-274. 56. Webb, D. S.; Shimizu, Y.; Van Seventer, G. A.; Shaw, S.; Gerrard, T. L. LFA-3, CD44, and CD45: physiologic triggers of human monocyte TNF and IL-1 release. Science. 1990;>249: 1295-1297. 57. McBride, W!' H^Extracellular hyaluronidase sensitive coats produced by tumor cells and their relevance |«ietastasis. Treatment of Metastasis: Problems and Prospects. Hellman, K.; Eccles,S. AT^ds. ; 1985: 351-354. 159 58. Hallahan, D. E.; Spriggs, D. R.; Beckett, M. A.; Kufe, D. W.; Weichselbaum, R. R. Increased tumor necrosis factor mRNA following ionizing radiation exposure. Proc Natl Acad Sci USA. 1989; 86: 10014-10017. 59 Scanneli G.; Waxman, K.; Kami, G. J.; loli, G.; Gatanaga, T.; Yamamoto, R.; Granger, G. A. Hypoxia induces a human macrophage cell line to release tumor necrosis factor-a and its soluble receptors in vitro. J Surg Res. 1993; 54: 281-285. 60 Matthews, N.; Neale, M. L.; Jackson, S. K.; Stark, J. M. Tumour cell killing by tumour necrosis factor: inhibition by anaerobic conditions, free- radical-scavengers and inhibitors of arachidonate metabolism. Immunology. 1987; 62: 153-155. 61 Sampson, L. K.; Chaplin, D. J. The influence of microenvironment on the cytotoxicity of TNFa in vitro. Int J Radiat Oncol Biol Phys. 1994; (in press). 160 CHAPTER 5 Conclusions and Future Directions Previous studies have suggested that tumour-associated macrophages are derived from both the differentiation of monocytes recruited into a tumour site from the peripheral blood (1,2), and the localized proliferation of macrophage progenitor cells within the tumour microenvironment (3, 4). At present, however, neither the relative importance of each of these processes nor the precise molecular mechanisms involved, have been well defined. It seems likely that any factor that is chemotactic or chemokinetic for monocytes and/or macrophage progenitor cell types, or which can promote the survival, proliferation and/or differentiation of these cells might play a role in determining the overall macrophage content of a tumour (2). The results presented in this Thesis indicate that fibroblasts and fibroblastoid tumour cells ate capable alone of providing all of the signals necessary to support the sustained generation of myeloid progenitor cells in vitro. Moreover, studies of the relative supportive ability of fibroblasts derived from various mutant murine strains (Sl/Sl and op/op) suggest an important role for fibroblast-derived M-CSF in this function. Interestingly, however, tumours generated from M-CSF-deficient tumour cell lines were found to contain approximately the same proportion of macrophages as tumours derived from equivalent M-CSF-producing cells, although the later population€id express higher levels of several M-CSF inducible genes including IL-ip, TNF-a, and FcRyll. In contrast, the macrophage marker 114/A10 was expressed at a similar level in both tumour types. These latter findings indicate that, although M-CSF does not seem to play an important role in regulating the macrophage content of at least these tumours, it can modulate their functional activity. Tumours derived from transformed M-CSF-producing fibroblasts also grew somewhat faster than tumours initiated from equivalent M-CSF-deficient cells suggesting the involvement of M-CIF, itself, or M-CSF-induced genes in tumour growth. A second fibroblast-derived cytokine, GM-CSF, has also been shown to promote monocyte survival and differentiation in vitro, and to induce the proliferation of peripheral blood-derived 161 CFU-M and (with varying efficiency and in synergy with other factors) other more primitive clonogenic cell types including blast colony-forming cells, CFU-GEMM, BFU-E and CFU-MEG. When administered in vivo, GM-CSF causes a dose-dependent increase in the number of circulating granulocytes (neutrophils and eosinophils) and monocytes. It is also chemotactic for monocytes, and can induce the expression on endothelial cells of various adhesion molecules thought to be involved in leukocyte extravasation. Importantly, tumours generated from cells genetically engineered to express high levels of GM-CSF have been found by others to contain increased numbers of macrophages and to exhibit dramatically reduced tumourigenicity and metastatic potential (5, 6). To determine the possible involvement of tumour cell-derived GM-CSF in regulating the accumulation of tumour-associated macrophages in vivo, the type of experiments described in this Thesis could be repeated using fibroblastoid cell lines derived from GM-CSF knockout animals (7) and M-CSF/GM-CSF double-deficient animals (8). It would also be of interest to compare the number and functional activity of macrophages isolated from tumours grown in normal and equivalent cytokine deficient hosts. Since tumour cells transformed with Polyoma Large T are likely to be immunogeneic in syngeneic hosts, another approach would need to be used to generate tumourigenic cell lines. In this regard, I (in collaboration with G.J. Dougherty, TFL) have already used a retroviral vector encoding a mutant ras to generate a number of transformed M-CSF-deficient op/op cell lines and found that these grow well in non-immunocompromised syngeneic hosts. Provisional data was also obtained in this study indicating that M-CSF-deficient op/op fibroblasts produce a soluble mediator that promotes monocyte survival in vitro. Although GM-CSF is capable of this activity it did not appear to be constitutively produced by either the op/op or op/+ fibroblasts used in these experiments. However, in order to formally rule out the involvement of GM-CSF in this process, media conditioned by fibroblastoid cells derived from GM-CSF knockout (7) and M-CSF/GM-CSF double-deficient animals (8) should be tested for their ability to promote monocyte survival in vitro. In the event that medium conditioned by these cells also supports monocyte survival, attempts could be made to identify, characterize and eventually clone, 162 the molecule involved. Such studies would be greatly facilitated by my observation that medium conditioned by COS7 cells does not have supportive activity. I would anticipate that an expression cloning stategy using a CDM8-based cDNA library prepared from RNA isolated from op/op cells might be employed. The other major finding reported in this study, is that the signals that operate within tumours to induce the expression of TNF-a (and IL-lp) by tumour-associated macrophages act via the same signal transduction pathway that is defective in Lpsd C3H/HeJ mice. Further characterization of these signals will probably have to await elucidation of the molecular nature of the Lps& defect. Recently, hypoxic conditions analogous to those found within Fsa-N and Fsa-R tumours have been to shown to trigger the production of TNF-a and soluble TNF-a receptors by macrophages (9). Moreover, preliminary studies indicate that C3H/HeJ mice also produce greatly reduced levels of TNF-a in response to ionizing radiation (Personal communication, Dr. William H. McBride, UCLA). One mechanism by which radiation appears to induce gene expression is via the generation of reactive oxygen intermediates which, in turn, directly or indirectly activate the transcription factor NF-KB (10-12). Treatment of cells with LPS has also been shown to induce the expression of reactive oxygen intermediates and to activate NF-KB (13), suggesting that radiation and LPS may feed into a common signal transduction pathway. It would thus be of considerable interest to examine the possibility that the expression of TNF-a by macrophages within tumours is controlled by changes in redox potential acting via NF-KB. If this is the case, then one would perhaps expect to see differences in the activation status of NF-KB (i.e. in the extent of an association between the p65 and p50 subunits of NF-KB and IKB) (12) in C3H/HeJ and C3H/HeN mice following exposure to ionizing radiation. 163 References 1. Mantovani, A. Tumor-associated macrophages. Curr Opin Immunol. 1990; 2: 689-692. 2. Mantovani, A.; Bottazzi, B.; Colotta, F.; Sozzani, S.; Ruco, L. The origin and function of tumor-associated macrophages. Immunol Today. 1992; 13: 265-270. 3. Stewart, C. C. Local proliferation of mononuclear phagocytes in tumors. J Reticuloendothel Soc. 1983;34:23-27. 4. Evans, R.; Cullen, R. T. In situ prolferation of intratumor macrophages. J Leukocyte Biol 1984;35:561-572. 5. Dranoff, G.; Jaffee, E.; Lazenby, A.; Golumbek, P.; Levitsky, H.; Brose, K.; Jackson, V.; Harnada, H.; Pardoll, D.; Mulligan, R. C. Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity. Proc Natl Acad Sci USA 1993-90: 3539-3543. 6. Tepper, R. I.; Mule, J. J. Experimental and clinical studies of cytokine gene-modified tumor cells. Human Gene Therapy. 1994; 5: 153-164. 7. Stanley, E.; Lieschke, G. J.; Grail, D.; Metcalf, D.; Hodgson, G.; Gall, J. A. M.; Maher, D. W.; Cebon, J.; Sinickas, V.; Dunn, A. R. Granulocyte-macrophage colony-stimulating factor-deficient mice show no major perturbation of hematopoiesis but develop a characteristic pulmonary pathology. Proc Natl Acad Sci USA. 1994; 91: 5592-5596. 8. Lieschke, G. J.; Stanley, E.; Grail, D.; Hodgson, G.; Sinickas, V.; Gall, J. A.; Sinclair, R. A.; Dunn, A. R. Mice lacking both macrophage- and granulocyte-macrophage colony-stimulating factor have macrophages and coexistent osteopetrosis and severe lung disease. Blood. 1994; 84: 27-35. 9. Scanneli, G.; Waxman, K.; Kami, G. J.; Ioli, G.; Gatanaga, T.; Yamamoto, R.; Granger, G. A. Hypoxia induces a human macrophage cell line to release tumor necrosis factor-cc and its soluble receptors in vitro. J Surg Res. 1993; 54: 281-285. 10. Weichselbaum, R. R.; Hallahan, D. E ; Sukhatme, V.; Dritschilo, A.; Sherman, M. L.; Kufe, D. W. Biological consequences of gene regulation after ionizing radiation. J Natl Cancer Inst. 1991; 83: 480-484. 11. Meyer, M.; Schresk, R.; Baeuerle, P. A. H2C>2 and antioxidants have opposite effects on the activation of NF-KB and AP-1 in intact cells: AP-1 as a secondary antioxidant-responsive element. EMBO J. 1993; 12: 2005-2015. 12. Beg, A. A.; Finco, T. S.; Nantermet, P. V.; Baldwin, A. S. Jr. Tumor necrosis factor and interleukin-1 lead to phosphorylation and loss of IKBCX: a mechanism for NF-KB activation. Mol Cell Biol. 1993; 13: 3301-3310. 13. Miiller J. M.; Loms Ziegler-Heitbrock, H. W.; Baeuerle, P. A. Nuclear factor kappa B, a mediator of lipopolysaccharide effects. Immunobiology. 1993; 187: 233-256. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088837/manifest

Comment

Related Items