In vivo role of macrophage growth factors as delineated using CSF-1 deficient op/op mouse

Total absence of CSF-1 in the op/op mouse leads to a profound and generalized deficiency of macrophages and to osteopetrosis subsequent to the absence of osteoclasts. These observations confirm that CSF-1 is a genuine regulator of macrophage and osteoclast formation in vivo. Further studies in affected animals have shown that the CSF-1 absence variably affects macrophage differentiation stages and different organ macrophage populations, and that functionally competent macrophages are produced in low numbers without CSF-1, presumably under the influence of GM-CSF and IL-3. The op/op mice have increased levels of both endogenous GM-CSF and IL-3, which apparently are not fully able to compensate for the absence of CSF-1. Macrophage deficiencies but not osteoclast deficiencies in the op/op mouse could be completely corrected by exogenous GM-CSF, while exogenous CSF-1 corrects both osteoclast and macrophage deficiencies, but only in those tissues which could be reached by CSF-1 from the circulation. Despite severe quantitative macrophage deficiencies, the op/op mice demonstrate normal in vivo phagocytosis and immune functions suggesting that CSF-1 dependent macrophages do not contribute significantly to those processes in vivo. On the other hand, the op/op mice demonstrate severe secondary deficiencies of TNF-alpha, IL-1 alpha, and G-CSF suggesting that major function of CSF-1 dependent macrophages is the release of monokines.

CSF-1 deficiency in the op/op mouse has differential effects on macrophage populations and differentiation stages

Osteopetrosis and the absence of colony-stimulating factor 1 (CSF-1) in op/op mice are associated with decreased cellularity of the bone marrow (to one tenth of the normal), a very significant reduction in the number of cells recovered from peritoneal, pleural, and alveolar lavages, moderate leukopenia, and a slight decrease in the number of cells per spleen and thymus. Furthermore, op/op mice possess deficiencies in the number of macrophages in various organs. These cells are apparently absent in the bone marrow, severely reduced (5%-15% of the normal number) in peritoneal and pleural cavities and in the lungs. In addition, a marked decrease in the frequency and total number of circulating monocytes is present (5% of the normal). The deficiency of macrophages is less severe in the liver, spleen, and thymus of op/op mice (approximately 30% of those seen in normal). There is a concomitant redistribution of macrophage progenitor cells (granulocyte-macrophage colony-forming units, CFU-GM) in op/op mice from the marrow to the spleen and liver, associated with an increased sensitivity to interleukin 3 (IL-3). Their total number is decreased at least threefold compared to control mice. Moreover, op/op mice have at least a fivefold reduction in the total number of day-11 spleen colony-forming units (CFU-S) associated with their redistribution to the spleen and liver. These data suggest that the macrophage system in op/op mice is reduced at all levels tested, that is, at the level of mature macrophages, the level of progenitors, and the level of stem cells, whereas the redistribution of progenitor and stem cells could be viewed as a secondary consequence of osteopetrosis. Furthermore, these data suggest that macrophage dependency in vivo on CSF-1 is limited and different in various organs. Particularly in the liver, spleen, and thymus, other growth factors may significantly compensate for CSF-1 deficiency. Based on the relative decrease in the number of CFU-GM in the op/op mice, it appears that the population size of these progenitors is less dependent on CSF-1 than the hematopoietic stem cell population size as evidenced by the day-11 CFU-S assay. The day-11 CFU-S population is severely reduced in op/op mice, suggesting a physiological involvement of CSF-1 in expanding its size. These data provide evidence that CSF-1, besides acting on the final and intermediate stages of macrophage maturation, may also play a role in early stages of hematopoiesis.

Distinct in vivo functions of two macrophage subpopulations as evidenced by studies using macrophage-deficient op/op mouse

The op/op mice totally lack macrophage growth factor colony-stimulating factor (CSF)-1 and thus, by definition are completely depleted of CSF-1-dependent functions of the macrophage cell lineage. Moreover, they possess a severe and generalized macrophage deficiency. However, residual macrophages of these mice should still have normal CSF-1-independent functions. Studies designed to elucidate this issue have revealed that op/op mice are capable of normal in vivo phagocytic function and demonstrate normal humoral and cellular response postimmunization with sheep red blood cells. However, release of monokines such as tumor necrosis factor and granulocyte CSF following administration of endotoxin is severely impaired in op/op mice as compared with littermate controls. These studies suggest that the CSF-1-dependent macrophage population (absent in the op/op mouse) is primarily responsible for regulatory functions of these cells mediated by monokines, while the CSF-1-independent macrophage population (present in the op/op mouse) is primarily responsible for the classical macrophage functions in immunity such as phagocytosis, antigen processing and presentation.

Administration of colony stimulating factor-1 corrects some macrophage, dental, and skeletal defects in an osteopetrotic mutation (toothless, tl) in the rat

The toothless (tl/tl) mutation in the rat results in a paucity of osteoclasts and osteopetrosis that cannot be corrected by bone marrow transplantation. In the present study we demonstrate that tl/tl rats also have profound deficiencies of femoral, peritoneal, and pleural cavity macrophages. Furthermore, the macrophage colony stimulating activity of post-endotoxin sera from tl/tl rats is substantially reduced, suggesting that, as in the case of the op mutation in mice, the basis of the tl mutation is a deficiency of the macrophage growth factor, colony stimulating factor-1 (CSF-1). Consistent with this suggestion, treatment of tl/tl rats from birth for up to six weeks with CSF-1 reduced the osteopetrosis, increased body weight, and permitted tooth eruption. In addition, CSF-1 treatment induced large numbers of osteoclasts in tl/tl bones and macrophages in the peritoneal cavity and bone marrow. Persistence of metaphyseal sclerosis, however, indicated that the disease was not totally corrected by this treatment. These studies indicate that the basis of the tl mutation is most likely another CSF-1 deficiency, and further emphasize the role of this growth factor in osteoclast differentiation.

A pregnancy defect in the osteopetrotic (op/op) mouse demonstrates the requirement for CSF-1 in female fertility

Correlative evidence suggests that maternal production of the mononuclear phagocyte growth factor colony stimulating factor-1 (CSF-1) regulates placental development. In order to study the role of CSF-1 in pregnancy the fertility of CSF-1-less osteopetrotic (op/op) mutant mice was investigated. Homozygous mutant crosses (op/op x op/op) were consistently infertile. As expected, op/op males were almost completely fertile when crossed with heterozygous females. Surprisingly, op/op females when mated to heterozygote males were fertile, although at a rate that was 46% of the rate for +/op females x op/op males. These data suggest that CSF-1 is required for pregnancy. However, a maternal CSF-1 source is not absolutely necessary in that pregnancies involving +/op fathers were partially rescued, suggesting that +/op fetuses and/or +/op seminal fluid provides CSF-1 or CSF-1-induced factors which compensate for the absence of maternally produced CSF-1. Despite the complete absence of CSF-1 in the uterus and placenta of op/op mice placental weights were normal, suggesting that proliferation of decidual cells and trophoblasts, both of which express the CSF-1 receptor, may not be solely regulated by CSF-1. Histochemical staining for F4/80 antigen was used to identify macrophages in the uterus and placenta. Uterine macrophages could not be detected in virgin op/op mice although they were abundant in +/op uteri. Interestingly, macrophages could be detected in op/op uteri as uncharacteristically rounded cells in early gestation, however, they were not maintained and no macrophages were apparent beyond Day 14 of pregnancy in op/op mice. Further studies in the osteopetrotic mouse will be useful in delineating those functions required for pregnancy that are regulated by CSF-1.

Correction by CSF-1 of defects in the osteopetrotic op/op mouse suggests local, developmental, and humoral requirements for this growth factor

Mice that are mutant at the op locus have a severe deficiency of mononuclear phagocytes due to an inactivating mutation in the CSF-1 (macrophage colony-stimulating factor, M-CSF) gene. op/op mice are toothless, possessing skeletal abnormalities, a low body weight, and compromised fertility; they are osteopetrotic due to a deficiency of osteoclasts. The congenital osteopetrosis, toothless phenotype, osteoclast deficit, and the defects in splenic and femoral macrophages were corrected by routes of administration of human recombinant CSF-1 that maintained normal circulating CSF-1 concentrations. Early restoration of circulating CSF-1 was required for rescue of the toothless phenotype, but only partially restored body weight. In contrast, the deficiencies of pleural and peritoneal cavity macrophages and the reduced female fertility were not corrected by restoration of circulating CSF-1. These results suggest that although circulating CSF-1 is required for osteoclast and macrophage production, local synthesis and action of the growth factor are important for certain target cell populations.

Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse

Osteopetrotic (op/op) mutant mice suffer from congenital osteopetrosis due to a severe deficiency of osteoclasts. Furthermore, the total number of mononuclear phagocytes is extremely low in affected mice. Serum, 11 tissues, and different cell and organ conditioned media from op/op mice were shown to be devoid of biologically active colony-stimulating factor 1 (CSF-1), whereas all of these preparations from littermate control +/+ and +/op mice contained the growth factor. The deficiency was specific for CSF-1 in that serum or conditioned media from op/op mice possessed elevated levels of at least three other macrophage growth factors. Partial correction of the op/op defect was observed following intraperitoneal implantation of diffusion chambers containing L929 cells, which in culture produce CSF-1 as their sole macrophage growth factor. No rearrangement of the CSF-1 gene in op/op mice was detected by Southern analysis. However, in contrast to control lung fibroblasts, which contained 4.6- and 2.3-kilobase CSF-1 mRNAs, only the 4.6-kilobase species was detected in op/op cells. An alteration in the CSF-1 gene is strongly implicated as the primary defect in op/op mice because they do not contain detectable CSF-1, their defect is correctable by administration of CSF-1, the op locus and the CSF-1 gene map within the same region of mouse chromosome 3, their CSF-1 mRNA biosynthesis is altered, and the op/op phenotype is consistent with the phenotype expected in a CSF-1 deficient mouse.

Interferon-gamma-treated K562 target cells distinguish functional NK cells from lymphokine-activated killer (LAK) cells

In vitro incubation of the erythroleukemic cell line K562 with interferon-gamma (IFN-gamma) renders these cells relatively resistant to natural killer (NK) cell lysis. However, such treatment does not alter their sensitivity to LAK cell lysis. Thus, the lytic susceptibility of interferon-gamma-treated K562 (I-K562) cells to LAK cells as opposed to its relative resistance to NK cell lysis provides a functional assay to help distinguish these two types of effector cells. The relative resistance of I-K562 for NK cell-mediated lysis was not secondary to the release of soluble factors or the frequency of Leu-19+, CD3+ T cells, residual IFN-gamma, or expression of MHC Class I molecules. Coincubation of I-K562 cells with NK or LAK cells overnight did not appreciably change the pattern of lytic responses against K562 and I-K562 target cells. However, incubation of PBMC in vitro with I-K562 but not native K562 in the presence of r-IL-2 leads to a marked decrease in the generation of LAK cells. The inhibition of LAK cell generation was not secondary to differences in the consumption of bioactive levels of IL-2. Differences in the lytic capability of NK and LAK effector cells suggest heterogeneity among cells that mediate such non-MHC-restricted lysis. Use was made of cells from a patient with a large granular lymphocyte lymphoproliferative disease (greater than 85% Leu-19+) to determine if such cells could be used to distinguish clonal population of cells which would represent NK or LAK cell function. Of interest was the finding that such cells, even after incubation in vitro with IL-2, showed lytic function representative of NK cells but not LAK cells. Data concerning the inhibition of LAK cell generation by I-K562 cells have important implications for future therapeutic trials of IFN-gamma and IL-2 in the treatment of human malignancies.

Postirradiation recovery of haemopoiesis in Steel mutant mice

The recovery of haemopoiesis in Steel mutant mice following 1 Gy sublethal irradiation is described. Steel homozygotes (S1/S1) did not display the abortive phase of erythropoietic recovery while the secondary phase of erythropoietic recovery was more pronounced in S1/S1 than in control (+/+) animals. On the contrary, the neutrophilopoietic recovery in S1/S1 mice was defective only during the secondary phase of recovery. Steel heterozygotes (S1/+) manifested similar, albeit less pronounced, defects. In the course of studies of recovery of eosioniphils it was observed that neither wild-type nor mutant animals expressed the abortive rise. Moreover, the kinetics of recovery of eosinophils was essentially different from both erythropoietic and neutrophilopoietic recovery, and the preirradiation level was reached in both normal and mutant animals on day 60 postirradiation as opposed to 24 and 35 days for erythropoiesis and neutrophils respectively.

Multiple lineage colony growth from human marrow in plasma clot diffusion chambers

Marrow cells from ten healthy adult donors were cultured in plasma clot diffusion chambers implanted intraperitoneally into mice. Host animals were conditioned by two injections of phenylhydrazine and 600 cGy of x-rays. Cultures (5 X 10(4) cells/chamber) were continued for between 2 and 40 days and the chambers were retransplanted into new host animals every 5 days. Following termination of cultures, plasma clots were stained with benzidine-hematoxylin and analyzed microscopically. Erythroid, neutrophil, monocyte, eosinophil, megakaryocyte, mixed, undifferentiated, and fibroblastoid colonies were grown with neutrophil, erythroid, monocyte, and eosinophil colonies being the most frequent. A total of between 25 and 60 colonies was observed per chamber at any time point.

Omental lymphoid organ as a source of macrophage colony stimulating activity in peritoneal cavity

To test the hypothesis that the omental lymphoid organ (OLO) made by peritoneal milky spots is either a source of haemopoietic (macrophage) progenitors or growth factors we attempted to culture OLO cells in vitro in a variety of assay combinations. With the culture in vitro in semisolid agar it was found that OLO cells do not form granulocyte-macrophage or macrophage colonies in response to stimulants. However, when in the same assay marrow cells were used as the targets and OLO-related preparations as stimulants it was observed that marrow cells formed exclusively macrophage colonies. These marrow cells, in response to stimulants derived from other organs, produced granulocyte-macrophage, granulocyte and macrophage colonies. OLO-related preparations tested for macrophage-colony stimulating activity included partly purified medium conditioned by OLO cells derived from mice, either injected with endotoxin or not, and medium conditioned by OLO cells after 14 days, liquid culture in vitro. While these results were observed in Swiss mice, C3H/W mice, which are genetically endotoxin-unresponsive, failed to show this reaction. These data may suggest that the local production of macrophage-colony stimulating activity in the peritoneal cavity could be one physiological role for OLO. OLO is the first organ in adult mice identified to stimulate exclusively macrophage colony growth, and not granulocyte-macrophage or pure granulocyte colonies.