Monocyte-enriched cells on calcified tissues

Enhanced osteolytic potential of monocytes/macrophages derived from bone marrow after particle stimulation

BACKGROUND

Total hip replacement can be complicated by periprosthetic osteolysis. Monocytes/macrophages play a major role in the formation of the foreign body granulomas induced by wear debris. We hypothesized that periprosthetic monocytes/macrophages do not only accelerate inflammatory and osteoclast-mediated osteolytic processes, but also resorb periprosthetic bone directly by themselves. This study was designed to evaluate the osteolytic potential in vitro of monocytes/macrophages derived from bone marrow.

METHODS

Monocytes/macrophages were produced by filtration of rat bone marrow cells, followed by culture in the presence of macrophage-colony stimulating factor (M-CSF). Monocyte/macrophage properties were ascertained using immunocytochemistry and phagocytic activity. Osteolytic cytokines and extracellular matrix degrading proteinases were quantified at the mRNA level.

RESULTS

Adherent cell fraction was immunoreactive for the monocyte/macrophage specific marker CD68 and active in the phagocytosis of carbon particles up to 72 h. They also showed immunoreactivity to cathepsin K, IL-1beta, IL-6, and M-CSF, but mostly did not react to TRAP. mRNA levels of osteolytic cytokines and extracellular matrix degrading proteinases were enhanced, but that of RANKL were not. Monocytes/macrophages resorbed dentine discs and carbonated calcium phosphate was very actively resorbed after stimulation with titanium particles.

DISCUSSION

Harvested bone marrow cells expressed monocyte/macrophage phenotype, but not osteoclastic markers. The capacity of these cathepsin-K-positive phagocytic cells to resorb dentine discs and carbonated calcium phosphate in vitro suggests a direct role of monocytes/macrophages in bone resorption and periprosthetic osteolysis. The finding supports our hypothesis and previous histomorphometric observations on the presence of such osteolytic macrophages in vivo around loosening prosthesis.

Rodent osteoblast-like cells support osteoclastic differentiation of human cord blood monocytes in the presence of M-CSF and 1,25 dihydroxyvitamin D3

Fracture repair requires the involvement of osteoclasts (OC), multinucleated cells which are responsible for bone resorption and form by fusion of circulating mononuclear haemopoietic precursors. The nature of these circulating precursor cells, in particular their relationship to blood monocytes, is uncertain. To define further the nature of the circulating human OC precursor, and to determine the role bone stromal cells and humoral factors play in the differentiation of OCs, we co-cultured human umbilical cord blood monocytes with UMR106.01 osteoblast-like cells in the presence and absence of 1,25 dihydroxyvitamin D3 [1,25 (OH)2D3], macrophage-colony stimulating factor (M-CSF) and dexamethasone on both bone slices and coverslips. Isolated cells were positive only for monocyte/macrophage markers (CD11a, CD11b, CD14 and HLA-DR) and negative for OC markers [tartrate resistant acid phosphatase (TRAP), vitronectin receptors (VNR) and calcitonin receptors (CT receptors)] and did not form resorption pits on bone slices after 24 hr in culture. However, after 14 days in co-culture with UMR106.01 cells, in the presence of 1,25 (OH)2D3 and M-CSF, numerous TRAP, CT receptor and VNR positive multinucleated cells capable of extensive lacunar bone resorption were formed in these co-cultures. The presence of 1,25 (OH)2D3, M-CSF and a bone-derived stromal cell population were absolute requirements for OC differentiation. It is concluded that mononuclear phagocytes are capable of differentiating into mature functional OCs when cultured under specific cellular and hormonal conditions. This is vitro model of human OC differentiation should prove useful in further analysing factors controlling OC generation in bone remodelling and repair.

Cells of the mononuclear phagocyte series differentiate into osteoclastic lacunar bone resorbing cells

Although the osteoclast shares several features with other cells of the mononuclear phagocyte system (MPS), its precise cellular ontogeny is unknown, and its membership of the MPS is controversial. This study examined whether various cells of the MPS can be induced to differentiate into cells capable of the highly specialized osteoclastic function of lacunar bone resorption. We isolated mouse and rat monocytes, mouse (liver, peritoneal, alveolar, brain) tissue macrophages, and spleen and marrow haemopoietic cells, as well as foreign body macrophages and macrophage polykaryons derived from subcutaneous granulomas formed by implantation of latex beads and coverslips in mice. When these cells were incubated with UMR106 osteoblast-like cells on glass coverslips and human cortical bone slices in the presence of 1,25-dihydroxy vitamin D3 [1,25(OH)2D3] for 7 and 14 days, numerous tartrate-resistant acid phosphatase-positive cells formed in these co-cultures and scanning electron microscopy revealed extensive lacunar resorption of the bone surface. Bone resorption was seen as early as 4 days after monocytes were co-cultured with UMR106 cells. With the exception of bone marrow-derived cells, lacunar resorption was not seen in the absence of UMR106 cells. These findings show that a bone-derived stromal cell element is necessary for differentiation of monocytes and tissue and inflammatory macrophages into osteoclast-like cells capable of extensive lacunar bone resorption, and would argue in favour of osteoclast membership of the MPS.

In vitro bone resorption by isolated multinucleated giant cells from giant cell tumour of bone: light and electron microscopic study

The behaviour of multinucleated giant cells (GCs), obtained from a giant cell tumour of the tibia and cultured on glass coverslips or on devitalized bone slices, was studied using light and electron microscopy. Monitoring the GCs on bone slices by phase-contrast microscopy revealed that they had removed calcified bone matrix resulting in excavation of lacunae, with subsequent lateral extension and perforation of the bone slices. Electron microscopy demonstrated for the first time that the GCs responsible for exavating lacunae had two specific membrane modifications, ruffled border and clear zone, and showed basically similar cytoplasmic fine structures to those of osteoclasts. Fluorescence images of the GCs on glass and on bone after rhodamine-conjugated phalloidin staining revealed that most of the GCs had an intensely fluorescent peripheral band composed of a number of F-actin dots called podosomes. Some GCs showed unusual arrangements of podosomes suggesting abortive attempts at GC formation. We have demonstrated that the band structure of the GCs cultured on bone is intimately involved in bone resorption. Two stromal cell types could be recognized. The predominant type, which seemed to be the only neoplastic element because of its proliferative capability, showed quite different fine structural and cytoskeletal features from the GCs. The other type, which was much less frequent and seemed not to proliferate, had morphological similarities to the GCs, and seemed to be their precursor. Importantly GCs cultured on bone and the osteoclasts share common structures for adhesion to and resorption of bone, strongly supporting the view that the GCs of the giant cell tumour of bone are potentially active bone resorbers and can be regarded as osteoclasts.

Osteoclast cytomorphometry demonstrates an abnormal population in B cell malignancies but not in multiple myeloma

Increased bone resorption in the vicinity of myeloma cells is mediated by local stimulating factors. Other malignancies of the B cell lineage are also able to produce resorbing factors responsible for increased bone resorption. We have studied three groups of subjects: 10 patients with overt multiple myeloma, 10 patients with a B cell malignancy, and 10 healthy human subjects as controls. Patients were studied at the time of diagnosis and had a transiliac bone biopsy. Osteoclasts were evident on histological sections by their acid phosphatase activity. A software was developed on an automatic image analyzer (Leitz TAS+) for measuring the maximal Feret's diameter (Oc.Le) of each osteoclast (corresponding to the osteoclast length). The histogram of Oc.Le frequency distribution was supplied in each group. In myeloma patients, the Oc.Le frequency distribution was similar to that in normal subjects and showed the histogram to be asymetric with a positive skew (maximum peak at 20-25 microns). With a graphical analysis, this distribution was shown to follow a lognormal distribution corresponding to a homogeneous osteoclast population. In other B cell malignancies, Oc.Le displayed a bimodal distribution with a peak at 20-25 microns and a lower peak at 10-15 microns. The graphical analysis showed that small (mononucleated?) osteoclasts are present in B cell malignancies with normal osteoclasts. This might reflect the secretion of different soluble factors by malignant cells of the B lymphocyte lineage.

Bone resorption in vitro: macrophages and giant cells from failed total hip replacement versus osteoclasts

Macrophages and giant cells which ingested material particles in loosening of total hip prostheses were tested for their ability to resorb bone in vitro, using osteoclasts as the control. Macrophages and giant cells did not form pits or resorption lacunae on the bone substrates as osteoclasts did. The results support the view that around implants also bone resorption is mediated by osteoclasts. A role of macrophages in the attachment phase of bone resorption is suggested.

Osteoclast-like cells form in long-term human bone marrow but not in peripheral blood cultures

Transplantation studies have suggested that peripheral blood mononuclear cells contain precursors for osteoclasts. Thus we tested the capacity of peripheral blood monocytes to form osteoclasts in long-term culture. We have reported previously that mononuclear cells from feline, baboon, and human marrow form osteoclast-like cells in long term cultures. Further, the formation of these cells is increased in response to bone resorption stimulatory agents such as PTH, interleukin 1, and transforming growth factor alpha. We now report that these cells show characteristic cytoplasmic contraction with calcitonin and form resorption lacunae when cultured on sperm whale dentine. Thus, these bone marrow-derived multinucleated cells fulfill the functional criteria for osteoclasts. Although cultured peripheral blood monocytes can be induced to form multinucleated cells with 1,25-dihydroxyvitamin D3, these cells did not show similar responses to the osteotropic factors as multinucleated cells formed in the bone marrow cultures multinucleated cells. These results indicate that osteoclasts or cells closely related to osteoclasts form in long-term human bone marrow cultures. In contrast, few mononuclear cells in the peripheral blood appear capable of forming osteoclasts under the culture conditions used in these experiments.

Bone cell biology: the regulation of development, structure, and function in the skeleton

Bone cells compose a population of cells of heterogeneous origin but restricted function with respect to matrix formation, mineralization, and resorption. The local, mesenchymal origin of the cells which form the skeleton contrasts with their extraskeletal, hemopoietic relatives under which bone resorption takes place. However, the functions of these two diverse populations are remarkably related and interdependent. Bone cell regulation, presently in its infancy, is a complicated cascade involving a plethora of local and systemic factors, including some components of the skeletal matrices and other organ systems. Thus, any understanding of bone cell regulation is a key ingredient in understanding not only the development, maintenance, and repair of the skeleton but also the prevention and treatment of skeletal disorders.

Ultrastructural study of the long-term development of two experimental cutaneous calcinoses (topical calciphylaxis and topical calcergy) in the rat

Skin calcification induced by topical calciphylaxis was provoked by a subcutaneous injection of iron chloride in rats previously sensitized by dihydrotachysterol. A cutaneous topical calcergy was induced by an injection of potassium permanganate. An electron-microscopical study of the long-term evolution of both these models of calcification was made. After the initial stages, mineralization of the connective tissue continued by a secondary nucleation process without matrix vesicles. The mineral composed of needle-like structures, apatite in nature, was mainly deposited between and around collagen fibrils, and showed various arrangements in calcified plaques. Intrafibrillar calcification was rarely observed and appeared only in the later stages. The extension of calcified deposits then stopped. Finally, there was a fragmentation of the mineralized area which was progressively surrounded by uncalcified collagen fibrils. A demineralization process, caused by cells such as macrophages and multinucleated giant cells, rather than a resorption of the calcified deposits, was noted. It is important to emphasize that, in both models of ectopic calcification, an evolution toward ectopic ossification was never observed, which is perhaps due to the absence of extensive resorption mechanisms.

Ultrastructure of the giant cell infiltrate of subcutaneously implanted bone particles in rats and mice

The giant cells of soft tissues and those of mineralized tissues (osteoclasts) have distinctly different cell surface receptors and ultrastructural characteristics. Recently, the removal of dead bone particles in a subcutaneous environment has been described as a prototype of bone resorption, and a major issue is whether the giant cells that surround these ectopic bone implants and the processes involved in the disruption of bone surfaces are the same as those in the skeleton. We have compared the cytology and ultrastructure of giant cells recruited to subcutaneously implanted isogeneic bone particles with similar features of osteoclasts in metaphyseal bone of young normal rats and mice. Giant cells on surfaces of bone particles 2, 3, and 4 weeks after implantation were multinucleated, had a homogeneous, nonvacuolated cytoplasm, and had a bone surface interface unremarkable by light microscopy. In a few cells randomly distributed, small cytoplasmic vacuoles were present and large vacuoles were noted next to the bone surface at high magnification. By transmission electron microscopy, folded membrane configurations forming extensive interdigitations with adjacent cells were prominent features on most surfaces of giant cells. In instances where these interdigitations abutted bone surfaces, configuration resembling a ruffled border were noted, but these regions were always part of two different cells when examined at lower magnification or in serial sections. Breakdown of bone particles appeared to be by phagocytosis of small pieces and subsequent intracellular digestion in electron-dense cytoplasmic vacuoles. Osteoclasts from these same young animals were smaller with fewer nuclei, had cytoplasmic vacuoles concentrated next to bone surfaces, and had characteristic ruffled borders and clear zones. These results confirm those of others that native osteoclasts and multinucleated giant cells on dead bone particles are distinctly different with respect to both ultrastructure and mechanism of disruption of bone surfaces.