Search

The immune system has profound effects on bone remodeling. IFN-gamma, a major product of immune cells, potently inhibits bone resorption, but its mechanism of action is unknown. We found in cultures of stroma-free mononuclear precursors that IFN-gamma strongly suppresses TRANCE/RANKL-induced osteoclast formation in a dose-dependent manner. This direct effect on osteoclast progenitors was not due to stimulation of NO production by IFN-gamma, as the NOS inhibitors 1400W and L-NAME were unable to reverse the suppression. However, TGFbeta(1), which has opposing actions to IFN-gamma on diverse cellular functions, was able to antagonize the effect of IFN-gamma. This suggests that IFN-gamma prevents osteoclast formation by actively directing the differentiation of osteoclastic progenitors toward an alternative cytocidal lineage to the osteoclast., (Copyright 2000 Academic Press.)

It is generally considered that osteoblastic cells are essential for osteoclast formation. We tested the ability of hemopoietic tissue to differentiate osteoclastic characteristics in the absence of osteoblastic cells. We found that large numbers of calcitonin-receptor positive (CTRP) cells can be induced by prostaglandin E2 and 1,25(OH)2 vitamin D3, in cultures of hemopoietic mouse spleen. Moreover, spleen stromal cells were equivalent to bone marrow stromal cells in CTRP-cell induction. The spleen CTRP cells did not resorb bone, but were rapidly induced to full resorptive activity upon osteoblast addition. Thus, bone cells may not be essential for osteoclast formation, but rather are required to activate and regulate the resorptive function of mature osteoclasts.

We have recently developed an experimental model whereby mechanical stimulation induces osteogenesis in the caudal vertebrae of rats. We used this model to assess expression of genes induced by mechanical loading. Bulk preparations of mRNA extracted after loading did not show > 2-fold increases in expression of mRNA for matrix proteins or growth factors in Northern blotting analysis. c-jun was undectable. However, c-fos showed a 4-fold increase in expression within 60 mins of loading, before returning to control levels by 4 hrs. This increase was associated with intense signals in in situ hybridization, not seen in any nonloaded vertebrae, for c-fos over cortical osteocytes: thus osteocytes respond to mechanical loading with c-fos expression so strongly as to be visible even in the bulk RNA preparations. The results represent persuasive evidence for a role for osteocytes, and for c-fos, in the osteogenic response of bone to mechanical stimulation.

Hepatocyte growth factor (HGF) stimulates the growth, motility and morphogenesis of a variety of cell types, including hemopoietic progenitors. We found that HGF is a potent inhibitor of bone resorption by isolated rat osteoclasts. However, in the presence of the osteoblastic cell line UMR 106, it stimulated osteoclastic resorption. HGF also increased osteoclastic motility and spread area, over a similar concentration range. We detected no effect on osteoclast formation or survival. Our data suggest that HGF may be involved in the recruitment of osteoclasts to sites of bone resorption, but that during migration resorptive functions are suppressed. Once on resorptive sites, the osteoclast response to HGF is modulated by osteoblastic cells and the bone resorptive activity of osteoclasts may be stimulated accordingly.

Osteoclasts have been shown to produce reactive oxygen intermediates (ROI) and it has been suggested that ROI are involved in the process of bone resorption. ROI have also been shown to play a central role in the activation of the multisubunit transcription factor NF-kappa B that enhances the transcription of genes encoding defence and signaling proteins. Therefore, we have assessed the effect of pyrrolidine dithiocarbamate (PDTC), an oxygen-radical scavenger and metal chelator that is a selective and potent inhibitor of NF-kappa B activation, on osteoclastic bone resorption in the bone slice assay. PDTC (0.001-0.1 mM) dose-dependently and non-cytotoxically inhibited osteoclast activity with an IC50 of 0.01 mM. PDTC (0.01 mM) caused no change in the ratio of resorption pit area to resorption pit depth as measured by Lasertec confocal microscopy, indicating that ROI are not involved in the resorptive process per se. This view is supported by time-course studies showing that addition of PDTC or N-acetyl cysteine (NAC; an ROI scavenger, but not metal chelator), 6 hr after the start of the assay had no significant effect on subsequent bone resorption. Desferal (100 microM), a chelator of iron and other metal ions, had no significant effect on bone resorption, indicating (along with the results with NAC) that ROI-scavenging rather than metal chelation is responsible for inhibition of osteoclastic bone resorption by PDTC. Taken together these results indicate that ROI produced by osteoclasts in the bone slice assay are not involved in the process of bone resorption, but are important during osteoclast activation for bone resorption, possibly being involved in activation of the transcription factor NF-kappa B.

Mithramycin is an mRNA synthesis inhibitor that has been used to decrease bone resorption in patients with humoral hypercalcemia and Paget's disease. During studies on the mechanism of action of mithramycin it became clear that the compound has a direct inhibitory effect on osteoclastic bone resorption in the in vitro bone slice assay. At concentrations of 0.1-100 nM mithramycin directly inhibited osteoclastic bone resorption dose-dependently up to 66 +/- 5% at 100 nM (mean +/- SEM, 3 expts.). Another mRNA synthesis inhibitor, actinomycin D (0.1-100 nM) and the protein synthesis inhibitor, cycloheximide (0.1-10 microM), also dose-dependently inhibited osteoclastic bone resorption by 78 +/- 7% at 100 nM and 76 +/- 7% at 10 microM, respectively. Mithramycin and actinomycin D at 100 nM did not affect osteoclast survival on bone slices and were therefore not cytotoxic at the concentrations used. Mithramycin (100 nM) and cycloheximide (10 microM) both slightly decreased osteoclast cytoplasmic spreading. Addition of 100 nM mithramycin 6 hr after osteoclast adhesion to bone slices still inhibited subsequent resorption by 50%, indicating a continued but lesser requirement for mRNA synthesis during bone resorption. These results show that approximately 75% of osteoclasts obtained from neonatal rat long bones are activated by adhesion to mineralized bone surfaces and require mRNA and protein synthesis in order to resorb bone in vitro.

It has recently been shown that M-CSF is essential for osteoclast formation and survival. Paradoxically, the cytokine inhibits bone resorption by isolated osteoclasts. This implies that it may stimulate some function in osteoclasts not directly related to the bone resorptive process itself. We now report that M-CSF stimulates migration in isolated rat osteoclasts. This was achieved predominantly by a 3-8-fold increase in the proportion of osteoclasts demonstrating migration, combined with an increase in the area of substrate covered (43-96%) by each migrating cell. This action of M-CSF on migration is the inverse of its effect on resorption, wherein it has been previously found to reduce the proportion of osteoclasts that resorb bone and suggests that migration and resorption might represent alternative states of osteoclast behaviour. The results also suggest that osteoblast-derived M-CSF might play a role in the induction and regulation of localisation of these cells in bone.

We have examined the kinetics of the effects of inhibitors of the Na+/H(+)-antiporter (dimethylamiloride) and the vacuolar H(+)-ATPase (bafilomycin A1) on bone resorption by disaggregated rat osteoclasts in the bone slice assay. Bafilomycin A1 (100 nM) inhibited resorption by approximately 95%, 75%, 80% and 60% respectively, when added at t = 0, 1, 3 or 6 hr after osteoclast adherence to bone slices, during a 24 hr culture period. The incomplete inhibition by bafilomycin A1 when added after the start of incubation was presumably accounted for by resorption that had occurred prior to addition of the compound. Dimethylamiloride (100 microM) inhibited bone resorption by 80% and 65% when added at t = 0 or 1 hr after osteoclast adherence, but was without effect when added at t = 3 or 6 hr. In addition, dimethylamiloride but not bafilomycin A1 strongly inhibited osteoclast cytoplasmic spreading. The results indicate that Na+/H(+)-antiporter activity is essential for controlling intracellular pH during early activation events stimulated by the adherence of osteoclasts to mineralized bone surfaces, which lead to cytoskeletal activation, cell spreading and bone resorption.

Osteoblasts mediate much of the hormonal responsiveness of osteoclasts. We and others have found that one mechanism through which this regulation is effected is by release of osteoclast resorption-stimulating activity (ORSA) into culture supernatants. However, although hormonal responsiveness is regularly observed in co-cultures, ORSA is not always detectable in conditioned media. We show here that one explanation for this finding is that ORSA may be retained by heparin-like glycosaminoglycans (GAGs) of the cell surface or extracellular matrix of osteoblasts. We found that protease-sensitive ORSA could be extracted from monolayers of the osteoblastic cell line UMR 106 with 2M NaCl or collagenase. Production of this activity was increased in response to 1,25(OH)2D3. The presence of the GAG heparin was required to reveal ORSA. Immobilisation of ORSA by GAGs may assist osteoblastic cells in the regulation of the complex patterns of osteoclastic activity observed during skeletal morphogenesis and restructuring.

The op/op mouse, in which the M-CSF gene is mutated, has greatly reduced numbers of macrophages and osteoclasts. We assessed the ability of M-CSF to induce osteoclast and macrophage formation in op/op hemopoietic cells in vitro. Osteoclast production was undetectable in op/op cell cultures, but was restored by M-CSF at concentrations approximately an order of magnitude higher than those that induced macrophages. In normal hemopoietic tissue M-CSF similarly increased macrophage numbers, but inhibited osteoclast formation. Despite cure of the macrophage defect, neither interleukin 3 nor granulocyte-macrophage CSF were able to induce osteoclastic differentiation in op/op cells. The results suggest that M-CSF induces osteoclastic differentiation but that macrophages, which are also induced by M-CSF, suppress osteoclast differentiation. Macrophages induced by other cytokines seem unable to contribute to osteoclast-formation.

A subclone of an osteoblast-like osteosarcoma cell line (UMR 106.01) has recently been shown to possess specific binding sites for calcitonin gene-related peptide (CGRP) linked to adenylate cyclase. The present study provides the first demonstration for the production of immunoreactive CGRP from CGRP-receptor positive osteosarcoma cells. Mean immunoreactive CGRP levels were 15 pmol/g and 1 pmol/l for acid extracts of cells and cell-exposed media respectively. On gel filtration and high performance liquid chromatography, a major proportion of immunoreactive CGRP was found to co-elute with synthetic rat CGRP(1-37). Only negligible quantities of calcitonin were detected in cell extracts or cell-exposed supernatant. The production of authentic CGRP from a CGRP-receptor positive tumour suggests that the peptide may have autocrine effects on its producer cell.